CONVERSIONS BETWEEN U.S. CUSTOMARY UNITS AND SI UNITS
Times conversion factor
U.S. Customary unit
Equals SI unit
Accurate
Acceleration (linear)
foot per second squared
inch per second squared
ft/s2
in./s2
Area
square foot
square inch
ft2
in.2
0.09290304*
645.16*
Density (mass)
slug per cubic foot
slug/ft3
Density (weight)
pound per cubic foot
pound per cubic inch
lb/ft3
lb/in.3
Practical
meter per second squared
meter per second squared
m/s2
m/s2
0.0929
645
square meter
square millimeter
m2
mm2
515.379
515
kilogram per cubic meter
kg/m3
157.087
271.447
157
271
newton per cubic meter
kilonewton per cubic
meter
N/m3
joule (N⭈m)
joule
megajoule
joule
J
J
MJ
J
newton (kg⭈m/s2)
kilonewton
N
kN
newton per meter
newton per meter
kilonewton per meter
kilonewton per meter
N/m
N/m
kN/m
kN/m
0.3048*
0.0254*
0.305
0.0254
kN/m3
Energy; work
foot-pound
inch-pound
kilowatt-hour
British thermal unit
ft-lb
in.-lb
kWh
Btu
Force
pound
kip (1000 pounds)
lb
k
Force per unit length
pound per foot
pound per inch
kip per foot
kip per inch
lb/ft
lb/in.
k/ft
k/in.
Length
foot
inch
mile
ft
in.
mi
0.3048*
25.4*
1.609344*
0.305
25.4
1.61
meter
millimeter
kilometer
m
mm
km
Mass
slug
lb-s2/ft
14.5939
14.6
kilogram
kg
Moment of a force; torque
pound-foot
pound-inch
kip-foot
kip-inch
lb-ft
lb-in.
k-ft
k-in.
newton meter
newton meter
kilonewton meter
kilonewton meter
N·m
N·m
kN·m
kN·m
1.35582
0.112985
3.6*
1055.06
4.44822
4.44822
14.5939
175.127
14.5939
175.127
1.35582
0.112985
1.35582
0.112985
1.36
0.113
3.6
1055
4.45
4.45
14.6
175
14.6
175
1.36
0.113
1.36
0.113
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CONVERSIONS BETWEEN U.S. CUSTOMARY UNITS AND SI UNITS (Continued)
Times conversion factor
U.S. Customary unit
Moment of inertia (area)
inch to fourth power
Equals SI unit
in.4
4
inch to fourth power
in.
Accurate
Practical
416,231
416,000
0.416231 ⫻ 10
⫺6
millimeter to fourth
power
meter to fourth power
mm4
m4
kilogram meter squared
kg·m2
watt (J/s or N·m/s)
watt
watt
W
W
W
47.9
6890
47.9
6.89
pascal (N/m2)
pascal
kilopascal
megapascal
Pa
Pa
kPa
MPa
16,400
16.4 ⫻ 10⫺6
millimeter to third power
meter to third power
mm3
m3
meter per second
meter per second
meter per second
kilometer per hour
m/s
m/s
m/s
km/h
cubic meter
cubic meter
cubic centimeter (cc)
liter
cubic meter
m3
m3
cm3
L
m3
⫺6
0.416 ⫻ 10
Moment of inertia (mass)
slug foot squared
slug-ft2
1.35582
1.36
Power
foot-pound per second
foot-pound per minute
horsepower (550 ft-lb/s)
ft-lb/s
ft-lb/min
hp
1.35582
0.0225970
745.701
1.36
0.0226
746
Pressure; stress
pound per square foot
pound per square inch
kip per square foot
kip per square inch
psf
psi
ksf
ksi
Section modulus
inch to third power
inch to third power
in.3
in.3
Velocity (linear)
foot per second
inch per second
mile per hour
mile per hour
ft/s
in./s
mph
mph
Volume
cubic foot
cubic inch
cubic inch
gallon (231 in.3)
gallon (231 in.3)
ft3
in.3
in.3
gal.
gal.
47.8803
6894.76
47.8803
6.89476
16,387.1
16.3871 ⫻ 10⫺6
0.3048*
0.0254*
0.44704*
1.609344*
0.0283168
16.3871 ⫻ 10⫺6
16.3871
3.78541
0.00378541
0.305
0.0254
0.447
1.61
0.0283
16.4 ⫻ 10⫺6
16.4
3.79
0.00379
*An asterisk denotes an exact conversion factor
Note: To convert from SI units to USCS units, divide by the conversion factor
Temperature Conversion Formulas
5
T(°C) ⫽ ᎏᎏ[T(°F) ⫺ 32] ⫽ T(K) ⫺ 273.15
9
5
T(K) ⫽ ᎏᎏ[T(°F) ⫺ 32] ⫹ 273.15 ⫽ T(°C) ⫹ 273.15
9
9
9
T(°F) ⫽ ᎏᎏT(°C) ⫹ 32 ⫽ ᎏᎏT(K) ⫺ 459.67
5
5
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
www.brookscole.com
www.brookscole.com is the World Wide Web
site for Brooks/Cole and is your direct
source to dozens of online resources.
At www.brookscole.com you can find out
about supplements, demonstration
software, and student resources. You can
also send email to many of our authors and
preview new publications and exciting new
technologies.
www.brookscole.com
Changing the way the world learns®
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
Mechanics of
Materials
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
Mechanics of Materials
SIXTH EDITION
James M. Gere
Professor Emeritus, Stanford University
Australia • Canada • Mexico • Singapore • Spain
United Kingdom • United States
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
Publisher: Bill Stenquist
Editorial Assistant: Julie Ruggiero
Technology Project Manager: Burke Taft
Executive Marketing Manager: Tom Ziolkowski
Marketing Assistant: Jennifer Gee
Advertising Project Manager: Vicky Wan
Editorial Production Project Manager: Kelsey McGee
Print/Media Buyer: Rebecca Cross
Permissions Editor: Sommy Ko
Production Service: RPK Editorial Services, Inc.
Interior Designer: Jeanne Calabrese
Cover Designer: Denise Davidson
Cover Photo: Hackenberg/Masterfile
Cover Printing, Printing and Binding: Quebecor, Inc.
Composition: Better Graphics, Inc.
COPYRIGHT © 2004 Brooks/Cole, a division of Thomson
Learning, Inc. Thomson LearningTM is a trademark used herein
under license.
Brooks/Cole–Thomson Learning
10 Davis Drive
Belmont, CA 94002
USA
ALL RIGHTS RESERVED. No part of this work covered by the
copyright hereon may be reproduced or used in any form or by
any means—graphic, electronic, or mechanical, including but not
limited to photocopying, recording, taping, Web distribution,
information networks, or information storage and retrieval
systems—without the written permission of the publisher.
Printed in the United States of America
1
2
3
4
5
6
7
08
07
06
05
04
For more information about our products, contact us at:
Thomson Learning Academic Resource Center
1-800-423-0563
For permission to use material from this text, contact us by:
Phone: 1-800-730-2214
Fax: 1-800-730-2215
Web: http://www.thomsonrights.com
Library of Congress Control Number: 2003113085
ISBN 0-534-41793-0
Asia
Thomson Learning
5 Shenton Way #01-01
UIC Building
Singapore 068808
Australia /New Zealand
Thomson Learning
102 Dodds Street
Southbank, Victoria 3006
Australia
Canada
Nelson
1120 Birchmount Road
Toronto, Ontario M1K 5G4
Canada
Europe/Middle East/Africa
Thomson Learning
High Holborn House
50/51 Bedford Row
London WC1R 4LR
United Kingdom
Latin America
Thomson Learning
Seneca, 53
Colonia Polanco
11560 Mexico D.F.
Mexico
Spain/Portugal
Paraninfo
Calle/Magallanes, 25
28015 Madrid, Spain
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
Contents
Preface xiii
Symbols xvii
Greek Alphabet xx
1
Tension, Compression, and Shear
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
2
1
Introduction to Mechanics of Materials 1
Normal Stress and Strain 3
Mechanical Properties of Materials 10
Elasticity, Plasticity, and Creep 20
Linear Elasticity, Hooke’s Law, and Poisson’s Ratio 23
Shear Stress and Strain 28
Allowable Stresses and Allowable Loads 39
Design for Axial Loads and Direct Shear 44
Problems 49
Axially Loaded Members 67
2.1
2.2
2.3
2.4
2.5
2.6
2.7
★2.8
★2.9
★2.10
★2.11
★2.12
★Stars
Introduction 67
Changes in Lengths of Axially Loaded Members 68
Changes in Lengths Under Nonuniform Conditions 77
Statically Indeterminate Structures 84
Thermal Effects, Misfits, and Prestrains 93
Stresses on Inclined Sections 105
Strain Energy 116
Impact Loading 128
Repeated Loading and Fatigue 136
Stress Concentrations 138
Nonlinear Behavior 144
Elastoplastic Analysis 149
Problems 155
denote specialized and advanced topics.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
vii
viii
3
CONTENTS
Torsion
185
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
3.10
★3.11
4
Shear Forces and Bending Moments 264
4.1
4.2
4.3
4.4
4.5
5
Introduction 185
Torsional Deformations of a Circular Bar 186
Circular Bars of Linearly Elastic Materials 189
Nonuniform Torsion 202
Stresses and Strains in Pure Shear 209
Relationship Between Moduli of Elasticity E and G 216
Transmission of Power by Circular Shafts 217
Statically Indeterminate Torsional Members 222
Strain Energy in Torsion and Pure Shear 226
Thin-Walled Tubes 234
Stress Concentrations in Torsion 243
Problems 245
Introduction 264
Types of Beams, Loads, and Reactions 264
Shear Forces and Bending Moments 269
Relationships Between Loads, Shear Forces, and
Bending Moments 276
Shear-Force and Bending-Moment Diagrams 281
Problems 292
Stresses in Beams (Basic Topics) 300
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.9
5.10
★5.11
★5.12
★5.13
Introduction 300
Pure Bending and Nonuniform Bending 301
Curvature of a Beam 302
Longitudinal Strains in Beams 304
Normal Stresses in Beams (Linearly Elastic Materials) 309
Design of Beams for Bending Stresses 321
Nonprismatic Beams 330
Shear Stresses in Beams of Rectangular Cross Section 334
Shear Stresses in Beams of Circular Cross Section 343
Shear Stresses in the Webs of Beams with Flanges 346
Built-Up Beams and Shear Flow 354
Beams with Axial Loads 358
Stress Concentrations in Bending 364
Problems 366
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CONTENTS
6
Stresses in Beams (Advanced Topics) 393
6.1
6.2
6.3
6.4
6.5
6.6
6.7
6.8
6.9
★6.10
7
Analysis of Stress and Strain 464
7.1
7.2
7.3
7.4
7.5
7.6
7.7
8
Introduction 393
Composite Beams 393
Transformed-Section Method 403
Doubly Symmetric Beams with Inclined Loads 409
Bending of Unsymmetric Beams 416
The Shear-Center Concept 421
Shear Stresses in Beams of Thin-Walled Open Cross
Sections 424
Shear Stresses in Wide-Flange Beams 427
Shear Centers of Thin-Walled Open Sections 431
Elastoplastic Bending 440
Problems 450
Introduction 464
Plane Stress 465
Principal Stresses and Maximum Shear Stresses 474
Mohr’s Circle for Plane Stress 483
Hooke’s Law for Plane Stress 500
Triaxial Stress 505
Plane Strain 510
Problems 525
Applications of Plane Stress
(Pressure Vessels, Beams, and Combined Loadings)
8.1
8.2
8.3
8.4
8.5
541
Introduction 541
Spherical Pressure Vessels 541
Cylindrical Pressure Vessels 548
Maximum Stresses in Beams 556
Combined Loadings 566
Problems 583
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
ix
x
9
CONTENTS
Deflections of Beams
594
9.1
9.2
9.3
9.4
9.5
9.6
9.7
9.8
★9.9
★9.10
★9.11
★9.12
★9.13
Introduction 594
Differential Equations of the Deflection Curve 594
Deflections by Integration of the Bending-Moment
Equation 600
Deflections by Integration of the Shear-Force and Load
Equations 611
Method of Superposition 617
Moment-Area Method 626
Nonprismatic Beams 636
Strain Energy of Bending 641
Castigliano’s Theorem 647
Deflections Produced by Impact 659
Discontinuity Functions 661
Use of Discontinuity Functions in Determining Beam
Deflections 673
Temperature Effects 685
Problems 687
10 Statically Indeterminate Beams 707
10.1
10.2
10.3
10.4
★10.5
★10.6
Introduction 707
Types of Statically Indeterminate Beams 708
Analysis by the Differential Equations of the Deflection
Curve 711
Method of Superposition 718
Temperature Effects 731
Longitudinal Displacements at the Ends of a Beam 734
Problems 738
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CONTENTS
11 Columns
xi
748
11.1
11.2
11.3
11.4
11.5
11.6
11.7
11.8
11.9
Introduction 748
Buckling and Stability 749
Columns with Pinned Ends 752
Columns with Other Support Conditions 765
Columns with Eccentric Axial Loads 776
The Secant Formula for Columns 781
Elastic and Inelastic Column Behavior 787
Inelastic Buckling 789
Design Formulas for Columns 795
Problems 813
12 Review of Centroids and Moments of Inertia 828
12.1
12.2
12.3
12.4
12.5
12.6
12.7
12.8
12.9
Introduction 828
Centroids of Plane Areas 829
Centroids of Composite Areas 832
Moments of Inertia of Plane Areas 835
Parallel-Axis Theorem for Moments of Inertia 838
Polar Moments of Inertia 841
Products of Inertia 843
Rotation of Axes 846
Principal Axes and Principal Moments of Inertia 848
Problems 852
References and Historical Notes 859
Appendix A Systems of Units and Conversion Factors 867
A.1
A.2
A.3
A.4
A.5
Systems of Units 867
SI Units 868
U.S. Customary Units 875
Temperature Units 877
Conversions Between Units 878
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
xii
CONTENTS
Appendix B
B.1
B.2
B.3
B.4
B.5
Problem Solving
881
Types of Problems 881
Steps in Solving Problems 882
Dimensional Homogeneity 883
Significant Digits 884
Rounding of Numbers 886
Appendix C Mathematical Formulas
Appendix D
887
Properties of Plane Areas 891
Appendix E Properties of Structural-Steel Shapes 897
Appendix F
Properties of Structural Lumber 903
Appendix G Deflections and Slopes of Beams 905
Appendix H
Properties of Materials 911
Answers to Problems 917
Name Index 933
Subject Index 935
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
Preface
Mechanics of materials is a basic engineering subject that must be
understood by anyone concerned with the strength and physical
performance of structures, whether those structures are man-made or
natural. The subject matter includes such fundamental concepts as
stresses and strains, deformations and displacements, elasticity and
inelasticity, strain energy, and load-carrying capacity. These concepts
underlie the design and analysis of a huge variety of mechanical and
structural systems.
At the college level, mechanics of materials is usually taught during
the sophomore and junior years. The subject is required for most
students majoring in mechanical, structural, civil, aeronautical, and aerospace engineering. Furthermore, many students from such diverse fields
as materials science, industrial engineering, architecture, and agricultural engineering also find it useful to study this subject.
About this Book
The main topics covered in this book are the analysis and design of
structural members subjected to tension, compression, torsion, and
bending, including the fundamental concepts mentioned in the first
paragraph. Other topics of general interest are the transformations of
stress and strain, combined loadings, stress concentrations, deflections
of beams, and stability of columns.
Specialized topics include the following: Thermal effects, dynamic
loading, nonprismatic members, beams of two materials, shear centers,
pressure vessels, discontinuity (singularity) functions, and statically
indeterminate beams. For completeness and occasional reference,
elementary topics such as shear forces, bending moments, centroids, and
moments of inertia also are presented.
Much more material than can be taught in a single course is
included in this book, and therefore instructors have the opportunity to
select the topics they wish to cover. As a guide, some of the more
specialized topics are identified in the table of contents by stars.
xiii
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
xiv
PREFACE
Considerable effort has been spent in checking and proofreading the
text so as to eliminate errors, but if you happen to find one, no matter
how trivial, please notify me by e-mail (jgere@ce.stanford.edu). Then
we can correct any errors in the next printing of the book.
Examples
Examples are presented throughout the book to illustrate the theoretical
concepts and show how those concepts may be used in practical
situations. The examples vary in length from one to four pages, depending
upon the complexity of the material to be illustrated. When the emphasis
is on concepts, the examples are worked out in symbolic terms so as to
better illustrate the ideas, and when the emphasis is on problem-solving,
the examples are numerical in character.
Problems
In all mechanics courses, solving problems is an important part of the
learning process. This textbook offers more than 1,000 problems for
homework assignments and classroom discussions. The problems are
placed at the end of each chapter so that they are easy to find and don’t
break up the presentation of the main subject matter. Also, an unusually
difficult or lengthy problem is indicated by attaching one or more stars
(depending upon the degree of difficulty) to the problem number, thus
alerting students to the time necessary for solution. Answers to all
problems are listed near the back of the book.
Units
Both the International System of Units (SI) and the U.S. Customary
System (USCS) are used in the examples and problems. Discussions of
both systems and a table of conversion factors are given in Appendix A.
For problems involving numerical solutions, odd-numbered problems
are in USCS units and even-numbered problems are in SI units. This
convention makes it easy to know in advance which system of units is
being used in any particular problem. (The only exceptions are problems
involving the tabulated properties of structural-steel shapes, because the
tables for these shapes are presented only in USCS units.)
References and Historical Notes
References and historical notes appear immediately after the last chapter
in the book. They consist of original sources for the subject matter plus
brief biographical information about the pioneering scientists, engineers,
and mathematicians who created the subject of mechanics of materials.
A separate name index makes it easy to look up any of these historical
figures.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
PREFACE
xv
Appendixes
Reference material appears in the appendixes at the back of the book.
Much of the material is in the form of tables—properties of plane areas,
properties of structural-steel shapes, properties of structural lumber,
deflections and slopes of beams, and properties of materials (Appendixes D through H, respectively).
In contrast, Appendixes A and B are descriptive—the former gives a
detailed description of the SI and USCS systems of units, and the latter
presents the methodology for solving problems in mechanics. Included
in the latter are topics such as dimensional consistency and significant
digits. Lastly, as a handy time–saver, Appendix C provides a listing of
commonly used mathematical formulas.
S. P. Timoshenko (1878–1972)
Many readers of this book will recognize the name of Stephen P.
Timoshenko—probably the most famous name in the field of applied
mechanics. Timoshenko appeared as co-author on earlier editions of this
book because the book began at his instigation. The first edition, published in 1972, was written by the present author at the suggestion of
Professor Timoshenko. Although he did not participate in the actual
writing, Timoshenko provided much of the book’s contents because the
first edition was based upon his earlier books titled Strength of Materials.
The second edition of this book, a major revision of the first, was written
by the present author, and each subsequent edition has incorporated
numerous changes and improvements.
Timoshenko is generally recognized as the world’s most outstanding
pioneer in applied mechanics. He contributed many new ideas and
concepts and became famous for both his scholarship and his teaching.
Through his numerous textbooks he made a profound change in the
teaching of mechanics not only in this country but wherever mechanics
is taught. (A brief biography of Timoshenko appears in the first
reference at the back of the book.)
Acknowledgments
To acknowledge everyone who contributed to this book in some manner
is clearly impossible, but I owe a major debt to my former Stanford
teachers, including (besides Timoshenko) those other pioneers in
mechanics, Wilhelm Flügge, James Norman Goodier, Miklós Hetényi,
Nicholas J. Hoff, and Donovan H. Young. I am also indebted to my
Stanford colleagues—especially Tom Kane, Anne Kiremidjian, Helmut
Krawinkler, Kincho Law, Peter Pinsky, Haresh Shah, Sheri Sheppard,
and the late Bill Weaver. They provided me with many hours of discussions about mechanics and educational philosophy. My thanks also to
Bob Eustis, friend and Stanford colleague, for his encouragement with
each new edition of this book.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
xvi
PREFACE
Finally, I am indebted to the many teachers of mechanics and
reviewers of the book who provided detailed comments concerning the
subject matter and its presentation. These reviewers include:
P. Weiss, Valpariso University
R. Neu, Georgia Tech
A. Fafitis, Arizona State University
M.A. Zikry, North Carolina State University
T. Vinson, Oregon State University
K.L. De Vries, University of Utah
V. Panoskaltsis, Case University
A. Saada, Case Western University
D. Schmucker, Western Kentucky University
G. Kostyrko, California State University—Sacramento
R. Roeder, Notre Dame University
C. Menzemer, University of Akron
G. Tsiatas, University of Rhode Island
T. Kennedy, Oregon State University
T. Kundu, Univerity of Arizona
P. Qiao, University of Akron
T. Miller, Oregon State University
L. Kjerengtroen, South Dakota School of Mines
M. Hansen, South Dakota School of Mines
T. Srivatsan, University of Akron
With each new edition, their advice has resulted in significant
improvements in both content and pedagogy.
The editing and production aspects of the book were a source of
great satisfaction to me, thanks to the talented and knowledgeable
personnel of the Brooks/Cole Publishing Company (now a part of
Wadsworth Publishing). Their goal was the same as mine—to produce
the best possible results without stinting on any aspect of the book,
whether a broad issue or a tiny detail.
The people with whom I had personal contact at Brooks/Cole and
Wadsworth are Bill Stenquist, Publisher, who insisted on the highest
publishing standards and provided leadership and inspiration throughout
the project; Rose Kernan of RPK Editorial Services, who edited the
manuscript and designed the pages; Julie Ruggiero, Editorial Assistant,
who monitored progress and kept us organized; Vernon Boes, Creative
Director, who created the covers and other designs throughout the book;
Marlene Veach, Marketing Manager, who developed promotional material;
and Michael Johnson, Vice President of Brooks/Cole, who gave us his
full support at every stage. To each of these individuals I express my
heartfelt thanks not only for a job well done but also for the friendly and
considerate way in which it was handled.
Finally, I appreciate the patience and encouragement provided by
my family, especially my wife, Janice, throughout this project.
To all of these wonderful people, I am pleased to express my gratitude.
James M. Gere
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
Symbols
A
a, b, c
C
c
D, d
E
Er , Et
e
F
f
fT
G
g
H, h
I
Ix , Iy , Iz
Ix1, Iy1
Ixy
I x1 y1
IP
I1, I2
J
K
k
kT
L
area
dimensions, distances
centroid, compressive force, constant of integration
distance from neutral axis to outer surface of a beam
diameter, dimension, distance
modulus of elasticity
reduced modulus of elasticity; tangent modulus of elasticity
eccentricity, dimension, distance, unit volume change (dilatation)
force
shear flow, shape factor for plastic bending, flexibility, frequency (Hz)
torsional flexibility of a bar
modulus of elasticity in shear
acceleration of gravity
height, distance, horizontal force or reaction, horsepower
moment of inertia (or second moment) of a plane area
moments of inertia with respect to x, y, and z axes
moments of inertia with respect to x1 and y1 axes (rotated axes)
product of inertia with respect to xy axes
product of inertia with respect to x1y1 axes (rotated axes)
polar moment of inertia
principal moments of inertia
torsion constant
stress-concentration factor, bulk modulus of elasticity, effective length
factor for a column
苶/E
苶I苶
spring constant, stiffness, symbol for 兹P
torsional stiffness of a bar
length, distance
xvii
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
xviii
SYMBOLS
LE
ln, log
M
MP , MY
m
N
n
O
O⬘
P
Pallow
Pcr
PP , PY
Pr , Pt
p
Q
q
R
r
S
s
T
TP , TY
t
U
u
ur , ut
V
v
v⬘, v⬙, etc.
W
w
x, y, z
xc, yc, zc
x , y苶, 苶z
苶
Z
effective length of a column
natural logarithm (base e); common logarithm (base 10)
bending moment, couple, mass
plastic moment for a beam; yield moment for a beam
moment per unit length, mass per unit length
axial force
factor of safety, integer, revolutions per minute (rpm)
origin of coordinates
center of curvature
force, concentrated load, power
allowable load (or working load)
critical load for a column
plastic load for a structure; yield load for a structure
reduced-modulus load for a column; tangent-modulus load for a column
pressure (force per unit area)
force, concentrated load, first moment of a plane area
intensity of distributed load (force per unit distance)
reaction, radius
苶苶 )
radius, radius of gyration (r ⫽ 兹I/A
section modulus of the cross section of a beam, shear center
distance, distance along a curve
tensile force, twisting couple or torque, temperature
plastic torque; yield torque
thickness, time, intensity of torque (torque per unit distance)
strain energy
strain-energy density (strain energy per unit volume)
modulus of resistance; modulus of toughness
shear force, volume, vertical force or reaction
deflection of a beam, velocity
dv/dx, d 2 v/dx 2, etc.
force, weight, work
load per unit of area (force per unit area)
rectangular axes (origin at point O)
rectangular axes (origin at centroid C)
coordinates of centroid
plastic modulus of the cross section of a beam
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SYMBOLS
a
b
bR
g
gxy, gyz, gzx
g x1 y1
gu
d
⌬T
dP , d Y
e
ex, ey, ez
ex1, ey1
eu
e1, e2, e3
e⬘
eT
eY
u
up
us
k
l
n
r
s
sx, sy, sz
sx1, sy1
su
s1, s2, s 3
sallow
scr
spl
sr
angle, coefficient of thermal expansion, nondimensional ratio
angle, nondimensional ratio, spring constant, stiffness
rotational stiffness of a spring
shear strain, weight density (weight per unit volume)
shear strains in xy, yz, and zx planes
shear strain with respect to x1y1 axes (rotated axes)
shear strain for inclined axes
deflection of a beam, displacement, elongation of a bar or spring
temperature differential
plastic displacement; yield displacement
normal strain
normal strains in x, y, and z directions
normal strains in x1 and y1 directions (rotated axes)
normal strain for inclined axes
principal normal strains
lateral strain in uniaxial stress
thermal strain
yield strain
angle, angle of rotation of beam axis, rate of twist of a bar in torsion
(angle of twist per unit length)
angle to a principal plane or to a principal axis
angle to a plane of maximum shear stress
curvature (k ⫽ 1/r)
distance, curvature shortening
Poisson’s ratio
radius, radius of curvature (r ⫽ 1/k), radial distance in polar
coordinates, mass density (mass per unit volume)
normal stress
normal stresses on planes perpendicular to x, y, and z axes
normal stresses on planes perpendicular to x1y1 axes (rotated axes)
normal stress on an inclined plane
principal normal stresses
allowable stress (or working stress)
critical stress for a column (scr ⫽ Pcr /A)
proportional-limit stress
residual stress
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
xix
xx
SYMBOLS
sT
sU , sY
t
txy, tyz, tzx
tx1y1
tu
tallow
tU , tY
f
c
v
thermal stress
ultimate stress; yield stress
shear stress
shear stresses on planes perpendicular to the x, y, and z axes and acting
parallel to the y, z, and x axes
shear stress on a plane perpendicular to the x1 axis and acting parallel to
the y1 axis (rotated axes)
shear stress on an inclined plane
allowable stress (or working stress) in shear
ultimate stress in shear; yield stress in shear
angle, angle of twist of a bar in torsion
angle, angle of rotation
angular velocity, angular frequency (v ⫽ 2p f )
★A
star attached to a section number indicates a specialized or advanced topic.
One or more stars attached to a problem number indicate an increasing level of
difficulty in the solution.
Greek Alphabet
〈
〉
⌫
⌬
⌭
⌮
⌯
⌰
⌱
⌲
⌳
⌴
a
b
g
d
e
z
h
u
i
k
l
m
Alpha
Beta
Gamma
Delta
Epsilon
Zeta
Eta
Theta
Iota
Kappa
Lambda
Mu
⌵
⌶
⌷
⌸
⌹
⌺
⌻
⌼
⌽
⌾
⌿
⍀
n
j
o
p
r
s
t
y
f
x
c
v
Nu
Xi
Omicron
Pi
Rho
Sigma
Tau
Upsilon
Phi
Chi
Psi
Omega
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
1
Tension, Compression,
and Shear
1.1 INTRODUCTION TO MECHANICS OF MATERIALS
Mechanics of materials is a branch of applied mechanics that deals
with the behavior of solid bodies subjected to various types of loading.
Other names for this field of study are strength of materials and
mechanics of deformable bodies. The solid bodies considered in this
book include bars with axial loads, shafts in torsion, beams in bending,
and columns in compression.
The principal objective of mechanics of materials is to determine the
stresses, strains, and displacements in structures and their components
due to the loads acting on them. If we can find these quantities for all values of the loads up to the loads that cause failure, we will have a
complete picture of the mechanical behavior of these structures.
An understanding of mechanical behavior is essential for the safe
design of all types of structures, whether airplanes and antennas, buildings and bridges, machines and motors, or ships and spacecraft. That is
why mechanics of materials is a basic subject in so many engineering
fields. Statics and dynamics are also essential, but those subjects deal
primarily with the forces and motions associated with particles and rigid
bodies. In mechanics of materials we go one step further by examining
the stresses and strains inside real bodies, that is, bodies of finite dimensions
that deform under loads. To determine the stresses and strains, we use the
physical properties of the materials as well as numerous theoretical laws
and concepts.
Theoretical analyses and experimental results have equally important
roles in mechanics of materials. We use theories to derive formulas
and equations for predicting mechanical behavior, but these expressions
cannot be used in practical design unless the physical properties of the
materials are known. Such properties are available only after careful
experiments have been carried out in the laboratory. Furthermore, not all
1
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
2
CHAPTER 1 Tension, Compression, and Shear
practical problems are amenable to theoretical analysis alone, and in such
cases physical testing is a necessity.
The historical development of mechanics of materials is a fascinating blend of both theory and experiment—theory has pointed the way to
useful results in some instances, and experiment has done so in others.
Such famous persons as Leonardo da Vinci (1452–1519) and Galileo
Galilei (1564–1642) performed experiments to determine the strength of
wires, bars, and beams, although they did not develop adequate theories
(by today’s standards) to explain their test results. By contrast, the
famous mathematician Leonhard Euler (1707–1783) developed the
mathematical theory of columns and calculated the critical load of a column in 1744, long before any experimental evidence existed to show the
significance of his results. Without appropriate tests to back up his theories, Euler’s results remained unused for over a hundred years, although
today they are the basis for the design and analysis of most columns.*
Problems
When studying mechanics of materials, you will find that your efforts
are divided naturally into two parts: first, understanding the logical
development of the concepts, and second, applying those concepts to
practical situations. The former is accomplished by studying the derivations, discussions, and examples that appear in each chapter, and the
latter is accomplished by solving the problems at the ends of the chapters. Some of the problems are numerical in character, and others are
symbolic (or algebraic).
An advantage of numerical problems is that the magnitudes of all
quantities are evident at every stage of the calculations, thus providing an
opportunity to judge whether the values are reasonable or not. The
principal advantage of symbolic problems is that they lead to
general-purpose formulas. A formula displays the variables that affect
the final results; for instance, a quantity may actually cancel out of the
solution, a fact that would not be evident from a numerical solution. Also,
an algebraic solution shows the manner in which each variable affects the
results, as when one variable appears in the numerator and another
appears in the denominator. Furthermore, a symbolic solution provides
the opportunity to check the dimensions at every stage of the work.
Finally, the most important reason for solving algebraically is to
obtain a general formula that can be used for many different problems. In
contrast, a numerical solution applies to only one set of circumstances.
Because engineers must be adept at both kinds of solutions, you will find a
mixture of numeric and symbolic problems throughout this book.
Numerical problems require that you work with specific units of
measurement. In keeping with current engineering practice, this book uti-
*The history of mechanics of materials, beginning with Leonardo and Galileo, is given in
Refs. 1-1, 1-2, and 1-3.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 1.2 Normal Stress and Strain
3
lizes both the International System of Units (SI) and the U.S. Customary
System (USCS). A discussion of both systems appears in Appendix A,
where you will also find many useful tables, including a table of
conversion factors.
All problems appear at the ends of the chapters, with the problem
numbers and subheadings identifying the sections to which they belong.
In the case of problems requiring numerical solutions, odd-numbered
problems are in USCS units and even-numbered problems are in SI units.
The only exceptions are problems involving commercially available
structural-steel shapes, because the properties of these shapes are tabulated in Appendix E in USCS units only.
The techniques for solving problems are discussed in detail in
Appendix B. In addition to a list of sound engineering procedures,
Appendix B includes sections on dimensional homogeneity and significant digits. These topics are especially important, because every equation
must be dimensionally homogeneous and every numerical result must be
expressed with the proper number of significant digits. In this book, final
numerical results are usually presented with three significant digits when
a number begins with the digits 2 through 9, and with four significant
digits when a number begins with the digit 1. Intermediate values are
often recorded with additional digits to avoid losing numerical accuracy
due to rounding of numbers.
1.2 NORMAL STRESS AND STRAIN
The most fundamental concepts in mechanics of materials are stress and
strain. These concepts can be illustrated in their most elementary form
by considering a prismatic bar subjected to axial forces. A prismatic
bar is a straight structural member having the same cross section
throughout its length, and an axial force is a load directed along the axis
of the member, resulting in either tension or compression in the bar.
Examples are shown in Fig. 1-1, where the tow bar is a prismatic member in tension and the landing gear strut is a member in compression.
Other examples are the members of a bridge truss, connecting rods in
automobile engines, spokes of bicycle wheels, columns in buildings, and
wing struts in small airplanes.
FIG. 1-1 Structural members subjected to
axial loads. (The tow bar is in tension
and the landing gear strut is in
compression.)
Landing gear strut
Tow bar
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
4
CHAPTER 1 Tension, Compression, and Shear
P
P
(a)
L
(b)
m
P
P
n
L+d
(c)
m
P
n
d
P
P
s=—
A
(d)
FIG. 1-2 Prismatic bar in tension:
(a) free-body diagram of a segment of
the bar, (b) segment of the bar before
loading, (c) segment of the bar after
loading, and (d) normal stresses in the
bar
For discussion purposes, we will consider the tow bar of Fig. 1-1
and isolate a segment of it as a free body (Fig. 1-2a). When drawing this
free-body diagram, we disregard the weight of the bar itself and assume
that the only active forces are the axial forces P at the ends. Next we
consider two views of the bar, the first showing the same bar before the
loads are applied (Fig. 1-2b) and the second showing it after the loads
are applied (Fig. 1-2c). Note that the original length of the bar is denoted
by the letter L, and the increase in length due to the loads is denoted by
the Greek letter d (delta).
The internal actions in the bar are exposed if we make an imaginary
cut through the bar at section mn (Fig. 1-2c). Because this section is taken
perpendicular to the longitudinal axis of the bar, it is called a cross section.
We now isolate the part of the bar to the left of cross section mn as a
free body (Fig. 1-2d). At the right-hand end of this free body (section
mn) we show the action of the removed part of the bar (that is, the part
to the right of section mn) upon the part that remains. This action consists of continuously distributed stresses acting over the entire cross
section, and the axial force P acting at the cross section is the resultant
of those stresses. (The resultant force is shown with a dashed line in
Fig. 1-2d.)
Stress has units of force per unit area and is denoted by the Greek
letter s (sigma). In general, the stresses s acting on a plane surface may
be uniform throughout the area or may vary in intensity from one point
to another. Let us assume that the stresses acting on cross section mn
(Fig. 1-2d) are uniformly distributed over the area. Then the resultant of
those stresses must be equal to the magnitude of the stress times the
cross-sectional area A of the bar, that is, P sA. Therefore, we obtain
the following expression for the magnitude of the stresses:
P
s
A
(1-1)
This equation gives the intensity of uniform stress in an axially loaded,
prismatic bar of arbitrary cross-sectional shape.
When the bar is stretched by the forces P, the stresses are tensile
stresses; if the forces are reversed in direction, causing the bar to be
compressed, we obtain compressive stresses. Inasmuch as the stresses
act in a direction perpendicular to the cut surface, they are called normal
stresses. Thus, normal stresses may be either tensile or compressive.
Later, in Section 1.6, we will encounter another type of stress, called
shear stress, that acts parallel to the surface.
When a sign convention for normal stresses is required, it is customary
to define tensile stresses as positive and compressive stresses as negative.
Because the normal stress s is obtained by dividing the axial force
by the cross-sectional area, it has units of force per unit of area. When
USCS units are used, stress is customarily expressed in pounds per
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 1.2 Normal Stress and Strain
5
square inch (psi) or kips per square inch (ksi).* For instance, suppose
that the bar of Fig. 1-2 has a diameter d of 2.0 inches and the load P has
a magnitude of 6 kips. Then the stress in the bar is
P
6k
P
1.91 ksi (or 1910 psi)
s
2
A
pd /4
p (2.0 in.)2/4
In this example the stress is tensile, or positive.
When SI units are used, force is expressed in newtons (N) and area
in square meters (m2). Consequently, stress has units of newtons per
square meter (N/m2), that is, pascals (Pa). However, the pascal is such a
small unit of stress that it is necessary to work with large multiples,
usually the megapascal (MPa).
To demonstrate that a pascal is indeed small, we have only to note
that it takes almost 7000 pascals to make 1 psi.** As an illustration, the
stress in the bar described in the preceding example (1.91 ksi) converts
to 13.2 MPa, which is 13.2 106 pascals. Although it is not recommended in SI, you will sometimes find stress given in newtons per
square millimeter (N/mm2), which is a unit equal to the megapascal
(MPa).
Limitations
b
P
P
FIG. 1-3 Steel eyebar subjected to tensile
loads P
The equation s P/A is valid only if the stress is uniformly distributed
over the cross section of the bar. This condition is realized if the axial
force P acts through the centroid of the cross-sectional area, as demonstrated later in this section. When the load P does not act at the centroid,
bending of the bar will result, and a more complicated analysis is necessary (see Sections 5.12 and 11.5). However, in this book (as in common
practice) it is understood that axial forces are applied at the centroids of
the cross sections unless specifically stated otherwise.
The uniform stress condition pictured in Fig. 1-2d exists throughout
the length of the bar except near the ends. The stress distribution at the
end of a bar depends upon how the load P is transmitted to the bar. If the
load happens to be distributed uniformly over the end, then the stress
pattern at the end will be the same as everywhere else. However, it is
more likely that the load is transmitted through a pin or a bolt, producing
high localized stresses called stress concentrations.
One possibility is illustrated by the eyebar shown in Fig. 1-3. In this
instance the loads P are transmitted to the bar by pins that pass through
the holes (or eyes) at the ends of the bar. Thus, the forces shown in the
figure are actually the resultants of bearing pressures between the pins
and the eyebar, and the stress distribution around the holes is quite complex. However, as we move away from the ends and toward the middle
*One kip, or kilopound, equals 1000 lb.
**Conversion factors between USCS units and SI units are listed in Table A-5, Appendix A.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
6
CHAPTER 1 Tension, Compression, and Shear
of the bar, the stress distribution gradually approaches the uniform distribution pictured in Fig. 1-2d.
As a practical rule, the formula s 5 P/A may be used with good
accuracy at any point within a prismatic bar that is at least as far away
from the stress concentration as the largest lateral dimension of the bar. In
other words, the stress distribution in the steel eyebar of Fig. 1-3 is uniform at distances b or greater from the enlarged ends, where b is the width
of the bar, and the stress distribution in the prismatic bar of Fig. 1-2 is
uniform at distances d or greater from the ends, where d is the diameter of
the bar (Fig. 1-2d). More detailed discussions of stress concentrations produced by axial loads are given in Section 2.10.
Of course, even when the stress is not uniformly distributed, the
equation s P/A may still be useful because it gives the average normal stress on the cross section.
Normal Strain
As already observed, a straight bar will change in length when loaded
axially, becoming longer when in tension and shorter when in compression.
For instance, consider again the prismatic bar of Fig. 1-2. The elongation d
of this bar (Fig. 1-2c) is the cumulative result of the stretching of all
elements of the material throughout the volume of the bar. Let us assume
that the material is the same everywhere in the bar. Then, if we consider half
of the bar (length L/2), it will have an elongation equal to d/2, and if we
consider one-fourth of the bar, it will have an elongation equal to d/4.
In general, the elongation of a segment is equal to its length divided
by the total length L and multiplied by the total elongation d. Therefore, a
unit length of the bar will have an elongation equal to 1/L times d. This
quantity is called the elongation per unit length, or strain, and is denoted
by the Greek letter e (epsilon). We see that strain is given by the equation
d
e 5
L
(1-2)
If the bar is in tension, the strain is called a tensile strain, representing
an elongation or stretching of the material. If the bar is in compression,
the strain is a compressive strain and the bar shortens. Tensile strain is
usually taken as positive and compressive strain as negative. The strain
e is called a normal strain because it is associated with normal stresses.
Because normal strain is the ratio of two lengths, it is a dimensionless quantity, that is, it has no units. Therefore, strain is expressed
simply as a number, independent of any system of units. Numerical values of strain are usually very small, because bars made of structural
materials undergo only small changes in length when loaded.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 1.2 Normal Stress and Strain
7
As an example, consider a steel bar having length L equal to 2.0 m.
When heavily loaded in tension, this bar might elongate by 1.4 mm,
which means that the strain is
d
1.4 mm
e 5 0.0007 700 106
L
2.0 m
In practice, the original units of d and L are sometimes attached to the
strain itself, and then the strain is recorded in forms such as mm/m,
m/m, and in./in. For instance, the strain e in the preceding illustration
could be given as 700 m/m or 700106 in./in. Also, strain is sometimes expressed as a percent, especially when the strains are large. (In
the preceding example, the strain is 0.07%.)
Uniaxial Stress and Strain
P
P
s =P
A
Line of Action of the Axial Forces
for a Uniform Stress Distribution
(a)
y
x
x–
A
dA
p1
O
–y
The definitions of normal stress and normal strain are based upon purely
static and geometric considerations, which means that Eqs. (1-1) and
(1-2) can be used for loads of any magnitude and for any material. The
principal requirement is that the deformation of the bar be uniform
throughout its volume, which in turn requires that the bar be prismatic,
the loads act through the centroids of the cross sections, and the material
be homogeneous (that is, the same throughout all parts of the bar). The
resulting state of stress and strain is called uniaxial stress and strain.
Further discussions of uniaxial stress, including stresses in directions
other than the longitudinal direction of the bar, are given later in Section
2.6. We will also analyze more complicated stress states, such as biaxial
stress and plane stress, in Chapter 7.
y
x
(b)
FIG. 1-4 Uniform stress distribution in a
prismatic bar: (a) axial forces P, and (b)
cross section of the bar
Throughout the preceding discussion of stress and strain in a prismatic
bar, we assumed that the normal stress s was distributed uniformly over
the cross section. Now we will demonstrate that this condition is met if
the line of action of the axial forces is through the centroid of the crosssectional area.
Consider a prismatic bar of arbitrary cross-sectional shape subjected
to axial forces P that produce uniformly distributed stresses s (Fig. 1-4a).
Also, let p1 represent the point in the cross section where the line of
action of the forces intersects the cross section (Fig. 1-4b). We construct
a set of xy axes in the plane of the cross section and denote the coordinates of point p1 by x and y. To determine these coordinates, we observe
that the moments Mx and My of the force P about the x and y axes,
respectively, must be equal to the corresponding moments of the uniformly distributed stresses.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
8
CHAPTER 1 Tension, Compression, and Shear
The moments of the force P are
Mx Py
P
P
s =P
A
(a)
Mx s y dA
x
x–
A
O
–y
y
x
(b)
FIG. 1-4 (Repeated)
My s xdA
(c,d)
These expressions give the moments produced by the stresses s.
Next, we equate the moments Mx and My as obtained from the
force P (Eqs. a and b) to the moments obtained from the distributed
stresses (Eqs. c and d):
dA
p1
(a,b)
in which a moment is considered positive when its vector (using the
right-hand rule) acts in the positive direction of the corresponding axis.*
The moments of the distributed stresses are obtained by integrating
over the cross-sectional area A. The differential force acting on an
element of area dA (Fig. 1-4b) is equal to sdA. The moments of this
elemental force about the x and y axes are sydA and sxdA, respectively,
in which x and y denote the coordinates of the element dA. The total
moments are obtained by integrating over the cross-sectional area:
y
My Px
Px s x dA
Py s y dA
Because the stresses s are uniformly distributed, we know that they are
constant over the cross-sectional area A and can be placed outside the
integral signs. Also, we know that s is equal to P/A. Therefore, we
obtain the following formulas for the coordinates of point p1:
y dA
y
A
x dA
x
A
(1-3a,b)
These equations are the same as the equations defining the coordinates
of the centroid of an area (see Eqs. 12-3a and b in Chapter 12). Therefore, we have now arrived at an important conclusion: In order to have
uniform tension or compression in a prismatic bar, the axial force must
act through the centroid of the cross-sectional area. As explained previously, we always assume that these conditions are met unless it is
specifically stated otherwise.
The following examples illustrate the calculation of stresses and
strains in prismatic bars. In the first example we disregard the weight of
the bar and in the second we include it. (It is customary when solving
textbook problems to omit the weight of the structure unless specifically
instructed to include it.)
*To visualize the right-hand rule, imagine that you grasp an axis of coordinates with your
right hand so that your fingers fold around the axis and your thumb points in the positive
direction of the axis. Then a moment is positive if it acts about the axis in the same direction as your fingers.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 1.2 Normal Stress and Strain
9
Example 1-1
A short post constructed from a hollow circular tube of aluminum supports a
compressive load of 26 kips (Fig. 1-5). The inner and outer diameters of the
tube are d1 4.0 in. and d2 4.5 in., respectively, and its length is 16 in. The
shortening of the post due to the load is measured as 0.012 in.
Determine the compressive stress and strain in the post. (Disregard the
weight of the post itself, and assume that the post does not buckle under the
load.)
26 k
16 in.
FIG. 1-5 Example 1-1. Hollow aluminum
post in compression
Solution
Assuming that the compressive load acts at the center of the hollow tube,
we can use the equation s 5 P/A (Eq. 1-1) to calculate the normal stress. The
force P equals 26 k (or 26,000 lb), and the cross-sectional area A is
p
p
A d 22 d 21 (4.5 in.)2 (4.0 in.)2 3.338 in.2
4
4
Therefore, the compressive stress in the post is
P
26,000 lb
s 2 7790 psi
A
3.338 in.
The compressive strain (from Eq. 1-2) is
d
0.012 in.
e 750 106
16 in.
L
Thus, the stress and strain in the post have been calculated.
Note: As explained earlier, strain is a dimensionless quantity and no units
are needed. For clarity, however, units are often given. In this example, e could
be written as 750 1026 in./in. or 750 in./in.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
10
CHAPTER 1 Tension, Compression, and Shear
Example 1-2
A circular steel rod of length L and diameter d hangs in a mine shaft and holds
an ore bucket of weight W at its lower end (Fig. 1-6).
(a) Obtain a formula for the maximum stress smax in the rod, taking into
account the weight of the rod itself.
(b) Calculate the maximum stress if L 40 m, d 8 mm, and W
1.5 kN.
L
Solution
(a) The maximum axial force Fmax in the rod occurs at the upper end and is
equal to the weight W of the ore bucket plus the weight W0 of the rod itself. The
latter is equal to the weight density g of the steel times the volume V of the rod,
or
d
FIG. 1-6 Example 1-2. Steel rod support-
ing a weight W
(1-4)
W0 gV gAL
W
in which A is the cross-sectional area of the rod. Therefore, the formula for the
maximum stress (from Eq. 1-1) becomes
Fmax
W gAL
W
smax gL
A
A
A
(1-5)
(b) To calculate the maximum stress, we substitute numerical values into
the preceding equation. The cross-sectional area A equals pd 2/4, where d 8
mm, and the weight density g of steel is 77.0 kN/m3 (from Table H-1 in Appendix H). Thus,
1.5 kN
(77.0 kN/m3)(40 m)
smax
p(8 mm)2/4
29.8 MPa 3.1 MPa 32.9 MPa
In this example, the weight of the rod contributes noticeably to the maximum
stress and should not be disregarded.
1.3 MECHANICAL PROPERTIES OF MATERIALS
The design of machines and structures so that they will function properly requires that we understand the mechanical behavior of the
materials being used. Ordinarily, the only way to determine how materials
behave when they are subjected to loads is to perform experiments in
the laboratory. The usual procedure is to place small specimens of the
material in testing machines, apply the loads, and then measure the
resulting deformations (such as changes in length and changes in diameter).
Most materials-testing laboratories are equipped with machines capable
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 1.3 Mechanical Properties of Materials
11
FIG. 1-7 Tensile-test machine with
automatic data-processing system.
(Courtesy of MTS Systems Corporation)
of loading specimens in a variety of ways, including both static and
dynamic loading in tension and compression.
A typical tensile-test machine is shown in Fig. 1-7. The test specimen is installed between the two large grips of the testing machine and
then loaded in tension. Measuring devices record the deformations, and
the automatic control and data-processing systems (at the left in the
photo) tabulate and graph the results.
A more detailed view of a tensile-test specimen is shown in Fig. 1-8
on the next page. The ends of the circular specimen are enlarged where
they fit in the grips so that failure will not occur near the grips themselves. A failure at the ends would not produce the desired information
about the material, because the stress distribution near the grips is not
uniform, as explained in Section 1.2. In a properly designed specimen,
failure will occur in the prismatic portion of the specimen where the
stress distribution is uniform and the bar is subjected only to pure tension. This situation is shown in Fig. 1-8, where the steel specimen has
just fractured under load. The device at the left, which is attached by
two arms to the specimen, is an extensometer that measures the elongation during loading.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
12
CHAPTER 1 Tension, Compression, and Shear
FIG. 1-8 Typical tensile-test specimen
with extensometer attached; the
specimen has just fractured in tension.
(Courtesy of MTS Systems Corporation)
In order that test results will be comparable, the dimensions of test
specimens and the methods of applying loads must be standardized.
One of the major standards organizations in the United States is the
American Society for Testing and Materials (ASTM), a technical society
that publishes specifications and standards for materials and testing.
Other standardizing organizations are the American Standards Association (ASA) and the National Institute of Standards and Technology
(NIST). Similar organizations exist in other countries.
The ASTM standard tension specimen has a diameter of 0.505 in.
and a gage length of 2.0 in. between the gage marks, which are the
points where the extensometer arms are attached to the specimen (see
Fig. 1-8). As the specimen is pulled, the axial load is measured and
recorded, either automatically or by reading from a dial. The elongation
over the gage length is measured simultaneously, either by mechanical
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 1.3 Mechanical Properties of Materials
13
gages of the kind shown in Fig. 1-8 or by electrical-resistance strain
gages.
In a static test, the load is applied slowly and the precise rate of
loading is not of interest because it does not affect the behavior of the
specimen. However, in a dynamic test the load is applied rapidly and
sometimes in a cyclical manner. Since the nature of a dynamic load
affects the properties of the materials, the rate of loading must also be
measured.
Compression tests of metals are customarily made on small specimens in the shape of cubes or circular cylinders. For instance, cubes
may be 2.0 in. on a side, and cylinders may have diameters of 1 in. and
lengths from 1 to 12 in. Both the load applied by the machine and the
shortening of the specimen may be measured. The shortening should be
measured over a gage length that is less than the total length of the specimen in order to eliminate end effects.
Concrete is tested in compression on important construction projects to ensure that the required strength has been obtained. One type of
concrete test specimen is 6 in. in diameter, 12 in. in length, and 28 days
old (the age of concrete is important because concrete gains strength as
it cures). Similar but somewhat smaller specimens are used when performing compression tests of rock (Fig. 1-9, on the next page).
Stress-Strain Diagrams
Test results generally depend upon the dimensions of the specimen being
tested. Since it is unlikely that we will be designing a structure having
parts that are the same size as the test specimens, we need to express
the test results in a form that can be applied to members of any size. A
simple way to achieve this objective is to convert the test results to
stresses and strains.
The axial stress s in a test specimen is calculated by dividing the
axial load P by the cross-sectional area A (Eq. 1-1). When the initial
area of the specimen is used in the calculation, the stress is called the
nominal stress (other names are conventional stress and engineering
stress). A more exact value of the axial stress, called the true stress, can
be calculated by using the actual area of the bar at the cross section
where failure occurs. Since the actual area in a tension test is always less
than the initial area (as illustrated in Fig. 1-8), the true stress is larger
than the nominal stress.
The average axial strain e in the test specimen is found by dividing
the measured elongation d between the gage marks by the gage length L
(see Fig. 1-8 and Eq. 1-2). If the initial gage length is used in the calculation (for instance, 2.0 in.), then the nominal strain is obtained. Since
the distance between the gage marks increases as the tensile load is
applied, we can calculate the true strain (or natural strain) at any value
of the load by using the actual distance between the gage marks. In
tension, true strain is always smaller than nominal strain. However, for
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
14
CHAPTER 1 Tension, Compression, and Shear
FIG. 1-9 Rock sample being tested in
compression. (Courtesy of MTS
Systems Corporation)
most engineering purposes, nominal stress and nominal strain are adequate, as explained later in this section.
After performing a tension or compression test and determining the
stress and strain at various magnitudes of the load, we can plot a diagram of stress versus strain. Such a stress-strain diagram is a
characteristic of the particular material being tested and conveys important information about the mechanical properties and type of behavior.*
*Stress-strain diagrams were originated by Jacob Bernoulli (1654–1705) and J. V. Poncelet (1788–1867); see Ref. 1-4.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
15
SECTION 1.3 Mechanical Properties of Materials
The first material we will discuss is structural steel, also known as
mild steel or low-carbon steel. Structural steel is one of the most widely
used metals and is found in buildings, bridges, cranes, ships, towers,
vehicles, and many other types of construction. A stress-strain diagram
for a typical structural steel in tension is shown in Fig. 1-10. Strains are
plotted on the horizontal axis and stresses on the vertical axis. (In order
to display all of the important features of this material, the strain axis in
Fig. 1-10 is not drawn to scale.)
The diagram begins with a straight line from the origin O to point A,
which means that the relationship between stress and strain in this initial
region is not only linear but also proportional.* Beyond point A, the
proportionality between stress and strain no longer exists; hence the
stress at A is called the proportional limit. For low-carbon steels, this
limit is in the range 30 to 50 ksi (210 to 350 MPa), but high-strength
steels (with higher carbon content plus other alloys) can have proportional limits of more than 80 ksi (550 MPa). The slope of the straight
line from O to A is called the modulus of elasticity. Because the slope
has units of stress divided by strain, modulus of elasticity has the same
units as stress. (Modulus of elasticity is discussed later in Section 1.5.)
With an increase in stress beyond the proportional limit, the strain
begins to increase more rapidly for each increment in stress. Consequently, the stress-strain curve has a smaller and smaller slope, until, at
point B, the curve becomes horizontal (see Fig. 1-10). Beginning at this
point, considerable elongation of the test specimen occurs with no
s
E'
Ultimate
stress
D
Yield stress
B
Proportional
limit
A
C
Fracture
O
FIG. 1-10 Stress-strain diagram for
a typical structural steel in tension
(not to scale)
Linear
region
Perfect
plasticity
or yielding
Strain
hardening
Necking
E
e
*Two variables are said to be proportional if their ratio remains constant. Therefore, a
proportional relationship may be represented by a straight line through the origin. However, a proportional relationship is not the same as a linear relationship. Although a
proportional relationship is linear, the converse is not necessarily true, because a relationship represented by a straight line that does not pass through the origin is linear but
not proportional. The often-used expression “directly proportional” is synonymous with
“proportional” (Ref. 1-5).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
16
CHAPTER 1 Tension, Compression, and Shear
Load
Region
of
necking
Region
of
fracture
Load
FIG. 1-11 Necking of a mild-steel bar in
tension
noticeable increase in the tensile force (from B to C). This phenomenon
is known as yielding of the material, and point B is called the yield
point. The corresponding stress is known as the yield stress of the steel.
In the region from B to C (Fig. 1-10), the material becomes perfectly
plastic, which means that it deforms without an increase in the applied
load. The elongation of a mild-steel specimen in the perfectly plastic
region is typically 10 to 15 times the elongation that occurs in the linear
region (between the onset of loading and the proportional limit). The
presence of very large strains in the plastic region (and beyond) is the
reason for not plotting this diagram to scale.
After undergoing the large strains that occur during yielding in the
region BC, the steel begins to strain harden. During strain hardening,
the material undergoes changes in its crystalline structure, resulting in
increased resistance of the material to further deformation. Elongation of
the test specimen in this region requires an increase in the tensile load,
and therefore the stress-strain diagram has a positive slope from C to D.
The load eventually reaches its maximum value, and the corresponding
stress (at point D) is called the ultimate stress. Further stretching of the
bar is actually accompanied by a reduction in the load, and fracture
finally occurs at a point such as E in Fig. 1-10.
The yield stress and ultimate stress of a material are also called the
yield strength and ultimate strength, respectively. Strength is a general
term that refers to the capacity of a structure to resist loads. For instance,
the yield strength of a beam is the magnitude of the load required to
cause yielding in the beam, and the ultimate strength of a truss is the
maximum load it can support, that is, the failure load. However, when
conducting a tension test of a particular material, we define load-carrying
capacity by the stresses in the specimen rather than by the total loads
acting on the specimen. As a result, the strength of a material is usually
stated as a stress.
When a test specimen is stretched, lateral contraction occurs, as
previously mentioned. The resulting decrease in cross-sectional area is
too small to have a noticeable effect on the calculated values of the
stresses up to about point C in Fig. 1-10, but beyond that point the
reduction in area begins to alter the shape of the curve. In the vicinity of
the ultimate stress, the reduction in area of the bar becomes clearly visible and a pronounced necking of the bar occurs (see Figs. 1-8 and 1-11).
If the actual cross-sectional area at the narrow part of the neck is
used to calculate the stress, the true stress-strain curve (the dashed line
CE⬘ in Fig. 1-10) is obtained. The total load the bar can carry does indeed
diminish after the ultimate stress is reached (as shown by curve DE), but
this reduction is due to the decrease in area of the bar and not to a loss in
strength of the material itself. In reality, the material withstands an
increase in true stress up to failure (point E⬘). Because most structures are
expected to function at stresses below the proportional limit, the conventional stress-strain curve OABCDE, which is based upon the original
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 1.3 Mechanical Properties of Materials
s (ksi)
80
D
60
E
C
40
A, B
20
0
0
0.05 0.10 0.15 0.20 0.25 0.30
e
FIG. 1-12 Stress-strain diagram for a
typical structural steel in tension (drawn
to scale)
s (ksi)
40
30
20
10
0
0
0.05 0.10 0.15 0.20 0.25
e
FIG. 1-13 Typical stress-strain diagram
for an aluminum alloy
s
A
0.002 offset
O
e
FIG. 1-14 Arbitrary yield stress
determined by the offset method
17
cross-sectional area of the specimen and is easy to determine, provides
satisfactory information for use in engineering design.
The diagram of Fig. 1-10 shows the general characteristics of the
stress-strain curve for mild steel, but its proportions are not realistic
because, as already mentioned, the strain that occurs from B to C may be
more than ten times the strain occurring from O to A. Furthermore, the
strains from C to E are many times greater than those from B to C. The
correct relationships are portrayed in Fig. 1-12, which shows a stressstrain diagram for mild steel drawn to scale. In this figure, the strains
from the zero point to point A are so small in comparison to the strains
from point A to point E that they cannot be seen, and the initial part of
the diagram appears to be a vertical line.
The presence of a clearly defined yield point followed by large plastic strains is an important characteristic of structural steel that is
sometimes utilized in practical design (see, for instance, the discussions
of elastoplastic behavior in Sections 2.12 and 6.10). Metals such as
structural steel that undergo large permanent strains before failure are
classified as ductile. For instance, ductility is the property that enables a
bar of steel to be bent into a circular arc or drawn into a wire without
breaking. A desirable feature of ductile materials is that visible distortions occur if the loads become too large, thus providing an opportunity
to take remedial action before an actual fracture occurs. Also, materials
exhibiting ductile behavior are capable of absorbing large amounts of
strain energy prior to fracture.
Structural steel is an alloy of iron containing about 0.2% carbon,
and therefore it is classified as a low-carbon steel. With increasing
carbon content, steel becomes less ductile but stronger (higher yield
stress and higher ultimate stress). The physical properties of steel are
also affected by heat treatment, the presence of other metals, and manufacturing processes such as rolling. Other materials that behave in a
ductile manner (under certain conditions) include aluminum, copper,
magnesium, lead, molybdenum, nickel, brass, bronze, monel metal,
nylon, and teflon.
Although they may have considerable ductility, aluminum alloys
typically do not have a clearly definable yield point, as shown by the
stress-strain diagram of Fig. 1-13. However, they do have an initial
linear region with a recognizable proportional limit. Alloys produced for
structural purposes have proportional limits in the range 10 to 60 ksi
(70 to 410 MPa) and ultimate stresses in the range 20 to 80 ksi (140 to
550 MPa).
When a material such as aluminum does not have an obvious yield
point and yet undergoes large strains after the proportional limit is
exceeded, an arbitrary yield stress may be determined by the offset
method. A straight line is drawn on the stress-strain diagram parallel to
the initial linear part of the curve (Fig. 1-14) but offset by some standard
strain, such as 0.002 (or 0.2%). The intersection of the offset line and
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
18
CHAPTER 1 Tension, Compression, and Shear
s (psi)
3000
2000
Hard rubber
1000
0
Soft rubber
0
2
4
e
6
8
FIG. 1-15 Stress-strain curves for two
kinds of rubber in tension
the stress-strain curve (point A in the figure) defines the yield stress.
Because this stress is determined by an arbitrary rule and is not an
inherent physical property of the material, it should be distinguished
from a true yield stress by referring to it as the offset yield stress. For a
material such as aluminum, the offset yield stress is slightly above the
proportional limit. In the case of structural steel, with its abrupt transition from the linear region to the region of plastic stretching, the offset
stress is essentially the same as both the yield stress and the proportional
limit.
Rubber maintains a linear relationship between stress and strain up to
relatively large strains (as compared to metals). The strain at the proportional limit may be as high as 0.1 or 0.2 (10% or 20%). Beyond the
proportional limit, the behavior depends upon the type of rubber (Fig. 1-15).
Some kinds of soft rubber will stretch enormously without failure, reaching
lengths several times their original lengths. The material eventually offers
increasing resistance to the load, and the stress-strain curve turns
markedly upward. You can easily sense this characteristic behavior by
stretching a rubber band with your hands. (Note that although rubber
exhibits very large strains, it is not a ductile material because the strains
are not permanent. It is, of course, an elastic material; see Section 1.4.)
The ductility of a material in tension can be characterized by its
elongation and by the reduction in area at the cross section where fracture occurs. The percent elongation is defined as follows:
L1 L0
Percent elongation (100)
L0
(1-6)
in which L0 is the original gage length and L1 is the distance between
the gage marks at fracture. Because the elongation is not uniform over
the length of the specimen but is concentrated in the region of necking,
the percent elongation depends upon the gage length. Therefore, when
stating the percent elongation, the gage length should always be given.
For a 2 in. gage length, steel may have an elongation in the range from
3% to 40%, depending upon composition; in the case of structural steel,
values of 20% or 30% are common. The elongation of aluminum alloys
varies from 1% to 45%, depending upon composition and treatment.
The percent reduction in area measures the amount of necking
that occurs and is defined as follows:
A0 A1
Percent reduction in area (100)
A0
(1-7)
in which A0 is the original cross-sectional area and A1 is the final area at
the fracture section. For ductile steels, the reduction is about 50%.
Materials that fail in tension at relatively low values of strain are
classified as brittle. Examples are concrete, stone, cast iron, glass,
ceramics, and a variety of metallic alloys. Brittle materials fail with only
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 1.3 Mechanical Properties of Materials
s
B
A
O
e
FIG. 1-16 Typical stress-strain diagram
for a brittle material showing the proportional limit (point A) and fracture stress
(point B)
19
little elongation after the proportional limit (the stress at point A in
Fig. 1-16) is exceeded. Furthermore, the reduction in area is insignificant, and so the nominal fracture stress (point B) is the same as the true
ultimate stress. High-carbon steels have very high yield stresses—over
100 ksi (700 MPa) in some cases—but they behave in a brittle manner
and fracture occurs at an elongation of only a few percent.
Ordinary glass is a nearly ideal brittle material, because it exhibits
almost no ductility. The stress-strain curve for glass in tension is essentially a straight line, with failure occurring before any yielding takes
place. The ultimate stress is about 10,000 psi (70 MPa) for certain kinds
of plate glass, but great variations exist, depending upon the type of
glass, the size of the specimen, and the presence of microscopic defects.
Glass fibers can develop enormous strengths, and ultimate stresses over
1,000,000 psi (7 GPa) have been attained.
Many types of plastics are used for structural purposes because of
their light weight, resistance to corrosion, and good electrical insulation
properties. Their mechanical properties vary tremendously, with some
plastics being brittle and others ductile. When designing with plastics it
is important to realize that their properties are greatly affected by both
temperature changes and the passage of time. For instance, the ultimate
tensile stress of some plastics is cut in half merely by raising the temperature from 50° F to 120° F. Also, a loaded plastic may stretch gradually
over time until it is no longer serviceable. For example, a bar of
polyvinyl chloride subjected to a tensile load that initially produces a
strain of 0.005 may have that strain doubled after one week, even
though the load remains constant. (This phenomenon, known as creep,
is discussed in the next section.)
Ultimate tensile stresses for plastics are generally in the range 2 to
50 ksi (14 to 350 MPa) and weight densities vary from 50 to 90 lb/ft3 (8
to 14 kN/m3). One type of nylon has an ultimate stress of 12 ksi (80
MPa) and weighs only 70 lb/ft3 (11 kN/m3), which is only 12% heavier
than water. Because of its light weight, the strength-to-weight ratio for
nylon is about the same as for structural steel (see Prob. 1.3-4).
A filament-reinforced material consists of a base material (or
matrix) in which high-strength filaments, fibers, or whiskers are embedded. The resulting composite material has much greater strength than the
base material. As an example, the use of glass fibers can more than double the strength of a plastic matrix. Composites are widely used in
aircraft, boats, rockets, and space vehicles where high strength and light
weight are needed.
Compression
Stress-strain curves for materials in compression differ from those in
tension. Ductile metals such as steel, aluminum, and copper have proportional limits in compression very close to those in tension, and the
initial regions of their compressive and tensile stress-strain diagrams are
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
20
CHAPTER 1 Tension, Compression, and Shear
s (ksi)
80
60
40
20
0
0
0.2
0.4
e
0.6
FIG. 1-17 Stress-strain diagram for
copper in compression
0.8
about the same. However, after yielding begins, the behavior is quite
different. In a tension test, the specimen is stretched, necking may occur,
and fracture ultimately takes place. When the material is compressed, it
bulges outward on the sides and becomes barrel shaped, because friction
between the specimen and the end plates prevents lateral expansion.
With increasing load, the specimen is flattened out and offers greatly
increased resistance to further shortening (which means that the stressstrain curve becomes very steep). These characteristics are illustrated in
Fig. 1-17, which shows a compressive stress-strain diagram for copper.
Since the actual cross-sectional area of a specimen tested in compression is larger than the initial area, the true stress in a compression test is
smaller than the nominal stress.
Brittle materials loaded in compression typically have an initial
linear region followed by a region in which the shortening increases at
a slightly higher rate than does the load. The stress-strain curves for
compression and tension often have similar shapes, but the ultimate
stresses in compression are much higher than those in tension. Also,
unlike ductile materials, which flatten out when compressed, brittle
materials actually break at the maximum load.
Tables of Mechanical Properties
Properties of materials are listed in the tables of Appendix H at the back
of the book. The data in the tables are typical of the materials and are
suitable for solving problems in this book. However, properties of materials and stress-strain curves vary greatly, even for the same material,
because of different manufacturing processes, chemical composition,
internal defects, temperature, and many other factors.
For these reasons, data obtained from Appendix H (or other tables
of a similar nature) should not be used for specific engineering or design
purposes. Instead, the manufacturers or materials suppliers should be
consulted for information about a particular product.
1.4 ELASTICITY, PLASTICITY, AND CREEP
Stress-strain diagrams portray the behavior of engineering materials
when the materials are loaded in tension or compression, as described in
the preceding section. To go one step further, let us now consider what
happens when the load is removed and the material is unloaded.
Assume, for instance, that we apply a load to a tensile specimen so
that the stress and strain go from the origin O to point A on the stressstrain curve of Fig. 1-18a. Suppose further that when the load is
removed, the material follows exactly the same curve back to the origin
O. This property of a material, by which it returns to its original dimen-
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 1.4 Elasticity, Plasticity, and Creep
s
F
E
Lo
ad
Un
in
loa
g
din
g
A
O
e
Elastic
Plastic
(a)
A
F
B
E
Unloa
Lo
ad
ding
in
g
s
C
O
D
e
Elastic
recovery
Residual
strain
(b)
FIG. 1-18 Stress-strain diagrams illustrating (a) elastic behavior, and (b) partially
elastic behavior
21
sions during unloading, is called elasticity, and the material itself is said
to be elastic. Note that the stress-strain curve from O to A need not be
linear in order for the material to be elastic.
Now suppose that we load this same material to a higher level, so
that point B is reached on the stress-strain curve (Fig. 1-18b). When
unloading occurs from point B, the material follows line BC on the diagram. This unloading line is parallel to the initial portion of the loading
curve; that is, line BC is parallel to a tangent to the stress-strain curve at
the origin. When point C is reached, the load has been entirely removed,
but a residual strain, or permanent strain, represented by line OC,
remains in the material. As a consequence, the bar being tested is longer
than it was before loading. This residual elongation of the bar is called
the permanent set. Of the total strain OD developed during loading
from O to B, the strain CD has been recovered elastically and the strain
OC remains as a permanent strain. Thus, during unloading the bar
returns partially to its original shape, and so the material is said to be
partially elastic.
Between points A and B on the stress-strain curve (Fig. 1-18b), there
must be a point before which the material is elastic and beyond which
the material is partially elastic. To find this point, we load the material
to some selected value of stress and then remove the load. If there is no
permanent set (that is, if the elongation of the bar returns to zero), then
the material is fully elastic up to the selected value of the stress.
The process of loading and unloading can be repeated for successively higher values of stress. Eventually, a stress will be reached such
that not all the strain is recovered during unloading. By this procedure, it
is possible to determine the stress at the upper limit of the elastic region,
for instance, the stress at point E in Figs. 1-18a and b. The stress at this
point is known as the elastic limit of the material.
Many materials, including most metals, have linear regions at the
beginning of their stress-strain curves (for example, see Figs. 1-10 and
1-13). The stress at the upper limit of this linear region is the proportional limit, as explained in the preceeding section. The elastic limit is
usually the same as, or slightly above, the proportional limit. Hence, for
many materials the two limits are assigned the same numerical value. In
the case of mild steel, the yield stress is also very close to the proportional limit, so that for practical purposes the yield stress, the elastic
limit, and the proportional limit are assumed to be equal. Of course, this
situation does not hold for all materials. Rubber is an outstanding example of a material that is elastic far beyond the proportional limit.
The characteristic of a material by which it undergoes inelastic
strains beyond the strain at the elastic limit is known as plasticity. Thus,
on the stress-strain curve of Fig. 1-18a, we have an elastic region followed by a plastic region. When large deformations occur in a ductile
material loaded into the plastic region, the material is said to undergo
plastic flow.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
22
CHAPTER 1 Tension, Compression, and Shear
A
E
Reloading of a Material
F
B
Lo
ad
Unloa
ding
in
g
s
C
O
D
e
Elastic
recovery
Residual
strain
(b)
FIG. 1-18b (Repeated)
s
F
B
ding
Reloa
Unloa
Lo
ad
ding
in
g
E
O
e
C
FIG. 1-19 Reloading of a material and
raising of the elastic and proportional
limits
Elongation
d0
O
P
(a)
t0
Time
(b)
FIG. 1-20 Creep in a bar under constant
load
If the material remains within the elastic range, it can be loaded,
unloaded, and loaded again without significantly changing the behavior.
However, when loaded into the plastic range, the internal structure of
the material is altered and its properties change. For instance, we have
already observed that a permanent strain exists in the specimen after
unloading from the plastic region (Fig. 1-18b). Now suppose that the
material is reloaded after such an unloading (Fig. 1-19). The new loading begins at point C on the diagram and continues upward to point B,
the point at which unloading began during the first loading cycle. The
material then follows the original stress-strain curve toward point F.
Thus, for the second loading, we can imagine that we have a new stressstrain diagram with its origin at point C.
During the second loading, the material behaves in a linearly elastic
manner from C to B, with the slope of line CB being the same as the slope
of the tangent to the original loading curve at the origin O. The proportional limit is now at point B, which is at a higher stress than the original
elastic limit (point E). Thus, by stretching a material such as steel or aluminum into the inelastic or plastic range, the properties of the material
are changed—the linearly elastic region is increased, the proportional
limit is raised, and the elastic limit is raised. However, the ductility is
reduced because in the “new material” the amount of yielding beyond
the elastic limit (from B to F ) is less than in the original material (from
E to F ).*
Creep
The stress-strain diagrams described previously were obtained from
tension tests involving static loading and unloading of the specimens,
and the passage of time did not enter our discussions. However, when
loaded for long periods of time, some materials develop additional
strains and are said to creep.
This phenomenon can manifest itself in a variety of ways. For
instance, suppose that a vertical bar (Fig. 1-20a) is loaded slowly by a
force P, producing an elongation equal to d0. Let us assume that the loading and corresponding elongation take place during a time interval of
duration t0 (Fig. 1-20b). Subsequent to time t0, the load remains constant.
However, due to creep, the bar may gradually lengthen, as shown in Fig.
1-20b, even though the load does not change. This behavior occurs with
many materials, although sometimes the change is too small to be of
concern.
*The study of material behavior under various environmental and loading conditions is
an important branch of applied mechanics. For more detailed engineering information
about materials, consult a textbook devoted solely to this subject.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 1.5 Linear Elasticity, Hooke’s Law, and Poisson’s Ratio
Wire
(a)
Stress
s0
O
t0
Time
(b)
FIG. 1-21 Relaxation of stress in a wire
under constant strain
23
As another manifestation of creep, consider a wire that is stretched
between two immovable supports so that it has an initial tensile stress s0
(Fig. 1-21). Again, we will denote the time during which the wire is
initially stretched as t0. With the elapse of time, the stress in the wire
gradually diminishes, eventually reaching a constant value, even though
the supports at the ends of the wire do not move. This process, is called
relaxation of the material.
Creep is usually more important at high temperatures than at
ordinary temperatures, and therefore it should always be considered in
the design of engines, furnaces, and other structures that operate at
elevated temperatures for long periods of time. However, materials such
as steel, concrete, and wood will creep slightly even at atmospheric
temperatures. For example, creep of concrete over long periods of time
can create undulations in bridge decks because of sagging between the
supports. (One remedy is to construct the deck with an upward camber,
which is an initial displacement above the horizontal, so that when creep
occurs, the spans lower to the level position.)
1.5 LINEAR ELASTICITY, HOOKE’S LAW, AND POISSON’S RATIO
Many structural materials, including most metals, wood, plastics, and
ceramics, behave both elastically and linearly when first loaded.
Consequently, their stress-strain curves begin with a straight line passing
through the origin. An example is the stress-strain curve for structural
steel (Fig. 1-10), where the region from the origin O to the proportional
limit (point A) is both linear and elastic. Other examples are the regions
below both the proportional limits and the elastic limits on the diagrams
for aluminum (Fig. 1-13), brittle materials (Fig. 1-16), and copper
(Fig. 1-17).
When a material behaves elastically and also exhibits a linear
relationship between stress and strain, it is said to be linearly elastic.
This type of behavior is extremely important in engineering for an obvious reason—by designing structures and machines to function in this
region, we avoid permanent deformations due to yielding.
Hooke’s Law
The linear relationship between stress and strain for a bar in simple
tension or compression is expressed by the equation
s Ee
(1-8)
in which s is the axial stress, e is the axial strain, and E is a constant of
proportionality known as the modulus of elasticity for the material. The
modulus of elasticity is the slope of the stress-strain diagram in the
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
24
CHAPTER 1 Tension, Compression, and Shear
linearly elastic region, as mentioned previously in Section 1.3. Since
strain is dimensionless, the units of E are the same as the units of stress.
Typical units of E are psi or ksi in USCS units and pascals (or multiples
thereof) in SI units.
The equation s Ee is commonly known as Hooke’s law, named
for the famous English scientist Robert Hooke (1635–1703). Hooke was
the first person to investigate scientifically the elastic properties of
materials, and he tested such diverse materials as metal, wood, stone,
bone, and sinew. He measured the stretching of long wires supporting
weights and observed that the elongations “always bear the same
proportions one to the other that the weights do that made them” (Ref.
1-6). Thus, Hooke established the linear relationship between the
applied loads and the resulting elongations.
Equation (1-8) is actually a very limited version of Hooke’s law
because it relates only to the longitudinal stresses and strains developed
in simple tension or compression of a bar (uniaxial stress). To deal with
more complicated states of stress, such as those found in most structures
and machines, we must use more extensive equations of Hooke’s law
(see Sections 7.5 and 7.6).
The modulus of elasticity has relatively large values for materials
that are very stiff, such as structural metals. Steel has a modulus of
approximately 30,000 ksi (210 GPa); for aluminum, values around
10,600 ksi (73 GPa) are typical. More flexible materials have a lower
modulus—values for plastics range from 100 to 2,000 ksi (0.7 to
14 GPa). Some representative values of E are listed in Table H-2,
Appendix H. For most materials, the value of E in compression is nearly
the same as in tension.
Modulus of elasticity is often called Young’s modulus, after
another English scientist, Thomas Young (1773–1829). In connection
with an investigation of tension and compression of prismatic bars,
Young introduced the idea of a “modulus of the elasticity.” However,
his modulus was not the same as the one in use today, because it
involved properties of the bar as well as of the material (Ref. 1-7).
Poisson’s Ratio
(a)
P
P
(b)
FIG. 1-22 Axial elongation and lateral
contraction of a prismatic bar in tension:
(a) bar before loading, and (b) bar after
loading. (The deformations of the bar
are highly exaggerated.)
When a prismatic bar is loaded in tension, the axial elongation is
accompanied by lateral contraction (that is, contraction normal to the
direction of the applied load). This change in shape is pictured in Fig. 1-22,
where part (a) shows the bar before loading and part (b) shows it after loading. In part (b), the dashed lines represent the shape of the bar prior to
loading.
Lateral contraction is easily seen by stretching a rubber band, but in
metals the changes in lateral dimensions (in the linearly elastic region)
are usually too small to be visible. However, they can be detected with
sensitive measuring devices.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 1.5 Linear Elasticity, Hooke’s Law, and Poisson’s Ratio
25
The lateral strain e⬘ at any point in a bar is proportional to the axial
strain e at that same point if the material is linearly elastic. The ratio of
these strains is a property of the material known as Poisson’s ratio. This
dimensionless ratio, usually denoted by the Greek letter n (nu), can be
expressed by the equation
lateral strain
e
n
axial strain
e
(1-9)
The minus sign is inserted in the equation to compensate for the fact that
the lateral and axial strains normally have opposite signs. For instance,
the axial strain in a bar in tension is positive and the lateral strain is
negative (because the width of the bar decreases). For compression we
have the opposite situation, with the bar becoming shorter (negative
axial strain) and wider (positive lateral strain). Therefore, for ordinary
materials Poisson’s ratio will have a positive value.
When Poisson’s ratio for a material is known, we can obtain the
lateral strain from the axial strain as follows:
e ne
(1-10)
When using Eqs. (1-9) and (1-10), we must always keep in mind that
they apply only to a bar in uniaxial stress, that is, a bar for which the
only stress is the normal stress s in the axial direction.
Poisson’s ratio is named for the famous French mathematician
Siméon Denis Poisson (1781–1840), who attempted to calculate this
ratio by a molecular theory of materials (Ref. 1-8). For isotropic
materials, Poisson found n 1/4. More recent calculations based upon
better models of atomic structure give n 1/3. Both of these values are
close to actual measured values, which are in the range 0.25 to 0.35 for
most metals and many other materials. Materials with an extremely low
value of Poisson’s ratio include cork, for which n is practically zero, and
concrete, for which n is about 0.1 or 0.2. A theoretical upper limit for
Poisson’s ratio is 0.5, as explained later in Section 7.5. Rubber comes
close to this limiting value.
A table of Poisson’s ratios for various materials in the linearly elastic
range is given in Appendix H (see Table H-2). For most purposes, Poisson’s ratio is assumed to be the same in both tension and compression.
When the strains in a material become large, Poisson’s ratio
changes. For instance, in the case of structural steel the ratio becomes
almost 0.5 when plastic yielding occurs. Thus, Poisson’s ratio remains
constant only in the linearly elastic range. When the material behavior
is nonlinear, the ratio of lateral strain to axial strain is often called the
contraction ratio. Of course, in the special case of linearly elastic behavior, the contraction ratio is the same as Poisson’s ratio.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
26
CHAPTER 1 Tension, Compression, and Shear
Limitations
(a)
P
P
(b)
FIG. 1-22 (Repeated)
For a particular material, Poisson’s ratio remains constant throughout
the linearly elastic range, as explained previously. Therefore, at any
given point in the prismatic bar of Fig. 1-22, the lateral strain remains
proportional to the axial strain as the load increases or decreases.
However, for a given value of the load (which means that the axial
strain is constant throughout the bar), additional conditions must be met
if the lateral strains are to be the same throughout the entire bar.
First, the material must be homogeneous, that is, it must have the
same composition (and hence the same elastic properties) at every point.
However, having a homogeneous material does not mean that the elastic
properties at a particular point are the same in all directions. For
instance, the modulus of elasticity could be different in the axial and
lateral directions, as in the case of a wood pole. Therefore, a second
condition for uniformity in the lateral strains is that the elastic properties
must be the same in all directions perpendicular to the longitudinal axis.
When the preceding conditions are met, as is often the case with metals,
the lateral strains in a prismatic bar subjected to uniform tension will be
the same at every point in the bar and the same in all lateral directions.
Materials having the same properties in all directions (whether
axial, lateral, or any other direction) are said to be isotropic. If the
properties differ in various directions, the material is anisotropic (or
aeolotropic).
In this book, all examples and problems are solved with the assumption that the material is linearly elastic, homogeneous, and isotropic,
unless a specific statement is made to the contrary.
Example 1-3
A steel pipe of length L 4.0 ft, outside diameter d2 6.0 in., and inside
diameter d1 4.5 in. is compressed by an axial force P 140 k (Fig. 1-23). The
material has modulus of elasticity E 30,000 ksi and Poisson’s ratio n 0.30.
Determine the following quantities for the pipe: (a) the shortening d, (b) the
lateral strain e, (c) the increase d2 in the outer diameter and the increase Dd1
in the inner diameter, and (d) the increase t in the wall thickness.
Solution
The cross-sectional area A and longitudinal stress s are determined as
follows:
p
p
A d 22 d 21 (6.0 in.)2 (4.5 in.)2 12.37 in.2
4
4
P
140 k
s 2 11.32 ksi (compression)
A
12.37 in.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 1.5 Linear Elasticity, Hooke’s Law, and Poisson’s Ratio
27
Because the stress is well below the yield stress (see Table H-3, Appendix H),
the material behaves linearly elastically and the axial strain may be found from
Hooke’s law:
P
s
11.32 ksi
e 377.3 106
E
30, 000 ksi
L
The minus sign for the strain indicates that the pipe shortens.
(a) Knowing the axial strain, we can now find the change in length of the
pipe (see Eq. 1-2):
d eL (377.3 106)(4.0 ft)(12 in./ft) 0.018 in.
d1
d2
FIG. 1-23 Example 1-3. Steel pipe in
compression
The negative sign again indicates a shortening of the pipe.
(b) The lateral strain is obtained from Poisson’s ratio (see Eq. 1-10):
e9 2ne 2(0.30)(377.3 106) 113.2 106
The positive sign for e9 indicates an increase in the lateral dimensions, as
expected for compression.
(c) The increase in outer diameter equals the lateral strain times the diameter:
d2 e9d2(113.2 106)(6.0 in.) 0.000679 in.
Similarly, the increase in inner diameter is
d1 e9d1 (113.2 106)(4.5 in.) 0.000509 in.
(d) The increase in wall thickness is found in the same manner as the
increases in the diameters; thus,
t e9t (113.2106)(0.75 in.) 0.000085 in.
This result can be verified by noting that the increase in wall thickness is equal
to half the difference of the increases in diameters:
d2 d 1
1
t (0.000679 in. 0.000509 in.) 0.000085 in.
2
2
as expected. Note that under compression, all three quantities increase (outer
diameter, inner diameter, and thickness).
Note: The numerical results obtained in this example illustrate that the
dimensional changes in structural materials under normal loading conditions
are extremely small. In spite of their smallness, changes in dimensions
can be important in certain kinds of analysis (such as the analysis of statically
indeterminate structures) and in the experimental determination of stresses and
strains.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
28
CHAPTER 1 Tension, Compression, and Shear
1.6 SHEAR STRESS AND STRAIN
In the preceding sections we discussed the effects of normal stresses
produced by axial loads acting on straight bars. These stresses are called
“normal stresses” because they act in directions perpendicular to the
surface of the material. Now we will consider another kind of stress,
called a shear stress, that acts tangential to the surface of the material.
As an illustration of the action of shear stresses, consider the bolted
connection shown in Fig. 1-24a. This connection consists of a flat bar A,
a clevis C, and a bolt B that passes through holes in the bar and clevis.
Under the action of the tensile loads P, the bar and clevis will press
against the bolt in bearing, and contact stresses, called bearing
stresses, will be developed. In addition, the bar and clevis tend to shear
the bolt, that is, cut through it, and this tendency is resisted by shear
stresses in the bolt.
To show more clearly the actions of the bearing and shear stresses,
let us look at this type of connection in a schematic side view (Fig.
1-24b). With this view in mind, we draw a free-body diagram of the bolt
(Fig. 1-24c). The bearing stresses exerted by the clevis against the bolt
appear on the left-hand side of the free-body diagram and are labeled 1
and 3. The stresses from the bar appear on the right-hand side and are
P
B
C
A
P
(a)
P
m
p
n
P
q
1
m
p
3
n
2
q
m
V
n
p
q
2
t
m
n
V
(b)
(c)
(d)
FIG. 1-24 Bolted connection in which the
bolt is loaded in double shear
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(e)
SECTION 1.6 Shear Stress and Strain
29
labeled 2. The actual distribution of the bearing stresses is difficult to
determine, so it is customary to assume that the stresses are uniformly
distributed. Based upon the assumption of uniform distribution, we can
calculate an average bearing stress sb by dividing the total bearing
force Fb by the bearing area Ab:
Fb
sb
Ab
(1-11)
The bearing area is defined as the projected area of the curved bearing
surface. For instance, consider the bearing stresses labeled 1. The projected area Ab on which they act is a rectangle having a height equal to
the thickness of the clevis and a width equal to the diameter of the bolt.
Also, the bearing force Fb represented by the stresses labeled 1 is equal
to P/2. The same area and the same force apply to the stresses labeled 3.
Now consider the bearing stresses between the flat bar and the bolt
(the stresses labeled 2). For these stresses, the bearing area Ab is a rectangle with height equal to the thickness of the flat bar and width equal
to the bolt diameter. The corresponding bearing force Fb is equal to the
load P.
The free-body diagram of Fig. 1-24c shows that there is a tendency
to shear the bolt along cross sections mn and pq. From a free-body
diagram of the portion mnpq of the bolt (see Fig. 1-24d), we see that
shear forces V act over the cut surfaces of the bolt. In this particular
example there are two planes of shear (mn and pq), and so the bolt is
said to be in double shear. In double shear, each of the shear forces is
equal to one-half of the total load transmitted by the bolt, that is, V
P/2.
The shear forces V are the resultants of the shear stresses distributed
over the cross-sectional area of the bolt. For instance, the shear stresses
acting on cross section mn are shown in Fig. 1-24e. These stresses act
parallel to the cut surface. The exact distribution of the stresses is not
known, but they are highest near the center and become zero at certain
locations on the edges. As indicated in Fig. 1-24e, shear stresses are customarily denoted by the Greek letter t (tau).
A bolted connection in single shear is shown in Fig. 1-25a, on the
next page, where the axial force P in the metal bar is transmitted to the
flange of the steel column through a bolt. A cross-sectional view of the
column (Fig. 1-25b) shows the connection in more detail. Also, a sketch
of the bolt (Fig. 1-25c) shows the assumed distribution of the bearing
stresses acting on the bolt. As mentioned earlier, the actual distribution of
these bearing stresses is much more complex than shown in the figure.
Furthermore, bearing stresses are also developed against the inside surfaces of the bolt head and nut. Thus, Fig. 1-25c is not a free-body
diagram—only the idealized bearing stresses acting on the shank of the
bolt are shown in the figure.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
30
CHAPTER 1 Tension, Compression, and Shear
P
(a)
P
m
FIG. 1-25 Bolted connection in which the
bolt is loaded in single shear
n
(b)
(c)
m
V
n
(d)
By cutting through the bolt at section mn we obtain the diagram
shown in Fig. 1-25d. This diagram includes the shear force V (equal to
the load P) acting on the cross section of the bolt. As already pointed
out, this shear force is the resultant of the shear stresses that act over the
cross-sectional area of the bolt.
The deformation of a bolt loaded almost to fracture in single shear
is shown in Fig. 1-26 (compare with Fig. 1-25c).
In the preceding discussions of bolted connections we disregarded
friction (produced by tightening of the bolts) between the connecting
elements. The presence of friction means that part of the load is carried
by friction forces, thereby reducing the loads on the bolts. Since friction
forces are unreliable and difficult to estimate, it is common practice to
err on the conservative side and omit them from the calculations.
The average shear stress on the cross section of a bolt is obtained
by dividing the total shear force V by the area A of the cross section on
which it acts, as follows:
V
taver
A
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(1-12)
SECTION 1.6 Shear Stress and Strain
31
Load
FIG. 1-26 Failure of a bolt in single shear
In the example of Fig. 1-25, which shows a bolt in single shear, the
shear force V is equal to the load P and the area A is the cross-sectional
area of the bolt. However, in the example of Fig. 1-24, where the bolt is
in double shear, the shear force V equals P/2.
From Eq. (1-12) we see that shear stresses, like normal stresses,
represent intensity of force, or force per unit of area. Thus, the units of
shear stress are the same as those for normal stress, namely, psi or ksi in
USCS units and pascals or multiples thereof in SI units.
The loading arrangements shown in Figs. 1-24 and 1-25 are examples
of direct shear (or simple shear) in which the shear stresses are created by
the direct action of the forces in trying to cut through the material. Direct
shear arises in the design of bolts, pins, rivets, keys, welds, and glued
joints.
Shear stresses also arise in an indirect manner when members are
subjected to tension, torsion, and bending, as discussed later in Sections
2.6, 3.3, and 5.8, respectively.
y
c
Load
a
t2
Equality of Shear Stresses on Perpendicular Planes
b
t1
x
z
FIG. 1-27 Small element of material
subjected to shear stresses
To obtain a more complete picture of the action of shear stresses, let us
consider a small element of material in the form of a rectangular parallelepiped having sides of lengths a, b, and c in the x, y, and z directions,
respectively (Fig. 1-27).* The front and rear faces of this element are
free of stress.
Now assume that a shear stress t1 is distributed uniformly over the
right-hand face, which has area bc. In order for the element to be in
equilibrium in the y direction, the total shear force t1bc acting on the
right-hand face must be balanced by an equal but oppositely directed
*A parallelepiped is a prism whose bases are parallelograms; thus, a parallelepiped has
six faces, each of which is a parallelogram. Opposite faces are parallel and identical parallelograms. A rectangular parallelepiped has all faces in the form of rectangles.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
32
CHAPTER 1 Tension, Compression, and Shear
y
a
c
t2
b
t1
x
z
shear force on the left-hand face. Since the areas of these two faces are
equal, it follows that the shear stresses on the two faces must be equal.
The forces t1bc acting on the left- and right-hand side faces (Fig.
1-27) form a couple having a moment about the z axis of magnitude
t1abc, acting counterclockwise in the figure.* Equilibrium of the element requires that this moment be balanced by an equal and opposite
moment resulting from shear stresses acting on the top and bottom faces
of the element. Denoting the stresses on the top and bottom faces as t2,
we see that the corresponding horizontal shear forces equal t2ac. These
forces form a clockwise couple of moment t2abc. From moment
equilibrium of the element about the z axis, we see that t1abc equals
t2abc, or
FIG. 1-27 (Repeated)
t1 t2
Therefore, the magnitudes of the four shear stresses acting on the element are equal, as shown in Fig. 1-28a.
In summary, we have arrived at the following general observations
regarding shear stresses acting on a rectangular element:
y
1. Shear stresses on opposite (and parallel) faces of an element are
equal in magnitude and opposite in direction.
2. Shear stresses on adjacent (and perpendicular) faces of an element
are equal in magnitude and have directions such that both stresses
point toward, or both point away from, the line of intersection of the
faces.
a
c
t
p
q
b
t
x
z
s
r
(a)
g
2 p
(1-13)
t
q
t
gr
2
s
p –g
2
p +g
2
These observations were obtained for an element subjected only to shear
stresses (no normal stresses), as pictured in Figs. 1-27 and 1-28. This
state of stress is called pure shear and is discussed later in greater detail
(Section 3.5).
For most purposes, the preceding conclusions remain valid even
when normal stresses act on the faces of the element. The reason is that
the normal stresses on opposite faces of a small element usually are
equal in magnitude and opposite in direction; hence they do not alter the
equilibrium equations used in reaching the preceding conclusions.
Shear Strain
Shear stresses acting on an element of material (Fig. 1-28a) are accompanied by shear strains. As an aid in visualizing these strains, we note
that the shear stresses have no tendency to elongate or shorten the
(b)
FIG. 1-28 Element of material subjected
to shear stresses and strains
*A couple consists of two parallel forces that are equal in magnitude and opposite in
direction.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 1.6 Shear Stress and Strain
33
element in the x, y, and z directions—in other words, the lengths of the
sides of the element do not change. Instead, the shear stresses produce a
change in the shape of the element (Fig. 1-28b). The original element,
which is a rectangular parallelepiped, is deformed into an oblique parallelepiped, and the front and rear faces become rhomboids.*
Because of this deformation, the angles between the side faces
change. For instance, the angles at points q and s, which were p/2 before
deformation, are reduced by a small angle g to p/2 g (Fig. 1-28b). At
the same time, the angles at points p and r are increased to p/2 g. The
angle g is a measure of the distortion, or change in shape, of the element
and is called the shear strain. Because shear strain is an angle, it is
usually measured in degrees or radians.
Sign Conventions for Shear Stresses and Strains
As an aid in establishing sign conventions for shear stresses and strains,
we need a scheme for identifying the various faces of a stress element
(Fig. 1-28a). Henceforth, we will refer to the faces oriented toward the
positive directions of the axes as the positive faces of the element. In
other words, a positive face has its outward normal directed in the positive direction of a coordinate axis. The opposite faces are negative faces.
Thus, in Fig. 1-28a, the right-hand, top, and front faces are the positive
x, y, and z faces, respectively, and the opposite faces are the negative x,
y, and z faces.
Using the terminology described in the preceding paragraph, we may
state the sign convention for shear stresses in the following manner:
A shear stress acting on a positive face of an element is positive if it acts in
the positive direction of one of the coordinate axes and negative if it acts in
the negative direction of an axis. A shear stress acting on a negative face of
an element is positive if it acts in the negative direction of an axis and negative if it acts in a positive direction.
Thus, all shear stresses shown in Fig. 1-28a are positive.
The sign convention for shear strains is as follows:
Shear strain in an element is positive when the angle between two positive
faces (or two negative faces) is reduced. The strain is negative when the
angle between two positive (or two negative) faces is increased.
Thus, the strains shown in Fig. 1-28b are positive, and we see that positive shear stresses are accompanied by positive shear strains.
*An oblique angle can be either acute or obtuse, but it is not a right angle. A rhomboid
is a parallelogram with oblique angles and adjacent sides not equal. (A rhombus is a parallelogram with oblique angles and all four sides equal, sometimes called a
diamond-shaped figure.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
34
CHAPTER 1 Tension, Compression, and Shear
Hooke’s Law in Shear
The properties of a material in shear can be determined experimentally
from direct-shear tests or from torsion tests. The latter tests are performed by twisting hollow, circular tubes, thereby producing a state of
pure shear, as explained later in Section 3.5. From the results of these
tests, we can plot shear stress-strain diagrams (that is, diagrams of
shear stress t versus shear strain g). These diagrams are similar in shape
to tension-test diagrams (s versus e) for the same materials, although
they differ in magnitudes.
From shear stress-strain diagrams, we can obtain material properties
such as the proportional limit, modulus of elasticity, yield stress, and
ultimate stress. These properties in shear are usually about half as large
as those in tension. For instance, the yield stress for structural steel in
shear is 0.5 to 0.6 times the yield stress in tension.
For many materials, the initial part of the shear stress-strain diagram
is a straight line through the origin, just as it is in tension. For this linearly elastic region, the shear stress and shear strain are proportional,
and therefore we have the following equation for Hooke’s law in shear:
t Gg
(1-14)
in which G is the shear modulus of elasticity (also called the modulus
of rigidity).
The shear modulus G has the same units as the tension modulus E,
namely, psi or ksi in USCS units and pascals (or multiples thereof) in SI
units. For mild steel, typical values of G are 11,000 ksi or 75 GPa; for
aluminum alloys, typical values are 4000 ksi or 28 GPa. Additional values
are listed in Table H-2, Appendix H.
The moduli of elasticity in tension and shear are related by the
following equation:
E
G
2(1 n)
(1-15)
in which n is Poisson’s ratio. This relationship, which is derived later in
Section 3.6, shows that E, G, and n are not independent elastic properties of the material. Because the value of Poisson’s ratio for ordinary
materials is between zero and one-half, we see from Eq. (1-15) that G
must be from one-third to one-half of E.
The following examples illustrate some typical analyses involving
the effects of shear. Example 1-4 is concerned with shear stresses in a
plate, Example 1-5 deals with bearing and shear stresses in pins and
bolts, and Example 1-6 involves finding shear stresses and shear strains
in an elastomeric bearing pad subjected to a horizontal shear force.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 1.6 Shear Stress and Strain
35
Example 1-4
A punch for making holes in steel plates is shown in Fig. 1-29a. Assume that a
punch having diameter d 20 mm is used to punch a hole in an 8-mm plate, as
shown in the cross-sectional view (Fig. 1-29b).
If a force P 110 kN is required to create the hole, what is the average
shear stress in the plate and the average compressive stress in the punch?
P
P = 110 kN
d = 20 mm
t = 8.0 mm
FIG. 1-29 Example 1-4. Punching a hole
in a steel plate
(a)
(b)
Solution
The average shear stress in the plate is obtained by dividing the force P by
the shear area of the plate. The shear area As is equal to the circumference of the
hole times the thickness of the plate, or
As pdt p(20 mm)(8.0 mm) 502.7 mm2
in which d is the diameter of the punch and t is the thickness of the plate. Therefore, the average shear stress in the plate is
P
110 kN
taver 2 219 MPa
As
502.7 mm
The average compressive stress in the punch is
P
P
110 kN
350 MPa
sc
Apunch
pd 2/4
p (20 mm)2/4
in which Apunch is the cross-sectional area of the punch.
Note: This analysis is highly idealized because we are disregarding impact
effects that occur when a punch is rammed through a plate. (The inclusion of
such effects requires advanced methods of analysis that are beyond the scope of
mechanics of materials.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
36
CHAPTER 1 Tension, Compression, and Shear
Example 1-5
A steel strut S serving as a brace for a boat hoist transmits a compressive force
P 12 k to the deck of a pier (Fig. 1-30a). The strut has a hollow square cross
section with wall thickness t 0.375 in. (Fig. 1-30b), and the angle u between
the strut and the horizontal is 40°. A pin through the strut transmits the
compressive force from the strut to two gussets G that are welded to the base
plate B. Four anchor bolts fasten the base plate to the deck.
The diameter of the pin is dpin 0.75 in., the thickness of the gussets is
tG 0.625 in., the thickness of the base plate is tB 0.375 in., and the diameter
of the anchor bolts is dbolt 0.50 in.
Determine the following stresses: (a) the bearing stress between the strut
and the pin, (b) the shear stress in the pin, (c) the bearing stress between the pin
and the gussets, (d) the bearing stress between the anchor bolts and the base
plate, and (e) the shear stress in the anchor bolts. (Disregard any friction
between the base plate and the deck.)
P
u = 40°
S
Pin
G
S
G
G
B
t
FIG. 1-30 Example 1-5. (a) Pin connec-
tion between strut S and base plate B.
(b) Cross section through the strut S.
(a)
(b)
Solution
(a) Bearing stress between strut and pin. The average value of the bearing
stress between the strut and the pin is found by dividing the force in the strut by
the total bearing area of the strut against the pin. The latter is equal to twice the
thickness of the strut (because bearing occurs at two locations) times the diameter of the pin (see Fig. 1-30b). Thus, the bearing stress is
P
12 k
sb1 21.3 ksi
2tdpin
This bearing stress is not excessive for a strut made of structural steel.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 1.6 Shear Stress and Strain
37
(b) Shear stress in pin. As can be seen from Fig. 1-30b, the pin tends to
shear on two planes, namely, the planes between the strut and the gussets.
Therefore, the average shear stress in the pin (which is in double shear) is equal
to the total load applied to the pin divided by twice its cross-sectional area:
P
12 k
13.6 ksi
tpin
2
2pd pin/4
2p(0.75 in.)2/4
The pin would normally be made of high-strength steel (tensile yield stress
greater than 50 ksi) and could easily withstand this shear stress (the yield stress
in shear is usually at least 50% of the yield stress in tension).
(c) Bearing stress between pin and gussets. The pin bears against the gussets at two locations, so the bearing area is twice the thickness of the gussets
times the pin diameter; thus,
P
12 k
sb2 12.8 ksi
2tG d pin
which is less than the bearing stress between the strut and the pin (21.3 ksi).
(d) Bearing stress between anchor bolts and base plate. The vertical component of the force P (see Fig. 1-30a) is transmitted to the pier by direct bearing
between the base plate and the pier. The horizontal component, however, is
transmitted through the anchor bolts. The average bearing stress between the
base plate and the anchor bolts is equal to the horizontal component of the force
P divided by the bearing area of four bolts. The bearing area for one bolt is
equal to the thickness of the base plate times the bolt diameter. Consequently,
the bearing stress is
P cos 40°
(12 k)(cos 40°)
sb3 12.3 ksi
4tB dbolt
(e) Shear stress in anchor bolts. The average shear stress in the anchor
bolts is equal to the horizontal component of the force P divided by the total
cross-sectional area of four bolts (note that each bolt is in single shear). Therefore,
P cos 40°
(12 k)(cos 40°)
11.7 ksi
tbolt 2
4p d bolt/4
4p (0.50 in.)2/4
Any friction between the base plate and the pier would reduce the load on the
anchor bolts.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
38
CHAPTER 1 Tension, Compression, and Shear
Example 1-6
A bearing pad of the kind used to support machines and bridge girders consists
of a linearly elastic material (usually an elastomer, such as rubber) capped by a
steel plate (Fig. 1-31a). Assume that the thickness of the elastomer is h, the
dimensions of the plate are a b, and the pad is subjected to a horizontal shear
force V.
Obtain formulas for the average shear stress taver in the elastomer and the
horizontal displacement d of the plate (Fig. 1-31b).
a
b
d
V
V
g
h
h
a
FIG. 1-31 Example 1-6. Bearing pad in
(a)
shear
(b)
Solution
Assume that the shear stresses in the elastomer are uniformly distributed
throughout its entire volume. Then the shear stress on any horizontal plane
through the elastomer equals the shear force V divided by the area ab of the
plane (Fig. 1-31a):
V
(1-16)
taver
ab
The corresponding shear strain (from Hooke’s law in shear; Eq. 1-14) is
taver
V
g
abGe
Ge
(1-17)
in which Ge is the shear modulus of the elastomeric material. Finally, the horizontal displacement d is equal to h tan g (from Fig. 1-31b):
V
d h tan g h tan
abGe
(1-18)
In most practical situations the shear strain g is a small angle, and in such cases
we may replace tan g by g and obtain
hV
d hg
abGe
(1-19)
Equations (1-18) and (1-19) give approximate results for the horizontal displacement of the plate because they are based upon the assumption that the
shear stress and strain are constant throughout the volume of the elastomeric
material. In reality the shear stress is zero at the edges of the material (because
there are no shear stresses on the free vertical faces), and therefore the deformation of the material is more complex than pictured in Fig. 1-31b. However, if
the length a of the plate is large compared with the thickness h of the elastomer,
the preceding results are satisfactory for design purposes.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 1.7 Allowable Stresses and Allowable Loads
39
1.7 ALLOWABLE STRESSES AND ALLOWABLE LOADS
Engineering has been aptly described as the application of science to the
common purposes of life. In fulfilling that mission, engineers design a
seemingly endless variety of objects to serve the basic needs of society.
These needs include housing, agriculture, transportation, communication, and many other aspects of modern life. Factors to be considered in
design include functionality, strength, appearance, economics, and environmental effects. However, when studying mechanics of materials, our
principal design interest is strength, that is, the capacity of the object to
support or transmit loads. Objects that must sustain loads include
buildings, machines, containers, trucks, aircraft, ships, and the like. For
simplicity, we will refer to all such objects as structures; thus, a structure is any object that must support or transmit loads.
Factors of Safety
If structural failure is to be avoided, the loads that a structure is capable
of supporting must be greater than the loads it will be subjected to when
in service. Since strength is the ability of a structure to resist loads, the
preceding criterion can be restated as follows: The actual strength of a
structure must exceed the required strength. The ratio of the actual
strength to the required strength is called the factor of safety n:
Actual strength
Factor of safety n
Required strength
(1-20)
Of course, the factor of safety must be greater than 1.0 if failure is to be
avoided. Depending upon the circumstances, factors of safety from
slightly above 1.0 to as much as 10 are used.
The incorporation of factors of safety into design is not a simple
matter, because both strength and failure have many different meanings.
Strength may be measured by the load-carrying capacity of a structure,
or it may be measured by the stress in the material. Failure may mean
the fracture and complete collapse of a structure, or it may mean that the
deformations have become so large that the structure can no longer perform its intended functions. The latter kind of failure may occur at loads
much smaller than those that cause actual collapse.
The determination of a factor of safety must also take into account
such matters as the following: probability of accidental overloading of
the structure by loads that exceed the design loads; types of loads (static
or dynamic); whether the loads are applied once or are repeated; how
accurately the loads are known; possibilities for fatigue failure; inaccuracies in construction; variability in the quality of workmanship;
variations in properties of materials; deterioration due to corrosion or
other environmental effects; accuracy of the methods of analysis;
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
40
CHAPTER 1 Tension, Compression, and Shear
whether failure is gradual (ample warning) or sudden (no warning); consequences of failure (minor damage or major catastrophe); and other
such considerations. If the factor of safety is too low, the likelihood of
failure will be high and the structure will be unacceptable; if the factor is
too large, the structure will be wasteful of materials and perhaps unsuitable for its function (for instance, it may be too heavy).
Because of these complexities and uncertainties, factors of safety
must be determined on a probabilistic basis. They usually are
established by groups of experienced engineers who write the codes and
specifications used by other designers, and sometimes they are even
enacted into law. The provisions of codes and specifications are
intended to provide reasonable levels of safety without unreasonable
costs.
In aircraft design it is customary to speak of the margin of safety
rather than the factor of safety. The margin of safety is defined as the
factor of safety minus one:
Margin of safety n 1
(1-21)
Margin of safety is often expressed as a percent, in which case the value
given above is multiplied by 100. Thus, a structure having an actual
strength that is 1.75 times the required strength has a factor of safety of
1.75 and a margin of safety of 0.75 (or 75%). When the margin of safety
is reduced to zero or less, the structure (presumably) will fail.
Allowable Stresses
Factors of safety are defined and implemented in various ways. For
many structures, it is important that the material remain within the
linearly elastic range in order to avoid permanent deformations when the
loads are removed. Under these conditions, the factor of safety is
established with respect to yielding of the structure. Yielding begins
when the yield stress is reached at any point within the structure. Therefore, by applying a factor of safety with respect to the yield stress (or
yield strength), we obtain an allowable stress (or working stress) that
must not be exceeded anywhere in the structure. Thus,
Yield strength
Allowable stress
Factor of safety
(1-22)
or, for tension and shear, respectively,
sY
sallow
n1
and
tY
tallow
n2
(1-23a,b)
in which sY and tY are the yield stresses and n1 and n2 are the
corresponding factors of safety. In building design, a typical factor of
safety with respect to yielding in tension is 1.67; thus, a mild steel
having a yield stress of 36 ksi has an allowable stress of 21.6 ksi.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 1.7 Allowable Stresses and Allowable Loads
41
Sometimes the factor of safety is applied to the ultimate stress
instead of the yield stress. This method is suitable for brittle materials,
such as concrete and some plastics, and for materials without a clearly
defined yield stress, such as wood and high-strength steels. In these
cases the allowable stresses in tension and shear are
sU
tU
sallow and tallow
n3
n4
(1-24a,b)
in which sU and tU are the ultimate stresses (or ultimate strengths).
Factors of safety with respect to the ultimate strength of a material are
usually larger than those based upon yield strength. In the case of mild
steel, a factor of safety of 1.67 with respect to yielding corresponds to a
factor of approximately 2.8 with respect to the ultimate strength.
Allowable Loads
After the allowable stress has been established for a particular material
and structure, the allowable load on that structure can be determined.
The relationship between the allowable load and the allowable stress
depends upon the type of structure. In this chapter we are concerned
only with the most elementary kinds of structures, namely, bars in
tension or compression and pins (or bolts) in direct shear and bearing.
In these kinds of structures the stresses are uniformly distributed (or
at least assumed to be uniformly distributed) over an area. For instance,
in the case of a bar in tension, the stress is uniformly distributed over the
cross-sectional area provided the resultant axial force acts through the
centroid of the cross section. The same is true of a bar in compression
provided the bar is not subject to buckling. In the case of a pin subjected
to shear, we consider only the average shear stress on the cross section,
which is equivalent to assuming that the shear stress is uniformly
distributed. Similarly, we consider only an average value of the bearing
stress acting on the projected area of the pin.
Therefore, in all four of the preceding cases the allowable load
(also called the permissible load or the safe load) is equal to the
allowable stress times the area over which it acts:
Allowable load (Allowable stress)(Area)
(1-25)
For bars in direct tension and compression (no buckling), this equation becomes
Pallow sallow A
(1-26)
in which sallow is the permissible normal stress and A is the crosssectional area of the bar. If the bar has a hole through it, the net area is
normally used when the bar is in tension. The net area is the gross
cross-sectional area minus the area removed by the hole. For compression,
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
42
CHAPTER 1 Tension, Compression, and Shear
the gross area may be used if the hole is filled by a bolt or pin that can
transmit the compressive stresses.
For pins in direct shear, Eq. (1-25) becomes
Pallow tallow A
(1-27)
in which tallow is the permissible shear stress and A is the area over which
the shear stresses act. If the pin is in single shear, the area is the crosssectional area of the pin; in double shear, it is twice the cross-sectional
area.
Finally, the permissible load based upon bearing is
Pallow sbAb
(1-28)
in which sb is the allowable bearing stress and Ab is the projected area
of the pin or other surface over which the bearing stresses act.
The following example illustrates how allowable loads are determined when the allowable stresses for the material are known.
Example 1-7
A steel bar serving as a vertical hanger to support heavy machinery in a factory
is attached to a support by the bolted connection shown in Fig. 1-32. The main
part of the hanger has a rectangular cross section with width b1 1.5 in. and
thickness t 0.5 in. At the connection the hanger is enlarged to a width b2
3.0 in. The bolt, which transfers the load from the hanger to the two gussets, has
diameter d 1.0 in.
Determine the allowable value of the tensile load P in the hanger based
upon the following four considerations:
(a) The allowable tensile stress in the main part of the hanger is 16,000 psi.
(b) The allowable tensile stress in the hanger at its cross section through the
bolt hole is 11,000 psi. (The permissible stress at this section is lower because
of the stress concentrations around the hole.)
(c) The allowable bearing stress between the hanger and the bolt is 26,000 psi.
(d) The allowable shear stress in the bolt is 6,500 psi.
Solution
(a) The allowable load P1 based upon the stress in the main part of the
hanger is equal to the allowable stress in tension times the cross-sectional area
of the hanger (Eq. 1-26):
P1 sallowA sallowb1t (16,000 psi)(1.5 in. 0.5 in.) 12,000 lb
A load greater than this value will overstress the main part of the hanger, that is,
the actual stress will exceed the allowable stress, thereby reducing the factor of
safety.
(b) At the cross section of the hanger through the bolt hole, we must make
a similar calculation but with a different allowable stress and a different area.
The net cross-sectional area, that is, the area that remains after the hole is drilled
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 1.7 Allowable Stresses and Allowable Loads
43
through the bar, is equal to the net width times the thickness. The net width is
equal to the gross width b2 minus the diameter d of the hole. Thus, the equation
for the allowable load P2 at this section is
P2 sallow A sallow(b2 d)t (11,000 psi)(3.0 in. 1.0 in.)(0.5 in.)
11,000 lb
(c) The allowable load based upon bearing between the hanger and the bolt
is equal to the allowable bearing stress times the bearing area. The bearing area
is the projection of the actual contact area, which is equal to the bolt diameter
times the thickness of the hanger. Therefore, the allowable load (Eq. 1-28) is
P3 sb A sbdt (26,000 psi)(1.0 in.)(0.5 in.) 13,000 lb
(d) Finally, the allowable load P4 based upon shear in the bolt is equal to
the allowable shear stress times the shear area (Eq. 1-27). The shear area is
twice the area of the bolt because the bolt is in double shear; thus:
P4 tallow A tallow(2)(pd 2/4) (6,500 psi)(2)(p)(1.0 in.)2/4 10,200 lb
We have now found the allowable tensile loads in the hanger based upon all
four of the given conditions.
Comparing the four preceding results, we see that the smallest value of the
load is
Pallow 10,200 lb
This load, which is based upon shear in the bolt, is the allowable tensile load in
the hanger.
b2 = 3.0 in.
d = 1.0 in.
Bolt
Washer
Gusset
Hanger
t = 0.5 in.
b1 = 1.5 in.
FIG. 1-32 Example 1-7. Vertical hanger
subjected to a tensile load P: (a) front
view of bolted connection, and (b) side
view of connection
P
(a)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
P
(b)
44
CHAPTER 1 Tension, Compression, and Shear
1.8 DESIGN FOR AXIAL LOADS AND DIRECT SHEAR
In the preceding section we discussed the determination of allowable
loads for simple structures, and in earlier sections we saw how to find
the stresses, strains, and deformations of bars. The determination of such
quantities is known as analysis. In the context of mechanics of materials, analysis consists of determining the response of a structure to loads,
temperature changes, and other physical actions. By the response of a
structure, we mean the stresses, strains, and deformations produced by
the loads.
Response also refers to the load-carrying capacity of a structure; for
instance, the allowable load on a structure is a form of response.
A structure is said to be known (or given) when we have a complete
physical description of the structure, that is, when we know all of its
properties. The properties of a structure include the types of members
and how they are arranged, the dimensions of all members, the types of
supports and where they are located, the materials used, and the properties
of the materials. Thus, when analyzing a structure, the properties are
given and the response is to be determined.
The inverse process is called design. When designing a structure,
we must determine the properties of the structure in order that the structure will support the loads and perform its intended functions. For
instance, a common design problem in engineering is to determine the
size of a member to support given loads. Designing a structure is usually
a much lengthier and more difficult process than analyzing it—indeed,
analyzing a structure, often more than once, is typically part of the
design process.
In this section we will deal with design in its most elementary form
by calculating the required sizes of simple tension and compression
members as well as pins and bolts loaded in shear. In these cases the
design process is quite straightforward. Knowing the loads to be transmitted and the allowable stresses in the materials, we can calculate the
required areas of members from the following general relationship
(compare with Eq. 1-25):
Load to be transmitted
Required area
(1-29)
This equation can be applied to any structure in which the stresses are
uniformly distributed over the area. (The use of this equation for finding
the size of a bar in tension and the size of a pin in shear is illustrated in
Example 1-8, which follows.)
In addition to strength considerations, as exemplified by Eq. (1-29),
the design of a structure is likely to involve stiffness and stability. Stiffness refers to the ability of the structure to resist changes in shape (for
instance, to resist stretching, bending, or twisting), and stability refers to
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 1.8 Design for Axial Loads and Direct Shear
45
the ability of the structure to resist buckling under compressive stresses.
Limitations on stiffness are sometimes necessary to prevent excessive
deformations, such as large deflections of a beam that might interfere with
its performance. Buckling is the principal consideration in the design of
columns, which are slender compression members (Chapter 11).
Another part of the design process is optimization, which is the
task of designing the best structure to meet a particular goal, such as
minimum weight. For instance, there may be many structures that will
support a given load, but in some circumstances the best structure will
be the lightest one. Of course, a goal such as minimum weight usually
must be balanced against more general considerations, including the
aesthetic, economic, environmental, political, and technical aspects of
the particular design project.
When analyzing or designing a structure, we refer to the forces that
act on it as either loads or reactions. Loads are active forces that are
applied to the structure by some external cause, such as gravity, water
pressure, wind, amd earthquake ground motion. Reactions are passive
forces that are induced at the supports of the structure—their magnitudes and directions are determined by the nature of the structure itself.
Thus, reactions must be calculated as part of the analysis, whereas loads
are known in advance.
Example 1-8, on the following pages, begins with a review of freebody diagrams and elementary statics and concludes with the design of
a bar in tension and a pin in direct shear.
When drawing free-body diagrams, it is helpful to distinguish reactions from loads or other applied forces. A common scheme is to place a
slash, or slanted line, across the arrow when it represents a reactive
force, as illustrated in Fig. 1-34 of the example.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
46
CHAPTER 1 Tension, Compression, and Shear
Example 1-8
The two-bar truss ABC shown in Fig. 1-33 has pin supports at points A and C,
which are 2.0 m apart. Members AB and BC are steel bars, pin connected at
joint B. The length of bar BC is 3.0 m. A sign weighing 5.4 kN is suspended
from bar BC at points D and E, which are located 0.8 m and 0.4 m, respectively,
from the ends of the bar.
Determine the required cross-sectional area of bar AB and the required
diameter of the pin at support C if the allowable stresses in tension and shear are
125 MPa and 45 MPa, respectively. (Note: The pins at the supports are in double shear. Also, disregard the weights of members AB and BC.)
A
2.0 m
C
D
E
0.9 m
B
0.9 m
0.8 m
W = 5.4 kN
0.4 m
FIG. 1-33 Example 1-8. Two-bar truss
ABC supporting a sign of weight W
Solution
The objectives of this example are to determine the required sizes of bar
AB and the pin at support C. As a preliminary matter, we must determine the
tensile force in the bar and the shear force acting on the pin. These quantities are
found from free-body diagrams and equations of equilibrium.
Reactions. We begin with a free-body diagram of the entire truss (Fig. 1-34a).
On this diagram we show all forces acting on the truss—namely, the loads from
the weight of the sign and the reactive forces exerted by the pin supports at A
and C. Each reaction is shown by its horizontal and vertical components, with
the resultant reaction shown by a dashed line. (Note the use of slashes across the
arrows to distinguish reactions from loads.)
The horizontal component RAH of the reaction at support A is obtained by
summing moments about point C, as follows (counterclockwise moments are
positive):
MC 0
RAH (2.0 m) (2.7 kN)(0.8 m) (2.7 kN)(2.6 m) 0
Solving this equation, we get
RAH 4.590 kN
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
47
SECTION 1.8 Design for Axial Loads and Direct Shear
RA
RAV
A
RAH
2.0 m
FAB
RCH
C
RC
RCV
D
0.8 m
E
B
RCH
C
RC
RCV
1.8 m
2.7 kN
D
E
0.8 m
2.7 kN 0.4 m
B
1.8 m
2.7 kN 0.4 m
2.7 kN
(b)
(a)
FIG. 1-34 Free-body diagrams for
Example 1-8
Next, we sum forces in the horizontal direction and obtain
Fhoriz 0
RCH RAH 4.590 kN
To obtain the vertical component of the reaction at support C, we may use
a free-body diagram of member BC, as shown in Fig. 1-34b. Summing moments
about joint B gives the desired reaction component:
MB 0
RCV (3.0 m) (2.7 kN)(2.2 m) (2.7 kN)(0.4 m) 0
RCV 2.340 kN
Now we return to the free-body diagram of the entire truss (Fig. 1-34a) and
sum forces in the vertical direction to obtain the vertical component RAV of the
reaction at A:
Fvert 0
RAV RCV 2.7 kN 2.7 kN 0
RAV 3.060 kN
As a partial check on these results, we note that the ratio RAV /RAH of the forces
acting at point A is equal to the ratio of the vertical and horizontal components
of line AB, namely, 2.0 m/3.0 m, or 2/3.
Knowing the horizontal and vertical components of the reaction at A, we
can find the reaction itself (Fig. 1-34a):
RA
2
2
(R
(RA
AH)
V) 5.516 kN
Similarly, the reaction at point C is obtained from its componets RCH and RCV,
as follows:
RC
2
2
(R
(RC
CH)
V) 5.152 kN
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
48
CHAPTER 1 Tension, Compression, and Shear
Tensile force in bar AB. Because we are disregarding the weight of bar AB,
the tensile force FAB in this bar is equal to the reaction at A (see Fig.1-34):
FAB RA 5.516 kN
Shear force acting on the pin at C. This shear force is equal to the reaction
RC (see Fig. 1-34); therefore,
VC RC 5.152 kN
Thus, we have now found the tensile force FAB in bar AB and the shear force VC
acting on the pin at C.
Required area of bar. The required cross-sectional area of bar AB is calculated by dividing the tensile force by the allowable stress, inasmuch as the stress
is uniformly distributed over the cross section (see Eq. 1-29):
FAB
5.516 kN
AAB 44.1 mm2
sallow
125 MPa
Bar AB must be designed with a cross-sectional area equal to or greater than
44.1 mm2 in order to support the weight of the sign, which is the only load we
considered. When other loads are included in the calculations, the required area
will be larger.
Required diameter of pin. The required cross-sectional area of the pin at C,
which is in double shear, is
VC
5.152 kN
Apin 57.2 mm2
2tallow
2(45 MPa)
from which we can calculate the required diameter:
dpin
4Apin /p 8.54 mm
A pin of at least this diameter is needed to support the weight of the sign without exceeding the allowable shear stress.
Notes: In this example we intentionally omitted the weight of the truss
from the calculations. However, once the sizes of the members are known, their
weights can be calculated and included in the free-body diagrams of Fig. 1-34.
When the weights of the bars are included, the design of member AB
becomes more complicated, because it is no longer a bar in simple tension.
Instead, it is a beam subjected to bending as well as tension. An analogous situation exists for member BC. Not only because of its own weight but also
because of the weight of the sign, member BC is subjected to both bending and
compression. The design of such members must wait until we study stresses in
beams (Chapter 5).
In practice, other loads besides the weights of the truss and sign would
have to be considered before making a final decision about the sizes of the bars
and pins. Loads that could be important include wind loads, earthquake loads,
and the weights of objects that might have to be supported temporarily by the
truss and sign.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 1 Problems
49
PROBLEMS CHAPTER 1
Normal Stress and Strain
1.2-3 A steel rod 110 ft long hangs inside a tall tower and
1.2-1 A solid circular post ABC (see figure) supports a load
holds a 200-pound weight at its lower end (see figure).
If the diameter of the circular rod is 1/4 inch, calculate
the maximum normal stress smax in the rod, taking into
account the weight of the rod itself. (Obtain the weight
density of steel from Table H-1, Appendix H.)
P1 2500 lb acting at the top. A second load P2 is
uniformly distributed around the shelf at B. The diameters
of the upper and lower parts of the post are dAB 1.25 in.
and dBC 2.25 in., respectively.
(a) Calculate the normal stress sAB in the upper part of
the post.
(b) If it is desired that the lower part of the post have
the same compressive stress as the upper part, what should
be the magnitude of the load P2?
P1
A
110 ft
1
— in.
4
dAB
P2
B
200 lb
dBC
PROB. 1.2-3
C
PROB. 1.2-1
1.2-2 Calculate the compressive stress sc in the circular
piston rod (see figure) when a force P 40 N is applied to
the brake pedal.
Assume that the line of action of the force P is parallel
to the piston rod, which has diameter 5 mm. Also, the other
dimensions shown in the figure (50 mm and 225 mm) are
measured perpendicular to the line of action of the force P.
50 mm
1.2-4 A circular aluminum tube of length L 400 mm is
loaded in compression by forces P (see figure). The outside and inside diameters are 60 mm and 50 mm,
respectively. A strain gage is placed on the outside of the
bar to measure normal strains in the longitudinal direction.
(a) If the measured strain is e 550 106, what is
the shortening d of the bar?
(b) If the compressive stress in the bar is intended to
be 40 MPa, what should be the load P?
5 mm
Strain gage
225 mm
P = 40 N
L = 400 mm
Piston rod
PROB. 1.2-2
P
P
PROB. 1.2-4
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
50
CHAPTER 1 Tension, Compression, and Shear
1.2-5 The cross section of a concrete pier that is loaded
uniformly in compression is shown in the figure.
(a) Determine the average compressive stress sc in the
concrete if the load is equal to 2500 k.
(b) Determine the coordinates x and y of the point
where the resultant load must act in order to produce uniform normal stress.
C
A
y
a
20 in.
b
B
16 in.
16 in.
48 in.
PROB. 1.2-7
16 in.
O
20 in.
x
16 in.
PROB. 1.2-5
1.2-6 A car weighing 130 kN when fully loaded is pulled
slowly up a steep inclined track by a steel cable (see figure).
The cable has an effective cross-sectional area of 490 mm2,
and the angle a of the incline is 30°.
Calculate the tensile stress st in the cable.
Cable
yy
;;
;;
yy
1.2-8 A long retaining wall is braced by wood shores set at
an angle of 30° and supported by concrete thrust blocks, as
shown in the first part of the figure. The shores are evenly
spaced, 3 m apart.
For analysis purposes, the wall and shores are idealized as shown in the second part of the figure. Note that the
base of the wall and both ends of the shores are assumed to
be pinned. The pressure of the soil against the wall is
assumed to be triangularly distributed, and the resultant
force acting on a 3-meter length of the wall is F 190 kN.
If each shore has a 150 mm 150 mm square cross
section, what is the compressive stress sc in the shores?
Soil
Retaining
wall
Concrete
Shore thrust
block
30°
a
B
F
30°
1.5 m
A
C
0.5 m
4.0 m
PROB. 1.2-8
PROB. 1.2-6
1.2-7 Two steel wires, AB and BC, support a lamp weighing 18 lb (see figure). Wire AB is at an angle a 34° to the
horizontal and wire BC is at an angle b 48°. Both wires
have diameter 30 mils. (Wire diameters are often expressed
in mils; one mil equals 0.001 in.)
Determine the tensile stresses sAB and sBC in the two
wires.
1.2-9 A loading crane consisting of a steel girder ABC supported by a cable BD is subjected to a load P (see the figure
on the next page). The cable has an effective crosssectional area A 0.471 in2. The dimensions of the crane
are H 9 ft, L1 12 ft, and L2 4 ft.
(a) If the load P 9000 lb, what is the average tensile
stress in the cable?
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 1 Problems
(b) If the cable stretches by 0.382 in., what is the average strain?
D
Cable
1.2-12 A round bar ACB of length 2L (see figure) rotates
about an axis through the midpoint C with constant angular
speed v (radians per second). The material of the bar has
weight density g.
(a) Derive a formula for the tensile stress sx in the bar
as a function of the distance x from the midpoint C.
(b) What is the maximum tensile stress smax?
★
v
H
A
C
L
Girder
B
A
51
B
x
L
C
PROB. 1.2-12
L1
L2
Mechanical Properties and Stress-Strain Diagrams
P
PROBS. 1.2-9 and 1.2-10
1.2-10 Solve the preceding problem if the load P 32 kN;
the cable has effective cross-sectional area A 481 mm2;
the dimensions of the crane are H 1.6 m, L1 3.0 m,
and L2 1.5 m; and the cable stretches by 5.1 mm.
★
1.2-11 A reinforced concrete slab 8.0 ft square and 9.0
in. thick is lifted by four cables attached to the corners, as
shown in the figure. The cables are attached to a hook at a
point 5.0 ft above the top of the slab. Each cable has an
effective cross-sectional area A 0.12 in2.
Determine the tensile stress st in the cables due to the
weight of the concrete slab. (See Table H-1, Appendix H,
for the weight density of reinforced concrete.)
Cables
ÀÀÀÀ
@@@@
;;;;
1.3-1 Imagine that a long steel wire hangs vertically from a
high-altitude balloon.
(a) What is the greatest length (feet) it can have without yielding if the steel yields at 40 ksi?
(b) If the same wire hangs from a ship at sea, what is
the greatest length? (Obtain the weight densities of steel
and sea water from Table H-1, Appendix H.)
1.3-2 Imagine that a long wire of tungsten hangs vertically
from a high-altitude balloon.
(a) What is the greatest length (meters) it can have
without breaking if the ultimate strength (or breaking
strength) is 1500 MPa?
(b) If the same wire hangs from a ship at sea, what is
the greatest length? (Obtain the weight densities of tungsten
and sea water from Table H-1, Appendix H.)
1.3-3 Three different materials, designated A, B, and C, are
tested in tension using test specimens having diameters of
0.505 in. and gage lengths of 2.0 in. (see figure). At failure,
the distances between the gage marks are found to be 2.13,
2.48, and 2.78 in., respectively. Also, at the failure cross
sections the diameters are found to be 0.484, 0.398, and
0.253 in., respectively.
Determine the percent elongation and percent reduction in area of each specimen, and then, using your own
judgment, classify each material as brittle or ductile.
P
P
Reinforced
concrete slab
PROB. 1.2-11
Gage
length
PROB. 1.3-3
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
52
CHAPTER 1 Tension, Compression, and Shear
ÀÀÀ
@@@
;;;
1.3-4 The strength-to-weight ratio of a structural material
is defined as its load-carrying capacity divided by its
weight. For materials in tension, we may use a characteristic tensile stress (as obtained from a stress-strain curve) as a
measure of strength. For instance, either the yield stress or
the ultimate stress could be used, depending upon the
particular application. Thus, the strength-to-weight ratio
RS/W for a material in tension is defined as
initial part of the stress-strain curve), and yield stress at
0.2% offset. Is the material ductile or brittle?
s
RS/W
g
STRESS-STRAIN DATA FOR PROBLEM 1.3-6
in which s is the characteristic stress and g is the weight
density. Note that the ratio has units of length.
Using the ultimate stress sU as the strength parameter,
calculate the strength-to-weight ratio (in units of meters)
for each of the following materials: aluminum alloy
6061-T6, Douglas fir (in bending), nylon, structural steel
ASTM-A572, and a titanium alloy. (Obtain the material
properties from Tables H-1 and H-3 of Appendix H. When
a range of values is given in a table, use the average value.)
1.3-5 A symmetrical framework consisting of three pinconnected bars is loaded by a force P (see figure). The
angle between the inclined bars and the horizontal is a
48°. The axial strain in the middle bar is measured as
0.0713.
Determine the tensile stress in the outer bars if they
are constructed of aluminum alloy having the stress-strain
diagram shown in Fig. 1-13. (Express the stress in USCS
units.)
A
B
C
a
D
P
PROB. 1.3-5
1.3-6 A specimen of a methacrylate plastic is tested in tension at room temperature (see figure), producing the
stress-strain data listed in the accompanying table.
Plot the stress-strain curve and determine the proportional limit, modulus of elasticity (i.e., the slope of the
P
P
PROB. 1.3-6
Stress (MPa)
Strain
8.0
17.5
25.6
31.1
39.8
44.0
48.2
53.9
58.1
62.0
62.1
0.0032
0.0073
0.0111
0.0129
0.0163
0.0184
0.0209
0.0260
0.0331
0.0429
Fracture
1.3-7 The data shown in the accompanying table were
obtained from a tensile test of high-strength steel. The test
specimen had a diameter of 0.505 in. and a gage length of
2.00 in. (see figure for Prob. 1.3-3). At fracture, the elongation between the gage marks was 0.12 in. and the minimum
diameter was 0.42 in.
Plot the conventional stress-strain curve for the steel
and determine the proportional limit, modulus of elasticity
(i.e., the slope of the initial part of the stress-strain curve),
yield stress at 0.1% offset, ultimate stress, percent elongation in 2.00 in., and percent reduction in area.
★
TENSILE-TEST DATA FOR PROBLEM 1.3-7
Load (lb)
Elongation (in.)
1,000
2,000
6,000
10,000
12,000
12,900
13,400
13,600
13,800
14,000
14,400
15,200
16,800
18,400
20,000
22,400
22,600
0.0002
0.0006
0.0019
0.0033
0.0039
0.0043
0.0047
0.0054
0.0063
0.0090
0.0102
0.0130
0.0230
0.0336
0.0507
0.1108
Fracture
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 1 Problems
Elasticity and Plasticity
1.4-1 A bar made of structural steel having the stress-strain
diagram shown in the figure has a length of 48 in. The yield
stress of the steel is 42 ksi and the slope of the initial linear
part of the stress-strain curve (modulus of elasticity) is
30 103 ksi. The bar is loaded axially until it elongates
0.20 in., and then the load is removed.
How does the final length of the bar compare with
its original length? (Hint: Use the concepts illustrated in
Fig. 1-18b.)
s (ksi)
60
40
53
shown in Fig. 1-13 of Section 1.3. The initial straight-line
part of the curve has a slope (modulus of elasticity) of
10 106 psi. The bar is loaded by tensile forces P 24 k
and then unloaded.
(a) What is the permanent set of the bar?
(b) If the bar is reloaded, what is the proportional
limit? (Hint: Use the concepts illustrated in Figs. 1-18b and
1-19.)
1.4-4 A circular bar of magnesium alloy is 800 mm long.
The stress-strain diagram for the material is shown in the
figure. The bar is loaded in tension to an elongation of
5.6 mm, and then the load is removed.
(a) What is the permanent set of the bar?
(b) If the bar is reloaded, what is the proportional
limit? (Hint: Use the concepts illustrated in Figs. 1-18b and
1-19.)
20
200
0
0
0.002
0.004
0.006
s (MPa)
e
100
PROB. 1.4-1
1.4-2 A bar of length 2.0 m is made of a structural steel
having the stress-strain diagram shown in the figure. The
yield stress of the steel is 250 MPa and the slope of the
initial linear part of the stress-strain curve (modulus of
elasticity) is 200 GPa. The bar is loaded axially until it
elongates 6.5 mm, and then the load is removed.
How does the final length of the bar compare with
its original length? (Hint: Use the concepts illustrated in
Fig. 1-18b.)
0
0.005
e
0.010
PROB. 1.4-4
★
1.4-5 A wire of length L 4 ft and diameter d 0.125
in. is stretched by tensile forces P 600 lb. The wire is
made of a copper alloy having a stress-strain relationship
that may be described mathematically by the following
equation:
18,000e
s 0 e 0.03 (s ksi)
1 300e
s (MPa)
300
200
100
0
0
0
0.002
0.004
0.006
e
PROB. 1.4-2
1.4-3 An aluminum bar has length L 4 ft and diameter
d 1.0 in. The stress-strain curve for the aluminum is
in which e is nondimensional and s has units of kips per
square inch (ksi).
(a) Construct a stress-strain diagram for the material.
(b) Determine the elongation of the wire due to the
forces P.
(c) If the forces are removed, what is the permanent set
of the bar?
(d) If the forces are applied again, what is the proportional limit?
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
54
CHAPTER 1 Tension, Compression, and Shear
Hooke’s Law and Poisson’s Ratio
When solving the problems for Section 1.5, assume that the
material behaves linearly elastically.
1.5-1 A high-strength steel bar used in a large crane has
diameter d 2.00 in. (see figure). The steel has modulus of
elasticity E 29 106 psi and Poisson’s ratio n 0.29.
Because of clearance requirements, the diameter of the bar
is limited to 2.001 in. when it is compressed by axial
forces.
What is the largest compressive load Pmax that is
permitted?
1.5-4 A prismatic bar of circular cross section is loaded by
tensile forces P (see figure). The bar has length L 1.5 m
and diameter d 30 mm. It is made of aluminum alloy
with modulus of elasticity E 75 GPa and Poisson’s ratio
n 1/3.
If the bar elongates by 3.6 mm, what is the decrease in
diameter d? What is the magnitude of the load P?
d
P
P
L
d
P
PROBS. 1.5-4 and 1.5-5
P
PROB. 1.5-1
1.5-2 A round bar of 10 mm diameter is made of
aluminum alloy 7075-T6 (see figure). When the bar is
stretched by axial forces P, its diameter decreases by 0.016
mm.
Find the magnitude of the load P. (Obtain the material
properties from Appendix H.)
d = 10 mm
P
P
7075-T6
1.5-5 A bar of monel metal (length L 8 in., diameter
d 0.25 in.) is loaded axially by a tensile force P 1500 lb
(see figure). Using the data in Table H-2, Appendix H, determine the increase in length of the bar and the percent
decrease in its cross-sectional area.
1.5-6 A tensile test is peformed on a brass specimen 10
mm in diameter using a gage length of 50 mm (see figure).
When the tensile load P reaches a value of 20 kN, the distance between the gage marks has increased by 0.122 mm.
(a) What is the modulus of elasticity E of the brass?
(b) If the diameter decreases by 0.00830 mm, what is
Poisson’s ratio?
PROB. 1.5-2
;;
@@
ÀÀ
@@
ÀÀ
;;
10 mm
1.5-3 A nylon bar having diameter d1 3.50 in. is placed
inside a steel tube having inner diameter d2 3.51 in. (see
figure). The nylon bar is then compressed by an axial
force P.
At what value of the force P will the space between
the nylon bar and the steel tube be closed? (For nylon,
assume E 400 ksi and n 0.4.)
Steel
tube
d1 d2
Nylon
bar
PROB. 1.5-3
50 mm
P
P
PROB. 1.5-6
1.5-7 A hollow steel cylinder is compressed by a force P
(see figure). The cylinder has inner diameter d1 3.9 in.,
outer diameter d2 4.5 in., and modulus of elasticity
E 30,000 ksi. When the force P increases from zero to
40 k, the outer diameter of the cylinder increases by
455 106 in.
(a) Determine the increase in the inner diameter.
(b) Determine the increase in the wall thickness.
(c) Determine Poisson’s ratio for the steel.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 1 Problems
55
P
p
b
L
d1
d2
PROB. 1.5-7
t
PROB. 1.6-1
1.5-8 A steel bar of length 2.5 m with a square cross section 100 mm on each side is subjected to an axial tensile
force of 1300 kN (see figure). Assume that E 200 GPa
and v 0.3.
Determine the increase in volume of the bar.
★
100 mm
100 mm
1.6-2 Three steel plates, each 16 mm thick, are joined by
two 20-mm diameter rivets as shown in the figure.
(a) If the load P 50 kN, what is the largest bearing
stress acting on the rivets?
(b) If the ultimate shear stress for the rivets is 180
MPa, what force Pult is required to cause the rivets to fail in
shear? (Disregard friction between the plates.)
1300 kN
1300 kN
P/2
P/2
P
P
P
2.5 m
PROB. 1.5-8
Shear Stress and Strain
PROB. 1.6-2
1.6-1 An angle bracket having thickness t 0.5 in. is
attached to the flange of a column by two 5/8-inch diameter
bolts (see figure). A uniformly distributed load acts on the
top face of the bracket with a pressure p 300 psi. The top
face of the bracket has length L 6 in. and width b 2.5 in.
Determine the average bearing pressure sb between
the angle bracket and the bolts and the average shear stress
taver in the bolts. (Disregard friction between the bracket
and the column.)
1.6-3 A bolted connection between a vertical column and a
diagonal brace is shown in the figure on the next page. The
connection consists of three 5/8-in. bolts that join two 1/4-in.
end plates welded to the brace and a 5/8-in. gusset plate
welded to the column. The compressive load P carried by
the brace equals 8.0 k.
Determine the following quantities: (a) The average
shear stress taver in the bolts, and (b) the average bearing
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
56
CHAPTER 1 Tension, Compression, and Shear
stress sb between the gusset plate and the bolts. (Disregard
friction between the plates.)
P
Column
Brace
1.6-5 The connection shown in the figure consists of
five steel plates, each 3/16 in. thick, joined by a single
1/4-in. diameter bolt. The total load transferred between
the plates is 1200 lb, distributed among the plates as shown.
(a) Calculate the largest shear stress in the bolt, disregarding friction between the plates.
(b) Calculate the largest bearing stress acting against
the bolt.
360 lb
600 lb
480 lb
End plates
for brace
PROB. 1.6-5
Gusset
plate
PROB. 1.6-3
1.6-4 A hollow box beam ABC of length L is supported at
end A by a 20-mm diameter pin that passes through the
beam and its supporting pedestals (see figure). The roller
support at B is located at distance L/3 from end A.
(a) Determine the average shear stress in the pin due to
a load P equal to 10 kN.
(b) Determine the average bearing stress between the
pin and the box beam if the wall thickness of the beam is
equal to 12 mm.
P
Cable sling
32°
B
L
—
3
1.6-6 A steel plate of dimensions 2.5 1.2 0.1 m is
hoisted by a cable sling that has a clevis at each end (see
figure). The pins through the clevises are 18 mm in diameter and are located 2.0 m apart. Each half of the cable is at
an angle of 32° to the vertical.
For these conditions, determine the average shear
stress taver in the pins and the average bearing stress sb
between the steel plate and the pins.
P
Box beam
A
600 lb
360 lb
32°
C
Clevis
2L
—
3
2.0 m
Steel plate
(2.5 × 1.2 × 0.1 m)
Box beam
Pin at support A
PROB. 1.6-6
1.6-7 A special-purpose bolt of shank diameter d
PROB. 1.6-4
0.50 in. passes through a hole in a steel plate (see figure on
the next page). The hexagonal head of the bolt bears
directly against the steel plate. The radius of the circumscribed
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
57
CHAPTER 1 Problems
circle for the hexagon is r 0.40 in. (which means that
each side of the hexagon has length 0.40 in.). Also, the
thickness t of the bolt head is 0.25 in. and the tensile force
P in the bolt is 1000 lb.
(a) Determine the average bearing stress sb between
the hexagonal head of the bolt and the plate.
(b) Determine the average shear stress taver in the head
of the bolt.
@@@
ÀÀÀ
;;;
;;;;
@@@@
ÀÀÀÀ
@@
ÀÀ
;;
@@@
ÀÀÀ
;;;
@@@@
ÀÀÀÀ
;;;;
@@
ÀÀ
;;
ÀÀ@@
@@
;;
;;
;;
ÀÀ
(a) What is the average shear strain gaver in the epoxy?
(b) What is the magnitude of the forces V if the shear
modulus of elasticity G for the epoxy is 140 ksi?
A
B
L
Steel plate
h
d
P
2r
t
t
d
A
PROB. 1.6-7
h
1.6-8 An elastomeric bearing pad consisting of two steel
plates bonded to a chloroprene elastomer (an artificial rubber) is subjected to a shear force V during a static loading
test (see figure). The pad has dimensions a 150 mm and
b 250 mm, and the elastomer has thickness t 50 mm.
When the force V equals 12 kN, the top plate is found to
have displaced laterally 8.0 mm with respect to the bottom
plate.
What is the shear modulus of elasticity G of the
chloroprene?
b
a
V
t
B
V
V
t
PROB. 1.6-9
1.6-10 A flexible connection consisting of rubber pads
(thickness t 9 mm) bonded to steel plates is shown in the
figure. The pads are 160 mm long and 80 mm wide.
(a) Find the average shear strain gaver in the rubber if
the force P 16 kN and the shear modulus for the rubber
is G 1250 kPa.
(b) Find the relative horizontal displacement d
between the interior plate and the outer plates.
160 mm
P
—
2
X
P
—
2
X
P
Rubber pad
80 mm
PROB. 1.6-8
t = 9 mm
t = 9 mm
Section X-X
1.6-9 A joint between two concrete slabs A and B is filled
with a flexible epoxy that bonds securely to the concrete
(see figure). The height of the joint is h 4.0 in., its length
is L 40 in., and its thickness is t 0.5 in. Under the
action of shear forces V, the slabs displace vertically
through the distance d 0.002 in. relative to each other.
Rubber pad
PROB. 1.6-10
1.6-11 A spherical fiberglass buoy used in an underwater
experiment is anchored in shallow water by a chain
[see part (a) of the figure on the next page]. Because the
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
58
CHAPTER 1 Tension, Compression, and Shear
buoy is positioned just below the surface of the water, it is
not expected to collapse from the water pressure. The
chain is attached to the buoy by a shackle and pin [see
part (b) of the figure]. The diameter of the pin is 0.5 in.
and the thickness of the shackle is 0.25 in. The buoy has a
diameter of 60 in. and weighs 1800 lb on land (not including the weight of the chain).
(a) Determine the average shear stress taver in the pin.
(b) Determine the average bearing stress sb between
the pin and the shackle.
d
Determine the average shear stress in the pin at C
when the load P 18 kN.
Pin
Shackle
;@À
c Line 2
P
Arm A
Arm B
Line 1
h
(b)
Arm A
C
P
(a)
PROB. 1.6-12
PROB. 1.6-11
1.6-12 The clamp shown in the figure is used to support a
load hanging from the lower flange of a steel beam. The
clamp consists of two arms (A and B) joined by a pin at C.
The pin has diameter d 12 mm. Because arm B straddles
arm A, the pin is in double shear.
Line 1 in the figure defines the line of action of the
resultant horizontal force H acting between the lower
flange of the beam and arm B. The vertical distance from
this line to the pin is h 250 mm. Line 2 defines the line
of action of the resultant vertical force V acting between the
flange and arm B. The horizontal distance from this line to
the centerline of the beam is c 100 mm. The force conditions between arm A and the lower flange are symmetrical
with those given for arm B.
★
1.6-13 A specially designed wrench is used to twist a circular shaft by means of a square key that fits into slots (or
keyways) in the shaft and wrench, as shown in the figure on
the next page. The shaft has diameter d, the key has a
square cross section of dimensions b b, and the length of
the key is c. The key fits half into the wrench and half into
the shaft (i.e., the keyways have a depth equal to b/2).
Derive a formula for the average shear stress taver in
the key when a load P is applied at distance L from the
center of the shaft.
Hints: Disregard the effects of friction, assume that the
bearing pressure between the key and the wrench is uni-
★
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 1 Problems
formly distributed, and be sure to draw free-body diagrams
of the wrench and key.
Links
59
Pin
c
12 mm
2.5 mm
Shaft
Key
T
F
Sprocket
Wrench
R
L
P
Chain
L
b
PROB. 1.6-14
1.6-15 A shock mount constructed as shown in the figure is used to support a delicate instrument. The mount
consists of an outer steel tube with inside diameter b, a central steel bar of diameter d that supports the load P, and a
hollow rubber cylinder (height h) bonded to the tube and
bar.
(a) Obtain a formula for the shear stress t in the rubber
at a radial distance r from the center of the shock mount.
(b) Obtain a formula for the downward displacement d
of the central bar due to the load P, assuming that G is the
shear modulus of elasticity of the rubber and that the steel
tube and bar are rigid.
★★
d
PROB. 1.6-13
Steel tube
1.6-14 A bicycle chain consists of a series of small
★★
links, each 12 mm long between the centers of the pins (see
figure). You might wish to examine a bicycle chain and
observe its construction. Note particularly the pins, which
we will assume to have a diameter of 2.5 mm.
In order to solve this problem, you must now make
two measurements on a bicycle (see figure): (1) the length
L of the crank arm from main axle to pedal axle, and (2) the
radius R of the sprocket (the toothed wheel, sometimes
called the chainring).
(a) Using your measured dimensions, calculate the tensile force T in the chain due to a force F 800 N applied
to one of the pedals.
(b) Calculate the average shear stress taver in the pins.
r
P
Steel bar
d
Rubber
h
b
PROB. 1.6-15
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
60
CHAPTER 1 Tension, Compression, and Shear
Allowable Loads
P
1.7-1 A bar of solid circular cross section is loaded in tension by forces P (see figure). The bar has length L 16.0
in. and diameter d 0.50 in. The material is a magnesium
alloy having modulus of elasticity E 6.4 106 psi. The
allowable stress in tension is sallow 17,000 psi, and the
elongation of the bar must not exceed 0.04 in.
What is the allowable value of the forces P?
d
P
P
L
ÀÀ
@@
;;
dB
dB
dW
t
dW
PROB. 1.7-3
PROB. 1.7-1
1.7-2 A torque T0 is transmitted between two flanged
shafts by means of four 20-mm bolts (see figure). The
diameter of the bolt circle is d 150 mm.
If the allowable shear stress in the bolts is 90 MPa,
what is the maximum permissible torque? (Disregard friction between the flanges.)
T0
d
1.7-4 An aluminum tube serving as a compression brace in
the fuselage of a small airplane has the cross section shown
in the figure. The outer diameter of the tube is d 25 mm
and the wall thickness is t 2.5 mm. The yield stress for
the aluminum is sY 270 MPa and the ultimate stress is
sU 310 MPa.
Calculate the allowable compressive force Pallow if the
factors of safety with respect to the yield stress and the
ultimate stress are 4 and 5, respectively.
t
T0
d
PROB. 1.7-2
PROB. 1.7-4
1.7-3 A tie-down on the deck of a sailboat consists of a
bent bar bolted at both ends, as shown in the figure. The
diameter dB of the bar is 1/4 in., the diameter dW of the
washers is 7/8 in., and the thickness t of the fiberglass deck
is 3/8 in.
If the allowable shear stress in the fiberglass is 300 psi,
and the allowable bearing pressure between the washer and
the fiberglass is 550 psi, what is the allowable load Pallow
on the tie-down?
1.7-5 A steel pad supporting heavy machinery rests on four
short, hollow, cast iron piers (see figure on the next page).
The ultimate strength of the cast iron in compression is
50 ksi. The outer diameter of the piers is d 4.5 in. and the
wall thickness is t 0.40 in.
Using a factor of safety of 3.5 with respect to the
ultimate strength, determine the total load P that may be
supported by the pad.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 1 Problems
t
61
Cables attached to the lifeboat pass over the pulleys
and wind around winches that raise and lower the lifeboat.
The lower parts of the cables are vertical and the upper
parts make an angle a 15° with the horizontal. The
allowable tensile force in each cable is 1800 lb, and the
allowable shear stress in the pins is 4000 psi.
If the lifeboat weighs 1500 lb, what is the maximum
weight that should be carried in the lifeboat?
d
PROB. 1.7-5
T T
1.7-6 A long steel wire hanging from a balloon carries a
weight W at its lower end (see figure). The 4-mm diameter
wire is 25 m long.
What is the maximum weight Wmax that can safely be
carried if the tensile yield stress for the wire is sY
350 MPa and a margin of safety against yielding of 1.5 is
desired? (Include the weight of the wire in the calculations.)
Davit
a = 15°
Pulley
Pin
Cable
PROB. 1.7-7
d
L
W
PROB. 1.7-6
1.7-7 A lifeboat hangs from two ship’s davits, as shown in
the figure. A pin of diameter d 0.80 in. passes through
each davit and supports two pulleys, one on each side of the
davit.
1.7-8 A ship’s spar is attached at the base of a mast by a
pin connection (see figure on the next page). The spar is a
steel tube of outer diameter d2 80 mm and inner diameter
d1 70 mm. The steel pin has diameter d 25 mm, and
the two plates connecting the spar to the pin have thickness
t 12 mm.
The allowable stresses are as follows: compressive
stress in the spar, 70 MPa; shear stress in the pin, 45 MPa;
and bearing stress between the pin and the connecting
plates, 110 MPa.
Determine the allowable compressive force Pallow in
the spar.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
62
CHAPTER 1 Tension, Compression, and Shear
Mast
Pin
0.75 m
P
2.5 m
0.75 m
1.75 m
Spar
Connecting
plate
1.75 m
W
A
PROB. 1.7-8
B
PROB. 1.7-10
1.7-11 Two flat bars loaded in tension by forces P are
1.7-9 What is the maximum possible value of the clamping
force C in the jaws of the pliers shown in the figure if a
3.75 in., b 1.60 in., and the ultimate shear stress in the
0.20-in. diameter pin is 50 ksi?
What is the maximum permissible value of the applied
load P if a factor of safety of 3.0 with respect to failure of
the pin is to be maintained?
P
spliced using two rectangular splice plates and two 5/8-in.
diameter rivets (see figure). The bars have width b 1.0 in.
(except at the splice, where the bars are wider) and thickness
t 0.4 in. The bars are made of steel having an ultimate
stress in tension equal to 60 ksi. The ultimate stresses in
shear and bearing for the rivet steel are 25 ksi and 80 ksi,
respectively.
Determine the allowable load Pallow if a safety factor
of 2.5 is desired with respect to the ultimate load that can
be carried. (Consider tension in the bars, shear in the rivets,
and bearing between the rivets and the bars. Disregard friction between the plates.)
Pin
b = 1.0 in.
P
P
a
b
P
Bar
Splice plate
t = 0.4 in.
PROB. 1.7-9
P
1.7-10 A metal bar AB of weight W is suspended by a system of steel wires arranged as shown in the figure. The
diameter of the wires is 2 mm, and the yield stress of the
steel is 450 MPa.
Determine the maximum permissible weight Wmax for
a factor of safety of 1.9 with respect to yielding.
P
PROB. 1.7-11
1.7-12 A solid bar of circular cross section (diameter d)
has a hole of diameter d/4 drilled laterally through the center
of the bar (see figure on the next page). The allowable average tensile stress on the net cross section of the bar is sallow.
★
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
63
CHAPTER 1 Problems
(a) Obtain a formula for the allowable load Pallow that
the bar can carry in tension.
(b) Calculate the value of Pallow if the bar is made of
brass with diameter d 40 mm and sallow 80 MPa.
(Hint: Use the formulas of Case 15, Appendix D.)
Cylinder
Piston
Connecting rod
A
P
M
d
B
R
L
d
P
–d
4
–d
4
P
C
PROB. 1.7-14
Design for Axial Loads and Direct Shear
d
1.8-1 An aluminum tube is required to transmit an axial
PROB. 1.7-12
1.7-13 A solid steel bar of diameter d1 2.25 in. has a
hole of diameter d2 1.125 in. drilled through it (see figure). A steel pin of diameter d2 passes through the hole and
is attached to supports.
Determine the maximum permissible tensile load
Pallow in the bar if the yield stress for shear in the pin is
ty 17,000 psi, the yield stress for tension in the bar is
sy 36,000 psi, and a factor of safety of 2.0 with respect
to yielding is required. (Hint: Use the formulas of Case 15,
Appendix D.)
★
tensile force P 34 k (see figure). The thickness of the
wall of the tube is to be 0.375 in.
What is the minimum required outer diameter dmin if
the allowable tensile stress is 9000 psi?
d
P
P
PROB. 1.8-1
1.8-2 A steel pipe having yield stress sy 270 MPa is to
d2
d1
d1
carry an axial compressive load P 1200 kN (see figure).
A factor of safety of 1.8 against yielding is to be used.
If the thickness t of the pipe is to be one-eighth of
its outer diameter, what is the minimum required outer
diameter dmin?
P
d
t =—
8
P
PROB. 1.7-13
1.7-14 The piston in an engine is attached to a connecting rod AB, which in turn is connected to a crank arm BC
(see figure). The piston slides without friction in a cylinder
and is subjected to a force P (assumed to be constant) while
moving to the right in the figure. The connecting rod,
which has diameter d and length L, is attached at both ends
by pins. The crank arm rotates about the axle at C with the
pin at B moving in a circle of radius R. The axle at C,
which is supported by bearings, exerts a resisting moment
M against the crank arm.
(a) Obtain a formula for the maximum permissible
force Pallow based upon an allowable compressive stress
sc in the connecting rod.
(b) Calculate the force Pallow for the following data:
sc 160 MPa, d 9.00 mm, and R 0.28L.
★
d
PROB. 1.8-2
1.8-3 A horizontal beam AB supported by an inclined strut
CD carries a load P 2500 lb at the position shown in the
figure on the next page. The strut, which consists of two
bars, is connected to the beam by a bolt passing through the
three bars meeting at joint C.
If the allowable shear stress in the bolt is 14,000 psi,
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
64
CHAPTER 1 Tension, Compression, and Shear
what is the minimum required diameter dmin of the bolt?
4 ft
A
4 ft
B
C
3 ft
P
D
Beam AB
1.8-6 A suspender on a suspension bridge consists of a
cable that passes over the main cable (see figure) and supports the bridge deck, which is far below. The suspender is
held in position by a metal tie that is prevented from sliding
downward by clamps around the suspender cable.
Let P represent the load in each part of the suspender
cable, and let u represent the angle of the suspender cable
just above the tie. Finally, let sallow represent the allowable
tensile stress in the metal tie.
(a) Obtain a formula for the minimum required crosssectional area of the tie.
(b) Calculate the minimum area if P 130 kN, u
75°, and sallow 80 MPa.
Bolt
Strut CD
Main
cable
PROB. 1.8-3
1.8-4 Two bars of rectangular cross section (thickness t
15 mm) are connected by a bolt in the manner shown in the
figure. The allowable shear stress in the bolt is 90 MPa and
the allowable bearing stress between the bolt and the bars is
150 MPa.
If the tensile load P 31 kN, what is the minimum
required diameter dmin of the bolt?
P
t
t
Suspender
Collar
u
u
Tie
Clamp
P
P
P
PROB. 1.8-6
P
P
PROBS. 1.8-4 and 1.8-5
1.8-5 Solve the preceding problem if the bars have thickness
t 5/16 in., the allowable shear stress is 12,000 psi, the
allowable bearing stress is 20,000 psi, and the load P
1800 lb.
1.8-7 A square steel tube of length L 20 ft and width
b2 10.0 in. is hoisted by a crane (see figure on the next
page). The tube hangs from a pin of diameter d that is held
by the cables at points A and B. The cross section is a
hollow square with inner dimension b1 8.5 in. and outer
dimension b2 10.0 in. The allowable shear stress in the
pin is 8,700 psi, and the allowable bearing stress between
the pin and the tube is 13,000 psi.
Determine the minimum diameter of the pin in order to
support the weight of the tube. (Note: Disregard the
rounded corners of the tube when calculating its weight.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 1 Problems
If the wall thickness of the post is 15 mm, what is the
minimum permissible value of the outer diameter d2?
d
A
B
Cable
Square
tube
Square
tube
65
Turnbuckle
Pin
L
d
b2
A
B
b1
b2
60°
d2
Post
60°
PROBS. 1.8-7 and 1.8-8
1.8-8 Solve the preceding problem if the length L of the
tube is 6.0 m, the outer width is b2 250 mm, the inner
dimension is b1 210 mm, the allowable shear stress in the
pin is 60 MPa, and the allowable bearing stress is 90 MPa.
1.8-9 A pressurized circular cylinder has a sealed cover plate
fastened with steel bolts (see figure). The pressure p of the
gas in the cylinder is 290 psi, the inside diameter D of the
cylinder is 10.0 in., and the diameter dB of the bolts is 0.50 in.
If the allowable tensile stress in the bolts is 10,000 psi,
find the number n of bolts needed to fasten the cover.
Cover plate
PROB. 1.8-10
1.8-11 A cage for transporting workers and supplies on a
construction site is hoisted by a crane (see figure). The
floor of the cage is rectangular with dimensions 6 ft by 8 ft.
Each of the four lifting cables is attached to a corner of the
cage and is 13 ft long. The weight of the cage and its
contents is limited by regulations to 9600 lb.
Determine the required cross-sectional area AC of a
cable if the breaking stress of a cable is 91 ksi and a factor
of safety of 3.5 with respect to failure is desired.
Steel bolt
p
Cylinder
D
PROB. 1.8-9
1.8-10 A tubular post of outer diameter d2 is guyed by two
cables fitted with turnbuckles (see figure). The cables are
tightened by rotating the turnbuckles, thus producing tension in the cables and compression in the post. Both cables
are tightened to a tensile force of 110 kN. Also, the angle
between the cables and the ground is 60°, and the allowable
compressive stress in the post is sc 35 MPa.
PROB. 1.8-11
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
66
CHAPTER 1 Tension, Compression, and Shear
1.8-12 A steel column of hollow circular cross section is
supported on a circular steel base plate and a concrete
pedestal (see figure). The column has outside diameter
d 250 mm and supports a load P 750 kN.
(a) If the allowable stress in the column is 55 MPa, what
is the minimum required thickness t? Based upon your result,
select a thickness for the column. (Select a thickness that is an
even integer, such as 10, 12, 14, . . ., in units of millimeters.)
(b) If the allowable bearing stress on the concrete
pedestal is 11.5 MPa, what is the minimum required
diameter D of the base plate if it is designed for the allowable
load Pallow that the column with the selected thickness can
support?
P
d
Column
P
Base plate
1.8-15 Two bars AB and BC of the same material support a vertical load P (see figure). The length L of the
horizontal bar is fixed, but the angle u can be varied by
moving support A vertically and changing the length of bar
AC to correspond with the new position of support A. The
allowable stresses in the bars are the same in tension and
compression.
We observe that when the angle u is reduced, bar AC
becomes shorter but the cross-sectional areas of both bars
increase (because the axial forces are larger). The opposite
effects occur if the angle u is increased. Thus, we see that
the weight of the structure (which is proportional to the volume) depends upon the angle u.
Determine the angle u so that the structure has minimum weight without exceeding the allowable stresses in
the bars. (Note: The weights of the bars are very small
compared to the force P and may be disregarded.)
★★
t
D
PROB. 1.8-12
1.8-13 A bar of rectangular cross section is subjected to an
axial load P (see figure). The bar has width b 2.0 in. and
thickness t 0.25 in. A hole of diameter d is drilled
through the bar to provide for a pin support. The allowable
tensile stress on the net cross section of the bar is 20 ksi,
and the allowable shear stress in the pin is 11.5 ksi.
(a) Determine the pin diameter dm for which the load P
will be a maximum.
(b) Determine the corresponding value Pmax of the
load.
d
1.8-14 A flat bar of width b 60 mm and thickness
t 10 mm is loaded in tension by a force P (see figure).
The bar is attached to a support by a pin of diameter d that
passes through a hole of the same size in the bar. The
allowable tensile stress on the net cross section of the bar is
sT 140 MPa, the allowable shear stress in the pin is
tS 80 MPa, and the allowable bearing stress between the
pin and the bar is sB 200 MPa.
(a) Determine the pin diameter dm for which the load P
will be a maximum.
(b) Determine the corresponding value Pmax of the
load.
★
A
P
b
θ
B
t
P
C
L
P
PROBS. 1.8-13 and 1.8-14
PROB. 1.8-15
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
2
Axially Loaded Members
2.1 INTRODUCTION
Structural components subjected only to tension or compression are
known as axially loaded members. Solid bars with straight longitudinal
axes are the most common type, although cables and coil springs also
carry axial loads. Examples of axially loaded bars are truss members,
connecting rods in engines, spokes in bicycle wheels, columns in buildings, and struts in aircraft engine mounts. The stress-strain behavior of
such members was discussed in Chapter 1, where we also obtained
equations for the stresses acting on cross sections (s P/A) and the
strains in longitudinal directions (e d /L).
In this chapter we consider several other aspects of axially loaded
members, beginning with the determination of changes in lengths caused
by loads (Sections 2.2 and 2.3). The calculation of changes in lengths is
an essential ingredient in the analysis of statically indeterminate structures, a topic we introduce in Section 2.4. Changes in lengths also must
be calculated whenever it is necessary to control the displacements of a
structure, whether for aesthetic or functional reasons. In Section 2.5, we
discuss the effects of temperature on the length of a bar, and we introduce
the concepts of thermal stress and thermal strain. Also included in this
section is a discussion of the effects of misfits and prestrains.
A generalized view of the stresses in axially loaded bars is presented
in Section 2.6, where we discuss the stresses on inclined sections (as
distinct from cross sections) of bars. Although only normal stresses act
on cross sections of axially loaded bars, both normal and shear stresses
act on inclined sections.
We then introduce several additional topics of importance in
mechanics of materials, namely, strain energy (Section 2.7), impact
loading (Section 2.8), fatigue (Section 2.9), stress concentrations (Section
2.10), and nonlinear behavior (Sections 2.11 and 2.12). Although these
67
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
68
CHAPTER 2 Axially Loaded Members
subjects are discussed in the context of members with axial loads, the
discussions provide the foundation for applying the same concepts to
other structural elements, such as bars in torsion and beams in bending.
2.2 CHANGES IN LENGTHS OF AXIALLY LOADED MEMBERS
When determining the changes in lengths of axially loaded members, it
is convenient to begin with a coil spring (Fig. 2-1). Springs of this type
are used in large numbers in many kinds of machines and devices—for
instance, there are dozens of them in every automobile.
When a load is applied along the axis of a spring, as shown in Fig. 2-1,
the spring gets longer or shorter depending upon the direction of the load.
If the load acts away from the spring, the spring elongates and we say that
the spring is loaded in tension. If the load acts toward the spring, the spring
shortens and we say it is in compression. However, it should not be
inferred from this terminology that the individual coils of a spring are
subjected to direct tensile or compressive stresses; rather, the coils act
primarily in direct shear and torsion (or twisting). Nevertheless, the overall
stretching or shortening of a spring is analogous to the behavior of a bar in
tension or compression, and so the same terminology is used.
P
FIG. 2-1 Spring subjected to an axial
load P
Springs
The elongation of a spring is pictured in Fig. 2-2, where the upper part
of the figure shows a spring in its natural length L (also called its
unstressed length, relaxed length, or free length), and the lower part of
the figure shows the effects of applying a tensile load. Under the action
of the force P, the spring lengthens by an amount d and its final length
becomes L d. If the material of the spring is linearly elastic, the load
and elongation will be proportional:
P k
L
d
P
FIG. 2-2 Elongation of an axially loaded
spring
fP
(2-1a,b)
in which k and f are constants of proportionality.
The constant k is called the stiffness of the spring and is defined as
the force required to produce a unit elongation, that is, k P/d. Similarly, the constant f is known as the flexibility and is defined as the
elongation produced by a load of unit value, that is, f d/P. Although
we used a spring in tension for this discussion, it should be obvious that
Eqs. (2-1a) and (2-1b) also apply to springs in compression.
From the preceding discussion it is apparent that the stiffness and
flexibility of a spring are the reciprocal of each other:
1
f
k
1
f
k
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(2-2a,b)
SECTION 2.2 Changes in Lengths of Axially Loaded Members
P
FIG. 2-3 Prismatic bar of circular
cross section
69
The flexibility of a spring can easily be determined by measuring the
elongation produced by a known load, and then the stiffness can be
calculated from Eq. (2-2a). Other terms for the stiffness and flexibility
of a spring are the spring constant and compliance, respectively.
The spring properties given by Eqs. (2-1) and (2-2) can be used in
the analysis and design of various mechanical devices involving springs,
as illustrated later in Example 2-1.
Prismatic Bars
Axially loaded bars elongate under tensile loads and shorten under
compressive loads, just as springs do. To analyze this behavior, let us
consider the prismatic bar shown in Fig. 2-3. A prismatic bar is a structural member having a straight longitudinal axis and constant cross
section throughout its length. Although we often use circular bars in our
illustrations, we should bear in mind that structural members may have a
variety of cross-sectional shapes, such as those shown in Fig. 2-4.
The elongation d of a prismatic bar subjected to a tensile load P is
shown in Fig. 2-5. If the load acts through the centroid of the end cross
section, the uniform normal stress at cross sections away from the ends
is given by the formula s P/A, where A is the cross-sectional area.
Furthermore, if the bar is made of a homogeneous material, the axial
strain is e d/L, where d is the elongation and L is the length of the bar.
Let us also assume that the material is linearly elastic, which means
that it follows Hooke’s law. Then the longitudinal stress and strain are
related by the equation s Ee, where E is the modulus of elasticity.
Combining these basic relationships, we get the following equation for
the elongation of the bar:
Solid cross sections
Hollow or tubular cross sections
Thin-walled open cross sections
FIG. 2-4 Typical cross sections of
structural members
PL
EA
L
d
P
FIG. 2-5 Elongation of a prismatic bar in
tension
(2-3)
This equation shows that the elongation is directly proportional to the
load P and the length L and inversely proportional to the modulus of
elasticity E and the cross-sectional area A. The product EA is known as
the axial rigidity of the bar.
Although Eq. (2-3) was derived for a member in tension, it applies
equally well to a member in compression, in which case d represents the
shortening of the bar. Usually we know by inspection whether a member
gets longer or shorter; however, there are occasions when a sign
convention is needed (for instance, when analyzing a statically indeterminate bar). When that happens, elongation is usually taken as positive
and shortening as negative.
The change in length of a bar is normally very small in comparison
to its length, especially when the material is a structural metal, such as
steel or aluminum. As an example, consider an aluminum strut that is
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
70
CHAPTER 2 Axially Loaded Members
75.0 in. long and subjected to a moderate compressive stress of 7000
psi. If the modulus of elasticity is 10,500 ksi, the shortening of the strut
(from Eq. 2-3 with P/A replaced by s) is d 0.050 in. Consequently,
the ratio of the change in length to the original length is 0.05/75, or
1/1500, and the final length is 0.999 times the original length. Under
ordinary conditions similar to these, we can use the original length of a
bar (instead of the final length) in calculations.
The stiffness and flexibility of a prismatic bar are defined in the
same way as for a spring. The stiffness is the force required to produce a
unit elongation, or P/d, and the flexibility is the elongation due to a unit
load, or d/P. Thus, from Eq. (2-3) we see that the stiffness and flexibility of a prismatic bar are, respectively,
EA
L
k
L
EA
f
(2-4a,b)
Stiffnesses and flexibilities of structural members, including those given
by Eqs. (2-4a) and (2-4b), have a special role in the analysis of large
structures by computer-oriented methods.
Cables
FIG. 2-6 Typical arrangement of strands
and wires in a steel cable
Cables are used to transmit large tensile forces, for example, when
lifting and pulling heavy objects, raising elevators, guying towers, and
supporting suspension bridges. Unlike springs and prismatic bars, cables
cannot resist compression. Furthermore, they have little resistance to
bending and therefore may be curved as well as straight. Nevertheless, a
cable is considered to be an axially loaded member because it is
subjected only to tensile forces. Because the tensile forces in a cable are
directed along the axis, the forces may vary in both direction and magnitude, depending upon the configuration of the cable.
Cables are constructed from a large number of wires wound in some
particular manner. While many arrangements are available depending
upon how the cable will be used, a common type of cable, shown in
Fig. 2-6, is formed by six strands wound helically around a central
strand. Each strand is in turn constructed of many wires, also wound
helically. For this reason, cables are often referred to as wire rope.
The cross-sectional area of a cable is equal to the total crosssectional area of the individual wires, called the effective area or
metallic area. This area is less than the area of a circle having the same
diameter as the cable because there are spaces between the individual
wires. For example, the actual cross-sectional area (effective area) of a
particular 1.0 inch diameter cable is only 0.471 in.2, whereas the area of
a 1.0 in. diameter circle is 0.785 in.2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
71
SECTION 2.2 Changes in Lengths of Axially Loaded Members
Under the same tensile load, the elongation of a cable is greater than
the elongation of a solid bar of the same material and same metallic
cross-sectional area, because the wires in a cable “tighten up” in the
same manner as the fibers in a rope. Thus, the modulus of elasticity
(called the effective modulus) of a cable is less than the modulus of the
material of which it is made. The effective modulus of steel cables is
about 20,000 ksi (140 GPa), whereas the steel itself has a modulus of
about 30,000 ksi (210 GPa).
When determining the elongation of a cable from Eq. (2-3), the
effective modulus should be used for E and the effective area should be
used for A.
In practice, the cross-sectional dimensions and other properties of
cables are obtained from the manufacturers. However, for use in solving
problems in this book (and definitely not for use in engineering applications), we list in Table 2-1 the properties of a particular type of cable.
Note that the last column contains the ultimate load, which is the load
that would cause the cable to break. The allowable load is obtained from
the ultimate load by applying a safety factor that may range from 3 to
10, depending upon how the cable is to be used. The individual wires in
a cable are usually made of high-strength steel, and the calculated
tensile stress at the breaking load can be as high as 200,000 psi (1400
MPa).
The following examples illustrate techniques for analyzing simple
devices containing springs and bars. The solutions require the use of
free-body diagrams, equations of equilibrium, and equations for changes
in length. The problems at the end of the chapter provide many additional examples.
TABLE 2-1 PROPERTIES OF STEEL CABLES*
Nominal
diameter
Approximate
weight
Effective
area
Ultimate
load
in.
(mm)
lb/ft
(N/m)
in.2
(mm2)
lb
(kN)
0.50
0.75
1.00
1.25
1.50
1.75
2.00
(12)
(20)
(25)
(32)
(38)
(44)
(50)
0.42
0.95
1.67
2.64
3.83
5.24
6.84
(6.1)
(13.9)
(24.4)
(38.5)
(55.9)
(76.4)
(99.8)
0.119
0.268
0.471
0.745
1.08
1.47
1.92
(76.7)
(173)
(304)
(481)
(697)
(948)
(1230)
23,100
51,900
91,300
144,000
209,000
285,000
372,000
(102)
(231)
(406)
(641)
(930)
(1260)
(1650)
* To be used solely for solving problems in this book.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
72
CHAPTER 2 Axially Loaded Members
Example 2-1
A rigid L-shaped frame ABC consisting of a horizontal arm AB (length b
10.5 in.) and a vertical arm BC (length c 6.4 in.) is pivoted at point B, as
shown in Fig. 2-7a. The pivot is attached to the outer frame BCD, which stands
on a laboratory bench. The position of the pointer at C is controlled by a spring
(stiffness k 4.2 lb/in.) that is attached to a threaded rod. The position of the
threaded rod is adjusted by turning the knurled nut.
The pitch of the threads (that is, the distance from one thread to the next) is
p 1/16 in., which means that one full revolution of the nut will move the rod
by that same amount. Initially, when there is no weight on the hanger, the nut is
turned until the pointer at the end of arm BC is directly over the reference mark
on the outer frame.
If a weight W 2 lb is placed on the hanger at A, how many revolutions of
the nut are required to bring the pointer back to the mark? (Deformations of the
metal parts of the device may be disregarded because they are negligible
compared to the change in length of the spring.)
b
B
A
Hanger
Frame
c
W
Spring
Knurled nut
Threaded
rod
D
C
(a)
W
b
A
B
F
W
c
FIG. 2-7 Example 2-1. (a) Rigid
L-shaped frame ABC attached to outer
frame BCD by a pivot at B, and
(b) free-body diagram of frame ABC
F
C
(b)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.2 Changes in Lengths of Axially Loaded Members
73
Solution
Inspection of the device (Fig. 2-7a) shows that the weight W acting downward will cause the pointer at C to move to the right. When the pointer moves to
the right, the spring stretches by an additional amount—an amount that we can
determine from the force in the spring.
To determine the force in the spring, we construct a free-body diagram of
frame ABC (Fig. 2-7b). In this diagram, W represents the force applied by the
hanger and F represents the force applied by the spring. The reactions at the
pivot are indicated with slashes across the arrows (see the discussion of reactions in Section 1.8).
Taking moments about point B gives
Wb
F
c
(a)
The corresponding elongation of the spring (from Eq. 2-1a) is
F
Wb
d
k
ck
(b)
To bring the pointer back to the mark, we must turn the nut through enough
revolutions to move the threaded rod to the left an amount equal to the elongation of the spring. Since each complete turn of the nut moves the rod a distance
equal to the pitch p, the total movement of the rod is equal to np, where n is the
number of turns. Therefore,
Wb
np d
ck
(c)
from which we get the following formula for the number of revolutions of the
nut:
Wb
n
ckp
(d)
Numerical results. As the final step in the solution, we substitute the given
numerical data into Eq. (d), as follows:
(2 lb)(10.5 in.)
Wb
n 12.5 revolutions
(6.4 in.)(4.2 lb/in.)(1/16 in.)
ckp
This result shows that if we rotate the nut through 12.5 revolutions, the threaded
rod will move to the left an amount equal to the elongation of the spring caused
by the 2-lb load, thus returning the pointer to the reference mark.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
74
CHAPTER 2 Axially Loaded Members
Example 2-2
The device shown in Fig. 2-8a consists of a horizontal beam ABC supported by
two vertical bars BD and CE. Bar CE is pinned at both ends but bar BD is fixed
to the foundation at its lower end. The distance from A to B is 450 mm and from
B to C is 225 mm. Bars BD and CE have lengths of 480 mm and 600 mm,
respectively, and their cross-sectional areas are 1020 mm2 and 520 mm2,
respectively. The bars are made of steel having a modulus of elasticity E 205
GPa.
Assuming that beam ABC is rigid, find the maximum allowable load Pmax
if the displacement of point A is limited to 1.0 mm.
A
B
C
P
450 mm
225 mm
600 mm
D
120 mm
E
(a)
A
B
P
H
C
FCE
FBD
450 mm
225 mm
(b)
B"
A"
B
A
d BD
a
B'
dA
A'
450 mm
225 mm
FIG. 2-8 Example 2-2. Horizontal beam
ABC supported by two vertical bars
(c)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
C'
d CE
C
SECTION 2.2 Changes in Lengths of Axially Loaded Members
75
Solution
To find the displacement of point A, we need to know the displacements of
points B and C. Therefore, we must find the changes in lengths of bars BD and
CE, using the general equation d PL/EA (Eq. 2-3).
We begin by finding the forces in the bars from a free-body diagram of
the beam (Fig. 2-8b). Because bar CE is pinned at both ends, it is a “two-force”
member and transmits only a vertical force FCE to the beam. However, bar
BD can transmit both a vertical force FBD and a horizontal force H. From equilibrium of beam ABC in the horizontal direction, we see that the horizontal force
vanishes.
Two additional equations of equilibrium enable us to express the forces
FBD and FCE in terms of the load P. Thus, by taking moments about point B and
then summing forces in the vertical direction, we find
FCE 2P
FBD 3P
(e)
Note that the force FCE acts downward on bar ABC and the force FBD acts
upward. Therefore, member CE is in tension and member BD is in compression.
The shortening of member BD is
FBDLBD
dBD
EABD
(3P)(480 mm)
6.887P 106 mm (P newtons)
(205 GPa)(1020 mm2)
(f)
Note that the shortening dBD is expressed in millimeters provided the load P is
expressed in newtons.
Similarly, the lengthening of member CE is
FCEL C E
dCE
E AC E
(2P)(600 mm)
11.26P 106 mm (P newtons)
(205 GPa)(520 mm2)
(g)
Again, the displacement is expressed in millimeters provided the load P is
expressed in newtons. Knowing the changes in lengths of the two bars, we can
now find the displacement of point A.
Displacement diagram. A displacement diagram showing the relative positions of points A, B, and C is sketched in Fig. 2-8c. Line ABC represents the
original alignment of the three points. After the load P is applied, member BD
shortens by the amount dBD and point B moves to B. Also, member CE elongates by the amount dCE and point C moves to C. Because the beam ABC is
assumed to be rigid, points A, B, and C lie on a straight line.
For clarity, the displacements are highly exaggerated in the diagram. In
reality, line ABC rotates through a very small angle to its new position ABC
(see Note 2 at the end of this example).
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
76
CHAPTER 2 Axially Loaded Members
Using similar triangles, we can now find the relationships between the
displacements at points A, B, and C. From triangles A⬘A⬙C⬘ and B⬘B⬙C⬘ we get
BB
dA dCE
dBD dCE
AA
or
A C
B C
450 225
225
(h)
in which all terms are expressed in millimeters.
Substituting for dBD and dCE from Eqs. (f) and (g) gives
B"
A"
B
A
d BD
C'
d CE
C
a
B'
dA
dA 11.26P 106
6.887P 106 11.26P 106
225
450 225
Finally, we substitute for dA its limiting value of 1.0 mm and solve the equation
for the load P. The result is
P Pmax 23,200 N (or 23.2 kN)
A'
450 mm
225 mm
(c)
FIG. 2-8c (Repeated)
When the load reaches this value, the downward displacement at point A is
1.0 mm.
Note 1: Since the structure behaves in a linearly elastic manner, the
displacements are proportional to the magnitude of the load. For instance, if the
load is one-half of Pmax, that is, if P 11.6 kN, the downward displacement of
point A is 0.5 mm.
Note 2: To verify our premise that line ABC rotates through a very small
angle, we can calculate the angle of rotation a from the displacement diagram
(Fig. 2-8c), as follows:
AA
dA dCE
tan a
A C
675 mm
(i)
The displacement dA of point A is 1.0 mm, and the elongation dCE of bar CE is
found from Eq. (g) by substituting P 23,200 N; the result is dCE 0.261 mm.
Therefore, from Eq. (i) we get
1.0 mm 0.261 mm
1.261 mm
tan a 0.001868
675 mm
675 mm
from which a 0.11°. This angle is so small that if we tried to draw the
displacement diagram to scale, we would not be able to distinguish between the
original line ABC and the rotated line ABC.
Thus, when working with displacement diagrams, we usually can consider
the displacements to be very small quantities, thereby simplifying the geometry.
In this example we were able to assume that points A, B, and C moved only
vertically, whereas if the displacements were large, we would have to consider
that they moved along curved paths.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.3 Changes in Lengths Under Nonuniform Conditions
77
2.3 CHANGES IN LENGTHS UNDER NONUNIFORM CONDITIONS
When a prismatic bar of linearly elastic material is loaded only at the
ends, we can obtain its change in length from the equation d PL /EA,
as described in the preceding section. In this section we will see how
this same equation can be used in more general situations.
Bars with Intermediate Axial Loads
Suppose, for instance, that a prismatic bar is loaded by one or more axial
loads acting at intermediate points along the axis (Fig. 2-9a). We can
determine the change in length of this bar by adding algebraically the
elongations and shortenings of the individual segments. The procedure
is as follows.
1. Identify the segments of the bar (segments AB, BC, and CD) as
segments 1, 2, and 3, respectively.
2. Determine the internal axial forces N1, N2, and N3 in segments 1, 2,
and 3, respectively, from the free-body diagrams of Figs. 2-9b, c,
and d. Note that the internal axial forces are denoted by the letter N
to distinguish them from the external loads P. By summing forces in
the vertical direction, we obtain the following expressions for the
axial forces:
N1 PB PC PD
N 2 PC PD
N 3 PD
In writing these equations we used the sign convention given in the
preceding section (internal axial forces are positive when in tension
and negative when in compression).
N1
A
PB
L1
B
PB
B
N2
L2
C
C
C
PC
N3
PC
PC
L3
D
D
D
D
FIG. 2-9 (a) Bar with external loads
acting at intermediate points; (b), (c),
and (d) free-body diagrams showing the
internal axial forces N1, N2, and N3
PD
(a)
PD
(b)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
PD
(c)
PD
(d)
78
CHAPTER 2 Axially Loaded Members
3. Determine the changes in the lengths of the segments from Eq. (2-3):
NL
d 1 11
EA
NL
d 2 22
EA
N L3
d 3 3
EA
in which L1, L2, and L3 are the lengths of the segments and EA is the
axial rigidity of the bar.
4. Add d1, d2, and d3 to obtain d, the change in length of the entire bar:
3
d di d1 d 2 d 3
i1
As already explained, the changes in lengths must be added algebraically, with elongations being positive and shortenings negative.
Bars Consisting of Prismatic Segments
PA
This same general approach can be used when the bar consists of several
prismatic segments, each having different axial forces, different dimensions, and different materials (Fig. 2-10). The change in length may be
obtained from the equation
A
L1
E1
n
Ni L i
i1 Ei Ai
PB
(2-5)
B
E2
in which the subscript i is a numbering index for the various segments of
the bar and n is the total number of segments. Note especially that Ni is
not an external load but is the internal axial force in segment i.
L2
Bars with Continuously Varying Loads or Dimensions
C
FIG. 2-10 Bar consisting of prismatic
segments having different axial forces,
different dimensions, and different
materials
Sometimes the axial force N and the cross-sectional area A vary continuously along the axis of a bar, as illustrated by the tapered bar of Fig.
2-11a. This bar not only has a continuously varying cross-sectional area
but also a continuously varying axial force. In this illustration, the load
consists of two parts, a single force PB acting at end B of the bar and
distributed forces p(x) acting along the axis. (A distributed force has units
of force per unit distance, such as pounds per inch or newtons per meter.)
A distributed axial load may be produced by such factors as centrifugal
forces, friction forces, or the weight of a bar hanging in a vertical position.
Under these conditions we can no longer use Eq. (2-5) to obtain the
change in length. Instead, we must determine the change in length of a
differential element of the bar and then integrate over the length of the bar.
We select a differential element at distance x from the left-hand end
of the bar (Fig. 2-11a). The internal axial force N(x) acting at this cross
section (Fig. 2-11b) may be determined from equilibrium using either
segment AC or segment CB as a free body. In general, this force is a
function of x. Also, knowing the dimensions of the bar, we can express
the cross-sectional area A(x) as a function of x.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
79
SECTION 2.3 Changes in Lengths Under Nonuniform Conditions
A
C
B
p(x)
cross-sectional area and
varying axial force
PB
C
p(x)
dx
x
FIG. 2-11 Bar with varying
A
C
N(x)
N(x)
N(x)
x
dx
L
(a)
(b)
(c)
The elongation dd of the differential element (Fig. 2-11c) may be
obtained from the equation d PL /EA by substituting N(x) for P, dx for
L, and A(x) for A, as follows:
N(x) d x
d
EA(x)
(2-6)
The elongation of the entire bar is obtained by integrating over the length:
L
d
L
d
0
0
N(x) d x
E A(x)
(2-7)
If the expressions for N(x) and A(x) are not too complicated, the integral
can be evaluated analytically and a formula for d can be obtained, as
illustrated later in Example 2-4. However, if formal integration is either
difficult or impossible, a numerical method for evaluating the integral
should be used.
Limitations
Equations (2-5) and (2-7) apply only to bars made of linearly elastic
materials, as shown by the presence of the modulus of elasticity E in the
formulas. Also, the formula d PL/EA was derived using the assumption
that the stress distribution is uniform over every cross section (because it is
based on the formula s P/A). This assumption is valid for prismatic bars
but not for tapered bars, and therefore Eq. (2-7) gives satisfactory results
for a tapered bar only if the angle between the sides of the bar is small.
As an illustration, if the angle between the sides of a bar is 20°, the
stress calculated from the expression s P/A (at an arbitrarily selected
cross section) is 3% less than the exact stress for that same cross section
(calculated by more advanced methods). For smaller angles, the error is
even less. Consequently, we can say that Eq. (2-7) is satisfactory if the
angle of taper is small. If the taper is large, more accurate methods of
analysis are needed (Ref. 2-1).
The following examples illustrate the determination of changes in
lengths of nonuniform bars.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
80
CHAPTER 2 Axially Loaded Members
Example 2-3
A vertical steel bar ABC is pin-supported at its upper end and loaded by a force
P1 at its lower end (Fig. 2-12a). A horizontal beam BDE is pinned to the vertical
bar at joint B and supported at point D. The beam carries a load P2 at end E.
The upper part of the vertical bar (segment AB) has length L1 20.0 in.
and cross-sectional area A1 0.25 in.2; the lower part (segment BC) has length
L2 34.8 in. and area A2 0.15 in.2 The modulus of elasticity E of the steel is
29.0 106 psi. The left- and right-hand parts of beam BDE have lengths a
28 in. and b 25 in., respectively.
Calculate the vertical displacement dC at point C if the load Pl 2100 lb
and the load P2 5600 lb. (Disregard the weights of the bar and the beam.)
A
A1
L1
a
b
B
D
E
P2
L2
A2
(a)
C
P1
RA
A
a
B
P3
P3
b
D
RD
(b)
FIG. 2-12 Example 2-3. Change in length
of a nonuniform bar (bar ABC)
E
B
P2
C
P1
(c)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.3 Changes in Lengths Under Nonuniform Conditions
81
Solution
Axial forces in bar ABC. From Fig. 2-12a, we see that the vertical displacement of point C is equal to the change in length of bar ABC. Therefore, we must
find the axial forces in both segments of this bar.
The axial force N2 in the lower segment is equal to the load P1. The axial
force N1 in the upper segment can be found if we know either the vertical reaction
at A or the force applied to the bar by the beam. The latter force can be obtained
from a free-body diagram of the beam (Fig. 2-12b), in which the force acting on
the beam (from the vertical bar) is denoted P3 and the vertical reaction at support
D is denoted RD. No horizontal force acts between the bar and the beam, as can be
seen from a free-body diagram of the vertical bar itself (Fig. 2-12c). Therefore,
there is no horizontal reaction at support D of the beam.
Taking moments about point D for the free-body diagram of the beam
(Fig. 2-12b) gives
(5600 lb)(25.0 in.)
P2b
P3 5000 lb
28.0 in.
a
This force acts downward on the beam (Fig. 2-12b) and upward on the vertical
bar (Fig. 2-12c).
Now we can determine the downward reaction at support A (Fig. 2-12c):
RA P3 P1 5000 lb 2100 lb 2900 lb
The upper part of the vertical bar (segment AB) is subjected to an axial
compressive force N1 equal to RA, or 2900 lb. The lower part (segment BC)
carries an axial tensile force N2 equal to Pl, or 2100 lb.
Note: As an alternative to the preceding calculations, we can obtain the
reaction RA from a free-body diagram of the entire structure (instead of from the
free-body diagram of beam BDE).
Changes in length. With tension considered positive, Eq. (2-5) yields
n
NiLi
N1L1
N2L2
d
EA1
EA2
i1 EiAi
(2900 lb)(20.0 in)
(2100 lb)(34.8 in.)
(29.0 106 psi)(0.25 in.2)
(29.0 106 psi)(0.15 in.2)
0.0080 in. 0.0168 in. 0.0088 in.
in which d is the change in length of bar ABC. Since d is positive, the bar
elongates. The displacement of point C is equal to the change in length of the
bar:
dC 0.0088 in.
This displacement is downward.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
82
CHAPTER 2 Axially Loaded Members
Example 2-4
A tapered bar AB of solid circular cross section and length L (Fig. 2-13a) is
supported at end B and subjected to a tensile load P at the free end A. The diameters of the bar at ends A and B are dA and dB, respectively.
Determine the elongation of the bar due to the load P, assuming that the
angle of taper is small.
x
A
B
A
P
dB
dA
dx B
dA
O
LA
d(x)
dB
L
L
LB
(a)
(b)
FIG. 2-13 Example 2-4. Change in length
of a tapered bar of solid circular cross
section
Solution
The bar being analyzed in this example has a constant axial force (equal to
the load P) throughout its length. However, the cross-sectional area varies
continuously from one end to the other. Therefore, we must use integration (see
Eq. 2-7) to determine the change in length.
Cross-sectional area. The first step in the solution is to obtain an expression for the cross-sectional area A(x) at any cross section of the bar. For this
purpose, we must establish an origin for the coordinate x. One possibility is to
place the origin of coordinates at the free end A of the bar. However, the integrations to be performed will be slightly simplified if we locate the origin of
coordinates by extending the sides of the tapered bar until they meet at point O,
as shown in Fig. 2-13b.
The distances LA and LB from the origin O to ends A and B, respectively,
are in the ratio
LA
dA
LB
dB
(a)
as obtained from similar triangles in Fig. 2-13b. From similar triangles we also
get the ratio of the diameter d(x) at distance x from the origin to the diameter dA
at the small end of the bar:
d(x)
x
dA
LA
or
dAx
d(x)
LA
(b)
Therefore, the cross-sectional area at distance x from the origin is
p d 2A x2
p[d(x)]2
A(x)
4
4L2A
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(c)
SECTION 2.3 Changes in Lengths Under Nonuniform Conditions
83
Change in length. We now substitute the expression for A(x) into Eq. (2-7)
and obtain the elongation d:
N(x)dx
d
EA(x)
LB
LA
4PL2A
Pdx(4 L 2A )
E(p d 2Ax 2) pEd 2A
LB
LA
dx
x2
(d)
By performing the integration (see Appendix C for integration formulas) and
substituting the limits, we get
4PL2A
1
d
pEd 2A
x
4PL 2A 1
1
pEd 2A LA
LB
LA
LB
(e)
This expression for d can be simplified by noting that
LB LA
1
1
L
LA
LB
LALB
LAL B
(f)
Thus, the equation for d becomes
4 P L LA
d
p E d 2A LB
(g)
Finally, we substitute LA/LBdA/dB (see Eq. a) and obtain
4P L
d
pE dAdB
(2-8)
This formula gives the elongation of a tapered bar of solid circular cross section.
By substituting numerical values, we can determine the change in length for any
particular bar.
Note 1: A common mistake is to assume that the elongation of a tapered
bar can be determined by calculating the elongation of a prismatic bar that has
the same cross-sectional area as the midsection of the tapered bar. Examination
of Eq. (2-8) shows that this idea is not valid.
Note 2: The preceding formula for a tapered bar (Eq. 2-8) can be reduced
to the special case of a prismatic bar by substituting dA dB d. The result is
PL
4PL
d 2
pEd
EA
which we know to be correct.
A general formula such as Eq. (2-8) should be checked whenever possible
by verifying that it reduces to known results for special cases. If the reduction
does not produce a correct result, the original formula is in error. If a correct
result is obtained, the original formula may still be incorrect but our confidence
in it increases. In other words, this type of check is a necessary but not sufficient condition for the correctness of the original formula.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
84
CHAPTER 2 Axially Loaded Members
2.4 STATICALLY INDETERMINATE STRUCTURES
The springs, bars, and cables that we discussed in the preceding sections
have one important feature in common—their reactions and internal
forces can be determined solely from free-body diagrams and equations
of equilibrium. Structures of this type are classified as statically
determinate. We should note especially that the forces in a statically
determinate structure can be found without knowing the properties of
the materials. Consider, for instance, the bar AB shown in Fig. 2-14. The
calculations for the internal axial forces in both parts of the bar, as well
as for the reaction R at the base, are independent of the material of
which the bar is made.
Most structures are more complex than the bar of Fig. 2-14, and
their reactions and internal forces cannot be found by statics alone. This
situation is illustrated in Fig. 2-15, which shows a bar AB fixed at both
ends. There are now two vertical reactions (RA and RB) but only one
useful equation of equilibrium—the equation for summing forces in the
vertical direction. Since this equation contains two unknowns, it is not
sufficient for finding the reactions. Structures of this kind are classified
as statically indeterminate. To analyze such structures we must supplement the equilibrium equations with additional equations pertaining to
the displacements of the structure.
To see how a statically indeterminate structure is analyzed, consider
the example of Fig. 2-16a. The prismatic bar AB is attached to rigid supports
at both ends and is axially loaded by a force P at an intermediate point C.
As already discussed, the reactions RA and RB cannot be found by statics
alone, because only one equation of equilibrium is available:
P1
A
P2
B
R
FIG. 2-14 Statically determinate bar
Fvert 0
RA P RB 0
(a)
An additional equation is needed in order to solve for the two unknown
reactions.
The additional equation is based upon the observation that a bar
with both ends fixed does not change in length. If we separate the bar
from its supports (Fig. 2-16b), we obtain a bar that is free at both ends
and loaded by the three forces, RA, RB, and P. These forces cause the bar
to change in length by an amount dAB, which must be equal to zero:
RA
A
P
dAB 0
B
RB
FIG. 2-15 Statically indeterminate bar
(b)
This equation, called an equation of compatibility, expresses the fact
that the change in length of the bar must be compatible with the conditions at the supports.
In order to solve Eqs. (a) and (b), we must now express the compatibility equation in terms of the unknown forces RA and RB. The
relationships between the forces acting on a bar and its changes in
length are known as force-displacement relations. These relations have
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.4 Statically Indeterminate Structures
RA
RA
A
A
P
a
P
various forms depending upon the properties of the material. If the
material is linearly elastic, the equation d PL /EA can be used to
obtain the force-displacement relations.
Let us assume that the bar of Fig. 2-16 has cross-sectional area A
and is made of a material with modulus E. Then the changes in lengths
of the upper and lower segments of the bar are, respectively,
R a
dAC A
EA
C
C
L
B
RB
RB
(a)
(b)
FIG. 2-16 Analysis of a statically
indeterminate bar
R b
dCB B
EA
(c,d)
where the minus sign indicates a shortening of the bar. Equations (c) and
(d) are the force-displacement relations.
We are now ready to solve simultaneously the three sets of equations
(the equation of equilibrium, the equation of compatibility, and the forcedisplacement relations). In this illustration, we begin by combining the
force-displacement relations with the equation of compatibility:
b
B
85
RAa
RBb
dAB dAC dCB 0
EA
EA
(e)
Note that this equation contains the two reactions as unknowns.
The next step is to solve simultaneously the equation of equilibrium
(Eq. a) and the preceding equation (Eq. e). The results are
Pb
RA
L
Pa
RB
L
(2-9a,b)
With the reactions known, all other force and displacement quantities can be determined. Suppose, for instance, that we wish to find the
downward displacement dC of point C. This displacement is equal to the
elongation of segment AC:
R a
Pab
dC dAC A
EA
L EA
(2-10)
Also, we can find the stresses in the two segments of the bar directly
from the internal axial forces (e.g., sACRA/APb/AL).
General Comments
From the preceding discussion we see that the analysis of a statically
indeterminate structure involves setting up and solving equations of
equilibrium, equations of compatibility, and force-displacement relations.
The equilibrium equations relate the loads acting on the structure to the
unknown forces (which may be reactions or internal forces), and the
compatibility equations express conditions on the displacements of
the structure. The force-displacement relations are expressions that use
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
86
CHAPTER 2 Axially Loaded Members
the dimensions and properties of the structural members to relate the
forces and displacements of those members. In the case of axially
loaded bars that behave in a linearly elastic manner, the relations are
based upon the equation d PL /EA. Finally, all three sets of equations
may be solved simultaneously for the unknown forces and displacements.
In the engineering literature, various terms are used for the
conditions expressed by the equilibrium, compatibility, and forcedisplacement equations. The equilibrium equations are also known as
static or kinetic equations; the compatibility equations are sometimes
called geometric equations, kinematic equations, or equations of
consistent deformations; and the force-displacement relations are often
referred to as constitutive relations (because they deal with the
constitution, or physical properties, of the materials).
For the relatively simple structures discussed in this chapter, the
preceding method of analysis is adequate. However, more formalized
approaches are needed for complicated structures. Two commonly
used methods, the flexibility method (also called the force method) and
the stiffness method (also called the displacement method), are
described in detail in textbooks on structural analysis. Even though
these methods are normally used for large and complex structures
requiring the solution of hundreds and sometimes thousands of simultaneous equations, they still are based upon the concepts described
previously, that is, equilibrium equations, compatibility equations, and
force-displacement relations.*
The following two examples illustrate the methodology for
analyzing statically indeterminate structures consisting of axially
loaded members.
*From a historical viewpoint, it appears that Euler in 1774 was the first to analyze a
statically indeterminate system; he considered the problem of a rigid table with four legs
supported on an elastic foundation (Refs. 2-2 and 2-3). The next work was done by the
French mathematician and engineer L. M. H. Navier, who in 1825 pointed out that
statically indeterminate reactions could be found only by taking into account the elasticity
of the structure (Ref. 2-4). Navier solved statically indeterminate trusses and beams.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
87
SECTION 2.4 Statically Indeterminate Structures
Example 2-5
A solid circular steel cylinder S is encased in a hollow circular copper tube C
(Figs. 2-17a and b). The cylinder and tube are compressed between the rigid
plates of a testing machine by compressive forces P. The steel cylinder has
cross-sectional area As and modulus of elasticity Es, the copper tube has area Ac
and modulus Ec, and both parts have length L.
Determine the following quantities: (a) the compressive forces Ps in the
steel cylinder and Pc in the copper tube; (b) the corresponding compressive
stresses ss and sc; and (c) the shortening d of the assembly.
Pc
P
Ps
P
L
C
Ac
As
L
S
Ps
Pc
(b)
(d)
(a)
(c)
FIG. 2-17 Example 2-5. Analysis of a
statically indeterminate structure
Solution
(a) Compressive forces in the steel cylinder and copper tube. We begin by
removing the upper plate of the assembly in order to expose the compressive
forces Ps and Pc acting on the steel cylinder and copper tube, respectively (Fig.
2-17c). The force Ps is the resultant of the uniformly distributed stresses acting
over the cross section of the steel cylinder, and the force Pc is the resultant of
the stresses acting over the cross section of the copper tube.
Equation of equilibrium. A free-body diagram of the upper plate is shown
in Fig. 2-17d. This plate is subjected to the force P and to the unknown
compressive forces Ps and Pc ; thus, the equation of equilibrium is
Fvert 0
Ps Pc P 0
(f)
This equation, which is the only nontrivial equilibrium equation available,
contains two unknowns. Therefore, we conclude that the structure is statically
indeterminate.
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
88
CHAPTER 2 Axially Loaded Members
Pc
P
Ps
P
L
C
Ac
As
L
S
Ps
Pc
(b)
(d)
(a)
(c)
FIG. 2-17 (Repeated)
Equation of compatibility. Because the end plates are rigid, the steel
cylinder and copper tube must shorten by the same amount. Denoting the shortenings of the steel and copper parts by ds and dc, respectively, we obtain the
following equation of compatibility:
(g)
ds dc
Force-displacement relations. The changes in lengths of the cylinder and
tube can be obtained from the general equation d PL /EA. Therefore, in this
example the force-displacement relations are
P L
ds s
Es As
P L
dc c
Ec Ac
(h,i)
Solution of equations. We now solve simultaneously the three sets of equations. First, we substitute the force-displacement relations in the equation of
compatibility, which gives
P L
P L
s c
Es As
Ec Ac
(j)
This equation expresses the compatibility condition in terms of the unknown
forces.
Next, we solve simultaneously the equation of equilibrium (Eq. f) and the
preceding equation of compatibility (Eq. j) and obtain the axial forces in the
steel cylinder and copper tube:
Es As
Ps P
Es As Ec Ac
(2-11a,b)
Ec Ac
Pc P
Es As Ec Ac
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.4 Statically Indeterminate Structures
89
These equations show that the compressive forces in the steel and copper parts
are directly proportional to their respective axial rigidities and inversely proportional to the sum of their rigidities.
(b) Compressive stresses in the steel cylinder and copper tube. Knowing the
axial forces, we can now obtain the compressive stresses in the two materials:
PEs
P
ss s
Es As Ec Ac
As
PEc
Pc
sc
(2-12a,b)
Es As Ec Ac
Ac
Note that the ratio ss /sc of the stresses is equal to the ratio Es /Ec of the moduli
of elasticity, showing that in general the “stiffer” material always has the larger
stress.
(c) Shortening of the assembly. The shortening d of the entire assembly can
be obtained from either Eq. (h) or Eq. (i). Thus, upon substituting the forces
(from Eqs. 2-11a and b), we get
P L
P L
PL
d s c
Es As
Ec Ac
Es As Ec Ac
(2-13)
This result shows that the shortening of the assembly is equal to the total load
divided by the sum of the stiffnesses of the two parts (recall from Eq. 2-4a that
the stiffness of an axially loaded bar is k EA/L).
Alternative solution of the equations. Instead of substituting the forcedisplacement relations (Eqs. h and i) into the equation of compatibility, we
could rewrite those relations in the form
E A
Ps s s ds
L
E A
Pc c c dc
L
(k, l)
and substitute them into the equation of equilibrium (Eq. f):
E A
E A
s s d s c c dc P
L
L
(m)
This equation expresses the equilibrium condition in terms of the unknown
displacements. Then we solve simultaneously the equation of compatibility
(Eq. g) and the preceding equation, thus obtaining the displacements:
PL
ds dc
Es As Ec Ac
(n)
which agrees with Eq. (2-13). Finally, we substitute expression (n) into Eqs. (k)
and (l) and obtain the compressive forces Ps and Pc (see Eqs. 2-11a and b).
Note: The alternative method of solving the equations is a simplified
version of the stiffness (or displacement) method of analysis, and the first
method of solving the equations is a simplified version of the flexibility (or
force) method. The names of these two methods arise from the fact that Eq. (m)
has displacements as unknowns and stiffnesses as coefficients (see Eq. 2-4a),
whereas Eq. (j) has forces as unknowns and flexibilities as coefficients (see
Eq. 2-4b).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
90
CHAPTER 2 Axially Loaded Members
Example 2-6
A horizontal rigid bar AB is pinned at end A and supported by two wires (CD and
EF) at points D and F (Fig. 2-18a). A vertical load P acts at end B of the bar. The
bar has length 3b and wires CD and EF have lengths L 1 and L 2, respectively.
Also, wire CD has diameter d1 and modulus of elasticity E1; wire EF has
diameter d2 and modulus E2.
(a) Obtain formulas for the allowable load P if the allowable stresses in
wires CD and EF, respectively, are s1 and s2. (Disregard the weight of the bar
itself.)
(b) Calculate the allowable load P for the following conditions: Wire CD is
made of aluminum with modulus El 72 GPa, diameter dl 4.0 mm, and
length L l 0.40 m. Wire EF is made of magnesium with modulus E2 45
GPa, diameter d2 3.0 mm, and length L 2 0.30 m. The allowable stresses in
the aluminum and magnesium wires are sl 200 MPa and 2 175 MPa,
respectively.
C
E
L1
A
RH
D
F
L2
A
D
T1
T2
F
B
B
P
RV
(b)
b
b
b
P
(a)
A
D
d1
F
B
d2
B'
FIG. 2-18 Example 2-6. Analysis of a
statically indeterminate structure
(c)
Solution
Equation of equilibrium. We begin the analysis by drawing a free-body
diagram of bar AB (Fig. 2-18b). In this diagram T1 and T2 are the unknown
tensile forces in the wires and RH and RV are the horizontal and vertical
components of the reaction at the support. We see immediately that the structure
is statically indeterminate because there are four unknown forces (Tl, T2, RH,
and RV) but only three independent equations of equilibrium.
Taking moments about point A (with counterclockwise moments being
positive) yields
MA 0
Tl b T2 (2b) 2 P(3b) 0
or
Tl 2T2 3P
(o)
The other two equations, obtained by summing forces in the horizontal direction
and summing forces in the vertical direction, are of no benefit in finding T1
and T2.
Equation of compatibility. To obtain an equation pertaining to the
displacements, we observe that the load P causes bar AB to rotate about the pin
support at A, thereby stretching the wires. The resulting displacements are
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.4 Statically Indeterminate Structures
91
shown in the displacement diagram of Fig. 2-18c, where line AB represents the
original position of the rigid bar and line AB⬘ represents the rotated position.
The displacements d1 and d 2 are the elongations of the wires. Because these
displacements are very small, the bar rotates through a very small angle (shown
highly exaggerated in the figure) and we can make calculations on the assumption
that points D, F, and B move vertically downward (instead of moving along the
arcs of circles).
Because the horizontal distances AD and DF are equal, we obtain the
following geometric relationship between the elongations:
d 2 2d1
(p)
Equation (p) is the equation of compatibility.
Force-displacement relations. Since the wires behave in a linearly elastic
manner, their elongations can be expressed in terms of the unknown forces T1
and T2 by means of the following expressions:
T1 L1
d1
E1 A1
T2 L 2
d2
E2 A2
in which Al and A2 are the cross-sectional areas of wires CD and EF, respectively; that is,
pd2
A1 1
4
pd2
A2 2
4
For convenience in writing equations, let us introduce the following
notation for the flexibilities of the wires (see Eq. 2-4b):
L1
f1
E1 A1
L2
f2
E 2 A2
(q,r)
Then the force-displacement relations become
d 1 f1T1
d 2 f2T2
(s,t)
Solution of equations. We now solve simultaneously the three sets of
equations (equilibrium, compatibility, and force-displacement equations).
Substituting the expressions from Eqs. (s) and (t) into the equation of compatibility
(Eq. p) gives
f2T2 2 f1T1
(u)
The equation of equilibrium (Eq. o) and the preceding equation (Eq. u) each
contain the forces T1 and T2 as unknown quantities. Solving those two equations
simultaneously yields
3f P
T1 2
4f1 f2
6f P
T2 1
4f1 f2
(v,w)
Knowing the forces T1 and T2, we can easily find the elongations of the wires
from the force-displacement relations.
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
92
CHAPTER 2 Axially Loaded Members
(a) Allowable load P. Now that the statically indeterminate analysis is
completed and the forces in the wires are known, we can determine the permissible value of the load P. The stress sl in wire CD and the stress s2 in wire EF
are readily obtained from the forces (Eqs. v and w):
3P
f2
T1
s1
A1 4 f1 f2
A1
6P
f1
T2
s2
A2 4 f1 f2
A2
From the first of these equations we solve for the permissible force Pl based
upon the allowable stress sl in wire CD:
s1 A1(4 f1 f2 )
P1
3 f2
(2-14a)
Similarly, from the second equation we get the permissible force P2 based
upon the allowable stress s2 in wire EF:
s 2 A2 (4 f1 f2 )
P2
6 f1
(2-14b)
The smaller of these two loads is the maximum allowable load Pallow.
(b) Numerical calculations for the allowable load. Using the given data
and the preceding equations, we obtain the following numerical values:
p d 12
p (4.0 mm)2
A1
12.57 mm2
4
4
p d 22
p (3.0 mm)2
A2 7.069 mm2
4
4
L1
0.40 m
f1
0.4420 106 m/N
E1 A1
(72 GPa)(12.57 mm 2 )
L2
0.30 m
f2
0.9431 106 m/N
E2 A2
(45 GPa)(7.069 mm 2 )
Also, the allowable stresses are
s 1 200 MPa
s 2 175 MPa
Therefore, substituting into Eqs. (2-14a and b) gives
P1 2.41 kN
P2 1.26 kN
The first result is based upon the allowable stress s 1 in the aluminum wire and
the second is based upon the allowable stress s 2 in the magnesium wire. The
allowable load is the smaller of the two values:
Pallow 1.26 kN
At this load the stress in the magnesium is 175 MPa (the allowable stress) and
the stress in the aluminum is (1.26/ 2.41)(200 MPa) 105 MPa. As expected,
this stress is less than the allowable stress of 200 MPa.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
93
SECTION 2.5 Thermal Effects, Misfits, and Prestrains
2.5 THERMAL EFFECTS, MISFITS, AND PRESTRAINS
External loads are not the only sources of stresses and strains in a structure.
Other sources include thermal effects arising from temperature changes,
misfits resulting from imperfections in construction, and prestrains that
are produced by initial deformations. Still other causes are settlements
(or movements) of supports, inertial loads resulting from accelerating
motion, and natural phenomenon such as earthquakes.
Thermal effects, misfits, and prestrains are commonly found in both
mechanical and structural systems and are described in this section. As a
general rule, they are much more important in the design of statically
indeterminate structures that in statically determinate ones.
Thermal Effects
A
B
FIG. 2-19 Block of material subjected to
an increase in temperature
Changes in temperature produce expansion or contraction of structural
materials, resulting in thermal strains and thermal stresses. A simple
illustration of thermal expansion is shown in Fig. 2-19, where the block
of material is unrestrained and therefore free to expand. When the block
is heated, every element of the material undergoes thermal strains in all
directions, and consequently the dimensions of the block increase. If we
take corner A as a fixed reference point and let side AB maintain its original
alignment, the block will have the shape shown by the dashed lines.
For most structural materials, thermal strain eT is proportional to the
temperature change ⌬T; that is,
⑀T ( T )
(2-15)
in which a is a property of the material called the coefficient of thermal
expansion. Since strain is a dimensionless quantity, the coefficient of
thermal expansion has units equal to the reciprocal of temperature
change. In SI units the dimensions of a can be expressed as either 1/K
(the reciprocal of kelvins) or 1/°C (the reciprocal of degrees Celsius).
The value of a is the same in both cases because a change in temperature
is numerically the same in both kelvins and degrees Celsius. In USCS
units, the dimensions of a are 1/°F (the reciprocal of degrees Fahrenheit).*
Typical values of a are listed in Table H-4 of Appendix H.
When a sign convention is needed for thermal strains, we usually
assume that expansion is positive and contraction is negative.
To demonstrate the relative importance of thermal strains, we will
compare thermal strains with load-induced strains in the following
manner. Suppose we have an axially loaded bar with longitudinal strains
*For a discussion of temperature units and scales, see Section A.4 of Appendix A.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
94
CHAPTER 2 Axially Loaded Members
given by the equation e s /E, where s is the stress and E is the
modulus of elasticity. Then suppose we have an identical bar subjected
to a temperature change T, which means that the bar has thermal
strains given by Eq. (2-15). Equating the two strains gives the equation
s 5 Ea( T )
From this equation we can calculate the axial stress s that produces the
same strain as does the temperature change T. For instance, consider a
stainless steel bar with E 30 106 psi and a 9.6 106/°F. A
quick calculation from the preceding equation for s shows that a change
in temperature of 100°F produces the same strain as a stress of 29,000
psi. This stress is in the range of typical allowable stresses for stainless
steel. Thus, a relatively modest change in temperature produces strains
of the same magnitude as the strains caused by ordinary loads, which
shows that temperature effects can be important in engineering design.
Ordinary structural materials expand when heated and contract
when cooled, and therefore an increase in temperature produces a positive
thermal strain. Thermal strains usually are reversible, in the sense that
the member returns to its original shape when its temperature returns to
the original value. However, a few special metallic alloys have recently
been developed that do not behave in the customary manner. Instead,
over certain temperature ranges their dimensions decrease when heated
and increase when cooled.
Water is also an unusual material from a thermal standpoint—it
expands when heated at temperatures above 4°C and also expands when
cooled below 4°C. Thus, water has its maximum density at 4°C.
Now let us return to the block of material shown in Fig. 2-19. We
assume that the material is homogeneous and isotropic and that the
temperature increase T is uniform throughout the block. We can calculate
the increase in any dimension of the block by multiplying the original
dimension by the thermal strain. For instance, if one of the dimensions
is L, then that dimension will increase by the amount
d T eT L a( T )L
L
∆T
dT
FIG. 2-20 Increase in length of a
prismatic bar due to a uniform increase
in temperature (Eq. 2-16)
(2-16)
Equation (2-16) is a temperature-displacement relation, analogous to
the force-displacement relations described in the preceding section. It
can be used to calculate changes in lengths of structural members
subjected to uniform temperature changes, such as the elongation dT of
the prismatic bar shown in Fig. 2-20. (The transverse dimensions of the
bar also change, but these changes are not shown in the figure since they
usually have no effect on the axial forces being transmitted by the bar.)
In the preceding discussions of thermal strains, we assumed that the
structure had no restraints and was able to expand or contract freely.
These conditions exist when an object rests on a frictionless surface or
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.5 Thermal Effects, Misfits, and Prestrains
A
∆T1
B
C
∆T2
FIG. 2-21 Statically determinate truss
with a uniform temperature change in
each member
FIG. 2-22 Statically indeterminate truss
95
hangs in open space. In such cases no stresses are produced by a
uniform temperature change throughout the object, although nonuniform
temperature changes may produce internal stresses. However, many
structures have supports that prevent free expansion and contraction, in
which case thermal stresses will develop even when the temperature
change is uniform throughout the structure.
To illustrate some of these ideas about thermal effects, consider the
two-bar truss ABC of Fig. 2-21 and assume that the temperature of bar
AB is changed by ⌬T1 and the temperature of bar BC is changed by ⌬T2.
Because the truss is statically determinate, both bars are free to lengthen
or shorten, resulting in a displacement of joint B. However, there are no
stresses in either bar and no reactions at the supports. This conclusion
applies generally to statically determinate structures; that is, uniform
temperature changes in the members produce thermal strains (and the
corresponding changes in lengths) without producing any corresponding
stresses.
B
C
A
D
subjected to temperature changes
A statically indeterminate structure may or may not develop
temperature stresses, depending upon the character of the structure and
the nature of the temperature changes. To illustrate some of the possibilities,
consider the statically indeterminate truss shown in Fig. 2-22. Because
the supports of this structure permit joint D to move horizontally, no
stresses are developed when the entire truss is heated uniformly. All
members increase in length in proportion to their original lengths, and
the truss becomes slightly larger in size.
However, if some bars are heated and others are not, thermal
stresses will develop because the statically indeterminate arrangement
of the bars prevents free expansion. To visualize this condition, imagine
that just one bar is heated. As this bar becomes longer, it meets
resistance from the other bars, and therefore stresses develop in all
members.
The analysis of a statically indeterminate structure with temperature
changes is based upon the concepts discussed in the preceding section,
namely equilibrium equations, compatibility equations, and displacement
relations. The principal difference is that we now use temperaturedisplacement relations (Eq. 2-16) in addition to force-displacement
relations (such as d PL/EA) when performing the analysis. The
following two examples illustrate the procedures in detail.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
96
CHAPTER 2 Axially Loaded Members
Example 2-7
A prismatic bar AB of length L is held between immovable supports (Fig. 2-23a).
If the temperature of the bar is raised uniformly by an amount ⌬T, what thermal
stress sT is developed in the bar? (Assume that the bar is made of linearly
elastic material.)
RA
RA
dT
A
A
∆T
L
B
FIG. 2-23 Example 2-7. Statically
dR
A
∆T
B
B
RB
indeterminate bar with uniform
temperature increase T
(a)
(b)
(c)
Solution
Because the temperature increases, the bar tends to elongate but is restrained
by the rigid supports at A and B. Therefore, reactions RA and RB are developed at
the supports, and the bar is subjected to uniform compressive stresses.
Equation of equilibrium. The only forces acting on the bar are the reactions
shown in Fig. 2-23a. Therefore, equilibrium of forces in the vertical direction
gives
Fvert 0
RB RA 0
(a)
Since this is the only nontrivial equation of equilibrium, and since it contains
two unknowns, we see that the structure is statically indeterminate and an additional equation is needed.
Equation of compatibility. The equation of compatibility expresses the fact
that the change in length of the bar is zero (because the supports do not move):
dAB 0
(b)
To determine this change in length, we remove the upper support of the bar and
obtain a bar that is fixed at the base and free to displace at the upper end (Figs.
2-23b and c). When only the temperature change is acting (Fig. 2-23b), the bar
elongates by an amount d T , and when only the reaction RA is acting, the bar
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.5 Thermal Effects, Misfits, and Prestrains
97
shortens by an amount dR (Fig. 2-23c). Thus, the net change in length is dAB
d T dR, and the equation of compatibility becomes
dAB d T d R 0
(c)
Displacement relations. The increase in length of the bar due to the
temperature change is given by the temperature-displacement relation (Eq.
2-16):
(d)
d T a( T)L
in which a is the coefficient of thermal expansion. The decrease in length due to
the force RA is given by the force-displacement relation:
R L
d R A
EA
(e)
in which E is the modulus of elasticity and A is the cross-sectional area.
Solution of equations. Substituting the displacement relations (d) and (e)
into the equation of compatibility (Eq. c) gives the following equation:
R L
d T d R a( T)L A 0
EA
(f)
We now solve simultaneously the preceding equation and the equation of equilibrium (Eq. a) for the reactions RA and RB:
(2-17)
RA RB EAa( T)
From these results we obtain the thermal stress sT in the bar:
RA
RB
Ea( T)
sT
A
A
(2-18)
This stress is compressive when the temperature of the bar increases.
Note 1: In this example the reactions are independent of the length of the
bar and the stress is independent of both the length and the cross-sectional area
(see Eqs. 2-17 and 2-18). Thus, once again we see the usefulness of a symbolic
solution, because these important features of the bar’s behavior might not be
noticed in a purely numerical solution.
Note 2: When determining the thermal elongation of the bar (Eq. d), we
assumed that the material was homogeneous and that the increase in temperature was uniform throughout the volume of the bar. Also, when determining the
decrease in length due to the reactive force (Eq. e), we assumed linearly elastic
behavior of the material. These limitations should always be kept in mind when
writing equations such as Eqs. (d) and (e).
Note 3: The bar in this example has zero longitudinal displacements, not
only at the fixed ends but also at every cross section. Thus, there are no axial
strains in this bar, and we have the special situation of longitudinal stresses
without longitudinal strains. Of course, there are transverse strains in the bar,
from both the temperature change and the axial compression.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
98
CHAPTER 2 Axially Loaded Members
Example 2-8
A sleeve in the form of a circular tube of length L is placed around a bolt and
fitted between washers at each end (Fig. 2-24a). The nut is then turned until it is
just snug. The sleeve and bolt are made of different materials and have different
cross-sectional areas. (Assume that the coefficient of thermal expansion aS of
the sleeve is greater than the coefficient aB of the bolt.)
(a) If the temperature of the entire assembly is raised by an amount ⌬T,
what stresses sS and sB are developed in the sleeve and bolt, respectively?
(b) What is the increase d in the length L of the sleeve and bolt?
Nut
Sleeve
Washer
Bolt head
Bolt
(a)
L
d1
d2
∆T
(b)
d
d4
d3
PB PS
FIG. 2-24 Example 2-8. Sleeve and bolt
assembly with uniform temperature
increase ⌬T
(c)
Solution
Because the sleeve and bolt are of different materials, they will elongate by
different amounts when heated and allowed to expand freely. However, when
they are held together by the assembly, free expansion cannot occur and thermal
stresses are developed in both materials. To find these stresses, we use the same
concepts as in any statically indeterminate analysis—equilibrium equations,
compatibility equations, and displacement relations. However, we cannot
formulate these equations until we disassemble the structure.
A simple way to cut the structure is to remove the head of the bolt, thereby
allowing the sleeve and bolt to expand freely under the temperature change ⌬T
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.5 Thermal Effects, Misfits, and Prestrains
99
(Fig. 2-24b). The resulting elongations of the sleeve and bolt are denoted d1 and
d 2, respectively, and the corresponding temperature-displacement relations are
d1 aS ( T)L d 2 aB( T )L
(g,h)
Since aS is greater than aB, the elongation d1 is greater than d2, as shown in
Fig. 2-24b.
The axial forces in the sleeve and bolt must be such that they shorten the
sleeve and stretch the bolt until the final lengths of the sleeve and bolt are the
same. These forces are shown in Fig. 2-24c, where PS denotes the compressive
force in the sleeve and PB denotes the tensile force in the bolt. The corresponding shortening d 3 of the sleeve and elongation d4 of the bolt are
P L
d 3 S
ES AS
P L
d 4 B
EB AB
(i,j)
in which ES AS and EB AB are the respective axial rigidities. Equations (i) and (j)
are the load-displacement relations.
Now we can write an equation of compatibility expressing the fact that the
final elongation d is the same for both the sleeve and bolt. The elongation of the
sleeve is d 1 d 3 and of the bolt is d 2 d4; therefore,
d d 1 d 3 d 2 d4
(k)
Substituting the temperature-displacement and load-displacement relations
(Eqs. g to j) into this equation gives
P L
P L
d aS( T )L S aB( T )L B
ES AS
EB AB
(l)
from which we get
P L
P L
S B aS( T )L aB( T)L
ES AS
EB AB
(m)
which is a modified form of the compatibility equation. Note that it contains the
forces PS and PB as unknowns.
An equation of equilibrium is obtained from Fig. 2-24c, which is a freebody diagram of the part of the assembly remaining after the head of the bolt is
removed. Summing forces in the horizontal direction gives
PS PB
(n)
which expresses the obvious fact that the compressive force in the sleeve is
equal to the tensile force in the bolt.
We now solve simultaneously Eqs. (m) and (n) and obtain the axial forces
in the sleeve and bolt:
(aS aB)( T )ES AS EB AB
PS PB
ES AS EB AB
(2-19)
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
100
CHAPTER 2 Axially Loaded Members
When deriving this equation, we assumed that the temperature increased and
that the coefficient aS was greater than the coefficient aB. Under these conditions,
PS is the compressive force in the sleeve and PB is the tensile force in the bolt.
The results will be quite different if the temperature increases but the
coefficient aS is less than the coefficient aB. Under these conditions, a gap will
open between the bolt head and the sleeve and there will be no stresses in either
part of the assembly.
(a) Stresses in the sleeve and bolt. Expressions for the stresses sS and sB in
the sleeve and bolt, respectively, are obtained by dividing the corresponding
forces by the appropriate areas:
PS (a S a B )( T )E S E B AB
sS
ES AS EB AB
AS
(2-20a)
(aS aB)( T )ES AS EB
PB
sB
ES AS EB AB
AB
(2-20b)
Under the assumed conditions, the stress sS in the sleeve is compressive and the
stress sB in the bolt is tensile. It is interesting to note that these stresses are
independent of the length of the assembly and their magnitudes are inversely
proportional to their respective areas (that is, sS /sB AB /AS).
(b) Increase in length of the sleeve and bolt. The elongation d of the
assembly can be found by substituting either PS or PB from Eq. (2-19) into Eq.
(l), yielding
(aS ES AS aB EB AB)( T )L
d
ES AS EB AB
(2-21)
With the preceding formulas available, we can readily calculate the forces,
stresses, and displacements of the assembly for any given set of numerical data.
Note: As a partial check on the results, we can see if Eqs. (2-19), (2-20),
and (2-21) reduce to known values in simplified cases. For instance, suppose
that the bolt is rigid and therefore unaffected by temperature changes. We can
represent this situation by setting aB 0 and letting EB become infinitely large,
thereby creating an assembly in which the sleeve is held between rigid supports.
Substituting these values into Eqs. (2-19), (2-20), and (2-21), we find
PS ES AS aS( T )
sS ESaS ( T)
d0
These results agree with those of Example 2-7 for a bar held between rigid
supports (compare with Eqs. 2-17 and 2-18, and with Eq. b).
As a second special case, suppose that the sleeve and bolt are made of the
same material. Then both parts will expand freely and will lengthen the same
amount when the temperature changes. No forces or stresses will be developed.
To see if the derived equations predict this behavior, we substitute S B
into Eqs. (2-19), (2-20), and (2-21) and obtain
PS PB 0
sS sB 0
which are the expected results.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
d a( T )L
SECTION 2.5 Thermal Effects, Misfits, and Prestrains
101
Misfits and Prestrains
C
L
A
B
D
(a)
C
L
A
B
D
P
(b)
FIG. 2-25 Statically determinate structure
with a small misfit
C
E
L
L
A
B
F
D
(a)
C
E
L
L
A
B
F
D
P
(b)
FIG. 2-26 Statically indeterminate
structure with a small misfit
Suppose that a member of a structure is manufactured with its length
slightly different from its prescribed length. Then the member will not
fit into the structure in its intended manner, and the geometry of the
structure will be different from what was planned. We refer to situations
of this kind as misfits. Sometimes misfits are intentionally created in
order to introduce strains into the structure at the time it is built.
Because these strains exist before any loads are applied to the structure,
they are called prestrains. Accompanying the prestrains are prestresses,
and the structure is said to be prestressed. Common examples of
prestressing are spokes in bicycle wheels (which would collapse if not
prestressed), the pretensioned faces of tennis racquets, shrink-fitted
machine parts, and prestressed concrete beams.
If a structure is statically determinate, small misfits in one or more
members will not produce strains or stresses, although there will be
departures from the theoretical configuration of the structure. To illustrate
this statement, consider a simple structure consisting of a horizontal
beam AB supported by a vertical bar CD (Fig. 2-25a). If bar CD has
exactly the correct length L, the beam will be horizontal at the time the
structure is built. However, if the bar is slightly longer than intended, the
beam will make a small angle with the horizontal. Nevertheless, there
will be no strains or stresses in either the bar or the beam attributable to
the incorrect length of the bar. Furthermore, if a load P acts at the end of
the beam (Fig. 2-25b), the stresses in the structure due to that load will
be unaffected by the incorrect length of bar CD.
In general, if a structure is statically determinate, the presence of
small misfits will produce small changes in geometry but no strains or
stresses. Thus, the effects of a misfit are similar to those of a temperature
change.
The situation is quite different if the structure is statically indeterminate, because then the structure is not free to adjust to misfits (just as
it is not free to adjust to certain kinds of temperature changes). To show
this, consider a beam supported by two vertical bars (Fig. 2-26a). If both
bars have exactly the correct length L, the structure can be assembled
with no strains or stresses and the beam will be horizontal.
Suppose, however, that bar CD is slightly longer than the prescribed
length. Then, in order to assemble the structure, bar CD must be
compressed by external forces (or bar EF stretched by external forces),
the bars must be fitted into place, and then the external forces must be
released. As a result, the beam will deform and rotate, bar CD will be in
compression, and bar EF will be in tension. In other words, prestrains
will exist in all members and the structure will be prestressed, even
though no external loads are acting. If a load P is now added (Fig. 2-26b),
additional strains and stresses will be produced.
The analysis of a statically indeterminate structure with misfits and
prestrains proceeds in the same general manner as described previously
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
102
CHAPTER 2 Axially Loaded Members
for loads and temperature changes. The basic ingredients of the analysis
are equations of equilibrium, equations of compatibility, force-displacement
relations, and (if appropriate) temperature-displacement relations. The
methodology is illustrated in Example 2-9.
Bolts and Turnbuckles
Prestressing a structure requires that one or more parts of the structure
be stretched or compressed from their theoretical lengths. A simple way
to produce a change in length is to tighten a bolt or a turnbuckle. In the
case of a bolt (Fig. 2-27) each turn of the nut will cause the nut to travel
along the bolt a distance equal to the spacing p of the threads (called the
pitch of the threads). Thus, the distance d traveled by the nut is
(2-22)
d np
in which n is the number of revolutions of the nut (not necessarily an
integer). Depending upon how the structure is arranged, turning the nut
can stretch or compress a member.
p
FIG. 2-27 The pitch of the threads is the
distance from one thread to the next
In the case of a double-acting turnbuckle (Fig. 2-28), there are two
end screws. Because a right-hand thread is used at one end and a lefthand thread at the other, the device either lengthens or shortens when
the buckle is rotated. Each full turn of the buckle causes it to travel a
distance p along each screw, where again p is the pitch of the threads.
Therefore, if the turnbuckle is tightened by one turn, the screws are
drawn closer together by a distance 2p and the effect is to shorten the
device by 2p. For n turns, we have
d 2np
(2-23)
Turnbuckles are often inserted in cables and then tightened, thus
creating initial tension in the cables, as illustrated in the following
example.
FIG. 2-28 Double-acting turnbuckle.
(Each full turn of the turnbuckle
shortens or lengthens the cable by 2p,
where p is the pitch of the screw
threads.)
P
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
P
103
SECTION 2.5 Thermal Effects, Misfits, and Prestrains
Example 2-9
The mechanical assembly shown in Fig. 2-29a consists of a copper tube, a rigid
end plate, and two steel cables with turnbuckles. The slack is removed from the
cables by rotating the turnbuckles until the assembly is snug but with no initial
stresses. (Further tightening of the turnbuckles will produce a prestressed
condition in which the cables are in tension and the tube is in compression.)
(a) Determine the forces in the tube and cables (Fig. 2-29a) when the turnbuckles are tightened by n turns.
(b) Determine the shortening of the tube.
Copper tube
Steel cable
Rigid
plate
Turnbuckle
(a)
L
d1
(b)
d1
d2
FIG. 2-29 Example 2-9. Statically
indeterminate assembly with a copper
tube in compression and two steel cables
in tension
(c)
d3
Ps
Pc
Ps
Solution
We begin the analysis by removing the plate at the right-hand end of the
assembly so that the tube and cables are free to change in length (Fig. 2-29b).
Rotating the turnbuckles through n turns will shorten the cables by a distance
d1 2np
(o)
as shown in Fig. 2-29b.
The tensile forces in the cables and the compressive force in the tube must
be such that they elongate the cables and shorten the tube until their final
lengths are the same. These forces are shown in Fig. 2-29c, where Ps denotes
the tensile force in one of the steel cables and Pc denotes the compressive force
in the copper tube. The elongation of a cable due to the force Ps is
PL
d 2 s
Es As
(p)
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
104
CHAPTER 2 Axially Loaded Members
in which Es As is the axial rigidity and L is the length of a cable. Also, the
compressive force Pc in the copper tube causes it to shorten by
PL
d 3 c
Ec Ac
(q)
in which Ec Ac is the axial rigidity of the tube. Equations (p) and (q) are the
load-displacement relations.
The final shortening of one of the cables is equal to the shortening d1
caused by rotating the turnbuckle minus the elongation d 2 caused by the force
Ps. This final shortening of the cable must equal the shortening d 3 of the tube:
(r)
d1 d 2 d 3
which is the equation of compatibility.
Substituting the turnbuckle relation (Eq. o) and the load-displacement
relations (Eqs. p and q) into the preceding equation yields
PL
P L
2np s c
Es As
Ec Ac
(s)
PL
PL
s c 2np
Es As
Ec Ac
(t)
or
which is a modified form of the compatibility equation. Note that it contains Ps
and Pc as unknowns.
From Fig. 2-29c, which is a free-body diagram of the assembly with the
end plate removed, we obtain the following equation of equilibrium:
(u)
2Ps Pc
(a) Forces in the cables and tube. Now we solve simultaneously Eqs. (t)
and (u) and obtain the axial forces in the steel cables and copper tube,
respectively:
2npE c A c E s A s
Ps
L(E c A c 2E s A s )
4npE c A c E s A s
Pc
L (E c A c 2E s A s )
(2-24a,b)
Recall that the forces Ps are tensile forces and the force Pc is compressive. If
desired, the stresses ss and sc in the steel and copper can now be obtained by
dividing the forces Ps and Pc by the cross-sectional areas As and Ac, respectively.
(b) Shortening of the tube. The decrease in length of the tube is the quantity d 3 (see Fig. 2-29 and Eq. q):
4npEs As
PL
d 3 c
Ec Ac 2Es As
Ec Ac
(2-25)
With the preceding formulas available, we can readily calculate the forces,
stresses, and displacements of the assembly for any given set of numerical data.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.6 Stresses on Inclined Sections
105
2.6 STRESSES ON INCLINED SECTIONS
In our previous discussions of tension and compression in axially loaded
members, the only stresses we considered were the normal stresses
acting on cross sections. These stresses are pictured in Fig. 2-30, where
we consider a bar AB subjected to axial loads P.
When the bar is cut at an intermediate cross section by a plane mn
(perpendicular to the x axis), we obtain the free-body diagram shown in
Fig. 2-30b. The normal stresses acting over the cut section may be
calculated from the formula sx P/A provided that the stress distribution is uniform over the entire cross-sectional area A. As explained in
Chapter 1, this condition exists if the bar is prismatic, the material is
homogeneous, the axial force P acts at the centroid of the crosssectional area, and the cross section is away from any localized stress
concentrations. Of course, there are no shear stresses acting on the cut
section, because it is perpendicular to the longitudinal axis of the bar.
For convenience, we usually show the stresses in a two-dimensional
view of the bar (Fig. 2-30c) rather than the more complex threedimensional view (Fig. 2-30b). However, when working with
y
P
m
O
P
x
z
A
B
n
(a)
y
P
O
sx = P
A
x
z
A
(b)
y
FIG. 2-30 Prismatic bar in tension
showing the stresses acting on cross
section mn: (a) bar with axial forces P,
(b) three-dimensional view of the cut
bar showing the normal stresses, and
(c) two-dimensional view
m
P
O
A
x
sx =
C
n
(c)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
P
A
106
CHAPTER 2 Axially Loaded Members
two-dimensional figures we must not forget that the bar has a thickness
perpendicular to the plane of the figure. This third dimension must be
considered when making derivations and calculations.
Stress Elements
The most useful way of representing the stresses in the bar of Fig. 2-30
is to isolate a small element of material, such as the element labeled C in
Fig. 2-30c, and then show the stresses acting on all faces of this element.
An element of this kind is called a stress element. The stress element at
point C is a small rectangular block (it doesn’t matter whether it is a
cube or a rectangular parallelepiped) with its right-hand face lying in
cross section mn.
The dimensions of a stress element are assumed to be infinitesimally small, but for clarity we draw the element to a large scale, as in
Fig. 2-31a. In this case, the edges of the element are parallel to the x, y,
and z axes, and the only stresses are the normal stresses sx acting on the
x faces (recall that the x faces have their normals parallel to the x axis).
Because it is more convenient, we usually draw a two-dimensional view
of the element (Fig. 2-31b) instead of a three-dimensional view.
Stresses on Inclined Sections
The stress element of Fig. 2-31 provides only a limited view of the
stresses in an axially loaded bar. To obtain a more complete picture, we
need to investigate the stresses acting on inclined sections, such as the
section cut by the inclined plane pq in Fig. 2-32a. Because the stresses
are the same throughout the entire bar, the stresses acting over the
inclined section must be uniformly distributed, as pictured in the freebody diagrams of Fig. 2-32b (three-dimensional view) and Fig. 2-32c
(two-dimensional view). From the equilibrium of the free body we know
that the resultant of the stresses must be a horizontal force P. (The resultant is drawn with a dashed line in Figs. 2-32b and 2-32c.)
y
y
P
sx = A
sx = P
A
O
FIG. 2-31 Stress element at point C of the
axially loaded bar shown in Fig. 2-30c:
(a) three-dimensional view of the
element, and (b) two-dimensional view
of the element
x
sx
sx
x
O
z
(a)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(b)
SECTION 2.6 Stresses on Inclined Sections
107
y
p
P
O
P
x
z
A
B
q
(a)
y
P
O
P
x
z
A
(b)
y
p
FIG. 2-32 Prismatic bar in tension
showing the stresses acting on an
inclined section pq: (a) bar with axial
forces P, (b) three-dimensional view of
the cut bar showing the stresses, and
(c) two-dimensional view
P
O
P
x
q
A
(c)
As a preliminary matter, we need a scheme for specifying the orientation of the inclined section pq. A standard method is to specify the
angle u between the x axis and the normal n to the section (see Fig.
2-33a on the next page). Thus, the angle u for the inclined section shown
in the figure is approximately 30°. By contrast, cross section mn (Fig.
2-30a) has an angle u equal to zero (because the normal to the section is
the x axis). For additional examples, consider the stress element of Fig.
2-31. The angle u for the right-hand face is 0, for the top face is 90°
(a longitudinal section of the bar), for the left-hand face is 180°, and for
the bottom face is 270° (or ⫺90°).
Let us now return to the task of finding the stresses acting on
section pq (Fig. 2-33b). As already mentioned, the resultant of these
stresses is a force P acting in the x direction. This resultant may be
resolved into two components, a normal force N that is perpendicular to
the inclined plane pq and a shear force V that is tangential to it. These
force components are
N P cos u
V P sin u
(2-26a,b)
Associated with the forces N and V are normal and shear stresses that are
uniformly distributed over the inclined section (Figs. 2-33c and d). The
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
108
CHAPTER 2 Axially Loaded Members
y
n
p
u
P
O
P
x
A
B
q
(a)
y
P
O
p
N
x
u
P
A
V
q
(b)
p
N
su = —
A1
A
A1 =
A
cos u
q
(c)
p
V
tu = – —
A1
FIG. 2-33 Prismatic bar in tension
showing the stresses acting on an
inclined section pq
A
A1 =
A
cos u
q
(d)
normal stress is equal to the normal force N divided by the area of the
section, and the shear stress is equal to the shear force V divided by the
area of the section. Thus, the stresses are
N
s
A1
V
t
A1
(2-27a,b)
in which A1 is the area of the inclined section, as follows:
A
A1
cos u
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(2-28)
109
SECTION 2.6 Stresses on Inclined Sections
As usual, A represents the cross-sectional area of the bar. The stresses s
and t act in the directions shown in Figs. 2-33c and d, that is, in the
same directions as the normal force N and shear force V, respectively.
At this point we need to establish a standardized notation and sign
convention for stresses acting on inclined sections. We will use a
subscript u to indicate that the stresses act on a section inclined at an
angle u (Fig. 2-34), just as we use a subscript x to indicate that the
stresses act on a section perpendicular to the x axis (see Fig. 2-30).
Normal stresses su are positive in tension and shear stresses tu are
positive when they tend to produce counterclockwise rotation of the
material, as shown in Fig. 2-34.
y
FIG. 2-34 Sign convention for stresses
acting on an inclined section. (Normal
stresses are positive when in tension and
shear stresses are positive when they
tend to produce counterclockwise
rotation.)
tu
P
O
su
u
x
For a bar in tension, the normal force N produces positive normal
stresses su (see Fig. 2-33c) and the shear force V produces negative
shear stresses tu (see Fig. 2-33d). These stresses are given by the
following equations (see Eqs. 2-26, 2-27, and 2-28):
N
P
su cos2u
A1
A
V
P
tu sinu cos u
A1
A
Introducing the notation sx P/A, in which sx is the normal stress on a
cross section, and also using the trigonometric relations
1
cos2u (1 cos 2u)
2
1
sinu cos u (sin 2u)
2
we get the following expressions for the normal and shear stresses:
x
cos2 x (1 cos 2)
2
x
sin cos x (sin 2)
2
(2-29a)
(2-29b)
These equations give the stresses acting on an inclined section oriented
at an angle u to the x axis (Fig. 2-34).
It is important to recognize that Eqs. (2-29a) and (2-29b) were
derived only from statics, and therefore they are independent of the
material. Thus, these equations are valid for any material, whether it
behaves linearly or nonlinearly, elastically or inelastically.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
110
CHAPTER 2 Axially Loaded Members
su or tu
sx
su
0.5sx
–90°
FIG. 2-35 Graph of normal stress s u and
shear stress tu versus angle u of the
inclined section (see Fig. 2-34 and
Eqs. 2-29a and b)
–45°
0
tu
45°
u
90°
–0.5sx
Maximum Normal and Shear Stresses
The manner in which the stresses vary as the inclined section is cut at
various angles is shown in Fig. 2-35. The horizontal axis gives the angle
u as it varies from ⫺90° to 90°, and the vertical axis gives the stresses
su and tu. Note that a positive angle u is measured counterclockwise
from the x axis (Fig. 2-34) and a negative angle is measured clockwise.
As shown on the graph, the normal stress su equals sx when u 0.
Then, as u increases or decreases, the normal stress diminishes until at
u 90° it becomes zero, because there are no normal stresses on
sections cut parallel to the longitudinal axis. The maximum normal
stress occurs at u 0 and is
smax sx
(2-30)
Also, we note that when u 45°, the normal stress is one-half the
maximum value.
The shear stress tu is zero on cross sections of the bar (u 0) as
well as on longitudinal sections (u 90°). Between these extremes,
the stress varies as shown on the graph, reaching the largest positive
value when u 45° and the largest negative value when u 45°.
These maximum shear stresses have the same magnitude:
s
tmax x
2
(2-31)
but they tend to rotate the element in opposite directions.
The maximum stresses in a bar in tension are shown in Fig. 2-36.
Two stress elements are selected—element A is oriented at u 0° and
element B is oriented at u 45°. Element A has the maximum normal
stresses (Eq. 2-30) and element B has the maximum shear stresses (Eq.
2-31). In the case of element A (Fig. 2-36b), the only stresses are the
maximum normal stresses (no shear stresses exist on any of the faces).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.6 Stresses on Inclined Sections
111
y
P
O
x
A
P
B
(a)
sx
2
sx
2
u = 45°
y
sx
O
y
x
sx
O
x
t max =
A
sx
2
FIG. 2-36 Normal and shear stresses
acting on stress elements oriented at
u 0° and u 45° for a bar in tension
B
(b)
sx
2
sx
2
(c)
In the case of element B (Fig. 2-36c), both normal and shear stresses
act on all faces (except, of course, the front and rear faces of the
element). Consider, for instance, the face at 45° (the upper right-hand
face). On this face the normal and shear stresses (from Eqs. 2-29a and b)
are sx /2 and ⫺sx /2, respectively. Hence, the normal stress is tension
(positive) and the shear stress acts clockwise (negative) against the
element. The stresses on the remaining faces are obtained in a similar
manner by substituting u 135°, ⫺45°, and ⫺135° into Eqs. (2-29a
and b).
Thus, in this special case of an element oriented at u 45°, the
normal stresses on all four faces are the same (equal to sx /2) and all four
shear stresses have the maximum magnitude (equal to sx /2). Also, note
that the shear stresses acting on perpendicular planes are equal in magnitude and have directions either toward, or away from, the line of
intersection of the planes, as discussed in detail in Section 1.6.
If a bar is loaded in compression instead of tension, the stress sx
will be compression and will have a negative value. Consequently, all
stresses acting on stress elements will have directions opposite to those
for a bar in tension. Of course, Eqs. (2-29a and b) can still be used for
the calculations simply by substituting sx as a negative quantity.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
112
CHAPTER 2 Axially Loaded Members
Load
FIG. 2-37 Shear failure along a 45° plane
of a wood block loaded in compression
Load
Even though the maximum shear stress in an axially loaded bar is only
one-half the maximum normal stress, the shear stress may cause failure if
the material is much weaker in shear than in tension. An example of a
shear failure is pictured in Fig. 2-37, which shows a block of wood that
was loaded in compression and failed by shearing along a 45° plane.
A similar type of behavior occurs in mild steel loaded in tension.
During a tensile test of a flat bar of low-carbon steel with polished
surfaces, visible slip bands appear on the sides of the bar at approximately 45° to the axis (Fig. 2-38). These bands indicate that the material
is failing in shear along the planes on which the shear stress is
maximum. Such bands were first observed by G. Piobert in 1842 and W.
Lüders in 1860 (see Refs. 2-5 and 2-6), and today they are called either
Lüders’ bands or Piobert’s bands. They begin to appear when the yield
stress is reached in the bar (point B in Fig. 1-10 of Section 1.3).
Load
Uniaxial Stress
Load
FIG. 2-38 Slip bands (or Lüders’ bands) in a
polished steel specimen loaded in tension
The state of stress described throughout this section is called uniaxial
stress, for the obvious reason that the bar is subjected to simple tension
or compression in just one direction. The most important orientations of
stress elements for uniaxial stress are u 0 and u 45° (Fig. 2-36b and
c); the former has the maximum normal stress and the latter has the
maximum shear stress. If sections are cut through the bar at other
angles, the stresses acting on the faces of the corresponding stress
elements can be determined from Eqs. (2-29a and b), as illustrated in
Examples 2-10 and 2-11 that follow.
Uniaxial stress is a special case of a more general stress state known
as plane stress, which is described in detail in Chapter 7.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
113
SECTION 2.6 Stresses on Inclined Sections
Example 2-10
A prismatic bar having cross-sectional area A 1200 mm2 is compressed by an
axial load P 90 kN (Fig. 2-39a).
(a) Determine the stresses acting on an inclined section pq cut through the
bar at an angle u 25°.
(b) Determine the complete state of stress for u 25° and show the
stresses on a properly oriented stress element.
y
p
u = 25°
P
O
P = 90 kN
x
13.4 MPa
28.7 MPa
28.7 MPa
b
61.6 MPa
q
c
(a)
25°
28.7 MPa
a
28.7 MPa
P
61.6 MPa
25°
61.6 MPa
28.7 MPa
d
13.4 MPa
(b)
(c)
FIG. 2-39 Example 2-10. Stresses on an
inclined section
Solution
(a) Stresses on the inclined section. To find the stresses acting on a section
at u 25°, we first calculate the normal stress sx acting on a cross section:
90 kN
P
sx
75 MPa
1200 mm2
A
where the minus sign indicates that the stress is compressive. Next, we calculate
the normal and shear stresses from Eqs. (2-29a and b) with u 25°, as follows:
su sx cos2 u (75 MPa)(cos 25°)2 61.6 MPa
tu sx sin u cos u (75 MPa)(sin 25°)(cos 25°) 28.7 MPa
These stresses are shown acting on the inclined section in Fig. 2-39b. Note that
the normal stress su is negative (compressive) and the shear stress tu is positive
(counterclockwise).
(b) Complete state of stress. To determine the complete state of stress, we
need to find the stresses acting on all faces of a stress element oriented at 25°
(Fig. 2-39c). Face ab, for which u 25°, has the same orientation as the
inclined plane shown in Fig. 2-39b. Therefore, the stresses are the same as those
given previously.
The stresses on the opposite face cd are the same as those on face ab, which
can be verified by substituting u 25° 180° 205° into Eqs. (2-29a and b).
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
114
CHAPTER 2 Axially Loaded Members
For face ad we substitute u 25° 90° 65° into Eqs. (2-29a and b)
and obtain
su 13.4 MPa
tu 28.7 MPa
These same stresses apply to the opposite face bc, as can be verified by substituting u 25° 90° 115° into Eqs. (2-29a and b). Note that the normal
stress is compressive and the shear stress acts clockwise.
The complete state of stress is shown by the stress element of Fig. 2-39c. A
sketch of this kind is an excellent way to show the directions of the stresses and
the orientations of the planes on which they act.
Example 2-11
A compression bar having a square cross section of width b must support a load
P 8000 lb (Fig. 2-40a). The bar is constructed from two pieces of material
that are connected by a glued joint (known as a scarf joint) along plane pq,
which is at an angle a 40° to the vertical. The material is a structural plastic
for which the allowable stresses in compression and shear are 1100 psi and 600
psi, respectively. Also, the allowable stresses in the glued joint are 750 psi in
compression and 500 psi in shear.
Determine the minimum width b of the bar.
Solution
For convenience, let us rotate a segment of the bar to a horizontal position
(Fig. 2-40b) that matches the figures used in deriving the equations for the
stresses on an inclined section (see Figs. 2-33 and 2-34). With the bar in this
position, we see that the normal n to the plane of the glued joint (plane pq)
makes an angle b 90° a, or 50°, with the axis of the bar. Since the angle u
is defined as positive when counterclockwise (Fig. 2-34), we conclude that u
50° for the glued joint.
The cross-sectional area of the bar is related to the load P and the stress sx
acting on the cross sections by the equation
P
A
sx
(a)
Therefore, to find the required area, we must determine the value of sx corresponding to each of the four allowable stress. Then the smallest value of sx
will determine the required area. The values of sx are obtained by rearranging
Eqs. (2-29a and b) as follows:
su
sx
cos2u
tu
sx
sin u cos u
We will now apply these equations to the glued joint and to the plastic.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(2-32a,b)
SECTION 2.6 Stresses on Inclined Sections
115
(a) Values of sx based upon the allowable stresses in the glued joint. For
compression in the glued joint we have su ⫺750 psi and u ⫺50°. Substituting into Eq. (2-32a), we get
750 psi
1815 psi
sx
(cos 50°)2
For shear in the glued joint we have an allowable stress of 500 psi.
However, it is not immediately evident whether tu is 500 psi or 500 psi.
One approach is to substitute both 500 psi and 500 psi into Eq. (2-32b) and
then select the value of sx that is negative. The other value of sx will be positive
(tension) and does not apply to this bar. Another approach is to inspect the bar
itself (Fig. 2-40b) and observe from the directions of the loads that the shear
stress will act clockwise against plane pq, which means that the shear stress is
negative. Therefore, we substitute tu 500 psi and u 50° into Eq.
(2-32b) and obtain
P
500 psi
sx 1015 psi
(sin 50°)(cos 50°)
p
a
q
(c)
(b) Values of sx based upon the allowable stresses in the plastic. The
maximum compressive stress in the plastic occurs on a cross section. Therefore,
since the allowable stress in compression is 1100 psi, we know immediately that
b
b
(b)
sx 1100 psi
(d)
The maximum shear stress occurs on a plane at 45° and is numerically equal to
sx / 2 (see Eq. 2-31). Since the allowable stress in shear is 600 psi, we obtain
sx 1200 psi
(a)
y
p
P O
P
x
a
q
n
b = 90° – a
(e)
This same result can be obtained from Eq. (2-32b) by substituting tu 600 psi
and u 45°.
(c) Minimum width of the bar. Comparing the four values of sx (Eqs. b, c,
d, and e), we see that the smallest is sx 1015 psi. Therefore, this value
governs the design. Substituting into Eq. (a), and using only numerical values,
we obtain the required area:
8000 lb
A 7.88 in.2
1015 psi
Since the bar has a square cross section (A b2), the minimum width is
a = 40°
b = 50°
u = –b = –50°
bmin A 7.88 in.2 2.81 in.
(b)
FIG. 2-40 Example 2-11. Stresses on an
Any width larger than bmin will ensure that the allowable stresses are not
exceeded.
inclined section
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
116
CHAPTER 2 Axially Loaded Members
2.7 STRAIN ENERGY
L
d
P
FIG. 2-41 Prismatic bar subjected to a
statically applied load
P
dP1
P
P1
O
d
dd1
d1
d
Strain energy is a fundamental concept in applied mechanics, and strainenergy principles are widely used for determining the response of
machines and structures to both static and dynamic loads. In this section
we introduce the subject of strain energy in its simplest form by considering
only axially loaded members subjected to static loads. More complicated
structural elements are discussed in later chapters—bars in torsion in
Section 3.9 and beams in bending in Section 9.8. In addition, the use
of strain energy in connection with dynamic loads is described in
Sections 2.8 and 9.10.
To illustrate the basic ideas, let us again consider a prismatic bar of
length L subjected to a tensile force P (Fig. 2-41). We assume that the
load is applied slowly, so that it gradually increases from zero to its
maximum value P. Such a load is called a static load because there are
no dynamic or inertial effects due to motion. The bar gradually elongates as the load is applied, eventually reaching its maximum elongation
d at the same time that the load reaches its full value P. Thereafter, the
load and elongation remain unchanged.
During the loading process, the load P moves slowly through the
distance d and does a certain amount of work. To evaluate this work, we
recall from elementary mechanics that a constant force does work equal
to the product of the force and the distance through which it moves.
However, in our case the force varies in magnitude from zero to its
maximum value P. To find the work done by the load under these
conditions, we need to know the manner in which the force varies. This
information is supplied by a load-displacement diagram, such as the
one plotted in Fig. 2-42. On this diagram the vertical axis represents
the axial load and the horizontal axis represents the corresponding
elongation of the bar. The shape of the curve depends upon the properties
of the material.
Let us denote by P1 any value of the load between zero and the
maximum value P, and let us denote the corresponding elongation of the
bar by d1. Then an increment dP1 in the load will produce an increment
dd1 in the elongation. The work done by the load during this incremental
elongation is the product of the load and the distance through which it
moves, that is, the work equals P1dd1. This work is represented in the
figure by the area of the shaded strip below the load-displacement curve.
The total work done by the load as it increases from zero to the
maximum value P is the summation of all such elemental strips:
P dd
d
FIG. 2-42 Load-displacement diagram
W
0
1
1
(2-33)
In geometric terms, the work done by the load is equal to the area below
the load-displacement curve.
When the load stretches the bar, strains are produced. The presence
of these strains increases the energy level of the bar itself. Therefore, a
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.7 Strain Energy
117
new quantity, called strain energy, is defined as the energy absorbed by
the bar during the loading process. From the principle of conservation of
energy, we know that this strain energy is equal to the work done by the
load provided no energy is added or subtracted in the form of heat.
Therefore,
P d
UW
1
1
(2-34)
0
in which U is the symbol for strain energy. Sometimes strain energy is
referred to as internal work to distinguish it from the external work
done by the load.
Work and energy are expressed in the same units. In SI, the unit of
work and energy is the joule (J), which is equal to one newton meter
(1 J 1 N·m). In USCS units, work and energy are expressed in footpounds (ft-lb), foot-kips (ft-k), inch-pounds (in.-lb), and inch-kips
(in.-k).*
Elastic and Inelastic Strain Energy
P
A
B
Inelastic
strain
energy
Elastic
strain
energy
O
D
C
d
FIG. 2-43 Elastic and inelastic strain
energy
If the force P (Fig. 2-41) is slowly removed from the bar, the bar will
shorten. If the elastic limit of the material is not exceeded, the bar
will return to its original length. If the limit is exceeded, a permanent set
will remain (see Section 1.4). Thus, either all or part of the strain energy
will be recovered in the form of work. This behavior is shown on the
load-displacement diagram of Fig. 2-43. During loading, the work done
by the load is equal to the area below the curve (area OABCDO). When
the load is removed, the load-displacement diagram follows line BD if
point B is beyond the elastic limit, and a permanent elongation OD
remains. Thus, the strain energy recovered during unloading, called the
elastic strain energy, is represented by the shaded triangle BCD. Area
OABDO represents energy that is lost in the process of permanently
deforming the bar. This energy is known as the inelastic strain energy.
Most structures are designed with the expectation that the material
will remain within the elastic range under ordinary conditions of service.
Let us assume that the load at which the stress in the material reaches
the elastic limit is represented by point A on the load-displacement curve
(Fig. 2-43). As long as the load is below this value, all of the strain
energy is recovered during unloading and no permanent elongation
remains. Thus, the bar acts as an elastic spring, storing and releasing
energy as the load is applied and removed.
*Conversion factors for work and energy are given in Appendix A, Table A-5.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
118
CHAPTER 2 Axially Loaded Members
P
Linearly Elastic Behavior
A
U = Pd
2
P
B
O
d
d
FIG. 2-44 Load-displacement diagram for
a bar of linearly elastic material
Let us now assume that the material of the bar follows Hooke’s law, so
that the load-displacement curve is a straight line (Fig. 2-44). Then
the strain energy U stored in the bar (equal to the work W done by the
load) is
P
U W
2
(2-35)
which is the area of the shaded triangle OAB in the figure.*
The relationship between the load P and the elongation d for a bar
of linearly elastic material is given by the equation
PL
d
EA
(2-36)
Combining this equation with Eq. (2-35) enables us to express the strain
energy of a linearly elastic bar in either of the following forms:
P 2L
U
2E A
E A 2
U
2L
(2-37a,b)
The first equation expresses the strain energy as a function of the load
and the second expresses it as a function of the elongation.
From the first equation we see that increasing the length of a bar
increases the amount of strain energy even though the load is unchanged
(because more material is being strained by the load). On the other hand,
increasing either the modulus of elasticity or the cross-sectional area
decreases the strain energy because the strains in the bar are reduced.
These ideas are illustrated in Examples 2-12 and 2-15.
Strain-energy equations analogous to Eqs. (2-37a) and (2-37b) can
be written for a linearly elastic spring by replacing the stiffness EA/L
of the prismatic bar by the stiffness k of the spring. Thus,
P2
U
2k
k 2
U
2
(2-38a,b)
Other forms of these equations can be obtained by replacing k by 1/f,
where f is the flexibility.
*The principle that the work of the external loads is equal to the strain energy (for the
case of linearly elastic behavior) was first stated by the French engineer B. P. E.
Clapeyron (1799–1864) and is known as Clapeyron’s theorem (Ref. 2-7).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.7 Strain Energy
119
Nonuniform Bars
A
The total strain energy U of a bar consisting of several segments is equal
to the sum of the strain energies of the individual segments. For
instance, the strain energy of the bar pictured in Fig. 2-45 equals the
strain energy of segment AB plus the strain energy of segment BC. This
concept is expressed in general terms by the following equation:
P1
B
n
U Ui
C
(2-39)
i1
P2
FIG. 2-45 Bar consisting of prismatic
segments having different crosssectional areas and different axial forces
in which Ui is the strain energy of segment i of the bar and n is the
number of segments. (This relation holds whether the material behaves
in a linear or nonlinear manner.)
Now assume that the material of the bar is linearly elastic and that
the internal axial force is constant within each segment. We can then use
Eq. (2-37a) to obtain the strain energies of the segments, and Eq. (2-39)
becomes
n
N 2L
U i i
i1 2 E i Ai
A
x
dx
(2-40)
in which Ni is the axial force acting in segment i and Li, Ei, and Ai are
properties of segment i. (The use of this equation is illustrated in Examples
2-12 and 2-15 at the end of the section.)
We can obtain the strain energy of a nonprismatic bar with
continuously varying axial force (Fig. 2-46) by applying Eq. (2-37a) to a
differential element (shown shaded in the figure) and then integrating
along the length of the bar:
L
B
P
L
U
FIG. 2-46 Nonprismatic bar with varying
axial force
0
[N(x)]2dx
2EA(x)
(2-41)
In this equation, N(x) and A(x) are the axial force and cross-sectional
area at distance x from the end of the bar. (Example 2-13 illustrates the
use of this equation.)
Comments
The preceding expressions for strain energy (Eqs. 2-37 through 2-41)
show that strain energy is not a linear function of the loads, not even
when the material is linearly elastic. Thus, it is important to realize that
we cannot obtain the strain energy of a structure supporting more than
one load by combining the strain energies obtained from the individual
loads acting separately.
In the case of the nonprismatic bar shown in Fig. 2-45, the total
strain energy is not the sum of the strain energy due to load P1 acting
alone and the strain energy due to load P2 acting alone. Instead, we must
evaluate the strain energy with all of the loads acting simultaneously, as
demonstrated later in Example 2-13.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
120
CHAPTER 2 Axially Loaded Members
Although we considered only tension members in the preceding
discussions of strain energy, all of the concepts and equations apply
equally well to members in compression. Since the work done by an
axial load is positive regardless of whether the load causes tension or
compression, it follows that strain energy is always a positive quantity.
This fact is also evident in the expressions for strain energy of linearly
elastic bars (such as Eqs. 2-37a and 2-37b). These expressions are
always positive because the load and elongation terms are squared.
Strain energy is a form of potential energy (or “energy of position”)
because it depends upon the relative locations of the particles or elements
that make up the member. When a bar or a spring is compressed, its particles
are crowded more closely together; when it is stretched, the distances
between particles increase. In both cases the strain energy of the member
increases as compared to its strain energy in the unloaded position.
Displacements Caused by a Single Load
The displacement of a linearly elastic structure supporting only one load
can be determined from its strain energy. To illustrate the method,
consider a two-bar truss (Fig. 2-47) loaded by a vertical force P. Our
objective is to determine the vertical displacement d at joint B where the
load is applied.
When applied slowly to the truss, the load P does work as it moves
through the vertical displacement d. However, it does no work as it
moves laterally, that is, sideways. Therefore, since the load-displacement
diagram is linear (see Fig. 2-44 and Eq. 2-35), the strain energy U stored
in the structure, equal to the work done by the load, is
A
B
C
Pd
U W
2
d
B'
P
from which we get
FIG. 2-47 Structure supporting a single
2U
P
(2-42)
load P
This equation shows that under certain special conditions, as outlined in
the following paragraph, the displacement of a structure can be determined directly from the strain energy.
The conditions that must be met in order to use Eq. (2-42) are as
follows: (1) the structure must behave in a linearly elastic manner, and
(2) only one load may act on the structure. Furthermore, the only
displacement that can be determined is the displacement corresponding
to the load itself (that is, the displacement must be in the direction of the
load and must be at the point where the load is applied). Therefore, this
method for finding displacements is extremely limited in its application
and is not a good indicator of the great importance of strain-energy
principles in structural mechanics. However, the method does provide
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.7 Strain Energy
121
an introduction to the use of strain energy. (The method is illustrated
later in Example 2-14.)
Strain-Energy Density
In many situations it is convenient to use a quantity called strainenergy density, defined as the strain energy per unit volume of material.
Expressions for strain-energy density in the case of linearly elastic
materials can be obtained from the formulas for strain energy of a
prismatic bar (Eqs. 2-37a and b). Since the strain energy of the bar is
distributed uniformly throughout its volume, we can determine the
strain-energy density by dividing the total strain energy U by the volume
AL of the bar. Thus, the strain-energy density, denoted by the symbol u,
can be expressed in either of these forms:
P2
u
2E A2
Ed 2
u
2 L2
(2-43a,b)
If we replace P/A by the stress s and d/L by the strain e, we get
2
u
2E
E 2
u
2
(2-44a,b)
These equations give the strain-energy density in a linearly elastic
material in terms of either the normal stress s or the normal strain e.
The expressions in Eqs. (2-44a and b) have a simple geometric
interpretation. They are equal to the area se/2 of the triangle below the
stress-strain diagram for a material that follows Hooke’s law (s Ee).
In more general situations where the material does not follow Hooke’s
law, the strain-energy density is still equal to the area below the stressstrain curve, but the area must be evaluated for each particular material.
Strain-energy density has units of energy divided by volume. The
SI units are joules per cubic meter (J/m3) and the USCS units are
foot-pounds per cubic foot, inch-pounds per cubic inch, and other
similar units. Since all of these units reduce to units of stress (recall that
1 J 1 N·m), we can also use units such as pascals (Pa) and pounds per
square inch (psi) for strain-energy density.
The strain-energy density of the material when it is stressed to the
proportional limit is called the modulus of resilience ur. It is found by
substituting the proportional limit spl into Eq. (2-44a):
s 2pl
ur
2E
(2-45)
For example, a mild steel having spl 36,000 psi and E 30 106 psi
has a modulus of resilience ur 21.6 psi (or 149 kPa). Note that the
modulus of resilience is equal to the area below the stress-strain curve
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
122
CHAPTER 2 Axially Loaded Members
up to the proportional limit. Resilience represents the ability of a material
to absorb and release energy within the elastic range.
Another quantity, called toughness, refers to the ability of a material
to absorb energy without fracturing. The corresponding modulus, called
the modulus of toughness ut, is the strain-energy density when the
material is stressed to the point of failure. It is equal to the area below
the entire stress-strain curve. The higher the modulus of toughness, the
greater the ability of the material to absorb energy without failing. A
high modulus of toughness is therefore important when the material is
subject to impact loads (see Section 2.8).
The preceding expressions for strain-energy density (Eqs. 2-43 to
2-45) were derived for uniaxial stress, that is, for materials subjected
only to tension or compression. Formulas for strain-energy density in
other stress states are presented in Chapters 3 and 7.
Example 2-12
Three round bars having the same length L but different shapes are shown in
Fig. 2-48. The first bar has diameter d over its entire length, the second has
diameter d over one-fifth of its length, and the third has diameter d over onefifteenth of its length. Elsewhere, the second and third bars have diameter 2d.
All three bars are subjected to the same axial load P.
Compare the amounts of strain energy stored in the bars, assuming linearly
elastic behavior. (Disregard the effects of stress concentrations and the weights
of the bars.)
2d
d
L
FIG. 2-48 Example 2-12. Calculation of
strain energy
P
(a)
2d
d
L ⁄5
P
(b)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
d
L ⁄ 15
P
(c)
SECTION 2.7 Strain Energy
123
Solution
(a) Strain energy U1 of the first bar. The strain energy of the first bar is
found directly from Eq. (2-37a):
P 2L
U1
2EA
(a)
in which A p d 2/4.
(b) Strain energy U2 of the second bar. The strain energy is found by
summing the strain energies in the three segments of the bar (see Eq. 2-40).
Thus,
n
N 2Li
P 2L
P 2(L /5)
P2(4L/5)
2U
U2 i
1
2
E
E
(4
A
)
5
E
A
2
E
A
2
A
5
i i
i1
(b)
which is only 40% of the strain energy of the first bar. Thus, increasing the
cross-sectional area over part of the length has greatly reduced the amount of
strain energy that can be stored in the bar.
(c) Strain energy U3 of the third bar. Again using Eq. (2-40), we get
n
N 2L
3P2L
P 2(L/15)
P 2(14L /15)
3U
U3 ii 1
2E A
2E(4 A)
20E A
10
i1 2E i A i
(c)
The strain energy has now decreased to 30% of the strain energy of the first bar.
Note: Comparing these results, we see that the strain energy decreases as
the part of the bar with the larger area increases. If the same amount of work is
applied to all three bars, the highest stress will be in the third bar, because the
third bar has the least energy-absorbing capacity. If the region having diameter
d is made even smaller, the energy-absorbing capacity will decrease further.
We therefore conclude that it takes only a small amount of work to bring
the tensile stress to a high value in a bar with a groove, and the narrower the
groove, the more severe the condition. When the loads are dynamic and the
ability to absorb energy is important, the presence of grooves is very damaging.
In the case of static loads, the maximum stresses are more important than
the ability to absorb energy. In this example, all three bars have the same maximum stress P/A (provided stress concentrations are alleviated), and therefore
all three bars have the same load-carrying capacity when the load is applied
statically.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
124
CHAPTER 2 Axially Loaded Members
Example 2-13
x
Determine the strain energy of a prismatic bar suspended from its upper end
(Fig. 2-49). Consider the following loads: (a) the weight of the bar itself, and (b)
the weight of the bar plus a load P at the lower end. (Assume linearly elastic
behavior.)
x
L
dx
L
Solution
dx
P
(a)
(b)
FIG. 2-49 Example 2-13. (a) Bar hanging
under its own weight, and (b) bar
hanging under its own weight and also
supporting a load P
(a) Strain energy due to the weight of the bar itself (Fig. 2-49a). The bar is
subjected to a varying axial force, the internal force being zero at the lower end
and maximum at the upper end. To determine the axial force, we consider an
element of length dx (shown shaded in the figure) at distance x from the upper
end. The internal axial force N(x) acting on this element is equal to the weight
of the bar below the element:
(d)
N(x) gA(L x)
in which g is the weight density of the material and A is the cross-sectional area
of the bar. Substituting into Eq. (2-41) and integrating gives the total strain
energy:
L
U
0
[N(x)]2 dx
2EA( x)
L
0
[gA(L x)]2 dx
g 2AL3
2EA
6E
(2-46)
(b) Strain energy due to the weight of the bar plus the load P (Fig. 2-49b).
In this case the axial force N(x) acting on the element is
(e)
N(x) gA(L x) P
(compare with Eq. d). From Eq. (2-41) we now obtain
L
U
0
g 2AL3
gPL2
P 2L
[gA(L x) P]2 dx
6E
2E
2E A
2EA
(2-47)
Note: The first term in this expression is the same as the strain energy of a
bar hanging under its own weight (Eq. 2-46), and the last term is the same as the
strain energy of a bar subjected only to an axial force P (Eq. 2-37a). However,
the middle term contains both g and P, showing that it depends upon both the
weight of the bar and the magnitude of the applied load.
Thus, this example illustrates that the strain energy of a bar subjected to
two loads is not equal to the sum of the strain energies produced by the individual loads acting separately.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.7 Strain Energy
125
Example 2-14
Determine the vertical displacement dB of joint B of the truss shown in Fig. 2-50.
Note that the only load acting on the truss is a vertical load P at joint B. Assume
that both members of the truss have the same axial rigidity EA.
A
C
b
b
H
B
FIG. 2-50 Example 2-14. Displacement
P
of a truss supporting a single load P
Solution
Since there is only one load acting on the truss, we can find the displacement corresponding to that load by equating the work of the load to the strain
energy of the members. However, to find the strain energy we must know the
forces in the members (see Eq. 2-37a).
From the equilibrium of forces acting at joint B we see that the axial force
F in either bar is
P
F
2 cos b
(f)
in which b is the angle shown in the figure.
Also, from the geometry of the truss we see that the length of each bar is
H
L1
cos b
(g)
in which H is the height of the truss.
We can now obtain the strain energy of the two bars from Eq. (2-37a):
F 2L
P 2H
U (2) 1
2E A
4EA c os3 b
(h)
Also, the work of the load P (from Eq. 2-35) is
PdB
W
2
(i)
where dB is the downward displacement of joint B. Equating U and W and
solving for dB, we obtain
PH
dB
2EA cos3 b
(2-48)
Note that we found this displacement using only equilibrium and strain energy—we
did not need to draw a displacement diagram at joint B.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
126
CHAPTER 2 Axially Loaded Members
Example 2-15
The cylinder for a compressed air machine is clamped by bolts that pass through
the flanges of the cylinder (Fig. 2-51a). A detail of one of the bolts is shown in
part (b) of the figure. The diameter d of the shank is 0.500 in. and the root diameter dr of the threaded portion is 0.406 in. The grip g of the bolts is 1.50 in. and
the threads extend a distance t 0.25 in. into the grip. Under the action of
repeated cycles of high and low pressure in the chamber, the bolts may eventually
break.
To reduce the likelihood of the bolts failing, the designers suggest two
possible modifications: (a) Machine down the shanks of the bolts so that the
shank diameter is the same as the thread diameter dr, as shown in Fig. 2-52a.
(b) Replace each pair of bolts by a single long bolt, as shown in Fig. 2-52b. The
long bolts are similar to the original bolts (Fig. 2-51b) except that the grip is
increased to the distance L 13.5 in.
Compare the energy-absorbing capacity of the three bolt configurations:
(1) original bolts, (2) bolts with reduced shank diameter, and (3) long bolts.
(Assume linearly elastic behavior and disregard the effects of stress
concentrations.)
Bolt
Cylinder
t
d
Chamber
FIG. 2-51 Example 2-15. (a) Cylinder
with piston and clamping bolts, and
(b) detail of one bolt
Piston
dr d
g
(a)(a)
(b)
Solution
(1) Original bolts. The original bolts can be idealized as bars consisting of
two segments (Fig. 2-51b). The left-hand segment has length g 2 t and diameter d, and the right-hand segment has length t and diameter dr. The strain
energy of one bolt under a tensile load P can be obtained by adding the strain
energies of the two segments (Eq. 2-40):
n
N 2L
P2(g t)
P2t
U1 ii
2EAs
2E Ar
i1 2E i A i
(j)
in which As is the cross-sectional area of the shank and Ar is the cross-sectional
area at the root of the threads; thus,
p d2
As
4
pd 2
Ar r
4
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(k)
SECTION 2.7 Strain Energy
127
Substituting these expressions into Eq. (j), we get the following formula for the
strain energy of one of the original bolts:
2 P 2t
2P2(g t)
2
U1
2
pEdr
p Ed
(l)
(2) Bolts with reduced shank diameter. These bolts can be idealized as prismatic bars having length g and diameter dr (Fig. 2-52a). Therefore, the strain
energy of one bolt (see Eq. 2-37a) is
P 2g
2P 2g
U2 2
2E Ar
pE dr
t
dr
dr d
The ratio of the strain energies for cases (1) and (2) is
gd 2
U2
U1
(g t)dr2 td 2
g
(m)
(n)
or, upon substituting numerical values,
(a)
(1.50 in.)(0.500 in.)2
U2
1.40
(1.50 in. 0.25 in.)(0.406 in.)2 (0.25 in.)(0.500 in.)2
U1
Thus, using bolts with reduced shank diameters results in a 40% increase in the
amount of strain energy that can be absorbed by the bolts. If implemented, this
scheme should reduce the number of failures caused by the impact loads.
(3) Long bolts. The calculations for the long bolts (Fig. 2-52b) are the same
as for the original bolts except the grip g is changed to the grip L. Therefore, the
strain energy of one long bolt (compare with Eq. l) is
2P 2(L t)
2P 2t
2
U3
2
p Ed
p Edr
L
(b)
FIG. 2-52 Example 2-15. Proposed
modifications to the bolts: (a) Bolts with
reduced shank diameter, and (b) bolts
with increased length
(o)
Since one long bolt replaces two of the original bolts, we must compare the
strain energies by taking the ratio of U3 to 2U1, as follows:
U3
(L t) d r2 td 2
2U 1
2(g t) d r2 2td 2
(p)
Substituting numerical values gives
(13.5 in. 0.25 in.)(0.406 in.)2 (0.25 in.)(0.500 in.)2
U3
4.18
2(1.50 in. 0.25 in.)(0.406 in.)2 2(0.25 in.)(0.500 in.)2
2U 1
Thus, using long bolts increases the energy-absorbing capacity by 318% and
achieves the greatest safety from the standpoint of strain energy.
Note: When designing bolts, designers must also consider the maximum
tensile stresses, maximum bearing stresses, stress concentrations, and many
other matters.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
128
CHAPTER 2 Axially Loaded Members
2.8 IMPACT LOADING
★
A
Sliding collar
of mass M
h
B
Flange
(a)
A
M
L
h
B
d max
(b)
FIG. 2-53 Impact load on a prismatic bar
AB due to a falling object of mass M
Loads can be classified as static or dynamic depending upon whether
they remain constant or vary with time. A static load is applied slowly,
so that it causes no vibrational or dynamic effects in the structure. The
load increases gradually from zero to its maximum value, and thereafter
it remains constant.
A dynamic load may take many forms—some loads are applied
and removed suddenly (impact loads), others persist for long periods of
time and continuously vary in intensity ( fluctuating loads). Impact loads
are produced when two objects collide or when a falling object strikes a
structure. Fluctuating loads are produced by rotating machinery, traffic,
wind gusts, water waves, earthquakes, and manufacturing processes.
As an example of how structures respond to dynamic loads, we will
discuss the impact of an object falling onto the lower end of a prismatic
bar (Fig. 2-53). A collar of mass M, initially at rest, falls from a height h
onto a flange at the end of bar AB. When the collar strikes the flange, the
bar begins to elongate, creating axial stresses within the bar. In a very
short interval of time, such as a few milliseconds, the flange will move
downward and reach its position of maximum displacement. Thereafter,
the bar shortens, then lengthens, then shortens again as the bar vibrates
longitudinally and the end of the bar moves up and down. The vibrations
are analogous to those that occur when a spring is stretched and then
released, or when a person makes a bungee jump. The vibrations of the
bar soon cease because of various damping effects, and then the bar
comes to rest with the mass M supported on the flange.
The response of the bar to the falling collar is obviously very
complicated, and a complete and accurate analysis requires the use of
advanced mathematical techniques. However, we can make an approximate analysis by using the concept of strain energy (Section 2.7) and
making several simplifying assumptions.
Let us begin by considering the energy of the system just before the
collar is released (Fig. 2-53a). The potential energy of the collar with
respect to the elevation of the flange is Mgh, where g is the acceleration
of gravity.* This potential energy is converted into kinetic energy as the
collar falls. At the instant the collar strikes the flange, its potential
energy with respect to the elevation of the flange is zero and its kinetic
energy is Mv 2/2, where v 2gh is its velocity.**
*In SI units, the acceleration of gravity g 9.81 m/s2; in USCS units, g 32.2 ft/s2. For
more precise values of g, or for a discussion of mass and weight, see Appendix A.
**In engineering work, velocity is usually treated as a vector quantity. However, since
kinetic energy is a scalar, we will use the word “velocity” to mean the magnitude of the
velocity, or the speed.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.8 Impact Loading
129
During the ensuing impact, the kinetic energy of the collar is transformed
into other forms of energy. Part of the kinetic energy is transformed into
the strain energy of the stretched bar. Some of the energy is dissipated in
the production of heat and in causing localized plastic deformations of
the collar and flange. A small part remains as the kinetic energy of the
collar, which either moves further downward (while in contact with the
flange) or else bounces upward.
To make a simplified analysis of this very complex situation, we will
idealize the behavior by making the following assumptions. (1) We
assume that the collar and flange are so constructed that the collar “sticks”
to the flange and moves downward with it (that is, the collar does not
rebound). This behavior is more likely to prevail when the mass of the
collar is large compared to the mass of the bar. (2) We disregard all
energy losses and assume that the kinetic energy of the falling mass is
transformed entirely into strain energy of the bar. This assumption
predicts larger stresses in the bar than would be predicted if we took
energy losses into account. (3) We disregard any change in the potential
energy of the bar itself (due to the vertical movement of elements of the
bar), and we ignore the existence of strain energy in the bar due to its own
weight. Both of these effects are extremely small. (4) We assume that the
stresses in the bar remain within the linearly elastic range. (5) We assume
that the stress distribution throughout the bar is the same as when the bar
is loaded statically by a force at the lower end, that is, we assume the
stresses are uniform throughout the volume of the bar. (In reality longitudinal
stress waves will travel through the bar, thereby causing variations in the
stress distribution.)
On the basis of the preceding assumptions, we can calculate the
maximum elongation and the maximum tensile stresses produced by the
impact load. (Recall that we are disregarding the weight of the bar itself
and finding the stresses due solely to the falling collar.)
Maximum Elongation of the Bar
The maximum elongation dmax (Fig. 2-53b) can be obtained from the
principle of conservation of energy by equating the potential energy lost
by the falling mass to the maximum strain energy acquired by the bar. The
potential energy lost is W(h dmax), where W Mg is the weight of the
collar and h dmax is the distance through which it moves. The strain
energy of the bar is EAd 2max/2L, where EA is the axial rigidity and L is the
length of the bar (see Eq. 2-37b). Thus, we obtain the following equation:
EAd 2max
(2-49)
W(h dmax)
2L
This equation is quadratic in dmax and can be solved for the positive
root; the result is
WL
dmax
EA
2
WEAL
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
WL
2h
EA
1/2
(2-50)
130
CHAPTER 2 Axially Loaded Members
Note that the maximum elongation of the bar increases if either the
weight of the collar or the height of fall is increased. The elongation
diminishes if the stiffness EA/L is increased.
The preceding equation can be written in simpler form by introducing
the notation
WL
MgL
dst
EA
EA
(2-51)
in which dst is the elongation of the bar due to the weight of the collar
under static loading conditions. Equation (2-50) now becomes
dmax dst (d 2st 2hdst )1/2
(2-52)
or
2h 1/2
dmax dst 1 1
(2-53)
dst
From this equation we see that the elongation of the bar under the impact
load is much larger than it would be if the same load were applied statically.
Suppose, for instance, that the height h is 40 times the static displacement
dst; the maximum elongation would then be 10 times the static elongation.
When the height h is large compared to the static elongation, we can
disregard the “ones” on the right-hand side of Eq. (2-53) and obtain
Mv2L
EA
dmax 2hdst
(2-54)
in which M W/g and v 2gh is the velocity of the falling mass when
it strikes the flange. This equation can also be obtained directly from
Eq. (2-49) by omitting dmax on the left-hand side of the equation and then
solving for dmax. Because of the omitted terms, values of dmax calculated
from Eq. (2-54) are always less than those obtained from Eq. (2-53).
Maximum Stress in the Bar
The maximum stress can be calculated easily from the maximum
elongation because we are assuming that the stress distribution is
uniform throughout the length of the bar. From the general equation
d PL/EA 5 s L /E, we know that
Edmax
smax
L
(2-55)
Substituting from Eq. (2-50), we obtain the following equation for the
maximum tensile stress:
W
smax
A
2
W
A
2WhE
AL
1/2
(2-56)
Introducing the notation
W
Mg
Edst
sst
A
A
L
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(2-57)
SECTION 2.8 Impact Loading
131
in which sst is the stress when the load acts statically, we can write Eq.
(2-56) in the form
2hE
smax sst s 2st sst
L
1/2
(2-58)
or
1/2
2hE
smax sst 1 1
Lsst
(2-59)
This equation is analogous to Eq. (2-53) and again shows that an impact
load produces much larger effects than when the same load is applied
statically.
Again considering the case where the height h is large compared to
the elongation of the bar (compare with Eq. 2-54), we obtain
smax
2hEsst
L
M v 2E
AL
(2-60)
From this result we see that an increase in the kinetic energy Mv 2/2 of
the falling mass will increase the stress, whereas an increase in the
volume AL of the bar will reduce the stress. This situation is quite
different from static tension of the bar, where the stress is independent
of the length L and the modulus of elasticity E.
The preceding equations for the maximum elongation and maximum
stress apply only at the instant when the flange of the bar is at its lowest
position. After the maximum elongation is reached in the bar, the bar will
vibrate axially until it comes to rest at the static elongation. From then on,
the elongation and stress have the values given by Eqs. (2-51) and (2-57).
Although the preceding equations were derived for the case of a
prismatic bar, they can be used for any linearly elastic structure
subjected to a falling load, provided we know the appropriate stiffness
of the structure. In particular, the equations can be used for a spring by
substituting the stiffness k of the spring (see Section 2.2) for the stiffness
EA/L of the prismatic bar.
Impact Factor
The ratio of the dynamic response of a structure to the static response
(for the same load) is known as an impact factor. For instance, the
impact factor for the elongation of the bar of Fig. 2-53 is the ratio of the
maximum elongation to the static elongation:
d ax
Impact factor m
dst
(2-61)
This factor represents the amount by which the static elongation is
amplified due to the dynamic effects of the impact.
Equations analogous to Eq. (2-61) can be written for other impact
factors, such as the impact factor for the stress in the bar (the ratio of
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
132
CHAPTER 2 Axially Loaded Members
smax to sst). When the collar falls through a considerable height, the
impact factor can be very large, such as 100 or more.
Suddenly Applied Load
A special case of impact occurs when a load is applied suddenly with no
initial velocity. To explain this kind of loading, consider again the prismatic
bar shown in Fig. 2-53 and assume that the sliding collar is lowered
gently until it just touches the flange. Then the collar is suddenly
released. Although in this instance no kinetic energy exists at the beginning
of extension of the bar, the behavior is quite different from that of static
loading of the bar. Under static loading conditions, the load is released
gradually and equilibrium always exists between the applied load and
the resisting force of the bar.
However, consider what happens when the collar is released
suddenly from its point of contact with the flange. Initially the elongation
of the bar and the stress in the bar are zero, but then the collar moves
downward under the action of its own weight. During this motion the
bar elongates and its resisting force gradually increases. The motion
continues until at some instant the resisting force just equals W, the
weight of the collar. At this particular instant the elongation of the bar is
dst. However, the collar now has a certain kinetic energy, which it
acquired during the downward displacement dst. Therefore, the collar
continues to move downward until its velocity is brought to zero by the
resisting force in the bar. The maximum elongation for this condition is
obtained from Eq. (2-53) by setting h equal to zero; thus,
dmax 2dst
(2-62)
From this equation we see that a suddenly applied load produces an
elongation twice as large as the elongation caused by the same load
applied statically. Thus, the impact factor is 2.
After the maximum elongation 2dst has been reached, the end of the
bar will move upward and begin a series of up and down vibrations,
eventually coming to rest at the static elongation produced by the weight
of the collar.*
Limitations
The preceding analyses were based upon the assumption that no energy
losses occur during impact. In reality, energy losses always occur, with
most of the lost energy being dissipated in the form of heat and localized
deformation of the materials. Because of these losses, the kinetic energy
of a system immediately after an impact is less than it was before the
impact. Consequently, less energy is converted into strain energy of the
*Equation (2-62) was first obtained by the French mathematician and scientist J. V.
Poncelet (1788–1867); see Ref. 2-8.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.8 Impact Loading
133
bar than we previously assumed. As a result, the actual displacement of
the end of the bar of Fig. 2-53 is less than that predicted by our simplified analysis.
We also assumed that the stresses in the bar remain within the
proportional limit. If the maximum stress exceeds this limit, the analysis
becomes more complicated because the elongation of the bar is no longer
proportional to the axial force. Other factors to consider are the effects
of stress waves, damping, and imperfections at the contact surfaces.
Therefore, we must remember that all of the formulas in this section
are based upon highly idealized conditions and give only a rough
approximation of the true conditions (usually overestimating the
elongation).
Materials that exhibit considerable ductility beyond the proportional
limit generally offer much greater resistance to impact loads than do
brittle materials. Also, bars with grooves, holes, and other forms of
stress concentrations (see Sections 2.9 and 2.10) are very weak against
impact—a slight shock may produce fracture, even when the material
itself is ductile under static loading.
Example 2-16
d = 15 mm
L = 2.0 m
M = 20 kg
A round, prismatic steel bar (E 210 GPa) of length L 2.0 m and diameter
d 15 mm hangs vertically from a support at its upper end (Fig. 2-54). A
sliding collar of mass M 20 kg drops from a height h150 mm onto the
flange at the lower end of the bar without rebounding.
(a) Calculate the maximum elongation of the bar due to the impact and
determine the corresponding impact factor.
(b) Calculate the maximum tensile stress in the bar and determine the
corresponding impact factor.
h = 150 mm
Solution
FIG. 2-54 Example 2-16. Impact load on
a vertical bar
Because the arrangement of the bar and collar in this example matches the
arrangement shown in Fig. 2-53, we can use the equations derived previously
(Eqs. 2-49 to 2-60).
(a) Maximum elongation. The elongation of the bar produced by the falling
collar can be determined from Eq. (2-53). The first step is to determine the static
elongation of the bar due to the weight of the collar. Since the weight of the
collar is Mg, we calculate as follows:
(20.0 kg)(9.81 m/s2)(2.0 m)
MgL
0.0106 mm
dst
(210 GPa)(p/4)(15 mm)2
EA
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
134
CHAPTER 2 Axially Loaded Members
From this result we see that
150 mm
h
14,150
dst
0.0106 mm
The preceding numerical values may now be substituted into Eq. (2-53) to
obtain the maximum elongation:
2h
dmax dst 1 1
dst
1/2
(0.0106 mm)[1 1 2(14,150)]
1.79 mm
Since the height of fall is very large compared to the static elongation, we
obtain nearly the same result by calculating the maximum elongation from Eq.
(2-54):
dmax 2hdst [2(150 mm)(0.0106 mm)]1/2 1.78 mm
The impact factor is equal to the ratio of the maximum elongation to the static
elongation:
dmax
1.79 mm
Impact factor 169
dst
0.0106 mm
This result shows that the effects of a dynamically applied load can be very
large as compared to the effects of the same load acting statically.
(b) Maximum tensile stress. The maximum stress produced by the falling
collar is obtained from Eq. (2-55), as follows:
(210 GPa)(1.79 mm)
Edmax
188 MPa
smax
2.0 m
L
This stress may be compared with the static stress (see Eq. 2-57), which is
(20 kg)(9.81 m/s2)
W
Mg
1.11 MPa
sst
(p/4)(15 mm)2
A
A
The ratio of smax to sst is 188/1.11169, which is the same impact factor as for
the elongations. This result is expected, because the stresses are directly proportional to the corresponding elongations (see Eqs. 2-55 and 2-57).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.8 Impact Loading
135
Example 2-17
A horizontal bar AB of length L is struck at its free end by a heavy block of
mass M moving horizontally with velocity v (Fig. 2-55).
(a) Determine the maximum shortening dmax of the bar due to the impact
and determine the corresponding impact factor.
(b) Determine the maximum compressive stress smax and the corresponding impact factor. (Let EA represent the axial rigidity of the bar.)
Solution
d max
v
M
A
B
L
FIG. 2-55 Example 2-17. Impact load on
a horizontal bar
The loading on the bar in this example is quite different from the loads on
the bars pictured in Figs. 2-53 and 2-54. Therefore, we must make a new
analysis based upon conservation of energy.
(a) Maximum shortening of the bar. For this analysis we adopt the same
assumptions as those described previously. Thus, we disregard all energy losses
and assume that the kinetic energy of the moving block is transformed entirely
into strain energy of the bar.
The kinetic energy of the block at the instant of impact is Mv 2/2. The strain
energy of the bar when the block comes to rest at the instant of maximum shortening is EAd 2max/2L, as given by Eq. (2-37b). Therefore, we can write the
following equation of conservation of energy:
EAd 2max
Mv 2
2
2L
(2-63)
Solving for dmax, we get
dmax
M v 2L
EA
(2-64)
This equation is the same as Eq. (2-54), which we might have anticipated.
To find the impact factor, we need to know the static displacement of the
end of the bar. In this case the static displacement is the shortening of the bar
due to the weight of the block applied as a compressive load on the bar (see Eq.
2-51):
WL
MgL
dst
EA
EA
Thus, the impact factor is
dmax
Impact factor
dst
EAv2
Mg2L
(2-65)
The value determined from this equation may be much larger than 1.
(b) Maximum compressive stress in the bar. The maximum stress in the bar
is found from the maximum shortening by means of Eq. (2-55):
Edmax
E
smax
L
L
M v 2L
EA
M v 2E
AL
(2-66)
This equation is the same as Eq. (2-60).
The static stress sst in the bar is equal to W/A or Mg/A, which (in combination with Eq. 2-66) leads to the same impact factor as before (Eq. 2-65).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
136
CHAPTER 2 Axially Loaded Members
2.9 REPEATED LOADING AND FATIGUE
★
Load
O
Time
(a)
Load
O
Time
(b)
Load
O
Time
(c)
FIG. 2-56 Types of repeated loads:
(a) load acting in one direction only,
(b) alternating or reversed load, and
(c) fluctuating load that varies about an
average value
The behavior of a structure depends not only upon the nature of the
material but also upon the character of the loads. In some situations the
loads are static—they are applied gradually, act for long periods of time,
and change slowly. Other loads are dynamic in character—examples are
impact loads acting suddenly (Section 2.8) and repeated loads recurring
for large numbers of cycles.
Some typical patterns for repeated loads are sketched in Fig. 2-56.
The first graph (a) shows a load that is applied, removed, and applied
again, always acting in the same direction. The second graph (b) shows
an alternating load that reverses direction during every cycle of loading,
and the third graph (c) illustrates a fluctuating load that varies about an
average value. Repeated loads are commonly associated with machinery,
engines, turbines, generators, shafts, propellers, airplane parts, automobile parts, and the like. Some of these structures are subjected to millions
(and even billions) of loading cycles during their useful life.
A structure subjected to dynamic loads is likely to fail at a lower
stress than when the same loads are applied statically, especially when
the loads are repeated for a large number of cycles. In such cases failure
is usually caused by fatigue, or progressive fracture. A familiar
example of a fatigue failure is stressing a metal paper clip to the
breaking point by repeatedly bending it back and forth. If the clip is bent
only once, it does not break. But if the load is reversed by bending the
clip in the opposite direction, and if the entire loading cycle is repeated
several times, the clip will finally break. Fatigue may be defined as the
deterioration of a material under repeated cycles of stress and strain,
resulting in progressive cracking that eventually produces fracture.
In a typical fatigue failure, a microscopic crack forms at a point of
high stress (usually at a stress concentration, discussed in the next
section) and gradually enlarges as the loads are applied repeatedly.
When the crack becomes so large that the remaining material cannot
resist the loads, a sudden fracture of the material occurs (see Fig. 2-57
on the next page). Depending upon the nature of the material, it may
take anywhere from a few cycles of loading to hundreds of millions of
cycles to produce a fatigue failure.
The magnitude of the load causing a fatigue failure is less than the
load that can be sustained statically, as already pointed out. To determine the failure load, tests of the material must be performed. In the
case of repeated loading, the material is tested at various stress levels
and the number of cycles to failure is counted. For instance, a specimen
of material is placed in a fatigue-testing machine and loaded repeatedly
to a certain stress, say s1. The loading cycles are continued until failure
occurs, and the number n of loading cycles to failure is noted. The test is
then repeated for a different stress, say s2. If s2 is greater than s1, the
number of cycles to failure will be smaller. If s2 is less than s1, the
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.9
Repeated Loading and Fatigue
137
FIG. 2-57 Fatigue failure of a bar loaded
repeatedly in tension; the crack spread
gradually over the cross section until
fracture occurred suddenly. (Courtesy of
MTS Systems Corporation)
Failure
stress
s
Fatigue limit
O
Number n of cycles to failure
FIG. 2-58 Endurance curve, or S-N
diagram, showing fatigue limit
number will be larger. Eventually, enough data are accumulated to plot
an endurance curve, or S-N diagram, in which failure stress (S) is
plotted versus the number (N) of cycles to failure (Fig. 2-58). The
vertical axis is usually a linear scale and the horizontal axis is usually a
logarithmic scale.
The endurance curve of Fig. 2-58 shows that the smaller the stress,
the larger the number of cycles to produce failure. For some materials
the curve has a horizontal asymptote known as the fatigue limit or
endurance limit. When it exists, this limit is the stress below which a
fatigue failure will not occur regardless of how many times the load is
repeated. The precise shape of an endurance curve depends upon many
factors, including properties of the material, geometry of the test
specimen, speed of testing, pattern of loading, and surface condition of
the specimen. The results of numerous fatigue tests, made on a great
variety of materials and structural components, have been reported in
the engineering literature.
Typical S-N diagrams for steel and aluminum are shown in Fig. 2-59.
The ordinate is the failure stress, expressed as a percentage of the
ultimate stress for the material, and the abscissa is the number of cycles
at which failure occurred. Note that the number of cycles is plotted on a
logarithmic scale. The curve for steel becomes horizontal at about 107
cycles, and the fatigue limit is about 50% of the ultimate tensile stress
for ordinary static loading. The fatigue limit for aluminum is not as
100
80
Failure stress
(Percent of
60
ultimate
tensile stress)
40
Steel
Aluminum
20
FIG. 2-59 Typical endurance curves for
steel and aluminum in alternating
(reversed) loading
0
103
104
105
106
107
Number n of cycles to failure
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
108
138
CHAPTER 2 Axially Loaded Members
clearly defined as that for steel, but a typical value of the fatigue limit is
the stress at 5108 cycles, or about 25% of the ultimate stress.
Since fatigue failures usually begin with a microscopic crack at a point
of high localized stress (that is, at a stress concentration), the condition
of the surface of the material is extremely important. Highly polished
specimens have higher endurance limits. Rough surfaces, especially those
at stress concentrations around holes or grooves, greatly lower the
endurance limit. Corrosion, which creates tiny surface irregularities, has a
similar effect. For steel, ordinary corrosion may reduce the fatigue limit by
more than 50%.
2.10 STRESS CONCENTRATIONS
★
P
s= P
bt
s= P
bt
s= P
bt
When determining the stresses in axially loaded bars, we customarily
use the basic formula s P/A, in which P is the axial force in the bar
and A is its cross-sectional area. This formula is based upon the assumption that the stress distribution is uniform throughout the cross section.
In reality, bars often have holes, grooves, notches, keyways, shoulders,
threads, or other abrupt changes in geometry that create a disruption in
the otherwise uniform stress pattern. These discontinuities in geometry
cause high stresses in very small regions of the bar, and these high
stresses are known as stress concentrations. The discontinuities themselves are known as stress raisers.
Stress concentrations also appear at points of loading. For instance,
a load may act over a very small area and produce high stresses in the
region around its point of application. An example is a load applied
through a pin connection, in which case the load is applied over the
bearing area of the pin.
The stresses existing at stress concentrations can be determined
either by experimental methods or by advanced methods of analysis,
including the finite-element method. The results of such research for
many cases of practical interest are readily available in the engineering
literature (for example, Ref. 2-9). Some typical stress-concentration data
are given later in this section and also in Chapters 3 and 5.
Saint-Venant’s Principle
b
FIG. 2-60 Stress distributions near the end
of a bar of rectangular cross section
(width b, thickness t) subjected to a
concentrated load P acting over a
small area
To illustrate the nature of stress concentrations, consider the stresses in a
bar of rectangular cross section (width b, thickness t) subjected to a
concentrated load P at the end (Fig. 2-60). The peak stress directly
under the load may be several times the average stress P/bt, depending
upon the area over which the load is applied. However, the maximum
stress diminishes rapidly as we move away from the point of load
application, as shown by the stress diagrams in the figure. At a distance
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.10 Stress Concentrations
139
from the end of the bar equal to the width b of the bar, the stress distribution is nearly uniform, and the maximum stress is only a few percent
larger than the average stress. This observation is true for most stress
concentrations, such as holes and grooves.
Thus, we can make a general statement that the equation s P/A
gives the axial stresses on a cross section only when the cross section is
at least a distance b away from any concentrated load or discontinuity in
shape, where b is the largest lateral dimension of the bar (such as the
width or diameter).
The preceding statement about the stresses in a prismatic bar is part
of a more general observation known as Saint-Venant’s principle.
With rare exceptions, this principle applies to linearly elastic bodies of
all types. To understand Saint-Venant’s principle, imagine that we have
a body with a system of loads acting over a small part of its surface. For
instance, suppose we have a prismatic bar of width b subjected to a
system of several concentrated loads acting at the end (Fig. 2-61a). For
simplicity, assume that the loads are symmetrical and have only a
vertical resultant.
Next, consider a different but statically equivalent load system acting
over the same small region of the bar. (“Statically equivalent” means the
two load systems have the same force resultant and same moment resultant.)
For instance, the uniformly distributed load shown in Fig. 2-61b is
statically equivalent to the system of concentrated loads shown in
Fig. 2-61a. Saint-Venant’s principle states that the stresses in the body
caused by either of the two systems of loading are the same, provided we
move away from the loaded region a distance at least equal to the largest
b
b
(a)
(b)
FIG. 2-61 Illustration of Saint-Venant’s
principle: (a) system of concentrated
loads acting over a small region of a bar,
and (b) statically equivalent system
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
140
CHAPTER 2 Axially Loaded Members
dimension of the loaded region (distance b in our example). Thus, the
stress distributions shown in Fig. 2-60 are an illustration of Saint-Venant’s
principle. Of course, this “principle” is not a rigorous law of mechanics but
is a common-sense observation based upon theoretical and practical
experience.
Saint-Venant’s principle has great practical significance in the design
and analysis of bars, beams, shafts, and other structures encountered in
mechanics of materials. Because the effects of stress concentrations are
localized, we can use all of the standard stress formulas (such as s P/A)
at cross sections a sufficient distance away from the source of the
concentration. Close to the source, the stresses depend upon the details
of the loading and the shape of the member. Furthermore, formulas that
are applicable to entire members, such as formulas for elongations,
displacements, and strain energy, give satisfactory results even when
stress concentrations are present. The explanation lies in the fact that
stress concentrations are localized and have little effect on the overall
behavior of a member.*
Stress-Concentration Factors
Now let us consider some particular cases of stress concentrations
caused by discontinuities in the shape of a bar. We begin with a bar of
rectangular cross section having a circular hole and subjected to a tensile
force P (Fig. 2-62a). The bar is relatively thin, with its width b being
much larger than its thickness t. Also, the hole has diameter d.
c/2
P
b
P
d
c/2
(a)
smax
P
FIG. 2-62 Stress distribution in a flat bar
with a circular hole
(b)
*Saint-Venant’s principle is named for Barré de Saint-Venant (1797–1886), a famous
French mathematician and elastician (Ref. 2-10). It appears that the principle applies
generally to solid bars and beams but not to all thin-walled open sections. For a discussion of the limitations of Saint-Venant’s principle, see Ref. 2-11.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.10 Stress Concentrations
141
The normal stress acting on the cross section through the center
of the hole has the distribution shown in Fig. 2-62b. The maximum
stress smax occurs at the edges of the hole and may be significantly
larger than the nominal stress s P/ct at the same cross section. (Note
that ct is the net area at the cross section through the hole.) The intensity
of a stress concentration is usually expressed by the ratio of the
maximum stress to the nominal stress, called the stress-concentration
factor K:
max
K
(2-67)
nom
For a bar in tension, the nominal stress is the average stress based upon
the net cross-sectional area. In other cases, a variety of stresses may be
used. Thus, whenever a stress concentration factor is used, it is important
to note carefully how the nominal stress is defined.
A graph of the stress-concentration factor K for a bar with a hole is
given in Fig. 2-63. If the hole is tiny, the factor K equals 3, which means
that the maximum stress is three times the nominal stress. As the hole
becomes larger in proportion to the width of the bar, K becomes smaller
and the effect of the concentration is not as severe.
From Saint-Venant’s principle we know that, at distances equal to
the width b of the bar away from the hole in either axial direction, the
stress distribution is practically uniform and equal to P divided by the
gross cross-sectional area (s P/bt).
3.0
2.5
K
P
c/2
b
2.0
c/2
s
K = s max
nom
1.5
FIG. 2-63 Stress-concentration factor K
for flat bars with circular holes
P
d
0
P
s nom = ct
0.1
t = thickness
0.2
0.3
d
b
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
0.4
0.5
142
CHAPTER 2 Axially Loaded Members
Stress-concentration factors for two other cases of practical interest
are given in Figs. 2-64 and 2-65. These graphs are for flat bars and
circular bars, respectively, that are stepped down in size, forming a
shoulder. To reduce the stress-concentration effects, fillets are used to
round off the re-entrant corners.* Without the fillets, the stress-concentration factors would be extremely large, as indicated at the left-hand
side of each graph where K approaches infinity as the fillet radius R
approaches zero. In both cases the maximum stress occurs in the smaller
part of the bar in the region of the fillet.**
3.0
1.5
2.5
R
b =2
c
P
c
1.3
K
1.1
P
bb
s
K = smax
nom
1.2
2.0
snom = P
ct
t = thickness
R= b–c
2
FIG. 2-64 Stress-concentration factor K
1.5
for flat bars with shoulder fillets. The
dashed line is for a full quarter-circular
fillet.
0
0.05
0.15
R
c
0.10
0.20
3.0
P
D2
1.5
s
K = s max
nom
1.2
K
0.30
R
D2
=2
D1
2.5
0.25
1.1
D1
s nom =
P
P
p D21/4
2.0
FIG. 2-65 Stress-concentration factor K
for round bars with shoulder fillets. The
dashed line is for a full quarter-circular
fillet.
R=
1.5
0
D2 – D1
2
0.05
0.10
0.15
R
D1
0.20
0.25
0.30
*A fillet is a curved concave surface formed where two other surfaces meet. Its purpose
is to round off what would otherwise be a sharp re-entrant corner.
**The stress-concentration factors given in the graphs are theoretical factors for bars of
linearly elastic material. The graphs are plotted from the formulas given in Ref. 2-9.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.10 Stress Concentrations
143
Designing for Stress Concentrations
Because of the possibility of fatigue failures, stress concentrations are
especially important when the member is subjected to repeated loading.
As explained in the preceding section, cracks begin at the point of
highest stress and then spread gradually through the material as the load
is repeated. In practical design, the fatigue limit (Fig. 2-58) is considered
to be the ultimate stress for the material when the number of cycles is
extremely large. The allowable stress is obtained by applying a factor of
safety with respect to this ultimate stress. Then the peak stress at the
stress concentration is compared with the allowable stress.
In many situations the use of the full theoretical value of the stressconcentration factor is too severe. Fatigue tests usually produce failure
at higher levels of the nominal stress than those obtained by dividing the
fatigue limit by K. In other words, a structural member under repeated
loading is not as sensitive to a stress concentration as the value of K
indicates, and a reduced stress-concentration factor is often used.
Other kinds of dynamic loads, such as impact loads, also require
that stress-concentration effects be taken into account. Unless better
information is available, the full stress-concentration factor should be
used. Members subjected to low temperatures also are highly susceptible
to failures at stress concentrations, and therefore special precautions
should be taken in such cases.
The significance of stress concentrations when a member is
subjected to static loading depends upon the kind of material. With
ductile materials, such as structural steel, a stress concentration can
often be ignored. The reason is that the material at the point of
maximum stress (such as around a hole) will yield and plastic flow
will occur, thus reducing the intensity of the stress concentration and
making the stress distribution more nearly uniform. On the other hand,
with brittle materials (such as glass) a stress concentration will remain
up to the point of fracture. Therefore, we can make the general
observation that with static loads and a ductile material the stressconcentration effect is not likely to be important, but with static loads
and a brittle material the full stress-concentration factor should be
considered.
Stress concentrations can be reduced in intensity by properly
proportioning the parts. Generous fillets reduce stress concentrations at
re-entrant corners. Smooth surfaces at points of high stress, such as on
the inside of a hole, inhibit the formation of cracks. Proper reinforcing
around holes can also be beneficial. There are many other techniques for
smoothing out the stress distribution in a structural member and thereby
reducing the stress-concentration factor. These techniques, which are
usually studied in engineering design courses, are extremely important
in the design of aircraft, ships, and machines. Many unnecessary structural failures have occurred because designers failed to recognize the
effects of stress concentrations and fatigue.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
144
CHAPTER 2 Axially Loaded Members
2.11 NONLINEAR BEHAVIOR
★
Up to this point, our discussions have dealt primarily with members and
structures composed of materials that follow Hooke’s law. Now we will
consider the behavior of axially loaded members when the stresses exceed
the proportional limit. In such cases the stresses, strains, and displacements depend upon the shape of the stress-strain curve in the region
beyond the proportional limit (see Section 1.3 for some typical stressstrain diagrams).
Nonlinear Stress-Strain Curves
For purposes of analysis and design, we often represent the actual stress-strain
curve of a material by an idealized stress-strain curve that can be expressed
as a mathematical function. Some examples are shown in Fig. 2-66. The
first diagram (Fig. 2-66a) consists of two parts, an initial linearly elastic
region followed by a nonlinear region defined by an appropriate mathematical
expression. The behavior of aluminum alloys can sometimes be represented
quite accurately by a curve of this type, at least in the region before the strains
become excessively large (compare Fig. 2-66a with Fig. 1-13).
In the second example (Fig. 2-66b), a single mathematical expression is used for the entire stress-strain curve. The best known expression
of this kind is the Ramberg-Osgood stress-strain law, which is described
later in more detail (see Eqs. 2-70 and 2-71).
s
=E
e
s=
ƒ(e
)
Nonlinear
s
Linearly elastic
O
e
(a)
s
s
s
=
in h
Stra
Perfectly plastic
s
Nonlinear
sY
Linearly elastic
O
O
e
(b)
eY
ning
arde
ƒ(
)
e
Linearly elastic
e
O
(c)
e
(d)
FIG. 2-66 Types of idealized material
behavior: (a) elastic-nonlinear
stress-strain curve, (b) general
nonlinear stress-strain curve, (c)
elastoplastic stress-strain curve,
and (d) bilinear stress-strain curve
The stress-strain diagram frequently used for structural steel is shown in
Fig. 2-66c. Because steel has a linearly elastic region followed by a region
of considerable yielding (see the stress-strain diagrams of Figs. 1-10 and
1-12), its behavior can be represented by two straight lines. The material is
assumed to follow Hooke’s law up to the yield stress sY, after which it
yields under constant stress, the latter behavior being known as perfect
plasticity. The perfectly plastic region continues until the strains are 10 or
20 times larger than the yield strain. A material having a stress-strain
diagram of this kind is called an elastoplastic material (or elastic-plastic
material).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.11 Nonlinear Behavior
145
Eventually, as the strain becomes extremely large, the stress-strain
curve for steel rises above the yield stress due to strain hardening, as
explained in Section 1.3. However, by the time strain hardening begins,
the displacements are so large that the structure will have lost its usefulness. Consequently, it is common practice to analyze steel structures on
the basis of the elastoplastic diagram shown in Fig. 2-66c, with the same
diagram being used for both tension and compression. An analysis made
with these assumptions is called an elastoplastic analysis, or simply,
plastic analysis, and is described in the next section.
Figure 2-66d shows a stress-strain diagram consisting of two lines
having different slopes, called a bilinear stress-strain diagram. Note
that in both parts of the diagram the relationship between stress and
strain is linear, but only in the first part is the stress proportional to the
strain (Hooke’s law). This idealized diagram may be used to represent
materials with strain hardening or it may be used as an approximation to
diagrams of the general nonlinear shapes shown in Figs. 2-66a and b.
Changes in Lengths of Bars
The elongation or shortening of a bar can be determined if the stress-strain
curve of the material is known. To illustrate the general procedure, we
will consider the tapered bar AB shown in Fig. 2-67a. Both the crosssectional area and the axial force vary along the length of the bar, and the
material has a general nonlinear stress-strain curve (Fig. 2-67b). Because
the bar is statically determinate, we can determine the internal axial forces
at all cross sections from static equilibrium alone. Then we can find the
stresses by dividing the forces by the cross-sectional areas, and we can
A
B
x
dx
L
(a)
s
FIG. 2-67 Change in length of a tapered
bar consisting of a material having a
nonlinear stress-strain curve
O
e
(b)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
146
CHAPTER 2 Axially Loaded Members
find the strains from the stress-strain curve. Lastly, we can determine the
change in length from the strains, as described in the following paragraph.
The change in length of an element dx of the bar (Fig. 2-67a) is e
dx, where e is the strain at distance x from the end. By integrating this
expression from one end of the bar to the other, we obtain the change in
length of the entire bar:
L
dx
(2-68)
0
where L is the length of the bar. If the strains are expressed analytically,
that is, by algebraic formulas, it may be possible to integrate Eq. (2-68)
by formal mathematical means and thus obtain an expression for the
change in length. If the stresses and strains are expressed numerically,
that is, by a series of numerical values, we can proceed as follows. We
can divide the bar into small segments of length ⌬x, determine the
average stress and strain for each segment, and then calculate the elongation of the entire bar by summing the elongations for the individual
segments. This process is equivalent to evaluating the integral in Eq.
(2-68) by numerical methods instead of by formal integration.
If the strains are uniform throughout the length of the bar, as in the
case of a prismatic bar with constant axial force, the integration of
Eq. (2-68) is trivial and the change in length is
(2-69)
d eL
as expected (compare with Eq. 1-2 in Section 1.2).
Ramberg-Osgood Stress-Strain Law
Stress-strain curves for several metals, including aluminum and magnesium,
can be accurately represented by the Ramberg-Osgood equation:
m
e
s
s
a
e0
s0
s0
(2-70)
In this equation, s and e are the stress and strain, respectively, and e 0,
s0, a, and m are constants of the material (obtained from tension tests).
An alternative form of the equation is
m
s
sa s
e 0
E
E s0
(2-71)
in which Es0 /e0 is the modulus of elasticity in the initial part of the
stress-strain curve.*
A graph of Eq. (2-71) is given in Fig. 2-68 for an aluminum
alloy for which the constants are as follows: E 10 106 psi,
*The Ramberg-Osgood stress-strain law was presented in Ref. 2-12.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
147
SECTION 2.11 Nonlinear Behavior
s (psi)
50,000
40,000
30,000
20,000
Aluminum alloy
E = 10 × 106 psi
10
s
e =
+ 1 s
614.0 38,000
10 × 106
10,000
FIG. 2-68 Stress-strain curve for an
aluminum alloy using the RambergOsgood equation (Eq. 2-72)
s = psi
0
0.010
0.020
0.030
e
s0 38,000 psi, a 3/7, and m 10. The equation of this particular
stress-strain curve is
s
1
s
e
6
10 10
614.0 38,000
10
(2-72)
where s has units of pounds per square inch (psi).
A similar equation for an aluminum alloy, but in SI units (E
70 GPa, s0 260 MPa, a 3/7, and m 10), is as follows:
s
1
s 10
(2-73)
e
70,000
628.2 260
where s has units of megapascals (MPa). The calculation of the change
in length of a bar, using Eq. (2-73) for the stress-strain relationship, is
illustrated in Example 2-18.
Statically Indeterminate Structures
If a structure is statically indeterminate and the material behaves
nonlinearly, the stresses, strains, and displacements can be found by
solving the same general equations as those described in Section 2.4 for
linearly elastic structures, namely, equations of equilibrium, equations of
compatibility, and force-displacement relations (or equivalent stress-strain
relations). The principal difference is that the force-displacement
relations are now nonlinear, which means that analytical solutions cannot
be obtained except in very simple situations. Instead, the equations must
be solved numerically, using a suitable computer program.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
148
CHAPTER 2 Axially Loaded Members
Example 2-18
A prismatic bar AB of length L 2.2 m and cross-sectional area A 480 mm2
supports two concentrated loads P1 108 kN and P2 27 kN, as shown in Fig.
2-69. The material of the bar is an aluminum alloy having a nonlinear stressstrain curve described by the following Ramberg-Osgood equation (Eq. 2-73):
10
s
1
s
e
70,000
628.2 260
in which s has units of MPa. (The general shape of this stress-strain curve is
shown in Fig. 2-68.)
Determine the displacement dB of the lower end of the bar under each of
the following conditions: (a) the load P1 acts alone, (b) the load P2 acts alone,
and (c) the loads P1 and P2 act simultaneously.
A
L
2
Solution
P2
(a) Displacement due to the load P1 acting alone. The load P1 produces a
uniform tensile stress throughout the length of the bar equal to P1/A, or 225
MPa. Substituting this value into the stress-strain relation gives e 0.003589.
Therefore, the elongation of the bar, equal to the displacement at point B, is (see
Eq. 2-69)
L
2
B
dB e L (0.003589)(2.2 m) 7.90 mm
P1
FIG. 2-69 Example 2-18. Elongation of a
bar of nonlinear material using the
Ramberg-Osgood equation
(b) Displacement due to the load P2 acting alone. The stress in the upper
half of the bar is P2/A or 56.25 MPa, and there is no stress in the lower half.
Proceeding as in part (a), we obtain the following elongation:
dB e L/2 (0.0008036)(1.1 m) 0.884 mm
(c) Displacement due to both loads acting simultaneously. The stress in
the lower half of the bar is P1/A and in the upper half is (P1 P2)/A. The corresponding stresses are 225 MPa and 281.25 MPa, and the corresponding strains
are 0.003589 and 0.007510 (from the Ramberg-Osgood equation). Therefore, the
elongation of the bar is
dB (0.003589)(1.1 m) (0.007510)(1.1 m)
3.95 mm 8.26 mm 12.2 mm
The three calculated values of dB illustrate an important principle pertaining to a
structure made of a material that behaves nonlinearly:
In a nonlinear structure, the displacement produced by two (or more) loads
acting simultaneously is not equal to the sum of the displacements produced by
the loads acting separately.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.12 Elastoplastic Analysis
149
2.12 ELASTOPLASTIC ANALYSIS
★
s
sY
Slope = E
O
eY
e
FIG. 2-70 Idealized stress-strain diagram
for an elastoplastic material, such as
structural steel
P
P
P
L
PY
PY = s Y A
PYL sYL
=
EA
E
EA
Slope =
L
dY =
O
dY
d
In the preceding section we discussed the behavior of structures when the
stresses in the material exceed the proportional limit. Now we will
consider a material of considerable importance in engineering design—
steel, the most widely used structural metal. Mild steel (or structural steel)
can be modeled as an elastoplastic material with a stress-strain diagram as
shown in Fig. 2-70. An elastoplastic material initially behaves in a linearly
elastic manner with a modulus of elasticity E. After plastic yielding
begins, the strains increase at a more-or-less constant stress, called the
yield stress sY. The strain at the onset of yielding is known as the yield
strain eY.
The load-displacement diagram for a prismatic bar of elastoplastic
material subjected to a tensile load (Fig. 2-71) has the same shape as the
stress-strain diagram. Initially, the bar elongates in a linearly elastic
manner and Hooke’s law is valid. Therefore, in this region of loading
we can find the change in length from the familiar formula d PL/EA.
Once the yield stress is reached, the bar may elongate without an
increase in load, and the elongation has no specific magnitude. The load
at which yielding begins is called the yield load PY and the corresponding elongation of the bar is called the yield displacement dY. Note that
for a single prismatic bar, the yield load PY equals sYA and the yield
displacement dY equals PY L/EA, or sY L/E. (Similar comments apply to
a bar in compression, provided buckling does not occur.)
If a structure consisting only of axially loaded members is statically
determinate (Fig. 2-72), its overall behavior follows the same pattern.
The structure behaves in a linearly elastic manner until one of its
members reaches the yield stress. Then that member will begin to elongate (or shorten) with no further change in the axial load in that
member. Thus, the entire structure will yield, and its load-displacement
diagram has the same shape as that for a single bar (Fig. 2-71).
FIG. 2-71 Load-displacement diagram for
a prismatic bar of elastoplastic material
P
FIG. 2-72 Statically determinate structure
consisting of axially loaded members
Statically Indeterminate Structures
The situation is more complex if an elastoplastic structure is statically
indeterminate. If one member yields, other members will continue to
resist any increase in the load. However, eventually enough members
will yield to cause the entire structure to yield.
To illustrate the behavior of a statically indeterminate structure, we
will use the simple arrangement shown in Fig. 2-73 on the next page.
This structure consists of three steel bars supporting a load P applied
through a rigid plate. The two outer bars have length L 1, the inner
bar has length L 2, and all three bars have the same cross-sectional
area A. The stress-strain diagram for the steel is idealized as shown in
Fig. 2-70, and the modulus of elasticity in the linearly elastic region
is E sY /eY.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
150
CHAPTER 2 Axially Loaded Members
As is normally the case with a statically indeterminate structure, we
begin the analysis with the equations of equilibrium and compatibility.
From equilibrium of the rigid plate in the vertical direction we obtain
2F1 F2 P
(a)
where F1 and F2 are the axial forces in the outer and inner bars, respectively.
Because the plate moves downward as a rigid body when the load is
applied, the compatibility equation is
d1 d 2
(b)
where d1 and d 2 are the elongations of the outer and inner bars, respectively.
Because they depend only upon equilibrium and geometry, the two
preceding equations are valid at all levels of the load P; it does not
matter whether the strains fall in the linearly elastic region or in the
plastic region.
When the load P is small, the stresses in the bars are less than the
yield stress sY and the material is stressed within the linearly elastic
region. Therefore, the force-displacement relations between the bar
forces and their elongations are
F1L1
d1
EA
F L
d 2 22
EA
(c)
Substituting in the compatibility equation (Eq. b), we get
F1L1 F2 L2
(d)
Solving simultaneously Eqs. (a) and (d), we obtain
PL2
F1
L1 2 L 2
F1
F1
L1
L1
L2
Rigid
plate
PL1
F
s 2 2
A(L1 2L 2)
A
FIG. 2-73 Elastoplastic analysis of a
statically indeterminate structure
(2-75a,b)
These equations for the forces and stresses are valid provided the
stresses in all three bars remain below the yield stress sY.
As the load P gradually increases, the stresses in the bars increase
until the yield stress is reached in either the inner bar or the outer bars.
Let us assume that the outer bars are longer than the inner bar, as
sketched in Fig. 2-73:
L1 L 2
P
(2-74a,b)
Thus, we have now found the forces in the bars in the linearly elastic
region. The corresponding stresses are
F
PL 2
s1 1
A(L1 2 L 2)
A
F2
PL1
F2
L1 2 L 2
(e)
Then the inner bar is more highly stressed than the outer bars (see Eqs.
2-75a and b) and will reach the yield stress first. When that happens, the
force in the inner bar is F2 sY A. The magnitude of the load P when
the yield stress is first reached in any one of the bars is called the yield
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.12 Elastoplastic Analysis
151
load PY. We can determine PY by setting F2 equal to sY A in Eq. (2-74b)
and solving for the load:
2L
PY s Y A 1 2
L1
PP
O
(2-76)
As long as the load P is less than PY, the structure behaves in a linearly
elastic manner and the forces in the bars can be determined from Eqs.
(2-74a and b).
The downward displacement of the rigid bar at the yield load, called
the yield displacement dY, is equal to the elongation of the inner bar
when its stress first reaches the yield stress sY:
P
PY
B
C
A
dY
dP
d
FIG. 2-74 Load-displacement diagram for
the statically indeterminate structure
shown in Fig. 2-73
F L
s L
s L
dY 2 2 22 Y2
EA
E
E
(2-77)
The relationship between the applied load P and the downward displacement d of the rigid bar is portrayed in the load-displacement diagram of
Fig. 2-74. The behavior of the structure up to the yield load PY is represented by line OA.
With a further increase in the load, the forces F1 in the outer bars
increase but the force F2 in the inner bar remains constant at the value
sY A because this bar is now perfectly plastic (see Fig. 2-71). When the
forces F1 reach the value sY A, the outer bars also yield and therefore
the structure cannot support any additional load. Instead, all three bars
will elongate plastically under this constant load, called the plastic load
PP. The plastic load is represented by point B on the load-displacement
diagram (Fig. 2-74), and the horizontal line BC represents the region of
continuous plastic deformation without any increase in the load.
The plastic load PP can be calculated from static equilibrium (Eq. a)
knowing that
F1 sY A
F2 sY A
(f)
Thus, from equilibrium we find
PP 3sY A
(2-78)
The plastic displacement d P at the instant the load just reaches the
plastic load PP is equal to the elongation of the outer bars at the instant
they reach the yield stress. Therefore,
FL
s L
s L
dP 11 11 Y1
EA
E
E
(2-79)
Comparing d P with d Y, we see that in this example the ratio of the
plastic displacement to the yield displacement is
dP
L
1
dY
L2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(2-80)
152
CHAPTER 2 Axially Loaded Members
Also, the ratio of the plastic load to the yield load is
PP
3L1
PY
L1 2L2
(2-81)
For example, if L 1 1.5L 2, the ratios are dP /dY 1.5 and
PP /PY 9/7 1.29. In general, the ratio of the displacements is always
larger than the ratio of the corresponding loads, and the partially plastic
region AB on the load-displacement diagram (Fig. 2-74) always has a
smaller slope than does the elastic region OA. Of course, the fully
plastic region BC has the smallest slope (zero).
General Comments
To understand why the load-displacement graph is linear in the partially
plastic region (line AB in Fig. 2-74) and has a slope that is less than in
the linearly elastic region, consider the following. In the partially plastic
region of the structure, the outer bars still behave in a linearly elastic
manner. Therefore, their elongation is a linear function of the load.
Since their elongation is the same as the downward displacement of the
rigid plate, the displacement of the rigid plate must also be a linear
function of the load. Consequently, we have a straight line between
points A and B. However, the slope of the load-displacement diagram in
this region is less than in the initial linear region because the inner bar
yields plastically and only the outer bars offer increasing resistance to
the increasing load. In effect, the stiffness of the structure has diminished.
From the discussion associated with Eq. (2-78) we see that the
calculation of the plastic load PP requires only the use of statics, because
all members have yielded and their axial forces are known. In contrast,
the calculation of the yield load PY requires a statically indeterminate
analysis, which means that equilibrium, compatibility, and forcedisplacement equations must be solved.
After the plastic load PP is reached, the structure continues to deform
as shown by line BC on the load-displacement diagram (Fig. 2-74). Strain
hardening occurs eventually, and then the structure is able to support
additional loads. However, the presence of very large displacements
usually means that the structure is no longer of use, and so the plastic load
PP is usually considered to be the failure load.
The preceding discussion has dealt with the behavior of a structure
when the load is applied for the first time. If the load is removed before
the yield load is reached, the structure will behave elastically and return
to its original unstressed condition. However, if the yield load is exceeded,
some members of the structure will retain a permanent set when the load
is removed, thus creating a prestressed condition. Consequently, the
structure will have residual stresses in it even though no external loads
are acting. If the load is applied a second time, the structure will behave
in a different manner.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 2.12 Elastoplastic Analysis
153
Example 2-19
The structure shown in Fig. 2-75a consists of a horizontal beam AB (assumed to
be rigid) supported by two identical bars (bars 1 and 2) made of an elastoplastic
material. The bars have length L and cross-sectional area A, and the material has
yield stress sY, yield strain eY, and modulus of elasticity E s Y /e Y. The beam
has length 3b and supports a load P at end B.
(a) Determine the yield load PY and the corresponding yield displacement
dY at the end of the bar (point B).
(b) Determine the plastic load PP and the corresponding plastic displacement dP at point B.
(c) Construct a load-displacement diagram relating the load P to the
displacement dB of point B.
;@À;@À
F1
1
F2
L
2
A
PP = 6 PY
5
PY
B
b
b
b
FIG. 2-75 Example 2-19. Elastoplastic
analysis of a statically indeterminate
structure
P
O
P
B
C
A
dY
dP = 2 d Y
(a)
dB
(b)
Solution
Equation of equilibrium. Because the structure is statically indeterminate,
we begin with the equilibrium and compatibility equations. Considering the
equilibrium of beam AB, we take moments about point A and obtain
MA 0
F1(b) F2(2b) P(3b) 0
in which F1 and F2 are the axial forces in bars 1 and 2, respectively. This equation simplifies to
F1 2F2 3P
(g)
Equation of compatibility. The compatibility equation is based upon the
geometry of the structure. Under the action of the load P the rigid beam rotates
about point A, and therefore the downward displacement at every point along
the beam is proportional to its distance from point A. Thus, the compatibility
equation is
d 2 2d1
(h)
where d2 is the elongation of bar 2 and d1 is the elongation of bar 1.
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
154
CHAPTER 2 Axially Loaded Members
(a) Yield load and yield displacement. When the load P is small and the
stresses in the material are in the linearly elastic region, the force-displacement
relations for the two bars are
F1L
d 1
EA
F2L
d 2
EA
(i,j)
Combining these equations with the compatibility condition (Eq. h) gives
FL
FL
2 2 1
EA
EA
or
F2 2F1
(k)
Now substituting into the equilibrium equation (Eq. g), we find
3P
F1
5
6P
F2
5
(l,m)
Bar 2, which has the larger force, will be the first to reach the yield stress. At
that instant the force in bar 2 will be F2 sY A. Substituting that value into Eq.
(m) gives the yield load PY, as follows:
5sY A
PY
6
(2-82)
The corresponding elongation of bar 2 (from Eq. j) is d 2 s Y L /E, and therefore the yield displacement at point B is
3sY L
32
dY
2E
2
(2-83)
Both PY and dY are indicated on the load-displacement diagram (Fig. 2-75b).
(b) Plastic load and plastic displacement. When the plastic load PP is
reached, both bars will be stretched to the yield stress and both forces F1 and F2
will be equal to s Y A. It follows from equilibrium (Eq. g) that the plastic load is
PP s Y A
(2-84)
At this load, the left-hand bar (bar 1) has just reached the yield stress; therefore,
its elongation (from Eq. i) is d1 s Y L/E, and the plastic displacement of point B
is
3sYL
dP 3d1
E
(2-85)
The ratio of the plastic load PP to the yield load PY is 6/5, and the ratio of the
plastic displacement dP to the yield displacement d Y is 2. These values are also
shown on the load-displacement diagram.
(c) Load-displacement diagram. The complete load-displacement behavior
of the structure is pictured in Fig. 2-75b. The behavior is linearly elastic in the
region from O to A, partially plastic from A to B, and fully plastic from B to C.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 2 Problems
155
PROBLEMS CHAPTER 2
Changes in Lengths of Axially Loaded Members
2.2-3 A steel wire and a copper wire have equal lengths
2.2-1 The T-shaped arm ABC shown in the figure lies in a
and support equal loads P (see figure). The moduli of
elasticity for the steel and copper are Es 30,000 ksi and
Ec 18,000 ksi, respectively.
(a) If the wires have the same diameters, what is the
ratio of the elongation of the copper wire to the elongation
of the steel wire?
(b) If the wires stretch the same amount, what is the
ratio of the diameter of the copper wire to the diameter of
the steel wire?
vertical plane and pivots about a horizontal pin at A. The
arm has constant cross-sectional area and total weight W.
A vertical spring of stiffness k supports the arm at point B.
Obtain a formula for the elongation of the spring due
to the weight of the arm.
k
A
B
b
C
b
Copper
wire
b
PROB. 2.2-1
2.2-2 A steel cable with nominal diameter 25 mm (see
Table 2-1) is used in a construction yard to lift a bridge
section weighing 38 kN, as shown in the figure. The cable
has an effective modulus of elasticity E 140 GPa.
(a) If the cable is 14 m long, how much will it stretch
when the load is picked up?
(b) If the cable is rated for a maximum load of 70 kN,
what is the factor of safety with respect to failure of the
cable?
PROB. 2.2-2
Steel
wire
P
P
PROB. 2.2-3
2.2-4 By what distance h does the cage shown in the
figure move downward when the weight W is placed
inside it? (See the figure on the next page.)
Consider only the effects of the stretching of the
cable, which has axial rigidity EA 10,700 kN. The
pulley at A has diameter dA 300 mm and the pulley at
B has diameter dB 150 mm. Also, the distance L1
4.6 m, the distance L2 10.5 m, and the weight W
22 kN. (Note: When calculating the length of the cable,
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
156
CHAPTER 2 Axially Loaded Members
include the parts of the cable that go around the pulleys at
A and B.)
L1
A
pinned end A of the pointer. The device is adjusted so that
when there is no load P, the pointer reads zero on the
angular scale.
If the load P 8 N, at what distance x should the load
be placed so that the pointer will read 3° on the scale?
P
x
A
B
C
0
L2
k
b
B
Cage
W
PROB. 2.2-4
2.2-5 A safety valve on the top of a tank containing steam
under pressure p has a discharge hole of diameter d (see
figure). The valve is designed to release the steam when the
pressure reaches the value pmax.
If the natural length of the spring is L and its stiffness
is k, what should be the dimension h of the valve? (Express
your result as a formula for h.)
PROB. 2.2-6
2.2-7 Two rigid bars, AB and CD, rest on a smooth horizontal surface (see figure). Bar AB is pivoted end A and bar
CD is pivoted at end D. The bars are connected to each
other by two linearly elastic springs of stiffness k. Before
the load P is applied, the lengths of the springs are such
that the bars are parallel and the springs are without stress.
Derive a formula for the displacement dC at point C
when the load P is acting. (Assume that the bars rotate
through very small angles under the action of the load P.)
b
b
b
A
B
C
h
P
D
d
PROB. 2.2-7
p
PROB. 2.2-5
2.2-6 The device shown in the figure consists of a pointer
ABC supported by a spring of stiffness k 800 N/m. The
spring is positioned at distance b 150 mm from the
2.2-8 The three-bar truss ABC shown in the figure has a
span L 3 m and is constructed of steel pipes having
cross-sectional area A 3900 mm2 and modulus of elasticity E 200 GPa. A load P act horizontally to the right at
joint C. (See the figure on the next page.)
(a) If P 650 kN, what is the horizontal displacement
of joint B?
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
157
CHAPTER 2 Problems
(b) What is the maximum permissible load Pmax if the
displacement of joint B is limited to 1.5 mm?
C
At what distance x from the left-hand spring should a
load P 18 N be placed in order to bring the bar to a horizontal position?
P
h
k1
L1
A
45°
45°
k2
L2
B
W
A
B
P
L
x
L
PROB. 2.2-8
PROB. 2.2-10
2.2-9 An aluminum wire having a diameter d 2 mm
and length L 3.8 m is subjected to a tensile load P
(see figure). The aluminum has modulus of elasticity
E 75 GPa.
If the maximum permissible elongation of the wire is
3.0 mm and the allowable stress in tension is 60 MPa, what
is the allowable load Pmax?
P
d
2.2-11 A hollow, circular, steel column (E 30,000 ksi)
is subjected to a compressive load P, as shown in the
figure. The column has length L 8.0 ft and outside diameter d 7.5 in. The load P 85 k.
If the allowable compressive stress is 7000 psi and the
allowable shortening of the column is 0.02 in., what is the
minimum required wall thickness tmin?
P
P
t
L
L
PROB. 2.2-9
d
2.2-10 A uniform bar AB of weight W 25 N is
supported by two springs, as shown in the figure. The
spring on the left has stiffness k1 300 N/m and natural
length L1 250 mm. The corresponding quantities for the
spring on the right are k2 400 N/m and L2 200 mm.
The distance between the springs is L 350 mm, and the
spring on the right is suspended from a support that is
distance h 80 mm below the point of support for the
spring on the left.
PROB. 2.2-11
2.2-12 The horizontal rigid beam ABCD is supported by
vertical bars BE and CF and is loaded by vertical forces
P1 400 kN and P2 360 kN acting at points A and D,
respectively (see figure on the next page). Bars BE and CF
are made of steel (E 200 GPa) and have cross-sectional
★
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
158
CHAPTER 2 Axially Loaded Members
areas ABE 11,100 mm2 and ACF 9,280 mm2. The
distances between various points on the bars are shown in
the figure.
Determine the vertical displacements dA and dD of
points A and D, respectively.
1.5 m
A
1.5 m
B
P
B
2.1 m
u
A
u
C
D
C
PROBS. 2.2-13 and 2.2-14
2.4 m
P1 = 400 kN
P2 = 360 kN
2.2-14 Solve the preceding problem for the following
data: b 200 mm, k 3.2 kN/m, 45°, and P 50 N.
★★
F
0.6 m
E
Changes in Lengths Under Nonuniform Conditions
2.3-1 Calculate the elongation of a copper bar of solid
PROB. 2.2-12
2.2-13 A framework ABC consists of two rigid bars AB
and BC, each having length b (see the first part of the figure
below). The bars have pin connections at A, B, and C and
are joined by a spring of stiffness k. The spring is attached
at the midpoints of the bars. The framework has a pin
support at A and a roller support at C, and the bars are at an
angle to the horizontal.
When a vertical load P is applied at joint B (see the
second part of the figure at the top of the next column) the
roller support C moves to the right, the spring is stretched,
and the angle of the bars decreases from to the angle .
Determine the angle and the increase in the
distance between points A and C. (Use the following data;
b 8.0 in., k 16 lb/in., 45°, and P 10 lb.)
★★
B
b
—
2
b
—
2
A
3.0 k
B
C
D
20 in.
50 in.
20 in.
3.0 k
PROB. 2.3-1
2.3-2 A long, rectangular copper bar under a tensile load P
b
—
2
b
—
2
k
a
A
circular cross section with tapered ends when it is stretched
by axial loads of magnitude 3.0 k (see figure).
The length of the end segments is 20 in. and the length
of the prismatic middle segment is 50 in. Also, the diameters
at cross sections A, B, C, and D are 0.5, 1.0, 1.0, and 0.5 in.,
respectively, and the modulus of elasticity is 18,000 ksi.
(Hint: Use the result of Example 2-4.)
a
C
hangs from a pin that is supported by two steel posts (see
figure on the next page). The copper bar has a length of
2.0 m, a cross-sectional area of 4800 mm2, and a modulus
of elasticity Ec 120 GPa. Each steel post has a height of
0.5 m, a cross-sectional area of 4500 mm2, and a modulus
of elasticity Es 200 GPa.
(a) Determine the downward displacement d of the
lower end of the copper bar due to a load P 180 kN.
(b) What is the maximum permissible load Pmax if the
displacement d is limited to 1.0 mm?
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
159
CHAPTER 2 Problems
b
—
4
P
Steel
post
t
b
L
—
4
P
L
—
2
L
—
4
PROBS. 2.3-4 and 2.3-5
2.3-5 Solve the preceding problem if the axial stress in the
Copper
bar
middle region is 24,000 psi, the length is 30 in., and the
modulus of elasticity is 30 106 psi.
P
PROB. 2.3-2
2.3-3 A steel bar AD (see figure) has a cross-sectional
area of 0.40 in.2 and is loaded by forces P1 2700 lb, P2
1800 lb, and P3 1300 lb. The lengths of the segments of
the bar are a 60 in., b 24 in., and c 36 in.
(a) Assuming that the modulus of elasticity E
30 106 psi, calculate the change in length d of the bar.
Does the bar elongate or shorten?
(b) By what amount P should the load P3 be increased
so that the bar does not change in length when the three
loads are applied?
P1
A
C
B
a
P2
b
D
2.3-6 A two-story building has steel columns AB in the
first floor and BC in the second floor, as shown in the
figure. The roof load P1 equals 400 kN and the secondfloor load P2 equals 720 kN. Each column has length L
3.75 m. The cross-sectional areas of the first- and secondfloor columns are 11,000 mm2 and 3,900 mm2,
respectively.
(a) Assuming that E 206 GPa, determine the total
shortening AC of the two columns due to the combined
action of the loads P1 and P2.
(b) How much additional load P0 can be placed at the
top of the column (point C) if the total shortening AC is not
to exceed 4.0 mm?
P1 = 400 kN
P3
P2 = 720 kN
c
C
L = 3.75 m
B
PROB. 2.3-3
L = 3.75 m
A
2.3-4 A rectangular bar of length L has a slot in the middle
half of its length (see figure). The bar has width b, thickness t, and modulus of elasticity E. The slot has width b/4.
(a) Obtain a formula for the elongation d of the bar due
to the axial loads P.
(b) Calculate the elongation of the bar if the material is
high-strength steel, the axial stress in the middle region is
160 MPa, the length is 750 mm, and the modulus of elasticity is 210 GPa.
PROB. 2.3-6
2.3-7 A steel bar 8.0 ft long has a circular cross section of
diameter d1 0.75 in. over one-half of its length and diameter d2 0.5 in. over the other half (see figure on the next
page). The modulus of elasticity E 30 106 psi.
(a) How much will the bar elongate under a tensile
load P 5000 lb?
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
160
CHAPTER 2 Axially Loaded Members
(b) If the same volume of material is made into a bar
of constant diameter d and length 8.0 ft, what will be the
elongation under the same load P?
d1 = 0.75 in.
P
d2 = 0.50 in.
4.0 ft
L
f
P = 5000 lb
P
4.0 ft
PROB. 2.3-7
2.3-8 A bar ABC of length L consists of two parts of equal
lengths but different diameters (see figure). Segment AB
has diameter d1 100 mm and segment BC has diameter
d2 60 mm. Both segments have length L/2 0.6 m. A
longitudinal hole of diameter d is drilled through segment
AB for one-half of its length (distance L/4 0.3 m).
The bar is made of plastic having modulus of elasticity
E 4.0 GPa. Compressive loads P 110 kN act at the
ends of the bar.
If the shortening of the bar is limited to 8.0 mm, what
is the maximum allowable diameter dmax of the hole?
A
B
L
—
4
L
—
4
2.3-10 A prismatic bar AB of length L, cross-sectional
area A, modulus of elasticity E, and weight W hangs vertically under its own weight (see figure).
(a) Derive a formula for the downward displacement
dC of point C, located at distance h from the lower end of
the bar.
(b) What is the elongation dB of the entire bar?
(c) What is the ratio of the elongation of the upper
half of the bar to the elongation of the lower half of the
bar?
A
d2
C
d1
P
PROB. 2.3-9
C
P
h
L
—
2
PROB. 2.3-8
L
B
PROB. 2.3-10
2.3-11 A flat bar of rectangular cross section, length L,
and constant thickness t is subjected to tension by forces P
(see figure on the next page). The width of the bar varies
linearly from b1 at the smaller end to b2 at the larger end.
Assume that the angle of taper is small.
(a) Derive the following formula for the elongation of
the bar:
★
2.3-9 A wood pile, driven into the earth, supports a load P
entirely by friction along its sides (see figure). The friction
force f per unit length of pile is assumed to be uniformly
distributed over the surface of the pile. The pile has length
L, cross-sectional area A, and modulus of elasticity E.
(a) Derive a formula for the shortening d of the pile in
terms of P, L, E, and A.
(b) Draw a diagram showing how the compressive
stress sc varies throughout the length of the pile.
PL
b
d ln 2
Et(b2 b1) b 1
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
161
CHAPTER 2 Problems
(b) Calculate the elongation, assuming L 5 ft,
t 1.0 in., P 25 k, b1 4.0 in., b2 6.0 in., and
E 30 106 psi.
b2
t
P
2.3-13 A long, slender bar in the shape of a right circular
cone with length L and base diameter d hangs vertically
under the action of its own weight (see figure). The weight
of the cone is W and the modulus of elasticity of the material
is E.
Derive a formula for the increase in the length of the
bar due to its own weight. (Assume that the angle of taper
of the cone is small.)
★
b1
d
L
P
PROB. 2.3-11
L
2.3-12 A post AB supporting equipment in a laboratory
is tapered uniformly throughout its height H (see figure).
The cross sections of the post are square, with dimensions
b b at the top and 1.5b 1.5b at the base.
Derive a formula for the shortening of the post due
to the compressive load P acting at the top. (Assume that
the angle of taper is small and disregard the weight of the
post itself.)
★
P
A
A
PROB. 2.3-13
2.3-14 A bar ABC revolves in a horizontal plane about
a vertical axis at the midpoint C (see figure). The bar,
which has length 2L and cross-sectional area A, revolves at
constant angular speed . Each half of the bar (AC and BC)
has weight W1 and supports a weight W2 at its end.
Derive the following formula for the elongation of
one-half of the bar (that is, the elongation of either AC or
BC):
★★
L22
(W1 3W2)
3g EA
b
b
in which E is the modulus of elasticity of the material of the
bar and g is the acceleration of gravity.
H
B
B
1.5b
1.5b
A
W2
C
W1
B
W1
L
PROB. 2.3-12
v
PROB. 2.3-14
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
L
W2
162
CHAPTER 2 Axially Loaded Members
2.3-15 The main cables of a suspension bridge [see
part (a) of the figure] follow a curve that is nearly parabolic
because the primary load on the cables is the weight of
the bridge deck, which is uniform in intensity along the
horizontal. Therefore, let us represent the central region
AOB of one of the main cables [see part (b) of the figure] as
a parabolic cable supported at points A and B and carrying
a uniform load of intensity q along the horizontal. The span
of the cable is L, the sag is h, the axial rigidity is EA, and
the origin of coordinates is at midspan.
(a) Derive the following formula for the elongation of
cable AOB shown in part (b) of the figure:
★★
qL3
16h 2
d (1 )
8hE A
3L2
Statically Indeterminate Structures
2.4-1 The assembly shown in the figure consists of a brass
core (diameter d1 0.25 in.) surrounded by a steel shell
(inner diameter d2 0.28 in., outer diameter d3 0.35 in.).
A load P compresses the core and shell, which have length
L 4.0 in. The moduli of elasticity of the brass and steel
are Eb 15 106 psi and Es 30 106 psi, respectively.
(a) What load P will compress the assembly by 0.003
in.?
(b) If the allowable stress in the steel is 22 ksi and the
allowable stress in the brass is 16 ksi, what is the allowable
compressive load Pallow? (Suggestion: Use the equations
derived in Example 2-5.)
P
(b) Calculate the elongation d of the central span of
one of the main cables of the Golden Gate Bridge, for
which the dimensions and properties are L 4200 ft, h
470 ft, q 12,700 lb/ft, and E 28,800,000 psi. The
cable consists of 27,572 parallel wires of diameter 0.196 in.
Hint: Determine the tensile force T at any point in the
cable from a free-body diagram of part of the cable; then
determine the elongation of an element of the cable of
length ds; finally, integrate along the curve of the cable to
obtain an equation for the elongation d.
Steel shell
Brass core
;@À ;;
;@À
ÀÀ
@@
L
d1
d2
d3
PROB. 2.4-1
(a)
2.4-2 A cylindrical assembly consisting of a brass core
y
L
—
2
A
L
—
2
B
h
q
O
(b)
PROB. 2.3-15
x
and an aluminum collar is compressed by a load P (see
figure on the next page). The length of the aluminum collar
and brass core is 350 mm, the diameter of the core is
25 mm, and the outside diameter of the collar is 40 mm.
Also, the moduli of elasticity of the aluminum and brass are
72 GPa and 100 GPa, respectively.
(a) If the length of the assembly decreases by 0.1%
when the load P is applied, what is the magnitude of the
load?
(b) What is the maximum permissible load Pmax if the
allowable stresses in the aluminum and brass are 80 MPa
and 120 MPa, respectively? (Suggestion: Use the equations
derived in Example 2-5.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 2 Problems
P
163
(b) Obtain a formula for the displacement dC of point C.
(c) What is the ratio of the stress s1 in region AC to
the stress s2 in region CB?
Aluminum collar
Brass core
A
A1
P
C
A2
B
350 mm
b1
25 mm
PROB. 2.4-4
40 mm
2.4-5 Three steel cables jointly support a load of 12 k (see
PROB. 2.4-2
2.4-3 Three prismatic bars, two of material A and one of
material B, transmit a tensile load P (see figure). The two
outer bars (material A) are identical. The cross-sectional
area of the middle bar (material B) is 50% larger than
the cross-sectional area of one of the outer bars. Also, the
modulus of elasticity of material A is twice that of
material B.
(a) What fraction of the load P is transmitted by the
middle bar?
(b) What is the ratio of the stress in the middle bar to
the stress in the outer bars?
(c) What is the ratio of the strain in the middle bar to
the strain in the outer bars?
A
B
b2
figure). The diameter of the middle cable is 3/4 in. and the
diameter of each outer cable is 1/2 in. The tensions in the
cables are adjusted so that each cable carries one-third of
the load (i.e., 4 k). Later, the load is increased by 9 k to a
total load of 21 k.
(a) What percent of the total load is now carried by the
middle cable?
(b) What are the stresses sM and sO in the middle and
outer cables, respectively? (Note: See Table 2-1 in Section
2.2 for properties of cables.)
P
A
PROB. 2.4-3
2.4-4 A bar ACB having two different cross-sectional
areas A1 and A2 is held between rigid supports at A and B
(see figure). A load P acts at point C, which is distance b1
from end A and distance b2 from end B.
(a) Obtain formulas for the reactions RA and RB at
supports A and B, respectively, due to the load P.
PROB. 2.4-5
¢¢
@@
ÀÀ
;;
QQ
@@
ÀÀ
;;
QQ
;@ÀQ¢¢¢
2.4-6 A plastic rod AB of length L 0.5 m has a diameter
d1 30 mm (see figure on the next page). A plastic sleeve
CD of length c 0.3 m and outer diameter d2 45 mm is
securely bonded to the rod so that no slippage can occur
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
164
CHAPTER 2 Axially Loaded Members
between the rod and the sleeve. The rod is made of an
acrylic with modulus of elasticity E1 3.1 GPa and the
sleeve is made of a polyamide with E2 2.5 GPa.
(a) Calculate the elongation of the rod when it is
pulled by axial forces P 12 kN.
(b) If the sleeve is extended for the full length of the
rod, what is the elongation?
(c) If the sleeve is removed, what is the elongation?
d1
d2
C
A
P
b
L
PROB. 2.4-6
2.4-7 The axially loaded bar ABCD shown in the figure is
held between rigid supports. The bar has cross-sectional
area A1 from A to C and 2A1 from C to D.
(a) Derive formulas for the reactions RA and RD at the
ends of the bar.
(b) Determine the displacements dB and dC at points B
and C, respectively.
(c) Draw a diagram in which the abscissa is the
distance from the left-hand support to any point in the bar
and the ordinate is the horizontal displacement at that
point.
A
B
L
—
4
C
L
—
4
A1
PB
A
PC
B
D
C
L2
L1
PROB. 2.4-8
2.4-9 The aluminum and steel pipes shown in the figure
are fastened to rigid supports at ends A and B and to a rigid
plate C at their junction. The aluminum pipe is twice as
long as the steel pipe. Two equal and symmetrically placed
loads P act on the plate at C.
(a) Obtain formulas for the axial stresses sa and ss in
the aluminum and steel pipes, respectively.
(b) Calculate the stresses for the following data: P 12 k,
cross-sectional area of aluminum pipe Aa 8.92 in.2, crosssectional area of steel pipe As 1.03 in.2, modulus of
elasticity of aluminum Ea 10 106 psi, and modulus of
elasticity of steel Es 29 106 psi.
A
Steel pipe
L
P
1.5A1
A1
P
A2
B
P
c
A1
L1
D
b
(b) Determine the compressive axial force FBC in the
middle segment of the bar.
P
C
D
L
—
2
Aluminum
pipe
2L
PROB. 2.4-7
B
2.4-8 The fixed-end bar ABCD consists of three prismatic
segments, as shown in the figure. The end segments have
cross-sectional area A1 840 mm2 and length L1 200 mm.
The middle segment has cross-sectional area A2 1260 mm2
and length L2 250 mm. Loads PB and PC are equal to
25.5 kN and 17.0 kN, respectively.
(a) Determine the reactions RA and RD at the fixed
supports.
PROB. 2.4-9
2.4-10 A rigid bar of weight W 800 N hangs from three
equally spaced vertical wires, two of steel and one of
aluminum (see figure on the next page). The wires also support
a load P acting at the midpoint of the bar. The diameter of the
steel wires is 2 mm, and the diameter of the aluminum wire is
4 mm.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 2 Problems
What load Pallow can be supported if the allowable
stress in the steel wires is 220 MPa and in the aluminum
wire is 80 MPa? (Assume Es 210 GPa and Ea 70 GPa.)
165
Determine the elongation AC of the bar due to the
load P. (Assume L1 2L3 250 mm, L2 225 mm, A1
2A3 960 mm2, and A2 300 mm2.)
A
S
A
A1
S
L1
Rigid bar
of weight W
B
L3
A3
D
P
PROB. 2.4-10
A2
2.4-11 A bimetallic bar (or composite bar) of square cross
section with dimensions 2b 2b is constructed of two
different metals having moduli of elasticity E1 and E2 (see
figure). The two parts of the bar have the same crosssectional dimensions. The bar is compressed by forces P
acting through rigid end plates. The line of action of the
loads has an eccentricity e of such magnitude that each part
of the bar is stressed uniformly in compression.
(a) Determine the axial forces P1 and P2 in the two
parts of the bar.
(b) Determine the eccentricity e of the loads.
(c) Determine the ratio s1/s2 of the stresses in the two
parts of the bar.
L2
C
P
PROB. 2.4-12
2.4-13 A horizontal rigid bar of weight W 7200 lb is
supported by three slender circular rods that are equally
spaced (see figure). The two outer rods are made of
aluminum (E1 10 106 psi) with diameter d1 0.4 in.
and length L1 40 in. The inner rod is magnesium
(E2 6.5 106 psi) with diameter d2 and length L2. The
allowable stresses in the aluminum and magnesium are
24,000 psi and 13,000 psi, respectively.
If it is desired to have all three rods loaded to their
maximum allowable values, what should be the diameter d2
and length L2 of the middle rod?
★
E2
P
e
b
b
E1
P
e
b
b
2b
d2
L2
d1
d1
PROB. 2.4-11
2.4-12 A circular steel bar ABC (E = 200 GPa) has crosssectional area A1 from A to B and cross-sectional area A2
from B to C (see figure). The bar is supported rigidly at end
A and is subjected to a load P equal to 40 kN at end C. A
circular steel collar BD having cross-sectional area A3
supports the bar at B. The collar fits snugly at B and D
when there is no load.
W = weight of rigid bar
PROB. 2.4-13
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
L1
166
★
CHAPTER 2 Axially Loaded Members
2.4-14 A rigid bar ABCD is pinned at point B and
supported by springs at A and D (see figure). The springs
at A and D have stiffnesses k1 10 kN/m and k2
25 kN/m, respectively, and the dimensions a, b, and c are
250 mm, 500 mm, and 200 mm, respectively. A load P
acts at point C.
If the angle of rotation of the bar due to the action of
the load P is limited to 3°, what is the maximum permissible load Pmax?
a = 250 mm
A
2.4-16 A trimetallic bar is uniformly compressed by an
axial force P 40 kN applied through a rigid end plate
(see figure). The bar consists of a circular steel core
surrounded by brass and copper tubes. The steel core has
diameter 30 mm, the brass tube has outer diameter 45 mm,
and the copper tube has outer diameter 60 mm. The
corresponding moduli of elasticity are Es 210 GPa,
Eb 100 GPa, and Ec 120 GPa.
Calculate the compressive stresses ss, sb, and sc in the
steel, brass, and copper, respectively, due to the force P.
★★
b = 500 mm
B
C
P = 40 kN
D
Copper tube
Brass tube
Steel core
c = 200 mm
P
k 2 = 25 kN/m
k1 = 10 kN/m
30
mm
PROB. 2.4-14
45
mm
2.4-15 A rigid bar AB of length L 66 in. is hinged to a
★★
support at A and supported by two vertical wires attached at
points C and D (see figure). Both wires have the same crosssectional area (A 0.0272 in.2) and are made of the same
material (modulus E 30 106 psi). The wire at C
has length h 18 in. and the wire at D has length twice
that amount. The horizontal distances are c 20 in. and
d 50 in.
(a) Determine the tensile stresses sC and sD in the
wires due to the load P 170 lb acting at end B of the bar.
(b) Find the downward displacement dB at end B of the
bar.
2h
h
A
C
D
P
L
PROB. 2.4-15
PROB. 2.4-16
Thermal Effects
2.5-1 The rails of a railroad track are welded together at
their ends (to form continuous rails and thus eliminate the
clacking sound of the wheels) when the temperature is
60°F.
What compressive stress s is produced in the rails
when they are heated by the sun to 120°F if the coefficient
of thermal expansion a 6.5 106/°F and the modulus
of elasticity E 30 106 psi?
2.5-2 An aluminum pipe has a length of 60 m at a
B
c
d
60
mm
temperature of 10°C. An adjacent steel pipe at the same
temperature is 5 mm longer than the aluminum pipe.
At what temperature (degrees Celsius) will the
aluminum pipe be 15 mm longer than the steel pipe?
(Assume that the coefficients of thermal expansion of
aluminum and steel are aa 23 106/°C and
as 12 106/°C, respectively.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
167
CHAPTER 2 Problems
2.5-3 A rigid bar of weight W 750 lb hangs from three
∆TB
∆T
equally spaced wires, two of steel and one of aluminum
(see figure). The diameter of the wires is 1/8 in. Before
they were loaded, all three wires had the same length.
What temperature increase T in all three wires will
result in the entire load being carried by the steel wires?
(Assume Es 30 106 psi, as 6.5 106/°F, and aa
12 106/°F.)
0
B
A
x
L
PROB. 2.5-5
S
A
S
W = 750 lb
PROB. 2.5-3
2.5-6 A plastic bar ACB having two different solid circular
cross sections is held between rigid supports as shown in
the figure. The diameters in the left- and right-hand parts
are 50 mm and 75 mm, respectively. The corresponding
lengths are 225 mm and 300 mm. Also, the modulus of
elasticity E is 6.0 GPa, and the coefficient of thermal
expansion a is 100 106/°C. The bar is subjected to a
uniform temperature increase of 30°C.
Calculate the following quantities: (a) the compressive
force P in the bar; (b) the maximum compressive stress sc;
and (c) the displacement dC of point C.
2.5-4 A steel rod of diameter 15 mm is held snugly (but
without any initial stresses) between rigid walls by the
arrangement shown in the figure.
Calculate the temperature drop T (degrees Celsius) at
which the average shear stress in the 12-mm diameter bolt
becomes 45 MPa. (For the steel rod, use a 12 106/°C
and E 200 GPa.)
12 mm diameter bolt
15 mm
PROB. 2.5-4
A
225 mm
B
300 mm
PROB. 2.5-6
2.5-7 A circular steel rod AB (diameter d1 1.0 in., length
L1 3.0 ft) has a bronze sleeve (outer diameter d2
1.25 in., length L2 1.0 ft) shrunk onto it so that the two
parts are securely bonded (see figure).
Calculate the total elongation d of the steel bar due to a
temperature rise T 500°F. (Material properties are as
follows: for steel, Es 30 106 psi and as 6.5 106/°F;
for bronze, Eb 15 106 psi and ab 11 106/°F.)
2.5-5 A bar AB of length L is held between rigid supports
and heated nonuniformly in such a manner that the temperature increase T at distance x from end A is given by the
expression T TBx3/L3, where TB is the increase in
temperature at end B of the bar (see figure).
Derive a formula for the compressive stress sc in the
bar. (Assume that the material has modulus of elasticity E
and coefficient of thermal expansion a.)
75 mm
50 mm C
d1
d2
A
B
L2
L1
PROB. 2.5-7
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
168
CHAPTER 2 Axially Loaded Members
2.5-8 A brass sleeve S is fitted over a steel bolt B (see
figure), and the nut is tightened until it is just snug. The bolt
has a diameter dB 25 mm, and the sleeve has inside and
outside diameters d1 26 mm and d2 36 mm, respectively.
Calculate the temperature rise T that is required to
produce a compressive stress of 25 MPa in the sleeve.
(Use material properties as follows: for the sleeve, aS
21 106/°C and ES 100 GPa; for the bolt, aB 10
106/°C and EB 200 GPa.) (Suggestion: Use the results
of Example 2-8.)
d2
d1
(Note: The cables have effective modulus of elasticity
E 140 GPa and coefficient of thermal expansion a
12 106/°C. Other properties of the cables can be found
in Table 2-1, Section 2.2.)
A
C
B
2b
dB
dC
dB
2b
D
b
Sleeve (S)
P
PROB. 2.5-10
2.5-11 A rigid triangular frame is pivoted at C and held
by two identical horizontal wires at points A and B (see
figure). Each wire has axial rigidity EA 120 k and coefficient of thermal expansion a 12.5 106/°F.
(a) If a vertical load P 500 lb acts at point D, what
are the tensile forces TA and TB in the wires at A and B,
respectively?
(b) If, while the load P is acting, both wires have their
temperatures raised by 180°F, what are the forces TA and TB?
(c) What further increase in temperature will cause the
wire at B to become slack?
★
Bolt (B)
PROB. 2.5-8
2.5-9 Rectangular bars of copper and aluminum are held
by pins at their ends, as shown in the figure. Thin spacers
provide a separation between the bars. The copper bars
have cross-sectional dimensions 0.5 in. 2.0 in., and the
aluminum bar has dimensions 1.0 in. 2.0 in.
Determine the shear stress in the 7/16 in. diameter
pins if the temperature is raised by 100°F. (For copper,
Ec 18,000 ksi and ac 9.5 106/°F; for aluminum,
Ea 10,000 ksi and aa 13 106/°F.) Suggestion: Use
the results of Example 2-8.
A
b
B
Copper bar
b
Aluminum bar
D
C
P
Copper bar
2b
PROB. 2.5-9
2.5-10 A rigid bar ABCD is pinned at end A and
PROB. 2.5-11
★
supported by two cables at points B and C (see figure). The
cable at B has nominal diameter dB 12 mm and the cable
at C has nominal diameter dC 20 mm. A load P acts at
end D of the bar.
What is the allowable load P if the temperature rises
by 60°C and each cable is required to have a factor of
safety of at least 5 against its ultimate load?
Misfits and Prestrains
2.5-12 A steel wire AB is stretched between rigid supports
(see figure). The initial prestress in the wire is 42 MPa
when the temperature is 20°C.
(a) What is the stress s in the wire when the temperature drops to 0°C?
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 2 Problems
(b) At what temperature T will the stress in the wire
become zero? (Assume a 14 106/°C and E 200
GPa.)
A
B
169
the bar is in a vertical position, the length of each wire is
L 80 in. However, before being attached to the bar, the
length of wire B was 79.98 in. and of wire C was 79.95 in.
Find the tensile forces TB and TC in the wires under the
action of a force P 700 lb acting at the upper end of the
bar.
Steel wire
700 lb
PROB. 2.5-12
2.5-13 A copper bar AB of length 25 in. is placed in
position at room temperature with a gap of 0.008 in.
between end A and a rigid restraint (see figure).
Calculate the axial compressive stress sc in the bar if
the temperature rises 50°F. (For copper, use a 9.6
106/°F and E 16 106 psi.)
B
b
C
b
b
80 in.
0.008 in.
A
PROB. 2.5-15
2.5-16 A rigid steel plate is supported by three posts of high-
25 in.
strength concrete each having an effective cross-sectional
area A 40,000 mm2 and length L 2 m (see figure).
Before the load P is applied, the middle post is shorter than
the others by an amount s 1.0 mm.
Determine the maximum allowable load Pallow if the
allowable compressive stress in the concrete is sallow
20 MPa. (Use E 30 GPa for concrete.)
B
PROB. 2.5-13
2.5-14 A bar AB having length L and axial rigidity EA is
;;
@
À
@
À
@;
À
;
@
À
P
fixed at end A (see figure). At the other end a small gap of
dimension s exists between the end of the bar and a rigid
surface. A load P acts on the bar at point C, which is twothirds of the length from the fixed end.
If the support reactions produced by the load P are to
be equal in magnitude, what should be the size s of the
gap?
S
s
C
2L
—
3
A
s
L
—
3
C
B
P
PROB. 2.5-14
2.5-15 Wires B and C are attached to a support at the lefthand end and to a pin-supported rigid bar at the right-hand
end (see figure). Each wire has cross-sectional area A
0.03 in.2 and modulus of elasticity E 30 106 psi. When
PROB. 2.5-16
C
C
L
2.5-17 A copper tube is fitted around a steel bolt and the
nut is turned until it is just snug (see figure on the next
page). What stresses s and c will be produced in the steel
and copper, respectively, if the bolt is now tightened by a
quarter turn of the nut?
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
170
CHAPTER 2 Axially Loaded Members
The copper tube has length L 16 in. and crosssectional area Ac 0.6 in.2, and the steel bolt has
cross-sectional area As 0.2 in.2 The pitch of the threads
of the bolt is p 52 mils (a mil is one-thousandth of an
inch). Also, the moduli of elasticity of the steel and copper
are Es 30 106 psi and Ec 16 106 psi, respectively.
Note: The pitch of the threads is the distance advanced
by the nut in one complete turn (see Eq. 2-22).
Copper tube
2.5-20 Prestressed concrete beams are sometimes manufactured in the following manner. High-strength steel wires
are stretched by a jacking mechanism that applies a force
Q, as represented schematically in part (a) of the figure.
Concrete is then poured around the wires to form a beam,
as shown in part (b).
After the concrete sets properly, the jacks are released
and the force Q is removed [see part (c) of the figure].
Thus, the beam is left in a prestressed condition, with the
wires in tension and the concrete in compression.
Let us assume that the prestressing force Q produces in
the steel wires an initial stress s0 620 MPa. If the moduli
of elasticity of the steel and concrete are in the ratio 12:1
and the cross-sectional areas are in the ratio 1:50, what are
the final stresses ss and sc in the two materials?
;;;
@@@
ÀÀÀ
ÀÀÀ
@@@
;;;
;;
@@
ÀÀ
Steel bolt
PROB. 2.5-17
Steel wires
2.5-18 A plastic cylinder is held snugly between a rigid
Q
plate and a foundation by two steel bolts (see figure).
Determine the compressive stress sp in the plastic
when the nuts on the steel bolts are tightened by one
complete turn.
Data for the assembly are as follows: length L
200 mm, pitch of the bolt threads p 1.0 mm, modulus of
elasticity for steel Es 200 GPa, modulus of elasticity for
the plastic Ep 7.5 GPa, cross-sectional area of one bolt
As 36.0 mm2, and cross-sectional area of the plastic
cylinder Ap 960 mm2.
;;
@@
ÀÀ
Steel
bolt
L
PROBS. 2.5-18 and 2.5-19
Q
(a)
Concrete
Q
Q
(b)
(c)
PROB. 2.5-20
Stresses on Inclined Sections
2.6-1 A steel bar of rectangular cross section (1.5 in.
2.0 in.) carries a tensile load P (see figure). The allowable
stresses in tension and shear are 15,000 psi and 7,000 psi,
respectively.
Determine the maximum permissible load Pmax.
2.5-19 Solve the preceding problem if the data for the
assembly are as follows: length L 10 in., pitch of the
bolt threads p 0.058 in., modulus of elasticity for
steel Es 30 106 psi, modulus of elasticity for the plastic
Ep 500 ksi, cross-sectional area of one bolt As
0.06 in.2, and cross-sectional area of the plastic cylinder
Ap 1.5 in.2
2.0 in.
P
1.5 in.
PROB. 2.6-1
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
P
CHAPTER 2 Problems
2.6-2 A circular steel rod of diameter d is subjected to a
tensile force P 3.0 kN (see figure). The allowable
stresses in tension and shear are 120 MPa and 50 MPa,
respectively.
What is the minimum permissible diameter dmin of the
rod?
d
P
P = 3.0 kN
PROB. 2.6-2
2.6-3 A standard brick (dimensions 8 in. 4 in. 2.5 in.)
is compressed lengthwise by a force P, as shown in the
figure. If the ultimate shear stress for brick is 1200 psi and
the ultimate compressive stress is 3600 psi, what force Pmax
is required to break the brick?
Q;À@¢¢Q;À@
171
(a) If the temperature is lowered by 50°F, what is the
maximum shear stress tmax in the wire?
(b) If the allowable shear stress is 10,000 psi, what is
the maximum permissible temperature drop? (Assume that
the coefficient of thermal expansion is 10.6 106/°F and
the modulus of elasticity is 15 106 psi.)
2.6-6 A steel bar with diameter d 12 mm is subjected to
a tensile load P 9.5 kN (see figure).
(a) What is the maximum normal stress smax in the
bar?
(b) What is the maximum shear stress tmax?
(c) Draw a stress element oriented at 45° to the axis of
the bar and show all stresses acting on the faces of this
element.
P
d = 12 mm
P = 9.5 kN
P
8 in.
PROB. 2.6-3
4 in.
PROB. 2.6-6
2.6-7 During a tension test of a mild-steel specimen (see
2.5 in.
2.6-4 A brass wire of diameter d 2.42 mm is stretched
tightly between rigid supports so that the tensile force is
T 92 N (see figure).
What is the maximum permissible temperature drop
T if the allowable shear stress in the wire is 60 MPa? (The
coefficient of thermal expansion for the wire is 20
106/°C and the modulus of elasticity is 100 GPa.)
T
d
figure), the extensometer shows an elongation of 0.00120
in. with a gage length of 2 in. Assume that the steel is
stressed below the proportional limit and that the modulus
of elasticity E 30 106 psi.
(a) What is the maximum normal stress smax in the
specimen?
(b) What is the maximum shear stress tmax?
(c) Draw a stress element oriented at an angle of 45° to
the axis of the bar and show all stresses acting on the faces
of this element.
T
2 in.
T
PROB. 2.6-7
T
2.6-8 A copper bar with a rectangular cross section is held
PROBS. 2.6-4 and 2.6-5
2.6-5 A brass wire of diameter d 1/16 in. is stretched
between rigid supports with an initial tension T of 32 lb
(see figure).
without stress between rigid supports (see figure on the
next page). Subsequently, the temperature of the bar is
raised 50°C.
Determine the stresses on all faces of the elements A
and B, and show these stresses on sketches of the elements.
(Assume a 17.5 106/°C and E 120 GPa.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
172
CHAPTER 2 Axially Loaded Members
(a) What is the shear stress on plane pq? (Assume a
60 106/°F and E 450 103 psi.)
(b) Draw a stress element oriented to plane pq and
show the stresses acting on all faces of this element.
45°
A
B
PROB. 2.6-8
p
u
2.6-9 A compression member in a bridge truss is fabricated
from a wide-flange steel section (see figure). The crosssectional area A 7.5 in.2 and the axial load P 90 k.
Determine the normal and shear stresses acting on all
faces of stress elements located in the web of the beam and
oriented at (a) an angle u 0°, (b) an angle u 30°, and
(c) an angle u 45°. In each case, show the stresses on a
sketch of a properly oriented element.
P
u
P
PROB. 2.6-9
2.6-10 A plastic bar of diameter d 30 mm is compressed
in a testing device by a force P 170 N applied as shown in
the figure.
Determine the normal and shear stresses acting on all
faces of stress elements oriented at (a) an angle u 0°,
(b) an angle u 22.5°, and (c) an angle u 45°. In each
case, show the stresses on a sketch of a properly oriented
element.
P = 170 N
100 mm
300 mm
q
PROBS. 2.6-11 and 2.6-12
2.6-12 A copper bar is held snugly (but without any initial
stress) between rigid supports (see figure). The allowable
stresses on the inclined plane pq, for which u 55°, are
specified as 60 MPa in compression and 30 MPa in shear.
(a) What is the maximum permissible temperature rise
T if the allowable stresses on plane pq are not to be
exceeded? (Assume a 17 106/°C and E 120 GPa.)
(b) If the temperature increases by the maximum
permissible amount, what are the stresses on plane pq?
2.6-13 A circular brass bar of diameter d is composed of
two segments brazed together on a plane pq making an
angle a 36° with the axis of the bar (see figure). The
allowable stresses in the brass are 13,500 psi in tension and
6500 psi in shear. On the brazed joint, the allowable
stresses are 6000 psi in tension and 3000 psi in shear.
If the bar must resist a tensile force P 6000 lb, what
is the minimum required diameter dmin of the bar?
P
a
d
p
P
q
u
PROB. 2.6-13
Plastic bar
d = 30 mm
PROB. 2.6-10
2.6-11 A plastic bar fits snugly between rigid supports at
room temperature (68°F) but with no initial stress (see
figure). When the temperature of the bar is raised to 160°F,
the compressive stress on an inclined plane pq becomes
1700 psi.
2.6-14 Two boards are joined by gluing along a scarf
joint, as shown in the figure. For purposes of cutting and
gluing, the angle a between the plane of the joint and the
faces of the boards must be between 10° and 40°. Under a
tensile load P, the normal stress in the boards is 4.9 MPa.
(a) What are the normal and shear stresses acting on
the glued joint if a 20°?
(b) If the allowable shear stress on the joint is
2.25 MPa, what is the largest permissible value of the angle a?
(c) For what angle a will the shear stress on the glued
joint be numerically equal to twice the normal stress on the
joint?
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
173
CHAPTER 2 Problems
P
P
Determine the maximum normal stress smax and
maximum shear stress tmax in the bar.
p
a
r b
PROB. 2.6-14
P
2.6-15 Acting on the sides of a stress element cut from a
bar in uniaxial stress are tensile stresses of 10,000 psi and
5,000 psi, as shown in the figure.
(a) Determine the angle u and the shear stress tu and
show all stresses on a sketch of the element.
(b) Determine the maximum normal stress smax and
the maximum shear stress tmax in the material.
5,000 psi
tu tu
su = 10,000 psi
u
10,000 psi
tu tu
5,000 psi
P
q
PROB. 2.6-17
2.6-18 A tension member is to be constructed of two
pieces of plastic glued along plane pq (see figure). For
purposes of cutting and gluing, the angle u must be
between 25° and 45°. The allowable stresses on the glued
joint in tension and shear are 5.0 MPa and 3.0 MPa, respectively.
(a) Determine the angle u so that the bar will carry the
largest load P. (Assume that the strength of the glued joint
controls the design.)
(b) Determine the maximum allowable load Pmax if the
cross-sectional area of the bar is 225 mm2.
★
ÀÀ;;
@@
;;
ÀÀ
@@
p
P
PROB. 2.6-15
2.6-16 A prismatic bar is subjected to an axial force that
produces a tensile stress su 63 MPa and a shear stress tu
21 MPa on a certain inclined plane (see figure).
Determine the stresses acting on all faces of a stress
element oriented at u 30° and show the stresses on a
sketch of the element.
s
u
P
q
PROB. 2.6-18
Strain Energy
When solving the problems for Section 2.7, assume that the
material behaves linearly elastically.
2.7-1 A prismatic bar AD of length L, cross-sectional area
63 MPa
u
21 MPa
A, and modulus of elasticity E is subjected to loads 5P, 3P,
and P acting at points B, C, and D, respectively (see
figure). Segments AB, BC, and CD have lengths L/6, L/2,
and L/3, respectively.
(a) Obtain a formula for the strain energy U of the bar.
(b) Calculate the strain energy if P 6 k, L 52 in.,
A 2.76 in.2, and the material is aluminum with E
10.4 106 psi.
5P
PROB. 2.6-16
2.6-17 The normal stress on plane pq of a prismatic bar
★
in tension (see figure) is found to be 7500 psi. On plane rs,
which makes an angle b 30° with plane pq, the stress is
found to be 2500 psi.
A
3P
B
L
—
6
P
C
L
—
2
PROB. 2.7-1
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
D
L
—
3
174
CHAPTER 2 Axially Loaded Members
2.7-2 A bar of circular cross section having two different
diameters d and 2d is shown in the figure. The length of
each segment of the bar is L/2 and the modulus of elasticity
of the material is E.
(a) Obtain a formula for the strain energy U of the bar
due to the load P.
(b) Calculate the strain energy if the load P 27 kN,
the length L 600 mm, the diameter d 40 mm, and the
material is brass with E 105 GPa.
2d
Q
A
P
B
C
L/2
d
P
(a) Determine the strain energy U1 of the bar when the
force P acts alone (Q 0).
(b) Determine the strain energy U2 when the force Q
acts alone (P 0).
(c) Determine the strain energy U3 when the forces P
and Q act simultaneously upon the bar.
P
L/2
PROB. 2.7-4
2.7-5 Determine the strain energy per unit volume (units
L
—
2
of psi) and the strain energy per unit weight (units of in.)
that can be stored in each of the materials listed in the
accompanying table, assuming that the material is stressed
to the proportional limit.
L
—
2
PROB. 2.7-2
2.7-3 A three-story steel column in a building supports
roof and floor loads as shown in the figure. The story
height H is 10.5 ft, the cross-sectional area A of the
column is 15.5 in.2, and the modulus of elasticity E of the
steel is 30 106 psi.
Calculate the strain energy U of the column assuming
P1 40 k and P2 P3 60 k.
P1
DATA FOR PROBLEM 2.7-5
Material
Weight
density
(lb/in.3)
Modulus of
elasticity
(ksi)
Mild steel
Tool steel
Aluminum
Rubber (soft)
0.284
0.284
0.0984
0.0405
30,000
30,000
10,500
0.300
Proportional
limit
(psi)
36,000
75,000
60,000
300
2.7-6 The truss ABC shown in the figure is subjected to a
P2
P3
H
H
horizontal load P at joint B. The two bars are identical with
cross-sectional area A and modulus of elasticity E.
(a) Determine the strain energy U of the truss if the
angle 60°.
(b) Determine the horizontal displacement dB of joint
B by equating the strain energy of the truss to the work
done by the load.
P
B
H
b
b
A
PROB. 2.7-3
2.7-4 The bar ABC shown in the figure is loaded by a
force P acting at end C and by a force Q acting at the
midpoint B. The bar has constant axial rigidity EA.
C
L
PROB. 2.7-6
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 2 Problems
2.7-7 The truss ABC shown in the figure supports a horizontal load P1 300 lb and a vertical load P2 900 lb.
Both bars have cross-sectional area A 2.4 in.2 and are
made of steel with E 30 106 psi.
(a) Determine the strain energy U1 of the truss when
the load P1 acts alone (P2 0).
(b) Determine the strain energy U2 when the load P2
acts alone (P10).
(c) Determine the strain energy U3 when both loads act
simultaneously.
175
2.7-9 A slightly tapered bar AB of rectangular cross
section and length L is acted upon by a force P (see figure).
The width of the bar varies uniformly from b2 at end A to
b1 at end B. The thickness t is constant.
(a) Determine the strain energy U of the bar.
(b) Determine the elongation d of the bar by equating
the strain energy to the work done by the force P.
A
B
b2
b1
P
A
L
PROB. 2.7-9
30°
C
B P1 = 300 lb
P2 = 900 lb
60 in.
PROB. 2.7-7
2.7-8 The statically indeterminate structure shown in the
figure consists of a horizontal rigid bar AB supported by
five equally spaced springs. Springs 1, 2, and 3 have stiffnesses 3k, 1.5k, and k, respectively. When unstressed, the
lower ends of all five springs lie along a horizontal line.
Bar AB, which has weight W, causes the springs to elongate
by an amount d.
(a) Obtain a formula for the total strain energy U of the
springs in terms of the downward displacement d of the
bar.
(b) Obtain a formula for the displacement d by
equating the strain energy of the springs to the work done
by the weight W.
(c) Determine the forces F1, F2, and F3 in the springs.
(d) Evaluate the strain energy U, the displacement
d, and the forces in the springs if W 600 N and k
7.5 N/mm.
1
3k
k
1.5k
2
3
A
PROB. 2.7-8
1
3k
B
W
P
s
L
PROB. 2.7-10
1.5k
2
2.7-10 A compressive load P is transmitted through a
rigid plate to three magnesium-alloy bars that are identical
except that initially the middle bar is slightly shorter than
the other bars (see figure). The dimensions and properties of
the assembly are as follows: length L 1.0 m, cross-sectional
area of each bar A 3000 mm2, modulus of elasticity
E 45 GPa, and the gap s 1.0 mm.
(a) Calculate the load P1 required to close the gap.
(b) Calculate the downward displacement d of the rigid
plate when P 400 kN.
(c) Calculate the total strain energy U of the three bars
when P 400 kN.
(d) Explain why the strain energy U is not equal to
Pd/2. (Hint: Draw a load-displacement diagram.)
★
2.7-11 A block B is pushed against three springs by a
force P (see figure on the next page). The middle spring has
stiffness k1 and the outer springs each have stiffness k2.
Initially, the springs are unstressed and the middle spring is
longer than the outer springs (the difference in length is
denoted s).
★★
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
176
CHAPTER 2 Axially Loaded Members
(a) Draw a force-displacement diagram with the force
P as ordinate and the displacement x of the block as
abscissa.
(b) From the diagram, determine the strain energy U1
of the springs when x 2s.
(c) Explain why the strain energy U1 is not equal to
Pd/2, where d 2s.
s
k2
P
k1
B
k2
Impact Loading
The problems for Section 2.8 are to be solved on the basis
of the assumptions and idealizations described in the text.
In particular, assume that the material behaves linearly
elastically and no energy is lost during the impact.
2.8-1 A sliding collar of weight W 150 lb falls from a
height h 2.0 in. onto a flange at the bottom of a slender
vertical rod (see figure). The rod has length L 4.0 ft,
cross-sectional area A 0.75 in.2, and modulus of elasticity E 30 106 psi.
Calculate the following quantities: (a) the maximum
downward displacement of the flange, (b) the maximum
tensile stress in the rod, and (c) the impact factor.
x
PROB. 2.7-11
Collar
2.7-12 A bungee cord that behaves linearly elastically
has an unstressed length L0 760 mm and a stiffness
k 140 N/m. The cord is attached to two pegs, distance b
380 mm apart, and pulled at its midpoint by a force P 80 N
(see figure).
(a) How much strain energy U is stored in the cord?
(b) What is the displacement dC of the point where the
load is applied?
(c) Compare the strain energy U with the quantity
PdC/2.
(Note: The elongation of the cord is not small
compared to its original length.)
L
★★★
Rod
Flange
;;;
@@@
ÀÀÀ
@@
ÀÀ
;;
@@@
ÀÀÀ
;;;
@@
ÀÀ
;;
b
h
PROBS. 2.8-1, 2.8-2, and 2.8-3
2.8-2 Solve the preceding problem if the collar has mass
M 80 kg, the height h 0.5 m, the length L 3.0 m, the
cross-sectional area A 350 mm2, and the modulus of elasticity E 170 GPa.
2.8-3 Solve Problem 2.8-1 if the collar has weight W
A
50 lb, the height h 2.0 in., the length L 3.0 ft, the crosssectional area A 0.25 in.2, and the modulus of elasticity
E 30,000 ksi.
B
2.8-4 A block weighing W 5.0 N drops inside a cylinder
C
P
PROB. 2.7-12
from a height h 200 mm onto a spring having stiffness
k 90 N/m (see figure).
(a) Determine the maximum shortening of the spring
due to the impact, and (b) determine the impact factor.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 2 Problems
177
If the allowable stress in the wood under an impact load
is 2500 psi, what is the maximum permissible height h?
Block
Cylinder
W = 4,500 l
h
h
d = 12 in.
k
L = 15 ft
PROBS. 2.8-4 and 2.8-5
2.8-5 Solve the preceding problem if the block weighs
W 1.0 lb, h 12 in., and k 0.5 lb/in.
2.8-6 A small rubber ball (weight W 450 mN) is
attached by a rubber cord to a wood paddle (see figure).
The natural length of the cord is L0 200 mm, its crosssectional area is A 1.6 mm2, and its modulus of elasticity
is E 2.0 MPa. After being struck by the paddle, the ball
stretches the cord to a total length L1 900 mm.
What was the velocity v of the ball when it left the
paddle? (Assume linearly elastic behavior of the rubber
cord, and disregard the potential energy due to any change
in elevation of the ball.)
PROB. 2.8-7
2.8-8 A cable with a restrainer at the bottom hangs vertically
from its upper end (see figure). The cable has an effective
cross-sectional area A 40 mm2 and an effective modulus
of elasticity E 130 GPa. A slider of mass M 35 kg
drops from a height h 1.0 m onto the restrainer.
If the allowable stress in the cable under an impact
load is 500 MPa, what is the minimum permissible length L
of the cable?
Cable
Slider
L
PROB. 2.8-6
Restrainer
2.8-7 A weight W 4500 lb falls from a height h onto a
vertical wood pole having length L 15 ft, diameter d
12 in., and modulus of elasticity E 1.6 106 psi (see
figure).
PROBS. 2.8-8 and 2.8-9
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
h
178
CHAPTER 2 Axially Loaded Members
2.8-9 Solve the preceding problem if the slider has weight
W 100 lb, h 45 in., A 0.080 in.2, E 21 106 psi,
and the allowable stress is 70 ksi.
2.8-10 A bumping post at the end of a track in a railway
yard has a spring constant k 8.0 MN/m (see figure). The
maximum possible displacement d of the end of the striking
plate is 450 mm.
What is the maximum velocity vmax that a railway car
of weight W 545 kN can have without damaging the
bumping post when it strikes it?
v
k
PROB. 2.8-12
d
2.8-13 A weight W rests on top of a wall and is attached
to one end of a very flexible cord having cross-sectional
area A and modulus of elasticity E (see figure). The other
end of the cord is attached securely to the wall. The weight
is then pushed off the wall and falls freely the full length of
the cord.
(a) Derive a formula for the impact factor.
(b) Evaluate the impact factor if the weight, when
hanging statically, elongates the band by 2.5% of its original length.
★
PROB. 2.8-10
2.8-11 A bumper for a mine car is constructed with a spring
of stiffness k 1120 lb/in. (see figure). If a car weighing
3450 lb is traveling at velocity v 7 mph when it strikes the
spring, what is the maximum shortening of the spring?
v
k
À;@À
@
;
W
W
PROB. 2.8-13
PROB. 2.8-11
2.8-12 A bungee jumper having a mass of 55 kg leaps
★
from a bridge, braking her fall with a long elastic shock
cord having axial rigidity EA 2.3 kN (see figure).
If the jumpoff point is 60 m above the water, and if it
is desired to maintain a clearance of 10 m between the
jumper and the water, what length L of cord should be
used?
2.8-14 A rigid bar AB having mass M 1.0 kg and
length L 0.5 m is hinged at end A and supported at end B
by a nylon cord BC (see figure on the next page). The cord
has cross-sectional area A 30 mm2, length b 0.25 m,
and modulus of elasticity E 2.1 GPa.
If the bar is raised to its maximum height and then
released, what is the maximum stress in the cord?
★★
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 2 Problems
C
b
A
B
W
179
2.10-2 The flat bars shown in parts (a) and (b) of the
figure are subjected to tensile forces P 2.5 kN. Each bar
has thickness t 5.0 mm.
(a) For the bar with a circular hole, determine the
maximum stresses for hole diameters d 12 mm and d
20 mm if the width b 60 mm.
(b) For the stepped bar with shoulder fillets, determine
the maximum stresses for fillet radii R 6 mm and R 10 mm
if the bar widths are b 60 mm and c 40 mm.
2.10-3 A flat bar of width b and thickness t has a hole of
diameter d drilled through it (see figure). The hole may
have any diameter that will fit within the bar.
What is the maximum permissible tensile load Pmax if
the allowable tensile stress in the material is st?
L
PROB. 2.8-14
Stress Concentrations
P
The problems for Section 2.10 are to be solved by considering
the stress-concentration factors and assuming linearly
elastic behavior.
2.10-1 The flat bars shown in parts (a) and (b) of the
figure are subjected to tensile forces P 3.0 k. Each bar
has thickness t 0.25 in.
(a) For the bar with a circular hole, determine the
maximum stresses for hole diameters d 1 in. and
d 2 in. if the width b 6.0 in.
(b) For the stepped bar with shoulder fillets, determine
the maximum stresses for fillet radii R 0.25 in. and R
0.5 in. if the bar widths are b 4.0 in. and c 2.5 in.
P
P
d
b
b
P
d
PROB. 2.10-3
2.10-4 A round brass bar of diameter d1 20 mm has
upset ends of diameter d2 26 mm (see figure). The
lengths of the segments of the bar are L1 0.3 m and L2
0.1 m. Quarter-circular fillets are used at the shoulders of
the bar, and the modulus of elasticity of the brass is E
100 GPa.
If the bar lengthens by 0.12 mm under a tensile load P,
what is the maximum stress smax in the bar?
P
d2
L2
d2
d1
L1
P
L2
PROBS. 2.10-4 and 2.10-5
(a)
R
P
c
b
(b)
PROBS. 2.10-1 and 2.10-2
P
2.10-5 Solve the preceding problem for a bar of monel
metal having the following properties: d1 1.0 in., d2
1.4 in., L1 20.0 in., L2 5.0 in., and E 25 106 psi.
Also, the bar lengthens by 0.0040 in. when the tensile load
is applied.
2.10-6 A prismatic bar of diameter d0 20 mm is being
compared with a stepped bar of the same diameter (d1
20 mm) that is enlarged in the middle region to a diameter
d2 25 mm (see figure on the next page). The radius of the
fillets in the stepped bar is 2.0 mm.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
180
CHAPTER 2 Axially Loaded Members
(a) Does enlarging the bar in the middle region make it
stronger than the prismatic bar? Demonstrate your answer
by determining the maximum permissible load P1 for the
prismatic bar and the maximum permissible load P2 for
the enlarged bar, assuming that the allowable stress for the
material is 80 MPa.
(b) What should be the diameter d0 of the prismatic bar
if it is to have the same maximum permissible load as does
the stepped bar?
Derive the following formula
gL2
gL
s0aL
d
2E
(m 1)E s0
m
for the elongation of the bar.
A
P1
L
P2
B
d0
d1
P1
PROB. 2.11-1
2.11-2 A prismatic bar of length L 1.8 m and crosssectional area A 480 mm2 is loaded by forces P1 30
kN and P2 60 kN (see figure). The bar is constructed of
magnesium alloy having a stress-strain curve described by
the following Ramberg-Osgood equation:
d2
d1
P2
PROB. 2.10-6
1
s
618 170
s
45,000
10
(s MPa)
e
2.10-7 A stepped bar with a hole (see figure) has widths
b 2.4 in. and c 1.6 in. The fillets have radii equal to
0.2 in.
What is the diameter dmax of the largest hole that can
be drilled through the bar without reducing the loadcarrying capacity?
P
d
c
b
P
in which s has units of megapascals.
(a) Calculate the displacement dC of the end of the bar
when the load P1 acts alone.
(b) Calculate the displacement when the load P2 acts
alone.
(c) Calculate the displacement when both loads act
simultaneously.
A
B
2L
—
3
PROB. 2.10-7
P1 C
P2
L
—
3
Nonlinear Behavior (Changes in Lengths of Bars)
PROB. 2.11-2
2.11-1 A bar AB of length L and weight density g hangs
2.11-3 A circular bar of length L 32 in. and diameter
d 0.75 in. is subjected to tension by forces P (see figure
on the next page). The wire is made of a copper alloy
having the following hyperbolic stress-strain relationship:
vertically under its own weight (see figure). The stressstrain relation for the material is given by the
Ramberg-Osgood equation (Eq. 2-71):
m
s
sa s
e 0
E
E s0
18,000e
1 3 0 0e
s
0 e 0.03
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(s ksi)
181
CHAPTER 2 Problems
(a) Draw a stress-strain diagram for the material.
(b) If the elongation of the wire is limited to 0.25 in.
and the maximum stress is limited to 40 ksi, what is the
allowable load P?
d
P
s
E2 = 2.4 × 106 psi
12,000
psi
P
E1 = 10 × 106 psi
L
PROB. 2.11-3
0
e
PROB. 2.11-5
2.11-4 A prismatic bar in tension has length L 2.0 m and
cross-sectional area A 249 mm2. The material of the bar
has the stress-strain curve shown in the figure.
Determine the elongation d of the bar for each of the
following axial loads: P 10 kN, 20 kN, 30 kN, 40 kN,
and 45 kN. From these results, plot a diagram of load P
versus elongation (load-displacement diagram).
200
2.11-6 A rigid bar AB, pinned at end A, is supported by a
wire CD and loaded by a force P at end B (see figure). The
wire is made of high-strength steel having modulus of
elasticity E 210 GPa and yield stress sY 820 MPa. The
length of the wire is L 1.0 m and its diameter is d
3 mm. The stress-strain diagram for the steel is defined by
the modified power law, as follows:
★
s Ee
s (MPa)
Ese
s sY
Y
100
0
0 s sY
n
0
0.005
e
s sY
(a) Assuming n 0.2, calculate the displacement dB at
the end of the bar due to the load P. Take values of P from
2.4 kN to 5.6 kN in increments of 0.8 kN.
(b) Plot a load-displacement diagram showing P
versus dB.
0.010
PROB. 2.11-4
C
L
2.11-5 An aluminum bar subjected to tensile forces P has
2
length L 150 in. and cross-sectional area A 2.0 in.
The stress-strain behavior of the aluminum may be represented approximately by the bilinear stress-strain diagram
shown in the figure.
Calculate the elongation d of the bar for each of the
following axial loads: P 8 k, 16 k, 24 k, 32 k, and 40 k.
From these results, plot a diagram of load P versus elongation (load-displacement diagram).
A
D
B
P
2b
PROB. 2.11-6
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
b
182
CHAPTER 2 Axially Loaded Members
Elastoplastic Analysis
The problems for Section 2.12 are to be solved assuming
that the material is elastoplastic with yield stress sY, yield
strain eY, and modulus of elasticity E in the linearly elastic
region (see Fig. 2-70).
R
2.12-1 Two identical bars AB and BC support a vertical
load P (see figure). The bars are made of steel having a
stress-strain curve that may be idealized as elastoplastic
with yield stress sY. Each bar has cross-sectional area A.
Determine the yield load PY and the plastic load PP.
A
B
P
PROB. 2.12-3
u
A
u
C
2.12-4 A load P acts on a horizontal beam that is supported
by four rods arranged in the symmetrical pattern shown in
the figure. Each rod has cross-sectional area A and the
material is elastoplastic with yield stress sY. Determine
the plastic load PP.
B
P
PROB. 2.12-1
2.12-2 A stepped bar ACB with circular cross sections is
held between rigid supports and loaded by an axial force P
at midlength (see figure). The diameters for the two parts of
the bar are d1 20 mm and d2 25 mm, and the material
is elastoplastic with yield stress Y 250 MPa.
Determine the plastic load PP.
A
d1
C
L
—
2
d2
P
B
a
L
—
2
PROB. 2.12-2
a
P
PROB. 2.12-4
2.12-3 A horizontal rigid bar AB supporting a load P is
hung from five symmetrically placed wires, each of crosssectional area A (see figure). The wires are fastened to a
curved surface of radius R.
(a) Determine the plastic load PP if the material of the
wires is elastoplastic with yield stress sY.
(b) How is PP changed if bar AB is flexible instead of
rigid?
(c) How is PP changed if the radius R is increased?
2.12-5 The symmetric truss ABCDE shown in the figure is
constructed of four bars and supports a load P at joint E. Each
of the two outer bars has a cross-sectional area of 0.307 in.2,
and each of the two inner bars has an area of 0.601 in.2 The
material is elastoplastic with yield stress sY 36 ksi.
Determine the plastic load PP.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
183
CHAPTER 2 Problems
21 in.
A
54 in.
C
B
2.12-8 A rigid bar ACB is supported on a fulcrum at C
and loaded by a force P at end B (see figure). Three identical wires made of an elastoplastic material (yield stress sY
and modulus of elasticity E) resist the load P. Each wire
has cross-sectional area A and length L.
(a) Determine the yield load PY and the corresponding
yield displacement dY at point B.
(b) Determine the plastic load PP and the corresponding displacement dP at point B when the load just reaches
the value PP.
(c) Draw a load-displacement diagram with the load P
as ordinate and the displacement dB of point B as abscissa.
★
21 in.
D
36 in.
E
P
PROB. 2.12-5
2.12-6 Five bars, each having a diameter of 10 mm,
support a load P as shown in the figure. Determine the
plastic load PP if the material is elastoplastic with yield
stress sY 250 MPa.
b
b
b
b
L
A
C
B
P
L
a
a
a
a
PROB. 2.12-8
2b
P
PROB. 2.12-6
2.12-7 A circular steel rod AB of diameter d 0.60 in. is
stretched tightly between two supports so that initially the
tensile stress in the rod is 10 ksi (see figure). An axial force
P is then applied to the rod at an intermediate location C.
(a) Determine the plastic load PP if the material is
elastoplastic with yield stress sY 36 ksi.
(b) How is PP changed if the initial tensile stress is
doubled to 20 ksi?
2.12-9 The structure shown in the figure consists of a
horizontal rigid bar ABCD supported by two steel wires,
one of length L and the other of length 3L/4. Both wires
have cross-sectional area A and are made of elastoplastic
material with yield stress sY and modulus of elasticity E. A
vertical load P acts at end D of the bar.
(a) Determine the yield load PY and the corresponding
yield displacement dY at point D.
(b) Determine the plastic load PP and the corresponding displacement dP at point D when the load just reaches
the value PP.
(c) Draw a load-displacement diagram with the load P
as ordinate and the displacement dD of point D as abscissa.
★
B
A
L
d
A
A
P
B
C
D
B
P
C
PROB. 2.12-7
3L
4
2b
PROB. 2.12-9
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
b
b
184
★★
CHAPTER 2 Axially Loaded Members
2.12-10 Two cables, each having a length L of approxi-
mately 40 m, support a loaded container of weight W (see
figure). The cables, which have effective cross-sectional
area A 48.0 mm2 and effective modulus of elasticity E
160 GPa, are identical except that one cable is longer than
the other when they are hanging separately and unloaded.
The difference in lengths is d 100 mm. The cables are
made of steel having an elastoplastic stress-strain diagram
with Y 500 MPa. Assume that the weight W is initially
zero and is slowly increased by the addition of material to
the container.
(a) Determine the weight WY that first produces
yielding of the shorter cable. Also, determine the corresponding elongation Y of the shorter cable.
(b) Determine the weight WP that produces yielding of
both cables. Also, determine the elongation P of the
shorter cable when the weight W just reaches the value WP.
(c) Construct a load-displacement diagram showing
the weight W as ordinate and the elongation of the shorter
cable as abscissa. (Hint: The load displacement diagram is
not a single straight line in the region 0 W WY.)
2.12-11 A hollow circular tube T of length L 15 in. is
uniformly compressed by a force P acting through a rigid
plate (see figure). The outside and inside diameters of the
tube are 3.0 and 2.75 in., repectively. A concentric solid circular bar B of 1.5 in. diameter is mounted inside the tube.
When no load is present, there is a clearance c 0.010 in.
between the bar B and the rigid plate. Both bar and tube are
made of steel having an elastoplastic stress-strain diagram
with E 29 103 ksi and Y 36 ksi.
(a) Determine the yield load PY and the corresponding
shortening Y of the tube.
(b) Determine the plastic load PP and the corresponding
shortening P of the tube.
(c) Construct a load-displacement diagram showing the
load P as ordinate and the shortening of the tube as
abscissa. (Hint: The load-displacement diagram is not a single
straight line in the region 0 P PY.)
★★
P
c
T
L
T
B
T
W
PROB. 2.12-10
PROB. 2.12-11
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
L
B
3
Torsion
3.1 INTRODUCTION
(a)
T
FIG. 3-1 Torsion of a screwdriver due to a
torque T applied to the handle
In Chapters 1 and 2 we discussed the behavior of the simplest type of
structural member—namely, a straight bar subjected to axial loads. Now
we consider a slightly more complex type of behavior known as torsion.
Torsion refers to the twisting of a straight bar when it is loaded by
moments (or torques) that tend to produce rotation about the longitudinal
axis of the bar. For instance, when you turn a screwdriver (Fig. 3-1a),
your hand applies a torque T to the handle (Fig. 3-1b) and twists the
shank of the screwdriver. Other examples of bars in torsion are drive
shafts in automobiles, axles, propeller shafts, steering rods, and drill bits.
An idealized case of torsional loading is pictured in Fig. 3-2a, on the
next page, which shows a straight bar supported at one end and loaded by
two pairs of equal and opposite forces. The first pair consists of the
forces P1 acting near the midpoint of the bar and the second pair consists
of the forces P2 acting at the end. Each pair of forces forms a couple that
tends to twist the bar about its longitudinal axis. As we know from statics, the moment of a couple is equal to the product of one of the forces
and the perpendicular distance between the lines of action of the forces;
thus, the first couple has a moment T1 ⫽ P1d1 and the second has a
moment T2 ⫽ P2d2.
Typical USCS units for moment are the pound-foot (lb-ft) and the
pound-inch (lb-in.). The SI unit for moment is the newton meter (N⭈m).
The moment of a couple may be represented by a vector in the form
of a double-headed arrow (Fig. 3-2b). The arrow is perpendicular to the
plane containing the couple, and therefore in this case both arrows are
parallel to the axis of the bar. The direction (or sense) of the moment is
indicated by the right-hand rule for moment vectors—namely, using
your right hand, let your fingers curl in the direction of the moment, and
then your thumb will point in the direction of the vector.
185
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
186
CHAPTER 3 Torsion
P2
P1
An alternative representation of a moment is a curved arrow acting
in the direction of rotation (Fig. 3-2c). Both the curved arrow and vector
representations are in common use, and both are used in this book. The
choice depends upon convenience and personal preference.
Moments that produce twisting of a bar, such as the moments T1 and
T2 in Fig. 3-2, are called torques or twisting moments. Cylindrical
members that are subjected to torques and transmit power through rotation are called shafts; for instance, the drive shaft of an automobile or
the propeller shaft of a ship. Most shafts have circular cross sections,
either solid or tubular.
In this chapter we begin by developing formulas for the deformations and stresses in circular bars subjected to torsion. We then analyze
the state of stress known as pure shear and obtain the relationship
between the moduli of elasticity E and G in tension and shear, respectively. Next, we analyze rotating shafts and determine the power they
transmit. Finally, we cover several additional topics related to torsion,
namely, statically indeterminate members, strain energy, thin-walled
tubes of noncircular cross section, and stress concentrations.
Axis
of bar
d1
P2
d2
P1
T1 = P1d1
T2 = P2 d 2
(a)
T1
T2
(b)
T1
T2
(c)
FIG. 3-2 Circular bar subjected to torsion
by torques T1 and T2
3.2 TORSIONAL DEFORMATIONS OF A CIRCULAR BAR
We begin our discussion of torsion by considering a prismatic bar of
circular cross section twisted by torques T acting at the ends (Fig. 3-3a).
Since every cross section of the bar is identical, and since every cross
section is subjected to the same internal torque T, we say that the bar is
in pure torsion. From considerations of symmetry, it can be proved that
cross sections of the bar do not change in shape as they rotate about the
longitudinal axis. In other words, all cross sections remain plane and
circular and all radii remain straight. Furthermore, if the angle of
rotation between one end of the bar and the other is small, neither the
length of the bar nor its radius will change.
f (x)
T
f
p
x
FIG. 3-3 Deformations of a circular bar in
pure torsion
q
f
q
q' r
T
q'
r
(b)
L
(a)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
187
SECTION 3.2 Torsional Deformations of a Circular Bar
To aid in visualizing the deformation of the bar, imagine that the
left-hand end of the bar (Fig. 3-3a) is fixed in position. Then, under the
action of the torque T, the right-hand end will rotate (with respect to the
left-hand end) through a small angle f, known as the angle of twist (or
angle of rotation). Because of this rotation, a straight longitudinal line pq
on the surface of the bar will become a helical curve pq⬘, where q⬘ is the
position of point q after the end cross section has rotated through the angle
f (Fig. 3-3b).
The angle of twist changes along the axis of the bar, and at intermediate
cross sections it will have a value f (x) that is between zero at the
left-hand end and f at the right-hand end. If every cross section of the
bar has the same radius and is subjected to the same torque (pure
torsion), the angle f(x) will vary linearly between the ends.
Shear Strains at the Outer Surface
Now consider an element of the bar between two cross sections distance
dx apart (Fig. 3-4a). This element is shown enlarged in Fig. 3-4b. On its
outer surface we identify a small element abcd, with sides ab and cd that
initially are parallel to the longitudinal axis. During twisting of the bar,
the right-hand cross section rotates with respect to the left-hand cross
section through a small angle of twist df, so that points b and c move to
b⬘ and c⬘, respectively. The lengths of the sides of the element, which is
now element ab⬘c⬘d, do not change during this small rotation.
However, the angles at the corners of the element (Fig. 3-4b) are no
longer equal to 90°. The element is therefore in a state of pure shear,
which means that the element is subjected to shear strains but no normal
strains (see Fig. 1-28 of Section 1.6). The magnitude of the shear strain
T
T
x
dx
L
(a)
gmax
g
b
a
T
b'
c
d
c'
FIG. 3-4 Deformation of an element of
length dx cut from a bar in torsion
df
df
r
T
r
dx
dx
(b)
(c)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
188
CHAPTER 3 Torsion
at the outer surface of the bar, denoted gmax, is equal to the decrease in
the angle at point a, that is, the decrease in angle bad. From Fig. 3-4b
we see that the decrease in this angle is
bb⬘
gmax ⫽ ᎏᎏ
ab
(a)
where gmax is measured in radians, bb⬘ is the distance through which
point b moves, and ab is the length of the element (equal to dx). With r
denoting the radius of the bar, we can express the distance bb⬘ as rdf,
where df also is measured in radians. Thus, the preceding equation
becomes
r df
gmax ⫽ ᎏᎏ
dx
(b)
This equation relates the shear strain at the outer surface of the bar to the
angle of twist.
The quantity df/dx is the rate of change of the angle of twist f with
respect to the distance x measured along the axis of the bar. We will
denote df /dx by the symbol u and refer to it as the rate of twist, or the
angle of twist per unit length:
df
u ⫽ ᎏᎏ
dx
(3-1)
With this notation, we can now write the equation for the shear strain at
the outer surface (Eq. b) as follows:
rd
gmax ⫽ ᎏᎏ ⫽ ru
dx
(3-2)
For convenience, we discussed a bar in pure torsion when deriving
Eqs. (3-1) and (3-2). However, both equations are valid in more general
cases of torsion, such as when the rate of twist u is not constant but
varies with the distance x along the axis of the bar.
In the special case of pure torsion, the rate of twist is equal to the
total angle of twist f divided by the length L, that is, u ⫽ f /L. Therefore, for pure torsion only, we obtain
r
gmax ⫽ ru ⫽ ᎏᎏ
L
(3-3)
This equation can be obtained directly from the geometry of Fig. 3-3a
by noting that gmax is the angle between lines pq and pq⬘, that is, gmax is
the angle qpq⬘. Therefore, gmaxL is equal to the distance qq⬘ at the end
of the bar. But since the distance qq⬘ also equals rf (Fig. 3-3b), we
obtain rf ⫽ gmax L, which agrees with Eq. (3-3).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 3.3 Circular Bars of Linearly Elastic Materials
189
Shear Strains Within the Bar
The shear strains within the interior of the bar can be found by the same
method used to find the shear strain gmax at the surface. Because radii in
the cross sections of a bar remain straight and undistorted during
twisting, we see that the preceding discussion for an element abcd at the
outer surface (Fig. 3-4b) will also hold for a similar element situated on
the surface of an interior cylinder of radius r (Fig. 3-4c). Thus, interior
elements are also in pure shear with the corresponding shear strains
given by the equation (compare with Eq. 3-2):
r
g ⫽ ru ⫽ ᎏᎏ gmax
r
(3-4)
This equation shows that the shear strains in a circular bar vary linearly
with the radial distance r from the center, with the strain being zero at
the center and reaching a maximum value gmax at the outer surface.
Circular Tubes
g max
A review of the preceding discussions will show that the equations for
the shear strains (Eqs. 3-2 to 3-4) apply to circular tubes (Fig. 3-5) as
well as to solid circular bars. Figure 3-5 shows the linear variation in
shear strain between the maximum strain at the outer surface and the
minimum strain at the interior surface. The equations for these strains are
as follows:
g min
r1
rf
gmax ⫽ ᎏ2ᎏ
L
r2
FIG. 3-5 Shear strains in a circular tube
r
rf
gmin ⫽ ᎏᎏ1 gmax ⫽ ᎏ1ᎏ
r2
L
(3-5a,b)
in which r1 and r2 are the inner and outer radii, respectively, of the tube.
All of the preceding equations for the strains in a circular bar are
based upon geometric concepts and do not involve the material
properties. Therefore, the equations are valid for any material, whether it
behaves elastically or inelastically, linearly or nonlinearly. However, the
equations are limited to bars having small angles of twist and small
strains.
3.3 CIRCULAR BARS OF LINEARLY ELASTIC MATERIALS
Now that we have investigated the shear strains in a circular bar in
torsion (see Figs. 3-3 to 3-5), we are ready to determine the directions
and magnitudes of the corresponding shear stresses. The directions of
the stresses can be determined by inspection, as illustrated in Fig. 3-6a
on the next page. We observe that the torque T tends to rotate the righthand end of the bar counterclockwise when viewed from the right.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
190
CHAPTER 3 Torsion
T
T
t
(a)
t
b
b'
a
t max
r
t
g
t
t
d
FIG. 3-6 Shear stresses in a circular bar in
torsion
t
(b)
r
c
c'
(c)
Therefore the shear stresses t acting on a stress element located on the
surface of the bar will have the directions shown in the figure.
For clarity, the stress element shown in Fig. 3-6a is enlarged in
Fig. 3-6b, where both the shear strain and the shear stresses are shown.
As explained previously in Section 2.6, we customarily draw stress
elements in two dimensions, as in Fig. 3-6b, but we must always
remember that stress elements are actually three-dimensional objects
with a thickness perpendicular to the plane of the figure.
The magnitudes of the shear stresses can be determined from the
strains by using the stress-strain relation for the material of the bar. If the
material is linearly elastic, we can use Hooke’s law in shear (Eq. 1-14):
t ⫽ Gg
(3-6)
in which G is the shear modulus of elasticity and g is the shear strain in
radians. Combining this equation with the equations for the shear strains
(Eqs. 3-2 and 3-4), we get
tmax ⫽ Gru
r
t ⫽ Gru ⫽ ᎏᎏ tmax
r
(3-7a,b)
in which tmax is the shear stress at the outer surface of the bar (radius r),
t is the shear stress at an interior point (radius r), and u is the rate of
twist. (In these equations, u has units of radians per unit of length.)
Equations (3-7a) and (3-7b) show that the shear stresses vary
linearly with the distance from the center of the bar, as illustrated by the
triangular stress diagram in Fig. 3-6c. This linear variation of stress is a
consequence of Hooke’s law. If the stress-strain relation is nonlinear, the
stresses will vary nonlinearly and other methods of analysis will be
needed.
The shear stresses acting on a cross-sectional plane are accompanied by shear stresses of the same magnitude acting on longitudinal
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 3.3 Circular Bars of Linearly Elastic Materials
t max
t max
FIG. 3-7 Longitudinal and transverse
shear stresses in a circular bar subjected
to torsion
191
planes (Fig. 3-7). This conclusion follows from the fact that equal shear
stresses always exist on mutually perpendicular planes, as explained in
Section 1.6. If the material of the bar is weaker in shear on longitudinal
planes than on cross-sectional planes, as is typical of wood when the
grain runs parallel to the axis of the bar, the first cracks due to torsion
will appear on the surface in the longitudinal direction.
The state of pure shear at the surface of a bar (Fig. 3-6b) is equivalent
to equal tensile and compressive stresses acting on an element oriented at
an angle of 45°, as explained later in Section 3.5. Therefore, a rectangular
element with sides at 45° to the axis of the shaft will be subjected to tensile
and compressive stresses, as shown in Fig. 3-8. If a torsion bar is made of a
material that is weaker in tension than in shear, failure will occur in tension
along a helix inclined at 45° to the axis, as you can demonstrate by twisting
a piece of classroom chalk.
The Torsion Formula
T
T
FIG. 3-8 Tensile and compressive stresses
acting on a stress element oriented at
45° to the longitudinal axis
dA
t
r
The next step in our analysis is to determine the relationship between the
shear stresses and the torque T. Once this is accomplished, we will be
able to calculate the stresses and strains in a bar due to any set of applied
torques.
The distribution of the shear stresses acting on a cross section is pictured in Figs. 3-6c and 3-7. Because these stresses act continuously
around the cross section, they have a resultant in the form of a moment—
a moment equal to the torque T acting on the bar. To determine this
resultant, we consider an element of area dA located at radial distance r
from the axis of the bar (Fig. 3-9). The shear force acting on this element
is equal to t dA, where t is the shear stress at radius r. The moment of
this force about the axis of the bar is equal to the force times its distance
from the center, or tr dA. Substituting for the shear stress t from Eq.
(3-7b), we can express this elemental moment as
t ax 2
r dA
dM ⫽ tr dA ⫽ ᎏmᎏ
r
r
FIG. 3-9 Determination of the resultant
of the shear stresses acting on a cross
section
The resultant moment (equal to the torque T ) is the summation over the
entire cross-sectional area of all such elemental moments:
T⫽
冮 dM ⫽ ᎏt rᎏ 冮 r
max
A
2
A
t ax
dA ⫽ ᎏmᎏ
IP
r
(3-8)
in which
IP ⫽
冮r
2
dA
A
is the polar moment of inertia of the circular cross section.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(3-9)
192
CHAPTER 3 Torsion
For a circle of radius r and diameter d, the polar moment of inertia is
pr 4
pd 4
IP ⫽ ᎏᎏ ⫽ ᎏᎏ
2
32
(3-10)
as given in Appendix D, Case 9. Note that moments of inertia have units
of length to the fourth power.*
An expression for the maximum shear stress can be obtained by
rearranging Eq. (3-8), as follows:
Tr
tmax ⫽ ᎏᎏ
IP
(3-11)
This equation, known as the torsion formula, shows that the maximum
shear stress is proportional to the applied torque T and inversely proportional to the polar moment of inertia IP.
Typical units used with the torsion formula are as follows. In SI, the
torque T is usually expressed in newton meters (N⭈m), the radius r in
meters (m), the polar moment of inertia IP in meters to the fourth power
(m4), and the shear stress t in pascals (Pa). If USCS units are used, T is
often expressed in pound-feet (lb-ft) or pound-inches (lb-in.), r in inches
(in.), IP in inches to the fourth power (in.4), and t in pounds per square
inch (psi).
Substituting r ⫽ d/2 and IP ⫽ p d 4/32 into the torsion formula, we
get the following equation for the maximum stress:
16T
tmax ⫽ ᎏᎏ3
pd
(3-12)
This equation applies only to bars of solid circular cross section,
whereas the torsion formula itself (Eq. 3-11) applies to both solid bars
and circular tubes, as explained later. Equation (3-12) shows that the
shear stress is inversely proportional to the cube of the diameter. Thus, if
the diameter is doubled, the stress is reduced by a factor of eight.
The shear stress at distance r from the center of the bar is
r
T
t ⫽ ᎏᎏ tmax ⫽ ᎏᎏ
r
IP
(3-13)
which is obtained by combining Eq. (3-7b) with the torsion formula
(Eq. 3-11). Equation (3-13) is a generalized torsion formula, and we see
once again that the shear stresses vary linearly with the radial distance
from the center of the bar.
*Polar moments of inertia are discussed in Section 12.6 of Chapter 12.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 3.3 Circular Bars of Linearly Elastic Materials
193
Angle of Twist
The angle of twist of a bar of linearly elastic material can now be related
to the applied torque T. Combining Eq. (3-7a) with the torsion formula,
we get
T
u ⫽ ᎏᎏ
GIP
(3-14)
in which u has units of radians per unit of length. This equation shows
that the rate of twist u is directly proportional to the torque T and
inversely proportional to the product GIP, known as the torsional
rigidity of the bar.
For a bar in pure torsion, the total angle of twist f, equal to the rate
of twist times the length of the bar (that is, f ⫽ uL), is
TL
f ⫽ ᎏᎏ
GIP
(3-15)
in which f is measured in radians. The use of the preceding equations in
both analysis and design is illustrated later in Examples 3-1 and 3-2.
The quantity GIP /L, called the torsional stiffness of the bar, is the
torque required to produce a unit angle of rotation. The torsional
flexibility is the reciprocal of the stiffness, or L/GIP, and is defined as
the angle of rotation produced by a unit torque. Thus, we have the
following expressions:
GIP
kT ⫽ ᎏ ᎏ
L
L
fT ⫽ ᎏ ᎏ
GIP
(a,b)
These quantities are analogous to the axial stiffness k ⫽ EA/L and axial
flexibility f ⫽ L/EA of a bar in tension or compression (compare with
Eqs. 2-4a and 2-4b). Stiffnesses and flexibilities have important roles in
structural analysis.
The equation for the angle of twist (Eq. 3-15) provides a convenient
way to determine the shear modulus of elasticity G for a material. By
conducting a torsion test on a circular bar, we can measure the angle of
twist f produced by a known torque T. Then the value of G can be
calculated from Eq. (3-15).
Circular Tubes
Circular tubes are more efficient than solid bars in resisting torsional
loads. As we know, the shear stresses in a solid circular bar are
maximum at the outer boundary of the cross section and zero at the
center. Therefore, most of the material in a solid shaft is stressed significantly below the maximum shear stress. Furthermore, the stresses near
the center of the cross section have a smaller moment arm r for use in
determining the torque (see Fig. 3-9 and Eq. 3-8).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
194
CHAPTER 3 Torsion
t
r2
tmax
r1
t
FIG. 3-10 Circular tube in torsion
By contrast, in a typical hollow tube most of the material is near the
outer boundary of the cross section where both the shear stresses and the
moment arms are highest (Fig. 3-10). Thus, if weight reduction and
savings of material are important, it is advisable to use a circular tube.
For instance, large drive shafts, propeller shafts, and generator shafts
usually have hollow circular cross sections.
The analysis of the torsion of a circular tube is almost identical to
that for a solid bar. The same basic expressions for the shear stresses
may be used (for instance, Eqs. 3-7a and 3-7b). Of course, the radial
distance r is limited to the range r1 to r2, where r1 is the inner radius and
r2 is the outer radius of the bar (Fig. 3-10).
The relationship between the torque T and the maximum stress is
given by Eq. (3-8), but the limits on the integral for the polar moment of
inertia (Eq. 3-9) are r ⫽ r1 and r ⫽ r2. Therefore, the polar moment of
inertia of the cross-sectional area of a tube is
p
p
IP ⫽ ᎏᎏ (r 42 ⫺ r 41) ⫽ ᎏᎏ (d 42 ⫺ d 41)
2
32
(3-16)
The preceding expressions can also be written in the following forms:
prt
pdt
IP ⫽ ᎏᎏ (4r 2 ⫹ t 2) ⫽ ᎏᎏ(d 2 ⫹ t 2)
2
4
(3-17)
in which r is the average radius of the tube, equal to (r1 ⫹ r2)/2; d is
the average diameter, equal to (d1 ⫹ d2)/2; and t is the wall thickness
(Fig. 3-10), equal to r2 ⫺ r1. Of course, Eqs. (3-16) and (3-17) give the
same results, but sometimes the latter is more convenient.
If the tube is relatively thin so that the wall thickness t is small
compared to the average radius r, we may disregard the terms t2 in Eq.
(3-17). With this simplification, we obtain the following approximate
formulas for the polar moment of inertia:
p d 3t
IP ⬇ 2p r 3t ⫽ ᎏᎏ
4
(3-18)
These expressions are given in Case 22 of Appendix D.
Reminders: In Eqs. 3-17 and 3-18, the quantities r and d are the
average radius and diameter, not the maximums. Also, Eqs. 3-16 and
3-17 are exact; Eq. 3-18 is approximate.
The torsion formula (Eq. 3-11) may be used for a circular tube of
linearly elastic material provided IP is evaluated according to Eq. (3-16),
Eq. (3-17), or, if appropriate, Eq. (3-18). The same comment applies to
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 3.3 Circular Bars of Linearly Elastic Materials
195
the general equation for shear stress (Eq. 3-13), the equations for rate of
twist and angle of twist (Eqs. 3-14 and 3-15), and the equations for
stiffness and flexibility (Eqs. a and b).
The shear stress distribution in a tube is pictured in Fig. 3-10. From
the figure, we see that the average stress in a thin tube is nearly as great
as the maximum stress. This means that a hollow bar is more efficient in
the use of material than is a solid bar, as explained previously and as
demonstrated later in Examples 3-2 and 3-3.
When designing a circular tube to transmit a torque, we must be
sure that the thickness t is large enough to prevent wrinkling or buckling
of the wall of the tube. For instance, a maximum value of the radius to
thickness ratio, such as (r2 /t)max ⫽ 12, may be specified. Other design
considerations include environmental and durability factors, which also
may impose requirements for minimum wall thickness. These topics are
discussed in courses and textbooks on mechanical design.
Limitations
The equations derived in this section are limited to bars of circular cross
section (either solid or hollow) that behave in a linearly elastic manner.
In other words, the loads must be such that the stresses do not exceed
the proportional limit of the material. Furthermore, the equations for
stresses are valid only in parts of the bars away from stress concentrations (such as holes and other abrupt changes in shape) and away from
cross sections where loads are applied. (Stress concentrations in torsion
are discussed later in Section 3.11.)
Finally, it is important to emphasize that the equations for the
torsion of circular bars and tubes cannot be used for bars of other
shapes. Noncircular bars, such as rectangular bars and bars having
I-shaped cross sections, behave quite differently than do circular bars.
For instance, their cross sections do not remain plane and their
maximum stresses are not located at the farthest distances from the
midpoints of the cross sections. Thus, these bars require more advanced
methods of analysis, such as those presented in books on theory of elasticity and advanced mechanics of materials.*
*The torsion theory for circular bars originated with the work of the famous French
scientist C. A. de Coulomb (1736–1806); further developments were due to Thomas
Young and A. Duleau (Ref. 3-1). The general theory of torsion (for bars of any shape) is
due to the most famous elastician of all time, Barré de Saint-Venant (1797–1886); see
Ref. 2-10.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
196
CHAPTER 3 Torsion
Example 3-1
A solid steel bar of circular cross section (Fig. 3-11) has diameter d ⫽ 1.5 in.,
length L ⫽ 54 in., and shear modulus of elasticity G ⫽ 11.5 ⫻ 106 psi. The bar
is subjected to torques T acting at the ends.
(a) If the torques have magnitude T ⫽ 250 lb-ft, what is the maximum
shear stress in the bar? What is the angle of twist between the ends?
(b) If the allowable shear stress is 6000 psi and the allowable angle of twist
is 2.5°, what is the maximum permissible torque?
d = 1.5 in.
T
T
FIG. 3-11 Example 3-1. Bar in pure
L = 54 in.
torsion
Solution
(a) Maximum shear stress and angle of twist. Because the bar has a solid
circular cross section, we can find the maximum shear stress from Eq. (3-12), as
follows:
16(250 lb-ft)(12 in./ft)
16T
⫽ ᎏᎏᎏ ⫽ 4530 psi
tmax ⫽ ᎏᎏ
pd 3
In a similar manner, the angle of twist is obtained from Eq. (3-15) with the polar
moment of inertia given by Eq. (3-10):
p (1.5 in.)4
pd 4
IP ⫽ ᎏᎏ ⫽ ᎏᎏ ⫽ 0.4970 in.4
32
32
TL
(250 lb-ft)(12 in./ft)(54 in.)
f ⫽ ᎏ ᎏ ⫽ ᎏᎏᎏ
⫽ 0.02834 rad ⫽ 1.62°
GIP
(11.5 ⫻ 106 psi)(0.4970 in.4)
Thus, the analysis of the bar under the action of the given torque is completed.
(b) Maximum permissible torque. The maximum permissible torque is
determined either by the allowable shear stress or by the allowable angle of
twist. Beginning with the shear stress, we rearrange Eq. (3-12) and calculate as
follows:
p
p d 3tallow
⫽ ᎏᎏ (1.5 in.)3(6000 psi) ⫽ 3980 lb-in. ⫽ 331 lb-ft
T1 ⫽ ᎏᎏ
16
16
Any torque larger than this value will result in a shear stress that exceeds the
allowable stress of 6000 psi.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
197
SECTION 3.3 Circular Bars of Linearly Elastic Materials
Using a rearranged Eq. (3-15), we now calculate the torque based upon the
angle of twist:
(11.5 ⫻ 106 psi)(0.4970 in.4)(2.5°)( p rad/180°)
GIPfallow
ᎏ ⫽ ᎏᎏᎏᎏᎏ
T2 ⫽ ᎏ
54 in.
L
⫽ 4618 lb-in. ⫽ 385 lb-ft
Any torque larger than T2 will result in the allowable angle of twist being
exceeded.
The maximum permissible torque is the smaller of T1 and T2:
Tmax ⫽ 331 lb-ft
In this example, the allowable shear stress provides the limiting condition.
Example 3-2
A steel shaft is to be manufactured either as a solid circular bar or as a circular
tube (Fig. 3-12). The shaft is required to transmit a torque of 1200 N⭈m without
exceeding an allowable shear stress of 40 MPa nor an allowable rate of twist of
0.75°/m. (The shear modulus of elasticity of the steel is 78 GP a.)
(a) Determine the required diameter d0 of the solid shaft.
(b) Determine the required outer diameter d2 of the hollow shaft if the
thickness t of the shaft is specified as one-tenth of the outer diameter.
(c) Determine the ratio of diameters (that is, the ratio d2/d0) and the ratio of
weights of the hollow and solid shafts.
t=
d0
d1
d2
FIG. 3-12 Example 3-2. Torsion of a steel
shaft
d2
10
(a)
(b)
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
198
CHAPTER 3 Torsion
t=
d2
10
Solution
(a) Solid shaft. The required diameter d0 is determined either from the
allowable shear stress or from the allowable rate of twist. In the case of the
allowable shear stress we rearrange Eq. (3-12) and obtain
16(1200 N⭈m)
16T
d 30 ⫽ ᎏᎏ ⫽ ᎏᎏ ⫽ 152.8 ⫻ 10⫺6 m3
p allow
p (40 MPa)
from which we get
d0
(a)
d1
d0 ⫽ 0.0535 m ⫽ 53.5 mm
d2
In the case of the allowable rate of twist, we start by finding the required polar
moment of inertia (see Eq. 3-14):
(b)
1200 N⭈m
T
IP ⫽ ᎏ ᎏ ⫽ ᎏᎏᎏᎏ ⫽ 1175 ⫻ 10⫺9 m4
(78
GPa)(0.75°/m)(
p rad/180°)
Guallow
FIG. 3-12 (Repeated)
Since the polar moment of inertia is equal to pd 4/32, the required diameter is
32(1175 ⫻ 10⫺9 m4)
32IP
⫽ ᎏᎏᎏ ⫽ 11.97 ⫻ 10⫺6 m4
d 40⫽ ᎏ ᎏ
p
p
or
d0 ⫽ 0.0588 m ⫽ 58.8 mm
Comparing the two values of d0, we see that the rate of twist governs the design
and the required diameter of the solid shaft is
d0 ⫽ 58.8 mm
In a practical design, we would select a diameter slightly larger than the calculated
value of d0; for instance, 60 mm.
(b) Hollow shaft. Again, the required diameter is based upon either the
allowable shear stress or the allowable rate of twist. We begin by noting that the
outer diameter of the bar is d2 and the inner diameter is
d1 ⫽ d2 ⫺ 2t ⫽ d2 ⫺ 2(0.1d2) ⫽ 0.8d2
Thus, the polar moment of inertia (Eq. 3-16) is
冤
冥
p
p
p
IP ⫽ ᎏᎏ (d 42 ⫺ d 41) ⫽ ᎏᎏ d 42 ⫺ (0.8d2)4 ⫽ ᎏᎏ (0.5904d 42) ⫽ 0.05796d 42
32
32
32
In the case of the allowable shear stress, we use the torsion formula
(Eq. 3-11) as follows:
Tr
T
T(d2/2)
ᎏ
tallow ⫽ ᎏᎏ ⫽ ᎏ
4 ⫽ ᎏᎏ
IP
0.05796d 2
0.1159d 32
Rearranging, we get
T
1200 N⭈m
d 32 ⫽ ᎏᎏ ⫽ ᎏᎏ ⫽ 258.8 ⫻ 10⫺6 m3
0.1159tallow 0.1159(40 MPa)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 3.3 Circular Bars of Linearly Elastic Materials
199
Solving for d2 gives
d2 ⫽ 0.0637 m ⫽ 63.7 mm
which is the required outer diameter based upon the shear stress.
In the case of the allowable rate of twist, we use Eq. (3-14) with u replaced
by uallow and IP replaced by the previously obtained expression; thus,
T
uallow ⫽ ᎏᎏ
G(0.05796d 24 )
from which
T
ᎏ
d 42 ⫽ ᎏ
0.05796Guallow
1200 N⭈m
⫽ ᎏᎏᎏᎏᎏ ⫽ 20.28 ⫻ 10⫺6 m4
0.05796(78 GPa)(0.75°/m)( p rad/180°)
Solving for d2 gives
d2 ⫽ 0.0671 m ⫽ 67.1 mm
which is the required diameter based upon the rate of twist.
Comparing the two values of d2, we see that the rate of twist governs the
design and the required outer diameter of the hollow shaft is
d2 ⫽ 67.1 mm
The inner diameter d1 is equal to 0.8d2, or 53.7 mm. (As practical values, we
might select d2 ⫽ 70 mm and d1 ⫽ 0.8d2 ⫽ 56 mm.)
(c) Ratios of diameters and weights. The ratio of the outer diameter of the
hollow shaft to the diameter of the solid shaft (using the calculated values) is
d
67.1 mm
ᎏ 2ᎏ ⫽ ᎏᎏ ⫽ 1.14
d0
58.8 mm
Since the weights of the shafts are proportional to their cross-sectional
areas, we can express the ratio of the weight of the hollow shaft to the weight of
the solid shaft as follows:
d 22 ⫺ d 21
Whollow
Ahollow
p(d 22 ⫺ d 21)/4
ᎏ
ᎏ⫽ᎏ
ᎏ⫽ᎏ
ᎏ
⫽
ᎏ
ᎏ
Wsolid
Asolid
p d 20/4
d 20
(67.1 mm)2 ⫺ (53.7 mm)2
⫽ ᎏᎏᎏ
⫽ 0.47
(58.8 mm)2
These results show that the hollow shaft uses only 47% as much material as
does the solid shaft, while its outer diameter is only 14% larger.
Note: This example illustrates how to determine the required sizes of both
solid bars and circular tubes when allowable stresses and allowable rates of
twist are known. It also illustrates the fact that circular tubes are more efficient
in the use of materials than are solid circular bars.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
200
CHAPTER 3 Torsion
Example 3-3
A hollow shaft and a solid shaft constructed of the same material have the same
length and the same outer radius R (Fig. 3-13). The inner radius of the hollow
shaft is 0.6R.
(a) Assuming that both shafts are subjected to the same torque, compare
their shear stresses, angles of twist, and weights.
(b) Determine the strength-to-weight ratios for both shafts.
R
R
0.6R
FIG. 3-13 Example 3-3. Comparison of
(a)
hollow and solid shafts
(b)
Solution
(a) Comparison of shear stresses. The maximum shear stresses, given by
the torsion formula (Eq. 3-11), are proportional to 1/IP inasmuch as the torques
and radii are the same. For the hollow shaft, we get
p R4
p(0.6R)4
IP ⫽ ᎏᎏ ⫺ ᎏᎏ ⫽ 0.4352pR4
2
2
and for the solid shaft,
pR4
IP ⫽ ᎏᎏ ⫽ 0.5pR4
2
Therefore, the ratio b1 of the maximum shear stress in the hollow shaft to that in
the solid shaft is
t
0.5p R 4
ᎏ ⫽ 1.15
b1 ⫽ ᎏHᎏ ⫽ ᎏ
tS
0.4352p R 4
where the subscripts H and S refer to the hollow shaft and the solid shaft,
respectively.
Comparison of angles of twist. The angles of twist (Eq. 3-15) are also
proportional to 1/IP, because the torques T, lengths L, and moduli of elasticity G
are the same for both shafts. Therefore, their ratio is the same as for the shear
stresses:
fH
.5pR 4
⫽ 0ᎏ
ᎏ ⫽ 1.15
b2 ⫽ ᎏᎏ
fS
0.4352pR 4
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 3.3 Circular Bars of Linearly Elastic Materials
201
Comparison of weights. The weights of the shafts are proportional to their
cross-sectional areas; consequently, the weight of the solid shaft is proportional
to pR2 and the weight of the hollow shaft is proportional to
pR2 ⫺ p(0.6R)2 ⫽ 0.64pR2
Therefore, the ratio of the weight of the hollow shaft to the weight of the solid
shaft is
WH
0.64p R 2
⫽ ᎏᎏ
⫽ 0.64
b3 ⫽ ᎏᎏ
WS
p R2
From the preceding ratios we again see the inherent advantage of hollow
shafts. In this example, the hollow shaft has 15% greater stress and 15% greater
angle of rotation than the solid shaft but 36% less weight.
(b) Strength-to-weight ratios. The relative efficiency of a structure is sometimes measured by its strength-to-weight ratio, which is defined for a bar in
torsion as the allowable torque divided by the weight. The allowable torque for
the hollow shaft of Fig. 3-13a (from the torsion formula) is
tmax(0.4352pR4)
tmaxIP
ᎏ⫽ᎏ
ᎏ ⫽ 0.4352pR3tmax
TH ⫽ ᎏ
R
R
and for the solid shaft is
tmaxIP
tmax(0.5pR 4)
ᎏ⫽ᎏ
ᎏ ⫽ 0.5pR3tmax
TS ⫽ ᎏ
R
R
The weights of the shafts are equal to the cross-sectional areas times the length
L times the weight density g of the material:
WH ⫽ 0.64pR2Lg
WS ⫽ pR2Lg
Thus, the strength-to-weight ratios SH and SS for the hollow and solid bars,
respectively, are
tmaxR
TH
SH ⫽ ᎏ ᎏ
⫽ 0.68 ᎏ
ᎏ
WH
gL
tmaxR
TS
SS ⫽ ᎏᎏ
⫽ 0.5 ᎏ
ᎏ
WS
gL
In this example, the strength-to-weight ratio of the hollow shaft is 36% greater
than the strength-to-weight ratio for the solid shaft, demonstrating once again
the relative efficiency of hollow shafts. For a thinner shaft, the percentage will
increase; for a thicker shaft, it will decrease.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
202
CHAPTER 3 Torsion
3.4 NONUNIFORM TORSION
T1
T2
A
T3
B
T4
C
LAB
D
LBC
LCD
(a)
T1
T2
T3
TCD
A
B
C
(b)
T2
T1
TBC
A
B
(c)
T1
TCD ⫽ ⫺T1 ⫺ T2 ⫹ T3
TAB
A
(d)
FIG. 3-14 Bar in nonuniform torsion
(Case 1)
As explained in Section 3.2, pure torsion refers to torsion of a prismatic
bar subjected to torques acting only at the ends. Nonuniform torsion
differs from pure torsion in that the bar need not be prismatic and the
applied torques may act anywhere along the axis of the bar. Bars in
nonuniform torsion can be analyzed by applying the formulas of pure
torsion to finite segments of the bar and then adding the results, or
by applying the formulas to differential elements of the bar and then
integrating.
To illustrate these procedures, we will consider three cases of
nonuniform torsion. Other cases can be handled by techniques similar to
those described here.
Case 1. Bar consisting of prismatic segments with constant torque
throughout each segment (Fig. 3-14). The bar shown in part (a) of the
figure has two different diameters and is loaded by torques acting at
points A, B, C, and D. Consequently, we divide the bar into segments in
such a way that each segment is prismatic and subjected to a constant
torque. In this example, there are three such segments, AB, BC, and CD.
Each segment is in pure torsion, and therefore all of the formulas
derived in the preceding section may be applied to each part separately.
The first step in the analysis is to determine the magnitude and
direction of the internal torque in each segment. Usually the torques can
be determined by inspection, but if necessary they can be found by
cutting sections through the bar, drawing free-body diagrams, and
solving equations of equilibrium. This process is illustrated in parts (b),
(c), and (d) of the figure. The first cut is made anywhere in segment CD,
thereby exposing the internal torque TCD. From the free-body diagram
(Fig. 3-14b), we see that TCD is equal to ⫺T1 ⫺ T2 ⫹ T3. From the next
diagram we see that TBC equals ⫺T1 ⫺ T2, and from the last we find that
TAB equals ⫺T1. Thus,
TBC ⫽ ⫺T1 ⫺ T2
TAB ⫽ ⫺T1 (a,b,c)
Each of these torques is constant throughout the length of its segment.
When finding the shear stresses in each segment, we need only the
magnitudes of these internal torques, since the directions of the stresses
are not of interest. However, when finding the angle of twist for the
entire bar, we need to know the direction of twist in each segment in
order to combine the angles of twist correctly. Therefore, we need to
establish a sign convention for the internal torques. A convenient rule in
many cases is the following: An internal torque is positive when its
vector points away from the cut section and negative when its vector
points toward the section. Thus, all of the internal torques shown in
Figs. 3-14b, c, and d are pictured in their positive directions. If the
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 3.4 Nonuniform Torsion
203
calculated torque (from Eq. a, b, or c) turns out to have a positive sign, it
means that the torque acts in the assumed direction; if the torque has a
negative sign, it acts in the opposite direction.
The maximum shear stress in each segment of the bar is readily
obtained from the torsion formula (Eq. 3-11) using the appropriate
cross-sectional dimensions and internal torque. For instance, the
maximum stress in segment BC (Fig. 3-14) is found using the diameter
of that segment and the torque TBC calculated from Eq. (b). The
maximum stress in the entire bar is the largest stress from among the
stresses calculated for each of the three segments.
The angle of twist for each segment is found from Eq. (3-15), again
using the appropriate dimensions and torque. The total angle of twist of
one end of the bar with respect to the other is then obtained by algebraic
summation, as follows:
f ⫽ f1 ⫹ f2 ⫹ . . . ⫹ fn
(3-19)
where f1 is the angle of twist for segment 1, f2 is the angle for segment
2, and so on, and n is the total number of segments. Since each angle of
twist is found from Eq. (3-15), we can write the general formula
n
n
T1L1
f ⫽ 冱 fi ⫽ 冱 ᎏ
ᎏ
i⫽1
i⫽1 Gi(IP)i
T
T
B
A
x
dx
L
FIG. 3-15 Bar in nonuniform torsion
(Case 2)
(3-20)
in which the subscript i is a numbering index for the various segments.
For segment i of the bar, Ti is the internal torque (found from equilibrium,
as illustrated in Fig. 3-14), Li is the length, Gi is the shear modulus, and
(IP)i is the polar moment of inertia. Some of the torques (and the
corresponding angles of twist) may be positive and some may be
negative. By summing algebraically the angles of twist for all segments,
we obtain the total angle of twist f between the ends of the bar. The
process is illustrated later in Example 3-4.
Case 2. Bar with continuously varying cross sections and constant
torque (Fig. 3-15). When the torque is constant, the maximum shear
stress in a solid bar always occurs at the cross section having the
smallest diameter, as shown by Eq. (3-12). Furthermore, this observation
usually holds for tubular bars. If this is the case, we only need to
investigate the smallest cross section in order to calculate the maximum
shear stress. Otherwise, it may be necessary to evaluate the stresses at
more than one location in order to obtain the maximum.
To find the angle of twist, we consider an element of length dx at
distance x from one end of the bar (Fig. 3-15). The differential angle of
rotation df for this element is
T dx
df ⫽ ᎏᎏ
GIP(x)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(d)
204
CHAPTER 3 Torsion
in which IP(x) is the polar moment of inertia of the cross section at
distance x from the end. The angle of twist for the entire bar is the
summation of the differential angles of rotation:
L
f⫽
L
x
T dᎏ
冮 df ⫽ 冮 ᎏ
GI (x)
0
0
(3-21)
P
If the expression for the polar moment of inertia IP(x) is not too
complex, this integral can be evaluated analytically, as in Example 3-5.
In other cases, it must be evaluated numerically.
Case 3. Bar with continuously varying cross sections and continuously varying torque (Fig. 3-16). The bar shown in part (a) of the figure
is subjected to a distributed torque of intensity t per unit distance along
the axis of the bar. As a result, the internal torque T(x) varies continuously along the axis (Fig. 3-16b). The internal torque can be evaluated
with the aid of a free-body diagram and an equation of equilibrium. As
in Case 2, the polar moment of inertia IP(x) can be evaluated from the
cross-sectional dimensions of the bar.
t
TA
TB
B
A
x
dx
L
(a)
t
TA
FIG. 3-16 Bar in nonuniform torsion
(Case 3)
A
T(x)
x
(b)
Knowing both the torque and polar moment of inertia as functions
of x, we can use the torsion formula to determine how the shear stress
varies along the axis of the bar. The cross section of maximum shear
stress can then be identified, and the maximum shear stress can be
determined.
The angle of twist for the bar of Fig. 3-16a can be found in the same
manner as described for Case 2. The only difference is that the torque,
like the polar moment of inertia, also varies along the axis. Consequently, the equation for the angle of twist becomes
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 3.4 Nonuniform Torsion
L
f⫽
205
L
T(x ) dx
ᎏ
冮 df ⫽ 冮 ᎏ
GI (x)
0
0
(3-22)
P
This integral can be evaluated analytically in some cases, but usually it
must be evaluated numerically.
Limitations
The analyses described in this section are valid for bars made of linearly
elastic materials with circular cross sections (either solid or hollow).
Also, the stresses determined from the torsion formula are valid in
regions of the bar away from stress concentrations, which are high localized stresses that occur wherever the diameter changes abruptly and
wherever concentrated torques are applied (see Section 3.11). However,
stress concentrations have relatively little effect on the angle of twist,
and therefore the equations for f are generally valid.
Finally, we must keep in mind that the torsion formula and the
formulas for angles of twist were derived for prismatic bars. We can
safely apply them to bars with varying cross sections only when the
changes in diameter are small and gradual. As a rule of thumb, the
formulas given here are satisfactory as long as the angle of taper (the
angle between the sides of the bar) is less than 10°.
Example 3-4
A solid steel shaft ABCDE (Fig. 3-17) having diameter d ⫽ 30 mm turns freely
in bearings at points A and E. The shaft is driven by a gear at C, which applies a
torque T2 ⫽ 450 N⭈m in the direction shown in the figure. Gears at B and D are
driven by the shaft and have resisting torques T1 ⫽ 275 N⭈m and T3 ⫽ 175 N⭈m,
respectively, acting in the opposite direction to the torque T2. Segments BC and
CD have lengths LBC ⫽ 500 mm and LCD ⫽ 400 mm, respectively, and the
shear modulus G ⫽ 80 GPa.
Determine the maximum shear stress in each part of the shaft and the angle
of twist between gears B and D.
T1
T2
T3
d
A
FIG. 3-17 Example 3-4. Steel shaft in
torsion
E
B
C
LBC
D
LCD
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
206
CHAPTER 3 Torsion
Solution
T2
T1
d
TCD
B
C
LBC
Each segment of the bar is prismatic and subjected to a constant torque
(Case 1). Therefore, the first step in the analysis is to determine the torques acting
in the segments, after which we can find the shear stresses and angles of twist.
Torques acting in the segments. The torques in the end segments (AB and
DE) are zero since we are disregarding any friction in the bearings at the
supports. Therefore, the end segments have no stresses and no angles of twist.
The torque TCD in segment CD is found by cutting a section through the
segment and constructing a free-body diagram, as in Fig. 3-18a. The torque is
assumed to be positive, and therefore its vector points away from the cut
section. From equilibrium of the free body, we obtain
TCD ⫽ T2 ⫺ T1 ⫽ 450 N⭈m ⫺ 275 N⭈m ⫽ 175 N⭈m
(a)
T1
TBC
The positive sign in the result means that TCD acts in the assumed positive
direction.
The torque in segment BC is found in a similar manner, using the free-body
diagram of Fig. 3-18b:
TBC ⫽ ⫺T1 ⫽ ⫺275 N⭈m
B
(b)
FIG. 3-18 Free-body diagrams for
Example 3-4
Note that this torque has a negative sign, which means that its direction is opposite
to the direction shown in the figure.
Shear stresses. The maximum shear stresses in segments BC and CD are
found from the modified form of the torsion formula (Eq. 3-12); thus,
16TBC
16(275 N⭈m)
⫽ 51.9 MPa
tBC ⫽ ᎏᎏ
3 ⫽ ᎏᎏ
pd
p (30 mm)3
16TCD
16(175 N⭈m)
⫽ ᎏᎏ
⫽ 33.0 MPa
tCD ⫽ ᎏᎏ
pd 3
p (30 mm)3
Since the directions of the shear stresses are not of interest in this example, only
absolute values of the torques are used in the preceding calculations.
Angles of twist. The angle of twist f BD between gears B and D is the algebraic sum of the angles of twist for the intervening segments of the bar, as given
by Eq. (3-19); thus,
fBD ⫽ fBC ⫹ fCD
When calculating the individual angles of twist, we need the moment of inertia
of the cross section:
p d4
p (30 mm)4
IP ⫽ ᎏᎏ ⫽ ᎏᎏ ⫽ 79,520 mm4
32
32
Now we can determine the angles of twist, as follows:
(⫺275 N⭈m)(500 mm)
TBCLBC
ᎏ ⫽ ᎏᎏᎏ
⫽ ⫺0.0216 rad
f BC ⫽ ᎏ
(80
GPa)(79,520 mm4)
GIP
(175 N⭈m)(400 mm)
TCD LCD
fCD ⫽ ᎏ
ᎏ ⫽ ᎏᎏᎏ
⫽ 0.0110 rad
(80
GPa)(79,520 mm4)
GIP
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 3.4 Nonuniform Torsion
207
Note that in this example the angles of twist have opposite directions. Adding
algebraically, we obtain the total angle of twist:
fBD ⫽ fBC ⫹ fCD ⫽ ⫺0.0216 ⫹ 0.0110 ⫽ ⫺0.0106 rad ⫽ ⫺0.61°
The minus sign means that gear D rotates clockwise (when viewed from the
right-hand end of the shaft) with respect to gear B. However, for most purposes
only the absolute value of the angle of twist is needed, and therefore it is
sufficient to say that the angle of twist between gears B and D is 0.61°. The
angle of twist between the two ends of a shaft is sometimes called the wind-up.
Notes: The procedures illustrated in this example can be used for shafts
having segments of different diameters or of different materials, as long as the
dimensions and properties remain constant within each segment.
Only the effects of torsion are considered in this example and in the problems
at the end of the chapter. Bending effects are considered later, beginning with
Chapter 4.
Example 3-5
T
B
A
T
x
dx
L
dA
dB
FIG. 3-19 Example 3-5. Tapered bar in
torsion
A tapered bar AB of solid circular cross section is twisted by torques T applied
at the ends (Fig. 3-19). The diameter of the bar varies linearly from dA at the
left-hand end to dB at the right-hand end, with dB assumed to be greater than dA.
(a) Determine the maximum shear stress in the bar.
(b) Derive a formula for the angle of twist of the bar.
Solution
(a) Shear stresses. Since the maximum shear stress at any cross section in
a solid bar is given by the modified torsion formula (Eq. 3-12), we know immediately that the maximum shear stress occurs at the cross section having the
smallest diameter, that is, at end A (see Fig. 3-19):
16T
tmax ⫽ ᎏᎏ
p d 3A
(b) Angle of twist. Because the torque is constant and the polar moment of
inertia varies continuously with the distance x from end A (Case 2), we will use
Eq. (3-21) to determine the angle of twist. We begin by setting up an expression
for the diameter d at distance x from end A:
dB ⫺ dA
ᎏx
d ⫽ dA ⫹ ᎏ
L
(3-23)
in which L is the length of the bar. We can now write an expression for the polar
moment of inertia:
冢
冣
pd 4
p
dB ⫺ dA
ᎏx
IP(x) ⫽ ᎏᎏ ⫽ ᎏᎏ dA ⫹ ᎏ
32
32
L
4
(3-24)
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
208
CHAPTER 3 Torsion
Substituting this expression into Eq. (3-21), we get a formula for the angle of
twist:
32T
T dx
dx
ᎏ ⫽ ᎏᎏ 冮 ᎏᎏ
冮ᎏ
GI (x)
pG
d ⫺d
L
f⫽
0
L
P
0
冢
(3-25)
冣
B
A
dA ⫹ ᎏ
ᎏ
x
L
4
To evaluate the integral in this equation, we note that it is of the form
dx
ᎏ
冮ᎏ
(a ⫹ bx)
4
in which
dB ⫺ dA
b⫽ᎏ
ᎏ
L
a ⫽ dA
(e,f)
With the aid of a table of integrals (see Appendix C), we find
dx
1
ᎏ ⫽ ⫺ᎏᎏ
冮ᎏ
(a ⫹ bx)
3b(a ⫹ bx)
4
3
This integral is evaluated in our case by substituting for x the limits 0 and L and
substituting for a and b the expressions in Eqs. (e) and (f). Thus, the integral in
Eq. (3-25) equals
冢
冣
1
1
L
ᎏᎏ ᎏᎏ
⫺ ᎏᎏ
3(dB ⫺ dA) d 3A
d 3B
(g)
Replacing the integral in Eq. (3-25) with this expression, we obtain
冢
冣
1
1
32TL
⫺ ᎏᎏ
f ⫽ ᎏᎏ ᎏᎏ
3pG(dB ⫺ dA) d 3A
d 3B
(3-26)
which is the desired equation for the angle of twist of the tapered bar.
A convenient form in which to write the preceding equation is
冢
冣
TL
b2 ⫹ b ⫹ 1
f ⫽ ᎏᎏ ᎏᎏ
G(IP)A
3b 3
(3-27)
p d 4A
(IP)A ⫽ ᎏᎏ
32
(3-28)
in which
dB
b ⫽ ᎏᎏ
dA
The quantity b is the ratio of end diameters and (IP)A is the polar moment of
inertia at end A.
In the special case of a prismatic bar, we have b ⫽ 1 and Eq. (3-27) gives
f ⫽ TL/G(IP)A, as expected. For values of b greater than 1, the angle of rotation
decreases because the larger diameter at end B produces an increase in the
torsional stiffness (as compared to a prismatic bar).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 3.5 Stresses and Strains in Pure Shear
209
3.5 STRESSES AND STRAINS IN PURE SHEAR
T
T
a b
d c
(a)
t
a
t
y b
O
d
t
x
c
t
(b)
FIG. 3-20 Stresses acting on a stress
element cut from a bar in torsion (pure
shear)
When a circular bar, either solid or hollow, is subjected to torsion, shear
stresses act over the cross sections and on longitudinal planes, as illustrated previously in Fig. 3-7. We will now examine in more detail the
stresses and strains produced during twisting of a bar.
We begin by considering a stress element abcd cut between two cross
sections of a bar in torsion (Figs. 3-20a and b). This element is in a state of
pure shear, because the only stresses acting on it are the shear stresses t
on the four side faces (see the discussion of shear stresses in Section 1.6.)
The directions of these shear stresses depend upon the directions of
the applied torques T. In this discussion, we assume that the torques
rotate the right-hand end of the bar clockwise when viewed from the
right (Fig. 3-20a); hence the shear stresses acting on the element have
the directions shown in the figure. This same state of stress exists for a
similar element cut from the interior of the bar, except that the magnitudes of the shear stresses are smaller because the radial distance to the
element is smaller.
The directions of the torques shown in Fig. 3-20a are intentionally
chosen so that the resulting shear stresses (Fig. 3-20b) are positive
according to the sign convention for shear stresses described previously
in Section 1.6. This sign convention is repeated here:
A shear stress acting on a positive face of an element is positive if it
acts in the positive direction of one of the coordinate axes and negative
if it acts in the negative direction of an axis. Conversely, a shear stress
acting on a negative face of an element is positive if it acts in the negative direction of one of the coordinate axes and negative if it acts in the
positive direction of an axis.
Applying this sign convention to the shear stresses acting on the
stress element of Fig. 3-20b, we see that all four shear stresses are
positive. For instance, the stress on the right-hand face (which is a positive face because the x axis is directed to the right) acts in the positive
direction of the y axis; therefore, it is a positive shear stress. Also, the
stress on the left-hand face (which is a negative face) acts in the negative
direction of the y axis; therefore, it is a positive shear stress. Analogous
comments apply to the remaining stresses.
Stresses on Inclined Planes
We are now ready to determine the stresses acting on inclined planes cut
through the stress element in pure shear. We will follow the same approach
as the one we used in Section 2.6 for investigating the stresses in uniaxial
stress.
A two-dimensional view of the stress element is shown in Fig. 3-21a
on the next page. As explained previously in Section 2.6, we usually draw a
two-dimensional view for convenience, but we must always be aware that
the element has a third dimension (thickness) perpendicular to the plane of
the figure.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
210
CHAPTER 3 Torsion
t
a
b
y
FIG. 3-21 Analysis of stresses on inclined
planes: (a) element in pure shear,
(b) stresses acting on a triangular stress
element, and (c) forces acting on the
triangular stress element (free-body
diagram)
O
t
x
d
su
tu
t
t
t A0
c
t
90°⫺ u
t A0 tan u
t
(a)
tu A0 sec u
su A0 sec u
u
u
(b)
(c)
We now cut from the element a wedge-shaped (or “triangular”) stress
element having one face oriented at an angle u to the x axis (Fig. 3-21b).
Normal stresses su and shear stresses tu act on this inclined face and are
shown in their positive directions in the figure. The sign convention for
stresses su and tu was described previously in Section 2.6 and is
repeated here:
Normal stresses su are positive in tension and shear stresses tu are
positive when they tend to produce counterclockwise rotation of the
material. (Note that this sign convention for the shear stress tu acting on
an inclined plane is different from the sign convention for ordinary shear
stresses t that act on the sides of rectangular elements oriented to a set
of xy axes.)
The horizontal and vertical faces of the triangular element (Fig. 3-21b)
have positive shear stresses t acting on them, and the front and rear faces
of the element are free of stress. Therefore, all stresses acting on the
element are visible in this figure.
The stresses su and tu may now be determined from the equilibrium
of the triangular element. The forces acting on its three side faces can be
obtained by multiplying the stresses by the areas over which they act.
For instance, the force on the left-hand face is equal to tA0, where A0 is
the area of the vertical face. This force acts in the negative y direction
and is shown in the free-body diagram of Fig. 3-21c. Because the
thickness of the element in the z direction is constant, we see that the
area of the bottom face is A0 tan u and the area of the inclined
face is A0 sec u. Multiplying the stresses acting on these faces by the
corresponding areas enables us to obtain the remaining forces and
thereby complete the free-body diagram (Fig. 3-21c).
We are now ready to write two equations of equilibrium for the
triangular element, one in the direction of su and the other in the direction of tu. When writing these equations, the forces acting on the
left-hand and bottom faces must be resolved into components in the
directions of su and tu. Thus, the first equation, obtained by summing
forces in the direction of su, is
su A0 sec u ⫽ tA0 sin u ⫹ tA0 tan u cos u
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 3.5 Stresses and Strains in Pure Shear
211
or
su ⫽ 2t sin u cos u
(3-29a)
The second equation is obtained by summing forces in the direction of
tu:
tu A0 sec u ⫽ tA0 cos u ⫺ tA0 tan u sin u
or
tu ⫽ t (cos2u ⫺ sin2u)
(3-29b)
These equations can be expressed in simpler forms by introducing the
following trigonometric identities (see Appendix C):
sin 2u ⫽ 2 sin u cos u
cos 2u ⫽ cos2 u ⫺ sin2 u
Then the equations for su and tu become
su ⫽ t sin 2u
su or tu
t
tu
–90°
– 45°
su
tu
0
45°
su
u
–t
FIG. 3-22 Graph of normal stresses su
and shear stresses tu versus angle u of
the inclined plane
90°
tu ⫽ t cos 2u
(3-30a,b)
Equations (3-30a and b) give the normal and shear stresses acting on any
inclined plane in terms of the shear stresses t acting on the x and y
planes (Fig. 3-21a) and the angle u defining the orientation of the
inclined plane (Fig. 3-21b).
The manner in which the stresses su and tu vary with the orientation
of the inclined plane is shown by the graph in Fig. 3-22, which is a plot
of Eqs. (3-30a and b). We see that for u ⫽ 0, which is the right-hand face
of the stress element in Fig. 3-21a, the graph gives su ⫽ 0 and tu ⫽ t. This
latter result is expected, because the shear stress t acts counterclockwise
against the element and therefore produces a positive shear stress tu.
For the top face of the element (u ⫽ 90°), we obtain su ⫽ 0 and
tu ⫽ ⫺t. The minus sign for tu means that it acts clockwise against the
element, that is, to the right on face ab (Fig. 3-21a), which is consistent
with the direction of the shear stress t. Note that the numerically largest
shear stresses occur on the planes for which u ⫽ 0 and 90°, as well as on
the opposite faces (u ⫽ 180° and 270°).
From the graph we see that the normal stress su reaches a maximum
value at u ⫽ 45°. At that angle, the stress is positive (tension) and equal
numerically to the shear stress t. Similarly, su has its minimum value
(which is compressive) at u ⫽ ⫺45°. At both of these 45° angles, the
shear stress tu is equal to zero. These conditions are pictured in Fig. 3-23
on the next page, which shows stress elements oriented at u ⫽ 0 and
u ⫽ 45°. The element at 45° is acted upon by equal tensile and compressive stresses in perpendicular directions, with no shear stresses.
Note that the normal stresses acting on the 45° element (Fig. 3-23b)
correspond to an element subjected to shear stresses t acting in the
directions shown in Fig. 3-23a. If the shear stresses acting on the
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
212
CHAPTER 3 Torsion
smin = – t
s max = t
t
y
45°
y
t
O
O
x
x
t
t
FIG. 3-23 Stress elements oriented at
(a)
u ⫽ 0 and u ⫽ 45° for pure shear
s min = – t
smax = t
(b)
element of Fig. 3-23a are reversed in direction, the normal stresses
acting on the 45° planes also will change directions.
If a stress element is oriented at an angle other than 45°, both
normal and shear stresses will act on the inclined faces (see Eqs. 3-30a
and b and Fig. 3-22). Stress elements subjected to these more general
conditions are discussed in detail in Chapter 7.
The equations derived in this section are valid for a stress element in
pure shear regardless of whether the element is cut from a bar in torsion
or from some other structural element. Also, since Eqs. (3-30) were
derived from equilibrium only, they are valid for any material, whether
or not it behaves in a linearly elastic manner.
The existence of maximum tensile stresses on planes at 45° to the
x axis (Fig. 3-23b) explains why bars in torsion that are made of materials
that are brittle and weak in tension fail by cracking along a 45° helical
surface (Fig. 3-24). As mentioned in Section 3.3, this type of failure is
readily demonstrated by twisting a piece of classroom chalk.
Strains in Pure Shear
Let us now consider the strains that exist in an element in pure shear. For
instance, consider the element in pure shear shown in Fig. 3-23a. The corresponding shear strains are shown in Fig. 3-25a, where the deformations
are highly exaggerated. The shear strain g is the change in angle between
two lines that were originally perpendicular to each other, as discussed
previously in Section 1.6. Thus, the decrease in the angle at the lower lefthand corner of the element is the shear strain g (measured in radians).
This same change in angle occurs at the upper right-hand corner, where
T
45° Crack
FIG. 3-24 Torsion failure of a brittle
material by tension cracking along a
45° helical surface
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
T
SECTION 3.5 Stresses and Strains in Pure Shear
213
the angle decreases, and at the other two corners, where the angles
increase. However, the lengths of the sides of the element, including the
thickness perpendicular to the plane of the paper, do not change when
these shear deformations occur. Therefore, the element changes its shape
from a rectangular parallelepiped (Fig. 3-23a) to an oblique parallelepiped
(Fig. 3-25a). This change in shape is called a shear distortion.
If the material is linearly elastic, the shear strain for the element
oriented at u ⫽ 0 (Fig. 3-25a) is related to the shear stress by Hooke’s
law in shear:
t
g ⫽ ᎏᎏ
G
(3-31)
where, as usual, the symbol G represents the shear modulus of elasticity.
smax = t
smin = – t
t
45°
t
t
p ⫺g
2
FIG. 3-25 Strains in pure shear: (a) shear
distortion of an element oriented at
u ⫽ 0, and (b) distortion of an element
oriented at u ⫽ 45°
t
smin = – t
smax = t
(a)
(b)
Next, consider the strains that occur in an element oriented at
u ⫽ 45° (Fig. 3-25b). The tensile stresses acting at 45° tend to elongate
the element in that direction. Because of the Poisson effect, they also
tend to shorten it in the perpendicular direction (the direction where
u ⫽ 135° or ⫺45°). Similarly, the compressive stresses acting at 135°
tend to shorten the element in that direction and elongate it in the 45°
direction. These dimensional changes are shown in Fig. 3-25b, where
the dashed lines show the deformed element. Since there are no shear
distortions, the element remains a rectangular parallelepiped even
though its dimensions have changed.
If the material is linearly elastic and follows Hooke’s law, we can
obtain an equation relating strain to stress for the element at u ⫽ 45°
(Fig. 3-25b). The tensile stress smax acting at u ⫽ 45° produces a positive normal strain in that direction equal to smax/E. Since smax ⫽ t, we
can also express this strain as t /E. The stress smax also produces a negative strain in the perpendicular direction equal to ⫺nt/E, where n is
Poisson’s ratio. Similarly, the stress smin ⫽ ⫺t (at u ⫽ 135°) produces a
negative strain equal to ⫺t/E in that direction and a positive strain in the
perpendicular direction (the 45° direction) equal to n t /E. Therefore, the
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
214
CHAPTER 3 Torsion
normal strain in the 45° direction is
t
t
nt
emax ⫽ ᎏᎏ ⫹ ᎏᎏ ⫽ ᎏᎏ (1 ⫹ n)
E
E
E
(3-32)
which is positive, representing elongation. The strain in the perpendicular direction is a negative strain of the same amount. In other words,
pure shear produces elongation in the 45° direction and shortening in the
135° direction. These strains are consistent with the shape of the
deformed element of Fig. 3-25a, because the 45° diagonal has lengthened
and the 135° diagonal has shortened.
In the next section we will use the geometry of the deformed
element to relate the shear strain g (Fig. 3-25a) to the normal strain emax
in the 45° direction (Fig. 3-25b). In so doing, we will derive the
following relationship:
g
emax ⫽ ᎏᎏ
2
(3-33)
This equation, in conjunction with Eq. (3-31), can be used to calculate
the maximum shear strains and maximum normal strains in pure torsion
when the shear stress t is known.
Example 3-6
T = 4.0 kN⋅m
T
A circular tube with an outside diameter of 80 mm and an inside diameter of
60 mm is subjected to a torque T ⫽ 4.0 kN⭈m (Fig. 3-26). The tube is made of
aluminum alloy 7075-T6.
(a) Determine the maximum shear, tensile, and compressive stresses in the
tube and show these stresses on sketches of properly oriented stress elements.
(b) Determine the corresponding maximum strains in the tube and show
these strains on sketches of the deformed elements.
Solution
(a) Maximum stresses. The maximum values of all three stresses (shear,
tensile, and compressive) are equal numerically, although they act on different
planes. Their magnitudes are found from the torsion formula:
60
mm
80
mm
FIG. 3-26 Example 3-6. Circular tube in
torsion
(4000 N⭈m)(0.040 m)
Tr
⫽ 58.2 MPa
tmax ⫽ ᎏᎏ ⫽ ᎏᎏᎏᎏ
p
IP
ᎏᎏ (0.080 m)4 ⫺ (0.060 m)4
32
冤
冥
The maximum shear stresses act on cross-sectional and longitudinal planes, as
shown by the stress element in Fig. 3-27a, where the x axis is parallel to the
longitudinal axis of the tube.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 3.5 Stresses and Strains in Pure Shear
215
The maximum tensile and compressive stresses are
st ⫽ 58.2 MPa
sc ⫽ ⫺58.2 MPa
These stresses act on planes at 45° to the axis (Fig. 3-27b).
(b) Maximum strains. The maximum shear strain in the tube is obtained
from Eq. (3-31). The shear modulus of elasticity is obtained from Table H-2,
Appendix H, as G ⫽ 27 GPa. Therefore, the maximum shear strain is
tmax
58.2 MPa
gmax ⫽ ᎏᎏ ⫽ ᎏᎏ ⫽ 0.0022 rad
G
27 GPa
The deformed element is shown by the dashed lines in Fig. 3-27c.
The magnitude of the maximum normal strains (from Eq. 3-33) is
gmax
emax ⫽ ᎏᎏ ⫽ 0.0011
2
Thus, the maximum tensile and compressive strains are
et ⫽ 0.0011
ec ⫽ ⫺0.0011
The deformed element is shown by the dashed lines in Fig. 3-27d for an element
with sides of unit length.
sc = 58.2 MPa
58.2 MPa
y
O
45°
y
x
t max =
58.2 MPa
O
x
st = 58.2 MPa
(a)
(b)
45°
FIG. 3-27 Stress and strain elements for
the tube of Example 3-6: (a) maximum
shear stresses, (b) maximum tensile and
compressive stresses; (c) maximum
shear strains, and (d) maximum tensile
and compressive strains
g max =
0.0022 rad
e t = 0.0011
(c)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
1
1
e c = 0.0011
(d)
216
CHAPTER 3 Torsion
3.6 RELATIONSHIP BETWEEN MODULI OF ELASTICITY E AND G
An important relationship between the moduli of elasticity E and G can
be obtained from the equations derived in the preceding section. For this
purpose, consider the stress element abcd shown in Fig. 3-28a. The front
face of the element is assumed to be square, with the length of each side
denoted as h. When this element is subjected to pure shear by stresses t,
the front face distorts into a rhombus (Fig. 3-28b) with sides of length h
and with shear strain g ⫽ t /G. Because of the distortion, diagonal bd is
lengthened and diagonal ac is shortened. The length of diagonal bd is
苶 h times the factor 1 ⫹ emax, where emax is
equal to its initial length 兹2
the normal strain in the 45° direction; thus,
Lbd ⫽ 兹2
苶 h(1 ⫹ emax)
(a)
This length can be related to the shear strain g by considering the geometry of the deformed element.
To obtain the required geometric relationships, consider triangle abd
(Fig. 3-28c) which represents one-half of the rhombus pictured in Fig.
3-28b. Side bd of this triangle has length Lbd (Eq. a), and the other sides
have length h. Angle adb of the triangle is equal to one-half of angle adc
of the rhombus, or p /4 ⫺ g /2. The angle abd in the triangle is the same.
Therefore, angle dab of the triangle equals p/2 ⫹ g. Now using the law
of cosines (see Appendix C) for triangle abd, we get
冢
冣
p
L 2bd ⫽ h2 ⫹ h2 ⫺ 2h2 cos ᎏᎏ ⫹ g
2
Substituting for Lbd from Eq. (a) and simplifying, we get
冢
冣
p
(1 ⫹ emax)2 ⫽ 1 ⫺ cos ᎏᎏ ⫹ g
2
By expanding the term on the left-hand side, and also observing that
cos(p/2 ⫹ g) ⫽ ⫺sin g, we obtain
1 ⫹ 2emax ⫹ e2max ⫽ 1 ⫹ sin g
t
b
a
t
FIG. 3-28 Geometry of deformed
element in pure shear
c
h
(a)
t
–p– – g
2
h
d
b
a
d
c
t
(b)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
a
–p– + g
2
h
h
L bd
d –p– – –g–
4 2
(c)
b
g
–p– – ––
4 2
SECTION 3.7 Transmission of Power by Circular Shafts
217
Because emax and g are very small strains, we can disregard e 2max in comparison with 2emax and we can replace sin g by g. The resulting expression is
g
emax ⫽ ᎏᎏ
2
(3-34)
which establishes the relationship already presented in Section 3.5 as
Eq. (3-33).
The shear strain g appearing in Eq. (3-34) is equal to t /G by Hooke’s
law (Eq. 3-31) and the normal strain emax is equal to t (1 ⫹ n)/E by Eq.
(3-32). Making both of these substitutions in Eq. (3-34) yields
E
G ⫽ ᎏᎏ
2(1 ⫹ n)
(3-35)
We see that E, G, and n are not independent properties of a linearly
elastic material. Instead, if any two of them are known, the third can be
calculated from Eq. (3-35).
Typical values of E, G, and n are listed in Table H-2, Appendix H.
3.7 TRANSMISSION OF POWER BY CIRCULAR SHAFTS
The most important use of circular shafts is to transmit mechanical
power from one device or machine to another, as in the drive shaft of an
automobile, the propeller shaft of a ship, or the axle of a bicycle. The
power is transmitted through the rotary motion of the shaft, and the
amount of power transmitted depends upon the magnitude of the torque
and the speed of rotation. A common design problem is to determine the
required size of a shaft so that it will transmit a specified amount of
power at a specified rotational speed without exceeding the allowable
stresses for the material.
Let us suppose that a motor-driven shaft (Fig. 3-29) is rotating at an
angular speed v, measured in radians per second (rad/s). The shaft transmits a torque T to a device (not shown in the figure) that is performing
useful work. The torque applied by the shaft to the external device has
the same sense as the angular speed v, that is, its vector points to the
Motor
v
T
FIG. 3-29 Shaft transmitting a constant
torque T at an angular speed v
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
218
CHAPTER 3 Torsion
left. However, the torque shown in the figure is the torque exerted on the
shaft by the device, and so its vector points in the opposite direction.
In general, the work W done by a torque of constant magnitude is
equal to the product of the torque and the angle through which it rotates;
that is,
W ⫽ Tc
(3-36)
where c is the angle of rotation in radians.
Power is the rate at which work is done, or
dW
dc
P ⫽ ᎏᎏ ⫽ T ᎏᎏ
dt
dt
(3-37)
in which P is the symbol for power and t represents time. The rate of
change dc/dt of the angular displacement c is the angular speed v, and
therefore the preceding equation becomes
P ⫽ Tv
(v ⫽ rad/s)
(3-38)
This formula, which is familiar from elementary physics, gives the
power transmitted by a rotating shaft transmitting a constant torque T.
The units to be used in Eq. (3-38) are as follows. If the torque T is
expressed in newton meters, then the power is expressed in watts (W).
One watt is equal to one newton meter per second (or one joule per
second). If T is expressed in pound-feet, then the power is expressed in
foot-pounds per second.*
Angular speed is often expressed as the frequency f of rotation,
which is the number of revolutions per unit of time. The unit of
frequency is the hertz (Hz), equal to one revolution per second (s⫺1).
Inasmuch as one revolution equals 2p radians, we obtain
v ⫽ 2pf
(v ⫽ rad/s, f ⫽ Hz ⫽ s⫺1)
(3-39)
The expression for power (Eq. 3-38) then becomes
( f ⫽ Hz ⫽ s⫺1)
P ⫽ 2p f T
(3-40)
Another commonly used unit is the number of revolutions per minute
(rpm), denoted by the letter n. Therefore, we also have the following relationships:
n ⫽ 60 f
(3-41)
and
2p nT
P ⫽ ᎏᎏ
60
(n ⫽ rpm)
*See Table A-1, Appendix A, for units of work and power.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(3-42)
SECTION 3.7 Transmission of Power by Circular Shafts
219
In Eqs. (3-40) and (3-42), the quantities P and T have the same units as
in Eq. (3-38); that is, P has units of watts if T has units of newton
meters, and P has units of foot-pounds per second if T has units of
pound-feet.
In U.S. engineering practice, power is sometimes expressed in
horsepower (hp), a unit equal to 550 ft-lb/s. Therefore, the horsepower H
being transmitted by a rotating shaft is
2p nT
2p nT
H ⫽ ᎏᎏ ⫽ ᎏᎏ
60(550)
33,000
(n ⫽ rpm, T ⫽ lb-ft, H ⫽ hp)
(3-43)
One horsepower is approximately 746 watts.
The preceding equations relate the torque acting in a shaft to the
power transmitted by the shaft. Once the torque is known, we can determine the shear stresses, shear strains, angles of twist, and other desired
quantities by the methods described in Sections 3.2 through 3.5.
The following examples illustrate some of the procedures for
analyzing rotating shafts.
Example 3-7
A motor driving a solid circular steel shaft transmits 40 hp to a gear at B (Fig. 330). The allowable shear stress in the steel is 6000 psi.
(a) What is the required diameter d of the shaft if it is operated at 500 rpm?
(b) What is the required diameter d if it is operated at 3000 rpm?
Motor
d
FIG. 3-30 Example 3-7. Steel shaft in
ω
T
B
torsion
Solution
(a) Motor operating at 500 rpm. Knowing the horsepower and the speed of
rotation, we can find the torque T acting on the shaft by using Eq. (3-43).
Solving that equation for T, we get
33,000(40 hp)
33,000H
T ⫽ ᎏᎏ ⫽ ᎏᎏ ⫽ 420.2 lb-ft ⫽ 5042 lb-in.
2p n
2p(500 rpm)
This torque is transmitted by the shaft from the motor to the gear.
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
220
CHAPTER 3 Torsion
Motor
d
ω
FIG. 3-30 (Repeated)
T
B
The maximum shear stress in the shaft can be obtained from the modified
torsion formula (Eq. 3-12):
16T
tmax ⫽ ᎏᎏ3
pd
Solving that equation for the diameter d, and also substituting tallow for tmax, we
get
16(5042 lb-in.)
16 T
d 3 ⫽ ᎏᎏ ⫽ ᎏᎏ ⫽ 4.280 in.3
ptallow
p(6000 psi)
from which
d ⫽ 1.62 in.
The diameter of the shaft must be at least this large if the allowable shear stress
is not to be exceeded.
(b) Motor operating at 3000 rpm. Following the same procedure as in part
(a), we obtain
33,000H
33,000(40 hp)
T ⫽ ᎏᎏ ⫽ ᎏᎏ ⫽ 70.03 lb-ft ⫽ 840.3 lb-in.
2pn
2p (3000 rpm)
16(840.3 lb-in.)
16 T
d 3 ⫽ ᎏᎏ ⫽ ᎏᎏ ⫽ 0.7133 in.3
p tallow
p (6000 p si)
d ⫽ 0.89 in.
which is less than the diameter found in part (a).
This example illustrates that the higher the speed of rotation, the smaller the
required size of the shaft (for the same power and the same allowable stress).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
221
SECTION 3.7 Transmission of Power by Circular Shafts
Example 3-8
A solid steel shaft ABC of 50 mm diameter (Fig. 3-31a) is driven at A by a
motor that transmits 50 kW to the shaft at 10 Hz. The gears at B and C drive
machinery requiring power equal to 35 kW and 15 kW, respectively.
Compute the maximum shear stress tmax in the shaft and the angle of twist
fAC between the motor at A and the gear at C. (Use G ⫽ 80 GPa.)
1.0 m
1.2 m
TA = 796 N ⋅ m
Motor
A
B
C
A
TB = 557 N ⋅ m
TC = 239 N ⋅ m
B
C
50 mm
(a)
(b)
FIG. 3-31 Example 3-8. Steel shaft in
torsion
Solution
Torques acting on the shaft. We begin the analysis by determining the
torques applied to the shaft by the motor and the two gears. Since the motor
supplies 50 kW at 10 Hz, it creates a torque TA at end A of the shaft (Fig. 3-31b)
that we can calculate from Eq. (3-40):
P
50 kW
TA ⫽ ᎏᎏ ⫽ ᎏᎏ ⫽ 796 N⭈m
2pf
2p(10 Hz)
In a similar manner, we can calculate the torques TB and TC applied by the gears
to the shaft:
P
35 kW
TB ⫽ ᎏᎏ ⫽ ᎏᎏ ⫽ 557 N⭈m
2pf
2p (10 Hz)
P
15 kW
TC ⫽ ᎏᎏ ⫽ ᎏᎏ ⫽ 239 N⭈m
2pf
2p (10 Hz)
These torques are shown in the free-body diagram of the shaft (Fig. 3-31b).
Note that the torques applied by the gears are opposite in direction to the torque
applied by the motor. (If we think of TA as the “load” applied to the shaft by the
motor, then the torques TB and TC are the “reactions” of the gears.)
The internal torques in the two segments of the shaft are now found (by
inspection) from the free-body diagram of Fig. 3-31b:
TAB ⫽ 796 N⭈m
TBC ⫽ 239 N⭈m
Both internal torques act in the same direction, and therefore the angles of twist
in segments AB and BC are additive when finding the total angle of twist. (To be
specific, both torques are positive according to the sign convention adopted in
Section 3.4.)
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
222
CHAPTER 3 Torsion
Shear stresses and angles of twist. The shear stress and angle of twist in
segment AB of the shaft are found in the usual manner from Eqs. (3-12) and
(3-15):
16TA B
16(796 N⭈m )
⫽ 32.4 MPa
tAB ⫽ ᎏᎏ
3 ⫽ ᎏᎏ
pd
p(50 mm)3
(796 N⭈m)(1.0 m)
TAB LAB
fAB ⫽ ᎏ
ᎏ ⫽ ᎏᎏᎏ ⫽ 0.0162 rad
p
G IP
(80 GPa) ᎏᎏ (50 mm)4
32
冢 冣
The corresponding quantities for segment BC are
16TBC
16(239 N⭈m)
⫽ 9.7 MPa
tBC ⫽ ᎏᎏ
3 ⫽ ᎏᎏ
pd
p (50 mm)3
TBC LBC
(239 N⭈m)(1.2 m)
fBC ⫽ ᎏᎏ ⫽ ᎏᎏᎏ ⫽ 0.0058 rad
p
GIP
(80 GPa) ᎏᎏ (50 mm)4
32
冢 冣
Thus, the maximum shear stress in the shaft occurs in segment AB and is
tmax ⫽ 32.4 MPa
Also, the total angle of twist between the motor at A and the gear at C is
fAC ⫽ fAB ⫹ fBC ⫽ 0.0162 rad ⫹ 0.0058 rad ⫽ 0.0220 rad ⫽ 1.26°
As explained previously, both parts of the shaft twist in the same direction, and
therefore the angles of twist are added.
3.8 STATICALLY INDETERMINATE TORSIONAL MEMBERS
The bars and shafts described in the preceding sections of this chapter
are statically determinate because all internal torques and all reactions
can be obtained from free-body diagrams and equations of equilibrium.
However, if additional restraints, such as fixed supports, are added to the
bars, the equations of equilibrium will no longer be adequate for determining the torques. The bars are then classified as statically
indeterminate. Torsional members of this kind can be analyzed by
supplementing the equilibrium equations with compatibility equations
pertaining to the rotational displacements. Thus, the general method for
analyzing statically indeterminate torsional members is the same as
described in Section 2.4 for statically indeterminate bars with axial
loads.
The first step in the analysis is to write equations of equilibrium,
obtained from free-body diagrams of the given physical situation. The
unknown quantities in the equilibrium equations are torques, either
internal torques or reaction torques.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 3.8 Statically Indeterminate Torsional Members
A
B
T
(a)
Bar (1)
d1
d2
Tube (2)
(b)
Tube (2)
A
f
B
Bar (1)
T
End
plate
L
(c)
A
d1
B
Bar (1)
T1
(d)
A
d2
f2
B
T2
(a)
Because this equation contains two unknowns (T1 and T2), we recognize
that the composite bar is statically indeterminate.
To obtain a second equation, we must consider the rotational
displacements of both the solid bar and the tube. Let us denote the angle
of twist of the solid bar (Fig. 3-32d) by f1 and the angle of twist of the
tube by f 2 (Fig. 3-32e). These angles of twist must be equal because the
bar and tube are securely joined to the end plate and rotate with it;
consequently, the equation of compatibility is
(b)
f1 ⫽ f 2
The angles f1 and f2 are related to the torques T1 and T2 by the torquedisplacement relations, which in the case of linearly elastic materials are
obtained from the equation f ⫽ TL /GIP. Thus,
Tube (2)
(e)
FIG. 3-32 Statically indeterminate bar in
torsion
The second step in the analysis is to formulate equations of
compatibility, based upon physical conditions pertaining to the angles
of twist. As a consequence, the compatibility equations contain angles of
twist as unknowns.
The third step is to relate the angles of twist to the torques by
torque-displacement relations, such as f ⫽ TL /GIP. After introducing
these relations into the compatibility equations, they too become equations
containing torques as unknowns. Therefore, the last step is to obtain the
unknown torques by solving simultaneously the equations of equilibrium
and compatibility.
To illustrate the method of solution, we will analyze the composite
bar AB shown in Fig. 3-32a. The bar is attached to a fixed support at end
A and loaded by a torque T at end B. Furthermore, the bar consists of
two parts: a solid bar and a tube (Figs. 3-32b and c), with both the solid
bar and the tube joined to a rigid end plate at B.
For convenience, we will identify the solid bar and tube (and their
properties) by the numerals 1 and 2, respectively. For instance, the diameter of the solid bar is denoted d1 and the outer diameter of the tube is
denoted d2. A small gap exists between the bar and the tube, and therefore the inner diameter of the tube is slightly larger than the diameter d1
of the bar.
When the torque T is applied to the composite bar, the end plate
rotates through a small angle f (Fig. 3-32c) and torques T1 and T2 are
developed in the solid bar and the tube, respectively (Figs. 3-32d and e).
From equilibrium we know that the sum of these torques equals the
applied load, and so the equation of equilibrium is
T1 ⫹ T2 ⫽ T
f1
223
TL
f1 ⫽ ᎏ1ᎏ
G1IP1
T L
f2 ⫽ ᎏ2ᎏ
G2 IP2
(c,d)
in which G1 and G2 are the shear moduli of elasticity of the materials
and IP1 and IP2 are the polar moments of inertia of the cross sections.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
224
CHAPTER 3 Torsion
When the preceding expressions for f1 and f2 are substituted into
Eq. (b), the equation of compatibility becomes
T1L
T2L
ᎏ ᎏ ⫽ ᎏᎏ
G1IP1
G2IP2
(e)
We now have two equations (Eqs. a and e) with two unknowns, so we
can solve them for the torques T1 and T2. The results are
冢 G I G⫹I G I 冣
P1
T1 ⫽ T ᎏ1ᎏ
1 P1
2 P2
冢 G I G⫹I G I 冣
P2
T2 ⫽ T ᎏ2ᎏ
1 P1
(3-44a,b)
2 P2
With these torques known, the essential part of the statically indeterminate
analysis is completed. All other quantities, such as stresses and angles of
twist, can now be found from the torques.
The preceding discussion illustrates the general methodology for
analyzing a statically indeterminate system in torsion. In the following
example, this same approach is used to analyze a bar that is fixed against
rotation at both ends. In the example and in the problems, we assume that
the bars are made of linearly elastic materials. However, the general
methodology is also applicable to bars of nonlinear materials—the only
change is in the torque-displacement relations.
Example 3-9
The bar ACB shown in Figs. 3-33a and b is fixed at both ends and loaded by a
torque T0 at point C. Segments AC and CB of the bar have diameters dA and dB,
lengths LA and LB, and polar moments of inertia IPA and IPB, respectively. The
material of the bar is the same throughout both segments.
Obtain formulas for (a) the reactive torques TA and TB at the ends, (b) the
maximum shear stresses tAC and tCB in each segment of the bar, and (c) the
angle of rotation fC at the cross section where the load T0 is applied.
Solution
Equation of equilibrium. The load T0 produces reactions TA and TB at the
fixed ends of the bar, as shown in Figs. 3-33a and b. Thus, from the equilibrium
of the bar we obtain
TA ⫹ TB ⫽ T0
(f)
Because there are two unknowns in this equation (and no other useful equations
of equilibrium), the bar is statically indeterminate.
Equation of compatibility. We now separate the bar from its support at end B
and obtain a bar that is fixed at end A and free at end B (Figs. 3-33c and d). When
the load T0 acts alone (Fig. 3-33c), it produces an angle of twist at end B that we
denote as f1. Similarly, when the reactive torque TB acts alone, it produces an
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
225
SECTION 3.8 Statically Indeterminate Torsional Members
TA
dA
A
angle f2 (Fig. 3-33d). The angle of twist at end B in the original bar, equal to the
sum of f1 and f2, is zero. Therefore, the equation of compatibility is
dB
C
B
T0
TB
(a)
TA
A
IPA
C
IPB B
(b)
C
f1
B
(h,i)
or
TBLA
T LB
T0 LA
⫽ ᎏ
ᎏ ⫹ ᎏBᎏ
IPB
(c)
C
TB LA
TB LB
f2 ⫽ ⫺ᎏ
ᎏ⫺ᎏ
ᎏ
GIPA
GIPB
T LA
T LA
T LB
ᎏ0 ᎏ
⫺ ᎏBᎏ
⫺ ᎏBᎏ
⫽0
GIPA
GIPA
GIPB
T0
A
T LA
f1 ⫽ ᎏ0 ᎏ
GIPA
The minus signs appear in Eq. (i) because TB produces a rotation that is opposite
in direction to the positive direction of f2 (Fig. 3-33d).
We now substitute the angles of twist (Eqs. h and i) into the compatibility
equation (Eq. g) and obtain
LB
L
A
(g)
Note that f1 and f2 are assumed to be positive in the direction shown in the
figure.
Torque-displacement equations. The angles of twist f1 and f2 can be
expressed in terms of the torques T0 and TB by referring to Figs. 3-33c and d
and using the equation f ⫽ TL /GIP. The equations are as follows:
TB
T0
LA
f1 f2 ⫽ 0
f2
B
TB
Solution of equations. The preceding equation can be solved for the torque
TB, which then can be substituted into the equation of equilibrium (Eq. f) to
obtain the torque TA. The results are
冢
LBIPA
TA ⫽ T0 ᎏᎏ
(d)
FIG. 3-33 Example 3-9. Statically
indeterminate bar in torsion
(j)
冣
冢
LAIPB
TB ⫽ T0 ᎏᎏ
冣
(3-45a,b)
Thus, the reactive torques at the ends of the bar have been found, and the statically
indeterminate part of the analysis is completed.
As a special case, note that if the bar is prismatic (IPA ⫽ IPB ⫽ IP) the
preceding results simplify to
T0LB
TA ⫽ ᎏ
T0LA
TB ⫽ ᎏ
(3-46a,b)
where L is the total length of the bar. These equations are analogous to those for
the reactions of an axially loaded bar with fixed ends (see Eqs. 2-9a and 2-9b).
Maximum shear stresses. The maximum shear stresses in each part of the
bar are obtained directly from the torsion formula:
T d
tAC ⫽ ᎏAᎏA
2 IPA
TB dB
tCB ⫽ ᎏ
ᎏ
2IPB
Substituting from Eqs. (3-45a) and (3-45b) gives
T0 LB dA
tAC ⫽ ᎏᎏ
T0 LAdB
tCB ⫽ ᎏᎏ
(3-47a,b)
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
226
CHAPTER 3 Torsion
By comparing the product LB dA with the product LAdB, we can immediately
determine which segment of the bar has the larger stress.
Angle of rotation. The angle of rotation fC at section C is equal to the
angle of twist of either segment of the bar, since both segments rotate through
the same angle at section C. Therefore, we obtain
TALA
TB LB
T0LA LB
fC ⫽ ᎏ ⫽ ᎏ ⫽ ᎏᎏ
(3-48)
In the special case of a prismatic bar (IPA ⫽ IPB ⫽ IP), the angle of rotation at
the section where the load is applied is
T0LALB
fC ⫽ ᎏ
(3-49)
This example illustrates not only the analysis of a statically indeterminate
bar but also the techniques for finding stresses and angles of rotation. In addition, note that the results obtained in this example are valid for a bar consisting
of either solid or tubular segments.
3.9 STRAIN ENERGY IN TORSION AND PURE SHEAR
A
f
L
B
T
FIG. 3-34 Prismatic bar in pure torsion
When a load is applied to a structure, work is performed by the load and
strain energy is developed in the structure, as described in detail in Section 2.7 for a bar subjected to axial loads. In this section we will use the
same basic concepts to determine the strain energy of a bar in torsion.
Consider a prismatic bar AB in pure torsion under the action of a
torque T (Fig. 3-34). When the load is applied statically, the bar twists
and the free end rotates through an angle f. If we assume that the material
of the bar is linearly elastic and follows Hooke’s law, then the relationship
between the applied torque and the angle of twist will also be linear, as
shown by the torque-rotation diagram of Fig. 3-35 and as given by the
equation f ⫽ TL /GIP.
The work W done by the torque as it rotates through the angle f is
equal to the area below the torque-rotation line OA, that is, it is equal to the
area of the shaded triangle in Fig. 3-35. Furthermore, from the principle
of conservation of energy we know that the strain energy of the bar is
equal to the work done by the load, provided no energy is gained or lost
in the form of heat. Therefore, we obtain the following equation for the
strain energy U of the bar:
Tf
U ⫽ W ⫽ ᎏᎏ
2
(3-50)
This equation is analogous to the equation U ⫽ W ⫽ Pd/2 for a bar
subjected to an axial load (see Eq. 2-35).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 3.9 Strain Energy in Torsion and Pure Shear
Using the equation f ⫽ TL /GIP, we can express the strain energy in
the following forms:
Torque
A
T
U=W=
O
227
Tf
2
f
Angle of rotation
FIG. 3-35 Torque-rotation diagram for a
bar in pure torsion (linearly elastic
material)
T 2L
U ⫽ ᎏᎏ
2G IP
GI f 2
U ⫽ ᎏPᎏ
2L
(3-51a,b)
The first expression is in terms of the load and the second is in terms of
the angle of twist. Again, note the analogy with the corresponding equations
for a bar with an axial load (see Eqs. 2-37a and b).
The SI unit for both work and energy is the joule (J), which is equal
to one newton meter (1 J ⫽ 1 N⭈m). The basic USCS unit is the footpound (ft-lb), but other similar units, such as inch-pound (in.-lb) and
inch-kip (in.-k), are commonly used.
Nonuniform Torsion
If a bar is subjected to nonuniform torsion (described in Section 3.4), we
need additional formulas for the strain energy. In those cases where the
bar consists of prismatic segments with constant torque in each segment
(see Fig. 3-14a of Section 3.4), we can determine the strain energy of
each segment and then add to obtain the total energy of the bar:
n
U ⫽ 冱 Ui
(3-52)
i⫽1
in which Ui is the strain energy of segment i and n is the number of
segments. For instance, if we use Eq. (3-51a) to obtain the individual
strain energies, the preceding equation becomes
n
T i2Li
U⫽冱 ᎏ
(3-53)
i⫽1
in which Ti is the internal torque in segment i and Li, Gi, and (IP)i are the
torsional properties of the segment.
If either the cross section of the bar or the internal torque varies
along the axis, as illustrated in Figs. 3-15 and 3-16 of Section 3.4, we
can obtain the total strain energy by first determining the strain energy
of an element and then integrating along the axis. For an element of
length dx, the strain energy is (see Eq. 3-51a)
[T(x) ]2dx
dU ⫽ ᎏᎏ
2GIP( x)
in which T(x) is the internal torque acting on the element and IP(x) is the
polar moment of inertia of the cross section at the element. Therefore,
the total strain energy of the bar is
[T(x) ] dx
ᎏ
冕ᎏ
2GI ( x)
L
U⫽
0
2
P
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(3-54)
228
CHAPTER 3 Torsion
Once again, the similarities of the expressions for strain energy in
torsion and axial load should be noted (compare Eqs. 3-53 and 3-54 with
Eqs. 2-40 and 2-41 of Section 2.7).
The use of the preceding equations for nonuniform torsion is illustrated in the examples that follow. In Example 3-10 the strain energy is
found for a bar in pure torsion with prismatic segments, and in Examples
3-11 and 3-12 the strain energy is found for bars with varying torques
and varying cross-sectional dimensions.
In addition, Example 3-12 shows how, under very limited conditions,
the angle of twist of a bar can be determined from its strain energy. (For
a more detailed discussion of this method, including its limitations, see
the subsection “Displacements Caused by a Single Load” in Section 2.7.)
Limitations
When evaluating strain energy we must keep in mind that the equations
derived in this section apply only to bars of linearly elastic materials with
small angles of twist. Also, we must remember the important observation
stated previously in Section 2.7, namely, the strain energy of a structure
supporting more than one load cannot be obtained by adding the strain
energies obtained for the individual loads acting separately. This
observation is demonstrated in Example 3-10.
Strain-Energy Density in Pure Shear
Because the individual elements of a bar in torsion are stressed in pure
shear, it is useful to obtain expressions for the strain energy associated with
the shear stresses. We begin the analysis by considering a small element of
material subjected to shear stresses t on its side faces (Fig. 3-36a). For
convenience, we will assume that the front face of the element is square,
with each side having length h. Although the figure shows only a twodimensional view of the element, we recognize that the element is actually
three dimensional with thickness t perpendicular to the plane of the figure.
Under the action of the shear stresses, the element is distorted so that
the front face becomes a rhombus, as shown in Fig. 3-36b. The change in
angle at each corner of the element is the shear strain g.
The shear forces V acting on the side faces of the element (Fig. 3-36c)
are found by multiplying the stresses by the areas ht over which they act:
V ⫽ t ht
(a)
These forces produce work as the element deforms from its initial shape
(Fig. 3-36a) to its distorted shape (Fig. 3-36b). To calculate this work we
need to determine the relative distances through which the shear forces
move. This task is made easier if the element in Fig. 3-36c is rotated as a
rigid body until two of its faces are horizontal, as in Fig. 3-36d. During
the rigid-body rotation, the net work done by the forces V is zero
because the forces occur in pairs that form two equal and opposite
couples.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 3.9 Strain Energy in Torsion and Pure Shear
t
229
t
t
t
h
–p– – g
2
t
t
t
h
t
(a)
(b)
V
V
d
V
–p– – g
2
V
g
V
FIG. 3-36 Element in pure shear
V
V
V
(c)
(d)
As can be seen in Fig. 3-36d, the top face of the element is
displaced horizontally through a distance d (relative to the bottom face)
as the shear force is gradually increased from zero to its final value V.
The displacement d is equal to the product of the shear strain g (which is
a small angle) and the vertical dimension of the element:
d ⫽ gh
(b)
If we assume that the material is linearly elastic and follows Hooke’s
law, then the work done by the forces V is equal to Vd/2, which is also
the strain energy stored in the element:
Vd
U ⫽ W ⫽ ᎏᎏ
2
(c)
Note that the forces acting on the side faces of the element (Fig. 3-36d)
do not move along their lines of action—hence they do no work.
Substituting from Eqs. (a) and (b) into Eq. (c), we get the total strain
energy of the element:
tgh2t
U⫽ ᎏ
Because the volume of the element is h2t, the strain-energy density u
(that is, the strain energy per unit volume) is
tg
u ⫽ ᎏᎏ
2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(d)
230
CHAPTER 3 Torsion
Finally, we substitute Hooke’s law in shear (t ⫽ Gg) and obtain the
following equations for the strain-energy density in pure shear:
t2
u ⫽ ᎏᎏ
2G
Gg 2
u ⫽ ᎏᎏ
2
(3-55a,b)
These equations are similar in form to those for uniaxial stress (see
Eqs. 2-44a and b of Section 2.7).
The SI unit for strain-energy density is joule per cubic meter (J/m3),
and the USCS unit is inch-pound per cubic inch (or other similar units).
Since these units are the same as those for stress, we may also express
strain-energy density in pascals (Pa) or pounds per square inch (psi).
In the next section (Section 3.10) we will use the equation for
strain-energy density in terms of the shear stress (Eq. 3-55a) to determine the angle of twist of a thin-walled tube of arbitrary cross-sectional
shape.
Example 3-10
A
B
Ta
A solid circular bar AB of length L is fixed at one end and free at the other
(Fig. 3-37). Three different loading conditions are to be considered: (a) torque
Ta acting at the free end; (b) torque Tb acting at the midpoint of the bar; and
(c) torques Ta and Tb acting simultaneously.
For each case of loading, obtain a formula for the strain energy stored in
the bar. Then evaluate the strain energy for the following data: Ta ⫽ 100 N⭈m,
Tb ⫽ 150 N⭈m, L ⫽ 1.6 m, G ⫽ 80 GPa, and IP ⫽ 79.52 ⫻ 103 mm4.
L
Solution
(a)
(a) Torque Ta acting at the free end (Fig. 3-37a). In this case the strain
energy is obtained directly from Eq. (3-51a):
C Tb
A
T 2L
Ua ⫽ ᎏaᎏ
2G IP
B
L
—
2
(b) Torque Tb acting at the midpoint (Fig. 3-37b). When the torque acts at
the midpoint, we apply Eq. (3-51a) to segment AC of the bar:
(b)
T b2(L/ 2)
T 2L
ᎏ ⫽ ᎏbᎏ
Ub ⫽ ᎏ
2 GIP
4G IP
C Tb
A
(e)
L
—
2
B
Ta
L
—
2
(c) Torques Ta and Tb acting simultaneously (Fig. 3-37c). When both loads
act on the bar, the torque in segment CB is Ta and the torque in segment AC is
Ta ⫹ Tb. Thus, the strain energy (from Eq. 3-53) is
n
(Ta ⫹ Tb)2(L/2)
T 2L i
T a2(L/ 2)
Uc ⫽ 冱 ᎏiᎏ
⫽ᎏ
ᎏ ⫹ ᎏᎏ
2 GIP
i⫽1 2G (IP )i
(c)
FIG. 3-37 Example 3-10. Strain energy
produced by two loads
(f)
TaTbL
T a2L
T b2L
⫽ᎏ⫹ᎏ⫹ᎏ
2GIP
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(g)
SECTION 3.9 Strain Energy in Torsion and Pure Shear
231
A comparison of Eqs. (e), (f), and (g) shows that the strain energy produced by
the two loads acting simultaneously is not equal to the sum of the strain energies
produced by the loads acting separately. As pointed out in Section 2.7, the
reason is that strain energy is a quadratic function of the loads, not a linear
function.
(d) Numerical results. Substituting the given data into Eq. (e), we obtain
(100 N⭈m)2(1.6 m)
T 2L
⫽ 1.26 J
Ua ⫽ ᎏaᎏ ⫽ ᎏᎏᎏᎏ
2(80 GPa)(79.52 ⫻ 103 mm4)
2G IP
Recall that one joule is equal to one newton meter (1 J ⫽ 1 N⭈m).
Proceeding in the same manner for Eqs. (f) and (g), we find
Ub ⫽ 1.41 J
Uc ⫽ 1.26 J ⫹ 1.89 J ⫹ 1.41 J ⫽ 4.56 J
Note that the middle term, involving the product of the two loads, contributes
significantly to the strain energy and cannot be disregarded.
Example 3-11
A prismatic bar AB, fixed at one end and free at the other, is loaded by a distributed torque of constant intensity t per unit distance along the axis of the bar
(Fig. 3-38).
(a) Derive a formula for the strain energy of the bar.
(b) Evaluate the strain energy of a hollow shaft used for drilling into the
earth if the data are as follows:
t ⫽ 480 lb-in./in., L ⫽ 12 ft, G ⫽ 11.5 ⫻ 106 psi, and IP ⫽ 17.18 in.4
t
A
FIG. 3-38 Example 3-11. Strain energy
dx
produced by a distributed torque
B
x
L
Solution
(a) Strain energy of the bar. The first step in the solution is to determine
the internal torque T(x) acting at distance x from the free end of the bar (Fig.
3-38). This internal torque is equal to the total torque acting on the part of the
bar between x ⫽ 0 and x ⫽ x. This latter torque is equal to the intensity t of
torque times the distance x over whhich it acts:
T(x) ⫽ tx
(h)
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
232
CHAPTER 3 Torsion
Substituting into Eq. (3-54), we obtain
1
[T(x)] dx
t L
ᎏ ⫽ ᎏ ᎏ 冕 (tx) dx ⫽ ᎏᎏ
冕ᎏ
2GI
6GI
2GI
L
U⫽
L
2
P
0
2 3
2
P
(3-56)
P
0
This expression gives the total strain energy stored in the bar.
(b) Numerical results. To evaluate the strain energy of the hollow shaft, we
substitute the given data into Eq. (3-56):
(480 lb-in./in.)2(144 in.)3
t 2L3
⫽ 580 in.-lb
U ⫽ ᎏᎏ ⫽ ᎏᎏᎏ
6(11.5 ⫻ 106 psi)(17.18 in.4)
6GIP
This example illustrates the use of integration to evaluate the strain energy of a
bar subjected to a distributed torque.
Example 3-12
A tapered bar AB of solid circular cross section is supported at the right-hand
end and loaded by a torque T at the other end (Fig. 3-39). The diameter of the
bar varies linearly from dA at the left-hand end to dB at the right-hand end.
Determine the angle of rotation fA at end A of the bar by equating the
strain energy to the work done by the load.
A
T
dA
FIG. 3-39 Example 3-12. Tapered bar in
fA
B
d(x)
x
dB
dx
L
torsion
Solution
From the principle of conservation of energy we know that the work done
by the applied torque equals the strain energy of the bar; thus, W ⫽ U. The work
is given by the equation
TfA
W ⫽ ᎏᎏ
2
(i)
and the strain energy U can be found from Eq. (3-54).
To use Eq. (3-54), we need expressions for the torque T(x) and the polar
moment of inertia IP(x). The torque is constant along the axis of the bar and
equal to the load T, and the polar moment of inertia is
p
IP(x) ⫽ ᎏᎏ 冤d(x)冥4
32
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 3.9 Strain Energy in Torsion and Pure Shear
233
in which d(x) is the diameter of the bar at distance x from end A. From the
geometry of the figure, we see that
dB ⫺ dA
ᎏx
d(x) ⫽ dA ⫹ ᎏ
L
(j)
and therefore
冢
冣
p
dB ⫺ dA
IP(x) ⫽ ᎏᎏ dA ⫹ ᎏ
ᎏx
32
L
4
(k)
Now we can substitute into Eq. (3-54), as follows:
dx
[T(x) ] dx
16T
ᎏ ⫽ ᎏᎏ 冕 ᎏᎏ
冕ᎏ
2GI ( x)
pG
d ⫺d
L
U⫽
0
2
2
P
L
冢d
0
A
B
A
⫹ᎏ
ᎏx
L
冣
4
The integral in this expression can be integrated with the aid of a table of
integrals (see Appendix C). However, we already evaluated this integral in
Example 3-5 of Section 3.4 (see Eq. g of that example) and found that
L
1
1
dx
⫽ ᎏᎏ 冢ᎏᎏ ⫺ ᎏᎏ冣
冕 ᎏᎏ
3(d ⫺ d ) d
d
d ⫺d
L
0
冢
B
A
dA ⫹ ᎏ
ᎏx
L
冣
4
B
3
A
A
3
B
Therefore, the strain energy of the tapered bar is
冢
冣
1
1
16T 2L
U ⫽ ᎏᎏ ᎏᎏ
⫺ ᎏᎏ
3pG(dB ⫺ dA) d 3A d B3
(3-57)
Equating the strain energy to the work of the torque (Eq. i) and solving for
fA, we get
冢
冣
1
1
32TL
fA ⫽ ᎏᎏ ᎏᎏ
⫺ ᎏᎏ
3pG(dB ⫺ dA) d 3A d B3
(3-58)
This equation, which is the same as Eq. (3-26) in Example 3-5 of Section 3.4,
gives the angle of rotation at end A of the tapered bar.
Note especially that the method used in this example for finding the angle of
rotation is suitable only when the bar is subjected to a single load, and then only
when the desired angle corresponds to that load. Otherwise, we must find angular
displacements by the usual methods described in Sections 3.3, 3.4, and 3.8.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
234
CHAPTER 3 Torsion
3.10 THIN-WALLED TUBES
The torsion theory described in the preceding sections is applicable to
solid or hollow bars of circular cross section. Circular shapes are the
most efficient shapes for resisting torsion and consequently are the most
commonly used. However, in lightweight structures, such as aircraft and
spacecraft, thin-walled tubular members with noncircular cross sections
are often required to resist torsion. In this section, we will analyze structural members of this kind.
To obtain formulas that are applicable to a variety of shapes, let us
consider a thin-walled tube of arbitrary cross section (Fig. 3-40a). The
tube is cylindrical in shape—that is, all cross sections are identical and
the longitudinal axis is a straight line. The thickness t of the wall is not
necessarily constant but may vary around the cross section. However,
the thickness must be small in comparison with the total width of the
tube. The tube is subjected to pure torsion by torques T acting at the
ends.
Shear Stresses and Shear Flow
The shear stresses t acting on a cross section of the tube are pictured in
Fig. 3-40b, which shows an element of the tube cut out between two
y
t
T
a
d
O
z
x
b
c
T
x
dx
L
(a)
t
T
a
d
b
c
tb
T
tc
a
b
d
arbitrary cross-sectional shape
tb
c
a
F1
(c)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
F1
c
Fc
dx
(b)
tb
b
d
tc
tc
FIG. 3-40 Thin-walled tube of
Fb
tb
(d)
tc
SECTION 3.10 Thin-Walled Tubes
235
cross sections that are distance dx apart. The stresses act parallel to the
boundaries of the cross section and “flow” around the cross section.
Also, the intensity of the stresses varies so slightly across the thickness
of the tube (because the tube is assumed to be thin) that we may assume
t to be constant in that direction. However, if the thickness t is not
constant, the stresses will vary in intensity as we go around the cross
section, and the manner in which they vary must be determined from
equilibrium.
To determine the magnitude of the shear stresses, we will consider a
rectangular element abcd obtained by making two longitudinal cuts ab
and cd (Figs. 3-40a and b). This element is isolated as a free body in
Fig. 3-40c. Acting on the cross-sectional face bc are the shear stresses t
shown in Fig. 3-40b. We assume that these stresses vary in intensity as
we move along the cross section from b to c; therefore, the shear stress
at b is denoted tb and the stress at c is denoted tc (see Fig. 3-40c).
As we know from equilibrium, identical shear stresses act in the
opposite direction on the opposite cross-sectional face ad, and shear
stresses of the same magnitude also act on the longitudinal faces ab and
cd. Thus, the constant shear stresses acting on faces ab and cd are equal
to tb and tc, respectively.
The stresses acting on the longitudinal faces ab and cd produce
forces Fb and Fc (Fig. 3-40d). These forces are obtained by multiplying
the stresses by the areas on which they act:
Fb ⫽ tbtb dx
Fc ⫽ tc tc dx
in which tb and tc represent the thicknesses of the tube at points b and c,
respectively (Fig. 3-40d).
In addition, forces F1 are produced by the stresses acting on faces
bc and ad. From the equilibrium of the element in the longitudinal direction (the x direction), we see that Fb ⫽ Fc, or
tb tb ⫽ tc tc
Because the locations of the longitudinal cuts ab and cd were selected
arbitrarily, it follows from the preceding equation that the product of the
shear stress t and the thickness t of the tube is the same at every point in
the cross section. This product is known as the shear flow and is
denoted by the letter f:
f ⫽ t t ⫽ constant
(3-59)
This relationship shows that the largest shear stress occurs where the
thickness of the tube is smallest, and vice versa. In regions where the
thickness is constant, the shear stress is constant. Note that shear flow is
the shear force per unit distance along the cross section.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
236
CHAPTER 3 Torsion
Torsion Formula for Thin-Walled Tubes
The next step in the analysis is to relate the shear flow f (and hence the
shear stress t) to the torque T acting on the tube. For that purpose, let us
examine the cross section of the tube, as pictured in Fig. 3-41. The
median line (also called the centerline or the midline) of the wall of the
tube is shown as a dashed line in the figure. We consider an element of
area of length ds (measured along the median line) and thickness t. The
distance s defining the location of the element is measured along the
median line from some arbitrarily chosen reference point.
The total shear force acting on the element of area is fds, and the
moment of this force about any point O within the tube is
s
ds
t
f ds
r
O
dT ⫽ rfds
FIG. 3-41 Cross section of thin-walled
tube
in which r is the perpendicular distance from point O to the line of
action of the force fds. (Note that the line of action of the force fds is
tangent to the median line of the cross section at the element ds.) The
total torque T produced by the shear stresses is obtained by integrating
along the median line of the cross section:
T⫽f
冕
Lm
(a)
r ds
0
in which Lm denotes the length of the median line.
The integral in Eq. (a) can be difficult to integrate by formal mathematical means, but fortunately it can be evaluated easily by giving it a
simple geometric interpretation. The quantity rds represents twice the
area of the shaded triangle shown in Fig. 3-41. (Note that the triangle
has base length ds and height equal to r.) Therefore, the integral represents twice the area Am enclosed by the median line of the cross section:
冕
Lm
r ds ⫽ 2Am
(b)
0
It follows from Eq. (a) that T ⫽ 2fAm, and therefore the shear flow is
T
f ⫽ ᎏᎏ
2Am
(3-60)
Now we can eliminate the shear flow f between Eqs. (3-59) and (3-60)
and obtain a torsion formula for thin-walled tubes:
T
t ⫽ ᎏᎏ
2tAm
(3-61)
Since t and Am are properties of the cross section, the shear stresses t
can be calculated from Eq. (3-61) for any thin-walled tube subjected to a
known torque T. (Reminder: The area Am is the area enclosed by the
median line—it is not the cross-sectional area of the tube.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 3.10 Thin-Walled Tubes
To illustrate the use of the torsion formula, consider a thin-walled
circular tube (Fig. 3-42) of thickness t and radius r to the median line.
The area enclosed by the median line is
t
r
Am ⫽ p r 2
FIG. 3-42 Thin-walled circular tube
(3-62)
and therefore the shear stress (constant around the cross section) is
T
t ⫽ ᎏᎏ
(3-63)
2p r 2t
This formula agrees with the stress obtained from the standard torsion
formula (Eq. 3-11) when the standard formula is applied to a circular
tube with thin walls using the approximate expression IP ⬇ 2pr3t for the
polar moment of inertia (Eq. 3-18).
As a second illustration, consider a thin-walled rectangular tube
(Fig. 3-43) having thickness t1 on the sides and thickness t2 on the top
and bottom. Also, the height and width (measured to the median line of the
cross section) are h and b, respectively. The area within the median line is
(3-64)
Am ⫽ bh
t2
t1
237
t1
h
and thus the shear stresses in the vertical and horizontal sides, respectively,
are
T
T
thoriz ⫽ ᎏᎏ
(3-65a,b)
tvert ⫽ ᎏᎏ
2t1bh
2t2bh
If t2 is larger than t1, the maximum shear stress will occur in the vertical
sides of the cross section.
Strain Energy and Torsion Constant
t2
b
FIG. 3-43 Thin-walled rectangular tube
The strain energy of a thin-walled tube can be determined by first
finding the strain energy of an element and then integrating throughout
the volume of the bar. Consider an element of the tube having area t ds
in the cross section (see the element in Fig. 3-41) and length dx (see the
element in Fig. 3-40). The volume of such an element, which is similar
in shape to the element abcd shown in Fig. 3-40a, is t ds dx. Because
elements of the tube are in pure shear, the strain-energy density of the
element is t 2/2G, as given by Eq. (3-55a). The total strain energy of the
element is equal to the strain-energy density times the volume:
t2
t 2t 2 ds
f 2 ds
dU ⫽ ᎏᎏ t ds dx ⫽ ᎏᎏ ᎏᎏ dx ⫽ ᎏᎏ ᎏᎏ dx
2G
2G t
2G t
(c)
in which we have replaced t t by the shear flow f (a constant).
The total strain energy of the tube is obtained by integrating dU
throughout the volume of the tube, that is, ds is integrated from 0 to Lm
around the median line and dx is integrated along the axis of the tube
from 0 to L, where L is the length. Thus,
冕 dU ⫽ ᎏ2fGᎏ 冕
2
U⫽
Lm
0
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
ds
ᎏᎏ
t
冕 dx
L
0
(d)
238
CHAPTER 3 Torsion
Note that the thickness t may vary around the median line and must
remain with ds under the integral sign. Since the last integral is equal to
the length L of the tube, the equation for the strain energy becomes
f 2L
U ⫽ ᎏᎏ
2G
冕
Lm
0
ds
ᎏᎏ
t
(e)
Substituting for the shear flow from Eq. (3-60), we obtain
T 2L
U⫽ ᎏ
冕
Lm
0
ds
ᎏᎏ
t
(3-66)
as the equation for the strain energy of the tube in terms of the torque T.
The preceding expression for strain energy can be written in simpler
form by introducing a new property of the cross section, called the
torsion constant. For a thin-walled tube, the torsion constant (denoted
by the letter J) is defined as follows:
4A2m
J⫽ ᎏ
Lm
ds
ᎏᎏ
t
0
冕
(3-67)
With this notation, the equation for strain energy (Eq. 3-66) becomes
T 2L
U ⫽ ᎏᎏ
2G J
(3-68)
which has the same form as the equation for strain energy in a circular
bar (see Eq. 3-51a). The only difference is that the torsion constant J has
replaced the polar moment of inertia IP. Note that the torsion constant
has units of length to the fourth power.
In the special case of a cross section having constant thickness t, the
expression for J (Eq. 3-67) simplifies to
t
r
FIG. 3-42 (Repeated)
4tA2m
J ⫽ ᎏᎏ
Lm
(3-69)
For each shape of cross section, we can evaluate J from either Eq. (3-67)
or Eq. (3-69).
As an illustration, consider again the thin-walled circular tube of
Fig. 3-42. Since the thickness is constant we use Eq. (3-69) and substitute Lm ⫽ 2p r and Am ⫽ p r 2; the result is
J ⫽ 2p r 3t
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(3-70)
SECTION 3.10 Thin-Walled Tubes
t2
t1
t1
h
which is the approximate expression for the polar moment of inertia (Eq.
3-18). Thus, in the case of a thin-walled circular tube, the polar moment
of inertia is the same as the torsion constant.
As a second illustration, we will use the rectangular tube of Fig. 3-43.
For this cross section we have Am ⫽ bh. Also, the integral in Eq. (3-67) is
冕
Lm
0
t2
b
239
ds
ᎏᎏ ⫽ 2
t
冕 ᎏdtᎏs ⫹ 2 冕 ᎏdtᎏs ⫽ 2冢ᎏthᎏ ⫹ ᎏtbᎏ冣
h
0
b
1
0
2
1
2
Thus, the torsion constant (Eq. 3-67) is
2b2h2t1t2
J⫽ ᎏ
FIG. 3-43 (Repeated)
(3-71)
Torsion constants for other thin-walled cross sections can be found in a
similar manner.
Angle of Twist
The angle of twist f for a thin-walled tube of arbitrary cross-sectional
shape (Fig. 3-44) may be determined by equating the work W done by
the applied torque T to the strain energy U of the tube. Thus,
W⫽U
or
Tf
T 2L
ᎏᎏ ⫽ ᎏᎏ
2
2G J
from which we get the equation for the angle of twist:
TL
f ⫽ ᎏᎏ
GJ
(3-72)
Again we observe that the equation has the same form as the corresponding equation for a circular bar (Eq. 3-15) but with the polar moment of
inertia replaced by the torsion constant. The quantity GJ is called the
torsional rigidity of the tube.
f
FIG. 3-44 Angle of twist f for a thin-
walled tube
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
T
240
CHAPTER 3 Torsion
Limitations
The formulas developed in this section apply to prismatic members
having closed tubular shapes with thin walls. If the cross section is thin
walled but open, as in the case of I-beams and channel sections, the
theory given here does not apply. To emphasize this point, imagine that
we take a thin-walled tube and slit it lengthwise—then the cross section
becomes an open section, the shear stresses and angles of twist increase,
the torsional resistance decreases, and the formulas given in this section
cannot be used.
Some of the formulas given in this section are restricted to linearly
elastic materials—for instance, any equation containing the shear
modulus of elasticity G is in this category. However, the equations for
shear flow and shear stress (Eqs. 3-60 and 3-61) are based only upon
equilibrium and are valid regardless of the material properties. The
entire theory is approximate because it is based upon centerline dimensions, and the results become less accurate as the wall thickness t
increases.*
An important consideration in the design of any thin-walled
member is the possibility that the walls will buckle. The thinner the
walls and the longer the tube, the more likely it is that buckling will
occur. In the case of noncircular tubes, stiffeners and diaphragms are
often used to maintain the shape of the tube and prevent localized buckling.
In all of our discussions and problems, we assume that buckling is
prevented.
*The torsion theory for thin-walled tubes described in this section was developed by
R. Bredt, a German engineer who presented it in 1896 (Ref. 3-2). It is often called
Bredt’s theory of torsion.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 3.10 Thin-Walled Tubes
241
Example 3-13
Compare the maximum shear stress in a circular tube (Fig. 3-45) as calculated
by the approximate theory for a thin-walled tube with the stress calculated by
the exact torsion theory. (Note that the tube has constant thickness t and radius r
to the median line of the cross section.)
t
r
Solution
Approximate theory. The shear stress obtained from the approximate
theory for a thin-walled tube (Eq. 3-63) is
FIG. 3-45 Example 3-13. Comparison of
approximate and exact theories of
torsion
T
T
⫽ ᎏ
t1 ⫽ ᎏᎏ
2p r 2t
(3-73)
r
b ⫽ ᎏᎏ
t
(3-74)
in which the relation
is introduced.
Torsion formula. The maximum stress obtained from the more accurate
torsion formula (Eq. 3-11) is
T(r ⫹ t/2)
t2 ⫽ ᎏᎏ
IP
(f)
where
p
IP ⫽ ᎏᎏ
2
冤冢r ⫹ ᎏ2tᎏ冣 ⫺ 冢r ⫺ ᎏ2tᎏ冣 冥
4
4
(g)
After expansion, this expression simplifies to
p rt
IP ⫽ ᎏᎏ (4r 2 ⫹ t 2)
2
(3-75)
and the expression for the shear stress (Eq. f) becomes
T(2r ⫹ t)
T(2b ⫹ 1)
t2 ⫽ ᎏᎏ
⫽ ᎏᎏ
p rt(4r 2 ⫹ t 2)
p t 3b(4b 2 ⫹ 1)
(3-76)
Ratio. The ratio t1/t2 of the shear stresses is
4b 2 ⫹ 1
t
ᎏᎏ1 ⫽ ᎏᎏ
t2
2b (2 b ⫹ 1)
(3-77)
which depends only on the ratio b.
For values of b equal to 5, 10, and 20, we obtain from Eq. (3-77) the values
t1/t2 ⫽ 0.92, 0.95, and 0.98, respectively. Thus, we see that the approximate
formula for the shear stresses gives results that are slightly less than those
obtained from the exact formula. The accuracy of the approximate formula
increases as the wall of the tube becomes thinner. In the limit, as the thickness
approaches zero and b approaches infinity, the ratio t1/t2 becomes 1.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
242
CHAPTER 3 Torsion
Example 3-14
A circular tube and a square tube (Fig. 3-46) are constructed of the same material and subjected to the same torque. Both tubes have the same length, same
wall thickness, and same cross-sectional area.
What are the ratios of their shear stresses and angles of twist? (Disregard
the effects of stress concentrations at the corners of the square tube.)
t
r
t
b
FIG. 3-46 Example 3-14. Comparison of
circular and square tubes
(a)
(b)
Solution
Circular tube. For the circular tube, the area Am1 enclosed by the median
line of the cross section is
Am1 ⫽ p r 2
(h)
where r is the radius to the median line. Also, the torsion constant (Eq. 3-70)
and cross-sectional area are
J1 ⫽ 2pr 3t
A1 ⫽ 2prt
(i,j)
Square tube. For the square tube, the cross-sectional area is
A2 ⫽ 4bt
(k)
where b is the length of one side, measured along the median line. Inasmuch as
the areas of the tubes are the same, we obtain b ⫽ p r/2. Also, the torsion
constant (Eq. 3-71) and area enclosed by the median line of the cross section are
p 3r 3t
J2 ⫽ b3t ⫽ ᎏᎏ
8
p 2r 2
Am2 ⫽ b2 ⫽ ᎏᎏ
4
(l,m)
Ratios. The ratio t1/t2 of the shear stress in the circular tube to the shear
stress in the square tube (from Eq. 3-61) is
t1
p 2r 2 /4 p
A 2
ᎏᎏ ⫽ ᎏmᎏ
⫽ ᎏᎏ
⫽ ᎏᎏ ⫽ 0.79
t2
4
Am1
pr2
(n)
The ratio of the angles of twist (from Eq. 3-72) is
f1
J2
p 3r 3t /8
p2
ᎏᎏ ⫽ ᎏ ᎏ ⫽ ᎏ ᎏ
3 ⫽ ᎏᎏ ⫽ 0.62
f2
J1
2 pr t
16
(o)
These results show that the circular tube not only has a 21% lower shear
stress than does the square tube but also a greater stiffness against rotation.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 3.11 Stress Concentrations in Torsion
243
3.11 STRESS CONCENTRATIONS IN TORSION
★
In the previous sections of this chapter we discussed the stresses in torsional members assuming that the stress distribution varied in a smooth
and continuous manner. This assumption is valid provided that there are
no abrupt changes in the shape of the bar (no holes, grooves, abrupt
steps, and the like) and provided that the region under consideration is
away from any points of loading. If such disruptive conditions do exist,
then high localized stresses will be created in the regions surrounding the
discontinuities. In practical engineering work these stress concentrations
are handled by means of stress-concentration factors, as explained
previously in Section 2.10.
The effects of a stress concentration are confined to a small region
around the discontinuity, in accord with Saint-Venant’s principle (see
Section 2.10). For instance, consider a stepped shaft consisting of two
segments having different diameters (Fig. 3-47). The larger segment has
diameter D2 and the smaller segment has diameter D1. The junction
between the two segments forms a “step” or “shoulder” that is machined
with a fillet of radius R. Without the fillet, the theoretical stress concentration factor would be infinitely large because of the abrupt 90°
reentrant corner. Of course, infinite stresses cannot occur. Instead, the
material at the reentrant corner would deform and partially relieve the
high stress concentration. However, such a situation is very dangerous
under dynamic loads, and in good design a fillet is always used. The
larger the radius of the fillet, the lower the stresses.
Fillet (R = radius)
A
B
T
C
D2
T
D1
B
C
A
(a)
t2
t1
tmax
D2
FIG. 3-47 Stepped shaft in torsion
Section A-A
(b)
D1
D1
Section B-B
(c)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
Section C-C
(d)
244
CHAPTER 3 Torsion
At a distance from the shoulder approximately equal to the diameter
D2 (for instance, at cross section A-A in Fig. 3-47a) the torsional shear
stresses are practically unaffected by the discontinuity. Therefore, the
maximum stress t2 at a sufficient distance to the left of the shoulder can
be found from the torsion formula using D2 as the diameter (Fig. 3-47b).
The same general comments apply at section C-C, which is distance D1
(or greater) from the toe of the fillet. Because the diameter D1 is less
than the diameter D2, the maximum stress t1 at section C-C (Fig. 3-47d)
is larger than the stress t2.
The stress-concentration effect is greatest at section B-B, which cuts
through the toe of the fillet. At this section the maximum stress is
冢 pD 冣
16T
Tr
IP
(3-78)
tmax ⫽ Ktnom ⫽ K ᎏᎏ ⫽ K ᎏᎏ3
1
In this equation, K is the stress-concentration factor and tnom (equal to
t1) is the nominal shear stress, that is, the shear stress in the smaller part
of the shaft.
Values of the factor K are plotted in Fig. 3-48 as a function of the
ratio R/D1. Curves are plotted for various values of the ratio D2/D1. Note
that when the fillet radius R becomes very small and the transition from
one diameter to the other is abrupt, the value of K becomes quite large.
Conversely, when R is large, the value of K approaches 1.0 and the effect
of the stress concentration disappears. The dashed curve in Fig. 3-48 is
for the special case of a full quarter-circular fillet, which means that D2
⫽ D1 ⫹ 2R. (Note: Problems 3.11-1 through 3.11-5 provide practice in
obtaining values of K from Fig. 3-48.)
2.00
R
T
1.2
K
D2
1.1
tmax = Ktnom
1.5
1.50
D1
16T
—
tnom = —
p D13
D2
—–
=
D1 2
D2 = D1 + 2R
FIG. 3-48 Stress-concentration factor K
for a stepped shaft in torsion. (The
dashed line is for a full quarter-circular
fillet.)
1.00
0
0.10
0.20
R–
—
D1
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
T
CHAPTER 3 Problems
245
Many other cases of stress concentrations for circular shafts, such as
a shaft with a keyway and a shaft with a hole, are available in the engineering literature (see, for example, Ref. 2-9).
As explained in Section 2.10, stress concentrations are important for
brittle materials under static loads and for most materials under dynamic
loads. As a case in point, fatigue failures are of major concern in the
design of rotating shafts and axles (see Section 2.9 for a brief discussion
of fatigue). The theoretical stress-concentration factors K given in this
section are based upon linearly elastic behavior of the material.
However, fatigue experiments show that these factors are conservative,
and failures in ductile materials usually occur at larger loads than those
predicted by the theoretical factors.
PROBLEMS CHAPTER 3
T
Torsional Deformations
T
3.2-1 A copper rod of length L ⫽ 18.0 in. is to be twisted
by torques T (see figure) until the angle of rotation between
the ends of the rod is 3.0°.
If the allowable shear strain in the copper is 0.0006
rad, what is the maximum permissible diameter of the rod?
d
T
L
r2
r1
T
PROBS. 3.2-3, 3.2-4, and 3.2-5
L
PROBS. 3.2-1 and 3.2-2
3.2-2 A plastic bar of diameter d ⫽ 50 mm is to be twisted
by torques T (see figure) until the angle of rotation between
the ends of the bar is 5.0°.
If the allowable shear strain in the plastic is 0.012 rad,
what is the minimum permissible length of the bar?
3.2-4 A circular steel tube of length L ⫽ 0.90 m is loaded
in torsion by torques T (see figure).
(a) If the inner radius of the tube is r1 ⫽ 40 mm and
the measured angle of twist between the ends is 0.5°, what
is the shear strain g1 (in radians) at the inner surface?
(b) If the maximum allowable shear strain is 0.0005
rad and the angle of twist is to be kept at 0.5° by adjusting
the torque T, what is the maximum permissible outer radius
(r 2)max?
3.2-5 Solve the preceding problem if the length L ⫽ 50 in.,
3.2-3 A circular aluminum tube subjected to pure torsion
by torques T (see figure) has an outer radius r2 equal to
twice the inner radius r1.
(a) If the maximum shear strain in the tube is measured
as 400 ⫻ 10⫺6 rad, what is the shear strain g1 at the inner
surface?
(b) If the maximum allowable rate of twist is 0.15
degrees per foot and the maximum shear strain is to be kept
at 400 ⫻ 10⫺6 rad by adjusting the torque T, what is the
minimum required outer radius (r2)min?
the inner radius r1 ⫽ 1.5 in., the angle of twist is 0.6°, and
the allowable shear strain is 0.0004 rad.
Circular Bars and Tubes
3.3-1 A prospector uses a hand-powered winch (see figure
on the next page) to raise a bucket of ore in his mine shaft.
The axle of the winch is a steel rod of diameter d ⫽ 0.625 in.
Also, the distance from the center of the axle to the center
of the lifting rope is b ⫽ 4.0 in.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
246
CHAPTER 3 Torsion
If the weight of the loaded bucket is W ⫽ 100 lb, what
is the maximum shear stress in the axle due to torsion?
P
P
9.0
d
W
in.
b
A
9.0
W
PROB. 3.3-1
in.
d = 0.5 in.
P = 25 lb
3.3-2 When drilling a hole in a table leg, a furniture maker
uses a hand-operated drill (see figure) with a bit of diameter
d ⫽ 4.0 mm.
(a) If the resisting torque supplied by the table leg is
equal to 0.3 N⭈m, what is the maximum shear stress in the
drill bit?
(b) If the shear modulus of elasticity of the steel is
G ⫽ 75 GPa, what is the rate of twist of the drill bit
(degrees per meter)?
PROB. 3.3-3
3.3-4 An aluminum bar of solid circular cross section is
twisted by torques T acting at the ends (see figure). The
dimensions and shear modulus of elasticity are as follows:
L ⫽ 1.2 m, d ⫽ 30 mm, and G ⫽ 28 GPa.
(a) Determine the torsional stiffness of the bar.
(b) If the angle of twist of the bar is 4°, what is the
maximum shear stress? What is the maximum shear strain
(in radians)?
d
T
d
T
L
PROB. 3.3-4
PROB. 3.3-2
3.3-3 While removing a wheel to change a tire, a driver
applies forces P ⫽ 25 lb at the ends of two of the arms of a
lug wrench (see figure). The wrench is made of steel with
shear modulus of elasticity G ⫽ 11.4 ⫻ 106 psi. Each arm
of the wrench is 9.0 in. long and has a solid circular cross
section of diameter d ⫽ 0.5 in.
(a) Determine the maximum shear stress in the arm
that is turning the lug nut (arm A).
(b) Determine the angle of twist (in degrees) of this
same arm.
3.3-5 A high-strength steel drill rod used for boring a hole
in the earth has a diameter of 0.5 in. (see figure).The allowable shear stress in the steel is 40 ksi and the shear modulus
of elasticity is 11,600 ksi.
What is the minimum required length of the rod so that
one end of the rod can be twisted 30° with respect to the
other end without exceeding the allowable stress?
T
d = 0.5 in.
T
L
PROB. 3.3-5
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
247
CHAPTER 3 Problems
3.3-6 The steel shaft of a socket wrench has a diameter of
8.0 mm. and a length of 200 mm (see figure).
If the allowable stress in shear is 60 MPa, what is the
maximum permissible torque Tmax that may be exerted with
the wrench?
Through what angle f (in degrees) will the shaft twist
under the action of the maximum torque? (Assume
G ⫽ 78 GPa and disregard any bending of the shaft.)
3.3-9 Three identical circular disks A, B, and C are welded
to the ends of three identical solid circular bars (see figure).
The bars lie in a common plane and the disks lie in planes
perpendicular to the axes of the bars. The bars are welded
at their intersection D to form a rigid connection. Each
bar has diameter d1 ⫽ 0.5 in. and each disk has diameter
d2 ⫽ 3.0 in.
Forces P1, P2, and P3 act on disks A, B, and C, respectively, thus subjecting the bars to torsion. If P1 ⫽ 28 lb, what
is the maximum shear stress tmax in any of the three bars?
P3
C
d = 8.0 mm
T
135°
P1
L = 200 mm
P3
d1
A
PROB. 3.3-6
D
3.3-7 A circular tube of aluminum is subjected to torsion
by torques T applied at the ends (see figure). The bar is 20
in. long, and the inside and outside diameters are 1.2 in.
and 1.6 in., respectively. It is determined by measurement
that the angle of twist is 3.63° when the torque is
5800 lb-in.
Calculate the maximum shear stress tmax in the tube,
the shear modulus of elasticity G, and the maximum shear
strain gmax (in radians).
T
T
20 in.
P1
d2
P2
P2
3.3-10 The steel axle of a large winch on an ocean liner is
subjected to a torque of 1.5 kN⭈m (see figure). What is the
minimum required diameter dmin if the allowable shear
stress is 50 MPa and the allowable rate of twist is 0.8°/m?
(Assume that the shear modulus of elasticity is 80 GPa.)
3.3-8 A propeller shaft for a small yacht is made of a
solid steel bar 100 mm in diameter. The allowable stress in
shear is 50 MPa, and the allowable rate of twist is 2.0° in
3 meters.
Assuming that the shear modulus of elasticity is
G ⫽ 80 GPa, determine the maximum torque Tmax that can
be applied to the shaft.
d
T
1.2 in.
PROB. 3.3-7
B
PROB. 3.3-9
T
1.6 in.
135°
90°
PROB. 3.3-10
3.3-11 A hollow steel shaft used in a construction auger has
outer diameter d2 ⫽ 6.0 in. and inner diameter d1 ⫽ 4.5 in.
(see figure on the next page). The steel has shear modulus of
elasticity G ⫽ 11.0 ⫻ 106 psi.
For an applied torque of 150 k-in., determine the
following quantities:
(a) shear stress t2 at the outer surface of the shaft,
(b) shear stress t1 at the inner surface, and
(c) rate of twist u (degrees per unit of length).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
248
CHAPTER 3 Torsion
Also, draw a diagram showing how the shear stresses
vary in magnitude along a radial line in the cross section.
3.3-14 Solve the preceding problem if the horizontal forces
have magnitude P ⫽ 5.0 kN, the distance c ⫽ 125 mm, and
the allowable shear stress is 30 MPa.
3.3-15 A solid brass bar of diameter d ⫽ 1.2 in. is
d2
subjected to torques T1, as shown in part (a) of the figure.
The allowable shear stress in the brass is 12 ksi.
(a) What is the maximum permissible value of the
torques T1?
(b) If a hole of diameter 0.6 in. is drilled longitudinally
through the bar, as shown in part (b) of the figure, what is
the maximum permissible value of the torques T2?
(c) What is the percent decrease in torque and the
percent decrease in weight due to the hole?
T1
d
T1
(a)
d1
d2
T2
d
T2
PROBS. 3.3-11 and 3.3-12
3.3-12 Solve the preceding problem if the shaft has outer
diameter d2 ⫽ 150 mm and inner diameter d1 ⫽ 100 mm.
Also, the steel has shear modulus of elasticity G ⫽ 75 GPa
and the applied torque is 16 kN⭈m.
3.3-13 A vertical pole of solid circular cross section is
twisted by horizontal forces P ⫽ 1100 lb acting at the ends
of a horizontal arm AB (see figure). The distance from the
outside of the pole to the line of action of each force is
c ⫽ 5.0 in.
If the allowable shear stress in the pole is 4500 psi,
what is the minimum required diameter dmin of the pole?
c
c
A
P
P
B
(b)
PROB. 3.3-15
3.3-16 A hollow aluminum tube used in a roof structure has
an outside diameter d2 ⫽ 100 mm and an inside diameter
d1 ⫽ 80 mm (see figure). The tube is 2.5 m long, and the
aluminum has shear modulus G ⫽ 28 GPa.
(a) If the tube is twisted in pure torsion by torques
acting at the ends, what is the angle of twist f (in degrees)
when the maximum shear stress is 50 MPa?
(b) What diameter d is required for a solid shaft (see
figure) to resist the same torque with the same maximum
stress?
(c) What is the ratio of the weight of the hollow tube to
the weight of the solid shaft?
d
d1
d2
PROBS. 3.3-13 and 3.3-14
PROB. 3.3-16
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
d
249
CHAPTER 3 Problems
3.3-17 A circular tube of inner radius r1 and outer radius
r2 is subjected to a torque produced by forces P ⫽ 900 lb
(see figure). The forces have their lines of action at a
distance b ⫽ 5.5 in. from the outside of the tube.
If the allowable shear stress in the tube is 6300 psi and
the inner radius r1 ⫽ 1.2 in., what is the minimum permissible outer radius r2?
★
P
3.4-2 A circular tube of outer diameter d3 ⫽ 70 mm and
inner diameter d2 ⫽ 60 mm is welded at the right-hand end to
a fixed plate and at the left-hand end to a rigid end plate (see
figure). A solid circular bar of diameter d1 ⫽ 40 mm is inside
of, and concentric with, the tube. The bar passes through a
hole in the fixed plate and is welded to the rigid end plate.
The bar is 1.0 m long and the tube is half as long as the
bar. A torque T ⫽ 1000 N⭈m acts at end A of the bar. Also,
both the bar and tube are made of an aluminum alloy with
shear modulus of elasticity G ⫽ 27 GPa.
(a) Determine the maximum shear stresses in both the
bar and tube.
(b) Determine the angle of twist (in degrees) at end A
of the bar.
;@QQ;@
Tube
P
End
plate
P
r2
r1
P
b
b
2r2
Bar
T
A
Tube
PROB. 3.3-17
Bar
Nonuniform Torsion
d1
d2
d3
3.4-1 A stepped shaft ABC consisting of two solid circular
segments is subjected to torques T1 and T2 acting in opposite
directions, as shown in the figure. The larger segment of the
shaft has diameter d1 ⫽ 2.25 in. and length L1 ⫽ 30 in.; the
smaller segment has diameter d2 ⫽ 1.75 in. and length L2 ⫽
20 in. The material is steel with shear modulus G ⫽ 11 ⫻ 106
psi, and the torques are T1 ⫽ 20,000 lb-in. and T2 ⫽
8,000 lb-in.
Calculate the following quantities: (a) the maximum
shear stress tmax in the shaft, and (b) the angle of twist fC
(in degrees) at end C.
T1
d1
d2
B
A
PROB. 3.4-2
3.4-3 A stepped shaft ABCD consisting of solid circular
segments is subjected to three torques, as shown in the
figure. The torques have magnitudes 12.0 k-in., 9.0 k-in.,
and 9.0 k-in. The length of each segment is 24 in. and
the diameters of the segments are 3.0 in., 2.5 in., and 2.0 in.
The material is steel with shear modulus of elasticity
G ⫽ 11.6 ⫻ 103 ksi.
(a) Calculate the maximum shear stress tmax in the shaft.
(b) Calculate the angle of twist fD (in degrees) at end D.
T2
12.0 k-in.
3.0 in.
9.0 k-in.
2.5 in.
C
L2
24 in.
PROB. 3.4-3
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
9.0 k-in.
2.0 in.
D
C
B
A
L1
PROB. 3.4-1
Fixed
plate
24 in.
24 in.
250
CHAPTER 3 Torsion
3.4-4 A solid circular bar ABC consists of two segments, as
80 mm
shown in the figure. One segment has diameter d1 ⫽ 50 mm
and length L1 ⫽ 1.25 m; the other segment has diameter
d2 ⫽ 40 mm and length L2 ⫽ 1.0 m.
What is the allowable torque Tallow if the shear stress is
not to exceed 30 MPa and the angle of twist between the
ends of the bar is not to exceed 1.5°? (Assume G ⫽ 80 GPa.)
d1
T
d2
A
L1
1.2 m
0.9 m
d
T
d
t=—
10
2.1 m
C
B
60 mm
PROB. 3.4-6
L2
PROB. 3.4-4
3.4-5 A hollow tube ABCDE constructed of monel metal is
subjected to five torques acting in the directions shown in the
figure. The magnitudes of the torques are T1 ⫽ 1000 lb-in.,
T2 ⫽ T4 ⫽ 500 lb-in., and T3 ⫽ T5 ⫽ 800 lb-in. The tube
has an outside diameter d2 ⫽ 1.0 in. The allowable shear
stress is 12,000 psi and the allowable rate of twist is 2.0°/ft.
Determine the maximum permissible inside diameter
d1 of the tube.
3.4-7 Four gears are attached to a circular shaft and
transmit the torques shown in the figure. The allowable
shear stress in the shaft is 10,000 psi.
(a) What is the required diameter d of the shaft if it has
a solid cross section?
(b) What is the required outside diameter d if the shaft
is hollow with an inside diameter of 1.0 in.?
8,000 lb-in.
19,000 lb-in.
T1 =
T2 =
1000 lb-in. 500 lb-in.
T3 =
T4 =
800 lb-in. 500 lb-in.
T5 =
800 lb-in.
4,000 lb-in.
A
A
B
C
D
d2 = 1.0 in.
7,000 lb-in.
B
E
C
D
PROB. 3.4-5
PROB. 3.4-7
3.4-6 A shaft of solid circular cross section consisting of
two segments is shown in the first part of the figure. The
left-hand segment has diameter 80 mm and length 1.2 m;
the right-hand segment has diameter 60 mm and length
0.9 m.
Shown in the second part of the figure is a hollow shaft
made of the same material and having the same length. The
thickness t of the hollow shaft is d/10, where d is the outer
diameter. Both shafts are subjected to the same torque.
If the hollow shaft is to have the same torsional stiffness
as the solid shaft, what should be its outer diameter d?
3.4-8 A tapered bar AB of solid circular cross section is
twisted by torques T (see figure). The diameter of the bar
varies linearly from dA at the left-hand end to dB at the
right-hand end.
For what ratio dB/dA will the angle of twist of the
tapered bar be one-half the angle of twist of a prismatic bar
of diameter dA? (The prismatic bar is made of the same material, has the same length, and is subjected to the same torque
as the tapered bar.) Hint: Use the results of Example 3-5.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
251
CHAPTER 3 Problems
T
B
A
B
A
T
T
T
L
dA
L
dB
t
t
PROBS. 3.4-8, 3.4-9, and 3.4-10
dA
dB = 2dA
PROB. 3.4-11
3.4-9 A tapered bar AB of solid circular cross section is
twisted by torques T ⫽ 36,000 lb-in. (see figure). The
diameter of the bar varies linearly from dA at the left-hand
end to dB at the right-hand end. The bar has length L ⫽
4.0 ft and is made of an aluminum alloy having shear
modulus of elasticity G ⫽ 3.9 ⫻ 106 psi. The allowable
shear stress in the bar is 15,000 psi and the allowable
angle of twist is 3.0°.
If the diameter at end B is 1.5 times the diameter at
end A, what is the minimum required diameter dA at end A?
(Hint: Use the results of Example 3-5).
3.4-12 A prismatic bar AB of length L and solid circular
cross section (diameter d) is loaded by a distributed torque
of constant intensity t per unit distance (see figure).
(a) Determine the maximum shear stress tmax in the
bar.
(b) Determine the angle of twist f between the ends of
the bar.
end A to end B and has a solid circular cross section. The
diameter at the smaller end of the bar is dA ⫽ 25 mm and
the length is L ⫽ 300 mm. The bar is made of steel with
shear modulus of elasticity G ⫽ 82 GPa.
If the torque T ⫽ 180 N⭈m and the allowable angle of
twist is 0.3°, what is the minimum allowable diameter dB
at the larger end of the bar? (Hint: Use the results of
Example 3-5.)
L
cross section is shown in the figure. The tube has constant
wall thickness t and length L. The average diameters at the
ends are dA and dB ⫽ 2dA. The polar moment of inertia may
be represented by the approximate formula IP ⬇ pd 3t/4
(see Eq. 3-18).
Derive a formula for the angle of twist f of the tube
when it is subjected to torques T acting at the ends.
B
PROB. 3.4-12
3.4-13 A prismatic bar AB of solid circular cross section
(diameter d) is loaded by a distributed torque (see figure on
the next page). The intensity of the torque, that is, the
torque per unit distance, is denoted t(x) and varies linearly
from a maximum value tA at end A to zero at end B. Also,
the length of the bar is L and the shear modulus of elasticity
of the material is G.
(a) Determine the maximum shear stress tmax in the
bar.
★
3.4-11 A uniformly tapered tube AB of hollow circular
t
A
3.4-10 The bar shown in the figure is tapered linearly from
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
252
CHAPTER 3 Torsion
(b) Determine the angle of twist f between the ends of
the bar.
(a) Determine the maximum tensile stress smax in the
shaft.
(b) Determine the magnitude of the applied torques T.
t(x)
T
A
d2
L
B
L
T
d1 d2
PROB. 3.4-13
★★
3.4-14 A magnesium-alloy wire of diameter d ⫽ 4 mm
and length L rotates inside a flexible tube in order to open
or close a switch from a remote location (see figure). A
torque T is applied manually (either clockwise or counterclockwise) at end B, thus twisting the wire inside the tube.
At the other end A, the rotation of the wire operates a
handle that opens or closes the switch.
A torque T0 ⫽ 0.2 N⭈m is required to operate the
switch. The torsional stiffness of the tube, combined with
friction between the tube and the wire, induces a distributed
torque of constant intensity t ⫽ 0.04 N⭈m/m (torque per
unit distance) acting along the entire length of the wire.
(a) If the allowable shear stress in the wire is tallow ⫽
30 MPa, what is the longest permissible length Lmax of the
wire?
(b) If the wire has length L ⫽ 4.0 m and the shear
modulus of elasticity for the wire is G ⫽ 15 GPa, what is
the angle of twist f (in degrees) between the ends of the
wire?
PROBS. 3.5-1, 3.5-2, and 3.5-3
3.5-2 A hollow steel bar (G ⫽ 80 GPa) is twisted by
torques T (see figure). The twisting of the bar produces a
maximum shear strain gmax ⫽ 640 ⫻ 10⫺6 rad. The bar has
outside and inside diameters of 150 mm and 120 mm,
respectively.
(a) Determine the maximum tensile strain in the bar.
(b) Determine the maximum tensile stress in the bar.
(c) What is the magnitude of the applied torques T ?
3.5-3 A tubular bar with outside diameter d2 ⫽ 4.0 in. is
twisted by torques T ⫽ 70.0 k-in. (see figure). Under the
action of these torques, the maximum tensile stress in the
bar is found to be 6400 psi.
(a) Determine the inside diameter d1 of the bar.
(b) If the bar has length L ⫽ 48.0 in. and is made of
aluminum with shear modulus G ⫽ 4.0 ⫻ 106 psi, what is
the angle of twist f (in degrees) between the ends of the
bar?
(c) Determine the maximum shear strain gmax (in
radians)?
;@;;
@@
T0 = torque
Flexible tube
B
3.5-4 A solid circular bar of diameter d ⫽ 50 mm (see
d
A
T
t
PROB. 3.4-14
figure) is twisted in a testing machine until the applied
torque reaches the value T ⫽ 500 N⭈m. At this value of
torque, a strain gage oriented at 45° to the axis of the bar
gives a reading e ⫽ 339 ⫻ 10⫺6.
What is the shear modulus G of the material?
Pure Shear
3.5-1 A hollow aluminum shaft (see figure) has outside
diameter d2 ⫽ 4.0 in. and inside diameter d1 ⫽ 2.0 in.
When twisted by torques T, the shaft has an angle of twist
per unit distance equal to 0.54°/ft. The shear modulus of
elasticity of the aluminum is G ⫽ 4.0 ⫻ 106 psi.
d = 50 mm
Strain gage
T
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
45°
T = 500 N·m
CHAPTER 3 Problems
PROB. 3.5-4
3.5-5 A steel tube (G ⫽ 11.5 ⫻ 106 psi) has an outer diameter d2 ⫽ 2.0 in. and an inner diameter d1 ⫽ 1.5 in. When
twisted by a torque T, the tube develops a maximum normal
strain of 170 ⫻ 10⫺6.
What is the magnitude of the applied torque T?
d ⫽ 40 mm is subjected to torques T ⫽ 300 N⭈m acting in
the directions shown in the figure.
(a) Determine the maximum shear, tensile, and
compressive stresses in the bar and show these stresses on
sketches of properly oriented stress elements.
(b) Determine the corresponding maximum strains
(shear, tensile, and compressive) in the bar and show these
strains on sketches of the deformed elements.
3.5-6 A solid circular bar of steel (G ⫽ 78 GPa) transmits a
torque T ⫽ 360 N⭈m. The allowable stresses in tension,
compression, and shear are 90 MPa, 70 MPa, and 40 MPa,
respectively. Also, the allowable tensile strain is 220 ⫻ 10⫺6.
Determine the minimum required diameter d of the bar.
253
T
T = 300 N·m
d = 40 mm
PROB. 3.5-10
3.5-7 The normal strain in the 45° direction on the surface
of a circular tube (see figure) is 880 ⫻ 10⫺6 when the torque
T ⫽ 750 lb-in. The tube is made of copper alloy with G ⫽
6.2 ⫻ 106 psi.
If the outside diameter d2 of the tube is 0.8 in., what is
the inside diameter d1?
T
d 2 = 0.8 in.
Strain gage
T = 750 lb-in.
Transmission of Power
3.7-1 A generator shaft in a small hydroelectric plant turns
at 120 rpm and delivers 50 hp (see figure).
(a) If the diameter of the shaft is d ⫽ 3.0 in., what is
the maximum shear stress tmax in the shaft?
(b) If the shear stress is limited to 4000 psi, what is the
minimum permissible diameter dmin of the shaft?
120 rpm
45°
d
PROB. 3.5-7
50 hp
3.5-8 An aluminum tube has inside diameter d1 ⫽ 50 mm,
shear modulus of elasticity G ⫽ 27 GPa, and torque T ⫽
4.0 kN⭈m. The allowable shear stress in the aluminum is
50 MPa and the allowable normal strain is 900 ⫻ 10⫺6.
Determine the required outside diameter d2.
3.5-9 A solid steel bar (G ⫽ 11.8 ⫻ 106 psi) of diameter
d ⫽ 2.0 in. is subjected to torques T ⫽ 8.0 k-in. acting in
the directions shown in the figure.
(a) Determine the maximum shear, tensile, and
compressive stresses in the bar and show these stresses on
sketches of properly oriented stress elements.
(b) Determine the corresponding maximum strains
(shear, tensile, and compressive) in the bar and show these
strains on sketches of the deformed elements.
T
d = 2.0 in.
PROB. 3.7-1
3.7-2 A motor drives a shaft at 12 Hz and delivers 20 kW
of power (see figure).
(a) If the shaft has a diameter of 30 mm, what is the
maximum shear stress tmax in the shaft?
(b) If the maximum allowable shear stress is 40 MPa,
what is the minimum permissible diameter dmin of the
shaft?
12 Hz
d
20 kW
T = 8.0 k-in.
PROB. 3.7-2
PROB. 3.5-9
3.5-10 A solid aluminum bar (G ⫽ 27 GPa) of diameter
3.7-3 The propeller shaft of a large ship has outside diameter 18 in. and inside diameter 12 in., as shown in the figure
on the next page. The shaft is rated for a maximum shear
stress of 4500 psi.
(a) If the shaft is turning at 100 rpm, what is the
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
254
CHAPTER 3 Torsion
maximum horsepower that can be transmitted without
exceeding the allowable stress?
(b) If the rotational speed of the shaft is doubled but
the power requirements remain unchanged, what happens
to the shear stress in the shaft?
18 in.
collar in order that the splice can transmit the same power
as the solid shaft?
d1
d
100 rpm
PROB. 3.7-7
12 in.
18 in.
PROB. 3.7-3
3.7-4 The drive shaft for a truck (outer diameter 60 mm
and inner diameter 40 mm) is running at 2500 rpm (see
figure).
(a) If the shaft transmits 150 kW, what is the maximum
shear stress in the shaft?
(b) If the allowable shear stress is 30 MPa, what is the
maximum power that can be transmitted?
2500 rpm
60 mm
3.7-8 What is the maximum power that can be delivered by
a hollow propeller shaft (outside diameter 50 mm, inside
diameter 40 mm, and shear modulus of elasticity 80 GPa)
turning at 600 rpm if the allowable shear stress is 100 MPa
and the allowable rate of twist is 3.0°/m?
3.7-9 A motor delivers 275 hp at 1000 rpm to the end of
a shaft (see figure). The gears at B and C take out 125 and
150 hp, respectively.
Determine the required diameter d of the shaft if the
allowable shear stress is 7500 psi and the angle of twist
between the motor and gear C is limited to 1.5°. (Assume
G ⫽ 11.5 ⫻ 106 psi, L1 ⫽ 6 ft, and L2 ⫽ 4 ft.)
★
Motor
A
C
d
B
40 mm
60 mm
PROB. 3.7-4
L1
3.7-5 A hollow circular shaft for use in a pumping station is
being designed with an inside diameter equal to 0.75 times
the outside diameter. The shaft must transmit 400 hp at 400
rpm without exceeding the allowable shear stress of 6000 psi.
Determine the minimum required outside diameter d.
3.7-6 A tubular shaft being designed for use on a construction site must transmit 120 kW at 1.75 Hz. The inside
diameter of the shaft is to be one-half of the outside diameter.
If the allowable shear stress in the shaft is 45 MPa,
what is the minimum required outside diameter d?
3.7-7 A propeller shaft of solid circular cross section and
diameter d is spliced by a collar of the same material (see
figure). The collar is securely bonded to both parts of the
shaft.
What should be the minimum outer diameter d1 of the
L2
PROBS. 3.7-9 and 3.7-10
3.7-10 The shaft ABC shown in the figure is driven by a
motor that delivers 300 kW at a rotational speed of 32 Hz.
The gears at B and C take out 120 and 180 kW, respectively. The lengths of the two parts of the shaft are L1 ⫽ 1.5
m and L2 ⫽ 0.9 m.
Determine the required diameter d of the shaft if the
allowable shear stress is 50 MPa, the allowable angle of
twist between points A and C is 4.0°, and G ⫽ 75 GPa.
★
Statically Indeterminate Torsional Members
3.8-1 A solid circular bar ABCD with fixed supports is acted
upon by torques T0 and 2T0 at the locations shown in the figure on the next page.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
255
CHAPTER 3 Problems
Obtain a formula for the maximum angle of twist fmax
of the bar. (Hint: Use Eqs. 3-46a and b of Example 3-9 to
obtain the reactive torques.)
TA
A
T0
2T0
B
C
D
Disk
A
TD
d
B
a
b
PROB. 3.8-3
3L
—
10
3L
—
10
3.8-4 A hollow steel shaft ACB of outside diameter 50 mm
4L
—
10
L
PROB. 3.8-1
3.8-2 A solid circular bar ABCD with fixed supports at ends
A and D is acted upon by two equal and oppositely directed
torques T0, as shown in the figure. The torques are applied at
points B and C, each of which is located at distance x from
one end of the bar. (The distance x may vary from zero to
L/2.)
(a) For what distance x will the angle of twist at points
B and C be a maximum?
(b) What is the corresponding angle of twist fmax?
(Hint: Use Eqs. 3-46a and b of Example 3-9 to obtain the
reactive torques.)
TA
A
T0
T0
B
C
and inside diameter 40 mm is held against rotation at ends
A and B (see figure). Horizontal forces P are applied at the
ends of a vertical arm that is welded to the shaft at point C.
Determine the allowable value of the forces P if the
maximum permissible shear stress in the shaft is 45 MPa.
(Hint: Use Eqs. 3-46a and b of Example 3-9 to obtain the
reactive torques.)
200 mm
A
P
200 mm
C
B
P
D
600 mm
TD
400 mm
x
x
PROB. 3.8-4
L
PROB. 3.8-2
3.8-5 A stepped shaft ACB having solid circular cross
3.8-3 A solid circular shaft AB of diameter d is fixed
against rotation at both ends (see figure). A circular disk is
attached to the shaft at the location shown.
What is the largest permissible angle of rotation fmax
of the disk if the allowable shear stress in the shaft is tallow?
(Assume that a ⬎ b. Also, use Eqs. 3-46a and b of Example
3-9 to obtain the reactive torques.)
sections with two different diameters is held against rotation at the ends (see figure).
If the allowable shear stress in the shaft is 6000 psi,
what is the maximum torque (T0)max that may be applied at
section C? (Hint: Use Eqs. 3-45a and b of Example 3-9 to
obtain the reactive torques.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
256
CHAPTER 3 Torsion
1.50 in.
0.75 in.
C
A
dA
B
A
dB
C
IPB
B
T0
T0
6.0 in.
IPA
a
15.0 in.
L
PROB. 3.8-5
PROB. 3.8-7
3.8-6 A stepped shaft ACB having solid circular cross
sections with two different diameters is held against rotation at the ends (see figure on the next page).
If the allowable shear stress in the shaft is 43 MPa,
what is the maximum torque (T0)max that may be applied at
section C? (Hint: Use Eqs. 3-45a and b of Example 3-9 to
obtain the reactive torques.)
3.8-8 A circular bar AB of length L is fixed against rotation
at the ends and loaded by a distributed torque t(x) that varies
linearly in intensity from zero at end A to t0 at end B (see
figure).
Obtain formulas for the fixed-end torques TA and TB.
t0
t(x)
20 mm
25 mm
B
C
A
TA
TB
A
T0
225 mm
B
x
450 mm
L
PROB. 3.8-8
PROB. 3.8-6
3.8-7 A stepped shaft ACB is held against rotation at ends
A and B and subjected to a torque T0 acting at section C
(see figure). The two segments of the shaft (AC and CB)
have diameters dA and dB, respectively, and polar moments
of inertia IPA and IPB, respectively. The shaft has length L
and segment AC has length a.
(a) For what ratio a/L will the maximum shear stresses
be the same in both segments of the shaft?
(b) For what ratio a /L will the internal torques be the
same in both segments of the shaft? (Hint: Use Eqs. 3-45a
and b of Example 3-9 to obtain the reactive torques.)
3.8-9 A circular bar AB with ends fixed against rotation has
a hole extending for half of its length (see figure). The outer
diameter of the bar is d2 ⫽ 3.0 in. and the diameter of the
hole is d1 ⫽ 2.4 in. The total length of the bar is L ⫽ 50 in.
At what distance x from the left-hand end of the bar
should a torque T0 be applied so that the reactive torques at
the supports will be equal?
25 in.
A
25 in.
T0
3.0 in.
B
x
2.4
in.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
3.0
in.
CHAPTER 3 Problems
PROB. 3.8-9
3.8-10 A solid steel bar of diameter d1 ⫽ 25.0 mm is
enclosed by a steel tube of outer diameter d3 ⫽ 37.5 mm and
inner diameter d2 ⫽ 30.0 mm (see figure on the next page).
Both bar and tube are held rigidly by a support at end A and
joined securely to a rigid plate at end B. The composite
bar, which has a length L ⫽ 550 mm, is twisted by a torque
T ⫽ 400 N⭈m acting on the end plate.
(a) Determine the maximum shear stresses t1 and t2 in
the bar and tube, respectively.
(b) Determine the angle of rotation f (in degrees) of
the end plate, assuming that the shear modulus of the steel is
G ⫽ 80 GPa.
(c) Determine the torsional stiffness kT of the composite bar. (Hint: Use Eqs. 3-44a and b to find the torques in the
bar and tube.)
(c) Determine the torsional stiffness kT of the
composite bar. (Hint: Use Eqs. 3-44a and b to find the
torques in the bar and tube.)
3.8-12 The composite shaft shown in the figure is manufactured by shrink-fitting a steel sleeve over a brass core so
that the two parts act as a single solid bar in torsion. The
outer diameters of the two parts are d1 ⫽ 40 mm for the
brass core and d2 ⫽ 50 mm for the steel sleeve. The shear
moduli of elasticity are Gb ⫽ 36 GPa for the brass and
Gs ⫽ 80 GPa for the steel.
Assuming that the allowable shear stresses in the brass
and steel are tb ⫽ 48 MPa and ts ⫽ 80 MPa, respectively,
determine the maximum permissible torque Tmax that may
be applied to the shaft. (Hint: Use Eqs. 3-44a and b to find
the torques.)
★
Tube
T
B
A
T
Bar
257
Steel sleeve
Brass core
T
End
plate
L
d1
d1 d2
d2
PROBS. 3.8-12 and 3.8-13
d3
PROBS. 3.8-10 and 3.8-11
3.8-11 A solid steel bar of diameter d1 ⫽ 1.50 in. is
enclosed by a steel tube of outer diameter d3 ⫽ 2.25 in. and
inner diameter d2 ⫽ 1.75 in. (see figure). Both bar and tube
are held rigidly by a support at end A and joined securely to
a rigid plate at end B. The composite bar, which has length
L ⫽ 30.0 in., is twisted by a torque T ⫽ 5000 lb-in. acting
on the end plate.
(a) Determine the maximum shear stresses t1 and t2 in
the bar and tube, respectively.
(b) Determine the angle of rotation f (in degrees) of
the end plate, assuming that the shear modulus of the steel is
G ⫽ 11.6 ⫻ 106 psi.
3.8-13 The composite shaft shown in the figure is manufactured by shrink-fitting a steel sleeve over a brass core so
that the two parts act as a single solid bar in torsion. The
outer diameters of the two parts are d1 ⫽ 1.6 in. for the
brass core and d2 ⫽ 2.0 in. for the steel sleeve. The shear
moduli of elasticity are Gb ⫽ 5400 ksi for the brass and
Gs ⫽ 12,000 ksi for the steel.
Assuming that the allowable shear stresses in the brass
and steel are tb ⫽ 4500 psi and ts ⫽ 7500 psi, respectively,
determine the maximum permissible torque Tmax that may
be applied to the shaft. (Hint: Use Eqs. 3-44a and b to find
the torques.)
★
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
258
★★
CHAPTER 3 Torsion
3.8-14 A steel shaft (Gs ⫽ 80 GPa) of total length L ⫽
4.0 m is encased for one-half of its length by a brass sleeve
(Gb ⫽ 40 GPa) that is securely bonded to the steel (see
figure on the next page). The outer diameters of the shaft
and sleeve are d1 ⫽ 70 mm and d2 ⫽ 90 mm, respectively.
(a) Determine the allowable torque T1 that may be
applied to the ends of the shaft if the angle of twist f
between the ends is limited to 8.0°.
(b) Determine the allowable torque T2 if the shear
stress in the brass is limited to tb ⫽ 70 MPa.
(c) Determine the allowable torque T3 if the shear
stress in the steel is limited to ts ⫽ 110 MPa.
(d) What is the maximum allowable torque Tmax if all
three of the preceding conditions must be satisfied?
Brass
sleeve
d2 = 90 mm
T
A
Steel
shaft
d1 = 70 mm
B
L
= 2.0 m
2
T
C
L
= 2.0 m
2
figure) has length L ⫽ 45 in., diameter d2 ⫽ 1.2 in., and
diameter d1 ⫽ 1.0 in. The material is brass with G ⫽
5.6 ⫻ 106 psi.
Determine the strain energy U of the shaft if the angle
of twist is 3.0°.
d2
d1
T
T
L
—
2
L
—
2
PROBS. 3.9-3 and 3.9-4
3.9-4 A stepped shaft of solid circular cross sections (see figure) has length L ⫽ 0.80 m, diameter d2 ⫽ 40 mm, and
diameter d1 ⫽ 30 mm. The material is steel with G ⫽ 80 GPa.
Determine the strain energy U of the shaft if the angle
of twist is 1.0°.
PROB. 3.8-14
Strain Energy in Torsion
3.9-1 A solid circular bar of steel (G ⫽ 11.4 ⫻ 106 psi)
with length L ⫽ 30 in. and diameter d ⫽ 1.75 in. is
subjected to pure torsion by torques T acting at the ends
(see figure).
(a) Calculate the amount of strain energy U stored in
the bar when the maximum shear stress is 4500 psi.
(b) From the strain energy, calculate the angle of twist
f (in degrees).
T
d
3.9-5 A cantilever bar of circular cross section and length L
is fixed at one end and free at the other (see figure). The bar
is loaded by a torque T at the free end and by a distributed
torque of constant intensity t per unit distance along the
length of the bar.
(a) What is the strain energy U1 of the bar when the
load T acts alone?
(b) What is the strain energy U2 when the load t acts
alone?
(c) What is the strain energy U3 when both loads act
simultaneously?
T
t
L
PROBS. 3.9-1 and 3.9-2
L
3.9-2 A solid circular bar of copper (G ⫽ 45 GPa) with
length L ⫽ 0.75 m and diameter d ⫽ 40 mm is subjected to
pure torsion by torques T acting at the ends (see figure).
(a) Calculate the amount of strain energy U stored in
the bar when the maximum shear stress is 32 MPa.
(b) From the strain energy, calculate the angle of twist
f (in degrees).
3.9-3 A stepped shaft of solid circular cross sections (see
T
PROB. 3.9-5
3.9-6 Obtain a formula for the strain energy U of the
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 3 Problems
statically indeterminate circular bar shown in the figure.
The bar has fixed supports at ends A and B and is loaded by
torques 2T0 and T0 at points C and D, respectively.
Hint: Use Eqs. 3-46a and b of Example 3-9, Section
3.8, to obtain the reactive torques.
2T0
259
end to a maximum value t ⫽ t0 at the support.
t0
T0
t
A
B
C
D
L
—
2
L
—
4
L
—
4
L
PROB. 3.9-6
PROB. 3.9-8
3.9-7 A statically indeterminate stepped shaft ACB is fixed
at ends A and B and loaded by a torque T0 at point C (see
figure). The two segments of the bar are made of the same
material, have lengths LA and LB, and have polar moments of
inertia IPA and IPB.
Determine the angle of rotation f of the cross section
at C by using strain energy.
Hint: Use Eq. 3-51b to determine the strain energy U in
terms of the angle f. Then equate the strain energy to the
work done by the torque T0. Compare your result with Eq.
3-48 of Example 3-9, Section 3.8.
3.9-9 A thin-walled hollow tube AB of conical shape has
constant thickness t and average diameters dA and dB at the
ends (see figure).
(a) Determine the strain energy U of the tube when it
is subjected to pure torsion by torques T.
(b) Determine the angle of twist f of the tube.
Note: Use the approximate formula IP ⬇ pd 3t/4 for a
thin circular ring; see Case 22 of Appendix D.
★
T
A
IPA
B
A
T
L
T0
C
IPB
LA
t
t
B
dA
LB
dB
PROB. 3.9-9
PROB. 3.9-7
3.9-10 A hollow circular tube A fits over the end of a
solid circular bar B, as shown in the figure. The far ends of
both bars are fixed. Initially, a hole through bar B makes an
angle b with a line through two holes in tube A. Then bar B
is twisted until the holes are aligned, and a pin is placed
through the holes.
When bar B is released and the system returns to equilibrium, what is the total strain energy U of the two bars?
★★
3.9-8 Derive a formula for the strain energy U of the
cantilever bar shown in the figure.
The bar has circular cross sections and length L. It is
subjected to a distributed torque of intensity t per unit
distance. The intensity varies linearly from t ⫽ 0 at the free
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
260
CHAPTER 3 Torsion
(Let IPA and IPB represent the polar moments of inertia of
bars A and B, respectively. The length L and shear modulus
of elasticity G are the same for both bars.)
IPA
IPB
Tube A
Bar B
L
L
subjected to a torque T ⫽ 1200 k-in.
Determine the maximum shear stress in the tube using
(a) the approximate theory of thin-walled tubes, and (b) the
exact torsion theory. Does the approximate theory give
conservative or nonconservative results?
10.0 in.
b
1.0 in.
Tube A
Bar B
PROB. 3.9-10
3.9-11 A heavy flywheel rotating at n revolutions per
minute is rigidly attached to the end of a shaft of diameter d
(see figure). If the bearing at A suddenly freezes, what will
be the maximum angle of twist f of the shaft? What is the
corresponding maximum shear stress in the shaft?
(Let L ⫽ length of the shaft, G ⫽ shear modulus of
elasticity, and Im ⫽ mass moment of inertia of the flywheel
about the axis of the shaft. Also, disregard friction in the
bearings at B and C and disregard the mass of the shaft.)
Hint: Equate the kinetic energy of the rotating flywheel
to the strain energy of the shaft.
★★
PROB. 3.10-1
3.10-2 A solid circular bar having diameter d is to be
replaced by a rectangular tube having cross-sectional
dimensions d ⫻ 2d to the median line of the cross section
(see figure).
Determine the required thickness tmin of the tube so
that the maximum shear stress in the tube will not exceed
the maximum shear stress in the solid bar.
t
t
d
d
2d
PROB. 3.10-2
A
d
3.10-3 A thin-walled aluminum tube of rectangular cross
n (rpm)
B
C
section (see figure) has a centerline dimensions b ⫽ 6.0 in.
and h ⫽ 4.0 in. The wall thickness t is constant and equal to
0.25 in.
(a) Determine the shear stress in the tube due to a
torque T ⫽ 15 k-in.
(b) Determine the angle of twist (in degrees) if the
length L of the tube is 50 in. and the shear modulus G is
4.0 ⫻ 106 psi.
PROB. 3.9-11
Thin-Walled Tubes
3.10-1 A hollow circular tube having an inside diameter of
10.0 in. and a wall thickness of 1.0 in. (see figure) is
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 3 Problems
t = 8 mm
r = 50 mm
261
r = 50 mm
t
h
b = 100 mm
b
PROB. 3.10-6
PROBS. 3.10-3 and 3.10-4
3.10-4 A thin-walled steel tube of rectangular cross section
(see figure) has centerline dimensions b ⫽150 mm and h ⫽
100 mm. The wall thickness t is constant and equal to 6.0 mm.
(a) Determine the shear stress in the tube due to a
torque T ⫽ 1650 N⭈m.
(b) Determine the angle of twist (in degrees) if the
length L of the tube is 1.2 m and the shear modulus G is
75 GPa.
3.10-5 A thin-walled circular tube and a solid circular bar
of the same material (see figure on the next page) are
subjected to torsion. The tube and bar have the same crosssectional area and the same length.
What is the ratio of the strain energy U1 in the tube to
the strain energy U2 in the solid bar if the maximum shear
stresses are the same in both cases? (For the tube, use the
approximate theory for thin-walled bars.)
3.10-7 A thin-walled steel tube having an elliptical cross
section with constant thickness t (see figure) is subjected to
a torque T ⫽ 18 k-in.
Determine the shear stress t and the rate of twist u (in
degrees per inch) if G ⫽ 12 ⫻ 106 psi, t ⫽ 0.2 in., a ⫽ 3 in.,
and b ⫽ 2 in. (Note: See Appendix D, Case 16, for the properties of an ellipse.)
t
2b
2a
PROB. 3.10-7
3.10-8 A torque T is applied to a thin-walled tube having a
Tube (1)
Bar (2)
cross section in the shape of a regular hexagon with
constant wall thickness t and side length b (see figure).
Obtain formulas for the shear stress t and the rate of
twist u.
t
PROB. 3.10-5
3.10-6 Calculate the shear stress t and the angle of twist f
(in degrees) for a steel tube (G ⫽ 76 GPa) having the cross
section shown in the figure. The tube has length L ⫽ 1.5 m
and is subjected to a torque T ⫽ 10 kN⭈m.
b
PROB. 3.10-8
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
262
CHAPTER 3 Torsion
3.10-9 Compare the angle of twist f1 for a thin-walled
circular tube (see figure) calculated from the approximate
theory for thin-walled bars with the angle of twist f2
calculated from the exact theory of torsion for circular bars.
(a) Express the ratio f1/f2 in terms of the nondimensional ratio b ⫽ r/t.
(b) Calculate the ratio of angles of twist for b ⫽ 5, 10,
and 20. What conclusion about the accuracy of the approximate theory do you draw from these results?
t
t
2 in.
2 in.
PROB. 3.10-11
r
3.10-12 A thin tubular shaft of circular cross section (see
figure) with inside diameter 100 mm is subjected to a
torque of 5000 N⭈m.
If the allowable shear stress is 42 MPa, determine the
required wall thickness t by using (a) the approximate
theory for a thin-walled tube, and (b) the exact torsion
theory for a circular bar.
★
C
PROB. 3.10-9
*3.10-10 A thin-walled rectangular tube has uniform thickness t and dimensions a ⫻ b to the median line of the cross
section (see figure on the next page).
How does the shear stress in the tube vary with the
ratio b ⫽ a/b if the total length Lm of the median line of the
cross section and the torque T remain constant?
From your results, show that the shear stress is
smallest when the tube is square (b ⫽ 1).
100 mm
t
PROB. 3.10-12
t
3.10-13 A long, thin-walled tapered tube AB of circular
cross section (see figure) is subjected to a torque T. The
tube has length L and constant wall thickness t. The diameter to the median lines of the cross sections at the ends A
and B are dA and dB, respectively.
Derive the following formula for the angle of twist of
the tube:
★★
b
a
PROB. 3.10-10
冢
冣
*3.10-11 A tubular aluminum bar (G ⫽ 4 ⫻ 10 psi)
2TL dA ⫹ dB
ᎏ
f ⫽ ᎏᎏ ᎏ
pGt
d A2 d 2B
of square cross section (see figure) with outer dimensions
2 in. ⫻ 2 in. must resist a torque T ⫽ 3000 lb-in.
Calculate the minimum required wall thickness tmin if
the allowable shear stress is 4500 psi and the allowable rate
of twist is 0.01 rad/ft.
Hint: If the angle of taper is small, we may obtain approximate results by applying the formulas for a thin-walled
prismatic tube to a differential element of the tapered tube
and then integrating along the axis of the tube.
6
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 3 Problems
T
Stress Concentrations in Torsion
B
A
T
The problems for Section 3.11 are to be solved by
considering the stress-concentration factors.
3.11-1 A stepped shaft consisting of solid circular segments
L
t
t
dA
263
having diameters D1 ⫽ 2.0 in. and D2 ⫽ 2.4 in. (see figure) is
subjected to torques T. The radius of the fillet is R ⫽ 0.1 in.
If the allowable shear stress at the stress concentration
is 6000 psi, what is the maximum permissible torque Tmax?
dB
PROB. 3.10-13
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
4
Shear Forces and
Bending Moments
4.1 INTRODUCTION
FIG. 4-1 Examples of beams subjected to
lateral loads
Structural members are usually classified according to the types of loads
that they support. For instance, an axially loaded bar supports forces
having their vectors directed along the axis of the bar, and a bar in torsion supports torques (or couples) having their moment vectors directed
along the axis. In this chapter, we begin our study of beams (Fig. 4-1),
which are structural members subjected to lateral loads, that is, forces or
moments having their vectors perpendicular to the axis of the bar.
The beams shown in Fig. 4-1 are classified as planar structures
because they lie in a single plane. If all loads act in that same plane, and
if all deflections (shown by the dashed lines) occur in that plane, then
we refer to that plane as the plane of bending.
In this chapter we discuss shear forces and bending moments in
beams, and we will show how these quantities are related to each other
and to the loads. Finding the shear forces and bending moments is an
essential step in the design of any beam. We usually need to know not
only the maximum values of these quantities, but also the manner in
which they vary along the axis. Once the shear forces and bending
moments are known, we can find the stresses, strains, and deflections, as
discussed later in Chapters 5, 6, and 9.
4.2 TYPES OF BEAMS, LOADS, AND REACTIONS
Beams are usually described by the manner in which they are supported.
For instance, a beam with a pin support at one end and a roller support at
the other (Fig. 4-2a) is called a simply supported beam or a simple
beam. The essential feature of a pin support is that it prevents translation
at the end of a beam but does not prevent rotation. Thus, end A of the
264
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 4.2 Types of Beams, Loads, and Reactions
P1
P2
q
a
HA A
B
a
RA
c
RB
b
L
(a)
P3
q2
12
HA
MA
q1
5
A
B
a
b
RA
L
(b)
P4
A
RA
B
a
M1
C
RB
L
(c)
FIG. 4-2 Types of beams: (a) simple
beam, (b) cantilever beam, and (c) beam
with an overhang
265
beam of Fig. 4-2a cannot move horizontally or vertically but the axis of
the beam can rotate in the plane of the figure. Consequently, a pin support
is capable of developing a force reaction with both horizontal and vertical
components (HA and RA), but it cannot develop a moment reaction.
At end B of the beam (Fig. 4-2a) the roller support prevents translation in the vertical direction but not in the horizontal direction; hence
this support can resist a vertical force (RB) but not a horizontal force. Of
course, the axis of the beam is free to rotate at B just as it is at A. The
vertical reactions at roller supports and pin supports may act either
upward or downward, and the horizontal reaction at a pin support may
act either to the left or to the right. In the figures, reactions are indicated
by slashes across the arrows in order to distinguish them from loads, as
explained previously in Section 1.8.
The beam shown in Fig. 4-2b, which is fixed at one end and free at
the other, is called a cantilever beam. At the fixed support (or clamped
support) the beam can neither translate nor rotate, whereas at the free
end it may do both. Consequently, both force and moment reactions may
exist at the fixed support.
The third example in the figure is a beam with an overhang (Fig.
4-2c). This beam is simply supported at points A and B (that is, it has a
pin support at A and a roller support at B) but it also projects beyond the
support at B. The overhanging segment BC is similar to a cantilever
beam except that the beam axis may rotate at point B.
When drawing sketches of beams, we identify the supports by
conventional symbols, such as those shown in Fig. 4-2. These symbols
indicate the manner in which the beam is restrained, and therefore they
also indicate the nature of the reactive forces and moments. However,
the symbols do not represent the actual physical construction. For
instance, consider the examples shown in Fig. 4-3 on the next page. Part
(a) of the figure shows a wide-flange beam supported on a concrete wall
and held down by anchor bolts that pass through slotted holes in the
lower flange of the beam. This connection restrains the beam against
vertical movement (either upward or downward) but does not prevent
horizontal movement. Also, any restraint against rotation of the longitudinal axis of the beam is small and ordinarily may be disregarded.
Consequently, this type of support is usually represented by a roller, as
shown in part (b) of the figure.
The second example (Fig. 4-3c) is a beam-to-column connection in
which the beam is attached to the column flange by bolted angles. This
type of support is usually assumed to restrain the beam against horizontal
and vertical movement but not against rotation (restraint against rotation is
slight because both the angles and the column can bend). Thus, this
connection is usually represented as a pin support for the beam (Fig. 4-3d).
The last example (Fig. 4-3e) is a metal pole welded to a base plate
that is anchored to a concrete pier embedded deep in the ground. Since
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
266
CHAPTER 4 Shear Forces and Bending Moments
@@
ÀÀ
;;
;;;
@@@
ÀÀÀ
@@;;
ÀÀ
;;
@@@
ÀÀÀ
;;;
ÀÀ
@@
Slotted hole
Beam
Anchor
bolt
Bearing
plate
Concrete
wall
(a)
Beam
(b)
Column
Beam
Beam
(c)
(d)
À@;
ÀÀ;À@
@@
;;
ÀÀÀ
@@@
;;;
¢¢;@ÀQ¢
QQ
ÀÀ
@@
;;
Pole
Base plate
Pole
Concrete pier
(e)
(f)
FIG. 4-3 Beam supported on a wall:
(a) actual construction, and
(b) representation as a roller support.
Beam-to-column connection: (c) actual
construction, and (d) representation as a
pin support. Pole anchored to a concrete
pier: (e) actual construction, and
(f) representation as a fixed support
the base of the pole is fully restrained against both translation and rotation, it is represented as a fixed support (Fig. 4-3f ).
The task of representing a real structure by an idealized model, as
illustrated by the beams shown in Fig. 4-2, is an important aspect of
engineering work. The model should be simple enough to facilitate
mathematical analysis and yet complex enough to represent the actual
behavior of the structure with reasonable accuracy. Of course, every
model is an approximation to nature. For instance, the actual supports of
a beam are never perfectly rigid, and so there will always be a small
amount of translation at a pin support and a small amount of rotation at
a fixed support. Also, supports are never entirely free of friction, and so
there will always be a small amount of restraint against translation at a
roller support. In most circumstances, especially for statically determinate
beams, these deviations from the idealized conditions have little effect
on the action of the beam and can safely be disregarded.
Types of Loads
Several types of loads that act on beams are illustrated in Fig. 4-2. When
a load is applied over a very small area it may be idealized as a concentrated load, which is a single force. Examples are the loads P1, P2,
P3, and P4 in the figure. When a load is spread along the axis of a beam,
it is represented as a distributed load, such as the load q in part (a) of the
figure. Distributed loads are measured by their intensity, which is
expressed in units of force per unit distance (for example, newtons per
meter or pounds per foot). A uniformly distributed load, or uniform
load, has constant intensity q per unit distance (Fig. 4-2a). A varying
load has an intensity that changes with distance along the axis; for
instance, the linearly varying load of Fig. 4-2b has an intensity that
varies linearly from q1 to q2. Another kind of load is a couple, illustrated
by the couple of moment M1 acting on the overhanging beam (Fig. 4-2c).
As mentioned in Section 4.1, we assume in this discussion that the
loads act in the plane of the figure, which means that all forces must
have their vectors in the plane of the figure and all couples must have
their moment vectors perpendicular to the plane of the figure. Furthermore, the beam itself must be symmetric about that plane, which means
that every cross section of the beam must have a vertical axis of symmetry. Under these conditions, the beam will deflect only in the plane of
bending (the plane of the figure).
Reactions
Finding the reactions is usually the first step in the analysis of a beam.
Once the reactions are known, the shear forces and bending moments
can be found, as described later in this chapter. If a beam is supported in
a statically determinate manner, all reactions can be found from freebody diagrams and equations of equilibrium.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 4.2 Types of Beams, Loads, and Reactions
P1
P2
q
a
HA A
B
a
RA
c
RB
b
L
(a)
267
As an example, let us determine the reactions of the simple beam
AB of Fig. 4-2a. This beam is loaded by an inclined force P1, a vertical
force P2, and a uniformly distributed load of intensity q. We begin by
noting that the beam has three unknown reactions: a horizontal force HA
at the pin support, a vertical force RA at the pin support, and a vertical
force RB at the roller support. For a planar structure, such as this beam,
we know from statics that we can write three independent equations of
equilibrium. Thus, since there are three unknown reactions and three
equations, the beam is statically determinate.
The equation of horizontal equilibrium is
Fhoriz 0
FIG. 4-2a Simple beam. (Repeated)
HA ⫺ P1 cos a 0
from which we get
HA P1 cos a
This result is so obvious from an inspection of the beam that ordinarily
we would not bother to write the equation of equilibrium.
To find the vertical reactions RA and RB we write equations of
moment equilibrium about points B and A, respectively, with counterclockwise moments being positive:
MB 0
⫺RAL (P1 sin a)(L a) P2(L b) qc2/2 0
MA 0
RBL (P1 sin a)(a) P2b qc(L c/2) 0
Solving for RA and RB, we get
(P1 sin a)(L a)
P2(L b)
qc2
RA
L
L
2L
(P1 sin a)(a)
Pb
qc(L c/2)
RB
2
L
L
L
P3
q2
12
HA
MA
q1
5
A
B
a
b
RA
As a check on these results we can write an equation of equilibrium in
the vertical direction and verify that it reduces to an identity.
As a second example, consider the cantilever beam of Fig. 4-2b.
The loads consist of an inclined force P3 and a linearly varying distributed load. The latter is represented by a trapezoidal diagram of load
intensity that varies from q1 to q2. The reactions at the fixed support are
a horizontal force HA, a vertical force RA, and a couple MA. Equilibrium
of forces in the horizontal direction gives
5P
HA 3
13
and equilibrium in the vertical direction gives
L
(b)
FIG. 4-2b Cantilever beam. (Repeated)
12P3
q1 q2
RA
13
2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
b
268
CHAPTER 4 Shear Forces and Bending Moments
In finding this reaction we used the fact that the resultant of the distributed load is equal to the area of the trapezoidal loading diagram.
The moment reaction MA at the fixed support is found from an equation
of equilibrium of moments. In this example we will sum moments about
point A in order to eliminate both HA and RA from the moment equation.
Also, for the purpose of finding the moment of the distributed load, we
will divide the trapezoid into two triangles, as shown by the dashed line in
Fig. 4-2b. Each load triangle can be replaced by its resultant, which is a
force having its magnitude equal to the area of the triangle and having its
line of action through the centroid of the triangle. Thus, the moment about
point A of the lower triangular part of the load is
ᎏq2ᎏb L ⫺ ᎏ23bᎏ
1
in which q1b/2 is the resultant force (equal to the area of the triangular
load diagram) and L ⫺ 2b/3 is the moment arm (about point A) of the
resultant.
The moment of the upper triangular portion of the load is obtained
by a similar procedure, and the final equation of moment equilibrium
(counterclockwise is positive) is
MA 0
12P3
qb
2b
qb
b
MA ⫺ ᎏᎏ
a ⫺ ᎏ1ᎏ L ⫺ ᎏᎏ ⫺ ᎏ2ᎏ L ⫺ ᎏᎏ 0
13
2
3
2
3
from which
12P3a
qb
2b
qb
b
MA ᎏᎏ
1 L 2 L
13
2
3
2
3
P4
A
RA
B
a
M1
RB
L
(c)
FIG. 4-2c Beam with an overhang.
(Repeated)
C
Since this equation gives a positive result, the reactive moment MA acts
in the assumed direction, that is, counterclockwise. (The expressions for
RA and MA can be checked by taking moments about end B of the beam
and verifying that the resulting equation of equilibrium reduces to an
identity.)
The beam with an overhang (Fig. 4-2c) supports a vertical force
P4 and a couple of moment M1. Since there are no horizontal forces acting on the beam, the horizontal reaction at the pin support is nonexistent
and we do not need to show it on the free-body diagram. In arriving at
this conclusion, we made use of the equation of equilibrium for forces in
the horizontal direction. Consequently, only two independent equations
of equilibrium remain—either two moment equations or one moment
equation plus the equation for vertical equilibrium.
Let us arbitrarily decide to write two moment equations, the first for
moments about point B and the second for moments about point A, as
follows (counterclockwise moments are positive):
MB 0
MA 0
RAL P4(L a) M1 0
P4a RBL M1 0
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 4.3 Shear Forces and Bending Moments
269
Therefore, the reactions are
P4(L ⫺ a)
M
RA ᎏ
ᎏ 1
L
L
P4 a
L
M1
L
RB
Again, summation of forces in the vertical direction provides a check on
these results.
The preceding discussion illustrates how the reactions of statically
determinate beams are calculated from equilibrium equations. We have
intentionally used symbolic examples rather than numerical examples in
order to show how the individual steps are carried out.
4.3 SHEAR FORCES AND BENDING MOMENTS
P
m
A
B
n
x
(a)
P
M
A
V
x
(b)
M
V
B
(c)
FIG. 4-4 Shear force V and bending
moment M in a beam
When a beam is loaded by forces or couples, stresses and strains are
created throughout the interior of the beam. To determine these stresses
and strains, we first must find the internal forces and internal couples
that act on cross sections of the beam.
As an illustration of how these internal quantities are found,
consider a cantilever beam AB loaded by a force P at its free end (Fig.
4-4a). We cut through the beam at a cross section mn located at distance
x from the free end and isolate the left-hand part of the beam as a free body
(Fig. 4-4b). The free body is held in equilibrium by the force P and by
the stresses that act over the cut cross section. These stresses represent
the action of the right-hand part of the beam on the left-hand part. At
this stage of our discussion we do not know the distribution of the
stresses acting over the cross section; all we know is that the resultant of
these stresses must be such as to maintain equilibrium of the free body.
From statics, we know that the resultant of the stresses acting on the
cross section can be reduced to a shear force V and a bending moment
M (Fig. 4-4b). Because the load P is transverse to the axis of the beam,
no axial force exists at the cross section. Both the shear force and the
bending moment act in the plane of the beam, that is, the vector for the
shear force lies in the plane of the figure and the vector for the moment
is perpendicular to the plane of the figure.
Shear forces and bending moments, like axial forces in bars and
internal torques in shafts, are the resultants of stresses distributed over
the cross section. Therefore, these quantities are known collectively as
stress resultants.
The stress resultants in statically determinate beams can be calculated
from equations of equilibrium. In the case of the cantilever beam of Fig.
4-4a, we use the free-body diagram of Fig. 4-4b. Summing forces in the
vertical direction and also taking moments about the cut section, we get
Fvert 0
M 0
P V 0 or V P
M Px 0 or M Px
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
270
CHAPTER 4 Shear Forces and Bending Moments
where x is the distance from the free end of the beam to the cross section
where V and M are being determined. Thus, through the use of a freebody diagram and two equations of equilibrium, we can calculate the
shear force and bending moment without difficulty.
Sign Conventions
M V
M
V
V
M V
M
FIG. 4-5 Sign conventions for shear
force V and bending moment M
V
V
V
V
(a)
M
M
M
(b)
FIG. 4-6 Deformations (highly
exaggerated) of a beam element
caused by (a) shear forces, and
(b) bending moments
M
Let us now consider the sign conventions for shear forces and bending
moments. It is customary to assume that shear forces and bending
moments are positive when they act in the directions shown in Fig. 4-4b.
Note that the shear force tends to rotate the material clockwise and the
bending moment tends to compress the upper part of the beam and
elongate the lower part. Also, in this instance, the shear force acts downward and the bending moment acts counterclockwise.
The action of these same stress resultants against the right-hand part
of the beam is shown in Fig. 4-4c. The directions of both quantities are
now reversed—the shear force acts upward and the bending moment
acts clockwise. However, the shear force still tends to rotate the material
clockwise and the bending moment still tends to compress the upper part
of the beam and elongate the lower part.
Therefore, we must recognize that the algebraic sign of a stress
resultant is determined by how it deforms the material on which it acts,
rather than by its direction in space. In the case of a beam, a positive
shear force acts clockwise against the material (Figs. 4-4b and c) and a
negative shear force acts counterclockwise against the material. Also, a
positive bending moment compresses the upper part of the beam (Figs.
4-4b and c) and a negative bending moment compresses the lower part.
To make these conventions clear, both positive and negative shear
forces and bending moments are shown in Fig. 4-5. The forces and
moments are shown acting on an element of a beam cut out between two
cross sections that are a small distance apart.
The deformations of an element caused by both positive and negative shear forces and bending moments are sketched in Fig. 4-6. We see
that a positive shear force tends to deform the element by causing the
right-hand face to move downward with respect to the left-hand face,
and, as already mentioned, a positive bending moment compresses the
upper part of a beam and elongates the lower part.
Sign conventions for stress resultants are called deformation sign
conventions because they are based upon how the material is deformed.
For instance, we previously used a deformation sign convention in dealing
with axial forces in a bar. We stated that an axial force producing elongation (or tension) in a bar is positive and an axial force producing
shortening (or compression) is negative. Thus, the sign of an axial force
depends upon how it deforms the material, not upon its direction in space.
By contrast, when writing equations of equilibrium we use static sign
conventions, in which forces are positive or negative according to their
directions along the coordinate axes. For instance, if we are summing forces
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 4.3 Shear Forces and Bending Moments
P
m
A
in the y direction, forces acting in the positive direction of the y axis are
taken as positive and forces acting in the negative direction are taken as
negative.
As an example, consider Fig. 4-4b, which is a free-body diagram of
part of the cantilever beam. Suppose that we are summing forces in the
vertical direction and that the y axis is positive upward. Then the load P
is given a positive sign in the equation of equilibrium because it acts
upward. However, the shear force V (which is a positive shear force) is
given a negative sign because it acts downward (that is, in the negative
direction of the y axis). This example shows the distinction between the
deformation sign convention used for the shear force and the static sign
convention used in the equation of equilibrium.
The following examples illustrate the techniques for handling sign
conventions and determining shear forces and bending moments in
beams. The general procedure consists of constructing free-body diagrams and solving equations of equilibrium.
B
n
x
(a)
P
M
A
V
x
(b)
V
M
271
B
(c)
FIG. 4-4 (Repeated)
Example 4-1
P
M0
B
A
Solution
L
—
2
L
—
4
L
—
4
RB
RA
A simple beam AB supports two loads, a force P and a couple M0, acting as
shown in Fig. 4-7a.
Find the shear force V and bending moment M in the beam at cross sections
located as follows: (a) a small distance to the left of the midpoint of the beam,
and (b) a small distance to the right of the midpoint of the beam.
(a)
Reactions. The first step in the analysis of this beam is to find the reactions
RA and RB at the supports. Taking moments about ends B and A gives two equations of equilibrium, from which we find, respectively,
3P
4
M0
L
RA ᎏᎏ ⫺ ᎏᎏ
P
M
V
(b)
RA
FIG. 4-7 Example 4-1. Shear forces and
bending moments in a simple beam
P
4
M0
L
RB ᎏᎏ
(a)
(a) Shear force and bending moment to the left of the midpoint. We cut the
beam at a cross section just to the left of the midpoint and draw a free-body diagram of either half of the beam. In this example, we choose the left-hand half of
the beam as the free body (Fig. 4-7b). This free body is held in equilibrium by
the load P, the reaction RA, and the two unknown stress resultants—the shear
force V and the bending moment M, both of which are shown in their positive
directions (see Fig. 4-5). The couple M0 does not act on the free body because
the beam is cut to the left of its point of application.
Summing forces in the vertical direction (upward is positive) gives
Fvert 0
RA P V 0
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
272
CHAPTER 4 Shear Forces and Bending Moments
from which we get the shear force:
P
4
P
(b)
M0
B
A
L
—
4
M0
L
V RA ⫺ P
L
—
2
L
—
4
This result shows that when P and M0 act in the directions shown in Fig. 4-7a,
the shear force (at the selected location) is negative and acts in the opposite
direction to the positive direction assumed in Fig. 4-7b.
Taking moments about an axis through the cross section where the beam is
cut (see Fig. 4-7b) gives
RB
RA
(a)
M 0
2 4
L
P
M
L
RA P M 0
in which counterclockwise moments are taken as positive. Solving for the bending moment M, we get
V
2 4
L
(b)
RA
P
L
PL
8
M0
2
M RA P
M0
M
V
(c)
RA
FIG. 4-7 Example 4-1. Shear forces and
bending moments in a simple beam
(Repeated)
(c)
The bending moment M may be either positive or negative, depending upon the
magnitudes of the loads P and M0. If it is positive, it acts in the direction shown
in the figure; if it is negative, it acts in the opposite direction.
(b) Shear force and bending moment to the right of the midpoint. In this
case we cut the beam at a cross section just to the right of the midpoint and
again draw a free-body diagram of the part of the beam to the left of the cut section
(Fig. 4-7c). The difference between this diagram and the former one is that the
couple M0 now acts on the free body.
From two equations of equilibrium, the first for forces in the vertical
direction and the second for moments about an axis through the cut section, we
obtain
P
4
M0
L
V
PL
8
M0
2
M
(d,e)
These results show that when the cut section is shifted from the left to the right
of the couple M0, the shear force does not change (because the vertical forces
acting on the free body do not change) but the bending moment increases
algebraically by an amount equal to M0 (compare Eqs. c and e).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 4.3 Shear Forces and Bending Moments
273
Example 4-2
A cantilever beam that is free at end A and fixed at end B is subjected to a
distributed load of linearly varying intensity q (Fig. 4-8a). The maximum intensity
of the load occurs at the fixed support and is equal to q0.
Find the shear force V and bending moment M at distance x from the free
end of the beam.
q0
q
A
B
x
L
(a)
q
M
A
V
x
FIG. 4-8 Example 4-2. Shear force and
(b)
bending moment in a cantilever beam
Solution
Shear force. We cut through the beam at distance x from the left-hand end
and isolate part of the beam as a free body (Fig. 4-8b). Acting on the free body
are the distributed load q, the shear force V, and the bending moment M. Both
unknown quantities (V and M) are assumed to be positive.
The intensity of the distributed load at distance x from the end is
q0 x
q ᎏᎏ
L
(4-1)
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
274
CHAPTER 4 Shear Forces and Bending Moments
Therefore, the total downward load on the free body, equal to the area of the
triangular loading diagram (Fig. 4-8b), is
1 q0 x
q0 x 2
ᎏᎏ ᎏᎏ (x) ᎏᎏ
2 L
2L
From an equation of equilibrium in the vertical direction we find
q0 x 2
V
2L
(4-2a)
At the free end A (x 0) the shear force is zero, and at the fixed end B (x L)
the shear force has its maximum value:
q0 L
Vmax
2
(4-2b)
which is numerically equal to the total downward load on the beam. The minus
signs in Eqs. (4-2a) and (4-2b) show that the shear forces act in the opposite
direction to that pictured in Fig. 4-8b.
Bending moment. To find the bending moment M in the beam (Fig. 4-8b),
we write an equation of moment equilibrium about an axis through the cut
section. Recalling that the moment of a triangular load is equal to the area of the
loading diagram times the distance from its centroid to the axis of moments, we
obtain the following equation of equilibrium (counterclockwise moments are
positive):
M 0
x
1 q0 x
M (x) 0
3
2 L
from which we get
q0 x 3
M
6L
(4-3a)
At the free end of the beam (x 0), the bending moment is zero, and at the
fixed end (x L) the moment has its numerically largest value:
q0 L2
Mmax
6
(4-3b)
The minus signs in Eqs. (4-3a) and (4-3b) show that the bending moments act in
the opposite direction to that shown in Fig. 4-8b.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 4.3 Shear Forces and Bending Moments
275
Example 4-3
A simple beam with an overhang is supported at points A and B (Fig. 4-9a). A
uniform load of intensity q 200 lb/ft acts throughout the length of the beam
and a concentrated load P 14 k acts at a point 9 ft from the left-hand support.
The span length is 24 ft and the length of the overhang is 6 ft.
Calculate the shear force V and bending moment M at cross section D
located 15 ft from the left-hand support.
P = 14 k
9 ft
q = 200 lb/ft
A
B
D
15 ft
C
RB
RA
24 ft
6 ft
(a)
14 k
200 lb/ft
M
A
D
11 k
V
(b)
M
200 lb/ft
V
D
FIG. 4-9 Example 4-3. Shear force and
bending moment in a beam with an
overhang
B
C
9k
(c)
Solution
Reactions. We begin by calculating the reactions RA and RB from equations
of equilibrium for the entire beam considered as a free body. Thus, taking
moments about the supports at B and A, respectively, we find
RA 11 k
RB 9 k
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
276
CHAPTER 4 Shear Forces and Bending Moments
Shear force and bending moment at section D. Now we make a cut at
section D and construct a free-body diagram of the left-hand part of the beam
(Fig. 4-9b). When drawing this diagram, we assume that the unknown stress
resultants V and M are positive.
The equations of equilibrium for the free body are as follows:
Fvert 0
MD 0
11 k ⫺ 14 k ⫺ (0.200 k/ft)(15 ft) ⫺ V 0
⫺(11 k)(15 ft) (14 k)(6 ft) (0.200 k/ft)(15 ft)(7.5 ft) M 0
in which upward forces are taken as positive in the first equation and counterclockwise moments are taken as positive in the second equation. Solving these
equations, we get
V 6 k
M 58.5 k-ft
The minus sign for V means that the shear force is negative, that is, its direction
is opposite to the direction shown in Fig. 4-9b. The positive sign for M means
that the bending moment acts in the direction shown in the figure.
Alternative free-body diagram. Another method of solution is to obtain V
and M from a free-body diagram of the right-hand part of the beam (Fig. 4-9c).
When drawing this free-body diagram, we again assume that the unknown shear
force and bending moment are positive. The two equations of equilibrium are
Fvert 0
MD 0
V 9 k (0.200 k/ft)(15 ft) 0
M (9 k)(9 ft) (0.200 k/ft)(15 ft)(7.5 ft) 0
from which
V 6 k
M 58.5 k-ft
as before. As often happens, the choice between free-body diagrams is a matter
of convenience and personal preference.
4.4 RELATIONSHIPS BETWEEN LOADS, SHEAR FORCES, AND BENDING MOMENTS
We will now obtain some important relationships between loads, shear
forces, and bending moments in beams. These relationships are quite
useful when investigating the shear forces and bending moments
throughout the entire length of a beam, and they are especially
helpful when constructing shear-force and bending-moment diagrams
(Section 4.5).
As a means of obtaining the relationships, let us consider an element
of a beam cut out between two cross sections that are distance dx apart
(Fig. 4-10). The load acting on the top surface of the element may be a
distributed load, a concentrated load, or a couple, as shown in Figs. 4-10a,
b, and c, respectively. The sign conventions for these loads are as follows:
Distributed loads and concentrated loads are positive when they act
downward on the beam and negative when they act upward. A couple act-
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 4.4 Relationships Between Loads, Shear Forces, and Bending Moments
q
M
V
M + dM
dx
V + dV
(a)
P
M
V
M + M1
dx
V + V1
(b)
M0
M
V
M + M1
277
ing as a load on a beam is positive when it is counterclockwise and negative when it is clockwise. If other sign conventions are used, changes may
occur in the signs of the terms appearing in the equations derived in this
section.
The shear forces and bending moments acting on the sides of the
element are shown in their positive directions in Fig. 4-10. In general,
the shear forces and bending moments vary along the axis of the beam.
Therefore, their values on the right-hand face of the element may be different from their values on the left-hand face.
In the case of a distributed load (Fig. 4-10a) the increments in V and
M are infinitesimal, and so we denote them by dV and dM, respectively.
The corresponding stress resultants on the right-hand face are V dV
and M dM.
In the case of a concentrated load (Fig. 4-10b) or a couple (Fig. 4-10c)
the increments may be finite, and so they are denoted V1 and M1. The
corresponding stress resultants on the right-hand face are V V1 and
M M1.
For each type of loading we can write two equations of equilibrium
for the element—one equation for equilibrium of forces in the vertical
direction and one for equilibrium of moments. The first of these equations
gives the relationship between the load and the shear force, and the second
gives the relationship between the shear force and the bending moment.
Distributed Loads (Fig. 4-10a)
dx
V + V1
(c)
FIG. 4-10 Element of a beam used in
deriving the relationships between
loads, shear forces, and bending
moments. (All loads and stress
resultants are shown in their
positive directions.)
The first type of loading is a distributed load of intensity q, as shown in
Fig. 4-10a. We will consider first its relationship to the shear force and
second its relationship to the bending moment.
(1) Shear force. Equilibrium of forces in the vertical direction
(upward forces are positive) gives
Fvert 0
V q dx (V dV) 0
or
dV
q
dx
(4-4)
From this equation we see that the rate of change of the shear force at
any point on the axis of the beam is equal to the negative of the intensity
of the distributed load at that same point. (Note: If the sign convention
for the distributed load is reversed, so that q is positive upward instead
of downward, then the minus sign is omitted in the preceding equation.)
Some useful relations are immediately obvious from Eq. (4-4). For
instance, if there is no distributed load on a segment of the beam (that is,
if q 0), then dV/dx 0 and the shear force is constant in that part of
the beam. Also, if the distributed load is uniform along part of the beam
(q constant), then dV/dx is also constant and the shear force varies
linearly in that part of the beam.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
278
CHAPTER 4 Shear Forces and Bending Moments
As a demonstration of Eq. (4-4), consider the cantilever beam with a
linearly varying load that we discussed in Example 4-2 of the preceding
section (see Fig. 4-8). The load on the beam (from Eq. 4-1) is
q0 x
q ᎏᎏ
L
which is positive because it acts downward. Also, the shear force
(Eq. 4-2a) is
q0 x 2
V
2L
Taking the derivative dV/dx gives
d
dV
q0 x 2
q0 x
q
dx
dx
2L
L
which agrees with Eq. (4-4).
A useful relationship pertaining to the shear forces at two different
cross sections of a beam can be obtained by integrating Eq. (4-4) along
the axis of the beam. To obtain this relationship, we multiply both sides
of Eq. (4-4) by dx and then integrate between any two points A and B on
the axis of the beam; thus,
B
A
q dx
B
dV
A
(a)
where we are assuming that x increases as we move from point A to point
B. The left-hand side of this equation equals the difference (VB VA) of
the shear forces at B and A. The integral on the right-hand side represents
the area of the loading diagram between A and B, which in turn is equal
to the magnitude of the resultant of the distributed load acting between
points A and B. Thus, from Eq. (a) we get
q dx
B
VB VA
A
(area of the loading diagram between A and B)
(4-5)
In other words, the change in shear force between two points along the
axis of the beam is equal to the negative of the total downward load
between those points. The area of the loading diagram may be positive
(if q acts downward) or negative (if q acts upward).
Because Eq. (4-4) was derived for an element of the beam subjected
only to a distributed load (or to no load), we cannot use Eq. (4-4) at a point
where a concentrated load is applied (because the intensity of load is not
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 4.4 Relationships Between Loads, Shear Forces, and Bending Moments
q
M
V
M + dM
dx
(a)
FIG. 4-10a (Repeated)
279
defined for a concentrated load). For the same reason, we cannot use
Eq. (4-5) if a concentrated load P acts on the beam between points A and B.
(2) Bending moment. Let us now consider the moment equilibrium
of the beam element shown in Fig. 4-10a. Summing moments about an
axis at the left-hand side of the element (the axis is perpendicular to the
plane of the figure), and taking counterclockwise moments as positive, we
obtain
V + dV
M 0
2
dx
⫺M ⫺ q dx ᎏᎏ ⫺ (V dV)dx M dM 0
Discarding products of differentials (because they are negligible compared
to the other terms), we obtain the following relationship:
dM
V
dx
(4-6)
This equation shows that the rate of change of the bending moment at
any point on the axis of a beam is equal to the shear force at that same
point. For instance, if the shear force is zero in a region of the beam,
then the bending moment is constant in that same region.
Equation (4-6) applies only in regions where distributed loads (or no
loads) act on the beam. At a point where a concentrated load acts, a sudden change (or discontinuity) in the shear force occurs and the derivative
dM/dx is undefined at that point.
Again using the cantilever beam of Fig. 4-8 as an example, we recall
that the bending moment (Eq. 4-3a) is
q0 x 3
M
6L
Therefore, the derivative dM/dx is
d
dM
q0 x 3
q0 x 2
dx
dx
6L
2L
which is equal to the shear force in the beam (see Eq. 4-2a).
Integrating Eq. (4-6) between two points A and B on the beam axis
gives
dM V dx
B
A
B
A
(b)
The integral on the left-hand side of this equation is equal to the difference (MB MA) of the bending moments at points B and A. To interpret
the integral on the right-hand side, we need to consider V as a function
of x and visualize a shear-force diagram showing the variation of V with
x. Then we see that the integral on the right-hand side represents the area
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
280
CHAPTER 4 Shear Forces and Bending Moments
below the shear-force diagram between A and B. Therefore, we can
express Eq. (b) in the following manner:
V dx
B
MB ⫺ MA
A
(area of the shear-force diagram between A and B) (4-7)
This equation is valid even when concentrated loads act on the beam
between points A and B. However, it is not valid if a couple acts between A
and B. A couple produces a sudden change in the bending moment, and the
left-hand side of Eq. (b) cannot be integrated across such a discontinuity.
Concentrated Loads (Fig. 4-10b)
P
M
V
M + M1
Now let us consider a concentrated load P acting on the beam element
(Fig. 4-10b). From equilibrium of forces in the vertical direction,
we get
V ⫺ P ⫺ (V V1) 0 or V1 P
dx
(b)
FIG. 4-10b (Repeated)
V + V1
(4-8)
This result means that an abrupt change in the shear force occurs at any
point where a concentrated load acts. As we pass from left to right
through the point of load application, the shear force decreases by an
amount equal to the magnitude of the downward load P.
From equilibrium of moments about the left-hand face of the
element (Fig. 4-10b), we get
2
dx
M P (V V1)dx M M1 0
or
2
dx
M1 P V dx V1 dx
(c)
Since the length dx of the element is infinitesimally small, we see from
this equation that the increment M1 in the bending moment is also infinitesimally small. Thus, the bending moment does not change as we pass
through the point of application of a concentrated load.
Even though the bending moment M does not change at a concentrated load, its rate of change dM/dx undergoes an abrupt change. At the
left-hand side of the element (Fig. 4-10b), the rate of change of the
bending moment (see Eq. 4-6) is dM/dx V. At the right-hand side, the
rate of change is dM/dx V V1 V P. Therefore, at the point of
application of a concentrated load P, the rate of change dM/dx of the
bending moment decreases abruptly by an amount equal to P.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 4.5 Shear-Force and Bending-Moment Diagrams
M0
M
V
Loads in the Form of Couples (Fig. 4-10c)
M + M1
dx
(c)
FIG. 4-10c (Repeated)
281
V + V1
The last case to be considered is a load in the form of a couple M0 (Fig.
4-10c). From equilibrium of the element in the vertical direction we
obtain V1 0, which shows that the shear force does not change at the
point of application of a couple.
Equilibrium of moments about the left-hand side of the element gives
⫺M M0 (V V1)dx M M1 0
Disregarding terms that contain differentials (because they are negligible
compared to the finite terms), we obtain
M1 M0
(4-9)
This equation shows that the bending moment decreases by M0 as we
move from left to right through the point of load application. Thus, the
bending moment changes abruptly at the point of application of a
couple.
Equations (4-4) through (4-9) are useful when making a complete
investigation of the shear forces and bending moments in a beam, as
discussed in the next section.
4.5 SHEAR-FORCE AND BENDING-MOMENT DIAGRAMS
When designing a beam, we usually need to know how the shear forces
and bending moments vary throughout the length of the beam. Of special importance are the maximum and minimum values of these
quantities. Information of this kind is usually provided by graphs in
which the shear force and bending moment are plotted as ordinates and
the distance x along the axis of the beam is plotted as the abscissa. Such
graphs are called shear-force and bending-moment diagrams.
To provide a clear understanding of these diagrams, we will explain
in detail how they are constructed and interpreted for three basic loading
conditions—a single concentrated load, a uniform load, and several concentrated loads. In addition, Examples 4-4 to 4-7 at the end of the
section provide detailed illustration of the techniques for handling various kinds of loads, including the case of a couple acting as a load on a
beam.
Concentrated Load
Let us begin with a simple beam AB supporting a concentrated load P
(Fig. 4-11a). The load P acts at distance a from the left-hand support
and distance b from the right-hand support. Considering the entire beam
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
282
CHAPTER 4 Shear Forces and Bending Moments
as a free body, we can readily determine the reactions of the beam from
equilibrium; the results are
P
a
b
B
A
Pb
L
RA ᎏᎏ
x
L
RB
RA
(a)
M
V RA ᎏᎏ
V
x
RA
(b)
P
M
a
A
V
x
RA
(c)
V
Pb
—
L
(4-10a,b)
We now cut through the beam at a cross section to the left of the load P
and at distance x from the support at A. Then we draw a free-body
diagram of the left-hand part of the beam (Fig. 4-11b). From the
equations of equilibrium for this free body, we obtain the shear force V
and bending moment M at distance x from the support:
Pb
L
A
Pa
L
RB ᎏᎏ
Pbx
L
M RAx ᎏᎏ
(0 x a)
(4-11a,b)
These expressions are valid only for the part of the beam to the left of
the load P.
Next, we cut through the beam to the right of the load P (that is, in
the region a x L) and again draw a free-body diagram of the lefthand part of the beam (Fig. 4-11c). From the equations of equilibrium
for this free body, we obtain the following expressions for the shear
force and bending moment:
Pb
L
Pa
(a x L)
L
Pbx
M RAx P(x a) P(x a)
L
Pa
(a x L)
(L x)
L
V RA P P
(4-12a)
(4-12b)
Note that these equations are valid only for the right-hand part of the beam.
The equations for the shear forces and bending moments (Eqs. 4-11
and 4-12) are plotted below the sketches of the beam. Figure 4-11d is
the shear-force diagram and Fig. 4-11e is the bending-moment diagram.
From the first diagram we see that the shear force at end A of the
beam (x 0) is equal to the reaction RA. Then it remains constant to the
point of application of the load P. At that point, the shear force
decreases abruptly by an amount equal to the load P. In the right-hand
part of the beam, the shear force is again constant but equal numerically
to the reaction at B.
As shown in the second diagram, the bending moment in the lefthand part of the beam increases linearly from zero at the support to Pab/L
at the concentrated load (x a). In the right-hand part, the bending
moment is again a linear function of x, varying from Pab/L at x a to
zero at the support (x L). Thus, the maximum bending moment is
0
Pa
–—
L
(d)
Pab
-–—
L
M
0
(e)
FIG. 4-11 Shear-force and bending-
moment diagrams for a simple beam
with a concentrated load
Pab
L
Mmax
and occurs under the concentrated load.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(4-13)
SECTION 4.5 Shear-Force and Bending-Moment Diagrams
283
When deriving the expressions for the shear force and bending
moment to the right of the load P (Eqs. 4-12a and b), we considered the
equilibrium of the left-hand part of the beam (Fig. 4-11c). This free body is
acted upon by the forces RA and P in addition to V and M. It is slightly simpler in this particular example to consider the right-hand portion of the
beam as a free body, because then only one force (RB) appears in the equilibrium equations (in addition to V and M). Of course, the final results are
unchanged.
Certain characteristics of the shear-force and bending moment
diagrams (Figs. 4-11d and e) may now be seen. We note first that the
slope dV/dx of the shear-force diagram is zero in the regions 0 x a
and a x L, which is in accord with the equation dV/dx q (Eq. 4-4).
Also, in these same regions the slope dM/dx of the bending moment
diagram is equal to V (Eq. 4-6). To the left of the load P, the slope of the
moment diagram is positive and equal to Pb/L; to the right, it is negative
and equal to Pa/L. Thus, at the point of application of the load P there
is an abrupt change in the shear-force diagram (equal to the magnitude
of the load P) and a corresponding change in the slope of the bendingmoment diagram.
Now consider the area of the shear-force diagram. As we move
from x 0 to x a, the area of the shear-force diagram is (Pb/L)a, or
Pab/L. This quantity represents the increase in bending moment between
these same two points (see Eq. 4-7). From x a to x L, the area of
the shear-force diagram is Pab/L, which means that in this region the
bending moment decreases by that amount. Consequently, the bending
moment is zero at end B of the beam, as expected.
If the bending moments at both ends of a beam are zero, as is usually
the case with a simple beam, then the area of the shear-force diagram
between the ends of the beam must be zero provided no couples act on
the beam (see the discussion in Section 4.4 following Eq. 4-7).
As mentioned previously, the maximum and minimum values of the
shear forces and bending moments are needed when designing beams.
For a simple beam with a single concentrated load, the maximum shear
force occurs at the end of the beam nearest to the concentrated load and
the maximum bending moment occurs under the load itself.
Uniform Load
A simple beam with a uniformly distributed load of constant intensity q
is shown in Fig. 4-12a. Because the beam and its loading are symmetric,
we see immediately that each of the reactions (RA and RB) is equal to
qL/2. Therefore, the shear force and bending moment at distance x from
the left-hand end are
qL
2
x
qLx
qx2
M RAx qx
2
2
2
V RA qx qx
(4-14a)
(4-14b)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
284
CHAPTER 4 Shear Forces and Bending Moments
q
A
B
x
L
RB
RA
(a)
These equations, which are valid throughout the length of the beam, are
plotted as shear-force and bending moment diagrams in Figs. 4-12b and
c, respectively.
The shear-force diagram consists of an inclined straight line having
ordinates at x 0 and x L equal numerically to the reactions. The
slope of the line is ⫺q, as expected from Eq. (4-4). The bending-moment
diagram is a parabolic curve that is symmetric about the midpoint of the
beam. At each cross section the slope of the bending-moment diagram is
equal to the shear force (see Eq. 4-6):
qL
—
2
d qLx
dM
qx2
qL
ᎏᎏ ᎏᎏ ᎏᎏ ⫺ ᎏᎏ ᎏᎏ ⫺ qx V
dx 2
dx
2
2
V
0
(b)
qL
–—
2
qL 2
—–
8
M
0
(c)
FIG. 4-12 Shear-force and bending-
moment diagrams for a simple beam
with a uniform load
The maximum value of the bending moment occurs at the midpoint of
the beam where both dM/dx and the shear force V are equal to zero.
Therefore, we substitute x L/2 into the expression for M and obtain
qL2
8
(4-15)
Mmax ᎏᎏ
as shown on the bending-moment diagram.
The diagram of load intensity (Fig. 4-12a) has area qL, and according to Eq. (4-5) the shear force V must decrease by this amount as we
move along the beam from A to B. We can see that this is indeed the
case, because the shear force decreases from qL/ 2 to ⫺qL/2.
The area of the shear-force diagram between x 0 and x L/2 is
qL2/8, and we see that this area represents the increase in the bending
moment between those same two points (Eq. 4-7). In a similar manner,
the bending moment decreases by qL2/8 in the region from x L /2
to x L.
Several Concentrated Loads
If several concentrated loads act on a simple beam (Fig. 4-13a), expressions for the shear forces and bending moments may be determined for
each segment of the beam between the points of load application. Again
using free-body diagrams of the left-hand part of the beam and measuring the distance x from end A, we obtain the following equations for the
first segment of the beam:
V RA
M RAx
(0 x a1)
(4-16a,b)
For the second segment, we get
V RA P1
M RAx P1(x a1)
(a1 x a2) (4-17a,b)
For the third segment of the beam, it is advantageous to consider the
right-hand part of the beam rather than the left, because fewer loads act
on the corresponding free body. Hence, we obtain
V RB P3
M RB(L x) P3(L b3 x) (a2 x a3)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(4-18a)
(4-18b)
SECTION 4.5 Shear-Force and Bending-Moment Diagrams
a1
a2
P1
Finally, for the fourth segment of the beam, we obtain
P3
a3
P2
b3
V RB
A
M RB(L x)
(a3 x L)
(4-19a,b)
B
x
L
RB
RA
(a)
RA
V
285
P1
P2
0
P3{
– RB
M1 RAa1
(b)
M1
M2
Equations (4-16) through (4-19) can be used to construct the shear-force
and bending-moment diagrams (Figs. 4-13b and c).
From the shear-force diagram we note that the shear force is constant in each segment of the beam and changes abruptly at every load
point, with the amount of each change being equal to the load. Also, the
bending moment in each segment is a linear function of x, and therefore
the corresponding part of the bending-moment diagram is an inclined
straight line. To assist in drawing these lines, we obtain the bending
moments under the concentrated loads by substituting x a1,
x a2, and x a3 into Eqs. (4-16b), (4-17b), and (4-18b), respectively.
In this manner we obtain the following bending moments:
M3
M
0
(c)
FIG. 4-13 Shear-force and bendingmoment diagrams for a simple beam
with several concentrated loads
M2 RAa2 P1(a2 a1)
M3 RBb3
(4-20a,b,c)
Knowing these values, we can readily construct the bending-moment
diagram by connecting the points with straight lines.
At each discontinuity in the shear force, there is a corresponding
change in the slope dM/dx of the bending-moment diagram. Also, the
change in bending moment between two load points equals the area of
the shear-force diagram between those same two points (see Eq. 4-7).
For example, the change in bending moment between loads P1 and P2 is
M2 M1. Substituting from Eqs. (4-20a and b), we get
M2 M1 (RA P1)(a2 a1)
which is the area of the rectangular shear-force diagram between x a1
and x a2.
The maximum bending moment in a beam having only concentrated
loads must occur under one of the loads or at a reaction. To show this,
recall that the slope of the bending-moment diagram is equal to the
shear force. Therefore, whenever the bending moment has a maximum
or minimum value, the derivative dM/dx (and hence the shear force)
must change sign. However, in a beam with only concentrated loads, the
shear force can change sign only under a load.
If, as we proceed along the x axis, the shear force changes from
positive to negative (as in Fig. 4-13b), then the slope in the bending
moment diagram also changes from positive to negative. Therefore, we
must have a maximum bending moment at this cross section.
Conversely, a change in shear force from a negative to a positive value
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
286
CHAPTER 4 Shear Forces and Bending Moments
indicates a minimum bending moment. Theoretically, the shear-force
diagram can intersect the horizontal axis at several points, although this
is quite unlikely. Corresponding to each such intersection point, there is
a local maximum or minimum in the bending-moment diagram. The
values of all local maximums and minimums must be determined in
order to find the maximum positive and negative bending moments in a
beam.
General Comments
In our discussions we frequently use the terms “maximum” and “minimum” with their common meanings of “largest” and “smallest.”
Consequently, we refer to “the maximum bending moment in a beam”
regardless of whether the bending-moment diagram is described by a
smooth, continuous function (as in Fig. 4-12c) or by a series of lines (as
in Fig. 4-13c).
Furthermore, we often need to distinguish between positive and
negative quantities. Therefore, we use expressions such as “maximum
positive moment” and “maximum negative moment.” In both of these
cases, the expression refers to the numerically largest quantity; that is,
the term “maximum negative moment” really means “numerically
largest negative moment.” Analogous comments apply to other beam
quantities, such as shear forces and deflections.
The maximum positive and negative bending moments in a beam
may occur at the following places: (1) a cross section where a concentrated
load is applied and the shear force changes sign (see Figs. 4-11 and 4-13),
(2) a cross section where the shear force equals zero (see Fig. 4-12),
(3) a point of support where a vertical reaction is present, and (4) a cross
section where a couple is applied. The preceding discussions and the
following examples illustrate all of these possibilities.
When several loads act on a beam, the shear-force and bendingmoment diagrams can be obtained by superposition (or summation) of
the diagrams obtained for each of the loads acting separately. For
instance, the shear-force diagram of Fig. 4-13b is actually the sum of
three separate diagrams, each of the type shown in Fig. 4-11d for a
single concentrated load. We can make an analogous comment for the
bending-moment diagram of Fig. 4-13c. Superposition of shear-force
and bending-moment diagrams is permissible because shear forces and
bending moments in statically determinate beams are linear functions of
the applied loads.
Computer programs are readily available for drawing shear-force
and bending-moment diagrams. After you have developed an understanding of the nature of the diagrams by constructing them manually,
you should feel secure in using computer programs to plot the diagrams
and obtain numerical results.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 4.5 Shear-Force and Bending-Moment Diagrams
287
Example 4-4
Draw the shear-force and bending-moment diagrams for a simple beam with a
uniform load of intensity q acting over part of the span (Fig. 4-14a).
q
A
B
Solution
Reactions. We begin the analysis by determining the reactions of the beam
from a free-body diagram of the entire beam (Fig. 4-14a). The results are
x
a
b
RA
RA ᎏ
RB
(a)
RA
V
0
x1
(b)
qb(b 2c)
2L
c
L
– RB
qb(b 2a)
2L
RB
(4-21a,b)
Shear forces and bending moments. To obtain the shear forces and bending
moments for the entire beam, we must consider the three segments of the beam
individually. For each segment we cut through the beam to expose the shear
force V and bending moment M. Then we draw a free-body diagram containing
V and M as unknown quantities. Lastly, we sum forces in the vertical direction
to obtain the shear force and take moments about the cut section to obtain the
bending moment. The results for all three segments are as follows:
V RA
M RAx
(0 x a)
(4-22a,b)
Mmax
M
0
V RA q(x a)
x1
V RB
(c)
FIG. 4-14 Example 4-4. Simple beam
with a uniform load over part of the
span
q(x a)2
M RAx
2
M RB(L x)
(a x a b)
(a b x L)
(4-23a,b)
(4-24a,b)
These equations give the shear force and bending moment at every cross section
of the beam. As a partial check on these results, we can apply Eq. (4-4) to the
shear forces and Eq. (4-6) to the bending moments and verify that the equations
are satisfied.
We now construct the shear-force and bending-moment diagrams (Figs.
4-14b and c) from Eqs. (4-22) through (4-24). The shear-force diagram consists
of horizontal straight lines in the unloaded regions of the beam and an inclined
straight line with negative slope in the loaded region, as expected from the equation dV/dx q.
The bending-moment diagram consists of two inclined straight lines in the
unloaded portions of the beam and a parabolic curve in the loaded portion. The
inclined lines have slopes equal to RA and RB, respectively, as expected from
the equation dM/dx V. Also, each of these inclined lines is tangent to the
parabolic curve at the point where it meets the curve. This conclusion follows
from the fact that there are no abrupt changes in the magnitude of the shear
force at these points. Hence, from the equation dM/dx V, we see that the slope
of the bending-moment diagram does not change abruptly at these points.
Maximum bending moment. The maximum moment occurs where the shear
force equals zero. This point can be found by setting the shear force V (from
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
288
CHAPTER 4 Shear Forces and Bending Moments
Eq. 4-23a) equal to zero and solving for the value of x, which we will denote by
x1. The result is
b
2L
x1 a (b 2c)
(4-25)
Now we substitute x1 into the expression for the bending moment (Eq.
4-23b) and solve for the maximum moment. The result is
qb
8L
Mmax 2 (b 2c)(4aL 2bc b2)
(4-26)
The maximum bending moment always occurs within the region of the uniform
load, as shown by Eq. (4-25).
Special cases. If the uniform load is symmetrically placed on the beam
(a c), then we obtain the following simplified results from Eqs. (4-25) and
(4-26):
L
x1
2
qb(2L b)
Mmax
8
(4-27a,b)
If the uniform load extends over the entire span, then b L and Mmax qL2/8,
which agrees with Fig. 4-12 and Eq. (4-15).
Example 4-5
Draw the shear-force and bending-moment diagrams for a cantilever beam with
two concentrated loads (Fig. 4-15a).
P2
P1
B
A
MB
0
V
0
M
–P1
x
a
–P1 – P2
b
L
(b)
RB
(a)
–P1a
–P1L – P2 b
(c)
FIG. 4-15 Example 4-5. Cantilever beam
with two concentrated loads
Solution
Reactions. From the free-body diagram of the entire beam we find the
vertical reaction RB (positive when upward) and the moment reaction MB
(positive when clockwise):
RB P1 P2
MB P1L P2b
(4-28a,b)
Shear forces and bending moments. We obtain the shear forces and bending
moments by cutting through the beam in each of the two segments, drawing the
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 4.5 Shear-Force and Bending-Moment Diagrams
289
corresponding free-body diagrams, and solving the equations of equilibrium.
Again measuring the distance x from the left-hand end of the beam, we get
V P1
V P1 P2
(0 x a)
M P1x
M P1x P2(x a)
(a x L)
(4-29a,b)
(4-30a,b)
The corresponding shear-force and bending-moment diagrams are shown in
Figs. 4-15b and c. The shear force is constant between the loads and reaches its
maximum numerical value at the support, where it is equal numerically to the
vertical reaction RB (Eq. 4-28a).
The bending-moment diagram consists of two inclined straight lines, each
having a slope equal to the shear force in the corresponding segment of the
beam. The maximum bending moment occurs at the support and is equal
numerically to the moment reaction MB (Eq. 4-28b). It is also equal to the area
of the entire shear-force diagram, as expected from Eq. (4-7).
Example 4-6
A cantilever beam supporting a uniform load of constant intensity q is shown in
Fig. 4-16a. Draw the shear-force and bending-moment diagrams for this beam.
q
A
MB
x
L
(a)
V
Solution
Reactions. The reactions RB and MB at the fixed support are obtained from
equations of equilibrium for the entire beam; thus,
B
RB
RB qL
qL2
2
Shear forces and bending moments. These quantities are found by cutting
through the beam at distance x from the free end, drawing a free-body diagram
of the left-hand part of the beam, and solving the equations of equilibrium. By
this means we obtain
0
– qL
V qx
qx2
2
0
qL2
– ——
2
The shear-force and bending-moment diagrams are obtained by plotting these
equations (see Figs. 4-16b and c). Note that the slope of the shear-force diagram
is equal to q (see Eq. 4-4) and the slope of the bending-moment diagram is
equal to V (see Eq. 4-6).
The maximum values of the shear force and bending moment occur at the
fixed support where x L:
(c)
FIG. 4-16 Example 4-6. Cantilever beam
with a uniform load
(4-32a,b)
M
(b)
M
(4-31a,b)
MB ᎏᎏ
Vmax ql
qL2
2
Mmax
(4-33a,b)
These values are consistent with the values of the reactions RB and MB (Eqs.
4-31a and b).
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
290
CHAPTER 4 Shear Forces and Bending Moments
Alternative solution. Instead of using free-body diagrams and equations
of equilibrium, we can determine the shear forces and bending moments by
integrating the differential relationships between load, shear force, and bending
moment. The shear force V at distance x from the free end A is obtained from
the load by integrating Eq. (4-5), as follows:
q dx qx
x
V ⫺ VA V ⫺ 0 V
(a)
0
which agrees with the previous result (Eq. 4-32a).
The bending moment M at distance x from the end is obtained from the
shear force by integrating Eq. (4-7):
M MA M 0 M
V dx qx dx q2x
x
x
0
0
2
(b)
which agrees with Eq. 4-32b.
Integrating the differential relationships is quite simple in this example
because the loading pattern is continuous and there are no concentrated loads or
couples in the regions of integration. If concentrated loads or couples were
present, discontinuities in the V and M diagrams would exist, and we cannot
integrate Eq. (4-5) through a concentrated load nor can we integrate Eq. (4-7)
through a couple (see Section 4.4).
Example 4-7
A beam ABC with an overhang at the left-hand end is shown in Fig. 4-17a. The
beam is subjected to a uniform load of intensity q 1.0 k/ft on the overhang AB
and a counterclockwise couple M0 12.0 k-ft acting midway between the
supports at B and C.
Draw the shear-force and bending-moment diagrams for this beam.
q = 1.0 k/ft
B
A
M0 = 12.0 k-ft
C
+1.25
V(k)
0
–4.0
b=
4 ft
RB
L
—
= 8 ft
2
(a)
L
—
= 8 ft
2
(b)
RC
M(k-ft)
0
FIG. 4-17 Example 4-7. Beam with an
overhang
+2.0
–8.0
–10.0
(c)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 4.5 Shear-Force and Bending-Moment Diagrams
291
Solution
Reactions. We can readily calculate the reactions RB and RC from a freebody diagram of the entire beam (Fig. 4-17a). In so doing, we find that RB is
upward and RC is downward, as shown in the figure. Their numerical values are
RB 5.25 k
RC 1.25 k
Shear forces. The shear force equals zero at the free end of the beam and
equals ⫺qb (or ⫺4.0 k) just to the left of support B. Since the load is uniformly
distributed (that is, q is constant), the slope of the shear diagram is constant and
equal to ⫺q (from Eq. 4-4). Therefore, the shear diagram is an inclined straight
line with negative slope in the region from A to B (Fig. 4-17b).
Because there are no concentrated or distributed loads between the supports, the shear-force diagram is horizontal in this region. The shear force is
equal to the reaction RC, or 1.25 k, as shown in the figure. (Note that the shear
force does not change at the point of application of the couple M0.)
The numerically largest shear force occurs just to the left of support B and
equals ⫺4.0 k.
Bending moments. The bending moment is zero at the free end and
decreases algebraically (but increases numerically) as we move to the right until
support B is reached. The slope of the moment diagram, equal to the value of
the shear force (from Eq. 4-6), is zero at the free end and ⫺4.0 k just to the left
of support B. The diagram is parabolic (second degree) in this region, with the
vertex at the end of the beam. The moment at point B is
2
qb
1
MB (1.0 k/ft)(4.0 ft)2 8.0 k-ft
2
2
which is also equal to the area of the shear-force diagram between A and B (see
Eq. 4-7).
The slope of the bending-moment diagram from B to C is equal to the shear
force, or 1.25 k. Therefore, the bending moment just to the left of the couple M0 is
8.0 k-ft (1.25 k)(8.0 ft) 2.0 k-ft
as shown on the diagram. Of course, we can get this same result by cutting
through the beam just to the left of the couple, drawing a free-body diagram,
and solving the equation of moment equilibrium.
The bending moment changes abruptly at the point of application of the
couple M0, as explained earlier in connection with Eq. (4-9). Because the couple
acts counterclockwise, the moment decreases by an amount equal to M0. Thus,
the moment just to the right of the couple M0 is
2.0 k-ft 12.0 k-ft 10.0 k-ft
From that point to support C the diagram is again a straight line with slope
equal to 1.25 k. Therefore, the bending moment at the support is
10.0 k-ft (1.25 k)(8.0 ft) 0
as expected.
Maximum and minimum values of the bending moment occur where the
shear force changes sign and where the couple is applied. Comparing the various high and low points on the moment diagram, we see that the numerically
largest bending moment equals 10.0 k-ft and occurs just to the right of the
couple M0.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
292
CHAPTER 4 Shear Forces and Bending Moments
PROBLEMS CHAPTER 4
Shear Forces and Bending Moments
4.3-5 Determine the shear force V and bending moment M
4.3-1 Calculate the shear force V and bending moment M
at a cross section located 16 ft from the left-hand end A of
the beam with an overhang shown in the figure.
at a cross section just to the left of the 1600-lb load acting
on the simple beam AB shown in the figure.
400 lb/ft
800 lb
1600 lb
A
200 lb/ft
B
A
C
B
10 ft
30 in.
60 in.
120 in.
30 in.
10 ft
6 ft
6 ft
PROB. 4.3-5
4.3-6 The beam ABC shown in the figure is simply sup-
PROB. 4.3-1
4.3-2 Determine the shear force V and bending moment M at
the midpoint C of the simple beam AB shown in the figure.
6.0 kN
2.0 kN/m
C
A
1.0 m
1.0 m
4.0 m
B
2.0 m
ported at A and B and has an overhang from B to C. The
loads consist of a horizontal force P1 4.0 kN acting at the
end of a vertical arm and a vertical force P2 8.0 kN
acting at the end of the overhang.
Determine the shear force V and bending moment M
at a cross section located 3.0 m from the left-hand support.
(Note: Disregard the widths of the beam and vertical arm
and use centerline dimensions when making calculations.)
P1 = 4.0 kN
P2 = 8.0 kN
PROB. 4.3-2
4.3-3 Determine the shear force V and bending moment M
at the midpoint of the beam with overhangs (see figure).
Note that one load acts downward and the other upward.
P
1.0 m
A
B
P
4.0 m
C
1.0 m
PROB. 4.3-6
b
L
b
4.3-7 The beam ABCD shown in the figure has overhangs
PROB. 4.3-3
4.3-4 Calculate the shear force V and bending moment M
at a cross section located 0.5 m from the fixed support of
the cantilever beam AB shown in the figure.
4.0 kN
1.5 kN/m
A
1.0 m
PROB. 4.3-4
at each end and carries a uniform load of intensity q.
For what ratio b/L will the bending moment at the midpoint of the beam be zero?
q
A
B
B
1.0 m
b
2.0 m
PROB. 4.3-7
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
D
C
L
b
@À;;À@
;À@
293
CHAPTER 4 Problems
4.3-8 At full draw, an archer applies a pull of 130 N to the
bowstring of the bow shown in the figure. Determine the
bending moment at the midpoint of the bow.
1600 N/m
2.6 m
70°
PROB. 4.3-10
4.3-11 A beam ABCD with a vertical arm CE is supported
as a simple beam at A and D (see figure). A cable passes
over a small pulley that is attached to the arm at E. One end
of the cable is attached to the beam at point B.
What is the force P in the cable if the bending moment
in the beam just to the left of point C is equal numerically
to 640 lb-ft? (Note: Disregard the widths of the beam and
vertical arm and use centerline dimensions when making
calculations.)
1400 mm
E
A
M
B
A
O
N
V
r
u
P
C
P
Cable
4.3-9 A curved bar ABC is subjected to loads in the form
of two equal and opposite forces P, as shown in the figure.
The axis of the bar forms a semicircle of radius r.
Determine the axial force N, shear force V, and bending moment M acting at a cross section defined by the angle
u.
P
1.0 m
2.6 m
350 mm
PROB. 4.3-8
900 N/m
P
u
8 ft
B
6 ft
C
6 ft
D
6 ft
PROB. 4.3-11
4.3-12 A simply supported beam AB supports a trapezoidally
distributed load (see figure). The intensity of the load
varies linearly from 50 kN/m at support A to 30 kN/m at
support B.
Calculate the shear force V and bending moment M at
the midpoint of the beam.
A
PROB. 4.3-9
50 kN/m
30 kN/m
4.3-10 Under cruising conditions the distributed load acting on the wing of a small airplane has the idealized
variation shown in the figure.
Calculate the shear force V and bending moment M at
the inboard end of the wing.
A
B
3m
PROB. 4.3-12
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
294
CHAPTER 4 Shear Forces and Bending Moments
y
4.3-13 Beam ABCD represents a reinforced-concrete
foundation beam that supports a uniform load of intensity
q1 3500 lb/ft (see figure). Assume that the soil pressure
on the underside of the beam is uniformly distributed with
intensity q2.
(a) Find the shear force VB and bending moment MB at
point B.
(b) Find the shear force Vm and bending moment Mm at
the midpoint of the beam.
L
b
a
B
W
C
A
PROB. 4.3-15
D
q2
8.0 ft
Shear-Force and Bending-Moment Diagrams
3.0 ft
PROB. 4.3-13
4.3-14 The simply-supported beam ABCD is loaded by a
weight W 27 kN through the arrangement shown in the
figure. The cable passes over a small frictionless pulley at
B and is attached at E to the end of the vertical arm.
Calculate the axial force N, shear force V, and bending
moment M at section C, which is just to the left of the
vertical arm. (Note: Disregard the widths of the beam and
vertical arm and use centerline dimensions when making
calculations.)
When solving the problems for Section 4.5, draw the shearforce and bending-moment diagrams approximately to
scale and label all critical ordinates, including the
maximum and minimum values.
Probs 4.5-1 through 4.5-10 are symbolic problems and
Probs. 4.5-11 through 4.5-24 are numerical problems. The
remaining problems (4.5-25 through 4.5-30) involve
specialized topics, such as optimization, beams with
hinges, and moving loads.
4.5-1 Draw the shear-force and bending-moment diagrams
for a simple beam AB supporting two equal concentrated
loads P (see figure).
E
Cable
A
B
W
x
q1 = 3500 lb/ft
3.0 ft
c
a
P
A
1.5 m
C
P
a
B
D
L
2.0 m
2.0 m
2.0 m
PROB. 4.5-1
4.5-2 A simple beam AB is subjected to a counterclockwise couple of moment M0 acting at distance a from the
left-hand support (see figure).
Draw the shear-force and bending-moment diagrams
for this beam.
W = 27 kN
PROB. 4.3-14
4.3-15 The centrifuge shown in the figure rotates in a
horizontal plane (the xy plane) on a smooth surface about
the z axis (which is vertical) with an angular acceleration a.
Each of the two arms has weight w per unit length and supports a weight W 2.0wL at its end.
Derive formulas for the maximum shear force and
maximum bending moment in the arms, assuming b L /9
and c L /10.
★
M0
A
B
a
L
PROB. 4.5-2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
295
CHAPTER 4 Problems
4.5-3 Draw the shear-force and bending-moment diagrams
for a cantilever beam AB carrying a uniform load of intensity q over one-half of its length (see figure).
q
A
4.5-7 A simply supported beam ABC is loaded by a vertical load P acting at the end of a bracket BDE (see figure).
Draw the shear-force and bending-moment diagrams
for beam ABC.
B
B
A
L
—
2
L
—
2
C
D
E
PROB. 4.5-3
P
4.5-4 The cantilever beam AB shown in the figure is
subjected to a concentrated load P at the midpoint and a
counterclockwise couple of moment M1 PL/4 at the free
end.
Draw the shear-force and bending-moment diagrams
for this beam.
PL
M1 = —–
4
P
A
B
L
—
2
L
—
2
L
—
4
PROB. 4.5-7
4.5-8 A beam ABC is simply supported at A and B and has
an overhang BC (see figure). The beam is loaded by two
forces P and a clockwise couple of moment Pa that act
through the arrangement shown.
Draw the shear-force and bending-moment diagrams
for beam ABC.
P
4.5-5 The simple beam AB shown in the figure is subjected
A
L
—
3
L
—
3
PROB. 4.5-5
4.5-6 A simple beam AB subjected to clockwise couples
M1 and 2M1 acting at the third points is shown in the figure.
Draw the shear-force and bending-moment diagrams
for this beam.
A
PROB. 4.5-6
C
B
a
a
a
4.5-9 Beam ABCD is simply supported at B and C and has
overhangs at each end (see figure). The span length is L and
each overhang has length L /3. A uniform load of intensity
q acts along the entire length of the beam.
Draw the shear-force and bending-moment diagrams
for this beam.
q
2M1
M1
B
L
—
3
A
Pa
PROB. 4.5-8
B
L
—
3
P
a
PL
M1 = —–
4
P
L
—
2
L
PROB. 4.5-4
to a concentrated load P and a clockwise couple M1
PL /4 acting at the third points.
Draw the shear-force and bending-moment diagrams
for this beam.
L
—
4
L
—
3
A
B
L
3
L
—
3
PROB. 4.5-9
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
D
C
L
L
3
296
CHAPTER 4 Shear Forces and Bending Moments
4.5-10 Draw the shear-force and bending-moment diagrams for a cantilever beam AB supporting a linearly
varying load of maximum intensity q0 (see figure).
Draw the shear-force and bending-moment diagrams
for this beam.
200 lb
q0
400 lb-ft
A
B
5 ft
A
5 ft
B
PROB. 4.5-13
L
PROB. 4.5-10
4.5-11 The simple beam AB supports a uniform load of
intensity q 10 lb/in. acting over one-half of the span
and a concentrated load P 80 lb acting at midspan (see
figure).
Draw the shear-force and bending-moment diagrams
for this beam.
4.5-14 The cantilever beam AB shown in the figure is subjected to a uniform load acting throughout one-half of its
length and a concentrated load acting at the free end.
Draw the shear-force and bending-moment diagrams
for this beam.
2.0 kN/m
B
A
2m
P = 80 lb
q = 10 lb/in.
2m
PROB. 4.5-14
A
B
L =
—
40 in.
2
2.5 kN
4.5-15 The uniformly loaded beam ABC has simple
supports at A and B and an overhang BC (see figure).
Draw the shear-force and bending-moment diagrams
for this beam.
L =
—
40 in.
2
25 lb/in.
PROB. 4.5-11
A
form load of intensity 3000 N/m acting over half the length
of the beam. The beam rests on a foundation that produces
a uniformly distributed load over the entire length.
Draw the shear-force and bending-moment diagrams
for this beam.
3000 N/m
A
B
C
B
4.5-12 The beam AB shown in the figure supports a uni72 in.
48 in.
PROB. 4.5-15
4.5-16 A beam ABC with an overhang at one end supports
a uniform load of intensity 12 kN/m and a concentrated
load of magnitude 2.4 kN (see figure).
Draw the shear-force and bending-moment diagrams
for this beam.
12 kN/m
0.8 m
1.6 m
0.8 m
2.4 kN
A
C
B
PROB. 4.5-12
1.6 m
1.6 m
4.5-13 A cantilever beam AB supports a couple and a
concentrated load, as shown in the figure.
PROB. 4.5-16
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
1.6 m
297
CHAPTER 4 Problems
4.5-17 The beam ABC shown in the figure is simply
supported at A and B and has an overhang from B to C. The
loads consist of a horizontal force P1 400 lb acting at the
end of the vertical arm and a vertical force P2 900 lb acting at the end of the overhang.
Draw the shear-force and bending-moment diagrams
for this beam. (Note: Disregard the widths of the beam and
vertical arm and use centerline dimensions when making
calculations.)
E
Cable
A
P2 = 900 lb
8 ft
B
C
6 ft
P1 = 400 lb
1800 lb
D
6 ft
6 ft
PROB. 4.5-19
1.0 ft
A
B
C
4.5-20 The beam ABCD shown in the figure has overhangs
4.0 ft
1.0 ft
PROB. 4.5-17
4.5-18 A simple beam AB is loaded by two segments of
uniform load and two horizontal forces acting at the ends of
a vertical arm (see figure).
Draw the shear-force and bending-moment diagrams
for this beam.
8 kN
4 kN/m
10.6 kN/m
5.1 kN/m
5.1 kN/m
A
B
D
C
4.2 m
4.2 m
1.2 m
4 kN/m
1m
A
that extend in both directions for a distance of 4.2 m from
the supports at B and C, which are 1.2 m apart.
Draw the shear-force and bending-moment diagrams
for this overhanging beam.
PROB. 4.5-20
B
1m
8 kN
2m
2m
2m
2m
4.5-21 The simple beam AB shown in the figure supports a
concentrated load and a segment of uniform load.
Draw the shear-force and bending-moment diagrams
for this beam.
PROB. 4.5-18
4.5-19 A beam ABCD with a vertical arm CE is supported
as a simple beam at A and D (see figure). A cable passes
over a small pulley that is attached to the arm at E. One end
of the cable is attached to the beam at point B. The tensile
force in the cable is 1800 lb.
Draw the shear-force and bending-moment diagrams
for beam ABCD. (Note: Disregard the widths of the beam
and vertical arm and use centerline dimensions when
making calculations.)
4.0 k
2.0 k/ft
C
A
5 ft
10 ft
20 ft
PROB. 4.5-21
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
B
298
CHAPTER 4 Shear Forces and Bending Moments
4.5-22 The cantilever beam shown in the figure supports a
concentrated load and a segment of uniform load.
Draw the shear-force and bending-moment diagrams
for this cantilever beam.
3 kN
1.0 kN/m
A
0.8 m
B
0.8 m
1.6 m
PROB. 4.5-22
4.5-25 A beam of length L is being designed to support a
uniform load of intensity q (see figure). If the supports of
the beam are placed at the ends, creating a simple beam, the
maximum bending moment in the beam is qL2/8. However,
if the supports of the beam are moved symmetrically
toward the middle of the beam (as pictured), the maximum
bending moment is reduced.
Determine the distance a between the supports so that
the maximum bending moment in the beam has the
smallest possible numerical value.
Draw the shear-force and bending-moment diagrams
for this condition.
q
4.5-23 The simple beam ACB shown in the figure is
subjected to a triangular load of maximum intensity
180 lb/ft.
Draw the shear-force and bending-moment diagrams
for this beam.
A
B
a
L
180 lb/ft
PROB. 4.5-25
A
C
4.5-26 The compound beam ABCDE shown in the figure
consists of two beams (AD and DE ) joined by a hinged
connection at D. The hinge can transmit a shear force but
not a bending moment. The loads on the beam consist of a
4-kN force at the end of a bracket attached at point B and a
2-kN force at the midpoint of beam DE.
Draw the shear-force and bending-moment diagrams
for this compound beam.
B
6.0 ft
7.0 ft
PROB. 4.5-23
4 kN
4.5-24 A beam with simple supports is subjected to a
trapezoidally distributed load (see figure). The intensity of
the load varies from 1.0 kN/m at support A to 3.0 kN/m at
support B.
Draw the shear-force and bending-moment diagrams
for this beam.
1m
2 kN
B
A
2m
3.0 kN/m
B
2.4 m
PROB. 4.5-24
2m
D
2m
E
2m
PROB. 4.5-26
1.0 kN/m
A
C
1m
4.5-27 The compound beam ABCDE shown in the figure
on the next page consists of two beams (AD and DE ) joined
by a hinged connection at D. The hinge can transmit a shear
force but not a bending moment. A force P acts upward at A
and a uniform load of intensity q acts downward on beam
DE.
Draw the shear-force and bending-moment diagrams
for this compound beam.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
299
CHAPTER 4 Problems
P
q
B
A
C
L
L
D
E
2L
L
4.5-30 A simple beam AB supports two connected wheel
loads P and 2P that are distance d apart (see figure). The
wheels may be placed at any distance x from the left-hand
support of the beam.
(a) Determine the distance x that will produce the
maximum shear force in the beam, and also determine the
maximum shear force Vmax.
(b) Determine the distance x that will produce the
maximum bending moment in the beam, and also draw the
corresponding bending-moment diagram. (Assume P
10 kN, d 2.4 m, and L 12 m.)
★
PROB. 4.5-27
4.5-28 The shear-force diagram for a simple beam is
shown in the figure.
Determine the loading on the beam and draw the
bending-moment diagram, assuming that no couples act as
loads on the beam.
P
x
2P
d
A
12 kN
B
V
0
L
–12 kN
2.0 m
1.0 m
PROB. 4.5-30
1.0 m
PROB. 4.5-28
4.5-29 The shear-force diagram for a beam is shown in
the figure. Assuming that no couples act as loads on the
beam, determine the forces acting on the beam and draw
the bending-moment diagram.
★
652 lb
580 lb
572 lb
500 lb
V
0
–128 lb
–448 lb
4 ft
16 ft
4 ft
PROB. 4.5-29
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
5
Stresses in Beams
(Basic Topics)
5.1 INTRODUCTION
P
A
B
(a)
y
v
B
A
x
(b)
FIG. 5-1 Bending of a cantilever beam:
(a) beam with load, and (b) deflection
curve
In the preceding chapter we saw how the loads acting on a beam create
internal actions (or stress resultants) in the form of shear forces and bending moments. In this chapter we go one step further and investigate the
stresses and strains associated with those shear forces and bending
moments. Knowing the stresses and strains, we will be able to analyze
and design beams subjected to a variety of loading conditions.
The loads acting on a beam cause the beam to bend (or flex), thereby
deforming its axis into a curve. As an example, consider a cantilever
beam AB subjected to a load P at the free end (Fig. 5-1a). The initially
straight axis is bent into a curve (Fig. 5-1b), called the deflection curve
of the beam.
For reference purposes, we construct a system of coordinate axes
(Fig. 5-1b) with the origin located at a suitable point on the longitudinal
axis of the beam. In this illustration, we place the origin at the fixed support. The positive x axis is directed to the right, and the positive y axis is
directed upward. The z axis, not shown in the figure, is directed outward
(that is, toward the viewer), so that the three axes form a right-handed
coordinate system.
The beams considered in this chapter (like those discussed in
Chapter 4) are assumed to be symmetric about the xy plane, which means
that the y axis is an axis of symmetry of the cross section. In addition, all
loads must act in the xy plane. As a consequence, the bending deflections
occur in this same plane, known as the plane of bending. Thus, the
deflection curve shown in Fig. 5-1b is a plane curve lying in the plane of
bending.
300
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.2
Pure Bending and Nonuniform Bending
301
The deflection of the beam at any point along its axis is the
displacement of that point from its original position, measured in the y
direction. We denote the deflection by the letter v to distinguish it from
the coordinate y itself (see Fig. 5-1b).*
5.2 PURE BENDING AND NONUNIFORM BENDING
M1
M1
A
B
(a)
M1
M
0
(b)
FIG. 5-2 Simple beam in pure bending
(M M1)
M2
A
B
M2
(a)
0
M
⫺M2
(b)
FIG. 5-3 Cantilever beam in pure
bending (M ⫺M2)
When analyzing beams, it is often necessary to distinguish between pure
bending and nonuniform bending. Pure bending refers to flexure of a
beam under a constant bending moment. Therefore, pure bending occurs
only in regions of a beam where the shear force is zero (because
V dM/dx; see Eq. 4-6). In contrast, nonuniform bending refers to
flexure in the presence of shear forces, which means that the bending
moment changes as we move along the axis of the beam.
As an example of pure bending, consider a simple beam AB loaded
by two couples M1 having the same magnitude but acting in opposite
directions (Fig. 5-2a). These loads produce a constant bending moment
M M1 throughout the length of the beam, as shown by the bending
moment diagram in part (b) of the figure. Note that the shear force V is
zero at all cross sections of the beam.
Another illustration of pure bending is given in Fig. 5-3a, where the
cantilever beam AB is subjected to a clockwise couple M2 at the free end.
There are no shear forces in this beam, and the bending moment M is
constant throughout its length. The bending moment is negative
(M ⫺M2), as shown by the bending moment diagram in part (b) of
Fig. 5-3.
The symmetrically loaded simple beam of Fig. 5-4a (on the next
page) is an example of a beam that is partly in pure bending and partly in
nonuniform bending, as seen from the shear-force and bending-moment
diagrams (Figs. 5-4b and c). The central region of the beam is in pure
bending because the shear force is zero and the bending moment is constant. The parts of the beam near the ends are in nonuniform bending
because shear forces are present and the bending moments vary.
In the following two sections we will investigate the strains and
stresses in beams subjected only to pure bending. Fortunately, we can
often use the results obtained for pure bending even when shear forces
are present, as explained later (see the last paragraph in Section 5.8).
*In applied mechanics, the traditional symbols for displacements in the x, y, and z directions are u, v, and w, respectively.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
302
CHAPTER 5 Stresses in Beams (Basic Topics)
P
P
V
A
0
B
(b)
a
a
(a)
FIG. 5-4 Simple beam with central region
in pure bending and end regions in
nonuniform bending
−P
Pa
M
0
(c)
5.3 CURVATURE OF A BEAM
P
A
B
(a)
O′
du
y
r
m1
A
B
m2
x
ds
x
dx
When loads are applied to a beam, its longitudinal axis is deformed into
a curve, as illustrated previously in Fig. 5-1. The resulting strains and stresses
in the beam are directly related to the curvature of the deflection curve.
To illustrate the concept of curvature, consider again a cantilever
beam subjected to a load P acting at the free end (Fig. 5-5a). The deflection curve of this beam is shown in Fig. 5-5b. For purposes of analysis,
we identify two points m1 and m2 on the deflection curve. Point m1 is
selected at an arbitrary distance x from the y axis and point m2 is located
a small distance ds further along the curve. At each of these points we
draw a line normal to the tangent to the deflection curve, that is, normal
to the curve itself. These normals intersect at point O, which is the
center of curvature of the deflection curve. Because most beams have
very small deflections and nearly flat deflection curves, point O is
usually located much farther from the beam than is indicated in the
figure.
The distance m1O from the curve to the center of curvature is
called the radius of curvature r (Greek letter rho), and the curvature
k (Greek letter kappa) is defined as the reciprocal of the radius of curvature. Thus,
1
k
r
(b)
FIG. 5-5 Curvature of a bent beam:
(a) beam with load, and (b) deflection
curve
(5-1)
Curvature is a measure of how sharply a beam is bent. If the load on a
beam is small, the beam will be nearly straight, the radius of curvature
will be very large, and the curvature will be very small. If the load is
increased, the amount of bending will increase—the radius of curvature
will become smaller, and the curvature will become larger.
From the geometry of triangle Om1m2 (Fig. 5-5b) we obtain
r du ds
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(a)
SECTION 5.3 Curvature of a Beam
303
in which du (measured in radians) is the infinitesimal angle between the
normals and ds is the infinitesimal distance along the curve between
points m1 and m2. Combining Eq. (a) with Eq. (5-1), we get
1
du
k
r
ds
(5-2)
This equation for curvature is derived in textbooks on calculus and holds
for any curve, regardless of the amount of curvature. If the curvature is
constant throughout the length of a curve, the radius of curvature will
also be constant and the curve will be an arc of a circle.
The deflections of a beam are usually very small compared to its
length (consider, for instance, the deflections of the structural frame of an
automobile or a beam in a building). Small deflections mean that the
deflection curve is nearly flat. Consequently, the distance ds along the
curve may be set equal to its horizontal projection dx (see Fig. 5-5b).
Under these special conditions of small deflections, the equation for the
curvature becomes
1
du
k
r
dx
(5-3)
y
Positive
curvature
x
O
(a)
y
Negative
curvature
x
O
(b)
FIG. 5-6 Sign convention for curvature
Both the curvature and the radius of curvature are functions of the distance x measured along the x axis. It follows that the position O of the
center of curvature also depends upon the distance x.
In Section 5.5 we will see that the curvature at a particular point on
the axis of a beam depends upon the bending moment at that point and
upon the properties of the beam itself (shape of cross section and type of
material). Therefore, if the beam is prismatic and the material is homogeneous, the curvature will vary only with the bending moment.
Consequently, a beam in pure bending will have constant curvature and
a beam in nonuniform bending will have varying curvature.
The sign convention for curvature depends upon the orientation of
the coordinate axes. If the x axis is positive to the right and the y axis is
positive upward, as shown in Fig. 5-6, then the curvature is positive when
the beam is bent concave upward and the center of curvature is above the
beam. Conversely, the curvature is negative when the beam is bent concave downward and the center of curvature is below the beam.
In the next section we will see how the longitudinal strains in a bent
beam are determined from its curvature, and in Chapter 9 we will see
how curvature is related to the deflections of beams.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
304
CHAPTER 5 Stresses in Beams (Basic Topics)
5.4 LONGITUDINAL STRAINS IN BEAMS
The longitudinal strains in a beam can be found by analyzing the curvature of the beam and the associated deformations. For this purpose, let us
consider a portion AB of a beam in pure bending subjected to positive
bending moments M (Fig. 5-7a). We assume that the beam initially has a
straight longitudinal axis (the x axis in the figure) and that its cross
section is symmetric about the y axis, as shown in Fig. 5-7b.
Under the action of the bending moments, the beam deflects in the
xy plane (the plane of bending) and its longitudinal axis is bent into a circular curve (curve ss in Fig. 5-7c). The beam is bent concave upward,
which is positive curvature (Fig. 5-6a).
Cross sections of the beam, such as sections mn and pq in Fig. 5-7a,
remain plane and normal to the longitudinal axis (Fig. 5-7c). The fact that
cross sections of a beam in pure bending remain plane is so fundamental
to beam theory that it is often called an assumption. However, we could
also call it a theorem, because it can be proved rigorously using only
rational arguments based upon symmetry (Ref. 5-1). The basic point is
y
y
A
e
M
s
O
p
m
dx
f
y
n
B
M
s
x
(b)
O′
r
du
M
e
bending: (a) side view of beam,
(b) cross section of beam, and
(c) deformed beam
f
dx
FIG. 5-7 Deformations of a beam in pure
n
B
p
m
s
O
q
(a)
A
z
(c)
y
q
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
s
M
SECTION 5.4 Longitudinal Strains in Beams
305
that the symmetry of the beam and its loading (Figs. 5-7a and b) means
that all elements of the beam (such as element mpqn) must deform in an
identical manner, which is possible only if cross sections remain plane
during bending (Fig. 5-7c). This conclusion is valid for beams of any
material, whether the material is elastic or inelastic, linear or nonlinear.
Of course, the material properties, like the dimensions, must be symmetric about the plane of bending. (Note: Even though a plane cross section
in pure bending remains plane, there still may be deformations in the
plane itself. Such deformations are due to the effects of Poisson’s ratio,
as explained at the end of this discussion.)
Because of the bending deformations shown in Fig. 5-7c, cross
sections mn and pq rotate with respect to each other about axes perpendicular to the xy plane. Longitudinal lines on the lower part of the beam
are elongated, whereas those on the upper part are shortened. Thus, the
lower part of the beam is in tension and the upper part is in compression.
Somewhere between the top and bottom of the beam is a surface in which
longitudinal lines do not change in length. This surface, indicated by the
dashed line ss in Figs. 5-7a and c, is called the neutral surface of the
beam. Its intersection with any cross-sectional plane is called the neutral
axis of the cross section; for instance, the z axis is the neutral axis for the
cross section of Fig. 5-7b.
The planes containing cross sections mn and pq in the deformed
beam (Fig. 5-7c) intersect in a line through the center of curvature O.
The angle between these planes is denoted du, and the distance from O
to the neutral surface ss is the radius of curvature r. The initial distance
dx between the two planes (Fig. 5-7a) is unchanged at the neutral surface
(Fig. 5-7c), hence r du dx. However, all other longitudinal lines
between the two planes either lengthen or shorten, thereby creating
normal strains ex.
To evaluate these normal strains, consider a typical longitudinal line
ef located within the beam between planes mn and pq (Fig. 5-7a). We
identify line ef by its distance y from the neutral surface in the initially
straight beam. Thus, we are now assuming that the x axis lies along the
neutral surface of the undeformed beam. Of course, when the beam
deflects, the neutral surface moves with the beam, but the x axis remains
fixed in position. Nevertheless, the longitudinal line ef in the deflected
beam (Fig. 5-7c) is still located at the same distance y from the neutral
surface. Thus, the length L1 of line ef after bending takes place is
y
L1 (r y) du dx dx
r
in which we have substituted du dx/r.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
306
CHAPTER 5 Stresses in Beams (Basic Topics)
Since the original length of line ef is dx, it follows that its elongation
is L1 dx, or y dx/r. The corresponding longitudinal strain is equal to
the elongation divided by the initial length dx; therefore, the straincurvature relation is
y
ex ky
r
(5-4)
where k is the curvature (see Eq. 5-1).
The preceding equation shows that the longitudinal strains in the
beam are proportional to the curvature and vary linearly with the distance
y from the neutral surface. When the point under consideration is above
the neutral surface, the distance y is positive. If the curvature is also
positive (as in Fig. 5-7c), then ex will be a negative strain, representing a
shortening. By contrast, if the point under consideration is below the
neutral surface, the distance y will be negative and, if the curvature is
positive, the strain ex will also be positive, representing an elongation.
Note that the sign convention for ex is the same as that used for normal
strains in earlier chapters, namely, elongation is positive and shortening
is negative.
Equation (5-4) for the normal strains in a beam was derived solely
from the geometry of the deformed beam—the properties of the material
did not enter into the discussion. Therefore, the strains in a beam in pure
bending vary linearly with distance from the neutral surface regardless of
the shape of the stress-strain curve of the material.
The next step in our analysis, namely, finding the stresses from the
strains, requires the use of the stress-strain curve. This step is described
in the next section for linearly elastic materials and in Section 6.10 for
elastoplastic materials.
The longitudinal strains in a beam are accompanied by transverse
strains (that is, normal strains in the y and z directions) because of the
effects of Poisson’s ratio. However, there are no accompanying
transverse stresses because beams are free to deform laterally. This stress
condition is analogous to that of a prismatic bar in tension or compression, and therefore longitudinal elements in a beam in pure bending are
in a state of uniaxial stress.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.4 Longitudinal Strains in Beams
307
Example 5-1
A simply supported steel beam AB (Fig. 5-8a) of length L 8.0 ft and height
h 6.0 in. is bent by couples M0 into a circular arc with a downward deflection
d at the midpoint (Fig. 5-8b). The longitudinal normal strain (elongation) on the
bottom surface of the beam is 0.00125, and the distance from the neutral surface
to the bottom surface of the beam is 3.0 in.
Determine the radius of curvature r, the curvature k, and the deflection d of
the beam.
Note: This beam has a relatively large deflection because its length is large
compared to its height (L/h 16) and the strain of 0.00125 is also large. (It is
about the same as the yield strain for ordinary structural steel.)
M0
M0
h
A
B
L
(a)
O′
y
u
u
r
r
C
A
d
B x
C′
FIG. 5-8 Example 5-1. Beam in pure
bending: (a) beam with loads, and
(b) deflection curve
L
—
2
L
—
2
(b)
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
308
CHAPTER 5 Stresses in Beams (Basic Topics)
Solution
Curvature. Since we know the longitudinal strain at the bottom surface of
the beam (ex 0.00125), and since we also know the distance from the neutral
surface to the bottom surface ( y ⫺3.0 in.), we can use Eq. (5-4) to calculate
both the radius of curvature and the curvature. Rearranging Eq. (5-4) and substituting numerical values, we get
y
⫺3.0 in.
r ⫺ᎏ ᎏ ⫺ᎏᎏ 2400 in. 200 ft
ex
0.00125
1
k ᎏᎏ 0.0050 ft–1
r
These results show that the radius of curvature is extremely large compared to
the length of the beam even when the strain in the material is large. If, as usual,
the strain is less, the radius of curvature is even larger.
Deflection. As pointed out in Section 5.3, a constant bending moment (pure
bending) produces constant curvature throughout the length of a beam.
Therefore, the deflection curve is a circular arc. From Fig. 5-8b we see that the
distance from the center of curvature O to the midpoint C of the deflected beam
is the radius of curvature r, and the distance from O to point C on the x axis is
r cos u, where u is angle BOC. This leads to the following expression for the
deflection at the midpoint of the beam:
d r (1 2 cos u)
(5-5)
For a nearly flat curve, we can assume that the distance between supports is the
same as the length of the beam itself. Therefore, from triangle BOC we get
L /2
sin u
r
(5-6)
Substituting numerical values, we obtain
(8.0 ft)(12 in./ft)
sin u 0.0200
2(2400 in.)
and
u 0.0200 rad 1.146°
Note that for practical purposes we may consider sin u and u (radians) to be equal
numerically because u is a very small angle.
Now we substitute into Eq. (5-5) for the deflection and obtain
d r(1 cos u ) (2400 in.)(1 0.999800) 0.480 in.
This deflection is very small compared to the length of the beam, as shown by
the ratio of the span length to the deflection:
L (8.0 ft)(12 in./ft)
200
d
0.480 in.
Thus, we have confirmed that the deflection curve is nearly flat in spite of the
large strains. Of course, in Fig. 5-8b the deflection of the beam is highly exaggerated for clarity.
Note: The purpose of this example is to show the relative magnitudes of the
radius of curvature, length of the beam, and deflection of the beam. However, the
method used for finding the deflection has little practical value because it is limited to pure bending, which produces a circular deflected shape. More useful
methods for finding beam deflections are presented later in Chapter 9.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.5 Normal Stresses in Beams (Linearly Elastic Materials)
309
5.5 NORMAL STRESSES IN BEAMS (LINEARLY ELASTIC MATERIALS)
In the preceding section we investigated the longitudinal strains ex in a
beam in pure bending (see Eq. 5-4 and Fig. 5-7). Since longitudinal
elements of a beam are subjected only to tension or compression, we can
use the stress-strain curve for the material to determine the stresses from
the strains. The stresses act over the entire cross section of the beam and
vary in intensity depending upon the shape of the stress-strain diagram
and the dimensions of the cross section. Since the x direction is longitudinal
(Fig. 5-7a), we use the symbol sx to denote these stresses.
The most common stress-strain relationship encountered in
engineering is the equation for a linearly elastic material. For such
materials we substitute Hooke’s law for uniaxial stress (s Ee) into
Eq. (5-4) and obtain
y
sx
M
x
O
(a)
y
dA
c1
y
z
c2
O
(b)
FIG. 5-9 Normal stresses in a beam of
linearly elastic material: (a) side view of
beam showing distribution of normal
stresses, and (b) cross section of beam
showing the z axis as the neutral axis of
the cross section
Ey
sx Eex ⫺ᎏᎏ ⫺Eky
r
(5-7)
This equation shows that the normal stresses acting on the cross section
vary linearly with the distance y from the neutral surface. This stress
distribution is pictured in Fig. 5-9a for the case in which the bending
moment M is positive and the beam bends with positive curvature.
When the curvature is positive, the stresses sx are negative (compression) above the neutral surface and positive (tension) below it. In the
figure, compressive stresses are indicated by arrows pointing toward the
cross section and tensile stresses are indicated by arrows pointing away
from the cross section.
In order for Eq. (5-7) to be of practical value, we must locate the
origin of coordinates so that we can determine the distance y. In other
words, we must locate the neutral axis of the cross section. We also need
to obtain a relationship between the curvature and the bending moment—
so that we can substitute into Eq. (5-7) and obtain an equation relating the
stresses to the bending moment. These two objectives can be accomplished by determining the resultant of the stresses sx acting on the cross
section.
In general, the resultant of the normal stresses consists of two
stress resultants: (1) a force acting in the x direction, and (2) a bending
couple acting about the z axis. However, the axial force is zero when a
beam is in pure bending. Therefore, we can write the following equations
of statics: (1) The resultant force in the x direction is equal to zero, and
(2) the resultant moment is equal to the bending moment M. The first
equation gives the location of the neutral axis and the second gives the
moment-curvature relationship.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
310
CHAPTER 5 Stresses in Beams (Basic Topics)
y
Location of Neutral Axis
sx
M
x
O
s dA EkydA 0
(a)
A
y
dA
c1
y
z
c2
O
(b)
FIG. 5-9 (Repeated)
To obtain the first equation of statics, we consider an element of area dA
in the cross section (Fig. 5-9b). The element is located at distance y from
the neutral axis, and therefore the stress sx acting on the element is given
by Eq. (5-7). The force acting on the element is equal to sx d A and is
compressive when y is positive. Because there is no resultant force acting
on the cross section, the integral of sx dA over the area A of the entire
cross section must vanish; thus, the first equation of statics is
x
A
(a)
Because the curvature k and modulus of elasticity E are nonzero
constants at any given cross section of a bent beam, they are not involved
in the integration over the cross-sectional area. Therefore, we can drop
them from the equation and obtain
y dA 0
(5-8)
A
This equation states that the first moment of the area of the cross section,
evaluated with respect to the z axis, is zero. In other words, the z axis
must pass through the centroid of the cross section.*
Since the z axis is also the neutral axis, we have arrived at the
following important conclusion: The neutral axis passes through the
centroid of the cross-sectional area when the material follows Hooke’s
law and there is no axial force acting on the cross section. This
observation makes it relatively simple to determine the position of the
neutral axis.
As explained in Section 5.1, our discussion is limited to beams for
which the y axis is an axis of symmetry. Consequently, the y axis also
passes through the centroid. Therefore, we have the following additional
conclusion: The origin O of coordinates (Fig. 5-9b) is located at the
centroid of the cross-sectional area.
Because the y axis is an axis of symmetry of the cross section, it
follows that the y axis is a principal axis (see Chapter 12, Section 12.9,
for a discussion of principal axes). Since the z axis is perpendicular to the
y axis, it too is a principal axis. Thus, when a beam of linearly elastic
material is subjected to pure bending, the y and z axes are principal
centroidal axes.
Moment-Curvature Relationship
The second equation of statics expresses the fact that the moment resultant
of the normal stresses sx acting over the cross section is equal to the
*Centroids and first moments of areas are discussed in Chapter 12, Sections 12.2 and
12.3.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.5 Normal Stresses in Beams (Linearly Elastic Materials)
311
bending moment M (Fig. 5-9a). The element of force sxdA acting on the
element of area dA (Fig. 5-9b) is in the positive direction of the x axis
when sx is positive and in the negative direction when sx is negative.
Since the element dA is located above the neutral axis, a positive stress
sx acting on that element produces an element of moment equal to
sx y dA. This element of moment acts opposite in direction to the positive bending moment M shown in Fig. 5-9a. Therefore, the elemental
moment is
dM ⫺sx y dA
The integral of all such elemental moments over the entire crosssectional area A must equal the bending moment:
M
s y dA
A
(b)
x
or, upon substituting for sx from Eq. (5-7),
M
kEy dA kE y dA
2
2
A
A
(5-9)
This equation relates the curvature of the beam to the bending moment M.
Since the integral in the preceding equation is a property of the crosssectional area, it is convenient to rewrite the equation as follows:
M kEI
(5-10)
in which
I
A
y 2 dA
(5-11)
This integral is the moment of inertia of the cross-sectional area with
respect to the z axis (that is, with respect to the neutral axis). Moments of
inertia are always positive and have dimensions of length to the fourth
power; for instance, typical USCS units are in.4 and typical SI units are
mm4 when performing beam calculations.*
Equation (5-10) can now be rearranged to express the curvature in
terms of the bending moment in the beam:
M
1
k
r EI
(5-12)
Known as the moment-curvature equation, Eq. (5-12) shows that the
curvature is directly proportional to the bending moment M and inversely
proportional to the quantity EI, which is called the flexural rigidity of
the beam. Flexural rigidity is a measure of the resistance of a beam to
bending, that is, the larger the flexural rigidity, the smaller the curvature
for a given bending moment.
*Moments of inertia of areas are discussed in Chapter 12, Section 12.4.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
312
CHAPTER 5 Stresses in Beams (Basic Topics)
y
+M
Positive
bending
moment
Comparing the sign convention for bending moments (Fig. 4-5) with
that for curvature (Fig. 5-6), we see that a positive bending moment
produces positive curvature and a negative bending moment produces
negative curvature (see Fig. 5-10).
+M
Positive
curvature
Flexure Formula
Now that we have located the neutral axis and derived the momentcurvature relationship, we can determine the stresses in terms of the
bending moment. Substituting the expression for curvature (Eq. 5-12)
into the expression for the stress sx (Eq. 5-7), we get
x
O
y
−M
Negative
bending
moment
Negative
curvature
My
sx
I
−M
O
x
FIG. 5-10 Relationships between signs of
bending moments and signs of
curvatures
(5-13)
This equation, called the flexure formula, shows that the stresses are
directly proportional to the bending moment M and inversely proportional to the moment of inertia I of the cross section. Also, the stresses
vary linearly with the distance y from the neutral axis, as previously
observed. Stresses calculated from the flexure formula are called bending stresses or flexural stresses.
If the bending moment in the beam is positive, the bending stresses
will be positive (tension) over the part of the cross section where y is negative, that is, over the lower part of the beam. The stresses in the upper
part of the beam will be negative (compression). If the bending moment
is negative, the stresses will be reversed. These relationships are shown
in Fig. 5-11.
Maximum Stresses at a Cross Section
The maximum tensile and compressive bending stresses acting at any
given cross section occur at points located farthest from the neutral axis.
Let us denote by c1 and c2 the distances from the neutral axis to the
y
y
Compressive stresses
s1
c1
FIG. 5-11 Relationships between
signs of bending moments and
directions of normal stresses:
(a) positive bending moment,
and (b) negative bending
moment
Positive bending
moment
⫹M
x
Tensile stresses
s1
c1
O
O
c2
c2
s2
Tensile stresses
(a)
Negative bending
moment
x
⫺M
s2
Compressive stresses
(b)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.5 Normal Stresses in Beams (Linearly Elastic Materials)
313
extreme elements in the positive and negative y directions, respectively
(see Fig. 5-9b and Fig. 5-11). Then the corresponding maximum normal
stresses s1 and s2 (from the flexure formula) are
Mc1
M
s1
I
S1
Mc2
M
s2
I
S2
(5-14a,b)
in which
I
S1
c1
y
b
—
2
z
h
—
2
O
h
I
S2
c2
(5-15a,b)
The quantities S1 and S2 are known as the section moduli of the crosssectional area. From Eqs. (5-15a and b) we see that each section modulus
has dimensions of length to the third power (for example, in.3 or mm3).
Note that the distances c1 and c2 to the top and bottom of the beam are
always taken as positive quantities.
The advantage of expressing the maximum stresses in terms of
section moduli arises from the fact that each section modulus combines
the beam’s relevant cross-sectional properties into a single quantity. Then
this quantity can be listed in tables and handbooks as a property of the
beam, which is a convenience to designers. (Design of beams using
section moduli is explained in the next section.)
Doubly Symmetric Shapes
If the cross section of a beam is symmetric with respect to the z axis as
well as the y axis (doubly symmetric cross section), then c1 c2 c and
the maximum tensile and compressive stresses are equal numerically:
b
(a)
Mc
M
s1 s2
I
S
y
or
M
smax ᎏᎏ
S
(5-16a,b)
in which
O
I
S ᎏᎏ
c
d
is the only section modulus for the cross section.
For a beam of rectangular cross section with width b and height h
(Fig. 5-12a), the moment of inertia and section modulus are
z
(b)
FIG. 5-12 Doubly symmetric cross-
sectional shapes
bh3
I ᎏᎏ
12
(5-17)
bh2
S ᎏᎏ
6
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(5-18a,b)
314
CHAPTER 5 Stresses in Beams (Basic Topics)
For a circular cross section of diameter d (Fig. 5-12b), these properties are
pd4
I ᎏᎏ
64
pd 3
S ᎏᎏ
32
(5-19a,b)
Properties of other doubly symmetric shapes, such as hollow tubes (either
rectangular or circular) and wide-flange shapes, can be readily obtained
from the preceding formulas.
Properties of Beam Cross Sections
Moments of inertia of many plane figures are listed in Appendix D for
convenient reference. Also, the dimensions and properties of standard sizes
of steel and wood beams are listed in Appendixes E and F and in many
engineering handbooks, as explained in more detail in the next section.
For other cross-sectional shapes, we can determine the location of
the neutral axis, the moment of inertia, and the section moduli by direct
calculation, using the techniques described in Chapter 12. This procedure
is illustrated later in Example 5-4.
Limitations
The analysis presented in this section is for pure bending of prismatic
beams composed of homogeneous, linearly elastic materials. If a beam is
subjected to nonuniform bending, the shear forces will produce warping
(or out-of-plane distortion) of the cross sections. Thus, a cross section
that was plane before bending is no longer plane after bending. Warping
due to shear deformations greatly complicates the behavior of the beam.
However, detailed investigations show that the normal stresses calculated
from the flexure formula are not significantly altered by the presence of
shear stresses and the associated warping (Ref. 2-1, pp. 42 and 48). Thus,
we may justifiably use the theory of pure bending for calculating normal
stresses in beams subjected to nonuniform bending.*
The flexure formula gives results that are accurate only in regions of
the beam where the stress distribution is not disrupted by changes in the
shape of the beam or by discontinuities in loading. For instance, the flexure formula is not applicable near the supports of a beam or close to a
concentrated load. Such irregularities produce localized stresses, or stress
concentrations, that are much greater than the stresses obtained from the
flexure formula (see Section 5.13).
*Beam theory began with Galileo Galilei (1564–1642), who investigated the behavior of
various types of beams. His work in mechanics of materials is described in his famous
book Two New Sciences, first published in 1638 (Ref. 5-2). Although Galileo made many
important discoveries regarding beams, he did not obtain the stress distribution that we
use today. Further progress in beam theory was made by Mariotte, Jacob Bernoulli, Euler,
Parent, Saint-Venant, and others (Ref. 5-3).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.5 Normal Stresses in Beams (Linearly Elastic Materials)
315
Example 5-2
R0
C
d
A high-strength steel wire of diameter d is bent around a cylindrical drum of
radius R0 (Fig. 5-13).
Determine the bending moment M and maximum bending stress smax in the
wire, assuming d 4 mm and R0 0.5 m. (The steel wire has modulus of elasticity E 200 GPa and proportional limit sp1 1200 MPa.)
Solution
FIG. 5-13 Example 5-2. Wire bent around
a drum
The first step in this example is to determine the radius of curvature r of the
bent wire. Then, knowing r, we can find the bending moment and maximum
stresses.
Radius of curvature. The radius of curvature of the bent wire is the distance
from the center of the drum to the neutral axis of the cross section of the wire:
d
r R0
2
(5-20)
Bending moment. The bending moment in the wire may be found from the
moment-curvature relationship (Eq. 5-12):
EI
2 EI
M
r
2R0 d
(5-21)
in which I is the moment of inertia of the cross-sectional area of the wire.
Substituting for I in terms of the diameter d of the wire (Eq. 5-19a), we get
pEd 4
M
32(2R0 d)
(5-22)
This result was obtained without regard to the sign of the bending moment, since
the direction of bending is obvious from the figure.
Maximum bending stresses. The maximum tensile and compressive stresses,
which are equal numerically, are obtained from the flexure formula as given by
Eq. (5-16b):
M
smax
S
in which S is the section modulus for a circular cross section. Substituting for M
from Eq. (5-22) and for S from Eq. (5-19b), we get
Ed
smax
2R0 d
(5-23)
This same result can be obtained directly from Eq. (5-7) by replacing y with d/2
and substituting for r from Eq. (5-20).
We see by inspection of Fig. 5-13 that the stress is compressive on the lower
(or inner) part of the wire and tensile on the upper (or outer) part.
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
316
CHAPTER 5 Stresses in Beams (Basic Topics)
Numerical results. We now substitute the given numerical data into Eqs.
(5-22) and (5-23) and obtain the following results:
p (200 GPa)(4 mm)4
pEd 4
M 5.01 Nm
32[2(0.5 m) 4 mm]
32(2R0 d)
(200 GPa)(4 mm)
Ed
smax ᎏ 797 MPa
2(0.5 m) 4 mm
2R0 ⫹ d
Note that smax is less than the proportional limit of the steel wire, and therefore
the calculations are valid.
Note: Because the radius of the drum is large compared to the diameter of
the wire, we can safely disregard d in comparison with 2R0 in the denominators
of the expressions for M and smax. Then Eqs. (5-22) and (5-23) yield the
following results:
M 5.03 Nm
smax 800 MPa
These results are on the conservative side and differ by less than 1% from the
more precise values.
Example 5-3
A simple beam AB of span length L 22 ft (Fig. 5-14a) supports a uniform load
of intensity q 1.5 k/ft and a concentrated load P 12 k. The uniform load
includes an allowance for the weight of the beam. The concentrated load acts at
a point 9.0 ft from the left-hand end of the beam. The beam is constructed of
glued laminated wood and has a cross section of width b 8.75 in. and height
h 27 in. (Fig. 5-14b).
Determine the maximum tensile and compressive stresses in the beam due
to bending.
Solution
Reactions, shear forces, and bending moments. We begin the analysis by
calculating the reactions at supports A and B, using the techniques described in
Chapter 4. The results are
RA 23.59 k
RB 21.41 k
Knowing the reactions, we can construct the shear-force diagram, shown in Fig.
5-14c. Note that the shear force changes from positive to negative under the concentrated load P, which is at a distance of 9 ft from the left-hand support.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.5 Normal Stresses in Beams (Linearly Elastic Materials)
317
P = 12 k
9 ft
q = 1.5 k/ft
A
B
V 23.59
(k)
10.09
0
–1.91
L = 22 ft
(a)
(c)
h = 27 in.
FIG. 5-14 Example 5-3. Stresses in a
simple beam
M
(k-ft)
–21.41
151.6
0
b = 8.75 in.
(d)
(b)
Next, we draw the bending-moment diagram (Fig. 5-14d) and determine the
maximum bending moment, which occurs under the concentrated load where the
shear force changes sign. The maximum moment is
Mmax 151.6 k-ft
The maximum bending stresses in the beam occur at the cross section of maximum moment.
Section modulus. The section modulus of the cross-sectional area is calculated from Eq. (5-18b), as follows:
bh2
1
S ᎏ (8.75 in.)(27 in.)2 1063 in.3
6
6
Maximum stresses. The maximum tensile and compressive stresses st and
sc, respectively, are obtained from Eq. (5-16a):
(151.6 k-ft)(12 in./ft)
M
st s 2 max
1710 psi
1063 in.3
S
M
sc s1 max
1710 psi
S
Because the bending moment is positive, the maximum tensile stress occurs at
the bottom of the beam and the maximum compressive stress occurs at the top.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
318
CHAPTER 5 Stresses in Beams (Basic Topics)
Example 5-4
q = 3.2 kN/m
A
C
B
3.0 m
1.5 m
(a)
4.8 kN
V 3.6 kN
The beam ABC shown in Fig 5-15a has simple supports at A and B and an overhang from B to C. The length of the span is 3.0 m and the length of the overhang
is 1.5 m. A uniform load of intensity q 3.2 kN/m acts throughout the entire
length of the beam (4.5 m).
The beam has a cross section of channel shape with width b 300 mm and
height h 80 mm (Fig. 5-16a). The web thickness is t 12 mm, and the average thickness of the sloping flanges is the same. For the purpose of calculating
the properties of the cross section, assume that the cross section consists of three
rectangles, as shown in Fig. 5-16b.
Determine the maximum tensile and compressive stresses in the beam due
to the uniform load.
Solution
Reactions, shear forces, and bending moments. We begin the analysis of this
beam by calculating the reactions at supports A and B, using the techniques
described in Chapter 4. The results are
0
1.125 m
−6.0 kN
RA 3.6 kN
(b)
M
RB 10.8 kN
From these values, we construct the shear-force diagram (Fig. 5-15b). Note that
the shear force changes sign and is equal to zero at two locations: (1) at a distance of 1.125 m from the left-hand support, and (2) at the right-hand reaction.
Next, we draw the bending-moment diagram, shown in Fig. 5-15c. Both the
maximum positive and maximum negative bending moments occur at the cross
sections where the shear force changes sign. These maximum moments are
2.025 kN.m
0
Mpos 2.025 kNm
1.125 m
−3.6 kN.m
(c)
FIG. 5-15 Example 5-4. Stresses in a
beam with an overhang
Mneg 3.6 kNm
respectively.
Neutral axis of the cross section (Fig. 5-16b). The origin O of the yz coordinates is placed at the centroid of the cross-sectional area, and therefore the
z axis becomes the neutral axis of the cross section. The centroid is located by
using the techniques described in Chapter 12, Section 12.3, as follows.
First, we divide the area into three rectangles (A1, A2, and A3). Second, we
establish a reference axis Z-Z across the upper edge of the cross section, and we
let y1 and y2 be the distances from the Z-Z axis to the centroids of areas A1 and
A2, respectively. Then the calculations for locating the centroid of the entire
channel section (distances c1 and c2) are as follows:
Area 1:
y1 t/2 6 mm
A1 (b – 2t)(t) (276 mm)(12 mm) 3312 mm2
Area 2:
y2 h/2 40 mm
A2 ht (80 mm)(12 mm) 960 mm2
Area 3:
y3 y2
A3 A2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
319
SECTION 5.5 Normal Stresses in Beams (Linearly Elastic Materials)
y
y
b = 300 mm
c1
Z
z
O
t = 12 mm
t = 12 mm
h=
80 mm
A1
y1
y2
z
t = 12 mm
O
A3
A2
(a)
t = 12 mm
h=
80 mm
d1
c2
Z
b = 300 mm
t=
12 mm
(b)
FIG. 5-16 Cross section of beam
discussed in Example 5-4. (a) Actual
shape, and (b) idealized shape for use
in analysis (the thickness of the beam
is exaggerated for clarity)
y1A1 2y2A2
yi Ai
c1 ᎏᎏ
A1 2A2
Ai
(6 mm)(3312 mm2) 2(40 mm)(960 mm2)
18.48 mm
3312 mm2 2(960 mm2)
c2 h 2 c1 80 mm 2 18.48 mm 61.52 mm
Thus, the position of the neutral axis (the z axis) is determined.
Moment of inertia. In order to calculate the stresses from the flexure
formula, we must determine the moment of inertia of the cross-sectional area
with respect to the neutral axis. These calculations require the use of the
parallel-axis theorem (see Chapter 12, Section 12.5).
Beginning with area A1, we obtain its moment of inertia (Iz)1 about the z axis
from the equation
(Iz )1 (Ic)1 A1d 21
(c)
In this equation, (Ic)1 is the moment of inertia of area A1 about its own centroidal
axis:
1
1
(Ic)1 (b2t)(t)3 (276 mm)(12 mm)3 39,744 mm4
12
12
and d1 is the distance from the centroidal axis of area A1 to the z axis:
d1 c1 t/2 18.48 mm 6 mm 12.48 mm
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
320
CHAPTER 5 Stresses in Beams (Basic Topics)
Therefore, the moment of inertia of area A1 about the z axis (from Eq. c) is
(Iz)1 39,744 mm4 (3312 mm2)(12.48 mm2) 555,600 mm4
Proceeding in the same manner for areas A2 and A3, we get
(Iz)2 (Iz)3 956,600 mm4
Thus, the centroidal moment of inertia Iz of the entire cross-sectional area is
Iz (Iz)1 (Iz)2 (Iz)3 2.469 106 mm4
Section moduli. The section moduli for the top and bottom of the beam,
respectively, are
I
S1 z 133,600 mm3
c1
Iz
S2
40,100 mm3
c2
(see Eqs. 5-15a and b). With the cross-sectional properties determined, we can
now proceed to calculate the maximum stresses from Eqs. (5-14a and b).
Maximum stresses. At the cross section of maximum positive bending
moment, the largest tensile stress occurs at the bottom of the beam (s2) and the
largest compressive stress occurs at the top (s1). Thus, from Eqs. (5-14b) and
(5-14a), respectively, we get
2.025 kNm
Mpos
st s 2 3 50.5 MPa
40,100 mm
S2
2.025 kNm
Mpos
sc s1 3 15.2 MPa
133,600 mm
S1
Similarly, the largest stresses at the section of maximum negative moment are
3.6 kNm
Mneg
st s1 3 26.9 MPa
133,600 mm
S1
3.6 kNm
Mneg
sc s 2 3 89.8 MPa
40,100 mm
S2
A comparison of these four stresses shows that the largest tensile stress in the
beam is 50.5 MPa and occurs at the bottom of the beam at the cross section of
maximum positive bending moment; thus,
(st)max 50.5 MPa
The largest compressive stress is 89.8 MPa and occurs at the bottom of the
beam at the section of maximum negative moment:
(sc)max 89.8 MPa
Thus, we have determined the maximum bending stresses due to the uniform
load acting on the beam.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.6 Design of Beams for Bending Stresses
321
5.6 DESIGN OF BEAMS FOR BENDING STRESSES
The process of designing a beam requires that many factors be considered,
including the type of structure (airplane, automobile, bridge, building, or
whatever), the materials to be used, the loads to be supported, the
environmental conditions to be encountered, and the costs to be paid.
However, from the standpoint of strength, the task eventually reduces
to selecting a shape and size of beam such that the actual stresses in the
beam do not exceed the allowable stresses for the material. In this
section, we will consider only the bending stresses (that is, the stresses
obtained from the flexure formula, Eq. 5-13). Later, we will consider
the effects of shear stresses (Sections 5.8, 5.9, and 5.10) and stress
concentrations (Section 5.13).
When designing a beam to resist bending stresses, we usually begin
by calculating the required section modulus. For instance, if the beam
has a doubly symmetric cross section and the allowable stresses are the
same for both tension and compression, we can calculate the required
modulus by dividing the maximum bending moment by the allowable
bending stress for the material (see Eq. 5-16):
M ax
S ᎏmᎏ
sallow
FIG. 5-17 Welding three large steel plates
into a single solid section (Courtesy of
The Lincoln Electric Company)
(5-24)
The allowable stress is based upon the properties of the material and the
desired factor of safety. To ensure that this stress is not exceeded, we
must choose a beam that provides a section modulus at least as large as
that obtained from Eq. (5-24).
If the cross section is not doubly symmetric, or if the allowable
stresses are different for tension and compression, we usually need to
determine two required section moduli—one based upon tension and the
other based upon compression. Then we must provide a beam that satisfies both criteria.
To minimize weight and save material, we usually select a beam that
has the least cross-sectional area while still providing the required section
moduli (and also meeting any other design requirements that may be
imposed).
Beams are constructed in a great variety of shapes and sizes to suit a
myriad of purposes. For instance, very large steel beams are fabricated by
welding (Fig. 5-17), aluminum beams are extruded as round or rectangular tubes, wood beams are cut and glued to fit special requirements, and
reinforced concrete beams are cast in any desired shape by proper construction of the forms.
In addition, beams of steel, aluminum, plastic, and wood can be
ordered in standard shapes and sizes from catalogs supplied by dealers
and manufacturers. Readily available shapes include wide-flange beams,
I-beams, angles, channels, rectangular beams, and tubes.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
322
CHAPTER 5 Stresses in Beams (Basic Topics)
Beams of Standardized Shapes and Sizes
The dimensions and properties of many kinds of beams are listed in engineering handbooks. For instance, in the United States the shapes and
sizes of structural-steel beams are standardized by the American Institute
of Steel Construction (AISC), which publishes manuals giving their
properties in both USCS aand SI units (Ref. 5-4). The tables in these
manuals list cross-sectional dimensions and properties such as weight,
cross-sectional area, moment of inertia, and section modulus.
Properties of aluminum and wood beams are tabulated in a similar
manner and are available in publications of the Aluminum Association
(Ref. 5-5) and the American Forest and Paper Association (Ref. 5-6).
Abridged tables of steel beams and wood beams are given later in
this book for use in solving problems using USCS units (see Appendixes
E and F).
Structural-steel sections are given a designation such as W 30 211,
which means that the section is of W shape (also called a wide-flange
shape) with a nominal depth of 30 in. and a weight of 211 lb per ft of
length (see Table E-1, Appendix E).
Similar designations are used for S shapes (also called I-beams) and
C shapes (also called channels), as shown in Tables E-2 and E-3. Angle
sections, or L shapes, are designated by the lengths of the two legs and
the thickness (see Tables E-4 and E-5). For example, L 8 6 1 denotes
an angle with unequal legs, one of length 8 in. and the other of length
6 in., with a thickness of 1 in. Comparable designations are used when
the dimensions and properties are given in SI units.
The standardized steel sections described above are manufactured by
rolling, a process in which a billet of hot steel is passed back and forth
between rolls until it is formed into the desired shape.
Aluminum structural sections are usually made by the process of
extrusion, in which a hot billet is pushed, or extruded, through a shaped
die. Since dies are relatively easy to make and the material is workable,
aluminum beams can be extruded in almost any desired shape. Standard
shapes of wide-flange beams, I-beams, channels, angles, tubes, and other
sections are listed in the Aluminum Design Manual (Ref. 5-5). In addition, custom-made shapes can be ordered.
Most wood beams have rectangular cross sections and are designated by nominal dimensions, such as 4 8 inches. These dimensions
represent the rough-cut size of the lumber. The net dimensions (or actual
dimensions) of a wood beam are smaller than the nominal dimensions if
the sides of the rough lumber have been planed, or surfaced, to make
them smooth. Thus, a 4 8 wood beam has actual dimensions 3.5 7.25
in. after it has been surfaced. Of course, the net dimensions of surfaced
lumber should be used in all engineering computations. Therefore, net
dimensions and the corresponding properties (in USCS units) are given
in Appendix F. Similar tables are available in SI units.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.6 Design of Beams for Bending Stresses
y
y
z
O
y
h
z
O
z
O
y
A
—
2
h
z
323
Flange
Web
O
Flange
b
d
(a)
(b)
FIG. 5-18 Cross-sectional shapes
of beams
(c)
A
—
2
(d)
Relative Efficiency of Various Beam Shapes
One of the objectives in designing a beam is to use the material as efficiently as possible within the constraints imposed by function,
appearance, manufacturing costs, and the like. From the standpoint of
strength alone, efficiency in bending depends primarily upon the shape of
the cross section. In particular, the most efficient beam is one in which
the material is located as far as practical from the neutral axis. The farther a given amount of material is from the neutral axis, the larger the
section modulus becomes—and the larger the section modulus, the larger
the bending moment that can be resisted (for a given allowable stress).
As an illustration, consider a cross section in the form of a rectangle
of width b and height h (Fig. 5-18a). The section modulus (from
Eq. 5-18b) is
Ah
bh2
S ᎏᎏ ᎏᎏ 0.167Ah
6
6
(5-25)
where A denotes the cross-sectional area. This equation shows that a rectangular cross section of given area becomes more efficient as the height
h is increased (and the width b is decreased to keep the area constant). Of
course, there is a practical limit to the increase in height, because the
beam becomes laterally unstable when the ratio of height to width
becomes too large. Thus, a beam of very narrow rectangular section will
fail due to lateral (sideways) buckling rather than to insufficient strength
of the material.
Next, let us compare a solid circular cross section of diameter d
(Fig. 5-18b) with a square cross section of the same area. The side h of a
. The corresquare having the same area as the circle is h (d/2)p
sponding section moduli (from Eqs. 5-18b and 5-19b) are
h3
p p
d 3
Ssquare ᎏᎏ ᎏᎏ 0.1160d 3
48
6
3
pd
Scircle ᎏᎏ 0.0982d 3
32
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(5-26a)
(5-26b)
324
CHAPTER 5 Stresses in Beams (Basic Topics)
from which we get
Ssquare
ᎏᎏ 1.18
Scircle
(5-27)
This result shows that a beam of square cross section is more efficient in
resisting bending than is a circular beam of the same area. The reason, of
course, is that a circle has a relatively larger amount of material located
near the neutral axis. This material is less highly stressed, and therefore
it does not contribute as much to the strength of the beam.
The ideal cross-sectional shape for a beam of given cross-sectional
area A and height h would be obtained by placing one-half of the area at
a distance h/2 above the neutral axis and the other half at distance h/2
below the neutral axis, as shown in Fig. 5-18c. For this ideal shape, we
obtain
冢 冣冢 冣
A h
I 2 ᎏᎏ ᎏᎏ
2 2
2
Ah2
ᎏᎏ
4
I
S ᎏᎏ 0.5Ah
h/2
(5-28a,b)
These theoretical limits are approached in practice by wide-flange
sections and I-sections, which have most of their material in the flanges
(Fig. 5-18d). For standard wide-flange beams, the section modulus is
approximately
S 0.35Ah
(5-29)
which is less than the ideal but much larger than the section modulus for
a rectangular cross section of the same area and height (see Eq. 5-25).
Another desirable feature of a wide-flange beam is its greater width,
and hence greater stability with respect to sideways buckling, when
compared to a rectangular beam of the same height and section modulus.
On the other hand, there are practical limits to how thin we can make the
web of a wide-flange beam. If the web is too thin, it will be susceptible
to localized buckling or it may be overstressed in shear, a topic that is
discussed in Section 5.10.
The following four examples illustrate the process of selecting a
beam on the basis of the allowable stresses. In these examples, only the
effects of bending stresses (obtained from the flexure formula) are
considered.
Note: When solving examples and problems that require the selection of a steel or wood beam from the tables in the appendix, we use the
following rule: If several choices are available in a table, select the lightest beam that will provide the required section modulus.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
@À;;;
@@
ÀÀ
;@À;
@@
ÀÀ
;;
SECTION 5.6 Design of Beams for Bending Stresses
325
Example 5-5
;@À;@À
q = 420 lb/ft
L = 12 ft
À
@
;
;@À;@À
FIG. 5-19 Example 5-5. Design of a
simply supported wood beam
A simply supported wood beam having a span length L 12 ft carries a uniform
load q 420 lb/ft (Fig. 5-19). The allowable bending stress is 1800 psi, the wood
weighs 35 lb/ft3, and the beam is supported laterally against sideways buckling
and tipping.
Select a suitable size for the beam from the table in Appendix F.
Solution
Since we do not know in advance how much the beam weighs, we will
proceed by trial-and-error as follows: (1) Calculate the required section modulus
based upon the given uniform load. (2) Select a trial size for the beam. (3) Add
the weight of the beam to the uniform load and calculate a new required section
modulus. (4) Check to see that the selected beam is still satisfactory. If it is not,
select a larger beam and repeat the process.
(1) The maximum bending moment in the beam occurs at the midpoint (see
Eq. 4-15):
(420 lb/ft)(12 ft)2(12 in./ft)
qL2
Mmax ᎏᎏ 90,720 lb-in.
8
8
The required section modulus (Eq. 5-24) is
Mmax
90,720 lb-in.
S ᎏᎏ ᎏᎏ 50.40 in.3
sallow
1800 psi
(2) From the table in Appendix F we see that the lightest beam that supplies
a section modulus of at least 50.40 in.3 about axis 1-1 is a 3 12 in. beam
(nominal dimensions). This beam has a section modulus equal to 52.73 in.3 and
weighs 6.8 lb/ft. (Note that Appendix F gives weights of beams based upon a
density of 35 lb/ft3.)
(3) The uniform load on the beam now becomes 426.8 lb/ft, and the corresponding required section modulus is
冢
冣
426.8 lb/ft
S (50.40 in.3) 51.22 in.3
420 lb/ft
(4) The previously selected beam has a section modulus of 52.73 in.3, which
is larger than the required modulus of 51.22 in.3
Therefore, a 3 12 in. beam is satisfactory.
Note: If the weight density of the wood is other than 35 lb/ft3, we can obtain
the weight of the beam per linear foot by multiplying the value in the last column
in Appendix F by the ratio of the actual weight density to 35 lb/ft3.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
326
CHAPTER 5 Stresses in Beams (Basic Topics)
Example 5-6
P = 12 kN
P = 12 kN
d2
d1
h = 2.5 m
h = 2.5 m
A vertical post 2.5-meters high must support a lateral load P 12 kN at its upper
end (Fig. 5-20). Two plans are proposed—a solid wood post and a hollow
aluminum tube.
(a) What is the minimum required diameter d1 of the wood post if the allowable bending stress in the wood is 15 MPa?
(b) What is the minimum required outer diameter d2 of the aluminum tube
if its wall thickness is to be one-eighth of the outer diameter and the allowable
bending stress in the aluminum is 50 MPa?
Solution
Maximum bending moment. The maximum moment occurs at the base of the
post and is equal to the load P times the height h; thus,
(a)
Mmax Ph (12 kN)(2.5 m) 30 kNm
(b)
FIG. 5-20 Example 5-6. (a) Solid wood
post, and (b) aluminum tube
(a) Wood post. The required section modulus S1 for the wood post (see Eqs.
5-19b and 5-24) is
p d 31
M m ax
30 kNm
S1 0.0020 m3 2106 mm3
32
sallow
15 MPa
Solving for the diameter, we get
d1273 mm
The diameter selected for the wood post must be equal to or larger than 273 mm
if the allowable stress is not to be exceeded.
(b) Aluminum tube. To determine the section modulus S2 for the tube, we
first must find the moment of inertia I2 of the cross section. The wall thickness
of the tube is d2/8, and therefore the inner diameter is d2 d2 /4, or 0.75d2. Thus,
the moment of inertia (see Eq. 5-19a) is
p
I2 d 42 (0.75d2)4 0.03356d 42
64
The section modulus of the tube is now obtained from Eq. (5-17) as follows:
I2
0.03356d 42
S2 0.06712d 32
c
d2/2
The required section modulus is obtained from Eq. (5-24):
Mmax
30 kNm
S2 0.0006 m3 600 103 mm3
sallow
50 MPa
By equating the two preceding expressions for the section modulus, we can solve
for the required outer diameter:
600 103 mm3
d 2
0.06712
1/ 3
208 mm
The corresponding inner diameter is 0.75(208 mm), or 156 mm.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.6 Design of Beams for Bending Stresses
327
Example 5-7
A simple beam AB of span length 21 ft must support a uniform load q 2000 lb/ft
distributed along the beam in the manner shown in Fig. 5-21a.
Considering both the uniform load and the weight of the beam, and also
using an allowable bending stress of 18,000 psi, select a structural steel beam of
wide-flange shape to support the loads.
q = 2,000 lb/ft
q = 2,000 lb/ft
B
A
12 ft
3 ft
6 ft
RA
RB
(a)
18,860
V
(lb)
0
x1
FIG. 5-21 Example 5-7. Design of a
simple beam with partial uniform loads
−5,140
−17,140
(b)
Solution
In this example we will proceed as follows: (1) Find the maximum bending
moment in the beam due to the uniform load. (2) Knowing the maximum
moment, find the required section modulus. (3) Select a trial wide-flange beam
from Table E-1 in Appendix E and obtain the weight of the beam. (4) With the
weight known, calculate a new value of the bending moment and a new value of
the section modulus. (5) Determine whether the selected beam is still satisfactory. If it is not, select a new beam size and repeat the process until a satisfactory
size of beam has been found.
Maximum bending moment. To assist in locating the cross section of maximum bending moment, we construct the shear-force diagram (Fig. 5-21b) using
the methods described in Chapter 4. As part of that process, we determine the
reactions at the supports:
RA 18,860 lb
RB 17,140 lb
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
328
CHAPTER 5 Stresses in Beams (Basic Topics)
The distance x1 from the left-hand support to the cross section of zero shear force
is obtained from the equation
V RA qx1 0
which is valid in the range 0
x
12 ft. Solving for x1, we get
RA
18,860 lb
x1 9.430 ft
q
2,000 lb/ft
which is less than 12 ft, and therefore the calculation is valid.
The maximum bending moment occurs at the cross section where the shear
force is zero; therefore,
qx 12
Mmax RA x1 88,920 lb-ft
2
Required section modulus. The required section modulus (based only upon
the load q) is obtained from Eq. (5-24):
(88,920 lb-ft)(12 in./ft)
M ax
S m
59.3 in.3
18,000 psi
sallow
Trial beam. We now turn to Table E-1 and select the lightest wide-flange
beam having a section modulus greater than 59.3 in.3 The lightest beam that provides this section modulus is W 12 50 with S 64.7 in.3 This beam weighs
50 lb/ft. (Recall that the tables in Appendix E are abridged, and therefore a lighter
beam may actually be available.)
We now recalculate the reactions, maximum bending moment, and required
section modulus with the beam loaded by both the uniform load q and its own
weight. Under these combined loads the reactions are
RA 19,380 lb
RB 17,670 lb
and the distance to the cross section of zero shear becomes
19,380 lb
x1 9.454 ft
2,050 lb/ft
The maximum bending moment increases to 91,610 lb-ft, and the new required
section modulus is
(91,610 lb-ft)(12 in./ft)
M ax
S m
61.1 in.3
18,000 psi
sallow
Thus, we see that the W 12 50 beam with section modulus S 64.7 in.3 is still
satisfactory.
Note: If the new required section modulus exceeded that of the W 12 50
beam, a new beam with a larger section modulus would be selected and the
process repeated.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.6 Design of Beams for Bending Stresses
329
Example 5-8
A temporary wood dam is constructed of horizontal planks A supported by
vertical wood posts B that are sunk into the ground so that they act as cantilever
beams (Fig. 5-22). The posts are of square cross section (dimensions b b) and
spaced at distance s 0.8 m, center to center. Assume that the water level behind
the dam is at its full height h 2.0 m.
Determine the minimum required dimension b of the posts if the allowable
bending stress in the wood is sallow 8.0 MPa.
b
b
b
B
s
B
h
A
B
A
Solution
Loading diagram. Each post is subjected to a triangularly distributed load
produced by the water pressure acting against the planks. Consequently, the loading diagram for each post is triangular (Fig. 5-22c). The maximum intensity q0
of the load on the posts is equal to the water pressure at depth h times the spacing
s of the posts:
(a)
q0 g hs
(a) Top view
(b) Side view
in which g is the specific weight of water. Note that q0 has units of force per unit
distance, g has units of force per unit volume, and both h and s have units of
length.
Section modulus. Since each post is a cantilever beam, the maximum bending moment occurs at the base and is given by the following expression:
gh3s
qh h
Mmax ᎏ0ᎏ ᎏᎏ ᎏᎏ
2 3
6
冢冣
h
(b)
Therefore, the required section modulus (Eq. 5-24) is
B
M ax
gh3s
S ᎏmᎏ
ᎏᎏ
sallow
6sallow
(c)
For a beam of square cross section, the section modulus is S b3/6 (see Eq.
5-18b). Substituting this expression for S into Eq. (c), we get a formula for the
cube of the minimum dimension b of the posts:
q0
(c) Loading diagram
gh3s
b 3 ᎏᎏ
sallow
FIG. 5-22 Example 5-8. Wood dam with
Numerical values. We now substitute numerical values into Eq. (d) and
obtain
horizontal planks A supported by
vertical posts B
(d)
(9.81 kN/m3)(2.0 m)3(0.8 m)
b3 0.007848 m3 7.848 106 mm3
8.0 MPa
from which
b 199 mm
Thus, the minimum required dimension b of the posts is 199 mm. Any larger
dimension, such as 200 mm, will ensure that the actual bending stress is less than
the allowable stress.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
330
CHAPTER 5 Stresses in Beams (Basic Topics)
5.7 NONPRISMATIC BEAMS
The beam theories described in this chapter were derived for prismatic
beams, that is, straight beams having the same cross sections throughout
their lengths. However, nonprismatic beams are commonly used to
reduce weight and improve appearance. Such beams are found in automobiles, airplanes, machinery, bridges, buildings, tools, and many other
applications (Fig. 5-23). Fortunately, the flexure formula (Eq. 5-13) gives
reasonably accurate values for the bending stresses in nonprismatic
beams whenever the changes in cross-sectional dimensions are gradual,
as in the examples shown in Fig. 5-23.
The manner in which the bending stresses vary along the axis of a
nonprismatic beam is not the same as for a prismatic beam. In a prismatic
beam the section modulus S is constant, and therefore the stresses vary in
direct proportion to the bending moment (because s M/S). However,
in a nonprismatic beam the section modulus also varies along the axis.
Consequently, we cannot assume that the maximum stresses occur at the
cross section with the largest bending moment—sometimes the maximum stresses occur elsewhere, as illustrated in Example 5-9.
(b)
(c)
FIG. 5-23 Examples of nonprismatic
beams: (a) street lamp, (b) bridge with
tapered girders and piers, (c) wheel strut
of a small airplane, and (d) wrench
handle
(d)
(a)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.7 Nonprismatic Beams
331
Fully Stressed Beams
To minimize the amount of material and thereby have the lightest possible beam, we can vary the dimensions of the cross sections so as to have
the maximum allowable bending stress at every section. A beam in this
condition is called a fully stressed beam, or a beam of constant strength.
Of course, these ideal conditions are seldom attained because of
practical problems in constructing the beam and the possibility of the
loads being different from those assumed in design. Nevertheless, knowing the properties of a fully stressed beam can be an important aid to the
engineer when designing structures for minimum weight. Familiar examples of structures designed to maintain nearly constant maximum stress
are leaf springs in automobiles, bridge girders that are tapered, and some
of the structures shown in Fig. 5-23.
The determination of the shape of a fully stressed beam is illustrated
in Example 5-10.
Example 5-9
A tapered cantilever beam AB of solid circular cross section supports a load P at
the free end (Fig. 5-24). The diameter dB at the large end is twice the diameter dA
at the small end:
dB
ᎏᎏ 2
dA
Determine the bending stress sB at the fixed support and the maximum bending
stress smax.
B
dB
A
dA
x
P
L
FIG. 5-24 Example 5-9. Tapered
cantilever beam of circular cross section
Solution
If the angle of taper of the beam is small, the bending stresses obtained from
the flexure formula will differ only slightly from the exact values. As a guideline
concerning accuracy, we note that if the angle between line AB (Fig. 5-24) and
the longitudinal axis of the beam is about 20°, the error in calculating the normal
stresses from the flexure formula is about 10%. Of course, as the angle of taper
decreases, the error becomes smaller.
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
332
CHAPTER 5 Stresses in Beams (Basic Topics)
Section modulus. The section modulus at any cross section of the beam can
be expressed as a function of the distance x measured along the axis of the beam.
Since the section modulus depends upon the diameter, we first must express the
diameter in terms of x, as follows:
x
(5-30)
dx dA (dB dA)
L
in which dx is the diameter at distance x from the free end. Therefore, the section
modulus at distance x from the end (Eq. 5-19b) is
p d x3
p
x 3
Sx dA(dBdA)
(5-31)
32
32
L
Bending stresses. Since the bending moment equals Px, the maximum
normal stress at any cross section is given by the equation
Mx
32Px
s1 3
p [dA (dB dA)(x/L)]
Sx
(5-32)
We can see by inspection of the beam that the stress s1 is tensile at the top of the
beam and compressive at the bottom.
Note that Eqs. (5-30), (5-31), and (5-32) are valid for any values of dA and
dB, provided the angle of taper is small. In the following discussion, we consider
only the case where dB 2dA.
Maximum stress at the fixed support. The maximum stress at the section of
largest bending moment (end B of the beam) can be found from Eq. (5-32) by
substituting x L and dB 2dA; the result is
4PL
sB 3
pdA
(a)
Maximum stress in the beam. The maximum stress at a cross section at distance x from the end (Eq. 5-32) for the case where dB 2dA is
32Px
s1
pdA3(1 x/L)3
(b)
To determine the location of the cross section having the largest bending stress
in the beam, we need to find the value of x that makes s1 a maximum. Taking the
derivative ds1/dx and equating it to zero, we can solve for the value of x that
makes s1 a maximum; the result is
L
x
2
(c)
The corresponding maximum stress, obtained by substituting x L /2 into
Eq. (b), is
128P L
4.741PL
smax 3
27pdA
pd A3
(d)
In this particular example, the maximum stress occurs at the midpoint of the
beam and is 19% greater than the stress sB at the built-in end.
Note: If the taper of the beam is reduced, the cross section of maximum
normal stress moves from the midpoint toward the fixed support. For small
angles of taper, the maximum stress occurs at end B.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.7 Nonprismatic Beams
333
Example 5-10
B
P
A
hB
hx
x
b
L
FIG. 5-25 Example 5-10. Fully stressed
beam having constant maximum normal
stress (theoretical shape with shear
stresses disregarded)
A cantilever beam AB of length L is being designed to support a concentrated
load P at the free end (Fig. 5-25). The cross sections of the beam are rectangular
with constant width b and varying height h. To assist them in designing this
beam, the designers would like to know how the height of an idealized beam
should vary in order that the maximum normal stress at every cross section will
be equal to the allowable stress sallow.
Considering only the bending stresses obtained from the flexure formula,
determine the height of the fully stressed beam.
Solution
The bending moment and section modulus at distance x from the free end of
the beam are
M Px
bh x2
S ᎏᎏ
6
where hx is the height of the beam at distance x. Substituting in the flexure
formula, we obtain
Px
M
6Px
sallow ᎏᎏ ᎏ2ᎏ ᎏᎏ2
bh x /6
S
bh x
(e)
Solving for the height of the beam, we get
hx
Px
ᎏ
冪ᎏ
莦
bs
6
(f)
allow
At the fixed end of the beam (x L), the height hB is
hB
6PL
ᎏ
冪ᎏ
莦
bs
(g)
allow
and therefore we can express the height hx in the following form:
冪莦
x
hx hB ᎏᎏ
L
(h)
This last equation shows that the height of the fully stressed beam varies with the
square root of x. Consequently, the idealized beam has the parabolic shape shown
in Fig. 5-25.
Note: At the loaded end of the beam (x 0) the theoretical height is zero,
because there is no bending moment at that point. Of course, a beam of this shape
is not practical because it is incapable of supporting the shear forces near the end
of the beam. Nevertheless, the idealized shape can provide a useful starting point
for a realistic design in which shear stresses and other effects are considered.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
334
CHAPTER 5 Stresses in Beams (Basic Topics)
5.8 SHEAR STRESSES IN BEAMS OF RECTANGULAR CROSS SECTION
When a beam is in pure bending, the only stress resultants are the bending moments and the only stresses are the normal stresses acting on the
cross sections. However, most beams are subjected to loads that produce
both bending moments and shear forces (nonuniform bending). In these
cases, both normal and shear stresses are developed in the beam. The
normal stresses are calculated from the flexure formula (see Section 5.5),
provided the beam is constructed of a linearly elastic material. The shear
stresses are discussed in this and the following two sections.
Vertical and Horizontal Shear Stresses
y
b
n
h
t
m
O
z
x
V
(a)
t
n
t
t
t
t
m
(b)
(c)
FIG. 5-26 Shear stresses in a beam of
rectangular cross section
Consider a beam of rectangular cross section (width b and height h)
subjected to a positive shear force V (Fig. 5-26a). It is reasonable to
assume that the shear stresses t acting on the cross section are parallel to
the shear force, that is, parallel to the vertical sides of the cross section.
It is also reasonable to assume that the shear stresses are uniformly
distributed across the width of the beam, although they may vary over the
height. Using these two assumptions, we can determine the intensity of
the shear stress at any point on the cross section.
For purposes of analysis, we isolate a small element mn of the beam
(Fig. 5-26a) by cutting between two adjacent cross sections and between
two horizontal planes. According to our assumptions, the shear stresses t
acting on the front face of this element are vertical and uniformly
distributed from one side of the beam to the other. Also, from the discussion of shear stresses in Section 1.6, we know that shear stresses acting
on one side of an element are accompanied by shear stresses of equal
magnitude acting on perpendicular faces of the element (see Figs. 5-26b
and c). Thus, there are horizontal shear stresses acting between horizontal
layers of the beam as well as vertical shear stresses acting on the cross
sections. At any point in the beam, these complementary shear stresses
are equal in magnitude.
The equality of the horizontal and vertical shear stresses acting on an
element leads to an important conclusion regarding the shear stresses at
the top and bottom of the beam. If we imagine that the element mn
(Fig. 5-26a) is located at either the top or the bottom, we see that the
horizontal shear stresses must vanish, because there are no stresses on the
outer surfaces of the beam. It follows that the vertical shear stresses
must also vanish at those locations; in other words, t 0 where
y h/2.
The existence of horizontal shear stresses in a beam can be demonstrated by a simple experiment. Place two identical rectangular beams on
simple supports and load them by a force P, as shown in Fig. 5-27a. If
friction between the beams is small, the beams will bend independently
(Fig. 5-27b). Each beam will be in compression above its own neutral
axis and in tension below its neutral axis, and therefore the bottom surface
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
335
SECTION 5.8 Shear Stresses in Beams of Rectangular Cross Section
of the upper beam will slide with respect to the top surface of the lower
beam.
Now suppose that the two beams are glued along the contact surface,
so that they become a single solid beam. When this beam is loaded, horizontal shear stresses must develop along the glued surface in order to
prevent the sliding shown in Fig. 5-27b. Because of the presence of these
shear stresses, the single solid beam is much stiffer and stronger than the
two separate beams.
P
(a)
Derivation of Shear Formula
P
We are now ready to derive a formula for the shear stresses t in a rectangular beam. However, instead of evaluating the vertical shear stresses
acting on a cross section, it is easier to evaluate the horizontal shear
stresses acting between layers of the beam. Of course, the vertical shear
stresses have the same magnitudes as the horizontal shear stresses.
With this procedure in mind, let us consider a beam in nonuniform
bending (Fig. 5-28a). We take two adjacent cross sections mn and m1n1,
distance dx apart, and consider the element mm1n1n. The bending
moment and shear force acting on the left-hand face of this element are
denoted M and V, respectively. Since both the bending moment and shear
(b)
FIG. 5-27 Bending of two separate beams
m
M
m1
m
s1
M ⫹ dM
V
m1
s2
M
p1
p
y1
x
h
M ⫹ dM —
2
x
h
—
2
V ⫹ dV
n
dx
n1
Side view of beam
(a)
m
n1
Side view of element
(b)
y
m1
s2
s1
p
dx
n
t
p1
y1
h
—
2
h
—
2
x
z
h
—
2
dA
y1
y
O
dx
b
FIG. 5-28 Shear stresses in a beam
of rectangular cross section
Side view of subelement
(c)
Cross section of beam at subelement
(d)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
336
CHAPTER 5 Stresses in Beams (Basic Topics)
force may change as we move along the axis of the beam, the corresponding quantities on the right-hand face (Fig. 5-28a) are denoted M ⫹ dM
and V ⫹ dV.
Because of the presence of the bending moments and shear forces,
the element shown in Fig. 5-28a is subjected to normal and shear stresses
on both cross-sectional faces. However, only the normal stresses are
needed in the following derivation, and therefore only the normal stresses
are shown in Fig. 5-28b. On cross sections mn and m1n1 the normal
stresses are, respectively,
My
(M⫹dM)y
s1 and s2ᎏ
(a,b)
I
I
as given by the flexure formula (Eq. 5-13). In these expressions, y is the
distance from the neutral axis and I is the moment of inertia of the crosssectional area about the neutral axis.
Next, we isolate a subelement mm1 p1 p by passing a horizontal plane
pp1 through element mm1n1n (Fig. 5-28b). The plane pp1 is at distance y1
from the neutral surface of the beam. The subelement is shown separately
in Fig. 5-28c. We note that its top face is part of the upper surface of the
beam and thus is free from stress. Its bottom face (which is parallel to the
neutral surface and distance y1 from it) is acted upon by the horizontal
shear stresses t existing at this level in the beam. Its cross-sectional faces
mp and m1 p1 are acted upon by the bending stresses s1 and s2, respectively, produced by the bending moments. Vertical shear stresses also act
on the cross-sectional faces; however, these stresses do not affect the
equilibrium of the subelement in the horizontal direction (the x direction),
so they are not shown in Fig. 5-28c.
If the bending moments at cross sections mn and m1n1 (Fig. 5-28b)
are equal (that is, if the beam is in pure bending), the normal stresses s1
and s2 acting over the sides mp and m1p1 of the subelement (Fig. 5-28c)
also will be equal. Under these conditions, the subelement will be in
equilibrium under the action of the normal stresses alone, and therefore
the shear stresses t acting on the bottom face pp1 will vanish. This
conclusion is obvious inasmuch as a beam in pure bending has no shear
force and hence no shear stresses.
If the bending moments vary along the x axis (nonuniform bending),
we can determine the shear stress t acting on the bottom face of the
subelement (Fig. 5-28c) by considering the equilibrium of the subelement
in the x direction.
We begin by identifying an element of area dA in the cross section at
distance y from the neutral axis (Fig. 5-28d). The force acting on this
element is sdA, in which s is the normal stress obtained from the flexure
formula. If the element of area is located on the left-hand face mp of the
subelement (where the bending moment is M), the normal stress is given
by Eq. (a), and therefore the element of force is
My
s1dA ᎏᎏ dA
I
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.8 Shear Stresses in Beams of Rectangular Cross Section
337
Note that we are using only absolute values in this equation because the
directions of the stresses are obvious from the figure. Summing these
elements of force over the area of face mp of the subelement (Fig. 5-28c)
gives the total horizontal force F1 acting on that face:
m
m1
p
p1
F1
F2
F3
y1
h
—
2
x
dx
FIG. 5-29 Partial free-body diagram of
subelement showing all horizontal
forces (compare with Fig. 5-28c)
My
(c)
F1 s1 dA ᎏᎏ dA
I
Note that this integration is performed over the area of the shaded part of
the cross section shown in Fig. 5-28d, that is, over the area of the cross
section from y y1 to y h/2.
The force F1 is shown in Fig. 5-29 on a partial free-body diagram of
the subelement (vertical forces have been omitted).
In a similar manner, we find that the total force F2 acting on the
right-hand face m1p1 of the subelement (Fig. 5-29 and Fig. 5-28c) is
(M⫹dM)y
F2 s2 dA ᎏᎏ dA
I
(d)
Knowing the forces F1 and F2, we can now determine the horizontal
force F3 acting on the bottom face of the subelement.
Since the subelement is in equilibrium, we can sum forces in the x
direction and obtain
F3 F2 – F1
or
(e)
(M ⫹ dM)y
My
(dM)y
F3 ᎏᎏ dA ⫺ ᎏᎏdA ᎏᎏ dA
I
I
I
The quantities dM and I in the last term can be moved outside the integral
sign because they are constants at any given cross section and are not
involved in the integration. Thus, the expression for the force F3 becomes
dM
F3 ᎏᎏ ydA
I
(5-33)
If the shear stresses t are uniformly distributed across the width b of the
beam, the force F3 is also equal to the following:
F3 t b dx
(5-34)
in which b dx is the area of the bottom face of the subelement.
Combining Eqs. (5-33) and (5-34) and solving for the shear stress t,
we get
dM 1
(5-35)
t ᎏᎏ ᎏᎏ y dA
dx I b
The quantity dM/dx is equal to the shear force V (see Eq. 4-6), and therefore the preceding expression becomes
V
t ᎏᎏ y dA
Ib
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(5-36)
338
CHAPTER 5 Stresses in Beams (Basic Topics)
The integral in this equation is evaluated over the shaded part of the cross
section (Fig. 5-28d), as already explained. Thus, the integral is the first
moment of the shaded area with respect to the neutral axis (the z axis). In
other words, the integral is the first moment of the cross-sectional area
above the level at which the shear stress t is being evaluated. This first
moment is usually denoted by the symbol Q:
Q y dA
(5-37)
With this notation, the equation for the shear stress becomes
VQ
t ᎏᎏ
Ib
(5-38)
This equation, known as the shear formula, can be used to determine
the shear stress t at any point in the cross section of a rectangular beam.
Note that for a specific cross section, the shear force V, moment of
inertia I, and width b are constants. However, the first moment Q
(and hence the shear stress t) varies with the distance y1 from the neutral
axis.
Calculation of the First Moment Q
If the level at which the shear stress is to be determined is above the
neutral axis, as shown in Fig. 5-28d, it is natural to obtain Q by calculating the first moment of the cross-sectional area above that level (the
shaded area in the figure). However, as an alternative, we could calculate
the first moment of the remaining cross-sectional area, that is, the area
below the shaded area. Its first moment is equal to the negative of Q.
The explanation lies in the fact that the first moment of the entire
cross-sectional area with respect to the neutral axis is equal to zero
(because the neutral axis passes through the centroid). Therefore, the
value of Q for the area below the level y1 is the negative of Q for the area
above that level. As a matter of convenience, we usually use the area
above the level y1 when the point where we are finding the shear stress is
in the upper part of the beam, and we use the area below the level y1 when
the point is in the lower part of the beam.
Furthermore, we usually don’t bother with sign conventions for V
and Q. Instead, we treat all terms in the shear formula as positive
quantities and determine the direction of the shear stresses by inspection, since the stresses act in the same direction as the shear force V itself.
This procedure for determining shear stresses is illustrated later in
Example 5-11.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.8 Shear Stresses in Beams of Rectangular Cross Section
y
Distribution of Shear Stresses in a Rectangular Beam
h
2
y1
z
O
h
2
We are now ready to determine the distribution of the shear stresses in a
beam of rectangular cross section (Fig. 5-30a). The first moment Q of the
shaded part of the cross-sectional area is obtained by multiplying the area
by the distance from its own centroid to the neutral axis:
冣冢
冢
冣
冢
h/2 y1
h
b h2
Q b ᎏᎏ y1 y1 y 12
2
2
2 4
冣
(f)
Of course, this same result can be obtained by integration using Eq. (5-37):
b
(a)
h/2
Q y dA
y1
b h2
yb dy y12
2 4
(g)
Substituting the expression for Q into the shear formula (Eq. 5-38), we
get
t
h
2
339
tmax
V h2
t y 12
2I 4
h
2
(b)
FIG. 5-30 Distribution of shear stresses in
a beam of rectangular cross section:
(a) cross section of beam, and
(b) diagram showing the parabolic
distribution of shear stresses over the
height of the beam
(5-39)
This equation shows that the shear stresses in a rectangular beam vary
quadratically with the distance y1 from the neutral axis. Thus, when plotted along the height of the beam, t varies as shown in Fig. 5-30b. Note
that the shear stress is zero when y1 h/2.
The maximum value of the shear stress occurs at the neutral axis
(y1 0) where the first moment Q has its maximum value. Substituting
y1 0 into Eq. (5-39), we get
Vh2
3V
tmax
8I
2A
(5-40)
in which A bh is the cross-sectional area. Thus, the maximum shear
stress in a beam of rectangular cross section is 50% larger than the average shear stress V/A.
Note again that the preceding equations for the shear stresses can be
used to calculate either the vertical shear stresses acting on the cross sections or the horizontal shear stresses acting between horizontal layers of
the beam.*
Limitations
The formulas for shear stresses presented in this section are subject to the
same restrictions as the flexure formula from which they are derived.
*The shear-stress analysis presented in this section was developed by the Russian
engineer D. J. Jourawski; see Refs. 5-7 and 5-8.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
340
CHAPTER 5 Stresses in Beams (Basic Topics)
Thus, they are valid only for beams of linearly elastic materials with
small deflections.
In the case of rectangular beams, the accuracy of the shear formula
depends upon the height-to-width ratio of the cross section. The formula
may be considered as exact for very narrow beams (height h much larger
than the width b). However, it becomes less accurate as b increases
relative to h. For instance, when the beam is square (b h), the true maximum shear stress is about 13% larger than the value given by Eq. (5-40).
(For a more complete discussion of the limitations of the shear formula,
see Ref. 5-9.)
A common error is to apply the shear formula (Eq. 5-38) to crosssectional shapes for which it is not applicable. For instance, it is not
applicable to sections of triangular or semicircular shape. To avoid
misusing the formula, we must keep in mind the following assumptions
that underlie the derivation: (1) The edges of the cross section must be
parallel to the y axis (so that the shear stresses act parallel to the y axis),
and (2) the shear stresses must be uniform across the width of the cross
section. These assumptions are fulfilled only in certain cases, such as
those discussed in this and the next two sections.
Finally, the shear formula applies only to prismatic beams. If a beam
is nonprismatic (for instance, if the beam is tapered), the shear stresses
are quite different from those predicted by the formulas given here (see
Refs. 5-9 and 5-10).
Effects of Shear Strains
m1
m
p1
p
n
n1
P
q
q1
FIG. 5-31 Warping of the cross sections
of a beam due to shear strains
Because the shear stress t varies parabolically over the height of a
rectangular beam, it follows that the shear strain g t/G also varies
parabolically. As a result of these shear strains, cross sections of the beam
that were originally plane surfaces become warped. This warping is shown
in Fig. 5-31, where cross sections mn and pq, originally plane, have
become curved surfaces m1n1 and p1q1, with the maximum shear strain
occurring at the neutral surface. At points m1, p1, n1, and q1 the shear strain
is zero, and therefore the curves m1n1 and p1q1 are perpendicular to the
upper and lower surfaces of the beam.
If the shear force V is constant along the axis of the beam, warping
is the same at every cross section. Therefore, stretching and shortening of
longitudinal elements due to the bending moments is unaffected by the
shear strains, and the distribution of the normal stresses is the same as in
pure bending. Moreover, detailed investigations using advanced methods
of analysis show that warping of cross sections due to shear strains does
not substantially affect the longitudinal strains even when the shear force
varies continuously along the length. Thus, under most conditions it is
justifiable to use the flexure formula (Eq. 5-13) for nonuniform bending,
even though the formula was derived for pure bending.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.8 Shear Stresses in Beams of Rectangular Cross Section
341
Example 5-11
q = 160 lb/in.
A
4 in.
3 in.
C
B
8 in.
A metal beam with span L 3 ft is simply supported at points A and B
(Fig. 5-32a). The uniform load on the beam (including its own weight) is
q 160 lb/in. The cross section of the beam is rectangular (Fig. 5-32b) with
width b 1 in. and height h 4 in. The beam is adequately supported against
sideways buckling.
Determine the normal stress sC and shear stress tC at point C, which is
located 1 in. below the top of the beam and 8 in. from the right-hand support.
Show these stresses on a sketch of a stress element at point C.
L = 3 ft
Solution
Shear force and bending moment. The shear force VC and bending moment
MC at the cross section through point C are found by the methods described in
Chapter 4. The results are
(a)
y
MC 17,920 lb-in.
h = 2.0 in.
—
2
1.0 in.
C
y = 1.0 in.
z
O
Normal stress at point C. The normal stress at point C is found from the
flexure formula (Eq. 5-13) with the distance y from the neutral axis equal to
1.0 in.; thus,
b = 1.0 in.
(b)
My
(17,920 lb-in.)(1.0 in.)
3360 psi
sC
5.333 in.4
I
450 psi
C
The signs of these quantities are based upon the standard sign conventions for
bending moments and shear forces (see Fig. 4-5).
Moment of inertia. The moment of inertia of the cross-sectional area about
the neutral axis (the z axis in Fig. 5-32b) is
bh3
1
I (1.0 in.)(4.0 in.)3 5.333 in.4
12
12
h = 2.0 in.
—
2
3360 psi
VC 1,600 lb
3360 psi
450 psi
(c)
The minus sign indicates that the stress is compressive, as expected.
Shear stress at point C. To obtain the shear stress at point C, we need
to evaluate the first moment QC of the cross-sectional area above point C
(Fig. 5-32b). This first moment is equal to the product of the area and its centroidal distance (denoted yC) from the z axis; thus,
AC (1.0 in.)(1.0 in.) 1.0 in.2
FIG. 5-32 Example 5-11. (a) Simple
beam with uniform load, (b) cross
section of beam, and (c) stress element
showing the normal and shear stresses
at point C
yC 1.5 in.
QC ACyC 1.5 in.3
Now we substitute numerical values into the shear formula (Eq. 5-38) and obtain
the magnitude of the shear stress:
(1,600 lb)(1.5 in.3)
VCQC
450 psi
tC
(5.333 in.4)(1.0 in.)
Ib
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
342
CHAPTER 5 Stresses in Beams (Basic Topics)
The direction of this stress can be established by inspection, because it acts in the
same direction as the shear force. In this example, the shear force acts upward on
the part of the beam to the left of point C and downward on the part of the beam
to the right of point C. The best way to show the directions of both the normal
and shear stresses is to draw a stress element, as follows.
Stress element at point C. The stress element, shown in Fig. 5-32c, is cut
from the side of the beam at point C (Fig. 5-32a). Compressive stresses
sC 3360 psi act on the cross-sectional faces of the element and shear stresses
tC 450 psi act on the top and bottom faces as well as the cross-sectional faces.
Example 5-12
P
P
B
A
a
a
(a)
A wood beam AB supporting two concentrated loads P (Fig. 5-33a) has a
rectangular cross section of width b 100 mm and height h 150 mm (Fig.
5-33b). The distance from each end of the beam to the nearest load is a 0.5 m.
Determine the maximum permissible value Pmax of the loads if the allowable stress in bending is sallow 11 MPa (for both tension and compression) and
the allowable stress in horizontal shear is tallow 1.2 MPa. (Disregard the
weight of the beam itself.)
Note: Wood beams are much weaker in horizontal shear (shear parallel to
the longitudinal fibers in the wood) than in cross-grain shear (shear on the cross
sections). Consequently, the allowable stress in horizontal shear is usually
considered in design.
Solution
y
The maximum shear force occurs at the supports and the maximum bending
moment occurs throughout the region between the loads. Their values are
Vmax P
z
O
h
Mmax Pa
Also, the section modulus S and cross-sectional area A are
bh2
S ᎏᎏ
6
b
A bh
The maximum normal and shear stresses in the beam are obtained from the
flexure and shear formulas (Eqs. 5-16 and 5-40):
(b)
FIG. 5-33 Example 5-12. Wood beam
with concentrated loads
Mmax
6P a
smax ᎏᎏ ᎏᎏ
S
bh2
3Vmax
3P
tmax ᎏᎏ ᎏᎏ
2A
2bh
Therefore, the maximum permissible values of the load P in bending and shear,
respectively, are
sallowbh2
Pbending ᎏᎏ
6a
2tallowbh
Pshear ᎏᎏ
3
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.9 Shear Stresses in Beams of Circular Cross Section
343
Substituting numerical values into these formulas, we get
(11 MPa)(100 mm)(150 mm)2
Pbending ᎏᎏᎏᎏ 8.25 kN
6(0.5 m)
2(1.2 MPa)(100 mm)(150 mm)
Pshear ᎏᎏᎏᎏ 12.0 kN
3
Thus, the bending stress governs the design, and the maximum permissible load
is
Pmax 8.25 kN
A more complete analysis of this beam would require that the weight of the beam
be taken into account, thus reducing the permissible load.
Notes: (1) In this example, the maximum normal stresses and maximum
shear stresses do not occur at the same locations in the beam—the normal stress
is maximum in the middle region of the beam at the top and bottom of the cross
section, and the shear stress is maximum near the supports at the neutral axis of
the cross section.
(2) For most beams, the bending stresses (not the shear stresses) control the
allowable load, as in this example.
(3) Although wood is not a homogeneous material and often departs from
linearly elastic behavior, we can still obtain approximate results from the flexure
and shear formulas. These approximate results are usually adequate for designing wood beams.
5.9 SHEAR STRESSES IN BEAMS OF CIRCULAR CROSS SECTION
y
m
r
t
z
p
tmax
O
q
FIG. 5-34 Shear stresses acting on the
cross section of a circular beam
When a beam has a circular cross section (Fig. 5-34), we can no longer
assume that the shear stresses act parallel to the y axis. For instance, we
can easily prove that at point m (on the boundary of the cross section) the
shear stress t must act tangent to the boundary. This observation follows
from the fact that the outer surface of the beam is free of stress, and therefore the shear stress acting on the cross section can have no component
in the radial direction.
Although there is no simple way to find the shear stresses acting
throughout the entire cross section, we can readily determine the shear
stresses at the neutral axis (where the stresses are the largest) by making
some reasonable assumptions about the stress distribution. We assume
that the stresses act parallel to the y axis and have constant intensity
across the width of the beam (from point p to point q in Fig. 5-34). Since
these assumptions are the same as those used in deriving the shear
formula t VQ/Ib (Eq. 5-38), we can use the shear formula to calculate
the stresses at the neutral axis.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
344
CHAPTER 5 Stresses in Beams (Basic Topics)
For use in the shear formula, we need the following properties pertaining to a circular cross section having radius r:
pr4
I ᎏᎏ
4
y
r1
z
p r 2 4r
2r 3
Q Ay ᎏᎏ ᎏᎏ ᎏᎏ
2
3p
3
b 2r
(5-41a,b)
The expression for the moment of inertia I is taken from Case 9 of
Appendix D, and the expression for the first moment Q is based upon the
formulas for a semicircle (Case 10, Appendix D). Substituting these
expressions into the shear formula, we obtain
r2
O
VQ
4V
V (2 r 3/3)
4V
tmax ᎏᎏᎏ
4 ᎏ ᎏᎏ
2 ᎏᎏ
Ib (p r /4 )(2r)
3A
3p r
FIG. 5-35 Hollow circular cross section
(5-42)
in which A p r 2 is the area of the cross section. This equation shows
that the maximum shear stress in a circular beam is equal to 4/3 times the
average vertical shear stress V/A.
If a beam has a hollow circular cross section (Fig. 5-35), we may
again assume with reasonable accuracy that the shear stresses at the
neutral axis are parallel to the y axis and uniformly distributed across the
section. Consequently, we may again use the shear formula to find the
maximum stresses. The required properties for a hollow circular section
are
p
I ᎏᎏ r 24⫺r 14
4
2
Q ᎏᎏ r 23⫺r 13
3
b2(r2⫺r1)
(5-43a,b,c)
in which r1 and r2 are the inner and outer radii of the cross section.
Therefore, the maximum stress is
2
2
VQ
4V r 2 ⫹ r2r1 ⫹ r 1
tmax ᎏᎏ ᎏᎏ ᎏᎏ
r 22 ⫹ r 12
Ib
3A
(5-44)
in which
A p r 22 ⫺ r 12
is the area of the cross section. Note that if r1 0, Eq. (5-44) reduces to
Eq. (5-42) for a solid circular beam.
Although the preceding theory for shear stresses in beams of circular
cross section is approximate, it gives results differing by only a few
percent from those obtained using the exact theory of elasticity (Ref. 5-9).
Consequently, Eqs. (5-42) and (5-44) can be used to determine the
maximum shear stresses in circular beams under ordinary circumstances.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.9 Shear Stresses in Beams of Circular Cross Section
345
Example 5-13
A vertical pole consisting of a circular tube of outer diameter d2 4.0 in. and
inner diameter d1 3.2 in. is loaded by a horizontal force P 1500 lb
(Fig. 5-36a).
(a) Determine the maximum shear stress in the pole.
(b) For the same load P and the same maximum shear stress, what is the
diameter d0 of a solid circular pole (Fig. 5-36b)?
d1
P
P
d2
FIG. 5-36 Example 5-13. Shear stresses
in beams of circular cross section
d0
(a)
(b)
Solution
(a) Maximum shear stress. For the pole having a hollow circular cross section
(Fig. 5-36a), we use Eq. (5-44) with the shear force V replaced by the load P and
the cross-sectional area A replaced by the expression p(r 22 ⫺ r 21); thus,
2
2
4P r 2 ⫹ r2r1 ⫹ r 1
tmax ᎏᎏ ᎏᎏ
4
4
r2 ⫺ r1
3p
冢
冣
(a)
Next, we substitute numerical values, namely,
P 1500 lb
r2 d2/2 2.0 in.
r1 d1/2 1.6 in.
and obtain
tmax 658 psi
which is the maximum shear stress in the pole.
(b) Diameter of solid circular pole. For the pole having a solid circular cross
section (Fig. 5-36b), we use Eq. (5-42) with V replaced by P and r replaced by
d0 /2:
4P
tmax ᎏᎏ
3p (d0/2)2
(b)
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
346
CHAPTER 5 Stresses in Beams (Basic Topics)
Solving for d0, we obtain
16P
16(1500 lb)
d 20 ᎏᎏ ᎏᎏ 3.87 in.2
3p tmax
3p (658 psi)
from which we get
d0 1.97 in.
In this particular example, the solid circular pole has a diameter approximately
one-half that of the tubular pole.
Note: Shear stresses rarely govern the design of either circular or rectangular beams made of metals such as steel and aluminum. In these kinds of
materials, the allowable shear stress is usually in the range 25 to 50% of the
allowable tensile stress. In the case of the tubular pole in this example, the maximum shear stress is only 658 psi. In contrast, the maximum bending stress
obtained from the flexure formula is 9700 psi for a relatively short pole of length
24 in. Thus, as the load increases, the allowable tensile stress will be reached
long before the allowable shear stress is reached.
The situation is quite different for materials that are weak in shear, such as
wood. For a typical wood beam, the allowable stress in horizontal shear is in
the range 4 to 10% of the allowable bending stress. Consequently, even though the
maximum shear stress is relatively low in value, it sometimes governs the design.
5.10 SHEAR STRESSES IN THE WEBS OF BEAMS WITH FLANGES
When a beam of wide-flange shape (Fig. 5-37a) is subjected to shear
forces as well as bending moments (nonuniform bending), both normal
and shear stresses are developed on the cross sections. The distribution of
the shear stresses in a wide-flange beam is more complicated than in a
rectangular beam. For instance, the shear stresses in the flanges of the
beam act in both vertical and horizontal directions (the y and z directions), as shown by the small arrows in Fig. 5-37b. The horizontal shear
stresses, which are much larger than the vertical shear stresses in the
flanges, are discussed later in Section 6.7.
y
FIG. 5-37 (a) Beam of wide-flange shape,
and (b) directions of the shear stresses
acting on a cross section
z
x
(a)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(b)
SECTION 5.10 Shear Stresses in the Webs of Beams with Flanges
347
The shear stresses in the web of a wide-flange beam act only in the
vertical direction and are larger than the stresses in the flanges. These
stresses can be found by the same techniques we used for finding shear
stresses in rectangular beams.
Shear Stresses in the Web
Let us begin the analysis by determining the shear stresses at line ef in the
web of a wide-flange beam (Fig. 5-38a). We will make the same assumptions as those we made for a rectangular beam; that is, we assume that the
shear stresses act parallel to the y axis and are uniformly distributed across
the thickness of the web. Then the shear formula t VQ/Ib will still
apply. However, the width b is now the thickness t of the web, and the area
used in calculating the first moment Q is the area between line ef and the
top edge of the cross section (indicated by the shaded area of Fig. 5-38a).
When finding the first moment Q of the shaded area, we will
disregard the effects of the small fillets at the juncture of the web and
flange (points b and c in Fig. 5-38a). The error in ignoring the areas of
these fillets is very small. Then we will divide the shaded area into two
rectangles. The first rectangle is the upper flange itself, which has area
冢
h1
h
A1 b ᎏᎏ ⫺ ᎏᎏ
2
2
冣
(a)
y
h
h
—
2
z
a
y1
c
f
b
e
d
O
h
—
2
t
h1
2
tmin
t
h1
2
h1
tmax
h1
2
h1
2
tmin
FIG. 5-38 Shear stresses in the web of a
wide-flange beam. (a) Cross section of
beam, and (b) distribution of vertical
shear stresses in the web
(b)
b
(a)
in which b is the width of the flange, h is the overall height of the beam,
and h1 is the distance between the insides of the flanges. The second rectangle is the part of the web between ef and the flange, that is, rectangle
efcb, which has area
冢
h1
A2 t ᎏᎏ ⫺ y1
2
冣
(b)
in which t is the thickness of the web and y1 is the distance from the
neutral axis to line ef.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
348
CHAPTER 5 Stresses in Beams (Basic Topics)
The first moments of areas A1 and A2, evaluated about the neutral
axis, are obtained by multiplying these areas by the distances from their
respective centroids to the z axis. Adding these first moments gives the
first moment Q of the combined area:
冢
冣
冢
h1
h/2 ⫺ h1/2
h1/2 ⫺ y1
Q A1 ᎏᎏ ⫹ ᎏᎏ ⫹ A2 y1 ⫹ ᎏᎏ
2
2
2
冣
Upon substituting for A1 and A2 from Eqs. (a) and (b) and then simplifying, we get
b
t
Q ᎏᎏ(h2 ⫺ h 12) ⫹ ᎏᎏ(h 12 ⫺ 4y 12)
8
8
(5-45)
Therefore, the shear stress t in the web of the beam at distance y1 from
the neutral axis is
VQ
V
t ᎏᎏ ᎏᎏ b(h2 ⫺ h 12) ⫹ t(h 12 ⫺ 4y 12)
It
8It
(5-46)
in which the moment of inertia of the cross section is
bh3
(b ⫺ t)h 3
1
I ᎏᎏ ⫺ ᎏᎏ1 ᎏᎏ (bh3 ⫺ bh 13 ⫹ th 13)
12
12
12
(5-47)
Since all quantities in Eq. (5-46) are constants except y1, we see immediately that t varies quadratically throughout the height of the web, as
shown by the graph in Fig. 5-38b. Note that the graph is drawn only for
the web and does not include the flanges. The reason is simple enough—
Eq. (5-46) cannot be used to determine the vertical shear stresses in the
flanges of the beam (see the discussion titled “Limitations” later in this
section).
Maximum and Minimum Shear Stresses
The maximum shear stress in the web of a wide-flange beam occurs at
the neutral axis, where y1 0. The minimum shear stress occurs where
h1/2). These stresses, found from
the web meets the flanges ( y1
Eq. (5-46), are
V
tmax (bh2 bh 12 ⫹ th 12)
8It
Vb
tmin (h2 h 12)
8It
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(5-48a,b)
SECTION 5.10 Shear Stresses in the Webs of Beams with Flanges
349
Both tmax and tmin are labeled on the graph of Fig. 5-38b. For typical
wide-flange beams, the maximum stress in the web is from 10% to 60%
greater than the minimum stress.
Although it may not be apparent from the preceding discussion, the
stress tmax given by Eq. (5-48a) not only is the largest shear stress in the
web but also is the largest shear stress anywhere in the cross section.
y
h
h
—
2
z
FIG. 5-38 (Repeated) Shear stresses in
the web of a wide-flange beam.
(a) Cross section of beam, and (b) distribution of vertical shear stresses
in the web
a
y1
c
f
b
e
d
O
h
—
2
t
h1
2
h1
h1
2
tmin
t
h1
2
tmax
h1
2
tmin
(b)
b
(a)
Shear Force in the Web
The vertical shear force carried by the web alone may be determined by
multiplying the area of the shear-stress diagram (Fig. 5-38b) by the thickness t of the web. The shear-stress diagram consists of two parts, a
rectangle of area h1tmin and a parabolic segment of area
2
ᎏᎏ (h1)(tmax ⫺ tmin)
3
By adding these two areas, multiplying by the thickness t of the web, and
then combining terms, we get the total shear force in the web:
th1
Vweb ᎏᎏ (2tmax ⫹ tmin)
3
(5-49)
For beams of typical proportions, the shear force in the web is 90% to
98% of the total shear force V acting on the cross section; the remainder
is carried by shear in the flanges.
Since the web resists most of the shear force, designers often calculate an approximate value of the maximum shear stress by dividing the
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
350
CHAPTER 5 Stresses in Beams (Basic Topics)
total shear force by the area of the web. The result is the average shear
stress in the web, assuming that the web carries all of the shear force:
V
taver ᎏᎏ
th1
(5-50)
For typical wide-flange beams, the average stress calculated in this
manner is within 10% (plus or minus) of the maximum shear stress
calculated from Eq. (5-48a). Thus, Eq. (5-50) provides a simple way to
estimate the maximum shear stress.
Limitations
The elementary shear theory presented in this section is suitable for
determining the vertical shear stresses in the web of a wide-flange beam.
However, when investigating vertical shear stresses in the flanges, we
can no longer assume that the shear stresses are constant across the width
of the section, that is, across the width b of the flanges (Fig. 5-38a).
Hence, we cannot use the shear formula to determine these stresses.
To emphasize this point, consider the junction of the web and upper
flange (y1 h1/2), where the width of the section changes abruptly from
t to b. The shear stresses on the free surfaces ab and cd (Fig. 5-38a) must
be zero, whereas the shear stress across the web at line bc is tmin. These
observations indicate that the distribution of shear stresses at the junction
of the web and the flange is quite complex and cannot be investigated by
elementary methods. The stress analysis is further complicated by the use
of fillets at the re-entrant corners (corners b and c). The fillets are necessary to prevent the stresses from becoming dangerously large, but they
also alter the stress distribution across the web.
Thus, we conclude that the shear formula cannot be used to determine
the vertical shear stresses in the flanges. However, the shear formula does
give good results for the shear stresses acting horizontally in the flanges
(Fig. 5-37b), as discussed later in Section 6.8.
The method described above for determining shear stresses in the
webs of wide-flange beams can also be used for other sections having
thin webs. For instance, Example 5-15 illustrates the procedure for a
T-beam.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.10 Shear Stresses in the Webs of Beams with Flanges
351
Example 5-14
A beam of wide-flange shape (Fig. 5-39a) is subjected to a vertical shear force
V 45 kN. The cross-sectional dimensions of the beam are b 165 mm,
t 7.5 mm, h 320 mm, and h1 290 mm.
Determine the maximum shear stress, minimum shear stress, and total shear
force in the web. (Disregard the areas of the fillets when making calculations.)
y
tmin =
17.4 MPa
h=
320 mm
z
O
h1 =
290 mm
tmax =
21.0 MPa
t = 7.5 mm
tmin
b=
165 mm
FIG. 5-39 Example 5-14. Shear stresses
in the web of a wide-flange beam
(b)
(a)
Solution
Maximum and minimum shear stresses. The maximum and minimum shear
stresses in the web of the beam are given by Eqs. (5-48a) and (5-48b). Before
substituting into those equations, we calculate the moment of inertia of the crosssectional area from Eq. (5-47):
1
I ᎏᎏ (bh3 ⫺ bh 31⫹th 31) 130.45 106 mm4
12
Now we substitute this value for I, as well as the numerical values for the shear
force V and the cross-sectional dimensions, into Eqs. (5-48a) and (5-48b):
V
tmax (bh2 bh 21 ⫹ th 21) 21.0 MPa
8It
Vb
tmin ᎏᎏ (h2 ⫺ h 21) 17.4 MPa
8It
In this case, the ratio of tmax to tmin is 1.21, that is, the maximum stress in the web
is 21% larger than the minimum stress. The variation of the shear stresses over the
height h1 of the web is shown in Fig. 5-39b.
Total shear force. The shear force in the web is calculated from Eq. (5-49)
as follows:
th1
Vweb ᎏᎏ (2tmax ⫹ tmin) 43.0 kN
3
From this result we see that the web of this particular beam resists 96% of the
total shear force.
Note: The average shear stress in the web of the beam (from Eq. 5-50) is
V
taver ᎏᎏ 20.7 MPa
th1
which is only 1% less than the maximum stress.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
352
CHAPTER 5 Stresses in Beams (Basic Topics)
Example 5-15
A beam having a T-shaped cross section (Fig. 5-40a) is subjected to a vertical
shear force V 10,000 lb. The cross-sectional dimensions are b 4 in.,
t 1.0 in., h 8.0 in., and h1 7.0 in.
Determine the shear stress t1 at the top of the web (level nn) and the maximum shear stress tmax. (Disregard the areas of the fillets.)
y
b = 4.0 in.
c1
n
z
t1
n
tmax
O
h = 8.0 in.
h1 = 7.0 in.
c2
h1
t = 1.0 in.
a
c2
a
FIG. 5-40 Example 5-15. Shear stresses
in the web of a T-beam
(a)
(b)
Solution
Location of neutral axis. The neutral axis of the T-beam is located by
calculating the distances c1 and c2 from the top and bottom of the beam to the
centroid of the cross section (Fig. 5-40a). First, we divide the cross section into
two rectangles, the flange and the web (see the dashed line in Fig. 5-40a). Then
we calculate the first moment Qaa of these two rectangles with respect to line aa
at the bottom of the beam. The distance c2 is equal to Qaa divided by the area A
of the entire cross section (see Chapter 12, Section 12.3, for methods for locating
centroids of composite areas). The calculations are as follows:
A Ai b(h ⫺ h1) ⫹ th1 11.0 in.2
h ⫹ h1
h1
Qaa yi Ai ᎏᎏ (b)(h ⫺ h1) ⫹ ᎏᎏ(th1) 54.5 in.3
2
2
Qaa
54 .5 in.3
c2 ᎏ ᎏ ᎏᎏ2 4.955 in.
A
11.0 in.
c1 h ⫺ c2 3.045 in.
Moment of inertia. The moment of inertia I of the entire cross-sectional area
(with respect to the neutral axis) can be found by determining the moment of
inertia Iaa about line aa at the bottom of the beam and then using the parallel-axis
theorem (see Section 12.5):
I Iaa ⫺ Ac 22
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.10 Shear Stresses in the Webs of Beams with Flanges
353
The calculations are as follows:
bh3
(b ⫺ t)h 3
Iaa ᎏᎏ ⫺ ᎏᎏ1 339.67 in.4
3
3
Ac22 270.02 in.4
I 69.65 in.4
Shear stress at top of web. To find the shear stress t1 at the top of the web
(along line nn) we need to calculate the first moment Q1 of the area above level
nn. This first moment is equal to the area of the flange times the distance from
the neutral axis to the centroid of the flange:
冢
冣
h ⫺ h1
Q1 b(h ⫺ h1) c1 ⫺ᎏᎏ
2
(4 in.)(1 in.)(3.045 in. ⫺ 0.5 in.) 10.18 in.3
Of course, we get the same result if we calculate the first moment of the area
below level nn:
冢
冣
h1
Q1 th1 c2 ⫺ ᎏᎏ (1 in.)(7 in.)(4.955 in. ⫺ 3.5 in.) 10.18 in.3
2
Substituting into the shear formula, we find
(10,000 lb)(10.18 in.3)
VQ1
1460 psi
t1 ᎏᎏ ᎏᎏᎏ
(69.65 in.4)(1 in.)
It
This stress exists both as a vertical shear stress acting on the cross section and as
a horizontal shear stress acting on the horizontal plane between the flange and
the web.
Maximum shear stress. The maximum shear stress occurs in the web at the
neutral axis. Therefore, we calculate the first moment Qmax of the cross-sectional
area below the neutral axis:
冢 冣
冢
冣
c2
4.955 in.
Qmax tc2 ᎏᎏ (1 in.)(4.955 in.) ᎏᎏ 12.28 in.3
2
2
As previously indicated, we would get the same result if we calculated the first
moment of the area above the neutral axis, but those calculations would be
slighter longer.
Substituting into the shear formula, we obtain
(10,000 lb)(12.28 in.3)
VQmax
tmax ᎏᎏ ᎏᎏᎏ
1760 psi
(69.65 in.4)(1 in.)
It
which is the maximum shear stress in the beam.
The parabolic distribution of shear stresses in the web is shown in
Fig. 5-40b.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
354
CHAPTER 5 Stresses in Beams (Basic Topics)
5.11 BUILT-UP BEAMS AND SHEAR FLOW
★
(a)
(b)
(c)
FIG. 5-41 Cross sections of typical
built-up beams: (a) wood box beam,
(b) glulam beam, and (c) plate girder
Built-up beams are fabricated from two or more pieces of material
joined together to form a single beam. Such beams can be constructed in
a great variety of shapes to meet special architectural or structural needs
and to provide larger cross sections than are ordinarily available.
Figure 5-41 shows some typical cross sections of built-up beams.
Part (a) of the figure shows a wood box beam constructed of two planks,
which serve as flanges, and two plywood webs. The pieces are joined
together with nails, screws, or glue in such a manner that the entire beam
acts as a single unit. Box beams are also constructed of other materials,
including steel, plastics, and composites.
The second example is a glued laminated beam (called a glulam
beam) made of boards glued together to form a much larger beam than
could be cut from a tree as a single member. Glulam beams are widely
used in the construction of small buildings.
The third example is a steel plate girder of the type commonly used
in bridges and large buildings. These girders, consisting of three steel
plates joined by welding, can be fabricated in much larger sizes than are
available with ordinary wide-flange or I-beams.
Built-up beams must be designed so that the beam behaves as a single
member. Consequently, the design calculations involve two phases. In
the first phase, the beam is designed as though it were made of one piece,
taking into account both bending and shear stresses. In the second phase,
the connections between the parts (such as nails, bolts, welds, and glue)
are designed to ensure that the beam does indeed behave as a single
entity. In particular, the connections must be strong enough to transmit
the horizontal shear forces acting between the parts of the beam. To
obtain these forces, we make use of the concept of shear flow.
Shear Flow
To obtain a formula for the horizontal shear forces acting between parts
of a beam, let us return to the derivation of the shear formula (see Figs.
5-28 and 5-29 of Section 5.8). In that derivation, we cut an element
mm1n1n from a beam (Fig. 5-42a) and investigated the horizontal equilibrium of a subelement mm1p1p (Fig. 5-42b). From the horizontal
equilibrium of the subelement, we determined the force F3 (Fig. 5-42c)
acting on its lower surface:
dM
F3 ᎏᎏ y dA
I
(5-51)
This equation is repeated from Eq. (5-33) of Section 5.8.
Let us now define a new quantity called the shear flow f. Shear flow
is the horizontal shear force per unit distance along the longitudinal axis
of the beam. Since the force F3 acts along the distance dx, the shear force
per unit distance is equal to F3 divided by dx; thus,
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
355
SECTION 5.11 Built-Up Beams and Shear Flow
冢 冣y dA
F3
dM 1
f ᎏᎏ ᎏᎏ ᎏᎏ
dx I
dx
Replacing dM/dx by the shear force V and denoting the integral by Q, we
obtain the following shear-flow formula:
VQ
f ᎏᎏ
I
(5-52)
This equation gives the shear flow acting on the horizontal plane pp1
shown in Fig. 5-42a. The terms V, Q, and I have the same meanings as in
the shear formula (Eq. 5-38).
If the shear stresses on plane pp1 are uniformly distributed, as we
assumed for rectangular beams and wide-flange beams, the shear flow f
equals tb. In that case, the shear-flow formula reduces to the shear
formula. However, the derivation of Eq. (5-51) for the force F3 does not
involve any assumption about the distribution of shear stresses in the
beam. Instead, the force F3 is found solely from the horizontal equilibrium of the subelement (Fig. 5-42c). Therefore, we can now interpret the
subelement and the force F3 in more general terms than before.
The subelement may be any prismatic block of material between
cross sections mn and m1n1 (Fig. 5-42a). It does not have to be obtained
m
m
m1
s1
s2
M
p1
p
M + dM
y1
m1
s2
s1
h
—
2
p
x
p1
t
h
—
2
n
dx
dx
n1
Side view of subelement
(b)
Side view of element
(a)
m
m1
p
p1
F1
F2
F3
FIG. 5-42 Horizontal shear stresses and
shear forces in a beam. (Note: These
figures are repeated from Figs. 5-28 and
5-29.)
dx
Side view of subelement
(c)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
y1
h
—
2
x
y1
h
—
2
x
356
CHAPTER 5 Stresses in Beams (Basic Topics)
by making a single horizontal cut (such as pp1) through the beam. Also,
since the force F3 is the total horizontal shear force acting between the
subelement and the rest of the beam, it may be distributed anywhere over
the sides of the subelement, not just on its lower surface. These same
comments apply to the shear flow f, since it is merely the force F3 per unit
distance.
Let us now return to the shear-flow formula f VQ/I (Eq. 5-52). The
terms V and I have their usual meanings and are not affected by the
choice of subelement. However, the first moment Q is a property of the
cross-sectional face of the subelement. To illustrate how Q is determined,
we will consider three specific examples of built-up beams (Fig. 5-43).
y
a
a
z
O
Areas Used When Calculating the First Moment Q
(a)
y
b
b
z
O
(b)
c
y
c
z
d
d
O
(c)
FIG. 5-43 Areas used when calculating
the first moment Q
The first example of a built-up beam is a welded steel plate girder
(Fig. 5-43a). The welds must transmit the horizontal shear forces that act
between the flanges and the web. At the upper flange, the horizontal
shear force (per unit distance along the axis of the beam) is the shear flow
along the contact surface aa. This shear flow may be calculated by taking
Q as the first moment of the cross-sectional area above the contact surface
aa. In other words, Q is the first moment of the flange area (shown
shaded in Fig. 5-43a), calculated with respect to the neutral axis. After
calculating the shear flow, we can readily determine the amount of welding
needed to resist the shear force, because the strength of a weld is usually
specified in terms of force per unit distance along the weld.
The second example is a wide-flange beam that is strengthened by
riveting a channel section to each flange (Fig. 5-43b). The horizontal
shear force acting between each channel and the main beam must be
transmitted by the rivets. This force is calculated from the shear-flow
formula using Q as the first moment of the area of the entire channel
(shown shaded in the figure). The resulting shear flow is the longitudinal
force per unit distance acting along the contact surface bb, and the rivets
must be of adequate size and longitudinal spacing to resist this force.
The last example is a wood box beam with two flanges and two
webs that are connected by nails or screws (Fig. 5-43c). The total horizontal shear force between the upper flange and the webs is the shear
flow acting along both contact surfaces cc and dd, and therefore the first
moment Q is calculated for the upper flange (the shaded area). In other
words, the shear flow calculated from the formula f VQ/I is the total
shear flow along all contact surfaces that surround the area for which Q
is computed. In this case, the shear flow f is resisted by the combined
action of the nails on both sides of the beam, that is, at both cc and dd, as
illustrated in the following example.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.11 Built-Up Beams and Shear Flow
357
Example 5-16
y
180 mm
15 mm
15 mm
20 mm
120 mm
40 mm
z
280 mm
O
(a)Cross
Crosssection
section
(a)
s
Solution
Shear flow. The horizontal shear force transmitted between the upper flange
and the two webs can be found from the shear-flow formula f VQ/I, in which
Q is the first moment of the cross-sectional area of the flange. To find this first
moment, we multiply the area Af of the flange by the distance df from its centroid
to the neutral axis:
40 mm
s
A wood box beam (Fig. 5-44) is constructed of two boards, each 40 180 mm
in cross section, that serve as flanges and two plywood webs, each 15 mm thick.
The total height of the beam is 280 mm. The plywood is fastened to the flanges
by wood screws having an allowable load in shear of F 800 N each.
If the shear force V acting on the cross section is 10.5 kN, determine the
maximum permissible longitudinal spacing s of the screws (Fig. 5-44b).
A f 40 mm 180 mm 7200 mm2
s
d f 120 mm
2
Q Af df (7200 mm )(120 mm) 864 103 mm3
x
The moment of inertia of the entire cross-sectional area about the neutral axis is
equal to the moment of inertia of the outer rectangle minus the moment of inertia
of the “hole” (the inner rectangle):
1
1
I (210 mm)(280 mm)3 (180 mm)(200 mm)3 264.2 106 mm4
12
12
Substituting V, Q, and I into the shear-flow formula (Eq. 5-52), we obtain
(b)Side
Sideview
view
(b)
FIG. 5-44 Example 5-16. Wood box
beam
(10,500 N)(864 103 mm3)
VQ
34.3 N/mm
f
264.2 106 mm4
I
which is the horizontal shear force per millimeter of length that must be transmitted between the flange and the two webs.
Spacing of screws. Since the longitudinal spacing of the screws is s, and
since there are two lines of screws (one on each side of the flange), it follows that
the load capacity of the screws is 2F per distance s along the beam. Therefore,
the capacity of the screws per unit distance along the beam is 2F/s. Equating 2F/s
to the shear flow f and solving for the spacing s, we get
2F
2(800 N)
s 46.6 mm
f
34.3 N/mm
This value of s is the maximum permissible spacing of the screws, based upon
the allowable load per screw. Any spacing greater than 46.6 mm would overload
the screws. For convenience in fabrication, and to be on the safe side, a spacing
such as s 45 mm would be selected.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
358
CHAPTER 5 Stresses in Beams (Basic Topics)
5.12 BEAMS WITH AXIAL LOADS
★
y
Q
P
x
S
L
(a)
y
V
M
Structural members are often subjected to the simultaneous action of
bending loads and axial loads. This happens, for instance, in aircraft
frames, columns in buildings, machinery, parts of ships, and spacecraft.
If the members are not too slender, the combined stresses can be obtained
by superposition of the bending stresses and the axial stresses.
To see how this is accomplished, consider the cantilever beam
shown in Fig. 5-45a. The only load on the beam is an inclined force P
acting through the centroid of the end cross section. This load can be
resolved into two components, a lateral load Q and an axial load S.
These loads produce stress resultants in the form of bending moments
M, shear forces V, and axial forces N throughout the beam (Fig. 5-45b).
On a typical cross section, distance x from the support, these stress
resultants are
M Q(L ⫺ x)
V Q
NS
x
N
x
(b)
+
+
–
–
+
+
+
+
+
(c)
(d)
(e)
(f)
(g)
FIG. 5-45 Normal stresses in a cantilever
beam subjected to both bending and
axial loads: (a) beam with load P acting
at the free end, (b) stress resultants N, V,
and M acting on a cross section at
distance x from the support, (c) tensile
stresses due to the axial force N acting
alone, (d) tensile and compressive
stresses due to the bending moment M
acting alone, and (e), (f), (g) possible
final stress distributions due to the
combined effects of N and M
in which L is the length of the beam. The stresses associated with each of
these stress resultants can be determined at any point in the cross section
by means of the appropriate formula (s My/I, t VQ/Ib, and
s N/A).
Since both the axial force N and bending moment M produce normal
stresses, we need to combine those stresses to obtain the final stress
distribution. The axial force (when acting alone) produces a uniform
stress distribution s N/A over the entire cross section, as shown by the
stress diagram in Fig. 5-45c. In this particular example, the stress s is
tensile, as indicated by the plus signs attached to the diagram.
The bending moment produces a linearly varying stress s My/I
(Fig. 5-45d) with compression on the upper part of the beam and tension
on the lower part. The distance y is measured from the z axis, which
passes through the centroid of the cross section.
The final distribution of normal stresses is obtained by superposing
the stresses produced by the axial force and the bending moment. Thus,
the equation for the combined stresses is
My
N
s
I
A
(5-53)
Note that N is positive when it produces tension and M is positive according to the bending-moment sign convention (positive bending moment
produces compression in the upper part of the beam and tension in the
lower part). Also, the y axis is positive upward. As long as we use these
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.12 Beams with Axial Loads
359
sign conventions in Eq. (5-53), the normal stress s will be positive for
tension and negative for compression.
The final stress distribution depends upon the relative algebraic
values of the terms in Eq. (5-53). For our particular example, the three
possibilities are shown in Figs. 5-45e, f, and g. If the bending stress at the
top of the beam (Fig. 5-45d) is numerically less than the axial stress
(Fig. 5-45c), the entire cross section will be in tension, as shown in
Fig. 5-45e. If the bending stress at the top equals the axial stress, the
distribution will be triangular (Fig. 5-45f ), and if the bending stress is
numerically larger than the axial stress, the cross section will be partially
in compression and partially in tension (Fig. 5-45g). Of course, if the
axial force is a compressive force, or if the bending moment is reversed
in direction, the stress distributions will change accordingly.
Whenever bending and axial loads act simultaneously, the neutral
axis (that is, the line in the cross section where the normal stress is zero)
no longer passes through the centroid of the cross section. As shown in
Figs. 5-45e, f, and g, respectively, the neutral axis may be outside the
cross section, at the edge of the section, or within the section.
The use of Eq. (5-53) to determine the stresses in a beam with axial
loads is illustrated later in Example 5-17.
y
A
e P
B
x
Eccentric Axial Loads
(a)
y
A
B
P
x
Pe
(b)
y
+
z
n
e
⫺y0
×P
s
C
n
n
⫺
(c)
(d)
FIG. 5-46 (a) Cantilever beam with an
eccentric axial load P, (b) equivalent
loads P and Pe, (c) cross section of
beam, and (d) distribution of normal
stresses over the cross section
An eccentric axial load is an axial force that does not act through the
centroid of the cross section. An example is shown in Fig. 5-46a, where
the cantilever beam AB is subjected to a tensile load P acting at distance
e from the x axis (the x axis passes through the centroids of the cross sections). The distance e, called the eccentricity of the load, is positive in the
positive direction of the y axis.
The eccentric load P is statically equivalent to an axial force P acting
along the x axis and a bending moment Pe acting about the z axis (Fig.
5-46b). Note that the moment Pe is a negative bending moment.
A cross-sectional view of the beam (Fig. 5-46c) shows the y and z
axes passing through the centroid C of the cross section. The eccentric
load P intersects the y axis, which is an axis of symmetry.
Since the axial force N at any cross section is equal to P, and since
the bending moment M is equal to ⫺Pe, the normal stress at any point
in the cross section (from Eq. 5-53) is
Pey
P
s ᎏᎏ ⫹ ᎏᎏ
I
A
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(5-54)
360
CHAPTER 5 Stresses in Beams (Basic Topics)
y
A
B
e P
x
(a)
y
A
B
P
x
Pe
(b)
y
+
z
n
e
y0
×P
s
C
n
n
FIG. 5-46 (Repeated)
(c)
(d)
in which A is the area of the cross section and I is the moment of inertia
about the z axis. The stress distribution obtained from Eq. (5-54), for the
case where both P and e are positive, is shown in Fig. 5-46d.
The position of the neutral axis nn (Fig. 5-46c) can be obtained from
Eq. (5-54) by setting the stress s equal to zero and solving for the coordinate y, which we now denote as y0. The result is
I
y0
Ae
(5-55)
The coordinate y0 is measured from the z axis (which is the neutral axis
under pure bending) to the line nn of zero stress (the neutral axis under
combined bending and axial load). Because y0 is positive in the direction
of the y axis (upward in Fig. 5-46c), it is labeled y0 when it is shown
downward in the figure.
From Eq. (5-55) we see that the neutral axis lies below the z axis
when e is positive and above the z axis when e is negative. If the eccentricity is reduced, the distance y0 increases and the neutral axis moves
away from the centroid. In the limit, as e approaches zero, the load acts
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.12 Beams with Axial Loads
361
at the centroid, the neutral axis is at an infinite distance, and the stress
distribution is uniform. If the eccentricity is increased, the distance y0
decreases and the neutral axis moves toward the centroid. In the limit, as
e becomes extremely large, the load acts at an infinite distance, the
neutral axis passes through the centroid, and the stress distribution is the
same as in pure bending.
Eccentric axial loads are analyzed in some of the problems at the end
of this chapter, beginning with Problem 5.12-12.
Limitations
The preceding analysis of beams with axial loads is based upon the
assumption that the bending moments can be calculated without considering the deflections of the beams. In other words, when determining the
bending moment M for use in Eq. (5-53), we must be able to use the original dimensions of the beam—that is, the dimensions before any
deformations or deflections occur. The use of the original dimensions is
valid provided the beams are relatively stiff in bending, so that the deflections are very small.
Thus, when analyzing a beam with axial loads, it is important to
distinguish between a stocky beam, which is relatively short and
therefore highly resistant to bending, and a slender beam, which is
relatively long and therefore very flexible. In the case of a stocky beam,
the lateral deflections are so small as to have no significant effect on the
line of action of the axial forces. As a consequence, the bending moments
will not depend upon the deflections and the stresses can be found from
Eq. (5-53).
In the case of a slender beam, the lateral deflections (even though
small in magnitude) are large enough to alter significantly the line of
action of the axial forces. When that happens, an additional bending
moment, equal to the product of the axial force and the lateral deflection,
is created at every cross section. In other words, there is an interaction,
or coupling, between the axial effects and the bending effects. This type
of behavior is discussed in Chapter 11 on columns.
The distinction between a stocky beam and a slender beam is
obviously not a precise one. In general, the only way to know whether
interaction effects are important is to analyze the beam with and without
the interaction and notice whether the results differ significantly.
However, this procedure may require considerable calculating effort.
Therefore, as a guideline for practical use, we usually consider a beam
with a length-to-height ratio of 10 or less to be a stocky beam. Only
stocky beams are considered in the problems pertaining to this section.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
362
CHAPTER 5 Stresses in Beams (Basic Topics)
Example 5-17
A tubular beam ACB of length L 60 in. is pin-supported at its ends and loaded
by an inclined force P at midlength (Fig. 5-47a). The distance from the point of
application of the load P to the longitudinal axis of the tube is d 5.5 in. The
cross section of the tube is square (Fig. 5-47b) with outer dimension b 6.0 in.,
area A 20.0 in.2, and moment of inertia I 86.67 in.4
Determine the maximum tensile and compressive stresses in the beam due
to a load P 1000 lb.
y
L
— = 30 in.
2
L
— = 30 in.
2
C
A
y
B
z
x
d = 5.5 in.
P = 1000 lb
b = 6 in.
b = 6 in.
60°
FIG. 5-47 Example 5-17. Tubular beam
(b)
subjected to combined bending and
axial load
(a)
Solution
Beam and loading. We begin by representing the beam and its load in
idealized form for purposes of analysis (Fig. 5-48a). Since the support at end A
resists both horizontal and vertical displacement, it is represented as a pin
support. The support at B prevents vertical displacement but offers no resistance
to horizontal displacement, so it is shown as a roller support.
The inclined load P is resolved into horizontal and vertical components PH
and PV, respectively:
PH P sin 60° (1000 lb)(sin 60°) 866 lb
PV P cos 60° (1000 lb)(cos 60°) 500 lb
The horizontal component PH is shifted to the axis of the beam by the addition
of a moment M0 (Fig. 5-48a):
M0 PH d (866.0 lb)(5.5 in.) 4760 lb-in.
Note that the loads PH, PV, and M0 acting at the midpoint C of the beam are
statically equivalent to the original load P.
Reactions and stress resultants. The reactions of the beam (RH, RA, and RB)
are shown in Fig. 5-48a. Also, the diagrams of axial force N, shear force V, and
bending moment M are shown in Figs. 5-48b, c, and d, respectively. All of these
quantities are found from free-body diagrams and equations of equilibrium,
using the techniques described in Chapter 4.
Stresses in the beam. The maximum tensile stress in the beam occurs at the
bottom of the beam (y 3.0 in.) just to the left of the midpoint C. We arrive
at this conclusion by noting that at this point in the beam the tensile stress due to
the axial force adds to the tensile stress produced by the largest bending moment.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 5.12 Beams with Axial Loads
y
M0 = 4760 lb-in.
A
x
RH
PH = 866 lb
B
C
PV = 500 lb
RA
L
— = 30 in.
2
L
— = 30 in.
2
(5110 lb-in.)(3.0 in.)
My
N
177 psi
(sc)right 0
86.67 in.4
I
A
0
Thus, the maximum compressive stress is
(b)
(sc)max 299 psi
329 lb
0
–171 lb
(c)
9870 lb-in.
M
The maximum compressive stress occurs either at the top of the beam ( y
3.0 in.) to the left of point C or at the top of the beam to the right of point C.
These two stresses are calculated as follows:
43 psi ⫺ 342 psi 299 psi
866 lb
V
(9870 lb-in.)(3.0 in.)
My
866 lb
N
(st)max ᎏᎏ ⫺ ᎏᎏ ᎏᎏ
2 ⫺
86.67 in.4
A
I
20.0 in.
(9870 lb-in.)(3.0 in.)
My
866 l b
N
(sc)left ᎏᎏ ⫺ ᎏᎏ ᎏᎏ2 ⫺
86.67 in.4
I
20.0 in.
A
RB = 171 lb
(a)
N
Thus, from Eq. (5-53), we get
43 psi ⫹ 342 psi 385 psi
RB
RH = 866 lb
RA = 329 lb
363
and occurs at the top of the beam to the left of point C.
Note: This example shows how the normal stresses in a beam due to combined bending and axial load can be determined. The shear stresses acting on
cross sections of the beam (due to the shear forces V ) can be determined independently of the normal stresses, as described earlier in this chapter. Later, in
Chapter 7, we will see how to determine the stresses on inclined planes when we
know both the normal and shear stresses acting on cross-sectional planes.
5110 lb-in.
0
(d)
FIG. 5-48 Solution of Example 5-17.
(a) Idealized beam and loading,
(b) axial-force diagram, (c) shear-force
diagram, and (d) bending-moment
diagram
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
364
CHAPTER 5 Stresses in Beams (Basic Topics)
5.13 STRESS CONCENTRATIONS IN BENDING
★
A
M
M
B
h
d
(a)
C
A
B
M
M
d
h
(b)
FIG. 5-49 Stress distributions in a beam
in pure bending with a circular hole at
the neutral axis. (The beam has a
rectangular cross section with height h
and thickness b.)
The flexure and shear formulas discussed in earlier sections of this chapter are valid for beams without holes, notches, or other abrupt changes in
dimensions. Whenever such discontinuities exist, high localized stresses
are produced. These stress concentrations can be extremely important
when a member is made of brittle material or is subjected to dynamic
loads. (See Chapter 2, Section 2.10, for a discussion of the conditions
under which stress concentrations are important.)
For illustrative purposes, two cases of stress concentrations in beams
are described in this section. The first case is a beam of rectangular cross
section with a hole at the neutral axis (Fig. 5-49). The beam has height
h and thickness b (perpendicular to the plane of the figure) and is in pure
bending under the action of bending moments M.
When the diameter d of the hole is small compared to the height h,
the stress distribution on the cross section through the hole is approximately as shown by the diagram in Fig. 5-49a. At point B on the edge of
the hole the stress is much larger than the stress that would exist at that
point if the hole were not present. (The dashed line in the figure shows
the stress distribution with no hole.) However, as we go toward the outer
edges of the beam (toward point A), the stress distribution varies linearly
with distance from the neutral axis and is only slightly affected by the
presence of the hole.
When the hole is relatively large, the stress pattern is approximately
as shown in Fig. 5-49b. There is a large increase in stress at point B and
only a small change in stress at point A as compared to the stress distribution in the beam without a hole (again shown by the dashed line). The
stress at point C is larger than the stress at A but smaller than the stress
at B.
Extensive investigations have shown that the stress at the edge of
the hole (point B) is approximately twice the nominal stress at that point.
The nominal stress is calculated from the flexure formula in the standard
way, that is, s My/I, in which y is the distance d/2 from the neutral axis
to point B and I is the moment of inertia of the net cross section at the
hole. Thus, we have the following approximate formula for the stress at
point B:
My
2Md
ᎏ
sB 2 ᎏᎏ 1ᎏ
I
b(h3 ⫺ d 3)
(5-56)
At the outer edge of the beam (at point C ), the stress is approximately equal to the nominal stress (not the actual stress) at point A (where
y h/2):
My
6 Mh
ᎏ
sC ᎏᎏ ᎏ
I
b(h3 ⫺ d 3)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(5-57)
365
SECTION 5.13 Stress Concentrations in Bending
From the last two equations we see that the ratio sB/sC is approximately
2d/h. Hence we conclude that when the ratio d/h of hole diameter to
height of beam exceeds 1/2, the largest stress occurs at point B. When d/h
is less than 1/2, the largest stress is at point C.
The second case we will discuss is a rectangular beam with
notches (Fig. 5-50). The beam shown in the figure is subjected to pure
bending and has height h and thickness b (perpendicular to the plane of
the figure). Also, the net height of the beam (that is, the distance between
the bases of the notches) is h1 and the radius at the base of each notch is
R. The maximum stress in this beam occurs at the base of the notches and
may be much larger than the nominal stress at that same point. The
nominal stress is calculated from the flexure formula with y h1/2 and
I bh 31/12; thus,
My
6M
snom ᎏᎏ ᎏᎏ
I
bh12
(5-58)
The maximum stress is equal to the stress-concentration factor K times
the nominal stress:
(5-59)
smax Ksnom
The stress-concentration factor K is plotted in Fig. 5-50 for a few values
of the ratio h/h1. Note that when the notch becomes “sharper,” that is, the
ratio R/h1 becomes smaller, the stress-concentration factor increases.
(Figure 5-50 is plotted from the formulas given in Ref. 2-9.)
The effects of the stress concentrations are confined to small regions
around the holes and notches, as explained in the discussion of SaintVenant’s principle in Section 2.10. At a distance equal to h or greater
from the hole or notch, the stress-concentration effect is negligible and
the ordinary formulas for stresses may be used.
3.0
2R
h
— = 1.2
h1
K
M
2.5
1.1
h = h1 + 2R
2.0
FIG. 5-50 Stress-concentration factor K
for a notched beam of rectangular cross
section in pure bending (h height of
beam; b thickness of beam, perpendicular to the plane of the figure). The
dashed line is for semicircular notches
(h h1 ⫹ 2R)
M
h
h1
s
K = smax snom = 6M2
nom
bh 1
b = thickness
1.05
1.5
0
0.05
0.10
0.15
R
—
h1
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
0.20
0.25
0.30
366
CHAPTER 5 Stresses in Beams (Basic Topics)
PROBLEMS CHAPTER 5
Longitudinal Strains in Beams
5.4-1 Determine the maximum normal strain emax produced
in a steel wire of diameter d 1/16 in. when it is bent
around a cylindrical drum of radius R 24 in. (see figure).
90°
d
R
PROB. 5.4-3
PROB. 5.4-1
5.4-4 A cantilever beam AB is loaded by a couple M0 at its
5.4-2 A copper wire having diameter d 3 mm is bent into
a circle and held with the ends just touching (see
figure). If the maximum permissible strain in the copper is
emax 0.0024, what is the shortest length L of wire that can
be used?
free end (see figure). The length of the beam is L 1.5 m
and the longitudinal normal strain at the top surface is 0.001.
The distance from the top surface of the beam to the neutral
surface is 75 mm.
Calculate the radius of curvature r, the curvature k, and
the vertical deflection d at the end of the beam.
d = diameter
A
L = length
d
L
B
M0
PROB. 5.4-2
PROB. 5.4-4
5.4-3 A 4.5 in. outside diameter polyethylene pipe designed
to carry chemical wastes is placed in a trench and bent
around a quarter-circular 90° bend (see figure). The bent
section of the pipe is 46 ft long.
Determine the maximum compressive strain emax in the
pipe.
5.4-5 A thin strip of steel of length L 20 in. and thickness
t 0.2 in. is bent by couples M0 (see figure on the next
page). The deflection d at the midpoint of the strip (measured
from a line joining its end points) is found to be 0.25 in.
Determine the longitudinal normal strain e at the top
surface of the strip.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 5 Problems
M0
5.5-2 A steel wire (E 200 GPa) of diameter
d 1.0 mm is bent around a pulley of radius R0
400 mm (see figure).
(a) What is the maximum stress smax in the wire?
(b) Does the stress increase or decrease if the radius of
the pulley is increased?
M0
t
d
L
—
2
L
—
2
367
PROB. 5.4-5
5.4-6 A bar of rectangular cross section is loaded and
supported as shown in the figure. The distance between supports is L 1.2 m and the height of the bar is h 100 mm.
The deflection d at the midpoint is measured as 3.6 mm.
What is the maximum normal strain e at the top and
bottom of the bar?
h
P
R0
d
d
P
PROB. 5.5-2
a
L
—
2
L
—
2
a
5.5-3 A thin, high-strength steel rule (E 30 106 psi)
PROB. 5.4-6
Normal Stresses in Beams
5.5-1 A thin strip of hard copper (E 16,400 ksi) having
length L 80 in. and thickness t 3/32 in. is bent into a
circle and held with the ends just touching (see figure).
(a) Calculate the maximum bending stress smax in the
strip.
(b) Does the stress increase or decrease if the thickness
of the strip is increased?
having thickness t 0.15 in. and length L 40 in. is bent
by couples M0 into a circular arc subtending a central angle
a 45° (see figure).
(a) What is the maximum bending stress smax in the
rule?
(b) Does the stress increase or decrease if the central
angle is increased?
L = length
3
t = — in.
32
M0
t
M0
a
PROB. 5.5-1
PROB. 5.5-3
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
368
CHAPTER 5 Stresses in Beams (Basic Topics)
5.5-4 A simply supported wood beam AB with span length
L 3.5 m carries a uniform load of intensity q 6.4 kN/m
(see figure).
Calculate the maximum bending stress smax due to the
load q if the beam has a rectangular cross section with width
b 140 mm and height h 240 mm.
P
P
A
B
d
d
R
b
R
L
b
PROB. 5.5-6
q
5.5-7 A seesaw weighing 3 lb/ft of length is occupied by
A
h
B
L
b
two children, each weighing 90 lb (see figure). The center of
gravity of each child is 8 ft from the fulcrum. The board is
19 ft long, 8 in. wide, and 1.5 in. thick.
What is the maximum bending stress in the board?
PROB. 5.5-4
5.5-5 Each girder of the lift bridge (see figure) is 180 ft long
and simply supported at the ends. The design load for each
girder is a uniform load of intensity 1.6 k/ft. The girders are
fabricated by welding three steel plates so as to form an
I-shaped cross section (see figure) having section modulus
S 3600 in3.
What is the maximum bending stress smax in a girder
due to the uniform load?
PROB. 5.5-7
5.5-8 During construction of a highway bridge, the main
girders are cantilevered outward from one pier toward the
next (see figure). Each girder has a cantilever length of 46 m
and an I-shaped cross section with dimensions as shown in
the figure. The load on each girder (during construction) is
assumed to be 11.0 kN/m, which includes the weight of the
girder.
Determine the maximum bending stress in a girder due
to this load.
50 mm
PROB. 5.5-5
5.5-6 A freight-car axle AB is loaded approximately as
shown in the figure, with the forces P representing the car
loads (transmitted to the axle through the axle boxes) and
the forces R representing the rail loads (transmitted to the
axle through the wheels). The diameter of the axle is d
80 mm, the distance between centers of the rails is L, and the
distance between the forces P and R is b 200 mm.
Calculate the maximum bending stress smax in the axle
if P 47 kN.
2400 mm
25 mm
600 mm
PROB. 5.5-8
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 5 Problems
369
5.5-9 The horizontal beam ABC of an oil-well pump has the
cross section shown in the figure. If the vertical pumping
force acting at end C is 8.8 k, and if the distance from the
line of action of that force to point B is 14 ft, what is the
maximum bending stress in the beam due to the pumping
force?
A
B
C
s
0.875 in.
L
20.0
in.
0.625
in.
8.0 in.
PROB. 5.5-9
PROB. 5.5-11
5.5-12 A small dam of height h 2.0 m is constructed of
vertical wood beams AB of thickness t 120 mm, as shown
in the figure. Consider the beams to be simply supported at
the top and bottom.
Determine the maximum bending stress smax in the beams,
assuming that the weight density of water is g 9.81 kN/m3.
A
5.5-10 A railroad tie (or sleeper) is subjected to two rail
loads, each of magnitude P 175 kN, acting as shown in
the figure. The reaction q of the ballast is assumed to be
uniformly distributed over the length of the tie, which has
cross-sectional dimensions b 300 mm and h 250 mm.
Calculate the maximum bending stress smax in the tie
due to the loads P, assuming the distance L 1500 mm and
the overhang length a 500 mm.
a
P
P
L
h
t
B
PROB. 5.5-12
a
b
h
q
5.5-13 Determine the maximum tensile stress st (due to
pure bending by positive bending moments M) for beams
having cross sections as follows (see figure): (a) a semicircle of diameter d, and (b) an isosceles trapezoid with
bases b1 b and b2 4b/3, and altitude h.
PROB. 5.5-10
b1
5.5-11 A fiberglass pipe is lifted by a sling, as shown in the
figure. The outer diameter of the pipe is 6.0 in., its thickness
is 0.25 in., and its weight density is 0.053 lb/in.3 The length
of the pipe is L 36 ft and the distance between lifting
points is s 11 ft.
Determine the maximum bending stress in the pipe due
to its own weight.
C
C
d
b2
(a)
(b)
PROB. 5.5-13
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
h
370
CHAPTER 5 Stresses in Beams (Basic Topics)
5.5-14 Determine the maximum bending stress smax (due to
pure bending by a moment M) for a beam having a cross
section in the form of a circular core (see figure). The circle
has diameter d and the angle b 60°. (Hint: Use the
formulas given in Appendix D, Cases 9 and 15.)
C
b
5.5-17 A cantilever beam AB, loaded by a uniform load and
a concentrated load (see figure), is constructed of a channel
section.
Find the maximum tensile stress st and maximum compressive stress sc if the cross section has the dimensions
indicated and the moment of inertia about the z axis (the
neutral axis) is I 2.81 in.4 (Note: The uniform load represents the weight of the beam.)
b
200 lb
20 lb/ft
d
PROB. 5.5-14
jected to two wheel loads acting at distance d 5 ft apart
(see figure). Each wheel transmits a load P 3.0 k, and the
carriage may occupy any position on the beam.
Determine the maximum bending stress smax due to the
wheel loads if the beam is an I-beam having section modulus S 16.2 in.3
P
d
P
B
A
5.5-15 A simple beam AB of span length L 24 ft is sub-
5.0 ft
3.0 ft
y
z
C
0.606 in.
2.133 in.
PROB. 5.5-17
A
C
B
L
PROB. 5.5-15
t and
maximum compressive stress sc due to the load P acting on
the simple beam AB (see figure).
Data are as follows: P 5.4 kN, L 3.0 m,
d 1.2 m, b 75 mm, t 25 mm, h 100 mm, and
h1 75 mm.
5.5-16 Determine the maximum tensile stress
5.5-18 A cantilever beam AB of triangular cross section has
length L 0.8 m, width b 80 mm, and height h 120 mm
(see figure). The beam is made of brass weighing 85 kN/m3.
(a) Determine the maximum tensile stress st and maximum compressive stress sc due to the beam’s own weight.
(b) If the width b is doubled, what happens to the
stresses?
(c) If the height h is doubled, what happens to the
stresses?
A
b
B
L
h
t
P
A
B
L
PROB. 5.5-16
PROB. 5.5-18
d
h1
h
b
5.5-19 A beam ABC with an overhang from B to C supports
a uniform load of 160 lb/ft throughout its length (see figure).
The beam is a channel section with dimensions as shown in
the figure. The moment of inertia about the z axis (the neutral
axis) equals 5.14 in.4
Calculate the maximum tensile stress st and maximum
compressive stress sc due to the uniform load.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 5 Problems
★
5.5-22 A cantilever beam AB with a rectangular cross
section has a longitudinal hole drilled throughout its length
(see figure). The beam supports a load P 600 N. The cross
section is 25 mm wide and 50 mm high, and the hole has a
diameter of 10 mm.
Find the bending stresses at the top of the beam, at the
top of the hole, and at the bottom of the beam.
160 lb/ft
A
C
B
10 ft
5 ft
y
371
0.674 in.
A
10 mm
50 mm
B
12.5 mm
z
C
2.496 in.
37.5 mm
P = 600 N
PROB. 5.5-19
L = 0.4 m
5.5-20 A frame ABC travels horizontally with an acceleration a0 (see figure). Obtain a formula for the maximum
stress smax in the vertical arm AB, which has length L, thickness t, and mass density r.
25 mm
PROB. 5.5-22
A
t
5.5-23 A small dam of height h 6 ft is constructed of
vertical wood beams AB, as shown in the figure. The wood
beams, which have thickness t 2.5 in., are simply supported by horizontal steel beams at A and B.
Construct a graph showing the maximum bending
stress smax in the wood beams versus the depth d of the
water above the lower support at B. Plot the stress smax (psi)
as the ordinate and the depth d (ft) as the abscissa. (Note:
The weight density g of water equals 62.4 lb/ft3.)
★★
a0 = acceleration
L
C
B
PROB. 5.5-20
5.5-21 A beam of T-section is supported and loaded
as shown in the figure. The cross section has width b
2 1/2 in., height h 3 in., and thickness t 1/2 in.
Determine the maximum tensile and compressive
stresses in the beam.
★
1
t=—
2 in.
P = 625 lb
L1 = 4 ft
L2 = 8 ft
PROB. 5.5-21
Steel beam
A Wood beam
t
Steel beam
Wood beam
h
q = 80 lb/ft
1
d
t=—
2 in.
L3 = 5 ft
t
h=
3 in.
B
1
b = 2—
2 in.
Side view
PROB. 5.5-23
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
Top view
372
CHAPTER 5 Stresses in Beams (Basic Topics)
Design of Beams
5.6-1 The cross section of a narrow-gage railway bridge is
shown in part (a) of the figure. The bridge is constructed
with longitudinal steel girders that support the wood cross
ties. The girders are restrained against lateral buckling by
diagonal bracing, as indicated by the dashed lines.
The spacing of the girders is s1 50 in. and the spacing of the rails is s2 30 in. The load transmitted by each
rail to a single tie is P 1500 lb. The cross section of a tie,
shown in part (b) of the figure, has width b 5.0 in. and
depth d.
Determine the minimum value of d based upon an
allowable bending stress of 1125 psi in the wood tie.
(Disregard the weight of the tie itself.)
P
5.6-3 A cantilever beam of length L 6 ft supports a uniform load of intensity q 200 lb/ft and a concentrated load
P 2500 lb (see figure).
Calculate the required section modulus S if sallow
15,000 psi. Then select a suitable wide-flange beam (W
shape) from Table E-1, Appendix E, and recalculate S taking
into account the weight of the beam. Select a new beam size
if necessary.
P 2500 lb
q 200 lb/ft
P
s2
Steel rail
Wood
tie
L = 6 ft
d
PROB. 5.6-3
b
Steel
girder
(b)
s1
(a)
PROB. 5.6-1
5.6-4 A simple beam of length L 15 ft carries a uniform
load of intensity q 400 lb/ft and a concentrated load P
4000 lb (see figure).
Assuming sallow 16,000 psi, calculate the required
section modulus S. Then select an 8-inch wide-flange beam
(W shape) from Table E-1, Appendix E, and recalculate S
taking into account the weight of the beam. Select a new
8-inch beam if necessary.
5.6-2 A fiberglass bracket ABCD of solid circular cross section has the shape and dimensions shown in the figure. A
vertical load P 36 N acts at the free end D.
Determine the minimum permissible diameter dmin of
the bracket if the allowable bending stress in the material is
30 MPa and b 35 mm. (Disregard the weight of the
bracket itself.)
P = 4000 lb
7.5 ft
q = 400 lb/ft
5b
A
L = 15 ft
B
PROB. 5.6-4
2b
D
P
PROB. 5.6-2
C
2b
5.6-5 A simple beam AB is loaded as shown in the figure on
the next page. Calculate the required section modulus S if
sallow 15,000 psi, L 24 ft, P 2000 lb, and q
400 lb/ft. Then select a suitable I-beam (S shape) from Table
E-2, Appendix E, and recalculate S taking into account the
weight of the beam. Select a new beam size if necessary.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 5 Problems
P
q
q
Planks
B
A
L
—
4
L
—
4
373
L
—
4
L
—
4
s
PROB. 5.6-5
5.6-6 A pontoon bridge (see figure) is constructed of two
longitudinal wood beams, known as balks, that span
between adjacent pontoons and support the transverse floor
beams, which are called chesses.
For purposes of design, assume that a uniform floor
load of 8.0 kPa acts over the chesses. (This load includes an
allowance for the weights of the chesses and balks.) Also,
assume that the chesses are 2.0 m long and that the balks are
simply supported with a span of 3.0 m. The allowable bending stress in the wood is 16 MPa.
If the balks have a square cross section, what is their
minimum required width bmin?
Chess
Pontoon
s
L
s
Joists
PROBS. 5.6-7 and 5.6-8
5.6-8 The wood joists supporting a plank floor (see figure)
are 40 mm 180 mm in cross section (actual dimensions)
and have a span length L 4.0 m. The floor load is 3.6 kPa,
which includes the weight of the joists and the floor.
Calculate the maximum permissible spacing s of the
joists if the allowable bending stress is 15 MPa. (Assume
that each joist may be represented as a simple beam carrying a uniform load.)
5.6-9 A beam ABC with an overhang from B to C is constructed of a C 10 30 channel section (see figure). The
beam supports its own weight (30 lb/ft) plus a uniform load
of intensity q acting on the overhang. The allowable stresses
in tension and compression are 18 ksi and 12 ksi, respectively.
Determine the allowable uniform load qallow if the
distance L equals 3.0 ft.
Balk
q
PROB. 5.6-6
5.6-7 A floor system in a small building consists of wood
planks supported by 2 in. (nominal width) joists spaced at
distance s, measured from center to center (see figure). The
span length L of each joist is 10.5 ft, the spacing s of the
joists is 16 in., and the allowable bending stress in the wood
is 1350 psi. The uniform floor load is 120 lb/ft2, which
includes an allowance for the weight of the floor system
itself.
Calculate the required section modulus S for the joists,
and then select a suitable joist size (surfaced lumber) from
Appendix F, assuming that each joist may be represented as
a simple beam carrying a uniform load.
A
C
B
L
L
3.033
in.
C
10.0 in.
PROB. 5.6-9
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
2.384 in.
0.649 in.
374
CHAPTER 5 Stresses in Beams (Basic Topics)
5.6-10 A so-called “trapeze bar” in a hospital room provides
a means for patients to exercise while in bed (see figure).
The bar is 2.1 m long and has a cross section in the shape of
a regular octagon. The design load is 1.2 kN applied at the
midpoint of the bar, and the allowable bending stress is
200 MPa.
Determine the minimum height h of the bar. (Assume
that the ends of the bar are simply supported and that the
weight of the bar is negligible.)
C
h
A
B
d
P
L
PROB. 5.6-12
5.6-13 A compound beam ABCD (see figure) is supported
at points A, B, and D and has a splice (represented by the pin
connection) at point C. The distance a 6.0 ft and the beam
is a W 16 57 wide-flange shape with an allowable bending stress of 10,800 psi.
Find the allowable uniform load qallow that may be
placed on top of the beam, taking into account the weight of
the beam itself.
q
PROB. 5.6-10
A
B
5.6-11 A two-axle carriage that is part of an overhead
traveling crane in a testing laboratory moves slowly across
a simple beam AB (see figure). The load transmitted to the
beam from the front axle is 2000 lb and from the rear axle is
4000 lb. The weight of the beam itself may be disregarded.
(a) Determine the minimum required section modulus S
for the beam if the allowable bending stress is 15.0 ksi, the
length of the beam is 16 ft, and the wheelbase of the carriage
is 5 ft.
(b) Select a suitable I-beam (S shape) from Table E-2,
Appendix E.
4000 lb
5 ft
A
2000 lb
B
4a
C
D
Pin
a
4a
PROB. 5.6-13
5.6-14 A small balcony constructed of wood is supported
by three identical cantilever beams (see figure). Each beam has
length L1 2.1 m, width b, and height h 4b/3. The dimensions of the balcony floor are L1 L2, with L2 2.5 m. The
design load is 5.5 kPa acting over the entire floor area. (This
load accounts for all loads except the weights of the cantilever
beams, which have a weight density g 5.5 kN/m3.) The
allowable bending stress in the cantilevers is 15 MPa.
Assuming that the middle cantilever supports 50% of
the load and each outer cantilever supports 25% of the load,
determine the required dimensions b and h.
16 ft
PROB. 5.6-11
5.6-12 A cantilever beam AB of circular cross section and
length L 450 mm supports a load P 400 N acting at the
free end (see figure). The beam is made of steel with an
allowable bending stress of 60 MPa.
Determine the required diameter dmin of the beam,
considering the effect of the beam’s own weight.
4b
h= —
3
L2
L1
PROB. 5.6-14
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
b
CHAPTER 5 Problems
5.6-15 A beam having a cross section in the form of an
unsymmetric wide-flange shape (see figure) is subjected to
a negative bending moment acting about the z axis.
Determine the width b of the top flange in order that the
stresses at the top and bottom of the beam will be in the ratio
4:3, respectively.
y
b
1.5 in.
1.25 in.
z
5.6-18 A horizontal shelf AD of length L 900 mm,
width b 300 mm, and thickness t 20 mm is supported
by brackets at B and C [see part (a) of the figure]. The
brackets are adjustable and may be placed in any desired
positions between the ends of the shelf. A uniform load of
intensity q, which includes the weight of the shelf itself, acts
on the shelf [see part (b) of the figure].
Determine the maximum permissible value of the load q
if the allowable bending stress in the shelf is sallow 5.0
MPa and the position of the supports is adjusted for
maximum load-carrying capacity.
★
12 in.
C
t
1.5 in.
A
16 in.
B
(a)
nel (see figure) is subjected to a bending moment acting
about the z axis.
Calculate the thickness t of the channel in order that the
bending stresses at the top and bottom of the beam will be
in the ratio 7:3, respectively.
q
A
D
B
C
y
L
t
(b)
t
C
t
PROB. 5.6-18
50 mm
120 mm
PROB. 5.6-16
5.6-19 A steel plate (called a cover plate) having crosssectional dimensions 4.0 in. 0.5 in. is welded along the
full length of the top flange of a W 12 35 wide-flange
beam (see figure, which shows the beam cross section).
What is the percent increase in section modulus (as
compared to the wide-flange beam alone)?
★
4.0 0.5 in. cover plate
5.6-17 Determine the ratios of the weights of three beams
that have the same length, are made of the same material, are
subjected to the same maximum bending moment, and have
the same maximum bending stress if their cross sections are
(1) a rectangle with height equal to twice the width, (2) a
square, and (3) a circle (see figures).
h = 2b
a
W 12 35
PROB. 5.6-19
5.6-20 A steel beam ABC is simply supported at A and B
and has an overhang BC of length L 150 mm (see figure
on the next page). The beam supports a uniform load of
intensity q 3.5 kN/m over its entire length of 450 mm.
★★
PROB. 5.6-17
b
L
5.6-16 A beam having a cross section in the form of a chan-
b
D
C
PROB. 5.6-15
z
a
375
d
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
376
CHAPTER 5 Stresses in Beams (Basic Topics)
The cross section of the beam is rectangular with width b
and height 2b. The allowable bending stress in the steel is
sallow 60 MPa and its weight density is g 77.0 kN/m3.
(a) Disregarding the weight of the beam, calculate the
required width b of the rectangular cross section.
(b) Taking into account the weight of the beam, calculate the required width b.
q
C
A
2b
B
2L
5.6-22 A beam of square cross section (a length of
each side) is bent in the plane of a diagonal (see figure). By
removing a small amount of material at the top and bottom
corners, as shown by the shaded triangles in the figure, we
can increase the section modulus and obtain a stronger
beam, even though the area of the cross section is reduced.
(a) Determine the ratio b defining the areas that should
be removed in order to obtain the strongest cross section in
bending.
(b) By what percent is the section modulus increased
when the areas are removed?
★★
y
b
L
a
PROB. 5.6-20
5.6-21 A retaining wall 5 ft high is constructed of horizontal wood planks 3 in. thick (actual dimension) that are
supported by vertical wood piles of 12 in. diameter (actual
dimension), as shown in the figure. The lateral earth
pressure is p1 100 lb/ft2 at the top of the wall and p2
400 lb/ft2 at the bottom.
Assuming that the allowable stress in the wood is 1200 psi,
calculate the maximum permissible spacing s of the piles.
(Hint: Observe that the spacing of the piles may be
governed by the load-carrying capacity of either the planks
or the piles. Consider the piles to act as cantilever beams
subjected to a trapezoidal distribution of load, and consider
the planks to act as simple beams between the piles. To be
on the safe side, assume that the pressure on the bottom
plank is uniform and equal to the maximum pressure.)
z
★★
C
a
5.6-23 The cross section of a rectangular beam having
width b and height h is shown in part (a) of the figure. For
reasons unknown to the beam designer, it is planned to add
structural projections of width b/9 and height d to the top
and bottom of the beam [see part (b) of the figure].
For what values of d is the bending-moment capacity of
the beam increased? For what values is it decreased?
★★
b
—
9
d
12 in.
diam.
s
5 ft
ba
PROB. 5.6-22
3 in.
p1 = 100 lb/ft2
12 in.
diam.
ba
h
b
3 in.
(a)
h
d
b
—
9
(b)
PROB. 5.6-23
Top view
Nonprismatic Beams
5.7-1 A tapered cantilever beam AB of length L has square
p2 =
Side view
PROB. 5.6-21
400 lb/ft2
cross sections and supports a concentrated load P at the free
end (see figure on the next page). The width and height of
the beam vary linearly from hA at the free end to hB at the
fixed end.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 5 Problems
Determine the distance x from the free end A to the
cross section of maximum bending stress if hB 3hA. What
is the magnitude smax of the maximum bending stress?
What is the ratio of the maximum stress to the largest stress
sB at the support?
B
hA
A
5.7-3 A tapered cantilever beam AB having rectangular
cross sections is subjected to a concentrated load P 50 lb
and a couple M0 800 lb-in. acting at the free end (see figure). The width b of the beam is constant and equal to 1.0 in.,
but the height varies linearly from hA 2.0 in. at the loaded
end to hB 3.0 in. at the support.
At what distance x from the free end does the maximum
bending stress smax occur? What is the magnitude smax of
the maximum bending stress? What is the ratio of the maximum stress to the largest stress sB at the support?
P = 50 lb
hB
x
P
hA =
2.0 in.
L
A M0 = 800 lb-in.
B
hB =
3.0 in.
x
PROB. 5.7-1
b = 1.0 in.
consisting of thin-walled, tapered circular tubes (see figure).
For purposes of this analysis, each beam may be represented
as a cantilever AB of length L 8.0 m subjected to a lateral
load P 2.4 kN at the free end. The tubes have constant
thickness t 10.0 mm and average diameters dA 90 mm
and dB 270 mm at ends A and B, respectively.
Because the thickness is small compared to the diameters, the moment of inertia at any cross section may be
obtained from the formula I pd 3t/8 (see Case 22,
Appendix D), and therefore the section modulus may be
obtained from the formula S pd 2t/4.
At what distance x from the free end does the maximum
bending stress occur? What is the magnitude smax of the
maximum bending stress? What is the ratio of the maximum
stress to the largest stress sB at the support?
PROB. 5.7-3
5.7-4 The spokes in a large flywheel are modeled as
beams fixed at one end and loaded by a force P and a couple M0 at the other (see figure). The cross sections of the
spokes are elliptical with major and minor axes (height and
width, respectively) having the lengths shown in the figure.
The cross-sectional dimensions vary linearly from end A to
end B.
Considering only the effects of bending due to the loads
P and M0, determine the following quantities: (a) the largest
bending stress sA at end A; (b) the largest bending stress sB
at end B; (c) the distance x to the cross section of maximum
bending stress; and (d) the magnitude smax of the maximum
bending stress.
★
P = 15 kN
M0 = 12 kN.m
P = 2.4 kN
Wind
load
b = 1.0 in.
L = 20 in.
5.7-2 A tall signboard is supported by two vertical beams
B
A
B
A
x
x
L = 1.10 m
L = 8.0 m
t = 10.0 mm
hA = 90 mm
dA = 90 mm
dB = 270 mm
hB = 120 mm
bA = 60 mm
bB = 80 mm
PROB. 5.7-2
377
PROB. 5.7-4
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
378
CHAPTER 5 Stresses in Beams (Basic Topics)
5.7-5 Refer to the tapered cantilever beam of solid circular cross section shown in Fig. 5-24 of Example 5-9.
(a) Considering only the bending stresses due to the
load P, determine the range of values of the ratio dB /dA for
which the maximum normal stress occurs at the support.
(b) What is the maximum stress for this range of
values?
★
P
A
h
B
C
x
L
—
2
L
—
2
Fully Stressed Beams
Problems 5.7-6 to 5.7-8 pertain to fully stressed beams of
rectangular cross section. Consider only the bending
stresses obtained from the flexure formula and disregard
the weights of the beams.
h
h
bB
bx
5.7-6 A cantilever beam AB having rectangular cross sections
with constant width b and varying height hx is subjected to
a uniform load of intensity q (see figure).
How should the height hx vary as a function of x
(measured from the free end of the beam) in order to have a
fully stressed beam? (Express hx in terms of the height hB at
the fixed end of the beam.)
q
B
A
hx
PROB. 5.7-7
5.7-8 A cantilever beam AB having rectangular cross sections with varying width bx and varying height hx is
subjected to a uniform load of intensity q (see figure). If the
width varies linearly with x according to the equation bx
bB x/L, how should the height hx vary as a function of x in
order to have a fully stressed beam? (Express hx in terms of
the height hB at the fixed end of the beam.)
hB
q
x
L
B
hx
hB
b
hB
hx
A
x
b
L
PROB. 5.7-6
5.7-7 A simple beam ABC having rectangular cross sections
with constant height h and varying width bx supports a concentrated load P acting at the midpoint (see figure).
How should the width bx vary as a function of x in order
to have a fully stressed beam? (Express bx in terms of the
width bB at the midpoint of the beam.)
hx
hB
bx
bB
PROB. 5.7-8
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 5 Problems
Shear Stresses in Rectangular Beams
5.8-1 The shear stresses t in a rectangular beam are given
by Eq. (5-39):
冢
V h2
t ᎏᎏ ᎏ ᎏ ⫺ y 21
2I 4
冣
in which V is the shear force, I is the moment of inertia of
the cross-sectional area, h is the height of the beam, and y1
is the distance from the neutral axis to the point where the
shear stress is being determined (Fig. 5-30).
By integrating over the cross-sectional area, show that
the resultant of the shear stresses is equal to the shear
force V.
5.8-4 A cantilever beam of length L 2 m supports a load
P 8.0 kN (see figure). The beam is made of wood with
cross-sectional dimensions 120 mm 200 mm.
Calculate the shear stresses due to the load P at points
located 25 mm, 50 mm, 75 mm, and 100 mm from the top
surface of the beam. From these results, plot a graph showing the distribution of shear stresses from top to bottom of
the beam.
P = 8.0 kN
200 mm
L=2m
120 mm
5.8-2 Calculate the maximum shear stress tmax and the
maximum bending stress smax in a simply supported wood
beam (see figure) carrying a uniform load of 18.0 kN/m
(which includes the weight of the beam) if the length is 1.75
m and the cross section is rectangular with width 150 mm
and height 250 mm.
18.0 kN/m
250 mm
1.75 m
150 mm
PROB. 5.8-4
5.8-5 A steel beam of length L 16 in. and cross-sectional
dimensions b 0.6 in. and h 2 in. (see figure) supports a
uniform load of intensity q 240 lb/in., which includes the
weight of the beam.
Calculate the shear stresses in the beam (at the cross
section of maximum shear force) at points located 1/4 in.,
1/2 in., 3/4 in., and 1 in. from the top surface of the beam.
From these calculations, plot a graph showing the distribution of shear stresses from top to bottom of the beam.
PROB. 5.8-2
q = 240 lb/in.
5.8-3 Two wood beams, each of square cross section
(3.5 in. 3.5 in., actual dimensions) are glued together to
form a solid beam of dimensions 3.5 in. 7.0 in. (see figure). The beam is simply supported with a span of 6 ft.
What is the maximum load Pmax that may act at the
midpoint if the allowable shear stress in the glued joint is
200 psi? (Include the effects of the beam’s own weight,
assuming that the wood weighs 35 lb/ft3.)
;@ÀQ¢;@ÀQ¢
¢¢
QQ
@@
;;
ÀÀ
¢Q;À@
h = 2 in.
L = 16 in.
b = 0.6 in.
PROB. 5.8-5
5.8-6 A beam of rectangular cross section (width b and
3.5 in.
P
7.0 in.
PROB. 5.8-3
379
6 ft
height h) supports a uniformly distributed load along its
entire length L. The allowable stresses in bending and shear
are sallow and tallow, respectively.
(a) If the beam is simply supported, what is the span
length L0 below which the shear stress governs the allowable load and above which the bending stress governs?
(b) If the beam is supported as a cantilever, what is the
length L0 below which the shear stress governs the allowable load and above which the bending stress governs?
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
380
CHAPTER 5 Stresses in Beams (Basic Topics)
5.8-7 A laminated wood beam on simple supports is built up
by gluing together three 2 in. 4 in. boards (actual dimensions) to form a solid beam 4 in. 6 in. in cross section, as
shown in the figure. The allowable shear stress in the glued
joints is 65 psi and the allowable bending stress in the wood
is 1800 psi.
If the beam is 6 ft long, what is the allowable load P
acting at the midpoint of the beam? (Disregard the weight of
the beam.)
3 ft
P
2 in.
5.8-9 A wood beam AB on simple supports with span
length equal to 9 ft is subjected to a uniform load of intensity 120 lb/ft acting along the entire length of the beam and
a concentrated load of magnitude 8800 lb acting at a point
3 ft from the right-hand support (see figure). The allowable
stresses in bending and shear, respectively, are 2500 psi and
150 psi.
(a) From the table in Appendix F, select the lightest
beam that will support the loads (disregard the weight of the
beam).
(b) Taking into account the weight of the beam (weight
density 35 lb/ft3), verify that the selected beam is satisfactory, or, if it is not, select a new beam.
★
2 in.
2 in.
8800 lb
L 6 ft
3 ft
120 lb/ft
4 in.
A
PROB. 5.8-7
B
9 ft
5.8-8 A laminated plastic beam of square cross section is
built up by gluing together three strips, each 10 mm
30 mm in cross section (see figure). The beam has a total
weight of 3.2 N and is simply supported with span length
L 320 mm.
Considering the weight of the beam, calculate the maximum permissible load P that may be placed at the midpoint
if (a) the allowable shear stress in the glued joints is
0.3 MPa, and (b) the allowable bending stress in the plastic
is 8 MPa.
P
PROB. 5.8-9
5.8-10 A simply supported wood beam of rectangular
cross section and span length 1.2 m carries a concentrated
load P at midspan in addition to its own weight (see figure).
The cross section has width 140 mm and height 240 mm.
The weight density of the wood is 5.4 kN/m3.
Calculate the maximum permissible value of the load P
if (a) the allowable bending stress is 8.5 MPa, and (b) the
allowable shear stress is 0.8 MPa.
★
q
P
10 mm
10 mm 30 mm
10 mm
30 mm
PROB. 5.8-8
240 mm
L
0.6 m
0.6 m
PROB. 5.8-10
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
140 mm
CHAPTER 5 Problems
★★
5.8-11 A square wood platform, 8 ft 8 ft in area, rests
on masonry walls (see figure). The deck of the platform is
constructed of 2 in. nominal thickness tongue-and-groove
planks (actual thickness 1.5 in.; see Appendix F) supported
on two 8-ft long beams. The beams have 4 in. 6 in. nominal dimensions (actual dimensions 3.5 in. 5.5 in.).
The planks are designed to support a uniformly distributed load w (lb/ft 2) acting over the entire top surface of the
platform. The allowable bending stress for the planks is
2400 psi and the allowable shear stress is 100 psi. When
analyzing the planks, disregard their weights and assume
that their reactions are uniformly distributed over the top
surfaces of the supporting beams.
(a) Determine the allowable platform load w1 (lb/ft 2)
based upon the bending stress in the planks.
(b) Determine the allowable platform load w2 (lb/ft 2)
based upon the shear stress in the planks.
(c) Which of the preceding values becomes the allowable load wallow on the platform?
(Hints: Use care in constructing the loading diagram
for the planks, noting especially that the reactions are distributed loads instead of concentrated loads. Also, note that
the maximum shear forces occur at the inside faces of the
supporting beams.)
¢¢
@@
ÀÀ
;;
QQ
@@
ÀÀ
;;
QQ
@@;À@¢¢
ÀÀ
;;
QQ
;@À¢¢
@@
ÀÀ
;;
QQ
;À@¢¢
8 ft
8 ft
5.8-12 A wood beam ABC with simple supports at A and
B and an overhang BC has height h 280 mm (see figure).
The length of the main span of the beam is L 3.6 m and
the length of the overhang is L/3 1.2 m. The beam supports a concentrated load 3P 15 kN at the midpoint of the
main span and a load P 5 kN at the free end of the overhang. The wood has weight density g 5.5 kN/m3.
(a) Determine the required width b of the beam based
upon an allowable bending stress of 8.2 MPa.
(b) Determine the required width based upon an allowable shear stress of 0.7 MPa.
★★
3P
L
—
2
A
P
h=
280 mm
C
B
L
L
—
3
b
PROB. 5.8-12
Shear Stresses in Circular Beams
5.9-1 A wood pole of solid circular cross section (d diameter) is subjected to a horizontal force P 450 lb (see
figure). The length of the pole is L 6 ft, and the allowable
stresses in the wood are 1900 psi in bending and 120 psi in
shear.
Determine the minimum required diameter of the pole
based upon (a) the allowable bending stress, and (b) the
allowable shear stress.
Bea
m
P
ll
Wa
PROB. 5.8-11
381
d
L
PROB. 5.9-1
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
d
382
CHAPTER 5 Stresses in Beams (Basic Topics)
5.9-2 A simple log bridge in a remote area consists of two
parallel logs with planks across them (see figure). The logs
are Douglas fir with average diameter 300 mm. A truck
moves slowly across the bridge, which spans 2.5 m. Assume
that the weight of the truck is equally distributed between
the two logs.
Because the wheelbase of the truck is greater than 2.5 m,
only one set of wheels is on the bridge at a time. Thus, the
wheel load on one log is equivalent to a concentrated load W
acting at any position along the span. In addition, the weight
of one log and the planks it supports is equivalent to a uniform load of 850 N/m acting on the log.
Determine the maximum permissible wheel load W
based upon (a) an allowable bending stress of 7.0 MPa, and
(b) an allowable shear stress of 0.75 MPa.
(a) Determine the minimum required diameter of the
poles based upon an allowable bending stress of 7500 psi in
the aluminum.
(b) Determine the minimum required diameter based
upon an allowable shear stress of 2000 psi.
b
h2
d
t=—
10
Wind
load
d
h1
PROBS. 5.9-3 and 5.9-4
5.9-4 Solve the preceding problem for a sign and poles
having the following dimensions: h1 6.0 m, h2 1.5 m,
b 3.0 m, and t d/10. The design wind pressure is
3.6 kPa, and the allowable stresses in the aluminum are
50 MPa in bending and 14 MPa in shear.
★
x
W
850 N/m
300 mm
Shear Stresses in Beams with Flanges
5.10-1 through 5.10-6 A wide-flange beam (see figure)
2.5 m
PROB. 5.9-2
5.9-3 A sign for an automobile service station is supported
by two aluminum poles of hollow circular cross section, as
shown in the figure. The poles are being designed to resist a
wind pressure of 75 lb/ft2 against the full area of the sign.
The dimensions of the poles and sign are h1 20 ft, h2 5
ft, and b 10 ft. To prevent buckling of the walls of the
poles, the thickness t is specified as one-tenth the outside
diameter d.
★
having the cross section described below is subjected to a
shear force V. Using the dimensions of the cross section,
calculate the moment of inertia and then determine the
following quantities:
(a) The maximum shear stress tmax in the web.
(b) The minimum shear stress tmin in the web.
(c) The average shear stress taver (obtained by dividing
the shear force by the area of the web) and the ratio
tmax/taver.
(d) The shear force Vweb carried in the web and the ratio
Vweb /V.
(Note: Disregard the fillets at the junctions of the web
and flanges and determine all quantities, including the
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
383
CHAPTER 5 Problems
moment of inertia, by considering the cross section to
consist of three rectangles.)
y
z
O h1
5.10-8 A bridge girder AB on a simple span of length
L 14 m supports a uniform load of intensity q that
includes the weight of the girder (see figure). The girder is
constructed of three plates welded to form the cross section
shown.
Determine the maximum permissible load q based upon
(a) an allowable bending stress sallow 110 MPa, and (b)
an allowable shear stress tallow 50 MPa.
h
450 mm
q
t
A
b
30 mm
B
L = 14 m
PROBS. 5.10-1 through 5.10-6
15 mm
5.10-1 Dimensions of cross section: b 6 in., t 0.5 in.,
1800 mm
h 12 in., h1 10.5 in., and V 30 k.
5.10-2 Dimensions of cross section: b 180 mm,
t 12 mm, h 420 mm, h1 380 mm, and V 125 kN.
30 mm
5.10-3 Wide-flange shape, W 8 28 (see Table E-1,
Appendix E); V 10 k.
450 mm
5.10-4 Dimensions of cross section: b 220 mm,
t 12 mm, h 600 mm, h1 570 mm, and V 200 kN.
PROB. 5.10-8
5.10-5 Wide-flange shape, W 18 71 (see Table E-1,
5.10-9 A simple beam with an overhang supports a uniform
Appendix E); V 21 k.
5.10-6 Dimensions of cross section: b 120 mm,
t 7 mm, h 350 mm, h1 330 mm, and V 60 kN.
5.10-7 A cantilever beam AB of length L 6.5 ft supports
a uniform load of intensity q that includes the weight of the
beam (see figure). The beam is a steel W 10 12 wideflange shape (see Table E-1, Appendix E).
Calculate the maximum permissible load q based upon
(a) an allowable bending stress sallow 16 ksi, and (b) an
allowable shear stress tallow 8.5 ksi. (Note: Obtain the
moment of inertia and section modulus of the beam from
Table E-1.)
q
q = 1200 lb/ft
C
B
W 10 12
12 ft
L = 6.5 ft
PROB. 5.10-7
P = 3000 lb
8 ft
A
B
A
load of intensity q 1200 lb/ft and a concentrated load
P 3000 lb (see figure). The uniform load includes an
allowance for the weight of the beam. The allowable
stresses in bending and shear are 18 ksi and 11 ksi, respectively.
Select from Table E-2, Appendix E, the lightest I-beam
(S shape) that will support the given loads.
Hint: Select a beam based upon the bending stress and
then calculate the maximum shear stress. If the beam is
overstressed in shear, select a heavier beam and repeat.
PROB. 5.10-9
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
4 ft
384
CHAPTER 5 Stresses in Beams (Basic Topics)
5.10-10 A hollow steel box beam has the rectangular cross
section shown in the figure. Determine the maximum allowable shear force V that may act on the beam if the allowable
shear stress is 36 MPa.
y
t
h1
z
20
mm
C
c
h
b
450
10 mm mm
10 mm
20
mm
PROBS. 5.10-12 and 5.10-13
5.10-13 Calculate the maximum shear stress tmax in the
web of the T-beam shown in the figure if b 10 in., t 0.6
in., h 8 in., h1 7 in., and the shear force V 5000 lb.
200 mm
Built-Up Beams
PROB. 5.10-10
5.10-11 A hollow aluminum box beam has the square
cross section shown in the figure. Calculate the maximum
and minimum shear stresses tmax and tmin in the webs of
the beam due to a shear force V 28 k.
5.11-1 A prefabricated wood I-beam serving as a floor joist
has the cross section shown in the figure. The allowable load
in shear for the glued joints between the web and the flanges
is 65 lb/in. in the longitudinal direction.
Determine the maximum allowable shear force Vmax for
the beam.
y
0.75 in.
1.0 in.
1.0 in.
z
0.625 in.
O
5 in.
12 in.
8 in.
0.75 in.
PROB. 5.11-1
PROB. 5.10-11
5.10-12 The T-beam shown in the figure has cross-sectional
dimensions as follows: b 220 mm, t 15 mm, h 300 mm,
and h1 275 mm. The beam is subjected to a shear force
V 60 kN.
Determine the maximum shear stress tmax in the web of
the beam.
5.11-2 A welded steel girder having the cross section shown
in the figure is fabricated of two 280 mm 25 mm flange
plates and a 600 mm 15 mm web plate. The plates are
joined by four fillet welds that run continuously for the
length of the girder. Each weld has an allowable load in
shear of 900 kN/m.
Calculate the maximum allowable shear force Vmax for
the girder.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 5 Problems
y
y
25 mm
25 mm
z
z
O
15 mm
280 mm
600 mm
O
50
mm
50
mm
260 mm
25 mm
385
260 mm
25 mm
PROB. 5.11-4
PROB. 5.11-2
5.11-3 A welded steel girder having the cross section shown
in the figure is fabricated of two 18 in. 1 in. flange plates
and a 64 in. 3/8 in. web plate. The plates are joined by
four longitudinal fillet welds that run continuously throughout the length of the girder.
If the girder is subjected to a shear force of 300 kips,
what force F (per inch of length of weld) must be resisted by
each weld?
5.11-5 A box beam constructed of four wood boards of size
6 in. 1 in. (actual dimensions) is shown in the figure. The
boards are joined by screws for which the allowable load in
shear is F 250 lb per screw.
Calculate the maximum permissible longitudinal
spacing smax of the screws if the shear force V is 1200 lb.
y
y
1 in.
1 in.
z
O
1 in.
z
O
1 in.
64 in.
1 in.
3
— in.
8
18 in.
6 in.
6 in.
1 in.
PROB. 5.11-3
5.11-4 A box beam of wood is constructed of two
260 mm 50 mm boards and two 260 mm 25 mm boards
(see figure). The boards are nailed at a longitudinal spacing
s 100 mm.
If each nail has an allowable shear force F 1200 N,
what is the maximum allowable shear force Vmax?
PROB. 5.11-5
5.11-6 Two wood box beams (beams A and B) have the
same outside dimensions (200 mm 360 mm) and the same
thickness (t 20 mm) throughout, as shown in the figure on
the next page. Both beams are formed by nailing, with each
nail having an allowable shear load of 250 N. The beams are
designed for a shear force V 3.2 kN.
(a) What is the maximum longitudinal spacing sA for
the nails in beam A?
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
386
CHAPTER 5 Stresses in Beams (Basic Topics)
(b) What is the maximum longitudinal spacing sB for
the nails in beam B?
(c) Which beam is more efficient in resisting the shear
force?
y
y
5.11-8 A beam of T cross section is formed by nailing
together two boards having the dimensions shown in the figure.
If the total shear force V acting on the cross section is
1600 N and each nail may carry 750 N in shear, what is the
maximum allowable nail spacing s?
y
200 mm
A
z
B
360
mm
O
z
t=
20 mm
O
50 mm
360
mm
z
C
t=
20 mm
200 mm
50 mm
200 mm
200 mm
PROB. 5.11-8
PROB. 5.11-6
5.11-7 A hollow wood beam with plywood webs has the
cross-sectional dimensions shown in the figure. The
plywood is attached to the flanges by means of small nails.
Each nail has an allowable load in shear of 30 lb.
Find the maximum allowable spacing s of the nails at
cross sections where the shear force V is equal to (a) 200 lb
and (b) 300 lb.
5.11-9 The T-beam shown in the figure is fabricated by
welding together two steel plates. If the allowable load for
each weld is 2.0 k/in. in the longitudinal direction, what is
the maximum allowable shear force V?
y
0.5 in.
3
— in.
16
3
— in.
16
z
3 in.
C
5 in.
y
z
3
in.
4
3
in.
4
PROB. 5.11-7
0.5 in.
PROB. 5.11-9
8 in.
O
6 in.
5.11-10 A steel beam is built up from a W 16 77 wideflange beam and two 10 in. 1/2 in. cover plates (see figure
on the next page). The allowable load in shear on each bolt
is 2.1 kips.
What is the required bolt spacing s in the longitudinal
direction if the shear force V 30 kips? (Note: Obtain the
dimensions and moment of inertia of the W shape from
Table E-1.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 5 Problems
y
1
10 in. —
2 in.
cover plates
387
Determine the maximum tensile and compressive
stresses st and sc, respectively, in the crank.
P = 25 lb
z
W 16 77
O
7
d= —
16 in.
7
b = 4—
8 in.
PROB. 5.11-10
5.11-11 Two W 10 45 steel wide-flange beams are
bolted together to form a built-up beam as shown in the
figure.
What is the maximum permissible bolt spacing s if the
shear force V 20 kips and the allowable load in shear on
each bolt is F 3.1 kips? (Note: Obtain the dimensions
and properties of the W shapes from Table E-1.)
PROB. 5.12-1
W 10 45
5.12-2 An aluminum pole for a street light weighs 4600 N
and supports an arm that weighs 660 N (see figure). The
center of gravity of the arm is 1.2 m from the axis of
the pole. The outside diameter of the pole (at its base) is
225 mm and its thickness is 18 mm.
Determine the maximum tensile and compressive
stresses st and sc, respectively, in the pole (at its base) due
to the weights.
W 10 45
W2 = 660 N
1.2 m
PROB. 5.11-11
W1 = 4600 N
Beams with Axial Loads
When solving the problems for Section 5.12, assume that
the bending moments are not affected by the presence of
lateral deflections.
18 mm
5.12-1 While drilling a hole with a brace and bit, you exert
a downward force P 25 lb on the handle of the brace (see
figure). The diameter of the crank arm is d 7/16 in. and its
lateral offset is b 4-7/8 in.
225 mm
PROB. 5.12-2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
388
CHAPTER 5 Stresses in Beams (Basic Topics)
5.12-3 A curved bar ABC having a circular axis (radius
r 12 in.) is loaded by forces P 400 lb (see figure). The
cross section of the bar is rectangular with height h and
thickness t.
If the allowable tensile stress in the bar is 12,000 psi
and the height h 1.25 in., what is the minimum required
thickness tmin?
h
B
45°
P2 = 100 lb
30 ft
C
A
P
Calculate the maximum tensile and compressive
stresses st and sc, respectively, at the base of the tree due to
its weight.
P
45°
r
12 ft
h
P1 = 900 lb
60°
t
PROB. 5.12-3
PROB. 5.12-5
5.12-4 A rigid frame ABC is formed by welding two steel
pipes at B (see figure). Each pipe has cross-sectional area
A 11.31 103 mm2, moment of inertia I 46.37 106
mm4, and outside diameter d 200 mm.
Find the maximum tensile and compressive stresses st
and sc, respectively, in the frame due to the load P 8.0 kN
if L H 1.4 m.
B
d
d
P
A
C
5.12-6 A vertical pole of aluminum is fixed at the base and
pulled at the top by a cable having a tensile force T (see
figure). The cable is attached at the outer surface of the pole
and makes an angle a 25° at the point of attachment. The
pole has length L 2.0 m and a hollow circular cross section with outer diameter d2 260 mm and inner diameter
d1 200 mm.
Determine the allowable tensile force Tallow in the cable
if the allowable compressive stress in the aluminum pole is
90 MPa.
H
d
L
a
T
L
L
d2
PROB. 5.12-4
5.12-5 A palm tree weighing 1000 lb is inclined at an angle
of 60° (see figure). The weight of the tree may be resolved
into two resultant forces, a force P1 900 lb acting at a
point 12 ft from the base and a force P2 100 lb acting at
the top of the tree, which is 30 ft long. The diameter at the
base of the tree is 14 in.
d1
d2
PROB. 5.12-6
5.12-7 Because of foundation settlement, a circular tower is
leaning at an angle a to the vertical (see figure on the next
page). The structural core of the tower is a circular cylinder
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
;@À;À@;;
ÀÀ
@@
;À@;@ÀQ¢;À@
;À@;À@
CHAPTER 5 Problems
of height h, outer diameter d2, and inner diameter d1. For
simplicity in the analysis, assume that the weight of the
tower is uniformly distributed along the height.
Obtain a formula for the maximum permissible angle a
if there is to be no tensile stress in the tower.
389
p
w
H
d1
d2
h
d1
d2
a
PROB. 5.12-7
5.12-8 A steel bar of solid circular cross section is subjected
to an axial tensile force T 26 kN and a bending moment
M 3.2 kNm (see figure).
Based upon an allowable stress in tension of 120 MPa,
determine the required diameter d of the bar. (Disregard the
weight of the bar itself.)
PROB. 5.12-9
5.12-10 A flying buttress transmits a load P 25 kN, acting
at an angle of 60° to the horizontal, to the top of a vertical
buttress AB (see figure). The vertical buttress has height
h 5.0 m and rectangular cross section of thickness
t 1.5 m and width b 1.0 m (perpendicular to the plane
of the figure). The stone used in the construction weighs
g 26 kN/m3.
What is the required weight W of the pedestal and
statue above the vertical buttress (that is, above section A) to
avoid any tensile stresses in the vertical buttress?
Flying
buttress
60°
A
A
M
—t
2
h
T
t
B
PROB. 5.12-8
5.12-9 A cylindrical brick chimney of height H weighs
w 825 lb/ft of height (see figure). The inner and outer
diameters are d1 3 ft and d2 4 ft, respectively. The wind
pressure against the side of the chimney is p 10 lb/ft2 of
projected area.
Determine the maximum height H if there is to be no
tension in the brickwork.
P
W
h
t
B
PROB. 5.12-10
5.12-11 A plain concrete wall (i.e., a wall with no steel
reinforcement) rests on a secure foundation and serves as a
small dam on a creek (see figure on the next page). The
height of the wall is h 6.0 ft and the thickness of the wall
is t 1.0 ft.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
390
CHAPTER 5 Stresses in Beams (Basic Topics)
(a) Determine the maximum tensile and compressive
stresses st and sc, respectively, at the base of the wall when
the water level reaches the top (d h). Assume plain concrete has weight density gc 145 lb/ft3.
(b) Determine the maximum permissible depth dmax of
the water if there is to be no tension in the concrete.
;
@
À
@
À
;
t
5.12-13 Two cables, each carrying a tensile force P
1200 lb, are bolted to a block of steel (see figure). The block
has thickness t 1 in. and width b 3 in.
(a) If the diameter d of the cable is 0.25 in., what are the
maximum tensile and compressive stresses st and sc,
respectively, in the block?
(b) If the diameter of the cable is increased (without
changing the force P), what happens to the maximum tensile
and compressive stresses?
b
P
P
t
h
d
PROB. 5.12-11
PROB. 5.12-13
5.12-14 A bar AB supports a load P acting at the centroid of
Eccentric Axial Loads
5.12-12 A circular post and a rectangular post are each
compressed by loads that produce a resultant force P acting
at the edge of the cross section (see figure). The diameter of
the circular post and the depth of the rectangular post are the
same.
(a) For what width b of the rectangular post will the
maximum tensile stresses be the same in both posts?
(b) Under the conditions described in part (a), which
post has the larger compressive stress?
the end cross section (see figure). In the middle region of the
bar the cross-sectional area is reduced by removing one-half
of the bar.
(a) If the end cross sections of the bar are square with
sides of length b, what are the maximum tensile and compressive stresses st and sc, respectively, at cross section mn
within the reduced region?
(b) If the end cross sections are circular with diameter
b, what are the maximum stresses st and sc?
b
—
2
A
P
b
b
P
b
b
—
2
m
(a)
b
—
2
n
B
P
b
d
PROB. 5.12-12
d
b
(b)
PROB. 5.12-14
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 5 Problems
5.12-15 A short column constructed of a W 10 30 wideflange shape is subjected to a resultant compressive load
P 12 k having its line of action at the midpoint of one
flange (see figure).
(a) Determine the maximum tensile and compressive
stresses st and sc, respectively, in the column.
(b) Locate the neutral axis under this loading condition.
P = 12 k
391
3
5.12-17 A tension member constructed of an L 4 4 4
inch angle section (see Appendix E) is subjected to a tensile
load P 15 kips that acts through the point where the
midlines of the legs intersect (see figure).
Determine the maximum tensile stress st in the angle
section.
2
y
3
3
L44—
4
z
C
1
C
1
P
3
2
W 10 30
PROB. 5.12-17
5.12-18 A short length of a C 811.5 channel is subjected
PROB. 5.12-15
5.12-16 A short column of wide-flange shape is subjected
to a compressive load that produces a resultant force P
60 kN acting at the midpoint of one flange (see figure).
(a) Determine the maximum tensile and compressive
stresses st and sc , respectively, in the column.
(b) Locate the neutral axis under this loading condition.
to an axial compressive force P that has its line of action
through the midpoint of the web of the channel (see figure).
(a) Determine the equation of the neutral axis under this
loading condition.
(b) If the allowable stresses in tension and compression
are 10,000 psi and 8,000 psi, respectively, find the maximum permissible load Pmax.
y
P
C 8 × 11.5
z
C
P = 60 kN
y
PROB. 5.12-18
P
Stress Concentrations
8 mm
z
C
12 mm
160
mm
PROB. 5.12-16
200
mm
The problems for Section 5.13 are to be solved considering
the stress-concentration factors.
5.13-1 The beams shown in the figure are subjected to
bending moments M 2100 lb-in. Each beam has a rectangular cross section with height h 1.5 in. and width
b 0.375 in. (perpendicular to the plane of the figure on the
next page).
(a) For the beam with a hole at midheight, determine
the maximum stresses for hole diameters d 0.25, 0.50,
0.75, and 1.00 in.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
392
CHAPTER 5 Stresses in Beams (Basic Topics)
(b) For the beam with two identical notches (inside
height h1 1.25 in.), determine the maximum stresses for
notch radii R 0.05, 0.10, 0.15, and 0.20 in.
M
M
h
d
(a)
2R
M
M
h
h1
(b)
PROBS. 5.13-1 through 5.13-4
5.13-2 The beams shown in the figure are subjected to
bending moments M 250 Nm. Each beam has a rectangular cross section with height h 44 mm and width b
10 mm (perpendicular to the plane of the figure).
(a) For the beam with a hole at midheight, determine
the maximum stresses for hole diameters d 10, 16, 22, and
28 mm.
(b) For the beam with two identical notches (inside
height h1 40 mm), determine the maximum stresses for
notch radii R 2, 4, 6, and 8 mm.
5.13-3 A rectangular beam with semicircular notches, as
shown in part (b) of the figure, has dimensions h 0.88 in.
and h1 0.80 in. The maximum allowable bending stress in
the metal beam is smax 60 ksi, and the bending moment is
M 600 lb-in.
Determine the minimum permissible width bmin of the
beam.
5.13-4 A rectangular beam with semicircular notches, as
shown in part (b) of the figure, has dimensions h 120 mm
and h1 100 mm. The maximum allowable bending stress
in the plastic beam is smax 6 MPa, and the bending
moment is M 150 Nm.
Determine the minimum permissible width bmin of the
beam.
5.13-5 A rectangular beam with notches and a hole (see figure) has dimensions h 5.5 in., h1 5 in., and width
b 1.6 in. The beam is subjected to a bending moment
M 130 k-in., and the maximum allowable bending stress
in the material (steel) is smax 42,000 psi.
(a) What is the smallest radius Rmin that should be used
in the notches?
(b) What is the diameter dmax of the largest hole that
should be drilled at the midheight of the beam?
2R
M
M
h1
h
PROB. 5.13-5
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
d
6
Stresses in Beams
(Advanced Topics)
6.1 INTRODUCTION
In this chapter we continue our study of the bending of beams by examining several specialized topics, including the analysis of composite
beams (that is, beams of more than one material), beams with inclined
loads, unsymmetric beams, shear stresses in thin-walled beams, shear
centers, and elastoplastic bending. These subjects are based upon the
fundamental topics discussed previously in Chapter 5—topics such as
curvature, normal stresses in beams (including the flexure formula), and
shear stresses in beams.
Later, in Chapters 9 and 10, we will discuss two additional subjects
of fundamental importance in beam design—deflections of beams and
statically indeterminate beams.
6.2 COMPOSITE BEAMS
Beams that are fabricated of more than one material are called composite beams. Examples are bimetallic beams (such as those used in
thermostats), plastic coated pipes, and wood beams with steel reinforcing plates (see Fig. 6-1 on the next page).
Many other types of composite beams have been developed in
recent years, primarily to save material and reduce weight. For instance,
sandwich beams are widely used in the aviation and aerospace industries, where light weight plus high strength and rigidity are required.
Such familiar objects as skis, doors, wall panels, book shelves, and
cardboard boxes are also manufactured in sandwich style.
393
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
394
CHAPTER 6 Stresses in Beams (Advanced Topics)
A typical sandwich beam (Fig. 6-2) consists of two thin faces of
relatively high-strength material (such as aluminum) separated by a
thick core of lightweight, low-strength material. Since the faces are at
the greatest distance from the neutral axis (where the bending stresses
are highest), they function somewhat like the flanges of an I-beam. The
core serves as a filler and provides support for the faces, stabilizing them
against wrinkling or buckling. Lightweight plastics and foams, as well
as honeycombs and corrugations, are often used for cores.
(a)
Strains and Stresses
The strains in composite beams are determined from the same basic
axiom that we used for finding the strains in beams of one material,
namely, cross sections remain plane during bending. This axiom is valid
for pure bending regardless of the nature of the material (see Section
5.4). Therefore, the longitudinal strains ex in a composite beam vary
linearly from top to bottom of the beam, as expressed by Eq. (5-4),
which is repeated here:
(b)
y
ex ky
r
(c)
FIG. 6-1 Examples of composite beams:
(a) bimetallic beam, (b) plastic-coated
steel pipe, and (c) wood beam reinforced
with a steel plate
(6-1)
In this equation, y is the distance from the neutral axis, r is the radius of
curvature, and k is the curvature.
Beginning with the linear strain distribution represented by Eq.
(6-1), we can determine the strains and stresses in any composite beam.
To show how this is accomplished, consider the composite beam shown
in Fig. 6-3. This beam consists of two materials, labeled 1 and 2 in the
figure, which are securely bonded so that they act as a single solid beam.
As in our previous discussions of beams (Chapter 5), we assume that
the xy plane is a plane of symmetry and that the xz plane is the neutral
plane of the beam. However, the neutral axis (the z axis in Fig. 6-3b) does
not pass through the centroid of the cross-sectional area when the beam
is made of two different materials.
If the beam is bent with positive curvature, the strains ex will vary as
shown in Fig. 6-3c, where eA is the compressive strain at the top of the
beam, eB is the tensile strain at the bottom, and eC is the strain at the
contact surface of the two materials. Of course, the strain is zero at
the neutral axis (the z axis).
The normal stresses acting on the cross section can be obtained
from the strains by using the stress-strain relationships for the two
materials. Let us assume that both materials behave in a linearly elastic
manner so that Hooke’s law for uniaxial stress is valid. Then the stresses
in the materials are obtained by multiplying the strains by the appropriate modulus of elasticity.
Denoting the moduli of elasticity for materials 1 and 2 as E1 and E2,
respectively, and also assuming that E2 E1, we obtain the stress
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 6.2 Composite Beams
395
diagram shown in Fig. 6-3d. The compressive stress at the top of the
beam is sA E1eA and the tensile stress at the bottom is sB E2eB.
At the contact surface (C ) the stresses in the two materials are
different because their moduli are different. In material 1 the stress is
s1C E1eC and in material 2 it is s2C E2eC.
Using Hooke’s law and Eq. (6-1), we can express the normal
stresses at distance y from the neutral axis in terms of the curvature:
sx1 E1ky
(a)
(6-2a,b)
sx2 E2ky
in which sx1 is the stress in material 1 and sx2 is the stress in material 2.
With the aid of these equations, we can locate the neutral axis and obtain
the moment-curvature relationship.
Neutral Axis
The position of the neutral axis (the z axis) is found from the condition
that the resultant axial force acting on the cross section is zero (see
Section 5.5); therefore,
(b)
1
sx1 d A
2
sx2 d A 0
(a)
where it is understood that the first integral is evaluated over the crosssectional area of material 1 and the second integral is evaluated over the
y
(c)
FIG. 6-2 Sandwich beams with:
(a) plastic core, (b) honeycomb core, and
(c) corrugated core
z
1
2
(a)
x
y
1
FIG. 6-3 (a) Composite beam of two
materials, (b) cross section of beam,
(c) distribution of strains ex throughout
the height of the beam, and (d) distribution of stresses sx in the beam for the
case where E2 E1
sA = E1eA
eA
A
eC s1C
C
z
2
O
(b)
B
eB
(c)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
s2C
sB = E2eB
(d)
396
CHAPTER 6 Stresses in Beams (Advanced Topics)
cross-sectional area of material 2. Replacing sx1 and sx2 in the preceding equation by their expressions from Eqs. (6-2a) and (6-2b), we get
E1kydA
1
2
E2kydA 0
Since the curvature is a constant at any given cross section, it is not
involved in the integrations and can be cancelled from the equation;
thus, the equation for locating the neutral axis becomes
(6-3)
E1 ydA E2 ydA 0
1
The integrals in this equation represent the first moments of the two
parts of the cross-sectional area with respect to the neutral axis. (If there
are more than two materials—a rare condition—additional terms are
required in the equation.)
Equation (6-3) is a generalized form of the analogous equation for
a beam of one material (Eq. 5-8). The details of the procedure for
locating the neutral axis with the aid of Eq. (6-3) are illustrated later in
Example 6-1.
If the cross section of a beam is doubly symmetric, as in the case
of a wood beam with steel cover plates on the top and bottom (Fig. 6-4),
the neutral axis is located at the midheight of the cross section and
Eq. (6-3) is not needed.
y
t
h
—
2
z
h
O
2
h
—
2
t
FIG. 6-4 Doubly symmetric cross section
Moment-Curvature Relationship
The moment-curvature relationship for a composite beam of two materials (Fig. 6-3) may be determined from the condition that the moment
resultant of the bending stresses is equal to the bending moment M
acting at the cross section. Following the same steps as for a beam of
one material (see Eqs. 5-9 through 5-12), and also using Eqs. (6-2a) and
(6-2b), we obtain
M
sx ydA
A
sx1 ydA
1
kE1 y 2dA kE2 y 2dA
1
sx2 ydA
2
(b)
2
This equation can be written in the simpler form
M k (E1I1 E2I2)
(6-4)
in which I1 and I2 are the moments of inertia about the neutral axis (the
z axis) of the cross-sectional areas of materials 1 and 2, respectively.
Note that I I1 I2, where I is the moment of inertia of the entire
cross-sectional area about the neutral axis.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 6.2 Composite Beams
397
Equation (6-4) can now be solved for the curvature in terms of the
bending moment:
1
M
k
r
E1I1 E2I2
(6-5)
This equation is the moment-curvature relationship for a beam of two
materials (compare with Eq. 5-12 for a beam of one material). The
denominator on the right-hand side is the flexural rigidity of the
composite beam.
Normal Stresses (Flexure Formulas)
The normal stresses (or bending stresses) in the beam are obtained by
substituting the expression for curvature (Eq. 6-5) into the expressions
for sx1 and sx2 (Eqs. 6-2a and 6-2b); thus,
MyE1
sx1
E1I1 E2I2
MyE2
sx2
E1I1 E 2I2
(6-6a,b)
These expressions, known as the flexure formulas for a composite
beam, give the normal stresses in materials 1 and 2, respectively. If the
two materials have the same modulus of elasticity (E1 E2 E), then
both equations reduce to the flexure formula for a beam of one material
(Eq. 5-13).
The analysis of composite beams, using Eqs. (6-3) through (6-6), is
illustrated in Examples 6-1 and 6-2 at the end of this section.
ÀÀ
;;
@@
;;
@@
ÀÀ
Approximate Theory for Bending of Sandwich Beams
Sandwich beams having doubly symmetric cross sections and composed
of two linearly elastic materials (Fig. 6-5) can be analyzed for bending
y
1
t
2
z
O
hc
h
1
FIG. 6-5 Cross section of a sandwich
beam having two axes of symmetry
(doubly symmetric cross section)
b
t
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
398
CHAPTER 6 Stresses in Beams (Advanced Topics)
using Eqs. (6-5) and (6-6), as described previously. However, we can
also develop an approximate theory for bending of sandwich beams by
introducing some simplifying assumptions.
If the material of the faces (material 1) has a much larger modulus
of elasticity than does the material of the core (material 2), it is reasonable to disregard the normal stresses in the core and assume that the
faces resist all of the longitudinal bending stresses. This assumption is
equivalent to saying that the modulus of elasticity E2 of the core is zero.
Under these conditions the flexure formula for material 2 (Eq. 6-6b)
gives sx2 0 (as expected), and the flexure formula for material 1
(Eq. 6-6a) gives
My
sx1
I1
(6-7)
which is similar to the ordinary flexure formula (Eq. 5-13). The quantity
I1 is the moment of inertia of the two faces evaluated with respect to the
neutral axis; thus,
b
I1 h3 h3c
12
(6-8)
in which b is the width of the beam, h is the overall height of the beam,
and hc is the height of the core. Note that hc h 2t where t is the
thickness of the faces.
The maximum normal stresses in the sandwich beam occur at the
top and bottom of the cross section where y h/2 and h/2, respectively. Thus, from Eq. (6-7), we obtain
Mh
stop
2I1
Mh
s bottom
2I1
(6-9a,b)
If the bending moment M is positive, the upper face is in compression
and the lower face is in tension. (These equations are conservative
because they give stresses in the faces that are higher than those
obtained from Eqs. 6-6a and 6-6b.)
If the faces are thin compared to the thickness of the core (that is, if
t is small compared to hc), we can disregard the shear stresses in the
faces and assume that the core carries all of the shear stresses. Under
these conditions the average shear stress and average shear strain in the
core are, respectively,
V
taver
bhc
V
gaver
bhcGc
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(6-10a,b)
SECTION 6.2 Composite Beams
399
in which V is the shear force acting on the cross section and Gc is the
shear modulus of elasticity for the core material. (Although the maximum
shear stress and maximum shear strain are larger than the average values,
the average values are often used for design purposes.)
Limitations
Throughout the preceding discussion of composite beams, we assumed
that both materials followed Hooke’s law and that the two parts of the
beam were adequately bonded so that they acted as a single unit. Thus,
our analysis is highly idealized and represents only a first step in understanding the behavior of composite beams and composite materials.
Methods for dealing with nonhomogeneous and nonlinear materials,
bond stresses between the parts, shear stresses on the cross sections,
buckling of the faces, and other such matters are treated in reference
books dealing specifically with composite construction.
Reinforced concrete beams are one of the most complex types of
composite construction (Fig. 6-6), and their behavior differs significantly from that of the composite beams discussed in this section.
Concrete is strong in compression but extremely weak in tension.
Consequently, its tensile strength is usually disregarded entirely. Under
those conditions, the formulas given in this section do not apply.
Furthermore, most reinforced concrete beams are not designed on
the basis of linearly elastic behavior—instead, more realistic design
methods (based upon load-carrying capacity instead of allowable
stresses) are used. The design of reinforced concrete members is a
highly specialized subject that is presented in courses and textbooks
devoted solely to that subject.
FIG. 6-6 Reinforced concrete beam
with longitudinal reinforcing bars and
vertical stirrups
¢¢¢
@@@
ÀÀÀ
;;;
QQQ
QQQ
¢¢¢
@@@
ÀÀÀ
;;;
¢Q;À@
@@@
ÀÀÀ
;;;
¢Q;À@
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
400
CHAPTER 6 Stresses in Beams (Advanced Topics)
Example 6-1
1
y
A composite beam (Fig. 6-7) is constructed from a wood beam (4.0 in. 6.0 in.
actual dimensions) and a steel reinforcing plate (4.0 in. wide and 0.5 in. thick).
The wood and steel are securely fastened to act as a single beam. The beam is
subjected to a positive bending moment M 60 k-in.
Calculate the largest tensile and compressive stresses in the wood
(material 1) and the maximum and minimum tensile stresses in the steel (material 2) if E1 1500 ksi and E2 30,000 ksi.
A
h1
z
h2
2
6 in.
Solution
O
4 in.
C
B
0.5 in.
Neutral axis. The first step in the analysis is to locate the neutral axis of
the cross section. For that purpose, let us denote the distances from the neutral
axis to the top and bottom of the beam as h1 and h2, respectively. To obtain these
distances, we use Eq. (6-3). The integrals in that equation are evaluated by
taking the first moments of areas 1 and 2 about the z axis, as follows:
FIG. 6-7 Example 6-1. Cross section of a
composite beam of wood and steel
1
2
ydA y1A1 (h1 3 in.)(4 in. 6 in.) (h1 3 in.)(24 in.2)
ydA y2 A2 (6.25 in. h1)(4 in. 0.5 in.) (h1 6.25 in.)(2 in.2)
in which A1 and A2 are the areas of parts 1 and 2 of the cross section, y1 and y2
are the y coordinates of the centroids of the respective areas, and h1 has units of
inches.
Substituting the preceding expressions into Eq. (6-3) gives the equation for
locating the neutral axis, as follows:
E1
1
ydA E2
ydA 0
2
or
(1500 ksi)(h1 3 in.)(24 in.2) (30,000 ksi)(h1 6.25 in.)(2 in.2) 0
Solving this equation, we obtain the distance h1 from the neutral axis to the top
of the beam:
hl 5.031 in.
Also, the distance h2 from the neutral axis to the bottom of the beam is
h2 6.5 in. hl 1.469 in.
Thus, the position of the neutral axis is established.
Moments of inertia. The moments of inertia I1 and I2 of areas A1 and A2
with respect to the neutral axis can be found by using the parallel-axis theorem
(see Section 12.5 of Chapter 12). Beginning with area 1 (Fig. 6-7), we get
1
Il (4 in.)(6 in.) 3 (4 in.)(6 in.)(h1 3 in.) 2 171.0 in.4
12
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 6.2 Composite Beams
401
Similarly, for area 2 we get
1
I2 (4 in.)(0.5 in.) 3 (4 in.)(0.5 in.)(h2 0.25 in.) 2 3.01 in.4
12
To check these calculations, we can determine the moment of inertia I of the
entire cross-sectional area about the z axis as follows:
1
1
I (4 in.)h31 (4 in.)h 32 169.8 4.2 174.0 in.4
3
3
which agrees with the sum of I1 and I2.
Normal stresses. The stresses in materials 1 and 2 are calculated from
the flexure formulas for composite beams (Eqs. 6-6a and b). The largest
compressive stress in material 1 occurs at the top of the beam (A) where y
h1 5.031 in. Denoting this stress by s1A and using Eq. (6-6a), we get
Mh1E1
s1A
E1I1 E2I2
(60 k-in.)(5.031 in.)(1500 ksi)
1310 psi
The largest tensile stress in material 1 occurs at the contact plane between the
two materials (C) where y (h2 0.5 in.) 0.969 in. Proceeding as in the
previous calculation, we get
(60 k-in.)(0.969 in.)(1500 ksi)
251 psi
s1C
(1500 ksi)(171.0 in.4) (30,000 ksi)(3.01 in.4)
Thus, we have found the largest compressive and tensile stresses in the wood.
The steel plate (material 2) is located below the neutral axis, and therefore
it is entirely in tension. The maximum tensile stress occurs at the bottom of the
beam (B) where y h2 1.469 in. Hence, from Eq. (6-6b) we get
M(h2)E2
s2B
E1I1 E2I2
(60 k-in.)(1.469 in.)(30,000 ksi)
7620 psi
(1500 ksi)(171.0 in.4) (30,000 ksi)(3.01 in.4)
The minimum tensile stress in material 2 occurs at the contact plane (C) where
y 0.969 in. Thus,
(60 k-in.)(0.969 in.)(30,000 ksi)
s2C
5030 psi
(1500 ksi)(171.0 in.4) (30,000 ksi)(3.01 in.4)
These stresses are the maximum and minimum tensile stresses in the steel.
Note: At the contact plane the ratio of the stress in the steel to the stress in
the wood is
s2C /s1C 5030 psi/251 psi 20
which is equal to the ratio E2 /E1 of the moduli of elasticity (as expected).
Although the strains in the steel and wood are equal at the contact plane, the
stresses are different because of the different moduli.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
402
CHAPTER 6 Stresses in Beams (Advanced Topics)
Example 6-2
A sandwich beam having aluminum-alloy faces enclosing a plastic core (Fig.
6-8) is subjected to a bending moment M 3.0 kNm. The thickness of the
faces is t 5 mm and their modulus of elasticity is E1 72 GPa. The height of
the plastic core is hc 150 mm and its modulus of elasticity is E2 800 MPa.
The overall dimensions of the beam are h 160 mm and b 200 mm.
Determine the maximum tensile and compressive stresses in the faces and
the core using: (a) the general theory for composite beams, and (b) the approximate theory for sandwich beams.
ÀÀ
;;
@@
;;
@@
ÀÀ
y
1
2
z
h
—
2
FIG. 6-8 Example 6-2. Cross section of
sandwich beam having aluminum-alloy
faces and a plastic core
O
hc =
150 mm
h=
160 mm
1
b = 200 mm
Solution
t = 5 mm
t = 5 mm
Neutral axis. Because the cross section is doubly symmetric, the neutral
axis (the z axis in Fig. 6-8) is located at midheight.
Moments of inertia. The moment of inertia I1 of the cross-sectional areas
of the faces (about the z axis) is
200 mm
b
I1 (h3 h3c ) (160 mm)3 (150 mm)3 12.017 106 mm4
12
12
and the moment of inertia I2 of the plastic core is
200 mm
b
I2 (h3c) (150 mm)3 56.250 106 mm4
12
12
As a check on these results, note that the moment of inertia of the entire crosssectional area about the z axis (I bh3/12) is equal to the sum of I1 and I2.
(a) Normal stresses calculated from the general theory for composite
beams. To calculate these stresses, we use Eqs. (6-6a) and (6-6b). As a preliminary matter, we will evaluate the term in the denominator of those equations
(that is, the flexural rigidity of the composite beam):
E1I1 E2I2 (72 GPa)(12.017 106 mm4) (800 MPa)(56.250 106 mm4)
910,200 Nm2
The maximum tensile and compressive stresses in the aluminum faces are found
from Eq. (6-6a):
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 6.3 Transformed-Section Method
(s 1)max
403
M(h/2)(E1)
E1I1 E2I2
(3.0 kNm)(80 mm)(72 GPa)
910,200 Nm2
19.0 MPa
The corresponding quantities for the plastic core (from Eq. 6-6b) are
(s2)max
M(hc /2)(E2)
E1I1 E2I2
(3.0 kNm)(75 mm)(800 MPa)
910,200 Nm2
0.198 MPa
The maximum stresses in the faces are 96 times greater than the maximum
stresses in the core, primarily because the modulus of elasticity of the aluminum
is 90 times greater than that of the plastic.
(b) Normal stresses calculated from the approximate theory for sandwich
beams. In the approximate theory we disregard the normal stresses in the core
and assume that the faces transmit the entire bending moment. Then the
maximum tensile and compressive stresses in the faces can be found from
Eqs. (6-9a) and (6-9b), as follows:
(s1)max
Mh
2I1
(3.0 kNm)(80 mm)
12.017 106 mm4
20.0 MPa
As expected, the approximate theory gives slightly higher stresses in the faces
than does the general theory for composite beams.
6.3 TRANSFORMED-SECTION METHOD
The transformed-section method is an alternative procedure for
analyzing the bending stresses in a composite beam. The method is
based upon the theories and equations developed in the preceding
section, and therefore it is subject to the same limitations (for instance, it
is valid only for linearly elastic materials) and gives the same results.
Although the transformed-section method does not reduce the calculating effort, many designers find that it provides a convenient way to
visualize and organize the calculations.
The method consists of transforming the cross section of a
composite beam into an equivalent cross section of an imaginary beam
that is composed of only one material. This new cross section is called
the transformed section. Then the imaginary beam with the transformed section is analyzed in the customary manner for a beam of one
material. As a final step, the stresses in the transformed beam are
converted to those in the original beam.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
404
CHAPTER 6 Stresses in Beams (Advanced Topics)
b1
Neutral Axis and Transformed Section
If the transformed beam is to be equivalent to the original beam, its
neutral axis must be located in the same place and its moment-resisting
capacity must be the same. To show how these two requirements are
met, consider again a composite beam of two materials (Fig. 6-9a). The
neutral axis of the cross section is obtained from Eq. (6-3), which is
repeated here:
y
1
z
2
O
b2
E1
(a)
y
ydA 0
(6-11)
2
E2
n
E1
1
1
In this equation, the integrals represent the first moments of the two
parts of the cross section with respect to the neutral axis.
Let us now introduce the notation
b1
z
ydA ⫹ E2
1
O
(6-12)
where n is the modular ratio. With this notation, we can rewrite Eq.
(6-11) in the form
nb2
(b)
1
FIG. 6-9 Composite beam of two
materials: (a) actual cross section, and
(b) transformed section consisting only
of material 1
y dA
yn dA 0
(6-13)
2
Since Eqs. (6-11) and (6-13) are equivalent, the preceding equation
shows that the neutral axis is unchanged if each element of area dA in
material 2 is multiplied by the factor n, provided that the y coordinate
for each such element of area is not changed.
Therefore, we can create a new cross section consisting of two
parts: (1) area 1 with its dimensions unchanged, and (2) area 2 with its
width (that is, its dimension parallel to the neutral axis) multiplied by n.
This new cross section (the transformed section) is shown in Fig. 6-9b
for the case where E2 E1 (and therefore n 1). Its neutral axis is in
the same position as the neutral axis of the original beam. (Note that all
dimensions perpendicular to the neutral axis remain the same.)
Since the stress in the material (for a given strain) is proportional to
the modulus of elasticity (s Ee), we see that multiplying the width of
material 2 by n E2 /E1 is equivalent to transforming it to material 1.
For instance, suppose that n 10. Then the area of part 2 of the cross
section is now 10 times wider than before. If we imagine that this part of
the beam is now material 1, we see that it will carry the same force as
before because its modulus is reduced by a factor of 10 (from E2 to E1)
at the same time that its area is increased by a factor of 10. Thus, the
new section (the transformed section) consists only of material 1.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 6.3 Transformed-Section Method
405
Moment-Curvature Relationship
The moment-curvature relationship for the transformed beam must be
the same as for the original beam. To show that this is indeed the case,
we note that the stresses in the transformed beam (since it consists only
of material 1) are given by Eq. (5-7) of Section 5.5:
sx E1ky
Using this equation, and also following the same procedure as for a
beam of one material (see Section 5.5), we can obtain the momentcurvature relation for the transformed beam:
sxy dA
M
A
E1k
1
sxy dA
1
y2 dA E1k
sxy dA
2
y 2 dA k (E1I1 E1 nI2)
2
or
M k(E1 I1 E2 I2)
(6-14)
This equation is the same as Eq. (6-4), thereby demonstrating that the
moment-curvature relationship for the transformed beam is the same as
for the original beam.
Normal Stresses
Since the transformed beam consists of only one material, the normal
stresses (or bending stresses) can be found from the standard flexure
formula (Eq. 5-13). Thus, the normal stresses in the beam transformed to
material 1 (Fig. 6-9b) are
My
sx1
IT
(6-15)
where IT is the moment of inertia of the transformed section with respect
to the neutral axis. By substituting into this equation, we can calculate
the stresses at any point in the transformed beam. (As explained later,
the stresses in the transformed beam match those in the original beam
in the part of the original beam consisting of material 1; however, in the
part of the original beam consisting of material 2, the stresses are
different from those in the transformed beam.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
406
CHAPTER 6 Stresses in Beams (Advanced Topics)
We can easily verify Eq. (6-15) by noting that the moment of inertia
of the transformed section (Fig. 6-9b) is related to the moment of inertia
of the original section (Fig. 6-9a) by the following relation:
E2
IT I1 nI2 I1 I2
E1
(6-16)
Substituting this expression for IT into Eq. (6-15) gives
MyE1
sx1
E1I1 E2I2
(a)
which is the same as Eq. (6-6a), thus demonstrating that the stresses in
material 1 in the original beam are the same as the stresses in the corresponding part of the transformed beam.
As mentioned previously, the stresses in material 2 in the original
beam are not the same as the stresses in the corresponding part of
the transformed beam. Instead, the stresses in the transformed beam
(Eq. 6-15) must be multiplied by the modular ratio n to obtain the
stresses in material 2 of the original beam:
My
sx 2 n
IT
(6-17)
We can verify this formula by noting that when Eq. (6-16) for IT is
substituted into Eq. (6-17), we get
MynE1
MyE2
sx 2
E1I1 E2I2
E1I1 E2I2
(b)
which is the same as Eq. (6-6b).
General Comments
In this discussion of the transformed-section method we chose to transform the original beam to a beam consisting entirely of material 1. It is
also possible to transform the beam to material 2. In that case the
stresses in the original beam in material 2 will be the same as the
stresses in the corresponding part of the transformed beam. However,
the stresses in material 1 in the original beam must be obtained by multiplying the stresses in the corresponding part of the transformed beam by
the modular ratio n, which in this case is defined as n E1/E2.
It is also possible to transform the original beam into a material
having any arbitrary modulus of elasticity E, in which case all parts of
the beam must be transformed to the fictitious material. Of course, the
calculations are simpler if we transform to one of the original materials.
Finally, with a little ingenuity it is possible to extend the transformedsection method to composite beams of more than two materials.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 6.3 Transformed-Section Method
407
Example 6-3
The composite beam shown in Fig. 6-10a is formed of a wood beam (4.0 in.
6.0 in. actual dimensions) and a steel reinforcing plate (4.0 in. wide and
0.5 in. thick). The beam is subjected to a positive bending moment M 60 k-in.
Using the transformed-section method, calculate the largest tensile and
compressive stresses in the wood (material 1) and the maximum and minimum
tensile stresses in the steel (material 2) if E1 1500 ksi and E2 30,000 ksi.
Note: This same beam was analyzed previously in Example 6-1 of Section 6.2.
1
y
1
A
4 in.
A
y
h1
FIG. 6-10 Example 6-3. Composite
beam of Example 6-1 analyzed by the
transformed-section method:
(a) cross section of original beam, and
(b) transformed section (material 1)
z
h2
2
6 in.
O
4 in.
z
C
B
O
h2
80 in.
0.5 in.
1
(a)
6 in.
h1
0.5 in.
C
B
(b)
Solution
Transformed section. We will transform the original beam into a beam of
material 1, which means that the modular ratio is defined as
E2
30,000 ksi
n 20
E1
1,500 ksi
The part of the beam made of wood (material 1) is not altered but the part made
of steel (material 2) has its width multiplied by the modular ratio. Thus, the
width of this part of the beam becomes
n(4 in.) 20(4 in.) 80 in.
in the transformed section (Fig. 6-10b).
Neutral axis. Because the transformed beam consists of only one material,
the neutral axis passes through the centroid of the cross-sectional area. Therefore, with the top edge of the cross section serving as a reference line, and with
the distance yi measured positive downward, we can calculate the distance h1 to
the centroid as follows:
(3 in.)(4 in.)(6 in.) (6.25 in.)(80 in.)(0.5 in.)
yi Ai
h1
(4 in.)(6 in.) (80 in.)(0.5 in.)
Ai
322.0 in.3
5.031 in.
64.0 in.2
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
408
CHAPTER 6 Stresses in Beams (Advanced Topics)
Also, the distance h2 from the lower edge of the section to the centroid is
h 2 6.5 in. h1 1.469 in.
Thus, the location of the neutral axis is determined.
Moment of inertia of the transformed section. Using the parallel-axis
theorem (see Section 12.5 of Chapter 12), we can calculate the moment of
inertia IT of the entire cross-sectional area with respect to the neutral axis as
follows:
1
IT (4 in.)(6 in.)3 (4 in.)(6 in.)(h1 3 in.) 2
12
1
(80 in.)(0.5 in.)3 (80 in.)(0.5 in.)(h2 0.25 in.) 2
12
171.0 in.4 60.3 in.4 231.3 in.4
1
Normal stresses in the wood (material 1). The stresses in the transformed
beam (Fig. 6-10b) at the top of the cross section (A) and at the contact plane
between the two parts (C) are the same as in the original beam (Fig. 6-10a).
These stresses can be found from the flexure formula (Eq. 6-15), as follows:
y
A
h1
z
h2
2
My
(60 k-in.)(5.031 in.)
s1A 1310 psi
IT
6 in.
O
4 in.
(60 k-in.)(0.969 in.)
My
s1C
251 psi
231.3 in.4
IT
C
B
0.5 in.
(a)
1
4 in.
A
y
h1
z
80 in.
1
(60 k-in.)(1.469 in.)
My
(20) 7620 psi
s2B n
231.3 in.4
IT
6 in.
O
h2
(b)
FIG. 6-10 (Repeated)
These are the largest tensile and compressive stresses in the wood (material 1)
in the original beam. The stress s1A is compressive and the stress s1C is tensile.
Normal stresses in the steel (material 2). The maximum and minimum
stresses in the steel plate are found by multiplying the corresponding stresses in
the transformed beam by the modular ratio n (Eq. 6-17). The maximum stress
occurs at the lower edge of the cross section (B) and the minimum stress occurs
at the contact plane (C):
0.5 in.
C
B
(60 k-in.)(0.969 in.)
My
s2C n
(20) 5030 psi
231.3 in.4
IT
Both of these stresses are tensile.
Note that the stresses calculated by the transformed-section method agree
with those found in Example 6-1 by direct application of the formulas for a
composite beam.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 6.4 Doubly Symmetric Beams with Inclined Loads
409
6.4 DOUBLY SYMMETRIC BEAMS WITH INCLINED LOADS
y
z
x
FIG. 6-11 Beam with a lateral load acting
in a plane of symmetry
y
In our previous discussions of bending we dealt with beams possessing a
longitudinal plane of symmetry (the xy plane in Fig. 6-11) and
supporting lateral loads acting in that same plane. Under these conditions the bending stresses can be obtained from the flexure formula (Eq.
5-13) provided the material is homogeneous and linearly elastic.
In this section, we will extend those ideas and consider what
happens when the beam is subjected to loads that do not act in the plane
of symmetry, that is, inclined loads (Fig. 6-12). We will limit our
discussion to beams that have a doubly symmetric cross section, that is,
both the xy and xz planes are planes of symmetry. Also, the inclined
loads must act through the centroid of the cross section to avoid twisting
the beam about the longitudinal axis.
We can determine the bending stresses in the beam shown in
Fig. 6-12 by resolving the inclined load into two components, one acting
in each plane of symmetry. Then the bending stresses can be obtained
from the flexure formula for each load component acting separately, and
the final stresses can be obtained by superposing the separate stresses.
Sign Conventions for Bending Moments
z
x
FIG. 6-12 Doubly symmetric beam with
an inclined load
y
My
z
As a preliminary matter, we will establish sign conventions for the
bending moments acting on cross sections of a beam.* For this purpose,
we cut through the beam and consider a typical cross section (Fig. 6-13).
The bending moments My and Mz acting about the y and z axes, respectively, are represented as vectors using double-headed arrows. The
moments are positive when their vectors point in the positive directions
of the corresponding axes, and the right-hand rule for vectors gives the
direction of rotation (indicated by the curved arrows in the figure).
From Fig. 6-13 we see that a positive bending moment My produces
compression on the right-hand side of the beam (the negative z side) and
tension on the left-hand side (the positive z side). Similarly, a positive
moment Mz produces compression on the upper part of the beam (where
y is positive) and tension on the lower part (where y is negative). Also, it
is important to note that the bending moments shown in Fig. 6-13 act on
the positive x face of a segment of the beam, that is, on a face having its
outward normal in the positive direction of the x axis.
Normal Stresses (Bending Stresses)
Mz
x
FIG. 6-13 Sign conventions for bending
moments My and Mz
The normal stresses associated with the individual bending moments My
and Mz are obtained from the flexure formula (Eq. 5-13). These stresses
*The directions of the normal and shear stresses in a beam are usually apparent from an
inspection of the beam and its loading, and therefore we often calculate stresses by ignoring
sign conventions and using only absolute values. However, when deriving general formulas
we need to maintain rigorous sign conventions to avoid ambiguity in the equations.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
410
CHAPTER 6 Stresses in Beams (Advanced Topics)
n
z
My
A
b
z
are then superposed to give the stresses produced by both moments
acting simultaneously. For instance, consider the stresses at a point in the
cross section having positive coordinates y and z (point A in Fig. 6-14). A
positive moment My produces tension at this point and a positive
moment Mz produces compression; thus, the normal stress at point A is
y
y
Mz
C
My z
Mz y
sx
Iy
Iz
n
FIG. 6-14 Cross section of beam
subjected to bending moments
My and Mz
(6-18)
in which Iy and Iz are the moments of inertia of the cross-sectional area
with respect to the y and z axes, respectively. Using this equation, we
can find the normal stress at any point in the cross section by substituting the appropriate algebraic values of the moments and the
coordinates.
Neutral Axis
The equation of the neutral axis can be determined by equating the
normal stress sx (Eq. 6-18) to zero:
My
Mz
z y 0
Iy
Iz
(6-19)
This equation shows that the neutral axis nn is a straight line passing
through the centroid C (Fig. 6-14). The angle b between the neutral axis
and the z axis is determined as follows:
MyIz
y
tan b
MzIy
z
(6-20)
Depending upon the magnitudes and directions of the bending moments,
the angle b may vary from 90° to 90°. Knowing the orientation of
the neutral axis is useful when determining the points in the cross
section where the normal stresses are the largest. (Since the stresses vary
linearly with distance from the neutral axis, the maximum stresses occur
at points located farthest from the neutral axis.)
Relationship Between the Neutral Axis and the Inclination
of the Loads
As we have just seen, the orientation of the neutral axis with respect to
the z axis is determined by the bending moments and the moments of
inertia (Eq. 6-20). Now we wish to determine the orientation of the
neutral axis relative to the angle of inclination of the loads acting on the
beam. For this purpose, we will use the cantilever beam shown in
Fig. 6-15a as an example. The beam is loaded by a force P acting in the
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
411
SECTION 6.4 Doubly Symmetric Beams with Inclined Loads
y
L
u
z
P
plane of the end cross section and inclined at an angle u to the positive y
axis. This particular orientation of the load is selected because it means
that both bending moments (My and Mz) are positive when u is between
0 and 90°.
The load P can be resolved into components P cos u in the positive y
direction and P sin u in the negative z direction. Therefore, the bending
moments My and Mz (Fig. 6-15b) acting on a cross section located at
distance x from the fixed support are
My (P sin u )(L x)
(a)
x
n
y
u
b
P
C
Mz
(6-22)
which shows that the resultant moment vector M is at the angle u from
the z axis (Fig. 6-15b). Consequently, the resultant moment vector is
perpendicular to the longitudinal plane containing the force P.
The angle b between the neutral axis nn and the z axis (Fig. 6-15b)
is obtained from Eq. (6-20):
u
z
(6-21a,b)
in which L is the length of the beam. The ratio of these moments is
My
tan u
Mz
My
M
Mz (P cos u )(L x)
My Iz
Iz
tan b tan u
MzIy
Iy
n
(b)
FIG. 6-15 Doubly symmetric beam with
an inclined load P acting at an angle u to
the positive y axis
(6-23)
which shows that the angle b is generally not equal to the angle u. Thus,
except in special cases, the neutral axis is not perpendicular to the
longitudinal plane containing the load.
Exceptions to this general rule occur in three special cases:
1. When the load lies in the xy plane (u 0 or 180°), which means
that the z axis is the neutral axis.
2. When the load lies in the xz plane (u 90°), which means that
the y axis is the neutral axis.
3. When the principal moments of inertia are equal, that is, when
Iy Iz.
In case (3), all axes through the centroid are principal axes and all
have the same moment of inertia. The plane of loading, no matter what
its direction, is always a principal plane, and the neutral axis is always
perpendicular to it. (This situation occurs with square, circular, and
certain other cross sections, as described in Section 12.9 of Chapter 12.)
The fact that the neutral axis is not necessarily perpendicular to the
plane of the load can greatly affect the stresses in a beam, especially if
the ratio of the principal moments of inertia is very large. Under these
conditions the stresses in the beam are very sensitive to slight changes in
the direction of the load and to irregularities in the alignment of the
beam itself. This characteristic of certain beams is illustrated later in
Example 6-5.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
412
CHAPTER 6 Stresses in Beams (Advanced Topics)
Example 6-4
A wood beam AB of rectangular cross section serving as a roof purlin (Figs.
6-16a and b) is simply supported by the top chords of two adjacent roof trusses.
The beam supports the weight of the roof sheathing and the roofing material,
plus its own weight and any additional loads that affect the roof (such as wind,
snow, and earthquake loads).
In this example, we will consider only the effects of a uniformly distributed
load of intensity q 3.0 kN/m acting in the vertical direction through the
centroids of the cross sections (Fig. 6-16c). The load acts along the entire length
of the beam and includes the weight of the beam. The top chords of the trusses
have a slope of 1 on 2 (a 26.57°), and the beam has width b 100 mm,
height h 150 mm, and span L 1.6 m.
Determine the maximum tensile and compressive stresses in the beam and
locate the neutral axis.
y
Roof
sheathing
b
A
A
a
Purlin
B
B
Roof truss
a
z
C
q
h
a
FIG. 6-16 Example 6-4. Wood beam of
rectangular cross section serving as a
roof purlin
1
(c)
(b)
(a)
2
a = 26.57°
Solution
Loads and bending moments. The uniform load q acting in the vertical
direction can be resolved into components in the y and z directions (Fig. 6-17a):
qy q cos a
qz q sin a
(6-24a,b)
The maximum bending moments occur at the midpoint of the beam and are
found from the general formula M qL2/8; hence,
qz L2
qL2sin a
My
8
8
qy L2
qL2cos a
Mz
8
(6-25a,b)
Both of these moments are positive because their vectors are in the positive
directions of the y and z axes (Fig. 6-17b).
Moments of inertia. The moments of inertia of the cross-sectional area
with respect to the y and z axes are as follows:
hb3
bh3
Iz
(6-26a,b)
Iy
12
12
Bending stresses. The stresses at the midsection of the beam are obtained
from Eq. (6-18) with the bending moments given by Eqs. (6-25) and the
moments of inertia given by Eqs. (6-26):
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 6.4 Doubly Symmetric Beams with Inclined Loads
My z
Mz y
qL2sin a
qL2cos a
sx 3 z 3y
Iz
8hb /12
8bh /12
y
D
z
3qL2 sin a
cos a
z
y
2bh b2
h2
C
a
qz
E
qy
(a)
y
My
M
b
C
q 3.0 kN/m
Mz
E
n
b
a
(6-28)
Numerical values. The maximum tensile and compressive stresses can be
calculated from the preceding equation by substituting the given data:
h
a
z
(6-27)
3qL2 sin a
cos a
sE sD
4bh
b
h
D
n
The stress at any point in the cross section can be obtained from this equation
by substituting the coordinates y and z of the point.
From the orientation of the cross section and the directions of the loads and
bending moments (Fig. 6-17), it is apparent that the maximum compressive
stress occurs at point D (where y h/2 and z b/2) and the maximum tensile
stress occurs at point E (where y h/2 and z b/2). Substituting these coordinates into Eq. (6-27) and then simplifying, we obtain expressions for the
maximum and minimum stresses in the beam:
q
a
413
L 1.6 m
b 100 mm
h 150 mm
a 26.57°
The results are
sE sD 4.01 MPa
(b)
FIG. 6-17 Solution to Example 6-4.
(a) Components of the uniform load, and
(b) bending moments acting on a cross
section
Neutral axis. In addition to finding the stresses in the beam, it is often
useful to locate the neutral axis. The equation of this line is obtained by setting
the stress (Eq. 6-27) equal to zero:
sin a
cos a
z
y0
b2
h2
(6-29)
The neutral axis is shown in Fig. 6-17b as line nn. The angle b from the z axis
to the neutral axis is obtained from Eq. (6-29) as follows:
y
h2
tan a
tan b
z
b2
(6-30)
Substituting numerical values, we get
(150 mm)2
h2
tan b 2 tan a 2 tan 26.57° 1.125
b
(100 mm)
b 48.4°
Since the angle b is not equal to the angle a, the neutral axis is inclined to the
plane of loading (which is vertical).
From the orientation of the neutral axis (Fig. 6-17b), we see that points D
and E are the farthest from the neutral axis, thus confirming our assumption that
the maximum stresses occur at those points. The part of the beam above and
to the right of the neutral axis is in compression, and the part to the left and
below the neutral axis is in tension.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
414
CHAPTER 6 Stresses in Beams (Advanced Topics)
Example 6-5
A 12-foot long cantilever beam (Fig. 6-18a) is constructed from an S 24 80
section (see Table E-2 of Appendix E for the dimensions and properties of this
beam). A load P 10 k acts in the vertical direction at the end of the beam.
Because the beam is very narrow compared to its height (Fig. 6-18b), its
moment of inertia about the z axis is much larger than its moment of inertia
about the y axis.
(a) Determine the maximum bending stresses in the beam if the y axis
of the cross section is vertical and therefore aligned with the load P (Fig. 6-18a).
(b) Determine the maximum bending stresses if the beam is inclined at a
small angle a 1° to the load P (Fig. 6-18b). (A small inclination can be
caused by imperfections in the fabrication of the beam, misalignment of the
beam during construction, or movement of the supporting structure.)
y
y
L = 12 ft
z
A
n
b = 41°
C
z
C
n
S 24 80
B
x
FIG. 6-18 Example 6-5. Cantilever beam
P = 10 k
with moment of inertia Iz much larger
than Iy
P
a = 1°
(b)
(a)
Solution
(a) Maximum bending stresses when the load is aligned with the y axis. If
the beam and load are in perfect alignment, the z axis is the neutral axis and the
maximum stresses in the beam (at the support) are obtained from the flexure
formula:
My
PL(h/2)
smax
Iz
Iz
in which M PL is the bending moment at the support, h is the height of the
beam, and Iz is the moment of inertia about the z axis. Substituting numerical
values, we obtain
(10 k)(12 ft)(12 in./ft)(12.00 in.)
smax 8230 psi
This stress is tensile at the top of the beam and compressive at the bottom of the
beam.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 6.4 Doubly Symmetric Beams with Inclined Loads
415
(b) Maximum bending stresses when the load is inclined to the y axis. We
now assume that the beam has a small inclination (Fig. 6-18b), so that the angle
between the y axis and the load is a 1°.
The components of the load P are P cos a in the negative y direction and
P sin a in the positive z direction. Therefore, the bending moments at the
support are
My (P sin a)L (10 k)(sin 1°)(12 ft)(12 in./ft) 25.13 k-in.
Mz (P cos a)L (10 k)(cos 1°)(12 ft)(12 in./ft) 1440 k-in.
The angle b giving the orientation of the neutral axis nn (Fig. 6-18b) is obtained
from Eq. (6-20):
My Iz
y
(25.13 k-in.)(2100 in.4)
0.8684
tan b
z
(1440 k-in.)(42.2 in.4)
b 41°
This calculation shows that the neutral axis is inclined at an angle of 41° from
the z axis even though the plane of the load is inclined only 1° from the y axis.
The sensitivity of the position of the neutral axis to the angle of the load is a
consequence of the large Iz /Iy ratio.
From the position of the neutral axis (Fig. 6-18b), we see that the
maximum stresses in the beam occur at points A and B, which are located at the
farthest distances from the neutral axis. The coordinates of point A are
zA 3.50 in.
yA 12.0 in.
Therefore, the tensile stress at point A (see Eq. 6-18) is
My zA
Mz yA
sA
Iz
(1440 k-in.)(12.0 in.)
(25.13 k-in.)(3.50 in.)
2100 in.4
42.2 in.4
2080 psi 8230 psi 10,310 psi
The stress at B has the same magnitude but is a compressive stress:
sB 10,310 psi
These stresses are 25% larger than the stress smax 8230 psi for the same
beam with a perfectly aligned load. Furthermore, the inclined load produces a
lateral deflection in the z direction, whereas the perfectly aligned load does not.
This example shows that beams with Iz much larger than Iy may develop
large stresses if the beam or its loads deviate even a small amount from their
planned alignment. Therefore, such beams should be used with caution, because
they are highly susceptible to overstress and to lateral (that is, sideways) bending and buckling. The remedy is to provide adequate lateral support for the
beam, thereby preventing sideways bending. For instance, wood floor joists in
buildings are supported laterally by installing bridging or blocking between the
joists.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
416
CHAPTER 6 Stresses in Beams (Advanced Topics)
6.5 BENDING OF UNSYMMETRIC BEAMS
y
z
M
x
(a)
y
z
In our previous discussions of bending, we assumed that the beams had
cross sections with at least one axis of symmetry. Now we will abandon
that restriction and consider beams having unsymmetric cross sections.
We begin by investigating beams in pure bending, and then in later
sections (Sections 6.6 through 6.9) we will consider the effects of lateral
loads. As in earlier discussions, it is assumed that the beams are made of
linearly elastic materials.
Suppose that a beam having an unsymmetric cross section is
subjected to a bending moment M acting at the end cross section
(Fig. 6-19a). We would like to know the stresses in the beam and the
position of the neutral axis. Unfortunately, at this stage of the analysis
there is no direct way of determining these quantities. Therefore, we will
use an indirect approach—instead of starting with a bending moment
and trying to find the neutral axis, we will start with an assumed neutral
axis and find the associated bending moment.
Neutral Axis
y
dA
z
C
(b)
FIG. 6-19 Unsymmetric beam subjected
to a bending moment M
We begin by constructing two perpendicular axes (the y and z axes) at an
arbitrarily selected point in the plane of the cross section (Fig. 6-19b).
The axes may have any orientation, but for convenience we will orient
them horizontally and vertically. Next, we assume that the beam is bent
in such a manner that the z axis is the neutral axis of the cross section.
Consequently, the beam deflects in the xy plane, which becomes the
plane of bending. Under these conditions, the normal stress acting on an
element of area dA located at distance y from the neutral axis (see
Fig. 6-19b and Eq. 5-7 of Chapter 5) is
sx Ek y y
(6-31)
The minus sign is needed because the part of the beam above the z axis
(the neutral axis) is in compression when the curvature is positive. (The
sign convention for curvature when the beam is bent in the xy plane is
shown in Fig. 6-20a.)
The force acting on the element of area dA is sx dA, and the
resultant force acting on the entire cross section is the integral of this
elemental force over the cross-sectional area A. Since the beam is in
pure bending, the resultant force must be zero; hence,
sx dA
A
Eky y dA 0
A
The modulus of elasticity and the curvature are constants at any given
cross section, and therefore
y dA 0
A
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(6-32)
SECTION 6.5 Bending of Unsymmetric Beams
417
This equation shows that the z axis (the neutral axis) passes through the
centroid C of the cross section.
Now assume that the beam is bent in such a manner that the y axis is
the neutral axis and the xz plane is the plane of bending. Then the
normal stress acting on the element of area dA (Fig. 6-19b) is
y
Positive
curvature ky
x
O
The sign convention for the curvature kz in the xz plane is shown in
Fig. 6-20b. The minus sign is needed in Eq. (6-33) because positive
curvature in the xz plane produces compression on the element dA. The
resultant force for this case is
(a)
z
sx dA
A
⫹
Positive
curvature kz
x
FIG. 6-20 Sign conventions for curvatures
ky and kz in the xy and xz planes,
respectively
y
My
A
C
FIG. 6-21 Bending moments My and Mz
acting about the y and z axes,
respectively
(6-34)
and again we see that the neutral axis must pass through the centroid.
Thus, we have established that the origin of the y and z axes for an
unsymmetric beam must be placed at the centroid C.
Now let us consider the moment resultant of the stresses sx. Once
again we assume that bending takes place with the z axis as the neutral
axis, in which case the stresses sx are given by Eq. (6-31). The corresponding bending moments Mz and My about the z and y axes,
respectively (Fig. 6-21), are
Mz
My
Mz
z dA 0
A
(b)
z
Ekz z dA 0
A
from which we get
O
(6-33)
sx Ekz z
⫹
A
y 2 dA ky EIz
(6-35a)
yz dA k y EIyz
(6-35b)
sx y dA ky E
A
sx z dA ky E
A
In these equations, Iz is the moment of inertia of the cross-sectional area
with respect to the z axis and Iyz is the product of inertia with respect to
the y and z axes.*
From Eqs. (6-35a) and (6-35b) we can draw the following conclusions: (1) If the z axis is selected in an arbitrary direction through the
centroid, it will be the neutral axis only if moments My and Mz act about
the y and z axes and only if these moments are in the ratio established by
Eqs. (6-35a) and (6-35b). (2) If the z axis is selected as a principal axis,
then the product of inertia Iyz equals zero and the only bending moment
is Mz. In that case, the z axis is the neutral axis, bending takes place in
the xy plane, and the moment Mz acts in that same plane. Thus, bending
occurs in a manner analogous to that of a symmetric beam.
In summary, an unsymmetric beam bends in the same general manner as a symmetric beam provided the z axis is a principal centroidal
*Products of inertia are discussed in Section 12.7 of Chapter 12.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
418
CHAPTER 6 Stresses in Beams (Advanced Topics)
axis and the only bending moment is the moment Mz acting about that
same axis.
If we now assume that the y axis is the neutral axis, we will arrive at
similar conclusions. The stresses sx are given by Eq. (6-33) and the
bending moments are
My
sx z dA kz E
A
Mz
z2dA kz EIy
(6-36a)
yz dA kzEIyz
(6-36b)
A
sx y dA kz E
A
A
in which Iy is the moment of inertia with respect to the y axis. Again we
observe that if the neutral axis (the y axis in this case) is oriented arbitrarily, moments My and Mz must exist. However, if the y axis is a
principal axis, the only moment is My and we have ordinary bending in
the xz plane. Therefore, we can state that an unsymmetric beam bends
in the same general manner as a symmetric beam when the y axis is a
principal centroidal axis and the only bending moment is the moment
My acting about that same axis.
One further observation—since the y and z axes are orthogonal, we
know that if either axis is a principal axis, then the other axis is automatically a principal axis.
We have now arrived at the following important conclusion: When
an unsymmetric beam is in pure bending, the plane in which the bending
moment acts is perpendicular to the neutral surface only if the y and z
axes are principal centroidal axes of the cross section and the bending
moment acts in one of the two principal planes (the xy plane or the xz
plane). In such a case, the principal plane in which the bending moment
acts becomes the plane of bending and the usual bending theory
(including the flexure formula) is valid.
Having arrived at this conclusion, we now have a direct method for
finding the stresses in an unsymmetric beam subjected to a bending
moment acting in an arbitrary direction.
y
n
M
b
z
Procedure for Analyzing an Unsymmetric Beam
My
u
Mz
C
n
FIG. 6-22 Unsymmetric cross section
with the bending moment M resolved
into components My and Mz acting about
the principal centroidal axes
We will now describe a general procedure for analyzing an unsymmetric
beam subjected to any bending moment M (Fig. 6-22). We begin by
locating the centroid C of the cross section and constructing a set of
principal axes at that point (the y and z axes in the figure).* Next, the
bending moment M is resolved into components My and Mz, positive in
the directions shown in the figure. These components are
My M sin u
Mz M cos u
(6-37a,b)
in which u is the angle between the moment vector M and the z axis
(Fig. 6-22). Since each component acts in a principal plane, it produces
*Principal axes are discussed in Sections 12.8 and 12.9 of Chapter 12.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 6.5 Bending of Unsymmetric Beams
419
pure bending in that same plane. Thus, the usual formulas for pure
bending apply, and we can readily find the stresses due to the moments
My and Mz acting separately. The bending stresses obtained from the
moments acting separately are then superposed to obtain the stresses
produced by the original bending moment M. (Note that this general
procedure is similar to that described in the preceding section for
analyzing doubly symmetric beams with inclined loads.)
The superposition of the bending stresses in order to obtain the
resultant stress at any point in the cross section is given by Eq. (6-18):
My z
Mz y
(M sin u )z
(M cos u )y
sx
Iy
Iz
Iy
Iz
(6-38)
in which y and z are the coordinates of the point under consideration.
Also, the equation of the neutral axis nn (Fig. 6-22) is obtained by
setting sx equal to zero and simplifying:
sin u
cos u
z y 0
Iy
Iz
(6-39)
The angle b between the neutral axis and the z axis can be obtained
from the preceding equation, as follows:
y
Iz
tan b tan u
z
Iy
(6-40)
This equation shows that in general the angles b and u are not equal,
hence the neutral axis is generally not perpendicular to the plane in
which the applied couple M acts. The only exceptions are the three
special cases described in the preceding section in the paragraph
following Eq. (6-23).
In this section we have focused our attention on unsymmetric
beams. Of course, symmetric beams are special cases of unsymmetric
beams, and therefore the discussions of this section also apply to
symmetric beams. If a beam is singly symmetric, the axis of symmetry
is one of the centroidal principal axes of the cross section; the other
principal axis is perpendicular to the axis of symmetry at the centroid. If
a beam is doubly symmetric, the two axes of symmetry are centroidal
principal axes.
In a strict sense the discussions of this section apply only to pure
bending, which means that no shear forces act on the cross sections.
When shear forces do exist, the possibility arises that the beam will twist
about the longitudinal axis. However, twisting is avoided when the shear
forces act through the shear center, which is described in the next
section.
The following examples illustrates the analysis of a beam having
one axis of symmetry. (The calculations for an unsymmetric beam
having no axes of symmetry proceed in the same general manner, except
that the determination of the various cross-sectional properties is much
more complex.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
420
CHAPTER 6 Stresses in Beams (Advanced Topics)
Example 6-6
y
A channel section (C 10 15.3) is subjected to a bending moment M
15 k-in. oriented at an angle u 10° to the z axis (Fig. 6-23).
Calculate the bending stresses sA and sB at points A and B, respectively,
and determine the position of the neutral axis.
A
C 10 15.3
Solution
Properties of the cross section. The centroid C is located on the axis of
symmetry (the z axis) at a distance
M = 15 k-in.
z
c 0.634 in.
C
from the back of the channel (Fig. 6-24).* The y and z axes are principal
centroidal axes with moments of inertia
u = 10°
Iy 2.28 in.4
Iz 67.4 in.4
Also, the coordinates of points A and B are as follows:
B
yA 5.00 in.
FIG. 6-23 Example 6-6. Channel section
subjected to a bending moment M acting
at an angle u to the z axis
zA 2.600 in. 0.634 in. 1.966 in.
yB 5.00 in.
zB 0.634 in.
Bending moments. The bending moments about the y and z axes
(Fig. 6-24) are
My M sin u (15 k-in.)(sin 10°) 2.605 k-in.
n
Mz M cos u (15 k-in.)(cos 10°) 14.77 k-in.
y
c = 0.634 in.
A
Bending stresses. We now calculate the stress at point A from Eq. (6-38):
My zA
Mz yA
sA
Iz
b = 79.1°
(2.605 k-in.)(1.966 in.)
(14.77 k-in.)(5.00 in.)
2.28 in.4
u = 10°
M
My
z
2246 psi 1096 psi 3340 psi
C
Mz
By a similar calculation we obtain the stress at point B:
My zB
Mz yB
sB
Iz
(14.77 k-in.)(5.00 in.)
(2.605 k-in.)(0.634 in.)
67.4 in.4
B
n
FIG. 6-24 Solution to Example 6-6
724 psi 1096 psi 1820 psi
These stresses are the maximum compressive and tensile stresses in the beam.
*See Table E-3, Appendix E, for dimensions and properties of channel sections.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 6.6 The Shear-Center Concept
421
Neutral axis. The angle b that locates the neutral axis (Eq. 6-40) is found
as follows:
Iz
67.4 in.4
tan b tan u 4 tan 10° 5.212
Iy
2.28 in.
b 79.1°
The neutral axis nn is shown in Fig. 6-24, and we see that points A and B are
located at the farthest distances from the neutral axis, thus confirming that
sA and sB are the largest stresses in the beam.
In this example, the angle b between the z axis and the neutral axis is much
larger than the angle u (Fig. 6-24) because the ratio Iz /Iy is large. The angle b
varies from 0 to 79.1° as the angle u varies from 0 to 10°. As discussed previously in Example 6-5 of Section 6.4, beams with large Iz /Iy ratios are very
sensitive to the direction of loading. Thus, beams of this kind should be
provided with lateral support to prevent excessive lateral deflections.
6.6 THE SHEAR-CENTER CONCEPT
In the preceding sections of this chapter we were concerned with
determining the bending stresses in beams under a variety of special
conditions. For instance, in Section 6.4 we considered symmetrical
beams with inclined loads, and in Section 6.5 we considered unsymmetrical beams. However, lateral loads acting on a beam produce shear
forces as well as bending moments, and therefore in this and the next
three sections we will examine the effects of shear.
In Chapter 5 we saw how to determine the shear stresses in beams
when the loads act in a plane of symmetry, and we derived the shear
formula for calculating those stresses for certain shapes of beams. Now
we will examine the shear stresses in beams when the lateral loads act in
a plane that is not a plane of symmetry. We will find that the loads must
be applied at a particular point in the cross section, called the shear
center, if the beam is to bend without twisting.
Consider a cantilever beam of singly symmetric cross section
supporting a load P at the free end (see Fig. 6-25a on the next page). A
beam having the cross section shown in Fig. 6-25b is called an
unbalanced I-beam. Beams of I-shape, whether balanced or unbalanced,
are usually loaded in the plane of symmetry (the xz plane), but in this
case the line of action of the force P is perpendicular to that plane. Since
the origin of coordinates is taken at the centroid C of the cross section,
and since the z axis is an axis of symmetry of the cross section, both the
y and z axes are principal centroidal axes.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
422
CHAPTER 6 Stresses in Beams (Advanced Topics)
y
z
P
x
(a)
y
S C
z
P
M0
(b)
FIG. 6-25 Cantilever beam with singly
symmetric cross section: (a) beam with
load, and (b) intermediate cross section
of beam showing stress resultants P and
M0, centroid C, and shear center S
y
z
C
S
A
T
P
P
FIG. 6-26 Singly symmetric beam with
load P applied at point A
Let us assume that under the action of the load P the beam bends
with the xz plane as the neutral plane, which means that the xy plane is
the plane of bending. Under these conditions, two stress resultants exist
at intermediate cross sections of the beam (Fig. 6-25b): a bending
moment M0 acting about the z axis and having its moment vector in the
negative direction of the z axis, and a shear force of magnitude P acting
in the negative y direction. For a given beam and loading, both M0 and P
are known quantities.
The normal stresses acting on the cross section have a resultant that
is the bending moment M0, and the shear stresses have a resultant that is
the shear force (equal to P). If the material follows Hooke’s law, the
normal stresses vary linearly with the distance from the neutral axis (the
z axis) and can be calculated from the flexure formula. Since the shear
stresses acting on a cross section are determined from the normal
stresses solely from equilibrium considerations (see the derivation of the
shear formula in Section 5.8), it follows that the distribution of shear
stresses over the cross section is also determined. The resultant of these
shear stresses is a vertical force equal in magnitude to the force P
and having its line of action through some point S lying on the z axis
(Fig. 6-25b). This point is known as the shear center (also called the
center of flexure) of the cross section.
In summary, by assuming that the z axis is the neutral axis, we can
determine not only the distribution of the normal stresses but also the
distribution of the shear stresses and the position of the resultant shear
force. Therefore, we now recognize that a load P applied at the end of
the beam (Fig. 6-25a) must act through a particular point (the shear
center) if bending is to occur with the z axis as the neutral axis.
If the load is applied at some other point on the z axis (say, at point
A in Fig. 6-26), it can be replaced by a statically equivalent system
consisting of a force P acting at the shear center and a torque T. The
force acting at the shear center produces bending about the z axis and
the torque produces torsion. Therefore, we now recognize that a lateral
load acting on a beam will produce bending without twisting only if it
acts through the shear center.
The shear center (like the centroid) lies on any axis of symmetry,
and therefore the shear center S and the centroid C coincide for a doubly
symmetric cross section (Fig. 6-27a). A load P acting through the
centroid produces bending about the y and z axes without torsion, and
the corresponding bending stresses can be found by the method
described in Section 6.4 for doubly symmetric beams.
If a beam has a singly symmetric cross section (Fig. 6-27b), both
the centroid and the shear center lie on the axis of symmetry. A load P
acting through the shear center can be resolved into components in the y
and z directions. The component in the y direction will produce bending
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 6.6 The Shear-Center Concept
y
C,S
z
P
(a)
y
S
z
C
P
(b)
FIG. 6-27 (a) Doubly symmetric beam
with a load P acting through the centroid
(and shear center), and (b) singly
symmetric beam with a load P acting
through the shear center
in the xy plane with the z axis as the neutral axis, and the component in
the z direction will produce bending (without torsion) in the xz plane
with the y axis as the neutral axis. The bending stresses produced by
these components can be superposed to obtain the stresses caused by the
original load.
Finally, if a beam has an unsymmetric cross section (Fig. 6-28),
the bending analysis proceeds as follows (provided the load acts through
the shear center). First, locate the centroid C of the cross section and
determine the orientation of the principal centroidal axes y and z. Then
resolve the load into components (acting at the shear center) in the y and
z directions and determine the bending moments My and Mz about the
principal axes. Lastly, calculate the bending stresses using the method
described in Section 6.5 for unsymmetric beams.
Now that we have explained the significance of the shear center and
its use in beam analysis, it is natural to ask, “How do we locate the shear
center?” For doubly symmetric shapes the answer, of course, is
simple—it is at the centroid. For singly symmetric shapes the shear
center lies on the axis of symmetry, but the precise location on that axis
may not be easy to determine. Locating the shear center is even more
difficult if the cross section is unsymmetric (Fig. 6-28). In such cases,
the task requires more advanced methods than are appropriate for this
book. (A few engineering handbooks give formulas for locating shear
centers; e.g., see Ref. 2-9.)
Beams of thin-walled open cross sections, such as wide-flange
beams, channels, angles, T-beams, and Z-sections, are a special case.
Not only are they in common use for structural purposes, they also are
very weak in torsion. Consequently, it is especially important to locate
their shear centers. Cross sections of this type are considered in the
following three sections—in Sections 6.7 and 6.8 we discuss how to find
the shear stresses in such beams, and in Section 6.9 we show how to
locate their shear centers.
y
z
C
S
FIG. 6-28 Unsymmetric beam with a load
P acting through the shear center S
423
P
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
424
CHAPTER 6 Stresses in Beams (Advanced Topics)
6.7 SHEAR STRESSES IN BEAMS OF THIN-WALLED OPEN CROSS SECTIONS
The distribution of shear stresses in rectangular beams, circular beams,
and in the webs of beams with flanges was described previously in
Sections 5.8, 5.9, and 5.10, and we derived the shear formula (Eq. 5-38)
for calculating the stresses:
VQ
t
Ib
(6-41)
In this formula, V represents the shear force acting on the cross section,
I is the moment of inertia of the cross-sectional area (with respect to the
neutral axis), b is the width of the beam at the location where the shear
stress is to be determined, and Q is the first moment of the crosssectional area outside of the location where the stress is being found.
Now we will consider the shear stresses in a special class of beams
known as beams of thin-walled open cross section. Beams of this type
are distinguished by two features: (1) The wall thickness is small
compared to the height and width of the cross section, and (2) the cross
section is open, as in the case of an I-beam or channel beam, rather than
closed, as in the case of a hollow box beam. Examples are shown in
Fig. 6-29. Beams of this type are also called structural sections or
profile sections.
We can determine the shear stresses in thin-walled beams of open
cross section by using the same techniques we used when deriving the
shear formula (Eq. 6-41). To keep the derivation as general as possible,
we will consider a beam having its cross-sectional centerline mm of
arbitrary shape (Fig. 6-30a). The y and z axes are principal centroidal
axes of the cross section, and the load P acts parallel to the y axis
through the shear center S (Fig. 6-30b). Therefore, bending will occur in
the xy plane with the z axis as the neutral axis.
Under these conditions, we can obtain the normal stress at any point
in the beam from the flexure formula:
Mz y
sx
Iz
(6-42)
where Mz is the bending moment about the z axis (positive as defined in
Fig. 6-13) and y is a coordinate of the point under consideration.
Now consider a volume element abcd cut out between two cross
sections distance dx apart (Fig. 6-30a). Note that the element begins at
the edge of the cross section and has length s measured along the center-
FIG. 6-29 Typical beams of thin-walled
open cross section (wide-flange beam or
I-beam, channel beam, angle section,
Z-section, and T-beam)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
425
SECTION 6.7 Shear Stresses in Beams of Thin-Walled Open Cross Sections
line mm (Fig. 6-30b). To determine the shear stresses, we isolate the
element as shown in Fig. 6-30c. The resultant of the normal stresses
acting on face ad is the force F1 and the resultant on face bc is the force
F2. Since the normal stresses acting on face ad are larger than those
acting on face bc (because the bending moment is larger), the force F1
will be larger than F2. Therefore, shear stresses t must act along face cd
in order for the element to be in equilibrium. These shear stresses act
parallel to the top and bottom surfaces of the element and must be
accompanied by complementary shear stresses acting on the crosssectional faces ad and bc, as shown in the figure.
To evaluate these shear stresses, we sum forces in the x direction for
element abcd (Fig. 6-30c); thus,
tt dx ⫹ F2 F1 0 or t t dx F1 F2
(a)
where t is the thickness of the cross section at face cd of the element. In
other words, t is the thickness of the cross section at distance s from the
free edge (Fig. 6-30b). Next, we obtain an expression for the force F1 by
using Eq. (6-42):
s
Mz1
F1 sx dA
Iz
0
s
(b)
y dA
0
where dA is an element of area on side ad of the volume element abcd,
y is a coordinate to the element dA, and Mz1 is the bending moment at
the cross section. An analogous expression is obtained for the force F2:
s
F2
0
Mz 2
sx dA
Iz
s
(c)
y dA
0
Substituting these expressions for F1 and F2 into Eq. (a), we get
Mz 2 Mz1 1
t
dx
Iz t
s
(d)
y dA
0
y
s
d
a
y
b
c
s
m
x
thin-walled open cross section. (The
y and z axes are principal centroidal
axes.)
x
(a)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
dx
C
S
F2
d
m
P
b
t
z
dx
FIG. 6-30 Shear stresses in a beam of
a
F1
z
t c
P
(b)
(c)
426
CHAPTER 6 Stresses in Beams (Advanced Topics)
The quantity (Mz 2 Mz1)/dx is the rate of change dM/dx of the bending
moment and is equal to the shear force acting on the cross section (see
Eq. 4-6):
Mz2 Mz1
dM
Vy
dx
dx
(6-43)
The shear force Vy is parallel to the y axis and positive in the negative
direction of the y axis, that is, positive in the direction of the force P
(Fig. 6-30). This convention is consistent with the sign convention previously adopted in Chapter 4 (see Fig. 4-5 for the sign convention for
shear forces).
Substituting from Eq. (6-43) into Eq. (d), we get the following equation for the shear stress t :
Vy
t
Iz t
s
y dA
(6-44)
0
This equation gives the shear stresses at any point in the cross section at
distance s from the free edge. The integral on the right-hand side represents the first moment with respect to the z axis (the neutral axis) of the
area of the cross section from s 0 to s s. Denoting this first moment
by Qz, we can write the equation for the shear stresses t in the simpler
form
VyQz
t
Iz t
(6-45)
which is analogous to the standard shear formula (Eq. 6-41).
The shear stresses are directed along the centerline of the cross
section and act parallel to the edges of the section. Furthermore, we
tacitly assumed that these stresses have constant intensity across the
thickness t of the wall, which is a valid assumption when the thickness is
small. (Note that the wall thickness need not be constant but may vary as
a function of the distance s.)
The shear flow at any point in the cross section, equal to the
product of the shear stress and the thickness at that point, is
VyQz
f t t
Iz
(6-46)
Because Vy and Iz are constants, the shear flow is directly proportional to
Qz. At the top and bottom edges of the cross section, Qz is zero and
hence the shear flow is also zero. The shear flow varies continuously
between these end points and reaches its maximum value where Qz is
maximum, which is at the neutral axis.
Now suppose that the beam shown in Fig. 6-30 is bent by loads that
act parallel to the z axis and through the shear center. Then the beam
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 6.8 Shear Stresses in Wide-Flange Beams
427
will bend in the xz plane and the y axis will be the neutral axis. In this
case we can repeat the same type of analysis and arrive at the following
equations for the shear stresses and shear flow (compare with Eqs. 6-45
and 6-46):
Vz Qy
t
Iy t
VzQy
f t t
Iy
(6-47a,b)
In these equations, Vz is the shear force parallel to the z axis and Qy is
the first moment with respect to the y axis.
In summary, we have derived expressions for the shear stresses in
beams of thin-walled open cross sections with the stipulations that the
shear force must act through the shear center and must be parallel to one
of the principal centroidal axes. If the shear force is inclined to the y and
z axes (but still acts through the shear center), it can be resolved into
components parallel to the principal axes. Then two separate analyses
can be made, and the results can be superimposed.
To illustrate the use of the shear-stress equations, we will consider
the shear stresses in a wide-flange beam in the next section. Later, in
Section 6.9, we will use the shear-stress equations to locate the shear
centers of several thin-walled beams with open cross sections.
6.8 SHEAR STRESSES IN WIDE-FLANGE BEAMS
We will now use the concepts and equations discussed in the preceding
section to investigate the shear stresses in wide-flange beams. For
discussion purposes, consider the wide-flange beam of Fig. 6-31a on the
next page. This beam is loaded by a force P acting in the plane of the
web, that is, through the shear center, which coincides with the centroid
of the cross section. The cross-sectional dimensions are shown in Fig.
6-31b, where we note that b is the flange width, h is the height between
centerlines of the flanges, tf is the flange thickness, and tw is the web
thickness.
Shear Stresses in the Upper Flange
We begin by considering the shear stresses at section bb in the righthand part of the upper flange (Fig. 6-31b). Since the distance s has its
origin at the edge of the section (point a), the cross-sectional area
between point a and section bb is stf. Also, the distance from the
centroid of this area to the neutral axis is h/2, and therefore its first
moment Qz is equal to stf h/2. Thus, the shear stress tf in the flange at
section bb (from Eq. 6-45) is
Vy Qz
P(stf h/2)
shP
t f
2Iz
Iz t
Iz t f
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(6-48)
428
CHAPTER 6 Stresses in Beams (Advanced Topics)
y
y
b
c
B
z
A
tf
d
r
z
b
s
a
h
—
2
d
C
h
tw —
2
tf
x
P
b
—
2
b
—
2
(b)
(a)
t1
F1
t2
b
A
a
dx
s
b
tmax
F2
t2
(c)
t1
FIG. 6-31 Shear stresses in a wide-flange
(d)
beam
The direction of this stress can be determined by examining the forces
acting on element A, which is cut out of the flange between point a and
section bb (see Figs. 6-31a and b).
The element is drawn to a larger scale in Fig. 6-31c in order to show
clearly the forces and stresses acting on it. We recognize immediately
that the tensile force F1 is larger than the force F2, because the bending
moment is larger on the rear face of the element than it is on the front
face. It follows that the shear stress on the left-hand face of element A
must act toward the reader if the element is to be in equilibrium. From
this observation it follows that the shear stresses on the front face of
element A must act toward the left.
Returning now to Fig. 6-31b, we see that we have completely determined the magnitude and direction of the shear stress at section bb,
which may be located anywhere between point a and the junction of the
top flange and the web. Thus, the shear stresses throughout the right-
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 6.8 Shear Stresses in Wide-Flange Beams
429
hand part of the flange are horizontal, act to the left, and have a magnitude given by Eq. (6-48). As seen from that equation, the shear stresses
increase linearly with the distance s.
The variation of the stresses in the upper flange is shown graphically in Fig. 6-31d, and we see that the stresses vary from zero at point a
(where s 0) to a maximum value t1 at s b/2:
bhP
t1
4Iz
(6-49)
The corresponding shear flow is
bh tf P
f1 t1tf
4Iz
(6-50)
Note that we have calculated the shear stress and shear flow at the junction of the centerlines of the flange and web, using only centerline
dimensions of the cross section in the calculations. This approximate
procedure simplifies the calculations and is satisfactory for thin-walled
cross sections.
By beginning at point c on the left-hand part of the top flange
(Fig. 6-31b) and measuring s toward the right, we can repeat the same
type of analysis. We will find that the magnitude of the shear stresses is
again given by Eqs. (6-48) and (6-49). However, by cutting out an
element B (Fig. 6-31a) and considering its equilibrium, we find that the
shear stresses on the cross section now act toward the right, as shown in
Fig. 6-31d.
Shear Stresses in the Web
The next step is to determine the shear stresses acting in the web.
Considering a horizontal cut at the top of the web (at the junction of the
flange and web), we find the first moment about the neutral axis to be
Qz btf h/2, so that the corresponding shear stress is
bh t f P
t 2
2Iztw
(6-51)
bh tf P
f2 t 2tw
2Iz
(6-52)
The associated shear flow is
Note that the shear flow f2 is equal to twice the shear flow f1, which is
expected since the shear flows in the two halves of the upper flange
combine to produce the shear flow at the top of the web.
The shear stresses in the web act downward and increase in magnitude until the neutral axis is reached. At section dd, located at distance r
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
430
CHAPTER 6 Stresses in Beams (Advanced Topics)
from the neutral axis (Fig. 6-31b), the shear stress tw in the web is
calculated as follows:
btf h
bt f h
tw h2
h
h/2 r
Qz r (tw) r 2
2
2
2
2
2 4
2
bt f h
h
P
(6-53)
tw r 2
4
2Iz
tw
When r h/2 this equation reduces to Eq. (6-51), and when r 0 it
gives the maximum shear stress:
b tf
h Ph
tmax
tw
4 2Iz
(6-54)
Again it should be noted that we have made all calculations on the basis
of the centerline dimensions of the cross section. For this reason, the
shear stresses in the web of a wide-flange beam calculated from
Eq. (6-53) may be slightly different from those obtained by the more
exact analysis made in Chapter 5 (see Eq. 5-46 of Section 5.10).
The shear stresses in the web vary parabolically, as shown in
Fig. 6-31d, although the variation is not large. The ratio of tmax to t 2 is
tmax
htw
1
t2
4b t f
(6-55)
For instance, if we assume h 2b and tf 2tw, the ratio is tmax/t2
1.25.
Shear Stresses in the Lower Flange
As the final step in the analysis, we can investigate the shear stresses in
the lower flange using the same methods we used for the top flange. We
will find that the magnitudes of the stresses are the same as in the top
flange, but the directions are as shown in Fig. 6-31d.
General Comments
From Fig. 6-31d we see that the shear stresses on the cross section
“flow” inward from the outermost edges of the top flange, then down
through the web, and finally outward to the edges of the bottom flange.
Because this flow is always continuous in any structural section, it
serves as a convenient method for determining the directions of the
stresses. For instance, if the shear force acts downward on the beam of
Fig. 6-31a, we know immediately that the shear flow in the web must
also be downward. Knowing the direction of the shear flow in the web,
we also know the directions of the shear flows in the flanges because of
the required continuity in the flow. Using this simple technique to get
the directions of the shear stresses is easier than visualizing the directions of the forces acting on elements such as A (Fig. 6-31c) cut out
from the beam.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 6.9 Shear Centers of Thin-Walled Open Sections
431
The resultant of all the shear stresses acting on the cross section is
clearly a vertical force, because the horizontal stresses in the flanges
produce no resultant. The shear stresses in the web have a resultant R,
which can be found by integrating the shear stresses over the height of
the web, as follows:
R
h /2
t dA 2
ttw dr
0
Substituting from Eq. (6-53), we get
h /2
2
btf h
b tf
h2
h h twP
P
r 2 dr (6-56)
4
6 2I z
tw
2I z
tw
0
The moment of inertia Iz can be calculated as follows (using centerline
dimensions):
t w h3
bt f h 2
Iz
(6-57)
12
2
in which the first term is the moment of inertia of the web and the
second term is the moment of inertia of the flanges. When this expression for Iz is substituted into Eq. (6-56), we get R P, which
demonstrates that the resultant of the shear stresses acting on the cross
section is equal to the load. Furthermore, the line of action of the
resultant is in the plane of the web, and therefore the resultant passes
through the shear center.
The preceding analysis provides a more complete picture of the
shear stresses in a wide-flange or I-beam because it includes the flanges
(recall that in Chapter 5 we investigated only the shear stresses in the
web). Furthermore, this analysis illustrates the general techniques for
finding shear stresses in beams of thin-walled open cross section. Other
illustrations can be found in the next section, where the shear stresses in
a channel section and an angle section are determined as part of the
process of locating their shear centers.
R 2t w
6.9 SHEAR CENTERS OF THIN-WALLED OPEN SECTIONS
In Sections 6.7 and 6.8 we developed methods for finding the shear
stresses in beams of thin-walled open cross section. Now we will use
those methods to locate the shear centers of several shapes of beams.
Only beams with singly symmetric or unsymmetric cross sections will
be considered, because we already know that the shear center of a
doubly symmetric cross section is located at the centroid.
The procedure for locating the shear center consists of two principal
steps: first, evaluating the shear stresses acting on the cross section when
bending occurs about one of the principal axes, and second, determining
the resultant of those stresses. The shear center is located on the line of
action of the resultant. By considering bending about both principal
axes, we can determine the position of the shear center.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
432
CHAPTER 6 Stresses in Beams (Advanced Topics)
As in Sections 6.7 and 6.8, we will use only centerline dimensions
when deriving formulas and making calculations. This procedure is
satisfactory if the beam is thin walled, that is, if the thickness of the
beam is small compared to the other dimensions of the cross section.
Channel Section
The first beam to be analyzed is a singly symmetric channel section
(Fig. 6-32a). From the general discussion in Section 6.6 we know immediately that the shear center is located on the axis of symmetry (the z
axis). To find the position of the shear center on the z axis, we assume
that the beam is bent about the z axis as the neutral axis, and then we
determine the line of action of the resultant shear force Vy acting parallel
to the y axis. The shear center is located where the line of action of Vy
intersects the z axis. (Note that the origin of axes is at the centroid C, so
that both the y and z axes are principal centroidal axes.)
Based upon the discussions in Section 6.8, we conclude that the
shear stresses in a channel vary linearly in the flanges and parabolically
in the web (Fig. 6-32b). We can find the resultant of those stresses if we
know the maximum stress t1 in the flange, the stress t2 at the top of the
web, and the maximum stress tmax in the web.
To find the stress t1 in the flange, we use Eq. (6-45) with Qz equal
to the first moment of the flange area about the z axis:
bt f h
Qz
2
(a)
in which b is the flange width, tf is the flange thickness, and h is the
height of the beam. (Note again that the dimensions b and h are
FIG. 6-32 Shear center S of a channel
section
t1
y
z
F1
t2
h
—
2
tw
z S
tmax
C
tf
b
y
y
tf
h
—
2
C
e
z
S
F2
t2
C
e
F1
Vy
t1
(a)
(b)
(c)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(d)
SECTION 6.9 Shear Centers of Thin-Walled Open Sections
433
measured along the centerline of the section.) Thus, the stress t1 in the
flange is
VyQz
bhV y
(6-58)
t1
Iz tf
2Iz
where Iz is the moment of inertia about the z axis.
The stress t2 at the top of the web is obtained in a similar manner
but with the thickness equal to the web thickness instead of the flange
thickness:
VyQz
btf hVy
t2
Iztw
2twIz
(6-59)
Also, at the neutral axis the first moment of area is
bt f h
htw h
ht w h
Qz btf
2
2 4
4 2
(b)
Therefore, the maximum stress is
VyQz
btf
h hVy
tmax
tw
4 2Iz
Iz tw
(6-60)
The stresses t1 and t2 in the lower half of the beam are equal to the
corresponding stresses in the upper half (Fig. 6-32b).
The horizontal shear force F1 in either flange (Fig. 6-32c) can be
found from the triangular stress diagrams. Each force is equal to the area
of the stress triangle multiplied by the thickness of the flange:
t1b
hb2tfVy
F1 (tf)
2
4Iz
(6-61)
The vertical force F2 in the web must be equal to the shear force Vy,
since the forces in the flanges have no vertical components. As a check,
we can verify that F2 Vy by considering the parabolic stress diagram
of Fig. 6-32b. The diagram is made up of two parts—a rectangle of area
t2h and a parabolic segment of area
2
(tmax t2)h
3
Thus, the shear force F2, equal to the area of the stress diagram times the
web thickness tw, is
2
F2 t2 htw (tmax t2)htw
3
Substituting the expressions for t2 and tmax (Eqs. 6-59 and 6-60) into
the preceding equation, we obtain
tw h3
bh 2t f Vy
F2
12
2
Iz
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(c)
434
CHAPTER 6 Stresses in Beams (Advanced Topics)
Finally, we note that the expression for the moment of inertia is
twh 3
bh 2t f
Iz
12
2
(6-62)
in which we again base the calculations upon centerline dimensions.
Substituting this expression for Iz into Eq. (c) for F2, we get
F2 Vy
(d)
as expected.
The three forces acting on the cross section (Fig. 6-32c) have a
resultant Vy that intersects the z axis at the shear center S (Fig. 6-32d).
Hence, the moment of the three forces about any point in the cross
section must be equal to the moment of the force Vy about that same
point. This moment relationship provides an equation from which the
position of the shear center may be found.
As an illustration, let us select the shear center itself as the center of
moments. In that case, the moment of the three forces (Fig. 6-32c) is
F1h F2e, where e is the distance from the centerline of the web
to the shear center, and the moment of the resultant force Vy is zero
(Fig. 6-32d). Equating these moments gives
F1h F2e 0
(6-63)
Substituting for F1 from Eq. (6-61) and for F2 from Eq. (d), and then
solving for e, we get
b2h2tf
e
4Iz
(6-64)
When the expression for Iz (Eq. 6-62) is substituted, Eq. (6-64) becomes
3b2tf
e
htw 6btf
(6-65)
Thus, we have determined the position of the shear center of a channel
section.
As explained in Section 6.6, a channel beam will undergo bending
without twisting whenever it is loaded by forces acting through the shear
center. If the loads act parallel to the y axis but through some point other
than the shear center (for example, if the loads act in the plane of the
web), they can be replaced by a statically equivalent force system consisting of loads through the shear center and twisting couples. We then
have a combination of bending and torsion of the beam. If the loads act
along the z axis, we have simple bending about the y axis. If the loads
act in skew directions through the shear center, they can be replaced by
statically equivalent loads acting parallel to the y and z axes.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
435
SECTION 6.9 Shear Centers of Thin-Walled Open Sections
Angle Section
The next shape to be considered is an equal-leg angle section (Fig.
6-33a), in which each leg of the angle has length b and thickness t. The
z axis is an axis of symmetry and the origin of coordinates is at the
centroid C; therefore, both the y and z axes are principal centroidal axes.
To locate the shear center, we will follow the same general procedure as that described for a channel section, because we wish to
determine the distribution of the shear stresses as part of the analysis.
However, as we will see later, the shear center of an angle section can be
determined by inspection.
We begin by assuming that the section is subjected to a shear force
Vy acting parallel to the y axis. Then we use Eq. (6-45) to find the corresponding shear stresses in the legs of the angle. For this purpose we
need the first moment of the cross-sectional area between point a at the
outer edge of the beam (Fig. 6-33b) and section bb located at distance s
from point a. The area is equal to st and its centroidal distance from the
neutral axis is
b s/2
ᎏ
2
Thus, the first moment of the area is
b s/2
Qz st
2
(6-66)
Substituting into Eq. (6-45), we get the following expression for the
shear stress at distance s from the edge of the cross section:
VyQz
Vy s
s
t b
Iz t
2
Iz 2
(6-67)
y
y
y
y
a
b
b
t
b
z
C
C
s
F
tmax
z
C
z
S
C
tmax
F
t
b
Vy
FIG. 6-33 Shear center of an
equal-leg angle section
(a)
(b)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(c)
(d)
436
CHAPTER 6 Stresses in Beams (Advanced Topics)
The moment of inertia Iz can be obtained from Case 24 of Appendix D
with b 45°:
tb3
tb3
Iz 2IBB 2
6
3
(6-68)
Substituting this expression for Iz into Eq. (6-67), we get
3Vy s
s
b
t
2
b3t 2
(6-69)
This equation gives the shear stress at any point along the leg of the
angle. The stress varies quadratically with s, as shown in Fig. 6-33c.
The maximum value of the shear stress occurs at the intersection
of the legs of the angle and is obtained from Eq. (6-69) by substituting
s b:
3Vy
tmax
2bt 2
(6-70)
The shear force F in each leg (Fig. 6-33d) is equal to the area of the
parabolic stress diagram (Fig. 6-33c) times the thickness t of the legs:
Vy
2
F (tmaxb)(t)
3
2
(6-71)
Since the horizontal components of the forces F cancel each other, only
the vertical components remain. Each vertical component is equal to
F/ 2, or Vy /2, and therefore the resultant vertical force is equal to the
shear force Vy, as expected.
Since the resultant force passes through the intersection point of the
lines of action of the two forces F (Fig. 6-33d), we see that the shear
center S is located at the junction of the two legs of the angle.
Sections Consisting of Two Intersecting Narrow Rectangles
In the preceding discussion of an angle section we evaluated the shear
stresses and the forces in the legs in order to illustrate the general
methodology for analyzing thin-walled open sections. However, if our
sole objective had been to locate the shear center, it would not have been
necessary to evaluate the stresses and forces.
Since the shear stresses are parallel to the centerlines of the legs
(Fig. 6-33b), we would have known immediately that their resultants are
two forces F (Fig. 6-33d). The resultant of those two forces is a single
force that passes through their point of intersection. Consequently, this
point is the shear center. Thus, we can determine the location of the
shear center of an equal-leg angle section by a simple line of reasoning
(without making any calculations).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 6.9 Shear Centers of Thin-Walled Open Sections
437
The same line of reasoning is valid for all cross sections consisting
of two thin, intersecting rectangles (Fig. 6-34). In each case the resultants of the shear stresses are forces that intersect at the junction of the
rectangles. Therefore, the shear center S is located at that point.
S
S
FIG. 6-34 Shear centers of sections
consisting of two intersecting narrow
rectangles
S
S
Z-Section
y
y
F1
z
C
C
z
F2
2F1
Vy
F1
(a)
(b)
FIG. 6-35 Shear center of a thin-walled
Z-section
Let us now determine the location of the shear center of a Z-section
having thin walls (Fig. 6-35a). The section has no axes of symmetry but
is symmetric about the centroid C (see Section 12.2 of Chapter 12 for a
discussion of symmetry about a point). The y and z axes are principal
axes through the centroid.
We begin by assuming that a shear force Vy acts parallel to the
y axis and causes bending about the z axis as the neutral axis. Then the
shear stresses in the flanges and web will be directed as shown in
Fig. 6-35a. From symmetry considerations we conclude that the forces
F1 in the two flanges must be equal to each other (Fig. 6-35b). The
resultant of the three forces acting on the cross section (F1 in the flanges
and F2 in the web) must be equal to the shear force Vy. The forces F1
have a resultant 2F1 acting through the centroid and parallel to the
flanges. This force intersects the force F2 at the centroid C, and therefore we conclude that the line of action of the shear force Vy must be
through the centroid.
If the beam is subjected to a shear force Vz parallel to the z axis, we
arrive at a similar conclusion, namely, that the shear force acts through
the centroid. Since the shear center is located at the intersection of the
lines of action of the two shear forces, we conclude that the shear center
of the Z-section coincides with the centroid.
This conclusion applies to any Z-section that is symmetric about the
centroid, that is, any Z-section having identical flanges (same width and
same thickness). Note, however, that the thickness of the web does not
have to be the same as the thickness of the flanges.
The locations of the shear centers of many other structural shapes
are given in the problems at the end of this chapter.*
*The first determination of a shear center was made by S. P. Timoshenko in 1913
(Ref. 6-1).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
438
CHAPTER 6 Stresses in Beams (Advanced Topics)
Example 6-7
A thin-walled semicircular cross section of radius r and thickness t is shown in
Fig. 6-36a. Determine the distance e from the center O of the semicircle to the
shear center S.
y
dA
y
t
a
f
b
2r
—
p
z
t
S
O
C e
b
df
u
S
z
C
O
e
Vy
r
FIG. 6-36 Example 6-7. Shear center of a
thin-walled semicircular section
(a)
(b)
Solution
We know immediately that the shear center is located somewhere on the
axis of symmetry (the z axis). To determine the exact position, we assume that
the beam is bent by a shear force Vy acting parallel to the y axis and producing
bending about the z axis as the neutral axis (Fig. 6-36b).
Shear stresses. The first step is to determine the shear stresses t acting on
the cross section (Fig. 6-36b). We consider a section bb defined by the distance
s measured along the centerline of the cross section from point a. The central
angle subtended between point a and section bb is denoted u. Therefore, the
distance s equals ru, where r is the radius of the centerline and u is measured in
radians.
To evaluate the first moment of the cross-sectional area between point a
and section bb, we identify an element of area dA (shown shaded in the figure)
and integrate as follows:
Qz
u
y dA
(r cos f)(tr df) r 2t sin u
(e)
0
in which f is the angle to the element of area and t is the thickness of the
section. Thus, the shear stress t at section bb is
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 6.9 Shear Centers of Thin-Walled Open Sections
Vy Qz
Vy r 2 sin u
t
Iz t
Iz
439
(f)
Substituting Iz pr 3t/2 (see Case 22 or Case 23 of Appendix D), we get
2Vy sin u
t
prt
(6-72)
When u 0 or u p, this expression gives t 0, as expected. When u p/2,
it gives the maximum shear stress.
Location of shear center. The resultant of the shear stresses must be the
vertical shear force Vy. Therefore, the moment M0 of the shear stresses about the
center O must equal the moment of the force Vy about that same point:
(g)
M0 Vy e
To evaluate M0, we begin by noting that the shear stress t acting on the element
of area dA (Fig. 6-36b) is
2Vy sin f
t
prt
as found from Eq. (6-72). The corresponding force is t dA, and the moment of
this force is
2Vy sin f dA
dM0 r(tdA)
pt
Since dA tr df, this expression becomes
2rVy sin f df
dM0
p
Therefore, the moment produced by the shear stresses is
M0
p
dM0
0
2rVy sin f df
4rVy
p
p
(h)
It follows from Eq. (g) that the distance e to the shear center is
M0
4r
e
Vy
p
1.27r
(6-73)
This result shows that the shear center S is located outside of the semicircular
section.
Note: The distance from the center O of the semicircle to the centroid C of
the cross section (Fig. 6-36a) is 2r/p (from Case 23 of Appendix D), which is
one-half of the distance e. Thus, the centroid is located midway between the
shear center and the center of the semicircle.
The location of the shear center in a more general case of a thin-walled circular section is determined in Problem 6.9-12.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
440
CHAPTER 6 Stresses in Beams (Advanced Topics)
6.10 ELASTOPLASTIC BENDING
★
s
sY
eY
e
O
eY
sY
FIG. 6-37 Idealized stress-strain diagram
for an elastoplastic material
y
In our previous discussions of bending we assumed that the beams were
made of materials that followed Hooke’s law (linearly elastic materials).
Now we will consider the bending of elastoplastic beams when the
material is strained beyond the linear region. When that happens, the
distribution of the stresses is no longer linear but varies according to
the shape of the stress-strain curve.
Elastoplastic materials were discussed earlier when we analyzed
axially loaded bars in Section 2.12. As explained in that section, elastoplastic materials follow Hooke’s law up to the yield stress sY and then
yield plastically under constant stress (see the stress-strain diagram of
Fig. 6-37). From the figure, we see that an elastoplastic material has a
region of linear elasticity between regions of perfect plasticity.
Throughout this section, we will assume that the material has the same
yield stress sY and same yield strain eY in both tension and compression.
Structural steels are excellent examples of elastoplastic materials
because they have sharply defined yield points and undergo large strains
during yielding. Eventually the steels begin to strain harden, and then
the assumption of perfect plasticity is no longer valid. However, strain
hardening provides an increase in strength, and therefore the assumption
of perfect plasticity is on the side of safety.
Yield Moment
z
M
x
FIG. 6-38 Beam of elastoplastic
material subjected to a positive bending
moment M
Let us consider a beam of elastoplastic material subjected to a bending
moment M that causes bending in the xy plane (Fig. 6-38). When the
bending moment is small, the maximum stress in the beam is less than
the yield stress sY, and therefore the beam is in the same condition as a
beam in ordinary elastic bending with a linear stress distribution, as
shown in Fig. 6-39b. Under these conditions, the neutral axis passes
through the centroid of the cross section and the normal stresses are
obtained from the flexure formula (s My/I ). Since the bending
moment is positive, the stresses are compressive above the z axis and
tensile below it.
The preceding conditions exist until the stress in the beam at the
point farthest from the neutral axis reaches the yield stress sY, either in
tension or in compression (Fig. 6-39c). The bending moment in the
beam when the maximum stress just reaches the yield stress, called the
yield moment MY, can be obtained from the flexure formula:
sY I
MY sYS
c
(6-74)
in which c is the distance to the point farthest from the neutral axis and S
is the corresponding section modulus.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
441
SECTION 6.10 Elastoplastic Bending
y
sY
s
y
z
sY
sY
sY
c
O
sY
(a)
(b)
(c)
(d)
sY
(e)
sY
(f)
FIG. 6-39 Stress distributions in a beam
of elastoplastic material
Plastic Moment and Neutral Axis
If we now increase the bending moment above the yield moment MY, the
strains in the beam will continue to increase and the maximum strain
will exceed the yield strain eY. However, because of perfectly plastic
yielding, the maximum stress will remain constant and equal to sY, as
pictured in Fig. 6-39d. Note that the outer regions of the beam have
become fully plastic while a central core (called the elastic core)
remains linearly elastic.
If the z axis is not an axis of symmetry (singly symmetric cross
section), the neutral axis moves away from the centroid when the yield
moment is exceeded. This shift in the location of the neutral axis is not
large, and in the case of the trapezoidal cross section of Fig. 6-39, it is
too small to be seen in the figure. If the cross section is doubly
symmetric, the neutral axis passes through the centroid even when the
yield moment is exceeded.
As the bending moment increases still further, the plastic region
enlarges and moves inward toward the neutral axis until the condition
shown in Fig. 6-39e is reached. At this stage the maximum strain in the
beam (at the farthest distance from the neutral axis) is perhaps 10 or 15
times the yield strain eY and the elastic core has almost disappeared.
Thus, for practical purposes the beam has reached its ultimate momentresisting capacity, and we can idealize the ultimate stress distribution as
consisting of two rectangular parts (Fig. 6-39f ). The bending moment
corresponding to this idealized stress distribution, called the plastic
moment MP , represents the maximum moment that can be sustained by
a beam of elastoplastic material.
To find the plastic moment MP, we begin by locating the neutral
axis of the cross section under fully plastic conditions. For this purpose,
consider the cross section shown in Fig. 6-40a on the next page and let
the z axis be the neutral axis. Every point in the cross section above the
neutral axis is subjected to a compressive stress sY (Fig. 6-40b), and
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
442
CHAPTER 6 Stresses in Beams (Advanced Topics)
y
dA
A1
FIG. 6-40 Location of the neutral axis and
C
c1
y1
z
sY
y
y1
O
c2
y2
A2
y2
T
sY
determination of the plastic moment MP
under fully plastic conditions
(b)
(a)
every point below the neutral axis is subjected to a tensile stress sY. The
resultant compressive force C is equal to sY times the cross-sectional
area A1 above the neutral axis (Fig. 6-40a), and the resultant tensile
force T equals sY times the area A2 below the neutral axis. Since the
resultant force acting on the cross section is zero, it follows that
(a)
T C or A1 A2
Because the total area A of the cross section is equal to A1 A2, we see
that
A
A1 A2
2
(6-75)
Therefore, under fully plastic conditions, the neutral axis divides the
cross section into two equal areas.
As a result, the location of the neutral axis for the plastic moment
MP may be different from its location for linearly elastic bending. For
instance, in the case of a trapezoidal cross section that is narrower at the
top than at the bottom (Fig. 6-40a), the neutral axis for fully plastic
bending is slightly below the neutral axis for linearly elastic bending.
Since the plastic moment MP is the moment resultant of the stresses
acting on the cross section, it can be found by integrating over the crosssectional area A (Fig. 6-40a):
MP
A
s y dA
(sY)y dA
A1
s Y y dA
A2
sY A(y1 y2)
sY ( y1A1) sY (y2 A2)
2
(b)
in which y is the coordinate (positive upward) of the element of area dA
and y1 and y2 are the distances from the neutral axis to the centroids c1
and c2 of areas A1 and A2, respectively.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 6.10 Elastoplastic Bending
443
An easier way to obtain the plastic moment is to evaluate the
moments about the neutral axis of the forces C and T (Fig. 6-40b):
MP C y1 T y2
(c)
Replacing T and C by sYA/2, we get
sY A(y1 y2)
MP
2
(6-76)
which is the same as Eq. (b).
The procedure for obtaining the plastic moment is to divide the
cross section of the beam into two equal areas, locate the centroid of
each half, and then use Eq. (6-76) to calculate MP.
Plastic Modulus and Shape Factor
The expression for the plastic moment can be written in a form similar
to that for the yield moment (Eq. 6-74), as follows:
MP sY Z
(6-77)
A(y1 y2)
Z
2
(6-78)
in which
is the plastic modulus (or the plastic section modulus) for the cross
section. The plastic modulus may be interpreted geometrically as the
first moment (evaluated with respect to the neutral axis) of the area of
the cross section above the neutral axis plus the first moment of the area
below the neutral axis.
The ratio of the plastic moment to the yield moment is solely a function of the shape of the cross section and is called the shape factor f :
MP
Z
f
S
MY
(6-79)
This factor is a measure of the reserve strength of the beam after
yielding first begins. It is highest when most of the material is located
near the neutral axis (for instance, a beam having a solid circular
section), and lowest when most of the material is away from the neutral
axis (for instance, a beam having a wide-flange section). Values of f for
cross sections of rectangular, wide-flange, and circular shapes are given
in the remainder of this section. Other shapes are considered in the problems at the end of the chapter.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
444
CHAPTER 6 Stresses in Beams (Advanced Topics)
Beams of Rectangular Cross Section
y
h
—
2
z
C
b
FIG. 6-41 Rectangular cross section
h
—
2
Now let us determine the properties of a beam of rectangular cross
section (Fig. 6-41) when the material is elastoplastic. The section
modulus is S bh2/6, and therefore the yield moment (Eq. 6-74) is
sY bh2
MY
6
(6-80)
in which b is the width and h is the height of the cross section.
Because the cross section is doubly symmetric, the neutral axis
passes through the centroid even when the beam is loaded into the plastic
range. Consequently, the distances to the centroids of the areas above
and below the neutral axis are
h
y1 y2
4
(d)
Therefore, the plastic modulus (Eq. 6-78) is
A(y1 y2)
bh h
h
bh2
Z
2 4
4
4
2
(6-81)
and the plastic moment (Eq. 6-77) is
sY bh2
MP
4
(6-82)
Finally, the shape factor for a rectangular cross section is
MP
Z
3
f
MY
S
2
(6-83)
which means that the plastic moment for a rectangular beam is 50%
greater than the yield moment.
Next, we consider the stresses in a rectangular beam when the
bending moment M is greater than the yield moment but has not yet
reached the plastic moment. The outer parts of the beam will be at the
yield stress sY and the inner part (the elastic core) will have a linearly
varying stress distribution (Figs. 6-42a and b). The fully plastic zones
are shaded in Fig. 6-42a, and the distances from the neutral axis to the
inner edges of the plastic zones (or the outer edges of the elastic core)
are denoted by e.
The stresses acting on the cross section have the force resultants C1,
C2, T1, and T2, as shown in Fig. 6-42c. The forces C1 and T1 in the
plastic zones are each equal to the yield stress times the cross-sectional
area of the zone:
h
C1 T1 s Y b e
2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(e)
445
SECTION 6.10 Elastoplastic Bending
y
sY
h
—
e
2
h
—
2
z
e
C
e
e
h
—
2
C1
h
—
2
e
C2
4e
h
—
e —
3
2
h
—
2
h
—
e
2
b
T2
T1
sY
(a)
(c)
(b)
FIG. 6-42 Stress distribution in a beam of
rectangular cross section with an elastic
core (MY M MP)
The forces C2 and T2 in the elastic core are each equal to the area of the
stress diagram times the width b of the beam:
sY e
C2 T2 b
2
(f)
Thus, the bending moment (see Fig. 6-42c) is
h
4e
M C1 e C2
2
3
s Y be 4e
h
h
s Y b e e
2
2
2
3
s Y bh2 3
2e2
3
2e2
M
Y
6
2
h2
2
h2
MY
M
MP
(6-84)
Note that when e h/2, the equation gives M MY, and when e 0, it
gives M 3MY /2, which is the plastic moment MP.
Equation (6-84) can be used to determine the bending moment
when the dimensions of the elastic core are known. However, a more
common requirement is to determine the size of the elastic core when
the bending moment is known. Therefore, we solve Eq. (6-84) for e in
terms of the bending moment:
1 3
M
e h
2 2
MY
MY
M
MP
(6-85)
Again we note the limiting conditions: When M MY, the equation
gives e h/2, and when M MP 3MY /2, it gives e 0, which is the
fully plastic condition.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
446
CHAPTER 6 Stresses in Beams (Advanced Topics)
Beams of Wide-Flange Shape
For a doubly symmetric wide-flange beam (Fig. 6-43), the plastic
modulus Z (Eq. 6-78) is calculated by taking the first moment about the
neutral axis of the area of one flange plus the upper half of the web and
then multiplying by 2. The result is
tf
h
h
1 h
Z 2 (bt f) (tw) tf tf
2
2
2
2 2
h
btf (h tf) tw tf
2
2
(g)
With a little rearranging, we can express Z in a more convenient form:
1
Z bh 2 (b tw)(h 2t f)2
4
(6-86)
After calculating the plastic modulus from Eq. (6-86), we can obtain the
plastic moment MP from Eq. (6-76).
Values of Z for commercially available shapes of wide-flange beams
are listed in the AISC manual (Ref. 5-4). The shape factor f for wideflange beams is typically in the range 1.1 to 1.2, depending upon the
proportions of the cross section.
Other shapes of elastoplastic beams can be analyzed in a manner
similar to that described for rectangular and wide-flange beams (see the
following examples and the problems at the end of the chapter).
y
tf
h
—
2
z
C
tw
tf
FIG. 6-43 Cross section of a wide-flange
beam
b
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
h
—
2
SECTION 6.10 Elastoplastic Bending
447
Example 6-8
y
Determine the yield moment, plastic modulus, plastic moment, and shape factor
for a beam of circular cross section with diameter d (Fig. 6-44).
Solution
z
As a preliminary matter, we note that since the cross section is doubly
symmetric, the neutral axis passes through the center of the circle for both
linearly elastic and elastoplastic behavior.
The yield moment MY is found from the flexure formula (Eq. 6-74) as
follows:
d
C
sYI
sY (p d 4/64)
pd 3
MY sY
c
d/2
32
FIG. 6-44 Example 6-8. Cross section of
a circular beam (elastoplastic material)
The plastic modulus Z is found from Eq. (6-78) in which A is the area of the
circle and y and y2 are the distances to the centroids c1 and c2 of the two halves
of the circle (Fig. 6-45). Thus, from Cases 9 and 10 of Appendix D, we get
y
pd 2
A
4
c1
z
(6-87)
y1
C
y2
c2
d
2d
y1 y2
3p
Now substituting into Eq. (6-78) for the plastic modulus, we find
A(y1 y2 )
d3
Z
2
6
FIG. 6-45 Solution to Example 6-8
(6-88)
Therefore, the plastic moment MP (Eq. 6-77) is
sY d 3
MP sY Z
6
(6-89)
and the shape factor f (Eq. 6-79) is
MP
16
f
MY
3p
1.70
(6-90)
This result shows that the maximum bending moment for a circular beam of
elastoplastic material is about 70% larger than the bending moment when the
beam first begins to yield.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
448
CHAPTER 6 Stresses in Beams (Advanced Topics)
Example 6-9
A doubly symmetric hollow box beam (Fig. 6-46) of elastoplastic material
(sY 33 ksi) is subjected to a bending moment M of such magnitude that the
flanges yield but the webs remain linearly elastic.
Determine the magnitude of the moment M if the dimensions of the cross
section are b 5.0 in., b1 4.0 in., h 9.0 in., and h1 7.5 in.
y
z
FIG. 6-46 Example 6-9. Cross section
h1
C
h
b1
of a hollow box beam (elastoplastic
material)
b
Solution
The cross section of the beam and the distribution of the normal stresses
are shown in Figs. 6-47a and b, respectively. From the figure, we see that the
stresses in the webs increase linearly with distance from the neutral axis and the
stresses in the flanges equal the yield stress sY. Therefore, the bending moment
M acting on the cross section consists of two parts:
(1) a moment M1 corresponding to the elastic core, and
(2) a moment M2 produced by the yield stresses sY in the flanges.
The bending moment supplied by the core is found from the flexure
formula (Eq. 6-74) with the section modulus calculated for the webs alone; thus,
(b b1)h21
S1
6
(6-91)
and
2
sY (b b1)h1
M1 sY S1
6
(6-92)
To find the moment supplied by the flanges, we note that the resultant force
F in each flange (Fig. 6-47b) is equal to the yield stress multiplied by the area of
the flange:
h h1
F sY b
2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(h)
449
SECTION 6.10 Elastoplastic Bending
y
sY
F
z
C
h1
h
—1
2
hh
—1
h
2
h
—1
2
sY
b1
FIG. 6-47 Solution to Example 6-9
b
(a)
F
(b)
The force in the top flange is compressive and the force in the bottom flange is
tensile if the bending moment M is positive. Together, the two forces create the
bending moment M2:
h h1
sY b(h2 h12)
M2 F
2
(6-93)
Therefore, the total moment acting on the cross section, after some rearranging,
is
sY
2
M M1 M2 3bh2 (b 2b1)h1
12
(6-94)
Substituting the given numerical values, we obtain
M 1330 k-in.
Note: The yield moment MY and the plastic moment MP for the beam in
this example have the following values (determined in Problem 6.10-13):
MY 1196 k-in.
MP 1485 k-in.
The bending moment M is between these values, as expected.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
450
CHAPTER 6 Stresses in Beams (Advanced Topics)
PROBLEMS CHAPTER 6
Composite Beams
When solving the problems for Section 6.2, assume that the
component parts of the beams are securely bonded by
adhesives or connected by fasteners. Also, be sure to use
the general theory for composite beams described in
Sect. 6.2.
6.2-3 A hollow box beam is constructed with webs of Douglas-fir plywood and flanges of pine as shown in the figure,
which is a cross-sectional view. The plywood is 1 in. thick
and 12 in. wide; the flanges are 2 in. 4 in. (actual size).
The modulus of elasticity for the plywood is 1,600,000 psi
and for the pine is 1,200,000 psi.
If the allowable stresses are 2000 psi for the plywood
and 1700 psi for the pine, find the allowable bending
moment Mmax when the beam is bent about the z axis.
@
À
;
;;
@@
ÀÀ
ÀÀÀ ;;
@@@
;;;
;À@
@@
ÀÀ
6.2-1 A composite beam consisting of fiberglass faces and
a core of particle board has the cross section shown in the
figure. The width of the beam is 2.0 in., the thickness of the
faces is 0.10 in., and the thickness of the core is 0.50 in.
The beam is subjected to a bending moment of 250 lb-in.
acting about the z axis.
Find the maximum bending stresses sface and score in
the faces and the core, respectively, if their respective
moduli of elasticity are 4 106 psi and 1.5 106 psi.
y
2 in.
y
z
0.10 in.
z
1 in.
0.10 in.
2.0 in.
12 in.
2 in.
0.50 in.
C
C
4 in.
1 in.
PROB. 6.2-3
PROB. 6.2-1
6.2-2 A wood beam with cross-sectional dimensions
200 mm 300 mm is reinforced on its sides by steel plates
12 mm thick (see figure). The moduli of elasticity for the
steel and wood are Es 204 GPa and Ew 8.5 GPa,
respectively. Also, the corresponding allowable stresses are
ss 130 MPa and sw 8.0 MPa.
Calculate the maximum permissible bending moment
Mmax when the beam is bent about the z axis.
6.2-4 A round steel tube of outside diameter d and an
aluminum core of diameter d/2 are bonded to form a composite beam as shown in the figure.
Derive a formula for the allowable bending moment M
that can be carried by the beam based upon an allowable
stress ss in the steel. (Assume that the moduli of elasticity
for the steel and aluminum are Es and Ea, respectively.)
y
S
A
z
12 mm
PROB. 6.2-2
C
200 mm
300 mm
d
—
2
d
12 mm
PROB. 6.2-4
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 6 Problems
6.2-5 A simple beam on a 10 ft span supports a uniform
load of intensity 800 lb/ft (see figure). The beam consists of
a wood member (4 in. 11.5 in. in cross section) that is
reinforced by 0.25 in. thick steel plates on top and bottom.
The moduli of elasticity for the steel and wood are Es
30 106 psi and Ew 1.5 106 psi, respectively.
Calculate the maximum bending stresses ss in the steel
plates and sw in the wood member due to the uniform load.
y
6.2-7 The cross section of a sandwich beam consisting of
aluminum alloy faces and a foam core is shown in the
figure. The width b of the beam is 8.0 in., the thickness t of
the faces is 0.25 in., and the height hc of the core is 5.5 in.
(total height h 6.0 in.). The moduli of elasticity are
10.5 106 psi for the aluminum faces and 12,000 psi for
the foam core. A bending moment M 40 k-in. acts about
the z axis.
Determine the maximum stresses in the faces and the
core using (a) the general theory for composite beams, and
(b) the approximate theory for sandwich beams.
ÀÀ
;;
@@
;;
@@
ÀÀ
y
t
0.25 in.
800 lb/ft
11.5 in.
z
z
C
0.25 in.
10 ft
451
hc
C
t
b
4 in.
h
PROBS. 6.2-7 and 6.2-8
PROB. 6.2-5
6.2-8 The cross section of a sandwich beam consisting of
ÀÀ
;;
@@
;;
@@
ÀÀ
6.2-6 A plastic-lined steel pipe has the cross-sectional
shape shown in the figure. The steel pipe has outer diameter
d3 100 mm and inner diameter d2 94 mm. The plastic
liner has inner diameter d1 82 mm. The modulus of elasticity of the steel is 75 times the modulus of the plastic.
Determine the allowable bending moment Mallow if the
allowable stress in the steel is 35 MPa and in the plastic is
600 kPa.
y
z
PROB. 6.2-6
C
d1
d2 d3
fiberglass faces and a lightweight plastic core is shown in
the figure. The width b of the beam is 50 mm, the thickness
t of the faces is 4 mm, and the height hc of the core is
92 mm (total height h 100 mm). The moduli of elasticity
are 75 GPa for the fiberglass and 1.2 GPa for the plastic.
A bending moment M 275 Nm acts about the z axis.
Determine the maximum stresses in the faces and the
core using (a) the general theory for composite beams, and
(b) the approximate theory for sandwich beams.
6.2-9 A bimetallic beam used in a temperature-control
switch consists of strips of aluminum and copper bonded
together as shown in the figure, which is a cross-sectional
view. The width of the beam is 1.0 in., and each strip has a
thickness of 1/16 in.
Under the action of a bending moment M 12 lb-in.
acting about the z axis, what are the maximum stresses sa
and sc in the aluminum and copper, respectively? (Assume
Ea 10.5 106 psi and Ec 16.8 106 psi.)
★
y
1
— in.
16
A
z
O
1.0 in.
PROB. 6.2-9
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
C
1
— in.
16
452
CHAPTER 6 Stresses in Beams (Advanced Topics)
6.2-10 A simply supported composite beam 3 m long carries a uniformly distributed load of intensity q 3.0 kN/m
(see figure). The beam is constructed of a wood member,
100 mm wide by 150 mm deep, reinforced on its lower side
by a steel plate 8 mm thick and 100 mm wide.
Find the maximum bending stresses sw and ss in the
wood and steel, respectively, due to the uniform load if the
moduli of elasticity are Ew 10 GPa for the wood and
Es 210 GPa for the steel.
★
y
6.3-2 A simple beam of span length 3.2 m carries a
uniform load of intensity 48 kN/m. The cross section of the
beam is a hollow box with wood flanges and steel side
plates, as shown in the figure. The wood flanges are 75 mm
by 100 mm in cross section, and the steel plates are
300 mm deep.
What is the required thickness t of the steel plates if
the allowable stresses are 120 MPa for the steel and
6.5 MPa for the wood? (Assume that the moduli of elasticity for the steel and wood are 210 GPa and 10 GPa,
respectively, and disregard the weight of the beam.)
q = 3.0 kN/m
150 mm
z
y
O
8
mm
3m
75 mm
100 mm
z
PROB. 6.2-10
Transformed-Section Method
300 mm
C
75 mm
When solving the problems for Section 6.3, assume that the
component parts of the beams are securely bonded by
adhesives or connected by fasteners. Also, be sure to use
the transformed-section method in the solutions.
6.3-1 A wood beam 8 in. wide and 12 in. deep is reinforced
100 mm
t
t
PROB. 6.3-2
on top and bottom by 0.5 in. thick steel plates (see figure).
Find the allowable bending moment Mmax about the
z axis if the allowable stress in the wood is 1,000 psi and in
the steel is 16,000 psi. (Assume that the ratio of the moduli
of elasticity of steel and wood is 20.)
6.3-3 A simple beam that is 15 ft long supports a
y
0.5 in
z
C
12 in.
0.5 in
8 in.
PROB. 6.3-1
uniform load of intensity q. The beam is constructed of an
S 8 18.4 section (I-beam section) reinforced with wood
beams that are securely fastened to the flanges (see the
cross section shown in the figure). The wood beams are
2 in. deep and 4 in. wide. The modulus of elasticity of the
steel is 20 times that of the wood.
If the allowable stresses in the steel and wood are
12,000 psi and 900 psi, respectively, what is the allowable
load qallow?
(Note: Disregard the weight of the beam, and see Table
E-2 of Appendix E for the dimensions and properties of the
steel beam.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
;@À;@À
CHAPTER 6 Problems
y
y
2 in.
t
S 8 18.4
z
453
C
z
d
h 4t 5d
3d
C
d
2 in.
4 in.
b
PROB. 6.3-3
PROBS. 6.3-5 and 6.3-6
6.3-4 The composite beam shown in the figure is simply
supported and carries a total uniform load of 50 kN/m on a
span length of 4.0 m. The beam is built of a wood member
having cross-sectional dimensions 150 mm 250 mm and
two steel plates of cross-sectional dimensions 50 mm
150 mm.
Determine the maximum stresses ss and sw in the steel
and wood, respectively, if the moduli of elasticity are
Es 209 GPa and Ew 11 GPa. (Disregard the weight of
the beam.)
6.3-6 Consider the preceding problem if the beam has
width b 75 mm, the aluminum strips have thickness
t 3 mm, the plastic segments have heights d 40 mm
and 3d 120 mm, and the total height of the beam is h
212 mm. Also, the moduli of elasticity are Ea 75 GPa
and Ep 3 GPa, respectively.
Determine the maximum stresses sa and sp in the
aluminum and plastic, respectively, due to a bending
moment of 1.0 kNm.
y
50 kN/m
50 mm
z
C
250 mm
50 mm
4.0 m
150 mm
6.3-7 A composite beam constructed of a wood beam
reinforced by a steel plate has the cross-sectional dimensions
shown in the figure. The beam is simply supported with a
span length of 6.0 ft and supports a uniformly distributed
load of intensity q 800 lb/ft.
Calculate the maximum bending stresses ss and sw in
the steel and wood, respectively, due to the uniform load if
Es /E w 20.
PROB. 6.3-4
y
6.3-5 The cross section of a beam made of thin strips of
aluminum separated by a lightweight plastic is shown in the
figure. The beam has width b 3.0 in., the aluminum strips
have thickness t 0.1 in., and the plastic segments have
heights d 1.2 in. and 3d 3.6 in. The total height of the
beam is h 6.4 in.
The moduli of elasticity for the aluminum and plastic are
Ea 11 106 psi and Ep 440 103 psi, respectively.
Determine the maximum stresses sa and sp in the
aluminum and plastic, respectively, due to a bending
moment of 6.0 k-in.
5.5 in.
z
O
0.5 in.
4 in.
PROB. 6.3-7
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
454
CHAPTER 6 Stresses in Beams (Advanced Topics)
6.3-8 The cross section of a composite beam made of
y
aluminum and steel is shown in the figure. The moduli of
elasticity are Ea 75 GPa and Es 200 GPa.
Under the action of a bending moment that produces a
maximum stress of 50 MPa in the aluminum, what is the
maximum stress ss in the steel?
z
O
3 mm
3 mm
B
y
10 mm
Aluminum
40 mm
Steel
z
O 80 mm
30 mm
PROB. 6.3-8
6.3-9 A composite beam is constructed of a wood beam
6 in. wide and 8 in. deep reinforced on the lower side by a
0.5 in. by 6 in. steel plate (see figure). The modulus of elasticity for the wood is E w 1.2 106 psi and for the steel
is Es 30 106 psi.
Find the allowable bending moment Mallow for the
beam if the allowable stress in the wood is sw 1200 psi
and in the steel is ss 10,000 psi.
PROB. 6.3-10
6.3-11 A W 12 50 steel wide-flange beam and a segment
of a 4-inch thick concrete slab (see figure) jointly resist a
positive bending moment of 95 k-ft. The beam and slab are
joined by shear connectors that are welded to the steel
beam. (These connectors resist the horizontal shear at the
contact surface.) The moduli of elasticity of the steel and
the concrete are in the ratio 12 to 1.
Determine the maximum stresses ss and sc in the steel
and concrete, respectively. (Note: See Table E-1 of
Appendix E for the dimensions and properties of the steel
beam.)
ÀÀ
@@
;;
y
30 in.
y
z
8 in.
z
A
0.5 in.
PROB. 6.3-9
O
W 12 50
O
6 in.
PROB. 6.3-11
6.3-12 A wood beam reinforced by an aluminum channel
section is shown in the figure. The beam has a cross section
of dimensions 150 mm by 250 mm, and the channel has a
uniform thickness of 6 mm.
If the allowable stresses in the wood and aluminum are
8.0 MPa and 38 MPa, respectively, and if their moduli of
elasticity are in the ratio 1 to 6, what is the maximum
allowable bending moment for the beam?
★
6.3-10 The cross section of a bimetallic strip is shown in
the figure. Assuming that the moduli of elasticity for metals
A and B are EA 168 GPa and EB 90 GPa, respectively,
determine the smaller of the two section moduli for the
beam. (Recall that section modulus is equal to bending
moment divided by maximum bending stress.) In which
material does the maximum stress occur?
4 in.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 6 Problems
455
y
150 mm
b
y
216 mm
h
250 mm
z
O
C
q
z
a
40 mm
6 mm
162 mm
PROBS. 6.4-2 and 6.4-3
PROB. 6.3-12
Beams with Inclined Loads
When solving the problems for Section 6.4, be sure to draw
a sketch of the cross section showing the orientation of the
neutral axis and the locations of the points where the
stresses are being found.
6.4-1 A beam of rectangular cross section supports an
inclined load P having its line of action along a diagonal of
the cross section (see figure). Show that the neutral axis lies
along the other diagonal.
y
z
C
6.4-3 Solve the preceding problem for the following
data: b 6 in., h 8 in., L 8.0 ft, tan a 1/3, and
q 375 lb/ft.
6.4-4 A simply supported wide-flange beam of span length
L carries a vertical concentrated load P acting through the
centroid C at the midpoint of the span (see figure). The
beam is attached to supports inclined at an angle a to the
horizontal.
Determine the orientation of the neutral axis and calculate the maximum stresses at the outside corners of the
cross section (points A, B, D, and E) due to the load P.
Data for the beam are as follows: W 10 30 section,
L 10.0 ft, P 4.25 k, and a 26.57°. (Note: See Table
E-1 of Appendix E for the dimensions and properties of the
beam.)
h
y
P
E
P
D
b
C
PROB. 6.4-1
6.4-2 A wood beam of rectangular cross section (see
figure) is simply supported on a span of length L. The
longitudinal axis of the beam is horizontal, and the cross
section is tilted at an angle a. The load on the beam is a
vertical uniform load of intensity q acting through the
centroid C.
Determine the orientation of the neutral axis and calculate the maximum tensile stress smax if b 75 mm,
h 150 mm, L 1.5 m, a 30°, and q 6.4 kN/m.
B
z
A
a
PROBS. 6.4-4 and 6.4-5
6.4-5 Solve the preceding problem using the following
data: W 8 21 section, L 76 in., P 4.8 k, and
a 20°.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
456
CHAPTER 6 Stresses in Beams (Advanced Topics)
6.4-6 A wood cantilever beam of rectangular cross section
and length L supports an inclined load P at its free end
(see figure).
Determine the orientation of the neutral axis and calculate the maximum tensile stress smax due to the load P.
Data for the beam are as follows: b 75 mm, h
150 mm, L 1.8 m, P 625 N, and a 36°.
y
6.4-9 A cantilever beam of wide-flange cross section and
length L supports an inclined load P at its free end (see
figure).
Determine the orientation of the neutral axis and calculate the maximum tensile stress smax due to the load P.
Data for the beam are as follows: W 10 45 section,
L 8.0 ft, P 1.5 k, and a 55°. (Note: See Table E-1 of
Appendix E for the dimensions and properties of the beam.)
y
P
h
z
a
a
C
z
P
C
b
PROBS. 6.4-6 and 6.4-7
PROBS. 6.4-9 and 6.4-10
6.4-7 Solve the preceding problem for a cantilever beam
with data as follows: b 4 in., h 8 in., L 7.5 ft,
P 320 lb, and a 45°.
6.4-8 A steel beam of I-section (see figure) is simply
supported at the ends. Two equal and oppositely directed
bending moments M0 act at the ends of the beam, so that
the beam is in pure bending. The moments act in plane mm,
which is oriented at an angle a to the xy plane.
Determine the orientation of the neutral axis and
calculate the maximum tensile stress smax due to the
moments M0.
Data for the beam are as follows: S 8 18.4 section,
M0 30 k-in., and a 30°. (Note: See Table E-2 of
Appendix E for the dimensions and properties of the beam.)
m
6.4-10 Solve the preceding problem using the following
data: W 8 35 section, L 5.0 ft, P 2.4 k, and a 60°.
6.4-11 A cantilever beam of W 12 14 section and
length L 9 ft supports a slightly inclined load P 500 lb
at the free end (see figure).
(a) Plot a graph of the stress sA at point A as a function
of the angle of inclination a.
(b) Plot a graph of the angle b, which locates the
neutral axis nn, as a function of the angle a. (When plotting
the graphs, let a vary from 0 to 10°.) (Note: See Table E-1
of Appendix E for the dimensions and properties of the
beam.)
★
y
y
A
n
b
z
C
z
C
M0
a
PROB. 6.4-8
n
m
P
PROB. 6.4-11
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
a
CHAPTER 6 Problems
457
2
Bending of Unsymmetric Beams
When solving the problems for Section 6.5, be sure to draw
a sketch of the cross section showing the orientation of the
neutral axis and the locations of the points where the
stresses are being found.
6.5-1 A beam of channel section is subjected to a bending
moment M having its vector at an angle u to the z axis (see
figure).
Determine the orientation of the neutral axis and calculate the maximum tensile stress st and maximum
compressive stress sc in the beam.
Use the following data: C 8 11.5 section, M
20 k-in., tan u 1/3. (Note: See Table E-3 of Appendix E
for the dimensions and properties of the channel section.)
M
1
C
1
2
PROBS. 6.5-3 and 6.5-4
6.5-4 Solve the preceding problem for an L 4 4 1/2
angle section with M 6.0 k-in.
6.5-5 A beam of semicircular cross section of radius r is
subjected to a bending moment M having its vector at an
angle u to the z axis (see figure).
Derive formulas for the maximum tensile stress st and
the maximum compressive stress sc in the beam for u 0,
45°, and 90°. (Note: Express the results in the form a M/r 3,
where a is a numerical value.)
★
y
M
z
y
u
C
M
z
u
O
C
PROBS. 6.5-1 and 6.5-2
r
PROB. 6.5-5
6.5-2 Solve the preceding problem for a C 6 13 channel
section with M 5.0 k-in. and u 15°.
6.5-3 An angle section with equal legs is subjected to a
bending moment M having its vector directed along the 1-1
axis, as shown in the figure.
Determine the orientation of the neutral axis and calculate the maximum tensile stress st and maximum
compressive stress sc if the angle is an L 6 6 3/4
section and M 20 k-in. (Note: See Table E-4 of Appendix
E for the dimensions and properties of the angle section.)
Shear Stresses in Wide-Flange Beams
When solving the problems for Section 6.8, assume that
the cross sections are thin-walled. Use centerline dimensions for all calculations and derivations, unless otherwise
specified.
6.8-1 A simple beam of wide-flange cross section supports
a uniform load of intensity q 3.0 k/ft on a span of length
L 10 ft (see figure on the next page). The dimensions
of the cross section are h 10.5 in., b 7 in., and
tf tw 0.4 in.
(a) Calculate the maximum shear stress tmax on cross
section A-A located at distance d 2.0 ft from the end of
the beam.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
458
CHAPTER 6 Stresses in Beams (Advanced Topics)
(b) Calculate the shear stress tB at point B on the cross
section. Point B is located at a distance a 2.0 in. from the
edge of the lower flange.
6.8-4 Solve the preceding problem for the following data:
b 145 mm, h 250 mm, tw 8.0 mm, tf 14.0 mm,
and V 30 kN.
y
b
—
2
q
A
b
—
2
h
—
2
tw
z
A
Shear Centers of Thin-Walled Open Sections
C
tf
B
h
—
2
a
d
L
When locating the shear centers in the problems for Section
6.9, assume that the cross sections are thin-walled and use
centerline dimensions for all calculations and derivations.
6.9-1 Calculate the distance e from the centerline of the
web of a C 12 20.7 channel section to the shear center S
(see figure).
(Note: For purposes of analysis, consider the flanges to
be rectangles with thickness tf equal to the average flange
thickness given in Table E-3, Appendix E.)
PROBS. 6.8-1 and 6.8-2
y
6.8-2 Solve the preceding problem for the following data:
L 3 m, q 40 kN/m, h 260 mm, b 170 mm, tf
12 mm, tw 10 mm, d 0.6 m, and a 60 mm.
6.8-3 A beam of wide-flange shape has the cross section
shown in the figure. The dimensions are b 5.25 in.,
h 7.9 in., tw 0.25 in., and tf 0.4 in. The loads on the
beam produce a shear force V 6.0 k at the cross section
under consideration.
(a) Using centerline dimensions, calculate the maximum shear stress in the web of the beam.
(b) Using the more exact analysis of Section 5.10 in
Chapter 5, calculate the maximum shear stress in the
web of the beam and compare it with the stress obtained in
part (a).
y
tf
z
tw
tf
b
PROBS. 6.8-3 and 6.8-4
h
C
S
z
C
e
PROBS. 6.9-1 and 6.9-2
6.9-2 Calculate the distance e from the centerline of the
web of a C 8 18.75 channel section to the shear center S
(see figure).
(Note: For purposes of analysis, consider the flanges to
be rectangles with thickness tf equal to the average flange
thickness given in Table E-3, Appendix E.)
6.9-3 The cross section of an unbalanced wide-flange beam
is shown in the figure. Derive the following formula for the
distance h1 from the centerline of one flange to the shear
center S:
t2b23h
h1
Also, check the formula for the special cases of a T-beam
(b2 t2 0) and a balanced wide-flange beam (t2 t1 and
b2 b1).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 6 Problems
y
459
y
t2
z
t1
b1
S
b2
C
S
z
C
h1 h2
e
h2
h1
h
PROB. 6.9-3
b
PROB. 6.9-5
6.9-4 The cross section of an unbalanced wide-flange
beam is shown in the figure. Derive the following formula
for the distance e from the centerline of the web to the
shear center S:
3tf (b 22 b 21 )
e
6.9-6 The cross section of a slit circular tube of constant
thickness is shown in the figure. Show that the distance e from
the center of the circle to the shear center S is equal to 2r.
y
Also, check the formula for the special cases of a channel
section (b1 0 and b2 b) and a doubly symmetric beam
(b1 b2 b/2).
y
r
z
S
tf
tw
C
e
h
—
2
PROB. 6.9-6
S
z
C
e
tf
b1
h
—
2
b2
6.9-7 The cross section of a slit square tube of constant
thickness is shown in the figure. Derive the following
formula for the distance e from the corner of the cross
section to the shear center S:
b
e
2 2
y
PROB. 6.9-4
b
6.9-5 The cross section of a channel beam with double
flanges and constant thickness throughout the section is
shown in the figure.
Derive the following formula for the distance e from
the centerline of the web to the shear center S:
3b2(h21 h22)
e
z
S
PROB. 6.9-7
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
e
C
460
CHAPTER 6 Stresses in Beams (Advanced Topics)
6.9-8 The cross section of a slit rectangular tube of
constant thickness is shown in the figure. Derive the
following formula for the distance e from the centerline of
the wall of the tube to the shear center S:
Also, check the formula for the special cases of a channel
section (a 0) and a slit rectangular tube (a h/2).
y
b(2h 3b)
e
2(h 3b)
a
h
—
2
y
S
z
C
e
h
—
2
a
S
z
C
e
h
—
2
h
—
2
b
PROB. 6.9-10
6.9-11 Derive the following formula for the distance e
from the centerline of the wall to the shear center S for the
hat section of constant thickness shown in the figure:
★
b
—
2
b
—
2
3bh2(b 2a) 8ba3
e
PROB. 6.9-8
6.9-9 A U-shaped cross section of constant thickness is
shown in the figure. Derive the following formula for the
distance e from the center of the semicircle to the shear
center S:
2(2r 2 b2 pbr)
e
4b p r
Also, plot a graph showing how the distance e (expressed
as the nondimensional ratio e/r) varies as a function of the
ratio b/r. (Let b/r range from 0 to 2.)
★
Also, check the formula for the special case of a channel
section (a 0).
y
a
h
—
2
y
b
S
z
e
C
h
—
2
r
z
S
O
a
C
b
PROB. 6.9-11
e
PROB. 6.9-9
6.9-12 A cross section in the shape of a circular arc of
constant thickness is shown in the figure. Derive the
following formula for the distance e from the center of the
arc to the shear center S:
2r (sin b b cos b )
e
b sin b cos b
★
6.9-10 Derive the following formula for the distance e
from the centerline of the wall to the shear center S for the
C-section of constant thickness shown in the figure:
★
3bh2(b 2a) 8ba3
e
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 6 Problems
in which b is in radians. Also, plot a graph showing how
the distance e varies as b varies from 0 to p.
461
y
y
r1
z
r
z
C
b
S
O
C
r2
b
PROB. 6.10-2
e
PROB. 6.9-12
Elastoplastic Bending
The problems for Section 6.10 are to be solved using the
assumption that the material is elastoplastic with yield
stress sY.
6.10-3 A cantilever beam of length L 54 in. supports a
uniform load of intensity q (see figure). The beam is made
of steel (sY 36 ksi) and has a rectangular cross section of
width b 4.5 in. and height h 6.0 in.
What load intensity q will produce a fully plastic
condition in the beam?
6.10-1 Determine the shape factor f for a cross section in
the shape of a double trapezoid having the dimensions
shown in the figure.
Also, check your result for the special cases of a
rhombus (b1 0) and a rectangle (b1 b2).
y
q
z
y
b1
z
L = 54 in.
h
—
2
C
C
h = 6.0 in.
b = 4.5 in.
PROB. 6.10-3
h
—
2
b1
b2
PROB. 6.10-1
6.10-2 (a) Determine the shape factor f for a hollow
circular cross section having inner radius r1 and outer
radius r2 (see figure). (b) If the section is very thin, what is
the shape factor?
6.10-4 A steel beam of rectangular cross section is 50 mm
wide and 80 mm high (see figure on the next page). The
yield stress of the steel is 210 MPa.
(a) What percent of the cross-sectional area is occupied by the elastic core if the beam is subjected to a
bending moment of 13.0 kNm acting about the z axis?
(b) What is the magnitude of the bending moment that
will cause 50% of the cross section to yield?
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
462
CHAPTER 6 Stresses in Beams (Advanced Topics)
6.10-10 Solve the preceding problem for a W 10 45
y
wide-flange beam.
6.10-11 A hollow box beam with height h 16 in., width
z
80 mm
C
b 8 in., and constant wall thickness t 0.75 in. is shown
in the figure. The beam is constructed of steel with yield
stress sY 32 ksi.
Determine the yield moment MY, plastic moment MP,
and shape factor f.
y
t
50 mm
PROB. 6.10-4
6.10-5 Calculate the shape factor f for the wide-flange
beam shown in the figure if h 12.0 in., b 6.0 in., tf
0.6 in., and tw 0.4 in.
z
h
C
t
y
b
tf
PROBS. 6.10-11 and 6.10-12
z
h
C
tw
tf
6.10-12 Solve the preceding problem for a box beam with
dimensions h 0.4 m, b 0.2 m, and t 20 mm. The
yield stress of the steel is 230 MPa.
6.10-13 A hollow box beam with height h 9.0 in., inside
b
PROBS. 6.10-5 and 6.10-6
6.10-6 Solve the preceding problem for a wide-flange
beam with h 400 mm, b 150 mm, tf 12 mm, and
tw 8 mm.
height h1 7.5 in., width b 5.0 in., and inside width
b1 4.0 in. is shown in the figure.
Assuming that the beam is constructed of steel with
yield stress sY 33 ksi, calculate the yield moment MY,
plastic moment MP, and shape factor f.
y
6.10-7 Determine the plastic modulus Z and shape factor f
for a W 10 30 wide-flange beam. (Note: Obtain the
cross-sectional dimensions and section modulus of the
beam from Table E-1, Appendix E.)
6.10-8 Solve the preceding problem for a W 8 28 wide-
z
flange beam.
b1
6.10-9 Determine the yield moment MY, plastic moment
MP, and shape factor f for a W 16 77 wide-flange beam if
sY 36 ksi. (Note: Obtain the cross-sectional dimensions
and section modulus of the beam from Table E-1, Appendix
E.)
h1
C
b
PROBS. 6.10-13 through 6.10-16
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
h
CHAPTER 6 Problems
6.10-14 Solve the preceding problem for a box beam with
dimensions h 200 mm, h1 160 mm, b 150 mm, and
b1 130 mm. Assume that the beam is constructed of steel
with yield stress sY 220 MPa.
6.10-19 A wide-flange beam of unbalanced cross section
has the dimensions shown in the figure.
Determine the plastic moment MP if sY 36 ksi.
★
y
6.10-15 The hollow box beam shown in the figure is
10 in.
subjected to a bending moment M of such magnitude that
the flanges yield but the webs remain linearly elastic.
(a) Calculate the magnitude of the moment M if the
dimensions of the cross section are h 14 in., h1
12.5 in., b 8 in., and b1 7 in. Also, the yield stress is
sY 42 ksi.
(b) What percent of the moment M is produced by the
elastic core?
6.10-16 Solve the preceding problem for a box beam with
dimensions h 400 mm, h1 360 mm, b 200 mm, and
b1 160 mm, and with yield stress sY 220 MPa.
6.10-17 A W 12 50 wide-flange beam is subjected to a
bending moment M of such magnitude that the flanges
yield but the web remains linearly elastic.
(a) Calculate the magnitude of the moment M if the
yield stress is sY 36 ksi.
(b) What percent of the moment M is produced by the
elastic core?
0.5 in.
z
O
7 in.
0.5 in.
0.5 in.
5 in.
PROB. 6.10-19
6.10-20 Determine the plastic moment MP for a beam
having the cross section shown in the figure if sY
210 MPa.
★
y
6.10-18 A singly symmetric beam of T-section (see figure)
has cross-sectional dimensions b 140 mm, a 200 mm,
tw 20 mm, and tf 25 mm.
Calculate the plastic modulus Z and the shape factor f.
★
120
mm
z
y
30 mm
PROB. 6.10-20
a
z
150
mm
O
250 mm
tw
463
O
tf
b
PROBS. 6.10-18
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
7
Analysis of Stress
and Strain
7.1 INTRODUCTION
Normal and shear stresses in beams, shafts, and bars can be calculated
from the basic formulas discussed in the preceding chapters. For
instance, the stresses in a beam are given by the flexure and shear formulas
(s My/I and t VQ/Ib), and the stresses in a shaft are given by the
torsion formula (t Tr /IP). The stresses calculated from these formulas
act on cross sections of the members, but larger stresses may occur on
inclined sections. Therefore, we will begin our analysis of stresses and
strains by discussing methods for finding the normal and shear stresses
acting on inclined sections cut through a member.
We have already derived expressions for the normal and shear
stresses acting on inclined sections in both uniaxial stress and pure shear
(see Sections 2.6 and 3.5, respectively). In the case of uniaxial stress, we
found that the maximum shear stresses occur on planes inclined at 45°
to the axis, whereas the maximum normal stresses occur on the cross
sections. In the case of pure shear, we found that the maximum tensile
and compressive stresses occur on 45° planes. In an analogous manner,
the stresses on inclined sections cut through a beam may be larger than
the stresses acting on a cross section. To calculate such stresses, we need
to determine the stresses acting on inclined planes under a more general
stress state known as plane stress (Section 7.2).
In our discussions of plane stress we will use stress elements to
represent the state of stress at a point in a body. Stress elements were
discussed previously in a specialized context (see Sections 2.6 and 3.5),
but now we will use them in a more formalized manner. We will begin
our analysis by considering an element on which the stresses are known,
and then we will derive the transformation equations that give the
stresses acting on the sides of an element oriented in a different
direction.
464
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 7.2 Plane Stress
465
When working with stress elements, we must always keep in mind
that only one intrinsic state of stress exists at a point in a stressed body,
regardless of the orientation of the element being used to portray that
state of stress. When we have two elements with different orientations at
the same point in a body, the stresses acting on the faces of the two
elements are different, but they still represent the same state of stress,
namely, the stress at the point under consideration. This situation is
analogous to the representation of a force vector by its components—
although the components are different when the coordinate axes are
rotated to a new position, the force itself is the same.
Furthermore, we must always keep in mind that stresses are not
vectors. This fact can sometimes be confusing, because we customarily
represent stresses by arrows just as we represent force vectors by arrows.
Although the arrows used to represent stresses have magnitude and
direction, they are not vectors because they do not combine according to
the parallelogram law of addition. Instead, stresses are much more complex quantities than are vectors, and in mathematics they are called
tensors. Other tensor quantities in mechanics are strains and moments
of inertia.
7.2 PLANE STRESS
FIG. 7-1 Elements in plane stress:
(a) three-dimensional view of an
element oriented to the xyz axes,
(b) two-dimensional view of the same
element, and (c) two-dimensional view
of an element oriented to the x1y1z1 axes
The stress conditions that we encountered in earlier chapters when
analyzing bars in tension and compression, shafts in torsion, and beams
in bending are examples of a state of stress called plane stress. To
explain plane stress, we will consider the stress element shown in
Fig. 7-1a. This element is infinitesimal in size and can be sketched either
as a cube or as a rectangular parallelepiped. The xyz axes are parallel to
the edges of the element, and the faces of the element are designated by
the directions of their outward normals, as explained previously in
y
y
sy
sy
sx
txy
O
x
z
sx
sx
O
tx1y1
x
txy
tyx
(a)
O
sy1
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(c)
u
x
ty1x1
sy
(b)
x1
sx1
sx1
tyx
sy
tx1y1
ty1x1
txy
sx
y
u
sy1
tyx
tyx
txy
y1
466
CHAPTER 7 Analysis of Stress and Strain
Section 1.6. For instance, the right-hand face of the element is referred
to as the positive x face, and the left-hand face (hidden from the viewer)
is referred to as the negative x face. Similarly, the top face is the positive
y face, and the front face is the positive z face.
When the material is in plane stress in the xy plane, only the x and y
faces of the element are subjected to stresses, and all stresses act parallel
to the x and y axes, as shown in Fig. 7-la. This stress condition is very
common because it exists at the surface of any stressed body, except at
points where external loads act on the surface. When the element shown
in Fig. 7-1a is located at the free surface of a body, the z axis is normal
to the surface and the z face is in the plane of the surface.
The symbols for the stresses shown in Fig. 7-1a have the following
meanings. A normal stress s has a subscript that identifies the face on
which the stress acts; for instance, the stress sx acts on the x face of the
element and the stress sy acts on the y face of the element. Since the
element is infinitesimal in size, equal normal stresses act on the opposite
faces. The sign convention for normal stresses is the familiar one,
namely, tension is positive and compression is negative.
A shear stress t has two subscripts—the first subscript denotes the
face on which the stress acts, and the second gives the direction on that
face. Thus, the stress txy acts on the x face in the direction of the y axis
(Fig. 7-1a), and the stress tyx acts on the y face in the direction of the x axis.
The sign convention for shear stresses is as follows. A shear stress
is positive when it acts on a positive face of an element in the positive
direction of an axis, and it is negative when it acts on a positive face of
an element in the negative direction of an axis. Therefore, the stresses
txy and tyx shown on the positive x and y faces in Fig. 7-la are positive
shear stresses. Similarly, on a negative face of the element, a shear stress
is positive when it acts in the negative direction of an axis. Hence, the
stresses txy and tyx shown on the negative x and y faces of the element
are also positive.
This sign convention for shear stresses is easy to remember if we
state it as follows: A shear stress is positive when the directions associated with its subscripts are plus-plus or minus-minus; the stress is
negative when the directions are plus-minus or minus-plus.
The preceding sign convention for shear stresses is consistent with
the equilibrium of the element, because we know that shear stresses on
opposite faces of an infinitesimal element must be equal in magnitude
and opposite in direction. Hence, according to our sign convention, a
positive stress txy acts upward on the positive face (Fig. 7-1a) and downward on the negative face. In a similar manner, the stresses tyx acting on
the top and bottom faces of the element are positive although they have
opposite directions.
We also know that shear stresses on perpendicular planes are equal
in magnitude and have directions such that both stresses point toward, or
both point away from, the line of intersection of the faces. Inasmuch as
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 7.2 Plane Stress
467
txy and tyx are positive in the directions shown in the figure, they are
consistent with this observation. Therefore, we note that
y
tx y tyx
sy
tyx
txy
sx
txy
O
sx
x
tyx
z
sy
(a)
y
tyx
txy
sx
O
x
txy
tyx
sy
y1
(b)
y
u
sy1
tx1y1
ty1x1
tx1y1
x1
sx1
O
sy1
(c)
FIG. 7-1 (Repeated)
u
x
ty1x1
sx1
This relationship was derived previously from equilibrium of the
element (see Section 1.6).
For convenience in sketching plane-stress elements, we usually
draw only a two-dimensional view of the element, as shown in Fig. 7-1b.
Although a figure of this kind is adequate for showing all stresses acting
on the element, we must still keep in mind that the element is a solid
body with a thickness perpendicular to the plane of the figure.
Stresses on Inclined Sections
sy
sx
(7-1)
We are now ready to consider the stresses acting on inclined sections,
assuming that the stresses sx, sy, and txy (Figs. 7-1a and b) are known.
To portray the stresses acting on an inclined section, we consider a new
stress element (Fig. 7-1c) that is located at the same point in the material
as the original element (Fig. 7-1b). However, the new element has faces
that are parallel and perpendicular to the inclined direction. Associated
with this new element are axes x1, y1, and z1, such that the z1 axis coincides with the z axis and the x1 y1 axes are rotated counterclockwise
through an angle u with respect to the xy axes.
The normal and shear stresses acting on this new element are
denoted sx1, sy1, tx1y1, and ty1x1, using the same subscript designations
and sign conventions described previously for the stresses acting on the
xy element. The previous conclusions regarding the shear stresses still
apply, so that
tx1y1 ty1x1
(7-2)
From this equation and the equilibrium of the element, we see that the
shear stresses acting on all four side faces of an element in plane stress
are known if we determine the shear stress acting on any one of those
faces.
The stresses acting on the inclined x1y1 element (Fig. 7-1c) can be
expressed in terms of the stresses on the xy element (Fig. 7-1b) by using
equations of equilibrium. For this purpose, we choose a wedge-shaped
stress element (Fig. 7-2a on the next page) having an inclined face that
is the same as the x1 face of the inclined element shown in Fig. 7-1c.
The other two side faces of the wedge are parallel to the x and y axes.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
468
CHAPTER 7 Analysis of Stress and Strain
y
y1
x1
tx1y1
u
sx
u
sx1
O
x
txy
tyx
sy
(a) Stresses
In order to write equations of equilibrium for the wedge, we need to
construct a free-body diagram showing the forces acting on the faces.
Let us denote the area of the left-hand side face (that is, the negative x
face) as A0. Then the normal and shear forces acting on that face are
sx A0 and txy A0, as shown in the free-body diagram of Fig. 7-2b. The
area of the bottom face (or negative y face) is A0 tan u, and the area of
the inclined face (or positive x1 face) is A0 sec u. Thus, the normal and
shear forces acting on these faces have the magnitudes and directions
shown in Fig. 7-2b.
The forces acting on the left-hand and bottom faces can be resolved
into orthogonal components acting in the x1 and y1 directions. Then we
can obtain two equations of equilibrium by summing forces in those
directions. The first equation, obtained by summing forces in the x1
direction, is
sx1 A 0 sec u ⫺ sx A 0 cos u ⫺ tx y A 0 sin u
⫺ sy A 0 tan u sinu ⫺ ty x A 0 tan u cosu 0
y
y1
In the same manner, summation of forces in the y1 direction gives
x1
tx1y1 A0 sec u
u
sx1 A0 sec u
sx A0
u
O
txy A0
tx1y1A 0 sec u sx A 0 sin u tx y A0 cos u
sy A0 tan u cos u ty x A0 tan u sin u 0
x
Using the relationship txy tyx, and also simplifying and rearranging,
we obtain the following two equations:
tyx A0 tan u
sy A0 tan u
sx1 sx cos2 u sy sin2 u 2txy sin u cos u
(7-3a)
tx1y1 (sx sy) sin u cos u tx y (cos2 u sin2 u)
(7-3b)
(b) Forces
FIG. 7-2 Wedge-shaped stress element in
plane stress: (a) stresses acting on the
element, and (b) forces acting on the
element (free-body diagram)
Equations (7-3a) and (7-3b) give the normal and shear stresses acting on
the x1 plane in terms of the angle u and the stresses sx, sy, and txy acting
on the x and y planes.
For the special case when u 0, we note that Eqs. (7-3a) and (7-3b)
give sx1 sx and tx1y1 tx y, as expected. Also, when u 90°, the
equations give sx1 sy and tx1y1 tx y tyx. In the latter case,
since the x1 axis is vertical when u 90°, the stress tx1y1 will be positive
when it acts to the left. However, the stress tyx acts to the right, and
therefore tx1y1 tyx .
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 7.2 Plane Stress
469
Transformation Equations for Plane Stress
Equations (7-3a) and (7-3b) for the stresses on an inclined section can
be expressed in a more convenient form by introducing the following
trigonometric identities (see Appendix C):
1
cos2 u ᎏᎏ(1 cos 2u )
2
1
sin2 u (1 2 cos 2u )
2
1
sin u cos u sin 2u
2
When these substitutions are made, the equations become
sx sy
sx 2 sy
sx1 cos 2u txy sin 2u
2
2
(7-4a)
sx sy
tx1y1 sin 2u txy cos 2u
2
(7-4b)
These equations are usually called the transformation equations for
plane stress because they transform the stress components from one set
of axes to another. However, as explained previously, the intrinsic state
of stress at the point under consideration is the same whether represented by stresses acting on the xy element (Fig. 7-1b) or by stresses
acting on the inclined x1y1 element (Fig. 7-1c).
Since the transformation equations were derived solely from equilibrium of an element, they are applicable to stresses in any kind of
material, whether linear or nonlinear, elastic or inelastic.
An important observation concerning the normal stresses can be
obtained from the transformation equations. As a preliminary matter, we
note that the normal stress sy1 acting on the y1 face of the inclined
element (Fig. 7-1c) can be obtained from Eq. (7-4a) by substituting
u 90° for u. The result is the following equation for sy1:
sx sy
sx 2 sy
sy1 cos 2u txy sin 2u
2
2
(7-5)
Summing the expressions for sx1 and sy1 (Eqs. 7-4a and 7-5), we obtain
the following equation for plane stress:
sx1 sy1 sx sy
(7-6)
This equation shows that the sum of the normal stresses acting on perpendicular faces of plane-stress elements (at a given point in a stressed
body) is constant and independent of the angle u.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
470
CHAPTER 7 Analysis of Stress and Strain
The manner in which the normal and shear stresses vary is shown
in Fig. 7-3, which is a graph of sx1 and tx1y1 versus the angle u (from
Eqs. 7-4a and 7-4b). The graph is plotted for the particular case of
sy 0.2sx and txy 0.8sx. We see from the plot that the stresses vary
continuously as the orientation of the element is changed. At certain
angles, the normal stress reaches a maximum or minimum value; at
other angles, it becomes zero. Similarly, the shear stress has maximum,
minimum, and zero values at certain angles. A detailed investigation of
these maximum and minimum values is made in Section 7.3.
sx1
or
tx1y1
s x1
sx1
sx
txy
tx1y1
0.5sx
tx1 y1
sy
–180°
FIG. 7-3 Graph of normal stress sx1 and
shear stress tx1y1 versus the angle u (for
sy 0.2sx and txy 0.8sx)
0
–90°
90°
u 180°
–0.5sx
–sx
Special Cases of Plane Stress
The general case of plane stress reduces to simpler states of stress under
special conditions. For instance, if all stresses acting on the xy element
(Fig. 7-1b) are zero except for the normal stress sx, then the element is
in uniaxial stress (Fig. 7-4). The corresponding transformation equations, obtained by setting sy and txy equal to zero in Eqs. (7-4a) and
(7-4b), are
y
sx
O
sx
FIG. 7-4 Element in uniaxial stress
x
sx
sx1 ᎏᎏ (1 cos 2u )
2
sx
tx1y1 (sin 2u )
2
(7-7a,b)
These equations agree with the equations derived previously in Section 2.6 (see Eqs. 2-29a and 2-29b), except that now we are using a
more generalized notation for the stresses acting on an inclined plane.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 7.2 Plane Stress
471
y
tyx
txy
O
txy
FIG. 7-5 Element in pure shear
x
tyx
Another special case is pure shear (Fig. 7-5), for which the transformation equations are obtained by substituting sx 0 and sy 0
into Eqs. (7-4a) and (7-4b):
sx1 txy sin 2u
tx1y1 txy cos 2u
(7-8a,b)
Again, these equations correspond to those derived earlier (see Eqs.
3-30a and 3-30b in Section 3.5).
Finally, we note the special case of biaxial stress, in which the xy
element is subjected to normal stresses in both the x and y directions but
without any shear stresses (Fig. 7-6). The equations for biaxial stress are
obtained from Eqs. (7-4a) and (7-4b) simply by dropping the terms
containing txy, as follows:
sx sy
sx 2 sy
sx1 ᎏ cos 2u
2
2
(7-9a)
sx sy
tx1y1 sin 2u
2
(7-9b)
Biaxial stress occurs in many kinds of structures, including thin-walled
pressure vessels (see Sections 8.2 and 8.3).
y
sy
sx
FIG. 7-6 Element in biaxial stress
sx
O
x
sy
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
472
CHAPTER 7 Analysis of Stress and Strain
Example 7-1
An element in plane stress is subjected to stresses sx 16,000 psi, sy
6,000 psi, and txy tyx 4,000 psi, as shown in Fig. 7-7a.
Determine the stresses acting on an element inclined at an angle u 45°.
y
y
sy = 6,000 psi
tyx
sx
O
txy = 4,000 psi
sx = 16,000 psi
x
y1
x1
sx1 = 15,000
psi
sy1 = 7,000 psi
u = 45°
tx1y1 = –5,000 psi
x
O
txy
FIG. 7-7 Example 7-1. (a) Element in
plane stress, and (b) element inclined
at an angle u 45°
tyx
sx1
s y1
sy
(a)
(b)
Solution
Transformation equations. To determine the stresses acting on an inclined
element, we will use the transformation equations (Eqs. 7-4a and 7-4b). From
the given numerical data, we obtain the following values for substitution into
those equations:
sx sy
sx 2 sy
ᎏ 11,000 psi
5,000 psi
txy 4,000 psi
2
2
sin 2u sin 90° 1
cos 2u cos 90° 0
Substituting these values into Eqs. (7-4a) and (7-4b), we get
sx sy
sx 2 sy
sx1 cos 2u txy sin 2u
2
2
11,000 psi (5,000 psi)(0) (4,000 psi)(1) 15,000 psi
sx 2 sy
tx1y1 sin 2u tx y cos 2u
2
(5,000 psi)(1) (4,000 psi)(0) 5,000 psi
In addition, the stress sy1 may be obtained from Eq. (7-5):
sx sy
sx 2 sy
sy1 cos 2u tx y sin 2u
2
2
11,000 psi (5,000 psi)(0) (4,000 psi)(1) 7,000 psi
Stress elements. From these results we can readily obtain the stresses acting
on all sides of an element oriented at u 45°, as shown in Fig. 7-7b. The
arrows show the true directions in which the stresses act. Note especially the
directions of the shear stresses, all of which have the same magnitude. Also,
observe that the sum of the normal stresses remains constant and equal to
22,000 psi (see Eq. 7-6).
Note: The stresses shown in Fig. 7-7b represent the same intrinsic state of
stress as do the stresses shown in Fig. 7-7a. However, the stresses have different
values because the elements on which they act have different orientations.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 7.2 Plane Stress
473
Example 7-2
y
A plane-stress condition exists at a point on the surface of a loaded structure,
where the stresses have the magnitudes and directions shown on the stress element
of Fig. 7-8a.
Determine the stresses acting on an element that is oriented at a clockwise
angle of 15° with respect to the original element.
12 MPa
46 MPa
O
Solution
x
19 MPa
The stresses acting on the original element (Fig. 7-8a) have the following
values:
sx 46 MPa
y1
1.4 MPa
sx sy
17 MPa
2
32.6 MPa
O
x
u = –15°
31.0 MPa
tx y 19 MPa
An element oriented at a clockwise angle of 15° is shown in Fig. 7-8b, where
the x1 axis is at an angle u 15° with respect to the x axis. (As an alternative,
the x1 axis could be placed at a positive angle u 75°.)
Transformation equations. We can readily calculate the stresses on the x1
face of the element oriented at u 15° by using the transformation equations
(Eqs. 7-4a and 7-4b). The calculations proceed as follows:
(a)
y
sy 12 MPa
x1
sin 2u sin (30°) 0.5
sx 2 sy
29 MPa
2
cos 2u cos (30°) 0.8660
Substituting into the transformation equations, we get
sx sy
sx 2 sy
sx1 cos 2u txy sin 2u
2
2
17 MPa (29 MPa)(0.8660) (19 MPa)(0.5)
(b)
FIG. 7-8 Example 7-2. (a) Element in
plane stress, and (b) element inclined at
an angle u 15°
32.6 MPa
sx 2 sy
tx1 y 1 sin 2u txycos 2u
2
(29 MPa)(0.5) (19 MPa)(0.8660)
31.0 MPa
Also, the normal stress acting on the y1 face (Eq. 7-5) is
sx sy
sx 2 sy
sy1 cos 2u txy sin 2u
2
2
17 MPa (29 MPa)(0.8660) (19 MPa)(0.5)
1.4 MPa
This stress can be verified by substituting u 75° into Eq. (7-4a). As a further
check on the results, we note that sx1 sy1 sx sy.
The stresses acting on the inclined element are shown in Fig. 7-8b, where
the arrows indicate the true directions of the stresses. Again we note that both
stress elements shown in Fig. 7-8 represent the same state of stress.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
474
CHAPTER 7 Analysis of Stress and Strain
7.3 PRINCIPAL STRESSES AND MAXIMUM SHEAR STRESSES
The transformation equations for plane stress show that the normal
stresses sx1 and the shear stresses tx1y1 vary continuously as the axes are
rotated through the angle u. This variation is pictured in Fig. 7-3 for a
particular combination of stresses. From the figure, we see that both the
normal and shear stresses reach maximum and minimum values at 90°
intervals. Not surprisingly, these maximum and minimum values are
usually needed for design purposes. For instance, fatigue failures of
structures such as machines and aircraft are often associated with the
maximum stresses, and hence their magnitudes and orientations should
be determined as part of the design process.
Principal Stresses
The maximum and minimum normal stresses, called the principal
stresses, can be found from the transformation equation for the normal
stress sx1 (Eq. 7-4a). By taking the derivative of sx1 with respect to u
and setting it equal to zero, we obtain an equation from which we can
find the values of u at which sx1 is a maximum or a minimum. The
equation for the derivative is
dsx1
ᎏ
(sx sy) sin 2u 2txy cos 2u 0
du
(7-10)
from which we get
2txy
tan 2up
sx sy
(7-11)
The subscript p indicates that the angle up defines the orientation of the
principal planes, that is, the planes on which the principal stresses act.
Two values of the angle 2up in the range from 0 to 360° can be
obtained from Eq. (7-11). These values differ by 180°, with one value
between 0 and 180° and the other between 180° and 360°. Therefore, the
angle up has two values that differ by 90°, one value between 0 and 90°
and the other between 90° and 180°. The two values of up are known as
the principal angles. For one of these angles, the normal stress sx1 is a
maximum principal stress; for the other, it is a minimum principal stress.
Because the principal angles differ by 90°, we see that the principal
stresses occur on mutually perpendicular planes.
The principal stresses can be calculated by substituting each of the
two values of up into the first stress-transformation equation (Eq. 7-4a)
and solving for sx1. By determining the principal stresses in this manner,
we not only obtain the values of the principal stresses but we also learn
which principal stress is associated with which principal angle.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 7.3 Principal Stresses and Maximum Shear Stresses
2xy
sy + t
–
s x 2
R=
2
txy
We can also obtain general formulas for the principal stresses. To do
so, refer to the right triangle in Fig. 7-9, which is constructed from
Eq. (7-11). Note that the hypotenuse of the triangle, obtained from the
Pythagorean theorem, is
R
2up
sx – sy
2
FIG. 7-9 Geometric representation of
Eq. (7-11)
475
ᎏᎏ
t
冣 莦
冪冢莦
2莦
sx ⫺sy
2
2
xy
(7-12)
The quantity R is always a positive number and, like the other two sides
of the triangle, has units of stress. From the triangle we obtain two additional relations:
sx sy
cos 2up
2R
txy
sin 2up
R
(7-13a,b)
Now we substitute these expressions for cos 2up and sin 2up into
Eq. (7-4a) and obtain the algebraically larger of the two principal
stresses, denoted by s1:
sx sy
sx sy
s1 sx1 cos 2up txy sin 2up
2
2
sx sy
sx sy sx sy
txy
txy
2
2
2R
R
冢
冣
冢 冣
After substituting for R from Eq. (7-12) and performing some algebraic
manipulations, we obtain
sx sy
s 1
2
t
冢2
冣 莦
冪莦
莦
sx sy
2
2
xy
(7-14)
The smaller of the principal stresses, denoted by s2, may be found from
the condition that the sum of the normal stresses on perpendicular planes
is constant (see Eq. 7-6):
s 1 s 2 sx sy
(7-15)
Substituting the expression for s 1 into Eq. (7-15) and solving for s 2, we
get
s 2 sx sy s1
sx sy
2
t
冢2
冣 莦
冪莦
莦
sx sy
2
2
xy
(7-16)
This equation has the same form as the equation for s1 but differs by the
presence of the minus sign before the square root.
The preceding formulas for s1 and s2 can be combined into a single
formula for the principal stresses:
sx sy
s1,2
2
t
冣 莦
冪冢莦
2莦
sx sy
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
2
2
xy
(7-17)
476
CHAPTER 7 Analysis of Stress and Strain
The plus sign gives the algebraically larger principal stress and the
minus sign gives the algebraically smaller principal stress.
Principal Angles
Let us now denote the two angles defining the principal planes as up1
and u p 2, corresponding to the principal stresses s1 and s2, respectively.
Both angles can be determined from the equation for tan 2up (Eq. 7-11).
However, we cannot tell from that equation which angle is up1 and
which is u p 2. A simple procedure for making this determination is to
take one of the values and substitute it into the equation for sx 1
(Eq. 7-4a). The resulting value of sx 1 will be recognized as either s1 or
s2 (assuming we have already found s1 and s2 from Eq. 7-17), thus
correlating the two principal angles with the two principal stresses.
Another method for correlating the principal angles and principal
stresses is to use Eqs. (7-13a) and (7-13b) to find up, since the only
angle that satisfies both of those equations is u p1. Thus, we can rewrite
those equations as follows:
sx 2 sy
cos 2up1 ᎏᎏ
2R
y
sx
sx
O
x
txy
sin 2u p1 ᎏᎏ
R
(7-18a,b)
Only one angle exists between 0 and 360° that satisfies both of these
equations. Thus, the value of u p1 can be determined uniquely from
Eqs. (7-18a) and (7-18b). The angle up 2, corresponding to s 2, defines a
plane that is perpendicular to the plane defined by up1. Therefore, up2
can be taken as 90° larger or 90° smaller than up1.
Shear Stresses on the Principal Planes
(a)
y
sy
sx
sx
O
x
An important characteristic of the principal planes can be obtained from
the transformation equation for the shear stresses (Eq. 7-4b). If we set
the shear stress t x1y1 equal to zero, we get an equation that is the same as
Eq. (7-10). Therefore, if we solve that equation for the angle 2u, we get
the same expression for tan 2u as before (Eq. 7-11). In other words, the
angles to the planes of zero shear stress are the same as the angles to the
principal planes.
Thus, we can make the following important observation: The shear
stresses are zero on the principal planes.
Special Cases
sy
(b)
FIG. 7-10 Elements in uniaxial and
biaxial stress
The principal planes for elements in uniaxial stress and biaxial stress
are the x and y planes themselves (Fig. 7-10), because tan 2up 0 (see
Eq. 7-11) and the two values of up are 0 and 90°. We also know that the
x and y planes are the principal planes from the fact that the shear
stresses are zero on those planes.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 7.3 Principal Stresses and Maximum Shear Stresses
y
For an element in pure shear (Fig. 7-11a), the principal planes are
oriented at 45° to the x axis (Fig. 7-11b), because tan 2up is infinite and
the two values of up are 45° and 135°. If txy is positive, the principal
stresses are s 1 txy and s 2 txy (see Section 3.5 for a discussion of
pure shear).
tyx
txy
txy
O
x
The Third Principal Stress
tyx
(a)
y
s2 = –txy
477
s1 = txy
up = 45°
O
x
(b)
FIG. 7-11 (a) Element in pure shear, and
(b) principal stresses.
The preceding discussion of principal stresses refers only to rotation of
axes in the xy plane, that is, rotation about the z axis (Fig. 7-12a). Therefore, the two principal stresses determined from Eq. (7-17) are called the
in-plane principal stresses. However, we must not overlook the fact
that the stress element is actually three-dimensional and has three (not
two) principal stresses acting on three mutually perpendicular planes.
By making a more complete three-dimensional analysis, it can be
shown that the three principal planes for a plane-stress element are the
two principal planes already described plus the z face of the element.
These principal planes are shown in Fig. 7-12b, where a stress element
has been oriented at the principal angle up1, which corresponds to the
principal stress s1. The principal stresses s1 and s 2 are given by Eq.
(7-17), and the third principal stress (s 3) equals zero.
By definition, s1 is algebraically larger than s 2, but s 3 may be
algebraically larger than, between, or smaller than s 1 and s 2. Of course,
it is also possible for some or all of the principal stresses to be equal.
Note again that there are no shear stresses on any of the principal
planes.*
y
y
y1
sy
s2
tyx
sx
txy
O
txy
s1
sx
O
x
z
FIG. 7-12 Elements in plane stress:
(a) original element, and (b) element
oriented to the three principal planes and
three principal stresses
tyx
sy
x1
u p1
x
s3 = 0
z, z1
(a)
(b)
*The determination of principal stresses is an example of a type of mathematical analysis
known as eigenvalue analysis, which is described in books on matrix algebra. The stresstransformation equations and the concept of principal stresses are due to the French mathematicians A. L. Cauchy (1789–1857) and Barré de Saint-Venant (1797–1886) and to the
Scottish scientist and engineer W. J. M. Rankine (1820–1872); see Refs. 7-1, 7-2, and
7-3, respectively.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
478
CHAPTER 7 Analysis of Stress and Strain
Maximum Shear Stresses
Having found the principal stresses and their directions for an element in
plane stress, we now consider the determination of the maximum shear
stresses and the planes on which they act. The shear stresses tx1y1 acting
on inclined planes are given by the second transformation equation
(Eq. 7-4b). Taking the derivative of tx1y1 with respect to u and setting it
equal to zero, we obtain
dtx1y1
ᎏᎏ
(s x sy) cos 2u 2tx y sin 2u 0
du
(7-19)
sx sy
tan 2us
2txy
(7-20)
from which
The subscript s indicates that the angle us defines the orientation of the
planes of maximum positive and negative shear stresses.
Equation (7-20) yields one value of us between 0 and 90° and
another between 90° and 180°. Furthermore, these two values differ by
90°, and therefore the maximum shear stresses occur on perpendicular
planes. Because shear stresses on perpendicular planes are equal in
absolute value, the maximum positive and negative shear stresses differ
only in sign.
Comparing Eq. (7-20) for us with Eq. (7-11) for up shows that
1
tan 2us cot 2up
tan 2up
(7-21)
From this equation we can obtain a relationship between the angles us
and up. First, we rewrite the preceding equation in the form
sin 2us
cos 2up
0
cos 2us
sin 2up
Multiplying by the terms in the denominator, we get
sin 2us sin 2up cos 2us cos 2up 0
which is equivalent to the following expression (see Appendix C):
cos (2us 2up) 0
Therefore,
2us 2up 90°
and
us up 45°
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(7-22)
SECTION 7.3 Principal Stresses and Maximum Shear Stresses
479
This equation shows that the planes of maximum shear stress occur at
45° to the principal planes.
The plane of the maximum positive shear stress tmax is defined by
the angle us1, for which the following equations apply:
txy
cos 2us1 ᎏᎏ
R
sx ⫺ sy
sin 2us1 ⫺ᎏᎏ
2R
(7-23a,b)
in which R is given by Eq. (7-12). Also, the angle us1 is related to the
angle up1 (see Eqs. 7-18a and 7-18b) as follows:
us1 up1 ⫺ 45°
(7-24)
The corresponding maximum shear stress is obtained by substituting the
expressions for cos 2us1 and sin 2us1 into the second transformation
equation (Eq. 7-4b), yielding
tmax
t
冣 莦
冪莦
冢2
莦
sx sy
2
2
xy
(7-25)
The maximum negative shear stress tmin has the same magnitude but
opposite sign.
Another expression for the maximum shear stress can be obtained
from the principal stresses s1 and s 2, both of which are given by
Eq. (7-17). Subtracting the expression for s 2 from that for s1, and then
comparing with Eq. (7-25), we see that
s1 s2
tmax
2
(7-26)
Thus, the maximum shear stress is equal to one-half the difference of the
principal stresses.
The planes of maximum shear stress also contain normal stresses.
The normal stress acting on the planes of maximum positive shear
stress can be determined by substituting the expressions for the angle us1
(Eqs. 7-23a and 7-23b) into the equation for sx1 (Eq. 7-4a). The resulting stress is equal to the average of the normal stresses on the x and y
planes:
sx sy
saver ᎏ
2
(7-27)
This same normal stress acts on the planes of maximum negative shear
stress.
In the particular cases of uniaxial stress and biaxial stress
(Fig. 7-10), the planes of maximum shear stress occur at 45° to the x and
y axes. In the case of pure shear (Fig. 7-11), the maximum shear
stresses occur on the x and y planes.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
480
CHAPTER 7 Analysis of Stress and Strain
In-Plane and Out-of-Plane Shear Stresses
The preceding analysis of shear stresses has dealt only with in-plane
shear stresses, that is, stresses acting in the xy plane. To obtain the maximum in-plane shear stresses (Eqs. 7-25 and 7-26), we considered
elements that were obtained by rotating the xyz axes about the z axis,
which is a principal axis (Fig. 7-12a). We found that the maximum shear
stresses occur on planes at 45° to the principal planes. The principal
planes for the element of Fig. 7-12a are shown in Fig. 7-12b, where s1
and s 2 are the principal stresses. Therefore, the maximum in-plane shear
stresses are found on an element obtained by rotating the x1y1z1 axes
(Fig. 7-12b) about the z1 axis through an angle of 45°. These stresses are
given by Eq. (7-25) or Eq. (7-26).
We can also obtain maximum shear stresses by 45° rotations about
the other two principal axes (the x1 and y1 axes in Fig. 7-12b). As a
result, we obtain three sets of maximum positive and maximum
negative shear stresses (compare with Eq. 7-26):
s2
s1
s1 s2
(tmax)x1 (tmax)y1 (tmax)z1 (7-28a,b,c)
2
2
2
in which the subscripts indicate the principal axes about which the 45°
rotations take place. The stresses obtained by rotations about the x1 and
y1 axes are called out-of-plane shear stresses.
The algebraic values of s1 and s2 determine which of the preceding
expressions gives the numerically largest shear stress. If s 1 and s 2 have
the same sign, then one of the first two expressions is numerically
largest; if they have opposite signs, the last expression is largest.
y
y
y1
s2
sy
tyx
sx
txy
O
txy
s1
sx
O
x
z
tyx
sy
FIG. 7-12 (Repeated)
s3 = 0
z, z1
(a)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(b)
x1
up1
x
481
SECTION 7.3 Principal Stresses and Maximum Shear Stresses
Example 7-3
An element in plane stress is subjected to stresses sx 12,300 psi, sy
⫺4,200 psi, and txy 4,700 psi, as shown in Fig. 7-13a.
(a) Determine the principal stresses and show them on a sketch of a properly oriented element.
(b) Determine the maximum shear stresses and show them on a sketch of a
properly oriented element. (Consider only the in-plane stresses.)
y
y
4,050 psi
up2 = 75.2°
4,200 psi
12,300 psi
x
4,050 psi
s2 = –5,440 psi
s1 = 13,540 psi
O
y
x
O
O
4,700 psi
us2 = 30.2°
x
9,490 psi
(a)
(b)
FIG. 7-13 Example 7-3. (a) Element in
plane stress, (b) principal stresses, and
(c) maximum shear stresses
(c)
Solution
(a) Principal stresses. The principal angles up that locate the principal
planes can be obtained from Eq. (7-11):
2txy
2( 4,700 psi)
tan 2up 0.5697
12,300 psi (4,200 psi)
sx sy
Solving for the angles, we get the following two sets of values:
2up 150.3° and up 75.2°
2up 330.3° and up 165.2°
The principal stresses may be obtained by substituting the two values of
2u p into the transformation equation for sx1 (Eq. 7-4a). As a preliminary calculation, we determine the following quantities:
sx sy
12,300 psi 4,200 psi
4,050 psi
2
2
sx sy
12,300 psi 4,200 psi
8,250 psi
2
2
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
482
CHAPTER 7 Analysis of Stress and Strain
Now we substitute the first value of 2up into Eq. (7-4a) and obtain
sx sy
sx sy
sx1 ᎏ cos 2u txy sin 2u
2
2
4,050 psi (8,250 psi)(cos 150.3°) (4,700 psi)(sin 150.3°)
5,440 psi
In a similar manner, we substitute the second value of 2up and obtain sx1
13,540 psi. Thus, the principal stresses and their corresponding principal angles
are
y
s1 13,540 psi and up1 165.2°
up2 = 75.2°
s2 5,440 psi and up2 75.2°
s2 = – 5,440 psi
s1 = 13,540 psi
O
x
Note that u p1 and u p2 differ by 90° and that s1 s2 sx sy .
The principal stresses are shown on a properly oriented element in Fig.
7-13b. Of course, no shear stresses act on the principal planes.
Alternative solution for the principal stresses. The principal stresses may
also be calculated directly from Eq. (7-17):
sx sy
s1, 2
2
t
冢2
冣 莦
冪莦
莦
sx sy
2
2
xy
4,050 psi 兹(8,250
苶苶
psi)2 苶
(4,7苶
00 psi)2苶
(b)
FIG. 7-13b (Repeated)
s1,2 4,050 psi 9,490 psi
Therefore,
s 1 13,540 psi
s 2 5,440 psi
The angle up1 to the plane on which s1 acts is obtained from Eqs. (7-18a)
and (7-18b):
sx sy
8,250 psi
cos 2u p1 0.869
2R
9,490 psi
txy
4,700 psi
sin 2up1 0.495
R
9,490 psi
in which R is given by Eq. (7-12) and is equal to the square-root term in the
preceding calculation for the principal stresses s1 and s2.
The only angle between 0 and 360° having the specified sine and cosine
is 2up1 330.3°; hence, up1 165.2°. This angle is associated with the
algebraically larger principal stress s1 13,540 psi. The other angle is 90°
larger or smaller than up1; hence, up2 75.2°. This angle corresponds to the
smaller principal stress s2 5,440 psi. Note that these results for the principal
stresses and principal angles agree with those found previously.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 7.4 Mohr’s Circle for Plane Stress
483
(b) Maximum shear stresses. The maximum in-plane shear stresses are
given by Eq. (7-25):
tmax
t
冢ᎏ2ᎏ
冣 莦
冪莦
莦
sx ⫺ sy
2
2
xy
兹(8,250
苶苶
psi)2 苶
(4,7苶
00 psi)2苶 9,490 psi
The angle us1 to the plane having the maximum positive shear stress is calculated from Eq. (7-24):
y
us1 up1 – 45° 165.2° 45° 120.2°
4,050 psi
4,050 psi
O
us2 = 30.2°
It follows that the maximum negative shear stress acts on the plane for which
us2 120.2° 90° 30.2°.
The normal stresses acting on the planes of maximum shear stresses are
calculated from Eq. (7-27):
sx sy
saver 4,050 psi
2
x
9,490 psi
(c)
FIG. 7-13c (Repeated)
Finally, the maximum shear stresses and associated normal stresses are shown
on the stress element of Fig. 7-13c.
As an alternative approach to finding the maximum shear stresses, we can
use Eq. (7-20) to determine the two values of the angles us, and then we can use
the second transformation equation (Eq. 7-4b) to obtain the corresponding shear
stresses.
7.4 MOHR’S CIRCLE FOR PLANE STRESS
The transformation equations for plane stress can be represented in
graphical form by a plot known as Mohr’s circle. This graphical representation is extremely useful because it enables you to visualize the
relationships between the normal and shear stresses acting on various
inclined planes at a point in a stressed body. It also provides a means for
calculating principal stresses, maximum shear stresses, and stresses on
inclined planes. Furthermore, Mohr’s circle is valid not only for stresses
but also for other quantities of a similar mathematical nature, including
strains and moments of inertia.*
*Mohr’s circle is named after the famous German civil engineer Otto Christian Mohr
(1835–1918), who developed the circle in 1882 (Ref. 7-4).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
484
CHAPTER 7 Analysis of Stress and Strain
Equations of Mohr’s Circle
The equations of Mohr’s circle can be derived from the transformation
equations for plane stress (Eqs. 7-4a and 7-4b). The two equations are
repeated here, but with a slight rearrangement of the first equation:
sx sy
tx1y1 sin 2u txy cos 2u
2
(7-29b)
冢s
R
sx1
2
2
冣
sx sy
x1
C
tx1y1
(7-29a)
From analytic geometry, we might recognize that these two equations
are the equations of a circle in parametric form. The angle 2u is the
parameter and the stresses sx1 and tx1y1 are the coordinates. However, it
is not necessary to recognize the nature of the equations at this stage—if
we eliminate the parameter, the significance of the equations will
become apparent.
To eliminate the parameter 2u, we square both sides of each equation and then add the two equations. The equation that results is
2u
O
sx sy
sx sy
sx1 ⫺ ᎏ cos 2u tx y sin 2u
2
2
(a)
t x2y
(7-30)
R
t
冣 莦
冪冢莦
2莦
sx sy
2
2
xy
(7-31a,b)
Equation (7-30) now becomes
2
(sx1 saver) t x21y1 R
tx1y1
2
(7-32)
which is the equation of a circle in standard algebraic form. The coordinates are sx1 and tx1y1, the radius is R, and the center of the circle has
coordinates sx1 saver and tx1y1 0.
2u
C
O
2
冣
This equation can be written in simpler form by using the following
notation from Section 7.3 (see Eqs. 7-27 and 7-12, respectively):
sx sy
saver
2
saver
冢
sx sy
t x21y1
2
R
sx1
Two Forms of Mohr’s Circle
saver
(b)
FIG. 7-14 Two forms of Mohr’s circle:
(a) tx1y1 is positive downward and the
angle 2u is positive counterclockwise,
and (b) tx1y1 is positive upward and the
angle 2u is positive clockwise. (Note:
The first form is used in this book.)
Mohr’s circle can be plotted from Eqs. (7-29) and (7-32) in either of two
forms. In the first form of Mohr’s circle, we plot the normal stress sx1
positive to the right and the shear stress tx1y1 positive downward, as
shown in Fig. 7-14a. The advantage of plotting shear stresses positive
downward is that the angle 2u on Mohr’s circle will be positive when
counterclockwise, which agrees with the positive direction of 2u in the
derivation of the transformation equations (see Figs. 7-1 and 7-2).
In the second form of Mohr’s circle, tx1y1 is plotted positive upward
but the angle 2u is now positive clockwise (Fig. 7-14b), which is opposite to its usual positive direction.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 7.4 Mohr’s Circle for Plane Stress
485
Both forms of Mohr’s circle are mathematically correct, and either
one can be used. However, it is easier to visualize the orientation of the
stress element if the positive direction of the angle 2u is the same in
Mohr’s circle as it is for the element itself. Furthermore, a counterclockwise rotation agrees with the customary right-hand rule for rotation.
Therefore, we will choose the first form of Mohr’s circle (Fig.
7-14a) in which positive shear stress is plotted downward and a positive
angle 2u is plotted counterclockwise.
Construction of Mohr’s Circle
Mohr’s circle can be constructed in a variety of ways, depending upon
which stresses are known and which are to be found. For our immediate
purpose, which is to show the basic properties of the circle, let us
assume that we know the stresses sx , sy, and txy acting on the x and y
planes of an element in plane stress (Fig. 7-15a). As we will see, this
information is sufficient to construct the circle. Then, with the circle
drawn, we can determine the stresses sx1, sy1, and tx1y1 acting on an
inclined element (Fig. 7-15b). We can also obtain the principal stresses
and maximum shear stresses from the circle.
y
sy
B
txy
sx
O
x
s1
A
sy
B(u = 90°)
S2
(a)
y
y1
–txy
D'
P1
sy1
D'
tx1y1
x1
sx1
O
u
x
C
P2
O
D
S1
sx + sy
saver =
2
sx
for plane stress
A(u = 0)
sx1
FIG. 7-15 Construction of Mohr’s circle
tx1y1
sx1
2up1
2u
s2
(b)
b
(c)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
tx1y1
txy
D(u = u)
sx – sy
2
486
CHAPTER 7 Analysis of Stress and Strain
y
sy
B
y
y1
sy1
D'
txy
O
tx1y1
x1
sx1
sx
O
x
u
x
D
A
(a)
(b)
s1
sy
B(u = 90°)
S2
–txy
D'
P1
C
P2
O
b
sx1
2up1
2u
s2
S1
sx + sy
saver =
2
sx
A(u = 0)
tx1y1
txy
D(u = u)
sx – sy
2
sx1
FIG. 7-15 (Repeated)
tx1y1
(c)
With sx , sy , and tx y known, the procedure for constructing
Mohr’s circle is as follows (see Fig. 7-15c):
1. Draw a set of coordinate axes with sx1 as abscissa (positive to the
right) and tx1y1 as ordinate (positive downward).
2. Locate the center C of the circle at the point having coordinates
sx1 saver and tx1y1 0 (see Eqs. 7-31a and 7-32).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 7.4 Mohr’s Circle for Plane Stress
487
3. Locate point A, representing the stress conditions on the x face
of the element shown in Fig. 7-15a, by plotting its coordinates
sx1 sx and tx1y1 tx y. Note that point A on the circle corresponds to u 0. Also, note that the x face of the element (Fig.
7-15a) is labeled “A” to show its correspondence with point A on the
circle.
4. Locate point B, representing the stress conditions on the y face
of the element shown in Fig. 7-15a, by plotting its coordinates
sx1 sy and tx1y1 txy. Note that point B on the circle corresponds to u 90°. In addition, the y face of the element
(Fig. 7-15a) is labeled “B” to show its correspondence with point B
on the circle.
5. Draw a line from point A to point B. This line is a diameter of the
circle and passes through the center C. Points A and B, representing
the stresses on planes at 90° to each other (Fig. 7-15a), are at opposite ends of the diameter (and therefore are 180° apart on the circle).
6. Using point C as the center, draw Mohr’s circle through points A
and B. The circle drawn in this manner has radius R (Eq. 7-31b), as
shown in the next paragraph.
Now that we have drawn the circle, we can verify by geometry that
lines CA and CB are radii and have lengths equal to R. We note that the
abscissas of points C and A are (sx sy)/2 and sx , respectively. The
difference in these abscissas is (sx sy)/2, as dimensioned in the figure. Also, the ordinate to point A is txy. Therefore, line CA is the
hypotenuse of a right triangle having one side of length (sx sy)/2 and
the other side of length txy. Taking the square root of the sum of the
squares of these two sides gives the radius R:
R
t
冢2
冣 莦
冪莦
莦
sx sy
2
2
xy
which is the same as Eq. (7-31b). By a similar procedure, we can show
that the length of line CB is also equal to the radius R of the circle.
Stresses on an Inclined Element
Now we will consider the stresses sx1, sy1, and tx1y1 acting on the faces
of a plane-stress element oriented at an angle u from the x axis
(Fig. 7-15b). If the angle u is known, these stresses can be determined
from Mohr’s circle. The procedure is as follows.
On the circle (Fig. 7-15c), we measure an angle 2u counterclockwise from radius CA, because point A corresponds to u 0 and is the
reference point from which we measure angles. The angle 2u locates
point D on the circle, which (as shown in the next paragraph) has coordinates sx1 and tx1y1. Therefore, point D represents the stresses on the x1
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
488
CHAPTER 7 Analysis of Stress and Strain
face of the element of Fig. 7-15b. Consequently, this face of the element
is labeled “D” in Fig. 7-15b.
Note that an angle 2u on Mohr’s circle corresponds to an angle u on
a stress element. For instance, point D on the circle is at an angle 2u
from point A, but the x1 face of the element shown in Fig. 7-15b (the
face labeled “D”) is at an angle u from the x face of the element shown
in Fig. 7-15a (the face labeled “A”). Similarly, points A and B are
180° apart on the circle, but the corresponding faces of the element
(Fig. 7-15a) are 90° apart.
To show that the coordinates sx1 and tx1y1 of point D on the circle
are indeed given by the stress-transformation equations (Eqs. 7-4a and
7-4b), we again use the geometry of the circle. Let b be the angle
between the radial line CD and the sx1 axis. Then, from the geometry of
the figure, we obtain the following expressions for the coordinates of
point D:
sx sy
sx1 ᎏ R cos b
2
tx1y1 R sin b
(7-33a,b)
Noting that the angle between the radius CA and the horizontal axis is
2u b, we get
sx sy
cos (2u b)
2R
txy
sin (2u b)
R
Expanding the cosine and sine expressions (see Appendix C) gives
sx sy
cos 2u cos b sin 2u sin b
2R
(a)
txy
sin 2u cos b cos 2u sin b
R
(b)
Multiplying the first of these equations by cos 2u and the second by
sin 2u and then adding, we obtain
冢
1 sx sy
cos b cos 2u tx y sin 2u
R
2
冣
(c)
Also, multiplying Eq. (a) by sin 2u and Eq. (b) by cos 2u and then subtracting, we get
冢
sx sy
1
sin b sin 2u txy cos 2u
R
2
冣
(d)
When these expressions for cos b and sin b are substituted into Eqs.
(7-33a) and (7-33b), we obtain the stress-transformation equations for
sx1 and tx1y1 (Eqs. 7-4a and 7-4b). Thus, we have shown that point D on
Mohr’s circle, defined by the angle 2u, represents the stress conditions
on the x1 face of the stress element defined by the angle u (Fig. 7-15b).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 7.4 Mohr’s Circle for Plane Stress
489
Point D, which is diametrically opposite point D on the circle, is
located by an angle 2u (measured from line CA) that is 180° greater than
the angle 2u to point D. Therefore, point D on the circle represents the
stresses on a face of the stress element (Fig. 7-15b) at 90° from the face
represented by point D. Thus, point D on the circle gives the stresses
sy1 and tx1y1 on the y1 face of the stress element (the face labeled
“D” in Fig. 7-15b).
From this discussion we see how the stresses represented by points
on Mohr’s circle are related to the stresses acting on an element. The
stresses on an inclined plane defined by the angle u (Fig. 7-15b) are
found on the circle at the point where the angle from the reference point
(point A) is 2u. Thus, as we rotate the x1y1 axes counterclockwise
through an angle u (Fig. 7-15b), the point on Mohr’s circle corresponding to the x1 face moves counterclockwise through an angle 2u.
Similarly, if we rotate the axes clockwise through an angle, the point on
the circle moves clockwise through an angle twice as large.
Principal Stresses
The determination of principal stresses is probably the most important
application of Mohr’s circle. Note that as we move around Mohr’s circle
(Fig. 7-15c), we encounter point P1 where the normal stress reaches its
algebraically largest value and the shear stress is zero. Hence, point P1
represents a principal stress and a principal plane. The abscissa s1 of
point P1 gives the algebraically larger principal stress and its angle 2up1
from the reference point A (where u 0) gives the orientation of the
principal plane. The other principal plane, associated with the algebraically smallest normal stress, is represented by point P2,
diametrically opposite point P1.
From the geometry of the circle, we see that the algebraically larger
principal stress is
sx sy
OC
苶苶
CP
s1 苶
苶1苶 R
2
which, upon substitution of the expression for R (Eq. 7-31b), agrees with
the earlier equation for this stress (Eq. 7-14). In a similar manner, we
can verify the expression for the algebraically smaller principal stress
s2.
The principal angle up1 between the x axis (Fig. 7-15a) and the
plane of the algebraically larger principal stress is one-half the angle
2up1, which is the angle on Mohr’s circle between radii CA and CP1.
The cosine and sine of the angle 2up1 can be obtained by inspection
from the circle:
sx sy
txy
cos 2up1
sin 2up1
2R
R
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
490
CHAPTER 7 Analysis of Stress and Strain
These equations agree with Eqs. (7-18a) and (7-18b), and so once again
we see that the geometry of the circle matches the equations derived
earlier. On the circle, the angle 2up2 to the other principal point (point
P2) is 180° larger than 2up1; hence, up2 up1 90°, as expected.
t
t
(a)
Maximum Shear Stresses
Points S1 and S2, representing the planes of maximum positive and maximum negative shear stresses, respectively, are located at the bottom and
top of Mohr’s circle (Fig. 7-15c). These points are at angles 2u 90°
from points P1 and P2, which agrees with the fact that the planes of
maximum shear stress are oriented at 45° to the principal planes.
The maximum shear stresses are numerically equal to the radius
R of the circle (compare Eq. 7-31b for R with Eq. 7-25 for tmax). Also,
the normal stresses on the planes of maximum shear stress are equal to
the abscissa of point C, which is the average normal stress saver (see
Eq. 7-31a).
(b)
Clockwise shear stresses
2u
C
sx1
O
Alternative Sign Convention for Shear Stresses
R
saver
Counterclockwise shear stresses
(c)
FIG. 7-16 Alternative sign convention for
shear stresses: (a) clockwise shear stress,
(b) counterclockwise shear stress, and
(c) axes for Mohr’s circle. (Note that
clockwise shear stresses are plotted
upward and counterclockwise shear
stresses are plotted downward.)
An alternative sign convention for shear stresses is sometimes used
when constructing Mohr’s circle. In this convention, the direction of a
shear stress acting on an element of the material is indicated by the
sense of the rotation that it tends to produce (Figs. 7-16a and b). If the
shear stress t tends to rotate the stress element clockwise, it is called a
clockwise shear stress, and if it tends to rotate it counterclockwise, it is
called a counterclockwise stress. Then, when constructing Mohr’s circle,
clockwise shear stresses are plotted upward and counterclockwise shear
stresses are plotted downward (Fig. 7-16c).
It is important to realize that the alternative sign convention produces
a circle that is identical to the circle already described (Fig. 7-15c). The
reason is that a positive shear stress tx1y1 is also a counterclockwise shear
stress, and both are plotted downward. Also, a negative shear stress tx1y1 is
a clockwise shear stress, and both are plotted upward.
Thus, the alternative sign convention merely provides a different
point of view. Instead of thinking of the vertical axis as having negative
shear stresses plotted upward and positive shear stresses plotted downward (which is a bit awkward), we can think of the vertical axis as
having clockwise shear stresses plotted upward and counterclockwise
shear stresses plotted downward (Fig. 7-16c).
General Comments about the Circle
From the preceding discussions in this section, it is apparent that we can
find the stresses acting on any inclined plane, as well as the principal
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 7.4 Mohr’s Circle for Plane Stress
491
stresses and maximum shear stresses, from Mohr’s circle. However, only
rotations of axes in the xy plane (that is, rotations about the z axis) are
considered, and therefore all stresses on Mohr’s circle are in-plane
stresses.
For convenience, the circle of Fig. 7-15 was drawn with sx, sy, and
txy as positive stresses, but the same procedures may be followed if one
or more of the stresses is negative. If one of the normal stresses is negative, part or all of the circle will be located to the left of the origin, as
illustrated in Example 7-6 that follows.
Point A in Fig. 7-15c, representing the stresses on the plane u 0,
may be situated anywhere around the circle. However, the angle 2u is
always measured counterclockwise from the radius CA, regardless of
where point A is located.
In the special cases of uniaxial stress, biaxial stress, and pure shear,
the construction of Mohr’s circle is simpler than in the general case of
plane stress. These special cases are illustrated in Example 7-4 and in
Problems 7.4-1 through 7.4-9.
Besides using Mohr’s circle to obtain the stresses on inclined planes
when the stresses on the x and y planes are known, we can also use the
circle in the opposite manner. If we know the stresses sx1, sy1, and tx1y1
acting on an inclined element oriented at a known angle u, we can easily
construct the circle and determine the stresses sx, sy, and tx y for the
angle u 0. The procedure is to locate points D and D from the known
stresses and then draw the circle using line DD as a diameter. By
measuring the angle 2u in a negative sense from radius CD, we can
locate point A, corresponding to the x face of the element. Then we can
locate point B by constructing a diameter from A. Finally, we can determine the coordinates of points A and B and thereby obtain the stresses
acting on the element for which u 0.
If desired, we can construct Mohr’s circle to scale and measure
values of stress from the drawing. However, it is usually preferable to
obtain the stresses by numerical calculations, either directly from the
various equations or by using trigonometry and the geometry of the
circle.
Mohr’s circle makes it possible to visualize the relationships
between stresses acting on planes at various angles, and it also serves as
a simple memory device for calculating stresses. Although many graphical techniques are no longer used in engineering work, Mohr’s circle
remains valuable because it provides a simple and clear picture of an
otherwise complicated analysis.
Mohr’s circle is also applicable to the transformations for plain
strain and moments of inertia of plane areas, because these quantities
follow the same transformation laws as do stresses (see Sections 7.7,
12.8, and 12.9).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
492
CHAPTER 7 Analysis of Stress and Strain
Example 7-4
At a point on the surface of a pressurized cylinder, the material is subjected to
biaxial stresses sx 90 MPa and sy 20 MPa, as shown on the stress element
of Fig. 7-17a.
Using Mohr’s circle, determine the stresses acting on an element inclined at
an angle u 30°. (Consider only the in-plane stresses, and show the results on a
sketch of a properly oriented element.)
72.5
D (u = 30°)
y
35
20
sy = 20 MPa
B
C
O
sx = 90 MPa
x
O
B
(u = 90°)
A
35
30.3
60°
35
A
(u = 0)
sx1
D'
(u = 120°)
(a)
55
90
tx1y1
FIG. 7-17 Example 7-4. (a) Element in
(b)
plane stress, and (b) the corresponding
Mohr’s circle. (Note: All stresses on the
circle have units of MPa.)
Solution
Construction of Mohr’s circle. We begin by setting up the axes for the
normal and shear stresses, with sx1 positive to the right and tx1y1 positive downward, as shown in Fig. 7-17b. Then we place the center C of the circle on the
sx1 axis at the point where the stress equals the average normal stress
(Eq. 7-31a):
sx sy
90 MPa 20 MPa
saver ᎏ 55 MPa
2
2
Point A, representing the stresses on the x face of the element (u 0), has
coordinates
sxl 90 MPa
t xl y l 0
Similarly, the coordinates of point B, representing the stresses on the y face
(u 90°), are
sxl 20 MPa
t xl y l 0
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 7.4 Mohr’s Circle for Plane Stress
493
Now we draw the circle through points A and B with center at C and radius
R (see Eq. 7-31b) equal to
R
冢 莦
冣 莦 冪莦
冢 莦莦莦
冣 0 35 MPa
冪莦
sx ⫺ sy 2
ᎏᎏ t x2y
2
90 MPa 20 MPa
2
2
Stresses on an element inclined at u 30°. The stresses acting on a plane
oriented at an angle u 30° are given by the coordinates of point D, which is at
an angle 2u 60° from point A (Fig. 7-17b). By inspection of the circle, we see
that the coordinates of point D are
(Point D)
sx l saver R cos 60°
55 MPa (35 MPa)(cos 60°) 72.5 MPa
tx1y1 R sin 60° (35 MPa)(sin 60°) 30.3 MPa
In a similar manner, we can find the stresses represented by point D, which corresponds to an angle u 120° (or 2u 240°):
(Point D)
sx l saver R cos 60°
55 MPa (35 MPa)(cos 60°) 37.5 MPa
tx1y1 R sin 60° (35 MPa)(sin 60°) 30.3 MPa
These results are shown in Fig. 7-18 on a sketch of an element oriented at an
angle u 30°, with all stresses shown in their true directions. Note that the
sum of the normal stresses on the inclined element is equal to sx sy, or
110 MPa.
y
D
37.5 MPa
D'
72.5 MPa
u = 30°
O
FIG. 7-18 Example 7-4 (continued).
x
30.3 MPa
Stresses acting on an element oriented at
an angle u 30°
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
494
CHAPTER 7 Analysis of Stress and Strain
Example 7-5
An element in plane stress at the surface of a large machine is subjected to
stresses sx 15,000 psi, sy 5,000 psi, and txy 4,000 psi, as shown in
Fig.7-19a.
Using Mohr’s circle, determine the following quantities: (a) the stresses
acting on an element inclined at an angle u 40°, (b) the principal stresses, and
(c) the maximum shear stresses. (Consider only the in-plane stresses, and show
all results on sketches of properly oriented elements.)
S2 (us2 = 64.3°)
5,000
4,000
5,000 psi
B
D (u = 40°)
B (u = 90°)
y
4,000 psi
4,000
03
80°
03 41.34°
4
,
6
5,000
P2 (up2 = 109.3°)
O
6,4
C
03
03
6,4
15,000 psi
x
O
6,4
38.66°
P1 (up1 = 19.3°)
sx1
4,000
A (u = 0)
D'
A
S1
(us1 = –25.7°)
(a)
10,000
FIG. 7-19 Example 7-5. (a) Element in
5,000
15,000
plane stress, and (b) the corresponding
Mohr’s circle. (Note: All stresses on the
circle have units of psi.)
tx1y1
(b)
Solution
Construction of Mohr’s circle. The first step in the solution is to set up the
axes for Mohr’s circle, with sx1 positive to the right and tx1y1 positive downward
(Fig. 7-19b). The center C of the circle is located on the sx1 axis at the point
where sx1 equals the average normal stress (Eq. 7-31a):
sx sy
15,000 psi 5,000 psi
saver ᎏ 10,000 psi
2
2
Point A, representing the stresses on the x face of the element (u 0), has
coordinates
sx l 15,000 psi
txl yl 4,000 psi
Similarly, the coordinates of point B, representing the stresses on the y face
(u 90°) are
sx l 5,000 psi
txl yl 4,000 psi
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
495
SECTION 7.4 Mohr’s Circle for Plane Stress
The circle is now drawn through points A and B with center at C. The radius of
the circle, from Eq. (7-31b), is
R
t
冢ᎏ2ᎏ
冣 莦
冪莦
莦
sx ⫺ sy
2
2
xy
冣 (4,000 psi) 6,403 psi
冪冢莦莦莦
莦莦 莦
2
15,000 psi 5,000 psi
2
2
(a) Stresses on an element inclined at u 40°. The stresses acting on a
plane oriented at an angle u 40° are given by the coordinates of point D,
which is at an angle 2u 80° from point A (Fig. 7-19b). To evaluate these coordinates, we need to know the angle between line CD and the sx1 axis (that is,
angle DCP1), which in turn requires that we know the angle between line CA
and the sx1 axis (angle ACP1). These angles are found from the geometry of the
circle, as follows:
4,000 psi
tan A
苶C
苶P
苶苶1 0.8
5,000 psi
AC
苶
苶P
苶苶1 38.66°
D苶
C苶
P1苶 80° 苶
AC
苶
苶P
苶1苶 80° 38.66° 41.34°
Knowing these angles, we can determine the coordinates of point D directly
from the figure:
(Point D)
sx 1 10,000 psi (6,403 psi)(cos 41.34°) 14,810 psi
tx1 y1 (6,403 psi)(sin 41.34°) 4,230 psi
FIG. 7-20 Example 7-5 (continued).
(a) Stresses acting on an element
oriented at u 40°, (b) principal
stresses, and (c) maximum shear stresses
y
5,190 psi
y
14,810 psi
10,000 psi
3,600 psi
u = 40°
D'
D
x
S2
16,400 psi
P2
4,230 psi
O
y
O
up1 = 19.3°
x
6,400 psi
O
x
us1 = – 25.7°
P1
10,000 psi
S1
(a)
(b)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(c)
continued
496
CHAPTER 7 Analysis of Stress and Strain
y
5,190 psi
In an analogous manner, we can find the stresses represented by point D, which
corresponds to a plane inclined at an angle u 130° (or 2u 260°):
14,810 psi
(Point D9 )
u = 40°
D'
4,230 psi
O
x
D
(a)
y
sx l 10,000 psi (6,403 psi)(cos 41.34°) 5,190 psi
tx1 y1 (6,403 psi)(sin 41.34°) 4,230 psi
These stresses are shown in Fig. 7-20a on a sketch of an element oriented at an
angle u 40° (all stresses are shown in their true directions). Also, note that the
sum of the normal stresses is equal to sx sy, or 20,000 psi.
(b) Principal stresses. The principal stresses are represented by points P1
and P2 on Mohr’s circle (Fig. 7-19b). The algebraically larger principal stress
(point P1) is
s1 10,000 psi 6,400 psi 16,400 psi
as seen by inspection of the circle. The angle 2u p1 to point P1 from point A is
the angle ACP1 on the circle, that is,
3,600 psi
16,400 psi
P2
A
苶C
苶P
苶1苶 2up1 38.66°
up1 = 19.3°
O
x
P1
u p1 19.3°
Thus, the plane of the algebraically larger principal stress is oriented at an angle
up1 19.3°, as shown in Fig. 7-20b.
The algebraically smaller principal stress (represented by point P2) is
obtained from the circle in a similar manner:
s 2 10,000 psi 6,400 psi 3,600 psi
(b)
y
The angle 2u p 2 to point P2 on the circle is 38.66° 180° 218.66°; thus, the
second principal plane is defined by the angle up2 109.3°. The principal
stresses and principal planes are shown in Fig. 7-20b, and again we note that the
sum of the normal stresses is equal to 20,000 psi.
(c) Maximum shear stresses. The maximum shear stresses are represented
by points S1 and S2 on Mohr’s circle; therefore, the maximum in-plane shear
stress (equal to the radius of the circle) is
10,000 psi
S2
6,400 psi
O
x
us1 = – 25.7°
10,000 psi
S1
(c)
FIG. 7-20 (Repeated)
tmax 6,400 psi
The angle ACS1 from point A to point S1 is 90° 38.66° 51.34°, and therefore the angle 2us1 for point S1 is
2us1 51.34°
This angle is negative because it is measured clockwise on the circle. The corresponding angle us1 to the plane of the maximum positive shear stress is one-half
that value, or us1 25.7°, as shown in Figs. 7-19b and 7-20c. The maximum
negative shear stress (point S2 on the circle) has the same numerical value as the
maximum positive stress (6,400 psi).
The normal stresses acting on the planes of maximum shear stress are equal
to saver, which is the abscissa of the center C of the circle (10,000 psi). These
stresses are also shown in Fig. 7-20c. Note that the planes of maximum shear
stress are oriented at 45° to the principal planes.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 7.4 Mohr’s Circle for Plane Stress
497
Example 7-6
At a point on the surface of a generator shaft the stresses are sx 50 MPa,
sy 10 MPa, and txy 40 MPa, as shown in Fig. 7-21a.
Using Mohr’s circle, determine the following quantities: (a) the stresses
acting on an element inclined at an angle u 45°, (b) the principal stresses, and
(c) the maximum shear stresses. (Consider only the in-plane stresses, and show
all results on sketches of properly oriented elements.)
50
y
A (u = 0)
10 MPa
S2
D'
A
B
O
90°
50 MPa
x
P2
(up2 = 26.6°)
P1 (up1 = 116.6°)
53.13°
36.87°
O
C
50
sx1
50
40 MPa
50
50
40
40
D (u = 45°)
B (u = 90°)
S1
(a)
(us1 = 71.6°)
FIG. 7-21 Example 7-6. (a) Element in
plane stress, and (b) the corresponding
Mohr’s circle. (Note: All stresses on the
circle have units of MPa.)
20
(b)
10
tx1y1
Solution
Construction of Mohr’s circle. The axes for the normal and shear stresses
are shown in Fig. 7-21b, with sx1 positive to the right and tx1y1 positive downward. The center C of the circle is located on the sx1 axis at the point where the
stress equals the average normal stress (Eq. 7-31a):
sx sy
50 MPa 10 MPa
saver 20 MPa
2
2
Point A, representing the stresses on the x face of the element (u 0), has
coordinates
sx l 50 MPa
tx l y l 40 MPa
Similarly, the coordinates of point B, representing the stresses on the y face
(u 90°), are
sx l 10 MPa
tx l y l 40 MPa
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
498
CHAPTER 7 Analysis of Stress and Strain
The circle is now drawn through points A and B with center at C and radius R
(from Eq. 7-31b) equal to
R
s ⫺s
ᎏ冣 t
冢ᎏ莦
冪莦
2 莦莦
x
y
2
2
xy
冣 (40 MPa) 50 MPa
冪冢莦莦莦
莦 莦莦
2
50 MPa 10 MPa
2
2
(a) Stresses on an element inclined at u 45°. The stresses acting on a
plane oriented at an angle u 45° are given by the coordinates of point D,
which is at an angle 2u 90° from point A (Fig. 7-21b). To evaluate these coordinates, we need to know the angle between line CD and the negative sx1 axis
(that is, angle DCP2), which in turn requires that we know the angle between
line CA and the negative sx1 axis (angle ACP2). These angles are found from the
geometry of the circle as follows:
40 MPa
4
C苶
P2苶
tan A
苶苶
30 MPa
3
A
苶C
苶P
苶苶2 53.13°
D
AC
苶C
苶P
苶苶2 90° 苶
苶P
苶苶2 90° 53.13° 36.87°
Knowing these angles, we can obtain the coordinates of point D directly from
the figure:
sx 1 20 MPa (50 MPa)(cos 36.87°) 60 MPa
(Point D)
tx 1 y 1 (50 MPa)(sin 36.87°) 30 MPa
FIG. 7-22 Example 7-6 (continued).
(a) Stresses acting on an element
oriented at u 45°, (b) principal
stresses, and (c) maximum shear stresses
y
y
30 MPa
20 MPa
30 MPa
50 MPa
60 MPa
D'
y
O
D
P1
70 MPa
u = 45°
x
up2 = 26.6°
O
P2
20 MPa
20 MPa
x
O
x
S2
(a)
(b)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
us1 = 71.6°
S1
(c)
SECTION 7.4 Mohr’s Circle for Plane Stress
499
In an analogous manner, we can find the stresses represented by point D, which
corresponds to a plane inclined at an angle u 135° (or 2u 270°):
(Point D9)
sx l 20 MPa (50 MPa)(cos 36.87°) 20 MPa
tx 1 y1 (50 MPa)(sin 36.87°) 30 MPa
These stresses are shown in Fig. 7-22a on a sketch of an element oriented at an
angle u 45° (all stresses are shown in their true directions). Also, note that the
sum of the normal stresses is equal to sx sy, or 40 MPa.
(b) Principal stresses. The principal stresses are represented by points P1
and P2 on Mohr’s circle. The algebraically larger principal stress (represented
by point P1) is
s1 20 MPa 50 MPa 30 MPa
as seen by inspection of the circle. The angle 2up1 to point P1 from point A is
the angle ACP1 measured counterclockwise on the circle, that is,
A苶
C苶
P苶1 2up1 53.13° 180° 233.13°
苶
up1 116.6°
Thus, the plane of the algebraically larger principal stress is oriented at an angle
up1 116.6°.
The algebraically smaller principal stress (point P2) is obtained from the
circle in a similar manner:
s2 20 MPa 50 MPa 70 MPa
The angle 2up2 to point P2 on the circle is 53.13°; thus, the second principal
plane is defined by the angle up2 26.6°.
The principal stresses and principal planes are shown in Fig. 7-22b, and
again we note that the sum of the normal stresses is equal to sx sy, or
40 MPa.
(c) Maximum shear stresses. The maximum positive and negative shear
stresses are represented by points S1 and S2 on Mohr’s circle (Fig. 7-21b). Their
magnitudes, equal to the radius of the circle, are
tmax 50 MPa
The angle ACS1 from point A to point S1 is 90° 53.13° 143.13°, and therefore the angle 2us1 for point S1 is
2us1 143.13°
The corresponding angle us1 to the plane of the maximum positive shear stress
is one-half that value, or us1 71.6°, as shown in Fig. 7-22c. The maximum
negative shear stress (point S2 on the circle) has the same numerical value as the
positive stress (50 MPa).
The normal stresses acting on the planes of maximum shear stress are equal
to saver, which is the coordinate of the center C of the circle (20 MPa). These
stresses are also shown in Fig. 7-22c. Note that the planes of maximum shear
stress are oriented at 45° to the principal planes.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
500
CHAPTER 7 Analysis of Stress and Strain
7.5 HOOKE’S LAW FOR PLANE STRESS
y
sy
txy
sx
O
x
z
FIG. 7-23 Element of material in plane
stress (sz 0)
y
cez
a
c
aex
bey
b
O
x
The stresses acting on inclined planes when the material is subjected to
plane stress (Fig. 7-23) were discussed in Sections 7.2, 7.3, and 7.4. The
stress-transformation equations derived in those discussions were
obtained solely from equilibrium, and therefore the properties of the
materials were not needed. Now, in this section, we will investigate the
strains in the material, which means that the material properties must be
considered. However, we will limit our discussion to materials that meet
two important conditions: first, the material is uniform throughout the
body and has the same properties in all directions (homogeneous and
isotropic material), and second, the material follows Hooke’s law (linearly elastic material). Under these conditions, we can readily obtain the
relationships between the stresses and strains in the body.
Let us begin by considering the normal strains ex, ey, and ez in
plane stress. The effects of these strains are pictured in Fig. 7-24, which
shows the changes in dimensions of a small element having edges of
lengths a, b, and c. All three strains are shown positive (elongation) in
the figure. The strains can be expressed in terms of the stresses (Fig. 7-23)
by superimposing the effects of the individual stresses.
For instance, the strain ex in the x direction due to the stress sx is
equal to sx /E, where E is the modulus of elasticity. Also, the strain ex
due to the stress sy is equal to ⫺nsy /E, where n is Poisson’s ratio (see
Section 1.5). Of course, the shear stress txy produces no normal strains
in the x, y, or z directions. Thus, the resultant strain in the x direction is
1
e x ᎏᎏ(sx ⫺ nsy)
E
z
FIG. 7-24 Element of material subjected
to normal strains ex, ey, and ez
O
z
FIG. 7-25 Shear strain gxy
In a similar manner, we obtain the strains in the y and z directions:
1
e y ᎏᎏ(sy ⫺ nsx)
E
n
ez (sx sy)
E
(7-34b,c)
These equations may be used to find the normal strains (in plane stress)
when the stresses are known.
The shear stress txy (Fig. 7-23) causes a distortion of the element
such that each z face becomes a rhombus (Fig. 7-25). The shear strain
gxy is the decrease in angle between the x and y faces of the element and
is related to the shear stress by Hooke’s law in shear, as follows:
y
p –g
xy
2
(7-34a)
x
p –g
xy
2
txy
gxy
G
(7-35)
where G is the shear modulus of elasticity. Note that the normal stresses
sx and sy have no effect on the shear strain gxy. Consequently, Eqs.
(7-34) and (7-35) give the strains (in plane stress) when all stresses
(sx, sy, and txy) act simultaneously.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 7.5 Hooke’s Law for Plane Stress
501
The first two equations (Eqs. 7-34a and 7-34b) give the strains ex
and ey in terms of the stresses. These equations can be solved simultaneously for the stresses in terms of the strains:
E
sx ᎏᎏ2 (ex ney)
1⫺n
E
sy
(ey nex)
1n2
(7-36a,b)
In addition, we have the following equation for the shear stress in terms
of the shear strain:
(7-37)
txy Ggxy
Equations (7-36) and (7-37) may be used to find the stresses (in plane
stress) when the strains are known. Of course, the normal stress sz in the
z direction is equal to zero.
Equations (7-34) through (7-37) are known collectively as Hooke’s
law for plane stress. They contain three material constants (E, G,
and n), but only two are independent because of the relationship
E
G
2(1 n)
(7-38)
which was derived previously in Section 3.6.
Special Cases of Hooke’s Law
In the special case of biaxial stress (Fig. 7-10b), we have txy 0, and
therefore Hooke’s law for plane stress simplifies to
1
ex (sx nsy)
E
1
ey (sy nsx)
E
n
(7-39a,b,c)
ez (sx sy)
E
E
E
sy 2 (e y nex) (7-40a,b)
sx 2 (ex ney)
1n
1 n
These equations are the same as Eqs. (7-34) and (7-36) because the
effects of normal and shear stresses are independent of each other.
For uniaxial stress, with sy 0 (Fig. 7-10a), the equations of
Hooke’s law simplify even further:
sx
ex
E
nsx
ey ez
E
sx Eex
(7-41a,b,c)
Finally, we consider pure shear (Fig. 7-11a), which means that sx
sy 0. Then we obtain
txy
gxy
(7-42a,b)
G
In all three of these special cases, the normal stress sz is equal to zero.
ex ey ez 0
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
502
CHAPTER 7 Analysis of Stress and Strain
Volume Change
y
cez
a
c
aex
bey
b
O
x
When a solid object undergoes strains, both its dimensions and its volume will change. The change in volume can be determined if the normal
strains in three perpendicular directions are known. To show how this is
accomplished, let us again consider the small element of material shown
in Fig. 7-24. The original element is a rectangular parallelepiped having
sides of lengths a, b, and c in the x, y, and z directions, respectively. The
strains ex, ey, and ez produce the changes in dimensions shown by the
dashed lines. Thus, the increases in the lengths of the sides are aex , bey ,
and cez.
The original volume of the element is
V0 abc
z
FIG. 7-24 (Repeated)
(a)
and its final volume is
V1 (a aex)(b bey)(c cez)
abc(1 ex)(1 ey)(1 ez)
(b)
By referring to Eq. (a), we can express the final volume of the element
(Eq. b) in the form
V1 V0(1 ex)(1 ey)(1 ez)
(7-43a)
Upon expanding the terms on the right-hand side, we obtain the following equivalent expression:
V1 V0(1 ex ey ez exey exez eyez exeyez)
(7-43b)
The preceding equations for V1 are valid for both large and small strains.
If we now limit our discussion to structures having only very small
strains (as is usually the case), we can disregard the terms in Eq. (7-43b)
that consist of products of small strains. Such products are themselves
small in comparison to the individual strains ex, ey, and ez. Then the
expression for the final volume simplifies to
V1 V0(1 ex ey ez)
(7-44)
and the volume change is
V V1 V0 V0(ex ey ez)
(7-45)
This expression can be used for any volume of material provided the
strains are small and remain constant throughout the volume. Note also
that the material does not have to follow Hooke’s law. Furthermore, the
expression is not limited to plane stress, but is valid for any stress conditions. (As a final note, we should mention that shear strains produce no
change in volume.)
The unit volume change e, also known as the dilatation, is defined
as the change in volume divided by the original volume; thus,
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 7.5 Hooke’s Law for Plane Stress
y
V
e ᎏ ex ey ez
V0
sy
sx
x
z
FIG. 7-23 (Repeated)
V
1 2n
e (sx sy)
V0
E
y
cez
a
c
aex
bey
b
O
x
z
FIG. 7-24 (Repeated)
(7-46)
By applying this equation to a differential element of volume and then
integrating, we can obtain the change in volume of a body even when
the normal strains vary throughout the body.
The preceding equations for volume changes apply to both tensile
and compressive strains, inasmuch as the strains ex, ey, and ez are algebraic quantities (positive for elongation and negative for shortening).
With this sign convention, positive values for V and e represent
increases in volume, and negative values represent decreases.
Let us now return to materials that follow Hooke’s law and are subjected only to plane stress (Fig. 7-23). In this case the strains ex, ey, and
ez are given by Eqs. (7-34a, b, and c). Substituting those relationships
into Eq. (7-46), we obtain the following expression for the unit volume
change in terms of stresses:
txy
O
503
(7-47)
Note that this equation also applies to biaxial stress.
In the case of a prismatic bar in tension, that is, uniaxial stress,
Eq. (7-47) simplifies to
V
sx
2n)
e (1
V0
E
(7-48)
From this equation we see that the maximum possible value of Poisson’s
ratio for common materials is 0.5, because a larger value means that the
volume decreases when the material is in tension, which is contrary to
ordinary physical behavior.
Strain-Energy Density in Plane Stress
y
p –g
xy
2
O
z
FIG. 7-25 (Repeated)
x
p –g
xy
2
The strain-energy density u is the strain energy stored in a unit volume
of the material (see the discussions in Sections 2.7 and 3.9). For an element in plane stress, we can obtain the strain-energy density by referring
to the elements pictured in Figs. 7-24 and 7-25. Because the normal and
shear strains occur independently, we can add the strain energies from
these two elements to obtain the total energy.
Let us begin by finding the strain energy associated with the normal
strains (Fig. 7-24). Since the stress acting on the x face of the element is
sx (see Fig. 7-23), we find that the force acting on the x face of the element (Fig. 7-24) is equal to sx bc. Of course, as the loads are applied to
the structure, this force increases gradually from zero to its maximum
value. At the same time, the x face of the element moves through the distance aex. Therefore, the work done by this force is
1
(sx bc)(aex)
2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
504
CHAPTER 7 Analysis of Stress and Strain
provided Hooke’s law holds for the material. Similarly, the force sy ac
acting on the y face does work equal to
1
ᎏᎏ (sy ac)(bey)
2
The sum of these two terms gives the strain energy stored in the
element:
abc
ᎏᎏ (sx ex sye y)
2
Thus, the strain-energy density (strain energy per unit volume) due to
the normal stresses and strains is
1
(c)
u1 (sxex syey)
2
The strain-energy density associated with the shear strains (Fig.
7-25) was evaluated previously in Section 3.9 (see Eq. d of that section):
txygxy
(d)
u 2
2
By combining the strain-energy densities for the normal and shear
strains, we obtain the following formula for the strain-energy density
in plane stress:
1
u (sxex syey txygxy)
2
(7-49)
Substituting for the strains from Eqs. (7-34) and (7-35), we obtain the
strain-energy density in terms of stresses alone:
2
txy
1
u (s 2x s y2 2nsx sy)
2E
(7-50)
In a similar manner, we can substitute for the stresses from Eqs. (7-36)
and (7-37) and obtain the strain-energy density in terms of strains alone:
2
Ggxy
E
2
2
u
2 (e x e y 2ne x e y)
2(1 n )
2
(7-51)
To obtain the strain-energy density in the special case of biaxial
stress, we simply drop the shear terms in Eqs. (7-49), (7-50), and (7-51).
For the special case of uniaxial stress, we substitute the following
values
sy 0
tx y 0
e y nex
gx y 0
into Eqs. (7-50) and (7-51) and obtain, respectively,
s 2x
u
2E
Ee 2x
u
These equations agree with Eqs. (2-44a) and (2-44b) of Section 2.7.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(e,f)
505
SECTION 7.6 Triaxial Stress
Also, for pure shear we substitute
sx sy 0
ex ey 0
into Eqs. (7-50) and (7-51) and obtain
2
t xy
u
Gg xy2
u
2
(g,h)
These equations agree with Eqs. (3-55a) and (3-55b) of Section 3.9.
7.6 TRIAXIAL STRESS
y
sy
sz
sx
sx
O
x
sz
z
sy
(a)
s
t
u
An element of material subjected to normal stresses sx, sy, and sz acting
in three mutually perpendicular directions is said to be in a state of
triaxial stress (Fig. 7-26a). Since there are no shear stresses on the x, y,
and z faces, the stresses sx , sy , and sz are the principal stresses in the
material.
If an inclined plane parallel to the z axis is cut through the element
(Fig. 7-26b), the only stresses on the inclined face are the normal stress
s and shear stress t, both of which act parallel to the xy plane. These
stresses are analogous to the stresses sx1 and tx1y1 encountered in our
earlier discussions of plane stress (see, for instance, Fig. 7-2a). Because
the stresses s and t (Fig. 7-26b) are found from equations of force equilibrium in the xy plane, they are independent of the normal stress sz .
Therefore, we can use the transformation equations of plane stress, as
well as Mohr’s circle for plane stress, when determining the stresses s
and t in triaxial stress. The same general conclusion holds for the normal
and shear stresses acting on inclined planes cut through the element
parallel to the x and y axes.
Maximum Shear Stresses
sx
sz
sy
(b)
FIG. 7-26 Element in triaxial stress
From our previous discussions of plane stress, we know that the maximum shear stresses occur on planes oriented at 45° to the principal
planes. Therefore, for a material in triaxial stress (Fig. 7-26a), the maximum shear stresses occur on elements oriented at angles of 45° to the x,
y, and z axes. For example, consider an element obtained by a 45°
rotation about the z axis. The maximum positive and negative shear
stresses acting on this element are
sx sy
(tmax)z
2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(7-52a)
506
CHAPTER 7 Analysis of Stress and Strain
Similarly, by rotating about the x and y axes through angles of 45°, we
obtain the following maximum shear stresses:
sy sz
(tmax)x
2
C
A
B
O
s
sz
sy
sx
t
FIG. 7-27 Mohr’s circles for an element
in triaxial stress
sx sz
(tmax)y
2
(7-52b,c)
The absolute maximum shear stress is the numerically largest of the
stresses determined from Eqs. (7-52a, b, and c). It is equal to one-half
the difference between the algebraically largest and algebraically smallest of the three principal stresses.
The stresses acting on elements oriented at various angles to the x,
y, and z axes can be visualized with the aid of Mohr’s circles. For
elements oriented by rotations about the z axis, the corresponding circle
is labeled A in Fig. 7-27. Note that this circle is drawn for the case in
which sx sy and both sx and sy are tensile stresses.
In a similar manner, we can construct circles B and C for elements
oriented by rotations about the x and y axes, respectively. The radii of
the circles represent the maximum shear stresses given by Eqs. (7-52a,
b, and c), and the absolute maximum shear stress is equal to the radius
of the largest circle. The normal stresses acting on the planes of maximum shear stresses have magnitudes given by the abscissas of the
centers of the respective circles.
In the preceding discussion of triaxial stress we only considered
stresses acting on planes obtained by rotating about the x, y, and z axes.
Thus, every plane we considered is parallel to one of the axes. For
instance, the inclined plane of Fig. 7-26b is parallel to the z axis, and its
normal is parallel to the xy plane. Of course, we can also cut through
the element in skew directions, so that the resulting inclined planes
are skew to all three coordinate axes. The normal and shear stresses
acting on such planes can be obtained by a more complicated threedimensional analysis. However, the normal stresses acting on skew
planes are intermediate in value between the algebraically maximum
and minimum principal stresses, and the shear stresses on those planes
are smaller (in absolute value) than the absolute maximum shear stress
obtained from Eqs. (7-52a, b, and c).
Hooke’s Law for Triaxial Stress
If the material follows Hooke’s law, we can obtain the relationships
between the normal stresses and normal strains by using the same procedure
as for plane stress (see Section 7.5). The strains produced by the stresses
sx, sy, and sz acting independently are superimposed to obtain the
resultant strains. Thus, we readily arrive at the following equations for
the strains in triaxial stress:
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 7.6 Triaxial Stress
sx
n
e x ᎏᎏ ⫺ ᎏᎏ (sy sz )
E
E
sy
n
e y (sz sx )
E
E
sz
n
e z (sx sy )
E
E
507
(7-53a)
(7-53b)
(7-53c)
In these equations, the standard sign conventions are used; that is, tensile
stress s and extensional strain e are positive.
The preceding equations can be solved simultaneously for the
stresses in terms of the strains:
E
sx 冤(1 n)e x n(e y e z )冥
(1 n)(1 2n)
(7-54a)
E
sy 冤(1 n)e y n(e z e x )冥
(1 n)(1 2n)
(7-54b)
E
sz 冤(1 n)e z n(e x e y )冥
(1 n)(1 2n)
(7-54c)
Equations (7-53) and (7-54) represent Hooke’s law for triaxial stress.
In the special case of biaxial stress (Fig. 7-10b), we can obtain the
equations of Hooke’s law by substituting sz 0 into the preceding
equations. The resulting equations reduce to Eqs. (7-39) and (7-40) of
Section 7.5.
Unit Volume Change
The unit volume change (or dilatation) for an element in triaxial stress is
obtained in the same manner as for plane stress (see Section 7.5). If the
element is subjected to strains ex, ey, and ez, we may use Eq. (7-46) for
the unit volume change:
e ex ey ez
(7-55)
This equation is valid for any material provided the strains are small.
If Hooke’s law holds for the material, we can substitute for the
strains ex, ey, and ez from Eqs. (7-53a, b, and c) and obtain
1 2n
e (sx sy sz)
E
(7-56)
Equations (7-55) and (7-56) give the unit volume change in triaxial
stress in terms of the strains and stresses, respectively.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
508
CHAPTER 7 Analysis of Stress and Strain
Strain-Energy Density
The strain-energy density for an element in triaxial stress is obtained by
the same method used for plane stress. When stresses sx and sy act
alone (biaxial stress), the strain-energy density (from Eq. 7-49 with the
shear term discarded) is
1
u ᎏᎏ (sx ex sy ey)
2
When the element is in triaxial stress and subjected to stresses sx, sy,
and sz, the expression for strain-energy density becomes
1
u (sx ex sy ey sz ez)
2
(7-57a)
Substituting for the strains from Eqs. (7-53a, b, and c), we obtain the
strain-energy density in terms of the stresses:
1
n
u (s x2 s y2 s z2 ) (sx sy sx sz sy sz ) (7-57b)
2E
E
In a similar manner, but using Eqs. (7-54a, b, and c), we can express the
strain-energy density in terms of the strains:
E
u [(1 n)(e 2x e 2y e 2z )
2(1 n)(1 2n)
2n(e xe y e xe z e ye z)]
(7-57c)
When calculating from these expressions, we must be sure to substitute
the stresses and strains with their proper algebraic signs.
Spherical Stress
y
A special type of triaxial stress, called spherical stress, occurs whenever all three normal stresses are equal (Fig. 7-28):
s0
s0
s0
s0
O
x
s0
z
s0
FIG. 7-28 Element in spherical stress
sx sy sz s0
(7-58)
Under these stress conditions, any plane cut through the element will be
subjected to the same normal stress s0 and will be free of shear stress.
Thus, we have equal normal stresses in every direction and no shear
stresses anywhere in the material. Every plane is a principal plane, and
the three Mohr’s circles shown in Fig. 7-27 reduce to a single point.
The normal strains in spherical stress are also the same in all directions, provided the material is homogeneous and isotropic. If Hooke’s
law applies, the normal strains are
s0
e 0 (1 2n)
E
as obtained from Eqs. (7-53a, b, and c).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(7-59)
SECTION 7.6 Triaxial Stress
509
Since there are no shear strains, an element in the shape of a cube
changes in size but remains a cube. In general, any body subjected to
spherical stress will maintain its relative proportions but will expand or
contract in volume depending upon whether s0 is tensile or compressive.
The expression for the unit volume change can be obtained from Eq.
(7-55) by substituting for the strains from Eq. (7-59). The result is
3s0(1 ⫺ 2n)
e 3e 0 ᎏᎏ
E
(7-60)
Equation (7-60) is usually expressed in more compact form by introducing a new quantity K called the volume modulus of elasticity, or bulk
modulus of elasticity, which is defined as follows:
E
K ᎏᎏ
3(1 ⫺ 2n)
(7-61)
With this notation, the expression for the unit volume change becomes
s0
e ᎏᎏ
K
(7-62)
s0
K ᎏᎏ
e
(7-63)
and the volume modulus is
Thus, the volume modulus can be defined as the ratio of the spherical
stress to the volumetric strain, which is analogous to the definition of the
modulus E in uniaxial stress. Note that the preceding formulas for e and
K are based upon the assumptions that the strains are small and Hooke’s
law holds for the material.
From Eq. (7-61) for K, we see that if Poisson’s ratio n equals 1/3,
the moduli K and E are numerically equal. If n 0, then K has the value
E/3, and if n 0.5, K becomes infinite, which corresponds to a rigid
material having no change in volume (that is, the material is incompressible).
The preceding formulas for spherical stress were derived for an
element subjected to uniform tension in all directions, but of course the
formulas also apply to an element in uniform compression. In the case
of uniform compression, the stresses and strains have negative signs.
Uniform compression occurs when the material is subjected to uniform
pressure in all directions; for example, an object submerged in water
or rock deep within the earth. This state of stress is often called
hydrostatic stress.
Although uniform compression is relatively common, a state of
uniform tension is difficult to achieve. It can be realized by suddenly
and uniformly heating the outer surface of a solid metal sphere, so that
the outer layers are at a higher temperature than the interior. The
tendency of the outer layers to expand produces uniform tension in all
directions at the center of the sphere.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
510
CHAPTER 7 Analysis of Stress and Strain
7.7 PLANE STRAIN
The strains at a point in a loaded structure vary according to the orientation of the axes, in a manner similar to that for stresses. In this section
we will derive the transformation equations that relate the strains in
inclined directions to the strains in the reference directions. These
transformation equations are widely used in laboratory and field investigations involving measurements of strains.
Strains are customarily measured by strain gages; for example,
gages are placed in aircraft to measure structural behavior during flight,
and gages are placed in buildings to measure the effects of earthquakes.
Since each gage measures the strain in one particular direction, it is
usually necessary to calculate the strains in other directions by means of
the transformation equations.
Plane Strain Versus Plane Stress
Let us begin by explaining what is meant by plane strain and how it
relates to plane stress. Consider a small element of material having sides
of lengths a, b, and c in the x, y, and z directions, respectively (Fig.
7-29a). If the only deformations are those in the xy plane, then three
strain components may exist—the normal strain ex in the x direction
(Fig. 7-29b), the normal strain ey in the y direction (Fig. 7-29c), and the
shear strain gxy (Fig. 7-29d). An element of material subjected to these
strains (and only these strains) is said to be in a state of plane strain.
It follows that an element in plane strain has no normal strain ez in
the z direction and no shear strains gxz and gyz in the xz and yz planes,
respectively. Thus, plane strain is defined by the following conditions:
ez 0
gx z 0
gy z 0
(7-64a,b,c)
The remaining strains (ex , e y , and gx y) may have nonzero values.
FIG. 7-29 Strain components ex, ey, and
gxy in the xy plane (plane strain)
y
a
y
c
y
y
bey
(a)
O
gxy
b
x
b
O
x
a
z
(a)
x
O
O
x
aex
(b)
(c)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(d)
511
SECTION 7.7 Plane Strain
From the preceding definition, we see that plane strain occurs when
the front and rear faces of an element of material (Fig. 7-29a) are fully
restrained against displacement in the z direction—an idealized condition that is seldom reached in actual structures. However, this does not
mean that the transformation equations of plane strain are not useful. It
turns out that they are extremely useful because they also apply to the
strains in plane stress, as explained in the following paragraphs.
The definition of plane strain (Eqs. 7-64a, b, and c) is analogous to
that for plane stress. In plane stress, the following stresses must be zero:
sz 0
tx z 0
tyz 0
(7-65a,b,c)
whereas the remaining stresses (sx , sy , and txy) may have nonzero values. A comparison of the stresses and strains in plane stress and plane
strain is given in Fig. 7-30.
It should not be inferred from the similarities in the definitions of
plane stress and plane strain that both occur simultaneously. In general,
an element in plane stress will undergo a strain in the z direction (Fig.
7-30); hence, it is not in plane strain. Also, an element in plane strain
usually will have stresses sz acting on it because of the requirement that
ez 0; therefore, it is not in plane stress. Thus, under ordinary conditions plane stress and plane strain do not occur simultaneously.
Plane stress
Plane strain
y
y
sy
ey
txy
gxy
ex
sx
O
z
sz = 0
Stresses
Strains
FIG. 7-30 Comparison of plane stress and
O
x
x
z
txz = 0
tyz = 0
txz = 0
tyz = 0
sx, sy, and txy may have
nonzero values
sx, sy, sz, and txy may have
nonzero values
gxz = 0
ez = 0
gyz = 0
gxz = 0
gyz = 0
ex, ey, ez, and gxy may have
ex, ey, and gxy may have
nonzero values
nonzero values
plane strain
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
512
CHAPTER 7 Analysis of Stress and Strain
An exception occurs when an element in plane stress is subjected to
equal and opposite normal stresses (that is, when sx ⫺sy) and
Hooke’s law holds for the material. In this special case, there is no
normal strain in the z direction, as shown by Eq. (7-34c), and therefore
the element is in a state of plane strain as well as plane stress. Another
special case, albeit a hypothetical one, is when a material has Poisson’s
ratio equal to zero (n 0); then every plane stress element is also in
plane strain because ez 0 (Eq. 7-34c).*
Application of the Transformation Equations
The stress-transformation equations derived for plane stress in the xy
plane (Eqs. 7-4a and 7-4b) are valid even when a normal stress sz is
present. The explanation lies in the fact that the stress sz does not enter
the equations of equilibrium used in deriving Eqs. (7-4a) and (7-4b).
Therefore, the transformation equations for plane stress can also be
used for the stresses in plane strain.
An analogous situation exists for plane strain. Although we will
derive the strain-transformation equations for the case of plane strain in
the xy plane, the equations are valid even when a strain ez exists. The
reason is simple enough—the strain ez does not affect the geometric
relationships used in the derivations. Therefore, the transformation
equations for plane strain can also be used for the strains in plane
stress.
Finally, we should recall that the transformation equations for plane
stress were derived solely from equilibrium and therefore are valid for
any material, whether linearly elastic or not. The same conclusion
applies to the transformation equations for plane strain—since they are
derived solely from geometry, they are independent of the material
properties.
Transformation Equations for Plane Strain
y1
y
u
x1
u
O
x
FIG. 7-31 Axes x1 and y1 rotated through
an angle u from the xy axes
In the derivation of the transformation equations for plane strain, we will
use the coordinate axes shown in Fig. 7-31. We will assume that the normal strains ex and ey and the shear strain gxy associated with the xy axes
are known (Fig. 7-29). The objectives of our analysis are to determine
the normal strain ex1 and the shear strain gx1 y1 associated with the x1y1
axes, which are rotated counterclockwise through an angle u from the xy
axes. (It is not necessary to derive a separate equation for the normal
strain ey1 because it can be obtained from the equation for ex1 by
substituting u 90° for u.)
*In the discussions of this chapter we are omitting the effects of temperature changes
and prestrains, both of which produce additional deformations that may alter some of our
conclusions.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
513
SECTION 7.7 Plane Strain
Normal strain ex1. To determine the normal strain ex1 in the x1
direction, we consider a small element of material selected so that the x1
axis is along a diagonal of the z face of the element and the x and y axes
are along the sides of the element (Fig. 7-32a). The figure shows a twodimensional view of the element, with the z axis toward the viewer. Of
course, the element is actually three dimensional, as in Fig. 7-29a, with
a dimension in the z direction.
Consider first the strain ex in the x direction (Fig. 7-32a). This strain
produces an elongation in the x direction equal to ex dx, where dx is the
length of the corresponding side of the element. As a result of this elongation, the diagonal of the element increases in length by an amount
e x dx cos u
(a)
as shown in Fig. 7-32a.
Next, consider the strain ey in the y direction (Fig. 7-32b). This
strain produces an elongation in the y direction equal to ey dy, where dy
is the length of the side of the element parallel to the y axis. As a result
of this elongation, the diagonal of the element increases in length by an
amount
e y dy sin u
(b)
which is shown in Fig. 7-32b.
y
y
ex dx cos u
y1
ds
O
u
x1
ey dy sin u
x1
y1
ey dy
dy
a1
dx
ex dx
ds
x
O
dx
(a)
(b)
y
gxy dy cos u
y1
x1
gxy dy
gxy
ds
u
FIG. 7-32 Deformations of an element in
plane strain due to (a) normal strain ex,
(b) normal strain ey, and (c) shear
strain gxy
O
a3
dx
(c)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
dy
x
a2
dy
u
x
514
CHAPTER 7 Analysis of Stress and Strain
Finally, consider the shear strain gxy in the xy plane (Fig. 7-32c).
This strain produces a distortion of the element such that the angle at the
lower left corner of the element decreases by an amount equal to the
shear strain. Consequently, the upper face of the element moves to
the right (with respect to the lower face) by an amount gxy dy. This
deformation results in an increase in the length of the diagonal equal to
gx y dy cos u
(c)
as shown in Fig. 7-32c.
The total increase d in the length of the diagonal is the sum of the
preceding three expressions; thus,
d e x dx cos u e y dy sin u gx y dy cos u
(d)
The normal strain ex1 in the xl direction is equal to this increase in length
divided by the initial length ds of the diagonal:
dy
dy
d
dx
e x1 e x cos u e y sin u gx y cos u
ds
ds
ds
ds
(e)
Observing that dx/ds cos u and dy/ds sin u, we obtain the following equation for the normal strain:
e x1 e x cos 2 u e y sin2 u gx y sin u cos u
y
y1
b
b
gx1y1 = a + b
a
x1
a
u
O
x
FIG. 7-33 Shear strain gx1y1 associated
with the x1y1 axes
(7-66)
Thus, we have obtained an expression for the normal strain in the x1
direction in terms of the strains ex, ey, and gxy associated with the xy
axes.
As mentioned previously, the normal strain e y1 in the y l direction is
obtained from the preceding equation by substituting u 90° for u.
Shear strain gx1y1. Now we turn to the shear strain gx1y1 associated
with the x1y1 axes. This strain is equal to the decrease in angle between
lines in the material that were initially along the x1 and yl axes. To clarify
this idea, consider Fig. 7-33, which shows both the xy and x1y1 axes,
with the angle u between them. Let line Oa represent a line in the
material that initially was along the xl axis (that is, along the diagonal of
the element in Fig. 7-32). The deformations caused by the strains ex, ey,
and gxy (Fig. 7-32) cause line Oa to rotate through a counterclockwise
angle a from the xl axis to the position shown in Fig. 7-33. Similarly,
line Ob was originally along the yl axis, but because of the deformations
it rotates through a clockwise angle b. The shear strain gx1y1 is the
decrease in angle between the two lines that originally were at right
angles; therefore,
gx1y1 a b
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(7-67)
515
SECTION 7.7 Plane Strain
Thus, in order to find the shear strain gx1y1, we must determine the
angles a and b.
The angle a can be found from the deformations pictured in Fig.
7-32 as follows. The strain ex (Fig. 7-32a) produces a clockwise rotation
of the diagonal of the element. Let us denote this angle of rotation as a1.
The angle a1 is equal to the distance ex dx sin u divided by the length ds
of the diagonal:
dx
a1 ex ᎏᎏ sin u
ds
(f)
Similarly, the strain ey produces a counterclockwise rotation of the diagonal through an angle a2 (Fig. 7-32b). This angle is equal to the distance
ey dy cos u divided by ds:
dy
a2 ey ᎏᎏ cosu
ds
(g)
Finally, the strain gx y produces a clockwise rotation through an angle a3
(Fig. 7-32c) equal to the distance gx y dy sin u divided by ds:
dy
a3 gx y ᎏᎏ sinu
ds
(h)
Therefore, the resultant counterclockwise rotation of the diagonal (Fig.
7-32), equal to the angle a shown in Fig. 7-33, is
a a1 a2 a3
dy
dy
dx
ex sin u e y cos u gx y sin u
ds
ds
ds
(i)
Again observing that dx/ds cos u and dy/ds sin u, we obtain
a (ex e y ) sin u cos u gx y sin2 u
(7-68)
The rotation of line Ob (Fig. 7-33), which initially was at 90° to line
Oa, can be found by substituting u 90° for u in the expression for a.
The resulting expression is counterclockwise when positive (because a
is counterclockwise when positive), hence it is equal to the negative of
the angle b (because b is positive when clockwise). Thus,
b (e x e y ) sin (u 90°) cos (u 90°) gx y sin2 (u 90°)
(ex e y ) sin u cos u gx y cos2 u
(7-69)
Adding a and b gives the shear strain gx1y1 (see Eq. 7-67):
gx1y1 2(ex e y ) sin u cos u gxy (cos2 u sin2 u )
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
( j)
516
CHAPTER 7 Analysis of Stress and Strain
To put the equation in a more useful form, we divide each term by 2:
gxl yl
gxy
(ex e y) sin u cos u (cos2 u sin2 u )
2
2
(7-70)
We have now obtained an expression for the shear strain gx1y1 associated with the x1y1 axes in terms of the strains ex, ey, and gxy associated
with the xy axes.
Transformation equations for plane strain. The equations for
plane strain (Eqs. 7-66 and 7-70) can be expressed in terms of the angle
2u by using the following trigonometric identities:
1
cos 2 u ᎏᎏ (1 cos 2u)
2
1
sin2 u (1 cos 2u)
2
1
sin u cos u sin 2u
2
Thus, the transformation equations for plane strain become
ex ey
ex ey
gxy
e x1 cos 2u sin 2u
2
2
2
(7-71a)
gx y
ex ey
gxy
l l
sin 2u cos 2u
2
2
2
(7-71b)
and
TABLE 7-1 CORRESPONDING VARIABLES IN
THE TRANSFORMATION EQUATIONS FOR
PLANE STRESS (EQS. 7-4a AND b) AND
PLANE STRAIN (EQS. 7-71a AND b)
Stresses
Strains
sx
ex
sy
ey
txy
gx y /2
sx1
e x1
tx1 y1
gx1 y1/2
These equations are the counterparts of Eqs. (7-4a) and (7-4b) for plane
stress.
When comparing the two sets of equations, note that exl corresponds
to sxl, gxl yl /2 corresponds to txl yl, ex corresponds to sx, ey corresponds to
sy, and gx y /2 corresponds to txy. The corresponding variables in the two
sets of transformation equations are listed in Table 7-1.
The analogy between the transformation equations for plane stress
and those for plane strain shows that all of the observations made in Sections 7.2, 7.3, and 7.4 concerning plane stress, principal stresses,
maximum shear stresses, and Mohr’s circle have their counterparts in
plane strain. For instance, the sum of the normal strains in perpendicular
directions is a constant (compare with Eq. 7-6):
e x1 e y1 e x e y
(7-72)
This equality can be verified easily by substituting the expressions for ex1
(from Eq. 7-71a) and ey1 (from Eq. 7-71a with u replaced by u 90°).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 7.7 Plane Strain
517
Principal Strains
Principal strains exist on perpendicular planes with the principal angles
up calculated from the following equation (compare with Eq. 7-11):
gxy
tan 2up
ex ey
(7-73)
The principal strains can be calculated from the equation
ex ey
e 1,2
2
冣 冢冣
冪冢莦
2 莦莦
2
ex ey
2
gxy
2
(7-74)
which corresponds to Eq. (7-17) for the principal stresses. The two principal strains (in the xy plane) can be correlated with the two principal
directions using the technique described in Section 7.3 for the principal
stresses. (This technique is illustrated later in Example 7-7.) Finally,
note that in plane strain the third principal strain is ez 0. Also, the
shear strains are zero on the principal planes.
Maximum Shear Strains
The maximum shear strains in the xy plane are associated with axes at
45° to the directions of the principal strains. The algebraically maximum
shear strain (in the xy plane) is given by the following equation (compare with Eq. 7-25):
gmax
2
冣 冢冣
冪冢莦
2 莦莦
2
ex ey
2
gxy
2
(7-75)
The minimum shear strain has the same magnitude but is negative. In
the directions of maximum shear strain, the normal strains are
ex ey
eaver ᎏ
2
(7-76)
which is analogous to Eq. (7-27) for stresses. The maximum out-ofplane shear strains, that is, the shear strains in the xz and yz planes, can
be obtained from equations analogous to Eq. (7-75).
An element in plane stress that is oriented to the principal directions
of stress (see Fig. 7-12b) has no shear stresses acting on its faces. Therefore, the shear strain gxl yl for this element is zero. It follows that the
normal strains in this element are the principal strains. Thus, at a given
point in a stressed body, the principal strains and principal stresses
occur in the same directions.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
518
CHAPTER 7 Analysis of Stress and Strain
Mohr’s Circle for Plane Strain
Mohr’s circle for plane strain is constructed in the same manner as the
circle for plane stress, as illustrated in Fig. 7-34. Normal strain ex1 is
plotted as the abscissa (positive to the right) and one-half the shear strain
(gxl y l /2) is plotted as the ordinate (positive downward). The center C of
the circle has an abscissa equal to eaver (Eq. 7-76).
Point A, representing the strains associated with the x direction
(u 0), has coordinates ex and gxy /2. Point B, at the opposite end of a
diameter from A, has coordinates ey and ⫺gx y /2, representing the strains
associated with a pair of axes rotated through an angle u 90°.
The strains associated with axes rotated through an angle u are
given by point D, which is located on the circle by measuring an angle
2u counterclockwise from radius CA. The principal strains are represented by points Pl and P2, and the maximum shear strains by points S1
and S2. All of these strains can be determined from the geometry of the
circle or from the transformation equations.
Strain Measurements
An electrical-resistance strain gage is a device for measuring normal
strains on the surface of a stressed object. These gages are quite small,
with lengths typically in the range from one-eighth to one-half of an
inch. The gages are bonded securely to the surface of the object so that
they change in length in proportion to the strains in the object itself.
e1
ey
B(u = 90°)
S2
–
gxy
2
D'
P1
P2
O
C
2up1
2u
e2
S1
e x + ey
eaver =
2
ex
A(u = 0)
ex1
FIG. 7-34 Mohr’s circle for plane strain
gx1y1
2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
gx1y1
2
D(u = u)
ex – ey
2
ex1
gxy
2
SECTION 7.7 Plane Strain
FIG. 7-35 Three electrical-resistance
strain gages arranged as a 45° strain
rosette (magnified view). (Courtesy of
Micro-Measurements Division,
Measurements Group, Inc., Raleigh, NC,
USA.)
519
Each gage consists of a fine metal grid that is stretched or shortened
when the object is strained at the point where the gage is attached. The
grid is equivalent to a continuous wire that goes back and forth from one
end of the grid to the other, thereby effectively increasing its length
(Fig. 7-35). The electrical resistance of the wire is altered when it
stretches or shortens—then this change in resistance is converted into a
measurement of strain. The gages are extremely sensitive and can
measure strains as small as 1 106.
Since each gage measures the normal strain in only one direction,
and since the directions of the principal stresses are usually unknown, it
is necessary to use three gages in combination, with each gage measuring the strain in a different direction. From three such measurements, it
is possible to calculate the strains in any direction, as illustrated in
Example 7-8.
A group of three gages arranged in a particular pattern is called a
strain rosette. Because the rosette is mounted on the surface of the
body, where the material is in plane stress, we can use the transformation equations for plane strain to calculate the strains in various
directions. (As explained earlier in this section, the transformation equations for plane strain can also be used for the strains in plane stress.)
Calculation of Stresses from the Strains
The strain equations presented in this section are derived solely from
geometry, as already pointed out. Therefore, the equations apply to any
material, whether linear or nonlinear, elastic or inelastic. However, if it
is desired to determine the stresses from the strains, the material properties must be taken into account.
If the material follows Hooke’s law, we can find the stresses using
the appropriate stress-strain equations from either Section 7.5 (for plane
stress) or Section 7.6 (for triaxial stress).
As a first example, suppose that the material is in plane stress and
that we know the strains ex, ey, and gxy, perhaps from strain-gage measurements. Then we can use the stress-strain equations for plane stress
(Eqs. 7-36 and 7-37) to obtain the stresses in the material.
Now consider a second example. Suppose we have determined the
three principal strains e1, e 2, and e 3 for an element of material (if the
element is in plane strain, then e 3 0). Knowing these strains, we can
find the principal stresses using Hooke’s law for triaxial stress (see
Eqs. 7-54a, b, and c). Once the principal stresses are known, we can find
the stresses on inclined planes using the transformation equations for
plane stress (see the discussion at the beginning of Section 7.6).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
520
CHAPTER 7 Analysis of Stress and Strain
Example 7-7
An element of material in plane strain undergoes the following strains:
e x 340
106
e y 110
106
gx y 180
106
These strains are shown highly exaggerated in Fig. 7-36a, which shows the
deformations of an element of unit dimensions. Since the edges of the element
have unit lengths, the changes in linear dimensions have the same magnitudes as
the normal strains e x and e y. The shear strain gxy is the decrease in angle at the
lower-left corner of the element.
Determine the following quantities: (a) the strains for an element oriented
at an angle u 30°, (b) the principal strains, and (c) the maximum shear
strains. (Consider only the in-plane strains, and show all results on sketches of
properly oriented elements.)
y
y
y1
110 × 10 –6
x1
90 × 10 –6
180 × 10 –6
u = 30°
1
1
110 × 10 –6
x
O
360 × 10 –6
O
x
340 × 10 –6
(b)
(a)
y
y
y1
x1
80 × 10 –6
x1
370 × 10 –6
O
up1 = 19.0°
x
us2 = 64.0°
y1
225 × 10 –6
225 × 10 –6
O
x
290 × 10 –6
(c)
FIG. 7-36 Example 7-7. Element of
material in plane strain: (a) element
oriented to the x and y axes, (b) element
oriented at an angle u 30°,
(c) principal strains, and (d) maximum
shear strains. (Note: The edges of the
elements have unit lengths.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(d)
521
SECTION 7.7 Plane Strain
Solution
(a) Element oriented at an angle u 5 30°. The strains for an element
oriented at an angle u to the x axis can be found from the transformation equations (Eqs. 7-71a and 7-71b). As a preliminary matter, we make the following
calculations:
ex ey
(340 110)106
ᎏ 225
2
106
ex ey
(340 110)106
115
2
106
gxy
90
2
106
Now substituting into Eqs. (7-71a) and (7-71b), we get
ex ey
ex ey
gxy
e x1 cos 2u sin 2u
2
2
2
106) (115
(225
106)(cos 60°) (90
106)(sin 60°)
106
360
gx y
ex ey
gxy
11 sin 2u cos 2u
2
2
2
(115
55
106)(sin 60°) (90
106)(cos 60°)
106
Therefore, the shear strain is
106
gx1 y1 110
The strain ey1 can be obtained from Eq. (7-72), as follows:
e y1 e x e y – e x1 (340 110 360)10–6 90
10–6
The strains ex1, ey1, and gx1 y1 are shown in Fig. 7-36b for an element oriented at
u 30°. Note that the angle at the lower-left corner of the element increases
because gx1y1 is negative.
(b) Principal strains. The principal strains are readily determined from
Eq. (7-74), as follows:
ex ey
e1,2
2
e e
g
冢2莦
冣 冢2莦冣
冪莦
x
y 2
xy 2
225
106 兹(115
苶苶
106)2苶
(9苶
0 106
苶
)2
225
106 146
106
Thus, the principal strains are
e1 370
10–6
e2 80
10–6
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
522
CHAPTER 7 Analysis of Stress and Strain
y
y
y1
x1
80 × 10 –6
370 × 10 –6
O
us2 = 64.0°
y1
x1
up1 = 19.0°
225 × 10 –6
225 × 10 –6
O
x
x
290 × 10 –6
(c)
(d)
FIG. 7-36c and d (Repeated)
in which e1 denotes the algebraically larger principal strain and e2 denotes the
algebraically smaller principal strain. (Recall that we are considering only inplane strains in this example.)
The angles to the principal directions can be obtained from Eq. (7-73):
gxy
180
tan 2up ᎏᎏ ᎏ ᎏ 0.7826
ex ⫺ ey
340 ⫺ 110
The values of 2up between 0 and 360° are 38.0° and 218.0°, and therefore the
angles to the principal directions are
up 19.0° and 109.0°
To determine the value of up associated with each principal strain, we substitute up 19.0° into the first transformation equation (Eq. 7-71a) and solve
for the strain:
ex ey
ex ey
gxy
ex1 ᎏ cos 2u sin 2u
2
2
2
(225
370
106) (115
106)(cos 38.0°) (90
106)(sin 38.0°)
106
This result shows that the larger principal strain e l is at the angle up1 19.0°.
The smaller strain e 2 acts at 90° from that direction (up2 109.0°). Thus,
e 1 370
10 –6 and up1 19.0°
e 2 80
10 –6 and up2 109.0°
Note that e1 e 2 ex ey.
The principal strains are portrayed in Fig. 7-36c. There are, of course, no
shear strains on the principal planes.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 7.7 Plane Strain
523
(c) Maximum shear strain. The maximum shear strain is calculated from
Eq. (7-75):
gmax
ᎏᎏ
2
冢ᎏ2ᎏ莦
冣 莦
冢2冣莦 146
冪莦
ex ⫺ ey
2
gxy
2
10 –6
gmax 290
10 –6
The element having the maximum shear strains is oriented at 45° to the principal directions; therefore, us 19.0° 45° 64.0° and 2us 128.0°. By
substituting this value of 2us into the second transformation equation (Eq.
7-71b), we can determine the sign of the shear strain associated with this direction. The calculations are as follows:
gx y
ex ey
gxy
11 sin 2u cos 2u
2
2
2
(115
146
106)(sin 128.0°) (90
106)(cos 128.0°)
106
This result shows that an element oriented at an angle us2 64.0° has the maximum negative shear strain.
We can arrive at the same result by observing that the angle us1 to the direction of maximum positive shear strain is always 45° less than up1. Hence,
us1 up1 45° 19.0° 45° 26.0°
us2 us1 90° 64.0°
The shear strains corresponding to us1 and us2 are gmax 290
10–6 and
–6
gmin 290 10 , respectively.
The normal strains on the element having the maximum and minimum
shear strains are
ex ey
eaver 225
2
106
A sketch of the element having the maximum in-plane shear strains is shown in
Fig. 7-36d.
In this example, we solved for the strains by using the transformation equations. However, all of the results can be obtained just as easily from Mohr’s
circle.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
524
CHAPTER 7 Analysis of Stress and Strain
Example 7-8
A 45° strain rosette (also called a rectangular rosette) consists of three electrical-resistance strain gages arranged to measure strains in two perpendicular
directions and also at a 45° angle between them, as shown in Fig. 7-37a. The
rosette is bonded to the surface of the structure before it is loaded. Gages A, B,
and C measure the normal strains ea, eb, and ec in the directions of lines Oa, Ob,
and Oc, respectively.
Explain how to obtain the strains ex1, ey1, and gx1 y1 associated with an element oriented at an angle u to the xy axes (Fig. 7-37b).
Solution
At the surface of the stressed object, the material is in plane stress. Since
the strain-transformation equations (Eqs. 7-71a and 7-71b) apply to plane stress
as well as to plane strain, we can use those equations to determine the strains in
any desired direction.
Strains associated with the xy axes. We begin by determining the strains
associated with the xy axes. Because gages A and C are aligned with the x and y
axes, respectively, they give the strains ex and ey directly:
e x ea
ey ec
(7-77a,b)
To obtain the shear strain gxy, we use the transformation equation for normal
strains (Eq. 7-71a):
y
c
b
45°
ex ey
ex ey
gxy
ex1 ᎏ cos 2u sin 2u
2
2
2
B
C
45°
A
O
a
x
For an angle u 45°, we know that ex1 e b (Fig. 7-37a); therefore, the preceding equation gives
(a)
ea ec
ea ec
gxy
e b (cos 90°) (sin 90°)
2
2
2
y
y1
x1
Solving for gxy, we get
u
O
x
(b)
FIG. 7-37 Example 7-8. (a) 45° strain
rosette, and (b) element oriented at an
angle u to the xy axes
gx y 2e b e a e c
(7-78)
Thus, the strains ex, ey, and gxy are easily determined from the given strain-gage
readings.
Strains associated with the x1y1 axes. Knowing the strains ex, e y , and g xy ,
we can calculate the strains for an element oriented at any angle u (Fig. 7-37b)
from the strain-transformation equations (Eqs. 7-71a and 7-71b) or from Mohr’s
circle. We can also calculate the principal strains and the maximum shear strains
from Eqs. (7-74) and (7-75), respectively.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 7 Problems
525
PROBLEMS CHAPTER 7
Plane Stress
3,400 psi
7.2-1 An element in plane stress is subjected to stresses
sx 6500 psi, sy 1700 psi, and txy 2750 psi, as
shown in the figure.
Determine the stresses acting on an element oriented at
an angle u 60° from the x axis, where the angle u is positive when counterclockwise. Show these stresses on a
sketch of an element oriented at the angle u.
3,600 psi
9,900 psi
y
sy = 1700 psi
sx = 6500 psi
x
O
txy = 2750 psi
PROB. 7.2-3
7.2-4 The stresses acting on element A in the web of a train
rail are found to be 42 MPa tension in the horizontal
direction and 140 MPa compression in the vertical direction
(see figure). Also, shear stresses of magnitude 60 MPa act
in the directions shown.
Determine the stresses acting on an element oriented at
a counterclockwise angle of 48° from the horizontal. Show
these stresses on a sketch of an element oriented at this angle.
140 MPa
PROB. 7.2-1
A
A
60 MPa
7.2-2 Solve the preceding problem for sx 80 MPa,
sy 52 MPa, txy 48 MPa, and u 25° (see figure).
Side
View
42 MPa
Cross
Section
52 MPa
PROB. 7.2-4
48 MPa
80 MPa
7.2-5 Solve the preceding problem if the normal and shear
stresses acting on element A are 7,500 psi, 20,500 psi, and
4,800 psi (in the directions shown in the figure) and the
angle is 30° (counterclockwise).
20,500 psi
PROB. 7.2-2
A
7,500 psi
4,800 psi
7.2-3 Solve Problem 7.2-1 for sx 9,900 psi,
sy 3,400 psi, txy 3,600 psi, and u 50° (see figure).
PROB. 7.2-5
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
526
CHAPTER 7 Analysis of Stress and Strain
7.2-6 An element in plane stress from the fuselage of an
airplane is subjected to compressive stresses of magnitude
25.5 MPa in the horizontal direction and tensile stresses of
magnitude 6.5 MPa in the vertical direction (see figure).
Also, shear stresses of magnitude 12.0 MPa act in the
directions shown.
Determine the stresses acting on an element oriented at
a clockwise angle of 40° from the horizontal. Show these
stresses on a sketch of an element oriented at this angle.
12 MPa
20 MPa
B
54 MPa
6.5 MPa
PROB. 7.2-8
25.5 MPa
12.0 MPa
PROB. 7.2-6
7.2-7 The stresses acting on element B in the web of a
wide-flange beam are found to be 11,000 psi compression
in the horizontal direction and 3,000 psi compression in the
vertical direction (see figure). Also, shear stresses of magnitude 4,200 psi act in the directions shown.
Determine the stresses acting on an element oriented at
a counterclockwise angle of 41° from the horizontal. Show
these stresses on a sketch of an element oriented at this
angle.
7.2-9 The polyethylene liner of a settling pond is subjected
to stresses sx 350 psi, sy 112 psi, and txy 120 psi,
as shown by the plane-stress element in the first part of the
figure.
Determine the normal and shear stresses acting on a
seam oriented at an angle of 30° to the element, as shown
in the second part of the figure. Show these stresses on a
sketch of an element having its sides parallel and perpendicular to the seam.
y
112 psi
30°
3,000 psi
O
350 psi
x
120 psi
B
B
Seam
11,000 psi
4,200 psi
Side
View
Cross
Section
PROB. 7.2-9
PROB. 7.2-7
7.2-8 Solve the preceding problem if the normal and shear
stresses acting on element B are 54 MPa, 12 MPa, and
20 MPa (in the directions shown in the figure) and the
angle is 42.5° (clockwise).
7.2-10 Solve the preceding problem if the normal and
shear stresses acting on the element are sx 2100 kPa,
sy 300 kPa, and txy 560 kPa, and the seam is
oriented at an angle of 22.5° to the element (see figure on
the next page).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
527
CHAPTER 7 Problems
y
300 kPa
22.5°
2100 kPa
x
O
Seam
560 kPa
7.2-13 At a point on the surface of a machine the material
is in biaxial stress with sx 3600 psi and sy 1600 psi,
as shown in the first part of the figure. The second part of
the figure shows an inclined plane aa cut through the same
point in the material but oriented at an angle u.
Determine the value of the angle u between zero and
90° such that no normal stress acts on plane aa. Sketch a
stress element having plane aa as one of its sides and show
all stresses acting on the element.
y
PROB. 7.2-10
5.0 in.
is formed by welding two triangular plates (see figure). The
plate is subjected to a tensile stress of 500 psi in the long
direction and a compressive stress of 350 psi in the short
direction.
Determine the normal stress sw acting perpendicular to
the line of the weld and the shear stress tw acting parallel to
the weld. (Assume that the normal stress sw is positive
when it acts in tension against the weld and the shear stress
tw is positive when it acts counterclockwise against the
weld.)
1600 psi
7.2-11 A rectangular plate of dimensions 3.0 in.
a
u
3600 psi
O
x
a
PROB. 7.2-13
350 psi
ld
We
3 in.
500 psi
5 in.
7.2-14 Solve the preceding problem for sx 32 MPa and
sy 50 MPa (see figure).
PROB. 7.2-11
7.2-12 Solve the preceding problem for a plate of dimensions 100 mm 250 mm subjected to a compressive stress
of 2.5 MPa in the long direction and a tensile stress of
12.0 MPa in the short direction (see figure).
y
50 MPa
a
12.0 MPa
O
ld
We
PROB. 7.2-12
100 mm
250 mm
u
32 MPa
x
2.5 MPa
a
PROB. 7.2-14
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
528
CHAPTER 7 Analysis of Stress and Strain
7.2-15 An element in plane stress from the frame of a
racing car is oriented at a known angle u (see figure). On
this inclined element, the normal and shear stresses have
the magnitudes and directions shown in the figure.
Determine the normal and shear stresses acting on an
element whose sides are parallel to the xy axes; that is,
determine sx, sy, and txy. Show the results on a sketch of
an element oriented at u 0°.
y
sy
txy
sx = 2000 psi
x
O
y
4,180 psi
2,360 psi
u = 30°
15,220 psi
O
x
PROB. 7.2-15
PROB. 7.2-17
★7.2-18 The surface of an airplane wing is subjected to
plane stress with normal stresses sx and sy and shear stress
txy, as shown in the figure. At a counterclockwise angle
u 30° from the x axis the normal stress is 35 MPa tension,
and at an angle u 50° it is 10 MPa compression.
If the stress sx equals 100 MPa tension, what are the
stresses sy and txy?
y
7.2-16 Solve the preceding problem for the element shown
in the figure.
sy
y 26.7 MPa
txy
O
u = 60°
66.7 MPa
O
sx = 100 MPa
x
x
25.0 MPa
PROB. 7.2-18
7.2-19 At a point in a structure subjected to plane stress,
the stresses are sx 4000 psi, sy 2500 psi, and
txy 2800 psi (the sign convention for these stresses is
shown in Fig. 7-1). A stress element located at the same
point in the structure, but oriented at a counterclockwise
angle u1 with respect to the x axis, is subjected to the
stresses shown in the figure (sb, tb, and 2000 psi).
Assuming that the angle u1 is between zero and 90°,
calculate the normal stress sb, the shear stress tb, and the
angle u1.
★
PROB. 7.2-16
7.2-17 A plate in plane stress is subjected to normal
stresses sx and sy and shear stress txy, as shown in the figure. At counterclockwise angles u 40° and u 80° from
the x axis the normal stress is 5000 psi tension.
If the stress sx equals 2000 psi tension, what are the
stresses sy and txy?
★
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 7 Problems
y
sb
tb
2000 psi u1
O
x
529
7.3-6 An element in plane stress is subjected to stresses
sx 25.5 MPa, sy 6.5 MPa, and txy 12.0 MPa
(see the figure for Problem 7.2-6).
Determine the maximum shear stresses and associated
normal stresses and show them on a sketch of a properly
oriented element.
7.3-7 An element in plane stress is subjected to stresses
sx 11,000 psi, sy 3,000 psi, and txy 4200 psi
(see the figure for Problem 7.2-7).
Determine the maximum shear stresses and associated
normal stresses and show them on a sketch of a properly
oriented element.
PROB. 7.2-19
Principal Stresses and Maximum Shear Stresses
When solving the problems for Section 7.3, consider only
the in-plane stresses (the stresses in the xy plane).
7.3-1 An element in plane stress is subjected to stresses
sx 6500 psi, sy 1700 psi, and txy 2750 psi (see the
figure for Problem 7.2-1).
Determine the principal stresses and show them on a
sketch of a properly oriented element.
7.3-2 An element in plane stress is subjected to stresses
sx 80 MPa, sy 52 MPa, and txy 48 MPa (see the
figure for Problem 7.2-2).
Determine the principal stresses and show them on a
sketch of a properly oriented element.
7.3-3 An element in plane stress is subjected to stresses
sx 9,900 psi, sy 3,400 psi, and txy 3,600 psi
(see the figure for Problem 7.2-3).
Determine the principal stresses and show them on a
sketch of a properly oriented element.
7.3-8 An element in plane stress is subjected to stresses
sx 54 MPa, sy 12 MPa, and txy 20 MPa (see
the figure for Problem 7.2-8).
Determine the maximum shear stresses and associated
normal stresses and show them on a sketch of a properly
oriented element.
7.3-9 A shear wall in a reinforced concrete building is
subjected to a vertical uniform load of intensity q and a
horizontal force H, as shown in the first part of the figure.
(The force H represents the effects of wind and earthquake
loads.) As a consequence of these loads, the stresses at
point A on the surface of the wall have the values shown in
the second part of the figure (compressive stress equal to
1100 psi and shear stress equal to 480 psi).
(a) Determine the principal stresses and show them on
a sketch of a properly oriented element.
(b) Determine the maximum shear stresses and associated normal stresses and show them on a sketch of a
properly oriented element.
q
7.3-4 An element in plane stress is subjected to stresses
sx 42 MPa, sy 140 MPa, and txy 60 MPa (see
the figure for Problem 7.2-4).
Determine the principal stresses and show them on a
sketch of a properly oriented element.
7.3-5 An element in plane stress is subjected to stresses
sx 7,500 psi, sy 20,500 psi, and txy 4,800 psi
(see the figure for Problem 7.2-5).
Determine the maximum shear stresses and associated
normal stresses and show them on a sketch of a properly
oriented element.
1100 psi
H
480 psi
A
A
PROB. 7.3-9
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
530
CHAPTER 7 Analysis of Stress and Strain
7.3-10 A propeller shaft subjected to combined torsion and
axial thrust is designed to resist a shear stress of 63 MPa
and a compressive stress of 90 MPa (see figure).
(a) Determine the principal stresses and show them on
a sketch of a properly oriented element.
(b) Determine the maximum shear stresses and associated normal stresses and show them on a sketch of a
properly oriented element.
7.3-11 sx 3500 psi, sy 1120 psi, txy 1200 psi
7.3-12 sx 2100 kPa, sy 300 kPa, txy 560 kPa
7.3-13 sx 15,000 psi, sy 1,000 psi, txy 2,400 psi
7.3-14 sx 16 MPa, sy 96 MPa, txy 42 MPa
7.3-15 sx 3,000 psi, sy 12,000 psi,
txy 6,000 psi
7.3-16 sx 100 MPa, sy 50 MPa, txy 50 MPa
7.3-17 At a point on the surface of a machine component
the stresses acting on the x face of a stress element are sx
6500 psi and txy 2100 psi (see figure).
What is the allowable range of values for the stress sy
if the maximum shear stress is limited to t0 2900 psi?
★
y
sy
90 MPa
txy = 2100 psi
63 MPa
sx = 6500 psi
x
O
PROB. 7.3-10
7.3-11 through 7.3-16 An element in plane stress (see
figure) is subjected to stresses sx, sy, and txy.
(a) Determine the principal stresses and show them on
a sketch of a properly oriented element.
(b) Determine the maximum shear stresses and associated normal stresses and show them on a sketch of a
properly oriented element.
PROB. 7.3-17
7.3-18 At a point on the surface of a machine component
the stresses acting on the x face of a stress element are
sx 45 MPa and txy 30 MPa (see figure).
What is the allowable range of values for the stress sy
if the maximum shear stress is limited to t0 34 MPa?
★
y
y
sy
sy
txy = 30 MPa
txy
O
PROBS. 7.3-11 through 7.3-16
sx
O
x
PROB. 7.3-18
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
sx = 45 MPa
x
CHAPTER 7 Problems
7.3-19 An element in plane stress is subjected to stresses
sx 6500 psi and txy 2800 psi (see figure). It is
known that one of the principal stresses equals 7300 psi in
tension.
(a) Determine the stress sy.
(b) Determine the other principal stress and the orientation of the principal planes; then show the principal
stresses on a sketch of a properly oriented element.
★
y
531
Mohr’s Circle
The problems for Section 7.4 are to be solved using Mohr’s
circle. Consider only the in-plane stresses (the stresses in
the xy plane).
7.4-1 An element in uniaxial stress is subjected to tensile
stresses sx 14,500 psi, as shown in the figure.
Using Mohr’s circle, determine (a) the stresses acting
on an element oriented at a counterclockwise angle u 24°
from the x axis and (b) the maximum shear stresses and
associated normal stresses. Show all results on sketches of
properly oriented elements.
sy
y
6500 psi
O
x
14,500 psi
O
2800 psi
x
PROB. 7.4-1
PROB. 7.3-19
7.3-20 An element in plane stress is subjected to stresses
★
sx 68.5 MPa and txy 39.2 MPa (see figure). It is
known that one of the principal stresses equals 26.3 MPa in
tension.
(a) Determine the stress sy.
(b) Determine the other principal stress and the orientation of the principal planes; then show the principal
stresses on a sketch of a properly oriented element.
7.4-2 An element in uniaxial stress is subjected to tensile
stresses sx 55 MPa, as shown in the figure.
Using Mohr’s circle, determine (a) the stresses acting
on an element oriented at an angle u 30° from the
x axis (minus means clockwise) and (b) the maximum shear
stresses and associated normal stresses. Show all results on
sketches of properly oriented elements.
y
y
sy
O
55 MPa
x
39.2 MPa
O
PROB. 7.3-20
68.5 MPa
x
PROB. 7.4-2
7.4-3 An element in uniaxial stress is subjected to compressive stresses of magnitude 5600 psi, as shown in the
figure on the next page.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
532
CHAPTER 7 Analysis of Stress and Strain
Using Mohr’s circle, determine (a) the stresses acting
on an element oriented at a slope of 1 on 2 (see figure) and
(b) the maximum shear stresses and associated normal
stresses. Show all results on sketches of properly oriented
elements.
y
1500 psi
6000 psi
y
O
x
1
2
O
5600 psi
x
PROB. 7.4-5
PROB. 7.4-3
7.4-4 An element in biaxial stress is subjected to stresses
sx 60 MPa and sy 20 MPa, as shown in the figure.
Using Mohr’s circle, determine (a) the stresses acting
on an element oriented at a counterclockwise angle u
22.5° from the x axis and (b) the maximum shear stresses
and associated normal stresses. Show all results on sketches
of properly oriented elements.
7.4-6 An element in biaxial stress is subjected to stresses
sx 24 MPa and sy 63 MPa, as shown in the figure.
Using Mohr’s circle, determine (a) the stresses acting
on an element oriented at a slope of 1 on 2.5 (see figure)
and (b) the maximum shear stresses and associated normal
stresses. Show all results on sketches of properly oriented
elements.
y
y
20 MPa
63 MPa
1
60 MPa
O
2.5
x
O
24 MPa x
PROB. 7.4-4
PROB. 7.4-6
7.4-5 An element in biaxial stress is subjected to stresses
sx 6000 psi and sy 1500 psi, as shown in the figure.
Using Mohr’s circle, determine (a) the stresses acting
on an element oriented at a counterclockwise angle u 60°
from the x axis and (b) the maximum shear stresses and
associated normal stresses. Show all results on sketches of
properly oriented elements.
7.4-7 An element in pure shear is subjected to stresses
txy 3000 psi, as shown in the figure on the next page.
Using Mohr’s circle, determine (a) the stresses acting
on an element oriented at a counterclockwise angle
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 7 Problems
u 70° from the x axis and (b) the principal stresses. Show
all results on sketches of properly oriented elements.
y
3000 psi
O
533
7.4-10 through 7.4-15 An element in plane stress is subjected to stresses sx, sy, and txy (see figure).
Using Mohr’s circle, determine the stresses acting on
an element oriented at an angle u from the x axis. Show
these stresses on a sketch of an element oriented at the
angle u. (Note: The angle u is positive when counterclockwise and negative when clockwise.)
x
y
sy
PROB. 7.4-7
txy
7.4-8 An element in pure shear is subjected to stresses
txy 16 MPa, as shown in the figure.
Using Mohr’s circle, determine (a) the stresses acting
on an element oriented at a counterclockwise angle u 20°
from the x axis and (b) the principal stresses. Show all
results on sketches of properly oriented elements.
O
sx
x
y
PROBS. 7.4-10 through 7.4-15
O
7.4-10 sx 21 MPa, sy 11 MPa, txy 8 MPa, u 50°
x
16 MPa
7.4-11 sx 4,500 psi, sy 14,100 psi, txy 3,100 psi,
PROB. 7.4-8
7.4-9 An element in pure shear is subjected to stresses
txy 4000 psi, as shown in the figure.
Using Mohr’s circle, determine (a) the stresses acting
on an element oriented at a slope of 3 on 4 (see figure) and
(b) the principal stresses. Show all results on sketches of
properly oriented elements.
u 55°
7.4-12 sx 44 MPa, sy 194 MPa, txy 36 MPa,
u 35°
7.4-13 sx 1520 psi, sy 480 psi, txy 280 psi,
y
3
u 18°
4
7.4-14 sx 31 MPa, sy 5 MPa, txy 33 MPa,
O
x
u 45°
4000 psi
7.4-15 sx 5750 psi, sy 750 psi, txy 2100 psi,
PROB. 7.4-9
u 75°
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
534
CHAPTER 7 Analysis of Stress and Strain
7.4-16 through 7.4-23 An element in plane stress is subjected to stresses sx, sy, and txy (see figure).
Using Mohr’s circle, determine (a) the principal
stresses and (b) the maximum shear stresses and associated
normal stresses. Show all results on sketches of properly
oriented elements.
y
in the figure. Strain gages A and B, oriented in the x and
y directions, respectively, are attached to the plate. The
gage readings give normal strains ex 0.0010 (elongation)
and ey 0.0007 (shortening).
Knowing that E 30 106 psi and n 0.3, determine the stresses sx and sy and the change t in the
thickness of the plate.
sy
sy
y
txy
O
B
sx
A
O
sx
x
x
PROBS. 7.5-1 and 7.5-2
7.5-2 Solve the preceding problem if the thickness of
PROBS. 7.4-16 through 7.4-23
7.4-16 sx 31.5 MPa, sy 31.5 MPa, txy 30 MPa
7.4-17 sx 8400 psi, sy 0, txy 1440 psi
7.4-18 sx 0, sy 22.4 MPa, txy 6.6 MPa
7.4-19 sx 1850 psi, sy 6350 psi, txy 3000 psi
7.4-20 sx 3100 kPa, sy 8700 kPa, txy 4500 kPa
the steel plate is t 10 mm, the gage readings are ex
480
106 (elongation) and ey 130 106 (elongation), the modulus is E 200 GPa, and Poisson’s ratio is
n 0.30.
7.5-3 Assume that the normal strains ex and ey for an
element in plane stress (see figure) are measured with
strain gages.
(a) Obtain a formula for the normal strain ez in the
z direction in terms of ex, ey, and Poisson’s ratio n.
(b) Obtain a formula for the dilatation e in terms of
ex, ey, and Poisson’s ratio n.
y
7.4-21 sx 12,300 psi, sy 19,500 psi, txy
sy
7700 psi
txy
sx
7.4-22 sx 3.1 MPa, sy 7.9 MPa, txy 13.2 MPa
O
x
7.4-23 sx 700 psi, sy 2500 psi, txy 3000 psi
z
Hooke’s Law for Plane Stress
When solving the problems for Section 7.5, assume that the
material is linearly elastic with modulus of elasticity E and
Poisson’s ratio n.
7.5-1 A rectangular steel plate with thickness t 0.25 in.
is subjected to uniform normal stresses sx and sy, as shown
PROB. 7.5-3
7.5-4 A magnesium plate in biaxial stress is subjected
to tensile stresses sx 24 MPa and sy 12 MPa (see
figure). The corresponding strains in the plate are ex
440 106 and ey 80 106.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 7 Problems
Determine Poisson’s ratio n and the modulus of elasticity E for the material.
sy
y
O
106 psi,
n 0.1) is compressed in biaxial stress by means of a
framework that is loaded as shown in the figure.
Assuming that each load F equals 20 k, determine the
change V in the volume of the cube and the strain energy
U stored in the cube.
7.5-9 A 4.0-inch cube of concrete (E 3.0
sx
x
535
;;
@@
ÀÀ
@@
ÀÀ
;;
F
PROBS. 7.5-4 through 7.5-7
7.5-5 Solve the preceding problem for a steel plate with
sx 10,800 psi (tension), sy 5,400 psi (compression),
ex 420 106 (elongation), and ey 300 106
(shortening).
7.5-6 A rectangular plate in biaxial stress (see figure) is
subjected to normal stresses sx 90 MPa (tension) and
sy 20 MPa (compression). The plate has dimensions
400 800 20 mm and is made of steel with
E 200 GPa and n 0.30.
(a) Determine the maximum in-plane shear strain gmax
in the plate.
(b) Determine the change t in the thickness of the
plate.
(c) Determine the change V in the volume of the
plate.
7.5-7 Solve the preceding problem for an aluminum plate
with sx 12,000 psi (tension), sy 3,000 psi (compression), dimensions 20 30 0.5 in., E 10.5 106 psi,
and n 0.33.
F
PROB. 7.5-9
7.5-10 A square plate of width b and thickness t is loaded
by normal forces Px and Py, and by shear forces V, as
shown in the figure. These forces produce uniformly
distributed stresses acting on the side faces of the plate.
Calculate the change V in the volume of the plate and
the strain energy U stored in the plate if the dimensions
are b 600 mm and t 40 mm, the plate is made of
magnesium with E 45 GPa and n 0.35, and the forces
are Px 480 kN, Py 180 kN, and V 120 kN.
Py
7.5-8 A brass cube 50 mm on each edge is compressed in
two perpendicular directions by forces P 175 kN (see
figure).
Calculate the change V in the volume of the cube and
the strain energy U stored in the cube, assuming
E 100 GPa and n 0.34.
t
V
y
Px
V
P = 175 kN
b O
b
x
V
Px
V
Py
P = 175 kN
PROB. 7.5-8
PROBS. 7.5-10 and 7.5-11
7.5-11 Solve the preceding problem for an aluminum plate
with b 12 in., t 1.0 in., E 10,600 ksi, n 0.33, Px
90 k, Py 20 k, and V 15 k.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
536
CHAPTER 7 Analysis of Stress and Strain
★7.5-12 A circle of diameter d 200 mm is etched on a
brass plate (see figure). The plate has dimensions 400
400 20 mm. Forces are applied to the plate, producing
uniformly distributed normal stresses sx 42 MPa and
sy 14 MPa.
Calculate the following quantities: (a) the change in
length ac of diameter ac; (b) the change in length bd of
diameter bd; (c) the change t in the thickness of the plate;
(d) the change V in the volume of the plate, and (e) the
strain energy U stored in the plate. (Assume E 100 GPa
and n 0.34.)
z
y
sy
d
sx
a
7.6-2 Solve the preceding problem if the element is steel
(E 200 GPa, n 0.30) with dimensions a 300 mm,
b 150 mm, and c 150 mm and the stresses are
sx 60 MPa, sy 40 MPa, and sz 40 MPa.
7.6-3 A cube of cast iron with sides of length a 4.0 in.
(see figure) is tested in a laboratory under triaxial stress.
Gages mounted on the testing machine show that the compressive strains in the material are ex 225 106 and
ey ez 37.5 106.
Determine the following quantities: (a) the normal
stresses sx, sy, and sz acting on the x, y, and z faces of the
cube; (b) the maximum shear stress tmax in the material;
(c) the change V in the volume of the cube; and (d) the
strain energy U stored in the cube. (Assume E 14,000 ksi
and n 0.25.)
y
c
sx
a
a
b
x
a
sy
O
x
PROB. 7.5-12
Triaxial Stress
When solving the problems for Section 7.6, assume that the
material is linearly elastic with modulus of elasticity E and
Poisson’s ratio n.
7.6-1 An element of aluminum in the form of a rectangular
parallelepiped (see figure) of dimensions a 6.0 in., b
4.0 in, and c 3.0 in. is subjected to triaxial stresses
sx 12,000 psi, sy 4,000 psi, and sz 1,000 psi
acting on the x, y, and z faces, respectively.
Determine the following quantities: (a) the maximum
shear stress tmax in the material; (b) the changes a, b,
and c in the dimensions of the element; (c) the change V
in the volume; and (d) the strain energy U stored in the
element. (Assume E 10,400 ksi and n 0.33.)
y
z
PROBS. 7.6-3 and 7.6-4
7.6-4 Solve the preceding problem if the cube is granite
(E 60 GPa, n 0.25) with dimensions a 75 mm and
compressive strains ex 720 106 and e y e z
270 106.
7.6-5 An element of aluminum in triaxial stress (see
figure) is subjected to stresses sx 5200 psi (tension),
sy 4750 psi (compression), and sz 3090 psi (compression). It is also known that the normal strains in the
x and y directions are ex 713.8 106 (elongation) and
ey 502.3 106 (shortening).
What is the bulk modulus K for the aluminum?
y
a
c
sy
sz
b
O
z
PROBS. 7.6-1 and 7.6-2
sx
sx
O
x
x
sz
z
PROBS. 7.6-5 and 7.6-6
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
sy
CHAPTER 7 Problems
7.6-6 Solve the preceding problem if the material is nylon
subjected to compressive stresses sx 4.5 MPa, sy
3.6 MPa, and sz 2.1 MPa, and the normal strains are
ex 740 106 and ey 320 106 (shortenings).
7.6-7 A rubber cylinder R of length L and cross-sectional
area A is compressed inside a steel cylinder S by a force F
that applies a uniformly distributed pressure to the rubber
(see figure).
(a) Derive a formula for the lateral pressure p between
the rubber and the steel. (Disregard friction between the
rubber and the steel, and assume that the steel cylinder is
rigid when compared to the rubber.)
(b) Derive a formula for the shortening d of the rubber
cylinder.
F
R
L
PROB. 7.6-7
7.6-8 A block R of rubber is confined between plane
parallel walls of a steel block S (see figure). A uniformly
distributed pressure p0 is applied to the top of the rubber
block by a force F.
(a) Derive a formula for the lateral pressure p between
the rubber and the steel. (Disregard friction between the
rubber and the steel, and assume that the steel block is rigid
when compared to the rubber.)
(b) Derive a formula for the dilatation e of the rubber.
(c) Derive a formula for the strain-energy density u of
the rubber.
F
PROB. 7.6-8
7.6-10 A solid steel sphere (E 210 GPa, n 0.3) is sub-
jected to hydrostatic pressure p such that its volume is
reduced by 0.4%.
(a) Calculate the pressure p.
(b) Calculate the volume modulus of elasticity K for
the steel.
(c) Calculate the strain energy U stored in the sphere if
its diameter is d 150 mm.
7.6-11 A solid bronze sphere (volume modulus of elasticity
K 14.5 106 psi) is suddenly heated around its outer
surface. The tendency of the heated part of the sphere to
expand produces uniform tension in all directions at the
center of the sphere.
If the stress at the center is 12,000 psi, what is the
strain? Also, calculate the unit volume change e and the
strain-energy density u at the center.
Plane Strain
When solving the problems for Section 7.7, consider only
the in-plane strains (the strains in the xy plane) unless
stated otherwise. Use the transformation equations of plane
strain except when Mohr’s circle is specified (Problems
7.7-23 through 7.7-28).
F
S
S
106 psi,
n 0.34) is lowered into the ocean to a depth of 10,000 ft.
The diameter of the ball is 11.0 in.
Determine the decrease d in diameter, the decrease
V in volume, and the strain energy U of the ball.
7.6-9 A solid spherical ball of brass (E 15
F
S
S
537
R
7.7-1 A thin rectangular plate in biaxial stress is subjected
to stresses sx and sy, as shown in part (a) of the figure on
the next page. The width and height of the plate are b
8.0 in. and h 4.0 in., respectively. Measurements show
that the normal strains in the x and y directions are ex
195 106 and e y 125 106, respectively.
With reference to part (b) of the figure, which shows a
two-dimensional view of the plate, determine the following
quantities: (a) the increase d in the length of diagonal Od;
(b) the change f in the angle f between diagonal Od and
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
538
CHAPTER 7 Analysis of Stress and Strain
the x axis; and (c) the change c in the angle c between
diagonal Od and the y axis.
y
845
106, and ey 211
106.
7.7-5 An element of material subjected to plane strain
(see figure) has strains as follows: ex 220 106, ey
480 106, and gxy 180 106.
Calculate the strains for an element oriented at an
angle u 50° and show these strains on a sketch of a properly oriented element.
sy
sx
h
7.7-4 Solve the preceding problem if b 225 mm, ex
b
y
x
z
ey
(a)
y
gxy
1
d
c
h
O
f
O
x
b
(b)
106, and ey 320
106.
stresses sx and sy, as shown in part (a) of the figure. The
width of the plate is b 12.0 in. Measurements show that
the normal strains in the x and y directions are ex
427 106 and ey 113 106, respectively.
With reference to part (b) of the figure, which shows a
two-dimensional view of the plate, determine the following
quantities: (a) the increase d in the length of diagonal Od;
(b) the change f in the angle f between diagonal Od and
the x axis; and (c) the shear strain g associated with diagonals
Od and cf (that is, find the decrease in angle ced ).
y
y
sy
c
e
b
f
x
z
(a)
PROBS. 7.7-3 and 7.7-4
d
b
O
b
(b)
e x 420 106, e y 170
and u 37.5°.
106, gx y 310
106,
7.7-7 The strains for an element of material in plane strain
7.7-3 A thin square plate in biaxial stress is subjected to
sx
x
7.7-6 Solve the preceding problem for the following data:
7.7-2 Solve the preceding problem if b 160 mm, h
b
ex
PROBS. 7.7-5 through 7.7-10
PROBS. 7.7-1 and 7.7-2
60 mm, ex 410
1
106, ey
(see figure) are as follows: ex 480
140 106, and gxy 350 106.
Determine the principal strains and maximum shear
strains, and show these strains on sketches of properly
oriented elements.
7.7-8 Solve the preceding problem for the following
106, ey 450
106, and gxy
strains: ex 120
360 10–6.
7.7-9 An element of material in plane strain (see figure) is
subjected to strains ex 480 106, ey 70 106, and
gxy 420 106.
Determine the following quantities: (a) the strains for
an element oriented at an angle u 75°, (b) the principal
strains, and (c) the maximum shear strains. Show the results
on sketches of properly oriented elements.
7.7-10 Solve the preceding problem for the following data:
f
x
106, ey 430
ex 1120
6
780 10 , and u 45°.
106, gxy
7.7-11 A steel plate with modulus of elasticity E 30
106 psi and Poisson’s ratio n 0.30 is loaded in biaxial
stress by normal stresses sx and sy (see figure). A strain
gage is bonded to the plate at an angle f 30°.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
539
CHAPTER 7 Problems
If the stress sx is 18,000 psi and the strain measured by
the gage is e 407 106, what is the maximum in-plane
shear stress (tmax)xy and shear strain (gmax)xy? What is the
maximum shear strain (gmax)xz in the xz plane? What is the
maximum shear strain (gmax)yz in the yz plane?
y
7.7-15 During a test of an airplane wing, the strain gage
readings from a 45° rosette (see figure) are as follows:
106; and gage C,
gage A, 520
106; gage B, 360
6
80 10 .
Determine the principal strains and maximum shear
strains, and show them on sketches of properly oriented
elements.
sy
y
45°
sx
f
B
C
x
45°
z
A
x
O
PROBS. 7.7-11 and 7.7-12
PROBS. 7.7-15 and 7.7-16
7.7-12 Solve the preceding problem if the plate is made
of aluminum with E 72 GPa and n 1/3, the stress
sx is 86.4 MPa, the angle f is 21°, and the strain e is
946 106.
7.7-13 An element in plane stress is subjected to
stresses sx 8400 psi, sy 1100 psi, and txy
1700 psi (see figure). The material is aluminum with
modulus of elasticity E 10,000 ksi and Poisson’s
ratio n 0.33.
Determine the following quantities: (a) the strains for
an element oriented at an angle u 30°, (b) the principal
strains, and (c) the maximum shear strains. Show the results
on sketches of properly oriented elements.
sy
surface of an automobile frame gives the following
106; and
readings: gage A, 310
106; gage B, 180
6
gage C, 160 10 .
Determine the principal strains and maximum shear
strains, and show them on sketches of properly oriented
elements.
7.7-17 A solid circular bar of diameter d 1.5 in. is subjected to an axial force P and a torque T (see figure). Strain
gages A and B mounted on the surface of the bar give read10–6 and eb 55
106. The bar is
ings ea 100
6
made of steel having E 30 10 psi and n 0.29.
(a) Determine the axial force P and the torque T.
(b) Determine the maximum shear strain gmax and the
maximum shear stress tmax in the bar.
txy
y
O
7.7-16 A 45° strain rosette (see figure) mounted on the
x
d
sx
T
P
C
PROBS. 7.7-13 and 7.7-14
B
7.7-14 Solve the preceding problem for the following data:
sx 150 MPa, sy 210 MPa, txy 16 MPa, and
u 50°. The material is brass with E 100 GPa and n
0.34.
C
PROB. 7.7-17
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
45°
A
540
CHAPTER 7 Analysis of Stress and Strain
7.7-18 A cantilever beam of rectangular cross section
(width b 25 mm, height h 100 mm) is loaded by a
force P that acts at the midheight of the beam and is
inclined at an angle a to the vertical (see figure). Two strain
gages are placed at point C, which also is at the midheight
of the beam. Gage A measures the strain in the horizontal
direction and gage B measures the strain at an angle b
60° to the horizontal. The measured strains are ea 125
106 and eb 375 106.
Determine the force P and the angle a, assuming the
material is steel with E 200 GPa and n 1/3.
7.7-21 On the surface of a structural component in a space
vehicle, the strains are monitored by means of three strain
gages arranged as shown in the figure. During a certain
maneuver, the following strains were recorded: ea
1100 106, eb 200 106, and ec 200 10–6.
Determine the principal strains and principal stresses in
the material, which is a magnesium alloy for which E
6000 ksi and n 0.35. (Show the principal strains and principal stresses on sketches of properly oriented elements.)
y
h
b
C
B
h
C
a
P
30°
b
O
x
A
PROB. 7.7-21
B
b
A
C
PROBS. 7.7-18 and 7.7-19
7.7-19 Solve the preceding problem if the cross-sectional
dimensions are b 1.0 in. and h 3.0 in., the gage angle
is b 75°, the measured strains are ea 171 106 and
eb 266 106, and the material is a magnesium alloy
with modulus E 6.0
106 psi and Poisson’s ratio n
0.35.
7.7-22 The strains on the surface of an experimental device
made of pure aluminum (E 70 GPa, n 0.33) and tested
in a space shuttle were measured by means of strain gages.
The gages were oriented as shown in the figure, and the
106, eb 1496
measured strains were ea 1100
6
–6
10 , and ec 39.44 10 .
What is the stress sx in the x direction?
y
B
7.7-20 A 60° strain rosette, or delta rosette, consists of
three electrical-resistance strain gages arranged as shown in
the figure. Gage A measures the normal strain ea in the
direction of the x axis. Gages B and C measure the strains eb
and ec in the inclined directions shown.
Obtain the equations for the strains ex, ey, and gxy associated with the xy axes.
y
B
60°
O
PROB. 7.7-20
60°
A
C
60°
x
O
40°
C
A
40°
x
PROB. 7.7-22
7.7-23 Solve
plane strain.
7.7-24 Solve
plane strain.
7.7-25 Solve
plane strain.
7.7-26 Solve
plane strain.
7.7-27 Solve
plane strain.
7.7-28 Solve
plane strain.
Problem 7.7-5 by using Mohr’s circle for
Problem 7.7-6 by using Mohr’s circle for
Problem 7.7-7 by using Mohr’s circle for
Problem 7.7-8 by using Mohr’s circle for
Problem 7.7-9 by using Mohr’s circle for
Problem 7.7-10 by using Mohr’s circle for
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
8
Applications of Plane Stress
(Pressure Vessels, Beams,
and Combined Loadings
8.1 INTRODUCTION
In the preceding chapter we analyzed the stresses and strains at a point
in a structure subjected to plane stress (see Sections 7.2 through 7.5).
Plane stress is a common stress condition that exists in all ordinary
structures, including buildings, machines, vehicles, and aircraft. In this
chapter we examine some practical applications involving plane stress,
and therefore the material presented in Sections 7.2 through 7.5 should
be thoroughly understood before proceeding.
The first structures to be discussed are pressure vessels, such as
compressed-air tanks and water pipes (Sections 8.2 and 8.3). We will
determine the stresses and strains in the walls of these structures due to
the internal pressures from the compressed gases or liquids. Next, we
return to the subject of stresses in beams and show how to determine
the principal stresses and maximum shear stresses at various points
(Section 8.4). Finally, in Section 8.5, we analyze structures subjected to
combined loadings, that is, combinations of axial loads, torsional loads,
bending loads, and internal pressure. Our objective is to determine the
maximum normal and shear stresses at various points in these structures.
8.2 SPHERICAL PRESSURE VESSELS
Pressure vessels are closed structures containing liquids or gases under
pressure. Familiar examples include tanks, pipes, and pressurized cabins
in aircraft and space vehicles. When pressure vessels have walls that are
thin in comparison to their overall dimensions, they are included within
a more general category known as shell structures. Other examples of
shell structures are roof domes, airplane wings, and submarine hulls.
541
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
542
CHAPTER 8 Applications of Plane Stress
Welded seam
FIG. 8-1 Spherical pressure vessel
t
r
p
In this section we consider thin-walled pressure vessels of spherical
shape, like the compressed-air tank shown in Fig. 8-1. The term thinwalled is not precise, but as a general rule, pressure vessels are
considered to be thin-walled when the ratio of radius r to wall thickness t
(Fig. 8-2) is greater than 10. When this condition is met, we can determine the stresses in the walls with reasonable accuracy using statics
alone.
We assume in the following discussions that the internal pressure p
(Fig. 8-2) exceeds the pressure acting on the outside of the shell. Otherwise, the vessel may collapse inward due to buckling.
A sphere is the theoretically ideal shape for a vessel that resists
internal pressure. We only need to contemplate the familiar soap bubble
to recognize that a sphere is the “natural” shape for this purpose. To
determine the stresses in a spherical vessel, let us cut through the sphere
on a vertical diametral plane (Fig. 8-3a) and isolate half of the shell and
its fluid contents as a single free body (Fig. 8-3b). Acting on this free
body are the tensile stresses s in the wall of the vessel and the fluid
pressure p. This pressure acts horizontally against the plane circular area
of fluid remaining inside the hemisphere. Since the pressure is uniform,
the resultant pressure force P (Fig. 8-3b) is
P p(p r 2)
FIG. 8-2 Cross section of spherical
pressure vessel showing inner radius r,
wall thickness t, and internal pressure p
(a)
where r is the inner radius of the sphere.
Note that the pressure p is not the absolute pressure inside the vessel
but is the net internal pressure, or the gage pressure. Gage pressure is
the internal pressure above the pressure acting on the outside of the vessel. If the internal and external pressures are the same, no stresses are
pr
s=—
2t
s
s
P = ppr2
s
s
(a)
(b)
FIG. 8-3 Tensile stresses s in the wall of
a spherical pressure vessel
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(c)
SECTION 8.2 Spherical Pressure Vessels
543
developed in the wall of the vessel—only the excess of internal pressure
over external pressure has any effect on these stresses.
Because of the symmetry of the vessel and its loading (Fig. 8-3b),
the tensile stress s is uniform around the circumference. Furthermore,
since the wall is thin, we can assume with good accuracy that the stress
is uniformly distributed across the thickness t. The accuracy of this
approximation increases as the shell becomes thinner and decreases as it
becomes thicker.
The resultant of the tensile stresses s in the wall is a horizontal
force equal to the stress s times the area over which it acts, or
s (2p rm t)
where t is the thickness of the wall and rm is its mean radius:
t
rm r
2
(b)
Thus, equilibrium of forces in the horizontal direction (Fig. 8-3b) gives
Fhoriz 0
s (2prm t) p(pr 2) 0
(c)
from which we obtain the tensile stresses in the wall of the vessel:
pr 2
s ᎏᎏ
2rm t
(d)
Since our analysis is valid only for thin shells, we can disregard the
small difference between the two radii appearing in Eq. (d) and replace r
by rm or replace rm by r. While either choice is satisfactory for this
approximate analysis, it turns out that the stresses are closer to the theoretically exact stresses if we use the inner radius r instead of the mean
radius rm. Therefore, we will adopt the following formula for calculating
the tensile stresses in the wall of a spherical shell:
pr
s ᎏᎏ
2t
(8-1)
As is evident from the symmetry of a spherical shell, we obtain the same
equation for the tensile stresses when we cut a plane through the center
of the sphere in any direction whatsoever. Thus, we reach the following
conclusion: The wall of a pressurized spherical vessel is subjected to
uniform tensile stresses s in all directions. This stress condition is represented in Fig. 8-3c by the small stress element with stresses s acting in
mutually perpendicular directions.
Stresses that act tangentially to the curved surface of a shell, such as
the stresses s shown in Fig. 8-3c, are known as membrane stresses.
The name arises from the fact that these are the only stresses that exist in
true membranes, such as soap films.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
544
CHAPTER 8 Applications of Plane Stress
Stresses at the Outer Surface
y
sy = s
sx = s
sx = s
O
x
z
sy = s
(a)
y
sy = s
sx = s
sz = – p
z
sx1 s and tx1y1 0
as expected. In other words, when we consider elements obtained by
rotating the axes about the z axis, the normal stresses remain constant
and there are no shear stresses. Every plane is a principal plane and
every direction is a principal direction. Thus, the principal stresses for
the element are
pr
s1 s2 ᎏᎏ
2t
sx = s
O
The outer surface of a spherical pressure vessel is usually free of any
loads. Therefore, the element shown in Fig. 8-3c is in biaxial stress. To
aid in analyzing the stresses acting on this element, we show it again in
Fig. 8-4a, where a set of coordinate axes is oriented parallel to the sides
of the element. The x and y axes are tangential to the surface of the
sphere, and the z axis is perpendicular to the surface. Thus, the normal
stresses sx and sy are the same as the membrane stresses s, and the
normal stress sz is zero. No shear stresses act on the sides of this element.
If we analyze the element of Fig. 8-4a by using the transformation
equations for plane stress (see Fig. 7-1 and Eqs. 7-4a and 7-4b of Section 7.2), we find
x
sy = s
(b)
FIG. 8-4 Stresses in a spherical pressure
vessel at (a) the outer surface and (b) the
inner surface
s3 0
(8-2a,b)
The stresses s1 and s2 lie in the xy plane and the stress s3 acts in the z
direction.
To obtain the maximum shear stresses, we must consider out-ofplane rotations, that is, rotations about the x and y axes (because all
in-plane shear stresses are zero). Elements oriented by making 45°
rotations about the x and y axes have maximum shear stresses equal to
s/2 and normal stresses equal to s/2. Therefore,
pr
s
tmax ᎏᎏ ᎏᎏ
2
4t
(8-3)
These stresses are the largest shear stresses in the element.
Stresses at the Inner Surface
At the inner surface of the wall of a spherical vessel, a stress element
(Fig. 8-4b) has the same membrane stresses s x and sy as does an
element at the outer surface (Fig. 8-4a). In addition, a compressive stress
s z equal to the pressure p acts in the z direction (Fig. 8-4b). This compressive stress decreases from p at the inner surface of the sphere to zero
at the outer surface.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 8.2 Spherical Pressure Vessels
545
The element shown in Fig. 8-4b is in triaxial stress with principal
stresses
pr
s 1 s 2 ᎏᎏ
2t
(e,f)
s3 p
The in-plane shear stresses are zero, but the maximum out-of-plane
shear stress (obtained by a 45° rotation about either the x or y axis) is
冢
sp
pr
p
p r
tmax ᎏ 1
2
4t
2
2 2t
(g)
When the vessel is thin-walled and the ratio r/t is large, we can disregard
the number 1 in comparison with the term r/2t. In other words, the principal stress s3 in the z direction is small when compared with the
principal stresses s1 and s2. Consequently, we can consider the stress
state at the inner surface to be the same as at the outer surface (biaxial
stress). This approximation is consistent with the approximate nature of
thin-shell theory, and therefore we will use Eqs. (8-1), (8-2), and (8-3) to
obtain the stresses in the wall of a spherical pressure vessel.
General Comments
Pressure vessels usually have openings in their walls (to serve as inlets
and outlets for the fluid contents) as well as fittings and supports that
exert forces on the shell (Fig. 8-1). These features result in nonuniformities in the stress distribution, or stress concentrations, that cannot
be analyzed by the elementary formulas given here. Instead, more
advanced methods of analysis are needed. Other factors that affect the
design of pressure vessels include corrosion, accidental impacts, and
temperature changes.
Some of the limitations of thin-shell theory as applied to pressure
vessels are listed here:
1. The wall thickness must be small in comparison to the other dimensions (the ratio r/t should be 10 or more).
2. The internal pressure must exceed the external pressure (to avoid
inward buckling).
3. The analysis presented in this section is based only on the effects of
internal pressure (the effects of external loads, reactions, the weight
of the contents, and the weight of the structure are not considered).
4. The formulas derived in this section are valid throughout the wall of
the vessel except near points of stress concentrations.
The following example illustrates how the principal stresses and
maximum shear stresses are used in the analysis of a spherical shell.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
546
CHAPTER 8 Applications of Plane Stress
Example 8-1
A compressed-air tank having an inner diameter of 18 inches and a wall
thickness of 1/4 inch is formed by welding two steel hemispheres (Fig. 8-5).
(a) If the allowable tensile stress in the steel is 14,000 psi, what is the
maximum permissible air pressure pa in the tank?
(b) If the allowable shear stress in the steel is 6,000 psi, what is the
maximum permissible pressure pb?
(c) If the normal strain at the outer surface of the tank is not to exceed
0.0003, what is the maximum permissible pressure pc? (Assume that Hooke’s
law is valid and that the modulus of elasticity for the steel is 29 106 psi and
Poisson’s ratio is 0.28.)
(d) Tests on the welded seam show that failure occurs when the tensile
load on the welds exceeds 8.1 kips per inch of weld. If the required factor of
safety against failure of the weld is 2.5, what is the maximum permissible
pressure pd?
(e) Considering the four preceding factors, what is the allowable pressure
pallow in the tank?
Weld
Solution
FIG. 8-5 Example 8-1. Spherical pressure
vessel. (Attachments and supports are
not shown.)
(a) Allowable pressure based upon the tensile stress in the steel. The maximum tensile stress in the wall of the tank is given by the formula s pr/2t (see
Eq. 8-1). Solving this equation for the pressure in terms of the allowable stress,
we get
2tsallow
2(0.25 in.)(14,000 psi)
pa ᎏᎏᎏ 777.8 psi
r
Thus, the maximum allowable pressure based upon tension in the wall of the
tank is pa 777 psi. (Note that in a calculation of this kind, we round downward, not upward.)
(b) Allowable pressure based upon the shear stress in the steel. The maximum shear stress in the wall of the tank is given by Eq. (8-3), from which we
get the following equation for the pressure:
4ttallow
4(0.25 in.)(6,000 psi)
pb ᎏᎏᎏ 666.7 psi
r
Therefore, the allowable pressure based upon shear is pb 666 psi.
(c) Allowable pressure based upon the normal strain in the steel. The
normal strain is obtained from Hooke’s law for biaxial stress (Eq. 7-39a):
1
ex (sx nsy)
E
(h)
Substituting sx sy s pr/2t (see Fig. 8-4a), we obtain
pr
s
ex ᎏᎏ (1 n) ᎏᎏ (1 n)
E
2tE
This equation can be solved for the pressure pc :
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(8-4)
SECTION 8.2 Spherical Pressure Vessels
547
2tEeallow
2(0.25 in.)(29 106 psi)(0.0003)
pc ᎏᎏ 671.3 psi
(9.0 in.)(1 0.28)
r(1 n)
Thus, the allowable pressure based upon the normal strain in the wall is
pc 671 psi.
(d) Allowable pressure based upon the tension in the welded seam. The
allowable tensile load on the welded seam is equal to the failure load divided by
the factor of safety:
Tfailure
8.1 k/in.
Tallow ᎏᎏ ᎏᎏ 3.24 k/in. 3240 lb/in.
n
2.5
The corresponding allowable tensile stress is equal to the allowable load on a
one-inch length of weld divided by the cross-sectional area of a one-inch length
of weld:
Tallow(1.0 in.)
(3240 lb/in.)(1.0 in)
sallow ᎏᎏ ᎏᎏᎏ 12,960 psi
(1.0 in.)(t)
Finally, we solve for the internal pressure by using Eq. (8-1):
2tsallow
2(0.25 in.)(12,960 psi)
pd ᎏᎏ ᎏᎏᎏ 720.0 psi
r
This result gives the allowable pressure based upon tension in the welded seam.
(e) Allowable pressure. Comparing the preceding results for pa, pb, pc, and
pd, we see that shear stress in the wall governs and the allowable pressure in the
tank is
pallow 666 psi
This example illustrates how various stresses and strains enter into the design of
a spherical pressure vessel.
Note: When the internal pressure is at its maximum allowable value
(666 psi), the tensile stresses in the shell are
pr
(666 psi)(9.0 in.)
s ᎏᎏ ᎏᎏ 12,000 psi
2t
Thus, at the inner surface of the shell (Fig. 8-4b), the ratio of the principal stress
in the z direction (666 psi) to the in-plane principal stresses (12,000 psi) is only
0.056. Therefore, our earlier assumption that we can disregard the principal
stress s3 in the z direction and consider the entire shell to be in biaxial stress is
justified.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
548
CHAPTER 8 Applications of Plane Stress
8.3 CYLINDRICAL PRESSURE VESSELS
Cylindrical pressure vessels with a circular cross section (Fig. 8-6) are
found in industrial settings (compressed air tanks and rocket motors), in
homes (fire extinguishers and spray cans), and in the countryside
(propane tanks and grain silos). Pressurized pipes, such as water-supply
pipes and penstocks, are also classified as cylindrical pressure vessels.
We begin our analysis of cylindrical vessels by determining the
normal stresses in a thin-walled circular tank AB subjected to internal
pressure (Fig. 8-7a). A stress element with its faces parallel and perpendicular to the axis of the tank is shown on the wall of the tank. The
normal stresses s1 and s 2 acting on the side faces of this element are
the membrane stresses in the wall. No shear stresses act on these faces
because of the symmetry of the vessel and its loading. Therefore, the
stresses s1 and s 2 are principal stresses.
Because of their directions, the stress s1 is called the circumferential stress or the hoop stress, and the stress s 2 is called the longitudinal
stress or the axial stress. Each of these stresses can be calculated from
equilibrium by using appropriate free-body diagrams.
(a)
(b)
FIG. 8-6 Cylindrical pressure vessels
with circular cross sections
Circumferential Stress
To determine the circumferential stress s1, we make two cuts (mn and
pq) perpendicular to the longitudinal axis and distance b apart (Fig.
8-7a). Then we make a third cut in a vertical plane through the longitudinal axis of the tank, resulting in the free body shown in Fig. 8-7b. This
free body consists not only of the half-circular piece of the tank but also
of the fluid contained within the cuts. Acting on the longitudinal cut
(plane mpqn) are the circumferential stresses s1 and the internal pressure p.
Stresses and pressures also act on the left-hand and right-hand faces
of the free body. However, these stresses and pressures are not shown in
the figure because they do not enter the equation of equilibrium that we
will use. As in our analysis of a spherical vessel, we will disregard the
weight of the tank and its contents.
The circumferential stresses s1 acting in the wall of the vessel have
a resultant equal to s1(2bt), where t is the thickness of the wall. Also,
the resultant force P1 of the internal pressure is equal to 2pbr, where r is
the inner radius of the cylinder. Hence, we have the following equation
of equilibrium:
s1 (2bt) 2pbr 0
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 8.3 Cylindrical Pressure Vessels
A
m p
s1
s2
549
B
b
n
q
(a)
m
p
t
s1
A
m
s2
r
P1 = 2 pbr
s1
q
n
b
FIG. 8-7 Stresses in a circular cylindrical
pressure vessel
P2 = ppr2
n
(c)
(b)
From this equation we obtain the following formula for the circumferential stress in a pressurized cylinder:
pr
s1 ᎏᎏ
t
(8-5)
This stress is uniformly distributed over the thickness of the wall,
provided the thickness is small compared to the radius.
Longitudinal Stress
The longitudinal stress s 2 is obtained from the equilibrium of a free
body of the part of the vessel to the left of cross section mn (Fig. 8-7c).
Again, the free body includes not only part of the tank but also its contents. The stresses s 2 act longitudinally and have a resultant force equal
to s 2(2prt). Note that we are using the inner radius of the shell in place
of the mean radius, as explained in Section 8.2.
The resultant force P2 of the internal pressure is a force equal to
pp r 2. Thus, the equation of equilibrium for the free body is
s2(2prt) ppr 2 0
Solving this equation for s2, we obtain the following formula for the
longitudinal stress in a cylindrical pressure vessel:
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
550
CHAPTER 8 Applications of Plane Stress
pr
s2 ᎏᎏ
2t
(8-6)
This stress is equal to the membrane stress in a spherical vessel
(Eq. 8-1).
Comparing Eqs. (8-5) and (8-6), we see that the circumferential
stress in a cylindrical vessel is equal to twice the longitudinal stress:
s1 2s 2
(8-7)
From this result we note that a longitudinal welded seam in a pressurized tank must be twice as strong as a circumferential seam.
Stresses at the Outer Surface
The principal stresses s1 and s2 at the outer surface of a cylindrical vessel are shown on the stress element of Fig. 8-8a. Since the third principal
stress (acting in the z direction) is zero, the element is in biaxial stress.
The maximum in-plane shear stresses occur on planes that are
rotated 45° about the z axis; these stresses are
s1 s2
s1
pr
(8-8)
(tmax)z ᎏᎏ ᎏᎏ ᎏᎏ
2
4
4t
The maximum out-of-plane shear stresses are obtained by 45° rotations
about the x and y axes, respectively; thus,
s1
pr
s2
pr
(tmax)y ᎏᎏ ᎏᎏ
(8-9a,b)
(tmax)x ᎏᎏ ᎏᎏ
2
2t
2
4t
Comparing the preceding results, we see that the absolute maximum
shear stress is
s1
pr
tmax ᎏᎏ ᎏᎏ
2
2t
(8-10)
This stress occurs on a plane that has been rotated 45° about the x axis.
y
sx = s2
O
FIG. 8-8 Stresses in a circular
cylindrical pressure vessel at (a) the
outer surface and (b) the inner
surface
z
y
sy = s1
sy = s1
sx = s2
sx = s2
x
sz = – p
z
(a)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
sy = s1
sx = s2
O
x
sy = s1
(b)
SECTION 8.3 Cylindrical Pressure Vessels
551
Stresses at the Inner Surface
The stress conditions at the inner surface of the wall of the vessel are
shown in Fig. 8-8b. The principal stresses are
pr
s1 ᎏᎏ
t
pr
s2 ᎏᎏ
2t
s3 p
(a,b,c)
The three maximum shear stresses, obtained by 45° rotations about the
x, y, and z axes, are
s1 s3
pr
p
(tmax)x ᎏᎏ ᎏᎏ
2
2t
2
s2 s3
pr
p
(tmax)y ᎏ ᎏᎏ
2
4t
2
s1 s2
pr
(tmax)z ᎏ ᎏᎏ
2
4t
(d,e)
(f)
The first of these three stresses is the largest. However, as explained in
the discussion of shear stresses in a spherical shell, we may disregard
the additional term p/2 in Eqs. (d) and (e) when the shell is thin-walled.
Equations (d), (e), and (f) then become the same as Eqs. (8-9) and (8-8),
respectively.
Therefore, in all of our examples and problems pertaining to
cylindrical pressure vessels, we will disregard the presence of the
compressive stress in the z direction. (This compressive stress varies
from p at the inner surface to zero at the outer surface.) With this
approximation, the stresses at the inner surface become the same as the
stresses at the outer surface (biaxial stress). As explained in the discussion
of spherical pressure vessels, this procedure is satisfactory when we
consider the numerous other approximations in this theory.
General Comments
The preceding formulas for stresses in a circular cylinder are valid in
parts of the cylinder away from any discontinuities that cause stress
concentrations, as discussed previously for spherical shells. An obvious
discontinuity exists at the ends of the cylinder where the heads are
attached, because the geometry of the structure changes abruptly. Other
stress concentrations occur at openings, at points of support, and
wherever objects or fittings are attached to the cylinder. The stresses at
such points cannot be determined solely from equilibrium equations;
instead, more advanced methods of analysis (such as shell theory and
finite-element analysis) must be used.
Some of the limitations of the elementary theory for thin-walled
shells are listed in Section 8.2.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
552
CHAPTER 8 Applications of Plane Stress
Example 8-2
Helical weld
a
FIG. 8-9 Example 8-2. Cylindrical
pressure vessel with a helical weld
A cylindrical pressure vessel is constructed from a long, narrow steel plate by
wrapping the plate around a mandrel and then welding along the edges of the
plate to make a helical joint (Fig. 8-9). The helical weld makes an angle
a 55° with the longitudinal axis. The vessel has inner radius r 1.8 m and
wall thickness t 20 mm. The material is steel with modulus E 200 GPa and
Poisson’s ratio n 0.30. The internal pressure p is 800 kPa.
Calculate the following quantities for the cylindrical part of the vessel:
(a) the circumferential and longitudinal stresses s1 and s2, respectively; (b) the
maximum in-plane and out-of-plane shear stresses; (c) the circumferential and
longitudinal strains e1 and e2, respectively; and (d) the normal stress sw and
shear stress tw acting perpendicular and parallel, respectively, to the welded
seam.
Solution
(a) Circumferential and longitudinal stresses. The circumferential and
longitudinal stresses s1 and s2, respectively, are pictured in Fig. 8-10a, where
they are shown acting on a stress element at point A on the wall of the vessel.
The magnitudes of the stresses can be calculated from Eqs. (8-5) and (8-6):
u = 35°
s1
A
B
s2
u
x
(a)
y
y
y1
sy = s1 = 72 MPa
47.8 MPa x1
60.2 MPa
u = 35°
sx = s2 = 36 MPa
x
O
A
FIG. 8-10 Solution to Example 8-2
x
O
16.9 MPa
B
(b)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(c)
SECTION 8.3 Cylindrical Pressure Vessels
pr
(800 kPa)(1.8 m)
s1 ᎏᎏ ᎏᎏ 72 MPa
t
20 m m
553
pr
s1
s2 ᎏᎏ ᎏᎏ 36 MPa
2t
2
The stress element at point A is shown again in Fig. 8-10b, where the x axis is in
the longitudinal direction of the cylinder and the y axis is in the circumferential
direction. Since there is no stress in the z direction (s 3 0), the element is in
biaxial stress.
Note that the ratio of the internal pressure (800 kPa) to the smaller in-plane
principal stress (36 MPa) is 0.022. Therefore, our assumption that we may disregard any stresses in the z direction and consider all elements in the cylindrical
shell, even those at the inner surface, to be in biaxial stress is justified.
(b) Maximum shear stresses. The largest in-plane shear stress is obtained
from Eq. (8-8):
s1 s2
s1
pr
(tmax) z ᎏᎏ ᎏᎏ ᎏᎏ 18 MPa
2
4
4t
Because we are disregarding the normal stress in the z direction, the largest
out-of-plane shear stress is obtained from Eq. (8-9a):
s1
pr
tmax ᎏᎏ ᎏᎏ 36 MPa
2
2t
This last stress is the absolute maximum shear stress in the wall of the vessel.
(c) Circumferential and longitudinal strains. Since the largest stresses are
well below the yield stress of steel (see Table H-3, Appendix H), we may
assume that Hooke’s law applies to the wall of the vessel. Then we can obtain
the strains in the x and y directions (Fig. 8-10b) from Eqs. (7-39a) and (7-39b)
for biaxial stress:
1
ex ᎏᎏ(sx nsy)
E
1
ey ᎏᎏ(sy nsx)
E
(g,h)
We note that the strain ex is the same as the principal strain e2 in the longitudinal
direction and that the strain ey is the same as the principal strain e1 in the circumferential direction. Also, the stress sx is the same as the stress s2, and the
stress sy is the same as the stress s1. Therefore, the preceding two equations can
be written in the following forms:
s2
pr
e2 ᎏᎏ (1 2n) ᎏᎏ (1 2n)
E
2tE
(8-11a)
s1
pr
e1 ᎏᎏ (2 n) ᎏᎏ (2 2n)
2E
2tE
(8-11b)
Substituting numerical values, we find
s2
(36 MPa)[1 2(0.30)]
e2 ᎏᎏ(1 2n) ᎏᎏᎏ 72 106
200 GPa
E
s1
(72 MPa)(2 0.30)
e1 ᎏᎏ (2 n) ᎏᎏᎏ 306 106
2(200 GPa)
2E
These are the longitudinal and circumferential strains in the cylinder.
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
554
CHAPTER 8 Applications of Plane Stress
(d) Normal and shear stresses acting on the welded seam. The stress element at point B in the wall of the cylinder (Fig. 8-10a) is oriented so that its
sides are parallel and perpendicular to the weld. The angle u for the element is
u 90° a 35°
as shown in Fig. 8-10c. Either the stress-transformation equations or Mohr’s
circle may be used to obtain the normal and shear stresses acting on the side
faces of this element.
Stress-transformation equations. The normal stress sx1 and the shear stress
tx1y1 acting on the x1 face of the element (Fig. 8-10c) are obtained from
Eqs. (7-4a) and (7-4b), which are repeated here:
y
y1
47.8 MPa x1
60.2 MPa
u = 35°
a
x
O
16.9 MPa
B
sx sy
sx sy
sx1 ᎏ ᎏ cos 2u txy sin 2 u
2
2
(8-12a)
sx sy
tx1 y1 ᎏᎏ sin 2u tx y cos 2 u
2
(8-12b)
Substituting sx s2 pr/2t, sy s1 pr/t, and txy 0, we obtain
(c)
FIG. 8-10c (Repeated)
pr
sx1 (3 cos 2u)
4t
pr
tx1 y1 ᎏᎏ sin 2u
4t
(8-13a,b)
These equations give the normal and shear stresses acting on an inclined plane
oriented at an angle u with the longitudinal axis of the cylinder.
Substituting pr/4t 18 MPa and u 35° into Eqs. (8-13a) and (8-13b),
we obtain
sx1 47.8 MPa
tx 1y1 16.9 MPa
These stresses are shown on the stress element of Fig. 8-10c.
To complete the stress element, we can calculate the normal stress sy1
acting on the y1 face of the element from the sum of the normal stresses on
perpendicular faces (Eq. 7-6):
s1 s2 sx1 sy1
(8-14)
Substituting numerical values, we get
sy1 s1 s2 sx1 72 MPa 36 MPa 47.8 MPa 60.2 MPa
as shown in Fig. 8-10c.
From the figure, we see that the normal and shear stresses acting
perpendicular and parallel, respectively, to the welded seam are
sw 47.8 MPa
tw 16.9 MPa
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
555
SECTION 8.3 Cylindrical Pressure Vessels
54
R = 18
A
(u = 0)
O
36
FIG. 8-11 Mohr’s circle for the biaxial
stress element of Fig. 8-10b. (Note: All
stresses on the circle have units of MPa.)
C
2u = 70°
B(u = 90°)
sx1
D(u = 35°)
72
tx1y1
Mohr’s circle. The Mohr’s circle construction for the biaxial stress element
of Fig. 8-10b is shown in Fig. 8-11. Point A represents the stress s2 36 MPa
on the x face (u 0) of the element, and point B represents the stress s1
72 MPa on the y face (u 90°). The center C of the circle is at a stress of
54 MPa, and the radius of the circle is
72 MPa 36 MPa
R ᎏᎏ 18 MPa
2
A counterclockwise angle 2u 70° (measured on the circle from point A)
locates point D, which corresponds to the stresses on the x1 face (u 35°) of
the element. The coordinates of point D (from the geometry of the circle) are
sx1 54 MPa R cos 70° 54 MPa (18 MPa)(cos 70°) 47.8 MPa
tx1y1 R sin 70° (18 MPa)(sin 70°) 16.9 MPa
p
u
d
These results are the same as those found earlier from the stress-transformation
equations.
Note: When seen in a side view, a helix follows the shape of a sine curve
(Fig. 8-12). The pitch of the helix is
p p d tan u
FIG. 8-12 Side view of a helix
(8-15)
where d is the diameter of the circular cylinder and u is the angle between a
normal to the helix and a longitudinal line. The width of the flat plate that wraps
into the cylindrical shape is
w p d sin u
(8-16)
Thus, if the diameter of the cylinder and the angle u are given, both the pitch
and the plate width are established. For practical reasons, the angle u is usually
in the range from 20° to 35°.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
556
CHAPTER 8 Applications of Plane Stress
8.4 MAXIMUM STRESSES IN BEAMS
The stress analysis of a beam usually begins by finding the normal and
shear stresses acting on cross sections. For instance, when Hooke’s law
holds, we can obtain the normal and shear stresses from the flexure and
shear formulas (Eqs. 5-13 and 5-38, respectively, of Chapter 5):
My
s ᎏᎏ
I
VQ
t ᎏᎏ
Ib
(8-17a,b)
In the flexure formula, s is the normal stress acting on the cross section,
M is the bending moment, y is the distance from the neutral axis, and I is
the moment of inertia of the cross-sectional area with respect to the
neutral axis. (The sign conventions for M and y in the flexure formula
are shown in Figs. 5-9 and 5-10 of Chapter 5.)
In the case of the shear formula, t is the shear stress at any point in
the cross section, V is the shear force, Q is the first moment of the crosssectional area outside of the point in the cross section where the stress is
being found, and b is the width of the cross section. (The shear formula
is usually written without regard to signs because the directions of the
shear stresses are apparent from the directions of the loads.)
The normal stresses obtained from the flexure formula have their
maximum values at the farthest distances from the neutral axis, whereas
the shear stresses obtained from the shear formula usually have their
highest values at the neutral axis. The normal stresses are calculated at
the cross section of maximum bending moment, and the shear stresses
are calculated at the cross section of maximum shear force. In most
circumstances, these are the only stresses that are needed for design
purposes.
However, to obtain a more complete picture of the stresses in a
beam, we need to determine the principal stresses and maximum shear
stresses at various points in the beam. We will begin by discussing the
stresses in a rectangular beam.
Beams of Rectangular Cross Section
We can gain an understanding of how the stresses in a beam vary by
considering the simple beam of rectangular cross section shown in
Fig. 8-13a. For the purposes of this discussion, we choose a cross section to the left of the load and then select five points (A, B, C, D, and E)
on the side of the beam. Points A and E are at the top and bottom of the
beam, respectively, point C is at the midheight of the beam, and points B
and D are in between.
If Hooke’s law applies, the normal and shear stresses at each of
these five points can be readily calculated from the flexure and shear
formulas. Since these stresses act on the cross section, we can picture
them on stress elements having vertical and horizontal faces, as shown
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 8.4 Maximum Stresses in Beams
A
557
B
DC
E
D
(a)
FIG. 8-13 Stresses in a beam of rectangular cross section: (a) simple beam with
points A, B, C, D, and E on the side of
the beam; (b) normal and shear stresses
acting on stress elements at points A, B,
C, D, and E; (c) principal stresses; and
(d) maximum shear stresses
A
A
A
B
B
B
C
C
C
D
D
D
E
E
E
(b)
(c)
(d)
in Fig. 8-13b. Note that all elements are in plane stress, because there
are no stresses acting perpendicular to the plane of the figure.
At point A the normal stress is compressive and there are no shear
stresses. Similarly, at point E the normal stress is tensile and again there
are no shear stresses. Thus, the elements at these locations are in uniaxial stress. At the neutral axis (point C) the element is in pure shear. At
the other two locations (points B and D), both normal and shear stresses
act on the stress elements.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
558
CHAPTER 8 Applications of Plane Stress
To find the principal stresses and maximum shear stresses at each
point, we may use either the transformation equations of plane stress or
Mohr’s circle. The directions of the principal stresses are shown in
Fig. 8-13c, and the directions of the maximum shear stresses are shown
in Fig. 8-13d. (Note that we are considering only the in-plane stresses.)
Now let us examine the principal stresses in more detail. From the
sketches in Fig. 8-13c, we can observe how the principal stresses change
as we go from top to bottom of the beam. Let us begin with the compressive principal stress. At point A the compressive stress acts in the
horizontal direction and the other principal stress is zero. As we move
toward the neutral axis the compressive principal stress becomes
inclined, and at the neutral axis (point C) it acts at 45° to the horizontal.
At point D the compressive principal stress is further inclined from the
horizontal, and at the bottom of the beam its direction becomes vertical
(except that its magnitude is now zero).
Thus, the direction and magnitude of the compressive principal
stress vary continuously from top to bottom of the beam. If the chosen
cross section is located in a region of large bending moment, the largest
compressive principal stress occurs at the top of the beam (point A), and
the smallest compressive principal stress (zero) occurs at the bottom of
the beam (point E). If the cross section is located in a region of small
bending moment and large shear force, then the largest compressive
principal stress is at the neutral axis.
Analogous comments apply to the tensile principal stress, which
also varies in both magnitude and direction as we move from point A to
point E. At point A the tensile stress is zero and at point E it has its maximum value. (Graphs showing how the principal stresses vary in
magnitude for a particular beam and particular cross section are given
later in Fig. 8-19 of Example 8-3.)
The maximum shear stresses (Fig. 8-13d) at the top and bottom of
the beam occur on 45° planes (because the elements are in uniaxial
stress). At the neutral axis, the maximum shear stresses occur on horizontal and vertical planes (because the element is in pure shear). At all
points, the maximum shear stresses occur on planes oriented at 45° to
the principal planes. In regions of high bending moment, the largest
shear stresses occur at the top and bottom of the beam; in regions of low
bending moment and high shear force, the largest shear stresses occur at
the neutral axis.
By investigating the stresses at many cross sections of the beam, we
can determine how the principal stresses vary throughout the beam.
Then we can construct two systems of orthogonal curves, called stress
trajectories, that give the directions of the principal stresses. Examples
of stress trajectories for rectangular beams are shown in Fig. 8-14. Part
(a) of the figure shows a cantilever beam with a load acting at the free
end, and part (b) shows a simple beam with a uniform load. Solid lines
are used for tensile principal stresses and dashed lines for compressive
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 8.4 Maximum Stresses in Beams
559
(a)
FIG. 8-14 Principal-stress trajectories
for beams of rectangular cross section:
(a) cantilever beam, and (b) simple
beam. (Solid lines represent tensile
principal stresses and dashed lines
represent compressive principal
stresses.)
(b)
principal stresses. The curves for tensile and compressive principal
stresses always intersect at right angles, and every trajectory crosses the
longitudinal axis at 45°. At the top and bottom surfaces of the beam,
where the shear stress is zero, the trajectories are either horizontal or
vertical.*
Another type of curve that may be plotted from the principal
stresses is a stress contour, which is a curve connecting points of equal
principal stress. Stress contours for a cantilever beam of rectangular
cross section are shown in Fig. 8-15 (for tensile principal stresses only).
The contour of largest stress is at the upper left part of the figure. As we
move downward in the figure, the tensile stresses represented by the
contours become smaller and smaller. The contour line of zero tensile
stress is at the lower edge of the beam. Thus, the largest tensile stress
occurs at the support, where the bending moment has its largest value.
Note that stress trajectories (Fig. 8-14) give the directions of the
principal stresses but give no information about the magnitudes of the
stresses. In general, the magnitudes of the principal stresses vary as we
move along a trajectory. In contrast, the magnitudes of the principal
stresses are constant as we move along a stress contour (Fig. 8-15 on the
next page), but the contours give no information about the directions of
the stresses. In particular, the principal stresses are neither parallel nor
perpendicular to a stress contour.
*Stress trajectories were originated by the German engineer Karl Culmann (1821–1881);
see Ref. 8-1.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
560
CHAPTER 8 Applications of Plane Stress
The stress trajectories and contours of Figs. 8-14 and 8-15 were
plotted from the flexure and shear formulas (Eqs. 8-17a and b). Stress
concentrations near the supports and near the concentrated loads, as well
as the direct compressive stresses caused by the uniform load bearing on
the top of the beam (Fig. 8-14b), were disregarded in plotting these
figures.
FIG. 8-15 Stress contours for a cantilever
beam (tensile principal stresses only)
Wide-Flange Beams
Beams having other cross-sectional shapes, such as wide-flange beams,
can be analyzed for the principal stresses in a manner similar to that
described previously for rectangular beams. For instance, consider the
simply supported wide-flange beam shown in Fig. 8-16a. Proceeding as
for a rectangular beam, we identify points A, B, C, D, and E from top to
bottom of the beam (Fig. 8-16b). Points B and D are in the web where it
meets the flange, and point C is at the neutral axis. We can think of these
points as being located either on the side of the beam (Figs. 8-16b and c)
or inside the beam along a vertical axis of symmetry (Fig. 8-16d). The
stresses determined from the flexure and shear formulas are the same at
both sets of points.
Stress elements at points A, B, C, D, and E (as seen in a side view of
the beam) are shown in parts (e) through (i) of Fig. 8-16. These elements
have the same general appearance as those for a rectangular beam
(Fig. 8-13b).
The largest principal stresses usually occur at the top and bottom of
the beam (points A and E) where the stresses obtained from the flexure
formula have their largest values. However, depending upon the relative
magnitudes of the bending moment and shear force, the largest stresses
sometimes occur in the web where it meets the flange (points B and D).
The explanation lies in the fact that the normal stresses at points B and D
are only slightly smaller than those at points A and E, whereas the shear
stresses (which are zero at points A and E) may be significant at points B
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
561
SECTION 8.4 Maximum Stresses in Beams
A
D
E
(a)
A
B
C
D
B
C
D
B
C
D
E
E
(b)
FIG. 8-16 Stresses in a wide-flange beam
A
A
E
(c)
(d)
A
B
C
D
E
(e)
(f)
(g)
(h)
(i)
and D because of the thin web. (Note: Figure 5-38 in Chapter 5 shows
how the shear stresses vary in the web of a wide-flange beam.)
The maximum shear stresses acting on a cross section of a wideflange beam always occur at the neutral axis, as shown by the shear
formula (Eq. 8-17b). However, the maximum shear stresses acting on
inclined planes usually occur either at the top and bottom of the beam
(points A and E) or in the web where it meets the flange (points B and
D) because of the presence of normal stresses.
When analyzing a wide-flange beam for the maximum stresses,
remember that high stresses may exist near supports, points of loading,
fillets, and holes. Such stress concentrations are confined to the region
very close to the discontinuity and cannot be calculated by elementary
beam formulas.
The following example illustrates the procedure for determining the
principal stresses and maximum shear stresses at a selected cross section
in a rectangular beam. The procedures for a wide-flange beam are
similar.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
562
CHAPTER 8 Applications of Plane Stress
Example 8-3
A simple beam AB with span length L 6 ft supports a concentrated load P
10,800 lb acting at distance c 2 ft from the right-hand support (Fig. 8-17).
The beam is made of steel and has a rectangular cross section of width b 2 in.
and height h 6 in.
Investigate the principal stresses and maximum shear stresses at cross
section mn, located at distance x 9 in. from end A of the beam. (Consider
only the in-plane stresses.)
y
P = 10,800 lb
x = 9 in.
c = 2 ft
y
m
O
A
x
B
n
z
h = 6 in.
b = 2 in.
L = 6 ft
FIG. 8-17 Example 8-3. Beam of
P = 3,600 lb
RA = —
3
rectangular cross section
Solution
We begin by using the flexure and shear formulas to calculate the stresses
acting on cross section mn. Once those stresses are known, we can determine
the principal stresses and maximum shear stresses from the equations of plane
stress. Finally, we can plot graphs of these stresses to show how they vary over
the height of the beam.
As a preliminary matter, we note that the reaction of the beam at support A
is RA P/3 3600 lb, and therefore the bending moment and shear force at
section mn are
M RA x (3600 lb)(9 in.) 32,400 lb-in.
V RA 3600 lb
Normal stresses on cross section mn. These stresses are found from the
flexure formula (Eq. 8-17a), as follows:
12(32,400 lb-in.)y
My
12My
ᎏᎏ
900y
sx ᎏᎏ ᎏᎏ
I
bh3
(2 in.)(6 in.)3
(a)
in which y has units of inches (in.) and sx has units of pounds per square inch
(psi). The stresses calculated from Eq. (a) are positive when in tension and negative when in compression. For instance, note that a positive value of y (upper
half of the beam) gives a negative stress, as expected.
A stress element cut from the side of the beam at cross section mn (Fig.
8-17) is shown in Fig. 8-18. For reference purposes, a set of xy axes is associated with the element. The normal stress sx and the shear stress txy are shown
acting on the element in their positive directions. (Note that in this example
there is no normal stress sy acting on the element.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 8.4 Maximum Stresses in Beams
563
Shear stresses on cross section mn. The shear stresses are given by the
shear formula (Eq. 8-17b) in which the first moment Q for a rectangular cross
section is
y
txy
sx
O
x
冢
h
h/2 y
b h2
Q b ᎏᎏ y y ᎏ ᎏᎏ ᎏᎏ y 2
2
2
2 4
Thus, the shear formula becomes
VQ
12V
b h2
6V h2
ᎏᎏ y 2
t ᎏᎏ ᎏ3ᎏ ᎏᎏ ᎏᎏ y2 ᎏᎏ
Ib
(bh )(b) 2 4
bh3 4
FIG. 8-18 Plane-stress element at cross
section mn of the beam of Fig. 8-17
(Example 8-3)
(8-18)
(8-19)
The shear stresses txy acting on the x face of the stress element (Fig. 8-18) are
positive upward, whereas the actual shear stresses t (Eq. 8-19) act downward.
Therefore, the shear stresses txy are given by the following formula:
6V h2
ᎏᎏ y 2
tx y ᎏᎏ
bh3 4
(8-20)
Substituting numerical values into this equation gives
6(3600 lb) (6 in.)2
ᎏᎏ y 2 50(9 y 2)
txy ᎏᎏ
(2 in.)(6 in.)3
4
(b)
in which y has units of inches (in.) and txy has units of pounds per square inch
(psi).
Calculation of stresses. For the purpose of calculating the stresses at cross
section mn, let us divide the height of the beam into six equal intervals and label
the corresponding points from A to G, as shown in the side view of the
beam (Fig. 8-19a on the next page). The y coordinates of these points are listed
in column 2 of Table 8-1 on the next page, and the corresponding stresses sx
and txy (calculated from Eqs. a and b, respectively) are listed in columns 3 and
4. These stresses are plotted in Figs. 8-19b and 8-19c. The normal stresses vary
linearly from a compressive stress of 2700 psi at the top of the beam (point A)
to a tensile stress of 2700 psi at the bottom of the beam (point G). The shear
stresses have a parabolic distribution with the maximum stress at the neutral
axis (point D).
Principal stresses and maximum shear stresses. The principal stresses at
each of the seven points A through G may be determined from Eq. (7-17):
sx sy
s1, 2 ᎏ
2
t
冢2
冪
sx sy
2
2
xy
(8-21)
Since there is no normal stress in the y direction (Fig. 8-18), this equation
simplifies to
sx
s1, 2
2
冢
冪
sx 2
tx2y
2
(8-22)
Also, the maximum shear stresses (from Eq. 7-25) are
tmax
冢
冪
sx sy 2
tx2y
2
(8-23)
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
564
CHAPTER 8 Applications of Plane Stress
y
m
– 2700
A
B
0
– 400
900
1800
sx
n
(a)
– 250
2700
txy
(b)
1350
–1834
34
934
–1052
152
602
– 450
450
450
–152
1052
602
– 34
1834
0
(c)
– 2700
0
Fig. 8-17 (Example 8-3). (a) Points A, B,
C, D, E, F, and G at cross section mn;
(b) normal stresses sx acting on cross
section mn; (c) shear stresses txy acting
on cross section mn; (d) principal tensile
stresses s1; (e) principal compressive
stresses s2; and (f) maximum shear
stresses tmax. (Note: All stresses have
units of psi.)
– 450
0
3 in.
F
G
FIG. 8-19 Stresses in the beam of
– 400
– 900
x
E
– 250
– 1800
3 in.
C
D
0
2700
934
1350
0
s1
s2
tmax
(d)
(e)
(f)
TABLE 8-1 STRESSES AT CROSS SECTION mn IN THE BEAM OF FIG. 8-17
(1)
(2)
(3)
(4)
(5)
(6)
(7)
Point
y
(in.)
sx
(psi)
txy
(psi)
s1
(psi)
s2
(psi)
tmax
(psi)
3
2
1
0
1
2
3
2700
0
250
900
400
0
900
1800
2700
450
0
34
152
450
1052
1834
2700
2700
1800
1350
934
602
450
602
934
1350
A
B
C
D
E
F
G
400
250
0
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
1834
1052
450
152
34
0
SECTION 8.4 Maximum Stresses in Beams
565
which simplifies to
tmax
t
冢2
冪
sx
2
2
xy
(8-24)
Thus, by substituting the values of sx and txy (from Table 8-1) into Eqs. (8-22)
and (8-24), we can calculate the principal stresses s1 and s2 and the maximum
shear stress tmax. These quantities are listed in the last three columns of
Table 8-1 and are plotted in Figs. 8-19d, e, and f.
The tensile principal stresses s1 increase from zero at the top of the beam
to a maximum of 2700 psi at the bottom (Fig. 8-19d). The directions of the
stresses also change, varying from vertical at the top to horizontal at the bottom.
At midheight, the stress s1 acts on a 45° plane. Similar comments apply to the
compressive principal stress s2, except in reverse. For instance, the stress is
largest at the top of the beam and zero at the bottom (Fig. 8-19e).
The maximum shear stresses at cross section mn occur on 45° planes at the
top and bottom of the beam. These stresses are equal to one-half of the normal
stresses sx at the same points. At the neutral axis, where the normal stress sx is
zero, the maximum shear stresses occur on the horizontal and vertical planes.
Note 1: If we consider other cross sections of the beam, the maximum
normal and shear stresses will be different from those shown in Fig. 8-19. For
instance, at a cross section between section mn and the concentrated load
(Fig. 8-17), the normal stresses sx are larger than shown in Fig. 8-19b because
the bending moment is larger. However, the shear stresses txy are the same as
those shown in Fig. 8-19c because the shear force doesn’t change in that region
of the beam. Consequently, the principal stresses s1 and s2 and maximum shear
stresses tmax will vary in the same general manner as shown in Figs. 8-19d, e,
and f but with different numerical values.
The largest tensile stress anywhere in the beam is the normal stress at the
bottom of the beam at the cross section of maximum bending moment. This
stress is
(stens)max 14,400 psi
The largest compressive stress has the same numerical value and occurs at the
top of the beam at the same cross section.
The largest shear stress txy acting on a cross section of the beam occurs to
the right of the load P (Fig. 8-17) because the shear force is larger in that region
of the beam (V RB 7200 lb). Therefore, the largest value of txy, which
occurs at the neutral axis, is
(txy)max 900 psi
The largest shear stress anywhere in the beam occurs on 45° planes at either the
top or bottom of the beam at the cross section of maximum bending moment:
14,400 psi
tmax 7200 psi
2
Note 2: In the practical design of ordinary beams, the principal stresses and
maximum shear stresses are rarely calculated. Instead, the tensile and compressive stresses to be used in design are calculated from the flexure formula at the
cross section of maximum bending moment, and the shear stress to be used in
design is calculated from the shear formula at the cross section of maximum
shear force.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
566
CHAPTER 8 Applications of Plane Stress
8.5 COMBINED LOADINGS
In previous chapters we analyzed structural members subjected to a
single type of loading. For instance, we analyzed axially loaded bars in
Chapters 1 and 2, shafts in torsion in Chapter 3, and beams in bending in
Chapters 4, 5, and 6. We also analyzed pressure vessels earlier in this
chapter. For each type of loading, we developed methods for finding
stresses, strains, and deformations.
However, in many structures the members are required to resist
more than one kind of loading. For example, a beam may be subjected
to the simultaneous action of bending moments and axial forces (Fig.
8-20a), a pressure vessel may be supported so that it also functions as a
beam (Fig. 8-20b), or a shaft in torsion may carry a bending load
(Fig. 8-20c). Known as combined loadings, situations similar to those
shown in Fig. 8-20 occur in a great variety of machines, buildings, vehicles, tools, equipment, and many other kinds of structures.
A structural member subjected to combined loadings can often be
analyzed by superimposing the stresses and strains caused by each load
acting separately. However, superposition of both stresses and strains is
permissible only under certain conditions, as explained in earlier chapters. One requirement is that the stresses and strains must be linear
functions of the applied loads, which in turn requires that the material
follow Hooke’s law and the displacements remain small.
A second requirement is that there must be no interaction between
the various loads, that is, the stresses and strains due to one load must
not be affected by the presence of the other loads. Most ordinary structures satisfy these two conditions, and therefore the use of superposition
is very common in engineering work.
Cable
Beam
(a)
Pressure vessel
(b)
B
(c)
Method of Analysis
FIG. 8-20 Examples of structures
subjected to combined loadings:
(a) wide-flange beam supported by a
cable (combined bending and axial
load), (b) cylindrical pressure vessel
supported as a beam, and (c) shaft in
combined torsion and bending
While there are many ways to analyze a structure subjected to more than
one type of load, the procedure usually includes the following steps:
1. Select a point in the structure where the stresses and strains are to be
determined. (The point is usually selected at a cross section where
the stresses are large, such as at a cross section where the bending
moment has its maximum value.)
2. For each load on the structure, determine the stress resultants at the
cross section containing the selected point. (The possible stress
resultants are an axial force, a twisting moment, a bending moment,
and a shear force.)
3. Calculate the normal and shear stresses at the selected point due to
each of the stress resultants. Also, if the structure is a pressure vessel, determine the stresses due to the internal pressure. (The stresses
are found from the stress formulas derived previously; for instance,
s P/A, t Tr/IP, s My/I, t VQ/Ib, and s pr/t.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 8.5 Combined Loadings
567
4. Combine the individual stresses to obtain the resultant stresses at the
selected point. In other words, obtain the stresses sx, sy, and txy acting on a stress element at the point. (Note that in this chapter we are
dealing only with elements in plane stress.)
5. Determine the principal stresses and maximum shear stresses at the
selected point, using either the stress-transformation equations or
Mohr’s circle. If required, determine the stresses acting on other
inclined planes.
6. Determine the strains at the point with the aid of Hooke’s law for
plane stress.
7. Select additional points and repeat the process. Continue until
enough stress and strain information is available to satisfy the
purposes of the analysis.
Illustration of the Method
C
D
A
B
T
P
b
(a)
A M = Pb
B
T
V=P
To illustrate the procedure for analyzing a member subjected to combined loadings, we will discuss in general terms the stresses in the
cantilever bar of circular cross section shown in Fig. 8-21a. This bar is
subjected to two types of load—a torque T and a vertical load P, both
acting at the free end of the bar.
Let us begin by arbitrarily selecting two points A and B for investigation (Fig. 8-21a). Point A is located at the top of the bar and point B is
located on the side. Both points are located at the same cross section.
The stress resultants acting at the cross section (Fig. 8-21b) are a
twisting moment equal to the torque T, a bending moment M equal to
the load P times the distance b from the free end of the bar to the cross
section, and a shear force V equal to the load P.
The stresses acting at points A and B are shown in Fig. 8-21c. The
twisting moment T produces torsional shear stresses
Tr
2T
t1 3
IP
pr
(b)
A
t1
r
B
sA
t1
t2
(c)
FIG. 8-21 Cantilever bar subjected to
combined torsion and bending: (a) loads
acting on the bar, (b) stress resultants at
a cross section, and (c) stresses at points
A and B
(a)
in which r is the radius of the bar and IP p r 4/2 is the polar moment of
inertia of the cross-sectional area. The stress t1 acts horizontally to the
left at point A and vertically downward at point B, as shown in the
figure.
The bending moment M produces a tensile stress at point A:
Mr
4M
(b)
sA 3
I
pr
in which I pr 4/4 is the moment of inertia about the neutral axis. However, the bending moment produces no stress at point B, because B is
located on the neutral axis.
The shear force V produces no shear stress at the top of the bar
(point A), but at point B the shear stress is as follows (see Eq. 5-42 in
Chapter 5):
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
568
CHAPTER 8 Applications of Plane Stress
C
D
4V
4V
t 2
3p r 2
3A
A
B
T
P
b
(a)
A M = Pb
B
T
V=P
(b)
A
t1
r
B
sA
t1
t2
(c)
FIG. 8-21 (Repeated)
(c)
in which A pr 2 is the cross-sectional area.
The stresses sA and t1 acting at point A (Fig. 8-21c) are shown
acting on a stress element in Fig. 8-22a. This element is cut from the top
of the bar at point A. A two-dimensional view of the element, obtained
by looking vertically downward on the element, is shown in Fig. 8-22b.
For the purpose of determining the principal stresses and maximum
shear stresses, we construct x and y axes through the element. The x axis
is parallel to the longitudinal axis of the circular bar (Fig. 8-21a) and the
y axis is horizontal. Note that the element is in plane stress with
sx sA, sy 0, and txy t1.
A stress element at point B (also in plane stress) is shown in Fig.
8-23a. The only stresses acting on this element are the shear stresses,
equal to t1 t2 (see Fig. 8-21c). A two-dimensional view of the stress
element is shown in Fig. 8-23b, with the x axis parallel to the longitudinal axis of the bar and the y axis in the vertical direction. The stresses
acting on the element are sx sy 0 and txy (t 1 t 2).
Now that we have determined the stresses acting at points A and B
and constructed the corresponding stress elements, we can use the transformation equations of plane stress (Sections 7.2 and 7.3) or Mohr’s
circle (Section 7.4) to determine principal stresses, maximum shear
stresses, and stresses acting in inclined directions. We can also use
Hooke’s law (Section 7.5) to determine the strains at points A and B.
The procedure described previously for analyzing the stresses at
points A and B (Fig. 8-21a) can be used at other points in the bar. Of
particular interest are the points where the stresses calculated from the
flexure and shear formulas have maximum or minimum values, called
critical points. For instance, the normal stresses due to bending are
largest at the cross section of maximum bending moment, which is at
the support. Therefore, points C and D at the top and bottom of the beam
at the fixed end (Fig. 8-21a) are critical points where the stresses should
be calculated. Another critical point is point B itself, because the shear
stresses are a maximum at this point. (Note that in this example the
y
A
A
t1
FIG. 8-22 Stress element at point A
sA
O
sA
(a)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
t1
(b)
x
SECTION 8.5 Combined Loadings
569
y
B
B
t1 + t2
x
O
t1 + t2
FIG. 8-23 Stress element at point B
(a)
(b)
shear stresses do not change if point B is moved along the bar in the
longitudinal direction.)
As a final step, the principal stresses and maximum shear stresses at
the critical points can be compared with one another in order to determine the absolute maximum normal and shear stresses in the bar.
This example illustrates the general procedure for determining the
stresses produced by combined loadings. Note that no new theories are
involved—only applications of previously derived formulas and concepts. Since the variety of practical situations seems to be endless, we
will not derive general formulas for calculating the maximum stresses.
Instead, we will treat each structure as a special case.
Selection of Critical Points
If the objective of the analysis is to determine the largest stresses anywhere in the structure, then the critical points should be selected at cross
sections where the stress resultants have their largest values. Furthermore, within those cross sections, the points should be selected where
either the normal stresses or the shear stresses have their largest values.
By using good judgment in the selection of the points, we often can be
reasonably certain of obtaining the absolute maximum stresses in the
structure.
However, it is sometimes difficult to recognize in advance where the
maximum stresses in the member are to be found. Then it may be necessary to investigate the stresses at a large number of points, perhaps even
using trial-and-error in the selection of points. Other strategies may also
prove fruitful—such as deriving equations specific to the problem at
hand or making simplifying assumptions to facilitate an otherwise
difficult analysis.
The following examples illustrate the methods used to calculate
stresses in structures subjected to combined loadings.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
570
CHAPTER 8 Applications of Plane Stress
Example 8-4
The rotor shaft of a helicopter drives the rotor blades that provide the lifting
force to support the helicopter in the air (Fig. 8-24a). As a consequence, the
shaft is subjected to a combination of torsion and axial loading (Fig. 8-24b).
For a 50-mm diameter shaft transmitting a torque T 2.4 kNm and a
tensile force P 125 kN, determine the maximum tensile stress, maximum
compressive stress, and maximum shear stress in the shaft.
(a)
y
s0
P
T
T
x
O
t0
P
FIG. 8-24 Example 8-4. Rotor shaft of a
helicopter (combined torsion and axial
force)
(b)
(c)
(c)
Solution
The stresses in the rotor shaft are produced by the combined action of the
axial force P and the torque T (Fig. 8-24b). Therefore, the stresses at any point
on the surface of the shaft consist of a tensile stress s0 and shear stresses t0, as
shown on the stress element of Fig. 8-24c. Note that the y axis is parallel to the
longitudinal axis of the shaft.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 8.5 Combined Loadings
571
The tensile stress s0 equals the axial force divided by the cross-sectional
area:
P
4P
4(125 kN )
s0
63.66 MPa
A
pd 2
p (50 mm)2
The shear stress t0 is obtained from the torsion formula (see Eqs. 3-11 and 3-12
of Section 3.3):
16T
Tr
16(2.4 kNm)
t0 3
97.78 MPa
pd
IP
p (50 mm)3
The stresses s0 and t0 act directly on cross sections of the shaft.
Knowing the stresses s0 and t0, we can now obtain the principal stresses
and maximum shear stresses by the methods described in Section 7.3. The principal stresses are obtained from Eq. (7-17):
sx sy
s1, 2
2
t
冢2
冪
sx sy
2
2
xy
(d)
Substituting sx 0, sy s0 63.66 MPa, and txy t0 97.78 MPa, we
get
s1,2 32 MPa 103 MPa
or
s1 135 MPa
s2 71 MPa
These are the maximum tensile and compressive stresses in the rotor shaft.
The maximum in-plane shear stresses (Eq. 7-25) are
tmax
t
冢2
冪
sx sy
2
2
xy
(e)
This term was evaluated previously, so we see immediately that
tmax 103 MPa
Because the principal stresses s 1 and s 2 have opposite signs, the maximum
in-plane shear stresses are larger than the maximum out-of-plane shear stresses
(see Eqs. 7-28a, b, and c and the accompanying discussion). Therefore, the maximum shear stress in the shaft is 103 MPa.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
572
CHAPTER 8 Applications of Plane Stress
Example 8-5
P
P
A
(a)
y
A thin-walled cylindrical pressure vessel with a circular cross section is subjected to internal gas pressure p and simultaneously compressed by an axial load
P 12 k (Fig. 8-25a). The cylinder has inner radius r 2.1 in. and wall thickness t 0.15 in.
Determine the maximum allowable internal pressure pallow based upon an
allowable shear stress of 6500 psi in the wall of the vessel.
Solution
sy
sx
O
x
A
The stresses in the wall of the pressure vessel are caused by the combined
action of the internal pressure and the axial force. Since both actions produce
uniform normal stresses throughout the wall, we can select any point on the surface for investigation. At a typical point, such as point A (Fig. 8-25a), we isolate
a stress element as shown in Fig. 8-25b. The x axis is parallel to the longitudinal
axis of the pressure vessel and the y axis is circumferential. Note that there are
no shear stresses acting on the element.
Principal stresses. The longitudinal stress sx is equal to the tensile stress
s2 produced by the internal pressure (see Fig. 8-7a and Eq. 8-6) minus the compressive stress produced by the axial force; thus,
pr
P pr
P
sx
A
2t
2t
2prt
(b)
FIG. 8-25 Example 8-5. Pressure vessel
subjected to combined internal pressure
and axial force
(f)
in which A 2prt is the cross-sectional area of the cylinder. (Note that for convenience we are using the inner radius r in all calculations.)
The circumferential stress sy is equal to the tensile stress s1 produced by
the internal pressure (Fig. 8-7a and Eq. 8-5):
pr
sy
t
(g)
Note that sy is algebraically larger than sx.
Since no shear stresses act on the element (Fig. 8-25), the normal stresses
sx and sy are also the principal stresses:
pr
s1 sy
t
pr
P
s2 sx
2t
2p rt
Now substituting numerical values, we obtain
pr
p(2.1 in.)
s1 14.0p
t
0.15 in.
pr
p(2.1 in.)
P
12 k
s2
2t
2p rt
2(0.15 in.)
7.0p 6063 psi
in which s1, s2, and p have units of pounds per square inch (psi).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(h,i)
SECTION 8.5 Combined Loadings
573
In-plane shear stresses. The maximum in-plane shear stress (Eq. 7-26) is
s1 s2
1
tmax (14.0p 7.0p 6063 psi) 3.5p 3032 psi
2
2
Since tmax is limited to 6500 psi, the preceding equation becomes
6500 psi 3.5p 3032 psi
from which we get
3468 psi
p 990.9 psi or ( p allow)1 990 psi
3.5
because we round downward.
Out-of-plane shear stresses. The maximum out-of-plane shear stress (see
Eqs. 7-28a and 7-28b) is either
s2
s1
tmax or tmax
2
2
From the first of these two equations we get
6500 psi 3.5p 3032 psi or ( pallow)2 2720 psi
From the second equation we get
6500 psi 7.0p or ( pallow)3 928 psi
Allowable internal pressure. Comparing the three calculated values for the
allowable pressure, we see that ( pallow)3 governs, and therefore the allowable
internal pressure is
pallow 928 psi
At this pressure the principal stresses are s1 13,000 psi and s2 430 psi.
These stresses have the same signs, thus confirming that one of the out-of-plane
shear stresses must be the largest shear stress (see the discussion following
Eqs. 7-28a, b, and c).
Note: In this example, we determined the allowable pressure in the vessel
assuming that the axial load was equal to 12 k. A more complete analysis would
include the possibility that the axial force may not be present. (As it turns out,
the allowable pressure does not change if the axial force is removed from this
example.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
574
CHAPTER 8 Applications of Plane Stress
Example 8-6
0.5 m
2.0 m
Bill’s
Bookstore
1.2 m
A sign of dimensions 2.0 m 1.2 m is supported by a hollow circular pole
having outer diameter 220 mm and inner diameter 180 mm (Fig. 8-26). The sign
is offset 0.5 m from the centerline of the pole and its lower edge is 6.0 m above
the ground.
Determine the principal stresses and maximum shear stresses at points A
and B at the base of the pole due to a wind pressure of 2.0 kPa against the sign.
Solution
6.0 m
Stress resultants. The wind pressure against the sign produces a resultant
force W that acts at the midpoint of the sign (Fig. 8-27a) and is equal to the
pressure p times the area A over which it acts:
W pA (2.0 kPa)(2.0 m 1.2 m) 4.8 kN
A
The line of action of this force is at height h 6.6 m above the ground and at
distance b 1.5 m from the centerline of the pole.
The wind force acting on the sign is statically equivalent to a lateral force
W and a torque T acting on the pole (Fig. 8-27b). The torque is equal to the
force W times the distance b:
B
C
B
T Wb (4.8 kN)(1.5 m) 7.2 kN·m
A
180 mm
The stress resultants at the base of the pole (Fig. 8-27c) consist of a bending moment M, a torque T, and a shear force V. Their magnitudes are
220 mm
M Wh (4.8 kN)(6.6 m) 31.68 kNm
FIG. 8-26 Example 8-6. Wind pressure
against a sign (combined bending,
torsion, and shear of the pole)
T 7.2 kNm
V W 4.8 kN
Examination of these stress resultants shows that maximum bending stresses
occur at point A and maximum shear stresses at point B. Therefore, A and B are
critical points where the stresses should be determined. (Another critical point
is diametrically opposite point A, as explained in the Note at the end of this
example.)
Stresses at points A and B. The bending moment M produces a tensile
stress sA at point A (Fig. 8-27d) but no stress at point B (which is located on the
neutral axis). The stress sA is obtained from the flexure formula:
M(d2/2)
sA
I
in which d2 is the outer diameter (220 mm) and I is the moment of inertia of the
cross section. The moment of inertia is
冢
p
p
4
4
I d 2 d 1 (220 mm)4 (180 mm)4 63.46 10 6 m4
64
64
in which d1 is the inner diameter. Therefore, the stress sA is
Md2
(31.68 kNm)(220 mm)
sA
54.91 MPa
2(63.46 10 6 m4)
2I
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 8.5 Combined Loadings
575
b = 1.5 m
W = 4.8 kN
T = 7.2 kN ⋅ m
h = 6.6 m
W
h = 6.6 m
(a)
(b)
T
M
C
t1
C
B
A
t2
sA
V
A
(c)
t1
t1
B
(d)
y
sy = sA
txy = t1
A
O
FIG. 8-27 Solution to Example 8-6
y
txy = t1 + t2
B
x
(e)
O
x
(f)
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
576
CHAPTER 8 Applications of Plane Stress
The torque T produces shear stresses t1 at points A and B (Fig. 8-27d). We
can calculate these stresses from the torsion formula:
T(d2/2)
t 1
IP
in which IP is the polar moment of inertia:
冢
p
4
4
IP d 2 d 1 2I 126.92 10 6 m4
32
Thus,
(7.2 kNm)(220 mm)
Td2
t1
6.24 MPa
2(126.92 10 6 m4)
2IP
Finally, we calculate the shear stresses at points A and B due to the shear
force V. The shear stress at point A is zero, and the shear stress at point B
(denoted t 2 in Fig. 8-27d) is obtained from the shear formula for a circular tube
(Eq. 5-44 of Section 5.9):
2
2
4V r 2 r2r1 r 1
t 2
3A
( j)
in which r2 and r1 are the outer and inner radii, respectively, and A is the crosssectional area:
d2
r2 110 mm
2
d1
r1 90 mm
2
A p (r 22 r 21) 12,570 mm2
Substituting numerical values into Eq. (j), we obtain
t2 0.76 MPa
The stresses acting on the cross section at points A and B have now been
calculated.
Stress elements. The next step is to show these stresses on stress elements
(Figs. 8-27e and f). For both elements, the y axis is parallel to the longitudinal
axis of the pole and the x axis is horizontal. At point A the stresses acting on the
element are
sx 0
sy sA 54.91 MPa
txy t1 6.24 MPa
At point B the stresses are
sx sy 0
txy t1 t2 6.24 MPa 0.76 MPa 7.00 MPa
Since there are no normal stresses acting on the element, point B is in pure
shear.
Now that all stresses acting on the stress elements (Figs. 8-27e and f) are
known, we can use the equations given in Section 7.3 to determine the principal
stresses and maximum shear stresses.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 8.5 Combined Loadings
577
Principal stresses and maximum shear stresses at point A. The principal
stresses are obtained from Eq. (7-17), which is repeated here:
sx sy
s1,2
2
t
冪冢
2
sx sy
2
2
xy
(k)
Substituting sx 0, sy 54.91 MPa, and txy 6.24 MPa, we get
s1,2 27.5 MPa 28.2 MPa
or
s1 55.7 MPa
s2 0.7 MPa
The maximum in-plane shear stresses may be obtained from Eq. (7-25):
tmax
t
冢2
冪
sx sy
2
2
xy
(l)
This term was evaluated previously, so we see immediately that
tmax 28.2 MPa
Because the principal stresses s1 and s2 have opposite signs, the maximum inplane shear stresses are larger than the maximum out-of-plane shear stresses
(see Eqs. 7-28a, b, and c and the accompanying discussion). Therefore, the maximum shear stress at point A is 28.2 MPa.
Principal stresses and maximum shear stresses at point B. The stresses at
this point are sx 0, sy 0, and txy 7.0 MPa. Since the element is in pure
shear, the principal stresses are
s1 7.0 MPa
s2 7.0 MPa
and the maximum in-plane shear stress is
tmax 7.0 MPa
The maximum out-of-plane shear stresses are half this value.
Note: If the largest stresses anywhere in the pole are needed, then we must
also determine the stresses at the critical point diametrically opposite point A,
because at that point the compressive stress due to bending has its largest value.
The principal stresses at that point are
s1 0.7 MPa
s2 55.7 MPa
and the maximum shear stress is 28.2 MPa. Therefore, the largest tensile stress
in the pole is 55.7 MPa, the largest compressive stress is 55.7 MPa, and the
largest shear stress is 28.2 MPa. (Keep in mind that only the effects of the wind
pressure are considered in this analysis. Other loads, such as the weight of the
structure, also produce stresses at the base of the pole.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
578
CHAPTER 8 Applications of Plane Stress
Example 8-7
d = 9 in.
P1 = 3,240 lb
A tubular post of square cross section supports a horizontal platform (Fig.
8-28). The tube has outer dimension b 6 in. and wall thickness t 0.5 in. The
platform has dimensions 6.75 in. 24.0 in. and supports a uniformly distributed load of 20 psi acting over its upper surface. The resultant of this distributed
load is a vertical force P1:
P1 (20 psi)(6.75 in. 24.0 in.) 3240 lb
This force acts at the midpoint of the platform, which is at distance d 9 in.
from the longitudinal axis of the post. A second load P2 800 lb acts horizontally on the post at height h 52 in. above the base.
Determine the principal stresses and maximum shear stresses at points A
and B at the base of the post due to the loads P1 and P2.
P2 = 800 lb
b b
Solution
h = 52 in.
Stress resultants. The force P1 acting on the platform (Fig. 8-28) is statically
equivalent to a force P1 and a moment M1 P1d acting at the centroid
of the cross section of the post (Fig. 8-29a). The load P2 is also shown in
this figure.
The stress resultants at the base of the post due to the loads P1 and P2
and the moment M1 are shown in Fig. 8-29b. These stress resultants are the
following:
A B
b = 6 in.
b = 3 in.
B 2
b = 3 in.
2
t = 0.5 in.
A
1.
2.
M1 P1d (3240 lb)(9 in.) 29,160 lb-in.
3.
4.
t = 0.5 in.
FIG. 8-28 Example 8-7. Loads on a post
(combined axial load, bending, and
shear)
An axial compressive force P1 3240 lb
A bending moment M1 produced by the force P1:
A shear force P2 800 lb
A bending moment M2 produced by the force P2:
M2 P2 h (800 lb)(52 in.) 41,600 lb-in.
Examination of these stress resultants (Fig. 8-29b) shows that both M1 and M2
produce maximum compressive stresses at point A and the shear force produces
maximum shear stresses at point B. Therefore, A and B are critical points where
the stresses should be determined. (Another critical point is diagonally opposite
point A, as explained in the Note at the end of this example.)
Stresses at points A and B. (1) The axial force P1 (Fig. 8-29b) produces
uniform compressive stresses throughout the post. These stresses are
P1
sP1
A
in which A is the cross-sectional area of the post:
A b2 (b 2t)2 4t(b t)
4(0.5 in.)(6 in. 0.5 in.) 11.00 in.2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
579
SECTION 8.5 Combined Loadings
P1
M1 = P1d
P2
h
A B
(a)
sM 2
P1
M1 = P1d
P2
A
sM1
sM 1
B
sP1
sP1
M2 = P2h
A
(b)
B
tP2
(c)
y
y
sA = sP1 + sM1 + sM2 = 4090 psi
sB = sP1 + sM1 = 1860 psi
B
A
O
x
O
x
tP2 = 160 psi
FIG. 8-29 Solution to Example 8-7
(d)
(e)
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
580
CHAPTER 8 Applications of Plane Stress
Therefore, the axial compressive stress is
P1
3 240 lb
sP1 2 295 psi
A
11.00 in.
The stress sP1 is shown acting at points A and B in Fig. 8-29c.
(2) The bending moment M1 (Fig. 8-29b) produces compressive stresses
sM1 at points A and B (Fig. 8-29c). These stresses are obtained from the flexure
formula:
M1(b/2)
M1b
sM1
I
2I
in which I is the moment of inertia of the cross-sectional area:
4
(b 2t)
b4
1
I (6 in.)4 (5 in.)4 55.92 in.4
12
12
12
Thus, the stress sM1 is
M1b
(29,160 lb-in.)(6 in.)
sM1 1564 psi
2I
(3) The shear force P2 (Fig. 8-29b) produces a shear stress at point B but
not at point A. From the discussion of shear stresses in the webs of beams with
flanges (Section 5.10), we know that an approximate value of the shear stress
can be obtained by dividing the shear force by the web area (see Eq. 5-50 in
Section 5.10). Thus, the shear stress produced at point B by the force P2 is
P2
P2
800 lb
tP2 160 psi
Aweb
2t(b 2t)
The stress tP2 acts at point B in the direction shown in Fig. 8-29c.
If desired, we can calculate the shear stress tP2 from the more accurate
formula of Eq. (5-48a) in Section 5.10. The result of that calculation is tP2
163 psi, which shows that the shear stress obtained from the approximate
formula is satisfactory.
(4) The bending moment M2 (Fig. 8-29b) produces a compressive stress at
point A but no stress at point B. The stress at A is
M2(b/2)
M2b
(41,600 lb-in.)(6 in.)
sM2 2232 psi
I
2I
This stress is also shown in Fig. 8-29c.
Stress elements. The next step is to show the stresses acting on stress
elements at points A and B (Figs. 8-29d and e). Each element is oriented so that
the y axis is vertical (that is, parallel to the longitudinal axis of the post) and the
x axis is horizontal. At point A the only stress is a compressive stress sA in the y
direction (Fig. 8-29d):
sA sP1 sM1 sM2
295 psi 1564 psi 2232 psi 4090 psi (compression)
Thus, this element is in uniaxial stress.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
581
SECTION 8.5 Combined Loadings
y
sy
txy
sx
x
O
FIG. 8-30 Notation for an element in
plane stress
At point B the compressive stress in the y direction (Fig. 8-29e) is
sB sP1 sM1 295 psi 1564 psi 1860 psi (compression)
and the shear stress is
tP2 160 psi
The shear stress acts leftward on the top face of the element and downward on
the x face of the element.
Principal stresses and maximum shear stresses at point A. Using the standard notation for an element in plane stress (Fig. 8-30), we write the stresses for
element A (Fig. 8-29d) as follows:
sx 0
sy sA 4090 psi
txy 0
Since the element is in uniaxial stress, the principal stresses are
s1 0
s2 4090 psi
and the maximum in-plane shear stress (Eq. 7-26) is
s1 s2
4090 psi
tmax 2050 psi
2
2
The maximum out-of-plane shear stress (Eq. 7-28a) has the same magnitude.
Principal stresses and maximum shear stresses at point B. Again using the
standard notation for plane stress (Fig. 8-30), we see that the stresses at point B
(Fig. 8-29e) are
sx 0
sy sB 1860 psi
txy tP2 160 psi
To obtain the principal stresses, we use Eq. (7-17), which is repeated here:
sx sy
s1,2
2
t
冪冢
2
sx sy
2
2
xy
(m)
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
582
CHAPTER 8 Applications of Plane Stress
Substituting for sx, sy, and txy, we get
s1,2 930 psi 944 psi
or
s1 14 psi
s2 1870 psi
The maximum in-plane shear stresses may be obtained from Eq. (7-25):
tmax
t
冢2
冪
sx sy
2
2
xy
(n)
This term was evaluated previously, so we see immediately that
tmax 944 psi
Because the principal stresses s1 and s2 have opposite signs, the maximum inplane shear stresses are larger than the maximum out-of-plane shear stresses
(see Eqs. 7-28a, b, and c and the accompanying discussion). Therefore, the maximum shear stress at point B is 944 psi.
Note: If the largest stresses anywhere at the base of the post are needed,
then we must also determine the stresses at the critical point diagonally opposite
point A (Fig. 8-29c), because at that point each bending moment produces the
maximum tensile stress. Thus, the tensile stress acting at that point is
sy sP1 sM1 sM2 295 psi 1564 psi 2232 psi 3500 psi
The stresses acting on a stress element at that point (see Fig. 8-30) are
sx 0
sy 3500 psi
txy 0
and therefore the principal stresses and maximum shear stress are
s1 3500 psi
s2 0
tmax 1750 psi
Thus, the largest tensile stress anywhere at the base of the post is 3500 psi, the
largest compressive stress is 4090 psi, and the largest shear stress is 2050 psi.
(Keep in mind that only the effects of the loads P1 and P2 are considered in this
analysis. Other loads, such as the weight of the structure, also produce stresses
at the base of the post.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 8 Problems
583
PROBLEMS CHAPTER 8
Spherical Pressure Vessels
When solving the problems for Section 8.2, assume that the
given radius or diameter is an inside dimension and that all
internal pressures are gage pressures.
8.2-4 A rubber ball (see figure) is inflated to a pressure of
60 kPa. At that pressure the diameter of the ball is 230 mm
and the wall thickness is 1.2 mm. The rubber has modulus
of elasticity E 3.5 MPa and Poisson’s ratio n 0.45.
Determine the maximum stress and strain in the ball.
8.2-1 A large spherical tank (see figure) contains gas at a
pressure of 400 psi. The tank is 45 ft in diameter and is
constructed of high-strength steel having a yield stress in
tension of 80 ksi.
Determine the required thickness (to the nearest
1/4 inch) of the wall of the tank if a factor of safety of 3.5
with respect to yielding is required.
PROBS. 8.2-4 and 8.2-5
8.2-5 Solve the preceding problem if the pressure is 9.0 psi,
the diameter is 9.0 in., the wall thickness is 0.05 in.,
the modulus of elasticity is 500 psi, and Poisson’s ratio is
0.45.
PROBS. 8.2-1 and 8.2-2
8.2-2 Solve the preceding problem if the internal pressure
is 3.5 MPa, the diameter is 18 m, the yield stress is
550 MPa, and the factor of safety is 3.0. Determine the
required thickness to the nearest millimeter.
8.2-6 A spherical steel pressure vessel (diameter 480 mm,
thickness 8.0 mm) is coated with brittle lacquer that cracks
when the strain reaches 150 106 (see figure).
What internal pressure p will cause the lacquer to
develop cracks? (Assume E 205 GPa and n 0.30.)
8.2-3 A hemispherical window (or viewport) in a decom-
Cracks in
coating
pression chamber (see figure) is subjected to an internal air
pressure of 80 psi. The port is attached to the wall of the
chamber by 18 bolts.
Find the tensile force F in each bolt and the tensile
stress s in the viewport if the radius of the hemisphere is
7.0 in. and its thickness is 1.0 in.
PROB. 8.2-6
PROB. 8.2-3
8.2-7 A spherical tank of diameter 50 in. and wall thickness 2.0 in. contains compressed air at a pressure of 2400
psi. The tank is constructed of two hemispheres joined by a
welded seam (see figure on the next page).
(a) What is the tensile load f (lb per in. of length of
weld) carried by the weld? (See the figure on the next page.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
584
CHAPTER 8 Applications of Plane Stress
(b) What is the maximum shear stress tmax in the wall
of the tank?
(c) What is the maximum normal strain e in the wall?
(For steel, assume E 31 106 psi and n 0.29.)
Weld
D0
PROBS. 8.2-7 and 8.2-8
PROB. 8.2-11
8.2-8 Solve the preceding problem for the following data:
diameter 1.0 m, thickness 50 mm, pressure 24 MPa, modulus 210 GPa, and Poisson’s ratio 0.29.
Cylindrical Pressure Vessels
8.2-9 A spherical stainless-steel tank having a diameter of
8.3-1 A scuba tank (see figure) is being designed for an
20 in. is used to store propane gas at a pressure of 2400 psi.
The properties of the steel are as follows: yield stress in
tension, 140,000 psi; yield stress in shear, 65,000 psi; modulus of elasticity, 30 106 psi; and Poisson’s ratio, 0.28.
The desired factor of safety with respect to yielding is 2.75.
Also, the normal strain must not exceed 1000 106.
Determine the minimum permissible thickness tmin of
the tank.
When solving the problems for Section 8.3, assume that the
given radius or diameter is an inside dimension and that all
internal pressures are gage pressures.
internal pressure of 1600 psi with a factor of safety of
2.0 with respect to yielding. The yield stress of the steel is
35,000 psi in tension and 16,000 psi in shear.
If the diameter of the tank is 7.0 in., what is the
minimum required wall thickness?
8.2-10 Solve the preceding problem if the diameter is
500 mm, the pressure is 16 MPa, the yield stress in tension
is 950 MPa, the yield stress in shear is 450 MPa, the factor
of safety is 2.4, the modulus of elasticity is 200 GPa, Poisson’s ratio is 0.28, and the normal strain must not exceed
1200 106.
PROB. 8.3-1
8.2-11 A hollow pressurized sphere having radius r
4.8 in. and wall thickness t 0.4 in. is lowered into a lake
(see figure). The compressed air in the tank is at a pressure
of 24 psi (gage pressure when the tank is out of the water).
At what depth D0 will the wall of the tank be subjected
to a compressive stress of 90 psi?
8.3-2 A tall standpipe with an open top (see figure) has
diameter d 2.0 m and wall thickness t 18 mm.
(a) What height h of water will produce a circumferential stress of 10.9 MPa in the wall of the standpipe?
(b) What is the axial stress in the wall of the tank due
to the water pressure?
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 8 Problems
585
What was the internal pressure p in the can? (Assume
E 10 106 psi and n 0.33.)
d
h
PROB. 8.3-2
12 FL OZ (355 mL)
8.3-3 An inflatable structure used by a traveling circus has
the shape of a half-circular cylinder with closed ends (see
figure). The fabric and plastic structure is inflated by a
small blower and has a radius of 40 ft when fully inflated.
A longitudinal seam runs the entire length of the “ridge” of
the structure.
If the longitudinal seam along the ridge tears open
when it is subjected to a tensile load of 540 pounds per inch
of seam, what is the factor of safety n against tearing when
the internal pressure is 0.5 psi and the structure is fully
inflated?
Longitudinal seam
PROB. 8.3-5
8.3-6 A circular cylindrical steel tank (see figure) contains
a volatile fuel under pressure. A strain gage at point A
records the longitudinal strain in the tank and transmits this
information to a control room. The ultimate shear stress in
the wall of the tank is 82 MPa and a factor of safety of 2 is
required.
At what value of the strain should the operators take
action to reduce the pressure in the tank? (Data for the steel
are as follows: modulus of elasticity E 205 GPa and
Poisson’s ratio n 0.30.)
Cylindrical tank
Pressure relief
valve
PROB. 8.3-3
8.3-4 A thin-walled cylindrical pressure vessel of radius r
A
is subjected simultaneously to internal gas pressure p and a
compressive force F acting at the ends (see figure).
What should be the magnitude of the force F in order
to produce pure shear in the wall of the cylinder?
F
F
PROB. 8.3-6
8.3-7 A cylinder filled with oil is under pressure from a
PROB. 8.3-4
8.3-5 A strain gage is installed in the longitudinal direction
on the surface of an aluminum beverage can (see figure).
The radius-to-thickness ratio of the can is 200. When the lid
of the can is popped open, the strain changes by e 0 170
106.
piston, as shown in the figure. The diameter d of the piston
is 1.80 in. and the compressive force F is 3500 lb. The maximum allowable shear stress tallow in the wall of the
cylinder is 5500 psi.
What is the minimum permissible thickness tmin of the
cylinder wall? (See the figure on the next page.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
586
CHAPTER 8 Applications of Plane Stress
Welded seams
Cylinder
F
p
Piston
PROBS. 8.3-10 and 8.3-11
PROBS. 8.3-7 and 8.3-8
8.3-8 Solve the preceding problem if d 90 mm, F
42 kN, and tallow 40 MPa.
8.3-9 A standpipe in a water-supply system (see figure) is
12 ft in diameter and 6 inches thick. Two horizontal pipes
carry water out of the standpipe; each is 2 ft in diameter
and 1 inch thick. When the system is shut down and water
fills the pipes but is not moving, the hoop stress at the
bottom of the standpipe is 130 psi.
(a) What is the height h of the water in the standpipe?
(b) If the bottoms of the pipes are at the same elevation
as the bottom of the standpipe, what is the hoop stress in
the pipes?
PROB. 8.3-9
8.3-10 A cylindrical tank with hemispherical heads is constructed of steel sections that are welded circumferentially
(see figure). The tank diameter is 1.2 m, the wall thickness
is 20 mm, and the internal pressure is 1600 kPa.
(a) Determine the maximum tensile stress sh in the
heads of the tank.
(b) Determine the maximum tensile stress sc in the
cylindrical part of the tank.
(c) Determine the tensile stress sw acting perpendicular to the welded joints.
(d) Determine the maximum shear stress th in the
heads of the tank.
(e) Determine the maximum shear stress tc in the
cylindrical part of the tank.
8.3-11 A cylindrical tank with diameter d 16 in. is subjected to internal gas pressure p 400 psi. The tank is
constructed of steel sections that are welded circumferentially (see figure). The heads of the tank are hemispherical.
The allowable tensile and shear stresses are 8000 psi and
3200 psi, respectively. Also, the allowable tensile stress perpendicular to a weld is 6400 psi.
Determine the minimum required thickness tmin of
(a) the cylindrical part of the tank, and (b) the hemispherical heads.
8.3-12 A pressurized steel tank is constructed with a helical weld that makes an angle a 60° with the longitudinal
axis (see figure). The tank has radius r 0.5 m, wall thickness t 15 mm, and internal pressure p 2.4 MPa. Also,
the steel has modulus of elasticity E 200 GPa and Poisson’s ratio n 0.30.
Determine the following quantities for the cylindrical
part of the tank: (a) the circumferential and longitudinal
stresses, (b) the maximum in-plane and out-of-plane shear
stresses, (c) the circumferential and longitudinal strains,
and (d) the normal and shear stresses acting on planes parallel and perpendicular to the weld (show these stresses on
a properly oriented stress element).
★
Helical weld
a
PROBS. 8.3-12 and 8.3-13
8.3-13 Solve the preceding problem for a welded tank
with a 65°, r 18 in., t 0.6 in., p 200 psi, E
30 106 psi, and n 0.30.
★
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
587
CHAPTER 8 Problems
Maximum Stresses in Beams
When solving the problems for Section 8.4, consider only
the in-plane stresses and disregard the weights of the
beams.
8.4-1 A cantilever beam of rectangular cross section is
subjected to a concentrated load P 15 k acting at the free
end (see figure). The beam has width b 4 in. and height
h 10 in. Point A is located at distance c 2 ft from the
free end and distance d 3 in. from the bottom of the
beam.
Calculate the principal stresses s1 and s2 and the maximum shear stress tmax at point A. Show these stresses on
sketches of properly oriented elements.
P
the overhang is L /2. The cross section has width b and
height h. Point D is located midway between the supports
at a distance d from the top face of the beam.
Knowing that the maximum tensile stress (principal
stress) at point D is s1 36.1 MPa, determine the magnitude of the load P. Data for the beam are as follows:
L 1.5 m, b 45 mm, h 180 mm, and d 30 mm.
P
d
D
A
C
B
L
—
2
L
—
2
h
b
L
—
2
PROBS. 8.4-4 and 8.4-5
8.4-5 Solve the preceding problem if the stress and dimen-
h
sions are as follows: s1 2320 psi, L 75 in., b 2.0 in.,
h 9.0 in., and d 1.5 in.
A
c
b
d
8.4-6 A beam of wide-flange cross section (see figure) has
PROBS. 8.4-1 and 8.4-2
8.4-2 Solve the preceding problem for the following data:
P 120 kN, b 100 mm, h 200 mm, c 0.5 m, and
d 150 mm.
8.4-3 A simple beam of rectangular cross section (width
4 in., height 8 in.) carries a uniform load of 1000 lb/ft on a
span of 10 ft (see figure).
Find the principal stresses s1 and s2 and the maximum
shear stress tmax at a cross section 1 ft from the left-hand
support at each of the following locations: (a) the neutral
axis, (b) 2 in. above the neutral axis, and (c) the top of the
beam. (Disregard the direct compressive stresses produced
by the uniform load bearing against the top of the beam.)
the following dimensions: b 120 mm, t 10 mm, h
300 mm, and h1 260 mm. The beam is simply supported
with span length L 3.0 m. A concentrated load P
120 kN acts at the midpoint of the span.
At a cross section located 1.0 m from the left-hand
support, determine the principal stresses s1 and s2 and the
maximum shear stress tmax at each of the following locations: (a) the top of the beam, (b) the top of the web, and
(c) the neutral axis.
b
t
h1 h
1000 lb/ft
8 in.
4 in.
1 ft
10 ft
PROB. 8.4-3
8.4-4 An overhanging beam ABC of rectangular cross section supports a concentrated load P at the free end (see
figure). The span length from A to B is L, and the length of
PROBS. 8.4-6 and 8.4-7
8.4-7 A beam of wide-flange cross section (see figure) has
the following dimensions: b 5 in., t 0.5 in., h 12 in.,
and h1 10.5 in. The beam is simply supported with span
length L 10 ft and supports a uniform load q 6 k/ft.
Calculate the principal stresses s1 and s2 and the maximum shear stress tmax at a cross section located 3 ft from
the left-hand support at each of the following locations:
(a) the bottom of the beam, (b) the bottom of the web, and
(c) the neutral axis.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
588
CHAPTER 8 Applications of Plane Stress
8.4-8 A W 12 14 wide-flange beam (see Table E-l,
Appendix E) is simply supported with a span length of 8 ft
(see figure). The beam supports a concentrated load of
20 kips at midspan.
At a cross section located 2 ft from the left-hand support, determine the principal stresses s1 and s2 and the
maximum shear stress tmax at each of the following locations: (a) the top of the beam, (b) the top of the web, and
(c) the neutral axis.
B
A
60°
7.5 kN
2.0 m
75 75
mm mm
25 mm
z
25 mm
20 k
W 12 × 14
2 ft
4 ft
PROB. 8.4-8
8.4-9 A W 8 28 wide-flange beam (see Table E-l,
Appendix E) is simply supported with a span length of
120 in. (see figure). The beam supports two symmetrically
placed concentrated loads of 6.0 k each.
At a cross section located 20 in. from the right-hand
support, determine the principal stresses s1 and s2 and the
maximum shear stress tmax at each of the following locations: (a) the top of the beam, (b) the top of the web, and
(c) the neutral axis.
6.0 k
8.4-11 A simple beam of rectangular cross section has
span length L 60 in. and supports a concentrated load
P 18 k at the midpoint (see figure). The height of the
beam is h 6 in. and the width is b 2 in.
Plot graphs of the principal stresses s1 and s2 and the
maximum shear stress tmax, showing how they vary over
the height of the beam at cross section mn, which is located
20 in. from the left-hand support.
P
m
h
n
b
L
PROBS. 8.4-11 and 8.4-12
6.0 k
8.4-12 Solve the preceding problem for a cross section
mn located 0.15 m from the support if L 0.7 m, P
144 kN, h 120 mm, and b 20 mm.
★
W 8 × 28
D
40 in.
Combined Loadings
20 in. 20 in.
120 in.
PROB. 8.4-9
8.4-10 A cantilever beam of T-section is loaded by an
inclined force of magnitude 7.5 kN (see figure). The line of
action of the force is inclined at an angle of 60° to the horizontal and intersects the top of the beam at the end cross
section. The beam is 2.0 m long and the cross section has
the dimensions shown.
Determine the principal stresses s1 and s2 and the
maximum shear stress tmax at points A and B in the web of
the beam.
★
150 mm
★
8 ft
40 in.
C
PROB. 8.4-10
D
2 ft
y
The problems for Section 8.5 are to be solved assuming that
the structures behave linearly elastically and that the
stresses caused by two or more loads may be superimposed
to obtain the resultant stresses acting at a point. Consider
both in-plane and out-of-plane shear stresses unless otherwise specified.
8.5-1 A bracket ABCD having a hollow circular cross section consists of a vertical arm AB, a horizontal arm BC
parallel to the x0 axis, and a horizontal arm CD parallel to
the z0 axis (see figure). The arms BC and CD have lengths
b1 3.2 ft and b2 2.4 ft, respectively. The outer and
inner diameters of the bracket are d2 8.0 in. and d1
7.0 in. A vertical load P 1500 lb acts at point D.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 8 Problems
Determine the maximum tensile, compressive, and
shear stresses in the vertical arm.
y0
b2
b1
B
C
589
Determine the maximum tensile, compressive, and
shear stresses in the drill pipe.
D
P
x0
A
z0
PROB. 8.5-1
PROB. 8.5-3
8.5-2 A gondola on a ski lift is supported by two bent arms,
as shown in the figure. Each arm is offset by the distance
b 180 mm from the line of action of the weight force W.
The allowable stresses in the arms are 100 MPa in tension
and 50 MPa in shear.
If the loaded gondola weighs 12 kN, what is the mininum diameter d of the arms?
W
8.5-4 A segment of a generator shaft is subjected to a
torque T and an axial force P, as shown in the figure. The
shaft is hollow (outer diameter d2 280 mm and inner
diameter d1 230 mm) and delivers 1800 kW at 4.0 Hz.
If the compressive force P 525 kN, what are the
maximum tensile, compressive, and shear stresses in the
shaft?
P
T
d
b
T
P
PROBS. 8.5-4 and 8.5-5
W
PROB. 8.5-2
8.5-3 The hollow drill pipe for an oil well (see figure) is
6.0 in. in outer diameter and 0.75 in. in thickness. Just
above the bit, the compressive force in the pipe (due to the
weight of the pipe) is 60 k and the torque (due to drilling) is
170 k-in.
8.5-5 A segment of a generator shaft of hollow circular
cross section is subjected to a torque T 220 k-in. (see figure). The outer and inner diameters of the shaft are 8.0 in.
and 6.0 in., respectively.
What is the maximum permissible compressive load P
that can be applied to the shaft if the allowable in-plane
shear stress is tallow 6500 psi?
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
590
CHAPTER 8 Applications of Plane Stress
8.5-6 A cylindrical tank subjected to internal pressure p is
8.5-9 Determine the maximum tensile, compressive, and
simultaneously compressed by an axial force F 72 kN
(see figure). The cylinder has diameter d 100 mm and
wall thickness t 4 mm.
Calculate the maximum allowable internal pressure
pmax based upon an allowable shear stress in the wall of the
tank of 60 MPa.
shear stresses at point A on the bicycle pedal crank shown
in the figure.
The pedal and crank are in a horizontal plane and point
A is located on the top of the crank. The load P 160 lb
acts in the vertical direction and the distances (in the horizontal plane) between the line of action of the load and
point A are b1 5.0 in. and b2 2.5 in. Assume that the
crank has a solid circular cross section with diameter
d 0.6 in.
F
F
P = 160 lb
PROB. 8.5-6
Crank
8.5-7 A cylindrical tank having diameter d 2.5 in. is subjected to internal gas pressure p 600 psi and an external
tensile load T 1000 lb (see figure).
Determine the minimum thickness t of the wall of the
tank based upon an allowable shear stress of 3000 psi.
d = 0.6 in.
A
b1 = 5.0 in.
T
b2 = 2.5 in.
T
PROB. 8.5-9
PROB. 8.5-7
8.5-8 The torsional pendulum shown in the figure consists
of a horizontal circular disk of mass M 60 kg suspended
by a vertical steel wire (G 80 GPa) of length L 2 m
and diameter d 4 mm.
Calculate the maximum permissible angle of rotation
fmax of the disk (that is, the maximum amplitude of torsional vibrations) so that the stresses in the wire do not
exceed 100 MPa in tension or 50 MPa in shear.
8.5-10 A cylindrical pressure vessel having radius r
300 mm and wall thickness t 15 mm is subjected to internal pressure p 2.5 MPa. In addition, a torque T
120 kN·m acts at each end of the cylinder (see figure).
(a) Determine the maximum tensile stress smax and the
maximum in-plane shear stress tmax in the wall of the cylinder.
(b) If the allowable in-plane shear stress is 30 MPa,
what is the maximum allowable torque T?
T
d = 4 mm
L=2m
T
PROB. 8.5-10
fmax
M = 60 kg
PROB. 8.5-8
8.5-11 An L-shaped bracket lying in a horizontal plane
supports a load P 150 lb (see figure). The bracket has a
hollow rectangular cross section with thickness t 0.125 in.
and outer dimensions b 2.0 in. and h 3.5 in. The
centerline lengths of the arms are b1 20 in. and b2
30 in.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
591
CHAPTER 8 Problems
Considering only the load P, calculate the maximum
tensile stress st, maximum compressive stress sc, and maximum shear stress tmax at point A, which is located on the
top of the bracket at the support.
A
z
y
A
x
t = 0.125 in.
B
b1 = 20 in.
h = 3.5 in.
C
b2 = 30 in.
P = 150 lb
b = 2.0 in.
PROB. 8.5-13
PROB. 8.5-11
8.5-12 A semicircular bar AB lying in a horizontal plane is
supported at B (see figure). The bar has centerline radius R
and weight q per unit of length (total weight of the bar
equals pqR). The cross section of the bar is circular with
diameter d.
Obtain formulas for the maximum tensile stress st,
maximum compressive stress sc, and maximum in-plane
shear stress tmax at the top of the bar at the support due to
the weight of the bar.
8.5-14 A pressurized cylindrical tank with flat ends is
loaded by torques T and tensile forces P (see figure). The
tank has radius r 50 mm and wall thickness t 3 mm.
The internal pressure p 3.5 MPa and the torque
T 450 N·m.
What is the maximum permissible value of the forces
P if the allowable tensile stress in the wall of the cylinder is
72 MPa?
P
T
T
P
O
A
B
R
PROB. 8.5-14
d
PROB. 8.5-12
8.5-15 A post having a hollow circular cross section sup-
8.5-13 An arm ABC lying in a horizontal plane and supported at A (see figure) is made of two identical solid steel
bars AB and BC welded together at a right angle. Each bar
is 20 in. long.
Knowing that the maximum tensile stress (principal
stress) at the top of the bar at support A due solely to the
weights of the bars is 932 psi, determine the diameter d of
the bars.
ports a horizontal load P 250 lb acting at the end of an
arm that is 4 ft long (see figure on the next page). The height
of the post is 25 ft, and its section modulus is S 10 in.3
(a) Calculate the maximum tensile stress smax and
maximum in-plane shear stress tmax at point A due to the
load P. Point A is located on the “front” of the post, that is,
at the point where the tensile stress due to bending alone is
a maximum.
(b) If the maximum tensile stress and maximum inplane shear stress at point A are limited to 16,000 psi and
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
592
CHAPTER 8 Applications of Plane Stress
6,000 psi, respectively, what is the largest permissible value
of the load P?
4 ft
P = 250 lb
25 ft
8.5-17 A sign is supported by a pole of hollow circular
cross section, as shown in the figure. The outer and inner
diameters of the pole are 10.0 in. and 8.0 in., respectively.
The pole is 40 ft high and weighs 3.8 k. The sign has
dimensions 6 ft 3 ft and weighs 400 lb. Note that its center of gravity is 41 in. from the axis of the pole. The wind
pressure against the sign is 30 lb/ft2.
(a) Determine the stresses acting on a stress element at
point A, which is on the outer surface of the pole at the
“front” of the pole, that is, the part of the pole nearest to the
viewer.
(b) Determine the maximum tensile, compressive, and
shear stresses at point A.
6 ft
A
Julie’s Office
PROB. 8.5-15
8.5-16 A sign is supported by a pipe (see figure) having
10 in.
outer diameter 100 mm and inner diameter 80 mm. The
dimensions of the sign are 2.0 m 0.75 m, and its lower
edge is 3.0 m above the base. Note that the center of gravity
of the sign is 1.05 m from the axis of the pipe. The wind
pressure against the sign is 1.5 kPa.
Determine the maximum in-plane shear stresses due to
the wind pressure on the sign at points A, B, and C, located
on the outer surface at the base of the pipe.
8 in.
40 ft
X
X
A
2.0 m
Rose’s
Editing Co.
0.75 m
100 mm
3.0 m
B
PROB. 8.5-16
X
A
B
C
A
C
A
Section X-X
PROB. 8.5-17
Pipe
X
3 ft
Section X-X
8.5-18 A horizontal bracket ABC (see figure on the next
page) consists of two perpendicular arms AB and BC, the
latter having a length of 0.4 m. Arm AB has a solid circular
cross section with diameter equal to 60 mm. At point C a
load P1 2.02 kN acts vertically and a load P2 3.07 kN
acts horizontally and parallel to arm AB.
Considering only the forces P1 and P2, calculate the
maximum tensile stress st, the maximum compressive
stress sc, and the maximum in-plane shear stress tmax at
point p, which is located at support A on the side of the
bracket at midheight.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 8 Problems
8.5-20 For purposes of analysis, a segment of the crankshaft in a vehicle is represented as shown in the figure. The
load P equals 1.0 kN, and the dimensions are b1 80 mm,
b2 120 mm, and b3 40 mm. The diameter of the upper
shaft is d 20 mm.
(a) Determine the maximum tensile, compressive, and
shear stresses at point A, which is located on the surface of
the upper shaft at the z0 axis.
(b) Determine the maximum tensile, compressive, and
shear stresses at point B, which is located on the surface of
the shaft at the y0 axis.
y0
A
p
x0
y0
B
z0
C
0.4 m
P2
p
y0
x0
O
b1 = 80 mm
P1
60 mm
B
Cross section at A
A
PROB. 8.5-18
8.5-19 A cylindrical pressure vessel with flat ends is subjected to a torque T and a bending moment M (see figure).
The outer radius is 12.0 in. and the wall thickness is 1.0 in.
The loads are as follows: T 800 k-in., M 1000 k-in.,
and the internal pressure p 900 psi.
Determine the maximum tensile stress st, maximum
compressive stress sc, and maximum shear stress tmax in
the wall of the cylinder.
y0
M
x0
z0
b2 = 120 mm
b3 = 40 mm
PROB. 8.5-20
T
M
T
593
x0
z0
PROB. 8.5-19
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
P = 1.0 kN
9
Deflections of Beams
9.1 INTRODUCTION
When a beam with a straight longitudinal axis is loaded by lateral
forces, the axis is deformed into a curve, called the deflection curve of
the beam. In Chapter 5 we used the curvature of the bent beam to determine the normal strains and stresses in the beam. However, we did not
develop a method for finding the deflection curve itself. In this chapter,
we will determine the equation of the deflection curve and also find
deflections at specific points along the axis of the beam.
The calculation of deflections is an important part of structural
analysis and design. For example, finding deflections is an essential
ingredient in the analysis of statically indeterminate structures (Chapter
10). Deflections are also important in dynamic analyses, as when investigating the vibrations of aircraft or the response of buildings to
earthquakes.
Deflections are sometimes calculated in order to verify that they are
within tolerable limits. For instance, specifications for the design of
buildings usually place upper limits on the deflections. Large deflections
in buildings are unsightly (and even unnerving) and can cause cracks in
ceilings and walls. In the design of machines and aircraft, specifications
may limit deflections in order to prevent undesirable vibrations.
9.2 DIFFERENTIAL EQUATIONS OF THE DEFLECTION CURVE
Most procedures for finding beam deflections are based on the differential equations of the deflection curve and their associated relationships.
Consequently, we will begin by deriving the basic equation for the
deflection curve of a beam.
594
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
595
SECTION 9.2 Differential Equations of the Deflection Curve
P
A
B
(a)
v
y
A
x
B
(b)
FIG. 9-1 Deflection curve of a cantilever
beam
For discussion purposes, consider a cantilever beam with a concentrated load acting upward at the free end (Fig. 9-1a). Under the
action of this load, the axis of the beam deforms into a curve, as shown
in Fig. 9-1b. The reference axes have their origin at the fixed end of the
beam, with the x axis directed to the right and the y axis directed
upward. The z axis is directed outward from the figure (toward the
viewer).
As in our previous discussions of beam bending in Chapter 5, we
assume that the xy plane is a plane of symmetry of the beam, and we
assume that all loads act in this plane (the plane of bending).
The deflection v is the displacement in the y direction of any point
on the axis of the beam (Fig. 9-1b). Because the y axis is positive
upward, the deflections are also positive when upward.*
To obtain the equation of the deflection curve, we must express the
deflection v as a function of the coordinate x. Therefore, let us now consider the deflection curve in more detail. The deflection v at any point
m1 on the deflection curve is shown in Fig. 9-2a. Point m1 is located at
distance x from the origin (measured along the x axis). A second point
m2, located at distance x dx from the origin, is also shown. The
deflection at this second point is v dv, where dv is the increment in
deflection as we move along the curve from m1 to m2.
When the beam is bent, there is not only a deflection at each point
along the axis but also a rotation. The angle of rotation u of the axis of
the beam is the angle between the x axis and the tangent to the deflection curve, as shown for point m1 in the enlarged view of Fig. 9-2b. For
O′
du
du
m2
r
y
v
ds
m1 ds
B
x
m1
v dv
m2
A
FIG. 9-2 Deflection curve of a beam
u du
(a)
v dv
v
x
dx
u
x
dx
x
(b)
*As mentioned in Section 5.1, the traditional symbols for displacements in the x, y, and z
directions are u, v, and w, respectively. The advantage of this notation is that it emphasizes the distinction between a coordinate and a displacement.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
596
CHAPTER 9 Deflections of Beams
our choice of axes (x positive to the right and y positive upward), the
angle of rotation is positive when counterclockwise. (Other names for
the angle of rotation are angle of inclination and angle of slope.)
The angle of rotation at point m2 is u du, where du is the increase
in angle as we move from point m1 to point m2. It follows that if we
construct lines normal to the tangents (Figs. 9-2a and b), the angle
between these normals is du. Also, as discussed earlier in Section 5.3,
the point of intersection of these normals is the center of curvature O9
(Fig. 9-2a) and the distance from O to the curve is the radius of curvature r. From Fig. 9-2a we see that
y
r du ds
Positive
curvature
in which du is in radians and ds is the distance along the deflection
curve between points m1 and m2. Therefore, the curvature k (equal to
the reciprocal of the radius of curvature) is given by the equation
x
O
(a)
(a)
1
du
k
r
ds
y
Negative
curvature
x
O
(b)
FIG. 9-3 Sign convention for curvature
(9-1)
The sign convention for curvature is pictured in Fig. 9-3, which is
repeated from Fig. 5-6 of Section 5.3. Note that curvature is positive
when the angle of rotation increases as we move along the beam in the
positive x direction.
The slope of the deflection curve is the first derivative dv/dx of the
expression for the deflection v. In geometric terms, the slope is the
increment dv in the deflection (as we go from point m1 to point m2 in
Fig. 9-2) divided by the increment dx in the distance along the x axis.
Since dv and dx are infinitesimally small, the slope dv/dx is equal to the
tangent of the angle of rotation u (Fig. 9-2b). Thus,
dv
tan u
dx
dv
u arctan
dx
(9-2a,b)
In a similar manner, we also obtain the following relationships:
dx
cosu
ds
dv
sinu
ds
(9-3a,b)
Note that when the x and y axes have the directions shown in Fig. 9-2a,
the slope dv/dx is positive when the tangent to the curve slopes upward
to the right.
Equations (9-1) through (9-3) are based only upon geometric
considerations, and therefore they are valid for beams of any material.
Furthermore, there are no restrictions on the magnitudes of the slopes
and deflections.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.2 Differential Equations of the Deflection Curve
597
Beams with Small Angles of Rotation
The structures encountered in everyday life, such as buildings, automobiles, aircraft, and ships, undergo relatively small changes in shape
while in service. The changes are so small as to be unnoticed by a casual
observer. Consequently, the deflection curves of most beams and
columns have very small angles of rotation, very small deflections, and
very small curvatures. Under these conditions we can make some mathematical approximations that greatly simplify beam analysis.
Consider, for instance, the deflection curve shown in Fig. 9-2. If the
angle of rotation u is a very small quantity (and hence the deflection
curve is nearly horizontal), we see immediately that the distance ds
along the deflection curve is practically the same as the increment dx
along the x axis. This same conclusion can be obtained directly from
Eq. (9-3a). Since cos ⬇ 1 when the angle u is small, Eq. (9-3a) gives
ds dx
(b)
With this approximation, the curvature becomes (see Eq. 9-1)
1
du
k
r
dx
(9-4)
Also, since tan u u when u is small, we can make the following
approximation to Eq. (9-2a):
dv
u tanu
dx
(c)
Thus, if the rotations of a beam are small, we can assume that the angle
of rotation u and the slope dv/dx are equal. (Note that the angle of rotation must be measured in radians.)
Taking the derivative of u with respect to x in Eq. (c), we get
du
d 2v
2
dx
dx
(d)
Combining this equation with Eq. (9-4), we obtain a relation between
the curvature of a beam and its deflection:
1
d 2v
k 2
r
dx
(9-5)
This equation is valid for a beam of any material, provided the rotations
are small quantities.
If the material of a beam is linearly elastic and follows Hooke’s
law, the curvature (from Eq. 5-12, Chapter 5) is
1
M
k
r
EI
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(9-6)
598
CHAPTER 9 Deflections of Beams
in which M is the bending moment and EI is the flexural rigidity of the
beam. Equation (9-6) shows that a positive bending moment produces
positive curvature and a negative bending moment produces negative
curvature, as shown earlier in Fig. 5-10.
Combining Eq. (9-5) with Eq. (9-6) yields the basic differential
equation of the deflection curve of a beam:
d 2v
M
2
dx
EI
M
M
V
V
q
q
FIG. 9-4 Sign conventions for bending
moment M, shear force V, and intensity
q of distributed load
(9-7)
This equation can be integrated in each particular case to find the deflection v, provided the bending moment M and flexural rigidity EI are
known as functions of x.
As a reminder, the sign conventions to be used with the preceding
equations are repeated here: (1) The x and y axes are positive to the right
and upward, respectively; (2) the deflection v is positive upward; (3) the
slope dv/dx and angle of rotation u are positive when counterclockwise
with respect to the positive x axis; (4) the curvature k is positive when
the beam is bent concave upward; and (5) the bending moment M is positive when it produces compression in the upper part of the beam.
Additional equations can be obtained from the relations between
bending moment M, shear force V, and intensity q of distributed load. In
Chapter 4 we derived the following equations between M, V, and q (see
Eqs. 4-4 and 4-6):
dV
dM
q
V
(9-8a,b)
dx
dx
The sign conventions for these quantities are shown in Fig. 9-4. By
differentiating Eq. (9-7) with respect to x and then substituting the preceding equations for shear force and load, we can obtain the additional
equations. In so doing, we will consider two cases, nonprismatic beams
and prismatic beams.
Nonprismatic Beams
In the case of a nonprismatic beam, the flexural rigidity EI is variable,
and therefore we write Eq. (9-7) in the form
d 2v
EIx 2 M
(9-9a)
dx
where the subscript x is inserted as a reminder that the flexural rigidity
may vary with x. Differentiating both sides of this equation and using
Eqs. (9-8a) and (9-8b), we obtain
(9-9b)
(9-9c)
d
d 2v
dM
EIx 2 V
dx
dx
dx
d2
d 2v
dV
2 EIx 2 q
dx
dx
dx
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.2 Differential Equations of the Deflection Curve
599
The deflection of a nonprismatic beam can be found by solving (either
analytically or numerically) any one of the three preceding differential
equations. The choice usually depends upon which equation provides
the most efficient solution.
Prismatic Beams
In the case of a prismatic beam (constant EI), the differential equations
become
d 3v
EI 3 V
dx
d 2v
EI 2 M
dx
d 4v
EI 4 q
dx
(9-10a,b,c)
To simplify the writing of these and other equations, primes are often
used to denote differentiation:
d 2v
v 2
dx
dv
v
dx
d 3v
v 3
dx
d 4v
v 4
dx
(9-11)
Using this notation, we can express the differential equations for a prismatic beam in the following forms:
EIv M
EIv V
EIv q
(9-12a,b,c)
We will refer to these equations as the bending-moment equation, the
shear-force equation, and the load equation, respectively.
In the next two sections we will use the preceding equations to find
deflections of beams. The general procedure consists of integrating the
equations and then evaluating the constants of integration from boundary and other conditions pertaining to the beam.
When deriving the differential equations (Eqs. 9-9, 9-10, and 9-12),
we assumed that the material followed Hooke’s law and that the slopes
of the deflection curve were very small. We also assumed that any shear
deformations were negligible; consequently, we considered only the
deformations due to pure bending. All of these assumptions are satisfied
by most beams in common use.
Exact Expression for Curvature
If the deflection curve of a beam has large slopes, we cannot use the
approximations given by Eqs. (b) and (c). Instead, we must resort to the
exact expressions for curvature and angle of rotation (see Eqs. 9-1 and
9-2b). Combining those expressions, we get
1
du
d(arctan v) dx
k
r
ds
dx
ds
(e)
From Fig. 9-2 we see that
1/2
ds 2 dx 2 dv 2 or ds [dx 2 dv 2]
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(f,g)
600
CHAPTER 9 Deflections of Beams
Dividing both sides of Eq. (g) by dx gives
1/2
dv 2 1/2
dx
ds
1
1
[1 (v)2]
or
(h,i)
dx
ds
dx
[1 (v)2]1/2
Also, differentiation of the arctangent function (see Appendix C) gives
d
v
(arctan v) 2
( j)
dx
1 (v)
Substitution of expressions (i) and ( j) into the equation for curvature
(Eq. e) yields
1
v
(9-13)
k
r
[1 (v )2]3/2
Comparing this equation with Eq. (9-5), we see that the assumption of
small rotations is equivalent to disregarding (v)2 in comparison to one.
Equation (9-13) should be used for the curvature whenever the slopes
are large.*
9.3 DEFLECTIONS BY INTEGRATION OF THE BENDING-MOMENT EQUATION
We are now ready to solve the differential equations of the deflection
curve and obtain deflections of beams. The first equation we will use is
the bending-moment equation (Eq. 9-12a). Since this equation is of second order, two integrations are required. The first integration produces
the slope v dv/dx, and the second produces the deflection v.
We begin the analysis by writing the equation (or equations) for the
bending moments in the beam. Since only statically determinate beams
are considered in this chapter, we can obtain the bending moments from
free-body diagrams and equations of equilibrium, using the procedures
described in Chapter 4. In some cases a single bending-moment expression holds for the entire length of the beam, as illustrated in Examples
9-1 and 9-2. In other cases the bending moment changes abruptly at one
or more points along the axis of the beam. Then we must write separate
bending-moment expressions for each region of the beam between
points where changes occur, as illustrated in Example 9-3.
Regardless of the number of bending-moment expressions, the general procedure for solving the differential equations is as follows. For
each region of the beam, we substitute the expression for M into the differential equation and integrate to obtain the slope v. Each such
*The basic relationship stating that the curvature of a beam is proportional to the bending
moment (Eq. 9-6) was first obtained by Jacob Bernoulli, although he obtained an incorrect value for the constant of proportionality. The relationship was used later by Euler,
who solved the differential equation of the deflection curve for both large deflections
(using Eq. 9-13) and small deflections (using Eq. 9-7). For the history of deflection
curves, see Ref. 9-1.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.3 Deflections by Integration of the Bending-Moment Equation
A
B
A
B
vA = 0
vB = 0
FIG. 9-5 Boundary conditions at simple
supports
A
B
A
B
vA = 0
v ′A = 0
FIG. 9-6 Boundary conditions at a fixed
support
A
C
B
A
B
C
At point C: (v)AC = (v)CB
(v′ )AC = (v′ )CB
FIG. 9-7 Continuity conditions at
point C
601
integration produces one constant of integration. Next, we integrate each
slope equation to obtain the corresponding deflection v. Again, each
integration produces a new constant. Thus, there are two constants of
integration for each region of the beam. These constants are evaluated
from known conditions pertaining to the slopes and deflections. The
conditions fall into three categories: (1) boundary conditions, (2) continuity conditions, and (3) symmetry conditions.
Boundary conditions pertain to the deflections and slopes at the
supports of a beam. For example, at a simple support (either a pin or a
roller) the deflection is zero (Fig. 9-5), and at a fixed support both the
deflection and the slope are zero (Fig. 9-6). Each such boundary condition supplies one equation that can be used to evaluate the constants of
integration.
Continuity conditions occur at points where the regions of integration meet, such as at point C in the beam of Fig. 9-7. The deflection
curve of this beam is physically continuous at point C, and therefore the
deflection at point C as determined for the left-hand part of the beam
must be equal to the deflection at point C as determined for the righthand part. Similarly, the slopes found for each part of the beam must be
equal at point C. Each of these continuity conditions supplies an equation for evaluating the constants of integration.
Symmetry conditions may also be available. For instance, if a simple beam supports a uniform load throughout its length, we know in
advance that the slope of the deflection curve at the midpoint must be
zero. This condition supplies an additional equation, as illustrated in
Example 9-1.
Each boundary, continuity, and symmetry condition leads to an
equation containing one or more of the constants of integration. Since
the number of independent conditions always matches the number of
constants of integration, we can always solve these equations for the
constants. (The boundary and continuity conditions alone are always
sufficient to determine the constants. Any symmetry conditions provide
additional equations, but they are not independent of the other equations. The choice of which conditions to use is a matter of convenience.)
Once the constants are evaluated, they can be substituted back into
the expressions for slopes and deflections, thus yielding the final equations of the deflection curve. These equations can then be used to obtain
the deflections and angles of rotation at particular points along the axis
of the beam.
The preceding method for finding deflections is sometimes called
the method of successive integrations. The following examples illustrate the method in detail.
Note: When sketching deflection curves, such as those shown in the
following examples and in Figs. 9-5, 9-6, and 9-7, we greatly exaggerate
the deflections for clarity. However, it should always be kept in mind
that the actual deflections are very small quantities.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
602
CHAPTER 9 Deflections of Beams
Example 9-1
Determine the equation of the deflection curve for a simple beam AB supporting
a uniform load of intensity q acting throughout the span of the beam (Fig. 9-8a).
Also, determine the maximum deflection dmax at the midpoint of the beam
and the angles of rotation uA and uB at the supports (Fig. 9-8b). (Note: The beam
has length L and constant flexural rigidity EI.)
q
A
B
L
Solution
(a)
Bending moment in the beam. The bending moment at a cross section distance x from the left-hand support is obtained from the free-body diagram of
Fig. 9-9. Since the reaction at the support is qL /2, the equation for the bending
moment is
y
d max
A
B
x
qL
qLx
qx 2
x
M (x) qx
2
2
2
2
uB
uA
L
—
2
L
—
2
(9-14)
Differential equation of the deflection curve. By substituting the expression for the bending moment (Eq. 9-14) into the differential equation
(Eq. 9-12a), we obtain
(b)
FIG. 9-8 Example 9-1. Deflections of a
simple beam with a uniform load
qLx
qx 2
EIv
2
2
q
M
This equation can now be integrated to obtain the slope and deflection of the
beam.
Slope of the beam. Multiplying both sides of the differential equation by
dx, we get the following equation:
A
qL x
qx2
EIv dx dx dx
2
2
V
qL
—
2
(9-15)
x
Integrating each term, we obtain
FIG. 9-9 Free-body diagram used in
qLx
qx2
EI v dx dx dx
2
2
determining the bending moment M
(Example 9-1)
or
qLx 2
qx 3
EIv C1
4
6
(a)
in which C1 is a constant of integration.
To evaluate the constant C1, we observe from the symmetry of the beam
and its load that the slope of the deflection curve at midspan is equal to zero.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
603
SECTION 9.3 Deflections by Integration of the Bending-Moment Equation
Thus, we have the following symmetry condition:
L
v 0 when x
2
This condition may be expressed more succinctly as
L
v 0
2
Applying this condition to Eq. (a) gives
qL L
0
4 2
2
q L
6 2
3
qL3
C1 or C1
24
The equation for the slope of the beam (Eq. a) then becomes
qL x 2
qx 3
qL3
EIv
4
6
24
or
q
v (L3 6Lx 2 4x 3)
24EI
(b)
(9-16)
As expected, the slope is negative (i.e., clockwise) at the left-hand end of the
beam (x 0), positive at the right-hand end (x L), and equal to zero at the
midpoint (x L /2).
Deflection of the beam. The deflection is obtained by integrating the equation for the slope. Thus, upon multiplying both sides of Eq. (b) by dx and
integrating, we obtain
qL x 3
qx4
qL 3x
EIv C2
12
24
24
(c)
The constant of integration C2 may be evaluated from the condition that the
deflection of the beam at the left-hand support is equal to zero; that is, v 0
when x 0, or
v(0) 0
Applying this condition to Eq. (c) yields C2 0; hence the equation for the
deflection curve is
qLx 3
q x4
qL 3x
EIv
12
24
24
or
qx
v (L3 2Lx 2 x 3)
24EI
(d)
(9-17)
This equation gives the deflection at any point along the axis of the beam. Note
that the deflection is zero at both ends of the beam (x 0 and x L) and negative elsewhere (recall that downward deflections are negative).
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
604
CHAPTER 9 Deflections of Beams
Maximum deflection. From symmetry we know that the maximum deflection occurs at the midpoint of the span (Fig. 9-8b). Thus, setting x equal to L/2
in Eq. (9-17), we obtain
5qL 4
L
v
2
384EI
in which the negative sign means that the deflection is downward (as expected).
Since d max represents the magnitude of this deflection, we obtain
L
dmax v
2
5qL 4
384 EI
(9-18)
Angles of rotation. The maximum angles of rotation occur at the supports
of the beam. At the left-hand end of the beam, the angle uA, which is a clockwise angle (Fig. 9-8b), is equal to the negative of the slope v9. Thus, by
substituting x 0 into Eq. (9-16), we find
qL3
uA v(0)
24EI
(9-19)
In a similar manner, we can obtain the angle of rotation uB at the right-hand end
of the beam. Since uB is a counterclockwise angle, it is equal to the slope at the
end:
qL3
uB v(L)
24EI
(9-20)
Because the beam and loading are symmetric about the midpoint, the angles of
rotation at the ends are equal.
This example illustrates the process of setting up and solving the differential equation of the deflection curve. It also illustrates the process of finding
slopes and deflections at selected points along the axis of a beam.
Note: Now that we have derived formulas for the maximum deflection and
maximum angles of rotation (see Eqs. 9-18, 9-19, and 9-20), we can evaluate
those quantities numerically and observe that the deflections and angles are
indeed small, as the theory requires.
Consider a steel beam on simple supports with a span length L 6 ft. The
cross section is rectangular with width b 3 in. and height h 6 in. The intensity of uniform load is q 8000 lb/ft, which is relatively large because it
produces a stress in the beam of 24,000 psi. (Thus, the deflections and slopes
are larger than would normally be expected.)
Substituting into Eq. (9-18), and using E 30 106 psi, we find that the
maximum deflection is dmax 0.144 in., which is only 1/500 of the span length.
Also, from Eq. (9-19), we find that the maximum angle of rotation is uA
0.0064 radians, or 0.37°, which is a very small angle.
Thus, our assumption that the slopes and deflections are small is validated.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
605
SECTION 9.3 Deflections by Integration of the Bending-Moment Equation
Example 9-2
Determine the equation of the deflection curve for a cantilever beam AB subjected to a uniform load of intensity q (Fig. 9-10a).
Also, determine the angle of rotation uB and the deflection dB at the free
end (Fig. 9-10b). (Note: The beam has length L and constant flexural rigidity
EI.)
y
q
B
A
A
B
L
FIG. 9-10 Example 9-2. Deflections of a
cantilever beam with a uniform load
x
dB
(b)
(a)
uB
Solution
Bending moment in the beam. The bending moment at distance x from the
fixed support is obtained from the free-body diagram of Fig. 9-11. Note that the
vertical reaction at the support is equal to qL and the moment reaction is equal
to qL2/2. Consequently, the expression for the bending moment M is
qL2
qx2
M qLx
2
2
(9-21)
Differential equation of the deflection curve. When the preceding expression for the bending moment is substituted into the differential equation
(Eq. 9-12a), we obtain
qL2
qx 2
EIv qLx
2
2
q
M
qL2
—
2 qL
A
V
x
FIG. 9-11 Free-body diagram used in
determining the bending moment M
(Example 9-2)
(9-22)
We now integrate both sides of this equation to obtain the slopes and deflections.
Slope of the beam. The first integration of Eq. (9-22) gives the following
equation for the slope:
qL2x
qL x 2
qx3
EIv C1
2
2
6
(e)
The constant of integration Cl can be found from the boundary condition
that the slope of the beam is zero at the support; thus, we have the following
condition:
v(0) 0
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
606
CHAPTER 9 Deflections of Beams
When this condition is applied to Eq. (e) we get C1 0. Therefore, Eq. (e)
becomes
qL2x
qL x 2
qx 3
EIv
2
2
6
(f)
and the slope is
qx
v (3L2 3Lx x 2)
6EI
(9-23)
As expected, the slope obtained from this equation is zero at the support (x 0)
and negative (i.e., clockwise) throughout the length of the beam.
Deflection of the beam. Integration of the slope equation (Eq. f) yields
qL2x 2
qL x 3
qx4
EIv C2
4
6
24
(g)
The constant C2 is found from the boundary condition that the deflection of the
beam is zero at the support:
v(0) 0
When this condition is applied to Eq. (g), we see immediately that C2 0.
Therefore, the equation for the deflection v is
qx 2
v (6L2 4Lx x2)
24EI
(9-24)
As expected, the deflection obtained from this equation is zero at the support
(x 0) and negative (that is, downward) elsewhere.
Angle of rotation at the free end of the beam. The clockwise angle of rotation uB at end B of the beam (Fig. 9-10b) is equal to the negative of the slope at
that point. Thus, using Eq. (9-23), we get
qL 3
uB v(L)
6EI
(9-25)
This angle is the maximum angle of rotation for the beam.
Deflection at the free end of the beam. Since the deflection dB is downward (Fig. 9-10b), it is equal to the negative of the deflection obtained from Eq.
(9-24):
qL 4
dB v(L)
8EI
This deflection is the maximum deflection of the beam.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(9-26)
SECTION 9.3 Deflections by Integration of the Bending-Moment Equation
607
Example 9-3
A simple beam AB supports a concentrated load P acting at distances a and b
from the left-hand and right-hand supports, respectively (Fig. 9-12a).
Determine the equations of the deflection curve, the angles of rotation
uA and uB at the supports, the maximum deflection dmax, and the deflection dC
at the midpoint C of the beam (Fig. 9-12b). (Note: The beam has length L and
constant flexural rigidity EI.)
y
P
A
a
uA
A
B
L
—
2
b
B x
d max
x1
L
FIG. 9-12 Example 9-3. Deflections of
a simple beam with a concentrated load
uB
C D
dC
(a)
(b)
Solution
Pb
—
L
Bending moments in the beam. In this example the bending moments are
expressed by two equations, one for each part of the beam. Using the free-body
diagrams of Fig. 9-13, we arrive at the following equations:
M
A
x
Pbx
M
L
V
x
a
P
a
M
A
V
x
x
x
(9-27a)
a)
Pbx
(a x L)
(9-27b)
M P(x a)
L
Differential equations of the deflection curve. The differential equations
for the two parts of the beam are obtained by substituting the bending-moment
expressions (Eqs. 9-27a and b) into Eq. (9-12a). The results are
(a)
Pb
—
L
(0
Pbx
EIv
L
(0
x
(9-28a)
a)
Pbx
EIv P(x a)
L
(a
x
(9-28b)
L)
Slopes and deflections of the beam. The first integrations of the two differential equations yield the following expressions for the slopes:
a
(b)
FIG. 9-13 Free-body diagrams used in
determining the bending moments
(Example 9-3)
Pb x 2
EIv C1
2L
(0
x
P(x a) 2
Pbx2
EIv C2
2L
2
(h)
a)
(a
x
L)
(i)
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
608
CHAPTER 9 Deflections of Beams
in which C1 and C2 are constants of integration. A second pair of integrations
gives the deflections:
Pb x3
EIv C1x C3
6L
(0
x
( j)
a)
P(x a)3
Pb x3
EIv C2 x C4
6L
6
(a
x
L)
(k)
These equations contain two additional constants of integration, making a total
of four constants to be evaluated.
Constants of integration. The four constants of integration can be found
from the following four conditions:
1.
2.
3.
4.
At x a, the slopes v for the two parts of the beam are the same.
At x a, the deflections v for the two parts of the beam are the same.
At x 0, the deflection v is zero.
At x L, the deflection v is zero.
The first two conditions are continuity conditions based upon the fact that
the axis of the beam is a continuous curve. Conditions (3) and (4) are boundary
conditions that must be satisfied at the supports.
Condition (1) means that the slopes determined from Eqs. (h) and (i) must
be equal when x a; therefore,
Pba 2
Pba 2
C1 C2 or C1 C 2
2L
2L
Condition (2) means that the deflections found from Eqs. ( j) and (k) must be
equal when x a; therefore,
Pba 3
Pba 3
C1a C3 C2a C4
6L
6L
Inasmuch as C1 C2, this equation gives C3 C4.
Next, we apply condition (3) to Eq. ( j) and obtain C3 0; therefore,
C3 C4 0
(l)
Finally, we apply condition (4) to Eq. (k) and obtain
PbL2
Pb 3
C2L 0
6
6
Therefore,
Pb (L2 b 2)
C1 C2
6L
(m)
Equations of the deflection curve. We now substitute the constants of integration (Eqs. l and m) into the equations for the deflections (Eqs. j and k) and
obtain the deflection equations for the two parts of the beam. The resulting
equations, after a slight rearrangement, are
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
609
SECTION 9.3 Deflections by Integration of the Bending-Moment Equation
Pbx
v (L2 b2 x 2)
6LEI
(0
x
(9-29a)
a)
P(x a)3
Pbx
v (L2 b 2 x 2)
6LEI
6EI
(a
x
L)
(9-29b)
The first of these equations gives the deflection curve for the part of the beam to
the left of the load P, and the second gives the deflection curve for the part of
the beam to the right of the load.
The slopes for the two parts of the beam can be found either by substituting
the values of Cl and C2 into Eqs. (h) and (i) or by taking the first derivatives of
the deflection equations (Eqs. 9-29a and b). The resulting equations are
Pb
v (L2 b2 3x 2 )
6LEI
(0
x
(9-30a)
a)
P(x a)2
Pb
v (L2 b 2 3x 2 )
6LEI
2EI
(a
x
L)
(9-30b)
The deflection and slope at any point along the axis of the beam can be calculated from Eqs. (9-29) and (9-30).
Angles of rotation at the supports. To obtain the angles of rotation uA and
uB at the ends of the beam (Fig. 9-12b), we substitute x 0 into Eq. (9-30a) and
x L into Eq. (9-30b):
Pb(L2 b 2)
Pab (L b)
uA v(0)
6LEI
6LEI
(9-31a)
Pab (L a)
Pb(2L2 3bL b 2)
uB v(L) (9-31b)
6LEI
6LEI
Note that the angle uA is clockwise and the angle uB is counterclockwise, as
shown in Fig. 9-12b.
The angles of rotation are functions of the position of the load and reach
their largest values when the load is located near the midpoint of the beam. In
the case of the angle of rotation uA, the maximum value of the angle is
PL2 3
(uA)max
27EI
(9-32)
and occurs when b L/ 3 0.577L (or a 0.423L). This value of b is
obtained by taking the derivative of uA with respect to b (using the first of the
two expressions for uA in Eq. 9-31a) and then setting it equal to zero.
Maximum deflection of the beam. The maximum deflection d max occurs at
point D (Fig. 9-12b) where the deflection curve has a horizontal tangent. If the
load is to the right of the midpoint, that is, if a b, point D is in the part of the
beam to the left of the load. We can locate this point by equating the slope v
from Eq. (9-30a) to zero and solving for the distance x, which we now denote as
x1. In this manner we obtain the following formula for xl:
x1
L2 b2
3
(a
b)
(9-33)
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
610
CHAPTER 9 Deflections of Beams
From this equation we see that as the load P moves from the middle of the beam
(b L /2) to the right-hand end (b 0), the distance xl varies from L /2 to
L/ 3 0.577L. Thus, the maximum deflection occurs at a point very close to
the midpoint of the beam, and this point is always between the midpoint of the
beam and the load.
The maximum deflection d max is found by substituting xl (from Eq. 9-33)
into the deflection equation (Eq. 9-29a) and then inserting a minus sign:
Pb(L2 b 2)3/2
d max (v) x x1
9 3 LEI
(a
b)
(9-34)
The minus sign is needed because the maximum deflection is downward
(Fig. 9-12b) whereas the deflection v is positive upward.
The maximum deflection of the beam depends on the position of the load
P, that is, on the distance b. The maximum value of the maximum deflection
(the “max-max” deflection) occurs when b L /2 and the load is at the midpoint of the beam. This maximum deflection is equal to PL3/48EI.
Deflection at the midpoint of the beam. The deflection dC at the midpoint
C when the load is acting to the right of the midpoint (Fig. 9-12b) is obtained by
substituting x L/2 into Eq. (9-29a), as follows:
L
Pb(3L2 4b 2)
dC v
2
48EI
(a
b)
(9-35)
Because the maximum deflection always occurs near the midpoint of the beam,
Eq. (9-35) yields a close approximation to the maximum deflection. In the most
unfavorable case (when b approaches zero), the difference between the maximum deflection and the deflection at the midpoint is less than 3% of the
maximum deflection, as demonstrated in Problem 9.3-7.
Special case (load at the midpoint of the beam). An important special case
occurs when the load P acts at the midpoint of the beam (a b L /2). Then
we obtain the following results from Eqs. (9-30a), (9-29a), (9-31), and (9-34),
respectively:
P
v (L2 4x2)
16EI
0
x
L
2
(9-36)
Px
v (3L2 4x2)
48EI
0
x
L
2
(9-37)
PL2
uA uB
16EI
(9-38)
PL3
dmax dC
48EI
(9-39)
Since the deflection curve is symmetric about the midpoint of the beam,
the equations for v and v are given only for the left-hand half of the beam
(Eqs. 9-36 and 9-37). If needed, the equations for the right-hand half can be
obtained from Eqs. (9-30b) and (9-29b) by substituting a b L /2.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.4 Deflections by Integration of the Shear-Force and Load Equations
611
9.4 DEFLECTIONS BY INTEGRATION OF THE SHEAR-FORCE AND LOAD EQUATIONS
The equations of the deflection curve in terms of the shear force V and
the load q (Eqs. 9-12b and c, respectively) may also be integrated to
obtain slopes and deflections. Since the loads are usually known quantities, whereas the bending moments must be determined from free-body
diagrams and equations of equilibrium, many analysts prefer to start
with the load equation. For this same reason, most computer programs
for finding deflections begin with the load equation and then perform
numerical integrations to obtain the shear forces, bending moments,
slopes, and deflections.
The procedure for solving either the load equation or the shear-force
equation is similar to that for solving the bending-moment equation,
except that more integrations are required. For instance, if we begin with
the load equation, four integrations are needed in order to arrive at the
deflections. Thus, four constants of integration are introduced for each
load equation that is integrated. As before, these constants are found
from boundary, continuity, and symmetry conditions. However, these
conditions now include conditions on the shear forces and bending
moments as well as conditions on the slopes and deflections.
Conditions on the shear forces are equivalent to conditions on the
third derivative (because EIv V). In a similar manner, conditions on
the bending moments are equivalent to conditions on the second derivative (because EIv M). When the shear-force and bending-moment
conditions are added to those for the slopes and deflections, we always
have enough independent conditions to solve for the constants of
integration.
The following examples illustrate the techniques of analysis in
detail. The first example begins with the load equation and the second
begins with the shear-force equation.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
612
CHAPTER 9 Deflections of Beams
Example 9-4
y
q0
x
B
A
Determine the equation of the deflection curve for a cantilever beam AB supporting a triangularly distributed load of maximum intensity q0 (Fig. 9-14a).
Also, determine the deflection dB and angle of rotation uB at the free end
(Fig. 9-14b). Use the fourth-order differential equation of the deflection
curve (the load equation). (Note: The beam has length L and constant flexural
rigidity EI.)
Solution
L
Differential equation of the deflection curve. The intensity of the distributed load is given by the following equation (see Fig. 9-14a):
(a)
q0(L x)
q
L
y
A
B
x
dB
uB
(b)
(9-40)
Consequently, the fourth-order differential equation (Eq. 9-12c) becomes
q0(L x)
EIv q
L
(a)
Shear force in the beam. The first integration of Eq. (a) gives
FIG. 9-14 Example 9-4. Deflections of a
cantilever beam with a triangular load
q0
EIv (L x)2 C1
2L
(b)
The right-hand side of this equation represents the shear force V (see Eq.
9-12b). Because the shear force is zero at x L, we have the following boundary condition:
v(L) 0
Using this condition with Eq. (b), we get C1 0. Therefore, Eq. (b) simplifies to
q0
EIv (L x) 2
2L
(c)
and the shear force in the beam is
q0
V EIv (L x) 2
2L
(9-41)
Bending moment in the beam. Integrating a second time, we obtain the
following equation from Eq. (c):
q0
EIv (L x) 3 C2
6L
(d)
This equation is equal to the bending moment M (see Eq. 9-12a). Since the
bending moment is zero at the free end of the beam, we have the following
boundary condition:
v(L) 0
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.4 Deflections by Integration of the Shear-Force and Load Equations
613
Applying this condition to Eq. (d), we obtain C2 0, and therefore the bending
moment is
q0
M EIv (L x) 3
6L
(9-42)
Slope and deflection of the beam. The third and fourth integrations yield
q0
EIv (L x) 4 C3
24L
(e)
q0
EIv (L x) 5 C3 x C4
120L
(f)
The boundary conditions at the fixed support, where both the slope and deflection equal zero, are
v(0) 0
v(0) 0
Applying these conditions to Eqs. (e) and (f), respectively, we find
q0 L3
C3
24
q0 L4
C 4
120
Substituting these expressions for the constants into Eqs. (e) and (f ), we obtain
the following equations for the slope and deflection of the beam:
q0 x
v (4L3 6L2x 4Lx 2 x3)
24LEI
(9-43)
q0 x2
v (10L3 10L2x 5Lx 2 x 3)
120LEI
(9-44)
Angle of rotation and deflection at the free end of the beam. The angle of
rotation uB and deflection dB at the free end of the beam (Fig. 9-14b) are
obtained from Eqs. (9-43) and (9-44), respectively, by substituting x L. The
results are
q0 L3
uB v(L)
24EI
q0 L4
30EI
dB v(L)
(9-45a,b)
Thus, we have determined the required slopes and deflections of the beam by
solving the fourth-order differential equation of the deflection curve.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
614
CHAPTER 9 Deflections of Beams
Example 9-5
A simple beam AB with an overhang BC supports a concentrated load P at the
end of the overhang (Fig. 9-15a). The main span of the beam has length L and
the overhang has length L / 2.
Determine the equations of the deflection curve and the deflection dC at the
end of the overhang (Fig. 9-15b). Use the third-order differential equation of the
deflection curve (the shear-force equation). (Note: The beam has constant
flexural rigidity EI.)
P
A
B
y
C
B
A
C
x
dC
P
—
2
FIG. 9-15 Example 9-5.
Deflections of a beam with
an overhang
L
3P
—
2
L
—
2
(b)
(a)
Solution
Differential equations of the deflection curve. Because reactive forces act
at supports A and B, we must write separate differential equations for parts AB
and BC of the beam. Therefore, we begin by finding the shear forces in each
part of the beam.
The downward reaction at support A is equal to P/2, and the upward reaction at support B is equal to 3P/2 (see Fig. 9-15a). It follows that the shear
forces in parts AB and BC are
P
V
2
VP
(0
L
x
L)
(9-46a)
(9-46b)
3L
2
x
in which x is measured from end A of the beam (Fig. 9-12b).
The third-order differential equations for the beam now become (see
Eq. 9-12b):
P
EIv
2
EIv P
(0
L
x
x
L)
(g)
(h)
3L
2
Bending moments in the beam. Integration of the preceding two equations
yields the bending-moment equations:
Px
M EIv C1
2
M EIv Px C2
(0
L
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
x
x
L)
(i)
(j)
3L
2
SECTION 9.4 Deflections by Integration of the Shear-Force and Load Equations
615
The bending moments at points A and C are zero; hence we have the following
boundary conditions:
3L
v 0
2
v(0) 0
Using these conditions with Eqs. (i) and ( j), we get
C1 0
3PL
C2
2
Therefore, the bending moments are
Px
M EIv
2
(0
x
L
P(3L 2x)
M EIv
2
(9-47a)
L)
3L
2
x
(9-47b)
These equations can be verified by determining the bending moments from freebody diagrams and equations of equilibrium.
Slopes and deflections of the beam. The next integrations yield the slopes:
Px2
EIv C3
4
(0
Px(3L x)
EIv C4
2
x
L)
L
x
3L
2
The only condition on the slopes is the continuity condition at support B.
According to this condition, the slope at point B as found for part AB of the
beam is equal to the slope at the same point as found for part BC of the beam.
Therefore, we substitute x L into each of the two preceding equations for the
slopes and obtain
PL2
C3 PL2 C4
4
This equation eliminates one constant of integration because we can express C4
in terms of C3:
3PL2
C4 C3
4
(k)
The third and last integrations give
Px3
EIv C3 x C5
12
(0
Px 2(9L 2x)
EIv C4 x C6
12
x
L
(l)
L)
x
3L
2
(m)
For part AB of the beam (Fig. 9-15a), we have two boundary conditions on the
deflections, namely, the deflection is zero at points A and B:
v(0) 0 and v(L) 0
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
616
CHAPTER 9 Deflections of Beams
P
A
B
C
y
B
A
C
x
dC
P
—
2
FIG. 9-15 (Repeated)
L
3P
—
2
L
—
2
(b)
(a)
Applying these conditions to Eq. (l), we obtain
C5 0
PL 2
C3
12
(n,o)
Substituting the preceding expression for C3 in Eq. (k), we get
5PL2
C4
6
(p)
For part BC of the beam, the deflection is zero at point B. Therefore, the boundary condition is
v(L) 0
Applying this condition to Eq. (m), and also substituting Eq. (p) for C4, we get
PL3
C6
4
(q)
All constants of integration have now been evaluated.
The deflection equations are obtained by substituting the constants of
integration (Eqs. n, o, p, and q) into Eqs. (l) and (m). The results are
Px
v (L 2 x2)
12EI
(0
P
v (3L 3 10L 2x 9Lx 2 2x 3)
12EI
x
L
(9-48a)
L)
x
3L
2
(9-48b)
Note that the deflection is always positive (upward) in part AB of the beam
(Eq. 9-48a) and always negative (downward) in the overhang BC (Eq. 9-48b).
Deflection at the end of the overhang. We can find the deflection dC at the
end of the overhang (Fig. 9-15b) by substituting x 3L /2 in Eq. (9-48b):
3L
PL3
dC v
2
8EI
(9-49)
Thus, we have determined the required deflections of the overhanging beam
(Eqs. 9-48 and 9-49) by solving the third-order differential equation of the
deflection curve.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.5 Method of Superposition
617
9.5 METHOD OF SUPERPOSITION
The method of superposition is a practical and commonly used technique for obtaining deflections and angles of rotation of beams. The
underlying concept is quite simple and may be stated as follows:
Under suitable conditions, the deflection of a beam produced by
several different loads acting simultaneously can be found by superposing the deflections produced by the same loads acting separately.
For instance, if v1 represents the deflection at a particular point on
the axis of a beam due to a load q1, and if v2 represents the deflection at
that same point due to a different load q2, then the deflection at that
point due to loads q1 and q2 acting simultaneously is v1 v2. (The loads
q1 and q2 are independent loads and each may act anywhere along the
axis of the beam.)
The justification for superposing deflections lies in the nature of the
differential equations of the deflection curve (Eqs. 9-12a, b, and c).
These equations are linear differential equations, because all terms containing the deflection v and its derivatives are raised to the first power.
Therefore, the solutions of these equations for several loading conditions may be added algebraically, or superposed.(The conditions for
superposition to be valid are described later in the subsection “Principle
of Superposition.”)
As an illustration of the superposition method, consider the simple
beam ACB shown in Fig. 9-16a. This beam supports two loads: (1) a
uniform load of intensity q acting throughout the span, and (2) a concentrated load P acting at the midpoint. Suppose we wish to find the
deflection dC at the midpoint and the angles of rotation uA and uB at the
ends (Fig. 9-16b). Using the method of superposition, we obtain the
effects of each load acting separately and then combine the results.
For the uniform load acting alone, the deflection at the midpoint and
the angles of rotation are obtained from the formulas of Example 9-1
(see Eqs. 9-18, 9-19, and 9-20):
P
q
A
B
C
L
—
2
L
—
2
(a)
y
C
A
B
dC
uA
qL3
(uA)1 (uB)1
24EI
in which EI is the flexural rigidity of the beam and L is its length.
For the load P acting alone, the corresponding quantities are
obtained from the formulas of Example 9-3 (see Eqs. 9-38 and 9-39):
uB
L
—
2
x
5qL4
(dC)1
384EI
L
—
2
PL3
(dC)2
48EI
(b)
FIG. 9-16 Simple beam with two loads
PL2
(uA)2 (uB)2
16EI
The deflection and angles of rotation due to the combined loading (Fig.
9-16a) are obtained by summation:
4
5qL
PL3
dC (dC )1 (dC )2
384EI
48EI
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(a)
618
CHAPTER 9 Deflections of Beams
qL3
PL2
uA uB (uA)1 (uA)2
(b)
24EI
16EI
The deflections and angles of rotation at other points on the beam axis
can be found by this same procedure. However, the method of superposition is not limited to finding deflections and angles of rotation at single
points. The method may also be used to obtain general equations for the
slopes and deflections of beams subjected to more than one load.
Tables of Beam Deflections
The method of superposition is useful only when formulas for deflections
and slopes are readily available. To provide convenient access to such
formulas, tables for both cantilever and simple beams are given in
Appendix G at the back of the book. Similar tables can be found in
engineering handbooks. Using these tables and the method of superposition, we can find deflections and angles of rotation for many different
loading conditions, as illustrated in the examples at the end of this section.
q0
C
A
B
L
—
2
Distributed Loads
L
—
2
(a)
q dx
y
C
A
x
B
x
dx
(b)
y
dC
C
A
B
uA
L
—
2
L
—
2
(c)
FIG. 9-17 Simple beam with a triangular
load
x
Sometimes we encounter a distributed load that is not included in a table
of beam deflections. In such cases, superposition may still be useful. We
can consider an element of the distributed load as though it were a concentrated load, and then we can find the required deflection by
integrating throughout the region of the beam where the load is applied.
To illustrate this process of integration, consider a simple beam ACB
with a triangular load acting on the left-hand half (Fig. 9-17a). We wish
to obtain the deflection dC at the midpoint C and the angle of rotation uA
at the left-hand support (Fig. 9-17c).
We begin by visualizing an element q dx of the distributed load as a
concentrated load (Fig. 9-17b). Note that the load acts to the left of the
midpoint of the beam. The deflection at the midpoint due to this concentrated load is obtained from Case 5 of Table G-2, Appendix G. The
formula given there for the midpoint deflection (for the case in which
a b) is
Pa
(3L2 4a 2 )
48EI
In our example (Fig. 9-17b), we substitute qdx for P and x for a:
(q dx)(x)
(3L2 4x 2 )
(c)
4 8EI
This expression gives the deflection at point C due to the element q dx of
the load.
Next, we note that the intensity of the uniform load (Figs. 9-17a
and b) is
2q0 x
(d)
q
L
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
619
SECTION 9.5 Method of Superposition
where q0 is the maximum intensity of the load. With this substitution for
q, the formula for the deflection (Eq. c) becomes
q0 x 2
(3L2 4x 2 )dx
24LEI
Finally, we integrate throughout the region of the load to obtain the
deflection dC at the midpoint of the beam due to the entire triangular
load:
L/2
dC
0
q0 x2
(3L2 4x 2 )dx
24LEI
q0
24LEI
L/2
0
q0 L4
(3L2 4x 2 ) x 2 dx
240EI
(9-50)
By a similar procedure, we can calculate the angle of rotation uA at
the left-hand end of the beam (Fig. 9-17c). The expression for this angle
due to a concentrated load P (see Case 5 of Table G-2) is
Pab(L b)
6LEI
Replacing P with 2q0 x dx/L, a with x, and b with L x, we obtain
2q 0 x 2(L x)(L L x)
q0
dx or 2 (L x)(2L x)x 2 dx
6L2EI
3L EI
Finally, we integrate throughout the region of the load:
L/2
uA
0
q0
41q0 L3
2 (L x)(2L x)x 2 dx
3L EI
2880EI
(9-51)
This is the angle of rotation produced by the triangular load.
This example illustrates how we can use superposition and integration to find deflections and angles of rotation produced by distributed
loads of almost any kind. If the integration cannot be performed easily
by analytical means, numerical methods can be used.
Principle of Superposition
The method of superposition for finding beam deflections is an example
of a more general concept known in mechanics as the principle of
superposition. This principle is valid whenever the quantity to be determined is a linear function of the applied loads. When that is the case, the
desired quantity may be found due to each load acting separately, and
then these results may be superposed to obtain the desired quantity due
to all loads acting simultaneously. In ordinary structures, the principle is
usually valid for stresses, strains, bending moments, and many other
quantities besides deflections.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
620
CHAPTER 9 Deflections of Beams
In the particular case of beam deflections, the principle of superposition is valid under the following conditions: (1) Hooke’s law holds
for the material, (2) the deflections and rotations are small, and (3) the
presence of the deflections does not alter the actions of the applied
loads. These requirements ensure that the differential equations of the
deflection curve are linear.
The following examples provide additional illustrations in which
the principle of superposition is used to calculate deflections and angles
of rotation of beams.
Example 9-6
q
A cantilever beam AB supports a uniform load of intensity q acting over part of
the span and a concentrated load P acting at the free end (Fig. 9-18a).
Determine the deflection dB and angle of rotation uB at end B of the beam
(Fig. 9-18b). (Note: The beam has length L and constant flexural rigidity EI.)
P
B
A
a
b
Solution
We can obtain the deflection and angle of rotation at end B of the beam by
combining the effects of the loads acting separately. If the uniform load acts
alone, the deflection and angle of rotation (obtained from Case 2 of Table G-1,
Appendix G) are
L
(a)
qa3
(dB)1 (4L a)
24EI
y
B
A
x
dB
(b)
uB
FIG. 9-18 Example 9-6. Cantilever beam
with a uniform load and a concentrated
load
qa3
(uB)1
6EI
If the load P acts alone, the corresponding quantities (from Case 4, Table G-1)
are
PL3
(dB) 2
3EI
PL2
(uB)2
2EI
Therefore, the deflection and angle of rotation due to the combined loading
(Fig. 9-18a) are
3
qa
PL3
dB (dB)1 (dB ) 2 (4L a)
24EI
3EI
(9-52)
3
qa
PL2
uB (uB )1 (uB )2
2EI
6EI
(9-53)
Thus, we have found the required quantities by using tabulated formulas and the
method of superposition.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
621
SECTION 9.5 Method of Superposition
Example 9-7
A cantilever beam AB with a uniform load of intensity q acting on the righthand half of the beam is shown in Fig. 9-19a.
Obtain formulas for the deflection dB and angle of rotation uB at the
free end (Fig. 9-19c). (Note: The beam has length L and constant flexural
rigidity EI.)
( )
y
q
A
B
A
B
L
—
2
q dx
x
dx
x
L
—
2
(b)
(a)
y
B
A
FIG. 9-19 Example 9-7. Cantilever beam
x
dB
with a uniform load acting on the righthand half of the beam
(c)
uB
Solution
In this example we will determine the deflection and angle of rotation by
treating an element of the uniform load as a concentrated load and then integrating (see Fig. 9-19b). The element of load has magnitude q dx and is located at
distance x from the support. The resulting differential deflection ddB and differential angle of rotation duB at the free end are found from the corresponding
formulas in Case 5 of Table G-1, Appendix G, by replacing P with q dx and a
with x; thus,
(qdx)(x 2)(3L x)
ddB
6EI
(q dx)(x 2)
duB
2EI
By integrating over the loaded region, we get
dB
q
ddB
6EI
uB
L
41qL 4
x 2(3L x) dx
384EI
L /2
q
duB
2EI
(9-54)
L
7qL 3
x 2 dx
48EI
L /2
(9-55)
Note: These same results can be obtained by using the formulas in Case 3
of Table G-1 and substituting a b L /2.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
622
CHAPTER 9 Deflections of Beams
Example 9-8
2a
—
3
A compound beam ABC has a roller support at A, an internal hinge at B, and a
fixed support at C (Fig. 9-20a). Segment AB has length a and segment BC has
length b. A concentrated load P acts at distance 2a/3 from support A and a uniform load of intensity q acts between points B and C.
Determine the deflection dB at the hinge and the angle of rotation uA at
support A (Fig. 9-20d). (Note: The beam has constant flexural rigidity EI.)
q
P
A
B
C
b
a
Solution
(a)
For purposes of analysis, we will consider the compound beam to consist
of two individual beams: (1) a simple beam AB of length a, and (2) a cantilever
beam BC of length b. The two beams are linked together by a pin connection
at B.
If we separate beam AB from the rest of the structure (Fig. 9-20b), we see
that there is a vertical force F at end B equal to 2P/3. This same force acts
downward at end B of the cantilever (Fig. 9-20c). Consequently, the cantilever
beam BC is subjected to two loads: a uniform load and a concentrated load. The
deflection at the end of this cantilever (which is the same as the deflection dB of
the hinge) is readily found from Cases 1 and 4 of Table G-1, Appendix G:
P
B
A
2P
F= —
3
(b)
F
q
B
4
qb
Fb3
dB
3EI
8EI
C
(c)
or, since F 2P/3,
y
dB
A
B
uA
4
C
x
qb
2Pb3
d B
9EI
8EI
(9-56)
The angle of rotation uA at support A (Fig. 9-20d) consists of two parts:
(1) an angle BAB produced by the downward displacement of the hinge, and
(2) an additional angle of rotation produced by the bending of beam AB (or
beam AB) as a simple beam. The angle BAB is
B′
(d)
FIG. 9-20 Example 9-8. Compound beam
with a hinge
4
dB
qb
2Pb3
(uA)1
9aEI
a
8aEI
The angle of rotation at the end of a simple beam with a concentrated load is
obtained from Case 5 of Table G-2. The formula given there is
Pab(L b)
6LEI
in which L is the length of the simple beam, a is the distance from the left-hand
support to the load, and b is the distance from the right-hand support to the load.
Thus, in the notation of our example (Fig. 9-20a), the angle of rotation is
2a a
a
P a
3 3
3
4Pa2
(uA)2
6aEI
81EI
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
623
SECTION 9.5 Method of Superposition
Combining the two angles, we obtain the total angle of rotation at support A:
4
qb
2Pb3
4Pa2
uA (uA)1 (uA) 2
8aEI
9aEI
81EI
(9-57)
This example illustrates how the method of superposition can be adapted to
handle a seemingly complex situation in a relatively simple manner.
Example 9-9
A simple beam AB of span length L has an overhang BC of length a (Fig.
9-21a). The beam supports a uniform load of intensity q throughout its length.
Obtain a formula for the deflection dC at the end of the overhang (Fig.
9-21c). (Note: The beam has constant flexural rigidity EI.)
q
A
C
B
a
L
(a)
P = qa
q
qa2
MB = —
2
A
B
L
(b)
y
A
Point of inflection
D
B
uB
C
x
d1
dC
d2
FIG. 9-21 Example 9-9. Simple beam
with an overhang
(c)
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
624
CHAPTER 9 Deflections of Beams
Solution
We can find the deflection of point C by imagining the overhang BC
(Fig. 9-21a) to be a cantilever beam subjected to two actions. The first action is
the rotation of the support of the cantilever through an angle uB, which is the
angle of rotation of beam ABC at support B (Fig. 9-21c). (We assume that a
clockwise angle uB is positive.) This angle of rotation causes a rigid-body rotation of the overhang BC, resulting in a downward displacement d1 of point C.
The second action is the bending of BC as a cantilever beam supporting a
uniform load. This bending produces an additional downward displacement d 2
(Fig. 9-21c). The superposition of these two displacements gives the total displacement dC at point C.
Deflection d1. Let us begin by finding the deflection d1 caused by the
angle of rotation uB at point B. To find this angle, we observe that part AB of the
beam is in the same condition as a simple beam (Fig. 9-21b) subjected to
the following loads: (1) a uniform load of intensity q, (2) a couple MB (equal to
qa2/2), and (3) a vertical load P (equal to qa). Only the loads q and MB produce
angles of rotation at end B of this simple beam. These angles are found from
Cases 1 and 7 of Table G-2, Appendix G. Thus, the angle uB is
qL3
MB L
qL3
qa2L
qL(4a2 L2)
uB
24EI
3EI
24EI
6EI
24EI
(9-58)
in which a clockwise angle is positive, as shown in Fig. 9-21c.
The downward deflection d1 of point C, due solely to the angle of rotation
uB, is equal to the length of the overhang times the angle (Fig. 9-21c):
qaL(4a2 L2)
d1 auB
24EI
(e)
q
P = qa
A
q
C
B
qa2
MB = —
2
A
B
a
L
L
(a)
y
A
(b)
Point of inflection
D
B
uB
C
x
d1
dC
d2
(c)
FIG. 9-21 (Repeated)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.5 Method of Superposition
625
Deflection d 2. Bending of the overhang BC produces an additional downward deflection d 2 at point C. This deflection is equal to the deflection of a
cantilever beam of length a subjected to a uniform load of intensity q (see
Case 1 of Table G-1):
qa4
d 2
8EI
(f)
Deflection dC. The total downward deflection of point C is the algebraic
sum of d1 and d 2:
qaL(4a2 L2)
qa 4
qa
d C d1 d 2 [L(4a 2 L2) 3a 3]
24EI
8EI
24EI
or
qa
d C (a L)(3a 2 aL L2)
24EI
(9-59)
From the preceding equation we see that the deflection dC may be upward
or downward, depending upon the relative magnitudes of the lengths L and a. If
a is relatively large, the last term in the equation (the three-term expression in
parentheses) is positive and the deflection dC is downward. If a is relatively
small, the last term is negative and the deflection is upward. The deflection is
zero when the last term is equal to zero:
3a 2 aL L2 0
or
L ( 13 1)
a 0.4343L
6
(g)
From this result, we see that if a is greater than 0.4343L, the deflection of point
C is downward; if a is less than 0.4343L, the deflection is upward.
Deflection curve. The shape of the deflection curve for the beam in this
example is shown in Fig. 9-21c for the case where a is large enough (a
0.4343L) to produce a downward deflection at C and small enough (a L) to
ensure that the reaction at A is upward. Under these conditions the beam has a
positive bending moment between support A and a point such as D. The deflection curve in region AD is concave upward (positive curvature). From D to C,
the bending moment is negative, and therefore the deflection curve is concave
downward (negative curvature).
Point of inflection. At point D the curvature of the deflection curve is zero
because the bending moment is zero. A point such as D where the curvature and
bending moment change signs is called a point of inflection (or point of contraflexure). The bending moment M and the second derivative d 2v/dx 2 always
vanish at an inflection point.
However, a point where M and d 2v/dx 2 equal zero is not necessarily an
inflection point because it is possible for those quantities to be zero without
changing signs at that point; for example, they could have maximum or minimum values.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
626
CHAPTER 9 Deflections of Beams
9.6 MOMENT-AREA METHOD
In this section we will describe another method for finding deflections
and angles of rotation of beams. Because the method is based upon two
theorems related to the area of the bending-moment diagram, it is called
the moment-area method.
The assumptions used in deriving the two theorems are the same as
those used in deriving the differential equations of the deflection curve.
Therefore, the moment-area method is valid only for linearly elastic
beams with small slopes.
First Moment-Area Theorem
To derive the first theorem, consider a segment AB of the deflection
curve of a beam in a region where the curvature is positive (Fig. 9-22).
Of course, the deflections and slopes shown in the figure are highly
exaggerated for clarity. At point A the tangent AA to the deflection
curve is at an angle uA to the x axis, and at point B the tangent BB is at
an angle uB. These two tangents meet at point C.
The angle between the tangents, denoted uB/A, is equal to the difference between uB and uA:
(9-60)
uB/A uB uA
Thus, the angle uB/A may be described as the angle to the tangent at B
measured relative to, or with respect to, the tangent at A. Note that the
angles uA and uB, which are the angles of rotation of the beam axis at
B′
y
uB
B
du
A
p2
p1
r
m2
m1 ds C
uB/A
du
A′
uA
x
O
M
—
EI
x
O
FIG. 9-22 Derivation of the first moment-
area theorem
x
dx
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.6 Moment-Area Method
627
points A and B, respectively, are also equal to the slopes at those points,
because in reality the slopes and angles are very small quantities.
Next, consider two points m1 and m2 on the deflected axis of the
beam (Fig. 9-22). These points are a small distance ds apart. The tangents to the deflection curve at these points are shown in the figure as
lines ml pl and m2 p 2. The normals to these tangents intersect at the center of curvature (not shown in the figure).
The angle du between the normals (Fig. 9-22) is given by the
following equation:
ds
(a)
du
r
in which r is the radius of curvature and du is measured in radians (see
Eq. 9-1). Because the normals and the tangents (m1 p1 and m2 p2) are
perpendicular, it follows that the angle between the tangents is also
equal to du.
For a beam with small angles of rotation, we can replace ds with dx,
as explained in Section 9.2. Thus,
dx
du
r
Also, from Eq. (9-6) we know that
1
M
r
EI
(b)
(c)
and therefore
M dx
(9-61)
du
EI
in which M is the bending moment and EI is the flexural rigidity of the
beam.
The quantity M dx/EI has a simple geometric interpretation. To see
this, refer to Fig. 9-22 where we have drawn the M/EI diagram directly
below the beam. At any point along the x axis, the height of this diagram
is equal to the bending moment M at that point divided by the flexural
rigidity EI at that point. Thus, the M/EI diagram has the same shape
as the bending-moment diagram whenever EI is constant. The term
M dx/EI is the area of the shaded strip of width dx within the M/EI diagram. (Note that since the curvature of the deflection curve in Fig. 9-22
is positive, the bending moment M and the area of the M/EI diagram are
also positive.)
Let us now integrate du (Eq. 9-61) between points A and B of the
deflection curve:
B
B
du
A
A
M dx
EI
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(d)
628
CHAPTER 9 Deflections of Beams
When evaluated, the integral on the left-hand side becomes uB uA,
which is equal to the angle uB/A between the tangents at B and A (Eq.
9-60).
The integral on the right-hand side of Eq. (d) is equal to the area of
the M/EI diagram between points A and B. (Note that the area of the
M/EI diagram is an algebraic quantity and may be positive or negative,
depending upon whether the bending moment is positive or negative.)
Now we can write Eq. (d) as follows:
B
uB/A
A
M dx
EI
Area of the M/EI diagram between points A and B
(9-62)
This equation may be stated as a theorem:
First moment-area theorem: The angle uB/A between the tangents to the
deflection curve at two points A and B is equal to the area of the M/EI
diagram between those points.
The sign conventions used in deriving the preceding theorem are as
follows:
1. The angles uA and uB are positive when counterclockwise.
2. The angle uB/A between the tangents is positive when the angle uB is
algebraically larger than the angle uA. Also, note that point B must
be to the right of point A; that is, it must be further along the axis of
the beam as we move in the x direction.
3. The bending moment M is positive according to our usual sign convention; that is, M is positive when it produces compression in the
upper part of the beam.
4. The area of the M/EI diagram is given a positive or negative sign
according to whether the bending moment is positive or negative. If
part of the bending-moment diagram is positive and part is negative,
then the corresponding parts of the M/EI diagram are given those
same signs.
The preceding sign conventions for uA, uB, and uB/A are often ignored
in practice because (as explained later) the directions of the angles of
rotation are usually obvious from an inspection of the beam and its loading. When this is the case, we can simplify the calculations by ignoring
signs and using only absolute values when applying the first momentarea theorem.
Second Moment-Area Theorem
Now we turn to the second theorem, which is related primarily to deflections rather than to angles of rotation. Consider again the deflection
curve between points A and B (Fig. 9-23). We draw the tangent at point
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.6 Moment-Area Method
629
y
B
dt
A
m1
du
m2
B1
x
x1
O
x
M
—
EI
C
O
FIG. 9-23 Derivation of the second
moment-area theorem
tB/A
x
x
dx
A and note that its intersection with a vertical line through point B is at
point B1. The vertical distance between points B and B1 is denoted tB/A
in the figure. This distance is referred to as the tangential deviation of
B with respect to A. More precisely, the distance tB/A is the vertical deviation of point B on the deflection curve from the tangent at point A. The
tangential deviation is positive when point B is above the tangent at A.
To determine the tangential deviation, we again select two points m1
and m 2 a small distance apart on the deflection curve (Fig. 9-23). The
angle between the tangents at these two points is du, and the segment on
line BB1 between these tangents is dt. Since the angles between the tangents and the x axis are actually very small, we see that the vertical
distance dt is equal to x1 du, where x1 is the horizontal distance from
point B to the small element m1 m 2. Since du M dx/EI (Eq. 9-61), we
obtain
M dx
dt x1 du x1
EI
(e)
The distance dt represents the contribution made by the bending of element m1m2 to the tangential deviation tB/A. The expression x1M dx/EI
may be interpreted geometrically as the first moment of the area of the
shaded strip of width dx within the M/EI diagram. This first moment is
evaluated with respect to a vertical line through point B.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
630
CHAPTER 9 Deflections of Beams
Integrating Eq. (e) between points A and B, we get
B
B
dt
A
A
M dx
x1
EI
(f)
The integral on the left-hand side is equal to tB/A, that is, it is equal to the
deviation of point B from the tangent at A. The integral on the right-hand
side represents the first moment with respect to point B of the area of the
M/EI diagram between A and B. Therefore, we can write Eq. (f) as follows:
B
tB/A
A
M dx
x1
EI
First moment of the area of the M/EI diagram
between points A and B, evaluated with respect to B
(9-63)
This equation represents the second theorem:
Second moment-area theorem: The tangential deviation tB/A of point B
from the tangent at point A is equal to the first moment of the area of the
M/EI diagram between A and B, evaluated with respect to B.
If the bending moment is positive, then the first moment of the M/EI
diagram is also positive, provided point B is to the right of point A.
Under these conditions the tangential deviation tB/A is positive and point
B is above the tangent at A (as shown in Fig. 9-23). If, as we move from
A to B in the x direction, the area of the M/EI diagram is negative, then
the first moment is also negative and the tangential deviation is negative, which means that point B is below the tangent at A.
The first moment of the area of the M/EI diagram can be obtained
by taking the product of the area of the diagram and the distance x from
point B to the centroid C of the area (Fig. 9-23). This procedure is usually more convenient than integrating, because the M/EI diagram usually
consists of familiar geometric figures such as rectangles, triangles, and
parabolic segments. The areas and centroidal distances of such figures
are tabulated in Appendix D.
As a method of analysis, the moment-area method is feasible only
for relatively simple kinds of beams. Therefore, it is usually obvious
whether the beam deflects upward or downward and whether an angle of
rotation is clockwise or counterclockwise. Consequently, it is seldom
necessary to follow the formal (and somewhat awkward) sign conventions described previously for the tangential deviation. Instead, we can
determine the directions by inspection and use only absolute values
when applying the moment-area theorems.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.6 Moment-Area Method
631
Example 9-10
P
A
Determine the angle of rotation uB and deflection dB at the free end B of a cantilever beam AB supporting a concentrated load P (Fig. 9-24). (Note: The beam
has length L and constant flexural rigidity EI.)
B
Solution
L
y
B
A
x
dB
uB
x
1
PL
PL2
A1 (L)
2
EI
2EI
O
C
Note that we are using only the absolute value of the area.
The relative angle of rotation between points A and B (from the first
theorem) is
PL
—
EI
FIG. 9-24 Example 9-10. Cantilever beam
with a concentrated load
By inspection of the beam and its loading, we know that the angle of rotation uB is clockwise and the deflection dB is downward (Fig. 9-24). Therefore,
we can use absolute values when applying the moment-area theorems.
M/EI diagram. The bending-moment diagram is triangular in shape with
the moment at the support equal to –PL. Since the flexural rigidity EI is constant, the M/EI diagram has the same shape as the bending-moment diagram, as
shown in the last part of Fig. 9-24.
Angle of rotation. From the first moment-area theorem, we know that the
angle uB/A between the tangents at points B and A is equal to the area of the
M/EI diagram between those points. This area, which we will denote as A1, is
determined as follows:
PL2
uB/A uB uA A1
2EI
Since the tangent to the deflection curve at support A is horizontal (uA 0), we
obtain
PL2
u B
2EI
(9-64)
This result agrees with the formula for uB given in Case 4 of Table G-1,
Appendix G.
Deflection. The deflection dB at the free end can be obtained from the second moment-area theorem. In this case, the tangential deviation tB/A of point B
from the tangent at A is equal to the deflection dB itself (see Fig. 9-24). The first
moment of the area of the M/EI diagram, evaluated with respect to point B, is
PL2 2L
PL3
Q1 A1 x
2EI 3
3EI
Note again that we are disregarding signs and using only absolute values.
From the second moment-area theorem, we know that the deflection dB is
equal to the first moment Q1. Therefore,
PL3
dB
3EI
This result also appears in Case 4 of Table G-1.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(9-65)
632
CHAPTER 9 Deflections of Beams
Example 9-11
q
A
C
Find the angle of rotation uB and deflection dB at the free end B of a cantilever
beam ACB supporting a uniform load of intensity q acting over the right-hand
half of the beam (Fig. 9-25). (Note: The beam has length L and constant flexural
rigidity EI.)
B
L
—
2
L
—
2
Solution
y
A
B
dB
x
uB
x2
2
qL3
1 L qL
A1
3 2 8EI
48EI
x1
O
A2
A3
The deflection and angle of rotation at end B of the beam have the directions shown in Fig. 9-25. Since we know these directions in advance, we can
write the moment-area expressions using only absolute values.
M/EI diagram. The bending-moment diagram consists of a parabolic curve
in the region of the uniform load and a straight line in the left-hand half of the
beam. Since EI is constant, the M/EI diagram has the same shape (see the last
part of Fig. 9-25). The values of M/EI at points A and C are 3qL2/8EI and
qL2/8EI, respectively.
Angle of rotation. For the purpose of evaluating the area of the M/EI diagram, it is convenient to divide the diagram into three parts: (1) a parabolic
spandrel of area A1, (2) a rectangle of area A2, and (3) a triangle of area A3.
These areas are
A1
qL2
—
8EI
x3
2
qL3
L qL
A2
2 8EI
16EI
2
qL 2
qL3
1 L 3qL
A3
2 2 8EI
8EI
16EI
3qL2
—
8EI
FIG. 9-25 Example 9-11. Cantilever beam
supporting a uniform load on the righthand half of the beam
According to the first moment-area theorem, the angle between the tangents at
points A and B is equal to the area of the M/EI diagram between those points.
Since the angle at A is zero, it follows that the angle of rotation uB is equal to
the area of the diagram; thus,
7qL 3
u B A1 A2 A3
48EI
(9-66)
Deflection. The deflection dB is the tangential deviation of point B with
respect to a tangent at point A (Fig. 9-25). Therefore, from the second momentarea theorem, dB is equal to the first moment of the M/EI diagram, evaluated
with respect to point B:
(g)
dB A1x 1 A2 x 2 A3 x 3
in which x 1, x 2, and x 3, are the distances from point B to the centroids of the
respective areas. These distances are
3 L
3L
x 1
4 2
8
L
L
3L
x2
2
4
4
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
L
2 L
5L
x 3
2
3 2
6
633
SECTION 9.6 Moment-Area Method
Substituting into Eq. (g), we find
qL3 3L
qL3 3L
qL3 5L
41qL 4
dB
48EI 8
16EI 4
16EI 6
384EI
(9-67)
This example illustrates how the area and first moment of a complex M/EI
diagram can be determined by dividing the area into parts having known properties. The results of this analysis (Eqs. 9-66 and 9-67) can be verified by using
the formulas of Case 3, Table G-1, Appendix G, and substituting a b L /2.
Example 9-12
A simple beam ADB supports a concentrated load P acting at the position
shown in Fig. 9-26. Determine the angle of rotation uA at support A and the
deflection dD under the load P. (Note: The beam has length L and constant
flexural rigidity EI.)
Solution
The deflection curve, showing the angle of rotation uA and the deflection
dD, is sketched in the second part of Fig. 9-26. Because we can determine the
directions of uA and dD by inspection, we can write the moment-area expressions using only absolute values.
M/EI diagram. The bending-moment diagram is triangular, with the maximum moment (equal to Pab/L) occurring under the load. Since EI is constant,
the M/EI diagram has the same shape as the moment diagram (see the third part
of Fig. 9-26).
Angle of rotation at support A. To find this angle, we construct the tangent
AB1 at support A. We then note that the distance BB1 is the tangential deviation
tB/A of point B from the tangent at A. We can calculate this distance by evaluating the first moment of the area of the M/EI diagram with respect to point B and
then applying the second moment-area theorem.
The area of the entire M/EI diagram is
1
Pab
Pab
A1 (L)
2
LE I
2EI
The centroid C1 of this area is at distance x1 from point B (see Fig. 9-26). This
distance, obtained from Case 3 of Appendix D, is
Lb
x1
3
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
634
CHAPTER 9 Deflections of Beams
P
D
A
B
a
b
L
y
A
D
dD
B
uA
D2
tD/A
tB/A
D1
A1
A1
Pab
—
LEI
B1
D1
Pab
—
LEI
B1
C1
O
Lb
x1 = —
3
P b
A2
O
FIG. 9-26 Example 9-12. Simple beam
Pab
—
LEI
C2
a
x2 = —
3
with a concentrated load
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
x
SECTION 9.6 Moment-Area Method
635
Therefore, the tangential deviation is
Pab L b
Pab
tB/A A1 x1 (L b)
2EI
3
6EI
The angle uA is equal to the tangential deviation divided by the length of the
beam:
tB /A
Pab
uA (L b)
L
6LEI
(9-68)
Thus, the angle of rotation at support A has been found.
Deflection under the load. As shown in the second part of Fig. 9-26, the
deflection dD under the load P is equal to the distance DD1 minus the distance
D2D1. The distance DD1 is equal to the angle of rotation uA times the distance a;
thus,
Pa2b
DD1 auA (L b)
6LEI
(h)
The distance D2D1 is the tangential deviation tD/A at point D; that is, it is
the deviation of point D from the tangent at A. This distance can be found from
the second moment-area theorem by taking the first moment of the area of the
M/EI diagram between points A and D with respect to D (see the last part of
Fig. 9-26). The area of this part of the M/EI diagram is
1
Pab
Pa 2 b
A 2 (a)
2
LE I
2LEI
and its centroidal distance from point D is
a
x 2
3
Thus, the first moment of this area with respect to point D is
Pa 2 b a
Pa 3b
tD/A A2 x 2
2LEI 3
6LEI
(i)
The deflection at point D is
dD DD1 D2D1 DD1 tD/A
Upon substituting from Eqs. (h) and (i), we find
Pa2b
Pa3b
Pa2b2
dD (L b)
6LEI
6LEI
3LEI
(9-69)
The preceding formulas for uA and dD (Eqs. 9-68 and 9-69) can be verified by
using the formulas of Case 5, Table G-2, Appendix G.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
636
CHAPTER 9 Deflections of Beams
9.7 NONPRISMATIC BEAMS
The methods presented in the preceding sections for finding deflections
of prismatic beams can also be used to find deflections of beams having
varying moments of inertia. Two examples of nonprismatic beams are
shown in Fig. 9-27. The first beam has two different moments of inertia,
and the second is a tapered beam having continuously varying moment
of inertia. In both cases the objective is to save material by increasing
the moment of inertia in regions where the bending moment is largest.
FIG. 9-27 Beams with varying moments
of inertia (see also Fig. 5-23)
Although no new concepts are involved, the analysis of a nonprismatic beam is more complex than the analysis of a beam with constant
moment of inertia. Some of the procedures that can be used are illustrated in the examples that follow (Examples 9-13 and 9-14).
In the first example (a simple beam having two different moments
of inertia), the deflections are found by solving the differential equation
of the deflection curve. In the second example (a cantilever beam having
two different moments of inertia), the method of superposition is used.
These two examples, as well as the problems for this section,
involve relatively simple and idealized beams. When more complex
beams (such as tapered beams) are encountered, numerical methods of
analysis are usually required. (Computer programs for the numerical
calculation of beam deflections are readily available.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
637
SECTION 9.7 Nonprismatic Beams
Example 9-13
P
C
B
A
L
—
4
L
—
4
D
L
—
4
E
L
—
4
Solution
L
Differential equations of the deflection curve. In this example we will
determine the slopes and deflections of the beam by integrating the bendingmoment equation, that is, the second-order differential equation of the
deflection curve (Eq. 9-12a). Since the reaction at each support is P/2, the bending moment throughout the left-hand half of the beam is
(a)
P
C
B
A
I
A beam ABCDE on simple supports is constructed from a wide-flange beam by
welding cover plates over the middle half of the beam (Fig. 9-28a). The effect
of the cover plates is to double the moment of inertia (Fig. 9-28b). A concentrated load P acts at the midpoint C of the beam.
Determine the equations of the deflection curve, the angle of rotation uA at
the left-hand support, and the deflection dC at the midpoint (Fig. 9-28c).
D
E
I
2I
Px
M
2
Px
EIv
2
Px
E(2I)v
2
y
A
B
D
L
2
x
(a)
Therefore, the differential equations for the left-hand half of the beam are
(b)
C
0
E x
dC
uA
(c)
FIG. 9-28 Example 9-13. Simple beam
with two different moments of inertia
0
L4
L
4
x
x
(b)
L
2
(c)
Each of these equations can be integrated twice to obtain expressions for
the slopes and deflections in their respective regions. These integrations produce four constants of integration that can be found from the following four
conditions:
1.
2.
3.
4.
Boundary condition: At support A (x 0), the deflection is zero (v 0).
Symmetry condition: At point C (x L /2), the slope is zero (v 0).
Continuity condition: At point B (x L /4), the slope obtained from part
AB of the beam is equal to the slope obtained from part BC of the beam.
Continuity condition: At point B (x L /4), the deflection obtained from
part AB of the beam is equal to the deflection obtained from part BC of the
beam.
Slopes of the beam. Integrating each of the differential equations (Eqs. b
and c), we obtain the following equations for the slopes in the left-hand half of
the beam:
Px 2
v C1
4EI
0
x
L
4
(d)
Px 2
v C2
8EI
L4
x
L
2
(e)
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
638
CHAPTER 9 Deflections of Beams
Applying the symmetry condition (2) to Eq. (e), we obtain the constant C2:
PL 2
C2
32EI
Therefore, the slope of the beam between points B and C (from Eq. e) is
P
v (L2 4x2)
32EI
L4
L
2
x
(9-70)
From this equation we can find the slope of the deflection curve at point B
where the moment of inertia changes from I to 2I:
L
3PL2
v
4
128EI
(f )
Because the deflection curve is continuous at point B, we can use the continuity
condition (3) and equate the slope at point B as obtained from Eq. (d) to the
slope at the same point given by Eq. (f). In this manner we find the constant C1:
P L
4EI 4
2
3PL 2
5P L 2
C1 or C1
128EI
128EI
Therefore, the slope between points A and B (see Eq. d) is
P
v (5L 2 32x 2)
128EI
0
L
4
x
(9-71)
At support A, where x 0, the angle of rotation (Fig. 9-28c) is
5PL2
uA v(0)
128EI
(9-72)
Deflections of the beam. Integrating the equations for the slopes (Eqs. 9-71
and 9-70), we get
0
32 x 3
P
v 5L 2x C3
3
128EI
4 x3
P
v L 2x C4
3
32EI
L4
x
L
4
(g)
(h)
L
2
x
Applying the boundary condition at the support (condition 1) to Eq. (g), we get
C3 0. Therefore, the deflection between points A and B (from Eq. g) is
Px
v (15L2 32x 2)
384EI
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
0
x
L
4
(9-73)
SECTION 9.7 Nonprismatic Beams
639
From this equation we can find the deflection at point B:
L
13PL3
v
4
1536EI
(i)
Since the deflection curve is continuous at point B, we can use the continuity
condition (4) and equate the deflection at point B as obtained from Eq. (h) to the
deflection given by Eq. (i):
P
L
4 L
L 2
32EI
4
3 4
3
13PL3
C4
1536EI
from which
PL3
C4
768EI
Therefore, the deflection between points B and C (from Eq. h) is
P
v (L 3 24L 2 x 32x 3)
768EI
L4
x
L
2
(9-74)
Thus, we have obtained the equations of the deflection curve for the left-hand
half of the beam. (The deflections in the right-hand half of the beam can be
obtained from symmetry.)
Finally, we obtain the deflection at the midpoint C by substituting x L/2
into Eq. (9-74):
L
3PL 3
dC v
2
256EI
(9-75)
All required quantities have now been found, and the analysis of the nonprismatic beam is completed.
Notes: Using the differential equation for finding deflections is practical
only if the number of equations to be solved is limited to one or two and only if
the integrations are easily performed, as in this example. In the case of a tapered
beam (Fig. 9-27), it may be difficult to solve the differential equation analytically because the moment of inertia is a continuous function of x. In such a case,
the differential equation has variable coefficients instead of constant coefficients, and numerical methods of solution are needed.
When a beam has abrupt changes in cross-sectional dimensions, as in this
example, there are stress concentrations at the points where changes occur.
However, because the stress concentrations affect only a small region of the
beam, they have no noticeable effect on the deflections.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
640
CHAPTER 9 Deflections of Beams
Example 9-14
P
B
C
A
A cantilever beam ACB having length L and two different moments of inertia I
and 2I supports a concentrated load P at the free end A (Figs. 9-29a and b).
Determine the deflection dA at the free end.
Solution
L
—
2
L
—
2
(a)
P
B
C
A
I
2I
(b)
P
P(L/2)3
PL 3
d1
3EI
24EI
C
A
d1
(c)
PL
—
2
P
B
C
dC
( j)
Deflection due to bending of part CB of the beam. Part CB of the beam also
behaves as a cantilever beam (Fig. 9-29d) and contributes to the deflection of
point A. The end of this cantilever is subjected to a concentrated load P and a
moment PL/2. Therefore, the deflection dC and angle of rotation uC at the free
end (Fig. 9-29d) are as follows (see Cases 4 and 6 of Table G-1):
P(L / 2)3
(PL / 2)(L /2)2
5PL3
dC
96EI
3(2EI)
2(2EI)
uC
2
P(L /2)
(PL / 2)(L /2)
3PL2
uC
16EI
2(2EI)
2EI
(d)
A
C
This deflection and angle of rotation make an additional contribution d 2 to the
deflection at end A (Fig. 9-29e). We again visualize part AC as a cantilever
beam, but now its support (at point C) moves downward by the amount dC and
rotates counterclockwise through the angle uC (Fig. 9-29e). These rigidbody displacements produce a downward displacement at end A equal to the
following:
dC
d2
uC
(e)
P
A
In this example we will use the method of superposition to determine the
deflection dA at the end of the beam. We begin by recognizing that the deflection consists of two parts: the deflection due to bending of part AC of the beam
and the deflection due to bending of part CB. We can determine these deflections separately and then superpose them to obtain the total deflection.
Deflection due to bending of part AC of the beam. Imagine that the beam is
held rigidly at point C, so that the beam neither deflects nor rotates at that point
(Fig. 9-29c). We can easily calculate the deflection d1 of point A in this beam.
Since the beam has length L/2 and moment of inertia I, its deflection (see Case
4 of Table G-1, Appendix G) is
C
B
L
5PL3
3PL2 L
7PL 3
d 2 d C uC
2
96EI
16EI 2
48EI
(k)
Total deflection. The total deflection dA at the free end A of the original
cantilever beam (Fig. 9-29f) is equal to the sum of the deflections d1 and d 2:
dA
(f)
FIG. 9-29 Example 9-14. Cantilever beam
with two different moments of inertia
7PL3
3PL3
PL3
dA d1 d 2
48EI
16EI
24EI
(9-76)
This example illustrates one of the many ways that the principle of superposition may be used to find beam deflections.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.8 Strain Energy of Bending
641
9.8 STRAIN ENERGY OF BENDING
M
B
A
M
L
(a)
u
r
r
The general concepts pertaining to strain energy were explained previously in our discussions of bars subjected to axial loads and shafts
subjected to torsion (Sections 2.7 and 3.9, respectively). In this section,
we will apply the same concepts to beams. Since we will be using the
equations for curvature and deflection derived earlier in this chapter, our
discussion of strain energy applies only to beams that behave in a linearly elastic manner. This requirement means that the material must
follow Hooke’s law and the deflections and rotations must be small.
Let us begin with a simple beam AB in pure bending under the
action of two couples, each having a moment M (Fig. 9-30a). The
deflection curve (Fig. 9-30b) is a nearly flat circular arc of constant
curvature k M/EI (see Eq. 9-6). The angle u subtended by this arc
equals L/r, where L is the length of the beam and r is the radius of
curvature. Therefore,
L
ML
u k L
r
EI
A
B
L
(b)
(9-77)
This linear relationship between the moments M and the angle u is
shown graphically by line OA in Fig. 9-31. As the bending couples gradually increase in magnitude from zero to their maximum values, they
perform work W represented by the shaded area below line OA. This
work, equal to the strain energy U stored in the beam, is
FIG. 9-30 Beam in pure bending by
Mu
W U
2
couples of moment M
(9-78)
This equation is analogous to Eq. (2-35) for the strain energy of an
axially loaded bar.
M
A
Mu
W=U= —
2
FIG. 9-31 Diagram showing linear
relationship between bending moments
M and the angle u
O
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
u
642
CHAPTER 9 Deflections of Beams
By combining Eqs. (9-77) and (9-78), we can express the strain
energy stored in a beam in pure bending in either of the following
forms:
M 2L
U
2EI
EIu 2
U
2L
(9-79a,b)
The first of these equations expresses the strain energy in terms of the
applied moments M, and the second equation expresses it in terms of the
angle u. The equations are similar in form to those for strain energy in
an axially loaded bar (Eqs. 2-37a and b).
If the bending moment in a beam varies along its length (nonuniform bending), then we may obtain the strain energy by applying
Eqs. (9-79a) and (9-79b) to an element of the beam (Fig. 9-32) and integrating along the length of the beam. The length of the element itself is
dx and the angle du between its side faces can be obtained from Eqs. (9-4)
and (9-5), as follows:
d 2v
dx
du k dx
dx2
(a)
Therefore, the strain energy dU of the element is given by either of the
following equations (see Eqs. 9-79a and b):
M 2dx
dU
2EI
2
EI d 2 v
EI(du)2
EI d 2 v 2
dU
dx
dx
2dx dx 2
2 dx 2
2dx
(b,c)
By integrating the preceding equations throughout the length of a beam,
we can express the strain energy stored in a beam in either of the following forms:
U
M 2dx
2EI
U
EI d 2v 2
dx
2 dx 2
(9-80a,b)
Note that M is the bending moment in the beam and may vary as a function of x. We use the first equation when the bending moment is known,
du
M
M
dx
FIG. 9-32 Side view of an element of a
beam subjected to bending moments M
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.8 Strain Energy of Bending
643
and the second equation when the equation of the deflection curve is
known. (Examples 9-15 and 9-16 illustrate the use of these equations.)
In the derivation of Eqs. (9-80a) and (9-80b), we considered only
the effects of the bending moments. If shear forces are also present,
additional strain energy will be stored in the beam. However, the strain
energy of shear is relatively small (in comparison with the strain energy
of bending) for beams in which the lengths are much greater than the
depths (say, L /d 8). Therefore, in most beams the strain energy of
shear may safely be disregarded.
Deflections Caused by a Single Load
If a beam supports a single load, either a concentrated load P or a couple
M0, the corresponding deflection d or angle of rotation u, respectively,
can be determined from the strain energy of the beam.
In the case of a beam supporting a concentrated load, the corresponding deflection d is the deflection of the beam axis at the point
where the load is applied. The deflection must be measured along the
line of action of the load and is positive in the direction of the load.
In the case of a beam supporting a couple as a load, the corresponding angle of rotation u is the angle of rotation of the beam axis at the
point where the couple is applied.
Since the strain energy of a beam is equal to the work done by the
load, and since d and u correspond to P and M0, respectively, we obtain
the following equations:
Pd
U W
2
M0u
U W
2
(9-81a,b)
The first equation applies to a beam loaded only by a force P, and the
second equation applies to a beam loaded only by a couple M0. It
follows from Eqs. (9-81a) and (9-81b) that
2U
d
P
2U
M0
u
(9-82a,b)
As explained in Section 2.7, this method for finding deflections and
angles of rotation is extremely limited in its application because only
one deflection (or one angle) can be found. Furthermore, the only
deflection (or angle) that can be found is the one corresponding to
the load (or couple). However, the method occasionally is useful and is
illustrated later in Example 9-16.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
644
CHAPTER 9 Deflections of Beams
Example 9-15
y
q
B
A
x
A simple beam AB of length L supports a uniform load of intensity q (Fig.
9-33). (a) Evaluate the strain energy of the beam from the bending moment in
the beam. (b) Evaluate the strain energy of the beam from the equation of the
deflection curve. (Note: The beam has constant flexural rigidity EI.)
Solution
L
FIG. 9-33 Example 9-15. Strain energy of
(a) Strain energy from the bending moment. The reaction of the beam at
support A is qL/2, and therefore the expression for the bending moment in the
beam is
a beam
qLx
qx2
q
M (Lx x 2)
2
2
2
(d)
The strain energy of the beam (from Eq. 9-80a) is
L
U
0
M 2dx
1
2EI
2EI
L
0
2
q
q2
(Lx x 2) dx
2
8EI
L
(L2 x 2 2L x 3 x 4)dx
0
(e)
from which we get
q2L5
U
240EI
(9-83)
Note that the load q appears to the second power, which is consistent with the
fact that strain energy is always positive. Furthermore, Eq. (9-83) shows that
strain energy is not a linear function of the loads, even though the beam itself
behaves in a linearly elastic manner.
(b) Strain energy from the deflection curve. The equation of the deflection
curve for a simple beam with a uniform load is given in Case 1 of Table G-2,
Appendix G, as follows:
qx
v (L 3 2Lx 2 x 3)
24EI
(f )
Taking two derivatives of this equation, we get
q
d 2v
2
2 (Lx x )
dx
2EI
q
dv
(L 3 6Lx 2 4x 3)
dx
24EI
Substituting the latter expression into the equation for strain energy (Eq. 9-80b),
we obtain
L
U
0
EI d 2 v 2
EI
dx
2 dx 2
2
q2
8EI
L
0
2EI (Lx x ) dx
q
2
2
L
(L 2 x 2 2Lx 3 x 4)dx
(g)
0
Since the final integral in this equation is the same as the final integral in Eq. (e),
we get the same result as before (Eq. 9-83).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
645
SECTION 9.8 Strain Energy of Bending
Example 9-16
P
A
B
x
L
Solution
(a)
M0
A
A cantilever beam AB (Fig. 9-34) is subjected to three different loading conditions: (a) a concentrated load P at its free end, (b) a couple M0 at its free end,
and (c) both loads acting simultaneously.
For each loading condition, determine the strain energy of the beam. Also,
determine the vertical deflection dA at end A of the beam due to the load P acting alone (Fig. 9-34a), and determine the angle of rotation uA at end A due to the
moment M0 acting alone (Fig. 9-34b). (Note: The beam has constant flexural
rigidity EI.)
B
(a) Beam with concentrated load P (Fig. 9-34a). The bending moment in
the beam at distance x from the free end is M Px. Substituting this expression for M into Eq. (9-80a), we get the following expression for the strain
energy of the beam:
x
L
L
U
0
P
A
0
(Px)2dx
P 2L3
2EI
6EI
(9-84)
PdA
P 2L 3
W U or
2
6EI
B
x
from which
L
PL3
d A
3EI
(c)
FIG. 9-34 Example 9-16. Strain energy of
a beam
L
To obtain the vertical deflection dA under the load P, we equate the work done
by the load to the strain energy:
(b)
M0
M 2dx
2EI
The deflection dA is the only deflection we can find by this procedure, because
it is the only deflection that corresponds to the load P.
(b) Beam with moment M0 (Fig. 9-34b). In this case the bending moment is
constant and equal to M0. Therefore, the strain energy (from Eq. 9-80a) is
L
U
0
M 2dx
2EI
L
0
(M0)2 dx
M 20 L
2EI
2EI
(9-85)
The work W done by the couple M0 during loading of the beam is M0uA /2,
where u A is the angle of rotation at end A. Therefore,
M0uA
M 20 L
W U or
2
2EI
and
M0 L
u A
EI
The angle of rotation has the same sense as the moment (counterclockwise in
this example).
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
646
CHAPTER 9 Deflections of Beams
(c) Beam with both loads acting simultaneously (Fig. 9-34c). When both
loads act on the beam, the bending moment in the beam is
M Px M0
Therefore, the strain energy is
L
U
0
M 2dx
1
2EI
2EI
L
(Px M0 )2 dx
0
PM0 L2
M 20 L
P 2L 3
2EI
2EI
6EI
(9-86)
The first term in this result gives the strain energy due to P acting alone
(Eq. 9-84), and the last term gives the strain energy due to M0 alone (Eq. 9-85).
However, when both loads act simultaneously, an additional term appears in the
expression for strain energy.
Therefore, we conclude that the strain energy in a structure due to two or
more loads acting simultaneously cannot be obtained by adding the strain energies due to the loads acting separately. The reason is that strain energy is a
quadratic function of the loads, not a linear function. Therefore, the principle of
superposition does not apply to strain energy.
We also observe that we cannot calculate a deflection for a beam with two
or more loads by equating the work done by the loads to the strain energy. For
instance, if we equate work and energy for the beam of Fig. 9-34c, we get
PdA2
M0uA2
PM0 L2
M 20 L
P 2L 3
W U or
2
6EI
2
2EI
2EI
(h)
in which dA2 and uA2 represent the deflection and angle of rotation at end A of
the beam with two loads acting simultaneously (Fig. 9-34c). Although the work
done by the two loads is indeed equal to the strain energy, and Eq. (h) is quite
correct, we cannot solve for either dA2 or uA2 because there are two unknowns
and only one equation.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.9 Castigliano’s Theorem
★9.9
647
CASTIGLIANO’S THEOREM
P
B
A
L
Castigliano’s theorem provides a means for finding the deflections of a
structure from the strain energy of the structure. To illustrate what we
mean by that statement, consider a cantilever beam with a concentrated
load P acting at the free end (Fig. 9-35a). The strain energy of this beam
is obtained from Eq. (9-84) of Example 9-16:
P 2L3
U
6EI
(a)
(a)
Now take the derivative of this expression with respect to the load P:
B
A
dA
dU
d P 2L3
PL 3
dP
dP 6EI
3EI
(b)
We immediately recognize this result as the deflection dA at the free end
A of the beam (see Fig. 9-35b). Note especially that the deflection dA
corresponds to the load P itself. (Recall that a deflection corresponding
to a concentrated load is the deflection at the point where the concentrated load is applied. Furthermore, the deflection is in the direction of
the load.) Thus, Eq. (b) shows that the derivative of the strain energy
with respect to the load is equal to the deflection corresponding to the
load. Castigliano’s theorem is a generalized statement of this observation, and we will now derive it in more general terms.
(b)
FIG. 9-35 Beam supporting a single
load P
Derivation of Castigliano’s Theorem
P1
P2
Pi
Pn
(a)
d1
d2
di
dn
(b)
FIG. 9-36 Beam supporting n loads
Let us consider a beam subjected to any number of loads, say n loads P1,
P2, . . ., Pi, . . ., Pn (Fig. 9-36a). The deflections of the beam corresponding to the various loads are denoted d1, d2, . . ., di, . . ., dn, as shown in
Fig. 9-36b. As in our earlier discussions of deflections and strain energy,
we assume that the principle of superposition is applicable to the beam
and its loads.
Now we will determine the strain energy of this beam. When the
loads are applied to the beam, they gradually increase in magnitude
from zero to their maximum values. At the same time, each load moves
through its corresponding displacement and does work. The total work
W done by the loads is equal to the strain energy U stored in the beam:
WU
(c)
Note that W (and hence U) is a function of the loads P1, P2, . . ., Pn acting
on the beam.
Next, let us suppose that one of the loads, say the ith load, is
increased slightly by the amount dPi while the other loads are held constant. This increase in load will cause a small increase dU in the strain
energy of the beam. This increase in strain energy may be expressed as
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
648
CHAPTER 9 Deflections of Beams
the rate of change of U with respect to Pi times the small increase in Pi.
Thus, the increase in strain energy is
]U
dU dPi
]Pi
(d)
where ]U/]Pi is the rate of change of U with respect to Pi. (Since U is a
function of all the loads, the derivative with respect to any one of the
loads is a partial derivative.) The final strain energy of the beam is
]U
U dU U dPi
]Pi
(e)
in which U is the strain energy referred to in Eq. (c).
Because the principle of superposition holds for this beam, the total
strain energy is independent of the order in which the loads are applied.
That is, the final displacements of the beam (and the work done by the
loads in reaching those displacements) are the same regardless of the
order in which the loads are applied. In arriving at the strain energy
given by Eq. (e), we first applied the n loads P1, P2, . . ., Pn, and then we
applied the load dPi. However, we can reverse the order of application
and apply the load dPi first, followed by the loads P1, P2, . . ., Pn. The
final amount of strain energy is the same in either case.
When the load dPi is applied first, it produces strain energy equal to
one-half the product of the load dPi and its corresponding displacement
ddi. Thus, the amount of strain energy due to the load dPi is
dPi dd i
2
(f)
When the loads P1, P2, . . ., Pn are applied, they produce the same
displacements as before (d1, d2, . . ., dn) and do the same amount of
work as before (Eq. c). However, during the application of these loads,
the force dPi automatically moves through the displacement di. In so
doing, it produces additional work equal to the product of the force and
the distance through which it moves. (Note that the work does not have
a factor 1/2, because the force dPi acts at full value throughout this
displacement.) Thus, the additional work, equal to the additional strain
energy, is
dPi di
(g)
Therefore, the final strain energy for the second loading sequence is
dPi dd i
U dPi d i
2
(h)
Equating this expression for the final strain energy to the earlier expression (Eq. e), which was obtained for the first loading sequence, we get
dPi dd i
]U
U dPi d i U dPi
]Pi
2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(i)
SECTION 9.9 Castigliano’s Theorem
649
We can discard the first term because it contains the product of two differentials and is infinitesimally small compared to the other terms. We
then obtain the following relationship:
]U
d i
]Pi
P
M0
B
A
x
(9-87)
This equation is known as Castigliano’s theorem.*
Although we derived Castigliano’s theorem by using a beam as an
illustration, we could have used any other type of structure (for example,
a truss) and any other kinds of loads (for example, loads in the form of
couples). The important requirements are that the structure be linearly
elastic and that the principle of superposition be applicable. Also, note
that the strain energy must be expressed as a function of the loads (and
not as a function of the displacements), a condition which is implied in
the theorem itself, since the partial derivative is taken with respect to a
load. With these limitations in mind, we can state Castigliano’s theorem
in general terms as follows:
The partial derivative of the strain energy of a structure with respect
to any load is equal to the displacement corresponding to that load.
The strain energy of a linearly elastic structure is a quadratic function of the loads (for instance, see Eq. a), and therefore the partial
derivatives and the displacements (Eq. 9-87) are linear functions of the
loads (as expected).
When using the terms load and corresponding displacement in connection with Castigliano’s theorem, it is understood that these terms are
used in a generalized sense. The load Pi and corresponding displacement
di may be a force and a corresponding translation, or a couple and a corresponding rotation, or some other set of corresponding quantities.
L
(a)
Application of Castigliano’s Theorem
B
A
dA
uA
(b)
FIG. 9-37 Application of Castigliano’s
theorem to a beam
As an application of Castigliano’s theorem, let us consider a cantilever
beam AB carrying a concentrated load P and a couple of moment M0
acting at the free end (Fig. 9-37a). We wish to determine the vertical
deflection dA and angle of rotation uA at the end of the beam
(Fig. 9-37b). Note that dA is the deflection corresponding to the load P,
and uA is the angle of rotation corresponding to the moment M0.
*Castigliano’s theorem, one of the most famous theorems in structural analysis, was discovered by Carlos Alberto Pio Castigliano (1847–1884), an Italian engineer (Ref. 9-2).
The theorem quoted here (Eq. 9-87) is actually the second of two theorems presented by
Castigliano and is properly called Castigliano’s second theorem. The first theorem is the
reverse of the second theorem, in the sense that it gives the loads on a structure in terms
of the partial derivatives of the strain energy with respect to the displacements.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
650
CHAPTER 9 Deflections of Beams
P
M0
B
A
The first step in the analysis is to determine the strain energy of the
beam. For that purpose, we write the equation for the bending moment
as follows:
x
M Px M0
L
in which x is the distance from the free end (Fig. 9-37a). The strain
energy is found by substituting this expression for M into Eq. (9-80a):
(a)
B
A
dA
uA
FIG. 9-37 (Repeated)
L
U
0
(b)
( j)
1
M 2dx
2EI
2EI
L
(Px M0)2 dx
0
PM0 L2
M2L
P 2L 3
0
6EI
2EI
2EI
(k)
in which L is the length of the beam and EI is its flexural rigidity. Note
that the strain energy is a quadratic function of the loads P and M0.
To obtain the vertical deflection dA at the end of the beam, we use
Castigliano’s theorem (Eq. 9-87) and take the partial derivative of the
strain energy with respect to P:
M0 L 2
]U
PL3
dA
]P
3EI
2EI
(l)
This expression for the deflection can be verified by comparing it with
the formulas of Cases 4 and 6 of Table G-1, Appendix G.
In a similar manner, we can find the angle of rotation uA at the end
of the beam by taking the partial derivative with respect to M0:
M0 L
]U
PL2
uA
]M0
2EI
EI
(m)
This equation can also be verified by comparing with the formulas of
Cases 4 and 6 of Table G-1.
Use of a Fictitious Load
The only displacements that can be found from Castigliano’s theorem
are those that correspond to loads acting on the structure. If we wish to
calculate a displacement at a point on a structure where there is no load,
then a fictitious load corresponding to the desired displacement must be
applied to the structure. We can then determine the displacement by
evaluating the strain energy and taking the partial derivative with respect
to the fictitious load. The result is the displacement produced by the
actual loads and the fictitious load acting simultaneously. By setting the
fictitious load equal to zero, we obtain the displacement produced only
by the actual loads.
To illustrate this concept, suppose we wish to find the vertical deflection dC at the midpoint C of the cantilever beam shown in
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.9 Castigliano’s Theorem
P
A
C
B
M0
L
—
2
L
—
2
(a)
Fig. 9-38a. Since the deflection dC is downward (Fig. 9-38b), the load
corresponding to that deflection is a downward vertical force acting at
the same point. Therefore, we must supply a fictitious load Q acting at
point C in the downward direction (Fig. 9-39a). Then we can use
Castigliano’s theorem to determine the deflection (dC)0 at the midpoint
of this beam (Fig. 9-39b). From that deflection, we can obtain the
deflection dC in the beam of Fig. 9-38 by setting Q equal to zero.
We begin by finding the bending moments in the beam of
Fig. 9-39a:
M Px M0
B
A
651
dC
0
L
M Px M0 Q x
2
(b)
x
L
2
L2
(n)
x
L
(o)
Next, we determine the strain energy of the beam by applying Eq.
(9-80a) to each half of the beam. For the left-hand half of the beam
(from point A to point C), the strain energy is
FIG. 9-38 Beam supporting loads P
and M0
L /2
UAC
0
M 2dx
1
2EI
2EI
L /2
(Px M0)2 dx
0
M 20 L
PM0 L2
P 2L 3
8EI
4EI
48EI
(p)
For the right-hand half, the strain energy is
L
UCB
P
M0
Q
C
A
B
x
L
—
2
L
—
2
(a)
M 2dx
1
2E
I
2
EI
L /2
L
L /2
Px M
0
L
Q x
2
dx
2
3PM0 L2
5PQL3
M 20 L
M0QL2
Q 2L3
7P 2L3
(q)
48EI
8EI
48EI
4EI
8EI
48EI
which requires a very lengthy process of integration. Adding the strain
energies for the two parts of the beam, we obtain the strain energy for
the entire beam (Fig. 9-39a):
U UAC UCB
PM 0 L2
5PQL3
M 02 L
M0QL2
Q 2L3
P 2L3
6EI
2EI
48EI
2EI
8EI
48EI
B
A
(dC)0
The deflection at the midpoint of the beam shown in Fig. 9-39a can now
be obtained from Castigliano’s theorem:
M 0 L2
QL3
]U
5PL3
(dC) 0
]Q
48EI
8 EI
24EI
(b)
FIG. 9-39 Beam with a fictitious load Q
(r)
(s)
This equation gives the deflection at point C produced by all three loads
acting on the beam. To obtain the deflection produced by the loads
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
652
CHAPTER 9 Deflections of Beams
P and M0 only, we set the load Q equal to zero in the preceding equation. The result is the deflection at the midpoint C for the beam with two
loads (Fig. 9-38a):
M 0 L2
5PL3
dC
48EI
8EI
(t)
Thus, the deflection in the original beam has been obtained.
This method is sometimes called the dummy-load method, because
of the introduction of a fictitious, or dummy, load.
Differentiation Under the Integral Sign
As we saw in the preceding example, the use of Castigliano’s theorem
for determining beam deflections may lead to lengthy integrations, especially when more than two loads act on the beam. The reason is
clear—finding the strain energy requires the integration of the square of
the bending moment (Eq. 9-80a). For instance, if the bending moment
expression has three terms, its square may have as many as six terms,
each of which must be integrated.
After the integrations are completed and the strain energy has been
determined, we differentiate the strain energy to obtain the deflections.
However, we can bypass the step of finding the strain energy by differentiating before integrating. This procedure does not eliminate the
integrations, but it does make them much simpler.
To derive this method, we begin with the equation for the strain
energy (Eq. 9-80a) and apply Castigliano’s theorem (Eq. 9-87):
]U
]
di
]Pi
]Pi
M 2dx
2EI
(u)
Following the rules of calculus, we can differentiate the integral by
differentiating under the integral sign:
]
di
]Pi
M 2dx
2EI
EMI ]]MPdx
(9-88)
i
We will refer to this equation as the modified Castigliano’s theorem.
When using the modified theorem, we integrate the product of the
bending moment and its derivative. By contrast, when using the standard Castigliano’s theorem (see Eq. u), we integrate the square of the
bending moment. Since the derivative is a shorter expression than the
moment itself, this new procedure is much simpler. To show this, we
will now solve the preceding examples using the modified theorem
(Eq. 9-88).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
653
SECTION 9.9 Castigliano’s Theorem
Let us begin with the beam shown in Fig. 9-37 and recall that we
wish to find the deflection and angle of rotation at the free end. The
bending moment and its derivatives (see Eq. j) are
M Px M0
]M
]M
x
1
]P
]M0
From Eq. (9-88) we obtain the deflection dA and angle of rotation uA:
1
dA
EI
1
uA
EI
L
0
M0 L2
PL3
(Px M0)(x)dx
3EI
2EI
L
0
(v)
M0 L
PL2
(Px M0)(1)dx
2EI
EI
(w)
These equations agree with the earlier results (Eqs. l and m). However,
the calculations are shorter than those performed earlier, because we did
not have to integrate the square of the bending moment (see Eq. k).
The advantages of differentiating under the integral sign are even
more apparent when there are more than two loads acting on the structure, as in the example of Fig. 9-38. In that example, we wished to
determine the deflection dC at the midpoint C of the beam due to the
loads P and M0. To do so, we added a fictitious load Q at the midpoint
(Fig. 9-39). We then proceeded to find the deflection (dC)0 at the midpoint of the beam when all three loads (P, M0, and Q) were acting.
Finally, we set Q 0 to obtain the deflection dC due to P and M0 alone.
The solution was time-consuming, because the integrations were
extremely long. However, if we use the modified theorem and differentiate first, the calculations are much shorter.
With all three loads acting (Fig. 9-39), the bending moments and
their derivatives are as follows (see Eqs. n and o):
L
M Px M0 Q x
2
L2
]M
0
0 x
]Q
]M
L
x
]Q
2
M Px M0
L
2
x
L
Therefore, the deflection (dC)0, from Eq. (9-88), is
1
(dC)0
EI
1
EI
L /2
(Px M0)(0)dx
0
L
L /2
Px M
0
L
Q x
2
x L2 dx
Since Q is a fictitious load, and since we have already taken the partial
derivatives, we can set Q equal to zero before integrating and obtain the
deflection dC due to the two loads P and M0 as follows:
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
654
CHAPTER 9 Deflections of Beams
1
dC
EI
L
M 0 L2
L
5PL3
[Px M0] x dx
2
48EI
8 EI
L /2
which agrees with the earlier result (Eq. t). Again, the integrations are
greatly simplified by differentiating under the integral sign and using the
modified theorem.
The partial derivative that appears under the integral sign in Eq.
(9-88) has a simple physical interpretation. It represents the rate of
change of the bending moment M with respect to the load Pi, that is, it is
equal to the bending moment M produced by a load Pi of unit value.
This observation leads to a method of finding deflections known as the
unit-load method. Castigliano’s theorem also leads to a method of structural analysis known as the flexibility method. Both the unit-load method
and the flexibility method are widely used in structural analysis and are
described in textbooks on that subject.
The following examples provide additional illustrations of the use
of Castigliano’s theorem for finding deflections of beams. However, it
should be remembered that the theorem is not limited to finding beam
deflections—it applies to any kind of linearly elastic structure for which
the principle of superposition is valid.
Example 9-17
P
q
A
B
C
L
—
2
L
—
2
FIG. 9-40 Example 9-17. Simple beam
with two loads
q
M
A simple beam AB supports a uniform load of intensity q 1.5 k/ft and a concentrated load P 5 k (Fig. 9-40). The load P acts at the midpoint C of the
beam. The beam has length L 8.0 ft, modulus of elasticity E 30 106 psi,
and moment of inertia I 75.0 in.4
Determine the downward deflection dC at the midpoint of the beam by
the following methods: (1) Obtain the strain energy of the beam and use
Castigliano’s theorem, and (2) use the modified form of Castigliano’s theorem
(differentiation under the integral sign).
Solution
Method (1). Because the beam and its loading are symmetrical about the
midpoint, the strain energy for the entire beam is equal to twice the strain
energy for the left-hand half of the beam. Therefore, we need to analyze only
the left-hand half of the beam.
The reaction at the left-hand support A (Figs. 9-40 and 9-41) is
A
qL
P
RA
2
2
V
x
and therefore the bending moment M is
qL
P
—
RA = —
2 2
qx2
qLx
qx2
Px
M RA x
2
2
2
2
FIG. 9-41 Free-body diagram for
determining the bending moment M in
the left-hand half of the beam
in which x is measured from support A.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(x)
SECTION 9.9 Castigliano’s Theorem
655
The strain energy of the entire beam (from Eq. 9-80a) is
U
M 2dx
2
2EI
L/2
0
qLx
qx2 2
1 Px
dx
2EI 2
2
2
After squaring the term in parentheses and performing a lengthy integration, we
find
5PqL4
q2L5
P2L3
U
384EI
96EI
240EI
Since the deflection at the midpoint C (Fig. 9-40) corresponds to the load P, we
can find the deflection by using Castigliano’s theorem (Eq. 9-87):
5PqL4
q2L5
5qL4
]U
] P 2L3
PL3
dC
]P
]P 96EI
384EI
240EI
48EI
384EI
(y)
Method (2). By using the modified form of Castigliano’s theorem (Eq.
9-88), we avoid the lengthy integration for finding the strain energy. The bending moment in the left-hand half of the beam has already been determined (see
Eq. x), and its partial derivative with respect to the load P is
]M
x
]P
2
Therefore, the modified Castigliano’s theorem becomes
dC
EMI ]]MP dx
L /2
2
0
qLx
qx2 x
5qL4
1 Px
PL3
dx
EI 2
2
2
2
48EI
384EI
(z)
which agrees with the earlier result (Eq. y) but requires a much simpler
integration.
Numerical solution. Now that we have an expression for the deflection at
point C, we can substitute numerical values, as follows:
5qL4
PL3
dC
48EI
384EI
(5 k)(96 in.)3
5(1.5 k/ft)(1/12 ft/in.)(96 in.)4
6
4
48(30 10 psi)(75.0 in. )
384(30 106 psi)(75.0 in.4)
0.0410 in. 0.0614 in. 0.1024 in.
Note that numerical values cannot be substituted until after the partial derivative
is obtained. If numerical values are substituted prematurely, either in the expression for the bending moment or the expression for the strain energy, it may be
impossible to take the derivative.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
656
CHAPTER 9 Deflections of Beams
Example 9-18
A simple beam with an overhang supports a uniform load of intensity q on span
AB and a concentrated load P at end C of the overhang (Fig. 9-42).
Determine the deflection dC and angle of rotation uC at point C. (Use the
modified form of Castigliano’s theorem.)
q
P
A
C
B
B
A
dC
L
—
2
L
C
uC
(b)
(a)
FIG. 9-42 Example 9-18. Beam with an
overhang
Solution
Deflection dC at the end of the overhang (Fig. 9-42b). Since the load P
corresponds to this deflection, we do not need to supply a fictitious load.
Instead, we can begin immediately to find the bending moments throughout the
length of the beam. The reaction at support A is
qL
P
RA
2
2
q
A
P
C
B
x1
L
as shown in Fig. 9-43. Therefore, the bending moment in span AB is
q x 21
qL x1
Px1
q x 21
MAB RA x1
2
2
2
2
(0
x1
L)
where x1 is measured from support A (Fig. 9-43). The bending moment in the
overhang is
x2
L
—
2
MBC Px2
qL
P
—
RA = —
2 2
FIG. 9-43 Reaction at support A and
coordinates x1 and x2 for the beam of
Example 9-18
0
x2
L
2
where x2 is measured from point C (Fig. 9-43).
Next, we determine the partial derivatives with respect to the load P:
]MAB
x1
]P
2
(0
x1
L)
]MBC
x2
]P
0
x2
L
2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
657
SECTION 9.9 Castigliano’s Theorem
Now we are ready to use the modified form of Castigliano’s theorem
(Eq. 9-88) to obtain the deflection at point C:
dC
EMI ]]MP dx
1
EI
L
0
]MAB
1
MAB dx
]P
EI
L /2
0
]MBC
MBC dx
]P
Substituting the expressions for the bending moments and partial derivatives,
we get
1
d C
EI
L
0
qLx1
Px1
q x 21
x1
1
dx1
2
2
2
2
EI
L /2
(Px2)(x2)dx2
0
By performing the integrations and combining terms, we obtain the deflection:
qL4
PL3
dC
8EI
48EI
(9-89)
Since the load P acts downward, the deflection dC is also positive downward. In
other words, if the preceding equation produces a positive result, the deflection
is downward. If the result is negative, the deflection is upward.
Comparing the two terms in Eq. (9-89), we see that the deflection at the
end of the overhang is downward when P qL/6 and upward when P qL/6.
Angle of rotation uC at the end of the overhang (Fig. 9-42b). Since there is
no load on the original beam (Fig. 9-42a) corresponding to this angle of rotation, we must supply a fictitious load. Therefore, we place a couple of moment
MC at point C (Fig. 9-44). Note that the couple MC acts at the point on the beam
where the angle of rotation is to be determined. Furthermore, it has the same
clockwise direction as the angle of rotation (Fig. 9-42).
We now follow the same steps as when determining the deflection at C.
First, we note that the reaction at support A (Fig. 9-44) is
qL
MC
P
RA
2
2
L
Consequently, the bending moment in span AB becomes
qLx1
Px1
MC x1
q x 21
q x 21
MAB RA x1
2
2
L
2
2
q
(0
P
MC
A
C
B
x2
x1
FIG. 9-44 Fictitious moment MC acting
on the beam of Example 9-18
RA
L
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
L
—
2
x1
L)
658
CHAPTER 9 Deflections of Beams
Also, the bending moment in the overhang becomes
0
MBC Px2 MC
L
2
x2
The partial derivatives are taken with respect to the moment MC, which is
the load corresponding to the angle of rotation. Therefore,
]MAB
x1
]MC
L
]MBC
1
]MC
0
0
x1
x2
L
L
2
Now we use the modified form of Castigliano’s theorem (Eq. 9-88) to
obtain the angle of rotation at point C:
EMI ]]MMdx
uC
C
1
EI
L
0
]MAB
1
MAB dx
EI
]MC
L /2
0
]MBC
MBC dx
]MC
Substituting the expressions for the bending moments and partial derivatives,
we obtain
1
uC
EI
1
EI
L
0
qLx1
Px1
MC x1
q x 21
x1
dx1
L
2
2
2
L
L /2
(Px2 MC)(1)dx 2
0
Since MC is a fictitious load, and since we have already taken the partial derivatives, we can set MC equal to zero at this stage of the calculations and simplify
the integrations:
1
uC
EI
L
0
qLx1
Px1
q x 21
x1
1
dx1
2
2
2
L
EI
L /2
(Px2)(1)dx2
0
After carrying out the integrations and combining terms, we obtain
qL3
7PL2
uC
24EI
24EI
(9-90)
If this equation produces a positive result, the angle of rotation is clockwise. If
the result is negative, the angle is counterclockwise.
Comparing the two terms in Eq. (9-90), we see that the angle of rotation is
clockwise when P qL/7 and counterclockwise when P qL/7.
If numerical data are available, it is now a routine matter to substitute
numerical values into Eqs. (9-89) and (9-90) and calculate the deflection and
angle of rotation at the end of the overhang.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.10 Deflections Produced by Impact
★9.10
659
DEFLECTIONS PRODUCED BY IMPACT
W
M= —
g
h
A
B
L
—
2
L
—
2
(a)
B
A
dmax
(b)
FIG. 9-45 Deflection of a beam struck by
a falling body
In this section we will discuss the impact of an object falling onto a
beam (Fig. 9-45a). We will determine the dynamic deflection of the
beam by equating the potential energy lost by the falling mass to the
strain energy acquired by the beam. This approximate method was
described in detail in Section 2.8 for a mass striking an axially loaded
bar; consequently, Section 2.8 should be fully understood before
proceeding.
Most of the assumptions described in Section 2.8 apply to beams as
well as to axially loaded bars. Some of these assumptions are as follows:
(1) The falling weight sticks to the beam and moves with it, (2) no
energy losses occur, (3) the beam behaves in a linearly elastic manner,
(4) the deflected shape of the beam is the same under a dynamic load as
under a static load, and (5) the potential energy of the beam due to its
change in position is relatively small and may be disregarded. In
general, these assumptions are reasonable if the mass of the falling object
is very large compared to the mass of the beam. Otherwise, this approximate analysis is not valid and a more advanced analysis is required.
As an example, we will consider the simple beam AB shown in Fig.
9-45. The beam is struck at its midpoint by a falling body of mass M and
weight W. Based upon the preceding idealizations, we may assume that
all of the potential energy lost by the body during its fall is transformed
into elastic strain energy that is stored in the beam. Since the distance
through which the body falls is h dmax, where h is the initial height
above the beam (Fig. 9-45a) and dmax is the maximum dynamic deflection of the beam (Fig. 9-45b), the potential energy lost is
Potential energy W(h dmax)
(a)
The strain energy acquired by the beam can be determined from the
deflection curve by using Eq. (9-80b), which is repeated here:
U
2
EI d v 2
dx
2 dx 2
(b)
The deflection curve for a simple beam subjected to a concentrated load
acting at the midpoint (see Case 4 of Table G-2, Appendix G) is
Px
v (3L 2 4x 2)
48EI
0
x
L
2
(c)
Also, the maximum deflection of the beam is
PL3
dmax
48EI
(d)
Eliminating the load P between Eqs. (c) and (d), we get the equation of
the deflection curve in terms of the maximum deflection:
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
660
CHAPTER 9 Deflections of Beams
0
dmax x
2
v (3L
4x2)
L3
L
2
x
(e)
Taking two derivatives, we find
24dmaxx
d 2v
2
(f )
dx
L3
Finally, we substitute the second derivative into Eq. (b) and obtain the
following expression for the strain energy of the beam in terms of
the maximum deflection:
L /2
U2
0
dx EI
EI d 2 v
2
2 dx
2
L /2
0
24EId 2max
L dx L
24dmaxx
2
3
3
(g)
Equating the potential energy lost by the falling mass (Eq. a) to the
strain energy acquired by the beam (Eq. g), we get
24EId 2max
W(h d max)
L3
(9-91)
This equation is quadratic in dmax and can be solved for its positive root:
WL3
d max
48EI
4W8L
EI
3
2
WL3
2h
48EI
1/2
(9-92)
We see that the maximum dynamic deflection increases if either the
weight of the falling object or the height of fall is increased, and it
decreases if the stiffness EI/L3 of the beam is increased.
To simplify the preceding equation, we will denote the static deflection of the beam due to the weight W as dst:
WL3
d st
48EI
(9-93)
Then Eq. (9-92) for the maximum dynamic deflection becomes
d max d st (d 2st 2hd st)1/2
(9-94)
This equation shows that the dynamic deflection is always larger than
the static deflection.
If the height h equals zero, which means that the load is applied
suddenly but without any free fall, the dynamic deflection is twice the
static deflection. If h is very large compared to the deflection, then the
term containing h in Eq. (9-94) predominates, and the equation can be
simplified to
d max
2hdst
(9-95)
These observations are analogous to those discussed previously in
Section 2.8 for impact on a bar in tension or compression.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.11 Discontinuity Functions
661
The deflection dmax calculated from Eq. (9-94) generally represents
an upper limit, because we assumed there were no energy losses during
impact. Several other factors also tend to reduce the deflection, including localized deformation of the contact surfaces, the tendency of the
falling mass to bounce upward, and inertia effects of the mass of the
beam. Thus, we see that the phenomenon of impact is quite complex,
and if a more accurate analysis is needed, books and articles devoted
specifically to that subject must be consulted.
★9.11
DISCONTINUITY FUNCTIONS
Discontinuity functions are utilized in a variety of engineering applications, including beam analysis, electrical circuits, and heat transfer.
These functions probably are the easiest to use and understand when
applied to beams, and therefore the study of mechanics of materials
offers an excellent opportunity to become familiar with them. The mathematics of the functions are described in this section, and we also will
see how the functions are used to represent loads on beams. Then, in the
next section, we will use them for finding slopes and deflections of
prismatic beams.
The unique feature of discontinuity functions is that they permit the
writing of a discontinuous function by a single expression, whereas the
more conventional approach requires that a discontinuous function be
described by a series of expressions, one for each region in which the
function is distinct.
For instance, if the loading on a beam consists of a mixture of concentrated and distributed loads, we can write with discontinuity
functions a single equation that applies throughout the entire length,
whereas ordinarily we must write separate equations for each segment
of the beam between changes in loading. In a similiar manner, we can
express shear forces, bending moments, slopes, and deflections of a
beam by one equation each, even though there may be several changes
in the loads acting along the axis of the beam.
These results are achievable because the functions themselves are
discontinuous; that is, they have different values in different regions of
the independent variable. In effect, these functions can pass through discontinuities in a manner that is not possible with ordinary continuous
functions. However, because they differ significantly from the functions
we are accustomed to, discontinuity functions must be used with care
and caution.
Two kinds of functions, called Macaulay functions and singularity
functions, will be discussed in this section. Although these functions
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
662
CHAPTER 9 Deflections of Beams
have different definitions and properties, together they form a family of
discontinuity functions.*
Macaulay Functions
Macaulay functions are used to represent quantities that “begin” at some
particular point on the x axis and have the value zero to the left of that
point. For instance, one of the Macaulay functions, denoted F1, is
defined as follows:
F1(x) x a1
0x a
when x
when x
a
a
(9-96)
In this equation, x is the independent variable and a is the value of x at
which the function “begins.” The pointed brackets (or angle brackets)
are the mathematical symbol for a discontinuity function.
In the case of the function F1, the pointed brackets (with the superscript 1) tell us that the function has the value zero when x is less than or
equal to a (that is, when the expression within the brackets is negative or
zero) and a value equal to x a when x is greater than or equal to a.
A graph of this function, called the unit ramp function, is given in
Figure 9-46.
1
F
1 (x
)=
x–
a
F1(x) = x – a
F1(x) = 0
O
FIG. 9-46 Graph of the Macaulay
a
x
function F1 (the unit ramp function)
In general terms, the Macaulay functions are defined by the following expressions:
Fn(x) x an
(x0 a)
n
when x
when x
a
a
(9-97)
We see from this definition that the Macaulay functions have the value
zero to the left of the point x a and the value (x a)n to the right of
that point. Except for the case n 0, which is discussed later, the function equals zero at x a.
*Sometimes both kinds of discontinuity functions are called singularity functions, but
this usage obscures the distinctions between the two functions, which obey different
mathematical laws. Furthermore, Macaulay functions do not have singularities.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.11 Discontinuity Functions
Another way to express this definition is the following: If the
quantity x a within the pointed brackets is negative or zero, the
Macaulay function has the value zero; if the function x a is positive or
zero, the Macaulay function has the value obtained by replacing the
pointed brackets with parentheses.
The preceding definition of the Macaulay functions holds for values
of n equal to positive integers and zero. When n 0, note that the function takes on the following special values:
y
y=3 x–2
4
0
3
2
1
–3
–2
–1
1
0
–1
2
3
4
x
1
y=–— x+1
4
1
0
–3
–2
2
–1
1
2
3
4
x
–2
–3
y
y=2 x+2 0
– x 1
+ x–3 1
2
1
–2
–1
0
–1
10
when x
when x
a
a
Fn(x) x an (x a)nx a0
–1
–3
F0(x) x a0
1
2
3
FIG. 9-47 Graphs of expressions
involving Macaulay functions
4
(9-98)
This function has a vertical “step” at the point of discontinuity x a;
thus, at x a it has two values: zero and one. The function F0, called
the unit step function, is pictured in the lower part of Table 9-1 along
with other Macaulay functions.*
The Macaulay functions of higher degree can be expressed in terms
of the unit step function, as follows:
y
2
663
x
(9-99)
This equation is easily confirmed by comparing Eqs. (9-97) and (9-98).
Some of the basic algebraic operations, such as addition, subtraction, and multiplication by a constant, can be performed on the
Macaulay functions. Illustrations of these elementary operations are
given in Fig. 9-47. The important thing to note from these examples is
that a function y having different algebraic expressions for different
regions along the x axis can be written as a single equation through the
use of Macaulay functions. The reader should verify each of the graphs
in Fig. 9-47 in order to become familiar with the construction of these
expressions.
The Macaulay functions can be integrated and differentiated according to the formulas given in the last column of Table 9-1. These
formulas can be verified by ordinary differentiation and integration of
the functions in the two regions x a and x a.
The units of the Macaulay functions are the same as the units of x n;
that is, F0 is dimensionless, F1 has units of x, F2 has units of x2, and so
forth.**
*The unit step function is also known as the Heaviside step function, denoted by H(x a),
after Oliver Heaviside (1850–1925), an English physicist and electrical engineer.
**The use of special brackets for discontinuity functions was introduced by the English
mathematician W. H. Macaulay in 1919 (Ref. 9-7), and the brackets are often called
Macaulay’s brackets. However, Macaulay actually used braces { } to identify the functions; the pointed brackets in common use today were introduced at a later time. The
general concept of combining two or more expressions into a single function by using a
special symbol predates Macaulay’s work (see Refs. 9-8 through 9-10).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
664
CHAPTER 9 Deflections of Beams
TABLE 9-1 DEFINITIONS OF DISCONTINUITY FUNCTIONS
Name
Definition
Derivative and
Integral
Graph
Singularity functions
F–2
Unit doublet function
F2 x a2
0
0
a
x
F–1
Unit impulse function
F1 x a1
F2 dx F1
F1dx F0
F0 dx F1
x
xa
xa
x
0
xa
x a
0
a
x
F0
Unit step function
0 x
F0 x a
1 x
0
a
a
x
1
0
a
x
d
F1 F0
dx
Macaulay functions
F1
Unit ramp function
F1 x a1
xa
0
x
x
a
a
0
a
x
F2 x a2
(x a)
0
2
x
x
a
a
0
Fn x an
(x a)
0
n
x
x
n 0, 1, 2, 3, . . .
x
a
x
a
a
x
0
F2
F1dx
2
F
F2 dx 3
3
d
Fn nFn1
dx
n 1, 2, 3, . . .
Fn
General Macaulay
function
d
F2 2F1
dx
F2
Unit second-degree
function
x
a
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
x
Fn1
Fndx
n1
n 0, 1, 2, 3, . . .
SECTION 9.11 Discontinuity Functions
665
Singularity Functions
The second kind of discontinuity functions consists of the singularity
functions, as defined by the following expressions:
Fn(x) x an
0
when x a
when x a
(9-100)
n 1, 2, 3, . . .
Note that singularity functions are defined for negative integer values of
n, whereas Macaulay functions are defined for positive integers and for
zero. The pointed brackets identify both kinds of functions, but the
brackets have different meanings in the two cases (compare Eqs. 9-97
and 9-100).
The singularity functions have the value zero everywhere except at
the singular point x a. The singularities arise because, when n is a
neative integer, the function (x a)n can be written as a fraction with
the expression x a in the denominator; thus, when x a, the function
becomes infinite.
The two most important singularity functions are pictured in the
upper part of Table 9-1, where we see that the kind of singularity
depends upon the value of n. The unit doublet function (n 2) has a
singularity that can be pictured by two arrows of infinite extent, one
directed upward and the other downward, with the arrows being infinitesimally close to each other. For convenience, these arrows can be
visualized as forces, and then the doublet can be represented by a curved
arrow that is the moment of the two forces. This moment is equal to the
product of an infinite force and a vanishingly small distance; the moment
turns out to be finite and equal to unity. For this reason, the doublet is
also known as the unit moment. (A third name in common usage is
dipole.)
The unit impulse function (n 1) is also infinite at x a but in
a different way. It can be pictured by a single arrow, as shown in Table 9-1.
If the arrow is visualized as a force, then the force has infinite intensity
and acts over an infinitesimally small distance along the x axis. The
force is equal to the intensity times the distance over which it acts; this
product also turns out to be finite and equal to unity. Thus, the name
unit force is also used for this function.*
*The terminology unit impulse function comes from the use of this function in dynamics,
where the x axis is the time axis. In physics and mathematics, this function is denoted
by d(x a) and is called the Dirac delta function, named for the English theoretical
physicist Paul A. M. Dirac (1902–1984) who developed it.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
666
CHAPTER 9 Deflections of Beams
Singularity functions are sometimes categorized as pathological
functions or improper functions, because they are neither continuous nor
differentiable at x a. However, singularity functions can be integrated
through the singularities; the integration formula is as follows:
x
x
Fn dx
x an dx x an1 Fn1
(9-101)
n 1, 2, 3, . . .
Note that this formula is not the same as the general integration formula
for the Macaulay functions, which is given in the last line and last column of Table 9-1.
Equation (9-101) shows that the integral of the unit doublet function
is the unit impulse function, and the integral of the unit impulse function
is the unit step function (see Table 9-1).
The units of the singularity functions, like those of the Macaulay
functions, are the same as the units of x n. Thus, the doublet function has
units of 1/x 2 and the impulse function has units of 1/x.
Representation of Loads on Beams
by Discontinuity Functions
The discontinuity functions listed in Table 9-1 are ideally suited for representing loads on beams, such as couples, forces, uniform loads, and
varying loads. The shapes of the various loading diagrams exactly match
the shapes of the corresponding functions F2, F1, F0, F1, and so on. It
is necessary only to multiply the functions given in the table (which are
the unit functions) by the appropriate load intensities in order to obtain
mathematical representations of the loads.
Several cases of loads on beams are listed in Table 9-2 for convenient reference. Each case is explained in detail in the following
paragraphs. More complicated cases of loading can be handled by superposition of these elementary ones.
To explain how the expressions in Table 9-2 are obtained, let us
consider a uniform load (Case 3). This load can be expressed in terms
of the unit step function F0, which is given by the following formula
(see Table 9-1):
F0(x) x a0
(a)
This function has the value 0 for x a and the value 1 for x a. If
the function is multiplied by the constant q0, representing the intensity
of the uniform load, it becomes an expression for the uniformly distributed load on a beam:
q(x) q0x a0
(b)
The load q(x) defined by this expression has the value 0 for x a and
a. Thus, at x a, the function equals zero if we
the value q0 for x
approach from the left, and it equals q0 if we approach from the right.
Note that Eq. (b) is listed in the last column of Table 9-2, Case 3.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.11 Discontinuity Functions
TABLE 9-2 LOAD INTENSITIES REPRESENTED BY DISCONTINUITY FUNCTIONS
Intensity q(x) of equivalent
distributed load (positive downard)
Load on beam
(shown positive)
Case
M0
q(x) M0 x a2
1
0
x
a
P
q(x) Px a1
2
0
x
a
q0
q(x) q0x a0
3
0
x
a
b
4
0
q0
q0
q(x) x a1
b
q0
q0
q(x) x
a2
b2
x
a
b
5
0
x
a
q0
q(x) q0x a10
6
0
a1
a2
b
x
q0
7
0
a1
q0
a2
x
a1
q0
q(x) x a11
b
q0
x a21
b
q0 x a20
q(x) q0x a10
b
8
0
q0 x a20
a2
x
q0
x a11
b
q0
x a21
b
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
667
668
CHAPTER 9 Deflections of Beams
The direction of the load represented by Eq. (b) could be either
upward or downward, depending upon the sign convention for distributed loads. Since we assume that downward uniform loads are positive
(see Fig. 9-4), the expression for q(x) represents the load pictured in Case 3.
Note that the uniform load q0 continues indefinitely to the right along the
x axis.
Cases 4 and 5 in Table 9-2 can be explained in a manner similar to
that for Case 3, using the ramp and second-degree functions. Both loading functions continue indefinitely to the right. To define the functions,
we need to know the coordinates of a particular point on each graph. A
convenient method is to give the ordinate q0 at distance b from the point
where x a, as shown in the table.
The loading in Case 6 is a segment of a uniform load that begins
at x a1 and ends at x a2. This loading can be expressed as the
superposition of two loads. The first load is a uniform load of intensity
q0 that begins at x a1 and continues indefinitely to the right (see
Case 3); the second load has intensity q0, begins at x a2, and also
continues indefinitely to the right. Thus, the second load cancels the first
load in the region to the right of the point x a2.
The loads in Cases 7 and 8 also consist of segments of distributed
loads and are obtained by combining more elementary load patterns.
The reader should verify the expressions given in the last column of the
table for these two cases. More complicated loading patterns involving
distributed loads are obtained by similar techniques of superposition
using the Macaulay functions.
Loads in the form of couples or forces (Cases 1 and 2) are handled
by the singularity functions, with the unit doublet function representing
a unit couple and the unit impulse function representing a unit force.
When the unit doublet function is multiplied by M0, it represents a couple as an equivalent distributed load of intensity q(x). The units of M0
are force times length, and the units of the unit doublet function are
length to the power 2. Thus, their product has units of force divided by
length, which are the correct units for intensity of distributed load.
The situation is similar for a concentrated load, because the product
of force P and the unit impulse function has units of load intensity.
Thus, the equations for q(x) given in Cases 1 and 2 are mathematical
expressions defining the equivalent load intensities for a couple and a
force, respectively.
The sign conventions for Cases 1 and 2 are as shown in Table 9-2,
namely, loads in the form of couples are positive when counterclockwise,
and loads in the form of forces are positive when downward. As mentioned previously, the equivalent loads q(x) are positive when downward.
The process of writing an expression for the equivalent distributed
load q(x) acting on a beam is illustrated in the following three examples
(Examples 9-19, 9-20, and 9-21). Then, in the next section (Section
9.12), we show how these same expressions for q(x) are integrated to
obtain shear forces, bending moments, slopes, and deflections.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.11 Discontinuity Functions
669
Example 9-19
y
P
M0
B
A
x
L/3
L/3
L/3
The simple beam AB shown in Fig. 9-48a supports a concentrated load P and a
couple M0.
(a) Write the expression for the intensity q(x) of the equivalent distributed
load acting on the beam in the region between the supports (0 x L).
(b) Write the expression for the intensity q(x) for the entire beam, including
the reactions at ends A and B (0 x L).
(a)
Solution
y
P
M0
B
A
x
RA
RB
(b)
FIG. 9-48 Example 9-19. Representation
of loads by discontinuity functions
(a) Intensity of the equivalent distributed load in the region between the
supports. The only loads acting on the beam between the supports are the concentrated load P and the couple M0. Their equivalent loads are obtained from
Table 9-2, Cases 2 and 1, respectively. Thus, with the origin of coordinates at
support A, we can write (by inspection) the following equation:
L
q(x) P x
3
1
2L
M0 x
3
2
(0
x
L)
(9-102)
This one equation gives the equivalent distributed load at every point along the
axis of the beam except at the ends. Note that the expression on the right-hand
side is zero everywhere except at the load points. At those points, the intensity
is infinite.
(b) Intensity of the equivalent distributed load for the entire beam, including the reactions at the ends. When we include the reactions, the resulting
expression for q(x) will be valid at all points along the axis of the beam.
To obtain q(x), we begin by showing the reactions on a free-body diagram
of the beam (Fig. 9-48b). Then, we can determine the reactions from static
equilibrium:
M0
2P
RA
L
3
M0
P
RB
L
3
(9-103a,b)
With the reactions known, we can write the equation for q(x) by inspection,
again using Cases 1 and 2 of Table 9-2:
L
q(x) RAx1 P x
3
1
2L
M0 x
3
2
RBx L1
(0
x
L) (9-104)
This equation gives the equivalent distributed load throughout the length of the
beam, including both ends.
Note: By integrating (Eq. (9-102) or Eq. (9-104), we can obtain the shear
forces, bending moments, slopes, and deflections of the beam, as described in
the next section.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
670
CHAPTER 9 Deflections of Beams
Example 9-20
y
q0
A
D
B
x
L/3
L/3
L/3
(a)
y
MA
A
Solution
q0
D
B
x
(a) Intensity of the distributed load in the region to the right of the fixed
support. The only load on the beam is the triangular load having an intensity of
zero at x L/3 and an intensity of q0 at x L. This load fits Table 9-2, Case 7,
from which we get the following expression for q(x) in terms of discontinuity
functions:
q0
L
q(x) x
2L / 3
3
RA
(b)
FIG. 9-49 Example 9-20. Representation
of loads by discontinuity functions
A cantilever beam ADB supports a triangular load of maximum intensity q0, as
shown in Fig. 9-49a.
(a) Write the expression for the intensity q(x) of the distributed load acting
on the beam in the region to the right of the fixed support (0 x L).
(b) Write the expression for the intensity q(x) of the equivalent distributed
load for the entire beam, including the reactions at the fixed support
(0 x L).
1
q0
x L1 q0x L0
2L / 3
(0
x
L) (9-105)
This equation describes the load pictured in Fig. 9-49a, with the load having
zero intensity for the first one-third of the beam’s length, then increasing to q0 at
the end of the beam, and finally dropping to zero at that same point. The reason
that the load drops to zero at the end of the beam is because we used Case 7
from Table 9-2. The last two terms in Case 7 cancel the part of the triangular
load that otherwise would continue indefinitely to the right of the actual beam.
In practice, there is no harm in letting the triangular load continue past the
end of the beam, because any loads for which x is greater than L have no physical meaning. Therefore, we can remove the two terms containing x L in
Eq. (9-105) and simplify the expression for q(x):
3q0
L
q(x) x
2L
3
1
(0
x
(9-106)
L)
Note that both of the removed terms have the value zero at every point along the
length of the beam, thus providing further evidence that their removal will have
no effect on any of our calculations.
Furthermore, we now recognize that we could have obtained our final
result (Eq. 9-106) in a simpler and more direct manner by using Case 4 of
Table 9-2, instead of using Case 7.
As a further observation, we note that the load on the beam (Eq. 9-106) can
be written in more conventional form by referring to the definition of the unit
ramp function in Table 9-1. From the definition given there, we obtain the following expressions for the load on the beam:
For 0
For L /3
L /3:
x
x
q(x) 0
3q0
L
L: q(x) x
2L
3
These expressions are the equivalent of Eq. (9-106) and serve as a check on its
accuracy.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.11 Discontinuity Functions
671
(b) Intensity of the equivalent distributed load for the entire beam, including the reactions at the fixed support. We begin by finding the reactions RA and
MA at end A of the beam (Fig. 9-49b). From equilibrium, we get
7q0L2
MA
27
q0L
RA
3
(9-107a,b)
Using Cases 2 and 1, respectively, of Table 9-2, we now obtain (by inspection)
the following equation for q(x):
1
3q0
L
q(x) RA x1 MAx2 x
2L
3
(0
x
L)
(9-108)
This equation takes into account all of the forces acting on the beam, and in
Example 9-22 in the next section we will integrate this equation to obtain the
shear forces, bending moments, slopes, and deflections of the beam.
Example 9-21
The beam ACBD shown in Fig. 9-50a has simple supports at A and B and an
overhang from B to D. The loads on the beam consist of a uniform load (q
800 lb/ft) extending from C to B and a concentrated load (P 1500 lb) at end
D. Point C is at the midpoint of span AB, which has a length of 12 ft. The length
of the overhang is 4 ft.
Write the expression for the intensity q(x) of the equivalent distributed load
for the beam, including the reactions.
P = 1500 lb
y
q = 800 lb/ft
A
C
B
D
x
6 ft
6 ft
4 ft
(a)
1500 lb
y
A
800 lb/ft
C
B
D
x
FIG. 9-50 Example 9-21. Representation
of loads by discontinuity functions
RA = 700 lb
RB = 5600 lb
(b)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
672
CHAPTER 9 Deflections of Beams
P = 1500 lb
y
q = 800 lb/ft
A
C
B
D
x
6 ft
6 ft
4 ft
(a)
1500 lb
y
A
800 lb/ft
C
B
D
x
RA = 700 lb
RB = 5600 lb
(b)
FIG. 9-50 (Repeated)
Solution
We begin by calculating the reactions of the beam from static equilibrium
using the free-body diagram of Fig. 9-50b. The results are
RA 700 lb
RB 5600 lb
Now we can write the expression for q(x) by inspection, using Cases 2 and 6 of
Table 9-2:
q(x) (700 lb)x1 (800 lb/ft)x 6 ft0 (800 lb/ft)x 12 ft0
(5600 lb)x 12 ft1 (1500 lb)x 16 ft1
(0 x 16 ft) (9-109)
In this equation, x has units of feet (ft) and q(x) has units of pounds per foot
(lb/ft).
Since it is a bit cumbersome to include the units in every term of Eq. (9-109),
we can omit the units and rewrite q(x) in simpler form:
q(x) 700x1 800x 60 800x 120
5600x 121 1500x 161
(0
x
16)
(9-110)
The units used in this equation are the same as for Eq. (9-109), that is, x has
units of feet and q(x) has units of pounds per foot.
Equation (9-110) can now be integrated to obtain shear forces, bending
moments, slopes, and deflections (see Example 9-23 in the next section).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.12 Use of Discontinuity Functions in Determining Beam Deflections
★9.12
673
USE OF DISCONTINUITY FUNCTIONS IN DETERMINING BEAM DEFLECTIONS
The conventional integration methods for finding beam deflections are
described earlier in this chapter (see Sections 9.3 and 9.4). In those
methods we begin by writing expressions for the loads, shear forces, or
bending moments for each segment of the beam between points where
the loads change. Then these expressions are integrated separately in
order to find slopes and deflections. Both boundary and continuity conditions are needed to determine the resulting constants of integration,
hence this method is satisfactory for simple loadings but may become
unwieldy if the number of loading segments exceeds two or three.
The use of discontinuity functions makes it possible to write one
expression that is valid throughout the entire length of a beam, as illustrated in the preceding section. When such an expression is integrated,
the number of constants of integration is greatly reduced. Sometimes
only one constant of integration is needed, and sometimes no constant is
required. Thus, in certain types of problems the use of discontinuity
functions can be very useful.
The general procedure for the use of discontinuity functions is
quite straightforward. First, we write the expression for the equivalent
distributed load acting on the beam, using the techniques described in
the preceding section. Then this expression is substituted into the basic
differential equation of the deflection curve, namely, the load equation
(Eq. 9-12c).
Next, the differential equation is integrated successively to obtain
the shear force V, bending moment M, slope v, and deflection v (see
Eqs. 9-12). The integrations produce constants of integration that can be
evaluated from boundary conditions. Thus, we ultimately arrive at a
single expression for each of the unknown quantities (V, M, v, and v),
and each expression is valid for every point along the axis of the beam.
There are many variations in the details of the integration
procedure, depending upon the particular beam being analyzed and the
personal preferences of the analyst. Some of these variations are brought
out in the following discussion of the procedure for using discontinuity
functions.
The analysis may begin with the load equation, the shear-force
equation, or the bending-moment equation. However, in our examples,
we always start with the load equation because then the shear-force and
bending-moment equations are automatically included in the solution.*
Note: Throughout this section, we consider only prismatic beams
having constant flexural rigidity EI, which is a requirement of the differential equations that we use (Eqs. 9-12a, b, and c). As a general
*The discontinuity method for determining beam deflections is sometimes called
Clebsch’s method, after the German engineer who developed it in 1883 (Refs. 9-8
through 9-10).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
674
CHAPTER 9 Deflections of Beams
observation, it is not practical to use discontinuity functions for nonprismatic beams.
Procedure for the Use of Discontinuity Functions
y
P
M0
B
A
x
L/3
L/3
L/3
(a)
y
P
M0
B
A
x
RA
RB
(b)
FIG. 9-51 Use of discontinuity functions in
beam analysis
To illustrate in detail the procedure for determining shear forces, bending moments, slopes, and deflections of a beam, let us consider again a
simple beam (Fig. 9-51) supporting a concentrated load P and a couple
M0. (This beam was discussed previously in Example 9-19 of the
preceding section.) The following seven steps serve as an outline of the
analysis.
1. Equivalent distributed loads. In Example 9-19 we obtained two
equations that describe the loads on this beam in terms of discontinuity
functions. The first of the two equations (Eq. 9-102) gives the intensity
q(x) of the equivalent distributed load acting on the beam in the region
between the supports, which means that the equation involves only the
applied loads P and M0. We refer to this equation as the applied-load
equation and repeat it here:
L
q(x) P x
3
1
2L
M0 x
3
2
(0
x
L)
(9-111)
The second equation (Eq. 9-104) gives the intensity q(x) for the entire
beam, including the reactions. We refer to this equation as the load-andreaction equation and repeat it also:
L
q(x) RAx1 P x
3
1
2L
M0 x
3
2
(0
RBx L1
x
L) (9-112)
The reactions appearing in this equation (from Eqs. 9-103a and b) are
M0
2P
RA
3
L
M0
P
RB
3
L
(9-113a,b)
Note that the terms in the load-and-reaction equation (Eq. 9-112) consist
of the complete set of forces and couples that hold the beam in equilibrium (Fig. 9-51b).
2. Shear forces. The shear forces in the beam are determined by
the integration of either of the two load equations given in Step 1. The
basic differential equations are (see Eqs. 9-12c and b, respectively):
EIv q
EIv V
(9-114a,b)
The procedure is to substitute the expression for the equivalent distributed load q(x) into Eq. (9-114a) and then integrate once to obtain the
shear forces (Eq. 9-114b).
If the applied-load equation is used as the starting point, a constant
of integration is required with the integration. The constant is needed in
order to account for the nature of the supports and reactions. When con-
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.12 Use of Discontinuity Functions in Determining Beam Deflections
675
sidered solely by themselves, as in the applied-load equation, the loads
could be acting on a simple beam, a cantilever beam, or some other
beam. However, when the loads are integrated to obtain shear forces, the
shear forces must be consistent with the supports and reactions of the
particular beam under consideration.
In other words, we must recognize that a given set of applied loads
will produce different shear forces on different beams. The consistency
between loads and shear forces is accomplished by including a constant
of integration that is evaluated from a boundary condition based upon
the supports of the beam.
If the load-and-reaction equation is used as the starting point for
finding the shear forces, no constant of integration is required. If a constant is included, it will turn out to have the value zero when evaluated
from one of the boundary conditions. A constant is not needed because
the load-and-reaction equation already includes the effects of the
supports.
The expression for the shear forces is the same no matter which
equivalent-load equation is used as the starting point.
3. Shear forces found from the applied-load equation. In this
case, we substitute q(x) from Eq. (9-111) into Eq. (9-114a) and obtain
L
EIv q P x
3
1
2L
M0 x
3
2
(0
x
L) (9-115)
Integration of this equation (see Eq. 9-114b) yields
M x 23L
L
EIv V P x
3
0
1
0
C0
(a)
in which C0 is the constant of integration. To evaluate C0, let us use the
boundary condition that the shear force at the left-hand support (or more
precisely, the shear force just to the right of the left-hand support) is
equal to the reaction RA:
(b)
V(0) RA
Applying this condition to Eq. (a), we obtain
V(0) RA P(0) M0(0) C0
or
C0 RA
(c)
Substituting into Eq. (a), we get
M x 23L
L
EIv V RA P x
3
0
1
0
(9-116)
This equation gives the shear forces in the beam in terms of discontinuity functions.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
676
CHAPTER 9 Deflections of Beams
If we wish, we can write the shear forces in more conventional
terms by applying the definitions of the step and impulse functions
(Table 9-1) to the preceding equation:
For 0
x
For L /3
For 2L/3
V RA P(0) M0(0) RA
(d)
2L /3
V RA P(1) M0(0) RA P
(e)
L
V RA P(1) M0(0) RA P
(f)
L/3:
x
x
These expressions for the shear forces are readily verified by statics.
4. Shear forces found from the load-and-reaction equation. In this
case, we substitute q(x) from Eq. (9-112) into Eq. (9-114a) and obtain
L
EIv q RAx1 P x
3
1
2L
M0 x
3
(0
2
RBx L1
L) (9-117)
x
The last term in this equation, representing the reaction at support B, is
zero at every point along the axis of the beam up to the support itself.
Consequently, this term will have no effect on the calculation of shear
forces, bending moments, slopes, and deflections, and we can omit it
from the equation.
Integrating the remainder of Eq. (9-117), we obtain
M x 23L
L
EIv V RAx0 P x
3
0
1
0
(9-118)
Note that no constant of integration is included. Also, note that the reaction RA is multiplied by x0, which is equal to unity at every point along
the axis of the beam. Therefore, substituting 1 for x0,we get
M x 23L
L
EIv V RA P x
3
0
1
0
(9-119)
which is identical to Eq. (9-116).
5. Bending moments. The bending moments in the beam are determined by the integration of the shear-force equation obtained in Step 4
(Eq. 9-119), using the following basic differential equations (see Eqs. 9-12b
and a, respectively):
EIv V
EIv M
(9-120a,b)
No constant of integration is required for the bending-moment integration because the shear-force equation already includes the effects of
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
677
SECTION 9.12 Use of Discontinuity Functions in Determining Beam Deflections
supports and reactions. Thus, integrating Eq. (9-119) yields the following equation for the bending moments:
M x 23L
L
EIv M RAx P x
3
1
0
(9-121)
0
Note that we could have obtained this equation directly from equilibrium and free-body diagrams, and then we could have started our slope
and deflection analyses with this equation instead of the load equation.
As in the case of shear forces, we can write the bending moments in
conventional terms by applying the definitions of the ramp and step
functions (Table 9-1) to the preceding equation; thus,
For 0
For L/3
x
L/3:
x
M RA x P(0) M0(0) RA x
(g)
2L/3
M RA x P(x L/3) M0(0) RA x P(x L/3) (h)
For 2L /3 x L
M RA x P(x L/3) M0 (1) RA x P(x L/3) M0
(i)
These expressions for the bending moments are readily verified by statics.
6. Slopes and deflections of the beam. Integrating the bending
moments leads to the slopes of the beam, and integrating the slopes
leads to the deflections. However, since the bending moments are
obtained from equations of statics that do not involve the deflection
curve of the beam, we will need a constant of integration with each of
these two integrations. These constants are determined from known conditions on the slopes and deflections.
The need for constants of integration can be seen from the following
example. The bending moments derived in Step 5 for the simple beam
of Fig. 9-51 would be unchanged if the beam was altered by moving the
left-hand support from the end of the beam to the point where the concentrated load acts and then supplying a load equal to RA at the left-hand
end, which is now a free end. This new beam would be subjected to the
same forces as the original beam and would have the same bending
moments. Other arrangements of the supports and forces would also
produce the same bending moments. However, the slopes and deflections would be quite different for each of these beams. Therefore, we
need constants of integration that take into account the support conditions for the particular beam being investigated.
The first integration of the bending-moment equation (Eq. 9-121)
yields the following equation for the slope:
x2
P
L
EIv RA x
2
2
3
M x 23L C
2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
1
0
1
(9-122)
678
CHAPTER 9 Deflections of Beams
y
P
M0
A
B
x
L/3
L/3
L/3
RA
RB
(a)
y
C
A
uA
B
x
dC
L/2
FIG. 9-52 Deflection curve for the beam
(b)
of Fig. 9-51
Since we do not have a known boundary condition for the slope, we
cannot find C1 at this stage. Therefore, we continue with the next integration, which yields the following equation for the deflection:
x3
P
L
EIv RA x
3
6
6
3
C xC
M0
2L
x
2
3
2
1
2
(9-123)
The deflection curve for this beam is portrayed in Fig. 9-52.
Now we can utilize the two boundary conditions that pertain to the
deflections; namely, the deflection is zero at x 0 and x L:
v(0) 0 and
v(L) 0
(j,k)
The first condition, when applied to Eq. (9-123), produces this equation:
RA
M0
P
0 (0) (0) (0) C1(0) C2
6
2
6
from which C2 0 (1)
The second condition leads to this equation:
3
L3
P 2L
0 RA
6
6 3
2
M L
0
2 3
C1(L) 0
from which we get
RAL2
M0 L
4PL2
C1
6
81
18
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(m)
SECTION 9.12 Use of Discontinuity Functions in Determining Beam Deflections
679
Now, substituting for RA from Eq. (9-113a) and then combining terms,
we obtain
M0 L
5PL2
C1
81
9
(n)
With both constants determined, we can now write the final equations for the slopes and deflections by substituting for C1, C2, and RA in
Eqs. (9-122) and (9-123). The results are
M0 x 2
P
2P
L
EIv x
3
3
2
L
2
2
5P81L M9L
(9-124)
5P81L M9L x
(9-125)
2L
M0 x
3
M0 x 3
2P
P
L
EIv x
3
L
6
6
3
1
2
0
3
M0
2L
x
2
3
2
2
0
7. Slopes and deflections at particular points of the beam. From
the preceding two equations we can obtain the slopes and deflections at
any desired positions along the axis of the beam. For instance, at the
left-hand support we find the following formula for the angle of rotation
uA (see Fig. 9-52b) by substituting x 0 into Eq. (9-124):
M0 L
5PL2
uA v(0)
81EI
9EI
(9-126)
in which uA is positive when clockwise.
At the midpoint of the beam the deflection dC (positive downward)
is obtained from Eq. (9-125) by substituting x L /2; the result is
5M0 L2
L
23PL3
dC v
2
1296EI
144EI
(9-127)
In a similar manner, we can find any other desired deflections and
angles of rotation for the beam pictured in Fig. (9-52).
The following two examples provide further illustrations of the use
of discontinuity functions in determining beam deflections.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
680
CHAPTER 9 Deflections of Beams
Example 9-22
y
q0
A
D
B
x
L/3
L/3
Solution
L/3
Equivalent distributed loads. The equivalent load q(x) for this beam,
including both reactions and applied loads, is given by Eq. (9-108) in Example
9-20. That equation is repeated here:
(a)
y
MA
A
Using discontinuity functions, determine the shear forces, bending moments,
slopes, and deflections for the cantilever beam ADB supporting the triangular
load shown in Fig. 9-53.
Note: The reactions and equivalent distributed loads for this beam were
determined in Example 9-20 of the preceding section.
q0
D
B
x
3q0
L
q(x) RAx1 MAx2 x
3
2L
1
(0
x
(9-128)
L)
in which the reactions are
RA
q0L
RA
3
(b)
FIG. 9-53 Example 9-22. Analysis of a
cantilever beam supporting a triangular
load
7q0 L2
MA
27
(9-129a,b)
Substituting the reactions into the equation for q(x) and then using the differential equation in terms of the load (Eq. 9-114a), we obtain
q0 L
7q0 L2
3q0
L
EIv q x1 x2 x
3
3
27
2L
1
(0
x
L)
(9-130)
This equation is the starting point for the integrations that follow.
Shear forces. We obtain the shear forces by integrating the load equation,
as follows:
q0L
7q0 L2
3q0
L
EIv V x0 x1 x
3
3
27
4L
2
(9-131)
No constant of integration is needed because the reactions are already included
in the load equation (Eq. 9-130).
Bending moments. Integrating the shear forces yields an expression for the
bending moments:
q0 L
7q0 L2
q0
L
EIv M x1 x0 x
3
27
4L
3
3
(9-132)
Again, no constant of integration is needed.
Slopes and deflections. Two more integrations are needed, each with a constant of integration, as explained previously:
C
q0L
7q0L2
q0
L
EIv x2 x1 x
6
27
16L
3
(o)
1
C xC
q0L
7q0L2
q0
L
EIv x3 x2 x
18
54
80L
3
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
4
5
1
2
(p)
SECTION 9.12 Use of Discontinuity Functions in Determining Beam Deflections
681
The constants C1 and C2 are evaluated from the conditions at the fixed support
(Fig. 9-53):
v(0) 0 and
v(0) 0
(q,r)
Substituting the first condition into Eq. (o) and the second into Eq. (p), we get
C1 0 and
C2 0
(s,t)
Therefore, the final equations for the slope and deflection, respectively, are
qL
7q L
q
L
EIv x x x
18
54
80L
3
q0L
7q0 L2
q0
L
EIv x2 x1 x
6
27
16L
3
0
0
3
2
(9-133)
5
0
2
4
(9-134)
Thus, we have obtained the equations for the entire deflection curve in terms of
discontinuity functions.
Deflections and angles of rotation. Using the two preceding equations, we
can determine the deflection and angle of rotation at any desired position along
the beam axis (see Fig. 9-54). For instance, at point D where x L/3, we obtain
q0 L4
L
dD v
3
81EI
11q0 L3
L
uD v
3
162EI
(9-135a,b)
Note that the deflection dD is positive downward and the angle of rotation uD is
positive clockwise.
Also, at the free end B(x L), the deflection and angle of rotation are
92q0 L4
dB v(L)
1215EI
17q0 L3
uB v(L)
162EI
(9-136a,b)
We leave it to the reader to verify these formulas by substituting into Eqs.
(9-133) and (9-134).
y
q0
A
D
B
x
L/3
L/3
L/3
(a)
y
L/3
A
D
B
dB
dD
FIG. 9-54 Example 9-22. Deflections and
angles of rotation for the cantilever
beam of Fig. 9-53
x
uD
uB
(b)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
682
CHAPTER 9 Deflections of Beams
Example 9-23
The steel beam ACBD shown in Fig. 9-55 consists of a simple span AB with an
overhang BD. Using discontinuity functions, determine the vertical deflections
dC and dD at points C and D, respectively (Fig. 9-55c). The modulus of elasticity of the steel is E 30 106 psi and the moment of inertia of the beam cross
section is I 25.92 in.4
Note: The reactions and equivalent distributed loads for this beam were
determined in Example 9-21 of the preceding section.
P = 1500 lb
y
q = 800 lb/ft
A
C
B
D
x
6 ft
6 ft
4 ft
(a)
1500 lb
y
800 lb/ft
A
C
B
D
x
RA = 700 lb
RB = 5600 lb
(b)
y
A
dD
B
C
x
dC
FIG. 9-55 Example 9-23. Analysis of a
(c)
simple beam with an overhang
Solution
Equivalent distributed loads. The equivalent load q(x) for this beam,
including both applied loads and reactions (Fig. 9-55b), is given by Eq. (9-110),
which is repeated here:
q(x) 700x1 800x 60 800x 120
5600x 121 1500x 161
(0
x
16) (9-137)
In this equation, x has units of feet (ft) and q(x) has units of pounds per foot
(lb/ft). The last term in the equation equals zero at all points along the axis of
the beam, except at the right-hand end (where x 16 ft). Therefore, the last
term may be omitted.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.12 Use of Discontinuity Functions in Determining Beam Deflections
683
Substituting Eq. (9-137) into the basic differential equation for the load
(Eq. 9-114a), we get the following fourth-order equation:
EIv q 700x1 800x 60 800x 120 5600x 121
(9-138)
To maintain consistent units throughout the calculations, we will use foot and
pound units for all variables in this equation:
E
I
v
v
q
x
pounds per square foot (lb/ft2)
feet to the fourth power (ft4)
feet (ft)
feet raised to the power minus 3 (ft3)
pounds per foot (lb/ft)
feet (ft)
In accord with the above units, we will use the following numerical values for E
and I:
E (30
106 psi)(144 in.2/ft2) 4320
106 lb/ft2
25.92 in.4
I 4 0.00125 ft4
(12 in./ft)
(u)
(v)
Shear forces and bending moments. The first two integrations give the
equations for shear forces and bending moments. No integration constants are
needed, because the load equation (Eq. 9-138) includes the effects of the reactions. The equations for shear and moment are
EIv V 700x0 800x 61 800x 121 5600x 120 (9-139)
EIv M 700x1 400x 62 400x 122 5600x 121 (9-140)
Slopes and deflections. The next two integrations, both of which require a
constant of integration, are
EIv 350x2 (400/3)x 63 (400/3)x 123 2800x 122 C1
(w)
EIv (350/3)x3 (100/3)x 64 (100/3)x 124
(2800/3)x 123 C1x C2
(x)
At supports A and B, the deflections are zero. Therefore, the boundary conditions are
v(0) 0 and
v(12 ft) 0
From the first condition we find C2 0, and from the second we find
0 (350/3)(123) (100/3)(64) (100/3)(0) (2800/3)(0) (C1)(12)
from which we get C1 13,200 lb-ft2.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
684
CHAPTER 9 Deflections of Beams
Thus, the final expressions for v and v are
y
A
dD
C
B
dC
(c)
x
EIv 350x2 (400/3)x 63 (400/3)(x 123
2800x 122 13,200
(9-141)
EIv (350/3)x3 (100/3)(x 64 (100/3)x 124
(2800/3)x 123 13,200x
(9-142)
From these equations we can determine the slope and deflection at any point
along the axis of the beam.
Deflection at point C. To find the deflection dC midway between the supports (Fig. 9-55c), we substitute x 6 ft into Eq. (9-142), as follows:
FIG. 9-55c (Repeated)
EIv(6 ft) 25,200 0 0 0 79,200 54,000 lb-ft3
from which we get
54,000 lb-ft3
v(6 ft)
EI
(y)
Finally, we solve for the deflection dC itself:
54,000 lb-ft3
dC v(6 ft)
0.0100 ft 0.120 in.
(4320 106 lb/ft2)(0.00125 ft4)
Note that in Fig. 9-55c we define the positive direction of dC as downward, and
therefore a minus sign is used in the preceding equation. Since the final calculations produce a positive value for dC, we know that the actual deflection is
downward.
Deflection at point D. Following a similar procedure for the deflection dD
at the end of the overhang (Fig. 9-55c), we substitute x 16 ft into Eq. (9-142)
and obtain
EIv(16 ft) (350/3)(4096) (100/3)(10,000) (100/3)(256)
(2800/3)(64) (13,200)
(16)
1600 lb-ft3
Therefore,
1600 lb-ft3
v(16 ft)
EI
(z)
and
1600 lb-ft3
dD v(16 ft)
0.0002963 ft
(4320 106 lb/ft2)(0.00125 ft4)
0.00356 in.
In this case, we defined the positive direction of dD as upward (Fig. 9-55c),
which is the same as the positive direction of v. Since the calculations produce a
positive value, we know that the actual deflection is upward.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 9.13 Temperature Effects
★9.13
685
TEMPERATURE EFFECTS
y
T1
x
h
T2
x
d T a(T )L
dx
L
(a)
T2
T1
(b)
du
T1
dx
T2
In the preceding sections of this chapter we considered the deflections
of beams due to lateral loads. In this section, we will consider the
deflections caused by nonuniform temperature changes. As a preliminary matter, recall that the effects of uniform temperature changes
have already been described in Section 2.5, where it was shown that a
uniform temperature increase causes an unconstrained bar or beam to
have its length increased by the amount
h
—
2
h
—
2
(c)
FIG. 9-56 Temperature effects in a beam
(9-143)
In this equation, a is the coefficient of thermal expansion, T is the
uniform increase in temperature, and L is the length of the bar (see
Fig. 2-20 and Eq. 2-16 in Chapter 2).
If a beam is supported in such a manner that longitudinal expansion
is free to occur, as is the case for all of the statically determinate beams
considered in this chapter, then a uniform temperature change will not
produce any stresses in the beam. Also, there will be no lateral deflections of such a beam, because there is no tendency for the beam to bend.
The behavior of a beam is quite different if the temperature is not
constant across its height. For example, assume that a simple beam,
initially straight and at a uniform temperature T0, has its temperature
changed to T1 on its upper surface and T2 on its lower surface, as
pictured in Fig. 9-56a. If we assume that the variation in temperature
is linear between the top and bottom of the beam, then the average
temperature of the beam is
T1 T2
(9-144)
Taver
2
and occurs at midheight. Any difference between this average temperature and the initial temperature T0 results in a change in length of the
beam, given by Eq. (9-143), as follows:
T1 T2
d T a(Taver T0)L a T0 L
(9-145)
2
In addition, the temperature differential T2 T1 between the bottom and
top of the beam produces a curvature of the axis of the beam, with the
accompanying lateral deflections (Fig. 9-56b).
To investigate the deflections due to a temperature differential,
consider an element of length dx cut out from the beam (Figs. 9-56a
and c). The changes in length of the element at the bottom and top are
a(T2 T0)dx and a(T1 T0)dx, respectively. If T2 is greater than T1,
the sides of the element will rotate with respect to each other through an
angle du, as shown in Fig. 9-56c. The angle du is related to the changes
in dimension by the following equation, obtained from the geometry of
the figure:
h du a(T2 T0)dx a(T1 T0)dx
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
686
CHAPTER 9 Deflections of Beams
from which we get
a(T2 T1)
du
h
dx
(9-146)
in which h is the height of the beam.
We have already seen that the quantity du/dx represents the curvature of the deflection curve of the beam (see Eq. 9-4). Since the
curvature is equal to d2v /dx2 (Eq. 9-5), we may write the following
differential equation of the deflection curve:
a(T2 T1)
d 2v
2
dx
h
(9-147)
Note that when T2 is greater than T1, the curvature is positive and the
beam is bent concave upward, as shown in Fig. 9-56b. The quantity
a (T2 T1)/h in Eq. (9-147) is the counterpart of the quantity M/EI,
which appears in the basic differential equation (Eq. 9-7).
We can solve Eq. (9-147) by the same integration techniques
described earlier for the effects of bending moments (see Section 9.3).
We can integrate the differential equation to obtain dv/dx and v, and we
can use boundary or other conditions to evaluate the constants of
integration. In this manner we can obtain the equations for the slopes
and deflections of the beam, as illustrated by Problems 9.13-1 through
9.13-4 at the end of this chapter.
If the beam is able to change in length and deflect freely, there will
be no stresses associated with the temperature changes described in this
section. However, if the beam is restrained against longitudinal expansion or lateral deflection, or if the temperature changes do not vary
linearly from top to bottom of the beam, internal temperature stresses
will develop. The determination of such stresses requires the use of
more advanced methods of analysis.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 9 Problems
687
PROBLEMS CHAPTER 9
Differential Equations of the Deflection Curve
The beams described in the problems for Section 9.2 have
constant flexural rigidity EI.
9.2-1 The deflection curve for a simple beam AB (see
figure) is given by the following equation:
q0 x
v (7L4 10L2x 2 3x4)
360LEI
(a) Describe the load acting on the beam.
(b) Determine the reactions RA and MA at the support.
Deflection Formulas
Problems 9.3-1 through 9.3-7 require the calculation of
deflections using the formulas derived in Examples 9-1,
9-2, and 9-3. All beams have constant flexural rigidity EI.
Describe the load acting on the beam.
y
B
A
q0 x 2
v
(45L4 40L3x 15L2x 2 x 4)
360L2EI
x
9.3-1 A wide-flange beam (W 12 35) supports a uniform
load on a simple span of length L 14 ft (see figure).
Calculate the maximum deflection dmax at the midpoint and the angles of rotation u at the supports if q
1.8 k/ft and E 30 106 psi. Use the formulas of Example 9-1.
L
q
PROBS. 9.2-1 and 9.2-2
h
9.2-2 The deflection curve for a simple beam AB (see
figure) is given by the following equation:
4
q0L
px
v 4 sin
p EI
L
L
(a) Describe the load acting on the beam.
(b) Determine the reactions RA and RB at the supports.
(c) Determine the maximum bending moment Mmax.
PROBS. 9.3-1, 9.3-2, and 9.3-3
9.2-3 The deflection curve for a cantilever beam AB (see
figure) is given by the following equation:
q0 x 2
v (10L3 10L2x 5Lx 2 x 3)
120LEI
9.3-2 A uniformly loaded steel wide-flange beam with simple supports (see figure) has a downward deflection
of 10 mm at the midpoint and angles of rotation equal to
0.01 radians at the ends.
Calculate the height h of the beam if the maximum
bending stress is 90 MPa and the modulus of elasticity is
200 GPa. (Hint: Use the formulas of Example 9-1.)
Describe the load acting on the beam.
9.3-3 What is the span length L of a uniformly loaded sim-
y
A
B
x
L
PROBS. 9.2-3 and 9.2-4
9.2-4 The deflection curve for a cantilever beam AB (see
figure) is given by the following equation:
ple beam of wide-flange cross section (see figure) if the
maximum bending stress is 12,000 psi, the maximum
deflection is 0.1 in., the height of the beam is 12 in., and the
modulus of elasticity is 30 106 psi? (Use the formulas of
Example 9-1.)
9.3-4 Calculate the maximum deflection dmax of a uniformly
loaded simple beam (see figure on the next page) if the span
length L 2.0 m, the intensity of the uniform load q
2.0 kN/m, and the maximum bending stress s 60 MPa.
The cross section of the beam is square, and the material
is aluminum having modulus of elasticity E 70 GPa.
(Use the formulas of Example 9-1.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
688
CHAPTER 9 Deflections of Beams
q = 2.0 kN/m
P
A
B
L = 2.0 m
a
b
L
PROB. 9.3-4
9.3-5 A cantilever beam with a uniform load (see figure)
has a height h equal to 1/8 of the length L. The beam is a
steel wide-flange section with E 28
106 psi and an
allowable bending stress of 17,500 psi in both tension and
compression.
Calculate the ratio d /L of the deflection at the free end
to the length, assuming that the beam carries the maximum
allowable load. (Use the formulas of Example 9-2.)
q
PROB. 9.3-7
Deflections by Integration of the Bending-Moment
Equation
Problems 9.3-8 through 9.3-16 are to be solved by
integrating the second-order differential equation of the
deflection curve (the bending-moment equation). The origin
of coordinates is at the left-hand end of each beam, and all
beams have constant flexural rigidity EI.
9.3-8 Derive the equation of the deflection curve for a
cantilever beam AB supporting a load P at the free end (see
figure). Also, determine the deflection dB and angle of
rotation uB at the free end. (Note: Use the second-order
differential equation of the deflection curve.)
h
L
PROB. 9.3-5
y
9.3-6 A gold-alloy microbeam attached to a silicon wafer
behaves like a cantilever beam subjected to a uniform load
(see figure). The beam has length L 27.5 m and rectangular cross section of width b 4.0 m and thickness
t 0.88 m. The total load on the beam is 17.2 N.
If the deflection at the end of the beam is 2.46 m,
what is the modulus of elasticity Eg of the gold alloy? (Use
the formulas of Example 9-2.)
q
t
b
L
P
A
B
x
L
PROB. 9.3-8
9.3-9 Derive the equation of the deflection curve for a simple beam AB loaded by a couple M0 at the left-hand support
(see figure). Also, determine the maximum deflection dmax.
(Note: Use the second-order differential equation of the
deflection curve.)
y
PROB. 9.3-6
9.3-7 Obtain a formula for the ratio dC /d max of the deflection at the midpoint to the maximum deflection for
a simple beam supporting a concentrated load P (see
figure).
From the formula, plot a graph of d C /d max versus the
ratio a /L that defines the position of the load (0.5
a /L 1). What conclusion do you draw from the graph?
(Use the formulas of Example 9-3.)
M0
B
A
L
PROB. 9.3-9
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
x
689
CHAPTER 9 Problems
9.3-10 A cantilever beam AB supporting a triangularly distributed load of maximum intensity q0 is shown in the
figure.
Derive the equation of the deflection curve and then
obtain formulas for the deflection dB and angle of rotation
uB at the free end. (Note: Use the second-order differential
equation of the deflection curve.)
9.3-13 Derive the equations of the deflection curve for a
simple beam AB loaded by a couple M0 acting at distance a
from the left-hand support (see figure). Also, determine the
deflection d0 at the point where the load is applied. (Note:
Use the second-order differential equation of the deflection
curve.)
y
y
M0
q0
B
A
B
A
x
x
a
b
L
L
PROB. 9.3-10
PROB. 9.3-13
9.3-11 A cantilever beam AB is acted upon by a uniformly
distributed moment (bending moment, not torque) of intensity m per unit distance along the axis of the beam (see
figure).
Derive the equation of the deflection curve and then
obtain formulas for the deflection dB and angle of rotation
uB at the free end. (Note: Use the second-order differential
equation of the deflection curve.)
y
9.3-14 Derive the equations of the deflection curve for a
cantilever beam AB carrying a uniform load of intensity
q over part of the span (see figure). Also, determine the
deflection dB at the end of the beam. (Note: Use the
second-order differential equation of the deflection curve.)
y
q
m
B
A
B
A
x
a
L
x
b
L
PROB. 9.3-11
9.3-12 The beam shown in the figure has a roller support at
PROB. 9.3-14
A and a guided support at B. The guided support permits
vertical movement but no rotation.
Derive the equation of the deflection curve and determine the deflection dB at end B due to the uniform load of
intensity q. (Note: Use the second-order differential equation of the deflection curve.)
9.3-15 Derive the equations of the deflection curve for a
cantilever beam AB supporting a uniform load of intensity
q acting over one-half of the length (see figure). Also,
obtain formulas for the deflections dB and dC at points B
and C, respectively. (Note: Use the second-order differential equation of the deflection curve.)
y
y
q
A
B
x
A
B
C
L
—
2
L
PROB. 9.3-12
q
PROB. 9.3-15
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
L
—
2
x
690
CHAPTER 9 Deflections of Beams
9.3-16 Derive the equations of the deflection curve for a
simple beam AB with a uniform load of intensity q acting
over the left-hand half of the span (see figure). Also, determine the deflection dC at the midpoint of the beam. (Note:
Use the second-order differential equation of the deflection
curve.)
Use the fourth-order differential equation of the deflection
curve (the load equation).
px
q = q0 sin —
L
y
B
A
y
x
q
B
A
L
—
2
L
x
C
PROB. 9.4-2
L
—
2
9.4-3 The simple beam AB shown in the figure has
moments 2M0 and M0 acting at the ends.
Derive the equation of the deflection curve, and then
determine the maximum deflection dmax. Use the thirdorder differential equation of the deflection curve (the
shear-force equation).
PROB. 9.3-16
Deflections by Integration of the Shear-Force
and Load Equations
y
The beams described in the problems for Section 9.4 have
constant flexural rigidity EI. Also, the origin of coordinates
is at the left-hand end of each beam.
2M0
M0
B
A
x
9.4-1 Derive the equation of the deflection curve for a
cantilever beam AB when a couple M0 acts counterclockwise
at the free end (see figure). Also, determine the deflection
dB and slope uB at the free end. Use the third-order
differential equation of the deflection curve (the shear-force
equation).
y
M0
B
A
x
L
PROB. 9.4-3
9.4-4 A simple beam with a uniform load is pin supported
at one end and spring supported at the other. The spring has
stiffness k 48EI/L3.
Derive the equation of the deflection curve by starting
with the third-order differential equation (the shear-force
equation). Also, determine the angle of rotation uA at
support A.
y
L
q
PROB. 9.4-1
B
A
x
48EI
—
k = L3
9.4-2 A simple beam AB is subjected to a distributed load
of intensity q q0 sin px/L, where q0 is the maximum
intensity of the load (see figure).
Derive the equation of the deflection curve, and then
determine the deflection dmax at the midpoint of the beam.
L
PROB. 9.4-4
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
691
CHAPTER 9 Problems
9.4-5 The distributed load acting on a cantilever beam AB
4q0 x
(L x)
q= —
L2
has an intensity q given by the expression q0 cos px /2L,
where q0 is the maximum intensity of the load (see figure).
Derive the equation of the deflection curve, and then
determine the deflection dB at the free end. Use the fourthorder differential equation of the deflection curve (the load
equation).
y
B
A
y
q0
x
L
px
q = q0 cos —
2L
PROB. 9.4-7
9.4-8 Derive the equation of the deflection curve for a
B
A
simple beam AB carrying a triangularly distributed load of
maximum intensity q0 (see figure). Also, determine the
maximum deflection dmax of the beam. Use the fourth-order
differential equation of the deflection curve (the load
equation).
x
L
9.4-6 A cantilever beam AB is subjected to a parabolically
varying load of intensity q q0(L2 x 2)/L2, where q0 is
the maximum intensity of the load (see figure).
Derive the equation of the deflection curve, and then
determine the deflection dB and angle of rotation uB at the
free end. Use the fourth-order differential equation of the
deflection curve (the load equation).
y
B
A
x
L
PROB. 9.4-8
9.4-9 Derive the equations of the deflection curve for an
overhanging beam ABC subjected to a uniform load of
intensity q acting on the overhang (see figure). Also, obtain
formulas for the deflection dC and angle of rotation uC at
the end of the overhang. Use the fourth-order differential
equation of the deflection curve (the load equation).
★
L2
x2
q = q0 —
L2
q0
q0
y
PROB. 9.4-5
B
A
x
y
q
B
A
L
C
x
PROB. 9.4-6
L
9.4-7 A beam on simple supports is subjected to a paraboli-
cally distributed load of intensity q 4q0 x(L x)/L2,
where q0 is the maximum intensity of the load (see figure).
Derive the equation of the deflection curve, and then
determine the maximum deflection d max. Use the fourthorder differential equation of the deflection curve (the load
equation).
L
—
2
PROB. 9.4-9
★9.4-10 Derive the equations of the deflection curve for a
simple beam AB supporting a triangularly distributed load
of maximum intensity q0 acting on the right-hand half of
the beam (see figure on the next page). Also, determine the
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
692
CHAPTER 9 Deflections of Beams
angles of rotation uA and uB at the ends and the deflection
dC at the midpoint. Use the fourth-order differential equation of the deflection curve (the load equation).
q0
y
C
A
B
9.5-3 The cantilever beam AB shown in the figure has an
extension BCD attached to its free end. A force P acts at the
end of the extension.
(a) Find the ratio a/L so that the vertical deflection of
point B will be zero.
(b) Find the ratio a/L so that the angle of rotation at
point B will be zero.
x
L
A
L
—
2
B
L
—
2
D
PROB. 9.4-10
P
a
C
Method of Superposition
PROB. 9.5-3
The problems for Section 9.5 are to be solved by the
method of superposition. All beams have constant flexural
rigidity EI.
9.5-4 Beam ACB hangs from two springs, as shown in the
P
figure. The springs have stiffnesses k1 and k2 and the beam
has flexural rigidity EI.
What is the downward displacement of point C,
which is at the midpoint of the beam, when the load P is
applied?
Data for the structure are as follows: P 5 8.0 kN,
L 1.8 m, EI 216 kNm2, k1 250 kN/m, and
k2 160 kN/m.
B
L = 1.8 m
9.5-1 A cantilever beam AB carries three equally spaced
concentrated loads, as shown in the figure. Obtain formulas
for the angle of rotation uB and deflection dB at the free end
of the beam.
P
P
A
L
—
3
L
—
3
L
—
3
k1 = 250 kN/m
PROB. 9.5-1
9.5-2 A simple beam AB supports five equally spaced loads
P (see figure).
(a) Determine the deflection d1 at the midpoint of the
beam.
(b) If the same total load (5P) is distributed as a uniform load on the beam, what is the deflection d 2 at the
midpoint?
(c) Calculate the ratio of d1 to d 2.
P
P
P
P
P
A
k2 = 160 kN/m
A
P = 8.0 kN
PROB. 9.5-4
9.5-5 What must be the equation y f (x) of the axis of the
slightly curved beam AB (see figure) before the load is
applied in order that the load P, moving along the bar,
always stays at the same level?
y
B
P
B
A
L
—
6
PROB. 9.5-2
L
—
6
L
—
6
L
—
6
L
—
6
B
C
L
—
6
L
PROB. 9.5-5
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
x
CHAPTER 9 Problems
9.5-6 Determine the angle of rotation uB and deflection dB
at the free end of a cantilever beam AB having a uniform
load of intensity q acting over the middle third of its length
(see figure).
q
B
A
L
—
3
L
—
3
9.5-9 A horizontal load P acts at end C of the bracket ABC
shown in the figure.
(a) Determine the deflection dC of point C.
(b) Determine the maximum upward deflection dmax of
member AB.
Note: Assume that the flexural rigidity EI is constant
throughout the frame. Also, disregard the effects of axial
deformations and consider only the effects of bending due
to the load P.
L
—
3
P
C
H
PROB. 9.5-6
6
2
flexural rigidity EI 2.1 10 k-in. Calculate the downward deflections dC and dB at points C and B, respectively,
due to the simultaneous action of the moment of 35 k-in.
applied at point C and the concentrated load of 2.5 k
applied at the free end B.
2.5 k
35 k-in.
A
L
PROB. 9.5-9
9.5-10 A beam ABC having flexural rigidity EI
75 kNm2 is loaded by a force P 800 N at end C and tied
down at end A by a wire having axial rigidity EA 900 kN
(see figure).
What is the deflection at point C when the load P is
applied?
B
C
48 in.
48 in.
PROB. 9.5-7
B
A
9.5-8 A beam ABCD consisting of a simple span BD and an
overhang AB is loaded by a force P acting at the end of the
bracket CEF (see figure).
(a) Determine the deflection dA at the end of the overhang.
(b) Under what conditions is this deflection upward?
Under what conditions is it downward?
2L
—
3
L
—
3
L
—
2
A
B
A
9.5-7 The cantilever beam ACB shown in the figure has
B
C
F
E
D
C
P = 800 N
0.5 m
D
0.5 m
0.75 m
PROB. 9.5-10
9.5-11 Determine the angle of rotation uB and deflection
dB at the free end of a cantilever beam AB supporting a
parabolic load defined by the equation q q0 x 2/L2 (see
figure).
q0
y
A
B
P
L
a
PROB. 9.5-8
693
PROB. 9.5-11
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
x
694
CHAPTER 9 Deflections of Beams
9.5-12 A simple beam AB supports a uniform load of intensity q acting over the middle region of the span (see figure).
Determine the angle of rotation uA at the left-hand
support and the deflection dmax at the midpoint.
For what stiffness k of the spring will the uniform load
produce no deflection at the free end C?
B
A
q
EI = 15 k-in.2
B
A
L
9.5-13 The overhanging beam ABCD supports two
concentrated loads P and Q (see figure).
(a) For what ratio P/Q will the deflection at point B be
zero?
(b) For what ratio will the deflection at point D be
zero?
P
9.5-16 A beam ABCD rests on simple supports at B and C
(see figure). The beam has a slight initial curvature so that
end A is 15 mm above the elevation of the supports and end
D is 10 mm above.
What loads P and Q, acting at points A and D, respectively, will move points A and D downward to the level of
the supports? (The flexural rigidity EI of the beam is
2.5 106 N·m2.)
Q
C
B
P
D
Q
A
15 mm
L
—
2
b = 15 in.
PROB. 9.5-15
PROB. 9.5-12
A
k
L = 30 in.
a
a
L
—
2
B
2.5 m
9.5-14 A thin metal strip of total weight W and length L is
placed across the top of a flat table of width L /3 as shown
in the figure.
What is the clearance d between the strip and the
middle of the table? (The strip of metal has flexural rigidity
EI.)
10 mm
2.5 m
2.5 m
PROB. 9.5-16
9.5-17 The compound beam ABCD shown in the figure has
fixed supports at ends A and D and consists of three members joined by pin connections at B and C.
Find the deflection d under the load P.
d
P
B
A
L
—
6
D
C
a
PROB. 9.5-13
L
—
3
C
L
—
6
L
—
3
PROB. 9.5-14
3b
C
b
b
D
b
PROB. 9.5-17
9.5-15 An overhanging beam ABC with flexural rigidity
EI 15 k-in.2 is supported by a pin support at A and by a
spring of stiffness k at point B (see figure). Span AB has
length L 30 in. and carries a uniformly distributed load.
The overhang BC has length b 15 in.
9.5-18 A compound beam ABCDE (see figure) consists of
two parts (ABC and CDE) connected by a hinge at C.
Determine the deflection dE at the free end E due to
the load P acting at that point.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 9 Problems
Note: Disregard the effects of axial deformations and
consider only the effects of bending due to the load P.
P
D
C
B
A
695
E
P
B
2b
b
b
C
b
c
PROB. 9.5-18
b
9.5-19 A steel beam ABC is simply supported at A and held
by a high-strength steel wire at B (see figure). A load P
240 lb acts at the free end C. The wire has axial rigidity
EA 1500 103 lb, and the beam has flexural rigidity
EI 36 106 lb-in.2
What is the deflection dC of point C due to the load P?
Wire
20 in.
P = 240 lb
Beam
B
A
C
A
PROB. 9.5-21
9.5-22 The frame ABCD shown in the figure is squeezed
by two collinear forces P acting at points A and D. What is
the decrease d in the distance between points A and D when
the loads P are applied? (The flexural rigidity EI is constant
throughout the frame.)
Note: Disregard the effects of axial deformations and
consider only the effects of bending due to the loads P.
★
P
20 in.
B
30 in.
A
a
PROB. 9.5-19
9.5-20 The compound beam shown in the figure consists
of a cantilever beam AB (length L) that is pin-connected to
a simple beam BD (length 2L). After the beam is constructed, a clearance c exists between the beam and a
support at C, midway between points B and D. Subsequently, a uniform load is placed along the entire length of
the beam.
What intensity q of the load is needed to close the gap
at C and bring the beam into contact with the support?
q
D
A
B
L
C
Pin
L
c
D
C
L
P
PROB. 9.5-22
9.5-23 A beam ABCDE has simple supports at B and D
and symmetrical overhangs at each end (see figure). The
center span has length L and each overhang has length b. A
uniform load of intensity q acts on the beam.
(a) Determine the ratio b/L so that the deflection dC at
the midpoint of the beam is equal to the deflections dA and
dE at the ends.
(b) For this value of b/L, what is the deflection dC at
the midpoint?
★★
q
L
A
PROB. 9.5-20
B
★9.5-21 Find the horizontal deflection d
h and vertical
deflection dv at the free end C of the frame ABC shown in
the figure. (The flexural rigidity EI is constant throughout
the frame.)
b
PROB. 9.5-23
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
C
L
E
D
b
696
CHAPTER 9 Deflections of Beams
★★★9.5-24 A frame ABC is loaded at point C by a force P
acting at an angle a to the horizontal (see figure). Both
members of the frame have the same length and the same
flexural rigidity.
Determine the angle a so that the deflection of point C
is in the same direction as the load. (Disregard the effects
of axial deformations and consider only the effects of bending due to the load P.)
Note: A direction of loading such that the resulting
deflection is in the same direction as the load is called a
principal direction. For a given load on a planar structure,
there are two principal directions, perpendicular to each
other.
P
L
q0
L
PROB. 9.6-2
9.6-3 A cantilever beam AB is subjected to a concentrated
load P and a couple M0 acting at the free end (see figure).
Obtain formulas for the angle of rotation uB and the
deflection dB at end B.
a
P
C
B
B
A
A
M0
B
L
L
PROB. 9.6-3
A
9.6-4 Determine the angle of rotation uB and the deflection
dB at the free end of a cantilever beam AB with a uniform
load of intensity q acting over the middle third of the length
(see figure).
PROB. 9.5-24
Moment-Area Method
q
The problems for Section 9.6 are to be solved by the
moment-area method. All beams have constant flexural
rigidity EI.
A
q
PROB. 9.6-4
9.6-5 Calculate the deflections dB and dC at points B and C,
B
A
L
—
3
L
—
3
L
—
3
9.6-1 A cantilever beam AB is subjected to a uniform load
of intensity q acting throughout its length (see figure).
Determine the angle of rotation uB and the deflection
dB at the free end.
B
respectively, of the cantilever beam ACB shown in the figure. Assume M0 36 k-in., P 3.8 k, L 8 ft, and EI
2.25 109 lb-in.2
L
PROB. 9.6-1
M0
P
C
B
A
9.6-2 The load on a cantilever beam AB has a triangular
distribution with maximum intensity q0 (see figure).
Determine the angle of rotation uB and the deflection
dB at the free end.
L
—
2
PROB. 9.6-5
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
L
—
2
CHAPTER 9 Problems
9.6-6 A cantilever beam ACB supports two concentrated
M0
loads P1 and P2 as shown in the figure.
Calculate the deflections dB and dC at points B and C,
respectively. Assume P1 10 kN, P2 5 kN, L 2.6 m,
E 200 GPa, and I 20.1 106 mm4.
A
P1
P2
C
B
697
A
B
L
PROB. 9.6-9
L
—
2
L
—
2
9.6-10 The simple beam AB shown in the figure supports
two equal concentrated loads P, one acting downward and
the other upward.
Determine the angle of rotation uA at the left-hand end,
the deflection d1 under the downward load, and the deflection d 2 at the midpoint of the beam.
★
PROB. 9.6-6
9.6-7 Obtain formulas for the angle of rotation uA at sup-
port A and the deflection dmax at the midpoint for a simple
beam AB with a uniform load of intensity q (see figure).
P
P
A
q
A
B
B
a
a
L
L
PROB. 9.6-10
PROB. 9.6-7
9.6-11 A simple beam AB is subjected to couples M0 and
2M0 acting as shown in the figure. Determine the angles of
rotation uA and uB at the ends of the beam and the deflection d at point D where the load M0 is applied.
★
9.6-8 A simple beam AB supports two concentrated loads P
at the positions shown in the figure. A support C at the midpoint of the beam is positioned at distance d below the
beam before the loads are applied.
Assuming that d 10 mm, L 6 m, E 200 GPa,
and I 198
106 mm4, calculate the magnitude of the
loads P so that the beam just touches the support at C.
P
d
A
L
—
4
2M0
B
D
L
—
3
P
B
E
L
—
3
L
—
3
PROB. 9.6-11
C
L
—
4
M0
A
L
—
4
L
—
4
PROB. 9.6-8
9.6-9 A simple beam AB is subjected to a load in the form
of a couple M0 acting at end B (see figure).
Determine the angles of rotation uA and uB at the supports and the deflection d at the midpoint.
Nonprismatic Beams
9.7-1 The cantilever beam ACB shown in the figure has
moments of inertia I2 and I1 in parts AC and CB, respectively.
(a) Using the method of superposition, determine the
deflection dB at the free end due to the load P.
(b) Determine the ratio r of the deflection dB to the
deflection d1 at the free end of a prismatic cantilever with
moment of inertia I1 carrying the same load.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
698
CHAPTER 9 Deflections of Beams
(c) Plot a graph of the deflection ratio r versus the ratio
I2 /I1 of the moments of inertia. (Let I2 /I1 vary from 1 to 5.)
P
I2
A
C
I1
B
9.7-4 A beam ABC has a rigid segment from A to B and
a flexible segment with moment of inertia I from B to C
(see figure). A concentrated load P acts at point B.
Determine the angle of rotation uA of the rigid segment, the deflection dB at point B, and the maximum
deflection dmax.
★★
L
—
2
L
—
2
Rigid
PROB. 9.7-1
I
A
9.7-2 The cantilever beam ACB shown in the figure supports a uniform load of intensity q throughout its length.
The beam has moments of inertia I2 and I1 in parts AC and
CB, respectively.
(a) Using the method of superposition, determine the
deflection dB at the free end due to the uniform load.
(b) Determine the ratio r of the deflection dB to the
deflection d1 at the free end of a prismatic cantilever with
moment of inertia I1 carrying the same load.
(c) Plot a graph of the deflection ratio r versus the ratio
I2 /I1 of the moments of inertia. (Let I2 /I1 vary from 1 to 5.)
q
A
C
9.7-5 A simple beam ABC has moment of inertia 1.5I
from A to B and I from B to C (see figure). A concentrated
load P acts at point B.
Obtain the equations of the deflection curves for both
parts of the beam. From the equations, determine the angles
of rotation uA and uC at the supports and the deflection dB
at point B.
★★
B
P
1.5I
A
9.7-3 A simple beam ABCD has moment of inertia I near
the supports and moment of inertia 2I in the middle region,
as shown in the figure. A uniform load of intensity q acts
over the entire length of the beam.
Determine the equations of the deflection curve for the
left-hand half of the beam. Also, find the angle of rotation
uA at the left-hand support and the deflection dmax at the
midpoint.
★
q
B
I
L
—
4
C
2I
L
D
I
L
—
4
I
C
B
2L
—
3
L
—
3
PROB. 9.7-2
A
2L
—
3
PROB. 9.7-4
L
—
2
L
—
2
C
B
L
—
3
I1
I2
PROB. 9.7-3
P
PROB. 9.7-5
9.7-6 The tapered cantilever beam AB shown in the
figure has thin-walled, hollow circular cross sections of
constant thickness t. The diameters at the ends A and B are
dA and dB 2dA, respectively. Thus, the diameter d and
moment of inertia I at distance x from the free end are,
respectively,
★★
dA
d (L x)
L
IA
pt d 3
p t d 3A
3
I
(L x)3
3 (L x)
L3
8
8L
in which IA is the moment of inertia at end A of the beam.
Determine the equation of the deflection curve and the
deflection dA at the free end of the beam due to the load P.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
699
CHAPTER 9 Problems
P
P
B
A
t
dA
B
A
3dA
dB = —
2
dA
dB = 2dA
d
x
x
L
d
b
L
PROB. 9.7-6
PROB. 9.7-8
9.7-9 A simple beam ACB is constructed with square
cross sections and a double taper (see figure). The depth of
the beam at the supports is dA and at the midpoint is
dC 2dA. Each half of the beam has length L. Thus, the
depth d and moment of inertia I at distance x from the lefthand end are, respectively,
★★
★★
9.7-7 The tapered cantilever beam AB shown in the
figure has a solid circular cross section. The diameters at
the ends A and B are dA and dB 2dA, respectively. Thus,
the diameter d and moment of inertia I at distance x from
the free end are, respectively,
dA
d (L x)
L
dA
d (L x)
L
4
d4
dA
IA
(L x)4
(L x)4
I
12
12L4
L4
p d 4A
IA
pd 4
I (L
x)4
(L x)4
4
64
64L
L4
in which IA is the moment of inertia at end A of the beam.
Determine the equation of the deflection curve and the
deflection dA at the free end of the beam due to the load P.
P
B
A
dB = 2dA
dA
x
in which IA is the moment of inertia at end A of the beam.
(These equations are valid for x between 0 and L, that is,
for the left-hand half of the beam.)
(a) Obtain equations for the slope and deflection of
the left-hand half of the beam due to the uniform load.
(b) From those equations obtain formulas for the angle
of rotation uA at support A and the deflection dC at the midpoint.
q
d
L
PROB. 9.7-7
C
A
★★9.7-8 A tapered cantilever beam AB supports a concentrated load P at the free end (see figure). The cross sections
of the beam are rectangular with constant width b, depth dA
at support A, and depth dB 3dA /2 at the support. Thus,
the depth d and moment of inertia I at distance x from the
free end are, respectively,
dA
(2L x)
d
2L
3
3
bd A
3 (2L
bd
I
12
96L
I
x)3 A(2L
x)3
8L3
in which IA is the moment of inertia at end A of the beam.
Determine the equation of the deflection curve and the
deflection dA at the free end of the beam due to the load P.
B
d
d
x
L
L
PROB. 9.7-9
Strain Energy
The beams described in the problems for Section 9.8 have
constant flexural rigidity EI.
9.8-1 A uniformly loaded simple beam AB (see figure on
the next page) of span length L and rectangular cross
section (b width, h height) has a maximum bending
stress smax due to the uniform load.
Determine the strain energy U stored in the beam.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
700
CHAPTER 9 Deflections of Beams
A
B
d
A
h
B
b
L
—
2
L
—
2
L
PROB. 9.8-1
PROB. 9.8-4
9.8-2 A simple beam AB of length L supports a concentrated load P at the midpoint (see figure).
(a) Evaluate the strain energy of the beam from the
bending moment in the beam.
(b) Evaluate the strain energy of the beam from
the equation of the deflection curve.
(c) From the strain energy, determine the deflection d
under the load P.
P
A
B
overhang BC supports a concentrated load P at the free end
C (see figure).
(a) Determine the strain energy U stored in the beam
due to the load P.
(b) From the strain energy, find the deflection dC under
the load P.
(c) Calculate the numerical values of U and dC if the
length L is 8 ft, the overhang length a is 3 ft, the beam is a
W 10 12 steel wide-flange section, and the load P
produces a maximum stress of 12,000 psi in the beam. (Use
E 29 106 psi.)
P
L
—
2
L
—
2
9.8-5 A beam ABC with simple supports at A and B and an
B
A
C
PROB. 9.8-2
load of intensity q (see figure).
(a) Evaluate the strain energy of the beam from the
bending moment in the beam.
(b) Evaluate the strain energy of the beam from the
equation of the deflection curve.
q
a
L
9.8-3 A cantilever beam AB of length L supports a uniform
PROB. 9.8-5
9.8-6 A simple beam ACB supporting a concentrated load
P at the midpoint and a couple of moment M0 at one end is
shown in the figure.
Determine the strain energy U stored in the beam due
to the force P and the couple M0 acting simultaneously.
P
A
A
B
M0
C
B
L
PROB. 9.8-3
9.8-4 A simple beam AB of length L is subjected to loads
that produce a symmetric deflection curve with maximum
deflection d at the midpoint of the span (see figure).
How much strain energy U is stored in the beam if the
deflection curve is (a) a parabola, and (b) a half wave of a
sine curve?
L
—
2
L
—
2
PROB. 9.8-6
9.8-7 The frame shown in the figure consists of a beam
ACB supported by a strut CD. The beam has length 2L and
is continuous through joint C. A concentrated load P acts at
the free end B.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 9 Problems
Determine the vertical deflection dB at point B due to
the load P.
Note: Let EI denote the flexural rigidity of the beam,
and let EA denote the axial rigidity of the strut. Disregard
axial and shearing effects in the beam, and disregard any
bending effects in the strut.
L
701
9.9-3 An overhanging beam ABC supports a concentrated
load P at the end of the overhang (see figure). Span AB has
length L and the overhang has length a.
Determine the deflection dC at the end of the overhang.
(Obtain the solution by determining the strain energy of the
beam and then using Castigliano’s theorem.)
L
P
B
A
C
A
P
B
C
L
L
D
a
PROB. 9.9-3
PROB. 9.8-7
Castigliano’s Theorem
The beams described in the problems for Section 9.9 have
constant flexural rigidity EI.
9.9-1 A simple beam AB of length L is loaded at the
left-hand end by a couple of moment M0 (see figure).
Determine the angle of rotation uA at support A.
(Obtain the solution by determining the strain energy of the
beam and then using Castigliano’s theorem.)
9.9-4 The cantilever beam shown in the figure supports a
triangularly distributed load of maximum intensity q0.
Determine the deflection dB at the free end B. (Obtain
the solution by determining the strain energy of the beam
and then using Castigliano’s theorem.)
q0
M0
A
B
B
A
L
PROB. 9.9-4
L
PROB. 9.9-1
9.9-2 The simple beam shown in the figure supports a
concentrated load P acting at distance a from the left-hand
support and distance b from the right-hand support.
Determine the deflection dD at point D where the load
is applied. (Obtain the solution by determining the strain
energy of the beam and then using Castigliano’s theorem.)
9.9-5 A simple beam ACB supports a uniform load of
intensity q on the left-hand half of the span (see figure).
Determine the angle of rotation uB at support B.
(Obtain the solution by using the modified form of
Castigliano’s theorem.)
q
P
A
B
D
a
b
L
—
2
L
PROB. 9.9-2
C
A
PROB. 9.9-5
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
B
L
—
2
702
CHAPTER 9 Deflections of Beams
9.9-6 A cantilever beam ACB supports two concentrated
loads P1 and P2, as shown in the figure.
Determine the deflections dC and dB at points C and B,
respectively. (Obtain the solution by using the modified
form of Castigliano’s theorem.)
P1
A
P2
C
9.9-9 A simple beam ABCDE supports a uniform load of
intensity q (see figure). The moment of inertia in the central
part of the beam (BCD) is twice the moment of inertia in
the end parts (AB and DE).
Find the deflection dC at the midpoint C of the beam.
(Obtain the solution by using the modified form of
Castigliano’s theorem.)
q
B
L
—
2
L
—
2
B
A
I
PROB. 9.9-6
9.9-7 The cantilever beam ACB shown in the figure is
subjected to a uniform load of intensity q acting between
points A and C.
Determine the angle of rotation uA at the free end A.
(Obtain the solution by using the modified form of
Castigliano’s theorem.)
q
C
A
B
L
—
2
L
—
2
C
D
E
I
2I
L
—
4
L
—
4
L
—
4
L
—
4
PROB. 9.9-9
9.9-10 An overhanging beam ABC is subjected to a couple
MA at the free end (see figure). The lengths of the overhang
and the main span are a and L, respectively.
Determine the angle of rotation uA and deflection dA at
end A. (Obtain the solution by using the modified form of
Castigliano’s theorem.)
MA
A
B
C
PROB. 9.9-7
9.9-8 The frame ABC supports a concentrated load P at
point C (see figure). Members AB and BC have lengths
h and b, respectively.
Determine the vertical deflection dC and angle of rotation uC at end C of the frame. (Obtain the solution by using
the modified form of Castigliano’s theorem.)
B
b
C
P
a
L
PROB. 9.9-10
9.9-11 An overhanging beam ABC rests on a simple
support at A and a spring support at B (see figure). A concentrated load P acts at the end of the overhang. Span AB
has length L, the overhang has length a, and the spring has
stiffness k.
Determine the downward displacement dC of the end
of the overhang. (Obtain the solution by using the modified
form of Castigliano’s theorem.)
P
B
A
h
C
k
A
PROB. 9.9-8
L
PROB. 9.9-11
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
a
CHAPTER 9 Problems
★9.9-12 A symmetric beam ABCD with overhangs at both
ends supports a uniform load of intensity q (see figure).
Determine the deflection dD at the end of the overhang. (Obtain the solution by using the modified form of
Castigliano’s theorem.)
W
h
q
A
A
B
L
—
4
L
—
4
L
B
D
C
703
L
—
2
L
—
2
PROB. 9.10-2
PROB. 9.9-12
Deflections Produced by Impact
The beams described in the problems for Section 9.10 have
constant flexural rigidity EI. Disregard the weights of the
beams themselves, and consider only the effects of the given
loads.
9.10-3 A cantilever beam AB of length L 6 ft is
constructed of a W 8 21 wide-flange section (see figure).
A weight W 1500 lb falls through a height h 0.25 in.
onto the end of the beam.
Calculate the maximum deflection dmax of the end of
the beam and the maximum bending stress smax due to the
falling weight. (Assume E 30 106 psi.)
W = 1500 lb
9.10-1 A heavy object of weight W is dropped onto the
midpoint of a simple beam AB from a height h (see figure).
Obtain a formula for the maximum bending stress
smax due to the falling weight in terms of h, sst, and dst,
where sst is the maximum bending stress and dst is the
deflection at the midpoint when the weight W acts on the
beam as a statically applied load.
Plot a graph of the ratio smax /sst (that is, the ratio of
the dynamic stress to the static stress) versus the ratio h /dst.
(Let h /dst vary from 0 to 10.)
W
h
A
B
L
—
2
L
—
2
PROB. 9.10-1
W8
21
h = 0.25 in.
A
B
L = 6 ft
PROB. 9.10-3
9.10-4 A weight W 20 kN falls through a height h
1.0 mm onto the midpoint of a simple beam of length L
3 m (see figure). The beam is made of wood with square
cross section (dimension d on each side) and E 12 GPa.
If the allowable bending stress in the wood is sallow
10 MPa, what is the minimum required dimension d?
W
h
A
9.10-2 An object of weight W is dropped onto the midpoint
of a simple beam AB from a height h (see figure). The beam
has a rectangular cross section of area A.
Assuming that h is very large compared to the deflection of the beam when the weight W is applied statically,
obtain a formula for the maximum bending stress smax in
the beam due to the falling weight.
d
B
d
L
—
2
L
—
2
PROB. 9.10-4
9.10-5 A weight W 4000 lb falls through a height
h 0.5 in. onto the midpoint of a simple beam of length
L 10 ft (see figure on the next page).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
704
CHAPTER 9 Deflections of Beams
Assuming that the allowable bending stress in the
beam is sallow 18,000 psi and E 30 106 psi, select
the lightest wide-flange beam listed in Table E-1 in Appendix E that will be satisfactory.
W = 4000 lb
h = 0.5 in.
A
Representation of Loads on Beams by
Discontinuity Functions
9.11-1 through 9.11-12 A beam and its loading are shown
in the figure. Using discontinuity functions, write the
expression for the intensity q(x) of the equivalent distributed load acting on the beam (include the reactions in the
expression for the equivalent load).
y
B
P
D
A
L
— = 5 ft
2
B
x
L
— = 5 ft
2
a
PROB. 9.10-5
b
L
9.10-6 An overhanging beam ABC of rectangular cross
section has the dimensions shown in the figure. A weight
W 750 N drops onto end C of the beam.
If the allowable normal stress in bending is 45 MPa,
what is the maximum height h from which the weight may
be dropped? (Assume E 12 GPa.)
PROBS. 9.11-1 and 9.12-1
y
q
D
A
B
x
40 mm
A
W
h
C
B
1.2 m
40 mm
500 mm
2.4 m
b
a
L
PROBS. 9.11-2 and 9.12-2
y
PROB. 9.10-6
q = 2 k/ft
P=4k
9.10-7 A heavy flywheel rotates at an angular speed v
★
(radians per second) around an axle (see figure). The axle is
rigidly attached to the end of a simply supported beam of
flexural rigidity EI and length L (see figure). The flywheel
has mass moment of inertia Im about its axis of rotation.
If the flywheel suddenly freezes to the axle, what will
be the reaction R at support A of the beam?
A
R
PROB. 9.10-7
A
6 ft
Im
3 ft
PROBS. 9.11-3 and 9.12-3
y
v
EI
B x
D
P
A
D
B
x
L
a
b
L
PROBS. 9.11-4 and 9.12-4
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
705
CHAPTER 9 Problems
y
M0
y
A
D
x
a
C
A
B
b
10 m
PROBS. 9.11-5 and 9.12-5
A
D
x
L
y
P = 120 kN
q = 20 kN/m
B
5m
5m
PROBS. 9.11-10 and 9.12-10
P
P
D
E
B
y
x
B
A
a
P=8k
M0 = 12 k-ft
D
x
C
a
L
6 ft
6 ft
6 ft
PROBS. 9.11-6 and 9.12-6
PROBS. 9.11-11 and 9.12-11
y
M0 = 20 k-ft
y
P = 18 k
D
A
B
x
16 ft
10 ft
B
1.2 m
x
1.2 m
Beam Deflections Using Discontinuity Functions
q
D
A
B
x
a
L
deflection curve for the cantilever beam ADB shown in the
figure. Also, obtain the angle of rotation uB and deflection
dB at the free end. (For the beam of Problem 9.12-3,
assume E 10 103 ksi and I 450 in.4)
q0
y
D
E
9.12-4, 9.12-5, and 9.12-6 Determine the equation of the
B
x
L/3
The problems for Section 9.12 are to be solved by using
discontinuity functions. All beams have constant flexural
rigidity EI. (Obtain the equations for the equivalent
distributed loads from the corresponding problems in
Section 9.11.)
9.12-1, 9.12-2, and 9.12-3 Determine the equation of the
PROBS. 9.11-8 and 9.12-8
PROBS. 9.11-9 and 9.12-9
D
PROBS. 9.11-12 and 9.12-12
y
L/3
C
A
1.2 m
PROBS. 9.11-7 and 9.12-7
A
q = 12 kN/m
L/3
deflection curve for the simple beam AB shown in the figure. Also, obtain the angle of rotation uA at the left-hand
support and the deflection dD at point D.
9.12-7 Determine the equation of the deflection curve for
the simple beam ADB shown in the figure. Also, obtain the
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
706
CHAPTER 9 Deflections of Beams
angle of rotation uA at the left-hand support and the
106 psi and
deflection dD at point D. Assume E 30
I 720 in.4
Determine the equation of the deflection curve of the
beam, the angle of rotation uB at end B, and the deflection
dB at end B.
y
9.12-8, 9.12-9, and 9.12-10 Obtain the equation of the
deflection curve for the simple beam AB (see figure). Also,
determine the angle of rotation uB at the right-hand support
and the deflection dD at point D. (For the beam of Problem
9.12-10, assume E 200 GPa and I 2.60 109 mm4.)
9.12-12 The overhanging beam ACBD shown in the figure
is simply supported at A and B. Obtain the equation of the
deflection curve and the deflections dC and dD at points
C and D, respectively. (Assume E 200 GPa and
I 15 106 mm4.)
PROB. 9.13-2
9.13-3 An overhanging beam ABC of height h is heated
to a temperature T1 on the top and T2 on the bottom (see
figure).
Determine the equation of the deflection curve of the
beam, the angle of rotation uC at end C, and the deflection
dC at end C.
y
goes a temperature change such that the bottom of the
beam is at temperature T2 and the top of the beam is at
temperature T1 (see figure).
Determine the equation of the deflection curve of the
beam, the angle of rotation uA at the left-hand support, and
the deflection dmax at the midpoint.
y
A
h
T1
T1
B
T2
T2
L
a
C
x
PROB. 9.13-3
9.13-4 A simple beam AB of length L and height h (see figure) is heated in such a manner that the temperature
difference T2 T1 between the bottom and top of the beam
is proportional to the distance from support A; that is,
T2 T1 T0 x
in which T0 is a constant having units of temperature
(degrees) per unit distance.
Determine the maximum deflection dmax of the beam.
B
x
T2
h
T1
The beams described in the problems for Section 9.13
have constant flexural rigidity EI. In every problem, the
temperature varies linearly between the top and bottom
of the beam.
9.13-1 A simple beam AB of length L and height h under-
x
L
A
Temperature Effects
B
T2
9.12-11 A beam ACBD with simple supports at A and B
and an overhang BD is shown in the figure. (a) Obtain the
equation of the deflection curve for the beam. (b) Calculate
the deflections dC and dD at points C and D, respectively.
(Assume E 30 106 psi and I 280 in.4)
h
T1
A
y
h
T1
A
L
x
T2
x
PROB. 9.13-1
L
9.13-2 A cantilever beam AB of length L and height h (see
figure) is subjected to a temperature change such that the
temperature at the top is T1 and at the bottom is T2.
B
PROB. 9.13-4
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
10
Statically Indeterminate
Beams
10.1 INTRODUCTION
In this chapter we will analyze beams in which the number of reactions
exceeds the number of independent equations of equilibrium. Since the
reactions of such beams cannot be determined by statics alone, the
beams are said to be statically indeterminate.
The analysis of statically indeterminate beams is quite different
from that of statically determinate beams. When a beam is statically
determinate, we can obtain all reactions, shear forces, and bending
moments from free-body diagrams and equations of equilibrium. Then,
knowing the shear forces and bending moments, we can obtain the
stresses and deflections.
However, when a beam is statically indeterminate, the equilibrium
equations are not sufficient and additional equations are needed. The
most fundamental method for analyzing a statically indeterminate beam
is to solve the differential equations of the deflection curve, as described
later in Section 10.3. Although this method serves as a good starting
point in our analysis, it is practical for only the simplest types of statically indeterminate beams.
Therefore, we also discuss the method of superposition (Section
10.4), a method that is applicable to a wide variety of structures. In the
method of superposition, we supplement the equilibrium equations with
compatibility equations and force-displacement equations. (This same
method was described earlier in Section 2.4, where we analyzed statically indeterminate bars subjected to tension and compression.)
In the last part of this chapter we discuss two specialized topics
pertaining to statically indeterminate beams, namely, beams with temperature changes (Section 10.5), and longitudinal displacements at the
ends of beams (Section 10.6). Throughout this chapter, we assume that
the beams are made of linearly elastic materials.
707
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
708
CHAPTER 10 Statically Indeterminate Beams
Although only statically indeterminate beams are discussed in this
chapter, the fundamental ideas have much wider application. Most of the
structures we encounter in everyday life, including automobile frames,
buildings, and aircraft, are statically indeterminate. However, they are
much more complex than beams and must be designed by very sophisticated analytical techniques. Many of these techniques rely on the
concepts described in this chapter, and therefore this chapter may be
viewed as an introduction to the analysis of statically indeterminate
structures of all kinds.
10.2 TYPES OF STATICALLY INDETERMINATE BEAMS
Statically indeterminate beams are usually identified by the arrangement
of their supports. For instance, a beam that is fixed at one end and
simply supported at the other (Fig. 10-1a) is called a propped
cantilever beam. The reactions of the beam shown in the figure consist
of horizontal and vertical forces at support A, a moment at support A,
and a vertical force at support B. Because there are only three independent equations of equilibrium for this beam, it is not possible to calculate
all four of the reactions from equilibrium alone. The number of reactions
in excess of the number of equilibrium equations is called the degree of
static indeterminacy. Thus, a propped cantilever beam is statically
indeterminate to the first degree.
The excess reactions are called static redundants and must be
selected in each particular case. For example, the reaction RB of the
propped cantilever beam shown in Fig. 10-1a may be selected as the
redundant reaction. Since this reaction is in excess of those needed to
maintain equilibrium, it can be released from the structure by removing
the support at B. When support B is removed, we are left with a
FIG. 10-1 Propped cantilever beam:
(a) beam with load and reactions;
(b) released structure when the reaction
at end B is selected as the redundant; and
(c) released structure when the moment
reaction at end A is selected as the
redundant
P
HA
MA
P
P
A
B
(b)
RA
(c)
(a)
RB
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
709
SECTION 10.2 Types of Statically Indeterminate Beams
cantilever beam (Fig. 10-1b). The structure that remains when the redundants are released is called the released structure or the primary
structure. The released structure must be stable (so that it is capable of
carrying loads), and it must be statically determinate (so that all force
quantities can be determined by equilibrium alone).
Another possibility for the analysis of the propped cantilever beam
of Fig. 10-1a is to select the reactive moment MA as the redundant.
Then, when the moment restraint at support A is removed, the released
structure is a simple beam with a pin support at one end and a roller support at the other (Fig. 10-1c).
A special case arises if all loads acting on the beam are vertical (Fig.
10-2). Then the horizontal reaction at support A vanishes, and three reactions remain. However, only two independent equations of equilibrium
are now available, and therefore the beam is still statically indeterminate
to the first degree. If the reaction RB is chosen as the redundant, the
released structure is a cantilever beam; if the moment MA is chosen, the
released structure is a simple beam.
P2
P1
B
A
MA
RA
FIG. 10-2 Propped cantilever beam with
RB
vertical loads only
FIG. 10-3 Fixed-end beam: (a) beam with
load and reactions; (b) released structure
when the three reactions at end B are
selected as the redundants; and
(c) released structure when the two
moment reactions and the horizontal
reaction at end B are selected as the
redundants
Another type of statically indeterminate beam, known as a fixedend beam, is shown in Fig. 10-3a. This beam has fixed supports at both
ends, resulting in a total of six unknown reactions (two forces and a
moment at each support). Because there are only three equations of
equilibrium, the beam is statically indeterminate to the third degree.
(Other names for this type of beam are clamped beam and built-in
beam.)
P
HA
MA
P
B
A
RA
RB
P
HB
MB
(b)
(a)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(c)
710
CHAPTER 10 Statically Indeterminate Beams
P1
A
FIG. 10-4 Fixed-end beam with vertical
loads only
MA
RA
P2
B
RB
MB
If we select the three reactions at end B of the beam as the redundants, and if we remove the corresponding restraints, we are left with a
cantilever beam as the released structure (Fig. 10-3b). If we release the
two fixed-end moments and one horizontal reaction, the released structure is a simple beam (Fig. 10-3c).
Again considering the special case of vertical loads only (Fig. 10-4),
we find that the fixed-end beam now has only four nonzero reactions
(one force and one moment at each support). The number of available
equilibrium equations is two, and therefore the beam is statically
indeterminate to the second degree. If the two reactions at end B are
selected as the redundants, the released structure is a cantilever beam; if
the two moment reactions are selected, the released structure is a simple
beam.
The beam shown in Fig. 10-5a is an example of a continuous
beam, so called because it has more than one span and is continuous
over an interior support. This particular beam is statically indeterminate
to the first degree because there are four reactive forces and only three
equations of equilibrium.
If the reaction RB at the interior support is selected as the redundant,
and if we remove the corresponding support from the beam, then there
remains a released structure in the form of a statically determinate
simple beam (Fig. 10-5b). If the reaction RC is selected as the redundant,
the released structure is a simple beam with an overhang (Fig. 10-5c).
In the following sections, we will discuss two methods for analyzing statically indeterminate beams. The objective in each case is to
determine the redundant reactions. Once they are known, all remaining
reactions (plus the shear forces and bending moments) can be found
from equations of equilibrium. In effect, the structure has become statically determinate. Therefore, as the final step in the analysis, the stresses
and deflections can be found by the methods described in preceding
chapters.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 10.3 Analysis by the Differential Equations of the Deflection Curve
711
FIG. 10-5 Example of a continuous beam:
(a) beam with loads and reactions;
(b) released structure when the reaction
at support B is selected as the redundant;
and (c) released structure when the
reaction at end C is selected as the
redundant
P1
P2
B
A
P1
P2
P1
P2
C
HA
RA
RB
RC
(b)
(c)
(a)
10.3 ANALYSIS BY THE DIFFERENTIAL EQUATIONS OF THE DEFLECTION CURVE
Statically indeterminate beams may be analyzed by solving any one of
the three differential equations of the deflection curve: (1) the
second-order equation in terms of the bending moment (Eq. 9-12a),
(2) the third-order equation in terms of the shear force (Eq. 9-12b), or
(3) the fourth-order equation in terms of the intensity of distributed load
(Eq. 9-12c).
The procedure is essentially the same as that for a statically determinate beam (see Sections 9.2, 9.3, and 9.4) and consists of writing the
differential equation, integrating to obtain its general solution, and then
applying boundary and other conditions to evaluate the unknown quantities. The unknowns consist of the redundant reactions as well as the
constants of integration.
The differential equation for a beam may be solved in symbolic
terms only when the beam and its loading are relatively simple and
uncomplicated. The resulting solutions are in the form of general purpose formulas. However, in more complex situations the differential
equations must be solved numerically, using computer programs
intended for that purpose. In such cases the results apply only to specific
numerical problems.
The following examples illustrate the analysis of statically indeterminate beams by solving the differential equations in symbolic terms.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
712
CHAPTER 10 Statically Indeterminate Beams
Example 10-1
y
q
MA
x
B
A
A propped cantilever beam AB of length L supports a uniform load of intensity q
(Fig. 10-6). Analyze this beam by solving the second-order differential equation
of the deflection curve (the bending-moment equation). Determine the reactions,
shear forces, bending moments, slopes, and deflections of the beam.
Solution
RA
L
RB
FIG. 10-6 Example 10-1. Propped
cantilever beam with a uniform load
Because the load on this beam acts in the vertical direction (Fig. 10-6), we
conclude that there is no horizontal reaction at the fixed support. Therefore, the
beam has three unknown reactions (MA, RA, and RB). Only two equations of
equilibrium are available for determining these reactions, and therefore the
beam is statically indeterminate to the first degree.
Since we will be analyzing this beam by solving the bending-moment
equation, we must begin with a general expression for the moment. This expression will be in terms of both the load and the selected redundant.
Redundant reaction. Let us choose the reaction RB at the simple support as
the redundant. Then, by considering the equilibrium of the entire beam, we can
express the other two reactions in terms of RB:
RA qL ⫺ RB
qL2
MA ⫺ RB L
2
(a,b)
Bending moment. The bending moment M at distance x from the fixed
support can be expressed in terms of the reactions as follows:
qx 2
M RA x ⫺ MA ⫺
2
(c)
This equation can be obtained by the customary technique of constructing a
free-body diagram of part of the beam and solving an equation of equilibrium.
Substituting into Eq. (c) from Eqs. (a) and (b), we obtain the bending
moment in terms of the load and the redundant reaction:
qL2
qx2
M qL x ⫺ RB x ⫺ RB L
2
2
(d)
Differential equation. The second-order differential equation of the deflection curve (Eq. 9-12a) now becomes
qL2
qx2
EIv M qLx RB x RBL
2
2
(e)
After two successive integrations, we obtain the following equations for the
slopes and deflections of the beam:
qL x 2
RB x 2
qL2x
q x3
EIv RB L x C1
2
2
2
6
(f )
qL x 3
RB x 3
qL2x 2
RB Lx 2
qx4
EIv C1 x C2
6
6
4
2
24
(g)
These equations contain three unknown quantities (C1, C2, and RB).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
713
SECTION 10.3 Analysis by the Differential Equations of the Deflection Curve
Boundary conditions. Three boundary conditions pertaining to the deflections and slopes of the beam are apparent from an inspection of Fig. 10-6. These
conditions are as follows: (1) the deflection at the fixed support is zero, (2) the
slope at the fixed support is zero, and (3) the deflection at the simple support is
zero. Thus,
v(0) 0
v(0) 0
v(L) 0
Applying these conditions to the equations for slopes and deflections (Eqs.
f and g), we find C1 0, C2 0, and
3qL
RB
8
Thus, the redundant reaction RB is now known.
Reactions. With the value of the redundant established, we can find the
remaining reactions from Eqs. (a) and (b). The results are
y
q
B
A
qL2
5qL
—
8
—
8
5qL
—
8
0
5L
—
8
M
3qL
—
8
9qL2
—
128
L
—
4
0
x
3qL
—
8
L
V
(10-1)
5qL
RA
8
qL2
—
8
FIG. 10-7 Shear-force and bendingmoment diagrams for the propped
cantilever beam of Fig. 10-6
(10-2a,b)
Knowing these reactions, we can find the shear forces and bending moments in
the beam.
Shear forces and bending moments. These quantities can be obtained by
the usual techniques involving free-body diagrams and equations of equilibrium.
The results are
5qL
V RA qx qx
8
(10-3)
qx2
5qL x
qL2
qx2
M RA x MA
2
8
8
2
(10-4)
Shear-force and bending-moment diagrams for the beam can be drawn with the
aid of these equations (see Fig. 10-7).
From the diagrams, we see that the maximum shear force occurs at the
fixed support and is equal to
5qL
Vmax
8
(10-5)
Also, the maximum positive and negative bending moments are
9qL2
Mpos
128
5L
—
8
qL2
MA
8
qL2
Mneg
8
(10-6a,b)
Finally, we note that the bending moment is equal to zero at distance x L /4
from the fixed support.
Slopes and deflections of the beam. Returning to Eqs. (f) and (g) for the
slopes and deflections, we now substitute the values of the constants of integration (C1 0 and C2 0) as well as the expression for the redundant RB
(Eq. 10-1) and obtain
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
714
CHAPTER 10 Statically Indeterminate Beams
y
d0
A
L
—
4
dmax uB
B
x
qx
v (6L2 15Lx 8x 2)
48EI
(10-7)
q x2
v (3L2 5Lx 2x 2)
48EI
(10-8)
The deflected shape of the beam as obtained from Eq. (10-8) is shown in
Fig. 10-8.
To determine the maximum deflection of the beam, we set the slope
(Eq. 10-7) equal to zero and solve for the distance x1 to the point where this
deflection occurs:
v 0 or 6L2 15Lx 8 x2 0
x1
from which
FIG. 10-8 Deflection curve for the
propped cantilever beam of Fig. 10-6
15 兹33
苶
x1 L 0.5785L
16
(10-9)
Substituting this value of x into the equation for the deflection (Eq. 10-8) and
also changing the sign, we get the maximum deflection:
qL4
dmax (v)xx1 (39 55兹33
苶)
65,536EI
qL4
qL4
0.005416
184.6EI
EI
(10-10)
The point of inflection is located where the bending moment is equal to
zero, that is, where x L /4. The corresponding deflection d 0 of the beam (from
Eq. 10-8) is
5qL 4
qL4
d 0 (v)xL /4 0.002441
2048EI
EI
(10-11)
Note that when x L /4, both the curvature and the bending moment are
negative, and when x L /4, the curvature and bending moment are positive.
To determine the angle of rotation uB at the simply supported end of the
beam, we use Eq. (10-7), as follows:
qL3
uB (v)xL
48EI
(10-12)
Slopes and deflections at other points along the axis of the beam can be
obtained by similar procedures.
Note: In this example, we analyzed the beam by taking the reaction
RB (Fig. 10-6) as the redundant reaction. An alternative approach is to take
the reactive moment MA as the redundant. Then we can express the bending
moment M in terms of MA, substitute the resulting expression into the
second-order differential equation, and solve as before. Still another approach is
to begin with the fourth-order differential equation, as illustrated in the next
example.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
715
SECTION 10.3 Analysis by the Differential Equations of the Deflection Curve
Example 10-2
MA
RA
y
P
A
C
B
x
The fixed-end beam ACB shown in Fig. 10-9 supports a concentrated load P at
the midpoint. Analyze this beam by solving the fourth-order differential
equation of the deflection curve (the load equation). Determine the reactions,
shear forces, bending moments, slopes, and deflections of the beam.
MB
L
2
L
2
RB
FIG. 10-9 Example 10-2. Fixed-end beam
with a concentrated load at the midpoint
Solution
Because the load on this beam acts only in the vertical direction, we know
that there are no horizontal reactions at the supports. Therefore, the beam has
four unknown reactions, two at each support. Since only two equations of equilibrium are available, the beam is statically indeterminate to the second degree.
However, we can simplify the analysis by observing from the symmetry of
the beam and its loading that the forces and moments at supports A and B are
equal, that is,
RA RB and MA MB
Since the vertical reactions at the supports are equal, we know from equilibrium
of forces in the vertical direction that each force is equal to P/2:
P
RA RB
2
(10-13)
Thus, the only unknown quantities that remain are the moment reactions MA and
MB. For convenience, we will select the moment MA as the redundant quantity.
Differential equation. Because there is no load acting on the beam between
points A and C, the fourth-order differential equation (Eq. 9-12c) for the lefthand half of the beam is
EIv q 0
(0 x L /2)
(h)
Successive integrations of this equation yield the following equations, which are
valid for the left-hand half of the beam:
EIv C1
(i)
EIv C1x C2
( j)
2
C1 x
EIv C2 x C3
2
(k)
C1 x 3
C2 x 2
EIv C3 x C4
6
2
(l)
These equations contain four unknown constants of integration. Since we
now have five unknowns (C1, C2, C3, C4, and MA), we need five boundary
conditions.
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
716
CHAPTER 10 Statically Indeterminate Beams
Boundary conditions. The boundary conditions applicable to the left-hand
half of the beam are as follows:
(1) The shear force in the left-hand segment of the beam is equal to RA, or
P/2. Therefore, from Eq. (9-12b) we find
P
EIv V
2
Combining this equation with Eq. (i), we obtain C1 P/2.
(2) The bending moment at the left-hand support is equal to MA. Therefore, from Eq. (9-12a) we get
EIv M MA at x 0
PL
—
8
y
P
A
C
L
—
2
P
—
2
Combining this equation with Eq. ( j), we obtain C2 MA.
(3) The slope of the beam at the left-hand support (x 0) is equal to zero.
Therefore, Eq. (k) yields C3 0.
(4) The slope of the beam at the midpoint (x L /2) is also equal to zero
(from symmetry). Therefore, from Eq. (k) we find
B
L
—
2
P
—
2
P
—
2
V
0
PL
—
8
(10-14)
Thus, the reactive moments at the ends of the beam have been determined.
(5) The deflection of the beam at the left-hand support (x 0) is equal to
zero. Therefore, from Eq. (l) we find C4 0.
In summary, the four constants of integration are
PL
C2 MA
8
C3 0
C4 0
(m,n,o,p)
Shear forces and bending moments. The shear forces and bending moments
can be found by substituting the appropriate constants of integration into Eqs.
(i) and (j). The results are
P
—
2
L
—
4
PL
—
8
P
C1
2
0
M
PL
MA MB
8
x
PL
—
8
P
EIv V
2
L
—
4
Px
PL
EIv M
2
8
PL
—
8
FIG. 10-10 Shear-force and bending-
moment diagrams for the fixed-end
beam of Fig. 10-9
(0 x L /2)
(0
x
L /2)
(10-15)
(10-16)
Since we know the reactions of the beam, we can also obtain these expressions
directly from free-body diagrams and equations of equilibrium.
The shear-force and bending moment diagrams are shown in Fig. 10-10.
Slopes and deflections. The slopes and deflections in the left-hand half of
the beam can be found from Eqs. (k) and (l) by substituting the expressions for
the constants of integration. In this manner, we find
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
717
SECTION 10.3 Analysis by the Differential Equations of the Deflection Curve
Px
v (L 2x)
8 EI
Px 2
v (3L 4x)
48EI
(0
(0
L /2)
x
(10-17)
L /2)
x
(10-18)
The deflection curve of the beam is shown in Fig. 10-11.
To find the maximum deflection dmax we set x equal to L /2 in Eq. (10-18)
and change the sign; thus,
PL3
dmax (v)xL /2
192EI
(10-19)
The point of inflection in the left-hand half of the beam occurs where the
bending moment M is equal to zero, that is, where x L /4 (see Eq. 10-16). The
corresponding deflection d 0 (from Eq. 10-18) is
PL3
d 0 (v)xL /4
384EI
(10-20)
which is equal numerically to one-half of the maximum deflection. A second
point of inflection occurs in the right-hand half of the beam at distance L /4 from
end B.
Notes: As we observed in this example, the number of boundary and other
conditions is always sufficient to evaluate not only the constants of integration
but also the redundant reactions.
Sometimes it is necessary to set up differential equations for more than one
region of the beam and use conditions of continuity between regions, as illustrated in Examples 9-3 and 9-5 of Chapter 9 for statically determinate beams.
Such analyses are likely to be long and tedious because of the large number of
conditions that must be satisfied. However, if deflections and angles of rotation
are needed at only one or two specific points, the method of superposition may
be useful (see the next section).
y
d0
A
FIG. 10-11 Deflection curve for the fixed-
end beam of Fig. 10-9
L
—
4
C
dmax
L
—
2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
d0
L
—
2
B
L
—
4
x
718
CHAPTER 10 Statically Indeterminate Beams
10.4 METHOD OF SUPERPOSITION
q
A
B
MA
RA
RB
L
(a)
B
A
(b)
q
(dB)1
(c)
(dB)2
RB
(d)
FIG. 10-12 Analysis of a propped
cantilever beam by the method of
superposition with the reaction RB
selected as the redundant
The method of superposition is of fundamental importance in the analysis of statically indeterminate bars, trusses, beams, frames, and many
other kinds of structures. We have already used the superposition
method to analyze statically indeterminate structures composed of bars
in tension and compression (Section 2.4) and shafts in torsion (Section
3.8). In this section, we will apply the method to beams.
We begin the analysis by noting the degree of static indeterminacy
and selecting the redundant reactions. Then, having identified the redundants, we can write equations of equilibrium that relate the other
unknown reactions to the redundants and the loads.
Next, we assume that both the original loads and the redundants act
upon the released structure. Then we find the deflections in the released
structure by superposing the separate deflections due to the loads and
the redundants. The sum of these deflections must match the deflections
in the original beam. However, the deflections in the original beam (at
the points where restraints were removed) are either zero or have known
values. Therefore, we can write equations of compatibility (or equations of superposition) expressing the fact that the deflections of the
released structure (at the points where restraints were removed) are the
same as the deflections in the original beam (at those same points).
Since the released structure is statically determinate, we can easily
determine its deflections by using the techniques described in Chapter 9.
The relationships between the loads and the deflections of the released
structure are called force-displacement relations. When these relations
are substituted into the equations of compatibility, we obtain equations
in which the redundants are the unknown quantities. Therefore, we can
solve those equations for the redundant reactions. Then, with the redundants known, we can determine all other reactions from the equations of
equilibrium. Furthermore, we can also determine the shear forces and
bending moments from equilibrium.
The steps described in general terms in the preceding paragraphs
can be made clearer by considering a particular case, namely, a propped
cantilever beam supporting a uniform load (Fig. 10-12a). We will make
two analyses, the first with the force reaction RB selected as the redundant and the second with the moment reaction MA as the redundant.
(This same beam was analyzed in Example 10-1 of Section 10.3 by
solving the differential equation of the deflection curve.)
Analysis with RB as Redundant
In this first illustration we select the reaction RB at the simple support
(Fig. 10-12a) as the redundant. Then the equations of equilibrium that
express the other unknown reactions in terms of the redundant are as follows:
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 10.4 Method of Superposition
719
qL2
MA RB L
(a,b)
2
These equations are obtained from equations of equilibrium that apply
to the entire beam taken as a free body (Fig. 10-12a).
The next step is to remove the restraint corresponding to the redundant (in this case, we remove the support at end B). The released
structure that remains is a cantilever beam (Fig. 10-12b). The uniform
load q and the redundant force RB are now applied as loads on the
released structure (Figs. 10-12c and d).
The deflection at end B of the released structure due solely to the
uniform load is denoted (dB)1, and the deflection at the same point due
solely to the redundant is denoted (dB)2. The deflection dB at point B in
the original structure is obtained by superposing these two deflections.
Since the deflection in the original beam is equal to zero, we obtain the
following equation of compatibility:
RA qL RB
dB (dB)1 (dB)2 0
(c)
The minus sign appears in this equation because (dB)1 is positive downward whereas (dB)2 is positive upward.
The force-displacement relations that give the deflections (dB)1 and
(dB)2 in terms of the uniform load q and the redundant RB, respectively,
are found with the aid of Table G-1 in Appendix G (see Cases 1 and 4).
Using the formulas given there, we obtain
qL 4
RBL3
(dB)2
(d,e)
(dB)1
8EI
3 EI
Substituting these force-displacement relations into the equation of compatibility yields
qL 4
RB L3
(f )
dB 0
8EI
3 EI
which can be solved for the redundant reaction:
3qL
RB
8
(10-21)
Note that this equation gives the redundant in terms of the loads acting
on the original beam.
The remaining reactions (RA and MA) can be found from the equilibrium equations (Eqs. a and b); the results are
5qL
RA
8
qL2
MA
8
(10-22a,b)
Knowing all reactions, we can now obtain the shear forces and bending
moments throughout the beam and plot the corresponding diagrams (see
Fig. 10-7 for these diagrams).
We can also determine the deflections and slopes of the original
beam by means of the principle of superposition. The procedure consists
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
720
CHAPTER 10 Statically Indeterminate Beams
of superposing the deflections of the released structure when acted upon
by the loads shown in Figs. 10-12c and d. For instance, the equations of
the deflection curves for those two loading systems are obtained from
Cases 1 and 4, respectively, of Table G-1, Appendix G:
qx 2
v1 (6L2 4Lx x 2)
24EI
RB x2
v2 (3L x)
6EI
Substituting for RB from Eq. (10-21) and then adding the deflections v1
and v2, we obtain the following equation for the deflection curve of the
original statically indeterminate beam (Fig. 10-12a):
qx 2
v v1 v2 (3L2 5Lx 2x 2)
48EI
This equation agrees with Eq. (10-8) of Example 10-1. Other deflection
quantitites can be found in an analogous manner.
Analysis with MA as Redundant
We will now analyze the same propped cantilever beam by selecting the
moment reaction MA as the redundant (Fig. 10-13). In this case, the
released structure is a simple beam (Fig. 10-13b). The equations of equilibrium for the reactions RA and RB in the original beam are
qL
MA
RA
2
L
qL
MA
RB
2
L
(g,h)
q
A
B
MA
A
RA
L
B
RB
(a)
(b)
q
(uA)2
FIG. 10-13 Analysis of a propped
cantilever beam by the method of
superposition with the moment reaction
MA selected as the redundant
MA
(uA)1
(c)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(d)
SECTION 10.4 Method of Superposition
721
The equation of compatibility expresses the fact that the angle of rotation uA at the fixed end of the original beam is equal to zero. Since this
angle is obtained by superposing the angles of rotation (uA)1 and (uA)2 in
the released structure (Figs. 10-13c and d), the compatibility equation
becomes
uA (uA)1 (uA)2 0
(i)
In this equation, the angle (uA)1 is assumed to be positive when
clockwise and the angle (uA)2 is assumed to be positive when counterclockwise.
The angles of rotation in the released structure are obtained from the
formulas given in Table G-2 of Appendix G (see Cases 1 and 7). Thus,
the force-displacement relations are
qL3
(uA)1
24EI
MAL
(uA)2
3EI
Substituting into the compatibility equation (Eq. i), we get
qL3
MAL
uA 0
24EI
3EI
( j)
Solving this equation for the redundant, we get MA qL2/8, which
agrees with the previous result (Eq. 10-22b). Also, the equations of equilibrium (Eqs. g and h) yield the same results as before for the reactions
RA and RB (see Eqs. 10-22a and 10-21, respectively).
Now that all reactions have been found, we can determine the shear
forces, bending moments, slopes, and deflections by the techniques
already described.
General Comments
The method of superposition described in this section is also called the
flexibility method or the force method. The latter name arises from the
use of force quantities (forces and moments) as the redundants; the
former name is used because the coefficients of the unknown quantities
in the compatibility equation (terms such as L3/3EI in Eq. f and L /3EI in
Eq. j) are flexibilities (that is, deflections or angles produced by a unit
load).
Since the method of superposition involves the superposition of
deflections, it is applicable only to linearly elastic structures. (Recall
that this same limitation applies to all topics discussed in this chapter.)
In the following examples, and also in the problems at the end of the
chapter, we are concerned primarily with finding the reactions, since this
is the key step in the solutions.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
722
CHAPTER 10 Statically Indeterminate Beams
Example 10-3
A two-span continuous beam ABC supports a uniform load of intensity q, as
shown in Fig. 10-14a. Each span of the beam has length L. Using the method of
superposition, determine all reactions for this beam.
q
A
A
RA
L
B
C
RB
RC
L
C
B
(b)
(a)
q
FIG. 10-14 Example 10-3. Two-span
continuous beam with a uniform load
(c)
RB
(dB)1
(dB)2
(d)
Solution
This beam has three unknown reactions (RA, RB, and RC). Since there are
two equations of equilibrium for the beam as a whole, it is statically indeterminate to the first degree. For convenience, let us select the reaction RB at the
middle support as the redundant.
Equations of equilibrium. We can express the reactions RA and RC in terms
of the redundant RB by means of two equations of equilibrium. The first equation, which is for equilibrium of moments about point B, shows that RA and RC
are equal. The second equation, which is for equilibrium in the vertical direction, yields the following result:
RB
RA RC qL
2
(k)
Equation of compatibility. Because the reaction RB is selected as the
redundant, the released structure is a simple beam with supports at A and C
(Fig. 10-14b). The deflections at point B in the released structure due to the uniform load q and the redundant RB are shown in Figs. 10-14c and d, respectively.
Note that the deflections are denoted (dB)1 and (dB)2. The superposition of these
deflections must produce the deflection dB in the original beam at point B. Since
the latter deflection is equal to zero, the equation of compatibility is
dB (dB)1 (dB) 2 0
(l)
in which the deflection (dB)1 is positive downward and the deflection (dB)2 is
positive upward.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 10.4 Method of Superposition
723
Force-displacement relations. The deflection (dB)1 caused by the uniform
load acting on the released structure (Fig. 10-14c) is obtained from Table G-2,
Case 1, as follows:
5q(2L)4
5qL 4
(dB)1
384EI
24EI
where 2L is the length of the released structure. The deflection (dB)2 produced
by the redundant (Fig. 10-14d) is
RB (2L)3
RB L3
(dB)2
48EI
6EI
as obtained from Table G-2, Case 4.
Reactions. The equation of compatibility pertaining to the vertical deflection at point B (Eq. l) now becomes
RB L3
5qL 4
dB 0
6EI
24EI
(m)
from which we find the reaction at the middle support:
5qL
RB
4
(10-23)
The other reactions are obtained from Eq. (k):
3qL
RA RC
8
(10-24)
With the reactions known, we can find the shear forces, bending moments,
stresses, and deflections without difficulty.
Note: The purpose of this example is to provide an illustration of the
method of superposition, and therefore we have described all steps in the analysis. However, this particular beam (Fig. 10-14a) can be analyzed by inspection
because of the symmetry of the beam and its loading.
From symmetry we know that the slope of the beam at the middle support
must be zero, and therefore each half of the beam is in the same condition as a
propped cantilever beam with a uniform load (see, for instance, Fig. 10-6). Consequently, all of our previous results for a propped cantilever beam with a
uniform load (Eqs. 10-1 to 10-12) can be adapted immediately to the continuous
beam of Fig. 10-14a.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
724
CHAPTER 10 Statically Indeterminate Beams
Example 10-4
P
B
D
A
MA
b
a
RA
A fixed-end beam AB (Fig. 10-15a) is loaded by a force P acting at an intermediate point D. Find the reactive forces and moments at the ends of the beam
using the method of superposition. Also, determine the deflection at point D
where the load is applied.
MB
RB
L
This beam has four unknown reactions (a force and a moment at each support), but only two independent equations of equilibrium are available.
Therefore, the beam is statically indeterminate to the second degree. In this
example, we will select the reactive moments MA and MB as the redundants.
Equations of equilibrium. The two unknown force reactions (RA and RB)
can be expressed in terms of the redundants (MA and MB) with the aid of two
equations of equilibrium. The first equation is for moments about point B, and
the second is for moments about point A. The resulting expressions are
(a)
P
(uA)1
(uB)1
MA
MB
Pb
R A
L
L
L
(b)
MA
(c)
MB
(uB)3
(uA)3
(d)
FIG. 10-15 Example 10-4. Fixed-end
beam with a concentrated load
MA
MB
Pa
RB
L
L
L
(n,o)
Equations of compatibility. When both redundants are released by removing the rotational restraints at the ends of the beam, we are left with a simple
beam as the released structure (Figs. 10-15b, c, and d). The angles of rotation at
the ends of the released structure due to the concentrated load P are denoted
(uA)1 and (uB)1, as shown in Fig. 10-15b. In a similar manner, the angles at the
ends due to the redundant MA are denoted (uA)2 and (uB)2, and the angles due to
the redundant MB are denoted (uA)3 and (uB)3.
Since the angles of rotation at the supports of the original beam are equal to
zero, the two equations of compatibility are
(uB)2
(uA)2
Solution
uA (uA)1 (uA)2 (uA)3 0
(p)
uB (uB)1 (uB)2 (uB)3 0
(q)
in which the signs of the various terms are determined by inspection from the
figures.
Force-displacement relations. The angles at the ends of the beam due to
the load P (Fig. 10-15b) are obtained from Case 5 of Table G-2:
Pab(L b)
(uA)1
6LEI
Pab(L a)
(uB)1
6LEI
in which a and b are the distances from the supports to point D where the load is
applied.
Also, the angles at the ends due to the redundant moment MA are (see
Case 7 of Table G-2):
MAL
(uA) 2
3EI
MAL
(uB) 2
6EI
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
725
SECTION 10.4 Method of Superposition
Similarly, the angles due to the moment MB are
MB L
(uA)3
6EI
MB L
(uB)3
3EI
Reactions. When the preceding expressions for the angles are substituted
into the equations of compatibility (Eqs. p and q), we arrive at two simultaneous
equations containing MA and MB as unknowns:
MAL
MB L
Pab(L b)
3EI
6EI
6LEI
(r)
MAL
MBL
Pab(L a)
6EI
3EI
6LEI
(s)
Solving these equations for the redundants, we obtain
Pab2
MA
L2
Pa2b
MB
L2
(10-25a,b)
Substituting these expressions for MA and MB into the equations of equilibrium
(Eqs. n and o), we obtain the vertical reactions:
Pb2
RA (L
2a)
L3
Pa2
RB (L
2b)
L3
(10-26a,b)
Thus, all reactions for the fixed-end beam have been determined.
The reactions at the supports of a beam with fixed ends are commonly
referred to as fixed-end moments and fixed-end forces. They are widely used
in structural analysis, and formulas for these quantities are listed in engineering
handbooks.
Deflection at point D. To obtain the deflection at point D in the original
fixed-end beam (Fig. 10-15a), we again use the principle of superposition. The
deflection at point D is equal to the sum of three deflections: (1) the downward
deflection (dD)1 at point D in the released structure due to the load P (Fig.
10-15b); (2) the upward deflection (dD)2 at the same point in the released structure due to the redundant MA (Fig. 10-15c), and (3) the upward deflection (dD)3
at the same point in the released structure due to the redundant MB (Fig.
10-15d). This superposition of deflections is expressed by the following
equation:
dD (dD)1 (dD)2 (dD)3
(t)
in which dD is the downward deflection in the original beam.
The deflections appearing in Eq. (t) can be obtained from the formulas
given in Table G-2 of Appendix G (see Cases 5 and 7) by making the appropriate substitutions and algebraic simplifications. The results of these
manipulations are as follows:
Pa2b2
(dD)1
3LEI
MAab
(dD) 2 (L b)
6LEI
MB ab
(dD) 3 (L a)
6LEI
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
726
CHAPTER 10 Statically Indeterminate Beams
Substituting the expressions for MA and MB from Eqs. (10-25a and b) into the
last two expressions, we get
Pa2b3
(dD)2 3 (L b)
6L EI
Pa3b2
(dD)3 3 (L a)
6L EI
Therefore, the deflection at point D in the original beam, obtained by substituting (dD)1, (dD)2, and (dD)3 into Eq. (t) and simplifying, is
Pa3b3
dD 3
3L EI
(10-27)
The method described in this example for finding the deflection dD can be used
not only to find deflections at individual points but also to find the equations of
the deflection curve.
Concentrated load acting at the midpoint of the beam. When the load P
acts at the midpoint C (Fig. 10-16), the reactions of the beam (from Eqs. 10-25
and 10-26 with a b L /2) are
PL
MA MB
8
P
RA RB
2
(10-28a,b)
Also, the deflection at the midpoint (from Eq. 10-27) is
PL3
dC
192EI
(10-29)
This deflection is only one-fourth of the deflection at the midpoint of a simple
beam with the same load, which shows the stiffening effect of clamping the
ends of the beam.
The preceding results for the reactions at the ends and the deflection at the
middle (Eqs. 10-28 and 10-29) agree with those found in Example 10-2 by solving the differential equation of the deflection curve (see Eqs. 10-13, 10-14, and
10-19).
P
A
MA
FIG. 10-16 Fixed-end beam with a
concentrated load acting at the midpoint
RA
C
L
—
2
B
L
—
2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
MB
RB
727
SECTION 10.4 Method of Superposition
Example 10-5
A fixed-end beam AB supports a uniform load of intensity q acting over part of
the span (Fig. 10-17a). Determine the reactions of this beam (that is, find the
fixed-end moments and fixed-end forces).
q dx
q
FIG. 10-17 Example 10-5. (a) Fixed-end
beam with a uniform load over part of
the span, and (b) reactions produced by
an element q dx of the uniform load
B
A
MA
MB
a
RA
RB
A
dMA
B
x
dRA
L
dx
dMB
dRB
(b)
(a)
Solution
Procedure. We can find the reactions of this beam by using the principle of
superposition together with the results obtained in the preceding example
(Example 10-4). In that example we found the reactions of a fixed-end beam
subjected to a concentrated load P acting at distance a from the left-hand end
(see Fig. 10-15a and Eqs. 10-25 and 10-26).
In order to apply those results to the uniform load of Fig. 10-17a, we will
treat an element of the uniform load as a concentrated load of magnitude q dx
acting at distance x from the left-hand end (Fig. 10-17b). Then, using the formulas derived in Example 10-4, we can obtain the reactions caused by this element
of load. Finally, by integrating over the length a of the uniform load, we can
obtain the reactions due to the entire uniform load.
Fixed-end moments. Let us begin with the moment reactions, for which we
use Eqs. (10-25a and b) of Example 10-4. To obtain the moments caused by the
element q dx of the uniform load (compare Fig. 10-17b with Fig. 10-15a), we
replace P with q dx, a with x, and b with L x. Thus, the fixed-end moments
due to the element of load (Fig. 10-17b) are
qx(L x)2dx
dMA
L2
qx 2(L x)dx
dMB
L2
Integrating over the loaded part of the beam, we get the fixed-end moments due
to the entire uniform load:
冕
8aL 3a )
冕 x(L x) dx 1q2a(6L
L
q
qa
冕dM 冕 x (L x)dx (4L 3a)
L
12L
q
MA dMA
L2
MB
a
2
2
2
2
2
(10-30a)
0
a
B
2
2
3
2
(10-30b)
0
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
728
CHAPTER 10 Statically Indeterminate Beams
Fixed-end forces. Proceeding in a similar manner as for the fixed-end
moments, but using Eqs. (10-26a and b), we obtain the following expressions
for the fixed-end forces due to the element q dx of load:
q(L x)2(L 2x)dx
dRA
qx2(3L 2x)dx
dRB
L3
Integration gives
冕
q
MA
A
RA
B
L
FIG. 10-18 Fixed-end beam with a
uniform load
冕
q a
qa
3
RA dRA 3 (L x)2(L 2x)dx (2L
2a2L a3)
L 0
2L3
MB
RB
冕
q
RB dRB 3
L
冕 x (3L 2x)dx 2qLa(2L a)
a
(10-31a)
3
2
3
(10-31b)
0
Thus, all reactions (fixed-end moments and fixed-end forces) have been found.
Uniform load acting over the entire length of the beam. When the load acts
over the entire span (Fig. 10-18), we can obtain the reactions by substituting
a L into the preceding equations, yielding
qL2
MA MB
12
qL
RA RB
2
(10-32a,b)
The deflection at the midpoint of a uniformly loaded beam is also of interest. The simplest procedure for obtaining this deflection is to use the method of
superposition. The first step is to remove the moment restraints at the supports
and obtain a released structure in the form of a simple beam. The downward
deflection at the midpoint of a simple beam due to a uniform load (from Case 1,
Table G-2) is
5qL4
(dC)1
384EI
(u)
and the upward deflection at the midpoint due to the end moments (from Case
10, Table G-2) is
MAL2
(qL2/12)L2
qL4
(dC)2
8EI
8EI
96EI
(v)
Thus, the final downward deflection of the original fixed-end beam (Fig. 10-18)
is
dC (dC)1 (dC)2
Substituting for the deflections from Eqs. (u) and (v), we get
qL 4
dC
384EI
(10-33)
This deflection is one-fifth of the deflection at the midpoint of a simple beam
with a uniform load (Eq. u), again illustrating the stiffening effect of fixity at the
ends of the beam.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 10.4 Method of Superposition
729
Example 10-6
A beam ABC (Fig. 10-19a) rests on simple supports at points A and B and is
supported by a cable at point C. The beam has total length 2L and supports a
uniform load of intensity q. Prior to the application of the uniform load, there is
no force in the cable nor is there any slack in the cable.
When the uniform load is applied, the beam deflects downward at point C
and a tensile force T develops in the cable. Find the magnitude of this force.
Cable
D
h
q
A
C
B
FIG. 10-19 Example 10-6.
Beam ABC with one end
supported by a cable
D
T
q
A
B
C
C
T
L
L
(b)
(a)
Solution
Redundant force. The structure ABCD, consisting of the beam and cable,
has three vertical reactions (at points A, B, and D). However, only two equations
of equilibrium are available from a free-body diagram of the entire structure.
Therefore, the structure is statically indeterminate to the first degree, and we
must select one redundant quantity for purposes of analysis.
The tensile force T in the cable is a suitable choice for the redundant. We
can release this force by removing the connection at point C, thereby cutting the
structure into two parts (Fig. 10-19b). The released structure consists of the
beam ABC and the cable CD as separate elements, with the redundant force T
acting upward on the beam and downward on the cable.
Equation of compatibility. The deflection at point C of beam ABC (Fig.
10-19b) consists of two parts, a downward deflection (dC)1 due to the uniform
load and an upward deflection (dC)2 due to the force T. At the same time, the
lower end C of cable CD displaces downward by an amount (dC)3, equal to the
elongation of the cable due to the force T. Therefore, the equation of compatibility, which expresses the fact that the downward deflection of end C of the
beam is equal to the elongation of the cable, is
(dC)1 (dC)2 (dC)3
(w)
Having formulated this equation, we now turn to the task of evaluating all three
displacements.
Force-displacement relations. The deflection (dC)1 at the end of the overhang (point C in beam ABC) due to the uniform load can be found from the
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
730
CHAPTER 10 Statically Indeterminate Beams
results given in Example 9-9 of Section 9.5 (see Fig. 9-21). Using Eq. (9-59) of
that example, and substituting a L, we get
qL4
(dC)1
4EbIb
(x)
where EbIb is the flexural rigidity of the beam.
The deflection of the beam at point C due to the force T can be taken from
the answer to Problem 9.8-5 or Problem 9.9-3. Those answers give the deflection (dC)2 at the end of the overhang when the length of the overhang is a:
Ta2(L a)
(dC) 2
3EbIb
Now substituting a L, we obtain the desired deflection:
2TL3
(dC) 2
3EbIb
(y)
Finally, the elongation of the cable is
Th
(dC) 3
Ec Ac
(z)
where h is the length of the cable and Ec Ac is its axial rigidity.
Force in the cable. By substituting the three displacements (Eqs. x, y, and
z) into the equation of compatibility (Eq. w), we get
qL4
2TL3
Th
4Eb Ib
3EbIb
Ec Ac
Solving for the force T, we find
3qL4E c A c
T
(10-34)
With the force T known, we can find all reactions, shear forces, and bending
moments by means of free-body diagrams and equations of equilibrium.
This example illustrates how an internal force quantity (instead of an
external reaction) can be used as the redundant.
Cable
D
h
q
A
C
B
T
q
A
B
L
L
FIG. 10-19 (Repeated)
D
(a)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(b)
C
C
T
SECTION 10.5 Temperature Effects
★10.5
TEMPERATURE EFFECTS
y
A
731
h
T1
B
x
T2
L
FIG. 10-20 Propped cantilever beam with
a temperature differential
Temperature changes may produce changes in lengths of bars and lateral
deflections of beams as discussed previously in Sections 2.5 and 9.13. If
these length changes and lateral deflections are restrained, thermal
stresses will be produced in the material. In Section 2.5 we saw how to
find these stresses in statically indeterminate bars, and now we will
consider some of the effects of temperature changes in statically indeterminate beams.
The stresses and deflections produced by temperature changes in a
statically indeterminate beam can be analyzed by methods that are
similar to those already described for the effects of loads. To begin
the discussion, consider the propped cantilever beam AB shown in Fig.
10-20. We assume that the beam was originally at a uniform temperature
T0, but later its temperature is increased to T1 on the upper surface and
T2 on the lower surface. The variation of temperature over the height h
of the beam is assumed to be linear.
Because the temperature varies linearly, the average temperature of
the beam is
T1 T2
Taver
2
(10-35)
and occurs at midheight of the beam. The difference between this
average temperature and the initial temperature T0 results in a tendency
for the beam to change in length. If the beam is free to expand longitudinally, its length will increase by an amount dT given by Eq. (9-145),
which is repeated here:
冢
冣
T1 T2
d T a(Taver T0)L a T0 L
2
(10-36)
In this equation, a is the coefficient of thermal expansion of the material
and L is the length of the beam. If longitudinal expansion is free to
occur, no axial stresses will be produced by the temperature changes.
However, if longitudinal expansion is restrained, axial stresses will
develop, as described in Section 2.5.
Now consider the effects of the temperature differential T2 T1,
which tends to produce a curvature of the beam but no change in length.
Curvature due to temperature changes is described in Section 9.13,
where the following differential equation of the deflection curve is
derived (see Eq. 9-147):
a(T2 T1)
d 2v
2
(10-37)
dx
h
This equation applies to a beam that is unrestrained by supports and
therefore is free to deflect and rotate. Note that when T2 is greater than
T1, the curvature is positive and the beam tends to bend concave upward.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
732
CHAPTER 10 Statically Indeterminate Beams
y
A
h
T1
B
x
MB
T2
MA
RA
Method of Superposition
RB
L
(a)
A
T1
B
T2
(b)
T1
T2
Deflections and rotations of simple beams and cantilever beams due to a
temperature differential can be determined with the aid of Eq. (10-37),
as discussed in Section 9.13. We can now use those results when analyzing statically indeterminate beams by the method of
superposition.
(uB)1
(dB)1
(c)
FIG. 10-21 (a) Fixed-end beam with a
temperature differential, (b) released
structure, and (c) deflection curve for the
released structure
To illustrate the use of superposition, let us determine the reactions of
the fixed-end beam of Fig. 10-21a due to the temperature differential. As
usual, we begin the analysis by selecting the redundant reactions.
Although other choices result in more efficient calculations, we will
select the reactive force RB and reactive moment MB as the redundants in
order to illustrate the general methodology.
When the supports corresponding to the redundants are removed,
we obtain the released structure shown in Fig. 10-21b (a cantilever
beam). The deflection and angle of rotation at end B of this cantilever
(due to the temperature differential) are as follows (see Fig. 10-21c):
a(T2 T1)L2
a(T2 T1)L
(uB)1
(dB)1
2h
h
These equations are obtained from the solution to Problem 9.13-2 in the
preceding chapter. Note that when T2 is greater than T1, the deflection
(dB)1 is upward and the angle of rotation (uB)1 is counterclockwise.
Next, we need to find the deflections and angles of rotation in the
released structure (Fig. 10-21b) due to the redundants RB and MB. These
quantities are obtained from Cases 4 and 6, respectively, of Table G-1:
RB L3
(dB)2
3 EI
MB L2
(dB)3
2EI
RB L2
(uB)2
2 EI
MB L
(uB)3
EI
In these expressions, upward deflection and counterclockwise rotation
are positive (as in Fig. 10-21c).
We can now write the equations of compatibility for the deflection
and angle of rotation at support B as follows:
dB (dB)1 (dB)2 (dB)3 0
(a)
uB (uB)1 (uB)2 (uB)3 0
(b)
or, upon substituting the appropriate expressions,
a (T2 T1)L2
RBL3
MBL2
0
2h
3EI
2 EI
(c)
a(T2 T1)L
RB L2
MB L
0
h
2 EI
EI
(d)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 10.5 Temperature Effects
733
These equations can be solved simultaneously for the two redundants:
RB 0
aEI(T2 T1)
MB
h
The fact that RB is zero could have been anticipated initially from the
symmetry of the fixed-end beam. If we had utilized this fact from the
outset, the preceding solution would have been simplified because only
one equation of compatibility would have been required.
We also know from symmetry (or from equations of equilibrium)
that the reaction RB is equal to the reaction RA and that the moment MA
is equal to the moment MB. Therefore, the reactions for the fixed-end
beam shown in Fig. 10-21a are as follows:
RA RB 0
aEI(T2 T1)
MA MB
h
(10-38a,b)
From these results we see that the beam is subjected to a constant bending moment due to the temperature changes.
Differential Equation of the Deflection Curve
We can also analyze the fixed-end beam of Fig. 10-21a by solving the
differential equation of the deflection curve. When a beam is subjected
to both a bending moment M and a temperature differential T2 T1, the
differential equation becomes (see Eqs. 9-7 and 10-37):
or
a(T2 T1)
d 2v
M
2
dx
EI
h
(10-39a)
aEI(T2 T1)
EIv M
h
(10-39b)
For the fixed-end beam of Fig. 10-21a, the expression for the bending moment in the beam is
M RAx MA
(e)
where x is measured from support A. Substituting into the differential
equation and integrating, we obtain the following equation for the slope
of the beam:
RAx 2
aEI(T2 T1)x
EIv MAx C1
2
h
(f )
The two boundary conditions on the slope (v 0 when x 0 and
x L) give C1 0 and
RAL
aEI(T2 T1)
MA
h
2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(g)
734
CHAPTER 10 Statically Indeterminate Beams
A second integration gives the deflection of the beam:
RAx 3
MAx 2
aEI(T2 T1)x 2
EIv C2
2h
6
2
(h)
The boundary conditions on the deflection (v 0 when x 0 and
x L) give C2 0 and
RAL
aEI(T2 T1)
MA
3
h
(i)
Solving simultaneously Eqs. (g) and (i), we find
RA 0
aEI(T2 T1)
MA
h
From the equilibrium of the beam, we obtain RB 0 and MB MA.
Thus, these results agree with those found by the method of superposition (see Eqs. 10-38a and b).
Note that we carried out the preceding solution without taking
advantage of symmetry because we wished to illustrate the general
approach of the integration method.
Knowing the reactions of the beam, we can now find the shear
forces, bending moments, slopes, and deflections. The simplicity of the
results may surprise you.
★10.6
LONGITUDINAL DISPLACEMENTS AT THE ENDS OF A BEAM
When a beam is bent by lateral loads, the ends of the beam move closer
together. It is common practice to disregard these longitudinal displacements because usually they have no noticeable effect on the behavior of
the beam. In this section, we will show how to evaluate these displacements and determine whether or not they are important.
Consider a simple beam AB that is pin-supported at one end and
free to displace longitudinally at the other (Fig. 10-22a). When this
beam is bent by lateral loads, the deflection curve has the shape shown
in part (b) of the figure. In addition to the lateral deflections, there is a
longitudinal displacement at end B of the beam. End B moves horizontally from point B to point B through a small distance l, called the
curvature shortening of the beam.
As the name implies, curvature shortening is due to bending of the
axis of the beam and is not due to axial strains produced by tensile or
compressive forces. As we see from Fig. 10-22b, the curvature shortening is equal to the difference between the initial length L of the straight
beam and the length of the chord AB of the bent beam. Of course, both
the lateral deflections and the curvature shortening are highly exaggerated in the figure.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
735
SECTION 10.6 Longitudinal Displacements at the Ends of a Beam
y
y
A
B
x
A
dx
B′
B
x
ds
L
l
(a)
y
FIG. 10-22 (a) Simple beam with lateral
loads, (b) horizontal displacement l at
the end of the beam, and (c) horizontal
reactions H for a beam with immovable
supports
(b)
B
A
x
H
H
(c)
Curvature Shortening
To determine the curvature shortening, we begin by considering an
element of length ds measured along the curved axis of the beam
(Fig. 10-22b). The projection of this element on the horizontal axis has
length dx. The relationship between the length of the element and the
length of its horizontal projection is obtained from the Pythagorean
theorem:
(ds)2 (dx)2 (dv)2
where dv is the increment in the deflection v of the beam as we move
through the distance dx. Thus,
冪莦冢 莦冣
dv
2
苶
苶
(dv)2 dx 1
ds 兹(dx)
dx
2
(a)
The difference between the length of the element and the length of its
horizontal projection is
冪莦冢 莦冣
dv
冢
1冥
冤冪1莦
d莦
x冣
dv 2
ds 2 dx dx 1 2 dx dx
dx
2
(b)
Now let us introduce the following binomial series (see Appendix C):
t
t2
t3
兹1苶
t 1 . . .
2
8
16
(10-40)
which converges when t is numerically less than 1. If t is very small
compared to 1, we can disregard the terms involving t 2, t 3, and so on, in
comparison with the first two terms. Then we obtain
t
兹1苶
t ⬇ 1
2
(10-41)
The term (dv/dx)2 in Eq. (b) is ordinarily very small compared to 1.
Therefore, we can use Eq. (10-41) with t (dv/dx)2 and rewrite Eq. (b) as
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
736
CHAPTER 10 Statically Indeterminate Beams
冢 冣
1 dv
ds 2 dx dx 1
2 dx
2
冥
冢 冣
1 dv 2
1 dx
2 dx
(c)
If the left- and right-hand sides of this expression are integrated over the
length of the beam, we obtain an expression for the difference between
the length of the beam and the length of the chord AB (Fig. 10-22b):
冕
L
苶B
苶苶
LA
0
冢 冣
1 dv 2
dx
2 dx
Thus, the curvature shortening is
冕冢
1
l
2
L
0
冣
dv 2
dx
dx
(10-42)
This equation is valid provided the deflections and slopes are small.
Note that when the equation of the deflection curve is known, we
can substitute into Eq. (10-42) and determine the shortening l.
Horizontal Reactions
Now suppose that the ends of the beam are prevented from translating
longitudinally by immovable supports (Fig. 10-22c). Because the ends
cannot move toward each other, a horizontal reaction H will develop at
each end. This force will cause the axis of the beam to elongate as bending occurs.
In addition, the force H itself will have an effect upon the bending
moments in the beam, because an additional bending moment (equal to
H times the deflection) will exist at every cross section. Thus, the deflection curve of the beam depends not only upon the lateral loads but also
upon the reaction H, which in turn depends upon the shape of the deflection curve, as shown by Eq. (10-42).
Rather than attempt an exact analysis of this complicated problem,
let us obtain an approximate expression for the force H in order to ascertain its importance. For that purpose, we can use any reasonable
approximation to the deflection curve. In the case of a pin-ended beam
with downward loads (Fig. 10-22c), a good approximation is a parabola
having the equation
4dx(L x)
v
L2
(10-43)
where d is the downward deflection at the midpoint of the beam. The
curvature shortening l corresponding to this assumed deflected shape
can be found by substituting the expression for the deflection v into Eq.
(10-42) and integrating; the result is
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
737
SECTION 10.6 Longitudinal Displacements at the Ends of a Beam
2
8d
l
3L
(10-44)
The horizontal force H required to elongate the beam by this amount is
EAl
8EAd 2
H
L
3L2
(10-45)
in which EA is the axial rigidity of the beam. The corresponding axial
tensile stress in the beam is
H
8Ed2
st
A
3L2
(10-46)
This equation gives a close estimate of the tensile stress produced by the
immovable supports of a simple beam.
General Comments
Now let us substitute some numerical values so that we can assess the
significance of the curvature shortening. The deflection d at the
midpoint of the beam is usually very small compared to the length; for
example, the ratio d/L might be 1/500 or smaller. Using this value, and
also assuming that the material is steel with E 30 106 psi, we find
from Eq. (10-46) that the tensile stress is only 320 psi. Since the allowable tensile stress in the steel is typically 15,000 psi or larger, it becomes
clear that the axial stress due to the horizontal force H may be disregarded when compared to the ordinary working stresses in the beam.
Furthermore, in the derivation of Eq. (10-46) we assumed that the
ends of the beam were held rigidly against horizontal displacements,
which is not physically possible. In reality, small longitudinal displacements always occur, thereby reducing the axial stress calculated from
Eq. (10-46).*
From the preceding discussions, we conclude that the customary
practice of disregarding the effects of any longitudinal restraints and
assuming that one end of the beam is on a roller support (regardless of
the actual construction) is justified. The stiffening effect of longitudinal
restraints is significant only when the beam is very long and slender and
supports large loads. This behavior is sometimes referred to as “string
action,” because it is analogous to the action of a cable, or string, supporting a load.
*For a more complete analysis of beams with immovable supports, see Ref. 10-1.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
738
CHAPTER 10 Statically Indeterminate Beams
PROBLEMS CHAPTER 10
Differential Equations of the Deflection Curve
The problems for Section 10.3 are to be solved by integrating the differential equations of the deflection curve. All
beams have constant flexural rigidity EI. When drawing
shear-force and bending-moment diagrams, be sure to label
all critical ordinates, including maximum and minimum
values.
10.3-1 A propped cantilever beam AB of length L is loaded
by a counterclockwise moment M0 acting at support B (see
figure).
Beginning with the second-order differential equation
of the deflection curve (the bending-moment equation),
obtain the reactions, shear forces, bending moments,
slopes, and deflections of the beam. Construct the shearforce and bending-moment diagrams, labeling all critical
ordinates.
10.3-3 A cantilever beam AB of length L has a fixed support at A and a roller support at B (see figure). The support
at B is moved downward through a distance dB.
Using the fourth-order differential equation of the
deflection curve (the load equation), determine the reactions of the beam and the equation of the deflection curve.
(Note: Express all results in terms of the imposed displacement dB.)
y
x
MA
A
dB
B
RA
RB
L
y
M0
A
PROB. 10.3-3
B
x
MA
RA
RB
L
PROB. 10.3-1
10.3-2 A fixed-end beam AB of length L supports a uniform load of intensity q (see figure).
Beginning with the second-order differential equation
of the deflection curve (the bending-moment equation),
obtain the reactions, shear forces, bending moments,
slopes, and deflections of the beam. Construct the shearforce and bending-moment diagrams, labeling all critical
ordinates.
10.3-4 A cantilever beam AB of length L has a fixed support at A and a spring support at B (see figure). The spring
behaves in a linearly elastic manner with stiffness k.
If a uniform load of intensity q acts on the beam, what
is the downward displacement dB of end B of the beam?
(Use the second-order differential equation of the deflection
curve, that is, the bending-moment equation.)
y
q
y
q
A
MA
RA
PROB. 10.3-2
MA
L
k
RA
x
B
B
A
RB
MB
L
RB
PROB. 10.3-4
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
x
CHAPTER 10 Problems
10.3-5 A propped cantilever beam AB of length L supports
a triangularly distributed load of maximum intensity q0 (see
figure).
Beginning with the fourth-order differential equation
of the deflection curve (the load equation), obtain the reactions of the beam and the equation of the deflection curve.
px
q = q0 sin —
L
y
A
MA
x
B
RA
y
739
MB
RB
L
PROB. 10.3-7
q0
10.3-8 A fixed-end beam AB of length L supports a
A
MA
x
B
RA
RB
L
triangularly distributed load of maximum intensity q0 (see
figure).
Beginning with the fourth-order differential equation
of the deflection curve (the load equation), obtain the reactions of the beam and the equation of the deflection curve.
y
PROB. 10.3-5
q0
10.3-6 The load on a propped cantilever beam AB of length
L is parabolically distributed according to the equation q
q0(1 x2/L2), as shown in the figure.
Beginning with the fourth-order differential equation
of the deflection curve (the load equation), obtain the reactions of the beam and the equation of the deflection curve.
MA
A
x
B
L
RA
MB
RB
PROB. 10.3-8
y
(
x2
q = q0 1 —2
L
q0
MA
A
)
x
B
RA
L
RB
10.3-9 A counterclockwise moment M0 acts at the midpoint of a fixed-end beam ACB of length L (see figure).
Beginning with the second-order differential equation
of the deflection curve (the bending-moment equation),
determine all reactions of the beam and obtain the equation
of the deflection curve for the left-hand half of the beam.
Then construct the shear-force and bending-moment
diagrams for the entire beam, labeling all critical ordinates.
Also, draw the deflection curve for the entire beam.
PROB. 10.3-6
y
M0
10.3-7 The load on a fixed-end beam AB of length L is distributed in the form of a sine curve (see figure). The
intensity of the distributed load is given by the equation
q q0 sin px/L.
Beginning with the fourth-order differential equation
of the deflection curve (the load equation), obtain the reactions of the beam and the equation of the deflection curve.
A
MA
RA
C
L
—
2
PROB. 10.3-9
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
x
B
L
—
2
MB
RB
740
CHAPTER 10 Statically Indeterminate Beams
10.3.-10 A propped cantilever beam AB supports a
concentrated load P acting at the midpoint C (see figure).
Beginning with the second-order differential equation
of the deflection curve (the bending-moment equation),
determine all reactions of the beam and draw the
shear-force and bending-moment diagrams for the entire
beam.
Also, obtain the equations of the deflection curves for
both halves of the beam, and draw the deflection curve for
the entire beam.
★★
10.4-2 The propped cantilever beam shown in the figure
supports a uniform load of intensity q on the left-hand half
of the beam.
Find the reactions RA, RB, and MA, and then draw the
shear-force and bending-moment diagrams, labeling all
critical ordinates.
q
A
MA
RA
y
C
x
B
RA
L
—
2
RB
PROB. 10.4-2
10.4-3 The figure shows a propped cantilever beam ABC
having span length L and an overhang of length a. A concentrated load P acts at the end of the overhang.
Determine the reactions RA, RB, and MA for this beam.
Also, draw the shear-force and bending-moment diagrams,
labeling all critical ordinates.
RB
L
—
2
L
—
2
L
—
2
P
A
MA
B
PROB. 10.3-10
P
MA
Method of Superposition
A
The problems for Section 10.4 are to be solved by the
method of superposition. All beams have constant flexural
rigidity EI unless otherwise stated. When drawing shearforce and bending-moment diagrams, be sure to label all
critical ordinates, including maximum and minimum
values.
C
B
RA
RB
a
L
PROB. 10.4-3
10.4-4 Two flat beams AB and CD, lying in horizontal
10.4-1 A propped cantilever beam AB of length L carries a
concentrated load P acting at the position shown in the
figure.
Determine the reactions RA, RB, and MA for this beam.
Also, draw the shear-force and bending-moment diagrams,
labeling all critical ordinates.
P
planes, cross at right angles and jointly support a vertical
load P at their midpoints (see figure). Before the load P is
applied, the beams just touch each other. Both beams are
made of the same material and have the same widths. Also,
the ends of both beams are simply supported. The lengths
of beams AB and CD are LAB and LCD, respectively.
What should be the ratio tAB/tCD of the thicknesses of
the beams if all four reactions are to be the same?
P
B
A
MA
RA
a
b
A
tAB
C
RB
B
tCD
L
PROB. 10.4-1
D
PROB. 10.4-4
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
741
CHAPTER 10 Problems
10.4-5 Determine the fixed-end moments (MA and MB) and
fixed-end forces (RA and RB) for a beam of length L supporting a triangular load of maximum intensity q0 (see
figure). Then draw the shear-force and bending-moment
diagrams, labeling all critical ordinates.
E
P
A
B
q0
C
D
A
MA
B
RA
L
—
2
L
—
2
MB
RB
P = 1700 lb
MA
PROB. 10.4-5
10.4-6 A continuous beam ABC with two unequal spans,
one of length L and one of length 2L, supports a uniform
load of intensity q (see figure).
Determine the reactions RA, RB, and RC for this beam.
Also, draw the shear-force and bending-moment diagrams,
labeling all critical ordinates.
RA
RB
RA
RB
L
— = 5 ft
2
L = 10 ft
C
B
L
k
PROB. 10.4-7
q
A
C
B
A
2L
RC
PROB. 10.4-6
10.4-7 Beam ABC is fixed at support A and rests (at point
B) upon the midpoint of beam DE (see the first part of the
figure). Thus, beam ABC may be represented as a propped
cantilever beam with an overhang BC and a linearly elastic
support of stiffness k at point B (see the second part of the
figure).
The distance from A to B is L 10 ft, the distance
from B to C is L /2 5 ft, and the length of beam DE is
L 10 ft. Both beams have the same flexural rigidity EI.
A concentrated load P 1700 lb acts at the free end of
beam ABC.
Determine the reactions RA, RB , and MA for beam
ABC. Also, draw the shear-force and bending-moment
diagrams for beam ABC, labeling all critical ordinates.
10.4-8 The beam ABC shown in the figure has flexural
rigidity EI 4.0 MN·m2. When the loads are applied to
the beam, the support at B settles vertically downward
through a distance of 6.0 mm.
Calculate the reaction RB at support B.
6 kN/m
3 kN
A
B
3m
6 mm settlement
PROB. 10.4-8
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
1m
1m
RB
C
742
CHAPTER 10 Statically Indeterminate Beams
10.4-9 A beam AB is cantilevered from a wall at one end
and held by a tie rod at the other end (see figure). The beam
is an S 6 12.5 I-beam section with length L 6 ft. The
tie rod has a diameter of 1/4 inch and length H 3 ft. Both
members are made of steel with E 30 106 psi. A uniform load of intensity q 200 lb/ft acts along the length of
the beam. Before the load q is applied, the tie rod just meets
the end of the cable.
(a) Determine the tensile force T in the tie rod due to
the uniform load q.
(b) Draw the shear-force and bending-moment diagrams for the beam, labeling all critical ordinates.
C
1
— in. tie rod
4
H = 3 ft
q = 200 lb/ft
S6
A
P
A
B
E
D
MA
RA
L
4
RD
L
4
RE
L
4
L
4
PROB. 10.4-11
10.4-12 A three-span continuous beam ABCD with three
equal spans supports a uniform load of intensity q (see
figure).
Determine all reactions of this beam and draw the
shear-force and bending-moment diagrams, labeling all
critical ordinates.
B
12.5
C
q
L = 6 ft
A
PROB. 10.4-9
B
D
C
10.4-10 The figure shows a nonprismatic, propped
cantilever beam AB with flexural rigidity 2EI from A to C
and EI from C to B.
Determine all reactions of the beam due to the uniform
load of intensity q. (Hint: Use the results of Problems 9.7-1
and 9.7-2.)
q
MA
RA
A
2EI
L
2
RA
L
RB
EI
L
2
B
RB
PROB. 10.4-10
10.4-11 A beam ABC is fixed at end A and supported by
beam DE at point B (see figure). Both beams have the same
cross section and are made of the same material.
(a) Determine all reactions due to the load P.
(b) What is the numerically largest bending moment in
either beam?
RC
L
RD
PROB. 10.4-12
10.4-13 A beam AC rests on simple supports at points A
and C (see figure). A small gap 0.4 in. exists between
the unloaded beam and a support at point B, which is
midway between the ends of the beam. The beam has
total length 2L 80 in. and flexural rigidity EI 0.4
109 lb-in.2
Plot a graph of the bending moment MB at the midpoint of the beam as a function of the intensity q of the
uniform load.
Hints: Begin by determining the intensity q0 of the
load that will just close the gap. Then determine the corresponding bending moment (MB)0. Next, determine the
bending moment MB (in terms of q) for the case where
q q0. Finally, make a statically indeterminate analysis
and determine the moment MB (in terms of q) for the case
where q q0. Plot MB (units of lb-in.) versus q (units of
lb/in.) with q varying from 0 to 2500 lb/in.
★
C
L
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 10 Problems
743
q
A
C
B
= 0.4 in.
RA
RC
RB
L = 40 in.
L = 40 in.
t = 1.5 in.
PROB. 10.4-13
10.4-14 A fixed-end beam AB of length L is subjected to
a moment M0 acting at the position shown in the figure.
(a) Determine all reactions for this beam.
(b) Draw shear-force and bending-moment diagrams
for the special case in which a b L /2.
★
d=
40 in.
B
h=
50 in.
A
PROB. 10.4-15
M0
A
B
MB
MA
RA
a
b
RB
L
PROB. 10.4-14
10.4-15 A temporary wood flume serving as a channel for
irrigation water is shown in the figure. The vertical boards
forming the sides of the flume are sunk in the ground,
which provides a fixed support. The top of the flume is held
by tie rods that are tightened so that there is no deflection
of the boards at that point. Thus, the vertical boards may be
modeled as a beam AB, supported and loaded as shown in
the last part of the figure.
Assuming that the thickness t of the boards is 1.5 in.,
the depth d of the water is 40 in., and the height h to the tie
rods is 50 in., what is the maximum bending stress s in the
boards? (Hint: The numerically largest bending moment
occurs at the fixed support.)
★10.4-16 Two identical, simply supported beams AB and
CD are placed so that they cross each other at their midpoints (see figure). Before the uniform load is applied, the
beams just touch each other at the crossing point.
Determine the maximum bending moments (MAB)max
and (MCD)max in beams AB and CD, respectively, due to the
uniform load if the intensity of the load is q 6.4 kN/m
and the length of each beam is L 4 m.
★
q
A
D
B
C
PROB. 10.4-16
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
744
CHAPTER 10 Statically Indeterminate Beams
★10.4-17 The cantilever beam AB shown in the figure is an
S 6 12.5 steel I-beam with E 30 106 psi. The simple
beam DE is a wood beam 4 in.
12 in. (nominal dimensions) in cross section with E 1.5
106 psi.
A steel rod AC of diameter 0.25 in., length 10 ft, and
E 30 106 psi serves as a hanger joining the two beams.
The hanger fits snugly between the beams before the uniform load is applied to beam DE.
Determine the tensile force F in the hanger and the
maximum bending moments MAB and MDE in the two
beams due to the uniform load, which has intensity q
400 lb/ft. (Hint: To aid in obtaining the maximum bending
moment in beam DE, draw the shear-force and bendingmoment diagrams.)
S6
A
10.4-19 The continuous frame ABC has a fixed support at
A, a roller support at C, and a rigid corner connection at B
(see figure). Members AB and BC each have length L and
flexural rigidity EI. A horizontal force P acts at midheight
of member AB.
(a) Find all reactions of the frame.
(b) What is the largest bending moment Mmax in the
frame? (Note: Disregard axial deformations in member AB
and consider only the effects of bending.)
★
L
P
12.5
B
Steel rod
A
10 ft
400 lb/ft
C
10.4-20 The continuous frame ABC has a pinned support
at A, a pinned support at C, and a rigid corner connection at
B (see figure). Members AB and BC each have length L and
flexural rigidity EI. A horizontal force P acts at midheight
of member AB.
(a) Find all reactions of the frame.
(b) What is the largest bending moment Mmax in the
frame? (Note: Disregard axial deformations in members AB
and BC and consider only the effects of bending.)
★
10 ft
10 ft
PROB. 10.4-17
10.4-18 The beam AB shown in the figure is simply supported at A and B and supported on a spring of stiffness k
at its midpoint C. The beam has flexural rigidity EI and
length 2L.
What should be the stiffness k of the spring in order
that the maximum bending moment in the beam (due to the
uniform load) will have the smallest possible value?
★
L
L
—
2
B
C
L
—
2
k
L
HC
C
B
VC
P
q
PROB. 10.4-18
MA
PROB. 10.4-19
Wood beam
L
HA
VA
E
A
VC
L
—
2
6 ft
D
C
B
L
—
2
A
HA
VA
PROB. 10.4-20
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
745
CHAPTER 10 Problems
★10.4-21 A wide-flange beam ABC rests on three identical
spring supports at points A, B, and C (see figure). The
flexural rigidity of the beam is EI 6912
106 lb-in.2,
and each spring has stiffness k 62,500 lb/in. The length
of the beam is L 16 ft.
If the load P is 6000 lb, what are the reactions RA, RB,
and RC? Also, draw the shear-force and bending-moment
diagrams for the beam, labeling all critical ordinates.
P
A
C
B
L
—
4
RA
L
—
4
RB
★★10.4-23 A beam supporting a uniform load of intensity q
throughout its length rests on pistons at points A, C, and B
(see figure). The cylinders are filled with oil and are connected by a tube so that the oil pressure on each piston is
the same. The pistons at A and B have diameter d1, and the
piston at C has diameter d2.
(a) Determine the ratio of d2 to d1 so that the largest
bending moment in the beam is as small as possible.
(b) Under these optimum conditions, what is the
largest bending moment Mmax in the beam?
(c) What is the difference in elevation between point C
and the end supports?
L
L
A
L
—
2
RC
PROB. 10.4-21
B
C
p
p
p
d1
d2
d1
PROB. 10.4-23
★★
10.4-22 A fixed-end beam AB of length L is subjected
to a uniform load of intensity q acting over the middle
region of the beam (see figure).
(a) Obtain a formula for the fixed-end moments MA
and MB in terms of the load q, the length L, and the length b
of the loaded part of the beam.
(b) Plot a graph of the fixed-end moment MA versus
the length b of the loaded part of the beam. For convenience, plot the graph in the following nondimensional form:
MA
b
2 versus
L
qL /12
with the ratio b/L varying between its extreme values of 0
and 1.
(c) For the special case in which a b L/ 3, draw
the shear-force and bending-moment diagrams for the
beam, labeling all critical ordinates.
10.4-24 A thin steel beam AB used in conjunction with
an electromagnet in a high-energy physics experiment is
securely bolted to rigid supports (see figure). A magnetic
field produced by coils C results in a force acting on the
beam. The force is trapezoidally distributed with maximum
intensity q0 18 kN/m. The length of the beam between
supports is L 200 mm and the dimension c of the trapezoidal load is 50 mm. The beam has a rectangular cross
section with width b 60 mm and height h 20 mm.
Determine the maximum bending stress smax and the
maximum deflection dmax for the beam. (Disregard any
effects of axial deformations and consider only the effects
of bending. Use E 200 GPa.)
★★★
c
q
q0
h
B
A
A
B
MB
MA
RA
a
b
a
C
RB
L
PROB. 10.4-22
c
L
PROB. 10.4-24
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
746
CHAPTER 10 Statically Indeterminate Beams
Temperature Effects
The beams described in the problems for Section 10.5 have
constant flexural rigidity EI.
10.5-1 A cable CD of length H is attached to the midpoint
of a simple beam AB of length L (see figure). The moment
of inertia of the beam is I, and the effective cross-sectional
area of the cable is A. The cable is initially taut but without
any initial tension.
Obtain a formula for the tensile force S in the cable
when the temperature drops uniformly by T degrees,
assuming that the beam and cable are made of the same
material (modulus of elasticity E and coefficient of thermal
expansion a). (Use the method of superposition in the
solution.)
10.5-4 A two-span beam with spans of lengths L and L /2 is
subjected to a temperature differential with temperature T1
on its upper surface and T2 on its lower surface (see figure).
Determine all reactions for this beam. (Use the method
of superposition in the solution. Also, if desired, use the
results from Problems 9.8-5 and 9.13-3.)
A
D
RB
H
C
RC
L
—
2
L
B
L
—
2
T1
B
T2
RA
C
L
—
2
Cable
h
T1
T2
Beam
T
A
10.5-3 Solve the preceding problem by integrating the
differential equation of the deflection curve.
PROBS. 10.5-4 and 10.5-5
10.5-5 Solve the preceding problem by integrating the
differential equation of the deflection curve.
★
PROB. 10.5-1
Longitudinal Displacements at the Ends of Beams
10.6-1 Assume that the deflected shape of a beam AB with
10.5-2 A propped cantilever beam, fixed at the left-hand
end A and simply supported at the right-hand end B, is subjected to a temperature differential with temperature T1 on
its upper surface and T2 on its lower surface (see figure).
Find all reactions for this beam. (Use the method of
superposition in the solution. Also, if desired, use the
results from Problem 9.13-1.)
y
A
MA
h
T1
B
y
x
T2
H
RA
L
PROBS. 10.5-2 and 10.5-3
immovable pinned supports (see figure) is given by the
equation v –d sin px/L, where d is the deflection at the
midpoint of the beam and L is the length. Also, assume that
the beam has constant axial rigidity EA.
(a) Obtain formulas for the longitudinal force H at
the ends of the beam and the corresponding axial tensile
stress st.
(b) For an aluminum-alloy beam with E 10
106
psi, calculate the tensile stress st when the ratio of the
deflection d to the length L equals 1/200, 1/400, and 1/600.
A
d
RB
L
PROB. 10.6-1
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
B
H
x
747
CHAPTER 10 Problems
10.6-2 (a) A simple beam AB with length L and height h
supports a uniform load of intensity q (see the first part of
the figure). Obtain a formula for the curvature shortening l
of this beam. Also, obtain a formula for the maximum
bending stress sb in the beam due to the load q.
(b) Now assume that the ends of the beam are pinned
so that curvature shortening is prevented and a horizontal
force H develops at the supports (see the second part of the
figure). Obtain a formula for the corresponding axial tensile
stress st.
(c) Using the formulas obtained in parts (a) and (b),
calculate the curvature shortening l, the maximum bending
stress sb, and the tensile stress st for the following steel
beam: length L 3 m, height h 300 mm, modulus of
elasticity E 200 GPa, and moment of inertia I
36
106 mm4. Also, the load on the beam has intensity
q 25 kN/m.
Compare the tensile stress st produced by the axial
forces with the maximum bending stress sb produced by
the uniform load.
q
A
h
B
L
q
A
B
H
h
L
PROB. 10.6-2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
H
11
Columns
11.1 INTRODUCTION
P
P
B
B
L
A
A
(a)
(b)
FIG. 11-1 Buckling of a slender column
due to an axial compressive load P
Load-carrying structures may fail in a variety of ways, depending upon
the type of structure, the conditions of support, the kinds of loads, and
the materials used. For instance, an axle in a vehicle may fracture
suddenly from repeated cycles of loading, or a beam may deflect excessively, so that the structure is unable to perform its intended functions.
These kinds of failures are prevented by designing structures so that the
maximum stresses and maximum displacements remain within tolerable
limits. Thus, strength and stiffness are important factors in design, as
discussed throughout the preceding chapters.
Another type of failure is buckling, which is the subject matter of
this chapter. We will consider specifically the buckling of columns,
which are long, slender structural members loaded axially in compression (Fig. 11-1a). If a compression member is relatively slender, it may
deflect laterally and fail by bending (Fig. 11-1b) rather than failing by
direct compression of the material. You can demonstrate this behavior by
compressing a plastic ruler or other slender object. When lateral bending
occurs, we say that the column has buckled. Under an increasing axial
load, the lateral deflections will increase too, and eventually the column
will collapse completely.
The phenomenon of buckling is not limited to columns. Buckling
can occur in many kinds of structures and can take many forms. When
you step on the top of an empty aluminum can, the thin cylindrical walls
buckle under your weight and the can collapses. When a large bridge
collapsed a few years ago, investigators found that failure was caused by
the buckling of a thin steel plate that wrinkled under compressive
stresses. Buckling is one of the major causes of failures in structures,
and therefore the possibility of buckling should always be considered in
design.
748
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 11.2
Buckling and Stability
749
11.2 BUCKLING AND STABILITY
To illustrate the fundamental concepts of buckling and stability, we will
analyze the idealized structure, or buckling model, shown in Fig. 11-2a.
This hypothetical structure consists of two rigid bars AB and BC, each
of length L/2. They are joined at B by a pin connection and held in a
vertical position by a rotational spring having stiffness bR.*
This idealized structure is analogous to the column of Fig. 11-1a,
because both structures have simple supports at the ends and are compressed by an axial load P. However, the elasticity of the idealized
structure is “concentrated” in the rotational spring, whereas a real column can bend throughout its length (Fig. 11-1b).
In the idealized structure, the two bars are perfectly aligned and
the axial load P has its line of action along the longitudinal axis
(Fig. 11-2a). Consequently, the spring is initially unstressed and the bars
are in direct compression.
Now suppose that the structure is disturbed by some external force
that causes point B to move a small distance laterally (Fig. 11-2b). The
rigid bars rotate through small angles u and a moment develops in the
spring. The direction of this moment is such that it tends to return the
structure to its original straight position, and therefore it is called a
restoring moment. At the same time, however, the tendency of the axial
P
P
P
C
C
C
Rigid
bar
bR
L
—
2
u
bR
u
B
B
MB
B
Rigid
bar
P
L
—
2
A
u
(c)
A
FIG. 11-2 Buckling of an idealized
structure consisting of two rigid bars
and a rotational spring
(a)
(b)
*The general relationship for a rotational spring is M bRu, where M is the moment acting on the spring, bR is the rotational stiffness of the spring, and u is the angle through
which the spring rotates. Thus, rotational stiffness has units of moment divided by angle,
such as lb-in./rad or N·m/rad. The analogous relationship for a translational spring is
F bd, where F is the force acting on the spring, b is the translational stiffness of the
spring (or spring constant), and d is the change in length of the spring. Thus, translational
stiffness has units of force divided by length, such as lb/in. or N/m.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
750
CHAPTER 11 Columns
compressive force is to increase the lateral displacement. Thus, these
two actions have opposite effects—the restoring moment tends to
decrease the displacement and the axial force tends to increase it.
Now consider what happens when the disturbing force is removed.
If the axial force P is relatively small, the action of the restoring moment
will predominate over the action of the axial force and the structure will
return to its initial straight position. Under these conditions, the structure
is said to be stable. However, if the axial force P is large, the lateral displacement of point B will increase and the bars will rotate through larger
and larger angles until the structure collapses. Under these conditions,
the structure is unstable and fails by lateral buckling.
Critical Load
The transition between the stable and unstable conditions occurs at a
special value of the axial force known as the critical load (denoted by
the symbol Pcr). We can determine the critical load of our buckling
model by considering the structure in the disturbed position (Fig. 11-2b)
and investigating its equilibrium.
First, we consider the entire structure as a free body and sum
moments about support A. This step leads to the conclusion that there is
no horizontal reaction at support C. Second, we consider bar BC as a
free body (Fig. 11-2c) and note that it is subjected to the action of the
axial forces P and the moment MB in the spring. The moment MB is
equal to the rotational stiffness bR times the angle of rotation 2u of the
spring; thus,
MB 2bRu
(a)
Since the angle u is a small quantity, the lateral displacement of point B
is uL /2. Therefore, we obtain the following equation of equilibrium by
summing moments about point B for bar BC (Fig. 11-2c):
冢 冣
uL
MB P 0
2
(b)
or, upon substituting from Eq. (a),
冢2b
R
冣
PL
u 0
2
(11-1)
One solution of this equation is u 0, which is a trivial solution and
merely means that the structure is in equilibrium when it is perfectly
straight, regardless of the magnitude of the force P.
A second solution is obtained by setting the term in parentheses
equal to zero and solving for the load P, which is the critical load:
4bR
Pcr
L
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(11-2)
SECTION 11.2
Buckling and Stability
751
At the critical value of the load the structure is in equilibrium regardless
of the magnitude of the angle u (provided the angle remains small,
because we made that assumption when deriving Eq. b).
From the preceding analysis we see that the critical load is the only
load for which the structure will be in equilibrium in the disturbed position. At this value of the load, the restoring effect of the moment in the
spring just matches the buckling effect of the axial load. Therefore, the
critical load represents the boundary between the stable and unstable
conditions.
If the axial load is less than Pcr, the effect of the moment in the
spring predominates and the structure returns to the vertical position
after a slight disturbance; if the axial load is larger than Pcr, the effect of
the axial force predominates and the structure buckles:
If P Pcr, the structure is stable
If P Pcr, the structure is unstable
From Eq. (11-2) we see that the stability of the structure is increased
either by increasing its stiffness or by decreasing its length. Later in this
chapter, when we determine critical loads for various types of columns,
we will see that these same observations apply.
Summary
P
Unstable equilibrium
B
Neutral equilibrium
Pcr
Stable equilibrium
O
u
FIG. 11-3 Equilibrium diagram for
buckling of an idealized structure
Let us now summarize the behavior of the idealized structure (Fig.
11-2a) as the axial load P increases from zero to a large value.
When the axial load is less than the critical load (0 P Pcr), the
structure is in equilibrium when it is perfectly straight. Because the
equilibrium is stable, the structure returns to its initial position after
being disturbed. Thus, the structure is in equilibrium only when it is perfectly straight (u 0).
When the axial load is greater than the critical load (P Pcr), the
structure is still in equilibrium when u 0 (because it is in direct compression and there is no moment in the spring), but the equilibrium is
unstable and cannot be maintained. The slightest disturbance will cause
the structure to buckle.
At the critical load (P Pcr), the structure is in equilibrium even
when point B is displaced laterally by a small amount. In other words,
the structure is in equilibrium for any small angle u, including u 0.
However, the structure is neither stable nor unstable—it is at the boundary between stability and instability. This condition is referred to as
neutral equilibrium.
The three equilibrium conditions for the idealized structure are
shown in the graph of axial load P versus angle of rotation u (Fig. 11-3).
The two heavy lines, one vertical and one horizontal, represent the
equilibrium conditions. Point B, where the equilibrium diagram
branches, is called a bifurcation point.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
752
CHAPTER 11 Columns
The horizontal line for neutral equilibrium extends to the left and
right of the vertical axis because the angle u may be clockwise or
counterclockwise. The line extends only a short distance, however,
because our analysis is based upon the assumption that u is a small angle.
(This assumption is quite valid, because u is indeed small when the
structure first departs from its vertical position. If buckling continues and
u becomes large, the line labeled “Neutral equilibrium” curves upward,
as shown later in Fig. 11-11.)
The three equilibrium conditions represented by the diagram of
Fig. 11-3 are analogous to those of a ball placed upon a smooth surface
(Fig. 11-4). If the surface is concave upward, like the inside of a dish,
the equilibrium is stable and the ball always returns to the low point
when disturbed. If the surface is convex upward, like a dome, the ball
can theoretically be in equilibrium on top of the surface, but the equilibrium is unstable and in reality the ball rolls away. If the surface is
perfectly flat, the ball is in neutral equilibrium and remains wherever it
is placed.
As we will see in the next section, the behavior of an ideal elastic
column is analogous to that of the buckling model shown in Fig. 11-2.
Furthermore, many other kinds of structural and mechanical systems fit
this model.
11.3 COLUMNS WITH PINNED ENDS
We begin our investigation of the stability behavior of columns by analyzing a slender column with pinned ends (Fig. 11-5a). The column is
loaded by a vertical force P that is applied through the centroid of the
end cross section. The column itself is perfectly straight and is made of a
linearly elastic material that follows Hooke’s law. Since the column is
assumed to have no imperfections, it is referred to as an ideal column.
For purposes of analysis, we construct a coordinate system with its
origin at support A and with the x axis along the longitudinal axis of the
column. The y axis is directed to the left in the figure, and the z axis (not
shown) comes out of the plane of the figure toward the viewer. We
assume that the xy plane is a plane of symmetry of the column and that
any bending takes place in that plane (Fig. 11-5b). The coordinate system is identical to the one used in our previous discussions of beams, as
can be seen by rotating the column clockwise through an angle of 90°.
FIG. 11-4 Ball in stable, unstable, and
neutral equilibrium
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
753
SECTION 11.3 Columns with Pinned Ends
x
x
P
P
B
B
x
P
L
M
v
x
y
A
y
A
y
A
FIG. 11-5 Column with pinned ends:
(a) ideal column, (b) buckled shape, and
(c) axial force P and bending moment M
acting at a cross section
(a)
(b)
(c)
When the axial load P has a small value, the column remains
perfectly straight and undergoes direct axial compression. The only
stresses are the uniform compressive stresses obtained from the equation
s P/A. The column is in stable equilibrium, which means that it
returns to the straight position after a disturbance. For instance, if we
apply a small lateral load and cause the column to bend, the deflection
will disappear and the column will return to its original position when
the lateral load is removed.
As the axial load P is gradually increased, we reach a condition of
neutral equilibrium in which the column may have a bent shape. The
corresponding value of the load is the critical load Pcr. At this load the
column may undergo small lateral deflections with no change in the
axial force. For instance, a small lateral load will produce a bent shape
that does not disappear when the lateral load is removed. Thus, the critical load can maintain the column in equilibrium either in the straight
position or in a slightly bent position.
At higher values of the load, the column is unstable and may
collapse by buckling, that is, by excessive bending. For the ideal case
that we are discussing, the column will be in equilibrium in the straight
position even when the axial force P is greater than the critical load.
However, since the equilibrium is unstable, the smallest imaginable
disturbance will cause the column to deflect sideways. Once that
happens, the deflections will immediately increase and the column will
fail by buckling. The behavior is similar to that described in the
preceding section for the idealized buckling model (Fig. 11-2).
The behavior of an ideal column compressed by an axial load P
(Figs. 11-5a and b) may be summarized as follows:
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
754
CHAPTER 11 Columns
If P Pcr, the column is in stable equilibrium in the straight position.
If P Pcr, the column is in neutral equilibrium in either the straight
or a slightly bent position.
If P Pcr, the column is in unstable equilibrium in the straight
position and will buckle under the slightest disturbance.
Of course, a real column does not behave in this idealized manner
because imperfections are always present. For instance, the column is
not perfectly straight, and the load is not exactly at the centroid. Nevertheless, we begin by studying ideal columns because they provide
insight into the behavior of real columns.
Differential Equation for Column Buckling
To determine the critical loads and corresponding deflected shapes for
an ideal pin-ended column (Fig. 11-5a), we use one of the differential
equations of the deflection curve of a beam (see Eqs. 9-12a, b, and c in
Section 9.2). These equations are applicable to a buckled column
because the column bends as though it were a beam (Fig. 11-5b).
Although both the fourth-order differential equation (the load equation) and the third-order differential equation (the shear-force equation)
are suitable for analyzing columns, we will elect to use the second-order
equation (the bending-moment equation) because its general solution is
usually the simplest. The bending-moment equation (Eq. 9-12a) is
(11-3)
EIv M
in which M is the bending moment at any cross section, v is the lateral
deflection in the y direction, and EI is the flexural rigidity for bending in
the xy plane.
x
x
P
P
B
B
x
P
L
M
v
x
y
FIG. 11-5 (Repeated)
A
y
(a)
A
(b)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
y
A
(c)
SECTION 11.3 Columns with Pinned Ends
755
The bending moment M at distance x from end A of the buckled column is shown acting in its positive direction in Fig. 11-5c. Note that the
bending moment sign convention is the same as that used in earlier
chapters, namely, positive bending moment produces positive curvature
(see Figs. 9-3 and 9-4).
The axial force P acting at the cross section is also shown in Fig.
11-5c. Since there are no horizontal forces acting at the supports, there
are no shear forces in the column. Therefore, from equilibrium of
moments about point A, we obtain
x
P
M Pv 0 or M Pv
B
x
P
M
v
x
y
A
y
A
(11-4)
where v is the deflection at the cross section.
This same expression for the bending moment is obtained if we
assume that the column buckles to the right instead of to the left
(Fig. 11-6a). When the column deflects to the right, the deflection itself
is v but the moment of the axial force about point A also changes
sign. Thus, the equilibrium equation for moments about point A (see Fig.
11-6b) is
M – P(v) 0
(a)
(b)
FIG. 11-6 Column with pinned ends
(alternative direction of buckling)
which gives the same expression for the bending moment M as before.
The differential equation of the deflection curve (Eq. 11-3) now
becomes
EIv0 Pv 0
(11-5)
By solving this equation, which is a homogeneous, linear, differential
equation of second order with constant coefficients, we can determine
the magnitude of the critical load and the deflected shape of the buckled
column.
Note that we are analyzing the buckling of columns by solving the
same basic differential equation as the one we solved in Chapters 9 and
10 when finding beam deflections. However, there is a fundamental difference in the two types of analysis. In the case of beam deflections, the
bending moment M appearing in Eq. (11-3) is a function of the loads
only—it does not depend upon the deflections of the beam. In the case
of buckling, the bending moment is a function of the deflections themselves (Eq. 11-4).
Thus, we now encounter a new aspect of bending analysis. In
our previous work, the deflected shape of the structure was not considered, and the equations of equilibrium were based upon the geometry
of the undeformed structure. Now, however, the geometry of the
deformed structure is taken into account when writing equations of
equilibrium.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
756
CHAPTER 11 Columns
Solution of the Differential Equation
For convenience in writing the solution of the differential equation (Eq.
11-5), we introduce the notation
P
k2
EI
or
k
冪莦EPI
(11-6a,b)
in which k is always taken as a positive quantity. Note that k has units of
the reciprocal of length, and therefore quantities such as kx and kL are
nondimensional.
Using this notation, we can rewrite Eq. (11-5) in the form
v k2v 0
(11-7)
From mathematics we know that the general solution of this equation is
v C1 sin kx C2 cos kx
(11-8)
in which C1 and C2 are constants of integration (to be evaluated from the
boundary conditions, or end conditions, of the column). Note that the
number of arbitrary constants in the solution (two in this case) agrees
with the order of the differential equation. Also, note that we can verify
the solution by substituting the expression for v (Eq. 11-8) into the differential equation (Eq. 11-7) and reducing it to an identity.
To evaluate the constants of integration appearing in the solution
(Eq. 11-8), we use the boundary conditions at the ends of the column;
namely, the deflection is zero when x 0 and x L (see Fig. 11-5b):
v(0) 0 and v(L) 0
(a,b)
The first condition gives C2 0, and therefore
v C1 sin kx
(c)
C1 sin kL 0
(d)
The second condition gives
P
Unstable equilibrium
B
Neutral equilibrium
Pcr
Stable equilibrium
O
v
FIG. 11-7 Load-deflection diagram for an
ideal, linearly elastic column
From this equation we conclude that either C1 0 or sin kL 0. We
will consider both of these possibilities.
Case 1. If the constant C1 equals zero, the deflection v is also zero
(see Eq. c), and therefore the column remains straight. In addition, we
note that when C1 equals zero, Eq. (d) is satisfied for any value of the
quantity kL. Consequently, the axial load P may also have any value (see
Eq. 11-6b). This solution of the differential equation (known in mathematics as the trivial solution) is represented by the vertical axis of the
load-deflection diagram (Fig. 11-7). It gives the behavior of an ideal column that is in equilibrium (either stable or unstable) in the straight
position (no deflection) under the action of the compressive load P.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 11.3 Columns with Pinned Ends
757
Case 2. The second possibility for satisfying Eq. (d) is given by the
following equation, known as the buckling equation:
sin kL 0
(11-9)
This equation is satisfied when kL 0, p, 2p, . . . . However, since
kL 0 means that P 0, this solution is not of interest. Therefore, the
solutions we will consider are
n 1, 2, 3, . . .
kL np
(e)
or (see Eq. 11-6a):
n2p 2EI
P
L2
n 1, 2, 3, . . .
(11-10)
This formula gives the values of P that satisfy the buckling equation and
provide solutions (other than the trivial solution) to the differential
equation.
The equation of the deflection curve (from Eqs. c and e) is
npx
v C1 sin kx C1 sin
L
n 1, 2, 3, . . .
(11-11)
Only when P has one of the values given by Eq. (11-10) is it theoretically possible for the column to have a bent shape (given by Eq. 11-11).
For all other values of P, the column is in equilibrium only if it remains
straight. Therefore, the values of P given by Eq. (11-10) are the critical
loads for this column.
Critical Loads
The lowest critical load for a column with pinned ends (Fig. 11-8a) is
obtained when n 1:
x
x
p 2EI
Pcr = —
L2
P
4p 2EI
Pcr = —
L2
B
B
B
C1
C1
L
C1
FIG. 11-8 Buckled shapes for an ideal
column with pinned ends: (a) initially
straight column, (b) buckled shape for
n 1, and (c) buckled shape for n 2
y
A
(a)
A
(b)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
y
A
(c)
758
CHAPTER 11 Columns
x
p 2EI
Pcr
L2
p 2EI
Pcr = —
L2
B
C1
A
y
(b)
x
4p 2EI
Pcr = —
L2
B
C1
C1
y
A
(c)
FIG. 11-8 (Repeated)
(11-12)
The corresponding buckled shape (sometimes called a mode shape) is
px
(11-13)
v C1 sin
L
as shown in Fig. 11-8b. The constant C1 represents the deflection at the
midpoint of the column and may have any small value, either positive or
negative. Therefore, the part of the load-deflection diagram corresponding to Pcr is a horizontal straight line (Fig. 11-7). Thus, the deflection
at the critical load is undefined, although it must remain small for our
equations to be valid. Above the bifurcation point B the equilibrium is
unstable, and below point B it is stable.
Buckling of a pinned-end column in the first mode is called the fundamental case of column buckling.
The type of buckling described in this section is called Euler
buckling, and the critical load for an ideal elastic column is often called
the Euler load. The famous mathematician Leonhard Euler (1707–
1783), generally recognized as the greatest mathematician of all time,
was the first person to investigate the buckling of a slender column and
determine its critical load (Euler published his results in 1744); see
Ref. 11-1.
By taking higher values of the index n in Eqs. (11-10) and (11-11),
we obtain an infinite number of critical loads and corresponding mode
shapes. The mode shape for n 2 has two half-waves, as pictured in
Fig. 11-8c. The corresponding critical load is four times larger than the
critical load for the fundamental case. The magnitudes of the critical
loads are proportional to the square of n, and the number of half-waves
in the buckled shape is equal to n.
Buckled shapes for the higher modes are often of no practical interest because the column buckles when the axial load P reaches its lowest
critical value. The only way to obtain modes of buckling higher than the
first is to provide lateral support of the column at intermediate points,
such as at the midpoint of the column shown in Fig. 11-8 (see Example
11-1 at the end of this section).
General Comments
From Eq. (11-12), we see that the critical load of a column is proportional to the flexural rigidity EI and inversely proportional to the square
of the length. Of particular interest is the fact that the strength of the
material itself, as represented by a quantity such as the proportional
limit or the yield stress, does not appear in the equation for the critical
load. Therefore, increasing a strength property does not raise the critical
load of a slender column. It can only be raised by increasing the flexural
rigidity, reducing the length, or providing additional lateral support.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 11.3 Columns with Pinned Ends
2
1
C
2
2
1
1
C
1
2
FIG. 11-9 Cross sections of columns
showing principal centroidal axes with I1
I2
759
The flexural rigidity can be increased by using a “stiffer” material
(that is, a material with larger modulus of elasticity E) or by distributing
the material in such a way as to increase the moment of inertia I of the
cross section, just as a beam can be made stiffer by increasing the moment
of inertia. The moment of inertia is increased by distributing the material
farther from the centroid of the cross section. Hence, a hollow tubular
member is generally more economical for use as a column than a solid
member having the same cross-sectional area.
Reducing the wall thickness of a tubular member and increasing its
lateral dimensions (while keeping the cross-sectional area constant) also
increases the critical load because the moment of inertia is increased.
This process has a practical limit, however, because eventually the wall
itself will become unstable. When that happens, localized buckling
occurs in the form of small corrugations or wrinkles in the walls of the
column. Thus, we must distinguish between overall buckling of a column, which is discussed in this chapter, and local buckling of its parts.
The latter requires more detailed investigations and is beyond the scope
of this book.
In the preceding analysis (see Fig. 11-8), we assumed that the xy
plane was a plane of symmetry of the column and that buckling took place
in that plane. The latter assumption will be met if the column has lateral
supports perpendicular to the plane of the figure, so that the column is
constrained to buckle in the xy plane. If the column is supported only at its
ends and is free to buckle in any direction, then bending will occur about
the principal centroidal axis having the smaller moment of inertia.
For instance, consider the rectangular and wide-flange cross sections shown in Fig. 11-9. In each case, the moment of inertia I1 is
greater than the moment of inertia I2; hence the column will buckle in
the 1-1 plane, and the smaller moment of inertia I2 should be used in the
formula for the critical load. If the cross section is square or circular, all
centroidal axes have the same moment of inertia and buckling may
occur in any longitudinal plane.
Critical Stress
After finding the critical load for a column, we can calculate the corresponding critical stress by dividing the load by the cross-sectional area.
For the fundamental case of buckling (Fig. 11-8b), the critical stress is
Pcr
p 2E I
scr
(11-14)
A
AL2
in which I is the moment of inertia for the principal axis about which
buckling occurs. This equation can be written in a more useful form by
introducing the notation
r
冪莦A
I
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(11-15)
760
CHAPTER 11 Columns
in which r is the radius of gyration of the cross section in the plane of
bending.* Then the equation for the critical stress becomes
scr (ksi)
50
spl = 36 ksi
40
Euler's curve
E = 30 103 ksi
30
20
10
0
100 150 200 250
L
—
r
FIG. 11-10 Graph of Euler’s curve (from
Eq. 11-16) for structural steel with E
30 103 ksi and spl 36 ksi
P
Pcr
B
C
A
(11-17)
Note that the slenderness ratio depends only on the dimensions of the
column. A column that is long and slender will have a high slenderness
ratio and therefore a low critical stress. A column that is short and
stubby will have a low slenderness ratio and will buckle at a high stress.
Typical values of the slenderness ratio for actual columns are between
30 and 150.
The critical stress is the average compressive stress on the cross
section at the instant the load reaches its critical value. We can a plot a
graph of this stress as a function of the slenderness ratio and obtain a
curve known as Euler’s curve (Fig. 11-10). The curve shown in the
figure is plotted for a structural steel with E 30 103 ksi. The curve
is valid only when the critical stress is less than the proportional limit of
the steel, because the equations were derived using Hooke’s law. Therefore, we draw a horizontal line on the graph at the proportional limit of
the steel (assumed to be 36 ksi) and terminate Euler’s curve at that level
of stress.**
Effects of Large Deflections, Imperfections,
and Inelastic Behavior
D
O
(11-16)
in which L/r is a nondimensional ratio called the slenderness ratio:
L
Slenderness ratio
r
91
50
p 2E
scr 2
(L/r)
v
FIG. 11-11 Load-deflection diagram for
columns: Line A, ideal elastic column
with small deflections; Curve B, ideal
elastic column with large deflections;
Curve C, elastic column with imperfections; and Curve D, inelastic column
with imperfections
The equations for critical loads were derived for ideal columns, that is,
columns for which the loads are precisely applied, the construction is
perfect, and the material follows Hooke’s law. As a consequence, we
found that the magnitudes of the small deflections at buckling were
undefined.*** Thus, when P Pcr, the column may have any small
deflection, a condition represented by the horizontal line labeled A in the
load-deflection diagram of Fig. 11-11. (In this figure, we show only the
right-hand half of the diagram, but the two halves are symmetric about
the vertical axis.)
*Radius of gyration is described in Section 12.4.
**Euler’s curve is not a common geometric shape. It is sometimes mistakenly called
a hyperbola, but hyperbolas are plots of polynomial equations of the second degree in two
variables, whereas Euler’s curve is a plot of an equation of the third degree in two variables.
***In mathematical terminology, we solved a linear eigenvalue problem. The critical load
is an eigenvalue and the corresponding buckled mode shape is an eigenfunction.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 11.3 Columns with Pinned Ends
761
The theory for ideal columns is limited to small deflections because
we used the second derivative v0 for the curvature. A more exact
analysis, based upon the exact expression for curvature (Eq. 9-13 in
Section 9.2), shows that there is no indefiniteness in the magnitudes of
the deflections at buckling. Instead, for an ideal, linearly elastic column,
the load-deflection diagram goes upward in accord with curve B of
Fig. 11-11. Thus, after a linearly elastic column begins to buckle, an
increasing load is required to cause an increase in the deflections.
Now suppose that the column is not constructed perfectly; for
instance, the column might have an imperfection in the form of a small
initial curvature, so that the unloaded column is not perfectly straight.
Such imperfections produce deflections from the onset of loading, as
shown by curve C in Fig. 11-11. For small deflections, curve C
approaches line A as an asymptote. However, as the deflections become
large, it approaches curve B. The larger the imperfections, the further
curve C moves to the right, away from the vertical line. Conversely, if
the column is constructed with considerable accuracy, curve C
approaches the vertical axis and the horizontal line labeled A. By
comparing lines A, B, and C, we see that for practical purposes the
critical load represents the maximum load-carrying capacity of an
elastic column, because large deflections are not acceptable in most
applications.
Finally, consider what happens when the stresses exceed the proportional limit and the material no longer follows Hooke’s law. Of course,
the load-deflection diagram is unchanged up to the level of load at
which the proportional limit is reached. Then the curve for inelastic
behavior (curve D) departs from the elastic curve, continues upward,
reaches a maximum, and turns downward.
The precise shapes of the curves in Fig. 11-11 depend upon the
material properties and column dimensions, but the general nature of the
behavior is typified by the curves shown.
Only extremely slender columns remain elastic up to the critical
load. Stockier columns behave inelastically and follow a curve such as
D. Thus, the maximum load that can be supported by an inelastic
column may be considerably less than the Euler load for that same
column. Furthermore, the descending part of curve D represents sudden
and catastrophic collapse, because it takes smaller and smaller loads to
maintain larger and larger deflections. By contrast, the curves for elastic
columns are quite stable, because they continue upward as the
deflections increase, and therefore it takes larger and larger loads to
cause an increase in deflection. (Inelastic buckling is described in more
detail in Sections 11.7 and 11.8.)
Optimum Shapes of Columns
Compression members usually have the same cross sections throughout
their lengths, and therefore only prismatic columns are analyzed in this
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
762
CHAPTER 11 Columns
chapter. However, prismatic columns are not the optimum shape if minimum weight is desired. The critical load of a column consisting of a
given amount of material may be increased by varying the shape so that
the column has larger cross sections in those regions where the bending
moments are larger. Consider, for instance, a column of solid circular cross section with pinned ends. A column shaped as shown in Fig.
11-12a will have a larger critical load than a prismatic column made
from the same volume of material. As a means of approximating this
optimum shape, prismatic columns are sometimes reinforced over part
of their lengths (Fig. 11-12b).
Now consider a prismatic column with pinned ends that is free to
buckle in any lateral direction (Fig. 11-13a). Also, assume that the
column has a solid cross section, such as a circle, square, triangle,
rectangle, or hexagon (Fig. 11-13b). An interesting question arises: For a
given cross-sectional area, which of these shapes makes the most
efficient column? Or, in more precise terms, which cross section gives
the largest critical load? Of course, we are assuming that the critical load
is calculated from the Euler formula Pcr p 2EI/L2 using the smallest
moment of inertia for the cross section.
While a common answer to this question is “the circular shape,” you
can readily demonstrate that a cross section in the shape of an equilateral triangle gives a 21% higher critical load than does a circular cross
section of the same area (see Problem 11.3-10). The critical load for an
equilateral triangle is also higher than the loads obtained for the other
shapes; hence, an equilateral triangle is the optimum cross section
(based only upon theoretical considerations). For a mathematical analysis of optimum column shapes, including columns with varying cross
sections, see Ref. 11-4.
P
P
P
(b)
(a)
(a)
(b)
FIG. 11-12 Nonprismatic columns
FIG. 11-13 Which cross-sectional shape is
the optimum shape for a prismatic
column?
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 11.3 Columns with Pinned Ends
763
Example 11-1
A long, slender column ABC is pin-supported at the ends and compressed by an
axial load P (Fig. 11-14). Lateral support is provided at the midpoint B in the
plane of the figure. However, lateral support perpendicular to the plane of the
figure is provided only at the ends.
The column is constructed of a steel wide-flange section (W 8 28) having modulus of elasticity E 29 103 ksi and proportional limit spl 42 ksi.
The total length of the column is L 25 ft.
Determine the allowable load Pallow using a factor of safety n 2.5 with
respect to Euler buckling of the column.
P
C
2
W8
X
X
L
— = 12.5 ft
2
28
1
1
B
L
— = 12.5 ft
2
2
Section X-X
A
FIG. 11-14 Example 11-1. Euler buckling
of a slender column
(b)
(a)
Solution
Because of the manner in which it is supported, this column may buckle in
either of the two principal planes of bending. As one possibility, it may buckle
in the plane of the figure, in which case the distance between lateral supports is
L /2 12.5 ft and bending occurs about axis 2-2 (see Fig. 11-8c for the mode
shape of buckling).
As a second possibility, the column may buckle perpendicular to the plane
of the figure with bending about axis 1-1. Because the only lateral support in
this direction is at the ends, the distance between lateral supports is L 25 ft
(see Fig. 11-8b for the mode shape of buckling).
Column properties. From Table E-1, Appendix E, we obtain the following
moments of inertia and cross-sectional area for a W 8 28 column:
I1 98.0 in.4
I2 21.7 in.4
A 8.25 in.2
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
764
CHAPTER 11 Columns
Critical loads. If the column buckles in the plane of the figure, the critical
load is
4p 2EI2
p 2EI2
Pcr
2
L2
(L / 2)
Substituting numerical values, we obtain
4p 2EI2
4p 2(29 10 3 ksi)(21.7 in.4)
276 k
Pcr
2
L
If the column buckles perpendicular to the plane of the figure, the critical load is
p 2EI1
p 2(29 10 3 ksi)(98.0 in.4)
Pcr
312 k
2
L
Therefore, the critical load for the column (the smaller of the two preceding
values) is
Pcr 276 k
and buckling occurs in the plane of the figure.
Critical stresses. Since the calculations for the critical loads are valid only
if the material follows Hooke’s law, we need to verify that the critical stresses
do not exceed the proportional limit of the material. In the case of the larger
critical load, we obtain the following critical stress:
Pcr
312 k
scr 2 37.8 ksi
8.25 in.
A
Since this stress is less than the proportional limit (spl 42 ksi), both criticalload calculations are satisfactory.
Allowable load. The allowable axial load for the column, based on Euler
buckling, is
Pcr
276 k
Pallow 110 k
n
2.5
in which n 2.5 is the desired factor of safety.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
765
SECTION 11.4 Columns With Other Support Conditions
11.4 COLUMNS WITH OTHER SUPPORT CONDITIONS
Buckling of a column with pinned ends (described in the preceding section) is usually considered as the most basic case of buckling. However,
in practice we encounter many other end conditions, such as fixed ends,
free ends, and elastic supports. The critical loads for columns with various kinds of support conditions can be determined from the differential
equation of the deflection curve by following the same procedure that
we used when analyzing a pinned-end column.
The procedure is as follows. First, with the column assumed to be
in the buckled state, we obtain an expression for the bending moment in
the column. Second, we set up the differential equation of the deflection
curve, using the bending-moment equation (EIv M). Third, we solve
the equation and obtain its general solution, which contains two constants of integration plus any other unknown quantities. Fourth, we apply
boundary conditions pertaining to the deflection v and the slope v and
obtain a set of simultaneous equations. Finally, we solve those equations
to obtain the critical load and the deflected shape of the buckled column.
This straightforward mathematical procedure is illustrated in the following discussion of three types of columns.
Column Fixed at the Base and Free at the Top
The first case we will consider is an ideal column that is fixed at the
base, free at the top, and subjected to an axial load P (Fig. 11-15a).* The
deflected shape of the buckled column is shown in Fig. 11-15b. From
this figure we see that the bending moment at distance x from the base is
(11-18)
M P(d v)
x
Pcr
P
Pcr
d
B
x
d
B
B
v
L
x
A
y
FIG. 11-15 Ideal column fixed at the base
and free at the top: (a) initially straight
column, (b) buckled shape for
n 1, (c) buckled shape for n 3,
and (d) buckled shape for n 5
A
p 2EI
Pcr = —
4L2
(a)
(b)
Pcr
y
d
L
—
3
d
L
—
3
A
L
—
3
9p 2EI
Pcr = —
4L2
(c)
x
d
B
d
L
—
5
d
A
y
25p 2EI
Pcr = —
4L2
(d)
*This column is of special interest because it is the one first analyzed by Euler in 1744.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
L
—
5
766
CHAPTER 11 Columns
where d is the deflection at the free end of the column. The differential
equation of the deflection curve then becomes
EIv M P(d v)
(11-19)
in which I is the moment of inertia for buckling in the xy plane.
Using the notation k2 P/EI (Eq. 11-6a), we can rearrange Eq.
(11-19) into the form
v k 2v k 2d
(11-20)
which is a linear differential equation of second order with constant
coefficients. However, it is a more complicated equation than the equation for a column with pinned ends (see Eq. 11-7) because it has a
nonzero term on the right-hand side.
The general solution of Eq. (11-20) consists of two parts: (1) the
homogeneous solution, which is the solution of the homogeneous equation obtained by replacing the right-hand side with zero, and (2) the
particular solution, which is the solution of Eq. (11-20) that produces
the term on the right-hand side.
The homogeneous solution (also called the complementary solution)
is the same as the solution of Eq. (11-7); hence
vH C1 sin kx C2 cos kx
(a)
where C1 and C2 are constants of integration. Note that when vH is substituted into the left-hand side of the differential equation (Eq. 11-20), it
produces zero.
The particular solution of the differential equation is
vP d
(b)
When vP is substituted into the left-hand side of the differential equation, it produces the right-hand side, that is, it produces the term k2d.
Consequently, the general solution of the equation, equal to the sum of
vH and vP, is
v C1 sin kx C2 cos kx d
(11-21)
This equation contains three unknown quantities (C1, C2, and d),
and therefore three boundary conditions are needed to complete the
solution.
At the base of the column, the deflection and slope are each equal to
zero. Therefore, we obtain the following boundary conditions:
v(0) 0
v (0) 0
Applying the first condition to Eq. (11-21), we find
C2 d
(c)
To apply the second condition, we first differentiate Eq. (11-21) to
obtain the slope:
v C1k cos kx – C2k sin kx
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(d)
SECTION 11.4 Columns With Other Support Conditions
767
Applying the second condition to this equation, we find C1 0.
Now we can substitute the expressions for C1 and C2 into the general solution (Eq. 11-21) and obtain the equation of the deflection
curve for the buckled column:
v d (1 cos kx)
(11-22)
Note that this equation gives only the shape of the deflection curve—the
amplitude d remains undefined. Thus, when the column buckles, the
deflection given by Eq. (11-22) may have any arbitrary magnitude,
except that it must remain small (because the differential equation is
based upon small deflections).
The third boundary condition applies to the upper end of the
column, where the deflection v is equal to d:
v(L) d
Using this condition with Eq. (11-22), we get
d cos kL 0
(11-23)
From this equation we conclude that either d 0 or cos kL 0. If
d 0, there is no deflection of the bar (see Eq. 11-22) and we have the
trivial solution—the column remains straight and buckling does not
occur. In that case, Eq. (11-23) will be satisfied for any value of the
quantity kL, that is, for any value of the load P. This conclusion is represented by the vertical line in the load-deflection diagram of Fig. 11-7.
The other possibility for solving Eq. (11-23) is
cos kL 0
(11-24)
which is the buckling equation. In this case, Eq. (11-23) is satisfied
regardless of the value of the deflection d. Thus, as already observed, d
is undefined and may have any small value.
The equation cos kL 0 is satisfied when
np
kL
2
n 1, 3, 5, . . .
(11-25)
Using the expression k2 P/EI, we obtain the following formula for the
critical loads:
n2p 2EI
Pcr
4L2
n 1, 3, 5, . . .
(11-26)
Also, the buckled mode shapes are obtained from Eq. (11-22):
冢
冣
np x
v d 1 cos
2L
n 1, 3, 5, . . .
(11-27)
The lowest critical load is obtained by substituting n 1 in
Eq. (11-26):
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
768
CHAPTER 11 Columns
p 2EI
Pcr
4L2
(11-28)
The corresponding buckled shape (from Eq. 11-27) is
冢
P
冣
px
(11-29)
v d 1 cos
2L
and is shown in Fig. 11-15b.
By taking higher values of the index n, we can theoretically obtain
an infinite number of critical loads from Eq. (11-26). The corresponding
buckled mode shapes have additional waves in them. For instance, when
n 3 the buckled column has the shape shown in Fig. 11-15c and Pcr is
nine times larger than for n 1. Similarly, the buckled shape for n 5
has even more waves (Fig. 11-15d) and the critical load is twenty-five
times larger.
P
Effective Lengths of Columns
L
Le = 2L
(a)
P
(b)
FIG. 11-16 Deflection curves showing the
effective length Le for a column fixed at
the base and free at the top
The critical loads for columns with various support conditions can be
related to the critical load of a pinned-end column through the concept
of an effective length. To demonstrate this idea, consider the deflected
shape of a column fixed at the base and free at the top (Fig. 11-16a).
This column buckles in a curve that is one-quarter of a complete sine
wave. If we extend the deflection curve (Fig. 11-16b), it becomes onehalf of a complete sine wave, which is the deflection curve for a
pinned-end column.
The effective length Le for any column is the length of the equivalent pinned-end column, that is, it is the length of a pinned-end column
having a deflection curve that exactly matches all or part of the deflection curve of the original column.
Another way of expressing this idea is to say that the effective
length of a column is the distance between points of inflection (that is,
points of zero moment) in its deflection curve, assuming that the curve
is extended (if necessary) until points of inflection are reached. Thus, for
a fixed-free column (Fig. 11-16), the effective length is
Le 2L
(11-30)
Because the effective length is the length of an equivalent pinnedend column, we can write a general formula for critical loads as follows:
p 2EI
Pcr
L 2e
(11-31)
If we know the effective length of a column (no matter how complex the
end conditions may be), we can substitute into the preceding equation
and determine the critical load. For instance, in the case of a fixed-free
column, we can substitute Le 2L and obtain Eq. (11-28).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
769
SECTION 11.4 Columns With Other Support Conditions
The effective length is often expressed in terms of an effectivelength factor K:
(11-32)
Le KL
where L is the actual length of the column. Thus, the critical load is
p 2EI
Pcr
(KL)2
(11-33)
The factor K equals 2 for a column fixed at the base and free at the top
and equals 1 for a pinned-end column. The effective-length factor is
often included in design formulas for columns, as illustrated later in
Section 11.9.
Column with Both Ends Fixed Against Rotation
Next, let us consider a column with both ends fixed against rotation
(Fig. 11-17a). Note that in this figure we use the standard symbol for the
fixed support at the base of the column. However, since the column is
free to shorten under an axial load, we must introduce a new symbol at
the top of the column. This new symbol shows a rigid block that is constrained in such a manner that rotation and horizontal displacement are
prevented but vertical movement can occur. (As a convenience when
drawing sketches, we often replace this more accurate symbol with the
standard symbol for a fixed support—see Fig. 11-17b—with the understanding that the column is free to shorten.)
The buckled shape of the column in the first mode is shown in
Fig. 11-17c. Note that the deflection curve is symmetrical (with zero
slope at the midpoint) and has zero slope at the ends. Because rotation at
x
P
P
P
M0
L
—
4
B
B
L
A
L
Le = —
2
L
A
L
—
4
y
M0
FIG. 11-17 Buckling of a column with
(a)
P
(b)
both ends fixed against rotation
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(c)
770
CHAPTER 11 Columns
the ends is prevented, reactive moments M0 develop at the supports.
These moments, as well as the reactive force at the base, are shown in
the figure.
From our previous solutions of the differential equation, we know
that the equation of the deflection curve involves sine and cosine functions. Also, we know that the curve is symmetric about the midpoint.
Therefore, we see immediately that the curve must have inflection points
at distances L/4 from the ends. It follows that the middle portion of the
deflection curve has the same shape as the deflection curve for a
pinned-end column. Thus, the effective length of a column with fixed
ends, equal to the distance between inflection points, is
L
Le
2
(11-34)
Substituting into Eq. (11-31) gives the critical load:
4p 2EI
Pcr
L2
(11-35)
This formula shows that the critical load for a column with fixed ends is
four times that for a column with pinned ends. As a check, this result
may be verified by solving the differential equation of the deflection
curve (see Problem 11.4-9).
Column Fixed at the Base and Pinned at the Top
The critical load and buckled mode shape for a column that is fixed at
the base and pinned at the top (Fig. 11-18a) can be determined by solving the differential equation of the deflection curve. When the column
buckles (Fig. 11-18b), a reactive moment M0 develops at the base
because there can be no rotation at that point. Then, from the equilibrium of the entire column, we know that there must be horizontal
reactions R at each end such that
M0 RL
(e)
The bending moment in the buckled column, at distance x from the base,
is
M M0 Pv Rx Pv R(L x)
(11-36)
and therefore the differential equation is
EIv M Pv R(L x)
(11-37)
Again substituting k P/EI and rearranging, we get
2
R
v k 2v (L x)
EI
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(11-38)
SECTION 11.4 Columns With Other Support Conditions
771
x
P
20.19 EI
Pcr = —
L2
P
R
B
B
B
v
Le = 0.699L
L
A
R
A
y
A
M0
FIG. 11-18 Column fixed at the base and
pinned at the top
P
(a)
(c)
(b)
The general solution of this equation is
R
v C1 sin kx C2 cos kx (L x)
P
(11-39)
in which the first two terms on the right-hand side constitute the homogeneous solution and the last term is the particular solution. This
solution can be verified by substitution into the differential equation (Eq.
11-37).
Since the solution contains three unknown quantities (C1, C2, and
R), we need three boundary conditions. They are
v(0) 0
v (0) 0
v(L) 0
Applying these conditions to Eq. (11-39) yields
RL
C2 0
P
R
C1k 0
P
C1 tan kL C2 0
(f,g,h)
All three equations are satisfied if Cl C2 R 0, in which case we
have the trivial solution and the deflection is zero.
To obtain the solution for buckling, we must solve Eqs. (f), (g), and
(h) in a more general manner. One method of solution is to eliminate R
from the first two equations, which yields
C1 kL C2 0 or C2 C1kL
(i)
Next, we substitute this expression for C2 into Eq. (h) and obtain the
buckling equation:
kL tan kL
The solution of this equation gives the critical load.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(11-40)
772
CHAPTER 11 Columns
Since the buckling equation is a transcendental equation, it cannot
be solved explicitly.* Nevertheless, the values of kL that satisfy the
equation can be determined numerically by using a computer program
for finding roots of equations. The smallest nonzero value of kL that satisfies Eq. (11-40) is
kL 4.4934
(11-41)
The corresponding critical load is
20.19EI
2.046p 2EI
Pcr
2
L
L2
(11-42)
which (as expected) is higher than the critical load for a column with
pinned ends and lower than the critical load for a column with fixed
ends (see Eqs. 11-12 and 11-35).
The effective length of the column may be obtained by comparing
Eqs. (11-42) and (11-31); thus,
Le 0.699L ⬇ 0.7L
(11-43)
This length is the distance from the pinned end of the column to the
point of inflection in the buckled shape (Fig. 11-18c).
The equation of the buckled mode shape is obtained by substituting C2 C1kL (Eq. i) and R/P C1k (Eq. g) into the general solution
(Eq. 11-39):
v C1[sin kx kL cos kx k(L x)]
(11-44)
in which k 4.4934/L. The term in brackets gives the mode shape for
the deflection of the buckled column. However, the amplitude of the
deflection curve is undefined because C1 may have any value (within the
usual limitation that the deflections must remain small).
Limitations
In addition to the requirement of small deflections, the Euler buckling
theory used in this section is valid only if the column is perfectly
straight before the load is applied, the column and its supports have no
imperfections, and the column is made of a linearly elastic material that
follows Hooke’s law. These limitations were explained previously in
Section 11.3.
*In a transcendental equation, the variables are contained within transcendental functions. A transcendental function cannot be expressed by a finite number of algebraic
operations; hence trigonometric, logarithmic, exponential, and other such functions are
transcendental.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
773
SECTION 11.4 Columns With Other Support Conditions
Summary of Results
The lowest critical loads and corresponding effective lengths for the four
columns we have analyzed are summarized in Fig. 11-19.
(a) Pinned-pinned column
(b) Fixed-free column
p 2 EI
Pcr = —
L2
(c) Fixed-fixed column
4p 2 EI
Pcr = —
L2
p 2 EI
Pcr = —
4L2
(d) Fixed-pinned column
2.046 p 2 EI
Pcr = —
L2
Le
L
L
Le
L
L
Le = L
Le = 2 L
Le = 0.5L
Le = 0.699L
K= 1
K= 2
K = 0.5
K = 0.699
FIG. 11-19 Critical loads, effective
lengths, and effective-length factors for
ideal columns
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
774
CHAPTER 11 Columns
Example 11-2
A viewing platform in a wild-animal park (Fig. 11-20a) is supported by a row of
aluminum pipe columns having length L 3.25 m and outer diameter d
100 mm. The bases of the columns are set in concrete footings and the tops of
the columns are supported laterally by the platform. The columns are being
designed to support compressive loads P 100 kN.
Determine the minimum required thickness t of the columns (Fig. 11-20b)
if a factor of safety n 3 is required with respect to Euler buckling. (For the
aluminum, use 72 GPa for the modulus of elasticity and use 480 MPa for the
proportional limit.)
d
t
d
(b)
L
FIG. 11-20 Example 11-2. Aluminum
(a)
pipe column
Solution
Critical load. Because of the manner in which the columns are constructed,
we will model each column as a fixed-pinned column (see Fig. 11-19d). Therefore, the critical load is
2.046p 2EI
Pcr
L2
( j)
in which I is the moment of inertia of the tubular cross section:
p
I 冤d 4 (d 2t)4冥
64
(k)
Substituting d 100 mm (or 0.1 m), we get
p
I 冤(0.1 m)4 (0.1 m 2t)4冥
64
in which t is expressed in meters.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(l)
SECTION 11.4 Columns With Other Support Conditions
775
Required thickness of the columns. Since the load per column is 100 kN
and the factor of safety is 3, each column must be designed for the following
critical load:
Pcr nP 3(100 kN) 300 kN
Substituting this value for Pcr in Eq. ( j), and also replacing I with its expression
from Eq. (l), we obtain
冢 冣
2.046p 2(72 10 9 Pa) p
300,000 N 冤(0.1 m)4 (0.1 m 2t)4冥
64
Note that all terms in this equation are expressed in units of newtons and meters.
After multiplying and dividing, the preceding equation simplifies to
44.40
or
6
10
m4 (0.1 m)4 (0.1 m 2t)4
(0.1 m 2t)4 (0.1 m)4 44.40
6
10
m4 55.60
6
10
m4
from which we obtain
0.1 m 2t 0.08635 m
and
t 0.006825 m
Therefore, the minimum required thickness of the column to meet the specified
conditions is
tmin 6.83 mm
Supplementary calculations. Knowing the diameter and thickness of the
column, we can now calculate its moment of inertia, cross-sectional area, and
radius of gyration. Using the minimum thickness of 6.83 mm, we obtain
p
I 冤d 4 (d 2t)4冥 2.18
64
p
A 冤d 2 (d 2t)2冥 1999 mm2
4
106 mm4
r
冪莦AI 33.0 mm
The slenderness ratio L/r of the column is approximately 98, which is in the
customary range for slender columns, and the diameter-to-thickness ratio d/t is
approximately 15, which should be adequate to prevent local buckling of the
walls of the column.
The critical stress in the column must be less than the proportional limit of
the aluminum if the formula for the critical load (Eq. j) is to be valid. The critical stress is
Pcr
300 kN
150 MPa
scr
A
1999 mm2
which is less than the proportional limit (480 MPa). Therefore, our calculation
for the critical load using Euler buckling theory is satisfactory.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
776
CHAPTER 11 Columns
11.5 COLUMNS WITH ECCENTRIC AXIAL LOADS
x
P
P
e
M0 = Pe
B
v
L
y
A
e
P
(a)
P
M0 = Pe
(b)
(11-45)
M M0 P(v) Pe Pv
FIG. 11-21 Column with eccentric axial
loads
In Sections 11.3 and 11.4 we analyzed ideal columns in which the axial
loads acted through the centroids of the cross sections. Under these conditions, the columns remain straight until the critical loads are reached,
after which bending may occur.
Now we will assume that a column is compressed by loads P that
are applied with a small eccentricity e measured from the axis of the
column (Fig. 11-21a). Each eccentric axial load is equivalent to a centric
load P and a couple of moment M0 Pe (Fig. 11-21b). This moment
exists from the instant the load is first applied, and therefore the column
begins to deflect at the onset of loading. The deflection then becomes
steadily larger as the load increases.
To analyze the pin-ended column shown in Fig. 11-21, we make the
same assumptions as in previous sections; namely, the column is initially perfectly straight, the material is linearly elastic, and the xy plane
is a plane of symmetry. The bending moment in the column at distance x
from the lower end (Fig. 11-21b) is
where v is the deflection of the column (positive when in the positive
direction of the y axis). Note that the deflections of the column are negative when the eccentricity of the load is positive.
The differential equation of the deflection curve is
EIv M Pe Pv
(11-46)
v k 2v k 2e
(11-47)
or
in which k P/EI, as before. The general solution of this equation is
2
v C1 sin kx C2 cos kx e
(11-48)
in which C1 and C2 are constants of integration in the homogeneous
solution and e is the particular solution. As usual, we can verify the solution by substituting it into the differential equation.
The boundary conditions for determining the constants C1 and
C2 are obtained from the deflections at the ends of the column
(Fig. 11-21b):
v(0) 0
v(L) 0
These conditions yield
C2 e
e(1 cos kL)
kL
C1 e tan
sin kL
2
Therefore, the equation of the deflection curve is
冢
kL
v e tan sin kx cos kx 1
2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
冣
(11-49)
SECTION 11.5 Columns With Eccentric Axial Loads
777
For a column with known loads P and known eccentricity e, we can use
this equation to calculate the deflection at any point along the x axis.
The behavior of a column with an eccentric load is quite different
from that of a centrally loaded column, as can be seen by comparing Eq.
(11-49) with Eqs. (11-13), (11-27), and (11-44). Equation (11-49) shows
that each value of the eccentric load P produces a definite value of the
deflection, just as each value of the load on a beam produces a definite
deflection. In contrast, the deflection equations for centrally loaded
columns give the buckled mode shape (when P Pcr) but with the
amplitude undefined.
Because the column shown in Fig. 11-21 has pinned ends, its critical load (when centrally loaded) is
p 2EI
Pcr
L2
(11-50)
We will use this formula as a reference quantity in some of the equations
that follow.
Maximum Deflection
P
The maximum deflection d produced by the eccentric loads occurs at the
midpoint of the column (Fig. 11-22) and is obtained by setting x equal to
L/2 in Eq. (11-49):
e
冢冣 冢
d
or, after simplifying,
冢
L
—
2
P
冣
L
kL
kL
kL
d v e tan sin cos 1
2
2
2
2
L
—
2
冣
kL
d e sec 1
2
(11-51)
This equation can be written in a slightly different form by replacing
the quantity k with its equivalent in terms of the critical load (see
Eq. 11-50):
e
FIG. 11-22 Maximum deflection d of a
column with eccentric axial loads
k
冪莦 冪莦
P
EI
Pp 2
p
2
PcrL
L
P
冪莦
P
(11-52)
cr
Thus, the nondimensional term kL becomes
kL p
P
冪莦
P
(11-53)
cr
and Eq. (11-51) for the maximum deflection becomes
P
1冥
冤 冢 冪莦
P 冣
p
d e sec
2
(11-54)
cr
As special cases, we note the following: (1) The deflection d is zero
when the eccentricity e is zero and P is not equal to Pcr, (2) the deflection
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
778
CHAPTER 11 Columns
P
Pcr
e=0
e=0
e = e1
e = e2
e2 e1 0
O
d
FIG. 11-23 Load-deflection diagram for a
column with eccentric axial loads (see
Fig. 11-22 and Eq. 11-54)
is zero when the axial load P is zero, and (3) the deflection becomes
infinitely large as P approaches Pcr. These characteristics are shown in
the load-deflection diagram of Fig. 11-23.
To plot the load-deflection diagram, we select a particular value e1
of the eccentricity and then calculate d for various values of the load P.
The resulting curve is labeled e e1 in Fig. 11-23. We note immediately
that the deflection d increases as P increases, but the relationship is nonlinear. Therefore, we cannot use the principle of superposition for
calculating deflections due to more than one load, even though the
material of the column is linearly elastic. As an example, the deflection
due to an axial load 2P is not equal to twice the deflection caused by an
axial load P.
Additional curves, such as the curve labeled e e2, are plotted in a
similar manner. Since the deflection d is linear with e (Eq. 11-54), the
curve for e e2 has the same shape as the curve for e e1 but the
abscissas are larger by the ratio e2/e1.
As the load P approaches the critical load, the deflection d increases
without limit and the horizontal line corresponding to P Pcr becomes
an asymptote for the curves. In the limit, as e approaches zero, the
curves on the diagram approach two straight lines, one vertical and one
horizontal (compare with Fig. 11-7). Thus, as expected, an ideal column
with a centrally applied load (e 0) is the limiting case of a column
with an eccentric load (e 0).
Although the curves plotted in Fig. 11-23 are mathematically
correct, we must keep in mind that the differential equation is valid only
for small deflections. Therefore, when the deflections become large, the
curves are no longer physically valid and must be modified to take into
account the presence of large deflections and (if the proportional limit of
the material is exceeded) inelastic bending effects (see Fig. 11-11).
The reason for the nonlinear relationship between loads and deflections, even when the deflections are small and Hooke’s law holds, can
be understood if we observe once again that the axial loads P are equivalent to centrally applied loads P plus couples Pe acting at the ends of the
column (Fig. 11-21b). The couples Pe, if acting alone, would produce
bending deflections of the column in the same manner as for a beam. In
a beam, the presence of the deflections does not change the action of the
loads, and the bending moments are the same whether the deflections
exist or not. However, when an axial load is applied to the member, the
existence of deflections increases the bending moments (the increases
are equal to the product of the axial load and the deflections). When the
bending moments increase, the deflections are further increased—hence
the moments increase even more, and so on. Thus, the bending moments
in a column depend upon the deflections, which in turn depend upon the
bending moments. This type of behavior results in a nonlinear relationship between the axial loads and the deflections.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 11.5 Columns With Eccentric Axial Loads
779
In general, a straight structural member subjected to both bending
loads and axial compressive loads is called a beam-column. In the case
of a column with eccentric loads (Fig. 11-21), the bending loads are the
moments M0 Pe and the axial loads are the forces P.
Maximum Bending Moment
The maximum bending moment in an eccentrically loaded column
occurs at the midpoint where the deflection is a maximum (Fig. 11-22):
(11-55)
Mmax P(e d)
Substituting for d from Eqs. (11-51) and (11-54), we obtain
P
冢 冪莦
P 冣
kL
p
Mmax Pe sec Pe sec
2
2
Mmax
Pe
O
Pcr
P
FIG. 11-24 Maximum bending moment in
(11-56)
cr
The manner in which Mmax varies as a function of the axial load P is
shown in Fig. 11-24.
When P is small, the maximum moment is equal to Pe, which
means that the effect of the deflections is negligible. As P increases, the
bending moment grows nonlinearly and theoretically becomes infinitely
large as P approaches the critical load. However, as explained before,
our equations are valid only when the deflections are small, and they
cannot be used when the axial load approaches the critical load. Nevertheless, the preceding equations and accompanying graphs indicate the
general behavior of beam-columns.
a column with eccentric axial loads (see
Fig. 11-22 and Eq. 11-56)
Other End Conditions
The equations given in this section were derived for a pinned-end column, as shown in Figs. 11-21 and 11-22. If a column is fixed at the base
and free at the top (Fig. 11-19b), we can use Eqs. 11-51 and 11-56 by
replacing the actual length L with the equivalent length 2L (see Problem
11.5-9). However, the equations do not apply to a column that is fixed at
the base and pinned at the top (Fig. 11-19d). The use of an equivalent
length equal to 0.699L gives erroneous results; instead, we must return
to the differential equation and derive a new set of equations.
In the case of a column with both ends fixed against rotation (Fig.
11-19c), the concept of an eccentric axial load acting at the end of the
column has no meaning. Any moment applied at the end of the column
is resisted directly by the supports and produces no bending of the
column itself.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
780
CHAPTER 11 Columns
Example 11-3
L
A
B
h
Brass bar
b
e
A brass bar AB projecting from the side of a large machine is loaded at end B by
a force P 1500 lb acting with an eccentricity e 0.45 in. (Fig. 11-25). The
bar has a rectangular cross section with height h 1.2 in. and width b 0.6 in.
What is the longest permissible length Lmax of the bar if the deflection at
the end is limited to 0.12 in.? (For the brass, use E 16 106 psi.)
P
Solution
FIG. 11-25 Example 11-3. Brass bar with
an eccentric axial load
Critical load. We will model this bar as a slender column that is fixed at
end A and free at end B. Therefore, the critical load (see Fig. 11-19b) is
p 2EI
Pcr
4L2
(a)
The moment of inertia for the axis about which bending occurs is
hb3
(1.2 in.)(0.6 in.)3
I 0.02160 in.4
12
12
Therefore, the expression for the critical load becomes
852,700 lb-in.2
p 2(16,000,000 psi)(0.02160 in.4)
Pcr
L2
(b)
in which Pcr has units of pounds and L has units of inches.
Deflection. The deflection at the end of the bar is given by Eq. (11-54),
which applies to a fixed-free column as well as a pinned-end column:
P
1冥
冤 冢 冪莦
P 冣
p
d e sec
2
(c)
cr
In this equation, Pcr is given by Eq. (a).
Length. To find the maximum permissible length of the bar, we substitute
for d its limiting value of 0.12 in. Also, we substitute e 0.45 in. and P
1500 lb, and we substitute for Pcr from Eq. (b). Thus,
1500 lb
1冥
冤 冢 冪
莦
852,70莦冣
0/L
p
0.12 in. (0.45 in.) sec
2
2
The only unknown in this equation is the length L (inches). To solve for L,
we perform the various arithmetic operations in the equation and then rearrange
the terms. The result is
0.2667 sec (0.06588 L) 1
Using radians and solving this equation, we get L 10.03 in. Thus, the maximum permissible length of the bar is
Lmax 10.0 in.
If a longer bar is used, the deflection will exceed the allowable value of 0.12 in.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 11.6 The Secant Formula for Columns
781
11.6 THE SECANT FORMULA FOR COLUMNS
In the preceding section we determined the maximum deflection and
maximum bending moment in a pin-ended column subjected to eccentric axial loads. In this section, we will investigate the maximum stresses
in the column and obtain a special formula for calculating them.
The maximum stresses in a column with eccentric axial loads occur
at the cross section where the deflection and bending moment have their
largest values, that is, at the midpoint (Fig. 11-26a). Acting at this cross
section are the compressive force P and the bending moment Mmax
(Fig. 11-26b). The stresses due to the force P are equal to P/A, where A
is the cross-sectional area of the column, and the stresses due to the
bending moment Mmax are obtained from the flexure formula.
Thus, the maximum compressive stress, which occurs on the concave side of the column, is
P Mmaxc
smax
A
I
(11-57)
in which I is the moment of inertia in the plane of bending and c is the
distance from the centroidal axis to the extreme point on the concave
side of the column. Note that in this equation we consider compressive
stresses to be positive, since these are the important stresses in a column.
The bending moment Mmax is obtained from Eq. (11-56), which is
repeated here:
P
冢 冪莦
P 冣
p
Mmax Pe sec
2
P
cr
e
L
—
2
d
P
Mmax
L
—
2
FIG. 11-26 Column with eccentric axial
loads
P
e
(a)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
L
—
2
P
e
(b)
782
CHAPTER 11 Columns
Since Pcr p 2EI/L2 for a pinned-end column, and since I Ar 2, where
r is the radius of gyration in the plane of bending, the preceding equation becomes
P
冢 冪莦
EA 冣
L
Mmax Pe sec
2r
(11-58)
Substituting into Eq. (11-57), we obtain the following formula for the
maximum compressive stress:
P
冢 冪莦
EA 冣
P Pec
L
smax sec
A
I
2r
or
冤
P
冢 冪莦
EA 冣冥
P
ec
L
smax 1 2 sec
A
r
2r
(11-59)
This equation is commonly known as the secant formula for an eccentrically loaded column with pinned ends.
The secant formula gives the maximum compressive stress in the
column as a function of the average compressive stress P/A, the modulus
of elasticity E, and two nondimensional ratios—the slenderness ratio L/r
(Eq. 11-17) and the eccentricity ratio:
ec
Eccentricity ratio
r2
(11-60)
As the name implies, the eccentricity ratio is a measure of the eccentricity of the load as compared to the dimensions of the cross section. Its
numerical value depends upon the position of the load, but typical values are in the range from 0 to 3, with the most common values being
less than 1.
When analyzing a column, we can use the secant formula to calculate the maximum compressive stress whenever the axial load P and its
eccentricity e are known. Then the maximum stress can be compared
with the allowable stress to determine if the column is adequate to
support the load.
We can also use the secant formula in the reverse manner, that is, if
we know the allowable stress, we can calculate the corresponding value
of the load P. However, because the secant formula is transcendental, it
is not practical to derive a formula for the load P. Instead, we can solve
Eq. (11-59) numerically in each individual case.
A graph of the secant formula is shown in Fig. 11-27. The abscissa
is the slenderness ratio L/r, and the ordinate is the average compressive
stress P/A. The graph is plotted for a steel column with modulus of elasticity E 30 103 ksi and maximum stress smax 36 ksi. Curves are
plotted for several values of the eccentricity ratio ec/r 2. These curves are
valid only when the maximum stress is less than the proportional limit of
the material, because the secant formula was derived using Hooke’s law.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 11.6 The Secant Formula for Columns
40
ec
—2 = 0
r
s max= 36 ksi
E = 30 103 ksi
0.1
0.2
30
P
— (ksi)
A
Euler’s curve
0.4
20
783
0.6
0.8
1.0
1.5
10
FIG. 11-27 Graph of the secant formula
0
50
100
(Eq. 11-59) for smax 36 ksi and
E 30 103 ksi
150
200
L
—
r
A special case arises when the eccentricity of the load disappears
(e 0), because then we have an ideal column with a centrally applied
load. Under these conditions the maximum load is the critical load
(Pcr p 2EI/L2) and the corresponding maximum stress is the critical
stress (see Eqs. 11-14 and 11-16):
Pcr
p 2EI
p 2E
scr
2
AL
A
(L/r)2
(11-61)
Since this equation gives the stress P/A in terms of the slenderness ratio
L /r, we can plot it on the graph of the secant formula (Fig. 11-27) as
Euler’s curve.
Let us now assume that the proportional limit of the material is the
same as the selected maximum stress, that is, 36 ksi. Then we construct
a horizontal line on the graph at a value of 36 ksi, and we terminate
Euler’s curve at that stress. The horizontal line and Euler’s curve
represent the limits of the secant-formula curves as the eccentricity e
approaches zero.
Discussion of the Secant Formula
The graph of the secant formula shows that the load-carrying capacity of
a column decreases significantly as the slenderness ratio L /r increases,
especially in the intermediate region of L/r values. Thus, long slender
columns are much less stable than short, stocky columns. The graph also
shows that the load-carrying capacity decreases with increasing eccentricity e; furthermore, this effect is relatively greater for short columns
than for long ones.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
784
CHAPTER 11 Columns
The secant formula was derived for a column with pinned ends, but
it can also be used for a column that is fixed at the base and free at the
top. All that is required is to replace the length L in the secant formula
with the equivalent length 2L. However, because it is based upon
Eq. (11-56), the secant formula is not valid for the other end conditions
that we discussed.
Now let us consider an actual column, which inevitably differs from
an ideal column because of imperfections, such as initial curvature of
the longitudinal axis, imperfect support conditions, and nonhomogeneity
of the material. Furthermore, even when the load is supposed to be centrally applied, there will be unavoidable eccentricities in its direction and
point of application. The extent of these imperfections varies from one
column to another, and therefore there is considerable scatter in the
results of laboratory tests performed with actual columns.
All imperfections have the effect of producing bending in addition
to direct compression. Therefore, it is reasonable to assume that the
behavior of an imperfect, centrally loaded column is similar to that of an
ideal, eccentrically loaded column. In such cases, the secant formula can
be used by choosing an approximate value of the eccentricity ratio ec/r 2
to account for the combined effects of the various imperfections. For
instance, a commonly used value of the eccentricity ratio for pinned-end
columns in structural-steel design is ec/r 2 0.25. The use of the secant
formula in this manner for columns with centrally applied loads provides a rational means of accounting for the effects of imperfections,
rather than accounting for them simply by increasing the factor of safety.
(For further discussions of the secant formula and the effects of imperfections, see Ref. 11-5 and textbooks on buckling and stability.)
The procedure for analyzing a centrally loaded column by means of
the secant formula depends upon the particular conditions. For instance,
if the objective is to determine the allowable load, the procedure is as
follows. Assume a value of the eccentricity ratio ec/r 2 based upon test
results, code values, or practical experience. Substitute this value into
the secant formula, along with the values of L/r, A, and E for the actual
column. Assign a value to smax, such as the yield stress sY or the proportional limit spl of the material. Then solve the secant formula for the
load Pmax that produces the maximum stress. (This load will always be
less than the critical load Pcr for the column.) The allowable load on the
column equals the load Pmax divided by the factor of safety n.
The following example illustrates how the secant formula may be
used to determine the maximum stress in a column when the load is
known, and also how to determine the load when the maximum stress is
given.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 11.6 The Secant Formula for Columns
785
Example 11-4
A steel wide-flange column of W 14 82 shape (Fig. 11-28a) is pin-supported
at the ends and has a length of 25 ft. The column supports a centrally applied
load P1 320 k and an eccentrically applied load P2 40 k (Fig. 11-28b).
Bending takes place about axis 1-1 of the cross section, and the eccentric load
acts on axis 2-2 at a distance of 13.5 in. from the centroid C.
(a) Using the secant formula, and assuming E 30,000 ksi, calculate the
maximum compressive stress in the column.
(b) If the yield stress for the steel is sY 42 ksi, what is the factor of safety
with respect to yielding?
1
W 14
2
82
C
2
1
(a)
Solution
P1 = 320 k
C
13.5 in.
P2 = 40 k
Bracket
Column
(b)
(a) Maximum compressive stress. The two loads P1 and P2 acting as shown
in Fig. 11-28b are statically equivalent to a single load P 360 k acting with an
eccentricity e 1.5 in. (Fig. 11-28c). Since the column is now loaded by a
single force P having an eccentricity e, we can use the secant formula to find the
maximum stress.
The required properties of the W 14 82 wide-flange shape are obtained
from Table E-1 in Appendix E:
A 24.1 in.2
r 6.05 in.
14.31 in.
c 7.155 in.
2
The required terms in the secant formula (Eq. 11-59) are calculated as follows:
e
P
360 k
2 14.94 ksi
A
24.1 in.
P = 360 k
ec
(1.5 in.)(7.155 in.)
0.2932
r2
(6.05 in.)2
L
(25 ft)(12 in./ft)
49.59
r
6.05 in.
P
360 k
497.9
EA
(c)
FIG. 11-28 Example 11-4. Column with
an eccentrically applied axial load
6
10
Substituting these values into the secant formula, we get
冤
P
冢 冪莦
E A 冣冥
P
ec
L
smax 1 2 sec
A
r
2r
(14.94 ksi)(1 0.345) 20.1 ksi
This compressive stress occurs at midheight of the column on the concave side
(the right-hand side in Fig. 11-28b).
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
786
CHAPTER 11 Columns
(b) Factor of safety with respect to yielding. To find the factor of safety, we
need to determine the value of the load P, acting at the eccentricity e, that will
produce a maximum stress equal to the yield stress sY 42 ksi. Since this value
of the load is just sufficient to produce initial yielding of the material, we will
denote it as PY.
Note that we cannot determine PY by multiplying the load P (equal to
360 k) by the ratio sY/smax. The reason is that we are dealing with a nonlinear
relationship between load and stress. Instead, we must substitute smax
sY 42 ksi in the secant formula and then solve for the corresponding load P,
which becomes PY. In other words, we must find the value of PY that satisfies
the following equation:
冢 冪莦
EA 冣冥
冤
PY
ec
L
sec
sY 1
A
r2
2r
PY
(11-62)
Substituting numerical values, we obtain
冤
冢
PY
49.59
42 ksi 2 1 0.2939 sec
2
24.1 in.
P
冪莦莦冣冥
Y
or
苶Y冣冥
1012 k PY 冤1 0.2939 sec 冢0.02916 兹P
in which PY has units of kips. Solving this equation numerically, we get
PY 716 k
This load will produce yielding of the material (in compression) at the cross
section of maximum bending moment.
Since the actual load is P 360 k, the factor of safety against yielding is
PY
716 k
n 1.99
360 k
P
This example illustrates two of the many ways in which the secant formula may
be used. Other types of analysis are illustrated in the problems at the end of the
chapter.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 11.7 Elastic and Inelastic Column Behavior
787
11.7 ELASTIC AND INELASTIC COLUMN BEHAVIOR
In the preceding sections we described the behavior of columns when
the material is stressed below the proportional limit. We began by
considering an ideal column subjected to a centrally applied load (Euler
buckling), and we arrived at the concept of a critical load Pcr. Then we
considered columns with eccentric axial loads and derived the secant
formula. We portrayed the results of these analyses on a diagram of
average compressive stress P/A versus the slenderness ratio L /r (see
Fig. 11-27). The behavior of an ideal column is represented in Fig. 11-27,
by Euler’s curve, and the behavior of columns with eccentric loads is
represented by the family of curves having various values of the eccentricity ratio ec/r 2.
We will now extend our discussion to include inelastic buckling,
that is, the buckling of columns when the proportional limit is exceeded.
We will portray the behavior on the same kind of diagram as before,
namely, a diagram of average compressive stress P/A versus slenderness
ratio L /r (see Fig. 11-29 on the next page). Note that Euler’s curve is
shown on this diagram as curve ECD. This curve is valid only in the
region CD where the stress is below the proportional limit spl of the
material. Therefore, the part of Euler’s curve above the proportional
limit is shown by a dashed line.
The value of slenderness ratio above which Euler’s curve is valid is
obtained by setting the critical stress (Eq. 11-61) equal to the proportional limit spl and solving for the slenderness ratio. Thus, letting (L /r)c
represent the critical slenderness ratio (Fig. 11-29), we get
p E
冢Lr冣 冪
莦
s
2
c
(11-63)
pl
As an example, consider structural steel with spl 36 ksi and
E 30,000 ksi. Then the critical slenderness ratio (L /r)c is equal to
90.7. Above this value, an ideal column buckles elastically and the Euler
load is valid. Below this value, the stress in the column exceeds the proportional limit and the column buckles inelastically.
If we take into account the effects of eccentricities in loading or
imperfections in construction, but still assume that the material follows
Hooke’s law, we obtain a curve such as the one labeled “Secant formula” in Fig. 11-29. This curve is plotted for a maximum stress smax
equal to the proportional limit spl.
When comparing the secant-formula curve with Euler’s curve, we
must keep in mind an important distinction. In the case of Euler’s curve,
the stress P/A not only is proportional to the applied load P but also is
the actual maximum stress in the column when buckling occurs. Consequently, as we move from C to D along Euler’s curve, both the
maximum stress P/A (equal to the critical stress) and the axial load P
decrease. However, in the case of the secant-formula curve, the average
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
788
CHAPTER 11 Columns
E
sult A
Strength
limit
B
P
—
A
Inelastic
stability
limit
C
spl
’
ler
Eu
formu
la fo
rs
rv
u
Short
columns
sc
Secant
ma
x=
Intermediate
columns
FIG. 11-29 Diagram of average com-
pressive stress P/A versus slenderness
ratio L/r
O
sp
e
Elastic
stability
limit
l
D
Long columns
(—Lr )
c
L
—
r
stress P/A decreases as we move from left to right along the curve (and
therefore the axial load P also decreases) but the maximum stress (equal
to the proportional limit) remains constant.
From Euler’s curve, we see that long columns with large slenderness ratios buckle at low values of the average compressive stress P/A.
This condition cannot be improved by using a higher-strength material,
because collapse results from instability of the column as a whole and
not from failure of the material itself. The stress can only be raised by
reducing the slenderness ratio L/r or by using a material with higher
modulus of elasticity E.
When a compression member is very short, it fails by yielding and
crushing of the material, and no buckling or stability considerations are
involved. In such a case, we can define an ultimate compressive stress
sult as the failure stress for the material. This stress establishes a
strength limit for the column, represented by the horizontal line AB in
Fig. 11-29. The strength limit is much higher than the proportional limit,
since it represents the ultimate stress in compression.
Between the regions of short and long columns, there is a range
of intermediate slenderness ratios too small for elastic stability to
govern and too large for strength considerations alone to govern. Such
an intermediate-length column fails by inelastic buckling, which means
that the maximum stresses are above the proportional limit when buckling occurs. Because the proportional limit is exceeded, the slope of the
stress-strain curve for the material is less than the modulus of elasticity;
hence the critical load for inelastic buckling is always less than the Euler
load (see Section 11.8).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 11.8 Inelastic Buckling
789
The dividing lines between short, intermediate, and long columns
are not precise. Nevertheless, it is useful to make these distinctions
because the maximum load-carrying capacity of columns in each category is based upon quite different types of behavior. The maximum
load-carrying capacity of a particular column (as a function of its length)
is represented by curve ABCD in Fig. 11-29. If the length is very small
(region AB), the column fails by direct compression; if the column is
longer (region BC), it fails by inelastic buckling; and if it is even longer
(region CD), it fails by elastic buckling (that is, Euler buckling). Curve
ABCD applies to columns with various support conditions if the length L
in the slenderness ratio is replaced by the effective length Le.
The results of load tests on columns are in reasonably good agreement with curve ABCD. When test results are plotted on the diagram,
they generally form a band that lies just below this curve. Considerable
scatter of test results is to be expected, because column performance is
sensitive to such matters as the accuracy of construction, the alignment
of loads, and the details of support conditions. To account for these
variables, we usually obtain the allowable stress for a column by
dividing the maximum stress (from curve ABCD) by a suitable factor of
safety, which often has a value of about 2. Because imperfections are
apt to increase with increase in length, a variable factor of safety
(increasing as L /r increases) is sometimes used. In Section 11.9 we will
give some typical formulas for allowable stresses.
11.8 INELASTIC BUCKLING
x
P
P
L
v
The critical load for elastic buckling is valid only for relatively long
columns, as explained previously (see curve CD in Fig. 11-29). If a
column is of intermediate length, the stress in the column will reach the
proportional limit before buckling begins (curve BC in Fig. 11-29). To
calculate critical loads in this intermediate range, we need a theory of
inelastic buckling. Three such theories are described in this section: the
tangent-modulus theory, the reduced-modulus theory, and the Shanley
theory.
Tangent-Modulus Theory
y
(a)
(b)
FIG. 11-30 Ideal column of intermediate
length that buckles inelastically
Let us again consider an ideal, pinned-end column subjected to an axial
force P (Fig. 11-30a). The column is assumed to have a slenderness ratio
L/r that is less than the critical slenderness ratio (Eq. 11-63), and therefore the axial stress P/A reaches the proportional limit before the critical
load is reached.
The compressive stress-strain diagram for the material of the
column is shown in Fig. 11-31. The proportional limit of the material is
indicated as spl, and the actual stress sA in the column (equal to P/A) is
represented by point A (which is above the proportional limit). If the
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
790
CHAPTER 11 Columns
Slope = Et
s
load is increased, so that a small increase in stress occurs, the relationship between the increment of stress and the corresponding increment of
strain is given by the slope of the stress-strain diagram at point A. This
slope, equal to the slope of the tangent line at A, is called the tangent
modulus and is denoted by Et; thus,
A
sA
spl
ds
Et
de
Slope = E
B
Slope = E
e
O
FIG. 11-31 Compression stress-strain
diagram for the material of the column
shown in Fig. 11-30
(11-64)
Note that the tangent modulus decreases as the stress increases beyond
the proportional limit. When the stress is below the proportional limit,
the tangent modulus is the same as the ordinary elastic modulus E.
According to the tangent-modulus theory of inelastic buckling, the
column shown in Fig. 11-30a remains straight until the inelastic critical
load is reached. At that value of load, the column may undergo a small
lateral deflection (Fig. 11-30b). The resulting bending stresses are superimposed upon the axial compressive stresses sA. Since the column starts
bending from a straight position, the initial bending stresses represent
only a small increment of stress. Therefore, the relationship between the
bending stresses and the resulting strains is given by the tangent modulus. Since the strains vary linearly across the cross section of the
column, the initial bending stresses also vary linearly, and therefore the
expressions for curvature are the same as those for linearly elastic bending except that Et replaces E:
1
d 2v
M
k
r
dx 2
Et I
(11-65)
(compare with Eqs. 9-5 and 9-7).
Because the bending moment M Pv (see Fig. 11-30b), the
differential equation of the deflection curve is
Et Iv Pv 0
(11-66)
This equation has the same form as the equation for elastic buckling
(Eq. 11-5) except that Et appears in place of E. Therefore, we can solve
the equation in the same manner as before and obtain the following
equation for the tangent-modulus load:
p 2Et I
Pt
L2
(11-67)
This load represents the critical load for the column according to the
tangent-modulus theory. The corresponding critical stress is
p 2Et
Pt
st 2
(L/r)
A
(11-68)
which is similar in form to Eq. (11-61) for the Euler critical stress.
Since the tangent modulus Et varies with the compressive stress
s P/A (Fig. 11-31), we usually obtain the tangent-modulus load by an
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 11.8 Inelastic Buckling
iterative procedure. We begin by estimating the value of Pt. This trial
value, call it P1, should be slightly larger than s pl A, which is the axial
load when the stress just reaches the proportional limit. Knowing P1, we
can calculate the corresponding axial stress s1 P1/A and determine the
tangent modulus Et from the stress-strain diagram. Next, we use Eq.
(11-67) to obtain a second estimate of Pt. Let us call this value P2. If P2
is very close to P1, we may accept the load P2 as the tangent-modulus
load. However, it is more likely that additional cycles of iteration will be
required until we reach a load that is in close agreement with the preceding trial load. This value is the tangent-modulus load.
A diagram showing how the critical stress st varies with the slenderness ratio L /r is given in Fig. 11-32 for a typical metal column with
pinned ends. Note that the curve is above the proportional limit and
below Euler’s curve.
The tangent-modulus formulas may be used for columns with various support conditions by using the effective length Le in place of the
actual length L.
P
scr = —cr
A
sr
st
spl
791
C
Eu
’s
l
er
O
(—rL)
c
cu
Reduced-Modulus Theory
rve
L
—
r
FIG. 11-32 Diagram of critical stress
versus slenderness ratio
The tangent-modulus theory is distinguished by its simplicity and ease
of use. However, it is conceptually deficient because it does not account
for the complete behavior of the column. To explain the difficulty, we
will consider again the column shown in Fig. 11-30a. When this column
first departs from the straight position (Fig. 11-30b), bending stresses
are added to the existing compressive stresses P/A. These additional
stresses are compressive on the concave side of the column and tensile
on the convex side. Therefore, the compressive stresses in the column
become larger on the concave side and smaller on the other side.
Now imagine that the axial stress P/A is represented by point A on
the stress-strain curve (Fig. 11-31). On the concave side of the column
(where the compressive stress is increased), the material follows the tangent modulus Et. However, on the convex side (where the compressive
stress is decreased), the material follows the unloading line AB on the
stress-strain diagram. This line is parallel to the initial linear part of the
diagram, and therefore its slope is equal to the elastic modulus E. Thus,
at the onset of bending, the column behaves as if it were made of two
different materials, a material of modulus Et on the concave side and a
material of modulus E on the convex side.
A bending analysis of such a column can be made using the bending
theories for a beam of two materials (Sections 6.2 and 6.3). The results
of such analyses show that the column bends as though the material had
a modulus of elasticity between the values of E and Et. This “effective
modulus” is known as the reduced modulus Er , and its value depends
not only upon the magnitude of the stress (because Et depends upon the
magnitude of the stress) but also upon the shape of the cross section of
the column. Thus, the reduced modulus Er is more difficult to determine
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
792
CHAPTER 11 Columns
than is the tangent modulus Et. In the case of a column having a rectangular cross section, the equation for the reduced modulus is
4EEt
Er
( E Et)2
(11-69)
For a wide-flange beam with the area of the web disregarded, the
reduced modulus for bending about the strong axis is
2EEt
Er
E Et
(11-70)
The reduced modulus Er is also called the double modulus.
Since the reduced modulus represents an effective modulus that
governs the bending of the column when it first departs from the straight
position, we can formulate a reduced-modulus theory of inelastic buckling. Proceeding in the same manner as for the tangent-modulus theory,
we begin with an equation for the curvature and then we write the differential equation of the deflection curve. These equations are the same as
Eqs. (11-65) and (11-66) except that Er appears instead of Et. Thus, we
arrive at the following equation for the reduced-modulus load:
p 2Er I
Pr
L2
(11-71)
The corresponding equation for the critical stress is
p 2Er
sr 2
(L/r)
(11-72)
To find the reduced-modulus load Pr, we again must use an iterative procedure, because Er depends upon Et. The critical stress according to the
reduced-modulus theory is shown in Fig. 11-32. Note that the curve for
sr is above that for st, because Er is always greater than Et.
The reduced-modulus theory is difficult to use in practice because
Er depends upon the shape of the cross section as well as the stressstrain curve and must be evaluated for each particular column.
Moreover, this theory also has a conceptual defect. In order for the
reduced modulus Er to apply, the material on the convex side of the column must be undergoing a reduction in stress. However, such a
reduction in stress cannot occur until bending actually takes place.
Therefore, the axial load P, applied to an ideal straight column, can
never actually reach the reduced-modulus load Pr. To reach that load
would require that bending already exist, which is a contradiction.
Shanley Theory
From the preceding discussions we see that neither the tangent-modulus
theory nor the reduced-modulus theory is entirely rational in explaining
the phenomenon of inelastic buckling. Nevertheless, an understanding of
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 11.8 Inelastic Buckling
P
Pcr
Pr
Euler load
Reduced-modulus load
Shanley
theory
Pt
Tangent-modulus load
O
v
FIG. 11-33 Load-deflection diagram for
elastic and inelastic buckling
793
both theories is necessary in order to develop a more complete and logically consistent theory. Such a theory was developed by F. R. Shanley in
1946 (see the historical note that follows) and today is called the Shanley
theory of inelastic buckling.
The Shanley theory overcomes the difficulties with both the tangentmodulus and reduced-modulus theories by recognizing that it is not
possible for a column to buckle inelastically in a manner that is analogous to Euler buckling. In Euler buckling, a critical load is reached at
which the column is in neutral equilibrium, represented by a horizontal
line on the load-deflection diagram (Fig. 11-33). As already explained,
neither the tangent-modulus load Pt nor the reduced-modulus load Pr can
represent this type of behavior. In both cases, we are led to a contradiction if we try to associate the load with a condition of neutral
equilibrium.
Instead of neutral equilibrium, wherein a deflected shape suddenly
becomes possible with no change in load, we must think of a column that
has an ever-increasing axial load. When the load reaches the tangentmodulus load (which is less than the reduced-modulus load), bending can
begin only if the load continues to increase. Under these conditions,
bending occurs simultaneously with an increase in load, resulting in a
decrease in strain on the convex side of the column. Thus, the effective
modulus of the material throughout the cross section becomes greater
than Et, and therefore an increase in load is possible. However, the effective modulus is not as great as Er, because Er is based upon full strain
reversal on the convex side of the column. In other words, Er is based
upon the amount of strain reversal that exists if the column bends without
a change in the axial force, whereas the presence of an increasing axial
force means that the reduction in strain is not as great.
Thus, instead of neutral equilibrium, where the relationship between
load and deflection is undefined, we now have a definite relationship
between each value of the load and the corresponding deflection. This
behavior is shown by the curve labeled “Shanley theory” in Fig. 11-33.
Note that buckling begins at the tangent-modulus load; then the load
increases but does not reach the reduced-modulus load until the deflection becomes infinitely large (theoretically). However, other effects
become important as the deflection increases, and in reality the curve
eventually goes downward, as shown by the dashed line.
The Shanley concept of inelastic buckling has been verified by
numerous investigators and by many tests. However, the maximum load
attained by real columns (see the dashed curve trending downward in
Fig. 11-33) is only slightly above the tangent-modulus load Pt. In addition, the tangent-modulus load is very simple to calculate. Therefore, for
many practical purposes it is reasonable to adopt the tangent-modulus
load as the critical load for inelastic buckling of columns.
The preceding discussions of elastic and inelastic buckling are
based upon idealized conditions. Although theoretical concepts are
important in understanding column behavior, the actual design of
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
794
CHAPTER 11 Columns
columns must take into account additional factors not considered in the
theory. For instance, steel columns always contain residual stresses produced by the rolling process. These stresses vary greatly in different
parts of the cross section, and therefore the stress level required to produce yielding varies throughout the cross section. For such reasons, a
variety of empirical design formulas have been developed for use in
designing columns. Some of the commonly used formulas are given in
the next section.
Over 200 years elapsed between the first calculation of a
buckling load by Euler (in 1744) and the final development of the theory by
Shanley (in 1946). Several famous investigators in the field of mechanics contributed to this development, and their work is described in this note.
After Euler’s pioneering studies (Ref. 11-1), little progress was made until
1845, when the French engineer A. H. E. Lamarle pointed out that Euler’s formula should be used only for slenderness ratios beyond a certain limit and that
experimental data should be relied upon for columns with smaller ratios (Ref.
11-6). Then, in 1889, another French engineer, A. G. Considère, published the
results of the first comprehensive tests on columns (Ref. 11-7). He pointed out
that the stresses on the concave side of the column increased with Et and the
stresses on the convex side decreased with E. Thus, he showed why the Euler
formula was not applicable to inelastic buckling, and he stated that the effective
modulus was between E and Et. Although he made no attempt to evaluate
the effective modulus, Considère was responsible for beginning the reducedmodulus theory.
In the same year, and quite independently, the German engineer F. Engesser
suggested the tangent-modulus theory (Ref. 11-8). He denoted the tangent modulus by the symbol T (equal to ds/de) and proposed that T be substituted for E
in Euler’s formula for the critical load. Later, in March 1895, Engesser again
presented the tangent-modulus theory (Ref. 11-9), obviously without knowledge
of Considère’s work. Today, the tangent-modulus theory is often called the
Engesser theory.
Three months later, Polish-born F. S. Jasinsky, then a professor in St.
Petersburg, pointed out that Engesser’s tangent-modulus theory was incorrect,
called attention to Considère’s work, and presented the reduced-modulus theory
(Ref. 11-10). He also stated that the reduced modulus could not be calculated
theoretically. In response, and only one month later, Engesser acknowledged the
error in the tangent-modulus approach and showed how to calculate the reduced
modulus for any cross section (Ref. 11-11). Thus, the reduced-modulus theory
is also known as the Considère-Engesser theory.
The reduced-modulus theory was also presented by the famous scientist
Theodore von Kármán in 1908 and 1910 (Refs. 11-12, 11-13, and 11-14),
apparently independently of the earlier investigations. In Ref. 11-13 he derived
the formulas for Er for both rectangular and idealized wide-flange sections (that
is, wide-flange sections without a web). He extended the theory to include the
effects of eccentricities of the buckling load, and he showed that the maximum
load decreases rapidly as the eccentricity increases.
The reduced-modulus theory was the accepted theory of inelastic buckling
until 1946, when the American aeronautical-engineering professor F. R. Shanley
pointed out the logical paradoxes in both the tangent-modulus and reduced-
Historical Note
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 11.9 Design Formulas for Columns
795
modulus theories. In a remarkable one-page paper (Ref. 11-15), Shanley not
only explained what was wrong with the generally accepted theories but also
proposed his own theory that resolved the paradoxes. In a second paper, five
months later, he gave further analyses to support his earlier theory and gave
results from tests on columns (Ref. 11-16). Since that time, many other investigators have confirmed and expanded Shanley’s concept.
For excellent discussions of the column-buckling problem, see the comprehensive papers by Hoff (Refs. 11-17 and 11-18), and for a historical account,
see the paper by Johnston (Ref. 11-19).
11.9 DESIGN FORMULAS FOR COLUMNS
In the preceding sections of this chapter we discussed the theoretical
load-carrying capacity of columns for both elastic and inelastic buckling. With that background in mind, we are now ready to examine some
practical formulas that are used in the design of columns. These design
formulas are based not only upon the theoretical analyses but also upon
the behavior of real columns as observed in laboratory tests.
The theoretical results are represented by the column curves shown
in Figs. 11-29 and 11-32. A common design approach is to approximate
these curves in the inelastic buckling range (low values of slenderness
ratio) by empirical formulas and to use Euler’s formula in the elastic
range (high values of slenderness ratio). Of course, a factor of safety
must be applied to obtain the allowable loads from the maximum loads
(or to obtain the allowable stresses from the maximum stresses).
The following examples of column design formulas are applicable
to centrally loaded columns of structural steel, aluminum, and wood.
The formulas give the allowable stresses in terms of the column properties, such as length, cross-sectional dimensions, and conditions of
support. Thus, for a given column, the allowable stress can be readily
obtained.*
Once the allowable stress is known, we can determine the allowable
load by multiplying by the cross-sectional area:
Pallow sallow A
(11-73)
The allowable load must be larger than the actual load if the allowable
stress is not to be exceeded.
The selection of a column often requires an iterative or trial-anderror procedure. Such a procedure is necessary whenever we don’t know
in advance which design formula to use. Since each formula is valid
*The design formulas given in this section are samples of the many formulas in use
around the world. They are intended for use in solving the problems at the end of the
chapter and should not be used in actual design, which requires many additional considerations. See the subsection titled “Limitations” at the end of this section.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
796
CHAPTER 11 Columns
only for a certain range of slenderness ratios, and since the slenderness
ratio is unknown until after the column is selected, we usually don’t
know which formula is applicable until we have made at least one trial.
A common trial-and-error procedure for selecting a column to
support a given axial load is the following:
1. Estimate the allowable stress sallow. (Note that an upper limit for
sallow is the allowable stress for a column of zero length. This stress
is readily obtained from the design formulas, and the estimated
stress should be equal to or less than this upper limit.)
2. Calculate an approximate value of the cross-sectional area A by
dividing the given axial load P by the estimated allowable stress.
3. Determine a column size and/or shape that supplies the required
area, either by calculating a required dimension or by selecting a
column from a table of available shapes.
4. Knowing the dimensions of a trial column from step (3), determine
the allowable stress sallow in the column from the appropriate design
formula.
5. Using Eq. (11-73), calculate the allowable load Pallow and compare it
with the actual load P.
6. If the column is not adequate to support the given load, select a
larger column and repeat the process. If the column appears to be
overdesigned (because the allowable load is much larger than the
given load), select a smaller column and repeat the process. A suitable column can usually be obtained with only two or three trials.
Many variations of this procedure are possible, depending upon the
type of column and what quantities are known in advance. Sometimes a
direct design procedure, bypassing the trial-and-error steps, can be
devised.
Structural Steel
Let us begin with design formulas for centrally loaded, structural-steel
columns. The following formulas were adopted by the American Institute of Steel Construction (AISC), a technical organization that prepares
specifications for structural-steel designers (Ref. 5-4) and provides many
other services to engineers. The AISC formulas for the allowable stress
in a column are obtained by dividing the maximum stress by an appropriate factor of safety. The term “maximum stress” means the stress
obtained by taking the maximum load (or ultimate load) the column can
carry and dividing it by the cross-sectional area.
When the slenderness ratio L/r is large, the maximum stress is
based upon the Euler load:
p 2E
smax 2
(KL /r)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(11-74)
SECTION 11.9 Design Formulas for Columns
797
where the effective length KL is used so that the formula may be applied
to a variety of support conditions.
Equation (11-74) is valid only when the stresses in the column are
less than the proportional limit spl. Under ordinary conditions, we
assume that the proportional limit of steel is equal to the yield stress sY.
However, rolled steel sections (such as wide-flange sections) contain
significant residual stresses—stresses that may be as large as one-half
the yield stress. For such a column, the proportional limit is reached
when the axial stress smax due to the compressive load equals one-half
the yield stress:
smax 0.5sY
(11-75)
To determine the smallest slenderness ratio for which Eq. (11-74) is
applicable, we set smax equal to 0.5sY and solve for the corresponding
value of KL/r, which is known as the critical slenderness ratio (compare
with Eq. 11-63):
KL
r c
冢 冣
冪莦
2p 2E
sY
(11-76)
If the actual slenderness ratio is equal to or larger than (KL /r)c , the
Euler formula for the maximum stress (Eq. 11-74) may be used. Thus,
the critical slenderness ratio given by Eq. (11-76) determines the boundary between elastic and inelastic buckling for rolled steel columns.
Equation (11-74) may be expressed in nondimensional form by
dividing by the yield stress sY and then substituting from Eq. (11-76):
smax
(KL /r)2c
p 2E
2 2
sY
s Y (KL /r)
2(KL /r)
KL
r
KL
冢r冣
c
(11-77)
This equation is plotted in Fig. 11-34 and labeled Euler’s curve.
For the region of inelastic buckling, where KL /r (KL /r)c, the
maximum stress is given by a parabolic formula:
smax
(KL /r)2
1 2
sY
2(KL /r)c
KL
r
KL
冢r冣
c
(11-78)
This empirical formula is also plotted in Fig. 11-34. Note that the curve
is a parabola with a horizontal tangent at KL/r 0, where the maximum
stress is equal to sY. At the critical slenderness ratio (KL /r)c the curve
merges smoothly with Euler’s curve (both curves have the same slope at
the point where they meet). Thus, the empirical formula provides a
design curve that fits the general shape of the theoretical curves (Figs.
11-29 and 11-32) while also being simple to use. The validity of the formula for use in design has been verified by numerous tests.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
798
CHAPTER 11 Columns
smax = sY
1
sallow
smax
—
sY and —
sY
0.6
Eq. (11-78)
smax
—
sY
sallow
—
sY
Eu
0.5
le
r’s
Eq. (11-81)
cu
0.261
FIG. 11-34 Design formulas for
rve
Eq. (11-77)
Eq. (11-82)
KL
—
r c
0
KL
—
r
( )
structural-steel columns
To obtain the allowable stresses from the maximum stresses, the
AISC adopted the following formulas for the factors of safety:
3(KL /r)
(KL /r)3
5
n1
3
8(KL /r)c
8(KL /r)3c
KL
r
23
n2 ⬇ 1.92
12
冢KrL 冣
KL
r
KL
冢r冣
c
(11-79)
(11-80)
c
Thus, the factor of safety is 5/3 when KL/r 0 and gradually increases
to 23/12 when KL/r (KL/r)c. For higher slenderness ratios, the factor
of safety remains constant at that value.
The allowable stresses are now obtained by dividing the maximum
stresses smax by the appropriate factor of safety (n1 or n2); thus,
冤
冥
sallow
(KL /r)2
1
1
sY
n1
2(KL /r)2c
sallow
(KL /r)2c
2
sY
2n2(KL /r)
KL
r
KL
r
冢KrL 冣
c
冢KrL 冣
c
(11-81)
(11-82)
These equations for the allowable stresses are also plotted in Fig. 11-34.
The AISC specifications place an upper limit of 200 on the slenderness ratio KL/r and specify the modulus of elasticity E as 29,000 ksi.
Also, the symbols used in the AISC specifications differ slightly from
those in the preceding formulas. For instance, the critical slenderness
ratio is denoted Cc, the allowable stress is denoted Fa, and the yield
stress if denoted Fy.
All of the preceding design formulas for structural steel may be
used with either USCS or SI units. The formulas are applicable to wideflange and other rolled shapes, as well as to columns with rectangular
and circular cross sections.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 11.9 Design Formulas for Columns
799
Aluminum
sallow
Eq. (11-83a)
Eq. (11-83b)
Eq. (11-83c)
O
S1
S2
FIG. 11-35 Design formulas for
aluminum columns
KL
r
The design formulas presented below for aluminum columns are taken
from the specifications of the Aluminum Association (Ref. 5-5). Like the
formulas for steel design, the formulas for aluminum are based upon the
theoretical curves given in Figs. 11-29 and 11-32.
The general shape of the design curves for aluminum is shown in
Fig. 11-35, where the ordinate is the allowable stress and the abscissa is
the effective slenderness ration KL/r. The slenderness ratio S1 separates
short and intermediate columns, and the ratio S2 separates intermediate
and long columns (compare with Fig. 11-29). The allowable stress in the
short-column region is based upon the yield strength of the material; in
the intermediate-column region it is based upon the tangent modulus
formula; and in the long-column region it is based upon Euler’s formula.
For aluminum columns in direct compression, the general design
formulas are expressed as follows:
sY
sallow
nY
冢
0
KL
1
sallow Bc 2 Dc
r
nu
冣
p 2E
sallow 2
nu(KL /r)
KL
r
(11-83a)
S1
S1
KL
r
KL
r
S2
S2
(11-83b)
(11-83c)
In these equations, KL/r is the effective slenderness ratio, the stress sY is
the compressive yield stress (0.2% offset), nY is the factor of safety with
respect to the yield stress, nu is the factor of safety with respect to the
ultimate stress, and Bc and Dc are constants.
The values of the various quantities appearing in Eqs. (11-83a, b,
and c) depend upon the particular aluminum alloy, the temper of the
finished product, and the use to which it will be put. Numerous alloys
and tempers are available, so the Aluminum Association gives tables of
values based upon the material and usage.
As examples, the following formulas apply to two alloys used in
buildings and aircraft structures. In these particular cases, the short-column
region is very small and can be combined with the intermediate-column
region; thus, for these materials, the slenderness ratio S1 is taken as
zero.
1. Alloy 2014-T6
S1 0, S2 55
冢 冣
KL
sallow 30.7 0.23 ksi
r
54,000 ksi
sallow
(KL/r)2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
0
KL
r
KL
r
55
55
(11-84a)
(11-84b)
800
CHAPTER 11 Columns
2. Alloy 6061-T6
S1 0, S2 66
冢 冣
KL
sallow 20.2 0.126 ksi
r
51,000 ksi
sallow
(KL/r)2
0
KL
r
KL
r
66
66 (11-85a)
(11-85b)
Note that these formulas give the allowable stresses, hence they already
incorporate the factors of safety, which are 1.65 and 1.95 for nY and nU,
respectively. Also, note that the allowable stresses have units of kips per
square inch (ksi). Finally, observe that the design curves (Fig. 11-35
on the preceding page) meet with distinctly different shopes at the
slenderness ratio S2.
Wood
Wood structural members are readily available in the form of sawn
lumber, glued-laminated timbers, and round poles and piles. Their
strength depends upon many factors, the most important being the
species (such as Douglas fir or southern pine) and the grade (such as
Select Structural or Construction). Among the other factors affecting
strength are moisture content and duration of loading (wood will support
greater loads for short durations than for long durations).
The design of wood structural members, like those of steel and
aluminum, is governed by codes and specifications. In the United States,
the most widely used design codes for wood are those of the American
Forest and Paper Association (Ref. 5-6), which publishes the National
Design Specifications for Wood Construction and related manuals. The
formulas and requirements described in this section are taken from those
specifications. We will limit our discussion to columns of rectangular cross
section constructed of either sawn lumber or glued-laminated timber.
The allowable stress in compression, parallel to the grain of the
wood, on the cross section of a column is denoted in the specifications
as F c, which is the same as allow in the notation of this book. Therefore,
the allowable axial load on a centrally loaded column is
Pallow sallow A F c A
(11-86)
in which A is the cross-sectional area of the column.
The allowable stress F c for use in the preceding equation is given
in the specifications as
F c FcC*CP F*c CP
(11-87)
in which Fc is the compressive design stress for the particular species
and grade of wood, C* is an adjustment factor for various service conditions, CP is the column stability factor, and F *c is the adjusted
compressive design stress (equal to the product of Fc and the adjustment
factor C*). Each of these terms will now be described.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 11.9 Design Formulas for Columns
801
The design stress Fc is based upon laboratory tests of wood specimens and is listed in tables in the specifications. For instance, typical
values of Fc for structural grades of Douglas fir and southern pine are in
the range of 700 to 2000 psi (5 to 14 MPa).
The adjustment factor C* takes into account the service conditions, that is, the actual conditions of use, including duration of loading,
wet conditions, and high temperatures. When solving problems in this
book, we will assume C* 1.0, which is not unreasonable for ordinary
indoor conditions.
The column stability factor CP is based upon buckling considerations analogous to those described in connection with Figs. 11-29 and
11-32. For wood columns, a single buckling formula has been devised
that covers the entire region of column behavior, including short, intermediate, and long columns. The formula, which follows as Eq. (11-89),
gives the stability factor CP in terms of several variables, one of which is
the wood slenderness ratio:
Le
Wood slenderness ratio
d
(11-88)
in which Le is the effective length for buckling and d is the depth of the
cross section in the plane of buckling.
The effective length Le appearing in the wood slenderness ratio is
the same as the effective length KL in our earlier discussions (see Fig.
11-19). However, note carefully that the slenderness ratio Le /d is not the
same as the slenderness ratio L/r used previously (see Eq. 11-17). The
dimension d is the depth of the cross section in the plane of buckling,
whereas r is the radius of gyration of the cross section in the plane of
buckling. Also, note that the maximum permissible value of the wood
slenderness ratio L e /d is 50.
The column stability factor CP is calculated from the following
formula:
1 (FcE /F*
c)
CP
2c
莦
莦
冥
冪冤莦
莦
2c
c 莦
1 (FcE/F c*)
2
FcE /F *c
(11-89)
in which FcE is the Euler buckling coefficient (Eq. 11-90), F*c is the
adjusted compressive design stress (see Eq. 11-87), and c is a constant
depending upon the type of column (for instance, c 0.8 for sawn lumber and 0.9 for glued-laminated timber).
The Euler buckling coefficient is defined as follows:
KcE E
FcE 2
(Le /d )
(11-90)
in which KcE is a buckling coefficient, E is an adjusted modulus of
elasticity, and L e /d is the wood slenderness ratio.
The coefficient KcE is based upon the method of grading and is
equal to 0.3 for visually graded lumber and 0.418 for glued-laminated
timber. The adjusted modulus E is equal to the modulus of elasticity E
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
802
CHAPTER 11 Columns
multiplied by an adjustment factor for service conditions. When solving
problems in this book, we will assume that these adjustment factors
equal 1.0, and therefore E E. Typical values of the modulus E for
structural lumber are in the range of 1,200,000 to 2,000,000 psi (8 to
14 GPa).
In summary, Eqs. (11-86) through (11-90) are the general equations for the buckling of wood columns. However, when solving
problems in this book, we assume the following specific conditions:
1. The columns have rectangular cross sections and are constructed of
either sawn lumber or glued-laminated timber.
2. The adjustment factor C * 1.0, and therefore the following three
relations may be used:
Fc sallow FcCP
F *c Fc
(11-91a,b)
(11-92)
Pallow Fc A FcCP A
3. The constant c 0.8 or 0.9 (for sawn lumber and glued-laminated
timber, respectively).
4. The coefficient KcE 0.3 or 0.418 (for sawn lumber and
glued-laminated timber, respectively).
5. The modulus E E.
With these conditions, the equation for the Euler buckling coefficient (Eq. 11-90) becomes
KcE E
FcE
(Le /d )2
(11-93)
and the nondimensional ratio FcE /F*
c, which we will denote by the
Greek letter f (phi), becomes
FcE
Kc E E
f
F c*
Fc(Le /d)2
(11-94)
With this simplified notation, the equation for the column stability factor
becomes
1f
CP
2c
冤2c莦
冥 c莦
冪莦
1f
2
f
(11-95)
Note that the slenderness ratio L e /d enters the calculation of CP through
the ratio f.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 11.9 Design Formulas for Columns
803
1.0
C* = 1.0
c = 0.8
KcE = 0.3
0.8
E′ = E
E
= 1500
Fc
0.6
CP
0.4
E
= 1000
Fc
0.2
FIG. 11-36 Typical curves for the column
0
10
20
30
40
50
Le
d
stability factor CP (rectangular wood
columns)
A graph of the stability factor is shown in Fig. 11-36. The curves
for CP are plotted for two values of the ratio E/Fc. Note that both curves
have zero slope for L e /d equal to zero, and both curves terminate at
Le /d 50, which is the upper limit permitted by the specifications.
Although these curves are plotted for specific values of the various
parameters, they show in general how the stability factor varies with the
slenderness ratio L e /d.
Limitations
The preceding formulas for the design of steel, aluminum, and wood
columns are intended solely for use in solving problems in this book.
They should not be used for the design of actual columns, because they
represent only a small part of the complete design process. Many factors
besides those discussed here enter into the design of columns, and therefore textbooks or other references on structural design should be
consulted before designing a column for a specific application.
Furthermore, all design formulas presented in specifications and
codes, such as the formulas given in this section, require informed judgment in their use. There are many cases of structures that “met the code”
but nevertheless collapsed or failed to perform adequately. Meeting the
code requirements is not enough for a safe design—practical design
experience is also essential.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
804
CHAPTER 11 Columns
Example 11-5
A steel column is constructed from a W 10
60 wide-flange section (Fig.
11-37). Assume that the column has pin supports and may buckle in any direction. Also, assume that the steel has modulus of elasticity E 29,000 ksi and
yield stress sY 36 ksi.
(a) If the length of the column is L 20 ft, what is the allowable axial
load?
(b) If the column is subjected to an axial load P 200 k, what is the maximum permissible length?
P
L
FIG. 11-37 Example 11-5. Steel wide-
flange column
P
Solution
We will use the AISC formulas (Eqs. 11-79 through 11-82) when analyzing
this column. Since the column has pin supports, the effective-length factor
K 1. Also, since the column will buckle about the weak axis of bending, we
will use the smaller radius of gyration:
r 2.57 in.
as obtained from Table E-1, Appendix E. The critical slenderness ratio (Eq.
11-76) is
2p E
2p (29,000 ksi)
冪
126.1
冢KrL 冣 冪
莦
莦
s
3莦
6 ksi 莦
2
c
2
(a)
Y
(a) Allowable axial load. If the length L 20 ft, the slenderness ratio of
the column is
L
(20 ft)(12 in./ft)
93.4
r
2.57 in.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 11.9 Design Formulas for Columns
805
which is less than the critical ratio (Eq. a). Therefore, we will use Eqs. (11-79)
and (11-81) to obtain the factor of safety and allowable stress, respectively:
3(93.4)
(93.4)3
5
3(KL /r)
(KL /r)3
5
n1 3 3 1.89
3
8(KL /r)c
8(KL /r)c
3
8(126.1)
8(126.1)
冤
冥
冤
冥
sallow
(KL /r)2
1
1
(93.4)2
1
0.384
2 1
sY
n1
1.89
2(126.1)2
2(KL /r) c
sallow 0.384sY 0.384(36 ksi) 13.8 ksi
Since the cross-sectional area of the column is A 17.6 in.2 (from Table E-1),
the allowable axial load is
Pallow sallow A (13.8 ksi)(17.6 in.2) 243 k
(b) Maximum permissible length. To determine the maximum length when
the axial load P 200 k, we begin with an estimated value of the length and
then use a trial-and-error procedure. Note that when the load P 200 k, the
maximum length is greater than 20 ft (because a length of 20 ft corresponds to
an axial load of 243 k). Therefore, as a trial value, we will assume L 25 ft.
The corresponding slenderness ratio is
L
(25 ft)(12 in./ft)
116.7
r
2.57 in.
which is less than the critical ratio. Therefore, we again use Eqs. (11-79) and
(11-81) to obtain the factor of safety and allowable stress:
3(KL /r)
(KL /r)3
3(116.7)
(116.7)3
5
5
n1 3 3 1.915
8(KL /r)c
8(KL /r)c
8 ( 1 2 6.1)
8(1 26.1)
3
3
冤
2
冥
冤
冥
(KL /r)
(116.7)2
sallow
1
1
1
0.299
2 1
sY
n1
1.915
2(1 26.1)2
2(KL /r) c
sallow 0.299sY 0.299(36 ksi) 10.8 ksi
Thus, the allowable axial load corresponding to a length L 25 ft is
Pallow sallow A (10.8 ksi)(17.6 in.2) 190 k
which is less than the given load of 200 k. Therefore, the permissible length is
less than 25 ft.
Performing similar calculations for L 24.0 ft and L 24.5 ft, we obtain
the following results:
L 24.0 ft
Pallow 201 k
L 24.5 ft
Pallow 194 k
L 25.0 ft
Pallow 190 k
Interpolating between these results, we see that a load of 200 k corresponds to a
length of 24.1 ft. Thus, the maximum permissible length of the column is
L max 24.1 ft
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
806
CHAPTER 11 Columns
Example 11-6
Find the minimum required thickness tmin for a steel pipe column of length
L 3.6 m and outer diameter d 160 mm supporting an axial load P
240 kN (Fig. 11-38). The column is fixed at the base and free at the top. (Use
E 200 GPa and sY 250 MPa.)
P
L
d
t
FIG. 11-38 Example 11-6. Steel pipe
column
Solution
We will use the AISC formulas (Eqs. 11-79 through 11-82) when analyzing
this column. Since the column has fixed-free end conditions, the effective length
is
Le KL 2(3.6 m) 7.2 m
Also, the critical slenderness ratio (Eq. 11-76) is
2p E
2p (200 GPa)
冪
125.7
冢KrL 冣 冪
莦
莦
s
250莦
MPa 莦
2
c
2
(b)
Y
First trial. To determine the required thickness of the column, we will use a
trial-and-error method. Let us start by assuming a trial value t 7.0 mm. Then
the moment of inertia of the cross-sectional area is
p
p
I 冤d 4 (d 2t) 4冥 冤(160 mm) 4 (146 mm) 4冥 9.866
64
64
10 6 mm4
Also, the cross-sectional area and radius of gyration are
p
p
A 冤 d 2 (d 2t ) 2 冥 冤(160 mm) 2 (146 mm) 2冥 3365 mm2
4
4
r
冪莦 冪莦莦莦
I
A
9.866 106 mm4
54.15 mm
3365 mm2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 11.9 Design Formulas for Columns
807
Therefore, the slenderness ratio of the column is
2(3.6 m)
KL
133.0
r
54.15 mm
Since this ratio is larger than the critical slenderness ratio (Eq. b), we obtain the
factor of safety and the allowable stress from Eqs. (11-80) and (11-82):
n2 1.92
(KL /r)2c
sallow
(125.7)2
2
0.2326
sY
2n 2(KL /r)
2(1.92)(133.0)2
sallow 0.2326sY 0.2326(250 MPa) 58.15 MPa
Thus, the allowable axial load is
Pallow sallow A (58.15 MPa)(3365 mm2) 196 kN
Since this load is less than the required load of 240 kN, we must try a larger
value of the thickness t.
Additional trials. Performing similar calculations for t 8 mm and t
9 mm, we get the following results:
t 7.0 mm
Pallow 196 kN
t 8.0 mm
Pallow 220 kN
t 9.0 mm
Pallow 243 kN
By interpolation, we see that t 8.9 mm corresponds to a load of 240 kN.
Therefore, the required thickness of the pipe column is
t min 8.9 mm
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
808
CHAPTER 11 Columns
Example 11-7
An aluminum tube (alloy 2014-T6) with an effective length L 16.0 in. is compressed by an axial force P 5.0 k (Fig. 11-39).
Determine the minimum required outer diameter d if the thickness t equals
one-tenth the outer diameter.
P
L
d
d
t= —
10
FIG. 11-39 Example 11-7. Aluminum
P
tube in compression
Solution
We will use the Aluminum Association formulas for alloy 2014-T6 (Eqs.
11-84a and b) for analyzing this column. However, we must make an initial
guess as to which formula is applicable, because each formula applies to a
different range of slenderness ratios. Let us assume that the slenderness ratio of
the tube is less than 55, in which case we use Eq. (11-84a) with K 1:
冢冣
L
sallow 30.7 0.23 ksi
r
(c)
In this equation, we can replace the allowable stress by the actual stress P/A,
that is, by the axial load divided by the cross-sectional area. The cross-sectional
area is
p
p
A 冤d 2 (d 2t ) 2 冥 冤d 2 (0.8d ) 2 冥 0.2827d 2
4
4
(d)
Therefore, the stress P/A is
P
5.0 k
17.69
A
0.2827d 2
d2
in which P/A has units of kips per square inch (ksi) and d has units of inches
(in.). Substituting into Eq. (c), we get
冢冣
17.69
L
30.7 0.23 ksi
d2
r
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(e)
SECTION 11.9 Design Formulas for Columns
809
The slenderness ratio L /r can also be expressed in terms of the diameter d.
First, we find the moment of inertia and radius of gyration of the cross section:
p
p
I 冤d 4 (d 2t) 4冥 冤d 4 (0.8d) 4冥 0.02898d 4
64
64
0.02898d
0.3202d
冪莦AI 冪
莦
0.282莦
7d
4
r
2
Therefore, the slenderness ratio is
16.0 in.
L
49.97 in.
0.3202d
r
d
(f)
where (as before) the diameter d has units of inches.
Substituting into Eq. (e), we obtain the following equation, in which d is
the only unknown quantity:
冢
17.69
49.97
30.7 0.23
d2
d
冣
With a little rearranging, this equation becomes
30.7d 2 11.49d 17.69 0
from which we find
d 0.97 in.
This result is satisfactory provided the slenderness ratio is less than 55, as
required for Eq. (c) to be valid. To verify that this is the case, we calculate the
slenderness ratio from Eq. (f):
L
49.97 in.
49.97 in.
51.5
r
d
0.97 in.
Therefore, the solution is valid, and the minimum required diameter is
dmin 0.97 in.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
810
CHAPTER 11 Columns
Example 11-8
A wood post of rectangular cross section (Fig. 11-40) is constructed of Douglas
fir lumber having a compressive design stress Fc 11 MPa and modulus of elasticity E 13 GPa. The length of the post is L and the cross-sectional dimensions
are b and h. The supports at the ends of the post provide pinned-end conditions,
so the length L becomes the effective length Le. Also, buckling is free to occur
about either principal axis of the cross section. (Note: Since the post is made of
sawn lumber, the constant c equals 0.8 and the coefficient KcE equals 0.3.)
(a) Determine the allowable axial load Pallow if L 1.8 m, b 120 mm,
and h 160 mm.
(b) Determine the maximum allowable length L max if the axial load
P 100 kN, b 120 mm, and h 160 mm.
(c) Determine the minimum width bmin of the cross section if the column is
square, P 125 kN, and L 2.6 m.
P
L
h
b
FIG. 11-40 Example 11-8. Wood post in
P
compression
Solution
(a) Allowable axial load. The allowable load (from Eq. 11-92) is
Pallow Fc A FcCP A
in which Fc 11 MPa and
A bh (120 mm)(160 mm) 19.2
103 mm2
To find the stability factor CP, we first calculate the slenderness ratio, as follows:
Le
1.8 m
15
d
120 mm
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
811
SECTION 11.9 Design Formulas for Columns
in which d is the smaller dimension of the cross section. Next, we obtain the
ratio f from Eq. (11-94):
FcE
KcE E
(0 .3)(13 GPa)
f 5 2 2 1.5758
F*c
Fc(L e/d)
(11 MPa)(15)
Then we substitute f into Eq. (11-95) for CP, while also using c 0.8, and we
obtain
1 1.5758
CP
1.6
1 1.5758
1.5758
莦
莦
0.8212
冥
冪冤莦
莦
1.6
0.8
2
Finally, the allowable axial load is
Pallow FcCP A (11 MPa)(0.8212)(19.2
103 mm2) 173 kN
(b) Maximum allowable length. We begin by determining the required
value of CP. Rearranging Eq. (11-92) and replacing Pallow by the load P, we
obtain the formula for CP shown below. Then, we substitute numerical values
and obtain the following result:
P
100 kN
CP 0.47348
Fc A
Substituting this value of CP into Eq. (11-95), and also setting c equal to 0.8, we
get the following equation in which f is the only unknown quantity:
1f
CP 0.47348
1.6
冥莦
莦
冪冤莦
1.6
0.8
1f
2
f
Solving numerically by trial and error, we find
f 0.55864
Finally, from Eq. (11-94), we get
L
d
(0.3)(13 GPa)
冪莦莦
25.19
冪
莦
fF
Kc E E
c
and
Lmax 25.19d (25.19)(120 mm) 3.02 m
continued
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
812
CHAPTER 11 Columns
Any larger value of the length L will produce a smaller value of CP and hence a
load P that is less than the actual load of 100 kN.
(c) Minimum width of square cross section. The minimum width bmin can
be found by trial and error, using the procedure described in part (a). The steps
are as follows:
1.
2.
3.
Select a trial value of b (meters)
Calculate the slenderness ratio L/d 2.6/b (nondimensional)
Calculate the ratio f from Eq. (11-94):
KcEE
( 0.3)(13 GPa)
52.448b2
f 2
Fc(Le /d)
(11 MPa)(2.6/b)2
4.
5.
(nondimensional)
Substitute f into Eq. (11-95) and calculate CP (nondimensional)
Calculate the load P from Eq. (11-92):
P FcCP A (11 MPa)(CP)(b2) 11,000 CP b2 (kilonewtons)
6.
Compare the calculated value of P with the given load of 125 kN. If P is
less than 125 kN, select a larger trial value for b and repeat steps (2)
through (7). If P is larger than 125 kN by a significant amount, select a
smaller value for b and repeat the steps. Continue until P reaches a satisfactory value.
Let us take a trial value of b equals to 130 mm, or 0.130 m. Then steps (2)
through (5) produce the following results:
L/d 2.6/b 20
f 52.448 b2 0.88637
CP 0.64791
P 11,000 CP b2 120.4 kN
Since the given load is 125 kN, we select a larger value of b, say 0.132 m, for
the next trial. Proceeding in this manner with successive trials, we obtain the
following results:
b 0.132 m; P 126.3 kN
b 0.131 m; P 123.4 kN
Therefore, the minimun width of the square cross section is
bmin 0.132 m 132 m
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 11 Problems
813
PROBLEMS CHAPTER 11
11.3-1 Calculate the critical load Pcr for a W 8
Idealized Buckling Models
11.2-1 through 11.2-4 The figure shows an idealized
structure consisting of one or more rigid bars with pinned
connections and linearly elastic springs. Rotational stiffness
is denoted bR and translational stiffness is denoted b.
Determine the critical load Pcr for the structure.
35 steel
column (see figure) having length L 24 ft and
E 30 106 psi under the following conditions:
(a) The column buckles by bending about its strong
axis (axis 1-1), and (b) the column buckles by bending
about its weak axis (axis 2-2). In both cases, assume that
the column has pinned ends.
P
2
C
P
B
B
b
L
L
C
1
1
a
bR
A
A
PROB. 11.2-1
PROB. 11.2-2
2
PROBS. 11.3-1 through 11.3-3
11.3-2 Solve the preceding problem for a W 10
column having length L 30 ft.
P
C
bR
P
11.3-3 Solve Problem 11.3-1 for a W 10
having length L 28 ft.
C
11.3-4 A horizontal beam AB is pin-supported at end A and
L
—
2
L
—
2
bR
B
B
bR
L
—
2
60 steel
bR
A
A
PROB. 11.2-3
PROB. 11.2-4
L
—
2
carries a load Q at end B, as shown in the figure. The beam
is supported at C by a pinned-end column. The column is a
solid steel bar (E 200 GPa) of square cross section having length L 1.8 m and side dimensions b 60 mm.
Based upon the critical load of the column, determine
the allowable load Q if the factor of safety with respect to
buckling is n 2.0.
A
B
C
2d
d
b
Critical Loads of Columns with Pinned Supports
The problems for Section 11.3 are to be solved using the
assumptions of ideal, slender, prismatic, linearly elastic
columns (Euler buckling). Buckling occurs in the plane of
the figure unless stated otherwise.
45 steel column
D
PROBS. 11.3-4 and 11.3-5
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
Q
L
814
CHAPTER 11 Columns
11.3-5 Solve the preceding problem if the column is
aluminum (E 10
106 psi), the length L 30 in., the
side dimension b 1.5 in., and the factor of safety n 1.8
(see the figure on the previous page).
P
C
11.3-6 A horizontal beam AB is pin-supported at end A and
X
carries a load Q at end B, as shown in the figure. The beam
is supported at C and D by two identical pinned-end
columns of length L. Each column has flexural rigidity EI.
What is the critical load Qcr? (In other words, at what
load Qcr does the system collapse because of Euler buckling of the columns?)
A
C
d
D
L
—
2
X
b
h
B
L
—
2
b
B
Section X-X
A
2d
d
Q
PROB. 11.3-8
L
PROB. 11.3-6
11.3-7 A slender bar AB with pinned ends and length L is
11.3-9 Three identical, solid circular rods, each of radius r
and length L, are placed together to form a compression
member (see the cross section shown in the figure).
Assuming pinned-end conditions, determine the critical load Pcr as follows: (a) The rods act independently as
individual columns, and (b) the rods are bonded by epoxy
throughout their lengths so that they function as a single
member.
What is the effect on the critical load when the rods act
as a single member?
held between immovable supports (see figure).
What increase T in the temperature of the bar will
produce buckling at the Euler load?
A
∆T
2r
B
PROB. 11.3-9
L
PROB. 11.3-7
11.3-8 A rectangular column with cross-sectional dimensions b and h is pin-supported at ends A and C (see figure).
At midheight, the column is restrained in the plane of the
figure but is free to deflect perpendicular to the plane of the
figure.
Determine the ratio h/b such that the critical load is the
same for buckling in the two principal planes of the
column.
11.3-10 Three pinned-end columns of the same material
have the same length and the same cross-sectional area
(see figure). The columns are free to buckle in any direction. The columns have cross sections as follows: (1) a
circle, (2) a square, and (3) an equilateral triangle.
Determine the ratios P1 : P2 : P3 of the critical loads
for these columns.
(1)
(2)
PROB. 11.3-10
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(3)
CHAPTER 11 Problems
11.3-11 A long slender column ABC is pinned at ends A
and C and compressed by an axial force P (see figure). At
the midpoint B, lateral support is provided to prevent
deflection in the plane of the figure. The column is a steel
wide-flange section (W 10
45) with E 30
106 psi.
The distance between lateral supports is L 18 ft.
Calculate the allowable load P using a factor of safety
n 2.4, taking into account the possibility of Euler buckling about either principal centroidal axis (i.e., axis 1-1 or
axis 2-2).
815
11.3-13 The hoisting arrangement for lifting a large pipe is
shown in the figure. The spreader is a steel tubular section
with outer diameter 2.75 in. and inner diameter 2.25 in. Its
length is 8.5 ft and its modulus of elasticity is 29 106 psi.
Based upon a factor of safety of 2.25 with respect to
Euler buckling of the spreader, what is the maximum
weight of pipe that can be lifted? (Assume pinned conditions at the ends of the spreader.)
F
P
Cable
2
C
7
7
X
X
L
W 10
10
10
45
1
1
B
A
B
Spreader
Cable
L
Pipe
2
Section X - X
A
PROB. 11.3-11
11.3-12 The multifaceted glass roof over the lobby of a
museum building is supported by the use of pretensioned
cables. At a typical joint in the roof structure, a strut AB is
compressed by the action of tensile forces F in a cable that
makes an angle a 75° with the strut (see figure). The
strut is a circular tube of aluminum (E 72 GPa) with
outer diameter d2 50 mm and inner diameter d1
40 mm. The strut is 1.0 m long and is assumed to be pinconnected at both ends.
Using a factor of safety n 2.5 with respect to the
critical load, determine the allowable force F in the cable.
F
A
PROB. 11.3-13
11.3-14 A pinned-end strut of aluminum (E 72 GPa)
with length L 1.8 m is constructed of circular tubing with
outside diameter d 50 mm (see figure). The strut must
resist an axial load P 18 kN with a factor of safety n
2.0 with respect to the critical load.
Determine the required thickness t of the tube.
t
d = 50 mm
PROB. 11.3-14
d2
11.3-15 The cross section of a column built up of two steel
a
Strut
a
B
Cable
F
PROB. 11.3-12
I-beams (S 6
17.25 sections) is shown in the figure on
the next page. The beams are connected by spacer bars, or
lacing, to ensure that they act together as a single column.
(The lacing is represented by dashed lines in the figure on
the next page.)
The column is assumed to have pinned ends and may
buckle in any direction. Assuming E 30
106 psi and
L 27.5 ft, calculate the critical load Pcr for the column.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
816
CHAPTER 11 Columns
Columns with Other Support Conditions
S6
17.25
The problems for Section 11.4 are to be solved using the
assumptions of ideal, slender, prismatic, linearly elastic
columns (Euler buckling). Buckling occurs in the plane of
the figure unless stated otherwise.
4 in.
PROB. 11.3-15
11.3-16 The truss ABC shown in the figure supports a vertical load W at joint B. Each member is a slender circular
steel pipe (E 200 GPa) with outside diameter 100 mm
and wall thickness 6.0 mm. The distance between supports
is 7.0 m. Joint B is restrained against displacement perpendicular to the plane of the truss.
Determine the critical value Wcr of the load.
11.4-1 An aluminum pipe column (E 10,400 ksi) with
length L 10.0 ft has inside and outside diameters
d1 5.0 in. and d2 6.0 in., respectively (see figure). The
column is supported only at the ends and may buckle in any
direction.
Calculate the critical load Pcr for the following end
conditions: (1) pinned-pinned, (2) fixed-free, (3) fixedpinned, and (4) fixed-fixed.
d2
d1
B
100 mm
W
40°
A
55°
PROBS. 11.4-1 and 11.4-2
11.4-2 Solve the preceding problem for a steel pipe column (E 210 GPa) with length L 1.2 m, inner diameter
d1 36 mm, and outer diameter d2 40 mm.
C
7.0 m
106 psi) of
W 12
87 shape (see figure) has length L 28 ft. It is
supported only at the ends and may buckle in any direction.
Calculate the allowable load Pallow based upon the critical load with a factor of safety n 2.5. Consider the
following end conditions: (1) pinned-pinned, (2) fixed-free,
(3) fixed-pinned, and (4) fixed-fixed.
11.4-3 A wide-flange steel column (E 30
PROB. 11.3-16
11.3-17 A truss ABC supports a load W at joint B, as
shown in the figure. The length L1 of member AB is fixed,
but the length of strut BC varies as the angle u is changed.
Strut BC has a solid circular cross section. Joint B is
restrained against displacement perpendicular to the plane
of the truss.
Assuming that collapse occurs by Euler buckling of
the strut, determine the angle u for minimum weight of the
strut.
★
A
2
1
B
1
u
W
2
C
PROBS. 11.4-3 and 11.4-4
L1
PROB. 11.3-17
11.4-4 Solve the preceding problem for a W 10
shape with length L 24 ft.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
60
CHAPTER 11 Problems
11.4-5 The upper end of a W 8
21 wide-flange steel
column (E 30 103 ksi) is supported laterally between
two pipes (see figure). The pipes are not attached to the column, and friction between the pipes and the column is
unreliable. The base of the column provides a fixed support, and the column is 13 ft long.
Determine the critical load for the column, considering
Euler buckling in the plane of the web and also perpendicular to the plane of the web.
W8
11.4-7 The horizontal beam ABC shown in the figure is
supported by columns BD and CE. The beam is prevented
from moving horizontally by the roller support at end A, but
vertical displacement at end A is free to occur. Each column
is pinned at its upper end to the beam, but at the lower ends,
support D is fixed and support E is pinned. Both columns
are solid steel bars (E 30 10 6 psi) of square cross section with width equal to 0.625 in. A load Q acts at distance
a from column BD.
(a) If the distance a 12 in., what is the critical value
Qcr of the load?
(b) If the distance a can be varied between 0 and
40 in., what is the maximum possible value of Qcr? What is
the corresponding value of the distance a?
a
21
Q
C
B
A
PROB. 11.4-5
40 in.
35 in.
45 in.
0.625 in.
11.4-6 A vertical post AB is embedded in a concrete foundation and held at the top by two cables (see figure). The
post is a hollow steel tube with modulus of elasticity
200 GPa, outer diameter 40 mm, and thickness 5 mm. The
cables are tightened equally by turnbuckles.
If a factor of safety of 3.0 against Euler buckling in the
plane of the figure is desired, what is the maximum allowable tensile force Tallow in the cables?
817
0.625 in.
D
E
PROB. 11.4-7
B
40 mm
11.4-8 The roof beams of a warehouse are supported by
Cable
2.1 m
Steel tube
Turnbuckle
A
2.0 m
PROB. 11.4-6
2.0 m
pipe columns (see figure on the next page) having outer
diameter d2 100 mm and inner diameter d1 90 mm.
The columns have length L 4.0 m, modulus E
210 GPa, and fixed supports at the base.
Calculate the critical load Pcr of one of the columns
using the following assumptions: (1) the upper end is
pinned and the beam prevents horizontal displacement;
(2) the upper end is fixed against rotation and the beam
prevents horizontal displacement; (3) the upper end is
pinned but the beam is free to move horizontally; and
(4) the upper end is fixed against rotation but the beam is
free to move horizontally.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
818
CHAPTER 11 Columns
Q 200 kN
Roof beam
B
Pipe column
1.0 m
d2
1.0 m
L
2.0 m
d 100 mm
PROB. 11.4-8
A
PROB. 11.4-10
11.4-9 Determine the critical load Pcr and the equation of
the buckled shape for an ideal column with ends fixed
against rotation (see figure) by solving the differential
equation of the deflection curve. (See also Fig. 11-17.)
P
B
L
A
PROB. 11.4-9
11.4-10 An aluminum tube AB of circular cross section is
fixed at the base and pinned at the top to a horizontal beam
supporting a load Q 200 kN (see figure).
Determine the required thickness t of the tube if its
outside diameter d is 100 mm and the desired factor of
safety with respect to Euler buckling is n 3.0. (Assume
E 72 GPa.)
11.4-11 The frame ABC consists of two members AB and
BC that are rigidly connected at joint B, as shown in part (a)
of the figure. The frame has pin supports at A and C.
A concentrated load P acts at joint B, thereby placing member AB in direct compression.
To assist in determining the buckling load for member
AB, we represent it as a pinned-end column, as shown in
part (b) of the figure. At the top of the column, a rotational
spring of stiffness bR represents the restraining action of
the horizontal beam BC on the column (note that the horizontal beam provides resistance to rotation of joint B when
the column buckles). Also, consider only bending effects in
the analysis (i.e., disregard the effects of axial deformations).
(a) By solving the differential equation of the deflection curve, derive the following buckling equation for this
column:
★
bRL
(kL cot kL 1) k 2L 2 0
EI
in which L is the length of the column and EI is its flexural
rigidity.
(b) For the particular case when member BC is identical to member AB, the rotational stiffness bR equals 3EI /L
(see Case 7, Table G-2, Appendix G). For this special case,
determine the critical load Pcr .
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 11 Problems
loads that have a resultant P 60 kN acting at the midpoint of one side of the cross section (see figure).
Assuming that the modulus of elasticity E is equal to
210 GPa and that the ends of the bar are pinned, calculate
the maximum deflection d and the maximum bending
moment Mmax.
x
P
P
C
B
819
bR
B
P = 60 kN
L
L
EI
y
A
(a)
m
50 mm 50 m
A
(b)
PROB. 11.4-11
PROB. 11.5-2
Columns with Eccentric Axial Loads
When solving the problems for Section 11.5, assume that
bending occurs in the principal plane containing the
eccentric axial load.
11.5-1 An aluminum bar having a rectangular cross section
(2.0 in. 1.0 in.) and length L 30 in. is compressed by
axial loads that have a resultant P 2800 lb acting at the
midpoint of the long side of the cross section (see figure).
Assuming that the modulus of elasticity E is equal to
10 10 6 psi and that the ends of the bar are pinned, calculate the maximum deflection d and the maximum bending
moment Mmax.
P = 2800 lb
11.5-3 Determine the bending moment M in the pinnedend column with eccentric axial loads shown in the figure.
Then plot the bending-moment diagram for an axial load
P 0.3Pcr .
Note: Express the moment as a function of the distance
x from the end of the column, and plot the diagram in
nondimensional form with M/Pe as ordinate and x /L as
abscissa.
x
P
P
e
M0 = Pe
B
v
.
1.0 in.
in
2.0
L
y
A
e
PROB. 11.5-1
P
11.5-2 A steel bar having a square cross section (50 mm
50 mm) and length L 2.0 m is compressed by axial
PROBS. 11.5-3 through 11.5-5
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
P
M0 = Pe
820
CHAPTER 11 Columns
11.5-4 Plot the load-deflection diagram for a pinned-end
column with eccentric axial loads (see figure on the preceding
page) if the eccentricity e of the load is 5 mm and the
column has length L 3.6 m, moment of inertia I
9.0 10 6 mm4, and modulus of elasticity E 210 GPa.
Note: Plot the axial load as ordinate and the deflection
at the midpoint as abscissa.
11.5-5 Solve the preceding problem for a column with e
0.20 in., L 12 ft, I 21.7 in.4, and E 30
11.5-9 The column shown in the figure is fixed at the base
and free at the upper end. A compressive load P acts at the
top of the column with an eccentricity e from the axis of the
column.
Beginning with the differential equation of the deflection curve, derive formulas for the maximum deflection d
of the column and the maximum bending moment Mmax in
the column.
10 6 psi.
x
P
11.5-6 A wide-flange member (W 8
15) is compressed
by axial loads that have a resultant P acting at the point
shown in the figure. The member has modulus of elasticity
E 29,000 ksi and pinned conditions at the ends. Lateral
supports prevent any bending about the weak axis of the
cross section.
If the length of the member is 20 ft and the deflection
is limited to 1/4 inch, what is the maximum allowable load
Pallow?
P
e
d
e
B
L
P
A
W8
15
y
(a)
(b)
PROB. 11.5-9
PROB. 11.5-6
11.5-7 A wide-flange member (W 10 30) is compressed
by axial loads that have a resultant P 20 k acting at the
point shown in the figure. The material is steel with modulus of elasticity E 29,000 ksi. Assuming pinned-end
conditions, determine the maximum permissible length
L max if the deflection is not to exceed 1/400th of the length.
11.5-10 An aluminum box column of square cross section
is fixed at the base and free at the top (see figure). The outside dimension b of each side is 100 mm and the thickness t
of the wall is 8 mm. The resultant of the compressive loads
acting on the top of the column is a force P 50 kN acting
at the outer edge of the column at the midpoint of one side.
What is the longest permissible length Lmax of the column if the deflection at the top is not to exceed 30 mm?
(Assume E 73 GPa.)
P
P = 20 k
t
W 10
30
A
A
L
b
Section A-A
PROB. 11.5-7
11.5-8 Solve the preceding problem (W 10
resultant force P equals 25 k.
30) if the
PROBS. 11.5-10 and 11.5-11
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
821
CHAPTER 11 Problems
11.5-11 Solve the preceding problem for an aluminum
A
column with b 6.0 in., t 0.5 in., P 30 k, and E
10.6 103 ksi. The deflection at the top is limited to 2.0 in.
B
D
h
E
q0
e = 100 mm
B
L = 4.0 m
Cable
d1
d2
a = 53.13°
C
d2
A
q0
E
11.5-12 A steel post AB of hollow circular cross section is
fixed at the base and free at the top (see figure). The inner
and outer diameters are d1 96 mm and d 2 110 mm,
respectively, and the length L 4.0 m.
A cable CBD passes through a fitting that is welded to
the side of the post. The distance between the plane of the
cable (plane CBD) and the axis of the post is e 100 mm,
and the angles between the cable and the ground are a
53.13°. The cable is pretensioned by tightening the turnbuckles.
If the deflection at the top of the post is limited to d
20 mm, what is the maximum allowable tensile force T in
the cable? (Assume E 205 GPa.)
C
L
Section E-E
PROB. 11.5-13
The Secant Formula
When solving the problems for Section 11.6, assume that
bending occurs in the principal plane containing the
eccentric axial load.
11.6-1 A steel bar has a square cross section of width b
2.0 in. (see figure). The bar has pinned supports at the ends
and is 3.0 ft long. The axial forces acting at the end of the
bar have a resultant P 20 k located at distance e
0.75 in. from the center of the cross section. Also, the
modulus of elasticity of the steel is 29,000 ksi.
(a) Determine the maximum compressive stress smax
in the bar.
(b) If the allowable stress in the steel is 18,000 psi,
what is the maximum permissible length L max of the bar?
a = 53.13°
P
e
D
PROB. 11.5-12
b
11.5-13 A frame ABCD is constructed of steel wide-flange
members ( W 8 21; E 30 10 6 psi) and subjected to
triangularly distributed loads of maximum intensity q0 acting along the vertical members (see figure). The distance
between supports is L 20 ft and the height of the frame is
h 4 ft. The members are rigidly connected at B and C.
(a) Calculate the intensity of load q0 required to produce a maximum bending moment of 80 k-in. in the
horizontal member BC.
(b) If the load q0 is reduced to one-half of the value
calculated in part (a), what is the maximum bending
moment in member BC? What is the ratio of this moment to
the moment of 80 k-in. in part (a)?
b
PROBS. 11.6-1 through 11.6-3
11.6-2 A brass bar (E 100 GPa) with a square cross section is subjected to axial forces having a resultant P acting
at distance e from the center (see figure). The bar is pin
supported at the ends and is 0.6 m in length. The side
dimension b of the bar is 30 mm and the eccentricity e of
the load is 10 mm.
If the allowable stress in the brass is 150 MPa, what is
the allowable axial force Pallow?
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
822
CHAPTER 11 Columns
11.6-3 A square aluminum bar with pinned ends carries a
load P 25 k acting at distance e 2.0 in. from the center
(see figure on the previous page). The bar has length
L 54 in. and modulus of elasticity E 10,600 ksi.
If the stress in the bar is not to exceed 6 ksi, what is
the minimum permissible width bmin of the bar?
11.6-4 A pinned-end column of length L 2.1 m is con-
structed of steel pipe (E 210 GPa) having inside diameter
d1 60 mm and outside diameter d 2 68 mm (see figure).
A compressive load P 10 kN acts with eccentricity e
30 mm.
(a) What is the maximum compressive stress smax in
the column?
(b) If the allowable stress in the steel is 50 MPa, what
is the maximum permissible length Lmax of the column?
11.6-7 A steel column (E 30
10 3 ksi) with pinned
ends is constructed of a W 10 60 wide-flange shape (see
figure). The column is 24 ft long. The resultant of the axial
loads acting on the column is a force P acting with an
eccentricity e 2.0 in.
(a) If P 120 k, determine the maximum compressive
stress smax in the column.
(b) Determine the allowable load Pallow if the yield
stress is sY 42 ksi and the factor of safety with respect to
yielding of the material is n 2.5.
P
e = 2.0 in.
W 10
60
P
e
PROB. 11.6-7
d1
d2
PROBS. 11.6-4 through 11.6-6
11.6-5 A pinned-end strut of length L 5.2 ft is con-
structed of steel pipe (E 30
10 3 ksi) having inside
diameter d1 2.0 in. and outside diameter d 2 2.2 in. (see
figure). A compressive load P 2.0 k is applied with
eccentricity e 1.0 in.
(a) What is the maximum compressive stress smax in
the strut?
(b) What is the allowable load Pallow if a factor of
safety n 2 with respect to yielding is required? (Assume
that the yield stress sY of the steel is 42 ksi.)
11.6-8 A W 16
57 steel column is compressed by a force
P 75 k acting with an eccentricity e 1.5 in., as shown
in the figure. The column has pinned ends
and length L. Also, the steel has modulus of elasticity E
30 103 ksi and yield stress sY 36 ksi.
(a) If the length L 10 ft, what is the maximum compressive stress smax in the column?
(b) If a factor of safety n 2.0 is required with respect
to yielding, what is the longest permissible length L max of
the column?
P = 75 k
e = 1.5 in.
W 16
11.6-6 A circular aluminum tube with pinned ends supports a load P 18 kN acting at distance e 50 mm from
the center (see figure). The length of the tube is 3.5 m and
its modulus of elasticity is 73 GPa.
If the maximum permissible stress in the tube is
20 MPa, what is the required outer diameter d 2 if the ratio
of diameters is to be d1 /d 2 0.9?
PROB. 11.6-8
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
57
CHAPTER 11 Problems
10 3 ksi) that is fixed at
the base and free at the top is constructed of a W 8
35
wide-flange member (see figure). The column is 9.0 ft long.
The force P acting at the top of the column has an eccentricity e 1.25 in.
(a) If P 40 k, what is the maximum compressive
stress in the column?
(b) If the yield stress is 36 ksi and the required factor
of safety with respect to yielding is 2.1, what is the allowable load Pallow?
11.6-9 A steel column (E 30
P
e
P2
s
823
P1
Wide-flange
column
PROBS. 11.6-11 and 11.6-12
11.6-12 The wide-flange pinned-end column shown in the
e
A
A
P
L
Section A-A
PROBS. 11.6-9 and 11.6-10
11.6-10 A W 12
50 wide-flange steel column with
length L 12.5 ft is fixed at the base and free at the top
(see figure). The load P acting on the column is intended
to be centrally applied, but because of unavoidable discrepancies in construction, an eccentricity ratio of 0.25
is specified. Also, the following data are supplied: E
30 10 3 ksi, sY 42 ksi, and P 70 k.
(a) What is the maximum compressive stress smax in
the column?
(b) What is the factor of safety n with respect to yielding of the steel?
11.6-11 A pinned-end column with length L 18 ft is
constructed from a W 12
87 wide-flange shape (see
figure). The column is subjected to a centrally applied load
P1 180 k and an eccentrically applied load P2 75 k.
The load P2 acts at distance s 5.0 in. from the centroid of
the cross section. The properties of the steel are E
29,000 ksi and sY 36 ksi.
(a) Calculate the maximum compressive stress in the
column.
(b) Determine the factor of safety with respect to
yielding.
figure carries two loads, a force P1 100 k acting at the
centroid and a force P2 60 k acting at distance s 4.0 in.
from the centroid. The column is a W 10 45 shape with
L 13.5 ft, E 29 10 3 ksi, and sY 42 ksi.
(a) What is the maximum compressive stress in the
column?
(b) If the load P1 remains at 100 k, what is the largest
permissible value of the load P2 in order to maintain a factor of safety of 2.0 with respect to yielding?
11.6-13 A W 14 53 wide-flange column of length L
15 ft is fixed at the base and free at the top (see figure). The
column supports a centrally applied load P1 120 k and a
load P2 40 k supported on a bracket. The distance from
the centroid of the column to the load P2 is s 12 in. Also,
the modulus of elasticity is E 29,000 ksi and the yield
stress is sY 36 ksi.
(a) Calculate the maximum compressive stress in the
column.
(b) Determine the factor of safety with respect to
yielding.
P1
P2
s
L
A
PROBS. 11.6-13 and 11.6-14
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
A
Section A-A
824
CHAPTER 11 Columns
11.6-14 A wide-flange column with a bracket is fixed at
the base and free at the top (see figure on the preceding
page). The column supports a load P1 75 k acting at the
centroid and a load P2 25 k acting on the bracket at
distance s 10.0 in. from the load P1. The column is a
W 12
35 shape with L 16 ft, E 29
103 ksi, and
sY 42 ksi.
(a) What is the maximum compressive stress in the
column?
(b) If the load P1 remains at 75 k, what is the largest
permissible value of the load P2 in order to maintain a factor of safety of 1.8 with respect to yielding?
Design Formulas for Columns
The problems for Section 11.9 are to be solved assuming
that the axial loads are centrally applied at the ends of the
columns. Unless otherwise stated, the columns may buckle
in any direction.
Steel Columns
11.9-1 Determine the allowable axial load Pallow for a
W 10 45 steel wide-flange column with pinned ends (see
figure) for each of the following lengths: L 8 ft, 16 ft,
24 ft, and 32 ft. (Assume E 29,000 ksi and sY 36 ksi.)
figure). The column has pinned ends and length L 14 ft.
Assume E 29,000 ksi and sY 36 ksi. (Note: The selection of columns is limited to those listed in Table E-1,
Appendix E.)
11.9-5 Select a steel wide-flange column of nominal depth
12 in. (W 12 shape) to support an axial load P 175 k (see
figure). The column has pinned ends and length L 35 ft.
Assume E 29,000 ksi and sY 36 ksi. (Note: The selection of columns is limited to those listed in Table E-1,
Appendix E.)
11.9-6 Select a steel wide-flange column of nominal depth
14 in. (W 14 shape) to support an axial load P 250 k (see
figure). The column has pinned ends and length L 20 ft.
Assume E 29,000 ksi and sY 50 ksi. (Note: The selection of columns is limited to those listed in Table E-1,
Appendix E.)
11.9-7 Determine the allowable axial load Pallow for a steel
pipe column with pinned ends having an outside diameter
of 4.5 in. and wall thickness of 0.237 in. for each of the following lengths: L 6 ft, 12 ft, 18 ft, and 24 ft. (Assume
E 29,000 ksi and sY 36 ksi.)
11.9-8 Determine the allowable axial load Pallow for a steel
pipe column with pinned ends having an outside diameter
of 220 mm and wall thickness of 12 mm for each of the following lengths: L 2.5 m, 5 m, 7.5 m, and 10 m. (Assume
E 200 GPa and sY 250 MPa.)
P
11.9-9 Determine the allowable axial load Pallow for a steel
L
A
A
Section A - A
pipe column that is fixed at the base and free at the top (see
figure) for each of the following lengths: L 6 ft, 9 ft,
12 ft, and 15 ft. The column has outside diameter d
6.625 in. and wall thickness t 0.280 in. (Assume E
29,000 ksi and sY 36 ksi.)
P
PROBS. 11.9-1 through 11.9-6
11.9-2 Determine the allowable axial load Pallow for a
W 12 87 steel wide-flange column with pinned ends (see
figure) for each of the following lengths: L 10 ft, 20 ft,
30 ft, and 40 ft. (Assume E 29,000 ksi and sY 50 ksi.)
t
A
L
d
11.9-3 Determine the allowable axial load Pallow for a
W 10 60 steel wide-flange column with pinned ends (see
figure) for each of the following lengths: L 10 ft, 20 ft,
30 ft, and 40 ft. (Assume E 29,000 ksi and sY 36 ksi.)
11.9-4 Select a steel wide-flange column of nominal depth
10 in. (W 10 shape) to support an axial load P 180 k (see
A
Section A-A
PROBS. 11.9-9 through 11.9-12
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 11 Problems
825
11.9-10 Determine the allowable axial load Pallow for a
steel pipe column that is fixed at the base and free at the top
(see figure) for each of the following lengths: L 2.6 m,
2.8 m, 3.0 m, and 3.2 m. The column has outside diameter
d 140 mm and wall thickness t 7 mm. (Assume E
200 GPa and sY 250 MPa.)
PROBS. 11.9-15 and 11.9-16
11.9-11 Determine the maximum permissible length Lmax
for a steel pipe column that is fixed at the base and free at
the top and must support an axial load P 40 k (see figure). The column has outside diameter d 4.0 in., wall
thickness t 0.226 in., E 29,000 ksi, and sY 42 ksi.
11.9-16 A W 10
45 steel wide-flange column with
pinned ends carries an axial load P. What is the maximum
permissible length Lmax of the column if (a) P 125 k, and
(b) P 200 k? (Assume E 29,000 ksi and sY 42 ksi.)
11.9-12 Determine the maximum permissible length Lmax
for a steel pipe column that is fixed at the base and free at
the top and must support an axial load P 500 kN (see
figure). The column has outside diameter d 200 mm,
wall thickness t 10 mm, E 200 GPa, and sY
250 MPa.
11.9-17 Find the required outside diameter d for a steel
pipe column (see figure) of length L 20 ft that is pinned
at both ends and must support an axial load P 25 k.
Assume that the wall thickness t is equal to d /20. (Use E
29,000 ksi and sY 36 ksi.)
11.9-13 A steel pipe column with pinned ends supports an
axial load P 21 k. The pipe has outside and inside diameters of 3.5 in. and 2.9 in., respectively. What is the
maximum permissible length Lmax of the column if E
29,000 ksi and sY 36 ksi?
t
d
11.9-14 The steel columns used in a college recreation
center are 55 ft long and are formed by welding three wideflange sections (see figure). The columns are pin-supported
at the ends and may buckle in any direction.
Calculate the allowable load Pallow for one column,
assuming E 29,000 ksi and sY 36 ksi.
W 12
11.9-18 Find the required outside diameter d for a steel
pipe column (see figure) of length L 3.5 m that is pinned
at both ends and must support an axial load P 130 kN.
Assume that the wall thickness t is equal to d /20. (Use E
200 GPa and sY 275 MPa).
87
W 24
W 12
PROBS. 11.9-17 through 11.9-20
162
87
11.9-19 Find the required outside diameter d for a steel
pipe column (see figure) of length L 11.5 ft that is
pinned at both ends and must support an axial load P 80
k. Assume that the wall thickness t is 0.30 in. (Use E
29,000 ksi and sY 42 ksi.)
PROB. 11.9-14
11.9-15 A W 8
28 steel wide-flange column with pinned
ends carries an axial load P. What is the maximum permissible length Lmax of the column if (a) P 50 k, and
(b) P 100 k? (Assume E 29,000 ksi and sY 36 ksi.)
11.9-20 Find the required outside diameter d for a steel
pipe column (see figure) of length L 3.0 m that is pinned
at both ends and must support an axial load P 800 kN.
Assume that the wall thickness t is 9 mm. (Use E
200 GPa and sY 300 MPa.)
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
826
CHAPTER 11 Columns
Aluminum Columns
11.9-26 A solid round bar of aluminum having diameter d
11.9-21 An aluminum pipe column (alloy 2014-T6) with
(see figure) is compressed by an axial force P 175 kN.
The bar has pinned supports and is made of alloy 2014-T6.
(a) If the diameter d 40 mm, what is the maximum
allowable length L max of the bar?
(b) If the length L 0.6 m, what is the minimum
required diameter d min?
(Hint: Convert the given data to USCS units, determine the required quantities, and then convert back to SI
units.)
pinned ends has outside diameter d 2 5.60 in. and inside
diameter d1 4.80 in. (see figure).
Determine the allowable axial load Pallow for each of
the following lengths: L 6 ft, 8 ft, 10 ft, and 12 ft.
d1 d2
11.9-27 A solid round bar of aluminum having diameter d
PROBS. 11.9-21 through 11.9-24
11.9-22 An aluminum pipe column (alloy 2014-T6) with
pinned ends has outside diameter d 2 120 mm and inside
diameter d1 110 mm (see figure).
Determine the allowable axial load Pallow for each of
the following lengths: L 1.0 m, 2.0 m, 3.0 m, and 4.0 m.
(Hint: Convert the given data to USCS units, determine the required quantities, and then convert back to SI
units.)
11.9-23 An aluminum pipe column (alloy 6061-T6) that is
fixed at the base and free at the top has outside diameter
d 2 3.25 in. and inside diameter d1 3.00 in. (see figure).
Determine the allowable axial load Pallow for each of
the following lengths: L 2 ft, 3 ft, 4 ft, and 5 ft.
11.9-24 An aluminum pipe column (alloy 6061-T6) that is
fixed at the base and free at the top has outside diameter
d 2 80 mm and inside diameter d1 72 mm (see figure).
Determine the allowable axial load Pallow for each of
the following lengths: L 0.6 m, 0.8 m, 1.0 m, and 1.2 m.
(Hint: Convert the given data to USCS units, determine the required quantities, and then convert back to SI
units.)
11.9-25 A solid round bar of aluminum having diameter d
(see figure) is compressed by an axial force P 60 k. The
bar has pinned supports and is made of alloy 2014-T6.
(a) If the diameter d 2.0 in., what is the maximum
allowable length L max of the bar?
(b) If the length L 30 in., what is the minimum
required diameter d min?
(see figure) is compressed by an axial force P 10 k. The
bar has pinned supports and is made of alloy 6061-T6.
(a) If the diameter d 1.0 in., what is the maximum
allowable length L max of the bar?
(b) If the length L 20 in., what is the minimum
required diameter dmin?
11.9-28 A solid round bar of aluminum having diameter d
(see figure) is compressed by an axial force P 60 kN.
The bar has pinned supports and is made of alloy 6061-T6.
(a) If the diameter d 30 mm, what is the maximum
allowable length Lmax of the bar?
(b) If the length L 0.6 m, what is the minimum
required diameter d min?
(Hint: Convert the given data to USCS units, determine the required quantities, and then convert back to SI
units.)
Wood Columns
When solving the problems for wood columns, assume that
the columns are constructed of sawn lumber (c 0.8 and
KcE 0.3) and have pinned-end conditions. Also, buckling
may occur about either principal axis of the cross section.
11.9-29 A wood post of rectangular cross section (see
figure) is constructed of 4 in.
6 in. structural grade,
Douglas fir lumber (Fc 2,000 psi, E 1,800,00 psi). The
net cross-sectional dimensions of the post are b 3.5 in.
and h 5.5 in. (see Appendix F).
Determine the allowable axial load Pallow for each of
the following lengths: L 5.0 ft, 7.5 ft, and 10.0 ft.
h
d
PROBS. 11.9-25 through 11.9-28
b
PROB. 11.9-29 through 11.9-32
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
CHAPTER 11 Problems
11.9-30 A wood post of rectangular cross section (see
figure) is constructed of structural grade, southern pine
lumber (Fc 14 MPa, E 12 GPa). The cross-sectional
dimensions of the post (actual dimensions) are b 100 mm
and h 150 mm.
Determine the allowable axial load Pallow for each of
the following lengths: L 1.5 m, 2.0 m, and 2.5 m.
11.9-31 A wood column of rectangular cross section (see
figure) is constructed of 4 in.
8 in. construction
grade, western hemlock lumber (Fc 1,000 psi, E
1,300,000 psi). The net cross-sectional dimensions of the
column are b 3.5 in. and h 7.25 in. (see Appendix F).
Determine the allowable axial load Pallow for each of
the following lengths: L 6 ft, 8 ft, and 10 ft.
11.9-32 A wood column of rectangular cross section (see
figure) is constructed of structural grade, Douglas fir lumber (Fc 12 MPa, E 10 GPa). The cross-sectional
dimensions of the column (actual dimensions) are b 140
mm and h 210 mm.
Determine the allowable axial load Pallow for each of
the following lengths: L 2.5 m, 3.5 m, and 4.5 m.
11.9-33 A square wood column with side dimensions b
(see figure) is constructed of a structural grade of Douglas
fir for which Fc 1,700 psi and E 1,400,000 psi. An
axial force P 40 k acts on the column.
(a) If the dimension b 5.5 in., what is the maximum
allowable length Lmax of the column?
(b) If the length L 11 ft, what is the minimum
required dimension bmin?
827
11.9-34 A square wood column with side dimensions b
(see figure) is constructed of a structural grade of southern
pine for which Fc 10.5 MPa and E 12 GPa. An axial
force P 200 kN acts on the column.
(a) If the dimension b 150 mm, what is the maximum allowable length Lmax of the column?
(b) If the length L 4.0 m, what is the minimum
required dimension bmin?
11.9-35 A square wood column with side dimensions b
(see figure) is constructed of a structural grade of spruce for
which Fc 900 psi and E 1,500,000 psi. An axial force
P 8.0 k acts on the column.
(a) If the dimension b 3.5 in., what is the maximum
allowable length Lmax of the column?
(b) If the length L 10 ft, what is the minimum
required dimension bmin?
11.9-36 A square wood column with side dimensions b
(see figure) is constructed of a structural grade of eastern
white pine for which Fc 8.0 MPa and E 8.5 GPa. An
axial force P 100 kN acts on the column.
(a) If the dimension b 120 mm, what is the maximum allowable length Lmax of the column?
(b) If the length L 4.0 m, what is the minimum
required dimension bmin?
b
b
PROBS. 11.9-33 through 11.9-36
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
12
Review of Centroids and
Moments of Inertia
12.1 INTRODUCTION
This chapter is a review of the definitions and formulas pertaining to
centroids and moments of inertia of plane areas. The word “review” is
appropriate because these topics are usually covered in earlier courses,
such as mathematics and engineering statics, and therefore most readers
will already have been exposed to the material. However, since centroids and moments of inertia are used repeatedly throughout the
preceding chapters, they must be clearly understood by the reader and
the essential definitions and formulas must be readily accessible.
Topics covered in this chapter include centroids and how to locate
them, moments of inertia, polar moments of inertia, products of inertia,
parallel-axis theorems, rotation of axes, and principal axes. Only plane
areas are considered.
A table of centroids and moments of inertia for a variety of common
geometric shapes is given in Appendix D for convenient reference.
The terminology used in this and earlier chapters may appear
puzzling to some readers. For instance, the term “moment of inertia” is
clearly a misnomer when referring to properties of an area, since no
mass is involved. Even the word “area” is used inappropriately. When
we say “plane area,” we really mean “plane surface.” Strictly speaking,
area is a measure of the size of a surface and is not the same thing as the
surface itself. In spite of its deficiencies, the terminology used in this
book is so entrenched in the engineering literature that it rarely causes
confusion.
828
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 12.2 Centroids of Plane Areas
829
12.2 CENTROIDS OF PLANE AREAS
The position of the centroid of a plane area is an important geometric
property. To obtain formulas for locating centroids, we will refer to
Fig. 12-1, which shows a plane area of irregular shape with its centroid
at point C. The xy coordinate system is oriented arbitrarily with its origin at any point O. The area of the geometric figure is defined by the
following integral:
y
x
C
x
dA
y
y
A
(12-1)
dA
x
O
FIG. 12-1 Plane area of arbitrary shape
with centroid C
in which dA is a differential element of area having coordinates x and y
(Fig. 12-1) and A is the total area of the figure.
The first moments of the area with respect to the x and y axes are
defined, respectively, as follows:
Qx
y dA
Qy
x dA
(12-2a,b)
Thus, the first moments represent the sums of the products of the differential areas and their coordinates. First moments may be positive or
negative, depending upon the position of the xy axes. Also, first
moments have units of length raised to the third power; for instance, in.3
or mm3.
The coordinates x and y of the centroid C (Fig. 12-1) are equal to
the first moments divided by the area:
x dA
Qy
x
A
dA
y
C
x
FIG. 12-2 Area with one axis of
symmetry
y dA
Q
y x
A
dA
(12-3a,b)
If the boundaries of the area are defined by simple mathematical expressions, we can evaluate the integrals appearing in Eqs. (12-3a) and
(12-3b) in closed form and thereby obtain formulas for x and y. The
formulas listed in Appendix D were obtained in this manner. In general,
the coordinates x and y may be positive or negative, depending upon the
position of the centroid with respect to the reference axes.
If an area is symmetric about an axis, the centroid must lie on that
axis because the first moment about an axis of symmetry equals zero.
For example, the centroid of the singly symmetric area shown in
Fig. 12-2 must lie on the x axis, which is the axis of symmetry. Therefore, only one coordinate must be calculated in order to locate the
centroid C.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
830
CHAPTER 12 Review of Centroids and Moments of Inertia
y
y
C
x
C
FIG. 12-3 Area with two axes of
x
FIG. 12-4 Area that is symmetric about a
symmetry
point
If an area has two axes of symmetry, as illustrated in Fig. 12-3, the
position of the centroid can be determined by inspection because it lies
at the intersection of the axes of symmetry.
An area of the type shown in Fig. 12-4 is symmetric about a point.
It has no axes of symmetry, but there is a point (called the center of
symmetry) such that every line drawn through that point contacts the
area in a symmetrical manner. The centroid of such an area coincides
with the center of symmetry, and therefore the centroid can be located
by inspection.
If an area has irregular boundaries not defined by simple mathematical expressions, we can locate the centroid by numerically
evaluating the integrals in Eqs. (12-3a) and (12-3b). The simplest procedure is to divide the geometric figure into small finite elements and
replace the integrations with summations. If we denote the area of the
ith element by ⌬Ai, then the expressions for the summations are
n
n
A ⫽ ⌬ Ai
n
Qx ⫽ yi⌬Ai
i⫽1
Qy ⫽ xi ⌬Ai
i⫽1
(12-4a,b,c)
i⫽1
in which n is the total number of elements, yi is the y coordinate of the
centroid of the ith element, and xi is the x coordinate of the centroid of
the ith element. Replacing the integrals in Eqs. (12-3a) and (12-3b) by
the corresponding summations, we obtain the following formulas for the
coordinates of the centroid:
n
xi ⌬Ai
Qy
i⫽1
x ⫽ ᎏAᎏ ⫽ ᎏ
n
⌬Ai
i⫽1
n
yi ⌬Ai
Qx
i⫽1
(12-5a,b)
y ⫽ ᎏAᎏ ⫽ ᎏᎏ
n
⌬Ai
i⫽1
The accuracy of the calculations for x and y depends upon how closely
the selected elements fit the actual area. If they fit exactly, the results are
exact. Many computer programs for locating centroids use a numerical
scheme similar to the one expressed by Eqs. (12-5a) and (12-5b).
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 12.2 Centroids of Plane Areas
831
Example 12-1
y
A
A parabolic semisegment OAB is bounded by the x axis, the y axis, and a parabolic curve having its vertex at A (Fig. 12-5). The equation of the curve is
y = f (x)
冢
x2
y f (x) h 1 2
b
dA
x
h
C
O
y
y
2
(a)
in which b is the base and h is the height of the semisegment.
Locate the centroid C of the semisegment.
y
B
dx
x
冣
b
FIG. 12-5 Example 12-1. Centroid of a
parabolic semisegment
x
Solution
To determine the coordinates x and y of the centroid C (Fig. 12-5), we will
use Eqs. (12-3a) and (12-3b). We begin by selecting an element of area dA in
the form of a thin vertical strip of width dx and height y. The area of this differential element is
冢
冣
x2
dA y dx h 1 2 dx
b
(b)
Therefore, the area of the parabolic semisegment is
b
冢
冣
x2
2bh
h 1 2 dx
b
3
A dA
0
(c)
Note that this area is 2/3 of the area of the surrounding rectangle.
The first moment of an element of area dA with respect to an axis is
obtained by multiplying the area of the element by the distance from its centroid
to the axis. Since the x and y coordinates of the centroid of the element shown in
Fig. 12-5 are x and y /2, respectively, the first moments of the element with
respect to the x and y axes are
Qx
y
dA
2
b
0
b
Qy
冢
冣
(d)
冢
冣
(e)
x2 2
h2
4bh2
1 2 dx
b
2
15
x dA
0
x2
b2h
hx 1 2 dx
b
4
in which we have substituted for dA from Eq. (b).
We can now determine the coordinates of the centroid C:
Qy
3b
x
A
8
Qx
2h
y
A
5
(f,g)
These results agree with the formulas listed in Appendix D, Case 17.
Notes: The centroid C of the parabolic semisegment may also be located by
taking the element of area dA as a horizontal strip of height dy and width
xb
y
冪1莦
h
This expression is obtained by solving Eq. (a) for x in terms of y.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
(h)
832
CHAPTER 12 Review of Centroids and Moments of Inertia
12.3 CENTROIDS OF COMPOSITE AREAS
In engineering work we rarely need to locate centroids by integration,
because the centroids of common geometric figures are already known
and tabulated. However, we frequently need to locate the centroids of
areas composed of several parts, each part having a familiar geometric
shape, such as a rectangle or a circle. Examples of such composite areas
are the cross sections of beams and columns, which usually consist of
rectangular elements (for instance, see Figs. 12-2, 12-3, and 12-4).
The areas and first moments of composite areas may be calculated
by summing the corresponding properties of the component parts. Let us
assume that a composite area is divided into a total of n parts, and let us
denote the area of the ith part as Ai. Then we can obtain the area and
first moments by the following summations:
y
n
A Ai
t
i1
n
Qx yi Ai
i1
n
Qy xi Ai
(12-6a,b,c)
i1
in which xi and yi are the coordinates of the centroid of the ith part.
The coordinates of the centroid of the composite area are
b
n
xi Ai
Qy
i1
x
n
A
Ai
t
O
x
c
i1
y
x
A1
A1 bt
x1
C1
A2 (c t)t
C
y1
A2 y
C2
y2
x2
(12-7a,b)
i1
t
x1 2
c⫹t
x2 2
b
y1
2
t
y2
2
Therefore, the area and first moments of the composite area (from Eqs.
12-6a, b, and c) are
x
(b)
FIG. 12-6 Centroid of a composite area
consisting of two parts
yi Ai
Qx
i1
y
n
A
Ai
Since the composite area is represented exactly by the n parts, the
preceding equations give exact results for the coordinates of the centroid.
To illustrate the use of Eqs. (12-7a) and (12-7b), consider the
L-shaped area (or angle section) shown in Fig. 12-6a. This area has side
dimensions b and c and thickness t. The area can be divided into two
rectangles of areas A1 and A2 with centroids C1 and C2, respectively
(Fig. 12-6b). The areas and centroidal coordinates of these two parts are
(a)
O
n
A A1 ⫹ A2 t(b ⫹ c t)
t
Qx y1A1 ⫹ y2 A2 (b2 ⫹ ct t 2)
2
t
Qy x1A1 ⫹ x2 A2 (bt ⫹ c2 t 2)
2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 12.3 Centroids of Composite Areas
b
a
c
f
g
e
h
d
(a)
(b)
FIG. 12-7 Composite areas with a cutout
and a hole
833
Finally, we can obtain the coordinates x and y of the centroid C of the
composite area (Fig. 12-6b) from Eqs. (12-7a) and (12-7b):
Qy
bt ⫹ c2 t 2
x A
2(b ⫹ c t)
Qx
b2 ⫹ ct t 2
y
A
2(b ⫹ c t)
(12-8a,b)
A similar procedure can be used for more complex areas, as illustrated
in Example 12-2.
Note 1: When a composite area is divided into only two parts, the
centroid C of the entire area lies on the line joining the centroids C1 and
C2 of the two parts (as shown in Fig. 12-6b for the L-shaped area).
Note 2: When using the formulas for composite areas (Eqs. 12-6
and 12-7), we can handle the absence of an area by subtraction. This
procedure is useful when there are cutouts or holes in the figure.
For instance, consider the area shown in Fig. 12-7a. We can analyze
this figure as a composite area by subtracting the properties of the inner
rectangle efgh from the corresponding properties of the outer rectangle
abcd. (From another viewpoint, we can think of the outer rectangle as a
“positive area” and the inner rectangle as a “negative area.”)
Similarly, if an area has a hole (Fig. 12-7b), we can subtract the
properties of the area of the hole from those of the outer rectangle.
(Again, the same effect is achieved if we treat the outer rectangle as a
“positive area” and the hole as a “negative area.”)
Example 12-2
The cross section of a steel beam is constructed of a W 18 71 wide-flange section with a 6 in. 1/2 in. cover plate welded to the top flange and a C 10 30
channel section welded to the bottom flange (Fig. 12-8).
Locate the centroid C of the cross-sectional area.
Plate 1
6 in. —
2 in.
y
C1
W 18 71
y1
C2
C
x
c
y3
C 10 30
FIG. 12-8 Example 12-2. Centroid of a
composite area
C3
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
continued
834
CHAPTER 12 Review of Centroids and Moments of Inertia
Solution
Plate 1
6 in. —
2 in.
y
Let us denote the areas of the cover plate, the wide-flange section, and the
channel section as areas A1, A2, and A3, respectively. The centroids of these
three areas are labeled C1, C2, and C3, respectively, in Fig. 12-8. Note that the
composite area has an axis of symmetry, and therefore all centroids lie on that
axis. The three partial areas are
C1
W 18 71
A1 (6 in.)(0.5 in.) 3.0 in.2
y1
C2
C
x
c
y3
A3 8.82 in.2
in which the areas A2 and A3 are obtained from Tables E-1 and E-3 of Appendix E.
Let us place the origin of the x and y axes at the centroid C2 of the wideflange section. Then the distances from the x axis to the centroids of the three
areas are as follows:
18.47 in.
0.5 in.
y1 ⫹ 9.485 in.
2
2
C 10 30
C3
FIG. 12-8 (Repeated)
A2 20.8 in.2
y2 0
18.47 in.
y3 2 ⫹ 0.649 in. 9.884 in.
in which the pertinent dimensions of the wide-flange and channel sections are
obtained from Tables E-1 and E-3.
The area A and first moment Qx of the entire cross section are obtained
from Eqs. (12-6a) and (12-6b) as follows:
n
A Ai A1 ⫹ A2 ⫹ A3
i1
3.0 in.2 ⫹ 20.8 in.2 ⫹ 8.82 in.2 32.62 in.2
n
Qx yi Ai y1 A1 ⫹ y2 A2 ⫹ y3 A3
i1
(9.485 in.)(3.0 in.2) ⫹ 0 (9.884 in.)(8.82 in.2) 58.72 in.3
Now we can obtain the coordinate y to the centroid C of the composite area
from Eq. (12-7b):
Qx
58.72 in.3
1.80 in.
y A
32.62 in.2
Since y is positive in the positive direction of the y axis, the minus sign means
that the centroid C of the composite area is located below the x axis, as shown
in Fig. 12-8. Thus, the distance c between the x axis and the centroid C is
c 1.80 in.
Note that the position of the reference axis (the x axis) is arbitrary; however, in
this example we placed it through the centroid of the wide-flange section
because it slightly simplifies the calculations.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 12.4 Moments of Inertia of Plane Areas
835
12.4 MOMENTS OF INERTIA OF PLANE AREAS
y
The moments of inertia of a plane area (Fig. 12-9) with respect to the x
and y axes, respectively, are defined by the integrals
C
x
y
x
O
FIG. 12-9 Plane area of arbitrary shape
y
dA
h
–
2
B
dy
y
x
C
h
–
2
b
–
2
b
–
2
B
FIG. 12-10 Moments of inertia of a
rectangle
y 2dA
Ix
dA
Iy
x2dA
(12-9a,b)
in which x and y are the coordinates of the differential element of area
dA. Because the element dA is multiplied by the square of the distance
from the reference axis, moments of inertia are also called second
moments of area. Also, we see that moments of inertia of areas (unlike
first moments) are always positive quantities.
To illustrate how moments of inertia are obtained by integration, we
will consider a rectangle having width b and height h (Fig. 12-10). The x
and y axes have their origin at the centroid C. For convenience, we use a
differential element of area dA in the form of a thin horizontal strip of
width b and height dy (therefore, dA b dy). Since all parts of the elemental strip are the same distance from the x axis, we can express the
moment of inertia Ix with respect to the x axis as follows:
Ix
y 2dA
bh3
y 2b dy
12
h/2
h/2
(a)
In a similar manner, we can use an element of area in the form of a vertical strip with area dA h dx and obtain the moment of inertia with
respect to the y axis:
Iy
x 2dA
b/2
hb3
x2h dx
12
b/2
(b)
If a different set of axes is selected, the moments of inertia will have
different values. For instance, consider axis BB at the base of the rectangle (Fig. 12-10). If this axis is selected as the reference, we must define
y as the coordinate distance from that axis to the element of area dA.
Then the calculations for the moment of inertia become
IBB
h
2
y dA
0
bh3
y2b dy
3
(c)
Note that the moment of inertia with respect to axis BB is larger than the
moment of inertia with respect to the centroidal x axis. In general, the
moment of inertia increases as the reference axis is moved parallel to
itself farther from the centroid.
The moment of inertia of a composite area with respect to any particular axis is the sum of the moments of inertia of its parts with respect
to that same axis. An example is the hollow box section shown in
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
836
CHAPTER 12 Review of Centroids and Moments of Inertia
y
h
x h1
C
b1
Fig. 12-11a, where the x and y axes are axes of symmetry through the
centroid C. The moment of inertia Ix with respect to the x axis is equal to
the algebraic sum of the moments of inertia of the outer and inner rectangles. (As explained earlier, we can think of the inner rectangle as a
“negative area” and the outer rectangle as a “positive area.”) Therefore,
b1h13
bh3
Ix
12
12
b
(a)
y
h
x
C
b1
b
(b)
h1
(d)
This same formula applies to the channel section shown in Fig. 12-11b,
where we may consider the cutout as a “negative area.”
For the hollow box section, we can use a similar technique to obtain
the moment of inertia Iy with respect to the vertical axis. However, in the
case of the channel section, the determination of the moment of inertia Iy
requires the use of the parallel-axis theorem, which is described in the
next section (Section 12.5).
Formulas for moments of inertia are listed in Appendix D. For
shapes not shown, the moments of inertia can usually be obtained by
using the listed formulas in conjunction with the parallel-axis theorem.
If an area is of such irregular shape that its moments of inertia cannot be
obtained in this manner, then we can use numerical methods. The procedure is to divide the area into small elements of area ⌬Ai, multiply each
such area by the square of its distance from the reference axis, and then
sum the products.
FIG. 12-11 Composite areas
Radius of Gyration
A distance known as the radius of gyration is occasionally encountered
in mechanics. Radius of gyration of a plane area is defined as the square
root of the moment of inertia of the area divided by the area itself; thus,
rx ⫽
冪ᎏ莦Aᎏ
Ix
ry ⫽
冪ᎏ莦Aᎏ
Iy
(12-10a,b)
in which rx and ry denote the radii of gyration with respect to the x and y
axes, respectively. Since moment of inertia has units of length to the
fourth power and area has units of length to the second power, radius of
gyration has units of length.
Although the radius of gyration of an area does not have an obvious
physical meaning, we may consider it to be the distance (from the reference axis) at which the entire area could be concentrated and still have
the same moment of inertia as the original area.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
SECTION 12.4 Moments of Inertia of Plane Areas
837
Example 12-3
y
A
Determine the moments of inertia Ix and Iy for the parabolic semisegment OAB
shown in Fig. 12-12. The equation of the parabolic boundary is
y = f (x)
冢
x2
y f (x) h 1
b2
dA
h
y
2
O
y
(e)
(This same area was considered previously in Example 12-1.)
B
dx
x
冣
x
Solution
b
FIG. 12-12 Example 12-3. Moments of
inertia of a parabolic semisegment
To determine the moments of inertia by integration, we will use Eqs. (12-9a)
and (12-9b). The differential element of area dA is selected as a vertical strip of
width dx and height y, as shown in Fig. 12-12. The area of this element is
冢
冣
x2
dx
dA y dx h 1
b2
(f)
Since every point in this element is at the same distance from the y axis, the
moment of inertia of the element with respect to the y axis is x 2dA. Therefore,
the moment of inertia of the entire area with respect to the y axis is obtained as
follows:
Iy
b
2
x dA
0
冢
冣
x2
2hb3
x2h 1 2 dx =
b
15
(g)
To obtain the moment of inertia with respect to the x axis, we note that the
differential element of area dA has a moment of inertia dIx with respect to the x
axis equal to
y3
1
dIx (dx)y 3 dx
3
3
as obtained from Eq. (c). Hence, the moment of inertia of the entire area with
respect to the x axis is
b
Ix
0
y3
dx
3
b
0
冢
冣
h3
x2 3
16bh3
1 2 dx
3
b
105
(h)
These same results for Ix and Iy can be obtained by using an element in the
form of a horizontal strip of area dA x dy or by using a rectangular element of
area dA dx dy and performing a double integration. Also, note that the preceding formulas for Ix and Iy agree with those given in Case 17 of Appendix D.
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
838
CHAPTER 12 Review of Centroids and Moments of Inertia
12.5 PARALLEL-AXIS THEOREM FOR MOMENTS OF INERTIA
yc
y
dA
d2
x
y
xc
C
d1
d
x
O
In this section we will derive a very useful theorem pertaining to
moments of inertia of plane areas. Known as the parallel-axis theorem,
it gives the relationship between the moment of inertia with respect to a
centroidal axis and the moment of inertia with respect to any parallel
axis.
To derive the theorem, we consider an area of arbitrary shape with
centroid C (Fig. 12-13). We also consider two sets of coordinate axes:
(1) the xc yc axes with origin at the centroid, and (2) a set of parallel xy
axes with origin at any point O. The distances between the two sets of
parallel axes are denoted d1 and d2. Also, we identify an element of
area dA having coordinates x and y with respect to the centroidal axes.
From the definition of moment of inertia, we can write the following equation for the moment of inertia Ix with respect to the x axis:
FIG. 12-13 Derivation of parallel-axis
theorem
Ix
(y ⫹ d1)2dA
y 2dA ⫹ 2d1 y dA ⫹ d 12 dA
(a)
The first integral on the right-hand side is the moment of inertia Ixc with
respect to the xc axis. The second integral is the first moment of the area
with respect to the xc axis (this integral equals zero because the xc axis
passes through the centroid). The third integral is the area A itself.
Therefore, the preceding equation reduces to
(12-11a)
Ix Ixc ⫹ Ad 12
Proceeding in the same manner for the moment of inertia with respect to
the y axis, we obtain
(12-11b)
Iy Iyc ⫹ Ad 22
Equations (12-11a) and (12-11b) represent the parallel-axis theorem
for moments of inertia:
y
dA
h
–
2
y
x
C
h
–
2
B
The moment of inertia of an area with respect to any axis in its plane is
equal to the moment of inertia with respect to a parallel centroidal axis
plus the product of the area and the square of the distance between the two
axes.
dy
b
–
2
b
–
2
B
FIG. 12-10 Moments of inertia of a
rectangle (Repeated)
To illustrate the use of the theorem, consider again the rectangle
shown in Fig. 12-10. Knowing that the moment of inertia about the x
axis, which is through the centroid, is equal to bh3/12 (see Eq. a of
Section 12.4), we can determine the moment of inertia IBB about the
base of the rectangle by using the parallel-axis theorem:
冢冣
bh3
h
IBB Ix ⫹ Ad 2 ⫹ bh
12
2
Copyright 2004 Thomson Learning, Inc. All Rights Reserved.
May not be copied, scanned, or duplicated, in whole or in part.
2
bh3
3
839
SECTION 12.5 Parallel-Axis Theorem for Moments of Inertia
This result agrees with the moment of inertia obtained previously by
integration (Eq. c of Section 12.4).
From the parallel-axis theorem, we see that the moment of inertia
increases as the axis is moved parallel to itself farther from the centroid.
Therefore, the moment of inertia about a centroidal axis is the least
moment of inertia of an area (for a given direction of the axis).
When using the parallel-axis theorem, it is essential to remember
that one of the two parallel axes must be a centroidal axis. If it is necessary to find the moment of inertia I2 about a noncentroidal axis 2-2
(Fig. 12-14) when the moment of inertia I1 about another noncentroidal
(and parallel) axis 1-1 is known, we must apply the parallel-axis
theorem twice. First, we find the centroidal moment of inertia Ixc from
the known moment of inertia I1:
yc
xc
C
d1
1 d2
1
2
2
FIG. 12-14 Plane area with two parallel
noncentroidal axes (axes 1-1 and 2-2)
Ixc I1 Ad 12
(b)
Then we find the moment of inertia I2 from the centroidal moment of
inertia:
I2 Ixc ⫹ Ad 22 I1 ⫹ A(d 22 d 21)
(12-12)
This equation shows again that the moment of inertia increases with
increasing distance from the centroid of the area.
Example 12-4
The parabolic semisegment OAB shown in Fig. 12-15 has base b and height h.
Using the parallel-axis theorem, determine the moments of inertia Ixc and Iyc
with respect to the centroidal axes xc and yc.
yc
y
A
Solution
x
h
C
O
y
xc
B
b
FIG. 12-15 Example 12-4. Parallel-axis
theorem
x
We can use the parallel-axis theorem (rather than integration) to find the
centroidal moments of inertia because we already know the area A, the centroidal coordinates x and y, and the moments of inertia Ix and Iy with respect to
the x and y axes. These quantities were obtained earlier in Examples 12-1 and
12-3. They also are listed in Case 17 of Appendix D and are repeated here:
2bh
A
3
3b
x 8
16bh3
Ix
105
2h
y 5
2hb3
Iy
15
To obtain the moment of inertia with respect to the xc axis, we use Eq. (b)
and write the parallel-axis theorem as follows:
冢 冣 81b7h5
16bh3
2bh 2h
Ixc Ix Ay 2
105
3
5
3
2
(12-13a)
In a similar manner