J. Aci PDF

Title No.
57-43
Rectangular Concrete Stress

Distribution in Ultimate
Strength Design
By ALAN H. MATTOCK, LADISLAV B. KRIZ,
and EIVIND HOGNESTAD
An ultimate strength design theory of broad applicability is developed,
based on an equivalent rectangular stress distribution in the concrete compression zone and in general accord with the Appendix to the 1956 ACI
Building Code. The theory is characterized by simplicity without significant
loss of accuracy.
The proposed method of ultimate strength design is applied to a wide
variety of structural concrete beams and columns, subject to various combinations of bending and axial load. Calculated ultimate strengths are
compared with experimentally determined ultimate strengths for a wide
range of variables, and an excellent agreement results.
It is concluded that the proposed extension of the rectangular stress
distribution theory permits prediction with sufficient accuracy of the ultimate strength in bending and compression of a II types of structural concrete sections likely to be encountered in structural design practice, including
odd-shaped sections and other unusual cases.
Part 1 -
Review of Basic Assumptions
IN THE OcTOBER, 1955, REPORT OF ACI-ASCE Committee 327 on ultimate strength design as abstracted in the Appendix to the 1956 ACI
Building Code (ACI 318-56), ultimate strength design methods were
given in specific terms only for the cases most frequently met in design practice. Specific design methods were not given for odd-shaped
cross sections and other special cases. Secondly, extensive researches
in the United States and abroad have been completed since the Committee 327 report was prepared. A re-evaluation of the design principles involved is therefore desirable. Finally, the rectangular stress distribution theory has become widely used in design practice and in general
accord with the 1956 Code. The extent to which this simple theory can
be safely extended to unusual design cases has previously not been
thoroughly studied.
675
876
JOURNAL OF THE AMERICAN CONCRETE INSTITUTE
February I 96 I
At the request of Subcommittee 15 on ultimate strength design of

ACI Committee 318, Standard Building Code, and in the light of recent research findings, the authors have studied the rectangular stress
distribution theory as commonly used in design practice under the
1956 Code. In addition, they have extended this theory into a generally applicable and reasonably simple design tool based on a single
set of assumptions.
This paper presents such assumptions and substantiating evidence
for consideration by ACI Committee 318 and the ACI membership as
a contribution toward development of a future revision of the ACI
Building Code. In this manner, the recommendations and views expressed are those of the authors and do not necessarily reflect the
collective judgment of Committee 318.
ASSUMPTIONS IN ULTIMATE STRENGTH DESIGN
By use of the following general assumptions, ultimate strength of

sections subjected to combined bending and axial load can be predicted with adequate accuracy, even for odd-shaped cross sections and
other unusual cases:
1. At ultimate strength, a concrete stress of intensity 0.85 times the concrete cylinder strength may be assumed uniformly distributed over an
equivalent compression zone bounded by the edges of the cross section
and a straight line located parallel to the neutral axis at a distance k1c
from the region of maximum compressive strain. The distance c from the
region of maximum strain to the neutral axis is measured in a direction
perpendicular to that axis. The fraction k1 is taken as 0.85 for concrete
cylinder strengths up to 4000 psi and is reduced continuously at a rate of
0.05 for each 1000 psi of strength in excess of 4000 psi.
2. Tensile strength of the concrete may be neglected in flexural calculations.
3. Strain in the concrete at the various section levels may be assumed
directly proportional to the distance from the neutral axis. Except in
anchorage regions, strain in reinforcing bars may be assumed equal to the
tensile or compressive strain in the concrete at the same distance from the
neutral axis as the centroid of each bar or group of bars considered.
4. The maximum strain at an extreme edge of the concrete compression
zone may be assumed equal to 0.003.
5. Stress in reinforcing bars below the yield point stress for the grade
of steel used may be taken as 30,000,000 psi times the steel strain. For
strain greater than that corresponding to the yield point stress, the reinforcement stress may be considered independent of strain and equal to the
yield point stress.
USE OF SIMPLIFIED EQUATIONS
The equations for ultimate strength design given in the Appendix

to the 1956 ACI Building Code, ACI 318-56, can almost all be derived
from the assumptions set out above. Certain of the Code equations
for the design of columns contain additional simplifying assumptions,
ULTIMATE STRENGTH DESIGN
877
Alan H. Mattock (ACI member) joined the staff of the Portland Cement Association
Research and Development Laboratories, Skokie, Ill., in 1957. After graduation from
London University, he served with the Fire Research Station at Elstree, England, and
in the Lancashire County Bridgmaster's design office. He returned to study and was
awarded MS (Eng), London University, in 1949. For the next 3 years he was a district engineer in British Guiana, South America, where the principal work was the
construction of sea and river walls, and drainage and irrigation works.
From 1952 to 1957 Dr. Mattock was a lecturer at Imperial College, University of
London, having charge of the Concrete Technology Laboratory; while there he
carried out research and consulting work, and in 1955 was awarded PhD, London
University. Dr. Mattock is a member of ACI-ASCE Committee 323, Prestressed
Concrete, and also of ACI-ASCE Committee 328, Limit Design.
Ladislav B. Kriz (ACI member) graduated in civil engineering from the University
of Illinois in 1957, with previous schooling at the Technical University in Prague,
Czechoslovakia, and at the School of Civil Engineering in Madrid, Spain. His undergraduate studies were interrupted by brief periods of employment as draftsman with
Zublin-Perriere, Paris, France; as a designer with Byrnes Associates, New York,
N. Y.; and a 4-year tour of duty as surveyor in the U. S. Air Force.
He joined the staff of the Structural Section of the Portland Cement Association
Research and Development Laboratories in 1957. During his employment with the
association, he earned his MS degree from Northwestern University.
Eivind Hognestad (ACI member), manager, Structural Development Section, Portland Cement Association Research and Development Laboratories, is well known to
ACI JOURNAL readers. Prior to joining the PCA staff in 1953, Dr. Hognestad was
research associate professor of theoretical and applied mechanics at the University of
Illinois.
He obtained his MS degree in civil engineering from the University of Illinois and
the degrees of civil engineer and doctor of technical sciences from the Norwegian
Institute of Technology.
Dr. Hognestad is chairman of ACI-ASCE Committee 326, Shear and Diagonal
Tension; a member of ACI Committee 318, Standard Building Code, and ACI Committee 336, Combined Footings.
For the straight-forward cases, for which these Code equations have
been derived, the equations are entirely adequate and may be used
with confidence. It should be noted however, in using Eq. (A3) from
the Appendix to the 1956 ACI Code for the design of a beam with
compression reinforcement, that the stress in the compression steel
at ultimate strength should be checked to confirm that yielding of the
steel has in fact occurred.
In unusual problems of design, the basic assumptions given may be
used directly in everyday practice. Cases not susceptible to direct
mathematical solution may be treated by suitable iteration procedures,
In mathematically complex cases, calculations based on the basic assumptions may be used together with experimental data to develop
and verify simplified procedures for design office use, such as those
proposed in the writings of P. M. Ferguson and C. S. Whitney.
Notation
Wherever possible the notation used in the ultimate strength design appendix
to the ACI Building Code (ACI 318-56) has been used in this paper. For convenience, the notation is summarized as follows:
878
A,,
A,,
a
b
b'
c
c
D
d
d'
E,
F,,
fc'
f,
f.
net concrete cross section, or

area of segment of circle covered by equivalent concrete
stress distribution
area of tensile reinforcement
area of compressive reinforcement
steel area required to develop
compressive strength of overhanging flange in T -sections
total area of longitudinal reinforcement
depth of equivalent rectangular stress distribution
width of a rectangular section, or over-all width of
flange in T -sections
width of web in T-sections
resultant concrete compressive force
distance from extreme compressive fiber to neutral axis
at ultimate strength
diameter of a circular section
distance from extreme compressive fiber to centroid of
tensile reinforcement
distance from extreme compressive fiber to centroid of
compressive reinforcement
modulus of elasticity of reinforcing steel
eccentricity of axial load,
measured from the centroid of
tensile reinforcement, unless
otherwise specified
ratio of change in strain of
steel of prestressed beam to
change in strain of concrete at
the level of the steel
resultant steel compressive
force
resultant steel tensile force
6 x 12-in. concrete cylinder
strength
stress in tensile reinforcement
at ultimate strength
stress in compressive reinforcement at ultimate strength
yield point stress of tensile reinforcement
yield point stress of compres-
$ive reinforcement
G,
G.oc
G,,
k,.
k,
k,
Mull
m
P,.
p
p'
p,,
Pn
Pr
p1l'b
q
q'
Qwb
February 1961
center of action of concrete

compressive force
center of action of steel compressive force
center of action of steel tensile
force
c/d
ratio of average stress to maximum stress
ratio of depth to resultant of
concrete compressive force, to
depth of neutral axis
ratio of maximum stress to
6 x 12-in cylinder strength, f/
ultimate moment of resistance
plastic modular ratio, E,E,/
0.85 klf/
ultimate strength of eccentrically loaded member
ultimate strength of concentrically loaded member
tensile steel ratio, A,/bd
compressive steel ratio, A.' !bd
steel ratio at balanced ultimate strength condition in a
beam without compression reinforcement
steel ratio at balanced ultimate
strength condition in beam
with both tension and compression reinforcement
A,,/b'd
tension steel ratio for T-beams,
A,/b'd
tension steel ratio for Tbeam with balanced ultimate
strength condition
tension reinforcement index,
pf.lfc'
compression reinforcement index, p'fIff o'
tension reinforcement index
for balanced ultimate strength
conditions, Pfvlfc'
for T-beams, Pwfvlfr.'
for T-beams with balanced
ultimate strength condition,
p,..f.Ifc'
flange thickness in T- beams,
also total depth of rectangular
$ection column

(1.
t
Ec
Ecp
Eu
= inclination of neutral axis to

horizontal in nonsymmetrical
section
= strain
= concrete tensile strain at ultimate at level of steel in prestressed beam
concrete precompression strain
at level of steel in a prestressed beam
maximum concrete compression strain at ultimate beam
strength
..
En
(}
879
tensile steel strain at ultimate

beam strength
tensile steel yield strain
= compression steel strain at ultimate strength
compression steel yield strain
effective steel prestrain
strain in steel distance, an,
from neutral axis
half included angle between
two upper faces of beams with
triangular ccmpression zones
CONCRETE STRESS DISTRIBUTION

The general form of the concrete compression stress distribution at
ultimate strength in a reinforced concrete member is shown in Fig. 1.
The properties of the "stress block" are represented by the following
coefficients:
kt = ratio of average stress to maximum stress
k,
ratio of depth to resultant of compressive force, to depth to neutral
axis
ks
ratio of maximum stress to 6 x 12-in. cylinder strength, f.'
=
=
Historical background
The use of a design theory based on the ultimate strength of sections is in effect a return to the original concept of design, in that
early design formulas were empirical, being based on the failure loads
of typical elements as found by experiment.
The first published ultimate load theory was that of Koenen1 who in
1886 assumed a straight line distribution of concrete stress and a neutral axis at middepth. Since that time about 30 theories have been
published. The salient points of many of these theories were set out
in Bulletin No. 399, 8 University of Illinois Engineering Experiment
Station.
c(Resultant . Concrete)
Compressrve Force
Fig. !-Concrete stress distribution at ultimate strength
Neutral
Axis
880
February 196 1
1.0 r----.-----.-----r---.------,------,
....
g' 0.8
-...
....
eo.6
....
Q)
C/)
Q)
Mattock from beams
k2
- - - -a..... ......___
'C
---....... - .___
-- --- ------- -sc - -
---.
c:
..,
0.2
0~----~------~------~----~------~----~
1000
2000
3000
4000
5000
6000
7000
Cylinder Strength in psi

Fig. 2-Design values from tests of reinforced concrete
Many different distributions of stress in the concrete compression

zone have been suggested, but in the light of recent experimentaJ2
and analyticaP investigations, it is considered that the use of the "equivalent rectangular stress block" in design calculations will yield sufficiently accurate results and will at the same time lead to considerable
simplification of design calculations. The use of a rectangular concrete
compression stress block was first proposed by von Emperger4 in 1904,
and since that time by several other engineers, the best known in this
country being C. S. Vlhitney, whose paper5 in 1937 was a notable contribution to the literature of ultimate strength design.
Reinforced concrete investigations
Until recently, most of the available information regarding stress
distribution in concrete was derived from tests of reinforced concrete
members. In the early 1930's the extensive ACI investigation6 of concentrically loaded columns led to the addition law which states that
the ultimate strength of a column is equal to 85 percent of the cylinder
strength times the concrete area plus the yield stress of the longitudinal
steel times its area. Thus, for concentrically loaded columns, the value
of k3 = 0.85 was derived experimentally.
881
Several papers on stress distribution and ultimate strength design

were published in Europe during the 1930's, and these were followed
by the studies of ultimate strength in this country by Whitney. 5 His
analytical approach involved an equivalent rectangular stress block
with a maximum stress of 0.85 fc', thus his value of ks was 0.85.
In 1943 an extensive study of the ultimate strength of beams was
presented by Jensen. 7 He proposed a trapezoidal idealized stress distribution, and derived the properties of this trapezoid as a function
of cylinder strength by analysis of the observed ultimate strength of
reinforced concrete beams. On this basis he found the values of k1ks
and k 2 shown in Fig. 2.
A study of the ultimate strength of eccentrically loaded columns
was reported by Hognestad 8 in 1951. The stress distribution used consists of a rising parabola and a descending straight line after the
maximum stress. From the results of concentrically loaded columns, a
maximum stress equal to 0.85 fc' was chosen, that is, k 3 = 0.85. The
slope of the descending straight line was chosen so as to give the best
statistical agreement between calculated and observed column strengths
in 120 tests. The corresponding values of k 1 k 3 and k 2 are also shown
in Fig. 2.
Additional experimental evidence as to the parameters of the concrete stress block was presented in 1956.9 Using the measured values
of depth to the neutral axis at failure and of the ulti~ate moment of
resistance, and assuming a safe limiting value 0.43 for k 2 , values of
k1ks were calculated for 69 beams tested at Imperial College, University
of London, and at the University of Illinois. Based on these results the
values of k1ks labeled "Mattock" in Fig. 2 were proposed as suitable
for ultimate strength design.
~ain
concrete investigations
In recent years various tests have been carried out on plain concrete specimens, using special testing techniques, in an effort to obtain
a true picture of the stress-strain relationship for concrete during
loading to failure. Independently, but almost simultaneously, tests of
eccentrically loaded prismatic concrete specimens were carried out
in the Portland Cement Association laboratories10 and by Riisch11 at
the Munich Institute of Technology.
The specimen used in the PCA tests is shown in Fig. 3. The two
thrusts P1 and P2 were varied independently, in such a manner that
the neutral axis was maintained at the bottom face of the specimen
throughout the test. By equating the internal and external forces and
moments, it was possible to calculate the values of k 1 k 3 and k 2 directly.
In the Munich tests, groups of about five identical prisms were
tested with a different and constant eccentricity for each test. By plotting, for all specimens within one group, strain measured at an outside
882
February 196 I
c = 8-in.
--L-~~--------L-~---~-----~--------~
Neutral Surface
lb=i-;"! ~~
iI
6-in. SR-4 Gage
Fig. 3-PCA eccentric load specimen
face at ultimate strength versus applied eccentricity, the eccentricity

corresponding to a position of the neutral axis at an edge of the section
was determined. By applying statistical methods, the magnitude and position of the internal concrete force at ultimate strength was determined.
The parameters of the concrete stress block obtained at the two
laboratories are summarized and compared in Fig. 4 with the values
proposed in this paper. The PCA tests with sand-gravel aggregates have
been reported in detail elsewhere, 10 a detailed report on the tests by
J. A. Hanson with lightweight aggregates has not yet been published.
The German tests 11 were made with sand-gravel aggregates. It can be
seen that test results for sand-gravel concretes are in good agreement,
and that the radical change in aggregate type to lightweight materials
caused only a minor change in the stress distribution properties.
Analytical investigations
In a recent analytical paper3 the problem of ultimate flexural strength
of reinforced concrete members was reduced to finding the maximum
value of a load function expressed in terms of the internal resisting
forces of the loaded member. Assuming only that concrete stress, f, is
some function of strain, E, given by f = F (E), and that plane sections
remain plane during bending, the moment of resistance of a reinforced
concrete section was expressed in terms of the extreme edge concrete
strain, E0 , the reinforcement yield point, f 11 , and the dimensional properties of the cross section. This expression for moment was differentiated
with respect to the extreme edge concrete strain and equated to zero.
The resulting equation demonstrates that, when the maximum moment
in a rectangular reinforced concrete beam failing in tension under
883
symmetrical bending is reached, then the concrete stress distribution

is such that the total compressive force is equal to that obtained from
an equivalent rectangular stress block in which the stress is equal
to the actual stress in the extreme compression fibers, and which has
a depth equal to twice the distance from the extreme compressive
fibers to the center of action of the resultant concrete compressive
force. From a study of concrete stress-strain curves obtained in the
PCA eccentrically loaded prism investigation, 10 it was deduced that
the relationship between concrete stress at extreme compressive fiber,
f,, and the cylinder strength, fc', can be expressed closely as
f,
= 5f.'""
1.2
Average Stress = k 1 k 3 f~
C'
~1.0
....
~0
en
-r-------!--2-..'b.__ ~Q<l.~~
~
cu'B
c
.;: .6
Proposed Value
=>
0 .4
..
r9
-- --
-----
..,
~.21-
.0
-;.s
-
..t::.'6
Depth to Centroid= k 2 c
C'
....
en
oooo
cu .4 fc
E
':E .3-
dJ
----~----~~---"'h'i.- ~
o o o0
""A--
----~
Proposed Value
f!.
-----
-- -
=>
o- P.C.A. Sand ~ Grav~--
c .2 f-
-RUsch's Tests
- P.C.A. Lightweight
.1~--~---~~--~--~--~----~---L--~'----L---~~
2000
4000
6000
8000
10,000
Cylinder Strength in psi
Fig. 4-Properties of concrete stress distribution at ultimate strength determined

from tests on plain concrete
884
h,,
r
f-k 3 f~
r-o.a5t~1
I
a
l_
(Rectangular
Compn. Zone)
Neutral
February 1961
Axis
1
--
kiC
l
c =0.85f~ ba
ie 0.85(-~}= k, k 3
k 2 c = JL
2
(Rectangular
_ _ _ _CQ_m..!l_Zone)
Fig. Sa (left)-Actual concrete stress distribution. Fig. Sb (right)-Equivalent rectangular stress distribution
However, for purposes of practical application a conservative straightline relationship, fu = 0.85fc' was proposed.
EQUIVALENT RECTANGULAR STRESS DISTRIBUTION

Nature of approximation
The essentials of the approximation are shown in Fig. 5 and may be
summarized as follows: At ultimate strength, the compressive stress .
in the concrete compression zone is, for purposes of computation, assumed to be uniformly distributed from the region of maximum compressive strain to a depth, a, measured in a direction perpendicular to
the neutral axis, where the depth, a, is less than the depth to the neutral
axis, c. The uniformly distributed stress is taken to be equal to 0.85
of the cylinder strength, fc'. The ratio, ale is taken to be equal to 0.85
for concrete cylinder strength up to 4000 psi, and thereafter is reduced
by 0.05 for each 1000 psi of strength in excess of 4000 psi.
For a rectangular concrete compression zone the average concrete
compressive stress k1k 3 fc' is equal to 0.85 (a/c) f c', that is k 1k 3 =
0.85(a/c), also k 2 = %(a/c). Since k 3 is taken as 0.85 fc', it follows
that ale = k 1 and a = k 1 c. For the values of ale proposed above, k 1 k 3
and k2 will have values as indicated in Fig. 2 and 4. It is seen that the
proposed values for k 1 k 3 , and k 2 correspond closely with the values
determined by experiment.
Assumptions in ultimate strength design
The assumptions necessary in ultimate strength design were set out
earlier in this paper, and will now be discussed item by item.
The nature of the first assumption has already been discussed above.
The second assumption is very nearly correct; any tension zone which
does exist in the concrete at ultimate strength, in a normally proper-
885
tioned reinforced concrete section, is very small indeed; and its lever
arm is also small. It is therefore reasonable to neglect any contribution
by concrete tension to the ultimate moment of resistance of a section.
The third assumption is not strictly correct for a reinforced concrete
section after cracking, since the strain in the concrete on the tension
side of the neutral axis will vary considerably, at any given level, due
to cracking. If, however, we measure the extension per unit length of
.006.---.---,----,---,----,---.----.---.---.
'0
c:
c....
Ui
.004
c....
;:I
)(
Q)
LL.
2.002
c
o Columns
Beams
0~--~~----~--~----~--~--~~----~--~--~
2000
4000
6000
8000
Compressive Strength, psi

Fig. 6a-Uitimate strain from tests of rei~forced members
.005
c:
....c .004
en
c....
;:I
.003
)(
Q)
LL.
.0021-
Q)
Flexural
Strain=
Eu
~
~
.,
L!.o
6.
L!.o
l1l
o
t;.
-----....------.!1-o-0.....0_._----
6.
6l
0
Proposed
Design Maximum
Rusch's Tests
- PCA Lightweight
L!. -
::::>
o - PCA Sand ~ Gravel
E .0011-
2000
4000
6000
8000
Concrete Prism Strength, f~, psi
Fig. 6b-Uitimate strain from tests of plain concrete specimen'
_L
10,000
886
February 1961
a gage length including several cracks, we find that this "apparent"

tensile strain varies linearly with distance from the neutral axis. If
for a particular section, we measure the compression strains and apparent tensile strains using the same gage length, then the distribution
of these strains will be very close to linear. By experiment it has also
been found that for the normal type of deformed bar reinforcement,
the strain in the steel is nearly equal to the strain in the adjacent conTABLE I-SUMMARY OF COMPARISONS OF CALCULATED AND
EXPERIMENTAL ULTIMATE STRENGTHS
Type of member
Reinforced concrete
rectangular beams
reinforced in tension
Reinforced concrete
rectangular beams
reinforced in tension
and compression
Loading
type and
number
tested
Simple
bending,
Failure mode
59 beams
Controlled by
crushing of
concrete
Simple
bending,
33 beams
Controlled by
yield of tension
reinforcement
Simple
bending,
11 beams
Controlled by
crushing of
concrete 1tnd
yield of compression steel
Simple
bending,
Average valPe
Test ultimate strength
Calc. ultimate strength
Standard
deviation
All beams:
Beams with fc'?=:,
1.06
0.113
2000 psi:
1.02
0.083
Beams using structural or intermediate

grade steel:
1.04
Beams using high
strength alloy steel: 1.21
0.170
0.042
0.072
All beams:
1.07
Varied
All beams:
1.03
0.077
Simple
bending,
Varied
All beams:
1.05
0.094
Symmetrical reinforced concrete

T-beams reinforced
in tension
Simple
bending,
15 beams
Controlled by
yield of the
reinforcement
All beams:
1.11
0.100
Unsymmetrical reinforced concrete

T-beams reinforced
in tension
Simple
bending,
Varied
12 beams
All beams:
1.10
Beams controlled by
1.04
yield of steel:
0.047
Prestressed beams
with well bonded
tendons
32 beams
Prestressed beams
with unhanded
tendons
24 beams
0.138
Reinforced concrete
beams with a triangular compression
zone reinforced in
tension
Simple
bending,
6 beams
Con trolled by
crushing of
concrete
All beams:
1.10
0.069
Reinforced concrete
rectangular columns
Axial
loading,
16 columns
Con trolled by
crushing of
concrete and
yield of steel
All columns:
1.00
0.074
Reinforced concrete
rectangular columns
Axial loading plus

bending
about one
pr\ncipal
axis,
86 columns
Varied
All columns:
0.97
0.059
Reinforced concrete
rectangular columns
Axial loading plus

bending
about two
principal
Controlled by
crushing of
concrete and
yield of some
of the steel
All columns:
0.99
0.046
Controlled by
crushing of
concrete and
yield of some
of the steel
All columns:
1.05
0.060
axes,
10 columns
Reinforced concrete
circular section
columns
Eccentric
load,
30 columns
887
crete. The second assumption is therefore sufficiently accurate for the

purpose of design calculations.
The maximum concrete compressive strain at ultimate strength has
been measured in many tests of plain and reinforced concrete members.
In Fig. 6a test results are plotted which were obtained at the University
of Illinois 8 12 in tests of reinforced concrete columns and beams, while
in Fig. 6b are shown values of maximum concrete compressive strain
measured in the PCA~ and Munich 11 tests of plain concrete. It can be
seen that the value of 0.003 in. per in. proposed for design is a reasonably conservative value.
The implication of the fifth assumption is that the effect of strain
hardening of the steel above the yield point is neglected. Compatibility
of strains can no longer be relied on at high steel strains beyond the
yield point, also the stress-strain curve beyond the yield point is not
(and probably cannot conveniently become) stipulated in American
specifications for manufacture of reinforcement. It is therefore considered unwise in design to rely on obtaining an increase in ultimate
strength through strain hardening of the reinforcement.
SUMMARY OF COMPARISONS WITH EXPERIMENTS

Development of design equations and comparison with test data are
presented in detail in Part 2 and the appendix, respectively. A summary
is presented below.
The correctness of the equation for ultimate strength of a reinforced
concrete beam reinforced only in tension, with the strength controlled
by tension, is verified by R. C. Elstner's statistical analysis of 364 tests
of this type of beam. This analysis leads to a value of 0.593 for the
coefficient of q in the equation, as against a value of 0.59 derived from
the assumptions set out in this paper, the differences being insignificant.
The comparisons of calculated and experimental ultimate strengths
for other types of member and conditions of loading are summarized
in Table 1. The data contained in the table were obtained from tests of
334 structural concrete elements of nine different types, in which the
elements were subjected to four different combinations of flexure and
axial loading. The tests included cases in which the concrete compression zone at ultimate strength was segmental, trapezoidal, or triangular
in shape, in addition to the more common rectangular shape. It will be
seen that there is close agreement between the experimental and calculated ultimate strengths. The mean value of "Test Ultimate Strength" I
"Calculated Ultimate Strength" for all 334 cases considered is 1.037, and
the standard deviation is 0.097.
CONCLUDING REMARKS
The validity of the basic assumptions, particularly the use of an
equivalent rectangular stress distribution as proposed herein, and their
888
February 1961
applicability to the calculation of the ultimate strength of structural

concrete sections has been confirmed by two findings:
1. The parameters proposed for use in ultimate strength design, as calculated from the basic assumptions, are in close agreement with the values
of these parameters determined from tests of plain and reinforced concrete.
2. There is excellent agreement between the ultimate strengths of a wide

range of structural concrete members determined experimentally, and the
calculated ultimate strengths of the members based on the assumptions
set out in this paper.
It is concluded, therefore, that the proposed method of ultimate

strength design permits prediction with sufficient accuracy of the
ultimate strength in bending, in compression, and in combinations of
the two, of all types of structural concrete sections likely to be encountered in practice.
Port 2 -
Design Equations
DESIGN OF RECTANGULAR BEAMS IN BENDING

Beams reinforced in tension only
Ultimate strength controlled by yielding of reinforcement-In this

case, steel stress at ultimate strength, f., equals the yield stress f!l.
From equilibrium of internal forces in Fig. 7
0.85ktfa' b c
= A, fv ....
(1.1)
From equilibrium of internal and external moments

Muzt
= A, fv(d- k.c) ..
Solving Eq. (1.1) and (1.2) and substituting q

Mu"
A,
f. d ( 1 -
( 1.2)
o.:;k, q }
f 11 /fc':
........................ (1.3)
or
M.,,.
Since k1
= bd"f.'q
( 1 -
_k_._q
0.85kt
......... (1.3A)
2k2 , we may write

Mull = A, fv d(l- 0.59q)
(1.4)
or
Mull
b d 2 fc' q (1 - 0.59q) ..
........................ (1.4A)
Eq. (1.4A) is, of course, identical with Eq. (Al) in the Appendix to
ACI 318-56.
The correctness of Eq. (1.4A) was demonstrated statistically by R.
C. Elstner. Using the general concrete stress distribution parameters
889
k1,k2,ka, Eq~ (1.3A) may also be written:

Mu!t
= bd f; q
2
k~~a
( 1 -
.......... ( 1.5)
q )
Examining the results of 364 beam tests by the method of least squares
k2/k1k:1 is given by:
_ ~ Mull q'
"'I bd2f c'
+ :Sq'
(1.6)
r-----b-1
T
-+----il- -
I
l
"''----
-- -
Axis
....
As
Strain
........._
Actual
Stresses
Assumed
Stresses
Fig. 7-Conditions at ultimate strength
0.35
0.30
0.25
- u
""0
' 0.20
.0
...........
=0.15
:::3
::;e
0.10
0.05
~I
I
0.05
0.10
0.15
0.20 0.25
q = Pfy/f~
0.30 0.35 0.40 0.45
Fig. 8-Tests of 364 beams controlled by tension
890
jOURNAL OF THE AMERICAN CONCRETE INSTITUTE
Fig. 9-Variation of plastic

modular ratio m with concrete cylinder strength
60
:e
2
0
February 1961
50
a:
0
"3
-g
::;;
40
u
~
a:
30
20 ~2~0~00~~3~00~0--~40~0~0--~5~00~0--~67
oo~o~~7~00~0~~87
oo~o~
Cylinder Strength f~ {psi)
Using this equation Elstner found k 2 /k 1 k 3 = 0.593. In Fig. 8, the curve

of Eq. (1.4) is drawn, together with the 364 test results considered in
Elstner's investigation.
Ultimate strength controlled by compression-In this case crushing

of the concrete takes place while the steel stress is below the yield
point.
From equilibrium of internal forces:
0.85 k,fc' b c = A, f, ...
..
( l.lA)
From equilibrium of internal and external moments:

M,. .. = A,f,(d- k.c).
(1.2A)
From assumption of linear strain distribution:

k
u-
d-
Eu
E,
Solving Eq. (l.lA) and (1.7) for k,:

k,.
pm
+ ( p~
.... ( 1.7)
Eu.
r-
p~
. ................ (1.8)
where m = (E. Eu) I (0.85 k 1 f/) as shown in Fig. 9.

The ultimate moment of resistance of a section may be calculated
by substituting the value of k, obtained from Eq. (1.8) in the following equation:
Mull
= (0.85 k,fc') b d
ku (1 - k2 ku) .
..................... ( 1.9)
Balanced conditions-In this case simultaneous yielding of steel and

crushing of concrete takes place.
The balanced steel ratio, pb, is obtained by solving Eq. (1.1) and (1.7):
Po
= 0.85kl
Ey
Eu
+
fc'..
-Eu
(1.10)
891
Fig. I0-Steel ratio qb for

balanced ultimate strength
0.51---------~
fy(psi)
40,000
0.3
0.2
0.1
1000
2000
3000
4000
Cylinder
5000
6000
7000
Strength f~ (psi)
We may also write:

q.
or, substituting for
11
= P
and
h._ = 0.85 k1
fc'
Eu
Ev
........................... ....
{1.11}
Eu
E11
q. = 0.85 k1 (
f v ~~~~OOO-) ............................. ( l.llA)
The ultimate moment for balanced failure may then be written:

Mull
= (0.85 klf,')b d
2 (
Ey
Eu
+
)[
Eu
1 -
k2 (
Ev
E,.
+
)] . . (1.12)
Eu
Alternatively the value of qb found using Eq. (l.llA) may be substituted in Eq. (1.4).
Fig. 10 shows the balanced steel ratios qb for various strengths of
concrete and steel. Also plotted in Fig. 10 is the limiting value of q to
be used in design of sections prescribed by ACI 318-56. It is seen that
the limiting value prescribed is safe for steel yield points not exceeding 60,000 psi as assumed in the Code, but the limit is in fact greater
than qb for high steel stresses. If it is considered desirable for design
purposes to establish a limiting value of q less than qb even for high
steel stresses, then it is proposed that this limiting value be expressed
as a fraction of qb and not in the form in current use. If, however, a
simple direct expression for the limiting value of q is desired, then
the following is proposed:
"For fc' < 4000 psi, qlim = 80/y j 11 ; for concrete strengths greater
than 4000 psi reduce qlim by 0.02 for each 1000 psi." This simplified
expression ensures that q 11m will be between 70 percent and 80 percent
of qb for an extremely wide range of concrete and steel strengths.
In Table A-1 of the appendix a comparison is made between the ultimate moment calculated using Eq. (1.9) and the ultimate moment meas-
892
February 1961
ured in 59 tests on singly reinforced beams failing in compression. The

average value of Mtest!Mcazc for all beams is 1.06 and the standard
deviation is 0.113. It is of interest from a practical design point of
view to note that the average value of Mtest!Mcalc for beams made
from concrete with a cylinder strength of 2000 psi or over is 1.02, and
that the standard deviation for this group of beams is 0.083. A histogram of these results is plotted in Fig. 11.
Beams reinforced in tension and compression
Ultimate strength controlled by yielding of tension reinforcementIn this case tensile steel stress at ultimate strength, f .. is equal to the
yield stress, f y
From equilibrium of internal forces in Fig. 12:
0.85 k1 fc' b c
A: f.'
= A, fy
...
(2.1)

M, ..
= 0.85 k,fa' b c(d- k2c) + A.'f.'(d- d').
(2.2)
In most normal doubly reinforced beams the compression steel is close

to the face of maximum concrete compression. The strain in the steel
Average
= 1.01
(/)
Q)
CD
0
0.8
0.9
1.0
1.1
MTEST
McALc
1.2
1.3
Fig. I I -Histogram of
(Mtest! Mcalc) for compression failures of singly reinforced concrete beams with
concrete strength greater
than 2000 psi
\
'
893
_L ----Neutral
Axis
"C
As
0000
..J I-
Asfs
Strain
Actual
Stresses
Assumed
Stresses
Fig. 12-Conditions at ultimate load, beam reinforced in tension and compression
will then be greater than the yield point strain Ey', and the steel stress
fs' will equal fy' In this case we may rewrite Eq. (2.1) and (2.2) as
follows:
0.85 k, fc' b c
A: f.' = A,
Mull= 0.85k,fo'bc(d-k.c)
f, .......................... (2.1A)
A.'f.'(d-d') ....... ...... (2.2A)
From Eq. (2.1A)

k.
P - f.'P')
( f. 0.85
k, fc'
....
...... (2.3)
or
- q' )
( q0.85
k,
..
............ (2.3A)
where
q'
= p' f.'fc'
f.'
Calculate c from Eq. (2.3) and check that

Ea
'
==
Eu
c - d'
c
L' using
(2.4)
If E.' :::::... Ey', the compression reinforcement has yielded and the ultimate moment may be calculated by solving Eq. (2.1A) and (2.2A),
this yields:
M., = (A, f.- A: fu') d
Since kt
M,.,
= (A. fv
1 -
O.~~k,
(f. P fc' f.' p' ) ~
A.' fu' {d- d'}
2k 2 , we may write:
-
A.' fu') d
1 -
Of~'g
(f, p -
fv'
p')
~ +
A .'f.' {d- d'}
(2.5)
I
894
lOURNAL OF THE AMERICAN CONCRETE INSTITUTE
/February 1961
This may also be written:

M.11 = (A, fy - A.' fy')d {1- 0.59 (q- q')}
If f v =
A.' f.' {d- d'} .... (2.5A)
A.' f. {d- d'} ..... (~.5B)
f u' this becomes
M,.., = (A,- A.')f. d
~ 1- 0.59
::. (p- p')
""-.
This is Eq. A3 of the Appendix to ACI 318-56, and is valid only if th.'e
compression steel has yielded. When using this equation a check should
be made on the compression steel stress using Eq. (2.3) and (2.4).
If Es' < Ey', then compatibility of strains in the compression zone
must be considered.
f.'
E.'
E,
Eu
Solving Eq. (2.1) and (2.6) for k
E, (
C -;;
11 :
)r
= ~ = V[';( p- :: p
+ pm ~
where, as before, m = (Es Eu) I (0.85 kt fc').
k.
d' )- ............................... (2.6)
';( p'-
:: p ) ... (2.7)
The compression steel stress, f.', may then be obtained from Eq.
(2.6) by substituting the value of c, calculated using Eq. (2.7). The
moment of resistance is obtained by substituting the calculated values
of c and fs' in Eq. (2.2).
Ultimate strength controlled by crushing of concrete-The tensile
reinforcement is in this case stressed elastically. From equilibrium of
internal forces:
0.85 k, fa' b c
= A, f,
A.' f.'
.................... (2.8)

Mull= 0.85k,fa'bc (d-k,c)
A.'f.' (d-d') .....
... (2.9)
If the compression reinforcement has yielded these equations become:

0.85 k, fa' b c
A.' fy' = A, f, .
. (2.8A)
and
Mull
= 0.85 k, fa' b
(d- k.c)
A.' fy' (d- d') ...
.. (2.9A)

k
_
u-
Eu
E,
Eu.
.. (2.10)
or
.
==
Eu
c)
( -d
-C-
... (2.10A)
~
.
895
Solving Eq. (2.10) and (2.8A) for ku.
-k.
= dc = 'V..t/)m(
12 P +
where m is
&::
e:
e'
p'
)t~
I +
pm
~(P
+ ~p')
2
Eu
(2.11)
before.
Check that Es' is > Ey' using E.' = Eu [ (c - d') /c].

The moment of resistance is calculated using Eq. (2.9A), substituting
the value of c, obtained from Eq. (2.11). If the compression reinforcement has not yielded, then the compatibility of strains across the entire
section must be considered. This yields:
==
f.s
Eu
c)
d
(-C-
(2.10A)
and
....... (2.6)
Solving Eq. (2.10A), (2.6), and (2.8A) for k,

k, =
1J1 1~
(p'
~~+
p)
m ( p'
~ +
p )
(p'
p)
(2.12)
The value of c obtained from Eq. (2.12) is substituted in Eq. (2.6) to

give Es' and hence fs'. These values of c and fs' are substituted in Eq.
(2.9) to yield the ultimate moment.
Balanced conditions-In this case simultaneous yielding of tensile

steel and crushing of concrete takes place. From assumption of linear
strain distribution:
C
= (
Ey
Eu
+
(2.13)
Eu
Solving Eq. (2.13) and (2.8):

Balanced steel ratio
Ps
= p' ~~ + P
...... (2.14)
where Pb is the balanced steel ratio for a beam having the same steel
and concrete strengths, but reinforced only in tension, as given by Eq.
(1.10).
If d' /d is
<
(Eu -
Ey') / (Eu
+ E11),
then E/
Pn = P' 1i_
t.
>
E/ and
fs'
(2.15)
896
If d' /d is
>
ey') I (Eu
(Eu PB
= p'
+ Ey),
~: ~
Eu -
then fs'
.~
(Ey
<
February 1961
fy'
+ Eu) } +
P ...................
(2.16)
The ultimate moment at balanced failure is given by:

M.,,=0.85k,fc'bd2 (
))l-k2(
Eu
Ev
Eu
)t+A,'f,'(d-d')(2.17)
Eu
Ev
Eu
where
f,'
= fv'
d'
d
if
and
f,'
= E, 1
Eu -
(Ev
Eu)
if
~ > ( ::
+:: )
The results of applying the above analysis to 44 doubly reinforced

beams is shown in detail in Table A-2 and is summarized below. (Histograms of these results are plotted in Fig. 13.)
Failure
mode
Reinforcement
Average M,., (test)

M,., (calc)
Standard
deviation
Tension
Structural or intermediate grade Steel
1.03
0.04
Tension
High strength alloy

steel
1.21
0.17
Structural or intermediate grade Steel
1.03
0.07
Com pression
It is of interest to note that the compression reinforcement yielded

in only eight of the 33 beams failing in tension. In four of the beams
the "compression" reinforcement was actually in tension at ultimate
load.
The reason for the relatively high value of Mtest!Mca.lc for beams failing in tension and reinforced with high strength alloy bars is probably
strain hardening of the tension steel in beams with only a small percentage of tension reinforcement. The yield plateau of this reinforcement is of the order of 0.2 percent whereas that of structural grade
reinforcement is about 1.5 percent. The alloy steel therefore quickly
passes into the strain hardening range with consequently higher steel
stresses at failure. Since neither the length of the yield plateau, nor
the shape of the steel stress-strain curve beyond the yield plateau are
included in steel specifications, it is the opinion of the writers that for
practical design purposes the calculation of the ultimate moment of
beams reinforced with high strength steel of this type should be based
on the specified yield stress. However, if it is desired to analyze the
897
80
70
;e
Average = 1.04
60
Average = 1.03
..... 50
Ill
E
~ 40
m
....0
d
30
20
10
0
0.9
1.1
1.0
1.2
0.9
1.0
1.1
1.2
MTEST
MTEST
McALC
McALC
Tension Failures
(Structural ~ Intermediate Steels}
Compression Failures
Fig. 13 -
Histograms of (Mtost!Mcazc) for reinforced concrete beams with

compression reinforcement
behavior of a specific beam, for which the stress-strain curve of the

reinforcement is known, the following approach can be used.
+ A.' f,' = A, f, ................................. (2.18)
0.85 k, fo' b c
From compatibility of strains:

E.
, = (c-d')
Eu
and
Solving the above equations:

f, = [0.85:,fo'
(e,
Eu) +
E,eu ~
1-
~(E ~
Eu )}J-.... (2.19)
Plot the curve of fs and E8 given by Eq. (2.19) on the same base as
the stress-strain curve for the reinforcement. The intersection of the
curves gives the steel stress and strain at ultimate moment, from
which the ultimate moment can be calculated. This approach was used
to analyze Beam IIIB-1 from Table A-2, using a typical stress-strain
curve for this type of reinforcement, and a value of Mtost!Mcazc of 1.23
was obtained, as against 1.44 if the calculated ultimate moment is based
on the yield strength of the reinforcement. The remaining hyper-
898
February 1961
-j0.85f~
1--.=..L
T
u
..ll:
u
~
t
...Lt-----J
.....
1-------'~
- -
- -
-1--~
As
Concrete
Strain
Actual
Assumed
Stresses at Ultimate
Fig. 14--Stress and strain conditions in prestressed beams
strength of 23 percent could be due to several as yet unpredictable

causes such as variation in length of yield plateau from bar to bar,
departure from linear strain relationship at a wide crack, etc.
Prestressed beams
The following reasoning applies to prestressed rectangular beams

and to T- or !-section beams in which the flange thickness is greater
than the depth of the equivalent rectangular stress block at failure.
From equilibrium of internal forces in Fig. 14:
= A,f............................................ (3.1)
0.85ktfc'bc

Mull
= A, f,
(d- k.c) .......................................... (3.2)
or
Mull
= 0.85 ktf.' b C (d- k.c) ................................ (3.2A)
From assumption of linear distribution of concrete strains:

ku =
e. ~ e,. ........................................... (3.3)
Solving Eq. (3.1) and (3.3) we obtain:

(1) Change in concrete strain at level of steel from zero load to failure
(e.
+ e,.)
l!op
+ e. (
! ~:f.'
0 8
- 1 } (3.4)
Because of high bond stresses and consequent bond slip at failure in this
type of beam, the change in strain in the steel during loading to failure
may not be equal to the change in strain in the adjacent concrete. We
will write therefore:
899
(2) Change in strain in steel during loading to failure:
= E, -
e..
=F 1 +
Ecp
Ev( 0 8 ~~:fo'-
)~
............... (3.5)
where F is the ratio of the actual change in strain of the steel, to the
change in strain of the concrete at the level of the steel calculated on
the basis of linear distribution of concrete strain.
It has been suggested23 that for well bonded tendons F = 1.0, while
for post-tensioned tendons without bond a suitable value deduced from
the experimental results 10 ~ 4 25 20 plotted in Fig. 15 is F = 0.8k,. Therefore, for prestressed beams with well bonded tendons:
E.
5~ Eu ( 0.85p k,
f. fc' - 1 )
Ecp
..... (3.6)
Eu ).
while for prestessed beams with unbonded tendons:

E,
= {O.~:~!fc'
Eu)
(Ecp -
+ 0.8
Eu
+ e,. ~. . .
......... (3.7)
Since the shapes of the stress-strain curves of prestressing steel do

not in general lend themselves to algebraic representation, the analysis
of a given section is best carried out using a process of iteration. Assuming a value of f., the steel stress at failure, the corresponding strain
E8 may be calculated using Eq. (3.6) or (3.7). The steel strain so obtained is entered on the stress-strain diagram for the reinforcing steel
and a new value of f. obtained. The process is repeated, using the new
value of f., as often as is necessary.
0.00160
x6
0.00120
FEu
X
t;:,.
oo
Janney (23)
Feldman (24)
Allen (25)
Mattock (26)
0.1
0.2
0.3
0.4
0.5
0
6
+
0.6
ku
Fig. IS-Post-tensioned prestressed beams without bond F Eu v ku
0.7
900
February 1961
When the steel stress has been determined, the neutral axis depth
may be calculated using Eq. (3.1) and the ultimate moment using Eq.
(3.2)
Value of Ecv: For a safe estimate of E8 , the minimum probable value
of Ecv should be used. It can be shown 26 that with present day allowable
concrete stresses a reasonable value of Ecv is 0.0004.
The analysis proposed above was applied to 32 prestressed beams
with well bonded tendons and to 24 post-tensioned beams without bond;
the results are detailed in Tables A-4 and A-5 and are summarized
below.
0
1. Beams with well bonded tendons:

Average M,.,,;M .,,. = 1.07; Standard deviation
0.077
2. Post-tensioned beams without bond:
Average M,.,,;M.,.,. = 1.05; Standard deviation
0.094
Histograms of these results are plotted in Fig. 16.
Note- Expressions for the strength of prestressed beams with non-rectangular compression zones may be derived by combining the approach
used above to take into account the influence of prestress, with the approach described below for reinforced concrete beams with nonrectangular
compression zones.
NONRECTANCULAR BEAMS
Symmetrical T-beams
Ultimate strength controlled by yielding of reinforcement- When

the flange thickness exceeds the depth of the equivalent rectangular
stress block, the equations developed for the singly reinforced rectangu70
Average = 1.07
I
-60
.,.,,,, 1.051
::,!!
a..
en
50
40
<II
-...
CD
0
Q>
30
.Q
::1
z 20
10
0
0.9
1.0
1.1
1.2
1.3
1.4
MTEST
MCALC
Beams With Banded Tendons
Fig. 16 -
0.9
0.8
0.9
1.0
1.1
1.2
MTEST
MCALC
Post-Tensioned Beams Without Bond
Histograms of (Mtest!Mcalc) for prestressed beams with and without

bond
_j_
J..
901
b +----t
Asfs
(a) Section
(b) Strains
(c) Stresses
Asfs
(d) Assumed
Stresses
Fig. 17-Conditions at ultimate strength in reinforced concrete T-beam
lar beams are valid for the T-beam. If the depth of the equivalent
rectangular stress block is greater than the flange thickness, i.e., if
k1c
= (A,f.)/(0.85fc'b)
>
t (Fig. 17),
then:

0.85 k1 fc' b' C
0.85 fc' (b- b') t = A, fv .
( 4.1)

M ... = 0.85 k1 fc' b' c ( d -
~1 c )
0.85 fc' (b- b') t ( d -
+) .... (4.2)
SolvingEq. (4.1) and (4.2):

M ., =(A,- A,,) fv d
11 - 0.59
b'~vfa'
(A,- A,,)}
A,rf, ( d -
+) (4.3)
In which A 81 = {0.85 fc' (b - b')t}/fy is the steel area necessary to

develop the compressive strength of the overhanging portions of the
flange. Eq. ( 4.3) may also be written as follows:
M ...
= (A,- A,,) fv d
11 - 0.59 (Pw- Pr) J:. } + A,,
fv ( d -
+) ... (4.4)
where: Pw = A./b'd and p1 = A 8 tfb'd. This is Eq. (A4) of ACI 318-56.

Ultimate strength controlled by crushing of concrete-While steel
stress is still below the yield point:
From equilibruim of internal forces:
0.85 k1 f.' b' c
0.85 fc' (b- b') t = A, f,
.......... (4.1A)

M ... = 0.85 k 1 fa' b' c ( d -
~1 c )
0.85 fc' (b- b') t ( d -
+)
(4.2)
902
February 1961
= dc =
ku
e,.
........................................... (4.5)
Eu
Solving Eq. ( 4.1A) and ( 4.5) for k,.:

ku
= ~ =
11
p,. m
+[ ~(
b' b' ) 2 ~. d ~ +
p,.2m
- [{e"bb')2~.d~ + p ..2mJ. ............................... (4.6)

where, as before, m = (E. e,) I (0.85kdc').
The ultimate moment of resistance of the section may be calculated by substituting the value of c obtained from Eq. (-!.6) in Eq. (4.2)
above.
Balanced conditions-i.e., simultaneous yielding of steel and crushing of concrete.
=(
Eu
E } ............ ( 4 .7 )

0.85k, ll._(
( A, ;A,,)=
d
f.
Eu
= Pb
.................... (4.8)
Eu
Where Pb is the balanced steel ratio for a rectangular beam made of

same concrete and steel.
The ultimate moment for balanced conditions may be written:
Mull
= 0.85 k, fc' b'

+
Eu
Eu
d2
1 -
0.85fc' (b- b') t ( d -
~(
2
Eu
+} . . . . . . .
Eu
Ev
(4.9)
The results of tests on 15 T-beams were analyzed and are detailed

in Table A-3. The mean value of Mtest!Mcalc is 1.11 and the standard
deviation 0.10. The results include some beams in which strain hardening of the reinforcement is known to have occurred. If desired the
strain hardening effect could be investigated in a particular case using
the approach outlined for doubly reinforced beams.
Nonsymmetrical T-beams
For most nonsymmetrical T-beams used in monolithically cast construction, lateral deflection is prevented. The neutral axis is then
horizontal, and the strength of an unsymmetrical T-beam equals that
of an equivalent symmetrical beam with the same total flange width.
When lateral restraint is absent, however, the beam may be considered to be loaded vertically and to be free to deflect laterally as shown
in Fig. 18. Since no moments are applied in a horizontal plane, the
903
center of concrete compression must be vertically above the centroid

of the tension reinforcement. The width of the stress block at the top
face of the beam will therefore be 1:5 times the width of the web of
the beam for this condition to be satisfied, assuming the reinforcement
is placed symmetrically within the web.
Ultimate strength controlled by yield of reinfoTcement-From equilibrium of internal forces:
lh (1.5 b' a 0.85 f.')
= A, f ....... ........................... (5.1)

M,..,
= A, f.
( d - : ) ...................................... (5.2)
or
= 0.75 b' a 0.85 f.'
( d -
; ) ............................ (5.2A)
Solving Eq. (5.1) and (5.2) we have:

M .. .,
A, f. d (1- 0.52 q .. ) ................................... (5.3)
or
Muu
= b' d" f.'
q,. (1- 0.52 q,.) ............................. (5.3A)
where
_ ( A,
b'd
q .. -
vf.)
rL..---....,
,
l __ _
"l . l
(a) Section
(b) Strains
(c) Actual
Stresses
(d) Assumed
Stresses
Fig. IS-Conditions at ultimate strength in a reinforced concrete unsymmetrical

T-beam
904
February 1961
Ultimate strength controlled by crushing of concrete-While the

steel stress is below the yield point.
From the geometry of the section, as shown in Fig. 18:
= 23k,. b' sin a ............................................... (5.4)
where a is the inclination of the neutral axis to the horizontal.

From the assumption of linear strain distribution:
~ =
(dcosa.
Ev
lhb'sina-c) .............................. (5.5)

C

.! (%b') tan a. 0.85 fc'
= A. E. e, ................................ (5.6)

Muu
= .! (%b') tan a. 0.85 fc' (d -
.! b' tan a) ..................... (5. 7)
Solving Eq. (5.4), (5.5), and (5.6) for tan a we have:
tan a
[v( ~r(
1 -
+ ( w.)
9 (b')"
- (1 -
~ )] ........... (5.8)
4k,mA.
where, as before, m = (Eu Es) I (0.85kdc').

The ultimate moment is calculated by substituting the value of tan a
from Eq. (5.8) in Eq. (5.7).
Balanced condition-Simultaneous crushing of concrete and yield of
steel.
.! (% b') tan a. 0.85 f.' = A. f, = A. E, e...................... (5.9)

~=
(d cos a
Eu
.! b' sin a)
b' sin a )
( 2 3k,. b' sin a.)
...
~
Ev
= 2 ~ ( ~. cot a+ .!) - 1 ........................... (5.10)
Solving Eq. (5.9) and (5.10) for the steel ratio

conditions:
Pwb
~: [
3 ( !-1:1
Pwb
for balanced ultimate
:1)- kt l ........ (5.11)
905
or
q,.b =
-;:--1.: ~') _k, ................................. (5.11A)
The ultimate moment of resistance is given by substituting

Eq. (5.3)
M,..,
qwb
in
= A, fu d (1- 0.52 q,.&) ................................. (5.12)
The results of tests on 12 unsymmetric T-beams are summarized

in Table A-6. The mean value of Mt t!Mcalo for all beams is 1.10, and
the standard deviation is 0.138. The mean value of Mtet!Moalo for beams
with ultimate strength controlled by tension is 1.04 and the standard
deviation is 0.047.
Beams with triangular compression zones
The following analysis was developed in connection with tests of
this type of beam, to check the applicability of the rectangular stress
block approximation to members having triangular shaped compression zones.
Ultimate strength controlled by yield of reinforcement-From equilibrium of internal forces (see Fig. 19)
0.85 f.' a tan()= A, fu .......................................... (6.1)

Muu
= A, fu
( d- 23a ) (6.2)
Solving Eq. (6.1) and (6.2) we have:

Muu
= A, f11 d (1- 0.723 '/q) .................................. (6.3)
or
f5 A1
(a) Section
(b) Strains
(c) Actual
Stresses
fsAs
(d) Assumed
Stresses
Fig. 19-Condition at ultimate strength in a reinforced concrete beam with a

triangular shaped compression zone
906
M .. .,
= d" tan
February 1961
f.' q (1- 0.723 -yq) ........................ (6.3A)
where
_ (
q_
fv)
A.
d"tanfJT.'
Ultimate strength controlled by crushing of the concrete-While the

steel stress is below the yield point.
0.85 f.' a tan 8
= A. f, ........................................ (6.4)

M....
= 0.85
f.' a tan
8 (
d -
2a) ............................. (6.5)

3
:: = (
d-; c
) ................................... (6.6)
From Eq. (6.4) and (6.6):

c (0.85~'kt" tan
E, Eu
8) + c _ d = 0 ............................ (6.7)
Solve Eq. (6.7) for c and substitute in Eq. (6.5A) below to compute
ultimate moment
Murt
= 0.85 f.' k," c" tan 8
d -
2~
C }
(6.5A)
Alternatively Eq. (6.4) and (6.6) can be solved for c using an iterative
procedure. The value of c obtained is then, as before, substituted in
Eq. (6.5A).
Balanced conditions-Simultaneous crushing of concrete and yielding
of steel.
0.85 f.' (k,c) tan 8
= A. fv .................................... (6.8)
= (e. ~ e. ) ............................................ (6.9)
Solving Eq. (6.8) and (6.9) for the steel ratio pb for balanced ultimate
conditions [p = A,/ (d2 tan 8) ].
p.
= 0.8:.f.' (
Evk';"l!u
y. . . . . . . . . . . . . . . . . .
(6.10)
907
or
......... (6.10A)
The ultimate moment of resistance is given by substituting qo in Eq.

(6.3)
Mu.,
= A, f, d (1- 0.723
-yq;;} ................... .
. ...... (6.11)
The results of tests on six beams with triangular compression zone

are summarized in Table A-7. The mean value of Mtest!Mcazc is 1.10 and
the standard deviation is 0.069.
DESIGN OF COLUMNS
Concentrically loaded columns
In this case it is assumed that at ultimate strength of a column the

entire concrete cross-section is uniformly stressed to 85 percent of the
cylinder strength of the concrete, and that the entire cross section of
the column is uniformly strained to 0.003 in. per in.
For columns reinforced with steels having a yield point of 90,000 psi
or less, ultimate strength will be controlled by simultaneous crushing
of the concrete and yielding of the steel. The ultimate strength is
therefore given by:
P.
= 0.85 f/ A. + fv A,, .................................... (7.1)
where A, is the net concrete cross section

A,, is the total longitudinal reinforcement cross section
In the above it is assumed that sufficient lateral ties are provided to

ensure that the reinforcement will not buckle before reaching its yield
stress.
Eq. (7.1) was used to calculate the ultimate strength of 16 concentrically loaded reinforced concrete columns with lateral ties, tested as
part of the ACI column investigation and reported in University of
Illinois, Engineering Experiment Station Bulletin No. 267. 6 The results
are set out in detail in Table A-8. The average value of Ptest!Pca.zc for all
beams is 1.00, and the standard deviation is 0.074.
It was concluded in the ACI column investigation 6 that Eq. (7.1)
also serves to calculate the yield point load of a concentrically loaded
circular column with spiral reinforcement. Further increase in load
is obtainable in this type of column when loaded concentrically due
to the lateral restraint of the compressed concrete by the spiral. However, Hognestad 8 has shown that if the load is eccentric to even a small
degree, then no further increase in load is obtained after yield in this
type of column. Since a truly concentrically loaded column is an extreme
rarity in practice, it is proposed that increase in load after yield of
908
February 1961
p
,0.85f~1
_.:::::.L!.!:I4!--
___::r
As .
tu
j_~
Asfs
Actual
Stresses
Strain
Asfs
Assumed
Stresses
Figure 20-Conditions in an eccentrically loaded column at failure, c
<d
spirally reinforced columns should be ignored, and that their ultimate

strength should be calculated using Eq. (7.1).
Eccentrically loaded rectangular column, reinforced on two faces
It is reasonable to assume that the compression reinforcement has

yielded. This may be verified by checking that Eu = [ (c- d') /c] is::::,.. Ey'.
If this is not the case, a solution may be developed by taking into account compatibility of strains, as was done for the case of a beam reinforced in compression in which the compression reinforcement did
not yield.
In what follows it is assumed that the above check has been made,
and that the compression reinforcement has been found to yield.
From equilibrium of internal and external forces in Fig. 20:
P,.
= 0.85klfo'bc +
A.'f.'- A,f,.
.. .... (8.1)

P.e
= 0.85 k1 fc' b c (d- k.c) + A,' f.'
(d- d')
..... (8.2)
Strength governed by tension steel yielding-Solving Eq. (8.1) and

(8.2) for c/d, and substituting
q =
pf.
fo'
f ~T -
~2-
2.35
and
q' =
~ q' ( e ~ d'
p'f/
fc'
1 ) -
q (
~)~
J. . .
(8.3)
909
Substituting this value of c/d in Eq. (8.1) yields:

Pu = fc' b d [ (q'- q)
v( - r~
0.85
2.35 q' ( e
~__<!_ -
For symmetrically reinforced columns, p

q = q', therefore:
P,. = 0.85fo'bd [
-(+- 1)
v\ ~-
- (
1 ) -
p'.
1)
+} l
Assuming
fv
(8.4)
= fv',
2.35q( ddd' )].(8.5)
In the above equations the displacement of concrete by the compression reinforcement is ignored. For values of p' > 2.0 percent the
ultimate load may be over-estimated by up to 6 percent. 8 This error
may be corrected by considering an effective yield point [fy' - 0.85 fc']
instead of fv' in the calculation q'.
35
30
25
~ 15
.1:1
:I
10
Fig. 21 -Histogram of
(Ptest!Pcazc) for eccentrically loaded rectangular columns
1.2
910
February 196 I
Strength governed by concrete crushing-From assumption of linear

distribution of strains:
dC
Eu
Eu
......................... (8.6)
)
E,
...
and,
f,
= E,
E,
= E,
d-;;
Eu (
)...
. ......................
{8.6A)
If k1 c is less than t, Eq. (8.1) and (8.2) apply.

If k 1 c is greater than t, these equations become:
P.
= 0.85 fo' b t +
= 0.85 fc' b t
P.e
( d -
A.' f.'
A. f, ......
...... (8.1A)
+) + A: f.' (d- d') ............. (8.2A)
Once again, if greater precision is desired in computation of the

ultimate load, an effective yield point [fv' - 0.85 fc'] should be used
instead of f y'.
Solving Eq. (8.1), (8.2), and (8.6A) yields a cubic equation in c/d
which does not lend itself to easy solution.
The following method of solution by successive approximations was
proposed by Hognestad. 8
With known section dimensions, material constants, and eccentricity e,
the problem is to find c, and hence Pu. The procedure is as follows.
Assume a value for c and use Eq. (8.2) to calculate Pu. Substituting
the assumed value for c in Eq. (8.6A) calculate f . Using the calculated
values of Pu and fs in Eq. (8.1) calculate c. The process is repeated
until the assumed and final value of "c" coincide or are sufficiently
close.
Balanced conditions - Simultaneous yielding of tension steel and
crushing of concrete.
In this case Es = Eu
.
~~)
(Eu
"{[
.................... .
........... (8.7)

P,.
= 0.85 k, fc' b d (
Eu
Eu
+
A.' f.' -
A, fv ..
....... (8.8)
Ev

P,.e
= 0.85
k1
fo' b d 2
Eu
E..
)
Ev
~ 1 - k2 (
Eu
f.,.
Ev
) (
\
A.' fv (d- d') .. (8.9)
The analysis proposed above was applied to 84 eccentrically loaded,

rectangular reinforced concrete columns tested by Hognestad. 8 The results are set out in detail in Table A-9. The average value of Pt.,t!Pcazc
for all columns is 0.97, and the standard deviation is 0.059. A histogram
911
~=I==Fsc tu' u
=IAsf.!.c
=I:Asfst
=I:Asfst
Assumed
Stresses
Actual
Stresses
Strains
Fst
Fst
rr
Fig. 22-Stress and strain conditions in circular column subject to eccentric loading
of these results is plotted in Fig. 21. The calculation of the ultimate

strength of this large number of columns was greatly facilitated by
use of an electronic computer.
Circular columns subject to eccentric load
Case 1-Equivalent rectangular stress distribution covers part only

of column cross section as shown in Fig. 22.
Let G., be center of action of steel compressive force.
Let G,, be center of action of steel tensile force.
Let. G. be center of action of concrete compressive force.
From condition of equilibrium of internal and external forces:

Pu
= 0.85f.' A.+
F -
F,, ...
. ... (9.1)
where
A.
area of segment of circle covered by equivalent rectangular stress
distribution.
Foe
= resultant steel compressive force
F,,
= resultant steel tensile force
(~A.f")
(~A.f,,)
From condition of equilibrium of internal and external moments:

P ..e
= 0.85 f.' A. y. +
F y,.
+ F,, y,, .......................... (9.2)
Where the eccentricity e is measured from the center of the column.

For the general case an iterative solution of the Eq. (9.1) and (9.2)
is proposed.
Assume a position of the neutral axis cc and calculate the stress in
each bar of the reinforcement using:
f,ft
= E.
Eon
= E,
E .. (
c;; ) ; which must
be
< f.
Calculate the resultant tensile and compressive steel forces, F. 1 and F.c,
and their centers of action, G. 1 (y,,); Gc(y,.).
912
February 1961
The first moment of area about center 0 of the segmental area covered
by the rectangular stress distribution, Acyc, is given by:
A. y.
(xx)
=- =%
12
8
{ D (k1 c) -
(k1 c) } 812
.........
. .....
(9.3)
Substitute the values calculated above in Eq. (9.2) and obtain P 11

Substitute this value of Pu together with the calculated values of F.c
and F.t in Eq. (9.1) to obtain Ac, and hence a value for k 1 c and c.
The calculation should be repeated as often as is necessary, until assumed and final values of c are equal or sufficiently close. The value
of Pu corresponding to this value of c is the ultimate load capacity of
the column.
50 '
Average
105
40 1-
~
~
1/'l
E 30 ::3
(.)
Q) 20 -
.&J
E
::3
10 I
0
0.9
1.0
1.1
pTEST
PeALe
1.2
Fig. 23 -
Histogram of
(Ptest! Pcalc) for 30 eccentrically loaded circular columns
913
10.85f~,
Actual
Stresses
Strains
Assumed
Stresses
Fig. 24-Conditions at ultimate load, eccentrically loaded rectangular column

with reinforcement on all four faces
Case 2-Equivalent rectangular stress distribution covers the entire

column cross section.
Pu
= 0.85 fc'
( ~
D) + F ................................... (9.4)

P,e
= F y, .................................................. (9.5)
As before assume a position for the neutral axis and calculate F

and Yc Using these values of F80 and Ysc in Eq. (9.5) calculate Pu. From
Eq. (9.4) calculate F 80 inserting value of Pu found above.
Adjust the assumed neutral axis position until the two values of Fsc
calculated are equal or sufficiently close.
The above method of analysis was applied to 30 eccentrically loaded
circular columns tested by Hognestad8 and the results are contained
in Table A-10. The average of Ptest!Pcazc for the whole series was 1.05,
and the standard deviation 0.060. A histogram of the results is plotted
in Fig. 23.
Rectangular columns, reinforced at all four faces (Fig. 24)
Let
Let
Let
Let
G be center of action of steel compressive force.

G,, be center of action of steel tensile force.
F be resultant steel compressive force (~A. f,.).
F,, be resultant steel tensile force (~A. f,,).

P. = 0.85 f.' b k, c
+ F
F ................................. (10.1)
91.(
lOURNAL OF THE AMERICAN CONCRETE INSTITUTE
February 1961

P.e
= 0.85 f.' b k1 c
+-
k,
T) + F,. y,. + F,, y,, ........ (10.2)
In the general case an iterative solution of the above equations is

proposed.
Assume a position of the neutral axis and calculate the stress in
each bar of the reinforcement using:
f,.
= E, e,,. = E, e.
~)
; which must be
~ f.
Calculate the resultant steel forces F.c and F.t, and their centers of
action G.c and G81 With these values of F8 c and F 8 t, calculate P,. using
Eq. (10.1). Substitute this value of P,. in Eq. (10.2) together with the
previously calculated values of F.o, F.t, Ysc, y.~, and hence calculate
the neutral axis depth c.
The calculation should be repeated as often as is necessary until the
assumed and final values of c coincide or are sufficiently close. The
value of P 11 corresponding to this value of c is the ultimate load ca:
pacity of the column for the particular eccentricity considered.
Rectangular columns with bending on both principal axes (Fig. 25)
Let G be center of action of steel compressive force.
Let G,, be center of action of steel tensile force.
Let G. be center of action of concrete compressive force.

P,.
= 0.85 f.' A. + F
F,, ................................. (11.1)
where
A.
= area of compression
zone covered by rectangular stress block
F = resultant steel compressive force
F,,
= resultant
(~A.
(~A.
steel tensile force
f,.)
f,,)

0.85 f.' A. x.
= (F,, x,,
F,. x -
P. e.) ..................... (11.2)
0.85 f.' A. y.
= (F,, y,,
F y,. -
P,. e.) ..................... (11.3)
In the general case an iterative solution of the above three equations

is proposed.
Assume a position of the neutral axis cc and calculate the stress in each
bar of the reinforcement using:
f.,.
= E, e,,. = E,
Eu (
which must be ~ f.
915
\~
'\~
r-e:. --1
...'t r
...~
~t-
Axis
---
c:
faT
Strains
FST
Actual
Assumed
Stress Distribution
Fig. 25-Conditions at ultimate load in column subject to axial load and bendind on both principal axes
Calculate the resultant tensile and compressive steel forces, F.t and
F.c, and their centers of action G.t (x sf, y ' ) and Gsc(x 1c, y a ) Using the
above values of F.t and F.c, find Pu for the assumed neutral axis position
using Eq. (11.1). From Eq. (11.2) and (11.3) obtain Ac Yc and Ac Xv,
using the values of P.,., F 80 , etc. already found.
It is then possible to calculate the dimensions of the area covered
by the rectangular stress distribution and hence to obtain the position
of the neutral axis. The calculation must be repeated as often as is
necessary until assumed and calculated positions of neutral axis coincide or are sufficiently close. The value of P u corresponding to this
neutral axis position is the ultimate strength of the column.
The calculation of the dimensions of the compression zone from values
of Ac Yc and Ac Xc obtained as outlined above is considered below.
Case 1-Line AA cuts adjacent sides OP and OR as in Fig. 26.
A.x.
= -xy
6-
A.y.
= -xy
6-
:.X= //6(A.x.)
'V . <All>
and
916
Fig. 26 -
February 1961
Geometry of compression
zone, Case I
Case 2-Line AA cuts opposite sides PQ and OR as in Fig. 27.
A,x, = b" [ Y,
+ TY. J
A,y, = +[ Y," + Y, (Y,

:. Y,
=[
3 (
Ab;') -
i-(A,
+~ )]
y,) -
3 (
Ab;'
rJ
and
Case 3-Line AA cuts opposite sides of OP and QR as in Fig. 28.

A,
Xo
= +[ X, +
A,y, =
:. X,
=[
(A~,)
X,
Tt' [ X,
(X, + ~2
X J
+--f
v+(A,
X. = 3 [ ;. A, y, -
x,) -
3 (
A~y,)]
X, ]
Case 4-Line AA cuts adjacent sides PQ and QR as in Fig. 29.

Use Q as origin for x, y, and e.
X.a
F,, x,, = 0.85 f.' A, x .................... (11.2A)
P.e,- F y
F,,y,,
P.e. -
F"
= 0.85f.'A.y, ................... (11.3A)
917
Fig. 27 -
zone, Case 2
c
c
R
. Y,
..
x, -
6 [ -b t
x,
2 - A.x.
2
J[
6
2
t - b2-
[ -2-b t"
A.x.
A.y.
]2
c
Pr---------.o
P~.;.....;.------;
Yc
Q
Fig. 28 -
zone, Case 3
Fig. 29 -
zone, Case 4
918
February 1961
Case 5-Equivalent rectangular stress distribution covers entire cross

section. As in Case 4, use Q as origin. Let neutral axis CC intersect x
and y axes at X' and Y' distance x' and y' from Q.
Again f,n
= E, e.
) , which must be
~ f .
Assume position of neutral axis. Calculate stress in each bar, and hence
total steel force F. and point of action G. (x. y.). Then:
P,.
= F. + 0.85 f.' b t ......................................... (11.4)

P,.(ez)
= {0.85f.'bt}
-%-+ F,x, ............................. (11.5)
P.(e.)
= {0.85fc'bt}
++
and
F,y ............................. (11.6)
Substituting the value of Pu from Eq. (11.4), Eq. (11.5) and (11.6)
may be solved for x. and y . If these values of x. and y. do not agree
with the values calculated above using the assumed position of neutral
axis, a new position of neutral axis must be assumed and the process
be repeated as often as is necessary. In a simple section it may be
possible to link x. and y, to x' and y' algebraically, in which case iteration may be used to solve the problem rather than trial and error.
Test results for columns subject to axial load and bending about
both principal axes are rather scarce. The analysis proposed has been
applied to ten columns tested by Andersen and Lee 27 and the results are
contained in Table A-11. The average value of Ptest!Pca!c for the series
is 0.99, and the standard deviation is 0.046.
REFERENCES
1. Koenen, M., "Fiir die Berechnung der Starke der Monierschen Cementplatten," Zentralblatt der Bauverwaltung (Berlin), V. 6, No. 47, Nov. 1886, p. 462.
2. Hognestad, E., "Confirmation of Inelastic Stress Distribution in Concrete,"
Proceedings, ASCE, ST 2, Mar. 1957, pp. 1-17. Also Bulletin D15, PCA Development Department.
3. Kriz, L. B., "Ultimate Strength Criteria for Reinforced Concrete," Proceedings, ASCE, EM 3, July 1959, pp. 95-110. Also Bulletin D31, PCA Development Department.
4. Emperger, F.v., "Ein graphischer Nachweis der Tragfahigkeit und aller
in einem Tragwerke aus Eisenbeton auftretenden Spannungen," Beton und Eisen
(Vienna), V. 4, No. 5, 1904, pp. 306-320.
5. Whitney, C. S., "Plastic Theory of Reinforced Concrete Design," Proceedings, ASCE, V. 66, No. 10, Dec. 1940, pp. 1749-1780.
6. Richart, F. E., and Brown, R. L., "An Investigation of Reinforced Concrete
Column," Bulletin No. 267, University of Illinois Engineering Experiment Station, June 1934, 91 pp.
919
7. Jensen, V. P., "The Plasticity Ratio of Concrete and Its Effects on the
Ultimate Strength of Beams," ACI JoURNAL, V. 14, No. 6, June 1943 (Proceedings
v. 39)' pp. 565-582.
8. Hognestad, E., "A Study of Combined Bending and Axial Load in Reinforced Concrete Member," BuHetin No. 399, University of Illinois Engineering
Experiment Station, Nov. 1951, 128 pp.
9. Mattock, A. H., "The Strength of Single Reinforced Beams in Bending,"
Symposium on the Strength of Concrete Structures, London, May 1956.
10. Hognestad, E.; Hanson, N. W.; and McHenry, D., "Concrete Stress Distribution in Ultimate Strength Design," ACI JoURNAL, V. 27, No. 4, Dec. 1955 (Proceedings V. 52), pp. 455-479; also Bulletin D6, Portland Cement Association Development Department.
11. Riisch, H., "Versuche zur Festigkeit der Biegedruckzone," Heft 120, Deutscher Ausschuss fur Stahlbeton (Berlin), 1955, 94 pp.
12. Billet, D. F., and Appleton, J. H., "Flexural Strength of Prestressed Concrete Beams," ACI JouRNAL, V. 25, No. 10, June 1954 (Proceedings V. 50), pp.
837-854.
13. Slater, W. A., and Lyse, I., "Compressive Strength of Concrete in Flexure
as Determined from Tests of Reinforced Beams," ACI JouRNAL, V. 1, No. 8, June
1930 (Proceedings V. 26), pp. 831-874.
14. Columbia University, "Comparative Tests of Concrete Beams Reinforced
with Isteg and Hot Rolled Deformed Bars," Report No. 2507, Columbia University,1941.
15. Cox, K. C., "Test of Reinforced Concrete Beams with Recommendations
for Attaining Balanced Design," ACI JoURNAL, V. 13, No. 1, Sept. 1941 (Proceedings V. 38), pp. 65-80.
16. Lash, S. D., and Brison, J. W., "The Ultimate Strength of Reinforced Concrete Beams," ACI JouRNAL V. 21, No. 7, Feb. 1950 (Proceedings V. 46), pp.
457-472.
17. Gaston, J. R., "An Investigation of the Load Deformation Characteristics
of Reinforced Concrete Beams up to the Point of Failure," MS Thesis, University
of Illinois, 1952.
18. McCollister, H. M., "Load Deformation Characteristics of Simulated Beam
Column Connections in Reinforced Concrete," MS Thesis, University of Illinois,
1954.
19. Guralnick, S. A., "An Investigation of High Strength Deformed Steel Bars
for Concrete Reinforcement," Cornell University School of Civil Engineering
Report No. TSR 4730-7146, Part I, July 1957, Part II, Nov. 1958.
20. Johnson, A., "Concrete Beams with Compression Reinforcement," (in
Swedish), Beton, V. 35, No. 1, 1950, pp. 113-125.
21. Talbot, A. N., "Tests of Reinforced Concrete T-Beams," Bulletin No. 12,
University of Illinois Engineering Experiment Station, 1907.
22. Gaston, J. R., and Hognestad, E., "Precast Concrete Girders Reinforced
with High Strength Deformed Bars," ACI JOURNAL V. 30, No. 4, 1958 (Proceedings V. 55), pp. 469-484; also Bulletin D24, Portland Cement Association Development Department.
23. Janney, J. R.; Hognestad, E.; and McHenry, D., "Ultimate Flexural Strength
of Prestressed and Conventionally Reinforced Concrete Beams," ACI JOURNAL
V. 27, No. 5, Jan. 1956 (Proceedings V. 52), pp. 601-620; also Bulletin D7, Portland Cement Association Development Department.
24. Feldman, A., "Bonded and Unbonded Prestressed Concrete Beams Failing
in Flexure," MS Thesis, University of Illinois, 1954.
25. Allen, R. J., "Flexural Strength of Unbonded Prestressed Concrete Beams,"
MS Thesis, University of Illinois, 1956.
920
February 1961
26. Mattock, A. H., "A Study of the Ultimate Moment of Resistance of Prestressed and Reinforced Concrete Beams," PhD Thesis, University of London, 1955.
27. Andersen, P., and Lee, H. N., "A Modified Plastic Theory of Reinforced
Concrete," Bulletin No. 33, University of Minnesota Engineering Experiment
Station.
28. Mattock, A. H., and Kriz, L. B., "Ultimate Strength of Structural Concrete
Members with Nonrectangular Compression Zones," ACI JoURNAL, V. 32, No. 7,
Jan. 1961 (Proceedings V. 57), pp. 737-766.
Received by the Institute Apr. 11, 1960. Title No. 57-43 is a part of copyrighted Journal of
the American Concrete Institute, V. 32, No. 8, Feb. 1961 (Proceedings V. 57l. Separate prints
are available at $1 each.
American Concrete Institute, P. 0. Box 4754, Redford Station, Detroit 19, Mich.
Discussion of this paper should reach ACI headquarters in triplicate by

May 1, 1961, for publication in the Part 2 September 1961 JOURNAL.
APPENDIX- TEST DATA

TABLE A-I-REINFORCED CONCRETE BEAMS REINFORCED IN
TENSION ONLY (Ultimate strength controlled by compression)
Source
Beam
No.
t,
PSI
t .
ksi
Muu
bd2 fo'
Test
Calculated
Test
Calculated
W. A. Slater
and
I. Lyseta
1
2
3
4
5
6
6A
7
8
9
10
lOA
1390
2790
4070
4800
5740
2590
4130
2950
2760
2820
2820
3810
64.8
64.8
64.8
64.8
64.8
64.8
64.8
64.8
64.8
64.8
64.8
64.8
0.021
0.028
0.037
0.047
0.056
0.030
0.039
0.028
0.031
0.032
0.030
0.040
0.506
0.337
0.326
0.339
0.320
0.422
0.327
0.341
0.380
0.391
0.346
0.354
0.362
0.342
0.336
0.333
0.323
0.349
0.337
0.339
0.348
0.347
0.345
0.344
1.40
0.99
0.97
1.02
0.99
1.21
0.97
1.01
1.09
1.13
1.00
1.03
Columbia
University"
Cl
Cll
C2
C12
3550
3550
3550
3550
61.37
62.00
63.04
64.28
0.0341
0.0345
0.0334
0.0328
0.406
0.394
0.386
0.365
0.339
0.340
0.338
0.337
1.20
1.16
1.14
1.08
921
TABLE A-I (cont.)- REINFORCED CONCRETE BEAMS REINFORCED IN

TENSION ONLY (Ultimate strength controlled by compression)
Source
Beam
No.
fo'
psi
ksi
Muu
bd f'
Test
K. C. Cox1
s.
L. Lash
and
J. W. Brison1
Calculated
Test
Calculated
4500
4500
4500
5800
5800
53.4
53.4
53.4
53.4
48.1
48.1
48.1
53.4
53.4
48.1
48.1
55.2
55.2
48.1
50.6
50.6
53.4
48.1
48.1
53.4
48.1
0.0176
0.0264
0.0352
0.0440
0.0244
0.0368
0.0488
0.0352
0.0440
0.0368
0.0488
0.0308
0.0388
0.0388
0.0348
0.0520
0.0440
0.0368
0.0488
0.0440
0.0488
0.405
0.423
0.455
0.471
0.442
0.426
0.490
0.343
0.353
0.370
0.375
0.374
0.372
0.381
0.379
0.388
0.320
0.297
0.342
0.293
0.300
0.343
0.363
0.374
0.382
0.359
0.376
0.385
0.348
0.359
0.350
0.363
0.341
0.353
0.353
0.348
0.366
0.336
0.326
0.341
0.309
0.314
1.18
1.16
1.22
1.23
1.23
1.13
1.27
0.98
0.98
1.06
1.03
1.10
1.05
1.08
1.09
1.06
0.95
0.91
1.00
0.94
0.95
1970
1930
3330
4170
4490
2150
2150
1950
2080
1915
2120
3290
2760
3200
2760
4490
4140
4190
4190
4870
4450
5450
39.2
44.4
42.8
40.8
43.4
88.0
75.8
75.8
73.5
75.2
75.2
72.0
75.8
74.0
75.2
75.8
74.0
75.2
62.1
75.8
65.0
75.8
0.0400
0.0475
0.0454
0.0367
0.0471
0.0140
0.0200
0.0225
0.0284
0.0385
0.0391
0.0147
0.0233
0.0286
0.0394
0.0226
0.0280
0.0390
0.0408
0.0233
0.0371
0.0458
0.449
0.464
0.367
0.305
0.346
0.346
0.333
0.415
0.400
0.391
0.407
0.270
0.347
0.338
0.359
0.263
0.293
0.317
0.314
0.247
0.299
0.292
0.374
0.380
0.357
0.333
0.339
0.316
0.338
0.349
0.357
0.373
0.370
0.292
0.332
0.335
0.359
0.296
0.318
0.336
0.338
0.289
0.327
0.318
1.20
1.22
1.03
0.91
1.02
1.09
0.99
1.19
1.12
1.05
1.10
0.92
1.04
1.01
1.00
0.89
0.92
0.94
0.93
0.85
0.91
0.92
122
123
124
125
142
143
144
224
225
243
244
214
215
235
252
253
325
343
344
425
444
1700
1700
1700
1700
1700
1700
1700
3100
3100
3100
3100
3100
3100
3100
3100
4205
4206
4308
4407
4408
6203
6204
6205
6206
6207
6208
6303
6304
6305
6306
6404
6405
6406
6407
6504
6505
6506
inoo
Avg, all beams (59) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.06

Standard deviation, all beams .................................... 0.113
Avg for beams with f' > 2000 psi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.02
Standard deviation for beams with f' 2000 psi . . . . . . . . . . . . . . . . . . . 0.083
-o
TABLE A-2-REINFORCED CONCRETE BEAMS REINFORCED IN TENSION AND COMPRESSION
Source
Gastonl7t
(tension
failures)
McCollister11
(tension
failures)
GuralnicklD
(tension
failures)
Beam
d'
A,
A,'
p,
p',
M
fo't
percent
percent
ps1
t.,
ksi
fv',
ksi
b(jjfT
Test
Calc
Test
Calc
tv
tv
Failure
modes
6
6
6
6
6
6
10.58
10.58
10.37
10.37
10.51
10.37
0.90
1.29
1.48
1.48
1.41
1.55
0.88
0.88
2.00
2.00
1.20
2.00
0.40
0.40
1.20
1.20
0.62
1.20
1.38
1.38
3.22
3.22
1.90
3.22
0.63
0.63
1.93
1.93
0.98
1.93
3940
4070
4310
3890
3330
4860
45.4
53.3
41.8
41.8
45.2
42.1
44.5
47.0
46.7
42.5
56.1
47.4
0.153
0.175
0.275
0.323
0.236
0.242
0.148
0.163
0.275
0.303
0.230
0.245
1.03
1.07
1.00
1.07
1.03
0.99
2
2
2
2t
2
2
c4xna
c4xnb
c4zn
c5yn
c6xm
c7w
6
6
6
6
6
6
10.51
10.51
10.37
9.38
9.38
9.38
1.22
1.51
1.48
1.57
1.57
1.57
1.20
1.20
2.00
3.16
3.16
3.16
0.62
0.88
1.20
1.58
1.58
1.58
1.90
1.90
3.22
5.61
5.61
5.61
0.98
1.39
1.93
2.80
2.80
2.80
2450
2430
3570
4480
3680
3480
45.5
46.4
41.3
44.0
41.8
41.6
41.4
44.1
46.4
43.4
40.2
43.6
0.310
0.324
0.329
0.457
0.531
0.552
0.313
0.319
0.326
0.463
0.526
0.552
0.99
1.02
1.01
0.99
1.01
1.00
lt
2
2
1t
1t
1t
B-34
B-35
T-1
T-2
T-3
T-7
T-10
6
6
6
6
6
6
6
10.58
10.51
10.58
10.37
10.20
10.58
10.58
1.40
1.41
1.28
1.51
1.56
1.42
1.33
0.88
1.20
0.88
2.00
3.12
0.88
0.88
0.40
0.62
0.40
1.20
1.58
0.88
0.22
1.39
1.90
1.39
3.21
5.10
1.39
1.39
0.63
0.98
0.63
1.93
2.58
1.39
0.35
4760
2345
3897
3858
4266
4540
4330
46.5
48.4
41.8
45.4
44.7
40.9
42.6
47.2
47.2
47.9
50.0
46.1
42.5
46.3
0.139
0.347
0.148
0.334
0.455
0.127
0.142
0.125
0.343
0.138
0.330
0.453
0.116
0.126
1.00
1.01
1.07
1.01
1.01
1.09
1.13
2
1t
2
2
1t
2
2
T-12
T-14
T-15
T-4
T-5
T-6
6
6
6
6
6
6
10.28
10.58
10.58
10.65
10.51
10.37
1.84
1.52
1.50
1.28
1.35
1.49
2.54
0.88
0.88
0.62
1.20
2.00
2.00
0.88
0.88
0.40
0.62
1.20
4.12
1.39
1.39
0.97
1.90
3.21
3.24
1.39
1.39
0.63
0.98
1.93
4367
4030
3700
2230
2021
1905
46.0
41.4
47.0
47.1
48.4
56.5
45.8
40.9
40.5
45.7
46.6
49.7
0.390
0.145
0.171
0.190
0.407
0.802
0.369
0.131
0.159
0.186
0.396
0.790
1.06
1.00
1.07
1.02
1.03
1.01
2
2
2
2
1t
2
IB-1
IIB-1
IID-1
IIIA-1
IIIB-1
23
23
23
23
23
11.81
11.81
12.05
11.81
11.81
2.31
2.31
2.81
2.74
2.31
1.20
1.20
2.08
2.08
1.20
1.20
1.20
2.08
2.08
1.20
0.442
0.442
0.750
0.765
0.442
0.442
0.442
0.750
0.765
0.442
2440
2440
5520
3130
3130
84.6
84.6
87.7
87.7
84.6
84.6
84.6
87.7
87.7
84.6
0.157
0.197
0.133
0.216
0.161
0.135
0.135
0.112
0.188
0.112
1.16
1.46
1.19
1.15
1.44
2
2
2
2
2
UIB-2
IIIC-1
IIID-1
23
23
23
12.12
11.78
11.81
2.63
2.68
2.72
1.20
3.73
2.08
1.20
3.73
2.08
0.430
1.380
0.766
0.430
1.380
0.766
3130
5300
5300
84.6
83.9
87.7
84.6
83.9
87.7
0.116
0.203
0.159
0.110
0.207
0.125
1.05
0.98
1.27
2
2
2
c2w
c2xm
c3w
c3xm
c3yna
c3ynb
---
;;o
)>
,0
-I
I
m
)>
s:m
;;o
()
)>
()
()
;;o
Vl
-I
=i
c
-I
m
CD
g-
.
c
-<
-o
0.
Johnson
(Compression
failures)
10
14
3
7
'
4.02
3.94
4.0Z
3.98
4.14
3.98
5.40
5.28
5.28
5.28
5.12
5.28
0.67
0.67
0.94
0.67
0.71
0.67
3.94
3.94
3.94
3.94
4.02
5.24
5.24
5.24
5.24
5.32
0.67
0.83
0.55
0.51
0.87
11
15
4
12
16
..
0.089
0.089
0.089
0.089
0.244
0.244
4.6
4.6
4.6
4.6
4.6
4.6
0.41
0.43
0.42
0.42
1.15
1.16
2860
2820
3200
2830
3240
2780
96.0
96.0
96.0
96.0
96.0
96.0
53.8
51.1
51.1
51.5
46.1
46.1
0.503
0.442
0.427
0.435
0.497
0.585
0.425
0.426
0.408
0.438
0.488
0.520
1.18
1.04
1.05
0.99
1.02
1.12
3
3
3
3
3
3
0.244
0.244
0.528
0.528
0.528
4.6
4.6
4.6
4.6
4.6
1.20
1.18
2.56
2.56
2.47
2780
3020
2900
2790
2810
96.0
96.0
96.0
96.0
96.0
39.0
39.0
41.8
43.0
42.5
0.530
0.493
0.637
0.712
0.590
0.498
0.478
0.663
0.693
0.650
1.06
1.03
0.96
1.03
0.91
3
3
3
3
3
-------
(1) Both tension and compression reinforcement yielded before ultimate strength; (2) Only tension reinforcement yielded; (3) Only compression reinforcement yielded.
t Test failure moment in this series is taken as maximum moment before strain hardening of tension reinforcement commenced.
: Test failure moment in this series is taken as the moment recorded at which the concrete in the compression zone crushed.
In both the above series a further small increase in moment was obtained in several cases due to strain hardening of the tension reinforcement. However, this increase in load was accompanied by extremely large deflections and therefore was neglected. It should be noted
that strain hardening of the structural or intermediate grade steel reinforcement only occurred in cases where either a very small percentage
of tension reinforcement was provided or where an excessive amount of compression reinforcement was provided such that the compression
reinforcement did not yield. For normal amounts of steel used in practice, the ultimate moment was not increased by ~train hardening,
~
~
~
;o
m
TABLE A-3-REINFORCED CONCRETE T-BEAMS REINFORCED IN TENSION ONLY

Source
A. N. Talbot"'
S. A. Guralnicku
Beam
in.
_d,
m.
t,
b',
in.
ksi
Test
Calc
1.22
1.08
1.12
1.15
1.28
2.20
1.76
2.20
3.08
0.92
1.10
0.92
0.97
1610
1450
1750
1610
38.3
38.3
40.7
38.3
0.200
0.249
0.187
0.195
0.193
0.244
0.187
0.197
1.03
1.02
1.00
0.99
4
4
4
4
2.08
1.20
3.72
2.08
0.766
0.442
1.375
0.766
3230
2440
4930
4930
87.7
84.6
83.9
87.7
0.200
0.184
0.205
0.138
0.182
0.139
0.202
0.125
1.10
1.32
1.02
1.10
2.75(avg)
2.75(avg)
1.20
1.60
0.82
1.11
4730
5230
90.0
90.0
0.149
0.185
0.134
0.159
1.11
1.16
10
10
10
10
8
8
8
8
3.25
3.25
3.25
3.25
IA-IR
23
23
23
23
11.81
11.81
11.78
11.81
7
7
7
7
9
9
16.25
16.00
3.5
3.5
VI
0.250
0.248
0.233
0.258
0.340
24
16
24
32
Mcalc
0.305
0.269
0.262
0.296
0.435
6
7
8
9
l::t
Mcut
54.9
53.8
52.7
38.3
53.4
3.25
3.25
3.25
3.25
3.25
=1.11.
psi
M
bd f'
t.
1890
1870
1760
1330
1190
8
8
8
8
8
Avg Mt .. t/Meal
percent
/o',
1.05
1.05
0.93
1.10
1.05
10
10
10
10
10
1
2
p,
1.68
3.36
2.24
1.76
3.36
16
32
24
16
32
m-IR
A,
in.
1
2
3
4
5
IC-IR
ID-IR
J. R. Gaston and
E. Hognestad22
b,
i5
Standard deviation = 0.10.
,
February 1961
924
__ _
TABLE A-4-PRESTRESSED CONCRETE BEAMS WITHOUT BOND

Mult
Beam
No.
b,
in.
d,
ln.
p,
percent
fae,
ksi
Steel
tensile
strength,
ksi
Test
Calc
J. R. Janney,
E. Hognestad,
and
D.McHenryaa
3-0.128
3-0.144
3-0.307
3-0.428
3-0.428
6.0
6.0
6.0
6.0
6.0
8.3
8.3
8.3
8.3
8.3
0.322
0.322
0.644
0.965
0.965
110
122
130
120
133
235
235
235
235
235
5900
5250
4930
5300
5300
0.084
0.102
0.200
0.229
0.238
0.079
0.093
0.184
0.233
0.241
A. Feldman..
U-1
U-3
U-4
U-5
U-7
U-8
5.99
6.00
6.16
6.09
6.00
6.00
8.42
7.85
8.33
8.24
7.50
7.65
0.346
0.183
0.782
0.685
0.382
0.624
127.5
120.0
118.9
120.7
121.9
118.9
250.76
250.76
250.76
250.76
250.76
250.76
4240
5620
4060
4060
5020
2565
0.139
0.055
0.274
0.273
0.120
0.285
0.135
0,048
0.250
0.271
0.124
0.290
1.03
1.15
1.10
1.01
0.97
0.95
U-9
U-10
U-11
U-12
U-14
U-16
6.00
6.02
5.90
6.00
5.95
5.96
7.35
7.55
8.20
8.34
7.84
7.56
0.650
0.820
0.355
0.401
0.184
0.509
120.0
118.8
122.8
125.8
122.1
121.1
250.76
250.76
250.76
250.76
250.76
250.76
3550
3390
5490
4020
7600
2190
0.273
0.328
0.121
0.173
0.049
0.308
0.246
0.292
0.107
0.145
0.043
0.287
1.11
1.12
1.13
1.19
1.14
1.07
U-17
U-18
U-19
U-21
U-22
U-23
U-24
6.20
6.10
6.04
6.20
6.10
6.00
6.00
7.39
7.45
6.85
7.60
7.32
6.99
7.85
0.527
0.797
0.365
0.384
0.743
0.288
0.192
118.8
110.5
123.7
119.8
118.2
117.7
119.0
255
255
255
255
255
255
255
2120
2770
6270
2450
4910
7580
5660
0.330
0.339
0.113
0.218
0.227
0.049
0.064
0.304
0.321
0.101
0.223
0.215
0.067
0.061
1.09
1.06
1.12
0.98
1.05
0.72
1.05
Source
R. J. Allen..
Avg Mteai!Mcalc for 24 beams= 1.05.
bd2 fc'
fc'!
PSl
Mtesf
Mcalc
~-
1.06
1.10
1.09
0.98
0.99
1--
Standard deviation = 0.094.
TABLE A-S-PRESTRESSED CONCRETE BEAMS WITH WELL BONDED

TENDONS
Mull
ksi
Steel
tensile
strength,
ksi
fc',
psi
0.198
0.101
0.942
0.953
0.418
116.8
120.0
112.8
112.9
19.1
246
246
248
248
240
9.01
9.21
8.15
7.99
9.29
0.107
0.419
0.656
0.916
0.418
19.0
20.4
21.2
20.2
150.0
6.01
6.00
6.00
6.08
6.08
9.00
9.09
8.29
8.27
9.05
0.108
0.429
0.647
0.873
0.284
B22
B23
B24
B25
B27
6.07
6.04
6.07
6.06
6.07
9.13
8.20
8.24
8.01
8.36
A. Feldman..
B28
B29
B30
B31
B32
B33
6.15
6.16
6.09
6.08
6.00
6.03
J. R. Janney,
E. Hognestad,
and
D. McHenry..
1-0.141
1-0.250
1-0.420
2-0.151
2-0.306
2-0.398
Source
D. F. Billet
and
H. H. Appleton,.
b,
in.
d,
in.
p,
percent
f11e,
B2
B3
B7
B8
B9
6.15
6.00
6.13
6.13
6.06
9.53
9.62
8.09
7.99
9.23
B10
B11
B13
B14
B15
6.06
6.06
6.02
6.00
6.03
B16
B17
B18
B19
B21
Beam
No.
6
6
6
6
6
6
7.93
8.07
8.08
8.23
9.32
9.08
18.3
8.3
8.3
8.3
8.3
8.3
bd2 fc'
Mtest
Mcalc
Test
Calc
5420
3750
5910
3280
6330
0.087
0.065
0.272
0.462
0.129
0.079
0.061
0.256
0.349
0.128
1.10
1.06
1.06
1.32
1.01
240
240
240
240
240
3530
3910
3750
3750
5710
0.068
0.208
0.283
0.327
0.143
0.068
0.190
0.242
0.266
0.143
1.00
1.09
1.17
1.23
1.00
150.3
151.0
148.8
151.3
118.0
240
240
240
240
248
3330
4580
4100
6220
6560
0.077
0.179
0.273
0.244
0.093
0.073
0.178
0.264
0.242
0.098
1.05
1.01
1.03
1.01
0.95
0.561
0.943
0.746
0.641
0.920
115.2
117.3
116.4
114.5
118.0
248
248
248
248
248
7630
8200
6120
3270
4590
0.153
0.213
0.235
0.349
0.319
0.150
0.206
0.221
0.294
0.295
1.02
1.05
1.06
1.19
1.08
0.475
0.815
0.177
0.579
0.510
0.312
92.5
92.7
101.1
94.1
115.3
116.9
186
186
248
248
256
256
2500
4280
2890
3450
7180
8320
0.281
0.261
0.135
0.279
0.154
0.093
0.260
0.240
0.129
0.250
0.158
0.091
0.322
0.644
0.965
0.322
0.644
0.965
119
113
117
126
118
117
235
235
235
235
235
235
5350
6050
5400
5000
4950
5700
0.144
0.218
0.260
0.146
0.237
0.271
0.127
0.186
0.258
0.134
0.214
0.251
Average M,.,,fMcalc: All 32 beams= 1.07.
Standard deviation= 0.077.
1.08
1.09
1.05
1.12
0.97
1.02
--1.13
1.17
1.01
1.07
1.11
1.08
925
TABLE A-6-UNSYMMETRIC REINFORCED CONCRETE T-BEAMS

Source
Beam
A. H. Mattock
and
L. B.Kriz2B
Flange
width,
B, in.
Ultimate
Effective
depth,
d, in.
A,,
t .
f/,
psi
sqin.
ksi
moment,
Mte.d
in.-kip
McaZc
Calc
Test
Ll
L2
L3
L4
L5
L6
8
8
8
8
12
12
13.8
13.8
12.9
12.9
13.8
13.8
2820
6200
3260
5570
2050
6320
1.20
1.20
2.40
2.40
1.20
1.20
45.6
46.1
45.6
51.0
47.9
47.5
647
710
952
1300
627
733
678
693
1062
1325
671
778
1.05
0.98
1.12
1.02
1.07
1.06
L7
L8
L9
L10
L11
L12
12
12
16
16
16
16
12.9
12.9
13.8
13.8
12.9
12.9
2225
6420
2680
4820
2530
5820
2.40
2.40
1.20
1.20
2.40
2.40
43.0
43.0
45.4
44.5
50.7
50.5
695
1160
637
678
775
1298
876
1173
663
757
1129
1265
1.26
1.01
1.04
1.12
1.46
0.97
Type
of
failure
T
T
c
CT
T
T
c
T
T
T
c
T
(a) All beams: Avg Mlui!Mcalo = 1.10; Standard deviation= 0.138.

(b) Beams with ultimate strength controlled by tension: Avg M,.,lfMcalc = 1.04; Standard
deviation= 0.047.
TABLE A-7-REINFORCED CONCRETE BEAMS WITH TRIANGULAR

COMPRESSION ZONES
Source
A. H.
Mattock
and
L. B.
Kriz28
f.,
f.',
Beam
psi
ksi
A,,
sq in.
T1
T2
T3
T4
T5
T6
3620
3455
6290
1735
3500
7000
49.0
49.0
45.0
49.5
50.5
45.0
0.88
1.32
1.80
0.88
1.32
1.80
Ultimate moment,
in.-kip
Calc
Test
194.5
215.0
320.0
116.3
219.0
323.0
196.8
240.9
326.8
131.8
258.4
368.3
M~e .,
Mcalo
1.01
1.12
1.02
1.13
1.18
1.14
For all beams, d = 7 m., 0 = 45 deg.

Ultimate strength of all beams controlled by crushing of concrete.
Avg M .. I!Mcalo = 1.10.
TABLE A-S-TIED REINFORCED CONCRETE COLUMNS

LOADED CONCENTRICALLY*
Column
Percentage reinforcement
Yield point
of reinforcement,
psi
psi
fc',
Ultimate load,
kip
P,.,,
Peale
Test
Calc
4
4
50,000
50,000
2860
3090
219
255
231
242
0.95
1.05
4
4
50,000
50,000
2650
2850
253
238
222
231
1.14
1.03
1.5
1.5
44,700
44,700
4700
4150
225
227
246
222
0.92
1.02
4
4
50,000
50,000
4670
4730
285
320
310
313
0.92
1.02
4
4
50,000
50,000
4225
4570
293
309
291
306
1.01
1.01
a
b
a
b
Continued on p. 926
926
1961
February
TABLE A-8 (cont.)-TIED REINFORCED CONCRETE COLUMNS

LOADED ECCENTRICALLY*
Column
Percentage reinforcement
Yield point
of reinforcement,
psi
psi
Ultimate load,
kip
fo',
P,.,,
Pclo
Test
Calc
6
6
42,200
42,200
4215
4985
317
291
315
348
1.01
0.84
4
4
50,000
50,000
5870
6950
353
387
363
410
0.97
0.94
379
391
1.08
1.08
410
4
50,000
6245
420
4
50,000
6530
Average value of p,.,,;p ,. = 1.00.
b
source: Series 3 of ACI column investigation-University of Illinois Bulletin No. 267.
TABLE A-9-RECTANGULAR REINFORCED CONCRETE COLUMNS

LOADED ECCENTRICALLY
Group I
Source
and
characteristics
E. Hognestad
b =t= lOin.
A. = 1.24 sq in.
A,' = 0.22 sq in.
fr = 43.6 ksi
E, = 28 X 10" psi
f/ = 60.0 ksi
d= 8.67 in.
d' = 1.33 in.
Column
No.
Concrete
cylinder
strength,
f' psi
Eccentricity
(total),
in.
A-la
A-lb
B-la
B-lb
C-la
C-lb
A-2a
A-2b
B-2a
B-2b
C-2a
C-2b
A-3a
A-3b
B-3a
B-3b
C-3a
C-3b
A-4a
A-4b
B-4a
B-4b
C-4a
C-4b
A-5a
A-5b
B-5a
B-5b
C-5a
C-6b
5280
5660
4250
4070
2270
2020
5280
5830
4250
4070
2270
1970
5660
5830
4630
4290
1880
1690
4810
5600
3800
4290
1690
1730
4810
5600
4290
4590
2310
1770
0.12
0.14
0.12
0.12
0.13
0.13
2.72
2.78
2.77
2.74
2.77
2.77
5.32
5.28
5.41
5.37
5.28
5.33
7.95
7.85
7.98
8.02
7.82
7.81
12.90
12.90
12.92
12.95
12.84
12.84
...
Ultimate load,
kip
p,.,,
p-;;;;-;;-
Mode of
failure
Test
Calc
388
441
343
352
222
191
239
253
213
190
118.5
100.0
133.5
140.0
125.9
116.0
60.5
64.0
84.5
81.0
80.0
81.0
50.5
52.0
48.2
42.8
46.1
45.5
39.0
32.8
452
481
372
356
212
192
240
260
206
197
116
103
154.0
158.0
134.3
129.4
66
52
85.7
92.1
77.7
81.1
48
49
45.6
46.9
44.5
45.0
37.7
34.0
0.86
0.92
0.92
0.99
1.05
1.00
1.00
0.97
1.03
0.96
1.02
0.97
0.87
0.89
0.94
0.90
0.92
1.03
0.99
0.88
1.03
1.00
1.05
1.06
1.06
0.91
1.04
1.01
1.04
0.96
T
T
T
T
CT
CT
456
420
225
202
437
436
268
222
1.04
0.96
0.84
0.91
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
T
T
CT
CT
c
c
T
T
T
T
c
c
Group II
E.
Hognestad1
b ::::ot = 10 in.
B-6a
B-6b
C-6a
C-6b
4080
4040
2020
1520
0,07
0.06
0.10
0.18
927
TABLE A-9 (cont.)-RECTANGULAR REINFORCED CONCRETE

COLUMNS LOADED ECCENTRICALLY
Source
and
characteristics
A, = 1.24 sq in.
A,' = 1.24 sq in.
t. =
43.6 psi
E, = 28 X Hl" psi
t. =
43.6 psi
d= 8.67 in.
d' = 1.33 in.
Column
No.
A-7a
A-7b
B-7a
B-7b
C-7a
C-7b
A-Sa
A-Bb
B-8a
B-Bb
C-Ba
C-8b
A-9a
A-9b
B-9a
B-9b
C-9a
C-9b
A-lOa
A-lOb
B-lOa
B-lOb
C-10a
C-lOb
Concrete
cylinder
strength,
fc', psi
5240
5810
4080
4040
1970
1520
5520
5810
4700
4260
1820
1820
5100
5170
4700
4370
1880
1730
5100
5170
4260
4370
2300
1770
Eccentriclty
(to.tal),
m.
3.44
2.76
2.75
2.74
2.78
2.77
5.34
5.40
5.35
5.32
5.32
5.39
7.87
7.89
7.85
7.82
7.88
7.85
12.78
12.75
12.78
12.79
12.85
12.88
Ultimate load,
kip
Ptut
Pca~o
Test
Calc
274
284
256
248
141
126.8
162
152
156
146
99
99
89.0
91.2
94.0
89.5
73.0
65.5
46.1
44.0
43.5
44.0
44.5
45.0
254
312
245
244
147
135
177.9
179.5
164.9
158.6
103
102
99.1
99.1
97.7
96.6
75.8
74.1
47.8
48.0
47.2
47.2
44.3
42.6
1.02
0.96
0.94
0.91
0.85
0.95
0.92
0.96
0.97
0.90
0.92
0.96
0.93
0.96
0.88
0.97
0.92
0.92
0.93
1.00
1.06
500
485
353
315
326
303
284
252
230
220
210
180
206
151
137
142
153
138.8
115.5
104.0
88.0
79.0
74.0
84.5
72.5
74.5
513
517
376
306
343
318
298
224
218
227
216
188
206
153
148
159.0
155.5
141.0
108
111
82.8
83.0
80.3
82.0
74
73.5
0.98
0.94
0.94
1.03
0.95
0.95
0.95
1.12
1.05
0.97
0.97
0.96
1.00
0.99
0.93
0.89
0.98
0.98
1.07
0.94
1.06
0.95
0.92
1.03
0.97
1.01
1.08
0.91
UK
Mode of
failure
c
c
c
c
c
c
T
T
CT
CT
c
c
T
T
T
T
CT
CT
T
T
T
T
T
T
Group Ill
E. Hognestad
b = t = 10 in.
A, = 2.40 sq in.
A,'= 2.40 sq in.
fv = 43.6 ksi
E, = 29 X 10 psi
fv = 43.6 ksi
d=8.50 in.
d' = 1.50 in.
B-11a
B-11b
C-11b
A-12a
A-12b
B-12a
B-12b
C-12a
C-12b
A-13a
A-13b
B-13a
B-13b
C-13a
C-13b
A-14a
A-14b
B-14a
C-14a
C-14b
A-15a
A-15b
B-15a
B-15b
C-15a
C-15b
3870
4070
2070
4150
5050
4300
4010
2300
2200
5350
4850
3580
4290
2300
2070
5350
5100
3580
1950
2070
5100
4850
3800
4630
1950
2070
0.08
0.10
0.00
2.70
2.72
2.72
2.76
2.76
2.72
5.36
5.34
5.35
5.34
5.33
5.28
7.87
7.93
7.89
7.84
7.87
12.92
12.85
12.91
12.92
12.89
12.91
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
T
T
CT
c
c
T
T
T
T
c
CT
Avg p,.,,;p ,. = 0.97.

Note: The eccentricity recorded in this table is measured from the center of the column.
928
February 1961
TABLE A-ID-CIRCULAR SECTION COLUMNS SUBJECT TO

ECCENTRIC LOAD*
Column
No.
Eccentricity,
e, in.
fc', psi
Ultimate load,
kips
Test
Calc
Pte I
~
Cross
section
A-16a
A-16b
5150
4640
0
0.01
760
693
704
655
1.08
1.06
All columns
12-in. diameter.
B-16a
B-16b
2990
3310
0.03
0.01
515
514
497
527
1.03
0.98
C-16a
C-16b
1590
1420
0.03
0.02
371
365
362
345
1.02
1.06
Reinforcement:
eight %-in.
diameter
A-17a
A-17b
5150
4640
3.30
3.29
343
283
308.6
291.9
1.11
0.97
B-17a
B-17b
3620
3310
3.34
3.34
253
238
251.7
239.3
1.01
0.99
C-17a
C-17b
1420
1600
3.55
3.50
187
179
160.0
164.9
1.17
1.08
A-lBa
A-18b
5020
5000
6.44
6.50
162
171
164.3
162.5
0.99
1.05
B-18a
B-18b
3380
3580
6.42
6.47
140
136
140.8
143
0.99
0.95
C-18a
C-18b
1680
1590
6.80
6.60
127
107
100.6
101.9
1.26
1.05
A-19a
A-19b
5020
5310
9.62
9.62
111.0
114.3
107.2
109.4
1.04
1.04
B-19a
B-19b
3380
3580
9.54
9.56
98.5
103.0
93.2
95.4
1.06
1.08
C-19a
C-19b
1680
1630
9.80
9.80
79.0
79.0
72.9
72.4
1.08
1.09
A-20a
A-20b
5310
5000
15.68
15.58
67.7
63.5
63.0
62.0
1.07
1.02
B-20a
B-20b
2990
3620
15.75
15.60
57.5
62.0
53.0
57.0
1.08
1.09
C-20a
C-20b
1630
1600
15.60
15.72
47.0
47.0
45.8
44.9
1.03
1.05
t. :=: 43.8 ksi
Avg P1es1/Pcalc for all 30 columns:=: 1.05

Standard deviation == 0.060
source: E. Hognestad.
TABLE A-ll- RECTANGULAR COLUMNS SUBJECT TO AXIAL THRUST

AND BENDING ABOUT TWO AXES*
Column
No.
b,
in.
r,
d',
in.
in.
SC1
SC6
0.75
SC2
SC7
SC3
sea
SC4
SC9
SC5
SClO
fc',
psi
ez',
in.
ev',
in.
Pte. I,
Pcalo,
PteH
psi
kip
kip
Peale
4-1/4
diameter
35.8
5435
3.14
3.14
5.27
4.68
5.10
1.03
0.92
0.75
4-5/16
diameter
39.2
5435
3.07
3.07
7.44
7.25
7.46
1.00
0.97
0.75
4-3/8
diameter
40.2
5435
3.06
3.05
9.50
9.81
9.13
l.D4
1.07
0.75
4-1/2
diameter
45.7
5435
3.03
3.03
13.50
14.30
13.88
0.97
1.03
0.75
4-5/8
diameter
40.7
5435
3.02
3.02
16.48
16.50
17.34
0.95
0.95
- - -1 -
Reinforcement
Avg P1eai/Pcalc :=: 0.99

Standard deviation== 0.046
source: P. Anderson and H. N. Lee.1

J. Aci PDF

Uploaded by

Copyright:

Available Formats

J. Aci PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

J. Aci PDF

Uploaded by

Copyright:

Available Formats

Title No.

Rectangular Concrete Stress

Review of Basic Assumptions

JOURNAL OF THE AMERICAN CONCRETE INSTITUTE

At the request of Subcommittee 15 on ultimate strength design of

By use of the following general assumptions, ultimate strength of

USE OF SIMPLIFIED EQUATIONS

The equations for ultimate strength design given in the Appendix

for the design of columns contain additional simplifying assumptions,

ULTIMATE STRENGTH DESIGN

JOURNAL OF THE AMERICAN CONCRETE INSTITUTE

net concrete cross section, or

center of action of concrete

ULTIMATE STRENGTH DESIGN

= inclination of neutral axis to

tensile steel strain at ultimate

CONCRETE STRESS DISTRIBUTION

JOURNAL OF THE AMERICAN CONCRETE INSTITUTE

Mattock from beams

Cylinder Strength in psi

Many different distributions of stress in the concrete compression

ULTIMATE STRENGTH DESIGN

Several papers on stress distribution and ultimate strength design

JOURNAL OF THE AMERICAN CONCRETE INSTITUTE

6-in. SR-4 Gage

Fig. 3-PCA eccentric load specimen

face at ultimate strength versus applied eccentricity, the eccentricity

ULTIMATE STRENGTH DESIGN

symmetrical bending is reached, then the concrete stress distribution

o- P.C.A. Sand ~ Grav~--

Cylinder Strength in psi

Fig. 4-Properties of concrete stress distribution at ultimate strength determined

JOURNAL OF THE AMERICAN CONCRETE INSTITUTE

EQUIVALENT RECTANGULAR STRESS DISTRIBUTION

does exist in the concrete at ultimate strength, in a normally proper-

ULTIMATE STRENGTH DESIGN

Compressive Strength, psi

o - PCA Sand ~ Gravel

Concrete Prism Strength, f~, psi

Fig. 6b-Uitimate strain from tests of plain concrete specimen'

JOURNAL OF THE AMERICAN CONCRETE INSTITUTE

a gage length including several cracks, we find that this "apparent"

Beams using structural or intermediate

Symmetrical reinforced concrete

Unsymmetrical reinforced concrete

Axial loading plus

Axial loading plus

ULTIMATE STRENGTH DESIGN

crete. The second assumption is therefore sufficiently accurate for the

SUMMARY OF COMPARISONS WITH EXPERIMENTS

JOURNAL OF THE AMERICAN CONCRETE INSTITUTE

applicability to the calculation of the ultimate strength of structural

2. There is excellent agreement between the ultimate strengths of a wide

It is concluded, therefore, that the proposed method of ultimate

DESIGN OF RECTANGULAR BEAMS IN BENDING

Ultimate strength controlled by yielding of reinforcement-In this

From equilibrium of internal and external moments

Solving Eq. (1.1) and (1.2) and substituting q

2k2 , we may write