MSC Nastran Implicit Nonlinear (SOL 600) User’s GuideMSC Nastran Verifica-

tion Guide

MSC Nastran 2021.4

Verification Guide
Corporate Europe, Middle East, Africa
5161 California Ave. Suite 200 Am Moosfeld 13
University Research Park 81829 Munich, Germany
Irvine, CA 92617 Telephone: (49) 89 431 98 70
Telephone: (714) 540-8900 Email: europe@mscsoftware.com
Email: americas.contact@mscsoftware.com

Japan Asia-Pacific
KANDA SQUARE 16F 100 Beach Road
2-2-1 Kanda Nishikicho, Chiyoda-ku #16-05 Shaw Tower
1-Chome, Shinjuku-Ku Singapore 189702
Tokyo 101-0054, Japan Telephone: 65-6272-0082
Telephone: (81)(3) 6275 0870 Email: APAC.Contact@mscsoftware.com
Email: MSCJ.Market@mscsoftware.com

Worldwide Web
www.mscsoftware.com, www.hexagon.com

Support
https://simcompanion.hexagon.com/customers/s/article/support-contact-information-kb8019304

Disclaimer
Hexagon reserves the right to make changes in specifications and other information contained in this document without prior notice.
The concepts, methods, and examples presented in this text are for illustrative and educational purposes only, and are not intended
to be exhaustive or to apply to any particular engineering problem or design. Hexagon assumes no liability or responsibility to any
person or company for direct or indirect damages resulting from the use of any information contained herein.
User Documentation: Copyright 2021 Hexagon AB and/or its subsidiaries. Printed in U.S.A. All Rights Reserved.
This notice shall be marked on any reproduction of this documentation, in whole or in part. Any reproduction or distribution of this
document, in whole or in part, without the prior written consent of Hexagon is prohibited.
This software may contain certain third-party software that is protected by copyright and licensed from Hexagon suppliers. Additional
terms and conditions and/or notices may apply for certain third party software. Such additional third party software terms and
conditions and/or notices may be set forth in documentation and/or at http://www.mscsoftware.com/thirdpartysoftware (or successor
website designated by Hexagon from time to time).
PCGLSS 8.0, Copyright © 1992-2016, Computational Applications and System Integration Inc. All rights reserved. PCGLSS 8.0 is
licensed from Computational Applications and System Integration Inc.
MSC, Dytran, Marc, MSC Nastran, Patran, Hexagon logo, MSC Software logo, e-Xstream, Digimat, and Simulating Reality are
trademarks or registered trademarks of Hexagon AB and/or its subsidiaries in the United States and/or other countries.
NASTRAN is a registered trademark of NASA. FLEXlm and FlexNet Publisher are trademarks or registered trademarks of Flexera
Software. All other trademarks are the property of their respective owners.
Use, duplicate, or disclosure by the U.S. Government is subjected to restrictions as set forth in FAR 12.212 (Commercial Computer
Software) and DFARS 227.7202 (Commercial Computer Software and Commercial Computer Software Documentation), as
applicable.
U.S. Patent 9,361,413
December 3, 2021
NA:V2021.4:Z:Z:Z:DC-VER-PDF
Documentation Feedback
At MSC Software, we strive to produce the highest quality documentation and welcome your feedback.
If you have comments or suggestions about our documentation, write to us at: documentation-
feedback@mscsoftware.com.
Please include the following information with your feedback:
 Document name
 Release/Version number
 Chapter/Section name
 Topic title (for Online Help)
 Brief description of the content (for example, incomplete/incorrect information, grammatical
errors, information that requires clarification or more details and so on).
 Your suggestions for correcting/improving documentation
You may also provide your feedback about MSC Software documentation by taking a short 5-minute
survey at: http://msc-documentation.questionpro.com.

Note: The above mentioned e-mail address is only for providing documentation specific
feedback. If you have any technical problems, issues, or queries, please contact Technical
Support.
 Contents
MSC Nastran Verification Guide

Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Section 1: MacNeal-Harder Problems

1.1. Straight Cantilever Beam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.2. Curved Beam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.3. Patch Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.4. Rectangular Plate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
1.5. Scordelis-Lo-Roof . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
1.6. Spherical Shell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
1.7. Thick Wall Cylinder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
1.8. Twisted Beam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

Section 2: Original Verification Manual Problems

2.1. Static Analysis of a Pin-Jointed Truss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
2.2. Heat Transfer from Cooling Fin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
2.3. Truss Analysis using Thermal Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
2.4. Inertia Load on Pilot During Landing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
2.5. Cantilever Beam, Statics, Checkpoint . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
2.6. Spring Supported Bar in a Skewed Coordinate System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
2.7. Punched Output of a Static Solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
2.8. Cantilever Beam Constructed of Plates, Stress Contours . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
2.9. Bar with Offsets in a Skewed Coordinate System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
2.10. Thick Walled Cylinder, Internal Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
2.11. Cantilevered Cylindrical Shell, Axisymmetric Force. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
2.12. Static Analysis of a Pin-Joined Truss Using Superelements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
2.13. Beam with Point Masses, Enforced Acceleration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
2.14. Two Degree of Freedom Modal Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
2.15. Normal Modes of a Beam Structure using Component Modal Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

Main Index
 2 MSC Nastran Verification Guide

2.16. Torsional Vibration of a Shaft with Three Disks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

2.17. Cantilever Beam, Normal Modes, Restart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
2.18. Normal Modes Analysis with Reduction Techniques. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
2.19. Modes of Fixed Circular Plate with Trapezoidal Ring Element . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
2.20. Two Degree of Freedom Modal Analysis Using Superelements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
2.21. Modes of Fixed Circular Plate with Superelement Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
2.22. Natural Frequencies of a Circular Plate Immersed in Fluid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
2.23. DDAM Method of Response Spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
2.24. Lateral Buckling of Cantilever Beam. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
2.25. Simple Frame Analysis with Buckling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
2.26. Beam with Gap/Lift-Off . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
2.27. Cantilever Beam with Plastic Hinge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
2.28. Complex Eigenvalues of Three-by-Three Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
2.29. Complex Roots as a Function of Gain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
2.30. Frequency Response of a Single Degree of Freedom System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
2.31. Transient Response of a Rocket - Direct Time Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
2.32. Response Spectra of a Single Oscillator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
2.33. Transient Response with Combined Loading Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
2.34. Direct Transient Response to an Internal Pressure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
2.35. Complex Eigenvalues of Three-by-Three Matrix, Modal Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
2.36. Modal Frequency Response of a Beam Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
2.37. Transient Response of a Beam Modal Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
2.38. Deflection of a Circular Plate, Cyclic Symmetry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
2.39. Normal Modes of a Circular Plate, Cyclic Symmetry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
2.40. Euler Buckling of a Simply Supported Beam, Cyclic Symmetry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
2.41. Frequency Response of a Beam-Mass Structure using Dihedral Cyclic Symmetry . . . . . . . . . . . . . . . . . . . . . . . . . 155
2.42. Stress Wave Propagation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
2.43. Impact Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
2.44. Flutter of a Two-Dimensional Airfoil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
2.45. Radiation Exchange Between Two Panels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
2.46. Transient Temperature Distribution in a Slab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
2.47. Transient Temperature Distribution in a Slab with Convective Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
2.48. Large Deflection of Uniformly Loaded Plate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
2.49. Large Deflection of a Beam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181

Main Index
 Contents 3

Section 3: Rotordynamics Problems

3.1 Flexible Rotor Critical Speed and Mode Shape. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .187
3.2. Jeffcott Rotor Eigenvalues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .192
3.3. Complex Eigenvalue solution for Jeffcott rotor with multiple disks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .196
3.4. Complex Eigenvalue Solution for Jeffcott Rotor with Overhung Disk and Anisotropic Bearings. . . . . . . . . . . . . . . . .200

Section 4: SOL 200 Optimization Problems

4.1. Optimization of a Three Bar Truss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .207
4.2. Design with Constraint on Minimum Frequency. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .210
4.3. Shape Design of a Culvert . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .213
4.4. Topology Design of the MBB Beam. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .216

Section 5: SOL 700 Explicit Dynamics Problems

5.1. Impulsively Loaded Strip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .221
5.2. Basic Blast Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .224
5.3. JWL Blast Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .227
5.4. Taylor Test (a rod impacted against a rigid wall). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .230

Section 6: NAFEMS Problems

Linear Elastic

6.1. LE1: Elliptic Membrane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .240

6.2. LE2: Cylindrical Shell Patch Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .243
6.3. LE3: Hemisphere-Point Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .246
6.4. LE5: Z-section Cantilever. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .248
6.5. LE6: Skew Plate Normal Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .251
6.6. LE7: Axisymmetric Cylinder/Sphere-Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .253
6.7. LE8: Axisymmetric Shell Pressure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .255
6.8. LE9: Axisymmetric Branched Shell-Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .257
6.9. LE10: Thick Plate Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .259
6.10. LE11: Solid Cylinder/ Taper/ Sphere-Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .262
Linear Thermo Elastic

6.11. T1: Membrane with Hot Spot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .265

Main Index
 4 MSC Nastran Verification Guide

Heat Transfer

6.12. T2: One Dimensional Heat Transfer with Radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269

6.13. T3: One Dimensional Transient Heat Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
6.14. T4: Two Dimensional Heat Transfer with Convection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
Free Vibration

6.15. FV2- Pin-Ended Double Cross: In-Plane Vibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276

6.16. FV4 - Cantilever with Off-Centre Point Masses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
6.17. FV5 - Deep Simply-supported Beam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
6.18. FV12- Free Thin Square Plate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
6.19. FV15- Clamped Thin Rhombic Plate. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
6.20. FV16- Cantilevered Thin Plate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
6.21. FV22 - Clamped Thick Rhombic Plate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
6.22. FV32 - Cantilevered Tapered Membrane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
6.23. FV41 - Axisymmetric Vibration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
6.24. FV42 - Thick Hollow Sphere . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
6.25. FV52 - Simply Supported Solid Square Plate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305
6.26. FV73 - Cantilevered Thin Square Plate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
Forced Vibration

6.27. Test 5 - Deep Simply Supported Beam. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311

6.28. Test 5H - Deep Simply Supported Beam Harmonic Forced Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
6.29. Test 5P - Deep Simply Supported Beam Periodic Forced Vibration Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315
6.30. Test 5T-Deep Simply Supported Beam Transient Forced Vibration Response. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318
6.31. Test 5R-Deep Simply Supported Beam Random Forced Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321
6.32. Test 13: Simply Supported Thin Square Plate: Frequency Extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
6.33. Test 13H: Simply Supported Thin Square Plate: Harmonic Forced Vibration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 328
6.34. Test 13P: Simply Supported Thin Square Plate: Periodic Forced Vibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332
6.35. Test 13T: Simply Supported Thin Square Plate: Transient Forced Vibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336
6.36. Test 13R: Simply Supported Thin Square Plate: Random Forced Vibration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340
6.37. Test 21: Simply Supported Thick Square Plate: Frequency Extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343
6.38. Test 21H: Simply Supported Thick Square Plate: Harmonic Forced Vibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 346
6.39. Test 21P: Simply Supported Thick Square Plate: Periodic Forced Vibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
6.40. Test 21T: Simply Supported Thick Square Plate: Transient Forced Vibration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353

Main Index
 Contents 5

6.41. Test 21R: Simply Supported Thick Square Plate: Random Forced Vibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .358
Nonlinear

6.42. NL4: Snap-back Under Displacement Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .362

6.43. NL6: Straight Cantilever with Axial End Point Load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .366
6.44. NL7: Lee's Frame Buckling Problem. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .369
Fracture Mechanics

6.45. Test 1.1: Center Cracked Plate in Tension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .373

6.46. Test 1.2: Center Cracked Plate with Thermal Load. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .376
6.47. Test 2.1: Single Edge Cracked Plate in Tension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .379
6.48. Test 3: Angle Crack Embedded in a Plate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .382
6.49. Test 4: Cracks at a Hole in a Plate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .385
6.50. Test 5: Axisymmetric Crack in a Bar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .388
6.51. Test 6: Compact Tension Specimen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .391
6.52. Test 7: T-Joint Weld Attachment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .394
6.53. Test 8: V-Notch Specimen in Tension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .397
Creep

6.54. Test 1A: 2-D Plane Stress - Uniaxial Load, Secondary Creep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .401
6.55. Test 1B: 2-D Plane Stress - Uniaxial Displacement, Secondary Creep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .405
6.56. Test 2A: 2-D Plane Stress - Biaxial Load, Secondary Creep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .409
6.57. Test 2B: 2-D Plane Stress - Biaxial Displacement, Secondary Creep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .412
6.58. Test 3B: 2-D Plane Stress - Biaxial (-ve) Displacement, Secondary Creep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .416
6.59. Test 8A: 2D Plane Stress - Uniaxial Load, Primary Creep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .420
6.60. Test 8B: 2-D Plane Stress - Uniaxial Displacement, Primary Creep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .423
6.61. Test 8C: 2D Plane Stress - Stepped Load Primary Creep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .426
6.62. Test 9A: 2-D Plane Stress - Biaxial Load, Primary Creep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .429
6.63. Test 9B: 2-D Plane Stress - Biaxial Displacement, Primary Creep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .432
6.64. Test 9C: 2-D Plane Stress - Biaxial Stepped Load, Primary Creep. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .435
Composite

6.65. R0031(1): Laminated Strip under Three-point Bending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .439

6.66. R0031(2): Wrapped Thick Cylinder under Pressure and Thermal Loading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .441
6.67. R0031 (3): Three-layer Sandwich Shell under Normal Pressure Loading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .444

Main Index
 6 MSC Nastran Verification Guide

Contact

6.68. R0094 (1): Benchmark 1: 2D Cylinder Roller Contact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 447

6.69. R0094 (2): Benchmark 2: 3D Punch (Rounded Edges). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450
6.70. R0094 (3): Benchmark 3: 3D Sheet Metal Forming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453
6.71. R0094 (4): Benchmark 4: 3D Loaded Pin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457
6.72. R0094 (5): Benchmark 5: 3D Steel Roller on Rubber. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 460
Geometric Nonlinear

6.73. 3DNLG1: Elastic Large Deflection Response of A Z-Shaped Cantilever under End Load . . . . . . . . . . . . . . . . . . . . . 465
6.74. 3DNLG2: Elastic Large Deflection Response of a Pear-Shaped Cylinder under End Shortening. . . . . . . . . . . . . . . . 468
6.75. 3DNLG6: Buckling of a Flat Plate When Subjected To In-Plane Shear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 471
6.76. 3DNLG7: Elastic Large Deflection Response of a Hinged Spherical Shell Under Pressure Loading . . . . . . . . . . . . . 474
6.77. 3DNLG9: Large Elastic Deflection of a Pinched Hemispherical Shell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 477
6.78. 3DNLG10: Elastic-Plastic Behavior of a Stiffened Cylindrical Panel under Compressive End Load. . . . . . . . . . . . . . 480

Main Index
 Preface


About this Guide 14

List of MSC Nastran Guides 15

Technical Support 16

Training and Internet Resources 16

Main Index
 14 MSC Nastran Verification Guide

About this Guide

The last update of the Verification Manual for MSC Nastran was in 1985. In the years since then, many
features have been added to the program, but the manual has not been updated. This version is an initial
update to bring the manual more up to date. The intent is for this to be a 'living document', with additional
problems added with each new release of MSC Nastran.
This manual not only contains the original problems from the Verification Manual, but it also contains many
additional problems.
The requirement for a problem to be included in this manual is that there must exist either a theoretical
solution, a solution accepted in published literature, or a published set of test results. MSC tests over 29,000
problems in MSC Nastran daily during the development process and this manual contains a small subset of
those problems.
A section named “MacNeal-Harder Problems” is added. This section contains the set of test problems
proposed by Dr. Richard MacNeal and Bob Harder in 1985. These are problems intended to test how well
the elements perform in varying situations which are known to be difficult for finite elements to handle
properly.
Three sections on newer features in MSC Nastran are added. These are Rotordynamics, SOL 200
Optimization, and SOL 700 Explicit Dynamics. It turns out that it is difficult to find documented examples
in literature with theoretical results or experimental results and a complete description of the structure/model
used, so these sections each have a small set of problems. Many more examples can be found in the associated
User Guides.
The final new section is called NAFEMS. This section contains many of the benchmark problems defined
by NAFEMS (https://www.nafems.org/).
A standard format is followed for all problems:
 Problem Description: a short description of the problem being solved.
 Solution Number: The solution used in MSC Nastran to run the problem
 Features Used: a short list of features used in solving the problem
 Reference: the reference used as a baseline for the problem
 Modeling Techniques Used: a short description of the problem and properties/features used
 Illustrations of the Model: one or more illustrations of the structure and/or finite element model
 Model File Name and Location: the files are delivered with the documentation in a directory called
'tpl' (test problem library). This provides the name(s) of the file(s) used and the directory where they
can be found.
 Results: a short presentation (usually in tabular form) of the results
 Conclusion: A summary of the findings

Main Index
 MSC Nastran Verification Guide 15
Preface

List of MSC Nastran Guides

A list of some of the MSC Nastran guides is as follows:

Installation and Release Guides

 Installation and Operations Guide
 Release Guide
Reference Guides
 Quick Reference Guide
 DMAP Programmer’s Guide
 Reference Guide
 Utilities Guide
Demonstration Guides
 Linear Analysis
 Implicit Nonlinear (SOL 400)
 Explicit Nonlinear (SOL 700)
 MSC Nastran Verification Guide
User’s Guides
 Getting Started
 Linear Static Analysis
 Dynamic Analysis
 Embedded Fatigue
 Embedded Vibration Fatigue
 Thermal Analysis
 Superelements and Modules
 Design Sensitivity and Optimization
 Rotordynamics
 Implicit Nonlinear (SOL 400)
 Explicit Nonlinear (SOL 700)
 Aeroelastic Analysis
 User Defined Services
 Non Linear (SOL 600)
 High Performance Computing
 DEMATD

Main Index
 16 MSC Nastran Verification Guide

You may find any of these documents from MSC Software at:
https://simcompanion.hexagon.com

Technical Support
For technical support phone numbers and contact information, please visit:
https://simcompanion.hexagon.com/customers/s/article/support-contact-information-kb8019304
Support Center (http://simcompanion.hexagon.com)
The SimCompanion link above gives you access to the wealth of resources for MSC Software products. Here
you will find product and support contact information, product documentations, knowledge base articles,
product error list, knowledge base articles and SimAcademy Webinars. It is a searchable database which allows
you to find articles relevant to your inquiry. Valid MSC customer entitlement and login is required to access
the database and documents. It is a single sign-on that gives you access to product documentation for
complete list of products from MSC Software, allows you to manage your support cases, and participate in
our discussion forums.

Training and Internet Resources

MSC Software (www.mscsoftware.com)

MSC Software corporate site with information on the latest events, products and services for the
CAD/CAE/CAM marketplace.
http://simcompanion.hexagon.com

The SimCompanion link above gives you access to the wealth of resources for MSC Software products. Here
you will find product and support contact information, product documentations, knowledge base articles,
product error list, knowledge base articles and SimAcademy Webinars. It is a searchable database which allows
you to find articles relevant to your inquiry. Valid MSC customer entitlement and login is required to access
the database and documents. It is a single sign-on that gives you access to product documentation for
complete list of products from MSC Software, allows you to manage your support cases, and participate in
our discussion forums.
http://www.mscsoftware.com/msc-training

The MSC Training link above will point you to schedule and description of MSC Seminars. Following
courses are recommended for beginning MSC Nastran users.
NAS120 - Linear Static Analysis using MSC Nastran and Patran

This seminar introduces basic finite element analysis techniques for linear static, normal modes, and buckling
analysis of structures using MSC Nastran and Patran. MSC Nastran data structure, the element library,
modeling practices, model validation, and guidelines for efficient solutions are discussed and illustrated with
examples and workshops. Patran will be an integral part of the examples and workshops and will be used to
generate and verify illustrative MSC Nastran models, manage analysis submission requests, and visualize

Main Index
 MSC Nastran Verification Guide 17
Preface

results. This seminar provides the foundation required for intermediate and advanced MSC Nastran
applications.

Main Index
 18 MSC Nastran Verification Guide

Main Index
 Section 1: MacNeal-Harder Problems

Main Index
Main Index
 1.1 Straight Cantilever Beam

Problem Description
This problem demonstrates the use of different elements to model a cantilevered beam with a rectangular
cross section subjected to unit forces at free end. The solution is compared with the analytical solution.

Solution Number
SOL 101

Features Used
 Linear Statics.

Reference
1. Macneal., R.H., Harder, R.L., 1985. A proposed standard set of problems to test finite element
accuracy. Finite Elem. Anal. Des. 1, 3-20.

Modeling Techniques Used

The model is analyzed using different types of elements such as HEXA, PENTA, QUAD, QUADR and
TRIA elements. The model details are given below:

Material Properties
The material is linear elastic with a Young's modulus, E = 1.0 × 107 and Poisson's ratio, = 0.30.

Geometrical properties
Length = 6.0
Width = 0.2
Depth = 0.1

Loading
A force of one unit applied at the free end.

Main Index
 10 MSC Nastran Verification Guide

Boundary Condition
The displacements ux = uy= uz = 0.0 at one end of the beam as shown in Figure 1.1-1.

Illustrations of the Model

Figure 1.1-1 Straight cantilever beam. (a) Regular shape elements (b) Trapezoidal shape elements
(c) Parallelogram shape elements

Main Index
 CHAPTER 1.1 11
Straight Cantilever Beam

Model File Name and Location

Model File Name Description Location

macneal013 Straight cantilever beam, HEXA8 elements tpl\verifman\macneal
macneal014 Straight cantilever beam, HEXA20 elements tpl\verifman\macneal
macneal015 Straight cantilever beam, PENTA elements tpl\verifman\macneal
macneal016 Straight cantilever beam, QUAD4 elements tpl\verifman\macneal
macneal017 Straight cantilever beam, QUAD8 elements tpl\verifman\macneal
macneal018 Straight cantilever beam, QUADR elements tpl\verifman\macneal
macneal019 Straight cantilever beam, TRIA elements tpl\verifman\macneal

Results
Rectangular Elements

Tip Displacement in Direction of Load

Tip
Loading Theoretical
Direction [1] HEXA8 HEXA20 PENTA QUAD4 QUAD8 QUADR TRIAR
Extension 3.0 × 10-5 2.9 × 10-5 3.0 × 10-5 3.0 × 10-5 3.0 × 10-5 3.0 × 10-5 3.0 × 10-5 3.0 × 10-5
In plane 0.1081 0.10599 0.10543 0.10542 0.10732 0.10667 0.10732 0.02559
shear
Out-of- 0.4321 0.42374 0.41901 0.41678 0.42649 0.42837 0.42542 0.42114
plane shear
Twist 0.03208 0.02917 0.02922 0.02963 0.03059 0.03046 0.03036 0.04238

Trapezoidal Elements

Tip Displacement in Direction of Load

Tip
Loading Theoretical
Direction [1] HEXA8 HEXA20 PENTA QUAD4 QUAD8 QUADR TRIAR
Extension 3.0 × 10-5 2.9 × 10-5 2.9 × 10-5 2.9 × 10-5 2.9 × 10-5 2.9 × 10-5 3.0 × 10-5 2.9 × 10-5
In plane 0.1081 0.00463 0.09701 0.04221 0.00569 0.10222 0.10449 0.02405
shear
Out-of- 0.4321 0.12793 0.41534 0.26323 0.42389 0.43117 0.41735 0.41813
plane shear
Twist 0.03208 0.02849 0.02930 0.28351 0.03085 0.03642 0.02951 0.04094

Main Index
 12 MSC Nastran Verification Guide

Parallelogram Elements

Tip Displacement in Direction of Load

Tip
Loading Theoretical
Direction [1] HEXA8 HEXA20 PENTA QUAD4 QUAD8 QUADR TRIAR
Extension 3.0 × 10-5 2.9 × 10-5 2.9 × 10-5 2.9 × 10-5 2.9 × 10-5 2.9 × 10-5 3.0 × 10-5 2.9 × 10-5
In plane 0.1081 0.02924 0.10575 0.08601 0.06834 0.10758 0.01046 0.03742
shear
Out-of- 0.4321 0.24852 0.41492 0.37517 0.42874 0.42552 0.42163 0.42051
plane shear
Twist 0.03208 0.026731 0.02923 0.028937 0.30652 0.03577 0.02820 0.043161

Conclusion
From the above observations, the rectangular element results from MSC Nastran are in better agreement with
analytical results than trapezoidal and parallelogram elements.

Main Index
 CHAPTER 1.2 13
Curved Beam

1.2 Curved Beam

Problem Description
This problem demonstrates the use of different elements to model a 90-degree section of a circular beam.
The beam is analyzed with two types of loading conditions:
a. loaded in its plane in a radial direction on its free end, and
b. loaded in out of plane direction on its free end.
The solution is compared to the analytical solution.

Solution Number
SOL 101

Features Used
 Linear Statics.

Reference
1. Macneal., R.H., Harder, R.L., 1985. A proposed standard set of problems to test finite element
accuracy. Finite Elem. Anal. Des. 1, 3-20.

Modeling Techniques Used

The model is analyzed using different types of elements such as HEXA, PENTA, QUAD, QUADR and
TRIAR elements. The model details are given below:

Material Properties
The material is linear elastic with a Young's modulus, E = 1.0 × 107 and Poisson's ratio  = 0.25.

Geometrical properties
Inner radius = 4.12
Outer radius=4.32
Thickness=0.1

Main Index
 14 MSC Nastran Verification Guide

Loading
A force of one unit applied at the tip shown in Figure 1.2-1.

Boundary Condition
The displacements ux = uy= uz = 0.0 at one end of the beam as shown in Figure 1.2-1.

Illustrations of the Model

a. Vertical Loading b. Out-of-plane Loading

Figure 1.2-1 Curved Beam

Model File Name and Location

Model File Name Description Location

macneal001 Curved beam, HEXA8 elements tpl\verifman\macneal
macneal002 Curved beam, PENTA elements tpl\verifman\macneal
macneal003 Curved beam, QUAD4 elements tpl\verifman\macneal
macneal004 Curved beam, QUAD8 elements tpl\verifman\macneal
macneal005 Curved beam, QUADR elements tpl\verifman\macneal
macneal006 Curved beam, TRIAR elements tpl\verifman\macneal

Main Index
 CHAPTER 1.2 15
Curved Beam

Results

Tip Displacement in Direction of Load

Tip Loading Theoretical
Direction [1] HEXA8 PENTA QUAD4 QUAD8 QUADR TRIAR
In plane 0.08734 0.07688 0.08191 0.077517 0.08797 0.08826 0.017425
(vertical)
Out-of-plane 0.5022 0.41940 0.46258 0.48018 0.47972 0.47069 0.500471

Conclusion
The results from MSC Nastran are in a good agreement with analytical results.

Main Index
 16 MSC Nastran Verification Guide

1.3 Patch Test

Problem Description
This problem contains basic patch test cases for Nastran elements. The solution is compared to the analytical
solution.

Solution Number
SOL 101

Features Used
 Linear Statics.

Reference
1. Macneal., R.H., Harder, R.L., 1985. A proposed standard set of problems to test finite element
accuracy. Finite Elem. Anal. Des. 1, 3-20.

Modeling Techniques Used

For the patch test, elements such as HEXA, QUAD, QUADR and TRIAR are used. The model details are
given below:

Material Properties
The material is elastic with a Young's modulus, E = 1.0 × 107 and Poisson's ratio = 0.25.

Geometrical properties
Membrane Plate
thickness=0.001
length=0.24
width=0.12
Solid Cube
A cube of unit length is used as shown in Figure 1.3-1.

Boundary Condition
Displacement boundary conditions at all exterior nodes for constant stress loading are given below:

Main Index
 CHAPTER 1.3 17
Patch Test

Membrane Plate
u = 10-3 (x+y/2)
v = 10-3 (y+x/2)
Solid Cube
u = 10-3 (2x+y+z)/2
v = 10-3 (x+2y+z)/2
w = 10-3 (x+y+2z)/2

Illustrations of the Model

a. plates b. solids

Figure 1.3-1 Patch Test Models

Main Index
 18 MSC Nastran Verification Guide

Model File Name and Location

Model File Name Description Location

macneal026 Patchtest, HEXA8 elements tpl\verifman\macneal
macneal027 Patchtest, HEXA20 elements tpl\verifman\macneal
macneal028 Patchtest, HEXA2R elements tpl\verifman\macneal
macneal029 Patchtest, QUAD4 elements tpl\verifman\macneal
macneal030 Patchtest, QUAD8 elements tpl\verifman\macneal
macneal031 Patchtest, QUADR elements tpl\verifman\macneal
macneal032 Patchtest, TRIAR elements tpl\verifman\macneal

Results
Theoretical Results

Membrane Plate Patch Test

x = y = 1333, xy = 400

Solid Patch Test

x = y = z = 2000, xy = yz = zx = 400

Maximum Error in Stress

HEXA8 HEXA20 HEXA20R QUAD4 QUAD8 QUADR TRIAR
Constant stress 0 0 0 0 15% 0 0
loading

Conclusion
All elements show exact solutions except for the QUAD8 which differ from the analytical solution by about
15%.

Main Index
 CHAPTER 1.4 19
Rectangular Plate

1.4 Rectangular Plate

Problem Description
This problem demonstrates the use of different elements to model a rectangular plate. The plate is analyzed
with two types of loading conditions:
a. uniform pressure, and
b. concentrated central load.
The solution is compared to the analytical solution.

Solution Number
SOL 101

Features Used
 Linear Statics.

Reference
1. Macneal., R.H., Harder, R.L., 1985. A proposed standard set of problems to test finite element
accuracy. Finite Elem. Anal. Des. 1, 3-20.

Modeling Techniques Used

The model is analyzed using different types of elements such as HEXA, PENTA, QUAD, QUADR and
TRIAR elements. The model details are given below:

Material Properties
The material is elastic with a Young's modulus, E = 1.7472 × 107 and Poisson's ratio = 0.3.

Geometrical properties
Two types of plate geometry are used for the analysis.
1. Square plate: a=2.0
2. Rectangular plate: a=2.0, b=10.0
3. Thickness= 0.0001(Shell element) and 0.01 (Solid element).

Main Index
 20 MSC Nastran Verification Guide

Loading
Two types of loadings are used for the analysis
1. Uniform pressure= 10-4
2. Central load = 4.0 × 10-4

Boundary Condition
The model is analyzed using two types of boundary conditions
1. Simply supported
2. Clamped

Main Index
 CHAPTER 1.4 21
Rectangular Plate

Illustrations of the Model

(b)
Figure 1.4-1 (a) Plate with Aspect Ratio 1.0 (b) Plate with Aspect Ratio 5.0

Main Index
 22 MSC Nastran Verification Guide

Model File Name and Location

Model File Name Description Location

macneal033 Rectangular Plate, HEX8 element, AR*=1, N*=2 tpl\verifman\macneal
macneal034 Rectangular Plate, HEX8 element, AR=1, N=4 tpl\verifman\macneal
macneal035 Rectangular Plate, HEX8 element, AR=1, N=6 tpl\verifman\macneal
macneal036 Rectangular Plate, HEX8 element, AR=1, N=8 tpl\verifman\macneal
macneal037 Rectangular Plate, HEX8 element, AR=5, N=2 tpl\verifman\macneal
macneal038 Rectangular Plate, HEX8 element, AR=5, N=4 tpl\verifman\macneal
macneal039 Rectangular Plate, HEX8 element, AR=5, N=6 tpl\verifman\macneal
macneal040 Rectangular Plate, HEX8 element, AR=5, N=8 tpl\verifman\macneal

macneal041 Rectangular Plate, PENTA element, AR=1, N=2 tpl\verifman\macneal

macneal042 Rectangular Plate, PENTA element, AR=1, N=4 tpl\verifman\macneal
macneal043 Rectangular Plate, PENTA element, AR=1, N=6 tpl\verifman\macneal
macneal044 Rectangular Plate, PENTA element, AR=1, N=8 tpl\verifman\macneal
macneal045 Rectangular Plate, PENTA element, AR=5, N=2 tpl\verifman\macneal
macneal046 Rectangular Plate, PENTA element, AR=5, N=4 tpl\verifman\macneal
macneal047 Rectangular Plate, PENTA element, AR=5, N=6 tpl\verifman\macneal
macneal048 Rectangular Plate, PENTA element, AR=5, N=8 tpl\verifman\macneal

macneal049 Rectangular Plate, QUAD4 element, AR=1, N=2 tpl\verifman\macneal

macneal050 Rectangular Plate, QUAD4 element, AR=1, N=4 tpl\verifman\macneal
macneal051 Rectangular Plate, QUAD4 element, AR=1, N=6 tpl\verifman\macneal
macneal052 Rectangular Plate, QUAD4 element, AR=1, N=8 tpl\verifman\macneal
macneal053 Rectangular Plate, QUAD4 element, AR=5, N=2 tpl\verifman\macneal
macneal054 Rectangular Plate, QUAD4 element, AR=5, N=4 tpl\verifman\macneal
macneal055 Rectangular Plate, QUAD4 element, AR=5, N=6 tpl\verifman\macneal
macneal056 Rectangular Plate, QUAD4 element, AR=5, N=8 tpl\verifman\macneal

macneal057 Rectangular Plate, QUAD8 element, AR=1, N=2 tpl\verifman\macneal

macneal058 Rectangular Plate, QUAD8 element, AR=1, N=4 tpl\verifman\macneal
macneal059 Rectangular Plate, QUAD8 element, AR=1, N=6 tpl\verifman\macneal
macneal060 Rectangular Plate, QUAD8 element, AR=1, N=8 tpl\verifman\macneal

Main Index
 CHAPTER 1.4 23
Rectangular Plate

Model File Name Description Location

macneal061 Rectangular Plate, QUAD8 element, AR=5, N=2 tpl\verifman\macneal
macneal062 Rectangular Plate, QUAD8 element, AR=5, N=4 tpl\verifman\macneal
macneal063 Rectangular Plate, QUAD8 element, AR=5, N=6 tpl\verifman\macneal
macneal064 Rectangular Plate, QUAD8 element, AR=5, N=8 tpl\verifman\macneal

macneal065 Rectangular Plate, QUADR element, AR=1, N=2 tpl\verifman\macneal

macneal066 Rectangular Plate, QUADR element, AR=1, N=4 tpl\verifman\macneal
macneal067 Rectangular Plate, QUADR element, AR=1, N=6 tpl\verifman\macneal
macneal068 Rectangular Plate, QUADR element, AR=1, N=8 tpl\verifman\macneal
macneal069 Rectangular Plate, QUADR element, AR=5, N=2 tpl\verifman\macneal
macneal070 Rectangular Plate, QUADR element, AR=5, N=4 tpl\verifman\macneal
macneal071 Rectangular Plate, QUADR element, AR=5, N=6 tpl\verifman\macneal
macneal072 Rectangular Plate, QUADR element, AR=5, N=8 tpl\verifman\macneal

macneal073 Rectangular Plate, TRIAR element, AR=1, N=2 tpl\verifman\macneal

macneal074 Rectangular Plate, TRIAR element, AR=1, N=4 tpl\verifman\macneal
macneal075 Rectangular Plate, TRIAR element, AR=1, N=6 tpl\verifman\macneal
macneal076 Rectangular Plate, TRIAR element, AR=1, N=8 tpl\verifman\macneal
macneal077 Rectangular Plate, TRIAR element, AR=5, N=2 tpl\verifman\macneal
macneal078 Rectangular Plate, TRIAR element, AR=5, N=4 tpl\verifman\macneal
macneal079 Rectangular Plate, TRIAR element, AR=5, N=6 tpl\verifman\macneal
macneal080 Rectangular Plate, TRIAR element, AR=5, N=8 tpl\verifman\macneal

*AR = aspect ratio, N= Number of node spaces per edge of model.

Main Index
 24 MSC Nastran Verification Guide

Results
1. Simply Supported and Uniform Load

a. Aspect Ratio 1.0

Lateral Deflection at the Center

Theoretical
N [1] HEXA8 PENTA QUAD4 QUAD8 QUADR TRIAR
2 4.062(shell) 4.01E-06 2.200 E-06 4.146 3.764 3.986 3.860
4 4.05E-06 3.203 E-06 4.117 4.064 4.078 4.037
4.062E-06
6 (solid) 4.05E-06 3.625 E-06 4.093 4.059 4.076 4.062
8 4.06E-06 3.828 E-06 4.081 4.061 4.071 4.068

b. Aspect Ratio 5.0

Lateral Deflection at the Center

Theoretical
N [1] HEXA8 PENTA QUAD4 QUAD8 QUADR TRIAR
2 12.97(shell) 12.3E-06 12.70E-06 13.04 15.86 13.64 12.873
4 12.6E-06 11.96 E-06 12.84 13.01 12.85 13.039
12.97E-06
6 (solid) 12.8E-06 13.43 E-06 12.91 12.97 12.92 13.018
8 12.9E-06 13.78 E-06 12.94 12.97 12.94 13.004

2. Simply Supported and Concentrated Load

a. Aspect Ratio 1.0

Lateral Deflection at the Center

Theoretical
N [1] HEXA8 PENTA QUAD4 QUAD8 QUADR TRIAR
2 11.60(shell) 11.68E-06 7.375 E-06 12.35 10.98 11.80 10.59
4 11.59E-06 8.171 E-06 11.94 11.50 11.78 11.29
11.60E-06
6 (solid) 11.59E-06 9.614 E-06 11.79 11.56 11.71 11.46
8 11.59E-06 10.39 E-06 11.72 11.58 11.68 11.52

Main Index
 CHAPTER 1.4 25
Rectangular Plate

b. Aspect Ratio 5.0

Lateral Deflection at the Center

Theoretical
N [1] HEXA8 PENTA QUAD4 QUAD8 QUADR TRIAR
2 16.96(shell) 15.31E-06 12.12 E-06 16.25 15.86 13.75 12.35
4 16.04E-06 6.338 E-06 16.65 13.01 15.80 16.00
16.96 E-06
6 (solid) 16.49E-06 7.634 E-06 16.97 12.97 16.50 16.47
8 16.67E-06 8.601 E-06 17.07 12.97 16.77 16.65

3. Clamped Supports and Uniform Load

a. Aspect Ratio 1.0

Lateral Deflection at the Center

Theoretical
N [1] HEXA8 PENTA QUAD4 QUAD8 QUADR TRIAR
2 1.26 (shell) 1.239E-06 2.07 E-08 1.360 1.458 1.269 1.203
4 1.258 E-06 1.74 E-07 1.323 1.225 1.300 1.255
1.26 E-06
6 (solid) 1.262 E-06 6.88 E-07 1.298 1.258 1.288 1.262
8 1.263E-06 9.69 E-07 1.286 1.263 1.280 1.263

b. Aspect Ratio 5.0

Lateral Deflection at the Center

Theoretical
N [1] HEXA8 PENTA QUAD4 QUAD8 QUADR TRIAR
2 2.56 (shell) 2.897E-06 7.64 E-08 3.406 3.866 3.364 2.695
4 2.501E-06 7.26 E-07 2.653 3.112 2.600 2.708
2.56 E-06
6 (solid) 2.532E-06 1.97 E-06 2.604 2.466 2.603 2.662
8 2.563E-06 2.46 E-06 2.603 2.648 2.603 2.639

Main Index
 26 MSC Nastran Verification Guide

4. Clamped Supports and Concentrated Load

a. Aspect Ratio 1.0

Lateral Deflection at the Center

Theoretical
N [1] HEXA8 PENTA QUAD4 QUAD8 QUADR TRIAR
2 5.60 (shell) 4.960 E-06 8.76E-08 5.612 6.030 5.234 4.354
4 5.447 E-06 5.74 E-07 5.793 5.429 5.662 5.199
5.60E-06
6 (solid) 5.538 E-06 2.69 E-06 5.740 5.560 5.674 5.407
8 5.571 E-06 4.04 E-06 5.701 5.588 5.662 5.488

b. Aspect Ratio 5.0

Lateral Deflection at the Center

Theoretical
N [1] HEXA8 PENTA QUAD4 QUAD8 QUADR TRIAR
2 7.23(shell) 2.322E-06 7.77 E-08 3.823 3.918 3.753 3.236
4 6.145E-06 6.30 E-07 6.692 5.451 6.241 5.658
7.23E-06
6 (solid) 6.705E-06 1.73 E-06 7.105 6.741 6.798 6.405
8 6.920E-06 2.42 E-06 7.242 7.046 7.031 6.708

Conclusion
The results from MSC Nastran are in a good agreement with analytical results.

Main Index
 CHAPTER 1.5 27
Scordelis-Lo-Roof

1.5 Scordelis-Lo-Roof

Problem Description
This problem demonstrates the deformation of thin curved membrane loaded under its own weight. The
computed maximum Z displacement is compared to the theoretical solution.

Solution Number
SOL 101

Features Used
 Linear Statics.

Reference
1. Macneal., R.H., Harder, R.L., 1985. A proposed standard set of problems to test finite element
accuracy. Finite Elem. Anal. Des. 1, 3-20.

Modeling Techniques Used

The model is analyzed using different types of elements such as HEXA, PENTA, QUAD, QUADR and
TRIAR elements. The model details are given below:

Material Properties
The material is linear elastic with a Young's modulus, E = 1.0 × 107 and Poisson's ratio, = 0.25.

Geometrical properties
Radius = 25.0
length=50.
Thickness=0.25
Angle = 40°

Loading
A force of 90.0 per unit area in the - Z direction is applied on the surface.

Main Index
 28 MSC Nastran Verification Guide

Boundary Condition
The displacements boundary conditions are applied as shown in Figure 1.5-1.

Illustrations of the Model

Figure 1.5-1 Scordelis-Lo-Roof

Main Index
 CHAPTER 1.5 29
Scordelis-Lo-Roof

Model File Name and Location

Model File Name Description Location

macneal081 Scordolis-lo-roof, HEXA8 element, N*=2 tpl\verifman\macneal
macneal082 Scordolis-lo-roof, HEXA8 element, N=4 tpl\verifman\macneal
macneal083 Scordolis-lo-roof, HEXA8 element, N=6 tpl\verifman\macneal
macneal084 Scordolis-lo-roof, HEXA8 element, N=8 tpl\verifman\macneal
macneal085 Scordolis-lo-roof, PENTA element, N=2 tpl\verifman\macneal
macneal086 Scordolis-lo-roof, PENTA element, N=4 tpl\verifman\macneal
macneal087 Scordolis-lo-roof, PENTA element, N=6 tpl\verifman\macneal
macneal088 Scordolis-lo-roof, PENTA element, N=8 tpl\verifman\macneal
macneal089 Scordolis-lo-roof, QUAD4 element, N=2 tpl\verifman\macneal
macneal090 Scordolis-lo-roof, QUAD4 element, N=4 tpl\verifman\macneal
macneal081 Scordolis-lo-roof, QUAD4 element, N=6 tpl\verifman\macneal
macneal082 Scordolis-lo-roof, QUAD4 element, N=8 tpl\verifman\macneal
macneal083 Scordolis-lo-roof, QUAD4 element, N=2 tpl\verifman\macneal
macneal084 Scordolis-lo-roof, QUAD8 element, N=4 tpl\verifman\macneal
macneal085 Scordolis-lo-roof, QUAD8 element, N=6 tpl\verifman\macneal
macneal086 Scordolis-lo-roof, QUAD8 element, N=8 tpl\verifman\macneal
macneal087 Scordolis-lo-roof, QUAD8 element, N=2 tpl\verifman\macneal
macneal088 Scordolis-lo-roof, QUAD8 element, N=4 tpl\verifman\macneal
macneal089 Scordolis-lo-roof, QUADR element, N=6 tpl\verifman\macneal
macneal100 Scordolis-lo-roof, QUADR element, N=8 tpl\verifman\macneal
macneal101 Scordolis-lo-roof, QUADR element, N=2 tpl\verifman\macneal
macneal102 Scordolis-lo-roof, QUADR element, N=4 tpl\verifman\macneal
macneal103 Scordolis-lo-roof, QUADR element, N=6 tpl\verifman\macneal
macneal104 Scordolis-lo-roof, TRIA element, N=8 tpl\verifman\macneal
macneal105 Scordolis-lo-roof, TRIA element, N=2 tpl\verifman\macneal
macneal106 Scordolis-lo-roof, TRIA element, N=4 tpl\verifman\macneal
macneal107 Scordolis-lo-roof, TRIA element, N=6 tpl\verifman\macneal
macneal108 Scordolis-lo-roof, TRIA element, N=8 tpl\verifman\macneal

* where N: Number of node spaces per edge of mode.

Main Index
 30 MSC Nastran Verification Guide

Results

Lateral Deflection at the Center

N* HEXA8 PENTA QUAD4 QUAD8 QUADR TRIAR
2 0.4556 0.0443 0.4868 0.3707 0.4730 0.3029
4 0.3550 0.1082 0.3658 0.3413 0.3643 0.2995
0.3024
6 0.3497 0.2000 0.3534 0.3474 0.3538 0.3229
8 0.3470 0.2735 0.3496 0.3463 0.3504 0.3329
10 0.3478 0.3454 0.3487 0.3377

* where N: Number of node spaces per edge of model

Conclusion
Except for PENTA elements, all other element results from MSC Nastran are in good agreement with the
theoretical solution.

Main Index
 CHAPTER 1.6 31
Spherical Shell

1.6 Spherical Shell

Problem Description
This problem demonstrates the use of different elements to model a thin hemispherical shell subjected to
concentrated loads. The solution is compared to the theoretical solution [1].

Solution Number
SOL 101

Features Used
 Linear Statics.

Reference
1. Macneal., R.H., Harder, R.L., 1985. A proposed standard set of problems to test finite element
accuracy. Finite Elem. Anal. Des. 1, 3-20.

Modeling Techniques Used

The model is analyzed using different types of elements such as HEXA, PENTA, QUAD, QUADR and
TRIAR elements. The model details are given below:

Material Properties
The material is linear elastic with a Young's modulus, E = 6.825 × 107 and a Poisson's ratio,  = 0.3.

Geometrical properties
Radius = 10.0
Thickness = 0.04

Loading
1.0-unit load is applied at quadrant.

Boundary Condition
The displacements boundary conditions are applied as shown in Figure 1.6-1.

Main Index
 32 MSC Nastran Verification Guide

Illustrations of the Model

Figure 1.6-1 Spherical shell modeled with eight hexahedral elements

Main Index
 CHAPTER 1.6 33
Spherical Shell

Model File Name and Location

Model File Name Description Location

macneal135 Spherical Shell, HEXA8 elements, N*=4 tpl\verifman\macneal
macneal136 Spherical Shell, HEXA8 elements, N=8 tpl\verifman\macneal
macneal137 Spherical Shell, HEXA8 elements, N=12 tpl\verifman\macneal

macneal138 Spherical Shell, PENTA elements, N=4 tpl\verifman\macneal

macneal109 Spherical Shell, PENTA elements, N=8 tpl\verifman\macneal
macneal110 Spherical Shell, PENTA elements, N=12 tpl\verifman\macneal

macneal111 Spherical Shell, QUAD4 elements, N=2 tpl\verifman\macneal

macneal112 Spherical Shell, QUAD4 elements, N=4 tpl\verifman\macneal
macneal113 Spherical Shell, QUAD4 elements, N=6 tpl\verifman\macneal
macneal114 Spherical Shell, QUAD4 elements, N=8 tpl\verifman\macneal
macneal115 Spherical Shell, QUAD4 elements, N=10 tpl\verifman\macneal
macneal116 Spherical Shell, QUAD4 elements, N=12 tpl\verifman\macneal

macneal117 Spherical Shell, QUAD8 elements, N=2 tpl\verifman\macneal

macneal118 Spherical Shell, QUAD8 elements, N=4 tpl\verifman\macneal
macneal119 Spherical Shell, QUAD8 elements, N=6 tpl\verifman\macneal
macneal120 Spherical Shell, QUAD8 elements, N=8 tpl\verifman\macneal
macneal121 Spherical Shell, QUAD8 elements, N=10 tpl\verifman\macneal
macneal122 Spherical Shell, QUAD8 elements, N=12 tpl\verifman\macneal

macneal123 Spherical Shell, QUADR elements, N=2 tpl\verifman\macneal

macneal124 Spherical Shell, QUADR elements, N=4 tpl\verifman\macneal
macneal125 Spherical Shell, QUADR elements, N=6 tpl\verifman\macneal
macneal126 Spherical Shell, QUADR elements, N=8 tpl\verifman\macneal
macneal127 Spherical Shell, QUADR elements, N=10 tpl\verifman\macneal
macneal128 Spherical Shell, QUADR elements, N=12 tpl\verifman\macneal

macneal129 Spherical Shell, TRIAR elements, N=2 tpl\verifman\macneal

macneal130 Spherical Shell, TRIAR elements, N=4 tpl\verifman\macneal

Main Index
 34 MSC Nastran Verification Guide

Model File Name Description Location

macneal131 Spherical Shell, TRIAR elements, N=6 tpl\verifman\macneal
macneal132 Spherical Shell, TRIAR elements, N=8 tpl\verifman\macneal
macneal133 Spherical Shell, TRIAR elements, N=10 tpl\verifman\macneal
macneal134 Spherical Shell, TRIAR elements, N=12 tpl\verifman\macneal

* where N: Number of node spaces per edge of model

Results

Lateral Deflection at the Center

N HEXA8 PENTA QUAD4 QUAD8 QUADR TRIAR
2 0.09373 0.00230 0.03863 0.07960
4 0.00384 0.00070 0.09716 0.01139 0.04019 0.10591
0.0940
6 0.09581 0.04644 0.06845 0.09543
8 0.06978 0.00379 0.09492 0.07739 0.08401 0.10850
10 0.09441 0.08975 0.08943 0.10482
12 0.08994 0.01446 0.09413 0.09323 0.09140 0.10966

Conclusion
Except for PENTA and CTRIAR elements, the results from MSC Nastran are in a good agreement with
theoretical results.

Main Index
 CHAPTER 1.7 35
Thick Wall Cylinder

1.7 Thick Wall Cylinder

Problem Description
This problem demonstrates the use of different elements to model a thick-walled cylinder subjected to
internal pressure. Nearly incompressible material is used. The solution is compared to the analytical solution.

Solution Number
SOL 101

Features Used
 Linear Statics.

Reference
1. Macneal., R.H., Harder, R.L., 1985. A proposed standard set of problems to test finite element
accuracy. Finite Elem. Anal. Des. 1, 3-20.

Modeling Techniques Used

The model is analyzed using different types of elements such as HEXA, PENTA, QUAD, QUADR and
TRIAR elements. The model details are given below:

Material Properties
The material is linear elastic with a Young's modulus, E = 1.0 × 107 and Poisson's ratio  = 0.49, 0.499 and
0.4999.

Geometrical properties
Inner radius = 3.0
Outer radius=9.0
Thickness=1

Loading
A unit pressure is applied at inner radius as shown in Figure 1.7-1.

Main Index
 36 MSC Nastran Verification Guide

Boundary Condition
uy= uz=Rx=Ry= Rz= 0.0 for all the nodes.

Illustrations of the Model

Figure 1.7-1 Thick-walled Cylinder

Model File Name and Location

Model File Name Description Location

macneal020 Thick Cylinder, HEXA8 elements tpl\verifman\macneal
macneal021 Thick Cylinder, PENTA elements tpl\verifman\macneal
macneal022 Thick Cylinder, QUAD4 elements tpl\verifman\macneal
macneal023 Thick Cylinder, QUAD8 elements tpl\verifman\macneal
macneal024 Thick Cylinder, QUADR elements tpl\verifman\macneal
macneal025 Thick Cylinder, TRIAR elements tpl\verifman\macneal

Results

Radial displace at inside radius

Poisson's Theoretical
ratio [1] HEXA8 PENTA QUAD4 QUAD8 QUADR TRIAR
0.49 0.00504 0.00497 0.00503 0.00497 0.00503 0.00496 0.00419
0.499 0.00506 0.00498 0.00504 0.00498 0.00504 0.00498 0.00404
0.4999 0.00506 0.00499 0.00504 0.00499 0.00489 0.00498 0.00402

Main Index
 CHAPTER 1.7 37
Thick Wall Cylinder

Conclusion
Except CTRIAR elements, the results from MSC Nastran are in a good agreement with analytical results.

Main Index
 38 MSC Nastran Verification Guide

1.8 Twisted Beam

Problem Description
This problem demonstrates the use of different elements to model a twisted beam with a rectangular cross
section subjected to unit forces at free end. The solution is compared with the analytical solution.

Solution Number
SOL 101

Features Used
 Linear Statics.

Reference
1. Macneal., R.H., Harder, R.L., 1985. A proposed standard set of problems to test finite element
accuracy. Finite Elem. Anal. Des. 1, 3-20.

Modeling Techniques Used

The model is analyzed using different types of elements such as HEXA, PENTA, QUAD, QUADR and
TRIAR elements. The model details are given below:

Material Properties
The material is linear elastic with a Young's modulus, E = 29.0 × 106 and Poisson's ratio  = 0.3.

Geometrical properties
Length = 12.0
Width = 1.1
Depth = 0.32
Twist= 90o (root to tip)

Loading
A force of one unit applied at the free end.

Main Index
 CHAPTER 1.8 39
Twisted Beam

Boundary Condition
The displacements ux= uy=uz = 0.0 at one end of the beam as shown in Figure 1.8-1.

Illustrations of the Model

Figure 1.8-1 Twisted Beam

Model File Name and Location

Model File Name Description Location

macneal007 Twisted Beam, HEXA8 elements tpl\verifman\macneal
macneal008 Twisted Beam, PENTA elements tpl\verifman\macneal
macneal009 Twisted Beam, QUAD4 elements tpl\verifman\macneal
macneal010 Twisted Beam, QUAD8 elements tpl\verifman\macneal
macneal011 Twisted Beam, QUADR elements tpl\verifman\macneal
macneal012 Twisted Beam, TRIA elements tpl\verifman\macneal

Results

Radial displace at inside radius

Tip loading Theoretical
direction [1] HEXA8 PENTA QUAD4 QUAD8 QUADR TRIAR
In plane 0.005424 0.00539 0.00537 0.00539 0.00541 0.00538 0.00531
(vertical)
Out-of-plane 0.001754 0.00173 0.00174 0.00173 0.00175 0.00173 0.00164

Conclusion
The results from MSC Nastran are in a good agreement with analytical results.

Main Index
 40 MSC Nastran Verification Guide

Main Index
 Section 2: Original Verification
Manual Problems

Main Index
Main Index
 2.1 Static Analysis of a Pin-
Jointed Truss

Problem Description
A truss consisting of two pin-jointed axial elements is loaded vertically. Find the vertical deflection of the end
and the internal forces and stresses in the elements.

Solution Number
SOL 101

Features Used
 Static Analysis

Reference
1. Egor P. Popov, Introduction to Mechanics of Solids (Englewood Cliffs: Prentice-Hall, Inc. 1968), P.
499

Modeling Techniques Used

 The pin jointed truss is modeled using rods (CONROD). The model details are given below:
Engineering Data
E = 3.0E+7 lb/in2
A1 = 0.15 in2
A2 = 0.25 in2
l = 60. in4
P = 500. lb

Main Index
 42 MSC Nastran Verification Guide

Illustrations of the Model

Figure 2.1-1 Pin Jointed Truss

Model File Location

Reference Deck
Model Name from Old Manual Model Description Location
v10101.dat v2401.dat Static analysis of pin jointed truss. tpl\verifman\oldverif\

Results
Theoretical Solution

At point B: U y = – 0.0444 in
Force in Element 1: P 1 = +2500. lb (tension)
Force in Element 2: P 2 = – 2500. lb (compression)
Stress in Element 1:  1 = P 1  A 1 = +16667. lb/in2
Stress in Element 2:  2 = P 2  A 2 = – 10000. lb/in2

Main Index
 CHAPTER 2.1 43
Static Analysis of a Pin-Jointed Truss

MSC Nastran Solution

At point B: U y = – 0.0444 in
Force in Rod 1: +2500. lb (tension)
Force in Element 2: -2500. lb (compression)
Stress in Element 1: +16667. lb/in2
Stress in Element 2: -10000. lb/in2

Comparison of Results

Solution Summary
Element Stress (lb/in2)
Displacement (in) Element Force (lb) Rod 1 Rod 2
Theory -0.0444  2500. +16667. -10000.
MSC Nastran -0.0444  2500. +16667. -10000.
% Difference 0% 0% 0% 0%

Conclusion
The results from MSC Nastran are comparable with theoretical results.

Main Index
 44 MSC Nastran Verification Guide

2.2 Heat Transfer from Cooling

Fin

Problem Description
A fin of circular cross section is maintained at 250o at one end. The fin is extended into the air. The air
temperature is 70o. Assuming a material conductivity of 132 Btu/hr-ft-oF. Determine the steady-state
temperature distribution along the rod.

Solution Number
SOL 400

Features Used
 Steady heat transfer

Reference
1. Alan J. Chapman, Heat transfer, 3rd ed. (New York: MacMillan Publishing Co., Inc., 1974) P. 76,
Ex. 3.8

Modeling Techniques Used

 The cooling fin is modeled using CBEAM element. The model details are given below:
Engineering Data
L = 1 ft
D = 0.04167 ft
A = 0.001365 ft2
I11=I22 = 1.48E-7 ft4
K = 132 Btu/hr-ft-oF
H = 1.6 Btu/hr-ft-oF
Tbase = 250oF
Tfluid = 70oF

Main Index
 CHAPTER 2.2 45
Heat Transfer from Cooling Fin

Illustrations of the Model

Figure 2.2-1 Cooling Fin

Model File Location

Reference Deck
Model Name from Old Manual Model Description Location
v40001.dat v2402.dat Heat transfer from cooling fin tpl\verifman\oldverif\

Results
Theoretical Solution
cosh  mL – x 
T =  T B – T F   ---------------------------------- + T F
 cosh mL 

where m = 4h
------ (for cylindrical rod)
kd

m = 4  1.6 -
--------------------------------
1
0.63  ---------------
2  12

1
m = 1.079 ----
ft

X/L T (oF)
0.25 217.0243
0.50 196.0559
0.75 183.7156
1.00 179.7002

Main Index
 46 MSC Nastran Verification Guide

Result comparison of theoretical solution and MSC Nastran results are as follows:

T (oF)
X/L 0.250 0.500 0.750 1.000
Theory 217.62 196.06 183.72 179.7
MSC Nastran 217.8 196.338 184.054 180.056
Difference 0.083% 0.142% 0.182% 0.198%

Conclusion
The results from MSC Nastran are comparable with theoretical results.

Main Index
 CHAPTER 2.3 47
Truss Analysis using Thermal Loads

2.3 Truss Analysis using Thermal

Loads

Problem Description
A pin joined truss is loaded with a force at one end of the component elements is heated. Considering thermal
effects, find the displacements of the truss joints and the forces and stresses in the axial elements.

Solution Number
SOL 101

Features Used
 Static analysis with thermal loads

Reference
1. J. S. Przemieniecki, Theory of Matrix Structural Analysis (New York: McGraw- Hill Book Company,
1968), P. 155

Modeling Techniques Used

 The Pin joined truss assembly is modeled using CONROD elements. The details of the model is as
follows.
Engineering Data
E = 1.0E+7 lb/in2
 = 1.0E-6/0F
A1 = 1.0 in2 (Elements 1, 3, 5, 6)
A2 = 0.7071068 in2 (Elements 2 and 4)
T = +100.0F (Elements 3 only)
F = 1000. lb
l = 20. in4

Main Index
 48 MSC Nastran Verification Guide

Illustrations of the Model

Figure 2.3-1 Truss with thermal loads

Model File Location

Reference Deck
Model Name from Old Manual Model Description Location
v10103.dat v2403.dat Truss analysis with thermal loads tpl\verifman\oldverif\

Results

Displacement (in)
u1 v1 u2 v2
Theory -1.272727E-03 6.363636E-03 7.272727E-04 3.636364E-03
MSC Nastran -1.272727E-03 6.363636E-03 7.272727E-04 3.636364E-03
Difference 0.00% 0.00% 0.00% 0.00%

Main Index
 CHAPTER 2.3 49
Truss Analysis using Thermal Loads

Forces (lb)
p1 p2 p3 p4 p5 p6
Theory -636.36 899.95 363.64 -514.26 363.64 0.00
MSC Nastran -636.36 899.95 363.64 -514.26 363.64 0.00
Difference 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%

Stresses (lb/in2)
1 2 3 4 5 6
Theory -636.36 1272.73 363.64 -727.27 363.64 0.00
MSC Nastran -636.36 1272.73 363.64 -727.27 363.64 0.00
Difference 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%

Conclusion
The results from MSC Nastran are comparable with theoretical results.

Main Index
 50 MSC Nastran Verification Guide

2.4 Inertia Load on Pilot During

Landing

Problem Description
Assuming an aircraft is making a level landing with the nose gear slightly off the ground, find the resulting
inertia loads on the pilot due to the main gear landing loads.

Solution Number
SOL 101

Features Used
 Inertia Relief Analysis

Reference
1. E. F. Bruhn, Analysis and Design of flight Vehicle Structures (Cincinnati: Tri- state Offset Company,
1965) P. A4.11

Modeling Techniques Used

 The aircraft and pilot assembly is modeled using beam (CBEAM) and concentrated mass element
(CONM2). The model details are given below:
Engineering Data
Aircraft: Weight = 100,000 lb, pitching mass moment of inertia = 40,000,000 lb.sec2in
Pilot: Weight = 180.0 lb
Forces (at main landing gear): Pz = 300,000 lb (Vertical), Px = 100,000 lb (aft)

Main Index
 CHAPTER 2.4 51
Inertia Load on Pilot During Landing

Illustrations of the Model

Figure 2.4-1 Aircraft with Pilot

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10104.dat v2404.dat Inertia load on Pilot during tpl\verifman\oldverif\
Landing

Results
Theoretical Solution
Loads on Pilot: Fx = 162.7 lb Fz = 378.84 lb

MSC Nastran Solution

The user must put all mass items into consistent units.

Aircraft

Ma = 100,000 lb ( 1--- = 0.002588 sec2/in)

g

Ma = (258.8 lb-sec2/in)

Pilot

Mp = 180 lb ( 1--- = 0.002588 sec2/in)

g

Mp = 0.4658 (lb-sec2/in)
Also the aircraft must be adjusted for the pilot.

Main Index
 52 MSC Nastran Verification Guide

 Ma = 258.8 – 0.4658

 Ma = 258.3342 (lb-sec2/in)
The MSC Nastran results are:
Fx = 162.62 lb
Fz = 378.46 lb

Result comparison of theoretical solution and MSC Nastran results are as follows:

Load on Pilot (lb)

Fx Fz
Theory 162.7 378.84
MSC Nastran 162.6 378.46
Difference -0.06% -0.10%

Conclusion
The results from MSC Nastran are comparable with theoretical results.

Main Index
 CHAPTER 2.5 53
Cantilever Beam, Statics, Checkpoint

2.5 Cantilever Beam, Statics,

Checkpoint

Problem Description
A simple cantilever beam is loaded with a vertical shear load at the free end. The solution results are saved
(checkpoint) for a restart in a modal solution (v10305r).

Solution Number
SOL 101

Features Used
 Static Analysis

Reference
1. R.J. Roark and W. C. Young, Formulas for stress and strain (New York: McGraw-Hill Book Co.
1975) P. 96

Modeling Techniques Used

 The cantilever beam is modeled using beam (CBEAM) element. The model details are given below:
Engineering Data
E = 1.0E+7 lb/in2
 = 0.3
l = 30.0 in
h = 1.0 in
w = 0.75 in
t = 0.10 in
A = 0.310 in2
I1 = 0.0390 in4
I2 = 0.0241 in4
J = 0.0631 in4

Main Index
 54 MSC Nastran Verification Guide

Mass = 1.0 lb-sec2/in

 = 1.0 in/sec2

Illustrations of the Model

Figure 2.5-1 Cantilever Beam

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10105d.dat v2405n.dat Cantilever beam, statics, tpl\verifman\oldverif\
v10305r.dat v0305r.dat checkpoint tpl\verifman\oldverif\

Results
Theoretical Solution
1. Displacement at free end:
3
Fl
 = ---------
3EI
3
100  30 
 = --------------------------------------------
7
-
3  1.0 10 0.0390

 = 2.3077 in
2. Moment at fixed end:
M = FL
M = 100(30)
M = 3000 lb-in

Main Index
 CHAPTER 2.5 55
Cantilever Beam, Statics, Checkpoint

3. Stress at extreme fibre locations:

MC
 = ---------
I

3000   0.50 
 = -------------------------------
 0.0390 

 =  38462 lb/in2
Result comparison of theoretical solution and MSC Nastran results are as follows:

Displacement at free Moment at Fixed end Maximum Stress

end (in) (lb-in) (lb/in2)
 M max
Theory 2.307700 3000.00 38462.00
MSC Nastran 2.307692 3000.00 38462.00
Difference 0.00% 0.00% 0.00%

Conclusion
The results from MSC Nastran are comparable with theoretical results.

Main Index
 56 MSC Nastran Verification Guide

2.6 Spring Supported Bar in a

Skewed Coordinate System

Problem Description
A coordinate system exists that is rotated 45o with respect to a basic frame of reference in the x-y plane. A bar
located in this system is simply supported by elastic supports. A load is applied to the bar at an intermediate
point. Calculate the reaction forces in the supports.

Solution Number
SOL 101

Features Used
 Static Analysis

Reference
1. Joseph Shigley, Mechanical Engineering Design (New York: McGraw-Hill Inc., 1977) P. 642

Modeling Techniques Used

 The spring supported bar is modeled using bar element (CBAR) and spring element (CELAS2). The
model details are given below:
Engineering Data
 = 450
a = 1 in
b = 2 in
k = 1.0E+4 lb/in
F = 600.0 lb
E = 3.0E+7 psi
A = 0.0625 in2
I = 3.255E-4 in4

Main Index
 CHAPTER 2.6 57
Spring Supported Bar in a Skewed Coordinate System

Illustrations of the Model

Figure 2.6-1 Spring supported bar

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10106.dat V2406.dat Spring supported bar in tpl\verifman\oldverif\
skewed system

Results
Result comparison of theoretical solution and MSC Nastran results are as follows:

Reaction Forces / Spring Forces (lb)

R1 R2
Theory 400 200
MSC Nastran 400 200
Difference 0.00% 0.00%

Conclusion
The results from MSC Nastran are comparable with theoretical results.

Main Index
 58 MSC Nastran Verification Guide

2.7 Punched Output of a Static

Solution

Problem Description
A simple structurally intermediate problem (see schematic below) is analyzed using MSC NASTRAN with
the resulting displacement and stress output punched.

Solution Number
SOL 101

Features Used
 Static Analysis

Modeling Techniques Used

 The model details are given below:
Engineering Data
E = 1.0E+7 lb/in2
 = 0.3
l = 1.0 in
M = 1000 in-lb
A = 1.0 in2
I1=I2 = 1.0E-4 in4

Main Index
 CHAPTER 2.7 59
Punched Output of a Static Solution

Illustrations of the Model

Figure 2.7-1 Spring supported bar

Model File Location

Reference Deck
Model Name from Old Manual Model Description Location
v10107.dat V2407.dat Punched output of static solution tpl\verifman\oldverif\

Main Index
 60 MSC Nastran Verification Guide

Results
Punched output of static solution by MSC NASTRAN is as follows.

Main Index
 CHAPTER 2.8 61
Cantilever Beam Constructed of Plates, Stress Contours

Cantilever Beam
2.8 Constructed of Plates,
Stress Contours

Problem Description
A Simple cantilever beam is constructed using plate element and is loaded with vertical shear force at one
end. Find the stress at the outside fibers.

Solution Number
SOL 101

Features Used
 Static analysis

Reference
1. Ferdinand L. Singer, Strength of Materials (Harper & Row, 1962) Art. 52, pp. 133.

Modeling Techniques Used

 The cantilever beam is modelled using CQUAD4 element. The model details are given below:
Engineering Data
E = 10.7E+06 lb/in2
 = 0.3
t = 0.1 in
l = 3.0 in
h = 0.6 in
p = 120.0 lb

Main Index
 62 MSC Nastran Verification Guide

Illustrations of the Model

Figure 2.8-1 Cantilever Beam

Model File Name and Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10108.dat V2408a.dat Cantilever beam tpl\verifman\oldverif
constructed of plates, stress
contours

Results
Theoretical Solution
Moment of Inertia:
3 3
th 0.10  0.6 
I = ------- = -------------------------- = 1.8E-3 in4
12 12

Bending Moment at fixed end:

M = 3.0p = 3.0  120  = 360.0 in-lb
Bending Stress:
MC 360.0C
S = --------- = ------------------ lb/in2
I 1.8E-3

C = Fibre distance from neutral axis

Main Index
 CHAPTER 2.8 63
Cantilever Beam Constructed of Plates, Stress Contours

MSC Nastran Solution

Stresses (psi)
Fibre Distance C = 0.0 C = 0.10 C = 0.20 C = 0.30
Theory 0 20000 40000 60000
MSC Nastran 0 19400 39350 54444
%Difference 0.00% -3.00% -1.63% -9.26%

Conclusion
The results from MSC Nastran are comparable to the theoretical results.

Main Index
 64 MSC Nastran Verification Guide

2.9 Bar with Offsets in a Skewed

Coordinate System

Problem Description
A coordinate system exists that is rotated 45o with respect to a basic frame of reference in the x-y plane. A bar
located in this system is simply supported by elastic supports. The grid points are located along the x1 axis
and the bar at an intermediate point. Calculate the reaction forces in the supports.

Solution Number
SOL 101

Features Used
 Static Analysis

Reference
1. Joseph Shigley, Mechanical Engineering Design (New York: McGraw-Hill Inc., 1977) P. 642

Modeling Techniques Used

 The spring bar system is modeled using CELAS and CBAR element, respectively. The model details
are given below:
Engineering Data
 = 450
a = 1 in
b = 2 in
k = 1.0E+4 lb/in
F = 600.0 lb
E = 3.0E+7 psi
A = 0.0625 in2
I = 3.255E-4 in4
H = 2 in

Main Index
 CHAPTER 2.9 65
Bar with Offsets in a Skewed Coordinate System

Illustrations of the Model

Figure 2.9-1 Bar with Spring system

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10109.dat V2409.dat Bar with offsets in skewed tpl\verifman\oldverif\
coordinate system

Results
Theoretical Solution
Fb 600  2 
R 1 = ------- = ----------------- = 400.0 lb
l 3

Fa 600  1 
R 2 = ------- = ----------------- = 200.0 lb
l 3

Result comparison of theoretical solution and MSC Nastran results are as follows:

Reaction Forces / Spring Forces (lb)

R1 R2
Theory 400 200
MSC Nastran 400 200
Difference 0.00% 0.00%

Main Index
 66 MSC Nastran Verification Guide

Conclusion
The results from MSC Nastran are comparable with theoretical results.

Main Index
 CHAPTER 2.10 67
Thick Walled Cylinder, Internal Pressure

2.10 Thick Walled Cylinder,

Internal Pressure

Problem Description
Find the stress distribution for a thick-walled cylinder subjected to internal pressure.

Solution Number
SOL 101

Features Used
 Static Analysis

Reference
1. S. H. Crandall, N. C. Dahl, and T. J. Lardner, An Introduction to the Mechanics of Solids, 2nd ed.
(New York: McGraw-Hill Book Company, 1972) P. 293-297

Modeling Techniques Used

 Thick Cylinder is modeled using CTRIAX6 element. The model details are given below:
Engineering Data
ri = 6.0 in
ro = 12.0 in
h = 8.0 in
E = 30.0E+6 psi
 = 0.28 lb/in3
 = 0.0
p = 10.0 lb/in2

Main Index
 68 MSC Nastran Verification Guide

Illustrations of the Model

Figure 2.10-1 Thick walled Cylinder

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10110.dat V2410.dat Thick walled cylinder, tpl\verifman\oldverif\
internal pressure

Results
Theoretical Solution
2
–Pi   ro  r  – 1 
 r = -------------------------------------------
2
-
 ro  ri  – 1

2
Pi   ro  r  + 1 
  = ----------------------------------------
2
-
 ro  ri  – 1

Result comparison of theoretical solution and MSC Nastran results are as follows:

r lb/in2  lb/in2
r r/ri Theory MSC Nastran % Error Theory MSC Nastran % Error
6.0 1 -10.00 16.67
6.50 1.083 -8.030 -8.095 0.811% 14.69 14.852 1.104%
7.0 1.167 -6.460 13.13
7.50 1.25 -5.200 -5.238 0.731% 11.87 11.956 0.722%

Main Index
 CHAPTER 2.10 69
Thick Walled Cylinder, Internal Pressure

r lb/in2  lb/in2
r r/ri Theory MSC Nastran % Error Theory MSC Nastran % Error
8.0 1.333 -4.170 10.83
8.50 1.417 -3.310 -3.333 0.704% 9.98 10.031 0.509%
9.0 1.5 -2.590 9.26
9.50 1.583 -1.990 -2.000 0.502% 8.65 8.686 0.421%
10.0 1.667 -1.470 8.13
10.50 1.75 -1.020 -1.030 1.009% 7.69 7.710 0.262%
11.0 1.833 -0.630 7.3
11.50 1.917 -0.300 -0.303 1.009% 6.96 6.979 0.271%
12.0 2 0.00 6.67

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 70 MSC Nastran Verification Guide

2.11 Cantilevered Cylindrical

Shell, Axisymmetric Force

Problem Description
Find the radial deflection and meridional moments in a cantilevered cylindrical shell.

Solution Number
SOL 101

Features Used
 Static Analysis

Reference
1. "Analysis of Axisymmetrical shells by the Direct Stiffness Method" P. E. Grafton and D. R. Strome,
AIAA Journal, 1(10): 2342-2347
2. "Evaluation of NASTRAN" J. W. Jones and H. H. Fong, Structural Mechanics Software Series, Vol.
IV (N. Perrone and W. Pilkey, eds.), 1982

Modeling Techniques Used

Engineering Data
E = 1.0E+7 psi
 = 0.3

3 3
Bh 1  0.01  –8
Moment of inertia per unit width: I = --------- = --------------------- = 8.33 10
12 12

Main Index
 CHAPTER 2.11 71
Cantilevered Cylindrical Shell, Axisymmetric Force

Illustrations of the Model

Figure 2.11-1 Cantilevered Cylinder shell

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10111.dat V2411.dat Cantilevered Cylinder shell tpl\verifman\oldverif\

Results
Theoretical Solution
Q - – X
Radial deflection: U r = -------------
3
e cos X
2 D

Q – X
Meridional moment: M = ---- e sin X


where,
 = 1.285  Rt = 5.746695 in-1
Result comparison of theoretical solution and MSC Nastran results are as follows:

Radial Deflection (10E-3 in)

Z (in) Theory MSC Nastran % Error

0 2.8769 2.814938 -2.15%
0.2 0.37291 0.3712448 -0.45%
0.4 -0.19215 -0.1812774 -5.66%
0.6 -0.08725 -0.08327 -4.56%

Main Index
 72 MSC Nastran Verification Guide

Bend moment (in/lbs)

Z (in) Theory MSC Nastran
0.1 5.32E-02 4.31E-02
0.3 3.07E-02 3.02E-02
0.5 2.61E-03 3.55E-03
0.7 -2.40E-03 -2.06E-03

Main Index
 CHAPTER 2.11 73
Cantilevered Cylindrical Shell, Axisymmetric Force

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 74 MSC Nastran Verification Guide

Static Analysis of a Pin-

2.12 Joined Truss Using
Superelements

Problem Description
A truss consisting of two pin-jointed axial elements is loaded vertically. Find the vertical deflection of the end
and the internal forces.

Solution Number
SOL 101

Features Used
 Static Analysis using Superelements

Reference
1. Egor P. Popov, Introduction to Mechanics of Solids (Englewood Cliffs: Prentice-Hall, Inc.. 1968) P.
499.

Modeling Techniques Used

 The conventional statics deck v10101 is converted by adding the SEALL, SUPER and SESET cards.
Their function is described by comments in the data deck. The model details are given below:
Engineering Data
E = 3.0E+7 lb/in2
A1 = 0.15 in2
A2 = 0.25 in2
l = 60.0 in
p = 3000 lb

Main Index
 CHAPTER 2.12 75
Static Analysis of a Pin-Joined Truss Using Superelements

Illustrations of the Model

Figure 2.12-1 Pin Joined Truss

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10112s.dat V6101s.dat Static analysis of pin joined truss tpl\verifman\oldverif

Results
Theoretical Solution

At point B: U y = – 0.0444 in
Force in Element 1: +2500. lb (tension)
Force in Element 2: -2500. lb (compression)

MSC Nastran Solution

At point B: U y = – 0.0444 in
Force in Rod 1: +2500. lb
Force in Rod 1: -2500. lb

Main Index
 76 MSC Nastran Verification Guide

Comparison of Results

Displacement (in) Element Force (lb)

Theory -0.444 ±2500.
MSC Nastran -0.444 ±2500.
% Difference 0% 0%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 CHAPTER 2.13 77
Beam with Point Masses, Enforced Acceleration

2.13 Beam with Point Masses,

Enforced Acceleration

Problem Description
A beam, with equal point masses on each end, is given a unit acceleration on its center point normal to its
axis. Find the reaction force, Q, required to produce this acceleration, and the resulting deflections relative
to the point of application of the force.

Solution Number
SOL 101

Features Used
 Superelement Statics with Inertial relief

Reference
1. Raymond J. Roark, Formulas for Stress and Strain, 2nd ed. (New York: mcGraaw-Hill Book
Company, 1965), P. 104.

Modeling Techniques Used

1. The BAR elements are given section properties for A, I1 and J, as well as I2, the section property in
the loaded direction, so that the model is fully defined in three dimensions.
2. The offsets of the masses provide moments of inertia of the total structure for all directions except
R1. A nominal I1 is therefore added to again give full three-dimensional capability to the model.
3. The enforced acceleration is input on the DMIG cards named UACCEL. The column numbers start
with 1, 0, and would be incremented to 2, 0; 3, 0;... if there were more loading conditions.
4. The loading condition is applied to the same grid point used as a reference point on the DMIG,
SUPORT, and PARAM, GRDPNT cards.
 The model details are given below:
Engineering Data
E = 1.0E+07 lb/in2
l = 10.0 in
I1 = I2 = I3 = 3.0 in4

Main Index
 78 MSC Nastran Verification Guide

Illustrations of the Model

Model File Name and Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10113s.dat V9101s.dat Beam with point masses, tpl\verifman\oldverif
enforced acceleration

Results
Theoretical Solution
From symmetry arguments, the loading condition is equivalent to a point load on a cantilevered beam of half
the total length of this beam. The reference gives the load-deflection equation:
3
p3 l
u 3 = ----------
3EI

Where P3 is the applied load equal to the unit tip mass times the unit acceleration, or unity.
Therefore,
3
10 - –5
u 3 = ------------------------
7
= 1.111...  10
3 10  3

The reaction force is twice the applied load for one side, therefore,
Q = – 2.0

Main Index
 CHAPTER 2.13 79
Beam with Point Masses, Enforced Acceleration

MSC Nastran Results

The reaction force, defined as the force applied by the structure on the supported point, appears in the
SPCFORCES. It agrees with the theoretical value to all places printed. The displacements also agree with the
theoretical values to all places printed.

Conclusion
The results from MSC Nastran are comparable to the theoretical results.

Main Index
 80 MSC Nastran Verification Guide

2.14 Two Degree of Freedom

Modal Analysis

Problem Description
A simple frictionless two degree of freedom system is constructed using springs and concentrated masses. Find
all natural frequencies and mode shapes.

Solution Number
SOL 103

Features Used
 Normal Modes Analysis
 Inverse Power Method and Lanczos Method

Reference
1. Donald. T. Greenwood, Principles of Dynamic (Englewood Cliffs: Prentice-Hall, Inc., 1965) p.459,
ex. 9-1.

Modeling Techniques Used

 The spring and concentrated mass assembly are modeled using rods (CONROD) and concentrated
masses (CONM2) respectively. The model details are given below:
Engineering Data
E 1.0E+5 lb/in2
A 0.1 in2
L 10. in
M1 4.0 lb sec2/in
M2 1.0 lb sec2/in

Main Index
 CHAPTER 2.14 81
Two Degree of Freedom Modal Analysis

Illustrations of the Model

Figure 2.14-1 Two degree of freedom system using spring and concentrated masses

Model File Name and Location

Model File Name Description Location

v10301.dat (Reference deck from TWO D.O.F. SYSTEM MODAL tpl\verifman\oldverif
old manual is v0301.dat) ANALYSIS - Inverse Power method
v10301l.dat TWO D.O.F. SYSTEM MODAL tpl\verifman\oldverif
ANALYSIS - Lanczos Method

Results
MSC Nastran by using rods (CONROD) and Concentrated mass (CONM2) the inverse power method
gives following results and are compared with theoretical results:

Theoretical MSC Nastran

INV LAN
Frequency Frequency Frequency
(rad/sec) Eigenvectors (rad/sec) Eigenvectors (rad/sec) Eigenvectors
Mode 1 10.83 1.000 10.83 1.000 10.83 1.000
0.531 0.531 0.531
Mode 2 46.18 -0.133 46.18 -0.133 46.18 -0.133
1.000 1.000 1.000

Conclusion
The results from MSC Nastran are comparable to the theoretical results.

Main Index
 82 MSC Nastran Verification Guide

Normal Modes of a Beam

2.15 Structure using Component
Modal Synthesis

Problem Description
Find all the natural frequencies of the clamped-clamped massless beam with the concentrated masses, shown
below. (Note that the model produces uncoupled modes with integer eigenvalues.)

Solution Number
SOL 103

Features Used
 Modal Analysis.
 Component Modal Synthesis
 Givens Method

Reference
1. J. P. Den Hartog, Mechanical Vibrations 4th ed., (New York: McGraw-Hill Book Company, 1956),
pp. 429-430.

Modeling Techniques Used

 As Generalized Dynamic Reduction is no longer supported, this file has been modified to use
Component Modal Synthesis. As the A-set defined contains no physical GRID points, the
component modes are identical to the system modes.
The beam and concentrated mass assembly are modeled using beam (CBEAM) and concentrated
masses (CONM2) respectively. The model details are given below:
Engineering Data
Beam Properties
A = 1.0 (Extensional Area)
I1 = 2. 0 (Bending moment of inertia in 1-3 plane)
I2 = 1.0 (Bending Moment of inertia in 1-2 plane)

Main Index
 CHAPTER 2.15 83
Normal Modes of a Beam Structure using Component Modal Synthesis

J = 1.25 (Moment of inertia for Torsion)

E = 1.0 (Young's Modulus)
G = 1.0 (Shear Modulus)
L = 1.0 (Length)
Mass Properties
M1 = 1.0 (Axial)
M1, M2 = 64.0 (Lateral)
I11 = 1.0
I22, I33 = 16.0
Note: Consistent, hypothetical units are used.

Illustrations of the Model

Figure 2.15-1 clamped-clamped massless beam with the concentrated masses

Model File Name and Location

Model File Name Description Location

v10302.dat (Reference deck from Beam-Mass Structure, Normal Modes, tpl\verifman\oldverif
old manual is v0302.dat) Gen. Dynamic Red

Main Index
 84 MSC Nastran Verification Guide

Results
Theoretical Solution
For single degree of freedom systems (see (2.15-18), for reference)
K
 = ---- (2.15-1)
N

 =  , rad/unit time (2.15-2)

u
f = ------ , cycles/ unit time (2.15-3)
2
Axial Extension Mode (T1)
(Axial springs in parallel)
EA EA 4EA
K 1 = --------- + --------- = ----------- = 4.0 (2.15-4)
l2 l2 l

 T1 = 4.0  1.0 = 4.0 ,  = 2.0 , f = 0.3183 (2.15-5)

Beam Bending - Translation Mode (T2, T3)

See (2.15-6) for reference.
192EI 2
K 2 = ------------------ = 192 (2.15-6)
3
l

m 2 = 64 (2.15-7)

T = 3. , u = 1.732 , f = 0.2757 (2.15-8)

2

192EI 1
K 3 = ------------------ = 384 (2.15-9)
3
l

m 3 = 64 (2.15-10)

T = 6. ,  = 2.4495 , f = 0.3898 (2.15-11)

3

Torsion Mode (R1)

See (2.15-19) for reference.
4GJ
K 1 = ---------- = 5 (2.15-12)
l

l 1 = 1.0 (2.15-13)

Main Index
 CHAPTER 2.15 85
Normal Modes of a Beam Structure using Component Modal Synthesis

R = 5. , u = 2.2361 , f = 0.3559 (2.15-14)

1

Beam Bending - Rotation Mode (R2, R3)

See (2.15-17) for reference.
16EI 2
K 3 = --------------- = 32 (2.15-15)
3
l

I 33 = 16 (2.15-16)

R = 1.0 ,  = 1.0 , f = 0.1592 (2.15-17)

3

16EI 1
K 2 = --------------- = 32 (2.15-18)
3
l

I 22 = 15. (2.15-19)

R = 2. ,  = 1.4142 , f = 0.2251 (2.15-20)

2

Comparison of MSC Nastran results with theoretical results is as follows:

Frequency (rad/sec)
Mode No. Theoretical Results MSC Nastran Results
1 1.00000E+00 1.00000E+00
2 1.41420E+00 1.41421E+00
3 1.73200E+00 1.73205E+00
4 2.00000E+00 2.00000E+00
5 2.23610E+00 2.23607E+00
6 2.24495E+00 2.44949E+00

Conclusion
The results from MSC Nastran are comparable with theoretical results.

Main Index
 86 MSC Nastran Verification Guide

2.16 Torsional Vibration of a Shaft

with Three Disks

Problem Description
Find the torsional modes of a shaft fixed at one end.

Solution Number
SOL 103

Features Used
 Modal Analysis.

Reference
1. Walter C. Hurty and Moshe F. Rubinstein, Dynamics of Structures (Englewood Cliffs: Prentice-
Hall, Inc.. 1964).

Modeling Techniques Used

 The shaft and three disk assembly is modelled using beam (CBEAM) and concentrated masses
(CONM2) respectively. The model details are given below:
Engineering Data
M = 100.0 lb-sec2 /in (mass)
I1 = I2 = I3 = 100.0 lb-sec2 /in (mass moment of inertia)
L = 10.0 in
G = 4.0E+6 lb/in2
 = 0.3
J = 1.0 in4 (torsional stiffness coefficient)
A = 1.0 in2
I2 = I3 = 0.5 in4 (area moment of inertia)

Main Index
 CHAPTER 2.16 87
Torsional Vibration of a Shaft with Three Disks

Illustrations of the Model

Figure 2.16-1 Shaft fixed at one end

Model File Name and Location

Model File Name Description Location

v10303.dat (Modified Givens Torsional Vibration of a shaft with three tpl\verifman\oldverif
Method) (Reference deck from old disks.
manual is v0303.dat)
v103031.dat (Lanczos Method) Torsional Vibration of a shaft with three tpl\verifman\oldverif
disks.

Results

MSC Nastran
Frequency (rad/sec) Theory MGIV LAN
1 89.00 89.008 89.008
2 249.40 249.396 249.396
3 360.40 360.388 360.388

Conclusion
The results from MSC Nastran are comparable with theoretical results.

Main Index
 88 MSC Nastran Verification Guide

2.17 Cantilever Beam, Normal

Modes, Restart

Problem Description
Find the two lowest natural frequencies of a cantilever beam, Restart the run from a previous statics solution,
v10105d.

Solution Number
SOL 103

Features Used
 Eigenvalue solution using Modified Givens method. (Restart)

Reference
1. R.J. Roark and W. C. Young, Formulas for stress and strain (New York : McGraw-Hill Book Co..
1975) P. 576

Modeling Techniques Used

 The cantilever beam is modelled using beam (CBEAM) element.
Engineering Data
E = 1.0E+7 lb/in2
 = 0.3
l = 30.0 in
h = 1.0 in
w = 0.75 in
t = 0.10 in
A = 0.310 in2
I1 = 0.0390 in4
I2 = 0.0241 in4
J = 0.0631 in4

Main Index
 CHAPTER 2.17 89
Cantilever Beam, Normal Modes, Restart

Mass = 1.0 lb-sec2 /in

 = 1.0 in/sec2

Illustrations of the Model

Figure 2.17-1 Cantilever Beam

Model File Name and Location

Model File Name Description Location

v10105d.dat (Reference deck from Cantilever Beam Statics. tpl\verifman\oldverif
old manual is v2405N.dat)
v10305r.dat (Reference deck from Restart from statics to calculate modes. tpl\verifman\oldverif
old manual is v0305r.dat)

Results
Theoretical Solution

1.732
f 1 = ------------- EIg
---------
2 wl 3

For I2 = 0.0241

1.732
f 1 = ------------- 1.0 + 7  0.0241 
---------------------------------------
2 1.0  30 
3

f1= 0.823 Hz
For I1 = 0.039

Main Index
 90 MSC Nastran Verification Guide

1.732
f 2 = ------------- 1.0 + 7  0.039 
------------------------------------
2 1.0  30 
3

f2= 1.048 Hz

Result comparison of MSC Nastran and theoretical results are as follows:

Frequency (Hz)
Theory MSC Nastran
Mode 1 0.823 0.823583
Mode 2 1.048 1.047686

Conclusion
The results from MSC Nastran are comparable with theoretical results. Note: Modified Givens method is
good for use on small problems, but for large problems, Lanczos method is preferred.

Main Index
 CHAPTER 2.18 91
Normal Modes Analysis with Reduction Techniques

2.18 Normal Modes Analysis with

Reduction Techniques

Problem Description
Find the two lowest modes of a simply supported shaft. Compare the results of the inverse power iteration
method without reduction, the Givens method with Guyan reduction and the Lanczos method.

Solution Number
SOL 103

Features Used
 Modal Analysis

Reference
1. J. P. Den Hartog, Mechanical Vibration 4th ed.. (New York: McGraw- Hill, 1956), P. 432

Modeling Techniques Used

1. General Comments
The data generator will produce 20 equal bar elements, whose associated grid point identification
numbers are 10000, 10001,...,10020.
2. Guyan Reduction
a. A rule of thumb for the number of degrees of freedom retained for dynamic analysis is two to four
times the number of modes required. Six degrees of freedom are selected by ASET1 cards.
b. The retained degrees of freedom are placed at the expected antinodes of the two modes. Both
translational and rotational degrees of freedom are used at the quarter spans.
3. Inverse Power Iteration
4. Lanczos Method
Engineering Data
L = 100.0 in (Shaft length)
r = 1.0 in (Shaft radius)
t = 0.05 in (Wall thickness)

Main Index
 92 MSC Nastran Verification Guide

E = 30E+7 lb/in2
 = 0.3
 = 7.764E-4 lb sec2 / in4 (Mass density)

Illustrations of the Model

Figure 2.18-1 Simply Supported Shaft

Model File Name and Location

Model File Name Description Location

v10306guy.dat (Reference deck Simply supported shaft using Guyan tpl\verifman\oldverif
from old manual is v03guy.dat) reduction
v10306inv.dat (Reference deck Simply supported shaft using inverse power tpl\verifman\oldverif
from old manual is v03inv.dat)
v10306lan.dat (Reference deck Simply supported shaft using Lanczos tpl\verifman\oldverif
from old manual is v03lan.dat)

Main Index
 CHAPTER 2.18 93
Normal Modes Analysis with Reduction Techniques

Results
Theoretical Solution
The reference gives the natural frequency equation as
a n EI
f n = ------ -----------
- (2.18-1)
2  L 4
1

2
where 1 is the mass per unit length, n is the mode index and a n =  n  . Approximate equations for thin
walled circular cylinder section properties are:
I = r2t = 0.15708 (2.18-2)
A = 2rt =0.31416 (2.18-3)

1 = A = 2.439 x 10-4 (2.18-4)

Substituting the terms

 r E  2
- n = 21.8335n2 cycles/second
f n =  ------ ----- (2.18-5)
 e 2 2

Theory MSC Nastran

Natural
Frequency Inverse % Difference Guyan % Difference % Difference
Mode (Hz) Closed Power No compared to Reduction compared to Lanczos compared to
No. Form Reduction theoretical A-set = 9 theoretical Method theoretical
1 21.8335 21.83346 0.000% 21.8398 0.029% 21.83346 0.000%
2 87.3339 87.33327 -0.001% 87.8204 0.557% 87.33327 -0.001%
3 196.501 196.49410 -0.004% 202.7124 3.161% 196.49410 -0.004%
4 349.335 349.29400 -0.012% 411.6080 17.826% 349.29400 -0.012%
5 545.837 545.67020 -0.031% 713.3204 30.684% 545.67020 -0.031%
6 786.005 785.47620 -0.067% 1269.1490 61.468% 785.47620 -0.067%

Conclusion
The model was run using three methods. The inverse power method and Lanczos method involve no
approximation and are used to give a measure of the discretization error caused by using 20 elements with
lumped masses. The error in natural frequency is less than one tenth of one percent, even for worst case.
For Guyan reduction, the error for the first and second mode is well below one percent, the criteria used for
picking the number of A-set points. However, the error rapidly climbs to 60 percent for the sixth mode.

Main Index
 94 MSC Nastran Verification Guide

Modes of Fixed Circular

2.19 Plate with Trapezoidal Ring
Element

Problem Description
Find the natural modes of fixed circular plate.

Solution Number
SOL 103

Features Used
 Modal Analysis.
 Inverse Power method with enhancements and Lanczos method

Reference
1. S. Timoshenko, D. H. Young and W. Weaver. Jr., Vibration Problems in Engineering, 4th ed. (New
York: John Wiley and Sons, 1947) p. 501

Modeling Techniques Used

 The circular plate is modeled using triangular axisymmetric element (CTRIAX6) elements.
Engineering Data
a = 3.5 in (Circular plate radius)
h = 0.125 in (Plate thickness)
E = 30E+6 lb/in2
 = 0.3
g = 386.4 in/sec2
 = 0.3 lb /in3

Main Index
 CHAPTER 2.19 95
Modes of Fixed Circular Plate with Trapezoidal Ring Element

Illustrations of the Model

Figure 2.19-1 Circular Plate

Model File Name and Location

Model File Name Description Location

v10307.dat (Reference deck from Modes of Fixed Circular Plate with tpl\verifman\oldverif
old manual is v0307.dat) Trapezoidal Ring Element using Inverse
Power Method with Enhancement
v10307lan.dat (Reference deck Modes of Fixed Circular Plate with tpl\verifman\oldverif
from old manual - NA) Trapezoidal Ring Element using Lanczos
Method

Results
Theoretical Solution
 D
 = ----- -----
- (2.19-1)
2
a h
3
Eh
D = --------------------------- (2.19-2)
2
12  1 –  
3
 30.E + 6   0.125 
D = ------------------------------------------------ (2.19-3)
2
12  1 – 0.3 
D = 5365.72 (2.19-4)

 5365.72 -
 = --------------- ---------------------------------------- (2.19-5)
2  7.764E-4 0.125
 3.5 

Main Index
 96 MSC Nastran Verification Guide

F = (606.988) (2.19-6)

Frequencies for N = 0
S   (rad/sec)
0 10.21 6197.35
1 39.78 24145.98
2 88.90 53961.23

Comparison of Results

MSC Nastran (Frequency rad/sec)

Mode No. Theory SINV %Difference LAN %Difference
Theory 6197.35 6191.7350 -0.091% 6191.7350 -0.091%
MSC Nastran 24145.98 23911.2700 -0.972% 23911.2700 -0.972%
Difference 53961.23 52928.0600 -1.915% 52928.0600 -1.915%

Conclusion
For MSC Nastran results using Inverse power method with enhancement and Lanzos, error is less than one
percent for first two modes and it rises to approximately two percent for third mode.

Main Index
 CHAPTER 2.20 97
Two Degree of Freedom Modal Analysis Using Superelements

Two Degree of Freedom

2.20 Modal Analysis Using
Superelements

Problem Description
A simple frictionless two degree of freedom system is constructed using spring and concentrated masses. Find
all natural frequencies and mode shapes.

Solution Number
SOL 103

Features Used
 Modal Analysis
 Superelements Method

Reference
1. Donald. T. Greenwood, Principles of Dynamic (Englewood Cliffs: Prentice-Hall, Inc., 1965)
P.459, Ex. 9-1.

Modeling Techniques Used

 The spring and concentrated mass assembly are modeled using rods (CONROD) and concentrated
masses (CONM2) respectively. The model details are given below:
Engineering Data
E = 1.0E+5 lb/in2
A = 0.10 in2
R = 10.0 in
M1 = 4.0 lb-sec2/in
M2 = 1.0 lb-sec2/in

Main Index
 98 MSC Nastran Verification Guide

Illustrations of the Model

Figure 2.20-1 Two degree of freedom system using spring and concentrated masses

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10308s.dat V6301s.dat Two D.O.F. system modal analysis tpl\verifman\oldverif
using superelement
(Inverse Power Method).

Results
MSC Nastran by using rods (CONROD) and Concentrated mass (CONM2) the inverse power method
gives following results and are compared with theoretical results.

Eigenvectors
Frequency
(rad/sec) Theoretical MSC Nastran (INV)
Mode 1 10.83 1.000 1.000
0.531 0.531
Mode 2 46.18 -0.133 -0.133
1.000 1.000

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 CHAPTER 2.21 99
Modes of Fixed Circular Plate with Superelement Method

Modes of Fixed Circular

2.21 Plate with Superelement
Method

Problem Description
Find the natural modes of fixed circular plate.

Solution Number
SOL 103

Features Used
 Modal Analysis
 Superelements Method

Reference
1. S. Timoshenko, D. H. Young and W. Weaver. Jr., Vibration Problems in Engineering, 4th ed. (New
York: John Wiley and Sons, 1947) P. 501

Modeling Techniques Used

 The circular plate is modeled using CTRIAX6 elements. The model details are given below:
Engineering Data
a = 3.5 in (Circular plate radius)
h = 0.125 in (plate thickness)
E = 30E+6 lb/in2
 = 0.3
g = 386.4 in/sec2
 = 0.3 lb/in3

Main Index
 100 MSC Nastran Verification Guide

Illustrations of the Model

Figure 2.21-1 Circular plate

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10309s.dat V6302s.dat Modes of fixed circular plate with tpl\verifman\oldverif
trapezoidal ring element using inverse
power method with enhancement

Results
Theoretical Solution
 D
 = ----- -----
- (2.21-1)
2
a h
3
Eh
D = --------------------------- (2.21-2)
2
12  1 –  
3
 30.E + 6   0.125 
D = ------------------------------------------------ (2.21-3)
2
12  1 – 0.3 
D = 5365.72 (2.21-4)

 5365.72 -
 = --------------- ---------------------------------------- (2.21-5)
2  7.764E-4 0.125
 3.5 
F = (606.988) (2.21-6)

Main Index
 CHAPTER 2.21 101
Modes of Fixed Circular Plate with Superelement Method

Frequencies for N = 0
S   (rad/sec)
0 10.21 6197.35
1 39.78 24145.98
2 88.90 53961.23

Comparison of Results

Frequency (rad/sec)
Mode No. Theory MSC Nastran %Difference
1 6197.35 6107.253 -1.4538%
2 24145.98 23568.68 -2.3909%
3 53961.23 52342.36 -3.0001%

Conclusion
MSC Nastran results are comparable with theoretical results using superelement method.

Main Index
 102 MSC Nastran Verification Guide

Natural Frequencies of a
2.22 Circular Plate Immersed in
Fluid

Problem Description
Find the natural modes of fixed circular plate immersed in fluid.

Solution Number
SOL 103

Features Used
 Modal Analysis

Reference
1. S. Timoshenko, D. H. Young and W. Weaver. Jr., Vibration Problems in Engineering, 4th ed. (New
York: John Wiley and Sons, 1947) P. 501

Modeling Techniques Used

 The circular plate is modeled using CTRIA3 and virtual mass:
Engineering Data
Plate:
r = 3.5 in (Circular plate radius)
t = 0.125 in (plate thickness)
E = 30E+6 lb/in2
 = 0.3
g = 386.4 in/sec2
p = 0.3 lb/in3

Fluid:
f = 0.0385 lb/in3 (1/g = 0.002588 sec2/in)

Main Index
 CHAPTER 2.22 103
Natural Frequencies of a Circular Plate Immersed in Fluid

Illustrations of the Model

Figure 2.22-1 Circular plate immersed in fluid

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10310s.dat V6303s.dat Natural frequency of a circular plate tpl\verifman\oldverif
immersed in fluid

Results
Theoretical Solution
First mode:
2
1 10.21 1 Et
f = ------ ------------------ ----- ------------------------------------------------------------------
- (2.22-1)
2 1 +  r 2  1 –  2   0.002588   12 
p

f r
 = 0.6689 ----- - (2.22-2)
p t

0.0385 3.5
 = 0.6689 ---------------- ------------- = 2.4036 (2.22-3)
0.3 0.125

Natural frequencies of a circular plate immersed in a fluid

7 2
1 10.21 1 3.0  10  0.125 
f = ------ -------------------------- ---------- ----------------------------------------------------------------------------- = 534.6 Hz (2.22-4)
2 1 + 2.4036 3.5 2  1 – 0.3 2   0.002588   0.3   12 

Main Index
 104 MSC Nastran Verification Guide

Comparison of Results

Frequency (rad/sec)
Mode No. Theory MSC Nastran %Difference
1 534.6 568.16 6.27%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 CHAPTER 2.23 105
DDAM Method of Response Spectra

2.23 DDAM Method of Response

Spectra

Problem Description
The purpose of this Verification Problem is to show how to calculate peak structural response estimates in
compliance with the U.S. Navy shock design modal summation convention (knowns as the NRL method).
The NRL convention calls for a choice among modal contributions to peak response
u ai =  ai q i

of the maximum, uaM, and summation according to the rule

2
u a = u aM +  u ai
iM

It is noted that the maximum contributing mode for a particular response is not necessarily the same for all
physical responses.
This problem demonstrates input of a shock spectrum to a frame structure, to calculate internal loads and
accelerations at points on the structure.

Solution Number
SOL 103

Features Used
 Modal Analysis

Reference
1. "Shock Design Criteria for Surface Ships", NAVSEA 0908-LP-000-3010, May 1976, P. 21.

Modeling Techniques Used

 Response spectra output may be computed in one run, without restart. A two-run sequence was
used here to print out the modal displacements and element forces for inspection. Note that for
efficiency the restart run only adds the SUBCASEs and bulk data entries for the shock spectrum.

Main Index
 106 MSC Nastran Verification Guide

Illustrations of the Model

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10311d.dat V6304d.dat DDAM method of response spectra tpl\verifman\oldverif
v10311r.dat V6304r.dat tpl\verifman\oldverif

Results
Theoretical Solution
Peak absolute modal responses were calculated for the eight flexible body modes (4-11). They are printed in
the run with the title “MATRIX UHVR”.

Mode Hz u hmax u· hmax u··hmax

4 26.655 2.3457E-5 3.9286E-3 6.4796E-1

5 69.617 7.6307E-6 3.3378E-3 1.4600E+0
6 78.709 2.3468E-6 1.1606E-3 5.7396E-1
7 281.127 9.1068E-9 1.6086E-5 2.8414E-2
8 328.144 2.7952E-9 5.7631E-6 1.1882E-2

Main Index
 CHAPTER 2.23 107
DDAM Method of Response Spectra

Mode Hz u hmax u· hmax u··hmax

9 337.648 1.6610E-8 3.5237E-5 7.4757E-2

10 499.944 7.7927E-10 2.4479E-6 7.6894E-3
11 562.488 2.2884E-10 8.0878E-7 2.8584E-3

Estimates for peak acceleration of grid point 6 in the x-direction and bending moment in element 104
(end A) are shown below on a mode-by-mode basis as well as the NRL sum. Corresponding results for the
standard SRSS and ABS conventions are also given.
The  6x i column is extracted from the “REAL EIGENVECTOR” output of the first run.
The u··6x i column can be calculated from the product of the u··h column times the  6x i data.
The  M104A column is extracted from the “FORCES IN BAR ELEMENT” output.
The M104A column can be calculated from the product of the uh column times the  M104A data.

Mode  6x i u··6x i  M104A i M 104A i

4 1.1815+0 0.7774 3.2118+6 7.5339+1

5 5.5595-1 0.8117 6.3625+6 4.8550+1
6 1.0234-0 0.5874 3.9504+6 9.2708+0
7 6.8641-2 1.9504-3 1.1626+7 1.0588-1
8 3.1245-2 3.7125-4 4.2000+6 1.1740-2
9 2.8401-3 2.1232-4 6.1150+6 1.0157-1
10 1.1122-2 8.5520-5 5.9823+6 4.6618-3
11 1.0800-2 3.0870-5 6.4437+6 1.4746-3

NRL 1.7861 124.77

SRSS 1.2684 90.11
ABS 2.1791 133.39

The enclosed modal quantities represent the largest absolute contributor among selected nodes.
Note that the second mode provides the highest acceleration component.

MSC Nastran Solution

The modes of the structure, and the eigenvectors and element forces per unit modal deflection are computed
in the first run. The displacement and internal loads due to the response spectrum input are computed in
the restart run.

Main Index
 108 MSC Nastran Verification Guide

Comparison of Results

Theory MSC Nastran

Mode Hz u hmax u· hmax u··hmax Hz u hmax u· hmax u··hmax
4 26.655 2.35E-05 3.93E-03 6.48E-01 26.6550 2.3457E-05 3.9286E-03 6.5796E-01
5 69.617 7.63E-06 3.34E-03 1.46E+00 69.6170 7.6307E-06 3.3378E-03 1.4600E+00
6 78.709 2.35E-06 1.16E-03 5.74E-01 78.7090 2.3468E-06 1.1606E-03 5.7396E-01
7 281.127 9.11E-09 1.61E-05 2.84E-02 281.1300 9.1068E-09 1.6086E-05 2.8414E-02
8 328.144 2.80E-09 5.76E-06 1.19E-02 328.1400 2.7952E-09 5.7631E-06 1.1882E-02
9 337.648 1.66E-08 3.52E-05 7.48E-02 337.6500 1.6610E-08 3.5237E-05 7.4757E-02
10 499.944 7.79E-10 2.45E-06 7.69E-03 499.9400 7.7927E-10 2.4479E-06 7.6894E-03
11 562.488 2.29E-10 8.09E-07 2.86E-03 562.4900 2.2884E-10 8.0878E-07 2.8584E-03

Theory MSC Nastran

Mode  6x i u··6x i  M104A i M 104A i  6x i u··6x i  M104A i M 104A i
4 1.1815E+0 0.7774 3.2118E+6 7.5339E+1 1.1815+0 0.7773797 -3.2118E+06 -7.5339E+01
5 5.5595E-1 0.8117 6.3625E+6 4.8550E+1 5.56E-01 0.8116918 6.3625E+06 4.8551E+01
6 1.0234E-0 0.5874 3.9504E+6 9.2708E+0 -1.02E+00 -0.587405 -3.9504E+06 -9.2709E+00
7 6.8641E-2 1.9504E-3 1.1626E+7 1.0588E-1 6.86E-02 0.0019504 1.1626E+07 1.0587E-01
8 3.1245E-2 3.7125E-4 4.2000E+6 1.1740E-2 -3.12E-02 -0.000371 1.1626E+07 3.2496E-02
9 2.8401-E3 2.1232E-4 6.1150+E6 1.0157E-1 -2.84E-03 -0.000212 -6.1150E+06 -1.0157E-01
10 1.1122-E2 8.5520E-5 5.9823+E6 4.6618E-3 1.11E-02 8.552E-05 5.9823E+06 4.6618E-03
11 1.0800-E2 3.0870E-5 6.4437+E6 1.4746E-3 -1.08E-02 -3.09E-05 -6.4437E+06 -1.4746E-03

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 CHAPTER 2.24 109
Lateral Buckling of Cantilever Beam

2.24 Lateral Buckling of Cantilever

Beam

Problem Description
A Cantilever beam of constant cross section is loaded at the free end with a vertical shear load. The critical
buckling condition occurs when the deflected shape in the vertical plane is unstable and lateral buckling
occurs. Determine the critical buckling eigenvalues and loads for the structure.

Solution Number
SOL 105

Features Used
 Buckling Analysis

Reference
1. Stephen P. Timoshenko and James M. Gere. Theory of Elastic Stability, 2nd ed. (New York:
Mc Graw-Hill Book Co. 1961), P. 257

Modeling Techniques Used

 The model must have sufficient nodes to accurately represent the buckled shape. Six or more grids
per half wavelength are recommended.
Engineering Data
b = 0.05 in
A = 0.05 in2
I1 = 0.00417 in4
J = 4.035E-5 in4
E = 1.0E+8 lb/in2
G = 3.0E+7 lb/in2
L = 20. in
h = 1.0 in
I2 = 1.0417 E-5 in4

Main Index
 110 MSC Nastran Verification Guide

Illustrations of the Model

Figure 2.24-1 Cantilever Beam

Model File Name and Location

Model File Name Description Location

v10501.dat (Reference deck from Lateral Buckling of Cantilever Beam. tpl\verifman\oldverif
old manual is v0501.dat)

Results
Theoretical Solution

P CR =  EI   GJ - N
----------------------- (2.24-1)
2 CR
L
Lateral Buckling of a Cantilever Beam
NCR = ±2.0063, ±5.1231, ±8.2580 (Derived in the reference) (2.24-2)
PCR = ±11.27, ±28.77, ±46.37 (2.24-3)
MSC Nastran Solution
P CR
Eigenvalue  = ---------- , solved from  K – K d   u  = 0
P
Letting P = 50, then

Main Index
 CHAPTER 2.24 111
Lateral Buckling of Cantilever Beam

 = ±0.2254, ±0.5783, ±0.9402

PCR = ±11.27, ±28.91, ±47.01

Comparison of Results

Solution for PCR

Buckling Mode (lb)
st
1 2nd 3rd
Theory ±11.27 ±28.77 ±46.37
MSC Nastran ±11.27 ±28.91 ±47.01
Difference 0.0% 0.50% 1.38%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 112 MSC Nastran Verification Guide

2.25 Simple Frame Analysis with

Buckling

Problem Description
A simple frame with lateral motion constrained, is loaded vertically. Using Static analysis, find the vertical
column forces and stresses. Then release the constraint to allow lateral motion and find the lowest critical
buckling load.

Solution Number
SOL 105

Features Used
 Static and Buckling Analysis

Reference
1. Stephen P. Timoshenko and James M. Gere. Theory of Elastic Stability, 2nd ed. (New York:
Mc Graw-Hill Book Co. 1961), P. 149

Modeling Techniques Used

 Simple frame is modeled using Bar elements (CBAR).
Engineering Data
E = 1.0E+6 lb/in2
A = 1.0 in2
l = 100. in
I1 = I2 = 1.0 in4
P = 500. lb

Main Index
 CHAPTER 2.25 113
Simple Frame Analysis with Buckling

Illustrations of the Model

Figure 2.25-1 Simple Frame

Model File Name and Location

Model File Name Description Location

v10502a.dat (Reference deck from Lateral Buckling of Cantilever Beam. tpl\verifman\oldverif
old manual is v0502a.dat)

Results
Theoretical Solution
Static Analysis of Symmetric Deformation Mode
Axial force in vertical column: -500.0 lb (comp.)
Axial stress of vertical column: -500.0 lb/in2
Buckling Analysis of Antisymmetric Deformation Mode

2 PC R
k = ----------
EI
where kL = -6 tan (kL)
lowest k: k1 = 2.71645974 x 10-2  p = k2EI = 737.9 lb

Main Index
 114 MSC Nastran Verification Guide

MSC Nastran Solution

Static Analysis
Axial force in vertical elements: -500.0 lb
Axial stress of vertical elements: -500.0 lb/in2
Buckling Analysis
Mode 1 Eigenvalue
 1 = 1.475288  P CR = P Ap p l i e d   1 = 737.6 lb

Comparison of Results

Force (lb) Stress (lb/in2) PCR (lb)

Theory -500 -500 737.9
MSC Nastran -500 -500 737.6
Difference 0.0% 0.0% 0.041%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 CHAPTER 2.26 115
Beam with Gap/Lift-Off

2.26 Beam with Gap/Lift-Off

Problem Description
A simply supported beam is hinged at one end and supported by lifting rollers at two other locations.
Allowing lift-off to occur at the rollers, determine the vertical deflections under the load points.

Solution Number
SOL 106

Features Used
 Nonlinear statics

Reference
1. Jack C. McCormac, Structural Analysis, 3rd ed. (New York: Intext Educational Publishers, 1957) P.
323, ex 17.12.

Modeling Techniques Used

 The model details are given below:
Engineering Data
E = 29.0E+06 lb/in2
L1 = 15 ft
L2 = 10 ft
I = 1000 in4
P1 = 40000 lb
P2 = 10000 lb

Main Index
 116 MSC Nastran Verification Guide

Illustrations of the Model

Model File Name and Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10601s.dat V6601s.dat Beam with gap/lift-off tpl\verifman\oldverif

Results
Theoretical Solution
(Lift-off occurs at point D only)
Point B: Uy = -1.01 in
Point D: Uy = +0.546 in

MSC Nastran Solution

Gap element were used to model the roller supports at points C and D. Iteration allowed lift-off to occur at
D. Final Deflection were:
Point B: Uy = -1.009449E+00 in
Point D: Uy = 5.443316E-01 in

Conclusion
The results from MSC Nastran are comparable to the theoretical results.

Main Index
 CHAPTER 2.27 117
Cantilever Beam with Plastic Hinge

2.27 Cantilever Beam with Plastic

Hinge

Problem Description
Given a cantilever beam with a concentrated end loading, find the collapse load using plastic hinge analysis.

Solution Number
SOL 106

Features Used
 Nonlinear statics

Reference
1. A. Mendelson, PLASTICITY: Theory and Application (New York: The Macmillan Company)P.
308

Modeling Techniques Used

 The model details are given below:
Engineering Data
E = 1.0E+07
 = 0.3
ET = 0.0
 yld = 1.0E+4
A = 0.1
b = 0.05
h = 0.5
F = F i + F

Fi = 16.0
F = 0.5

Main Index
 118 MSC Nastran Verification Guide

Illustrations of the Model

Model File Name and Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10603s.dat V6603s.dat Cantilever beam with tpl\verifman\oldverif
plastic hinge

Results
Theoretical Solution
With a concentrated end load, the maximum moment occurs at the base. Collapse occurs when the moment
at the base causes the root cross section to become fully plastic.
2
Collapse Load: F = 2bh  yld  L = 25.0

MSC Nastran Solution

Using a Beam element with the elastoplastic, Von Mises criteria, the beam is loaded ( F = 0.5 ) until collapse
occurs (large tip deflection).
Collapse Load: F = 25.0

Conclusion
The results from MSC Nastran are comparable to the theoretical results.

Main Index
 CHAPTER 2.28 119
Complex Eigenvalues of Three-by-Three Matrix

2.28 Complex Eigenvalues of

Three-by-Three Matrix

Problem Description
Given the real, unsymmetric matrix [A], find its characteristic values (“eigenvalues”) and modal matrix
(“matrix of eigenvectors").

Solution Number
SOL 107

Features Used
 Direct Complex Eigenvalues

Reference
1. Paul M. DeRusso, Rob J. Roy, and Charles M. Close, State Variables for Engineers (New York: John
Wiley & Sons, 1967), ex. 4.7-2

Modeling Techniques Used

1. The Program requires at least one structural element. When using DMIG input, put in at least one
ELAS or other element.
2. The standard form of the eigenproblem, used in the reference, is  A – I   u  = 0 , where [I] is the
identity matrix. The program solves the equation [p2M+pB+K]{u} = 0, One Method to 'model' the
original equation is to input [1] for [M], let [B] be null, and input [A] for [K]. The output quantity
pi is then converted to  i units by,
2
i = –pi

Engineering Data

2 –2 3
A= 1 1 1
1 3 –1

Main Index
 120 MSC Nastran Verification Guide

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10701.dat V2801.dat Complex eigenvalues of three-by- tpl\verifman\oldverif
three matrix

Results
Theoretical Solution
Evaluate  A – I  = 0 to produce the characteristic equation  3 – 2 2 – 5 + 6 = 0 , whose roots are
tabulated below. Their validity can be determined by substitute into the characteristic equation. The reference
calculates the eigenvector matrix [M], tabulated below, by use of adjoint matrix techniques. Its validity can
be checked by back-substitution into the equation  A –  i I   M  = 0 , where {Mi} is the ith column of M.

Results First Second Third

Eigenvalues  i  -2 1 3
Roots of quadratic  pi  1.414214 (0, 1.0i) (0, 1.732051i)
Eigenvectors 11 -1 1
1 1 1
-14 1 1

Eigenvectors are arbitrary to a scaling constant. The MSC Nastran results were renormalized by hand to
obtain the results tabulated above.

Conclusion
The closed form and tabulated solutions agree exactly to all printed digits. (MSC NASTRAN outputs 6 or 7
digits depending on machine type.)

Main Index
 CHAPTER 2.29 121
Complex Roots as a Function of Gain

2.29 Complex Roots as a

Function of Gain

Problem Description
Find the root-locus of the control system shown in the figure.

Solution Number
SOL 107

Features Used
 Direct Complex Eigenvalues

Reference
1. C. J. Savant, Jr., Basic Feedback Control System Design (New York: McGraw-Hill Book Company,
Inc., 1958)

Modeling Techniques Used

1. The equation is multiplied by t. The coefficient of the p2 term is then t, and the coefficient of the p
term is unity.
2. Three uncoupled oscillators are modeled, to simulate changes in gain. Values were chosen that
correspond to under-damped, critically damped, and over-damped systems.
Engineering Data
t = 0.01 (Time constant)
Km = 5.0 (Motor constant)

Solve for AKs = 1.0, 5.0, 10.0

Main Index
 122 MSC Nastran Verification Guide

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10702.dat V2802.dat Complex roots as a function of gain tpl\verifman\oldverif

Results
Theoretical Solution
The denominator of the transfer function is,
2 1 AK m K s
p + --- p + ------------------ = 0
 

Solutions of this quadratic equation are tabulated for three values of AKs as given on page 82 of the reference.
They are:

Root Location
AKs P1 P2
1.0 -94.7 -5.3
5.0 -50.0 -50.0
10.0 -50 + j50 -50 - j50

MSC Nastran Solution

All roots agree to all places printed, when rounded off.

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 CHAPTER 2.30 123
Frequency Response of a Single Degree of Freedom System

Frequency Response of a
2.30 Single Degree of Freedom
System

Problem Description
A 3000-pound weight is attached to a platform by a spring and a damper. If the platform is moved up and
down at the system's resonant frequency with an amplitude of one inch, find the maximum amplitude of the
3000-pound weight.

Solution Number
SOL 108

Features Used
 Direct Frequency Response

Reference
1. J.P. Den Hartog, Mechanical vibrations (New York: McGraw-Hill Book Company,1956), pp. 49-55

Modeling Techniques Used

Engineering Data
K = 333.0 lb/in
C = 28.0 lb/in/sec
a0 = 1 in

3000 lb
M 1 = -----------------------------------------
32.2 ft- 12
---------------- inch
 ------------------
sec
2 1 ft

M 1 = 7.764 lb-sec2/in

n = k - = 6.5491 rad/sec = 1.042 cyc/sec

------
M1

Main Index
 124 MSC Nastran Verification Guide

Illustrations of the Model

Figure 2.30-1 Single Degree of Freedom System

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10801.dat V2601.dat Frequency response of tpl\verifman\oldverif\
single degree of system

Results
Theoretical Solution
The motion of M1 is given by equation 2.28a of the reference (p. 49)
P0  k
u = ----------------------------------------------------------------------------------------------
2 2 2 2
 1 –   n  +  2  c  cc    n 

where,
2 2
P0 =  ka 0  +  ca 0  

c c = 2 km

 = n = excitation frequency (Hz)

Solving the above equations yield,
P 0 = 380.1517 lb
P0 380.1517
Then, u = ------- = -------------------------------- = 2.0731 inch
c 28.0  6.5491 

Main Index
 CHAPTER 2.30 125
Frequency Response of a Single Degree of Freedom System

Amplitude (inches)
Theory 2.0731
MSC Nastran 2.073576
% Error 0.023%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 126 MSC Nastran Verification Guide

Transient Response of a
2.31 Rocket - Direct Time
Integration

Problem Description
A rocket is subjected to axial loading for a finite duration of time. Using axial elements in a free-free structure,
find the displacement histories of the midpoint and two ends, considering both the rigid body motion and
the structural vibration.

Solution Number
SOL 109

Features Used
 Direct Transient Response

Reference
1. J.S. Przemieniecki, Theory of Matrix structural Analysis (New York: McGraw-Hill Book Co., 1968)
P. 340

Modeling Techniques Used

 The following is details of the model which is modeled using CONROD element. The model
details are given below:
Engineering Data
E = 1.0E+4 lb/in2
A = 1.0 in2
L = 140.0 in
 = 0.1
P0 = 100.0 lb
t0 = 1.0 sec

Main Index
 CHAPTER 2.31 127
Transient Response of a Rocket - Direct Time Integration

Illustrations of the Model

Figure 2.31-1 Single Degree of Freedom System

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10901.dat V2701.dat Transient response of a rocket - direct tpl\verifman\oldverif
time integration

Results
Theoretical Solution
Using the matrix forms for the equations of motion, with zero initial conditions, the base response
for t  t 0 can be written as:

P 0 t 0  2t – t 0  P 0 3  2 1 
- ------  cos  2  t – t 0  – cos  2 t  + ------  cos  3  t – t 0  – cos  3 t  
u 1  t  = -------------------------------- + ------- -----
2AL 4A E    2 3 

where,

2 = 12E
---------- ; 3 = 48E
----------
2 2
L L

For the given engineering data, the base displacement is:

at t=2.00; u1=10.89972 in
at t=2.15; u1=11.73228 in

Main Index
 128 MSC Nastran Verification Guide

Vertical Displacement of Base,

u1
t=2.0 t=2.15
Theory 10.89972 11.73228
MSC Nastran 10.08471 11.06770
% Difference -7.477% -5.665%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 CHAPTER 2.32 129
Response Spectra of a Single Oscillator

2.32 Response Spectra of a

Single Oscillator

Problem Description
A large mass, M1, is subjected to a force doublet. A very small mass, M2, is attached to it with a spring and
damper, tuned to an undamped natural frequency of one Hz. A response spectrum is computed for the large
mass at this same frequency. A variable, v3, is connected by a multipoint constraint and is a model of the
relative displacement between the large and small mass. As the small mass-spring combination has the same
physical model as the response spectrum calculation, it should produce the same results as the inertial motion
component (variable 2) and the relative motion component (variable 3).

Solution Number
SOL 109

Features Used
 Direct Transient Response

Reference
1. Harris, Cyril M. and Crede, Charles E., Shock and Vibration Handbook, (New York: McGraw-Hill
Book Company, Inc., 1961)

Modeling Techniques Used

 The model details are given below:
Engineering Data
M1 = 1.0E+6
M2 = 1.0
k12 = 22
c12 = 0.01  2M 2  2

= 0.125664
0 = 1.0E+6

Main Index
 130 MSC Nastran Verification Guide

Illustrations of the Model

Figure 2.32-1 Single Degree of Freedom System

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10902.dat V2702.dat Response spectra of a single oscillator tpl\verifman\oldverif

Results
Magnitude of Peak Response, Inertial Coordinates

Theory MSC Nastran % Difference

Displacement 2.000315 2.0000 -0.02%
Velocity 1.000107 1.0000 -0.01%
Acceleration 1.00237 1.0019 -0.05%

Magnitude of Peak Response, Relative Coordinates

Theory MSC Nastran % Difference

Displacement 0.025406 2.5377E-02 -0.1136%
Velocity 0.051437 5.1440E-02 0.0055%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 CHAPTER 2.33 131
Transient Response with Combined Loading Functions

Transient Response with

2.33 Combined Loading
Functions

Problem Description
Three identical structures are given different time-varying loads. The loads on the third structure are a linear
combination of the loads on the other two structures. It is shown that the response of the third structure is
also the same linear combination of the response of the first two structures.

Solution Number
SOL 109

Features Used
 Direct Transient Response

Modeling Techniques Used

 The model details are given below:
Engineering Data
E = 1.0E+7 lb/in2
L = 10 in
A = 104 in2
I1 = 1.0 in4

Main Index
 132 MSC Nastran Verification Guide

Illustrations of the Model

Figure 2.33-1 Single Degree of Freedom System

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10903.dat V2703.dat Transient response with tpl\verifman\oldverif
combined loading function

Results
All three systems are sampled at t = 1.5, with their displacements tabulated below seen that the response of
the third structure is the sum of the other two.

Response at t=1.5
Mass 1 1.576766E-05
Mass 2 -1.361804E-05
sum 2.149620E-06
Mass 3 2.149617E-06
% Difference 0.000%

Conclusion
It is proved by MSC Nastran that response of the third structure is also same linear combination of the
response of the first two structures.

Main Index
 CHAPTER 2.34 133
Direct Transient Response to an Internal Pressure

2.34 Direct Transient Response

to an Internal Pressure

Problem Description
Find the peak displacements in a thick-walled cylinder resulting from a suddenly applied internal pressure
load.

Solution Number
SOL 109

Features Used
 Direct Transient Response

Reference
1. S. H. Crandall, N. C. Dahl, and T.J. Lardner, An Introduction to the Mechanics of Solids, 2nd ed.
(New York: McGraw-Hill book Company, 1972) pp. 293-297

Modeling Techniques Used

 The model is the same thick-walled cylinder given in Verification Problem No. v10110. The model
details are given below:
Engineering Data
ri = 6.0 in
ro = 12.0 in
h = 8.0 in
E = 30.0E+6 psi
P = 0.28 lb/in3
 = 0.0
p = 10.0 lb/in2

Main Index
 134 MSC Nastran Verification Guide

Illustrations of the Model

Figure 2.34-1 Thick walled Cylinder

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v10904.dat V2704.dat Direct transient response to an tpl\verifman\oldverif
internal pressure

Results
Theoretical Solution
Radial displacements from a static analysis (SOL 101) were: Grid 1, 3.326E-6 in.; Grid 7, 2.773E-6 in.; and
Grid 11, 2.673E-6 in.
MSC Nastran Solution
The force was applied in an instantaneous manner, as shown below. Several time steps of zero load were input
to ensure simulation of a step function in the loading. The integration time step was chosen to be 1.0
microsecond. Approximately three cycles of response were simulated, and damping was neglected. Note that

Main Index
 CHAPTER 2.34 135
Direct Transient Response to an Internal Pressure

for a single degree-of-freedom system, an instantaneously applied load will give a maximum displacement of
twice the static displacement.

Comparison of Results

Static Peak Transient Ratio

Node Displacement Displacement Transient/Static
1 3.33E-06 6.56238E-06 1.97
7 2.77E-06 5.65275E-06 2.04
11 2.67E-06 5.74974E-06 2.15

Conclusion
For a Single degree of freedom system, an instantaneously applied load will give a maximum displacement
of twice the static displacement.

Main Index
 136 MSC Nastran Verification Guide

Complex Eigenvalues of
2.35 Three-by-Three Matrix,
Modal Method

Problem Description
Given the same unsymmetric matrix [A] used in problem v10701, find its eigen solution in two steps.
1. Add enough terms to [A] to produce a symmetric matrix. Find the real eigensolution of this matrix
to use as modal coordinates.
2. Add in the terms to reconstruct [A], and compute its eigensolution.
Use the so-called 'particular' matrix [AB] of reference 2 as the symmetric matrix, as it has a closed-form
solution for its eigenvalues. Use the negative of this matrix, as this is more typical of the values encountered
on structures.

Solution Number
SOL 110

Features Used
 Modal Complex Eigenvalues

Reference
1. Verification Problem No. v10701
2. Marvin Marcus, 'Basic Theorems in Matrix Theory, "National Bureau of Standards Applied
Mathematics Series, 57 (1964)"

Modeling Techniques Used

1. The mass matrix is input to DMIG cards as the identity matrix.
2. The [D] matrix is input on the system after the transformations based on the real eigensolutions are
made on the symmetric portion. As there was no truncation of the modal coordinates, an exact answer
can be expected, even though the terms in [K2PP] are of the same magnitude as [AB].
Engineering Data

Main Index
 CHAPTER 2.35 137
Complex Eigenvalues of Three-by-Three Matrix, Modal Method

[A] is tabulated in v10701. [AB] has -2 on the diagonal, +1 above and below the diagonal, and zeros
elsewhere. The difference matrix [D] is then
[D] = [A] - [-AB] = [A] + [AB]

0 –1 3 2 –2 3 –2 1 0
2 –1 2 = 1 1 1 + 1 –2 1
1 4 –3 1 3 –1 0 1 –2

-[AB] is equivalent to the physical system

Illustrations of the Model

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v11001.dat V2901.dat Complex eigenvalues of three-by- tpl\verifman\oldverif
three matrix, modal method

Results
Theoretical Solution
The theory of the complex eigensolution is described in v10701. A closed form solution for the real
eigensolution to [AB] is given in Reference 2 as
2
 i = – 4 sin  i   2  N + 1   

where N is the size of the matrix.

For N = 3, the roots of [-AB] are:

 2
 1 = 4 sin  --- = 0.585786
 8

 2
 2 = 4 sin  --- = 2.0
 4

3 2
 3 = 4 sin  ------ = 3.414214
 8

Main Index
 138 MSC Nastran Verification Guide

Comparison of Results

Eigenvalues
Theory MSC Nastran % Difference
Mode 1 0.5858 0.5858 0.00%
Mode 2 2.0000 2.0000 0.00%
Mode 3 3.4142 3.4142 0.00%

Both the intermediate solution to the real eigenvalue problem, tabulated above, and the final complex eigen
solution, tabulated in v10701, agree to all places printed.

Conclusion
The MSC Nastran results are comparable with theoretical results.

Main Index
 CHAPTER 2.36 139
Modal Frequency Response of a Beam Structure

2.36 Modal Frequency Response

of a Beam Structure

Problem Description
Find the response at response of the model described in problem v10302 for six values of damping.

Solution Number
SOL 111

Features Used
 Modal Frequency Response

Reference
1. J. P. Den Hartog, Mechanical Vibrations, 4th ed. (New York: McGraw- Hill Book Company, 1956)
pp. 47-49

Modeling Techniques Used

Please refer problem v10302.
Engineering Data
The structure is described in Problem v10302. In that problem the natural frequency of each of the
structure's uncoupled modes were found. Here we apply damping to each of these modes as follows:

Natural Frequency % Critical Damping

Mode (cyc/sec)* (c/cc)
Axial (T1) 0.3183 50%
Bending - Translational (T2) 0.2757 20%
Bending - Translational (T3) 0.3898 200%
Torsion (R1) 0.3559 100%
Bending - Rotational (R2) 0.2251 2%
Bending - Rotational (R3) 0.1592 1%

Main Index
 140 MSC Nastran Verification Guide

Illustrations of the Model

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v11101.dat V3001.dat Modal frequency response of a beam tpl\verifman\oldverif
structure

Results
Theoretical Solution
The response of a single degree of freedom oscillator at resonance is given in the reference by equation 2.28a
(page 49). Simplifying that equation for our case gives,
P0  k
x = --------------- (2.36-1)
 = n 2c  c c

Where,
x = Magnitude of displacement
P0 = Magnitude of applied load
c = viscous damping coefficient
cc = viscous damping coefficient required for critical damping

Main Index
 CHAPTER 2.36 141
Modal Frequency Response of a Beam Structure

Phase angle data is plotted on page 51 of the reference. The graph shows a value of 900 for all values of
damping.
As shown in problem v0302, each degree of freedom shows different stiffness. Here we apply a load
proportional to each stiffness at its corresponding degree of freedom ( P 0 = k ).
With this assumption, equation reduces to
1 -
x - --------------
------------- = (2.36-2)
x static 2c  c c

P0
Where, x static = ------
k

We obtain for each degree of freedom:

Mode Response (x/xstatic)

Axial (T1) 1.0000
Bending - Translational (T2) 2.5000
Bending - Translational (T3) 0.2500
Torsion (R1) 0.5000
Bending - Rotational (R2) 25.000
Bending - Rotational (R3) 50.000

Comparison of Results

Frequency Mode Theory MSC Nastran % Difference

3.1831E-01 Axial (T1) 1.00E+00 1.00E+00 0.00%
2.7566E-01 Bending - Translational (T2) 5.00E+00 5.00E+00 0.00%
3.8985E-01 Bending - Translational (T3) 2.50E-01 2.50E-01 0.00%
3.5588E-01 Torsion (R1) 5.00E-01 5.00E-01 0.00%
2.2508E-01 Bending - Rotational (R2) 2.50E+01 2.50E+01 0.00%
1.5915E-01 Bending - Rotational (R3) 5.00E+01 5.00E+01 0.00%

Conclusion
The MSC Nastran response solutions agree with the above closed-form solutions to all places printed. The
MSC Nastran phase angles are also in agreement with the theoretical results, after accounting for the MSC
Nastran sign convention.

Main Index
 142 MSC Nastran Verification Guide

2.37 Transient Response of a

Beam Modal Method

Problem Description
A simply supported beam is mounted on elastic springs. Given a time-dependent symmetric distributed
loading function, find the transient response of the midspan deflections. Use the Modal method, considering
symmetric modes only.

Solution Number
SOL 112

Features Used
 Modal Transient Response

Reference
1. Walter C. Hurty and Moshe F. Rubinstein, Dynamics of Structure, (Englewood Cliffs: Prentice-Hall,
Inc., 1964), P. 294.

Modeling Techniques Used

 The model details are given below:
Engineering Data
m = 0.1 (mass per length)
E = 1.0E+07
l = 100
4 3
K =  EI  l = 974.1
 1 – 2x 2P
P  x t  = ------ --------0-  f  t  0  x  --l-
 l l 2

t0 = 1.0
P0 = 100.0

Main Index
 CHAPTER 2.37 143
Transient Response of a Beam Modal Method

Illustrations of the Model

Model File Name and Location

Reference Deck
from Old
Model Name Manual Model Description Location
v11201.dat V3101.dat Modal frequency response tpl\verifman\oldverif
of a beam structure

Results
Theoretical Solution
Using a Rayleigh-Ritz approximation of the first three symmetric modes, an equation for the center
deflection from the modal method is
3
–P0 l
 =   0 t  = -------------
4
-  2.603D 1  t  + 0.050D 2  t   (2.37-1)
 EI

where,
1
D 1  t  = -----------  sin  1  t – t 0  – sin  1 t  + cos  1  t – t 0  (2.37-2)
1 t0

1
D 2  t  = -----------  sin  2  t – t 0  – sin  2 t  + cos  2  t – t 0  (2.37-3)
2 t0

2
 EI
 1 = 0.725 -----
2
------ = 7.155 (2.37-4)
l m

2
 EI
 2 = 2.860 -----
2
------ = 28.3 (2.37-5)
l m

Local maximum midspan deflections:

Main Index
 144 MSC Nastran Verification Guide

t = 1.45 ,  = 0.23382 (2.37-6)

t = 1.9 ,  = – 0.24283 (2.37-7)

MSC Nastran Solution

Since only symmetrical modes are considered, only half of the beam is modelled. Four CBEAM elements for
the half span are used, and one CONROD is used to model the spring. The loading was shifted by 0.1 to
ensure zero initial conditions.

Local maximum midspan deflections:

t = t N – 0.1 = 1.45 ,  = 0.23418 (2.37-8)

t = t N – 0.1 = 1.90 ,  = – 0.24119 (2.37-9)

Comparison of Results

Local max,  Natural Frequency

t = 1.45 t = 1.9 1 2
Theory (Rayleigh-Ritz) 0.2338 -0.2428 7.1550 28.2300
MSC Nastran 0.2342 -0.2412 7.1808 29.0297
% Difference 0.15% -0.67% 0.36% 2.83%

Conclusion
The results from MSC Nastran are comparable to the theoretical results.

Main Index
 CHAPTER 2.38 145
Deflection of a Circular Plate, Cyclic Symmetry

2.38 Deflection of a Circular Plate,

Cyclic Symmetry

Problem Description
Find the maximum displacement of a circular plate with a uniform pressure load and fixed edges.

Solution Number
SOL 114

Features Used
 Cyclic Static

Reference
1. Fred G. Seely and James O. Smith, Advanced Mechanics of Materials, 2nd ed. (New York: John
Wiley & Sons, 1952) P. 227

Modeling Techniques Used

 The model details are given below:
Engineering Data
R = 11.0 in
T = 0.10 in
E = 10.0E+6 lb/in2
 = 0.3
p = 1.0 psi

Main Index
 146 MSC Nastran Verification Guide

Illustrations of the Model

Model File Location

Reference Deck from

Model Name Old Manual Model Description Location
v11401.dat V4701.dat Deflection of a circular tpl\verifman\oldverif
plate, cyclic symmetry

Results
Theoretical Solution
For a circular plate with the edges fixed and a uniform load
4
3 2 PR
 max = ------  1 –   ---------3- (2.38-1)
16 Et

4
3 2  1.0   11.0 
 max = ------  1 – 0.3  ------------------------------------3- (2.38-2)
16  10E+6   0.1 

 max = 0.25 in (2.38-3)

MSC Nastran Solution

Using Cyclic Symmetry, a 300 segment was modeled as shown.

Main Index
 CHAPTER 2.38 147
Deflection of a Circular Plate, Cyclic Symmetry

The resulting displacement at grid point one is z = 0.2526 in.

Comparison of Results

z (in)
Theory 0.2500
MSC Nastran 0.2526
Difference 1.03%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 148 MSC Nastran Verification Guide

2.39 Normal Modes of a Circular

Plate, Cyclic Symmetry

Problem Description
Find the natural frequencies of a free circular plate.

Solution Number
SOL 115

Features Used
 Cyclic Modes

Reference
1. S. Timoshenko, D. H. Young and W. Weaver, Jr., Vibration Problems in Engineering, 4th ed. (New
York: John Wiley & Sons, 1947) P. 501

Modeling Techniques Used

 The model details are given below:
Engineering Data
a = 11.0 in
h = 0.10 in
E = 10.0E+6 lb/in2
 = 0.3
p = 0.10 (1/g = 0.002588 sec2/in) = 0.0002588 ib-sec/in4

Main Index
 CHAPTER 2.39 149
Normal Modes of a Circular Plate, Cyclic Symmetry

Illustrations of the Model

Model File Location

Reference Deck
from Old
Model Name Manual Model Description Location
v11501.dat V4801.dat Normal modes of a circular tpl\verifman\oldverif
plate- cyclic symmetry

Results
Theoretical Solution
Frequency of vibration in rad/sec
 D
 = ----2- ------ (2.39-1)
a h

3
Eh
D = --------------------------
3
- (2.39-2)
12  1 –  

D = 915.75 (2.39-3)

Main Index
 150 MSC Nastran Verification Guide

Table of  and  in rad/sec for N nodal diameters and S nodal circles

-
---


S N=0 N=1 N=2 N=3

0 - - 5.251 12.23

(258.14) (601.23)
1 9.076 20.52 35.24 52.91

(446.18) (1008.78) (1732.43) (2601.11)

2 38.52 59.86 - -

(1893.68) (2937.86)

MSC Nastran Solution

Using Cyclic Symmetry, a 300 segment was modeled as shown.

Comparison of Results
Following table shows frequency (rad/sec) comparison of MSC Nastran and theoretical results.

S N=0 N=1 N=2 N=3

Theory 0 258.14 601.23
MSC Nastran 262.966 603.5286
% Difference 1.87% 0.38%

Main Index
 CHAPTER 2.39 151
Normal Modes of a Circular Plate, Cyclic Symmetry

S N=0 N=1 N=2 N=3

Theory 1 446.18 1008.78 1732.43 2601.11
MSC Nastran 437.708 1008.877 1688.29 2447.44
% Difference -1.90% 0.01% -2.55% -5.91%

Theory 2 1893.68 2937.86

MSC Nastran 1842.125 2862.61
% Difference -2.72% -2.56%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 152 MSC Nastran Verification Guide

Euler Buckling of a Simply

2.40 Supported Beam, Cyclic
Symmetry

Problem Description
A simply supported beam of constant cross sections is loaded at the ends with an axial force. Determine the
critical buckling load.

Solution Number
SOL 116

Features Used
 Cyclic Symmetry Buckling

Reference
1. Stephen P. Timoshenko and James M. Gere, Theory of Elastic Stability, 2nd ed. (New York:
McGrawHill Book Co., 1961) P.49.

Modeling Techniques Used

 Four equal beams are used in MSC Nastran Solution. The model details are given below:
Engineering Data
I = bh3/12 = 0.08333 in4
A = bh = 1.00 in2
L = 100 in
E = 10.0E+06 lb/in2
 = 0.3

Main Index
 CHAPTER 2.40 153
Euler Buckling of a Simply Supported Beam, Cyclic Symmetry

Illustrations of the Model

Model File Name and Location

Reference Deck
from Old
Model Name Manual Model Description Location
v11601.dat V7701s.dat Euler Buckling of a simply tpl\verifman\oldverif
supported beam

Results
Theoretical Solution
2
 EI-
P CR = -----------
2
l

2 6
  10.0 10   0.08333 -
P CR = ----------------------------------------------------------
2
100

P CR = 822.467 lb

MSC Nastran Solution

P CR
Eigenvalue  = ---------- , solved from  k – k d   u  = 0
P

P applied = 822.467

 = 0.99982

P CR = 0.99982  822.467 

P CR = 822.319 lb

Main Index
 154 MSC Nastran Verification Guide

Comparison of Results

Solutions for PCR

Theory 822.467
MSC Nastran 822.319
% Difference 0.018%

Conclusion
The results from MSC Nastran are comparable to the theoretical results.

Main Index
 CHAPTER 2.41 155
Frequency Response of a Beam-Mass Structure using Dihedral Cyclic Symmetry

Frequency Response of a
2.41 Beam-Mass Structure using
Dihedral Cyclic Symmetry

Problem Description
Find the response at resonance for the model described in v10302, modeling one-half of the model, using
cyclic symmetry analysis.

Solution Number
SOL 118

Features Used
 Cyclic Frequency Response

Reference
1. J. P. Den Hartog, Mechanical Vibrations, 4th ed. (New York: McGraw-Hill Book Company, Inc.,
1956) pp. 429-431.

Modeling Techniques Used

1. The model of V10302 is cut through the center grid point, with the plane of symmetry being normal
to the bar centerline. Only the right-hand half is modeled.
2. The only elements entirely in the plane of symmetry are the mass elements. Their values are halved,
because the mirrored component also has these values implicitly. The dihedral symmetry option is
selected.
3. All loads are applied on the plane of symmetry. Full loads are placed on the right hand side, and none
on the left side.
4. Structural damping that provides critical damping at resonance is used. This required a value of twice
the fraction of critical damping, as is input on the card PARAM, G, 2.0.
5. The loads on each degree of freedom are proportional to the stiffness in that direction, to
nondimensionalize the problem (see DAREA cards)

Main Index
 156 MSC Nastran Verification Guide

6. The default format for frequency response analysis output is the Cartesian form of the complex
number (“RE_IMAG”). By requesting (“PHASE”) on the DISP case control card a polar form
(“MAG_PHASE”) is output instead. The SORT2 option on the DISP card produces a more compact
output.
 The model details are given below:
Engineering Data
Beam properties:
A = 1.0 (Extensional Area)
I1 = 2. 0 (Bending moment of inertia in 1-3 plane)
I2 = 1.0 (Bending Moment of inertia in 1-2 plane)
J = 1.25 (Moment of inertia for Torsion)
E = 1.0 (Young's Modulus)
G = 1.0 (Shear Modulus)
L = 1.0 (Length)

Mass properties
M1 = 1.0 (Axial)
M2, M3 = 64.0
I11 = 1.0
I22, I33 = 16.0

Note: Consistent, hypothetical units are used.

Main Index
 CHAPTER 2.41 157
Frequency Response of a Beam-Mass Structure using Dihedral Cyclic Symmetry

Illustrations of the Model

Model File Name and Location

Reference Deck
from Old
Model Name Manual Model Description Location
v11801.dat V7801s.dat Frequency response of a beam-mass tpl\verifman\oldverif
structure using Dihedral cyclic symmetry

Results
Theoretical Solution
The response of a single degree of freedom oscillator at resonance is given in equation 2.28a (page 49 of the
reference) as
P0  k
X 0   =  n  = --------------- (2.41-1)
2c  c c

The phase angle between applied force and displacement is plotted on page 51 and shows a 900 value for all
values of damping. The symbols are defined as:
X0 = Magnitude of displacement
P0 = Magnitude of applied load
k = Stiffness of oscillator
c = Viscous damping coefficient

Main Index
 158 MSC Nastran Verification Guide

cc = Viscous damping required for critical damping

MSC Nastran Solution
MSC Nastran uses a different sign convention for the phase angle than does the reference. For example, in
the reference, the phase angle for a single degree of freedom system starts at a small positive angle at low
frequency, passes through 900 at resonance, and approaches 1800 at high frequency. In MSC Nastran the
Phase angle traverses the range 3600 through 2700, approaching 1800.
As shown in Problem v10302, each degree of freedom of these single degree of freedom problems has a
different stiffness. A load proportional to stiffness is used to nondimensionalize the solution. (see the
definition of the static deflection Xs on page 49 of the reference and the nondimensional quantity X/Xs.)

The expected solutions should all have the magnitude of one-half and a phase of 2700.

Conclusion
The results from MSC Nastran are comparable to the theoretical results.

Main Index
 CHAPTER 2.42 159
Stress Wave Propagation

2.42 Stress Wave Propagation

Problem Description
A rod of uniform cross section is fixed at one end and a constant force is suddenly applied to its free end. A
stress wave propagates along the length of the rod. Analyze this wave propagation phenomenon by SOL 129
for transient response analysis.

Solution Number
SOL 129

Features Used
 Transient Response

Reference
1. Stephen P. Timoshenko and J. N. Goodier, Theory of Elastic Stability, 2nd ed. (New York:
McGraw-Hill Book Co. 1951), pp. 492-496.
2. R. C. Juvinall, Engineering Consideration of Stress, Strain and Strength, McGraw-Hill, New York,
1967, pp. 185-188.

Modeling Techniques Used

 The model details are given below:
Engineering Data
L = 100.0 m (Rod length)
A = 1.0 m2, (Cross-sectional area of the rod)
E = 103 N/m2, (Young's Modulus)
 = 0.3, (Poisson’s ratio)
 = 0.1 kg/m3, (Mass density)
c = E = 100.0 m/s = L/s, (Stress wave velocity)

Main Index
 160 MSC Nastran Verification Guide

Illustrations of the Model

Model File Name and Location

Reference Deck from

Model Name Old Manual Model Description Location
v12901.dat V9901s.dat Stress wave propagation tpl\verifman\oldverif

Results
Theoretical Solution
Transient responses at two points of interest, the free end and the fixed end points are analyzed.
Due to constant force F, applied at the free end, the stress at the free end  = – F  A is constant at all times.
Before the wave front first reaches the fixed end, the stress at the fixed end is zero. As soon as the wave reaches
the fixed end, it is reflected and produces a stress of 2 at the fixed end. The stress remains unchanged until
the wave front returns to this point. The compressive wave leaves the fixed end and upon reaching the free
end it is reflected as a tensile stress wave. When this tensile wave arrives at the fixed end and is reflected there,
the stress at this point becomes zero. The zero stress is again unchanged for a complete cycle of the stress wave
motion.
The displacement-time history at the free end can be expressed, in general, as:
L
  x t 
ut =  ----------------
E
dx
0

Or, expressing ut separately for each time interval,

– FL
u  t  = ---------- t for 0  t  2
AE

FL
u  t  = u  2  + -------  t – 2  for 2  t  4
AE

etc.

Main Index
 CHAPTER 2.42 161
Stress Wave Propagation

This shows that the displacement for the free end is a linear function of time in each time interval.

MSC Nastran Solution

Using twenty CONROD elements, the rod is loaded axially at its free end with a step forcing function. The
upper bound stability limit for the time increment is given by
1
t = -------------------------
4.   n  max

Where,   n  max is the highest natural frequency in cycles/sec of all the interested modes of the corresponding
model. Introduction of small damping in the rod elements is recommended to reduce the high frequency
oscillations in the time history plots.

Comparison of Results
The stress time history plots given in Figure 2.42-1 and Figure 2.42-2 for the free end and fixed end, respectively
show that the MSC Nastran results are consistent with the theoretical ones in their global trends and
characters. The oscillation about the exact solution may be attributed to higher natural frequency modes.

Figure 2.42-1 Comparison of Free End Stress Time History

Main Index
 162 MSC Nastran Verification Guide

Figure 2.42-2 Comparison of Fixed End Stress Time History

Figure 2.42-3 Comparison of Free End Displacement Time History

Conclusion
The displacement-time plot for the free end (Figure 2.42-3) Shows a close agreement with the theoretical
solution.

Main Index
 CHAPTER 2.43 163
Impact Analysis

2.43 Impact Analysis

Problem Description
A rod with fixed end is struck by a moving mass at the other end. Let v0 be the initial velocity of the mass.
Consider the mass body as absolutely rigid. The velocity of particles at the end of the rod at the instant of
impact (t = 0) is v0. Determine the time history of stress and displacement at various cross sections of the rod
during the impact. Find the duration of the impact and the maximum stress in the rod for various values of
mass ratios.

Solution Number
SOL 129

Features Used
 Transient Response

Reference
1. Stephen P. Timoshenko and J. N. Goodier, Theory of Elastic Stability, 2nd ed. (New York:
McGraw-Hill Book Co. 1951), pp. 497-504.
2. April 1986 Application Note, Application Manual Section.

Modeling Techniques Used

 The model details are given below:
Engineering Data
L = 100.0 m (Rod length)
A = 1.0 m2, (Cross-sectional area of the rod)
E = 103 N/m2, (Young's Modulus)
 = 0.3, (Poisson’s ratio)
 = 0.1 kg/m3, (Mass density)
m = AL = 10.0 kg, (Rod mass)

Main Index
 164 MSC Nastran Verification Guide

Illustrations of the Model

Model File Name and Location

Reference Deck
from Old
Model Name Manual Model Description Location
v12902.dat V9902s.dat Impact analysis tpl\verifman\oldverif

Results
MSC Nastran Solution
The rod is modeled with twenty CONRODs and a GAP element. One end of the rod is constrained
completely and the other is left free axially. One CONM2 is used to represent the mass body, and the mass
value is varied for various ratios of M to m. A gap element connects the CONM2 and the free end of the rod
with a zero initial opening. An initial velocity of v0 = -0.1 m/sec is specified for both the CONM2 and the
free end grid point using TIC cards. A time step is chosen so that twenty vibration modes of the rod model
can be adequately represented. A small damping (0.4% at resonance) is introduced in the rod elements to
smooth out the high frequency oscillations. The consistent mass matrix option (PARAM, COUPMASS) is
recommended for more stable responses.

Comparison of Results
The instant when the stress at the free end (or force in the gap element) vanishes signifies the end of the
impact. The durations of the impact for four distinct values of M/m are determined from MSC Nastran
analyses and compared with the calculations of Saint-Venant in the following table.

Table 2.43-1 Duration of impact (sec.)

M/m 1 2 4 6
Theory 3.068 4.708 5.900 7.419
MSC Nastran 3.065 4.708 6.700* 7.420
% Difference 0.1% 0.00% 13.56% 0.01%

*The discrepancy is due to a second contact on the rebound which is not accounted for in the theory as
explained in the references.

Main Index
 CHAPTER 2.43 165
Impact Analysis

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 166 MSC Nastran Verification Guide

2.44 Flutter of a Two-Dimensional

Airfoil

Problem Description
A two-dimensional airfoil is mounted on bending and torsion springs that are aft of the aerodynamic center
as shown in the figure below. A critical velocity will cause either a static instability (torsional divergence) or
an oscillatory instability (flutter). Find both the divergence and flutter speeds for the system.

Solution Number
SOL 145

Features Used
 Flutter

Reference
1. Raymond L. Bisplinghoff, Holt Ashley, and Robert L. Halfman, Aeroelasticity (Reading: Addison-
Wesley Publishing Co., Inc., 1955) pp. 538-541

Modeling Techniques Used

 Large values of reduced frequency k may be chosen to illustrate the low speed behavior of the
frequencies and artificial damping. Moderate values of k define the flutter speed and very low values
of k lead to the divergence speed. The model details are given below:
Engineering Data
2
Mass Ratio  = m   b  = 20.0
Elastic Axis at 40% a = -0.2
Centre of Gravity at 45% x = 0.1
Aerodynamic centre at 25%
Dimensionless radius of gyration ra = 0.5
Uncoupled bending frequency h = 10 rad/sec
Uncoupled torsion frequency  = 25 rad/sec

Main Index
 CHAPTER 2.44 167
Flutter of a Two-Dimensional Airfoil

Illustrations of the Model

Model File Name and Location

Reference Deck
Model Name from Old Manual Model Description Location
v14501q.dat V7501q.dat Flutter of two dimensional Airfoil tpl\verifman\oldverif

Results
Theoretical Solution
The data correspond to Case (r) on page 538 of the reference. The divergence speed is given there as
U D  b  = 2.88 (2.44-1)

Using an airfoil chord of c = 2b = 6.0 ft., the divergence speed is

U D = 216 fps (2.44-2)

The flutter speed is found from the graph (r) on page 541 of the reference. For X  = 0.1 and frequency ratio
 h    = 0.4 , we find

U F  b  = 2.25 (2.44-3)

from which
U F = 169 fps (2.44-4)

MSC Nastran Solution

Additional dimensional data are required and derived as follows:
Sea level density  = 0.002378 slugs/cu.ft.
2
Mass m = b  = 1.3447 slugs

Main Index
 168 MSC Nastran Verification Guide

Static unbalance s = mbx  = 0.40342 slugs

Moment of inertia ia = 2 2
mb r  = 3.0256 slug-ft2
2
Bending stiffness kh = m h = 134.47 lbs/ft
2
Torsional stiffness k = mi    = 1891.0 ft-lbs/rad

The solution is carried out by the k-method using the exact Theodorsen function. The plotted output is
shown in the following figure. Interpolation of the printed output for the flutter speed at zero damping gives
VF = 166 fps; extrapolation of the printed output for the divergence speed at zero damping gives VD = 216
fps.

Comparison of Results

VF VD
Theory 169 216
MSC Nastran 166 216
% Difference 1.8% 0.0%

Conclusion
The results from MSC Nastran are comparable to the theoretical results.

Main Index
 CHAPTER 2.44 169
Flutter of a Two-Dimensional Airfoil

Main Index
 170 MSC Nastran Verification Guide

2.45 Radiation Exchange

Between Two Panels

Problem Description
Two Parallel 'grey' planes have emissivities of 0.8 and 0.7 and are maintained at 500oF and 1000oF
respectively. Assuming all energy radiated from each plate is irradiated onto the other (View factors = 1.0)
Calculate the net radiation exchange.

Solution Number
SOL 153

Features Used
 Nonlinear steady state Heat transfer

Reference
1. Alan J., Chapman, Heat Transfer, 3rd ed. (New York: Macmillan Publishing Co., Inc., 1974) P. 447,
ex. 11.4

Modeling Techniques Used

 The model details are given below:
Engineering Data
a=b = 1 ft
d = 2 ft
T1 = 5000 F
T2 = 10000 F
1 = 0.7
2 = 0.8
TABS = 460.0 0 R
 = 1.714E-9 Btu/hr-ft2-0 R4

Main Index
 CHAPTER 2.45 171
Radiation Exchange Between Two Panels

Illustrations of the Model

Model File Name and Location

Reference Deck
from Old
Model Name Manual Model Description Location
v15301s.dat V7401s.dat Radiation exchange between tpl\verifman\oldverif
two panels

Results
Theoretical Solution
The net radiation exchange between two “gray” surfaces is given by,
4 4
T1 – T2
q =  ---------------------------------------------------- (2.45-1)
 1  1  +  1  2  – 1

Solving this equation yields,

2
q = 3772.347 Btu/hr-ft

Note: This is the exact answer obtained when above values are used in the above equation. Text example
contains some round-off error.

MSC Nastran Solution

Total radiation exchange is calculated to be
q = 3772.347 Btu/hr
Since the model plates are one square foot each, the flux per square foot is
q = 3772.347 Btu/hr-ft2

Conclusion
The results from MSC Nastran are comparable to the theoretical results.

Main Index
 172 MSC Nastran Verification Guide

2.46 Transient Temperature

Distribution in a Slab

Problem Description
A 10-inch-thick slab, initially at a uniform temperature of 180o F, suddenly has its end surfaces reduced to
60o F. Find the temperature at a plane four inches from the end surface after an elapsed time of two hours.

Solution Number
SOL 159

Features Used
 Transient Heat Transfer

Reference
1. Alan J., Chapman, Heat Transfer, 3rd ed. (New York: Macmillan Publishing Co., Inc., 1974) P. 126,
ex. 4.2.

Modeling Techniques Used

 The model details are given below:
Engineering Data
a=b = 12 in
l = 10 in
k = 0.4 Btu/hr-ft-0F
 = 100.0 lbm/ft2
Cn = 0.2 Btu/lbm/0F
k-
 = --------- = 0.2 ft2 /hr
C p

Tinitial = 1800F
T ends time = 0
= 600F

Main Index
 CHAPTER 2.46 173
Transient Temperature Distribution in a Slab

Illustrations of the Model

Model File Name and Location

Reference Deck from

Model Name Old Manual Model Description Location
v15901s.dat V8901s.dat Transient temperature tpl\verifman\oldverif
distribution in a slab

Results
Theoretical Solution
For the given conditions, from a graph in the reference it can be determined that
T–T
----------------s = 0.685 (2.46-1)
Ti – Ts

where, Ti = Initial temperature distribution

Ts = Applied surface temperature at time = 0

Thus, T = (0.685)(180-60) + 60 = 142.20 0F

MSC Nastran Solution

Using HEXA elements to divide the slab into five two inch thick sections, use a time step = 0.02 hr. For a
plane four inches from the end surface, as shown in the figure, at time = 2 hours, the result is:
T = 142.05 oF

Main Index
 174 MSC Nastran Verification Guide

Comparison of Results

Temperature
Theory 142.2
MSC Nastran 142.05
% Difference -0.10549%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 CHAPTER 2.47 175
Transient Temperature Distribution in a Slab with Convective Effects

Transient Temperature
2.47 Distribution in a Slab with
Convective Effects

Problem Description
A same slab as in problem v15901s (10 inch thick, initially at 180oF) has a fluid at 60oF placed in contact
with its end surfaces at time equal 0.0. Assuming a convective film coefficient of 4.0 Btu/hr-ft-F, find the
temperature at a depth of one inch after 10 hours have passed.

Solution Number
SOL 159

Features Used
 Transient Heat Transfer

Reference
1. Alan J., Chapman, Heat Transfer, 3rd ed. (New York: Macmillan Publishing Co., Inc., 1974) P. 126,
ex. 4.3.

Modeling Techniques Used

 The model details are given below:
Engineering Data
a=b = 12 in
l = 10 in
k = 0.4 Btu/hr-ft-0F
 = 100.0 lbm/ft2
Cn = 0.2 Btu/lbm/0F
k-
 = --------- = 0.2 ft2 /hr
C p

Main Index
 176 MSC Nastran Verification Guide

Tinitial = 1800F
T ends time = 0
= 600F

Illustrations of the Model

Model File Name and Location

Reference Deck
Model Name from Old Manual Model Description Location
v15902s.dat V8902s.dat Transient temperature distribution in tpl\verifman\oldverif
a slab with convective effects

Results
Theoretical Solution
For the given conditions, it can be graphically determined that
T–T
----------------f = 0.13 (2.47-1)
Ti – Tf

where, Tf = Fluid temperature

Ti = Initial temperature distribution

Thus, T = (0.13)(180-60) + 60 = 75.6 0F

at depth of one inch after an elapsed time of 10 hours.

MSC Nastran Solution

Using HEXA elements to divide the slab into ten one inch thick sections, use a time step = 0.05 hr. For a
plane one inch from the end surface, as shown in the figure, at time = 10 hours, the result is:
T = 71.89 oF

Main Index
 CHAPTER 2.47 177
Transient Temperature Distribution in a Slab with Convective Effects

Comparison of Results

Temperature
Theory 75.6
MSC Nastran 71.89
% Difference -4.90741%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 178 MSC Nastran Verification Guide

2.48 Large Deflection of Uniformly

Loaded Plate

Problem Description
A square plate with clamped edges is loaded uniformly such that the center deflection is more than the plate
thickness. Considering large deflection (small strain theory) find the deflection of the center of the plate.

Solution Number
SOL 400

Features Used
 Geometric Nonlinearity

Reference
1. S. Timoshenko and S. Woinowsky - Krieger, Theory of plates and Shells (New York: McGraw-Hill
Book Company, Inc., 1959), P. 422.

Modeling Techniques Used

 The model details are given below:
Engineering Data
a=b = 100
h = 1
E = 2.0E+11
 = 0.3
q = 2.0E+04
3
Eh
D = --------------------------
2
= 1.83 x 1010
12  1 –  

Main Index
 CHAPTER 2.48 179
Large Deflection of Uniformly Loaded Plate

Illustrations of the Model

Model File Name and Location

Reference Deck
Model Name from Old Manual Model Description Location
V40002.dat V6401.dat Large deflection of uniformly loaded tpl\verifman\oldverif
plate

Results

Theoretical Solution

4
For qb  Dh = 109.3 , W max  h = 1.20

4
For qb  Dh = 218.6 , W max  h = 1.66

Main Index
 180 MSC Nastran Verification Guide

Comparison of Results

Wmax/h
4
qb  Dh 109.3 218.6
Theory 1.2 1.66
MSC Nastran 1.26 1.75
% Difference 5.00% 5.42%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 CHAPTER 2.49 181
Large Deflection of a Beam

2.49 Large Deflection of a Beam

Problem Description
Find the displaced position of a post-buckled column, fixed at one end with an applied axial load at other
end.

Solution Number
SOL 400

Features Used
 Geometric Nonlinearity

Reference
1. Stephen P. Timoshenko and James M. Gere, "Theory of Elastic Stability" (New York: McGraw-Hill
Book Company, Inc., 1961), P. 48.

Modeling Techniques Used

 The model details are given below:
Engineering Data
L = 12.0
A = 0.15
I1 = I2 = 2.8125E-04
E = 2.0E+07
G = 1.0E+07

Main Index
 182 MSC Nastran Verification Guide

Illustrations of the Model

Model File Name and Location

Reference Deck
from Old
Model Name Manual Model Description Location
V40003.dat V6402s.dat Large deflection of a beam tpl\verifman\oldverif

Results
Theoretical Solution
The critical Euler buckling load is found using equation 2-4 on page 48 of the reference.
2
 EI-
P CR = -----------
2
(2.49-1)
4L

2 7 –4
  2.0  10   2.8125  10 -
P CR = -------------------------------------------------------------------------
2
(2.49-2)
4  12 

P CR = 96.38286 (2.49-3)

For a load greater than P CR , the following tip rotations can be determined:

P/PCR  (deg)  (rad)

0
1.152 60 1.0472
1.293 800 1.3963

Main Index
 CHAPTER 2.49 183
Large Deflection of a Beam

MSC Nastran Solution

The column is modeled using 12 BEAMs, each of unit length. The final rotations for a converged solution
are:

R3
Subcase Load ID (rad) (deg)
14 200 1.04385 59.808
26 300 1.3919 79.750

Comparison of Results

 (deg)
P/PCR 1.152 1.293
Theory 600 800
MSC Nastran 59.808 79.750
% Difference 0.32% 0.31%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 184 MSC Nastran Verification Guide

Main Index
 Section 3: Rotordynamics Problems

Main Index
Main Index
 3.1 Flexible Rotor Critical Speed
and Mode Shape

Problem Description
Complex eigenvalue analysis for Rotors.

Solution Number
SOL 107

Features Used
 The analysis is performed in fixed coordinate systems. For the purpose of comparison, same rotor is
modeled using 1-D elements (ROTORG), axisymmetric harmonic elements (ROTORAX), and
solid elements (ROTOR).

Reference
1. Geradin M. and Kill N., (1984), “A new Approach to Finite Element Modeling of Flexible Rotors.”
Engineering Computations, Vol. 1, 52-64.

Modeling Techniques Used

 Apex/Patran was used to create the Finite Element model. The same rotors are modeled by 1D
element CBEAM for ROTORG, by 2D element CQUADX for ROTORAX, and by 3D element
CHEXA/CPENTA for ROTOR. The bearing s are modeled by CBUSH:

Main Index
 188 MSC Nastran Verification Guide

Illustrations of the Model

Model File Name and Location

Model File Name Location

rug_7_9.dat (1D); qa\nastran\tpl\verifman\rotor\
rug_7_11.dat (2D);
rug_7_12.dat (3D)

Results
Eigenvalue comparison to the paper [1]:

MSC Nastran Paper [1]

Frequencies (Hz) Frequencies (Hz)
Mode Type Beam Model Axi Model Solid Model Beam Model Axi Model
st
1 Backward 231.6 231.1 232.3 231.3 228.7
st
1 Forward 274.1 270.8 271.3 275.8 272.3
nd
2 Backward 655.4 643.1 647.5 666.6 656.4
2nd Forward 793.6 772.7 776.0 806.3 794.1

Main Index
 CHAPTER 3.1 189
Flexible Rotor Critical Speed and Mode Shape

MSC Nastran Paper [1]

Frequencies (Hz) Frequencies (Hz)
Mode Type Beam Model Axi Model Solid Model Beam Model Axi Model
Torsional 1006.1 1004.3
rd
3 Backward 1025.4 1026.6 1037.2 995.3 963.6
rd
3 Forward 1580.2 1481.9 1456.0 1543.1 1479.4

Mode shapes of solid rotor

Main Index
 190 MSC Nastran Verification Guide

Main Index
 CHAPTER 3.1 191
Flexible Rotor Critical Speed and Mode Shape

Conclusion
The results obtained here compare well with those obtained by Geradin et al. (Geradin and Kill, 1984).

Main Index
 192 MSC Nastran Verification Guide

3.2 Jeffcott Rotor Eigenvalues

Problem Description
Solve a Jeffcott rotor eigenvalues and plot Campbell Diagram.

Solution Number
SOL 107

Features Used
 Massless shaft flexible rotor with lumped mass from disk - Jeffcott Rotor
 Rigid supported bearing - Pinned at bearing location
 Campbell Diagram - Spin speed vs Natural Frequency

Reference
1. Friswell, M. I., Penny, J. E. T., Garvey, S. D. and Lees, A. W. (2010). Dynamics of rotating
machines. Cambridge University Press, P114~116.

Modeling Techniques Used

 Apex/Patran was used to create the Finite Element 1D rotor model (ROTORG): Beam element for
shaft (CBEAM), lumped mass/inertia for disk (CONM2); displacement constrains at two ends
(SPC1).

Illustrations of the Model

The flexible rotor with a single disk [1] is shown as following:
Example - A 38 mm diameter solid shaft is supported in self aligning bearings 1.1 m apart. A disk, 650 mm
in diameter and 100 mm in thickness is shrunk onto the shaft 0.8 m from one bearing. The material
properties of the shaft and disk are density = 7810 kg/m3 and modulus of elasticity E = 211 GPa.

Main Index
 CHAPTER 3.2 193
Jeffcott Rotor Eigenvalues

The finite element models used in the analysis is shown in the following figure:

Massless shaft material:

Lumped mass added to Grid 10:

Pinned ends (bearings):

Rotor define:

Main Index
 194 MSC Nastran Verification Guide

Model File Name and Location

Model File Name Location

friswell_3-9-1.dat qa\nastran\tpl\verifman\rotor\

Results
The Campbell Diagram from MSC Nastran's output is shown as:

Main Index
 CHAPTER 3.2 195
Jeffcott Rotor Eigenvalues

The analytical solution of Campbell Diagram from [1] is:

The comparison of MSC Nastran results and analytical solution from [1]:

Speed, rpm Freq, Hz Freq, Hz Freq, Hz Freq, Hz

SOL 107 0 10.32 10.32 34.59 34.59
Analytical [1] 0 10.34 10.34 34.72 34.72
Error 0.15% 0.16% 0.39% 0.38%
SOL 107 900 8.30 11.86 25.42 50.95
Analytical [1] 900 8.31 11.88 25.55 51.06
Error 0.09% 0.19% 0.51% 0.22%

Conclusion
The Jeffcott rotor eigenvalues results from Nastran SOL 107 match the analytical results. The natural
frequencies' maximum error is 0.39% at 0 rpm and 0.51% at 900 rpm.

Main Index
 196 MSC Nastran Verification Guide

Complex Eigenvalue solution

3.3 for Jeffcott rotor with
multiple disks

Problem Description
Solve a Jeffcott rotor with multiple disks for complex eigenvalues.

Solution Number
SOL 107

Features Used
 Massless shaft flexible rotor with lumped mass from disk - Jeffcott Rotor
 Rigid supported bearing - Pinned at bearing location.

Reference
1. Friswell, M. I., Penny, J. E. T., Garvey, S. D. and Lees, A. W. (2010). Dynamics of rotating
machines. Cambridge University Press, P205-206.

Modeling Techniques Used

 Apex/Patran was used to create the Finite Element 1D rotor model (ROTORG): Beam element for
shaft (CBEAM), lumped mass/inertia for disk (CONM2); displacement constrains at two ends
(SPC1).

Illustrations of the Model

The flexible rotor with a multiple disk [1] is shown as following:
Example - A 1.5m long shaft, has a diameter of 0.05m. The disks are keyed to the shaft at 0.5 and 1 m from
one end. The left disk is 0.07m thick with a diameter of 0.28m; the right disk is 0.07m thick with a diameter
of 0.35m. For the shaft, E = 211GN/m2 and G=81.2GN/m2. There is no internal shaft damping. For both
the shaft and the disks,  = 7810 kg/m3. The shaft is supported by identical bearings at its ends.

Main Index
 CHAPTER 3.3 197
Complex Eigenvalue solution for Jeffcott rotor with multiple disks

The finite element model used in the analysis is shown in the following figure.

Massless shaft material:

Pinned ends (bearings)

Main Index
 198 MSC Nastran Verification Guide

Lumped mass added to Grids 7 & 10:

Rotor define

Campbell define:

Model File Name and Location

Model File Name Description Location

Friswell_5-9-1.dat Jeffcott Rotor with Multiple Disks \verifman\rotor\

Results
The comparison of MSC Nastran results and analytical solution from [1]:

Speed,
Mode Type rpm Freq, Hz Freq, Hz Freq, Hz Freq, Hz Freq, Hz Freq, Hz
SOL107 0 13.65 13.65 43.26 43.26 113.53 113.53
Analytical [1] 0 13.79 13.79 43.66 43.66 114.08 114.08
Error 1.02% 1.02% 0.92% 0.92% 0.48% 0.48%
SOL107 4000 13.45 13.83 39.75 46.44 95.24 130.63
Analytical [1] 4000 13.59 13.97 40.07 46.90 95.52 131.63
Error 1.03% 1.00% 0.80% 0.80% 0.29% 0.76%

Main Index
 CHAPTER 3.3 199
Complex Eigenvalue solution for Jeffcott rotor with multiple disks

Conclusion
The Jeffcott rotor eigenvalues results from Nastran SOL 107 match the analytical results. The natural
frequencies' maximum error is 1.02% at 0 rpm and 1.03% at 4000 rpm.

Main Index
 200 MSC Nastran Verification Guide

Complex Eigenvalue
3.4 Solution for Jeffcott Rotor
with Overhung Disk and
Anisotropic Bearings

Problem Description
Solve a Jeffcott rotor with overhung disk supported by anisotropic bearings for complex eigenvalues.

Solution Number
SOL 107

Features Used
 Massless shaft flexible rotor with lumped mass from disk - Jeffcott Rotor
 Rigid anisotropic bearing - Pinned at bearing location.

Reference
1. Friswell, M. I., Penny, J. E. T., Garvey, S. D. and Lees, A. W. (2010). Dynamics of rotating
machines. Cambridge University Press, P278-279.

Modeling Techniques Used

 Apex/Patran was used to create the Finite Element 1D rotor model (ROTORG): Beam element for
shaft (CBEAM), lumped mass/inertia for disk (CONM2); displacement constrains at two ends
(SPC1).

Illustrations of the Model

The flexible rotor with a multiple disk [1] is shown as following:
Example - A 1.5m long shaft, has a diameter of 25mm. Shaft is supported with bearings at 0.0m & 1.0m.
The bearings are short in that they present insignificant angular stiffness to the shaft, but they present finite
translational stiffness. A disk is overhung is 250mm is diameter and 40mm thick. The shaft and disk are made
of steel, and a mass density = 7810 kg/m3, modulus of elasticity E = 211GN/m2 and Poisson's ratio of 0.3
are assumed.

Main Index
 CHAPTER 3.4 201
Complex Eigenvalue Solution for Jeffcott Rotor with Overhung Disk and Anisotropic Bearings

The finite element model used in the analysis is shown in the following figure.

Several different variants of this system are considered in which the difference line in the bearing (and bearing
support) properties as shown in below table.

Stiffness (NM/m) Damping (kNs/m)

Case Kxx Kyy Cxx Cyy
1 10 10 0 0
2 10 20 0 0
3 10 20 60 60
4 10 20 400 400

Main Index
 202 MSC Nastran Verification Guide

Different bearing setups are as shown:

Main Index
 CHAPTER 3.4 203
Complex Eigenvalue Solution for Jeffcott Rotor with Overhung Disk and Anisotropic Bearings

Massless shaft material

Lumped mass added to Grids 7 & 10

Pinned ends (bearings)

Rotor define

Campbell define:

Model File Name and Location

Model File Name Description Location

Friswell_6-8-3.dat Jeffcott Rotor with Overhung Disk and \verifman\rotor\
Anisotropic Bearing Supports

Results
The comparison of MSC Nastran results and analytical solution from [1]:

Speed Bearing Setup Pair of Roots (rad/s)

rev/mi
n Left Right First Second Third Fourth
0 SOL107 1 1 0 ± 44.15j 0 ± 44.15j 0 ± 397.36j 0 ± 397.36j
0 Analytical [1] 1 1 0 ± 44.28j 0 ± 44.28j 0 ± 397.6j 0 ± 397.6j

Main Index
 204 MSC Nastran Verification Guide

Speed Bearing Setup Pair of Roots (rad/s)

rev/mi
n Left Right First Second Third Fourth
0 SOL107 2 2 0 ± 44.15j 0 ± 44.35j 0 ± 397.36j 0 ± 399.86j
0 Analytical [1] 2 2 0 ± 44.28j 0 ± 44.36j 0 ± 397.6j 0 ± 400.2j

0 SOL107 2 3 -0.037 ± 44.16j -0.0099 ± 44.35j -1.219 ± 400.4j -0.850 ± 400.9j

0 Analytical [1] 2 3 -0.038 ± 44.29j -0.010 ± 44.36j -1.220 ± 400.6j -0.851 ± 401.2j

0 SOL107 2 4 -0.065 ± 44.26j -0.038 ± 44.26j -0.211 ± 400.8j -0.208 ± 401.5j

0 Analytical [1] 2 4 -0.066 ± 44.39j -0.038 ± 44.40j -0.211 ± 401.1j -0.209 ± 401.9j

1000 SOL107 1 1 0 ± 42.10j 0 ± 46.21j 0 ± 389.8j 0 ± 402.1j

1000 Analytical [1] 1 1 0 ± 42.22j 0 ± 46.35j 0 ± 389.9j 0 ± 402.3j

1000 SOL107 2 3 -0.020 ± 42.13j -0.028 ± 46.30j -0.625 ± 391.9j -1.352 ± 406.3j
1000 Analytical [1] 2 3 -0.020 ± 42.26j -0.028 ± 46.40j -0.619 ± 392.0j -1.358 ± 406.6j

2000 SOL107 2 3 -0.016 ± 40.10j -0.033 ± 48.33j -0.201 ± 378.0j -1.595 ± 410.1j
2000 Analytical [1] 2 3 -0.017 ± 40.22j -0.033 ± 48.47j -0.190 ± 377.9j -1.605 ± 410.3j

3000 SOL107 2 3 -0.137 ± 38.12j -0.038 ± 50.37j -0.0001 ± 357.8j -1.774 ± 412.8j
3000 Analytical [1] 2 3 -0.139 ± 38.23j -0.039 ± 50.51j -0.0004 ± 357.3j -1.786 ± 413.0j

Conclusion
The Jeffcott rotor eigenvalues results from Nastran SOL 107 match the analytical results for different bearing
stiffness & damping settings.

Main Index
 Section 4: SOL 200 Optimization
Problems

Main Index
Main Index
 4.1 Optimization of a Three Bar
Truss

Problem Description
Optimization of a three bar truss.

Solution Number
SOL 200

Features Used
 Sizing optimization with design constraints on displacements and stresses in the simple structure
under two static load conditions.

Reference
1. DOT-Design Optimization Tools, User's Manual ,Vanderplaats Research and Development,
Colorado Springs, CO, 1999.

Modeling Techniques Used

 The simple model can be constructed using bulk data entries directly, or from within any
preprocessor such as PATRAN. Patran can also create the design task as outlined in Chapter 8 of the
MSC Nastran Design Sensitivity and Optimization User's Guide (DSOUG). Figure 4.1-1 shows the
finite element model which is made up of three rod elements loaded statically with two load cases.
The design task is to minimize the weight of the structure while limiting the displacement in the y
direction at node 4 to be between -0.2 ≤≤ 0.2 inches and the stress in the bars to be
-15000 ≤≤20000 psi. The loading for example of this manual differ from that of the DSOUG in
that the 20000 lb loads are applied at a 45° angle.

Main Index
 208 MSC Nastran Verification Guide

Illustrations of the Model

10” 10”

1 2 3
X
3
1
A3 10”
A1
2

20,000 lbs (Subcase 2) 20,000 lbs (Subcase 1)

Figure 4.1-1 Three-Bar Truss

Model File Name and Location

Model File Name Location

vandy.dat tpl\verifman\SOL200

Results
The table below shows a comparison between the MSC Nastran result and the result in the reference.

Term Reference Result MSC Nastran Results

Weight 2.633 2.636
Design Variable 1 0.799 0.771
Design Variable 2 0.372 0.456
Design Variable 3 0.799 0.771
Maximum Constraint .0028 .0028
Number of Function Evaluations 29 5

Main Index
 CHAPTER 4.1 209
Optimization of a Three Bar Truss

Conclusion
Slight differences are observed between the MSC Nastran result and the Referenced Results, which is typical
for the nonlinear programming tasks involved with optimization. MSC Nastran employs sensitivities to
make the optimization search process more efficient so that a converged result is achieved in 4 design cycles
(5 function evaluations) while the reference uses gradients computed using finite differences, and therefore,
many more function evaluations.

Main Index
 210 MSC Nastran Verification Guide

4.2 Design with Constraint on

Minimum Frequency

Problem Description
Optimization of a cantilevered structure with a constraint on the first natural frequency.

Solution Number
SOL 200

Features Used
 Sizing optimization with a design constraint on a natural frequency.

Reference
1. M.J. Turner "Design of Minimum Mass Structures with Specified Natural Frequencies," AIAA
Journal , Vol 5, No. 3, March 1967.

Modeling Techniques Used

 The simple model can be constructed using bulk data entries directly, or from within any
preprocessor such as PATRAN. Patran can also create the design task as outlined in Chapter 8 of the
MSC Nastran Design Sensitivity and Optimization User's Guide (DSOUG). The finite element model
is made up of three shear webs with rod elements acting as spar caps on the top and bottom of the
webs. Six CONM2 elements provide nonstructural mass and act to make the design task meaningful
(without these masses, the structural mass and stiffness would be proportional and the design task
would lead to design variables reaching their lower bounds). The design task is to minimize the
weight of the structure while ensuring that the first natural frequency is greater than 20 Hz.
Analysis Model
• Web is modeled by QUAD4 elements
• Caps are modeled by ROD elements
• Material: E = 1.03E7 psi
• Poisson ratio = 0.3
• Nonstructural mass: Lumped masses at top and bottom nodes, 15 lbs each.

Main Index
 CHAPTER 4.2 211
Design with Constraint on Minimum Frequency

Illustrations of the Model

Z
A1 A2 A3

6” t1 t2 t3 X

20” 20” 20”

Figure 4.2-1 Cantilever Beam Vibration Model

Model File Name and Location

Model File Name Location

dsoug2.dat tpl\verifman\SOL200

Results
The referenced analysis is from 1967 and used innovative techniques of the time to attack the design task.
The webs of the structure were taken to be piecewise linear with the design variables the thickness at four
spanwise stations. MSC Nastran considered all the elements to be constant thickness so there are three design
variables for the web and another three for the caps (top and bottom caps are taken to be equal). The table
below averages the web thickness from the two ends to provide a direct comparison.
The table below shows a comparison between the MSC Nastran result and the result in the reference.

Term Reference Result MSC Nastran Results

Weight 6.91 6.97
Web 1 0.91 0.771
Web 2 0.485 0.456
Web 3 0.037 0.771
Cap 1 0.037 0.045
Cap 2 0.034 0.040
Cap 3 0.023 0.025
Maximum Constraint 0.0 .0026

Main Index
 212 MSC Nastran Verification Guide

Conclusion
It is seen that the two methods produce the same basic design but there are differences between them. There
are significant differences in the modeling techniques used and the algorithms employed, so the strong
qualitative agreement is considered acceptable.

Main Index
 CHAPTER 4.3 213
Shape Design of a Culvert

4.3 Shape Design of a Culvert

Problem Description
Optimization of a Culvert Structure using Shape Variables to provide a least weight design with design
constraints on element stresses.

Solution Number
SOL 200

Features Used
 Shape optimization with constraints on the stress in the elements.

Reference
1. Belegundu, A.D. and Rajan, S.P., "Shape Optimal Design Using Isoparametric Elements,”
Proceedings, 29th AIAA/ASME/ASEE/AMS/ASC Structures, Dynamics and Materials Conference,
pp. 696-701, Williamsburg, VA, April 1988.

Modeling Techniques Used

 The 2-d culvert model is meshed using 27 CQUAD4 elements. A pressure load is applied to the
“road” surface.
The simple model can be constructed using bulk data entries directly, or from within any
preprocessor such as PATRAN. Figure 4.3-1 shows the finite element model. The design task is to
find the shape that minimizes the volume of the structure while modifying the shape of the initially
circular hole. Grid points on the plane of symmetry can move in the y-direction while those at the
base can move in the x-direction. A constraint is placed on the maximum allowable von Mises stress
≤31000 psi. A separate “auxiliary” model is used to create the candidate shape vectors that deforms
the structure while separately applying a radial load to each of the seven grids on the cutout of the
culvert. Stiff beams are added to cutout of this auxiliary model to create smooth deformation due to
the applied point loads.

Main Index
 214 MSC Nastran Verification Guide

Illustrations of the Model

Figure 4.3-1 Initial Culvert Design

Model File Name and Location

Model File Name Description Location

Dsoug5a.dat Auxiliary model for Culvert Example \verifman\SOL200
Dsoug5.dat Shape optimization of culvert \verifman\SOL200

Results
The final design, including stress contours, is shown in Figure 4.3-2. An exact comparison with the reference
cannot be made because
a. the number of elements used to mesh the surface differs between the referenced paper and the
current analysis and
b. the reference does not contain data (such as grid locations) that can be compared directly.

Main Index
 CHAPTER 4.3 215
Shape Design of a Culvert

However there is good qualitative agreement in that the reference indicates the final design has reduced the
weight by 19% and MSC Nastran shows an 18.9% reduction.

Figure 4.3-2 Final Culvert Design

Main Index
 216 MSC Nastran Verification Guide

4.4 Topology Design of the MBB

Beam

Problem Description
A classic problem of topology optimization that seeks the optimal topology design of a simply supported
beam under a point load.

Solution Number
SOL 200

Features Used
 Topology optimization with constraints on the weight of the structure and minimal compliance as
the design objective

Reference
1. Sigmund Ole, Morphology-Based Black and White Filters for Topology Optimization. Structural
and Multidisciplinary Optimization, 33(4-5) 401-424.https://doi.org/10.1007/s-158-006-0087-x.

Modeling Techniques Used

 This MBB example is widely used by academic and industrial researchers for Topology
Optimization validation. A mesh of 4800 CQUAD4 elements represents the structure while a point
load is applied vertically at the top left side of the model. The literature contains numerous
representations of this model with differing dimensions and loading. However, they all have the
same aspect ratio for the beam and apply the load at the same location and direction so that the
topology results can be compared qualitatively regardless of the dimensions/loading/materials.
Design Model Description
Objective: Minimize compliance
Topology design region: PSHELL
Constraints: Mass target = 0.5 (i.e. mass savings 50%)
Minimum member size control
The simple model can be constructed from within any preprocessor such as PATRAN. The design
model information is very terse in the case of topology optimization and can be generated manually
or through the use of Patran.

Main Index
 CHAPTER 4.4 217
Topology Design of the MBB Beam

Illustrations of the Model

Due to symmetry, a half span model is appropriate as shown in Figure 4.4-1

Figure 4.4-1 Creation and Selection of Subcase

Model File Name and Location

Model File Name Location

mbb.dat tpl\verifman\SOL200

Results
The final design is shown in Figure 4.4-2. The minimum thickness, which applies to the members generated
by the design, is very important in constructing this figure as it prevents the possibility of very small
members. An exact comparison with the reference cannot be made because of differences in dimensions in
the two studies and wide number of variations applied to the design task. Nonetheless, many of the designs
of the reference show a similar final design with the result of Figure 5g in the reference very similar to the
one shown here.

Main Index
 218 MSC Nastran Verification Guide

Figure 4.4-2 MBB Beam Topology Design with “Minimum Member Size”

Main Index
 Section 5: SOL 700 Explicit Dynamics
Problems

Main Index
Main Index
 5.1 Impulsively Loaded Strip

Problem Description
An aluminum strip clamped at its left and right edges is subjected to an impulsive initial velocity (vz = -132
m/sec) over its central portion (one-fourth of the length).
The purpose is to investigate the sensitivity of the response to Poisson's ratio ( =0.01,  = 0.3) and to check
results against data available from experiments performed at the Air Freight Flight Dynamics Laboratory.
This problem is also presented in the Demonstration Problem Manual- Explicit Nonlinear (SOL 700).

Solution Number
SOL 700

Features Used
 Explicit dynamics.
 Elastic-plastic material.
 Impact (initial velocity)

Reference
1. Balmer, H. A. and Witmer, E. A. “Theoretical-experimental Correlation of Large Dynamic and
Permanent Deformation of Impulsively Loaded Simple Structure,” 1964, Air Force Flight Dynamics
Laboratory report FDRTDR-64-108.

Modeling Techniques Used

Due to the symmetry of the problem, only half of the strip needs to be modeled (see Figure 5.1-1). The left
hand half strip is discretized by a regular mesh of (30x3) quadrilateral elements.
The MAT1 and MATEP entries are used to input the aluminum elastic-plastic material data.
Since plastic deformations occur, the PSHELL1 entry is used to specify five integration points across the
thickness. The PSHELL1 entry defaults to three integration points.
The SPC entry is used to impose the zero displacement/rotations of the damped left-hand edge and zero out
of-plane displacement/rotations of the right-hand edge (symmetry plane).

Main Index
 222 MSC Nastran Verification Guide

To better represent the actual experimental conditions, the initial velocity distribution is slightly modified so
as to have a smooth transition to the remainder of the strip. The TIC1 entry is used to input initial velocities.
Using the PARAM, INISTEP entry, the initial time step is set to 0.1e-6 sec according to the COURANT
criterion.

length L = 0.254m
width W = 0.0305m
thickness t = 0.00318m
density 2791kg/m3
Young’s modulus E = 7.17  1010N/m2
Poisson’s ratio  = 0.3
yield stress y = 2.854  108N/m2
hardening modulus Eh = 0N/m2

The ENDTIME Case Control entry is used to follow the dynamic behavior in the range up to 0.001 seconds.
The Case Control entries TYPE, SAVE, GPOUT, GRIDS, and TIMES are used to build the z displacement
time history of Grid Point 31 (in the plane of symmetry) by saving results every 0.01 msec. The experimental
results are available for every 0.1 msec.

Illustrations of the Model

Figure 5.1-1 Numerical Model Layout

Main Index
 CHAPTER 5.1 223
Impulsively Loaded Strip

Model File Name and Location

Model File Name Description Location

Impulse_a.dat  = 0.01 Verif\sol700\impulsly_lded_strip
doc\SOL700_input_files\Structural_Dynamics\impulsly_lded_strip
Impulse_b.dat  = 0.3 Verif\SOL700\impulsly_lded_strip
doc\SOL700_input_files\Structural_Dynamics\impulsly_lded_strip

where the input file is in tpl directory.

Results

Figure 5.1-2 Vertical Z-displacement Time History of the (Free) Corner Grid Point (31) of the Half Strip

Conclusion
As viewed in the above figure, the analytical results agree well with the test results.

Main Index
 224 MSC Nastran Verification Guide

5.2 Basic Blast Analysis

Problem Description
The effect of a detonation on the environment can be simulated by assuming that the detonated material can
be idealized by a sphere of hot gas with a homogeneous density and specific internal energy. This approach
is suited for problems in which the processes inside of the explosive material are not to be investigated.
In this problem, the propagation of a blast wave will be simulated starting from the initial shock front radius
R0=0.05 m at the time t = 0 sec until it reaches a radius of R = 10 R0. Both, the gas in the sphere and the
surrounding environment behave as an ideal gas (Gas Constant = 1.4).

Solution Number
SOL 700

Features Used
 Transient explicit dynamic analysis.
 8-node hex elements for Euler
 Two regions (high and low pressure) are defined with transient initial values

Reference
1. Baker, W. E., Explosions in Air, University of Texas Press, 1973, Austin and London.

Modeling Techniques Used

The motion of the gas is radial. Therefore, only a part of the area of interest need to be analyzed. Here, a mesh
of 20 x 20 x 20 hexahedral Eulerian elements has been used for investigating the volume (0 < x < 0.3 m, 0 <
y < 0.5 m, 0 < z < 0.5 m). The center of the blast wave is located at the origin of the global coordinate system.
A higher level has to be used to indicate the higher priority of the spherical condition. The value for the
parameter INISTEP has to be below the minimal time step that follows from the Courant Criterion:
l
t = S ----- (5.2-1)
c0

where l denotes the smallest element dimension, c0 the initial speed of sound, and S is a safety factor Dytran’s
default value is 2/3).

Main Index
 CHAPTER 5.2 225
Basic Blast Analysis

With l = 10R 0  20 = 0.025 m and c 0 =    – 1 e 1 Equation (5.2-1) yields for the time step
t = 1.96E-7 second. Therefore, the value 1.96E-7 sec is used for the INISTEP parameter.

Figure 5.2-1 Radial Pressure Distribution for Several Time Steps and Theoretical Maximum Pressure

Illustrations of the Model

Figure 5.2-2 SOL 700 CTETRA Model - Impact of Cylinder - Quarter Model

Main Index
 226 MSC Nastran Verification Guide

Model File Name and Location

Model File Name Description Location

blast_sol700.dat  =0.1 Verifman\sol700\Basic_Blast

doc\SOL700_input_files\Fluid_Dynamics\Basic_Blast

Results
Figure 5.2-3 shows pressure profiles in the elements 1 to 20, which are located along the edge y = z = 0 m of
the control volume. The theoretical value of the shock front pressure is also included in this plot. Though
this mesh is very coarse, the analysis results is a fairly good approximation of the theoretical values. Note that
at t = 0, element 2 is not completely inside of the sphere of high energy gas. Therefore, its pressure is below
that of element 1 at t = 0 sec.

Conclusion

Figure 5.2-3 Pressure Contour Plot for Step 50

The contour plot of the pressure in Figure 5.2-3 shows an almost spherical shape of the shock front. Deviations
are due to the fact that mass transport takes place along element faces only. Thus, the relief of pressure is
hampered for the elements on the diagonal.

Main Index
 CHAPTER 5.3 227
JWL Blast Analysis

5.3 JWL Blast Analysis

Problem Description
This example demonstrates detonation of high explosives.

Solution Number
Transient explicit dynamic analysis (SOL 700)

Features Used
 Detonation of High Explosives (HE)
 Hydrodynamic Theory of Steady-state Plane Detonation
 Steady-state Detonation Modeling with SOL 700

Modeling Techniques Used

Due to symmetry, the reaction zone (detonation front) is a plane traveling along the length of the slab.
Therefore, it is sufficient to model only a portion of the slab with all boundary faces closed to transport.
A mesh of 200 elements along the 50 cm of slab length is used (element thickness of 0.25 cm).

Figure 5.3-1 Geometrical Layout of Explosive and Detonation Point

At the start time, all the elements are filled with explosive material. Therefore, they all reference a EOSJWL
equation of state defined by a DMAT entry.
The specific chemical energy q0 is assigned as the initial specific internal energy of the explosive by using the
TICEL entry.

Main Index
 228 MSC Nastran Verification Guide

The Chapman-Jouguet detonation velocity Dcj, the ignition point, and the ignition time are specified by the
DETSPH entry. The ignition time is taken as the start time of the analysis and the ignition point is the center
of the left face of the mesh. See Figure 5.3-1.
The duration of the analysis is set to 60 sec necessary to burn the slab (50 cm) at a detonation speed of 7980
m/sec.
Edits of pressure profiles are requested every 10 sec.
Boundary conditions: Two regions (high and low pressure) are defined with transient initial values
Element type: 8-node hex elements for Euler
FE results: Pressure distribution plot

Illustrations of the Model

Model File Name and Location

Model File Name Description Location

ep3_3.dat  =0.1 Verifman\sol700\JWL_Blast\ep3_3.dat

doc\SOL700_input_files\Fluid_Dynamics\JWL_Blast\ep3_3.dat

Results
The solution of the detonation analysis is shown in Figure 5.3-2.
The peak pressure of each profile corresponding to the pressure behind the detonation front is depicted in
Figure 5.3-3 in terms of the Chapman-Jouguet pressure fraction.
From the hydrodynamic theory of steady-state plane detonation for a -law gas, the CJ pressure is

Main Index
 CHAPTER 5.3 229
JWL Blast Analysis

P cj = 2   c j – 1 q 0  0

with cj = 2.706 for TNT, this leads to

10
P cj = 2.9 * 10 Pa

The code needs about 60 elements to build up the detonation front at which time the pressure reaches
approximately 0.85 of Pcj. Subsequently, the front propagates with only a small increase of pressure.

Figure 5.3-2 Peak Pressure Profiles at Different Distances

Figure 5.3-3 Pressure Distribution Plot

Conclusion
These results are acceptable if the fact is taken into account that SOL 700 is a first-order code that smears
the shock front over a number of elements (always conserving momentum and energy). The result of that
will be a reduction in peak pressure.

Main Index
 230 MSC Nastran Verification Guide

5.4 Taylor Test (a rod impacted

against a rigid wall)

Problem Description
The Taylor “bar test” is an important laboratory test in the science of ballistics. It enables to determine an
average value of the dynamic yield stress of a material. It consists of accelerating a cylindrical bar (the velocity
being parallel to the axis of symmetry), and then let the bar hit a rigid target. As a result, the bar shortens and
the impact side expands radially acquiring a mushroom like shape.
Much research work has been done on these impact tests. In this example, the experimental work done by
Johnson/Cook is validated with a number of simulations in SOL 700 with different element types. On top
of this work, the results are compared against the theoretical solution developed by Taylor.
Johnson and Cook developed a material model that represents a constitutive model for materials subjected
to large strains, high strain rates and high temperatures. The SOL 700 implementation is validated for a set
of constitutive constants presented in Reference 1.

Solution Number
SOL 700

Features Used
A list of the functionalities used in this problem.

References
1. Johnson, G.R. and Hook, W.H. “A constitutive model and data for metals subjected to large strains,
high strain rates and high temperatures”, April 1983, 7th Ballistic Symposium, The Hague, The
Netherlands
2. Wilkins, L.M and Guinan M.W. “Impact of cylinders on a rigid boundary”, August 1972, Lawrence
Livermore Laboratory

Modeling Techniques Used

Johnson-Cook Material Model
This material model is described in the Dytran Reference and User manuals. For more detailed information
and data sets for various materials, see Reference 1.

Main Index
 CHAPTER 5.4 231
Taylor Test (a rod impacted against a rigid wall)

Theoretical Approach - Taylor

Taylor developed a simple expression for the final length of the bar, Lf, as function of the initial length L0,
the impact velocity V, the yield strength y and the density 
2
L f L 0 = exp  – V /2 y 

The assumptions made in this expression are:

 There is a stationary yield front near the wall.
 A quasi-steady process is assumed.

Theoretical - Taylor
In Figure 5.4-9, Figure 5.4-10, and Figure 5.4-11 the theoretical reduction in length according to Taylor is
plotted:
Lf /L0 = 0.78
This theoretical reduction is based on an assumption of constant yield of 700 MPa.

Figure 5.4-1 Schematic Overview of the Taylor Impact Test

In this Taylor test, the process stops when all the kinetic energy has converted to plastic work. It is interesting
to note that the governing parameters of the process give a non-dimensional quantity, therefore, the result
depends on this combination and not on the specific value of each parameter.

SOL 700 Model

A cylindrical rod with a length of 25.4 mm and a diameter of 3.82 impacts a rigid wall with a velocity of 190
mm/s. The material data:
  = 8.96e-9 tonne/mm3
 Bulk Modulus K = 143e3 MPa
 Shear Modulus G = 47.7e3 MPa
 Two different material models have been used:
• Johnson/Cook
• Constant von Mises: y = 600 MPa (elastic-rigid-plastic)
The interface between the wall and the rod are assumed frictionless.

Main Index
 232 MSC Nastran Verification Guide

Illustrations of the Model

Figure 5.4-2 SOL 700 CHEXA Model - Impact of Cylinder - Quarter Model

Figure 5.4-3 SOL 700 CTETRA Model - Impact of Cylinder - Quarter Model

Main Index
 CHAPTER 5.4 233
Taylor Test (a rod impacted against a rigid wall)

Model File Name and Location

Model File Name Description Location

sol700_hex2x2 INT=2, Verif\sol700\ Taylor_test_rod_wall\Hexa2x2x2
x2-full.dat ISOP=FULL
doc\SOL700_input_files\Structural_Dynamics\Taylor_test_rod_wal
and l\Hexa2x2x2
hexs-
2x2x2full.bdf
sol700_Hexs- INT=3 Verif\sol700\ Taylor_test_rod_wall\Hexs-EPM-default-JC
JC_barimpact.
dat doc\SOL700_input_files\Structural_Dynamics\Taylor_test_rod_wal
l\Hexs-EPM-default-JC
and
hexs-jc.bdf
sol700_Hexs- INT=2 Verif\sol700\ Taylor_test_rod_wall\Hexs-EPM-default-JC
VM_2.dat
doc\SOL700_input_files\Structural_Dynamics\Taylor_test_rod_wal
and l\Hexs-EPM-default-JC
hexs-jc.bdf

Results
The deformed shape is compared in Figure 5.4-4 through .

Figure 5.4-4 SOL 700 Simulation - Johnson Cook Material Model with CHEXA - One Point Gauss

Main Index
 234 MSC Nastran Verification Guide

Figure 5.4-5 SOL 700 Simulation - von Mises Material Model with CHEXA - One Point Gauss

Figure 5.4-6 SOL 700 Simulation - von Mises Material Model with CHEXA - 2x2x2

Figure 5.4-7 SOL 700 Simulation - von Mises Material Model with CTETRA - Old Element

Main Index
 CHAPTER 5.4 235
Taylor Test (a rod impacted against a rigid wall)

Figure 5.4-8 SOL 700 Simulation - von Mises Material Model with CTETRA - New Element

Figure 5.4-9 Comparison of SOL 700 CHEXA Elements

Main Index
 236 MSC Nastran Verification Guide

Figure 5.4-10 Comparison of Different Element Types in SOL 700

Figure 5.4-11 Detailed Comparison of Different Element Types in SOL 700

Conclusion
In Figure 5.4-11, a detailed section of Figure 5.4-10 is shown. The results all show an oscillating behavior that is
common in all transient dynamic events and is physical. Furthermore, the new 2x2x2 CHEXA element is
deviating from the experimental result. and further developments are targeted to improve the behavior of the
new CHEXA element in impact events involving metal plasticity. The 2x2x2 CHEXA element has
specifically been implemented and tested to reduce the hour-glassing phenomenon in rubber and foam
modeling.

Main Index
 Section 6: NAFEMS Problems

Main Index
Main Index
 Linear Elastic

Main Index
 242 MSC Nastran Verification Guide

6.1 LE1: Elliptic Membrane

Problem Description
Find the tangential edge stress at D (refer below figure) of elliptical membrane which is uniformly loaded with
outward pressure.

Solution Number
SOL 101

Features Used
 Linear Static Analysis.

Reference
1. Test LE1 from NAFEMS Publication NNB, Rev. 3, NAFEMS Linear Benchmarks, 5 Oct 1990.

Modeling Techniques Used

Elliptical membrane is modeled using plane stress elements CQUAD4. GPSTRESS is used to find out nodal
stress output.

Loading
Uniform outward pressure of 10MPa at outer edge BC, Inner curved edge AD unloaded.

Boundary Condition
Edge AB, symmetry about Y axis, e.g. zero x displacement
Edge CD, symmetry about X axis, e.g. zero y displacement.

Main Index
 CHAPTER 6.1 243
LE1: Elliptic Membrane

Engineering Data
E = 210E+03 MPa
 = 0.3
P = 10 Mpa

Illustrations of the Model

Figure 6.1-1 Elliptical membrane

Main Index
 244 MSC Nastran Verification Guide

Figure 6.1-2 Elliptical membrane created in Patran using CQUAD4 elements

Model File Name and Location

Model File Name Description Location

nafem_lin01.dat Linear Static Analysis of Elliptical membrane \verifman\nafems

Results
Theoretical Results
Tangential edge stress (yy) at D: 92.7 MPa

MSC Nastran results vs theoretical results are as follows:

Tangential Edge Stress (MPa)

Theory MSC Nastran % Error
92.7 91.16 -1.661%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 CHAPTER 6.2 245
LE2: Cylindrical Shell Patch Test

6.2 LE2: Cylindrical Shell Patch

Test

Problem Description
Find the outer surface tangential (- ) stress at E (refer below figure) of cylindrical shell with two load cases
as follow.
1. Uniform normal edge moment, on DC of 1.0 kNm/m
2. Uniform outward normal pressure, at mid-surface ABCD, of 0.6 MPa and Tangential outward
normal pressure, on edge DC, of 60.0 MPa.

Solution Number
SOL 101

Features Used
 Linear Static Analysis.

Reference
1. Test LE2 from NAFEMS report TSBM, Publication NNB, Rev. 3, NAFEMS Linear Benchmarks,
5 Oct 1990.

Modeling Techniques Used

Cylindrical shell is modeled using plane stress elements CQUAD4. GPSTRESS is used to find out nodal
stress output.

Main Index
 246 MSC Nastran Verification Guide

Loading

Boundary Condition
Edge AB, all translations and rotations zero
Edge AD and edge BC, symmetry about r  plane e.g. Z translations and normal rotations all zero.

Engineering Data
E = 210E+03 MPa and = 0.3

Figure 6.2-1 Cylindrical shell

Main Index
 CHAPTER 6.2 247
LE2: Cylindrical Shell Patch Test

Model File Name and Location

Model File Name Description Location

nafem_lin02.dat Cylindrical Shell Patch Test \verifman\nafems

Results
Theoretical Results
Outer Convex surface tangential stress at point E: 60 MPa
MSC Nastran results vs theoretical results are as follows:

Tangential Edge Stress (MPa)

Theory MSC Nastran % Error
Case 1 60 57.3 -4.5%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 248 MSC Nastran Verification Guide

6.3 LE3: Hemisphere-Point

Loads

Problem Description
Find the displacement in x- direction of point A of hemisphere shown in Figure 6.3-1.

Solution Number
SOL 101

Features Used
 Linear Static Analysis.

Reference
1. NAFEMS Finite Element Methods & Standards, The Standard NAFEMS Benchmarks, Test No.
LE3. Glasgow: NAFEMS, Rev. 3, 1990.

Modeling Techniques Used

Hemisphere shell is modeled using plane stress elements CQUAD4.

Loading
Uniform normal pressure of 1 MPa on the upper surface of the plate.

Boundary Condition
Edge AB, all translations and rotations zero
Edge AD and edge BC, symmetry about r  plane, e.g. Z translations and normal rotations all zero.

Engineering Data
E = 68.25E+03 MPa and  = 0.3

Main Index
 CHAPTER 6.3 249
LE3: Hemisphere-Point Loads

Illustrations of the Model

Figure 6.3-1 Hemisphere Shell

Model File Name and Location

Model File Name Description Location

nafem_lin03.dat Hemisphere shell, point loads \verifman\nafems

Results
Theoretical Results
X displacement at point A: 0.185 m
MSC Nastran results vs theoretical results are as follows:

X- displacement at Point A (m)

Theory MSC Nastran % Error
0.185 0.17979 -2.81%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 250 MSC Nastran Verification Guide

6.4 LE5: Z-section Cantilever

Problem Description
A z-section cantilever beam is subjected to a torsion load. Find the axial stress (X-X) at mid-surface at Point
A shown in Figure 6.4-1.

Solution Number
SOL 101

Features Used
 Linear Static Analysis.

Reference
1. NAFEMS Finite Element Methods & Standards, The Standard NAFEMS Benchmarks, Test No.
LE5. Glasgow: NAFEMS, Rev. 3, 1990.

Modeling Techniques Used

Z Shaped Cantilever Beam is modeled using plane stress elements CQUAD4.

Loading
Torque of 1.2 MNm applied at end x = 10 by two uniformly distributed edge shears, S = 0.6 at each flange.

Engineering Data
E = 210E+03 MPa
= 0.3

Boundary Condition
Edge x = 0, all displacements are zero.

Main Index
 CHAPTER 6.4 251
LE5: Z-section Cantilever

Illustrations of the Model

Figure 6.4-1 Z Shaped Cantilever Beam

Model File Name and Location

Model File Name Description Location

nafem_lin05.dat Z- Section Cantilever, Torque \verifman\nafems

Results
Theoretical Results
Axial (x-x) stress at mid-surface, point A: -108 MPa (Compression)
MSC Nastran results vs theoretical results are as follows:

Axial (x-x) stress at mid-surface, point A (MPa)

Theory MSC Nastran % Error
-108 -103 -4.63%

Main Index
 252 MSC Nastran Verification Guide

Figure 6.4-2 Axial Stress Result in X- direction of Z-section Cantilever Beam by Nastran

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 CHAPTER 6.5 253
LE6: Skew Plate Normal Pressure

6.5 LE6: Skew Plate Normal

Pressure

Problem Description
A skew plate is subjected uniform normal pressure in the vertical z- direction. Find the maximum principal
at plate center.

Solution Number
SOL 400

Features Used
 Linear Static Analysis.

Reference
1. NAFEMS Finite Element Methods & Standards, The Standard NAFEMS Benchmarks, Test No.
LE6. Glasgow: NAFEMS, Rev. 3, 1990.

Modeling Techniques Used

Skew plate is modeled using different shell elements. Refer to the Results section.

Loading
Normal pressure of -0.7KPa in vertical Z-direction.

Engineering Data
E = 210E+03 MPa and = 0.3

Boundary Condition
Simple supports (no Z-displacement) for all edges AB, BC, CD, DA.

Main Index
 254 MSC Nastran Verification Guide

Illustrations of the Model

Figure 6.5-1 Skew Plate

Model File Name and Location

Model File Name Description Location

nafem_lin06q4.dat Skew Plate, Pressure, CQUAD4 tpl\verifman\nafems
nafem_lin06q8.dat Skew Plate, Pressure, CQUAD8 tpl\verifman\nafems
nafem_lin06q8a.dat Skew Plate, Pressure, CQUAD8 Advance tpl\verifman\nafems
nafem_lin06qr.dat Skew Plate, Pressure, CQUADR tpl\verifman\nafems

Results
Theoretical Results
Maximum principal stress on the lower surface at the plate center: 0.802 MPa
MSC Nastran results vs theoretical results are as follows:

Maximum Principal Stress, at point E (MPa)

Elements CQUAD4 CQUAD8 CQUAD8 Advance CQUADR
Theory 0.802 0.802 0.802 0.802
MSC Nastran 0.768 0.742 0.778 0.805
% Error -4.30% -7.50% -3.0% 0.4%

Conclusion
MSC Nastran result Maximum principal stress at Center of plate for decks with different elements
CQUAD4, CQUAD8, CQUAD8 with advance and CQUADR are comparable with Theoretical results.

Main Index
 CHAPTER 6.6 255
LE7: Axisymmetric Cylinder/Sphere-Pressure

6.6 LE7: Axisymmetric

Cylinder/Sphere-Pressure

Problem Description
Axisymmetric cylinder-sphere shell is subjected uniform internal pressure. Find the axial stress yy on the
outer surface of the upper cylinder at D.

Solution Number
SOL 400

Features Used
 Linear Static Axisymmetric.

Reference
1. NAFEMS Finite Element Methods & Standards, The Standard NAFEMS Benchmarks, Test No.
LE7. Glasgow: NAFEMS, Rev. 3, 1990.

Modeling Techniques Used

Cylindrical-Sphere Shell is modeled using axisymmetric elements CAXISYM.

Boundary Condition
Point A, zero radial displacement and zero rotation. Point F, zero y-displacement.

Loading
Uniform internal pressure of 1 MPa.

Engineering Data
E = 210E+03 MPa and = 0.3

Main Index
 256 MSC Nastran Verification Guide

Illustrations of the Model

Figure 6.6-1 Z Shaped Cantilever Beam

Model File Name and Location

Model File Name Description Location

nafem_lin07.dat Axisymmetric cylinder-sphere shell, uniform pressure \verifman\nafems

Results
Theoretical Results
Axial (yy) stress on the outer surface of the upper cylinder, at point D: 25.86 MPa
Nastran result is calculated using interpolation formula.
MSC Nastran results vs theoretical results are as follows:

Axial (x-x) stress at mid-surface, point A (MPa)

Theory MSC Nastran % Error
25.86 25.6866 -0.67%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 CHAPTER 6.7 257
LE8: Axisymmetric Shell Pressure

6.7 LE8: Axisymmetric Shell

Pressure

Problem Description
Axisymmetric shell is subjected uniform internal pressure. Find the hoop stress on the outer surface of the
shell at D.

Solution Number
SOL 400

Features Used
 Linear Static Axisymmetric.

Reference
1. NAFEMS Finite Element Methods & Standards, The Standard NAFEMS Benchmarks, Test No.
LE8. Glasgow: NAFEMS, Rev. 3, 1990.

Modeling Techniques Used

Shell is modeled using axisymmetric elements CAXISYM. The model details are given below:
E = 210E+03 MPa;  = 0.3

Loading
Uniform internal pressure of 1 MPa.

Boundary Condition
Point A, zero y displacement. Point E, zero radial displacement and zero rotation.

Main Index
 258 MSC Nastran Verification Guide

Illustrations of the Model

Figure 6.7-1 Axisymmetric Shell

Model File Name and Location

Model File Name Description Location

nafem_lin08.dat Axisymmetric shell, uniform pressure \verifman\nafems

Results
Theoretical Solution
Hoop stress (tt) stress on the outer surface of shell, at point D: 94.55 MPa
Nastran result is calculated using extrapolation formula:
83.97 + 110.14 -
-----------------------------------------
Hoop Stress at node 33: = 97.055 MPa
2

Result comparison of theoretical solution and MSC Nastran results are as follows:

Hoop stress tt, at point D (MPa)

Theory MSC Nastran % Error
94.55 97.055 2.64%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 CHAPTER 6.8 259
LE9: Axisymmetric Branched Shell-Pressure

6.8 LE9: Axisymmetric Branched

Shell-Pressure

Problem Description
Axisymmetric branched shell is subjected uniform internal pressure over edge BCD shown below in figure.
Find the axial stress  yy on the outer surface of the upper cylinder at C.

Solution Number
SOL 400

Features Used
 Linear Static Axisymmetric.

Reference
1. NAFEMS Finite Element Methods & Standards, The Standard NAFEMS Benchmarks, Test No.
LE9. Glasgow: NAFEMS, Rev. 3, 1990.

Modeling Techniques Used

Cylindrical Shell is modelled using axisymmetric elements CAXISYM. The model details are given below:
E = 210E+03 MPa
 = 0.3

Loading
Uniform internal pressure of 1 MPa over edge BCD

Boundary Condition
Point A, fully fixed

Main Index
 260 MSC Nastran Verification Guide

Illustrations of the Model

Figure 6.8-1 Axisymmetric Shell

Model File Name and Location

Model File Name Description Location

nafem_lin09.dat Axisymmetric branched shell, uniform pressure \verifman\nafems

Results
Theoretical Solution
Axial (yy) stress on the outer surface of the upper cylinder, at point C: -319.9 MPa
Nastran result is calculated using extrapolation formula as follows:
0.0234375
Axial Stress at node 33: 139.515 +  -------------------------   248.565 – 139.515  = 303.09 MPa
 0.015625 

Result comparison of theoretical solution and MSC Nastran results are as follows:

Axial stress (yy), at point C (MPa)

Theory MSC Nastran % Error
-319.19 -303.09 -5.25%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 CHAPTER 6.9 261
LE10: Thick Plate Pressure

6.9 LE10: Thick Plate Pressure

Problem Description
A elliptical thick plate is subjected uniform normal pressure on the upper surface of the plate. Find the direct
stress yy at Point D shown in Figure 6.9-1.

Solution Number
SOL 101

Features Used
 Linear Static Analysis.

Reference
1. NAFEMS Finite Element Methods & Standards, The Standard NAFEMS Benchmarks, Test No.
LE10. Glasgow: NAFEMS, Rev. 3, 1990.

Modeling Techniques Used

Thick elliptical plate is modelled using solid elements CHEXA. The model details are given below:
E = 210E+03 MPa
 = 0.3

Loading
Uniform normal pressure of 1 MPa on the upper surface of the Plate

Boundary Condition
Face DCD'C' zero y-displacement
Face ABA'B' zero x- displacement
Face BCB'C' x and y displacements fixed, z displacements fixed along mid-plane

Main Index
 262 MSC Nastran Verification Guide

Illustrations of the Model

Figure 6.9-1 Axisymmetric Shell

Model File Name and Location

Model File Name Description Location

nafem_lin10.dat Elliptical Thick Plate, Pressure \verifman\nafems

Results
Theoretical Solution
Normal (yy) stress at point D: 5.38 MPa

Result comparison of theoretical solution and MSC Nastran results are as follows:

Normal stress (yy), at point D (MPa)

Theory MSC Nastran % Error Element
5.38 6 11.52% HEXA 8
5.38 6.36 18.21% HEXA 20

Main Index
 CHAPTER 6.9 263
LE10: Thick Plate Pressure

Figure 6.9-2 a) Model with CHEXA element with 8 node b) Model with CHEXA element with 20 node

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 264 MSC Nastran Verification Guide

6.10 LE11: Solid Cylinder/ Taper/

Sphere-Temperature

Problem Description
A solid spherical taper cylinder is subjected to temperature loading. Find the direct stress  zz at Point A shown
in below figure.

Solution Number
SOL 101

Features Used
 Linear Static Analysis.

Reference
1. NAFEMS Finite Element Methods & Standards, The Standard NAFEMS Benchmarks, Test No.
LE11. Glasgow: NAFEMS, Rev. 3, 1990.

Modeling Techniques Used

Solid cylinder is modeled using solid elements CHEXA. The model details are given below:
E = 210E+03 MPa
 = 0.3
a = 2.3E-04 /0C

Loading

Linear temperature gradient in the radial and axial direction T (°C) = 2 2

x + y  + z

Boundary Condition
Symmetry on x-z plane i.e., zero Y-displacement
Symmetry on y-z plane i.e., zero X- displacement
Face on x-y plane zero Z- displacement
Face HIH’I’ zero Z- displacement

Main Index
 CHAPTER 6.10 265
LE11: Solid Cylinder/ Taper/ Sphere-Temperature

Illustrations of the Model

Figure 6.10-1 Spherical Taper Cylinder

Model File Name and Location

Model File Name Description Location

nafem_lin11.dat Spherical taper cylinder, temperature \verifman\nafems

Results
Theoretical Solution
Direct stress (zz) at point A: -105 MPa

Result comparison of theoretical solution and MSC Nastran results are as follows:

Direct stress (zz), at point A (MPa)

Theory MSC Nastran % Error
-105 -99.477 -5.26%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 266 MSC Nastran Verification Guide

Linear Thermo Elastic

Main Index
 CHAPTER 6.11 267
T1: Membrane with Hot Spot

6.11 T1: Membrane with Hot Spot

Problem Description
Find the direct stress in y- direction at point D in membrane as shown below figure which is subjected to
thermal load (hot-spot) at center.

Solution Number
SOL 400

Features Used
 Heat Transfer

Reference
1. NAFEMS Finite Element Methods & Standards, The Standard NAFEMS Benchmarks, Test No.
T1. Glasgow: NAFEMS, Rev. 3, 1990.

Modeling Techniques Used

Membrane is modeled using 2D element CQUAD4. The model details are given below:
Isotropic, E = 210E+03 MPa
 = 0.3

Loading
Within hot-spot ( 0  r  1.0 mm )

Thermal strain ( T ) = 1.0 x 10-3

Outboard of hot-spot ( r  1.0 mm )
Thermal strain ( T ) = 0

Boundary Condition
On quarter model,
at y=0, symmetric about x, e.g. y displacement zero

Main Index
 268 MSC Nastran Verification Guide

at x=0, symmetric about y, e.g. x displacement zero

Illustrations of the Model

Figure 6.11-1 Membrane with Hot Spot

Main Index
 CHAPTER 6.11 269
T1: Membrane with Hot Spot

Model File Name and Location

Model File Name Description Location

nafem_te01.dat Membrane with Hot Spot \verifman\nafems

Results
Theoretical Solution
Direct stress at Point D in Y direction: 50 MPa
Result comparison of theoretical solution and MSC Nastran results are as follows:

Direct stress at D in Y- Direction (MPa)

Theory MSC Nastran % Error
50 46.5 -7.0%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 270 MSC Nastran Verification Guide

Heat Transfer

Main Index
 CHAPTER 6.12 271
T2: One Dimensional Heat Transfer with Radiation

6.12 T2: One Dimensional Heat

Transfer with Radiation

Problem Description
Find the material temperature at end of the bar when it is subjected to thermal load at other end.

Solution Number
SOL 400

Features Used
 Heat Transfer

Reference
1. NAFEMS Finite Element Methods & Standards, The Standard NAFEMS Benchmarks, Test No.
T2. Glasgow: NAFEMS, Rev. 3, 1990.

Modeling Techniques Used

Bar is modeled using 2D element CBAR. The model details are given below:
Conductivity = 55.6 W/m0C
Specific Heat = 460.0 J/Kg0C
Density = 7850 Kg/m3

At boundary B
Emissivity = 0.98
Stefan-Boltzman Constant = 5.67 x 10-8 Wm2 0K-4

Loading
Zero internal heat generation

Boundary Condition
At A, prescribed temperature of 1000 0K

Main Index
 272 MSC Nastran Verification Guide

At B, radiation to ambient temperature of 300 0K

No heat flux perpendicular to AB

Illustrations of the Model

Figure 6.12-1 Bar

Model File Name and Location

Model File Name Description Location

nafem_te02.dat One Dimensional Heat Transfer with Radiation \verifman\nafems

Results
Theoretical Solution
Material temperature at B: 927 oK
Result comparison of theoretical solution and MSC Nastran results are as follows:

Temperature at B in K
Theory MSC Nastran % Error
927 912.7 -1.54%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 CHAPTER 6.13 273
T3: One Dimensional Transient Heat Transfer

6.13 T3: One Dimensional

Transient Heat Transfer

Problem Description
Find the Material Temperature at point C of the bar when it is subjected to transient thermal load at B. (Refer
Figure 6.13-1)

Solution Number
SOL 400

Features Used
 Transient Heat Conduction

Reference
1. NAFEMS Finite Element Methods & Standards, The Standard NAFEMS Benchmarks, Test No.
T3. Glasgow: NAFEMS, Rev. 3, 1990.

Modeling Techniques Used

Bar is modeled using 2D element CROD. The model details are given below:
Conductivity = 35.0 W/m0C
Specific Heat = 440.5 J/Kg0C
Density = 7200 Kg/m3

Loading
Zero internal heat generation

Boundary Condition
At time t = 0, All temperature = 0 0C
At time t > 0, All temperature = 0 0C

At B, temperature = nt- 0C
100 sin -----
40

Main Index
 274 MSC Nastran Verification Guide

No heat flux perpendicular to AB

Illustrations of the Model

Figure 6.13-1 Bar Subjected to Transient Load

Model File Name and Location

Model File Name Description Location

nafem_te03.dat One Dimensional Transient Heat Transfer \verifman\nafems

Results
Theoretical Solution
Material temperature at C: 36.60 oC
Result comparison of theoretical solution and MSC Nastran results are as follows:

Material Temperature at Point C

Theory MSC Nastran % Error
36.60 34.85 -4.78%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 CHAPTER 6.14 275
T4: Two Dimensional Heat Transfer with Convection

6.14 T4: Two Dimensional Heat

Transfer with Convection

Problem Description
Natural Convection heat transfer occurs in one edge insulted plate which is subjected to thermal loading at
one edge. Find the temperature at point E. (Refer Figure 6.14-1)

Solution Number
SOL 400

Features Used
 Heat Conduction

Reference
1. NAFEMS Finite Element Methods & Standards, The Standard NAFEMS Benchmarks, Test No.
T4. Glasgow: NAFEMS, Rev. 3, 1990.

Modeling Techniques Used

Plate is modeled using 2D elements CQUAD4. The model details are given below:
Conductivity = 52.0 W/m0C
Surface convective heat transfer = 750.0 W/m2 0C
coefficient (edges BC, CD)

Loading
Zero internal heat generation

Boundary Condition
Edge AB, temperature = 100 0C
Edge DA, Zero heat flux
Edges BC, CD, convection to ambient temperature = 0 0C

Main Index
 276 MSC Nastran Verification Guide

Illustrations of the Model

Figure 6.14-1 Insulated Plate

Model File Name and Location

Model File Name Description Location

nafem_te04.dat Two Dimensional Transient Heat Transfer \verifman\nafems

Results
Theoretical Solution
Material temperature at E: 18.3 oC
Result comparison of theoretical solution and MSC Nastran results are as follows:

Material Temperature at Point E

Theory MSC Nastran % Error
18.3 19.3 5.46%

Conclusion
MSC Nastran results are comparable with theoretical results.

Main Index
 CHAPTER 6.14 277
T4: Two Dimensional Heat Transfer with Convection

Free Vibration

Main Index
 278 MSC Nastran Verification Guide

6.15 FV2- Pin-Ended Double

Cross: In-Plane Vibration

Problem Description
NAFEMS Benchmark FV2 - pin-ended double cross normal modes

Solution Number
SOL 103

Features Used
 Normal modes
 Beam and bar elements (Euler and considering shear flexibility)
 Coupling between flexural and extensional behavior
 Repeated eigenvalues

Reference
1. Problem FV2 from The Standard NAFEMS Benchmarks, Oct 5, 1990.

Modeling Techniques Used

 This is a simple model using beam or bar elements. Coordinate systems are used to simplify the
modeling.
 Modeled using 2- and 3-noded beam elements.

Main Index
 CHAPTER 6.15 279
FV2- Pin-Ended Double Cross: In-Plane Vibration

Illustrations of the Model

From the NAFEMS reference:

Geometry and Mesh

Exact beam: 4 elements per arm

Iso-parametric beam: 4 elements per arm

All arms of equal length

Boundary Conditions:
• Out-of plane motion is constrained at all nodes.
• Nodes A through H are constrained for in-plane translation.

Material Properties
E = 200  109N/m2
8000kg/m3

Main Index
 280 MSC Nastran Verification Guide

Model File Name and Location

Model File Name Description Location

fv2_2e.dat 2-noded Euler beams \verifman\nafems
fv2_2ec 2-noded Euler beams with coupled mass \verifman\nafems
fv2_2s.dat 2-noded beams with shear flexibility \verifman\nafems
fv2_3e.dat 3-noded Euler beams \verifman\nafems
fv2_3s.dat 3-noded beams with shear flexibility \verifman\nafems

Results
Natural frequencies (Hz):

2-Noded 2-Noded Euler 2-Noded Beams

NAFEMS Euler Beams using with Shear 3-Noded Euler
Mode Value Beams Coupled Mass Flexibility Beams
1 11.336 11.333 11.339 11.323 11.337
2 and 3 17.687 17.662 17.692 17.621 17.685
4 through 8 17.715 17.69 17.720 17.650 17.715
9 45.477 45.016 45.524 45.841 45.856
10 and 11 57.364 56.058 57.427 55.682 58.043
12 through 16 57.683 56.343 57.748 55.961 58.387

Conclusion
Although not specifically mentioned, the NAFEMS values are obviously based on Euler beam theory and
ignore the shear flexibility. If Euler beams are used, the results compare well with the NAFEMS expected
values. If shear flexibility is included, then the natural frequencies are slightly lower.

Main Index
 CHAPTER 6.16 281
FV4 - Cantilever with Off-Centre Point Masses

6.16 FV4 - Cantilever with Off-

Centre Point Masses

Problem Description
Calculation of modes of a cantilever beam with offset masses at the free end.

Solution Number
SOL 103 - normal modes

Features Used
 BAR elements (Euler beam and inclusion of shear flexibility)
 3-noded Beam elements
 CONM2 with offset
 Normal modes
 Coupling between bending and torsion
 Close eigenvalues

Reference
1. NAFEMS Finite Element Methods & Standards. Abbassian, F., Dawswell, D. J., and Knowles, N.
C.Selected Benchmarks for Natural Frequency Analysis, Test No. 4. Glasgow: NAFEMS, Nov.,
1987.

Modeling Techniques Used

 Simple cantilever beam model.
 CONM2's with offsets used to model offset masses.

Main Index
 282 MSC Nastran Verification Guide

Illustrations of the Model

M1 = 10000 kg (along X, Y, Z)
Geometry and Mesh
M2 = 1000 kg (along X, Y, Z)
Exact beam: 5 elements along cantilever

Iso-parametric beam: 3 elements along cantilever

Model File Name and Location

Model File Name Description Location

fv4_2e.dat 2-noded Euler beams \verifman\nafems
fv4_2ec 2-noded Euler beams with coupled mass \verifman\nafems
fv4_2s.dat 2-noded beams with shear flexibility \verifman\nafems
fv4_3e.dat 3-noded Euler beams \verifman\nafems
fv2_3s.dat 3-noded beams with shear flexibility \verifman\nafems

Results
Natural frequencies (Hz) of the first 6 modes:

2-Noded 2-Noded Euler

NAFEMS Euler Beams using 2-Noded Beams with 3-Noded Euler
Mode Value Beams Coupled Mass Shear Flexibility Beams
1 1.723 1.715 1.723 1.713 1.722
2 1.727 1.719 1.727 1.718 1.726
3 7.413 7.428 7.45 7.415 7.423
4 9.972 9.978 9.975 9.952 9.965
5 18.155 17.748 18.205 17.658 18.226
6 26.957 27.058 27.001 26.762 27.188

Conclusion
While not an exact match, the calculated natural frequencies agree well with the expected values.

Main Index
 CHAPTER 6.17 283
FV5 - Deep Simply-supported Beam

6.17 FV5 - Deep Simply-

supported Beam

Problem Description
Calculation of modes of a simply supported beam

Solution Number
SOL 103 - normal modes

Features Used
 BEAM elements (Timoshenko beam formulation)
 2- and 3-noded Beam elements
 Normal modes - Lanczos method
 Shear deformation and torsion

Reference
1. NAFEMS Finite Element Methods & Standards. Abbassian, F., Dawswell, D. J., and Knowles, N.
C.Selected Benchmarks for Natural Frequency Analysis, Test No. 5. Glasgow: NAFEMS, Nov.,
1987.

Modeling Techniques Used

 Simply-supported beam model using BEAM elements.

Main Index
 284 MSC Nastran Verification Guide

Illustrations of the Model

From the NAFEMS reference:
Geometry and Mesh

Exact beam: 5 elements

Iso-parametric beam: 5 elements

Model File Name and Location

Model File Name Description Location

fv5_2.dat 2-noded beams using lumped mass \verifman\nafems
fv5_2c 2-noded beams using coupled mass \verifman\nafems
fv5_3.dat 3-noded beams \verifman\nafems

Results
Natural frequencies (Hz) of the first 9 modes

2-Noded Beams
NAFEMS using lumped 2-Noded Beams using
Mode Value mass Coupled Mass 3-Noded Euler Beams
1 and 2 42.649 43.111 43.226 38.635
3 77.542 77.204 77.842 77.524
4 125.0 124.487 125.515 125.002
5 and 6 148.31 149.393 154.519 172.888
7 233.1 224.056 241.242 232.811
8 and 9 284.55 269.578 305.392 253.929

Conclusion
While not an exact match, the calculated natural frequencies agree well with the expected values. While the
bending modes for the 3-noded beam are different than those of the 2-noded beams, they agree well with the
modes one obtains if the models using 2-noded beams are refined to have 10 elements.

Main Index
 CHAPTER 6.18 285
FV12- Free Thin Square Plate

6.18 FV12- Free Thin Square Plate

Problem Description
Out of plane free vibration of a square plate with in-plane motion constrained.

Solution Number
SOL 103 - normal modes

Features Used
 Rigid-body modes (3)
 Repeated eigenvalues

Reference
1. NAFEMS Finite Element Methods & Standards, Abbassian, F., Dawswell, D. J., and Knowles, N.
C., Selected Benchmarks for Natural Frequency Analysis, Test No. 12. Glasgow: NAFEMS, Nov.,
1987.

Modeling Techniques Used

 A simple flat plate model is created using either QUAD4 or QUAD8 elements and the out-of plane
modes are calculated.

Illustrations of the Model

t = 0.05 m
Geometry and Mesh

H.O.E. 4 × 4 (as shown)

L.O.E. 8 × 8

Main Index
 286 MSC Nastran Verification Guide

Boundary Conditions
x- and y-translation and z-rotation are constrained to 0.0

Material Properties
Young’s modulus, E = 200  109N/m2
Poisson's ratio, 0.3

Model File Name and Location

Model File Name Description Location

fv12_q4.dat QUAD4 with lumped mass \verifman\nafems
fv12_q4c.dat QUAD4 with coupled mass \verifman\nafems
fv12_q8.dat QUAD8 with lumped mass \verifman\nafems
fv12_q8c.dat QUAD8 with coupled mass \verifman\nafems

Results
Natural frequencies (Hz) of modes 4 through 10 (modes 1-3 are rigid-body modes).

Main Index
 CHAPTER 6.18 287
FV12- Free Thin Square Plate

QUAD4 results

NAFEMS NAFEMS Expected QUAD4 Lumped QUAD4 Coupled

Mode Theoretical Value Value Mass Mass
4 7.938 8.142 7.832 7.967
5 12.835 13.891 12.851 13.400
6 17.941 20.035 18.078 19.261
7 and 8 19.133 20.165 18.585 19.338
9 24.009 27.950 24.310 26.837
10 27.922 32.046 27.644 29.702

QUAD8 results

NAFEMS NAFEMS Expected QUAD4 Lumped QUAD4 Coupled

Mode Theoretical Value Value Mass Mass
4 7.938 7.873 7.992 8.012
5 12.835 12.480 13.051 13.177
6 17.941 17.312 18.667 19.112
7 and 8 19.133 18.738 19.238 19.383
9 24.009 22.704 25.927 27.081
10 27.922 25.883 28.421 29.060

Conclusion
In all but one case (QUAD8, coupled mass, mode 5) the results are closer to the theoretical values than the
expected results. It should be noted that a 45 degree skew angle is well beyond the recommended maximum
skew of 30 degrees, but the elements still performed well.

Main Index
 288 MSC Nastran Verification Guide

6.19 FV15- Clamped Thin

Rhombic Plate

Problem Description
Solution to find the first 6 modes of a clamped flat plate, which is skewed 45 degrees.

Solution Number
SOL 103 normal modes

Features Used
 Lanczos eigenvalue solution

Reference
1. NAFEMS Finite Element Methods & Standards, Abbassian, F., Dawswell, D. J., and Knowles, N.
C., Selected Benchmarks for Natural Frequency Analysis, Test No. 15. Glasgow: NAFEMS, Nov.,
1987.

Modeling Techniques Used

 Flat QUAD4 and QUAD8 elements modeling a clamped plate.

Illustrations of the Model

t = 0.05 m
Geometry and Mesh

H.O.E. 6 × 6 (as shown)

L.O.E. 12 × 12

Main Index
 CHAPTER 6.19 289
FV15- Clamped Thin Rhombic Plate

Boundary Conditions
Clamped edges

Model File Name and Location

Model File Name Description Location

fv15_q4.dat QUAD4 with lumped mass \verifman\nafems
fv15_q4c.dat QUAD4 with coupled mass \verifman\nafems
fv15_q8.dat QUAD8 with lumped mass \verifman\nafems
fv15_q8c.dat QUAD8 with coupled mass \verifman\nafems

Main Index
 290 MSC Nastran Verification Guide

Results

QUAD4 results

NAFEMS NAFEMS Expected QUAD4 Lumped QUAD4 Coupled

Mode Theoretical Value Value Mass Mass
1 7.938 8.142 7.832 7.967
2 12.835 13.891 12.851 13.400
3 17.941 20.035 18.078 19.261
4 19.133 20.165 18.585 19.338
5 24.009 27.950 24.310 26.837
6 27.922 32.046 27.644 29.702

QUAD8 results

NAFEMS NAFEMS Expected QUAD8 Lumped QUAD8 Coupled

Mode Theoretical Value Value Mass Mass
1 7.938 7.873 7.992 8.012
2 12.835 12.480 13.051 13.177
3 17.941 17.312 18.667 19.112
4 19.133 18.738 19.238 19.383
5 24.009 22.704 25.927 27.081
6 27.922 25.883 28.421 29.060

Conclusion
In all but one case (QUAD8, coupled mass, mode 5) the results are closer to the theoretical values than the
expected results. It should be noted that a 45 degree skew angle is well beyond the recommended maximum
skew of 30 degrees, but the elements still performed well.

Main Index
 CHAPTER 6.20 291
FV16- Cantilevered Thin Plate

6.20 FV16- Cantilevered Thin

Plate

Problem Description
Normal modes of a cantilevered thin plate modeled using QUAD8 elements.

Solution Number
SOL 103 - normal modes using Lanczos method.

Features Used
 Normal modes calculation using QUAD8 elements, distorted in-plane.

Reference
1. NAFEMS Finite Element Methods & Standards, Abbassian, F., Dawswell, D. J., and Knowles, N.
C., Selected Benchmarks for Natural Frequency Analysis, Test No. 16. Glasgow: NAFEMS, Nov.,
1987.

Modeling Techniques Used

 A simple flat plate model created using QUAD8 elements (NAFEMS lists 8-noded semi-loof
elements, which do not exist in MSC Nastran). Four different versions are created, using different
mesh refinement and distortion. The first 6 out-of-plane modes are calculated.

Illustrations of the Model

The numbering and locations shown for Test2 are not consistent with the axes and diagrams in the reference.
As such, the locations have been applied to the closest GRID points to the offsets. Also, although it is not
mentioned in the reference, straight lines are drawn between the offset GRID points, so the mid-side nodes
will also be offset to maintain straight edges.

Main Index
 292 MSC Nastran Verification Guide

Illustrations of models from the reference:

Locations of Points from the reference (and corrections):

Coordinates (m)
Node No. x y
1 4.0 4.0
2 2.25 2.25
3 4.75 2.5
4 7.25 2.75
5 7.5 4.75
6 7.75 7.25
7 5.25 7.25
8 2.25 7.25
9 2.5 4.75

Table 6-1 Locations for Test 2

Coordinates (m)
Node No. x y
1 4.0 4.0

Table 6-2 Locations for Test 4

Main Index
 CHAPTER 6.20 293
FV16- Cantilevered Thin Plate

Correct locations in illustration for Test 2:

NAFEMS Node No. Correct Node No.

1 1
2 8
3 9
4 2
5 3
6 4
7 5
8 6
9 7

Main Index
 294 MSC Nastran Verification Guide

Model File Name and Location

Model File Name Description Location

Fv16_t1.dat Test 1 - CQUAD8, lumped mass \verifman\nafems
Fv16_t1c.dat Test 1 - CQUAD8, coupled mass \verifman\nafems
Fv16_t2.dat Test 2 - CQUAD8, lumped mass \verifman\nafems
Fv16_t2c.dat Test 2 - CQUAD8, coupled mass \verifman\nafems
Fv16_t3.dat Test 3 - CQUAD8, lumped mass \verifman\nafems
Fv16_t3c.dat Test 3 - CQUAD8, coupled mass \verifman\nafems
Fv16_t4.dat Test 4 - CQUAD8, lumped mass \verifman\nafems
Fv16_t4c.dat Test 4 - CQUAD8, coupled mass \verifman\nafems

Results
Test 1 results:

NAFEMS Expected QUAD8 Lumped QUAD8 Coupled

Mode Value Mass Mass
1 0.421 0.4153 0.4183
2 1.029 1.005 1.023
3 2.582 2.485 2.579
4 3.306 3.132 3.298
5 3.753 3.622 3.765
6 6.555 6.292 6.719

Test 2 results:

NAFEMS Expected QUAD8 Lumped QUAD8 Coupled

Mode Value Mass Mass
1 0.421 0.4153 0.4193
2 1.029 1.007 1.026
3 2.582 2.509 2.604
4 3.306 3.164 3.321
5 3.753 3.664 3.815
6 6.555 6.327 6.801

Main Index
 CHAPTER 6.20 295
FV16- Cantilevered Thin Plate

Test 3 results:

NAFEMS Expected QUAD8 Lumped QUAD8 Coupled

Mode Value Mass Mass
1 0.421 0.4069 0.4186
2 1.029 0.9648 1.021
3 2.582 2.200 2.662
4 3.306 2.894 3.402
5 3.753 3.348 3.904
6 6.555 5.072 6.780

Test 4 results:

NAFEMS Expected QUAD8 Lumped QUAD8 Coupled

Mode Value Mass Mass
1 0.421 0.4067 0.4190
2 1.029 0.9554 1.026
3 2.582 2.200 2.673
4 3.306 2.800 3.437
5 3.753 3.387 4.006
6 6.555 4.941 6.892

Conclusion
Considering the number of elements in the models, the program gave excellent results. With square elements,
the results show excellent correlation, considering the mesh and the results for distorted elements are
acceptable, considering the mesh and element distortions. Using the recommendation of 5 GRID points per
half sine-wave of the deformed shape, Test 1 and 2 might be expected to predict mode 1 and possibly mode
2 well and Test 3 and 4 might be expected to predict the first mode well.

Main Index
 296 MSC Nastran Verification Guide

6.21 FV22 - Clamped Thick

Rhombic Plate

Problem Description
Solution to find the first 6 modes of a clamped flat plate, which is skewed 45 degrees.

Solution Number
SOL 103 - normal modes

Features Used
 Lanczos Eigenvalue Solution

Reference
1. NAFEMS Finite Element Methods & Standards, Abbassian, F., Dawswell, D. J., and Knowles, N.
C., Selected Benchmarks for Natural Frequency Analysis, Test No. 22. Glasgow: NAFEMS, Nov.,
1987.

Modeling Techniques Used

 Distorted Flat QUAD4 and QUAD8 elements modeling a thick clamped plate.

Illustrations of the Model

t = 1.0 m
Geometry and Mesh

H.O.E. 6 × 6 (as shown)

L.O.E. 10 × 10

Boundary conditions
Clamped edges.

Main Index
 CHAPTER 6.21 297
FV22 - Clamped Thick Rhombic Plate

Model File Name and Location

Model File Name Description Location

fv22_q4.dat QUAD4 with lumped mass \verifman\nafems
fv22_q4c.dat QUAD4 with coupled mass \verifman\nafems
fv22_q8.dat QUAD8 with lumped mass \verifman\nafems
fv22_q8c.dat QUAD8 with coupled mass \verifman\nafems

Results
QUAD4

NAFEMS NAFEMS Expected QUAD4 Lumped QUAD4 Coupled

Mode Theoretical Value Value Mass Mass
1 133.95 137.80 131.22 134.27
2 201.41 218.48 200.37 211.89
3 265.81 295.42 262.03 286.62
4 282.74 296.83 273.59 287.03
5 334.45 383.56 327.01 373.25
6 NA 426.59 372.22 410.58

QUAD8

NAFEMS NAFEMS Expected QUAD4 Lumped QUAD4 Coupled

Mode Theoretical Value Value Mass Mass
1 133.95 133.86 134.91 135.24
2 201.41 203.34 204.42 206.27
3 265.81 271.38 270.27 276.39
4 282.74 283.68 286.89 289.07
5 334.45 346.41 374.45 353.77
6 NA 386.62 384.73 393.96

Conclusion
Distortion of elements to 45 degrees skew angle is well beyond the recommended maximum value of 30
degrees. However, the elements still performed well in this example. As the mesh is not that refined, the
results are expected to be less accurate after the first mode or two.

Main Index
 298 MSC Nastran Verification Guide

6.22 FV32 - Cantilevered Tapered

Membrane

Problem Description
Solution to find the first 6 modes of a tapered membrane plate.

Solution Number
SOL 103-normal modes

Features Used
 Lanczos eigenvalue solution

Reference
1. NAFEMS Finite Element Methods & Standards, Abbassian, F., Dawswell, D. J., and Knowles, N.
C., Selected Benchmarks for Natural Frequency Analysis, Test No. 32. Glasgow: NAFEMS, Nov.,
1987.

Modeling Techniques Used

 Distorted Flat QUAD4 and QUAD8 elements modeling a tapered membrane. Note - the thickness
is not provided, but for finite elements, any thickness will give the same answers.

Illustrations of the Model

Geometry and Mesh

H.O.E. 8 × 4 (as shown)

L.O.E. 16 × 8

Main Index
 CHAPTER 6.22 299
FV32 - Cantilevered Tapered Membrane

Boundary Conditions
Clamped edges

Model File Name and Location

Model File Name Description Location

fv32_q4.dat QUAD4 with lumped mass \verifman\nafems
fv32_q4c.dat QUAD4 with coupled mass \verifman\nafems
fv32_q8.dat QUAD8 with lumped mass \verifman\nafems
fv32_q8c.dat QUAD8 with coupled mass \verifman\nafems

Main Index
 300 MSC Nastran Verification Guide

Results

QUAD4 results

NAFEMS NAFEMS Expected QUAD4 Lumped QUAD4 Coupled

Mode Theoretical Value Value Mass Mass
1 44.623 44.636 44.52 44.65
2 130.03 130.14 129.55 131.04
3 162.70 162.72 162.56 162.80
4 246.05 246.63 244.13 250.33
5 379.90 382.02 374.46 391.54
6 391.44 391.55 389.60 393.10

QUAD8 results

NAFEMS NAFEMS Expected QUAD4 Lumped QUAD4 Coupled

Mode Theoretical Value Value Mass Mass
1 44.623 44.905 44.54 44.63
2 130.03 132.12 129.71 130.11
3 162.70 162.83 162.66 162.71
4 246.05 252.99 245.14 246.44
5 379.90 393.31 377.87 381.41
6 391.44 396.26 390.92 391.52

Conclusion
Distortion of elements is a way to test their accuracy. In this case, the results are acceptably close to the
predicted results.

Main Index
 CHAPTER 6.23 301
FV41 - Axisymmetric Vibration

6.23 FV41 - Axisymmetric

Vibration

Problem Description
Free modes of a cylinder using axisymmetric elements.

Solution Number
SOL 103 - normal modes

Features Used
 Axisymmetric elements
 Normal modes - Lanczos method

Reference
1. NAFEMS Finite Element Methods & Standards, Abbassian, F., Dawswell, D. J., and Knowles, N.
C., Selected Benchmarks for Natural Frequency Analysis, Test No. 41. Glasgow: NAFEMS, Nov.,
1987.

Modeling Techniques Used

 Simply-supported beam model using BEAM elements.

Illustrations of the Model

Geometry and Mesh

H.O.E. 8 × 1 (as shown)

L.O.E. 16 × 3

Main Index
 302 MSC Nastran Verification Guide

Model File Name and Location

Model File Name Description Location

fv41_q4.dat 4-noded axisymmetric quad elements \verifman\nafems
using lumped mass
Fv41_q4c.dat 4-noded axisymmetric quad elements \verifman\nafems
using coupled mass
Fv41_q8.dat 8-noded axisymmetric quad elements \verifman\nafems
using lumped mass
Fv41_q4c.dat 8-noded axisymmetric quad elements \verifman\nafems
using coupled mass

Results

Natural frequencies (hz) of modes 2-6

Note: The first mode is a rigid-body mode.

Main Index
 CHAPTER 6.23 303
FV41 - Axisymmetric Vibration

4-noded CQUADX

4-noded QUADX
NAFEMS NAFEMS Expected Using Lumped 4-noded QUADX
Mode Theoretical Value Value Mass Using Coupled Mass
2 243.53 244.01 243.23 244.01
3 377.41 379.41 374.47 379.42
4 394.11 395.41 389.47 395.44
5 397.72 397.94 389.49 401.38
6 405.28 421.87 396.10 421.89

8-noded CQUADX

8-noded QUADX
NAFEMS NAFEMS Expected Using Lumped 8-noded QUADX
Mode Theoretical Value Value Mass Using Coupled Mass
2 243.53 243.50 243.24 243.57
3 377.41 377.46 356.49 377.46
4 394.11 394.28 356.88 394.30
5 397.72 397.94 375.85 397.97
6 405.28 406.41 393.65 406.44

Conclusion
While the results using lumped mass are not bad, the results using coupled mass fall within 1% of the
expected values for all modes.

Main Index
 304 MSC Nastran Verification Guide

6.24 FV42 - Thick Hollow Sphere

Problem Description
Radial modes of a hollow sphere using axisymmetric elements.

Solution Number
SOL 103 - normal modes

Features Used
 Axisymmetric Elements
 Cylindrical Coordinate System
 MPC Equations
 Normal Modes - Lanczos Method

Reference
1. NAFEMS Finite Element Methods & Standards, Abbassian, F., Dawswell, D. J., and Knowles, N.
C., Selected Benchmarks for Natural Frequency Analysis, Test No. 42. Glasgow: NAFEMS, Nov.,
1987.

Modeling Techniques Used

 Simply-supported beam model using BEAM elements.

Illustrations of the Model

Geometry and Mesh

H.O.E. 5 × 1 (as shown)

 = 10°

L.O.E. 10 × 1
 = 5°

Constraints:
z' displacements are constrained to 0.0 and GRIDs at the same radius are constrained to have the same radial
displacement.

Main Index
 CHAPTER 6.24 305
FV42 - Thick Hollow Sphere

Model File Name and Location

Model File Name Description Location

fv42_q4.dat 4-noded axisymmetric quad elements \verifman\nafems
using lumped mass
Fv42_q4c.dat 4-noded axisymmetric quad elements \verifman\nafems
using coupled mass
Fv42_q8.dat 8-noded axisymmetric quad elements \verifman\nafems
using lumped mass
Fv42_q8c.dat 8-noded axisymmetric quad elements \verifman\nafems
using coupled mass

Results
Natural frequencies (hz) of the first 5 modes

4-noded CQUADX

4-noded QUADX
NAFEMS NAFEMS Expected Using Lumped 4-noded QUADX
Mode Theoretical Value Value Mass Using Coupled Mass
1 369.91 370.64 370.47 370.64
2 838.03 841.2 833.49 841.20
3 1451.2 1473.1 1423.77 1473.06
4 2117.0 2192.2 2033.83 2192.20
5 2795.8 2975.7 2050.78 2975.67

8-noded CQUADX

NAFEMS NAFEMS Expected 8-noded QUADX Using 8-noded QUADX

Mode Theoretical Value Value Lumped Mass Using Coupled Mass
1 369.91 370.01 368.54 370.01
2 838.03 838.08 817.72 841.20
3 1451.2 1453.0 1387.84 1453.0
4 2117.0 2131.7 1475.40 1942.55
5 2795.8 2852.8 1974.77 2131.74

Main Index
 306 MSC Nastran Verification Guide

Conclusion
While the results using lumped mass are not bad, the results using coupled mass on the 4-noded elements
matches the baseline values within a fraction of a percent. The 8-noded elements match the first 3 modes
almost exactly, but fall off on the last two modes, indicating a need for a more refined mesh.

Main Index
 CHAPTER 6.25 307
FV52 - Simply Supported Solid Square Plate

6.25 FV52 - Simply Supported

Solid Square Plate

Problem Description
Calculate the first 10 modes of a plate which is supported in the Z-direction on its edges.

Solution Number
SOL 103 normal modes

Features Used
 Solid elements
 Normal modes - Lanczos method

Reference
1. NAFEMS Finite Element Methods & Standards. Abbassian, F., Dawswell, D. J., and Knowles, N.
C.Selected Benchmarks for Natural Frequency Analysis, Test No. 51. Glasgow: NAFEMS, Nov.,
1987.

Modeling Techniques Used

 Simply-supported beam model using BEAM elements.

Illustrations of the Model

Geometry and Mesh

H.O.E. 4 × 4 × 1 (as shown)

L.O.E. 8 × 8 × 3

Main Index
 308 MSC Nastran Verification Guide

Constraints
z' displacements are constrained to 0.0 along the bottom edges

Model File Name and Location

Model File Name Description Location

fv52_h8.dat 8-noded HEXA elements using lumped mass \verifman\nafems
fv52_h8c.dat 8-noded HEXA elements using coupled mass \verifman\nafems
fv52_q20.dat 20-noded HEXA elements using lumped mass \verifman\nafems
fv52_q20c.dat 20-noded HEXA elements using coupled mass \verifman\nafems

Results
Natural frequencies (Hz) of modes 4-10. Modes 1-3 are rigid-body modes.

8-noded HEXA

NAFEMS NAFEMS Expected 8-noded HEXA using 8-noded HEXA using

Mode Theoretical Value Value Lumped Mass Coupled Mass
4 45.897 51.564 44.62 45.24
5, 6 109.44 132.73 107.83 113.70
7 167.89 194.37 157.55 172.25
8 193.59 197.18 193.64 196.76
9, 10 206.19 210.55 200.39 209.56

20 noded HEXA

NAFEMS NAFEMS Expected 20-noded HEXA 20-noded HEXA

Mode Theoretical Value Value using Lumped Mass using Coupled Mass
4 45.897 44.762 43.81 44.16
5, 6 109.44 110.52 105.24 107.93
7 167.89 169.08 156.26 163.92
8 193.59 193.93 193.97 193.91
9, 10 206.19 206.64 193.52 206.60

Conclusion
In most cases, the values from MSC Nastran were closer to the theoretical values than the NAFEMS expected
values. As expected with a coarse mesh model, the results using coupled mass were closer to the theoretical
values than those using lumped mass.

Main Index
 CHAPTER 6.26 309
FV73 - Cantilevered Thin Square Plate

6.26 FV73 - Cantilevered Thin

Square Plate

Problem Description
Calculate the first 6 modes of a plate which is simply supported along the Y-axis.

Solution Number
SOL 103 normal modes

Features Used
 8-noded QUAD elements (NAFEMS lists 8-noded semi-loof elements, which do not exist in MSC
Nastran)
 Normal modes - Lanczos method

Reference
1. NAFEMS Finite Element Methods & Standards. Abbassian, F., Dawswell, D. J., and Knowles, N.
C.Selected Benchmarks for Natural Frequency Analysis, Test No. 73. Glasgow: NAFEMS, Nov.,
1987.

Modeling Techniques Used

 Simply-supported plate model using QUAD8 elements.

Main Index
 310 MSC Nastran Verification Guide

Illustrations of the Model

Constraints
X, Y, and Z translation and Y rotation are constrained to 0.0 on the GRID points along the Y-axis

Model File Name and Location

Model File Name Description Location

fv73_q8_t1.dat Test 1 using Lumped Mass \verifman\nafems
fv73_q8_t1c.dat Test 1 using Coupled Mass \verifman\nafems
fv73_q8_t2.dat Test 2 using Lumped Mass \verifman\nafems
fv73_q8_t2c.dat Test 2 using Coupled Mass \verifman\nafems
fv73_q8_t3.dat Test 3 using Lumped Mass \verifman\nafems
fv73_q8_t3c.dat Test 3 using Coupled Mass \verifman\nafems
fv73_q8_t4.dat Test 4 using Lumped Mass \verifman\nafems
fv73_q8_t4c.dat Test 4 using Coupled Mass \verifman\nafems

Results
Natural frequencies (Hz) of modes 1-6.

Main Index
 CHAPTER 6.26 311
FV73 - Cantilevered Thin Square Plate

NAFEMS Values

Mode 1 Mode 2 Mode 3 Mode 4 Mode 5 Mode 6

Reference 0.421 0.129 2.582 3.306 3.753 6.555
Solution
Test 1 0.4174 1.020 2.564 3.302 3.769 6.805
Test 2 0.4174 1.020 2.597 3.345 3.888 7.517
Test 3 0.4175 1.021 2.677 3.365 4.035 7.495
Test 4 0.4184 1.032 2.850 3.571 5.466 -

Lumped Mass

Mode 1 Mode 2 Mode 3 Mode 4 Mode 5 Mode 6

Test 1 0.415 1.005 2.485 3.150 3.622 6.292
Test 2 0.415 1.006 2.509 3.180 3.713 6.901
Test 3 0.415 1.007 2.563 3.196 3.828 6.879
Test 4 0.416 1.015 2.711 3.272 4.935 -

Coupled Mass

Mode 1 Mode 2 Mode 3 Mode 4 Mode 5 Mode 6

Test 1 0.418 1.023 2.579 3.298 3.765 6.719
Test 2 0.418 1.023 2.605 3.332 3.862 7.399
Test 3 0.418 1.024 2.672 3.344 4.000 7.387
Test 4 0.419 1.034 2.836 3.535 5.360 -

Conclusion
While the results using lumped mass are good, the results using coupled mass is closer to the reference
solution than the NAFEMS expected values in all cases.

Main Index
 312 MSC Nastran Verification Guide

Forced Vibration

Main Index
 CHAPTER 6.27 313
Test 5 - Deep Simply Supported Beam

6.27 Test 5 - Deep Simply

Supported Beam

Problem Description
Normal modes of a simply supported beam model.

Solution Number
SOL 103 - normal modes using Lanczos method

Features Used
 Normal modes calculation using 3-noded BEAM3 elements

Reference
1. NAFEMS Finite Element Methods & Standards, Abbassian, F., Dawswell, D. J., and Knowles, N.
C.Selected Benchmarks for Natural Frequency Analysis, Test No. 5. Glasgow: NAFEMS, Nov.,
1987.

Modeling Techniques Used

A simply supported beam model is solved for the first 9 modes.

Illustrations of the Model

Geometry and Mesh
Exact beam: 5 elements
Iso-parametric: 5 elements

Main Index
 314 MSC Nastran Verification Guide

Model File Name and Location

Model Name Description Location

Test5.dat Simply supported Beam model - lumped mass \verifman\nafems
Test5c.dat Simply supported Beam model - coupled mass \verifman\nafems

Results

NAFEMS
Theoretical NAFEMS Iso BEAM3
Mode Value Value Results
1 and 2 42.650 42.657 43.185
3 71.20 71.202 71.209
4 125.00 125.00 125.002
5 and 6 148.15 148.71 154.424
7 213.61 213.72 213.848
8 and 9 283.47 286.91 307.523

Conclusion
The normal recommendation is to have 8 elements per sine wave of the deformed shape. With only 5
elements in the model, the first mode is the only mode which can be expected to be accurate. Modes 1
through 4 are the first bending, axial, and torsional modes and are predicted well using this model. The other
modes are not predicted as well, but mesh refinement would improve the accuracy of these modes.

Main Index
 CHAPTER 6.28 315
Test 5H - Deep Simply Supported Beam Harmonic Forced Response

Test 5H - Deep Simply

6.28 Supported Beam Harmonic
Forced Response

Problem Description
Forced harmonic vibration response of a simply supported beam model.

Solution Number
SOL 108 and 111 - Direct and Modal Frequency response analysis

Features Used
 Frequency response calculation using 3-noded BEAM3 elements

Reference
1. NAFEMS Finite Element Methods & Standards, Abbassian, F., Dawswell, D. J., and Knowles, N.
C.Selected Benchmarks for Natural Frequency Analysis, Test No. 5H. Glasgow: NAFEMS, Nov.,
1987.

Modeling Techniques Used

A simply supported beam model is solved for a harmonic loading. In the direct method, Rayleigh damping
is used with a0=5.35 and a1=7.46 ×10-5. In the modal approach, modal damping of 2% of critical is used on
the first 16 modes.
A uniform load of 106N/m is applied in the Y-direction over the frequency range from 0.0 to 60.0Hz.
As the NAFEMS baseline does not use residual vectors, they are turned off in the modal solution. In a normal
run, they would be enabled to improve the accuracy of the solution.

Illustrations of the Model

Geometry and Mesh
Exact beam: 5 elements

Main Index
 316 MSC Nastran Verification Guide

Iso-parametric: 5 elements

Model File Name and Location

Model Name Description Location

Test5hd.dat Simply supported Beam model - Forced Harmonic \verifman\nafems
Vibration, direct solution.
Test5hm.dat Simply supported Beam model - Forced Harmonic \verifman\nafems
Vibration, modal solution.

Results

Peak Disp Peak Stress Frequency

Mode (mm) (N/mm2) (Hz)
Reference Solution 13.45 241.9 42.65
Target Solution (modal) 13.44 240.3 42.60
Target Solution (direct) 13.39 242.0 42.58
Modal Solution 13.14 238.4 43.185
Direct Solution 13.27 240.9 43.185

Conclusion
The maximum response occurs at the first natural frequency of the system, which for the 3-noded beam
model is 43.185hz. The peak values are within engineering accuracy of the predicted values.

Main Index
 CHAPTER 6.29 317
Test 5P - Deep Simply Supported Beam Periodic Forced Vibration Response

Test 5P - Deep Simply

6.29 Supported Beam Periodic
Forced Vibration Response

Problem Description
Forced periodic vibration response of a simply supported beam model.

Solution Number
SOL 108 and 111 - Direct and Modal Frequency response analysis

Features Used
 Frequency response calculation using 3-noded BEAM3 elements

Reference
1. NAFEMS Finite Element Methods & Standards, Abbassian, F., Dawswell, D. J., and Knowles, N.
C.Selected Benchmarks for Natural Frequency Analysis, Test No. 5P. Glasgow: NAFEMS, Nov.,
1987.

Modeling Techniques Used

A simply supported beam model is solved for a periodic loading. In the direct method, Rayleigh damping is
used with a0=5.35 and a1=7.46 ×10-5. In the modal approach, modal damping of 2% of critical is used on
the first 16 modes.
A uniform load of 106 N/m is applied in the Y-direction using the formula:
F=F0(sint - sin3t)
where,
F0 = 106 N/m
 = 2f
f = 20 Hz

Main Index
 318 MSC Nastran Verification Guide

Although this appears to be a transient solution, it is really a frequency response solution, using Fourier
transforms and inverse Fourier transforms.

Illustrations of the Model

Illustration of the model from the reference:

Geometry and Mesh

Exact beam: 5 elements
Iso-parametric: 5 elements

Model File Name and Location

Model Name Description Location

Test5pd.dat Simply supported Beam model - Forced Periodic \verifman\nafems
Vibration, direct solution.
Test5pm.dat Simply supported Beam model - Forced Periodic \verifman\nafems
Vibration, modal solution.

Results

Mode Peak Disp (mm) Peak Stress (N/mm2)

Reference Solution 0.951 17.10
Target Solution (modal) 0.949 17.05
Target Solution (direct) 0.953 17.33
Modal Solution 0.953 17.38
Direct Solution 0.953 17.38

Main Index
 CHAPTER 6.29 319
Test 5P - Deep Simply Supported Beam Periodic Forced Vibration Response

Conclusion
As residual vectors were used, the modal and direct methods gave the same results (to the number of
significant digits shown), which were within engineering accuracy of the reference solution.

Main Index
 320 MSC Nastran Verification Guide

Test 5T-Deep Simply

6.30 Supported Beam Transient
Forced Vibration Response

Problem Description
Forced transient vibration response of a simply supported beam model.

Solution Number
SOL 109 and 112 - Direct and Modal transient response analysis

Features Used
 Transient response calculation using 3-noded BEAM3 elements

Reference
1. NAFEMS Finite Element Methods & Standards, Abbassian, F., Dawswell, D. J., and Knowles, N.
C.Selected Benchmarks for Natural Frequency Analysis, Test No. 5T. Glasgow: NAFEMS, Nov.,
1987.

Modeling Techniques Used

A simply supported beam model is solved for a periodic loading. In the direct method, Rayleigh damping is
used with a0=5.35 and a1=7.46x10-5. In the modal approach, modal damping of 2% of critical is used on the
first 16 modes.
A load of 106N/m is suddenly applied in the Y-direction:

Forcing Function (Temporal)

Suddenly applied step load

Main Index
 CHAPTER 6.30 321
Test 5T-Deep Simply Supported Beam Transient Forced Vibration Response

F0 = 106 N/m

Illustrations of the Model

Geometry and Mesh
Exact beam: 5 elements
Iso-parametric: 5 elements

Model File Name and Location

Model Name Description Location

Test5td.dat Simply supported Beam model - Forced Transient \verifman\nafems
Vibration, direct solution
Test5tm.dat Simply supported Beam model - Forced Transient \verifman\nafems
Vibration, modal solution

Results

Peak Disp PSD Tp Peak Stress Static displacement

Solution (mm) (sec) (N/mm2) (mm)
Reference Solution 1.043 0.0117 18.76 0.538
Target Solution (modal) 1.040 0.0116 18.71 0.536
Target Solution (direct) 1.057 0.0117 18.77 0.536
Modal Solution 1.016 0.0117 18.15 0.525
Direct Solution 1.016 0.0117 18.15 0.525

Main Index
 322 MSC Nastran Verification Guide

Plot of displacement vs time for the mid-point of the beam

Conclusion
As residual vectors were used, the modal and direct methods gave the same results (to the number of
significant digits shown), which were within engineering accuracy of the reference solution.

Main Index
 CHAPTER 6.31 323
Test 5R-Deep Simply Supported Beam Random Forced Response

Test 5R-Deep Simply

6.31 Supported Beam Random
Forced Response

Problem Description
Forced random vibration response of a simply supported beam model.

Solution Number
SOL 108 and 111 - Direct and Modal Frequency response analysis

Features Used
 Frequency response calculation using BEAM elements
 Random response analysis

Reference
1. NAFEMS Finite Element Methods & Standards, Abbassian, F., Dawswell, D. J., and Knowles, N.
C.Selected Benchmarks for Natural Frequency Analysis, Test No. 5R. Glasgow: NAFEMS, Nov.,
1987.

Modeling Techniques Used

A simply supported beam model is solved for a harmonic loading, followed by random response analysis. In
the direct method, Rayleigh damping is used with a0=5.35 and a1=7.46x10-5. In the modal approach, modal
damping of 2% of critical is used on the first 16 modes.
A uniform load of 106 N/m is applied in the Y-direction over the frequency range from 0.0 to 60.0 Hz to
calculate the transfer functions, then the PSD provided is applied to obtain the random response.

Illustrations of the Model

Geometry and Mesh
Exact beam: 5 elements

Main Index
 324 MSC Nastran Verification Guide

Iso-parametric: 5 elements

Forcing Function (Temporal)

Random forcing with uniform power spectral density (of force)
PSD = (106 N/m)2 / Hz

Main Index
 CHAPTER 6.31 325
Test 5R-Deep Simply Supported Beam Random Forced Response

Model File Name and Location

Model Name Description Location

Test5rd.dat Simply supported Beam model - Forced Random \verifman\nafems
Vibration, direct solution
Test5rm.dat Simply supported Beam model - Forced Random \verifman\nafems
Vibration, modal solution

Results

Displacement PSD from Modal Run

Peak Disp PSD Peak Stress PSD Frequency

Solution (mm2/hz) (N/mm2)2/hz (Hz)
Reference Solution 180.90 58516 42.65
Target Solution (modal) 179.35 53722 42.58

Main Index
 326 MSC Nastran Verification Guide

Peak Disp PSD Peak Stress PSD Frequency

Solution (mm2/hz) (N/mm2)2/hz (Hz)
Target Solution (direct) 179.29 42.53
Modal Solution 182.6 59388 43.166
Direct Solution 186.4 60623 43.166

Conclusion
The maximum response occurs at the first natural frequency of the system, which for this model is 43.166hz.
The peak values are within engineering accuracy of the predicted values.

Main Index
 CHAPTER 6.32 327
Test 13: Simply Supported Thin Square Plate: Frequency Extraction

Test 13: Simply Supported

6.32 Thin Square Plate:
Frequency Extraction

Problem Description
Modal analysis of a thin square plate for extraction of frequencies.

Solution Number
SOL 103

Features Used
 Modal analysis using QUAD4 and QUAD8 shell elements

Reference
1. Test 13 from NAFEMS Selected Benchmarks for Forced Vibration, R0016, March 1993.

Modeling Techniques Used

A simply supported thin plate is solved for a modal. The model details are given below:

Material Properties
E = 200 x 109 N/m2  = 0.3  = 8000 kg/m3

Loading
Steady state harmonic forcing function F = F0 sint

F0 = 100 N/m2 over whole plate with  = 2f and f = 4.16 Hz.

Boundary Conditions
Ux = Uy = Rz = 0.0 at all nodes and Uz = 0 along all 4 edges
Rx = 0.0 along edges x=0 and x=10m and Ry = 0.0 along edges y=0 and y=10m.

Main Index
 328 MSC Nastran Verification Guide

Illustrations of the Model

Figure 6.32-1 Simply Supported Thin Square Plate

Model File Name and Location

Model Name Description Location

test13_q4.dat Test13: QUAD4 modal analysis \verifman\nafems
test13_q8.dat Test13: QUAD8 modal analysis \verifman\nafems

Main Index
 CHAPTER 6.32 329
Test 13: Simply Supported Thin Square Plate: Frequency Extraction

Results

Figure 6.32-2 Mode Shapes of Thin Square Plate

NAFEMS MSC Nastran (QUAD4) MSC Nastran (QUAD8)

Mode Hz Hz % Error Hz % Error
1 2.377 2.33179 -1.90% 2.375694 -0.05%
2 and 3 5.942 5.79671 -2.45% 5.938033 -0.07%
4 9.507 8.96345 -5.72% 9.747177 2.53%
5 and 6 11.884 11.67295 -1.78% 11.87156 -0.10%
7 and 8 15.449 14.45277 -6.45% 16.56407 7.22%

Conclusion
The results from MSC Nastran are compared with the NAFEMS results in the above tables.

Main Index
 330 MSC Nastran Verification Guide

Test 13H: Simply Supported

6.33 Thin Square Plate: Harmonic
Forced Vibration

Problem Description
Forced harmonic vibration response of a thin square plate with modal and direct solutions.

Solution Number
SOL 108 and SOL 111

Features Used
 Harmonic response calculation using QUAD4 and QUAD8 shell elements

Reference
1. Test 13H from NAFEMS Selected Benchmarks for Forced Vibration, R0016, March 1993.

Modeling Techniques Used

A simply supported thin plate is solved for a harmonic loading. In the direct method, Rayleigh damping is
used with a0 = 0.299 and a1 = 1.339×10-3. In the modal approach, modal damping of 2% of critical is used
on the first 16 modes. The model details are given below:

Material Properties
E = 200 x 109 N/m2  = 0.3  = 8000 kg/m3

Loading
Steady state harmonic forcing function F = F0 sint

F0 = 100 N/m2 over whole plate with  = 2f and f = 4.16 Hz.

Boundary Conditions
Ux = Uy = Rz = 0.0 at all nodes and Uz = 0 along all 4 edges
Rx = 0.0 along edges x=0 and x=10m and Ry = 0.0 along edges y=0 and y=10m.

Main Index
 CHAPTER 6.33 331
Test 13H: Simply Supported Thin Square Plate: Harmonic Forced Vibration

Illustrations of the Model

Figure 6.33-1 Simply Supported Thin Square Plate

Main Index
 332 MSC Nastran Verification Guide

Model File Name and Location

Model Name Description Location

test13hd_q4.dat Test13H: QUAD4 direct solution \verifman\nafems
test13hd_q8.dat Test13H: QUAD8 direct solution \verifman\nafems
test13hm_q4.dat Test13H: QUAD4 modal solution \verifman\nafems
test13hm_q8.dat Test13H: QUAD8 modal solution \verifman\nafems

Results

Peak Displacement  (mm)

MSC Nastran
NAFEMS Values % Error
QUAD4 (Modal) 45.42 45.9934 1.26%
QUAD4 (Direct) 45.42 45.9632 1.20%
QUAD8 (Modal) 45.42 45.2967 -0.27%
QUAD8 (Direct) 45.42 45.2756 -0.32%

Peak Stress  (N/mm2)

MSC Nastran
NAFEMS Values % Error
QUAD4 (Modal) 30.03 31.0656 3.45%
QUAD4 (Direct) 30.03 31.0442 3.38%
QUAD8 (Modal) 30.03 35.0256 16.64%
QUAD8 (Direct) 30.03 35.0075 16.58%

Frequency f (Hz)
MSC Nastran
NAFEMS Values % Error
QUAD4 (Modal) 2.377 2.3318 -1.90%

Main Index
 CHAPTER 6.33 333
Test 13H: Simply Supported Thin Square Plate: Harmonic Forced Vibration

Frequency f (Hz)
MSC Nastran
NAFEMS Values % Error
QUAD4 (Direct) 2.377 2.3318 -1.90%
QUAD8 (Modal) 2.377 2.3757 -0.05%
QUAD8 (Direct) 2.377 2.3757 -0.05%

Conclusion
The results from MSC Nastran are compared with the NAFEMS results in the above tables.

Main Index
 334 MSC Nastran Verification Guide

Test 13P: Simply Supported

6.34 Thin Square Plate: Periodic
Forced Vibration

Problem Description
Forced periodic vibration response of a thin square plate with modal and direct solutions.

Solution Number
SOL 108 and SOL 111

Features Used
 Periodic response calculation using QUAD4 and QUAD8 shell elements

Reference
1. Test 13P from NAFEMS Selected Benchmarks for Forced Vibration, R0016, March 1993.

Modeling Techniques Used

A simply supported thin plate is solved for a periodic loading. In the direct method, Rayleigh damping is used
with a0 = 0.299 and a1 = 1.339×10-3. In the modal approach, modal damping of 2% of critical is used on the
first 16 modes. The model details are given below:

Material Properties
E = 200 x 109 N/m2  = 0.3  = 8000 kg/m3

Loading
Steady state periodic forcing function F = F0 (sint - sin3t)

F0 = 100 N/m2 over whole plate with  = 2f and f = 1.2 Hz.

Boundary Conditions
Ux = Uy = Rz = 0.0 at all nodes and Uz = 0 along all 4 edges
Rx = 0.0 along edges x=0 and x=10m and Ry = 0.0 along edges y=0 and y=10m.

Main Index
 CHAPTER 6.34 335
Test 13P: Simply Supported Thin Square Plate: Periodic Forced Vibration

Illustrations of the Model

Figure 6.34-1 Simply Supported Thin Square Plate

Model File Name and Location

Model Name Description Location

test13pd_q4.dat Test13P: QUAD4 direct solution \verifman\nafems
test13pd_q8.dat Test13P: QUAD8 direct solution \verifman\nafems
test13pm_q4.dat Test13P: QUAD4 modal solution \verifman\nafems
test13pm_q8.dat Test13P: QUAD8 modal solution \verifman\nafems

Main Index
 336 MSC Nastran Verification Guide

Results

Figure 6.34-2 Displacement versus Time plot at node 41

Figure 6.34-3 Stress versus Time plot at node 41

Main Index
 CHAPTER 6.34 337
Test 13P: Simply Supported Thin Square Plate: Periodic Forced Vibration

Peak Displacement  (mm) Peak Stress  (N/mm2)

MSC Nastran MSC Nastran
NAFEMS Values % Error NAFEMS Values % Error
QUAD4 (Modal) 2.863 2.8519 -0.39% 2.018 1.9526 -3.24%
QUAD4 (Direct) 2.863 2.8513 -0.41% 2.018 1.9510 -3.32%
QUAD8 (Modal) 2.863 2.8682 0.18% 2.018 2.2714 12.56%
QUAD8 (Direct) 2.863 2.8675 0.16% 2.018 2.2690 12.44%

Conclusion
The displacement and stress results from the direct solution with QUAD4 elements are shown in Figure 6.34-2
and Figure 6.34-3. The results from MSC Nastran are compared with the NAFEMS results in the above table.

Main Index
 338 MSC Nastran Verification Guide

Test 13T: Simply Supported

6.35 Thin Square Plate: Transient
Forced Vibration

Problem Description
Forced transient vibration response of a thin square plate with modal and direct solutions.

Solution Number
SOL 109 and SOL 112

Features Used
 Transient response calculation using QUAD4 and QUAD8 shell elements.

Reference
1. Test 13T from NAFEMS Selected Benchmarks for Forced Vibration, R0016, March 1993.

Modeling Techniques Used

A simply supported thin plate is solved for a transient loading. In the direct method, Rayleigh damping is
used with a0 = 0.299 and a1 = 1.339×10-3. In the modal approach, modal damping of 2% of critical is used
on the first 16 modes. The model details are given below:

Material Properties
E = 200 x 109 N/m2  = 0.3  = 8000 kg/m3

Loading
Suddenly applied step load F0 = 100 N/m2 over whole plate.

Boundary Conditions
Ux = Uy = Rz = 0.0 at all nodes and Uz = 0 along all 4 edges
Rx = 0.0 along edges x=0 and x=10m and Ry = 0.0 along edges y=0 and y=10m.

Main Index
 CHAPTER 6.35 339
Test 13T: Simply Supported Thin Square Plate: Transient Forced Vibration

Illustrations of the Model

Figure 6.35-1 Simply Supported Thin Square Plate

Model File Name and Location

Model Name Description Location

test13td_q4.dat Test13T: QUAD4 direct solution \verifman\nafems
test13td_q8.dat Test13T: QUAD8 direct solution \verifman\nafems
test13tm_q4.dat Test13T: QUAD4 modal solution \verifman\nafems
test13tm_q8.dat Test13T: QUAD8 modal solution \verifman\nafems

Main Index
 340 MSC Nastran Verification Guide

Figure 6.35-2 Displacement versus Time plot at node 41

Figure 6.35-3 Stress versus Time plot at node 41

Main Index
 CHAPTER 6.35 341
Test 13T: Simply Supported Thin Square Plate: Transient Forced Vibration

Results

Peak Displacement p (mm) Peak Displacement tp (sec)

MSC Nastran MSC Nastran
NAFEMS Values % Error NAFEMS Values % Error
QUAD4 (Modal) 3.523 3.4951 -0.79% 0.21 0.2150 2.38%
QUAD4 (Direct) 3.523 3.5082 -0.42% 0.21 0.2150 2.38%
QUAD8 (Modal) 3.523 3.4329 -2.56% 0.21 0.2110 0.48%
QUAD8 (Direct) 3.523 3.4479 -2.13% 0.21 0.2110 0.48%

Peak Stress  (N/mm2) Static Displacement s (mm)

MSC Nastran MSC Nastran
NAFEMS Values % Error NAFEMS Values % Error
QUAD4 (Modal) 2.484 2.2601 -9.01% 1.817 1.8038 -0.73%
QUAD4 (Direct) 2.484 2.2534 -9.28% 1.817 1.8038 -0.73%
QUAD8 (Modal) 2.484 2.4977 0.55% 1.817 1.7647 -2.88%
QUAD8 (Direct) 2.484 2.4727 -0.45% 1.817 1.7647 -2.88%

Conclusion
The displacement and stress results from the direct solution with QUAD4 elements are shown in Figure 6.35-2
and Figure 6.35-3. The results from MSC Nastran are compared with the NAFEMS results in the above tables.

Main Index
 342 MSC Nastran Verification Guide

Test 13R: Simply Supported

6.36 Thin Square Plate: Random
Forced Vibration

Problem Description
Forced random vibration response of a thin square plate with modal and direct solutions.

Solution Number
SOL 108 and SOL 111

Features Used
 Random response calculation using QUAD4 and QUAD8 shell elements

Reference
1. Test 13R from NAFEMS Selected Benchmarks for Forced Vibration, R0016, March 1993.

Modeling Techniques Used

A simply supported thin plate is solved for a random loading. In the direct method, Rayleigh damping is used
with a0 = 0.299 and a1 = 1.339×10-3. In the modal approach, modal damping of 2% of critical is used on the
first 16 modes. The model details are given below:

Material Properties
E = 200 x 109 N/m2  = 0.3  = 8000 kg/m3

Loading
Random forcing with uniform power spectral density (of force) PSD = (100 N/m2)2 / Hz over whole plate.

Boundary Conditions
Ux = Uy = Rz = 0.0 at all nodes and Uz = 0 along all 4 edges
Rx = 0.0 along edges x=0 and x=10m and Ry = 0.0 along edges y=0 and y=10m.

Main Index
 CHAPTER 6.36 343
Test 13R: Simply Supported Thin Square Plate: Random Forced Vibration

Illustrations of the Model

Figure 6.36-1 Simply Supported Thin Square Plate

Model File Name and Location

Model Name Description Location

test13rd_q4.dat Test13R: QUAD4 direct solution \verifman\nafems
test13rd_q8.dat Test13R: QUAD8 direct solution \verifman\nafems
test13rm_q4.dat Test13R: QUAD4 modal solution \verifman\nafems
test13rm_q8.dat Test13R: QUAD8 modal solution \verifman\nafems

Main Index
 344 MSC Nastran Verification Guide

Results

Peak Displacement  PSD (mm2/Hz)

MSC Nastran
NAFEMS Values % Error
QUAD4 (Modal) 2063.2 2115.3920 2.53%
QUAD4 (Direct) 2063.2 2112.6130 2.39%
QUAD8 (Modal) 2063.2 2051.7860 -0.55%
QUAD8 (Direct) 2063.2 2049.8790 -0.65%

Peak Stress  PSD ((N/mm2)2/Hz)

MSC Nastran
NAFEMS Values % Error
QUAD4 (Modal) 1025.44 913.7688 -10.89%
QUAD4 (Direct) 1025.44 912.5110 -11.01%
QUAD8 (Modal) 1025.44 1226.8230 19.64%
QUAD8 (Direct) 1025.44 1225.5580 19.52%

Frequency f (Hz)
MSC Nastran
NAFEMS Values % Error
QUAD4 (Modal) 2.377 2.3318 -1.90%
QUAD4 (Direct) 2.377 2.3318 -1.90%
QUAD8 (Modal) 2.377 2.3757 -0.05%
QUAD8 (Direct) 2.377 2.3757 -0.05%

Conclusion
The results from MSC Nastran are compared with the NAFEMS results in the above tables.

Main Index
 CHAPTER 6.37 345
Test 21: Simply Supported Thick Square Plate: Frequency Extraction

Test 21: Simply Supported

6.37 Thick Square Plate:
Frequency Extraction

Problem Description
Modal analysis of a thick square plate for extraction of frequencies.

Solution Number
SOL 103

Features Used
 Modal analysis using QUAD4 and QUAD8 shell elements.

Reference
1. Test 21 from NAFEMS Selected Benchmarks for Forced Vibration, R0016, March 1993.

Modeling Techniques Used

A simply supported thick plate is solved for a modal. The model details are given below:

Material Properties
E = 200 x 109 N/m2  = 0.3  = 8000 kg/m3

Boundary Conditions
Ux = Uy = Rz = 0.0 at all nodes and Uz = 0 along all 4 edges
Rx = 0.0 along edges x=0 and x=10m and Ry = 0.0 along edges y=0 and y=10m.

Main Index
 346 MSC Nastran Verification Guide

Illustrations of the Model

Figure 6.37-1 Simply Supported Thick Square Plate

Model File Name and Location

Model Name Description Location

test21_q4.dat Test21: QUAD4 modal analysis \verifman\nafems
test21_q8.dat Test21: QUAD8 modal analysis \verifman\nafems

Main Index
 CHAPTER 6.37 347
Test 21: Simply Supported Thick Square Plate: Frequency Extraction

Figure 6.37-2 Mode Shapes of Thick Square Plate

Results

NAFEMS MSC Nastran (QUAD4) MSC Nastran (QUAD8)

Mode Hz Hz % Error Hz % Error
1 45.897 45.22394 -1.47% 46.14619 0.54%
2 and 3 109.44 107.28680 -1.97% 110.25390 0.74%
4 167.89 157.62500 -6.11% 166.90790 -0.58%
5 and 6 204.51 200.16360 -2.13% 204.93380 0.21%
7 and 8 256.5 234.84710 -8.44% 248.16460 -3.25%

Conclusion
The results from MSC Nastran are compared with the NAFEMS results in the above table.

Main Index
 348 MSC Nastran Verification Guide

Test 21H: Simply Supported

6.38 Thick Square Plate:
Harmonic Forced Vibration

Problem Description
Forced harmonic vibration response of a thick square plate with modal and direct solutions.

Solution Number
SOL 108 and SOL 111

Features Used
 Harmonic response calculation using QUAD4 and QUAD8 shell elements.

Reference
1. Test 21H from NAFEMS Selected Benchmarks for Forced Vibration, R0016, March 1993.

Modeling Techniques Used

A simply supported thin plate is solved for a harmonic loading. In the direct method, Rayleigh damping is
used with a0 = 5.772 and a1 = 6.929×10-5. In the modal approach, modal damping of 2% of critical is used
on the first 16 modes. The model details are given below:

Material Properties
E = 200 x 109 N/m2  = 0.3  = 8000 kg/m3

Loading
Steady state harmonic forcing function F = F0 sint

F0 = 106 N/m2 over whole plate with  =2f and f = 0 to 78.17 Hz.

Boundary Conditions
Ux = Uy = Rz = 0.0 at all nodes and Uz = 0 along all 4 edges
Rx = 0.0 along edges x=0 and x=10m and Ry = 0.0 along edges y=0 and y=10m.

Main Index
 CHAPTER 6.38 349
Test 21H: Simply Supported Thick Square Plate: Harmonic Forced Vibration

Illustrations of the Model

Figure 6.38-1 Simply Supported Thick Square Plate

Model File Name and Location

Model Name Description Location

Test21hd_q4.dat Test21H: QUAD4 direct solution \verifman\nafems
Test21hd_q8.dat Test21H: QUAD8 direct solution \verifman\nafems
Test21hm_q4.dat Test21H: QUAD4 modal solution \verifman\nafems
Test21hm_q8.dat Test21H: QUAD8 modal solution \verifman\nafems

Main Index
 350 MSC Nastran Verification Guide

Results

Peak Displacement  (mm)

MSC Nastran
NAFEMS Values % Error
QUAD4 (Modal) 58.33 61.1372 4.81%
QUAD4 (Direct) 58.33 61.1334 4.81%
QUAD8 (Modal) 58.33 60.0149 2.89%
QUAD8 (Direct) 58.33 60.0176 2.89%

Peak Stress  (N/mm2)

MSC Nastran
NAFEMS Values % Error
QUAD4 (Modal) 800.8 776.6374 -3.02%
QUAD4 (Direct) 800.8 776.5670 -3.03%
QUAD8 (Modal) 800.8 877.5689 9.59%
QUAD8 (Direct) 800.8 877.5689 9.59%

Frequency f (Hz)
MSC Nastran
NAFEMS Values % Error
QUAD4 (Modal) 45.9 45.2339 -1.47%
QUAD4 (Direct) 45.9 45.2339 -1.47%
QUAD8 (Modal) 45.9 46.1462 0.54%
QUAD8 (Direct) 45.9 46.1462 0.54%

Conclusion
The results from MSC Nastran are compared with the NAFEMS results in the above tables.

Main Index
 CHAPTER 6.39 351
Test 21P: Simply Supported Thick Square Plate: Periodic Forced Vibration

Test 21P: Simply Supported

6.39 Thick Square Plate: Periodic
Forced Vibration

Problem Description
Forced periodic vibration response of a thick square plate with modal and direct solutions.

Solution Number
SOL 108 and SOL 111

Features Used
 Harmonic response calculation using QUAD4 and QUAD8 shell elements.

Reference
1. Test 21P from NAFEMS Selected Benchmarks for Forced Vibration, R0016, March 1993.

Modeling Techniques Used

A simply supported thin plate is solved for a harmonic loading. In the direct method, Rayleigh damping is
used with a0 = 5.772 and a1 = 6.929×10-5. In the modal approach, modal damping of 2% of critical is used
on the first 16 modes. The model details are given below:

Material Properties
E = 200 x 109 N/m2  = 0.3  = 8000 kg/m3

Loading
Steady state harmonic forcing function F = F0 (sint - sin3t)

F0 = 106 N/m2 over whole plate with  =2f and f = 20 Hz.

Main Index
 352 MSC Nastran Verification Guide

Boundary Conditions
Ux = Uy = Rz = 0.0 at all nodes and Uz = 0 along all 4 edges
Rx = 0.0 along edges x=0 and x=10m and Ry = 0.0 along edges y=0 and y=10m.

Illustrations of the Model

Figure 6.39-1 Simply Supported Thick Square Plate

Model File Name and Location

Model Name Description Location

Test21pd_q4.dat Test21H: QUAD4 direct solution \verifman\nafems
Test21pd_q8.dat Test21H: QUAD8 direct solution \verifman\nafems
Test21pm_q4.dat Test21H: QUAD4 modal solution \verifman\nafems
Test21pm_q8.dat Test21H: QUAD8 modal solution \verifman\nafems

Main Index
 CHAPTER 6.39 353
Test 21P: Simply Supported Thick Square Plate: Periodic Forced Vibration

Results

Figure 6.39-2 Displacement versus Time plot at node 41

Figure 6.39-3 Stress versus Time plot at node 41

Main Index
 354 MSC Nastran Verification Guide

Results

Peak Displacement  (mm) Peak Stress  (N/mm2)

MSC Nastran MSC Nastran
NAFEMS Values % Error NAFEMS Values % Error
QUAD4 (Modal) 4.929 4.9828 1.09% 67.67 64.1107 -5.26%
QUAD4 (Direct) 4.929 4.8454 -1.70% 67.67 62.2009 -8.08%
QUAD8 (Modal) 4.929 5.1885 5.27% 67.67 77.4392 14.44%
QUAD8 (Direct) 4.929 4.9322 0.06% 67.67 73.5257 8.65%

Conclusion
The displacement and stress results from the direct solution with QUAD4 elements are shown in Figure 6.39-2
and Figure 6.39-3. The results from MSC Nastran are compared with the NAFEMS results in the above table.

Main Index
 CHAPTER 6.40 355
Test 21T: Simply Supported Thick Square Plate: Transient Forced Vibration

Test 21T: Simply Supported

6.40 Thick Square Plate: Transient
Forced Vibration

Problem Description
Forced transient vibration response of a thick square plate with modal and direct solutions.

Solution Number
SOL 109 and 112

Features Used
 Random transient calculation using QUAD4 and QUAD8 shell elements

Reference
1. Test 21T from NAFEMS Selected Benchmarks for Forced Vibration, R0016, March 1993.

Modeling Techniques Used

A simply supported thick plate is solved for a transient loading. In the direct method, Rayleigh damping is
used with a0 = 5.772 and a1 = 6.929 x 10-5. In the modal approach, modal damping of 2% of critical is used
on the first 16 modes. The model details are given below:

Material Properties
E = 200 x 109 N/m2
 = 0.3
 = 8000 kg/m3

Loading
Suddenly applied step load F0 = 106 N/m2 over whole plate

Boundary Condition
ux = uy = Rz = 0.0 at all nodes and Uz = 0 along all 4 edges

Main Index
 356 MSC Nastran Verification Guide

Rx = 0.0 along edges x=0 and x=10m and Ry = 0.0 along edges y=0 and y=10m

Illustrations of the Model

Figure 6.40-1 Simply Supported Thick Square Plate

Main Index
 CHAPTER 6.40 357
Test 21T: Simply Supported Thick Square Plate: Transient Forced Vibration

Model File Name and Location

Model Name Description Location

Test21td_q4.dat Test21T: QUAD4 direct solution tpl\verifman\nafems
Test21td_q8.dat Test21T: QUAD8 direct solution
Test21tm_q4.dat Test21T: QUAD4 modal solution
Test21tm_q8.dat Test21T: QUAD8 modal solution

Results

Figure 6.40-2 Displacement versus Time plot at node 41

Main Index
 358 MSC Nastran Verification Guide

Figure 6.40-3 Stress versus Time plot at node 41

Peak Displacement p (mm)

MSC Nastran
NAFEMS Values % Error
QUAD4 (Modal) 4.524 4.7026 3.95%
QUAD4 (Direct) 4.524 4.6947 3.77%
QUAD8 (Modal) 4.524 4.6047 1.78%
QUAD8 (Direct) 4.524 4.5977 1.63%

Peak Stress  (N/mm2)

MSC Nastran
NAFEMS Values % Error
QUAD4 (Modal) 62.11 59.1492 -4.77%
QUAD4 (Direct) 62.11 58.6496 -5.57%
QUAD8 (Modal) 62.11 66.3350 6.80%
QUAD8 (Direct) 62.11 65.7299 5.83%

Main Index
 CHAPTER 6.40 359
Test 21T: Simply Supported Thick Square Plate: Transient Forced Vibration

Peak Response Time (tp)

MSC Nastran
NAFEMS Values % Error
QUAD4 (Modal) 0.0108 0.0108 0.00%
QUAD4 (Direct) 0.0108 0.0109 0.93%
QUAD8 (Modal) 0.0108 0.0105 -2.78%
QUAD8 (Direct) 0.0108 0.0107 -0.93%

Static Displacement s (mm)

MSC Nastran
NAFEMS Values % Error
QUAD4 (Modal) 2.333 2.3816 2.08%
QUAD4 (Direct) 2.333 2.3816 2.08%
QUAD8 (Modal) 2.333 2.3273 -0.25%
QUAD8 (Direct) 2.333 2.3274 -0.24%

Conclusion
The displacement and stress results from the direct solution with QUAD4 elements are shown in Figure 6.40-2
and Figure 6.40-3. The results from MSC Nastran are compared with the NAFEMS results in the above tables.

Main Index
 360 MSC Nastran Verification Guide

Test 21R: Simply Supported

6.41 Thick Square Plate: Random
Forced Vibration

Problem Description
Forced random vibration response of a thick square plate with modal and direct solutions.

Solution Number
SOL 108 and 111

Features Used
 Random response calculation using QUAD4 and QUAD8 shell elements

Reference
1. Test 21R from NAFEMS Selected Benchmarks for Forced Vibration, R0016, March 1993.

Modeling Techniques Used

A simply supported thick plate is solved for a random loading. In the direct method, Rayleigh damping is
used with a0 = 5.772 and a1 = 6.929 x 10-5. In the modal approach, modal damping of 2% of critical is used
on the first 16 modes. The model details are given below:
Material Properties
E = 200 x 109 N/m2
0.3
8000 kg/m3

Loading
Random forcing with uniform power spectral density (of force) PSD = (106 N/m2)2 / Hz over the whole
plate.

Boundary Condition
ux = uy = Rz = 0.0 at all nodes and Uz = 0 along all 4 edges

Main Index
 CHAPTER 6.41 361
Test 21R: Simply Supported Thick Square Plate: Random Forced Vibration

Rx = 0.0 along edges x=0 and x=10m and Ry = 0.0 along edges y=0 and y=10m

Illustrations of the Model

Figure 6.41-1 Simply Supported Thick Square Plate

Model File Name and Location

Model Name Description Location

Test21rd_q4.dat Test21R: QUAD4 direct solution tpl\verifman\nafems
Test21rd_q8.dat Test21R: QUAD8 direct solution
Test21rm_q4.dat Test21R: QUAD4 modal solution
Test21rm_q8.dat Test21R: QUAD8 modal solution

Main Index
 362 MSC Nastran Verification Guide

Results

Peak  PSD (mm2/Hz)

MSC Nastran
NAFEMS Values % Error
QUAD4 (Modal) 3401.81 3737.7530 9.88%
QUAD4 (Direct) 3401.81 3737.2870 9.86%
QUAD8 (Modal) 3401.81 3601.7850 5.88%
QUAD8 (Direct) 3401.81 3602.1140 5.89%

Peak  PSD ((N/mm2)2/Hz)

MSC Nastran
NAFEMS Values % Error
QUAD4 (Modal) 641200 603173.1 -5.93%
QUAD4 (Direct) 641200 603063.7 -5.95%
QUAD8 (Modal) 641200 770148.1 20.11%
QUAD8 (Direct) 641200 770148.2 20.11%

Frequency f (Hz)
MSC Nastran
NAFEMS Values % Error
QUAD4 (Modal) 45.9 45.2239 -1.47%
QUAD4 (Direct) 45.9 45.2239 -1.47%
QUAD8 (Modal) 45.9 46.1462 0.54%
QUAD8 (Direct) 45.9 46.1462 0.54%

Conclusion
The results from MSC Nastran are compared with the NAFEMS results in the above tables.

Main Index
 CHAPTER 6.41 363
Test 21R: Simply Supported Thick Square Plate: Random Forced Vibration

Nonlinear

Main Index
 364 MSC Nastran Verification Guide

6.42 NL4: Snap-back Under

Displacement Control

Problem Description
This is a NAFEMS nonlinear benchmark to demonstrate a geometric nonlinearity solution procedure test of
snap-back under displacement control. It is modeled using the modified Riks-Ramm arc length procedure
available in MSC Nastran.

Solution Number
SOL 400

Features Used
 Implicit nonlinear analysis.
 Modified Riks/Ramm arc length method.

Reference
1. Test NL4 from NAFEMS Publication NNB, Rev. 1, NAFEMS Non-Linear Benchmarks, October
1989 from a previous report Benchmark Tests for Solution Procedures for Geometric Non-Linearity
by Crisfield, Duxbury & Hunt. NAFEMS Report SPGNL, October 1987.

Modeling Techniques Used

The spring and element assembly are modeled using CELAS1 and CROD elements respectively. The model
details are given below:

Material Properties
AE = 5.0x107, L = 2500, L = 25, K1 = 1.5, K2 = AE/L(1 + 2)1/2 = 19999.0, K3 = 0.25, K4 = 1.0

Loading
Incremental load P=10000.0 is applied to node A in the X-direction using arc length method.

Boundary Condition
The displacements uy = 0.0 at node A and B, along with the displacement ux = 0.0 at node C.

Main Index
 CHAPTER 6.42 365
NL4: Snap-back Under Displacement Control

Illustrations of the Model

Figure 6.42-1 Snap-back under displacement control

Main Index
 366 MSC Nastran Verification Guide

Model File Name and Location

Model File Name Description Location

nafemnl4.dat NL4: Snap-back Under Displacement Control \verifman\nafems

Results

Figure 6.42-2 P versus UA Solution Path

Main Index
 CHAPTER 6.42 367
NL4: Snap-back Under Displacement Control

Figure 6.42-3 P versus UA Solution Path

Figure 6.42-4 P versus V Solution Path

Main Index
 368 MSC Nastran Verification Guide

Conclusion
The results from MSC Nastran are compared with NAFEMS results at loads P= 649.9, 1300.0, 1949.0,
2599.0, 3243.0 and -1099.0 in the above figures. The largest deviation from the NAFEMS solution is less
than 1% at all these reference points.

Main Index
 CHAPTER 6.43 369
NL6: Straight Cantilever with Axial End Point Load

6.43 NL6: Straight Cantilever with

Axial End Point Load

Problem Description
A cantilever beam with an axial compressive load combines bending and membrane deformation with
bifurcation of initially straight elements using the modified Riks-Ramm arc length procedure available in
MSC Nastran.

Solution Number
SOL 400

Features Used
 Implicit nonlinear analysis.
 Modified Riks/Ramm arc length method.

Reference
1. Test NL6 from NAFEMS Publication NNB, Rev. 1, NAFEMS Non-Linear Benchmarks, October
1989 as well as a previous report Finite Element Benchmarks for 2D beams and Axisymmetric shells
involving Geometric Non-Linearity by P Lyons and S Holsgrove NAFEMS Report FEBNLGBAS,
March 1989.

Modeling Techniques Used

The cantilever beam is modeled using 32 plane stress quadratic CQUAD8 elements. The model details are
given below:

Material Properties
The material is elastic with a Young's modulus of E = 210 x 109 N/m2 and a Poisson's ratio of 0.0.

Loading
The loading at end A is applied in increments up to a maximum value of PL2/EI = 22.493.

Boundary Condition
The displacements ux = uy = 0.0 at node 1 and ux = 0.0 at nodes 2 and 3 of the end B.

Main Index
 370 MSC Nastran Verification Guide

Illustrations of the Model

Figure 6.43-1 Straight Cantilever Beam with Axial End Point Load

Model File Name and Location

Model File Name Description Location

nafemnl6.dat NL6: Straight Cantilever With Axial End Point \verifman\nafems
Load

Results

Figure 6.43-2 PL2/EI versus UX/L Solution Path

Main Index
 CHAPTER 6.43 371
NL6: Straight Cantilever with Axial End Point Load

Figure 6.43-3 PL2/EI versus UY/L Solution Path

Conclusion
The results from MSC Nastran are matching well with NAFEMS results as shown in the above figures.

Main Index
 372 MSC Nastran Verification Guide

6.44 NL7: Lee's Frame Buckling

Problem

Problem Description
This example demonstrates the ability of MSC Nastran to model a beam column structure composed of
initially straight beams that will suddenly snap through and back. Modified Riks/Ramm & Crisfield arc
length methods are used to trace the solution path to this unstable problem.

Solution Number
SOL 400

Features Used
 Implicit nonlinear analysis.
 Modified Riks/Ramm and Crisfield arc length methods.

Reference
1. Finite Element Benchmarks for 2D beams and Axisymmetric shells involving Geometric Non-
Linearity by P Lyons and S Holsgrove NAFEMS Report FEBNLGBAS, March 1989.

Modeling Techniques Used

The frame is modeled using 100 CBAR elements. The model details are given below:

Material Properties
The material is elastic with a Young's modulus of E = 71.74 x 109 N/m2 and a Poisson's ratio of 0.0.

Loading
A concentrated load P at point C with an ultimate value of 50000 N is applied incrementally.

Boundary Condition
The beams are pinned at points B and D.

Main Index
 CHAPTER 6.44 373
NL7: Lee's Frame Buckling Problem

Illustrations of the Model

Figure 6.44-1 Lee's Frame Buckling Problem

Main Index
 374 MSC Nastran Verification Guide

Model File Name and Location

Model File Name Description Location

nafemnl7a.dat NL7: Lee's Frame Buckling Problem with MRIKS \verifman\nafems
method
nafemnl7b.dat NL7: Lee's Frame Buckling Problem with \verifman\nafems
Crisfield method

Results

Figure 6.44-2 Lee's Frame Buckling with Modified Riks/Ramm Method

Main Index
 CHAPTER 6.44 375
NL7: Lee's Frame Buckling Problem

Figure 6.44-3 Lee's Frame Buckling with Crisfield Method

Conclusion
The results from MSC Nastran are matching well with NAFEMS results as shown in the above figures for
both modified Riks/Ramm and Crisfield methods.

Main Index
 376 MSC Nastran Verification Guide

Fracture Mechanics

Main Index
 CHAPTER 6.45 377
Test 1.1: Center Cracked Plate in Tension

6.45 Test 1.1: Center Cracked

Plate in Tension

Problem Description
The following test case describe finite element analysis of a centre cracked plate tension specimen.
Normalized stress intensity factor for the test case is calculated using virtual crack extension method. The
results are compared to the analytical solution [1].

Solution Number
Sol 400

Features Used
 Implicit Nonlinear

Reference
1. Test 1.1 from NAFEMS Publication, Ref: R0020, 2D Test Cases in Linear Elastic Fracture
Mechanics, Jan,1992.

Modeling Techniques Used

Due to symmetry in loading and boundary conditions one-quarter of test geometry along the symmetry axis
is used for the analysis. The finite element model is shown in Figure 6.45-1. The model is meshed with
QUAD8 and quarter point TRIA6 elements. The quarter point elements are used at the crack tip. VCCT
bulk data entry is used to obtain the energy release rate.

Material Properties
The material is linear elastic with a Young's modulus, E = 20700 and Poisson's ratio  = 0.3

Geometrical properties
a/b = 0.5; h/b = 1.0; b = 20

Loading
Uniform stress of 100 unit is applied on top face as shown in Figure 6.45-1.

Main Index
 378 MSC Nastran Verification Guide

Boundary Condition
Symmetry boundary condition as shown in Figure 6.45-1.
uy = 0 on x axis
ux = 0 on y axis

Illustrations of the Model

Figure 6.45-1 One Quarter of Centre Cracked Plate

Model File Name and Location

Model Name Description Location

nvcct001 Centre Cracked Plate in Tension, Plain Strain tpl\verifman\nafems

Results
K 0.5
Reference solution [1] of the normalized stress intensity factor is ------I = 1.325 where K o =   a 
Ko

Virtual Crack Extension Method

The details of this method is explained in NAFEMS fracture mechanics benchmarks article [1].

Main Index
 CHAPTER 6.45 379
Test 1.1: Center Cracked Plate in Tension

Energy release rate (G)= 2.4665

GE 0.5
Stress intensity factor K I =  --------------2- = 749.0
 
1–

0.5
K o =   a  = 560.5

K
------I = 1.3364
Ko

Conclusion
The results from MSC Nastran are in a good agreement with the reference solution.

Main Index
 380 MSC Nastran Verification Guide

6.46 Test 1.2: Center Cracked

Plate with Thermal Load

Problem Description
The following test case describe finite element analysis of a center cracked plate subjected to quadratic
temperature distribution. Normalized stress intensity factor for the test case is calculated using virtual crack
extension method. The results are compared to the analytical solution [1].

Solution Number
Sol 400

Features Used
 Implicit Nonlinear

Reference
1. Test 1.2 from NAFEMS Publication, Ref: R0020, 2D Test Cases in Linear Elastic Fracture
Mechanics, Jan,1992.

Modeling Techniques Used

Due to symmetry in loading and boundary conditions one-quarter of test geometry along the symmetry axis
is used for the analysis. The finite element model is shown in Figure 6.45-1. The model is meshed with
QUAD8 and quarter point TRIA6 elements. The quarter point elements are used at the crack tip. VCCT
bulk data entry is used to obtain the energy release rate.

Material Properties
The material is linear elastic with a Young's modulus E = 20700 and a Poisson's ratio  = 0.3 and coefficient
of thermal expansion  =1.35 × 10-5

Geometrical properties
a/b = 0.1; h/b = 2.5; b = 100

Loading
Quadratic thermal distribution
2
T = 0.01  x

Main Index
 CHAPTER 6.46 381
Test 1.2: Center Cracked Plate with Thermal Load

Boundary Condition
Symmetry boundary condition as shown in Figure 6.45-1.
uy = 0 on x axis
ux = 0 on y axis

Illustrations of the Model

Figure 6.46-1 One Quarter of Centre Cracked Plate

Main Index
 382 MSC Nastran Verification Guide

Model File Name and Location

Model Name Description Location

nvcct002 Centre cracked plate with quadratic thermal distribution, Plane tpl\verifman\nafems
Stress

Results
K 2 2
------I = 1.00 ET  b----- a----- 0.5
Reference solution [1] of the normalized stress intensity factor is where K o = -----------
2  3
–  a 
Ko c 2

Virtual Crack Extension Method

The details of this method is explained in NAFEMS fracture mechanics benchmarks article [1].
Energy release rate (G)= 1.3004
Stress intensity factor K1 = (G)0.5 =518.9

K0 = (a)0.5 =514.3
K
------I = 1.008
Ko

Conclusion
The results from MSC Nastran are in a good agreement with the reference solution.

Main Index
 CHAPTER 6.47 383
Test 2.1: Single Edge Cracked Plate in Tension

6.47 Test 2.1: Single Edge

Cracked Plate in Tension

Problem Description
The following test case describe finite element analysis of a Single Edge Cracked Plate subjected to uniform
stress. Normalized stress intensity factor for the test case is calculated using virtual crack extension method.
The results are compared to the analytical solution [1].

Solution Number
Sol 400

Features Used
 Implicit Nonlinear

Reference
1. Test case 2.1 from NAFEMS Publication, Ref: R0020, 2D Test Cases in Linear Elastic Fracture
Mechanics, Jan,1992.

Modeling Techniques Used

Due to symmetry in loading and boundary conditions one-quarter of test geometry along the symmetry axis
is used for the analysis. The finite element model is shown in Figure 6.47-1. The model is meshed with
QUAD8 and quarter point TRIA6 elements. The quarter point elements are used at the crack tip. VCCT
bulk data entry is used to obtain the energy release rate.

Material Properties
The material is linear elastic with a Young's modulus of E = 20700 and a Poisson's ratio  = 0.3

Geometrical properties
a/b = 0.5; h/b = 0.5; b = 20

Loading
Uniform stress of 100 unit is applied on top face as shown in Figure 6.47-1.

Main Index
 384 MSC Nastran Verification Guide

Boundary Condition
Symmetry boundary condition as shown in Figure 6.47-1.
uy = 0 on x axis
ux = 0 at point A

Illustrations of the Model

Figure 6.47-1 Half of the Single Edge Cracked Plate

Model File Name and Location

Model Name Description Location

nvcct003 Single edge cracked plate: uniform tensile stress, Plane Strain tpl\verifman\nafems

Results
K 0.5
Reference solution [1] of the normalized stress intensity factor is ------I = 3.00 where K o =   a 
Ko

Virtual Crack Extension Method

The details of this method is explained in NAFEMS fracture mechanics benchmarks article [1].
Energy release rate (G)= 12.468
GE 0.5
Stress intensity factor K I =  --------------2- = 1684.1
 
1–

0.5
K o =   a  = 560.5

Main Index
 CHAPTER 6.47 385
Test 2.1: Single Edge Cracked Plate in Tension

K
------I = 3.00
Ko

Conclusion
The results from MSC Nastran are in a good agreement with the reference solution.

Main Index
 386 MSC Nastran Verification Guide

6.48 Test 3: Angle Crack

Embedded in a Plate

Problem Description
The following test case describe finite element analysis of an angle crack plate subjected to uniform stress.
Normalized stress intensity factor for the test case is calculated using virtual crack extension method. The
results are compared to the analytical solution [1].

Solution Number
Sol 400

Features Used
 Implicit Nonlinear

Reference
1. Test case 3.1 from NAFEMS Publication, Ref: R0020, 2D Test Cases in Linear Elastic Fracture
Mechanics, Jan,1992.

Modeling Techniques Used

The finite element model is shown in Figure 6.48-1. The model is meshed with QUAD8 and quarter point
TRIA6 elements. The quarter point elements are used at the crack tip. VCCT bulk data entry is used to
obtain the energy release rate.

Material Properties
The material is linear elastic with a Young's modulus of E = 20700 and a Poisson's ratio  = 0.3

Geometrical properties
a/b = 0.5; h/b = 1.25; b = 50

Loading
Uniform stress of 100 unit is applied on top face as shown in Figure 6.48-1.

Boundary Condition
Boundary condition is shown in Figure 6.48-1.

Main Index
 CHAPTER 6.48 387
Test 3: Angle Crack Embedded in a Plate

uy = 0 on x axis at y = 0
ux = 0 at point A

Illustrations of the Model

Figure 6.48-1 Angle Crack Plate

Model File Name and Location

Model Name Description Location

nvcct004 Angle crack plate embedded in a plate: Uniaxial tension, tpl\verifman\nafems
Plane Strain

Results
K K
Reference solution [1] of the normalized stress intensity factor is ------I = 0.190 ; ------II- = 0.405 where
Ko Ko
0.5
K o =   a 

Main Index
 388 MSC Nastran Verification Guide

Virtual Crack Extension Method

The details of this method is explained in NAFEMS fracture mechanics benchmarks article [2].
Energy release rate (G)= 0.6717
U
Ratio of mode II to mode I separation of crack face nodes ( R ) = -------- =2.1426
V

Stress intensity factor GE

K I =  --------------------------------------------- 0.5 = 165.32
 2 2 
 1 –    1 – R 

K II = RK I = 354.21

0.5
K o =   a  = 886.2

K K II
------I = 0.186 , ------- = 0.400
Ko Ko

Conclusion
The results from MSC Nastran are in a good agreement with the reference solution.

Main Index
 CHAPTER 6.49 389
Test 4: Cracks at a Hole in a Plate

6.49 Test 4: Cracks at a Hole in a

Plate

Problem Description
The following test case describe finite element analysis of cracks at a hole in a plate and subjected to uniform
stress. Normalized stress intensity factor for the test case is calculated using virtual crack extension method.
The results are compared to the analytical solution [1].

Solution Number
Sol 400

Features Used
 Implicit Nonlinear

Reference
1. Test case 4.1 from NAFEMS Publication, Ref: R0020, 2D Test Cases in Linear Elastic Fracture
Mechanics, Jan,1992.

Modeling Techniques Used

Due to symmetry in loading and boundary conditions, one-quarter of test geometry along the symmetry axis
is used for the analysis. The finite element model is shown in Figure 6.45-1. The model is meshed with
QUAD8 and quarter point TRIA6 elements. The quarter point elements are used at the crack tip. VCCT
bulk data entry is used to obtain the energy release rate.

Material Properties
The material is linear elastic with a Young's modulus E = 207000 and a Poisson's ratio = 0.3

Geometrical properties
a/b = 0.3; R/b = 0.25; b = 10

Loading
Uniform stress of 100 unit is applied on top face as shown in Figure 6.45-1.

Main Index
 390 MSC Nastran Verification Guide

Boundary Condition
Boundary condition as shown in Figure 6.45-1.
uy = 0 on x axis
ux = 0 on y axis

Illustrations of the Model

Figure 6.49-1 Cracks at a Hole in a Plate

Main Index
 CHAPTER 6.49 391
Test 4: Cracks at a Hole in a Plate

Model File Name and Location

Model Name Description Location

nvcct005 Cracks at a Hole in a Plate, Plane Stress tpl\verifman\nafems

Results
K 0.5
Reference solution [1] of the normalized stress intensity factor is ------I = 1.050 where K o =   a 
Ko

Virtual Crack Extension Method

The details of this method are explained in NAFEMS fracture mechanics benchmarks article [1].
Virtual Crack Extension Method:
Energy release rate (G)= 0.5198
0.5
Stress intensity factor K I =  GE  = 328.02

0.5
K o =   a  = 307.0

K
------I = 1.068
Ko

Conclusion
The results from MSC Nastran are in a good agreement with the reference solution.

Main Index
 392 MSC Nastran Verification Guide

6.50 Test 5: Axisymmetric Crack

in a Bar

Problem Description
The following test case describe finite element analysis of axisymmetric crack in a bar and subjected to
uniform stress. Normalized stress intensity factor for the test case is calculated using virtual crack extension
method. The results are compared to the analytical solution [1].

Solution Number
Sol 400

Features Used
 Implicit Nonlinear

Reference
1. Test case 5 from NAFEMS Publication, Ref: R0020, 2D Test Cases in Linear Elastic Fracture
Mechanics, Jan,1992.

Modeling Techniques Used

Due to symmetry in loading and boundary conditions one-quarter of test geometry along the symmetry axis
is used for the analysis. The finite element model is shown in Figure 6.45-1. The model is meshed with
CQUADX and quarter point CTRIAX elements. The quarter point elements are used at the crack tip. VCCT
bulk data entry is used to obtain the energy release rate.

Material Properties
The material is linear elastic with a Young's modulus, E = 207000 and Poisson's ratio, = 0.3

Geometrical properties
b/R = 0.5; h/R = 1.5; R = 20

Loading
Uniform stress of 100 unit is applied on top face as shown in Figure 6.45-1.

Main Index
 CHAPTER 6.50 393
Test 5: Axisymmetric Crack in a Bar

Boundary Condition
Boundary condition as shown in Figure 6.45-1.
uy = 0 on x axis
ux = 0 on y axis

Illustrations of the Model

Figure 6.50-1 Cracks at a Hole in a Plate

Main Index
 394 MSC Nastran Verification Guide

Model File Name and Location

Model Name Description Location

nvcct006 Axisymmetric Crack in a Bar, Plane Strain tpl\verifman\nafems

Results
K P 0.5
Reference solution [1] of the normalized stress intensity factor is ------I = 0.475 where K o = --------2-  b 
Ko b

Virtual Crack Extension Method

The details of this method are explained in NAFEMS fracture mechanics benchmarks article [1].
Virtual Crack Extension Method:
Energy release rate (G)= 5.181
GE 0.5
Stress intensity factor K I =  --------------2- = 1085.6
 
1–

P 0.5
K o = --------2-  b  = 2242.0
b

K
------I = 0.4842
Ko

Conclusion
The results from MSC Nastran are in a good agreement with the reference solution.

Main Index
 CHAPTER 6.51 395
Test 6: Compact Tension Specimen

6.51 Test 6: Compact Tension

Specimen

Problem Description
The following test case describe finite element analysis of compact tension specimen subjected to a point
load. Normalized stress intensity factor for the test case is calculated using virtual crack extension method.
The results are compared to the analytical solution [1].

Solution Number
SOL 400

Features Used
 Implicit Nonlinear

Reference
1. Test case 6 from NAFEMS Publication, Ref: R0020, 2D Test Cases in Linear Elastic Fracture
Mechanics, Jan,1992.

Modeling Techniques Used

Due to symmetry in loading and boundary conditions one half of the test geometry along crack plane is used
for the analysis. The finite element model is shown in Figure 6.45-1. The model is meshed with CQUAD8
and quarter point CTRIA6 elements. The quarter point elements are used at the crack tip. VCCT bulk data
entry is used to obtain the energy release rate.

Material Properties
The material is linear elastic with a Young's modulus E = 207000 and a Poisson's ratio = 0.3

Geometrical properties
a/W=0.5; W= 50.0; t = 1.0.

Loading
Point load of 100 unit is applied on top face as shown in Figure 6.45-1.

Main Index
 396 MSC Nastran Verification Guide

Boundary Condition
Boundary condition as shown in Figure 6.45-1.
uy = 0 on x axis
ux = 0 at point A.

Illustrations of the Model

Figure 6.51-1 Compact Tension Specimen

Model File Name and Location

Model Name Description Location

nvcct007 Compact Tension Specimen, Plane Strain tpl\verifman\nafems

Results
K P-
Reference solution [1] of the normalized stress intensity factor is ------I = 9.659 where K o = -----------
Ko tw
0.5

Virtual Crack Extension Method

The details of this method are explained in NAFEMS fracture mechanics benchmarks article [2].
Virtual Crack Extension Method:
Energy release rate (G)= 8.224
GE 0.5
Stress intensity factor K I =  --------------2- = 1367.7
 
1–

P-
K o = -----------
0.5
= 141.42
tw

Main Index
 CHAPTER 6.51 397
Test 6: Compact Tension Specimen

K
------I = 9.671
Ko

Conclusion
The results from MSC Nastran are in a good agreement with the reference solution.

Main Index
 398 MSC Nastran Verification Guide

6.52 Test 7: T-Joint Weld

Attachment

Problem Description
The following test case describe finite element analysis of T joint subjected to uniform stress at the top edge.
Normalized stress intensity factor for the test case is calculated using virtual crack extension method. The
results are compared to the analytical solution [1].

Solution Number
SOL 400

Features Used
 Implicit Nonlinear

Reference
1. Test 7.1 from NAFEMS Publication, Ref: R0020, 2D Test Cases in Linear Elastic Fracture
Mechanics, Jan,1992.

Modeling Techniques Used

The finite element model is shown in Figure 6.45-1. The model is meshed with CQUAD8 and quarter point
CTRIA6 elements. The quarter point elements are used at the crack tip. VCCT bulk data entry is used to
obtain the energy release rate.

Material Properties
The material is linear elastic with a Young's modulus, E = 207000 and Poisson's ratio, = 0.3

Geometrical properties
a/T=0.1; W/T= 0.5; t/T=1.0; L/T = 12.0; =45°; T=50.

Loading
Point load of 100 unit is applied on top face as shown in Figure 6.45-1.

Boundary Condition
Boundary condition as shown in Figure 6.45-1.

Main Index
 CHAPTER 6.52 399
Test 7: T-Joint Weld Attachment

uy = 0 on base
ux = 0 on base and top.

Illustrations of the Model

Figure 6.52-1 Compact Tension Specimen

Main Index
 400 MSC Nastran Verification Guide

Model File Name and Location

Model Name Description Location

nvcct008 Weld Attachment, T-Joint, Plane Strain tpl\verifman\nafems

Results
K 0.5
Reference solution [1] of the normalized stress intensity factor is ------I = 1.317 where K o =   b 
Ko

Virtual Crack Extension Method

The details of this method are explained in NAFEMS fracture mechanics benchmarks article [1].
Virtual Crack Extension Method:
Energy release rate (G)= 1.233
GE 0.5
Stress intensity factor K I =  --------------2- = 529.60
 
1–

0.5
K o =   b  = 396.3

K
------I = 1.335
Ko

Conclusion
The results from MSC Nastran are in a good agreement with the reference solution.

Main Index
 CHAPTER 6.53 401
Test 8: V-Notch Specimen in Tension

6.53 Test 8: V-Notch Specimen

in Tension

Problem Description
The following test case describe finite element analysis of V-Notch cracked plate subjected to uniform stress
at the top edge. Normalized stress intensity factor for the test case is calculated using virtual crack extension
method. The results are compared to the analytical solution [1].

Solution Number
SOL 400

Features Used
 Implicit Nonlinear

Reference
1. Test case 8.1 from NAFEMS Publication, Ref: R0020, 2D Test Cases in Linear Elastic Fracture
Mechanics, Jan,1992.

Modeling Techniques Used

Due to symmetry in loading and boundary conditions, one-quarter of test geometry along the plate axis is
used for the analysis. The finite element model is shown in Figure 6.45-1. The model is meshed with
CQUAD8 and quarter point CTRIA6 elements. The quarter point elements are used at the crack tip. VCCT
bulk data entry is used to obtain the energy release rate.

Material Properties
The material is linear elastic with a Young's modulus, E = 207000 and Poisson's ratio, = 0.3

Geometrical properties
a/d=0.2; d/W= 0.1; h/W=1.0; =45°; W=250.

Loading
Uniform stress of 100 unit is applied on top face as shown in Figure 6.45-1.

Main Index
 402 MSC Nastran Verification Guide

Boundary Condition
Boundary condition as shown in Figure 6.45-1.
uy = 0 on X axis.
ux = 0 on Y axis.

Illustrations of the Model

Figure 6.53-1 V-Notch Cracked Plate

Model File Name and Location

Model Name Description Location

nvcct009 V-Notch Cracked Plate (Plane Strain) tpl\verifman\nafems

Results
K 0.5
Reference solution [1] of the normalized stress intensity factor is ------I = 2.740 where K o =   a 
Ko

Virtual Crack Extension Method

The details of this method are explained in NAFEMS fracture mechanics benchmarks article [1].

Main Index
 CHAPTER 6.53 403
Test 8: V-Notch Specimen in Tension

Virtual Crack Extension Method:

Energy release rate (G)= 5.461
GE 0.5
Stress intensity factor K I =  --------------2- = 1114.6
 
1–

0.5
K o =   a  = 396.3

K
------I = 2.81
Ko

Conclusion
The results from MSC Nastran are in a good agreement with the reference solution.

Main Index
 404 MSC Nastran Verification Guide

Creep

Main Index
 CHAPTER 6.54 405
Test 1A: 2-D Plane Stress - Uniaxial Load, Secondary Creep

Test 1A: 2-D Plane Stress -

6.54 Uniaxial Load, Secondary
Creep

Problem Description
This NAFEMS creep test involves secondary creep of a two dimensional thin plate subjected to uniaxial load.

Solution Number
SOL 400

Features Used
 Secondary Creep Analysis with Plane Stress Elements

Reference
1. Test 1A from NAFEMS Publication Ref: R0027, NAFEMS Fundamental Tests of Creep Behavior,
June 1993.

Modeling Techniques Used

A thin square plate subjected to secondary creep under uniaxial tensile load is modeled using CQUAD8
plane stress elements. The model details are given below:

Material Properties
Young’s Modulus E = 200 × 103 N/mm2 Poisson's ratio = 0.3
Creep Law:  = A n t A = 3.125 × 10-14 per hour ( in N/mm2) n=5

Loading
Prescribed tensile stress 1 = 200 N/mm2 on line BC.

Boundary Condition
Ux = 0.0 at line AD
Uy = 0.0 on midpoint of line AD.

Main Index
 406 MSC Nastran Verification Guide

Illustrations of the Model

Figure 6.54-1 Plane Stress Model under Uniaxial Load and Secondary Creep

Main Index
 CHAPTER 6.54 407
Test 1A: 2-D Plane Stress - Uniaxial Load, Secondary Creep

Model File Name and Location

Model Name Description Location

nctest1a.dat Test1A: Plane stress - Uniaxial Load and Secondary tpl\verifman
Creep

Results

Figure 6.54-2 Creep Strain versus Time Plot

Conclusion
C C
The reference solutions from NAFEMS are defined as and  YY =  ZZ =  – 0.005 t .

C C
 X X =  e ff = 0.01t

Main Index
 408 MSC Nastran Verification Guide

The X and Y components of creep strain values from MSC Nastran are compared to these reference results
from NAFEMS in Figure 6.54-2 and also in the following table:

C
 XX C
 YY
Time NAFEMS MSC Nastran % Error NAFEMS MSC Nastran % Error
0.1 0.001 0.001 0.0% -0.0005 -0.0005 0.0%
1.0 0.01 0.01 0.0% -0.005 -0.005 0.0%
5.0 0.05 0.05 0.0% -0.025 -0.025 0.0%
10.0 0.1 0.1 0.0% -0.05 -0.05 0.0%
50.0 0.5 0.5 0.0% -0.25 -0.25 0.0%
100.0 1.0 1.0 0.0% -0.50 -0.50 0.0%
500.0 5.0 5.0 0.0% -2.5 -2.5 0.0%
1000.0 10.0 10.0 0.0% -5.0 -5.0 0.0%

Main Index
 CHAPTER 6.55 409
Test 1B: 2-D Plane Stress - Uniaxial Displacement, Secondary Creep

Test 1B: 2-D Plane Stress -

6.55 Uniaxial Displacement,
Secondary Creep

Problem Description
This NAFEMS creep test involves secondary creep of a two dimensional thin plate subjected to uniaxial
displacement.

Solution Number
SOL 400

Features Used
 Secondary Creep Analysis with Plane Stress Elements

Reference
1. Test 1B from NAFEMS Publication Ref: R0027, NAFEMS Fundamental Tests of Creep Behavior,
June 1993.

Modeling Techniques Used

A thin square plate subjected to secondary creep under uniaxial displacement load is modeled using
CQUAD8 plane stress elements. The model details are given below:

Material Properties
Young’s Modulus E = 200 × 103 N/mm2 Poisson's ratio = 0.3
Creep Law:  = A n t A = 3.125 × 10-14 per hour ( in N/mm2) n=5

Loading
None.

Boundary Condition
Ux = 0.0 at line AD

Main Index
 410 MSC Nastran Verification Guide

Uy = 0.0 on midpoint of line AD

Uniaxial displacement u1 = 0.1 mm on line BC

Illustrations of the Model

Figure 6.55-1 Plane Stress Model under Uniaxial Displacement and Secondary Creep

Main Index
 CHAPTER 6.55 411
Test 1B: 2-D Plane Stress - Uniaxial Displacement, Secondary Creep

Model File Name and Location

Model Name Description Location

nctest1b.dat Test1B: Plane stress - Uniaxial Displacement and tpl\verifman
Secondary Creep

Results

Figure 6.55-2 Stress versus Time Plot

Conclusion
The X component of stress values from MSC Nastran are compared to the reference results from NAFEMS
in Figure 6.54-2 and also in the following table.

SIGMA - XX (Time Hardening)

Time NAFEMS MSC Nastran % Error
0.0 200.00 200.00 0.00%
0.1 139.71 133.70 -4.30%
1.0 79.04 79.00 -0.05%
5.0 53.12 53.07 -0.09%
10.0 44.69 44.65 -0.09%
50.0 29.91 29.86 -0.16%

Main Index
 412 MSC Nastran Verification Guide

SIGMA - XX (Time Hardening)

Time NAFEMS MSC Nastran % Error
100.0 25.15 25.11 -0.15%
500.0 16.82 16.78 -0.24%
1000.0 14.14 14.11 -0.21%

Main Index
 CHAPTER 6.56 413
Test 2A: 2-D Plane Stress - Biaxial Load, Secondary Creep

Test 2A: 2-D Plane Stress -

6.56 Biaxial Load, Secondary
Creep

Problem Description
This NAFEMS creep test involves secondary creep of a two dimensional thin plate subjected to biaxial load.

Solution Number
SOL 400

Features Used
 Secondary Creep Analysis with Plane Stress Elements

Reference
1. Test 2A from NAFEMS Publication Ref: R0027, NAFEMS Fundamental Tests of Creep Behavior,
June 1993.

Modeling Techniques Used

A thin square plate subjected to secondary creep under biaxial tensile load is modeled using CQUAD8 plane
stress elements. The model details are given below:

Material Properties
Young’s Modulus E = 200 × 103 N/mm2 Poisson's ratio = 0.3
Creep Law:  = A n t A = 3.125 × 10-14 per hour ( in N/mm2) n=5

Loading
Prescribed tensile stress 1 = 200 N/mm2 on line BC and 2 = 200 N/mm2 on line CD

Boundary Condition
Ux = 0.0 at line AD
Uy = 0.0 on midpoint of line AB

Main Index
 414 MSC Nastran Verification Guide

Illustrations of the Model

Figure 6.56-1 Plane Stress Model under Biaxial Load and Secondary Creep

Model File Name and Location

Model Name Description Location

nctest2a.dat Test2A: Plane stress - Biaxial Load and Secondary Creep tpl\verifman

Results

Figure 6.56-2 Creep Strain versus Time Plot

Main Index
 CHAPTER 6.56 415
Test 2A: 2-D Plane Stress - Biaxial Load, Secondary Creep

Conclusion
C C C
The reference solutions from NAFEMS are defined as  X X =  e ff = 0.005t and  e ff = 0.01t .
The X and Y components of creep strain values from MSC Nastran are compared to these reference results
from NAFEMS in Figure 6.54-2 and also in the following table:

C
 XX C
 e ff
Time NAFEMS MSC Nastran % Error NAFEMS MSC Nastran % Error
0.1 0.0005 0.0005 0.0% 0.001 0.001 0.0%
1.0 0.005 0.005 0.0% 0.01 0.01 0.0%
5.0 0.025 0.025 0.0% 0.05 0.05 0.0%
10.0 0.05 0.05 0.0% 0.10 0.10 0.0%
50.0 0.25 0.25 0.0% 0.50 0.50 0.0%
100.0 0.50 0.50 0.0% 1.00 1.00 0.0%
500.0 2.50 2.50 0.0% 5.00 5.00 0.0%
1000.0 5.00 5.00 0.0% 10.00 10.00 0.0%

Main Index
 416 MSC Nastran Verification Guide

Test 2B: 2-D Plane Stress -

6.57 Biaxial Displacement,
Secondary Creep

Problem Description
This NAFEMS creep test involves secondary creep of a two dimensional thin plate subjected to biaxial
displacement.

Solution Number
SOL 400

Features Used
 Secondary Creep Analysis with Plane Stress Elements

Reference
1. Test 2B from NAFEMS Publication Ref: R0027, NAFEMS Fundamental Tests of Creep Behavior,
June 1993.

Modeling Techniques Used

A thin square plate subjected to secondary creep under biaxial displacement load is modeled using CQUAD8
plane stress elements. The model details are given below:

Material Properties
Young’s Modulus E = 200 × 103 N/mm2 Poisson's ratio = 0.3
Creep Law:  = A n t A = 3.125 × 10-14 per hour ( in N/mm2) n=5

Loading
None.

Boundary Condition
Ux = 0.0 at line AD
Uy = 0.0 on midpoint of line AD

Main Index
 CHAPTER 6.57 417
Test 2B: 2-D Plane Stress - Biaxial Displacement, Secondary Creep

Biaxial displacement u1 = 0.1 mm on line BC and u2 = 0.1 mm on line CD

Illustrations of the Model

Figure 6.57-1 Plane Stress Model under Biaxial Displacement and Secondary Creep

Main Index
 418 MSC Nastran Verification Guide

Model File Name and Location

Model Name Description Location

nctest2b.dat Test2B: Plane stress - Biaxial Displacement and tpl\verifman
Secondary Creep

Results

Figure 6.57-2 Stress versus Time Plot

Conclusion
The X component of stress values from MSC Nastran are compared to the reference results from NAFEMS
in Figure 6.54-2 and also in the following table.

SIGMA - XX (Time Hardening)

Time NAFEMS MSC Nastran % Error
0.0 285.71 285.71 0.00%
0.1 148.35 150.72 1.60%
1.0 84.77 86.28 1.78%
5.0 56.76 57.78 1.80%
10.0 47.79 48.59 1.69%
50.0 31.96 32.49 1.66%

Main Index
 CHAPTER 6.57 419
Test 2B: 2-D Plane Stress - Biaxial Displacement, Secondary Creep

SIGMA - XX (Time Hardening)

Time NAFEMS MSC Nastran % Error
100.0 26.86 27.31 1.67%
500.0 17.99 18.25 1.44%
1000.0 15.07 15.35 1.86%

Main Index
 420 MSC Nastran Verification Guide

Test 3B: 2-D Plane Stress -

6.58 Biaxial (-ve) Displacement,
Secondary Creep

Problem Description
This NAFEMS creep test involves secondary creep of a two dimensional thin plate subjected to biaxial
displacement.

Solution Number
SOL 400

Features Used
 Secondary Creep Analysis with Plane Stress Elements

Reference
1. Test 3B from NAFEMS Publication Ref: R0027, NAFEMS Fundamental Tests of Creep Behavior,
June 1993.

Modeling Techniques Used

A thin square plate subjected to secondary creep under biaxial displacement load is modeled using CQUAD8
plane stress elements. The model details are given below:

Material Properties
Young’s Modulus E = 200 × 103 N/mm2 Poisson's ratio = 0.3
Creep Law:  = A n t A = 3.125 × 10-14 per hour ( in N/mm2) n=5

Loading
None.

Boundary Condition
Ux = 0.0 at line AD
Uy = 0.0 on midpoint of line AB

Main Index
 CHAPTER 6.58 421
Test 3B: 2-D Plane Stress - Biaxial (-ve) Displacement, Secondary Creep

Biaxial displacement u1 = 0.1 mm on line BC and u2 = -0.1 mm on line CD

Illustrations of the Model

Figure 6.58-1 Plane Stress Model under Biaxial (Negative) Displacement and Secondary Creep

Main Index
 422 MSC Nastran Verification Guide

Model File Name and Location

Model Name Description Location

nctest3b.dat Test3B: Plane stress - Biaxial (Negative) tpl\verifman
Displacement and Secondary Creep

Results

Figure 6.58-2 Stress versus Time Plot

Conclusion
The X component of stress values from MSC Nastran are compared to the reference results from NAFEMS
in Figure 6.54-2 and also in the following table.

 XX

Time NAFEMS MSC Nastran % Error

0.0 153.85 153.85 0.00%
0.1 76.29 77.46 1.53%
1.0 43.42 44.20 1.80%
5.0 29.11 29.59 1.65%
10.0 24.45 24.88 1.76%
50.0 16.33 16.64 1.90%

Main Index
 CHAPTER 6.58 423
Test 3B: 2-D Plane Stress - Biaxial (-ve) Displacement, Secondary Creep

 XX

Time NAFEMS MSC Nastran % Error

100.0 13.78 13.99 1.52%
500.0 9.20 9.34 1.52%
1000.0 7.73 7.86 1.68%

Main Index
 424 MSC Nastran Verification Guide

Test 8A: 2D Plane Stress -

6.59 Uniaxial Load, Primary
Creep

Problem Description
This NAFEMS creep test involves primary creep of a two dimensional thin plate subjected to uniaxial load.

Solution Number
SOL 400

Features Used
 Primary Creep Analysis with Plane Stress Elements

Reference
1. Test 8A from NAFEMS Publication Ref: R0027, NAFEMS Fundamental Tests of Creep Behavior,
June 1993.

Modeling Techniques Used

A thin square plate subjected to primary creep under uniaxial tensile load is modeled using CQUAD8 plane
stress elements. The model details are given below:

Material Properties
Young’s Modulus E = 200 × 103 N/mm2 Poisson's ratio = 0.3
Creep Law:  = A n tm A = 3.125 × 10-14 per hour ( in N/mm2) n=5 m = 0.5

Loading
Prescribed tensile stress 1 = 200 N/mm2 on line BC.

Boundary Condition
Ux = 0.0 at line AD
Uy = 0.0 on midpoint of line AD.

Main Index
 CHAPTER 6.59 425
Test 8A: 2D Plane Stress - Uniaxial Load, Primary Creep

Illustrations of the Model

Figure 6.59-1 Plane Stress Model under Uniaxial Load and Primary Creep

Model File Name and Location

Model Name Description Location

lefm008a.dat Weld Attachment, T-Joint, Plane Strain tpl\verifman

Results

Figure 6.59-2 Creep Strain versus Time Plot

Main Index
 426 MSC Nastran Verification Guide

Conclusion
C C 0.5
The reference solutions from NAFEMS are defined as  X X =  e ff = 0.01t and
C C 0.5
 YY =  ZZ =  – 0.005 t .
The X & Y components of creep strain values from MSC Nastran are compared to these reference results
from NAFEMS in Figure 6.54-2 and also in the following table:

C
 XX C
 YY
Time NAFEMS MSC Nastran % Error NAFEMS MSC Nastran % Error
0.1 0.00316 0.003278 3.74% -0.00158 -0.001639 3.74%
1.0 0.01 0.010141 1.41% -0.005 -0.005071 1.42%
5.0 0.0224 0.022530 0.58% -0.0112 -0.011265 0.58%
10.0 0.0316 0.031806 0.65% -0.0158 -0.015903 0.65%
50.0 0.0707 0.070914 0.30% -0.0354 -0.035458 0.16%
100.0 0.1 0.100224 0.22% -0.05 -0.050112 0.22%
500.0 0.2236 0.223871 0.12% -0.112 -0.111935 -0.06%
1000.0 0.3162 0.316511 0.10% -0.158 -0.158251 0.16%

Main Index
 CHAPTER 6.60 427
Test 8B: 2-D Plane Stress - Uniaxial Displacement, Primary Creep

Test 8B: 2-D Plane Stress -

6.60 Uniaxial Displacement,
Primary Creep

Problem Description
This NAFEMS creep test involve primary creep of a two dimensional thin plate subjected to uniaxial
displacement.

Solution Number
SOL 400

Features Used
 Primary Creep Analysis with Plane Stress Elements

Reference
1. Test 8B from NAFEMS Publication Ref: R0027, NAFEMS Fundamental Tests of Creep Behavior,
June 1993.

Modeling Techniques Used

A thin square plate subjected to primary creep under uniaxial displacement load is modeled using CQUAD8
plane stress elements. The model details are given below:

Material Properties
Young’s Modulus E = 200 × 103 N/mm2 Poisson's ratio = 0.3
Creep Law:  = A n tm A = 3.125 × 10-14 per hour ( in N/mm2) n=5 m = 0.5

Loading
None.

Boundary Condition
Ux = 0.0 at line AD and Uy = 0.0 on midpoint of line AD

Main Index
 428 MSC Nastran Verification Guide

Uniaxial displacement u1 = 0.1 mm on line BC.

Illustrations of the Model

Figure 6.60-1 Plane Stress Model under Uniaxial Displacement and Primary Creep

Model File Name and Location

Model Name Description Location

nctest8b.dat Test8B: Plane Stress - Uniaxial Displacement and Primary tpl\verifman
Creep

Results

Figure 6.60-2 Creep Strain versus Time Plot

Main Index
 CHAPTER 6.60 429
Test 8B: 2-D Plane Stress - Uniaxial Displacement, Primary Creep

Conclusion
The X component of stress values from MSC Nastran are compared to the reference results from NAFEMS
in Figure 6.54-2 and also in the following table:

SIGMA - XX (Time Hardening)

Time NAFEMS MSC Nastran % Error
0.0 200 200.000000 0.00%
0.1 103.74 103.592590 -0.14%
1.0 78.9 78.619415 -0.36%
5.0 64.76 64.521416 -0.37%
10.0 59.45 59.268791 -0.30%
50.0 48.65 48.407439 -0.50%
100.0 44.6 44.357185 -0.54%
500.0 36.48 36.316412 -0.45%
1000.0 33.69 33.351093 -1.01%

Main Index
 430 MSC Nastran Verification Guide

Test 8C: 2D Plane Stress -

6.61 Stepped Load Primary
Creep

Problem Description
This NAFEMS creep test involve primary creep of a two dimensional thin plate subjected to stepped load.

Solution Number
SOL 400

Features Used
 Primary Creep Analysis with Plane Stress Elements

Reference
1. Test 8C from NAFEMS Publication Ref: R0027, NAFEMS Fundamental Tests of Creep Behavior,
June 1993.

Modeling Techniques Used

A thin square plate subjected to primary creep under stepped load is modeled using CQUAD8 plane stress
elements. The model details are given below:

Material Properties
Young’s Modulus E = 200 × 103 N/mm2 Poisson's ratio = 0.3
Creep Law:  = A n tm A = 3.125 × 10-14 per hour ( in N/mm2) n=5 m = 0.5

Loading
Prescribed tensile stress 1 on line BC

1 = 200 N/mm2 for t = 0 to 100 hoursand 1 = 250 N/mm2 for t > 100 hours

Boundary Condition
Ux = 0.0 at line AD

Main Index
 CHAPTER 6.61 431
Test 8C: 2D Plane Stress - Stepped Load Primary Creep

Uy = 0.0 on midpoint of line AD

Illustrations of the Model

Figure 6.61-1 Plane Stress Model under Stepped Load and Primary Creep

Model File Name and Location

Model Name Description Location

nctest8c.dat Test8C: Plane Stress - Stepped Load and Primary Creep tpl\verifman

Main Index
 432 MSC Nastran Verification Guide

Results

Figure 6.61-2 Creep Strain - XX versus Time Plot

Conclusion
The X component of creep strain values from MSC Nastran are compared to the reference results from
NAFEMS in Figure 6.54-2 and also in the following table:

Creep Strain - XX (Time Hardening)

Time NAFEMS MSC Nastran % Error
1.0 0.01 0.010141 1.41%
10.0 0.03162 0.031809 0.60%
50.0 0.07071 0.070935 0.32%
100.0 0.1 0.100238 0.24%
110.0 0.1149 0.115134 0.20%
120.0 0.1291 0.129365 0.21%
150.0 0.1686 0.168827 0.13%
200.0 0.2264 0.226660 0.11%

Main Index
 CHAPTER 6.62 433
Test 9A: 2-D Plane Stress - Biaxial Load, Primary Creep

6.62 Test 9A: 2-D Plane Stress -

Biaxial Load, Primary Creep

Problem Description
This NAFEMS creep test involve primary creep of a two dimensional thin plate subjected to biaxial load.

Solution Number
SOL 400

Features Used
 Primary Creep Analysis with Plane Stress Elements

Reference
1. Test 9A from NAFEMS Publication Ref: R0027, NAFEMS Fundamental Tests of Creep Behavior,
June 1993.

Modeling Techniques Used

A thin square plate subjected to primary creep under biaxial tensile load is modeled using CQUAD8 plane
stress elements. The model details are given below:

Material Properties
Young’s Modulus E = 200 × 103 N/mm2 Poisson's ratio = 0.3
Creep Law:  = A n tm A = 3.125 × 10-14 per hour ( in N/mm2) n=5 m = 0.5

Loading
Prescribed tensile stress 1 = 200 N/mm2 on line BC and 2 = 200 N/mm2 on line CD.

Boundary Condition
Ux = 0.0 at line AD
Uy = 0.0 at line AB

Main Index
 434 MSC Nastran Verification Guide

Illustrations of the Model

Figure 6.62-1 Plane Stress Model under Stepped Load and Primary Creep

Model File Name and Location

Model Name Description Location

nctest9a.dat Test9A: Plane Stress - Biaxial Load and Primary Creep tpl\verifman

Main Index
 CHAPTER 6.62 435
Test 9A: 2-D Plane Stress - Biaxial Load, Primary Creep

Results

Figure 6.62-2 Creep Strain - XX versus Time Plot

Conclusion
C C 0.5 C 0.5
The reference solutions from NAFEMS are defined as  X X =  YY = 0.005t and  e ff = 0.01t .
The X component and effective creep strain values from MSC Nastran are compared to these reference results
from NAFEMS in Figure 6.54-2 and also in the following table:

C
 XX C
 e ff
Time NAFEMS MSC Nastran % Error NAFEMS MSC Nastran % Error
0.1 0.00158 0.001642 3.93% 0.00316 0.003280 3.80%
1.0 0.005 0.005076 1.53% 0.01 0.010200 2.00%
5.0 0.01112 0.011273 1.38% 0.0224 0.022500 0.45%
10.0 0.0158 0.015912 0.71% 0.0316 0.031800 0.63%
50.0 0.0354 0.035467 0.19% 0.0707 0.071325 0.88%
100.0 0.05 0.050121 0.24% 0.1 0.100273 0.27%
500.0 0.1118 0.111942 0.13% 0.2236 0.223990 0.17%
1000.0 0.1581 0.158268 0.11% 0.3162 0.317000 0.25%

Main Index
 436 MSC Nastran Verification Guide

Test 9B: 2-D Plane Stress

6.63 - Biaxial Displacement,
Primary Creep

Problem Description
This NAFEMS creep test involve primary creep of a two dimensional thin plate subjected to biaxial
displacement.

Solution Number
SOL 400

Features Used
 Primary Creep Analysis with Plane Stress Elements

Reference
1. Test 9B from NAFEMS Publication Ref: R0027, NAFEMS Fundamental Tests of Creep Behavior,
June 1993.

Modeling Techniques Used

A thin square plate subjected to primary creep under biaxial displacement load is modeled using CQUAD8
plane stress elements. The model details are given below:

Material Properties
Young’s Modulus E = 200 × 103 N/mm2 Poisson's ratio = 0.3
Creep Law:  = A n tm A = 3.125 × 10-14 per hour ( in N/mm2) n=5 m = 0.5

Loading
None.

Boundary Condition
Ux = 0.0 at line AD and Uy = 0.0 at line AB
Biaxial displacement: u1 = 0.1 mm on line BC and u2 = 0.1 mm on line CD.

Main Index
 CHAPTER 6.63 437
Test 9B: 2-D Plane Stress - Biaxial Displacement, Primary Creep

Illustrations of the Model

Figure 6.63-1 Plane Stress Model under Biaxial Displacement and Primary Creep

Model File Name and Location

Model Name Description Location

nctest9b.dat Test9B: Plane Stress - Biaxial Displacement and Primary Creep tpl\verifman

Main Index
 438 MSC Nastran Verification Guide

Results

Figure 6.63-2 Stress vs Time Plot

Conclusion
The X component of stress values from MSC Nastran are compared to the reference results from NAFEMS
in Figure 6.63-2 and also in the following table:

SIGMA - XX (Time Hardening)

Time NAFEMS MSC Nastran % Error
0.0 285.71 285.714290 0.00%
0.1 112.52 114.182870 1.48%
1.0 84.78 85.921959 1.35%
5.0 69.37 70.340996 1.40%
10.0 63.59 64.572617 1.55%
50.0 52.02 52.715802 1.34%
100.0 47.74 48.303667 1.18%
500.0 39.04 39.523781 1.24%
1000.0 35.71 36.291130 1.63%

Main Index
 CHAPTER 6.64 439
Test 9C: 2-D Plane Stress - Biaxial Stepped Load, Primary Creep

Test 9C: 2-D Plane Stress -

6.64 Biaxial Stepped Load,
Primary Creep

Problem Description
This NAFEMS creep test involve primary creep of a two dimensional thin plate subjected to biaxial stepped
load.

Solution Number
SOL 400

Features Used
 Primary Creep Analysis with Plane Stress Elements

Reference
1. Test 9C from NAFEMS Publication Ref: R0027, NAFEMS Fundamental Tests of Creep Behavior,
June 1993.

Modeling Techniques Used

A thin square plate subjected to primary creep under biaxial stepped load is modeled using CQUAD8 plane
stress elements. The model details are given below:

Material Properties
Young’s Modulus E = 200 × 103 N/mm2 Poisson's ratio = 0.3
Creep Law:  = A n tm A = 3.125 × 10-14 per hour ( in N/mm2) n=5 m = 0.5

Loading
Prescribed tensile stress 1 on line BC and 2 on line CD.

1 = 1 = 200 N/mm2 for t = 0 to 100 hours, and

1 = 1 = 250 N/mm2 for t > 100 hours.

Main Index
 440 MSC Nastran Verification Guide

Boundary Condition
Ux = 0.0 at line AD and Uy = 0.0 at line AB

Illustrations of the Model

Figure 6.64-1 Plane Stress Model under Biaxial Stepped Load and Primary Creep

Model File Name and Location

Model Name Description Location

nctest9c.dat Test9C: Plane Stress - Biaxial Stepped Load and Primary Creep tpl\verifman

Main Index
 CHAPTER 6.64 441
Test 9C: 2-D Plane Stress - Biaxial Stepped Load, Primary Creep

Results

Figure 6.64-2 Creep Strain - XX versus Time Plot

Conclusion
The X component of creep strain values from MSC Nastran are compared to the reference results from
NAFEMS in Figure 6.63-2 and also in the following table:

Creep Strain - XX (Time Hardening)

Time NAFEMS MSC Nastran % Error
1.0 0.005 0.005077 1.53%
10.0 0.0158 0.015912 0.71%
50.0 0.0354 0.035473 0.21%
100.0 0.05 0.050126 0.25%
110.0 0.0574 0.057577 0.31%
120.0 0.0646 0.064692 0.14%
150.0 0.0843 0.084424 0.15%
200.0 0.1132 0.113340 0.12%

Main Index
 442 MSC Nastran Verification Guide

Composite

Main Index
 CHAPTER 6.65 443
R0031(1): Laminated Strip under Three-point Bending

6.65 R0031(1): Laminated Strip

under Three-point Bending

Problem Description
This NAFEMS composite benchmark is a simply supported 7 layer symmetric strip with a central line load.
A quarter model, with appropriate boundary conditions, is used and predictions for stress and displacement
are obtained using composite shell elements of Nastran.

Solution Number
Sol 101

Features Used
 Linear Static Analysis
 Layered Composite Elements

Reference
1. Test R0031/1 from NAFEMS publication R0031, Composites Benchmarks, February 1995.

Modeling Techniques Used

One quarter of 7 layer laminated strip is modeled using CQUAD4 composite shell elements of Nastran. The
model details are given below:

Material Properties
E1 = 100GPa 12 = 0.4 G12 = 3GPa
E2 = 5GPa 13 = 0.3 G13 = 2GPa
E3 = 5GPa 23 = 0.3 G23 = 2GPa

Loading
Line load of 10 N/mm at point C.

Boundary Condition
Uz = 0.0 at point A

Main Index
 444 MSC Nastran Verification Guide

Ux = y = z = 0.0 at x = 25 and Uy = x = z = 0.0 at y = 5

Illustrations of the Model

Figure 6.65-1 Laminated Strip in a Three-point Bending

Model File Name and Location

Model Name Description Location

nafem_r00311.dat R0031(1): Laminated Strip under Three-point Bending \verifman\nafems

Results

MSC Nastran
NAFEMS Values % Error
11 at E (MPa) 683.9 682.7 -0.18%
13 at D (MPa) -4.1 -4.0908 -0.22%
uz at E (mm) -1.06 -1.0625 0.24%

Conclusion
The results from MSC Nastran are compared with the NAFEMS results in the above table. The value of 11
at point E are obtained by extrapolating stress values at element centroid of two elements adjacent to point
E. More variants of this model with other composite element types are demonstrated with SOL400 in
Chapter 7: Laminated Strip under Three-point Bending of MSC Nastran Implicit Nonlinear Demonstration Guide.

Main Index
 CHAPTER 6.66 445
R0031(2): Wrapped Thick Cylinder under Pressure and Thermal Loading

R0031(2): Wrapped Thick

6.66 Cylinder under Pressure
and Thermal Loading

Problem Description
This NAFEMS composite benchmark is a long thick cylinder made from isotropic material onto which
external hoop windings of orthotropic material have been added. This problem demonstrates the ability to
model pressure and thermal loading for composite laminated material.

Solution Number
Sol 101

Features Used
 Linear Static Analysis
 Laminated Composites

Reference
1. Test R0031/2 from NAFEMS publication R0031, Composites Benchmarks, February 1995.

Modeling Techniques Used

Due to symmetry, one eighth of the cylinder is modeled with composite CQUAD4 shell elements. The
cylinder consists of two layers with layer thickness and orientation as shown in Figure 6.66-1. The model
details are given below:

Material Properties

E = 210GPa  = 0.3  = 2.0 × 10-5/°C for inner cylinder and

E1 = 130GPa 12 = 0.25 G12 = 10GPa  = 3.0 × 10-6/°C

Main Index
 446 MSC Nastran Verification Guide

E2 = 5GPa 13 = 0.25 G13 = 10GPa  = 2.0 × 10-5/°C

E3 = 5GPa 23 = 0.0 G23 = 5GPa  = 2.0 × 10-5/°C for outer cylinder

Loading
Case-1: Internal Pressure of 200 MPa
Case-2: Internal Pressure of 200 MPa and temperature rise of 130°C. The temperature loading simulates the
pre-stress due to tension in the hoop windings.

Boundary Condition
Axial displacement Uz = 0 at z = 0.

Illustrations of the Model

Figure 6.66-1 Wrapped Thick Cylinder under Pressure and Thermal Loading

Main Index
 CHAPTER 6.66 447
R0031(2): Wrapped Thick Cylinder under Pressure and Thermal Loading

Model File Name and Location

Model Name Description Location

nafem_r00312.dat R0031(2): Wrapped Thick Cylinder under Pressure and \verifman\nafems
Thermal Loading

Results

MSC Nastran
Hoop Stress
(MPa) NAFEMS Values % Error
Case 1 at radius = 24 1483 1414.2 -4.64%
at radius = 26 822 874.8 6.42%
Case 2 at radius = 24 1309 1235.9 -5.58%
at radius = 26 994 1052.9 5.93%

Conclusion
The hoop stress values in the inner and outer cylinders for the two load cases at r = 24 and r = 26 from MSC
Nastran are compared with the NAFEMS results in the above table. SOL400 variant of this model is
demonstrated in Chapter 8: Wrapped Thick Cylinder under Pressure and Thermal Loading of MSC Nastran Implicit
Nonlinear Demonstration Guide.

Main Index
 448 MSC Nastran Verification Guide

R0031 (3): Three-layer

6.67 Sandwich Shell under
Normal Pressure Loading

Problem Description
This NAFEMS composite benchmark is a simply supported square sandwich plate subjected to a normal
pressure loading. The orthotropic materials of the sandwich plate are defined in such a way that the facing
sheets carry the bending and the shear is carried by the core.

Solution Number
Sol 101

Features Used
 Linear Static Analysis
 Laminated Composites

Reference
1. Test R0031/3 from NAFEMS publication R0031, Composites Benchmarks, February 1995.

Modeling Techniques Used

One quarter of the square sandwich plate is modeled with composite CQUAD4 shell elements. It consists of
two outer facing sheets and a thick central core as shown in Figure 6.67-1. The model details are given below:

Material Properties
E1 = 10×105 psi E2 = 4×106 psi 12 = 0.3 and
G12 = 1.875×106 psi G13 = 1.875×106 psi G23 = 1.875×106 psi for face sheets
E1 = 10 psi E2 = 10 psi 12 = 0.0 and
G12 = 10 psi G13 = 3 ×104 psi G23 = 1.2×104 psi for core

Loading
Plate is subjected to uniform pressure of 100 psi.

Main Index
 CHAPTER 6.67 449
R0031 (3): Three-layer Sandwich Shell under Normal Pressure Loading

Boundary Condition
The plate is simply supported at all four edges and fixed at four corners. Symmetric boundary conditions are
applied at the symmetric planes of one quarter model.

Illustrations of the Model

Figure 6.67-1 Wrapped Thick Cylinder under Pressure and Thermal Loading

Model File Name and Location

Model Name Description Location

nafem_r00313.dat R0031(3): Three-layer Sandwich Shell under Normal \verifman\nafems
Pressure Loading

Results
MSC Nastran
NAFEMS Values % Error
Uz at C (in) -0.123 -0.1226 -0.33%
11 at C (kpsi) 34.45 34.03 -1.22%
22 at C (kpsi) 13.93 13.29 -4.59%
12 at E (kpsi) -5.07 -5.04 -0.59%

Conclusion
The Z-displacement at point C and stress values at points C and E from MSC Nastran are compared with
the NAFEMS results in the above table. The variants of this model with other composite element types are
demonstrated with SOL400 in Chapter 9: Three-layer Sandwich Shell under Normal Pressure Loading of MSC
Nastran Implicit Nonlinear Demonstration Guide.

Main Index
 450 MSC Nastran Verification Guide

Contact

Main Index
 CHAPTER 6.68 451
R0094 (1): Benchmark 1: 2D Cylinder Roller Contact

6.68 R0094 (1): Benchmark 1:

2D Cylinder Roller Contact

Problem Description
In this NAFEMS contact benchmark, a steel cylinder is pressed into an aluminum block. The solution with
a 2-D plane strain approximation of this problem from Nastran is compared to an analytical solution for a
frictionless case.

Solution Number
SOL 400

Features Used
 Nonlinear static analysis.
 Advancing contact area with curved contact surfaces
 Deformable to deformable contact

Reference
1. Benchmark 1 from NAFEMS publication R0094, Advanced Finite Element Contact Benchmarks,
2006.

Modeling Techniques Used

Steel cylinder and aluminum block are modeled as two contact bodies with either linear or quadratic plane
strain elements. The model details are given below:

Material Properties
Ecylinder = 210 kN/mm2
 cylinder = 0.3
Eblock = 70 kN/mm2
 block = 0.3

Loading
Vertical point load F = 35kN

Main Index
 452 MSC Nastran Verification Guide

Boundary Condition
Symmetric displacement constraints along vertical symmetric line
Bottom surface of the block is fixed (Ux = Uy = 0.0)
Touching contact between cylinder and block with segment to segment contact

Illustrations of the Model

Figure 6.68-1 2D Cylinder Roller Contact

Main Index
 CHAPTER 6.68 453
R0094 (1): Benchmark 1: 2D Cylinder Roller Contact

Model File Name and Location

Model Name Description Location

nafem_r00941a.dat Linear element \verifman\nafems
nafem_r00941b.dat Quadratic element \verifman\nafems

Results

amin aavg amax

Element (mm) (mm) (mm) % Error Pmax % Error
Linear 6.698577 7.0811275 7.463678 14.03% 3335.70 -6.96%
Quadratic 5.952867 6.325722 6.698577 1.86% 3486.60 -2.75
NAFEMS 6.21 3585.37

Conclusion
The contact pressure results along the touching surfaces of steel cylinder from MSC Nastran are compared
with the NAFEMS results in the above plot. The length of contact zone and maximum contact pressure
values from MSC Nastran are compared with NAFEMS results in the above table. More details of this
contact benchmark model with and without friction are demonstrated using node to segment contact in 2-
D Cylindrical Roller Contact of MSC Nastran Implicit Nonlinear Demonstration Problems Guide.

Main Index
 454 MSC Nastran Verification Guide

R0094 (2): Benchmark 2:

6.69 3D Punch (Rounded
Edges)

Problem Description
In this NAFEMS contact benchmark, an axisymmetric steel punch is compressed on an aluminum
foundation. The solution with a 2-D axisymmetric model of this problem from Nastran is compared to
NAFEMS solution for a frictionless case.

Solution Number
SOL 400

Features Used
 Nonlinear static analysis.
 Axisymmetric contact
 Deformable to deformable contact

Reference
1. Benchmark 2 from NAFEMS publication R0094, Advanced Finite Element Contact Benchmarks,
2006.

Modeling Techniques Used

Steel punch and aluminum foundation are modeled as two contact bodies with either linear axisymmetric
CQUADX elements. The model details are given below:

Main Index
 CHAPTER 6.69 455
R0094 (2): Benchmark 2: 3D Punch (Rounded Edges)

Material Properties
Epunch = 210 kN/mm2
 punch = 0.3
Efoundation = 70 kN/mm2
 foundation = 0.3

Loading
Uniform pressure in punch P = 100 N/mm2

Boundary Condition
Bottom surface of the foundation is fixed (Ux = Uy = 0.0)
Touching contact between punch and foundation with node to segment contact

Illustrations of the Model

Figure 6.69-1 3D Punch

Main Index
 456 MSC Nastran Verification Guide

Model File Name and Location

Model Name Description Location

nafem_r00942a.dat Without friction \verifman\nafems
nafem_r00942b.dat With friction \verifman\nafems

Results

Conclusion
The radial displacement results along the touched surfaces of aluminum foundation from MSC Nastran are
compared with the NAFEMS results in the above plot for the cases with and without friction. More details
of this contact benchmark model with 3D solid elements are demonstrated in Chapter 2: 3-D Punch (Rounded
Edges) Contact of MSC Nastran Implicit Nonlinear Demonstration Guide.

Main Index
 CHAPTER 6.70 457
R0094 (3): Benchmark 3: 3D Sheet Metal Forming

6.70 R0094 (3): Benchmark 3:

3D Sheet Metal Forming

Problem Description
This NAFEMS contact benchmark is an approximation of Numisheet 2002 - Benchmark B problem.
Simulations are carried out with MSC Nastran with 2D plane strain and 3D shell models to find the angles
before and after spring back.

Solution Number
SOL 400

Features Used
 Nonlinear static analysis.
 Large sliding contact around circular surface
 Deformable to deformable contact

Reference
1. Benchmark 3 from NAFEMS publication R0094, Advanced Finite Element Contact Benchmarks,
2006.

Modeling Techniques Used

Sheet is modeled with either 2D plane strain or shell elements. The model details are given below:

Main Index
 458 MSC Nastran Verification Guide

Material Properties
E = 70.5 kN/mm2
 = 0.342
Initial yield stress 0 = 194 N/mm2
n
Hollomon hardening:  = k

with K = 550.4 N/mm2 & n =0.223

Loading
Prescribed vertical displacement of punch = -28.5 mm

Boundary Condition
Symmetric displacement restraints (half symmetry)
Rigid surface for Die is fixed
Touching contact of sheet with punch and die using node to segment contact

Illustrations of the Model

Figure 6.70-1 3D Sheet Forming

Main Index
 CHAPTER 6.70 459
R0094 (3): Benchmark 3: 3D Sheet Metal Forming

Model File Name and Location

Model Name Description Location

nafem_r00943a.dat 2D Without friction \verifman\nafems
nafem_r00943b.dat 2D With friction \verifman\nafems
nafem_r00943c.dat 3D Without friction \verifman\nafems
nafem_r00943d.dat 3D With friction \verifman\nafems

Results

Main Index
 460 MSC Nastran Verification Guide

Conclusion
The force-displacement curves of punch from MSC Nastran for 2D plane strain and 3D shell models are
compared with the NAFEMS results in the above plots for the cases with and without friction. More details
of this contact benchmark model are presented in Chapter 3: 3-D Sheet Metal Forming of MSC Nastran Implicit
Nonlinear Demonstration Guide.

Main Index
 CHAPTER 6.71 461
R0094 (4): Benchmark 4: 3D Loaded Pin

6.71 R0094 (4): Benchmark 4:

3D Loaded Pin

Problem Description
This NAFEMS contact benchmark evaluates the performance of contact algorithms at the curved
boundaries. A cylindrical pin is located in the cylindrical hole of a strip, with identical diameters of the hole
and the pin. Two equal point loads are applied to the center of the pin, resulting to a loss of contact at one
side of the pin and a localized contact area on the other side. A 2D plane strain representation of the model
is used to evaluate the contact normal stress in the contact surfaces of strip and pin.

Solution Number
SOL 400

Features Used
 Nonlinear Static Analysis.
 Receding Contact Area with Curved Contact Surfaces
 Deformable to Deformable Contact with Friction

Reference
1. Benchmark 4 from NAFEMS publication R0094, Advanced Finite Element Contact Benchmarks,
2006.

Modeling Techniques Used

Both pin and strip are modeled with either 2D plane strain quadratic elements. The model details are given
below:

Main Index
 462 MSC Nastran Verification Guide

Material Properties
Epin = 210 kN/mm2
pin = 0.3
Esheet = 70 kN/mm2
sheet = 0.3

Loading
Point load of 100kN is applied at the center of pin.

Boundary Condition
Symmetric displacement restraints (half symmetry).
Left side of sheet is fixed.
Touching contact of strip with pin (Node to Segment contact).

Illustrations of the Model

Figure 6.71-1 3D Loaded Pin

Main Index
 CHAPTER 6.71 463
R0094 (4): Benchmark 4: 3D Loaded Pin

Model File Name and Location

Model Name Description Location

nafem_r00944a.dat 2D Without friction \verifman\nafems

Results

Main Index
 464 MSC Nastran Verification Guide

Conclusion
The results from MSC Nastran for the variation of contact normal stress along the contact surfaces of pin and
strip are compared with the results from Marc, which is also used with the node to segment contact in the
NAFEMS publication for this benchmark. Segment to segment contact will give more smooth variation of
the contact normal stress in both the pin and strip surfaces. Further details of this contact benchmark model
are presented in Chapter 4: 3-D Loaded Pin with Friction of MSC Nastran Implicit Nonlinear Demonstration
Problems Guide.

Main Index
 CHAPTER 6.72 465
R0094 (5): Benchmark 5: 3D Steel Roller on Rubber

6.72 R0094 (5): Benchmark 5:

3D Steel Roller on Rubber

Problem Description
In this NAFEMS contact benchmark, a steel cylindrical roller rotates on a rubber base. It tests the ability of
contact algorithms to model the movement of two surfaces when they are in contact, as the rubber will stretch
compared to the steel cylinder.

Solution Number
SOL 400

Features Used
 Nonlinear Static Analysis.
 Rolling Contact with Friction
 Deformable and Rigid Contact Bodies

Reference
1. Benchmark 5 from NAFEMS publication R0094, Advanced Finite Element Contact Benchmarks,
2006.

Modeling Techniques Used

Both 2D plane strain and 3D solid representations are used for this contact simulation of this benchmark.
The model details are given below:

Material Properties
Esteel = 210 kN/mm2
steel = 0.3
C10, rubber = 10 kN/mm2 with Neo-Hookean material description

Loading
Move bottom surface of rubber 3 mm up during time period 0-1 second and held at this position from 1-2
second. Prescribed rotation of steel roller by 360 degrees during time period 1-2 second.

Main Index
 466 MSC Nastran Verification Guide

Boundary Condition
No rotation of roller and sheet x-displacement fixed during time period 0-1 second.
Sheet is free to move in horizontal direction during time period 1-2 second.
Touching contact of roller and sheet and rotating rigid body is glued to roller (Segment to Segment contact).

Illustrations of the Model

Figure 6-1 3D Steel Roller on Rubber

Main Index
 CHAPTER 6.72 467
R0094 (5): Benchmark 5: 3D Steel Roller on Rubber

Model File Name and Location

Model Name Description Location

nafem_r00945a.dat 2D with friction \verifman\nafems
nafem_r00945b.dat 2D with friction \verifman\nafems

Results

Main Index
 468 MSC Nastran Verification Guide

Conclusion
The results from MSC Nastran for the variation of vertical force on roller are compared with the results from
Marc, which is also used in the NAFEMS publication for this benchmark. Further details of this contact
benchmark model are presented in Chapter 5: Steel Roller on Rubber of MSC Nastran Implicit Nonlinear
Demonstration Problems Guide.

Main Index
 CHAPTER 6.72 469
R0094 (5): Benchmark 5: 3D Steel Roller on Rubber

Geometric Nonlinear

Main Index
 470 MSC Nastran Verification Guide

3DNLG1: Elastic Large

6.73 Deflection Response of A
Z-Shaped Cantilever under
End Load

Problem Description
This NAFEMS geometric nonlinear benchmark on Z-shaped cantilever under an end load is used to test
beam element's formulation for large displacement and rotations. The model is subjected to significant stress
stiffening accompanied by stress reversal.

Solution Number
SOL 400

Features Used
 Implicit nonlinear analysis.
 3D Large rotation - large displacement with membrane and bending actions.
 Tension stiffening with change in sign of bending moment at section A.

Reference
1. Test 3DNLG-1 from NAFEMS Publication 'A Review of Benchmark Problems for Geometric Non-
Linear Behaviour of 3D Beams & Shells (Summary)' by Prinja. N & Clegg. R, R0024, August 1993.

Modeling Techniques Used

The Z- shaped cantilever is modeled using CBEAM elements with different mesh densities (number of
elements varying from 30 to 150). The model details are given below:

Material Properties
E = 2.0x107  = 0.3

Loading
Total load P = 4000.0 is applied at the free end in the Z-direction in 3 different steps.

Main Index
 CHAPTER 6.73 471
3DNLG1: Elastic Large Deflection Response of A Z-Shaped Cantilever under End Load

Boundary Condition
All degrees of freedom are constrained at the built-in end.

Illustrations of the Model

Figure 6.73-1 Elastic Large Deflection Response of A Z-Shaped Cantilever

Model File Name and Location

Model File Name Description Location

nafem_3dnlg1n.dat 3DNLG1: Elastic Large Deflection Response of A \verifman\nafems
Z-Shaped Cantilever

Results

Tip Z-Displacement
Number of Elements
Target
Load Value 30 60 90 120 150
104.5 80.4 80.690 80.686 80.685 80.685 80.685
1263 133.1 133.146 133.126 133.122 133.121 133.120
4000 143.4 143.534 143.482 143.472 143.469 143.467

Main Index
 472 MSC Nastran Verification Guide

Moment at Section A-A

Number of Elements
Target
Load Value 30 60 90 120 150
104.5 -8309 -8359.355 -8359.653 -8359.739 -8359.780 -8359.803
1263 -4966 -5125.586 -5145.153 -5148.734 -5149.957 -5150.507
4000 10743 11064.210 143.482 11060.750 11059.610 11059.120

Conclusion
The results for tip Z-displacement and moment at section A-A from MSC Nastran are compared with
NAFEMS results at loads P = 104.5, 1263.0 and 4000.0 in the above tables. The largest deviation of tip Z-
displacement from the NAFEMS solution is less than 0.4% and largest deviation of moment at section A-A
from the NAFEMS solution is less than 3.8%.The results of MSC Nastran is also found to be accurate even
for the course mesh of 30 elements for this model.

Main Index
 CHAPTER 6.74 473
3DNLG2: Elastic Large Deflection Response of a Pear-Shaped Cylinder under End Shortening

3DNLG2: Elastic Large

6.74 Deflection Response of a
Pear-Shaped Cylinder
under End Shortening

Problem Description
This NAFEMS benchmark involves initial softening in the load defection path due to the collapse of the flat
plate sections. This is accompanied by stress redistribution and subsequent stiffening to a maximum load,
which causes buckling of curved plate sections.

Solution Number
SOL 400

Features Used
 Implicit nonlinear analysis with elastic large defection and large rotation
 Initial softening behavior followed by stiffening to final buckling
 Interaction of bending, membrane and shear effects.

Reference
1. Test 3DNLG-2 from NAFEMS Publication 'A Review of Benchmark Problems for Geometric Non-
Linear Behavior of 3D Beams & Shells (Summary)' by Prinja. N & Clegg. R, R0024, August 1993.

Modeling Techniques Used

The pear shaped cylinder is modeled using 960 linear CQUAD4 shell elements. The model details are given
below:

Material Properties
E = 1.0×107 and  = 0.3

Main Index
 474 MSC Nastran Verification Guide

Loading
Uniform negative z-displacement at nodes initially lying on plane z =0.4. Due to limitation of arc length
method's implementation in MSC Nastran for enforced displacement, the end shortening is applied with a
maximum load of 2600 in -z direction.

Boundary Condition
Ux = y = z = 0.0 on plane x = 0.0, Uz = x = y = 0.0 on plane z = 0.0 and Ux = Uy = 0.0 on plane z = 0.4.

Illustrations of the Model

Figure 6.74-1 Pear Shaped Cylinder under End Shortening

Model File Name and Location

Model File Name Description Location

nafem_3dnlg2.dat 3DNLG2: Elastic Large Deflection Response of a \verifman\nafems
Pear-Shaped Cylinder under End Shortening

Main Index
 CHAPTER 6.74 475
3DNLG2: Elastic Large Deflection Response of a Pear-Shaped Cylinder under End Shortening

Results

Conclusion
The results from MSC Nastran are compared with NAFEMS results at loads 249.26, 1285.44, 2255.87,
2437 in the above figure. The largest deviation from the NAFEMS solution around 3.3% at all these
reference points. The collapse load predicted by MSC Nastran is 2511.97 which is 3.1% higher compared
to the NAFEMS estimated collapse load of 2437.

Main Index
 476 MSC Nastran Verification Guide

3DNLG6: Buckling of a Flat

6.75 Plate When Subjected To
In-Plane Shear

Problem Description
This NAFEMS benchmark on buckling of imperfect flat plate demonstrates shell element's formulation for
modeling shear behavior.

Solution Number
SOL 400

Features Used
 Implicit nonlinear analysis.
 Large rotation - large displacement response with in-plane shear
 Initial imperfection.

Reference
1. Test 3DNLG-6 from NAFEMS Publication 'A Review of Benchmark Problems for Geometric Non-
Linear Behavior of 3D Beams & Shells (Summary)' by Prinja. N & Clegg. R, R0024, August 1993.

Modeling Techniques Used

The square plate is modeled using 64 quadratic CQUAD8 shell elements. The model details are given below:

Material Properties
E = 6.4x106  = 0.3

Loading
Maximum shear load/edge = 2.12x108 is applied in all four edges using modified Riks arc length method.

Boundary Condition
Uz = 0.0 in all four edges. Ux = Uy = z = 0.0 at central node.

Main Index
 CHAPTER 6.75 477
3DNLG6: Buckling of a Flat Plate When Subjected To In-Plane Shear

Initial Imperfection
The initial imperfection is defined by the following equations:
z = 0.5  t  f(x y)  f(L  2 L  2)
where;

mx ny
f(x y) =   a mn sin  ----------
L 
- sin  ---------
 L 
m n

a11=1.0a22 = 0.2897
a13 = -0.0706a31 = -0.0691
a13 = 0.0384a42 = 0.0032
These imperfection amplitudes calculated at all grid points are used to define imperfection shape of the
model. The bulk data entry 'IMPGEOM' is used to define this imperfection geometry in MSC Nastran
input file.

Illustrations of the Model

Figure 6.75-1 Buckling of flat plate

Main Index
 478 MSC Nastran Verification Guide

Model File Name and Location

Model File Name Description Location

nafem_3dnlg6.dat 3DNLG6: Buckling of a Flat Plate when \verifman\nafems
nafem_3dnlg6.impf Subjected To In-Plane Shear

Results

Conclusion
The results from MSC Nastran are compared with NAFEMS results at edge loads 0.5442E8, 0.7585E8,
1.074E8, 1.3521E8, 1.7718E8 and 2.1117E8 in the above figure. The largest deviation from the NAFEMS
solution is less than 1.0% at all these reference points.

Main Index
 CHAPTER 6.76 479
3DNLG7: Elastic Large Deflection Response of a Hinged Spherical Shell Under Pressure Loading

3DNLG7: Elastic Large

6.76 Deflection Response of a
Hinged Spherical Shell
Under Pressure Loading

Problem Description
This NAFEMS benchmark involves snap through buckling response of hinged spherical shell under pressure
load. It demonstrates the ability of Riks arc length method to trace the positive and negative tangents in the
load-displacement equilibrium path.

Solution Number
SOL 400

Features Used
 Implicit nonlinear analysis.
 Snap through buckling with arc length method.
 Large displacement - large rotation response.

Reference
1. Test 3DNLG-7 from NAFEMS Publication 'A Review of Benchmark Problems for Geometric Non-
Linear Behaviour of 3D Beams & Shells (Summary)' by Prinja. N & Clegg. R, R0024, August 1993.

Modeling Techniques Used

The hinged spherical shell is modelled using 1024 linear CQUAD4 shell elements. The shell mid surface is
defined as Z = 2.0285x10-4[X(1570-X)+Y(1570-Y)]. Other details of the model are given below:

Material Properties
E = 69  = 0.3

Loading
Maximum distributed pressure of 0.1 applied normal to shell surface using Riks arc length method.

Main Index
 480 MSC Nastran Verification Guide

Boundary Condition
Ux = Uy = Uz = 0.0 on all four edges.

Illustrations of the Model

Figure 6.76-1 Hinged Spherical Shell under Pressure Loading

Model File Name and Location

Model File Name Description Location

nafem_3dnlg7.dat 3DNLG7: Elastic Large Deflection Response of a \verifman\nafems
Hinged Spherical Shell Under Pressure Loading

Main Index
 CHAPTER 6.76 481
3DNLG7: Elastic Large Deflection Response of a Hinged Spherical Shell Under Pressure Loading

Results

Conclusion
The results from MSC Nastran are compared with NAFEMS results at loads 0.0624, 0.0468, 0.02885 and
0.1 in the above figure. The largest deviations from the NAFEMS solution are around 1.83% and -0.93%
at limit points 1 and 2 respectively.

Main Index
 482 MSC Nastran Verification Guide

3DNLG9: Large Elastic

6.77 Deflection of a Pinched
Hemispherical Shell

Problem Description
This NAFEMS benchmark is a good test case for membrane locking of doubly curved shell and it is an
extension of linear NAFEMS benchmark on pinched hemispherical shell to incorporate the effect of large
displacement and large rotation.

Solution Number
SOL 400

Features Used
 Implicit nonlinear analysis.
 Large rigid body rotation with stress redistribution.
 Membrane, bending and twisting structural actions.

Reference
1. Test 3DNLG-9 from NAFEMS Publication 'A Review of Benchmark Problems for Geometric Non-
Linear Behavior of 3D Beams & Shells (Summary)' by Prinja. N & Clegg. R, R0024, August 1993.

Modeling Techniques Used

The hemispherical shell is modeled using 256 linear CQUAD4 shell elements. Other details of the model are
given below:

Material Properties
E = 6.825x107  = 0.3

Loading
Maximum load of Px =.-100 at point A and Py = 100 at point B

Main Index
 CHAPTER 6.77 483
3DNLG9: Large Elastic Deflection of a Pinched Hemispherical Shell

Boundary Condition
Ux = y = z = 0.0 on plane x = 0, Uy = x = z = 0.0 on plane y = 0, Uz = 0 at point A to prevent rigid
body motion.

Illustrations of the Model

Figure 6.77-1 Hinged Spherical Shell under Pressure Loading

Model File Name and Location

Model File Name Description Location

nafem_3dnlg9.dat 3DNLG-9: Large Elastic Deflection of a Pinched \verifman\nafems
Hemispherical Shell

Main Index
 484 MSC Nastran Verification Guide

Results

Conclusion
The results from MSC Nastran are compared with NAFEMS results in the above figure. The maximum
deviation of 2.1% and 1.2% are seen at points A and B respectively.

Main Index
 CHAPTER 6.78 485
3DNLG10: Elastic-Plastic Behavior of a Stiffened Cylindrical Panel under Compressive End Load

3DNLG10: Elastic-Plastic
6.78 Behavior of a Stiffened
Cylindrical Panel under
Compressive End Load

Problem Description
This NAFEMS benchmark illustrate the snap through behavior of stiffened shell panel shell with interaction
between geometric and material nonlinearities.

Solution Number
SOL 400

Features Used
 Implicit nonlinear analysis.
 Snap through with elasto-plastic large deflection behavior.
 Initial imperfection.

Reference
1. Test 3DNLG-10 from NAFEMS Publication 'A Review of Benchmark Problems for Geometric
Non-Linear Behaviour of 3D Beams & Shells (Summary)' by Prinja. N & Clegg. R, R0024, August
1993.

Modeling Techniques Used

The cylindrical panel is modelled using 40 quadratic CQUAD8 shell elements. The model details are given
below:

Material Properties
E = 2.1x105,  = 0.3, y=350

Loading
Compressive load of 12875 distributed evenly at end A.

Main Index
 486 MSC Nastran Verification Guide

Boundary Condition
Ucirc = radial = z = 0.0 at edges 1 and 2
Ucirc = Uradial = radial = 0.0 at end A.
Uz = circ = radial = 0.0 for panel at Z = L/2
Uz = X = Y = 0.0 for stiffener at Z = L/2

Initial Imperfection
The initial imperfection for cylindrical panel and stiffener are defined by the following equations:
2
 Z
R = 0.569 sin  ------------ for cylindrical panel (cylindrical system -RZ )
 7200 

where;
with -9 ≤  ≤ 9 and 0 ≤ Z ≤ 200

x y
and z =  – 0.3  sin  --------- sin  ------
 400  20

with 0 ≤ x ≤ 200 and 0 ≤ y ≤ 10.

These imperfection amplitudes calculated at all grid points are used to define imperfection shape of the
model. The bulk data entry IMPGEOM is used to define this imperfection geometry in MSC Nastran input
file.

Illustrations of the Model

Figure 6.78-1 Buckling of flat plate

Main Index
 CHAPTER 6.78 487
3DNLG10: Elastic-Plastic Behavior of a Stiffened Cylindrical Panel under Compressive End Load

Model File Name and Location

Model File Name Description Location

nafem_3dnlg10.dat 3DNLG-10: Elastic-Plastic Behavior of a \verifman\nafems
nafem_3dnlg10.impf Stiffened Cylindrical Panel under Compressive
End Load

Results

Conclusion
The results from MSC Nastran are compared with NAFEMS results at loads 23355, 15785 and 18122 in
the above figure. As shown in the following table, the load at limit points 1 and 2 from MSC Nastran show
the deviation of less than 2.0% compared to the NAFEMS solution.

NAFEMS Results MSC Nastran Results

P UZ at point D P UZ at point D % Error
Limit Point 1 23742.0 0.1679 24208.8 0.1710 1.97%
Limit Point 2 15456.0 0.1377 15715.5 0.1390 1.68%

Main Index
 488 MSC Nastran Verification Guide

Main Index