Advanced Nonlinear TMG 1 200

NX Nastran 12
Advanced Nonlinear Theory and Modeling

Guide
Proprietary & Restricted Rights Notice
© 2017 Siemens Product Lifecycle Management Software Inc. All Rights Reserved.
This software and related documentation are proprietary to Siemens Product Lifecycle
Management Software Inc. Siemens and the Siemens logo are registered trademarks of
Siemens AG. NX is a trademark or registered trademark of Siemens Product Lifecycle
Management Software Inc. or its subsidiaries in the United States and in other countries.
NASTRAN is a registered trademark of the National Aeronautics and Space
Administration. NX Nastran is an enhanced proprietary version developed and
maintained by Siemens Product Lifecycle Management Software Inc.
MSC is a registered trademark of MSC.Software Corporation. MSC.Nastran and
MSC.Patran are trademarks of MSC.Software Corporation.
All other trademarks are the property of their respective owners.
TAUCS Copyright and License
TAUCS Version 2.0, November 29, 2001. Copyright (c) 2001, 2002, 2003 by Sivan
Toledo, Tel-Aviv University, stoledo@tau.ac.il. All Rights Reserved.
TAUCS License:
Your use or distribution of TAUCS or any derivative code implies that you agree to this
License.
THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY
EXPRESSED OR IMPLIED. ANY USE IS AT YOUR OWN RISK.
Permission is hereby granted to use or copy this program, provided that the Copyright,
this License, and the Availability of the original version is retained on all copies. User
documentation of any code that uses this code or any derivative code must cite the
Copyright, this License, the Availability note, and "Used by permission." If this code or
any derivative code is accessible from within MATLAB, then typing "help taucs" must
cite the Copyright, and "type taucs" must also cite this License and the Availability note.
Permission to modify the code and to distribute modified code is granted, provided the
Copyright, this License, and the Availability note are retained, and a notice that the code
was modified is included. This software is provided to you free of charge.
Availability (TAUCS)
As of version 2.1, we distribute the code in 4 formats: zip and tarred-gzipped (tgz), with
or without binaries for external libraries. The bundled external libraries should allow you
to build the test programs on Linux, Windows, and MacOS X without installing additional
software. We recommend that you download the full distributions, and then perhaps
replace the bundled libraries by higher performance ones (e.g., with a BLAS library that
is specifically optimized for your machine). If you want to conserve bandwidth and you
want to install the required libraries yourself, download the lean distributions. The zip
and tgz files are identical, except that on Linux, Unix, and MacOS, unpacking the tgz
file ensures that the configure script is marked as executable (unpack with tar zxvpf),
otherwise you will have to change its permissions manually.
HDF5 (Hierarchical Data Format 5) Software Library and Utilities Copyright

2006-2016 by The HDF Group
NCSA HDF5 (Hierarchical Data Format 5) Software Library and Utilities Copyright
1998-2006 by the Board of Trustees of the University of Illinois. All rights reserved.
Redistribution and use in source and binary forms, with or without modification, are
permitted for any purpose (including commercial purposes) provided that the following
conditions are met:
1. Redistributions of source code must retain the above copyright notice, this list of
conditions, and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright notice, this list
of conditions, and the following disclaimer in the documentation and/or materials
provided with the distribution.
3. In addition, redistributions of modified forms of the source or binary code must carry
prominent notices stating that the original code was changed and the date of the
change.
4. All publications or advertising materials mentioning features or use of this software

are asked, but not required, to acknowledge that it was developed by The HDF
Group and by the National Center for Supercomputing Applications at the University
of Illinois at Urbana-Champaign and credit the contributors.
5. Neither the name of The HDF Group, the name of the University, nor the name of any
Contributor may be used to endorse or promote products derived from this software
without specific prior written permission from The HDF Group, the University, or the
Contributor, respectively.
DISCLAIMER: THIS SOFTWARE IS PROVIDED BY THE HDF GROUP AND THE

CONTRIBUTORS "AS IS" WITH NO WARRANTY OF ANY KIND, EITHER EXPRESSED
OR IMPLIED.
In no event shall The HDF Group or the Contributors be liable for any damages suffered
by the users arising out of the use of this software, even if advised of the possibility of
such damage
Table of contents
Table of contents
1. Introduction ............................................................................................................ 1
1.1 Objective of this manual.................................................................................... 1
1.2 Overview of Advanced Nonlinear Solution ...................................................... 2
1.2.1 Choosing between Solutions 601 and 701.................................................. 4
1.2.2 Units ........................................................................................................... 7
1.3 Structure of Advanced Nonlinear Solution ....................................................... 7
1.3.1 Executive Control ....................................................................................... 7
1.3.2 Case Control ............................................................................................... 9
1.3.3 Bulk Data.................................................................................................. 11
1.3.4 Terminology used in Advanced Nonlinear Solution ................................ 14
2. Elements ................................................................................................................ 15
2.1 Rod elements ................................................................................................... 22
2.1.1 General considerations ............................................................................. 22
2.1.2 Material models and formulations ............................................................ 23
2.1.3 Numerical integration ............................................................................... 23
2.1.4 Mass matrices ........................................................................................... 23
2.1.5 Heat transfer capabilities .......................................................................... 24
2.2 Beam elements ................................................................................................ 24
2.2.1 Beam geometry and cross-sections ........................................................... 25
2.2.2 Beam element formulations....................................................................... 29
2.2.2.1 Kinematics ......................................................................................... 29
2.2.2.2 Linear formulation............................................................................ 31
2.2.2.3 Materially-nonlinear-only formulation ............................................. 31
2.2.2.4 Large displacement formulation ....................................................... 32
2.2.2.5 Mass matrices ................................................................................... 34
2.2.2.6 Elastic beam element ........................................................................ 36
2.2.2.7 Elastic-plastic beam element ............................................................ 37
2.2.4 Pin flag option .......................................................................................... 42
2.2.5 Beam element modeling hints ................................................................... 45
2.3 Shell elements.................................................................................................. 46
2.3.1 Basic assumptions in element formulation ............................................... 48
2.3.3 Shell nodal point degrees of freedom ....................................................... 55
2.3.4 Composite shell elements (Solution 601 only) ......................................... 62
2.3.6 Mass matrices ........................................................................................... 66
Advanced Nonlinear Solution  Theory and Modeling Guide iii

Table of contents
2.3.8 Selection of elements for analysis of thin and thick shells ....................... 69
2.3.9 3D-shell element....................................................................................... 70
2.4 Surface elements – 2-D solids (Solution 601 only) ......................................... 77
2.4.4 Mass matrices ........................................................................................... 87
2.4.6 Recommendations on use of elements ..................................................... 88
2.5 Solid elements – 3-D ....................................................................................... 88
2.5.2 Material models and nonlinear formulations ............................................ 95
2.5.4 Mass matrices ........................................................................................... 97
2.5.6 Recommendations on use of elements ..................................................... 98
2.6 Scalar elements – Springs, masses and dampers ............................................. 99
2.6.1 CELAS1, CELAS2, CMASS1, CMASS2, CDAMP1, CDAMP2 ........... 99
2.6.2 6-DOF spring element (Solution 601 only) .............................................. 99
2.7 R-type elements ............................................................................................. 104
2.7.1 Rigid elements ........................................................................................ 104
2.7.2 RBE3 element......................................................................................... 111
2.8 Potential-based fluid elements (Solution 601,106 only) ............................... 115
2.8.1 Theory .................................................................................................... 115
2.8.2 Elements ................................................................................................. 120
2.8.3 Fluid boundary conditions ...................................................................... 121
2.8.4 Loads ...................................................................................................... 123
2.8.5 Phi model completion ............................................................................. 123
2.9 Other element types ....................................................................................... 136
2.9.1 Gap element............................................................................................ 136
2.9.2 Concentrated mass element .................................................................... 137
2.9.3 Bushing element ..................................................................................... 138
3. Material models and formulations .................................................................... 139
3.1 Stress and strain measures ............................................................................. 143
3.1.1 Kinematic formulations .......................................................................... 143
3.1.2 Strain measures....................................................................................... 145
3.1.3 Stress measures....................................................................................... 147
3.1.4 Large strain thermo-plasticity analysis with the ULH formulation ........ 148
3.1.5 Large strain thermo-plasticity analysis with the ULJ formulation ......... 152
3.1.6 Thermal strains ....................................................................................... 154
iv Advanced Nonlinear Solution  Theory and Modeling Guide

Table of contents
3.2 Linear elastic material models ....................................................................... 157

3.2.1 Elastic-isotropic material model ............................................................. 158
3.2.2 Elastic-orthotropic material model ......................................................... 159
3.3 Nonlinear elastic material model ................................................................... 162
3.3.1 Nonlinear elastic material for rod element ............................................. 167
3.4 Isothermal plastic material models ................................................................ 169
3.4.1 Plastic-bilinear and plastic-multilinear material models ........................ 169
3.4.2 Plastic-cyclic material model ................................................................. 180
3.4.2.1 Fundamental concepts .................................................................... 181
3.4.2.2 Specification of input ..................................................................... 198
3.5 Temperature-dependent elastic material models ........................................... 200
3.6 Thermal elasto-plastic and creep material models ........................................ 202
3.6.1 Evaluation of thermal strains .................................................................. 207
3.6.2 Evaluation of plastic strains ................................................................... 207
3.6.3 Evaluation of creep strains ..................................................................... 212
3.6.4 Computational procedures ...................................................................... 215
3.7 Hyperelastic material models ........................................................................ 217
3.7.1 Mooney-Rivlin material model .............................................................. 219
3.7.2 Ogden material model ............................................................................ 222
3.7.3 Arruda-Boyce material model ................................................................ 223
3.7.4 Hyperfoam material model ..................................................................... 225
3.7.5 Sussman-Bathe material model .............................................................. 227
3.7.6 Thermal strain effect .............................................................................. 235
3.7.7 Viscoelastic effects (Solution 601 only)................................................. 237
3.7.8 Mullins effect (Solution 601 only) ......................................................... 246
3.8 Gasket material model (Solution 601 only)................................................... 251
3.9. Shape memory alloy (Solution 601 only) ...................................................... 258
3.10 Viscoelastic material model (Solution 601 only) ........................................ 266
3.11 Heat transfer materials (Solution 601 only) ................................................ 269
3.11.1 Constant isotropic material properties .................................................. 270
3.11.2 Constant orthotropic conductivity ........................................................ 270
3.11.3 Temperature dependent thermal properties .......................................... 270
4. Contact conditions .............................................................................................. 271
4.1 Overview ....................................................................................................... 273
4.2 Contact algorithms for Solution 601 ............................................................. 282
4.2.1 Constraint-function method .................................................................... 282
4.2.2 Segment (Lagrange multiplier) method.................................................. 285
4.2.3 Rigid target method ................................................................................ 285
4.2.4 Selection of contact algorithm ................................................................ 285
4.3 Contact algorithms for Solution 701 ............................................................. 285
Advanced Nonlinear Solution  Theory and Modeling Guide v

Table of contents
4.3.1 Kinematic constraint method.................................................................. 286

4.3.2 Penalty method ....................................................................................... 287
4.3.3 Rigid target method ................................................................................ 288
4.3.4 Selection of contact algorithm ................................................................ 289
4.4 Contact set properties .................................................................................... 289
4.5 Friction .......................................................................................................... 300
4.5.1 Basic friction model ............................................................................... 300
4.5.2 Pre-defined friction models (Solution 601 only) .................................... 300
4.5.3 Frictional heat generation ....................................................................... 302
4.6 Contact analysis features ............................................................................... 303
4.6.1 Dynamic contact/impact ......................................................................... 303
4.6.2 Contact detection .................................................................................... 304
4.6.3 Suppression of contact oscillations (Solution 601) ................................ 305
4.6.4 Restart with contact ................................................................................ 306
4.6.5 Contact damping..................................................................................... 306
4.7 Modeling considerations ............................................................................... 307
4.7.1 Contactor and target selection ................................................................ 307
4.7.2 General modeling hints .......................................................................... 309
4.7.3 Modeling hints specific to Solution 601 ................................................. 313
4.7.4 Modeling hints specific to Solution 701 ................................................. 314
4.7.5 Convergence considerations (Solution 601 only)................................... 316
4.7.6 Handling improperly supported bodies .................................................. 318
4.8 Rigid target contact algorithm ....................................................................... 321
4.8.1 Introduction ............................................................................................ 321
4.8.2 Basic concepts ........................................................................................ 323
4.8.2.1 Contactor surfaces .......................................................................... 323
4.8.2.2 Target surfaces ............................................................................... 326
4.8.2.3 Determination of contact between contactor and target ................. 326
4.8.2.4 Frictional contact ............................................................................ 333
4.8.3 Modeling considerations ........................................................................ 335
4.8.4 Rigid target contact reports for Solution 601 ......................................... 345
4.8.5 Rigid target contact report for Solution 701 ........................................... 347
4.8.6 Modeling hints and recommendations.................................................... 348
5. Loads, boundary conditions and constraint equations ................................... 352
5.1 Introduction ................................................................................................... 352
5.2 Concentrated loads ........................................................................................ 360
5.3 Pressure and distributed loading.................................................................... 362
5.4 Inertia loads ─ centrifugal and mass proportional loading............................ 365
5.5 Enforced motion ............................................................................................ 372
5.6 Applied temperatures .................................................................................... 374
vi Advanced Nonlinear Solution  Theory and Modeling Guide

Table of contents
5.7 Bolt preload ................................................................................................... 376

5.8 Constraint equations ...................................................................................... 377
5.9 Mesh glueing ................................................................................................. 379
5.10 Convection boundary condition .................................................................. 385
5.11 Radiation boundary condition ..................................................................... 386
5.12 Boundary heat flux load .............................................................................. 388
5.13 Internal heat generation ............................................................................... 389
6. Static and implicit dynamic analysis ................................................................ 390
6.1 Linear static analysis ..................................................................................... 390
6.2 Nonlinear static analysis ................................................................................ 391
6.2.1 Solution of incremental nonlinear static equations................................. 392
6.2.2 Line search ............................................................................................. 393
6.2.3 Low speed dynamics feature .................................................................. 395
6.2.4 Automatic-Time-Stepping (ATS) method .............................................. 396
6.2.5 Total Load Application (TLA) method and Stabilized TLA (TLA-S)
method .............................................................................................................. 399
6.2.6 Load-Displacement-Control (LDC) method .......................................... 402
6.2.7 Convergence criteria for equilibrium iterations...................................... 405
6.2.8 Selection of incremental solution method .............................................. 410
6.2.9 Example .................................................................................................. 413
6.3 Linear dynamic analysis ................................................................................ 419
6.3.1 Mass matrix ............................................................................................ 422
6.3.2 Damping ................................................................................................. 423
6.4 Nonlinear dynamic analysis .......................................................................... 424
6.4.1 Step-by-step implicit time integration .................................................... 424
6.4.2 Global mass matrix ................................................................................. 426
6.5 Solvers ........................................................................................................... 427
6.5.1 Direct sparse solver ................................................................................ 427
6.5.2 Iterative multigrid solver ........................................................................ 429
6.5.3 3D-iterative solver .................................................................................. 431
6.6 Tracking solution progress ............................................................................ 433
7. Explicit dynamic analysis .................................................................................. 435
7.1 Formulation ................................................................................................... 435
7.1.1 Mass matrix ............................................................................................ 437
7.1.2 Damping ................................................................................................. 437
7.2 Stability ......................................................................................................... 438
7.3 Time step management .................................................................................. 441
7.4 Tracking solution progress ............................................................................ 442
8. Heat transfer analysis (Solution 601 only) ....................................................... 444
Advanced Nonlinear Solution  Theory and Modeling Guide vii

Table of contents
8.1 Formulation ................................................................................................... 444

8.2 Loads, boundary conditions, and initial conditions ....................................... 446
8.3 Steady state analysis ...................................................................................... 448
8.4 Transient analysis .......................................................................................... 449
8.5 Choice of time step and mesh size ................................................................ 450
8.6 Automatic time stepping method................................................................... 453
9. Coupled thermo-mechanical analysis (Solution 601 only) .............................. 455
9.1 One-way coupling ......................................................................................... 457
9.2 Iterative coupling ........................................................................................... 457
10. Additional capabilities ..................................................................................... 461
10.1 Initial conditions .......................................................................................... 461
10.1.1 Initial displacements and velocities ...................................................... 461
10.1.2 Initial temperatures ............................................................................... 461
10.2 Restart.......................................................................................................... 462
10.3 Element birth and death feature ................................................................... 464
10.4 Element death due to rupture....................................................................... 472
10.5 Reactions calculation ................................................................................... 473
10.6 Stiffness stabilization (Solution 601 only) .................................................. 474
10.7 Bolt feature (Solution 601 only) .................................................................. 476
10.7.1 Beam-bolt element ............................................................................... 478
10.7.2 3D-bolt ................................................................................................. 480
10.7.3 Usage of bolt loadings .......................................................................... 484
10.7.4 Usage of beam-bolts ............................................................................. 487
10.7.5 Usage of 3D-bolts ................................................................................. 487
10.7.6 Modeling issues .................................................................................... 492
10.8 Direct matrix input (Solution 601 only) ...................................................... 492
10.9 Parallel processing ....................................................................................... 493
10.10 Usage of memory and disk storage ........................................................... 493
Additional reading.................................................................................................. 495
Index ........................................................................................................................ 496
viii Advanced Nonlinear Solution  Theory and Modeling Guide

1.1: Objective of this manual
1. Introduction
1.1 Objective of this manual
This Theory and Modeling Guide serves two purposes:
To provide a concise summary of the theoretical basis of

Advanced Nonlinear Solution as it applies to Solution 601 and
Solution 701. This includes the finite element procedures used,
the elements and the material models. The depth of coverage of
these theoretical issues is such that the user can effectively use
Solutions 601 and 701. A number of references are provided
throughout the manual which give more details on the theory
and procedure used in the program. These references should be
consulted for further details. Much reference is made however
to the book Finite Element Procedures (ref. KJB).
ref. K.J. Bathe, Finite Element Procedures, 2nd ed.,

Cambridge, MA, Klaus-Jürgen Bathe, 2014.
 To provide guidelines for practical and efficient modeling using

Advanced Nonlinear Solution. These modeling guidelines are
based on the theoretical foundation mentioned above, and the
capabilities and limitations of the different procedures,
elements, material models and algorithms available in the
program. NX Nastran commands and parameter settings needed
to activate different analysis features are frequently mentioned.
It is assumed that the user is familiar with NX Nastran

fundamentals pertaining to linear analysis. This includes general
knowledge of the NX Nastran structure, commands, elements,
materials, and loads.
We intend to update this report as we continue our work on
Advanced Nonlinear Solution. If you have any suggestions
regarding the material discussed in this manual, we would be glad
to hear from you.
Advanced Nonlinear Solution  Theory and Modeling Guide 1

Chapter 1: Introduction
1.2 Overview of Advanced Nonlinear Solution

 Advanced Nonlinear Solution is a Nastran solution option
focused on nonlinear problems. It is capable of treating geometric
and material nonlinearities as well as nonlinearities resulting from
contact conditions. State-of-the-art formulations and solution
algorithms are used which have proven to be reliable and efficient.
 Advanced Nonlinear Solution supports static and implicit

dynamic nonlinear analysis via Solution 601, and explicit dynamic
analysis via Solution 701. Solution 601 also supports heat transfer
analysis and coupled structural heat transfer analysis.
 Advanced Nonlinear Solution supports many of the standard

Nastran commands and several commands specific to Advanced
Nonlinear Solution that deal with nonlinear features such as
contact. The NX Nastran Quick Reference Guide provides more
details on the Nastran commands and entries that are supported in
Advanced Nonlinear Solution.
 Advanced Nonlinear Solution supports many of the commonly

used features of linear Nastran analysis. This includes most of the
elements, materials, boundary conditions, and loads. Some of these
features are modified to be more suitable for nonlinear analysis,
and many other new features are added that are needed for
nonlinear analysis.
 The elements available in Advanced Nonlinear Solution can be

broadly classified into rods, beams, 2-D solids, 3-D solids, shells,
2-D fluids, 3-D fluids, scalar elements and rigid elements. The
formulations used for these elements have proven to be reliable and
efficient in linear, large displacement, and large strain analyses.
Chapter 2 provides more details on the elements.
 The material models available in Advanced Nonlinear Solution

are elastic isotropic, elastic orthotropic, plastic bilinear/multilinear,
plastic-cyclic, hyperelastic, gasket, nonlinear elastic isotropic,
shape memory alloy, viscoelastic and fluid. Thermal and creep
effects can be added to some of these materials. Chapter 3 provides
more details on these material models.
2 Advanced Nonlinear Solution  Theory and Modeling Guide

1.2: Overview of Advanced Nonlinear Solution
 Advanced Nonlinear Solution has very powerful features for

contact analysis. These include several contact algorithms and
different contact types such as single-sided contact, double-sided
contact, self-contact, and tied contact. Chapter 4 provides more
details on contact.
 Loads, boundary conditions and constraints are addressed in

Chapter 5. Time varying loads and boundary conditions are
common to nonlinear analysis and their input in Advanced
Nonlinear Solution is slightly different from other Nastran
solutions, as discussed in Chapter 5.
 Solution 601 of Advanced Nonlinear Solution currently

supports two nonlinear structural analysis types: static and implicit
transient dynamic. Details on the formulations used are provided in
Chapter 6. Other features of nonlinear analysis, such as time
stepping, load displacement control (arc length method), line
search, and available solvers are also discussed in Chapter 6.
 Solution 701 of Advanced Nonlinear Solution is dedicated to

explicit transient dynamic analysis. Details on the formulations
used are provided in Chapter 7. Other features of explicit analysis,
such as stability and time step estimation, are also discussed in
Chapter 7.
 Solution 601 of Advanced Nonlinear Solution also supports two

heat transfer or coupled structural heat transfer analysis types. The
first type 153 is for static structural with steady state heat transfer,
or just steady state heat transfer. The second analysis type 159 is
for cases when either the structural or heat transfer models are
transient (dynamic). This type can also be used for just transient
heat transfer analysis. Details of the heat transfer analysis are
provided in Chapter 8, and details of the thermo-mechanical
coupled (TMC) analysis are provided in Chapter 9.
 Additional capabilities present in Advanced Nonlinear Solution

such as restarts, stiffness stabilization, initial conditions, and
parallel processing are discussed in Chapter 10.

 Most of the global settings controlling the structural solutions in

Advanced Nonlinear Solution are provided in the NXSTRAT bulk
data entry. This includes parameters that control the solver
selection, time integration values, convergence tolerances, contact
settings, etc. An explanation of these parameters is found in the NX
Nastran Quick Reference Guide.
 Similarly, most of the global settings controlling the heat

transfer or coupled solutions in Advanced Nonlinear Solution are
provided in the TMCPARA bulk data entry.
1.2.1 Choosing between Solutions 601 and 701


 The main criterion governing the selection of the implicit
(Solution 601) or explicit (Solution 701) formulations is the time
scale of the solution.
ref. KJB  The implicit method can use much larger time steps since it is
Section 9.2 unconditionally stable. However, it involves the assembly and
solution of a system of equations, and it is iterative. Therefore, the
computational time per load step is relatively high. The explicit
method uses much smaller time steps since it is conditionally
stable, meaning that the time step for the solution has to be less
than a certain critical time step, which depends on the smallest
element size and the material properties. However, it involves no
matrix solution and is non-iterative. Therefore, the computational
time per load step is relatively low.
 For both linear and nonlinear static problems, the implicit

method is the only option.
 For heat transfer and coupled structural heat transfer problems,

the implicit method is the only option.
 For low-speed dynamic problems, the solution time spans a

period of time considerably longer than the time it takes the wave
to propagate through an element. The solution in this case is
dominated by the lower frequencies of the structure. This class of
problems covers most structural dynamics problems, certain metal
forming problems, crush analysis, earthquake response and
biomedical problems. When the explicit method is used for such

problems the resulting number of time steps will be excessive,

unless mass-scaling is applied, or the loads are artificially applied
over a shorter time frame. No such modifications are needed in the
implicit method. Hence, the implicit method is the optimal choice.
 For high-speed dynamic problems, the solution time is

comparable to the time required for the wave to propagate through
the structure. This class of problems covers most wave propagation
problems, explosives problems, and high-speed impact problems.
For these problems, the number of steps required with the explicit
method is not excessive. If the implicit method uses a similar time
step it will be much slower and if it uses a much larger time step it
will introduce other solution errors since it will not be capturing the
pertinent features of the solution (but it will remain stable). Hence,
the explicit method is the optimal choice.
 A large number of dynamics problems cannot be fully classified

as either low-speed or high-speed dynamic. This includes many
crash problems, drop tests and metal forming problems. For these
problems both solution methods are comparable. However,
whenever possible (when the time step is relatively large and there
are no convergence difficulties) we recommend the use of the
implicit solution method.
 Note that the explicit solution provided in Solution 701 does not
use reduced integration with hour-glassing. This technique reduces
the computational time per load step. However, it can have
detrimental effect on the accuracy and reliability of the solution.
 Since the explicit time step size depends on the length of the
smallest element, one excessively small element will reduce the
stable time step for the whole model. Mass-scaling can be applied
to these small elements to increase their stable time step. The
implicit method is not sensitive to such small elements.
 Since the explicit time step size depends on the material

properties, a nearly incompressible material will also significantly
reduce the stable time step. The compressibility of the material can
be increased in explicit analysis to achieve a more acceptable
solution time. The implicit method is not as sensitive to highly

incompressible materials (provided that a mixed formulation is

used).
 Higher order elements such as the 10-node tetrahederal, 20 and

27 node brick elements are only available in implicit analysis. They
are not used in explicit analysis because no suitable mass-lumping
technique is available for these elements.
 Model nonlinearity is another criterion influencing the choice

between implicit and explicit solutions. As the level of nonlinearity
increases, the implicit method requires more time steps, and each
time step may require more iterations to converge. In some cases,
no convergence is reached. The explicit method however, is less
sensitive to the level of nonlinearity.
Note that when the implicit method fails it is usually due to non-
convergence within a time step, while when the explicit method
fails it is usually due to a diverging solution.
 The memory requirements is another factor. For the same mesh,

the explicit method requires less memory since it does not store a
stiffness matrix and does not require a solver. This can be
significant for very large problems.
 Since Advanced Nonlinear Solution handles both Solution 601

and Solution 701 with very similar inputs, the user can in many
cases restart from one analysis type to the other. This capability can
be used, for example, to perform implicit springback analysis
following an explicit metal forming simulation, or to perform an
explicit analysis following the implicit application of a gravity
load.
It can also be used to overcome certain convergence difficulties
in implicit analyses. A restart from the last converged implicit
solution to explicit can be performed, then, once that stage is
passed, another restart from explicit to implicit can be performed to
proceed with the rest of the solution.

1.2.2 Units
In Advanced Nonlinear Solution, it is important to enter all

physical quantities (lengths, forces, masses, times, etc.) using a
consistent set of units. For example, when working with the SI
system of units, enter lengths in meters, forces in Newtons, masses
in kg, times in seconds. When working with other systems of units,
all mass and mass-related units must be consistent with the length,
force and time units. For example, in the USCS system (USCS is
an abbreviation for the U.S. Customary System), when the length
unit is inches, the force unit is pound and the time unit is second,
the mass unit is lb-sec2/in, not lb.
Rotational degrees of freedom are always expressed in radians.
1.3 Structure of Advanced Nonlinear Solution

 The input data for Advanced Nonlinear Solution follows the
standard Nastran format consisting of the following 5 sections:
1. Nastran Statement (optional)

2. File Management Statements (optional)
3. Executive control Statements
4. Case Control Statements
5. Bulk Data Entries
The first two sections do not involve any special treatment in

Advanced Nonlinear Solution. The remaining three sections
involve some features specific to Advanced Nonlinear Solution, as
described below.
1.3.1 Executive Control
 Solution 601 is invoked by selecting solution sequence 601 in

the SOL Executive Control Statement. This statement has the
following form:
SOL 601,N
where N determines the specific analysis type selected by Solution

601.

Currently, static and direct time-integration implicit dynamic

structural analyses are available as shown in Table 1.3-1. In
addition, two analysis types are available for thermal or coupled
thermal-mechanical problems.
 Solution 701 is invoked by selecting solution sequence 701 in

the SOL Executive Control Statement. This statement has the
following form:
SOL 701
and is used for explicit dynamic analyses.
 In many aspects Solution 601,106 is similar to Solution 106 for

nonlinear static analysis. However, it uses the advanced nonlinear
features of Solution 601. Likewise, Solution 601,129 is similar to
Solution 129 for nonlinear transient response analysis. Solution 701
provides an alternative to the implicit nonlinear dynamic analysis
of Solution 601,129.
Table 1.3-1: Solution 601 Analysis Types
N Solution 601 Analysis Type

106 Static
129 Transient dynamic
Steady state thermal + static
153
structural
Transient thermal + dynamic
159
structural1
1
N = 159 also allows either of the structure or the thermal parts to
be static or steady state.

1.3: Structure of Advanced Nonlinear Solution
1.3.2 Case Control
 The Case Control Section supports several commands that

control the solution, commands that select the input loads,
temperatures and boundary conditions, commands that select the
output data and commands that select contact sets. Table 1.3-2 lists
the supported Case Control Commands.
Table 1.3-2: Case Control Commands
Case Control Command Description

Solution control
SUBCASE1 Subcase delimiter
TSTEP2 Time step set selection
ANALYSIS3 Subcase analysis type solution
Loads and boundary conditions
LOAD Static load set selection
DLOAD4 Dynamic load set selection
SPC Single-point constraint set selection
MPC Multipoint constraint set selection
TEMPERATURE5 Temperature set selection
TEMPERATURE(LOAD) Temperature load
TEMPERATURE(INITIAL) Initial temperature
IC Transient initial condition set selection
BGSET Glue contact set selection
BOLTLD Bolt preload set selection
DTEMP Time-dependent temperature set selection
DMIG related
B2GG Selects direct input damping matrices
K2GG Selects direct input stiffness matrices
M2GG Selects direct input mass matrices
Element related
EBDSET Element birth/death selection

Table 1.3-2: Case Control Commands (continued)
Output related
SET Set definition
DISPLACEMENT Displacement output request
VELOCITY Velocity output request
ACCELERATION Acceleration output request
STRESS Element stress/strain output request
SPCFORCES Reaction force output request
GPFORCE Nodal force output request
GKRESULTS Gasket results output request
TITLE Output title
SHELLTHK Shell thickness output request
THERMAL Temperature output request
FLUX Heat transfer output request
OLOAD Applied load output request
BGRESULTS Glue result output request
Contact related
BCSET Contact set selection
BCRESULTS Contact results output request

Notes for Table 1.3-2:
1. Only one subcase is allowed in structural analysis Advanced Nonlinear

Solution (N = 106, 129). In coupled TMC analyses (N = 153, 159), two
subcases are required, one for the structural and one for the thermal sub-
model.
2. TSTEP is used for all analysis types in Advanced Nonlinear Solution. In
explicit analysis with automatic time stepping it is used for determining the
frequency of output of results.
3. Supports ANALYSIS = STRUC and ANALYSIS = HEAT for SOL 601,153
and SOL 601,159.
4. DLOAD is used for time-varying loads for both static and transient dynamic
analyses.
5. TEMPERATURE, TEMPERATURE(BOTH) and TEMPERATURE(MAT)
are not allowed for Advanced Nonlinear Solution. Use
TEMPERATURE(INIT) and TEMPERATURE(LOAD) instead.
1.3.3 Bulk Data
 The Bulk Data section contains all the details of the model.
Advanced Nonlinear Solution supports most of the commonly used
Bulk Data entries. In many cases, restrictions are imposed on some
of the parameters in a Bulk Data entry, and in some other cases,
different interpretation is applied to some of the parameters to
make them more suitable for nonlinear analysis. Several Bulk Data
entries are also specific to Advanced Nonlinear Solution.
 Table 1.3-3 lists the supported Bulk Data entries.

Table 1.3-3: Bulk Data entries supported by Advanced Nonlinear Solution
Element Connectivity
CBAR CMASS1 CPLSTS3 CQUADX4 CTRAX3
CBEAM CMASS2 CPLSTS4 CQUADX8 CTRAX6
CBUSH CONM1 CPLSTS6 CROD CTETRA
CBUSH1D CONM2 CPLSTS8 CTRIA3 RBAR
CDAMP1 CONROD CPYRAM CTRIA6 RBE2
CDAMP2 CPENTA CQUAD4 CTRIAR RBE3
CELAS1 CPLSTN3 CQUAD8
CELAS2 CPLSTN4 CQUADR
CGAP CPLSTN6
CHEXA CPLSTN8
Element Properties
EBDSET PBCOMP PCOMP PLPLANE PROD
EBDADD PBEAM PCOMPG PLSOLID PSHELL
PBAR PBEAML PDAMP PMASS PSHL3D
PBARL PBUSH PELAS PPLANE PSOLID
PBUSH1D PELAST
PGAP
Material Properties
CREEP MAT8 MATHE MATT1 PCONV
MAT1 MAT9 MATHEV MATT2 PLCYISO
MAT2 MAT10 MATHEM MATT3 PLCYKIN
MAT3 MAT11 MATHP MATT4 PLCYRUP
MAT4 MATCID MATPLCY MATT5 RADM
MAT5 MATCRP MATS1 MATT8 RADMT
MATG MATSMA MATT9 TABLEM1
MATSR MATT11 TABLES1
MATTC TABLEST
MATVE

Table 1.3-3: Bulk Data entries supported by Advanced Nonlinear Solution

(continued)
Loads, Boundary Conditions and Constraints

BCPROPS FORCE MPC PLOADX1 TABLED1
BEDGE FORCE1 MPCADD RFORCE TABLED2
BFLUID FORCE2 PLOAD RFORCE1 TEMP
BGSET GRAV PLOAD1 SPC TEMPD
BOLT LOAD PLOAD2 SPC1 TIC
BOLTFOR MOMENT PLOAD4 SPCADD TLOAD1
BSURFS MOMENT1 PLOADE1 SPCD
DLOAD MOMENT2
DTEMP
Heat Transfer Loads and Boundary Conditions
BDYOR CONV QHBDY TEMPBC
CHBDYE QBDY1 QVOL
CHBYDG QBDY2 RADBC
Contact
BCPROP BCRPARA BCTPARA BEDGE BSURF
BCPROPS BCTADD BCTSET BLSEG BSURFS
Direct Matrix Input
DMIG
Other Commands
CORD1C CORD1S CORD2R GRID PARAM2
CORD1R CORD2C CORD2S GROUP TMCPARA3
NXSTRAT1 TSTEP4
Notes:
1. NXSTRAT is the main entry defining the solution settings for Advanced
Nonlinear Solution.
2. Only a few PARAM variables are supported. Most are replaced by
NXSTRAT variables.
3. TMCPARA is the main entry defining the solution settings for heat
transfer and TMC models.
4. TSTEP is used for both static and dynamic analyses.

1.3.4 Terminology used in Advanced Nonlinear Solution
The terminology used in Advanced Nonlinear Solution is for the most

part the same as that used in other Nastran documents.

2. Elements
2. Elements
 Advanced Nonlinear Solution supports most of the commonly
used elements in linear Nastran analyses. Some of these elements
are modified to be more suitable for nonlinear analysis.
 The Advanced Nonlinear Solution elements are generally

classified as line, surface, solid, scalar, R-type or potential-based
fluid.
Line elements are divided into 2 main categories – rod

elements and beam elements. Rod elements only possess
axial stiffness, while beam elements also possess bending,
shear and torsional stiffness.
Surface elements are also divided into 2 main categories –
2-D solids and shell elements.
 3-D solid elements are the only solid elements in Advanced
Nonlinear Solution.
 The scalar elements are spring, mass, damper elements and
6-DOF spring elements.
 R-type elements impose constraints between nodes, such as
rigid elements.
 Potential-based fluid elements are used in static analysis to
model the compressibility of fluids.
 Other element types available in Advanced Nonlinear
Solution are the gap element, concentrated mass element,
and the bushing element.
 This chapter outlines the theory behind the different element

classes, and also provides details on how to use the elements in
modeling. This includes the materials that can be used with each
element type, their applicability to large displacement and large
strain problems, their numerical integration, etc.
 More detailed descriptions of element input and output are

provided in several other manuals, including:
- NX Nastran Reference Manual

- NX Nastran Quick Reference Guide
- NX Nastran DMAP Programmer’s Guide

Chapter 2: Elements
 Table 2-1 below shows the different elements available in

Advanced Nonlinear Solution, and how they can be obtained from
Nastran element connectivity and property ID entries. Restrictions
related to Solution 701 are noted.
Table 2-1: Elements available in Advanced Nonlinear Solution
Element Property ID Entry Advanced Nonlinear

Connectivity Entry Solution Element
Rod Elements
CROD PROD 2-node rod element
CONROD None 2-node rod element
Beam Elements
CBAR PBAR, PBARL 2-node beam element
CBEAM PBEAM, PBEAML, 2-node beam element
PBCOMP
Shell Elements3
CQUAD4 PSHELL1, PCOMP2, 4-node quadrilateral shell
PSHL3D7, PCOMPG element
CQUAD8 PSHELL1, PCOMP2, 4-node to 8-node quadrilateral
PCOMPG shell element
CQUADR PSHELL, PCOMP2, 4-node quadrilateral shell
PCOMPG element
CTRIA3 PSHELL1, PCOMP2, 3-node triangular shell
CTRIA6 PSHELL1, PCOMP2, 3-node to 6-node triangular
PCOMPG shell element
CTRIAR PSHELL, PCOMP2, 3-node triangular shell

2. Elements
Table 2-1: Elements available in Advanced Nonlinear Solution (continued)
2D Solid and Fluid Elements4

CPLSTN3 PPLANE, PLPLANE 3-node triangular 2D plane
strain element
CPLSTN4 PPLANE, PLPLANE 4-node quadrilateral 2D plane
strain element
CPLSTN6 PPLANE, PLPLANE 6-node triangular 2D plane
strain element
CPLSTN8 PPLANE, PLPLANE 8-node quadrilateral 2D plane
strain element
CPLSTS3 PPLANE, PLPLANE 3-node triangular 2D plane
stress element
CPLSTS4 PPLANE, PLPLANE 4-node quadrilateral 2D plane
stress element
CPLSTS6 PPLANE, PLPLANE 6-node triangular 2D plane
stress element
CPLSTS8 PPLANE, PLPLANE 8-node quadrilateral 2D plane
stress element
CQUAD6 PLPLANE 4-node to 9-node quadrilateral
2D plane strain element with
hyperelastic material
CQUAD4 PLPLANE, PSHELL1 4-node quadrilateral 2D plane
strain element
CQUAD8 PLPLANE, PSHELL1 4-node to 8-node 2D plane
strain element
CTRIA3 PLPLANE, PSHELL1 3-node triangular 2D plane
strain element
CTRIA6 PLPLANE, PSHELL1 3-node to 6-node triangular
2D plane strain element

Chapter 2: Elements
2D Solid and Fluid Elements (continued)

CQUADX4 PSOLID, PLSOLID 4-node quadrilateral 2D
axisymmetric element
CQUADX8 PSOLID, PLSOLID 8-node quadrilateral 2D
CTRAX3 PSOLID, PLSOLID 3-node triangular 2D
CTRAX6 PSOLID, PLSOLID 6-node triangular 2D
3D Solid and Fluid Elements5
CHEXA PSOLID, PLSOLID 8-node to 20-node brick 3D
solid element
CPENTA PSOLID, PLSOLID 6-node to 15-node wedge 3D
solid element
CTETRA PSOLID, PLSOLID 4-node to 10-node tetrahedral
3D solid element
CPYRAM PSOLID, PLSOLID 5-node to 13-node pyramid
3D solid element
Scalar Elements
CELAS1; CELAS2 PELAS; None Spring element
CDAMP1; CDAMP2 PDAMP; None Damper element
CMASS1; CMASS2 PMASS; None Mass element
R-Type Elements
RBAR None Single rigid element
RBE2 None Multiple rigid elements
RBE3 None Interpolation constraint
element

2. Elements
Other Elements
CGAP PGAP 2-node gap element
CONM1, CONM2 None Concentrated mass element
CBUSH1D PBUSH1D Rod Type Spring-and-Damper
Connection
CBUSH PBUSH Generalized Spring-and-
Damper Connection
Notes:
1. CQUAD4, CQUAD8, CTRIA3, and CTRIA6 with a PSHELL property ID are

treated as either 2D plane strain elements or shell elements depending on the
MID2 parameter.
2. Elements with PCOMP or PCOMPG property ID entries are treated as multi-
layered shell elements. These elements are not supported in Solution 701.
3. Only 3-node and 4-node single layer shells are supported in Solution 701.
4. 2-D solid elements are not supported in Solution 701.
5. Only 4-node tetrahedral, 6-node wedge and 8-node brick 3-D solid elements are
supported in Solution 701.
6. CQUAD cannot be used for potential-based fluid elements.
7. PSHL3D is used to specify a 3D-shell element.

Chapter 2: Elements
 Table 2-2 lists the acceptable combination of elements and

materials for Solution 601. Thermal effects in this table imply
temperature dependent material properties. Thermal strains are
usually accounted for in isothermal material models.
For potential-based fluid elements, only the fluid material can
be used.
Table 2-2: Element and material property combinations in Solution 601
Rod Beam Shell 2D Solid 3D Solid

Elastic isotropic     
...Thermal    
...Creep    
Elastic orthotropic   
...Thermal   
Plastic bilinear/
    
multilinear
Plastic-cyclic     
Plastic thermal    
Plastic creep    
Hyperelastic 2  
Gasket 
Nonlinear elastic
1 1 1 1
isotropic
Shape memory
   
alloy
Viscoelastic    
Notes:
1. No thermal strains in these material models.

2. 3D-shell element only.

2. Elements
 Table 2-3 lists the acceptable combination of elements and

materials for Solution 701. Thermal effects in this table imply
temperature dependent material properties. Thermal strains are
usually accounted for in isothermal material models. Note that
interpolation of temperature dependent material properties is only
performed at the start of the analysis in Solution 701.
The potential-based fluid elements are not available in Solution
701.
Table 2-3: Element and material property combinations in Solution 701
Rod Beam Shell 2D Solid 3D Solid

Elastic isotropic    
...Thermal   
...Creep
Elastic orthotropic  
...Thermal  
Plastic bilinear/
   
multilinear
Plastic-cyclic    
Plastic thermal   
Plastic creep
Hyperelastic 2 
Gasket 
Nonlinear elastic
1
isotropic
Shape memory
alloy
Viscoelastic
Notes:
1. No thermal strains in these material models.

2. 3D-shell element only.

Chapter 2: Elements
2.1 Rod elements

2.1.1 General considerations
 Rod elements are generated using the CONROD and CROD

entries. These line elements only possess axial stiffness. Fig. 2.1-1
shows the nodes and degrees of freedom of a rod element. Note
that the rod element only has 2 nodes.
Z
w2
Y G2
w1 l v2
X
G1 u2
l
v1
u1
(u, v, w) are nodal translational

degrees of freedom
Fig. 2.1-1: Rod element
ref. KJB  Note that the only force transmitted by the rod element is the
Sections 5.3.1, longitudinal force as illustrated in Fig. 2.1-2. This force is constant
6.3.3 throughout the element.
Z P
Y
X
area A
s
Stress constant over
P cross-sectional area
Fig. 2.1-2: Stresses and forces in rod elements

2.1: Rod elements
2.1.2 Material models and formulations
 See Tables 2-2 and 2-3 for a list of the material models that are
compatible with rod elements.
 The rod elements can be used with small displacement/small

strain or large displacement/small strain kinematics. In the small
displacement case, the displacements and strains are assumed
infinitesimally small. In the large displacement case, the
displacements and rotations can be very large. In all cases, the
cross-sectional area of the element is assumed to remain
unchanged, and the strain is equal to the longitudinal displacement
divided by the original length.
All of the compatible material models listed in Tables 2-2 and
2-3 can be used with both the small and large displacement
formulations.
2.1.3 Numerical integration
 The rod elements use one point Gauss integration.
2.1.4 Mass matrices
 The consistent mass matrix is calculated using Eq. (4.25) in ref.

KJB, p. 165.
 The lumped mass matrix for the rod element is formed by

dividing the element’s mass M among its nodes. The mass assigned
 i 
to each node is M    , in which L = total element length,  i =
L
fraction of the total element length associated with element node i
L L
(i.e., for the 2-node rod element, 1  and  2  ). The
2 2
element has no rotational mass.
 The same lumped mass matrix is used for both Solution 601 and
Solution 701.

Chapter 2: Elements
2.1.5 Heat transfer capabilities
 The rod element supports 1-D heat conductivity, heat capacity

and heat generation features in heat transfer and coupled TMC
analyses.
 One temperature degree of freedom is present at each node.
 The heat capacity matrix can be calculated based on a lumped or

consistent heat capacity assumption.
 In the lumped heat capacity assumption, each node gets a heat

capacity of cAL/2.
 This element can also be used as a general thermal link element

between any two points in space.
2.2 Beam elements

 The beam element is a 2-node Hermitian beam with a constant
cross-section. The element is initially straight.
 The beam element can be employed in the following analysis

conditions:
 Linear analysis, in which case the displacements, rotations

and strains are infinitesimally small, and the material is
linear elastic.
 Materially nonlinear analysis, in which case the
displacements, rotations and strains are infinitesimally
small, but the material is nonlinear.
 Large displacement/large rotation analysis, in which case
the displacements and rotations can be large, but the strains
are small. The material can either be linear or nonlinear.
 The beam element can be used in Solution 601 (statics and

implicit dynamics) and in Solution 701 (explicit dynamics).

2.2: Beam elements
 The beam element can optionally include a bolt feature. The

bolt feature is fully described in Section 10.7.
 Throughout this section, the element formulations of the current

version of Advanced Nonlinear Solution are described. It is,
however, possible to choose the element formulations used in
Advanced Nonlinear Solution of NX 8.5, using the BEAMALG
parameter of the NXSTRAT entry. (However, if the elasto-plastic
material model is chosen, then the current beam element algorithms
are always used.) BEAMALG=0 corresponds to the current version
(the default) and BEAMALG=1 corresponds to NX 8.5.
The results obtained will vary depending upon whether the
current formulations or the NX 8.5 formulations are used. The
major improvements are in the large displacement/large rotation
formulation and in the elastic-plastic formulation.
 Beam elements are generated using the CBAR and CBEAM

entries. The properties for a CBAR entry are defined using PBAR
or PBARL entries while the properties for CBEAM are defined
using the PBEAM, PBEAML or PBCOMP entries. See Tables 2-2
and 2-3 for a list of the material models that are compatible with
the beam element.
2.2.1 Beam geometry and cross-sections
 Figs 2.2-1 and 2.2-2 show the beam element along with its local
coordinate system (r,s,t). The r direction always lies along the
neutral line of the beam (line connecting nodes GA and GB). The
orientation of the s and t directions is defined using the v vector
defined in the CBAR or CBEAM entries.
Notice that, for the Hermitian beam element, (r,s,t) are not
isoparametric coordinates, rather (r,s,t) have the same units as the
global coordinates.
Fig 2.2-2 also shows the degrees of freedom at the local nodes.
These degrees of freedom are defined in the local coordinate
system.

Chapter 2: Elements
End B
Plane 1 r (xelem)
s (yelem) l
node GB
End A t (zelem)
node GA
Fig. 2.2-1: Beam element
_
qs
-
_ qr
v _u
s
l
_
r GB w _
Z
l qt
GA t
X Y
Fig. 2.2-2: Degrees of freedom and local axes

for beam element
 The forces / moments in the beam element are shown in Fig.

2.2-3. These forces / moments are also defined in the element local
coordinate system.
 The s and t directions give the orientation of the beam element

cross-section. Care must be used in defining the s and t directions
so that the beam element cross-section has the desired orientation.

2.2: Beam elements
s
S11
S 10
S5 S8
S7
l
S2 S9
S1 GB
S4 l S12
GA S3
S6 t
Z
X Y
S1 = r-direction force at node GA (axial force, positive in compression)

S2 = s-direction force at node GA (shear force)
S3 = t-direction force at node GA (shear force)
S4 = r-direction moment at node GA (torsion)
S5 = s-direction moment at node GA (bending moment)
S6 = t-direction moment at node GA (bending moment)
S7 = r-direction force at node GB (axial force, positive in tension)

S8 = s-direction force at node GB (shear force)
S9 = t-direction force at node GB (shear force)
S10 = r-direction moment at node GB (torsion)
S11 = s-direction moment at node GB (bending moment)
S12 = t-direction moment at node GB (bending moment)
Fig. 2.2-3: Element end forces/moments
 The basic beam geometric properties are:
A : Cross-sectional area
L : Beam length
Assh , Atsh , effective shear cross-section areas in the s and t

directions. These are calculated as Assh  K1 A , Atsh  K 2  A .

Chapter 2: Elements
I rr : Polar moment of inertia: I rr   s 2  t 2 dA

A
(do not confuse the polar moment of inertia I rr with the Saint-
Venant torsional constant J )
I ss : Inertia for bending about the s-axis:

I ss   t 2 dA
A
I tt : Inertia for bending about the t-axis:

I tt   s 2 dA
A
I st : Product of inertia:
I st   stdA
A
If s-t axes are coincident with the principal axes of the section then
I st  0 .
J : Saint-Venant torsional constant
Cross-sections
 The following PBARL and PBEAML cross-sections are

supported by Advanced Nonlinear Solution: BAR, BOX, CHAN,
CHAN1, CHAN2, H, I, I1, ROD, T, T1, T2, TUBE
 Table 2.2-1 shows the beam capabilities available for the

available cross sections.

2.2: Beam elements
Table 2.2-1: Table of beam capabilities for the available cross-sections
Formulation General cross-section Cross-sections BOX, Cross-sections

and material entered using PBAR, CHAN, CHAN1, BAR, ROD,
model PBEAM or CHAN2, H, I, I1, T, TUBE, entered
PBCOMP T1, T2, entered using using PBARL or
PBARL or PBEAML PBEAML
Linear   
elastic
Large   
displacement
elastic
MNO plastic --- --- 
Large --- --- 
displacement
plastic
2.2.2 Beam element formulations
2.2.2.1 Kinematics
 The element is formulated based on the Bernoulli-Euler beam

theory, corrected for shear deformation effects if requested.
 The beam displacements in the beam local coordinate system

are u (axial r-direction displacement), v (transverse s-direction
displacement) and w (transverse t-direction displacement).
 It is assumed that the cross-section rotates rigidly. Therefore the

displacements at an arbitrary point on the beam cross-section can
be written in terms of the displacements and rotations of the beam
neutral axis:
u  un   t s   s t
v  vn   r t
w  wn   r s

Chapter 2: Elements
In addition, the transverse rotations of the beam neutral axis are

assumed to be equal to the slopes of the transverse displacements:
dwn
s  
dr
dv
t  n
dr
The neutral axis displacements and rotations are interpolated from

the nodal displacements and rotations using
un  L1u 1  L2u 2
vn  H1v 1  H 2t1  H 3v 2  H 4t 2
wn  H1w1  H 2 s1  H 3 w2  H 4 s2
 r  L1 r1  L2 r2
dH1 1 dH 2 1 dH 3 2 dH 4 2
s   w  s  w  s
dr dr dr dr
dH dH dH dH
t  1 v 1  2 t1  3 v 2  4 t 2
dr dr dr dr
where the nodal displacements and rotations are

 u , v , w ,
1 1 1
r
1
,  s1 ,t1  for node GA and  u 2 , v 2 , w2 , r2 , s2 ,t 2 
for node GB, and in which
r r
L1  1  , L2 
L L
r2 r3 r2 r3
H1  1  3 2  2 3 , H 2  r  2  2 ,
L L L L
2 3 2 3
r r r r
H3  3 2  2 3 , H4    2
L L L L
are the linear and cubic interpolation functions (Hermitian

displacement functions). It is seen that the transverse neutral axis
displacements vn and wn are cubic and that the axial neutral axis

2.2: Beam elements
displacement un and torsional rotation  r are linear.
 In all cases, the centroid of the cross-section is located at the

origin of the cross-section axes (s,t), and, in addition, the principal
axes of the cross-section are aligned with the cross-section axes
(s,t). Thus I st  0 .
 Note that axial forces applied at nodes GA or GB are assumed

to be acting along the beam’s centroid and hence cause no bending.
Also shear forces applied at nodes GA or GB are assumed to be
acting through the beam’s shear center and hence cause no
twisting.
 The kinematics given above do not account for the torsional

warping of non-circular cross-sections, or for shear deformations.
For the linear elastic material model, the torsional warping is
accounted for by choice of the torsional constant J, and shear
deformations are accounted for by choice of the shear areas. For the
elasto-plastic material model with BAR cross-section, the torsional
warping is approximated as described in Section 2.2.2.7.

 The stiffness matrix, mass matrix and force vector are
formulated in the local degrees of freedom (in the r, s, t axes).
These matrices and vectors are then transformed to the nodal
degrees of freedom (either global or skew) and assembled into the
global system matrices.
2.2.2.2 Linear formulation
 It is assumed that the displacements, rotations, and strains are

infinitesimally small, and the elastic-isotropic material is used.
2.2.2.3 Materially-nonlinear-only formulation
 It is assumed that the displacements, rotations, and strains are

infinitesimally small. Either the elastic-isotropic material is used
(for example, in conjunction with element birth-death) or the
plastic material model is used.

Chapter 2: Elements
2.2.2.4 Large displacement formulation
 It is assumed that large displacements/rotations can occur, but

only small strains. Any beam material model, linear or nonlinear,
can be used. The input of beam cross-sections and material data is
exactly the same as when using the linear or materially-nonlinear-
only beam elements.

 If a stressed large displacement beam element is subjected to a
rigid body rotation, the stresses/forces (expressed in the local
coordinate system) do not change during the rigid body rotation.
 The geometry of the large displacement beam element is shown

in Fig 2.2-4. The shape of the beam neutral axis is completely
specified by the positions of the end-nodes, and by the orientation
of the end-node triads. Each end-node triad consists of three
orthogonal unit vectors Vr , Vs , Vt . Initially the triads are identical
to the element coordinate axes unit vectors, e.g, Vr is a unit vector
in the r direction, etc.
 During the deformations, the positions of the end-nodes are
updated, as usual, by the nodal displacements. The orientations of
the end-node triads are updated incrementally by the increments in
nodal rotations. The end-node triads can rotate independently of
each other, however it is assumed that the relative rotation of the
end-node triads remains small.
The shape of the beam neutral axis becomes a curved space
curve. The bending and torsion in the beam is obtained from the
relative orientations of the end-node triads.
 In the large displacement formulation, the coordinate system in

which the beam local displacements and rotations are measured is
updated during the solution. In the current version of Advanced
Nonlinear Solution, this coordinate system is taken from the
orientations of the end-node triads. In Advanced Nonlinear
Solution of NXN 8.5, this coordinate system is taken from the
coordinates of the end-nodes. Hence the solution output in the
current version of Advanced Nonlinear Solution will in general be
different than the solution output in Advanced Nonlinear Solution
of NXN 8.5.

2.2: Beam elements
z
Vs2
Vr2
x y
GB Vt2
Vs1
Vr1
GA Vt1
a) Before deformation
Vs1
Vr1 Vs2
GA
Vt1 Vr2
GB Vt2
Beam neutral axis

b) After deformation
Fig. 2.2-4: Geometry of large displacement beam element
Fig 2.2-5 shows an example for in-plane bending. A single

beam element is clamped at node GA and subjected to a prescribed
rotation  at node GB. The coordinate system x , y used to
measure beam local displacements and rotations is located halfway
between the nodes, and is rotated an angle  / 2 with respect to the
global coordinate system x, y . Notice that the x direction does not
coincide with the line between the end-nodes.
The rotation at local node GB with respect to the local
coordinate system is  / 2 , and the rotation at local node GA with
respect to the local coordinate system is  / 2 . Also, the
transverse displacement at local node GB with respect to the local
coordinate system is v and the transverse displacement at local
node GA with respect to the local coordinate system is v .

Chapter 2: Elements
GB
y
y x
v
Beam neutral axis

GA x
a) Viewed in global coordinate system
y
v x
-
-v
b) Viewed in local coordinate system
Fig. 2.2-5: Coordinate system update for large displacement beam

element, in-plane bending example
The advantage of using a coordinate system defined from the

orientations of the end-node triads is that this coordinate system is
uniquely defined even for general 3-D deformations including
torsion. A coordinate system defined from the end-nodes
coordinates is not uniquely defined for 3-D deformations including
torsion.
2.2.2.5 Mass matrices
 The beam element can be used with a lumped or a consistent

mass matrix, except for explicit dynamic analysis (Solution 701)
which always uses a lumped mass matrix.

2.2: Beam elements
 The consistent mass matrix of the beam element is evaluated in

closed form, and does not include the effect of shear deformations.
The matrix is defined in the local coordinate system using
1 1 
3 0 0 0 0 0 0 0 0 0 0 
6
 
 13 6 I tt 11L I 9 6I 13L I tt 
 0 0 0  tt 0  tt 0 0 0  
 35 5 AL2 210 10 AL 70 5 AL2 420 10 AL 
 
 13 6 I ss 11L I 9 6I 13L I 
 0   ss 0 0 0  ss 0  ss 0
 35 5 AL2 210 10 AL 70 5 AL2 420 10 AL 
 I rr I rr 
 0 0 0 0 0 0 0 
 3A 6A 
 2
L 2I 13L I L 2
I 
  ss 0 0 0   ss 0   ss 0 
 105 15 A 420 10 AL 140 30 A 
 L2 2I 13L I L2 I tt 
  tt 0  tt 0 0 0   
105 15 A 420 10 AL 140 30 A 
M   AL 
 1 
 0 0 0 0 0 
 3 
 13 6 I tt 11L I tt 
  0 0 0   
 35 5 AL2 210 10 AL 
 13 6 I ss 11L I 
  0  ss 0 
35 5 AL2 210 10 AL
 
 I rr 
0 0
 3A 
 2 
 L 2I 
 ss 0
 105 15 A 
 
 L2 2I
symmetric  tt 
 105 15 A 
This matrix is derived in the following reference.
ref. J.S. Przemieniecki, Theory of Matrix Structural

Analysis, McGraw-Hill Book Co., 1968.
 The lumped mass for translational degrees of freedom is M / 2

where M is the total mass of the element.
 The rotational lumped mass for static analysis is 0.
The rotational lumped mass for implicit dynamic analysis

2 M I rr
(Solution 601) is M rr    . This lumped mass is applied
3 2 A
to all rotational degrees of freedom in order to obtain a diagonal
mass matrix in any coordinate system.

Chapter 2: Elements
The rotational lumped mass for explicit dynamic analysis

M Im
(Solution 701) is M rr  3   where I m  max  I ss , I tt  is
2 A
the maximum bending moment of inertia of the beam. This lumped
mass is applied to all rotational degrees of freedom. Note that this
scaling of rotational masses ensures that the rotational degrees of
freedom do not affect the critical stable time step of the element.
2.2.2.6 Elastic beam element
 The beam element stiffness matrix is evaluated in closed form.

The stiffness matrix used is discussed in detail in the following
reference:
ref. J.S. Przemieniecki, Theory of Matrix Structural

Analysis, McGraw-Hill Book Co., 1968.
The stiffness matrix, in the local coordinate system, is
 AE 
 L 0 0 0 0 0 k11 0 0 0 0 0 
 
 12 EI tt 6 EI tt 
0 0 0 0  k22 0 0 0 k26
 L3 (1  2 ) L2 (1  2 ) 
 
 12 EI ss 6 EI ss 
 0 0 0 0  k33 0 k35 0 
L3 (1  3 ) L2 (1  3 )
 
 GJ 
 0 0 0 0 0  k44 0 0 
L
 
 EI ss (4  3 ) EI ss (2  3 ) 
K 0 0 0  k35 0 0
L(1  3 ) L(1  3 ) 
 
 EI tt (4  2 ) EI tt (2  2 ) 
 0  k26 0 0 0
L(1  2 ) L(1  2 ) 
 
 k11 0 0 0 0 0 
 k22 0 0 0 k26 
 
 k33 0  k35 0 
 
 k44 0 0 
 k55 0 
 
 symmetric k66 
in which E = Young's modulus,  = Poisson's ratio, G = shear

E
modulus = . Also
2(1   )

2.2: Beam elements
24(1   )
2  I tt , Assh  0
L2 Assh
 0, Assh  0
24(1   )
3  2 sh
I ss , Atsh  0
L At
 0, Atsh  0
are factors that are used to approximate the effects of shear

deformations.
 Note that torsional warping effects are only taken into account
by the selection of the torsional constant J.
 The coefficient of thermal expansion can be specified as a

material property. The coefficient of thermal expansion is constant
(independent of the temperature). In addition, the beam temperature
is taken as the average of the temperatures of the beam end-nodes.
Temperature gradients at beam nodes are ignored.
 Stress and strain output is not supported in elastic beam

elements.
.
2.2.2.7 Elastic-plastic beam element
 The element is used with the plastic-cyclic material model

described in Section 3.4.2. The element can also be used with a
plastic-bilinear material with isotropic hardening, but then the
program automatically converts the material input into the
equivalent plastic-cyclic material input.
The stress-strain law used incorporates the assumptions that the
stresses  ss ,  tt ,  st are zero.
 The beam element, using an elastic-plastic material model, can

only be employed for the BAR, ROD and TUBE cross-sections.
 The material model can be used either with the materially-

nonlinear-only formulation or with the large displacement

Chapter 2: Elements
formulation (in which case the displacements/rotations can be

large). In all cases, the strains are assumed to be small.
 Shear deformation effects are not included.
 For beams with a BAR cross-section, warping effects are

included by modifying the longitudinal displacement as follows:
u  ...  st1   s 3t  st 3   2
where 1 ,  2 are additional degrees of freedom. The additional

degrees of freedom are condensed out at the element level.
This procedure is described in the following reference:
ref. K.J. Bathe and A. Chaudhary, "On the Displacement

Formulation of Torsion of Shafts with Rectangular
Cross-Sections", Int. Num. Meth. in Eng., Vol. 18, pp.
1565-1568, 1982.
ref. KJB  All element matrices in elasto-plastic analysis are calculated

Section 6.6.3 using numerical integration. The default integration orders are
given in Tables 2.2-2 and 2.2-3. The locations and the labeling of
the integration points are given in Fig. 2.2-6 and 2.2-7. For the
ROD and TUBE sections, polar coordinates are used:
t
R  s 2  t 2 ,   tan 1 .
s
Table 2.2-2: Integration orders in elasto-plastic beam analysis,

BAR section
Coordinate Integration order1

r 5
s,t 7
1) Newton-Cotes integration is used in all coordinate directions

2.2: Beam elements
Table 2.2-3: Integration orders in elasto-plastic beam analysis,

ROD and TUBE sections
Coordinate Integration Integration order

scheme
r Newton-Cotes 5
radius R Newton-Cotes 3
tangential angle Composite 8
 trapezoidal rule
 Note that the stiffness matrix is identical to the one used in

linear elastic analysis if
 The cross-section of the beam is circular (or rectangular)

with b  a or b  a , (because the exact warping functions are
employed for b  a and b  a , and the appropriate torsional
rigidity is used).
 The numerical integration is of high enough order
 The material is (still) elastic
 Stress and strain output is not supported in elastic-plastic beam

elements.

Chapter 2: Elements
s Integration points
equally spaced
1 2 3 ...
Ä Ä Ä Ä
r
r=0 r=L
t
a) Integration point locations in r-direction
Integration points
t equally spaced
HEIGHT s
Ä
Ä
Ä
1 2 3
WIDTH
b) Integration point locations in s-direction
t Integration points
Ä
equally spaced
s
ÄÄÄ
3
2
1
c) Integration point locations in t-direction
Fig. 2.2-6: Integration point locations in elasto-

plastic beam analysis, BAR cross-section

2.2: Beam elements
Integration points
s
equally spaced
1 2 3 ...
Ä Ä Ä Ä
r
r=0 r=L
t
a) Integration point locations in r-direction
t
Integration points
equally spaced
3 Ä
...
2Ä
R 1Ä
Ä s
b) Integration point locations in radial direction
t
Integration points
equally spaced
... Ä 3 2
R Ä
1
Ä s
Ä
c) Integration point locations in tangential direction
Fig. 2.2-7: Integration point locations in elasto-

plastic beam analysis, ROD and TUBE sections

Chapter 2: Elements
 The beam element has the same heat transfer capabilities of the
rod element. See Section 2.1.5 for details.
2.2.4 Pin flag option
To model beam internal hinges, a moment and force release option

can be used (see Fig. 2.2-8). Twisting moments and axial forces
can also be released.
Shear force
Internal hinge released
l l
r r r r
Element 1 Element 2 Element 1 Element 2
Specify 456 in the PB field in Specify 23 in the PB field in

CBAR for element 1, or CBAR for element 1, or
specify 456 in the PA field in specify 23 in the PA field in
CBAR for element 2 CBAR for element 2
(a) Moment to be released at (b) Shear force to be released

internal hinge
Fig. 2.2-8: Use of pin flag options
 A brief description of the theory of pin flags follows:
For static linear elastic analysis without warping degrees of

freedom, the stiffness matrix and internal force vector for a beam
element can be written as
Ku  F
In this expression, u contains the displacements and rotations of

the beam nodes, in the beam local coordinate system:
uT  u1r u1s ut1  r1  s1 t1 ur2 us2 ut2  r2  s2 t2  ,
F contains the forces and moments of the beam nodes,

2.2: Beam elements
FT   S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 S11 S12 
and K is the stiffness matrix. These expressions use the same

notations as Figs. 2.2-1 and 2.2-2.
Now suppose that one or more of the local displacements or
rotations is to be released. “Releasing” means that the
corresponding force is set equal to zero. For example, if the axial
displacement at local node GB is to be released, then force S7 is
set to zero. The corresponding pin flag is PB=1 in the CBAR bulk
data entry.
In order to release the selected local displacements or rotations,
the following procedure is used:
1) u is rearranged and partitioned: uT  uTA uTB  , so that u A

contains all of the local displacements / rotations to be retained, and
u B contains all of the local displacements / rotations to be released.
For example, if the axial displacement at local node 2 is to be
released, then
uTA  u1r u1s ut1  r1  s1 t1 us2 ut2  r2  s2 t2 

uTB  ur2 
2) F and K are similarly rearranged and partitioned. The

resulting system of equations is
 K AA K AB  u A  FA 
 KT 
 AB K BB  u B  FB 
3) The pin flag condition is now expressed as FB  0 , and the

resulting system of equations becomes
 K AA K AB  u A  FA 
 T 
 K AB K BB  u B   0 

Chapter 2: Elements
4) This system of equations is satisfied by choosing
K AA  K AB K BB1 K TAB  u A  FA
u B  K BB1 K TAB u A
Therefore the stiffness matrix that is assembled into the global

system of equations is K AA  K AB K BB1 K TAB .
 Any external forces acting onto the released degrees of freedom

are ignored.
 The pin flag procedure is also implemented for nonlinear

analysis, both for materially nonlinear elements and also for
geometrically nonlinear elements.
 The pin flag procedure is available for static, implicit dynamic

and explicit dynamic analysis.
 Pin flags only affect the stiffness matrix and force vector, not
the mass matrix. Therefore inertial forces and moments (forces and
moments due to mass matrix effects) are not released in the above
procedure.
 If pin flags are used in dynamic analysis, the elements in which

pin flags are specified should be very short, in order to minimize
the inertial forces and moments.
 Pin flags are applied to the element local nodes (not to the
global nodes). Therefore, to model the hinge shown in Fig.
2.2-8(a), pin flags that release the moments can be applied to local
node GB of element 1, or to local node GA of element 2 (but not to
both local nodes).
 When the beam elements are geometrically nonlinear, the global

directions corresponding to the released degrees of freedom change
as the model deforms. By default, this effect is fully included in the
pin flag calculations. However, this effect can occasionally slow
down convergence.
The pin flags are evaluated in a local coordinate system
corresponding to the configuration of the beam in the previous

2.2: Beam elements
equilibrium iteration. Therefore, at convergence, the pin flags are

evaluated in (almost) the current configuration of the beam. (At
convergence, the difference between the current equilibrium
iteration and the previous equilibrium iteration is very small.)
In implicit static and implicit dynamic analysis, when the
elements are geometrically nonlinear, the program performs at least
two equilibrium iterations in each time step.
 In explicit dynamic analysis, the use of pin flags will cause the
program to run considerably more slowly than if this option were
not used.
2.2.5 Beam element modeling hints
 For modeling tapered beams or curved beams, the user needs to

divide such members into several elements. For the tapered beams
the user needs to divide the beam into several elements and use an
appropriate constant cross-section for each of these elements.
 In order to model the bending due to an off-centroidal axial

force or a shear force applied away from the shear center, the
resulting moments can be applied directly or the forces can be
applied at an offset location using rigid elements.
 Off-centered beam elements can be modeled using rigid

elements (see Fig. 2.2-9 and Section 2.7-1).
Physical problem: Finite element model:

Rigid panel
I-beam
I-beam Beam elements
Beam element
Rigid elements
Hollow square section
Fig. 2.2-9: Use of rigid elements for modeling off-centered beams

Chapter 2: Elements
2.3 Shell elements

 Shell elements in Advanced Nonlinear Solution are generated
when a PSHELL, PSHL3D, PCOMP or PCOMPG property ID
entry is referenced by one of the following Nastran shell entries:
CQUAD4, CTRIA3, CQUAD8, CTRIA6, CQUADR, or CTRIAR.
The elements are shown in Fig. 2.3-1.
l
l
l
l
l
l
(a) 3-node element (b) 4-node element
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
(c) 6-node element (d) 8-node element
Fig. 2.3-1: Shell elements in Advanced

Nonlinear Solution
 The PSHELL entry results in a single-layered shell, while

PCOMP and PCOMPG produces a composite shell.
 The PSHL3D entry results in a single-layered 3D-shell. 3D-

shell elements are described in Section 2.3.9.
 Shell elements are classified based on the number of nodes in

the element. Table 2.3-1 shows the correspondence between the
different shell elements and the NX element connectivity entries.
 Solution 701 only supports 3-node and 4-node single-layered

shell elements.

2.3: Shell elements
Table 2.3-1: Correspondence between shell elements and NX

element connectivity
Shell element NX element connectivity entry

3-node CTRIA3, CTRIAR
4-node CQUAD4, CQUADR
1
6-node CTRIA6
8-node1 CQUAD8
1
9-node CQUAD82
Notes:
1. Only for Solution 601
2. With ELCV = 1 in NXSTRAT entry

 The extra middle node in the 9-node shell element is
automatically added by the program when ELCV is set to 1 in the
NXSTRAT entry. This extra node improves the performance of the
shell element. The boundary conditions at the added node are
predicted from the neighboring nodes.
 Incompatible modes (bubble functions) can be used with 4-node

shell elements. Additional displacement degrees of freedom are
introduced which are not associated with nodes; therefore the
condition of displacement compatibility between adjacent elements
is not satisfied in general. The addition of the incompatible modes
(bubble functions) increases the flexibility of the element,
especially in bending situations. For theoretical considerations, see
reference KJB, Section 4.4.1. Note that these incompatible-mode
elements are formulated to pass the patch test. Also note that
element distortions deteriorate the element performance when
incompatible modes are used.
The incompatible modes feature can only be used with 4-node
single layer shell elements. The feature is available in linear and
nonlinear analysis.
The incompatible modes feature is set through ICMODE in the
NXSTRAT entry.
 Table 2.3-2 lists the features and capabilities available for the
shell element types mentioned above.

Chapter 2: Elements
Table 2.3-2: Features available for shell elements
Large Large Large

Shell Bubble 3D- Soln
displacement/ strain ULJ strain ULH
element functions shell 701
small strain formulation formulation
3-node     
4-node      
6-node   
8-node 
9-node   
2.3.1 Basic assumptions in element formulation
 The basic equations used in the formulation of the shell

elements in Advanced Nonlinear Solution are given in ref. KJB.
These elements are based on the Mixed Interpolation of Tensorial
Components (MITC). Tying points are used to interpolate the
transverse shear strain and the membrane strains if necessary.
These elements show excellent performance.
 The shell element formulation treats the shell as a three-

dimensional continuum with the following two assumptions used in
the Timoshenko beam theory and the Reissner-Mindlin plate
theory:
Assumption 1: Material particles that originally lie on a

straight line "normal" to the midsurface of the structure remain
on that straight line during deformation.
Assumption 2: The stress in the direction normal to the

midsurface of the structure is zero.
For the Timoshenko beam theory, the structure is the beam, and
for the Reissner/Mindlin plate theory, the structure is the plate
under consideration. In shell analysis, these assumptions
correspond to a very general shell theory. See the reference below
for more details:

2.3: Shell elements
ref. D. Chapelle and K.J. Bathe, The Finite Element Analysis

of Shells — Fundamentals, 2nd ed, Springer, 2011.
 In the calculations of the shell element matrices the following

geometric quantities are used:
 The coordinates of the node k that lies on the shell element

midsurface at t xik , i  1, 2,3 (see Fig. 2.3-2); (the left
superscript denotes the configuration at time t)
t
G4
G8 s
G7
G3
G1
G6 r
G5
Midsurface nodes G2
Fig. 2.3-2: Some conventions for the shell element;

local node numbering; local element coordinate system
 The director vectors t Vnk pointing in the direction

"normal" to the shell midsurface
ref. KJB  The shell thickness, ak , at the nodal points measured in the
Fig. 5.33
page 437
direction of the director vectors t Vnk (see Fig. 2.3-3).

Chapter 2: Elements
No stiffness
for rotation k
Vn
about Vnk
wk
l Z
l
k
l V2
vk Y
l bk
X
uk
ak
k
V1
Fig. 2.3-3: Shell degrees of freedom at node k
 Based on these quantities the geometry of the shell is

interpolated as follows:
q
t q
xi   hk x   ak hk Vnik
t
t t k
i (i  1, 2,3)
k 1 2 k 1
where q is the number of element nodes, t Vnik (i  1, 2, 3) are the

components of the shell director vector t Vnk and hk (r , s ) are the
2-D interpolation functions.
 At the element level the shell has 5 independent degrees of

freedom per node: 3 displacements about the displacement
coordinate system resulting from the displacement of the shell
midsurface and 2 rotations resulting from the motion of the shell
direction vector Vnk :
q
t q
t
ui   hk t uik   ak hk  t Vnik 0 Vnik 
k 1 2 k 1

2.3: Shell elements
The motion of the director vector at node k is described using 2

rotational degrees of freedom about V1k and V2k which are 2 axes
perpendicular to the shell director Vnk as shown in Fig. 2.3-3.
Y  Vnk
V1k 
Y  Vnk
2
V2k  Vnk  V1k
For the special case when the Vnk vector is parallel to the Y axis,
the program uses the following conventions:
V1k  Z V2k  X when Vnk   Y
and
V1k  Z V2k  X when Vnk  Y
The two rotational degrees of freedom named  k and  k are about

axes V1k and V2k respectively.
When using the large displacement formulations, the definitions
of V1k and V2k are only used at time = 0 (in the initial
configuration) after which the vectors t Vnk and t V1k are updated
using incremental rotations at the nodal points, and t V2k is
calculated by the cross-product t V2k  t Vnk  t V1k .
Note that a shell node may however be assigned 3 rotational
degrees of freedom. In this case, the element’s two rotational
degrees of freedom are transformed to the displacement coordinate
system before assembly.
 Assumption 1 on the kinematic behavior of the shell enters the

finite element solution in that the particles along the director vector
t
Vn (interpolated from the nodal point director vectors t Vnk )
remain on a straight line during deformation.

Chapter 2: Elements
Note that in the finite element solution, the vector t Vn is not

necessarily exactly normal to the shell midsurface. Fig. 2.3-4(a)
demonstrates this observation for a very simple case, considering
the shell initial configuration. Furthermore, even if t Vn is
originally normal to the shell midsurface, after deformations have
taken place this vector will in general not be exactly perpendicular
to the midsurface because of shear deformations (see Fig. 2.3-4(b)).
( L1 L+2L2 ) a
0 k+1
a Vn
0
0 k
Vn Vn
l l
k k+1
L1 L2
a) Due to initial geometry

0
Vn
Initial configuration
Angle = 90°
t
l
Vn
l
l
Z Final configuration
Angle ¹ 90°
Y
X
b) Due to displacements and deformations (with shear)
Fig. 2.3-4: Examples of director vectors not normal to the shell

midsurface

2.3: Shell elements
 The assumption 2 on the stress situation enters the finite

ref. KJB element solution in a manner that is dependent on the formulation
Section 5.4.2
page 440
employed:
All formulations except for the large displacement/large strain

shell element: The stress in the t-direction (i.e., in the direction of
t
Vn ) is imposed to be zero. This is achieved by using the stress-
strain relationship in the r , s , t coordinate system, shown in Fig.
2.3-5(a), with the condition that the stress in the direction t is zero.
t
s
t
s
s
r
r
r= s´t s= t ´ r
s´t 2 t´r 2 r
(a) Definition of the local Cartesian system at an

integration point in the shell
Fig. 2.3-5: Local coordinate systems in shell element
Large displacement/large strain shell element: The stress in the

tˆ -direction (not necessarily in the direction of t Vn ) is imposed to
be zero. This is achieved by using the stress-strain relationship in
the rˆ, sˆ, tˆ coordinate system, shown in Fig. 2.3-5(b), with the
condition that the stress in the direction tˆ is zero.

Chapter 2: Elements
t t
s
t
s
s
r
r
s ´ t , s=t´r
t= r´s , r=
r´s 2 s´t 2 r
b) Definition of the midsurface Cartesian system ( r, s, t )

at an integration point
Fig. 2.3-5: (continued)
 The interpolation of the geometry of the shell element is always

as described above, but for a specific solution time the current
coordinates of the midsurface nodal points are used, and the current
director vectors are employed. The midsurface nodal point
coordinates are updated by the translational displacements of the
nodes and the director vectors are updated using the rotations at the
nodes (rotation increments in large displacement analysis).
ref. KJB  The transverse shear deformations are assumed to be constant

pp. 399, 440 across the shell thickness.
 In large displacement analysis, the midsurface nodal point

coordinates are updated by adding the translational displacements
of the nodes, and the director vectors are updated using the
incremental rotations at the nodes by applying the large rotation
update transformation described in p. 580 of ref. KJB (Exercise
6.56).

2.3: Shell elements
compatible with shell elements.
 The shell element can be used with
small displacement/small strain kinematics,

large displacement/small strain kinematics, or
large displacement/large strain kinematics.
ref. KJB In the small displacement/small strain case, the displacements

Section 6.6 and rotations are assumed to be infinitesimally small. Using a
linear material results in a linear element formulation, and using a
nonlinear material results in a materially-nonlinear-only
formulation.
In the large displacement/small strain case, the displacements
and rotations can be large, but the strains are assumed to be small.
In this case, a TL formulation is used.
The large displacement/large strain formulation for shells can be
either a ULJ (updated Lagrangian Jaumann) formulation or a ULH
formulation (updated Lagrangian Hencky) depending on the
ULFORM parameter in the NXSTRAT entry. In the ULJ
formulation, the total strains can be large, but the incremental strain
for each time step should be small (< 1%). The ULH formulation
requires more computations, however, it has no such restriction on
the size of the incremental strains.
The large displacement/large strain kinematics can be only used
with single layer shell elements with a plastic bilinear/multilinear
or plastic-cyclic material. See Table 2.3-2 for a list of the supported
shell elements.
2.3.3 Shell nodal point degrees of freedom
 Shell nodes can have either 2 or 3 rotational degrees of freedom

which results in nodes having either 5 or 6 degrees of freedom.

 The criterion for determining whether a shell node is assigned 5
or 6 degrees of freedom is as follows. 5 degrees of freedom are
initially assigned to all shell midsurface nodes. The following cases
change the node to 6 degrees of freedom:

Chapter 2: Elements
Geometry. Shell elements at that node intersecting at an

angle greater than a specified tolerance (SDOFANG parameter
in the NXSTRAT entry).
Other elements. If the node also has other elements with

rotational degrees of freedom, i.e., beam elements, rotational
springs, rotational masses or rotational dampers.

Rotational loads, constraints or boundary conditions.
This includes the following cases:
- applied moment at the node
- rotational fixed boundary condition at the node
- rigid link connected to the node
- constraint equation involving constrained rotations
connected to the node
- enforced rotations at the node
 Shell nodes with 6 degrees of freedom may be a potential source

for singularity. In this case, very weak springs are automatically
added to prevent the singularity. The cases in which this happens
are discussed later in this section.
 Fig. 2.3-6 shows examples of 5 and 6 degree of freedom shell
nodes.
 Note that for both 5 and 6 degree of freedom shell nodes, the
translations uk, vk, wk are referred to the chosen displacement
coordinate system.
5 degrees of freedom node: A node "k" that is assigned 5 degrees

of freedom incorporates the following assumption:
 Only one director vector (denoted at time = 0 as 0 Vnk ) is

associated with the node. The program calculates the director
vector by taking the average of all normal vectors (one normal
vector is generated per shell element attached to node k) at the
node. This is illustrated in Fig. 2.3-7.

2.3: Shell elements
4-node shell elements

No X,Y,Z translations
No Y, Z rotations
FX
Concentrated
l forces
l l 9 FZ
4 3 2 1
l
l l l
l l
5 MZ
6
l
MY
8l 7 l
l l l Concentrated
X moments
Rigid element (Node 3
beam elements as independent node)
Z
Y
Number Potential
Node
of DOFs singularity
1 6 Yes
2 5 No
3 6 Yes
4 6 No
5 5 No
6 6 No
7 6 No
8 6 Yes
9 5 No
Fig. 2.3-6: Examples of shell nodes with 5 or 6 degrees of freedom
If two (or more) elements attached to the node have oppositely

directed normals, the program reverses the oppositely directed
normals, so that all normals attached to the node have (nearly) the
same direction.

Chapter 2: Elements
0 k
Vn is the average of all element director vectors
0 k
Vn
l k
element 1 l
element 2
0 k 0 k
Vn Vn
k k
l l
l
element 1 l
element 2
Fig. 2.3-7: 5 degree of freedom shell node with unique vector at node k
6 degrees of freedom node: A node "k" that is assigned 6 degrees

of freedom incorporates the following assumption:
 The program generates as many normal vectors at node k as

there are shell elements attached to the node. Hence each individual
shell element establishes at node k a vector normal to its
midsurface. This is illustrated in Fig. 2.3-8. The components of the
shell element matrices corresponding to the rotational degrees of
freedom at this node are first formulated in the local midsurface
system defined by the normal vector and then rotated to the
displacement coordinate system.

2.3: Shell elements
director for element 1

director for element 2
l k
element 1 l
element 2
0 k 0 k
Vn Vn
element 1 element 2
k k
l l
l
element 1 l
element 2
Fig. 2.3-8: 6 degree of freedom shell node with separate director

vectors at node k (each vector is used as a director vector for the
respective element)
 The three rotational degrees of freedom at node k referred to the

displacement coordinate system can be free or constrained.
Singularity at 6 degree of freedom shell nodes
 When a shell node is forced to have 6 degrees of freedom due to

the reasons explained above, there may be a singularity at one of
the rotational degrees of freedom. In this case, a weak rotational
spring is added to the 3 rotational degrees of freedom. This is done
automatically by Advanced Nonlinear Solution and usually does
not require user intervention. The stiffness of the spring is set to be
a small fraction of the average rotational stiffnesses at the shell
node. This fraction is can be changed via the DRILLKF parameter
in NXSTRAT.
 Not all the cases that lead to a shell node possessing 6 degrees
of freedom (listed at the beginning of this section) may introduce a
singularity at the node.

Chapter 2: Elements
Geometry. No potential singularity exists in this case, since

the shell is curved.
Other elements. Beam-stiffened shells will have a

singularity at the shell nodes only if the beam is perpendicular
to a flat shell surface. Otherwise, a singularity can still exist in
the model if the beam is not properly restrained (see Fig. 2.3-9
(a)). The same applies to rotational springs, masses and
dampers.
Rotational loads, constraints or boundary conditions. All

the items listed earlier for this feature result in a potential
singularity (see Fig. 2.3-9 (b)) except when all rotational
degrees of freedom at the node are fixed.
 If multiple factors lead to the presence of 6 degrees of freedom

at a shell node, no singularity is present if any of the factors
eliminates the singularity. For example, if a shell node has an
applied moment and is attached to non-perpendicular beam
elements there is no singularity.
 Fig. 2.3-6 shows examples where shell singularity may or may

not occur.
 The singularity that may result from beams attached to shells

requires some clarification. If a beam connects two shell structures
as shown in Fig. 2.3-10, and it is perpendicular to both shells, then
the beam is free to rotate about this perpendicular direction (the
z-direction in this example). If the beam intersects the shells at an
angle, this singularity is not present.

2.3: Shell elements
Zero pivot in
stiffness matrix.
F t
Vn
a) Structural element/rigid link attached to node,

structural element/rigid link is unsupported.
Moment causes
z infinite rotation.
y Mz
x t
Vn
Reactions at fixities
are zero.
b) Moment applied in shell normal direction.
Fig. 2.3-9: Flat shell with 6 DOFs at a node with

singularity
 An alternative to using the drilling stiffness option is to connect

the 6 DOF nodes on flat shells to neighboring shell nodes using
soft beam elements (so-called “weld elements”). This idea is shown
in Fig 2.3-11. Then moments applied into the t Vn direction will be
taken by the weld elements, and these moments will cause
equilibrating reactions at the fixities. The weld elements also
provide stiffness in the t Vn direction, so that there will be no zero
pivots.

Chapter 2: Elements
Perpendicular
to both shells
Beams
Not perpendicular
to shells
z
Shell elements y
x
Fig. 2.3-10: Beams intersecting shell elements
Soft beam element.

Mz t
Vn
Reactions at fixities
are nonzero.
Fig. 2.3-11: Soft beam element takes applied

moment
2.3.4 Composite shell elements (Solution 601 only)
 Composite shell elements are generated when a PCOMP or

PCOMPG property ID is referenced by one of the following
Nastran shell connectivity entries: CQUAD4, CTRIA3, CQUAD8,
CTRIA6.



2.3: Shell elements
 The composite shell elements are kinematically formulated in

the same way as the single layer shell elements, but
 An arbitrary number of layers can be used to make up the

total thickness of the shell, and each layer can be assigned a
different thickness.
 Each layer can be assigned one of the different material
models available. The element is nonlinear if any of the
material models is nonlinear, or if the large displacement
formulation is used.
 Large displacement/large strain kinematics are not
supported for composite shell elements.
 The conventions for defining the director vectors, the local axes
V1 and V2, and the 5/6 degree of freedom selection are all the same
as those for the single layer shell.

 In order to take into account the change of material properties
from one layer to another, numerical integration of the mass and
stiffness matrices is performed layer by layer using reduced natural
coordinates through the thickness of the element (see Figs. 2.3-12
and 2.3-13). The relation between the element natural coordinate t
and the reduced natural coordinate tn of layer n is:
1  n i n n 
t  1   2       1  t   (2.3-1)
a   i 1  
with
t = element natural coordinate through the thickness

tn = layer n natural coordinate through the thickness
i = thickness of layer i
a = total element thickness
a and i are functions of r and s.

Chapter 2: Elements
l
1
Y t l
8
5l
X s l 4
r l
a2 2l 7
V3n
l
6
3l V 32
b3
V31 a3
Fig. 2.3-12: 8-node composite shell element
ln E nxx n
G xz nnxy Layer n
a
Layer 1
Fig. 2.3-13: Multilayered shell
The geometry of layer n is given by:
N N
 n  t
t
xi   hk t xik    mkn  t n k  hk Vnik (2.3-2)
k 1 k 1  2

2.3: Shell elements
with
t
xi = coordinate of a point inside layer n in direction i
N = number of nodes
hk = interpolation functions
t
xik = Cartesian coordinates of node k
t
Vnik = component of normal vector t Vnk at node k
ak = total element thickness at node k
 kj = thickness of layer j at node k
mkn = distance between element midsurface and midsurface of
layer n at node k
In the above formula, mkn is given by
ak n
n
mkn      kj  k (2.3-3)
2 j 1 2
 Gauss numerical integration is used in the in-plane directions of

the shell. For the 4-node shell element, 22 integration is used. For
the 8-node and 9-node elements, 33 point integration is used. The
3-node triangular shell element uses 4-point Gauss integration in
the in-plane directions, and the 6-node triangular shell element uses
7-point Gauss integration.
 Numerical integration through the shell thickness is as follows:

ref. KJB
Section 6.8.4
 For elastic materials, 2-point Gaussian integration is always
used.
 For elasto-plastic materials and the nonlinear elastic
material, 5-point Newton-Cotes integration is the default.
Although using 5-point integration is computationally more
expensive, it gives much more accurate results for elasto-
plastic shells.

Chapter 2: Elements
 For composite shells with elasto-plastic materials, 3-point

Newton-Cotes integration is the default.
The order of through-thickness integration can be modified via the

TINT parameter in the NXSTRAT entry. If TINT is specified, it
will be applied to both single-layered and composite elasto-plastic
shells.
 The same integration order is used for both Solution 601 and
701.
2.3.6 Mass matrices
 In Solution 601 shell elements can be employed with a lumped

or a consistent mass matrix. Only a lumped mass matrix is allowed
in Solution 701.
 The consistent mass matrix is calculated using the isoparametric

formulation with the shell element interpolation functions.
 The lumped mass for translational degrees of freedom of

midsurface nodes is M / n where M is the total element mass and
n is the number of nodes. No special distributory concepts are
employed to distinguish between corner and midside nodes, or to
account for element distortion.
The rotational lumped mass for implicit analysis (Solution 601)
M 1 2
is   t  , where tav is the average shell thickness. The same
n 12 av
rotational mass matrix is assumed for 5- and 6-degree of freedom
nodes, and is applied to all rotational degrees of freedom.
The rotational lumped mass for explicit analysis (Solution 701)
M 1 2
is   t  A  , where tav is the average shell thickness and A is
n 12 av
the cross-sectional area. The rotational masses are scaled up to
ensures that the rotational degrees of freedom will not reduce the
critical time step for shell elements. The same rotational mass
matrix is assumed for 5- and 6-degree of freedom nodes and is
applied to all rotational degrees of freedom.

2.3: Shell elements
 Heat transfer capabilities are available for all supported shell

elements, including composite shells.
ref KJB
 The shell heat transfer capabilities are formulated by assuming
Section 5.4.2 that the temperature varies linearly through the shell thickness
direction. Two degrees of freedom are therefore assigned at each
shell node, one for the top shell surface and one for the bottom
shell surface.
 The element matrices are integrated numerically by Gauss

integration using the same integration order as the structural
matrices.
 In the calculation of the top and bottom shell surfaces, the

following geometric quantities are used:
 The coordinates of the nodes that lie on the shell element

midsurface.
 The director vectors Vn normal to the shell midsurface.
 The shell thicknesses a at the nodal points measured in the

direction of the vector Vnk (see Fig. 2.3-14)
 Fig. 2.3-14 shows a 4-node thermal shell element with the shell
midsurface nodes, the nodal director vectors and constructed top
and bottom nodes. The director vectors are automatically calculated
by the program, see Fig. 2.3-15.
 In the calculation of the shell element matrices, i.e.,

conductivity, heat capacity, and heat generation, the top and bottom
shell surfaces are used instead of the midsurface.
 The shell heat capacity matrix can be calculated based on a

lumped or a consistent formulation, similar to the mass matrix in
structural analysis.

Chapter 2: Elements
All director vectors must

t point towards the same
side of the shell element
l
l
Shell
l
midsurface
s
l Vkn
l node k
l
l
ak
Z
Thickness at a mid-
l
r surface node is
Y l measured along the
X l director vector Vnk
l Input midsurface nodes Generated top and bottom nodes
Fig. 2.3-14: Description of the thermal shell element
 Thermal loads and boundary conditions such as applied

temperatures, heat flux, convection and radiation can all be applied
to either the top or bottom shell surfaces.

2.3: Shell elements
Director vector is
average of all midsurface
normal vectors at the node
Normal vector to midsurface
Node k
Normal vector to
Element 1 l midsurface
Element 2
l
l
l
l
Thickness l
l
l
l
Thickness
l Input midsurface nodes l Generated top and bottom nodes
Vkn Vkn
Element 1
l l Element 2
Thickness input refer

to these directions
Fig. 2.3-15: Program-calculated director vector at thermal shell nodes
2.3.8 Selection of elements for analysis of thin and thick shells
 The most effective element for analysis of general shells is

usually the 4-node element. This element does not lock and has a
high predictive capability and hence can be used for effective
analysis of thin and thick shells.
 The phenomenon of an element being much too stiff is, in the

ref. KJB literature, referred to as element locking. In essence, the
pp. 403-408 phenomenon arises because the interpolation functions used for an
element are not “abundantly” able to represent zero (or very small)
shearing or membrane strains. If the element cannot represent zero
shearing strains, but the physical situation corresponds to zero (or

Chapter 2: Elements
very small) shearing strains, then the element becomes very stiff as
its thickness over length ratio decreases. The MITC elements are
implemented to overcome the locking problem. More details on the
interpolations used for the transverse and membrane terms are
provided in ref. KJB, pp. 403 – 406.
 The 8-node element is not as effective as the other MITC

elements, and its use is not recommended in general.
 In order to arrive at an appropriate element idealization of a thin

shell, it may be effective to consider the behavior of a single
element in modeling a typical part of the shell. As an example, if a
shell of thickness h and principal radii of curvatures R1 and R2 is
to be analyzed, a single element of this thickness and these radii
and supported as a cantilever could be subjected to different simple
stress states. The behavior of the single element when subjected to
the simple stress states (e.g., constant bending moments) tells what
size of element, and hence element idealization, can be used to
solve the actual shell problem.
 For cases where the ratio thickness/radius of curvature is large

(in the original configuration or in the deformed configuration in
large deformation analysis) it is best to use 5 dofs at each shell
node. Then all elements will represent the hyperbolic stress
distributions through the element thickness. If 6 dofs per node are
used, then only the higher order elements will represent the
hyperbolic stress distribution because the nodal director vectors are
constructed from the geometry of the elements.
 Geometrically nonlinear incompatible modes elements with

large aspect ratio should not be used, because spurious modes may
be present in the finite element solution.
2.3.9 3D-shell element
Overview
 One characteristic of the shell elements described earlier is that

the change in thickness of the element is not explicitly calculated
from the element degrees of freedom. This is because the zero
stress through the shell thickness assumption is used in the material

2.3: Shell elements
descriptions.
 However, in large strain analysis, the change in thickness can

become important. For example, during out-of-plane bending, the
material in compression thickens, and the material in tension thins
(for nonzero Poisson's ratio). Hence the midsurface is no longer
exactly halfway between the top and bottom surfaces. This effect
is shown in Fig 2.3-16.
 This change in thickness is naturally modeled when quadratic

3D solid elements are used, because the nodes on the top and
bottom surfaces can move relative to the nodes on the midsurface,
as shown in Fig 2.3-17. However 3D solid elements tend to lock
when they are very thin, so that they are unsuitable for bending
analysis of thin structures.
 In the 3D-shell element, the change in thickness of the element

is modeled using control vectors, as shown for the case of pure
bending in Fig 2.3-18. The motion of the control vectors is
controlled by element degrees of freedom at the shell midsurface,
as discussed in more detail below.
 Because the change in thickness is explicitly calculated from

element degrees of freedom, the assumption of zero stress through
the shell thickness is not used in the 3D-shell element.
 In addition, the 3D-shell elements use MITC tying rules to

relieve shear locking. Therefore these elements are suitable for
out-of-plane bending analysis of thin structures, even for large
bending strains.
 The 3D-shell elements can be used with 3 or 4 nodes (CTRIA3

or CQUAD4). The 4-node element is recommended for general
use.
 The 3D-shell elements can be used in static, implicit dynamic or

explicit dynamic analysis.

Chapter 2: Elements
Midsurface
a) Undeformed configuration
Material particles
initially on midsurface
b) Deformed configuration
Fig. 2.3-16: Kinematics of pure bending
 The following features available for other shell elements are not
available for the 3D-shell element: 6-, 8-node elements, composite
(multilayer) shells.
 When using contact groups with true offsets (BCTPARA entry,

OFFTYPE=2), the 3D-shell elements contribute to the offsets.
However, the offset at a 3D-shell element node is computed as half
the current shell thickness at the node, and this offset is used
regardless of whether the contact occurs on the shell top surface or
on the shell bottom surface.

2.3: Shell elements
Fig. 2.3-17: Pure bending modeled with quadratic 3D elements
 The 3D-shell is more fully described in the following reference:
ref. T. Sussman & K.J. Bathe, 3D-shell elements for

structures in large strains, Computers & Structures, 122,
2-12, 2013.

Chapter 2: Elements
Top control vector
Bottom control vector
Top control vector shrinks
Bottom control vector elongates
Fig. 2.3-18: Pure bending modeled with 3D-shell elements
Kinematics and degrees of freedom of the 3D-shell element
Fig. 2.3-19 shows the corner of a 3D-shell element, with its top and
bottom surfaces described by control vectors. Initially the control
vectors are equal and opposite. During deformations, the control
vectors can evolve independently. Thus in the deformed
configuration, the control vectors are in general not equal and
opposite, as shown in Fig 2.3-19(b).

2.3: Shell elements
Top surface
Bottom surface
a) Undeformed configuration b) Deformed configuration
Fig. 2.3-19: Control vectors at 3D-shell element node
The control vector motions are governed by element degrees of

freedom. For ease of use of the element, the element degrees of
freedom include the same degrees of freedom as for the usual shell
elements:
x, y, z translations
 ,  rotations (5 DOF node), or  x ,  y ,  z rotations (6 DOF
node)
and extra degrees of freedom:
constant and linear thickness incremental strains
The elongations of the control vectors are governed by the constant

and linear thickness incremental strains. The control vectors
always point in opposite directions, but in general have different
lengths after deformations. And material particles that were
initially on a straight line remain on a straight line after
deformations.
All of the considerations for selection of 5 and 6 DOF nodes

discussed in Section 2.7.3 directly apply to the 3D-shell element
nodes.
It is not allowed to prescribe or fix any of the extra degrees of

freedom.

Chapter 2: Elements
Material models and formulations for the 3D-shell element
Material models for the usual shell elements are developed using
the assumption of zero stress through the shell thickness. Hence
these models do not directly apply to the 3D-shell element.
The following material models are implemented for the 3D-shell

element: elastic-isotropic (MAT1), plastic-cyclic (MATPLCY),
hyperelastic (MATHE, MATHP).
Small displacement/small strain, large displacement/small

strain or large displacement/large strain formulations can be
used with these material models, as described under the material
models descriptions.
Material model notes
The thermal strains are assumed to be small in all material models,

including the hyperelastic material models.
When stresses are output in the shell local coordinate system, the
coordinate system is the rˆ, sˆ, tˆ system shown in Fig. 2.3-5(b).
Mixed u/p formulation
The mixed u/p formulation is used for the plastic-cyclic material

and all of the hyperelastic materials except for the hyperfoam
material. When the u/p formulation is used, the assumed pressure
field is p  p0  p1t where p0 and p1 are the pressure degrees of
freedom, and t is the isoparametric coordinate through the shell
thickness. Note that it is necessary to allow a linear variation of
pressure through the shell thickness in order to model out-of-plane
bending.
Incompatible modes
Incompatible modes can be used in the 4-node 3D-shell element.

However incompatible modes and the u/p formulation cannot be
used together, therefore incompatible modes are not used for the
materials in the last paragraph.

2.4: Surface elements — 2-D solids
2.4 Surface elements – 2-D solids (Solution 601 only)

 2-D solid elements are obtained in the following cases:
 PPLANE property ID that is referenced by CPLSTS3,

CPLSTS4, CPLSTS6 or CPLSTS8 plane stress elements, or
the CPLSTN3, CPLSTN4, CPLSTN6 or CPLSTN8 plane
strain elements. This leads to 2D plane strain or plane stress
elements that must be oriented in either the X-Y or X-Z
plane.
 PSOLID or PLSOLID property ID entry that is referenced
by the axisymmetric elements CQUADX4, CQUADX8,
CTRAX3, or CTRAX6. This leads to an axisymmetric 2-D
element which must be oriented in either the X-Y or X-Z
plane. This is the preferred form for axisymmetric elements
since elastic, plastic and hyperelastic materials can be used
with these elements. Contact analysis can also be performed
with these elements.
 PLPLANE property ID that is referenced by the CPLSTS3,
CPLSTS4, CPLSTS6 or CPLSTS8 plane stress elements, or
the CPLSTN3, CPLSTN4, CPLSTN6 or CPLSTN8 plane
strain elements. This leads to a hyperelastic 2-D plane strain
or plane stress element which must be oriented in either the
X-Y or X-Z plane.
 PLPLANE property ID entry that is referenced by the
CQUAD, CQUAD4, CQUAD8, CTRIA3 or CTRIA6 shell
elements. This leads to a hyperelastic plane strain or
axisymmetric 2-D element which must be oriented in the X-
Y plane.
 PSHELL property ID entry with MID2 = -1 that is
referenced by the shell elements CQUAD4, CQUAD8,
CTRIA3, or CTRIA6. This leads to a plane strain 2-D
element which must be oriented in the X-Y plane.
 2-D elements are not supported in Solution 701.

Chapter 2: Elements
 The following kinematic assumptions are available for two-

dimensional elements in Solution 601: plane strain, plane stress
and axisymmetric. Fig. 2.4-1 and Fig. 2.4-2 show some typical
2-D elements and the assumptions used in the formulations.
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
(a) 8- & 9-node quadrilateral elements

l
l l
l l
l l
(b) 3-node triangular element (c) 4-node quadrilateral element

l l
l
l l l
l
l l
l l
l l
l l
(d) 6- & 7-node triangular elements
Fig. 2.4-1: 2-D solid elements
 2-D solid elements in Solution 601 are classified based on the

number of nodes in the element and the element shape. Table 2.4-1
shows the correspondence between the different 2-D solid elements
and the NX element connectivity entries.

y l l e zz = 0
y
l l s zz = 0
g xz = 0 s zx = 0
z x l l g yz = 0 z x l l s zy = 0
a) Plane strain element b) Plane stress element
ezz = u/x
l l
y (linear analysis)
g xz = 0
z x l l
g yz = 0
c) Axisymmetric element
Fig. 2.4-2: Basic assumptions in 2-D analysis (assuming

element lies in X-Y plane)
 Note that the extra middle node in the 7-node and 9-node 2-D
elements is automatically added by the program when ELCV is set
to 1 in the NXSTRAT entry. These extra nodes improve the
performance of the 2-D elements as explained later in this section.
The boundary conditions at the added node are predicted from the
neighboring nodes.
 The axisymmetric element must lie in the +X half plane.
 2-D solid elements can be combined with any other elements

available in Advanced Nonlinear Solution.
 The axisymmetric element represents one radian of the

ref. KJB structure, and defines the stiffness, mass and forces accordingly.
Sections 5.3.1
and 5.3.2 Hence, when this element is combined with other elements, or
when concentrated loads are defined, these must also refer to one
radian, see ref. KJB, Examples 5.9 and 5.10, p. 356.
 The plane strain element provides for the stiffness of a unit

thickness of the structure, and defines the stiffness, mass and forces
accordingly.

Chapter 2: Elements
Table 2.4-1: Correspondence between 2-D solid elements and NX element

connectivity entries
2-D solid element NX element connectivity entry
CPLSTN31, CPLSTS32
3-node triangle
CTRIA34, CTRAX35
CPLSTN41, CPLSTS42
4-node quadrilateral CQUAD44, CQUADX45
CQUAD3 (with 4 input nodes)
CPLSTN61, CPLSTS62
6-node triangle
CTRIA64, CTRAX65
CPLSTN61,6, CPLSTS62,6
7-node triangle
CTRIA64,6, CTRAX65,6
CPLSTN81, CPLSTS82
8-node quadrilateral CQUAD84, CQUADX85
CPLSTN81,6, CPLSTS82,6
CQUAD84,6, CQUADX85,6
9-node quadrilateral
CQUAD3,6 (with 8 input nodes)
Notes: 1. Plane strain

2. Plane stress
3. Plane strain hyperelastic only
4. Plane strain hyperelastic
5. Axisymmetric with no restriction on material
6. With ELCV = 1 in NXSTRAT entry
 The plane stress 2-D element has an element thickness that is

defined either in the CPLSTSi element entry or in the PPLANE or
PLPLANE entry. The element can have a varying thickness, that is,
the thickness can be different at each node. However, the thickness
at a mid-side node is always taken as the average of the thickness
of the corresponding corner nodes.

 The basic 2-D elements used in Solution 601 are isoparametric

displacement-based elements, and their formulation is described in
detail in ref. KJB, Section 5.3.
 The basic finite element assumptions for the coordinates are

(see Fig. 2.4-3):
q q
x   hi xi ; y   hi yi
i 1 i 1
and for the displacements:
q q
u   hi ui ; v   hi vi
i 1 i 1
where
hi(r,s) = interpolation function corresponding to node i

(r,s) = isoparametric coordinates
q = number of element nodes
xi, yi = nodal point coordinates
ui , vi = nodal point displacements
The equations above are for 2-D solid elements that lie in the X-Y
plane. A simple change of variable from y to z describes the 2-D
solid elements in the X-Z plane.

 In addition to the displacement-based elements, special mixed-
interpolated elements are also available, in which the displacements
and pressure are interpolated separately. These elements are
effective and should be preferred in the analysis of incompressible
media and inelastic materials (elastic materials with Poisson's ratio
close to 0.5, rubber-like materials, creep and elasto-plastic
materials). The mixed formulation is only available for plane strain
and axisymmetric 2-D elements. It is not available (and not needed)
for plane stress 2-D elements.

Chapter 2: Elements
Y
s
4l l
7 3
l
Displacement degrees
of freedom
8l l
9 6 r
l
v2
1l l
5 2 u2
l
y2
Z X
x2
Fig. 2.4-3: Conventions used for the 2-D solid element
(assuming element lies in X-Y plane)
 Table 2.4-2 shows the number of pressure degrees of freedom

used for each 2-D element type. For more details on the number of
degrees of freedom ideal for each element, see the ref. KJB,
Section 4.4.3 and Table 4.6, pp. 292-295.
Table 2.4-2: Mixed formulation settings for 2-D solid elements
Number of pressure
2-D solid element
DOFs
3-node triangle -
4-node quadrilateral 1
6-node triangle 3
7-node triangle 3

 The mixed interpolation is the default setting for hyperelastic

materials. It can be activated for other materials, such as elastic
with Poisson’s ratio close to 0.5, elasto-plastic, and creep, via the
UPFORM flag in the NXSTRAT entry.
 The 4-node element (1 pressure degree of freedom) and 9-node

element (3 pressure degrees of freedom) are recommended for use
with the mixed formulation.
ref. T. Sussman and K.J. Bathe, "A Finite Element

Formulation for Nonlinear Incompressible Elastic and
Inelastic Analysis," J. Computers and Structures, Vol.
26, No. 1/2, pp. 357-409, 1987.
 In addition to the displacement-based and mixed-interpolated

elements, Advanced Nonlinear Solution also includes the
possibility of including incompatible modes (bubble functions) in
the formulation of the 4-node 2-D solid element. Within this
element, additional displacement degrees of freedom are
introduced. These additional displacement degrees of freedom are
not associated with nodes; therefore the condition of displacement
compatibility between adjacent elements is not satisfied in general.
The addition of the incompatible modes (bubble functions)
increases the flexibility of the element, especially in bending
situations. For theoretical considerations, see reference KJB,
Section 4.4.1. Note that these incompatible-mode elements are
formulated to pass the patch test. Also note that element distortions
deteriorate the element performance when incompatible modes are
used.
For plane stress and plane strain elements, the incompatible
modes feature is activated by settting ICMODE=1 in the
NXSTRAT entry. For axisymmetric elements, the incompatible
modes feature is activiated by setting IN=BUBBLE in the PSOLID
entry.
The incompatible modes feature should be used with caution
when using large displacement/small strain or large displacement/
large strain kinematics, in conjunction with large aspect ratio
elements, because meshes of incompatible modes elements can
contain spurious modes under these conditions.
The incompatible modes feature cannot be used in conjunction
with the mixed-interpolation formulation.

Chapter 2: Elements
 The interpolation functions used for 2-D solid elements are

defined in ref. KJB, Fig. 5.4, p. 344.
ref. KJB  The 6-node spatially isotropic triangle is obtained by correcting

Section 5.3.2 the interpolation functions of the collapsed 8-node element. It then
uses the same interpolation functions for each of the 3 corner nodes
and for each of the midside nodes.
The 3-node triangular element is obtained by collapsing one
side of the 4-node element. This element exhibits the constant
strain conditions (except that the hoop strain in axisymmetric
analysis varies over the element).
 The stresses/strains can be output either at the center and corner

grid points (PSOLID STRESS=blank or GRID), or at the center
and corner Gauss points (PSOLID STRESS=1 or GAUSS). The
option for output at the Gauss points is only available for
axisymmetric elements.
compatible with 2-D solid elements.
 Advanced Nonlinear Solution automatically uses the mixed

interpolation formulation for hyperelastic materials. The mixed
formulation is also recommended for elastic-plastic materials and
also elastic materials with a Poisson ratio close to 0.5. For these
materials, the u/p mixed formulation can be activated by setting
UPFORM = 1 in the NXSTRAT entry.
 The two-dimensional elements can be used with
- small displacement/small strain kinematics,
- large displacement/small strain kinematics, or
- large displacement/large strain kinematics.
 The small displacement/small strain and large

displacement/small strain kinematics can be used with any of
the compatible material models, except for the hyperelastic

material. The use of a linear material with small

displacement/small strain kinematics corresponds to a linear
formulation, and the use of a nonlinear material with the
small displacement/small strain kinematics corresponds to a
materially-nonlinear-only formulation.
 The program uses the TL (total Lagrangian) formulation
when the large displacement/small strain formulation is
selected.
 The large displacement/large strain kinematics can be used
with plastic materials including those with thermal and creep
effects, as well as hyperelastic materials. The ULH (updated
Lagrangian Hencky) formulation or ULJ (updated
Lagrangian Jaumann) formulation can be used for all
compatible material models except the hyperelastic material.
For the hyperelastic material, the TL (total Lagrangian)
formulation is used. The ULFORM parameter in the
NXSTRAT entry determines the ULH/ULJ setting.
ref. KJB
 The basic continuum mechanics formulations of 2-D solid
Sections 6.2 and elements are described in ref. KJB, pp. 497-537, and the finite
6.3.4 element discretization is given in ref. KJB pp. 538-542, 549-555.
ref. KJB  Note that all these formulations can be mixed in the same finite
Section 6.8.1 element model. If the elements are initially compatible, then they
will remain compatible throughout the analysis.
ref. KJB  The 4-node quadrilateral element uses 22 Gauss integration for
Sections 5.5.3, the calculation of element matrices. The 8-node and 9-node
5.5.4 and 5.5.5 elements use 33 Gauss integration. See Fig 2.4-4(a).

Chapter 2: Elements
s
s
r r
a) Rectangular elements
r
1-point 4-point 7-point
b) Triangular elements
Fig. 2.4-4: Integration point positions for 2-D solid elements
 The 3-node, 6-node and 7-node triangular elements are spatially

isotropic with respect to integration point locations and
interpolation functions (see Section 5.3.2, ref. KJB). The 3-node
element uses a single point integration in plane strain and 4-point
Gauss integration in the axisymmetric case. The 6-node and 7-node
triangular elements use 7-point Gauss integration. See Fig 2.4-4(b).
 Note that in geometrically nonlinear analysis, the spatial

positions of the Gauss integration points change continuously as
the element undergoes deformations, but throughout the response
the same material particles are at the integration points.

2.4.4 Mass matrices
 The consistent mass matrix is always calculated using either

33 Gauss integration for rectangular elements or 7-point Gauss
integration for triangular elements.
 The lumped mass matrix of an element is formed by dividing

the element’s mass M equally among its n nodes. Hence, the mass
assigned to each node is M / n . No special distributory concepts
are employed to distinguish between corner and midside nodes, or
to account for element distortion.
 Heat transfer capabilities are available for all 2-D solid

elements.
 The planar 2-D solid heat transfer elements may be defined

using any of the 2-D solid plane strain or plane stress element
entries. There is no difference between plane strain and plane stress
for heat transfer analysis, except for the element thickness, see
below.
 The axisymmetric elements must be defined using the

CQUADXi or CTRAXi entries, and they cover one radian of the
physical domain.
 The element matrices are integrated numerically by Gauss

matrices.
 The planar 2-D heat transfer element assumes the same

thickness as the underlying plane stress or plane strain element.
The axisymmetric element always extends one radian in the
circumferential direction.


Chapter 2: Elements
 The lumped heat capacity matrix of an element is formed by

dividing the element’s total heat capacity C equally among its n
nodes. Hence, the mass assigned to each node is C / n . No special
distributory concepts are employed to distinguish between corner
and midside nodes, or to account for element distortion.

2.4.6 Recommendations on use of elements
 The 9-node element is usually the most effective.
 The linear interpolation elements (3-node and 4-node) should

only be used in analyses when bending effects are not dominant. If
the 4-node element is used in problems where bending effects are
significant, incompatible modes should be activated.
 For nearly incompressible elastic materials, elasto-plastic

materials and creep materials, and when using plane strain or
axisymmetric elements, the use of the u/p mixed formulation
elements is recommended.
2.5 Solid elements – 3-D

 3-D solid elements are generated using the CHEXA, CPENTA ,

CTETRA and CPYRAM element connectivity entries. They
generate 6-, 5- and 4-sided 3-D elements. Typical 3-D solid
elements are shown in Fig 2.5-1.
 The PSOLID property ID entry is used for all of the supported

materials, except hyperelastic, which uses PLSOLID.

2.5: Solid elements — 3-D
l l l ll l l l l
l
l l l l l
l l l
l l l
l l l l
l l l l l
l l
l
l
l l l l l
l l l l l l l l
l l l l l
l l l
l l l l l l
(a) 8-, 20- and 27-node brick elements (CHEXA)
l l l
l l
l l l l
l l l
l
l
l l l l
l l l l l l
l l
(b) 4-, 10- and 11-node tetrahedral elements (CTETRA)

l
l l
l l l
l l l
l l
l l l
l l l
l ll
l
l
l l l l l l l
l l ll
l l
l l
l l l l
l l
(c) 6-, 15-, and 21-node wedge elements (CPENTA)
l l l
l l
l l
l l
l l l l
l l l l l
l l
l l
l
(d) 5-, 13-, and 14-pyramid elements (CPYRAM)
Fig. 2.5-1: 3-D solid elements

Chapter 2: Elements
 3-D solid elements in Advanced Nonlinear Solution are

classified based on the number of nodes in the element, and the
element shape.
 Table 2.5-1 shows the correspondence between the different

3-D solid elements and the NX element connectivity entries. Note
that the elements are frequently referred to just by their number of
nodes.

 Solution 701 only supports linear elements (4-node tetrahedron,
6-node wedge and 8-node brick elements).
Table 2.5-1: Correspondence between 3-D solid elements and NX

element connectivity entries
3-D solid element NX element connectivity entry

4-node tetrahedron CTETRA
1
10-node tetrahedron CTETRA
11-node tetrahedron1 CTETRA and ELCV = 1 in NXSTRAT
6-node wedge CPENTA
15-node wedge1 CPENTA
1
21-node wedge CPENTA and ELCV = 1 in NXSTRAT
8-node brick CHEXA
1
20-node brick CHEXA
27-node brick1 CHEXA and ELCV = 1 in NXSTRAT
5-node pyramid CPYRAM
13-node pyramid1 CPYRAM
1
14-node pyramid CPYRAM and ELCV = 1 in NXSTRAT
Note:
1. Only for Solution 601



 Advanced Nonlinear Solution supports incomplete quadratic

3-D elements for tetrahedral and pyramid elements. Incomplete
quadratic elements are not supported for brick and wedge elements.
For example, a CHEXA entry can only have 8 nodes or 20 nodes.
Anything in between is not supported. Also, a CTETRA can have
any of its midside nodes removed.
 For nonlinear analysis, stress/strain results for 3-D solid

elements are output in the element coordinate system. ELRESCS =
1 in NXSTRAT may be used to request output of nonlinear
stress/strain results in the material coordinate system. The option is
useful for post-processors that do not perform any transformation
of the stress/strain coordinate system when importing the op2 file.
 Note that the mid-volume and midsurface nodes in the 27-node,

21-node, 14-node and 11-node elements are automatically added by
Advanced Nonlinear Solution when ELCV is set to 1 in the
NXSTRAT entry. The boundary conditions at the added nodes are
predicted from the neighboring nodes.
ref. KJB  The elements used in Advanced Nonlinear Solution are

Section 5.3 isoparametric displacement-based elements, and their formulation
is described in ref. KJB, Section 5.3.
 The basic finite element assumptions for the coordinates are

(see Fig. 2.5-2, for the brick element):
q q q
x   hi xi y   hi yi z   hi zi
i 1 i 1 i 1
and for the displacements:
q q q
u   hi ui v   hi vi w   hi wi
i 1 i 1 i 1

Chapter 2: Elements
t
Z 7
l
w8 l
18
19
l
l
6
v8 15 l
8l l
20 l l
l l 17
l 5 14
l l l
u8 l 16
3 l
l
l l 10
11 13 l
4l l l 2
s
l
l 9
z4 r 12
l
1
x4 Y
y4
X
Fig. 2.5-2: Conventions used for the nodal coordinates and

displacements of the 3-D solid element
where
hi (r, s, t) = interpolation function corresponding to node i
r, s, t = isoparametric coordinates
q = number of element nodes
xi, yi, zi = nodal point coordinates
ui, vi, wi = nodal point displacements

 In addition to the displacement-based elements, special mixed-
interpolated elements are also available, in which the displacements
and pressure are interpolated separately. These elements are
effective and should be preferred in the analysis of incompressible
media and inelastic materials (specifically for materials in which
Poisson's ratio is close to 0.5, for rubber-like materials and for
elasto-plastic materials). Table 2.5-2 shows the number of pressure
degrees of freedom for each 3-D element type. For more details on
the mixed interpolation of pressure and displacement degrees of
freedom for 3-D solids, see Section 4.4.3, p. 276, and Tables 4.6
and 4.7, pp. 292 - 295 in ref. KJB.

Table 2.5-2: Mixed u/p formulations available for 3-D solid elements
Number of
3-D solid element
pressure DOFs
4-node tetrahedron -
10-node, 11-node tetrahedron 4
6-node wedge 1
15-node, 21-node wedge 4
8-node brick 1
20-node, 27-node brick 4
5-node pyramid 1
13-node, 14-node pyramid 1
 The mixed interpolation is the default setting for hyperelastic

materials. It can be activated for other materials, such as elastic
with Poisson’s ratio close to 0.5, elasto-plastic, and creep, via the
UPFORM flag in the NXSTRAT entry.
 The 8-node element (1 pressure degree of freedom) and 27-node

element (4 pressure degrees of freedom) are recommended for use
with the mixed formulation.

 Note that 4 pressure degrees of freedom are used for the 10-
node tetrahedron, the 15-node wedge and the 20-node brick
element. Even though this setting does not satisfy the inf-sup test,
the elements generally perform better than with a single pressure
degree of freedom. Still, it is better to add the midside nodes if
possible. This is done by setting ELCV = 1 in the NXSTRAT
entry.
 In addition to the displacement-based and mixed-interpolated

elements, Advanced Nonlinear Solution also includes the
possibility of including incompatible modes (bubble functions) in
the formulation of the 5-node pyramid, 6-node wedge and the 8-
node brick element. Within this element, additional displacement
degrees of freedom are introduced. These additional displacement
degrees of freedom are not associated with nodes; therefore the
condition of displacement compatibility between adjacent elements

Chapter 2: Elements
is not satisfied in general. The addition of the incompatible modes

(bubble functions) increases the flexibility of the element,
especially in bending situations. For theoretical considerations, see
reference KJB, Section 4.4.1. Note that these incompatible-mode
elements are formulated to pass the patch test. Also note that
element distortions deteriorate the element performance when
incompatible modes are used.
The incompatible modes feature is activiated by setting
IN=BUBBLE in the PSOLID entry.
The incompatible modes feature should be used with caution
when using large displacement/small strain or large displacement/
large strain kinematics, in conjunction with large aspect ratio
elements, because meshes of incompatible modes elements can
contain spurious modes under these conditions.
The incompatible modes feature cannot be used in conjunction
with the mixed-interpolation formulation.
 Table 2.5-3 shows which elements support incompatible modes

(bubble functions). The incompatible modes feature is only
available for the 5-node pyramid, 6- node wedge and the 8-node
brick elements.
Table 2.5-3: Incompatible modes (bubble functions) available for

3-D solid elements
Support for
3-D solid element incompatible
modes
4-node tetrahedron No
5- to 11-node tetrahedron No
6-node wedge Yes
15-node, 21-node wedge No
8-node brick Yes
20-node, 27-node brick No
5-node pyramid Yes
6- to 14-node pyramid No

 The interpolation functions used for 3-D brick solid elements

for q ≤ 20 are shown in Fig. 5.5, ref. KJB, p. 345 (note that ref KJB
uses a different local node numbering convention).
 The 10-node tetrahedron (see Fig. 2.5-1(c)) is obtained by

collapsing nodes and sides of rectangular elements. Spatially
isotropic 10-node and 11-node tetrahedra are used in Solution 601.
The 4-node tetrahedron (see Fig. 2.5-1(c)) is obtained by
collapsing nodes and sides of the 8-node brick element. This
element exhibits constant strain conditions.
 The stresses/strains can be output either at the center and corner

grid points (PSOLID STRESS=blank or GRID), or at the center
and corner Gauss points (PSOLID STRESS=1 or GAUSS).
2.5.2 Material models and nonlinear formulations
compatible with 3-D solid elements.
 Advanced Nonlinear Solution automatically uses the mixed

interpolation formulation for hyperelastic materials. The mixed
formulation is also recommended for elastic-plastic materials and
elastic materials with a Poisson ratio close to 0.5. It can be
activated by setting UPFORM = 1 in the NXSTRAT entry.
 The 3-D elements can be used with
 - small displacement/small strain kinematics,
 - large displacement/small strain kinematics, or
 - large displacement/large strain kinematics.
The small displacement/small strain and large

displacement/small strain kinematics can be used with any of
the compatible material models, except for the hyperelastic
material. The use of a linear material with small
displacement/small strain kinematics corresponds to a linear
formulation, and the use of a nonlinear material with the
small displacement/small strain kinematics corresponds to a

Chapter 2: Elements
materially-nonlinear-only formulation.
The program uses the TL (total Lagrangian) formulation
when large displacement/small strain kinematics is selected.
The large displacement/large strain kinematics can be used
with plastic materials including thermal and creep effects, as
well as hyperelastic materials. The ULH (updated
Lagrangian Hencky) formulation or the ULJ (updated
Lagrangian Jaumann) formulation can be used for all
compatible material models except the hyperelastic material.
For the hyperelastic material models, a TL (total Lagrangian)
formulation is used. The ULFORM parameter in the
NXSTRAT entry determines the ULH/ULJ setting.
ref. KJB  The basic continuum mechanics formulations are described in

Sections 6.2 ref. KJB, pp. 497-568. The finite element discretization is
and 6.3.5 summarized in Table 6.6, p. 555, ref. KJB.
ref. KJB  Note that all these formulations can be used in the same finite
Section 6.8.1 element model. If the elements are initially compatible, they will
remain compatible throughout the analysis.
ref. KJB  The 8-node brick element uses 222 Gauss integration for the
Sections 5.5.3,
5.5.4 and 5.5.5 calculation of element matrices. The 20-node and 27-node elements
use 333 Gauss integration.
 Tetrahedral elements are spatially isotropic with respect to

integration point locations and interpolation functions. By default,
for the 4-node tetrahedral element, 1-point Gauss integration is
used, for the 10-node tetrahedral element, 5-point Gauss integration
is used, and 17-point Gauss integration is also used for the 11-node
tetrahedral element. The Gauss integration order for tetrahedral
elements can also be chosen using TETINT in the NXSTRAT
entry.
 Note that in geometrically nonlinear analysis, the spatial

positions of the Gauss integration points change continuously as
the element undergoes deformations, but throughout the response
the same material particles are at the integration points.

 The same integration order is used for both Solution 601 and
701.
2.5.4 Mass matrices
 The consistent mass matrix is always calculated using 333

Gauss integration except for the tetrahedral 4-node, 10-node and
11-node elements which use a 17-point Gauss integration.
 The lumped mass matrix of an element is formed by dividing

the element’s mass M equally among each of its n nodal points.
Hence the mass assigned to each node is M / n . No special
distributory concepts are employed to distinguish between corner
and midside nodes, or to account for element distortion.
 The same lumped matrix is used for both Solution 601 and
Solution 701.
 Heat transfer capabilities are available for all 3-D solid

elements.
 The element matrices are integrated numerically by Gauss

matrices.

 The lumped heat capacity matrix of an element is formed by

dividing the element’s total heat capacity C equally among each of
its n nodal points. Hence the mass assigned to each node is C / n .
No special distributory concepts are employed to distinguish
between corner and midside nodes, or to account for element
distortion.

Chapter 2: Elements
2.5.6 Recommendations on use of elements
 The linear interpolation elements (4- to 8-node) usually perform

better in contact problems.
 The linear interpolation elements (5-node, 6-node and 8-node

brick elements, without incompatible modes) should only be used
in analyses when bending effects are not dominant. If bending
effects are insignificant, it is usually best to not use incompatible
modes.
 Since the 4-node tetrahedron is a constant strain element, many

elements (fine meshes) must usually be used in analyses.
 For nearly incompressible elastic materials, elasto-plastic

materials and creep materials, the use of the u/p mixed formulation
elements is recommended.

 When the structure to be modeled has a dimension which is
extremely small compared with the others, e.g., thin plates and
shells, the use of the 3-D solid element usually results in too stiff a
model and a poor conditioning of the stiffness matrix. In this case,
the use of shell elements, particularly the 4-node shell element (see
Section 2.3), is more effective.
Recommendations specific to Solution 601
ref. KJB  The 27-node element is the most accurate among all available
Page 383
elements. However, the use of this element can be costly.
 The 20-node element is usually the most effective, especially if

the element is rectangular (undistorted).

2.6: Scalar elements — Springs, masses and dampers
2.6 Scalar elements – Springs, masses and dampers

2.6.1 CELAS1, CELAS2, CMASS1, CMASS2, CDAMP1, CDAMP2
 Scalar elements in Advanced Nonlinear Solution either connect

2 degrees of freedom together or just a single degree of freedom to
the ground. There are three forms of scalar elements: springs,
masses, and dampers.
 Spring elements are defined using the CELAS1 and

CELAS2 element connectivity entries.
 Mass elements are defined using the CMASS1 and
CMASS2 element connectivity entries.
 Damper elements are defined using the CDAMP1 and
CDAMP2 element connectivity entries.
 Fig. 2.6-1 shows the spring, mass and damper single degree of
freedom elements available in Advanced Nonlinear Solution. They
correspond to a grounded spring, a concentrated mass, and a
grounded damper, respectively.
 Fig. 2.6-2 shows the available scalar elements connecting two

degrees of freedom. Only the translational version of the spring and
damper are shown in the figure, but they can connect rotational
degrees of freedom as well.
2.6.2 6-DOF spring element (Solution 601 only)
 The 6-DOF spring element is a generalized spring-damper

element which can be linear or materially-nonlinear only (MNO).
This element is defined using the CBUSH element connectivity
entries. It can have single node, two coincident or two non-
coincident nodes. In each degree of freedom, the element stiffness
can be defined as a constant or using a force-displacement curve in
the element coordinate system. The damping coefficients are
always constants in units of force per unit velocity.

Chapter 2: Elements
u q ü
k
l
l
k m
Single Single Single Single

translational rotational translational rotational
DOF spring DOF spring DOF mass DOF mass
Stiffness K = [k] Mass M = [m]

(a) spring element (b) mass element
.
u
c c
l l
Single Single
translational rotational
DOF damper DOF damper
Damping C = [c]
(c) damper element
Fig. 2.6-1: Single degree of freedom scalar elements
 A displacement (skew) system can be used in the 6-DOF spring

element to prescribe loads and constraints. Element birth/death is
also supported.
 Currently, the 6-DOF spring element is not supported in

Solution 701.
 If a 6-DOF spring element has single node or two coincident

nodes, its element coordinate system must be defined using a CID
as shown in Fig. 2.6-3 and Fig. 2.6-4. A single node 6-DOF spring
element corresponds to a grounded spring acting in the user-
specified degree of freedom.

k -k
k
K = 1 2
-k k
U1 U2
a) spring element
c -c c
C = 1 2
-c c
U1 U2
b) damper element
m/2 0
Mlumped =
0 m/2 m
m/3 m/6 U1 U2
Mconsistent =
m/6 m/3
c) mass element
Fig. 2.6-2: Two-degrees-of-freedom scalar elements
y-element
z-element x-element
CID
z y
GA
x
Fig. 2.6-3: 6-DOF spring element with

single node

Chapter 2: Elements
y-element
z-element x-element
CID
y
z
GA, GB x
Fig. 2.6-4: 6-DOF spring element with two

coincident nodes
 If a 6-DOF spring element has two non-coincident nodes, its

element coordinate system can be defined using a CID, an
orientation vector or its axial direction as shown in Figs. 2.6-5 to
2.6-7.
In Fig. 2.6-5, the element coordinate system is defined by a
CID. Note that GA and GB might or might not have displacement
(skew) coordinate systems.
`In Fig. 2.6-6, the element coordinate system is defined by an
orientation vector using GO or X1, X2, X3. Note that X1, X2, X3
refers to the displacement (skew) coordinate system of GA.
In Fig. 2.6-7, a 6-DOF spring element is defined with two non-
coincident nodes without GO, X1, X2, X3 or CID. This defines a 1-
D axial/torsional spring/damper. In this case, axial stiffness (or
damping) or torsional stiffness (or damping) or both must be
specified but all other stiffness (or damping) must not be specified.

y-element
x-element
z-element
CID
y
z
x
GA
GB
Fig. 2.6-5: 6-DOF spring element with two

non-coincident nodes
z-element y-element
(normal to plane) (in plane) GO or (X1,X2,X3)
Plane of GA, GB, GO

GA x-element
GB
Fig. 2.6-6: 6-DOF spring element with two non-coincident

nodes
GA x-element
GB
Fig. 2.6-7: 1-D spring element with two non-

coincident nodes

Chapter 2: Elements
2.7 R-type elements

 R-type elements impose multipoint constraints on one or more
nodes. The constraints are created automatically by the program
based on the element’s input. The following R-type elements are
supported in Advanced Nonlinear Solution: RBAR, RBE2 and
RBE3.
 Rigid elements are a subset of R-type elements that include

RBAR and RBE2.
 RBE3 is an interpolation constraint element which also

produces constraint equations.
2.7.1 Rigid elements
 Solution 601 provides several options for modeling the Rigid

elements. They can be modeled as perfectly rigid elements using
constraint equations or as flexible (but stiff) elements. The
EQRBAR or EQRBE2 parameters in the NXSTRAT entry
determine how the Rigid elements are treated.
 Solution 701 does not support the flexible option.
 The RBAR entry generates a single Rigid element between two

nodes.
 The RBE2 entry generates multiple Rigid elements. They

connect one independent node to several nodes.
 If the perfectly rigid option is selected, Rigid elements are

internally represented either as standard multipoint constraints, or
as rigid links (see Section 5.8 for enforcement of constraint
equations). Multipoint constraints have constant constraint
coefficients and therefore do not give accurate results in large
displacements (unless the 2 nodes are coincident or the constraints
do not involve rotational degrees of freedom). Rigid links also
create multipoint constraints but with variable coefficients that are
updated based on the deformation of the structure. This is
illustrated in Fig. 2.7-1. Therefore, whenever possible, large-
displacement rigid links are used.

2.7: R-type elements
l l l
qrotation l l l
l
using MPC — small rotations
l l l
l l l
l
3 RBAR elements with
l
one independent node
l
l
l
l
using large displacement
rigid links
Fig. 2.7-1: Difference between small displacement MPC and

large displacement rigid links
 Rigid elements that are internally represented as multipoint

constraints are affected by the general constraint setting (GENMPC
parameter in NXSTRAT). If constraints are set to general
constraints (GENMPC=1), the constraint is enforced using
Lagrange multipliers. Rigid elements represented by rigid links
(which have variable constraint coefficients) are not influenced by
the general constraint flag. They are always enforced using the
default master-slave constraint approach.
 If the flexible option is selected for Rigid elements, Solution

601 internally generates beam or spring elements depending on the
Rigid element parameters and the distance between the nodes
(RBLCRIT parameter in NXSTRAT), or a spring element
translation can be always requested (in the EQRBAR parameter in
NXSTRAT).
 The stiffness of the internal springs and the Young’s modulus

and cross-sectional area of the internal beams can be automatically
determined by Solution 601 or set by the user (see SPRINGK,
BEAME and BEAMA parameters in NXSTRAT entry).

Chapter 2: Elements
 The rigid option results in more accurate enforcement of the

constraint. However, the compliance introduced in the model when
using the flexible option can lead to easier convergence in
nonlinear problems.
 The flexible option results in none of the degrees of freedom

becoming dependent. This allows multiple constraints to be defined
at a node, and it is sometimes beneficial for contact.
 A dependent degree of freedom of a constraint (standard, not

general constraint) or rigid link cannot be used in another constraint
or rigid link as an independent degree of freedom. Hence, chaining
of constraints is not allowed. Chaining of rigid links is enabled by
internally replacing the dependent node of each rigid link by the
first node in the chain (to avoid the restriction mentioned above).
Classification of Rigid elements
 The internal representation of an RBAR rigid element depends

on the options present in CNA, CNB, CMA and CMB, as shown in
Fig. 2.7-2.
GB
l
GA
l Independent: CNB
Dependent: CMB
Independent: CNA
Dependent: CMA
Fig. 2.7-2: Relevant parameters in the RBAR

rigid element
 Currently, Advanced Nonlinear Solution identifies 5 classes of

RBAR settings. Each class gets a different internal representation.
Checking for each class is done in sequence starting with Class 1.
Class 1:
All 6 degrees of freedom of one point are dependent on those of the
other point. In other words,
CNA = 123456, CNB = 0, CMA = 0, CMB = 123456

or
CNA = 0, CNB = 123456, CMA = 123456, CMB = 0
Class 2:
One point has all the dependent degrees of freedom (but not all 6 of
them), and all those that are not dependent (missing terms in CMA
or CMB) involve degrees of freedom that do not exist at the slave
node. For example,
CNA = 123456, CNB = 0, CMA = 0, CMB = 123
where node B is attached only to 3D solid elements (so degrees of

freedom 456 do not exist).
Another example,
CNA = 0, CNB = 123456, CMA = 12, CMB = 0
where node A is attached only to 2D solid elements (so degrees of

freedom 3456 do not exist)
Note that this only applies to non-existent degrees of freedom (not

fixed ones). If an excluded DOF is fixed then the rigid element
does not belong to this Class.
Class 3:
One point has all the dependent degrees of freedom (but not all 6 of
them). In other words,
CNA = 123456, CNB = 0, CMA = 0, CMB = Q
or
CNA = 0, CNB = 123456, CMA = Q, CMB = 0
where Q is any combination of the 6 DOFs except “0” and

“123456” (“0” is not allowed, and “123456” belongs to Class 1).
Note that if the degrees of freedom not included in Q are all non-
existent at the node, then the rigid element belongs to Class 2.

Chapter 2: Elements
Class 4:
All 6 degrees of freedom active but not all dependent degrees of
freedom belong to 1 point. For example,
CNA CNB CMA CMB

123 456 0 0
12346 5 5 12346
Class 5:
Not all the 6 degrees of freedom are active in the constraint and
rigid element fails criteria for Classes 2 and 3. For example,
CNA CNB CMA CMB

123 456 4 3
Note that there are some other valid settings for RBAR that are
not supported in Advanced Nonlinear Solution.
 The internal representation of Rigid elements for each class is

described in Table 2.7-1.
 RBE2 is interpreted in the same manner as RBAR except that it

produces multiple Rigid elements. These elements can only belong
to Class 1 or 3, and their internal representation is dictated by the
EQRBE2 parameter in NXSTRAT.
Table 2.7-1: Internal representation of Rigid elements
Rigid option Flexible option

L < Lcrit L > Lcrit L < Lcrit L > Lcrit
Class 1 MPC Rigid link1 Springs Beam1
Class 2 MPC Rigid link1 Springs Beam1
Class 3 MPC Rigid link1 Springs Springs
Class 4 MPC MPC Springs Beam1
Class 5 MPC MPC Springs Springs
1
This constraint is accurate in large displacement analysis

Rigid elements and continuum elements: Care must be used when

connecting continuum elements (2-D and 3-D solid elements) with
rigid elements. In order for the rigid element to rotate, the master
DOF of the rigid element must have its rotational DOFs free.
This situation is illustrated in Fig. 2.7-3. Two assemblages of
2-D solid elements are connected together with large displacement
rigid elements. If the z rotational DOFs are fixed, then when the
lower element is rotated, the rigid elements maintain both their
length and their angle, and the upper element assemblage distorts
as shown in Fig 2.7-3(b). If the z rotational DOFs are free, then
when the lower element is rotated, the rigid elements are free to
rotate and the upper element assemblage can also rotate without
distortion as shown in Fig 2.7-3(c).
Solution 601 does not fix the rotational DOFs of all nodes
connected to continuum elements. Thus, the model in Fig. 2.7-3
will behave as shown in Fig 2.7-3(c), as expected.

Chapter 2: Elements
S S M = Master node
S = Slave node
M M
M S
Rigid link
a) Two element assemblages connected with rigid elements
Elements must distort
b) Master rotational DOFs are fixed, rigid elements cannot rotate
Elements remain undistorted
c) Master rotational DOFs are free, rigid elements can rotate
Fig. 2.7-3: Rigid elements used with continuum elements 

2.7.2 RBE3 element
 The RBE3 R-type element defines the motion of a reference

node as a weighted average of the motion of a set of other nodes.
This element is a useful tool for distributing applied load and mass
in a model.
 he reference node is denoted REFGRID, and only the

components REFC of this node are connected to the RBE3
element. The other nodes in the RBE3 element are denoted Gi,j.
The double indexing i,j is used so that weights WTi and
components Ci can be associated with sets of nodes Gi,j. Only the
components Ci of nodes Gi,j are connected to the RBE3 element.
 The element is implemented as a set of constraint equations in

which the motions of the reference node REFGRID are constrained
to the motions of the other nodes Gi,j.
 In the following, we briefly outline the derivation of the RBE3
element. This derivation is based upon TAN 4494 "Mathematical
Specification of the RBE3 Element". For simplicity, we consider
just the case in which all components are given in the basic
coordinate system, but the RBE3 element allows for the
components of the reference node, and for the components of each
of the connected nodes, to all be in a different coordinate system.
Let superscript q denote the reference node REFGRID and
superscript k denote one of the connected nodes Gi,j, with
associated weight WTi.
The derivation is based on the transmission of forces/moments
from the connected nodes k to the reference node q, and on the
transmission of forces/moments from the reference node q to the
connected nodes k.
Firstly, if the forces/moments at a given connected node k are
known, then these forces are transmitted to reference node q using
the equilibrium equation

Chapter 2: Elements
 Fxq   1 0 0 0 0 0   Fxk 
 q   
 Fy   0 1 0 0 0 0   Fyk 
 Fz  
q
0 0 1 0 0 0   Fzk 
 q    
M x   0 ( z k  z q ) ( y k  y q ) 1 0 0   M xk 
M q   (zk  zq ) 0 ( x k  x q ) 0 1 0   M yk 
 q 
y
 
 M z   ( y  y ) ( x  x )
k q k q
0 0 0 1   M zk 
or, in matrix form,
F q  (S k )T F k (2.7-1)
where F q and F k are column vectors containing the forces and

moments, and S k is the transpose of the square matrix in the above
equation.
Therefore if the forces/moments at all connected nodes k are
known, then the forces/moments at reference node q is found using
F q   (S k )T F k (2.7-2)
k
Secondly, we now postulate that, if the forces/moments at

reference node q are known, that these forces/moments are
transmitted to connected node k using
F k  W k S k XF q (2.7-3)
 w1k 
 k 
 w2 
 w3k 
where W k    is a diagonal
 w4k 
 w5k 
 
 w6k 
matrix with weighting factors given by WTi and Ci as follows:

For each component l, l=1,2,3,4,5,6, then

If component l is not in the Ci list, then
wlk  0
else
if l=1,2,3, then
wlk  WTi
else
wlk  WTi  L2c
endif
endif
end
In this calculation, Lc is the average distance between the

reference node and all of the connected nodes. If this average
distance is zero, then Lc  1 . Lc is introduced into the weighting
matrix so that the element has dimensional independence.
Matrix X is a 6x6 matrix that is the same for all of the
connected nodes. Matrix X is determined as follows. Combining
(2.7-2) and (2.7-3) gives
F q   (S k )T W k S k XF q (2.7-4)
k
and for this to be satisfied for all F q ,
1
 
X    (S k )T W k S k  (2.7-5)
 k 
Now let
G k  Wk Sk X (2.7-6)
From (2.7-3),
Fk  G k Fq (2.7-7)

Chapter 2: Elements
Finally, we apply the principle of virtual work to the RBE3

element:
(F q )T  u q   (F k )T  u k
k
(2.7-8)
  (F q )T (G k )T  u k
k
(in which  u are the virtual displacements/rotations), and for

(2.7-8) to be satisfied for all F q ,
 u q   (G k )T  u k (2.7-9)
k
(2.7-9) shows that the virtual displacements/rotations of the

reference node must be related to the virtual displacements/
rotations of the connected nodes. This can only occur if the actual
displacements/rotations of the reference node are related to the
actual displacements/rotations of the connected nodes:
u q   (G k )T u k (2.7-10)
k
(2.7-10) has six rows. Each row l selected by a component in REFC

corresponds to a multipoint constraint equation, in which ulq is
dependent and u k is independent. Rows not selected by a
component in REFC are discarded.
 The RBE3 element assumes that the displacements of the

reference and connected nodes are small. This is seen in the
definition of matrix S k , since the original coordinates of the nodes
are used in this matrix.
 If a point mass is attached to the reference node, and the mass

matrix is consistent, the point mass will be coupled to all connected
nodes. In the worst case, when the reference node is connected to
all nodes in the model, the mass matrix will be full, leading to a
very slow solution and high memory usage. To avoid this, use a
lumped mass matrix.

 Similarly, if a nonzero stiffness is attached to the reference

node, the stiffness will be coupled to all connected nodes.
2.8 Potential-based fluid elements (Solution 601,106 only)

 The elements discussed in this section incorporate the following
assumptions:
 Inviscid, irrotational medium with no heat transfer

 Compressible or almost incompressible medium
 Relatively small displacements of the fluid boundary
 The potential-based fluid elements can be used in static

analyses, where the pressure distribution in the fluid and the
displacement and stress distribution in the structure is of interest.
 The potential-based fluid elements can be employed in 2-D and

3-D analyses. Two-dimensional elements can be employed in
planar and axisymmetric analyses. Two-dimensional elements
must be defined in the XZ plane, and axisymmetric elements must
lie in the +X half plane (all nodal point coordinates must have non-
negative x values).
 The potential-based fluid elements can be coupled with

structural elements, as described in detail below. The structural
motions cause fluid pressure, and the fluid pressure causes
additional forces to act on the structure.
 The potential-based fluid elements can be coupled to a pressure

boundary condition (i.e., no structure adjacent to the potential-
based fluid element boundary). This feature can be used to model
free surfaces.
2.8.1 Theory
Fluid
Fig. 2.8-1 shows a generic fluid region. It is assumed that the fluid
is inviscid and irrotational with no heat transfer, and that the
velocities and density changes are infinitesimally small. Under
these assumptions, the continuity equation in the fluid is

Chapter 2: Elements
0 p
  (  )    0 2   0 2  0 (2.8-1)

the momentum/equilibrium equation is
p
h    x    (2.8-2)

and the pressure-density relationship is
 p
 1 (2.8-3)
0 
where  is the density,  is the velocity potential ( v   where

v is the fluid velocity), h is the specific enthalpy (defined as
dp
h ), p is the pressure,  is the bulk modulus and  0 is the

nominal density. Also   x  is the potential of the (conservative)
body force accelerations at position x . For example, when the
body forces are due to gravity,   g , where g is the
acceleration due to gravity.
Body force acceleration g,

body force acceleration potential W
Bounding surface S
Unknowns on S:
f = fluid potential Fluid region V
u = displacements n
Unknowns in V:
f = fluid potential
Essential boundary condition: f prescribed

Natural boundary condition: ru× ×n prescribed
Fig. 2.8-1: Fluid region

2.8: Potential-based fluid elements
The momentum/equilibrium equation can also be written

p   0   x     (2.8-4)
Substituting (2.8-4) into (2.8-3) gives
  0     2   0 
 (2.8-5)
Equation (2.8-5) is a special form of the wave equation. It is linear

in the solution variable  . (2.8-5) can be written in variational
form using standard techniques. The result is
   0   dV       dV    u  n  dS
V V S
  dV
   0  (2.8-6)
V
Structure
 We assume that part of the boundary S is adjacent to the

structure (Fig. 2.8-2). The part of the boundary adjacent to the
structure is denoted S1 .
Surface S1 (Fluid and structure

are separated for
clarity)
Fluid Structure
n
Pressure p Traction -pn
Fig. 2.8-2: Forces on structure from fluid

Chapter 2: Elements
The fluid pressure on S1 provides additional forces on the structure

adjacent to S1 :
  
 Fu   pn u dS1    0   0 u   0  n u dS1 (2.8-7)
S1 S1 
x 
The finite element contributions to the system matrices

corresponding to (2.8-6) and (2.8-7) are
0 0  U    0    K UU 
CTFU   U 0  U 
 0 M          0
S

 FF     CFU 0   K FF    
(2.8-8)
 R    0 
=  UB S     
 0    R FB 
where
M FF = matrix from   term in (2.8-6)

K FF = matrix from     term in (2.8-6)
CFU = matrix from u  n term in (2.8-6)
 
 K UU S = matrix from  0 u  n u term in (2.8-7)
 x 
 R UB S = loads vector from  0  n u term in (2.8-7)
R = loads vector from     term in (2.8-6)
FB 0
U = vector containing unknown nodal displacements
 = vector containing unknown nodal fluid potentials.
We note that the term  K UU S is numerically very small

compared with the rest of the structural stiffness matrix, when there
is a structure adjacent to the fluid. But  K UU S is important in the
case when there is no structure adjacent to the fluid.

 The left-hand-side of equation (2.8-8) with the exception of the

term  K UU S , is identical to the formulation presented in the
following reference:
ref. L.G. Olson and K.J. Bathe, “Analysis of fluid-structure

interactions. A direct symmetric coupled formulation
based on the fluid velocity potential”, J. Computers and
Structures, Vol 21, No. 1/2, pp 21-32, 1985.
Static conditions
The static equations of motion can be formally derived from

(2.8-8) by applying the Laplace transform to both sides of (2.8-8)
and applying the final value theorem. The result is
 K UU S CTFU   U   R UB S 
  = (2.8-9)
 CFU M FF       R FB 
together with the condition
K FF = 0 (2.8-10)
 There are a number of unusual characteristics of (2.8-9) and

(2.8-10):
 The solution involves  instead of  . This makes sense as

(2.8-4) then implies that p is constant (in time) in a static
solution.
 The condition (2.8-10) must be satisfied. This condition is
satisfied whenever  = constant within each separate fluid
region. Hence the number of unknown potential degrees of
freedom in static analysis is equal to the number of separate
fluid regions in the analysis.
 The condition  = constant within each separate fluid
region implies that p  0   C where C is a constant
determined from the solution. Hence the variation of
pressure within each separate fluid region is contained

Chapter 2: Elements
within C and any choice of constant of integration within

0  (recall that  is a potential and therefore includes an
arbitrary constant of integration) is balanced by an opposite
change in C .
 It is necessary to enter the density of the fluid in static
analysis, even when the solution does not depend on the
density.
 The 2nd row of (2.8-9) represents mass conservation within
each separate fluid region. Any structural motion that does
not change the fluid volume of the fluid region is not given
stiffness by the fluid. As examples, consider motions
tangential to the fluid, and also consider a "ripple" on a free
surface in which the fluid volume is not changed.
2.8.2 Elements
 The volume V of the fluid domain is modeled using two-

dimensional or three-dimensional fluid elements. These elements
are analogous to the two-dimensional or three-dimensional solid
elements and the nodal point numbering of the fluid elements is the
same as the nodal point numbering of the solid elements. Thus for
example, CHEXA, CTETRA, CPENTA, CPYRAM entries are
used to define three-dimensional fluid elements.
 The two-dimensional elements are either planar (unit thickness

of fluid assumed) or axisymmetric (1 radian of fluid assumed).
CPLSTNi entries are used to define planar elements and
CQUADXi, CTRAXi entries are used to define axisymmetric
elements. The elements must lie in the X-Z plane and the
axisymmetric elements must lie in the +X half plane.
 The bounding surface S of the fluid domain is modeled with

fluid boundaries, as discussed in detail below.
 It is required that each separate fluid domain be modeled with

separate fluid element PIDs. This is because Solution 601
constrains the potential degrees of freedom of the elements with the
same fluid element PIDs together in static analysis during phi
model completion, step 7, see Section 2.8.5.

 It is not permitted to have fluid regions of different densities

sharing the same potential degrees of freedom. This is because the
nodal pressure would be different as computed from the fluid
regions connected to the node.
 The fluid elements and adjacent structure elements must be

compatible.
 It is recommended that the fluid and adjacent structure not share

the coincident nodes. This allows Solution 601 to construct
constraint equations between the fluid and structural degrees of
freedom that are most appropriate during phi model completion,
see Section 2.8.5.
 The fluid material properties are specified via the MAT10 entry.
2.8.3 Fluid boundary conditions
For ease of modeling, fluid boundary conditions of various

types can be defined along the surface of the fluid domain. A fluid
boundary is specified using the BFLUID entry, by referencing a
BSURFS, BCPROPS or BEDGE entry.
There are several types of fluid boundary:
Fluid-structure: Place a fluid-structure fluid boundary on the

boundary between a potential-based fluid and the adjacent
structure.
In many cases, Solution 601 can automatically generate fluid-
structure interface elements along the boundary between the fluid
and structure during phi model completion, step 1, see Section
2.8.5. So fluid-structure boundaries typically need not be defined.
We emphasize that the potential-based fluid elements must be
compatible with the adjacent structural elements.
It is assumed that the structure provides stiffness to all
translational degrees of freedom, because the fluid-structure
boundary does not provide stiffness to the translational directions
that are tangential to the fluid boundary.

Chapter 2: Elements
Free surface: Place a free surface fluid boundary on the

boundary where the pressures are to be prescribed and the
displacements are desired, for example, on the free surface of a
fluid.
It is necessary to fix all displacements that are tangential to the
free surface fluid boundary, because the free surface fluid bondary
does not provide stiffness to the translational directions that are
tangential to the fluid boundary.
In many cases, Solution 601 can generate skew systems and
fixities corresponding to the tangential directions during phi model
generation, see Section 2.8.5.
Fluid-fluid: Place a fluid-fluid boundary on the boundary

between two potential-based fluid elements of two different fluid
regions.
We emphasize that the fluid elements must be compatible
between the two different fluid regions.
Note that only one fluid-fluid boundary need be defined for
each boundary. Solution 601 generates a fluid-fluid interface
element for each of the two elements that share a common
boundary during phi model completion, step 1, see Section 2.8-5.
Rigid-wall: Place a rigid-wall potential interface wherever the

fluid is not to flow through the boundary.
Interface elements
Each fluid boundary type has an associated interface element type.

The difference between fluid boundaries and interface elements
is that a fluid boundary is specified as part of the model definition
on a surface, but an interface element covers only the edge or face
of a single fluid element. Thus a single fluid boundary corresponds
to many interface elements.
Interface elements are automatically generated by Solution 601
as part of the "phi model completion" process.
A rigid-wall fluid boundary suppresses any automatic
generation of interface elements along the rigid-wall boundary.
Solution 601 uses the rigid-wall boundary during phi model
completion, step 2, in constructing structural normals, see Section
2.11.15.

2.8.4 Loads
Concentrated forces, pressure loads, prescribed displacements
Concentrated forces, pressure loads and/or prescribed

displacements can be applied directly to any part of the fluid
boundary on which there are fluid-structure, free surface or fluid-
fluid boundaries. However, when applying concentrated forces,
remember that Solution 601 can apply skew systems to certain
nodes on the fluid boundary during phi model completion, see
Section 2.8.5. Therefore, make sure that the nodes on which you
apply concentrated forces have the anticipated degree of freedom
directions.
Mass-proportional loads
Mass-proportional loads applied to fluid elements are

interpreted as physical body forces (and not, for example, as
ground accelerations). These loads are used in the construction of
 , and therefore these loads must be constant in time.
Centrifugal loads
Centrifugal load effects are not included in the potential-based

fluid elements.
2.8.5 Phi model completion
As can be seen above, there are many restrictions and conditions

that must be considered when specifying boundary conditions on
potential-based fluid elements. These conditions have been
automated in Solution 601 in the following way. Solution 601
performs “phi model completion” before beginning an analysis in
which potential-based fluid elements are used. The steps in phi
model completion are:
1) Solution 601 loops over all fluid element sides on the

boundary of each fluid element region. If the fluid element side
has a fluid boundary, an interface element of the appropriate type is
generated. Otherwise, the side is checked to see if it is attached to
a structure (shares structural degrees of freedom with structural

Chapter 2: Elements
elements) or is close to a structure (nodes coincident with nodes of

a structural element); and, if any of the above conditions are met,
an interface element of the appropriate type is generated. The
intent of this step is to cover as much of the fluid boundary as
possible with interface elements.
2) Solution 601 loops over all nodes attached to interface

elements. If the node is attached to structural elements, the node is
skipped. Otherwise the types of the attached interface elements are
determined. Then
a) If the node is attached only to a free surface interface or fluid-

fluid interface, then the node has a free normal direction (normal to
the interface) and zero stiffness directions that are tangential to the
free normal. The free normal and zero stiffness directions are
identified, and if they are not aligned with the global directions, a
skew system is generated that is aligned with the free normal and
zero stiffness directions.
b) If the node is attached to a free surface interface or fluid-fluid

interface, and is also attached to a fluid-structure interface or rigid-
wall interface, Solution 601 proceeds as follows. The node has a
free normal direction (determined from the free surface interface or
fluid-fluid interface), a structural normal direction (determined
from the fluid-structure interface or rigid-wall interface), and, in
3D, another direction orthogonal to the free normal and structural
normal directions, which may be a zero stiffness direction or
another structural normal direction. The free normal direction is
modified to be orthogonal to the structural normal directions. The
free normal, structural normal and zero stiffness directions are
identified, and, if they are not aligned with the global directions, a
skew system is generated that is aligned with the free normal,
structural normal and zero stiffness directions.
The intent of step 2 is to identify the zero stiffness and free
normal directions of the nodes.
3) Solution 601 loops over all nodes attached to interface

elements. If the node is attached to a structural element, the node is
skipped. If the node (node A) is attached to a fluid-structure
interface element and is close to a structural node B, node A is
constrained to node B as follows. Each displacement degree of

freedom for node A is constrained to the corresponding degrees of

freedom for node B, accounting for differences in skew systems
between A and B, accounting for B possibly being a slave node in a
constraint equation (but not accounting for B possibly being a slave
in a rigid link), accounting for B possibly being fixed. But a
displacement degree of freedom for node A is not constrained if the
degree of freedom is a free normal direction (see 2 above).
The intent of step 3) is to connect the fluid mesh with the
structural mesh, when different nodes are used for the fluid and
structure. The connection still allows the fluid nodes to slip
relative to the structural nodes on intersections between free
surfaces and the structure.
4) In static analysis, when there are no body force loads, Solution

601 loops over all nodes on a free surface or fluid-fluid interface.
If the node is attached to a structural element, the node is skipped.
Otherwise, constraint equations are defined for all nodes so that the
displacements in the direction of the free normal are equal.
The intent is to remove the zero pivots in the stiffness matrix
that are otherwise present (see the discussion after equation (2.8-
10)).
5) When there are body force loads, Solution 601 loops over all
nodes on a fluid-fluid interface. If the node is attached to a
structural element, the node is skipped. Otherwise constraint
equations are defined for all pairs of nodes, so that the
displacements in the direction of the free normal are compatible.
The intent is to enforce displacement compatibility between the
fluids.
6) Solution 601 then loops over all nodes with zero stiffness
degrees of freedom and defines fixities for each zero stiffness
degree of freedom.
7) Solution 601 constrains all of the potential degrees of freedom

for fluid elements with the same PID together.

Chapter 2: Elements
Example 1: We now present a detailed example for a 2-D fluid

filled basin with flexible walls.
Fig. 2.8-3(a) shows the model before phi model completion.
The model is defined with separate nodes for the fluid and the
structure. A fluid boundary of type free-surface is defined on the
horizontal line, as shown.
Fluid boundary of type

free surface is defined on
this line
16 3 2 1 10 Potential-based
Fluid nodes are at the same fluid elements
coordinates as structural nodes.
6 5 4 11
The fluid nodes and structural Beam elements
nodes are separated in this figure 15
for clarity. 9 8 7
30o
14 13 12
a) Finite element model before phi model completion
Fig. 2.8-3: Example of phi model completion
In step 1 of phi model completion, Solution 601 generates fluid-

structure interface elements where the fluid is adjacent to the
structure, and free surface interface elements corresponding to the
fluid boundary (Fig. 2.8-3(b)).
In step 2 of phi model completion, Solution 601 classifies the
displacement directions on the free surface (Fig. 2.8-3(c)). Notice
that the free normal for node 2 is taken from the free surface, but
the free normal for node 3 is modified by the presence of the
adjacent structure. A skew system is defined for node 3 because
the free normal and structural normal are not aligned with the
global coordinate directions. The zero stiffness direction of node 2
will be fixed in step 6 below.

Interface elements
of type free surface
3 2 1
6 4
9 8 7
Interface elements of
type fluid-structure interface
b) Step 1 of phi model completion;

interface elements are created
b z z
x x
c 3 2 1
Node 1: x = structural normal direction,

z = free normal direction
Node 2: x = zero stiffness direction,
z = free normal direction
Node 3: b = free normal direction,
c = structural normal direction
c) Step 2 of phi model completion: classification

of displacement directions on free surface
In step 3 of phi model completion, Solution 601 constrains the

fluid displacement directions to the structure (Fig. 2.8-3(d)). At
node 1, only the x displacement direction is constrained; the z
displacement is left free so that the free surface can slip along the
wall. At node 3, only the c displacement direction is constrained;

Chapter 2: Elements
the b displacement is left free so that the free surface can slip along
the wall.
Notice that at node 4, both the x and z displacements are
constrained to the structure. The fluid still slips in the z direction
because only the normal displacement (the x displacement in this
case) is used by the fluid equations. Similar statements hold for
nodes 6 and 8.
Nodes 7 and 9 are fixed because corresponding nodes 12 and 14
are fixed.
If the analysis is static without body forces, then Solution 601
performs step 4 of phi model completion (Fig. 2.8-3(e)). The free
surface can only translate vertically as a rigid body.
16 3 1 10
6 4 11
15
9 8 7
14 13 12
Node 1: ux=u10
x
o o
Node 3: uc=-cos30 ux16 - sin 30 uz16
Node 4: ux=u11 11
x , uz=uz Node 6: ux=u15 15
x , uz=uz
Node 7: ux=uz =fixed Node 8: ux=u13 13
x , uz=uz
Node 9: ux=uz =fixed
d) Step 3: Creation of constraint equations and fixities

b z z
x x
c 3 2 1
Node 2: uz=u1z
o
Node 3: ub= uz1 / cos30
Step 4 is only performed in static analysis when
there are no body forces.
e) Step 4 of phi model completion: defining constraint

equations to set normal displacements equal on free surface
This motion affects the total volume of the fluid region, so that
there is no zero pivot in the system matrices.
If there are body forces, then step 4 is not necessary because all
boundary motions are given stiffness by the matrix  K UU S .
Step 5 of phi model completion is skipped because there are no
fluid-fluid boundaries.
In step 6 of phi model completion, the zero stiffness direction at
node 2 is fixed (Fig. 2.8-3(f)). Vertical motions of the nodes
attached only to free surface interface elements are allowed, but
horizontal motions of these nodes are not allowed (because the
fluid does not provide stiffness, damping or mass to horizontal
motions).

Chapter 2: Elements
z
x
2
Node 2: ux=fixed
f) Step 6 of phi model completion: defining fixities to

eliminate zero stiffness degrees of freedom
Fig. 2.8-3: (continued) .
Then Solution 601 performs step 7 of phi model completion

(Fig. 2.8-3g). Only constant (in space) potentials are allowed in
static analysis.
3 2 1
6 5 4
9 8 7
f2=f1, ..., f9=f1
g) Step 7: defining constraint equations to set all potential

degrees of freedom equal
Example 2: In 3-D analysis of a fluid-filled basin, there is one

additional consideration. Consider the model shown in Fig. 2.8-4,
in which only the free surface and the adjacent structural nodes are
shown.

y
x
13 14 15
3 Sides of 3-D potential-based

1 2
fluid elements on free surface
16
22
4 5 6 Sides of shell elements
21 17 Fluid nodes are at the same

7 8 9 coordinates as structural nodes.
The fluid nodes and structural
10 11 12 nodes are separated in this figure
for clarity.
20 19 18
a) Top view of finite element model
Fig. 2.8-4: Example 2 of phi model completion
During step 2 of phi model completion, Solution 601 determines

the structural normal direction(s), zero stiffness direction(s) and
free normal direction for the nodes of the free surface (Fig.
2.8-4(b)). Notice that node 3 has two structural normals, but node
6 has only one structural normal. That is because the angle
between the two structural normals for node 3 is greater than 30
degrees, but the angle between the two structural normals for node
6 is less than 30 degrees.
Nodes 3, 6, 9 and 12 are assigned skew systems because the
structural normal directions are not aligned with the global system.

Chapter 2: Elements
n12 n23
1 2 3
n14 n36
4 5 6
n69
7 8 9
y
10 11 12 x
Node 1: Structural normal 1 = n14 Node 2: Structural normal = n12 = n23

Structural normal 2 = n12 Zero stiffness direction = x
Free normal = z Free normal = z
Node 3: Structural normal 1 = n23 Node 5: Zero stiffness direction 1 = x
Structural normal 2 = n36 Zero stiffness direction 2 = y
Free normal = z Free normal = z
Node 6: Structural normal = average of n36 and n69
Free normal = z
Zero stiffness direction = remaining orthogonal direction
b) Classification of structural normals and zero stiffness directions

for some nodes on the free surface
During step 3 of phi model completion, Solution 601 creates

constraint equations for the fluid nodes adjacent to the structural
nodes. For example, node 1 is constrained in both the x and y
directions to node 13, because both directions are structural normal
directions. Node 2 is also constrained in both the x and y directions
to node 14, here because the y direction is a structural normal
direction and the x direction is a zero stiffness direction. (It is
assumed that the structure provides stiffness in the x direction.)
During step 6 of phi model completion, Solution 601 fixes the x
and y directions for nodes 5 and 8, because these directions are
zero stiffness directions, and there is no adjacent structure.

Example 3: We now present some of the steps for phi model

completion of an enclosure with two distinct fluid regions. Fig.
2.8-5(a) shows the model before phi model completion.
Fluid nodes are at the same coordinates as structural nodes.

The fluid nodes and structural nodes are separated in this figure
for clarity.
15 14 13 2-D potential-based
fluid elements, PID 1
3 2 1
Beam elements
6 5 4 Enclosed nodes all lie on

17 16 the same line, fluid boundary
of type fluid-fluid is defined
9 8 7 on this line
2-D potential-based
12 11 10 fluid elements, PID 2
20 19 18
a) Finite element model before phi model completion
Fig. 2.8-5: Example 3 of phi model completion
In step 1 of phi model completion, Solution 601 generates fluid-

structure interface elements where the fluid is adjacent to the
structure, and fluid-fluid interface elements corresponding to the
fluid boundary. Four fluid-fluid interface elements are generated,
two for each shared element side.
In step 2 of phi model completion, Solution 601 classifies the
displacement directions on the fluid-fluid interface. Here the free
normal is always in the z direction and the zero stiffness directions
for nodes 5 and 8 are in the x direction.
In step 3 of phi model completion, the fluid nodes are
constrained to the adjacent structural nodes (Fig. 2.8-5(b)). Notice
that the nodes on the fluid-fluid interface are allowed to slip
relative to the structure.

Chapter 2: Elements
z
15 14 13 x
3 2 1
6 5 4
17 16
9 8 7
12 11 10
20 19 18
Node 1: ux=u13 13
x , uz=uz Node 2: ux=u14 14
x , uz=uz
Node 3: ux=u15 15
x , uz=uz Node 4: ux=u16
x
Node 6: ux=u17
x Node 7: ux=u16
x
Node 9: ux=u17
x Node 10: ux=u18 18
x , uz=uz
Node 11: ux=u19 19 20 20

x , uz=uz Node 12: ux =ux , uz=uz
b) Step 3: constraining fluid nodes to adjacent structural nodes
If the analysis is static without body forces, then Solution 601

performs step 4 of phi model completion. In this case, the z
displacements of nodes 5 to 9 are constrained to be equal to the z
displacement of node 4. The free surface can only translate
vertically as a rigid body.
In step 5 of phi model completion, the fluid nodes on the fluid-
fluid interface are constrained to each other (Fig. 2.8-5(c)). Step 5
is not performed if step 4 was performed. Notice that the potential
degrees of freedom are not constrained.

15 14 13
3 2 1
6 5 4
17 16
9 8 7
12 11 10
20 19 18
Node 4: uz=u7z Node 5: uz=u8z

Node 6: uz=u9z
c) Step 5: constraining adjacent nodes of the two fluid regions together
In step 6 of phi model completion, the zero stiffness direction of

nodes 5 and 8 are fixed. Vertical motions of the nodes attached
only to fluid-fluid interface elements are allowed, but horizontal
motions are not allowed (because the fluid does not provide
stiffness, damping or mass to horizontal motions).
Then Solution 601 performs step 7 of phi model completion
(Fig. 2.8-5(d)). Only constant (in space) potentials are allowed in
static analysis, but the potential can be different for the regions
with different PIDs.

Chapter 2: Elements
15 14 13
3 2 1
6 5 4
17 16
9 8 7
12 11 10
20 19 18
f2=f1, ..., f6=f1 f8=f7, ..., f12=f7
d) Step 7: defining constraint equations to set the potential

degrees of freedom of each fluid region equal
2.9 Other element types

2.9.1 Gap element
 The gap element is used in Advanced Nonlinear Solution to

connect two nodes as shown in Fig. 2.9-1. Gap elements are
defined using the CGAP element connectivity entry.
 The initial gap opening is U0. When the gap is closed the
element has a stiffness of KA (should be stiff), and when it is open
the stiffness is KB (should be soft).
 The tangential behavior of the gap element represented by KT,

MUI and MUZ is not supported.

2.9: Other element types
t
s
GA
r U0
KB
KA - KB
GB
Fig. 2.9-1: CGAP element coordinate

system
2.9.2 Concentrated mass element
 Advanced Nonlinear Solution supports the CONM1 and

CONM2 entries for defining concentrated masses.
 For CONM1, only the diagonal mass terms are supported, and
the resulting mass matrix is given by:
 M 11 0 0 0 0 0 
 0 M 22 0 0 0 0 

 0 0 M 33 0 0 0 
M 
 0 0 0 M 44 0 0 
 0 0 0 0 M 55 0 
 
 0 0 0 0 0 M 66 
 For CONM2, the off-diagonal mass moments of inertia terms

are neglected, and the resulting mass matrix is

Chapter 2: Elements
M 0 0 0 0 0
0 M 0 0 0 0 

0 0 M 0 0 0
M 
0 0 0 I11 0 0
0 0 0 0 I 22 0
 
 0 0 0 0 0 I 33 
2.9.3 Bushing element
 The one-dimensional bushing element CBUSH1D is used in

Advanced Nonlinear Solution to provide an axial stiffness and
damping between two nodes as shown in Fig. 2.9-2.
GA GB
Fig. 2.9-2: BUSH1D element
 The stiffness and damping act along the axis of the element,
which is the line connecting its two nodes. In large displacement
analysis the element axis is updated with deformation. A fixed
element axis can be specified via the CID parameter in the
CBUSH1D entry.
 The element can have a constant or a nonlinear stiffness defined

via a lookup table.
 Any mass assigned to the BUSH1D element is lumped at the

grid points.

3: Material models and formulations
3. Material models and formulations

The objective of this chapter is to summarize the theoretical basis and practical
use of the material models and formulations available in Advanced Nonlinear
Solution.
The stress and strain measures used by different materials and formulations
are first summarized in Section 3.1.
The table below lists the material models available in Advanced Nonlinear
Solution, and how they can be obtained from the material entry cards. Note that
Tables 2-2 and 2-3 list the acceptable combinations of elements and material
properties for Solutions 601 and 701.
Table 3.1: Material models available in Advanced Nonlinear Solution
Material Entries Advanced Nonlinear Solution Sol 701

material availability1
MAT1 Elastic isotropic 
MAT1, CREEP or Elastic-creep
MAT1, MATCRP
MAT1, CREEP, MATTC or Thermal elastic-creep
MAT1, MATCRP
MAT1, MATG Gasket
2
MAT1, MATS1 Elastic isotropic nonlinear3 
MAT1, MATS14, MATSR Elasto-plastic 
6
MAT1, MATS1 Thermal elasto-plastic 
MAT1, MATS1, MATT16 Thermal elasto-plastic, 
temperature-dependent elastic
properties
MAT1, MATS15 Thermal elasto-plastic, 
temperature-dependent plastic
properties
MAT1, MATS1, MATT15 Thermal elasto-plastic, 
temperature-dependent elastic
and plastic properties

Chapter 3: Material models and formulations

(continued)

MAT1, MATS1, Plastic-creep
CREEP6
MAT1, MATS1, CREEP, Plastic-creep with temperature-
MATT16 dependent properties
MAT1, MATS1, Thermal plastic-creep, temperature-
CREEP5 dependent plastic properties
MATT1, CREEP5 dependent elastic and plastic
properties
MAT1, MATS1, CREEP, Thermal plastic-creep, temperature-
MATTC6 dependent creep properties
MATT1, CREEP, dependent elastic and creep
MATTC6 properties
MAT1, MATS1, CREEP, Thermal plastic-creep, temperature-
MATTC5 dependent plastic and creep
properties
MATT1, CREEP, dependent elastic, plastic and creep
MATTC5 properties
MAT2 Elastic orthotropic (surface 
elements)
MAT2, MATT2 Thermal elastic orthotropic (surface 
elements)
MAT3 Elastic orthotropic (2D elements)
MATT3, MAT3 Thermal elastic orthotropic (2D
elements)
MAT4 Isotropic heat transfer

3: Material models and formulations

(continued)

MAT4, MATT4 Temperature dependent isotropic
heat transfer
MAT5 Orthotropic heat transfer
MAT5, MATT5 Temperature dependent orthotropic
heat transfer
MAT8 Elastic orthotropic (surface 
elements)
MAT8, MATT8 Thermal elastic orthotropic (surface 
elements)
MAT9 Elastic orthotropic (solid elements) 
MAT9, MATT9 Thermal elastic orthotropic (solid 
elements)
MAT10 Fluid
MAT11 Elastic orthotropic (solid elements) 
MAT11, MATT11 Thermal elastic orthotropic (solid 
elements)
MATHE, MATHEM, Hyperelastic (Mooney-Rivlin, 7
MATHEV Ogden, Arruda-Boyce, Sussman-
Bathe and Hyperfoam)
MATHP Hyperelastic (Mooney-Rivlin only) 
MATPLCY, PLCYISO, Plastic-cyclic 
PLCYKIN, PLCYRUP


(continued)

MATSMA Shape memory alloy (SMA)
MATVE Viscoelastic
Notes:
1. Temperature interpolation at the start of the analysis only in Solution 701.
2. With MATS1 TYPE=NELAST.
3. Cannot be used with beam element for SOL 601. Can only be used with rod element
for SOL 701.
4. With MATS1 TYPE=PLASTIC.
5. With MATS1 TYPE=PLASTIC and TID pointing to a TABLEST entry.
6. With MATS1 TYPE=PLASTIC and TID pointing to a TABLES1 entry.
7. Only Mooney-Rivlin, Ogden and Sussman-Bathe hyperelastic materials are available
in Solution 701.

3.1: Stress and strain measures
3.1 Stress and strain measures

 It is important to recognize which stress and strain measures are
employed in each material model: this is necessary in the
preparation of the input data and the interpretation of the analysis
results.
 This section summarizes the stress and strain measures in

Advanced Nonlinear Solution and how they are used with the
different element types and nonlinear features. More details on
stress/strain measures are provided in ref. KJB, Section 6.2.
3.1.1 Kinematic formulations
Small displacement/small strain kinematics
Input of material parameters: All elements and material models

use the engineering stress-engineering strain relationship.
Output: All elements and material models output engineering

stresses and engineering strains. Note that, as long as the
displacements and strains are small, Cauchy stresses and
engineering stresses are nealy equal.
 Using a linear material model with small displacement/small

strain kinematics results in a linear finite element formulation.
 Using a nonlinear material model with small displacement/small

strain kinematics results in a materially-nonlinear only (MNO)
formulation.
Large displacement/small strain kinematics
Input of material parameters: 2nd Piola-Kirchhoff stresses and

Green-Lagrange strains. Note that under small strain conditions, 2nd
Piola-Kirchhoff stresses are nearly equal to engineering stresses,
and Green-Lagrange strains are nearly equal to engineering strains.
Strains should be less than 2%.

Output: The output depends on the element type. Note that as long
as the strains are small, Green-Lagrange strains are practically the
same as engineering strains in the element coordinate system.
Similarly, 2nd Piola-Kirchhoff stresses are practically the same as
Cauchy stresses in the element coordinate system.
(1) 2-D, 3-D solid elements: all supported material models

output Cauchy stresses and Green-Lagrange strains.
(2) Shell elements: all supported material models output 2nd

Piola-Kirchhoff stresses and Green-Lagrange strains.
(3) Rods and beams: all supported material models output

Cauchy stresses and engineering strains in the element
coordinate system.
Large displacement/large strain kinematics
This kind of formulation can only be used with 2-D and 3-D solid
elements and with shell elements.
For 2-D and 3-D solid elements
(1) Both the updated Lagrangian Hencky formulation and the

updated Lagrangian Jaumann formulation can be used with
elastic-plastic materials (including thermal and creep effects). In
this case,
Input of material parameters: Cauchy (true) stresses and

logarithmic (true) strains. For the multilinear stress-strain
curves, it is also possible to enter engineering stress-strain
data along with the input NXSTRAT CVSSVAL=1, see
detailed description of the CVSSVAL feature in Section
3.4.1.
Output:
ULH formulation: Cauchy stresses and logarithmic
strains in the element coordinate system.
ULJ formulation: Cauchy stresses and Jaumann strains.

(2) For hyperelastic materials a total Lagrangian formulation is

used. In this case,
Input of material parameters: Hyperelastic material

constants.
Output: Cauchy stresses and Green-Lagrange strains in the

element coordinate system.
For shell elements
Both the updated Lagrangian Jaumann (ULJ) formulation and the

updated Lagrangian Hencky (ULH) formulation can be used. For
more details on how these formulations apply to shell elements, see
Section 2.3.
Input of material parameters: Cauchy (true) stresses and

logarithmic (true) strains. For the multilinear stress-strain
curves, it is also possible to enter engineering stress-strain
data along with the input NXSTRAT CVSSVAL=1, see
detailed description of the CVSSVAL feature in Section
3.4.1.
When the ULJ formulation is used:
Output: Cauchy stresses and Jaumann strains.
When the ULH formulation is used:
Output: Kirchhoff stresses and left Hencky strains

(practically equivalent to Cauchy stresses and logarithmic
strains).
3.1.2 Strain measures
The strain measures used in Advanced Nonlinear Solution are

illustrated here in the simplified case of a rod under uniaxial
tension (see Fig. 3.1-1).

A
´
A0 D
A0 = initial cross-sectional area
S = final cross-sectional
F = applied force
0 = 0 +D
Fig. 3.1-1: Rod under uniaxial tension
  0
Engineering strain: e0 
0
1  2   20
Green-Lagrange strain: 
2  20
Logarithmic strain, Hencky strain, Jaumann strain:
     d 
e  ln      
  0    0  

Stretch: 
0
 Note that for the small strains assumption to be valid, the strains
should be less than about 2%.

ref. KJB  Green-Lagrange strains are used in the large displacement/small
Sec. 6.2.2 strain formulations. This is because this strain measure is invariant
with respect to rigid-body rotations. Therefore, for small strains,
Green-Lagrange strains and the rotated engineering strains are

equivalent.
 Engineering strains are also called nominal strains in the

literature.
 Logarithmic strains are also known as true strains.
3.1.3 Stress measures
The stress measures used in Advanced Nonlinear Solution include

engineering stresses, 2nd Piola-Kirchhoff stresses, Kirchhoff
stresses, and Cauchy stresses (see ref. KJB). These stress measures
are illustrated here in the simplified case of a rod under uniaxial
tension (see Fig. 3.1-1).
F
Engineering stress: 
A0
F A
Cauchy stress:   0
A A
F0  0
2nd Piola-Kirchhoff stress: S  
A0  
F  
Kirchhoff stress: J  
A0  0  0
 Cauchy stresses are also called true stresses in the literature.
 For the case in which the material is incompressible,


  J  can be used to compute the Cauchy stress and the
0
Kirchhoff stress from the engineering stress.
 When the strains are small, the 2nd Piola-Kirchhoff stresses are
nearly equal to the Cauchy stresses from which the rigid body
rotations of the material have been removed.
 When the volume change of the material is small, the Kirchhoff

stresses are nearly equal to the Cauchy stresses.

 Since Kirchhoff stresses are input/output only for large strain

analysis with materials that are nearly incompressible, practically
speaking, the differences between Kirchhoff and Cauchy stresses
are negligible.
3.1.4 Large strain thermo-plasticity analysis with the ULH formulation
 This section discusses the ULH formulation for large strain

analysis. ULH stands for updated Lagrangian Hencky.
 The following is a quick summary of the theory of large strain

inelastic analysis with the ULH formulation. For further
information, see ref KJB, Section 6.6.4 and also the following
references:
ref. F.J. Montáns and K.J. Bathe, "Computational issues in

large strain elasto-plasticity: an algorithm for mixed
hardening and plastic spin", Int. J. Numer. Meth. Engng,
2005; 63;159-196.
ref. M. Kojić and K.J. Bathe, Inelastic Analysis of Solids and

Structures, Springer-Verlag, 2003.
Total deformation gradient tensor: Let X be the total

deformation gradient tensor at time t with respect to an initial
configuration taken at time 0. For ease of writing, we do not
include the usual left superscripts and subscripts.
Polar decomposition into rotation and right stretch tensor: The

total deformation gradient tensor X can be decomposed into a
material rigid-body rotation tensor R and a symmetric positive-
definite (right) stretch tensor U (polar decomposition):
X  RU (3.1-1)
Principal directions of right stretch tensor: The right stretch

tensor U can be represented in its principal directions by a diagonal
tensor Λ , such that
U  R L Λ R TL (3.1-2)

where R L is a rotation tensor with respect to the fixed global axes

(see Fig. 3.1-2).
y R
(t) RL
(0)
x
(0) Initial configuration at time 0

Configuration at time t not including rigid body rotation
(t) Configuration at time t
Directions of maximum/minimum total stretches
and strains
R Material rigid-body rotation between time 0 and time t
R L Directions of initial configuration fibers with

maximum/minimum total stretches and strains
Fig. 3.1-2: Directions of maximum/minimum total stretches and

strains
(Note that the rotation R L does not correspond to a material

rigid-body rotation, but to a rotation of the coordinate system: U
and Λ are two representations of the same deformed state,
respectively in the basic coordinate system and in the U principal
directions coordinate system.)
Right Hencky strain tensor: The Hencky strain tensor (computed

in the right basis) is given by
E R  ln U  R L ln Λ R TL (3.1-3)
The superscript “R” symbolizes the right basis.

Polar decomposition into rotation and left stretch tensor: The

total deformation gradient tensor X can also be decomposed into a
material rigid-body rotation R and a symmetric positive-definite
(left) stretch tensor V (polar decomposition):
X VR (3.1-4)
R in (3.1-4) is the same as R in (3.1-1).
Principal directions of left stretch tensor: The left stretch tensor

V can be represented in its principal directions by a diagonal tensor
Λ , such that
V  R E Λ R TE (3.1-5)
where R E is a rotation tensor with respect to the fixed global axes.

Note that R E  R R L .
Left Hencky strain tensor: The Hencky strain tensor (computed

in the left basis) is given by
E L  ln V  R E ln Λ R TE (3.1-6)
The superscript “L” symbolizes the left basis.
Comparison of left and right Hencky strain tensors: The

principal values of the left and right Hencky strain tensors are
identical, and equal to the logarithms of the principal stretches.
Hence both of these strain tensors can be considered to be
logarithmic strain tensors. However, the principal directions of the
left and right Hencky strain tensors are different. The principal
directions of the right Hencky strain tensor do not contain the rigid
body rotations of the material, but the principal directions of the
left Hencky strain tensor contain the rigid body rotations of the
material.
Therefore, for a material undergoing rigid body rotations, the
principal directions of the right Hencky strain tensor do not rotate,
however the principal directions of the left Hencky strain tensor
rotate with the material. Hence, the left Hencky strain tensor is
preferred for output and visualization of the strain state.

Multiplicative decomposition of deformation gradient in

inelastic analysis: In inelastic analysis, the following
multiplicative decomposition of the total deformation gradient into
an elastic deformation gradient X E and an inelastic deformation
gradient X P is assumed:
X  XE XP (3.1-7)
To understand (3.1-7), consider a small region of material under a

given stress state with deformation gradient X. If this region of
material is separated from the rest of the model and subjected to the
same stress state, the deformation gradient is still X. Now if the
stress state is removed, (3.1-7) implies that the deformation
gradient of the unloaded material is X P . The stresses are due
entirely to the strains associated with the elastic deformation
gradient X E .
It can be shown (see Montáns and Bathe), that (3.1-7) is
equivalent to the additive decomposition of the displacements into
elastic displacements and plastic displacements.
For the materials considered here, det X P  1 .
Polar decomposition of elastic deformation gradient: The elastic

deformation gradient can be decomposed into an elastic rotation
tensor R E and elastic right and left stretch tensors U E , V E :
XE  R E U E  V E R E (3.18-a,b)
Elastic Hencky strain tensors: The elastic Hencky strain tensors

in the right and left bases are given by
E ER  ln U E , E EL  ln V E (3.1-9a,b)
Stress-strain relationships: The stresses are computed from the

elastic Hencky strain tensors using the usual stress-strain law of
isotropic elasticity. However, the stress measures used depend upon
the strain measures used. When the right Hencky strain measure is
used, the stress measure used is the rotated Kirchhoff stress
τ   R E  Jτ R E
T
(3.1-10)

and when the left Hencky strain measure is used, the stress measure
is the (unrotated) Kirchhoff stress Jτ . J  det X is the volume
change of the material, and, using det X P  1, J  det X E .
With these choices of stress and strain measures, the stresses
and strains are work-conjugate.
The choice of right Hencky strain and rotated Kirchhoff stresses
gives the same numerical results as the choice of left Hencky strain
and (unrotated) Kirchhoff stresses.
Implementation notes: For 2-D and 3-D solid elements, the

difference between the Cauchy and Kirchhoff stresses is neglected.
The stress measure used with the right Hencky strains is
τ   R E  τ R E . The input of material properties is assumed to be
T
in terms of Cauchy stresses, and the output of stresses is in terms of

Cauchy stresses.
For shell elements, Kirchhoff stresses are used throughout. The
input of material properties is assumed to be in terms of Kirchhoff
stresses, and the output of stresses is in terms of Kirchhoff stresses.
These assumptions are justified because they are used with
material models in which the plastic deformations are
incompressible and the plastic deformations are generally much
larger than the elastic deformations.
3.1.5 Large strain thermo-plasticity analysis with the ULJ formulation
 This section discusses the ULJ formulation for large strain

inelastic analysis (ULJ formulation). ULJ stands for updated
Lagrangian Jaumann.
 The following is a quick summary of the theory of large strain

inelastic analysis with the ULJ formulation:
For further information, see ref KJB, Section 6.2.2 and also the
following reference:
ref. M. Kojić and K.J. Bathe, Inelastic Analysis of Solids and

Structures, Springer-Verlag, 2003.

Velocity gradient tensor: The velocity gradient tensor is defined

as
  t u 
L   t i   XX 1 (3.1-11)
  x 
j 
Notice that the derivative is taken with respect to the current

coordinates.
Rate of deformation tensor, spin tensor: The rate of deformation

tensor is defined as
1
D
2
 L  LT  (3.1-12)
and the spin tensor is defined as
1
W
2
 L  LT  (3.1-13)
D is the symmetric part of L and W is the skew-symmetric part of

L.
Rate of change of Jaumann strain tensor: The rate of change of

the Jaumann strain is defined as
ε J  D Wε J  ε J W (3.1-14)
The quantity ε J is termed the Jaumann strain in analogy with the

more often-used Jaumann stress. But we do not use the Jaumann
stress in the ULJ formulation.
Jaumann strain tensor: In practice, increments are used in

computing the Jaumann strain tensor, i.e.
t t
ε J  t ε J   Dt   Wt  t ε J  t ε J  Wt  (3.1-15)

Comparison of Jaumann strain with left Hencky strain: When

the rate of change of the principal directions of the left stretch
tensor V is zero, the rate of change of the left Hencky strain is the
same as the rate of change of the Jaumann strain. Hence the
Jaumann strain can be used as an approximate replacement for the
left Hencky strain. The Jaumann strain can be computed more
efficiently than the left Hencky strain, because it is not necessary to
take the square root or logarithm of a tensor when computing the
Jaumann strain. On the other hand, the time step size affects the
Jaumann strain, so that finite time step sizes lead to an error in the
calculation of the Jaumann strain.
For a uniaxial deformation, the Jaumann strain approaches the
logarithmic strain as the step size is reduced. For a rigid-body
rotation, the Jaumann strain also rotates, with the rotation of the
Jaumann strain approaching the expected rotation as the step size is
reduced.
It can also be shown that the Jaumann strain is path-dependent
in general, so that a deformation history in which the final
deformations equal the initial deformations can produce (non-
physical) non-zero Jaumann strains, even in the limit of
infinitesimally small time steps.
Stress-strain relationships: In elasto-plasticity, the same

algorithms are used as in small-strain elasto-plasticity. The
mechanical strains are computed as the total strains minus the
plastic strains (and also any thermal strains), in which the total
strains are the Jaumann strains.
As in the ULH formulation, the stresses are Cauchy stresses for
2-D / 3-D elements, and are Kirchhoff stresses for shell elements.
3.1.6 Thermal strains
 Calculation of thermal strains is needed for temperature-

dependent material models (thermo-elastic isotropic, thermo-elastic
orthotropic, thermo-plastic), as well as temperature-invariant
material models with non-zero thermal expansion coefficients.
 The current temperature

t
 and the initial temperature
0
 (corresponding to zero thermal strains) are both needed for the
calculation of thermal strains. The current temperature field is set

via the TEMPERATURE(LOAD) case control entry, while the

initial temperature is set via the TEMPERATURE(INITIAL) case
control entry. See Section 5.6 for more details.
 The temperature at an integration point is evaluated based on

the nodal temperatures and the element shape functions, and then
used to calculate the thermal strains.
 For isotropic temperature independent materials, the following

expression is used for thermal expansion.
t TH
e ij    t  0   ij (3.1-16)
where  ij is the Kronecker delta (  ij = 1 for i = j and  ij = 0 for

i  j ).
 If the thermal expansion is temperature dependent and isotropic,

the thermal strains are calculated as follows:
t TH
ije  t  t  0   ij (3.1-17)
where
t


1
  
t 0 
  t  t   REF     0  0   REF  
(3.1-18)
and  REF is the material reference temperature.
 For temperature independent orthotropic materials Eq. (3.1-16)

is replaced by a thermal expansion coefficient vector,
t TH
e
ij   i  t  0   ij (no summation over i ) (3.1-19)
 For temperature dependent orthotropic materials Eq. (3.1-17)

and Eq. (3.1-18) are modified for each direction similar to Eq.
(3.1-19).

 Equations (3.1-17) and (3.1-18) are derived as follows: Suppose

that, from experimental data, the dependence of the length of a bar
as a function of temperature is obtained, as shown in Fig. 3.1-3.
L
Secant to curve
Length, L
LREF
qREF q
Temperature, q
Fig. 3.1-3: Length of bar vs. temperature
The thermal strain with respect to the reference length may be

calculated as
L  LREF
 TH 
LREF
Then we define the mean coefficient of thermal expansion for a

given temperature as follows:
 TH  
   
   REF
With this definition, the secant slope in Fig. 3.1-3 is LREF    .

Now, in Solution 601, we assume that the thermal strains are
initially zero. To do this, we subtract the thermal strain
corresponding to 0 to obtain
t
 TH    t  t   REF     0  0   REF 
which leads to Equations (3.1-17) and (3.1-18).

Notice that if the mean coefficient of thermal expansion is

constant,  REF no longer enters into the definition of t and the
equations reduce to Eq. 3.1-16. In general, when the mean
coefficient of thermal expansion is not constant,  REF must be
chosen based on knowledge of the experiment used to determine
   since for the same material curve, different choices of  REF
yield different values of    .
3.2 Linear elastic material models

 The following material models are discussed in this section:
Elastic-isotropic: isotropic linear elastic non-thermal dependent

material model obtained with MAT1
Elastic-orthotropic: orthotropic linear elastic non-thermal

dependent material model obtained with MAT2 and MAT8 for
surface elements and MAT9 and MAT11 for 3-D solid elements
 These models can be employed using small displacement/

small strain or large displacement/small strain kinematics. The
strains are always assumed to be small.
 Thermal strains are supported for the elastic isotropic materials

and the elastic-orthotropic materials.
 When the elastic-isotropic and elastic-orthotropic materials are

used with the small displacement formulation, the formulation is
linear.
 In the small displacement formulation, the stress-strain

relationship is
t
0 σ  C t0e
in which t0 σ = engineering stresses and t0 e = engineering strains.

 In the large displacement formulation used by 2-D solid, 3D-

solid and shell elements, the stress-strain relationship is
ref. KJB
Section 6.6.1 t
0 S  C t0ε
in which t0 S = second Piola-Kirchhoff stresses and t0 ε = Green-

Lagrange strains.
 In the large displacement formulation used by rod and beam

elements, the stress-strain relationship is
t
τ  C tt ε*
in which t τ = Cauchy stresses and tt ε* = rotated engineering

strain.
 In the presence of thermal strains the following stress-strain

relationship is used instead in small displacement analysis:
t
0 σ  C  t0 e  t0eTH 
where t0 eTH are the thermal strains. A similar t0 εTH or tt ε*TH term is
added for the large displacement formulations. The calculation of
thermal strain is detailed in Section 3.1.6.
 The same matrix C is employed in all of these formulations. As

long as the strains remain small, the difference in the responses is
negligible.
 However, if the strains are large, the difference in the response

predictions is very significant (see ref. KJB, pp 589-590). If the
strains are large, it is recommended that these linear elastic material
models not be used.
3.2.1 Elastic-isotropic material model
 This material model is available for the rod, 2-D solid, 3-D
solid, beam, and shell type elements.

3.2: Linear elastic material models
 The two material constants used to define the constitutive

relation (the matrix C) are
E = Young's modulus, v = Poisson's ratio
 The thermal expansion coefficient  is also used if thermal

strains are present.
3.2.2 Elastic-orthotropic material model
 The elastic-orthotropic material model is available for the 2-D

solid, 3-D solid and shell elements.
 The thermal expansion coefficient  is also used if thermal

strains are present.
3-D solid elements: The orthotropic 3-D material is defined either

using the MAT9 entry or the MAT11 entry. When the MAT9 entry
is used, the following assumptions are made:
C14 = C15 = C16 = C24 = C25 =C26 = C34 = C35 = C36 = 0
and
C45 = C46 = C56 = 0.
resulting in
 11   C11 C12 C13 0 0 0   e11 

   C22 C23 0 0 0   e22 
 22  
 33   C33 0 0 0   e33 
   
 12   C44 0 0   12 
 23  symmetric C55 0    23 
    
 31   C66    31 

When the MAT11 entry is used, the following inverse stress-strain

relationship is used:
 e11   1/ E1 12 / E1 13 / E1 0 0 0   11 

 e    / E 1/ E2  23 / E2 0 0 0   22 
 22   21 2
 e33    31 / E3  32 / E3 1/ E3 0 0 0   33 
   
 12   0 0 0 1/ G12 0 0   12 
  23   0 0 0 0 1/ G23 0   23 
    
  31   0 0 0 0 0 1/ G13   31 
The MATCID entry can be used to define the material coordinate

system (when using either the MAT9 or MAT11 entries).
Shell elements: The orthotropic shell material is preferably defined

using the MAT8 entry, which leads to the following inverse stress-
strain relationship defined in the shell material coordinate system
(1,2,3):
 e11   1/ E1 12 / E1 0 0 0   11 

 e    / E 1/ E2 0 0 0   22 
 22   21 2
 12    0 0 1/ G12 0 0   12 
    
 13   0 0 0 1/ G1z 0   13 
  23   0 0 0 0 1/ G2 z    23 
The MAT2 entry can also be used to define a shell material with
only in-plane orthotropy:
 11   C11 C12 0 0 0  e11 

   C 0  e22 
 22   12 C22 0 0
 12    0 0 C33 0 0   12 
    
 13   0 0 0 C33 0   13 
 23   0 0 0 0 C33    23 

3.2: Linear elastic material models
2-D solid elements: The orthotropic 2-D material is defined using

the MAT3 command.
For the axisymmetric element, the stress-strain relationship
defined in the (x,θ,z) plane is
 ex   1  x  zx   x 
     0  
   Ex E Ez
 
    x 1  z  
 e     0    
    Ex E Ez
 
    xz  z 1  
 ez     0   z 
   Ex E Ez  
   1  
  zx   0 0 0   
   Gzx   zx 
For the plane stress element, the stress-strain relationship defined in

the (x,y,z) plane, with the y direction as the transverse direction, is:
 1  zx 
 ex    0   x 
   Ex Ez  
    1  
 ez     xz 0   z 
   Ex Ez  
   1  
  zx   0 0   
 Gzx   zx 
For the plane strain element, the stress-strain relationship defined in

the (x,y,z) plane, with the y direction as the transverse direction, is:
 1   yx  xy  zx   yx zy 
 ex    0   x 
   Ex Ez  
      1   yz  zy  
 ez     xz yx xy
0   z 
   Ex Ez  
   1  
  zx   0 0   
 Gzx   zx 

All elements when Young's moduli, Poisson's ratios and shear

moduli are specified: The material constants must be defined so
that the stress-strain constitutive matrix is positive-definite. The
conditions are illustrated for the MAT11 entry. In this case, the
following conditions must be satisfied:
1
 E 2
 ji   j  , i, j  1, 2,3
 Ei 
 E1 E E 
 21 32 13  0.5 1  212  322 2  132 3   0.5
 E2 E3 E1 
Based on the input values for  ij , the remaining constants  ji are

calculated so as to have a symmetric constitutive matrix; i.e.,
 ji  ij

Ej Ei
3.3 Nonlinear elastic material model

 Advanced Nonlinear Solution supports the nonlinear elastic
material for the rod, 2-D solid, 3-D solid and shell elements. The
nonlinear effect is obtained with a MATS1 entry which has
TYPE = ‘NELAST’. The formulations used for the rod element are
slightly different (and simpler) and are detailed in Section 3.3.1.
 This material uses a nonlinear elastic uniaxial stress-strain data

input in tabular form and shown in Fig. 3.3-1. This material is not
based on the classical theory of finite elasticity, and is not intended
for large strain analysis. However, it is a useful material model
when used appropriately, and with awareness of its limitations.

3.3: Nonlinear elastic material model
Fig. 3.3-1: Stress-strain behavior of nonlinear

elastic material model
 Note that the material unloads along the same curve, so that no
permanent inelastic strains are introduced.
 The material can have different stress-strain curves in tension

and compression. Under predominantly uniaxial tension or
compression, the material response will follow the input curve
exactly. Under shear dominated loading, the stress is interpolated
from both tension and compression parts of the material stress-
strain curve.

 In order to use the unaxial stress-stain data  ( ) of Fig. 3.3-1,

the effective stress and strain (  and  ) must be calculated based
on the 2-D or 3-D total stress and strain tensors ( σ and ε ). The
von-Mises stress is used as the effective stress, while the effective
strain is based on
  d   σ
T
dε (3.3-1)
which equates the deformation work per unit volume in unaxial

loading to the multi-dimensional state. This results in a unique
equation for  as a function of ε ,  and the stress-strain state that
depends on the element type.
 The effective strain,  , is defined by
1 1
E0 2  εT C0ε (3.3-2)
2 2
where E0 is Young’s modulus which is determined by the most

stiff region of the input stress-strain curve, C0 is the elastic stress-
strain matrix obtained using E0 and  . ( E0 cancels out from both
sides of Eq. (3.3-2))
Differentiating Eq. (3.3-2) with respect to the total strain, we
have
1 T
d  ε C0 dε (3.3-3)
E0
Substituting Eq. (3.3-3) into Eq. (3.3-1), the stresses can be

expressed in terms of total strains, i.e.,

σ C0ε (3.3-4)
E0
or


σ σe (3.3-5)
E0
where σ e  C0 ε which is the elastic trial stress.
 The effective stress  is taken from the tensile part of the

stress-strain curve for predominantly tensile loading, from the
compression part of the stress-strain curve for predominantly
compression loading and is interpolated between the two curves
otherwise.
 The consistent tangent stress-strain matrix is obtained by

differentiating Eq. (3.3-4) or (3.3-5) with respect to the total strain
tensor. The stress-strain matrix is symmetric in predominant tensile
or compression loading (when only one of the two material curves
is used), and is non-symmetric otherwise (when interpolation
between the curves is required). The constitutive matrix is
symmetrized and in most cases reasonable convergence is obtained.
 Note that discontinuities are not allowed in the user-supplied

stress-strain curve. The table look-up is performed using linear
interpolation within the table and linear extrapolation outside the
table using the two starting or ending points.
Stress update algorithm
For an iteration i , given t σ , t ε , t t

u (i ) , E0 ,  , update t t
σ (i ) ,
t t
ε (i )
t t
Step 1. Calculate the new total strain state ε (i ) based on
displacements t t u (i )
Step 2. Calculate the elastic trial stress,
t t
σ e  C0 t t ε (3.3-6)
Step 3. Compute the magnitude of the effective strain,  .

Step 4. Calculate the ratio
I1e
r C (3.3-7)
t t
e
where C is a constant that biases the general stress state towards

the pure tension or compression curves and is internally set to 3/2,
I1e is the first elastic stress invariant, and t t  e is the effective
elastic stress which is calculated as follows,
t t
 e  E0 t t  (3.3-8)
Restrict r to be between -1 and 1.
Step 5. Calculate the effective stress in tension  t and in

compression  c , based on the user-supplied stress-strain curve and
 , as follows:
t t
 t t t  (t t  ) (3.3-9)
t t
 c  t t  (t t  ) (3.3-10)
Step 6. Calculate the actual effective stress, t t  , as
1  r t t 1  r t t
t t
 t  c (3.3-11)
2 2
Step 7. Evaluate the new stress state by
t t
t t
 t t
σ σe (3.3-12)
E0 t t

Step 8. Evaluate the tangential stress-strain matrix and symmetrize

it.

3.3.1 Nonlinear elastic material for rod element
 For the rod element, the stress-strain relationship is defined as a

piecewise linear function, as shown in Fig. 3.3-2.
Stress
s6 l
s5 l
s4 l
e1 e2 e3
s3 e4
l
e5 e6 Strain
l s2
l s1
Fig. 3.3-2: Nonlinear elastic material for rod elements
Note that the stress is uniquely defined as a function of the

strain only; hence for a specific strain te, reached in loading or
unloading, a unique stress is obtained from the curve in Fig. 3.3-2.
 A sufficient range (in terms of the strain) must be used in the

definition of the stress-strain relation so that the element strain
evaluated in the solution lies within that range; i.e., referring to Fig.
3.3-2, we must have e1  t e  e6 for all t.
 The stress-strain curve does not necessarily have to pass through

the origin.
 A typical example of the nonlinear elastic model for rod

elements is shown in Fig. 3.3-3. This example corresponds to a
cable-like behavior in which the rod supports tensile but no
compressive loading.

Stress
Point 3
´
Point 1 Point 2
´ ´
Strain
Fig. 3.3-3: Nonlinear elastic material model corresponding

to a tension-only cable
 The rod element with this nonlinear elastic material model is

particularly useful in modeling gaps between structures. This
modeling feature is illustrated in Fig. 3.3-4. Note that to use this
element to simulate a contact gap, it is necessary to know which
node of one body will come into contact with which node of the
other body, and then to connect these two nodes with a rod
element.
n
l
Gap D
Gap element L
(between nodes m and n)
m
l
Stress
D
_
L
Strain
Fig. 3.3-4: Modeling of gaps

3.4: Isothermal plastic material models
3.4 Isothermal plastic material models

 This section describes the following material models:
Plastic bilinear, plastic multilinear: von Mises model with

isotropic, kinematic hardening or mixed hardening
Plastic-cyclic: von Mises model with hardening rules suitable

for modeling cyclic plasticity.
 All elasto-plasticity models use the flow theory to describe the

elastic-plastic response; the basic formulations for the von Mises
models are summarized on pp. 596-604, ref. KJB.
3.4.1 Plastic-bilinear and plastic-multilinear material models
 Elasto-plastic materials with bilinear or multilinear hardening

are defined using the MATS1 material entry with TYPE =
‘PLASTIC’.
 These material models are based on
 The von Mises yield condition (see p. 597, ref. KJB)

 An associated flow rule using the von Mises yield function
 Isotropic, kinematic, or mixed hardening
 Bilinear or multilinear stress-strain curves (based on H and
TID fields in MATS1)
 Figs. 3.4-1 to 3.4-3 summarize some important features of these

material models. Point A marks the first onset of plasticity, point B
marks unloading into elasticity, point C marks reverse loading to
plasticity and point D marks unloading into elasticity.
 These models can be used with the rod, 2-D solid, 3-D solid,
beam (plastic-bilinear only), and shell elements.
 All elastic and plastic material constants are thermally invariant.

However, thermal strains can be present when there is a
temperature load and a coefficient of thermal expansion.

Stress
Multilinear stress-strain curve
0s
y
Bilinear stress-strain curve
Strain
Fig. 3.4-1: von Mises model
s3
(1,1,1) s3
Ö 23
0
sy
Ö 23
0
sy
s2 s1 s2
Elastic region
s1
a) Principal stress space b) Deviatoric stress space
Fig. 3.4-2: von Mises yield surface

Stress Stress
B ET B ET
A A
0s ts 0s
y y y
E E 2 0sy
Strain E Strain
E ts
y E
E D C
D ET
C
ET
a) Bilinear isotropic hardening b) Bilinear kinematic hardening
Stress Stress
B B
A A
ts
0s y
0s 2 0sy
y y
Strain Strain
ts D C
y
D C
c) Multilinear isotropic hardening d) Multilinear kinematic hardening
Fig. 3.4-3: Isotropic and kinematic hardening
 These models can be used with small displacement/small

strain, large displacement/small strain and large
displacement/large strain kinematics.
When used with small displacements/small strain kinematics, a
materially-nonlinear-only formulation is employed.
When used with large displacements/small strain kinematics,
either a TL or a UL formulation is employed (depending on
element type).
When used with large displacement/large strain kinematics, a
ULH formulation or a ULJ formulation can be employed. Large
displacement/large strain kinematics can only be used with the 2-D

solid, 3-D solid and shell elements (only single layer shell
elements).
If geometrically nonlinear effects are to be included, the large
displacements/large strain kinematics are preferred to the large
displacement/small strain kinematics, even when the strains are
numerically small. The large displacement/small strain kinematics
should be used only when the large displacement/large strain
kinematics are not supported by the element.
 For multilinear plasticity, there is no restriction on the number
of stress-strain points in the stress-strain curve.
 Mixed hardening is available only for bilinear plasticity.
 Plane strain, axisymmetric or 3-D solid elements that reference

these material models should preferably employ the mixed
displacement-pressure (u/p) element formulation. This is done by
setting UPFORM = 1 in the NXSTRAT command.
 In the von Mises model with isotropic hardening, the following
yield surface equation is used:
1 t t 1
t
fy 
2
 s  s   t y2  0
3
where ts is the deviatoric stress tensor and 0 y2 the updated yield

stress at time t.
In the von Mises model with kinematic hardening, the following
1 t 1
t
fy 
2
 s  t α    t s  t α   0 y2  0
3
where t α is the shift of the center of the yield surface (back stress
tensor) and 0  y2 is the virgin, or initial, yield stress.
In the von Mises model with mixed hardening, the following

1 t 1
t
fy 
2
 s  t α    t s  t α   t y2  0
3
where
t
 y  0 y  ME p e p
The back stress t α is evolved by
dα  C p 1  M  de p
Cp is Prager’s hardening parameter, related to the plastic

modulus Ep and M is the factor used in general mixed hardening
(0 < M < 1) which is currently restricted to 0.5.
The formulation for the von Mises model with mixed hardening
is given in the following reference:
ref K.J. Bathe and F.J. Montáns, “On Modeling Mixed

Hardening in Computational Plasticity”, Computers and
Structures, Vol. 82, No. 6, pp. 535–539, 2004.
The yield stress is a function of the effective plastic strain,

which defines the hardening of the material. The effective plastic
strain is defined as
t
P 2 p
t
e  de  de p
0
3
in which de p is the tensor of differential plastic strain increments

and in which de p  de p is calculated as deijp deijp (see ref. KJB, p.
P
599). In finite element analysis, t e is approximated as the sum of
all of the plastic strain increments up to the current solution time:

P
t
e  e p
all solution steps

P 2 p
where  e  e  e p and e p is the tensor of plastic strain
3
increments in a solution step. Because of the summation over the
P
solution steps, the calculated value of t e is referred to as the
accumulated effective plastic strain.
 If a thermal load is applied to the structure, the thermal strains

are taken into account but the material characteristics are
considered to be temperature independent.
 Notice that for the multilinear material models, the plastic

tangent modulus at point C in Fig. 3.4-3 is smaller than the plastic
tangent modulus at point A. The plastic tangent modulus decreases
for each successive cycle. Hence these multilinear material models
are not well suited for modeling cyclic behavior.
If cyclic behavior is to be modeled, either the bilinear kinematic
model can be used (because the plastic tangent modulus is
constant), or the plastic-cyclic material model can be used.
Stress-strain input data for multilinear plasticity in large strain

analysis
In large strain analysis, Advanced Nonlinear Solution works

internally with true (Cauchy) stresses and true (logarithmic) strains.
However, typical tension tests used to determine experimental data
return forces and displacements. These forces and displacements
are used to compute engineering stresses (force per unit original
area) and engineering strain (change in length per unit length).
Therefore it is necessary to convert engineering stress-strain data to
true stress-strain data. This conversion is either done by the user,
or is done automatically by Advanced Nonlinear Solution ( using
the CVSSVAL=1 entry on the NXSTRAT card). We now discuss
this conversion process in detail.
Consider an experiment in which a fully incompressible
material is put into uniaxial tension. The force-displacement curve
is determined, and the following information extracted from the
force-displacement curve:

The engineering stress is computed as the force / original area.

The engineering strain is computed as the displacement /
original length.
This data is determined for a number of points on the force-

displacement curve, starting at point 2, which is considered to be
the elastic limit (point 1 is at the origin). And the Young’s
modulus is also determined as the ratio of engineering stress /
engineering strain at the elastic limit.
Now consider duplicating the experimental results with a finite
element model that uses a large strain formulation. The tension test
finite element model should return the correct force for each
displacement point on the force-displacement curve.
Because the stress and strain measures used in the large strain
finite element formulation are true stress and true strain, it is
necessary to convert the engineering stress / engineering strain data
to true stress / true strain data.
The conversion can be done using an algorithm similar to
{
For (each stress-strain point i, i=1, 2, ... ) {
ei  engineering strain
 i  engineering stress
 i   i (1  ei ) = true stress
 i  ln(1  ei ) = true strain
}
E  2 / 2
}
Notice that E also needs to be converted. The reason is as

follows. If E is not converted, Advanced Nonlinear Solution
computes  2   2 / E but now  2 is no longer the true strain at the
elastic limit.
Here is a numerical example:
e2  0.1 ,  2  30 MPa , E   2 / e2  300 MPa

The conversion given above produces
 2  0.09531 ,  2  33 MPa
This point is assumed to be at the elastic limit. If E is kept at 300,

then Advanced Nonlinear Solution computes
 2  33 / 300  0.11
and this  2 is no longer at the elastic limit. Therefore E must be

recalculated as part of the conversion process:
E  33 / 0.09531  346.2 MPa
Assuming that the above conversion is performed, then the tension

test finite element model will return the correct force for each of
the displacements corresponding to the points on the original force-
displacement curve.
Assumptions in the conversion formulas
There are a number of assumptions in the above formulas, as

follows:
Elastic response
It is assumed that the stress-strain behavior is linearly elastic up

to the elastic limit. However, since Advanced Nonlinear Solution
uses true stress / true strain data internally, the linear elastic
behavior is also based on true stress / true strain data. In the above
example, the linear elastic response computed by the tension test
finite element model is based on a Young’s modulus of 346.2 MPa.
This response is, of course, quite different than a linear elastic
response based on a Young’s modulus of 300 MPa.
The reason that the Young’s modulus is so different is because
the strain at the elastic limit is “large”. When the strain at the
elastic limit is small, then the change in Young’s modulus is also
small. The change in Young’s modulus caused by the conversion is

3
approximately equal to E  e2 E .
2
Compressible elastic material
The above conversion assumes that the material response is

fully incompressible, under both elastic and plastic conditions.
However, in most cases, the material is compressible under elastic
conditions.
Nevertheless, the above formulas are frequently used anyway.
The error thus incurred will be largest for point 2, and will diminish
for larger values of strain.
Response in uniaxial compression
The above formulas use tension data to convert from

engineering to true values. However, it should be recognized that
Advanced Nonlinear Solution uses the same stress-strain curve in
both tension and compression. Therefore, if a large strain finite
element model is put into uniaxial compression, the observed
(compressive) force will not be equal in magnitude to the expected
(tensile) force, at a given (compressive) displacement level.
Here is an example. Suppose that in a tensile test, the following
data is obtained:
Point Engineering Engineering True strain True

strain stress (MPa) stress
...
3 0.05 50 0.04879 52.5
4 0.05263 51 0.05129 53.7
...
First consider a materially-nonlinear-only analysis. Enter the

engineering stress and strain values for the two points as part of the
stress-strain input data. When the model is run in compression to
an engineering strain of -0.05, the engineering stress is -50 MPa.
The force-displacement response is symmetric in tension and
compression.
Next consider a large strain analysis. Enter the true stress and
strain values for the two points as part of the stress-strain input
data. Now run the model in compression to an engineering strain of

-0.05. An engineering strain of -0.05 corresponds to a true strain of

-0.05129, therefore the true stress is -53.7 and the engineering
stress is -56.5. The force-displacement response is not symmetric
in tension and compression.
Homogeneous deformation
The above conversion assumes that the specimen is uniform and

that the specimen deforms homogeneously under load. Therefore
the conversion is only valid up to the ultimate tensile strength of
the material, because beyond that point the deformation of the
specimen might be no longer homogeneous due to localized
necking.
NXSTRAT CVSSVAL feature
When CVSSVAL=0 on the NXSTRAT card (the default),

Advanced Nonlinear Solution does not perform any conversion.
True stress-strain data should be input, and the user could compute
this true stress-strain data using the formulas given above.
When CVSSVAL=1, then Advanced Nonlinear Solution
performs this conversion using the formulas given above.
Therefore engineering stress-strain data should be input.
The CVSSVAL=1 feature should only be used when all of the
elements that use multilinear plastic materials also use large strain
formulations.
The CVSSVAL=1 feature does not perform any conversions for
the bilinear plastic material models, or for the plastic-cyclic
material models.
Material behavior beyond the last point of the stress-strain

curve in multilinear plasticity
The material behavior beyond the last point of the stress-strain

curve in multilinear plasticity can be considered ruptured, or the
curve can be extended indefinitely with the slope of its final
segment. This depends on a global setting of the XTCURVE
parameter in NXSTRAT with indefinite extension as the default.
Modeling of rupture: Rupture conditions can also be modeled for

the multilinear stress-strain curve. The rupture plastic strain

corresponds to the effective plastic strain at the last point input for
the stress-strain curve. No rupture strain exists for the bilinear case.
When rupture is reached at a given element integration point,
the corresponding element is removed from the model (see Section
10.4).
Rate-dependent plasticity: The rate-dependent model is used to

simulate the increase in the yield stress with an increase in strain
rate.
The rate-dependent model only applies to the isotropic plasticity
models with isotropic hardening (bilinear or multilinear).
The rate-dependent model is implemented for 2-D solid, 3-D
solid, rod and shell elements.
The effective yield stress including strain rate effects is
   P  
 y   y0 1  b ln 1  
  0  
where  y0 is the static yield stress, 0 is the transition strain rate
and b is the strain hardening parameter. Here  P is calculated
 P
using  P  .
t
For more information on this formula, see the following
reference
ref W.H. Drysdale and A.R. Zak, “Mechanics of Materials

and Structural Theories  A Theory for Rate Dependent
Plasticity”, Computers and Structures, Vol. 20, pp. 259-
264, 1985.
Rate-dependent plasticity is specified using the MATSR entry,

in which the MID number refers to the MATS1 entry for the
material. The MATSR entry contains BVALUE (corresponds to b )
and TSRATE (corresponds to 0 ). MADSR also contains TID, the
identification number of a TABLEST entry. If TID is specified, and
the referenced TABLEST entry contains stress-strain curves at one
or more strain rates, then Solution 601 calculates BVALUE by
curve-fitting, overwriting any specified BVALUE. If TID is

specified, and the referenced TABLEST entry contains stress-strain

curves at two or more strain rates, then Solution 601 calculates
BVALUE and TSRATE by curve-fitting, overwriting any specified
BVALUE and TSRATE.
In the TABLEST entry, the Ti are interpreted as plastic strain-
rates  P , not as temperatures. In the TABLES1 entries referred to
by the TIDi parameters, (xi, yi) are interpreted as (strain, stress).
The curve-fitting is done as follows: For each plastic strain rate
y
 P , the average overstress ratio is determined from the
 y0
associated user-input stress-strain curve (in TABLES1). Then the
curve-fitting is performed using the plastic strain rates and
overstress ratios.
3.4.2 Plastic-cyclic material model
 This material model is based on
 The von Mises yield condition (see p. 597, ref. KJB)

 A flow rule using the von Mises yield function
 An isotropic and/or kinematic hardening rule. The isotropic
and kinematic hardening rules used in the plastic-cyclic
model are suitable for use in modeling cyclic plasticity.
 The plastic-cyclic material model differs from the plastic-

bilinear and plastic-multilinear material models because the
isotropic and kinematic hardening rules are different. However, the
plastic-cyclic material model can reproduce the plastic-bilinear and
plastic-multilinear models when suitable material constants are
chosen.
 This material model can be used with the rod, 2-D solid, 3-D
solid, Hermitian beam and shell elements. However this material
model cannot be used with the CQUAD8 shell element, or with the
composite shell element.
 This model can be used with the small displacement/small

strain, large displacement/small strain and large
displacement/large strain formulations . Large displacement/large
strain kinematics can only be used with the 2-D solid, 3-D solid

and shell elements.
3.4.2.1 Fundamental concepts
 Many of the ideas used in the plastic-cyclic material model are

given in the following reference:
ref J. Lemaitre and J.-L. Chaboche, Mechanics of Solid

Materials, Cambridge University Press, 1990.
We abbreviate this reference as ref LC in the discussion below. As

an aid to understanding the model, whenever our notation differs
from the notation used in ref LC, we give the equivalent Lemaitre
and Chaboche notation.
 The motivation for the plastic-cyclic material model is

illustrated in Figs. 3.4-4 and 3.4-5. Fig. 3.4-4 shows a bar subjected
to uniaxial cycling, with the strain prescribed. Fig. 3.4-5 shows
response predictions from the plastic-bilinear and plastic-
multilinear material models of Section 3.4.1. On repeated cyclic
loading, perfect plasticity and multilinear hardening plasticity
produce stabilized plastic cycles with no additional hardening.
Bilinear isotropic hardening does not produce a stable plastic cycle
and bilinear kinematic hardening produces a very rough
approximation to a stable plastic cycle.

Strain
Time
a) Uniaxial cycling of a bar b) Prescribed strain time history
Fig. 3.4-4: Uniaxial cycling example

Stress
Stress
Strain Strain
a) Perfect plasticity b) Bilinear isotropic hardening
Stress
Stress
Strain Strain
c) Bilinear kinematic hardening d) Multilinear isotropic hardening
Stress
Strain
e) Multilinear kinematic hardening
Fig. 3.4-5: Response predictions using the plastic-bilinear and plastic-multilinear models

 Response predictions from the plastic-cyclic material model are

illustrated in Fig. 3.4-6. When nonlinear kinematic hardening is
used without isotropic hardening, a stable plastic cycle is reached
after one cycle. In this stable plastic cycle, the transition from
elastic to plastic conditions occurs more gradually than the
corresponding transition from bilinear kinematic hardening. Cyclic
hardening and cyclic softening are obtained by combining the
nonlinear kinematic hardening with isotropic hardening or
softening.
Stress Stress
Strain Strain
a) Nonlinear kinematic hardening, b) Nonlinear kinematic hardening,

no isotropic hardening or softening isotropic hardening
Stress
Strain
c) Nonlinear kinematic hardening,

isotropic softening
Fig. 3.4-6: Response predictions using the plastic-cyclic material model

 The plastic-cyclic material model can optionally contain a strain

memory surface. The motivation for using the strain memory
surface is shown in Fig. 3.4-7. Considering an increase in the
prescribed strain amplitude, if no strain memory surface is used,
then no additional cyclic hardening takes place, whereas if a strain
memory surface is used, additional cyclic hardening takes place.
Strain
Time
a) Prescribed strain time history

Additional
Stress Stress cyclic
hardening
Strain Strain
b) Stress-strain curve without c) Stress-strain curve with

strain memory surface strain memory surface
Fig. 3.4-7: Response predictions with and without

strain memory surface
 Ratchetting occurs when prescribed stresses with non-zero mean

stress are applied to the bar, as shown in Fig. 3.4-8.

Stress Stress
Strain
Time
a) Prescribed stress time history b) Stress-strain curve
Fig. 3.4-8: Ratchetting using the plastic-cyclic material model
Stresses, strains, stress-strain law:
t t
τ t t s  t t m I
where t t τ is the stress tensor ( σ in ref LC), t t

s is the deviatoric
stress tensor ( σ in ref LC) and t t
 m is the mean stress
(hydrostatic stress  H in ref LC).
t t
e t t
e  t t em I
where t t e is the strain tensor ( ε in ref LC), t t

e is the
deviatoric strain tensor ( ε in ref LC) and t t
em is the mean strain
(hydrostatic strain  H in ref LC).
 m  3 t t em
t t
t t
s  2G ( t t e  t t e P )
where  and G are the bulk modulus and shear modulus, and
t t P
e is the plastic strain ( ε P in ref LC). Thermal strains are not
included in the above equations, but are included in the program
when there are thermal effects.

Yield condition: The von Mises yield condition is
1 2 1
t t
fy  t t
s  t t α  t t
 y2  0
2 3
where α is the back stress tensor ( X and X in ref LC, note

t t
that back stress is always deviatoric) and t t  y is the yield stress
(  Y  R or k in ref LC). The norm of a symmetric tensor a is
defined as a  a : a . The yield condition is always evaluated
at time t  t .
Fig. 3.4-9 shows the yield condition.
s3
t+Dt
fy=0
t+Dt
s
t+Dt
a
2 t+Dts
y
3
s1 s2
Fig. 3.4-9: von Mises yield surface in principal

deviatoric stress space
Flow rule: The flow rule states that the direction of plastic strain
increments is normal to the yield surface.
directions of ΔeP  directions of  t   t

s  t  t α 
where Δe P is the increment in plastic strain, and  is a constant

used to choose the yield surface configuration (   0 corresponds

to the configuration at time t ,   1 corresponds to the

configuration at time t  t , other values of  correspond to
intermediate configurations). (The concept is similar to that used
in alpha-integration; we would have used  instead of  , except
that  is used for the back stresses.) The directions of a symmetric
a
tensor a are defined as directions of a  .
a
Fig. 3.4-10 shows the evolution of the yield surface using the
stress-strain law, yield condition and flow rule.
Strain memory surface: The memory-exponential isotropic

hardening rule uses the concept of a strain memory surface with
additional internal variables t t ξ and t t q . The strain memory
surface is now briefly described.
The strain memory surface is defined in the space of plastic

strains as the surface of a sphere centered at position t t ξ with
3 t t t t
radius q . (In ref LC, ξ is written ζ ). This surface can
2
be written as
2 t t 2
fm  eP  t t ξ  t t q 2  0
3
Fig. 3.4-11 shows the strain memory surface.

s3
Ds t+Dt
ts s
t+Dt
Da a
ta
2 t+Dts
y
2 ts 3
3 y
s1 s2
a) Overview
DeP parallel to n
2GDe’ 2GDe’
ts -2GDeP ts
n -2GDeP
t+Dt
t+Dt
s s
Yield surface Yield surface Yield surface Yield surface

at time t at time t + Dt at time t at time t + Dt
Yield surface
at time t + Dt/2
b) b=1, DeP normal to yield c) b=1/2, DeP normal to yield

surface at time t + Dt surface at time t + Dt/2
Fig. 3.4-10: Incremental update of yield surface and plastic strains

e3P
t+Dt
fm=0
t+Dt
eP
t+Dt
x
3 t+Dtq
2
e1P
e2P
Fig. 3.4-11: Strain memory surface in principal

plastic strain space
The differential rules used for the evolution of the strain memory
surface are
dq   (directions of deP ) : (directions of ( t eP  t ξ )) de P
and
directions of dξ  directions of ( t eP  t ξ )
in which the symbol <> means that <u>=0 when u<0, and <u>=u
when u>0. The incremental versions of these rules are
q   (directions of ΔeP ) : (directions of ( t  t eP  t  t ξ )) e P
and
directions of Δξ  directions of ( t  t eP  t  t ξ )
Again, the  notation is used to denote a configuration between t

and t  t . These concepts are illustrated in Fig. 3.4-12.

n
t+Dt
t
eP De P eP
t+Dt
Dx x
tx
3 t+Dtq
2
3 tq
2 Dx parallel to n
Fig. 3.4-12: Incremental update of strain memory

surface, for b=1
1
 is a material constant that must be between 0 and . Typically
2
1
 .
2
These rules ensure that the strain memory surface at time t  t
encloses the strain memory surface for all preceding times. The
evolution of the strain memory surface in 1D uniaxial straining is
shown in Fig. 3.4-13.

Advanced Nonlinear TMG 1 200

Uploaded by

Copyright:

Available Formats

Advanced Nonlinear TMG 1 200

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Advanced Nonlinear TMG 1 200

Uploaded by

Copyright:

Available Formats

NX Nastran 12

Advanced Nonlinear Theory and Modeling

HDF5 (Hierarchical Data Format 5) Software Library and Utilities Copyright

4. All publications or advertising materials mentioning features or use of this software

DISCLAIMER: THIS SOFTWARE IS PROVIDED BY THE HDF GROUP AND THE

Advanced Nonlinear Solution  Theory and Modeling Guide iii

iv Advanced Nonlinear Solution  Theory and Modeling Guide

3.2 Linear elastic material models ....................................................................... 157

Advanced Nonlinear Solution  Theory and Modeling Guide v

4.3.1 Kinematic constraint method.................................................................. 286

vi Advanced Nonlinear Solution  Theory and Modeling Guide

5.7 Bolt preload ................................................................................................... 376

Advanced Nonlinear Solution  Theory and Modeling Guide vii

8.1 Formulation ................................................................................................... 444

viii Advanced Nonlinear Solution  Theory and Modeling Guide

To provide a concise summary of the theoretical basis of

ref. K.J. Bathe, Finite Element Procedures, 2nd ed.,

 To provide guidelines for practical and efficient modeling using

It is assumed that the user is familiar with NX Nastran

Advanced Nonlinear Solution  Theory and Modeling Guide 1

1.2 Overview of Advanced Nonlinear Solution

 Advanced Nonlinear Solution supports static and implicit

 Advanced Nonlinear Solution supports many of the standard

 Advanced Nonlinear Solution supports many of the commonly

 The elements available in Advanced Nonlinear Solution can be

 The material models available in Advanced Nonlinear Solution

2 Advanced Nonlinear Solution  Theory and Modeling Guide

 Advanced Nonlinear Solution has very powerful features for

 Loads, boundary conditions and constraints are addressed in

 Solution 601 of Advanced Nonlinear Solution currently

 Solution 701 of Advanced Nonlinear Solution is dedicated to

 Solution 601 of Advanced Nonlinear Solution also supports two

 Additional capabilities present in Advanced Nonlinear Solution

Advanced Nonlinear Solution  Theory and Modeling Guide 3

 Most of the global settings controlling the structural solutions in

 Similarly, most of the global settings controlling the heat

1.2.1 Choosing between Solutions 601 and 701

 For both linear and nonlinear static problems, the implicit

 For heat transfer and coupled structural heat transfer problems,

 For low-speed dynamic problems, the solution time spans a

4 Advanced Nonlinear Solution  Theory and Modeling Guide

problems the resulting number of time steps will be excessive,

 For high-speed dynamic problems, the solution time is

 A large number of dynamics problems cannot be fully classified

 Since the explicit time step size depends on the material

Advanced Nonlinear Solution  Theory and Modeling Guide 5

incompressible materials (provided that a mixed formulation is

 Higher order elements such as the 10-node tetrahederal, 20 and

 Model nonlinearity is another criterion influencing the choice

 The memory requirements is another factor. For the same mesh,

 Since Advanced Nonlinear Solution handles both Solution 601

6 Advanced Nonlinear Solution  Theory and Modeling Guide

In Advanced Nonlinear Solution, it is important to enter all

1.3 Structure of Advanced Nonlinear Solution

1. Nastran Statement (optional)

The first two sections do not involve any special treatment in

1.3.1 Executive Control

 Solution 601 is invoked by selecting solution sequence 601 in

where N determines the specific analysis type selected by Solution