MD R2 Nastran Release Guide

MD R2 Nastran
Release Guide
Corporate
MSC.Software Corporation
2 MacArthur Place
Santa Ana, CA 92707 USA
Telephone: (800) 345-2078
Fax: (714) 784-4056
Europe
MSC.Software GmbH
Am Moosfeld 13
81829 Munich, Germany
Telephone: (49) (89) 43 19 87 0
Fax: (49) (89) 43 61 71 6
Asia Pacific
MSC.Software Japan Ltd.
Shinjuku First West 8F
23-7 Nishi Shinjuku
1-Chome, Shinjyku-Ku
Tokyo 160-0023, JAPAN
Telephone: (03)-6911-1200
Fax: (03)-6911-1201
Worldwide Web
www.mscsoftware.com
Disclaimer
MSC.Software Corporation reserves the right to make changes in specifications and other information contained
in this document without prior notice.
The concepts, methods, and examples presented in this text are for illustrative and educational purposes only,
and are not intended to be exhaustive or to apply to any particular engineering problem or design. MSC.Software
Corporation assumes no liability or responsibility to any person or company for direct or indirect damages resulting
from the use of any information contained herein.
User Documentation: Copyright

2007 MSC.Software Corporation. Printed in U.S.A. All Rights Reserved.
This notice shall be marked on any reproduction of this documentation, in whole or in part. Any reproduction or
distribution of this document, in whole or in part, without the prior written consent of MSC.Software Corporation is
prohibited.
This software may contain certain third-party software that is protected by copyright and licensed from
MSC.Software suppliers.
MSC, MD, Dytran, Marc, MSC Nastran, MD Nastran, MSC Patran, MD Patran, the MSC.Software corporate logo,
and Simulating Reality are trademarks or registered trademarks of the MSC.Software Corporation in the United
States and/or other countries.
NASTRAN is a registered trademark of NASA. PAMCRASH is a trademark or registered trademark of ESI Group.
SAMCEF is a trademark or registered trademark of Samtech SA. LS-DYNA is a trademark or registered trademark
of Livermore Software Technology Corporation. ANSYS is a registered trademark of SAS IP, Inc., a wholly owned
subsidiary of ANSYS Inc. ABAQUS is a registered trademark of ABAQUS Inc. All other brand names, product
names or trademarks belong to their respective owners.
jak^WoOWwWwWwWa`Jobi
Cont ent s
MD R2 Nastran Release Guide
List of Books xii
Technical Support xiii
Internet Resources xv
1 Overview of MD R2 Nastran
Overview 2
Contact in Linear and Nonlinear Solutions 2
Advanced Integrated Nonlinear Analysis 2
Implicit Nonlinear Analysis (SOL 600) 3
Explicit Nonlinear Analysis (SOL 700) 3
NVH & Acoustics 4
Numerical Enhancements 4
Elements & Connectors 4
Optimization 5
Rotor Dynamics & Aeroelasticity 5
List of MD Nastran Documents Released with MD R2 Nastran 6
2 Contact in Linear and Nonlinear Solutions
Linear Contact, Permanent Glued Contact, and Nonlinear Contact 8
Introduction 8
Linear Contact Modeling in SOL 101 8
Permanent Glued Contact Modeling in SOLs 101, 103, 105, 107, 108, 109, 110,
111, and 112 8
Nonlinear Contact Modeling in SOL 400 8
Benefits 9
Input 9
Output 10
Guidelines and Limitations 10
Examples 15
ja=oO=k~~=
o~=d
Table of Contents
==
iv
3 Advanced Integrated Nonlinear Analysis
Advanced Integrated Nonlinear (SOL 400) 26
Introduction 26
Benefits 26
Limitations for the Current Release 27
A New General Nonlinear Contact Analysis Capability 27
Kinematic Elements 28
Introduction 28
Benefits 28
Type of Elements that can be Combined with the Kinematic Elements 28
The Scale Factors LMFACT and PENFN 29
Thermal Effects 29
RJOINT and the Pin Flags of CBEAM Element 29
User Interfaces 30
Examples 31
SOL 400 Material and Elements 36
Nonlinear Property Extension 36
The Nonlinear QUADR/TRIAR Elements 44
Introduction 44
Benefits 44
Limitations 44
Type of Analysis 44
Drilling Degrees of Freedom 45
Off-plane Shear Formulations 45
Green Strain 46
Stress Output Location 46
Differential Stiffness 46
CBAR Element to CBEAM Element Conversion 47
CSHEAR Element to CQUAD4 Element Conversion 48
SOL 400 Pretension Modeling Description 49
Pretension Modeling 49
VCCT Virtual Crack Closure Method 54
Cohesive Zone Modeling 55
Nonlinear Iteration Algorithms 58
Analysis Chaining SUBCASE, STEP, ANALYSIS, and NLIC 62
v Contents
Adaptive Time Stepping Scheme 65
AUTO Step 65
Recycling Criterion 65
Exceptions 66
Post Files Output 66
Defaults 66
The Numerical Integration Method for the Transient analysis the HHT
method 68
Nonlinear Iteration Summary Table for Nonlinear Transient Analysis in
SOL 400 69
Output Data Grouping: NLPACK 72
Restarts 73
File Management Commands 73
Case Control Modifications 73
Bulk Data Modifications 75
Initial Conditions 76
Transient Temperature Loads 78
Boundary Condition (SPC and MPC) Changes 80
Direct Matrix Input Changes 81
Rotor Dynamics 82
Coordinate System for the Nonlinear Stresses of 3D Elements 83
Outputs 84
Error Handling 85
User Interfaces 86
Examples 88
4 Implicit & Explicit Nonlinear Analysis
MD R2 Nastran Implicit Nonlinear - SOL 600 106
Heat Transfer 106
Heat Transfer Examples 109
Creep Simulations 109
Element Selection 109
Other SOL 600 Items 109
==
vi
Fracture Mechanics J-Integral (LORENZI) 111
Delamination 111
MATEP Extensions 112
SOL 600 Failure Description MATF 112
Element Birth and Death 116
Unglue 116
Composite Element Numerical Analysis 116
PLOAD4 Extensions 117
Large Rotation RBE 117
Streaming Input 117
CONNECTOR TECHNOLOGY 118
COMPUTATIONAL ENHANCEMENTS 119
SUPER ELEMENTS / DMIG 119
CONTACT ENHANCEMENTS 119
New SOL 600 Parameters 120
Platform Specific Notes 121
For 64-bit Windows EM64T platforms: 122
For 32-bit Windows platforms: 122
Supported Systems for SOL 600 in MD R2 Nastran 123
MD R2 Nastran Explicit Nonlinear - SOL 700 124
Introduction 124
Latest Capabilities of MD Nastran Explicit Nonlinear - SOL 700 124
Airbags and Occupant Safety 124
Inflator in AirBags 126
Occupant Dummy Models 126
FTSS (First Technology Safety Systems) ATDs 128
Time Domain NVH 128
Time Domain NVH Example 130
Prestressing (Implicit to Explicit Sequential Simulation) 139
Additional Capabilities of MD Nastran Explicit Nonlinear - SOL 700 140
5 NVH & Acoustics
Frequency Response Function (FRF) and FRF Based Assembly (FBA)
Feature in MD R2 Nastran 144
Introduction 144
FRF Concept 144
FBA Concept 145
Usage of the FRF/FBA Feature 145
Exterior Acoustics 148
Introduction 148
Benefits 148
vii Contents
Input 148
Definition of Infinite Elements 148
Definition of Field Point Meshes 151
Case Control Commands 153
Output 155
Guidelines 156
Limitations 156
Input File 159
Excerpt of fluid1.bdf 160
Results 161
Rigid Porous Absorber 163
Introduction 163
Porous Materials 163
Inputs 164
Discussion 165
References 165
Example 165
6 Numerical Enhancements
New SPARSESOLVER Executive Statement 172
Introduction 172
Benefits 172
Method and Theory 172
Inputs 172
Outputs 172
Improved Performance with New Sparse Solvers 175
Introduction 175
The TAUCS Sparse Solver 175
User Interface 175
Use Cases 175
Limitations 176
The UMFPACK Sparse Solver 177
Improved Memory Usage in Lanczos (Pre-release) 180
New MAXRATIO Information Output (Pre-release) 181
Introduction 181
Benefits 181
Method and Theory 181
Inputs 181
==
viii
Outputs 181
Demonstration Example 182
Example Input Data 182
Example Output 184
Performance Improvements and Expanded Capabilities for ACMS 185
Introduction 185
MDACMS for Upstream Superelements 185
Automatic FASTFR Decision Logic 186
Examples 187
Limitations for the FASTFR method 188
Miscellaneous Performance Improvements 189
Relaxed Restrictions for CASI Solver Usage 192
System Dependent Performance Improvements 193
Introduction 193
Model Description 193
Performance Improvements for Intel x86_64 194
Performance Improvements for AMD due to Atlas 194
Improved Selection of Reordering Methods 195
7 Elements & Connectors
Enhancements to Connector Elements 198
Introduction 198
CSEAM Elements 198
SWLDPRM Enhancements 201
Displacement Output of GA and GB for CWELD and CFAST Elements 202
Composite Beam Using VAM 206
Introduction 206
Benefits 206
Input 206
Output 207
Guidelines 208
Limitations 210
Example 210
References 212
Nonhomogeneous Multipoint Constraint 213
Introduction 213
Benefits 213
ix Contents
Output 213
Theory 213
Examples 215
8 Optimization
Topology Optimization Enhancements 220
Introduction 220
Benefits 220
Modified Type One Responses - Fractional Mass 222
Example 1 Wheel (wheeltop.dat) 223
Automatic External Superelement Optimization (AESO) 226
Introduction 226
Benefits 226
Methodology 226
Input 227
Outputs 229
Examples 234
Miscellaneous 237
Randomization of a Users Input Data File (Pre-release) 241
Introduction 241
Benefits 241
Random Elimination of Element Types (Pre-release) 243
Introduction 243
Benefits 243
9 Rotor Dynamics & Aeroelasticity
Changes to Rotordynamics for MD R2 Nastran 246
Unbalance Entry for Frequency Response 246
New Parameter and Hybrid Damping Specifications 246
Updating/Summing of Monitor Points 247
Introduction 247
Benefits 247
Examples (monsum.dat and monsum3.dat) 247
==
x
Stripwise Aerodynamic Results 249
Introduction 249
Benefits 249
Examples (moncncm and monbodi) 250
Input of an Aerodynamic Mesh 251
Introduction 251
Benefits 251
Example (aegridf) 251
Rigid Body Spline 255
Introduction 255
Benefits 255
Example (splinrb) 255
Wendland Spline Functions for the Spline4/5 256
Introduction 256
Benefits 256
Example (ha144c_ris) 256
Spline Blending 257
Introduction 257
Benefits 257
Example (ha145e_blnd) 257
Export of the Spline Matrix 259
Introduction 259
Benefits 259
Examples (splinopch and splinoop2) 259
The 2005 New Template
m~
List of Books
Technical Support
Internet Resources
MD Nastran R2 Release Guide

xii
List of Books
Below is a list of some of the MD Nastran and MSC Nastran documents. You may order any of these
documents from the MSC.Software BooksMart site at http://store.mscsoftware.com/.
f~~=~=o~=d
Installation and Operations Guide
Release Guide
o=_
Quick Reference Guide
DMAP Programmers Guide
Reference Manual
r=d
Getting Started
Linear Static Analysis
Basic Dynamic Analysis
Advanced Dynamic Analysis
Design Sensitivity and Optimization
Thermal Analysis
Numerical Methods
Aeroelastic Analysis
Superelement
User Modifiable
Toolkit
Implicit Nonlinear (SOL 600)
Explicit Nonlinear (SOL 700)
xiii
Preface
Technical Support
For help with installing or using an MSC.Software product, contact your local technical support services.
Our technical support provides the following services:
Resolution of installation problems
Advice on specific analysis capabilities
Advice on modeling techniques
Resolution of specific analysis problems (e.g., fatal messages)
Verification of code error.
If you have concerns about an analysis, we suggest that you contact us at an early stage.
You can reach technical support services on the web, by telephone, or e-mail.
t Go to the MSC.Software website at www.mscsoftware.com, and click on Support. Here you can find
a wide variety of support resources including application examples, technical application notes, training
courses, and documentation updates at the MSC.Software Training, Technical Support, and
Documentation web page.
m=
~=
c~
r=p~
Telephone: (800) 732-7284
Fax: (714) 784-4343
cI=`~
pI=r=h
Telephone: (44) (1276) 60 19 00
Fax: (44) (1276) 69 11 11
jI=d~
Telephone: (49) (89) 43 19 87 0
Fax: (49) (89) 43 61 71 6
qI=g~~
Telephone: (81) (03) 6911 1200
Fax: (81) (03) 6911 1201
oI=f~
Telephone: (390) (6) 5 91 64 50
Fax: (390) (6) 5 91 25 05
m~I=c~
Telephone: (33) (1) 69 36 69 36
Fax: (33) (1) 69 36 45 17
jI=o~
Telephone: (7) (095) 236 6177
Fax: (7) (095) 236 9762
d~I=q=k~
Telephone: (31) (18) 2543700
Fax: (31) (18) 2543707
j~I=p~
Telephone: (34) (91) 5560919
Fax: (34) (91) 5567280

xiv
b~ Send a detailed description of the problem to the email address below that corresponds to the product you
are using. You should receive an acknowledgement that your message was received, followed by an
email from one of our Technical Support Engineers.
q~
The MSC Institute of Technology is the world's largest global supplier of CAD/CAM/CAE/PDM training
products and services for the product design, analysis, and manufacturing markets. We offer over 100
courses through a global network of education centers. The Institute is uniquely positioned to optimize
your investment in design and simulation software tools.
Our industry experienced expert staff is available to customize our course offerings to meet your unique
training requirements. For the most effective training, The Institute also offers many of our courses at our
customer's facilities.
The MSC Institute of Technology is located at:
2 MacArthur Place
Santa Ana, CA 92707
Phone: (800) 732-7211
Fax: (714) 784-4028
The Institute maintains state-of-the-art classroom facilities and individual computer graphics laboratories
at training centers throughout the world. All of our courses emphasize hands-on computer laboratory
work to facility skills development.
We specialize in customized training based on our evaluation of your design and simulation processes,
which yields courses that are geared to your business.
In addition to traditional instructor-led classes, we also offer video and DVD courses, interactive
multimedia training, web-based training, and a specialized instructor's program.
Course Information and Registration. For detailed course descriptions, schedule information,
and registration call the Training Specialist at (800) 732-7211 or visit www.mscsoftware.com.
ja=m~~=p
ja=k~~=p
a~=p
jp`=c~=p
j~=p
jp`=f=`=f~
~~K]~K
~~K]~K
~K]~K
~K]~K
~K]~K
~K]~K
xv
Preface
Internet Resources
MSC.Software (www.mscsoftware.com)
MSC.Software corporate site with information on the latest events, products, and services for the
CAD/CAE/CAM marketplace.

xvi
Ch. 1: Overview of MD R2 Nastran MD R2 Nastran Release Guide
1
Overview of MD R2 Nastran
Overview

2
Overview
The MD R2 Nastran release brings powerful new features and enhancements for enterprise solutions in
the areas of contact, nonlinear analysis, NVH and acoustics, Automotive Powertrain, Crash and
Occupant Safety, high performance computing, assembly modeling, optimization, rotor dynamics and
aeroelasticity.
Contact in Linear and Nonlinear Solutions
MD R2 Nastran introduces contact in linear and nonlinear solutions with the ability to easily convert from
linear contact in SOL 101 to nonlinear contact in SOL 400.
Nonlinear Contact in SOL 400. Provides general multi body contact under full nonlinear
conditions of large contact motion, nonlinear materials, large rotation, and large strain. It
features deformable-deformable contact, deformable-rigid contact, and contact with friction.
Linear Contact in SOL 101. A similar algorithm as nonlinear contact for use in SOL 101 linear
analysis. A SOL 101 model with linear contact can be converted to simulate nonlinear behavior
in SOL 400 without having to redefine contact.
Permanent Glued contact in linear solution sequences. A special type of contact model for
condition where the contacting surfaces have no relative normal or tangential motion. Permanent
Glued contact is available in nearly all linear solutions (ex. SOL 200) and provides a convenient
way for joining dissimilar meshes. Glued, Linear and Nonlinear contact utilize the same user
interface.
Advanced Integrated Nonlinear Analysis
SOL 400 has been extensively enhanced for multi-step nonlinear analysis of large strain and material
behavior with contact.
Nonlinear Contact Modeling. A new contact algorithm has been implemented in SOL 400 to
perform general multi-body contact for large contact motion, nonlinear materials, large rotations
and large strains. It is available under both nonlinear static and nonlinear transient analysis.
Nonlinear Element and Materials. Enhancements include orthotropic material properties for 3-
dimensional and plane strain behavior, nonlinear gasket material properties for compression
behavior, elasto-plastic material properties for use in large deformation analysis, and failure
model properties for linear elastic material. Where appropriate, most finite elements in MD
Nastran have been extended to include the effects of finite strain. Also layered axisymmetric,
shell and solid composite elements are now available. The QUADR/TRIAR elements have been
extended to nonlinear analysis.
Nonlinear Procedures. New analysis chaining procedures to conveniently define multiple load
steps, running multiple independent cases, and specifying multiple and mixed types of analyses
in one job. This allows, for example, to pretension a bolt structure as a first load step in a multi-
step nonlinear analysis. Adaptive stepping schemes have been introduced that automatically
modify load increment size or time step size to accelerate convergence.
3
CHAPTER 1
Crack and Delamination. A new virtual crack closure method (VCCT) calculates energy release
rates for sharp cracks. In addition, a new library of interface elements (cohesive zone modeling)
can be used to simulate the onset and progress of delamination.
Kinematic Elements. New rigid elements with Lagrange formulation are now available for use
in geometric nonlinear analysis involving large rotations. They have been implemented in both
nonlinear static and nonlinear transient analysis.
Implicit Nonlinear Analysis (SOL 600)
Heat Transfer. Advanced thermal analysis is now available in SOL 600 including an efficient
hemi-cube viewfactor calculation method and automated procedures for thermal stress analysis
based upon the heat transfer simulation. Thermal analysis of composites may include accurate
calculation of the thermal gradient through the thickness.
Modeling Enhancements. Connector technology has been enhanced to include large deformation
formulations of CFAST, CWELD and CBUSH. Enhancements for facture mechanics include the
calculation of stress intensity factors using either the VCCT or Lorenzi method and the
prediction of delamination.
Performance Improvements. A new streaming input option is available that eliminates transfer
files. The analysis of composite shells has been improved such that assembly time has been
often reduced by a factor of 10 and memory requirements have also substantially been reduced.
Other performance improvements include an out-of-core iterative and direct PLOAD4 support.
Explicit Nonlinear Analysis (SOL 700)
Several new analysis capabilities are introduced in this release.
Airbags and Occupant Safety. Fluid Structure Interaction (FSI) is now available in SOL 700 to
simulate the complex behavior of multi-compartment airbags and their interaction with the
ATDs (Anthropomorphic Test Devices) during crash scenarios. The airbag inflation is based on
full gas dynamics and finite volume (Eulerian) technology with adaptive meshing capability.
The airbag surface acts as a coupling surface, adapts itself and follows the expanding eulerian
elements as the gas jet flows inside the bag. In addition to full gas dynamics, SOL700 also
supports the conventional Uniform Pressure Method to simulate the airbag behavior.
Time Domain NVH. This technique is based on FFT (Fast Fourier Transform) method to
compute frequencies and mode shapes of highly nonlinear and dynamic systems such as a
vehicle running over a rough road. The advantage of Time Domain NVH is its ability to capture
the nonlinear effects of a system such as contact, springs, bushings, tire and suspension
behavior.
Prestress. Prestressing is performed by double precision version of the Implicit solver and the
results are used as pre-state for Explicit simulation such as bird strike and blade out applications.
Nastran Native Output - This release produces native MD Nastran output files for seamlessly
post-processing explicit events with MD Patran.

4
New Material and Element formulations. Thirteen new material models and five element
formulations are added in SOL 700.
Contact Features. SOL 700 now supports many new contact capabilities such as Glued contact,
Tied nodes to surface, Tied Shell Edge to Surface, Tied Surface to Surface, Spotweld contact,
Single Edge contact and Force Transducers.
NVH & Acoustics
Frequency Response Functions (FRFs) and FRF Based Assembly (FBA). A powerful new
capability in MD R2 Nastran for frequency response analysis based on representations of
components by their FRFs and assemblies of such components by the FBA process. FRF based
assemblies provide an effective means for revealing how excitations on one component affect
responses at other components in the assembly.
Exterior Acoustics. MD R2 Nastran is the production release of exterior acoustics. It extends
acoustics applications to unbounded regions such as radiated acoustics from an engine. A new
sparse solver is available for efficient solution of the unsymmetric matrices encountered in
exterior acoustics frequency response.
Numerical Enhancements
Sparse Solvers. Two new sparse solvers have been introduced; TAUCS (statics) and UMFPACK
(unsymmetric). The UMFPACK solver provides scalable performance for exterior acoustics. In
addition, Lanczos has been enhanced to take advantage of available memory. Automatic optimal
reordering selection has been implemented for solid models to eliminate having the user set
flags.
Iterative Solvers. Restrictions to the CASI iterative solver have been relaxed for statics of large
solid models (engines). This includes an expanded list of supported element types.
ACMS. Automated Component Modal Synthesis (ACMS) has been extended to External
Superelements to provide significant reductions in compute time, I/O and scratch space. One
typical case study demonstrates an order of magnitude improvement.
Other HPC enhancements. MD R2 Nastran has been ported to Microsoft Compute Cluster. The
compute kernels for x86_64 platforms have been optimized for both Intel and AMD based
systems. Improved user diagnostics provides a pivot ratio bar chart to localize model
singularities.
Elements & Connectors
Connectors. A new seamweld (CSEAM) connector element is now available for assembly
modeling. It features extended capabilities for connecting higher order elements, mesh
independent connections to top/bottom shell patches defined either by property IDs or Element
IDs, tailored parts connection, and support for anisotropic material properties. For spot weld
5
CHAPTER 1
elements (CWELD, CFAST) end point displacement output can be obtained to view the
relationship between the spot weld and the connecting shells. A new connector type RBE2GS is
introduced to optionally search and connect independent grids of the two closest RBE2 elements
with a specified search radius.
Composite Beam Using VAM. Arbitrary beam cross section has been enhanced to support
composite materials. The 3-noded composite beam element represents the cross-section and lay-
up of plies, and accounts for interaction among plies. It provides an efficient alternative to
conventional 3-D modeling for beam-like composite structures such as rotor blades and shell
stiffeners.
Optimization
Topology Optimization. This release provides combined topology, sizing and shape
optimization simultaneously to find possible better designs. Different mass targets can now be
applied on multiple design parts of the structure. Symmetry constrains have been extended to
cyclical applications such as car wheels. An adjoint design sensitivity analysis method has been
implemented for inertia relief sizing optimization. Significant performance enhancement with
minimum member size control has been achieved particularly for large number of sizing design
variables.
Automatic External Superelement Optimization (AESO). This new feature automatically
partitions the model into a designed and non-designed part (external Superelement) for efficient
optimization. Order of magnitude speedup can be achieved without requiring user knowledge of
superlements.
Randomization (Pre-release). This randomization capability provides a way to stochastically
introduce uncertainty into a model such as tolerances in connectivity, properties and loads. The
user selects the outputs to monitor. This beta capability is a first step in developing a multi-run
environment to spawn multiple jobs, collect the results and perform statistical post-processing.
Rotor Dynamics & Aeroelasticity
Rotor Dynamics. Unbalance loading can now be used for frequency response with the rotor
dynamics option. Frequency response case control can be used directly in SOL 146 since we can
now handle multiple RGYRO subcases in rotor dynamics. Damping specification has been
simplified and allow for new damping formulations such as hybrid damping. As a prerelease
only capability: the effects of rotor stiffness, mass, and damping effects can be included in SOL
200 optimization.
Aeroelasticity. Monitoring points can now be updated and summed. A new type of monitoring
point (MONCNCM) has been introduced for monitoring stripwise aerodynamic results such as
lift and pitching moments. Other enhancements include various splining techniques for
aerodynamic structural applications.

6
List of MD Nastran Documents Released with MD R2 Nastran
Along with this Guide, the following documents are updated for the MD R2 Nastran release.
MD Nastran Quick Reference Guide
MD Users Guide
MD Nastran Installation and Operations Guide
MD Nastran Explicit Nonlinear (SOL 700) Users Guide
Ch. 2: Contact in Linear and Nonlinear Solutions MD R2 Nastran Release Guide
2
Contact in Linear and Nonlinear
Solutions
Linear Contact, Permanent Glued Contact, and Nonlinear Contact
Input
Output
Guidelines and Limitations
Examples

8
Linear Contact, Permanent Glued Contact, and
Nonlinear Contact
Introduction
MD R2 Nastran introduces linear analysis two new types of 3D contact and into SOL 400 nonlinear
analysis general 3D contact along with 2D solid edge to edge contact.
Linear Contact Model i ng i n SOL 101
Linear contact is defined as the full nonlinear contact algorithm of SOL 400 without material
nonlinear requirements and the usual linear requirements of small strain and small rotation imposed.
The contact bodies need not be in i ni tial contact, and multiple contact bodies are allowed. Only
surface to surface 3D contact is currently supported. The grids of the contacting bodies need not be
aligned, and the contact algorithm may be used to join dissimilar meshes. Both deformable-
deformable and deformable-rigid contact is allowed.
Bilinear Coulomb or bilinear shear friction is allowed. I n the general bilinear friction algorithms the
relative tangental displacement is split into stick (elastic contribution) and slip (plastic contribution)
components. I n SOL 101, only an elastic contribution (stick) to relative tangental displacement is
computed.
Note that the linear GAP contact defined by PARAM,CDITER,n is still supported, but should not be used
in association with the surfaces defined for linear contact.
Permanent Glued Contact Model i ng i n SOLs 101, 103,
105, 107, 108, 109, 110, 111, and 112
Permanent Glued contact is defined as a special type of contact model which imposes the condition
that between the contacting surfaces, there is no relative normal or tangental motion.
For Permanent Glued contact, MD Nastran will form the required constraints without entering the
full nonlinear contact algorithm. I f there is no initial contact between the contacting bodies these
constraints can not be formed and the run will fail. I f you cannot explicitly establish the contact
surfaces, you can set contact surface tolerances and any grids falling within these tolerances will be
GLUED. With this option you can also require that the algorithm actually bring the surfaces in to
contact before applying the constraints.
Nonl i near Contact Model i ng i n SOL 400
A full nonlinear 3D surface to surface contact algorithm and 2D solid edge to edge contact
algorithm is available in SOL 400 that works in conjunction with all the large strain - large
r ot at i on and material nonlinearity now available in SOL 400. The contact bodies need not be in
initial contact, and multiple contact bodies are allowed. Both deformable-deformable and
9
CHAPTER 2
deformable-rigid contact is allowed. The algorithm works for both nonlinear static and nonlinear
transient analysis in SOL 400.
Bilinear Coulomb or bilinear shear friction is allowed. I n the general bilinear friction algorithm,
the relative tangental displacement is split into stick (elastic contribution) and slip (a plastic
contribution) component.
The permanent glued contact is available for joining dissimilar meshes.
Benefits
The primary benefit of Linear Contact Modeling in SOL 101 is that you can use the full SOL 400 contact
algorithm for linear structural models where the only nonlinearity is the contact. Also, if after running
such a model in SOL 101 the user determines that there are other nonlinear effects such as material
nonlinearity or large rotation, the model can simply be switched to SOL 400 without having to redefine
the contact surfaces as both SOL 400 and SOL 101 use the same contact definitions.
The primary benefit of the Permanent Glued contact is the joining of two dissimilar meshes.
The primary benefit of Nonlinear Contact Modeling in SOL 400 is that you can perform general
multibody contact under full nonlinear conditions of large contact motion, nonlinear materials, large
rotation, and large strain.
All of the new contact capabilities use the same user interface.
Input
The Linear Contact, Permanent Glued contact, and General Nonlinear Contact are all controlled by the
Case Control command, BCONTACT. The following lists the appropriate user interface for SOL 101,
SOL 400, and Permanent Glued contact.
1. When used with SOL 101, SOL 400, or Permanent Glued contact in the previously described
linear solutions, only the BCONTACT = n or BCONTACT=NONE (default) are valid. For
BCONTACT = n, n can refer to the BCTABLE, BCHANGE, or BCMOVE entries.
2. For SOL 101 and SOL 400, the following associated Bulk contact entries are supported:
a. BCBODY is supported except for the following options:
HEAT, POLY, CYLIND, SPHERE, LINE, ARC, and NURBS2.
b. BCHANGE is supported except for the following options:
EXCLUDE
c. BCMOVE
d. BCPARA is supported except for the following options:
NBODIES, MAXENT, MAXNOD, ISPLIT, MAXSEP, ICHECK, IPRINT
FTYPE Only supports 0 (default for SOL 400), 6 and 7

10
FKIND, BEAMB, FSSMULT, FSSTOL, LINQUAD, INITCON
e. BCPROP
f. BCTABLE with the following NOT SUPPORTED
HHHB entry
FK through TBLCID
3. Prior to bodies coming into contact due to load application it may be desirable to initially identify
contacting bodies or specify initial stress-free contact. Thus Bulk Data entries such as BCPARA
have an ID field which may be 0 or blank (For the BCPARA entry, BCPARA,0 is usually the
typical and only usage as it is used to define initial contact parameters.) For other entries such as
a BCTABLE entry, a BCTABLE,0 entry would be used to identify the contact bodies that can
possibly contact each other at the start of the analysis. Then BCTABLE, n with n > 0 and called
out in Case control by a BCONTACT = n entry is used to control contact during loading. In SOL
400, a BCONTACT=0 is allowed above all subcases but is not required. ANY of the contact Bulk
Data entries that allow a 0 and HAVE a 0 value ID field are automatically sensed by SOL 400
with or without a BCONTACT = 0 command.
4. The PARAM,LMFACT,value and PARAM,PENFN,value can be used to control the penalty
function value for contact.
5. The PARAM, MARCREVR, value can be used to reverse the motion of rigid contact surfaces if
you have entered the patch information in the wrong order.
Output
The Case Control command BOUTPUT controls contact output. For the BOUTPUT Case Control
command in SOLs 101 and 400, SORT2 and PLOT are not currently supported.
1. Surface-to-surface 3D contact is currently supported in SOL 101 and SOL 400. Additionally, 2D
solid contact is supported in SOL 400. For solids contacting solids, the contact surfaces contain
the grids. For shell surfaces contacting each other, the contacting surfaces do not contain the grids.
Shells have thickness. For contacting shell surfaces, the distance between grids is ,
where and are the thickness of the two contacting shells. For deformable-rigid contact,
contact would occur when the position of the shell grid plus or minus half the thickness projected
with the normal comes into contact with the rigid segment. It is important to note that, if you
specifiy that the shell grids are touching, the algorithm will assume contact penetration.
2. SOL 101 allows linear contact. This means the SOL 400 full nonlinear contact algorithm
without material nonlinearity and standard linear small strain and small rotation is assumed.
3. Permanent Glued Contact is defined when the IGLUE field of the BCTABLE is set to 1. In this
option, all degrees of freedom of the contact grids are MPCd in the case of deformable-
deformable contact once the grids have come in contact. The relative tangential motion of a
contact grid is zero in the case of deformable-rigid contact. Permanent Glued Contact is available
t
1
t
2
H ( ) 2
t
1
t
2
11
CHAPTER 2
in SOLs 101, 103, 105, 107, 108, 109, 110, 111, and 112, as well as SOL 400. If IGLUE=1 on
the BCTABLE, MD Nastran will form the required constraints without entering the full nonlinear
contact algorithm. If there is no initial contact between the contacting bodies, these constraints
can not be formed and the run will fail. Therefore special cases arise, described as follows:
a. You want to run permanent glue contact in SOL 101 or SOL 400 but there is no initial contact
between the contacting bodies. Use entry BCPARA,0,NLGLUE,1 to turn on the general SOL
400 contact algorithm. Use Case Control BCONTACT etc. to define possible contacting
surfaces. The algorithm will then determine the contacting surfaces and glue the bodies
together.
b. If, in SOL 400 on the BCTABLE, there are multiple GLUE and non-GLUE entries associated
with different SLAVE entries, then, currently BCPARA,0,NLGLUE,1 must be used.
c. Only SOL 101 and SOL 400 can call the general nonlinear contact algorithm. If in SOLs 101
(optional method), 103, 105, 107, 108, 109, 110, 111, and 112 the user wishes to form a
permanent glue contact and there is no initial contactbetween the contacting bodies, on the
BCTABLE,0, entry specify a value for the ERROR field. Any grid within this error tolerance
will be considered to be in contact. If the user sets ICOORD=1 on the BCTABLE,0 entry then
the grids will be physically moved so that the surfaces are actually in contact. You may also
use the bulk entry BCPARA,0 field ISHELL to have the algorithm ignore shell thickness
when computing contact penetration.
Do not use BCPARA,0,NLGLUE,1 with this option.
4. Each grid and element should be, at most, in one body. Solids and shells should not be mixed in
the same body.
5. Deformable bodies can contact rigid bodies. Contact between rigid bodies is not supported.
6. In the following text, examples of contact bulk data entries and their fields are given. Only one
Entry, Field is given. The field however may appear on multiple Entries. Field values on
the BCPARA entry are overridden by Field values on the BCTABLE.
7. When defining deformable-deformable contact, the default rule is that the grids of a body may
contact any other body and the searching starts with the body having the lowest ID number and
ends with the body having the highest ID number. Depending on the contact constraints found, it
is possible that grids of a body with a higher ID number contact surfaces of bodies with a lower
ID number. This default behavior (so called double sided contact) can be overruled by the user in
different ways, namely via the BCBODY option or the BCTABLE option.
Via the BCBODY option, two different global search orders can be defined. The first is so called
single sided contact, which means that the search for contact for grids of a deformable body is
only done with respect to surfaces of the body itself and of bodies with a higher ID number. The
other activates a process which tries to optimize the set of contacting grids and contacted surfaces
and is only recommended for rare cases where a proper set of MASTER and SLAVE bodies
cannot be defined by the user.

12
The recommended way of overruling the default search process is via the BCTABLE option,
where a local search method can be defined per body combination. The BCTABLE option allows
the user to define which pair of contact bodies can potentially come into contact, and if they can
come into contact, it is also possible to specify that grids of the body indicated as SLAVE can
only contact surfaces of the body indicated as MASTER, and not the other way around. In this
way one can define single sided contact per body combination, thus overruling the default double
sided contact.
8. When defining deformable-deformable contact, it is important to ensure that grids of the body
with the finer mesh contacting surfaces of a body with a coarser mesh. When defining
deformable-rigid contact, it is important that the deformable bodies all have ID numbers lower
than the rigid bodies.
9. Softer bodies should be SLAVEs. E.g., rubber should be a SLAVE and steel should be a
MASTER.
10. Avoid sharp corners in contacted surfaces. It is better to have a finer mesh and smooth surface
where possible.
11. In sliding contact try to have smooth surfaces. Use the SPLINE option, which generates a Coons
surface (where the mesh surfaces are patched). This helps considerably by avoiding scattering
and accelerating convergence.
12. Because contact can entail large motion, it is recommended that a full Newton iteration scheme
be used. This has can be facilitated on the NLPARM Bulk Data entry by the addition to the
KMETHOD field of the keyword FNT, and to the TSTEPNL Bulk Data entry by the addition
to the METHOD field of the keyword FNT.
13. The motion of deformable bodies is prescribed by applying displacements or loads to the bodies.
It is recommended that enforced displacements or point loads not be placed on grids that might
come into contact with a rigid body.
14. It is recommended that MPC relationships be avoided for grids that might come into contact.
15. In SOL 400 2D Solid edge to edge contact, it is not recommended that a higher order CQUAD,
TRIAX, CQUADX element be used with the contact algorithm unless they have an associated
PSHLN2 entry.
16. Rigid bodies may have prescribed velocity, position, or load. See BCBODY entry.
17. During the contact process, it is unlikely that a grid exactly contacts the surface. For this reason,
a contact tolerance is associated with each contact body pair. If a grid is within the tolerance, it is
considered to be in contact. The algorithm computes the tolerance as the smaller of 5% of the
smallest element size (Solids) or 25% of the smallest element thickness (Shells). The user can
manually set the tolerance, see BCTABLE entry field ERROR. In general, the default is
recommended.
18. The tolerance value lies equally above and below the contacted surface. That is to say, the
thickness of the contacted surface is twice the value of the Tolerance. Often the grids are almost
touching the contacted surface. The algorithm provides for a biased tolerance with smaller
distance on the outside (the side nearest the contacting surface) and a larger distance on the inside.
This avoids the close grids from coming into contact and separating again. There is a default bias
13
CHAPTER 2
of 0.9 in the SOL 400 algorithm. The user may set the bias. See BCTABLE entry field BIAS. The
outside contact area is computed as (1-BIAS)*ERROR and the inside contact area is
(1+BIAS)*ERROR. To turn off the BIAS enter a small value on the BCTABLE entry for BIAS
such as 1.0-18 do not enter 0.0 value as blank or 0.0 reverts to a default 0.9 value.
19. In SOL 101 and SOL 400, bilinear Coulomb (see BCPARA entry, FTYPE = 6) or bilinear shear
(see BCPARA entry, FTYPE = 7) friction is available. Bilinear shear is usually used for forging
processes. Bilinear Coulomb is used for most friction applications.
20. Currently, quadratic (8-noded) shell elements are not recommended for use with contact.
21. Bilinear Coulomb friction is based on relative tangential displacements. The model assumes that
the stick and slip conditions correspond to reversible (elastic) and permanent (plastic) relative
displacements. The model defines a slip surface defined as:
where is the tangential force and is the normal force. is the friction coefficient (See
BCTABLE entry field FRIC). Stick is given by .
The rate of relative tangential displacement is split into elastic and plastic contributions according
to:
and the rate of change of friction force vector is related to the elastic tangential displacement by:
where the matrix is given by:
where is the slip threshold below which sticking is simulated. The algorithm
determines as 0.0025 times the average edge length of the elements defining the deformable
contact bodies.
The physically impossible case where , implies that the plastic or slip contribution must be
determined.
The determining relationship is:
ft Z f
n
ft f
n

0 <
u
t u
t
e
Z u
t
p
H
f
t Du
e
t Z
D
D
f
n
JJJJJJJJ 0
0
f
n
JJJJJJJJ
Z
0 >
f
t
D u
t
u
t

Z

14
The bilinear Coulomb model then assumes a slip flow potential given by and further
assumes that the direction of the slip displacement rate is given by the normal to the slip flow
potential. Note that this is similar to a nonassociative rule, as and are different functions.
The bilinear Coulomb model then defines:
where is the slip displacement rate. Also, since the friction force must always lie on the slip
surface:
Combining the preceding two equations the slip displacement rate is determined as:
The rate of change of friction force vector then becomes:
The bilinear Coulomb model also uses the check on convergence to determine if convergence is
achieved when
where is the current total friction force vector and is the total friction force vector from
the previous iteration. is the friction force tolerance, which has a value of 0.05.
22. The bilinear Coulomb model may not correlate well with experiment if the frictional tractions
exceed the flow stress or the stress of the material. If this occurs, the user can modify the frictional
stress limit (see BCTABLE entry field FRLIM). In the case of the frictional stress limit if
the surface traction reaches the limit value, the applied friction force is reduced, so that the
maximum shear traction is given by . The user may also switch to the bilinear shear
friction model.
23. The bilinear shear friction model states that the frictional stress is a fraction of the equivalent
stress in the material:
f
t
Z

u
p
t

ft
JJJJJJJJ Z

ft
JJJJJJ

T
f
t 0 Z Z
ft
JJJJJJ

T
Du
ft
JJJJJJ

T
D
ft
JJJJJJJ
JJJJJJJJJJJJJJJJJJJJJJJJJJJJ Z
f
t D
D
ft
JJJJJJJJ

ft
JJJJJJ

T
D
ft
JJJJJJ

T
D
ft
JJJJJJJJ
JJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJ

u
t D D
( )u
t Z Z
F
t
F
t
previous
F
t
JJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJ e
F
t
F
t
previous
e
t
limit
mi n
n

t
limit
, ( )
15
CHAPTER 2
for stick and for slip
where is the surface tangent vector. is the friction coefficient (See BCTABLE entry field
FRIC). is computed by:
Examples
Below are examples for Glued Contact and Linear Contact. Other examples of contact can be found in
the MD Users Guide.
Example 1: Glued Contact (s103perg.dat)
As an example of permanent glued contact in SOL 103, Figure 2-1 shows a deformable cylinder made
up of 10 noded TETRA elements and a bracket structure also made up of 10 noded TETRA elements
sitting on top of the cylinder. We wish to glue the two structures together and do a ground check on the
structures before proceeding with further analysis. The gluing is accomplished with the use of the Case
Control entry BCONTACT=1 as shown in the abbreviated bulk data file. The BCONTACT=1 Case entry
points to a BCTABLE with ID=1 and 1 value for the NGROUP field of the BCTABLE entry indicating
that 1 pair of structures are to be glued.
Figure 2-1 Glued Contact
t

3
JJJJJJJ <
t

3
JJJJJJJ Z t
t
t
min
n

3
JJJJJJJ ,

Z

16
SOL 103
CEND
TITLE = Gluing with TETRA10 with flat interface
$
echo = sort(except ctetra,grid,bsurf,pload4)
autospc(noprint) = yes
SUBCASE 1
SUBTITLE= Modal analysis
LABEL = Ground Check for rigid body modes
BCONTACT = 1
METHOD=1
GROUNDCHECK(SET=ALL,DATAREC=YES)=YES
BEGIN BULK
PARAM,WTMASS,.00259
PARAM PRTMAXIM NO
BCPARA,0,BIAS,0.0
$
$ 1 >< 2 >< 3 >< 4 >< 5 >< 6 >< 7 >< 8 >< 9 >< 10 >
EIGRL120
BCTABLE 1 1
SLAVE 3 0.0001 0. 0. 0. 1
1
MASTERS 2
$ Deform Body Contact LBC set: cylinder
BCBODY 2 3D DEFORM 2 0
BSURF 2 1 2 3 4 5 6 7
8 9 10 11 12 13 14 15
.
.
.
4960 4961 4962 4963 4964 4965 4966 4967
4968 4969 4970 4971 4972 4973
$ Deform Body Contact LBC set: bracket
BSURF 3 4974 4975 4976 4977 4978 4979 4980
4981 4982 4983 4984 4985 4986 4987 4988
.
.
.
9413 9414 9415 9416 9417 9418 9419 9420
9421 9422 9423 9424 9425 9426 9427 9428
$ Model definition follows:
$
GRID 1 .446672 2.59695 -.063605
GRID 2 .50144 -.5775 -.085552
GRID 3 .323733 2.69367 -.571285
.
.
.
$
PSOLID 1 1 0
$ Pset: cylinder will be imported as: psolid.1
CTETRA 1 1 617 616 615 614 621 622
623 618 619 620
CTETRA 2 1 613 612 611 610 627 628
629 624 625 626
17
CHAPTER 2
CTETRA 3 1 617 616 609 608 621 633
634 630 631 632
.
.
.
$ Material Record : al
$ 1 >< 2 >< 3 >< 4 >< 5 >< 6 >< 7 >< 8 >< 9 >< 10 >
MAT1 1 1.+7 .33 0.1
$
ENDDATA
The second line of entry on the BCTABLE shows that the bracket structure is going to be the contact
body or the touching body. This is accomplished by the IDSLA1=3 field entry pointing to a BCBODY
entry with the same BID=3.
Most importantly in field 8 of the SLAVE entry record, a value of 1 for IGLUE field is set. This entry
is what allows SOL 103 to perform a permanent glued contact.
The continuation entry after the SLAVE entry show a 1 value for ISEARCH. This tell the algorithm to
search touching to touched. The entry MASTER defines the cylinder as the touched body. Its IDMA1=2
field entry points to a corresponding BCBODY entry with BID=2 value.
The BCBODY entries show that both structures are to be considered as deformable and that the 3D
contact algorithm is to be used. (This is the only algorithm valid in linear analysis.) The BCBODY entry
with field BID=3 has its field 5 BSID =3 which corresponds to a BSURF Bulk Data entry of the same
value for BSID. The BSURF then lists the surface elements of the bracket that can come into contact.
A similar BSURF is present for the cylinder.
Note that the Bulk Data also contains a BCPARA,0 entry. The ID=0 on this entry sets parameter values
for the contact prior to any analysis beginning. If there is a BCONTACT=n in any case control, MD
Nastran will automatically search for the existence of a BCPARA,0 entry to initialize the contact
algorithm.
The results for the model show that all GROUNDCHECK requirements PASS and the first 20 modes are.

18
*** USER INFORMATION MESSAGE 7570 (GPWG1D)
RESULTS OF RIGID BODY CHECKS OF MATRIX KGG (G-SET) FOLLOW:
PRINT RESULTS IN ALL SIX DIRECTIONS AGAINST THE LIMIT OF 2.674629E-03
DIRECTION STRAIN ENERGY PASS/FAIL
--------- ------------- ---------
1 1.100904E-07 PASS
2 7.325037E-08 PASS
3 4.130997E-08 PASS
4 5.545296E-07 PASS
5 1.048790E-07 PASS
6 3.550209E-07 PASS
*** USER INFORMATION MESSAGE 7570 (GPWG1D)
RESULTS OF RIGID BODY CHECKS OF MATRIX KAA1 (A-SET) FOLLOW:
PRINT RESULTS IN ALL SIX DIRECTIONS AGAINST THE LIMIT OF 2.674629E-03
DIRECTION STRAIN ENERGY PASS/FAIL
--------- ------------- ---------
1 1.096489E-07 PASS
2 7.120107E-08 PASS
3 3.338044E-08 PASS
4 5.495763E-07 PASS
5 1.451471E-07 PASS
6 5.887807E-07 PASS

R E A L E I G E N V A L U E S
MODE EXTRACTION EIGENVALUE RADIANS CYCLES GENERALIZED GENERALIZED
NO. ORDER MASS STIFFNESS
1 1 -3.343832E-05 5.782588E-03 9.203274E-04 1.000000E+00 -3.343832E-05
2 2 -1.705077E-05 4.129258E-03 6.571918E-04 1.000000E+00 -1.705077E-05
3 3 -1.452048E-06 1.205010E-03 1.917832E-04 1.000000E+00 -1.452048E-06
4 4 2.856250E-06 1.690044E-03 2.689789E-04 1.000000E+00 2.856250E-06
5 5 1.440756E-05 3.795729E-03 6.041091E-04 1.000000E+00 1.440756E-05
6 6 2.289412E-05 4.784780E-03 7.615214E-04 1.000000E+00 2.289412E-05
7 7 6.371218E+07 7.981991E+03 1.270373E+03 1.000000E+00 6.371218E+07
8 8 4.260703E+08 2.064147E+04 3.285192E+03 1.000000E+00 4.260703E+08
9 9 1.158678E+09 3.403937E+04 5.417533E+03 1.000000E+00 1.158678E+09
10 10 1.949990E+09 4.415869E+04 7.028074E+03 1.000000E+00 1.949990E+09
11 11 4.246696E+09 6.516668E+04 1.037160E+04 1.000000E+00 4.246696E+09
12 12 4.305021E+09 6.561266E+04 1.044258E+04 1.000000E+00 4.305021E+09
13 13 4.692808E+09 6.850407E+04 1.090276E+04 1.000000E+00 4.692808E+09
14 14 6.424801E+09 8.015486E+04 1.275704E+04 1.000000E+00 6.424801E+09
15 15 7.120649E+09 8.438394E+04 1.343012E+04 1.000000E+00 7.120649E+09
16 16 9.036405E+09 9.506001E+04 1.512927E+04 1.000000E+00 9.036405E+09
17 17 9.057384E+09 9.517029E+04 1.514682E+04 1.000000E+00 9.057384E+09
18 18 9.837994E+09 9.918666E+04 1.578605E+04 1.000000E+00 9.837994E+09
19 19 1.080527E+10 1.039484E+05 1.654390E+04 1.000000E+00 1.080527E+10
20 20 1.156105E+10 1.075223E+05 1.711271E+04 1.000000E+00 1.156105E+10

1 * * * END OF JOB * * *
19
CHAPTER 2
Example 2: 3D Contact (s101bmcn.dat)
As an Example of 3D contact in SOL 101 Figure 2-2 shows two beams. The left beam is clamped at its
left end and is 10 units of length long. The right beam is clamped at its right end and is 0.15 units above
the left beam. It is 10 units of length long and over laps the right end of the left beam by two units of
length. Both beams are modeled using CQUAD4 elements. The upper right beam has a uniform pressure
load applied along its entire length. It is desired to know how the upper beam contacts the lower beam
under the assumption that the beams do not under go large rotation or experience large strain or nonlinear
material behavior. (If on running such a model in SOL 101 the user determines that these assumptions
are incorrect, He or She can without changing the contact definition run the model in SOL 400.)
Figure 2-2 3D Contact in SOL 101
The abbreviated Bulk Data file is used for discussion:
SOL 101
CEND
SUBCASE 1
BOUTPUT = ALL
BCONTACT = 1
NLPARM = 1
SPC = 2
LOAD = 3
DISPLACEMENT(SORT1,REAL)=ALL
SPCFORCES(SORT1,REAL)=ALL
STRESS(SORT1,REAL,VONMISES,BILIN)=ALL
FORCE = ALL
BEGIN BULK

20
PARAM POST 0
PARAM PRTMAXIM YES
$
NLPARM 1 20 FNT
BCTABLE 1 1
SLAVE 3 0. 0. 0. 0. 0
1 0 0
MASTERS 4
$
$ Deform Body Contact LBC set: left
$
BSURF 3 1 2 3 4 5 6 7
8 9 10 11 12 13 14 15
16 17 18 19 20 21 22 23
24 25 26 27 28 29 30 31
32 33 34 35 36 37 38 39
40
$
$ Deform Body Contact LBC set: right
$
BSURF 4 101 102 103 104 105 106 107
108 109 110 111 112 113 114 115
116 117 118 119 120 121 122 123
124 125 126 127 128 129 130 131
132 133 134 135 136 137 138 139
140
$
$ Elements and Element Properties for left beam
$
PSHELL 1 1 .1 1 1
$
CQUAD4 1 1 1 2 23 22
CQUAD4 2 1 2 3 24 23
CQUAD4 3 1 3 4 25 24
.
.
.
$
$ Elements and Element Properties for right beam
$
PSHELL 2 2 .1 2 2
$ Pset: right will be imported as: pshell.2
CQUAD4 101 2 101 102 123 122
CQUAD4 102 2 102 103 124 123
CQUAD4 103 2 103 104 125 124
.
.
.
$
$ Material Record : left
MAT1 1 1.+6 .3 2.59-4
$ Material Record : right
MAT1 2 1.+6 .3 2.59-4
$ Nodes Left Beam
GRID 1 0. 0. 0.
GRID 2 .5 0. 0.
GRID 3 1. 0. 0.
GRID 4 1.5 0. 0.
21
CHAPTER 2
.
.
.
$ Nodes Right Beam
GRID 101 8. 0. .15
GRID 102 8.5 0. .15
GRID 103 9. 0. .15
.
.
.
$ Loads and Boundary Contitions
$
SPCADD 2 1 3
LOAD 3 1. 1. 2
SPC1 1 123456 1 22 43
SPC1 3 123456 121 142 163
$ Pressure Loads right beam
PLOAD4 2 101 -.5 THRU 140
$ Referenced Coordinate Frames
CORD2R 1 0. 0. .05 0. 0. 1.05
1. 0. .05
ENDDATA
The BOUTPUT = ALL entry requests contact force information
The BCONTACT=1 Case entry points to a BCTABLE with ID=1 and 1 value for the NGROUP field of
the BCTABLE entry indicating that 1 pair of structures are to be glued. (The number of
SLAVE,MASTERS groups.)
The second line of entry on the BCTABLE,1 entry identifies the SLAVE body as IDSLA1=3 and the
ISEARCH=1 value on the continuation line following the SLAVE entry line indicates to the search
algorithm that touching to touched is to be invoked. The IDSLA1=3 field entry points to a corresponding
BCBODY entry with BID=3 value as the touched body.
The entry MASTERS with its IDMA1=4 field entry points to a corresponding BCBODY entry with
BID=4 value as the touched body.
The BCBODY entries show that both structures are to be considered as deformable and that the 3D
contact algorithm is to be used. (This is the only algorithm valid in linear analysis.) The BCBODY entry
with field BID=3 has its field 5 BSID =3 which corresponds to a BSURF Bulk entry of the same value
for BSID. The BSURF then lists the elements of the left beam.
A similar BSURF is present for the right beam.
One additional Case Control and Bulk Data pair shown is the Case Control NLPARM=1 command and
its corresponding NLPARM,1 Bulk Data entry. These entries are not required and MD Nastran will
generate internally, based on the presence of a BCONTACT=n command, a default NLPARM entry. The
presence of this Case Control NLPARM=1 command and its corresponding NLPARM,1 Bulk Data entry
allows the user to change the defaults as deemed necessary. In this case 20 iterations are requested using,
the recommended for contact, the full Newton method (FNT).

22
The following partial f06 file shows some output unique to SOL101 when running contact and
BOUTPUT is present in Case Control. A status of 1 in the contact table indicates touching nodes and a
status of 0 in the table indicate the grids of the touched surface.
N O N - L I N E A R I T E R A T I O N M O D U L E O U T P U T
STIFFNESS UPDATE TIME-1892160000.00 SECONDS SUBCASE 1 STEP 0
ITERATION TIME 0.01 SECONDS
LOAD NO. - - ERROR FACTORS - - CONV ITR MAT NO. AVG TOTL - - - - - DISP - - - - - - LINE_S NO. TOT TOT
STEP INC ITR DISP LOAD WORK RATE DIV DIV BIS R_FORCE WORK AVG MAX AT GRID C FACT NO QNV KUD ITR
% 1.0000 20 3 4.09E-05 4.53E-03 5.91E-05 0.700 0 1 0 1.18E-05 2.731E+00 1.72E-01 -3.423E+00 42 3 1.00 0 0 0
64

0 SUBCASE 1
N O D A L F O R C E S A N D S T R E S S E S I N 3 D C O N T A C T A N A L Y S I S
LOAD STEP = 1.00000E+00 SUBCASE 1 STEP 0
GRID STATUS CONTACT FORCES - RESIDUALS FRICTION FORCES NORMAL FRICTION FRICTION
EXCLUDING FRICTION STRESS STRESS 1 STRESS 2
- IN GLOBAL SYSTEM - - IN GLOBAL SYSTEM - - IN CONTACT LOCAL SYSTEM -
17 1 1.1705E-01 -1.8131E-05 -4.4983E-01 0.0000E+00 0.0000E+00 0.0000E+00 3.8423E+00 0.0000E+00
0.0000E+00
38 1 8.6045E-02 -1.3307E-05 -3.3009E-01 0.0000E+00 0.0000E+00 0.0000E+00 1.4101E+00 0.0000E+00
0.0000E+00
59 1 1.1705E-01 1.8087E-05 -4.4983E-01 0.0000E+00 0.0000E+00 0.0000E+00 3.8423E+00 0.0000E+00
0.0000E+00
101 0 -1.1503E-01 1.7818E-05 4.4206E-01 0.0000E+00 0.0000E+00 0.0000E+00 7.5518E+00 0.0000E+00
0.0000E+00
102 0 -2.0232E-03 3.1340E-07 7.7754E-03 0.0000E+00 0.0000E+00 0.0000E+00 6.6415E-02 0.0000E+00
0.0000E+00
122 0 -8.4557E-02 1.3077E-05 3.2438E-01 0.0000E+00 0.0000E+00 0.0000E+00 2.7713E+00 0.0000E+00
0.0000E+00
123 0 -1.4867E-03 2.2991E-07 5.7031E-03 0.0000E+00 0.0000E+00 0.0000E+00 2.4363E-02 0.0000E+00
0.0000E+00
143 0 -1.1503E-01 -1.7774E-05 4.4205E-01 0.0000E+00 0.0000E+00 0.0000E+00 7.5518E+00 0.0000E+00
0.0000E+00
144 0 -2.0232E-03 -3.1263E-07 7.7753E-03 0.0000E+00 0.0000E+00 0.0000E+00 6.6414E-02 0.0000E+00
0.0000E+00
STATUS DEFINITION -----
= 0 node is a retained node
= 1 node is a tied node (touching node)
*** JOB CONVERGES FOR THE CURRENT STEP.
*** SUBCASE 1 STEP 0 IS COMPLETED.
23
CHAPTER 2
Figure 2-3 shows the deformation of the two beams. Note that it appears that the two beams are not
actually touching. However, in 3D contact shells have actual thickness and the contact algorithm
accounts for this fact.
Figure 2-3

24
Ch. 3: Advanced Integrated Nonlinear Analysis MD R2 Nastran Release Guide

3
Advanced Integrated Nonlinear
Analysis
Advanced Integrated Nonlinear (SOL 400)
Kinematic Elements
SOL 400 Material and Elements
The Nonlinear QUADR/TRIAR Elements
CBAR Element to CBEAM Element Conversion
CSHEAR Element to CQUAD4 Element Conversion
SOL 400 Pretension Modeling Description
VCCT Virtual Crack Closure Method
Cohesive Zone Modeling
Nonlinear Iteration Algorithms
Analysis Chaining SUBCASE, STEP, ANALYSIS, and NLIC
Adaptive Time Stepping Scheme
The Numerical Integration Method for the Transient analysis the

HHT method
Nonlinear Iteration Summary Table for Nonlinear Transient Analysis

in SOL 400
Output Data Grouping: NLPACK


26
Advanced Integrated Nonlinear (SOL 400)
Introduction
This release extends the capabilities of SOL 400, including nonlinear static analysis and nonlinear
transient analysis. In this release guide, all items pertaining to SOL 400 are discussed. Some of the items
may have been discussed in the MD R1 release guide, but are included here for completeness and clarity.
Benefits
The benefits of SOL 400 are discussed in this section for the current release. Some of the benefits may
be subject to the limitations discussed in next section. The major benefits are:
A new general nonlinear contact analysis capability.
Extensive enhancements for the nonlinear large strain and material behavior.
The kinematic elements such as RBAR, RBE2 etc. can be combined with shell and beam
elements.
The QUADR/TRIAR elements are extended to nonlinear analysis in SOL 400.
The linear CBAR element can be converted into a geometric nonlinear CBEAM element.
The linear CSHEAR element can be converted into a geometric nonlinear CQUAD4 element.
Improved nonlinear iteration algorithms make solution easier and faster to converge. These
includes ADAPT, AUTO, ITER, SEMI, FNT, and PFNT methods.
Analysis chaining SUBCASE, STEP, ANALYSIS, and NLIC Case Control commands allow
flexible loading and solution sequences. Also, different nonlinear analysis types can
communicate with each other in a meaningful physical sequence in a SUBCASE
An adaptive time stepping scheme is implemented into SOL 400 to automatically modify the
size of a load increment or time step.
Other benefits are:
A transient solution integration method, the HHT (Hilbert-Hughes-Taylor) method, to give
stable transient solution.
Comprehensive nonlinear summary tables for the static analysis and the transient analysis.
Support initial conditions (IC) for the transient analysis.
Case Control command NLIC allows any previous converged nonlinear static solution as the
preload for the first nonlinear transient step.
The boundary conditions (SPC and MPC) are allowed to change between load steps.
Direct matrix input, such as K2PP, M2PP, B2PP, and TFL, are allowed to change between load
steps.
Support grid based reordering for a faster decomposition after a time step size change or
stiffness update in the nonlinear iterations.
27
CHAPTER 3
Support thermal loads in the transient analysis. Two new bulk data entries, TTEMP and
TMPSET, have been added for this new feature in the nonlinear transient analysis.
Support rotor dynamics in the nonlinear transient analysis.
A user-friendly restart procedure.
A more flexible output method for the nonlinear transient analysis. Users can use a new
parameter NLPACK to control the output and restart time steps.
Solution output is at user requested time for transient analysis. Also, allow the simulation of the
same output logic and format of SOL 129 by specifying a negative NO on the TSTEPNL bulk
data card.
Support linear superelements.
The nonlinear stresses for 3D elements selected by NLSTRESS can now be output in user
selected material coordinate system.
The OTIME output request is supported (except NLSTRESS).
Limitations for the Current Release
The followings are limitations for the current release of the nonlinear transient analysis:
Omitted degrees-of-freedom (o-set) may not be used.
Nonlinear normal modes and nonlinear buckling analysis are not supported.
Arch-length method (input by NLPCI bulk data entry) is not supported.
Restart is not available for the contact analysis and the newly enhanced large strain and material
behavior (elements with PSHNLi, PSLNL1, or PSHEARN bulk data entry).
These limitations will be remedied in a future release.
A New General Nonlinear Contact Analysis Capability
A general nonlinear contact analysis is implemented into SOL400, which is applicable for both the
nonlinear static analysis and the nonlinear transient analysis. This capability can be used with existing
Nastran nonlinear elements and the new enhanced large strain and material behavior. Both deformable-
deformable and deformable-rigid contacts are allowed. For details, please refer to section SOL 400
Contact.
Extensive Enhancements for Nonlinear Large strain and Material Behavior.
For this release, extensive enhancements for nonlinear large strain and material behavior have been
implemented in SOL400. For details, please refer to section SOL 400 Material and Elements on
page 36.

28
Kinematic Elements
The kinematic elements have been implemented in both nonlinear static analysis and transient analysis
for SOL 400. These elements are the rigid elements with Lagrange formulation used in the geometric
nonlinear analysis of large rotations. They include RBAR, RBAR1, RJOINT, RBE1, RBE2, RBE3,
RTRPLT, and TRTPLT1. This type of elements becomes linkage if any degree of free connecting to the
element is released. Appropriate constraints must be provided, otherwise the structure model will be
singular and the solution will diverge.
Introduction
The kinematic elements are the rigid elements with Lagrange formulation used in the geometric nonlinear
analysis of large rotations. They include RBAR, RBAR1, RJOINT, RBE1, RBE2, RBE3, RTRPLT, and
TRTPLT1. This type of elements becomes linkage if any degree of free connecting to the element is
released. Appropriate constraints must be provided, otherwise the structure model will be singular and
the solution will diverge. In the follow sections, kinematic elements and rigid elements are used
interchangeably.
The kinematic elements have been implemented into both nonlinear static analysis and nonlinear
transient analysis for SOL 400.
Benefits
The benefits of the kinematic elements are discussed in this section. Some of the benefits may be subject
to the limitations discussed in next section
The kinematic elements can be combined with shell, beam, and solid elements.
Differential stiffness for the kinematic elements is computed to facilitate the convergence of the
solution.
Appropriate scale factors for the Lagrange multiplier and penalty function are computed
automatically.
Allowed thermal loads to be used with the kinematic elements.
Both force and GPFORCE outputs can be requested for output.
Type of Elements that can be Combined with the Kinematic
Elements
The following type of elements can be combined with the Kinematic elements:
Shells and Beam CQUAD4, CQUADR, CTRIA3, CTRIA3, and CBEAM.
Solid elements CHEXA, CPENTA, and both 4 node and 10 node CTETRA.
The newly enhanced shell and solid elements Elements with the PSHLN1 and PSLDN1 Bulk
Data Entries.
29
CHAPTER 3
The CBAR element is a linear element. But user can convert it into a CBEAM element by the
Bulk Data entry MDLPRM, BRTOBM, 1. Example EX02 demonstrates this.
The CSHEAR element is a linear element. But user can convert it into a CQUAD4 element by
the Bulk Data entry MDLPRM, SHRTOQ4, 1.
The Scale Factors LMFACT and PENFN
The parameters LMFACT and PENFN are the scale factor and penalty function for the kinematic
elements. The purpose of LMFACT and PENFN is to make the values of stiffness matrix of the kinematic
elements and/or the contact components about the same relative magnitude as those of the other elements
in the model. Too small a value will produce inaccurate results and too large a value will produce
numerical difficulties. The same value is usually assigned to both LMFACT and PENFN. Under special
requirement, user may assign different values for LMFACT and PENFN. For example, if PENFN=0.0
and LMFACT0.0, then the solution method for the rigid elements becomes the pure Lagrange multiplier
method instead of the augmented Lagrangian method. However, user must exercise caution if different
values are assigned to LMFACT and PENFN.
The default value for the kinematic element is computed automatically by SOL 400 based geometry of
the individual element and the average magnitude of the stiffness matrix. This means that actual scale
factor is different for each kinematic element.
You can use PARAM, LMFACT, F1 and PARAM, PENFN, F2 to modify the SOL 400 computed value.
Thermal Effects
The thermal effects for the kinematic elements are computed if thermal load is requested. In this case,
the ALPHA field on the Bulk Data entry for element must be entered.
RJOINT and the Pin Flags of CBEAM Element
The Bulk Data entry for RJOINT element is:
GA is the independent grid point and all of its six degrees-of-freedom are the independent
degrees-of-freedom. GB is the dependent grid point. The length between points GA and GB
must be zero. Since its length is zero, the thermal load effect is not applicable.
If CB=123456 or blank, then the grid point GB is constrained to move with the grid point GA
and two the grid points move as a single point. If any degree-of-freedom is released on CB, then
RJOINT becomes a mechanical joint. A mechanical joint is a mechanical system that has two
bodies jointed at a point. The two bodies can rotate relatively about one, two or three axis of a
local coordinate system at the point. For example:
RJOINT EID GA GB CB

30
Hinge - A hinge is a mechanical joint that rotates freely about one axis of the local coordinate
system. It can be simulated by RJOINT with one rotational degree of freedom released, i.e., CB
= 12356, 12346, or 12345.
Universal joint - A universal joint is a mechanical joint that rotates freely in two axes. It can be
simulated by RJOINT with two rotational degrees for freedom released, i.e., CB = 1234, 1235,
or 1236.
Spherical joint - A spherical joint is a mechanical joint that rotates freely about all three axes. It
can be simulated by a zero length RBAR with all rotational degrees of freedom released, i.e. CB
= 123.
Prismatic joint - A prismatic joint is a mechanical system with two blocks that are constrained to
have the same rotations, but translate relatively with each other along a local axis. It can be
simulated by RJOINT with one translational degree of freedom released, i.e., CB = 23456,
13456, 12456.
Cylindrical joint - A cylindrical joint is a mechanical system that allows two grid points have
relative translation along a moving axis and, at the same, have relative rotation about the same
axis. It can be simulated by RJOINT with one translational degree of freedom and one rotational
degree of freedom released, i.e. CB = 2356, 1346, 1245.
The pin flag on the CBEAM element should not be used in the nonlinear analysis. It may or may not give
correct results depending structural model. Also, the differential stiffness for pin flag is not computed to
facilitate the convergence. Instead, the RJOINT should be used to simulate the pin flag. Example EX04
demonstrates this.
User Interfaces
Case Control Command RIGID
The Control Case command RIGID selects the type of rigid element. It has the following format:
LINEAR will select the linear rigid elements, LAGRAN will select the Lagrange rigid element with the
Lagrange multiplier method, and LGELIM will select the Lagrange element with the Lagrange
elimination method.
If RIGID command does not exist in the user Case Control file, for SOL 400, the selection
RIGID=LAGRAN is assumed. Example EX03 shows that RIGID=LAGRAN is not required in the Case
Control Packet.
Please note that, for rotor dynamics, the theory requires the linear rigid elements. Otherwise, the solution
will not converge. In this case, the RIGID=LINEAR must be given in the Case Control packet.

LINEAR
RIGID = LAGRAN
LGELIM
31
CHAPTER 3
For nonlinear solution sequences, the LGELIM is not available. If requested, a user fatal error will be
issue. Also for the nonlinear solution sequences, RIGID=LAGRAN is implemented for SOL 400 only.
A user fatal error will be issued if it is used in SOL 106 or 129.
Force and GPFORCE Output
The force output for the kinematic elements is requested by the MPCF Case Control command.
The GPFORCE output is requested by the GPFOR Case Control command.
Both force and GPFORCE outputs are requested in examples EX01 and EX02.
Bulk Data Entries
The kinematic elements are input using the Bulk Data entries: RBAR, RBAR1, RBE1, RBE2, RBE3,
RTRPLT, and RTRPLT1. If the thermal loads are used, the ALPHA field of these entries must be entered.
Examples
The following four examples show the inputs and capabilities of the rigid elements in combination with
other elements for nonlinear static or nonlinear transient analysis. The intention of these examples is to
show the input structure. The model itself and the detailed entries in the Bulk Data Deck are not
important.
Example 1
Example EX01 is a cantilever beam with two CQUAD4 elements, connecting to a RBE2 element with
two branches. The load is applied at the end of the RBE2. This is a nonlinear static geometric analysis
with 3D motion. The force and GPFORCE output for the RBE2 are requested by MPCF=all and
GPFOR=all.
ID MSC, EX01
SOL 400
CEND
TITLE = NONLINEAR STATIC ANALYSIS
SUBTITLE = QUAD4 + RBE2 LARGE ROTATION MPCF/GPFOR
RIGID = LAGRAN $ This entry is not required.
NLPARM = 10
SUBCASE 1
LOAD = 100
SPC = 10
DISP = ALL
SPCF = ALL
MPCF = ALL
GPFOR= ALL
BEGIN BULK
NLPARM,10,10,,,,,,,+NLP1
PARAM,LGDISP,+1
FORCE,100,7,0,50.0,0.0,0.0,-1.0
FORCE,100,8,0,50.0,0.0,0.0,-1.0
GRID,1,,0.0,0.0,0.0
GRID,2,,1.0,0.0,0.0
GRID,3,,2.0,0.0,0.0
GRID,4,,0.0,1.0,0.0
GRID,5,,1.0,1.0,0.0
GRID,6,,2.0,1.0,0.0
GRID,7,,2.0,2.0,0.0
GRID,8,,3.0,2.0,0.0
MAT1,1,4.+6,,0.3
CQUAD4,11,10,1,2,5,4

32
CQUAD4,12,10,2,3,6,5
PSHELL,10,1,0.1,1
RBE2,21,6,123456,7,8
SPC1,10,123456,1,4
ENDDATA
Example 2
Example EX02 is a cantilever frame with six CBAR elements and a RBE3 element connected to three
independent grids. Extra grid and RBAR are added to give the direction of follower force. The CBAR
element is a linear element. The entry MDLPRM, BRTOBM, 1 is added to convert the linear CBAR
element to the nonlinear CBEAM element. This is a static geometric nonlinear with 3D motions.
ID MSC, EX02
SOL 400
CEND
SUBTITLE = BEAM + RBE3 LARGE ROTATION FOLLOWER FORCE MPCF/GPFOR
RIGID = LAGR $ This entry is not required.
NLPARM = 10
SUBCASE 1
LOAD = 100
SPC = 10
DISP = ALL
SPCF = ALL
MPCF = ALL
GPFOR= ALL
BEGIN BULK
NLPARM,10,10,,,,,,,+NLP1
MDLPRM,BRTOBM,1
PARAM,LGDISP,+1
FORCE1,100,7,100.0,17,7
GRID,1,,0.0,0.0,0.0
GRID,2,,1.0,0.0,0.0
GRID,3,,2.0,0.0,0.0
GRID,4,,0.0,1.0,0.0
GRID,5,,1.0,1.0,0.0
GRID,6,,2.0,1.0,0.0
GRID,7,,3.0,2.0,0.0
GRID,17,,3.0,2.0,1.0
MAT1,1,1.+3,,0.3
CBAR,11,10,1,2,0.0,1.0,0.0
CBAR,12,10,2,3,0.0,1.0,0.0
CBAR,13,10,4,5,0.0,1.0,0.0
CBAR,14,10,5,6,0.0,1.0,0.0
CBAR,15,10,2,5,1.0,0.0,0.0
CBAR,16,10,3,6,1.0,0.0,0.0
PBAR,10,1,1.0,0.1,0.1,0.5
RBE3,21, ,7,123456,1.0,123456,3,5,+RBE3
+RBE3,6
RBAR,31,7,17,123456, , ,123456
SPC1,10,123456,1,4
ENDDATA
Example 3
Example EX03 is a model with solid elements and RBE2/3 elements. This is a nonlinear transient
analysis with 3D motions. There is no RIGID=LAGRAN in the Case Control packet to show that, for
SOL 400, RIGID=LAGRAN is default.
ID MSC, EX03
SOL 400
CEND
TITLE = NONLINEAR TRANSIENT ANALYSIS
SUBTITLE = HEXA/BEAM + RBE2/3
TSTEPNL = 10
SET 1000 = 11, 21, 221, 231, 100000
ANALYSIS=NLTRAN
SUBCASE 1
DLOAD = 200
33
CHAPTER 3
SPC = 10
DISP = 1000
BEGIN BULK
$
TSTEPNL 10 40 0.05 AUTO
PARAM,LGDISP,+1
$
$ BOUNDARY CONDITION
$
SPC1 10 123456 1 211 232 442
$
$ RBE ELEMENTS
$
RBE3 21 100000 123456 1.0 123 21 221 +RBE3
+RBE3 231
$
RBE2 252 252 123456 463
RBE2 357 357 123456 468
RBE2 462 462 123456 473
$
$ BEAM ELEMENTS
$
PBEAM 10 1 1.0 0.1 0.1 0.0 0.5
$
CBEAM 251 10 247 252 0.0 1.0 0.0
CBEAM 256 10 352 357 0.0 1.0 0.0
CBEAM 261 10 457 462 0.0 1.0 0.0
$
CBEAM 252 10 252 357 1.0 0.0 0.0
CBEAM 357 10 357 462 1.0 0.0 0.0
$
$ SOLID ELEMENTS AND ELEMENT PROPERTIES
$
PSOLID 1 1 0
$
CHEXA 1 1 1 6 111 106 232 237
342 337
CHEXA 2 1 6 11 116 111 237 242
347 342
CHEXA 3 1 11 16 121 116 242 247
352 347
CHEXA 4 1 16 21 126 121 247 252
357 352
CHEXA 5 1 106 111 216 211 337 342
447 442
CHEXA 6 1 111 116 221 216 342 347
452 447
CHEXA 7 1 116 121 226 221 347 352
457 452
CHEXA 8 1 121 126 231 226 352 357
462 457
$
$ Referenced Material Records
$
MAT1 1 2.+5 .3 0.10 0.10
$
$ NODES OF THE ENTIRE MODEL
$
GRID 100000 21. 11. 0.
GRID 1 0. 0. 0.
GRID 6 5. 0. 0.
GRID 11 10. 0. 0.
GRID 16 15. 0. 0.
GRID 21 20. 0. 0.
GRID 106 0. 5. 0.
GRID 111 5. 5. 0.
GRID 116 10. 5. 0.
GRID 121 15. 5. 0.
GRID 126 20. 5. 0.
GRID 211 0. 10. 0.
GRID 216 5. 10. 0.
GRID 221 10. 10. 0.
GRID 226 15. 10. 0.
GRID 231 20. 10. 0.
GRID 232 0. 0. 1.
GRID 237 5. 0. 1.
GRID 242 10. 0. 1.

34
GRID 247 15. 0. 1.
GRID 252 20. 0. 1.
GRID 337 0. 5. 1.
GRID 342 5. 5. 1.
GRID 347 10. 5. 1.
GRID 352 15. 5. 1.
GRID 357 20. 5. 1.
GRID 442 0. 10. 1.
GRID 447 5. 10. 1.
GRID 452 10. 10. 1.
GRID 457 15. 10. 1.
GRID 462 20. 10. 1.
GRID 463 0 21. 0. 1.
GRID 468 0 21. 5. 1.
GRID 473 0 21. 10. 1.
$
$ LOADS FOR LOAD CASE
$
$-------2-------3-------4-------5-------6-------7-------8-------9-------0-------
TLOAD1 200 100 0 0 500
TABLED1 500 +TAB1
+TAB1 0. 0. 1. 1. 1.2 0. 10. 0. +TAB2
+TAB2 ENDT
$
FORCE 100 100000 0 100.0 0.0 0.0 -1.0
FORCE 100 463 0 12.0 0.0 0.0 -1.0
FORCE 100 468 0 20.0 0.0 0.0 -1.0
FORCE 100 473 0 12.0 0.0 0.0 -1.0
$
ENDDATA
Example 4
Example EX04 is a frame modeled with CBEAM, RBE2, and RJOINT elements. The large motions of
the frame are provided by the PIN ends of the BEAM. But in the nonlinear analysis, the PIN flag should
not be used. Therefore the RJOINT elements are used to simulate the PIN ends. This is static geometric
analysis with 3D motion.
ID MSC, EX04
SOL 400
CEND
SUBTITLE = BEAM + RBE2 + RJOINT LARGE ROTATION
NLPARM = 10
RIGID=LAGRAN
LOAD = 100
SPC = 10
DISP = ALL
NLPARM,10,10,,,,,,,+NLP1
PARAM,LGDISP,+1
$-------2-------3-------4-------5-------6-------7-------8-------9-------0-------
RBE210 3 1234562
RBE220 6 1234567
RJOINT 23 3 23 1234
RJOINT 67 6 67 1234
$-------2-------3-------4-------5-------6-------7-------8-------9-------0-------
GRID1 0.0.0.
GRID2 5.0 0.00.
GRID3 10.0 0.00.
GRID4 20.0 0.00.
GRID5 20.0 10.00.
GRID6 30.0 0.00.
GRID7 35.0 0.00.
GRID8 40.0 0.00.
GRID23 10.0 0.00.
GRID67 30.0 0.00.
$-------2-------3-------4-------5-------6-------7-------8-------9-------0-------
CBEAM 1 10 1 2 0.0 0.0 1.0
CBEAM 2 10 23 4 0.0 0.0 1.0
$ 156
CBEAM 3 10 4 5 0.0 0.0 1.0
CBEAM 4 10 4 67 0.0 0.0 1.0
$ 56
CBEAM 5 10 7 8 0.0 0.0 1.0
35
CHAPTER 3
$-------2-------3-------4-------5-------6-------7-------8-------9-------0-------
PBEAM 10 1 1.0 0.1 0.1 0.0 0.5
MAT111.+6.3
$-------2-------3-------4-------5-------6-------7-------8-------9-------0-------
SPC1 10 123456 1 8
$-------2-------3-------4-------5-------6-------7-------8-------9-------0-------
FORCE1002 5000.0.0-1.00.0
FORCE1005 5000.0.0 0.0 1.0
FORCE1007 5000.0.0-1.00.0
ENDDATA

36
SOL 400 Material and Elements
Nonlinear Property Extension
Introduction
MD R2 Nastran introduces into SOL 400 extensive enhancements for nonlinear large strain and material
behavior.
Now available in MD Nastran SOL 400 are: Orthotropic material properties for 3-dimensional and plane
strain behavior via the MATORT Bulk Data entry, nonlinear gasket material properties for compression
behavior via the MATG Bulk Data entry, elasto-plastic material properties for use in large deformation
analysis via the MATEP Bulk Data entry, and failure model properties for linear elastic material via the
MATF Bulk Data entry. To take advantage of these material descriptions the new PSHLN1, PSHLN2,
and PSLDN1 Bulk Data entries must be utilized.
The existing composite capabilities in MD Nastran invoked through use of the PCOMP or PCOMG Bulk
Data entries are now enhanced to use the newly added material behavior by associating these entries with
a PSHLN1 Bulk Data entry.
Composites are now also available for plane strain and axisymmetric elements via the PLCOMP Bulk
Data entry and also for layered solid composites via the PCOMPLS Bulk Data entry.
Another enhancement is the introduction of hexahedral, pentahedral, quadrilateral, and axisymmetric
quadrilateral interface elements which may be used to simulate the onset or progress of delamination. The
cohesive material is defined using cohesive energy (also called the critical energy release rate). Mixed
mode delamination is incorporated by converting the normal and shear components of relative
displacement into an equivalent relative displacement using the shear-normal weighting factor. This
capability is invoked via the CIFPENT, CIFHEX, CIFQUAD, and CIFQDX Bulk Data entries and their
associated PCOHE and MCOHE Bulk Data entries.
Another new capability called Virtual Crack Closure Technique via the VCCT Bulk Data entry has been
introduced for evaluating energy release rates. Multiple cracks can be defined and results will be obtained
for each crack separately. Each crack consists of a crack tip grid for shells and a crack front for solids. A
crack is also allowed to grow. This can occur if the crack is in a glued contact interface. You can enter a
crack growth resistance (fracture toughness) for the crack. If the calculated energy release rate is larger
than this value the crack will grow. This is done by automatically releasing the glued contact interface
segment by segment.
Benefits
With these material enhancements, MD Nastran, SOL 400 is in a better position to support modern
manufacturing techniques which require the use of composite materials and the study of nonlinear
material behavior dealing with failure and delamination.
Further, where appropriate, most of the finite elements in MD Nastran are now extended to include the
effects of finite strain.
37
CHAPTER 3
Input
To take advantage of the new large strain, new material, and Virtual Crack Closure Technique
enhancements the following Bulk Data entries are utilized:
1. PSHLN1 Bulk Data entry: This entry extends the large strain and new material capabilities to the
general shells defined by CQUAD4, CQUADR, CQUAD8, CTRIA3, or CTRIAR elements. This
entry MUST have the same property ID as the PSHELL, PCOMP, or PCOMPG associated with
the element. If any GRID of a shell element is listed on the new VCCT Virtual Crack Closure
Technique Bulk Data entry, that shell element MUST have a PSHLN1 entry associated with it.
2. PSHLN2 Bulk Data entry: This entry extends the large strain and new material capabilities to the
2-dimensional solid plane strain, plane stress, or axisymmetric elements defined by the
CQUAD4, CQUAD8, CTRIA6, CQUAD and CQUADX with either four or eight grids, or
CTRIAX with six grids. This entry MUST have the same property ID as the PLPLANE
associated with the element. These element MUST lie in the basic X-Y plane. If any GRID of a
2-dimensional solid element is listed on the new VCCT Virtual Crack Closure Technique Bulk
Data entry, that 2-dimensional solid element MUST have a PSHLN2 entry associated with it.
This augments previous hyperelastic element technology.
3. PSLDN1 Bulk Data entry: This entry extends the large strain and new material capabilities to the
3-dimensional solid elements defined by the CHEXA and CTETRA. This entry MUST have the
same property ID as the PSOLID associated with the element. If any GRID of a 3-dimensional
solid element is listed on the new VCCT Virtual Crack Closure Technique Bulk entry, that 3-
dimensional solid element MUST have a PSLDN1 entry associated with it.
4. PLCOMP Bulk Data entry: This entry extends composites to the 2-dimensional solid plane strain,
plane stress, or axisymmetric elements defined by CQUAD4, CQUAD8, CQUAD and CQUADX
with either four or eight grids.
5. PCOMPLS Bulk Data entry: This entry extends composites to the 3-dimensional solid element
defined by CHEXA. A solid shell formulation is available with this entry.
6. CIFPENT, CIFHEX, CIFQUAD, and CIFQDX Bulk Data entries: These are new MD Nastran
interface elements (currently valid only in SOL 400) used to simulate the onset or progress of
delamination.
7. PCOHE Bulk Data entry: The property interface or the CIFPENT, CIFHEX, CIFQUAD, and
CIFQDX elements.
8. MCOHE Bulk Data entry: This entry specifies material cohesive properties used to simulate the
onset or progress of delamination.
9. NLMOPTS Bulk Data entry: This entry controls parameters associated with PSHLN1, PSHLN2,
PLCOMP, PCOMPLS, and PCOHE.
10. VCCT Case Control command: By specifying VCCT=n, this command selects the VCCT Bulk
Data entry to be used in a given STEP.
11. VCCT Bulk Data entry: This entry specifies the Virtual Crack Closure Technique entry for
evaluating energy release rates.

38
12. PSHEARN Bulk Data entry: This entry extends large membrane rotation to the CSHEAR
element. Stringer effectiveness is ignored. The Bulk Data entry MDLPRM,SHRTOQ4,1
CANNOT be used with this entry.
13. MATORT and MATG Bulk Data entries: These existing primary material entries have been
extended for use with SOL 400. Their associated MATTORT and MATTG entries are also valid
for specifying temperature dependent materials.
14. MATEP and MATF Bulk Data entries: These existing associated material entries have been
extended for use with SOL 400. The associated MATTEP entry is also valid for specifying
temperature dependent materials.
15. MAT3 Bulk Data entry: This existing entry may also be used in conjunction with PSHLN2 and
PLCOMP axisymmetric elements. The associated MATT3 entry is also valid for specifying
temperature dependent materials.
16. CQUAD, CQUADX, and CTRIAX Bulk Data entries: These three existing entries have had a
(THETA/MCID) field added to their description. This new field is only applicable if the
PLPLANE entry has an associated PSHLN2 entry. If the element only has a PLPLANE property,
the field is ignored.
Output
The element output is obtained via standard MD Nastran STRESS=n and NLSTRESS=n commands.
Both linear formatted nonlinear stress and nonlinear stress/strain output is available.
The Virtual Crack Closure Technique output data is automatically placed on file OFVCCT. In this
release, if the VCCT is utilized in the run it is automatically output to the .F06 file.
1. For the shell elements, the 2-dimensional solid elements and 3-dimensional solid elements there
are two types of property entries:
a. The primary property entries are the PSHELL, PCOMP, PCOMPG, PLPLANE, PSOLID,
PLCOMP, PCOMPLS, and PSHEAR.
b. An associated property such as a PSHLN1, PSHLN2, PSLDN1, and PSHEARN.
c. The associated property is matched to the primary property by having the same ID.
LOAD STEP = 1.00000E+00
V C C T C R A C K R E S U L T S
CRACK TIP ------------- ENERGY RELEASE RATE ------------ ESTIMATED CRACK GROWTH DIRECTION
CRACK ID GRID ID TOTAL MODE I MODE II MODE III X Y Z
100 1 4.5493E+01 4.5493E+01 7.6309E-16 0.0000E+00 1.0000E+00 1.3297E-16 0.0000E+00
100 2 4.5484E+01 4.5484E+01 1.4533E-15 0.0000E+00 1.0000E+00 1.0262E-16 0.0000E+00
100 3 4.5484E+01 4.5484E+01 4.3923E-17 0.0000E+00 1.0000E+00 1.6459E-16 0.0000E+00
100 4 4.5493E+01 4.5493E+01 1.6856E-15 0.0000E+00 1.0000E+00 9.2419E-17 0.0000E+00
39
CHAPTER 3
d. The PSHLN1 invokes the enhanced nonlinear capability for shell elements whose PID points
to a PSHELL, PCOMP, or PCOMPG. The PSHLN2 invokes the enhanced nonlinear
capability for 2-dimensional solid elements whose PID points to a PLPLANE. The PSLDN1
invokes the enhanced nonlinear capability for 3-dimensional solid elements whose PID points
to a PSOLID.
2. In MD Nastran there are two types of material entries:
a. A primary material entry whose ID may appear on an appropriate PSHELL, PLPLANE,
PSOLID, PCOMP(G), PSHLN1, PSHLN2, PSLDN1, PLCOMP, PCOMPLS, PSHEAR etc.,
entry.
b. An associated material entry whose ID must appropriately match one of the primary material
entry IDs.
c. The primary material entry MATORT ID may only appear on PSHLN2, PSLDN1, PLCOMP,
and PCOMPLS. If its ID appears on say a PSOLID in the MID field it will be ignored and
the run will fail with no material defined error. The primary material entry MATG ID may
only appear on PSHLN2 and or PSLDN1.
d. If the associated materials MATEP or MATF point to a primary material ID for shell elements
and there is no associated PSHLN1 pointing to a PSHELL, PCOMP, or PCOMPG the
associated material will not be used. If the associated materials MATEP or MATF point to a
primary material ID for 2-dimensional solid elements that have a PLPLANE as their primary
property and there is no associated PSHLN2 pointing to a PLPLANE the associated material
will not be used. If the associated materials MATEP or MATF point to a primary material ID
for 3-dimensional solid elements that have a PSOLID as their primary property and there is
no associated PSLDN1 pointing to a PSOLID the associated material will not be used.
3. The PSHLN1 entry allows the user to change the material ID associated with the MID1 or MID2
or both on the PSHELL. If these entries are left blank on the PSHLN1 then the MID1 and MID2
values on the PSHELL are used. The flow diagram below shows the PSHLN1s relationship to
the shell elements.

40
4. The PSHLN2 entry allows the user to change the material ID associated with the MID on the
PLPLANE. There is no default. The PLPLANE requires a MATHP and the user MUST override
with a MAT1, MAT2, MAT3, MAT8, MATORT, or if appropriate a MATG. The flow diagram
below shows its relationship to the 2-dimensional solid using a PLPLANE entry as its primary
property entry. On the PSHLN2 entry the BEHi codes are sensitive to the required primary
material used. MAT1 is applicable to all BEHi codes. MAT2 anisotropic and MAT8 orthotropic
are applicable to BEHi=PSTRS codes only. MAT3 axisymmetric orthotropic is applicable to
BEHi=AXSOLID code only. MATORT orthotropic are applicable to BEHi=PLSTRN code only.
MATG is applicable for BEH4=COMPS or AXCOMP with INT4=L codes only. The
BEH4=COMPS or AXCOMP with INT4=L should not be used with MAT1, MAT2, MAT3,
MAT8, or MATORT as they will suffer hour-glassing. In SOL 400, if a PLPLANE entry has an
associated PSNLN2 entry, it can directly refer to an appropriate MAT1, MAT2, etc., material
entry and not have a MATHP referral. However, in this case all elements referring to the
PLPLANE entry will fail in all other solution sequences.
PSHLN1
Property ID Material Used
PSHELL PCOMP or PCOMPG MAT1 MAT2, MAT8
MAT1 MAT2, MAT8 MAT1, MAT2, MAT8 MATEP, MATF, MATS1 MATEP, MATF
MATS1
Assoc. Material on PSHELL
Assoc. Material only if PSHLN1 present
Optional Primary Material Points to
41
CHAPTER 3
5. The PSLDN2 entry allows the user to change the material ID associated with the MID on the
PSOLID. If the MID field is left blank on the PSLDN1 then the MID value on the PSLOID is
used. The flow diagram below shows its relationship to the 3-dimensional solid using a PSOLID
entry as its primary property entry.
PSHLN2
PLPLANE MAT1 MAT2, MAT3, MAT8, MATORT
MATEP, MATF, MATS1 MATEP, MATF MATHP
Required entry
Assoc. Material only if PSHLN2 present
Required Primary Material Override Points to
MATG

42
6. The key word field entries on the PSHLN1, PSSHLN2, PSLDN1, PLCOMP, and PCOMPLS
Bulk Data have default integration schemes and needs not be used if the user is willing to use these
defaults.
7. The MATG gasket material requires a special integration scheme. It is only available for elements
using a PSHLN2 or PSLDN1 Bulk Data entry. For the PSHLN2, the C4 keyword entry with
BEH4=COMPS or AXCOMP and INT4=L would be required. For the solid BEH8=SLCOMP,
INT8=L would be required. For example assume an eight noded HEXA element with a PSOLID
of PID=24 and it is desired to use a MATG entry with MID of 13. Then the following would be
required.
8. For composites, a solid shell element formulation is available. The DIRECT field entry must be
DIRECT=1 (the default). For the linear and quadratic formulations, no key word entry is
required. The sample below shows the solid shell element request.
PSLDN1
PSOLID MAT1 MAT9, MATORT
MATEP, MATF, MATS1 MATEP, MATF MAT1
Associated Material
Assoc. Material only if PSLDN1 present
Optional Primary Material Points to
MATG
MAT9
MATS1
PSLDN1 24 13
C8 SLCOMP L
43
CHAPTER 3
9. Because these new material features often involve large deformation, it is recommended that a
full Newton iteration scheme be used. This has been facilitated on the NLPARM Bulk Data entry
by the addition to the KMETHOD field the key word FNT and the TSTEPNL Bulk Data entry
by the addition to the METHOD field the key word FNT.
10. Any shell element that has non structural mass (NSM) that utilizes any PSHELN1 or PSHLN2
entry will lose the associated non structural mass.
Examples
Many examples of new element materials can be found in the MD R2 Users Guide.
PCOMPLS 782 3
C8 SLCOMP ASTN
100 171 .3 12.3
101 175 .7 77.7

44
The Nonlinear QUADR/TRIAR Elements
Introduction
The QUADR/TRIAR elements have been implemented in all solution sequences except the nonlinear
solution sequences. In this release, these elements are implemented in SOL 400 only. If QUADR/TRIAR
is requested in other nonlinear solution sequence (SOL 106 or 129), a user fatal error will be issued.
Benefits
The benefits for the QUADR/TRIAR elements are:.
The drilling dofs are defined for the shell elements so that the parameter K6ROT is not required
in a geometric nonlinear analysis.
With the drilling dofs, the drilling loads can be applied.
The membrane performance of QUADR/TRIAR is superior to that of QUAD4/TRIA3.
For material nonlinear analysis, 4 (QUADR) or 3 (TRIAR) integration points in the plane of the
element are used. This will give better results.
The nonlinear temperature composites are implemented for both the smeared and the non-
smeared methods
Green strains are implemented for QUADR/TRIAR.
The stress output location can be at either corner or integration points.
Limitations
The following nonlinear analysis material models have not been implemented for the QUADR/TRIAR
elements:
Creep,
Hyper-elasticity.
Type of Analysis
The following types of analysis have been implemented for the QUADR/TRIAR element.:
Material nonlinear analysis the material models are elastoplastic and nonlinear elastic. The
method to invoke the material nonlinear analysis is the same as that of QUAD4/TRIA3. See
Bulk Data entry MAT1S.
Geometric nonlinear analysis method to invoke this capability is to use PARAM, LGDISP, 1
Bulk Data entry.
45
CHAPTER 3
Geometric nonlinear analysis for temperature dependent composites both the smear method
and non-smear method are implemented. The method to invoke this capability is to use
PARAM, COMPMATT, value and PARAM, EPSILONT, value Bulk Data entries. Please
refer to Section 4.1, Temperature-Dependent Composites Support Extended to Unsymmetric
Laminates, MSC.Nastran Release Guide 2005 for details. The user must use SOL 400 for this
capability with QUADR/TRIAR, though it was SOL 106 stated in the Release Guide.
The QUADR/TRIAR have been implemented into both nonlinear static analysis and nonlinear
transient analysis.
Drilling Degrees of Freedom
The QUADR/TRIAR elements have rotational stiffness computed for the drilling degree-of-freedom
(DOF). Options are provided to deactivate the drilling DOF. Here, deactivate means that the rotational
stiffness for drilling DOF is not computed. QUADR/TRIAR without drilling DOF become elements
similar to the QUAD4/TRIA3 elements. This option is provided by the Bulk Data entry:
MDLPRM, QR6ROT, n
n = 0 The drilling DOF is active. Default.
1 The drilling DOF is deactivated for all QUADR/TRIAR elements in the model.
2 The drilling DOF is deactivated for those QUADR/TRIAR elements that have
membrane stiffness only (MID2 and MID3 are blank on the PSHELL entry.
If the drilling DOF is deactivated, small drilling DOF stiffness may be supplied by PARAM, K6ROT,
value. For default, value = 0.0 if it is a membrane element and value = 100.0 if it is a shell element. Note
that, if the drill DOF is active, the K6ROT parameter has no effect on the QUADR/TRIAR elements.
The QR6ROT option is available all solution sequences.
Off-plane Shear Formulations
For QUADR/TRIAR, there are two methods to compute the off-plane shear stiffness: the stiffness
method and the flexibility method. The stiffness method is a new method implemented in QUADR. The
flexibility method was the method implemented in the QUAD4 element. Therefore, if the flexibility
method is selected, the solution results of QUADR are closer to those of QUAD4. Option is provided to
section the off-plane shear computation method by the Bulk Data entry:
MDLPRM, QRSHEAR, n
n =0 Use stiffness method if MID3 0 on the PSHELL Bulk Data entry. Use the
flexibility method if MID3 = 0. Default.
1 Use the flexibility method.
2 Use the stiffness method.

46
QRSHEAR option is available for all solution sequences.
Green Strain
In addition to the small strain formulation, the Green strain formulation is also computed for the
QUADR/TRIAR elements. Option is provide to select the strain formulation by the Bulk Data entry:
MDLPRM, GNLSTN, n
n =0 Small strain. Default.
1 Green strain.
The GNLSTN option is available for SOL 400 only.
Stress Output Location
For QUADR/TRIAR elements, the stress/strain/force output location can be either at the corner points or
at the integration points. Option is provided to select the output location by the Bulk Data entry:
MDLPRM, INTOUT, n
n =0 Corner point output. Default.
1 Integration point output.
The INTOUT option is available for all solution sequences.
Differential Stiffness
The differential stiffness is computed for the QUADR/TRIAR elements in the geometric nonlinear
analysis. However, the differential stiffness is not required to obtain the correct results. If the solution
converges, the results are correct. The differential stiffness will facilitate the convergence if the axial
forces are large. This is true for problem that computes the nonlinear buckling load or problem with
tension stiffening effect.
However, there are situations in which the differential stiffness hinders the convergence or even leads to
solution divergence. For this reason, SOL 400 provides an option to turn off the generation of differential
stiffness. The option is given by the Bulk Data entry:
MDLPRM, NLDIFF, n
n = 0 The differential stiffness is computed. Default.
1 Turn off the computation of differential stiffness.
The NLDIFF option is available for SOL 400 only. But it is also available for other elements such as
QUADR/TRIA3.
47
CHAPTER 3
CBAR Element to CBEAM Element Conversion
The CBAR element is a linear element. It can be converted to a geometric nonlinear CBEAM by the Bulk
Data entry:
MDLPRM, BRTOBM, n
n =0 do not convert CBAR to CBEAM. Default.
1 Convert CBAR to CBEAM.
-1 Same as 1, but print the converted Bulk Data entries on f06 file.
Note that, after conversion, all outputs are in CBEAM format.

48
CSHEAR Element to CQUAD4 Element Conversion
The CSHEAR element is a linear element. It can be converted to a geometric nonlinear CQUAD4 with
membrane shear stiffness only. The conversion is requested by the Bulk Data entry:
MDLPRM, SHRTOQ4, n
n =0 do not convert CSHEAR to CQUAD4. Default.
1 Convert CSHEAR to CQUAD4.
2 Same as 1, but print the converted Bulk Data entries on f06 file.
Note that, for irregular shapes, CSHEAR to CQUAD4 conversion is only an approximation. Also, after
conversion, all outputs are in CQUAD4 format.
49
CHAPTER 3
SOL 400 Pretension Modeling Description
Pretension Modeling
Introduction
It may be convenient for the user to pretension a part of his or her structure. In MD Nastran this is done
with a combination of MPC and SPC entries using what is traditionally call in MD Nastran SLACK
variables. In SOL 400 MD Nastran this is automated so that the user can apply in the first load STEP the
pretension load as a physical load and in subsequent steps automatically pick up the associated
displacement the initial load produced to maintain the initial pretension under other general structural
loading conditions.
Benefits
In SOL400 an automatic pretension capability is provided. This is useful say in applying a pretension to
a bolt structure.
Input
The method simply requires standard MPC and SPC input as described in the guideline section below.
Output
Standard MD Nastran output is available.
The procedure in SOL 400 is as follows:
1. No SPC or SPCD with a displacement value Di other than 0.0 is allowed in residual structure
when this method is used.
2. The user identifies a cut between two surfaces where it is desired to pretension the structure.
3. The element grids that lie above the cut and lie on the cut plane will be identified as and should
be in the independent set of MPC equations.
4. The element grids that are connected to the elements that lie below the cut and lie on the cut plane
will be identified as and should be in the dependent set of the MPC equations.
5. Associate with the cut a single control grid, identified as , that will be in the independent set of
MPC equations.
6. In the first STEP apply a pretension load to the control or slack grid in the appropriate direction
to apply the desired pretension load. While either SPOINT or GRID can be used to define a
slack variable, in SOL 400, a GRID is recommended.
7. In the first STEP and all subsequent STEPs write a series of MPC equations for the cut plane of
the form:
u
t
u
b
u
s

50
One such equation for each top and bottom grid along the cut plane.
8. In the first STEP also apply any SPCs necessary to constrain the overall structure against rigid
body motion. The Di value on any SPC or SPCD must be 0.0
9. In the second and any subsequent STEPs, apply the cut plane MPCs and additionally add to the
existing SPC applied in the first STEP an SPC for each control grid in the direction of the load
application with a Di value of 0.0
10. MD Nastran will automatically apply incrementally the displacement of the control grid in the
direction of the pretension applied in the first STEP.
Examples
The figure shows two blocks of solid elements. The lower block is a 10x2x2 mesh. The upper block is
also a 10x2x2 mesh. The three Spider arrangements of MPCs shown in the Figure 3-1 will apply a
pretension of 100 load units in each Spider. At the center of each Spider a control grid is introduced.
The pretension load for each Spider will be applied at these control grids. These control grids were
placed at the center of each Spider for convenience only. They could have been placed at any
reasonable location. The partial input file below shows important modeling features for pretensioning.
Figure 3-1 Two Block Sections with Pretension Applied by MPC at each Spiders
In STEP 10 a Pretension load is applied at each Spider. In STEP 20, the structure has a thermal load
applied and the Pretension displacement at each control grid is carried forward from STEP 10 to STEP
20 and any subsequent STEPs.
u
b
u
t
u
s
0 Z
51
CHAPTER 3
SOL 400
CEND
TEMP(INIT)=3001
DISP = ALL
MPCF = ALL
SPCF = ALL
NLPARM=1
ANALYSIS=NLSTAT
MPC = 22 <----------- MPCs carried through each STEP
STEP 10
SPC = 1
LOAD = 100
STEP 20
SPC = 3
TEMP(LOAD)=6002
BEGIN BULK
$
NLPARM,1,2,,ITER,1,10,,NO,+NLPRM1
+NLPRM1,,,,,0
$
TEMPD,3001,0.
TEMPD,6002,100.
$
GRDSET,,,,,,,456
$
$ Lower 10x2x2 mesh grids
$
GRID 1 0 0.0 0.0 0.0 0
.
.
.
GRID 99 0 10.000 2.000 2.000 0
$
$ Upper 10x2x2 mesh grids
$
GRID 1067 0 0.0 0.0 2.000 0
.
.
.
GRID 165 0 10.000 2.000 4.000 0
$
$ Properties and material for all CHEXA elements
$
PSOLID 1 1 0 0 0 0
$
MAT1* 1 2.0680E+05 0.28999999166+MA 1
*MA 1 1.00000000 1.169999996E-05 +MB 1
*MB 1 1500000.00 1500000.00 68000.00 +MC 1
*MC 1
$
MATS1 1 PLASTIC 1.+7 1 1 3.+4
$
$ Lower 10x2x2 mesh elements
$
CHEXA 1 1 1 2 13 12 34 35
46 45
.
.
.

52
CHEXA 40 1 54 55 66 65 87 88
99 98
$
$ Upper 10x2x2 mesh elements
$
CHEXA 41 1 1067 1068 1079 1078 100 101
112 111
.
.
.
CHEXA 80 1 120 121 132 131 153 154
165 164
$
$ Clamp lower section of block
$
SPC 1 1 123 0.0
.
.
.
SPC 1 89 123 0.0
$
$ Simply support upper section of block
$
SPC 1 133 123 0.0
SPC 1 143 3 0.0
SPC 1 155 13 0.0
$
$ In STEP 20 apply supports plus SLACK variables to impose Pretension
$
SPCADD312
$
$ u_bottom - u_top - u_control = 0
$
$ Pretension input required for center Spider of figure
$
$ BOLT 2
$
$ Control grid
$
GRID 2083 0 5.000 1.0 2.000 0
$
$ Preload on 2083 applied in STEP 10
$
FORCE1002083-100.0.0.1.
$
$ Displacement overlapp - often call a SLACK variable in MD Nastran
$ Disp applied in STEP 20
$ -1.598639E-02 will be picked up from STEP 10
for this SPC
$
SPC2208330.0
$
$ MPCs for Pretension on center Spider of figure Spider
$
MPC 22 1071 1 1. 71 1 -1.
2083 1 -1.
.
.
.
MPC 22 1095 1 1. 95 1 -1.
53
CHAPTER 3
2083 1 -1.
$
MPC 22 1071 2 1. 71 2 -1.
2083 2 -1.
.
.
.
MPC 22 1095 2 1. 95 2 -1.
2083 2 -1.
$
MPC 22 1071 3 1. 71 3 -1.
2083 3 -1.
.
.
.
MPC 22 1095 3 1. 95 3 -1.
2083 3 -1.
$
ENDDATA

54
VCCT Virtual Crack Closure Method
The VCCT option provides a method to calculate the energy release rate for sharp cracks. The energy
release rate is equivalent to the so-called J-integral for the linear elastic case. The minimum required user
input is the crack tip in a 2D analysis or for a shell with a line crack and the grids defining the crack front
for the 3D case. The capability supports line cracks in 2D solid elements and for shell elements and
surface cracks in 3D and for stacked shells. The VCCT results are available on the OFVCCT file. The
printed output to .F06 consists of the energy release rate and an estimation of the anticipated crack growth
direction. The crack growth direction is calculated using the maximum hoop stress criterion. In addition,
the table contains three vectors defining the current crack tip system. This system is updated with the
deformations and follows the crack tip. The vectors are available for post processing.
If the crack is placed in an interface between two glued contact bodies, it can propagate along this glued
interface. This is illustrated in Figure 3-3 for a simple 2D case. The user specifies the grid ID of the initial
crack tip and a crack growth resistance (fracture toughness or critical energy release rate). When, during
the analysis, the calculated energy release rate is larger than the limit the crack will grow. The program
automatically finds how much of the glued interface that should be released for each load level. If
multiple choices of grid release are available, the program will choose the one that is closest to the
estimated crack growth direction.
Figure 3-2 Design of model for crack propagation.
The 3D case is similar, but the crack is defined as a crack front of multiple grids and the glued interface
is a surface. Each point along the crack front can grow independently.
The crack growth resistance value can be given a variation with the distance the crack has grown. This
is done by associating it with a TABLEM1 entry in the VCCT bulk data entry. Thus it is possible to model
a so-called r-curve behavior, where the resistance to further growth changes (usually increases) as the
crack grows. For 3D this is traced for each crack front point separately.
glued interface
initial crack tip
first contact body
second contact body
55
CHAPTER 3
`=w=j
The new library of so-called interface elements can be used to simulate the onset and progress of
delamination. The constitutive behavior of these elements is expressed in terms of tractions versus
relative displacements between the top and bottom edge/surface of the elements.
Considering a 3-D interface element, the relative displacement components are given by one normal and
two shear components, expressed with respect to the local element system (see Volume B for the
definition of the local element systems):
(3-1)
Based on the relative displacement components, the effective opening displacement is defined as:
(3-2)
in which is called the shear-normal weighting coefficient, which equals the ratio between the critical
shear and normal tractions. Equation 3-2 is strictly valid under the assumption that the interface element
is loaded in tension ( ). At the end of this section the modification for compressive loading will be
discussed.
The effective traction is introduced as a function of the effective opening displacement and is
characterized by an initial reversible response followed by an irreversible response as soon as a critical
effective opening displacement has been reached. Three standard functions are currently available,
namely a bilinear, an exponential and a combined linear-exponential function (see Figure 3-3):
if
if (bilinear) (3-3)
if
(exponential) (3-4)
if
v
n
u
1
t op
u
1
bot t om
Z
v
s
u
2
t op
u
2
bot t om
Z
v
t
u
3
t op
u
3
bot t om
Z
v v
n
2
2
v
s
2
2
v
t
2
H H Z
v
n
0
t
v
c
t
2G
c
v
m
JJJJJJJJJJ
v
v
c
JJJJJ Z 0 v v
c

t
2G
c
v
m
JJJJJJJJJJ
v
m
v
v
m
v
c
JJJJJJJJJJJJJJJJJJ

Z v
c
v < v
m
t 0 Z v v
m
>
t G
c
v
v
c
2
JJJJJ e
v v
c
Z
t
2qG
c
v
c
q 2 H ( )
JJJJJJJJJJJJJJJJJJJJJJJ
v
v
c
JJJJJ Z 0 v v
c


56
(linear-exponential) (3-5)
if
in which is the critical energy release rate (also called the cohesive energy), is the maximum
effective opening displacement (which is only used by the bilinear model) and is the exponential decay
factor (which is only used by the linear-exponential model).
Figure 3-3 Bilinear (left), Exponential (middle) and Linear-exponential (right) Cohesive
Material Model
It can be easily verified that the maximum effective traction , corresponding to the critical effective
opening displacement , is given by:
(bilinear) (3-6)
(exponential) (3-7)
So, if the maximum effective traction is known, the critical energy release rate can be determined by:
(linear) (3-9)
(exponential) (3-10)
t
2qG
c
v
c
q 2 H ( )
JJJJJJJJJJJJJJJJJJJJJJJ e
q 1 v v
c
( )
Z v v
c
>
G
c
v
m
q
v v
t t
v
c
v
c
v
m
v
t
v
c
t
c
v
c
t
c
2G
c
v
m
JJJJJJJJJJ Z
t
c
G
c
ev
c
JJJJJJJJ Z
t
c
2qG
c
v
c
q 2 H ( )
JJJJJJJJJJJJJJJJJJJJJJJ Z
G
c
t
c
v
m
2
JJJJJJJJJJ Z
G
c
et
c
v
c
Z
57
CHAPTER 3
In order to avoid convergence problems in a finite element simulation of delamination, one may activate
so-called viscous energy dissipation. The basic idea of the dissipation model is that when delamination
starts, the rate of deformation may suddenly increase. This increase is used to augment the constitutive
behavior with a viscous contribution being equivalent to this rate of deformation:
(3-12)
in which is the viscous energy factor, is effective opening displacement rate and is the reference
value of the effective opening displacement rate. This reference value can either be user defined or
calculated by the program. In the latter case, the reference value is given by the maximum effective
opening displacement rate in any interface element, as long as the response in all the interface elements
is reversible. The viscous energy dissipation model does not directly have a physical background, but is
basically numerical in nature.
As has been mentioned before, the assumption has been that the interface element is loaded in tension.
Assuming that in compression the behavior will remain reversible, equation 3-2 will be adapted as:
So far, the constitutive behavior has been discussed in terms of an effective traction versus an effective
opening displacement. The traction components follow from the effective traction according to:
; (3-13)
G
c
t
c
v
c
q 2 H ( )
2q
JJJJJJJJJJJJJJJJJJJJJJJJJJJJ Z
t
vi s
t
c
v
0
JJJJJJJJJJ Z
v
0
v max v
n
0 , ( ) [ ]
2
2
v
s
2
2
v
t
2
H H Z
t
n
t
v
v
n
JJJJJJJJ Z t
s t ,
t
v
v
s t ,
JJJJJJJJJJJ Z

58
Nonlinear Iteration Algorithms
In this section, we are going to discuss the nonlinear iteration algorithm used in SOL 400 to obtain an
equilibrium solution.
The equilibrium solution is achieved by a combination of the following methods:
Time step size adjustment for transient analysis at the beginning of a SUBCASE or at
convergence of a time step, the time step size is adjusted based on the estimated current nature
frequency of the structural model or the input loads.
Load iteration at each time step, quasi Newton method and line search technique are employed
repeatedly to obtain the balance of internal forces and external loads. This is called the load
iteration, or simply the iteration. The time step or a load increment is converged when the
balance of internal forces and external loads is obtained. The user can deselect the quasi Newton
method (MAXQN) or the line search (MAXLS) by using parameters on the
TSTEPNL/NLPARM bulk data entry
Stiffness update in many situations, the load iteration will not be able to achieve equilibrium at
a time step. In this situation, the stiffness matrix can be recomputed using the current geometric
and material state of the structure model to facilitate convergence. This is called the stiffness
update. Consecutive stiffness updates without load iteration is called the full Newton method.
Bisection when the program senses that it is impossible for the solution to converge, a
procedure called divergence processing is employed. One technique used in this procedure is
bisection. Bisection means cutting the time step size or load increment size by half. Divergence
processing uses a combination of stiffness updates and bisections to facilitate convergence
Externally, bisection and time step adjustment are very similar. However, they are initiated due to
different reasons. Time step adjustment is based on the requirement of the natural frequency of the
structure, which may or may not be nonlinear. On the other hand, bisection is due to large nonlinearity in
the solution. Another difference is that the time step adjustment is performed at the beginning of a
SUBCASE or at a converged time step. Bisection is performed during a time step when the program
determines that the solution is diverging.
When the program determines the large nonlinearity is gone and the solution is stabilized toward
convergence, reversal of the bisection is performed in order to maintain the time step size required by
the time step adjustment method given above.
The nonlinear iteration algorithm is controlled by NLPARM/TSTEPNL entries. These entries are given
here:
1 2 3 4 5 6 7 8 9 10
NLPAR ID NINC DT KMET KSTEP MAXI CONV INTOU
EPSU EPSP EPSW MAXD MAXQ MAXL FSTRE LSTOL
MAXBIS MAXR RTOL MINIT
59
CHAPTER 3
The KMETHOD selects the nonlinear iteration method. There are six options available:
ADAPT the convergence of a time step is obtained chiefly by the time step adjustment and the
load iteration. No stiffness update is performed during the normal iteration. Stiffness updates are
performed only for divergent processing and at solution convergence of a time step with number
of iterations greater than 3*MAXITR. MAXITR is the maximum number of iterations given on
the TSTEPNL bulk entry. Even for divergence processing, the main method to correct the
divergence is bisection. This method is available for the transient analysis only.
AUTO the convergence of a load increment or a time step is achieved by automatically
selecting the load iterations and the stiffness updates, in combination with the ADAPT method.
The ADAPT method can be deselected by setting ADJUST=0 on the TSTEPNL bulk data entry.
This is the default method. For divergence processing, the divergence is corrected by a
combination of stiffness updates and bisections. The AUTO method always tries to maintain the
time step size required by the time step adjustment method.
ITER the convergence of a load increment or a time step is achieved by performing a stiffness
update at every KSTEP load iterations, in combination with the ADAPT method. Again, the
ADAPT method can be deselected by setting ADJUST=0. The divergence processing is the
same as that of the AUTO method.
SEMI same as the AUTO method, except that a stiffness update is performed at first iteration
of a new time step.
FNT the Full Newton method is used to obtain the convergence. For this method, the stiffness
is updated at every iteration. In comparison with PFNT method, the defaults for FNT are
EPSU=0.01, EPSW=0.01 and MAXLS=4.
PFNT the Pure Full Newton method is used to obtain the solution convergence. The PFTN is
the same as the FNT method except that the defaults for PFNT are EPSU=-0.01, EPSW=-0.01
and MAXLS=0. This means that PFNT is more stringent than the FNT and the line search is not
selected by default for PFNT.
In performing the convergence tests, three error factors are computed: the displacement error (U), the
load error (P), and the work or energy error (W), which are printed in the Nonlinear Iteration Summary
Table. These three error factors must satisfy the error tolerance rules specified by CONV, EPSU, EPSP,
and EPSW fields.
The CONV field selects convergence criteria to be used. It can select U, P, and/or W. It is highly
recommended that CONV=UPW, which is the default, should be used. If U is selected together with P
or W, the U is not checked at the first iteration of a load increment or time step.
1 2 3 4 5 6 7 8 9 10
TSTEPN ID NDT DT NO KMET KSTEP MAXI CONV
EPSU EPSP EPSW MAXD MAXQ MAXL FSTRE
MAXBIS ADJUST MSTEP RB MAXR UTOL RTOL MINIT

60
For U and W, there are two ways to normalize the errors: the errors are computed respect to the total
quantities or to the incremental quantities of a load increment. In older nonlinear solution sequences
(SOL106 and 129), the errors are computed respect to the total quantities. In SOL400, we computed both
ways, which could be selected by the user. If EPSU>0, the U error is computed respect to the total
displacements. If EPSU<0, the U error is computed respect to the delta displacements of a load
increment. Similarly, if EPSW<0, the W error is computed respect to the total energy. If EPSW<0, the W
error is computed to the delta energy of a load increment. If CONV=UPW, both methods give similar
results for most problems.
Another major modification in comparison to SOL 106 or 129 is the computation of the work error. In
SOL 106 or 129, the work error is computed based on the multiplication of the residual force and the
displacement change. During iteration, both the residual force and the displacement change become
smaller; therefore, the convergence rate of this value is proportional to the square of the convergence rate
of the solution. Thus it becomes very small near convergence. Also, it does not have a counter part in the
physical world. In SOL 400, the total work done to the structure model is computed at each iteration and
the work error is estimated based on the total work. In this way, the work error gives an estimation of the
error in the actual work done to the structural model. The total work for each iteration is printed on the
Nonlinear Iteration Summary Table. Please note that this total work is only an approximation. Users who
specify the work error criterion and are switching from SOL 106/129 to SOL 400 may need to change
the value
The MAXITER and MAXDIV fields control how the solution will be ended if it is divergent. If
MAXDIV>0 (default), SOL 400 will try to compute the best solutions for 4 consecutive un-convergent
load increments and then terminate the run with a fatal error message. If MAXDIV<0, the run will
terminated right away. However, in many cases, the user may want to compute the solution to the end of
a step even if the solution is divergent. In this case, he can set MAXITER<0 and the solution will continue
to the end of a step. The default value of MAXBIS is zero if MAXITER<0. The reason for this is that we
want the solution to end quickly without bisections.
MINITER is new field for SOL400 only. It specifies the minimum iterations to be performed for each
load increment. The default is 1 for all cases except the contact analysis, for which the default is 2.
AUTO and ADAPT are two major methods for the transient analysis in SOL400. From testing problems,
it is observed that the AUTO method tries to maintain the time step size required by the time adjustment
method or the user time step. The ADAPT method usually converges with a time step size smaller than
the time step required by the time step adjustment method or the user time step. For most problems,
AUTO method gives better solutions; therefore, it is selected as the default method. However, if the time
step estimated by the time step adjustment method or the user time step is too large, the ADAPT method
gives a better solution. A better solution here means that the solution converges faster or does not diverge.
For highly nonlinear problems, FNT and PFNT are the recommended methods. For the contact analysis
and new enhanced large strain or new material model (elements with PSHNLi, PSLDN1, or PSHEARN
bulk data entry), it is recommended to use FNT or PFNT. PFNT uses negative EPSU and EPSW as
defaults; therefore, PFNT is more stringent than FNT. However, if CONV=UPW, both methods give
similar results for most problems.
For the transient analysis, the DT and NO fields on the TSTEPNL bulk data entry need more explanation
here. The DT field defines the user time step size. NO (which may be positive or negative) defines that
61
CHAPTER 3
at every |NO| time steps the results are output. Since both time step size adjustment and bisection may
modify the time step size, options are given to the user to select whether to output at user time step size
DT or at internally computed time step size by the following:
If NO>0, the output will be at the user time step size or multiples of the user time step size. Also,
the time step size computed by the time step adjustment will never be greater then the user time
step size DT. For default NO=1.
If NO<0, the output will be at the internally computed time step size, which may or may not be
at the user time point determined by DT. Also, the time step size computed by the time step
adjustment can be either greater or less than the use time step size DT.

62
Analysis Chaining SUBCASE, STEP, ANALYSIS, and
NLIC
The combination of SUBCASE, STEP, ANALYSIS and NLIC commands provide a mechanism for
defining the multiple load steps, running multiple independent cases, and specifying multiple (and
mixed) types of analyses in one job.
SUBCASE and STEP define load cases for a job. SUBCASE defines multiple load cases, which are
independent from each other, i.e., the load history is not passed from one SUBCASE to next. In a
SUBCASE, a number of STEPs can be defined. The solution of one STEP is a continuation of the
solution of its previous STEP.
The user can specify the type of analysis for each SUBCASE and/or STEP by using the Case Control
command ANALYSIS. ANALYSIS supports the following two keywords for SOL 400. They are:
NLSTATIC Nonlinear static analysis. Default.
NLTRAN Nonlinear transient analysis.
NLIC selects the initial condition from a previous static analysis step for the current nonlinear transient
analysis step; see section Initial Conditions below for detail.
The following examples illustrate the manner in which the SUBCASE, STEP and ANALYSIS
commands are used.
With one SUBCASE and multiple steps, each step defines the total external load and other
characteristics for the step, which will be applied by the completion of the step. The solution of
any STEP is a continuation of the solution of its previous STEP. The following is a typical
example:
SUBCASE 1 $ This line can be omitted
ANALYSIS = NLTRAN
TSTEPNL = 200
STEP 10
DLOAD = 10
STEP 20
DLOAD = 20
STEP 30
DLOAD = 30
In the above example, the output time will be continues from step 10 to step 20 and step 30.
Multiple SUBCASEs may be executed in one job where the types of analysis, loads and
boundary conditions can be changed. All SUBCASEs are independent from each other, i.e., no
load history information is transmitted from one SUBCASE to the next. At the start of each
SUBCASE, the displacements, stresses and strains throughout the model are zero if there is no
initial condition specified. For example:
SUBCASE 1
ANALYSIS = NLSTAT $ This line can be omitted
NLPARM = 100
63
CHAPTER 3
STEP 110
LOAD = 110
STEP 120
LOAD = 120
SUBCASE 2
ANALYSIS = NLTRAN
TSTEPNL = 200
STEP 210
DLOAD = 210
STEP 220
DLOAD = 220
In above example, the solutions of SUBCASE 1 and SUBCASE 2 are independent of each other.
In case that the solution divergence is detected in a step, SOL 400 will terminate the solution of
the current subcase and jump to the next subcase.
A case control command placed below the step level allows that command to vary from on step
to another. If it is placed above the step level, the command becomes the default for all steps in
the subcase. Most of the case control commands, which can be placed below the subcase level,
can also placed below the step level. For example, all steps in above examples use the same Case
Control command NLPARM = 100 in SUBCASE 1 and TSTEPNL = 200 in SUBCASE 2.
In the current release, NLSTATIC and NLTRAN analyses can be mixed in a single SUBCASE.
For example:
SUBCASE 10
STEP 1
ANALYSIS = NLSTAT
LOAD = 10
NLAPRM = 110
STEP 2
ANALYSIS = NLSTAT
LOAD = 20
NLPARM = 120
STEP 3
ANALYSIS = NLTRAN
DLOAD = 30
TSTEPNL = 130
In above example, SUBCASE 10 has three steps: the first two steps request nonlinear static analyses and
the third step requests a nonlinear transient analysis. Since there is no NLIC request in the 3
rd
STEP, the
final result of the 2
nd
STEP becomes the initial condition of the 3
rd
STEP automatically. Note that in one
SUBCASE, all STEPs of NLSTAT must be prior to the STEPs of NLTRAN.
The meaning of multiple SUBCASEs without STEP is dependent on the system cell
NASTRAN SYSTEM (366) as following:
0 - The solutions of all SUBCASE are independent of each other. This is consistent to the new
SOL400 procedure. SOL400 will keep all SUBCASE commands in the Case Control file and
insert internally a STEP 1 for each SUBCASE.

64
1 - The solution of each SUBCASE is a continuation of the previous SUBCASE. This is similar
to the solution sequence SOL106 or 129 procedure. SOL400 will convert internally all the
SUBCASE identification numbers to STEP identification numbers and insert a SUBCASE 1
before the first STEP.
The default is 0.
Both SUBCASE and STEP are used in Examples 1, 2, 3, 4, 5, 7, and 8. SUBCASE only is used in
Example 9 on page 101. No SUBCASE and no STEP are used in Example 6 on page 97Example 6.
65
CHAPTER 3
Adaptive Time Stepping Scheme
AUTO Step
An adaptive time stepping scheme has been introduced in Sol400 using the NLAUTO method to modify
automatically the load increment size or the time step size. The primary control of the load stepping is
based upon the number of recycles needed to obtain convergence if full Newton method (FNT or PFNT)
is used. Recycle means either a stiffness update or a load iteration. For modified Newton methods
(ADAPT, AUTO, ITER, or SEMI), both the number of recycles and the number of new stiffness
formations are taken into account.
The adaptive time stepping is invoked if a NLAUTO Bulk Data entry has the same ID as that of a
NLPARM or TSTEPNL Bulk Data entry.
Recycling Criterion
The default recycle based criterion works as follows: The user specifies a desired number of recycles.
For most problems, it is sufficient to provide a value in the range of three to five. For problems with
severe nonlinearities or for problems with very small convergence tolerances, it may be necessary to
increase this number. This number is used as a target value for the load stepping scheme. If the number
of recycles required in the current increment is less than the desired number, the load step for the next
increment is increased. The time step increase is based on a factor, , that can also be specified by the
user. Typical values for are in the range of 1.2 to 1.5. While the time step increase is obviously more
aggressive with larger scale factors, it should be noted that there may be excessive recycling and
cutbacks if sudden nonlinearities are encountered. In order to avoid this, the following logic is used for
higher scale factors: If the actual number of recycles in an increment is greater than 60% of the desired
number of recycles (i.e., the current increment did not converge easily), the increase scale factor for the
next increment is limited to 1.25 for scale factor values between 1.25 and 1.5625, and to 80% of the value
for scale factors above 1.5625.
Time Step Cutback Scheme
The load step is never increased during an increment. If the number of recycles needed to obtain
convergence exceeds the desired number, the load step size is scaled back, the recycling cutback number
is incremented by 1 and the increment is performed again with the new load step. The scaleback
factor for the th cutback is taken as , where the factor is calculated from the expression
; where is the maximum number of recycling related cutbacks for
the increment and is calculated from , is the time increment before any
recycling related cutbacks occur for the increment and is the minimum possible time step for the
S
u
S
u
N
r
N
r s
Nr
s
s T
s
T
m
[ ]
2 N
r m
N
r m
1 + ( ) ( )
= N
r m
N
r m
log
10
10
5
*
T
s
T
m
( ) = T
s
T
m

66
increment. is equal to the value set by the user ( by default) if there is no quasi-static inertial
damping and is equal to times the value set by the user ( by default) if there is quasi-static
inertial damping. The scaleback factor for any cutback is the smaller of ( , ). This scheme
guarantees that no matter what the starting time step for an increment, the minimum time step is reached
in a reasonable number of cutbacks if the increment consistently fails to converge.
In addition, the averaged convergence ratios of previous few iterations are stored and compared against
the current iterations residual. In case of divergence the time step is cut down.
Exceptions
There are some exceptions to the basic scheme outlined above. If an increment is consistently converging
with the current load step and the number of recycles exceeds the desired number, the number of recycles
is allowed to go beyond the desired number until convergence is achieved or up to the user specified
maximum number. The time step is then decreased for the next increment by . An increment is
determined to be converging if the convergence ratio was decreasing in three previous recycles.
Special rules also apply in a contact analysis. During the recycles, the contact status can keep changing
(new nodes come in contact, nodes slide to new segments, separate etc.). Whenever the contact status
changes during an increment, a new set of contact constraints are incorporated into the equilibrium
equations and more recycles are necessary in order to find equilibrium. These extra recycles, due to
contact changes, are not counted when the recycle number is checked against the desired number for
determining if the load step needs to be decreased within the increment. Thus, only true Newton-Raphson
iterations are taken into account. For the load step of the next increment, the accumulated number of
recycles during the previous increment is used. This ensures that the time step is not increased when there
are many changes in contact during the previous increment.
Post Files Output
In many analyses it is convenient to obtain post file results at specified time intervals. This is naturally
obtained with a fixed load stepping scheme but not with an automatic scheme. Traditionally, the post
output frequency is given as every nth increment. Using the NLPARM option, you can request post
output to be obtained at equally spaced time intervals. In this case, the time step is temporarily modified
to exactly reach the time for output. The time step is then restored in the following increment.
Defaults
The defaults of the NLAUTO option are carefully chosen to be adequate in a wide variety of applications.
There are cases, however, when the settings may need to be modified. Assume that the default settings
are used, which means that the recycle based control is active with an initial load of one per cent of the
total. If the structure is weakly nonlinear, convergence is obtained in just a few recycles and the time steps
for successive increments get progressively larger. This can lead to problems if the initially weakly
T
m 10
5
10
3
10
8
s
Nr
1 S
u
1 S
u
67
CHAPTER 3
nonlinear structure suddenly exhibits stronger nonlinearities; for instance, occurrence of plasticity or
parts coming into contact. Possible remedies to this problem include:
decrease the time step scale factor to a smaller number so the step size does not grow so rapidly;
use the maximum time step to limit large steps;
decrease the desired and maximum number of recycles to decrease the load step if more recycles
are needed.
Another situation is if the structure is highly nonlinear and convergence is slow. In this case, it may be
necessary to increase the desired number and maximum number of recycles. In general, there is a close
connection between the convergence tolerances used and the desired number and maximum number of
recycles.

68
The Numerical Integration Method for the Transient
analysis the HHT method
To solve the equation of motion for a nonlinear transient analysis, two aspects must be resolved. The first
is how to integrate the equation of motion for the transient analysis and second is how to obtain an
equilibrium solution for the nonlinear analysis. This section discusses the numerical integration of the
equation of motion and the section Nonlinear Iteration Algorithms discusses the nonlinear iteration
methods for obtaining the balanced solution.
For numerical integration, the 2
nd
order HHT (Hilbert-Hughes-Taylor) method is used in the nonlinear
transient analysis in SOL 400. It provides a user definable parameter , which defines a numerical
damping associated with higher frequency modes, but also maintains the accuracy in the essential lower
frequency modes. Since it is a key feature of HHT method, it is sometimes called the HHT- method.
When <0, the numerical damping is introduced into the system. This leads to an unconditionally stable
integration time scheme when it is in the following range
When comparing with other numerical damping method even at (the maximum numerical
damping value), HHT method still introduces less damping. Therefore, the solution is more accurate
theoretically. Note that when =0, this method is equivalent to the central finite difference method. The
numerical damping, , can be specified using parameter NDAMP (default is -0.05 in SOL 400).
Parameter NDAMP is used in Example 1 on page 88 and Example 7 on page 98.
0
3
1

3
1
=
69
CHAPTER 3
Nonlinear Iteration Summary Table for Nonlinear
Transient Analysis in SOL 400
In order to allow the user to track the solution sequence during the nonlinear iteration, a detailed
Nonlinear Iteration Summary Table is output. A line for each iteration is output on the F06 file. Printing
of the average and the maximum displacements allows the user to know the solution status before the
end of the job. This is useful for large nonlinear problems. Even for small problems, the user will be able
to know approximately how the analysis of a structural model is performing by examining this table. An
example of this table the transient analysis is given below and the descriptions of information given in
this table are shown in Table1.
The summary tables are printed in real time, i.e., each line is printed in the f06 file when the analysis for
the line is completed. Thus, for very large problems, the user can read the fo6 to monitor the progress of
the solution. Also, each linear of the summary table begins with a %, which may be used to locate this
table.
0 N O N - L I N E A R I T E R A T I O N M O D U L E O U T P U T
STIFFNESS UPDATE TIME 0.02 SECONDS SUBCASE 100 STEP 10
- TIME STEP - - - ERROR FACTORS - - CONV ITR MAT AVG TOTL
TIME NO. BIS ADJUST ITR DISP LOAD WORK RATE DIV DIV R_FORCE WORK
1.00000E-01 10 0 1.0000 6 6.09E-05 4.59E-03 1.11E-04 0.36 0 1 2.4E-02 1.487E+00
1.10000E-01 11 0 1.0000 8 3.55E-05 8.29E-03 7.79E-05 0.33 0 1 4.9E-02 2.456E+00
1.20000E-01 12 0 1.0000 11 1.18E-05 4.51E-03 1.14E-05 0.40 0 1 3.1E-02 4.000E+00
1.30000E-01 13 0 1.0000 11 4.16E-04 1.07E-01 1.82E-04 0.38 0 1 6.9E-01 7.460E+00
- - - - - DISP - - - - - - NO. TOT TOT
AVG MAX AT GRID C QNV KUD ITR
7.60E-02 5.538E-01 10701 2 5 1 46
1.00E-01 7.398E-01 10701 2 12 1 54
1.29E-01 9.615E-01 10701 2 19 1 65
1.63E-01 1.221E+00 10701 2 16 1 76
Table 1. Information In Nonlinear Iteration Summary Table
TIME The Current Time. Starts from 0.0 at the beginning of the 1
st
STEP and
accumulate the value until at the end of the last STEP.
For each STEP, the total time is determined by NDT and DT on TSTEPNL Bulk
Data entry.
TIME STEP NO Number of time step, including bisection. Initialized to 0 in the beginning of each
STEP.
TIME STEP BIS Number of bisections performed.
TIME STEP ADJUST The ratio of the current time increment to the original DT on TSTEPNL Bulk Data
entry.
ITR Number of iteration at each time increment.
ERROR FACTORS:
DISP
LOADWORK
There are three error factors: displacement, load, and works. In order for an
increment to converge, these factors must satisfy the error tolerance rules
specified by CONV, EPSU, EPSP, and EPSP on the TSTEPNL Bulk Data entry.

70
NON - LINEAR ITERATION MODULE OUTPUT
For a large problem, TIME STEP NO, TOT KUD, and TOT ITR in this table may be too large to be
printed in the allocated fields, resulting in print overflow. If any of these values overflows, an additional
line is printed to show the offsets of these values. In the table below, TOT ITR of the first line is 15,504.
The offset of TOT ITR is shown as TOT ITR= 10000+XXX, where XXX is the number shown under the
TOT ITR column.
0 N O N - L I N E A R I T E R A T I O N M O D U L E O U T P U T
TIME STEP NO.= 0+XXX TOT KUD= 0+XXX TOT ITR= 10000+XXX
- TIME STEP - - - ERROR FACTORS - - CONV ITR MAT AVG TOTL
TIME NO. BIS ADJUST ITR DISP LOAD WORK RATE DIV DIV R_FORCE WORK
1.99906E+00 2374 0 0.0313 5 1.48E-06 3.21E-03 7.01E-07 0.21 0 1 4.3E-01 5.917E+05
1.99937E+00 2375 0 0.0313 5 1.37E-06 4.75E-03 7.24E-07 0.07 0 1 5.3E-01 5.934E+05
1.99969E+00 2376 0 0.0313 7 6.14E-07 8.13E-04 1.48E-07 0.13 0 1 9.5E-02 5.935E+05
2.00000E+00 2377 0 0.0313 5 1.40E-06 6.48E-03 2.09E-06 0.09 0 1 7.0E-01 5.918E+05
- - - - - DISP - - - - NO. TOT TOT
AVG MAX AT GRID C QNV KUD ITR
2.44E+00 2.022E+01 11100 2 12 614 5504
2.44E+00 2.020E+01 11100 2 13 614 5509
2.44E+00 2.023E+01 11200 2 16 614 5516
2.44E+00 2.026E+01 11200 2 11 614 5521
CONV RATE Converge rate, which denote how fast the solution converges for the current
increment. A value of 0.0 means fast converges and a value > 1.0 means that the
solution will never converge.
ITR DIV Number of iteration divergences. Action to correction solution divergence will be
taken if ITRDIV > MAXDIV.
MAT DIV Number of material divergence + 1, i.e., it will be 1 if there is no material
divergence. Material divergence is due to bad creep strain or excessive sub-
increments in plasticity.
AVG R_FORCE Average residual force. In order for a time step to converge, this value must
become very small.
TOTAL WORK Accumulated total work done to the structure model. This value is only an
approximation.
DISP
AVG
MAX AT GRID C
The average displacement, the maximum displacement and its grid point
identification number and component number.
NO. QNV Number Quasi Newton vectors stored and used.
TOT KUD Total number of stiffness updates performed
TOT ITER Total number of iterations performed, including the number of stiffness updates
and time steps.
Table 1. Information In Nonlinear Iteration Summary Table
71
CHAPTER 3
For static analysis, this table is similar. In stead of TIME STEP, LOAD STEP and NO. INC (no. of
increments) are given in the table. An example is giving as the following.
N O N - L I N E A R I T E R A T I O N M O D U L E O U T P U T
LOAD NO. - - ERROR FACTORS - - CONV ITR MAT NO. AVG TOTL
STEP INC ITR DISP LOAD WORK RATE DIV DIV BIS R_FORCE WORK
0.1000 1 1 1.00E+00 1.62E+02 1.62E+02 1.000 0 1 0 1.56E+02 9.852E-01
0.1000 1 2 6.90E-01 2.66E+01 5.18E-01 0.164 0 1 0 3.12E+01 1.739E+00
0.1000 1 3 2.83E-01 1.63E+01 4.54E-01 0.227 0 1 0 1.91E+01 2.678E+00
0.1000 1 4 5.64E-02 6.76E+00 9.73E-02 0.537 0 1 0 7.91E+00 2.552E+00
0.1000 1 5 6.32E-02 5.88E+00 6.51E-02 0.549 0 1 0 6.82E+00 2.467E+00
0.1000 1 6 1.70E-01 6.23E+00 1.75E-02 0.796 0 1 0 6.99E+00 2.456E+00
- - - - - - - DISP - - - - - - - - LINE_S NO. TOT TOT
AVG MAX AT GRID C FACT NO QNV KUD ITR
1.32E-02 9.800E-02 8 1 0.36 3 0 0 1
2.31E-02 2.047E-01 6 1 0.04 1 1 0 2
2.96E-02 2.627E-01 6 1 1.00 0 2 0 3
2.89E-02 2.561E-01 6 1 1.00 0 3 0 4
2.88E-02 2.521E-01 6 1 2.00 1 4 0 5
2.97E-02 2.532E-01 6 1 2.00 1 5 0 6

72
Output Data Grouping: NLPACK
For the nonlinear transient analysis of a large problem with many time steps, the amount of data to be
stored in the database will be huge if we do not devise a scheme to group the output data. Data grouping
will save CPU time, IO time, and disk space to store the data. Another advantage is that, for a large
problem, if accident happens in the middle of a run, data can be salvaged for later restart.
There are two types of data required to be saved:
Output data are displacements, stresses, and strains, etc. at each output time step to be printed on
the f06 file or plotted by the post processors. They are controlled by the NO field on the
TSTEPNL bulk data entry. For this type of data, there is nothing much we can do except to
output them as requested by the user.
Restart data are used to described the material and geometric state of the structure model. We use
them to reconstruct the stiffness, the mass, the damping matrices and other tables required in
later usage. Restart data are usually much larger than the output data if the structural model is
large. This type of data can be saved selectively without degrading the effectiveness of the
transient analysis.
For this purpose, in the current release, a user modifiable parameter: PARAM, NLPACK,N is available.
N means that SOL 400 will pack output data for N output time steps and restart data for the last time step
as a single data package. For example, if N=100 (the default), then one data package has output data for
100 output time steps and restart data for the last time step. Later usage, such as restart or initial condition
for later step, can be performed only at NLPACK data group boundaries.
Some Ns have special meaning:
N= -1, all output data for a STEP and restart data for the end of the STEP are grouped into a
single package. This is the SOL 129 grouping method. In this case, the restart can be performed
only at STEP boundaries.
N=0, this is illegal.
N=1, each package of data on the database includes the output data for one output time step and
restart data. This is the NLSTATIC grouping method. Therefore, for the nonlinear static analysis,
the restart can be performed at each user output load increment, which is controlled by INTOUT
on the NLPARM bulk data entry.
NLPACK is used in Example 1.
For the nonlinear static analysis, NLPACK=1.
73
CHAPTER 3
Restarts
The purpose of a nonlinear restart is to allow the user to use the material and geometrical properties of a
previously converged solution as a new starting point to continue the analysis. This is useful when the
user want to change the loading sequence, the solution criteria, or to extend the analysis.
For SOL 400, a user-friendly restart procedure has been implemented. For restarts, the following
principles are noted:
The restart must be continued at a previous converged solution point in a nonlinear transient
analysis or a static analysis by specifying a SUBCASE, STEP, and/or TIME (LOADFAC). This
is accomplished by using the Case Control command NLRESTART; please refer to the Quick
Reference Guide, section (4A).
When the cold start is ANALYSIS=NLSTAT, it can be restarted at any user specified output load
increment (controlled by NOUT in NLPARM bulk data deck).
When the cold start is ANALYSIS=NLTRAN, it must be restarted from a saved or check-
pointed time step. The checkpoint times are dependent on DT and NO values on the
TESTEPNL bulk data entry, and PARAM, NLPACK of the cold start run. The checkpoint times
are integer multiples of (DT x NO) x NLPACK. For example, if DT=0.001 second, NO=10, and
NLPACK=100, the possible times that can be used for restart are at 1.0, 2.0, etc. If a requested
restart time does not match a checkpoint time, the closest checkpoint time will be used
The geometry and the initial material properties of the structural model cannot be modified. This
is obvious because any modification to the geometry or the initial material properties would
invalidate the previous analysis and require the nonlinear solution to start from the very
beginning. In such cases, it is simpler to initiate another cold start.
Performing restart is described in the following sub-sections.
File Management Commands
For a restart, the data of the cold start must be made available by using the File Management commands.
For nonlinear restart, two commands are needed: ASSIGN and RESTART. These two commands are
existing commands and special requirements are not needed for SOL 400.
There are many methods to retrieve data for a restart. One method is given in the example below. For
other methods, please refer to the file manage section of the Quick Reference Guide or chapter 12 of the
Reference Manual.
Case Control Modifications
The presence of a Case Control command NLRESTART indicates that the current run is a restart
execution. The Case Control file contains both subcases and steps, which have been executed in the cold
start, and those that are to be executed in the restart. The first subcase, step and/or load factor to be
executed in the restart is indicated by the options on the NLRESTART command. This is shown by the
following example:

74
NLRESTART SUBCASE 1, STEP 2, TIME 0.3
SUBCASE 1
STEP 1
LOAD = 10
STEP 2
LOAD = 20
STEP 3
LOAD = 30
In the above example, the first step through time 0.3 of the second step has been previously executed.
The restart execution begins with time 0.3 of the second step, and continues through the end of the third
step. If time 0.3 is not a restart point saved by NLPACK on the cold start, SOL 400 will search for the
nearest restart point on the data base and use that point to begin the restart. For restart, the Case Control
file structure for SUBCASE and STEP commands must be the same as the cold start up to the restart
point. After the restart point, the user may modify the Case Control file structure for SUBCASE and
STEP commands. For example, in above example, steps 1 and 2 must exist in the cold start. However,
step 3 may or may not exist in the cold start.
The following Case Control commands may be modified in a nonlinear restart:
Boundary conditions such as MPC and SPC.
Nonlinear solution control, NLPARM/TSTEPNL.
The LOAD requests.
Output request such as DISP and NLSTRESS.
The analysis type ANALYSIS.
Depending on the option selected with the NLRESTART command, the nonlinear restart may be
logically divided into three types: a case restart, a step restart, or a time restart:
The case restart begins the execution with a SUBCASE. All five types of modification described
above are legal for a case restart.
The step restart begins the execution with a STEP, which may be a new step or a previously
executed step. Although boundary condition and analysis type modifications are allowed, the
user has the responsibility to determine whether they are meaningful. Special attention should be
given to the analysis type modification; it may not be meaningful in many situations and, thus,
leads to erroneous results.
The time restart begins execution with a user specified TIME. For a time restart, the user should
not modify the analysis type, boundary conditions, or load requests. The user needs to exercise
discretion when attempting other types of modification at this level. Also, in order to perform
this type of restart, the specified TIME must be at the NLPACK data group boundary. If it is not,
SOL 400 will search for the nearest data boundary and use this boundary as the restart point.
75
CHAPTER 3
Bulk Data Modifications
The Bulk Data file for a nonlinear restart contains only those entries that are to be added to the cold start.
The deletion Bulk Data entry / cannot be used. This is to serve as a reminder that the geometry and the
initial properties cannot be modified. The user may make modifications to the Bulk Data file by
introducing new entries, which may be copies of the original entries with appropriate changes and new
identification numbers. The following list of entries can be added in a restart:
Load entries such as LOAD, FORCE, PLOAD4, and SPCD.
NLPARM/TSTEPNL entries.
Boundary condition entries such as SPC, SPC1, and MPC.
Examples for cold start and restart are given in Example 4 and Example 5.

76
Initial Conditions
In SOL 400, the traditional way of requesting initial conditions by using the Case Control
command IC and it associated bulk data entry TIC to assign initial displacements and velocities
is supported.
In the current release, a user-friendly interface to define initial condition is implemented. This is
the new Case Control command NLIC. It allows users to assign the result of any previous STEP
of NLSTAT analysis as the initial condition for the 1
st
STEP of NLTRAN analysis in a
SUBCASE. The following is an example:
SUBCASE 10
STEP 1
ANALYSIS = NLSTAT
LOAD = 10
NLAPRM = 110
STEP 2
ANALYSIS = NLSTAT
LOAD = 20
NLPARM = 120
STEP 3
ANALYSIS = NLTRAN
NLIC STEP 1 LOADFAC 0.5
DLOAD = 30
TSTEPNL = 130
Now, the 3
rd
STEP will use the result of the 1
st
STEP at 50% load increment as the initial condition.
Please note that the NLIC can only be defined at a load increment whose output flag is on - an available
restart point in static analysis. Otherwise, a fatal error message will be issued and job will be terminated.
Here, SOL 400 will not search for the nearest available restart point, because we want the user to know
the precise restart point.
In above example, if LOADFAC 0.5 is left out, then STEP 3 will take last state of STEP 1 as its initial
condition. In this case, the job will always run because the last state of a step is always a restart point.
For NLIC, there are some rules and limitations:
SOL 400 requires that all STEPs of NLSTAT must be located before all STEPs of NLTRAN. In
other words, analysis change can only occur once between NLSTAT and NLTRAN in one
SUBCASE.
In one SUBCASE, the beginning time of a transient STEP will be reset to 0.0 when the analysis
type is changed from a static STEP to transient STEP (or NLIC Case Control card is detected).
If the user does not specify a NLIC card when the analysis type changes from NLSTAT to
NLTRAN, the final result of the last STEP of NLSTAT will be picked up as the initial condition.
77
CHAPTER 3
The NLIC (or IC) can only appear in the first transient analysis STEP (ANALYSIS=NLTRAN)
in a SUBCASE. Otherwise, it will be ignored.
The new Case Control command NLIC can only specify restart-able NLSTAT location as the
initial condition. (Please read the Restart section for the definition of Restart-able.)
It is not allowed to use NLIC to select the initial condition from any previous STEP of
NLTRAN, SOL 400 will issue a fatal error message and user should run a restart job instead.
In the same STEP, the NLIC cannot appear together with an IC. A fatal error message will be
issue when NLIC and IC appear in the same STEP.
The NLIC can only be used in SOL 400 (NONLIN).
Parameter ICOPT is used with the NLIC and IC Case Control Commands. At the beginning of a
NLTRAN step, the user input loads may or may not be in equilibrium with the results of the previous
preload step. When , SOL 400 will compute the initial acceleration based on users inputs.
Otherwise, it will be assumed that the initial acceleration is null. In other words, when
(the default), it is assumed that the whole structure is in equilibrium automatically. Theoretically,
gives better performance algorithm since it guarantees the equilibrium in the beginning to
avoid a suddenly jump of loads or displacements. The drawback of comes from the
characteristics of the mass matrix, whose inverse matrix is required when computing the initial
acceleration. The mass matrix is usually highly singular for a lumped mass matrix or for a model with
only solid 3D elements, a large amount of CPU times may be required and the accuracy of the result may
be in doubt.
An alternative way to resolve the problem of with a suddenly large jump of loads or
displacements is to insert a NLTRAN step in between the analysis step and its previous preload step. This
step will have very short time duration in comparison with the analysis step and will provide a transition
between the preload step and the analysis step.
NLIC is used in Example 4 and Example 5. ICOPT is also used in these two examples.
ICOPT = 0
ICOPT=1
ICOPT=0
ICOPT=0
ICOPT=1

78
Transient Temperature Loads
A new capability, which has never been supported in the original nonlinear transient analysis (SOL 129),
is added into SOL 400 when ANALYSIS=NLTRAN. It is the time-dependent dynamic thermal effect,
which is applied to all the nonlinear elements in the residual.
The time-dependent thermal-elastic Equation can be written as follows
where:
T
(t) = The thermal strain,
T(t) = The current temperature is defined in Equation

{T
p
}is the temperature field and f(t) is the time function,
T
ref
= The reference temperature,
T
0
= The stress free temperature (initial temperature), and
(T) = The coefficient of thermal expansion.
For all nonlinear elements, the temperature effect, in both static and transient, is directly handled as
thermal strain in SOL 400 when computing the element forces.
To support it in nonlinear transient analysis, two new bulk data entries are created for the current release.
They are TTEMP and TMPSET. Basically, TTEMP is to define a time-dependent dynamic thermal field,
T(t), which includes a spatial temperature distribution (TMPSET) and a time function (TABLEDi), in the
same form as TLOAD1. TMPSET defines the spatial distribution by referencing a set of grid points. The
temperatures (T
p
) of these grid points are defined by TEMPD, TEMP, TEMPP1, or TEMPRB in the
normal way. Please see the Quick Reference Guide, section 8 Bulk Data, for the details of these two new
bulk data cards. By using TTEMP and TMPSET, the whole model can be separated into finite sub-
regions and each sub-region can have its own temperature distribution pattern. If it is necessary, the user
can also make every grid point as an independent sub-region or make the whole model as a single sub-
region.
As in nonlinear static analysis, TEMP(INIT) and TEMP(LOAD) commands are used in the Case Control
Deck to define the temperature input in nonlinear transient analysis. The SID of TEMP(LOAD) can refer
to TTEMP to define the transient temperature load for a STEP. The spatial temperature distribution
defined by TEMP, TEMPD, etc., must have the same SID as that of the associated TTEMP. If
TEMP(INIT) refers to TTEMP entry, only the spatial temperature distribution of the entry is used and the
time function is ignored.
The case control command TEMP(LOAD) can also refers to a spatial distribution (TEMP, etc.) directly
without TTEMP. In this case, the temperature time functions are linearly interpolated for the current step
) T T ( ) T ( ) T ) t ( T ( )) t ( T ( ) t (
ref 0 0 ref T
=
) ( } { ) ( t f T t T
P
=
79
CHAPTER 3
by using the last values of the previous step and the values of the spatial distribution referenced by the
TEMP(LOAD) command.
The thermal effects computed by the above method depend on the current material state and geometric
shape of the structural model. Therefore, they are called the nonlinear transient temperature loads.
For all up stream superelements and all linear elements in the residual, the thermal effects are computed
using the conventional method the user can use the DLOAD bulk data entry to combine multiple
TLOAD1 and TLAOD2s, whose EXCITE_ID reference thermal loads. The DLOAD bulk data entry
must be referenced by a DLOAD Case Control command to be selected for analysis. The transient
temperature loads computed by this method depend on the initial stiffness matrix only and are called the
linear transient temperature load. The TEMP(LOAD) and all its corresponding temperature related
bulk data entries introduced above can only describe the thermal effect for the nonlinear elements in the
residual. If there is no DLOAD Case Control command to select the temperature load for the linear
temperature load, the temperature effect for the linear part of the structure will be lost.
An example for the temperature load is given in Example 3 on page 90.

80
Boundary Condition (SPC and MPC) Changes
In SOL 400, the SPC and MPC are allowed to change from one step to next. This is accomplished by
placing the SPC or MPC case control command below step level.
For more detail of boundary condition changes, please refer to section SOL 400 Pretension Modeling
Description.
An example for boundary condition changes is given in Example 8 on page 100.
81
CHAPTER 3
Direct Matrix Input Changes
In SOL 400, the direct input matrices, K2PP, M2PP, B2PP and TFL, are allowed to change between steps
for nonlinear transient analysis. This is accomplished by the place these case control commands below
step level.
An example for the direct matrix changes is given in Example 7 on page 98

82
Rotor Dynamics
The nonlinear transient rotor dynamics is implemented in SOL 400. For details, please refer to Chapter
4 Rotor dynamics in MSC.Nastran Release Guide V2004.
An example for the rotor dynamics is given in Example 6 on page 97.
83
CHAPTER 3
Coordinate System for the Nonlinear Stresses of 3D
Elements
The nonlinear stresses are selected by the NLSTRESS Case Control command. The coordinate system
for the nonlinear stresses of 3D elements is used to be the element coordinate. In previous release, no
user options were allowed. In this release, we modify the coordinate system for these nonlinear stresses
to be output in user selected material coordinate system. This coordinate system is defined by CORDM
on the PSOLID Bulk Data entry.
Please note that the default coordinate system for CORDM is the basic coordinate system (0) instead of
the element coordinate system, therefore, the values for the nonlinear stresses will change for any exiting
deck.

84
Outputs
The outputs are requested by using the Case Control commands. All existing output Case Control
commands such DISPLACEMENT, VELOCITY, ACCELERATION, STRESS, NLSTRESS, OLOAD,
SPCFORCE, etc., are also allowed in the nonlinear analysis in SOL 400.
Two special outputs, Nonlinear Iteration Summary Table and PARAM, PH2OUT, are also available
for both nonlinear static and transient analysis in SOL 400. In addition, a new output control, PARAM,
NLPACK, n, is added for the current release for nonlinear transient analysis. For static analysis, only
PARAM, NLPACK, 1 is allowed and it is the default.
This new parameter, NLPACK (=100 is the default fro NLTRAN), is used to control the packed output
in SOL 400. The value of NLPACK represents the total number of output time steps in one output
package. SOL 400 will process the output procedure only after collecting all "NLPACK" output time
steps or at the end of each STEP. For detail, please see section Output Data Grouping: NLPACK above.
Because the matrix and table trailers output volume may be extremely high, the diagnostic output
requests of DIAG 8 and DIAG 15 have been turned off automatically when preparing the output data in
the solver of SOL 400 if NLPACK1. This action can reduce the .F04 file output size tremendously,
especially, when there is a large output time steps requested. User can force them on by setting DIAG 56
but it is not recommended.
85
CHAPTER 3
Error Handling
In SOL 400, there are three type fatal errors:
User or system fatal errors.
Fatal error due to solution divergence
Fatal error due to CPU time not enough.
For user or system fatal errors, if the error occurs before the solution iteration phase, the run will
terminate immediately without output the stored data. If the error occurs during the solution iteration
phase, SOL will try to output all stored data for the current subcase and terminate the run. The solution
will not continue into next subcase if there are multiple subcases.
For fatal errors due to solution divergence, SOL 400 will try to output all stored data for the current
subcase and terminate the subcase. The solution sequence will jump to perform the next subcase if there
are multiple subcases.
For fatal errors due to not enough CPU time, SOL 400 will try to output all stored data and terminate the
run.

86
User Interfaces
The user interfaces, which are important or new to the nonlinear transient analyses in SOL 400, are
summarized in this section. For detail, please refer to the Quick Reference Guide.
Nastran System Cells
STPFLG (SYSTEM (366)) Selects the SUBCASE or STEP layout when there are a number of
SUBCASE commands and no STEP command in a Case Control file.
TZEROMAX(SYSTEM (373)) Controls initial time step adjustment in nonlinear transient analysis.
File Management Commands
The following File Management commands are required for restarts. Please refer to the File Management
Section of the Quick Reference Guide or Chapter 12 of the MSC.Nastran Reference Manual for detail.
ASSIGN Assigns physical file names to database files that are used by a Nastran data deck to run a job.
RESTART Requests that data stored in a previous run be used in the current run.
Executive Control Command
SOL 400 or SOL NONLIN Requests the SOL 400 advanced integrated nonlinear solution sequence
Parameters
PARAM, LGDISP Requests a geometric nonlinear analysis. The default is 0, no geometric nonlinear
effect.
PARAM, FOLLOWK Requests whether the follower force stiffness will be used in a geometric
nonlinear analysis. The default is YES.
PARMAM, FKSYMFAC Controls whether the symmetrical follower force stiffness will be used in a
geometric nonlinear analysis. Default=0.24.
PARAM, LMFACT and PENFN Modify the scale factors to be used in kinematic elements and/or
contact analysis.
PARAM, MAXLP Specifies maximum number of iterations for element relaxation and material point
sub-increment process. Default=10.
PARAM, NLAYERS Specifies the number of layers for integration through shell thickness of the
material nonlinear properties. Default=5 for all except QUADR/TRIAR. For QUADR/TRIAR,
default=7.
PARAM, NLTOL Selects defaults for CONV, EPSU, EPSP, and EPSW for the Bulk Data entries
NLPARM and TSTEPNL. Default=2.
PARAM, PH2OUT Requests phase II outputs for a nonlinear analysis. Default=0, phase III output only.
87
CHAPTER 3
PARAM, NLPACK Control the total output time step in one output package; see section Output Data
Grouping above. Default=100.
PARAM, NDAMP Specifies the values when the HHT- method using in SOL 400, see section A
New Numerical Integration Method above. Default=-0.05.
PARAM, ICOPT Select how to handle the equilibrium when dealing with the initial condition, see
section Initial Conditions above.
Case Control Commands
ANALYSIS Selects solution method for an analysis step, see section Case Control Structure above.
NLIC Selects the initial condition from any static analysis for the nonlinear transient analysis
NLPARM Selects NLPARM Bulk Data entry.
NLRESTART Requests a restart execution at a specific solution point for SOL 400, see section Restarts
above.
NLSTRESS Requests the form and type of the nonlinear element stress output.
STEP Delimits and identifies an analysis step, see section Case Control Structure above.
TSTEPNL Selects the TSTEPNL Bulk Data entry.
VCCT Selects the VCCT Bulk Data entry.
Bulk Data Entries
MATHP- Specifies the hyperelastic material properties for an element.
MATS1 Specifies the stress-dependent material properties for an element.
MDLPRM Defines various model parameters for the analysis.
NLPARM Defines a set of parameters for the nonlinear static analysis iteration strategy.
TSTEPNL Defines a set of parameters for the nonlinear transient analysis iteration strategy.
TTEMP Defines a time-dependent temperature distribution for use in the nonlinear transient response.
TMPSET Defines a spatial temperature distribution for use in the TTEMP bulk data entry.
VCCT Selects the Virtual Crack Closure Technique to evaluate energy release rates.
The Bulk data entries for the newly enhanced large strain and material models. Please refer to SOL 400
Material and Elements on page 36 for description of these entries.

88
Examples
The following nine examples show the inputs and capabilities of the nonlinear static and transient
analysis. The intention of these examples is to show the input structure for SOL 400. The model itself
and the detailed entries in the Bulk Data Deck are not important.
Example 1
Example one, EX01, is a simplified version of the standard QA deck, NLTSUB02. This model only has
QUAD4 elements. It has both material nonlinearity (MATS1) and geometrical nonlinearity (PARAM,
LGDISP, 1). The 1
st
STEP will process the output data at every 5 output time steps and the 2
nd
STEP do
it only once because of the settings of the parameter NLPACK. All bold-font statements are entries
pertaining to the nonlinear analysis.
ID MSC, EX01 $
TIME 150 $
SOL 400 $
CEND
TITLE=ISOTROPIC MATERIAL & MATS1, ELLIPTIC CYLINDER UNDER EX01
SUBTITLE =SPC CHANGE IN EACH STEP, NLPACK's
SET 10 = 10000,11200
SET 20 = 101
SEALL = ALL
DISPL = ALL
STRESS = 20
$
SUBCASE 100
ANALYSIS=NLTRAN
STEP 10
PARAM,NLPACK,5
DLOAD = 100
SPC = 200
TSTEPNL = 310
STEP 20
PARAM,NLPACK,-1
DLOAD = 100
SPC = 400
TSTEPNL = 320
$
BEGIN BULK
PARAM NDMAP -0.05
PARAM LGDISP 1
TSTEPNL 310 100 0.01 10 AUTO
TSTEPNL 320 100 0.01 10 AUTO
$
PLOAD4 510 101 5. THRU 112
$
TLOAD1 100 510 0 0 120
TABLED1 120 +TBD1
+TBD1 0. 0. 5. 1. 16. 1. ENDT
MAT1 100 3.+7 0.3 .283-2
MAT1 101 3.+7 0.3 .283-2
MATS1 100 PLASTIC 3.+5 500000.
$
GRID 10000 100. 0.0 10. 345
GRID 10001 100. 0.0 0.0 345
GRID 10100 99.3625 3.30491 10. 345
89
CHAPTER 3
GRID 10101 99.3625 3.30491 0.0 345
GRID 10200 96.8149 6.51543 10. 345
GRID 10201 96.8149 6.51543 0.0 345
GRID 10300 92.5105 9.59323 10. 345
GRID 10301 92.5105 9.59323 0.0 345
GRID 10400 86.6025 12.5 10. 345
GRID 10401 86.6025 12.5 0.0 345
GRID 10500 79.2443 15.1974 10. 345
GRID 10501 79.2443 15.1974 0.0 345
GRID 10600 70.5889 17.6472 10. 345
GRID 10601 70.5889 17.6472 0.0 345
GRID 10700 60.7898 19.8111 10. 345
GRID 10701 60.7898 19.8111 0.0 345
GRID 10800 50. 21.6506 10. 345
GRID 10801 50. 21.6506 0.0 345
GRID 10900 38.3729 23.1276 10. 345
GRID 10901 38.3729 23.1276 0.0 345
GRID 11000 26.0617 24.2037 10. 345
GRID 11001 26.0617 24.2037 0.0 345
GRID 11100 13.2197 24.8406 10. 345
GRID 11101 13.2197 24.8406 0.0 345
GRID 11200 0.0 25. 10. 345
GRID 11201 0.0 25. 0.0 345
$
CQUAD4 101 100 10000 10001 10101 10100
CQUAD4 102 100 10100 10101 10201 10200
CQUAD4 103 100 10200 10201 10301 10300
CQUAD4 104 100 10300 10301 10401 10400
CQUAD4 105 100 10400 10401 10501 10500
CQUAD4 106 100 10500 10501 10601 10600
CQUAD4 107 100 10600 10601 10701 10700
CQUAD4 108 100 10700 10701 10801 10800
CQUAD4 109 100 10800 10801 10901 10900
CQUAD4 110 100 10900 10901 11001 11000
CQUAD4 111 100 11000 11001 11101 11100
CQUAD4 112 100 11100 11101 11201 11200
$
PSHELL 100 100 0.10 100 101
$
SPC1 200 16 11200 11201
SPC1 200 26 10000 10001
$
SPC1 400 16 11200 11201
SPC1 400 26 10000 10001
SPC1 400 1 10700
SPC1 400 2 10701
$
ENDDATA
Example 2
Example two, EX02, is modified version of the standard QA deck, NLTSUB02. It shows two different
types of analyses in the same job. This model is similar to the Example one except for adding some static
loads and the required NLPARMs. All bold-font statements are entries that show the difference in the
two different analysis types.
ID MSC, EX02 $
TIME 150 $
SOL 400 $

90
CEND
TITLE=TEST MIXED ANALYSES - NLSTAT AND NLTRAN EX02
SUBTITLE =SPC CHANGE IN THE STEPS IN EACH SUBCASE
SET 10 = 10000,11200
SET 20 = 101
SEALL = ALL
DISPL = ALL
STRESS = 20
$
SUBCASE 100
ANALYSIS=NLSTAT
STEP 10
LOAD = 800
SPC = 200
NLPARM = 110
STEP 20
LOAD = 900
SPC = 400
NLPARM = 120
$
SUBCASE 200
ANALYSIS=NLTRAN
STEP 10
DLOAD = 100
SPC = 200
TSTEPNL = 310
STEP 20
DLOAD = 100
SPC = 400
TSTEPNL = 320
$
BEGIN BULK
NLPARM 110 10 AUTO YES
NLPARM 120 10 AUTO YES
$
LOAD 800 0.01 1.0 510
LOAD 900 0.05 1.0 510
( The rest is same as what in the Bulk Data Deck in the 1st Example)
ENDDATA
Example 3
Example three, EX03, is modified form of the standard QA deck, NLTTL002. This model only has 1
QUAD4 element and 2 TRAI3 elements. Its major purpose is to show the various combinations of
TTEMP and TMPSET inputs in nonlinear transient analysis for the thermal effect. All the bold-font
statements are entries related to the temperature related inputs.
ID MSC, EX03 $
SOL 400
DIAG 8,15
TIME 60
CEND
SEALL = ALL
SUPER = ALL
TITLE = THERMAL LOAD TEST FOR NONLINEAR TRANSIENT ANALYSIS EX03
SUBTITLE = Q4/T3 MODEL, TTEMP AND TMPSET
$ECHO = NONE
MAXLINES = 999999999
$
91
CHAPTER 3
TEMPERATURE(INITIAL) = 1
SUBCASE 1
analysis=NLTRAN
step 1
TSTEPNL= 1
SPC = 2
TEMPERATURE(LOAD) = 3
nlstress = all
stress = all
step 2
TSTEPNL= 1
SPC = 2
nlstress = all
stress = all
SUBCASE 2
analysis=NLTRAN
step 3
TSTEPNL= 1
SPC = 2
nlstress = all
stress = all
step 4
TSTEPNL= 1
SPC = 2
nlstress = all
stress = all
SUBCASE 3
analysis=NLTRAN
step 5
TSTEPNL= 1
SPC = 2
nlstress = all
stress = all
step 6
TSTEPNL= 1
SPC = 2
nlstress = all
stress = all
SUBCASE 4
analysis=NLTRAN
step 7
TSTEPNL= 1
SPC = 2
nlstress = all
stress = all
step 8
TSTEPNL= 1

92
SPC = 2
nlstress = all
stress = all
$
BEGIN BULK
PARAM POST -1
PARAM COUPMASS 1
PARAM LGDISP 1
PARAM K6ROT 100.
PARAM,NOCOMPS,-1
PARAM PRTMAXIM YES
PARAM,COMPMATT,YES
PARAM,EPSILONT,INTEGRAL
PARAM NLTOL 0
TSTEPNL,1,4,0.25,1,AUTO
$
PCOMP 1 79. 0.
* 1 .04875 0. YES
$
MAT8 1 7.15+6 2.9+6 .29 1.4+6 1.9-4
2.9-6 6.-6 79.
MATT8 1 3 5 4 6
1 2
$
TABLEM1 1 + CR
+ CR 60. 2.9-6 70. 2.9-6 80. 3.24-6 100. 3.86-6 + CS
+ CS 120. 4.01-6 140. 3.89-6 150. 3.78-6 160. 3.68-6 + CT
+ CT 180. 3.52-6 200. 3.47-6 220. 3.55-6 240. 3.76-6 + CU
+ CU 250. 3.87-6 260. 3.99-6 280. 4.12-6 300. 4.24-6 + CV
+ CV 320. 4.24-6 ENDT
$
TABLEM1 2 + CW
+ CW 60. 6.-6 70. 6.-6 80. 7.67-6 100. 1.168-5+ CX
+ CX 120. 1.341-5 140. 1.37-5 150. 1.349-5 160. 1.328-5+ CY
+ CY 180. 1.266-5 200. 1.222-5 220. 1.218-5 240. 1.259-5+ CZ
+ CZ 250. 1.296-5 260. 1.334-5 280. 1.415-5 300. 1.46-5 + DA
+ DA 320. 1.46-5 ENDT
$
TABLEM1 3 + BX
+ BX 60. 7.15+6 70. 7.15+6 80. 7.15+6 100. 7.13+6 + BY
+ BY 120. 7.11+6 140. 7.08+6 150. 7.07+6 160. 7.07+6 + BZ
+ BZ 180. 7.06+6 200. 7.05+6 220. 7.05+6 240. 7.04+6 + CA
+ CA 250. 7.04+6 260. 7.05+6 280. 7.06+6 300. 7.08+6 + CB
+ CB 320. 7.08+6 ENDT
$
TABLEM1 4 + CM
+ CM 60. .29 70. .29 80. .29 100. .29 + CN
+ CN 120. .29 140. .29 150. .29 160. .29 + CO
+ CO 180. .29 200. .29 220. .29 240. .29 + CP
+ CP 250. .29 260. .29 280. .29 300. .29 + CQ
+ CQ 320. .29 ENDT
$
TABLEM1 5 + CC
+ CC 60. 2.9+6 70. 2.9+6 80. 2.9+6 100. 2.82+6 + CD
+ CD 120. 2.75+6 140. 2.68+6 150. 2.64+6 160. 2.58+6 + CE
+ CE 180. 2.47+6 200. 2.35+6 220. 2.22+6 240. 2.09+6 + CF
+ CF 250. 2.03+6 260. 1.95+6 280. 1.8+6 300. 1.65+6 + CG
+ CG 320. 1.65+6 ENDT
$
93
CHAPTER 3
TABLEM1 6 + CH
+ CH 60. 1.4+6 70. 1.4+6 80. 1.4+6 100. 1.34+6 + CI
+ CI 120. 1.29+6 140. 1.24+6 150. 1.22+6 160. 1.2+6 + CJ
+ CJ 180. 1.15+6 200. 1.1+6 220. 980000. 240. 870000.+ CK
+ CK 250. 810000. 260. 750000. 280. 620000. 300. 500000.+ CL
+ CL 320. 500000. ENDT
$
cquad4,1,1,1,2,5,4
ctria3,2,1,1,2,4
ctria3,3,1,2,5,4
$
GRID 1 0.00000 0.00000 0.00000
GRID 2 1.00000 0.00000 0.00000
GRID 4 0.00000 1.00000 0.00000
GRID 5 1.00000 1.00000 0.10000
$
SPCADD 2 1
SPC1 1 123456 1 2
spc1 1 123456 4
$
TTEMP,3,111,300
TMPSET,111,4,5
TTEMP,3,101,310
TMPSET,101,1,2
$
TTEMP,4,102,400
TMPSET,102,1,2,4,5,7,8,9,
,10,11,12
$
TTEMP,5,201,500
TMPSET,201,1,2,4,5
$
TTEMP,6,-1,400
$
TTEMP,7,202,700
TMPSET,202,1,2
$
TTEMP,8,204,800
TMPSET,204,1,2
$
TTEMP,9,402,900
TMPSET,402,1,2,4,5
$
TEMP 1 1 79.
TEMP 1 2 79.
TEMP 1 4 79.
TEMP 1 5 79.
$
TEMP 3 1 80.
TEMP 3 2 80.
TEMP 3 4 80.
TEMP 3 5 80.
TABLED1 300
0.0 .9875 1.0 1.0 ENDT
TABLED1 310
0.0 .9875 1.0 1.0 ENDT
$
TEMP 4 1 81.
TEMP 4 2 81.
TEMP 4 4 81.

94
TEMP 4 5 81.
TABLED1 400
1.0 .9876542 2.0 1.0 ENDT
$
TEMP 5 1 80.
TEMP 5 2 80.
TEMP 5 4 80.
TEMP 5 5 80.
TABLED1 500
0.0 .9875 1.0 1.0 ENDT
$
TEMP 6 1 81.
TEMP 6 2 81.
TEMP 6 4 81.
TEMP 6 5 81.
$
$
TEMP 7 1 80.
TEMP 7 2 80.
TEMP 7 4 80.
TEMP 7 5 80.
TABLED1 700
0.0 .9875 1.0 1.0 ENDT
$
TEMP 8 1 81.
TEMP 8 2 81.
TEMP 8 4 81.
TEMP 8 5 81.
TABLED1 800
1.0 .9876542 2.0 1.0 ENDT
$
TEMP 9 1 80.
TEMP 9 2 80.
TEMP 9 4 80.
TEMP 9 5 80.
TABLED1 900
0.0 .9875 1.0 1.0 ENDT
$
TEMP 10 1 81.
TEMP 10 2 81.
TEMP 10 4 81.
TEMP 10 5 81.
ENDDATA
Example 4
Example four, EX04, is modified from the standard QA deck, NLTIC19. This model only has 1 HEXA
element. Its purpose is to shows two different types of analyses in the same SUBCASE, the model itself
is not important. All the bold-font statements are entries that show the difference between those analyses
and how to set the initial condition for the nonlinear transient analysis after static analysis. Note that the
nonlinear transient analysis does not use the final results of the closest static analysis as the initial
condition; instead, it asks the results of the 50% load increment in the 1
st
STEP to be the initial condition.
Also, parameter ICPOT=0 is selected, which will compute the initial acceleration at the beginning (t=0.0)
of the transient analysis when it is not in balance.
ID MSC, EX04 $
DIAG 8,15
95
CHAPTER 3
TIME 60
SOL 400 $
CEND
TITLE= ELASTIC-PLASTIC STATIC & TRANSIENT RESPONSE, EX04
SUBTI= INITIAL ACCELERATION COMPUTED - PARAM,ICOPT,0
SET 1 = 1111
SET 2 = 100
DISP = 1
VELO = 1
ACCE = 1
OLOAD = 1
$ STRESS(PLOT) = 2
SUBCASE 1130
step 1
LABEL=UNIAXIAL TENSION (LOADING)
ANALYSIS=NLSTAT
SPC=100
LOAD=1130
NLPARM = 1
step 2
LABEL=UNIAXIAL TENSION (UNLOADING)
ANALYSIS=NLSTAT
SPC=100
NLPARM = 1
step 10
LABEL=I.C. FROM THE FIRST NLSTAT STEP(50%) - UNBALANCED CASE (NLIC)
ANALYSIS=NLTRAN
SPC=100
DLOAD=2130
TSTEPNL=10
param,icopt,0
BEGIN BULK
PARAM,LANGLE,3
PARAm,LGDISP,1
PARAM,W4,1.0
$
NLPARM 1 4 AUTO ALL
1.-6
TSTEPNL 10 2000 0.001 AUTO
$
MAT1 1 30.0+6 11.5+6 0.3 7.332-2 0.01
PSOLID 1 1
$
SPC1 100 123456 1000
SPC1 100 1 1010
SPC1 100 2 1001
SPC1 100 3 1100
GRDSET 456
$
TLOAD1 2130 2130 0 500
TABLED1 500 +TAB1
+TAB1 0. 0. 1. -1. 1.2 0. 10. 0. +TAB2
+TAB2 ENDT
$
LOAD 1130 -1.6 2. 121
LOAD 2130 -1.6 1. 121
$
GRID 1000 0. 0. 0.
GRID 1100 1. 0. 0.

96
GRID 1110 1. 1. 0.
GRID 1010 0. 1. 0.
GRID 1001 0. 0. 1.
GRID 1101 1. 0. 1.
GRID 1111 1. 1. 1.
GRID 1011 0. 1. 1.
$
CHEXA 100 1 1000 1100 1110 1010 1001 1101 +HX100
+HX100 1111 1011
$
PLOAD4 121 100 36.+3 1100 1111
PLOAD4 121 100 36.+3 1000 1011
ENDDATA
Example 5
Example five, EX05, is a modified version of the standard QA deck, NLTIC19R. This model is a restart
run of the Example four, EX04. Since there is no structure change in the SUBCASE 1130 and there is no
parameter in NLRESTART command, this restart job will start from the 2
nd
SUBCASE. Its purpose is to
shows how the Case Control commands NLRESTART and NLIC can work together. All bold-font
statements are entries that show the key Case Control commands in this example. Note that the nonlinear
transient analysis in the 2
nd
SUBCASE asks the final results in the 1
st
STEP of the 1
st
SUBCASE to be
the initial condition. Also, parameter ICPOT=1 (the default) is selected, which will NOT compute the
initial acceleration but loads in the beginning (t=0.0) of the transient analysis, that assumes the whole
model is in equilibrium automatically when initial conditions are applied.
ASSIGN RSFILE='DBSDIR:ex04.MASTER' $
RESTART LOGICAL=RSFILE $
$
ID MSC, EX05 $
DIAG 8,15
TIME 60
SOL 400 $
CEND
TITLE= ELASTIC-PLASTIC STATIC & TRANSIENT RESPONSE, EX05
SUBTI= NO INITIAL ACCELERATION COMPUTED - PARAM,ICOPT1
SET 1 = 1111
SET 2 = 100
DISP = 1
VELO = 1
ACCE = 1
OLOAD = 1
$ STRESS = 2
NLRESTART
SUBCASE 1130
step 1
LABEL=UNIAXIAL TENSION (LOADING)
ANALYSIS=NLSTAT
SPC=100
LOAD=1130
NLPARM = 1
step 2
LABEL=UNIAXIAL TENSION (UNLOADING)
ANALYSIS=NLSTAT
SPC=100
NLPARM = 1
97
CHAPTER 3
step 10
LABEL=I.C. FROM THE FIRST NLSTAT STEP - UNBALANCED CASE (NLIC)
ANALYSIS=NLTRAN
SPC=100
DLOAD=2130
TSTEPNL=10
param,icopt,0
SUBCASE 1131
LABEL=I.C. FROM THE 1st NLSTAT STEP OF PREVIOUS SUBCASE (NLIC)
ANALYSIS=NLTRAN
NLIC SUBCASE 1130 STEP 1
SPC=100
DLOAD=2130
TSTEPNL=10
param,icopt,1
BEGIN BULKk
ENDDATA
Example 6
Example six, EX06, is modified from the standard QA deck, NLTROT01. This model simply shows how
to run Rotor Dynamics in SOL 400. All bold-font statements are the basic entries that may be required
in rotor dynamic analysis. For details, please refer to the MSC.NASTRAN V2004 Release Guide.
ID MSC, EX06 $
SOL 400
DIAG 8, 15
CEND
$
ANALYSIS=NLTRAN
RIGID=LINEAR
$
RGYRO= 100
TSTEP= 100
SET 99= 101
disp= 99
$
BEGIN BULK
$
UNBALNC, 100, 2.0, 101, 0., 1., 0.,
, 1.0, 0.0, 0.0, 0.0, 1000., none
$
TSTEPNL, 100, 5000, 3.0E-4, 10
$
ROTORG 10 101 THRU 103
$
RSPINT 10 101 102 FREQ 100
0.01
TABLED1, 100,
, 0.0, 22.5, 100.0, 22.5, endt
$
$ ROTOR 1
$
GRID, 101, , 0., 0., 0.
GRID, 102, , 1., 0., 0., , 14
GRID, 103, , 2., 0., 0.
GRID, 104, , 0., 0., 0.

98
GRID, 105, , 2., 0., 0.
$
RBE2, 1001, 102, 123456, 101, 103
RBE2, 1002, 101, 123456, 104
RBE2, 1003, 103, 123456, 105
$
CONM2, 1004, 102, , 50.,
, 5.0, , 15.0, , , 15.0
$
CELAS1, 1005, 1000, 104, 2
CELAS1, 1006, 1000, 104, 3
CELAS1, 1007, 1000, 105, 2
CELAS1, 1008, 1000, 105, 3
PELAS, 1000, 1.0E+5, 0.0
$
param, g, 0.05
param, w3, 141.3
$
enddata
Example 7
Example seven, EX07, is modified from the standard QA deck, NLTK2PP1. This model shows how to
apply K2PP in SOL 400, such as using different sets of K2PP in different STEP. For example, the 1
st

STEP requests the K2MAT matrix as a K2PP input and the 2
nd
STEP requests the combination of
K2MAT and K3MAT matrices as a K2PP input. All the bold-font statements are the entries that are
required in this kind of analysis.
ID MSC, EX07 $
SOL 400 $
DIAG 8,15
TIME 50
CEND
$
TITLE = LINEAR TRANSIENT RESP. (DIRECT METHOD) TEST NLTK2PP1
SUBTITLE = COULOMB FRICTION LESS F3(0),F4(0)SEALL=ALL
$ ECHO = NONE
SET 10= 1,7
DISP=10
SEALL = ALL
MPC = 20
SPC = 10
analysis=NLTRAN
SUBCASE 1
STEP 1
DLOAD = 12
K2PP = K2MAT
TSTEPNL = 10
STEP 2
DLOAD = 12
K2PP = K2MAT, K3MAT
TSTEPNL = 10
BEGIN BULK
param,ndamp,-0.055
$
CORD2R 1 0 0. 0. 0. 0. 0. 1. +CORD
+CORD 1. 0. 0.
$
99
CHAPTER 3
GRDSET 1 1 1246
GRID 1 -460. 0. 0.
GRID 3 -460. 0. 100.
GRID 5 -460. 0. 200.
GRID 2 290. 0. 0.
GRID 4 290. 0. 100.
GRID 6 290. 0. 200.
GRID 7 0. 0. 200.
$
CONM2 3 3 1 0.6867
CONM2 4 4 1 1.2735
CONM2 7 7 1 15.445 +CONM
+CONM 3.0E5
$
CELAS2 1 2200. 1 3 3 3
CDAMP2 1 3.2 1 3 3 3
CELAS2 2 4200. 2 3 4 3
CDAMP2 2 6.2 2 3 4 3
CELAS2 3 550. 3 3 5 3
CDAMP2 3 13.3 3 3 5 3
CELAS2 4 2000. 4 3 6 3
CDAMP2 4 0. 4 3 6 3
$
MPC 10 7 3 750. 5 3 -290. +MPC10
+MPC10 6 3 -460.
MPC 11 7 5 750. 5 3 -1. +MPC11
+MPC11 6 3 1.
MPCADD 20 10 11
$
SPC1 10 5 1 THRU 6
$
TABLED1 12 +TBLD11
+TBLD11 -100. 0. 0. 0. 0.005 3. 100. 3. +TBLD12
+TBLD12 ENDT
$
EPOINT 101
$
DMIG K2MAT 0 1 1 1
DMIG K2MAT 1 3 101 2.
DMIG K2MAT 101 1 3 -2.
$
DMIG K3MAT 0 1 1 1
DMIG K3MAT 1 3 101 3.
DMIG K3MAT 101 1 3 -3.
$
DAREA 12 101 1.
DELAY 12 101 0.
$
EPOINT 202
$
DMIG K2MAT 2 3 202 2.
DMIG K2MAT 202 2 3 -2.
$
DMIG K3MAT 2 3 202 3.
DMIG K3MAT 202 2 3 -3.
$
DAREA 12 202 1.
DELAY 12 202 0.54
$
TLOAD1 12 12 12 12

100
$
TSTEPNL 10 1000 0.002 1 AUTO 10000
ENDDATA
Example 8
Example eight, EX08, is modified from the standard QA deck, NLTSTP06. This model shows the
example of the SPC-Set change between STEPs. All the bold-font statements are the entries that are
required .
ID MSC, EX08 $
TIME 150 $
SOL 400 $
DIAG 8,15 $
CEND
TITLE=ANISOTROPIC MATERIAL & MATS1, ELLIPTIC CYLINDER UNDER NLTSTP06
SUBTITLE = TWO STEP'S TEST, SPC SET CHANGE AND SAME DT
SET 10 = 10000,11200
SET 20 = 101
SEALL = ALL
DISPL = ALL
STRESS = 20
SPCF = ALL
OLOAD = ALL
$
SUBCASE 100
ANALYSIS=NLTRAN
STEP 10
DLOAD = 100
SPC = 200
TSTEPNL = 310
STEP 20
DLOAD = 100
SPC = 400
TSTEPNL = 320
$
BEGIN BULK
PARAM LGDISP 1
TSTEPNL 310 100 0.01 10 AUTO +TS11
TSTEPNL 320 100 0.01 10 AUTO +TS21
$
PLOAD4 510 101 5. THRU 112
$
TLOAD1 100 510 0 0 120
TABLED1 120 +TBD1
+TBD1 0. 0. 5. 1. 16. 1. ENDT
MAT2 100 3.2967+79.8901+60. 3.2967+70. 1.1538+70.283-2
MAT2 101 3.2967+79.8901+60. 3.2967+70. 0.283-2
MATS1 100 PLASTIC 3.+5 500000.
$
GRID 10000 100. 0.0 10. 345
GRID 10001 100. 0.0 0.0 345
GRID 10100 99.3625 3.30491 10. 345
GRID 10101 99.3625 3.30491 0.0 345
GRID 10200 96.8149 6.51543 10. 345
GRID 10201 96.8149 6.51543 0.0 345
GRID 10300 92.5105 9.59323 10. 345
GRID 10301 92.5105 9.59323 0.0 345
GRID 10400 86.6025 12.5 10. 345
GRID 10401 86.6025 12.5 0.0 345
101
CHAPTER 3
GRID 10500 79.2443 15.1974 10. 345
GRID 10501 79.2443 15.1974 0.0 345
GRID 10600 70.5889 17.6472 10. 345
GRID 10601 70.5889 17.6472 0.0 345
GRID 10700 60.7898 19.8111 10. 345
GRID 10701 60.7898 19.8111 0.0 345
GRID 10800 50. 21.6506 10. 345
GRID 10801 50. 21.6506 0.0 345
GRID 10900 38.3729 23.1276 10. 345
GRID 10901 38.3729 23.1276 0.0 345
GRID 11000 26.0617 24.2037 10. 345
GRID 11001 26.0617 24.2037 0.0 345
GRID 11100 13.2197 24.8406 10. 345
GRID 11101 13.2197 24.8406 0.0 345
GRID 11200 0.0 25. 10. 345
GRID 11201 0.0 25. 0.0 345
$GRID 20000 0. 0. 0. 123456
$GRID 20001 100. 0. 0. 123456
$GRID 20100 0. 100. 0. 123456
$
CQUAD4 101 100 10000 10001 10101 10100
CQUAD4 102 100 10100 10101 10201 10200
CQUAD4 103 100 10200 10201 10301 10300
CQUAD4 104 100 10300 10301 10401 10400
CQUAD4 105 100 10400 10401 10501 10500
CQUAD4 106 100 10500 10501 10601 10600
CQUAD4 107 100 10600 10601 10701 10700
CQUAD4 108 100 10700 10701 10801 10800
CQUAD4 109 100 10800 10801 10901 10900
CQUAD4 110 100 10900 10901 11001 11000
CQUAD4 111 100 11000 11001 11101 11100
CQUAD4 112 100 11100 11101 11201 11200
$
PSHELL 100 100 0.10 100 101
$
SPC1 200 16 11200 11201
SPC1 200 26 10000 10001
$
SPC1 400 16 11200 11201
SPC1 400 26 10000 10001
SPC1 400 1 10700
SPC1 400 2 10701
$
ENDDATA
Example 9
Example nine, EX09, is modified from the standard QA deck, NLTSTP07. This model shows that the
upper stream superelement can have output requests that are different from the residual. Also, a case
control command OTIME is used here to select a subset of all output time steps, which is selected by the
TSTEPNL bulk data card. It can reduce the output data dramatically. For example, in the following deck,
TSTEPNL asks output at time=0.0, 0.1,, 1.0 second but OTIME overwrites this request and only
makes output at time=0.5 second. All the bold-font statements are the entries that are required to
complete all above requests in this example.
ID MSC, EX09 $
TIME 150 $
SOL 400 $

102
DIAG 8,15 $
CEND
TITLE=Test For Upperstream Superelement Output Request EX09
SUBTITLE = OTIME Output Request
SET 1 = 0.5
SET 10 = 11200
SET 20 = 101
LOADSET = 500
DISPL = 10
OTIME = 1
$
SUBCASE 10
SUPER=10
METHOD=10
SET 10001 = 10001
DISP =10001
SUBCASE 100
ANALYSIS=NLTRAN
DLOAD = 100
SPC = 200
TSTEPNL = 310
$
BEGIN BULK
PARAM LGDISP 1
$
SESET,10,10000,thru,10301
SEQSET1 10 0 10500 THRU 11001
EIGRL 10
$
TSTEPNL 310 100 0.01 10 AUTO +TS11
$+TS11 1.E-2
LSEQ 500 110 510
PLOAD4 510 101 5. THRU 112
$
TLOAD1 100 110 0 0 120
TABLED1 120 +TBD1
+TBD1 0. 0. 5. 1. 16. 1. ENDT
$MAT1 100 3.+7 0.3 .283-2
MAT2 100 3.2967+79.8901+60. 3.2967+70. 1.1538+70.283-2
MAT2 101 3.2967+79.8901+60. 3.2967+70. 0.283-2
MATS1 100 PLASTIC 3.+5 500000.
$
GRID 10000 100. 0.0 10. 345
GRID 10001 100. 0.0 0.0 345
GRID 10100 99.3625 3.30491 10. 345
GRID 10101 99.3625 3.30491 0.0 345
GRID 10200 96.8149 6.51543 10. 345
GRID 10201 96.8149 6.51543 0.0 345
GRID 10300 92.5105 9.59323 10. 345
GRID 10301 92.5105 9.59323 0.0 345
GRID 10400 86.6025 12.5 10. 345
GRID 10401 86.6025 12.5 0.0 345
GRID 10500 79.2443 15.1974 10. 345
GRID 10501 79.2443 15.1974 0.0 345
GRID 10600 70.5889 17.6472 10. 345
GRID 10601 70.5889 17.6472 0.0 345
GRID 10700 60.7898 19.8111 10. 345
GRID 10701 60.7898 19.8111 0.0 345
GRID 10800 50. 21.6506 10. 345
GRID 10801 50. 21.6506 0.0 345
GRID 10900 38.3729 23.1276 10. 345
103
CHAPTER 3
GRID 10901 38.3729 23.1276 0.0 345
GRID 11000 26.0617 24.2037 10. 345
GRID 11001 26.0617 24.2037 0.0 345
GRID 11100 13.2197 24.8406 10. 345
GRID 11101 13.2197 24.8406 0.0 345
GRID 11200 0.0 25. 10. 345
GRID 11201 0.0 25. 0.0 345
$
CQUAD4 101 100 10000 10001 10101 10100
CQUAD4 102 100 10100 10101 10201 10200
CQUAD4 103 100 10200 10201 10301 10300
CQUAD4 104 100 10300 10301 10401 10400
CQUAD4 105 100 10400 10401 10501 10500
CQUAD4 106 100 10500 10501 10601 10600
CQUAD4 107 100 10600 10601 10701 10700
CQUAD4 108 100 10700 10701 10801 10800
CQUAD4 109 100 10800 10801 10901 10900
CQUAD4 110 100 10900 10901 11001 11000
CQUAD4 111 100 11000 11001 11101 11100
CQUAD4 112 100 11100 11101 11201 11200
$
PSHELL 100 100 0.10 100 101
$
SPC1 200 16 11200 11201
SPC1 200 26 10000 10001
$
SPC1 400 16 11200 11201
SPC1 400 26 10000 10001
SPC1 400 1 10700
SPC1 400 2 10701
$
ENDDATA

104
Ch. 4: Implicit & Explicit Nonlinear Analysis MD R2 Nastran Release Guide

4
Implicit & Explicit Nonlinear
Analysis
MD R2 Nastran Implicit Nonlinear - SOL 600
MD R2 Nastran Explicit Nonlinear - SOL 700


106
MD R2 Nastran Implicit Nonlinear - SOL 600
The MD R2 release contains significant enhancements to functionality and performance. This includes
the addition on heat transfer capability, support of Nastran fastener technology (CBUSH, CFAST and
CWELD) in a native mode, improved support of PLOAD4, and an enhanced composite capabilities.
Fracture mechanics may also be performed to obtain the stress intensity factors at a crack. A new
mechanism has been added for the direct transfer of data that is known as streaming input.
Heat Transfer
For heat transfer, most of the capabilities in Nastran SOL 153 and 159 are supported by SOL 600 with
the exception of CHBDYP and forced convection, the equivalents of which are not currently available in
Marc. The main advantage of using SOL 600 for heat transfer over SOL 153 or 159 is that thermal contact
is available directly and that radiation view factors may possibly be calculated faster. The user needs to
weight the drawbacks of not having CHBDYP and forced convection. Because of these alternatives,
SOL 600 offers two ways to perform a heat transfer analysis. The direct (new) method uses Marc to
perform all of the calculations and can support thermal contact that varies during the run. The other
(indirect) method is to calculate the thermal contact conditions (if they are needed) at the start of the run
and perform the rest of the calculations using Nastran SOL 153/159. This option is addressed using a new
option on the SOL 600 entry TSOLVE=M or TSOLVE=N respectively. A typical SOL 600 Executive
Control statement for heat transfer using the direct method would be:
SOL 600,153 TSOLVE=M
A typical Executive Control statement for heat transfer using the indirect method would be:
SOL 600,153 TSOLVE=N
To use the Thermal Contact capability released with the Nastran 2005 version either TSOLVE=N
should be used or the TSOLVE option should be left blank (which will support most existing input
decks).
SOL 600 heat transfer addresses conduction, free convection, radiation to space, cavity radiation, thermal
contact and latent heat. Steady state or transient heat transfer calculation may be obtained. All material
properties may be temperature dependent, and the material may be isotropic, orthotropic or anisotropic.
For the direct method, Marcs table input is used for all applicable input items. The direct method requires
postprocessing using the Marc t16 file. All standard output forms (op2, xdb, f06 and/or punch) are
available using the indirect method.
The temperature history obtained may then be used in a subsequent thermal stress simulation by using
the MINSTAT and MCHSTAT bulk data options to read the temperatures off the t16 file. When used in
conjunction with the MTHERM bulk data option the time steps will be either subdivided or merged to
satisfy the accuracy and convergence requirements of the nonlinear mechanical analysis.
For the directly solutions, when CTRIA3 or CQUAD4 elements are used, the thermal conduction can be
based upon either two methods which is selected on the PSHELL option. Similar to conventional
Nastran, the thermal behavior may be membrane like only, in which case there is no thermal gradient
107
CHAPTER 4
Implicit & Explicit Nonlinear Analysis
through the thickness. To support this, new heat transfer elements have been added and are used when
appropriate. These new elements are:
Element 196 Three-node, Bilinear Heat Transfer Membrane), 904
Element 197 Six-node, Biquadratic Heat Transfer Membrane), 907
Element 198 Four-node, Isoparametric Heat Transfer Element), 911
Element 199 Eight-node, Biquadratic Heat Transfer Membrane, 915
The second method is that the element has a thermal gradient through the thickness, which may be
required for composite simulation or thermal shock type problems. This is activated by specifying a
nonzero MID2 entry. The MPHEAT options is used to specify, whether the temperature gradient is linear
or quadratic through the total thickness of the shell, or linear or quadratic variation is specified per layer
basis. In the later case if a composite shell has n layers the number of degrees of freedom per grid is n+1
or 2*n+1 for the quadratic case.
The MHEATSHL parameter may also be used to control this behavior.
Additions/changes to a standard Nastran SOL 153 or SOL 159 heat transfer input file are as follows:
Executive Control
Change SOL statement as described above.
Case Control
No changes
Bulk Data
BCBODY, BCTABLE (BCPARA if necessary to change defaults) if there is thermal contact
MPHEAT New entry that maps to Marcs HEAT parameter. See, MPHEAT (SOL 600) on
page 1973 of the MD Nastran Quick Reference Guide for more information.
NLHEATC - Defines numerical analysis parameters for SOL 600 Heat Transfer Analysis. For more
information please see, NLHEATC (SOL 600) on page 1988 of the MD Nastran Quick Reference
Guide.
MCHSTAT - Option to change state variables for SOL 600 Used in SOL 600 only. For more
information please see, MCHSTAT (SOL 600) on page 1927 of the MD Nastran Quick Reference
Guide.
MINSTAT - Option to define initial state variables for SOL 600 Used in SOL 600 only. Please see,
MINSTAT (SOL 600) on page 1941 of the MD Nastran Quick Reference Guide for more information.
Bulk Data Parameters
PARAM,MARCHEAT is obsolete starting with this release.

108
Bulk Data
BCBODY, BCTABLE (BCPARA if necessary to change defaults) if there is thermal contact. Please see
the MD Nastran Quick Reference Guide for more details on entries, DMIGOUT, MCHSTAT, MINSTAT,
MTHERM, MPHEAT, NLHEATC.
PARAM,MARHTPRT (Integer) Control heat transfer output in the Marc .out file
0 = Do not print any output except for summary tables
1 = Print the nodal temperatures
2 = Print all possible nodal heat transfer output
PARAM,MRADUNIT (Integer) Controls the units used in radiation heat transfer for SOL 600
1 = Degrees Celsius
2 = Degrees Kelvin (default if parameter not entered)
3 = Degrees Fahrenheit
Remark: Degrees Rankin are not available
PARAM,MHEMIPIX (Integer) Controls the number of pixels used in radiation heat transfer for
SOL 600 using the hemi-cube method. The default, if this parameter is not
entered is 500.
PARAM,MARVFCUT (Real) Controls the fraction of the maximum view factor that is to be used
as a cutoff. View factors calculated below this cutoff are ignored. Default
is 0.0001 if this parameter is not entered (Used in SOL 600 radiation heat
transfer only)
PARAM,MRVFIMPL (Real) Controls the fraction of the maximum view factor that is to be
treated implicitly (contribute to operator matrix). View factor values
smaller than this cutoff are treated explicitly. Default is 0.01 if this
parameter is not entered using this parameter reduces the size of the heat
transfer operator matrix, which reduces the computational costs associated
with decomposition. (Used in SOL 600 radiation heat transfer only)
PARAM,MRSTEADY (Integer) Controls the solution method for SOL 600 steady state heat
transfer
1 Marc STEADY STATE is used with TIME STEP of 1.0 (default if
parameter not entered) The specific heat matrix is not formed.
2 AUTO STEP is used.
Remark: Requires that a sufficiently large time period to be simulated for the
solution to reach steady state.
109
CHAPTER 4
Heat Transfer Examples
The following heat transfer examples are located in the tpl1 directory:
Creep Simulations
Creep is an important phenomena in high temperature applications. To facilitate this type of analysis
several new options have been added into SOL 600 in the MD r2 release.
The MPCREEP allows you to select the procedure used to perform the time integration.If the explicit
method is used, then the time step must be small, but is the material is elastic and small deformation, no
reassembly of the stiffness matrix is required. If the implicit method is chosen, then larger time steps may
be used, but reassembly occurs at every increment.
In thermal creep simulations it is necessary that the time step be chosen to satisfy accuracy of both the
rate independent thermal stress problem and the rate dependent creep problem. To insure that this occurs
the MTCREEP bulk data option has been introduced.
Element Selection
To increase the flexibility in performing simulations the MRALIAS param has been augmented by the
ALIASM bulk data entry. This allows one to map the Marc element type to be used for a selection of
elements. All Marc element types may be used if they are topologically similar to the Nastran element
type. See Marc Volume B for greater details.
This is analogous to the PCOMPLS, PLCOMP, PSHLN1, PSHLN2 and PSLDN1 options in SOL 400.
Other SOL 600 Items
Membrane Elements
SOL 600 now directly supports membrane-only shells without the user having to add any alias bulk data
entries or parameters. This was made possible due to a new element which was added in Marc:
Element 200 Six-node, Biquadratic Isoparametric Membrane), 920
Conduction mhqbd1, mhqbd1a, mhqbd1c, mhqbd1s, mhqbd2, mhqbd2c, mhqbd2s,
mhbc01, mhbc02, mhtepe
Free Convection mhcbv1, mhcbv1a, mhcbv1b, mhcbv1c, mhcbv1d, mhcbv1e
Radiation to Space mhrad1, mhrad2, mhrad3
Cavity Radiation mhrcv1, mhrcv1a, mhrcv2, mhrhx0, mhrdhx, mhrhx4, mhrc1t
Thermal Contact mhcnoc, mhtc07, mhtc7a
Latent Heat mtlat1, mtlat2

110
This new element completes the set of shell elements necessary to support membranes. Users should be
careful when using membranes with nonlinear analyses as they are unstable under certain conditions
because of a lack of bending stiffness.
Fracture Mechanics
New capabilities have been added to SOL 600 to allow greater insight in determining the fracture
resistance of your designs. These methods complement the material damage models that existed
previously and are entered through the MATHED option. The new capabilities include:
Calculation of Energy Release Rates and Stress Intensity Factors VCCT and LORENZI
Option
Crack Propagation VCCT
Delamination COHESIV
New Failure Criteria MATF
Birth and death of elements
Virtual Crack Closure Technique (VCCT)
Marcs revised VCCT capability is fully supported by this release of SOL 600. It involves a anew VCCT
Case Control entry and a new VCCT Bulk Data entry (see the MD Nastran Quick Reference Guide for
details). This option defines that the virtual crack closure technique is to be used for evaluating energy
release rates. The user defines the node (in 2-D or for shells) or nodes (in 3-D) that define each crack tip.
The supported elements are lower- and higher-order 2-D solids and 3-D shells, lower- and higher-order
3-D hexahedral solids and lower order 3-D tetrahedral solids. For 3-D solids it is important that a regular
mesh around the crack front is used.
Multiple cracks can be defined and results obtained for each crack separately. Each crack consists of a
crack tip node in 2-D and for shells and a list of nodes along the crack front for 3-D solids. Shell elements
can be used for defining a 2-D style line crack and also be connected to the face of another shell or 3-D
solid to form a 3-D style surface crack. The different cases are automatically identified.
The VCCT method is advantageous because it may be used with any material model including
orthotropic or anisotropic behavior, and because it automatically obtains the mode I, II, and III stress
intensity factors. This makes is applicable to composite structures.
For crack propagation, there are two modes of growth: fatigue and direct growth. For fatigue style, the
user specifies a load sequence time period. During the load sequence, the largest energy release rate and
the corresponding estimated crack growth direction is recorded. At the end of the load sequence, the
crack is grown using the specified method. For direct growth, the crack grows as soon as the calculated
energy release rate is larger than the user-specified Gc. Note that Gc can be made a function of the
accumulated crack growth length to model a crack growth resistance behavior. This release does not
support large crack propagation which requires remeshing.
111
CHAPTER 4
Fracture Mechanics J-Integral (LORENZI)
This option gives an estimation of the J-Integral for a crack configuration using the domain integration
method. The domain integration method has the advantage that it can also be used for problems with
thermal behavior and for dynamic analysis. This procedure is only available for continuum elements.
Only the nodes defining the crack front (crack tip in two dimensions) need to be defined. The program
automatically finds integrations paths according to the format below. The complete J-Integral is
evaluated and output. For the case of linear elastic material with no external loads on the crack faces, the
program automatically separates mode I, mode II, and mode III (3-D only) stress intensity factors from
the J-Integral. for isotropic materials.
A new Bulk Data entry, LORENZI, is necessary to activate this capability and if entered applies to all
subcases in the analysis. For more details the MD Nastran Quick Reference Guide.
Delamination
An alternative method to model failure is to use the COHESIV bulk data option in conjunction with
special delamination or interface elements. Three different models are available along with a user
subroutine. The user defines the traction versus the relative separation. The area under the curve is the
cohesive energy, often known as the critical energy release rate.
Bilinear model
Exponential Model
Linear-Exponential Model
Element Type Number of Nodes Characteristic
186 4 Planar
187 8 Planar
188 8 3-D
189 20 3-D

112
MATEP Extensions
Material description MATEP was extended to add Chaboche, Power Law, Kumar, Johnson Cook and
other options. For more details the MD Nastran Quick Reference Guide.
SOL 600 Failure Description MATF
For SOL 600 failure indices or actual material failure is only described using the MATF entry. For this
release, MATF has been totally revised to accommodate additional types of failure and improve the input
and user understanding of the input. The user should be aware that other Nastran solutions can specify
failure index calculation on various MAT entries. These specifications are not available in SOL 600
only MATF may be used. To activate the new entries, the PARAM,MRMATFSB, 1 must also be
included. For more details on the MATF entry, see the MD Nastran Quick Reference Guide.
190 4 Axisymmetric
191 8 Axisymmetric
192 6 3-D
193 15 3-D
Element Type Number of Nodes Characteristic
113
CHAPTER 4
Primary Format (SOL 400, 600)
Primary Format Example:
(Note: The 4th and 6th lines cannot be entirely blank and the last line of the
3rd criteria has been omitted)
1 2 3 4 5 6 7 8 9 10
MATF MID ITYPE SB
"CRI" Criteria Xt Xc Yt Yc Zt Zc 1st
Sxy Syz Szx Find Fxy Fyz Fzx Ext
Exc Eyt Eyc Ezt Ezc Gxy Gyz Gzx
"CRI" Criteria Xt Xc Yt Yc Zt Zc 2nd
"CRI" Criteria Xt Xc Yt Yc Zt Zc 3rd
1 2 3 4 5 6 7 8 9 10
MATF 100 0
+ CRI 1 2500. 4000. 2500. 4000. 2000. 3000. 1st
+ 4500. 4500. 4500.
+
+ CRI 2 2nd
+
+ .11 .06 .1 .05 .075 .03 .03 .03
+ CRI 4 2500. 4000. 2500. 4000. 2000. 3000. 3rd
+ 4500. 4500. 4500. 0.90

114
Alternate Format (SOL 600, 700 Only)
Example Alternate Format
1 2 3 4 5 6 7 8 9 10
MATF MID CRI67 Xt Xc Yt Yc Zt Zc
1 2 3 4 5 6 7 8 9 10
1 2 3 4 5 6 7 8 9 10
MATF 100 1 2500. 4000. 2500. 4000. 2000. 3000.
4500. 4500. 4500.
Field Contents
MID Identification number of a MAT1, MAT2, MAT8, MATORT or MAT9 entry
(Integer>0, no default). See Remarks 1 and 2.
ITYPE Flag to invoke progressive failure (Integer, default=0 for SOL 400 and 1 for SOL 600)
0 No progressive failure compute failure indices only (Default)
2 Progressive failure (presently available only for Hashin and Puck methods)
SB Allowable shear stress of the bonding material between layers (composites only) (Real,
No default)
"CRI" For the alternate format, enter the character string "CRI" to start each criteria
(Character, Required)
CRI67 Used with the alternate format only (Integer, no default, required). It is highly
recommended that only one criterion be used. However, up to three criteria from the
list under Criteria below can be specified in a packed list as follows:
1000000*ITYPE+10000*C3+100*C2+C1
Where C1, C2, C3 are integer values for the various Criteria listed below. For SOL
700, only criteria number 6 is available and progressive failure will occur if the limits
are reached regardless of the value of ITYPE.
Criteria Select an integer corresponding to the failure criteria to be applied (integer, no default)
Up to three failure criteria may be specified for each MID for SOL 400 or 600. Only
one failure criteria may be used for SOL 700 and the primary format should be used.
115
CHAPTER 4
1. Maximum stress criterion. (SOL 400, 600 only, See Remark 3)
2. Maximum strain criterion. (SOL 400, 600 only, See Remark 4)
3. Hill failure criterion. (SOL 400, 600 only, See Remark 5)
4. Hoffman failure criterion. (SOL 400, 600 only, See Remark 6)Tsai-Wu failure
criterion. (SOL 400, 600 only, See Remark 7)
5. Chang-Chang failure criterion (SOL 700 only)
6. Hashin failure criteria (SOL 400, 600 only, remark 8)
7. Puck failure criteria (SOL 400, 600 only, remark 11) - must not be combined
with any other method
8. User defined failure criteria (SOL 400, 600 only, remark 12)
9. Hashin-Tape (SOL 400, 600 only, remark 13)
10. Hashin-Fabric (SOL 400, 600 only, remark 14)
Xt Maximum tensile stress in x-direction (Real>0. or blank)
Xc Maximum compressive stress (absolute value) in x-direction (Real>0., default=Xt)
Yt Maximum tensile stress in y-direction (Real>0., default=Xt)
Yc Maximum compressive stress (absolute value) in y-direction (Real>0., default=Yt)
Zt Maximum tensile stress in z-direction (Real>0., default=Xt)
Zc Maximum compressive stress (absolute value) in z-direction (Real>0., default=Zt)
Sxy Maximum shear stress in xy-plane (Real>0. or blank)
Syz Maximum shear stress in yz-plane (Real>0., default=Sxy)
Szx Maximum shear stress in zx-plane (Real>0., default=Sxy)
Find Failure index (Real>0., default=1.) See Remarks 5-7.
Fxy Interactive strength constant for xy-plane (Real<0., default= ).
Fyz Interactive strength constant for yz-plane (Real<0., default= ).
Fzx Interactive strength constant for zx-plane (Real<0., default= ).
Ext Maximum tensile strain in x-direction (Real>0. or blank)
Field Contents
c t c t
Y Y X X
1 1
2
1
c t c t
Z Z Y Y
1 1
2
1
c t c t
X X Z Z
1 1
2
1

116
Element Birth and Death
Starting with this release it is possible to deactivate and re-activate elements in the model that have failed
or for some other reason needs to be deactivated or re-activated. This is accomplished using Case Control
commands DEACTEL and ACTIVAT as well as matching Bulk Data entries DEACTEL and ACTIVAT.
Once an element is deactivated or activated it stays that way during the entire subcase case unless it fails
due to a MATF criteria. Please see the MD Nastran Quick Reference Guide for further details.
Unglue
Frequently in contact analysis it is known beforehand that two surfaces will never separate once they
contact. To prevent numerical chattering contact between these surfaces is frequently described using
glued contact. In order to perform VCCT analysis of such surfaces it might be necessary to unglue those
nodes near a crack. A new Bulk Data entry, UNGLUE, is available for such purposes. Please see the MD
Nastran Quick Reference Guide for further details.
Composite Element Numerical Analysis
In previous versions, SOL 600 provided two options for composite analyses (1) complete through the
thickness integration at every iteration and (2) the smeared approach as used in other Nastran solution
sequences. The first approach is more accurate particularly for nonlinear analyses where local buckling
takes place and the analysis needs to extend well into the post-bucking regime. The second approach is
usually satisfactory for small deformation linear static and dynamic analyses. Method 1, complete
integration through the thickness has been modified such that the accuracy has been retained, but the
computational times and memory requirements have been significantly reduced. These are known as
fast integration techniques and are described by the new Bulk Data entry PCOMPF. The limitation is
that using these fast integration procedures the material may not exhibit any nonlinear behavior. Large
deformation and buckling is supported using these procedures.
Exc Maximum compressive strain (absolute value) in x-direction (Real>0., default=Ext)
Eyt Maximum tensile strain in y-direction (Real>0., default=Ext)
Eyc Maximum compressive strain (absolute value) in y-direction (Real>0., default=Eyt)
Ezt Maximum tensile strain in z-direction (Real>0., default=Ext)
Ezc Maximum compressive strain (absolute value) in z-direction (Real>0., default=Ezt)
Gxy Maximum shear strain in xy-plane (Real>0.)
Gyz Maximum shear strain in yz-plane (Real>0., default=Gxy)
Gzx Maximum shear strain in zx-plane (Real>0., default=Gxy)
Field Contents
117
CHAPTER 4
The following table indicates typical performance improvements with this release
when using this option.
Please see the MD Nastran Quick Reference Guide for further details.
PLOAD4 Extensions
SOL 600 versions previous to this one did not support the PLOAD4 continuation line. In addition, if
corner pressures with different values were entered, they were averaged. SOL 600 now fully supports
different corner pressures, pressures specified by the CID, N1, N2, and N3 fields on the continuation line
and line loads specified by the CID, N1, N2, and N3 fields. The SOLR field is fully supported. The LDIR
field is not supported. Line loads must be specified using the CID, N1, N2, and N3 fields rather than
LDIR. For SOL 600, the SORL field applies to CQUAD4 and CTRIA3 as well as CQUAD4R and
CTRIAR elements. The CID field may reference an ID of any CORD1R, CORD1C, CORD1S,
CORD2R, CORD2C or CORD2S entry but not CORD3G.
All of these new PLOAD4 extensions are activated by entering PARAM,MRPLOAD4,2 in the bulk data
or by placing this parameter in one of the RC files. For this release these capabilities are not the default
and must be activated using this parameter. For the next release, it is anticipated that MRPLOAD4=2
will become the SOL 600 default.
Large Rotation RBE
Improved large motion RBE capabilities have been added to SOL 600, however it was decided to retain
the small rotation formulation as the default. This will allow models used with previous versions to
obtain the same results. To activate the large RBE rotation capability, add the following parameter to the
bulk data:
PARAM,marc7601,1
Streaming Input
A new capability available in this release is known as streaming input. Normally SOL 600 will form
a Marc input file from the Nastran input file, then execute Marc to compute the results. With streaming
input Marc is not executed. Instead, the Marc subroutines which are now in Nastran are used directly
and called by the main SOL 600 routine. The Marc input file is still formed and saved on disk for possible
Model Number of Element
Maximum Number of
Layers
Improvement in CPU
for Stiffness Matrix
Memory
Reduction
1 10000 46 1425 % 1550 %
2 88854 182 2013 % 957 %
3 48858 33 1085 % 795 %
4 420015 48 1263 % 886 %

118
future use by the customers, but the same information is passed in memory from the main SOL 600
routine to the Marc initiation routine, thus saving computer time because a physical Marc input file on
disk does not need to be opened and read. Streaming input is activated using the Bulk Data entry:
PARAM,MRSTREAM,1
Streaming input is not available with DDM (parallel processing) or if user subroutines are necessary. For
those cases, do not include param,mrstream and run SOL 600 as before. Streaming input should also not
be used if a special version of Marc is necessary for your particular application. In that case, specify one
of the PATH options on the SOL 600 Executive statement in conjunction with a file to point the analysis
to the location of the version of Marc that is desired.
CONNECTOR TECHNOLOGY
CBUSH, CWELD and CFAST have been added to Marc as nonlinear (large deformation and rotation)
elements. These formulations are now available in SOL 600 using PARAM,MARCWELD (see the QRG
for additional details). The user should beware that CBUSH, CWELD, CFAST, RBE2, RBE3 in SOL 600
and Marc are truly nonlinear elements while in Nastran they are linear small deformation elements. Thus
different results will sometimes be obtained. In addition, solutions that converge with SOL 106, 129 or
400 may not converge with SOL 600 (the opposite may also occur). The user should be careful when
using such elements to make sure they are applicable to both linear and nonlinear solutions when
constructing the model.
The new formulation of the CBUSH is activated using the MARCBUSH,-1, param. If CBUSH entries
are used with a nonzero CID, then the MRCOORDS,0, param should also be included.
The new formulation of the CFAST and CWELD is activated using the MARCWELD,1 param.
The other new parameters that control CWELD behavior are:
MARIPROJ
MRCWANGL
MRFACEA
MRFACEB
MRHERRMN
MRITTYPE
The following CWELD parameters are not supported in SOL 600
CWDIAGP
CWRANDEL
The following CWELD parameters are supported in SOL 600
CWLDIGNR
If the new CWELD is used, then MSPEEDCW is ignored
119
CHAPTER 4
The following CFAST parameters are not supported in SOL 600
CFDIAGP
CFRANDELT
COMPUTATIONAL ENHANCEMENTS
The Direct Iterative Solver (MARCSOLV=2) can now be used with out-of-core assembly to allow larger
models to be analyzed. It should be noted that there is a decrease in performance. The program will
automatically use the out-of-core option if necessary or you can select it by using the MARCOOCC
param.
SUPER ELEMENTS / DMIG
The use of super elements in conjunction with SOL 600 has increased over the last year, and several
changed have been made in this area. In addition to the MDMIOUT that may be used to create a reduced
stiffness matrix (super element) or an Adams MNF file, the DMIGOUT option has been added.
The DMIGOUT option may be used to output either the complete global matrices or individual element
matrices at the element levels. These global matrices include:
Stiffness matrix This is the total stiffness matrix, including geometric stiffness, follower force
and friction contributions.
Differential stiffness This is only available in a buckle sub case
Mass matrix Available in dynamics
Damping matrix Available in dynamics
Conductivity matrix Available in heat transfer
Specific Heat matrix Available in transient heat transfer
The element matrices correspond to the above and can be output in either the basic system or in the
transformed system. The global stiffness matrices are always output in the transformed system.
It should be noted that the output of these matrices may be huge. To reduce the size of the DMIG files it
is possible to filter out small values. It should be noted that doing this may influence subsequent
calculations.
CONTACT ENHANCEMENTS
The spline option used with deformable bodies via the BCBODY option can now be used with higher
order elements. This improves the calculation of when contact is to occur and the normal to the surface.

120
New SOL 600 Parameters
The following new SOL 600 parameters have been introduced as described below. Please see the MD
Nastran Quick Reference Guide for more details:
PARAM,MARMTLCK Determines whether a check of various property-material combinations
for SOL 600 will be made or not. This slightly slows down the input
processing.
PARAM,MARLDCMB Determines whether extraneous loads in the input file will be combined
to save computer time.
PARAM,MARLDRMV Determines whether extraneous FORCE, MOMENT and/or PLOAD4
entries in the input file will be filtered out at an early stage to save
computer time.
PARAM,MARNOCID SOL 600 by default does not support MCID defined by cylindrical or
spherical coordinate systems. This parameter determines whether MCID
defined by cylindrical or spherical coordinate systems will be ignored or
fataled out for shell and solid elements depending on this parameter.
Inclusion of this parameter overcomes this problem.
PARAM,MRCPENTA This entry determines how CPENTA will be mapped to Marc degenerate
solid elements. Marc does not presently have wedge elements, so
CPENTA elements must be mapped to degenerate hexa elements such as
type 7.
PARAM,MRHERRMN This entry controls whether extra grids created for such items as
hyperelastic Herrmann elements, CWELD, etc. are output or not in the
op2, f06, punch and/or xdb files. When Herrmann grids are output, the
displacement value is actually pressure which might be confusing when
looking at an f06 file.
PARAM,MARCMID3 This entry controls whether MID3 will be set to the same value as MID2
when the Marc PSHELL option is used (designated by
PARAM,MRPSHELL,1 or when the SMEAR option is used on the SOL
600 Executive Control statement)
PARAM,MFORDUPE This entry controls how duplicate forces encountered for the same load
case are handled in SOL 600.
PARAM,MARBK105 This entry controls whether linear buckling or nonlinear buckling
eigenvalues are calculated for SOL 600,105.
PARAM,MRCTRIA3 This entry controls the default Marc element type for CTRIA3
elements (75 or 138) in SOL 600
PARAM,MRCQUAD4 This entry controls the default Marc element type for CQUAD4
elements (75 or 139) in SOL 600
121
CHAPTER 4
Platform Specific Notes
For Linux IA64 and EM64T platforms:
The default MPI for these platforms is HP MPI.
Intel MPI is also supported and can be used by switching to it using the maintain script under the tools
directory.
To use the Intel MPI (iMPI), please observe the following.
PARAM,MARCDUPE This entry controls whether SOL 600 will check for duplicate entries for
most every type of bulk data card. SOL 600 does not allow duplicate
entries, but the portion of IFP that runs prior to spawning Marc does not
usually check for duplicate entries.
PARAM,MARCSTOP This entry controls whether a check model run will be performed, no
actual simulation will occur, and the analysis will stop with a Marc Exit 7.
PARAM,MRRSTOP2 Normally op2, xdb, punch and f06 output is not available for SOL 600
restart analyses. Setting this parameter to 1 will allow the program to
attempt to create one or more of these files. Only in limited cases will the
job be successful.
PARAM,MRBDYCVT Determines if CHBDYG is converted to CHBDYE for SOL 600 heat
transfer
PARAM,MRMATFSB Determines if the version of Marc being used supports the new MATF SB
field or not
PARAM,MRDELTTT Determines how delta time is set for each "step" of a SOL 600 transient
nonlinear analysis.
PARAM,MARCFEAT,N If entered will add FEATURE,N to the Marc input file in the parameters
section
PARAM,MRCOORDS Determines whether Marc COORD SYSTEM will be added if any
CORD1i or CORD2i entries are in the model and if CBUSH elements are
present in the model
PARAM,MARCWELD Determines how CWELD/PWELD and CFAST/PFAST elements will be
translated to Marc
PARAM,MRITTYPE Type of constraint used to connect the auxiliary nodes in all CWELDs
PARAM,MRCWANGL Angle in degrees over which to rotate the CWELD cross-section about
the beam axis to obtain its final orientation.
PARAM,MARIPROJ Flag to determine if auxiliary nodes of a CWELD will be projected on the
model
PARAM,MARFACEA Face number for A side of weld if welds are made of solid elements
PARAM,MARFACEB Face number for B side of weld if welds are made of solid elements

122
a. Create a .mpd.conf file in your home directory that contains the following line.
secretword=<your mpd password>
where <your mpd password> can be any arbitrary string.
Change mode of the .mpd.conf to 600, i.e. do a
chmod 600 $HOME/.mpd.conf
b. Setup a mpd.hosts file in your home directory consists of the names of nodes in your cluster
(it can have only 1 node, i.e. 1 line):
clusternode1
clusternode2
clusternode3
The rest is taken care of by the run_marc script.
However, if your cluster requires password to perform ssh or rsh between nodes, you may need to enter
your password every time you are running a parallel job. You can disable the password requirement in
the use of ssh or rsh. Please consult your system administrator.
For 64-bit Windows EM64T platforms:
The default MPI for this platform is MPICH2. MS MPI is not supported for this release of MD Nastran
for SOL 600.
To install MPICH2, go into the mpichx64\bin directory and type
smpd -install
For 32-bit Windows platforms:
The default MPI for this platform is Argonne National Labs MPICH2.
To use the MPICH2 please observe the following:
a. Go into the mpich2\bin directory and type
smpd install
If smpd could not be installed, you may need to reboot your system.
b. The first time your run a parallel job, you will be required to enter your login ID and password.
If your system is rebooted or your command prompt is closed, you will need to repeat step b.
123
CHAPTER 4
Supported Systems for SOL 600 in MD R2 Nastran
1 Hardware MPI version also available (via maintain in /tools directory).
2 Supports Solver 6.
3 Supports multi-threading.
4 Supports true 64-bit version.
5 Supports the Intel MPI 3.0
Vendor OS Hardware
FORTRAN
Version
C
Version
Parallel
Enabled Default MPI
Also Works
On
HP-Alpha
(DEC)
4
Tru64 5.1 Alpha
Server 4100
f90 5.5 cc 6.4 yes HP MPI 2.0
1
HP (64-bit)
2,4
HP (64-bit)
2,4
HPUX 11.0
HPUX 11.23
PA2.0
Itanium 2
f90 2.9.2
f90 2.8.7
C.03.50
A.06.02
yes
yes
HP MPI 2.0
HP MPI 2.2
IBM (64-bit)
4
AIX 5.2 RS/6000 &
RS/6000 SP
xlf 8.1.1 cc 6.0.0 yes MPICH
1
IBM POE 4.1
SGI (mips4
64-bit)
2,3,4
SGI (Altix 64-
bit)
2, 4
IRIX 6.5
Linux 2.4.21-
sgi303r2
R12000
Itanium 2
(Propack
3.0)
f90 7.4
Intel 8.1
cc 7.4
Intel 8.1
yes
yes
MPICH
1
SGI MPT 1.10.1 Propack 4.0
Sun (64-bit)
4
Solaris 2.8 UltraSPARC
III
f90 8.1 cc 5.7 yes MPICH
1
Linux (32-bit) RedHat 9 Intel
Pentium or
equiv.
Intel 8.1 Intel 8.1 yes HP MPI 2.2.5 RedHat AS 3.0
Linux (64-bit)

4,5
RedHat AS 3.0 Itanium 2 Intel 8.1 Intel 8.1 yes HP MPI 2.2.5
5
Linux (64-bit)

4,5
RedHat WS 3.0 Intel EM64T Intel 8.1 Intel 8.1 yes HP MPI 2.2.5
5
AMD Opteron,
RedHat WS 4.0
Intel (32-bit)
Intel (64-bit)
4
Windows 2000
Windows
Server 2003
x64
Intel
Pentium or
equivalent
Intel EM64T
Intel 8.1
Intel 8.1
Intel 8.1
Intel 8.1
yes
yes
MPICH2
MPICH2
Windows XP,
Intel 9.1
MPICH2,
Windows XP 64

124
MD R2 Nastran Explicit Nonlinear - SOL 700
Introduction
SOL 700 is the second release of powerful Explicit Nonlinear Solution available in MD Nastran and
offers an advanced technology to analyze transient dynamic events of short duration with severe
geometric and material nonlinearities.
MD Nastran SOL 700 allows users to work within one common modeling environment using the same
Bulk Data interface. The NVH, linear and nonlinear models can be used for explicit applications such as
crash, crush, and drop test, blade out and bird strike simulations. This dramatically reduces the time spent
to build different models for implicit and explicit analysis and prevents the users from making mistakes
because of unfamiliarity between different programs.
Latest Capabilities of MD Nastran Explicit Nonlinear - SOL 700
SOL 700 have been dramatically improved to include the following new capabilities in this release:
Fluid Structure Interaction - Airbag and Occupant Safety simulation
Time Domain NVH (FFT Method)
Pre-stress with implicit solver
Many additional Elements and Material Models
Support for MD Nastran native output files
Airbags and Occupant Safety
In this release, SOL 700 includes Fluid Structure Interaction (FSI) capability that is based on the
advanced Finite Volume (Eulerian) and General Coupling Technology available in Dytran. The FSI
capability in MD R2 release, however, is only limited to airbag and occupant safety simulation.
The objective of fluid-structure interaction using the coupling algorithm is to enable the material
modeled in Eulerian and Lagrangian meshes to interact. Initially, the two solvers are entirely separate.
Lagrangian elements that lie within an Eulerian mesh do not affect the flow of the Eulerian material and
no forces are transferred from the Eulerian material back to the Lagrangian structure. The coupling
algorithm computes the interaction between the two sets of elements. It thus enables complex fluid-
structure interaction problems to be analyzed.
The first task in coupling the Eulerian and Lagrangian sections of a model is to create a surface on the
Lagrangian structure. This surface is used to transfer the forces between the two solver domains. The
surface acts as a boundary to the flow of material in the Eulerian mesh. At the same time, the stresses in
the Eulerian elements cause forces to act on the coupling surface, distorting the Lagrangian elements.
The method used for airbag simulation is based on General Coupling with adaptive Euler. SOL 700
computes the fluid flow and Coupling based on Dytran solver while the Contact and Fabric models are
co-simulated by LS-DYNA. Unlike other techniques such as ALE (Arbitrary Lagrange Euler) technique
125
CHAPTER 4
where the Eulerian mesh is fixed in space, or GBAG method where the gas flow is modeled by applying
a pre-determined pressure profile to inflate the bag, in General Coupling technique, the Eulerian mesh
will adapt itself to the Lagrangian fabric model as the airbag is inflated. In other words, initially when
the airbag is at folded stage, there is a small Eulerian domain encapsulating the lagrangian mesh. When
the airbag is inflated, the Eulerian mesh expands as the gas jet flows through the airbag compartments
and adapts itself to follow the airbag fabric. This technique is unique in MD R2 and is considered the
most accurate method to predict the complex airbag behavior such as Out-of-Position (OOP) simulation,
as required by FMVSS 208, where the occupant is leaning forward when the airbag is inflated and
Eulerian fluid deployment of multi-compartmented airbags, using multiple, fully automatic, adaptive
Euler domains. The following capabilities are available:
Analyze multiple compartments with the Eulerian fluid approach.
Simulate flow from one Eulerian fluid domain into another Eulerian fluid domain.
The individual Eulerian fluid domains are dynamic and adaptive. The user does not need to
mesh the Eulerian fluid domains, nor does he have to worry about the size, since the Eulerian
fluid domains will automatically follow the deploying air bag compartments.
Flow through both small and large holes is accurately calculated.
Inflator Models
Figure 4-1 Shape of side curtain air bag at start of simulation
Figure 4-2 Euler meshes for all 6 regions.

126
Figure 4-3 Shape of bag after 20 milliseconds
Inflator in AirBags
There are several methods available to define an inflator in air bag analyses. The most general and
extended inflator definitions are:
INFLATR1 - Standard inflator defined by mass flow rate and static temperature of a single inflowing gas.
INFLHYB1 - Hybrid inflator defined by mass flow rate and static temperature of multiple inflowing
gasses.
Figure 4-4 Airbag and Occupant Safety using SOL 700
The new entry AIRBAG in MD Nastran facilitates the airbag simulation. Basically all required entries
for airbags can be defined by using the AIRBAG entry.
Occupant Dummy Models
The occupant dummy models, also known as ATDs (Anthropomorphic Test Devices) were introduced in
the previous release of SOL 700 for those applications that airbag was not needed. Many applications
such as sled test and aircraft seat design, or armored vehicle design where the occupant behavior is
studied when a land mine is detonated do not require airbag simulation.
127
CHAPTER 4
SOL 700 supports occupant dummies that are readily available in LS-DYNA *key file format. These
include:
LS-DYNA Public Domain Dummies
5th percentile deformable female dummy
50th percentile deformable male dummy
5th percentile rigid female dummy
50th percentile rigid male dummy
ETA (Engineering Technology Associates) Calibrated Dummies
The calibrated dummies are similar to LS-DYNA dummies except that they are validated against a set
of standard tests. The calibration tests are conducted on all Hybrid III, SID and EUROSID models to
assure model fidelity.
5th percentile deformable female dummy
5th percentile rigid female dummy
EUROSID - 1 (Euro Side Impact Dummy)
US DOT SID
EEVC Upper Legform
FMVSS 201 Headform
Figure 4-5 Typical Occupant Dummies in VPG (courtesy of ETA)
There are many other ATD models available through ETA VPG (Virtual Proving Ground) pre-post
processor. In addition VPG supports the following barriers:

128
Barriers
FMVSS & ECE Side Impact Barriers
0-Degree Front Rigid Barrier
30deg Rigid Front Barrier
Front Offset Deformable Barrier
Rear Impact Barrier
Impactors
Rams
Pendulums
Head Forms
For more details, contact ETA in Troy, Michigan.
FTSS (First Technology Safety Systems) ATDs
The FTSS ATDs are high fidelity dummies and are available with additional licensing. The following
FTSS dummies are supported through SimX Crash or ETA VPG:
5% Female dummy
SID II
Hybrid III - 3% Child Dummy (W.I.P.)
Hybrid III - 6% Child Dummy (W.I.P.)
BIOSID (W.I.P.)
Pre- and Post-Processing
Even though MD Patran supports SOL 700, it does not have dummy positioner or airbag folder. Dummy
positioning can be done using pre-post processing tools such as MSC Software SimX Crash and ETA-
VPG, both of which support SOL 700. For more information for airbag and occupant analysis, please
refer to MD Nastran SOL 700 Users Guide and the MD Nastran Quick Reference Guide.
Time Domain NVH
This is a new methodology in MD Nastran SOL 700 that is used to compute the natural frequencies of
a structure similar to what is done in experimental modal identification using impact or other transient
testing. The user applies an impact, simulates driving a vehicle over a rough road or some other type of
transient loading to the structure such that it is likely to excite all the important modes. A standard SOL
700 analysis is run to output accelerations, velocities and/or displacements at selected grids using a
reasonably fine output delta time. After the nonlinear analysis finishes, a postprocessing operation is used
to transform the selected outputs from the time domain to the frequency domain using Fast Fourier
Transforms. Various criteria are available to select which peaks will be approximately defined as modes
129
CHAPTER 4
(eigenvectors). This approximation is reasonably accurate if the damping is small. The amplitudes of all
the selected degrees of freedom for each of the selected modal frequencies are normalized and become
the eigenvectors and saved on a file. A restart capability is available to change the criteria selections.
In addition, a user may pick modes from plots made from data saved during the first run and then
compute the eigenvectors associated with the chosen natural frequencies.
The advantages of Time Domain NVH include the ability to consider the material and structural
nonlinearity, damping and contact interface between various components. In other words, instead of
utilizing a linear implicit method, an explicit approach is under taken to first, predict the nonlinear
behavior of the structure, and then compute the NVH characteristics by FFT method. The double
precision version of LS-DYNA is used to ensure higher fidelity of the solution.
The disadvantage of time domain NVH technique is the excessive CPU time since an analysis may be
required to run for a few seconds, considered a long runtime for an explicit simulation, before an FFT is
performed to compute frequencies and mode shapes. However, this problem is becoming less significant
due to the advanced Distributed Memory Parallel (DMP) technology in SOL 700 and dramatic hardware
performance improvements.
Figure 4-6 Time Domain NVH of vehicle running on proving grounds (courtesy of ETA)
To define a time domain NVH simulation, refer to the new entries of TIMNVH, TIMSML and TIMNAT
in QRG and see examples in SOL 700 User Guide. A simple example is also included here.
Time Domain Data
Frequency Domain
FFT Process

130
Time Domain NVH Example
SOL 700 entries included:
SOL 700
TSTEPNL
PARAM,S700NVH1
TIMNVH
TIMNAT
Example: Time Domain NVH Example
Filename: timnvh.bdf, timnvh1.bdf
Description
This is an example of a virtual dynamic test. A plate was modeled and a impulse loading was applied at
one of corner points. Time histories were obtained at six points and they were translated by Fast Fourier
Transformation (FFT) method to a frequency domain to obtain modal frequencies and shapes.
131
CHAPTER 4
Model
The plate was modeled in this example. A fixed boundary condition was applied along one end. The
model has a total of 231 grid points and 200 quadratic elements. All shell elements were Belytschko-
Wong-Chiang formulation. The impact loading at the corner was modeled by defining the load time
history shown below. The simulation time is 1.024 seconds.
Time NVH scheme

132
Input
On node 231, the loading which is specified by FORCE and TABLE1 cards was applied. Using
TIMNVH, TIMNAT and PARAM, S700NVH1, dynamic properties of the plate were obtained. There
were three steps to acquire the dynamic properties as follows:
Step 1: Find modal properties roughly using the TIMNVH entry
Step 2: Check the obtained modal properties and select required natural frequencies
Step 3: Re-run with selected natural frequencies
Input File timnvh.bdf
SOL 700,NLTRAN path=3 stop=1
SOL 700 is a Executive Control statement similar to SOL 600. It activates an explicit nonlinear transient
analysis.
Case Control Section follows.
CEND
TITLE = MD Nastran job created on 12-Dec-06 at 11:21:25
LOADSET = 1
$ Direct Text Input for Global Case Control Data
SUBCASE 1
TITLE=This is a default subcase.
TSTEPNL = 1
SPC = 2
DLOAD = 2
SPCFORCES(SORT1,REAL)=ALL
Bulk Data Section follows.
BEGIN BULK
TSTEPNL 1 100 .01024 1 ADAPT 2 10
TIMNVH,1,,,1.0,500.,3,0.00005,2,+
+,0,3,1,0.015,0,3,13,.0030,+
+,11,21,116,126,221,231
TSTEPNL is a SOL 700 Bulk Data entry which describes the number of Time Steps (100) and Time
Increment (10.24 ms) of the simulation. End time is the product of the two entries. Notice here the Time
Increment is only for the first step. The actual number of Time Increments and the exact value of the Time
Steps is determined by Dytran LS-Dyna during the analysis. The time step is a function of the smallest
element dimension during the simulation.
TIMNVH is an entry for Time NVH analysis.
Details for TIMNVH entry:
TIMNVH, 1, , , 1.0, 500., 3, 0.00005, 2,+
The range of natural frequencies to obtain is from 1.0 Hz to 500 Hz and translational DOF of z-direction
is only considered (3). The sampling rate is 0.00005 seconds. The peaking criterion is two, which means
a peak is selected if the number of increasing and decreasing amplitude around a peak is over 2.
133
CHAPTER 4
+, 0, 3, 1, 0.015, 0, 3, 13, .0030,+
Acceleration is selected for the response (0) and translational eigenvectors are only requested as ASCII
format (3). Eigenvalues are normalized by 1.0 (1) and 0.015 is selected as CLOSE value which means if
there are two modes which distance is smaller than 0.015 Hz, it is assumed to be the same mode. ACII
file format of natural frequencies and eigenvalues are asked (0) and translational time histories of z-
direction are requested (3). Frequency-amplitude data of z-direction are asked (13) and a peak which
amplitude is less than 0.0030the maximum amplitude is ignored (.0030)
+,11,21,116,126,221,231
The grid points 11, 21, 116, 126, 221 and 231 are only considered for Time NVH analysis.
The Bulk Data entry that defines grid points and elements.
CQUAD4 1 1 1 2 23 22
..
$ Nodes of the Entire Model
GRID 1 0. 0. 0.
..
_______________________________________________________________
127 148 169 190 211
Bulk data entry that defines boundary conditions along the center line.
Bulk data entry that defines material properties.
$ Referenced Material Records
$ Material Record : steel
$ Description of Material : Date: 12-Dec-06 Time: 11:09:13
MAT1 is an isotropic-elastic material entry of SOL 700 Bulk Data entry.
Bulk data entry that defines properties for shell elements with .001 thickness.
PSHELL 1 1 .001 1 1
__________________________________________________________
DLOAD 2 1. 1. 4
$ Nodal Forces of Load Set : impact
Bulk data entry that defines tables.
$ Dynamic Load Table : hammer
TABLED1 1
-10. 0. 0. 0. .001 1. .002 0.
10. 0. ENDT

134
Input Tile timnvh1.bdf:
This input is for the refinement of the selection of modal frequencies and mode shapes.
Only different part is shown:
PARAM,S700NVH1,1
TSTEPNL 1 100 .01024 1 ADAPT 2 10
TIMNVH,1,,,1.0,500.,3,0.00005,-2,+
+,0,3,1,0.015,0,3,13,.0030,+
PARAM, S700NVH is for the re-run of Time NVH analysis. Using 1 as option Time NVH analysis is
carried out without dytran-lsdyna re-running.
The PEAK value (original: 2) in TIMNVH entry is changed to -2 to use TIMNAT entry.
The TIMNAT entry is for the control of the natural frequency selection. In this job, 20.5, 88, 129, 285,
360 Hz are selected to get the results.
Results
There are three types of new results file from Time NVH analysis.
1. mode.out: the natural frequencies and eigenvalues selected are restored.
2. ampl-freq-00000116-3.txt: amplitude-frequency output of DOF =3 at grid point 116.
3. time-hist-00000116-3.txt: time history output of DOF =3 at grid point 116.
135
CHAPTER 4
From the ampl-freq-*** files, the frequency-amplitude plots are obtained.
1.00E-02
1.00E-01
1.00E+00
1.00E+01
1.00E+02
1.00E+03
1.00E+04
0 100 200 300 400 500
Frequency (Hz)
A
m
p
l
i
t
u
d
e

(
a
c
c
e
l
e
r
a
t
i
o
n
)
node 11
node 21
node 116
node 126
node 221
node 231
1st mode = 20.5
2nd mode = 88.
3rd mode = 129.
4th mode = 285.

136
In mode.out file,
Comparison of natural frequencies between SOL 103 and SOL 700 (Hz).
Mode SOL 103 SOL 700 Diff(%)
1 2.0508E+01 2.0864E+01 1.74
2 8.7894E+01 8.8639E+01 0.85
3 1.2891E+02 1.2966E+02 0.58
4 2.8516E+02 2.8771E+02 0.89
5 3.6036E+02 3.6269E+02 0.65
MODES 1 18
EIGV 1 2.050821E+01
11-9.13892833E-06 3.86241316E-06 3.19331321E-01
21 2.14690732E-05-1.14613790E-05 9.95118635E-01
116-9.93770531E-06 2.13863912E-06 3.30875181E-01
126 7.27358928E-06-2.94809676E-06 1.00000000E+00
221-1.23613298E-05-3.06943202E-06 3.20993777E-01
231 1.35196599E-05 6.87379272E-06 9.99026322E-01
EIGV 2 8.789234E+01
11-5.69607103E-06 1.65922185E-06-6.09298304E-01
21 5.33656071E-06-7.67633319E-06-9.90453563E-01
116-3.47585317E-06 8.00428604E-07 1.97687095E-02
126 5.34868995E-06-3.96739670E-06-7.95414322E-03
221-4.07797686E-06-2.98546031E-06 5.90684541E-01
231 3.21720171E-06 7.30456344E-06 1.00000000E+00
Total number of modes
Natural Frequency
Node number Node number Node number Node number
1st mode
137
CHAPTER 4
Comparison of mode shapes between SOL 103 and SOL 700.
Mode SOL 103 SOL 700
1
2

138
3
4
139
CHAPTER 4
Prestressing (Implicit to Explicit Sequential Simulation)
Many applications require a prestress analysis prior to transient analysis. For example in bird strike and
blade out analysis, the blades of a running jet engine have residual stresses due to high rotational
velocities. SOL 700 uses an implicit solver for prestress analysis to dramatically reduce the CPU time.
Prestress calculations are performed by LS-DYNA implicit double precision version where initial state
of the model is written into a file that can only be used for subsequent SOL 700 transient run. This is an
automated process and there is no need for dynamic relaxation to eliminate the high frequency
oscillations.
Please refer to the new PRESTRS entry in the MD Nastran Quick Reference Guide and the SOL 700
Users Guide for more details.
5
6 points sampling
12 points sampling

140
Additional Capabilities of MD Nastran Explicit Nonlinear -
SOL 700
MD R2 has many additional material models and elements to support a variety of different applications.
The following are the new capabilities that are added in this release:
New Material Models
Force Limited Material MATD029
Fabric Material MATD034
Nonlinear Orthotropic Material MATD040
Concrete Damage Material MATD072
Concrete Damage Material REL3 MATD72R
Plasticity Polymer MATD089
Simplified Johnson Cook Material MATD098
Simplified Johnson Cook Orthotropic Damage Material MATD099
Finite Elastic Strain Plasticity Material MATD112
Layered Linear Plasticity Material MATD114
Modified Piecewise Linear Plasticity MATD123
Rate Sensitive Composite Fabric Material MATD158
General Spring Discrete Spring Material MATD196
Elements & Boundary Conditions
Breakable Join BJOIN
Geometric Rigid Wall WALLGEO
Spotweld Beam Property PBSPOT
Spotweld Material MATDSW1 to MATDSW5
Connection Spotweld CONSPOT
Properties
Shell Property PSHELLD
Mass Damping per Property DAMPMAS
Stiffness Damping per Property DAMPSTF
Airbags
AIRBAG Defines an air bag
PARAM, UGASC - Defines a value for the universal gas constant.
PARAM, SBOLTZ - Defines a value for the Stephan-Boltzmann constant
141
CHAPTER 4
PARAM, DYDEFAUL - : Controls the default setting of the simulation
Hybrid inflator gas fraction definition INFLFRC
Gamma Law Gas Equation of State EOSGAM
Tabulated Equation of State with Compaction - EOSTABC
Tabulated Equation of State EOSTAB
Grid Point in Air Bag Reference Geometry GRIA
Prestress
ISTRSBE - Initialize stresses and plastic strains in the Hughes-Liu beam elements
ISTRSSH - Initialize stresses, history variables and the effective plastic strain for shell elements
ISTRSTS - Initialize stresses, history variables and the effective plastic strain for thick shell
elements.
ISTRSSO - Initialize stresses and plastic strains for solid elements
PRESTRS - Perform prestress run to calculate an initially stressed model and write out the initial
state to a file that can be used for a subsequent explicit SOL 700 run
New Contact Features
IGLUE, JGLUE - Glued contact features
TIEDNSCO: Tied Nodes to Surface Constrained Offset
TIEDESCO: Tied Shell Edge to Surface Constrained Offset
TIEDSSCO: Tied Surface to Surface Constrained Offset
SPOTWELD: Spotweld Contact
SPOTWTOR: Spotweld with Torsion Contact
EDGE: Single Edge Contact (Master Body not allowed)
FTRANSC: Force Transducer Constraint
FTRANSP: Force Transducer Penalty
Parameters
DYPARAM, LSDYNA, SOLID, [option], <value> - Define control parameters that can improve
the accuracy of the calculation
DYPARAM, LSDYNA, ACCURACY, [option], <value> - Define control parameters that can
improve the accuracy of the calculation
DYPARAM, LSDYNA, DATABASE, FORMAT, <value> - Define the type of result output files
DYPARAM, LSDYNA, DATABASE, [option], <value> - Define control parameters that
control output generation of the calculation

142
Miscellaneous
ACC usage is no longer recommended and will be removed from the code in a future version.
Use ACCMETR.
MATD100 usage is no longer recommended and will be removed from the code in a future
version. Use MATDSW1-5 in combination with PBSPOT.
Ch. 5: NVH & Acoustics MD R2 Nastran Release Guide

5
NVH & Acoustics
Frequency Response Function (FRF) and FRF Based Assembly

(FBA) Feature in MD R2 Nastran
Exterior Acoustics
Rigid Porous Absorber


144
Frequency Response Function (FRF) and FRF Based
Assembly (FBA) Feature in MD R2 Nastran
The FRF / FBA (Frequency Response Function / FRF Based Assembly) capability is a new feature
available in MD R2 Nastran. The following sections describe this feature and explain its usage.
Introduction
Frequency Response Functions (FRFs) can be used to represent components. The FRFs of such
components can then be combined to yield the FRFs of assemblies of such components. Such an FRF
based assembly (FBA) process represents an alternative to component mode synthesis (or modal
substructuring). This method also lends itself to tracking load paths or energy flows through a structure.
This approach may be regarded as an alternative to mode participation studies. This latter feature,
commonly referred as Transfer Path Analysis (TPA), is not currently available, but will be incorporated
in a post-MD R2 Nastran release.
FRF Concept
Using the usual symbols, the equation of motion for frequency response analysis is given by:
(5-1)
where is the dynamic stiffness matrix of the system.
Solving the above equation for the response displacements , we obtain
(5-2)
The matrix in the above equation is the receptance matrix of the system. It is also commonly referred
to as the FRF (Frequency Response Function) of the system. It should be noted that the FRF of a system
is a function of the forcing frequency . Thus there is a different FRF for the system for each forcing
frequency.
The rows of the FRF matrix represent response DOFs while its columns represent excitation DOFs. Any
term of the FRF matrix represents the response displacement at the i-th DOF due to a unit load applied
to the j-th DOF. Thus, the rows in , the j-th column of the FRF, represent the displacements at all
response DOFs due to a unit load applied to the j-th DOF. Similarly, the columns in the i-th row of the
FRF represent the response displacements at the i-th DOF due to unit loads applied to all excitation
DOFs.
In practice, the FRF matrix is computed by solving Eq. (5-2) for a series of unit loads. Thus, consider the
following equation which represents Eq. (5-2) for a series of unit loads on the right hand side:
(5-3)
2
M [ ] i B [ ] K [ ] H H ( ) u } Z [ ] u } P } Z Z
Z [ ]
u }
u } Z [ ]
1
P } H [ ] P } Z Z
H [ ]
H
i j
H { }
j
U [ ] Z [ ]
1
I [ ] Z
145
CHAPTER 5
NVH & Acoustics
The solution of the above equation for the matrix of displacements yields the desired FRF matrix
.
The matrix of displacements resulting from the solution of Eq. (5-3), which is equivalent to the
receptance or FRF matrix of Eq. (5-2), is square because Eq. (5-3) assumes that the responses are
computed at all DOFs of the system and also that unit loads are applied at all DOFs. However, this is
rarely the case in practice. In most practical cases, responses are computed at only a subset of the total
points in a configuration and loads are also applied to only a small subset of the total DOFs in the system.
Therefore, in practice, the actual FRF matrix to be considered for analysis purposes is really a subset (or
smaller partition) of the matrix given by Eq. (5-3). In general, this FRF matrix, which is what is
actually employed in FRF analysis, is rectangular and can be regarded as an (r x e) matrix where r, the
number of rows, is the number of actual response DOFs and e, the number of columns, is the number of
actual excitation DOFs.
FBA Concept
It is possible to obtain the FRFs of an assembly of components from the FRFs of the individual
components comprising the assembly. This process is referred to as FRF Based Assembly (FBA).
The FRFs of an assembly of components is computed in Nastran by coupling the FRFs of its constituent
components and enforcing conditions of displacement compatibility and force equilibrium at the
connected DOFs. The final result of the FBA process will be a set of FRF matrices for the assembly, with
each such matrix corresponding to a forcing frequency. The number of rows of each such matrix will be
equal to the total number of response DOFs of the assembly and the number of columns will be equal to
the total number of excitation DOFs of the assembly.
Using the FRFs resulting from the FBA process, the program recovers the FRFs for the individual FRF
components comprising the FRF assembly. These results will reveal clearly how excitations on one
component affect the responses at other components of the assembly.
Usage of the FRF/FBA Feature
The FRF/FBA capability is available in both SOL 108 (Direct Frequency Response) and SOL 111
(Modal Frequency Response). The feature involves the use of a new Case Control command (FRF) and
three new Bulk Data entries (FRFCOMP, FRFXIT and FRFXIT1). Descriptions of these are all given in
the MD Nastran Quick Reference Guide.
The generation of FRFs for a component and their use in a subsequent FBA process using the FRF Case
Control command involves running a standard SOL 108 or SOL 111 job, with the following additional
data:
U [ ]
H [ ]
U [ ]
H [ ]
U [ ]

146
1. The DOFs where unit loads are to be applied must be specified either indirectly via the DLOAD
Case Control request and/or directly via the FRFXIT/FRFXIT1 Bulk Data entries. The DLOAD
Case Control points to appropriate Bulk Data loading entries. All DOFs where there are non-zero
load values will have unit loads applied to them. The FRFXIT entry permits specification of unit
load for a single DOF with a label. The FRFXIT1 entry permits specification of unit loads at
multiple DOFs.
2. There is no requirement that unit loading data as above be defined for every component for which
FRFs are generated since some components in a configuration may not have any loads applied on
them.
3. If the FRFs of a component are to be employed in a subsequent FBA process, the user must
specify the points where this component will be connected to other components in the FBA
process. This is done via the CONNPTS keyword in the FRF command which references a set of
these connection points. Also, in this case, the FRFs and other related information for the
component are saved on the medium (database or OP2 file) specified in the FRF command and
the assembly punch (.asm) file will be generated and saved with a single FRFCOMP Bulk Data
entry in it for subsequent use in the FBA process.
4. Regardless of whether an FRF component has unit loads explicitly specified for it, as in 1., or not,
as in 2., the program will internally apply unit loads automatically at all DOFs for all connection
points comprising the set referenced by the CONNPTS keyword of the FRF command. This
ensures that correct results are obtained from subsequent FBA processes.
5. The specific points for which FRFs are computed in a FRF generation run comprise the following:
a. All points specified via DISP, VELO and ACCE requests
b. All points associated with elements for which STRESS/FORCE requests are specified
c. All points where unit loads are applied (as per the scheme indicated in 1.)
d. All points comprising the set referenced by the CONNPTS keyword in the FRF command
6. Component ID of 0 is assigned to the assembled FRF configuration resulting from the FBA
process.
7. The FBA process generates a connection information table in the .f06 file indicating the
relationship between the internal point IDs of the assembled FRF configuration (component 0)
and the external point IDs of the associated FRF components.
8. The normal output from the FRF generation and FBA process runs, implied by the default of
XITOUT = USER, gives the results only for unit loads explicitly specified by the user as indicated
in 1. above. If output for the unit loads internally applied automatically by the program as
indicated in 4. is also desired, then XITOUT = ALL must be specified in the FRF command to
obtain such expanded output.
147
CHAPTER 5
NVH & Acoustics
9. The normal output from an FBA process run, implied by the default of ASMOUT = COMP, gives
the results for the individual FRF components that comprise the assembled FRF configuration. If
output is also desired for the assembled configuration as a separate entity (component 0 as
indicated in 6.), then ASMOUT = ALL must be specified in the FRF command to obtain such
expanded output. However, in this case, the output for component 0 will be limited to
displacements, velocities and accelerations and these will be output in terms of the internal point
IDs mentioned in 7.
10. It is currently assumed that the FRFs of all FRF components are generated at the same forcing
frequencies and that these are also the forcing frequencies at which the FBA process is performed.
As a result, the FBA process as currently implemented derives these forcing frequencies from the
saved data of the first of the FRF components being assembled and employs them for the FBA
process. This restriction on forcing frequencies will be removed in a later release.
The usage of the FRF/FBA feature as described can involve one of four scenarios as follows. All of these
scenarios are illustrated by job setup examples given under the description of the FRF Case Control
command in the MD Nastran Quick Reference Guide.
1. Generate the FRFs for a one shot configuration with no subsequent FBA process.
2. Generate the FRFs for a component with the purpose of employing them in a subsequent FBA
process.
3. Compute the FRFs of an assembly of components by the FBA process by using the FRFs of
components that have been previously generated.
4. Generate the FRFs for a component and, in the same execution, compute the FRFs of an assembly
of components via the FBA process by combining the FRFs of this component with those of other
components whose FRFs have been previously generated.
Limitations of the FBA Process Implementation
The current implementation of the FBA process has several limitations as indicated below. These
limitations will be removed in future releases of MD Nastran.
The FBA process connects a particular point in an FRF component with points in other FRF
components only if such points are coincident (that is, their basic coordinates are the same).
There is no provision to allow for connection of non-coincident points.
All connections between and among FRF components are considered as rigid. There is no
provision to allow for non-rigid (that is, spring or bushing) connections.
The FBA process has no provision for user specified loads.
The FBA process assumes that the FRFs of all FRF components are generated at the same
forcing frequencies and that these are also the forcing frequencies at which the FBA process is
performed.

148
Exterior Acoustics
Introduction
Nastran has been used to analyze interior acoustic problems for a long time. With interior acoustic
problems, the acoustic domain considered is bounded. A typical example is the determination of the
sound pressure inside a car.
With exterior acoustic problems, the acoustic domain is unbounded. The analysis studies the sound
pressure in the vicinity of the vibrating body or far away from the body. A further important result is the
acoustic power radiated from the vibrating structure. A typical example is the determination of the
radiated acoustic power of an engine.
In MD R2 Nastran the infinite elements from MSC Actran have been integrated to allow the analysis of
exterior acoustic problems. The infinite elements are attached to the boundary of the acoustic finite
element mesh to provide the correct non-reflecting boundary condition.
Apart from standard acoustic results within the finite region it is possible to compute acoustic results at
points within the infinite elements. These points, the so-called field points, may be connected by elements
to form a field point mesh. If a field point mesh is defined, also the acoustic power through this field point
mesh is computed.
Benefits
The integration of the proven and tested infinite elements from MSC Actran into MD Nastran largely
facilitates the analysis of exterior acoustic problems. It is no longer necessary to transfer data between
different programs but the analysis can be done completely within MD Nastran.
Input
New Bulk Data entries, CACINF3 and CACINF4, are used to define the connectivity of the infinite
elements. The properties of the infinite elements are defined on PACINF Bulk Data entries.
Field points and field point meshes are defined in separate sections of the bulk data file. These sections
must follow the main bulk data section.
Output of radiated power from the wetted surface and intensities on the wetted surface is controlled by
Case Control commands ACPOWER and INTENSITY. Output of field point mesh results is controlled
by Case Control command ACFPMRESULT.
Definition of Infinite Elements
The geometry of an infinite element is defined by the geometry of its base and the location of the pole,
see Figure 5-1. The base of the infinite element is that part that is in contact with the finite domain. The
geometry of the base is defined by its connectivity and the location of the corner grids. In order to avoid
149
CHAPTER 5
NVH & Acoustics
overlapping of the infinite elements, the surface they are attached to must be convex. However, it is not
necessary that this surface is smooth.
Figure 5-1 Geometry of Infinite Element
Within an infinite element, the acoustic pressure is expanded into a power series of (1/r) where r is the
distance from the pole.The radial interpolation order is the number of terms kept in this series.
The connectivity of the base is defined on the CACINF3 and CACINF4 Bulk Data entries. The
orientation should be such that the normal vector on the base points into radial direction. However, MD
Nastran will automatically change the orientation if the normal vector does not point away from the pole.
The location of the pole as well as the radial interpolation order are defined on PACINF Bulk Data
entries.

150
Defines an acoustic conjugate infinite element with triangular base
Format:
Defines an acoustic conjugate infinite element with quadrilateral base
Format:
CACINF3
Acoustic Conjugate Infinite Element Base Connection
1 2 3 4 5 6 7 8 9 10
CACINF3 EID PID G1 G2 G3
Field Contents
EID Element Identification Number (Integer > 0)
PID Property Identification Number of a PACINF entry (Integer > 0)
Gi Grid Point Identification Numbers of Element Base Connection Points (Integer > 0)
CACINF4
Acoustic Conjugate Infinite Element Base Connection
1 2 3 4 5 6 7 8 9 10
CACINF4 EID PID G1 G2 G3 G4
Field Contents
EID Element Identification Number (Integer > 0)
PID Property Identification Number of a PACINF entry (Integer > 0)
Gi Grid Point Identification Numbers of Element Base Connection Points (Integer > 0)
151
CHAPTER 5
NVH & Acoustics
Defines the properties of acoustic conjugate infinite elements.
Format:
Definition of Field Point Meshes
Acoustic field point meshes are defined in separate sections of the bulk data file. These sections follow
the main bulk data section. Each of the sections begins with
BEGIN BULK AFPM = afpmid
or
BEGIN AFPM = afpmid
where afpmid is the acoustic field point mesh identifier (integer > 0). Acoustic field points are defined
using the standard GRID Bulk Data entry. Because all grid points defined in this section of the bulk data
file are acoustic field points, it is not necessary to place a -1 into field 7.
The grid points can be connected by any type of elements. However, only CTRIA3 and CQUAD4
elements define a field point mesh that is used to compute normal components of the acoustic intensity
and the power through the field point mesh. Legal property identifiers have to be specified on the
CTRIA3 and CQUAD4 Bulk Data entries. However, the referenced PSHELL entries need not be
defined.
If output to an .op2 file is requested, parameter POST has to be defined within the section of the acoustic
field point mesh. Output of different field point meshes can be sent to different .op2 files using parameter
OUNIT2 within the sections of the acoustic field point meshes.
Example
BEGIN AFPM=100
$
PARAM, POST, -1
$
PACINF
Acoustic Conjugate Infinite Element Property
1 2 3 4 5 6 7 8 9 10
PACINF PID MID RIO X1 X2 X3
Field Contents
PID Property Identification Number of PACINF entry (Integer > 0)
MID Material Identification Number of a MAT10 entry (Integer > 0)
XP, YP, ZP Coordinates of the Pole of the Infinite Elements (in the Basic Coordinate System)

152
$ Isolated Field Points along a Line
$
GRID 1001 2 5. 0. 0.
GRID 1002 2 6. 0. 0.
GRID 1003 2 7. 0. 0.
GRID 1004 2 8. 0. 0.
GRID 1005 2 9. 0. 0.
GRID 1006 2 10. 0. 0.
CORD2R, 2,, 0., 0., 0., -.5, -.5, .707107
, .5, .5, .707107
$
BEGIN AFPM = 200
$
PARAM, POST, -1
$
GRID, 1,, -1., -1., 2.
GRID, 2,, 0., -1., 2.
GRID, 3,, 1., -1., 2.
GRID, 4,, -1., 0., 2.
GRID, 5,, 0., 0., 2.
GRID, 6,, 1., 0., 2.
GRID, 7,, -1., 1., 2.
GRID, 8,, 0., 1., 2.
GRID, 9,, 1., 1., 2.
$
GRID, 11,, -1., -1., 2.
GRID, 12,, 0., -1., 3.
GRID, 13,, 1., -1., 2.
GRID, 14,, -1., 0., 3.
GRID, 16,, 1., 0., 3.
GRID, 17,, -1., 1., 2.
GRID, 18,, 0., 1., 3.
GRID, 19,, 1., 1., 2.
$
CQUAD4, 1, 1, 1, 2, 5, 4
CQUAD4, 2, 1, 2, 3, 6, 5
CQUAD4, 3, 1, 4, 5, 8, 7
CQUAD4, 4, 1, 5, 6, 9, 8
$
CQUAD4, 5, 1, 12, 16, 18, 14
CTRIA3, 6, 1, 11, 12, 14
CTRIA3, 7, 1, 12, 13, 16
CTRIA3, 8, 1, 16, 19, 18
CTRIA3, 9, 1, 14, 18, 17
$
ENDDATA
153
CHAPTER 5
NVH & Acoustics
Case Control Commands
The following new case control commands control postprocessing which is related to exterior acoustics.
Requests output of the power radiated from the wetted surface.
Format:
ACPOWER Acoustic Power Output Request
Describer Meaning
SORT1 Output will be presented as tabular listing of panels for each excitation frequency
SORT2 Output will be presented as tabular listing of excitation frequencies for each panel
(Default)
PRINT The printer will be the output medium (Default).
PUNCH The punch file will be the output medium.
PLOT Results are generated but not output.
CSV Results will be written to a .csv file.
unit Unit of the .csv file as used on the ASSIGN statement
ALL Radiated power will be processed for the wetted surface and all panels.
n Set identification of a previously defined set of panels. Radiated power will be
processed for the wetted surface and all panels in the referenced set.
NONE Radiated power will not be processed.
ACPOWER
SORT1
SORT2
PRINT,PUNCH
PLOT
CSV unit Z [ ] , ,
'

| |
ALL
n
NONE
)

`

Z

154
Requests output of acoustic intensity on wetted surface.
Format:
INTENSITY
Acoustic Intensity Output Request
Describer Meaning
SORT1 Output will be presented as tabular listing of grid points for each excitation
frequency (Default)
SORT2 Output will be presented as tabular listing of excitation frequencies for each grid
point
ALL Intensities will be computed for all grid points of the wetted surface.
n Set identification of a previously defined set of grid points. Intensities will be
computed for the grid points in this set only.
NONE Acoustic intensities will not be processed.
INTENSITY
SORT1
SORT2
PRINT,PUNCH
PLOT
,
'

| |
ALL
n
NONE
)

`

Z
155
CHAPTER 5
NVH & Acoustics
Requests output of field point mesh results.
Format:
Output
In addition to the standard acoustic results, the following results may be requested:
ACFPMRESULT
Acoustic Field Point Mesh Results Output Request
Describer Meaning
SORT1 Output will be presented as tabular listing of grid points for each excitation
frequency (Default)
SORT2 Output will be presented as tabular listing of excitation frequencies for each grid
point
VELOCITY Requests output of particle velocities (Default: NO)
REAL or IMAG Requests rectangular format (real and imaginary) of complex output. Use of either
REAL of IMAG yields the same output.
PHASE Requests polar format (magnitude and phase) of complex output. Phase output is in
degrees.
POWER Requests output of power through field point mesh (Default: YES)
ALL Results of all field point meshes will be processed
n Set identification of a previously defined set of field point mesh identifiers. Results
will be processed for the field point meshes in this set only.
NONE Field point mesh results will not be processed.
ACFPMRESULT
SORT1
SORT2
PRINT,PUNCH
PLOT
VELOCITY
YES
NO
)
`

Z , ,
'
|
REAL or IMAG
PHASE
POWER
YES
NO )
`

Z
|
,
ALL
n
NONE
)

`

Z

156
Acoustic energy radiated from the wetted surface or from panels
Normal component of the acoustic intensity at grid points of the wetted surface
Acoustic pressure and intensity at the field points
Acoustic velocity at the field points
Acoustic energy flowing through the field point mesh
Field point mesh results can be written to the .f06 file or the .op2 file. The .xdb file is not yet supported.
Guidelines
1. The surface the infinite elements are attached to must be convex but it need not be smooth. Infinite
elements have to be connected to acoustic grid points. Thus it is necessary to model the vicinity
of the vibrating structure with acoustic finite elements until a convex surface is reached.
2. The radial interpolation order required depends on the directivity of the pressure field. Usually, a
higher order is needed for higher frequencies.
3. It is recommended to study the sensitivity of the results with respect to the radial interpolation
order by repeating the analysis with a different radial interpolation order.
4. Infinite elements are supported in direct and modal frequency response analysis (SOL 108 and
SOL 111). Experience shows that a large number of fluid modes is needed to get converged
results. Thus, modal reduction of the fluid is not recommended. However, modal reduction can
be applied to the structure.
5. Efficiency can be increased if the iterative solver is used. It is recommended to use the Jacobi
preconditioner with an epsilon of 10
-4
.
Limitations
1. Infinite elements are supported in a frequency response analysis only (SOL 108 and SOL 111).
2. In this prerelease, case control commands ACPOWER and INTENSITY are not yet supported.
Example
In this example, the sound transmission through an elastic plate embedded in an infinite rigid wall is
analyzed. Figure 5-2 shows the geometry of the plate and of the fluid region meshed with finite elements.
The finite element mesh of the structure and of one quarter of the fluid can be seen in Figure 5-3.
157
CHAPTER 5
NVH & Acoustics
Figure 5-2 Geometry
Infinite Elements are attached to the top and the four lateral faces but not to the bottom. The pole of all
infinite elements is at the center of the plate. The finite element mesh of the structure together with one
quarter of the infinite element mesh is shown in Figure 5-4.
The field point mesh is a cube which completely encloses the fluid mesh. Thus, the power through this
field point mesh equals the total power radiated from the plate. Figure 5-5 shows the fluid mesh and one
quarter of the field point mesh.

158
Figure 5-3 Structure and Quarter of Fluid Mesh
Figure 5-4 Structure and Quarter of Infinite Element Mesh
159
CHAPTER 5
NVH & Acoustics
Figure 5-5 Fluid Mesh and Quarter of Field Point Mesh
The excitation is a uniform pressure applied to the plate. The acoustic pressure in the finite element mesh
and acoustic results at some discrete field points located along the z-axis and at the field points of the
field point mesh are computed. Acoustic results at the field points include the acoustic pressure and the
acoustic intensities.
Input File
SOL 108
CEND
TITLE = Vibrating Plate Example
SUBTITLE = Direct Frequency Response
$
ECHO = SORT(EXCEPT, GRID, CHEXA, CQUAD4, CACINF4)
$
DLOAD = 10
FREQ = 20
SMETHOD = 30
SPC = 1
$
DISP(PLOT) = ALL
ACFPMRESULT(PHAS) = ALL
$
BEGIN BULK
$
PARAM, POST, -1
$

160
ACMODL, IDENT
$
RLOAD1, 10, 200,,, 300
PLOAD2, 200, 1., 1, THRU, 150
TABLED1, 300
, 0., 1., 1000., 1., ENDT
$
FREQ, 20, 10.
$
ITER, 30
PRECOND = J,
ITSEPS = 1.0E-4
$
$ Fluid
$
GRDSET,,,,,, -1
INCLUDE 'fluid1.bdf'
$
$ Structure
$
INCLUDE 'structure.bdf'
$
BEGIN AFPM=100
$
$ Isolated Field Points
GRID, 10001,, 0., 0., 3.5
GRID, 10002,, 0., 0., 5.
GRID, 10003,, 0., 0., 10.
$
BEGIN AFPM=200
$
$ Field Point Mesh
INCLUDE 'fpm.bdf'
$
ENDDATA
Excerpt of fluid1.bdf
$ Exterior Acoustics - Vibrating Plate Example
$ Fluid Model of Domain 1: Finite and Infinite Elements
$
$ -------------------------------------------------------------------
$
$ PID MID RIO XP YP ZP
PACINF 10 20 10 0. 0. 0.
PSOLID 2 20 PFLUID
MAT10 20 1.21 340.
$
CHEXA 1001 2 1001 1002 1018 1017 1177 1178
1194 1193
CHEXA 1002 2 1002 1003 1019 1018 1178 1179
1195 1194

161
CHAPTER 5
NVH & Acoustics
$ Infinite Elements
$
CACINF4 13901 10 3641 3642 3658 3657
CACINF4 13902 10 3642 3643 3659 3658
CACINF4 13903 10 3643 3644 3660 3659
CACINF4 13904 10 3644 3645 3661 3660
CACINF4 13905 10 3645 3646 3662 3661
CACINF4 13906 10 3646 3647 3663 3662

Results
Figure 5-6 shows the acoustic pressure in the Finite Element mesh and Figure 5-7 shows some results
printed to the .f06.
Figure 5-6 Acoustic Pressure in the Finite Element Mesh

162
Figure 5-7 Results in .f06
FREQUENCY = 1.000000E+01 ACOUSTIC FIELD POINT MESH = 100
A C O U S T I C F I E L D P O I N T M E S H R E S U L T S

ACOUSTIC PRESSURE INTENSITY COMP. INTENSITY INTENSITY INTENSITY
POINT ID. MAGNITUDE PHASE NORMAL TO FPM X Y Z
10001 5.867592E-03 3.224626E+02 0.0 3.901852E-16 1.102822E-09 4.101243E-08
10002 4.148942E-03 3.068220E+02 0.0 1.168128E-16 3.946748E-10 2.071128E-08
10003 2.089879E-03 2.542063E+02 0.0 1.290767E-17 5.147245E-11 5.294778E-09
1 VIBRATING PLATE EXAMPLE DECEMBER 6, 2005 MSC.NASTRAN 12/ 5/05
PAGE 16
DIRECT FREQUENCY RESPONSE
0
*** USER INFORMATION MESSAGE 3119 (AFPINI)
DATA RECOVERY OF ACOUSTIC FIELD POINT MESH 200 INITIATED
1 VIBRATING PLATE EXAMPLE DECEMBER 6, 2005 MSC.NASTRAN 12/ 5/05
PAGE 17
DIRECT FREQUENCY RESPONSE
0
FREQUENCY = 1.000000E+01 ACOUSTIC FIELD POINT MESH
= 200
A C O U S T I C F I E L D P O I N T M E S H R E S U L T S

ACOUSTIC PRESSURE INTENSITY COMP. INTENSITY INTENSITY INTENSITY
POINT ID. MAGNITUDE PHASE NORMAL TO FPM X Y Z
46567 2.959861E-03 2.856447E+02 7.442431E-09 -7.647572E-09 7.442431E-09 1.545178E-10
46568 2.930327E-03 2.848879E+02 7.258169E-09 -7.408595E-09 7.258169E-09 1.340973E-09
46569 2.846820E-03 2.826632E+02 6.691689E-09 -6.762688E-09 6.691689E-09 2.608988E-09
46570 2.722373E-03 2.790949E+02 5.876647E-09 -5.877888E-09 5.876647E-09 3.492858E-09
46571 2.572942E-03 2.743544E+02 4.970530E-09 -4.929979E-09 4.970530E-09 3.970974E-09
46572 2.412938E-03 2.686276E+02 4.189022E-09 -4.042182E-09 4.189022E-09 4.023463E-09
46573 3.274509E-03 2.928127E+02 1.022698E-08 -8.121219E-09 1.022698E-08 2.107736E-10
46574 3.234508E-03 2.919741E+02 9.893179E-09 -7.820792E-09 9.893179E-09 1.809509E-09
163
CHAPTER 5
NVH & Acoustics
Rigid Porous Absorber
Introduction
A new capability to model basic rigid skeleton porous absorber characteristics in acoustic response
analysis is now available. The capability allows some types of absorbent material to be modelled, such
as vehicle seat structures or lining materials which exhibit stiff carcasses. The absorber material is
considered using an equivalent fluid analogy and so is modelled in the same manner as a typical fluid,
using solid CHEXA, CPENTA or CTETRA elements, the GRID points for which have their CD field set
to -1. The porous absorber elements reference a PSOLID property entry with field 8 set to PFLUID in
the usual way, with field 3 of the PSOLID entry referencing a MAT10 entry which has been modified
with an additional field 7 dedicated to porous absorber materials.
Porous Materials
If a material is not completely solid, but contains voids or air pockets, then it is said to be porous. There
are a great many materials which exhibit porosity, the term given to the degree of openness of the
material, including materials generally considered solid like brick or stone. If the voids in the
substance are large enough, they may form an interconnected maze of passage ways allowing air (or any
other fluid) to pass through the material. However, depending on the degree of convolution in the
passage ways (known as tortuosity), the air will encounter some resistance as it passes through the
substance, requiring pressure to be exerted to push the air against the resistance. Sound waves striking
the material do not cause air to flow through the material, but they do cause local perturbations that exert
pressure and cause the air to move in the vicinity of the material; the oscillating movement of air caused
by the sound waves encounters resistance (called impedance) which uses up some of the sound energy
and damps the level of sound. This energy is eventually dissipated as heat.
If the porous material is enclosed in a frame which is considered as rigid, as will be the case for example
for a porous medium which has a high skeleton density or very large elastic modulus or weak fluid-
structure coupling, the porous material can be considered as an equivalent fluid with both density and
bulk modulus being complex frequency dependent parameters. It is possible to obtain values for these
parameters by empirical methods as introduced, for example, by Delany and Bazley
1
, methods which
have been widely used to describe sound propagation in fibrous materials.
In the equivalent fluid approach, the equation of motion reads
where is the equivalent density, the equivalent bulk modulus, the complex pressure amplitude
and the circular excitation frequency, in which it can be shown
1
e
JJJJJ
2
P

2
B
e
JJJJJJ P H 0 Z
e
B
e
P
e
JJJJJ
1
JJJ 1 iGE H ( ) Z

164
and
if it is assumed that the parameters are not frequency dependent, a reasonable assumption for the study
of frequencies in a narrow band. Here, , and are the values of RHO, BULK and GE
respectively of the MAT10 entry for the porous absorber material. A new field 7 has been added to the
MAT10 entry to allow the value of , the normalized admittance coefficient
2
, to be entered.
Inputs
Defines material properties for fluid elements in coupled fluid-structural analysis.
Format:
The following may be used to calculate the equivalent fluid property values to be entered on the MAT10
entry starting from the complex density and complex bulk modulus describing the rigid porous absorber.
If a complex density and complex speed of sound are determined for the porous material, the complex
bulk modulus must first be calculated.
MAT10 density
MAT10 damping coefficient
MAT10 bulk modulus
MAT10
Fluid Material Property Definition
1 2 3 4 5 6 7 8 9 10
MAT10 MID BULK RHO C GE ALPHA
1
B
e
JJJJJJ
1
B
JJJ Z i
A
JJJJ
1
B
JJJ 1
i
JJJJJ
'
| |
Z
B GE

r
2
i
2
H
r
JJJJJJJJJJJJJJJJJJJ Z
i
complex density, real part
complex density, imaginary part
GE
GE

i
r
JJJJJ Z
B
B
B
r
2
B
i
2
H
B
r
JJJJJJJJJJJJJJJJJJJJ Z
complex bulk modulus, real part
complex bulk modulus, imaginary part
B
r
B
i
165
CHAPTER 5
NVH & Acoustics
MAT10 normalized admittance coefficient
Discussion
The implementation implies that if the complex density and bulk modulus are constant, the normalized
admittance coefficient is a function of frequency. However, frequency dependent is not supported, so
a reference frequency must be selected. Typically, this frequency will be either in the mid-range of the
desired frequency range to be studied, or will correspond to the frequency at a response peak of interest.
As frequencies progressively further away from the reference frequency are considered, the response
becomes increasingly subject to the limitations of the frequency independent formulation; the extent will
depend somewhat on the nature of the problem, and it may be necessary to study discrete frequency
bands in order to mitigate against this effect.
The use of a non-zero value in field 7 of the MAT10 entry causes the generation of a damping matrix
because the normalized admittance coefficient is multiplied by the imaginary operator i. Consequently,
the use of modal methods on the fluid are not appropriate and frequency response analysis must be
carried out using the direct method, at least for the fluid.
References
1. M.E. Delany & E.N. Bazley, Acoustical Characteristics of Fibrous Absorbent Materials,
National Physics Laboratory, Aerodynamics Division, NPL Aero Report Ac 37, March 1969.
2. J. Wandinger, Possible Implementations of Porous Absorbers in Nastran, MSC internal memo,
April 2006.
Example
Consider the following unbounded fluid (air) and porous absorber medium domains as in Figure 5-8. An
acoustic source is placed at the location indicated and the acoustic response (pressure) at the centre of
the fluid is monitored.

B
i
B
r
JJJJJ Z

166
Figure 5-8 Porous medium
The following properties were determined using experimental methods.
Air
Porous Absorber
Density 1.225 + i0
Speed of sound 340.0 + i3.4
Density 3.8663 + i14.2204
Speed of sound 92.7076 + i70.2854
Porous medium
Response
Acoustic source
Air
167
CHAPTER 5
NVH & Acoustics
From which the following equivalent bulk moduli were obtained
Air
Porous Absorber
A frequency of 250 Hz was selected to calculate the values of alpha for air and the porous absorber.
Using the equations illustrated above, the following data is entered on the MAT10 entries.
MAT10 for air
MAT10 for the porous absorber material
Notice that the values of bulk modulus, GE damping coefficient and alpha are all negative; this is a
normal characteristic of the implementation.
The response at the centre of the air domain is calculated using the new capability and the results
compared with the same model run in Actran. Both HEXA-20 (Figure 5-9) and HEXA-8 (Figure 5-10)
elements are compared.
Bulk modulus 141595.8 + i2832.2
Bulk modulus -171190.0 + i102356.3
MID BULK RHO C GE ALPHA
MAT10 1 141652.5 1.225 0.0 31.41907
MID BULK RHO C GE ALPHA
MAT10 2 -232389. 56.16948 -3.67804 -939.196

168

Figure 5-9 HEXA-20
Figure 5-10 HEXA-8
HEXA- 20
60
80
100
120
140
160
180
0 100 200 300 400 500
Fr e que nc y ( Hz )
Actran Nastran
P
r
e
s
s
u
r
e

(
r
e
a
l
)
HEXA- 8
60
80
100
120
140
160
180
0 100 200 300 400 500
Fr e que nc y ( Hz )
Actran Nastran
P
r
e
s
s
u
r
e

(
r
e
a
l
)
169
CHAPTER 5
NVH & Acoustics
The results compare very favorably with those from Actran. Notice the gradual departure from the
expected Actran response for frequencies progressively further away from the reference frequency of
250 Hz, as the response becomes increasingly subject to the limitations of the frequency independent
formulation.

170
Ch. 6: Numerical Enhancements MD R2 Nastran Release Guide
=
6
New SPARSESOLVER Executive Statement
Improved Performance with New Sparse Solvers
Improved Memory Usage in Lanczos (Pre-release)
New MAXRATIO Information Output (Pre-release)
Performance Improvements and Expanded Capabilities for ACMS
Relaxed Restrictions for CASI Solver Usage
System Dependent Performance Improvements
Improved Selection of Reordering Methods


172
New SPARSESOLVER Executive Statement
Introduction
A new interface, the SPARSESOLVER Executive Statement, is now available for analysts to better
control the options and processes associated with sparse matrix solution methods. The interface permits
specification of the solver to be used as well as matrix reordering methods and compression techniques
to be used. Furthermore, improved control of matrix diagonal term ratio output is provided.
Benefits
The new interface provides analysts more succinct control over the process than the existing method
selection process that usually involves defining values for some SYSTEM cells via the NASTRAN
statement. In addition, a new output data option is available for matrix diagonal term ratios in the form
of a simple bar chart that provides a more comprehensive view of the ratio data.
Method and Theory
No new theory is involved. The method involves simply the specification of solver options to be used by
various DMAP modules during the solution process. The specified options are checked against feature
availability tables to ensure that they do not conflict with any limitations posed by the specification of
feature combinations. For example, the specification of a particular ordering method may not be available
for a particular solver specification. The options are available only for the DCMP, DECOMP, FRRD1,
READ, SOLVE and TRD1 modules.
Inputs
The sparse solution options are controlled by keywords specified on the SPARSESOLVER Executive
statement. See the MD Nastran Quick Reference Guide for a complete description of this statement.
Outputs
There are no new outputs associated with this feature other than informational and diagnostic messages.
The ability to specify particular sparse matrix solution options is sometimes useful in determining
whether one method is more effective than another in obtaining the solution. Other features can also be
useful as in obtaining diagnostic data output. For example, one might be interested in reviewing matrix
diagonal term ratios. In general, high ratios indicate a loss of accuracy. The feature can be used by taking
all of the program defaults for the various control variables. These defaults produce both the table and
bar outputs. The table is limited to 25 ratios that exceed 1.0E+05. The bar chart uses powers of ten for
segment widths. This can be done by simply adding:
173
CHAPTER 6
SPARSESOLVER DCMP ( MDTRATIO )
to the Executive Section of the input data file. The use of this new feature is currently limited to sparse
symmetric matrix operations in the DCMP module.
Several different sparse matrix factorization methods are available. Specification of a particular method
should be done only after thoughtful consideration. The following table summarizes the advantages and
disadvantages of the various factorization methods.
Similarly, re-ordering methods can also be specified, but only should be done after consideration of the
potential effects. The following table summarizes the advantages and disadvantages of the various re-
ordering methods.
Method Advantages Limitation
MSCLDL Small memory requirements; handles
indefinite matrices
None
MSCLU Small memory requirements None
TAUCSCHL Generally superior performance
compared to MSCLDL
High memory requirements; real positive
definite matrices only
UMFLU Generally superior performance
compared to MSCLU
High memory requirements
Method Advantage Limitation
AMF Similar to BEND but with smaller
memory requirements
Produces less optimal reordering
compared to BEND
BEND Optimal for small matrices and for large
matrices from models dominated by 1-
and 2-dimensional finite elements
None
MD Very low memory requirements Produces sub-optimal reordering
METIS Good for large matrices dominated by 3-
dimensional finite elements
Large memory requirements
MMD Low memory requirements Produces sub-optimal reordering

174
Similarly, compression methods can also be specified, but only should be done after consideration of the
potential effects. The following table summarizes the advantages and disadvantages of the various
compression methods.
Method Advantage Limitation
GRID Utilizes USET and SIL information from the
Nastran Database
Relies on USET and SIL
SUPER Does not rely on USET and SIL tables; can
produce better orderings for models
dominated by 3-dimensional elements
None
GRDSUPER Can produce better orderings for models
dominated by 3-dimensional finite elements
Requires USET and SIL tables
175
CHAPTER 6
Improved Performance with New Sparse Solvers
Introduction
As part of MSC Software HPC development, surveys of solver technology from industry and academia
are conducted. This has lead to the integration of two solvers developed outside of MSC Software:
TAUCS and UMFPACK.
In addition, MSC has enhanced the Lanczos eigensolver to take better advantage of available memory.
The TAUCS Sparse Solver
A new sparse Cholesky solver is available. The solver is derived from the TAUCS package of linear
solvers.
User Interface
The new solver can be selected for the DCMP, DECOMP, and SOLVE modules through the
SPARSESOLVER Executive statement:
SPARSESOLVE DCMP (FACTMETH=TAUCSCHL)
It may also be selected by setting system cell 166 to 8192.
Use Cases
The new solver keeps the matrix data in memory, and may show improved performance over the default
solver for models which are dominated by three dimensional elements.
The four models described below demonstrate the potential performance enhancement with the new
solver.
Each model was run on one node with 2 dual core 2.4GHz Opteron processors, 8Gb of memory and 56Gb
of scratch space. Each job was run with mem=7gb, and the CPU time of the DCMP module is displayed
in the following chart.
Model Name DOF Elements
zzz3 537876 108675 TETRAs
vl0sst 408999 710768 TETRAs
xl0rst 739815 113217 HEXAs
Solidcyl 604800 176400 HEXAs

176

Limitations
The new solver only works for real, symmetric matrices which are positive definite. Modeling techniques
which lead to indefinite matrices, such as Lagrange multipliers, are not supported by the new solver. If
the new solver detects an indefinite matrix, or has insufficient memory to perform the factorization, or
encounters any other error, the out-of-core LDL
T
solver is used.
The new solver is not recommended for multiple superelement analysis.
The TAUCS code is used by MSC under the following license agreement. Please note that in the
agreement, this program and this software refers only to the TAUCS code, available at
http://www.tau.ac.il/~stoledo/taucs.
TAUCS version 2.0, November 29, 2001. Copyright 2001 by Sivan Toledo, Tel-Aviv University,
stoledo@tau.ac.il. All Rights Reserved.
Used by permission.
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 site 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 the type taucs must also cite this License and the Availability note.
Permission to modify the code and to distribute modified code is granted, provided that the Copyright,
this License, and the Availability note are retained, and a notice that the code was modified is included.
This software is provided free of charge.
zzz3 vl0sst xl0rst solidcyl
0
50
100
150
200
250
300
350
400
450
500
550
600
650
700
750
800
New Cholesky Solver Performance
MSCLDL
TAUCSCHL
Model
D
C
M
P

t
i
m
e

(
c
p
u

s
e
c
)
177
CHAPTER 6
The UMFPACK Sparse Solver
UMFPACK is a set of routines for solving unsymmetric sparse linear systems using an unsymmetric
multi-frontal method. The UMFPACK factorization has been implemented in Nastran as a licensed
software product from the University of Florida. It operates on both real and complex matrices.
UMFPACK has been implemented for solving unsymmetric frequency response problems. In addition,
it is implemented in the SOLVE and DECOMP modules. It is activated by an explicit user request via
the SPARSESOLVER Executive statement. For example:
SPARSESOLVER FRRD1 (FACMETH=UMFLU)
This command specifies that UMFPACK is to be run in the FRRD1 module, which is used for frequency
response analysis problems. UMFPACK may also be selected by setting system cell 209 to 16.
The UMFPACK solver has shown to be very useful for problems in Exterior Acoustics, which generates
unsymmetric matrices for frequency response.
Examples:
Two simple example models in Exterior Acoustics are shown, to demonstrate the usefulness of
UMPACK for unsymmetric applications. Model 2 is meshed twice: once with a 64 x 64 plate structure,
and once with a 128 x 128 plate structure.
The model properties are described in the following table. Each model was run with SOL 111. There are
200 forcing frequencies in each run.
Model 1
Model 2

178
All jobs were run on a 2.6GHz Intel Woodcrest processor system with 16GB main memory, running
under Linux.
In order to use UMFPACK in a user-written DMAP sequence via the DECOMP module, permutations
must be applied by the DMAP writer, using MATMOD option 45. In the following DMAP sequence
fragment, the matrix equation [A][x]=[b] is solved.
DECOMP A/L,U,ROWPERM,COLPERM/4 $ Option 4 for UMFPACK
MATMOD B,ROWPERM,,,,/BPERM,/45/1 $ Permute the RHS
FBS L,U,BPERM/XPERM $ Solve for permuted X
MATMOD XPERM,CPERM,,,,/X,/45/-1 $ Un-permute X
Element Distribution
Elements Model1 Model2a Model2b
Air 6566 13440 13440
Infinite 768 1664 1664
Air_low n/a 8192 8192
Plate 4096 4096 16384
Total 11520 27392 39680
Note: for the DECOMP module to call UMFPACK, the first parameter (KSYM) must be set to 4.
179
CHAPTER 6
UMFPACK Memory Guidelines
The UMFPACK solver operates entirely in memory. Therefore, memory requirements for UMFPACK
may be considerably higher than for the default Nastran sparse direct unsymmetric solver.
Additionally, it is important to note that UMFPACK operates entirely outside the control of Nastran
memory management. Each time it is invoked, UMFPACK will obtain memory from the operating
system, use the new memory, and then return the new memory back to the operating system. If there is
insufficient memory to complete the UMFPACK factorization, Nastran terminates with a Fatal Error
message, and it attempts to inform the user of memory requirements. It is not possible to allocate
additional memory for UMFPACK via the nastran command line option "mem=".
It is the user's responsibility to ensure that sufficient unsubscribed memory is available from the
operating system for UMFPACK to complete successfully. This may mean lowering the amount of
memory requested at job submittal with the "mem=" option. For example, suppose there is 8GB of
memory on your computer, and you want to use Nastran with UMFPACK. If the memory requirement
for UMFPACK to complete is 4GB, then you should not submit your Nastran job with more than 4GB.
In fact, UMFPACK will be the most memory intensive portion of your Nastran analysis, so that the actual
Nastran memory could be set much lower by comparison.
This memory limitation will be removed in a future release, so that all UMFPACK memory operations
will be controlled by Nastran memory management.

180
Improved Memory Usage in Lanczos (Pre-release)
The Lanczos method has been improved to better take advantage of the given memory.
This beta feature may be selected by setting system cell 146 to -1.
To illustrate the performance improvements, the following models were run on an SGI Altix with four
1.4GHz itanium2 processors, 20Gb of ram and 470Gb of disk space. Each model was run with
mem=4200mb.
Model Name DOF Elements
xl0u 595303 55925 HEXAs, 18483 PENTAs
xl0h 335680 54509 QUAD4s, 15523 TRIA3s
xl0j 590474 94979 QUAD4s, 13406 TRIA3s
xl0m 529199 96185 QUAD4s
xl0u xl0j xl0h xl0m
0
50
100
150
200
250
300
350
400
450
500
550
600
650
700
Improved Lanczos Performance
2006R1
2006R2
Model
R
E
A
D

t
i
m
e

(
c
p
u

s
e
c
)
181
CHAPTER 6
New MAXRATIO Information Output (Pre-release)
Introduction
A new interface is now available for analysts to better control the generation of matrix diagonal term
ratio statistics produced by the sparse symmetric matrix decomposition process in the DCMP module.
The matrix diagonal term ratio statistics are sometimes useful in determining the quality of the matrix
decomposition process. In general, for linear static analysis, high or negative ratios indicate a loss of
accuracy and could be indicative of a modeling error.
Benefits
The new interface provides analysts more control over the process than the existing method of supplying
a value for the MAXRATIO DMAP parameter. In addition, a new output data option is available in the
form of a simple bar chart that provides a more comprehensive view of the ratio data.
Method and Theory
No new theory is involved. The method involves simply the computation of a ratio defined to be the
original matrix diagonal term divided by the decomposed matrix diagonal term. These ratios are placed
in a table together with the external identifier associated with the row/column of the term. This table is
then processed according to the options requested by the user.
Inputs
The matrix diagonal term ratio output options are controlled by keywords specified on the
SPARSESOLVER Executive statement. See New SPARSESOLVER Executive Statement on page 172
for a complete description of this statement.
Outputs
The matrix diagonal term ratios can be presented in two different views. The first view is the table view
in which each ratio is listed together with the external identifier of the row/column of the matrix as well
as the original input matrix diagonal term. This format is virtually identical to that produced now when
any ratio exceeds the value of the MAXRATIO input parameter. The second view of the ratios is
statistical in nature. It is similar to a bar chart. A series of bar segments are generated. There are two
options for specifying the segment widths of the bars. The default option uses powers of 10 as the widths
(e.g. 10.0 to 100.0 and 100.0 to 1000.0). The second option allows the user to specify how many
segments are desired. The program will compute the segment width using the maximum and minimum
ratios. For each bar in the chart, the total number of terms in the range is tabulated together with a visual
indication of the percentage number of terms in that particular bar.

182
Note that when negative matrix diagonal term ratios are detected, they will always be output if the
TABLE option is specified.
These new views of the ratios do not replace any existing diagnostics generated by the DCMP module if
a problem is detected. Under these conditions, output from the table view may duplicate previous output
generated by DCMP module error processing.
The matrix diagonal term ratio statistics are sometimes useful in determining the quality of the matrix
decomposition process. In general, high ratios indicate a loss of accuracy. The feature can be used by
taking all of the program defaults for the various control variables. These defaults produce both the table
and bar outputs. The table is limited to 25 ratios that exceed 1.0E+05. The bar chart uses powers of ten
for segment widths. This can be done by simply adding
SPARSESOLVER DCMP (MDTRATIO)
to the Executive Section of the input data file.
The use of this new feature is currently limited to sparse symmetric matrix operations in the DCMP
module.
If there are scalar-type points present in the problem, the degrees of freedom associated with these points
will be grouped into the results for the translational degrees of freedom output.
Demonstration Example
A simple example is presented that demonstrates the use of some of the new features available for output
of the matrix diagonal term ratios. The SPARSESOLVER Executive statement is used to specify the
desired features. The example problem is used for demonstration purposes only and is not representative
of anything in particular. The model data consists of a simple plate structure subject to an end load.
Example Input Data
$
$ Example problem to demonstrate matrix diagonal term ratios
$
id test,case
sol 101
SPARSESOLVER DCMP (MDTRATIO) $
cend
spc=100
load=1000
disp=all
begin bulk
grdset,,,,,,,6
cquad4,101,101,1,2,52,51
cquad4,102,101,2,3,53,52
cquad4,103,101,3,4,54,53
cquad4,104,101,4,5,55,54
cquad4,105,101,5,6,56,55
183
CHAPTER 6
cquad4,106,101,6,7,57,56
cquad4,107,101,7,8,58,57
cquad4,108,101,8,9,59,58
cquad4,109,101,9,10,60,59
cquadr,1101,101,1,2,52,51
cquadr,1102,101,2,3,53,52
cquadr,1103,101,3,4,54,53
cquadr,1104,101,4,5,55,54
cquadr,1105,101,5,6,56,55
cquadr,1106,101,6,7,57,56
cquadr,1107,101,7,8,58,57
cquadr,1108,101,8,9,59,58
cquadr,1109,101,9,10,60,59
grid, 1,, 0.0,0.0,0.0
grid, 2,, 1.0,0.0,0.0
grid, 3,, 2.0,0.0,0.0
grid, 4,, 3.0,0.0,0.0
grid, 5,, 4.0,0.0,0.0
grid, 6,, 5.0,0.0,0.0
grid, 7,, 6.0,0.0,0.0
grid, 8,, 7.0,0.0,0.0
grid, 9,, 8.0,0.0,0.0
grid,10,, 9.0,0.0,0.0
grid,51,, 0.0,1.0,0.0
grid,52,, 2.4,1.0,0.0
grid,53,, 3.5,1.0,0.0
grid,54,, 4.6,1.0,0.0
grid,55,, 5.7,1.0,0.0
grid,56,, 6.8,1.0,0.0
grid,57,, 7.9,1.0,0.0
grid,58,, 9.0,1.0,0.0
grid,59,,10.1,1.0,0.0
grid,60,,11.2,1.0,0.0
$
ctria3,201,101,101,102,151
ctria3,202,101,102,152,151
ctria3,203,101,102,103,152
ctria3,204,101,103,153,152
ctria3,205,101,103,104,153
ctria3,206,101,104,154,153
ctria3,207,101,104,105,154
ctria3,208,101,105,155,154
ctriar,1201,101,101,102,151
ctriar,1202,101,102,152,151
ctriar,1203,101,102,103,152
ctriar,1204,101,103,153,152
ctriar,1205,101,103,104,153
ctriar,1206,101,104,154,153
ctriar,1207,101,104,105,154
ctriar,1208,101,105,155,154
grid,101,, 0.0,0.0,0.0
grid,102,, 1.0,0.0,0.0
grid,103,, 2.0,0.0,0.0
grid,104,, 3.0,0.0,0.0
grid,105,, 4.0,0.0,0.0
grid,151,, 0.0,1.0,0.0
grid,152,, 3.4,1.0,0.0
grid,153,, 4.5,1.0,0.0
grid,154,, 5.6,1.0,0.0
grid,155,, 6.7,1.0,0.0

184
$
pshell,101,1,0.05,1
mat1,1,10.+6,,0.33
spc1,100,123,1,101
spc1,100,3,5,55,105,155
spc1,100,1,55,155
spc1,100,2,1,101
force,1000,10,,1000.0,1.0,0.0,0.0
force,1000,60,,1000.0,1.0,0.0,0.0
force,1000,105,,1000.0,1.0,0.0,0.0
force,1000,155,,1000.0,1.0,0.0,0.0
enddata
Example Output
The output generated by the previous example is shown following. Notice that there are two separate
sections of output: one for translational degrees of freedom and one for rotational. Within each section,
both a bar chart and table of matrix diagonal term ratios are output.
TRANSLATIONAL DOF DIAGONAL TERM RATIO STATISTICS CHART FOLLOWS FOR THE DECOMPOSITION OF MATRIX KLL
------------------------------------------------|---------------------------------------------------------------------------
DIAGONAL TERM RATIO RANGE #TERMS % TOT |MAXIMUM RATIO = 6.90963E+02 MINIMUM RATIO = 1.00000E+00
------------------------------------------------|---------------------------------------------------------------------------
1.0000E+00 TO 1.0000E+01 62 79.49 |**************************************************************************>
1.0000E+01 TO 1.0000E+02 12 15.38 |***************
1.0000E+02 TO 1.0000E+03 4 5.13 |*****

0
0 MATRIX/FACTOR DIAGONAL TERMS RATIO SUMMARY TABLE FOR TRANSLATIONAL DOF SORTED ON DIAGONAL RATIO
GRID POINT ID DEGREE OF FREEDOM MATRIX/FACTOR DIAGONAL RATIO MATRIX DIAGONAL
(TOP 1 RATIOS>MAXRAT= 6.90963E+02)
58 T3 6.90963E+02 5.65535E+04

ROTATIONAL DOF DIAGONAL TERM RATIO STATISTICS CHART FOLLOWS FOR THE DECOMPOSITION OF MATRIX KLL
------------------------------------------------|---------------------------------------------------------------------------
DIAGONAL TERM RATIO RANGE #TERMS % TOT |MAXIMUM RATIO = 3.35974E+02 MINIMUM RATIO = 1.00000E+00
------------------------------------------------|---------------------------------------------------------------------------
1.0000E+00 TO 1.0000E+01 38 63.33 |***************************************************************
1.0000E+01 TO 1.0000E+02 18 30.00 |******************************
1.0000E+02 TO 1.0000E+03 4 6.67 |*******
0
0 MATRIX/FACTOR DIAGONAL TERMS RATIO SUMMARY TABLE FOR ROTATIONAL DOF SORTED ON DIAGONAL RATIO
GRID POINT ID DEGREE OF FREEDOM MATRIX/FACTOR DIAGONAL RATIO MATRIX DIAGONAL
(TOP 1 RATIOS>MAXRAT= 3.35974E+02)
58 R2 3.35974E+02 2.14135E+04
185
CHAPTER 6
Performance Improvements and Expanded
Capabilities for ACMS
Introduction
Automated Component Modal Synthesis (ACMS) is a powerful tool for a variety of large modal based
analyses. ACMS functions in two domains, the Matrix Domain and the Geometric Domain. Since its
introduction in 2005, Matrix Domain ACMS (MDACMS) has become the default ACMS method.
Developments described in this section apply to MDACMS.
MDACMS has been extended to more thoroughly interact with the Nastran External Superelement
capability. See "MDACMS for Upstream Superelements" below.
An automatic decision logic has been implemented to select which path to take for frequency response
calculations. See "Automatic FASTFR Decision Logic" below.
For other enhancements, see "Miscellaneous Performance Improvements" below.
MDACMS for Upstream Superelements
MDACMS has been extended to more thoroughly interact with the Nastran External Superelement
capability. Specifically, MDACMS reduction calculations have been expanded to include all quantities
being reduced to the residual, namely stiffness, mass, and damping. (Previously, only the component
modes were computed by MDACMS.) By reducing these quantities while simultaneously computing
component modes, a significant reduction in compute time, disk I/O, and scratch disk space is realized.
One typical case study is presented here. Note that in this example, the resource requirements for MD
Nastran R1 exceeded availability and the job did not complete, even though it had run five times longer
than the current release at the time of its failure.
No. of Grid
Points
No. of global
DOF
No. of O-set
DOF
No. of A-set
DOF Analysis Description
375,502 2.25 million 1.9 million 3526 Normal modes, frequency range
0-550Hz (1627 modes)

186
The above jobs were run on an IBM Power4 computer running the AIX operating system.
Automatic FASTFR Decision Logic
In MSC.Nastran 2004, the FASTFR method was introduced for modal frequency response analysis. It
can be selected via the Bulk Data entry PARAM,FASTFR,YES and shows significant performance
improvement for certain models in the mid-frequency range. However, the user has to make the decision
as to whether to use the new method or whether to run with the standard FRRD1 method with or without
the iterative solver before starting to run the job.
With this release, automatic decision logic has been implemented which eliminates the need for the user
to make that decision. The program will decide automatically which solution method will be most
efficient for the frequency response portion in a SOL 111 analysis. Based on the size of the modal space
and some other heuristic criteria, either the FASTFR solution method will be run, or the FRRD1 module
with or without the iterative solver will be used.
Currently, the FASTFR method is selected or deselected in the Bulk Data Section using param,fastfr,yes
or param,fastfr,no. To activate the automatic decision logic, specify
PARAM, FASTFR, AUTO
in the Bulk Data Section. This will cause the program to run as if the FASTFR method was selected until
it reaches the point where the decision will be made. If it is decided that it would be faster to not run with
the FASTFR method, the following message will be printed in the F06 file, and the program will continue
with the standard frequency response method.
Also, if the FASTFR method is deselected for a different reason, a similar message will be printed. If the
^^^ SYSTEM INFORMATION MESSAGE 9157 (GMA)
^^^ FASTFR OPTION REQUESTED, BUT THE MODEL DID NOT MEET THE FOLLOWING CRITERIA:
^^^ THE FASTFR OPTION MAY BE TOO EXPENSIVE.
^^^ USER INFORMATION: STANDARD FREQUENCY RESPONSE METHOD WILL BE USED
187
CHAPTER 6
FASTFR method will actually be used, then no message will be printed.
This new option for PARAM,FASTFR is helpful for all modal frequency response analysis jobs,
especially in the mid to upper frequency range.
Examples
The following charts show the benefit of the automatic decision logic for three examples. Please note
that the times given are the elapsed minutes for the complete job.
Example 1
For this automotive example, the automatic decision logic decided that the FASTFR method might be
too expensive. So it switched to the FRRD1 module with the iterative solver.

Example 2
For this automotive example, the automatic decision logic decided to go with the FASTFR method, and
rightly so. Both, the FRRD1 module with and without the iterative solver proved to be much slower.
Job Statistics: 1,208,135 grid points;
matrix size in FRRD1 (i.e. H-size) is 9,974;
modes below 1,000 Hz;
2 frequencies.
Job Statistics: 233,094 grid points;
matrix size in FRRD1 is 4,350;
modes below 600 Hz;
260 frequencies.
0
200
400
600
800
1000
1200
1400
FASTFR FRRD1 (iter) FRRD1
(direct)
Decision
Time

188
Example 3
For this automotive example, the automatic decision logic decided to go with the FASTFR method which
proves to be significantly faster than the conventional method (i.e. FRRD1 with the direct method). The
iterative method inside FRRD1 is not practical for this job because there are 147 loads which
significantly slow down the iterative solver.
Limitations for the FASTFR method
1. Fluid Damping can only be specified via the param,gfl entry and/or the sdamping(fluid)
command.
a. The CAABSF element (acoustic absorbers) is not supported;
b. The FASTFR method will be turned off automatically for fluid K4 and for fluid viscous
damping.
Job Statistics: 321,597 grid points;
matrix size in FRRD1 is 5,424;
modes below 600 Hz;
501 frequencies.
0
50
100
150
200
250
300
350
FASTFR FRRD1 (iter) FRRD1
(direct)
Decision
Time
0
500
1000
1500
2000
2500
FASTFR FRRD1(direct) Decision
Time
189
CHAPTER 6
2. All matrices must be symmetric
a. Unsymmetric formulation for acoustic coupling is not supported;
b. EPOINT Bulk Data entries are not supported.
3. The FASTFR method work only for SOL 111 and for SOL 200 with ANAL=MFREQ.
4. SESDAMP and FASTFR are not allowed in the same run.
Miscellaneous Performance Improvements
MDACMS has been enhanced to reduce the amount of disk I/O required for a successful analysis. This
happens automatically and there is no user action required to realize the reduction in I/O. Typically,
reduced I/O requirements results in reduced elapsed time and greater CPU utilization, which increases
the effectiveness of Shared Memory Parallel (SMP) computations.
To demonstrate some of these performance improvements, three example jobs were run and the
performance results were compared to the previous version. The results are displayed below.
Model Description:
Model
No.
No. of grid
Points
No. of global
DOF Analysis Description
1 44,314 265884 Auto body normal modes run w/ACMS. Frequency range 0-400Hz
(360 modes)
2 44,314 265884 Model (1) run in SOL 111 w/ACMS; 75 load cases, 250 forcing
frequencies
3 1,243,651 7.4 million Mid frequency acoustic analysis of automotive trimmed body. 9389
structure modes below 750Hz; 1486 fluid modes below 1200Hz; 4
load cases, 290 forcing frequencies

190
Example 1 shows time and I/O resource requirements for model one run on an IBM Power4 system under
AIX.
191
CHAPTER 6
Example 2 shows DMP scalability improvement for model two. These jobs were run on a Altix machine
with IA64 processors under Linux.
Example 3 shows I/O and elapsed time improvement for model three. These jobs were run on an IBM
Power4 system under AIX.

192
Relaxed Restrictions for CASI Solver Usage
The use of the element-based CASI iterative solver is limited by several restrictions noted in the remarks
for the ITER Bulk Data entry. Two of these restrictions have been somewhat relaxed so that the solver is
usable over a wider range of problems.
The first is a reduction in the number of element types that the solver does not recognize. The solver
recognizes only a sub-set of all of the element types available, but the list has been expanded to include
the FAST, SEAM and WELD element types as supported element types.
The second is a relaxation of the restriction associated with the processing of direct input matrices via the
K2GG Case Control command. K2GG matrix input is now allowed by the CASI solver interface.
However, the matrix size is limited to 100 grid/scalar points, since it must be treated as an unknown finite
element type. This should be sufficient for most cases where the input K2GG matrix represents some
form of attachment to, or base for, the structural model being analyzed.
193
CHAPTER 6
System Dependent Performance Improvements
Introduction
MSC has implemented new versions of the Basic Linear Algebra Subroutine library (BLAS) for Nastran,
on platforms supported by the "x86-64" architecture.
Version 9 of Intel's Math Kernel Library (MKL) provides optimal performance for analysis solutions that
make good use of the BLAS. This includes the ACMS solution for automotive NVH and acoustic
analysis, and the Exterior Acoustics capability in Nastran. In addition, MKL provides automatic
multithreading support. Shared memory parallel speedup is available via the smp=n command supplied
to the nastran command processor.
The ATLAS project (Automatically Tuned Linear Algebra Software) provides a portable, efficient
version of the BLAS. An ATLAS version of the BLAS library has been implemented for use on AMD
Opteron processors. ATLAS is public domain software available on the internet at
http://math-atlas.sourceforge.net/.
For demonstration purposes, two analysis model were selected and run with the current and prior release
of MD Nastran. Performance results are shown on the following page.
Model Description
Model
No.
No of Grid
Points
No. of
Global DOF Analysis Description
1 268,486 1.6 million Auto body normal modes run w/ACMS. Frequency range
0-200Hz (1043 modes)
2 603,266 2.8 million Low frequency acoustic analysis of automotive trimmed
body

194
Performance Improvements for Intel x86_64
Performance Improvements for AMD due to Atlas
Processor Clock Speed OS Real Memory Scratch Filesystem
Woodcrest 2666MHz Linux RHWS4 16GB ext2 not striped
Processor Clock Speed OS Real Memory Scratch Filesystem
Opteron 2400MHz Linux SuSE 10 8GB xfs not striped
Intel Woodcrest Performance Comparison
Model 2
0
2000
4000
6000
8000
10000
12000
14000
MD R1 MD R2 MD R2 2smp
17%CPU Ti me Speedup(seri al )
T
i
m

i
n

S
e
c
o
n
d
s
elapsed
cpu
Intel Woodcrest Performance Comparison
Model 1
0
500
1000
1500
2000
MD R1 MD R2 MD R2 2smp
17%CPU Ti me Speedup (seri al )
T
i
m

i
n

S
e
c
o
n
d
s
elapsed
cpu
AMD Opteron Performance Comparison
Model 1
0
500
1000
1500
2000
2500
3000
MD R1 MD R2
17%CPU Ti me Speedup
T
i
m
e

i
n

s
e
c
o
n
d
s elapsed
cpu
AMD Opteron Performance Comparison
Model 2
0
5000
10000
15000
20000
MD R1 MD R2
18% CPU Time Speedup
T
i
m
e

i
n

s
e
c
o
n
d
s
elapsed
cpu
195
CHAPTER 6
Improved Selection of Reordering Methods
Nastran contains five methods for reordering (permuting) sparse matrices in preparation for a sparse
symmetric factorization. The five methods are minimum degree (MD), multiple minimum degree
(MMD), and three nested dissection/minimum degree hybrid methods, called BEND, AMF, and METIS.
A judicious choice of a reordering method can lead to a dramatic improvement in the performance of the
sparse factorization.
The hybrid methods (BEND, AMF, and METIS) are generally superior to MD and MMD. The previous
default behavior was to use BEND unless the USET and SIL tables were not available, in which case
MD was used. If the BEND algorithm failed, MD was also used in this case. The new default behavior
is to use BEND except for large (> 50,000 DOF) models, which are dominated by three-dimensional
elements; for those models METIS is the default. If the default method fails, the program will attempt to
use another hybrid method (for example, if BEND fails, the program tries METIS). The minimum degree
algorithms are selected only as a last resort.
To illustrate the performance improvements, the following models were run on an SGI Altix with four
1.4GHz itanium2 processors, 20Gb of ram and 470Gb of disk space.
The chart below shows the factorization time for each job. The improvement for the xx0e, xx0d, and xx0f
jobs stems from the automatic selection of METIS rather than BEND reordering. The improvement in
lg0q is due to the selection of METIS rather than MD.
Model Name Solution DOF Elements
xx0e Normal Modes 3308298 656570 TETRAs
xx0d Normal Modes 1920855 402441 TETRAs
xx0f Statics 1237999 246066 TETRAs
lg0q Direct Freq. Response 93375 28140 HEXAs

196
xx0e xx0f xx0d lg0q
0
250
500
750
1000
1250
1500
1750
2000
2250
2500
2750
Reordering Selection Performance Improvement
2006R1
2006R2
Model
F
a
c
t
o
r
i
z
a
t
i
o
n

t
i
m
e

(
c
p
u

s
e
c
)
Ch. 7: Elements & Connectors MD R2 Nastran Release Guide

7
Enhancements to Connector Elements
Composite Beam Using VAM
Nonhomogeneous Multipoint Constraint


198
Enhancements to Connector Elements
Introduction
In MD R2 Nastran, a new seam weld element is introduced to allow for the definition of a seam line. This
new CSEAM element replaces the existing CWSEAM element and is modeled by the new CSEAM and
PSEAM Bulk Data entries. In addition, the SWLDPRM Bulk Data entry is enhanced to support element
type specific control parameters and different diagnostic output format.
For spot weld elements CWELD and CFAST, the displacements of the projected grids GA and GB are
now computed and output in this release. These data will facilitate users to view the relationship between
the spot weld and the connecting shells.
CSEAM Elements
The new CSEAM element provides the following key features to extend the analysis capabilities, provide
more flexibility in modeling, and improve the robustness of the analysis results.
A seam line is considered continuous between two CSEAM elements that have a common face
based on either common GS/GE points or XYZ coordinates. Note that the SMLN label on a
CSEAM element does not determine the definition of a seam line. It is only intended for ease of
seam line visualization.
This element can connect up to 64 shell grids, which allows the connection of higher order shell
elements.
Besides selecting the connected surface patches by property IDs, the user may define the
connection by specifying shell element IDs directly.
Each of the eight auxiliary points must have a projection onto the connected shell element. This
requirement prevents the generation of ill conditioned stiffness matrix in down stream
processing.
The user can model tailored blank tapering by specifying different property IDs at the start and
end points.
This element type supports the MAT9 anisotropic material properties.
Inputs
The seam connection is modeled by the new CSEAM and PSEAM Bulk Data entries and the modified
SWLDPRM Bulk Date entry. The details of these entries are described in the MD Nastran Quick
Reference Guide.
Outputs
The connecting record based on seam line label is written as a SEAMLN record in GEOM2 data block,
which is included in OP2 file for post-processing. The contents of the SEAMLN record for each seam
line label are listed:
199
CHAPTER 7
Limitations
Each CSEAM element can connect a maximum of three shell elements on patch A and three
shell elements on patch B.
Only line type of seam is supported.
Super-element modeling is not supported.
FORCE, STRESS and STRAIN output requests are not supported.
Example A Symmetric Hat Profile (cseam_hut.dat)
This example demonstrates the application of CSEAM elements to analyze an unconstrained symmetric
hat profile model, see Figure 7-1. Each edge of the hat is connected by 27 CSEAM elements. The grids
with identification numbers 6000 to 6028 and 8000 to 8028 are used as the piercing points to define the
seams, see Figure 7-2.
Hat Profil e
Figure 7-1 Hat Profile
Word Type Description
1-2 C Seam line ID
3 I Total number of elements for this seam line
4 I CSEAM element ID
5 I GS grid ID for the start point
6 I GE grid ID for the end point
: I Repeat words 4-6 for each CSEAM element
: I End the seam line data with -1

200
Piercing Points and Seams
Figure 7-2 Piecing Points and Seams
The input for the seam welds is listed:
CSEAM 7000 500 SMLN_A PSHELL 100 200
6000 6001
6001 6002
:
6026 6027
6027 6028
$
CSEAM 9000 500 SMLN_B PSHELL 100 200
8000 8001
8001 8002
:
201
CHAPTER 7
8026 8027
8027 8028
$
PSEAM 500 1 LINE 1.0
The normal mode analysis results are shown:
SWLDPRM Enhancements
Element Specific Control Parameters
The SWLDPRM Bulk Data entry is enhanced to support element type specific control parameters. Two
new parameters, CNRAGLI and CNRAGLO, are introduced to define the angle limits for checking the
geometry of seam elements. These parameters replace the GSPROJ parameter to define the allowable
angles for corner check. The GSPROJ parameter specified in the previous seam weld models must be
replaced by CNRAGLI and CNRAGLO to obtain identical results.
For example, if the SWLDPRM Bulk Data entry is defined as
SWLDPRM,PROJTOL,0.0,GSMOVE,0,GSPROJ,-1.,NREDIA,4,
,GMCHK,1,CHKRUN,1
then this entry must be changed into
SWLDPRM,PROJTOL,0.0,GSMOVE,0,GSPROJ,-1.,NREDIA,4,
,GMCHK,1,CHKRUN,1,cnraglo,-1.
to get the same results as the results running from previous version.

R E A L E I G E N V A L U E S
MODE EXTRACTION EIGENVALUE RADIANS CYCLES GENERALIZED GENERALIZED
NO. ORDER MASS STIFFNESS
1 1 -5.927181E-05 7.698819E-03 1.225305E-03 1.000000E+00 -5.927181E-05
2 2 -2.941204E-05 5.423287E-03 8.631429E-04 1.000000E+00 -2.941204E-05
3 3 2.051645E-05 4.529509E-03 7.208938E-04 1.000000E+00 2.051645E-05
4 4 2.372748E-05 4.871086E-03 7.752574E-04 1.000000E+00 2.372748E-05
5 5 5.330599E-05 7.301095E-03 1.162005E-03 1.000000E+00 5.330599E-05
6 6 1.118197E-04 1.057448E-02 1.682981E-03 1.000000E+00 1.118197E-04
7 7 1.154947E+07 3.398451E+03 5.408803E+02 1.000000E+00 1.154947E+07
8 8 1.585229E+07 3.981493E+03 6.336742E+02 1.000000E+00 1.585229E+07
9 9 2.653947E+07 5.151647E+03 8.199102E+02 1.000000E+00 2.653947E+07
10 10 2.959784E+07 5.440389E+03 8.658648E+02 1.000000E+00 2.959784E+07
11 11 3.002028E+07 5.479077E+03 8.720222E+02 1.000000E+00 3.002028E+07
12 12 3.041736E+07 5.515193E+03 8.777703E+02 1.000000E+00 3.041736E+07
13 13 3.328896E+07 5.769658E+03 9.182696E+02 1.000000E+00 3.328896E+07
14 14 3.757468E+07 6.129819E+03 9.755909E+02 1.000000E+00 3.757468E+07
15 15 4.259932E+07 6.526814E+03 1.038775E+03 1.000000E+00 4.259932E+07
16 16 4.738684E+07 6.883810E+03 1.095592E+03 1.000000E+00 4.738684E+07

202
Real Format of Diagnostic Output
For the PRTSW parameter of SWLDPRM Bulk Data entry, two options are added to support real format
of diagnostic output so that more significant digits will show (quality 1-25861871). This parameter now
has five options.
Displacement Output of GA and GB for CWELD and CFAST
Elements
The displacements of the projected grids GA and GB for CWELD elements with GRIDID, ELEMID,
ELPAT or PARTPAT format and CFAST elements are computed to display the relationship between these
elements and their connecting shell elements. As a result, the displacement output of GA and GB for
CWELD element with GRIDID or ELEMID format and MSET=OFF are no longer dummy zero values.
These displacements are calculated from the constraint equations described in the Basic Theory and
Methods section.
If GA or GB is not specified for CFAST elements or CWELD elements with ELPAT or PARTPAT option,
the program will create a grid internally, with the grid ID number starting from OSWPPT+1 (OSWPPT
is a parameter specified by PARAM Bulk Data entry). The user may request a positive PRTSW parameter
(1, 2, 11, or 12) in SWLDPRM Bulk Data entry to view the grid ID of GA or GB in the diagnostic output.
Input
The displacements of GA and GB are requested using the CONNECTOR keyword of the
DISPLACEMENT Case Control command. See MD Nastran Quick Reference Guide for a detail
description of this command.
Outputs
The output is integrated with the displacements of the general grid points.
Basic Theory and Methods
For CWELD elements with GRIDID or ELEMID format, the displacements of the projected grids GA
and GB in basic coordinate are computed by the following equations:
PRTSW Output
PRTSW = 0 No Diagnostic Output
PRTSW = 1 Print diagnostic output in exponential format to .f06 file
PRTSW = 2 Punch diagnostic output in exponential format to .pch file
PRTSW = 11 Print diagnostic output in real format to .f06 file
PRTSW = 12 Punch diagnostic output in real format to .pch file
203
CHAPTER 7
Where and are displacements at GA and GB. and are constraint matrices. and are
displacements of the connected shell grids.
For CWELD elements with ELPAT or PARTPAT format, the displacements of the auxiliary points in
basic coordinate are calculated first.
Where and are displacements of the auxiliary points. and are RBE3 type constraint
matrices. and are displacements of the connected shell grids.
Then the displacements of the projected grids GA and GB in basic coordinate are computed by the same
constraint equations used for GRIDID and ELEMID options.
Where and are displacements at GA and GB. and are constraint matrices. and are
displacements of the connected auxiliary points.
Current Limitation
The displacement output of GA and GB are only available in Solution Sequences 101 and 103.
Example
This example demonstrates the various displacement output requests and their results for a small model
with two CWELD elements.
The input file follows:
nastran mesh
SOL 101
TIME 60
CEND
TITLE= two elements, identical location for GA, GB, GS
d
A
A
I
u
I
I
=
d
B
B
J
u
J
J
=
d
A
d
B
A
I
B
J
u
I
u
J
u
I
G
I K
u
K
K
=
u
J
G
JL
u
L
L
=
u
I
u
J
G
IK
G
JL
u
K
u
L
d
A
A
I
u
I
I
=
d
B
B
J
u
J
J
=
d
A
d
B
A
I
B
J
u
I
u
J

204
OLOAD= ALL
FORCE = ALL
SUBCASE 1
SUBTITLE= shear the weld
SPC= 1
LOAD= 1
DISP= ALL
SUBCASE 2
SUBTITLE= in plane twist
set 21 = 1002,1003,2011,thru,2014
spc= 1
LOAD= 2
DISP(CONN=ALL)=21
SUBCASE 3
SUBTITLE= upper bending
set 32 = 4
set 33 = 1012,1013,2001,thru,2004
spc= 1
LOAD= 3
DISP(CONN=32)=33
BEGIN BULK
$
$ Grids of lower shell
grid, 1001, , 0., 0., 0.
grid, 1002, , 20., 0., 0.
:
grid, 1013, , 20., 10., 5.
grid, 1014, , 0., 10., 5.
$ Grids of upper shell
grid, 2001, , 0., 0., 0.0
grid, 2002, , 20., 0., 0.0
:
grid, 2013, , 20., 10., 6.0
grid, 2014, , 0., 10., 6.0
$ spot weld grid
grid, 3001, , 10.0, 5.0, 0.0
grid, 3011, , 10.0, 5.0, 10.0
$ quad4s
cquad4, 4001, 10, 1001, 1002, 1003, 1004
cquad4, 5001, 10, 2001, 2002, 2003, 2004
cquad4, 4011, 10, 1011, 1012, 1013, 1014
cquad4, 5011, 10, 2011, 2012, 2013, 2014
:
$ property and material
pshell, 10, 10, 1.0, 10
mat1, 10, 2.e+5 , , 0.3, 0.785e-8
$
$ spot welds
$
cweld, 4, 4, 3001, elemid, , , , , +cw4
+cw4, 4001, 5001
cweld, 5, 4, 3011, elemid, , , , , +cw5
+cw5, 4011, 5011
pweld, 4, 10, 5.0
$
enddata
205
CHAPTER 7
The displacement results are shown as follows:
0 SUBCASE 1
D I S P L A C E M E N T V E C T O R
POINT ID. TYPE T1 T2 T3 R1 R2 R3
1001 G 0.0 0.0 0.0 0.0 0.0 0.0
1002 G 4.887498E-03 5.624979E-04 0.0 0.0 0.0 1.406245E-05
1003 G 4.887498E-03 -5.624979E-04 0.0 0.0 0.0 -1.406245E-05
1004 G 0.0 0.0 0.0 0.0 0.0 0.0
1011 G 0.0 0.0 0.0 0.0 0.0 0.0
1012 G 4.887498E-03 5.624979E-04 -7.685000E-01 -6.750000E-03 5.864999E-02 1.406245E-05
1013 G 4.887498E-03 -5.624979E-04 -7.685000E-01 6.750000E-03 5.864999E-02 -1.406245E-05
1014 G 0.0 0.0 0.0 0.0 0.0 0.0
2001 G 6.058339E-04 3.750000E-04 0.0 0.0 0.0 7.733938E-19
2002 G 5.605834E-03 3.750000E-04 0.0 0.0 0.0 1.084202E-1
2003 G 5.605834E-03 -3.750000E-04 0.0 0.0 0.0 1.191827E-18
2004 G 6.058339E-04 -3.750000E-04 0.0 0.0 0.0 7.623363E-19
2011 G 3.907157E-02 3.750000E-04 8.148733E-04 -2.562901E-16 3.850649E-02 2.836059E-17
2012 G 4.407157E-02 3.750000E-04 -7.693148E-01 -5.232793E-15 3.850649E-02 2.851452E-17
2013 G 4.407157E-02 -3.750000E-04 -7.693148E-01 -5.356609E-15 3.850649E-02 2.956961E-17
2014 G 3.907157E-02 -3.750000E-04 8.148733E-04 -2.636780E-16 3.850649E-02 2.803595E-17
3001 G 0.0 0.0 0.0 0.0 0.0 0.0
3011 G 0.0 0.0 0.0 0.0 0.0 0.0
101000001 G 2.443749E-03 8.917563E-18 0.0 0.0 0.0 8.402567E-19
101000002 G 3.105834E-03 8.944668E-18 0.0 0.0 0.0 9.774760E-19
101000003 G 2.443749E-03 2.517788E-16 -3.842500E-01 -2.775558E-15 3.842500E-02 2.303930E-17
101000004 G 4.157158E-02 3.029749E-15 -3.842500E-01 -2.789435E-15 3.850649E-02 2.786400E-17
0 SUBCASE 2
1002 G 1.499616E-02 3.324066E-02 0.0 0.0 0.0 2.330633E-03
1003 G -1.499616E-02 3.324066E-02 0.0 0.0 0.0 2.330633E-03
2011 G 6.107051E-03 -8.968769E-03 -1.542189E-16 3.303767E-17 1.479144E-16 1.890160E-03
2012 G 2.110705E-02 4.220943E-02 -3.132043E-15 5.117434E-17 1.498367E-16 3.390160E-03
2013 G -2.110705E-02 4.220943E-02 -2.615099E-15 5.202362E-17 1.383761E-16 3.390160E-03
2014 G -6.107051E-03 -8.968769E-03 1.645747E-16 3.118436E-17 1.398878E-16 1.890160E-03
101000001 G 6.505213E-18 1.662033E-02 0.0 0.0 0.0 1.580825E-03
101000002 G 6.722053E-18 1.662033E-02 0.0 0.0 0.0 2.640160E-03
101000003 G 3.035766E-18 1.662033E-02 -1.434168E-15 4.168935E-17 1.434168E-16 1.580825E-03
101000004 G 1.461505E-16 1.662033E-02 -1.434197E-15 4.178690E-17 1.439374E-16 2.640160E-03
0 SUBCASE 3
1012 G 3.460202E-16 9.473882E-16 -1.537000E+00 -1.350000E-02 1.173000E-01 6.294140E-17
1013 G -4.478185E-16 9.592263E-16 -1.537000E+00 1.350000E-02 1.173000E-01 6.130147E-17
2001 G 0.0 0.0 8.021409E-04 -1.965454E-14 7.693148E-02 0.0
2002 G 0.0 0.0 -1.537827E+00 6.397660E-15 7.693148E-02 0.0
2003 G 0.0 0.0 -1.537827E+00 6.758049E-15 7.693148E-02 0.0
2004 G 0.0 0.0 8.021409E-04 -2.021495E-14 7.693148E-02 0.0
101000001 G 0.0 0.0 -7.685000E-01 -6.591949E-15 7.685000E-02 0.0
101000002 G 0.0 0.0 -7.685127E-01 -6.664110E-15 7.693148E-02 0.0

206
Composite Beam Using VAM
Introduction
Arbitrary beam cross section, ABCS, has been implemented in MSC.Nastran 2005. With the increasing
popularity of composite material in the aerospace and auto industries, composite support for arbitrary
beam cross section has been requested by users of MD Nastran. The Variational Asymptotic Method,
VAM, is utilized to compute the beam properties of arbitrary cross section. VAM is available for both
isotropic and composite material. The theoretical background of VAM can be found in many
publications. Literatures significant to this implementation is listed in the references.
Benefits
Beam like structure made of composite material, such as rotorblade, can be modeled with 3D elements,
such as HEXA, PENTA and TETRA, to account for the interaction among plies. An obvious drawback
of 3D elements technique for composite beam is the increase of FE model size which can slow down the
total CAE productivity. The composite beam using VAM provides an alternative to the conventional 3D
modeling technique and permits engineers to use CBEAM3, 3-noded beam element, to model composite
beam. The lay-up of plies of composite beam can be described via PCOMP/PCOMPG Bulk Data entries.
In addition, the cross section of the composite beam can be expressed conveniently with the CP/OP
options of PBMSECT Bulk Data entry.
Input
Modification of PBMSECT for Composite Beam
The following KEYWORDs are added to the PBMSECT Bulk Data entry.
Core(id)=[PCID,PT=(pid1,pid2)]; specifies the composite layup for CORE part of composite. PCID is
the ID of a PCOMPi/PCOMPG Bulk Data entry. PT=(pid1,pid2) defines the start and end points of line
segment(s) which utilizes PCID.
Layer(id)=[PCID,SETID]; specifies the composite layup for additional Layer(s) that are placed on top
or at bottom of Core. PCID is the ID of a PCOMP/PCOMPG Bulk Data entry. SETID selects a
SET1/SET3with POINT IDs.
OUTM=value(Integer > 0), points to the ID of BEGIN BULK ARBMODEL. OUTM is designed
specifically for ABCS with finite element discretization already available. Note that OUTM must not
appear together with other keyword, such as OUTP or INP, on a PBMSECT.
It should be noted that:
1. Keyword Core can be abbreviated as C. Similarly, keyword Layer can be abbreviated as L.
2. If PT=(pid1,pid2) clause is utilized, pid1 and pid2 do not have to be two neighboring POINT in
a SET. However, PID1 must be ahead of PID2 in the sequence.
207
CHAPTER 7
3. If Core and/or Layer appears in PBMSECT, the PID of PBMSECT should not be referenced on
CBEAM. Instead, it should be referenced on CBEAM3.
4. Keyword T (for thickness of isotropic beam) and Core (and Layer; for composite beam) can not
coexist on a PBMSECT entry.
5. If OUTM=arbid is utilized on PBMSECT, element connection, grid location, PSHELL, and
material bulk data entries must be provided after BEGIN ARBMODEL=arbid.
6. The arbid used under BEGIN ARBMODEL is considered global and can be referenced by
PBMSECT with OUTM=arbid in different BEGIN SUPER Bulk Data Section for Part
Superelement(SE).
7. PBMSECT with Core or Layer must be utilized along with
PARAM,ARBMSTYP,TIMOSHEN in the Bulk Data Section.
Selection of VAM for Properties of Arbitrary Beam
PARAM,ARBMSTYP,TIMOSHEN
The default value for PARAMeter ARBMSTYP is VKI which does not support composite beam.
PARAM,ARBMASP,2
The default value for PARAMeter ARBMASP is 2 (integer). Since the thickness of a ply is usually the
lessor dimension, ARBMASP affects the size of QUAD4 lengthwise along a segment.
PARAM,ARBMASP is functional only when PARAM,ARBMSTYP=TIMOSHEN.
Full Stress Recovery for Cross Section
Similar to ABCS with isotropic material, full stress recovery must be requested. The required input for
ABCS logic to perform stress recovery for whole cross section is the element force. Hence, to activate
stress recovery for whole cross section, following input must be present in the input file.
1. ELFORCE (or simply FORCE) output request in the Case Control and
2. PARAM,ARBMSS,YES (Default = NO) as Bulk Data entry
Due to the potential of huge amount of output, these stresses are only available in OUTPUT2 format. In
addition, the direct stresses corresponds to maximum failure index computed during full ABCS stress
recovery are collected for each grid of a CBEAM3 element and made available in f06 file.
Output
Generation of FEM for Cross Sections
Similar to ABCS with isotropic material, FEM for the composite beam can be requested with
PARAM,ARBMFEM,YES. The file generated has following naming convention
AAA_xxyyy_zz.bdf

208
where AAA By default, it assume the input file name. To alter AAA to a name other than input file
name, use
ASSIGN opcase=any character string $
As naming implies, there should have as many files holding FEM of cross sections as
PBRSECT/PBMSECT bulk data entries at conclusion of a job. Note that PBMSECT supports constant
section beam only. For composite beam, a PSHELL entry for each ply of PCOMP/PCOMPG is also part
of the FEM. However, the material entries, such as MAT1, MAT2 and/or MAT8, are not available in the
bdf file of FEM.
The FEM in bdf can be modified (with material entries inserted) and utilized via ARBMODEL=arbid
under OUTM option of PBMSECT. A word of caution on modification. The connection of CQUAD4
elements in the FEM are ordered in such a way that line from GID1 to GID2 defines the X-axis of element
coordinate. Current implementation does not support reordering of element connection based on user
provided coordinate system.
Stress Output for the Whole Cross Section
The file holding stresses for whole cross section in OP2 format has same naming convention as the FEM
for cross section. For stresses, the file has the extension of .op2 instead.
For screened stresses, an output example is shown as follows:
Guidelines
1. CORE (or C) should point to a PCOMP/PCOMPG that has the common plies for most, if not all,
segments. For example,
xx - character string of BR for PBRSECT and BM for PBMSECT
yyy - ID of PBRSECT or PBMSECT
zz - station ID. 01 for end A of PBMSECT. No zz section for PBRSECT.
1 VAB ALGORITHM USING CORE ON PBMSECT JANUARY 8, 2007 MD NASTRAN 1/ 8/07 PAGE 819
TRANSVERSE TIP LOAD
0 SUBCASE 1

S T R E S S E S I N L A Y E R E D C O M P O S I T E E L E M E N T S ( B E A M 3 )
ELEMENT GRID PLY D I R E C T S T R E S S E S FAILURE MAXIMUM STRENGTH
ID ID ID NORMAL-1 NORMAL-2 NORMAL-3 SHEAR-12 SHEAR-23 SHEAR-13 THEORY FAIL. INDEX RATIO FLAG
2 302 2 2.468E+01 1.601E+01 2.570E+00 2.323E+01 4.991E-01 3.724E+00 TSAI-WU 7.161E-04 4.035E+02
102 2 1.685E+01 1.610E+01 -7.230E-01 1.993E+01 -1.377E-01 -5.572E-01 TSAI-WU 7.258E-04 4.470E+02
1301 2 1.558E+01 1.594E+01 -7.167E-01 1.938E+01 -1.162E-01 -5.280E-01 TSAI-WU 7.193E-04 4.569E+02
209
CHAPTER 7
2. PBMSECT,32 is a box beam made of composite material. All segments has a common
CORE=204 with 4 plies. Segments from POINT,2 to 5 has one ply on top, layer=(210,101),
and one ply at bottom, L(2)=(210,103).
3. Negative number is SET1,103 indicates it traverses the top of segments which has the opposite
sequence of that marches at the bottom.
4. Center line of a profile, defined by OUTP and/or BRP, must fall in between plies.
5. Plies of a PCOMP/PCOMPG are always placed from bottom to the top.
6. During transition from OUTP to BRP (or vice versa), all plies above or below centerline, defined
via OUTP/BRP/SET1 entries, must be carried over.
7. Interior ply insertion or deletion between neighboring segments is not supported. Ply/plies
addition on top of the outer most ply can be made via Layer keyword of PBMSECT.
8. PCOMP/PCOMPG referenced on PBMSECT via Core/Layer is converted to PSHELL entries.
The value of THETA field on PCOMP/PCOMPG is placed in T(thickness) field of PSHELL. Due
to non-negative requirement of T field of PSHELL entry, negative THETA is converted to a
supplementary positive value. An example of conversion is shown as follows:
9. Use only matching .bdf and .op2 files from the same job to visualize the stress pattern.
$.......2.......3.......4.......5.......6.......7.......8.......9.......10.....
PBMSECT 32 OP +
OUTP=101,BRP(1)=102,CORE=204, +
layer=(210,101),L(2)=(210,103) +
PCOMP 204 -0.254
501 0.127 -15.0 501 0.127 -15.0
501 0.127 -15.0 501 0.127 -15.0
PCOMP 210 0.0
501 0.127 -15.0
point 2 11.724 6.349
point 3 11.724 -6.349
point 4 -11.724 -6.349
point 5 -11.724 6.349
SET1 101 2 thru 5
SET1 102 5 2
SET1 103 -5 -4 -3 -2
$
pcomp 101 -0.1
501 0.05 0.0 501 0.05 90.0
501 0.05 -45.0 501 0.05 45.0
501 0.05 0.0
PSHELL 10101 501 0.000
PSHELL 10102 501 90.000
$PSHELL 10103 501 -45.000
PSHELL 10103 501 135.000
PSHELL 10104 501 45.000
PSHELL 10105 501 0.000

210
Limitations
1. Although BEAM3 has the curved beam capability, BEAM3 must be straight when utilized for
composite beam.
2. TEMPRB is not supported with current implementation of VAM.
3. SOL 200 does not yet support composite beam.
Example
A simple file, vabcore1, with two PBMSECT entries is utilized here to demonstrate the features
implemented. A bracket used in auto industry is modeled with PBMSECT,31 with isotropic material and
PBMSECT,32 with composite material. Some key bulk data entries are shown as follows:
$
$
param,arbmstyp,timoshen
param,arbmss ,yes
.
.
$ Elements and Element Properties for region : prop2
CBEAM3 2 32 302 102 1301 1204
.
.
$
$<<<rocker at b pillar>>>
$ U profile
point 1 0.8 14.5
point 2 0.9 13.5
point 3 1.0 13.0
point 4 1.1 12.5
point 5 1.5 12.0
point 6 1.75 11.5
point 7 2.6 10.9
point 8 2.7 10.0
point 9 2.6 9.0
point 10 2.4 8.0
point 11 1.9 6.0
point 12 1.1 4.0
point 13 0.1 1.4
point 14 3.0 1.8
point 15 4.5 2.0
point 16 6.1 2.2
point 17 6.7 5.6
point 18 7.4 9.56
point 19 7.35 10.4
point 20 7.7 13.0
point 21 7.6 15.4
SET1 101 1 thru 21
$
$ center branch
$15 point 31 4.5 2.0
point 32 5.5 2.8
point 33 6.8 9.6
point 34 6.6 15.1
SET1 103 15 32 33 34
$
211
CHAPTER 7
$ Isotropic case
PBMSECT 31 1 OP 0.015
OUTP=101,T=0.1,brp=103
$
$ Composite case
PBMSECT 32 1 OP 0.015
OUTP=101,C=101,brp=103,c(1)=[201,pt=(15,34)]
pcomp 101 -0.1
501 0.05 0.0 501 0.05 90.0
501 0.05 -45.0 501 0.05 45.0
501 0.05 0.0
pcomp 201 SYM
501 0.05 -45.0 501 0.05 45.0
501 0.05 0.0
MAT1 501 3.+6 .3
.
Vabcore1 uses default value for PARAM,ARBMPS and PARAM,ARBMFEM. Finite element model for
composite beam of PBMSECT,32 is shown as follows:
The properties of Composite PBMSECT,32 computed as a 6x6 D matrix. A PBEAM3 is created
internally and printed in f06 if requested. The equivalent PBEAM3 for PBMSECT,32 is shown as
follows:
*** USER INFORMATION MESSAGE 4379 (IFP9B)
THE USER SUPPLIED PBMSECT BULK DATA ENTRIES ARE REPLACED BY THE FOLLOWING PBEAM3 ENTRIES.
CONVERSION METHOD FOR PBARL/PBEAML - .
PBEAM3 32 0 4.7202E+00 8.3059E+01 2.9578E+01 -1.5664E+01 3.2316E+01 0.0000E+00
1.8014E+01 4.2136E+00 1.7100E+01 -2.7858E+00 3.8881E+00 -3.5404E+00 4.7202E+00 2.6994E+00
0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00
0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00
0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00 0.0000E+00
0.0000E+00 0.0000E+00
1.2253E+08 -2.1160E+05 8.1193E+04 -2.4761E+06 -3.7193E+06 7.9040E+06 -2.1160E+05 2.1792E+06
-1.7859E+06 1.9780E+07 5.4643E+05 -3.5845E+05 8.1193E+04 -1.7859E+06 2.7228E+07 1.7190E+07
2.9835E+04 2.1407E+06 -2.4761E+06 1.9780E+07 1.7190E+07 2.2332E+08 5.8182E+06 -1.2186E+06
-3.7193E+06 5.4643E+05 2.9835E+04 5.8182E+06 2.1349E+09 -4.0706E+08 7.9040E+06 -3.5845E+05
2.1407E+06 -1.2186E+06 -4.0706E+08 7.5602E+08

212
The 6x6 D matrix starts in the Wi(j) section of PBEAM3 (7
th
line of above PBEAM3). Note that above
PBEAM3 is for information only. It does not have a punched version for later use. In addition, the 0
for MID field is probably not suitable for MD.Nastran input file processor, IFP.
The stress output for composite CBEAM3, if PARAM,ARBMSS,YES is in the input, is shown as
follows:
References
1. Palacios R, Asymptotic Models of Integrally-Strained Slender Structures for High-Fidelity
Nonlinear Aeroelastic Analysis, Ph.D. Dissertation, 2005
2. Cesnik CES, Palacios R, UM/VABS Theoretical Manual, Release 1.20, May 2004
TRANSVERSE TIP LOAD
0 SUBCASE 1

2 302 2 2.468E+01 1.601E+01 2.570E+00 2.323E+01 4.991E-01 3.724E+00 TSAI-WU 7.161E-04 4.035E+02
102 2 1.685E+01 1.610E+01 -7.230E-01 1.993E+01 -1.377E-01 -5.572E-01 TSAI-WU 7.258E-04 4.470E+02
1301 2 1.558E+01 1.594E+01 -7.167E-01 1.938E+01 -1.162E-01 -5.280E-01 TSAI-WU 7.193E-04 4.569E+02
LATERAL TIP LOAD
0 SUBCASE 2

2 302 2 1.206E+02 1.181E+02 -4.676E+00 1.455E+02 -1.072E+00 -7.170E+00 TSAI-WU 5.479E-03 6.114E+01
102 2 1.242E+02 1.185E+02 -4.690E+00 1.473E+02 -1.119E+00 -7.235E+00 TSAI-WU 5.500E-03 6.059E+01
1301 2 1.224E+02 1.183E+02 -4.683E+00 1.464E+02 -1.096E+00 -7.202E+00 TSAI-WU 5.490E-03 6.086E+01
213
CHAPTER 7
Nonhomogeneous Multipoint Constraint
Introduction
In the past, if a nonhomogeneous multipoint constraint was desired then the method of SLACK
variable was required. That is, the nonhomogeneous right hand side of the MPC equation was written
using a scalar or grid point and an SPC or SPCD specifying the right hand side value. In this release a
MPCY Bulk Data entry is introduced that allows the user to enter in a right hand side value directly into
the MPC equation. The entry defines an equation of the form
Associated with the MPCY entry is another new Bulk Data entry MPCD used to define a load selectable
value for of nonhomogeneous multipoint constraint.
Benefits
The user can define a nonhomogeneous multipoint constraint directly.
Input
1. The MPCY Bulk Data entry is used to define a nonhomogeneous multipoint constraint directly.
This entry can also be used to define a standard homogenous multipoint constraint as well.
2. The MPCD Bulk Data entry is used to define a load selectable value for of nonhomogeneous
multipoint constraint.
Output
Standard MD Nastran MPCF output is available.
Currently this method is not supported in the Dynamic solution sequences because the dynamic load
generation modules have not yet been taught to automatically expand into the required number of
load columns.
Theory
The basic relations are as follows:
The unconstrained stiffness matrix equation in MD Nastran is
A
m
u
m
A
i
u
i
i
+ Y
m
=
Y
m
Y
m
Y
m

214
(7-1)
If we apply a nonhomogeneous MPC constraint
(7-2)
Then Eq. (7-1) becomes
(7-3)
where and are the forces of constraint necessary to impose Eq. (7-2).
Partition Eq. (7-2) as
(7-4)
solve for to get
(7-5)
where and
Substitute Eq. (7-5) into Eq. (7-3) to get
(7-6)
or
(7-7)
Since the of Eq. (7-5) are prescribed, any virtual variation of Eq. (7-5) the results by definition of
. Hence the standard conjugate force transformation holds. Or
K
gg
[ ] U
g
{ } P
g
{ } =
R
g
[ ] U
g
{ } Y
{ } =
K
NN
K
NM
K
MN
K
MM
U
N
U
M

P
N
Q
N
+
P
M
Q
M
+

=
Q
N
Q
M
R
N
R
M
U
N
U
M

Y
{ } =
U
M
U
M
{ }
M
G
MN
U
N
+ { } =
M
R
M
1
Y
Z G
MN
R
M
1
R
N
Z
K
MN
K
MM
G
MN
+ [ ] U
N
{ } P
M
Q
M
K
MM
M
+ { } =
Q
M
{ } K
MN
K
MM
G
MN
+ [ ] U
N
{ } P
M
{ } = K
MM
M
{ } +

current new
M
0 Z
215
CHAPTER 7
(7-8)
where the minus sign comes from the fact we are imposing constraints and that forces are reactive to
the constraint forces.
Then the upper Eq. (7-3) with Eq. (7-8) becomes
and using Eq. (7-7) to expand the result to get
or
(7-9)
Examples
As a simple example consider a horizontal rod structure. The left and right ends are clamped. However
as Figure 7-3 shows, there is a break in the structure of 0.1 units of length. If the grid id at the left end of
the break is and the grid id at the right end of the break is , we can tie the structure by the MPC
equation
In the example, we show how to write the MPC equation in two ways. The first way uses the standard
SLACK variable method. The second way uses the new MPCY entry.
Figure 7-3 Example Rod Structure for MPCY
SOL 101
CEND
TITLE = DEMONSTRATE USE OF MPCY
SPC = 300
MPC = 300
LOAD= 300
DISPL= ALL
OLOAD= ALL
MPCFO = ALL
Q
N
{ } G
MN
T
Q
M

=
Q
N
Q
M
K
NN
U
N
K
NM
G
MN
U
N

M
+ [ ] + { } P
N
G
MN
T
Q
M

=
K
NN
K
NM
G
MN
G
MN
T
K
MN
G
MN
T
K
MM
G
MN
H H H [ ] U
N
{ } P
N
G
MN
T
P
M
H { } Z K
NM
G
MN
T
K
MM
[ ]
M
{ }
K
NN
[ ] U
N
{ } P
N
{ } =

current
K
NM
G
MN
T
K
MM
+ [ ]
M
{ }

new
A B
U
B
U
A
0.1 =

216
SPCFO= ALL
ELFOR = ALL
BEGIN BULK
$
PROD 1 1 1.
MAT1 1 1.+7 0.
GRDSET 3456
$
$
$ STANDARD SLACK VARIABLE METHOD
$
$ |1 2 3 4 5|
$ |O---------------O---------------O O---------------O|
$ | 1 2 3 |
$
$ GRIDs 3 and 4 have a 0.1 unit gap between them
$ We wish to impose relationship in x-direction
$ U4 = U3 + 0.1
$ When GRID 3 has a x-direction load of 1.+5
$
GRID 1 0. 0. 0. 123456
GRID 1 100. 0. 0. 23456
GRID 1 200. 0. 0. 23456
GRID 1 200.1 0. 0. 23456
GRID 1 300.1 0. 0. 123456
$
SPOINT 6
$
CROD 1 1 1 2
CROD 2 1 2 3
CROD 3 1 4 5
$
FORCE 300 3 1.+5 1.
$
SPC 300 6 1 0.1
MPC 300 4 1 -1. 3 1 1.
6 1 1.
$ ^
$ |
$ |
$ |
$ SPOINT with 0.1 in displacement gap
$
$
$ ----------------> NEW MPCY METHOD <----------------
$
$ |11 12 13 14 15|
$ |O---------------O---------------O O---------------O|
$ | 11 12 13 |
$
$ GRIDs 13 and 14 have a 0.1 unit gap between them
$ We wish to impose relationship in x-direction
$ U14 = U13 + 0.1
$ When GRID 13 has a x-direction load of 1.+5
$
GRID 11 0. 1. 0. 123456
GRID 12 100. 1. 0. 23456
GRID 13 200. 1. 0. 23456
GRID 14 200.1 1. 0. 23456
GRID 15 300.1 1. 0. 123456
217
CHAPTER 7
$
CROD 11 1 11 12
CROD 12 1 12 13
CROD 13 1 14 15
$
FORCE 300 13 1.+5 1.
$
MPCY 300 14 1 1. 0.1
13 1 -1.
ENDDATA
The results are:

1 G 0.0 0.0 0.0 0.0 0.0 0.0
2 G 3.000000E-01 0.0 0.0 0.0 0.0 0.0
3 G 6.000000E-01 0.0 0.0 0.0 0.0 0.0
4 G 7.000000E-01 0.0 0.0 0.0 0.0 0.0
5 G 0.0 0.0 0.0 0.0 0.0 0.0
6 S 1.000000E-01
11 G 0.0 0.0 0.0 0.0 0.0 0.0
12 G 3.000000E-01 0.0 0.0 0.0 0.0 0.0
13 G 6.000000E-01 0.0 0.0 0.0 0.0 0.0
14 G 7.000000E-01 0.0 0.0 0.0 0.0 0.0
15 G 0.0 0.0 0.0 0.0 0.0 0.0

L O A D V E C T O R

3 G 1.000000E+05 0.0 0.0 0.0 0.0 0.0
6 S 0.0
13 G 1.000000E+05 0.0 0.0 0.0 0.0 0.0
F O R C E S O F S I N G L E - P O I N T C O N S T R A I N T

1 G -3.000000E+04 0.0 0.0 0.0 0.0 0.0
5 G -7.000000E+04 0.0 0.0 0.0 0.0 0.0
6 S 7.000000E+04
11 G -3.000000E+04 0.0 0.0 0.0 0.0 0.0
15 G -7.000000E+04 0.0 0.0 0.0 0.0 0.0

F O R C E S I N R O D E L E M E N T S ( C R O D )
ELEMENT AXIAL ELEMENT AXIAL
ID. FORCE TORQUE ID. FORCE TORQUE
1 3.000000E+04 0.0 2 3.000000E+04 0.0
3 -7.000000E+04 0.0 11 3.000000E+04 0.0
12 3.000000E+04 0.0 13 -7.000000E+04 0.0

F O R C E S O F M U L T I P O I N T C O N S T R A I N T

3 G -7.000000E+04 0.0 0.0 0.0 0.0 0.0
4 G 7.000000E+04 0.0 0.0 0.0 0.0 0.0
6 S -7.000000E+04
13 G -7.000000E+04 0.0 0.0 0.0 0.0 0.0
14 G 7.000000E+04 0.0 0.0 0.0 0.0 0.0

* * * END OF JOB * * *

218
Ch. 8: Optimization MD R2 Nastran Release Guide

8
Optimization
Topology Optimization Enhancements
Automatic External Superelement Optimization (AESO)
Randomization of a Users Input Data File (Pre-release)
Random Elimination of Element Types (Pre-release)


220
Topology Optimization Enhancements
Introduction
Topology optimization capability was first released in 2005 and a number of manufacturability
constraints were added to MD Nastran R1. New features have been added to this release based on
feedback from clients. With these enhancements, MD R2 Nastran SOL200 is able to support: combined
topology, sizing, and shape optimization, multiple mass reduction targets, cyclical symmetry constraints,
and adjoint design sensitivity analysis for inertia relief of static analyses. A major performance
enhancement of module DOPR1 has been made to speed up minimum member size control and sizing
optimization with many thousands variables.
Benefits
Combined Topology, Sizing, and Shape Optimization
It is often recommend that topology optimization is first used to find efficient design concepts at the early
design stage whereas sizing and/or shape optimization is used for detail design based on the topology
design proposals at a later design stage. The use of topology, sizing, and shape optimization
simultaneously may find possible better design since the interaction of sizing and/or shape variables with
topology optimization is considered during a single design optimization process. Another benefit of this
feature is that the DRESP2 BETA function (minimize the maximum responses) is now available to
topology optimization.
Cyclical Symmetry Constraints
A mirror symmetry constraint was added to MD Nastran R1. It is also desirable to design a rotational
symmetric component or system. By using cyclical symmetry constraints in topology optimization, a
rotational symmetric design can be obtained regardless of the boundary conditions or loads. This cyclical
symmetric constraint capability can be used for irregular finite element meshes.
Adjoint Design Sensitivity Support for Inertia Relief
A direct design sensitivity analysis method is only available for inertia relief of static analyses in previous
versions. The direct method is not affordable computationally for topology optimization since many
thousands of variables are often involved. The adjoint design sensitivity analysis method has been
developed to benefit not only inertia relief topology optimization but also inertia relief sizing
optimization with many design variables. The adjoint method is automatically selected when it is more
efficient computationally.
Multiple Mass Target
Type one response FRMASS (DRESP1=FRMASS) used to be total fraction mass of topological designed
properties. Feedback from industrial users have shown that it is desirable to set up different mass
221
CHAPTER 8
Optimization
reduction targets on multiple designed parts for a built-up structures. This enhancement reflects this
requirement.
Module DOPR1 Performance Enhancements
When a large value is given for minimum member size TDMIN, a significant CPU time is spent on
module DOPR1 in previous versions. With this enhancement, a substantial performance speedup (3-100
times) is achieved for minimum member size control. Module DOPR1 is also enhanced to efficiently
support optimization problems with many thousands sizing design variables (for example, a sizing
optimization deck with 320,000 design variables, the DOPR1 module in previous MSC Nastran versions
and other Nastran products requires 16,000 seconds, MD R2 Nastran DOPR1 module requires only 10
seconds).
Input
The TOPVAR Bulk Data entry has been enhanced to provide cyclical symmetry constraints. To select a
topologically designable region, the user needs to specify a group of elements by using a Bulk Data entry,
TOPVAR. The cyclical symmetry constraints are then applied on all elements referencing the given
property on TOPVAR entry.
The enhanced TOPVAR format is:
N 2 3 4 5 6 7 8 9 10
TOPVAR ID LABEL PTYPE XINIT XLB DELXV POWER PID
SYM CID MSi MSi MSi CS NCS
CAST CID DDi DIE
EXT: CID EDi
TDMIN TV
Field Contents
PID Property entry identifier (Integer > 0). This PID must be unique for PIDs
referenced by other TOPVAR, DVPREL1 and DVPREL2 entries. Topology
and sizing variables cannot share the same properties. (Integer > 0)
SYM Indicates that this line defines symmetry constraints.
CID Rectangular coordinate system ID used for specifying manufacturing
constraints. See Remark 2. (Blank or Integer > 0; Default = 0)
CS Cyclical symmetry axis (character X, Y, Z). See Remark 3.
NCS Number of cyclical symmetric segments in 360 degrees. (Integer > 0). The
angle for one segment is calculated by 360/NCS. See Remark 3..

222
New Remarks:
1. The topologically designable element properties include PROD, PBAR, PBARL, PBEND,
PBEAM, PBEAML, PSHELL, PSHEAR, PSOLID, and PWELD. Multiple TOPVARs are
allowed in a single file. Combined topology, sizing, and shape optimization is supported in a
single file. However, TOPVAR cannot be used with DVMREL1 and DVMREL2 entries.
2. Only CORD1R and CORD2R can be used as a referenced coordinate system to specify topology
manufacturing constraints. Only one reference coordinate system CID is allowed for each
TOPVAR entry.
3. The first cyclical symmetry segment starts at the X-axis when CS=Z (at Z-axis when CS = Y, and
at the Y-axis when CS = X). One cyclical symmetry constraint can be combined with one mirror
symmetry constraint as long as the axis of cyclic symmetry is normal to the plane of mirror
symmetry. For example, MSi = YZ and CS = X, MSi = XZ and CS = Y, and MSi = XY and CS
= Z. This feature can also be used for rotational parts with < 360 degrees but NCS must be given
in 360 degrees.
4. For EXT constraints, possible combinations are (ED=X, MSi=XY, and/or ZX or CS=X),
(ED=Y, MSi=YZ, and/or XY or CS=Y), (ED=Z, MSi=ZX, and/or YZ or CS=Z).
5. For CAST constraints, possible combinations are (DD=X or X-, MSi=XY and/or ZX or CS=X),
(DD=Y or Y-, MSi=YZ and/or XY or CS=Y), (DD=Z or Z-, MSi=ZX and/or YZ or CS=Z).
Modified Type One Responses - Fractional Mass
To allow the user to put different mass reduction constraints on multiple designed properties,
RTYPE=FRMASS entry has been extended to provide property ID (PID) at attribute field ATTi. The
format for this extension is shown in Table 8-1.
Remark:
1. RTYPE= FRMASS (mass fraction of topological designed elements) entries are used for
topology optimization or combined topology, sizing and or shape optimization. ATTi=Blank is
for total mass fraction of all topological designed properties. ATTi=PID is the mass fraction of
topological designed property PID.
Table 8-1 Modified RTYPE=FRMASS
Response Type
(RTYPE)
Response Attributes
ATTA (Integer >
0)
ATTB (Integer > 0 or
Real > 0.0) ATTI (Integer > 0)
FRMASS
Remark 1
BLANK BLANK BLANK or Property
ID (PID)
223
CHAPTER 8
Optimization
Although combined topology and sizing optimization is supported, TOPVAR and DVPREL1/2
entries cannot reference the same property ID (PID).
While FRMASS is calculated for topological designed properties only, RTYPE=WEIGHT
computes total weight including all designed and non-designed parts. For combined topology
and sizing/shape optimization problems, it is recommended that RTYPE=FRMASS is used for
topological designed property mass reduction constraints and RTYPE=WEIGHT is used for
total mass reduction constraints.
Example 1 Wheel (wheeltop.dat)
A wheel model shown in Figure 8-1 is used to demonstrate MD Nastran topology optimization cyclical
symmetry capabilities. The wheel is modeled with six-sided solid elements (CHEXA). The wheel outer
layers and bolts are non-designable. One load case is considered. The structural compliance is minimized
(i.e., minimize the total strain energy of the structure) with a mass target 0.1 (i.e., remove 90% of the
material). Although the load is not cyclically symmetric about the Y-axis, the design is required to be
cyclically symmetric about the Y-axis with five segments.
Figure 8-1 Wheel

224
Input
The input data for this example related to topology optimization model is given in Listing 8-1. The
coordinate system CORD2R = 1 is created to be used to specify cyclical symmetric constraints. The field
CS on the SYM line is Y-axis with NCS=5.
Listing 8-1 Input File for Example 1
DESOBJ = 10
DESGLB = 1
ANALYSIS = STATICS
SMETHOD = ELEMENT
SUBCASE 1
SPC = 2
LOAD = 2
BEGIN BULK
CORD2R 1 10.512 33.3312 12.9921 -22.209833.3312 4.88385
28.388 33.3313 -19.7297
DCONSTR 1 2 .1
TOPVAR 1 PSOLID PSOLID .1 2
SYM 1 Y 5
DRESP1 2 FRM FRMASS
DRESP1 10 COMP COMP
Output
Figure 8-2 shows the topology optimized result that is smoothed by using MD Patran. It is noticed that
cyclical symmetry is obtained even though the loading is not cyclically symmetric.
225
CHAPTER 8
Optimization
Figure 8-2 Wheel Topology Design

226
Automatic External Superelement Optimization
(AESO)
Introduction
The Manual External Superelement Optimization capability (MESO) has been available since the
MSC.Nastran 2004. In this technique, the user manually partitions the analysis model into two parts: a
designed part and a non-designed part. The latter is treated as one external part superelement while the
former is defined as a residual structure. A creation run is performed which applies Component Mode
Synthesis (CMS) or Static Condensation to the part superelement and stores the resulting boundary
matrices in a database or a punch file. The original optimization task is performed on the assembly run
that assembles those boundary matrices into the residual model for solving system solutions. The strategy
is most efficient when the size of the design model (or residual model) is much smaller than the size of
the original analysis model. However, although the feature is efficient in CPU time, since both files of
both creation and assembly runs must be created by the user, significant effort in manual partitioning the
model might outweigh the performance gain.
The AESO capability presented here extends the MESO in an important way: rather than requiring the
user to segregate his large model into a designed and non-designed part, the process does this
automatically by identifying which parts of the finite element model are affected by the design task. In
essence, the new AESO capability provides an efficient and accurate solution in a user friendly way.
Benefits
Several major benefits are:
1. Removes a tedious and error prone task from the user in preparing the user input data.
2. Does not require the user to be knowledgeable in the specialized area of superelements in general
and external superelements in particular.
3. Provides an efficient and accurate approach for large-scale design optimization tasks.
4. Enable the performance of various design studies rapidly once the model has been divided into a
designed and non-designed part. Examples of this are the setting up of different design constraints
and objective in the studies to gain insight into the design and the available trade-offs or the
applying of various frequency excitation loadings in the frequency response analyses.
Methodology
A complete AESO task involves two separate MD Nastran job runs: 1) the first run is an AESO creation
run (or simply creation run) whose logical flow is described in Figure 7-3 and 2) the second run is an
AESO assembly run (or simply assembly run) whose logical flow is described in Figure 7-5.
As shown in Figure 7-3, the creation run automatically partitions the original analysis model into the
residual (the designed part) and external SE (the non-designed part). This automatic partition procedure
will assign the following grid points to the residual:
227
CHAPTER 8
Optimization
1. All grid points that belong to a design model consisting of DRESP1, DVGRID, DVPRELi,
DVMRELi and DVCRELi entries;
2. All grid points that are referenced on all static or dynamic loading entries such as DAREA,
DPHASE, FORCE, MOMENT, PLOADi, TEMP entries;
3. All grid points for a rigid element that has one or more connecting grid belonging to the residual;
4. Any grid point that is a dependent grid on an MPC entry
After the automatic partition procedure, a new user input file is created from the residual (Figure 7-4).
Then, the remainder of the creation run is the application of Static Condensation and/or Component
Modes Synthesis procedures to produce stiffness, mass, damping boundary matrices. After the creation
run is complete, a Nastran database is saved to store the boundary matrices and a .asm file is also created
to include superelement boundary connection information.
The assembly run is similar to a conventional SOL 200 task as shown in Figure 7-5 by utilizing all three
types of data generated from the creation run (Figure 7-4). The original optimization problem is solved
by assembling the boundary matrices into the residual for the system solutions.
Input
1. Two parameters are added on the DOPTPRM entry for an AESO job:
AUTOSE - flag to request an AESO job (integer 0, 1, Default = 0). AUTOSE = 1 activates an
AESO creation run.
DRATIO - the threshold value that is used to turn off an active AESO job if the ratio of the size
of the design model to that of the analysis model is greater than DRATIO (Real > 0.; Default =
0.1).
Optimizer
Automatic
Model Partition
ExternaI SE
CMS to create
boundary matrices
ResiduaI
MSC.Nastran
EXTSE
Database
.asm file
k=f=c=
========
^=o=

|nitiaI User
|nput FiIe
|nitiaI User
|nput FiIe
Figure 7-3 AESO Creation Run Figure 7-4 Output of
Creation Run and
Input of Assembly
Run
Figure 7-5 AESO Assembly Run

228
2. An ASSIGN statement with a logical key name AESO is placed in the FMS to specify an input
file name for the assembly run.
To illustrate ideas behind the input and output for an AESO task, a test problem (aeso1.dat) is used here.
Figure 8-6 shows a sample model whose upper left portion covering elements 18 to 42 (SE 1) is the non-
designed part while the rest of structure is the residual structure.
The listing below is a condensed version of the creation run file (aeso1.dat) that only shows the required
user input to invoke an AESO creation run. The assign aeso='aeso1_2.dat' statement is specified in the
FMS section and autose 1 and dratio 0.9 are requested on the DOPTPRM entry. The DRATIO=0.9 here
overrides the default.

Figure 8-6 A Sample Model
SE1
229
CHAPTER 8
Optimization
Figure 8-7 Condensed Version of the Creation Run File, dseoptl.dat
Outputs
As shown in Figure 7-4, three types of data generated from the creation run are saved in the working
directory: a Nastran database, a .asm file and a new input file for the assembly run (or an assembly file).
This section describes each of these items and explains how they are used in the assembly run. In
addition, some special print outputs from the creation run are shown that display the model partition
information.
Nastran Database Files
Two Nastran database files: aeso1.MASTER and aeso1.DBALL are automatically saved when the
creation run is submitted with SCR=NO option. Notice the size of the database is much smaller than the
regular Nastran database because only boundary matrices are stored. The Master file will be referenced
by an ASSIGN statement in the new input file described below.
The .asm File
The aeso1.asm file that includes the boundary connection information follows. As shown in Figure 8-6,
this problem has only two boundary points, 35 and 36 between the residual and the external superelement
1. This file is accessed through an INCLUDE command in the assembly file.
assign aeso='aeso1_2.dat'
SOL 200
CEND
desobj(max) = 1
analysis = modes
..
Begin Bulk
doptprm desmax 10 autose 1 dratio 0.9

enddata

230
New Input File for the Assembly Run
Notice that the name of this file, aeso1_2.dat is specified on the ASSIGN AESO statement in the creation
run (Figure 8-7). It is a standard Nastran input file. The AESO specific contents in the Executive Control,
Case Control and Bulk Data Sections are listed in Figure 8-8 and are described below.
Figure 8-8 Highlights of the Assembly Run File, aeso1_2.dat
1. Executive Control Section
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$
$ ASSEMBLY PUNCH (.ASM) FILE FOR EXTERNAL SUPERELEMENT 1
$
$ THIS FILE CONTAINING BULK DATA ENTRIES PERTAINING TO
$ EXTERNAL SUPERELEMENT 1 IS MEANT FOR INCLUSION
$ ANYWHERE IN THE MAIN BULK DATA PORTION OF THE ASSEMBLY RUN
$
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
SEBULK 1EXTERNAL MANUAL
$
SECONCT 1 0 5.0E-05
35 35 36 36
$
$ BOUNDARY GRID DATA
$
GRID 35 -3.6 6. 0.
GRID 36 -2.8 6. 0.
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
nastran rseqcont=1
assign se1= './aeso1.MASTER'
dblocate datablk(EXTDB) logical=se1,
CONVERT(SEID=1)
SOL 200
CEND
Executive Control Section
desobj(max) = 1
analysis = modes
.
subcase 10
method = 1
spc = 10
$
Case Control Section
begin bulk
include './aeso1.asm'
.
doptprm desmax 10
enddata
Bulk Data Section
}
}
}
231
CHAPTER 8
Optimization
The Nastran rseqcont=1 statement instructs the input file processor to ignore all continuation
fields. This statement is automatically created in this file regardless of whether the creation run
has it or not.
The next two statements assign the Nastran Master database file and locate the EXTDB datablock
that stores various boundary matrices.
Notice that the other statements in Executive Control Section of the creation run are not retained.
2. Case Control Section
The whole Case Control section of the creation run is retained in the assembly file.
3. Bulk Data Section
This section completely defines the residual structure. The include ./aeso1.asm command
allows the assembly run to access the .asm file created from the creation run. In addition, the
autose 1 and dratio 0.9 have been removed from the doptprm entry of the creation run.
Special Print Outputs from Creation Run
The following output is taken from the aeso1.f06 file. It displays detailed information about the model
partition. You may use Figure 8-6 to help read the printout here. Notice that the Superelement 1 covers
the non-designed part while the residual (or Superelement 0) covers the designed part.

232
Listing 8-2 Printout Showing Model Partition of Designed and Non-Designed Parts
You may adjust DRATIO to allow an assembly run with larger or smaller residual model. The
UIM 7824 provides brief information about the sizes of your analysis model and design model in
terms of number of the grid points.
BOUNDARY SEQUENCE ASSIGNMENT TABLE
BOUNDARY
SEQUENCE ID -------------------------- ASSIGNED TO POINT ID (SUPERELEMENT) ---------------------------------
1B 35 ( 0) 35 ( 1)
2B 36 ( 0) 36 ( 1)
SUPERELEMENT 0
LIST OF INTERIOR POINTS ( TOTAL NO. OF INTERIOR POINT = 70 )
INDEX -1- -2- -3- -4- -5- -6- -7- -8- -9- -10-
1 1 2 3 4 5 6 7 8 9 10
11 11 12 13 14 15 16 17 18 19 20
21 21 22 23 24 25 26 27 28 29 30
31 31 32 39 40 41 42 43 44 51 52
41 53 54 55 56 63 64 65 66 67 68
51 75 76 77 78 79 80 87 88 89 90
61 91 92 99 100 101 102 103 104 1B 2B
SUPERELEMENT 0
LIST OF ELEMENTS ( TOTAL NO. OF ELEMENTS = 42 )
INDEX -1- -2- -3- -4- -5- -6- -7- -8- -9- -10-
1 1 2 3 4 5 6 7 8 9 10
11 11 12 13 14 15 16 17 43 44 45
21 46 47 48 49 50 51 52 53 54 55
31 56 57 58 59 60 61 62 63 64 65
41 66 67
SUPERELEMENT 1
LIST OF EXTERIOR POINTS ( TOTAL NO. OF EXTERIOR POINT = 2 )
INDEX -1- -2- -3- -4- -5- -6- -7- -8- -9- -10-
1 1B 2B
SUPERELEMENT 1
LIST OF INTERIOR POINTS ( TOTAL NO. OF INTERIOR POINT = 34 )
INDEX -1- -2- -3- -4- -5- -6- -7- -8- -9- -10-
1 33 34 37 38 45 46 47 48 49 50
11 57 58 59 60 61 62 69 70 71 72
21 73 74 81 82 83 84 85 86 93 94
31 95 96 97 98
SUPERELEMENT 1
LIST OF ELEMENTS ( TOTAL NO. OF ELEMENTS = 25 )
INDEX -1- -2- -3- -4- -5- -6- -7- -8- -9- -10-
1 18 19 20 21 22 23 24 25 26 27
11 28 29 30 31 32 33 34 35 36 37
21 38 39 40 41 42
*** USER INFORMATION MESSAGE 7824 (DSGRDM)
THE NUMBER OF GRID POINTS IN THE ANALYSIS MODEL = 104.
THE NUMBER OF GRID POINTS IN THE DESIGN MODEL = 70.
THE DESIGN MODEL COMPRISES 67.3 PERCENT OF THE ANALYSIS MODEL.
233
CHAPTER 8
Optimization
For an AESO job with Analysis=MODES or MFREQ, it is recommended to activate the matrix
domain based decomposition with domainsolver acms(partopt=dof) in the Executive Control
section to speed up the CMS procedure.
Always specify the ASSIGN AESO=filename.ext statement in the creation run to define the
name of the assembly file. Directly assigning the original job name to filename should be
avoided and will cause the assembly run to fail with User Fatal Message 713. A good practice is
to add some suffix to the original file name such as myjob_2nd.dat where myjob is the original
file name.
When submitting the AESO creation run, use SCR=NO option. Otherwise, the Nastran database
will not be retained after the creation run is done. However, it is optional for submitting an
assembly run.
After the creation run is complete, check the following user information message in the f06 file
to ensure the job is terminated successfully.
The assembly input file may be modified to perform various design studies as long as the
changes do not affect the boundary matrices stored in the database.
If the AESO creation run includes a GRAV entry, it will be terminated with the following
message. The same applies to TEMPD entry. They should be removed from the file if they are
temporary inactive for the current task. Gravity and TEMPD loads are not supported with AESO
since they apply loads to the entire structure and therefore block any partitioning.
Duplicate GRID entries are allowed in the .asm file and in the assembly input file. They will be
automatically removed during the assembly run within the location tolerance specified by the
TOL field on the SECONCT entry. For the AESO jobs, the default of the location tolerance has
been increased to 5.E-5 from the original 1.E-5. However, due to numerical imperfection, this
tolerance may need to be adjusted particularly for the cases in which the boundary grid points
are defined in one or multi-levels of coordinate systems.
*** USER FATAL MESSAGE 732 (OPFUNT)
LOGICAL NAMES 'INPUT ' AND 'AESO ' ARE ASSIGNED TO THE SAME PHYSICAL FILE.
USER INFORMATION: PHYSICAL FILE NAME 1: ./abc.aeso
PHYSICAL FILE NAME 2: ./abc.aeso
USER ACTION: CHANGE FILE NAME ON ONE OF THE ASSOCIATED ASSIGN STATEMENTS.
^^^
^^^ USER INFORMATION MESSAGE 9181 (FEA)
^^^ THE JOB IS TERMINATED FOR AN AUTO EXTERNAL CREATION RUN
^^^
*** USER FATAL MESSAGE 7699 (DSGRDM)
A GRAV Bulk Data entry is specified in an AESO creation run.
USER INFORMATION: The AESO run does not support the GRAV entry.

234
Bulk Data parameter SEMAPPRT can be used to control the printout of the model partition
information shown in Listing 8-2. For example, Setting SEMAPPRT to 0 will turn off the
printout.
If the AESO task includes DVGRID entries, make sure that the grid points referenced by
DVGRID entries are inside the residual. Since the grid points on DVGRID entries vary during
the design process, including them as part of boundary grid points will invalidate the invariance
of those boundary matrices. Currently, the grid points on the DVGRID entries will not be
automatically assigned to be inside the residual. ASET and ASET1 entries can be used to create
an enclosure or a barrier to ensure the grid points referenced on the DVGRID entries are always
placed inside the residual.
Since all the CORD1i entries are automatically converted to CORD2i entries during the AESO
creation run. The DVGRID entry should not reference the grid points that define the CORD1i
entry.
The AESO tasks do not support acoustics response.
Examples
Road Response Optimization with CAMARO Model with Analysis=MFREQ
Figure 8-9 CAMARO Model
A CAMARO model provided by GM as shown in Figure 8-9 is modeled with 23K grid points, 37K
elements and 137K DOFs, random inputs applied on left and right suspension including cross-
correlation, to simulate road conditions.
The design task is to vary 9 spring constants of engine mounts modeled by elastic elements in two design
cases:
Case A: minimize the sum of RMS acceleration at Drivers seat and passengers seat and limiting the
PSD response at steering column and
Case B: minimize the RMS acceleration at Drivers seat and maintain frequency dependent limits on
drivers seat.
235
CHAPTER 8
Optimization
Because these design variables and responses are limited to a small part of the total vehicle, the size ratio
of design model vs. analysis model is <1.% (the residual structure has 298 elements, 163 interior points
and 130 boundary points while the external superelement has 33345 elements and 22761 interior points).
To demonstrate the efficiency of the AESO capability, the tables below present the results from a regular
run and the AESO job. First, the same final designs are achieved by both jobs. However, the AESO job
takes 1/5 of the time to complete Case A. Furthermore, since Case B just changes the design objective
and constraint formulation, the boundary matrices from the creation run are invariant. Therefore, only
the assembly run for Case B is required. Comparing the time spent on this assembly run with the single
run as shown in Case B in the bottom half of the table, the speed up is 16X.
Road Response Optimization with CAMARO Model with Analysis=DFREQ
In general, direct frequency approach is more expensive than the modal approach. However, when the
size of the residual model is small, the Direct Frequency approach can also be performed efficiently as
shown here. Case A of the same road response optimization problem are solved with both a regular direct
frequency response optimization and the direct frequency AESO. The table below presents the results.
Case A
Initial
OBJ
Final
OBJ
Init Max
Const.
Final Max
Const.
# Design
Cycle
Clock Time
(Minutes)
Full Model Run 0.1534 -0.0639 0.1329 -0.2102 9 76
AESO Creation
Run
N/A N/A N/A N/A N/A 10
AESO Assembly
Run
0.1534 -0.0639 0.1329 -0.2102 9 7
Total time of two
AESO runs
16
Performance
Ratio
5
Case B
Initial
OBJ
Final
OBJ
Init Max
Const
Final Max
Const
# Design
Cycle
Clock Time
(Minutes)
Full Model
Run
0.0713 0.0586 0.2855 0.0081 14 110
AESO
Creation
Run
N/A N/A N/A N/A N/A 0
AESO
assembly
Run
0.0713 0.0584 0.2855 -0.0108 14 7

236
Again, the same final designs are achieved. However, the speed up obtained by the AESO over the
regular job is 27X.
The following figure shows the plot of the total clock time spent by two runs vs. design cycle and it
clearly shows that the AESO is much more efficient than the regular run because the cost per design cycle
for AESO (or the slope of the curve for the AESO run) is much smaller than that for a regular run (the
slope of the curve for the regular run). Although for this example, the one time cost for the AESO due to
the creation run is smaller than that for a regular run, in general, this one time cost may be larger because
the time spent by CMS procedure on mass matrix reductions could be expensive.
Case A
Initial
OBJ
Final
OBJ
Init Max
Const
Final Max
Const
# Design
Cycle
Clock Time
(Minute)
Full Model Run 0.1535 0.1327 -0.0631 -0.2073 9 388
AESO Creation Run N/A N/A N/A N/A N/A 10
AESO Assembly Run 0.1534 0.1327 -0.0636 -0.2062 9 5
Total time of two
AESO runs
15
Performance Ratio 27
0
50
100
150
200
250
300
350
400
450
0 2 4 6 8 10
Design Cycle
C
l
o
c
k

T
i
m
e

(
M
i
n
u
t
e
s
)
Total Time (Regular Run) Total Time (AESO)
237
CHAPTER 8
Optimization
Miscellaneous
Prior to this release, SOL200 supports a limited Part optimization capability while it fully supports the
traditional Superelement optimization. Notice that the traditional SE is defined with SESET Bulk Data
entry while the Part is defined with Begin Super Bulk Data entry. In addition, the Part may be defined
either as External or Internal.
The enhancement added in this release provides a robust Part optimization capability as long as the
design model is in the residual (or an upstream Part can not be designed or constrained). One example
of its application would be: an airplane wing is modeled as an internal Part while the engine is modeled
as an external Part. Then these two are attached to the fuselage, a residual that will be designed.
The Table below summarizes the features of Superelement Optimization and Part Optimization in
SOL200 in this new release.
To illustrates how to create a Part optimization job, the sample model as shown in Fig 7-6 is used here.
The top left squared region is modeled as Part 1 (elements 18 to 42) while the top right squared region
is modeled as Part 2 (elements 43 to 67). The rest is modeled as a residual (elements 1 to 17). Grid points
1 and 2 are fully constrained. The design task is to maximize the first natural frequency by varying the
thickness of elements 1 to 17 within a given range.
The following listing is a condensed version of dseopt18.dat that shows three major Bulk Data sections.
Its full version can be obtained from the Test Problem Library. The unique feature of a Part optimization
job is the multiple Bulk Data sections and how the design model definition is placed. Three sections in
this job are:
1. Main section (between Begin Bulk and Begin Super = 1) that defines the residual model. Notice
that all the designed entries such as DESVAR, DVPREL1, DRESP1 entries are defined in this
section because the design model must be within the residual.
2. Section for Part 1 (between Begin Super = 1 and Begin Super =2). This section defines Part 1
model.
3. Section for Part 2 (between Begin Super = 2 and Enddata). It defines Part 2 model.
Notice that the design model must be placed in the main Bulk Data section because the design model is
required to be in the residual. The detailed steps to define Parts and residual model are not skipped here.
Item Feature Support in SOL 200
Design Model in Upstream
SE or Part
1 Traditional SE Yes Yes
2 Ext. Part Yes No
3 Int. Part Yes No
4 2+3 Yes No
1+2,1+3,1+2+3 No No

238
SOL 200
diag 8,15,56
CEND
TITLE = Test for Internal Part Optimization
echo = sort
desobj(max) = 1
analysis = modes
subcase 10
spc = 10
method = 1
$
$ residual structure model
$
BEGIN BULK
eigrl,1,,,9
param,post,0
PARAM,GRDPNT,0
PARAM,WTMASS,.00259
CQUAD4 5 1 13 14 24 23
$
$GRDSET 6
GRID 13 -.4 3.6 0.
GRID 14 .4 3.6 0.
GRID 23 -.4 4.4 0.
GRID 24 .4 4.4 0.
$
MAT1,1,30.+6,,.3,.283
PSHELL,1,1,.05,1,,1
$
CQUAD4 14 1 19 20 30 29
CQUAD4 15 1 29 30 36 35
$$
spc1,10,123456,1,2
$
$ design model definition must be in the main BULK Data section
$ 2 3 4 5
desvar 1 X1 0.05 0.01 2.0
dvprel11 pshell 1 T
1 1.0
DRESP1 1 F1 FREQ 1
Main Bulk
Data Section
239
CHAPTER 8
Optimization
$
$ Part 1
$
begin super = 1
$
$ define modal coordinates for CMS
$
$ define q-set for component modes and residual vectors
SPOINT 11001 THRU 11020
QSET1 0 11001 THRU 11020
$
$ define which dofs will be retained (i.e. which dofs
$ will form the attachment to the system model when we
$ create SE10 in se10.dat)
$
ASET1 123456 35 36
$
$ part1.dat
$
CQUAD4 18 9 33 34 46 45
CQUAD4 42 9 85 86 98 97
$
param,k6rot,100.
$
$ boundary grids
$
GRID 35 -3.6 6. 0.
GRID 36 -2.8 6. 0.
$
GRID 97 -2. 10. 0.
GRID 98 -1.2 10. 0.
$
MAT1,1,30.+6,,.3,.283
PARAM,WTMASS,.00259
PARAM,AUTOSPC,YES
PSHELL,9,1,.05,1,,1
$
$ plotels to outline component in assembly run
$
plotel,101,33,35
plotel,102,33,93
plotel,103,93,98
plotel,104,98,38
plotel,105,38,36
plotel,106,35,36
eigrl,1,,,9
Section for
Part 1 Model

240
$ Part 2
$
begin super = 2
$
$ define modal coordinates for CMS
$
$ define q-set for component modes and residual vectors
SPOINT 21001 THRU 21020
QSET1 0 21001 THRU 21020
$
$ define which dofs will be retained (i.e. which dofs
$ will form the attachment to the system model when we
$ create SE10 in se10.dat)
$
ASET1 123456 4142
$
$
$ part2.dat
$
CQUAD4 43 2 39 40 52 51

CQUAD4 66 2 90 91 103 102
CQUAD4 67 2 91 92 104 103
$
param,k6rot,100.
$
$ boundary grids
$
GRID 41 2.8 6. 0.
GRID 42 3.6 6. 0.
$

GRID 103 4.4 10. 0.
GRID 104 5.2 10. 0.
$
$
$
MAT1,1,30.+6,,.3,.283
PARAM,WTMASS,.00259
PARAM,AUTOSPC,YES
PSHELL,2,1,.05,1,,1
$
$ plotels to outline component in assembly run
$
plotel,201,39,41
plotel,202,39,99
plotel,203,99,104
plotel,204,104,44
plotel,205,44,42
plotel,206,41,42
$
eigrl,1,,,9
enddata
Section for
Part 2 Model
241
CHAPTER 8
Optimization
Randomization of a Users Input Data File (Pre-
release)
Introduction
The stochastic capability in MD Nastran is the first step towards a complete and automatic self-
randomization of a Finite Element model. The capability currently offers the user the possibility to
automatically distribute tolerances and uncertainties with minimum effort. This reduces dramatically the
burden on a user wishing to perform large-scale stochastic simulations. In fact, once the stochastic option
is triggered, the entire Bulk Data Deck is randomized automatically and without further user
intervention. The resulting model, which needs to be incorporated in a Monte Carlo Simulation loop -
there are numerous off-the-shelf products which support this capability - possesses unprecedented levels
of realism.
In order to make full use of this new development, it is necessary to resort to a multi-run environment,
which can spawn a certain number of independent MD Nastran executions, collect the results, allow the
user to perform statistical post-processing. With the self-randomization capability in MD Nastran, all the
user needs to define are the outputs he wishes to monitor, such as stresses, eigen-frequencies,
temperatures, displacements, etc. There is no need to define inputs, as these are defined automatically by
MD Nastran. The process is, essentially, error-free.
Benefits
A basic assumption of MD Nastran is that the inputs to the analysis are known exactly so that the
computed responses are also known exactly. This is, of course, an invalid assumption in that there will
always be some uncertainty in the inputs with a corresponding variations in the outputs. The MD R2
release of Nastran provides a way of introducing this uncertainty into the analysis process by
automatically randomizing user input real numbers based on the input values and statistical quantities
that characterize the variation.
Input
The randomization capability is driven by a new STOCHASTICS case control command as shown in the
Quick Release Guide. If STOCHASTICS=ALL is used, all real quantities on connectivity (those starting
with C), Material and Property entries as well as any loads and SPCD quantities are modified based on
a covariance factor of 0.05. A Gaussian distribution is used to randomly select the perturbed quantity
with the restriction that the value can be no more that a specified number of standards deviations from
the user input mean value. The default number of maximum standard deviations is 3.
Alternatively, the STOCHASTICS command can point to a STOCHAS bulk data entry that provides the
ability to selectively randomize different types of input quantities using user specified covariance values
and number of allowed standard deviations. In this case, only the types of input specified are randomized
so that, for example, it is possible to randomize the loads input while leaving the property values
unchanged.

242
Output
There is no new output produced by this capability at present.
The randomization algorithm involves using a random number generator, a Gaussian distribution, the
prescribed covariance and a mean value based on the user input to come up with a randomized value that
is to be used in the analysis. In order to avoid physically meaningless properties, the random value is
prescribed to be within m standard deviations of the input value, where m is a user input value with a
default value of 3.0.
The product of m * COV should not be greater than 1.0 to eliminate the possibility of the property
changing sign.
The full benefit of this beta capability requires submitting multiple runs with the same randomization
parameters. Each would produce a unique randomization and it is possible to collect the results of each
of these analyses and produce statistical information on the variability of the responses. At present, MSC
does not have software that performs these functions and it is unlikely that users would carry out this type
of multiple run analysis on their own.
If the user input property value is 0.0, no randomization occurs. If it is recommended that any property
values that are 0.0 (say orientation angle on a PCOMP entry) be set to some non-zero value that is not
negligible.
243
CHAPTER 8
Optimization
Random Elimination of Element Types (Pre-release)
Introduction
There has been a long time capability in Nastran that allows the user to specify the random elimination
of a specified percentage of the CWELD elements that are contained in a bulk data file. This was done
using Nastran PARAM CWRANDEL, with an additional CWDIAGP PARAM providing the option of
printing the IDs of the deleted elements. This capability has been extended to the CELASi, CFAST,
CSEAM, and 1-D mass (CMASSi, CONM1 and CONM2) elements. Further, the user interface has been
changed from the NASTRAN PARAM input to the MDLPRM entry.
Benefits
The ability to randomly delete various 1-D elements provides the user with some assessment of the
integrity of the design. For instance, if randomly deleting 20%, say, of the CWELDs from a model
caused a negligible change in the first ten natural frequencies, this was taken as an indication of the
robustness of the structure. Extending this to other element types provides that many more options in this
type of analysis. Placing the input on the MDLPRM entry consolidates that input so that the user does
not have to deal with the PARAM entry.
Input
The MDLPRM entry has 10 new PARAMi names that support this capability. Five of them (e.g,
DELELAS) select the element type to which this random elimination applies and the ratio to be deleted
while an additional five (e.g., PRTELAS) provide control as to whether the IDs of the deleted elements
are to be printed. The default is that the IDs will not be printed.
Output
There is no new output produced by this capability.
The deletion ratio is input as a real number between 0.0 and 1.0 with 0.0 indicating no deletion is to take
place while 1.0 would eliminate all elements of the specified type.
It is possible that the elimination of a series of elements will introduce mechanisms in the structure that
will cause the analysis to fail. It is the users responsibility to determine whether this failure has occurred.
A likely scenario for the use of this capability would be to submit the same deck multiple times and
determine in the variation in the results. MSC does not offer an automated way of doing this at this time.

244
Ch. 9: Rotor Dynamics & Aeroelasticity MD R2 Nastran Release Guide
9
Changes to Rotordynamics for MD R2 Nastran
Updating/Summing of Monitor Points
Stripwise Aerodynamic Results
Input of an Aerodynamic Mesh
Rigid Body Spline
Wendland Spline Functions for the Spline4/5
Spline Blending
Export of the Spline Matrix


246
Changes to Rotordynamics for MD R2 Nastran
Unbalance Entry for Frequency Response
The UNBALNC loading has been available for rotordynamic transient response for several versions.
Unfortunately, unbalance loading for frequency response had to be input using RLOADi entries. This
manual input was uncommonly prone to error. And backward (opposite the rotor spin direction) loading
was easily input. To prevent this problem and provide a more natural input, the UNBALNC Bulk Data
entry can now be used for frequency response with the rotordynamics option. The entry is selected using
the DLOAD Case Control command (note that the transient response solutions must still specify the
transient unbalance using the RGYRO Case Control command).
The format for the UNBALNC in frequency response is the same for transient response, except the
continuations for force output via EPOINTs are ignored. For further information on the UNBALNC Bulk
Data entry see the MD Nastran Quick Reference Guide.
New Parameter and Hybrid Damping Specifications
The specification of damping using parameter input (PARAM,G for structural damping and PARAMs
ALPHA1 and ALPHA2 for Rayleigh damping) often resulted in users not knowing what damping, if any,
was actually being specified in the model. This is because parameters can be entered anywhere in the
Case Control or Bulk Data Sections. For large models with many include files, a rigorous search would
be required. To make the damping specifications easier and also allow new damping formulations, such
as hybrid damping, a new damping entry has been developed that is selected in the case control. The new
entry allows selection of parameter damping, such as structural damping (PARAM, G) and Rayleigh
damping (PARAMs ALPHA1 and ALPHA2), and a new hybrid damping. This entry also allows the
scaling of material damping (GE on material entries, MATi). The new Bulk Data entry is named
DAMPING and is selected by new Case Control commands:
For specifying damping of superelements, use
SEDAMP= n
For specifying damping of the residual structure, use
RSDAMP(STRUCTURE, FLUID, or BOTH)= n
Default= STRUCTURE
Where n specifies the DAMPING Bulk Data entry.
For more information on the DAMPING Bulk Data entry see the MD Nastran Quick Reference Guide.
247
CHAPTER 9
Updating/Summing of Monitor Points
Introduction
MD R1 Nastran greatly expanded the concept of monitor points by modifying the MONPNT1 and
introducing the MONPNT2, MONPNT3 and MONDSP1. The MD R2 release of MD Nastran further
expands the monitor points by providing two additional capabilities
1. Allowing for the modification of existing component results for MONDSP1, MONPNT1 and
MONPNT3 by a scalar multiple to allow, for example, a change in sign or units.
2. Enabling the weighted summation or two or more MONDSP1, MONPNT1 or MONPNT3s that
are of the same type.
Benefits
The ability to update an existing monitor point result is done by applying factors to specified
components. An example of the utility of this feature is if the model has been constructed in one set of
units (e.g., centimeters) and it is desired to see the results in a different set (e.g, millimeters).
The ability to sum existing monitor point results could be used; e.g., to better present running results
along a wing or fuselage.
Input
The new MONSUM Bulk Data entry is used to implement both updating and summing of monitor point
results. The description of this entry in the MD Nastran Quick Reference Guide includes examples and
numerous remarks that help in understanding the capabilities of this entry.
Output
The output of the monitor point results from the MONSUM Bulk Data entry is identical to that of the
underlying monitor point. In the updating scenario, if the NAME field on the entry, is the same as the
NAMEij fields, only the final updated result is given. If a new NAME is used, both results are given.
Examples (monsum.dat and monsum3.dat)
Two small test files demonstrate the use of the MONSUM Bulk Data entry. They are both variations of
the familiar forward swept wing and are:
monusm.dat- contains some MONSUM Bulk Data entries that are simple enough that the results can be
checked by hand
monsum3.dat sums MONPNT3 results. An interesting feature of this example is that it demonstrates
the DMAP enhancements that were required to sum MONPNT3 results produced by using what is
referred to as the mini-ema method with results using the grid point force recovery method.

248
The remarks for the MONSUM Bulk Data entry in the MD Nastran Quick Release Guide provide a
number of guidelines that should be reviewed before applying this new entry. Because it performs both
an update-in-place and a combining function, it represents a powerful, compact capability.
The two examples shown in the MD Nastran Quick Reference Guide provide a further explanation of this
entry. In the first, a new monitor point is being synthesized from three existing monitor point results to
provide the user with a blended result that has special meaning to the designer. In this case, results at three
stations along the root chord are being summed with different weights for the desired shear and bending
moment resultants. Since the COEFi fields are left blank for the shear components, the default of 1.0 is
applied. In the second example, the MONSUM is used to convert the units of the monitor point output
from English to metric units. Note that in the first case, monitor point results will be presented for the
summed quantity as well as for each of the original MONPNT1 quantities that contribute to the sum. In
the second case, only the final result is presented since its NAME is shared with the NAMEij attributes.
If output in both units systems is desired, one needs to simply provide a unique NAME in order to get
both results.
249
CHAPTER 9
Stripwise Aerodynamic Results
Introduction
An additional monitor point, MONCNCM, has been provided that streamlines the task of providing
stripwise lift and pitching moment results for a doublet-lattice type of aerodynamic model.
Benefits
Stripwise aerodynamics; i.e., aerodynamic lift and pitching moments, can be quite useful in visualizing
the aerodynamic results on lifting surfaces. In particular, these can be compared with wind tunnel results
with the possibility of weighting the computed aerodynamics to match the test results. It is possible to
generate these results using an aerodynamic MONPNT1, but the preparation of the input data is a
tedious, error-prone process. The new entry automates this process so that it is possible to obtain the
results for each aerodynamic strip in the model with a single MONCNCM entry.
Input
The new MONCNCM Bulk Data entry is used to provide the stripwise aerodynamic results. The remarks
of this entry in the MD Nastran Quick Release Guide provide guidance in its use. In particular, use of
CAERID1=ALL produces results at each aerodynamic strip. The code internally determines if the flat
plate panels are abutting and, if they are, considers this a single strip from the leading edge to the trailing
edge of the surface made up of the multiple CAERO1 entries.
Output
A new table in the .f06 file that is included as part of the aerodynamic monitor point output. A sample
output from the moncncm test case discussed below is shown here:
A E R O D Y N A M I C M O N I T O R P O I N T I N T E G R A T E D L O A D S
CONFIGURATION = AEROSG2D XY-SYMMETRY = ASYMMETRIC XZ-SYMMETRY = SYMMETRIC
MACH = 9.000000E-01 Q = 4.000000E+01
MONCNCM NAME = ALL INT GROUP ID = 1 CLASS = STRIP
LABEL = ALL STRIPS FOR THE AIRPLANE
CP = 1 AERODYNAMIC COORDINATE SYSTEM
FORCE MOMENT
STRP YS ZS XREF REFC REFS RIGID ELASTIC RIGID ELASTIC
____ __________ __________ __________ ___________ ___________ ___________ ___________ ___________ ___________
1 1.250E+00 0.000E+00 0.000E+00 1.0000E+01 2.5000E+01 1.5715E+00 1.5735E+00 1.4352E-01 1.4300E-01
2 1.250E+00 0.000E+00 1.428E+01 1.0000E+01 2.5000E+01 -3.1044E-01 -2.9946E-01 -6.2361E-02 -6.1159E-02
3 3.750E+00 0.000E+00 0.000E+00 1.0000E+01 2.5000E+01 1.2319E+00 1.2336E+00 1.1194E-01 1.1147E-01
4 3.750E+00 0.000E+00 1.283E+01 1.0000E+01 2.5000E+01 4.9708E-03 1.6573E-02 -9.2159E-03 -8.0990E-03
5 6.250E+00 0.000E+00 1.139E+01 1.0000E+01 2.5000E+01 1.1896E+00 1.2036E+00 1.7237E-01 1.7369E-01

250
6 8.750E+00 0.000E+00 9.948E+00 1.0000E+01 2.5000E+01 1.1173E+00 1.1319E+00 9.4866E-02 9.5348E-02
7 1.125E+01 0.000E+00 8.505E+00 1.0000E+01 2.5000E+01 1.0168E+00 1.0308E+00 3.8398E-02 3.8538E-02
8 1.375E+01 0.000E+00 7.061E+00 1.0000E+01 2.5000E+01 8.8463E-01 8.9719E-01 3.4616E-03 3.4251E-03
9 1.625E+01 0.000E+00 5.618E+00 1.0000E+01 2.5000E+01 7.1419E-01 7.2468E-01 -1.3503E-02 -1.3636E-02
10 1.875E+01 0.000E+00 4.175E+00 1.0000E+01 2.5000E+01 4.8032E-01 4.8760E-01 -1.6023E-02 -1.6161E-02
The strips are arranged based on increasing y-coordinate of the strip in the aerodynamic coordinate
system. The z coordinate and the x location of the point at which the pitching moment is computed (based
on the MREF field of the MONCNCM entry) is given as are the local chord length and strip area. This
is followed by the lift and pitching moment at the trim state for the rigid vehicle and for the elastically
deforming vehicle.
Examples (moncncm and monbodi)
Two small test files demonstrate the use of the MONCNCM Bulk Data entry. They are both variations of
the familiar forward swept wing and are:
moncncm provides tests of various ways of merging data across strips that may not be practical from an
engineering standpoint but that exercise different code paths.
monbodi - demonstrates the application of MONCNCM when there are multiple interference groups and
aerodynamic bodies. It is seen that different interference groups are not merged across aerodynamic
panels and that there are no results given for CAERO2 entries.
The MONCNCM entry only supports Doublet Lattice like models and only for lifting surfaces.
Aerodynamic bodies are not supported. The output is in the aerodynamic panel coordinate systems so
that the sign of the forces and moments are a function of the numbering of the corners of the panel (see
the description of the CAERO1 entry in the MD Nastran Quick Reference Guide). The results are given
at the aeroelastic trim state.
251
CHAPTER 9
Input of an Aerodynamic Mesh
Introduction
Four new bulk data entries have been provided that allow the user to define an aerodynamic mesh that is
distinct from the traditional aerodynamic model that is input using a combination of CAEROi/PAEROi
entries.
Benefits
The ability to input an aerodynamic mesh into MD Nastran opens up various capabilities that are
considered embryonic at this point. Among these are:
It enables the viewing of an external aero mesh on an MD Nastran structural model using a tool such as
MD Patran.
It is possible to spline results from the structural mesh to the new aerodynamic mesh and vice versa. The
spline export feature described in section 1.6 below provides this spline for manipulation outside of MD
Nastran
The aerodynamic grid points can be loaded using the AEFORCE entry, thereby enabling the import of
rigid aerodynamic loads from an external aerodynamics program.
The AESCALE entry described below opens up some interesting possibility for morphing models that
have yet to be explored.
Input
The aerodynamic mesh can be input using four new bulk data entries
The AEGRID provides coordinates of the mesh.
The AEQUAD4 entry connects the mesh using a quad element
The AETRIA3 connects the mesh using a triangular element
The AESCALE entry allows for a scaling of the AEGRID data on a grid-by-grid basis
Note that the element data (AEQUAD4 and AETRIA3) are strictly for display and do not participate in
the creation of any aerodynamic matrices.
Output
There is no new output.
Example (aegridf)
This simple example is a variation on the 15 degree swept wing model named ha144c.dat in the
MSC.Nastran Users Guide for Aeroelasticity. Relative to the example in the guide, the aegridf.dat
example:

252
Removes the doublet lattice model
Adds AEGRID quantities that coincide with all the structural grids in the model plus four more
that represent an aerodynamic stabilizer
Uses a SPLINE4 entry to spline the wing forces at the AEGRIDs to the structure
Uses a SPLINRB (see ?.4) to spline the AEGRIDs on the stabilizer to the support point
Removes the FORCE/GRAV artifice used to simulate constrained wind tunnel condition. Instead
a 1g cruise condition is specified with the aircraft free to pitch and plunge.
AEPARM entries specify ALPHA and STAB states.
AEFORCE entries are used to set up load vectors for the intercept, unit angle of attack and unit
stabilizer setting.
The stability derivative print from running the example includes:
N O N - D I M E N S I O N A L S T A B I L I T Y A N D C O N T R O L D E R I V A T I V E C O E F F I C I E N T S
MACH = 4.5000E-01 Q = 2.0000E+00
CHORD = 2.0705E+00 SPAN = 1.1050E+01 AREA = 1.1440E+01
CONTROLLER STATE: INTERCEPT ONLY, ALL CONTROLLERS ARE ZERO
TRIM VARIABLE COEFFICIENT RIGID ELASTIC INERTIAL
UNSPLINED SPLINED RESTRAINED UNRESTRAINED RESTRAINED UNRESTRAINED
REF. COEFF. CX 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00
CY 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00
CZ 3.496504E-02 3.496504E-02 3.496504E-02 3.496504E-02 0.000000E+00 3.496504E-02
CMX 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00
CMY -1.249997E-02 -1.249997E-02 -1.249997E-02 -1.249997E-02 0.000000E+00 -1.249997E-02
CMZ 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00
ALPHA CX 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00
CY 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00
CZ 2.596154E+00 2.596154E+00 2.596154E+00 2.596154E+00 0.000000E+00 2.596154E+00
CMX 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00
CMY -1.296420E+00 -1.296420E+00 -1.296420E+00 -1.296420E+00 0.000000E+00 -1.296420E+00
CMZ 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00
STAB CX 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00
253
CHAPTER 9
CY 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00
CZ 7.517483E-01 7.517483E-01 7.517483E-01 7.517483E-01 0.000000E+00 7.517483E-01
CMX 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00
CMY -2.760425E+00 -2.760425E+00 -2.760425E+00 -2.760425E+00 0.000000E+00 -2.760425E+00
CMZ 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00
URDD3 CX 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00
CY 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00
CZ 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 1.793749E-03 0.000000E+00
CMX 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00
CMY 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 -6.412643E-04 0.000000E+00
CMZ 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00
It is seen that the splined and unsplined rigid aerodynamic terms are identical, which means the splines
are functioning well, and that the rigid and elastic results are the same, which reflects the fact that there
is no aeroelastic feedback in this case.
Trim results from the run include:
A E R O S T A T I C D A T A R E C O V E R Y O U T P U T T A B L E S
MACH = 4.500000E-01 Q = 2.000000E+00
CHORD = 2.0705E+00 SPAN = 1.1050E+01 AREA = 1.1440E+01
TRIM ALGORITHM USED: LINEAR TRIM SOLUTION WITHOUT REDUNDANT CONTROL SURFACES.
AEROELASTIC TRIM VARIABLES
ID LABEL TYPE TRIM STATUS VALUE OF UX
INTERCEPT RIGID BODY FIXED 1.000000E+00
501 ALPHA GENERAL CONTROL FREE -1.334841E-02 RADIANS
502 STAB GENERAL CONTROL FREE 1.973043E-03 RADIANS
503 URDD3 RIGID BODY FIXED 1.000000E+00 LOAD FACTOR
505 URDD5 RIGID BODY FIXED 0.000000E+00 RAD/S/S PER G
It is seen that the very lightweight structure is in trim based on intercept aerodynamics, a negative angle
of attack and a slightly positive stabilizer.
The aerodynamic mesh cannot be used in conjunction with the existing CAEROi format for creating an
aero mesh.

254
The CMPID listed on the aerodynamic elements can be reference on an AECOMP that is used in the
aerodynamic splining. However, the splining is only done based on the AEGRID data that are provided
on an AELIST entry identified with the AECOMP.
Unlike the CAEROi format, there is no central grid on an aerodynamic element. The only aerodynamic
grids are at the element vertices.
There is no integration of user input pressures or downwashes, so the AEFORCE entry is the only means
of loading the aerodynamic mesh and, again, this can only be done at the AEGRID locations.
AICs are not supported for this type of mesh.
255
CHAPTER 9
Rigid Body Spline
Introduction
A new spline technique has been introduced that enables the splining of aerodynamic data to exactly six
structural degrees of freedom in a rigid body fashion.
Benefits
The primary application of the rigid body spline is to introduce aerodynamic loads into the structural
analysis process when there is no structure underlying the aero mesh. For example, this enables
introducing loads from the tail component in a wing design when only the wing has structural detail.
Input
The new rigid spline method is enabled via the SPLINRB entry.
Output
There is no new output.
Example (splinrb)
A very simple example is provided that is based on the ha144e test case of Section 7.5 of the
MSC.Nastran Users Guide for Aeroelastic Analysis. The example in the Users Guide uses separate 1D
splines (the SPLINE2 entry) to distribute the aerodynamic load on the beam structure. This example
replaces these two splines with SPLINRB entries that spline all the loads to the support point. The effect
of this is that these loads no longer create an aeroelastic affect and the rigid and splined stability
derivatives are identical.
In addition to the loads being transferred to the 6 structural degrees of freedom, it should also be noted
that the motion of the aerodynamic surface corresponds to the rigid body motion defined by these 6 dofs.

256
Wendland Spline Functions for the Spline4/5
Introduction
Options have been added to the SPLINE4 and 5 bulk data entries to support selection of radial
interpolation functions. Reference 1 contains the mathematical development of these functions while
Reference 2 further develops them in the context of an aeroelastic application.
Benefits
The references indicate that these alternative splining techniques should provide smooth results with
improved performance relative to the existing methods. As indicated in the Guidelines and Limitations
portion of this subsection, MSC does not have enough experience with the new methods to quantify these
benefits.
Input
The METH field for the SPLINE4 now supports RIS in addition to the existing IPS, TPS and FPS
methods. If RIS is selected, the order of the interpolation method is either defaulted to WF2 or can be set
to WF0. RCORE is used to define the radius of support. For the SPLINE5, METH is a new field, with
BEAM (infinite beam spline) the default. If METH=RIS is specified, FTYPE (default=WF2) and
RCORE also require input.
Example (ha144c_ris)
The ha144c example has been modified to change the spline from SPLINE1 to SPLINE4 with
METH=RIS, FTYPE=WF2 and RCORE=5.0. Running the example produces answers that are
equivalent but not identical to ha144c. The differences are comparable to those that results from replacing
the infinite plate spline with a finite plate spline so that it is deemed these results are reasonable.
MSC has had limited experience with the radial interpolation functions. They are being provided in a
Beta state so that clients can exercise these new options. This is best done in conjunction with the
Spline Verify option of MSC.Flightloads so that a visual check of the quality of the fit can be made.
Because the radial interpolation functions act over a limited area, specified using RCORE, it is believed
that a single spline can be effectively applied to a large area. For example, with a high aspect ratio wing,
a single spline may be able to suffice.
If the linear SPLINE5 includes the rotational flexibility (DTHX/DTHY > -1.0) from the structural
deformations, the FTYPE =WF2 will be used regardless of the user input value of FTYPE.
257
CHAPTER 9
Spline Blending
Introduction
A capability has been added to blend spline effects across multiple splines.
Benefits
Prior to this MD R2 Nastran release, it was an error if the same aerodynamic grid was invoked on
multiple splines. This could result in discontinuities at the panel boundaries so this rule has been relaxed
so that an aero grid can be selected on multiple splines and the results blended in some user specified
fashion. This results in smoother displacement patterns which will be of particularly of benefit as the
aeroelastic analysis includes more CFD aerodynamics.
Input
A parameter has been provided to indicate that blending is allowed to occur. The parameter is
MPTSPLIN and is input using the MDLPRM Bulk Data entry. This parameter has the following
meaning:
MPTSPLIN = 0 Do not allow an aerodynamic grid to be referenced by more than one spline (Default)
= 1 Allow an aerodynamic grid to be reference by multiple splines.
This parameter does not apply to aerodynamic grids defined with AEGRID discussed above. These grids
can always be referenced on multiple splines.
Two new Bulk Data entries (SPBLND1 and SPBLND2) have been provided to support spline blending.
If neither of these is used, the splines are averaged. The SPBLND1 performs strip based blending that
can be either averaged (with a user defined weighting), linear or cubic in nature. The SPBLND2
performs a curve based blending based on user input curve.
Example (ha145e_blnd)
The existing ha145e test deck has a single SPLINE1 entry. In this example, this has been replaced by two
splines that overlap the two center rows of boxes. MDLPRM,MPTSPLIN,1 is used to enable this overlap
when using the CAERO1 generated aerodynamic meshes. A SPBLND1 entry is then used to blend the
results in a linear way. If the SPBLND1 entry had not been used, the splines would have been averaged.
For this simple example, the blending has little affect on the flutter results. The intent of the example is
to demonstrated the new features.
This is another area where MSC has limited experience so there is limited guidance. It can be said that
the blending developments have been made in concert with the alternative aeromeshing of section 1.3
above and is therefore pointed at applications that use external aerodynamics. It is felt that the blending

258
will be most useful in areas like wing fuselage junctions were the standard splining techniques are likely
to result in gaps in the aerodynamic deformations.
259
CHAPTER 9
Export of the Spline Matrix
Introduction
A capability has been provided to export the spline matrix to either an .op2 file or to the punch file.
Benefits
Users can now easily obtain the spline matrix in a format that can be readily applied in another procedure
or reintroduced into another Nastran run.
Input
The SPLINOUT case control command is used to produce the exported spline matrix. If the .op2 format
is used, the user should also assign a file that will contain the matrix in OUTPUT2 format.
Output
If the punch output is requested, the GPGK spline matrix is written to the .pch file using the DMIG
format. The DMIG print is preceded by a direct table input (DTI) data that provides a map of the matrix
columns to the aerodynamic degrees of freedom. With the default .op2 method, an output2 file is written.
Examples (splinopch and splinoop2)
These two examples show the use of the .pch and .op2 options for exporting a spline mat
References:
Wendland, H., Piecewise Polynomial, Positive Definite and Compactly Supported Radial Functions of
Minimal Degree, Adv. Computational Mathematics, 4(1995), 389-396.
Beckert, A. and Wendland, H,. Multivariate Interpolation for Fluid-Structure-Interaction Problems using
Radial Basis Functions, Aerospace Sci. Technology, 00 (2001), pp 1-11.

260

MD R2 Nastran Release Guide

Diunggah oleh

Informasi Dokumen

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

MD R2 Nastran Release Guide

Diunggah oleh

Hak Cipta:

Format Tersedia

MD R2 Nastran

Linear Contact, Permanent Glued Contact, and Nonlinear Contact

Guidelines and Limitations

Advanced Integrated Nonlinear (SOL 400)

SOL 400 Material and Elements

The Nonlinear QUADR/TRIAR Elements

CBAR Element to CBEAM Element Conversion

CSHEAR Element to CQUAD4 Element Conversion

SOL 400 Pretension Modeling Description

VCCT Virtual Crack Closure Method

Cohesive Zone Modeling

Nonlinear Iteration Algorithms

Analysis Chaining SUBCASE, STEP, ANALYSIS, and NLIC

Adaptive Time Stepping Scheme

The Numerical Integration Method for the Transient analysis the

Nonlinear Iteration Summary Table for Nonlinear Transient Analysis

Output Data Grouping: NLPACK

MD R2 Nastran Implicit Nonlinear - SOL 600

MD R2 Nastran Explicit Nonlinear - SOL 700

MD R2 Nastran Release Guide

Frequency Response Function (FRF) and FRF Based Assembly

Rigid Porous Absorber

MD R2 Nastran Release Guide

New SPARSESOLVER Executive Statement

Improved Performance with New Sparse Solvers

Improved Memory Usage in Lanczos (Pre-release)

New MAXRATIO Information Output (Pre-release)

Performance Improvements and Expanded Capabilities for ACMS

Relaxed Restrictions for CASI Solver Usage

System Dependent Performance Improvements

Improved Selection of Reordering Methods

Enhancements to Connector Elements

Composite Beam Using VAM

Nonhomogeneous Multipoint Constraint

Topology Optimization Enhancements

Automatic External Superelement Optimization (AESO)

Randomization of a Users Input Data File (Pre-release)

Random Elimination of Element Types (Pre-release)

doptprm desmax 10 autose 1 dratio 0.9

Changes to Rotordynamics for MD R2 Nastran

Updating/Summing of Monitor Points

Stripwise Aerodynamic Results

Input of an Aerodynamic Mesh

Rigid Body Spline

Wendland Spline Functions for the Spline4/5

Export of the Spline Matrix

Anda mungkin juga menyukai