Are you sure?
This action might not be possible to undo. Are you sure you want to continue?
,
Although not updated for Version 68 of MSCINASTRAN, the MSCINASTRAN Handbook for Nonlinear Analysis, Version 67, is still being offered for a couple of reasons. The first is that the new MSCINASTRAN Nonlinear Analysis User's Guide for Version 68 IS currently in progress and not yet available. The goals for this new guide are to present the range of nonlinear analysis in capabilities in a userfriendly, tutorial format and to update the wealth of technical inforrnatlon ava~lable the current handbook. It is intended to act as a singlepoint source of reference for all MSCINASTRAN nonlinear analysis capabilities. Secondly, the MSCINASTRAN Handbook for NonlinearAnalysis contains indepth technical inforrnatlon and a thorough review of implementation details that are simply unavailable anywhere else. Much of thls information has been of particular interest to advanced analysts, and as such still enjoys w~de appeal. In addition to this handbook, we recommended that you also consult the MSCINASTRAN Reference Manual, Section 15.3, for information pertaining specifically to Version 68 nonlinear analysis capabilitles.
Greg Moore Manager, Technical Publications
MSCINASTRAN Nonlinear Analysis
HANDBOOK
Version 67
Volume I
Sang H. Lee
Corporate Headquarters The MacNealSchwendler Corporation 815 Colorado Boulevard Los Angeles, CA 900411777 Tel: (213) 25891 11 or (800) 3364858 FAX: (213) 2593838 Headquarters, European Operations MacNealSchwendler GmbH lnnsbrucker Ring 15 Postfach 801240 D  81 612 Miinchen, GERMANY Tel: (89) 431 9870 Telex: 523 784 MSG D FAX: (89) 436 1716 Headquarters, Far East Operations MSC Japan Ltd. EntsujiGadelius Building 239, Akasaka 5chome Minatoku, Tokyo 107, JAPAN Tel: (03) 35050266 Telex: J23363 MSCWATA FAX: (03) 3505091 4
DISCLAIMER
I
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. The MacNealSchwendler 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.
a1992 by The MacNealSchwendierCorporation Printed in U.S.A. All rights reserved.
I
MSC and MSCI are registered trademarks and service marks of The MacNealSchwendler Corporation. NASTRAN is a registered trademark of the National Aeronautics and Space Administration. MSCINASTRAN is an enhanced, proprietary version developed and maintained by The MacNealSchwendlerCorporation.
NA *V67 * Z * Z * Z * HBNLN
PREFACE
This handbook is developed as tutorial material for MSC/NASTRAN users t o solve nonlinear structural problems. It is assumed that the reader have prior knowledge of MSCINASTRAN fundamentals as well as the proficient knowledge of solid mechanics. The text covers the entire spectrum of nonlinear structural analysis: statics, dynamics and stability. Different aspects of nonlinearities are exemplified. The rudimentary theories are reviewed and extended t o practical problems. Special features of nonlinear finite elements are highlighted. Some details of the program structure are discussed in terms of algorithms, modules and DMAP. The handbook is selfcontained for nonlinear analysis, providing a full description of the nonlinear features of MSCINASTRAN with emphasis on the userinterface. While maintaining consistency and continuous flow throughout the book, efforts are made to keep each chapter independent of others so that the reader need not read the entire volume to start. This handbook will be expanded as new capabilities are introduced. The initial edition of this handbook consists of eleven chapters. It is recommended that the reader review the first two chapters t o start. Then he may continue to read any chapter on the subject of his interest. The user interface is collectively described in Chapters 7 and 9 for static and dynamic analysis, respectively. MSCINASTRAN nonlinear capabilities have matured through the wide utilization, constructive critiques, friendly advice and favorable acceptance of numerous users in our esteemed clientele community. With this document, MSC responds t o our enthusiastic users for their patronage with an acknowledgement of their contribution t o the growth of MSCINASTRAN. I hope that this handbook proves t o be an easytofollow tutorial text for novice users and a comprehensive reference for expert users, clarifying their understanding of our code. Dr. R. H. MacNeal and Messrs. D. N. Herting and R. L. Harder are acknowledged for the conception and the initial design of the nonlinear capabilities. Wherever applicable, contributors t o the sections of this handbook involving various authors are listed in the Table of Contents. Contributions t o the development efforts by many engineers and programmers whose names are not mentioned here are acknowledged. Thanks are expressed t o A. Raiten, C. S. Kim and C. J. Curley for their assistance in preparing Figures and Tables. Appreciation is extended t o K. Blakely and A. Raiten for proofreading the manuscript. March 1, 1992
Sang H. Lee, Dr. Eng., P.E.
Principal Engineer
CONTENTS
Note: Contributors of various sections denoted in italics.
Preface List of Tables and List of F i g u r e s
1
Introduction 1.1 Evolution 1.2 Overview of Nonlinear Capabilities 1.3 Program Architecture 1.4 Rudiments of User Interface 1.4.1 Case Control Section 1.4.2 Bulk Data Section 1.4.3 Parameters 1.4.4 Nonlinear Element Output Codes for Plotting 1.5 Nonlinear Characteristics and General Recommendations Basic C o n s i d e r a t i o n s for Nonlinear Analysis 2.1 Discrete System for a Nonlinear Continuum Model 2.2 Finite Element Formulation for Equilibrium Equations 2.3 Coordinate Transformations 2.4 Displacement Sets and Reduction of Equations 2.5 Nonlinear Solution Procedure Solution M e t h o d s 3.1 Adaptive Solution Strategies 3.2 Newton's Method of Iteration 3.3 Stiffness Update Strategies 3.3.1 Update Principles 3.3.2 Divergence Criteria 3.3.3 Time Expiration Criteria 3.4 Line Search Method 3.4.1 Theoretical Basis 3.4.2 Search Criteria 3.4.3 Implementation of Search Procedure 3.4.4 OneDimensional Example 3.4.5 Other Provisions for Line Search 3.4.6 Quadratic Interpolation 3.5 QuasiNewton Method 3.5.1 Evolution of QuasiNewton Method 3.5.2 Criteria for BFGS Update 3.5.3 Implementation of the BFGS Update Strategy 3.5.4 OneDimensional Example
2
3
3.5.5 BFGS Updates for Arclength Methods 3.6 Convergence Criteria 3.6.1 Rudimentary Considerations 3.6.2 Convergence Conditions 3.6.3 Error Functions and Weighted Normalization 3.6.4 Implementation 3.6.5 Some Observations 3.7 ArcLength Methods for PostBuckling Analysis 3.7.1 Basic Theory for ArcLength Methods 3.7.2 Riks' Method and Its Variations 3.7.3 Crisfield's Method 3.7.4 Adaptive Arclength Method 3.7.5 Verification Problems 4
Finite Elements in Nonlinear Analysis 4.1 Overview 4.1.1 Line Elements 4.1.2 Surface Elements 4.1.3 Solid Elements 4.1.4 Other Elements 4.2 Formulation of Isoparametric Finite Elements 4.2.1 Isoparametric Coordinates 4.2.2 Shape Functions 4.2.3 An Example of Element Matrix 4.2.4 Volume Integration of Element Matrices 4.2.5 Element Loads and Equilibrium 4.2.6 Element Coordinates 4.2.7 Stress Data Recovery 4.3 Gap and Friction Element 4.3.1 Introduction 4.3.2 Basic Considerations for Penalty GAP Element 4.3.3 User Interface for GAP Element 4.3.4 Theoretical Basis for Frictional Behavior 4.3.5 Subincremental Algorithm for Stiffness Update and Bisection 4.3.6 Adaptive Adjustment of Penalty Values 4.3.7 Nonadaptive GAP Element with Anisotropic Friction 4.3.8 Verification and Validation: Bouncing Mass Problem 4.3.9 Verification and Validation: Vibration with Coulomb Damping 4.3.10 Hertzian Problem: Contact Between Sphere and Rigid Plane 4.3.11 Dynamic Case: Impact of a Sphere with Rigid Plane 4.3.12 Recommendations Geometric Nonlinearity 5.1 Overview and User Interface
S. H. Lee
5
S. H . Lee 5.2 Updated Element Coordinates 5.2.1 Concept of Convective Coordinates 5.2.2 Updated Coordinates and Net Deformation 5.2.3 Provisions for Global Operation S. H. Lee, D. N . Herting 5.3 Follower Forces 5.3.1 Basic Definition 5.3.2 Implementation 5.3.3 Dynamic Follower Forces 5.3.4 Verification: Elliptic Cylinder Subject to an Internal Pressure 5.4 Treatment of Large Rotation D. V. Wallerstein,R.Allahabadi 5.4.1 Gimbal Angle Approach 5.4.2 Rotation Vector Approach 5.4.3 Bisection Due to Large Rotations 5.4.4 Output Interpretation A . Raiten 5.5 Verification: Large Rotation of a Beam 5.5.1 Problem Description 5.5.2 Finite Element Model 5.5.3 Analysis Procedure 5.5.4 Analysis Results S. H. Lee Canonical Approach t o the Geometric Nonlinearity 5.6 5.6.1 Fundamentals of Continuum Mechanics 5.6.2 Incremental Formulations for Large Displacements and Strains 5.6.3 Total Lagrangian vs. Updated Lagrangian in Finite Elements 5.6.4 Constitutive Relations for Large Deformation
6
Material Nonlinearity and Constitutive Relations 6.1 Introduction 6.1.1 Overview 6.1.2 User Interface 6.2 Plasticity 6.2.1 Some Preliminaries 6.2.2 Yield Criteria 6.2.3 Yield Function 6.2.4 Strain Hardening 6.2.5 PrandtlReuss StressStrain Relations 6.2.6 Associated Flow Rule 6.2.7 Generalized Effective Plastic Strain Increment 6.2.8 Yield Function Derivatives 6.2.9 Degenerate Cases: Plane Strain, Plane Stress and Uniaxial Stress 6.2.10 Solution Algorithm for ElastoPlastic Material 6.2.11 Verification and Validation 6.3 Nonlinear Elasticity 6.3.1 Theoretical Basis 6.3.2 Solution Alogorithm
S. H. Lee
S.H. Lee
S. H. Lee
6.3.3 Adaptation of Uniaxial Compression StressStrain Curve 6.3.4 Computational Procedure for Bilateral StressStrain Relations 6.4 Creep and Viscoelasticity 6.4.1 Introduction 6.4.2 Effects of Stress and Temperature 6.4.3 Equillibrium of a KelvinMaxwell Model for a Stress Component 6.4.4 Adaptation of Rheological Model to Multiaxial Stress State 6.4.5 Coupling of Plasticity 6.4.6 Implementation 6.4.7 Verification and Validation 6.4.8 NozzletoSpherical Shell Attachment as Benchmark Problem 6.5 ThermoElasticity 6.5.1 Theoretical Basis 6.5.2 User Interface 6.5.3 Validation Problem 6.6 Initial Anisotropy 6.6.1 Theoretical Basis 6.6.2 Validation Problems
7
S. H. Lee
S.S. Hsieh
S.S. Hsieh
Nonlinear Static Analysis 7.1 User Interface 7.1.1 Case Control 7.1.2 Iteration Control: NLPARM Data 7.1.3 User Interface for Arclength Methods: NLPCI data 7.1.4 Iteration Related Output Data 7.2 Iteration Module and Solution Sequence 7.2.1 DMAP Interface of NLITER Module 7.2.2 Input Data Blocks 7.2.3 Output Data Blocks 7.2.4 Parameters 7.2.5 Diagnostic Outputs 7.3 Loads Overview 7.3.1 Concentrated Loads 7.3.2 Distributed Loads 7.3.3 Mass Related Loads 7.3.4 Thermal Loads 7.3.5 Loads Combination 7.4 Nonproportional Loading 7.4.1 Validation of Nonproportional Loading Case 7.4.2 Biaxial Loading Case For Neutral Loading 7.5 Constraints and Enforced Motion 7.5.1 Boundary Condition 7.5.2 Multipoint Constraint 7.5.3 Enforced Motion
S. H. Lee
S.H.Lee, T.L. Bock
D. V. Wallerstein
S.H. Lee
S.S. Hsieh
7.5.4 Example 7.6 Restarts 7.6.1 Preliminaries 7.6.2 Restart Procedure 7.6.3 Example 7.7 Verification Problem: Analysis of a Pressure Vessel
8
S. Hsieh S.
Direct Time Integration S. H. Lee 8.1 Overview 8.2 ThreePoint Method (NLTRD Module) D. N. Herting 8.2.1 Basic Equations 8.2.2 Nonlinear Iterations 8.2.3 Stability Analysis 8.2.4 DMAP Interface of NLTRD Module 8.3 TwoPoint Method (NLTRD2 Module) S.H. Lee 8.3.1 Newmark Integration 8.3.2 Adaptation of Newmark's Method 8.3.3 Startup Algorithm 8.3.4 DMAP Interface of NLTRD2 Module 8.4 Adaptive Time Stepping S.H. Lee 8.4.1 Introduction 8.4.2 Automatic Time Stepping Algorithm 8.4.3 Bisection Algorithm 8.5 Expedient Iteration Strategies S. H. Lee 8.6 Verification Problems S.H. Lee 8.6.1 ElasticPlastic Rod with an End Mass Subjected to a Step Loading 8.6.2 Impulsively Loaded Clamped Plate 8.6.3 Simply Supported Beam with a Restrained Motion 8.6.4 Bouncing Weight on an Elastic Platform 8.6.5 Shallow Spherical Cap under a Concentrated Apex Load Nonlinear Transient Response Analysis 9.1 User Interface 9.1.1 Case Control 9.1.2 Implicit Integration Control: TSTEPNL Data 9.1.3 Iteration Related Output Data 9.2 Restarts 9.2.1 Restarting from SOL 66 into SOL 99 9.2.2 Restarting within SOL 99 9.2.3 Example 9.3 Dynamic Effects: Mass and Damping 9.3.1 Mass Modeling 9.3.2 Mass Matrix 9.3.3 Damping Modeling
9
S.H. Lee
S. S. Hsieh
A. Raiten
9.3.4 Damping Matrix 9.4 Transient Loads and Initial Conditions 9.4.1 Transient Loads 9.4.2 Enforced Motion 9.4.3 Initial Conditions 9.5 Diagnostic Outputs 9.5.1 NLTRD Module for AUTO or TSTEP Method 9.5.2 NLTRD2 Module for ADAPT Method
10 Special Applications
D. Herting, S. H. Lee
T.L. Bock
10.1 Nonlinear Buckling Analysis 10.1.1 Introduction 10.1.2 Formulation of Nonlinear Buckling Analysis 10.1.3 Analysis Procedure 10.1.4 Verification Problem: Euler Column Buckling 10.1.5 Example of a Nonlinear Buckling 10.2 Nonlinear Modal Analysis 10.2.1 Analysis Procedure 10.2.2 Bending Vibration of Linear Material Beam 10.2.3 Bending Vibration of Nonlinear Material Beam 10.2.4 Plate Vibration 10.2.5 Beam Bending Vibration with Superelements 10.3 Static Analysis using SOL 99 or 129 10.3.1 Input Data for SOL 99 or 129 Static Analysis 10.3.2 A ZShaped Beam Loaded with A Static Force 10.3.3 Rotation of a Beam with Friction
11 Example Problems 11.1 SnapThrough Analysis of a Spherical Diaphragm in SOL 66 11.1.1 Problem Description 11.1.2 Finite Element Modeling and MSCINASTRAN Input Data 11.1.3 Analysis Results 11.1.4 Solution Progression 11.1.5 Concluding Remarks 11.2 Parametric Study on BFGS Updates and Line Search Method in SOL 66 11.2.1 Introduction 11.2.2 The Cologne Challenge: a ZShaped Cantilever Beam 11.2.3 SnapThrough and SnapBack: a Spherical Shell with Backing Plate 11.2.4 PreBuckling Behavior: an Imperfect Spherical Cap 11.2.5 Concluding Remarks 11.3 Creep Analysis of a ThickWalled Pressure Vessel 11.3.1 Introduction 11.3.2 Problem Description 11.3.3 Model Description
11.3.4 Solution Procedure 11.3.5 Results and Discussion 11.3.6 Input Data Listing of Pressure Vessel 11.4 Transient Response of an Impulsively Loaded Cylindrical Panel 11.4.1 Problem Description 11.4.2 Finite Element Model 11.4.3 Analysis Results and Discussion 11.4.4 Input Data Listing for the 8x16 Model R. Louwers 11.5 Contact Analysis for the ShaftJournal Interface in SOL 66 11.5.1 Introduction 11.5.2 Problem Description 11.5.3 Selection of Gap Stiffness 11.5.4 Superelement and ASET 11.5.5 Analysis Procedure 11.5.6 Summary of Results 11.5.7 Friction in the GAP 11.6 Impact Analysis in SOL 99: Stress Wave Propagation in an Elastic Rod 11.6.1 Problem Description 11.6.2 Theoretical Solution 11.6.3 MSCINASTRAN Solution 11.6.4 Comparison of Results 11.6.5 Input Data Listing 11.7 Effects of KGROT on Geometric Stiffening of the Thin Shell Model 11.7.1 Theoretical Basis for Parameter K6ROT 11.7.2 Difficulties in the Thin Shell Model 11.7.3 Parametric Studies 11.7.4 QUAD4 Model of Rectangular Plate with Line Load 11.7.5 TRIA3 Model of Rectangular Plate with Line Load 11.7.6 QUAD4 Model of Square Plate with Uniform Pressure 11.7.7 QUAD4 Model of Square Plate with Concentrated Load 11.7.8 Observations 11.8 Adaptation of Creep Analysis Capability to General Viscoelastic Materials 11.8.1 Introduction 11.8.2 Formulation for Least Square Fit of Rheological Parameters 11.8.3 Computational Process 11.8.4 Analysis Procedure and Results 11.8.5 Remarks
Bibliography Appendix A. Index
Nonlinear Bulk Data Description
LIST OF TABLES
Summary of Properties of the Nonlinear Elements Example Input Data for Nonlinear Static Analysis Example Input Data for Nonlinear Dynamic Analysis Summary of Nonlinear Case Control Data Summary of Parameters in Nonlinear Analysis PARAMeters used in Nonlinear Solution Sequences Input Data Listing for Bouncing Mass Problem Input Data Listing for Coulomb Damping Problem SOL 66 Input Data Listing for Elliptic Cylinder SOL 99 Input Data Listing for Elliptic Cylinder SOL 64 Input Data Listing for Beam with Axial End Force SOL 66 Input Data Listing for Beam with Axial End Force SOL 64 Input Data Listing for Beam with End Moment SOL 66 Input Data Listing for Beam with End Moment Input Data Listing for Stress Path Dependent Case Input Data Listing for Strain Path Dependent Case Input Data Listing for Biaxial Loading Case Input Data Listing for Elastic Clip Input Data Listing for Cold Start Input Data Listing for First Restart Input Data Listing for Second Restart Input Data Listing for Third Restart Input Data Listing for Pressure Vessel Model Input Data Listing for ElasticPlastic Rod Input Data Listing for Impulsively Loaded Plate Input Data Listing for Beam with a Restrained Motion Input Data Listing for Bouncing Weight Model Input Data Listing for Shallow Spherical Cap Input Data Listing for Simply Supported Beam Verification Problem Input Data MSCINASTRAN Output Summary Example Problem Input Data DMAP Alters for SOL 63 (Version 66) SOL 66 Input Data Listing for Uniform Nonlinear Beam SOL 63 Input Data Listing for Uniform Nonlinear Beam SOL 66 Input Data Listing for Plate SOL 63 Input Data Listing for Plate SOL 66 Input Data Listing for Stepped Beam SOL 63 Input Data Listing for Stepped Beam SOL 66 Input Data Listing for ZShaped Beam SOL 99 Input Data Listing for ZShaped Beam
SOL 66 Input Data Listing for Beam with Friction SOL 99 Input Data Listing for Beam with Friction Input Data for the Clamped Case Partial Listing of the Input Data for the Simply Supported Case Solution Strategy and Performance for Clamped Case Effects of NLPARM Parameters on Convergence for Subcase 2lIncrement 4 of Clamped Case Solution Strategy and Performance for Simply Supported Case Effects of NLPARM Parameters on Convergence for Subcase 2lIncrement 3 of Simply Supported Case Effects of NLPARM Parameters on Convergence for Subcase 4lIncrement 1 of Simply Supported Case Solution Strategy and Performance for Sliding and Simply Supported Case Numerical Performance as a function of MAXQN in ZPLATE Numerical Performance as a function of MAXQN in ZBEAM Numerical Performance as a function of MAXQN in ZPLATE Numerical Performance as a function of MAXLS in ZBEAM Numerical Performance as a function of MAXQN in the spherical diaphragm with clamped boundary Numerical Performance as a function of MAXLS in the spherical diaphragm with clamped boundary Numerical Performace as a function of MAXQN in the spherical diaphragm with simply supported boundary Numerical Performance as a function of MAXQN in the spherical diaphragm with sliding/simply supported boundary Numerical Performance as a function of MAXLS in the spherical diaphragm with sliding/simply supported boundary Numerical Performance as a function of LSTOL in the spherical diaphragm with sliding/simply supported boundary Numerical Performance as a functioin of MAXQN in the imperfect spherical cap Numerical Performance as a function of MAXLS in the imperfect spherical cap Numerical Performance as a function of LSTOL in the imperfect spherical cap Normalized values of elapsed CPU time for analyses with and without ASETs and/or SUPERELEMENTS Variation of Iteration Strategy MSCINASTRAN Input Data Listing for Case 16 Duration of Impact Input Data Listing for QUAD4 Model with Line Load Results of Parametric Study for K6ROT: QUAD4 Model with Line Load
PAM File Content) Run Progression of the Program "DISPLY" Source Listing for the Program "RHEOLOGY" Source Listing for the Program "DISPLY" . Stress (RHEOLOGY ./CpJli. vs.OUT File Content) C.Input Data Listing for TRIA3 Model with Line Load Results of Parametric Study for KGROT: TRIA3 Model with Line Load Input Data Listing for QUAD4 Model with Uniform Pressure Results of Parametric Study for K6ROT: QUAD4 Model with Uniform Pressure Input Data Listing for QUAD4 Model with Concentrated Load Results of Parametric Study for K6ROT: QUAD4 Model with Concentrated Load Run Progression of the Program "RHEOLOGY7' StrainJStrainRate vs. Time (RHEOLOGY.
Vertical Displacement at the Center (Clamped Case) Imperfect Spherical Shell MSCJNASTRAN Model for Imperfect Spherical Shell LoadDeflection Curve of Imperfect Spherical Shell Deformed Shapes at Different Buckling Stages BEAM Element Coordinate System and Element Forces Plate and shell Element Coordinate System Forces and Stresses in Plate Elements Volume Coordinate for Tetrahedron Gaussian Quadrature for Tetrahedron Shape Functions for Pressure Load Gaussian Quadrature for Pressure Load Element Coordinate System Definition The Gap Element Gap Axial Force vs. Line Search Error Flow Diagram for Line Search Algorithm Flow Diagram for Line Search Iteration Examples for the Line Search Procedure Quadratic Interpolation Algorithm Flow Diagram for Iteration with Quasi Newton Updates Comparison of Iteration Methods (QuasiNewton method with or without line search) Riks' Method Modified Riks' Method Crisfield's Methods Spherical Diaphragm Model with Axisymmetric Boundary Condition and Constraints by Gap Elements Load vs. Deflection Gap Lateral Force vs.LIST OF FIGURES Hierarchy of Nonlinear Looping Newton's Methods for Iteration Hybrid Newton's Method Based on Stiffness Matrix Update Strategy Iteration Trap Condition Flow Diagram for K Update Strategies (to set NEWK) Divergence and Retreat Procedure CPU Time Expiration Logic for NOGO Condition Line Search Parameter vs. Deflection Schema for Subincremental Process (m = 3) Physical Model Symbolic MSCJNASTRAN Model Displacement Response Adaptive Penalty Value Adjustment on GAP 22 .
051.. as a Function of Stress a Representation of Hardening Rules Geometric Representation of von Mises Yield Surface Mohr Circle Representation of the MohrCoulomb Yield Criterion Von Mises Yield Locus for Plane Stress Tresca's Yield Locus for Plane Stress Isotropic Yield Surfaces in Principal Stress Space Isotropic Hardening Kinematic Hardening Flow Diagram for ElastoPlastic Material Uniaxial Loading and Apparent Poisson's Ratio StrainControlled Cyclic Loading Monotonic and Cyclic StressStrain Curves for Gray Cast Iron TABLES1 INPUT for StressStrain Rheological Model xiii .2 sec. Displaced Coordinates Computation of Net Deformation Elliptic Cylinder subject t o an Internal Pressure Displacement Responses for Elliptic Cylinder Deformed Shapes of the Elliptic Cylinder Beam with an Axial End Force Beam with an End Moment Load vs. Criticallydamped. Deformed Shape at P = 2160 Contact Radius & Approach vs. Net Deformation of a Rod Element Coordinates vs.072 and 0. 0. Displacement at Free End of Beam with Axial End Force Deflection Curves for Beam with Axial End Force Deflection Curves for Beam with End Moment StressStrain Curve Definition with TABLES1 Entry StressStrain Curve Definition for a Single H Uniaxial Rheological Model Creep Model Parameter K p as a Function of Stress a Creep Model Parameter Cp as a Function of Stress a Creep Model Parameter C. Total Force in Static Analysis MSCINASTRAN Model for Dynamic Analysis Reaction Force vs. 0. Overdamped Horizontal Displacement during Static Loading Contact Between Elastic Sphere and Rigid Plane MSCINASTRAN Model for Static Analysis Underformed Shape vs. Time Vertical Displacement Response a t Select Points on Center Line Snap Shots a t t = O.Physical Model for Vibration with Coulomb Damping Horizontal Displacement Response Comparison of Responses: Underdamped.
Flow Diagrams for Creep Process Creep Capability Verification for Variable Temperature Isochronous StressStrain Curve Comparison of Creep Relaxation Predictions Effective Stress Distribution after 3 hours of Creep Finite Element Model by MSCINASTRAN Total Circumferential Strain at Gauge Location during Pressure Loading Total Circumferential Strain at Gauge Location (1) during Moment Loading Total Circumferential Strain at Gauge Location (2) during Moment Loading Effective Stress Contour Plot at t=2546 hours upon Loading Effective Stress Contour Plot at t=3400 hours upon Loading Cube Subjected to Uniaxial Tension Flow Diagram for Nonlinear Process in SOL 66 Flow Diagram for Appending Process in SOL 66 Hierarchy of Load Dat'a Validation of Nonproportional Loading Case Finite Element Model for Biaxial Loading Material Curve for Biaxial Loading Biaxial Loading Path Corresponding Strain Path for Biaxial Loading StressStrain Response Curve Elastic Clip Forced into a Cavity Model of Rod with Spring ForceDisplacement Curve Steel Pressure Vessel MSCINASTRAN Model (10% Sector of the Steel Pressure Vessel) Experimental Results vs. MSCINASTRAN Flow Diagram for Nonlinear Process in SOL 99 Flow Diagram for Appending Process in SOL 99 Illustration of the Bisection Strategy Flow Diagram for the Automatic Time Step Adjustment Flow Diagram for the Bisection Algorithm Flow Diagram for the NLTRD2 Module ElasticPlastic Rod Subjected t o a Step Loading ElasticPlastic Rod: Displacement Response and Time Step Variation Impulsively Loaded Clamped Plate Clamped Plate: Central Displacement Response and Time Step Variation Simply Supported Beam with a Restrained Motion Simply Supported Beam: Loading Point Displacement Time History Bouncing Weight on an Elastic Platform Bouncing Weight: Displacement Time History Shallow Spherical Cap under a Concentrated Apex Load Spherical Cap: Apex Displacement Response xiv .
) at End of Beam During Dynamic Unloading Spherical Diaphragm Subjected t o a Uniform Pressure Spherical Diaphragm Model with Axisymmetric Boundary Condition and Constraints by Gap Elements Load vs. Vertical Displacement a t the Center (Clamped Case) Deformed Shapes of the Diaphragm (Clamped Case) Load vs. Vertical Displacement a t the Center (Clamped Case) Finite Element Model of Pressure Vessel Initial Stress Distribution Effective Stress Distribution After 3 Hours of Creep . Due to Symmetry) Stepped Beam 2Shaped Cantilever Beam Nonlinear StressStrain Relationship for 2Shaped Beam Deformed Shapes of 2Shaped Beam LoadDeflection Curve for 2Shaped Beam Uniform Beam with a Gap Deformed Shapes of Beam During Static Loading (SOL 66 and SOL 99) Deformed Shapes of Beam During Static Unloading (SOL 66) Rotation (8.Simply Supported Beam with a Restrained Motion Simply Supported Beam: Loading Point Response Dynamic Loads for Transient Analysis Example of Combining Loads Options in TLOADi (Types 1. Vertical Displacement at the Center (Simply Supported Case) Deformed Shapes of the Diaphragm (Simply Supported Case) Vertical and Radial Displacement at the Center and Boundary Points vs. 2 and 3) Euler Buckling for Verification ElasticPlastic Buckling of a Clamped Spherical Cap with Flat Spot (Load vs. Vertical Displacement a t the Center (Sliding and Simply Supported Case) Cantilevered 2Shaped Beam (QUAD4 Model) Deformed Shapes (QUAD4 model) LoadDeflection Curve Spherical Diaphragm Subjected t o a Uniform Pressure Spherical Diaphragm Model with Axisymmetric Boundary Condition and Constraints by Gap Elements Load vs. SPC Forces at the Periphery Load vs. Central Deflection) Deformed Shape on Buckling Deformed Shape Upon Buckling with a Solid Model Simply Supported Beam Nonlinear StressStrain Relationship for Simply Supported Beam Clamped Plate (Quarter Only.
vs. Time C. Stress I<p vs.28 in) Comparison of Gauss Integration Schemes Using 8 x 16 MSCINASTRAN Model Finite Element Model Exposing Gap Locations Detail of Contact Region for Determining Local Stiffness Values Ir'. Axial Gap Forces for 66element and 132element Models Sections from Figure 11. vs.Stress Relaxation at the Junction of the Cylinder and End Closure Effective Creep Strain ( x 1 0 .1 Indicating which GAPS Close Displacement vs.5. Stress . Time for M l m = 1 Comparison of Free End Stress (Time History for M l m = 1) Comparison of Impact Stresses QUAD4 Model of Rectangular Plate with Line Load LoadDeflection at Center of QUAD4 Model with Line Load TRIA3 Model of Rectangular Plate with Line Load LoadDeflection a t Center of TRIA3 Model with Line Load QUAD4 Model of Square Plate with Uniform Pressure LoadDeflection at Grid 23 of QUAD4 Model with Uniform Pressure QUAD4 Model of Square Plate with Concentrated Load LoadDeflection at Grid 19 of QUAD4 Model with Concentrated Load Creep Strain vs. Time at 1000 psi Creep StrainRate vs. Stress C. vs.~ )After 3 Hours of Creep Impulsively Loaded Cylindrical Panel Displacement Response at Midspan (z=6.
was added for nonlinear dynamic analysis in Version 62 [1. had no nonlinear capabilities to mention [1. The current nonlinear capabilities were embodied in Version 62. The automated iteration method with convergence criteria was developed and released in 1981 as Version 61 with a new solution sequence. The GAP element and the nonlinear BEAM element were also introduced in this version. the method was not automatic and required the user's intervention for every iteration. Nonlinear material capabilities (plasticity and nonlinear elasticity) and the automated modifiedNewton's iteration method were included in SOL 66 and SOL 99. This was the first time the convective element coordinates (in ROD. released in 1982. another solution sequence. HEXA.4]. SOL 64. have been improved . However. released in 1970. released in MSCINASTRAN Version 62. In addition t o SOL 64 and SOL 66 for nonlinear static analysis.2. was developed for this purpose and released in 1978 t o provide a tool for such an iterative and incremental procedure in the geometric nonlinear analysis. and PENTA) and the gimbal angle (in QUAD4 and TRIA3) were provided for large displacements in MSCINASTRAN. TRIA3. SOL 99. The primary feature of nonlinear analysis is the requirement of the iterative and incremental processes t o obtain a solution. During the same period. A prototype solution sequence. SOL 66 [1.1 EVOLUTION The original version of MSCINASTRAN. Some groundwork for nonlinear capabilities had begun in 1973 and necessary precursors had been implemented in the element routines by 1977.31. The nonlinear capabilities. other than some limited usage of nonlinear loads and piecewiselinear analysis with a differential stiffness.1]. Limited nonlinear material capabilities became available for static analysis in this version. QUAD4. which facilitated nonlinear capability development. 1.Chapter 1 INTRODUCTION 1. the database technology had been introduced in MSCINASTRAN.
Major enhancements in Version 63 [1.and extended in the versions that followed. A new element.5] include creep analysis capability. resulting in reduced database storage space with a parameter SMALLDB. stiffness matrix updates. Additional capabilities for the nonlinear dynamic analysis in Version 64 consist of options for initial conditions (TIC). primarily for the snapthrough and postbuckling analyses. and unitindependent convergence criteria. implementation of the planestrain element using shell elements was most notable. and the quasiNewton updatelline search method (LSQN). bisection. essentially to provide better convergence. In addition. SOL 66 was reenforced with the Crisfield's arclength method. SOL 66 DMAP was also improved extensively. Enhancements in Version 65 pertain t o the iteration algorithm in static analysis. Large rotation capability has also been improved by introducing an auxiliary angle set (to remove the gimbal lock at 90') and a rotation vector approach (an option selectable by a . Adaptive bisection and recovery processes were also implemented for the Newton's iteration. discriminate use of line searches. including the creep time bisection. and builtin nonlinear buckling analysis capability controlled by a parameter BUCKLE. quasiNewton update method. adaptive arclength adjustment and bisection process. timely stiffness updates. Among the element enhancements. composite elements and the follower forces implemented in SOL 66. and staticlkinetic frictions. pertinent divergence criteria. was also added t o the nonlinear element library. Solution methods were further improved and expanded in Version 66. numerous minor enhancements and error corrections were incorporated in Version 64 for computational efficiency and userfriendliness. SOL 99 was overhauled with an adaptive time stepping (automatic adjustment and bisection) and twopoint direct time integration in concert with improved iteration strategies (quasiNewton updates. line search and dynamic matrix update scheme). The data recovery was enhanced by implementing new modules in SOL 66 and SOL 99 to provide the output for upstream superelements and the special output format for nonlinear element stresses and strains. Two additional options (RIKS and MRIKS) for arclength method were added and elaborated with BFGS updates. Adaptive solution algorithms were extended t o SOL 66 and the GAP element in Version 67. Major enhancements for the static analysis in Version 64 were improved PLOT module and SOL 66 for consolidated data recovery by combining linear and nonlinear elements. subincrements. nonlinear loads (NOLINi). Also included in SOL 66 for this version were follower centrifugal force (RFORCE) and the temperature dependent material capabilities (thermoelastic and creep under variable temperature). Additional capabilities include extended nonlinearelastic material capability to allow a bilateral stressstrain curve. TETRA. Options for static analysis (using a parameter TSTATIC) and follower forces were also added to SOL 99 in this version. The basic thrust of the adaptive GAP element was t o improve the convergence and accuracy in the contact problems using GAP elements by introducing adaptive penalty value adjustment. The nonlinear buckling analysis was supplemented in SOL 66 by RF66D84. and the composite beam element (PBCOMP) to allow user specification of the crosssectional shape. flexible restarts. Highlights include full utilization of the quasiNewton updates.
creep. and modification of internal relaxation iteration in solid elements (HEXA and PENTA). Minor enhancements in this version include introduction of numerical damping (by a parameter NDAMP) in the adaptive time stepping.parameter LANGLE 2). thermoelastic materials combined with MAT2 or MATS). linear analysis capability in SOL 99. . Structured solution sequences (SOL 106 and SOL 129) were also introduced in this version. initial anisotropy in nonlinear materials (plastic.
Material nonlinearity is an inherent property of any engineering material. plate and shell modeling. Material nonlinear effects may be classified into many catagories. Geometric nonlinear effects are prominent in two different aspects: geometric stiffening due t o initial displacements and stresses. and the internal relaxation iteration for material nonlinearity. HEXAs. For the transient response analysis. The geometric nonlinearity becomes discernible when the structure is subjected t o large displacement and rotation. The nonlinear element library of MSCINASTRAN consists of: r RODS. and TETRAS for solid modeling. Included in the current release of MSCINASTRAN are plasticity. r QUAD4s and TRIA3s for membrane.2 OVERVIEW OF NONLINEAR CAPABILITIES Nonlinear effects in structures occur mainly due to nonlinear materials and large rotations. implemented as a generalized nonlinear viscoelastic capability. The stiffness matrix updates are performed occasionally to improve the computational efficiency. r BEAMS for axially and laterally deforming line members. may be coupled with plasticity. . a number of options are available for implicit direct time integration. and follower forces due to a change in loads as a function of displacements.1. Creep and viscoelasticity. CONRODs. Contact problems exhibit nonlinear effects due to changes in boundary conditions. nonlinear elasticity. Refer to Table 1.2. iterations with convergence tests for acceptable equilibrium error. and stiffness matrix updates. r GAPS for contact and friction simulation. The primary solution operations are gradual load or time increments. These effects are included in the current release of MSCINASTRAN. and may be overridden at the user's discretion. A number of options of arclength methods are also available for snapthrough or postbuckling analysis of the static problems. Many sophisticated options are available for yield criteria and hardening behavior in plasticity. All of these attributes may be represented by nonlinear elements. and TUBES for unidirectional truss members. PENTAs. combined with adaptive and expedient iteration strategies similar t o those implemented for static analysis. and viscoelasticity. but the large deformation effect resulting in large strains is yet t o be implemented.1 for a summary of the nonlinear element properties. creep. The iterative process is based on the modifiedNewton's method combined with optional expeditious met hods such as the quasiNewton (BFGS) update and the line search. Primary operations for nonlinear elements are updating element coordinates and applied loads for large displacements. Nonlinear elements may be combined with linear elements for computational efficiency if the nonlinear effects can be localized.
because SOL 66 can be substituted for any capabilities in SOL 64. geometric. which are equivalent t o SOL 66 and SOL 99. was developed only t o analyze geometric nonlinear problems. and nonlinear buckling analyses. SOL 66 (or 106) is applicable t o static. SOL 64 will be eventually removed. and boundary nonlinearities in addition t o elements. Both of these solution sequences can accommodate superelements and provide easy restarts from the database. SOL 64. and userfriendly. They are SOL 106 and SOL 129. . Enhancements and new capabilities will be implemented t o fulfill this goal in the areas of material. efficient.Solution sequences 66 (or 106) and 99 (or 129) consolidate all the nonlinear features described above. taking advantage of the new executive system. quasistatic. Structured solution sequences are created in Version 67. and solution algorithms. The future goal for MSCINASTRAN nonlinear capabilities is t o provide a program that is more capable. the forerunner of these solution sequences. respectively. No new capabilities have been added t o SOL 64 since Version 63. SOL 99 (or 129) is primarily applicable t o dynamic transient response analysis with some limited static analysis capability.
* may be changed by the user. Edge nodes are not applicable to nonlinear elements. .Table 1.1 Summary of Properties of the Nonlinear Elements 3.2. 4.
the hierarchy of the nonlinear looping is shown in Fig.1. The Executive System processes the input data by IFP (Input File Processor) and the It general initialization.3 PROGRAM ARCHITECTURE MSCINASTRAN has a modular structure to separate functional capabilities which are organized under an efficient executive system. then establishes and controls the sequence of module executions in the OSCAR (Operation Sequence Control Array) based on the userspecified DMAP (Direct Matrix Abstraction Program) or solution sequence. Confining the discussion t o SOL 66 and SOL 99. the module uses scratch files. The Executive System allocates system files t o the data blocks in the FIAT (File Allocation Table) and maintains a parameter table for module interface. The solution sequence is a collection of module statements written in the DMAP language tailored t o process a sequential series of operations. If the size of the main memory is insufficient to complete an operation. DMAP is a kind of macro program using a data block oriented language. called functional modules. resulting in a specific type of structural analysis. solution of equations. The functional module consists of a number of subroutines. Modules communicate with each other only through secondary storage files. respectively. load increments. nonlinear solution processes comprise a number of internal iteration loops. and data recovery. . A functional module is capable of performing a predefined subset of operations. Nested in this DMAP loop. 1. This is known as a spill operation. The nonlinear solution sequences have builtin loops in the second phase for subcase changes.1. The Executive System is also responsible for the database management and all the input and output operations by GIN0 (General Input /Output Routines). which are known as Preface~operations. The program is divided into a series of independent subprograms. A typical solution sequence consists of three phases of functional operations: formation. and reduction of matrices. It is the Executive System that identifies every module t o execute by MPL (Module Properties List). and stiffness matrix updates. A module may communicate with the Executive System and with other modules through parameters. Modules utilize main memory dynamically. These modules are selfcontained t o perform iterations for converged solutions. Central t o the nonlinear processes are modules NLITER and NLTRD2 for SOL 66 (or 106) and SOL 99 (or 129). which reside in the secondary storage as an extension of the main memory. Solution sequences that process superelements have builtin superelement loops in the first and the last phases. which may be input and/or output variables of the module.3. called data blocks (matrix or table). assembly. Each module performs a certain function with input data blocks and produces output data blocks.
.AND PENTA) .1 Hierarchy of Nonlinear Looping 1... LOADS...3..INTERNAL RELAXATION (FOR BEAM.INTEGRATION I I r STRAIN SUBINCREMENTS.. ...3 . TEMPERATURES..(NMATD) ..... .. . .. . OUTPUTS)  r I LOAD..  r I r VOLUME . .. . ... I I 1 I STRESS COMPONENTS Figure 1. .(NLITER FOR STATIC) . Module Control Loops Subroutine Control Loops r I I I I SUBCASES (BOUNDARIES.. . ...HEXA.I I I r .DMAP Control Loops .INCREMENTS . . (GAUSS POINTS) ....2 .TIME STEPS (NLTRD2 FOR TRANSIENT) r I I l STIFFNESS MATRIX UPDATES  l ITERATIONS (VECTOR ARITHMETIC) LINE SEARCHES ELEMENTS (NLEMG) . .
4. Mechanical design is dictated by the strength. Case Control Section.1. Examples of input data are illustrated in Tables 1. dynamic. SOL 99 (or 129) provides noqlinear transient response analysis. Needless t o say. Any exceptions for nonlinear analysis will be explained in the relevant chapters that follow. and stability characteristics of the structure. The basic input data required for a finite element analysis may be classified as follows: a Geometric data a Element data a Material data Boundary conditions and constraints Loads and enforced motions Solution methods The first three classes of data may not be changed during the course of an analysis whereas the last three classes of data may be changed in midcourse via subcases under the Case Control Section. and buckling analyses. SOL 66 (or 129) is designed for static. . Executive Control Section.2 for nonlinear static and dynamic analyses.1 and 1. respectively. The type of desired analysis is specified in the Executive Control Section by using a solution sequence identification. each of which is designed for specific applications. still has the most popular format consisting of an optional NASTRAN header. quasistatic. Review Reference [I.1 for details of the general features.4.4 RUDIMENTS of USER INTERFACE The input data structure. and Bulk Data Section. as was designed originally. All the features and principles for the user interface are common in MSCINASTRAN. all the features of the user interface for nonlinear analysis 1 are compatible with those for the linear analysis. MSCINASTRAN provides the analysis capabilities of these characteristics with solution sequences.
4.50 $ DIAGNOSTIC PRINTOUT TIME 8 $ FOR VAX $ END OF EXECUTIVE CONTROL DATA CEND TITLE = TEST OF CTETRA ELEMENT (CUBE SUBJECT TO UNIAXIAL LOADING) SEALL = ALL $ FOR ALL MATRICIES DISP = ALL STRESS = ALL SPC = 100 $ RESTART PARAMETERS PARAM SUBID PARAM LOADINC PARAM LOOPID SUBCASE 1 SUBTITLE = ELASTIC .LOAD TO 850. PSI LABEL = LOAD BEYOND YIELD LOAD = 100 NLPARM = 100 SUBCASE 4 SUBTITLE = ELASTIC .LOAD TO 1000.V65 $ CJS 18JULY85 SOL 66 $ NONLINEAR STATIC ANALYSIS DIAG 8.UNLOAD COMPLETELY TO 0.Table 1. PSI LABEL = FULL UNLOAD LOAD = 200 NLPARM = 200 $OUTPUT(PLOT) $ PLOT DATA ADDED FOLLOWING THIS LINE IF NEEDED $ END OF CASE CONTROL DATA BEGIN BULK $ PARAMETERS FOR NONLINEAR ITERATION N0 UPW 1 AUTO NLPARM 50 UPW N0 8 SEMI NLPARM 100 UPW N0 2 AUTO NLPARM 200 ENDDATA .1 Example Input Data for Nonlinear Static Analysis ID TETRAC. PSI LABEL = LOAD TO YIELD LOAD = 50 NLPARM = 50 SUBCASE 2 SUBTITLE = PLASTIC .
V65 $ SHL SOL 99 $ NONLINEAR TRANSIENT ANALYSIS $ DIAGNOSTIC PRINTOUT DIAG 8. YMAX = 15.23 SET 4 = 20 $ CONROD DISP = 1 STRESS = 3 $ FOR GAP FORCES ELFORCE = 4 LOADSET = 20 'SPC = 100 $ RESTART PARAMETERS $PARAM LOOPID $PARAM STIME SUBCASE 1 $ UP TO 0.Table 1.200 SET 3 = 22 $ .2 Example Input Data for Nonlinear Dynamic Analysis ID VGAP.5 SECONDS DLOAD = 210 TSTEPNL = 22 SUBCASE 2 $ UP TO 1 SECOND DLOAD = 210 TSTEPNL = 21 OUTPUT(XYPLOT) CSCALE = 1.202. XYPLOT DISP RESP/201 (TI) .200(TI) $ END OF CASE CONTROL DATA BEGIN BULK .4.50 $ FOR VAX TIME 5 CEND $ END O EXECUTIVE CONTROL DATA F TITLE = GAP VERIFICATION BY NONLINEAR TRANSIENT ANALYSIS SEALL = ALL $ FOR ALL MATRICIES SET 1 = 201 $ .3 XTITLE = TIME IN SECONDS YTITLE = DISPLACEMENT IN INCHES YMIN = 0.0025 2 AUTO TSTEPNL 21 200 ENDDATA .202(TI) .
loads and solution methods may change from subcase to subcase on an incremental basis. Confining the discussion to SOL 66 (or 106) and SOL 99 (or 129). . However. constraints can be changed from subcase t o subcase only in the static solution sequence. Selections made in an individual subcase supersede the selections made above the subcases.4.1 Case Control Section The Case Control Section may be regarded as a central control panel where select switches for input and output are set. The primary purpose of the Case Control ~ i t is to define a subcases. Table 1. by specifying set numbers with keywords.1. The subcase structure in MSCINASTRAN provides a unique means of changing loads. and solution methods by making selections from the Bulk Data. plot.3 summarizes the Case Control Data for nonlinear analysis. boundary conditions.4. etc. The subcase structure also allows the user to select and change output requests for printout. As a result. Any selections made above the subcase specifications are applicable t o all the subcases.. the subcase structure determines a sequence of loading and constraint paths in a nonlinear analysis.
4.3 Summary of Nonlinear Case Control Data Load Selection LOAD CLOAD DLOAD LOADSET NONLINEAR IC Solution Met hod Selection METHOD NLPARM TSTEPNL Output Requests DISPLACEMENT VELOCITY ACCELERATION ELFORCE STRESS SPCFORCES NLLOAD OUTPUT (PLOT) Superelement Control SUPER Selects static loading condition Selects static load combination f ~ r ' s u ~ e r e l e m e n t s Selects dynamic loading conditions Selects static load sets defined on the Bulk data LSEQ Selects nonlinear loading (NOLINi) for transient response Selects initial conditions for transient response Selects methods for eigenvalue analysis Selects iteration methods for nonlinear static analysis Selects iteration methods for nonlinear transient analysis Requests output for displacements of physical points Rquests output for velocities of physical points Requests output for acceleration of physical points Requests output for element forces Requests output for element stresses Requests output for constraint forces of SPC points Requests output for NOLINi in transient response Requests the beginning of the plotter output SEALL SEKR SELG SELR SEMG SEMR Specifies the superelement identification number and the load sequence number Combines the functions of SEMG. SEMR. SELG. and damping matrices will be generated Specifies the superelement identification numbers for which : the mass and d a m ~ i n ematrices will be assembled and reduced . mass. and SELR Specifies the superelement identification numbers for which stiffness matrices will be assembled and reduced Specifies the superelement identification numbers for which load vectors will be generated Specifies the superelement identification numbers for which the static load matrices will be assembled and reduced Specifies the superelement identification numbers for which stiffness.Table 1. SEKR.
1.4. CPENTA and CTETRA Defines properties for CTUBE .4 Summary of Nonlinear Bulk Data Element Connectivity CBEAM CGAP CHEXA CONROD CPENTA CQUAD4 CROD CTETRA CTRIA3 CTUBE Defines connection for beam element Defines connection for gap or frictonal element Defines connection for sixsided solid element Defines connection and properties for rod Defines connection for fivesided solid element Defines connection for quadrilateral element with bending and membrane stiffness Defines connection for rod with axial and torsional stiffness Defines connection for foursided solid element Defines connection for triangular element with bending and membrane stiffness Defines connection for a tube Element Properties PBCOMP PBEAM PCOMP PGAP PROD PSHELL PSOLID PTUBE Defines properties for composite CBEAM Defines properties for CBEAM Defines properties for composite material laminate Defines properties for CGAP Defines properties for CROD Defines properties for CTRIA3 and CQUAD4 Defines properties for CHEXA.4. All the input data specially designed for nonlinear analysis will be described in the relevant chapters that follow. The nonlinear Bulk Data are summarized in Table 1.4. where all details of the input data are specified.4. Descriptions of the Bulk Data pertinent t o nonlinear analysis are listed in Appendix A. Some of the Bulk Data may not be used during a program execution if they are not selected in the Case Control Data.2 Bulk Data Section The primary input media is the Bulk Data. Table 1.
QUAD4 and TRIA3 Defines pressure loads on surfaces of HEXA. TETRA TRIA3. and QUAD4 elements Defines load due to centrifugal force field Defines temperature at grid points Defines temperature field for surface elements Defines temperature field for line elements Specifies initial values for displacement and velocity Defines loads as a function of time Defines iteration methods for nonlinear static analysis Defines arclength methods for nonlinear static analysis Specifies integration and iteration methods for nonlinear transient analysis Defines eigenvalue extraction method for buckling analysis .Material P r o p e r t i e s 2REEP MAT2 MAT8 MAT9 MATS1 rABLESl TABLEST Defines creep material properties Defines anisotropic material properties for shell elements Defines orthotropic material properties for shell elements Defines anisotropic material properties for solid elements Defines properties for plastic and nonlinear elastic material Defines a function for stressdependent material properties Combines many TABLES1 entries for temperature dependent material properties Constraints SPC SPCl MPC Loads CLOAD FORCEi LSEQ MOMENTi NOLINi PLOAD PLOAD2 PLOAD4 RFORCE TEMP TEMPPi TEMPRB TIC TLOADi Solution M e t h o d s NLPARM NLPCI TSTEPNL EIGB Defines singlepoint constraints and enforced displacements Defines singlepoint constraints Defines a linear relationship for two or more degrees of freedom Defines a static load combination for superelement loads Defines concentrated load at grid point Defines static load sets for dynamic analysis Defines moment at a grid point Defines nonlinear transient load Defines pressure loads on QUAD4 and TRIA3 Defines pressure loads on shell elements. PENTA.
and 129 If K6ROT > 0.4. 106. There are two different types of parameters in MSCINASTRAN: user parameters (V. users are authorized t o change the initial values by specifying PARAM data in the Bulk Data Section or ocassionally in the Case Control Section.name in the DMAP) and DMAP (nonuser) parameters.1. then stiffness is added t o the normal rotation for QUAD4 and TRIA3 elements. and 129 By default in SOLs 66. 106. This capability should be used in the restart runs.0. which can be overridden by a DMAP initialization.3 Parameters Parameters constitute an important part of input data in their own right because they maintain communications among the user. and is intended primarily for geometric nonlinear analysis.333333 in SOL 99 or 129 The value of BETA is the transient integration control factor for the NewmarkBeta method with AUTO and TSTEP methods.5 lists the parameters used in nonlinear solution sequences 64. and 129. 99. 99. The recommended value is in the range of 10. 106 and 129. Default = 1 for SOL 64 only This parameter is used t o control restarts but not recommended. Default = 0. DBDRNL.0 for SOLS 66. Parameters are initialized in the MPL. used for requesting special features or specifying miscellaneous data. For the user parameters. . Parameters are singlevalued input data. Many options available in MSCINASTRAN are controlled by userspecified parameters.Y. nonlinear buckling analysis is activated. K6ROT Default = 100. This is an alternative method t o suppress the grid point singularities. BETA BUCKLE CNSTRT DBDRNL Default = 0 for SOLS 66. Table 1. Default = 1 If a value of 1 is specified in SOL 66 or 106. the database directory is printed upon completion of the nonlinear analysis phase.4.0 t o l. and modules. 99. 66. PARAM. 99. 1 will suppress this printout. Key parameters which allow options for nonlinear analyses are listed below: AUTOSPCR Default = NO in SOL 66 or 106 Provides for automatic singlepoint constraints for the residual structure.OE+4. Modules may change the parameter values while the program is running. 106. DMAP.
which does not allow user's specification of any other value.0 to 0. Default = 1 If LGDISP=l. Default = 0 Defines the loop number for initial conditions in SOL 66 or SOL 99 for restarts. Set NLAYERS=l for efficiency if no bending is selected (MID2 = 0 or 1 on all PSHELL data entries). with zero being no numerical damping. The numerical damping increases as the value of NDAMP increases. Default = 0. maximum = 12) This parameter defines the number of layers used to integrate through the thickness of isotropic material nonlinear QUAD4 and TRIA3 elements. LGDISP LOADINC LOOPID MAXLP NDAMP NLAYERS . Default = 5 (minimum = 1. large rotations (using PARAM.01 is adequate. If LGDISP=1. Recommended range is from 0.LGDISP) are treated by the gimbal angle approach through a default option. Default = 1 Defines the beginning load step of the subcase in which the restart analysis is to be initiated in SOL 66.1. A larger value of NLAYERS will give greater accuracy at the cost of computing time and storage requirements.LANGLE Default = 1 In SOL 66 (or 106) and SOL 99 (or 129). This parameter is set t o 10 in SOL 66 (or 106).LANGLE with a value of 2. Default = 5 Maximum number of iterations allowed internally for element relaxation and material point subincrement processes in SOL 99 (or 129). no large displacement effects will be considered.025 for SOLs 99 and 129 only Numerical damping introduced in the twopoint integration (ADAPT method in SOL 99 or 129) for numerical stability (implemented in Version 67). follower force effects will be ignored while large displacement is accounted for. A rotation vector approach is implemented in Version 67. and can be selected by PARAM. If LGDISP=2. For most cases the value of 0. all the nonlinear elements allowed in SOLs 66 and 99 will be assumed to have large displacement effects (updated element coordinates and follower forces).
set SKPLOAD= . If the matrix of the present subcase is the same as in the previous case (sometimes called right side iterations). nonlinear modal analysis capability is built in SOL 106 with PARAM. Default = 1 In SOL 99 (or 129). In SOL 106. set OUTPUT=1 (and usually set NODATA=1). as an efficiency feature. The printout LOOPID=x is given for each subase and load increment in SOL 66. skipping nonlinear computation loop.1. The default causes recalculation of the loads. Default = 1 This parameter is used t o control the calculation of loads. even if they do not change (SOL 64 only). NMLOOP. This option also requires the presence of the METHOD command in the Case Control section and EIGR or EIGRL entry in the Bulk Data section. OUTPUT SDATA SKPLOAD SKPMTRX SLOOPID . Default = 0 In SOL 99 (or 129) transient analysis. normal modes can be computed with the updated stiffness if PARAM. but will usually require more iterations for convergence. If the loads of the present subcase are the same as in the previous case. an efficiency feature in geometric nonlinear analysis (SOL 64). If it is desired t o skip operations in the loop prior to stress recovery. Setting SLOOPID=x and providing the static data base will cause SOL 99 t o start from the static deformed position. this parameter identifies the initial conditions from a previous SOL 66 (or 106) nonlinear static solution. set SKPMTRX=1. SDATA=1 with the proper value (last value) for LOOPID allows data recovery restarts. n is specified where n is the desired LOOPID. Use of this parameter results in a lower solution cost per iteration.NMLOOP Default = 0 In Version 67. NMLOOP. Default = 0 In SOL 64 this parameter is used in geometric nonlinear analysis to control restarts in a loop. Default = 1 This parameter is used t o control the matrix reduction and decomposition.
the database will retain only those data blocks corresponding to the load steps for which an output request is made by the "INTOUT" field in the NLPARM entry. PARAM. STIME Default = 0. Default = 0. Default = 0 This parameter is used t o control restarts in a loop for geometric nonlinear analysis (SOL 64). TABS. 273. TABS. SUBID is the subcase identification number (from a previous SOL 64 run) used for an initial guess. Restarts will be restricted t o those starting from these steps. i. restart after time limit).SMALLDB Default = 1 (No longer available from Version 66) This parameter is provided in SOL 66 t o allow a small database option in connection with the "INTOUT" field of the NLPARM entry. With a value of 1.69 when Celsius is used when Fahrenheit is used SUBID SUBSKP TABS Refer t o the Bulk Data entry CREEP for a creep analysis with SOL 66 (or 106). not the subcase ID number. . Default = 0 Controls the skipping of subcases which are not desired in geometric nonlinear analysis. The default discards all previous results and is equivalent t o a cold start. SUBSKP can be used on restart t o skip completed cases or can be used t o delete the static (first) or differential (second) subcase. Default = 1 Defines the record number of the case control data (for SEID=O) in which the beginning load step is defined for the restart in SOL 66 (or 106). If it is desired t o skip operations in the loop prior t o FBS (for example.0 In SOL 99 (or 129) restarts from previous SOL 99 runs. Refer t o PARAM. set SSG3 = 1 (and usually set NODATA = 1).. the user provides STIME=tN. where t~ is the last time step of the subcase to be continued with a new or changed subcase in the new run. Thus the loading and printout will start from t~ as though the original run contained the new subcase data. In SOL 64.16 PARAM. The number of subcases t o be skipped is input.SIGMA for heat transfer analysis.e. 459.0 This is used t o convert units of the temperature input (OF or OC) t o the absolute temperature.
Default = 1 (SOL 99 or 129 only) If TSTATIC=l. The units for W 3 and W4 are radians per unit time. Negative terms indicate that the differential stiffness has introduced a structural instability. For subcases 3 onward. 2 for Newton's method in SOL 66. The default values of 0. the solution iterates within the subcase until the incremental strain energy is less than TESTSE. but a stable solution exists).0 x Used t o control iterations within subcases for SOL 64. TSTATIC W3. the output is skipped if the strain energy is greater than TESTSE. Continue if negative terms occur. The instability may be real (structural buckling) or mathematical (the current iteration appears unstable.0 The damping matrix for transient analysis is assembled from the equation The values of W3 and W4 are used for w3 and w4. respectively. do not use differential stiffness.0 for W3 and W4 cause the associated terms in the damping matrix t o be set equal t o zero. This option is available only with the adaptive method. Do not use differential stiffness.W4 Default = 0. For subcases 1 and 2. . regardless of the presence of the PARAMeter G or [ K j d ] terms. If negative terms exist. specified by ADAPT in the TSTEPNL. +1 for arclength method in SOL 66 and SOL 106 Controls SOL 64 and SOL 66 solution sequence when negative terms are encountered on the factor diagonal of matrix decomposition.TESTNEG Default = 1 for SOL 64. TESTSE Default = 1. a static solution may be obtained in SOL 99 by ignoring inertial and damping forces. Value 1 +lor 0 2 +2 Result Stop if negative terms occur.
0 1 1 1 1 E I DBCLEAN DBDICT DBDRNL DBINIT DBNAME DBNBLKS DBSETi DBSORT DLOAD DOPT I B B E B B B B B B E IBI .2 CNSTRT COUPMASS CURV CURVPLOT B B E E I E E E I B E B E I NO 0 0.O 1.2.4.3 CM1.2 CK1.Table 1. B B E B B B B B B I 0 1 0 0 I Empties database Prints database dictionary at the end and beginning Prints database dictionary in the nonlinear loop Initializes database with empty data blocks Specifies database name Sets maximum number of GIN0 blocks Specifies database subsets for storing and fetching Sorts printout of database dictionary Indicates no other changes except for loads Controls x spacing of curves for CURVPLOT I B E B B B B B B DBOl 4000 0 5 0 0 .5 PARAMeters used in Nonlinear Solution Sequences NOTES: B  Usable in t h e Bulk Data Section only Usable in either t h e Bulk D a t a or Case Control D a t a Section SOLUTION E I PARAMETER 1 I NAME ASING AUTOSPC ( SEQUENCE I DEFAULT / B B E E E E 0 NO I 1 DESCRIPTION Removes singularities by appropriate techniques Specifies automatic singlepointconstraints I AUTOSPCR I BAILOUT BETA BUCKLE CB1.O I Specifies control factor for NewmarkBeta method Selects nonlinear buckling analysis for restarts Selects superelement damping matrix coefficients Selects superelement stiffness matrix coefficients Selects superelement mass matrix coefficients Controls restarts for constraint change Generates coupled rather than lumped mass matrices Computes stress data in a material coordinate system Requests xy (curve) plots E E E E E 1.33333 I Specifies automatic SPC7sfor residual structure Terminates superelement program with singularities 1 1 1.
OE8 1.OE+5 1 .0 1.0 100.OE8 1.OE8 1 DESCRIPTION Prints singularities of stiffness matrix Turning parameter for generalized dynamic reduction Turning parameter for generalized dynamic reduction Turning parameter for generalized dynamic reduction Determines printed singularities compared to EPPRT Terminates run for superelement errors Requests element measure and volume computation Generates SEQUID on SEQGP entry Prints all elements connected to each grid point 66 99 E E E E E E B B B B E E E B E E E E B E 2 10.OEf30 0 1.PARAMETER NAME EPPRT EPSBIG EPSMALC EPSMALU EPZERO ERROR EST FACTOR GPECT SOLUTION SEQUENCE 64 DEFAULT 1.0 1 E E E E LANGLE LFREQ LGDISP LMODES LOADINC LOOPID MAXLP II II E E l I \ E l 0.0 1 0 IE/ I E E ( E E E E E I 1 I I I 1 0 5 MAXRATIO MODEL MPCX NDAMP NEWDYN I IB IB I B B I 1.025 1 B ( B E E 1 Stores several models in the database Controls MPC and rigid element processing on restarts I I Executes MTRXIN. and GKAM modules Specifies numerical damping in ADAPT method .~ e t e r m i n ethe singularity of the stiffness matrix s 0 0 0.OE10 1. GKAD.OE+8 1.000 1 I GRDPNT G HFREQ INRLM KDIAG KGROT  I B I E I E E E E I 1 1 Executes Grid Point Weight Generator Specifies uniform structural damping coefficient Defines upper limit of frequency range Requests appending of inertia relief modes Applies value of a spring constant to all DOF7s Assigns stiffness to normal rotation of QUAD4 and TRIA3 Selects gimbal angle approach for large rotations if 1 and rotation vector approach if 2 Defines lower limit of frequency range Selects large displacement effects Defines number of lowest modes Specifies load increment ID for restarts Specifies loop ID in the database for restarts Maximum number of iterations for internal loop 1 0.
~ 8 .Ld.~ pas? s1o~puo3 su1syuoy3am sno!~nds s ~ o q u o 3 0 3 a3XLON TO 3 8 8 8 H33INON VLVaON spa y ~ n gy sa%usy:. 3 .P'T uo!qDnpal 3!mvubp pmua% JOOJx y s m .CNIN a~npom o s s a 3 o ~ d ~ Bupuanbas u pasn uorldo spalas y I 8 8 8 b 3 ~ ~ 3 ~ .10$3a~ua%!a slu!~d ~g+30'1 3 3 3 3 3 8 8 ZAIHdXd JLnO8d LSd3Xd LSOd 3SINXtd I . ~m0.C T 3 3 3 1 3 131 I I T 0 T I I I I I 1 3 13181 dnSLOld LO?d uo!pnpa. ou sals3rpu1 u   sluamaia 11s l a ~ alvlodlalu! 03 o ~ x n slsanbax 3 01 3 S.q lnopur~d 73 puv uo!pn3axa sassaxddns qnoquy~d dlpsln8u!s sassa~ddns assqsqop u! sy301q s l v p %u!ssa30~dqsod salops saBvssam $old pam~ojapunsassa~ddns s?uawala~adnso j s ~ o pamlojapun saysm l ~d $old pam~ojapun sdurnf op TS~.91 .
0 121 1.OE+36 E E E E 0 1 B B B E E E E E E E 0 1.W4 I B B B I E B E I B E I 1 I Specifies subcase ID for restarts Skips a number of subcases for restarts Deletes ungrouped grid points Controls punch for response spectra Converts units of temperature input t o absolute Tests for negative terms on factor diagonal of matrix Tests for strain energy for iteration control Selects static solution with TSTATIC=l in SOL 99 using ADAPT method in TSTEPNL Selects output type for degrees of freedom Controls sets for row list output option Multiplies terms of structural mass matrix by value Selects frequency for conversion of damping I 0 0 2 0.0 .PARAMETER NAME SEMAPOPT SEMAPPRT ) I SOLUTION SEQUENCE 64 66 B B I 99 B B 1 I I DEFAULT 42 3 SENAME 0 1 1 DESCRIPTION Places exterior superelement grids on CSUPER Prints SEMAP table I SENAME SEQOUT SKPLOAD SKPMTRX SLOOPID I B E E E B I E B I Identifies superelement by name Controls output options for NEWSEQ Skips load vector calculation Skips matrix reduction and decomposition Specifies loop ID from SOL66 database for restart I B 0 1 I SMALLDB SOLID SPCGEN SSG3 START STIME ( B B B B I E B E B I B E B E I Selects small database option Stores several solutions in same database Places automatic SPC's in S P C l format Skips operations prior t o FBS Defines number of grid points for NEWSEQ Specifies time step for restarts I 0 0 0 0 0. O 0.0 I SUBID SUBSKP SUPER TABID TABS TESTNEG TESTSE TSTATIC USETPRT USETSEL WTMASS W3.
(All items are stresses unless otherwise denoted) Word or Element Name Item Component (Code) External grid point ID CBEAM 2 C (alphanumeric value) 3 (94) Long. stress at point C I / 1 I Equivalent stress Total strain Effective plastic strain Effective creep strain (Item codes 3 thru 8 are repeated for points D. The codes for linear elements can be found in the User's Manual.1. Vol. Then the entire record (from 2 through N) is repeated for end B of the element.) CGAP (86) COMPX SHEARY SHEARZ AXIALU SHEARV SHEARW SLIPv SLIPW Grid/ Gauss External grid ID (0 = center) STRESSX STRESSY STRESSZ STRESSXY STRESSYZ STRESSZX CHEXA (93) . E and I?.4 Nonlinear Element Output Codes for Plotting Nonlinear elements (which reference nonlinear geometry and/or material) have different codes from linear elements. The following codes may be used in SOL 66 (or 106) and SOL 99 (or 129).4. 11.
if plane stress) STRESSY (at 21.Element Name (Code) CHEXA (93) (cont .) Word or Component Item Equivalent stress Effective plastic strain Effective creep strain STRAINX STRAINY STRAINZ STRAINXY STRAINYZ STRAINZX Items 3 through 17 repeated for 8 corners Same as CROD CONROD (92) CPENTA (91) CQUAD4 (90) Same as CHEXA but with only 6 corners Z1 = Fibre distance 1 (Plane stress only) STRESSX (at 21. if plane stress) STRESSZ (Plane strain only) STRESSXY (at 21. if plane stress) STRAINX (at 21. if plane stress) Equivalent stress (at 21. if plane stress) Items 2 through 13 repeated for fibre distance 22 (Plane stress only) . if plane stress) Creep strain (at 21. if plane stress) Plastic strain (at 21. if plane stress) STRAINY (at Z1. if plane stress) STRAINZ (Plane strain only) STRAINXY (at Z l .
Element Name (Code) CROD (89) Word or Component 2 3 4 5 6 7 Item Axial stress Equivalent stress Total strain Effective plastic strain Effective creep strain Linear torsional stress Same as CHEXA except with only 4 corners CTETRA (85) CTRIA3 (88) CTUBE (87) Same as CQUAD4 Same as CROD .
If there is a change in constraints due to contact during loading. r The size of the model should be determined based on the purpose of the analysis.. the tradeoffs between accuracy and efficiency. the model is accepted for nonlinear analysis by SOL 66 or SOL 99. r The model should be verified prior t o the analysis by some visual means. The key t o this issue is t o know where the loading point is in the loaddeflection curve of the critical area. and occasionally on environmental conditions.1.1]. Intuitively. use TRIA3 and TETRA only for geometric or topological reasons. a finer mesh in the area of stress concentrations. Nonlinear analysis requires better insight into structural behavior. The material nonlinearity is characterized by material properties. the material nonlinear effects may or may not be significant depending on the magnitude and duration of the loading. First of all. symmetric considerations for simplification. a simple model should be the starting point. such as plots and graphic displays. i. The anticipated stress level would be a key t o this issue. r Substructuring should be considered for the modularity of the model and/or synergism between projects and agencies involved. r Element types and the mesh size should be judiciously chosen.e. if it consists of: . the type of nonlinearities involved must be determined. For example. avoid highly distorted and/or stretched elements (with high aspect ratio). the problem may be classified as a boundary nonlinear problem and would require GAP elements or some surface contact algorithm. There is no distinct limit for large displacements because geometric nonlinear effects are related t o the dimensions of the structure and the boundary conditions.5 NONLINEAR CHARACTERISTICS AND GENERAL RECOMMENDATIONS Modeling for nonlinear analysis is not exempted from the guidelines for good modeling practice pertaining t o linear analysis [1. and systematic numbering of nodal points and elements for easy classification of locality. and the scheduled deadline. r Prior contemplation of the geometric modeling will increase efficiency in the long run. geometric nonlinear effects should be significant if the deformed shape of the structure appears distinctive from the original geometry without amplifying the displacements. The geometric nonlinearity is characterized by large rotations which usually cause large displacements. However. As for MSC/NASTRAN. otherwise. Factors t o be considered include selection of coordinate systems. r Discretization should be based on the anticipated stress gradient.which are summarized as follows: r The analyst should have some insight into the behavior of the structure to be modeled.
other potentially nonlinear elements in the residual structure become actively nonlinear only if a parameter. Such a nonlinearity is characterized by a loading path. r Segregate the linear region by using superelements and/or linear elements if possible. additional recommendations are imperative for nonlinear analysis: r Identify the type of nonlinearity and localize the nonlinear region for computational efficiency. perform a linear analysis by SOL 61 (or 24) or SOL 69 (or 27) prior t o the nonlinear analysis by SOL 66 or SOL 99. Use a finer mesh if severe element distortions or stress concentrations are anticipated. If unsure. and contact problems with frictional forces. makes the problem more complicated with another nonlinearity known as the Baushinger effect. r any GAP element (CGAP and PGAP). r any active nonlinear material data (specified on MATSl and/or CREEP). r Be prepared for restarts with the database properly stored in cases of divergence and changing constraints or loading paths via subcases. however. LGDISP. The cyclic loading. All the GAP elements are always actively nonlinear if included in the residual structure. r The subcase structure should be utilized properly t o divide the load or time history for conveniences in restarts. not to exceed 20 load steps (for SOL 66) or 200 time steps (for SOL 99) in each subcase. and database storage control. which can be controlled by multiple subcases in SOLS 66 and 99. This nonlinearity is manifested in the pathdependent problems such as plastic deformation. However. data recovery. . which belongs t o this category of nonlinearity. r The load or time for the subcase should then be further divided into increments. Notice that the potentially nonlinear elements can be used as linear elements. r The nonlinear region usually requires a finer mesh. With these points in mind. respectively. not t o mention changing constraints and loading paths. The model may consist of superelements. an incremental process may not be required. mixed with any type of linear elements. When the analysis involves a pathindependent nonlinear problem such as a purely geometric nonlinear problem. is used and/or if they use the nonlinear material data specified on the MATSl and/or CREEP data entries. The nonproportional loading introduces additional nonlinearity to the structural mechanics problem. r or any combination thereof. but only the residual structure (superelement 0) may consist of nonlinear elements. For computational efficiency. the analysis of such a problem does frequently need incremental solutions. creep deformation.r a parameter LGDISP (for geometric nonlinearity).
. r Understand the basic theory of plasticity or creep before using these capabilities. In particular. r For the transient analysis by SOL 99. refer to errors and limitations listed in the User's Manual Volume 11. because they are unit dependent.r Many options are available in solution methods to be specified on NLPARM (for SOL 66) or TSTEPNL (for SOL 99) data entries. Caution should be exercised in preparing input data for creep. the closed gap stiffness should not exceed the stiffness of the adjacent degree of freedom by 1000 times. r The time step size for a transient response analysis should be carefully determined based on the highest natural frequency of interest because it has significant effects on the efficiency as well as the accuracy. The defaults should be used on all options before gaining experience. some damping is desirable and the massless degree of freedom should be avoided. Normal rotation for QUAD4 and TRIA3 should be restrained by a parameter KGROT when the geometric nonlinearity is involved. For any anomalies. r Caution should be exercised in specifying GAP properties.
g.=O axj where aij are stress tensor components. b. These systems of governing differential equations must be satisfied for every infinitesimal element throughout the domain of the continuum.j +b. may be determined by solving these systems of equations supplemented by . where u.g. and xj are space coordinates. a The constitutive relations represented by stressstrain relations.Chapter 2 BASIC CONSIDERATIONS FOR NONLINEAR ANALYSIS 2. are displacements. are body forces. for linear elasticity. namely displacements. for a small defor mation. The complete set of state variables. e. where ~ k are l strain tensor components and Dijkl are elastic constants..1 DISCRETE SYSTEM FOR A NONLINEAR CONTINUUM MODEL Theories in solid mechanics are dictated by three governing relationships [2. a The compatibility represented by straindisplacement relations.1]: a The state equilibrium that requires da. e..
. The system may even be nonconservative. namely discretization. The first phase of the structural analysis is the idealization of a physical system into a simpler and more manageable engineering problem. The idealization process involves simplifications of the geometry. For the discrete system. boundary and joint conditions. the real nature is more complicated and inherently nonlinear. Even the nonlinear solutions that we seek deal with only a small subset of special cases in a general category of nonlinear problems. When the nonlinear system is confronted. State variables are the displacements (displacement method or stiffness approach) of the nodal points which carry fictitious forces representing distributed stresses actually acting on the element boundaries. If the idealized structural system renders a problem that cannot be resorted t o a classical method of analysis. The material nonlinearity is manifested in the constitutive relations. and in dynamic situations by initial conditions as well. the governing equations should be satisfied throughout the history of load application.boundary conditions. however. load. The compatibility is ensured by the displacement continuity between elements. the discretization is applied t o time. and loading conditions. Changes in constraints affect the boundary conditions. etc. however. empirical observations. The material constitutive laws are satisfied at the integration points of the element. it introduces numerous mathematical singularities which may complicate computational processes. governing differential equations are converted t o algebraic equations. the efficiency of modern digital computers makes it feasible t o apply complicated computational procedures t o the complex systems of engineering problems. Finite elements represent spatial discretization of a continuum. which constitute contact problems. no general mathematical solutions exist and superposition no longer applies. The finite element model represents a structure by an assemblage of finite elements interconnected a t nodal points. The linear system is a very particular case of a general problem. experimental data. they do not immediately impose nonlinearity. Fortunately. However. and classical solutions. That is. for numerical analysis. and material properties by using piecewise linear curves. a numerical model poses new dimensions t o the discretization in addition t o the ndimensional Euclidean space. As such. The geometric nonlinearity is pronounced in the straindisplacement relations. but it also affects the equilibrium equation by changing applied loads. that the compatibility of the nonconforming elements is ensured by a patch test. While discretization allows approximate solutions by numerical methods. further idealization is required. Most of the known solutions for the solid mechanics problems are based on ideal geometry and linear approximations. using engineering intuitions. For the nonlinear problems. The equilibrium requirements are satisfied at nodal points by the nodal force balance. When nonlinearity has t o be taken into account for large displacements and/or stresses. It is noted.
Now it remains t o determine admissible functions expressing the arguments of the functional I in terms of state variables {u). can be expressed in terms of nodal displacements using interpolation functions known as shape functions. representing a total potential of a continuum [2. which have simple geometrical shapes. which aggregates into a whole region when assembled. (such as a gravity load).22.. Considering a threedimensional continuum for a nonlinear problem. traction forces ti a t the boundary surface (such as pressure loads).e. {u). The finite element method can be characterized by the following features distinguished from the conventional Ritz methods or the matrix method for frame structures: r The whole region of the system is divided into numerous subdomains. called finite elements. N. i. the stationarity condition results in where the dots and 6 denote infinitesimal increments and arbitrary variations.e. and concentrated forces pi. where U is the strain energy of the system and W is the potential energy of the external loads. which are valid throughout the whole region and I satisfy the boundary conditions. The functional II is so called because it involves the integral of implicit functions of the state variables.entia1 of the system must be stationary with respect to the state variables (displacement) for equilibrium to be ensured. The equilibrium equations can be obtained by invoking the principle of virtual work or the Ritz method. where {u) is a displacement vector consisting of all nodal points of the element. . The lefthand side represents variations in the strain energy increment and the righthand side represents variations in the external work which consists of body forces b.. respectively.51. {fi). which implies that the total pot. The variational process is limited t o each finite element. r The admissible displacement field within each element.2 FINITE ELEMENT FORMULATION FOR EQUILIBRIUM EQUATIONS The variational principle renders the system governing equilibrium equations when applied t o a functional IT. i. i.2..e.
4). The nodal forces of an element can then be expressed as . (2. Then the element boundary stresses are statically equivalent t o the nodal forces which balance the applied external loads. Notice that the equilibrium equation for an incremental load may be expressed as where {u) should be components of corotational stress which is independent of a rigid body rotation. MSCINASTRAN employs an approximate updated Lagrangian approach for geometric nonlinear problems.e.2.e.4) and (2. i.. based on the small deformation theory.2.. (2.5). where [N.8) where {oIT = < o x by 02 Txy Tyz Tzx > and [Dl is a material tangent matrix.] is an appropriate interpolation function for the traction force.2. The element stiffness matrix can be obtained by substituting the constitutive relations into Eq.e. where = < Px iy Pz jxy jyZjZx > and the element matrix [ B ]consists of derivatives of the shape functions. (2.2. i. (2. by which linear strains are computed in the updated element coordinate system in order t o eliminate the effects of the rigid body rotation but the equilibrium is established at the final position in the stationary coordinate system.3) after the substitution of Eqs.2.2. { F I e = {Pie (2. i. Notice that the geometric linear problem requires that the element matrix be evaluated only at the initial geometry. This method does not require reevaluation of the element matrix [ B ](constant in the absence of large strains) while the element coordinates are reevaluated continuously.The strain displacement relations for the element can then be established in terms of nodal displacements using the shape functions in Eq. Equilibrium equations for an element may be obtained by reducing Eq. = [Dl {El (2..2.6) with {F}' = and 1 v [BIT {o} dV {PIe = 1 v [ N ] b~ dV { ) + [NSlT i t ) d s + {PI. evaluated at the current deformed geometry.7).
.where the element stiffness is [KIe = 1 v [BIT[Dl [ B ] dV .11) is satisfied exactly. The convergence. For the incremental process. This convergence condition is required and ensured by element formulations with regard to the element convergence criteria. the equilibrium equation may be rewritten as with where {a0}represents an initial stress or the stress state at the preceding load step. Notice that this expression represents an element stiffness due t o the material stiffness without geometric nonlinear effects. Consequently. the finite element model provides an approximate solution even if the equilibrium equation (2. . The equilibrium must be satisfied in the whole region throughout the complete history of load application. where C over m denotes a summation over all elements. By virtue of the nonconforming elements and the reduced integration. i. the finite element model will have added flexibility and compensate the stiffening effects by the displacement method. Equilibrium equations for the global discrete system are obtained when all the elements are assembled. the differential equations of equilibrium are not satisfied exactly even for linear problems. As will be shown later. Because of the approximations involved in the interpolation functions. but the error decreases as the finite element mesh is refined. an additional stiffness [ K ~ ] to initial due stresses should be included for an incremental process because the initial stresses exist from the second increment. the finite element model is generally known to produce a stiffer structure than in reality.e. may not be monotonic due to nonconforming elements or reduced integration.2. however. With a displacement approach.
where Notice that T is an orthogonal matrix and thus .3 COORDINATE TRANSFORMATIONS The coordinate transformation is one of the most frequent operations in the finite element method [2. hence the system matrices. all the displacements and forces. Coordinate systems involved in MSCINASTRAN are: a Basic coordinate system: a cartesian coordinate system on which local coordinate systems are defined Local coordinate system: defined by the user in the Bulk Data.e. The transformation matrix T consists of direction cosines of unit vectors of the unprimed coordinate system. such as the stiffness matrix. Element and material coordinate systems are defined in the element connectivity description. In MSCINASTRAN.6]. Let us consider a coordinate transformation between the primed and unprimed systems which are righthanded cartesian coordinates. i.2. are expressed in the global coordinates. Vectors and matrices defined in a particular coordinate system can be transformed into another system. Now we only have t o consider linear transformations between cartesian coordinate systems. which may include special coordinates such as cylindrical and spherical coordinate systems Global coordinate system: a collective coordinate system which comprises all the local coordinate systems specified for output quantities Element coordinate system: a cartesian coordinate system unique t o each element Displaced element coordinate system: similar t o element coordinate system but defined in the displaced position Material coordinate system: a cartesian coordinate sytem used t o orient anisotropic material properties Modal coordinate system: a generalized coordinate system defined for each eigenmode. This implies that all the major computations involved in the analysis are processed in cartesian coordinates.. although noncartesian coordinate systems are adopted to orient the local cartesian coordinates for output quantities. a a a a a a It is noted that the global system is a cartesian coordinate system.
e. .7) Kgg = T$ The Icee Tbg ~b: for which the building blocks of Tbe and Tbgare (3x3) matrices formed for each nodal point and have t o be assembled for an entire element. transforms from element t o basic coordinates. the forces and displacements are transformed from element to global coordinates and vice versa. Tbgtransforms from global t o basic coordinates. Then the equilibrium equation.3.Since the work and energy are invariants with respect to the coordinate transformation.. It is noted that Tbeis identical for all the nodes of an element but Tbgmay vary from node t o node in the same element. for a threenoded triangular shell element where the superscript is used t o associate each (3x3) matrix with the nodal point and is repeated for the rotational degrees of freedom. i. e.g. The element stiffness matrix is transformed into global coordinates by (2.e.3. In MSCINASTRAN.. (2.5) {ue> = T g {ug) b TL where Tb. K'ul = P I may be expressed in the unprimed coordinate system by with K = T~K'T. i. it follows that PI = T P . It is noted that the modal matrix is used as a transformation matrix for a modal transformation which is not elaborated here.
: degrees of freedom eliminated by a Bulk Data SUPORT t o suppress rigid body motion. and thus system matrices. uo: degrees of freedom omitted by static condensation. The multipoint constraint equations are formed in the module GP4 as follows: where [Rmg] = [Rmm 1 Rmn] . .e. u. and reduce it t o the nset.. YES) is used.uo . For convenience. AUTOSPC. The module MCEl partitions [Rmg]and solves for a transformation matrix [Gmn]. Mutually exclusive subsets of the global displacement set.2.}. ur: degrees of freedom which remain for solution after reductions. Since the setreduction operations involve many basic modules and DMAP blocks.4 DISPLACEMENT SETS AND REDUCTION OF SYSTEM EQUATIONS The equilibrium equations. are reduced in size using the displacement set. are defined as follows: urn: degrees of freedom eliminated by multipoint constraints. complementary sets are defined as follows: Un Uf Ua = ug . us: degrees of freedom eliminated by singlepoint constraints. but they are not selectable in the subcases as for MPCs. which is a unique feature in MSCINASTRAN.urn = = Un Uf Ua Ul = . where [GmnI = [Rrnrn]' [Rmn] Then the module MCE2 partitions the global stiffness matrix.i. [I&. the mathematics for elimination of constraints and static condensation is reviewed here. {u.'Us .].UT The subsets are defined by the user with a possible exception in the sset if (PARAM. Notice that the rigid elements are equivalent to the multipoint constraints internally in the program.
+ K.. and Q.] Qn = ~ 2 Q. and QL to distinguish from K. i. . = Ys. which are The primes are used in K. and T [KmmG. known as static condensation. The fset is partitioned by a module UPARTN as follows: from which {uO) = {u:) + [Goal { ~ a ) where uz = K : and Po (by SSG3) Go.. {. (by DECOMP and FBS).m + Qk P..4. (by SSG2) Notice that the effects of constraint forces (Q.from which the system is reduced to where I .6). resulting matrices after the reduction. The Further reduction of equations in the fset is performed by an elimination of the oset... Equations in the nset can be further reduced by eliminating singlepoint constraints. ... (2..] + [IT:.. Y.Icj. G . + K.) ~ingle~point constraint forces are recovered by are not visible in Eq. = G. PA..e. which is reduced to ri.. and Q.Uf = Pj where Fj = with Us P. = K: KO..
AUTOSPCR is available in SOL 66 (or 106) and applicable t o the oset in the residual structure. r In SOL 99 (or 129). They are: The rset is not supported in SOLS 66 (or 106) and 99 (or 129). MPCs and SPCs can not be changed in the subcase level. r PARAM. Therefore. the oset may only be used in the linear part of the structure. r Rigid elements are formulated with linear multipoint constraint equations and do not have large displacement capability.Then the reduced system of equations in the aset is obtained as where and Pa = P i + GF~ . r In SOL 66 (or 106). but they will be changed on an incremental basis. Po There are some rules t o remember regarding the displacement sets in nonlinear analysis. Consequently. erroneous results will be obtained if the rigid element undergoes a large rotation. stiff elements should be used in place of rigid elements for large displacement analysis. SPCs and MPCs can be changed from subcase to subcase. . AUTOSPC is applicable only t o the superelements. r PARAM. To avoid this. all the constraints should be specified above subcases. Do not use a Bulk Data SUPORT. However.
While an excessively small increment reduces the computing efficiency without any significant improvement in accuracy. criteria and parameter values with numerous test problems resulted in a succinct implementation. User specifications for solution methods in nonlinear analyses are allowed via: Bulk Data NLPARM for the static analysis in SOL 66 or 106 (to be selected by a Case Control Data NLPARM) Bulk Data TSTEPNL for the transient response analysis in SOL 99 or 129 (to be selected by a Case Control Data TSTEPNL) Bulk Data EIGB for the nonlinear buckling analysis in SOL 66 or 106 (to be selected by a Case Control Data METHOD) The increment size can be varied from subcase t o subcase by specifying different NLPARM or TSTEPNL. algorithms. Based on the extensive numerical experiments. Needless t o say. . a large increment may deteriorate the efficiency as well as the accuracy. the size of the load increment (or time increment for creep analysis) should be so chosen t o yield a uniform rate of change in strains or stresses for the material nonlinear problems and a uniform rate of change in displacements for geometric nonlinear problems. no incremental load steps are required when the response is linear. Determination of the proper incremental time step will be treated in chapter 8. an attempt was made to establish a general strategy suitable for most problems without requiring insight or experience. The best engineering judgement should be exercised t o determine the increment size based on the severity of the nonlinearity. The increment size for loads or time steps has the most significant effect on the efficiency and the accuracy of the computation. so that changes can be accommodated in the subcase level as needed. The main issue is how t o choose the most efficient method from the options available for the incremental and iterative processes in the solution of nonlinear equilibrium equations. while maintaining flexibility by allowing the user to specify optional parameters.5 NONLINEAR SOLUTION PROCEDURE The generalpurpose program developer faces the task of providing the best workable solution method for a wide spectrum of problems. It is recommended to define separate NLPARM or TSTEPNL for every subcase even if the same values are specified. It is impossible to optimize the incremental step size in the absence of prior knowledge of the structural response. Some adaptive methods are available. The major feature of the nonlinear analysis is the requirement for the incremental and iterative processes to obtain a solution. it may even cause divergence. Variations in combining theories. particularly in the pathdependent problems. because it is dictated by dynamic characteristics as well as the loading history. such as an automatic time step adjustment and bisection of loads upon divergence. In principle. The incremental and iterative processes are complementary t o each other because the larger the increment size the more iterations the solution requires.2.
2.6) with {PIe = and Jy [BIT dV [NSlT{ t ) d s t {PIe = /v [ N ]{ ~ ) d V t b {PI..4) and (2. based on the small deformation theory. The nodal forces of an element can then be expressed as .The strain displacement relations for the element can then be established in terms of nodal displacements using the shape functions in Eq.e. where {elT = < Px iy e.2. Equilibrium equations for an element may be obtained by reducing Eq.2. jXyjyz jzx > and the element matrix [B] consists of derivatives of the shape functions. This method does not require reevaluation of the element matrix [B] (constant in the absence of large strains) while the element coordinates are reevaluated continuously.4).7). i. (2. (2. where [N.2. Then the element boundary stresses are statically equivalent to the nodal forces which balance the applied external loads..5). evaluated at the current deformed geometry. i. The element stiffness matrix can be obtained by substituting the constitutive relations into Eq. Notice that the geometric linear problem requires that the element matrix be evaluated only at the initial geometry. (2. (2. i.3) after the substitution of Eqs..8) Tyz fly Try Trr > and [Dl is a material tangent matrix.e. {F)e = {PIe (2.2. MSCJNASTRAN employs an approximate updated Lagrangian approach for geometric nonlinear problems.e. by which linear strains are computed in the updated element coordinate system in order to eliminate the effects of the rigid body rotation but the equilibrium is established at the final position in the stationary coordinate system.2.2. Notice that the equilibrium equation for an incremental load may be expressed as {F) = /v [BIT {ti) dV = {P) where {ti) should be components of corotational stress which is independent of a rigid body rotation. (4 where {ulT = <Ox = [Dl (2) 02 (2.] is an appropriate interpolation function for the traction force.
However. Effectiveness may be interpreted as reliability or userfriendliness. The adaptive algorithm alleviates this difficulty. The adaptive algorithm is related t o cybernetics or artificial intelligence in a broad sense. The control criterion is most critical and tuning the control parameters is the major design task. solution accuracy and effectiveness. and various global solution strategies. The computational procedure involves incremental and iterative processes ranging from local subincrements to global solution processes. Favorable performance of the automatic control system is obtained when control parameters are tuned t o process dynamics. The effectiveness of the program is significantly improved by the adaptive algorithms. The size of the load or time increment has the most profound effect on the efficiency as well as accuracy.1 ADAPTIVE SOLUTION STRATEGIES Nonlinear finite element computations comprise material processes. the engineering time can be reduced considerably. element force computations. All of these attributes of the nonlinear program can be improved by adaptive algorithms.Chapter 3 SOLUTION METHODS 3. the number of stiffness matrix updates and the number of iterations should be minimized. The adaptive algorithms can help t o reduce these processes or t o choose a more desirable iteration path. which may be conflicting requirements. There is a broad range of processes for which adaptive algorithms may be adopted in the computational procedure of nonlinear finite element analysis. Performance of the finite element program can be scrutinized from three different perspectives: computational efficiency. From the efficiency point of view. The most CPU consuming processes in nonlinear analysis are the stiffness matrix update operation and element force calculation. it is difficult t o determine optimal load or time increment size. If the program provides a plausible solution or direction despite the poor estimates or selection of control parameters. . the adaptive system is defined t o be an automatic control system which has the capability of automatic measurement of process dynamics and of automatic readjustment (or redesign) of the control algorithm. In engineering applications.
. which included the load step bisection strategy for static analysis. The groundwork was carried out during Version 65 development.Adaptive switching algorithm for limiting cases. adaptive GAPIfriction element.Twopoint Newmark integration. a Promote effectiveness in terms of engineering time. Adaptive features implemented in MSCINASTRAN nonlinear capabilities as of Version 67 are as follows: a Newton's iteration for static and implicit dynamic analysis . . trialanderror).Adaptive bisection and recovery of load increment. and the adaptive bisection strategies based on the large rotation and material processes. Further development of the adaptive capabilities continued for Version 67. . .Selective BFGS updates.The adaptive algorithm is analogous t o the automatic control system with the process to be controlled being numerical rather than physical. and plasticity induced bisection strategy.Selective line search processes.Adaptive stiffness matrix update strategies. a Direct time integration for transient response analysis . .Convergence test: error functions and convergence criteria.Riks and modified Riks methods. which included the automatic time stepping. . . .Adaptive correction in case of path reversal. a Maintain accuracy and efficiency without user intervention throughout the analysis. a Arclength methods for static postbuckling and snapthrough problems .Reiteration process.Selective BFGS Updates.Crisfield's arclength method. Major efforts have been made to implement adaptive solution algorithms for nonlinear analysis in MSCINASTRAN.Adaptive arclength adjustment. As a result. The adaptive capabilities for the transient response analysis were implemented in Version 66. . The design objectives are: a Relieve users from uncertainties (guessing. . automatic arclength adjustment. . creep time step bisection. a cascade of computational feedback control loops have been implemented in the incremental and iterative processes in conjunction with the adaptive subincremental processes in the local level processes. adaptive update and bisection. .Divergence criteria.
Adaptive algorithms in nonlinear analysis have shone some light toward this goal.Rotation vector approach. . .Adaptive bisection strategy. is not the computational efficiency but the overall engineering efficiency.Adaptive bisections in plasticity. Processes for Material Nonlinearity .Gimbal angle approach with auxiliary angle. the goal of improving userfriendliness.Adaptive subincremental scheme in plasticity and creep routines. . .Adaptive bisection on creep time increment.  Error control and stiffness updates for unloading. The main goal of the selfadaptive method is t o relieve users from the chores and uncertainties of guessing or trialanderror searches.Adaptive penalty value adjustment.Adaptive bisection based on rotations. an effective algorithm design is automatically attained for efficiency as well as for solution accuracy. . Automatic time step adjustment.Adaptive stiffness matrix update strategy.Adaptive subincremental process. By achieving this goal. . Although the optimal combination of all parameters is yet to be found. It is desirable to have a fully automated program which efficiently renders accurate solutions t o nonlinear problems. As a result.Internal relaxation iteration for select elements. however. the success of the adaptive algorithms in various levels presents the future direction of nonlinear capabilities development. Processes in Gap/Friction Element for contact problems .  Adaptive friction t o simulate stickslip forces. solution efficiency and the accuracy has been accomplished. . . The primary virtue of the selfadaptive method. make it possible t o maintain an acceptable accuracy while maintaining the solution efficiency without user intervention throughout the dynamic environment of nonlinear computation. .Adaptive bisection and recovery. This is because the adaptability makes it possible to produce a solution at the very first trial without requiring advanced knowledge of structural behavior. The adaptive algorithms in the lower level processes (element or material) in concert with adaptive solution methods in global incremental/iterative processes. Large rotations for Geometric Nonlinearity . .
Since the equilibrium condition is not immediately attained in the presence of nonlinear elements.10) with a material tangential matrix for [Dl. without regard t o the coordinate transformation. {Q. Element nodal forces are nonlinear functions of displacements for nonlinear elements. which is signified by the convergence criteria. is defined as follows: . the unbalanced forces acting at nodal points at any iteration step are conveniently defined as an error vector by (3.{Fa} Notice that the aset is equivalent t o the 1set in the nonlinear analysis because the rset does not exist.2. constraint forces.). The equation t o solve at the ith iteration is where { n u i > = {ui) . The subscript a will be dropped for simplicity in the following discussion. i. 3. a linearized system of equations is solved for incremental displacements by Gaussian elimination in succession.1. The Jacobian of the error vector emerges as the tangential stiffness matrix.{ui1) and { R ~ )= {P) .3. respectively.).21. The tangential stiffness consists of the geometric stiffness in addition t o the material stiffness. and {F.2) {Ra} = {Pa} . The material stiffness is given in Eq.{ ~ ( u " } . respectively. where [ K m ]and [ K d ] refer t o the material and the differential stiffnesses.) represent vectors of applied loads. Since the error vanishes at constrained points and the constraint forces vanish at free points.. an iterative scheme such as the NewtonRaphson method is required. The differential stiffness. Based on Newton's method [3. which is caused by the initial stress. and element nodal forces.2 NEWTON'S METHOD OF ITERATION The equilibrium equations in the gset may be written as where {P. (2.e.2. The iteration continues until the residual error {R) and the incremental displacements {Au) become negligible.
however. one may resort to the modified Newton's method which requires the tangential stiffness to be evaluated just once at the initial position.2. (3. {RO} = P . Notice that the initial displacement stiffness is not included in [I<T] because its effects are already eliminated in the element formulation. determination of the tangential stiffness and its inverse at each iteration entails a considerable amount of computation..where [ B N ] represents the second order effects in the straindisplacement relations. Then the residual load error is automatically carried over to the next incremental process.[I<] {u}. From a practical standpoint.1) suggests. [GI consists of derivatives of shape functions and [MI is a function of stresses.. and used thereafter t o solve for {Aui}. i. The merit of the NewtonRaphson method is the quadratic rate of convergence. Then the successive error vectors can be evaluated by It is noted that the corrective force vector vanishes for linear elements. As Fig. . + where {AP) is an incremental load vector and {Re} is a residual load error carried over from the last converged solution. more iterations are required for a given accuracy by the modified Newton's method. where u* is a true value of {u}. The iteration starts with initial values {uO) = last converged displacement.~ ( u ' ) = {AP} {Re}. The Gaussian elimination method is better suited for this approach than the iterative descent method because the decomposition is performed only once. Recalling that the element forces for linear elements are expressed as a corrective force vector may be defined as {C) = {F} . However.e. Newton's procedure is implemented using a corrective force in MSCINASTRAN until Version 65.e. and 11 11 represents a vector norm. The corrective force calculation was removed in Version 66 and the error vector is computed directly from the internal forces. i. q is a constant. {uO).
(a) NewtonRaphston Method LOAD ITERATL\ COUNT k DEFLECTIGN (b) Modified Newton's Method Figure 3.1 Newton's Methods for Iteration .2.
most of which have been determined empirically: a The ITER method is specified for KMETHOD and the iteration count (i) is an integer multiple of KSTEP. an adaptive matrix update method is indispensible.1 Update Principles Newton's method could be trapped in an infinite loop. To this end. the nonpositive definite stiffness matrix will be retained and used for iteration if the arclength method is used for the static solution. the stiffness update process is primarily controlled by the input data in KMETHOD (AUTO. This difficulty is overcome during the Newton's iteration by when discarding the differential stiffness. stiffness update strategies are reviewed with respect t o the static analysis. KSTEP where ik is the number of iterations t o convergence after the stiffness update. a The first iteration is finished with the SEMI method.3. The value of NEWK will be set t o 1 under the following conditions. the modified Newton's method could lead t o divergence when the stiffness changes drastically. 3.3. 1. however. 3. oscillating about the local maximum as illustrated in Fig. From the user's point of view. as demonstrated in Fig. However.3.1. However. a The solution has converged with the AUTO or SEMI method and ik 2. SEMI or ITER) and KSTEP (integer) fields of the NLPARM Bulk Data entry. [lid]. 3. the stiffness matrix update has probably the most profound effect on the success of the nonlinear solution.2(a). the tangential stiffness is not positive definite as shown in Fig. unless the tangential stiffness is reevaluated at the critical point. . In spite of its significance. The new stiffness is formed whenever the iteration module NLITER returns a positive value for NEWK. Stiffness update strategies are established t o update the stiffness matrix on an as needed basis [3. Variations of the modified Newton's method are adopted in MSCJNASTRAN.2(b). In this section.or 2. A variation of the modified Newton's methods is t o update the stiffness matrix at every few iterations. 3.3.3 STIFFNESS UPDATE STRATEGIES Among other features of the solution algorithm.TESTNEG. The stiffness matrix is updated by a DMAP loop in SOL 66 (or 106) controlled by a nonuser parameter NEWK which has a value of 1.3] such as probable divergence. it is very difficult to implement a robust algorithm for update strategy due to the lack of a priori information regarding the right timing for an update. This problem is controlled by PARAM.3.
(printed under the heading ENIC at every iteration) is estimated by N. x (N. and Ni is the expected number of iterations for convergence without a K update. The KMETHOD is AUTO or SEMI and if ik > 2 and tk < t.1) where tk and ti are the CPU time required for a K update and an iteration. EPSP the specified : tolerance for E.99) + where the convergence rate Xi (printed under the heading LAMBDA at every iteration) is computed by with an initial value XI = 0.3. 3. for the ith iteration.3.lX. The KMETHOD is AUTO or SEMI and if ik > 2 and (N. 5 ) where NQNV is the number of quasiNewton vectors accumulated and MAXQN (integer parameter t o be specified in NLPARM) represents the maximum number of quasiNewton vectors allowed. The value of N. The absolute divergence is treated as two occurrences of the probable divergence. There are two classifications in the divergence condition: probable and absolute.2 Divergence Criteria The value of NEWK will be set to 2 if the solution tends t o diverge. Then.i) where i is an iteration count and MAXITER (integer parameter in NLPARM) is the maximum number of iterations allowed for an increment. Divergence criteria are defined based on the ratio of energy errors evaluated before and after the iteration. 0.. and Xf = min (A. The MAXDIV field in the NLPARM entry requires an integer to specify a limit on the probable divergence conditions allowed for each iteration t o continue. 0. .7 O... i. respectively. .7) > (MAXITER .The solution has converged with the AUTO or SEMI method and NQNV > MAX(MAXQN .6 .7) (3.) log Xf with E being the relative error in terms of loads at the ith iteration.e. the divergence criteria are established as follows: . = log (EPSPIE.1.
It is noted that NDIV is incremented by 2 if the absolute divergence is encountered. To reserve a minimum of 5% of the userspecified CPU time (t. .e. KMATUP > (I + j ) 4 and KMETHOD=AUTO where KMATUP is a DMAP parameter representing the number of stiffness updates within an incremental step. When the solution is diverging.elapsed CPU time) . or NEWK=2. Another absolute divergence condition is established based on the heuristic observations t o prevent thrashing with excessive K updates.Absolute divergence if E i> 1 + or E i< 1012 Probable divergence if 1012 < E i< 1 or excessive line search. If the divergence is detected a t the first iteration ( i k = l ) after a new [K] is formed. the remaining time is defined as t. Recalling those criteria.). Sufficiency of the remaining time (t.3.2tk when the solution has not converged and the stiffness matrix is scheduled t o be updated.3 Time Expiration Criteria If the remaining CPU time is not sufficient to continue the computation.. If the absolute divergence condition (NEWK=2) is encountered in two successive iterations in spite of the new stiffness. i.4. the current solution base is not retracted t o prevent a repetition of the same diverging iteration.) for data recovery. where the excessive line search condition is defined in Section 3.05tS. when the solution needs more iterations without a new stiffness. t. The purpose is to allow adequate time for the data recovery with solutions previously obtained. . Then the NOGO flag will be set under the following conditions: tT < 1. and if the value of MAXDIV is positive.0. < lot. the computation will be terminated with a fatal message (UFM 4551) t o that effect.) is determined based on the K update time (tk) and the iteration time (t. the current solution base (displacement) is retracted so that the stiffness matrix is evaluated in reference t o the preceding iteration step. however. depending on the succeeding process anticipated. = (t. the solution is considered diverging if NDIV 2 MAXDIV where NDIV is the number of probable divergences encountered during the iteration. the NLITER module will set the NOGO flag (DMAP parameter NSKIP=2) and exit with a UFM 4311 to that effect. 3.
This condition includes pseudo convergence. t. the divergence a n d r e t r e a t proced u r e .5. where the number of iterations has reached MAXITER (>O). If the number of iterations reaches MAXITER while MAXITER is negative. respectively.3.3. < 1. 3. < 0. the NOGO flag will be set regardless of the remaining time. and the time expiration logic are shown for clarity in Figures 3. .3.8tk when the solution has converged and the stiffness is to be updated.t.4.8tk when the solution has converged and the stiffness update is not required.3. Flow diagrams for the stiffness update strategies. and 3.
 /' I' 1 I I I U1 I ..(Path taken with F I stiffness matrix update) A y P Y (Path taken without stiffness matrix update) . u u" u* Figure 3..3.1 Hybrid Newton's Method based on Stiffness Matrix Update Strategy ..
Omitted for these two iterations b DEFLECTION (b) Iteration sequence drops the differential stiffness with PARAM.TESTNEG=2 when the negative diagonal terms appear in the stiffness matrix. LOAD ? I I ~ c .3.2 Iteration Trap Condition . Figure 3. Iteration will be trapped if continued with TESTNEG=+l or 0.TESTNEG=1 if the stiffness becomes negative.b DEFLECTION (a) Iteration stops at 3 with PARAM.
3. I 4 2 I Figure 3.Update t i m e t t I t e r a t i o n tim ti NEYK * 1 1 0 MXITER > 0 NSKIP 9 In.3 Flow Diagram for K Update Strategies (to set NEWK) .
4 no R 9 1 .4 Divergence and Retreat Procedure .e Hasage for Diwrgrtce u 6 w k lip a step i1 i1 ...3. Figure 3.
3.3 .NOGO Figure 3.' I Remaining t i m e t r K update time tk I t e r a t i o n t i m e ti I EXIT1: EXIT2: EX I T 3 : No k update K update .9 .5 CPU Time Expiration Logic for NOGO Condition 3.
.
where {d) denotes a direction vector. the implementation scheme was not well elaborated in the finite element analysis [3.4 LINE SEARCH METHOD The line search method is well established as a basic descent method in nonlinear programming [3.4. 3.2) results in an iteration scheme for Newton's method with {ui+') = {ui) I~''{R~} + where the unbalanced force {Ri) represents the negative of the gradient vector at {ui). [K] at {ui). i. Then. The function is convex if and only if the Hessian matrix is positive semidefinite. The stationarity condition of Eq..4. The function II must be convex in the vicinity of a solution t o have a minimum point. some insights were gained for making the option more viable and effective.91. (3. .53.4]. While the line search method was implemented and used in MSCJNASTRAN. corresponding to {u .ui). The basic idea behind Newton's method for the nonlinear iteration is that the objective function (II) is approximated by a quadratic function in the vicinity of the trial solution point.e. a feasible direction {d) must satisfy the necessary and sufficient condition: where the equality should be satisfied at the solution.1 Theoretical Basis The functional t o minimize in the finite element analysis is the total potential of a continuum. and the tangential stiffness matrix. This is the theoretical basis of the line search method. which is reduced t o a quadratic function of displacements {u) in linear problems as follows: where U is the strain energy of the system and W is the potential energy of the external loads.10]. represents the Hessian matrix of the total potential.3. The line search method was implemented and modified t o optimize its usefulness for a general class of problems in conjunction with quasiNewton updates using MSCJNASTRAN [3. Although the method has been used for many years to improve the rate of convergence in the nonlinear iterations.
which is known as the method of false position or the secant method [3.9) with respect to a disappears. where nonquadratic terms may not be negligible. By a line search procedure satisfying Eq. the solution set {u) is moved t o a relative minimum point in the feasible direction { d ) .l ~ i . (3.2) is not appropriate. When the derivative of Eq. The total potential may be expressed in terms of the search parameter by substituting Eq.7) vanish. (3. which results in This equation represents a linear interpolation in terms of { u ) and {R).4. i.4. (3.4. (3. the solution set is placed at a superior position for the next iteration.e.4. the objective function to be minimized can be regarded as a function of a single variable. which is the basis of the line search method. (3.l where a positive search parameter ( a ) is determined such that The search parameter is introduced to expedite the convergence when Newton's method is used. The local minimum point can be obtained by making the derivative of Eq. Along the direction of descent {d). because the approximation employed in Eq. the search parameter. Considering the ith iteration. the search parameter for a local minimum is obtained by . The supplementary approximation for a quadratic function can be made by utilizing the gradients known at two successive iterations to estimate the Hessian matrix [K]. the new solution set is determined by with di = K .7).6).2].4. Thus. The line search should be particularly effective when the modified Newton's method or the adaptive update method is used. (3.4. It could be effectively used at a distance from the solution.5) into Eq.The process of determining the local minimum point in a given direction is called the line search..
LSTOL 5 Ek 5 LSTOL where LSTOL is a tolerance specified in the NLPARM entry. comparable to that of a modified Newton's iteration. the solution is diverging and the stiffness matrix should be updated without the line search. If Ek is initially less than unity but outside the tolerance. Furthermore..e.4. a criterion is established to end the line search procedure if . to evaluate element forces. An excessively large value of a (compared to unity) does not validate the underlying assumption of close proximity t o the solution point and may cause a numerical instability. (3.4. it is often desirable to sacrifice accuracy in the line search and to conserve the overall computing time by terminating the search procedure before it has actually converged t o the local minimum. = { d i ) T { ~ i . If it is too small (i. each line search costs a considerable amount of computing time.e. the search procedure is initiated by utilizing one of the following two schemes: .2 Search Criteria It is impractical in general to obtain the exact minimum point for the line search procedure because it does not render a final solution.e.13) where ATOL is a tolerance.11). with E. respectively.4. a t a = 1) greater than unity. the search parameter should have a value close to unity.4. close t o zero)...12) In the neighborhood of the solution point. If Ek is initially (i. anew 1 = (sold t 2 anew) if anew < ATOL (3.e... Another criterion for limiting line searches is based on the divergence rate defined by a ratio of energy errors before and after the iteration. i. 3.3 Implementation of Search Procedure The search method was varied depending on the value of the relative error Ek in Eq.3. One method of restricting the line search is t o limit the maximum number of consecutive searches at each iteration cycle. the iteration is not effective because the solution vector retracts t o the old value. The line search will result in a dilemma if the selected search parameter is too small or too large. i. (3.4. Then. the bisection process can be adopted when a approaches zero. To circumvent such a situation. Therefore.l ) where i and k are indices for the iteration and the line search (within the ith iteration).
the error Ek is allowed t o increase in magnitude if it is negative because the root (Ek= 0) is guaranteed t o exist in this case.Eold The Illinois algorithm was introduced for the subsequent searches t o achieve faster convergence in the succeeding search process.1. As shown in the example.4. 3.4. (3.4. The domains for different operations are shown schematically in Fig. respectively.14). there should be an a. Eq. i.4. For this reason. The relative error Ek is normalized such that Ek is unity at a = 0. a doubling scheme was employed as a conservative approach by simply doubling the old value of a. the line search is blocked by assigning zero t o E l .10). where Ek is expressed as a function of a. Eq. as exemplified in Fig.14) should be rewritten as: ak+l = a n e w . With k = 1. This value is predicted by a linear interpolation. An example of the Illinois algorithm is illustrated in Fig. Thus. If Ek becomes negative (< LSTOL).E n e w anew Enew . between 0 and 1.. 3. These two points were so chosen t o maintain two most recent points with the errors in opposite sign and the extrapolation was avoided.14) are not necessarily the most recent data points acquired from the preceding line searches.11) approaches zero.4. If E l is negative. where k is a line search index. 3. Note that the line search would not be meaningful if the denominator (Eo) in Eq.5.(3. the next line search uses onehalf the value of the error in opposite sign for interpolation. This scheme is employed because the error function ( E k ) may not be a monotonic function of a . This extrapolation may result in a very large value of a. the solution should have almost converged and the line search would not be required. corresponding to a = 1.4. the doubling scheme is switched to the interpolation method in the subsequent searches as shown in Fig. (3. If it is close t o zero. (3.1) involved in Eq.e.4.e. if El is close to 1. (3.14) is equivalent t o Eq. However. two points (corresponding t o indices k and k .4. in fact.1. the zerocrossing could be predicted by a linear extrapolation based on the same mathematical reasoning as interpolation. If two successive line searches result in the error of the same sign.4. . A linear extrapolation can also be conceived by Eq.4.. before using Eq.2.14a). 3.Linear interpolation with Illinois algorithm if Ek < LSTOL Doubling scheme if LSTOL < Ek < 1 where LSTOL has a default value of 0.4. The initial and the final values of Ek are printed for every iteration under the headings EFIRST and EFINAL.sold .1. (3. For a positive value of E l (for a = I ) . i. Under such circumstances. The search procedure is terminated if the doubling scheme results in an increasing relative error Ek. that makes the error Ek disappear. (3.
i. the line search process is terminated before accurate convergence is achieved. Starting from this point. and R. ending at point 1. The bisection process is initiated based on the criterion: a < ATOL in which 1 10 t MAXLS During the line search iteration. 3. R0=pFO=85=3 which results in and . MAXLS.. but expedited by virtue of the Illinois algorithm.2 and 3. Upper and lower bounds of a are effectively set by combined effects of MAXLS and the doubling scheme or the bisection. and the convergence rate as such is rather immaterial. the external load (P) is increased from 5 to 8.4.4 OneDimensional Example It is instructive to examine the line search procedure in a single degree of freedom system. However. ATOL = The method of false position is not globally convergent.3. or rather the process converges only if it is started in close proximity to the solution point. is defaulted to 5 for static analysis and 2 for dynamic analysis. The first iteration. the golden mean [3.4].4.4.4.4. The order of its convergence approaches 1. represents a Newton's iteration with the initial stiffness (KO= 6).The line search algorithm and its iteration scheme are shown succinctly by flow diagrams in Figs. Consider a fictitious structure that responds to the displacement with the internal force described by a parabolic function: F(u) = U 2 6u. 3..The number of consecutive line searches allowed for each iteration is limited by a userspecified parameter. This parameter. + Suppose that the structure is in equilibrium at u = 1 with an internal force (F) of 5.e.amin. the data relevant to the computed minimum error (Emin.618. MAXLS.. The convergence rate of the current line search process is somewhat reduced by the doubling scheme. 3. as shown by a dashed line in Fig.) are resumed in case the line search is terminated due to reaching MAXLS or increasing error ( E k ) .
g. KO= 2: The first line search will result in: .F 2 = 8 . the line search will normally be skipped.. the doubling scheme will be employed since El is positive. however. The first iteration is a modified Newton's iteration with a stiffness evaluated at certain point previously.4.3 This point corresponds t o point 2.g. (3. The doubling scheme will result in: which is an exact solution and the iteration is finished. For the sake of comparison... LSTOL= 0.7 = 0.714 * dl = 1.e. the relative error in Eq. = uO+ 1.86 R2 = P .5). The line search seems to be more effective in the stiffening structure. If the line search is enforced (e.7. e. which is illustrated with another parabolic function representing an internal force: Suppose that the structure is in equilibrium at u = 1 with F = 2 and an external force (P) of 6 is applied as shown by a solid line in Fig.14): which results in: and u.4.4. (3.4. i. which can be obtained by connecting points 0 and 1 on the dashed line. based on Eq.4).11) is estimated: Since this error is smaller than the line search error tolerance by default (LSTOL= 0.To perform the line search at point 1. let us employ an extrapolation scheme for the line search. 3.
while the residual error vector (R). to be minimized in the iterative process. not to mention the numerical difficulty .) and sset (AY.5. This is because dg includes AY. (3. ug = ug adg. without requiring additional iterations. Let us assume that LSTOL= 0.5.4. but is set t o zero at the second iteration and on.. which is not sufficient to render a solution to a multivariable system unless the direction vector happened to be oriented in the right direction. conversions from the Iset to the gset and vice versa should be performed at every iteration.11). comprises only the 1set. the solution converges to u = 2 with a = 0. The line search has a single parameter. The line search uses the direction vector in the gset to avoid additional set conversions in the search process. In this context. there is another conversion of displacement increments to treat the cumulative gimbal angle for large rotations. It is noted that the line search cannot be performed at the first iteration if the enforced displacements (SPCD or SPC) are specified. In fact. For this reason.The second line search will result in: The second line search would not have been performed if LSTOL= 0. The third line search would result in: Now that E3 is positive again. the Illinois algorithm is invoked: As the line search continues. Such a convergence by using the line search alone is a peculiar behavior in the single degree of freedom problem.) to the Iset displacement increment from the Newton's iteration.02 to continue discussion. at the first iteration. + where dg can be obtained by merging displacement increments in the mset (Au.e. Considering Eq. if the denominator is close .4.5 Other Provisions for Line Search Notice that the displacement vector in the gset is required to compute internal element forces (F) at every iteration. 3. the line search would not be meaningful if the denominator (Eo) approaches zero. i.
therefore. (Ek at the end of line search) NLS : LSCT (line search count k at the end) . the results will be printed with other iteration information under the following heading: FACTOR : cr (final value for the search parameter) EFIRST : El (relative error at a=l. excessive line searches are treated like a probable divergence condition by incrementing NDIV by 1. the quasiNewton update is not used if E1=O. the solution should have almost converged and the line search should not be required. There are provisions to block the line search under such circumstances as follows: Reset E1=O if ABS(Eo) < EPSW x E* for the first iteration. Reset E1=O if ABS(Eo) < EPSWxE* for subsequent iterations. The excessive line search is defined to satisfy TLSCT > (2 x MAXLS . To alleviate exhausting CPU resources in such cases.to zero. As will become clear in Section 3. Reset E1=LSTOL if ABS(Ek) > LSTOL and ABS(Eo) < 0. The first condition is employed for the creep analysis because the line search or the quasiNewton update is not appropriate for the first iteration of the creep which always causes a small value of Eo. where EPSW is a convergence tolerance in terms of work from NLPARM and E* is introduced for dimensionless criteria (defined in Section 3.1) where TLSCT is the cumulative number of line searches after the new stiffness matrix is formed within an increment. The third condition is. introduced to allow quasiNewton updates while suppressing the line search. It was discovered that the solution algorithm tends to thrash line searches when the structure becomes physically unstable.01 EPSWXE*. Upon termination of the line search procedure. such as at the collapse load.5. divergence rate) EFINAL : E.6).
4.1 Line Search Parameter vs. Line Search Error .A Before Line Search (k=l) EK (El= Divergence Rate) 0 Second Line Search (k=3) C] First Line Search (k=2) NOTES: 0 obtained by a linear interpolation obtained by a linear interpolation with Illinois algorithm Figure 3.
4.2 Flow Diagram for the Line Search Algorithm .Figure 3.
I + .~i R'  P .. E. ROOT NQ a.3 Flow Diagram for the Line Search Iteration . Emin 10'2.4. di Save R ~ * if LSC~ o '  Update R . n 1    0 I t .F Iu9 u i .T START EmL  1 d 2 'ABS (u'll Initialire LSCl 0 a 1. / < m~ (Ri Resume amin Emin Rmin  NDlV  NDlV + 1 Compute enon NDlV  Print iteration information NDlV + 3 Convergence test Figure 3. if flag is set E ' E' Oil (di A Exln Line Search Algorithm . ~a IE.
4 Examples for the Line Search Procedure .Load (F) Displacement (u) Figure 3.4.
25.4. The quadratic interpolation is implemented as an optional feature t o be used instead of the linear interpolation. SYSTEM(141)=1 must be included in the NASTRAN definition at the beginning of the input data. Considering the previous search criteria.5.4. El = E ( l ) = dT R1 The derivative of E can be expressed as where KO is a stiffness matrix formed at some preceeding step or iteration.3. Real roots can exist for _< 0. it is necessary t o introduce the quadratic interpolation only for < 0. In reference to Fig. Let the error (generalized residual force vector) be defined (in terms of a ) by E ( a ) = dT R ( a ) Known values of this function are for a >0. What follows is the theoretical basis for the quadratic interpolation. Since the doubling scheme is safe and conservative. the doubling scheme is used for 0.e. the coefficients for the quadratic curve fitting can be obtained as where cz = El from E ( l ) = El.5 < < 1 and the linear interpolation with Illinois algorithm is used for < 0. i.16) Eo = E(0) = dT RO and for a = 0 for a = 1 . NASTRAN SYSTEM(141)=1$ which will revive the linear interpolation on the first iteration as in Version 66.5.6 Quadratic Interpolation A quadratic interpolation scheme is added to the line search method in Version 67. In this domain the determinant should be 3 3 . In order to suppress the quadratic interpolation. (3. A quadratic line search on the first iteration is the default option. Solving for E ( a ) = 0..5. 3.4.
9 Figure 3. Notice that the geometric stiffening case belongs to this domain..19) should be discarded.4.4. i.Since a > 0. which is valid for 5 0.5 Quadratic Interpolation Algorithm . the positive sign in Eq.e. The quadratic interpolation is deemed effective for the geometric stiffening case. (3.
3. the second iteration would lead t o the next feasible direction: d2 = K' {R(ul)) . the BFGS method renders a stiffness matrix resembling tangential stiffness in the limit.5. the performance of the BFGS update with respect to effectiveness and efficiency depends largely on the implementation.10]. They have been developed for efficient nonlinear iterations by approximating the inverse Hessian matrix in place of the true inverse that is required in Newton's method. QuasiNewton methods have been widely and successfully used in nonlinear optimization applications. the information acquired during the iteration is used to modify the inverse stiffness matrix. By the BFGS update. expressed as . 3. a feasible direction of descent is given by where {R) represents an error vector to be minimized and KO is a Hessian matrix.8]. for the general class of problems. Upon the first iteration. which may be problem dependent. QuasiNewton update methods are considered the most sophisticated methods of the Newton's iterative solution schema and represent the culmination of extensive algorithm development for Newton's iterative process [3. One of the quasiNewton techniques. This approximate update to the inverse stiffness matrix results in a secant modulus in the search direction.4]. with u1 = u0 + d1 Consider a Taylor series expansion of the load error {R) about {ui). our experience showed that the implementation method and adjustment of parameters had a significant impact on the program's performance [3. When combined with the line search [3. It is extremely difficult to tune the tolerances and parameters to the optimal condition. The current approximation is used at the next iteration to define the next feasible direction in the modified Newton's method.11]. In the absence of line searches or quasiNewton updates.5 QUASINEWTON METHOD The quasiNewton update method is employed as an option to accelerate the convergence and improve the overall effectiveness of the modified Newton's iteration. As these updates accumulate.1 Evolution of QuasiNewton Method The basic concept of quasiNewton methodology is to construct an approximate inverse Hessian (stiffness) matrix using information gathered during the descent process. known as BFGS (BroydenFletcherGoldfarbShanno) update. was introduced into the finite element method by Matthies and Strang [3. When the BFGS method was implemented and tested in MSC/NASTRAN.
2). Another way of finding the formula for the inverse stiffness matrix is to invert Eq. (3. It is also possible t o update approximations t o the stiffness matrix itself.where R(u) = P . It is natural t o attempt constructing successive approximations t o [K]' based on the data obtained during the iterative procedure.4]. i. By this scheme which is referred to as the DFP method [3.5.e..5. {ui') and {ui).1).. Recalling the complementary roles of [K] and [K]' with respect to the quasiNewton vectors in Eq.F ( u ) with F ( u ) being the aggregate vector of element nodal forces. rather than its inverse. Therefore. This can be done by applying the general inversion identity known as the ShermanMorrison (BFGS) formula. . (3.R.. the scheme is a rank two correction procedure.e. The resulting formula is referred to as the BroydenFletcherGoldfarbShanno update for the inverse Hessian.e. i. the formula for [K] is found by interchanging y and 6. because they should satisfy: and = R"' . Notice that the formula simply satisfies Eq.2) while preserving positive definiteness and symmetry of [K]'. Assuming that the load stiffness due to follower forces is negligible..5. i. the stiffness matrix is formed by In view of Eq. The earliest quasiNewton scheme was proposed by Davidon and later elaborated by Fletcher and Powell.5. (3. (3.6). the data from two points. the inverse Hessian is updated by adding two symmetric rank one matrices a t each iteration. should provide some information about [K].
When the change in [K]* due to BFGS update is negligible. Case 3. Case 5.2 Criteria for BFGS Update There are cases where the BFGS update is not appropriate.Numerical experiments have indicated that the performance of the BFGS method is superior t o that of the DFP method [3. When the BFGS update makes [ K ]near singular. (3.5. The stiffness matrix (therefore its inverse) is assumed t o be symmetric and positive definite throughout the derivation.8) where [ j = cI 1 = V 6 j I T {yj) ' This is a recurrence formula employed in MSCINASTRAN. Case 4. An equivalent form of Eq.5.5.8) can be expressed in an equivalent form as follows: where with and . Case 2. These cases include the following: Case 1. When the solution diverges at the current iteration. The BFGS update formula in Eq. Z' and 3. Characteristics of such cases should be identified so that undesirable updates can be avoided. Notice that the index j for the BFGS update may be different from the iteration index i. (3.4]. When the BFGS update makes [K]I near singular.7) can be expressed for the jth BFGS update as: [~j]'= [cjlT [K~I]' [Cj]+ ~j {6j) {6jjT [ I ] . When two successive data points involved in QN vector generation are too closely spaced. which is applicable t o every pair of quasiNewton(QN) vectors.zj {yj) {6jIT (3.5.
(3. Case 2. This nonunity eigenvalue itself or its reciprocal becomes the condition number.5. (3. Eigenvalues of [A] consist of multiple roots of unity and a single nonunity eigenvalue. Hence the undesired BFGS updates can be filtered based on the squared value of the condition number in Eq. In MSCINASTRAN. {6j)T{yj) approaches zero) makes the condition num ber approach infinity. If Ei approaches 1.11). Case 4. (3. the condition number approaches infinity.5.12). i.1. which causes [Kj] t o become singular. in which the updates are performed: T O L l < X~ < 1 . Based on Eq.12). As the condition number approaches zero.3 Implementation of the BFGS Update Strategy The overall picture of the iteration module.5.5.5.e. the following observations can be made: Case 1. 3. in which the BFGS update method is implemented. (3. The BFGS should not be needed near the convergence or in the linear case where X is close to unity. is shown by a flow diagram in Fig. which makes [Kj] tend t o be singular. the criteria for BFGS updates are established based on the squared value of Eq.. Case 5.TOL2 for E% 0 .e. will become nearly singular. A small change in energy (i.10].The condition number (ratio of the largest to the smallest eigenvalue) of the matrix [A] can be used t o determine the characteristics of [K]l. which can be simplified as: where Ei is the divergence rate defined as with {di) = [K]' {R~') and a is the line search parameter [3.. In case of divergence ( E i > I). 3.12). [Kj]'l Case 3. Undesirable BFGS updates can be avoided by establishing two admissible domains for the expression in Eq. X has an imaginary value.5.
MAXQN.14) where L is a lower triangular matrix and D is a diagonal matrix with the pivot elements in the diagonal. K ~ L D L ~ = (3. Notice that vectors {di) and {Ril) are generated before the line search is initiated and no QN vectors are created during the line search process. Then. the number of QN vector pairs to be stored is limited by a userspecified parameter. where the tolerances are determined based on the numerical experiments [3. expressed symbolically as d. If the candidate quasiNewton vectors satisfy the criterion based on the condition number. multiple QN vectors can be processed as follows: Letting {qjl) = [CjI {qjI and with initially .1 0 The quasiNewton vectors are determined based on the data from two iteration steps.. = K1 ~ i . the direction vector is obtained by the forward and backward substitution. where a is unity if the line search is not performed. The BFGS update can be suppressed by specifying MAXQN=O. It is noted that the stiffness matrix is updated upon convergence if MAXQN has been exceeded in order to stop dragging too many QN vectors. If MAXQN has been exceeded.e. Using Eq.14].e.5.5. The BFGS update is a sequence of contiguous processes of dot products (4n multiplications for each pair of QN vectors with n components). combined with a forward and backward substitution (FBS). (3.. i. However. which has already been decomposed. the QNV file is not altered until it is purged upon a stiffness matrix update.8) for the BFGS update.and 1 TOL2 + < X~ < TOL3 for E~ > 0. these vectors will be stored for the subsequent iterations by appending them in a file (named QNV). and {7j} = {Ri'}  {Ri). which is defaulted t o the maximum number of iterations allowed (defaulted t o 20). i. The algorithm begins with an initial stiffness [KO] (evaluated at some previous step).
the FBS operation applied to {qo) t o obtain { s o ) . This formula is processed recursively for pj and { q j .2.5. represents a modified Newton's iteration with the initial stiffness (Ko=6). which results in u1=u0+d1=1..4 OneDimensional Example It is instructive to examine the effects of the BFGS update in the analysis of a single degreeoffreedom system. i. the QN vectors will reduce to: = u 1 . 3.1 = 0. sweeping the index backward from j through 1. The computational procedure can be completed by applying the following recurrence formula: with Tj = ~j {~j~}. For this procedure.uO = 1.Pj {^lj). If the BFGS update is performed at this point. The first process. the external load ( P ) is increased from 5 to 8 .75. i. ending at point 1. The iteration process is illustrated with data points labeled 0 through 4. as shown in Fig. the recursion is processed forward for rj and s j ..e. Consider a fictitious structure that responds t o the displacement with the internal force described by a parabolic function: Suppose that the structure is in equilibrium at u=l with an internal force ( F ) of 5.5 and ~'=6.5 .5. sweeping the index from 1 to j. Starting from this point. 3.e.5 . Then.the recurrence formula is obtained as: where {qj1) = [Cj]{ q j ) = { q j ) .l ) . The end result { s j ) represents the direction vector { d Z ) .
(3. This is a significant improvement compared to the old stiffness (Ko=6) and the tangential stiffness (of a value 3) at point 1.5.639 (equivalent t o connecting points 1and 2) is arrived at by recursive application of Eq. (3.5.R1 = 3 .R2 = 1.5. Both of the QN vectors satisfy the criteria t o be used for BFGS updates in the subsequent iterations.12) and (3. From this observation.5.and 71 = R0 .10] with the BFGS update. the relative error is estimated: If the line search is enforced by using a linear extrapolation with Eo=l. from which the secant stiffness at 1 (connecting points 0 and 1) is found to be 3. Let us now examine the effects of combining the line search [3. Invoking the BFGS update criteria using Eqs.ao=O.(8 .8).(8 .70. the line search factor is found to be: .25 .5. will end at point 2 resulting in i: and p2 = 7. respectively.75 Introducing the BFGS update formula in Eq.75) = 1.6.7) = 0.7.95 which results in a modified [K] with BFGS update as: Notice that the stiffness value of 2.8) with two sets of QN vectors. To perform the line search at point 1. values of X2 for the first and the second QN vectors are 1.13). it can be deduced that the order of QN vector application is not commutative. The second iteration using l. although the first QN vector set has no effect on the end result in this case of a single degreeoffreedom problem.and a l = l .714 and 1.316. (3. The second pair of QN vectors can now be computed: and 72 = R' .
i.09. However. .714 x d l = 1..5.F3 = 0. the third iteration using BFGS updates without the line search would have given a more accurate result than point 3. which is the very point that was predicted by the second iteration using the BFGS update. point 3 shows a significant improvement.86 . This result occurs because the concept of line search methodology is developed based on the secant method using a single parameter.14. This point corresponds to point 2. 3.e. the quasiNewton vectors will be reduced to: and which indicates a secant stiffness (connecting points 0 and 2) having a value of 3.which results in: ul* = u0 $1. If the BFGS update is applied at this point.2. The next iteration will result in point 3 in Fig. Compared to point 2 obtained without the line search. with R3 = P . and the BFGS update is equivalent t o the line search in a single degreeoffreedom model.
=a lE' A NO T Compute errors Print iteration information Convergence Test ( C O N V ) Set K update flag ( N E W K ) YES FIGURE 3.Zj to Q N V A T YES A  .Z.i=i+l l ui = uil + a d ' R = P .1 ~f .[R'] = j Z l =1 1 Append [bj]. RO [bj]= ~ [ d ' ] [rj] [Ri'1 . ~.l= K.9 .' TOL1 = TOL2 = T O L 3 = 100 NQNV=O.[ y j ] . i=O uO. l l l v ~2 Compute E' = Line Search Iteration. 305 . .bT  NO .F(u') ' di+l = K .Flow Diagram for Iteration with Quasi Newton Updates.K. 6. 4 I If Q N V exists.5.t l l l NQNV = NQNV +1 Initialize: = PF(IJ~).J1 C .1.
0 1.0 Derived from Starting Point First modified Newton iteration After line search or BFGS update Second iteration with LS and BFGS Solution Point 1.5 l.0 3.088 2.2.0 6.70 7. Comparison of Iteration Methods (QuasiNewton method with or without line search) .0 3 4 1.75 K 4.0 Load (F) 10 I Applied load (P = 8 )   Displacement ( u ) FIGURE 3.91 8.29 2.Point 0 1 u F 5.956 2.5.0 2.8G 2 7.
(3. This is because the effect of Aup diminishes as the solution converges (Ap approaches zero).22). For a constant external applied load.1) is shown to be: where the follower matrix may be omitted. The BFGS does not have to be implemented in Eq. Two different approaches for adopting BFGS updates can be contemplated.21).5. (3.5. (3. where k implies that the BFGS update is incoporated into K . However. (3. (3.5.5.5 BFGS Updates for Arclength Methods The BFGS update is required to make the arclength methods more effective. The applied load varies as a function of the load factor (p) at every iteration in the arclength . (3.5.22) needs to be solved only once upon stiffness matrix update in the absence of the BFGS update. the quasiNewton vector is defined as in Eq. (3.5. The BFGS updates have been implemented for the arclength methods in Version 67. FBS has to be processed on both equations at every iteration.5. especially when the stiffness matrix is deficient. It is also possible to adopt the BFGS update only on Eq. It is necessary to modify the quasiNewton vector y for the arclength method. It has been proven that the BFGS update is effective and efficient for convergence.22).20) into two parts: and KAup = A P Then the trial solution is obtained by with AU = AuR + ApAup where A p is obtained from the constraint equation.5.21). Then. Eq.21) and (3. but the convergence will be expedited.7.e. The BFGS update may be applied to Eqs. The iterative process can be established by decomposing Eq. The iteration equation for the arclength method (Section 3. or Aup does not have to be reevaluated unless a new stiffness matrix is formulated. i. The solution should not be altered by introducing the BFGS. is evaluated at every iteration and Eq.5. The residual error..5.3. The BFGS updates compensate for the deficiency and frequently provide converged solutions in difficult problems. (3.4). R in Eq.21) is solved again by FBS.
The changes in p during the iteration can be accommodated in the quasiNewton vector (y) by where NINC is the number of increments for the subcase. Various strategies of the BFGS updates for the arclength methods are implemented with a control by SYSTEM(145). . The NASTRAN definition at the top of the input data may include NASTRAN SYSTEM(145) = n $ where n = 0 (default) t o update AuR and Aup at every iteration with y* n = l t o update AuR only with y n =2 to update AuR only with 7 ' It is noted that these options are available in Version 67.methods.
while the true solution is not known. render consistent accuracy. Tolerances should be realistic for the solution scheme to be efficient. a a a be reliable (cancellation of errors are not acceptable). In this context.7. the displacements could be in gross error while the residual load error is negligible. because improper criteria could cause inefficiency as well as inaccuracy. However. In order t o ensure accurate and consistent convergence.8. and for the solution t o be accurate. Both aspects must be defined properly for the criteria t o be effective. a be independent of structural units. or vice versa. It is rather astounding to find a scarcity of publications on this subject. The energy error accommodates both quantities and is usually adequate for most problems. The convergence criteria should: a be satisfied for linear cases a t all times. variations are considered in search of the best workable combinations of error functions and tolerances for a wide class of structural problems. .6 CONVERGENCE CRITERIA The convergence test is an important factor that affects accuracy and overall efficiency in nonlinear finite element analysis. 3.13] has revealed that there are no universally accepted convergence criteria t o date in the field of finite element analysis. 3. The convergence criteria are extremely important for the incrementalliterative solution strategy to be effective and efficient. A literature search [3. Conditions t o be met by ideal convergence criteria for a generalpurpose finite element analysis have been contemplated. Two distinct aspects are involved in the convergence criteria: a a Error functions t o be minimized by the iteration Tolerances of error functions within which errors are acceptable. multiple criteria with errors measured in terms of displacements. Outofbalance forces and changes in displacements should vanish upon convergence in an iterative process. and energy should be combined.3. for the solution scheme to be efficient.12. 3.6. on which termination of the iterative process is based. be independent of structural characteristics (stiffening or softening). It is the error function and the convergence tolerance that characterize the criteria. Error functions are formulated using the weighted normalization so that the error measures are dimensionless. loads. considering the significant impact of the convergence criteria on the accuracy and the efficiency of the computation.1 Rudimentary Considerations The convergence test is a decisionmaking process.3.
some distinct errors are not detected with this criterion. For example. When the convergence criteria are satisfied. Nominally. or vice versa. r excessively loose tolerances cause not only inaccuracy but convergence difficulties in the subsequent steps due t o cumulative errors. combinations of forces and moments or translations and rotations. by visualizing the loaddeflection curve for a onedimensional case. displacements are in gross error while the residual load error is negligible. The most natural and reasonable criterion for the convergence test is formulated in terms of an energy error. 3. The fundamental difficulty of the convergence tests for a structural analysis lies in the fact that the base vectors (forces and displacements) involve inconsistent units. on the succeeding incremental steps.e. Such cases compel the need for criteria in terms of loads and displacements. it can be noticed that the convergence criterion in terms of loads governs the stiffening structure and the criterion in terms of displacements governs the softening structure.be able to handle all the loading cases including constant loading. i. Scalar error functions for these criteria are formulated t o be dimensionless by introducing the weighted normalization. This would be the case if the degreesoffreedom in error have a very small or a very large stiffness.6. the outofbalance forces and the changes in displacements should be sufficiently small so that the remaining error is not physically significant nor will it cause any detrimental effects. while an error in forces is dominant when the model is expressed in newtonmeter. unloading. Although the convergence test in terms of energy errors is usually adequate. These conditions dictate the formulation of error functions to be discussed [3. energy quantities do not pose problems of inconsistent units due t o mixed units associated with translations and rotations.14].. namely. and no external loading (applicable to creep analysis). r have smooth transitions after the stiffness updates and loading changes. Indiscriminate use of these vectors will cause unitdependent convergence criteria. The energy error is the logical choice because both the outofbalance forces { R ) and the change in displacements {Au) should be minimized by the iteration process. . Convergence tolerances have the following effects: r excessively tight tolerances cause a waste of computing resources for unnecessary accu racy. numerically or physically.2 Convergence Conditions The iteration continues until the convergence is attained by satisfying the convergence criteria and the residual error vector at convergence is carried over to the next incremental step. the error would be dominated by moments if the same model is described in newtonmillimeter. Furthermore.
3) are not dot products and should be viewed as weighted vector norms. utilizing the triangle inequality. (3. the contraction factor (q) of { A u ) should be introduced because the error in displacements is not known.3).6.. The weighting functions are chosen so that the products have common and consistent structural units. The error function in terms of loads (printed a t every iteration under the heading EPI) is made dimensionless by a weighted normalization using the total displacement vector. in which the numerator and denominator are defined respectively as in Eqs. and {u) and { A u ) represent the total and incremental displacements at the current iteration. the energy units resulting from the products are common and cancel each other.3 Error Functions and Weighted Normalization All the error functions are defined as relative errors in terms of vector norms. In order t o formulate an error function in terms of displacements.6.6. thus. Norms as defined in Eqs. i. Assuming . (3.2) and (3. are not raw vectors but weighted vectors using other vector quantities as weighting functions.6.6.6.4). where subscript 1's are omitted for simplicity in the following development.e. based on the concept of the CauchySchwarz in equality. with and where { A P ) represents the incremental load (including increments in reaction forces due t o enforced displacements) and { P ) pertains t o the total load at the previous loading step. The vector norm is defined by a sum of absolute values of the components (II norm or 11 [I1). Notice that the current displacement vector is used as a weighting function. i.3.e.2) and (3. however. To define a relative error in terms of energy (or work) which is printed at every iteration under the heading EWI. They are devised t o prevent cancellation of errors. Notice that smooth transitions between incremental steps upon load changes are attempted by defining an absolute sum for { P I ) as in Eq. (3. the energy error is normalized by another energy quantity ( E * ) that is relatively stable throughout the analysis. Thus the normalization factor does not vanish unless the incremental load as well as the total load disappear simultaneously. Cancellation of errors due to orthogonality is thereby prevented effectively. The vectors used in these norms..
(3.6.ui11 2 11 Allz 11 1q The convergence criterion in terms of displacements is not accurate. The vector {L) and its reciprocal.6.. (3. the upper bound of the absolute error in displacements can be estimated.i.6. we adopted: with an initial value q1 = 0. All the product terms have the same unit of the square root of the energy. After trial and error. wj = (3. The weighting functions in this case will be formed using a mapping vector { L ) consisting of 1's for the translational degreeoffreedom and the specified characteristic length for the rotational degreeoffreedom.10) 6 and the norms of the weighted displacements are defined similarly t o Eq.99. we have where qn denotes the nth power of q.a geometric progression of the displacement increments.. the above expression is reduced to: 4 (3. where i is an iteration index. in which the weighting function {w) is formed by collecting the square root of the diagonal terms of [I<].e.e. the contraction factor is defined as a ratio of displacement increments between two successive steps.e. i. Taking the limit as n approaches infinity.6. the contraction factor may be illconditioned and does indeed fluctuate erratically. denoted by { L ' ) are used as weighting functions for errors . If the contraction factor is assumed to remain constant with a value less than unity.. The error function in displacements (printed at every iteration under the heading EUI) is made dimensionless by a weighted normalization using the stiffness matrix. i.7) 11 u . An alternative approach has been considered t o formulate error functions in loads and displacements using a characteristic length specified by the user. For all positive integers n.2). which implies more equitable participation of all terms in the error estimation. An averaging scheme is employed to smooth a fluctuating q. Contrary t o the assumption.7) with many assumptions. as is obvious from the derivation of Eq. Note that the contraction factor is not used for the divergence check because of its erratic and unreliable behavior.
However. The goal of the default tolerances is t o consistently provide sufficiently accurate solutions to a wide spectrum of problems without sacrificing efficiency. It is difficult to choose optimal default values for the convergence tolerances. 3.in displacements and loads. Three error functions (in terms of displacements. respectively.P.4 Implementation The convergence tolerance determines the efficiency of the solution scheme as well as the accuracy of the solution. EPSP. < EPSU (= l o w 3 by default) Ep < EPSP (= E. loads. i. and energy) are computed and compared to tolerances. Thus. The default values have been determined as a result of numerous tests on 47 problems with a wide variety of characteristics and options.. The tolerance should be realistic. Thus. efforts have been made t o set the default values to provide reliable solutions to the general class of problems. default tolerances should be adhered t o until good reasons are found to change them. With this approach. The default option is designed to comply with this requirement. The solution should converge in a single iteration if the structural response is linear.5) and (3. * EPSP (= 10lo by default). However. it may be difficult or cumbersome to determine what the characteristic length should be. i. the solution t o the linear case should still converge regardless of the convergence criteria. The convergence test is performed at every iteration after the line search process (if applicable).9) using {L') and {L) . respectively. . However. E.e.. if the user inadvertently specifies an undesirable condition such as an extremely small tolerance.6. Ep < It is noted that divergence conditions are established independent of convergence criteria.6.6. and/or W ) are designed t o be satisfied for convergence. This method is intuitively appealing from an engineering perspective. However. only those criteria chosen by the user (combinations of U. and EPSW are tolerances specified in the NLPARM entry. the units are unified and errors become dimensionless.e. (3. Then the error functions in terms of loads and displacements are defined similarly to Eqs. the absolute convergence criterion is designed for this purpose using the error function in terms of the load error. < EPSW (= by default) by default) where EPSU. because the weighting functions are known functions and the errors are measured in terms of physical units of length and force. not too toght nor too loose.
(3. Without { P ) in Eq. the current convergence criteria combined with line search [3.8). This effect is pronounced in three problems which gave erroneous solutions in Version 64 and ran correctly in Version 65.6. have been analyzed t o completion successfully in Version 65. and tests were conducted in comparison with Version 64. for instance. however. and E p are amplified when the load increment is decreased.6.5) are satisfied. This is the case when the creep relaxation is analyzed under residual stresses. The error in displacements is not an appropriate measure for the first iteration because of an arbitrarily assigned initial value for the contraction factor in Eq. as a complement t o the usual divergence classified as absolute divergence. Introducing divergence criteria independent of the convergence rate is proven effective because of the erratic nature of convergence behavior.3)) errors E.10] and BFGS updates [3. as indicated in Table 3. (3. Notice that 11problems out of 50.3.6. which can cause improper weight distribution. The significant contribution of the current convergence criteria to the overall performance of the program is demonstrated in providing reliable solutions with consistent accuracy for all the converged solutions.9) if criteria in Eqs.4). and Ep will fail if no external loads are applied in more than two consecutive loading cases.11] improved the convergence characteristics remarkably. (3. In view of Eq.2. covering a wide spectrum of problems. Numerical experimentation proves. in effect. the convergence tests will fail in the constant loading cases as in a creep analysis.6. The probable divergence is designed to expedite the convergence of the stiffening structure.4). The divergence criteria play an important role in trapping the runaway solution.5 Some Observations Some objectives of the ideal convergence criteria have not been fully met by the criteria as currently implemented. (3.2. the convergence tests will fail in the unloading cases. the reduced load increment effectively results in a tightened convergence test. bypassed by assigning large values to the tolerances. Convergence tests can be. .6. Two categories of divergence (absolute and probable) help t o handle cases with different structural characteristics. (3.6. resulting in a null vector for {P') in Eq. Admitting that the test data are not available in a systematic and statistical form. the performance measure in terms of convergence test is rather qualitative and heuristic than quantitative.6.1 and 3.6.6. (3. The work was carried out during the development of MSCINASTRAN Version 65.1) and (3.4).6. which failed t o converge in Version 64. Another potential problem exists in the displacement error for using the stiffness matrix as a weighting function. the results presented here are derived from testing and tuning of parameters and tolerances with fifty nonlinear problems as described in Tables 3. (3.6. however.6.2. With the modified Newton method. this approach uses an old stiffness matrix while the tangential stiffness may change significantly. Without { A P } in Eq. As illustrated in Table 3. Therefore. Due t o the nature of the subject. Convergence tests on E.6. that most problems do not require convergence criterion in Eq. Such difficulties can be avoided by introducing a fictitious load t o a disjoint nodal point in the model.
includes buckling analysis 31.Effects of stress reversal 47. GOODRICH .Imperfect spherical shell with shell element .Superelement model of two tips plus a residual(QUAD4) in sort 2 35.Curved boxbeam test for large displacements 36. compression 19. NLELAS3 27.Beam with Gap/Liftoff 5. CSR3065 .1 HEXA element to test plastic material 10.Static analysis of nozzle to sphere attachment 8. NLELAS . CTHEXA 11.Boundary condition test 2. BCTEST .Creep under variable temperature . SHLBEAM .Cantilever beam with boundary condition changes 6. VGAP .for Euler buckling . BEAMZ 4.Basic nonlinear test with rod elements 42. SBUCKL . NLSHELL .Elastoplasticcreep.compression properties 24.Nonsuperelement model of two tips plus a residual . UNIAX2R . HAMMER . CBEAM .1 HEXA and 2 PENTA elements to test plastic material 9. UNIAX2P . LGDTEST . APLATE .Gap error with friction .relaxation prediction 46. ZBEAM .Snapthru of a spherical dome 38. loading and unloading . UNIAX4 . SWAVE . TRIPOD 43. NLELAS2 . UNIAXlP . FOLFO .Nonproportional loading specified by strain path 28.Gap elements .Cantilevered rod with a step forcing function at the end.tension.2 DOE model with thermal load 20.Creep at constant stress 44. MNEWTON .Follower force test problem with CRDEMO 14.Code test for creep .Gap verification with nonlinear transient analysis 48. NLPATH2 . T C R P . RODTEST .Cantilevered beam with shell elements .Zshaped beam analysis with 10 QUAD4 elements 50. CLUTCH . GAPTST 17.mass connected to rod by gap . CTSHELL .Table 3.Nonlinearelastic HEXA with hydrostatic load 26.2 PENTA elements to test plastic material 12.Imperfect spherical shell with solid model.10BEAM model to test SMALLDB . SE66S2 .Nonlinearelastic material with differnet tension.Modified Newton method with line search on RODTEST 23. SFRAME 37.Superelement model of two tips plus a residual(TRIA3) in sort 1 34.Shell elements with bilateral nonlinearelastic material 25.2 layers of HEXA's to model bimetal 18.unloadiong.2 DOE test problem for BEGS method with line search . BMLIFT . GAPS . NLPATHl .opening and closing in sequence 15. DRUM .Gap test with electromechanical clutch 16.reduced from ZBEAM for short run 3.1D plasticity in rod 29. SUPLCOP .1 Test Problem Description I . NLROD 30. HDRULE .Nonproportional loading specified by stress path .A single QUAD4 to test plastic material.Cantilevered rod hit by a mass with initial velocity specified . SNAPS 39.for buckling analysis 33.Tire model with composite element and large displacement effects . reproducing isochronous curve 45.Drum head model with thermal load 13. CTESHP . BBUCKL .Zshaped beam analysis with 10 BEAM elements 49. CRDEMOS .Electromechanical clutch 7.Kinematic hardening rule test . CTPENTA .6.3 shell elements to test LGDISP 22. SE66S1 . NLCTIRE . testing stress wave 40. UNIAX2T . NLSOLID 32. LGDROD 21.Beam .Creep of plate under thermal load 41.NLSHELL .
.K Updates Problem V65 V65 V64 V64 Name 0 0 5 5 BBUCKL 0 0 2 2 BBUCKL(R) 0 0 22 26 BCTEST 6 30 BEAMZ 2 10 BEAMZ(R) 2 2 2 2 BMLIFT 0 6 0 6 CBEAM 12 54 3 CLUTCH 9+ 6 141 CLUTCH@) 0 16 0 14 CRDEMOS 5 5 8 CSR3065 8+ 2 15 2 CTESHP 5t 3 15 3 CTHEX A 5+ 1 9 3 CTPENTA 12+ 0 15 23 2 CTSHELL 0 0 3 3 CTSHELL(R) 3 DRUM 36+ 2 15 DRUM(R) 0 12 0 9 FOLFO 1 39 3 26 GAPS 1 29 3 24 GAPTST 14 3 155 GOODRICH 11+ HAMMER 2 15 9 HDRULE 21+ 1 13 5 6 HDRULE(R) 3 51 67 17 LGDROD 5 3 LGDTEST 17+ 40+ 1 1 LGDTEST 4+ 15+ 15 1 79 MNEWTON 4+ 3 13 6 39 NLCTIRE 1 35 NLELAS 1 0 40 NLELAS2 16+ 0 11 NLELAS3 4 25 20 4 NLPATHl 4 35 28 9 NLPATH2 3 10 9 3 NLROD 4 78 68 13 NLSHELL 11 21 NLSOLD 74+ 77+ 3 15 NLSOLID(R) 47(46) 14 RODTEST 12(9) 35(+) 13 51 60 13 SBUCKL 2 10 11 2 SBUCKL(R) 2 12 13 2 SE66S1 0 11 13 0 SE66S2 15 19 SFRAME 54+ 71+ 3 SFRAME(R) 7+ 0 SHLBEAM 82 2 55+ 4 SHLBEAM(R) 60 41 6 2 36 40 5 SNAPS 0 9 10 SUPLCOP 0 SWAVE 0 TCRP 14 13 1 1 2 14 22 TRIPOD 2 4 20 13 UNIAXlP 3 5 75 89 UNIAXZP 31 46 0 UNIAX2R 0 2 146 179 3 UNIAX2T 4 31 37 5 UNIAX4 22 22 4 VGAP 4 200 267+ ZBEAM 1907 54 335 822 2621 ZPLATE 292 (R) : Restart run. I Remarks 0 0 0 0 0 0 56 8 1 0 3 0 4 0 0 31 10 0 44 34 20 1 2 3 4 5 6 7 8 9  6 1 31 36 5 18 5 8 7 1 0 3 0 0 20 4 29(30) 6 0 0 0 13 5 4 0 15 0 10 1 1 12 13 14 .Iters No.S. V65 No. : Diverging solution. + 6 0 15 14 3 21 15 0 0 0 6 0 16 310 16 291 ble for this analysis  .: V64 No.Table 3. L.2 Test Results Cornparision : V64 vs.6.
and the total external load P is expressed as where Po denotes the applied load at the end of the preceding subcase.1 Basic Theory for ArcLength Method The concept of the arclength method is to modulate the applied loads in order t o produce solutions with displacement increments of manageable size for a given load step. In addition. Consider a residual load {R} where F represents the internal forces. . Although the postbuckling state is not usually allowed in the structural design. There are various approaches t o providing a constraint equation [3. 3.153. p + Ap) dictates that where [Elis the follower matrix. for instance.31. known as Crisfield's method.and is = AP. and p is the load factor varying from 0 to 1 (but not t o be limited to this range) within the subcase. In order t o modulate the applied load. [g] the stiffness matrix [I<].7. It is noted that they are not efficient solution methods.7. but effective methods t o render solutions in the unstable regime which is not necessarily a physically viable state. some convergence difficulties have been identified and resolved in Version 66A.7. In the design process.7. A P represents the load increment in the current subcase. an additional variable (the load factor) and a constraint equation are introduced.181. An arclength method. (3. Linearizing { R } about ( u . was implemented in Version 66 of MSCINASTRAN. (3.1) can be expressed as Based on Eqs.1) and (3. a number of options for arclength methods and the adaptive bisection algorithm have been implemented in Version 67.7 ARCLENGTH METHODS FOR POSTBUCKLING ANALYSIS The static nonlinear response of a structure can not be analyzed beyond the critical limit load by the conventional Newton's method.3. it may be desirable t o trace the response of the snapthrough or postbuckling behavior. the equilibrium condition at ( u + Au. the prediction of such response could be of interest in some cases. However. p ) . Eq. The arclength methods allow solutions in the unstable regime for such class of problems.
3.6) with Au = AuR + ApAup where A p can be obtained from the constraint equation.7. Therefore the subsequent iterations (i > 1) must satisfy {AU')~{AU~) * api= 0 (3.7.5) into two parts: Ir'AuR = R(u.7.9) has a disparity in the dimension by mixing the displacements with the load factor. For this reason.7.7.4): where the follower matrix may be omitted.7. 3. The iterative process can be established by decomposing Eq. 3.2 Riks' Method and Its Variations The displacement increment is limited by a constraint equation: where w is a userspecified scaling factor (to be specified in NLPCI) and A t is defined by if the initial value of A p is given. (3.10) may be reduced t o from which the load factors for the subsequent iterations are determined by .p ) Then the trial solution is obtained by and Ir'Aup = A P (3.10) + Recalling that the first iteration should result in Eq.7. (3.7. the scaling factor (w) is introduced so that the user can scale p to the appropriate dimension or delete the A p term. The constraint of Eq.3b).7.The iteration equation could be derived by rearranging Eq.1. The default value of w is zero (Fig. (3. The iteration follows the path on the plane normal t o the initial tangent as shown in Fig. (3.
7. 3. {nup} remains constant if the iteration process is the modified Newton's method.9) with an arclength of A as depicted in Fig. If the normal plane is t o be normal t o the cumulative incremental displacements for the preceding iterations as shown in Fig.and pi = pil + Notice that the normal plane does not change during the iteration by Riks' method. if w=0.7..7. the solution is sought on the surface defined by Eq. ! This constraint can be interpreted as keeping the incremental displacement constant.. the selection of the proper root of this equation becomes the most critical process for the success of this method. as shown in Fig. the normal plane may be updated at every iteration.7.7. There are two .3b. 3. In addition.3a. we obtain a quadratic equation in terms of Ap: where Since Crisfield's method leads t o a quadratic equation.7.7. (3. This variation of Riks' method has an advantage over Crisfield's method as it avoids the solution of a quadratic equation. 3.2. i. i.7.e. (3. Alternatively. (3. Substituting Eq. 3.10) should be modified to: The increment in the load factor for i > 1 is obtained by solving Eq.e. (3.8) into the preceding equation.10b).3 Crisfield's Method Instead of iterating on the normal plane. the orthogonality condition in Eq.
(3. Such is the case when the trial solution is far from the true solution and stays outside the region covered by the arclength. Such cases can be detected by checking the orthogonality of the incremental displacements of the two successive solutions.12).e. i. the root is chosen so that the angle between {ul. One method is t o reduce the arclength if it requires an excessive number of iterations to attain a converged solution.e. with . If this case is detected.uO)and {ui uO)is less than 90 degrees. It is also desirable to decrease the arclength if the stiffness ratio increases.e.uO)and {ui. In this case.7.7.4 Adaptive ArcLength Method It would be difficult for a user to estimate a proper arclength for multidegreeoffreedom problems. is the number of iterations required for convergence at the preceding step.. The wrong choice of the root could cause an unintentional loading path reversal.uO)is an acute angle. The arclength should be continuously updated at every increment using the information gathered during the preceding increment. by which the solution returns t o the previous state. i.. i. Ap 1 = p1= 1  where Id is the desired number of iterations for convergence and I... The initial arclength for the Crisfield method can be determined by with NINC Notice that NINC is a userspecified value for the number of increments in the subcase.The root is chosen so that the angle between two vectors {uiI.. 3. the trial solution vector is scaled so that the direction vector intersects with the surface defined by Eq. There are cases where no roots can be found.
In the stiffening case this limit may be exceeded at the first increment of the subcase. defaulted to 0.19) where MINALR and MAXALR can be userspecified parameters. lee. j . Atnew ALRATIO = Atold Combining two criteria. (3. I The adaptive process should be based on the arclength ratio.7. Then the arclength can be updated at the beginning of the next step based on ALRATIO as follows: = ALRATIO * Atold A .20) at every iteration with p* = C 1 NINC . i.at.e. If the adjusted ALRATIO falls outside the bounds.7.where the subscripts ( j . ! If the iteration produces pi which exceeds the value of p for output points. Imaz In order t o maintain the stability for the adaptive process. This limiting process may be implemented as follows: Process Eq.2) denote the converged solution steps. respectively. ALRATIO should be reset t o the limit.20) +A~*AU$ where p* is the specified value for output (end of subcase or intermediate points).1.25 and 4. the computation is simplified as Ir'..e. MINALR < ALRATIO < MAXALR (3.7. Api should be adjusted t o the specified value of p for the output. j . the new arclength ratio may be adapted to the nonlinearity by If If (iz (dZ 1 and Ii7Tatio then ALRATIO = MIN Imax and  then ALRATIO = MAX else ALRATIO = 1.. p* . Ap* = MIN(Ap.. i. ALRATIO should also be bounded.pi1) and Au* = Auk (3.. In terms of the arclength..o = ABS Apn * A!.
p 31 . the sign of Apl should be reversed. However. Ap'3 = . . In the unstable regime where the stiffness is negative.. i. The bisection procedure will continue until the iterative process is stabilized and a converged solution is found. The arclength should be updated upon convergence whenever this limiting process takes place. the load factor decreases with a forward step.3 .UO ( ( < ne + Api = Ap* and Aui = Au* otherwise ignore Ap* and Au* and proceed t o update ui and pi. the number of contiguous bisections is limited by a userspecified parameter. otherwise p* = 1. The variable arclength at every increment invokes the recovery from the bisection process once the difficulties in convergence are overcome. The sign can be determined by the sign of a dot product.e. This possibility should be examined at the beginning of each increment. Notice that this limiting process in effect reduces the arclength method to the conventional Newton's iteration.if INTOUT=ALL. Then if 1) ui nu*. When this happens. MAXBIS in the NLPARM Bulk Data entry. The bisection is combined in concert with the stiffness matrix update strategy. If the iterative process using the arclength method tends to diverge.( p . ) An adaptive bisection algorithm is also incorporated to cope with divergent cases. the arclength is bisected.
Riks' Method u Figure 3.U Figure 3.7. Modified Riks' Method .7.1.2.
(a) Arclength in terms of Combined Variables (b) Arclength in terms of Displacements Figure 3. . Crisfield's Methods.7.3.
the obtained solutions were discontinuous. the unloading curve should follow the reversed loading curve. SNAPTHROUGH AND SNAPBACK: A SPHERICAL SHELL WITH BACKING PLATE A spherical diaphragm resting on a rigid backing plate is subjected to a uniform pressure. and changes in constraints due to the backing plate. and the corresponding displacement is 0. During the snapthrough process (between points A and B). the pressure rose rapidly as the solution proceeded. and the deformation pattern is assumed t o be axisymmetric because the diaphragm is connected t o a plunger which prevents rotation of the apex.98 psi. This model was analyzed by Crisfield's method and the results are compared to those obtained by the conventional Newton's method [3. Such examples are illustrated below. .) despite the increasing pressure.18 in. Since the apex point came into contact with the backing plate at 0.5. the displacement increased with a decreasing load until the stiffness became positive again when a tensile stress started t o build up in the membrane direction of the shell. the point did not translate beyond the gap distance (0.757 in. The contact process between the diaphragm and the backing plate was simulated by gap elements as shown in Fig.7. therefore.5 Verification Problems The most notable applications of the arclength method are analyses of the snapthrough behavior and the postbuckling behavior. since Newton's method is unable t o trace the structural response throughout the snapthrough process. A 10degree sector of the diaphragm was modeled using shell elements with axisymmetric boundary conditions.3.61 in. Note that the unloading steps did not coincide with the loading steps since the load increment varied adaptively during the analyses. the convergence behavior was rather erratic and unpredictable around the critical loads. 3.97 psi. This problem does not involve the material nonlinearity and. The analysis objective is to find the structural response for snapthrough and snapback under the clamped boundary condition.10]. In addition. This proved the validity of Crisfield's method in MSC/NASTRAN. The snapthrough occurs at around 0.4.7. follower forces. In the latter. The main features of these problems were geometric nonlinearities due t o large displacements. The material is linear elastic. Additional constraints are applied after the snapthrough by the backing plate. Beyond point B.7. This turning point (point B) corresponds t o the snapback point with a pressure of 0. 3. at the apex point. The loaddeflection curve at the apex point. The missing solution portions were conjectured and represented by dashed lines.41 psi and a displacement of 0. is shown in Fig. constructed by combining loading and unloading paths.
BOUNDARY CONDITIONS G I 7 SHELL SPHERICAL A SECTION OF A Figure 3.7. . Spherical Diaphragm Model with Axisymmetric Boundary Condition and Constraints by Gap Elements.4.
.. i . . Vertical Displacement at the Center (Clamped Case).. 0.?.. / ... ! ....5.. . . ....?..x Loading (Newton) Unloading (Newton) Loading (Crisfield) Unloading (Crisfield) Conjectured 0 X ! A . : I . .7. j .B ~ ~ . 0 x 0 Load (psi) \ v rt & A ' \A\ M 3 g .r: ..7.75 Figure 3. $ .5 Displacement (in) 0. Load vs.
In the analysis. Since a finite number of load steps was used. By using the arclength method.8 shows the loaddeflection curve at the apex.. and 4000 psi at the end of SUBCASE 1. The peak pressure and its corresponding displacement are identified as 3574 psi and 0. the buckling load as well as the postbuckling behavior can be predicted in a single run.19]. The run automatically stopped when the number of load increments reached the maximum limit MXINC. respectively. On the other hand. and SUBCASE 3. Fig.7. The buckling pressure and displacement estimated by this method are 3547 psi and 0.6. and 0.. starting from the solution at P=3300 psi.0139 in. The initial imperfection was introduced by making the radius of curvature near the apex greater than the shell radius. This solution is not as accurate as that predicted by the arclength method. 3. When the same model was analyzed by using Newton's iteration method in all three subcases. the maximum displacement at buckling lies between 0.9. The finite element model shown in Fig.7. The problem was assumed to remain axisymmetric in geometry and loading throughout the deformation. the external pressure was gradually increased: 2000 psi. respectively. it was decided to use Newton's method in the first two subcases and Crisfield's in the third. The solution proceeded successfully and passed beyond the peak point. while the periphery was clamped. 3.0139 in. The details of the geometry and the material are shown in Fig.7 represents a 10degree sector with axisymmetric boundary conditions. the predicted solution would not have passed through the exact peak point.POSTBUCKLING BEHAVIOR: AN IMPERFECT SPHERICAL CAP A spherical shell with an initial imperfection was analyzed [3. The large displacement effect was also included in the analysis. respectively. but the error in the buckling pressure should be small because the curve is plateaued in the vicinity of the peak. 3000 psi. . the solution diverged in the third subcase. A restart run was made in SOL 66 for the eigenvalue analysis [3. SUBCASE 2. 3. Since Newton's method is usually more efficient than the arclength methods if the stiffness matrix is positivedefinite and not close to being singular.. where the DECOMP module encountered negative terms in the factor diagonal of the stiffness matrix.0159 in. 3. Solutions a t P=3400 psi and 3500 psi were repeated and a buckling analysis was performed using the stiffnesses between those two steps. unlike Newton's method.0122 in. The deformed shapes are shown in Fig. The material was elastoplastic with von Mises yield criterion and kinematic hardening.20].7. The shell was subjected t o an external uniform pressure.7.
267 in.8 x 104 psi Strain Hardening: Slope E. = 1.3 q = 7. P = 37.7.32908 in. y = 20' b = 0. Imperfect Spherical Shell. R.8251 in.3065' c = 0.7612'  Material: 7075T6 Aluminum E = 10.6.1506 in. R 0. a 14.1 x 1O6 psi Kinematic  Boundary Condition: Periphery Clamped Figure 3. . = 1.8 x 1O6 psi u = 0.Geometw: h = 0.0251 in. a = 0.14328 in.
Figure 3.7.7. . MSCJNASTRAN Model for Imperfect Spherical Shell.
. (psi)  i $r i ? t O Newton's method $ A.  0. A  * A a a C B 3000.bl I o. LoadDeflection Curve of Imperfect Spherical Shell. Buckling point (Newton) Crisfield's method Peak point (Crisfield) Kao A 4  1000.   Pressure 2000. 0.05 Central Deflection (in) Figure 3.8.00 I I o.7. b2 1 I I I 0.04 0.b3 0.4000.
0286 in.0: Undeformed Shape A: at q=3574 psi. u=0.0139 in.7.9. . u=0. Deformed Shapes at Different Buckling Stages. Figure 3.0495 in. B: a t q=3243 psi. u=0. C : at q=2863 psi.
quadrilateral and hexahedral elements are generally preferred over triangular. pentahedral or tetrahedral elements.. Offset options in beam. rather than quadratic or cubic elements. roller or ball joints. For lack of better information. If improved accuracy is required at such joints. In such cases. and caution is needed when using these elements in bulk. Ideal boundaries are represented as free. Triangular and tetrahedral elements can exhibit excessively stiff behavior. Modeling the joints (such as bolted. lowerorder elements are often preferred over higherorder ones because of their robustness and reasonable accuracy at reduced costs. especially for material nonlinearity. some provision (e.g. and if these elements are incompatible (e. MSCINASTRAN employs linear elements. Modeling of the boundary conditions at the supports poses similar difficulties. OVERVIEW In nonlinear finite element analysis. Alternatives to these offset options are .g. appropriate constraints) may be necessary at the interface boundary. When using lowerorder elements. pinned. the joints are usually modeled as rigid or free in certain degreesoffreedom. clamped. the characteristics of the joint (stiffness and/or damping) may have to be identified from experiments or the local analysis of a detailed model at the joint. The reality tends to be in smeared condition. riveted. These offset options are not fully validated for nonlinear analysis. Another common modeling problem involves offsets a t the interface boundary. t o process nonlinearity. or welded) is particularly difficult. HEXA and QUAD$).Chapter 4 FINITE ELEMENTS IN NONLINEAR ANALYSIS 41 . plate and shell elements are provided by MSCINASTRAN. Caution is also needed when different element types are combined in a model..
1.RBAR Notes: * denotes elements capable of geometric and material nonlinear processes. Linear elements [4.1]. However. CONROD and CTUBE) differ only in input convenience.CTRIA3 CONM2 GENEL RBE2. The ROD element is a straight element connecting two GRID points. As for geometric nonlinearity. Active nonlinear elements may only be used in the residual superelement. however.using very stiff rod elements for the offset distance using rigid elements or MPC relations if the large rotation is not involved.1 Line Elements The truss type elements (CROD. The ROD element can sustain axial force P and torque T. Elements which may be used and are recommended for nonlinear analysis are tabulated below: Line element Surface element Solid element 0t her elements CGAP CELASi CTRIAX6 * * * CROD CONROD CTUBE CBAR * CQUAD4 CQUAD8 * CTRIA3 CTRIA6 CSHEAR RTRPLT * * * CHEXA CPENTA CTETRA +* CQUAD4. The MSCINASTRAN element library includes about forty elements. Elements that are potentially capable of nonlinear processes (those denoted by *) become actively nonlinear if the parameter LGDISP is turned on or nonlinear material is assigned t o those via MATS1 or CREEP bulk data. 4. with axial and torsional .RBES * CBEAM CBEND RROD. MSC /NASTRAN does not currently support large strain capability. of which some are considered obsolete and some are not applicable t o the nonlinear analysis. producing axial and shear stresses as follows: . denotes a planestrain element. large displacement is treated effectively by computing element stresses and strains in the updated element coordinates. All the upstream superelements are assumed t o remain linear throughout the analysis. may be used in the residual superelement for the region that remains linear throughout the analysis. stiffnesses in displacements u and 8 respectively. + All the elements listed above may be used in the nonlinear analysis (SOL 66 and SOL 99) if they are combined properly in the model.
Geometric nonlinear effects are reflected in the stiffness via the differential stiffness and the internal forces using the method of updated element coordinates. properties do not vary along the length). therefore the stress and strain are assumed constant throughout the length. It provides exact linear solutions for end loads of any kind.. . and that the torsional stiffening due t o warping is neglected. The strain is computed based on the small strain theory. respectively. i. where L and Lo denote the current and original length of the element. However.e.. A L GJ ke = L and Aax kd = L The BAR element is derived from classical beam theory. (4.e. i. These limitations are relaxed with the BEAM element in addition to the nonlinear capabilities to be described in the following section.e.and with ox = E dcx and Toz =GYB~ where E is the Young's modulus or an appropriate modulus for the nonlinear material and G is the shear modulus. Its major limitations are that it is prismatic (i.= E . and c the userspecified value to determine the stress. The torsional stress is assumed linear and not coupled t o the axial stress. The stress is determined based on a single Gauss integration.1). the linear shear stress is simply computed by where T is the torsional moment. The stiffness matrix for a ROD element is the (8x8) matrix shown below: with where k. nonlinear material behavior will only be exhibited in the axial deformation.. the plane crosssection remains planar under deformation.1. that the shear center and the neutral axis coincide. J the torsional rigidity. Geometrical properties and the directions of internal forces of the BEAM are shown in Fig.
The SHEAR element can be used t o model the shear panel in the linear part of the structure. Linear solid elements applicable t o the nonlinear region will be discussed in the following section. no material nonlinear capability exists in the composite elements.3 Solid Elements Solid elements (HEXA. are always recorded and output in the element coordinate system. as shown in Fig.The BEND element forms a circular arc. The different material coordinate system with respect to the element coordinate system may be defined in the shell element connection card as shown in Fig. In other words.2 Surface Elements The plate and shell elements (QUAD4.1. Stresses and strains for these elements are recorded in the material coordinate system. Properties of shell elements are specified in the PSHELL data. this is not the case in the doublycurved shells. . the quadrilateral elements are more reliable than the triangular elements. However. The element coordinate systems are implicitly defined by positions and the order of the nodal connectivity (see description of HEXA. The material coordinate system (defaulted to the element coordinates) can be defined in the PSOLID entry.1. The quadratic elements (QUAD8 and TRIAG) are applicable only for the linear part of the structure and provide midside nodes that may be deleted. corresponding to ROD and BAR. connecting two nodal points. Components of forces and stresses. The RTRPLT element may be used to model a rigid plate in the linear structure. TRIA3. respectively. In general. TRIAG) can be used to model membranes. The element coordinate systems are implicitly defined by positions and the order of the nodal connectivity. no nonlinear material may be referenced. QUAD8. (4. If the PCOMP is selected. plates and shells. 4. but PCOMP can be alternatively used to model the composite layers. Various schemes of Gaussian quadrature may be selected for the volume integration in the element property data specified in the PSOLID entry.2). These quadratic elements can represent the singlycurved shells or linear curvature shells more accurately than the linear shell elements at the same cost. PENTA and TETRA for details). RROD and RBAR are rigid elements. Linear shell elements (QUAD4 and TRIA3) may be used for nonlinear analysis with geometric and material nonlinearities.3). because the TRIA3 element exhibits behavior too stiff in membrane action.1. The quadratic element with midside nodes may not be used in the nonlinear region. PENTA and TETRA) are quadratic isoparametric elements that may be degenerated to linear elements. (4. 4. Its primary use is to model a piping elbow.1. however.
include a lumped elastic spring (ELASi). Plane strain elements were implemented by suppressing bending and enforcing the plane strain condition instead of the plane stress condition. The input d a t a may be obtained either from another computer run or from test data. an axisymmetric element (TRIAXG). but without effects of large rotation. Other elements. which are linear but available for nonlinear analysis.4.4 Other Elements The GAP element simulates a pointtopoint contact including the friction effect. The GENEL element is used t o represent the stiffness of a substructure which has an arbitrary number of connected grid points. RBE3). The GAP element is a nonlinear element and loses its significance if included in a linear analysis. Plane strain elements are available for nonlinear analysis using QUAD4 and TRIA3 elements by specifying 1 in the MID2 field of the associated PSHELL entry. a general element (GENEL). and rigid elements (RBE2. a lumped mass (CONM2).1. The formulation is based on the penalty method. .
1 BEAM Element Coordinate System and Element Forces 4.Figure 4.1.6 .1 .
1.(a) TRIA3 (b) QUAD4 Figure 4.2 Plate and Shell Element Coordinate Systems 4.7 .1 .
(a) Forces and moments (b) Stresses Figure 4.3 Forces and Stresses in Plate Elements 4.1  8 .1.
. 4.2) .7 ) is used for a quadrilateral surface in which each corner node has an extremum value of $1 or 1 in and 7 . (4.e. parametric mapping is frequently used t o map an irregular region into a regular one.4. element matrix describing the straindisplacement relations. Then. In what follows. volume integration for stiffness and mass matrices by Gauss quadrature.2.2.2.etc. For instance. i. N2 = Lz .e t c . a natural coordinate system (I..2 FORMULATION OF ISOPARAMETRIC ELEMENTS In the finite element method. ..2. The mapping is isoparametric if the same shape function is used t o interpolate the displacement field as well as the geometry. The coordinate system used in the parametric mapping is a natural coordinate system suitable for the geometry. The mathematical expression for this parametric mapping or interpolation is called a shape function. the geometry of the internal points of the finite element can be described in terms of the nodal coordinates by the parametric mapping. The displacement field inside the element should also be interpolated in terms of nodal displacements. The merit of isoparametric mapping is that the displacement field is invariant t o the orientation of the cartesian coordinate system x and y. N1 = L1 because x = C Nix.1) that the shape functions are simply the volume coordinates.1 Isoparametric Coordinates Cartesian coordinates are not convenient t o describe the geometry or the displacement field of a tetrahedron element. Let us introduce a set of volume coordinates (L1 L2 L3 L4). Most of the MSCINASTRAN elements are isoparametric elements. shape functions. The relation between volume and cartesian coordinates can be established for a linear tetrahedron (without midside nodes) as follows: It is obvious from Eq. and element loads are derived for a tetrahedron element t o illustrate element related operations.. . (4.1. 4. Volume (1234) as shown in Fig. such that L1 = Volume (P234) . derivations are shown for the isoparametric coordinates. . in which the shape functions are expressed in terms of isoparametric coordinates.
. . .BlO] nodal displacement vector with Since the shape functions are defined in terms of local coordinates. and u= C N . Bi . Notice that these shape functions. (4. where strain vector {&lT = < EX ~y EZ Y X ~ yz y Yrx > element matrix [ B ] = [BlBS . Volume coordinates involve four coordinates . Eqs. a coordinate transformation is required t o obtain global derivatives.2. These are N1 = and (2L1  l)L1. With the isoparametric element.e. i. . the same shape functions are used to describe the displacement field as well as the geometry. etc. x= The strains are calculated as = [Bl{ul C N ~ x .3).2) and (4. . satisfy element convergence criteria: integrability (en'continuity for nth derivative) and completeness (no straining by a rigid body mode.2. U ~ etc. . constant strain condition. etc.Shape functions for the quadratic tetrahedron can be derived using Lagrangian interpolation.etc. and continuous displacement field). . for corner nodes for midside nodes Ns = 4L1L2.
deletion of any or all of. d x d y d z = det J d t d q d c 4.2 (4.2].. 7. To avoid this difficulty.e. i. i.the midside nodes is permitted [4.2.10) Shape Functions To make tetrahedron element compatible with other solid elements (HEXA and PENTA). from which the element matrix is formed.. let us introduce local coordinates (J.2. The shape functions are modified with Kroneker deltas (65 . <) as follows: Derivatives with respect to the local coordinates can be expressed in terms of global derivatives using the chain rule. Notice that the determinant of the Jacobian matrix is called Jacobian which represents a volume change. and the Jacobian matrix will become rectangular.e.blO) where Si = 0 if the midside node i is deleted = 1 if the midside node i is not deleted . where the Jacobian matrix is expressed in terms of shape functions as This Jacobian matrix must be inverted t o obtain global derivatives.(one of which is dependent).
The goal is t o construct functions which are unity at the associated node and zero at all other nodes, regardless of any combination of deleted midside nodes, by the following scheme: at the corner nodes
at the midside nodes
N5 = 4t7S5,etc.
Shape functions are identified as follows:
Derivatives of the shape functions with respect to the local coordinates are obtained as follows:
These shape functions and derivatives may be reduced to those for 4noded and 10noded tetrahedron elements. It can be verified that, for any combination of deleted midside nodes, there exist a condition which satisfies the constant strain requirement.
4.2.3
An Example of Element Matrix
To illustrate the computational procedure, an element matrix for a linear tetrahedron (4noded) element is explicitly derived here. The shape functions and the derivatives are tabulated below:
Node
N i
arv,arv,m aq at
1 0 0

1 2 3 4
E
rl
c
1
1
0 1 0 1
0 0 1 1
Then the Jacobian matrix may be found as
[J]=
[
1 0 0  1 0 1 0 1 0 0 1  1
]1:; ;: j
Xl
Yl
Zl
214
Y14
= [x2.
234
y24
Y34
I:
214
24
Y4
where
(x;j =
x;  xj) is used for convenience. Upon inverting the Jacobian matrix, we have
[ J ]= det J
where
[ :; pi ): ]
a1 a2 a3
The global derivatives of the shape functions are
arv,arv,%aiv,

7
a2
a3 (a1
ax
ax
ax
ax
1 det J
bl
arv,arv,arv,arJ,=
ay ay ay ay arv,arv,arv,* az az az az
b2
b3
(bl

el
c2
c3
(el
+ + + f + +
a2
a3)
b2
b3) ~ 3 )
c2
Hence the element matrix will be
Notice that the rank of [B] matrix is 6.
4.2.4
Volume Integration of Element Matrices
By virtue of variational principles, the element stiffness matrix is derived as follows:
where D is a (6x6) material tangential matrix. While analytical integration is possible, there are some advantages in using numerical integration. The Gaussian quadrature is used for the tetrahedron as usual. The minimum number of intergration points required for nonsingular stiffness matrix may be determined based on Total dof  rigid body dof minimum n = No. of strain components for each point
These numbers are found t o be 1 and 4 for linear and quadratic tetrahedron, respectively. Using the integration formulas, shown in Fig. 4.2.2, the element stiffness matrices for linear and quadratic tetrahedron may be computed as
[Ice]  JdetJI B ~ D B for 4noded TETRA = 6 [Ke]  x = 6
1 1 .
ad
1
x
ldet J ( B ~ D B for 510 noded TETRA
gauss
where the scaling factor 116 is introduced t o compensate I J = 6 x (Volume). Onepoint scheme I should be used for 4noded tetrahedron (all the midside nodes deleted) and fourpoint scheme otherwise. It is difficult to determine lumped masses when some of the midside nodes are deleted. However, the consistent mass matrix may be obtained using the same shape functions, i.e.,
where [ N ] = [NIT N a I . . . NloI] with I being (3x3) identity matrix. Again the Gaussian quadrature 4point formula is used to find
[MIe=p,
1
x
gauss
ldet Jl [ N ] ~ [ N ]
Notice that 4point scheme is to be used even for 4noded tetrahedron. For computational convenience, the consistent mass may be converted to the lumped mass. One way t o achieve this is t o take the diagonal terms and scale them so that the total mass is preserved.
4.2.5
Element Loads and Equilibrium
Nodal forces are computed from the element stresses using element matrix, i.e.,
{ F J e=
The Gaussian integration is performed as
/v
B~{u)~v
(4.2.16)
{F)' =
for 4noded tetrahedron and
12 x 1
1 6 ldet
J [BIT{u) I
12x6
{ F J e=

1
24 g a u s s
x
ldet J [ B T ] { a ) I
for (510)noded tetrahedron. Thermal load is like an initial strain. Therefore, nodal forces due to thermal load are obtained by
where { E ~ = aAT < 1 1 1 0 0 0 ) ~ AT < crl, a2 . . .a6 > in general.
>. With anisotropic thermal properties,
{ E ~ becomes ) ~
The pressure load applied t o any surface of the tetrahedron may be distributed to the grid points using the shape functions applicable to the 6noded triangular element with removable midside nodes, i.e.,
where { n ) is a unit direction vector associated with a pressure (p) and
[Nt]= [nyrN;I . . . N;I]
with I being ( 3 x 3) identity matrix. Shape functions ( N i ) are shown in Fig. 4.2.3. Area integration should be performed using Gaussian quadrature formulas shown in Fig. 4.2.4, i.e.,
ac
V
gauss
During the nonlinear iteration process, the equilibrium is not reached until convergence is achieved. The equilibrium is sought in the global level when the residual load R approaches zero. The residual load vector is defined as
linear nonlinear
{ R ) = {P}e
{lieue} 
(4.2.21)
where {P) is the applied load vector including the thermal load and the global coordinate system (coordinate transformation required).
C implies assemblage in
4.2.6
Element Coordinates
The element coordinate system for the tetrahedron element is shown in Fig. 4.2.5. It is defined with the initial element geometry such that
e
The origin is at the first grid point in the connectivity, GI. The xaxis is determined by connecting the origin G1 to node G2, i.e.,
The yaxis is determined by orthonormalization (GramSchmidt process) of the edge with respect to xaxis, direction (FI3)
a
The zaxis is orthogonal to x and y according to the righthand rule, i.e.?
Then, the transformation from element coordinates to the basic coordinates is simply
{:I
basic = I T b e j { : ]
element
+{:;I
]
basic
where
[Tbel
=
[
i x j x kz i! . k y I :
2,
(4.2.25)
Jz
kz
and < z e ,ye, ze is the position vector of the element coordinate system with respect t o the basic coordinate system. Notice that [The] is an orthogonal matrix, i.e., [Tbe]l = [TbelT
>Fasic
4.2.7
Stress Data Recovery
In case of linear analysis, the stresses at the integration point are recovered as
with
AT = {N;}*{T~  T,;)
where {a), [B], {u)and{a) are defined in the element coordinate system, and the shape function {N;} interpolates Gauss point temperatures from the nodal temperatures. In case of nonlinear analysis, stresses are computed again upon convergence, starting from the last converged st ate (last converged solution of a and u), i.e.,
Given {sold) and {A&}= { E ) 
with
{&I [Bl{u)  { 4 A T =
the nonlinear material routine computes updated stresses { , , , which are stored in ESTNL. o,} To output grid point stresses, the stresses at the integration points must be extrapolated t o the nodal points. A linear extrapolation can be applied using stresses at the element c.g. and the corner Gauss point associated with a grid point. For a tetrahedron, referring to Fig. 4.2.2,
01
=
'Ja CU 
1 'Jo (1  4) 'Jo 114
+
where
00 =  ( ~ a
1
4
+ + +
' ~ b
'JC
' ~ d )
This yields
'J
=
0,
1 P 14p

P ( ' J b + 'Jc + ' J d )
14p
In matrix form for all the grid points in the element
where
(i]
grid
A
B A
B B
A
B B B
A 


Sym.
(i]
gauss
1P A =  P and B = 1  4p 1  4p This process must be operated on every component of stress vector. No extrapolation is required for 1point integration.
Figure 4.2.1. Volume Coordinates for Tetrahedron
4.2  11
Linear
Quadratic
1 point integration
4 point integration
Figure 4.2.2 Gaussian Quadrature for Tetrahedron
* Area coordinate system (t 7 C) such that
* Shape Functions:
Figure 4.2.3 Shape Functions for Pressure Load
4.2  13
Linear
Quadratic
1
1 point integration
3 point integration
Figure 4.2.4 Gaussian Quadrature for Pressure Load
Origin at G1 xaxis: in the direction of the edge from G1 to G2 yaxis: orthogonalized from edge (GlG3),
K3
zaxis: orthogonal t o x and y according t o the right hand rule
Figure 4.2.5 Element Coordinate System Definition
4.3
GAP AND FRICTION ELEMENT
One of the major areas in nonlinear analysis that poses difficulties in convergence is the application of GAP elements. This has been particularly true when friction is involved. Numerical difficulty is inherent in nature for the penalty method which is employed in the MSCJNASTRAN GAP element. Furthermore, the GAP element simulates a pointtopoint contact, and is often used to simulate surface contact problems. Deficiencies of the penalty GAP element are coped with by implementing an adaptive GAP element in Version 67 of MSCINASTRAN. The adaptability is based on the subincremental scheme with the capability t o update the stiffness at proper timing, t o bisect the increment when it is excessive, and t o adjust the penalty stiffnesses to the proper values in concert with the nonlinear and dynamic environment. In addition, the kinetic friction effect (different from the static friction) is introduced to simulate the physics more accurately by distinguishing stickslip forces.
*
4.3.1
Introduction
Contact problems are common in engineering practice. The GAP element simulates a unidirectional pointtopoint contact using the Bulk Data CGAP and PGAP. The GAP element is one of the major areas that pose difficulties to the convergence of the iterative process, particularly when the friction is involved. The penalty method is used t o simulate the rigidity between two degrees of freedom by adding a large value t o the stiffness so that the two have approximately the same displacements. Penalty values are introduced t o avoid penetration and t o enforce the sticking condition (static friction) between two contact points. Difficulties (or inaccuracies) arise when the penalty values are not properly chosen. The success of this method is highly dependent on the user's choice of penalty values and requires a compromise between accuracy and numerical performance. The Lagrange multiplier method has been used by researchers [4.34.51 in contact problems. The method enforces the constraints exactly. However, it also poses some difficulties. The main drawbacks of the Lagrange multiplier method are: (1) the problem size increases by Lagrange multiplier variables and (2) the stiffness matrix has zero diagonal terms, and requires pivoting and resequencing. Some variations of the Lagrange multiplier method have been proposed [4.64.81 t o mitigate the deficiencies. Another school of thought for the dilemma is t o alleviate the difficulties by implementing the adaptive penalty method t o choose a proper penalty value and adjust it to control the numerical stability and the accuracy [4.94.101. Although the constraint is not enforced exactly, the effectiveness of the penalty method lies in preserving the size and the bandwidth of the stiffness matrix. MSCINASTRAN employs this approach in the adaptive GAP element implemented in Version 67.
When the GAP is closed and sticking (static friction). dictated by the static coefficient of friction (p. F = Ktv.3.) and the transverse shear stiffness ( K t ) . . no lateral stiffness).1) When the GAP is closed and sliding (no friction). and Fy = F = 0. where Kb is an arbitrary open stiffness.u.) and the friction forces in lateral directions (F. Force components in the GAP element are the axial compressive force (F. where   with the friction dictated by the kinetic coefficient of friction pk. w) in the element coordinate system.3.4. . F = ICau > 0. and F = Ktw.2 Basic Considerations for Penalty GAP Element The GAP element changes its status when the load is applied. It is noted that the sticking condition is . (4. . When the GAP is closed and slipping (kinetic friction). and F.2) where Iiais the penalty value for the closed stiffness. . (4. 0. Kb sym. F = ICbu 5 0. v. and Fy= F = 0 . The element stiffness matrix can be formed as with I ={ a{F> a@> where the stiffness components are derived for each GAP status as follows: When the GAP is open.3. These internal forces can be computed based on the GAP status as follows: When the GAP is open (no contact. Consider the internal forces of a GAP element with an isotropic friction. .) in terms of displacements (u. where Fx = K. 0.
Since the Newton's method is sensitive t o abrupt changes in stiffness during iterations. Optimal penalty values ( K . The stickslip behavior of the frictional GAP element is pathdependent in a manner similar t o the plasticity. I{ = sym.) are difficult to assess. sym. they have t o be updated continuously during the deformation process. When the GAP is closed and slipping. When the GAP is closed and sticking. Furthermore. adaptive stiffness update alone cannot stabilize the solution and the increment size should be reduced.11].When the GAP is closed and sliding. It is noted that the slipping due to friction introduces an unsymmetric stiffness matrix. The timing of the stiffness update is crucial to the convergence. which necessitates an adaptive stiffness update criterion based on the GAP status. In order t o trace the pathdependent solution. and I<. T h e lateral forces due to friction should be considered for equilibrium. because the stiffness of the adjacent structure changes during the incremental process. It becomes necessary t o adjust penalty values adaptively by the program throughout the analysis.  ["a R] 0. it the GAP status undergoes two or more consecutive changes. . If the internal forces in the GAP elements are computed in a single step. the accuracy deteriorates when the increment produces large changes in displacements with friction. 0. it could cause a divergent or oscillatory solution when the GAP changes its status. the subincremental scheme within an incremental load step is desirable [4. Furthermore. There are some inherent difficulties involved in the penalty GAP element. Ir'.
4. The element coorO dinate system does not rotate as a result of deflections.3.3. 4. If GA and GB are coincident in space.3. CID is used t o define the element coordinate system.1. Properties that characterize these curves are specified in the Bulk Data PGAP. .4. In other words.1 The G a p Element The axial and lateral forcedeflection curves of the GAP element are shown in Figs.2 and 4.3.3. X3) defining the xy plane. I Go (orientation vector) Figure 4. The orientation of the y and z axes is determined by G (or XI.GO X2 X3 CID 3 Line GAGB defines the GAP element xaxis as shown in Fig. the element coordinate system is not updated even when the geometric nonlinearity is considered (LGDISP=l). X2.3.3 User Interface for G A P Element Connectivity is specified in the Bulk Data entries CGAP as shown below: 1 2 3 4 5 6 7 8 9 10  CGAP CG AP EID 17 PID 2 GA 110 GB 112 X1.
F. (compression) t tension I . Ka 136 Kb Kt 1.3 Gap Lateral Force vs. F o 0.2 Gap Axial Force vs.I + UA .3.001 +PG +PG 0.UB (compression) 1 Figure 4.PGAP PGAP PID 2 u o 0. +PG TMAX MAR TRMIN 0.UB 2 Uo ) Nonlinear Shear Figure 4.E5 PI 112 l. loo. Deflection . 0. Deflection ( slope K .3. is used when UA .E2 0.
value for li. If TMAX element is chosen. which corresponds t o the GAP element coordinate system.OIIia 5 lit I<. Upper and lower bounds of the adrespectively.) are the input for p1 and pg.. When TMAX = 0. When the GAP is closed and if there is friction.g. defaulted t o 100) is used only for penalty value adjustment of the adaptive GAP element. If the nonadaptive GAP element is chosen. the adaptive GAP of GAP elements: adaptive GAP and nonadaptive GAP. the friction effect is ignored (equivalent t o p1 = p2 = 0).The initial GAP opening is defined by Uo (defaulted t o 0)) not by the separation distance.0 5 TRMIN 5 1. should be chosen t o be three orders of magnitude higher than the stiffness of the neighboring grid points.001).UB 2 Uo). For most contact problems.) in the axial direction. there is a small open stiffness ( K b .0. where IiO the userspecified is justed penalty value are MAR * I i Oand I i O / M A ~ . value may result in an inaccuracy. the axial stiffness ( K . and pk.0. the GAP has the elastic stiffness ( K t ) in the transverse direction until the lateral force exceeds the friction force and the slip starts t o occur. The value of TMAX = 1. the anisotropic coefficients of friction ( p y and p. or lit. TMAX should be determined such that the penetration is two or three orders less than the elastic deformation of the solid model. The parameter TRMIN is a fractional number (0. of the overall length. Notice that positive values of force and displacement represent compression in the GAP element coordinates. representing a lower bound for the allowable penetration. The penalty values are decreased if the penetration is below the lower bound. > & The maximum adjustment ratio MAR ( 1 < MAR < lo6. and subincremental process). The penalty value adjustment will thrash if TMAX is too small.. The recommended range of the transverse stiffness value is O. Directions y and z are defined by the orientation vector in the CGAP entry. p1 and p.0 (default option). There are two kinds 0. If K t = 0. Penalty values (I<. Ii.0. gap induced bisection. defaulted t o 0. respectively. e. The preload is defined by Fo (defaulted t o 0). When the GAP is open. while a much smaller K . there is no transverse stiffness.! represent static and kinetic coefficients of friction p. When the GAP is open. penalty values will not be adjusted but other adaptive features will be active (the gap induced stiffness update. and no adjustment will occur if TMAX is too large. The factor may be reduced t o facilitate convergence a t the expense of the accuracy. ) has a large value relative t o the adjacent structure. . The lower bound for the allowable penetration is computed by TRMIN * TMAX. The transverse shear stiffness Iit (defaulted t o p l K a ) is used when the friction is active upon contact. If the GAP is closed (UA . A much larger K . The recommended allowable penetration TMAX is about 10% of the element thickness for plates and shells (or equivalent thickness for other elements) which are connected t o the GAP element. and l i t ) are adjusted adaptively if TMAX > 0. If the structure is a massive solid.0 selects the nonadaptive (old) GAP element. < When the adaptive GAP (default option) is used. The initial estimate of TMAX could be made such that the penetration is inconspicuous relative t o the overall dimension of the structure.which is defaulted t o 10*I<. value may slow down convergence or cause divergence.
the magnitude of the slip center translation (SLIPV.0 AXIALU .D I S P L A C E M E N T S IN E L E M SYSTTOTALV TOTALW 0. since pk < p. SLIPW) should be smaller than the magnitude of the total displacement. Slipping does not occur as long as the external force exerted in the lateral direction (FT) is smaller than the friction force. the reaction force due t o friction is proportional to the normal force (FN).0 SLIPV 2. When the GAP is open (regardless of friction) or sliding without friction. TOTALW (representing total lateral displacements) and SLIPV. FORCE IN NEWTON SEPTEMBER 5.0 STATUS SLIDE COMPX 1.00000E01 S T R E S S E S ( F O R C E S ) IN G A P E L E M E N T S ( C G A P ) ELEMENT ID 31 . Positive values of the axial force (Fx) and'the axial displacement ( U ) represent compression. Lateral displacements in the y and z directions are printed under TOTALV. SLIPW) have the same value.4 Theoretical Basis for Frictional Behavior According to the Coulomb friction law. STICK or SLIP.GAP element output may be obtained by the STRESS ouput request in the Case Control Section. TOTALW) and the slip center displacements (SLIPV. 4. the total displacements (TOTALV. Slipping starts when the external lateral force exceeds the static friction force.F O R C E S IN E L E M SYSTSHEARY 0. is the static coefficient of friction.3891301 SLIPW 0. i.0 SHEARZ 0. An example of the GAP output format is shown below: ELECTROMECHANICAL CLUTCH DISPLACEMENT IN MM. The last column of the output shows the GAP status (available in V67): OPEN.2977E02 2.3.3891301 The element forces and displacements are computed in the element coordinate system.. In order t o simulate the stickslip behavior of the current GAP element. 1990 MSCINASTRAN 8/30/90 PAGE 66 SUBCASE 1 LOAD S T E P = 2.e. and printed as SHEARY and SHEARZ. the frictional yield function is formed as follows: with . Lateral forces are computed for the frictional case. SLIDE.. in general. The friction force is reduced t o the kinetic friction force ( p k F N ) during the slip. For the sticking or slipping cases with friction. SLIPW (representing translation of the slip center).2051E06 1. where p.
which can be represented by a circular cone in the element force space.11) and (4. (4.14) and (4. is kept uptodate by updating slip forces and the slip center. the lateral displacement can be decomposed into two parts: recoverable sticking part (denoted by superscript st) and irrecoverable slipping part (denoted by superscript s l ) . and (v.3.. (4. we obtain where Substituting Eq. we can express nyS' below: as Combining Eq. where (Fy.11]. The GAP is sticking. (4.. if f < 0.3.15): .3.3.10).13) into Eq. i. (4. Adopting the associated flow rule.14)) the final form of Eq.3.12) into (4. (4. Slipping is similar to the plastic deformation [4. (4.12)) (4. When the GAP starts to slip (f 2 O).where Kt is a penalty value (representing an elastic stiffness) during sticking. w.13) and (4. Substituting Eqs.3. represent the forces in the normal direction t o the slip locus.3.9) with f = 0.3.3. The slip yield surface.) represents a translation of the slip center in the yz plane.3.13) becomes The slip center must be updated accordingly using Eq.e.3.) Notice that slipping occurs along the direction normal to the slip locus. the slip displacements can be defined in terms of the effective slip increment (AyS') by where and F.
open to stick. Bisect the load or time step if the GAP undergoes two state changes (e. The scale factor . ) which can be rearranged to solve for + + P K ~ A V+ ~ + ~ ) (F: l i ~ ~ w ) (4. The frictional yield function should be satisfied (f = 0) at the stick/slip boundary. The first step is t o find the open/close boundary. When the friction is present. resulting in a quadratic function of P: ap2 +2b/?+ c = 0 (4. the status of the GAP element may change from open t o stick and further to slip.3. i. open t o slip) in a single step.B 5 1)is used t o determine the point where the GAP begins to slip. Let the displacement increment for the ith iteration be defined as follows: where n represents the previously converged step. In order to achieve the adaptability.g.. and vice versa.. or slip to stick). 210 u= if U.3. 5 uo uZ (open + closed) < 0 if Un L UO (closed + closed) where uo represents the initial opening of the GAP element. Assume that the GAP changes status from stick to slip a t the ith iteration.e. Assume that the GAP closes from the opening state at the ith iteration. it is necessary t o use the subincremental approach within a global increment. General principles adopted in the adaptive stiffness matrix update and bisection strategies induced by GAP elements are: Allow only one state change in any GAP element within each increment. the GAP element may change its status from open to sliding and vice versa.19) .5 Subincremental Algorithm for Stiffness Update and Bisection In the absence of friction..3. slipping with friction is a pathdependent process and the subincremental scheme allows a more accurate solution t o the pathdependent process. Two major steps are required to implement the subincrement method.B (0 5 .18) ~ P.g. The scale factor a (0 5 cr 5 1) is used t o determine the point at which the GAP starts t o close as follows: . Update the stiffness matrix when the GAP changes its status (e.4. p s ( ~ z P K ~ A U = J(F.11]. Furthermore. Subdivide the increment for the slip process [4. The second step is t o find the stick/slip boundary.
the following inequalities can be obtained: ac 2 a+2b+c<0 and b2ac>(l) 2 0 These inequalities lead to a proper root and for a  0 Once the initial slip point is found.3.where Fz. and F. i. and the the slip force ( F . The subincremental process is schematically described in Fig. the element stiffness and the slip center are updated. It can be shown that the quadratic equation is guaranteed to produce a real and positive root. It is noted that the "sliptoslip" case is catagorized into two groups. F. . The number of subdivisions are determined adaptively by The displacement increment for each subincrement becomes The GAP normal force (F. and FF represent GAP internal forces at the last converged solution. the slip region is subdivided into m subincrements (1 < m 5 10). Since the GAP is slipping at the ith iteration.3.20) using the relation in Eq..4. The condition for the slip in the same direction can be determined by the dot product of the last converged friction force and the friction force increment.3. Expanding Eq. One is the slip in the same direction and the other is the slip in the opposite direction.e. After the subincremental process. (4.) is updated in m subincrements. 4.19).) is gradually increased. (4.
If the slip occurs in the opposite direction. for final position Figure 4.Slipping in the same direction is also signified by a double root of /? at zero. It is highly desirable to reduce the penalty values in such cases to improve the numerical performance.4 Schema for Subincremental Process (m = 3) 4.3. /? has two roots with one being zero. On the other hand. The adaptive penalty value adjustment is designed to adjust penalty values automatically.3. if necessary. If the penalty value is too small and produce an appreciable penetration. numerical difficultiesarise if the penalty values are set too high. the root at zero is discarded and the subincrement is processed from the slip locus a t the opposite side signified by nonzero P. the accuracy of the solution may not be acceptable.6 Adaptive Adjustment of Penalty Values Penalty values are adaptively adjusted upon convergence for the next increment. using the userspecified initial values and the maximum allowable penetration . In this case. Then the penalty value (K.) should be increased to reduce the penetration.
.. Assume that the lateral displacement under the sticking condition is E when the GAP starts t o slip. the magnitude of shear stiffness K t has t o be related t o Ii. starting from the userspecified initial values.and K t are adjusted independently. (4.3. t o K t constitutes the slip criterion. The slip condition requires that If it is desired t o limit the lateral displacement under the sticking condition (6) by the same magnitude as the normal penetration (u . IiO MAR  < I( < MAR * K O where KO is the userspecified value for Ii.uo).. i.e. and Iit)are increased. The stepping function is designed such that the adjustment ratio is in powers of 10.. and K t ) are adjusted by the same ratio. penalty values (Ii... the penalty value is fixed and the adjustment algorithm is not invoked.23) If the user inputs TMAX = 0. The global stiffness matrix is updated whenever the penalty values are adjusted. and vice versa if the penetration is smaller than the minimum allowable value. I. i. an incompatibility will occur in the constitutive relations between two consecutive solutions if I<.1.e.uo < TMAX * TRMIN TMAX * TRMIN where TRMIN represents the userspecified parameter t o define a lower bound for allowable penetration. It is necessary t o introduce a stepping function in order t o avoid thrashing [4. and n denotes the incremental step with R=U  ug TMAX U  if u  uo > TMAX U o if u . by K t 2: p k K a . The proper ratio between two penalty values can be established logically. .12].. a fraction of TMAX. 10. Since the ratio of Ii. The penalty value adjustment is limited by an upper and lower bound defined by the userspecified maximum adjustment ratio (MAR). 0.. . or Ii.. If the GAP penetration exceeds the allowable value supplied by the user. where f ( R ) is . The adjustments are made a t converged points.(TMAX). R = It is noted that two penalty values (Ii.
3. If the GAP is sticking. the friction forces become with the associated stiffness as follows: In the case of slipping. are coefficients of friction in the y and z direction. When the GAP is in contact (closed).7 Nonadaptive GAP Element with Anisotropic Friction The original GAP element can have an anisotropic friction by specifying directiondependent friction coefficients associated with the GAP orientation. the lateral forces are given by The associated GAP stiffness is given by where the symmetry is enforced. . and p.4. the compressive force in the axial direction is defined by where Fo is the preload and u is the initial GAP opening. The slip yield function may be o modified for anisotropic friction as follows: where p.
The platform is released.3.4.8 Verification and Validation: Bouncing Mass Problem A 4lb weight is resting on a springsupported platform which is initially compressed 4 inches in the position shown in Fig. = mgh = 8 lbin 1 from which the velocity of the mass a t the time of departure from the platform can be obtained Up t o this point. and xo = 0 in/sec. 4. T h e change in energy from the initial state when the platform is stopped is AV.5.3. Sprl ng k1OlblIn. and after moving 2 inches.xi) = 60 lbin 2 AV. the platform is halted by a stopper. = k(x. Figure 4. the motion of the mass is governed by the equation With the initial conditions xo = 4 in.13]. The spring modulus of the platform support is 10 lb/in. The response of the weight is t o be determined [4. the response of the mass is expressed as .5 Physical Model From the conservation of energy AV. + A T = 0. + AV.3. .
= 2 = 13. Elasped time when the mass reaches the peak is obtained as 6 t2 = tl + . The theory predicts that one full cycle will take 0. the motion of the mass is described by h = xot .5905 sec.2595 = 0.0 in 29 ' X from the stopper.= 0. tl = 0 Wn ( +) Tip xo + = 0..6 Symbolic MSCINASTRAN Model .e. As the mars departs from the platform. = = 31.1 2 (4.035'7 ser where w.3. " 9 The height of the mass at this moment is h.. i.2953 sec.0644 radlsec.34) relative t o the stopper.0357 + 0.3.The elapsed time at the moment the platform hits the stopper may be obtained by solving the above equation for t with xl = 2 in. I GAP (22) I Load t o s l m l a t e f n l t l a l preload Plrtfotn COWRR) (20) S P ng ~ Gnvlty 0 ~ r Points ~ d Figure 4.
7. Small viscous damping (0.1. T h e automatic time stepping algorithm caused the reduction of At t o a half of the specified value (0. A small mass CONM2 24 (0. for which the stiffness is given as k = E (Young's modulus) = 100 psi.0 in2.7 Displacement Response . The base (GRID 200) is fixed t o the ground.3. t o which the platform is connected by GAP 23. Static loads are converted to the dynamic loads using Bulk Data LSEQ and TLOAD2 with a Case Control command LOADSET.) is introduced and adaptive time stepping is used.) of GAP 22 was adjusted to lo5.5 % a t 31 radlsec. the automatic time stepping algorithm quadrupled the time step size. T h e bisection was also activated by the adaptive time stepping during the initial phase. T h e analysis started with initial At = 0. T h e input d a t a listing is given in Table 4. and 1 (length) = 10 in.6. G A P elements had the initial stiffness of l o 4 lb/in for the closed position with TMAX of 0. t = 0. the penalty value (K. An initial displacement (4 inches) is simmulated by a step force (40 lbs) applied a t GRID 201.3. The stiffness matrix was updated three times due t o G A P status changes in the vicinity of t = 0. and t = 0. 4.0025 sec.035 sec.) while the platform was in motion around t = 0. After the initial settlements. The gravity field is applied by a GRAV entry. for the duration of 1 sec. At the onset. Figure 4.58 sec.00125 sec. The solution t o the MSCINASTRAN model is obtained by the nonlinear transient analysis method (SOL 99). The spring is modeled with A (area) = 1. t i m e in sec.614 sec. which is connected t o the platform (GRID 201) by G A P 22. for adaptive adjustment. The stopper is represented by a fixed point (GRID 203).3.. The mass is modeled as CONM2 21.545 sec.1 % of the mass block) is provided at the platform t o suppress extraneous jittering. with a CONROD element.005 in.The MSCINASTRAN model is depicted schematically in Fig.
Notice that penalty values were accepted in a wide range ( l o 4 t o lo6) by the adaptive adjustment scheme.56 inches and the period of approximately 0. '. a parametric study is conducted with the penalty value (closed stiffness) of GAP 22.6 0 . ~ . 0.E 1 Time [sec Figure 4.3. Due t o the finite interval of the time increment. the penalty value was adjusted to the narrower range ( l o 5 t o lo6).3.005.The displacement response in Fig. .4 0. T h e value was varied from lo3 t o lo7 with TMAX = 0.8. the time stepping scheme can not be synchronized exactly with the GAP closing/opening. ~ ~ ~ . The artificial stiffness introduced to the GAP elements changes the dynamic characteristics slightly.17 .583 sec.001) was also tested t o examine its effect. v 9 # f 0 .3 . Indeed.7 shows a peak height of approximately 14. a a In order t o verify the adaptive penalty value adjustment algorithm. rum 0 l ' ' ' ' l ~ r ~ . Discrepancies between the MSC/NASTRAN solution and the analytical solution may be attributed to the following: a Some momentum is dissipated due to the dimping and the small mass introduced at the platform.3.8 Adaptive Penalty Value Adjustment o n G A P 22 4. 4. A tighter tolerance (TMAX = 0. The effectiveness of adaptive penalty value adjustment is shown in Fig. 4.
E+4 +PG3 5.Table 4.50 $ PRINT MATRIX TRAILERS DIAG SOL 99 CEND TITLE = GAP VERIFICATION SUBTITLE = NONLINEAR TRANSIENT ANALYSIS SEALL = ALL ECHO = UNSORT SET 1 = 201.3 PGAP 3 1. LSEQ 20 301 202 TLOADI 210 301 0 220 TABLED1 220 +TAB1 0. 0.001 1. . 1. ENDT .3. LOAD 202 I. 1. AND ITERATIONS 0. 1.V66 $ SHL 121190 $ CPU MINUTES FOR VAX TIME 5 8. 0.E+4 +PG4 5. 200 I.0025 I ADAPT TSTEPNL 21 200 1 +TSI +TS2 0 $ PROPERTIES PARAM W4 31.3 PGAP 4 2. 201 0. .1 Input Data Listing for Bouncing Mass Problem ID VGAPM . I.23 SET 4 = 20 $ CONROD DISP = I STRESS = 3 $ FOR GAP FORCES ELFORCE = 4 LOADSET = 20 SUBCASE I DLOAD = 210 TSTEPNL = 21 SUBCASE 2 DLOAD = 210 TSTEPNL = 21 OUTPUT(XYPLOT) CSCALE = 1.3 PLOTTER = NAST XTITLE = TIME IN SEC YTITLE = DISP IN INCH XYPLOT DISP RESP/201(TI) . 100.3 $ LOADING FORCE 201 201 40.202(TI) BEGIN BULK $ SOLUTION STRATEGY . GRAV 200 386.202 SET 3 = 22. MAT1 I 100. 0. I.
$ BOUNDARY CONDITIONS G DE RST $ GEOMETRY GI RD 200 6. for for u > 0.3. GI RD 202 4.9 Physical Model for Vibration with Coulomb Damping [4. The solution while u < 0 is .9. 4.3. GI RD 201 4. u < 0.1 6. GI RD 203 $ CONNECTIVITY CONROD 20 200 201 CN 2 OM 24 20 1 CN 2 OM 21 202 CGAP 22 3 201 CGAP 23 4 201 ENDDATA ' 4.14] Suppose the block is displaced by uo >> Flk and released. (4.3.9 Verification and Validation: Vibration with Coulomb Damping A mass block (m) connected to a spring (k) is resting on a frictional surface as shown in Fig.3.35) Figure 4.141. The motion of the mass block is governed by mu+ku=F mii+ku= F where F is the friction force 14.
because of the shear stiffness ( k t ) in the GAP element. . which apply the preloads (gravity and a horizontal force of P=220 N). The first two subcases use PARAM. The analysis is performed in three subcases using SOL 99. the solution is m. A rod element is used t o simulate an elastic spring with k = lo3 N/m.623 radlsec. T 5 wt < 27r while It is noted that the acceleration ( u )is discontinuous at t = 5 . For u > 0. = lo6. 5 a. the amplitude decreases t o After two and a half cycles. resulting in an initial displacement The third subcase uses PARAM. and u = (uo . 1 1 F 1. TSTATIC. = . Ideally. 1 for static analyses.3. %.3 and pk = 0.4 F l k ) a t t = %.10. which gives rise t o a normal force (FN) of 100 N. p. / v m = 31. %. .where w = The mass block comes t o a stop ( u = 0) at w t = n. 4. . = 0. the mass block should come t o a stop with a residual displacement due t o friction. while the velocity is continuous. From the observation that u = (uo . 0 for dynamic response upon release of the load P. The MSCINASTRAN model uses a GAP element t o simulate friction with k. the mass block is stagnated. Let m (CONM2) be 100 kg in a gravity field of lm/sec2. the decay is a linear decrement at the rate of 4 F l k per cycle until I u.2.. kt = l o 5 .2 F I k ) at t = 2 .987 sec.3.2. TSTATIC. However. The natural frequency (w) of the system is 3. This effect is shown as ripples after two and a half cycles in Fig. The input data listing is given in Table 4. the mass block oscillates a t w. After the first cycle.1623 rad/sec and the period is 1.
governed by p. First.10 Horizontal Displacement Response In order t o examine the effects of the frictional force. 4. the normal force ( F N )is increased to 366.45 and pk = 0. The mass will come to rest when u = 0. the mass (m) is increased to 366. in turn. Despite the excessive steps. this is analogous to the overdamped case in viscous damping. It is noted that an excessive number of load steps was taken during the static loading in order to show the reduction in the frictional force at the stick/slip transition.3. the peak at 165 N is not shown due to the discrete load increment size.3. .3) and the horizontal applied force (P) is increased to 300 N.12.03 from Eq.19 m. 4. As shown in Fig.7 N. In this case. the initial displacement (uo) becomes 0..3. complying with pk. (4. the GAP will be sticking up to the friction force of 165 N. = 0.Figure 4. This is analogous to the criticallydamped case in viscous damping.11.36). the analysis is performed with some variations in the model. As the block starts to slip. the decay rate is such is that the amplitude is reduced to zero in a half cycle (wt = T ) as shown in Fig.9 sec.3. The residual displacement should have an asymptotic value of 0. I the friction coefficients are f also increased ( p .7 kg and. 4. the friction force is reduced to 110 N.3. Then the initial displacement (uO)is decreased to 0.11.1467 and the natural frequency (a) reduced to 1.6514 rad/sec. which occurs at t = $ = 1. For the overdamped case. The effect of distinctive friction coefficients (static and kinetic) during the static loading is shown in Fig.
.
12(Ti) .Table 4. 22 SET 2 = 10. 198 ECHO = BOTH SEALL = ALL SET I = 2.2 Input Data Listing for Coulomb Damping Problem ID COULOMB. P.3. YTITLE = DISPLACEMENT AT THE MASS BLOCK XYPLOT DISP RESP /2(TI).220(2) YTITLE = SHEAR FORCE IN GAP XYPLOT STRESS RESP 30/10(3) . 120. XMAX=8.22(TI) YTITLE = VELOCITY AT THE MASS BLOCK XYPLOT VELO RESP /2(TI) . 120(2) .5 XTITLE = TIME (SEC) XGRID LINES = YES YGRID LINES = YES XMIN=2.50 $ PRINT MATRIX TRAILERS AND ITERATIONS CEND TITLE = COULOMB FRICTION VIBRATION PROBLEM SUBTITLE = NONLINEAR STATIC AND TRANSIENT ANALYSIS LABEL = TIMOSHENKO. 210 SET 3 = 20. 12. 110. YOUNG.12(Ti) YTITLE = ACCELERATION AT THE MASS BLOCK XYPLOT ACCE RESP /2(Ti) YTITLE = STRESS IN ROD XYPLOT STRESS RESP 30/20(2). AND WEAVER. 220 DISP = 1 VELO = I ACCE = I STRES = 2 FORCE = 3 LOADSET = 1000 SUBCASE 10 $ COMPRESS GAP PARAM TSTATIC 1 DLOAD = 100 TSTEPNL = 10 SUBCASE 20 $ SHEAR GAP PARAM TSTATIC I DLOAD = 200 TSTEPNL = 20 SUBCASE 30 $ DYNAMIC ANALYSIS PARAM TSTATIC 0 DLOAD = I00 TSTEPNL = 30 OUTPUT(XYPLOT) CSCALE = 1. V67 $ SHL 122190 $ CPU MINUTES FOR VAX TIME 5 SOL 99 $ NONLINEAR TRANSIENT ANALYSIS DIAG 8.
100. 100.+5 123456 3456 123456 .+6 1. 110 12 13 I. 0.+6 1. 210 22 23 21 22 220 I. $ LSEQ 1000 211 . I. 0. 0.+5 0. 2 0. 1.3 . 0. 1.210(8) BEGIN BULK $ SOLUTION CONTROL TSTEPNL 10 4 . I 2 20 1.7 0. 0.+5 0. 10 2 3 1. 0.2 $ $ CONNECTIVITY FOR OVERDAMPED CASE 0. 11 12 120 I. 0. 100. 100.3 . 0. 0. 0. 123456 100.3 $ $ 110 110 210 111 I. 1.04 I ADAPT TSTEPNL 30 30 . 0. 12 366. 1. 1. 22 366.45 .25 I ADAPT TSTEPNL 20 25 .2 I ADAPT $$ GEOMETRY AND I GRID GRID 2 GRID 3 10 CGAP CONROD 20 CONM2 30 $ PROPERTIES PGAP 10 MAT1 20 $ 20 20 $ CONNECTIVITY FOR OSCILLATORY SYSTEM 0. 0.+5 . 0. 100.+5 123456 3456 123456 .7 0. 0. 0. 0. 1. 100.2 $ $ GEOMETRY AND I1 GRID 12 GRID 13 GRID 110 CGAP CONROD 120 CONM2 130 $ PROPERTIES PGAP 110 MAT1 120 $ GEOMETRY AND GRID 21 GRID 22 GRID 23 CGAP 210 CONROD 220 CONM2 230 $ PROPERTIES PGAP 210 MAT1 220 $ LOADING LSEQ 1000 GRAV Ill TLOAD2 100 CONNECTIVITY FOR CRITICALLY DAMPED CASE 0. IlO(8) . 3456 123456 1. 0.YTITLE = TOTALV DISP IN GAP XYPLOT STRESS RESP 30/10(6) YTITLE = SLIPV DISP IN GAP XYPLOT STRESS RESP 30/10(8). I. 100. 0 1. 1.+6 1. 0.+5 0.
I. 220. and rolling wheels. 0. I. 0. 0. I. Stresses at the contact area depend on the deformation of the bodies in contact. gears. 0. 300.3. 2000 0.3. 100 0 0. 4. 0. cams. Hertz presented solutions to a family of twobody contact problems with general curved surfaces for the loading normal to the surface. Hertz's analysis revealed that the pressure distribution on the contact surface between two curved bodies is represented by a semiellipsoid constructed on the surface of contact.FORCE FORCE FORCE DLOAD TLOADI TABLED1 tTBD2I ENDDATA 211 211 211 200 2000 201 0. 220. 2 12 22 1. 1. The hemispherical pressure distribution on the surface of contact with a radius a gives the maximum pressure (qo) at the center by . Hertz and published in 1881.13 Contact Between Elastic Sphere and Rigid Plane Let us confine our discussion t o the contact of an elastic sphere with a rigid plane as shown in Fig. 20 1 2.3.10 Hertzian Problem: Contact Between Sphere and Rigid Plane Mechanics of the frictionless contact between two elastic bodies were first solved by H. 1. ENDT 4. 1. 210 0. I. Typical examples are ball and roller bearings. z 1 Figure 4.13.
v. the circumferential stress (as) is compressive with the same magnitude as a. The solution for two balls in contact with radii R1 and R2 is given [4. respectively. which represents a pure shear stress state. Stresses at the center of the contact surface are given by aT= = (1 1 2 + 2v)o.v. k2 = 7rE2 for two balls (1 and 2) with different Young's moduli and Poisson's ratios.3.39) and (4. given as At this boundary.where P is the total compressive force.15] with the radius of contact area as follows: and for the rigid body motion (or approach distance) with 1 . Uniform vertical displacement of the symmetric plane is maintained by tying the vertical displacements of all the nodes .3. 4. given as TTz = ~0~ = 0. and = 40 The maximum shear stress occurs on the zaxis at a distance of 0 . For the contact of an elastic sphere with a rigid flat plane.40) are reduced with R2 = cc and E2 = cc to and where cr is identical to the displacement of points on the sphere remote from the deformed area.. k1 = . (4.xE1 and 1 .31q0 The maximum tensile stress occurs at the circular contact boundary. 4 7 ~ below the surface of contact. For the static analysis.3. Eqs. Contact with a rigid plane is simulated by adaptive GAP elements.14. a symmetric half of a 10 degree sector of the sphere is modeled by HEXA and PENTA elements with axisymmetric boundary conditions as shown in Fig.
The analysis ran t o completion in 27 iterations with four GAP induced stiffness updates. All cases ran to completion without requiring any user intervention.on which 1/36 of the total compressive force is applied.3.3. It is noted that the frictional case appears more accurate at the initial stage. The deformed shape at the maximum loading is shown in Fig. E = 1000 psi.15. which shows the MSC/NASTRAN solution in comparison with the Hertz solution.3 Figure 4. 4.3 and pk = 0. This is because the frictional case took much smaller increments due to bisections during the initial phase. Notice that the effects of friction and geometric nonlinearity are gradually manifested as the load increases.on the top plane (loading plane) by MPC relations to the one at the centerline (zaxis).2).15 in comparison with the undeformed model. 4. = 0. 4. however.14 MSC/NASTRAN Model for Static Analysis The static analysis was performed in 20 increments up to the maximum loading (P = 2160).3. Two additional cases were analyzed: one with geometric nonlinearity (PARAM. In the frictional case. and the point of contact could be more closely predicted. and 30 incremental steps were required.0 in. . R = 8. Deformed shapes for other cases were not discernible from the one in Fig. v = 0.3. The radius of contact surface and the approach distance (vertical displacement at the loading plane) are plotted as functions of jhe total force (P) in Fig.16.LGDISP) and the other with friction (p. bisections were activated three times during the initial phase.
.3.. Total Force in Static Analysis 4......"' .... ......l i ~ * 1 ~I ~ m 0 I t I 588 1000 1500 2008 2508 Load [ l b u Figure 4.... ......."*...16 Contact Radius & Approach vs.81.28 ...15 Undeformed Shape vs.. 1 ' ' ' 1 ' ' ' ' ..."' ....m...3......Figure 4....3 ......... approach dlstance ( H e r t z ' MSC/NRSTRQN so l u t i o n MSC/NQSTRQN ( w / f r 1c t i o n MSC/NQSTRQN + ' Iw/LG@ISP) C .. Deformed Shape at P = 2160 CU  TCI ~ n ' 0 X c o n t a c t r o d jus (Hprt7) ...
The duration of impact is given by For the impact analysis. In the meantime the GAP induced . respectively.3. The adaptive time stepping algorithm adjusted At t o 2 and 4 msec. at the maximum compression can be found by substituting di = 0 into Eq. is specified for the sphere a t the beginning of the impact. Contact with a rigid plane is simulated by adaptive GAP elements. The applied force P can be obtained from where m and v are the mass and the velocity of the sphere. a 10 degree sector of the sphere is modeled by HEXA and PENTA elements with axisymmetric boundary conditions as shown in Fig. Vibrations and the stress wave can therefore be neglected.4. The analysis started with the initial time step size of 1 msec. In the impact of an elastic sphere with the rigid plane.3. (4. The initial velocity of 10 injsec.41).. The value of the maximum compressive force is obtained by where a.11 Dynamic Case: Impact of a Sphere with Rigid Plane The Hertzian problem shown in Fig.17. the duration of contact is very long compared t o the period of lowest mode of vibration in the sphere.39a) and (4.13 is considered for dynamic response. 4.40a) are assumed to be valid during impact [4.3. resulting in where p is the mass density of the sphere..3.3. and the statical solutions in Eqs. during the initial stage. 4. and progressed to completion without any difficulty.3.25 msec. The bisection was activated during the maximum impact to adjust At to 0.15]. Introducing a notation the acceleration of approach becomes from which where vo is the velocity of the sphere at the beginning of impact. (4.
(4.18 indicates that the duration is around 0.051 sec.42) and (4. which occurs when the approach distance (a. The maximum indentation occurs in the vicinity of t = 0.44).3. Fig. MSC/NASTRAN anal.3. the time step size fluctuated between 0. 4.. and 4. with the contact radius of approximately 1.71 while Eq.5 lbs. v = 0.3. 4..18 and 4..072 sec.3 p = 0.39a) predicts 1. from Eq..0 in...) reaches 0..19.3.3. = 1440 and a. 0.stiffness updates were activated (a total of 15 times) when GAPs were closing.25 msec.3.01 lbsec2/in4 Figure 4.1544 sec.3.3. After the peak impact.5251.0 msec. = . the maximum compression force (P. R = 5. This thrashing has been remedied by introducing the numerical damping (PARAM.17 MSCINASTRAN Model for Dynamic Analysis According t o Eqs.2 sec. The duration of impact is 0. which seems t o have been caused by opening GAPs.) is 1246.62..20 shows snap shots of the deformed configuration at time 0. respectively. (4.NDAMP). Small discrepancies are attributed to the discrete contact rather than continuous contact due t o the discreteness in space and time. 4. E = 1000 psi. MSC/NASTRAN analysis as shown in Fig.53 as shown in Figs.072 sec.143 sec. ysis shows P. due to thrashing of bisection and automatic adjustment.43).3. and 0. (4. 0.
NONLINEAR I R A N S I E ~ Il M P A C l V E L O C l l V I N I H E 2 1 s 1 0 . Time .18 Reaction Force vs. Of A SPHERt ON A R I S I 0 P L A I P L I N E Figure 4.3.
19 Vertical Displacement Response a t Select Points o n Center Line .3.Curves in the order of Near contact point Center point on midplane Point at the top NONLINEAR I R A I S I E N I I M P A C l V E L O C L I V IN INE Z I S l a . OF A SPHERE ON A R I G 1 8 F L A l P L A I E Figure 4.
2. 0.051 .Clockwise from Upperleft corner: t = O .0.051. 0.2 sec. t = 0.072 and 0.072 Figure 4.20 Snap Shots at t=O..3. .0.
This guess has t o be a conservative one a t the expense of the computational efficiency. because the proper value varies throughout the analysis. which dictates the accuracy. is not the computing efficiency but the overall engineering efficiency. A limitation exists.16]. in concert with adaptive solution methods [4. that one of the contact surfaces should not rotate by a large angle because the GAP element orientation is not updated for large rotations. Penalty values should be chosen as large as possible for solution accuracy but as small as possible for solution efficiency.3.12 Recommendations The most crucial task in using the GAP element is t o determine proper penalty values. The adaptive GAP element implemented in Version 67.12. It is noted that there is an advantage in maintaining the original orientation of GAP elements when a contact with a fixed flat surface is simulated. By achieving this goal. however. Even with the advanced knowledge of the structural characteristics of the contact region. The recommended value for the penalty stiffness is 1000 times the stiffness of the adjacent structure. user specification of the penalty stiffness is merely an educated guess. A selfadaptive algorithm has been proven effective and efficient for nonlinar analysis in MSCJNASTRAN. makes it possible t o maintain an acceptable accuracy while maintaining the solution efficiency without user intervention throughout the dynamic environment of nonlinear computation. This is because the adaptability makes it possible t o obtain the solution a t the very first trial. efficiency and the effectiveness of the analysis.1%. The main goal of the selfadaptive method is t o relieve users from the chores and uncertainties of guessing or trialanderror searches.4. The primary virtue of the adaptive method. which will produce an error of 0. The applicability of the present algorithm is deemed viable t o simulation of surface contacts with friction. an optimum algorithm design is automatically attained for efficiency as well as for solution accuracy. . however.4.
. Finite strain formulation is required to treat the problems in this category. However. moderately large strains (20 to 30 %) appear in local areas if there is any large deformation. In inost structural applications.1]. and there is no distinct limit for large displacements because geometric nonlinear effects are related t o the boundary conditions as well as the dimensions of the structure. If the loaddeflection curve of the critical point can be estimated. whether rigid body rotations or deformation inc'. are selfevident. The inotion of a pendulum under gravity is caused by geometric (differential) stiffness. Large strain effects are pronounced in metal forming problems which could have strains exceeding 100%. geometric stiEening due to initial stress effect (as s result of large rotations). such as pressure load and thermal load applied on the surface that rotates. the loading point should be in the nonlinear portion of the curve. stiffness in a pendulum or snapthrough of an arch belong to this category. Centrifugal force is another example of follower forces. Geometric nonlinear effects should be significant if the deformed shape of the strccture appears distinctive from the original geometry by a visual inspection. MSCINASTRAN does not currently support the large strain capability.uced rotations. this definition for seems to be a little co~servative numerical analysis.MSC/NASTRAN can be used for that category of problems. Other geometric nonlinear effects are treated by updated element coordinates.2] in the structure involving large rotations.1 OVERVIEW AND USER INTERFACE Geometric nonlinearities are manifested in problems involving large rotations and large deformation. The characteristics are follower forces due to large rotations. Follower forces are manifested when the applied loads are displacement dependent.Chapter 5 GEOMETRIC NONLINEARITY 5. A more rigorous and quantitative definition for the large displacements can be derived from the plate theory of Kirchhoff and Love: the small deflection theory is valid for a maximum deflection of less :$an 20% of the plate thickness or 2% of the small span length [5. Stiffening of a membrare. and large strains due to large defarmation. Geometric nonlinear effects [5. however. gimbal angles .
e.1.1.1. MSC/NASTRAN has a distinct approach t o the large rotation. (5.] du D BLdv and . all the potentially nonlinear elements become actively nonlinear elements unlike the material nonlinear model. (5. The geometric nonlinearity is controlled by only one parameter (PARAM. Substituting Eq. (5.1).4) becomes d F = [ K L KR with KL=JVB.(or rotation vector).3) and do = D d~ = D B du. but without follower force effect With values of 1 or 2 for LGDISP. Upon differentiation of Eq. (5.T tangential matrix + + K. i. LGDISP) with the following values in SOL 66 and SOL 99: a 0 for geometrically linear analysis 1 for geometrically nonlinear analysis 2 for geometrically nonlinear.1. The equilibrium is sought in the deformed position.3]: F = The element matrix pTo dV (5.1. Consider the internal force computation as follows [5..5) Eq. for which the element coordinates are continuously updated to the current configuration during the iteration. and the differential stiffness [ K ~ ] . we have where {o) represents stresses with reference to the original coordinates.1) (B) defined from the strain definition as is in which [B]could be divided into two parts (linear and nonlinear).
) and nodal displacements { u ) (using an example of a 10noded tetrahedron).. a geometric stiffness dependent on the initial stress level.e. The definition of finite strains based on the Lagrangian formulation (referred t o the initial configuration) is as follows: with other components obtained similarly. where [A] = 6 x 9 a: 0 0 a: 0 T a. displacement derivatives are expressed by and . In matrix notation where { E L ) is the usual infinitesimal strain vector and { E N } is the nonlinear strain vector consisting of the second order terms. and K . K R a stiffness due t o large rotation. 0 a: 0 a: a$ 0 0 0 a: o a: a: b and Introducing shape functions (N. i.in which K L represents the usual linear stiffness matrix. Now it remains t o define the nonlinear part of the element matrix (BN).
da = in which /V d ~ u dV = . dV .where and Fkom the properties of matrices A and 0. it can be shown that from which can be derived as follows: The initial stress stiffness [I<o] K .
. Geometric stiffness matrix [Ii. and [MI with the following observations: [GI is dependent upon the initial geometry.] is equivalent t o the differential stiffness [Kd] in MSC/NASTRAN.]takes into account the effects of the initial stresses. This matrix is used in forming [KR] and [I'.]. is dependent on the rotations and should be updated [A] continuously.where with I being the (3x3) identity matrix. This effect becomes important with geometric stiffening. [MI is used in forming [I{. and is used for instability analysis. [GI. The geometric stiffness matrix [Ii. Finally the geometric stiffness is It has been found that stiffness matrices caused by geometric nonlinearity (I<R and Ii. due t o rigid body translation and rotation. hence stays constant unless the geometry is updated.]. [MI is dependent on the stresses and should be updated continuously. The primary functions of nonlinear stiffness matrices can be interpreted as follows: takes into account the effects of large rotations. are treated effectively in the absence of large strains by updating element coordinates in MSCINASTRAN. [A] is used in forming [I<R].) can be computed from the matrices [A]. The large displacement The matrix effects.
.
5.. . Consider a rod which underwent rigid body motion as well as deformation as shown below: Figure 5.1 Net Deformation of a Rod d The net displacement u is measured in the displaced element coordinate system by overlaying the original element on top of the deformed element. and the equilibrium is computed in the deformed configuration.1 Concept of Convective Coordinates The concept is based on the fact that the rigid body motion does not contribute to the strain energy and is eliminated from the internal force computation. i.2 UPDATED ELEMENT COORDINATES When the large displacement effect is included in the nonlinear analysis. MSC/NASTRAN employs a method of displaced element coordinate system. where the summation sign implies an assembly operation.e. and Tbd Tbg transformation and are matrices from displaced to basic and from global t o basic coordinate systems. Then the element forces should be transformed into the common coordinate system (namely global coordinate system denoted by a subscript g) before assembly for global operations.1) where the superscript e denotes an elemental operation and the subscript d denotes the vectors in the displaced element coordinate system. The element force can simply be computed by {FdIe = [KIe{ud) (5. This method allows large rotations by updating element coordinates to the deformed geometry. 5. respectively.2.2.2.
.2.. i. Figure 5. ze and transformation matrix [The] is composed of direction cosines of unit vectors of the element coordinate system with respect to the basic coordinate system. ye.2 Updated Coordinates and Net Deformation Referring to Fig. 5. >casi. Transformation from the element coordinate sytem t o basic coordinates is simply basic element basic where the position vector (Xiin Fig. respectively) with reference to the basic coordinate system (denoted by a subscript b).2.2) of the element coordinate system with respect to the basic coordinate system is denoted by < ze.2.2 Element Coordinates vs.2.e.2. 5. . Displaced Coordinates The element coordinate system is established by bisecting the diagonals of the quadrilateral. a quadrilateral element is shown in its original and deformed positions (denoted by subscripts e and d.5.
As the element deforms or displaces. {}{I}~{)~ (5. the element coordinate system moves and this is defined as a displaced coordinate system.2) and [Tbd] formed simliarly to [T*.2. Again the transformation should be performed similarly.3 Computation of Net Deformation The net displacements can be computed by substracting the original nodal coordinates in the element coordinate system from the displaced nodal coordinates in the displaced element coordinate system.e.2..2. is In order to isolate the deformation from the rigid body displacements. The displaced coordinate system is established in the same manner as the element coordinate system. nodal displacements are computed in the displaced element coordinate system by overlaying the original element as shown in Fig.2. i. 5.e.]. i.6) .. zd yd. element defor~ation Figure 5. { ) basic =lTbdI{} displaced +{::} basic is the position vector of the displaced element coordinate system with where < sd. respect t o the basic coordinate system ( ~in fFig. 5.3.
. Effectively. The internal forces are computed using net displacements and rotations.3 Provisions for Global Operation It is noted that the net rotations (8.in which the nodal coordinates in the element and displaced element coordinate systems can be computed by the following transformations: { i Ie and = [TbelT [{ i } { i! I b aSi ] basic  bnsic global basic Substitution of Eqs.. ud.. the second order effect . (5. if the material is linear or if nonlinear material is involved. the element forces have t o be transformed to the global coordinate system before assembly for equilibrium check. l 5.e.2. (5.2.2.:I = d s { u . (5.7) and (5. Subsequently. In the absence of the large displacement effect.2.i.6) results in where { u g ) is a total displacement (translational components only) in the global coordinates. the net displacement u d in Eq. By and 8.e. ud The update process is performed at every iteration and the updated nodal displacements are used whenever strains and stresses are computed. of each node associated with the shell and beam elements) are computed by a gimbal angle approach (or rotation vector approach) before computing element forces. i.9) is reduced to: {.2.8) into Eq. Consequently. the tangent stiffness matrix is formed in the global system by assembling the element stiffness matrices transformed into the global coordinate system from the displaced coordinate system.
This approach can be interpreted as approximate updated Lagrangian method.due to large rigid body motion is eliminated.. which is inherently different from the current method of updating the coordinate system. the transformation [Tbg] computed for each nodal point and it is not stored but recomputed whenever it is needed. are available from the data block BGPDT. . this method of displaced coordinate system is a unique and salient feature in MSCINASTRAN. Xb. The transformation matrix [Tbd]is computed for each element after each iteration and stored in the ESTNL data block for stiffness matrix update when required by the stiffness matrix update strategy. The referential geometry in the updated Lagrangian method is brought uptodate at every incremental step upon convergence but fixed during the iterative process. However.e. The nodal coordinates in the undeformed geometry. where the superscript i denotes operations on each nodal point. since the motion of the body follows Lagrangian description. However. Stresses are computed in the deformed geometry just like Cauchy stress. the displacement output shows the total displacements in the global coordinates. i. On the is other hand.
.
PENTA and TETRA) or shell elements (QUAD4 and TRIA3) defined by the GRID points and will change in magnitude if the area changes. where p is the magnitude of the pressure on the surface A . PLOAD. 5. They generally occur with fluid pressures such as the pressurized balloon. and centrifugal forces. Forces and moments defined by the location of one or more GRID points. .5. Other physical applications involve kinematics such as the classical fire hose instability problem or inertia loads on spinning bodies. includes the following Bulk Data inputs: FORCEl. TEMP.g. the MSCJNASTRAN static loads belong to one of two catagories. etc: Thermal loads are built into the element force calculations and will automatically follow the elements. follower force catagory. e. interpolated by a shape function N .1 Basic Definition For geometrical nonlinear analysis. the term applies to specific load inputs as defined below. Also note that upstream superelements are assumed t o be linear and therefore the upstream loads will remain fixed in magnitude and direction. The second.FORCE2.MOMENT2: direction changes with displacements The of the referenced GRID points. The magnitudes of these concentrated loads are constant.PLOADB. The effect may be destabilizing if large motions occur.3. inflated tire or the lift load on the airplane wing.. TEMPI. In MSCJNASTRAN. It is recommended that lumped masses be used with these loads.PLOAD4: The pressure loads follow the surface of the solid elements (HEXA.MOMENTl. namely: Forces and moments defined by fixed vector inputs. forces on omitted degrees bf freedom (when ASET or OMIT data are present) should not be follower forces. The first catagory includes simple forces. which may be calculated once per run and cannot change direction or magnitude.3 FOLLOWER FORCES The term "follower force" usually refers to the applied loads that change direction and magnitude with structural displacements and rotations. moments and enforced displacements. thermal loads. RFORCE: Centrifugal loads change in magnitude and direction with motion of the masses attached t o the GRID points. which include pressures. and iidA changes as a function of u. In addition.
3 Dynamic Follower Forces All of the follower forces listed above. However.{ ~ a ( u ~ ) } where the applied load vector {Pa)is now a variable.. becomes: (5. defined by Eq.3.3) using derivatives of the loads.2.2) {R:} = {pa(ui)} .3.2. it could become a major concern in thin shell models with pressure loads causing large rotations.3. Also the buckling solutions or modal analysis on preloaded structure with pressure load may not be correct due to the approximate tangent matrix if the effect of the follower matrix is significant. which are available in static nonlinear analysis (SOL 66 and SOL 106). 5.) = P(O) f (4 + where f (u) P ( u ) . primarily t o verify the effects of the follower forces.e. where the converged solutions will be correct but the rate of convergence may be slow or cause divergence.5. = Note that thermal effects are included in the vector { F } . (3. i. The third option was made available from Version 66. Corrective Loads are computed based on the updated geometry and added to the initially applied loads to account for the follower forces. P(.P(0).4. In turn. are also available in nonlinear transient analysis (SOL 99 and SOL 129) except for the following restrictions: . the tangent matrix could be calculated from Eq. 5. the nonlinear solution process in MSCINASTRAN ignores the stiffness effects of the changing loads and use the approximation: The effect of the approximation is minor in most cases. (3. However. The follower force effects in the analysis can be controlled by the parameter LGDISP.2). Three options are available in PARAM LGDISP: 0 for no geometric nonlinearity 1 for full geometric nonlinearity (including follower forces) 2 for deactivating follower forces while maintaining other geometric effects (rotation of element coordinates and differential stiffness).51.2 Implementation The follower forces depend on the GRID displacements and therefore must be recalculated for each nonlinear iteration and line search. The basic equation for residual error. which is termed follower matrix [5.
The effect of follower forces is a conspicuous feature in this problem. Thermal effects are not supported in the nonlinear transient analysis.2] is subjected to an increasing internal pressure. In the static analysis. Fig. The solution strategy is provided by default methods in the NLPARM Bulk Data entry in SOL 66 and by the TSTEPNL Bulk Data entry in SOL 99. where pi(u) represents the follower forces. g(t) is a forcing function in terms of time. 5. Table 5. The loading history and the numerical performance of below.Centrifugal forces (RFORCE) are not recalculated in SOL 99 or SOL 129. the pressure load of 5 psi is applied increments) with gradually increasing incremental pressure.e. in 5 subcases (total of 51 bisection is activated seven up t o 0.3. The times while the geometric effects are predominent during loading subcase). and N ( u ) ~ the applied load via NOLINi options. 5.1 shows one quarter of the crosssection of the elliptic cylinder consisting of a row of 12 QUAD4 elements.2 list the input data for SOL 66 and SOL 99.015 psi (in the first each subcase are tabulated . Dynamic loading with DAREA Bulk Data and initial conditions do not follow the structure and are treated as linear loads. with the plasticity modulus being 1%of the Young's modulus.4 Verification: Elliptic Cylinder Subject to an Internal Pressure A thinwalled cylinder having an elliptical shape [5. The TLOADl and TLOAD2 data define the time dependency and refer t o the DAREA IDS of the LSEQ data. Geometric nonlinearity effects are included in the analysis by specifying PARAM.3. i. The pressure loading is applied both statically (SOL 66) and dynamically (SOL 99) by PLOAD4 Bulk Data entries and the displacement responses are observed. up to a maximum of 5 psi. respectively. LGDISP.1 and 5. The material of the cylinder is elasticplastic obeying the von Mises yield criterion and the isotropic hardening rule. 1.3. The Load Set Identification number is deis fined on LSEQ Bulk Data. All the translational degreesoffreedom in the longitudinal direction are constrained t o represent a typical segment of the long cylinder.3.. The material properties are specified by the MAT1 and MATS1 Bulk Data entries. The applied load is a function of displacement (follower forces) as well as time.
2 shows the displacement responses at two points (denoted by A and B) on the minor and major axes of the ellipse. The vibration frequency increases as the internal pressure increases. bisections take place due to rapid changes in geometric effects. Fig. The automatic time step adjustment is activated 35 times during the first second of the analysis. and the dynamic deformation is dominated by the ovaling mode.3. mainly to recover from the bisection process by stretching the previously reduced time step size. 5. The static deformation turns the elliptical into a nearlycircular cylinder.3. The effect of follower forces is verified and validated by the fact that the static deformed shape is approaching a circular shape. of line searches 40 1 81 57 16 13 2 3 4 5 20 10 5. The static and dynamic deformed shapes (snap shots at the extreme magnitudes of the vibration) in the vicinity of 5 psi are shown in Fig.0 psi/sec. 5. of increments iterations I< updates bisections (psi) 0. but the amplitude diminishes due to the geometric stiffening. .3.1). It is noted that the plastic reloading occurs after the unloading while approaching the minimum peak displacements. of no. It is noted that the circular deformation pattern is formed at less than 0. The static response is plotted in the same figure. of no.subcase ID 1 Numerical Performance During Static Analysis no.O no. 5. and the geometric nonlinearity disappears beyond that point.0 8 5 8 656 143 126 53 60 174 26 26 7 0 0 8 7 0 0 The dynamic analysis is performed in five subcases of equal duration (1 sec) with an initial time step size of 10 msec. which shows good agreement between the static response and the mean dynamic amplitude. of load no.5 psi in static analysis. The bisection is again activated when the plastic deformation starts while approaching the peak (maximum and minimum) displacements.05 0. The bisection process is activated 37 times during the first second of the time history.3. In the beginning of the analysis.1 0.5 1. The plastic deformation occurs in the region near the small radius on the major axis of the ellipse (point B in Fig. The internal pressure of the cylinder is increasing linearly at the rate of 1.
Young's modulus 0. Poisson's ratio 50000 psi.t = E = Y = a = .1 Elliptic Cylinder Subject to an Internal Pressure . plasticity modulus 0.3. thickness 30 x lo6 psi.3 ..1 in. yield stress 3 x 10' psi. H = p = to 0.00283 lbsec2/in4. mass density =5 sec Figure 5.
I displacement of point A  dynamic response static response displacement of point B 0 1 Pressure (psi) Figure 5.2 Displacement Responses for Elliptic Cylinder .3.
3.3  7 .Major Axis Figure 5.3 Deformed Shapes of the Elliptic Cylinder 5.
XTITLE = TIME IN SEC YTITLE = DISP OF GRID 1000 AND 1120 XYPLOT DISP RESP/l000(Tl).3 PLOTTER NAST YMIN = 60.ll20(T2) .. KIM 2/87 ID SOL 66 $ NONLINEAR STATIC ANALYSIS $ CPU MINUTES TIME 30 DIAG 8.0 MAXI DEFO .I SET 1 PLOT STATIC 0 4 RANGE 4.50 $ PRINT MATRIX TRAILERS AND ITERATIONS CEND TITLE = ELLIPTIC CYLINDER UNDER INTERNAL PRESSURE SEALL = ALL ECHO = BOTH SET I = 1000. V67 $ APR 62591. SSH 4/20/88.I SET I PLOT STATIC 0 2 RANGE 2.0 MAXI DEFO . Y MAXI DEFO .I SET I PLOT STATIC 0 5 RANGE 5. O. SHL 3/9/88.I FIND SCALE ORIGIN I SET I PLOT STATIC 0 I RANGE 1.1 SOL 66 Input Data Listing for Elliptic Cylinder ELLIP66.I SET I OUTPUT(XYPLOT) CSCALE 1.. YMAX = 80.0 MAXI DEFO .0 MAXI DEFO .3. X .1120 DISP = I SPC = 200 SUBCASE I LOAD = 10 NLPARM = 100 SUBCASE 2 LOAD = 20 NLPARM = 200 SUBCASE 3 LOAD = 30 NLPARM = 300 SUBCASE 4 LOAD = 40 NLPARM = 400 SUBCASE 5 LOAD = 50 NLPARM = 500 OUTPUT(PLOT) CSCALE 1. 0. AXES Z .0 MAXI DEFO .Table 5.3 PLOTTER NAST SET I PLOTEL VIEW O.I SET I PLOT STATIC 0 3 RANGE 3.
BEGIN B L UK PRM AA LGDISP 1 $$ SOLUTION CONTROL NP R L A M 100 20 NP R L A M 200 10 NP R L A M 300 8 NP R L A M 400 5 NP R L A M 500 8 $$ LOADING PLOAD4 10 101 PLOAD4 20 101 PLOAD4 30 101 PLOAD4 40 101 PLOAD4 50 101 $$ G O E R E MTY GRID 1000 GRID 1001 GRID 1010 GRID 1011 GRID 1020 GRID 1021 GRID 1030 GRID 1031 GRID 1040 GRID 1041 GRID 1050 GRID 1051 GRID 1060 GRID 1061 GRID 1070 GRID 1071 GRID 1080 GRID 1081 GRID 1090 GRID 1091 GRID 1100 GRID 1101 GRID 1110 GRID 1111 GRID 1120 GRID 1121 GRID 2000 GRID 3000 GRID 4000 $$ CONNECTIVITY CQUAD4 101 100 CQUAD4 102 100 CQUAD4 103 100 CQUAD4 104 100 AT UO AT UO AT UO AT UO AT UO TR HU TR HU TR HU TR HU TR HU YES YES YES YES YES 112 112 112 112 112 .
I 16 26 1000 1010 1020 1030 1040 1050 1060 1070 1080 1090 1100 ill0 2000 2000 ENDDATA .CQUAD4 105 CQUAD4 106 CQUAD4 107 CQUAD4 108 CQUAD4 109 CQUAD4 110 CQUAD4 Ill CQUAD4 112 $$ PROPERTIES PSHELL 100 MAT1 110 MATS1 110 $$ CONSTRAINTS SPCl 200 SPCI 200 $$ FOR PLOT PLOTEL 1000 PLOTEL 1001 PLOTEL 1002 PLOTEL 1003 PLOTEL 1004 PLOTEL 1005 PLOTEL 1006 PLOTEL 1007 PLOTEL 1008 PLOTEL 1009 PLOTEL 1010 PLOTEL 1011 PLOTEL 1012 PLOTEL 1013 $ 100 100 100 100 100 100 100 100 110 3.+5 .+7 110 .3 PLASTIC 3 .
11 . 0. X . V67 SOL 99 $ NONLINEAR TRANSIENT ANALYSIS $ CPU MINUTES TIME 200 $ PRINT MATRIX TRAILERS AND ITERATIONS DIAG 8. SHL 3/9/88.I FIND SCALE ORIGIN 1 SET I PLOT TRANS 0 TIME 4.50 CEND TITLE = ELLIPTIC CYLINDER UNDER INTERNAL PRESSURE SUBTITLE = PROPORTIONALLY LOADED UP TO 5 PSI IN 5 SECONDS SEALL = ALL ECHO = BOTH SET I = 1000.Ol TSTEPNL 100 5.3 .0 MAXI DEFO . Y MAXI DEFO . O. SSH 4/28/88.3 PLOTTER NAST SET 1 PLOTEL VIEW O.I SET I OUTPUT(XYPLOT) CSCALE 1.3 PLOTTER NAST XTITLE = TIME IN SEC YTITLE = DISP OF GRID 1000 AND 1120 XYPLOT DISP RESP/IOOO(Tl)..2 SOL 99 Input Data Listing for Elliptic Cylinder $ APR 62591.Table 5. AXES Z . KIOOK KIM 2/87 ID ELLIP99.8 5.3. 1120(T2) BEGIN BULK PARAM LGDISP 1 $$ SOLUTION CONTROL ADAPT 100 .1120 DISP = I SPC = 200 LOADSET = 1000 SUBCASE I DLOAD = I 0 0 TSTEPNL = 100 SUBCASE 2 DLOAD = I00 TSTEPNL = 200 SUBCASE 3 DLOAD = 100 TSTEPNL = 300 SUBCASE 4 DLOAD = 100 TSTEPNL = 400 SUBCASE 5 DLOAD = 100 TSTEPNL = 500 OUTPUT(PL0T) CSCALE 1..
I. ED NT . $$ G O E R E MTY GRID 1000 GRID 1001 GRID 1010 GRID 1011 GRID 1020 GRID 1021 GRID 1030 GRID 1031 GRID 1040 GRID 1041 GRID 1050 GRID 1051 GRID 1060 GRID 1061 GRID 1070 GRID 1071 GRID 1080 GRID 1081 GRID 1090 GRID 1091 GRID 1100 GRID 1101 GRID I110 GRID 1111 GRID 1120 GRID 1121 GRID 2000 GRID 3000 GRID 4000 $$ CONNECTIVITY CqUAD4 101 100 CQUAD4 102 100 CqUAD4 103 100 CqUAD4 104 100 CQUAD4 105 100 CqUAD4 106 100 CQUAD4 107 100 CqUAD4 108 100 A AT DP A AT DP A AT DP ADAPT TR HU 112 +TBDI 5. 0.TSTEPNL 200 100 TSTEPNL 300 100 TSTEPNL 400 100 TSTEPNL 500 100 $$ LOADING LSEq 1000 900 PLOAD4 110 101 TLOADI 100 900 TABLED1 120 +TBDI 0.
3 PLASTIC 3 .1 110 .+5 .CqUAD4 109 CqUAD4 110 CqUAD4 111 CqUAD4 112 $$ PROPERTIES PSHELL 100 MAT1 110 MATS1 110 $$ CONSTRAINTS SPCI 200 SPCi 200 $$ FOR PLOT PLOTEL 1000 PLOTEL 1001 PLOTEL 1002 PLOTEL 1003 PLOTEL 1004 PLOTEL 1005 PLOTEL 1006 PLOTEL 1007 PLOTEL 1008 PLOTEL 1009 PLOTEL 1010 PLOTEL 1011 PLOTEL 1012 PLOTEL 1013 $ 100 100 100 100 110 3.2832 50000 16 26 1000 1010 1020 1030 1040 1050 1060 1070 1080 1090 I100 I110 2000 2000 ENDDATA .+7 110 .
.
The output is in terms of Ox. about the global zaxis. In MSCINASTRAN. The rotation matrix R corresponding t o the gimbal angles is given by Ry.LANGLE.1 Gimbal Angle Approach The gimbal angle approach is selected by specifying PARAM. gimbal angle approach and rotation vector approach (implemented in Version 67). and RZ are where the matrices Rx..LANGLE. 8.. The three angles Ox. 5. then by an amount of 8.4. are called gimbal angles in MSCINASTRAN [5. In the gimbal angle approach. The user has the option t o select either of the two approaches. and finally it is rotated about the twice reoriented xaxis by an amount 0.LANGLE. about the global yaxis. there are two different approaches t o update rotations: namely. the three components of rotations at a grid point (O. the three gimbal angles. It can be shown that the same final orientation of the rigid body can also be obtained by first rotating the body by an amount of Ox about the global xaxis.6]..4 'TREATMENT OF LARGE ROTATIONS Geometric nonlinearities may produce large displacements and rotations. 1 0 0 .. 0.2 for the rotation vector approach in either the Bulk Data or the Case Control. and is determined at the time of the cold start.l.5. This is done by specifying PARAM. Unlike displacements. and O..) are interpreted as finite rotations of a rigid body attached to the grid point subjected t o three successive rotations as follows: first the body is rotated about the grid point's global zaxis by an amount 0. finite rotations (components 4 . and finally by an amount of 0. The value of LANGLE can not be changed between subcases or at the restart. 5 and 6 of the MSCINASTRAN displacement vector) can not be updated by simple vector addition. 0. O.l (default option) for the gimbal angle approach and PARAM. and 0.. then it is rotated about the reoriented yaxis by an amount 0.
called auxiliary angles. we have It should be noted that matrices Rz and Ry are based on the gimbal angles at the beginning of the iteration.z) are small. The gimbal lock in MSCINASTRAN is avoided by switching temporarily t o a different set of angles. . This is called gimbal lock. and then reverting back to the gimbal angles. Form rotation matrix R(6) based on the three gimbal angles at the beginning of the iteration.with c.y. and s. t o avoid gimbal lock is as follows: 1. It can be seen from the above equations that there is a singularity at Oy = 90'. and Rx one obtains Three tasks are performed for every iteration in the gimbal angle approach as follows: Step 1. Multiplying Rz. about the global yaxis. Update the Gimbal Angles The gimbal angles are updated using The expressions for the incremental gimbal angles are derived by transforming gimbal angles to global angles and equating them with those obtained from the equilibrium solution. first rotating the body by an amount of 4 about the global xaxis. Solving for the incremental gimbal angles. and s. This is possible because the rotation matrix is unique and there exist several ways t o obtain the final orientation of a rigid body attached t o a grid point. then by an amount of 4. for i=x. and finally by an amount of 4 about the global xaxis. the following expressions are obtained where c. This is reasonable because of the small angle assumption.z correspond t o the gimbal angles a t the beginning of iteration. i=x. The algorithm . = sin 6..y. If we assume the global incremental angles (SOi. are the three gimbal angles. = cos 6. The interpretation of the auxiliary angles is that the rotation of a rigid body attached to a grid point is computed by .z and 6. i=x. Ry.y.
Oi+'.2 Rotation Vector Approach The rotation vector approach is selected by specifying PARAM.e. about an axis of rotation p: l. denoted as $ $. 4. The matrices T b d . +if = + $ A4. and Tbe transform the deformed element coordinates.~ b& = arcsin i ( ~ 2 . $ t o distinguish from the gimbal angles. and the undeformed element coordinates. = arcsin i ( ~ 3 . In the rotation . to basic coordinates. + A+). Recover the updated gimbal angles.. the global coordinates. Tbg. i.. Eliminate Rigid Body Rotation The rigid body rotation is removed from the rotation matrix as follows: where R is the rotation matrix corresponding to angular deformations. respectively. Update the auxiliary angles. The matrices Tdg and Teg transform the global coordinates t o deformed and undeformed element coordinates.e.~ 2 6ey = arcsin i ( 8 1 3 . This algorithm may sometimes lead to sudden changes in the gimbal angles. . 5. where . l 3 . However. Hence. Form updated rotation matrix from the updated auxiliary angles. i.LANGLE. Step 2.B. respectively.e. Step 3.. it does not affect the solution because the element routines use the rotation matrix which is unique. Calculate Angular Deformation The gimbal angle approach assumes that the angular deformations are small.2. The orientation of a rigid body attached t o a grid point is obtained by rotating the body by an amount !! called the magnitude of rotation.4.~ l 2 3 ) 3 1 ) 1 2 ) These angular deformations are then used t o calculate the element forces. i. vector approach the three components of rotation at a grid point. are interpreted as three components of a rotation vector (+). Obtain the auxiliary angles q5 from R(9). R(q5 5 . the angular deformations are recovered as follows: 69..
p.15) where S R is the rotation matrix for the current iteration. sin Q + p. are recovered from given Qi. First update the rotation matrix. three tasks are also performed in the rotation vector approach.cos Q) + cos Q + p t ( l .SR R~ l (5. Let p. and $ the three components of . Comparing the elements of the matrix Eq.4.cos Q) R= + p. i.4. $.4. rotation vector +.cos Q ) p.and (5.4.3]: cos Q p:(l .py sin Q + p.cos Q) . ~ i + . That is SR can be approximated by 1 S$z 6.19) The scalar increment. Then the rotation matrix is as follows [5.(l .e.py(l .cos Q) cos XQ + p:(l .4.13) The output from MSCINASTRAN is in terms of $ . A p .. sin Q + p. is obtained from equilibrium iteration.(l .(l . The details of the three tasks are: Step 1. $  S$x 1 In this algorithm the updated rotation vector zl.i+l is split into a scalar and a vector (no longer unit vector): +i+i = ~ i + Pi+l 1 (5.p. pi and Szl.y.. Update the Rotation Vector The rotation vector can be updated by successive rotations. and the vector increment.15) and after some transformations the following expressions are obtained: .4.p.p.cos Q) + p. sin Q p. sin Q + pyp.(l . AQ.cos Q) (5.. where Szl.14) Like the gimbal angle approach.(l .z) in any iteration are small. be a component of the unit vector parallel to the axis of rotation and Q be the magnitude of rotation.cos !P) p.cos Q ) p. sin Q p. and then recover the updated rotation vector from the updated rotation matrix R~+'.MSCINASTRAN employs an algorithm which assumes that the incremental rotations (S$. i=x. (5.17) with Qi+l = Qi and pi+' = pi + AQ +Ap (5.
4.A Q pX) (W. Calculate Angular Deformation The angular deformations for the current iteration are calculated from the rotation matrix R.16) is used. one needs an estimate for the updated angle +i+l through vector addition of $i and S$: APZ = + + + +i+l = +i S$ + (5. For the singularity at !4 = 0 the small angle rotation matrix in Eq.) sin 2 (sin D AQ cos @ ) ! Special care has t o be taken at the singularity points Q = 0 and J = n.14). The increment in @ is limited t o n per iteration. p. n. which is of the same form as given in Eq. and the singularity a t D = n is avoided by restricting I @ to ! O < . Eliminate Rigid Body Rotation The procedure to eliminate rigid body rotation is the same as that used in the gimbal angle approach. and therefore is only valid in the range 0 < Q < n. < ? (5.24) The singularity points are checked for the angle $i and or Whenever $i $i+l are close to a singularity with a tolerance of kc. p.4.S$. (5.I212 1 Pz = 2 sin P ! It becomes obvious that the recovery procedure outlined above has singularities at 6! = 0. .4. py) sin \.I! (5.the angles are updated through simple vector addition according t o Eq.AQ py) + (S$. That is: The components of the rotation vector p are obtained from the offdiagonal components of R: 1 ~ 1~ 3 3 1 p ~ = 2s i n g 1 ~ 2 . as follows: the magnitude $ is obtained from the first invariant of R. The angles +i as well ! as the updated angles l have t o be checked for the singularity points. (5.A g pz) (SgXpy . (5.4.+ (1 + cosg)(S$y . Therefore. Step 2. APY = APX = (1 + cos Q)(b$.21) + px . The vector addition is consistent with the first order update.4. pZ) sin tlj A@ cos Q) (1 cos Q)(S$z .24). 2 (sin llj + A P cos !J) 2 (sin Q  S$. and does not affect the converged solution.29) 2 This is not only reasonable but necessary t o avoid unrealistic element distortions.S$.4. . Step 3.
Therefore. In nonlinear problems the solution is obtained by iterations.0 0.4683 x 10I 2.698 radians) about the (0. in the gimbal angle approach.4683 x 10I . The output from MSCINASTRAN for the three rotations. 5 and 6 of the displacement vector for the grid point at the end of each rotation is as follows: 1. 0 . The user can choose t o bisect the load or time step if bei.4683 x 10I Qz Gimble angle approach Rotation vector approach 0. Consider a rigid body attached t o a grid point.4 Output Interpretation In MSCINASTRAN the user can specify enforced nonzero angular displacements. Let the rigid body be subjected to two $) axis. Using a larger value of RTOLB may result in more iterations.4.4. 5. i=x.z exceeds RTOLB specified in degrees in the eighth field of the second continuation entry for the NLPARM and the TSTEPNL bulk data entries.0 2.y. After First Rotation: Q z A. The current default value for RTOLB is 20'.. The total rotation is 40' successive rotations of 20' (0. This should not cause any alarm t o the user. However. components 4. A) 5. Therefore. (5. it is more appropriate t o use the rotation vector approach if the angular deformations are expected to be large. The gimbal angles in general for 3D rotations are not the same as the global angles. in actual computations this may not be true.5. However. as the results represent gimbal angles while the specified rotation is about the global axes.16) is a first order approximation.e. no accuracy is lost if the solution converges. However.349 radians) about the (0. Therefore. These angular displacements are always enforced about the global axes.4. Some accuracy in the results may be lost if the angular deformations are large. the printed results may not be equal t o the specified value.3 Bisection Due to Large Rotations Both the rotation vector and gimbal angle approaches assume that the incremental rotations from the equilibrium iterations satisfy the small angle theory. The gimbal angle approach also assumes small angular deformations (in performing Step 3). i. in MSCINASTRAN the user has an option (default) t o bisect the load if the incremental rotations are large.4683 x 10I 2. axis. (0. 2. in many problems it may be a computing advantage t o use a larger value for RTOLB. the printed results from the gimbal angle approach are the three gimbal angles. and since the small angle rotation matrix in Eq.
First Rotation Increment: Gimbal angles at the beginning of increment Incremental angles from global iteration Incremental gimbal angles from Eq.0885 x 10I 4. After Second Rotation: Ox Gimble angle approach Rotation vector approach 6.9365 x 10I 82 5. Second Rotation Increment: Gimbal angles at the beginning of increment Incremental angles from global iteration .8) Updated gimbal angles a t the end of increment 2. 4.2.0 0.2192 x 0.9365 x 10I The details of the calculations for the two approaches are as follows: Gimbal Angle Approach 1.4.8617 x 10I 4. (5.
= 2. (5.8) ne. = 2. First Rotation Increment: r At the beginning of increment r Incremental angles from global iteraiion r Updated angles at the end of increment from Eq.4.24) 2.3941 x 10I Ae. = 6.4. (5. Second Rotation Increment: r At the beginning of increment for which .2192 x 102 A@.r Incremental gimbal angles from Eq.6202 x 10I r Updated gimbal angles at the end of increment Rotation Vector Approach 1.
= 3.4907 x 10I updated angles = ll.p + 3.9814 x 10I As a result .4907 x 10I = 6.Incremental angles from global iteration for which At the end of increment $ = $ + All.
.
SOL 64 and SOL 66 in Version 66A are used for the analysis of the two problems.4.lPcr (initial disturbance) at the free end as shown in Fig. The main feature of these problems is geometric nonlinearity due to large rotations. 5. It is therefore preferable t o use SOL 66 (or 106) for the geometric nonlinear analysis.2.5.1. Figure 5. and a small lateral force Pi=O.518. Beam with Axial End Force Initially. SOL 64 is an older trial DMAP developed for geometric nonlinear analysis only. is 2. The Young's modulus is 20 x lo6 lb/in2 and the moment of inertia I. Two cases are analyzed according t o the loading conditions at the free end of the beam: axial end force and end moment.1 Problem Description A uniform beam is clamped at one end and free at the other end.116 in order t o compare the results with those of Timoshenko and Gere.152. and 9. The lateral force is subsequently removed and large rotations of the beam are produced when the axial force is increased above the critical value PC. SOL 66 is also simpler to use because the solution is controlled by the NLPARM Bulk Data instead of subcases and parameters as in SOL 64.5 VERIFICATION: LARGE ROTATION OF A BEAM This example compares geometric nonlinear analysis using SOL 64 and SOL 66.5. whereas SOL 66 (or 106) is more recently developed and contains many features not available in SOL 64.1 Beam with an Axial End Force. the beam is subjected t o an axial force P=0.5.7] is The beam rotation is determined for the ratios of P/PcT of 1. The beam has a length of 12 inches and a crosssectional area of 0.813 x in4.5. Two problems are studied: a beam with an axial end force and a beam with an end moment.1.541. For this problem. the critical load which is calculated by Timoshenko and Gere [5. . 5.15 in2.029...9Pc.
3 Analysis Procedure SOL 64 Analysis Subcases and parameters control the number of iterations.l.) are contained in the first two subcases. for the axial end force case and in Tables 5.3 and 5.5.2. The analysis is performed in seven subcases for the beam with axial end force. 5.5.4. and lateral force of O. The input data for SOL 64 and SOL 66 are given in Tables 5.5. the solution is controlled by subcases. with no lateral force. 5. PARAM. The initial loading conditions (axial force of 0. 5.5.5. respectively.2. The clamped boundary condition at one end of the beam is imposed by constraining all six components of GRID 100.lPc. . respectively. The solution strategy is provided by the NLPARM Bulk Data entry in SOL 66. and large rotations are produced..5. The applied load at the free end of the beam (GRID 112) is specified using the FORCE Bulk Data entry for the axial end force case and the MOMENT Bulk Data entry for the end moment case. as shown in Fig. and stiffness matrix updates in SOL 64. for the end moment case. respectively.2 Beam with an End Moment. I I I ~ I I I I I I IM = 3000 inlb Figure 5.Beam with End Moment A moment with a magnitude of 3000 inlb is applied in the +z direction at the free end of the beam.2 Finite Element Model The beam is modeled by twelve BEAM elements with linear elastic material. The first two subcases correspond t o the linear elastic and the differential stiffness solutions. Five additional subcases are included t o specify the five axial force ratios of PIP.5.9Pc. load vector updates.TESTSE.5.l. Geometric nonlinearity effects are included in the SOL 66 analysis by specifying PARAM. In the SOL 64 analysis. The stiffness matrix is updated after every iteration and there is one load increment per subcase.1 and 5.10 is included in the Bulk Data for an automatic convergence test based on strain energy.LGDISP.
. The displaced shapes of the beam are shown in Fig. Automatic convergence tests are provided and no special parameters need t o be included to control the solution strategy. Runs are made with different KMETHOD options (ITER. when the AUTO option is selected with line search and BFGS updates.. the NLPARM Bulk Data controls the load increment size and the iteration strategy. The distance yend is the displacement of the free end of the beam in the vertical direction and xend is the horizontal distance from the free end of the beam t o its clamped end. of iterations AUTO 5. For the beam with end moment.. SOL 66 Analysis In SOL 66. SEMI. . The MSCINASTRAN results are in good agreement with the values of Timoshenko and Gere.For the beam with end moment. The analysis is performed in six subcases for the beam with axial end force. Performance for Beam with E n d Moment No. However.5.5. the analysis is performed in twelve subcases. The tangent at the free end of the beam has an angle Bend with the horizontal. triggers large bending beyond the critical buckling load. 5. the analysis is performed in one subcase which has ten increments.3.7] are tabulated below for the ratios of PIP. The relation between the axial and lateral displacement of the free end of the beam (ul12 and vllz) and the axial force is shown in Fig.4 Analysis Results The MSCINASTRAN analysis results for the beam with axial end force and the values calculated by Timoshenko and Gere [5. the solution diverges at the first load increment. 5.5. The first two subcases correspond to the linear elastic and differential solutions. and the last ten subcases correspond t o the ten moment increments. or AUTO). of KMETHOD K updates No.4. while KSTEP=l. which contains the initial loading. which specify the stiffness matrix update strategy. specifying the full Newton's method.. Each subcase has one increment and the ITER option of KMETHOD is selected with KSTEP=l. The lateral force is removed and each of the remaining subcases correspond t o an increasing axial force ratio of P/P. The performance in terms of the number of iterations and stiffness updates is tabulated below. The first subcase. The analysis executes successfully with each of the KMETHOD options when the line search and BFGS updates are suppressed (MAXLS=O and MAXQN=O).
7] Figure 5. . Displacement at Free End of Beam with Axial End Force.3 Load vs.LoadDeflection Data of Beam with Axial End Force Notes: values in ( ) denote solution in Timoshenko and Gere [5.5.
Figure 5.5. .4 Deflection Curves for Beam with Axial End Force.
Note that after ten increments.5 Deflection Curves for Beam with End Moment.5 . the angle that the tangent at the free end of the beam has with the horizontal. The deflection curves are labeled with the increment number (from 1 t o 10). the beam has deformed into a complete circle.3" 110. 5.7" 73.7" 183. inc 1 2 3 4 5 6 7 8 9 10 moment (inlb) 300 600 900 1200 1500 1800 2100 2400 2700 3000 Bend 36.OO 366.6 . The moment magnitude and Oend.6" Figure 5.3" 330 .5.0" 256.5.6" 293.5. are listed for each incremental step.0" 146.The displaced shapes of the beam with end moment are shown in Fig. 5.3" 2'20.
Y PTITLE = TOP VIEW FIND SCALE ORIGIN 1 SET 1 PLOT STATIC 0 SET 1 ORIGIN 1 SYMBOL 1 BEGIN BULK $ PARAMETERS PARAM TESTSE 1. 0. 0. 0. 0.5*PI/L)**2 ECHO = BOTH SEALL = ALL DISP = ALL SPCF = ALL ELFOR = ALL SPC = 100 SUBCASE 1 $ P/P(CR)=0.116 LOAD = 9116 OUTPUT(PLOT) CSCALE 1. GRID 106 6. GRID 105 5. .900 PLUS LATERAL LOAD = 900 SUBCASE 3 $ P/P(CR)=l. 0. GRID 103 3. 0.900.Table 5. 0.5. 152 LOAD = 1152 SUBCASE 4 $ P/P(CR)=l.V6402S SUBTITLE = TIMOSHENKO AND GERE. GRID 101 1. GRID 104 4.76 = 96. 0.029 LOAD = 4029 SUBCASE 7 $ P/P(CR)=9.1 SOL 64 Input Data Listing for Beam with Axial End Force PSTBUCK64. AXES Z . 0. GRID 102 2. 0.10 $ GEOMETRY GRID 100 0. P.PLUS LATERAL LOAD = 900 SUBCASE 2 $ P/P(CR)=0.518 LOAD = 1518 SUBCASE 5 $ P/P (CR) =2. THEORY OF ELASTIC STABILITY.3 PLOTTER NAST SET 1 = ALL VIEW 0. 0. V66A $ APR 20MAR90 ID TIME 5 $ CPU MINUTES $ GEOMETRIC NONLINEAR ANALYSIS SOL 64 8 DIAG CEND TITLE = LARGE DEFLECTIONS OF A BEAM . 0. .4 LABEL = P(CR1T) = EI*(. . 0.541 LOAD = 2541 SUBCASE 6 $ P/P (CR) =4. X . 0. 0. 0.
GRID Ill GRID 112 12.GRID 107 7.+7 1. $ CONNECTIVITY CBEAM 1 10 100 CBEAM 2 10 101 CBEAM 3 10 102 4 10 103 CBEAM CBEAM 5 10 104 CBEAM 6 10 105 CBEAM 7 10 106 CBEAM 8 10 107 10 108 CBEAM 9 10 109 CBEAM 10 CBEAM 11 10 110 CBEAM 12 10 Ill $ PROPERTIES PBEAM 10 11 . GRID 108 8.4 FORCE 900 112 112 FORCE 1152 FORCE 1518 112 FORCE 2541 112 FORCE 4029 112 FORCE 9116 112 $ I01 102 103 104 105 106 107 108 109 110 Ill 112 ENDDATA .P(CRIz96. 11. GRID 110 10. GRID 109 9.I5 MAT1 I1 2.+7 $ CONSTRAINTS SPCI 100 126 100 $ LOADING .
0.541 LOAD = 2541 SUBCASE 5 $ P/P(CR)=4.518 LOAD = 1518 SUBCASE 4 $ P/P (CR) =2. V66A $ APR 20MAR90 5 $ CPU MINUTES TIME $ NONLINEAR STATIC ANALYSIS SOL 66 8.2 SOL 66 Input Data Listing for Beam with Axial End Force ID PSTBUCK66. 3. 0. 0.152 LOAD = 1152 SUBCASE 3 $ P/P(CR)=I. 0.116 LOAD = 9116 OUTPUT(PLOT) CSCALE 1. X . 0. 0. I 0 YES +NLP 0. Y PTITLE = TOP VIEW FIND SCALE ORIGIN I SET I PLOT STATIC 0 SET I ORIGIN I SYMBOL I BEGIN BULK $ PARAMETERS PARAM LGDISP I $ NLPARM 1000 +NLP $ GEOMETRY GRID 100 101 GRID GRID 102 GRID 103 GRID 104 I 0.029 LOAD = 4029 SUBCASE 6 $ ~/~(CR)=9. .76 LABEL = P(cRIT) = EI*(. I.V6402S SUBTITLE = TIMOSHENKO AND GERE. 2. .5. 0. 4. ITER 0.3 PLOTTER NAST SET I = ALL VIEW 0.900 PLUS LATERAL LOAD = 900 SUBCASE 2 $ P/P (CR) =I.5*PI/L)**2 = 96.50 $ PRINT MATRIX TRAILERS AND ITERATIONS DIAG CEND TITLE = LARGE DEFLECTIONS OF A BEAM . 0.4 ECHO = BOTH SEALL = ALL DISP = ALL SPCF = ALL ELFOR = ALL SPC = 100 NLPARM = 1000 SUBCASE I $ P/P(CR)=0. 0.Table 5. P. . 0. THEORY OF ELASTIC STABILITY. AXES Z .
9.+7 126 100 101 102 103 104 105 106 107 108 109 110 11 1 .+7 100 100 LOADING .P(CR)=96.38286 FORCE 900 112 FORCE 1152 112 FORCE 1518 112 FORCE 2541 112 FORCE 4029 112 FORCE 9116 112 $ ENDDATA .GRID 105 GRID 106 GRID 107 GRID 108 GRID 109 GRID 110 GRID 111 GRID 112 $ CONNECTIVITY CBEAM 1 CBEAM 2 CBEAM 3 CBEAM 4 CBEAM 5 CBEAM 6 CBEAM 7 CBEAM 8 CBEAM 9 CBEAM 10 CBEAM 11 CBEAM 12 $ PROPERTIES PBEAM 10 MAT1 11 $ CONSTRAINTS SPC1 $ 5. 10 10 10 10 10 10 10 10 10 10 10 10 11 2.15 1. 11. 12. 7. 10. 6. 8.
0. V66A $ APR 20MAR90 ID $ CPU MINUTES TIME 5 64 $ GEOMETRIC NONLINEAR ANALYSIS SOL 8 DIAG CEND TITLE = LARGE DEFLECTION DEMO ECHO = BOTH SEALL = ALL DISP = ALL SPCF = ALL ELFOR = ALL SPC = 100 SUBCASE I LOAD = 103 SUBCASE 2 LOAD = 103 SUBCASE 3 LOAD = 103 SUBCASE 4 LOAD = 106 SUBCASE 5 LOAD = 109 SUBCASE 6 LOAD = 112 SUBCASE 7 LOAD = 115 SUBCASE 8 LOAD = I 1 8 SUBCASE 9 LOAD = 121 SUBCASE 10 LOAD = 124 SUBCASE 11 LOAD = 127 SUBCASE 12 LOAD = 130 OUTPUT(PLOT) CSCALE 1..5.10 .Table 5.3 SOL 64 Input Data Listing for Beam with End Moment BMMOM64.3 PLOTTER NAST SET I = ALL AXES Z. X. O. Y VIEW O. PTITLE = TOP VIEW FIND SCALE ORIGIN I SET 1 PLOT STATIC 0 SET 1 ORIGIN I SYMBOL 1 BEGIN BULK $ PARAMETERS PARAM TESTSE 1..
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. ENDDATA .$ GEOMETRY GRID 100 GRID 101 GRID 102 GRID 103 GRID 104 GRID 105 GRID 106 GRID 107 GRID 108 GRID 109 GRID 110 GRID Ill GRID 112 $ CONNECTIVITY CBEAM I CBEAM 2 CBEAM 3 CBEAM 4 CBEAM 5 CBEAM 6 CBEAM 7 CBEAM 8 CBEAM 9 CBEAM 10 CBEAM I1 CBEAM 12 $ PROPERTIES PBEAM 10 MAT1 11 $ CONSTRAINTS SPCI 100 $ LOADING MOMENT 103 MOMENT 106 MOMENT 109 MOMENT 112 MOMENT 115 MOMENT 118 MOMENT 121 MOMENT 124 MOMENT 127 MOMENT 130 $ 100 I01 102 103 104 105 106 107 108 109 110 Ill 101 102 103 104 105 106 107 108 109 110 Ill 112 0.
PTITLE = TOP VIEW FIND SCALE ORIGIN I SET I PLOT STATIC 0 SET I ORIGIN I SYMBOL I BEGIN BULK $ PARAMETERS PARAM LGDISP I $ NLPARM 1000 +NLI $ GEOMETRY GRID 100 GRID 101 GRID 102 GRID 103 GRID 104 GRID 105 GRID 106 GRID 107 GRID 108 GRID 109 110 GRID GRID Ill GRID 112 $ CONNECTIVITY CBEAM I CBEAM 2 CBEAM 3 10 ITER I 0 YES +NLI 10 10 10 . Y VIEW O.4 SOL 66 Input Data Listing for Beam with End Moment BMMOM66. X. O.5.50 $ PRINT MATRIX TRAILERS AND ITERATIONS DIAG CEND TITLE = VERY LARGE DEFLECTION DEMO ECHO = BOTH SEALL = ALL DISP = ALL SPCF = ALL ELFOR = ALL SPC = 100 SUBCASE 10 LOAD = 120 NLPARM = 1000 OUTPUT(PLOT) CSCALE 1.Table 5.3 PLOTTER NAST SET I = ALL AXES Z. 0. V66A $ APR 20MAR90 ID TIME 5 $ CPU MINUTES $ NONLINEAR STATIC ANALYSIS SOL 66 8...
. 0. 0. 0. 0. 0. 0.CBEAM 4 CE M BA 5 CE M BA 6 CE M BA 7 CE M BA 8 CE M BA 9 CE M BA 10 CEM BA 11 CBEAM 12 $ PROPERTIES PEM BA 10 MT A1 11 $ CONSTRAINTS SPCI 100 $ LOADING M MN O E T 120 $ E D AA NDT 103 104 105 106 107 108 109 110 Ill 104 105 106 107 108 109 I10 Ill 112 0. 0. 0.
6. dxT dx = (F d ~ ) (*F d X ) = d x T C d X (5.6. As is .e. the spatial deformation gradient can be defined by Fl. If the motion of a body is described in terms of the undeformed configuration using a referential coordinate system (denoted by X ) attached t o the object material. ax. i. or Eulerian method treat these effects of large displacements by starting with the finite strain. referred t o as canonical approach.121 such as the total Lagrangian. as well as the applied external loads vary with large displacements. Conversely. the method is known as the Lagrangian (or material) formulation.161.t).4) with c = F ~ F where Green's deformation tensor is defined by C.6. in contrast t o an approximate updated Lagrangian approach employed by MSCJNASTRAN based on the infinitesimal strain. X = X(x.3) Then. the Eulerian formulation gives the initial position of the particle at time t in terms of the current position x.6. The Lagrangian description may be interpreted as a mapping of the initial configuration into the current configuration.e.85. The deformation tensor similarly defined for the Eulerian formulation is called Cauchy's deformation tensor and is denoted by c. However. updated Lagrangian. Conventional methods [5. These conventional methods which are based on the classical continuum mechanics theory [5. axz ax3 & & (5.6 CANONICAL APPROACH TO GEOMETRIC NONLINEARITY In geometric linear analysis where the displacements are assumed small regardless of the material nonlinearity. [B]. && ax2 && axl ax3 ax. 5. (5.1 Fundamentals of Continuum Mechanics The kinematics of continua may be described from two different points of view. are reviewed in this section. the element matrix. the description of the motion in terms of the deformed configuration using spatial coordinates fixed in space (denoted by x) is called the Eulerian (or spatial) formulation. z axl ax.135. it is implied that the straindisplacement relations [B] stay constant.e. i. Consider a squared length of an infinitesimal line element in terms of the deformation gradient. i.2) The most fundamental quantity of these descriptions is the material deformation gradient defined as F = [$I [ = a.5. In contrast.
According t o the polar decomposition theorem. Within the framework of the infinitesimal theory. the finite strain tensors can be expressed in terms of the displace and where the repeated indices represent summation. in which the Lagrangian (also called Green's) finite strain tensor is defined t o be The Eulerian formulation can be derived likewise.apparent from the derivation. it can no longer be reduced t o a strain and a rotation by an additive decomposition. these deformation tensors are symmetric and positivedefinite.e. however. i. the symmetric part is the linear strain tensor and the skewsymmetric part is the linear rotation tensor. The displacement gradient can be decomposed into a symmetric and a skewsymmetric part. i. in which the Eulerian (also called Almansi's) finite strain tensor is defined t o be Noting that x = X ment gradient.e.e. which is the basis of the small deformation theory. i. . When components of the displacement gradient are sufficiently small compared t o unity. the multiplicative decomposition is possible if the deformation gradient is not a singular tensor. i. Notice that the expression for the finite strain tensor is exact and not a second order approximation.e. The change in the squared length is used as a measure of deformation. Then. + u.e. the distinction between the Lagrangian description and the Eulerian description disappears. When the displacement gradient is finite. the product terms are negligible and the finite strain tensors reduce t o the infinitesimal strain definition. i.
it can be shown that the finite strain tensors are invariants with respect to the rigidbody motion.6. which is known as the stretch ratio (A = dx/dX).6. it has advantages of being symmetric and independent of rigidbody rotation.e. and M and N are positive definite symmetric tensors. This stress tensor is easy t o define but has the disadvantage of being nonsymmetric.16) [Sij] = ( det F ) FI [aG]F .6.where R is the orthogonal rotation tensor. each principal value of M. A significant interpretation can be given to Eq.15) ( det F ) loij] = F [Sij] F~ = F [Tij] or inversely..e. Based on the polar decomposition.' ) ~ ( where the determinant of the deformation gradient F represents the change in volume due t o the deformation. is defined to give the actual force acting on the deformed surface element with an area dS. dxl dx2 dx3 . (5. Stresses are usually measured in the current configuration. however. The stress in the deformed state is known as the Cauchy stress. The first PiolaKirchhoff stress tensor (also known as Lagrangian stress). These PiolaKirchhoff stress tensors are related t o the Cauchy stress tensor as follows: (5. however. i.. Two PiolaKirchhoff stress tensors are defined in the undeformed (or referential) configuration. and the principal directions of C and M coincide. but the force is assigned t o the unit area of the corresponding undeformed surface dSo. but the force is related to the actual force in the deformed surface by the deformation gradient.18) .v dXldX2dX3 Vo' The second PiolaKirchhoff stress tensor has little physical meaning. because the equilibrium must be established in the deformed configuration. e. denoted by Tij. respectively. (5.14). The second PiolaKirchhoff stress tensor (also called Kirchhoff stress). denoted by Sij. the strain energy is preserved by det F = Sij SEij dV = (5.. i.] ( F . It is the second PiolaKirchhoff stress tensor that is energetically conjugate t o the Lagrangian strain tensor. consistent with the Eulerian strain tensor. It follows that by virtue of RTR = I. is the square root of the corresponding principal value of C. stresses should be defined similarly. In the course of analysis.' ) ~ = [Ti. known as the right and left stretch tensors.g.6. if the strains are referred t o the original position of particles as in the Lagrangian description. is defined to give the corresponding force acting on and assigned t o the undeformed surface element dSo.
When the velocity is described only in terms of spatial variables x which are implicit functions of time.( L . respectively. . then the acceleration can be calculated by . i. For instance.21) The rate of deformation and the vorticity tensors can be interpreted as the material derivatives of the Eulerian linear strain and rotation tensors.5.6.e. (5.2 Incremental Formulations for Large Displacements and Strains The kinematic tensors in the incremental process of flow problems such as in plasticity. 21 = "(2. Consider an acceleration which is a material derivative of the velocity. the time rate of change measured at a fixed point in space is called the local rate of change or the spatial derivative. respectively. t). the vorticity tensor represents the angular velocity of the rigidbody rotation.6.LT) 2 where D and W are called the rate of deformation (or stretching) tensor and the vorticity (or spin) tensor. the velocity gradient rather than the displacement gradient is used.(L 1 2 + LT) = .e. The material derivative is the time rate of change of any property of a moving continuum measured by an observer traveling with a specific particle under consideration. da = D dt . i. While the rate of deformation tensor is independent of rigidbody motion. creep or viscoplasticity are frequently expressed in terms of time rates in the process of linearization t o follow the deformation paths.e. i. with a symmetric part: D = and a skew symmetric part: .e. i. The rate of deformation tensor can be used for a 3dimensional generalization of the naturalstrain increment in time dt. In contrast. Recalling that the naturalstrain increment is defined as w 1 t o give the following logarithmic strain by an integration over the length.
Functions (vectors or tensors) are called frameindifferent if their values are unchanged by the superposed rigidbody motion when the frame of reference is changed from {x. despite the fact that the velocity is not. + the vectors and tensors transform according to v* = Q(t) v and (for vectors) (for tensors) (5.6. Specifically. t ) . Finite strains have a similar problem because the material derivative of the finite strain is not frameindifferent. As pointed out earlier.26) S* = Q(t) s ~ ( t ) ~ where Q(t) is a rotation tensor and c(t) represents a translation of the coordinate origin.6.=&WO+OW (5. However. and R is the angular velocity tensor of the unstarred frame relative to the starred frame.where the subscripts identify the variables held constant. It is noted that the spin tensor is not frameindifferent and transforms by with R = QQT = . The material derivative of the deformation gradient can be obtained by . when the change of frame is described by (5. it can be demonstrated that the frameindifferent i stress rate is defined as . The first term gives the local rate of change and the second term gives the convective rate of change. The principle of material frameindifference states that constitutive equations must be invariant under changes of reference frame.29) which is known as the Jaumann stress rate or the corotational stress rate tensor and is denoted by v over stress. av F== ax ax ax av ax = LF. Thus.Q Q' T = QT where Q is the material derivative of Q. t*). Of our particular interest is the time rate of change of the stress among constitutive variables in the incremental formulation because the material derivative of the stress (irij) is not frameindifferent even though U j is. t) to {x*. The first term on the right side represents the material rate of change with respect t o a fixed coordinate system and the last two terms represent the convective rate of change.24) x*(X. t*) = c(t) &(t) x ( X . that the deformation gradient F transforms like a vector. the rate of deformation tensor is a material derivative of the linear strain. However. It can be shown.6.25) (5.6. it can be shown that the rate of deformation tensor [D) is frameindifferent. however. the frameindifferent .
treating the previously converged configuration as an initial geometry. From the principle of virtual displacements. In the total Lagrangian formulation. When the choice of the referential configuration is the preceding incremental load (or time) step. the updated Lagrangian formulation may be regarded as an application of the Lagrangian method t o a single load step with the referential configuration brought uptodate continuously. the basic equation to solve in accordance with the Eulerian approach is which is to be satisfied in the currently deformed configuration.6.constitutive relations for the incremental formulation can be obtained by using the Jaumann stress rate and the infinitesimal strain rate.e. where D represents the constitutive relations. Now it remains to linearize Eqs. i. with 0. respectively. all the variables are referred t o the initial configuration. the method is known as the updated Lagrangian formulation.31a) and (5. such a natural procedure of the Lagrangian formulation may be applied with an arbitrary choice of the referential configuration.6. The fundamental difficulty with the Eulerian approach is that the current geometry of the body.3110) to find the tangential stiffness. and t' denoting the initial. both based on the Lagrangian description of kinematics. the incremental process of multiple load steps is applied to the Lagrangian formulation. The basic equation can be rewritten for the total Lagrangian formulation as and for the updated Lagrangian formulation as where the left superscripts refer to the current configuration of the body and the left subscripts refer t o the referential coordinate axes. It is noted that the expression for the updated Lagrangian formulation is obtained by substituting t for the left subscripts C in the total Lagrangian formulation. It will be assumed for simplicity that the applied load is independent of deformation. (5. On the other hand. In the total Lagrangian formulation. is yet to be determined. which is the referential configuration. Two popular approaches to the incremental formulations are considered here: the total Lagrangian and the updated Lagrangian approaches. however. . Without loss of generality. t. the tth incremental. Now we are ready to derive the incremental formulations for solving general nonlinear problems with large displacements and large strains. and the (t+l)th incremental steps.
0E2 is the nonlinear due strain increment causing a stiffness [ K ~ ] to the initial stress.5 6.6.6.6. the substitution of Eqs. the stress and strain (omitting indices) may be decomposed as follows: and with t' .3 Total Lagrangian vs.E = . If the straindisplacement relations are linearized by [%It = . Updated Lagrangian in Finite Elements Recalling that t'u = tu+u for the total Lagrangian formulation. and 0E3 is a crossproduct term which causes a stiffness [ K g ] due t o the initial displacement. (5.31a) gives the following equation with some rearrangement of terms: Notice that 0El is the linear strain increment causing a linear stiffness [I<'].B = b~~ + ~ B N . the equilibrium equation can be rewritten in matrices as .E + o E 1 + oE2 + oE3 Noting that S ~ = S(oE1 E and + oE2 + oE3) St'U = Su .32) and (5.33) into Eq. (5.
. 6s For the updated Lagrangian process. Thus. i. (5. the unbalanced force vanishes as the solution approaches convergence.6. many researchers reported difficulties with the total Lagrangian method in those problems.37) must be defined with reference t o the initial geometry. the displacements are updated continuously. The total Lagrangian approach is not appropriate for the pathdependent inelastic problems such as a plasticity problem. which is meaningful only for elastic materials such as a hyperelastic material.where (omitting the leftsuperscript t and leftsubscript 0 of B).40) + and t' E = El + E2 with and . for the ith iteration with which 6 BN and are reevaluated to yield $ BN and $5'. In fact. The reason is that the constitutive relations OD in Eq.6. we have to resort t o the updated Lagrangian approach for general geometric nonlinear problems. stresses and strains may be decomposed (omitting the leftsubscript t ) as follows: = tS S (5.e.e. with a constitutive relation is = OD i E . During the iteration. i. Thus. The concept of the updated Lagrangian method is realized by updating the state variables at the end of each incremental step with respect to the converged configuration or convected coordinates.
40) and (5. Introducing the constitutive relations at t by s = t ~ ( ~E l~ ) + and linearizing Eq.31b) gives.42) by neglecting E 2 in the first term. the substitution of Eqs. . the equilibrium equation can be written in a matrix form (omitting the leftsubscript t of B and D ) as [I(' with + gd] {u) I(' = = t' P  lV BE {a) dV ly[BE tD BL]dV and where the current straindisplacement relation is obtained by and the volume integration is performed over the current volume. (5.Noting that t S = a.6. (5.6. (5.6.41) into Eq. with some rearrangement of terms.6.
5.6.4
Constitutive Relations for Large Deformat ion
Difficulties arise in using the Lagrangian constitutive relations in Eq. (5.6.43). First of all, the Lagrangian strain increments are not additive while the elastoplastic deformation process requires (5.6.46) &total = d,se d ~ .~
+
However, since the rate of deformation tensor is additive, we can establish the rate constitutive v relations in terms of the Jaumann stress rate ( a ) and the rate of deformation tensor as presented in Eq. (5.6.30). The rate of deformation tensor (D) is related to the Lagrangian strain rate by
By allowing the coordinate axes of the referential configuration to be momentarily coincident with that of the current geometry, the distinction disappears and the Lagrangian strain rate may be used as a constitutive variable, i.e.
Recalling the generalization of the naturalstrain increment defined in Eq. (5.6.21), an integration of Eq. (5.6.48) may be interpreted t o give the logarithmic strain which is denoted by E*, i.e. E* = J ~ d t . (5.6.49) This is another advantage of using the rate equations since we can now formulate the constitutive relations [tD]based on the true stressstrain curve obtained from the uniaxial test. It is noted that the generalized logarithmic strain, which has a physical meaning in the updated Lagrangian coordinates, is different from either the Lagrangian or the Eulerian strain. For every iteration, displacements are updated by
and the stresses on the righthand side of Eq. (5.6.44) by
However, the frameindifferent constitutive relations involve the Jaumann stress rate, which is related t o 6 by Eq. (5.6.29), in order t o account for the rotation of the stress during the increment. Since the Cauchy stress and the Kirchhoff stress are related by Eq. (5.6.15), the material derivative of the Cauchy stress can be obtained by taking a material derivative of Eq. (5.6.15), i.e. F S F ~ (detF)(divva = ir  L a  O L ~ ) (5.6.52)
+
using Eq. (5.6.23) and the material derivative of (det F ) from
d  (det F ) = J = J dt
vi,i
= J div v
.
(5.6.53)
By using Eq. (5.6.29) and the momentarily coincident referential configuration, Eq. (5.6.52) reduces t o v s =a DoaD+divva. (5.6.54) For incompressible materials, as in plasticity, Eq. (5.6.54) further reduces to
using Eq. (5.6.19) with a substitution of E for D. Then the stress increment is calculated by
It is noted that the volume integration is performed over the configuration at t in Eq. (5.6.42). As the solution converges, the stress becomes tS, and should be transformed to the Cauchy stress a t t' according t o Eq. (5.6.15). For the hyperelastic material, the Eulerian constitutive relations may be obtained from the Lagrangian constitutive relations by
= F F ~ D F T F T . det F
The Lagrangian strain can be transformed t o the Eulerian strain at t' by
(5.6.57)
where [ t F ]is the deformation gradient at t' with reference t o the configuration at t.
Chapter 6
MATERIAL NONLINEARITY AND CONSTITUTIVE RELATIONS
6.1
6.1.1
INTRODUCTION
Overview
Material nonlinearity becomes an issue in the analysis if the stressstrain relationship or the constitutive equations are nonlinear. Constitutive equations characterize the macroscopic behavior of the material, which stems from the microscopic constitution of the material. All the engineering materials are inherently nonlinear. From the mathematical point of view, it is not feasible t o characterize a nonlinear material by a single set of equations for the entire range of environmental conditions, such as the loading, temperature, and the rate of deformation. Even in the restricted region of these conditions for specific needs, the mathematical model of the material is still an idealization (e.g., incompressibility) of more general nonlinearity. Depending on the application or the situation, the material behavior may be idealized or simplified t o account for certain effects which are important in the analysis. The linear elastic (Hookean) material is the simplest case of all in solid mechanics. As the strain exceeds the infinitesimal range, the material exhibits nonlinear behavior: nonlinear elastic if it is recoverable and plastic if it is irrecoverable. If the temperature effects on the material properties become an important issue, coupling between thermal and mechanical behavior should be properly taken into consideration, which is the realm of thermoelasticity and thermoplasticity. If the strainrate has significant effects on the material response, we have to resort t o the theories of viscoelasticity and viscoplasticity. Anisotropic effects add complexity t o the material models of all kinds in solid mechanics.
Development of material constitutive models has been prolific in recent years. Detailed discussion in this book will be confined to the capabilities provided in MSCINASTRAN.
6.1.2
User Interface
The data for nonlinear elastic or plastic material applications are contained in the Bulk Data entry MATSl, in connection with the MATl entry with the same MID. The MATSl entry specifies the stressdependent material properties according to the format shown below. A brief description of each of the fields is provided, along with default values and further elaborating remarks as applicable.
MATS1 MATS1
MID 17
TID 28
TYPE PLASTIC
H 0.0
YF
1
HR.
1
LIMIT1 2.+4
LIMIT2
MID Identification number of a MATl entry (Integer > 0). TID Identification number of a TABLES1 or TABLEST entry (Integer 2 0). If H is given,
then this field should be blank. See Remark 1.
TYPE Specifies the type of material nonlinearity: NLELAST (Nonlinear elastic) or PLASTIC (Elastoplastic). See Remarks 2 and 3. H Work hardening slope (slope of stress vs. plastic strain) in units of stress (Real). For elasticperfectly plastic cases, H=O.O (default). For more than a single slope in the plastic range, the stressstrain data must be supplied on a TABLES1 entry referenced by TID and this field must be blank. See Remark 3.
YF Yield function (Integer) is selected by one of the following values:
1 2 3 4
von Mises (Default) Tresca MohrCoulomb DruckerPrager
HR Hardening Rule (Integer) is selected by one of the following values:
1 Isotropic hardening (Default) 2 Kinematic hardening 3 Combined hardening
LIMIT1 Parameter representing an initial yield point (Real). See Remark 4. LIMIT2 Parameter representing the internal friction angle for the MohrCoulomb and DruckerPrager yield criteria. (0
5 Real < 45'). See Remark 4.
Remarks:
1. If TID is given, TABLES1 entries (xi, y;) of stressstrain data ( E ~ , must conform t o Yk) the following rules (see Figure 6.1.1):
(a) If TYPE is PLASTIC, the curve must be defined in the first quadrant. The first point must be at the origin (xl = 0, yl = 0) and the second point (x2, y2) must be at the initial yield point (Y1 or 2c) specified on the MATS1 entry. The slope of the line joining the origin to the yield stress must be equal to the value of E on the MAT1 entry. (b) If TYPE is NLELAST, the full stressstrain curve (co < x < co) may be defined in the first and the third quadrant to accommodate different uniaxial compression data. If the curve is defined only in the first quadrant, then the curve must start at the origin (xl = 0, yl = 0) and the compression properties will be assumed identical t o tension properties. (c) If TYPE is PLASTIC and TID is given, it may not reference a TABLEST entry. 2. If TYPE is NLELAST, the stressstrain data given in the TABLES1 entry will be used t o determine the stress for a given value of strain. The values H, YF, HR, LIMIT1, and LIMIT2 will not be used in this case. For nonlinear elastic temperaturedependent material properties, TID must reference a TABLEST entry and TABLEST subsequently references TABLES1 entries for different temperatures. Also a MATT1 entry should be supplied to specify the temperature dependence of material properties.
3. If T Y P E is PLASTIC, either the table identification, TID, or the work hardening slope, H, may be specified but not both. If the table ID is omitted, the work hardening slope, H, should be specified in field 5 unless the material is perfectlyplastic. The plasticity modulus (H) is related t o the tangential modulus (ET) by
where E is the elastic modulus and ET (= curve in the plastic region. See Figure 6.1.2.
g) the slope of the uniaxial stressstrain is
4. LIMIT1 and LIMIT2 are parameters used in the yield function and must be defined as follows:
Yield Function von Mises (1) or Tresca (2) MohrCoulomb (3) or DruckerPrager (4)
LIMIT1 Yield stress in tension, Y1
LIMIT2 Not used
2cCohesion, 2c (in stress units)
Angle of internal friction
4 (in degrees)
Y (or a)
1
.; y2 1
Y3
y1 .'
!
/
j
/
, , '
,,< AT ; , '
H1
I
I I
/
/I
I
I
Hz
H3
I
k=3
k=2
II
I
k=l
I
I I
I
1
~f TYPE is PLASTIC:
F: = Effective Plastic Strain
I
I
I
E
I
I I
II II
I
1 I
II I I
H k = & Y kq I  y k =
I
I
I
I
I
I
1 ....?.....
0
El
z;
i
E2
F;
E3
E
Figure 6.1.l. StressStrain Curve Definition with TABLES1 Entry.
Y (or F)
Figure 6.1.2. StressStrain Curve Definition for a Single H.
The parameters in the MATS1 Bulk Data entry pertain only t o nonlinear elastic and plastic material nonlinearities. The required data for creep analysis are contained in the Bulk Data entry CREEP, which defines the creep characteristics based on experimental data or known empirical creep laws. In order t o activate the creep analysis, the CREEP entry must contain the same MID as an associated MAT1 Bulk Data entry and the NLPARM Bulk Data entry must have a positive real value in the DT field. The DT value specifies the incremental time interval for creep analysis. The time unit in the CREEP entry used t o define the creep characteristics with coefficients a through g must be consistent with the DT time unit. The CREEP entry format is shown below with default values, followed by a brief description of each field and applicable remarks.
CREEP CREEP
MID
8
RT 0.
EXP 1.E9
FORM CRLAW
TIDKP
TIDCP
TIDCS
THRESH
TYPE 121
i
a
b
2.444
c
d
0.1072
e
f
0.1479
g
6.98536
7.03234
6.7339
3.
MID Identification number of a MAT1 entry (Integer
0.0. See Remark 1.
> 0).
RT Reference temperature (To)at which creep characteristics are defined (Real). Default =
EXP Temperaturedependent term, em, in the creep rate expression (0.0 < Real 5 1.0). Default = 1.OE9. See Remark 1. FORM Specifies the form of the input data defining creep characteristics: CRLAW (empirical creep law) or TABLE (tabular input data of creep model parameters). See Remark 2. TIDKP Identification number of a TABLES1 entry for specification of the creep model parameter K p ( a ) (Integer > 0). See Remark 3. TIDCP Identification number of a TABLES1 entry for specification of the creep model parameter C,(a) (Integer > 0). See Remark 3. TIDCS Identification number of a TABLES1 entry for specification of the creep model parameter C,(a) (Integer > 0). See Remark 3. THRESH Threshold limit for creep process corresponding t o the threshold strain (0.0 < Real < 1.OE3). Default = 1.OE5. Threshold stress under which creep does not occur is computed as THRESH*E, where E is the Young's Modulus. TYPE Specifies the identification number (threedigit integer) of the empirical creep law type (111 5 Integer 5 300). See Remark 4.
ag Coefficients of the empirical creep law specified in the TYPE field (Real). See Remark 4.
AH
Remarks:
1. Creep law coefficients (ag) are usually determined by least squares fit of experimental data obtained under a constant temperature. The reference temperature (To)at which creep behavior is characterized must be specified in the RT field if the temperature of the structure is different from this reference temperature. The unit of the temperature input (OF or OC) must be indicated in the PARAM entry TABS as follows:
PARAM, TABS, 273.16 PARAM, TABS, 459.69 (if Celsius is used) (if Fahrenheit is used)
When the correction for the temperature effect is required, the temperature distribution must be defined in the Bulk Data entries (TEMP, TEMPP1, and/or TEMPRB) which are selected by the Case Control command, TEMP(L0AD) = SID, within the subcase. From the thermodynamic consideration, the creep rate is expressed as
where A H is the free energy of activation R is the gas constant (= 1.98 cal/mole OK) T is the absolute temperature A is strainlsec per activation If the creep characteristics are defined at the temperature To (RT field), the creep rate at the temperature T is corrected by a factor
AH
where ( e  m ) is t o be specified in the EXP field. 2. If FORM = CRLAW, the creep law type and the coefficients (ag) must be specified in the CREEP continuation fields. If FORM = TABLE, the creep model parameters (Kp, Cp, and C,) are to be specified in the TABLES1 entries whose identification numbers appear in the TIDKP, TIDCP, and TIDCS fields, respectively. In this case, the CREEP continuation field should not be used. 3. The creep model parameters (K,, Cp, and C,) represent parameters in the uniaxial rheological model shown in Figure 6.1.3. Tabular entries (xi, in the TABLES1 entry will y;) (a;, Ii,;), (a;, Cpi), or (a;, C,;) for the input of 14, Cp, and C,, respectively. consist of In the case of a linear viscoelastic material, the parameter (hrp, C,, and C,) values are constant and two data points with an identical value should be specified for each parameter. The creep model parameters must have positive values as shown in Figures 6.1.46.1.6. If the table lookup results in a negative value, the value will be reset t o zero and a warning message (TABLE LOOKUP RESULTS IN NEGATIVE VALUE OF CREEP MODEL PARAMETER IN ELEMENT ID = ****) will be detected. 4. Two different classes of empirical creep laws are available at present. The first class of creep law is ~ " ( at) = A(u)[l  eR(u)t] h'(u)t ,
+
Oak Ridge National Laboratory recommends the parameters A(u), R(u), and K ( u ) in the form given in the following table: Parameter 40) R(a) K(u) Type 1 aub cExp(du) e[sinh(f u)]g Type 2 aExp(ba) cud eExp( f a )
Each of the three digits (e.g., 111, 121, etc.) in the TYPE field selects type 1 or 2 for the parameters A(u), R(u), and K ( u ) , respectively.
• he
second class of creep law (TYPE = 300) is expressed as
where the values b and d are limited t o the ranges
and 0.2
< d < 1.0
T h e coefficient g should be blank for creep law type xx2 material, and coefficients c, e, f, and g should be blank for creep law type 300. T h e coefficients (ag) are dependent on the structural units. Caution must be exercised t o make these units consistent with the rest of the input data.
5. Creep analysis requires an initial static solution a t t=O, which can be obtained by a subcase with an NLPARM on which DT=O.
Primary Elastic
Figure 6.1.3. Uniaxial Rheological Model.
Figure 6.1.4.
Creep Model Parameter I(, as a Function of Stress o.
Figure 6.1.5. Creep Model Parameter C as a Function of Stress u. ,
c s
Kipshours in3
a(ksi)
Figure 6.1.6. Creep Model Parameter C as a Function of Stress ,
0.
6.2
PLASTICITY
Two broad categories of material nonlinearity are nonlinear elastic analysis and plastic analysis. In both cases, the stressstrain relation is nonlinear; however, the unloading follows the stressstrain curve in the case of nonlinear elastic analysis, whereas elastic unloading takes place in plasticity. The plasticity is characterized by its path dependency. The strain is not a function of the current state of stress alone, but depends on the previous loading or stress history. This is exemplified by the simple case of zero stress, when permanent sets of differing magnitude can be established by varying histories in which the stress starts and finishes at zero.
6.2.1
Some Preliminaries
Some of the commonly used terms in plasticity [6.1, 6.21 are given below.
Yield Stress: Yield stress is usually measured as the value of stress which produces the
smallest measurable permanent strain.
Yield Criteria: In simple tension, there exists a yield point at which the material will begin
to deform plastically. However, if the stress state a t a point is not uniaxial but consists of stress components in different directions, a criterion is required t o define which combination of multiaxial stresses will cause yielding. Such criteria are called yield criteria. The first step of any plasticity analysis is to decide a yield criterion.
Yield Surface: Any yield criterion is expressible in the form
where f(a)is a function of the stress state and Y ( K is a function of the strain hardening. ) Geometric representation of the yield function is a surface in the stress space (al a;! as coordinates). This surface is called the yield surface.
Loading and Unloading: If the stress is changed so that the stress tends t o move out of the
elastic region, the process is called loading. If the stress value is changed such that the stress value moves into the elastic region, the process is called unloading.
Proportional Loading: Loading condition under which all the stress components increase
in the same proportion.
Strain Hardening: When the material undergoes plastic deformation or change in strain, the
yield surface expands and/or translates with increasing strain. This is known as strain hardening or work hardening and the material which does not exhibit strain hardening is called perfectlyplastic material.
Bauschinger Effect: When a plastically deformed specimen is unloaded, residual stresses on a microscopic scale remain and influence the plastic yielding for the different loading. If the previous strain was a uniform extension and the specimen is then reloaded in compression in the opposite direction, it is observed that yielding occurs a t a reduced stress. This is known as the Bauschinger effect.
The phenomenon of the Bauschinger effect can be described with reference t o Fig. 6.2.1, which shows an idealized stressstrain curve of a ductile material: first deformed by uniform tension, the load removed, and reloaded in compression. According t o the kinematic hardening model (which is one extreme viewpoint), it is assumed that the elastic unloading range will be double the initial yield stress. If the initial yield stress in tension is ay and the specimen is loaded up t o a stress a and unloaded, the plastic yielding will begin in compression at 1 given by a = a  20,. At the other extreme is the model that shows the hardening 2 1 stress a;! mechanism acting equally in tension and compression. Thus compressive yielding will occur 2 when a =  0 1 . This model shows no Bauschinger effect. This is the simplest of the theories to apply and is most frequently used. This behavior is called the isotropic hardening rule. Between these theories there are models that compromise the attributes of the kinematic and isotropic hardening for the onset of compressive yield after tensile hardening. This behavior can be modeled in MSCINASTRAN using the combined hardening option.
Figure 6.2.1. Representation of Hardening Rules
2. 2 23 23 1 = [( 01 .0 2 ) ~ 6 + (02  03)~ + (03 .2. Jl) exceeds a critical value. The von Mises criterion is most commonly used in plastic analysis of ductile materials.2.01)~]. this equation represents a right circular cylinder perpendicular to the octahedral ( n ) plane.2.3 . where J' 2 1 = O!. This cylinder is the von Mises yield surface. MohrCoulomb and DruckerPrager. . The criteria by MohrCoulomb and DruckerPrager are suitable for frictional materials such as soil and concrete. The Tresca criterion is more suitable for brittle materials.1) may be expressed in terms of the general stress state.O!.6. (6. (6. 6 stress in simple tension.2. where a is termed the effective stress. and..e.3)..2.e. i. The von Mises yield criterion states that yielding begins when the elastic distortion energy (or second invarient of deviatoric stress.61: von Mises. from Eqs. Tresca.2) Geometrically.2. as shown in Fig. i.6. yield stress in pure shear is 1 times the yield . 6. The lefthand side of Eq.2 Yield Criteria MSCINASTRAN provides four different yield criteria [6. i.e. this critical value is related to the yield stress (Y). (6. In simple tension. For pure shear of the plane stress case. V where Y is the yield stress of the material.1) and (6..
Figure 6.2.2. Geometric Representation of von Mises Yield Surface.
The T r e s c a yield c r i t e r i o n (sometimes referred t o as the maximum shear theory) assumes that yielding will occur when the maximum shear stress exceeds the value of the maximum shear stress occurring under simple tension, i.e.,
where a1 and a3 are principal stresses such that a s < a 2 < a l . Geometric representation of this equation is a hexagonal surface perpendicular t o the .rr plane (al 0 2 a3 = 0) in the stress space.
+ +
For frictional materials, the cohesion c and the internal friction angle 4 are the governing properties. The yield criterion for such materials is given by M o h r  C o u l o m b as
where
r
a, c
4
is the magnitude of the shear stress, is the normal stress, is the cohesion, and the angle of internal friction. (0 4
<
< 4) 5 '
Graphically Eq. (6.2.7) represents a straight line tangent t o the largest Mohr stress circle as shown in Fig. 6.2.3. In terms of principal stresses (0 > a1 2 a 2 _> as), Eq. (6.2.7) becomes
1 a1 (a1  as) cos 4 2 c  ( 2 2
which may be reduced t o
a1  a3
+ +  sin 4 ) tan 4 2
a 3
a 1
03
+ (al + as) sin q5 = 2c
cos 4
(6.2.9)
This may be reduced t o Tresca's criterion with
4 = 0 and Y = 2c.
An approximation t o the MohrCoulomb yield criterion was presented by Drucker and Prager, known as the D r u c k e r  P r a g e r yield criterion, which is a modification of the von Mises yield criterion, i.e., (6.2.10) 3a0, ( J : ) ~ ' ~= K
+
where
2 sin 4 &(3  sin 4) 6c cos 4 K = &(3  sin 4 ) This yield surface has the form of a circular cone coinciding with the outer apices of the MohrCoulomb hexagonal cone at any section. Rearranging terms of Eq. (6.2.10), the DruckerPrager yield function is given as
a=
6 sin 4 6c cos 4 am &( J ; ) ~ ' ~ > 3  sin 4  3  sin4
+
which is reduced to von Mises criterion for 4 = 0.
I

 (a, + a312
01
7
Figure 6.2.3. Mohr Circle Representation of t h e MohrCoulomb Yield Criterion.
6.2.3 Yield Function
Associated with the yield criteria, there exist yield functions in the form of
where and
f (a) is an effective stress, a function of the stress state Y ( P ) is a tensile yield stress, a function of the strain hardening.
.
Geometric representation of the yield function is a surface in the stress space (al  a2 a3 coordinates), which is called a yield surface. The von Mises yield surface is a circular cylinder and Tresca's is a hexagonal cylinder. The yield surface is reduced to the yield locus in the case of plane stress, as shown in Fig. 6.2.4 and 6.2.5 for von Mises and Tresca criteria. respectively. Both the MohrCoulomb and DruckerPrager yield functions represent conical surfaces as shown in Fig. 6.2.6.
Figure 6.2.4 Von Mises Yield Locus for Plane Stress.
02
u2  01 = uy
..........
. . . . . . . .. ... . . . .. ... ... . . . . . . . . . . . . . . . . . . . . . ....
uy
u1
..
7....L . .
0 , ,, ... ,, ...
,. '
,, .'
,, /'
, . '
,' . ,' .
4
1
/
1
02
= fly
.. .
..;
;u1
= uy
/" , ,/'
,,,,"by
,/ ,' ,
= uy +/
.
,/ ,'
j
:
I
," ,/'
, /
,.'
+
61
 02 = uy
+,'. !
m2
= uy
uy
!
Figure 6.2.5 Tresca's Yield Locus for Plane Stress.
6.2  7
DruckerPrager @ > 0
u r = q = q
J at
(a) von Mises and DruckerPrager
MohrCoulomb @ > 0 (1882) (1733)
Space
The stress function f ( a ) represents the effective stress or equivalent stress, which is equivalent to the uniaxial stress in tension. The value of f ( a ) can be computed as follows: i) ii) iii) For von Mises criterion For Tresca's criterion f(o) = For MohrCoulomb criterion
01
1 112 f(4 = fi(~2)

03
iv)
For DruckerPrager criterion f ( ~ .= 3 ) 6 sin 4  sin 4 a m
+ A(J;)~'~
The function Y ( P ) is a measure of the tensile yield stress. The value of Y ( P ) can be found from the userspecified stressstrain curve by an interpolation, i .e.,
with
c;
=
&k
Y k  E
where ( E ~ , is a data point such that Z 5 P < z:+~, Yl is the initial yield point, and ,G' = 0.) Yk) : 0.5 or 1. (for kinematic, combined, or isotropic hardening, respectively). Since Yl = 2c for frictional materials, it is obvious from Eqs. (6.2.9) and (6.2.11) that Eq. (6.2.16) must be modified by multiplying correction factors as follows: Y ( P ) = cosq5 Y(P) =
*
Y(P)
for MohrCoulomb for DruckerPrager
(6.2.18)
3
3 cos 4  sin 4
*
Y(P)
Notice that Y > 0.0, otherwise the program issues a user error message: "Implausible stressstrain curve in EID=xxxn.
6.2.4
Strain Hardening
The yield surfaces discussed previously correspond t o the yield surface of a material when yielding starts. For a material that strain hardens, the yield surface must change for continued straining. Such a change of the yield surface is governed by the hardening rules. The isotropic hardening theory, due to Hill, assumes that the loading surface expands uniformly about the origin in the stress space during plastic flow, maintaining the same shape, center and orientation as the original yield surface. Fig. 6.2.7 shows the initial and subsequent yield surfaces on a twodimensional plot when the stress state shifts from point 1 to point 2. Unloading and subsequent reloading in the reverse direction will result in yielding at the stress state represented by point 3. The path 23 will be elastic and 02 is equal to 03. The main drawback t o this theory is its inability to account for the Bauschinger effect exhibited by most materials. The equation for subsequent yield surfaces is given by
where Y ( P )is a measure of the expansion of the yield surface in the stress space, and the total effective plastic strain (3') a monotonically increasing scalar function during plastic is deformation. The kinematic hardening theory, due t o Prager, assumes that the loading surface translates as a rigid body in stress space during plastic deformation, maintaining the size, shape and orientation of the yield surface (see Fig. 6.2.8). Thus, this theory accounts for the Bauschinger effect. The yield function for kinematic hardening is expressed as
where a;j represents the translation of the center of the constant. The theory as set forth by Prager postulates that loading surface in the stress space occurs in the direction of at the instantaneous stress state. This relation is expressed
yield surface, and Y is assumed the incremental translation of the the exterior normal to the surface analytically by
where c is a scalar parameter called the hardening modulus, and d&Yj is the increment of plastic strain. However, inconsistencies arise when Prager's theory is applied in various subspaces of stress. In other words, the loading surface will not, in general, translate in the direction of the exterior normal in the stress subspace with the full nine dimensional stress space. To avoid this difficulty, Ziegler has proposed a modification of Prager's rule as
The scalar d p in Eq. (6.2.23) is determined from the condition that the stress state must remain on the translated loading surface during plastic deformation. Substituting Eq. (6.2.23) into
the total differential of Eq. (6.2.21),
and solving for d p ,
dp =
Z d c , ( o . . a . . 2 i
'3
a31
A theory that combines kinematic and isotropic work hardening states that
for which d p is found t o be
dp =
do
4HdP
a.f(~~.) (0 a ) aa
where H is the instantaneous slope of the uniaxial stressstrain curve and = 0.5. Eq. (6.2.25) reduces t o kinematic hardening when Y ( P ) a constant and t o isotropic hardening is when a;j = 0 and Y ( P )is monotonically increasing. The specification of the initial yield surface (by means of a choice of the yield criterion) defines the stress state at which plastic flow will begin. The specification of a hardening rule defines the stress state for subsequent yielding.
(g)
Figure 6.2.7. Isotropic Hardening.
Loadi ng Curve
Y i e l d Curve
Figure 6.2.8. Kinematic Hardening.
6.2.5
PrandtlReuss StressStrain Relations
One of the major differences between stressstrain relations in the elastic range and those in the plastic range is that strains in the plastic range are, in general, not uniquely determined by the stresses but are dependent on the past history as well. Due to the pathdependent nature of the plastic strains, increments of plastic strain are calculated by an incremental procedure and the total strains obtained by integration or summation. After Saint Venant proposed that the principal axes of the strain increment coincided with the axes of principal stress, a general relationship between components of the strain increment and stress was established by Levy and von Mises, independently [6.1]. LevyMises equations, however, use the total strain increments and are applicable only t o rigidplastic materials. LevyMises equations were modified t o allow applications for elastoplastic materials by Prandtl and Reuss. These equations are known as the PrandtlReuss stressstrain relations, i.e.,
where
1  [(ox uy)2 ( u y  uz)2 2 and
+
+
(0, 
+ 3 ( ~ ,+ ~ + 2
T ; ~
T,"~)
These stressstrain relations are consistent with the von Mises yield criterion. For other yield criteria, a general derivation follows.
6.2.6
Associated Flow Rule
Based on the condition that the principal axes of the plastic strainincrement must coincide with the principal stress axes, and defining a function of stress components ( a i j )as the plastic
potential function (Q), the plastic strainincrements can be expressed as
where dX is a Lagrange multiplier. For a socalled stable plastic material which has a stationary value of the plastic work done in a given increment, such a function (Q) exists and is assumed identical to the yield function (F),i.e.,
where F is given by Eq. (6.2.25). This is called the associated flow rule. The geometric implication of this relationship is that the plastic strainincrement vector is normal t o the yield surface, known as normality principle. During the incremental loading, the incremental strain is assumed t o be equal t o the sum of the incremental elastic and incremental plastic strains, i.e.,
But
{d&Ie= [D,]'{do)
with
[D,] being the elasticity matrix;
and upon substituting Eqs. (6.2.30) and (6.2.32) into (6.2.31), it is found that
{ d ~ = [D,]' { d a ) )
d + dX{ }F do
When plastic yield is in progress, the stresses are on the yield surface given by the yield function, which, when differentiated, provides the expression
where K (workhardening parameter) can be a;j or P , whichever case it may be. Eq. (6.2.34) may be conveniently expressed as
Then
dF {ao}
d o )  H*dX = 0
1 arc dX where H* has a physical implication which will be shown later.
8F H* = drc
After a few matrix manipulations t o eliminate {do) from Eqs. (6.2.33) and (6.2.35), the scalar multiplier dX can be obtained as
is reduced to where of dX from Eq. (6.2.33) gives
{s)
{ g } and the terms H* and 1%)are functions of {a}. Elimination ,
{do) = [DepI{d~) (6.2.38)
with the elastoplastic matrix defined as [6.5]
6.2.7
Generalized Effective Plastic Strain Increment
The effective plastic strain increment can be defined using the workhardening hypothesis. Workhardening is postulated as the amount of work done during plastic deformation, i.e.,
According t o the Euler's theorem, if the function f(aij) is homogeneous and of degree one, e.g.7 (6.2.41) f (Xaij) = X (aij) f then the following equality exists:
All four yield criteria, provided in MSCINASTRAN, satisfy Eq. (6.2.41). By virtue of Euler's theorem, Eq. (6.2.40) may be restated as
The stressstrain data from the uniaxial tension test (in xdirection) may be related t o Eq. (6.2.43) by considering the work done by the plastic deformation during the test, i.e.,
When the effective stress ( a ) corresponds to the stress under uniaxial tension (Y), the corresponding effective strain increment must be equivalent t o dX by equating corresponding terms of Eqs. (6.2.43) and (6.2.44). Hence,
and dX is defined by Eq. (6.2.37). In fact, it can be shown that dX in Eq. (6.2.37) is identical t o d P as defined in Eq. (6.2.28) with the von Mises yield function. It is necessary to clarify H* in Eq. (6.2.36) using Eq. (6.2.45). Combining Eqs. (6.2.30), (6.2.35) and (6.2.45), Eq. (6.2.36) can be restated as
It is revealed that H * is identical t o H which is the slope of the uniaxial stressstrain curve,
dY F.
This is a natural consequence of the workhardening, which should follow the simple tension test data regardless of the hardening rule.
6.2.8
Yield Function Derivatives
The derivatives of the yield function with respect t o stress components are frequently required in the computation of plastic strains. This frequent computation is conveniently simplified by adopting a unified approach to the various yield criteria. All the yield functions are expressed in terms of stress invariants [6.6], and the gradient vector is obtained by differentiating the yield function with respect t o these invariants first. This unified approach also facilitates t o avoid the difficulty of singularities in the Tresca and the MohrCoulomb yield criteria. The principal stresses can be expressed in terms of stress invariants as follows:
where
and
with
and
J; = aLaha:
+
~
T
~
~
T
+ a & ~ , +~ 2
~ ~ T
1
2
o~ ~
~ T ~
~
)
Substitution of this expression into Eqs. (6.2.13) and (6.2.14) results in the Tresca and MohrCoulomb yield criteria in terms of stress invariants:
f ( a ) = 2 (J;)'I2 cos 0
and
(6.2.13~)
f ( a ) = 2 a , sin 4
+ 2 ( J 2 )112[cos 0 1
1
(6.2.14~)
The gradient vector is computed by differentiating the stress function f ( a ) representing effective stress for each yield criterion as follows:
with
a a f ae  f  
a ~ ; aea~,
and
af af aJ;
a ~+ ;
.
af aeaJ;
ae
The gradient vectors of the stress invariants are found independent of yield criteria as follows:
1 =  < 1 1 1 0 0 0 >
= =
< a Ix a& o 2rXy 2rYz 2~~~ > :
2 < ( a & d  rYz +
J;
(&:

2 T~~
+ 3),
J::
The derivatives of the effective stress with respect t o the stress invariants are found as below: i) Tresca's criterion:
af 
a J;

& sin 0
J; cos 30
it is not possible t o compute af a J.2. a provision is made t o use the expressions for von Mises and 0 DruckerPrager yield criteria when 1 1 2 29'.sin $ (tan 36' . the general expressions derived so far must be modified with and E~ = = 0 . Plane Stress and Uniaxial Stress In the case of plane strain. This difficulty arises because the gradient vector cannot be uniquely defined at these corners.ii) von Mises' criterion: iii) MohrCoulomb criterion: darn f  'f  2 sin 4 cos 6' [(I + tan 6' tan 36') a aJ: af a J.9 Degenerate Cases: Plane Strain. @ + 1 . To prevent such difficulties. and f  a a J. representing corners of the Tresca and MohrCoulomb yield Notice that when 0 = surfaces.tan 6')] fi  sin 6' + sin $ cos 6' J. 6. cos 39 iv) DruckerPrager criterion: 30°.
aux auy au. The deviatoric stress invariants can be reduced t o r 1 J2 = .(060: + $). (oLo. + A&: thermal . The deviatoric stress invariants can be computed with a. = AE: and + Aef: # 0 Aef: = A&: . . all the stress components become zero except a.L 2 L 2.)..A&.Aei. The elasticity matrix is reduced to: In the case of uniaxial loading.f.L 2.OAT = A&.but Ae. deviatoric stress invariants become simply The 1 2 J2 = 1. which can {*jT {$jT {%jT < a: ah a: 2r.(ai2 2 + oL2 +  0:') + rXy 2 1 1 Jh = at (a. aTxy >.gz 3 and 1 J3 = 2a3: 3 27 .2o:r.  r2. J' > The elasticity matrix for the plane strain case is In the case of plane stress. + +). does not have to be stored as in the plane strain case.. > < (aha: = + $).. = 0. Notice that E.. The gradient vector be computed with ( 2 )will consist of four components < = = 1< 1 1 1 0 > 3 2. the equations can be reduced with but Ae.1 2 r.a.
* Yl. {anew}. for von Mises and Tresca for MohrCoulomb for DruckerPrager The elasticity matrix is reduced t o a scalar: 6. The procedure can be summarized in steps as follows: Step 1.The effective stresses become f ( a ) = ABS lax1 for von Mises and Tresca f ( a ) = ABS lax) + a.]. [De. {anew}. Initialize Zero = { a } = {a} E? = p. {A&} Output: Gew. i.10 Solution Algorithm for ElastoPlastic Material The material routine interfaces with element routines with the following 110 data: r Input: P. a The governing equation for the incremental process of elastoplastic deformation is as follows: with where H and {g)are functions of {a}.. Dep = De . and The computational process in the plasticity routine is depicted in the flow diagram in Fig. sin 4 + 2 sin 4 3 .sin 4 a x for MohrCoulomb for DruckerPrager f (0) = ABS lox1 The gradient vector can be reduced to a scalar.9.e. 6. {a}.2.2.{a}.
the material is on the yield surface. the process is elastic.5. Go t o Step 10 with {ali = {ao) and y = 0 . iv) Make corrections for the case of MohrCoulomb or DruckerPrager criterion. Estimate the stress increment using the elasticity matrix. If lFol 5 0. iii) Compute the current yield stress: Y = Yl + p (Yk . Return with: Step 7. combined.) P = 0. for kinematic. If Fl 5 0. ( + {Aae} Step 3. Translate the yield surface for kinematic or combined hardening: Step 4. or where Yl is the initial yield point and isotropic hardening.) 0. Determine whether the material is yielding.Step 2. Step 5. Find the current yield stress (Y) from the input stressstrain curve as follows: where E: = i) Increment lc until C 5 9 < : ' ii) Compute the current plasticity modulus:  %. or 1. Estimate the yield function: Step 6.E. respectively. Determine whether the plastic deformation is involved.Yl) + /3 Hk ( P . Then the trial stress is {at} = {.
If m indicates excessive increment size. Find the number of incremental steps required as follows: m = Int. Prior t o Version 66. Step 11. Check the error.Step 8. i) Compute { g J i a n d Hi based on {a). Update the stress components t o place the stress state on the yield point. set the flag t o activate the bisection in order t o reduce the load or time increment. a warning message was issued instead of bisection: "Excessive incremental load is applied in element ID=xxx. Bring the stress state t o the first yield point. The incremental stress (equivalent t o the equally divided subincrement of Ae) is where {Aa. and q ii) Find ." Step 12. Step 13. {a}. Step 9. if the material is elastic initially. Repeat the following loop m times.} = De {A&). If Fo < 0. which is based on where bisection is activated if M > 10. = (00) + 7 {AaeI where y =  Fo Fl  Fo Step 10. ( F s T R E s s F1 and * Y) + I 2 < m < LOOPCT where FSTRESS is a user parameter in the NLPARM data entry and is defaulted t o 0. material goes to plastic from the elastic state. Reduce the load increment by a factor of M for better solution. exit with a message "Effective stress is greater than yield stress in element ID =XXY" . If Fo > 0. The governing equation will be integrated by the Euler method of step by step integration in m subincrements.2.
+.p H AX. and A p = MAX(&. 0.0) {Aa).. = {Au)  AX. = MAX(AX.) ii) If 161 FSTRESS * Y in the converging state. ($1. "Error exceeded xx% of current yield stress in element ID=xxx".+1 = {a).) . . = Ap{a). < 5 x set the unloading flag up (notice that AX = 0 by the end of this loop) and eliminate negative AX. {a). EXIT with a fatal error message. {Au}. vii) Update the stress and the plastic strain and { E ~ )= { E ~ ) + AX. () 2: (0).iii) Compute iv) If AX.. i. . iii) If 161 < 0. AX.Y(Z:+.{Act}. ignore the error and jump t o Step 15. Step 14. = {Aa). +' {Aa). . 0. vi) If the hardening rule is not isotropic Ap = {%IT {Au).e. {d). Error Correction step i) Compute the error: 6 = F3 = f (a.0) v) Compute the stress subincrement {Au).
return with the elastoplastic tangential matrix: . Otherwise. S Step 16. Offset the center of the yield surface.+. make a correction as follows [6.e. the material is assumed to be unloaded.7]: {a>.+m Y Step 1 5 . return with the elastic tangential matrix. Since the material becomes elas tic during unloading.iv) If S 2 0.. Step 17. If the unloading flag is up. = { ~ } i + m. i.{a>.
7= 0 Yes Fatal Error I L i Number of incremental steps fi I = Int( 1' FSTAESSW F1 4 Elastic to Plastic Iransition <a11 = <a..... = E {<>I = {<Il DIP = DE Estimated stress increment {ME) DE{&€> = Trial stress {a?) = {a) + {ME> r. LOOPCI) Incremental Stress 1 {ha> = 4 1 ...) Fl = f(ai) Y Y I Return I I Haterial on Yield Surface {@I = {a..'"..#\.2......... Find current yield stress Y I Compute Fe = f(o...Fe I 1 I I I  I I I I I I No ' t 2 5 m ! Min(i0...r){&r> m I I I I I I Uarning Message ! 1 I I B I I ..}................. ............'........> + r{hr> Fe with7 = Fi .9a Flow Diagram for ElastoPlastic Material (NMEP) .................. I Figure 6.Initialize Zero = i~'* Y i p P €.....
Cofipute {a).9b Flow Diagram for ElastoPlastic Material (NMEP) . = 0 I Stress Incre~ent I I h Update Figure 6. and HI w Find {dl. :DE{~)I Compute AXI : {a):{&) HI + {){) a:dI  Yes + Set Unloading Flag AX.2. = (3.Based on {a} and €.
) lbl FSlRESSwY and C N = O Ul ) Fatal Error Offset yield center {0II+"= {o}~...2 .+ [ Yes I Figure 6 ..4 = Compute error P F3 = f(gI+. 9 ~ Flow Diagram for ElastoPlastic Material (NMEP) 6.27 .) .Y(EI+. 2 .
shown in Fig. = ..e.f3 approaches infinity. The strain components in simple tension for the elasticplastic deformation can be decomposed into elastic and plastic parts. and programming errors with a wide variety of problems in MSC/NASTRAN. Pma. The lateral dimension of the specimen shrinks during the simple tension test as in the elastic domain.e.2.11 Verification and Validation The elastoplastic material capabilities were verified with respect t o the algorithm. P has a finite value limited by f material properties (E and H ) .41.f3 with a value of v for .2. in width direction The apparent Poisson's ratio then becomes with The variation of v* with respect to P is plotted in Fig. However. 2D and 3D) have been tested and known solutions have been reproduced [6. 6. &a = E: + E$ VE.6. 6.5~: . in tension direction and E. Some of the example problems of interest are illustrated below: Uniaxial Loading and Apparent Poisson's Ratio Experiments have shown that the volume of material does not change due to plastic strain. ..10(a).3.10(b). All types of elements (1D. i.5 as . The Poisson's ratio is defined as v = E./E. It is shown that the apparent Poisson's ratio v* is a continuous and smooth function of .2.0.2. i.5 due to plastic strain.10(b).5 in the PrandtlReuss stressstrain relations. If the Poisson's ratio is to be measured from the elasticplastic deformation (termed apparent Poisson's ratio). a transition would be observed from an initially elastic (v) t o a smeared value with 0. 6. 6.3 = 0 and asymptotic t o 0. . accuracy. = *+m lim H(P) L l A finite element simulation of a simple tension test by MSCINASTRAN resulted in data points circled in Fig. This incompressibility is signified by a Poisson's ratio of 0.
MSC/NASTRAN Data Points : (3 i C F J CL . i v E last i c Strain (b) Apparent Poisson's Ratio Figure 6.10 Uniaxial Loading and Apparent Poisson's Ratio . . '9 I I I I I I I 1 0 2 4 6 8 10 12 14 .2.I 6 Ratio of P l a s t .(a) Uniaxial Tension Specimen U ~LDu n Analyticd +' 0 iu .
Suppose a material has a low initial yield stress and a high rate of hardening. In order t o avoid such an anomaly. involving successive loading. as expected. makes the problem more complicated due t o a phenomenon known as the Bauschinger effect. The cyclic loading. such a material does not exist. A finite element simulation of the cyclic loading has been performed with a thin plate specimen subjected t o repetitive uniaxial tension and compression. i.e.2.11(a) using the isotropic. unloading and reloading. Note that not all the flexure points are represented in the plot because of discrete solution points. kinematic and combined hardening rules. an identical modulus is resumed upon reloading from the previously loaded point.Cyclic Loading and Hysteresis If the stress state is changed such that the stress value moves into the elastic region. as some materials may possess. The combined hardening rule is complementary for such cases. During the first tensile loading phase shown in the first quadrant (0 to A). the reloading in compression could occur while the applied stress is still in tension as shown in Fig. The combined hardening curve falls in between the isotropic and kinematic hardening curves. In reality. kinematic hardening only describes a perfect kinematic hardening material for which the size of the yield surface does not change. The loading cycle was controlled by strain using the enforced displacements. .2. the initial yield point may have t o be artificially elevated. However. the responses for the three hardening rules are identical t o the original strainstrain curve specified in the input data. the process is called unloading. Kinematic hardening has an advantage over isotropic hardening for the Bauschinger effect. The plasticity modulus is kept continuous upon reloading for all three hardening rules.. This behavior is physically unrealizable. The difficulty of such problems lies in the fact that the Bauschinger effect is known as a qualitative and not a quantitative description. If kinematic hardening is used t o model such a material exactly. while the isotropic hardening shows a monotonic increase in the yield stress. The solution process progressed without any difficulty caused by unloading. 6.11(b). The stressstrain responses of the material obeying the von Mises yield criterion are shown in Fig. 6. The kinematic hardening exhibits the Bauschinger effect upon reloading in the reverse direction.
2.~ ~ i n el a += l + I ~ J L ~ U lc 17ernat li Harder l n g t i o r d e r ll.11 StrainControlled Cyclic Loading .y m H a r d e r incj I ( a ) Stress Strain Hysteresis (b) Anomaly in Kinematic Hardening Figure 6.u. ( k s i ) *~ .
1 Theoretical Basis The nonlinear elastic capability in MSCJNASTRAN was designed t o satisfy the equivalence of the deformation work per unit volume in the simple tension t o the strain energy per unit volume (conservation of energy). for which only the simple tension stressstrain data are available. i. (6. modified in Version 65. some of the constitutive relations may be violated in the multiaxial stress cases.10]. (6. This design is considered adequate for the multiaxial deformation if the strain remains in the small range. Intrinsic deficiency due t o oversimplification of the theory will introduce errors.3.6.3. stresses may be expressed in terms of total strains.3) into Eq. (6. but the error may not be noticeable unless large strains are applied in the multiaxial deformation.4) The tangential matrix for such material may be obtained by differentiating Eq. The current nonlinear elastic capability.3. For large strain applications the generalized hyperelastic material based on the classical theory of finite elasticity is being planned to be implemented.1). It was further assumed that the effective strain (E) may be defined by From the total differential of Eq. i.2).e.3.3 NONLINEAR ELASTICITY The nonlinear elastic capability was developed for Version 62 to predict the multiaxial stressstrain behavior for the nonlinear elastic material..3. Nevertheless. while the work done for deformation may be defined by a stressstrain curve in simple tension. i. Consequently. 6. (6.4). we obtain Substituting Eq.e.3. this capability can be justified by stating that the design is not intended for the largestrain deformation.e.. the theory is not based on the classical theory of finite elasticity [6. 0 ( 0 ) = EIDeI{~) (6. However. can accommodate bilateral properties (uniaxial tensioncompression) accurately for the uniaxial deformations. ... The theory and algorithm are adequate to trace the stressstrain curve accurately for the uniaxial loading cases.
the new strain state is computed by Step 2... in the TABLES1 entry along with MATl and MATS1 entries. The effective strain ( 2 ) is computed based on {E).v2 + E: + ~ V & X E ~ I~xy] + 1v 2 for plane stress + 2 for 3D Solid Step 3. Upon entry t o the subroutine NMEVD. The new stress state is determined by Step 5. which calls the subroutine NMEVD for the nonlinear material. The effective stress (a) is determined by lookingup the userspecified stressstrain curve for E. by where or g2 1 [E: 1 . The tangential matrix is determined by .3.6. The element routine calls the material routine driver NMATD.2 Solution Algorithm Users specify nonlinear stressstrain curve. Young's modulus and the Poisson's ratio are available from MATl entry. a(<). Step 4. The material routine interfaces with element routines with the following data: The computational procedure is described below: Step 1.
Some method of interpolation or extrapolation is required t o predict the effective stress for the general stress state using two known data points. E.e.. < 6. exhibit appreciably different behavior in compression from that in tension even in the small strain range.1. Some materials..for which $$ is the slope at E.e. .).3. Hydrostatic tension and compression cases will impose lower and upper bounds for extrapolation.3. namely the effective stress for uniaxial tension (at) and the effective stress for uniaxial compression (a. E= E= E.3 Adaptation of Uniaxial Compression StressStrain Curve Until Version 64. i.e. it seems appropriate t o use the first stress invariant. ( g )should be interpolated or extrapolated in the same man . where (xk.i.1 for E 2 x.3. There are two known data points as shown in Fig. the effective stressstrain curve is determined only by the uniaxial tension data in two and three dimensional cases.. for uniaxial tension/compression Il = 0 for pure shear Il = 3p for hydrostatic pressure The instantaneous modulus ner. For uniaxial loading. It is noted that k = l for E < xl and k = k . the magnitude of the strain in that direction becomes the effective strain. i. i... Considering that the pure shear is in the midway between simple tension and simple compression.yk) is the kth data point in the TABLES1 entry and k is determined such that xk E < xk+1.2. 6.. for uniaxial tension in x for uniaxial compression in x We need to find the effective stress ( a ) corresponding t o Z. Il = a. The uniaxial compression data will be ignored if provided for two and three dimensional elements. ..e. The first stress invariant ( I l ) is adopted for interpolation/extrapolation. 6. A typical case of such material is shown is Fig. however.
... ....4 Computational Procedure for Bilateral StressStrain Relations The new stress state is proportional in magnitude to the effective stress determined as follows: 1.. 0............e.......002 0.( o Monotonic . 24 ..1..which should be (a....)based on {o...o MonotonicCompression o CyclicCompression 40 ... StressStrain Curves for Gray Cast Iron 6..........006 " . ............ i...)....... .3..3....... ..........008 STRAIN Figure 6.........004 0......Tension o Cyclic .............i STRESS (ksi) 32 .. !... Compute the effective stress (a).. 0 0.Tension D Fracture 0 .
i. and r . The hydrostatic load). i. (E).. Therefore.3.. It would be implausible to process a large value of r (such is the case with a 1.and i. Determine ($$) based on ($$)t. < 4.2. Compute the instantaneous slope at E for tension. Look up the userspecified stressstrain curve in the TABLES1 entry and determine and a. i.. 3.e..for plane stress for plane strain 2. Determine the ratio (r) by normalizing Il by a. and at at B.yj xj+l ..e. . i.e.+1 1. < E < xj+l 3. for xi where (x. Compute the instantaneous solpe at (E) for compression. i. 6% (I .): I = ox 1 + + oz +y where a. y.a.2. Compute the first invariant of (0. 1 5 r value will be reset to the limit ( r = &1 ) if r lies outside the range.. = 0 for plane stress.x j for x j T... For the tangent matrix. r will be confined t o a plausible range. where r signifies the relative distance from the midpoint of a.t = Yi+l .) is the ith data point in the TABLES1 entry. aa ( z ) c = yj+l .e.. as shown in Fig 6. Determine a based on at.e. 2.Yi x.+1 .e. at 5. the instantaneous modulus same ratio ( r ) as follows: (g) be determined using the should 5 B < x..
3. TABLES1 INPUT for StressStrain 6.Figure 6.2.6 .3 .
If the plastic deformation is coupled with creep. The formulation is based on the stepbystep time integration of the KelvinMaxwell rheological model with nonconstant parameters. The recoverable portion of the creep deformation is called primary creep and the nonrecoverable portion. When the creep characteristics are specified in terms of empirical creep laws. using a generalized viscoelastic model [6.4. The concept of a rheological model is extended t o the multiaxial stresses by the PrandtlReuss stressstrain relationship. is introduced to process the creep behavior coupled with elastoplastic deformation. For a generalization of the viscoelastic material behavior. similar to necking in plasticity. The tertiary creep. secondary creep and the tertiary creep to rupture. the program converts the empirical formula t o the corresponding rheological model. A number of empirical creep laws. the creep behavior is coupled with plastic deformation and the variable temperature effect is included.4 CREEP A N D VISCOELASTICITY A creep analysis capability. which is beyond the scope of this handbook. recommended by the Oak Ridge National Laboratory (ORNL). the creep phenomenon is best observed in the uniaxial test under constant load and the relaxation test under constant strain a t constant temperature. the algorithm will seek a solution in two distinct steps.6. The stepbystep integration is performed assuming that the strainrate varies linearly during each time increment while the rheological model parameters remain constant for a short time interval. A specimen subjected t o a constant uniaxial tension at an elevated temperature exhibits three distinct phases in a time frame: primary creep. is considered as a localized instability phenomenon. If the specimen is unloaded after some creep deformation. . secondary creep. the elastic strain is immediately recovered and a portion of the creep strain is gradually recovered.11]. are provided along with options for general tabular input of the rheological model parameters as functions of effective stress. the rheological model parameters are treated as nonlinear functions of the effective stress and the temperature. Elaborating on the KelvinMaxwell rheological creep model developed by Badani [6. from which the tangential stiffness matrix is formed for Newton's iteration.12]. The Kelvin element represents the primary creep behavior and the Maxwell element describes the secondary creep. Various choices of empirical creep laws are available and small variations in temperature are allowed as implemented in MSCINASTRAN. 6.1 Introduction At the macroscopic level.
with where t denotes time and parameters Kp. and C. The creep hardening (and softening) effects. 6. however. the creep deformation is an integrated effect of dislocations of the crystal structure primarily due t o thermal activation and stress. are identified in Fig.4. i. and CP (To) Cp(T) = Fc . Due t o this virtue.2 Effects of Stress and Temperature An analytical solution t o the KelvinMaxwell model subjected to a constant stress (a)is given by c . the creep model parameters measured at a reference temperature (To) can be used t o compute the creep strain rate a t the temperature T in the vicinity of To.AH/RT (6. in general. On this basis.e.u (6. Cp.4. For a varying stress case.1. assumed t o be exhibited by the primary creep.. The creep model parameters are corrected as Cs(T) = where c (To) s F. wherein creep strain rates are expressed in terms of primary creep strain rather than time. an instantaneous strainrate should be employed to have the creep hardening effects properly accounted. The creep strain rate is. a function of absolute temperature. where Notice that the creep strain rate reverses the sign with proper effects of the cummulative primary creep when the sign of the stress reverses.4.1) Etotal .3) where AH is the energy of activation. are accounted for by using the creep hardening law. R is the gas constant. T is the absolute temperature and A is in strainlunit time.4. allowing small variations in the ambient temperature.6. przmary c.t t EC . the stress reversal effects are properly taken into account without any special provision in the algorithm. Microscopically. The effects of temperature at the microscopic level can be quantified as i c = A.
6.4.3 Equillibrium of a KelvinMaxwell Model for a Stress Component Suppose that the KelvinMaxwell model is applied t o a typical component of a deviatoric stressstrain pair.1.4. determined by The pseudoincremental stress represents the change in deviatoric stress component due t o creep relaxation. This value of As' can be converted t o the equivalent psuedoincremental . the state equilibrium equation of the model a t any instant is expressed as: [C] {A&} + [Ii] {Ae} = {As} where (6.4. (6. el).4. 6.7) for Ael gives kc Ael = As' + As1 where As' is a pseudoincremental stress.4. Referring t o Fig. (6. denoted by (Asl. the strainrate increment can be expressed as n by which Eq.5) and AS}^ = < Asl 0 > Introducing a linearly varying strainrate during the time interval.5) is reduced t o Defining the stiffness of the primary and secondary creep elements by kl = K p + 2Cp At and k2 = 2cs  At the equivalent creep stiffness (LC)for a deviatoric stressstrain pair can be determined by Solving Eq.
1. i.i2) + s i z ] bl in which the first and the second terms are contributions of the Maxwell and the Kelvin elements..4. 0 u 2 r x y 2 ~ y Z 2rzx >. Notice that as At approaches zero.4.11) reduces to Ae' cl At il = At iio.10) are identical to the ordinary strain components.{ s ) 2 a deviatoric stresses { s l T = < 0. and the effective stress (a) and the effective creep strain rate ( p )are defined analogously t o those in plasticity with von Mises yield criterion.e. which represents the change in deviatoric strain component due to creep.. (6.4. Primary Creep (K. which makes it easier to comprehend Eq.4 Adaptation of Rheological Model to Multiaxial Stress State The concept of the rheological model for a stressstrain pair is extended to the general multiaxial stresses with the aid of effective stress and effective strainrate by adopting the PrandtlReuss stressstrain relationship.) Secondary Creep (K2) r=Je=rel * b o(t) or AS. respectively.4. By virtue of creep volume constancy. Hence the pseudoincremental strain can be expressed as AE' = As' = kc 2 .11). . (6. (6.. deviatoric strains (Ae') and the strain rates (11 and i 2 ) in Eq. Rheological model 6. Figure 6.4. Eq.strain (Ae'). {iC) = where creep strain rates 32 .
which can be related t o all the deviatoric stress components and the corresponding creep strain rates. However..4.4. and c .14) {Ace Aec) = [DL' D. ) based on the effective stress. In terms of effective creep model parameters. Eqs.]{Ae .Implied by this relationship. (6. and The creep strain rates in Eq. we obtain {ACT}= [Be. (6.14) and (6.11a) is reduced to In the absence of plastic deformation. the total strain increment must be corrected with a pseudoincremental strain {A&'}. in terms of deviatoric stresses where Then the pseudoincremental strain in Eq.e. respectively.4. (6. Combining Eqs.16) where the elasticcreep tangent matrix may be conveniently obtained for an isotropic material by K + $kec K .4. there is a unique set of rheological parameters (Kp.11) may be rewritten as follows: cp.' k ec K 3 0 0 0 itec K+ikec Kikec [Dec] = I o o o l k ec 2 c o 0 o o 0 0 SYM L ikec  .15). the total strain increment (elastic and creep) may be expressed as (6. i.8) and (6.4.4. (6.4..4.A&'} (6. i.e.'] {ACT} + + where De and Dc are material matrices for elasticity and creep.2) may be rewritten likewise.
' (6.4. For this reason.4.and the function ( f ) defining effective stress (a).4.e. Eq. 1 2G+  1 kc 6.20) with Ax = {gjTDeC{A&. therefore.AE') = A P > O H + {gjT Dec { g ) where the effective plastic strain increment is denoted by A P The creep deformation tends to relax the stress gradient in the absence of further increments in external loads. defined by and kec the stiffness of the elasticcreep component 1   kec where G denotes the shear modulus.4.' + D.16). it solves for the incremental stress components with an elasticcreep material as in Eq. The plastic strain increment may be obtained by { A E ~ } D {Aa) = .. the elastoplasticcreep stressstrain relations are established as [DL' + D.18) into Eq.Introducing da Eq. In the creepdominant process.4. (6. (6. if the new stress state ( a = a. i.19) from which the stress increment {Aa) can be obtained. + .4.'] {Ao) = {AE  AE') (6.4.with li being a bulk modulus (K = ).14). First.18) where da with plasticity modulus H = . However. plastic deformation can be induced only by creep deformation to alleviate stresses in the neighboring material. (6.19) may be rearranged as a (6. Then.ld Aa) exceeds the current yield stress.5 Coupling of Plasticity When the plastic deformation is involved in the creep analysis. (6. the plastic strain increment should be included in the total strain increment.4. the material routine employs a solution scheme which seeks a solution in two distinct steps when the plastic deformation is coupled with creep.
.4.with F being a yield function.= H + aTd Compute Aa = Aa . The algorithm for this process is depicted by a flow diagram in Fig.A l d Comctlon for klnematlc hardening Update a = a + Aa 6P = 69 + M Error Correct~onlf requrrcd :{1 I I I I I I I I I  I I I I I I I I I I Figure 6.. based on ri (Process Bulk Data cards CREEP TABLES I I c. (6. . Comp I Interpolate Fp.' + AE" Compute 1 Update ( 4 .2.4.2. Flow Diagram for Creep Process .= [ G )+ { A G . irst Yes Plast~c~ty Routine I I I I I Compute a = I I I I I I I I I I I I_______I Compute ( d ) = D. C. 6. a correction is made on a previously obtained incremental stress based on Eq.4. j Compute A&' Update E' = E . I Form [ D ..{a} aTAa Compute b.20) and the plastic strains are computed. ] matrix i No 1 Solve elasticcreep problem for j Ao) Compute A&'(A&') Compute Ad Store [Dec] 1 Compute (A&.
The nonproportional loading cases including neutral loading have been examined and validated by controlling either the stress or strain path in twodimensional problems with the elastoplastic material. All the potentially nonlinear elements (ROD. The plasticity routine again subdivides the given increment into smaller increments if the increment in the yield stress exceeds a certain fraction of the current yield stress. t ) = A(a) [1 . The material routines are implemented to adopt an adaptive subincremental scheme by which the computations in material routines are carried out in small increments within a global incremental step.eR(u)t] + Ir'(0)t Type 2 a Exp (ba) cad for all combinations of types 1 and 2 for A. The creep routine subdivides an incremental time step such that the effective creep strain increments and effective stress increments are bounded by a fraction of the total creep strain and effective stress. The fraction which defines the bounds is a user specified parameter with a default value of 0. the creep routine calls the plasticity routine.6 Implementation The quasistatic nonlinear analysis capability for the generalized viscoelastic model coupled with the plastic deformation is implemented into the generalpurpose nonlinear static analysis method in MSCINASTRAN.4. QUAD4. and K . HEXA. TRIA3. If the creep behavior is prescribed by a creep law. PENTA and TETRA) are applicable t o creep analysis. Users are allowed t o either specify an empirical creep law or provide direct input of rheological parameter values as functions of the effective stress. If the plastic deformation occurs during the creep process. (6.6. creep law parameters are converted t o rheological parameters whenever a new stress state is computed for the creep analysis. R. the nonproportional loading or unloading does not present any difficulty.4. but not all the nonlinear elements in the model need to be composed of creep material. The empirical creep laws provided in MSCINASTRAN are in the following equation forms with user specified coefficients (ag): and c C ( a . BEAM. By virtue of adaptive subincremental scheme within the material routines.4) if the operating temperature is different from the temperature at which creep behavior is characterized. The solution method employs an adaptive Newton's iteration with accelerated convergence schemes such as the BFGS updates and the line search process. respectively. Rheological parameters are corrected for changes in temperature based on Eq.2. A nonlinear static analysis can readily be converted t o the quasistatic analysis and vice versa by adding or subtracting a few related data. which are expressed as Parameter 40) R(a) WJ) Type 1 aab c Exp (do)  e [sinh (f a)]g e EXP(f0 ) .
However.4. The material routine computes the elasticcreep tangential matrix for the formation of a global stiffness matrix.3. where 6. and programming errors with a wide variety of problems in MSC/NASTRAN. Various creep laws in the form of Eq. is applied at the operating temperature of 1200°F with corrections for variable temperature. defining the creep behavior at llOO°F. (6. which is used for the iteration combined with the BFGS update and line searches. (6. accuracy.4.4.21) to the rheological model is not so simple due to lack of similarity. equivalent rheological parameters are computed by matching the total creep strain and their first and second derivatives of Eqs.22) t o the rheological model is performed by simply matching equivalent terms in Eq. however.1% in the creep strain a t the end of 70 steps.4. All types of elements (ID.4. i.7 Verification and Validation All aspects of creep capabilities were verified with respect t o the algorithm.4. (6. Creep laws in Eq. The total creep strain increment can be recovered by subtracting the total {Ae) in Eq.e. (6. The time increment should be selected so that the strain and/or stress do not change excessively in a single step. (6. 6.2). However.e. i. and {AeP) from The primary creep strain increment is saved in the database to take into account creep hardening/softening effects.21).Conversion of creep laws in the form of Eq. There are two different creep laws (types 111 and 121) for the same material (type 304 stainless steel) established a t different temperatures (llOO°F And 1200°F) as shown in Fig. (6. The creep law type 111.4. conversion of the popular and classical creep law in Eq. 2D and 3D) have been tested and known solutions have been reproduced. An equilibrium condition is achieved within the convergence tolerance at every timeincrement. (6.21). The numerical solution scheme employed in the creep formulation is unconditionally stable owing to the implicit method adopted for integration.4.1) and (6. a limit should be placed on the time increment to achieve an accurate solution.22) and the direct input of rheological parameters reproduced analytical solutions accurately with errors less than 0. produced a cumulative error of 1519% in the creep strain a t the end of 70 steps. In this case.4... . because of the mismatch between the empirical formula and the rheological model.4. The creep analysis capability under variable temperatures is verified using empirical creep laws recommended by ORNL.19).
Figure 6.4. Coupling of the plastic deformation with creep is verified by reproducing the isochronous stressstrain curve for stainless steel. as shown in Fig. type 304.5 compares relaxation predictions by various methods. 6. MSCINASTRAN reproduced the solution very accurately using nine HEXA elements with axisymmetric and plane strain boundary conditions. The effects of the creep hardeninglsoftening were exhibited properly. They employed an incompressible material with a simple empirical formula which accounts only for secondary creep.stress reversal was thoroughly investigated. which predicts proper creep behavior at 1200°F. Nevertheless this feature is considered useful and essential.4. The creep response of various elements to the. For a model five times coarser. The creep behavior is manifested in the relaxation process under constant strain. However. Results from MSCINASTRAN analysis fall between experimental data and the closedform solution. It is demonstrated that the current method is superior t o the initialstrain method. 6.13] under an internal pressure of 445 psi with a material obeying an empirical creep law in the form of A finite element model of the same pressure vessel was analyzed by MSCJNASTRAN using 72 solid elements with 355 active DOFs.4. Fig. The accuracy of the varying temperature case was found satisfactory only for a short period of time or for small temperature variations.13]. to which an analytical solution exists. The data points are obtained in 15 steps to the creep time of 100 hours.13]. The solution t o the creep behavior of an infinitely long thickwalled cylinder subjected t o internal pressure was presented by Greenbaum and Rubinstein [6. The creep behavior of a thickwalled pressure vessel with a flatend closure was also analyzed by Greenbaum and Rubinstein [6.6 shows stress contours at t=3 hours. the convergence occasionally required a stiffness matrix update and a smaller time increment a t the onset of stress reversal. .The results are compared with the creep law type 121.4. this solution represents an excellent agreement with that in Reference [6.
Strain Cur .lriablr Temperature 24 X 4 Original Material MSCINASTRAN ASME Code Prediction I 0 0 0 1 I I I a 0.6.4.4 Figure 6. Creep Capability Verification for \..3.4.0 2.8 1.6 TOTAL STRAIN (Oh) 2.4 0.4. Isochronous Stress .2 1. 0   Creep Law 121 at 1200°F X Y Creep Law 1 11 Defined at 1 1OO°F 0 100 200 300 CREEP TIME (HOURS) 400 500 Figure 6. 5.
700..4.. Level I .. Note: Stress Units in ksi Figure 6...... Level E ..MSCINASTRAN ......0 .......Badani Experimental 0 . Level F .. Closed Form initial Strain Method 0 20 40 60 80 100 TIME (HOURS) Figure 6.......... Level H ......... 1000..........400... Effective Stress Distribution after 3 hours of Creep ..... 800.6............5.... 600.. Level J ..... 900.4. 500.. Level G ..... Cornparision of Creep Relaxation Predictions Level D ...
and 6. instead of a kinematic hardening rule as employed by Levy [6. respectively.9(b).8 NozzletoSpherical Shell Attachment as Benchmark Problem The HighTemperature Structural Design (HTSD) program of ORNL was instigated to develop design methodologies for breeder reactor structures. 6. ~ where the stress ( a ) and the time ( t )are measured in ksi and hour.4.10720) K ( a ) = 6. are compared with data in References [6.221 as shown in Figs.141 and analytical [6.22) with ORNL recommended values for parameters as follows: A(a) = 6.15. The model depicted in Fig.985 x a2. respectively. It is noted that MSCINASTRAN used an isotropic. (6.4.4. a nozzletospherical shell attachment model was presented as a benchmark problem with extensive experimental 16. 6. Under the ORNL program. which is analyzed by MSCJNASTRAN with elastoplasticcreep material and geometric nonlinear effects.9(a) and 6.6. The creep characteristics are defined by a creep law in the form of Eq.4. 6.444 R ( a ) = 7. where the plastic deformation is significant.10(a) and 6. These results demonstrate proficient capabilities of MSCJNASTRAN for a complex problem with combined nonlinearities. made of type 304 stainless steel.4. 1 4 7 9 a ) ] ~ .4.7 represents this component using 1552 DOFs.15.032 x low4Exp (0.73 x lo' [sinh ( 0 . Figs.161 results.6.10(b) show effectivestress contours upon unloading a t t=2546 and t=3400 hours. Typical results at selected points. 6.4. .4.15].14. 6. subjected t o successive cycles of loading with internal pressure and end moment at llOO°F.8. The model represents a liquidmetal fast breeder reacter component.
7 Finite element model by hfSC/NASTRAN 10 Experiment MSC/NASTRAN . Nozzle Radius: 1.4.33 in.03 in.8 Total circumferential strain at gauge location during pressure loading . TIME (HOURS) Figure 6. Nozzle Thickness:0.984 in.112 in.06 in. Fillet Radius: 1.425 in. Pressure Loading.4. 304 GRIDS 264 QUAD4s MSC/GRASP ( UAX 4 25JAN85 16:04:26 Figure 6...1 NOZZLETOSPHERICAL SHELL m Overall Height: 24. Shell Thickness: 0.. Sphere Radius: 12.
I u 1..1.5.26 .0 . 1 I M Experimental data exh~brted dr~ft I I .9(a) Total circumferential strain at gauge location (1) during moment loading 1800 2000 2200 2400 2600 TlME 2800 3000 3200 3400 3600 (HOURS) Figure 6.4..Levy 1 1Q ?a f I I s Unloaded I0 / 4 c. ' Moment Load I I 1 I I 20.9(b) Total circumferential strain a t gauge location (2) during moment loading .0 I 1800 2000 2200 2400 2600 2800 3000 3200 3400 3 1 TlME (HOURS) Figure 6.0 0.4..MSCINASTRAN .0 1  ) /((  r4' I .
4.4.10(a) Effective stress contour plot at t=2546 hours upon loading Numbers Represent Stress in ksi Figure 6.11(b) Effective stress contour plot a t t=3400 hours upon loading .Figure 6.
is used in Eq. To look up the material properties at temperature To for linear and upstream superelements (these properties remain unchanged through the entire analysis).5. (3. In the linear elastic case. With this capability.4) as 1 a(E. To.5. Eq.T) {a} = E z [De(T)]{€ .5.2 User Interface The thermalstress free temperature distribution. 6.3.6. (6. The thermal strain is computed by where T TREF T o a(T) = current temperature = reference temperature at which coefficient thermal expansion is measured = stress free temperature (initial temperature) = coefficient of thermal expansion The total stress and total strain relations are modified from Eq.ET}. the temperature dependent material properties for linear elastic and nonlinear elastic materials are updated as the temperature distribution in the structure changes.1 Theoretical Basis Thermoelasticity in MSC/NASTRAN is an extension of nonlinear elasticity as described in Section 6. The initial temperature.1) and other parts of the computation. (6.5 THERMOELASTICITY Thermoelastic capability was implemented in SOL 66 (or SOL 106) of MSC/NASTRAN Version 66.2) reduces t o By neglecting the rate of change of the instantaneous tangent t o the uniaxial curve with respect to temperature. the tangent material stiffness matrix may be obtained by 6. is used as a starting point for linear and nonlinear thermal elastic analyses. .5. The set ID of this temperature distribution is specified above the Case Control section by the TEMP(IN1T) command.3..e. i. To.
T . E the effective elasticity modulus from TABLES1 data. the MATSl and contains the TID of TABLES1 cards for given temperatures. The TABLEST entry is shown below followed by brief description of the fields: =. but it can only be applied to the upstream superelements. is specified in the Case Control by TEMP(L0AD). TABLEST. the material properties are determined at the very beginning of the analysis based on the temperature distribution given in the first subcase. The coefficient of thermal expansion reference temperature.(T)] the elasticity matrix from MATT1 are data. TEMPPI.TABLEST.TABLESl) where {a) and (€1 the stress and strain vectors. Different options of thermoelastic constitutive relations may be obtained by different combinations of input data: r {a) = W [ D ~ ( T ) ] { ~ ) {a} = [D.(T)]{C) r {a} = W [ D e ] { c ) (MAT1. Some old features such as TREF in MATi and TEMP(MAT) which were valid prior t o Version 66 should not be used in versions following Version 65. the TABLEST may be used to input the temperThe TABLEST is referenced by ature dependent uniaxial stressstrain relations.e. there is no default value for To and a fatal error message will be issued if TEMP(1NIT) is not present. The material properties will conform to TEMP(L0AD) without requiring TEMP(MAT) or TEMP(B0TH). MATTi. are input by a combination of the MATi. A fatal error message will be issued if both TABLES1 and TABLEST with the same TID are present. Each TEMP(1NIT) or TEMP(L0AD) points to the Bulk Data TEMP. For nonlinear elements. i. TEMPD. TABLESl) (MATl.(T)].e. i. [D. From Version 67. But these properties will stay constant throughout the analysis. For example. r To be the reference (stressfree) temperature when TEMP(L0AD) is specified. The set ID for the load temperature distribution.MATSl. the TEMP(L0AD) also specify the temperatures at which material properties are determined if Bulk Data entries MATTi are present. TEMP(MAT) without TEMP(IN1T) would run through in Version 67. They are recommended to be used for all types of thermoelastic analyses available in SOL 66 and SOL 106 for Version 66 and beyond. For the nonlinear elastic material. TREFis input by the TREF field on the MATi. TEMPRB on which the temperature distribution is specified. which is different from Version 65. TEMP(MAT) may also be used in the nonlinear transient analsis (SOL 99 or SOL 129). MATS1. . However. A fatal message will be issued if TEMP(MAT) appears in a residual subcase. [D. and E the elasticity modulus from MAT1 data. which is specified by the MATSl and TABLESl.To be the default temperature distribution whenever TEMP(L0AD) is not specified for a subcase. MATT1. Temperature dependent properties for elasticity matrix. and TABLEMi entries. Notice that the combination of TEMP(IN1T) and TEMP(L0AD) is new user interface introduced in Version 66.
but nonlinear elastic composite materials (MATSlINONLINEAR with MAT2 or MATS) are not.TEMPD. Ii : Table identification number of TABLES1 entries (integer > 0).TEMPD. Ii must be unique with respect to all TABLES1 and TABLEST table identification numbers. no TEMP(L0AD) = n each subcase temperature dependent material is present thermal loads temperature same as above dependent linear materials Entries Bulk Data TEMP. Possible thermal stress analysis types and the corresponding required input cards are shown in tlie table below. MATS1 TABLEST TABLES1 same as above but a=O. Analysis capability for the creep (or viscoelastic) deformation under the variable temperature became available from Version 66.TEMPD. Ti : Temperature values (real) which must be listed in ascending order. ENDT : The endoftable flag.. In addition. Thermoplasticity is also not available in MSCINASTRAN Version 67. MATi. The former . This number must be identical to TID of MATSl to be active.TID T1 I1 T2 I2 T3 I3 ENDT TID : Table identification number (integer > 0). Types of Thermal Stress Analyses and Required Input Case Control Type of analysis TEMP(IN1T) = m once thermal loads only.. temperature dependent material only + same as above same as above Notice that temperature dependent linear composite materials (MAT2 with MATT2 or MATS with MATTS) are available in MSCINASTRAN. MATTi TABLEMi TEMP. MATi..O on MATi + thermal loads temperature dependent nonlinear elastic materials no thermal load. the creep characteristics and the elasticity properties both can be temperature dependent. MATi TEMP.
000 psi 0 psi 0 psi 1. a set of temperature dependent stressstrain curves were specified using a combination of MATS1 and TABLES1 entries.75 x 2.are specified by the CREEP entry (see Section 6. assuming that the former is the stressfree temperature. 6. as shown in Fig. The Young's modulus is 8. Figure 6. Temperature 100°F 200°F OX UY EX EY 10. The MATTl entry present in the second model is a dummy. and the latter are input with the same procedure as described above. two identical plate models were created and solved using different approaches as shown in the attached MSCINASTRAN input data. the thermal expansion coefficient is assumed to be 0.5. the reference properties were still input through a MATTl entry. respectively.0 x lo6 psi at 200°F.1. which is required in V66 and V66A t o activate the thermal elastic capability. In the second model. however.3 for all temperatures.1 Cube Subjected t o Uniaxial Tension For verification purpose.4 for more details). In the first model. For simplicity.25 x low3 3.5. is determined. The results given by these two models are identical as shown below. The structural response at 10o°F and 200°F. and their variations with different temperatures were input through MATTl and TABLEMl entries. the reference material properties were defined on a MAT1 entry.000 lbs and a temperature change from 10o°F to 200°F.3 Validation Problem A 1inch cube. is subjected to a uniaxial tension of 10. and the Poisson's ratio is 0. 6.50 x .5.000 psi 10. This requirement has been removed in V67 and beyond.50 x 7.0 x lo6 psi at 100°F and 4.
0. 8.0000+6ENDT . I. I. I. 0.+6 151 1. 0.0769+60. $ NONLINEAR STATIC SOLUTION SEQUENCE SOL 66 8. V66A $ SSH 12/12/90 ID TIME 5 $ CPU MIN. 0. 0. 0.3 152 PSHELL 100 $ MAT1 MATT1 $ 120 120 TABLEMI 151 0.0 3. TEMPD $ $ ITERATION STRATEGY $ NLPARM 100 NLPARM 200 $ $ $ I 5 YES YES FIRST MODEL 101 102 103 104 0 0 0 0 0.0000t6200.0000+6100. I. TABLEMI 152 8. GRID GRID GRID GRID $ CQUAD4 101 $ 100 101 102 103 104 120 8. 4. 0.THPLATE.50 $ PRINT MATRIX TRAILERS AND TRACE NL ITERATION DIAG CEND TITLE = 2 DISCRETE PLATE MODELS UNDER UNIAXIAL TENSION ECHO = SORT SEALL = ALL = ALL DISP = ALL STRESS ELFORCE = ALL SPCFORCES = ALL = ALL OLOAD TEMP(INIT)=IOO SUBCASE 100 NLPARM = 100 LOAD = I 0 1 SUBCASE 200 TEMP(L0AD) = 200 NLPARM = 200 LOAD = 101 BEGIN BULK $ $ $ TEMPERATURE DISTRIBUTION 100 100. 200 200.
0.+6 221 25 1 8.PSHELL 200 $ 220 8. 0.0769+6100. 253 1. 252 1. 0. 0. 0.0769+6200. 0.OE3 200. I. 251 1.OE3 5.0 1. 0.t6 ENDT ENDT ENDT ENDT FORCE 101 FORCE 101 ENDDATA .0 3. 0 0 MAT1 MATT1 MATS1 TABLEST +ST1 TABLES1 +SSI TABLESI +SS2 TABLESI tSS3 $ 220 220 220 221 0. 252 0. 0.+6 4. 5.+6 8.+6 8. I. I. 102 103 0 0 3.+6 202 203 1. 0.+6 4.OE3 5.0 1.0. 5.0769+60. 0. 0. 1.3 NLELAST 100.0 FORCE FORCE 101 101 $ $ SECOND MODEL $ GRID GRID GRID GRID $ 201 202 203 204 0 0 0 0 0.OE3 1. 0.0 253 8. I. $ 3. I. I. I. CQUAD4 201 200 201 202 203 204 $ .
Each deck contains two discrete models. This capability may be used t o model nonlinear materials which exhibit significant initial anisotropy. 6. ANISOEP is used t o test the elasticplastic analyses. and viscoelastic materials as shown in the preceding sections. respectively. are listed at the end of this section as examples.6. in which the yield stress is assumed to be 5000 psi and the plastic hardening modulus t o be 5 x lo5 psi.] appears in the incremental constitutive relations for elasticplastic.g.6. a plate and a solid.6. First. namely ANISOEP. Then. When they are combined with plastic or creep material.6 INITIAL ANISOTROPY This material capability became available in SOL 66 (or 106) and SOL 99 (or 129) since Version 66A. ANISOC. which are subjected t o a simple tension force of 10. the elastic properties are assumed to be isotropic so that may the elasticity matrix [D. two subcases are specified.1 Theoretical Basis The elasticity matrix [D. In ANISOEC. The input data of three test decks. Young's modulus E and Poisson's ratio v.2 Validation Problems A cube with dimensions of one unit was modeled with one QUAD4 element for testing the 2D case and one HEXA element for testing the 3D case. The first subcase is for a static response analysis which is followed by the second subcase for a creep analysis of 100 hours. e. The material is assumed t o remain linear elastic throughout the analysis.000 lb. genuine anisotropic properties were used and the results from the plate and solid models were compared with each other. It allows the following combinations of material property entries: Elasticplastic: MAT2 or MAT9 combined with MATSl(TYPE=PLASTIC) Thermoelastic: MAT2 combined with MATT2 or MAT9 with MATT9 Viscoelastic: MAT2 or MAT9 combined with CREEP Notice that MAT2 or MAT9 represent elastic anisotropy. ANISOT is used t o test the thermoelastic . and ANISOT.] an anisotropic elasticity matrix specified by MAT2 or MAT9 Bulk Data entries. they represent initial anisotropy because the plastic or creep (viscoelastic) deformation remains isotropic. isotropic properties were input using MAT2 and MAT9 Bulk Data entries and the results were compared with the analytical solutions. The by orthotropic material (MAT8) is not applicable to this capability. In those constitutive relations. 6. Inplane anisotropic elastic properties are assumed so that the plate and solid models are expected t o give compatable results. thermoelastic. This assumption was removed by simply replacing the isotropic [D.] be determined by two material constants.
147 x 1.000 solid model 10.910 x .054 x 7.560 x 2.550 x lo* 9.910 x 1.941 x 4.000 ANISOC : plate model 10. The results of these three test problems are tabulated below: ox ex E~ YSY ANISOEP : plate model 10.analyses. A tension force of 10.486 x 9.526 x 4. the mechanical load remains unchanged but the temperature is increased.487 x 9.050 x 9.000 solid model 10.540 x 9.551 x 1.000 1.000 solid model 10.560 x 2.941 x 2.147 x 2.000 ANISOT : plate model 10.548 x 9.526 x 7. In the second subcase. which reduces the value of each elasticity modulus by one half.000 lb is applied in the first subcase.
0 0. 3456 0 0. UNIAXIAL TENSION CASE. 2 DISCRETE MODELS EACH CONTAINS I QUAD4 OR HEXA ELEMENT.0824E6 PLASTIC 5 .9812E66. 0.0 1.0 1.0 1.0 1.0 0. 0. V66 TIME 5 SOL 66 CEND $ $ SSH 3/25/91 TITLE SUBTITLE LABEL SEALL DISP STRESS $ = = = = MAT2 OR MAT9 COMBINED WITH MATSI(TYPE=PLASTIC).0 0 1 .0 0. ALL = ALL = ALL = 100 = 101 NLPARM LOAD $ BEGIN BULK $123456789.0 0.9812E66.5136E64. 4.0 0. 0. 0.0 0 1.0 SOLID MODEL 501 502 503 504 511 512 513 514 501 513 500 520 0 0. 0.OE3 1.0 0 1.5136E63.0 1.E5 5.0 0 0.149E50.0 203 0 5.0 500 501 502 514 520 7.9539E61.0 7.149E51.0 1.0 0 0.OE3 1. 0. 123456 0 I.E3 202 0 5.O 0 1. I.0 0.0 503 123456 3456 3456 13456 12456 456 456 1456 511 GRID GRID GRID GRID GRID GRID GRID GRID CHEXA +E51 PSOLID MAT9 504 512 +E51 1.0 0. I. 13456 200 201 202 203 204 220 1.0 0.ID ANISOEP.0 0. 0. 3456 0 1.0 0 0.0 0. 0.0 0.9539E6+M51 .0 1.0 1.lo$ $ $ ITERATION STRATEGY $ NLPARM 100 10 AUTO $ $ PLATE MODEL $ GRID GRID GRID GRID CQUAD4 PSHELL MAT2 MATS1 FORCE FORCE $ $ $ 201 202 203 204 201 200 220 220 101 101 0 0.5136E60.
3. 0.E5 5.OE3 5.0 0.5 0.+M51 tM52 MATS1 FORCE FORCE FORCE FORCE ENDDATA 0.5 0.5136E60.0 0.0 0. 0.OE3 1.5 0.E5 0. 5.0 0. 1. 520 PLASTIC 101 502 0 101 503 0 101 512 0 101 513 0 0.E3 0.OE3 5.E6 0.0 0. 3. 3.0824E60.0 0.E5 5.0 0.5 0. +M52 .0 0.OE3 5.
0.08511 2. I. 123456 3456 3456 13456 12456 456 456 GRID GRID GRID GRID GRID GRID GRID . 0. 3456 0 0. 0. 0. 0.5136E63. 1.9812E66.5136E64. 3456 0 1. 1. I. 0.ID ANISOC.0824E6 GRID GRID GRID GRID CQUAD4 PSHELL MAT2 CREEP +CP2I CRLAW 3. 0. V66 $ SSH 3/25/91 TIME 5 SOL 66 CEND TITLE = MAT2 OR MAT9 COMBINED WITH CREEP SUBTITLE = 2 DISCRETE MODELS. 0. I. 0. 0.0 7.094 1. 13456 200 20 1 202 203 204 220 1.149E51.0 0.084 2. 0 1. 0.0 FORCE FORCE $ $ $ I I SOLID MODEL 401 402 403 404 411 412 413 0 0 0 0 0 0 0 0. 0.0211 7. EACH CONTAINS ONE QUAD4 OR HEXA ELEMENT LABEL = UNAXIAL TENSION CASE = ALL SEALL = ALL DISP = ALL STRESS $ SUBCASE 10 = I LOAD NLPARM = 10 SUBCASE 11 LOAD = I NLPARM = 20 $ $ ELASTIC $ CREEP FOR 100 HOURS BEGIN BULK $123456789$ $ $ lo ITERATION STRATEGY I 5 20.4764 2. 1. I.0 0. 0.+4 0.0 203 0 0.5 1. NLPARM 10 NLPARM 20 $ $ $ AUTO AUTO PLATE MODEL 201 202 203 204 201 200 220 220 222 123456 0 0. 1.+4 0.9539E61. 0.5 1. I. 0. I. 0. I.434 202 0 0. I.
25 1. 1.25 1.0 0.E6 0.E5 420 CL W RA 222 3.0211 I 402 0 0.0 I 412 0 0.094 1.4764 2. 1.0 0. 3.0 .+4 0.434 0. 3.5136E60. 1. +M42 +CP41 7.+4 0.9539E6+M41 0. 3. 0.+4 0.084 2. 1. 0.5136E60. 1.E5 0.25 1.GRID CEA HX +E41 PSOLID MT A9 +M41 +M42 CREEP +CP41 FORCE FORCE FORCE FORCE E D AA NDT 414 0 0.+4 0. 0.0 0. 4.08511 2. 0.0 1 403 0 0.0 1 413 0 0.149E50.0824E60. 401 400 40 1 402 403 404 413 414 400 420 420 7.25 1.0 1456 41 1 412 +E41 0.9812E66.
FORCE 101 FORCE 101 . TABLEMI 252 100. TABLEMI 251 100.ID ANISOT. YES $ $ PLATE MODEL $ GRID 201 202 GRID GRID 203 GRID 204 CQUAD4 201 PSHELL 200 MAT2 220 MATT2 220 TABLEMI 255 100. EACH CONTAINS A QUAD4 OR HEXA ELEMENT = UNIAXIAL TENSION LABEL = SORT ECHO = ALL SEALL DISP = ALL = ALL STRESS TEMP(INIT)=IOO $ SUBCASE 100 NLPARM = I00 LOAD = I01 SUBCASE 200 TEMP (LOAD)=200 NLPARM = 100 =I01 LOAD $ BEGIN BULK $ TEMPD NLPARM 100 100 100. I 200 200. TABLEMI 254 100. V66 $ SSH 3/25/91 TIME 5 SOL 66 CEND $ = MAT2 COMBINED WITH MATT2 AND MAT9 WITH MATT9 TITLE SUBTITLE = 2 DISCRETE MODELS. TABLEMI 253 100.
25 0.0 402 0 1.0 411 0 0.0 1.0 1.25 0.0 FORCE 101 413 0 l.5136E60. 3.0 ENDDATA .5136E60. 3. 1.5412+6ENDT TABLEMi 454 100.0 401 400 40I 402 403 404 413 414 400 420 420 7. 1.0 1. TABLEMI 452 100.4770t6ENDT TABLEMI 451 7.OE4 0.9906t6ENDT 100.0 404 0 0.0 0.$ $ SOLID MODEL $ GRID GRID GRID GRID GRID GRID GRID GRID CHEXA +E41 PSOLID MAT9 +M41 +M42 MATT9 tMT41 tMT42 TABLEMi 401 0 0. 6. 3.0 0. 3.0 414 0 0. 2.9539t6200. 0.5136+6200.E5 420 451 452 453 453 455 455 100.0 FORCE 101 412 0 1.149t5200.0 1.7568+6ENDT 100. FORCE 101 402 0 l.075t5ENDT TABLEMi 453 1.0 0.E5 0. 1.0 0. 0.E6 0.0 1.25 0.0 412 0 1. 3.0 0.OE4 0.OE4 0.0 1.0 1.0 0.0 403 0 1.0 0.149E50.0 0. 0.0824E60.0 413 0 1.0824+6200.0 1.9812E66.9812t6200.OE4 0.25 0. 4. 3. 0.0 FORCE 101 403 0 i. i.
The load step is labeled by the cumulative load factor. 2.1.e. PLOADi. 7. The enforced motion is also subdivided in the subcase like the incremental loads. i.. The enforced motion can be applied using SPCD or nonzero value in SPC. MOMENTi. etc. gap elements (GAP) for nonlinear interface. CREEP and TABLESl). 3. The functionality of SOL 106 is identical t o the old DMAP.g.e. FORCEi.Chapter 7 NONLINEAR STATIC ANALYSIS 7. i. SOL 106 is a structured DMAP (utilizing the new executive system constructs) implemented in Version 67. and PARAMeter LGDISP for geometric nonlinearity. which can be subdivided into a number of increments. etc.. Constraints can be specified using SPCs and MPCs. Loading conditions are specified using the same loads as in linear static analysis.. There are advantages t o divide the entire loading history into . Then the load step will end with 1. A SOL 61 model can be converted t o SOL 66 analysis with a few additional data entries relevant to the nonlinear analysis. Modeling options are compatible with linear analysis. The load factor varies from 0 t o 1 in each subcase. for the first. The nonlinear properties and/or effects are defined by nonlinear material data (e. MATS1. respectively. the second and the third subcase. The unique data required for SOL 66 or 106 is supplied on the NLPARM data entry. TEMP. which controls the incremental and iterative solution processes. The specified value for SPCD or SPC is a measure of displacement with reference t o the undeformed postion. which can be changed from subcase to subcase.1 Case Control Each subcase defines a set of loading and boundary conditions. SOL 66. GRAV. The snapthrough analysis or postbuckling analysis can be performed using arclength methods which are controlled by parameters in the NLPCI Bulk Data entry.1 USER INTERFACE Nonlinear static analysis can be performed using SOLS 66 or 106.
SUBID and LOADINC) which can be specified either in the Case Control section or Bulk Data section. e. ELFORCE. which defines strategies for the incremental and iterative solution processes.g. STRESS. Restarts are controlled by parameters (LOOPID. However. More extensive line searches may be exercised by a large value of MAXLS and/or a smaller value of LSTOL. OLOAD.g. the nonproportional loading can only be achieved by using multiple subcases. However. Output requests for each subcase are processed independently. MAXLP) are provided for additional control or capabilities. 7. users with little insight or experience in a specific application should start with the default option. Some optional parameters (BUCKLE.2 Iteration Control: NLPARM Data The most crucial data for successful nonlinear static solutions are contained in the Bulk Data NLPARM entry. On the other hand.many subcases so that each subcase does not have excessive number of increments. Available outputs are DISPLACEMENT. It is difficult to choose the optimal combination of all the options for a specific problem. Therefore. creep analysis. LANGLE. The data blocks containing solutions can be generated at each increment or the end of each subcase depending on the intermediate output option specified in the INTOUT field of NLPARM data entry. e. line searches may have adverse effects in some problems. element force and GRID point stresses are available only for linear elements. Default tolerances for the convergence criteria may be somewhat conservative. However. Furthermore. However.. As such. not t o exceed 20 steps. users with some experience and insight in a specific problem may change the default values based on the following observations: Computing cost for each line search is comparable t o that of an iteration.1. the database size or the output quantities are better controlled with multiple subcases. GRID point stresses and SPCFORCE. based on a considerable number of numerical experiments. . e.. The line search method is effective t o cope with difficulties in convergence in some problems.. loose tolerances may cause difficulties in the subsequent steps. Requested output quantities for all the subcases are appended after the computational process for actual output operation. All the superelement model generation options and matrix reduction options are allowed for the linear portion of the structure. plane stress plasticity. However. the default option was intended to provide the best workable method for a general class of problems. it seems t o have adverse effects in some problems. The quasiNewton method is effective in most problems. Data blocks are stored in the database for output process and restarts.g. The SEMI method usually provides better convergence than the AUTO method a t the expense of higher computing cost.
g.OE3 EPSW 1. Use of a moderate value has advantages in controlling database size. The larger the creep rate. The NLPARM data format is shown below with default values wherever applicable: NLPARM NLPARM ID NINC DT 0.2 LSTOL 0.5 +NP2 MAXBIS +NP2 5 M AXR 20. It is noted that a creep analysis should be preceded by a static analysis with at least one subcase. This . The total load specified in the subcase minus the load specified in the preceding subcase is equally divided by this integer (NINC) t o obtain the incremental load for the current subcase.OE7 MAXDIV 3 MAXQN MAXITER MAXLS 4 FSTRESS 0. The ID field requires an integer t o be referenced by a Case Control command NLPARM. which is an auxiliary to the KMETHOD field. the program automatically determines when to reevaluate the stiffness matrix based on the rate of convergence. RTOLB 20. However. SEMI. Another subcase should be defined to change constraints or loading paths. Stiffness matrix update strategies are determined by a combination of the data specified in the two fields KMETHOD and KSTEP. No creep deformation is considered with the default value of 0. The NINC field is an integer which specifies the number of increments to be processed in the subcase. The creep time increment should be determined properly for efficiency and accuracy based on the creep rate represented by the creep law to be used.OE3 EPSP 1. With the AUTO option. The KSTEP field. not to exceed 20) for NINC. the smaller the time increment should be. KMETHOD AUTO KSTEP 5 MAXITER 25 CONV PW INTOUT NO +NP1 EPSU +NP1 1. should have an integer greater than 1. output size. the computing time for convergence without the stiffness matrix update is estimated and compared with the computing time for the matrix update in order t o determine whether the update is more efficient. Then the Bulk Data entry NLPCI must be attached.The arclength method should be used if the problem involves snapthrough or postbuckling deformation. because the creep deformation requires stresses in the material. or ITER. The D T field requires a real number specifying the time increment for each load step in the case of creep analysis. multiple subcases may be required in the absence of any changes in constraints or loads to use a moderate value (e. and restarts. At each iteration. Options for KMETHOD are AUTO. The time unit must be consistent with the unit used in the CREEP data t o define the creep characteristics.
With a small database option. the output requests are processed for every computed and user specified load increment. NO or ALL. If KSTEP > MAXITER. and W(for work) may be specified. It is KSTEP noted that the original modified Newton's method is selected when KSTEP = MAXITER. Then. specified in the Case Control Data.. etc. If the solution tends to diverge.). The INTOUT field requires a specification of YES. There are two classifications in the divergence .). If ALL is specified. P. The convergence tolerances are specified in the fields EPSU. Any combination of U(for displacement). NLPCI command is specified). However. and W criteria. the stiffness matrix is updated at every KSTEP iterations. the output requests will be processed only for the last load step of the subcase. will be processed for every computed load step in the subcase. The convergence test is performed at every iteration with the criteria specified in the CONV field. etc.g. If MAXDIV is negative. the load increment is bisected and the analysis is repeated. For Newton's iteration methods (i. without NLPCI). the best attainable solution is computed and the analysis is continued to the next load increment. < The MAXITER field is an integer representing the number of iterations allowed for each load increment. If YES is specified.e. MAXBIS is reached or MAXBIS=O) and MAXDIV is positive. The option ALL allows the user t o obtain solutions at the desired intermediate load increments. This subject will be elaborated in the following section. The stiffness matrix will be updated upon convergence if the number of iterations required for convergence is greater than KSTEP. the output requests (DISPLACEMENT. The SEMI option is identical t o the AUTO option except for one additional stiffness update after the first iteration which always occurs unless the solution converges in a single iteration. time expiration.. EPSP. If the load increment cannot be bisected (i. the stiffness will never be updated. if the job is terminated with an incomplete subcase for some reason (e.. except for an absolute convergence condition. the computed load increment in general is not the same as the user specified load increment. All the specified criteria must be satisfied t o achieve convergence. the option ALL is equivalent t o the option YES.decision is deferred in the first two iterations after a new stiffness is obtained. the update decision will be made effective immediately. STRESS. since the computed load increment is equal t o the user specified load increment. under which the solution is converged regardless of criteria. The INTOUT data also affects the database if the PARAMeter SMALLDB (valid only in Version 65) has a value of 1. Thus. respectively. diverging. ELFORCE. the restarts will be restricted t o those starting from these steps. The MAXDIV field requires an integer to specify a limit on the probable divergence conditions allowed for each iteration to continue. If the number of iterations reaches MAXITER without convergence. and EPSW for U. data blocks for all the interim steps of that subcase will be stored in the database to allow flexible restarts. The ITER method also allows a stiffness matrix update upon convergence (since Version 66) when MAXITER. With the ITER option. If NO is specified. however. the database will retain only those data blocks corresponding t o the incremental load steps for which output is requested by an INTOUT field.e. P(for load). For arclength methods (i. the full NewtonRaphson iteration is exercised if KSTEP is 1.e. the analysis is terminated. and is not known in advance.
the program will terminate the job with a fatal error message "ERROR EXCEEDED 20% OF CURRENT YIELD STRESS. the stiffness matrix is updated on the first divergence and the load increment is bisected on the second divergence. different actions are taken depending on the sign of MAXBIS. When the solution diverges.condition: probable and absolute.01 and 0.2 for a detailed description of divergence criteria.3. the line search continues until the number of line searches reaches MAXLS. the current solution base (displacements) is retracted and the stiffness matrix is updated in reference to the preceding iteration step. QuasiNewton vectors are accumulated. If MAXBIS is positive. All the QN vectors already accumulated are purged upon a stiffness update and the accumulation is reinitiated. the analysis is terminated on second divergence. bisection will be activated under this condition. When the probable divergence occurs MAXDIV times or more and if MAXDIV is positive. If the number of accumulated QN vectors is greater than or close to MAXQN upon convergence. The MAXR field requires a real number to specify the bounds on the ratio of the adjusted arclength increment relative t o the initial value.. Details of the quasiNewton method will be presented in the following section. The LSTOL field requires a real number between 0. if desired. In the adaptive load/arclength increment. . the load increment is bisected every time the solution diverges until the limit on bisection is reached.9 t o specify a tolerance for the line search operation. The FSTRESS field requires a real number (0. The absolute divergence is treated as two occurrences of the probable divergence. the best attainable solution is computed and the analysis is continued to the next load increment." If the bisection option is selected. the stiffness matrix will be updated before the next increment is processed.0 < FSTRESS < 1. If the line search convergence is not achieved.0). If MAXDIV is negative. If MAXBIS is negative. The line search operation will be conducted if the error defining the divergence rate is greater than LSTOL. The maximum number of line searches allowed for each iteration is specified in the MAXLS field. The line search process may be suppressed with a value of 0 for MAXLS. Refer to Section 3. This data is also used t o establish an error tolerance in the yield function t o alleviate accumulation and propagation of the error. If the solution does not converge after (MAXBISI bisections. The number of subincrements in the material routines for elastoplastic or creep deformation processes is determined such that a subincrement in the effective stress is approximately FSTRESS*a. The MAXQN field requires an integer to specify the maximum number of quasiNewton vectors t o be saved in the database. If the solution diverges again in the same load increment while MAXDIV is positive. The MAXBIS field requires an integer t o specify the number of bisections allowed for a load or arclength increment. If the error in the yield function exceeds FSTRESS*B at the converging state. the analysis is continued or terminated depending on the sign of MAXDIV. representing a fraction of the effective stress ( a ) used to limit the subincrement size in the material routines. until MAXQN is reached. The quasiNewton update may be suppressed with a value of 0 for MAXQN. The line search is controlled by the fields MAXLS and LSTOL.
followed by brief description of the fields: NLPCI NLPCI ID TYPE CRIS MINALR 0. or MRIKS for modified Riks method) MINALR. The NLPCI entry is shown below with default values. This bisection strategy based on the incremental rotation is controlled by the MAXBIS field. The bisection is activated if the incremental rotation for any degreeoffreedom (Ad. 5 5 MAXALR Atold > 1.Al0 where Al.the overall upper and lower bounds on the load/arclength increment in the subcase are defined as 1 Aln < . MAXLS is not applicable because the line search procedure is not coupled with arclength methods currently.< MAXR MAXR . MAXALR : For variable arclength method. TYPE : Constraint type (CRIS for Crisfield. is the arclength at step n and Alo is the original arclength. 0.1.) excceeds the value specified for RTOLB. The arclength method for load increments is selected by the NLPCI Bulk Data entry. 7..25 MAXALR SCALE DESITER 7 MXINC 20 4. The parameters which are applicable only t o the arclength methods are specified in the Bulk Data entry NLPCI. ID : must be identical to ID of NLPARM to be active. or Ad. Ad. which must have the same ID as the NLPARM Bulk Data entry. RIKS for Riks. The RTOLB field requires a real number to specify the incremental rotation (in degrees) allowed per iteration. and MAXALR MAXALR=MINALR=l.3 User Interface for ArcLengt h Met hods: NLPCI Most of the parameters in the NLPARM Bulk Data entry are used t o control the iteration strategy for arclength methods. The automatic adjustment is not activated if . the arclength is bounded by MINALR where MINALR 5 1. However. in connection with the NLPARM data by the same ID.
In the limiting case of infinite w.6.SCALE : Scaling factor for the magnitude of the load term in arc length.3) : relative error in terms of work (See Section 3. initial error before line search (See Section 3. the analysis cannot be completed without bounding the maximum number of increments. this value should be confined to a reasonable limit. t o prevent inadvertent depletion of the computing resources.6.g.3. the relevant data from the iteration process are printed under the following heading: ITERATION : iteration count i EUI EPI EWI LAMBDA DLMAG FACTOR EFIRST EFINAL NQNV NLS ENIC : relative error in terms of displacements (See Section 3. the constraint equation is gradually dominated by the load term. p .1) : absolute norm of the load error vector : final value of the line search parameter (See Section 3.1) . less than 100. In these cases.1. the arclength method would be degenerated t o the conventional Newton's method. Real O.. The scaling factor (w) is introduced as user input so that the user can make constraint equation unitindependent by a proper scaling of the load factor. Default = 0.. For such cases.4. The constraint equation has a disparity in the dimension by mixing the displacements with the load factor. > DESITER : Desired number of iterations for convergence to be used for adaptive arclength adjustment.4 Iteration Related Output Data At the end of every iteration. e.3) : relative error in terms of loads (See Section 3.6.3) : number of line searches performed during the iteration : expected number of iterations for convergence (See Section 3. As the value of w is increased.5.3) : number of quasiNewton vectors appended (See Section 3. MXINC : Maximum number of controlled increments allowed within the specified load step. There can be cases where the structure never snaps back and collapses t o failure.4.3.3) : rate of convergence (defined in Section 3.4) : divergence rate.3) : error at the end of line search (See Section 3. 7.
3.NDV tion 3.2) : number of occurrences of probable divergence during the iteration (See Sec MDV : number of occurrences of bisection conditions in the material routine (excessive stress increment) or in the rotation angle (excessive rotation) during the iteration .
2. else only material stiffness unless the arclength methods).N./ S. FFGVH.NEWK=+l/ S.LGDISP/ S. LLLT.ITERID/ S.1 and 7.N. MESTNL. DIT. ssize. + . TWODIV=O/ V.LANGLES $ Note: New data in Version 67 are highlighted in italics.ARCLG=O.N. G M N L Multipoint constraint equation matrix. there is a nonlinear module which performs incremental and iterative processes.N.N.Y. diag(KLLT). K A A L Reduced element stiffness matrix (linear element only). E L D A T A Element combination data. P L M A T Load vectors for subcase (initial and final).N./ S. CSTMS. USET / UGVMO.N.RSTEP/ S.SOLCUR/ V.NEWP=+l/ S.N.1 NLITER DMAP Interface of NLITER Module (Version 67) CASES.NSKIP/ V.N.KTIME/ S.TABS/ S. QNV.LOADFAC/ S. BGPDTS.ARCSIGN=l/ S.2 Input Data Blocks C A S E S Case Control data (Case Control subset for residual). GMNL. 7. FGNL. NLITER.N.N. DUGVMO / S.KMATUP/ S.N.CONV=O/ S. !size.N. 7.N.N.2 ITERATION MODULE AND SOLUTION SEQUENCE In the heart of the nonlinear static solution sequences 66 or 106. ELDATA. The DMAP interface. KAAL. 7.2. YSMAT.POUTF/ S. NBIS=O/ S. CIDATA.Y. has evolved for a decade since Version 61.DELMU=l. PLMAT. LLLT Lower Triangular Factor for nonlinear elements (material and differential if positive definite..2. The solution sequence for nonlinear static analysis (SOL 66) is depicted by flow diagrams in Figs.N. CURMU/ S. KELMNL. input/output data blocks and parameters as of Version 67 are described below. C N V T S T Convergence test vectors : AP.N. The module.N. Pg Qg. MUGV.N. ESTNLH.2.N. consisting of element thermal loads and element connectivity (used in SDRNL). SLT1.N.MPTSIZE/ S.7. MJJ. CNVTST. SILS. K E L M N L Nonlinear element stiffness matrix. MPT. NBISO=O/ S.2. Y S M A T Constraint displacement vectors for subcase (initial and final).N.LSTEP/ S.KFLAG=+I/ S.
used for the second Crisfield rootselection algorithm No data store for the modified Riks method QNV QuasiNewton vector file.  YSMAT (on first solution). DENOML is set t o zero. DIT Direct Input Table. CSTMS. ESTNLH Input/output. Constraint status. BGPDTS. Note: The following data blocks may be purged SLT1. the amount by which the projected value of p exceeds the specified output value of p. 2 that it is off the constraint. the last denominator in the equation for the computation of A p . 7. For the other methods. FGNL Nonlinear element force from last iteration. USET Displacement Set Definition Table. The current value of p. The value of the arc length at the beginning of the subcase. GMNL. CIDATA Input /output. SECALG FULL Special rootselection flag for the false reversal case. miscellaneous data for arclength methods. For the Riks method. FFGVH Follower force vector for OLOAD output. MJJ Mass matrix (gsize). DIT. BGPDTS Basic Grid Point Definition Table.3 Output Data Blocks UGVMO Input/output. CSTMS Coordinate System Transformation Table. SILS Scalar Index List.uO). 1 signifies that the solution is on the constraint. currently used only with modified Riks. CURDEL ORGDEL NEWMU DMUAST DENOML { A u : ) for the Riks method { u l . SLTl Static Load Table for follower forces. updated table of ESTNL at converged step. If nonzero.2. The current value of the arc length. (normalized). displacement vector in gset at converged step.MPT Material Properties Table (NLPARM and NLPCI data included). .
On Output : 1 if old stiffness matrix t o be retained. On Output : 1 if stresses are not t o be printed. When LSTEP = NINC from the NLPARM card. 1 otherwise. Absolute temperature conversion constant for creep analysis. be stored at the output steps.MUGV Displacement vector for Kupdate. nonlinear Element Summary Table for Kupdate. The DMAP loop number at the current iteration step. The current count of the stiffness matrix updates within the increment. On Output : 1 if solution did not converge. NEWP Input /outputinteger. to 7. On output from the module. KTIME Input/outputinteger. RSTEP Input/outputinteger. On Input : +1 for geometric nonlinear problem. KMATUP Input/outputinteger. TABS Inputreal. new stiffness matrix required. $1 if solution converged. Problem nonlinear type flag. The value is set to 2 upon NOGO flag LGDISP Inputinteger. The timetogo value a t the start of the stiffness matrix formulation loop within the DMAP. ITERID Input/outputinteger. +1 for a new subcase. LSTEP Input/outputinteger.2. (No longer used in Version 67). SOLCUR Input/outputinteger. The value of the load step within a subcase. $1 if stresses are t o be output. MPTSIZE Input/outputintegerdefault 0. 1 otherwise.4 Parameters LOADFAC Input/outputcomplex. CONV Input/outputinteger. Iteration count. DUGVMO Input/output. The count of arclength increments. . (material nonlinear elements) NSKIP Input/outputinteger. denoting the increment of time between stiffness computation and module start. NEWK Input/outputinteger. The value is assumed t o be one at the start of a subcase and is incremented by one each time the module converges. The number of records to skip into the Case Control record (subcase ID). > 1 if POUTF Outputinteger. incremental displacement vector in gset between converged steps (17. Unl). Load factor for current module iteration. fractional value between 0 to 1. a negative value is stored. MESTNL Input/output. On Input : Last iteration value (0 for initial iteration). NEWP is set fl.
LANGLES Inputinteger. NBIS Input/outputinteger. Used to select the rotation vector approach t o large rotations by a value of 2. ARCSIGN Input/outputinteger. all the data are printed at each iteration. Initialized as $1 for the cold start and 1 for the restart in the DMAP. Number of diverging solutions in the same load step. If this number reaches two. Current value of the load factor in the subcase. t o be stored in the database. The data that may be produced are listed below. Because of the volume of output produced. The current count of the bisection. The ratio of current load increment t o the user specified load increment.2. For each entry into NLITER: + + Subcase status data NLPARM/NLPCI data . The value is set to 0 at the start of each user specified load step and is set equal t o NBIS when the solution converges. The value is set to 0 at the beginning of the subcase. The previous count of the bisection. CURMU Input/outputreal. DELMU Input/outputreal. NBISO Input/outputinteger. The value is set t o 0 at the start of each user specified load step. or K is updated with CONV=1.5 Diagnostic Outputs Diagnostic output is available in SOL 66 (or 106) if DIAG 50 or 51 is specified in the Executive Control Section.KFLAG Input/outputinteger. and users in general are cautioned against its use. The arclength at converged step. ) This is used in restarts in the postbuckling region. the use of DIAG 51 is recommended only for small test problems. In the module it will be set t o + l when the iteration is converged and K is updated. The data marked with an asterisk (*) are used only for the arclength method whereas the data marked with a bullet ( 0 ) are used only for Newton's method. ARCLG Input/outputreal. the solution process is terminated. The value is set to 1 at the start of each user specified load step. The sign of ( P ) ~ { A P at the beginning of the subcase. allowing the user t o follow the nonlinear iteration calculations step by step. When DIAG 51 is selected. 7. Flag to invoke the K update at last converged step. When DIAG 50 is requested. The default (= 1) option is the gimbal angle approach. Set t o 1 when the iteration is converged but K is not updated. TWODIV Input /outputinteger . DIAG 51 is intended for debugging purposes. only the data marked with a plus (+) are printed.
) Problem statistics (gsize.u.) Projected NEWMU: pi+1 For each iteration: Displacement increment: Aul Initial energy New displacement vector: ug Nonlinear force vector: F.u. etc.) File control block Input file status External load increment for subcase: A P Displacement increment: Aup Initial arclength Initial nonlinear force vector: Fg Initial sum of nonlinear forces including follower forces: Fl Initial displacement vector: ul KFSNL*DELYS: [ L ~ ~ ] ~ { A u ~ } Initial unbalanced load: Rl Cumulative displacement increment: {u.+~ . etc. Sum of nonlinear forces including follower forces: Fl New displacement vector: ul New unbalanced load: R 1 Denominator of EUI Denominator of EPI Contraction factor: q Remaining time Displacement increment: A U ~ Arclength ratio d p calculations * * * For each quasiNewton vector set: Condition number: X2 quasiNewton vector: 6 quasiNewton vector: y 1 Energy error: z = 67. + ~ .* *+ * * * Core statistics (ICORE.} Dot product determining sign: { A ~ ~ } ~ { u .3 For each line search: Previous line search factor: ak Previous error: Ek New line search factor: ak+1 .
*sTNL. rorn K upate KPICTNL. m L ) S O U = SOLCW*l  + .1. Fetch 9 . W W T O @5 EQUIV PTELtWL.O Icc. PPTEVII i Fur* d a t w e for 9 1 A U St paruieters Process M 8 and i f clone 4 (END) Fetc?~W. delta blocks (W. KGCNL naft tor LGDIS clock for I KMmL t o KAANL KLLT = M L + KAFINL Form KFST and KSST 4 R B C U KDMi to KDLL for LWISP K L U = KLLT + KDLL fOT W I S P &cmrne KLLT or KLLR b U L T L YES Figure 7.GV.pr L ns 9 Set the F O T ~EUWL. N(.Save wid k p ~ e n df i l s Initialize parameters I 4. tiCJECT1 conpute load factor Cenerhte 1086 vector Chea ccrtstraint m q e (DCttC.la Flow Diagram for Nonlinear Process in SOL 66 . P E W L ) 1 9 sinplsri ty process for +set poinu Reset LET parameters USTPRT i f reqWteU RWuce KGG to K M L I ' FB6UC. a L LW PPTEW. NPGX.2. MLFG. ITERID * 0.
8 MLPL = MERRPL Fetm M U . m P U = ~+~ DELF'M = D L G U J EP M G Reduce trELPC.EUHTA. QW Incrmt S O M Figure 7.UGWtD. PTEUINL. ERW'L. WX. U W .lb Flow Diagram for Nonlinear Process in SOL 66 . DFGNL..2. t O DPLG h l y for tiw Subcase 1 L . . uplate W C T I store CVECT1. Reset parmeters ELDATA 0for all Update UGV = HUGVMWS Store only i f oset exists hpute = F P +K U +K Y S S f S f S s s . W .1 t o ELPLU E a d MLPLV i n t o W V i l mn A + lRUiL #la PFI. SLTi + + 4 E m L . = PWTT*I ISELPL = D E L P L V + K W L POUTOLD + YES 1 v . PPTEVVI NLCOHB NUTER .
2 Flow Diagram for Appending Process in SOL 66 7.2.Fetcn ELDATk 8BCID = 0 nwm = o ED N = S)LPMNU)@ LEND = K S 1 oxpsi te e l m t w sues recovery force.R QCV b QGS A p n C'uECTl b C 1 Fed D6 A p n N F N U + OEFNLU Fed qpem OESNU + OESNUX' Figure 7.2  8 . failure im A#lend O E S 4 U X AppClM 6~ A ed mn 4 + OESHU 1C + 1V PGX +P.
the nonlinear solution algorithm will begin in an incremental fashion to remove the load. MOMENT. The load for a subcase is subdivided into the number of increments specified for the subcase.3. FORCE2 defines the direction of the force as a vector which is the vector product of two other vectors. The loads described below make up the valid subset applicable for nonlinear static analysis. and MOMENT2. The second category consists of the following: FORCEl. which defines an enforced motion of a grid in a specific fixed direction. If the load reaches the desired level in a subcase and if the load description is left out in the subsequent subcase inadvertently. The first category consists of the . The loads are selected in each subcase by a Case Control command LOAD. MOMENTI.1 Concentrated Loads The concentrated loads consist of two categories: those which are stationary in direction through out the analysis and those which follow the grid motion. The three FORCE entries differ only in the manner in which the direction of the force is specified.2 Distributed Loads The distributed loads consist of two categories: those which are stationary in direction through out the analysis and those which follow the element motion. 7. not necessarily the same as the loaded grid points. FORCE uses the components of a vector.3. The solution strategy in nonlinear is to apply the loads in an incremental fashion until the desired load level is reached. The first category consists of the following: FORCE.7. The distinctions between the three MOMENT entries are similar.3 LOADS OVERVIEW The static loads in nonlinear analysis consist of a subset of the total MSCINASTRAN static load set. Most of the relevant loads data applicable t o the linear static analysis are also applicable t o nonlinear static analysis except for the Bulk Data entry DEFORM. Each grid point with an enforced displacement must also appear on an SPC or SPCl entry. The algorithms "remember" the loads from one subcase to the next. and SLOAD. 7. The SLOAD entry is used t o apply loads t o scalar points. FORCE1 uses two grid points. FORCE2. The enforced displacements can be applied as loads using the Bulk Data entry SPCD.
and PLOAD4. The centrifugal loads t o be specified in the Bulk Data entry RFORCE is used t o define the components of a spin vector which is used internally t o compute centrifugal forces. However. TEMPD.3. TEMPPI. The components of the gravity vector are multiplied by the mass matrix t o obtain the components of the gravity force a t each grid point. PLOAD2. The gravity loads which can be specified in the Bulk Data entry GRAV are stationary in direction through out the analysis. TRIA3. The TEMPRB provides for average temperature at ends A and B for the ROD and BEAM element. The PLOAD entry is exceptional in that it references three or four arbitrarily located grids rather than an element. It is used to define the direction and magnitude of the gravity vector in some identified coordinate system.4 Thermal Loads Thermal loads are selected in each subcase by the Case Control command TEMP(L0AD). The PLOADl entry defines concentrated and linearly distributed forces and moments to the BEAM element. and TEMPRB The TEMP and TEMPD provide for grid point temperature specification which are then interpolated to points within elements. the stress calculation for nonlinear BEAM element ignores the distributed loads. 7. and QUAD4 elements which need not be normal to the surface. which require specification of mass data.3. The PLOAD2 entry defines a uniform static pressure load applied to the QUAD4 or TRIA3 elements. The PLOAD4 defines a pressure load on surfaces of HEXA.The second category consists of the following: PLOAD. PENTA. Each component of the spin vector is multiplied by a scale factor. 7. For the nonlinear QUAD4 and TRIA3 elements the T E M P P l provides for an average temperature and thermal gradient in the thickness direction and the TEMPP3 provides for temperature at points through the thickness.3 Mass Related Loads The mass related loads are gravity and centrifugal loads. TEMPP3. but lump them onto the nodal points. Loads may be applied to the interiors of finite elements. Thermal loading which is a follower loading is accomplished through the following entries: TEMP. Aver .
The CLOAD defines a static load combination for superelement loads acting on residual boundary and the LSEQ defines a static load combination for superelement upstream loads. FORCE. The static loads (LOAD. etc. TEMP(LOAD). The LID and TID field entries point t o the appropriate load and temperature distribution entries. The Case Control TEMP(IN1T) is required above subcase level. This limitation was removed in Version 66. Segregating these temperatures is not meaningful in the nonlinear analysis. This limitation was caused by the initial temperature which was specified in the material reference temperature field. Subcases that do not contain a TEMP(L0AD) will default to TEMP(INIT). The CLOAD entry is designed t o apply static loads t o upstream superelements by combining loads defined in LSEQ entries. In addition. Version 66 allowed TEMP(MAT) option in SOL 66. until Version 65. The CLOAD defines a static load as a linear combination of previously calculated superelement loads defined by the LSEQ. In addition. which caused a number of errors. the material temperature. The IDVi fields correspond directly t o the numeric labels defined via the DAREA fields of the LSEQ entries. TEMP(MAT) was defaulted t o TEMP(L0AD) in Version 66. Any loads referenced by CLOAD should not be again referenced by LOAD until Ver . The LOAD provides for the linear combination of static load sets. which implies no thermal load. 7. For upward compatibility. TEMP(MAT). LSEQ). For these reasons.5 Loads Combination Various load specifications can be combined using LOAD and (CLOAD. The (CLOAD.3. LSEQ) combination allows the nonlinear algorithms t o apply in an incremental fashion on the upstream superelement loads to the boundary of the residual structure. a Case Control command TEMP(IN1T) was created for the initial temperature distribution.age temperature specified directly for an element will take precedence over the temperature interpolated from the elements connected grid points. was distinguished from the loading temperature. There was a limitation until Version 65 which required specification of only one increment in the subcase where thermal load is first applied or removed.) applied t o the upstream superelements cannot be referenced by a Case Control command LOAD. TEMP(MAT) option was abandoned in Version 67 and TEMP(IN1T) was made a required data in case any thermal load is present in the analysis. In order to maintain a consistent temperature field. The LSEQ assigns load vectors t o the superelements and numerically labels them via the DAREA field entry value. Thermal loads are often used t o provide initial strains. The CLOAD must be selected in the residual solution subcase via the case control CLOAD entry. The use of TEMP(MAT) or TEMP(B0TH) will cause a fatal error.
This deficiency has been corrected in Version 67.. otherwise the load will be doubled: e. An example for the Case Control setup is shown below: SEALL=ALL LOADSET = 1000 $ Selects LSEQ 1000 for upstream loads SUPER = ALL $ Identify SEs to process DISP = ALL $ NONLINEAR SOLUTIONS FOR RESIDUAL SUPERELEMENTS SUBCASE 10 CLOAD = 1001 $ Refers to CLOAD Bulk Data NLPARM = 12 $ Iteration control . 7. SUBCASE 20 CLOAD = 1002 LOAD = 10 $ Residual SE load NLPARM = 22 . The LSEQ is selected by a LOADSET Case Control command above any residual subcase.g. The hierarchy of the loads data is shown schematically in Fig.3. GRAV and TEMP. Only one LOADSET may appear in Case Control.1.sion 66..
3.1 Hierarchy of Load Data .Upstream SE 7 CLOAD ( Bulk CLOAD 1 defined on LSEQ DAREAl correlated t o CLOAD DAREA2 LSEQ + L L Static Loads Figure 7.
respectively. In the stress path dependent problem.4 NONPROPORTIONAL LOADING The loading condition under which all the stress components increase in the same proportion is called proportional loading.2 for the stress and strain path dependent problems. 7. 7. The input data are shown in Tables 7. in which the stress path was measured while the strain path was traced. First. 7. By virtue of the subincremental scheme within the material routine.l(b).4.1 signify the elastic limit and the turning point into the nonproportional loading. the solution was obtained for a prescribed stress path (shown in Fig. the loading was applied via the LOAD and PLOAD4 Bulk Data entries while in the strain path dependent problem. It can be shown that the stressstrain relations reduce to stepwiselinear equations in the case of proportional loading. As illustrated in the figures.l(a)). the nonproportional loading did not pose any difficulty.4. Notice that the linear strain path does not always result in the proportional loading condition.1 and 7.7. The solution was validated by obtaining a solution to an inverted problem. Notice that points A and B in Fig. 7. a parametric study was performed by taking a various number of global increments (2 t o 10) between points B and D. discrepancy between the two solutions is not discernible. and the strain path was plotted in Fig.e.4.. respectively. Two classes of problems in this category can be contemplated: a strain path dependent problem controlled by a geometric (or essential) boundary condition and a stress path dependent problem controlled by a force (or natural) boundary condition.4.1 Validation of Nonproportional Loading Case The general cases of nonproportional loading add more complexity to nonlinearity. Proportional loading is in general difficult t o achieve except for very simple and special cases such as uniaxial loading. In order t o examine the accuracy with respect t o the number of increments in the nonproportional loading region. The stress gradient vector 1%) remains unchanged during proportional loading.4. and the stress state may be expressed in terms of any one stress component.4. i. The material properties were specified using the MAT1 and MATS1 Bulk Data entries and the solution strategy was provided by the NLPARM Bulk Data entries. Errors are measured at points C and D for five different cases as tabulated below: . Both cases were validated by a simple model of a single HEXA element with eight Gauss quadrature points using the von Mises yield criterion and the isotropic hardening rule. the SPC and S P C l Bulk Data entries specified the loading conditions.
a.2(a) and 7. Appropriate boundary conditions are provided by the S P C l and MPC Bulk Data entries. The solution of the strain path (E. the entire loading path could have been analyzed in three steps (one in proportional and two in nonproportional loading) for engineering accuracy.2 Biaxial Loading Case for Neutral Loading Neutral loading.2(d). it is slightly different because of the kinematic hardening rule employed in the analysis. The plate is modeled by four QUAD4 elements with the material properties specified in the MAT1 and MATS1 Bulk Data entries.Error in Stresses (%) NO. The fact that the solution did not improve from eight to ten steps indicates that the number of subincrements is reduced when the global increment size is sufficiently small due t o the adaptive subincremental scheme. 5 0 2 Point C 0. E.4. + The analysis was performed in 36 increments (7 in radial loading and 29 in nonproportional loading). neutral loading. It is observed that smaller increments tend to reduce the error. 7.4.1]. . the second loading phase is nearly.2(b). is verified by a biaxially loaded plate under the inplane. Effectively. which is a special case of nonproportional loading. 7.4.2(e). followed by a slight unloading to 40 ksi.0001 %. vs. . The stressstrain response curves in x and y components are plotted in Fig. Therefore. respectively) and line searches (7 and 18. The subsequent at = 402. which ran t o completion without any difficulty in convergence.4. where the incremental step size is larger than other steps. t o be a yield locus.208 I I I I I I Note: Errors in B are less than 0. required more iterations (7 and 10.162 Point C 0. Although the locus seems loading path is defined by an elliptic locus a.3. 145 iterations and 36 line searches.185 Point D 0. The loading sequence starts with a uniaxial tension in the xdirection t o 40. as shown in Fig. 2 (. The finite element model and the material properties are shown in Figures 7.5 ksi. The input data are given in Table 7. The incremental process progressed with 11 stiffness matrix updates. The response is highly nonlinear during the second loading phase due t o the nonproportional loading path.2(c). but not exactly. The loading path is shown in Fig. an acceptable solution was achieved with only two increments. The loading is applied via the FORCE Bulk Data entries and the solution strategy is specified in the NLPARM Bulk Data entries.486 Point D 0.4.4.) corresponding t o the loading path is compared with the one presented by Chern [7.4. 7.a. of Inc. respectively). However. planestress condition using the von Mises yield criterion and kinematic hardening. 7. It is noted that increments 12 and 13.
1 Validation of Nonproport ional Loading Case . (ksi) & Stress Path Specified + Strain P a t h Specified (Inverted Case) (a) Stress Path A (b) Strain Path Figure 7. (ksi) 7 Initial Yield Locus a.4.a.
2(b) Material Curve for Biaxial Loading + I I 1 I MPC RELATIONS (ux3 = u x 6 = u x 9 ) . inch Thickness = 0.2(a) Finite Element Model for Biaxial Loading E = 30.5 x lo3 ksi Kinematic Hardening 7.001 inch Figure 7.3 H = 1.4.0 x lo3 ksi v = 0.4.MPC RELATIONS(uy7 = uYs 28 9 1 BI 5 El l I II p4 1 &2 P I II II I I I Width = Length = 1.
Figure 7.2(c) Biaxial Loading Path STEP NO. 1 2 3 4 5 6 7 8 to 36 .bv (ksi) w a (ksi) .4.
2(d) Corresponding Strain Path for Biaxial Loading .Chern 0 MSCINASTRAN Figure 7.4.
2(e) StressStrain Response Curves .4.a. (ksi) Figure 7. or a.
PTITLE = FRONT VIEW FIND SCALE. SET I PLOT SET 1. 0. LABEL BOTH PLOT STATIC 0.V66 $ CJS 7MAY85. SET I PLOT SET I. O. 0. I. 0. ORIGIN I AXES Z. 20.50 $ PRINT MATRIX TRAILERS AND ITERATIONS 5 $ CPU MIN FOR VAX TIME CEND TITLE = CODE TEST FOR NONLINEAR ELEMENTS WITH STRESS PATH 2 ECHO = UNSORT SEALL = ALL DISP = ALL OLOAD = ALL SPCF = ALL STRESS(PL0T) = ALL SPC = 100 SUBCASE 1195 LABEL = UNIAXIAL COMPRESSION LOAD = 1195 NLPARM = 10 SUBCASE 1196 LABEL = PLANE STRAIN TENSION LOAD = 1196 NLPARM = 20 OUTPUT(PLOT) CSCALE = 1. 0. SET I. AUTO SEMI 0.. 0. I.4. ORIGIN I BEGIN BULK $ NLPARM 10 NLPARM 20 $$ GEOMETRY 1000 GRID 1100 GRID 1110 GRID 2 4 0. 0.. ORIGIN1 PLOT STATIC 0. Y. SHL 1981 66 $ NONLINEAR STATIC ANALYSIS SOL DIAG 8. 20 UPW UPW YES YES ..3 PLOTTER NAST SET I = ALL VIEW 30. ORIGIN I PLOT STATIC 0. I. SET 1 PLOT SET 1. MX PTITLE = TOP VIEW FIND SCALE. PTITLE = ISOMETRIC VIEW FIND SCALE. ORIGIN I.Table 7. ORIGIN I. SET I.1 Input Data Listing for Stress Path Dependent Case ID NLPATH2. ORIGIN I. SET I. ORIGIN I. ORIGIN I VIEW O..
5 $$ MATERIAL AND GEOMETRIC PROPERTIES MAT1 10 30.+6 PSOLID I 10 0 . I. I. I. 9999 SLOAD 100 9999 1.4 I. 121 I100 1000 I000 1001 llll 1011 ill0 llll .+3 PLOAD4 123 100 36.3 MATS1 10 PLASTIC 3. 0. 0.GRID 1010 1001 GRID GRID 1101 1111 GRID GRID 1011 GRDSET $$ CONNECTIVITY 100 I CHEXA +HXIOO 1111 1011 $ 0. I. 0. I.O $$ BOUNDARY CONDITIONS SPCI 100 123456 I000 SPCl 100 I 1010 SPCI 100 2 1001 SPCI 100 3 II00 $$ LOADING CONDITIONS LOAD 1195 2.+3 ENDDATA 1.+6 11. 123 $ UNIAXIAL TENSION IN X PLOAD4 121 100 36. 123 LOAD 1196 2.+3 $ UNIAXIAL TENSION IN Z PLOAD4 123 100 36.5+6 . 0. 1000 1100 CELAS4 9999 1.4 I.+3 PLOAD4 121 100 36.
2 Input Data Listing for Strain Path Dependent Case ID NLPATH3.50 $ PRINT MATRIX TRAILERS AND ITERATIONS DIAG 5 $ FOR VAX TIME CEND TITLE = CODE TEST FOR NONLINEAR ELEMENTS WITH STRAIN PATH 3 ECHO = UNSORT SEALL = ALL DISP = ALL OLOAD = ALL SPCF = ALL STRES = ALL SUBCASE 10 LABEL = SUBCASE 10 SPC =I0 NLPARM = 10 SUBCASE 20 LABEL = SUBCASE 20 SPC = 20 NLPARM = 20 SUBCASE 30 LABEL = SUBCASE 30 SPC = 30 NLPARM = 20 SUBCASE 40 LABEL = SUBCASE 40 SPC = 40 NLPARM = 20 SUBCASE 50 LABEL = SUBCASE 50 = 50 SPC NLPARM = 20 SUBCASE 60 LABEL = SUBCASE 60 SPC = 60 NLPARM = 20 SUBCASE 70 LABEL = SUBCASE 70 SPC = 70 NLPARM = 20 SUBCASE 80 LABEL = SUBCASE 80 SPC = 80 NLPARM = 20 SUBCASE 90 LABEL = SUBCASE 90 SPC = 90 NLPARM = 20 . V66 $ QYC 8/24/90 SOL 66 $ NONLINEAR STATIC ANALYSIS 8.4.Table 7.
PTITLE = FRONT VIEW FIND SCALE.. LABEL BOTH PLOT STATIC 0. ORIGIN 1. HX PTITLE = TOP VIEW FIND SCALE. 1. 1. 1000 1100 . SET 1. 0. ORIGIN 1 PLOT STATIC 0. 0. ORIGIN 1. 0.. 1. 0. ORIGIN 1 AXES Z. SET 1 PLOT SET 1... SET 1. SET 1 PLOT SET 1. ORIGIN 1. 0. I. 1. 1.ORIGIN 1 PLOT STATIC 0. 0.3 PLOTTER NAST SET 1 = ALL VIEW 30. SET 1 PLOT SET 1.SUBCASE 100 LABEL = SUBCASE 100 = 100 SPC NLPARM = 20 SUBCASE 110 LABEL = SUBCASE 110 SPC = 110 NLPARM = 20 SUBCASE 120 LABEL = SUBCASE 120 = 120 SPC NLPARM = 20 OUTPUT (PLOT) CSCALE = 1. 0. PTITLE = ISOMETRIC VIEW FIND SCALE. ORIGIN 1 VIEW O. 20 UPW UPW YES YES 1. ORIGIN 1. 0. O. SET 1. Y. 20. 0. 1. ORIGIN 1 BEGIN BULK $ NLPARM 10 1 NLPARM 20 1 $$ GEOMETRY 1000 GRID 1100 GRID 1110 GRID 1010 GRID 1001 GRID GRID 1101 1111 GRID GRID 1011 GRDSET $$ CONNECTIVITY CHEXA 100 1 +HXlOO 1111 1011 $ SEMI SEMI 0.
122 I010 1100 1100 7.542 1010 1100 SPCl SPCl SPCl SPC SPC SPC SPC $ SPCl SPCl SPCl SPC SPC SPC SPC $ SPCl SPCl SPCl SPC SPC SPC SPC $ SPCl SPCl .674 $ 7.213 1. 9999 1.073 1.923 6.122 1.+6 11.173 1.673 2.4 SPC 10 I101 1 6.3 MATS1 1 PLASTIC 3 .123 6.4 1 10 1100 SPC 2.+6 PSOLID 1 1 0 0 $$ LOADING CONDITIONS 1010 1000 10 I SPCI 1100 1000 10 2 SPCl I100 1000 10 3 SPCl 2.3321 1001 1001 1010 1110 1111 1101 1111 1001 1001 1010 1110 llll 1101 1111 1001 1001 1010 1110 1111 1101 1111 1001 1001 1010 1110 1111 1101 1111 1001 1001 1010 1110 1111 1101 1111 1001 1001 SPCl SPCl SPCl SPC SPC SPC SPC $ 20 20 20 20 20 20 20 30 30 30 30 30 30 30 40 40 40 40 40 40 40 50 50 60 50 50 50 50 60 60 1 2 3 1100 1101 1001 1011 1 2 3 1100 1101 1001 1011 1 2 3 1100 1101 1001 1011 1 2 3 1100 1101 1001 1011 1 2 1000 1000 1000 1 1 3 3 1000 1000 1000 1 1 3 3 1000 1000 1000 1 1 3 3 1000 1000 1000 1 1 3 3 1000 1000 1010 I100 I100 1.CELAS4 9999 1.123 3.5+6 .4 SLOAD 100 9999 $$ MATERIAL AND GEOMETRIC PROPERTIES MAT1 1 30.923 1010 1100 1100 5.213 7.674 SPC 10 1001 3 SPC 10 1011 3 6.173 5.073 2.673 1010 I100 1100 3.542 1.
922 1010 1100 1100 2.682 1.922 2.282 3.263 9.282 1001 1001 1010 1110 llll 1101 1111 1001 1001 1010 1110 1111 1101 llll SPCl SPCl SPCl SPC SPC SPC SPC $ SPCl SPCl SPCl SPC SPC SPC SPC $ SPCl SPCl SPCl .263 1.0212 2.0212 3.972 1.682 2.972 1010 1110 1111 1101 llll SPCl SPCl SPCl SPC SPC SPC SPC $ SPCl SPCl SPCl SPC SPC SPC SPC $ SPCl SPCl SPCl SPC SPC SPC SPC $ 1010 1100 1100 1.SPCl SPC SPC SPC SPC $ 60 60 60 60 60 70 70 70 70 70 70 70 80 80 80 80 80 80 80 90 90 90 90 90 90 90 100 100 100 100 100 100 100 110 110 110 110 110 110 110 120 120 120 1100 9.
792 4.032 4.032 ill0 ill1 1101 ill1 1 1 3 3 2.032 .792 4.792 2.SPC SPC SPC SPC ENDDATA 120 120 120 120 1100 1101 1001 1011 1 1 3 3 2.032 4.792 2.
MCAUTO) SUBTITLE = BIAXIAL LOADING CONDITION LABEL = REF. P.Table 7.4. A50 SEALL = ALL STRESS = ALL SPC = 10 MPC = 10 SUBCASE 1 LOAD = 1 NLPARM = 10 SUBCASE 2 LOAD = 2 NLPARM = 10 SUBCASE 3 LOAD = 3 NLPARM = 10 SUBCASE 4 LOAD = 4 NLPARM = 10 SUBCASE 5 LOAD = 5 NLPARM = 10 SUBCASE 6 LOAD = 6 NLPARM = 10 SUBCASE 7 LOAD = 7 NLPARM = 10 SUBCASE 8 LOAD = 8 NLPARM = 10 SUBCASE 9 LOAD = 9 NLPARM = 10 SUBCASE 10 = 10 LOAD NLPARM = 10 SUBCASE 11 LOAD = 11 NLPARM = 10 SUBCASE 12 LOAD = 12 NLPARM = 12  . MARCH 1972.50 $ PRINT MATRIX TRAILERS AND ITERATIONS 5 $ CPU MIN FOR VAX TIME CEND TITLE = QUAD4 PLASTIC ANALYSIS CHECK (M. V66 $ FEB84 SOL 66 $ NONLINEAR STATIC ANALYSIS DIAG 8.: FOSTER WHEELER REPORT FWR27. WASYLCZAK.3 Input Data Listing for Biaxial Loading Case ID BIAXIAL.
SUBCASE 13 LOAD = NLPARM = SUBCASE 14 LOAD = NLPARM = SUBCASE 15 LOAD = NLPARM = SUBCASE 16 LOAD = NLPARM = SUBCASE 17 LOAD = NLPARM = SUBCASE 18 LOAD = NLPARM = SUBCASE 19 LOAD = NLPARM = SUBCASE 20 LOAD = NLPARM = SUBCASE 21 LOAD = NLPARM = SUBCASE 22 LOAD = NLPARM = SUBCASE 23 LOAD = ILPARM = SUBCASE 24 LOAD = NLPARM = SUBCASE 26 LOAD = NLPARM = SUBCASE 26 LOAD = NLPARM = SUBCASE 27 LOAD = NLPARM = SUBCASE 28 LOAD = NLPARM = SUBCASE 29 13 13 14 14 I 5 10 16 10 17 10 18 10 19 10 20 10 21 10 22 10 23 10 24 10 25 10 26 10 27 10 28 10 .
0 0.5 1.001 .0 I 2 4 5 .0 0.0 0.0 0.533 NLPARM 14 I $$ GEOMETRY GRID I GRID 2 GRID 3 GRID 4 GRID 5 GRID 6 GRID 7 GRID 8 GRID 9 $$ CONNECTIVITY CQUAD4 I I CQUAD4 2 I CQUAD4 3 1 CQUAD4 4 I $$ P R O P E R T I E S PSHELL I I AUTO AUTO AUTO AUTO 0.0 0.5 1.5 1.LOAD = 29 NLPARM = 10 SUBCASE 30 LOAD =30 NLPARM = 10 SUBCASE 3 1 LOAD = 31 NLPARM = 10 SUBCASE 32 LOAD = 32 NLPARM = 10 SUBCASE 33 LOAD = 33 NLPARM = 10 SUBCASE 34 LOAD = 34 NLPARM = 10 SUBCASE 35 LOAD = 35 NLPARM = 10 SUBCASE 36 LOAD = 36 NLPARM = 10 B E G I N BULK $$ PARAMETERS PARAM NLAYERS I $ NLPARM 10 I NLPARM 12 1 NLPARM 13 1 +ILP13 3.
07 30. 32.05 15.5 38.5 3. 36.MAT1 I 30. I.5 32.5 39. 42. 37.5 11.44 34. I.53 46.5 40.1 41. +6 MATS1 I $$ CONSTRAINTS MPC I 6 MPC I 9 MPC I 7 MPC I 8 SPCI 1 3456 SPCl 1 2 SPC1 I I $$ LOADING I 3 FORCE FORCE 2 3 FORCE 3 3 FORCE 4 3 FORCE 5 3 FORCE 6 3 FORCE 7 3 FORCE 8 3 FORCE 8 9 FORCE 9 3 FORCE 9 9 FORCE 10 3 FORCE 10 9 FORCE 1 1 3 FORCE 11 9 FORCE 12 3 FORCE 12 9 FORCE 13 3 FORCE 13 9 FORCE 14 3 FORCE 14 9 FORCE 15 3 FORCE 15 9 FORCE 16 3 FORCE 16 9 FORCE 17 3 FORCE 17 9 FORCE 18 3 FORCE 18 9 FORCE 19 3 FORCE 19 9 FORCE 20 3 FORCE 20 9 FORCE 2 1 3 FORCE 21 9 FORCE 22 3 FORCE 22 9 FORCE 23 3 . 6. 46.188 23. 2 I I I THRU 2 4 30.48 40. 1.3 PLASTIC 1. 37.5+6 I I 2 I. 40. 40. 45.89 44.19 43. 40.966 43. . 46. 44.02 41.5 40.13 20.05 44.5 1.5 35.3 40.41 25.
FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE $ 23 24 24 25 25 26 26 27 27 28 28 29 29 30 30 31 31 32 32 33 33 34 34 36 35 36 36 ENDDATA .
7. Each SPC entry may define enforced displacements for up t o two grid or scalar points. 7. Similar to SPC.5. only the SPCADD entry will be used. In the former case. Notice that SPCADD entries take precedence over SPC or SPCl entries. the enforced displacement values may be input directly in the D field of the SPC entry. An MPC entry in the Case Control section is used t o select a multipoint constraint set t o be applied t o the structural model. Each MPC Bulk Data entry may be used to define a constraint equation involving a group of degrees of freedom in which the first degree of freedom is assumed to be the dependent degree of freedom and included in the mset. If both have the same SID.5. All the degree of freedom in mset will be condensed out prior to the matrix operations.1 Boundary Conditions In the Case Control section.2 Multipoint Constraint While SPC is t o constrain the motion by specifying a known value. Several SPC entries which reference the same SID may be used if enforced . an SPC entry is used to select a single point constraint set (sset) which will be applied t o the structural model.5 CONSTRAINTS AND ENFORCED MOTION In addition to the loading conditions. SOL 66 (or SOL 106) allows the user to change the boundary conditions or enforce the displacements at the specified grid or scalar points. Their response will be directly recovered from those of the independent degrees of freedom according t o the specified constraint equation. 7. SPCl or SPCADD Bulk Data entry. But these constraints are socalled permanent constraints which can not be changed during the analysis. The PS field in the GRID entry is also able to specify singlepoint constraints associated with a grid point. The specified MPC set identification must appear at least in one MPC or MPCADD Bulk Data entry. MPC can also be changed from subcase t o subcase by specifying a different MPC set ID inside each subcase in static analysis (SOL 66 or SOL 106).3 Enforced Motion The enforced displacements may be specified in the Bulk Data section using SPC or SPCD entries. MPC is to tie the motion of a degree of freedom t o other degrees of freedom. The specified set identification must be identical to the SID field of an SPC.7. A significant application of SPC is the imposition of boundary conditions.5. An advantage of using SPC to specify boundary conditions is that these boundary conditions can be changed from subcase to subcase by simply selecting a different SPC set inside each subcase.
then the former will override the latter.0 inches.1 shows a steel clip which will be pushed into a cavity whose height is 5. 7. In this case. 7. in the first restart run the enforced displacement at GRID 18 is released and another enforced . The only disadvantage of this method is that the entire sset must be redefined if the enforced displacement conditions vary among subcases. The table below summarizes the data entries required for the enforced motions or the boundary condition changes in MSC/NASTRAN. It is obvious from Fig. may also be combined with SPCD. it should be noted that the Bulk Data LOAD combination entry will not combine an SPCD load.displacements for more than two grid or scalar points are desired. Since only the LOAD set needs to be changed instead of the entire sset.264 inches. this method of applying the enforced displacements is more efficient than the SPC entry method when more than one enforced displacement condition is applied. Therefore. The result shows that the height of the deformed clip a t GRID 19 is 5. all the enforced degreesoffreedom must be included in the sset by SPC or S P C l entries. and therefore a LOAD set must also be selected in the Case Control section. Since the height of the clip at GRID 18 is 5.1 that GRID 18 will be the first grid point to come in contact with the ceiling of the cavity. The SPCD entry is requested as a LOAD. an enforced displacement of 0. which is still greater than the height of the cavity. PLOAD.4 Example Fig.5 inches in the y direction is applied at GRID 18 in the cold start run. The goal of the analysis is t o predict the final configuration of the clip. It is assumed that there is no friction force a t the contact surface. Since a part of the clip bottom might lift off the ground during the process.5.0 inches. Option 1 Case Control Entries SUBCASE 1 SPC = 100 SUBCASE 2 SPC = 200 SUBCASE 1 SPC = 100 LOAD = 10 SUBCASE 2 SPC = 100 LOAD = 20 Bulk Data Entries SPC 100 SPC 200 2 SPC(or SPC1) 100 SPCD 10 SPCD 20 7. Other Bulk Data load entries such as FORCE. etc. FORCE1. However. The clip is modeled with 24 elastic BEAM elements.5. One way t o avoid this condition is t o define the enforced displacements using SPCD entries. PLOADi.5 inches and that of the cavity is 5. If the enforced displacements are specified on both the SPCD and SPC entries. 7 GAP elements are provided at the clipground interface.5.
Undeformed 21 'Y 20 [ I l l I Figure 7. . A second restart is performed to release the constraint at GRID 19 and enforce the displacement at GRID 20 by 1.0) is applied in the y direction at GRID 19.0 . At this time.0 = 1. 7. the analysis results indicate that no grid point in the model has a height over 5..0 inches (i.1. Elastic Clip Pushed into a Cavity .6...5 inches in the y direction.displacement of 1. Notice that the specified enforced displacements should always refer to the undeformed configuration and MSCJNASTRAN will automatically compute the increments.e.5.5.. 5.156 inches. The final deformed shape of the clip is shown in Fig.. . which is still greater than the cavity height... The predicted heights for some of the grid points are tabulated below at different stages with the values greater than 5.. which means that the entire clip is inside the cavity.0 inches. The analysis from this run predicts that GRID 20 will attain a height of 5.0 in italics.1. nr 0  22 '0 / e.
LOOPID.Table 7.50 DIAG CEND TITLE = INSERTING A SEALL = ALL ECHO = BOTH DISP = ALL SPCF = ALL $ $ added for the restart runs $ SSH 12/12/90 $ $ NONLINEAR STATIC ANALYSIS $ CLIP INTO A CAVITY (BEAM MODEL) $PARAM.2 $ $ $ used in the 1st restart $PARAM. 2 $PARAM. 0. V67 5 TIME SOL 66 8. 0.LOOPID.SUBID.5.1.SUBID.7 $ $ used in the 2nd restart $ SUBCASE 10 SPC = 10 LOAD = 100 NLPARM = 1 $ $SUBCASE 20 $ SPC=20 $ LOAD=200 $ NLPARM = $ 2 $ $ added for the 1st restart $ $ $ $ added for the 2nd restart $ $ $SUBCASE 30 $ SPC=30 $ LOAD=300 $ NLPARM = $ 3 BEGIN BULK PARAM LGDISP 1 $$ SOLUTION CONTROL 2 NLPARM I 5 NLPARM 2 NLPARM 3 4 $ $$ $ 0.3 $PARAM. SEMI AUTO AUTO 1 GEOMETRY I 2 3 GRID GRID GRID . Input Data Listing for Elastic Clip $RESTART ID CLIPB.
GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID $ $$ $ 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 106 107 108 109 110 Ill 112 ELEMENT DEFINITION 101 102 103 104 105 106 107 108 109 110 Ill 112 113 114 100 100 100 100 100 100 100 100 100 100 100 100 100 100 CBEAM CBEAM CBEAM CBEAM CBEAM CBEAM CBEAM CBEAM CBEAM CBEAM CBEAM CBEAM CBEAM CBEAM .
206 207 208 209 210 211 212 PROPERTIES 100 100 200 210 PBEAM MAT1 PGAP PGAP $ $$ $ LOADING I0 20 30 100 200 300 SPCI SPCI SPCI SPCD SPCD SPCD $ ENDDATA .CBEAM CBEAM CBEAM CBEAM CBEAM CBEAM CBEAM CBEAM CBEAM CBEAM CORD2R +CORI CGAP CGAP CGAP CGAP CGAP CGAP CGAP $ $$ $ 115 116 117 118 119 120 121 122 123 124 I 0.
SUBID.l) is only applicable in Version 65.) terminated analysis. In Version 66. The present section will concentrate on these two types of restarts. Note that the small database option (PARAM. Consequently. In addition. The most common reason is to extend a normally or abnormally (due t o divergence or CPU time expiration. one may use SOL 66 t o preload a nonlinear structure and then restart into either SOL 66 for a nonlinear buckling analysis. Therefore.. If PARAM. In this case. solution data blocks corresponding to the intermediate load steps not requested for output (by INTOUT field in NLPARM) are deleted from the database after a subcase is completed. the DMAP was modified to store only the data blocks for the output load steps without requiring parameter SMALLDB. the solution can only be restarted from the end of a subcase if the default (INTOUT=NO) is used.e.6. Restarts may also be used only for obtaining additional output. 7. Further improvements were made in Version 67 so that the data blocks for the last converged load step will be saved if the solution is terminated in the middle of a subcase due t o divergence or consumption of the specified CPU time.76 .2 Restart Procedure Restarts in SOL 66 are controlled by three user parameters which may be specified either in the Case Control section prior t o the subcases or in the Bulk Data section: PARAM. any converged load step in the current subcase or from the end of any previous subcase.SMALLDB is specified as 1. LOOPID.1 Preliminaries In Version 65 the data blocks corresponding to every converged load step in SOL 66 are saved for restarts by default. however. n (only applicable in V65 and V66) . the restarts are restricted t o those starting from the solution with the corresponding data blocks stored in the database. etc. 7. t PARAM. In addition. or SOL 99 for a nonlinear transient analysis.6. RESTARTS Restarts may be performed for various purposes.SMALLDB. restarts from the unstable region for the arclength methods were also made available in Version 67. LOADINC. i. SOL 63 for a normal mode analysis. the solution can always be restarted from the termination point regardless of the value of INTOUT in the NLPARM entry. m PARAM.
one may set the new parameter values LOOPID=3. SUBID selects the subcase in which the restart analysis is initiated and LOADINC selects the load step number in that subcase.e. LOADINC=4 and leave out the second subcase. Assuming that the second subcase specifies the same LOAD and NLPARM entries as the first subcase (i.. Assume that the solution diverges and the third load increment (assume its LOOPID = 3) is the last solution step saved for the restart. . SUBID=2. This means that the solution will restart into the very first load step of a new subcase. The forcedisplacement curve of the rod is shown in Fig. 7. the solution will continue with the new LOAD and NLPARM data specified in that subcase. the loading is increased from 0 t o 2 lbs in two load increments and the analysis executed t o completion.3 x 0. the solution will continue with a load increment of ( P .6. load P with 5 increments). SUBID=l. one may also set LOOPID=3.e. In the restart data deck. The default values for LOADINC and SUBID are 1.2P for each increment.6. In Versions 65 and 66. PARAMeters SUBID and LOADINC refer t o the initial load step for the restart. the solution will restart into the fourth load increment to complete the remaining part of the first subcase. 7. LOADINC=l and add a second subcase in the Case Control section. + For example. a value greater than 1 may be specified for LOADINC if the new SUBID value is the same as that which appeared in the previous run.1 shows a nearlyhorizontal simple rod whose left end is hinged and right end is simply supported. where a vertical force is applied. If the second subcase is not added. it is removed from SOL 66 in Version 67. Using the example described in the preceding paragraph. only the data recovery will be processed.g.6. which may be used as initial condition for a subsequent restart. A new subcase is added in the restart run to increase the loading t o 4 lbs using two load increments. If the new subcase is not added to the restart data deck. i. This is a snapthrough problem with a single degreeoffreedom. If a subcase with the new SUBID value is added to the Case Control section of the restart data deck. new SUBID = last SUBID 1) and t o use the default for LOADINC (i. 0..1. Since this feature has only limited applications and is very difficult to use with the adaptive load/arclength method. a load P is applied in one subcase with 5 increments.2. is connected in the vertical direction to the right end of the rod. In this case. This analysis diverges in the second load increment.3 Example Fig.e. only the data recovery (without any computation) will be processed in the rest art analysis.6.2P)/5 = 0.2.The integer 1 identifies data blocks for a specific converged solution stored in the data base..6. An elastic spring of 3 lbs/in. LOADINC = 1).08P. The value of LOOPID as well as SUBID and LOADINC for each solution step is printed with the iteration information in the output.. 7. The simplest way t o perform a restart analysis is t o specify a new value for SUBID which is greater than the SUBID value appearing in the last converged solution step of the previous run (e. In the cold start run whose input data is given in Table 7. The input data for this run is given in Table 7.
0 Figure 7.._ _ _ _ _ _ ( . T h e solution is restarted again using the arclength method.6.3. Mi4SINC is intentionally set t o 2 in the SLPC'I entry so that the analysis would terminate with a unstable solution.6. ______ .6. Figure 7..1 Model of Rod with Spring 4oa. the solution is restarted a third time t o reach the loading of 4 lbs.e.5) is the last converged solution step which is saved for the restart (except Version 66).4.6. In this run. whose input d a t a is given in Table 7.2 ForceDisplacement Curve .5 1 0 0 1 0 s 2. LOOPID 3 (LOAD S T E P 1. Finally.i..3rd Restart 0. The input d a t a for the third restart is given in Table 7.
1. GRID 2 100.6..E7 $ LOADING FORCE I 2 $ PARAMETERS PARAM LGDISP +I NLPARM I 2 $ ENDDATA . 0. 0.V67 $ SSH 6/17/91 $ TIME 5 SOL 66 $ DIAG 8.I PROD 10 PELAS 20 3. $ CONNECTIVITY 10 I 2 CROD 10 CELASI 20 20 2 2 0 0 $ PROPERTIES I . COMP. AND STRUCT. 7:125136 (1977) ECHO=UNSORT SEALL=ALL OLOAD=ALL DISP=ALL SUBCASE 10 $ LOAD = 2 LB. LOAD=I NLPARM=I BEGIN BULK $ GEOMETRY GRID I 0. MAT1 1 10. 0.Table 7.1 Input Data Listing for Cold Start ID SPROD.50 $ CEND TITLE=SIMPLE ONE DOF GEOMETRIC NONLINEAR PROBLEM SUBTITLE= COLD START RUN LABEL=REF: STRICKLIN AND HAISLER.
0 ENDDATA . COMP.50 $ CEND TITLE=SIMPLE ONE DOF GEOMETRIC NONLINEAR PROBLEM SUBTITLE= FIRST RESTART (INCREASES THE LOAD UP TO 4 LBS) LABEL=REF: STRICKLIN AND HAISLER. LOAD=:! NLPARM=2 BEGIN BULK $ GEOMETRY I 0. LOAD=1 NLPARM=I SUBCASE 20 $ LOAD = 4 LB.Table 7. MAT1 I 10. 7:125136 (1977) $ ECHO=UNSORT SEALL=ALL OLOAD = ALL DISP = ALL PARAM. GRID $ CONNECTIVITY 10 10 I CROD CELASI 20 20 2 $ PROPERTIES 10 I ..2 Input Data Listing for First Restart RESTART ID SPRODRI. 0.6. 1. SUBID.I PROD PELAS 20 3. GRID 2 100. LOOPID. 2 SUBCASE 10 $ LOAD = 2 LB. AND STRUCT.E7 $ LOADING 2 FORCE I FORCE 2 2 $ PARAMETERS PARAM LGDISP +I NLPARM I 2 2 NLPARM 2 +NLP21 +NLP22 0 $ 0.V67 $ SSH 6/17/91 $ TIME 5 SOL 66 $ DIAG 8. 2 2 0. 2 PARAM.
I PROD PELAS 20 3. 0...50 $ CEND TITLE=SIMPLE ONE DOF GEOMETRIC NONLINEAR PROBLEM SUBTITLE= 2ND RESTART (SWITCHES TO THE ARCLENGTH METHOD) LABEL=REF: STRICKLIN AND HAISLER. $ CONNECTIVITY 10 10 I 2 CROD CELASI 20 20 2 2 0 $ PROPERTIES 10 I .V67 $ SSH 6/17/91 $ TIME 5 SOL 66 $ DIAG 8. 0.E7 $ LOADING 2 FORCE I FORCE 2 2 $ PARAMETERS PARAM LGDISP ti NLPARM 1 2 NLPARM 2 2 +NLP2I tNLP22 0 NLPARM 3 I NLPCI 3 $ ENDDATA . SUBID.6. 7:125136 (1977) ECHO=UNSORT SEALL=ALL OLOAD = ALL DISP = ALL PARAM. GRID 2 100. 3 SUBCASE 10 $ LOAD = 2 LB. LOAD=I NLPARM=I SUBCASE 20 $ LOAD = 4 LB. MAT1 I 10.Table 7. LOOPID.3 Input Data Listing for Second Restart RESTART ID SPRODR2. AND STRUCT. LOAD=:! NLPARM=2 SUBCASE 30 $ LOAD = 4 LB. 0. COMP. ARCLENGTH METHOD LOAD=2 NLPARM=3 BEGIN BULK $ GEOMETRY I 0. 3 PARAM. GRID 1.
. 0. 0. RESTART IN UNSTABLE REGION LOAD=2 NLPARM=4 BEGIN BULK $ GEOMETRY I 0. 7:125136 (1977) $ ECHO=UNSORT SEALL=ALL OLOAD = ALL DISP = ALL PARAM. $ CONNECTIVITY CROD 10 10 I 2 CELASI 20 20 2 2 0 $ PROPERTIES .. GRID 2 100. COMP. 0. GRID 1.l PROD 10 I PELAS 20 3. MAT1 I 10. SUBID. NORMAL RESTART LOAD=2 NLPARM=2 SUBCASE 30 $ LOAD = 4 LB. 4 PARAM. AND STRUCT.6. COLD START LOAD=I NLPARM=I SUBCASE 20 $ LOAD = 4 LB..E7 $ LOADING 2 FORCE I FORCE 2 2 $ PARAMETERS PARAM LGDISP t1 NLPARM 1 2 NLPARM 2 2 +NLP2I . RESTART WITH THE ARCLENGTH METHOD LOAD=:! NLPARM=3 SUBCASE 40 $ LOAD = 4 LB. LOOPID.Table 7. 5 SUBCASE 10 $ LOAD = 2 LB.50 $ CEND TITLE=SIMPLE ONE DOF GEOMETRIC NONLINEAR PROBLEM SUBTITLE= 3RD RESTART (RESTARTS IN THE UNSTABLE REGION TO COMPLETION) LABEL=REF: STRICKLIN AND HAISLER.4 Input Data Listing for Third Restart RESTART ID SPRODR3.V67 $ SSH 6/17/91 $ TIME 5 SOL 66 $ DIAG 8.
tNLP22 NLPARM NLPCI NLPARM NLPCI $ 0 3 3 4 4 1 1 ENDDATA .
T h e pressure vessel is illustrated in Fig.1. No s t r a i n hardening Zierrkiewicz. T h e MSC/NASTRAN results agree reasonably well with the experimental results of Dinno and Gill [7. 1971. 1 0 . This difficulty is emphasized here because changes in the plate thicknesses a t the junction result in large differences in the final answers.7.7. F i g u r e 1 8 . The F i n i t e Element Method McGrawHill.7. 7.1.3] as shown in Fig. 7. 7. 4 7 4 Figure 7. Steel Pressure Vessel. T h e pressure vessel consists of a spherical shell joined by a thin cylindrical shell a t the top. T h e intersection of the spherical and cylindrical shells of the vessel is difficult t o model using plate elements. A 10' sector of the steel pressure vessel is modeled using QUAD4 and PETU'TA elements as shown in Fig.7. .7. P.2. M . T h e input d a t a for the model is listed in Table 7.7 VERIFICATION PROBLEM: ANALYSIS of a PRESSURE VESSEL This problem is used t o illustrate the nonlinear static analysis of pressure vessel using QUAD4 elements.3.1.7.
2. MSCINASTRAN Model (10% Sector of the Steel Pressure Vessel) .7.Figure 7.
7. MSCINASTRAN 7.3. Experimental Results vs.Vertical deflectlon of point 'A' ( x 10'~ In.7  3 .) Figure 7.
0.00 FIND SCALE ORIGIN 1 SET I PLOT STATIC 0 SET I AXES MY. 23. 0.. 0.V65 $ CJS 27SEP1985.27.17.7. PSI LOAD = 40 NLPARM = 30 OUTPUT(PLOT) CSCALE 1. FIND SCALE ORIGIN I SET I PLOT STATIC 0 SET I BEGIN BULK $ PARAMETERS PARAM AUTOSPC YES $ PARAMETERS FOR NONLINEAR ITERATION NLPARM 10 I AUTO NLPARM 20 3 AUTO 5 AUTO NLPARM 30 $ DEFINE GRID LOCATIONS CORD2C 10 0. 15MAY1984 SOL 66 DIAG 8. 0. 0. S. +C2C1 I.1 Input Data Listing for Pressure Vessel Model ID VESSEL.913 PSI LOAD = 10 NLPARM = 10 SUBCASE 2 LABEL = PRESSURE LOAD OF 1080. 0.. PSI LOAD = 20 NLPARM = 20 SUBCASE 3 LABEL = PRESSURE LOAD OF 1200. X VIEW O. PSI LOAD = 30 NLPARM = 30 SUBCASE 4 LABEL = PRESSURE LOAD OF 1400.3 SET I ALL MAXI DEFO 1.50 TIME 30 CEND TITLE = PRESSURE VESSEL SUBTITLE = MATERIAL NONLINEAR ANALYSIS ECHO = UNSORT SEALL = ALL DISP = ALL ELST = ALL SUBCASE I LABEL = PRESSURE LOAD OF 749.0 VIEW 34. W W W YES YES YES . MZ.Table 7.
GRID I 20 GRID 2 20 GRID 3 20 GRID 4 20 GRID 6 20 GRID 5 20 GRID 7 20 GRID 8 20 GRID 9 20 GRID 10 20 GRID 11 10 GRID 12 10 GRID 13 10 GRID 14 10 GRID 15 10 GRID 16 10 GRID 17 I0 GRID 18 10 GRID 19 10 GRID 20 10 GRID I01 20 GRID 102 20 GRID 103 20 GRID 104 20 GRID 105 20 GRID 106 20 GRID 107 20 GRID 108 20 GRID 109 20 GRID 110 20 GRID Ill 10 GRID 112 10 GRID 113 10 GRID 114 10 GRID 115 10 GRID 116 10 GRID 117 10 GRID 118 10 GRID 119 10 GRID 120 10 GRID 201 GRID 202 $ qUAD4 ELEMENTS CqUAD4 I I CqUAD4 2 I CqUAD4 3 1 CqUAD4 4 I . 0.CORD2S 20 +C2S2 I.
25 0.3 $ DEFINE LOADING CONDITIONS PLOAD2 10 749.15625 0.125 0.28125 CQUAD4 10 3 10 110 I11 I1 +CQl 0.054t4 MATS1 I PLASTIC I I $ PENTA ELEMENT 120 CPENTA 101 4 20 1 19 119 202 20 $ PENTA CHARACTERISTICS PSOLID 4 2 0 MAT1 2 2.912+7 0.125 I PSHELL 3 I 0.3 4.25 0.912t7 0.15625 CQUAD4 12 5 12 112 113 13 tCQ3 0.28125 0.3125 0.913 PLOAD4 20 101 1080.15625 0.28125 I PSHELL 5 I 0. PLOAD4 40 101 1400.15625 I MAT1 1 2. I THRU 18 PLOAD2 30 1200.125 CQUAD4 13 2 13 113 114 14 CQUAD4 I4 2 14 114 115 15 CQUAD4 15 2 15 115 116 16 CQUAD4 16 2 16 116 117 17 CQUAD4 17 2 17 117 118 18 CQUAD4 I8 2 18 118 119 19 $ QUAD4 CHARACTERISTICS PSHELL I 1 0.28125 0.25 I PSHELL 2 I 0. ENDDATA .3125 0.3125 I PSHELL 4 I 0.913 I THRU 18 PLOAD2 20 1080.3125 CQUAD4 11 3 I1 ill 112 12 tCQ2 0.28125 0.3125 0.CQUAD4 5 I 5 105 106 6 CQUAD4 6 I 6 106 107 7 CQUAD4 7 1 7 107 108 8 CQUAD4 8 1 8 108 109 9 CQUAD4 9 4 9 109 110 10 +CQO 0. I THRU 18 PLOAD4 10 101 749.15625 0. PLOAD4 30 I01 1200. I THRU 18 PLOAD2 40 1400.
.
MSCINASTRAN Nonlinear Analysis HANDBOOK Version 67 Volume I1 Sang H. Lee .
81 612 Miinchen. JAPAN Tel: (03) 35050266 Telex: J23363 MSCWATA FAX: (03) 3505091 4 DISCLAIMER I The concepts. European Operations MacNealSchwendler GmbH lnnsbrucker Ring 15 Postfach 801240 D . Akasaka 5chome Minatoku. GERMANY Tel: (89) 431 9870 Telex: 523 784 MSG D FAX: (89) 436 1716 Headquarters. methods. All rights reserved. NA *V67 * Z * Z * Z * HBNLN .A. CA 900411777 Tel: (213) 25891 11 or (800) 3364858 FAX: (213) 2593838 Headquarters. Tokyo 107. The MacNealSchwendler 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. Far East Operations MSC Japan Ltd.S. NASTRAN is a registered trademark of the National Aeronautics and Space Administration. MSCINASTRAN is an enhanced. 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.Corporate Headquarters The MacNealSchwendler Corporation 815 Colorado Boulevard Los Angeles. a1992 by The MacNealSchwendierCorporation Printed in U. I MSC and MSCI are registered trademarks and service marks of The MacNealSchwendler Corporation. proprietary version developed and maintained by The MacNealSchwendlerCorporation. EntsujiGadelius Building 239.
By direct numerical integration. A selfadaptive time stepping method has been implemented in MSCINASTRAN Version 66 for the transient analysis of nonlinear structures. The equilibrium is ensured by iterations until the solution converges to the preset error tolerance. such as the quasiNewton update and line search process. The primary concept of automatic time step adjustment . Newmark's direct time integration method is implemented using the twopoint recurrence (or onestep) formula as a foundation of this selfadaptive time stepping algorithm [8. The optimal time step size.1. an adaptive time stepping algorithm has been implemented for the nonlinear transient analysis.1) is satisfied at discrete time steps with an interval of At.2]. The goal is to provide an automatic method of implicit time integration. which renders accurate solutions efficiently t o any nonlinear transient response problem. changes continuously in the transient dynamic environment. These procedures.Chapter DIRECT TIME INTEGRATION 81 . provide users with additional options to adjust incremental and iterative processes in search of more efficient and effective strategy for obtaining solutions. which is required for accuracy and efficiency. Since Version 62. The goal of the generalpurpose program is to provide users with tools that are easy t o use. (8. several methods of improving the convergence of nonlinear problems have been adopted in MSCINASTRAN [8. the equilibrium of Eq.1]. the nonlinear transient response is analyzed by a stepbystep integration of an equation of motion. This goal was achieved by combining the automatic time step adjustment with the bisection process in a coordinated manner. As a step toward this goal. OVERVIEW If the effects of inertia and damping forces are to be included in the nonlinear analysis.
changes in nonlinearity cannot be predicted from the deformation pattern at the previous time step. SOL 99. has evolved since Version 62. The iteration process for the equilibrium employs expeditious methods such as BFGS updates and line searches as well as an adaptive stiffness matrix update strategy. Although these solution sequences are developed for nonlinear transient response analysis.is that the proper size of the time step can be predicted based on the dominant frequency in the incremental deformation pattern at the previous time step. and userfriendliness. which was implemented in Version 67 t o utilize the new executive system constructs.2. Efforts have been made to minimize the thrashing in the algorithm (e. This deficiency is overcome by the bisection process. SOLS 99 and 129. This concept presents a deficiency of time lag involved in the prediction process. linear transient response analysis can be performed in these solution sequences in order t o take advantage of the automatic time stepping and restart capabilities.1 and 8. which is activated when any difficulties arise in terms of convergence during the iteration. Furthermore. SOL 129 is a structured DMAP with the same functionality as SOL 99. Frequent decomposition of the dynamic stiffness matrix is inevitable in adaptive time stepping using the implicit integration method..g. effectiveness.1. The transient response analysis capability is provided in the selfcontained solution sequences.1. aimed at efficiency. The present selfadaptive time stepping algorithm for the nonlinear transient analysis is proving to be a robust and practical method. The core part of SOL 99 are the nonlinear modules (NLTRD and NLTRD2) which perform the incremental and iterative processes for implicit direct time integration. which is depicted by flow diagrams in Figs. . 8. too frequent or repetitive adjustment of the time step size) and yet to provide a solution reliably t o any type of problem.
1 . UHVNL. OESNLI PNL Print LOOPID .1. IMETHOD Generate PDT. TEL Process initial conditions and store UHVNL 1 Update stiffness KANL Reduce KJJNL KDDNL KRDD = KDD + KDDNL  * . MPT T E L UHVNL. ESTNL Store and pnnt database Fetch MESTNL SOLCUR SOLPRE + 1  CASE  CASESXP. Create FDV tor IC Create KSGG for SPCF 1 1 AM3=2MDD+ZiiiBDD AM2 = zBDD AM1 = PKBDD + AM3 Decompose AM1 Run NLCOMB and NLTRD Partition FSGV FSV if end of rubcar. 1 I . Save andparameters Append F~les ln~t~al~ze Generate FDV. Reduce KDJJ + KDDD KDDD KBDD KRDD 4 = . SOLPRE = SOLPRE + 1 Figure 8.1 Flow Diagram for Nonlinear Process in SOL 99 8. . CASESX2. NLCOMB NLTRDP Print OLDDT I I Print STlME Smra FSV. FDV.3 ..
TEL. PNL. NOLlN Force I Figure 8.1. CASESX2. GOD. OESNL2.MLOOP  Output Solution Set.2 Flow Diagram for Appending Process in SOL 99 8.FSV Find the number of output time steps TOLAPP TEL t OLB Append PPNL t PP Append PSV PSSD Append UHVNL UHVF Append OESNL1 OESNLX Append PNL PNU MLOOP = MLOOP + 1 MLAST = LOOPID . GMD. OESNL2. FSV. PST PSV = PST. OESNLI .1  4 . NOLlN Force XYplot Solution Set.9 SDATREC P OLTOP Fetch UHVNL.SMPT TRLG t PPNL.
The Newmark Beta method based on the twostep integration is combined with the modifications t o Newton's method for nonlinear solutions [8.t . we obtain a load equilibrium error vector. This method was implemented in the NLTRD module in Version 62.O = Newmark Beta operator and F ( u n ) is the nonlinear force due t o a generalized displacement vector {u. tnl < t < tn+1.) at time step n by the equation: where {P.2. An identical definition occurs for { P .). From central finite differences. This method can be used by selecting the AUTO or TSTEP method in the TSTEPNL Bulk Data entry.~ n for small angles and At = tn+1 . For large angle changes we require that {ii}.5) { ~ n + l= { u n ) [ R n ] { A u n + l ) ) + .2.}.2) { F ) = { P F ( u n + l ) (1 . is the time step size. {u). ( F = K'lL for linear problems. The additional iteration steps provide equilibrium solutions at each time step.) { U{ {Fn = Average load over the time period (tnl < t . the static forces are (8.) The above equation is solved at the reduced ( u d ) displacement vector size. u = Average elastoplastic element total force vector.2 THREEPOINT METHOD (NLTRD MODULE) The threepoint method is chosen to be compatible with MSCJNASTRAN linear transient integration method.I ) ) + where .1 Basic Equations Adding dynamic loads t o the basic equation given previously. < tn+l) = Corresponding acceleration and velocity vectors.2 P ) F ( u n ) f P F ( u n . The vector { u ) contains gimbal angles requiring a transformation [R(u. thereby guaranteeing stability and accuracy for arbitrary time step size. The approximation errors due t o dynamic reduction methods are not included in the error vector {R. {R.8. Applying Newmark's averaging method over a finite time period. and { A u ) are vectors in the glpbal coordinate direction. the acceleration and velocity vectors are: ) l ) where { A ~ n + l = { ~ n + . ) from the applied loads a t each time step.1].)] such that (8.2. 8.
1) defines the load error in terms of the current vectors estimates: and {Aun+l} = {nun + dn) At any time step the vectors {P.2 Nonlinear Iterations For a nonlinaer solution. With .dk} = 0 dn I.) equals [l<]{u.2.6) and (8.2.8) is where [I. the estimated displacement change is The algorithm is identical t o the static case except that instead of a tangent stiffness matrix. {u.).3) and (8.2.} = {Au. 8.+l) or by an iterative search.2.7) may be solved with Newton's method (or a modified version).+. {F.D = 113 this will produce the linear SOL 27 MSCINASTRAN results. Note that for a linear solution {F.).6) and (8.2. (8.2. {nu.2.}.2.[ Then.).} as the primary solution variable provides the following iteration algorithm.+~} directly by setting {R. and {Fnl) are known. Eqs. (8. obtained from Eqs.Substituting Eqs.2.> = (0) (8. where [ K ]is the reduced stiffness matrix.2.2. In general the vectors {d.6) and (8.2.}. Using {d. (8. (8. (8. the left hand side matrix. First assume a linear approximation: {RL+'} Let  {R. At a new time step the load iterations may be started by assuming that {Au. and {Fn+l} = {F(u.} and therefore: {do. (8.+~).4) into Eqs.']= is the current tangent stiffness matrix.+l)) must be found either by approximating {F. In this case Eqs.2).12) .7) may be used t o solve for {u.} + dRn {dk+' .) = (0).
} passes the convergence tests or the number of passes reaches an iteration limit.2. 8.2. assuming that the matrices are reduced t o equivalent scalar modal quantities. After lengthy calculations we may summarize the various criteria.3 Stability Analysis In transient analysis.13) into Eq. (8. (8. The second is the divergence which grows with time in the transient integration.12) and (8. {R). ( i > 0) Eq. {Au.6) for i = 0 results in the first estimate: where 1 Here [A2] is a precalculated matrix. and/or the time step size may be reduced.2.2. the matrices may be updated. The first is the familiar nonlinear load iteration divergence which also occurs in static analysis.10) and Eq.16) until {R. (8. For faster convergence the iterations may continue. are assumed t o grow at the rate A. (8. defined as: Note that if IAI > 1 the system will be defined as unstable. two types of instability could occur. results in MS C/NASTRAN.2. from von Neumann.18) where A I c the difference between the tangent stiffness matrix and its approximation 1CT.6) becomes: where 1 [A31 = I S M +z 1 B 1 (8. calculating only Ro.2.2.However.2.)}  [IiT A I ~ ~ ~ ] { u ) + (8.) are previous results. Both instabilities are caused by uncorrected nonlinear equilibrium errors.17) At each time step the code will iterate on Eq. It is assumed that the nonlinearity has a first order approximation: IF(. (8. With a single step.) and {F.2. i = i. the results will be identical t o the existing NOLIN. A convenient method. For subsequent iterations. we must extrapolate the nonlinear forces: Substituting Eqs. the criterian for stable solutions are: . t o be consistent with displacements. is ~ ~ ~ The error vectors. for analyzing the stability limits is to assume that the nonlinear forces are nearly linear and the error vector has a constant convergence.
MESTNL Nonlinear element summary table PDT Applied loads.RSTIME/S.NSTEP/V. as in the standard Newmark Beta method: 4 n K N L 5 (4p .MPTS. For time step integration with no intermediate static iterations.N.N.ELDATA.Y.Y.N.LGDISP/S.N.22) At2 This restriction is more severe than criteria (8. The better stability of the method and the capability t o use larger time steps outweights the cost of a few internal iterations on the static element forces.1)I<' M (8. proving that the internal iterations are more stable than the Newmark integration.N.2.TEL/ V.Y.2.DLTl.NOGONL//V.FDV.NLF'T.PNL. 8. BGPDTS.KTIME/S.GM. one column vector for each time step. KRDD Reduced stiffness in dset .KRDD.N.USETD. + In summary.21) above.4 NLTRD DMAP Interface of NLTRD Module (Version 66) CASESX2.AMS.MAXLP $ ttV66 Input Data Blocks: CASESX2 Case control data.N.AMlLL. For static load iterations: This states that the mass and damping add to the effective linear stiffness [ K T ]and improve the stability.YS.N.PDT.OESNLl.SILS.NEWK// S. KELMNL.LASTUPD/S.N. the method will have fewer divergence problems than either the static nonlinear solution or the single step transient nonlinear methods.For time steps with converged static iterations: This is the same criteria as linear analysis.BETA/S.2.CONV/S.NEWP/S.CSTMS.ITERIDS// S.N.MESTNL.DIT.AM2.MPTSIZE/S.FSGV.KSGG/ UHVNL.ESTNL.AMlUU. YS Enforced displacement.
CSTMS Coordinate system transformation table. DIT Direct input tables.) for matrix update.. FDV Input/output. AMlLL Lower triangular factor for nonlinear elements GM Transformation matrix for multipoint constraint reduction. SILS Scalar index list.]{Y. BGPDTS Basic GRID point definition table. KELMNL Nonlinear element stiffness matrix required for nonlinear force ( F ) computation. Output Data Blocks: + B. and acceleration vectors a t output time steps upon completion of the subcase. F ~at the end of subcases. displacement and velocity vectors at converged time steps during the direct integration. . ESTNL Input/output. OESNLl Output stress data for nonlinear elements. nonlinear Element Summary Table at the last converged step. KSGG Partition of stiffness matrix for forces of constraints.ELDATA Element data used to generate OESNL file. F. AM2 Damping matrix in dset divided by At.) at output time steps. Purged at the beginning of subcases. MPTS Material properties table containing TSTEPNL data. DLTl Dynamic loads table for follower forces. USETD Displacement set definition table. velocity. PNL Nonlinear load vector appended at output time steps. TEL Appended output time step list. Element forces and ] the current and the preceding steps (F. AM3 Matrix consisting of mass and damping terms: &Ad NLFT Nonlinear transient loads for NOLIN. UHVNL InputJoutput. equivalent t o [K.. Two column vectors [F. UHVNL consists of output displacement. A M l U U Upper triangular matrix for nonlinear elements.l. FSGV Nonlinear element forces at constrained points.
set t o +1 to continue. iteration count within a time step. MAXLP Inputinteger. set t o 1 if not converged. set t o $1 for new subcase. NEWP Input/outputinteger. initialized t o 0 in MPL and retained during the problem solution. set t o f l or +3 for stiffness update. on output from the module. The value 0 is used t o indicate if the stiffness update is performed due to the GAP element during the iteration.Parameters: BETA Inputreal. STIME Input/outputreal. NOGONL Outputinteger. current time step position for subcase. NEWK Input/outputinteger. LGDISP Inputinteger. 1 t o terminate. set t o +1 for geometric nonlinearity. CONV Input/outputinteger. f 1 if converged. NSTEP InputJoutputinteger. MPTSIZE Input/outputinteger. 1 otherwise. value from the last iteration for an input (0 for the initial iteration). denoting the increment of time between the K computation and the start of the module. ITERIDS Input /outputinteger. 1 otherwise. 1 for no subcase switch. integration parameter. set t o 0 at the beginning of the subcase. LASTUPD Input/outputinteger. maximum limit allowed for element relaxation iteration and the material subincrement processes. starting time or accumulated time used for restarts by user input. the time step number of the last stiffness update. default = 113. timetogo value at the beginning of K update in the DMAP loop. KTIME Input/outputinteger. a negative value is stored. . the amount of opencore words required to hold the material data.
1 Newmark Integration For the adaptive scheme.ui } n+l where By solving Eqs. In addition t o the extra memory requirement.8.+I + . and restarting.1) and (8. An incrementalliterative form of Eq.e. Newmark's method [8.3. ending time step. and {F) and {Pn+l) denote internal and external forces. (8. {u}. i. These difficulties become more complicated and cumbersome when the adaptive time stepping is employed due t o the unequal time intervals involved. (8. respectively.3) can be recast for the ith iteration (denoted by superscripts) as: satisfying ' + I i. { u ) and At denote displacement. Therefore.3] is employed with the twopoint recurrence formula for onestep integration. the twopoint integration method [8. these variables a t the ith iteration of the time step ( n + l ) can be expressed in terms of displacements as shown below: . acceleration and the time step increment.2) for the velocity and the acceleration vectors.3.2]. is introduced in Version 66 with module NLTRD2. 8.un> (8.+ n .. This method can be selected by specifying the ADAPT method in the TSTEPNL Bulk Data entry. the multistep integration scheme also introduces difficulties in the starting algorithm.3. velocity.3 TWOPOINT METHOD (NLTRD2 MODULE) The multistep implicit integration method [8.2) where {U). The subscript n designates the time step and the parameters (P and y ) are to be determined. respectively. or onestep integration.1] is not suitable for the adaptive method.3. An equilibrium equation t o be satisfied a t time step (ni1) is: where M and C denote mass and damping matrices. and = {un) + At { ~ n ) + 7 At {&+I .3.
3.2 Adaptation of Newmark's Method An alternative expression for the load vector can be derived for y=0. (8. (modified Newton's iteration) or 2 which is a stiffness matrix evaluated at any preceding time step.6) may be replaced by Kn Notice that the tangential stiffness matrix. {u:+~ .where { u }= {u:+. i.3.C { P At +. (8. .Un  &) + ($ .3.7) in Eq.4) and (8. (8.5 by introducing ~ { u n } c { u ~ )= {Pn +  Fn) (8. + ~is expressed as ) I {R:+~} { R : + ~ } = {Pn+l  ~:+1}  M pat. Notice that the acceleration term in Eq.3.e.3.} + {aui+l} The governing equation for the NewtonRaphson iteration is obtained by substituting Eqs.3.25.} + (  1) c { u } .3 parameter L has the recommended value of 0. b' 1 M where the residual load vector { R . 8.3a) to obtain: Y + CAt + 1iicl { h u i + l } = [pat.3.8) By virtue of Eq.3.1) M A {un} Y . while the computation is significantly simplified.3.7a) conveniently disappears when the . (8.8) the residual load error at each time step is effectively carried over t o the next step and the error propagation is reduced.(1  { n } (8.. (8.5) into Eq.
{On)  * At n {&I)] (8.If follower forces (i. t n + l ) = g(tn+l>{pa(u) P(O)J where g(t) is a time dependent forcing function specified in the TLOADi entries and pi(^) denotes static follower forces.3. + l ( ~ ) I = IPn+l(O)J + {fk(u7 tn+l)} + { ~ ' ( u tn+l)} .5b) for 0 < n < N .4)) (8. (8.9) where N' represents NOLINi forces at the ith iteration. (8. the Eqs.3. i.3.un) In this method of direct integration (based on the average acceleration scheme).O+l .e.7b) can be computed without computing velocities and accelerations at every iteration.3.6) are reduced to: { u n + ~ )= 4 {Un+l un} . (8.+.3.10) + 1cn{u. {UO) and {uN). applied forces change as a function of displacements) are involved.e. The initial and the final accelerations.{Un} At 4 Since the acceleration vector does not appear in Eq.5a).  and {Fn) {~:+1) = {K) (8. the external load {Pn+l) { p . It is noted that the residual vector in Eq. 8. (8.3. {uo). {u) needs t o be computed only for output purposes.3.5a) requires an initial value. are evaluated by an extrapolation using Eq. (8. + ~ ) a correction for is follower forces as expressed below: f k ( u .+.)Qtn Atn At.. ..3. and f k ( ~ . the modified expression for the acceleration is {On+l} + (At. t .7b).3.5) and (8.{On} .3. Eq.8) without the need for computing initial acceleration Uo. For irregular time step intervals. it has been determined that accelerations are most effectively evaluated by the central difference method using velocities at three consecutive time steps at the end of the integration process. (8. with N being the last time step.3 Startup Algorithm The first iteration is performed with initial conditions {u:+i} which satisfy {~:+l) = {un} It is noted that the initial equilibrium is satisfied by virtue of Eq. .3. After some numerical experiments. should be updated at every iteration by however. (8.3. However.
BDD. MUHVNL.NSTEP/ V.4 DMAP Interface of NLTRD2 Module (Version 67) NLTRD2 CASESX2.STIME/ S. USETD Displacement set definition table. OESNL1.MPTSIZE/ S.Y.MAXLP=5/ V.NEWDT=O/ S. CSTMS.. GM.CONV/ S. YS.NDAMP $ Input Data Blocks: CASESX2 Case control data. BDD Damping matrix in dset.ITIME/ S.KRATIO=(l.N.OLDDT/ S. KELMNL. GM Transformation matrix for multipoint constraint reduction. SILS Scalar index list. KDD.NOGONL/ S. DLT1.N. FDV. ELDATA.N. PNL.3. KDD Stiffness matrix for linear portion in dset.N.N.Y.N.LANGLES/ V.)/ S.NBIS/ V.N. KBDD Tangential stiffness of residual structure in dset.O. BGPDTS. P D T . one column vector for each time step. NLFT Nonlinear transient loads for NOLIN . BGPDTS Basic GRID point definition table. MPT Material properties table containing TSTEPNL data. NLFT/ UHVNL. USETD.N.N.KTIME/ S.NEWP/ S.LASTUPD/ S.ITERIDS/ V. ESTNL. KBDD. FSV. CSTMS Coordinate system transformation table.TSTATIC=1/ V.Y.8. DLTl Dynamic load table for follower forces. ELDATA Element data used to generate OSENL file.N. DIT. DIT Direct input tables. MDD. SILS. MESTNL/ V.N. YS Enforced displacement. MDD Mass matrix in dset. MPT.N. PDT Applied loads.N.Y. TEL.Y.N.LGDISP/ S. KELMNL Nonlinear element stiffness matrix required for nonlinear force (F) computation.
stiffness ratio t o be used for time step adjustment.. ESTNL Input/output. initialized t o (l. STIME Input/outputreal. MUHVNL Input/output. +1 if converged. MESTNL Input/output.O. Purged at the beginning of subcases. used t o detect if the time step is adjusted in the previous subcase. and acceleration vectors a t output time steps upon completion of the subcase. P. NSTEP Input/outputinteger. 1 otherwise. MPTSIZE Input/outputinteger. . initialized t o 0 in MPL and retained during the problem solution. NEWDT Input/outputinteger.)..Output Data Blocks: UHVNL Input/output. displacement and velocity vectors a t converged time steps during the direct integration. It is deleted at the beginning of a subcase (MUHVNL r UHVNL). NEWP Input/outputinteger. LGDISP Inputinteger. PNL Nonlinear load vector appended at output time steps. displacement and velocity vectors (2 columns) for K update. velocity. current time step position for subcase. nonlinear Element Summary Table for K update. OESNLl Output stress data for nonlinear elements. TEL Appended output time step list. the amount of opencore words required to hold the material data. equivalent to [liss]{Ys) output time at steps. set to 0 at the beginning of the subcase. UHVNL consists of output displacement. The value is set at the end of each subcase: 1 if ADJUST>O and 0 if ADJUST=O. time step increment used in the previous iteration or time step. 1 for no subcase switch. FDV Input/output. Parameters: KRATIO Input/outputcomplex. set t o 1 if not converged. t o be used after the matrix update or subcase switch. set t o +1 for geometric nonlinearity. CONV Input/outputinteger. OLDDT Input/outputreal. starting time or accumulated time used for restarts by user input. Two column vectors [F. FSV Nonlinear element forces at constrained points.] at the last converged step. set t o +1 for new subcase. nonlinear Element Summary Table at the last converged step. value from the last iteration for an input (0 for the initial iteration).
LASTUPD Input/outputinteger. maximum limit allowed for element relaxation iteration and the material subincrement processes. denoting the increment of time between the K computation and the start of the module. used t o select the rotation vector approach t o large rotations by a value of 2. NDAMP Inputreal. NBIS Input/outputinteger.ITERIDS Input/outputinteger. current count of bisection. set t o +1 to continue. MAXLP Inputinteger. TSTATIC Inputinteger. The value 0 is used t o indicate if the stiffness update is performed due to the GAP element during the iteration. which is used in where 17 represents NDAMP. ITIME Inputreal. The default (= 1) option is the gimbal angle approach. LANGLES Inputinteger. Introduces numerical damping with a value greater than zero. . +1 allows static analysis by ignoring inertia and damping forces. KTIME Input/outputinteger. NOGONL Outputinteger. the time step number of the last stiffness update. default=1. default= 0. initial time at the beginning of a subcase. a negative value is stored. on output from the module. timetogo value at the beginning of K update in the DMAP loop.025. 1 t o terminate. iteration count within a time step.
The bisection process is coordinated with the stiffness matrix update so that changes in nonlinearities are properly reflected while the bisection is in progress. the usual procedure will be resumed for the next time step. which is coped with by combining it with the bisection method in a coordinated manner.4 ADAPTIVE TIME STEPPING 8. the adjustment is designed t o be adaptive to the severity of the nonlinearity in the problem by monitoring the changes in the stiffness.4] have shone some light on this goal.2 Automatic Time Stepping Algorithm The proper time step size. 8. However. changes continuously in the transient dynamic environment. The dominant frequency is estimated at every converged time step using the displacement increment (as an assumed vibration shape) and internal forces by . The dynamic response characteristics of the structure may change due t o the nonlinearity (geometric or material) or the mode of excitation. the time step size should be adjusted.4] did not distinguish between these two cases.4]. In such cases an undesired time step adjustment can be caused by a drastic change in the stiffness. but their method has inherent flaws with regard to nonlinear problems. The authors [8. because such a method is yet t o be found. In the present implementation of the automatic time stepping for nonlinear applications. However. requiring a decomposition. The deficiency of this method stems from the time lag.8. the bisection method is chosen to be activated only at the time when divergence occurs. This size can be predicted from the dominant frequency. their idea is salient and can be applied t o nonlinear problems with some modification t o cope with the difficulties in adjusting the time step size which fluctuates with the deformation pattern.1 Introduction It is desirable t o have a fully automated method of time integration which renders effective and accurate solutions of nonlinear problems. which are intrinsically different and therefore require distinct treatment as discussed below. This premise is wishful thinking. In addition. When the mode of response changes. Divergence activates the bisection process which is valid within the time interval in which the process is activated. on. After the solution is achieved. [8. Bergan et al. when plasticity or GAP elements are involved. required for accuracy and efficiency. A filtering scheme has been devised t o suppress the effects of the spurious mass or stiffness in the automatic time step adjustments. which may be a temporary situation. difficulties in the automatic time stepping have been discovered. in the incremental deformation pattern based on the Rayleigh's method [8.4.4. Nevertheless. the stiffness matrix does not have to be reformulated unless it is so required due t o nonlinearity.
To prevent thrashing.0 = 4. At may have to be bounded t o represent the external loads properly.25 = 0.. Since the automatic time stepping scheme ignores user specified D T in time marching. At. Since At is allowed t o be quadrupled or quadrisected a t each time step. the impulsive loading case. However..+. and MAXR and MAXBIS are additional parameters to be specified in the TSTEPNL card.l ) . which is effectively detected by II where ' ' II max < UTOL (defaulted t o 0.75. the userspecified DT will be used only as an initial time step size and the automatic time stepping scheme will be activated. Upper and lower bounds of At will be defined as At. is the maximum value of the norms computed throughout the preceding steps. DT/MAXR) where D T is the user specified step size. .where {AU. .) approaches zero.0 = 2. a stepping function f ( r ) is introduced such that (8.0 for for for for for r < 0. where f f f f f = 0.. If the predicted time step size falls outside the bounds. i.1) 11 u I(. however.4) At. While the adjustment algorithm is based on the mode of response. a termination criterion is required t o wrap up the subcase.1 27 1 r=m w.3) At. and { F ..Fnl) represents a change in the internal forces which are readily available in memory.. and = MAXR * DT (8. = f ( T ) At. the time step will be quickly adapted if the original At is not pertinent. At n where it is assumed that the dominant period is marched through with m steps.4.5Rb 0.g. the time step will be set to the limit. e. The last step may be determined by terminating .U n . = MIN (DT/2**MAXBIS. At. The next time step can be predicted by multiplying the current time step by a scaling factor r . When the automatic time stepping option is selected.5Rb 5 r < Rb Rbir<2 2 5 r < 3/Rb ' 1 3/Rb 7 with Rb being a user specified parameter defaulted t o 0. continuous adjustment of the step size at every time step is not a viable option because it requires too frequent matrix decomposition.5 = 1. .) = { U . A provision should be made to skip the time step adjustment if {AU..e.4.+.
Fnl} and {M AUn} . no further step will be taken if where the right hand side represents the total duration for a subcase.. The dominant frequency should then be computed by where {AU. It has also been observed that m of 20.} is a modified vector with zeros inserted for the undesired DOF.1 ] n where wi is the apparent angular frequency of the ith DOF. The value of m can be a function of the stiffness ratio.4. which is accomplished by checking the "apparent frequency" of each degreeoffreedom by comparing components of {Fn .e. Undesired time step adjustment due t o the drastic change in stiffness of the plastic deformation is also properly suppressed by this method. Severity of the nonlinearity may be represented by the stiffness ratio defined as: where I?' represents the last updated stiffness matrix. It has been observed that the criterion in Eq. Notice that matrix decomposition is required whenever the time step is adjusted..4. (8. (8. The bounds are determined based on the reference frequency as follows: where wTef is the dominant frequency previously computed..e. i. This is implemented as a default option in the following manner: A* A* = = 1 a if if X2 2 1 X2 < 1 .1) is dictated by spurious mass or stiffness terms when artificially low or high masses or stiffnesses are used. The adjusted time step will remain effective across the subcases.the normal time stepping scheme when the currently processed step is close to the last step. This undesired effect should be eliminated. in Eq. i. The central idea is t o eliminate the effects of the DOF that has extremely small or large apparent frequency compared with the dominant frequency. w9 = ABS [{ F Y 2. Large mass or large spring approach t o the base motion is an example. This is because the plastic deformation casts implications similar t o the GAP in the automatic time stepping algorithm. It could be caused by massless point or GAP elements as well.2)) is adequate if the deformation remains linear but should be greater for nonlinear problems.
In addition. i. the bisection is activated when the maximum number of iterations allowed for an increment is reached without convergence. 1 2 Bisection continues until the solution converges.g. The bisection is activated by a material routine if an incremental stress (Ai?) exceeds the specified tolerance (e. In addition. at.e.3 Bisection Algorithm The automatic time stepping option is implemented to adapt the time step size primarily to the dynamic environment based on the response pattern. =  Gl.4. Bisection normally begins upon the second detection of divergence within a At. Then. however. Such difficulties are resolved by using the bisection method. The divergence criterion is defined as NDIV > MAXDIV (8.At.e. i. 20 % of the effective stress).7) where MAXDIV is a userspecified parameter (defaulted t o 2 ) and NDIV is an internal variable which is initialized to zero at the beginning of a new time step and incremented as follows: NDIV = NDIV and NDIV=NDIV+l with if +2 if Ef > 1 or E f < 10+12 orif 1012<~f<1 I E ~ I > ~ Ef = {auiIT R ~ ) { {AUijT {Ri1) and E. E. reaching MAXITER without convergence will be treated as a diverging case.4. . The material processing routine can detect a change in nonlinearity. This method. an iteration is tried without bisection. is not likely to alleviate difficulties due t o the change in nonlinearity or divergence....m = 20 m = 40 No Adjust for for for A* < 5 5 5 A* < 1000 1000 5 A* This option will be superseded if the user specifies a value for m (MSTEP). = . The stiffness will be updated at the last converged step when the first divergence is encountered. 8. The bisection could also be initiated when a divergence is detected or anticipated.
The maximum number of bisections is limited by a userspecified parameter MAXBIS (defaulted to 5 ) . the integration proceeds t o the next time step with the same increment ( A t k ) .Further bisection may be required after achieving converged solution at the intermediate time steps as shown in Fig. The average error is defined by averaging relative errors in displacements. some recovery process such as doubling time step may be activated. If the bisection is required more than MAXBIS times. If the ith iteration yields the least average error. The bisection process is activated on an asneeded basis. . This is the time when the automatic time step adjustment is resumed as illustrated in Fig. the automatic adjustment procedure will be deferred until the divergence is trapped and the solution process is stabilized. 8. When the automatic time stepping is combined with the bisection method and if the bisection is activated. any adjustment t o the time step would require additional matrix decomposition which offsets the advantages of adjusting time step size. The normal process should be resumed when the userspecified time step at which bisection is activated is fully processed. it has been determined t o continue stepping through the specified time interval to completion. If no further bisection is required and the solutions converge without any difficulty. the reiteration procedure will end a t the ith iteration and the normal time stepping procedure will be resumed. Therefore. Once the bisection is successful (rendering a converged solution). the solution process will continue without bisection by activating the reiteration procedure. However. The solution process is considered stabilized when two successive solutions have converged without requiring further bisection. loads and energy.where k is a bisection count.1.4.1.4. by which the same iteration process is repeated t o find the best attainable solution. The best attainable solution is a solution obtained from the iteration that yields the least average error. Users also have an option to suppress bisection by specifying MAXBIS=O. 8.
4. X denotes converged solutions during the time step ~t . 8. Figure 8. 0 denotes reference point for k update. denotes the step which triggers decomposition. Automatic time stepping resumes at # if selected. 5. Numbers above the baseline denote a sequence of trial solutions. 2. 3. Numbers below the baseline denote a sequence of bisections.1 Illustration of the Bisection Strategy . 7/~denotes critical period.Notes: 1. 6.4. 7. The baseline represents a duration of a time step before bisection starts.
{P.5 EXPEDIENT ITERATION STRATEGIES The dynamic tangent matrix may change for the following reasons: r Stiffness (K) changes due t o geometric. additional Kupdates are allowed within a given At at r Decomposition is performed within the module at every bisection or time step adjust ment.6] processes are fully operational during the iteration.[ M I . the QN vector pair. where KSTEP is a userspecified parameter in the TSTEPNL entry. Examples of various cases for the update relative to the bisection strategy are illustrated in Fig.4. { U n ) . r Bisection or timestep adjustment causes changes in the effective stiffness. r If the bisection is in progress. 8.l(ad). While the second case only requires decomposition. r Stiffness matrix update is allowed only once within the same time step.[C]. material and/or kinematic nonlinearity. which renders the condition number closest t o unity.). the QN vectors will be updated selectively based on the condition number of the QN update matrix. and the triangularized matrices ( L D L ~for [A] = ) M 81. r The stiffness matrix is updated at the beginning of each subcase.5] and the line search [8. {F.). is replaced or eliminated when a new candidate QN vector pair becomes available. r Whenever the decomposition is performed. [& + c + 2. the first case requires reformulation of the new stiffness matrix as well. Given at the beginning of the time step are At. As the condition number approaches unity.8. Once the number of QN vector pairs reaches MAXQN (defaulted t o l o ) . QuasiNewton (QN) vectors are continuously accumulated up t o MAXQN (userspecified parameter) pairs until the new stiffness matrix is evaluated. the QN vectors t o be accumulated are stored on a temporary basis until a converged solution is obtained. every KSTEPth converged bisection. As a requirement for the reiteration procedure. Therefore.4. See Figure 8. Adjust the time step size.l(d) for an example. . the iteration starts from the previously con verged position. The BFGS update [8. the QN update matrix approaches identity matrix. if applicable..). which causes little effect on the matrix t o be modified. The iterative process for time step ( n + l ) can be summarized as follows: 1. w T e j . {u. The adaptive method is based on the following matrix update strategies: Stiffness is updated at the previously converged position when the divergence is detected for the first time at a given time step. All the accumulated QN vectors are purged if the stiffness matrix update or the decomposition (due t o the change in At) is scheduled. {AF.).
13. {F. [A] = L D L ~if .+~) {P. The adaptive time stepping method is implemented as an optional feature in SOL 99 with a new module NLTRD2 for the nonlinear transient response analysis C8. 18. If it is the first divergence or the divergence after the KSTEPth converged bisection step. if the maximum number of bisections allowed for the time step is reached. If divergence occurs. Some efforts were required to bring together all the data blocks in compatible size and format. proceed to the next step. Another aspect t o consider in the DMAP is that the output time steps cannot be predetermined in the variable time stepping. 12. Compute {Ri.O+l). However. 5. 17. Then. 15.1 and 8. However. 10.2. 14. # At. 8. Perform the reiteration procedure t o find the best attainable solution. proceed t o the next step.. Compute the velocity u n + l . proceed to the next step. Otherwise. purge the QN vector file. Compute global error functions and check convergence. 16. These algorithms are combined into the module NLTRD2 as shown in . Proceed t o the next step if the line search error is less than the tolerance. 7. 6. The automatic time stepping and the bisection algorithms are flowcharted in Figs. go to step 16. Advance t o the next time step with n = n + 1. Accept the solution (Un+l and Fn+1) and append temporary QN vectors to the permanent QN vector file. go t o step 12. respectively. Go to step 5 to continue iteration. if the maximum number of iterations allowed for an increment is reached.71.3. Among numerous enhancements from the DMAP standpoint. Initialize {U. Otherwise. Otherwise. Decompose. update the stiffness matrix. Bisect At and go back t o step 3.5. using the BE'GS update if applicable.+l). the efficiency of the adaptive time stepping method stems from skipping about 50 statements by virtue of the internal decomposition within NLTRD2 module. 8. perform the line search process and go back to step 7. If the solution has converged. and 4.5. Solve for {nuif1)by forward and backward substitution. Compute the line search error. 9.O+l). and go back t o step 4. save the QN vector on a temporary basis (if applicable) and go t o the next iteration in step 5. 11.
some combination of user parameters could cause thrashing such as excessive bisection. If necessary. Thrashing of the decomposition is effectively prevented by adopting a stepping function in the adjustment criterion.5. Efforts have been made to minimize the thrashing in the algorithm and yet to provide a reliable solution efficiently t o nonlinear problems.Fig.3. the user should make a judicious selection of the parameter values t o prevent any thrashing. 8. However. or repeated cycle of timestep adjustment and bisection. . Default values are tuned to near optimal condition to the best of our experience. Control parameters for these algorithms t o be specified by the user are gathered in the TSTEPNL Bulk Data entry.
=m : I IAFnXI (AU. M(Aun} .=0: .calculate K(Au. t = MAXRDT NBISO=NBIS.) WORK [49. NBISK=NBISK+l Atn.! =MIN ( A..9] [49.}I {MAU.).l 1 [49. L 0.1 Flow Diagram for Automatic Time Step Adjustment ..5.la] A. MAX ( d i n bf+t 1) Figure 8.10] A DENOM+ { A < ) ~ {KAU.) DENOMOS (AFn )=(FnJ{Fn.) I I WORK={AU~ J{AF..
5. NBIS=NBIS+l NBISK=NBISK+l if NBISK>O EXlT A: Reiterate for best solution EXlT B: Continue bisection EXlT C: Update stiffness EXlT D: Terminate job Figure 8..2 Flow Diagram for the Bisection Algorithm .
.$ 3 .m[dcur. .' . (x x) 6nol E 9 n S H L 4 m dWL40 NDIVNDIVtlll I E ~ I I U 10"dh< 9 > . ' NL21NT Waale memmory S KTIUE .~: . .ylm .  NBlS4 NBlS I WTERNO NUIML .ET~)I S. ROOTno or0 ._________. E (1. ) P(O)]+N NUUOG 'I + I R I ~ YES .) 41 EMIN.lz.. RITER YES fs.A. ..~(uV .W E' kr €FIRST LSCT 0 e..(d. Fy Fy1. LSCT 0 1.. W ONV r.1 ._$i_ _ ... FSV PNL Y * 4) canp*I Fa 4 M) . U .en.~p:"(~)*lp.l~ool~~andlE'I>LST~ .. . .3" f ."(o)+~. nd p d lor STATIC1 .d.)( U : ~ .I@ NDIV 0  YES .(t)[du". am.}'{fW: .~s~ol~IE.A. Pol .~ c) { _ _ _ _+(p: _ [ M $ ] CdcUa*N(u.1 It ~ t A p p d OESNLI..  MolV.e .&lo" ~4)I ..LDC' . + CONV paramelm Q FIC\TTT n i 3 ~ ~ ~the for m NTJTRD2 Module .A'* " h . ( hNSTEP  NLPlNC a 1(+A1 4 = A ." ( 0 ) c'"~p. 4 *Ilapolan P.renuimp~~(aaum*llanutm EXIT 1 K up6.[Q~. ..n EXIT z m ( w a r e oa EXIT 3 ~bmmu( . DELW pot.m.$c]@+&~ a* w7b1 Fam DGMAG. R.l<E~amdil) ~(IE~I<(O'.? NBISK NBISK t l . NUML v F$[&Y.*. sa . _ *_&_ +_ c_I c_ _ _i_I. .
6. Also shown in Fig. there is a point in time where the stress in the rod reaches the yield point. A quarter section of the plate is modeled by five QUAD4 elements with symmetric boundary conditions using MSCJNASTRAN.66Fy. The input data for this problem is given in Table 8. All four cases are included in a single run by four disjoint models to test the effectiveness of the automatic time step adjustment. the program reduced the time step t o 0. The material is elasticperfectly plastic and the large displacement effect is also included in the response analysis for the duration of one msec.l(b) is the variation of the time step increment.8. 8.8] as shown by A. In the second analysis when the initial time step was 0.. 0.. In the first analysis. 8.l(b).001 second and the other with 0.6.004 second. respectively. If P > F. in which the rod is made of an elasticperfectly plastic material.6. The impulsive load was simulated by initial velocities applied a t the nodal points as shown in Fig. B. The displacement responses of the two analyses are almost identical and agree with the closedform solution [8. As a result.2(a).l(a) shows the model subjected t o a step loading with four different magnitudes: namely. the system is governed by an elastic equation for rebound. and rebounding (unloading) regime. the mass responds with different patterns.6 VERIFICATION PROBLEMS Five example problems are presented in this section t o validate the effectiveness of the algorithm and to verify the program's performance.88Fy and F. 8. C and D in Fig.5Fy. During the plastic deformation. no rebound occurs and the mass will break away from the rod.6. 0. 8.01 second. For .2 Impulsively Loaded Clamped Plate An elasticplastic plate clamped at two ends is subjected t o an impulsive load over the center portion as shown in Fig.coswt) which is valid throughout the time history if P 5 $F. = P using proper initial conditions. plastic. 8. the ~olution may be obtained from mu F. 8. the program detected that the time step is too small and automatically quadrupled it t o 0. When the displacement reaches the maximum value. 0. Fig. The input data for this problem is given in .6.1 ElasticPlastic Rod with an End Mass Subjected to a Step Loading Consider a singledegree system (massrod model). 8. Bisection did not occur in this analysis..6. in which Fy is the axial force corresponding t o the initial yield of the rod. The theoretical solution can be considered in three different regimes: the elastic regime. < P 5 F.6. No unnecessary time step adjustments were performed during the elasticplastic vibration.2(a).F. + Two analyses have been made: one with an initial time step of 0.6. In the elastic regime the solution can be obtained simply by u = f ( 1 .01 second..005 second.1.
It is demonstrated that the time step size is adjusted to an equitable value (fluctuating between 0. The model is modified for the performance comparison t o include 200 beam elements. Each case was analyzed twice: once with and once without using the automatic time step adjustment.6. from either the high or low value of a specified initial increment size. Except for the case of 0. 0.02 inch. 8.6. Bisection was not activated in this analysis. The displacement responses at the loading point for the four cases are compared in Fig.001 sec) regardless of the initial value.0002 sec and 0.6. When the automatic time step adjustment is not activated.002 sec) have been used t o investigate the .9] and the solution from the explicit integration obtained by Belytschko et al. however.1 sec.6. Five different time step sizes (0. A question arose if a coarse mesh resulted in a more accurate solution by beneficially filtering the high frequency response. 0. 0. 8.2(b). the analysis was repeated with a 10QUAD4 model for comparison with the 10element model of Belytschko et al. This figure demonstrates that the MSC/NASTRAN solution agrees more closely with the test results.3.10].6. the solution process either wastes computing time by using time increments which are too small or renders inaccurate solutions by using time step sizes which are too coarse.0025 msec.6. This step size was maintained until approximately 0. In all of the cases the time step is adjusted to approximately 0.001 sec.2(b) also shows the variation of time increments due to the automatic time step adjustment. Fig. 8. 0. the same model was analyzed with two different initial time step sizes: 0.002 sec. The beam is modeled using 20 linear BEAM elements. Predicted time histories of the midspan displacement of the four cases are nearly identical as shown in Fig. Also shown in Fig.0002 sec.0125 msec.0025 msec) were used in the analysis. and 0.10]. Four different initial time step sizes (namely.6.Table 8.6.4 msec. [8.2.0005 sec. bisection processes were activated at the beginning to reduce At below 0. [8. This gap is simulated by a displacement dependent force which is active only when the beam is in contact with the snubber.2(b). The input data for this problem is given in Table 8. 0. In order to study this point.02 msec and 0. The result is even more favorable as shown in Fig. To demonstrate the effectiveness of the automatic time step adjustment.3(a). The duration of the analysis time span is also extended t o 0.3(b). A stopper (snubber) is present underneath the center of the beam with a clearance of 0.05 msec.3 Simply Supported Beam with a Restrained Motion A simply supported elastic beam is subjected t o a single cycle of a sinusoidal forcing function at the quarter span as illustrated in Fig. 8.6. in which the time steps are indicated by the symbol x in the figure.1 msec.0001 sec. 8.2(b) are the experimental data obtained by Balmer and Witmer [8.02 msec. 8. The problem is linear except for the effect of the stopper. that the time step variation is quite different from the 5element model. when At was doubled and remained unchanged to completion in all four cases. It is noted. All the solutions converged without activating the bisection process.0004 sec and 0. 8. 0.
4(b). This velocity becomes the initial condition for the equation h = vot .6. and after moving 2 inches. because the proper time step size is not only difficult t o predict in advance but it also changes throughout the analysis history. .6. It is noted that the CPU time is increasing almost proportionally t o the decrease in the time step size if the time step size is fixed. All computer runs were made on the VAX 8700 system at its offpeak time. A closedform solution is obtained for the massless platform. it is likely to be more efficient in terms of computing time than the fixed time stepping.4(a). While the weight is riding on the platform. The results are tabulated below: CPU Time Comparison Userspecified At(sec) Analysis with fixed time step Analysis with adaptive time step algorithm 0. The input data for this problem is given in Table 8. bisection was activated when the block was departing from the platform and changing the status of the GAP. 8.01 sec a t the initial adjustment phase as shown in Fig.001 86 145 0. In this analysis. the motion is governed by mx bx = mg with initial conditions xo = 4 and k o = 0. However.002 56 135 0. 8.0001 66 1 158 Note : Accuracy of the CPU time is estimated t o have a 5% deviation.t g t 2 .6.4(b).6. 8.4(a). The gravity field is applied and a small mass is assigned to the platform to suppress its jittering.0025 sec. Furthermore. 8.0002 327 150 0. The adaptive time stepping method provides the definite advantage which is the avoidance of the trialanderror process. The initial time step size was specified as 0.6.efficiency and effectiveness of the adaptive method. 8. The velocity of the weight block at the time of its departure from the platform can be obtained by the conservation of energy.0005 140 144 0.4. in the case of the adaptive time stepping method the CPU time remains approximately constant regardless of the initial time step size. which governs the motion after the weight departs from the platform. The time steps are indicated by the symbol x in the figure. is halted by a stopper.6.4 Bouncing Weight on an Elastic Platform A 4lb weight is resting on a springsupported platform which is initially compressed 4 inches in the position as shown in Fig. The platform is then released. The finite element model consists of one ROD element representing the spring. a mass for the weight block. The stopper is represented by a fixed nodal point and the initial deformation in the spring by an equivalent step force. and two GAP elements simulating contacts between the massplatform and the platformstopper interfaces as shown in Fig. + The response predicted by MSCINASTRAN is compared with the theoretical solution as shown in Fig. which was automatically adjusted to 0.
but the material properties are assumed to remain elastic througthout the analysis. however. 20 psec. the solution diverged at the very first time step and the algorithm activated the bisection process. The input data for this problem is given in Table 8.6. After obtaining converged solutions at two consecutive time steps. Hence. 8. the program detected that the time step size was still too large and adjusted it t o a smaller size.11] in Fig.5(b) also shows the variation of time step size due to the adaptive time step adjustment. .6. 8.005 sec. 8.5 Shallow Spherical Cap under a Concentrated Apex Load A thin shell in the shape of a shallow dome is clamped at the periphery and subjected t o a step loading of a concentrated force at the apex as shown in Fig. Fig. The analysis continued without any adjustment until the displacement approached the peak.6. 8. As expected.6. The displacement response a t the apex is compared with the results by Mondkar and Powell [8.5(a). When the same model was analyzed again with a smaller At(i. This time step size is 10 times larger than the size adopted in the previous study by Mondkar and Powell using an axisymmetric model with ten 8noded quadrilateral elements [8. which was later adjusted back t o 2 psec before completion. The geometric nonlinear effects are included. No further bisection was required t o complete the analysis.5(b) except for the minor details of spikes. 2.resulting in 0. the analysis was deliberately started with a relatively larger time step size. This analysis proves the effectiveness of the current adaptive method for the problems involving the gaps and/or drastic changes of the structural stiffness. Considering the symmetry. 8.5.5 psec. The step size was again adjusted to 4 psec past the peak response.6.5 psec during the analysis.6.. the bisection did not occur. 2 psec).5(b).11]. resulting in At = 10 psec.5(a). only one quarter of the cap is modeled using 27 QUAD4 elements with symmetric boundary conditions as illustrated in Fig. In this analysis. The resulting response agrees with that of the previous case shown in Fig. the adjusted time step size was reduced t o as low as 0. This problem is chosen to demonstrate the performance of the automatic time step adjustment and the bisection method combined.6.e. 8. where At was adjusted to 2 psec (which was the lower bound allowed for adjustment).
ElasticPlastic Rod Subjected to a Step Loading .Figure 8.6.1 (a).
........ .. 4... DT = 0. (in) 0...".... . 1...............01 sec At (sec) 0............005 ....2 DT = 0......... ElasticPlastic Rod: Displacement Response and Time Step Variation .1 (b).................01 0....Displ............001 sec Figure 8......1 Time (sec) 0.......6..
6. v = 3904 inlsec.61 x lo' lbsec2/i$ 10. yield stress 5000 in/sec.2(a). = v = p 2.Finite Element Mesh = E = a. Impulsively Loaded Clamped Plate .4 x lo6 psi 41400 psi. Figure 8.
Clamped Plate: Central Displacement Response and Time Step Variation 8.2 (b).Figure 8.6 .6.8 .
3(a).20 Beam Elements 0.01 1 (sec) (~10010) Forcing Function Representing GAP Figure 8. Simply Supported Beam with a Restrained Motion .6.
0002 Sec.) with adjustment .6. Simply Supported Beam: Loading Point Displacement Time History . Time(sec.) Time(sec.) Time(sec. Figure 8.) with adjustment without adjustment Userspecified At = 0.002 Sec. without adjustment Userspecified At = 0.3(b).Time(sec .
4(a).i I i I stopper 1. (initially compressed 4 ins.) P1 atform Structural Configuration w closed gap 40 (lbs) k = 10 lb/in MSCJNASTRAN Model Figure 8.11 . Sprt ng Mass (4 lbs.6 . Bouncing Weight on an Elastic Platform 8.) k = 10 lb/in.6.
Bouncing Weight: Displacement Time History + 16. ( : 8 8tt I b " 8 i t '* " 8 t t t < f + 0.4(b). 12.a t3 6. + 0. a 2. z . t /da // 0 Theoretical + MSCJNASTRAN \ . " Weight " *.5 Figure 8. 2 W a +.. 8.6.1 0.2 0. 0. W Z . 2. 4.6 TIME (SEC) Bisection . g 10. P U ' o m \ 7.A A 4 14.4 0.5 0.3 0.
75 in. lbs. 0.01576 in.90 in. constant applied force Figure 8.5(a).R = a = H = t = E = v = 4.6. Young's modulus 0.08589 in 0. Poisson's ratio p = 2.3. Shallow Spherical Cap under a Concentrated Apex Load . 10000 ksi. 0.45 x lo' lbsec2/in4 P = 100..
6.5(b).+ 0 MSCINASTRAN Mondkar and Powell Figure 8. Spherical Cap: Apex Displacement Response .
21(2) . 21 2 22 21 CROD CMASS2 21 10000.31(~2).31. 21 2 GRID 31 GRID 32 240.1 Input Data Listing for ElasticPlastic Rod ID TBIGGS4.5/.41 DISP=I ACCE=I VELO=1 STRESS=I SEALL=ALL SUBCASE I DLOAD=I00 TSTEPNL=IOO OUTPUT(XYPLOT) CSCALE = 1.5 XTITLE = TIME IN SECS XGRID LINES = YES YGRID LINES = YES YTITLE = DISPLACEMENT GRID 1 XYPLOT DISP REsP/II (T2) .41(2) YTITLE = STRESS IN ROD XYPLOT STRESS RESP /11(2) .31(2) . M.88/1.21(2) .Table 8.7. 6971 ECHO=BOTH SET I = 11. 11 2 21 GRID GRID 22 240.V66 $ SSH 10/22/87 $ PRINT MATRIX TRAILERS AND ITERATIONS DIAG 8. 11 2 12 II CROD CMASS2 I1 10000. BIGGS.25 SEC.50 $ VAX TETRA TIME 5 $ SOL 99 CEND TITLEzELASTOPLASTIC VIBRATION PROBLEM (LOAD = . 1 2 ~41(2) 3() BEGIN BULK $ GEOMETRY AND CONNECTIVITY 11 GRID GRID 12 240.21(~2) . P.6.4i(~2) YTITLE = STRESS IN RODS XYPLOT STRESS RESP/11(2) . SECTION 2.4i(~2) XYPLOT VELO RESP/II(T2). 31 2 32 31 CROD .41(~2) YTITLE = VELOCITY GRID 1 ) ~ . 66/. 21( ~ 231(~2) YTITLE = ACCELERATION GRID 1 XYPLOT ACCE RESP/II(T2) $21 ~ 231(~2) ( ) ~ . LABEL=J.21. FYI SUBTITLE=NONLINEAR TRANSIENT ANALYSIS FOR 0.
0 0 0 0 1. 41 2 2 10000. 30000.0025907 50 . 31 2 240. . 50 50 50 50 ED NT 10. +06 42 41 .0050 I ADAPT . 0 I.CMASS2 31 GRID 41 GRID 42 CROD 41 CMASS2 41 $ PROPERTIES PO RD 2 MT A1 2 MATS1 2 $ LOADING DO D LA 100 +DL1 I. 2 30. 104 11 21 31 41 201 202 203 204 1. $ PARAMETERS PRM AA WM S T AS TSTEPNL 100 tTS2I +TS22 E D AA NDT 10000. 45300. 40000. I. 2 2 2 2 101 22650.6672 PLASTIC 0 . DAREA 201 D RA AE 202 D RA AE 203 D RA AE 204 TLOADl 101 TLOADI 102 TLOADI 103 TLOADI 104 TABLED1 50 +ABC 0.
3 XGRID LINES = YES YGRID LINES = YES XTIT = XMAX = 1.2 Input Data Listing for Impulsively Loaded Plate ID GMNL08.50 $ PRINT MATRIX TRAILERS AND ITERATION CEND TITLE = NONLINEAR IMPULSIVELY LOADED CLAMPED PLATE (~6=1. GRID I1 1. TIME 5 $ NLTRANSIENT SCR=YES SOL 99 DIAG 8.0 t.6 0.06 I 0.6. GRID 22 2.03 $ TIME YTITLE = D I S P L 1 XYPLOT DISP / l T ) (3 YTITLE = V E L 0 C I XYPLOT VELO / 1(T3) YTITLE = A C C E L I XYPLOT ACCE / l T ) (3 YTITLE = D I S P L 3 I XYPLOT DISP / 31(T3) YTITLE = V E L 0 C 3 1 1T) XYPLOT VELO / 3 ( 3 YTITLE = A C C E L 3 I XYPLOT ACCE / 31(T3) YTITLE = S T R E S S I XYPLOT ELST / 1 3 () BEGIN BULK TSTEPNL 400 200 5.0 t.Table 8. GRID 12 GRID 21 2. 6 0. I.0 +. GRID GRID 2 0.6 0.+4) 99Iq4 SEALL = ALL SPC = 200 = 300 IC TSTEP = 400 sTRE(PL0T) = ALL $ SET I = I DISP = I VELO = I ACCE = I OUTPUT (XYOUT) CSCALE 1. V66 $ SHL 5/21/87 $IMPULSIVE LOAD $ VAX8700(TETRA) CPU MIN. GRID 31 3.0 ADAPT . I 0.
0 MATS1 100 PARAM COUPMASSI PARAM K6ROT 1.05 .0+4 PARAM NLAYERS 5 PARAM LGDISP 1 PARAM W 4 1.GRID 32 4 1 GRID GRID 42 GRID 51 GRID 52 CQUAD4 I 100 CQUAD4 2 100 CQUAD4 3 100 CQUAD4 4 100 CQUAD4 5 100 PSHELL 100 100 +PS .4+6 PLASTIC 0.3 MAT1 100 10.2E4 SPCI 200 246 SPCI 200 123456 SPCI 200 156 TIC 300 I TIC 300 2 TIC 300 11 TIC 300 12 ENDDATA .05 +.
......50 $ PRINT MATRIX TRAILERS AND ITERATIONS CEND TITLE= DYNAMICS HANDBOOK DEMONSTRATION PROBLEM SUBTITLE= DIRECT TRANSIENT RESPONSE..4.0 0.....2..0 0.0 0.. V66 $ SHL 5/19/87 $ MSC.....Table 8.0 0.10010 DISPL=3 $ ACCEL=2 VELO=3 OLOAD=I NLLOAD=2 SUBCASE I $ DLOAD=30 TSTEPNL=20 NONLINEAR=I3 $ SELECT NONLINEAR FORCE OUTPUT(XYPLOT) CSCALE=I....... $ MODELING INFORMATION FOR BEAM 101 100 I0000 CBAR 100 10001 CBAR 102 CBAR 103 100 10002 100 10003 CBAR 104 100 10004 CBAR 105 ONLY 10001 10002 10003 10004 10005 0. NONLINEAR FORCE LABEL= NOLIN IN SOL 99 SEALL=ALL ECHO=UNSORT SPC=IOO2 SET 1=10005 SET 2=10010 SET 3=10005..... D9918 $ KDB 12JUN84 TIME 5 $ VAX8700(TETRA) CPU MIN... $ NONLINEAR TRANSIENT ANALYSIS SOL 99 DIAG 8.0 0..8 PLOTTER NAST CURVELINESYMBOL = I XGRID=YES YGRID=YES XTITLE=TIME YTITLE=DISP 10005 T3 XYPLOT DISP /10005(T3) YTITLE=DISP 10010 T3 XYPLOT DISP /10010(T3) YTITLE=APPLIED LOAD 10005 T3 XYPLOT OLOAD RESPONSE/l0005(T3) $ PLOT THE INPUT YTITLE=NONLINEAR LOAD 10010 T3 XYPLOT NONLINEAR RESPONSE/IOO~O(T~) BEGIN BULK $.6..0 .3.5.7....3 Input Data Listing for Beam with a Restrained Motion ID TNOLIN.0 0..0 0..0 0....0 0...6.
0 0. 20. 10006 10007 10008 10009 I0010 10011 10012 10013 10014 10015 10016 10017 10018 10019 10020 .0 0. 0.0 0. 95.0 0.0 0.0 0. 15.0 100 ++0000011. 10.0 0.0 0.0 0.15708 0.0 0.0 0.0 0.0 0.0 0.0 0.0 5.0 0.0 0.0 0.0 0.0 0. 85.0 0. 10 123456 10020 3 10000 3 $ $ MODELING INFORMATION FOR CENTER S P R I N G CROD 10 10 10 I0010 MAT1 10 10. 90.0 0. 80. 50.0 0.0 0.0 0.0 0.0 0.3 0.0 0. 25.0 0.0 0. 0.0 0. 55.0 .0 0.0 0.0 0.0 1.002588 1000 0.0 0.0 .0 0.I 0.0 0.0 0.0 0. SPC 1002 SPC 1002 3.0 0. 30.0 0.0 0.0 0.0 0. 100.0 0.31416 0.0 0.0 0.3 0.0 0. 60. 40.0 0.0 0.0 0.0 0.0 0. 65.0 0.+7 10010 . 35.0 0.0 0. 0.0 0. 75.CBAR CBAR CBAR CBAR CBAR CBAR CBAR CBAR CBAR CBAR CBAR CBAR CBAR CBAR CBAR CONM2 GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID MAT1 PARAM PARAM PBAR 106 107 108 109 110 Ill 112 113 114 115 116 117 118 119 120 12 10 10000 10001 10002 10003 10004 10005 10006 10007 10008 10009 10010 lOO11 10012 10013 10014 10015 10016 10017 I0018 10019 10020 1000 GRDPNT WTMASS 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 10010 10005 10006 10007 10008 10009 10010 10011 10012 10013 10014 10015 10016 10017 10018 10019 50.0 0.0 0.0 0.0 0. 70.0 0.0 0.0 0. .0 0.0 0.0 0.0 0. 45.
+TAB ENDDATA C J 3 Y3 T 13 +ABC +TAB 0.I0.5E2 4.3.OE2 0.2 .0E8 $ LOADING AND SOLUTION INFORMATION TLOAD2 30 33 DAREA 33 10005 3 47.0.95 2.0 10010 NOLINI 13 $TABLED1 ID $+ABC XI Yl X2 Y2 X3 TABLED1 13 2.0 ADAPT MODELING INFORMATION FOR NONLINEAR SPRING $NOLINI SID GI CI S G J 10010 3 1..PLASTIC. ENDT . 0.PROD 10 10 1.0DIDI.0002 I TSTEPNL 20 199 +TSI +TS2 $ $ 0. MATSI.
01 MAT1 I 100.202 SET 2=200 $.3 . $PARAM U3 1000.23 SET 4=20 $ CONROD DISP=I $ OLOAD=I $ SPCF=2 ELFORCE=4 STRESS=3 $ FOR GAP FORCES SEALL=ALL LOADSET=20 SUBCASE I DLOAD=210 TSTEPNL=21 SUBCASE 2 DLOAD=210 TSTEPNL=2I OUTPUT (XYPLOT) CSCALE=I.0025 I +TSPI +TSP2 0 $ PROPERTIES PARAM W4 31.203 SET 3=22 .3 PLOTTER=NAST XTITLE=TIME IN SEC.4 Input Data Listing for Bouncing Weight Model ID VGAPM. XYPLOT DISP RESP/~OI I .3 4 2. IN INCH $ YMIN=O. $ YMAX=I5. SSH 1/13/88 $ FOR VAX TIME 5 DIAG 8. 202(TI) ( ) T BEGIN BULK ADAPT TSTEPNL 21 200 0. YTITLE=DISPL.6.Table 8.50 $ PRINT MATRIX TRAILERS AND ITERATIONS SOL 99 CEND TITLE=GAP VERIFICATION SUBTITLE=NONLINEAR TRANSIENT ANALYSIS ECHO=UNSORT SET 1=201. PGAP 3 +PG3 5. $PARAM G 0. PGAP +PG4 5.V66 $ SHL 12/11/82.
0. CONROD 20 200 201 CN 2 OM 24 201 CN 2 OM 21 202 CA GP 22 3 201 CA GP 23 4 20 1 E D AA NDT 1. GRID 202 4. 0. GRID 201 4. LSEQ 20 30 1 202 $TLOAD2 210 301 TLOADI 210 30 1 TABLED1 220 +TAB1 0.001 $ BOUNDARY CONDITIONS GRDSET $ G O E R AND CONNECTIVITY E MTY GRID 200 6.$ LOADING FORCE 201 20 1 GRAV 200 386.1 GRID 203 6. LA OD 202 I. ED NT 23456 123456 . I.
SAAB4 $ NONLINEAR TRANSIENT RESPONSE SOL 99 $ TRACE ITERATION PROCESS DIAG 50 TIME 60 $ CPU TIME IN MINUTES CEND TITLE = TEST PROBLEM 1.6.. S PLOTTER NAST XPAPER = 29. 23. GAD MSC. 0. 31. 18.V66 $KIOOK KIM $ MAY 1987. 25. X 21. MX. FIND SCALE ORIGIN 1 SET I PLOT SET I LABEL BOTH SHRINK FIND SCALE ORIGIN 1 SET I PLOT STATIC 0 SET I AXES Y. 32 AXES X.5 Input Data Listing for Shallow Spherical Cap ID SCAP.5 PLOTTER NAST PAPER SIZE 29. 24. IJNME 11:499520 (1977) SEALL = ALL SET 130 = 1 $ THRU 30 DISP = 130 SET 14 = 1 $ THRU 4 OLOAD = 14 SPC = 1234 SUBCASE I DLOAD = 100 TSTEPNL = 250 OUTPUT(PLOT) CSCALE 1. CURVELINESYMBOL = 2 XGRID = YES YGRID = YES XTITLE = TIME YTITLE = VERTICAL DISPLACEMENT AT GRID I XYPLOT DISP /I(T3) BEGIN BULK 5. Z VIEW O.Table 8. SET I = ALL SET 2 = I THRU 6. YPAPER = 21.4 SUBTITLE= SHALLOW SPHERICAL CAP WITH CONCENTRATED APEX LOAD LABEL = REF: MONDKAR AND POWELL. 10. 11. FIND SCALE ORIGIN I SET 2 PLOT STATIC 0 SET 2 OUTPUT(XYPLOT) CSCALE I. 0. 17. O.17.. Y.27. Z VIEW 34.6 I ADAPT TSTEPNL 250 80 .
10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 I0 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 1 I I I I I .' +TSI +TS2 PRM AA CORD2S +I GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID $ CqUAD4 CqUAD4 CqUAD4 CqUAD4 CUD QA4 CqUAD4 LGDISP 10 1. I 2 3 4 5 6 7 8 9 10 I1 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 I 2 3 4 5 6 0 +I 0.
CQUAD4 7 CQUAD4 8 CQUAD4 9 CQUAD4 10 CQUAD4 I1 CqUAD4 12 CqUAD4 13 CQUAD4 14 CQUAD4 15 CQUAD4 16 CQUAD4 17 CQUAD4 18 CQUAD4 19 CQUAD4 20 CQUAD4 21 CQUAD4 22 CQUAD4 23 CqUAD4 24 CQUAD4 25 CQUAD4 26 CQUAD4 27 $ PSHELL I MAT1 I $ SPCI I SPCI 2 SPCI 2 SPCl 3 SPCI 4 SPCI 4 SPCI 4 SPCI 4 SPCADD 1234 $ DAREA 10 TLOADI 100 TABLED1 2 +TBDI 0.0 EIDDATA 5 9 THRU 6 THRU THRU THRU 3 ENDT .
The nonlinear properties and/or effects are defined by nonlinear material data (MATS1 and TABLESI). 9. etc. including all previous subcases. elements and material data). NDAMP) are provided for additional control or capabilities. DAREA.1 USER INTERFACE The input data required for SOL 99 or SOL 129 is a combination of direct time integration control data. similar to SOL 69 (for direct linear transient with superelements). and nonlinear modeling data similar t o SOL 66 (for nonlinear statics). The output time is labeled by the cumulative time. LSEQ.Chapter 9 NONLINEAR TRANSIENT RESPONSE ANALYSIS 9. Restarts are controlled by parameters (LOOPID. SLOOPID and SDATA) which can be specified either in the Case Control section or Bulk Data section. and PARAMeter LGDISP for geometric nonlinearity. and mass properties. Massless degree of freedom should be avoided for numerical stability and the small damping is also recommended.1 Case Control Each subcase defines a time interval starting from the last time step of the previous subcase and the interval is subdivided into small time steps. The unique data required for SOL 99 or 129 is supplied on the TSTEPNL Bulk Data entry. The TSTEPNL entry in itself is a combination of the TSTEP entry for direct time integration and the NLPARM entry for nonlinear iteration control.1. gap elements (GAP) for nonlinear interface.). STIME. Some optional parameters (TSTATIC. damping (parameters. There are advantages t o divide the total duration of analysis into many subcases so that each subcase does not have excessive number of time . The transient effects are produced by timedependent loading functions (TLOADi.
Initial conditions (displacement or velocity) can be specified by the Bulk Data TIC selectable by the Case Control command IC. The input loading functions may be changed for each subcase or continued by repeating the same DLOAD request. The data blocks containing solutions are generated at the end of each subcase t o store in the database for output process and restarts. As such. The residual superelement may contain scalar degree of freedom representing linear modal formulations. e. Nonlinear forces as function of displacements or velocities (NOLINi) may be selected and printed by Case Control commands NONLINEAR and NLLOAD. If initial conditions were given. Each subcase may have a different time step size. and T F . The generalized dynamic reduction. MOMENTi) may be associated with timedependent functions by the Bulk Data LSEQ which can be selected by a Case Control command LOADSET. converged solutions are apt t o be saved at many intermediate steps in case of divergence and more flexible control becomes possible with multiple subcases. the PARAMeter NDAMP is used t o control the stability in the ADAPT method. VELOCITY. and Guyan reduction may be performed for upstream superelements.. However. all of the nonlinear element forces and stresses must be computed to satisfy equilibrium with the prescribed initial conditions. SVELOCITY. However. OLOAD. FORCEi. The Case Control requests which may not be changed after the first subcase are: SPC. initial conditions could be a part of the nonlinear analysis by applying static analysis for the preload using PARAM. Then the transient analysis can be performed in the ensuing subcases. The parameter NDAMP represents the numerical damping (recommended value for usual case is 0. MPC. SACCELERATION.01) which often required t o improve the stability and convergence in the contact problems. TSTATIC in the first subcase. because TLOADi data defines the loading history as a function of cumulative time. Static loads (PLOADi. SDISPLACEMENT. FORCE. not t o exceed 200 steps.g. no thermal loads or enforced displacements (SPCD) are not allowed in the nonlinear transient analysis. NONLINEAR (NLLOAD) and SPCFORCE. Requested output quantities for all the subcases are appended after the computational process for actual output operation. respectively. and iteration control selected by the TSTEPNL request. Available outputs are DISPLACEMENT. ACCELERATION. All the superelement model generation options and matrix reduction options are allowed for the linear portion of the structure. Associated with the adaptive time stepping method. element force output and GRID point stresses are not available for nonlinear elements. . time interval. STRESS.steps. However. component mode synthesis. On the other hand. DMIG. Output requests for each subcase are processed independently. it is recommended to use the same TLOAD Bulk Data for all the subcases in order t o maintain the continuity between subcases.
BFGS updates and line searches similar t o those on the NLPARM Bulk Data entry. the initial time step size should be estimated by the user according t o the aforementioned principles. The TSTEPNL Bulk Data is selectable by the Case Control command TSTEPNL. However. Modes with shorter periods (higher frequency) will be attenuated by the numerical process. Although the same TSTEPNL Bulk Data may be selected by more than one subcase.2 Implicit Integration Control: TSTEPNL Data The input fields of the TSTEPNL Bulk Data entry specify the time step size. The drastic change. It is advised that a drastic change in the time step size between subcases be avoided. it is recommended t o select different TSTEPNL entry for each subcase in preparation for changes in the restarts.g. The controls are applicable t o the automatic time step adjustment and bisection process in addition t o stiffness matrix updates. Highly nonlinear problems may require smaller step size. Since the automatic time step adjustment algorithm does not consider the loading history. short pulses could be skipped if the time step is automatically adjusted t o a larger value than the pulse duration. A caution is necessary in using the automatic time step adjustment if the forcing function is a short duration pulse. could cause a loss of accuracy upon subcase switch. The parameters for controlling the accuracy and stability of the incremental and iterative process can be specified in the TSTEPNL Bulk Data entry.9. the number of steps. Nevertheless. e.1. The choice of time step size is determined primarily by the frequency content of the input load and the highest frequency modeofinterest. the adaptive time stepping capability is the vital part of SOL 99 or 129 and is recommended for any problem (linear or nonlinear). Since default values are resulted from numerous test runs. The TSTEPNL data format (Version 67) is shown below with default values: . A general guideline is that 7 or more steps per cycle be provided for reasonable accuracy. In such case. and the output interval as well as the nonlinear iteration options. ratio exceeding 1000. the analysis should be started with the default setting and changed if necessary. an intermediate subcase should be provided for a transition period of short interval t o reduce the ratio..
etc. stiffness is updated at every K S T E P ~successful bisection. The data will be output at steps 0. Multiple subcases are assumed t o occur sequentially in time. The METHOD field selects an option for direct time integration and the stiffness matrix update strategies among ADAPT.e. If the TSTEP option is selected..e. but AUTO or TSTEP *ill abort the run if the model does not have any data representing nonlinearity. The total duration for the subcase can be assessed by multiplying NDT with DT (i. the time increment (At) changes during the analysis in the ADAPT option and the actual number of time steps will not be equal t o NDT. The ADAPT method allows linear transient analysis. and is equal t o DT..75 The TSTEPNL Bulk Data entry is selected using ID by the Case Control command TSTEPNL. irrespective of the option selected. the program automatically updates the stiffness matrix t o improve convergence while KSTEP value is ignored. the program updates the stiffness matrix every KSTEP~~ increment of time.0 2 2 0. the initial conditions of each subcase are defined by the end conditions of the previous subcase..E2 5 1. Therefore. i. The stiffness matrix is always updated for a new subcase or restart. AUTO and TSTEP.Format: 1 2 ID EPSU MAXBIS 3 4 5 NO MAXDIV RB 6 METHOD MAXQN MAXR 7 KSTEP MAXLS UTOL 8 9 10 +TNLl +TNL2 TSTEPNL +TNLl +TNL2 NDT EPSP ADJUST DT EPSW MSTEP MAXITER FSTRESS RTOLB CONV Exam~le with defaults: TSTEPNL +TNLl +TNL2 250 l. The Case Control command OTIME may also be used t o control the output points. . NDT*DT). The NDT field specifies the number of time steps with DT as the size of each time step. However. If the ADAPT option is selected. NO. Each subcase (residual superelement solutions only) requires a TSTEPNL entry.E6 0 2 0.2 20. 2N0. PW +TNL2 +TNLl 1. . the program automatically adjusts the incremental time and uses the bisection algorithm in case of divergence. The NO field specifies the time step interval for output. The time increment (At) remains constant during the analysis in AUTO and TSTEP options.1 10 0. and the last converged step for printing and plotting purposes. During ~ the bisection process in ADAPT option... If the AUTO option is selected.E3 5 1 ADAPT 10 16. In ADAPT option DT is used as an initial value for At. every ~0~~ step solution is saved for output.
the stress state is adjusted to the current yield surface. FSTRESS is also used to establish a tolerance for error correction in the elastoplastic material. The BFGS quasiNewton updates and the line search process work in the way as in static analysis except for the default setting.e. If NDIV reaches MAXDIV twice within the same time step. The convergence test is controlled by convergence test flags (U for displacement error test. The MAXQN field defines the maximum number of quasiNewton vectors t o be saved on the database and the MAXLS defines the number of line searches allowed per iteration.. i. If MAXITER is positive. The number of bisections for a time increment is limited t o (MAXBISI. The sign of MAXITER provides a control over the ultimate recourse (reiteration) in case of failure in convergence or bisection. If MAXBIS is negative and the solution does not converge in JMAXBISI bisection. the process is treated as a divergent process. It should be noted that at least two iterations are necessary to check the displacement convergence criterion. the analysis is terminated when the divergence condition is encountered twice during the same time step or the solution diverges for five consecutive time steps. and/or W ) are satisfied upon convergence.. Otherwise. If the solution does not converge in MAXITER iterations. P for load equilibrium error test. P. the analysis is terminated. If the MAXITER is negative. The MAXDIV field provides control over diverging solutions. respectively.e. Depending on the rate of divergence. i. either a bisection or stiffness matrix update takes place based on the value of MAXBIS. The solution is assumed to be divergent when NDIV reaches MAXDIV during the iteration. the analysis is terminated with a fatal message. All requested criteria (combination of U. the solution for the time step is repeated with a new stiffness based on the converged state at the beginning of the time step. MAXBIS is the maximum number of bisections allowed for each time step (9 5 MAXBIS 5 9). The number of subincrements in the material routines is determined such that the subincrement size is approximately FSTRESS * a (equivalent stress). . W for work error test) and the error tolerances (EPSU. Otherwise. Error in yield function < FSTRESS * a If the limit is exceeded at the converging state. which is used to limit the The FSTRESS field defines a fraction of the effective stress (5) subincrement size in the material routine. The bisection process is activated when divergence occurs and MAXBIS # 0. the program computes the best attainable solution and continues the analysis on second divergence. Nonzero values of MAXQN and MAXLS activate the quasiNewton update and the line search process. the program will exit with a fatal error message. the number of diverging solutions (NDIV) is incremented by 1 or 2. the best solution is computed and the analysis is continued t o the next time step. EPSP and EPSW) which define the convergence criteria. If the bisection option is used with the ADAPT method. If MAXBIS is positive and the solution does not converge after MAXBIS bisections.The number of iterations for a time step is limited t o MAXITER. the time step is bisected upon divergence.
0). MSTEP defines the desired number of time steps to obtain the dominant period response accurately (10 _< Integer 5 200): RB defines bounds for maintaining the same time step for the stepping function in the automatic time step adjustment method (0. Parameters MSTEP and RB are used to adjust the time increment during the analysis in the ADAPT option.e. It is user's responsibility to ensure that the loading history is properly traced with ADJUST option. the program automatically computes the value of MSTEP based on the changes in the stiffness. The MAXR field defines the maximum ratio for the adjusted incremental time relative t o DT allowed for time step adjustment (1. If ADJUST is positive.5 = 1. A value of ADJUST an order greater than NDT will turn off adjustements after the initial adjustement.1 Real 5 1.5 * RB 0.. the time increment is continually adjusted for the first few steps until a good value of At is obtained.001 < Real 5 1.The parameter ADJUST allows the user t o control the automatic time stepping in the ADAPT option. UTOL is used t o filter undesirable time step ./RB r 3. In the default option.0 for for for for < 0. Since the automatic time step adjustment is based on the mode of response and not on the loading pattern. A larger value (e.g.0).5*RB 5 r < RB RB r < 2 2.0 = 4. If unsure. The ADJUST option should be suppressed for the duration of short pulse loading.0 5 Real 32. < DT (2MAXBIS ' < At 5 s) MAXR * D T DT The UTOL field defines the tolerance on displacement increment below which there is no time step adjustment (0. The time increment is adjusted as follows: < where 1 = MSTEP with f f f f f = 0. start with DT which is much smaller than the pulse duration in order to properly represent the loading pattern. MAXR is used to define the upper and lower bounds for adjusted time step size. The adjustment is based on the number of time steps desired to capture the dominant frequency response accurately. the time increment is adjusted every ADJUST time steps only./RB < > The recommmended value of MSTEP for nearly linear problems is 20. it may be necessary to limit the adjustable step size when the period of the forcing function is much shorter than the period of dominant response frequency of the structure. After this initial adjustment. 40) is required for highly nonlinear problems.0 = 2.25 for r = 0. A value of zero for ADJUST turns off the automatic adjustment completely. i..0). 5 r < 3.
. no time step adjustment is performed if The RTOLB field defines the maximum value of incremental rotation (in degrees) allowed per iteration t o activate bisection (Real > 2...0).. 9. the process information consisting of iteration data is printed at the end of each iteration or time step.) exceeds the value specified for RTOLB.1. or A0. II~211. This bisection strategy based on the incremental rotation is controlled by MAXBIS field. LOAD: Relative error in terms of load vectors defined as for ADAPT method for AUTO or TSTEP where tively. They are printed under the following heading: TIME: Cumulative time for the duration of the analysis ITER: Iteration count for each time step DISP: Relative error in terms of displacements defined as for ADAPT method for AUTO or TSTEP where urnax= max(llulll... respec WORK: Relative error in terms of work defined as . F and P are threepoint average values for internal and external forces.. i.e. IIunll) . AO.adjustment. The bisection is activated if the incremental rotation for any degreeoffreedom (AB.3 Iteration Related Output Data During the incremental and iterative computation.
computed only for AUTO or TSTEP method DLMAG: Absolute norm of the load vector. error before line search EFINAL: Error at the end of line search NQNV: Number of quasiNewton vectors appended NLS: Number of line searches performed during the iteration ITR DIV: Number of occurrences of divergence detected during the adaptive iteration by the iteration module NLTRD2 MAT DIV: Number of occurrences of bisection conditions in the material routine (excessive stress increment) or in the rotation angle (excessive rotation) during the iteration using the ADAPT method NO. I( R 11.LAMBDA(1): Rate of Convergence in iteration where LAMBDA(T): Ratio of the load error for two consecutive time steps computed only for AUTO or TSTEP method Pn LAMBDABAR: Average of LAMBDA(T) over the last 3 steps. DLMAG by 1) R I(< 1012. The absolute convergence is defined using FACTOR: Final value of the line search parameter EFIRST: Divergence rate. BIS: Number of bisections executed for the current time interval ADJUST: Ratio of time step adjustment relative to DT within a subcase .
can be directly read from the printout of the previous run. SLOOPID. direct input matrices. It is recommended for the user t o restart from a copy of the input SOL 66 database.2. If the same model is t o be reexecuted. This deficiency is corrected in Version 67. 9. only the first subcase is affected. The normal restart for a transient run is t o be continued from the last step of a previous subcase with different loads and/or TSTEPNL data.2 RESTARTS Since SOL 66 (or 106) and SOL 99 (or 129) share the same database storage for nonlinear tables and matrices. the restart system for transient analysis can use either a previous static or transient nonlinear analysis as its initial conditions. only the residual superelement needs to be reassembled (SEMA. static t o dynamic.2 Restarting within SOL 99 Restarting within SOL 99 allows static to static. Restarts from a previous nonlinear transient execution are available for a number of cases.9. N where N is the printed value of LOOPID for the desired static solution. Simply provide a database created in SOL 66 and specify the parameter . mass. For the normal restart provide the following parameters: LOOPID = N STIME = t : Start from the Nth subcase : Start from time t Note that constraint sets should not be changed to avoid incompatible matrix sizes. Constraint sets. add N dummy SUBCASE commands t o start the residual Case Control execution and STIME = 0. PARAM.2. If the final results from the previous transient run are to be used as the initial conditions at t=O. 9. The values of LOOPID and STIME.1 Restarting from SOL 66 into SOL 99 The options for a restart from SOL 66 into SOL 99 are static to static and static t o dynamic analysis. and dynamic t o dynamic analysis. The initial transient load should be identical t o static loads at the restart state. It is noted that SOL 99 destroyed (until Version 66) the database for any subsequent static analysis once a SOL 66 database is used for restart. SELA = 0). For a restart from a previous static analysis. If the adaptive time stepping algorithm . and damping may be changed. which are printed with the iteration information for each subcase.
the NO on the TSTEPNL entry). If a SOL 99 run is terminated abnormally in the middle of a subcase. SVELO and SACCE sets are not.e. Once STIME and LOOPID are known. however. If METHOD = ADAPT (NLTRD2 module). The latter case also requires parameters LOOPID and STIME.is not activated. Insert a new subcase which references the new TSTEPNL entry prior t o the remaining subcases in the Case Control section. OLOAD. In the second case the solution can only be restarted from the end of a subcase. determine the number of remaining time steps in the subcase and create a new TSTEPNL entry for the remaining time. the last converged solution is treated as an output step and is always saved for the restart so that STIME can be the time of the last converged step. STIME is the time corresponding t o the last output step which may be calcualted based on the output skip factor (i. the input value for STIME differs depending on the value of METHOD specified on the TSTEPNL entry. it may or may not be restartable depending upon the cause of the abnormal exit. the solution can be restarted from the termination point as in the solution diverging case.. and DT. are the number of time steps and the time increment of the ith subcase. respectively. In the first case. The restart procedure for the former is identical t o that for the normal restart as described in the preceding paragraph. which is a new feature implemented in Version 66. When completed. SPCF (printout and plotting) and NLLOAD (plotting only) are recoverable while STRESS. the value o f t may also be calculated by the following equation: t= NDTi r DT. or (2) the CPU time exceeds the limit specified in the computer system which leads to a sudden job abortion by the system. If the job is stopped due to a diverging solution.. ACCE. A solution may be terminated in the middle of a subcase due t o insufficient CPU time: (1) the CPU time specified in the Executive Control section is insufficient so the run is forced t o exit by the program. if the ADAPT method is used. . The values of STIME and LOOPID can also be found in the printout. Restarts may also be performed solely for data recovery by providing the following parameters: SDATA = 1 : Recover data without running the solution module LOOPID = N : from the 1st through the Nth subcases. it can be restarted either from the end of a previous subcase or from the last saved solution step. If METHOD = AUTO or TSTEP (the NLTRD module). Note that solution sets DISP. MSCINASTRAN is able t o detect the specified CPU time in the Executive Control section and automatically activate the wrappingup procedure before time expiration. VELO. where NDT. SDISP.
02 in. NLLOAD and SPC output.0 min. The third restart was simply t o recover the DISP. ACCE. The input data for this problem is given in Table 9. .7 msec instead of the specified time span.i 50 in 50 in +/ p = 0. (~10010) Forcing Function NOLINl Representing GAP Figure9. 9. the execution time was changed t o 5.1. VELO. As expected. the job ran t o completion at 9.2. A new subcase with 600 time steps was added in the first restart run t o extend the time span for another 30 msec. The beam is modeled using 20 linear BEAM elements.9. In the second restart run.02 inches. the run was aborted a t 36 msec due t o insufficient time. OLOAD. 9. The adaptive time stepping method was turned off and the execution time was specified to be 0.2.3 lb/in3 20 Beam Element 0.2.2. A stopper (snubber) is present underneath the center of the beam with a clearance of 0.3 Example A simply supported elastic beam is subjected t o a single cycle of a sinusoidal forcing function at the quarter span as illustrated in Fig. The problem is linear except for the effect of the stopper. The response at the loading point of the beam is shown in Fig.1 msec was analyzed in the cold start run. 1 7 t.8 min. 100 x 0.2.2. One subcase with 100 time steps and an initial time step size of 0. and another subcase with 164 time steps and a time increment of 0.1. Since the adaptive time stepping algorithm was activated. This gap is simulated by a displacement dependent force which is active only when the beam is in contact with the snubber.1.1 = 10 msec.SimplySupportedBeamWithaRestrainedMotion .1 msec was added t o extend the analysis time span up t o 40 msec.
Simply Supported Beam: Loading Point Response . AT POINT A Figure 9.2.VERT.2. DISPL.
V66 $ SSH 7/10/91 TIME 5 $ SOL 99 $ NONLINEAR TRANSIENT ANALYSIS $ PRINT MATRIX TRAILERS AND ITERATIONS DIAG 8.Table 9.LOOPID. I. 1. CBAR CBAR 102 100 I0001 10002 0. NONLINEAR FORCE LABEL = NOLIN IN SOL 99 SEALL = ALL ECHO = UNSORT SET I = 10005 SET 2 = 10010 SET 3 = 10000 SPC = 1002 $ $PARAM. Input Data Listing for Simply Supported Beam $ ***added for the restart runs $RESTART ID TNOLIN.STIME.STIME. 0.LOOPID.0.O093 $ *** used for the 1st restart $PARAM.3 $PARAM.2.O.2 $PARAM.LOOPID.SDATA.1. 0.I $ *** used for the 3rd restart SUBCASE 1 DLOAD = 30 TSTEPNL = 10 NONLINEAR = 13 $ SELECT NONLINEAR FORCE SUBCASE 2 DLOAD = 30 TSTEPNL = 20 NONLINEAR = 13 $ SELECT NONLINEAR FORCE SUBCASE 3 DLOAD = 30 TSTEPNL = 30 NONLINEAR = 13 $ SELECT NONLINEAR FORCE BEGIN BULK $ MODELING INFORMATION FOR BEAM ONLY 101 100 I0000 I0001 0.50 CEND TITLE = DYNAMICS HANDBOOK DEMONSTRATION PROBLEM SUBTITLE = DIRECT TRANSIENT RESPONSE.I $PARAM. .0236 $$ $ $ $ $ $ *** used for the 2nd restart DISPL=2 VELO=2 ACCEL=2 OLOAD=I NLLOAD=2 SPCF=3 $PARAM.
0. GRID 10016 80. CBAR 105 100 10004 10005 0. 0. 0. 0. 0. 0. 1. GRID 10017 85. 0. CBAR 113 100 10012 10013 0. 0. 0. GRID 10001 5. 0. GRID 10002 10. 0. 0. CBAR 110 100 10009 I0010 0. SPC 1002 10 123456 SPC 1002 10020 3 10000 $ . 0. CBAR 107 100 10006 10007 0. CBAR 114 100 10013 10014 0.3 PARAM GRDPNT 10010 PARAM WTMASS . 0. GRID 10015 75. 0. CONM2 12 I0010 .002588 PBAR 100 1000 . 0. 0. GRID 10019 95. GRID 10007 35. 0. 0. GRID 10014 70. 0. 0. 0.+7 . 0. CBAR 117 100 10016 10017 0. 0. 0. CBAR 112 100 I0011 10012 0.I5708 + P B I I.3 . 0. 0. 0. CBAR 108 100 10007 10008 0. 0. GRID 10012 60. CBAR 104 100 10003 10004 0.I GRID 10 50. GRID 10008 40. GRID 10020 100. 0. GRID 10013 65. CBAR 116 100 10015 10016 0. MAT1 1000 3.31416 . 0. GRID 10006 30. 0. CBAR 118 100 10017 10018 0. 0.CBAR 103 100 10002 10003 0. GRID 10018 90. GRID 10004 20. 0. GRID 10011 55. CBAR 115 100 10014 10015 0. CBAR 120 100 10019 10020 0. GRID 10005 25. 0. 0. 0. CBAR 119 100 10018 I0019 0. CBAR 109 100 10008 I0009 0. GRID 10000 0. CBAR Ill 100 10010 10011 0. 0. 0. GRID 10010 50. CBAR 106 100 10005 10006 0. GRID 10003 15. 0. GRID 10009 45. 0. 0. 0.
0.33 ADAPT $ $ MODELING INFORMATION FOR NONLINEAR SPRING NOLINI 13 I0010 TABLED1 13 +TAB 2.00010 I +TS3I +TS32 0 I 0.2 TSTEPNL 10 100 . 13 ENDT 2. 3 0.+8 90. . MATS1 10 PLASTIC 0. $ LOADING AND SOLUTION INFORMATION TLOAD2 30 33 DAREA 33 10005 3 47. ADAPT ADAPT I 3.0001 1 TSTEPNL 20 600 .00005 1 +TS21 +TS22 0 TSTEPNL 30 164 .95 $ 3 I.$ MODELING INFORMATION FOR CENTER SPRING CROD 10 10 10 10010 MAT1 10 10.011451 87. PROD 10 10 I. ENDDATA .02 0. 10010 0.52 4.
9. Nonstructural mass distributed over elements (mass per unit length for line elements or mass per unit area for surface elements) is specified in the NSM field of property Bulk Data entries. As an option. including mass elements (CONMI. parameters (WTMASS. Applications of this feature are floor loads. or in other words. They are described below: The density (mass per unit volume) is specified in the RHO field of a MATi Bulk Data entry and is used t o automatically compute the mass for all structural elements which reference the MATi entry. the inertia force fi = . .  The CONMl Bulk Data entry provides a general description of a concentrated mass at a grid point of the structural model by defining a 6x6 symmetric matrix of mass coefficients at the grid point. GRDPNT) and fields in material or property Bulk Data entries. and M is the mass coefficient (specified on the CMASSi entry or on the PMASS entry if i = 2 or 4). COUPMASS. then inertia forces will result only when u l is not equal t o u2.The CONM2 Bulk Data entry provides a more convenient definition of a concentrated mass at a grid point. If this is the case. . M is added t o the mass matrix in the diagonal position corresponding t o ul.M u l is generated. The mass elements have inertia properties directly specified by the user. the center of mass may be measured from the origin of the basic coordinate system rather than as an offset from the grid point. Applications of CONMl are rare. Another application is for modeling an inertia term in relative coordinates. If the ul and u terms are both 2 defined on the CMASSi entry. the offset of the center of mass from the grid point.3. An important application of the CMASSi entry occurs in the recommended method for specifying enforced motion a t grid points. electrical cables or thermal insulating materials. CONM2. In most applications. The form of the relationship is where fl and f2 are the inertia forces acting at degrees of freedom 1 and 2. and the moments and products of inertia about the center of mass. the second degree of freedom is not specified. respectively.  The CMASSi Bulk Data entries provide mass coupling between any two degrees of freedom.1 DYNAMIC EFFECTS: MASS AND DAMPING Mass Modeling Mass properties of the finite element model may be specified in several different ways. The inputs t o CONM2 are the mass.3 9. CMASSi).
Unlike the stiffness matrix. the distributed mass is replaced by a nondiagonal mass matrix which couples the connected grid points. In the lumped mass option. 9. and the mass input is specified in weight pounds. V1 causes the Grid Point Weight Generator to be executed.PARAM. using the mass density on the MATi entry a Nonstructural mass for all elements specifying a value on the property entry a Scalar masses defined on CMASSi entries and M2 corresponds t o the direct input matrices which are generated by transfer functions specified by T F Bulk Data entries or supplied directly via DMIG or DMIAX Bulk Data entries. the distributed mass of an element is simply divided up and the parts are located at the connected grid points. The mass matrix for the deformed structure is the same for the undeformed structure in material nonlinear analysis. then V1 = & . . but uses more computer time. the mass matrix is formulated as where M is the mass matrix. 1 requests that the coupled mass option be used rather than the lumped mass option (default) for all elements which support the coupled mass option. During large displacement analysis.3. COUPMASS. if the inlbsec system is used = 0. GRDPNT. except plate elements without membrane stiffness. The lumped mass option is therefore prefered for computational speed. The coupled mass option is generally more accurate. V1 specifies a factor by which t o multiply the mass input data t o obtain dimensionally correct mass. the mass matrix is generated only once and is not updated during the analysis. . For example.PARAM. This situation should not cause much concern. This capability can be used as a check on mass and geometric input data. WTMASS.00259. precautions can be taken so that the mass matrix for the deformed structure is very similar t o the one for the undeformed structure.a The three parameters which relate to the specification of mass properties are  PARAM. The value of V1 identifies a grid point at which the rigid body mass properties of the structure will be computed.2 Mass Matrix For transient response analysis. In the coupled mass option. Ml contains mass terms from the following sources A 6x6 matrix of mass coefficients at a grid point defined on a CONMl entry a A concentrated mass element defined on a CONM2 entry a Structural mass for all elements.
This change. As with element rotational inertias. the rotational inertia terms defined in the global coordinate system change. cm2 .. 3. only translational terms are generated. This rigid structure should emulate the mass properties of the concentrated mass. concentrated rotational inertia terms should not be used. Do not use the coupled mass option (PARAM. a user wishes t o simulate a rigid mass with the following properties: Total mass = 10 kg I. The x component of the elastic force will not be resisted. Initially the elastic bending and shear forces are inline with the inertial forces of the mass.COUPMASS. By using the lumped mass option (default). As it rotates. then specify the same mass on all three translational degrees of freedom. This causes the element mass distribution and rotational inertias t o change. As the elements rotate. 2. As the beam is deformed by an end load the elastic forces follow the deformation (stiffness matrix is updated). As the actual structure rotates at the concentrated mass point. For this reason. Rotational inertias about the three directions of the global coordinate are usually different. If rotational inertia terms are necessary then a rigid structure with concentrated translational masses should be used. 1. is not reflected in the mass matrix since it is formulated only once and is not updated. If concentrated masses are used.The mass matrix is defined in the global coordinate system using the initial undeformed geometry. This change in rotational inertia is not taken into account by the mass matrix.cm2 Izz = 30 kg . For example. Translational terms are unaffected by element distortions. The coupled mass option generates rotational inertia terms for many of the elements. the inertias about the directions of the global system change.l). the concentrated mass will also rotate with respect t o the global coordinate system. The element mass distribution due to element deformation is usually minimal and will in most cases cause little error. The mass is only defined in the ydirection and will resist only the y component of the elastic force. 4. the same situation exists for concentrated rotational inertias. In the extreme case the elastic force is unresisted. Consider the following example: a cantilevered beam with a mass at the free end defined only in the ydirection. During large displacement the structure deforms and rotates relative t o the global coordinate system. cm2 Ivy= 20 kg . This is required t o keep the inertial forces consistent with the elastic forces of the structure. however. These terms are defined in the global coordinate system. The changes in the rotational inertia due t o element rotation and elongation are more severe and may cause large errors. = 10 kg . The following guidelines will help avoid most problems when using the large displacement option.
cm2) 2m. my and m. and I . There .2 = I. (20 kg . cm2) 2m. Solving for I. 1 = 4 cm..) at the ends of rigid bars (stiff CBEAMs)... = 3 kg were selected. The two types of damping perform different functions. = total mass (10 kg) 2myli + 2m. y and zaxes. In this example m. are lengths of the rigid bars in the x. The two damping types can be combined t o model the damping characteristics of a structure.... They are used mainly t o model actual damping components in a structure.1.la = I. The bars will extend an equal distance in the positive and negative directions from the grid along the global coordinate system axes. The rotational inertia will be modeled by lumped masses (two mass points of equal magnitude in each direction: m.12 + 2m. the CVISC and CDAMPi elements are not used t o add stability t o a structure during transient response analysis. (30 kg .l2 + 2m. .3 Damping Modeling Damping represents the energy dissipation observed in the structure and results from many sources including Viscous effects (dashpot.where the principal axes of the mass align with the global coordinate system. such as hydraulic dampers and viscous interface material. 1 and 1. (10 kg . i.3. = 1 kg. 9. where l. results in I. When using the discrete elements in this situation.li = I. are more unknowns than equations. respectively. shock absorber) Internal friction (characteristic of material type. hysterisis) External friction (slippage in structural joints) Structural nonlinearities (plasticity) In MSC/NASTRAN.e.. damping is divided into two types: viscous and structural. = f i cm. = 0 cm. cm2) 2m. = 1 kg. Because stable damping situations are difficult t o model using discrete elements. This rigid structure will simulate the translational and rotational properties of the concentrated mass as the structure rotates because the motion of the rigid bars is taken into account during the stiffness matrix updating procedure.. Viscous damping is input using discrete damping elements (CVISC and CDAMPi). I. The bar lengths and concentrated masses must satisfy the following requirements: + 2my + 2m. m. I. the major relative motion is assumed t o be between the coupled degrees of freedom or along the initial axis connecting the grid points. Viscous damping is used t o simulate actual structural damping elements whereas structural damping is used t o simulate the inherent damping properties of materials. and m. and .
Damping due to internal material friction and damping due t o bolted or riveted joints behave more like structural damping than viscous damping. whereas a composite material may have a 4% damping value.1 compares the viscous and structural damping forces as a function of frequency.03 x W4 x At is recommended for stability. Structural damping is included in a model in an attempt to simulate the intrinsic material damping. regardless of the material.. A lightweight metal structure may only have a 0. i..e.e. and PARAM W4. PARAM W3. The user should choose a realistic value (0.Structural damping is a global damping proportional to the stiffness. A change in the damping value may have a significant effect on the calculation of the response. .5% damping value.5% t o 4%).3. and the structural force is proportional t o displacement. i. The viscous damping force is proportional to velocity. dependent upon the material in the model. 9. The matrix is formed from the input in the GE field of MATi or via PARAM G. A minimum value of 0. Viscous and structural damping are related by with which results in an equivalence a t resonance with Fig.
then fi = B u l .3. and B is the damping coefficient (specified on the CDAMPi entry or on the PDAMP entry if i = 2 or 4). If the second degree of freedom is not specified.all of the linear damping options may be used in nonlinear transient analysis. B is added t o the damping matrix in the diagonal position corresponding to ul. The form of the relationship is where fi and f2 are the damping forces acting at degrees of freedom 1 and 2. The CDAMPi Bulk Data entries define a scalar damping element using a 2x2 viscous damping matrix between any two degrees of freedom.e. The effect of GE is t o multiply the stiffness matrix for an element which references the MATi entry by the scalar factor (1 ig) where g is the value of the GE field and i is the operator for imaginary number.. + . Two degrees of freedom can be in any direction and may not act like a rod element. Structural damping is specified in the GE field of a MATi Bulk Data entry. respectively. It acts like a rod element with viscous extensional and torsional properties rather than elastic properties. The damping input options in the finite element model are described below: The CVISC Bulk Data entry defines a viscous damping element that may be connected between any two grid points.1 Comparison of Viscous and Structural Damping Except for the overall structural damping parameter ( G ) . i.Damping Force I Wn W Figure 9.
W4 selects the frequency (default = 0.0) to convert the element structural damping (including nonlinear elements) to viscous damping. NDAMP provides numerical damping in the ADAPT method of SOL 99 for numerical stability (implemented in Version 67).025). DMIAX) G : Overall structural damping coefficient.3. . W3 selects the frequency (default = 0.a PARAM. : initial element stiffness matrix. : structural damping constant.4 : frequency of interest. For the geometric nonlinear analysis. for instance. specified on MATi W.4 Damping Matrix For transient response analysis. For material nonlinear analysis. specified by PARAM G W3 : frequency of interest. specified by PARAM W3 KI : global linear stiffness matrix G . the viscous damping matrix is generated only once and is not UPdated.0) for the linear part of the stiffness matrix. PARAM. with zero being no numerical damping (default is 0. a ' a 9. in nonlinear transient analysis. In both cases caution must be exercised when using discrete damping elements in transient analysis. G specifies an overall structural damping factor (default = 0. specified by PARAM W4 K. PARAM. the damping matrix is formulated as where B ' : contains terms from viscous damping elements (CVISC.0) to convert the overall structural damping on the linear elements to viscous damping. PARAM. The numerical damping increases as the value of NDAMP increases. generated for t=O Like the mass matrix. the viscous damping matrix will be inaccurate for the deformed structure with large rotation because the direction has been changed and the damping matrix is not rotated. CDAMP) B~ : direct input matrices generated by transfer functions (TF entries) or supplied by direct matrix input (DMIG. the viscous damping matrix for the deformed structure remains constant as for the undeformed structure based on the elastic material.
May not be changed between solution subcases. TLOADl defines the load history in TABLED format while TLOAD2 defines the load history by an analytic expression. NLLOAD: Required for NOLINi output request. the traditional nonlinear loads (NOLINi option) may be used in both types of transient solutions. Each stage of the transient solution may be defined with a separate SUBCASE. These are useful for simulating nonlinear damping mechanisms and rotational coupling not provided by the GAP elements. Input Bulk Data associated with the transient loads in SOL 99 (or 129) are briefly described below: TLOADl. They can be a load applied at a particular degree of freedom. NOLINEAR: Used t o select a family of NOLINi functions.1 Transient Loads Transient loads define the loadings as functions of time and the location. Functions may be tabular such as an earthquake or a booster liftoff. or the body force simulating an acceleration. A single degree of freedom or a set of GRID points may be loaded with force pattern that varies with time. for each degree of freedom.4 TRANSIENT LOADS AND INITIAL CONDITIONS The methods employed t o define loads in nonlinear transient analysis are similar t o those used in the linear solutions. or coefficients). LOADSET: Required for LSEQ processing. TLOAD2: Required data defining the coupling between Load factors (DAREA Sets) and functions of time (TABLED. The time history is provided by TLOADi Bulk Data and the static loads are converted t o dynamic loads using LSEQ option. DAREA: Optional Bulk Data defining simple load scale factors. u j requested in set i. pressure over the surface area. The DLOAD data is restricted to act only in the time defined in its subcase. Fi(t). Note that each TLOADi input must have a unique identification number.9. The DAREA factors may also be defined with static load . They may be scaled and combined with other loads t o simulate complex loading problems. For special problems involving simple scalar nonlinearities. each with its own DLOAD request. A i j . Simple static load sets generated in both upstream superelements and the default residual may be used to create the dynamic loads. The transient dynamic loads are selected in the Case Control section. or they may be simple analytic functions such as a sine wave. 9. Must be placed above the solution subcases and/or in the Superelement Subcases. DELAY. controlled by the following commands: DLOAD: Required t o select TLOADi or DLOAD Bulk Data loads.4. dynamic load i. Specifies the time funcfor tions.
9. Basic analog nonlinear transfer functions are defined. The name was chosen by a dynamicist who viewed the problem in terms of timedependent pressures and locationdependent "areas".4 : Used with TLOADl input to specify general time varying tabular functions.1. DELAY: Specifies the delay time for applying the forcing function defined in TLOADi t o each GRID point.data with the LSEQ Bulk Data.2. TABLEDi. 9.4. DLOAD: Combines different TLOADi functions into a single set and scales each by a factor. Line search and BFGS solution methods are recommended for these models. The Bulk Data DELAY is associated with the degree of freedom and not the forcing function in each TLOADi.4. is: where the factors are defined by various Bulk Data inputs as shown in Fig. uj . LSEQ: Generates transient load history for static loads and controls the assembly of static load vectors t o be used in dynamic analysis.2. i=1. The general form of the transient load.3. Since extrapolation is used beyond the range of data the user is cautioned to be careful with the end points. t o be generated and will label it with the DAREA identification. This option is useful for defining loads that travel across a structure such as an oblique wave along a dam or a bump in the road exciting the wheels of a traveling vehicle. Unlike statics. this is the only method to combine loads in dynamic analysis. It will cause a static load vector A. However. This saves the analyst the effort of defining multiple TLOADi inputs for simple time lags.3. NOLINl : Nonlinear transient load as a tabular function NOLIN2 : Nonlinear transient load as products of two variables NOLIN3 : Nonlinear transient load as a positive variable raised t o a power NOLIN4 : Nonlinear transient load as a negative variable raised t o a power A diagram of input data linkages is shown in Fig. for each degree of freedom. Performs the same function as the static LOAD Bulk Data input. these functions do not generate tangent matrix terms and the Newton iteration method may experience difficulties in convergence. Pkj. i=1.2. . NOLINi. This is useful for modeling problems with distributed transient loading such as GRAV loads or pressurized areas.4 : Define specific nonlinear scalar loads as functions of velocity and/or displacement. Each set is equivalent to a Subcase in Case Control. C k i .
Case Control Section Q DLOAD 1 LOADSET 1 I I i""" Bulk Data Section Figure 9.4.1 Dynamic Loads for Transient Analysis .
 I SLT I Note : Each load is scaled by [S x Si x f ( t ) ]at each time step in dynamic loads.. ...... Figure 9.. .... . : I I 1 I I 1 MOMENT PLOAD4 I I L ....... ._ ......_ _ .._ _ _ _ _ _ _ DAREA IDAREA ..4. ( ~ f ( t ) ): I 4 _ _ .. ..DLOAD ( x S > : :  TLOADl TABLED1 ( X f(t))  TLOADl TABLED2 ( X f(t)) TLOAD2 . I I FORCE1 DAREA I LOAD PLOAD2 l .... .... . .. .2 Example of Combining Loads ....... I : I : .....
2 Enforced Motion Excitation of the base of a structure with a timedependent motion (displacement. which in turn are converted to acceleration by the program t o compute the inertia force as depicted in Fig. It may be necessary t o reduce the error tolerances on the TSTEPNL inputs depending on the size of the large base mass. The large stiffness method has the advantage in case of enforced displacement. the "large mass or large stiffness approach" requires special caution in choosing convergence error tolerances. velocity. or acceleration) is provided by MSCJNASTRAN indirectly via the "large mass method". The actual errors will be normalized by terms proportional t o the large base forces. because it avoids the roundoff error of numerical differentiation which occurs while computing accelerations from displacements in the large mass approach. .3. the load of magnitude Pb will produce an approximate base acceleration: The accuracy of this approximation increases as Mb is made larger in comparison to the mass of the structure. It is recommended that the value of Mb be approximately 1000 times the mass of the structure. whichever is more convenient.9. However. For Nonlinear problems. and the value of Mb is input on DAREA or DLOAD Bulk Data entry. the applied force would be computed by Pb = Kbub where Kb is the stiff stiffness and ub is the enforced displacement. CMASSi or CONM2 entry should be used to input the large mass. CMASSi entries allow different masses for different directions. numerical difficulties arise if Mb is too large. In effect the user places a large mass element on the degrees of freedom representing the base and then provides a large force proportional to the desired accelerations.4. The factor 1000 is a safe limit that will produce three digits of numerical accuracy.4. A large stiffness method could be used in liu of the large mass approach. When the large mass method is used. If a very large mass (Mb) is connected t o a degree of freedom on an unconstrained structure. The TLOADi inputs allow the specification of displacements and velocities. In this case. 9. the function iib is normally specified as input loads.
4. the initial loads on the structure will not be in equilibrium.3 Options in TLOADi (Types 1. As in linear analysis. 2 and 3) 9. the equivalent loads are: where [K] and [C] are the linear stiffness and damping matrices. If u and vo are the initial displacement and velocity vectors. TIC Bulk Data are used to specify pointbypoint initial displacements and velocities. This will . three methods are available for providing a set of initial displacements and/or velocities for Nonlinear Transient Analysis: 1.4.2nd derivative 1st derivative ( I EQUIVALENT ACCELERATION 1 LARGE MASS 1 4 0 EQUIVALENT LOAD Figure 9.3 Initial Conditions Initial velocities or displacements can be imposed by using the Bulk Data TIC which are selected by a Case Control command IC. respectively. Note that if the initial state is nonlinear. the system also calculates internal equivalent loads to provide o equilibrium loads at the start. Alternatively.
The static solution (in SOL 66 or 106) may use enforced displacements (SPCD) and different boundaries t o obtain the desired initial shape. an initial velocity may be generated by a large impulsive load applied for 510 small time steps. Solutions should be obtained as follows: a Restart from a Nonlinear static solution using the initial load. To be safe. 3. The user is also cautioned against supplying large initial loads in the structure that may produce initial nonlinearities. An Initial LLdummy" Subcase may be used to generate complicated initial conditions without a restart. An extra subcase is added in front of the normal transient subcases with loading functions designed to produce the desired initial condition by matching the momentum. The initial static equilibrium will be preserved if the initially supplied transient loads are identical t o the static loads.require iterations and may cause difficulty for the convergence. An initial deformation may be produced by a subcase with large time steps (larger than the lowest mode's period) which eliminate the transient behavior and produces a static solution if it converges. This method requires that the other loads be shifted in time and will also cause a shift in the output response. For instance. Start with a smaller load and increase it over a small time interval. Reduce the size of the time steps by a few orders of magnitude for the first subcase. The job may not be able t o complete the first time step. a . Restarts from static analysis (SOL 66 or SOL 99 with TSTATIC) are highly recommended for initially displaced and loaded structures. use the TIC data only for initial velocities and on linear sections of the structure. 2.
) File control block Input file status For each iteration: Initial energy for line search Nonlinear internal force: Fg Sum of nonlinear forces including follower forces: Fd NOLINi vector: Nd Displacement vector: ud Total internal force: Fd Loading error vector: Rd Iteration summary (Convergence factors.) Problem statistics (gsize. etc. and users in general are cautioned against its use. only the data marked with a plus ( f ) are printed. etc. When DIAG 51 is selected. Because of the volume of output produced. When DIAG 50 is requested. all the data are printed at each iteration. allowing the user to follow the nonlinear iteration calculations step by step. the use of DIAG 51 is recommended only for small test problems.5 Diagnostic Outputs Diagnostic output is available in SOL 99 (or 129) if DIAG 50 or 51 is specified in the Executive Control section. 9. DIAG 51 is intended for debugging purposes.9.1 NLTRD Module for AUTO or TSTEP Method For each entry into NLTRD: + + Subcase status data TSTEPNL data Core statistics (ICORE. etc. line search data.5.) + For each quasiNewton vector set: Condition number: X2 QuasiNewton vector: 6 QuasiNewton vector: y 1 Energy error: z = 7 j' 7 3 For each line search: Previous line search factor: crk Previous error: Ek New line search factor: a k + l .
9.) Problem statistics (gsize.) + For each quasiNewton vector set: Condition number: X2 QuasiNewton vector: 6 QuasiNewton vector: y 1 Energy error: z = 6T.5. etc.) File control block Input file status For each time step: NOLINi vector: Nd External load vector: P d Load vector including follower forces and NOLINs: Pd Constant portion of residual vector: R& Total internal force: Fd Initial loading error vector: Rd For each iteration: Initial energy for line search Nonlinear internal force: F' Displacement vector: ud Nonlinear internal force: Fd Total internal force: F d NOLINi vector: Nd Load vector including follower forces and NOLINs: Pd Loading error vector: Rd Iteration summary (Convergence factors.3 For each line search: Previous line search factor: crk Previous error: Ek New line search factor: a k + l . line search data.2 NLTRD2 Module for ADAPT Method For each entry into NLTRD2: + + Subcase status data TSTEPNL data Core statistics (ICORE. etc. etc.
Generalized mass: DENOM2 = Au: M Au.+* Generalized stiffness: DENOMl = Au: I< A u . AF. Magnitude of the new velocity vector: u.For each converged time step: Velocity vector: vd For each time step adjustment: Magnitude of the old velocity vector: u . Work = Au. Square of dominant frequency: wi Stiffness ratio Number of steps for the period of dominant frequency: MSTEP Controlling ratio for time step adjustment: T .
= Pn XAP. The distinction is that MSCINASTRAN uses two converged solutions in the vicinity of the buckling point to form a differential stiffness [AIi]. The tangent stiffness matrix is proportional to the displacement increments. Restart is required for a nonlinear buckling analysis after the nonpositive definite stiffness matrix is detected. + It appears that Fujikake has chosen the first approach.Chapter 10 SPECIAL APPLICATIONS 10. Since the modified Newton's method is employed with quasiNewton updates in the general purpose program. tangential stiffness matrices in the two successive increments are not immediately available in general. the internal loads are quadratic function of displacements. + 2.1]. The tangent stiffness matrix is proportional to the external loads. A similar approach to the limit point calculations was published by Fujikake [10.1. whereas Fujikake uses a converged point and an estimated data point beyond the buckling point.. he did not show how he obtains the critical displacements.. The distinction between the two approaches disappears in the linear bifurcation problem. To find an instability. which implies that the critical displacements may be obtained by extrapolating from the current state.2]. Since the tangent matrix is assumed t o change linearly. Details of the method are given below.point within a small range of nonlinear domain. However.1 10.e. i. . = Un XAU. Uc. MSCINASTRAN has chosen the second approach.e. which implies that the critical load may be linearly interpolated.1 NONLINEAR BUCKLING ANALYSIS Introduction A simple approach to nonlinear buckling analysis was introduced in SOL 66 (or 106) by using a quadratic extrapolation [10. PC. i. two methods of idealization can be contemplated: 1.
1. This capability has been available with a control parameter BUCKLE in SOL 66 since Version 65.10.4) {AP} = {Pn) where cu = {Au)~ [Kn X AK] {AU} {AUIT {AP) This procedure was first implemented with DMAP ALTER in Version 64 of MSCINASTRAN by a simple DMAP (Direct Matrix Abstraction Program) language.1. + .1.2 Formulation of Nonlinear Buckling Analysis It can be shown that the eigenvalue problem for a nonlinear buckling analysis may be approximated by (10. {un)  {unI> where The critical buckling load may be estimated as {Per) = {Pn) with + a {AP) {Pn1) (10.1) [Kn t X AK] ( 4 ) = (0) with AK = Kn  Knvl where Kn and Knvl are the stiffness matrices evaluated at the known solution points in the vicinity of the instability. The critical displacements upon instability may be estimated as with {Au) = Based on the virtual work principle.
3 Analysis Procedure Nonlinear buckling analysis capability is provided in SOL 66 (or 106) and activated by a userspecified parameter BUCKLE with a value of + l . The column with dimensions shown in Fig.4 Verification Problem: Euler Column Buckling This method should be applicable to linear problems because linear buckling is a special case of nonlinear buckling. (f) Include PARAM. of Both procedures produced identical solutions with the critical buckling load (PC. 10. The buckling analysis was performed by two independent methods as follows: 1.1.1. an Euler buckling problem of a pinended column is illustrated. It is recommended t o continue the nonlinear static analysis with small steps until a negative determinant of the stiffness matrix is encountered. For verification. (b) Provide two small loading steps (below buckling). The operational steps are 1. BUCKLE. 2. Run SOL 66 for static analysis and create a database for restarts. 10. SINV method is recommended for the eigenvalue extraction.38 lbs. Make a restart run as follows: (a) Select PARAM entries SUBID.1 is modeled with 6 CBEAM elements and subjected to a compressive axial load. . BUCKLE. The input data listing is shown in Table 10. restarted from SOL 66 database.1. Using SOL 65 (linear buckling analysis). skipping the rest of the increments of the subcase in which the stiffness matrix becomes singular. restarted from SOL 61 (linear static analysis) database. 2. 1. (d) Include EIGB Bulk Data entry with a METHOD command in the Case Control section for the eigenvalue analysis.) 120.10. (e) Provide mode shape PLOT data if desired. LOADINC and LOOPID for restart. This may be provided by adding a new subcase. Using SOL 66 with PARAM.1.1. (c) Use "KSTEP=lV in the NLPARM entry for two load steps t o be solved so that the stiffness matrix is updated for each solution.
1 Euler Buckling for Verification .48 1bs . Figure 10.r buckled shape where n = l E = 10.069 x in4 L = 6in P = 120.8 x lo6 psi I = 4.1.
50 TIME 5 $ CPU MINUTES CEND TITLE = EULER BUCKLING OF BEAMCOLUMN SUBTITLE = AXIAL FORCE. 20 +EIGB MAX UPW PW 2 2 YES . PTITLE = SIDE VIEW FIND SCALE ORIGIN I SET I PLOT SET I ORIGIN I LABEL BOTH PLOT STATIC 0 MAXI DEFO . 3..1.V65 $ CJS. Y VIEW 0. SHL 5/2/84 SOL 66 $ NONLINEAR STATIC ANALYSIS $ PRINT MATRIX TRAILERS AND ITERATIONS DIAG 8. PINENDED BOUNDARY ECHO = UNSORT SEALL = ALL DISP = ALL OLOAD = ALL SPCF = ALL STRESS = ALL METHOD = 30 $* $PARAM $PARAM $PARAM $PARAM BUCKLE SUBID LOADINC LOOPID I 2 I 1 $* * * * * CHANGES REQUIRED FOR BUCKLING SUBCASE I LOAD = 10 NLPARM = 10 SUBCASE 2 LOAD = 20 NLPARM = 20 OUTPUT(PLOT) CSCALE = 1.Table 10..0.1 Verification Problem Input Data ID BBUCKL.05 SET I ORIGIN I PLOT MODAL 0 SET I ORIGIN I BEGIN BULK $ PARAMETERS PARAM LGDISP I $ SOLUTION CONTROL NLPARM 10 I AUTO 4 AUTO I NLPARM 20 $ EIGENVALUE ANALYSIS EIGB 30 INV 0.05 AXES Z. 7MAY85.0.3 PLOTTER NAST SET I = ALL EXCEPT PLOTEL MAXI DEFO . X.
0625 0. 1.$ MATERIAL AND GEOMETRIC PROPERTIES MAT1 1 10. 123456 345 12345 . GRID 14 4. 0. 0. 0. 0.I25 $ LOADING FORCE 10 16 100. 0. 0. FORCE 20 16 140. GRID 15 5. 0. GRID 11 1. $ CONNECTIVITY CBEAM I 3 10 II 1000 CBEAM 2 3 1 1 12 1000 CBEAM 3 3 12 13 I000 CBEAM 4 3 13 14 I000 CBEAM 5 3 14 15 1000 CBEAM 6 3 15 16 I000 $ 7. 0.0625 . 125 . GRDSET GRID 10 0.0625 . 1.1252 4. 1. 0. 0. $ GEOMETRY GRID 1000 0. GRID 12 2. 0.6284 +PB1 .8+4 4.8+4 7.I25 .3 +MATI 7. 0.8+6 . GRID 16 6.225+6 1 2 PBEAM 3 I 3. 0. 0.0695 1. GRID 13 3. 0. 125 ENDDATA . 0.8+4 .0625 . 0. 0.5+4 MATS1 1 PLASTIC 1.
2. where a negative determinant of [I<]was first detected. 10. Central Deflection) .1. the nonlinear static analysis determined that buckling would occur between 3500 psi and 3600 psi.2 and Table 10.5 Example of Nonlinear Buckling An elasticplastic buckling analysis of a thin spherical shell is presented [10.1.10. The results are summarized in Fig.1.2 ElasticPlastic Buckling of a Clamped Spherical Cap with Flat Spot (Load vs.1.1]. With a model consisting of 16 shell elements (10 Q U A D 4 and one TRIAS). h s m m a 1000 = 2000 I/ @ buckling point by MSCINASTRAN + Finite Element by MSCINASTRAN @ b Finite Difference by ~ a o buckling pdnt by ~ujikakel E x ~ e r i m n t aBuckling point2 l ~ Central Deflection. W (in) Figure 10.
20 0.60 0.375 0.024+5 1.7393 3.50 S t a r 1000 2000 2200 2400 2600 2800 3000 3100 3200 3300 3400 3500 3600 5 7 7 8 lo* 1. .Table 10.2 MSCINASTRAN Output Summary Subcase 1 C Load Factor Load (psi) No. initiated at 2600 psi.125 0.0 0.279+4 8. The execution was terminated after seven iterations due t o two consecutive diverging **** Plastic deformation initiated. prescribed maximum limit of 10 with e = 3.4183 8. Reference 1 reports that the yielding K [K] update during iteration.894+4**** 8.7193 4.061+4 8.098+5   o 1.80 1.0913 8.75 1 d 2 t  lo* R e s 20 7K 7 9 17 8K 2*** 3 t a r t  * Max limit is reached. the number of iterations reached the .500 0.532+4 7.460+4 T9.25 0. *** A negative term on factor diagonal in [K] detected.343732.555+4 7. ** In this stage of the cold start run.179+4 9.0 0.625 0.1543 4.6403 5. solutions with stiffness updating.0323 11.8463 7.564+4 8.2173 5.9423 3.944+4 9. u100(in) max ale (psi) 0.40 0.9183 10.9273 6. of Iter.795+4 1.1. The iteration limit was changed t o 20 and the convergence tolerances were relaxed before the job restarted.
Solutions a t P=3400 psi and 3500 psi were repeated and a buckling analysis was performed using the differential stiffness between those two loading steps.01224 in.466 from which the critical buckling load is calculated by Pc.466 x 100 Uc.2. Notice that Fujikake [10.1. 10.1. T h e results of this eigenvaiue analysis are X = 0. = Un = 3546.6 psi and the maximum deflection a t the crown by + AAU = 0. Fig.As shown in Table 10. = P n + Q A P = 3500 + 0.2] overestimated the maximum deflection at buckling. BUCKLE starting from the solution a t P=3300 psi. Undef ormed shape Oefonned shape on buckling Figure 10.3 Deformed Shape on Buckling .1. a restart run was made with PARAM. The loading history to buckling point was obtained as shown in Fig.1.775 and cr = 0.3 shows the deformed shape on buckling.3. 10.
This structure was analyzed again with a solidelement model. the prediction becomes less reliable. a t the crown.. It seems that the deformed shape a t buckling is as significant and useful as the buckling mode shape in the largedisplacement problem.4 shows the deformed shape on buckling. . the solution may be easily validated by repeating the process with new trial loads closer t o the estimated buckling load. The buckling point is somewhat underestimated with solid model. Fig.4 Deformed Shape Upon Buckling with a Solid Model A few more analyses were attempted t o predict buckling point from the solutions at the earlier stage of loading. In such cases.1. When the solution points are farther removed from the actual buckling point. in which the predicted buckling point is not close t o the instability point detected by the static solution. = 3520. Note that the mode shapes for bifurcation buckling may be quite different.1. resulting in PC. .  UndetorW Shape Figure 10.0122 in. 10. Thus the user is cautioned t o be suspicious when the cr: is greater than unity. = 0.8 psi and U.
3 PLOTTER NAST SET 1 = ALL EXCEPT PLOTEL SET 2 = 1000 SET 3 = ALL VIEW 30.05 FIND SCALE ORIGIN 1 SET 3 PTITLE = ISOMETRIC VIEW PLOT SET 1 ORIGIN 1 SET 2 SYMBOL 3 VIEW 90.0..0.. MAXI DEFO .1.50 $ CPU MINUTES TIME 15 CEND TITLE = ELASTICPLASTIC BUCKLING OF IMPERFECT SPHERICAL SHELL A1185G SUBTITLE = HYDROSTATIC PRESSURE APPLIED.V65 $ SHL 5/7/85 N6609 $ NONLINEAR STATIC ANALYSIS SOL 66 $ PRINT MATRIX TRAILERS AND ITERATIONS DIAG 8. FIND SCALE ORIGIN 1 SET 1 PTITLE = SIDE VIEW . PERIPHERY CLAMPED ECHO = UNSORT SEALL = ALL DISP = ALL OLOAD = ALL SPCF = ALL STRESS = ALL = 10 SPC METHOD = 30 . $ $PARAM $PARAM $PARAM $PARAM BUCKLE SUBID LOADINC LOOPID 1 4 1 10 $* * * * * * CHANGES REQUIRED FOR BUCKLING SUBCASE 1 LOAD = 1 0 NLPARM = 10 SUBCASE 2 LOAD = 20 BLPARM = 20 SUBCASE 3 LOAD = 30 NLPARM = 30 $ ADDED FOR BUCKLING ANALYSIS SUBCASE 4 LOAD = 40 NLPARM = 40 OUTPUT(PLOT) CSCALE = 1.3 Example Problem Input Data ID SBUCKL..0.Table 10.20..
1506 . 10 THRU 25 PLOAD2 30 3800.29 5. I.005 5.1506 2. GRID 113 200 1.72 5.1506 0. GRID 116 200 1. CORD2S 200 0. GRID ill 200 1. 10 THRU 25 PLOAD2 40 3500.225+6 I 2 7. I. 0. 5. GRID GRID 106 200 1.1506 1. GRID 114 200 1. 0. GRDSET 100 345 1000 0. 0.72 5.43 5. I.86 5.3 MATS1 I PLASTIC 1. 32908 0.1506 . +C2S2 I. 0 12456 GRID 101 200 1. 1506 2. GRID 112 200 1. .1506 5.1506 4.0251 I $$ BOUNDARY AND LOADING SPCI 10 123456 131 132 PLOAD2 10 2000. 10 THRU 25 $$ GEOMETRY AND CONNECTIVITY CORD2S 100 0. GRID 115 200 1. 2.1506 1. GRID 107 200 1.43 5.8+4 PSHELL 2 I . 0.05 SET I ORIGIN I SET 2 SYMBOL 3 PLOT MODAL 0 SET I ORIGIN I SET 2 SYMBOL 3 PLOT BEGIN BULK $$ PARAMETERS PARAM SMALLDB I PARAM LGDISP 1 $$ SOLUTION CONTROL PW NLPARM 10 2 AUTO NLPARM 20 5 AUTO PW NO YES NLPARM 30 8 AUTO YES I NLPARM 40 2 AUTO $$ EIGENVALUE ANALYSIS EIGB 30 SINV 0.86 5. 0.1506 5. 0. 20 2 2 +EIGB MAX $$ PROPERTIES MAT1 I 10.PLOT SET I ORIGIN I SET 2 SYMBOL 3 STATIC 0 MAXI DEFO .715 5.145 5. 0.145 5. 0.1506 4. GRID 108 200 1.575 5.1506 3. GRID 110 200 1. 0. +C2SI I.29 5.575 5.8+6 .715 GRID 103 200 1. I0 THRU 25 PLOAD2 20 3000. 123456 GRID GRID 100 200 1. +EIGB .1506 5. GRID 102 200 1.1506 2.1606 2.005 5 . 0. 1. 105 200 I. GRID 104 200 1.1506 3.1506 5. 0. GRID 109 200 1.
8251 .8251 .1506 .8251 .1506 1.8251 .8251 .8251 .8251 .8251 .8251 .8251 .8251 .8251 .GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID PLOTEL CTRIA3 CQUAD4 CUD QA4 CUD QA4 CUD QA4 CUD QA4 CUD QA4 CUD QA4 CUD QA4 CUD QA4 CUD QA4 CQUAD4 CUD QA4 CUD QA4 CUD QA4 CUD QA4 E D AA NDT 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 1000 10 11 12 13 14 15 16 17 18 I9 20 21 22 23 24 25 1.8251 .8251 100 100 101 103 105 107 109 Ill 113 115 117 119 121 123 125 127 129 101 103 105 107 109 Ill 113 115 117 I19 121 123 125 127 129 131 102 104 106 108 110 112 114 116 118 120 122 124 126 128 130 132 .
.
This is an alternate method to the SOL 64 SOL 63 sequence for nonlinear normal mode analysis. The rigid format DMAP alter for SOL 63. LOOPID is used as a parameter t o choose the step at which the modal analysis is desired. For example.1 Analysis Procedure In nonlinear static analysis using SOL 106. when dynamic reduction is t o be applied in SOL 63. the nonlinear matrix generations for superelements will be skipped. the linear normal modes will be computed. RF63D89. n is specified where n is the desired LOOPID. this capability is built in SOL 106 with PARAM. the same scalar points should be defined in SOL 66. demonstrated in the Application Note of May 1981. NMLOOP. the value of KSTEP in the NLPARM entry should be set t o 1 with the AUTO. and (d) vibration of a nonlinear stepped beam. This rigid format alter can be used in Version 66. 10. The main advantage of using SOL 66 instead of SOL 64 is that both geometric and material nonlinearities can be included. In Version 67.1.2.1 for other input items required for SOL 63. However. is shown in Table 10. NMLOOP." dated January 1986.10. (b) vibration of a nonlinear material beam. The structure of the matrices used in SOL 66 and SOL 63 should be consistent. SEALL = n for the residual structure or SEALL = ALL should be used to initiate the nonlinear stiffness matrix generation. If the restart procedure (from SOL 66 t o SOL 63) is used with RF63D89. The stiffness matrix must be updated after the load step where the normal modes are desired. normal modes can be computed with the updated stiffness if PARAM.2. This option also requires the presence of the METHOD command in the Case Control section and EIGR or EIGRL entry in the Bulk Data section. SEMI or ITER method. See the examples and the comments in Table 10. Since the superelement analysis does not have the nonlinear capability. The parameter LGDISP needs t o be set to 1in both SOLS 66 and 63 to obtain the differential stiffness matrix t o account for the prestressed condition. In order t o force the update. The DMAP alter for Version 65 is listed in the Application Note. . data blocks (ESTNL and UGV) are fetched for the nonlinear stiffness from the SOL 66 run. SOL 63. (c) vibration of a plate.2 NONLINEAR MODAL ANALYSIS The modal analysis DMAP. If LOOPID is not specified in the SOL 63 run or ESTNL is not found in the data base. has been modified for nonlinear modal analysis with the nonlinear stiffness matrices obtained from the SOL 66 database. even though these points are not actually used. This capability is demonstrated by four problems: (a) vibration of a linear material beam.2. "Nonlinear Normal Mode Analysis.
2. first the theoretical natural frequencies of a simply supported beam without axial force are compared with those obtained from MSCINASTRAN.10. The Young's modulus is 2.0684 x l o 5 MPa and the mass density per unit length is 3.9167 x P = 1. The total length of the beam is 1000 mm and the width and height of its cross section is 100 mm and 50 mm. .9043 x lo2 lo3 lo3 lo3 x 6.2 Vibration of a Linear Material Beam Fig.9167 x Nsec2/mm2 (tonlmm).1589 x l o 4 The above table shows that MSCJNASTRAN values compare very well with the theoretical values.2.5079 x 8.3020 2.0 x l o 7 N is applied and the beam remains linear. the model is quite accurate.5883 x lo3 1. E = 2. An axial force of 1.0 x l o 7 N Figure 10.0385 x x lo4 1. To check the accuracy of the finite element beam model.9281 x lo3 6. The bending vibration frequencies will increase because of the axial force. The beam is modeled using five BEAM elements. Mode 1 w (Theory) w (MSC/NASTRAN) Mode 1st bending 2nd bending 3rd bending 1st longitudinal 4th bending 7. Therefore.1 Simply Supported Beam. 10. respectively.0684 x lo5 MPa m = 3.2.1 shows a simply supported beam under axial force that is used to test this capability. The theoretical values are obtained by Nsec2/mm2 EI The comparison is shown below: Beam Vibration with P = 0.1712 7.3208 2 3 4 5 x lo2 2.
0 x l o 7 N .001 .2.0684 x lo5 MPa.0453 1.0385 x lo3 8. Mode 1st bending 2nd bending 1st longitudinal 3rd bending 4th bending P = 1. The input data for this model are shown in Tables 10. 10.3189 x.2 Nonlinear StressStrain Relationship for Simply Supported Beam .3323 x l o 4 The above tables show that the axial load affects the lower modes more than the higher modes. .2. 10. 10.3.005 & . Mode 1 w (Theory) w (MSCJNASTRAN) 1. E = 2.3207 x lo3 5 1.2828 x lo3 x lo3 x lo3 x lo4 2 3 4 4. lo3 8. and with the nonlinear elastic stressstrain relationship shown in Fig.1) subjected t o an axial load of 1.2 and 10.2.2.2. x lo7 N.7481 x lo3 1.2.003 .The natural frequencies of the simply supported beam subjected t o constant axial force are compared with the theoretical values obtained by The comparison is shown below: Linear Beam with P = 1.7339 4.3 Vibration of a Nonlinear Material Beam This example calculates the natural frequencies of a simply supported beam (Fig.1292 8.011 Figure 10. x l o 7 N. .2.
6277 3. E = 5 .4 Plate Vibration This example demonstrates geometric nonlinear modal analysis for a rectangular plate. 10. 5.7593 8.4859 x lo3 x lo3 x lo3 x lo3 2 3 4  5. 10.0 x lo3 kg/m3. The plate has a thickness of 1. This is due t o the fact that the softening material effect is larger on the higher modes than the longitudinal tension. respectively. Due t o symmetry.Analysis results are compared with the theoretical solution as tabulated below.4 and 10. The theoretical values are calculated by using the tangent modulus at the final solution ( E = 0. and 6 on side AB and components 1.0 m.0 x lo5 P a normal t o its plane. The axisymmetric boundary condition is imposed by constraining components 2. x lo7 N .9099 x lo3 1.3. .5456 x lo3 9.031).2.2.7978 x lo3 6. x lo4 MPa. x lo7 N.0684 x lo5 MPa.2. Mode 1 w (Theory) w (MSC/NASTRAN) Mode 1st bending 2nd bending 1st longitudinal 3rd bending 4th bending 1.0180 x lo3 6. The clamped boundary condition is applied t o sides BC and CD by constraining all six DOFs.2.5717 5 x lo3 3. only one quarter of the plate is modeled using 50 QUAD4 elements as shown in Fig. and 6 on side AD. .0 x 10'' N/m2 and the mass density is 8. P = 1. Eo = 2. The plate is clamped on all sides and is subjected t o a pressure loading of 2. 4.0 m x 4.5 for nonlinear static analysis and modal analysis. . Nonlinear Beam with P = 1.0 cm and the dimension is 2. The Young's modulus is 2. The input data for this model are listed in Tables 10.6768 The frequencies of the lower modes have increased while those of the higher modes decreased.
Due t o Symmetry) At first.1118 9.2595 x lo1 1. The results are tabulated along with frequencies of the linear system without the pressure load.5 Vibration of a Stepped Beam This example demonstrates the geometric and material nonlinear modal analysis for a simply supported stepped beam.4842 x l o 2 I 5th bending 1 10.2. The input data for this model are listed in Tables 10. The stepped beam is modeled using five BEAM elements.6710 x l o 2 3. The Young's modulus is 2.5690 7.7648 5. dynamic reduction is used.0174 x lo2 x lo2 x lo2 x lo2 / 1st bending 2nd bending 3rd bending 4th bending I 1 9. for which elements 1 and 2 remain linear.2. 10. The material is nonlinear elastic with the properties shown in Fig.2. The total length of the beam is 1000 mm and its cross sectional areas vary from 120 mm to 40 mm.4.5 x lo6 N is per unit length is 3. which indicates the same tendency as in the beam problem. in Fig. In the SOL 63 restart.2. An axial force of 1.Figure 10.9167 x applied. 10.2. SOL 66 provides information about the nonlinear stiffness matrix.2.2677 x l o 2 4. The lower modes are affected more by the pressure load than the higher modes.7. Mode w (Linear) w (Nonlinear) Mode 1 1 2 3 4 5 1 1 9.3 Clamped Plate (Quarter Only.6 and 10. Angular Frequencies of Plate.6600 x l o 2 5.2.3634 x lo2 / 4. .0684 x lo5 MPa and the mass density Nsec2/mm2 (tonlmm).
1614 x lo4 1.2 and 10.9748 x 1.9167 x Nsec2/mm2 P = 1. have decreased.5 x l o 6 N.3).  Mode w (Linear) w (Nonlinear) Mode 1 1 1 1 2 3 4 5 1 1 1 1 1 x lo3 1.9544 x l o 4 1.0684 x l o 5 MPa m = 3.1089 x lo3 4. . The second column shows the nonlinear system with the axial load of 1. except the first.4. The bending vibration frequencies are tabulated below.2. E = 2.L = 1.2. P = 1.2437 x l o 4 1. .0684 x l o 5 MPa. This is because the softening material effect is larger than the longitudinal tension.0806 x 1.8154 x 1. Stepped Beam.5 x lo6 N Figure 10. This result is consistent with the previous analyses (Section 10.1142 x 5. The first column represents the linear system without the axial load.2.2179 x 1.0 x lo3 mm E = 2.0112 1 lo3 lo3 lo4 lo4 lo4 1 I1 / 1st bending / I1 2nd bending 3rd bending 1st longitudinal 4th bending ( / 1 / The nonlinear frequencies. Angular Frequencies of Stepped Beam.5 x lo6 N.
GO TO LINEAR STIFFNESS COND $DBFETCH /UGV./SOLID/LOOPID//DBSET3 $ DATA FROM SOL 66 RUN DBVIEW UGV66 =UGV (WHERE VERSION=RESTART AND SOLFETCH=LOOPID) $ +V66NX DBVIEW ESTNL66=ESTNL (WHERE VERSION=RESTART AND SOLFETCH=LOOPID) $ +V66NX EQUIV UGV66..LUSETS/S.NOKGXX/O/O/O/////////////K6ROT GPECT.GEOM2S..NLAYERS=5 $ GENERATE TABLE FOR LINEAR ELEMENT JMPKGG .BGPDTS./KELMNL.N.CSTMS.LOOPID=O/l $ READ LOOPID CARD PARAM LNSTIF.Y..UGV/ALWAYS/ESTNL66.CSTMS...CSTMS.K6ROT GPECT..CSTMS/KDLGG.DIT/ EST.N.CSTMS/KBJJZ $ ESTL.DESTNL.N.MPT.ECTS. LOCATION = DBUP LNSTIF./KELM.KDICT..NP/2/S.GPECT.DIT..Table 10.DIT.KELM.UGV.NOD=I/O/O//NP $ DIFFERENTIAL STIFFNESS FOR LINEAR ELEMENTS GPECT.N. NONLK $ NO ESTNL.KDLDI.KDLDI.KELMNL..ESTL/V.Y.KDICTNL./ S.ALTER FOR NORMAL MODES WITH NONLINEAR STIFFNESS $ $ JCL ATTACH DATABASE FROM SOL 66 COLD START $ $ $ EXECUTIVE DECK INPUT $ NO SPECIAL CARDS REQUIRED $ $ CASE CONTROL DECK INPUT $ SEALL = ALL $ SUPER = N $ $ BULK DATA DECK INPUT PARAM. GO TO LINEAR STIFFNESS COND TAI.2.KDLGG/KLTOT/$ ESTNL.KDICT.1 DMAP Alters for SOL 63 (Version 66) $ BEGINNING OF RF ALTER 63D89 $ + + + + + 22DEC1988 +++++++ $ $ RF63D89 .LGDISP.KDICTNL.SILS.KDLEL.N..NOGENL/SEID/S.DIT.MPT. COND EMG EMA EMG EMA ADD EMG EMA ADD MPT..BGPDTS..GEOM2S.ESTNL/ALWAYS $ +V66NX PARAML ESTNL//JPRES'////S.CSTMS.SILS.EPT.KBJ JZNL/KBI/ $ ./$ KLTOT .N.BGPDTS.Y./ $ STIFFNESS FOR NONLINEAR ELEMENTS I/O/O//////////////V.NONLK$ CHECK PRESENCE OF ESTNL LNSTIF.SELECT LARGE DISPLACEMENTEFFECTS (DEFAULT=I) $ $ PARAM.NOKGGX $ ESTL.LOOPID.NOLOOP/V.BGPDTS.N.Y./I/$ KB J JZ.LGDISP=I/ V.ETT... NOLOOP $ NO LOOPID.ESTNL.I SELECTS NONLINEAR STIFFNESS (DEFAULT=O) $ ALTER 483 $ JUST AFTER LABEL JMPK4GG $ ACTUALLY JUST AFTER OLD DBSTORE K4JJ $ NOTE KELM AND KDICT IF IN HERE NEVER STORED TYPE DB. ACON $ SKIP THE ALTERS FOR SUPERELEMENTS COND //lLT1/S.I.UGV $ PATH = SSNL.GEI.MPT./ $ STIFFNESS FOR LINEAR ELEMENTS S.N.SILS..SILS.NOESTL/ S.EDT/~LEL.ESTNL.ETT..CSTMS/KBJJZNL.
..CSTMS/KBDJJ.UGV...EDT/KDELM.KDDICT.MPT.CSTMS.KDICT.KBDJJ/KJJZ/$ JUMP JMPKGG $ SKIP LINEAR STIFFNESS GENERATION LABEL LNSTIF $ LABEL FOR LINEAR STIFFNESS GENERATION $ END OF RF ALTER 63D89 EMG .KELM.SILS..KDDICT.BGPDTS.ETT.DIT.ESTNL.KDELM./ $ ++V66NX ADD KBI./ I/O/O//NP/ $ DIFFERENTIAL STIFFNESS FOR NONLINEAR ELEMENTS EMA GPECT./I/$ DELETE /KJJZ.
0. 0. CBEAM 3 I 4 5 0.0417+6 PBEAM I $ MATERIAL PROPERTIES MAT1 I 2. 0. 0.2.+3 4. . V66 $ SHL 2/13/90 SOL 66 TIME 5 CEND $ GENERATE NONLINEAR STIFFNESS TERMS FOR SOL 63 RUN TITLE=EXTENSION OF SS BEAM SUBTITLE=FIVE BEAM ELEMENTS SET 5 = 5 SET 6 = 6 DISP = 6 STRESS = 5 SEALL = ALL SUBCASE I LOAD=IOO NLPARM=I0 OUTPUT(PLOT) CSCALE 1.3 PLOTTER NAST SET I ALL VIEW 0. YES . CBEAM 2 I 3 4 0. 0. CBEAM I I 2 3 0. GRID 5 800. GRID 2 200. GRID GRID 6 1000. GRID 3 400. . 0. 0. 0. X . I. CBEAM 4 5 6 0.Table 10. 0. I.83349 MATS1 1 101 NLELAST TABLES1 I01 PW 12346 246 246 246 246 2346 0. 0. $ CONNECTIVITY 1 1 2 0. Z FIND SCALE ORIGIN I SET I PTITLE=SIDE VIEW PLOT MODAL 0 SET I ORIGIN 1 SYMBOL 1 LABEL GRID BEGIN BULK $ PARAMETERS PARAM LGDISP I $ NLPARM 10 1 AUTO I $ GEOMETRY GRID I 0. 0. 0. 0. 0. AXES MY . 1. I. 4 600.3 7. 0. 0. 0.0684+5 0. ID BEAMWT. 1.1667+61. CBEAM 5 1 $ BEAM PROPERTIES I 5. 0.2 SOL 66 Input Data Listing for Uniform Nonlinear Beam CSR6347 $ JFC 13DEC1989 $ A0186H.
+TBI 0. 6 . .Oil $ LOADING FORCE 100 $ 0. 1000. +TB2 .003 500.005 700.+7 1.001 ENDT 206.84 . ENDDATA . +TB2 1.
CBEAM 5 $ BEAM PROPERTIES 1 5. 0. I. 246 0. I. 0. Z FIND SCALE ORIGIN I SET I PTITLE=SIDE VIEW PLOT MODAL 0 SET 1 ORIGIN 1 SYMBOL 1 LABEL GRID BEGIN BULK $ PARAMETERS PARAM LGDISP 1 PARAM LOOPID I PARAM COUPMASS1 $ GEOMETRY I 0. 0.+3 4.1667+61. 0. 0. 246 5 800. 0. 12346 GRID GRID 2 200. CBEAM 4 I 4 5 0. CSR6347R $ JFC 13DEC1989 DIAG 8 TIME 5 SOL 63 $ NORMAL MODES COMPILE SOL63. 0. 0. 0. CBEAM 2 CBEAM 3 I 3 4 0. I.3 SOL 63 Input Data Listing for Uniform Nonlinear Beam RESTART ID BEAMWT63.0417+6 PBEAM 1 $ MATERIAL PROPERTIES . 0.SOUIN=MSCSOU. 246 GRID 3 400. X . 0. 0. V66 $ SHL 2/13/90 $ A01861. 0. 1 2 3 0. I.LIST. I 5 6 0.3 PLOTTER NAST SET I ALL VIEW 0. I. 0. 246 GRID GRID 6 1000. 0. 0. 0. $ FILE EST=OVRWRT/GEI=OVRWRT/CPECT=OVRWRT/KELM=OVRWRT/ICTRWRT $ ALTER 483 $ FILE EST=OVRWRT/GEI=OVRWRT/GPECT=OVRWRT/KELM=~~R~RT/KDICT=OVRWRT $ RFALTER RF63D89 CEND TITLE=NONLINEAR VIBRATION OF SS BEAM SUBTITLE=FIVE BEAM ELEMENTS DISP = ALL SEALL = ALL METHOD = 10 OUTPUT(PLOT) CSCALE 1. AXES MY . GRID 4 600. .REF $ RFALTER RF63D89 IN V66 AND V66A NEEDS THE FOLLOWING LINE INSERTED IN $ THE ALTER.2. .Table 10. 2346 $ CONNECTIVITY CBEAM 1 1 1 2 0. 0.
3 4000.MAT1 I 2. 7.83349 8 8 ENDDATA .0684+5 $ EIGENVALUE EXTRACTION EIGR 10 SINV 0. $ 0.
. 0. 0. 0. FIND SCALE ORIGIN I SET I PTITLE=ISOMETRIC VIEW PLOT MODAL 0 SET 1 ORIGIN 1 SYMBOL I VIEW 0. NOTE. . V66 $ SHL 2/13/90 $ MSC. Y . SOL 66 Input Data Listing for Plate NASTRAN MESH $ PREFORT=2 ID PLATEWP. I PW PW YES YES 1. 2. APP. $ NLPARM 10 2 2 NLPARM 20 $ VERTEX GRID POINTS EGRID 1 EGRID 2 EGRID 3 EGRID 4 AUTO AUTO 2. . I.4. AXES Y . MAY 1981 SEALL = ALL DISP = ALL STRESS = ALL = 10 SPC SUBCASE I LOAD=I NLPARM=10 SUBCASE 2 LOAD=2 NLPARM=20 OUTPUT (PLOT) CSCALE 1.. Z FIND SCALE ORIGIN I SET I PTITLE=SIDE VIEW PLOT MODAL 0 SET I ORIGIN I SYMBOL I LABEL GRID AXES X . X .3 PLOTTER NAST SET I ALL VIEW 30. Z FIND SCALE ORIGIN I SET I PTITLE=FRONT VIEW PLOT MODAL 0 SET I ORIGIN I SYMBOL I LABEL GRID BEGIN BULK $ PARAMETERS PARAM LGDISP I PARAM K6ROT 100.AOI86J $ KOK 11/85 SOL 66 TIME 5 CEND $ GENERATE NONLINEAR STIFFNESS MATRIX FOR SOL 63 RESTART TITLE= GEOMETRIC NONLINEAR ANALYSIS OF A FLAT PLATE SUBTITLE= PRESSURE LOAD WITH CLAMPED EDGES LABEL= REF. . 20.2.Table 10.
GRIDGI THRU THRU 250 250 20012 $ ENDDATA .+5 $ THESE POINTS ARE FOR SOL 63 SPOINT 20001 THRU 20012 SPCl 3 0 20001 THRU .+5 PLOAD4 2 201 2.+11 $ CONSTRAINTS SPCG 2 I 156 1 246 SPCG 2 I 123456 SPCG I 1 2 SPCADD 10 $ APPLIED LOADS PLOAD4 I 201 1.01 PSHELL 200 MAT1 I 2.$ GENERATE GRID POINT ARRAY GRIDG 1 +RIDGI 5 4 $ GENERATE QUAD4 ELEMENTS CGEN QUAD4 201 200 $ PROPERTIES I .
LIST. EGRID 2 I. . Y . EGRID 3 2.2.REF ALTER 483 $ FILE EST=OVRWRT/GEI=OVRWRT/GPECT=OVRWRT/KELM=OVRWRT/KDICTOVRWRT RFALTER RF63D89 CEND TITLE=SYMMETRIC MODES SUBTITLE=PRELOAD OF UNIFORM PRESSURECLAMPED EDGES LABEL = REF..Table 10. 20. EGRID 4 I. . X .SOUIN=MSCSOU. 0. $ VERTEX GRID POINTS EGRID 1 2.. AXES Y .3 PLOTTER NAST SET I ALL VIEW 30. $ GENERATE GRID POINT ARRAY 1 2 3 10 GRIDG I $ GRIDGI . FIND SCALE ORIGIN I SET I PTITLE=ISOMETRIC VIEW PLOT MODAL 0 SET I ORIGIN I SYMBOL 1 VIEW 0. 0. V66 $ SHL 2/13/90 $ A0581B $ JAJ 8 DEC 82 TIME 5 SOL 63 COMPILE SOL63. 0. NOTE OF MAY 1981 SEALL = ALL = 15 SPC SET 1 = 0 DISP = ALL METHOD = I DYNRED = 1 OUTPUT(PLOT) CSCALE 1. Z FIND SCALE ORIGIN I SET I PTITLE=SIDE VIEW PLOT MODAL 0 SET I ORIGIN I SYMBOL I LABEL GRID AXES X .5 SOL 63 Input Data Listing for Plate NASTRAN MESH $ PREFORT=2 RESTART ID PLATEWP63. APP. Z FIND SCALE ORIGIN 1 SET 1 PTITLE=FRONT VIEW PLOT MODAL 0 SET I ORIGIN I SYMBOL I LABEL GRID BEGIN BULK $ PARAMETERS PARAM LOOPID 4 PARAM LGDISP I PARAM K6ROT 100.
DYNRED 1 EIGR 1 MGIV 0.+11 $ CONSTRAINTS SPCG 2 1 156 2 i 246 SPCG 1 1 123456 SPCG $ SYMMETRIC BOUNDARY CONDITION SPCADD 15 1 2 $ DATA FOR GENERALIZED DYNAMIC SPOINT 20001 THRU 20012 160. +IGRI MASS ASETl 0 20001 THRU qSETi 0 20001 THRU $ 1 1 . 20012 20012 EIGRi ENDDATA . AD AB BC CD FOR SOL 63 REDUCTION AND EIGENVALUE EXTRACTION 8  SOL 63 160.tRIDG1 5 4 $ GENERATE QUAD4 ELEMENTS QUAD4 201 200 CGEN $ PROPERTIES PSHELL 200 1 .3 8000.01 MAT1 1 2.
133+6 7. 0. .44+4 PBEAM 11 PBEAM 12 I 1. 0.080+6 2.60+3 $ MATERIAL PROPERTIES MAT1 1 2.00+4 PBEAM 13 1 6.133+6 0. V66 $ SHL 1/13/90 $ A0186L KOK 12/85 TIME 5 SOL 66 CEND $ GENERATE NONLINEAR STIFFNESS TERMS FOR SOL 63 RUN TITLE=EXTENSION OF SS STEPPED BEAM SUBTITLE=FIVE BEAM ELEMENTS DISP = A L L STRESS = ALL SEALL = ALL SUBCASE I LOAD=IOO NLPARM=10 OUTPUT(PLOT) CSCALE 1.080+6 2.60+3 PBEAM 15 I 1. 0. 0. SOL 66 Input Data Listing for Stepped Beam ID SBEAMWT.333+6 3. 1.413+6 1. 0.728+7 8. . GRID 5 800. GRID 6 1000. 0.83349 I.413+6 1.3 PLOTTER NAST SET I ALL VIEW 0. GRID 4 600. 0. 0. 0. X .6. 1.40+3 PBEAM 14 1 3. I. Z FIND SCALE ORIGIN 1 SET 1 PTITLE=SIDE VIEW PLOT MODAL 0 SET I ORIGIN I SYMBOL I LABEL GRID BEGIN BULK $ PARAMETERS PARAM LGDISP 1 $ NLPARM 10 I $ GEOMETRY 1 0. $ CONNECTIVITY II I CBEAM I CBEAM 2 12 2 CBEAM 3 13 3 CBEAM 4 14 4 CBEAM 5 15 5 $ BEAM PROPERTIES I 1.0684+5 AUTO 0. 0. PW YES 2 3 4 5 6 1. 0.333+6 3.3 I. 0. GRID 3 400. 0.Table 10. 0. AXES MY .2. 1. 0. 1 0.728+7 8. 0. GRID GRID 2 200. 0. .
ENDDATA .001 ENDT 206. 1000.84 . +TB2 .003 500. +TB1 +TB2 1.Oil $ LOADING FORCE 100 $ 101 0.5+6 I.MATS1 I TABLES1 101 +TBI 0.005 700. . 6 NLELAST .
0.60+3 1. GRID 6 1000. X . GRID 5 800. CBEAM 5 15 5 6 0. GRID 4 600.413+6 3.080+6 1. .133+6 2. GRID 3 400. V66 $ SHL 2/13/90 $ A0186M $ KOK 12/85 DIAG 8 TIME 5 SOL 63 COMPILE SOL63.2. SOL 63 Input Data Listing for Stepped Beam RESTART ID SBEAMWT63.REF ALTER 483 $ FILE EST=OVRWRT/GEI=OVRWRT/GPECT=OVRWRT/KELM=OVRWRT/KDIC=OVR RFALTER RF63D89 CEND TITLE=VIBRATION OF SS STEPPED BEAM SUBTITLE=FIVE BEAM ELEMENTS SEALL = ALL DISP = A L L METHOD = 10 OUTPUT(PLOT) CSCALE 1.44+4 1.00+4 8. CBEAM 3 13 3 4 0. 0.728+7 PBEAM I1 PBEAM 12 I 1. .3 PLOTTER NAST SET I ALL VIEW 0. 0. $ BEAM PROPERTIES I 1.333+6 PBEAM 13 I 6. 0. 0. 0. 0.60+3 2. Z FIND SCALE ORIGIN I SET I PTITLE=SIDE VIEW PLOT MODAL 0 SET I ORIGIN I SYMBOL 1 LABEL GRID BEGIN BULK $ PARAMETERS PARAM LOOPID 1 PARAM LGDISP I PARAM COUPMASSI $ GEOMETRY I 0.080+6 PBEAM 15 I 1. $ CONNECTIVITY 11 1 2 0. 0. 0. AXES MY . 0. 0.40+3 3. 0. 0.333+6 8.413+6 PBEAM 14 I 3.133+6 $ .728+7 1.7. CBEAM I CBEAM 2 12 2 3 0.Table 10. CBEAM 4 14 4 5 0. 0. GRID GRID 2 200.SOUIN=MSCSOUYLIST.
7. $ 0.3 4000.0684+5 $ EIGENVALUE EXTRACTION EIGR 10 SINV 0.83349 8 8 ENDDATA .$ MATERIAL PROPERTIES MAT1 1 2.
The MAXITER and MAXLS fields should be assigned a larger value than the defaults (e. The TSTATIC option was implemented in Version 66 when the bisection was not available in SOL 66. references static load entries (i. material nonlinearity is combined with geometric nonlinear effects in the analysis of a Zshaped beam with a static load.) and provides static t o dy r r . the automatic time stepping algorithm is not applicable. and illustrates SOL 99 static and dynamic analysis using the same input data.3.3 STATIC ANALYSIS USING SOL 99 Static analysis can be performed using SOL 99 (or 129). thus allowing a static and dynamic analysis in a single run from the same input data. Since the effects of inertia and damping forces are ignored in the static analysis. However.1 Input Data for SOL 99 Static Analysis The important input data for static analysis using SOL 99 are described below: r The parameter TSTATIC with a value of 1 must be included in the Case Control or Bulk Data section in order t o activate the static analysis. MPC) after the first subcase.e. etc. selected by LOADSET in the Case Control section. the nonlinear transient solution sequence. The solution strategy is controlled by the TSTEPNL Bulk Data entry..g. it lacks some features provided by SOL 66. MOMENT.. which is the general strategy adopted in the NLPARM for static analysis. FORCE. 25 and 4) in order t o allow more iterations and line searches. SOL 99 static capability is confined to what SOL 99 can perform and abides by its limitations. in which the ADAPT method (NLTRD2 module) must be specified for static analysis. The value of TSTATIC can be varied from subcase t o subcase. Restrictions on the SOL 99 input data include the following: No constraint changes (SPC. In the first example. including restarts r r No enforced displacements (SPCD) allowed No CREEP or thermal loads allowed The SOL 99 static analysis is compared with SOL 66 for two example problems. the bisection algorithm is active with the adaptive stiffness matrix update strategy and other expedient iteration schemes such as the BFGS updates and the line search process. 10. The bisection algorithm is implemented in SOL 66 (or 106) in Version 67.10. PLOAD. A minimum of 4 time steps is required in the NDT field for each subcase. TSTATIC with a value of 1 (default) activates the dynamic analysis. The second example combines frictional contact with geometric nonlinear effects in the rotation of a beam with a GAP element. The Bulk Data LSEQ. While the static capability in SOL 99 provides convenience for transient response analysis of a preloaded structure.
3. The problem is geometrically nonlinear. The default value for dynamic analysis is 5. shown in Fig. E v = 68 m m 2 . Poisson's ratio a. MAXLP with a value of 10 should be included in the input d a t a t o conform with the static default. The parameter MAXLP determines the maximum number of internal subincrements in the plasticity process.3. The static loading history should be defined following the same convention using TLOADi. moment of inertia = 2 x lo5 n'/rnm2.e. i. The timedependent function for the dynamic load is specified by a TLOADi entry which is defined by a tabular function via TABLED input or by an analytic function.. A I.1. One advantage in SOL 99 static analysis is that the loading path does not have t o be increased linearly within a subcase as in SOL 66.5067 m m 4 .3.4].1 ZShaped Cantilever B e a m . 10. involving almost 90 degree rotation. The material is elasticplastic obeying the von Mises yield criterion and the isotropic hardening rule. The Zshaped beam is loaded statically with a vertical force of 2000 N a t one end and clamped a t the other end. Young's modulus = 0.3. but the follower force effect is absent. crosssectional a r e a = 65.2 A ZShaped Beam Loaded with a Static Force A Zshaped cantilever beam. yield stress Figure 10. If a plastic material is used in the finite element model. 10.. = 450 ~ / r n r n ' . time implies the static load step in the subcase with TSTATIC=l. LSEQ also selects DAREA which is internally generated t o define the points of dynamic load application and is referenced by TLOADi. was proposed as a benchmark problem for testing nonlinear finite element analysis codes [10.namic load conversion.
Smaller elements are used near the fixed end because of the large plastic deformation anticipated in this area. The NLPARM and TSTEPNL Bulk Data entries provide the solution strategy in SOL 66 and SOL 99. LSEQ. The numerical performance of each subcase. Figure 10. The clamped boundary condition at the end of the beam is imposed by constraining all six components of GRID 1.3. Material properties are given by the MAT1. Each subcase specifies 10 increments.3.The beam is modeled by ten BEAM elements of various lengths as seen in Fig. and TABLES1 Bulk Data entries representing the elastoplastic material as shown in Fig.3. TLOAD1. is tabulated below. .3. Fig. is divided into 20 subcases. MATS1. which is identical for SOL 66 and SOL 99. 10. respectively.2. comparing SOL 66 and SOL 99. The load applied at the free end of the beam is specified by FORCE and LOAD Bulk Data entries in SOL 66 and by FORCE. The loaddeflection curve at the free end is shown in Fig.4. except for subcase 2 and 3 which contain 25 increments to cope with the severe rotation and plastic yielding in this domain.3. 10.2. and TABLED1 Bulk Data entries in SOL 99.1 and 10. The solution produced by SOL 66 and SOL 99 was identical. The input data are given in Table 10.3 shows the deformed shapes of the beam at progressively increasing loading stages.3.3.1. 10.2 Nonlinear StressStrain Relationship for ZShaped Beam The analysis is challenging because the substantial loading results in very large rotation and severe plastic deformation at the fixed end. The loading history. 10.
of No. of SOL 99 No. of No.Performance Comparison for 2Shaped Beam SOL 66 Subcase No. of ID Iter K Update Bis Iter K Update Bis 18 19 20 26 21 20 1495 1 1 0 0 0 0 32 26 24 2880 1 1 1 0 0 0 TOTAL 179 23 74 59 . of No. of No.
3 Deformed Shapes of ZShaped Beam Displacement (mrn) Figure 10.4 LoadDeflection Curve for ZShaped Beam .3.Figure 10.3.
SHL 9MAY85.1 SOL 66 Input Data Listing for 2Shaped Beam ID ZBEAM. 50 $ PRINT MATRIX TRAILERS AND ITERATIONS CEND TITLE = ZBEAM ANALYSIS WITH 10 BEAM ELEMENTS SUBTITLE = COLOGNE CHALLENGE. CYP 19OCT87 TIME 30 $ CPU MINUTES FOR VAX $ NONLINEAR STATIC ANALYSIS SOL 66 DIAG 8.Table 10. V67 $ KOK 5APR85.3. 1984 SEALL = ALL DISP = ALL SUBCASE I LOAD = 10 NLPARM = 10 SUBCASE 2 = 20 LOAD NLPARM = 25 SUBCASE 3 LOAD = 30 NLPARM = 25 SUBCASE 4 = 40 LOAD NLPARM = 10 SUBCASE 5 LOAD = 50 NLPARM = 10 SUBCASE 6 = 60 LOAD NLPARM = 10 SUBCASE 7 LOAD = 70 NLPARM = 10 SUBCASE 8 = 80 LOAD NLPARM = 10 SUBCASE 9 LOAD = 90 NLPARM = 10 SUBCASE 10 LOAD = 100 NLPARM = 10 SUBCASE I1 LOAD = 110 NLPARM = 10 SUBCASE 12 LOAD = 120 NLPARM = 10 SUBCASE 13 LOAD = 130 NLPARM = 10 . AUG.
GRID 3 20. AXES Z... GRID 5 60. 0. SET 1 = ALL VIEW O. 0. 0. O. X. 0. 0. 0.SUBCASE 14 LOAD = 140 NLPARM = 10 SUBCASE 15 LOAD = 150 NLPARM = 10 SUBCASE 16 LOAD = 160 NLPARM = 10 SUBCASE 17 LOAD = 170 NLPARM = 10 SUBCASE 18 LOAD = 180 NLPARM = 10 SUBCASE 19 LOAD = 190 NLPARM = 10 SUBCASE 20 LOAD = 200 NLPARM = 10 OUTPUT(PLOT) CSCALE 1. Y PTITLE = TOP VIEW FIND SCALE ORIGIN 1 SET I PLOT STATIC 0 MAXI DEFO 1.I1 BEGIN BULK PARAM LGDISP 1 $$ SOLUTION CONTROL 10 AUTO 7 NLPARM 10 NLPARM 25 25 AUTO 7 $$ GEOMETRY I 0. 0. 0.3 PLOTTER NAST XTITLE = FORCE (N) YTITLE = TIP DISPLACEMENT (MM) AT NODES 5. GRID 4 40. 8. 0. GRID GRID 2 10. GRID 6 80.8(~2) (T2) . 0. 10. . 0. 11 XYPLOT = DISP RESP /5 (T2) . SET I ORIGIN 1 OUTPUT(XYPLOT) CSCALE 1.3 PLOTTER NAST MAXI DEFO 1. 0.
LOAD 60 LOAD 70 I. 8. LOAD 40 I. 5. ENDDATA 68. LOAD 160 LOAD 170 I. I. I. LOAD 100 I. LOAD 110 I. LOAD 20 I. LOAD 120 1. LOAD 130 I. LOAD 200 I.5067 2266. +TBSI +TBS2 2. LOAD 140 I. 0.8252 580.GRID 7 GRID 8 GRID 9 GRID 10 11 GRID GRID 99 $$ CONNECTIVITY 100 CBEAM 100 CBEAM 110 100 100 CBEAM 120 CBEAM 130 100 CBEAM 140 100 CBEAM 150 100 CBEAM 160 100 CBEAM 170 100 CBEAM 180 100 100 CBEAM 190 PLOTEL 200 I $$ PROPERTIES PBEAM 100 110 MAT1 110 2. +TBS2 ENDT $$ LOADING 1000 I1 FORCE LOAD 10 I.+5 MATS1 110 111 TABLES1 Ill +TBS1 0.667 . LOAD 150 I. LOAD 190 I. LOAD 30 I.253 450. LOAD 80 I.3 450. LOAD 50 I. PLASTIC 65. . LOAD 180 I.253 570. LOAD 90 1.
3. SHL 9MAY85. 1984 SEALL = ALL DISP = ALL $$ $ $ LOADSET = 1000 $ SUBCASE I DLOAD = TSTEPNL = SUBCASE 2 DLOAD = TSTEPNL = SUBCASE 3 DLOAD = TSTEPNL = SUBCASE 4 DLOAD = TSTEPNL = SUBCASE 5 DLOAD = TSTEPNL = SUBCASE 6 DLOAD = TSTEPNL = SUBCASE 7 DLOAD = TSTEPNL = SUBCASE 8 DLOAD = TSTEPNL = SUBCASE 9 DLOAD = TSTEPNL = SUBCASE 10 DLOAD = TSTEPNL = SUBCASE I 1 DLOAD = TSTEPNL = SUBCASE 12 DLOAD = TSTEPNL = 100 10 100 25 100 25 100 10 100 10 100 10 100 10 100 10 100 10 100 10 100 10 100 10 .2 SOL 99 Input Data Listing for 2Shaped Beam $ SSH 3DEC87.Table 10. CYP 19OCT87 ID ZBEAMD. AUG.V67 TIME 30 $ CPU MINUTES FOR VAX $ NONLINEAR TRANSIENT ANALYSIS SOL 99 DIAG 8.50 $ PRINT MATRIX TRAILERS AND ITERATIONS CEND TITLE = 2BEAM ANALYSIS WITH 10 BEAM ELEMENTS SUBTITLE = COLOGNE CHALLENGE.
l ( 2 8T) lT) BEGIN BULK $$ $ PARAM TSTATIC 1 $ PARAM LGDISP 1 PARAM MAXLP 10 $$ SOLUTION CONTROL 10 TSTEPNL 10 +TSI TSTEPNL 25 25 .. ( 2 . 8.3 PLOTTER NAST XTITLE = FORCE (N) YTITLE = TIP DISPLACEMENT (MM) AT NODES 5. SET I ORIGIN I OUTPUT(XYPLOT) CSCALE 1..SUBCASE 13 DLOAD = 100 TSTEPNL = 10 SUBCASE 14 DLOAD = 100 TSTEPNL = 10 SUBCASE 15 DLOAD = 100 TSTEPNL = 10 SUBCASE 16 DLOAD = 100 TSTEPNL = 10 SUBCASE 17 DLOAD = 100 TSTEPNL = 10 SUBCASE 18 DLOAD = 100 TSTEPNL = 10 SUBCASE 19 DLOAD = 100 TSTEPNL = 10 SUBCASE 20 DLOAD = 100 TSTEPNL = 10 OUTPUT(PLOT) CSCALE 1. AXES Z. O.3 PLOTTER NAST MAXI DEFO I. X.04 ADAPT 25 4 ADAPT 25 .I . 0. Y FIND SCALE ORIGIN 1 SET 1 PLOT TRANS 0 MAXI DEFO 1. SET I = ALL VIEW O. I1 XYPLOT DISP RESP /5(T2) .
60. 2 3 4 5 6 7 8 9 10 11 0. 0. 0. 8. 80. I. 200 20. 0 1.667 . 0. 0. 10. 0.8252 580. 200. 10. 0. 0. 20. 30. 110 2000. 40.+5 MATS1 110 II I TABLES1 I l l +TBSI 0. ' 65. 0. 0. 0. 180. 0. 0.253 $ 910 1I 910 0. 450. 5. +TBSl +TBS2 $ PLASTIC 2. 30. 0.05 1.3 450.253 570. 0. +TBDI . 0.5067 2266. 20. 100. 0. 0. $$ 0. 140. 99 99 99 99 99 99 99 99 99 99 2 3 4 5 6 7 8 9 10 99 68. 30. 1 0.+TS2 $$ G O E R E MTY GRID I GRID 2 GRID 3 GRID 4 GRID 5 GRID 6 GRID 7 GRID 8 GRID 9 GRID 10 GRID 11 GRID 99 $$ CONNECTIVITY 100 CEM BA 100 CEM BA 110 100 100 CEM BA 120 CEM BA 130 100 CE M BA 140 100 CEM BA 150 100 CBEAM 160 100 CEM BA 170 100 CEM BA 180 100 CE M BA 190 100 PLOTEL 200 1 $$ PROPERTIES PEM BA 100 110 MT A1 110 2. ED NT $ $ E D AA NDT . +TBS2 ED NT LOADING LSEQ 1000 FORCE 110 TLOADI 100 TABLED1 200 +TBD1 0. 30. 0. 200. 0. 120.
) is l o 4 lb/in. The coefficient of static friction (p.3. The solution strategy is provided by the NLPARM Bulk Data entry in SOL 66 and by the TSTEPNL Bulk Data entry in SOL 99. The second subcase statically loads the beam with a 1000 inlb moment applied in 10 increments. One end of the beam is clamped and the other end is attached t o a GAP element as shown in Fig.3. The deformed shapes of the beam during static loading are identical for SOL 66 and SOL 99 and are shown in Fig. the beam is statically loaded by a 1000 inlb moment which is applied in the +z direction.5.) is 0. Friction at one end of the beam is simulated by a GAP element. respectively. The deformed shapes are labeled from 1 t o 6. The unloading phase was analyzed statically (SOL 66) and dynamically (SOL 99).8' at the end.0 x lo3 lb/in (default value). corresponding to . while in SOL 99 the Bulk Data entries LSEQ.15 in2 E = 20 x l o 6 lb/in2 I = 2.10. connected a t GRID 112 and 212. and the transverse shear stiffness ( l i t ) is 5. and the beam responses were observed. In the second phase. TLOADl and TABLED1 are used in addition t o FORCE and MOMENT entries. In the first phase. resulting in a total rotation of 115. A force of 50 lb acting in the z direction is applied as a preload providing a friction force at the end attached to the GAP.15 in2. 10.3 Rotation of a Beam with Friction A uniform beam has a length of 12 inches and a crosssectional area of 0.813 x in4 P = 50 lb M = 1000 inlb Figure 10.3.5.5 Uniform Beam with a G a p The beam is modeled by twelve BEAM elements of equal length with linear elastic material. MOMENT and LOAD Bulk Data entries in SOL 66. The applied load (at GRID 112) is specified using FORCE.6.TSTATIC. Notice that in SOL 99 the first and second subcase use PARAM. / ix I l l l I I l t I i" . The first subcase applies the preload (force of 50 lb) in 4 increments.3. the moment is removed while the applied force remains. which is initially closed. The clamped boundary condition at the end of the beam is imposed by constraining all six components of GRID 100.3 and 10. the coefficient of kinetic friction ( p k ) is 0. the closed axial stiffness (li.4.3. The same loading history is specified in the first and second subcase of the SOL 66 and SOL 99 analysis. 10.3. The analysis is performed in three subcases.3. The input data for SOL 66 and SOL 99 are given in Table 10.l for the static analysis. DLOAD. 4 1' L = 12 in A = 0.
The third subcase in SOL 99 uses PARAM. The numerical performance of the second subcase in terms of the number of iterations. the end point starts to move as evidenced by the subsequent deformed shapes in the figure.of No. of bisections iterations K updates I / SOL 66. .7. The first deformed shape does not exhibit translation of the end point because the lateral force induced by the applied moment is less than the friction force. 10. In SOL 66 the third subcase statically unloads the beam by removing the applied moment while retaining the 50 lb preload. which have significant effects on the efficiency of the iterative process. V67 246 66 63 1533 2 25 The table shows a variation in the efficiency of the two solutions. Different stiffness update strategies are provided in SOL 66 and SOL 99. the rotation of the end of the beam about the zaxis. and bisections is tabulated below for SOL 66 and SOL 99. Notice that the bisection algorithm was implemented in SOL 66 in Version 67 and was activated in the SOL 66 analysis. maintaining the preload on the GAP element. V67 SOL 99. allowing numerous intermediate stiffness updates as required for convergence based on the global divergence criteria in addition t o the GAP induced updates.of No.8' and diminishes as the beam vibrates due to friction.1 for dynamic unloading of the beam after the moment is released.TSTATIC. The angle (Bend)starts from 115. The resulting deformed shapes are shown in Fig. 10. stiffness updates.3. When the induced lateral force exceeds the frictional force.8 shows the rotation of the end of the beam in radians about the zaxis during dynamic unloading. This signifies the effects of timely stiffness updates on the efficiency of the iterative process. SOL 66 has a more liberal stiffness update strategy. Performance Comparison During Static Loading (Subcase 2) No. Fig.the tabulated applied moment and Bend.3. SOL 99 limits the number of stiffness updates within an increment and updates in reference to the last converged step except for GAP induced update.
8' Figure 10.1' 17.8" 104.1' 55.3' 80.2' Figure 10.curve moment (inlb) 1 2 3 4 5 6 Oend 100 200 400 600 800 1000 3.7 Deformed Shapes of Beam During Static Unloading (SOL 66) .3.1' 115.6' 1000 800 600 400 200 0 7.3~ 41.8' 66.4' 91.3.9' 31.6 Deformed Shapes of Beam During Static Loading (SOL 66 and SOL 99) curve moment (inlb) 1 2 3 4 5 6 Oend 115.
Wl T H I S 1 0 0 0 I NLB MOMENT A GAP AND A P P L l ED 2 .3. 0 4 T I ME F l XED/ F R E E BEAM.2 .8 Rotation (0. 0 6 I N SEC AT F R E E END MOMENT Figure 10.) at End of Beam During Dynamic Unloading .
0.001 SET I = ALL EXCEPT GAP SET 2 = ALL EXCEPT PLOTEL. EXCLUDE GRID POINTS 1000 THRU 1003 VIEW O. 0. 0. GRID 104 4. GRID 105 5. . 0. 0.Table 10. 0. 0.3 PLOTTER NAST MAXI DEFO . O. 0. GRID 103 3. 0. GRID 102 2.001 SET 1 ORIGIN I BEGIN BULK PARAM LGDISP I $$ SOLUTION CONTROL YES NLPARM 10 4 AUTO YES 10 AUTO NLPARM 20 YES 10 AUTO NLPARM 30 $$ GEOMETRY 100 0.3. 0..50 $ PRINT MATRIX TRAILERS AND ITERATIONS CEND TITLE = FIXED/FREEBEAM. GRID GRID 101 I. V67 $ APR 15AUG90 TIME 30 $ CPU MINUTES 66 $ NONLINEAR STATIC ANALYSIS SOL DIAG 8. Y FIND SCALE ORIGIN I SET I PTITLE = TOP VIEW PLOT STATIC 0 MAXI DEFO .3 SOL 66 Input Data Listing for Beam with Friction ID FRIBEAM66. 0. AXES MZ . X . 0. 0.. WITH A GAP AND APPLIED MOMENT AT FREE END SUBTITLE = MOMENT IS 1000 INLB ECHO = BOTH SEALL = ALL SET 1 = 101 THRU 112 SET 2 = 100 SET 3 = 200 DISP = I SPCF = 2 STRES = 3 SUBCASE I $ FORCE = 50 LB NLPARM = 10 LOAD = I 0 1 SUBCASE 2 $ MOMENT = 1000 INLB NLPARM = 20 LOAD = 2 0 1 SUBCASE 3 $ UNLOADING NLPARM = 30 LOAD = I 0 1 OUTPUT(PLOT) CSCALE 1.
0. 0. FORCE 100 112 MOMENT 200 112 $$ FRAME FOR PLOT GRID 1000 GRID 1001 GRID 1002 GRID 1003 PLOTEL 200 1000 PLOTEL 201 I001 PLOTEL 202 1002 PLOTEL 203 1003 $ I01 102 103 104 105 106 107 108 109 110 Ill 112 112 0. 0. I.E+7 PGAP 20 0. 0. 0. 0. ENDDATA .GRID 106 GRID 107 GRID 108 GRID 109 GRID 110 GRID Ill GRID 112 GRID 212 $$ CONNECTIVITY CBEAM I 10 CBEAM 2 10 CBEAM 3 10 CBEAM 4 10 CBEAM 5 10 CBEAM 6 10 CBEAM 7 10 CBEAM 8 10 CBEAM 9 10 CBEAM 10 10 CBEAM 11 10 CBEAM 12 10 CGAP 200 20 $$ PROPERTIES PBEAM 10 II MAT1 I1 2. 0. 0. LOAD 201 I. 0. 0.02 $$ LOADING LOAD 101 I. +PGI . 0.
3 PLOTTER NAST MAXI DEFO . O... Y FIND SCALE ORIGIN I SET I PTITLE = TOP VIEW PLOT TRANS 0 TIME 0.0 MAXI DEFO .0 2.50 $ PRINT MATRIX TRAILERS AND ITERATIONS CEND TITLE = FIXED/FREE BEAM. EXCLUDE GRID POINTS 1000 THRU 1003 VIEW O. X . 0. WITH A GAP AND APPLIED MOMENT AT FREE END SUBTITLE = MOMENT IS 1000 INLB ECHO = BOTH SEALL = ALL SET I = 101 THRU 112 SET 2 = 100 SET 3 = 200 DISP = I SPCF = 2 STRES = 3 $ $ $ $ LOADSET = 1000 $ $ SUBCASE I $$ FORCE = 50 LB $ $ PARAM TSTATIC I $ TSTEPNL = 10 DLOAD = 101 SUBCASE 2 $ MOMENT = 1000 INLB $$$ $ PARAM TSTATIC I $ TSTEPNL = 20 DLOAD = 201 SUBCASE 3 $ DYNAMIC UNLOADING $$$ $ PARAM TSTATIC 1 $ TSTEPNL = 30 DLOAD = 101 OUTPUT(PLOT) CSCALE 1.001 SET 1 ORIGIN I OUTPUT(XYPLOT) .3.Table 10.001 SET 1 = ALL EXCEPT GAP SET 2 = ALL EXCEPT PLOTEL. AXES MZ .4 SOL 99 Input Data Listing for Beam with Friction ID FRIBEAM99. V67 $ APR 15AUG90 TIME 90 $ CPU MINUTES $ NONLINEAR STATIC ANALYSIS SOL 99 DIAG 8.
0. GRID 102 2. GRID Ill 11. 0. 10 III 112 0. GRID 108 8. CBEAM 3 CBEAM 4 10 103 104 0. 0.0 XGRID LINES = YES YGRID LINES = YES XTITLE = TIME IN SEC YTITLE = DISP AT GRID 112 ALONG Y XYPLOT DISP REsP/II~(T~) YTITLE = DISP AT GRID 112 ABOUT Z XYPLOT DISP RESP/II2(R3) BEGIN BULK PARAM LGDISP I $$ SOLUTION CONTROL TSTEPNL 10 4 .25 ADAPT +TSI 4 10 . $$ PROPERTIES PBEAM 10 II .3 PLOTTER NAST XMIN = 2. 0. CBEAM 9 10 108 109 0. 0. CBEAM 8 10 107 108 0.01 $$ CONNECTIVITY CBEAM I 10 100 I01 0. 0. GRID 103 3. 0. . GRID 110 10. 0. CBEAM 2 10 101 102 0. 0. 10 102 103 0. CBEAM 11 10 110 Ill 0. 0. 0.8134 . 0. 0. 0. GRID 112 12. 0. 0. 0. 0. GRID 106 6. 0.CSCALE 1. 0.002 ADAPT TSTEPNL 30 50 $$ GEOMETRY GRID 100 0. 0. 0. 0. 10 105 106 0.8134 2.I5 2. 0. 0. GRID 107 7. GRID 212 12. 0. 0. 0. CBEAM 5 10 104 105 0. 0. 0. 0. 0. 0. CBEAM 6 CBEAM 7 10 106 107 0. 0. 0. 0. GRID 105 5. 0. GRID 104 4. CBEAM 10 10 109 II0 0. 0. 0. GRID 109 9. CBEAM 12 CGAP 200 20 212 112 I. GRID 101 I. 0.I ADAPT TSTEPNL 20 +TS2 4 .
. $$ 50. 211 2. 0. 3 7. I. 1001 1002 1003 1000 2. 0. 123456 123456 123456 123456 . I.5 . I. 2. 0. 0. 10. $$ LOADING (SUBCASE 2) LSEQ 1000 920 200 M M N 200 O ET 112 DO D LA 201 I. E+7 0. I. ED NT 1000.Et4 5. I 1I 10. 0. I. 0.254 I . 10. +TB2 0.Et3 . ED NT $ $ F A E F R PLOT RM O GRID 1000 GRID 1001 GRID 1002 GRID 1003 PLOTEL 200 1000 PLOTEL 201 1001 PLOTEL 202 1002 PLOTEL 203 1003 $ E D AA NDT 2. TLOADI 2001 920 TABLED1 21 1 +TB2 0. 1. 2001 I. +13.3 +PGI $ $ $$ LOADING (SUBCASE 1 AND 3) LSEQ 1000 910 100 FORCE 100 112 TLOADI 101 910 TABLED1 I l l +TBI 0.02 2. +13. I. I.MT A1 PGAP +PG1 I1 20 . 101 0. 2.
e. The geometry and the loadings are axisymmetric.1. i.Chapter 11 EXAMPLE PROBLEMS 11. It is desired to find the critical pressure loads for snapthrough and snapback. For simplicity. Clamped boundary Simply supported boundary (and release of constraints) Sliding and simply supported boundary .15 in.1. 11. Material properties of the diaphragm are given as follows: Young's modulus: E = 2.757 in. The diaphragm has a thickness of 0. it is further assumed that the deformation pattern is confined to axisymmetric shapes.1 A spherical diaphragm resting on a rigid backing plate is subjected to a uniform pressure as shown in Fig.28 The objective of this analysis is t o study the effects of NLPARM parameters on the convergence characteristics when difficulties arise due to snapthrough.1 SNAPTHROUGH ANALYSIS OF A SPHERICAL DIAPHRAGM BY NEWTON'S ITERATION Problem Description 11.. Three cases are analyzed in this study according to the boundary conditions of the diaphragm edge. and the gap a t the center between the diaphragm and the backing plate is 0.0 x 105psi Poisson's ratio: v = 0.1.
and 6 fixed All grid points (GRIDs 200 . The typical input data for the clamped and the simply supported cases are listed in Tables 11. The contact process between the diaphragm and the backing plate is simulated by gap elements as shown in Fig. Subcase 7 is supplemented t o remove the constraints at the periphery after full unloading. 4. The apex point is allowed t o move only in the axis of symmetry. 3. The flange portion is omitted in the model.1. .1.1.2.2. and 6 of the grid points on the planes of symmetry. Case Control Section Subcases are used t o change the loading path and/or boundary conditions.1. As shown in Tables 11. The geometry and the output data are described in the cylindrical coordinate system. The contact process between the diaphragm and the backing plate is simulated by 13 GAP elements as shown in Fig. and the number of active DOFs is 37 for the clamped case.2. Within each subcase.1 and 11.1 and 11. In the case of the simply supported boundary in Table 11.1. Various boundary conditions are applied t o the GRID points on the periphery (GRIDs 170 and 171) as follows: Clamped case: all components (16) fixed Simply supported case: components 1.11. Executive Control Section SOL 66 (in Version 65) is used for the static analysis of the snapthrough phenomena.1. Various boundary conditions are imposed on the periphery for the three different cases to be investigated.2 Finite Element Modeling and MSCINASTRAN Input Data The main features of this problem are geometric nonlinearities due t o large displacements and changes in constraints due t o the backing plate. Bulk Data Section A 10degree sector of the diaphragm is modeled by one TRIA3 and six QUAD4 elements with linear elastic material. 2. 11.1. the loads are subdivided into equal increments and each load increment requires iterations for convergence. 4. 3. A total of 28 GRID points are used in the model.2.2. 4. respectively. 11. The number of total DOFs of the problem is 168. A 10degree sector of the diaphragm is modeled using shell elements with axisymmetric boundary conditions. The axisymmetric boundary condition is imposed by constraining components 2. and 6 fixed r Sliding and simply supported case: components 2.261) simulating the backing plate are fixed. Case Control data include loading and unloading phases.1.
The solution strategy is provided by the NLPARM Bulk Data. The external pressure applied t o the diaphragm surface in each subcase is specified using PLOAD2. . which are varied from subcase t o subcase depending on the convergence characteristics. +I. Smaller load increments are used around the loads where snapthrough and snapback are anticipated to occur. LGDISP.Effects of geometric nonlinearity are included in the analysis by specifying PARAM.
1.1. Spherical Diaphragm Sujected to a Uniform Pressure (a) Side View (b) Isometric View .AFTER SNAPTHROUGH 7 SPHERICAL DIAPHRAGM (UNDEFORMED) I' RIGID BACKING PLATE Side View Isometric View Figure 11.
2. Spherical Diaphragm Model with Axisymmetric Boundary Condition and Constraints by Gap Elements .BOUNDARY CONDITIONS Case 1 m I Case 2 Figure 11.1.
001 FIND SCALE ORIGIN 1 SET 1 PTITLE=ISOMETRIC VIEW PLOT SET 1. 0. . 0. Z FIND SCALE ORIGIN 1 SET 2 . ORIGIN 1. .90 PSI NLPAR = 1 LOAD = 90 SUBCASE 2 SUBTI = 1.6 PSI NLPAR = 4 LOAD = 60 SUBCASE 5 SUBTI = 0.3 PLOTTER NAST PAPER SIZE 29.5 PSI NLPAR = 3 LOAD = 150 SUBCASE 4 SUBTI = 0.0 PSI NLPAR = 6 OUTPUT(PLOT) CSCALE 1.1 Input Data for the Clamped Case ID SNAPC.Table 11. X .1.50 CEND TITLE = SNAPTHROUGH ANALYSIS OF A SPHERICAL DOME SUBTI = EDGE CLAMPED ECHO = UNSORT SEALL = ALL SPC = 200 SET 1 = 100 110 120 130 140 150 160 SET 2 = 1 THRU 7 SET 3 = 170. SET 1 ALL SET 2 ALL EXCEPT GAP MAXI DEFO .0 PSI NLPAR = 2 LOAD = 100 SUBCASE 3 SUBTI = 1. LABEL VIEW 0. AXES MY .35 PSI NLPAR = 5 LOAD = 35 SUBCASE 6 SUBTI = 0.V66 $ SHL 5/2/85.200 DISPL = 1 SPCFO = 3 SUBCASE 1 SUBTI = 0. X 21. GBL 11/22/86 TIME 30 SOL 66 DIAG 8.
0 0.0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0.315 . 0.0 1. 0.0 10.445 . 0.0 0.238 .238 .479 7.I53 .0 10.082 2.082 1.486 .322 5.064 .445 .393 .0 10.0 $ GRID POINTS GRID 100 GRID 110 GRID 111 GRID 120 GRID 121 GRID 130 GRID 131 GRID 140 GRID 141 GRID 150 GRID 151 GRID 160 GRID 161 GRID 170 GRID 171 $ ELEMENTS CTRIA3 1 CqUAD4 2 CqUAD4 3 CqUAD4 4 CqUAD4 5 CqUAD4 6 CqUAD4 7 0.90 1 THRU 7 PLOAD2 100 1.0 1 THRU 7 PLOAD2 150 1 .557 100 110 120 130 140 150 160 0.064 110 121 131 141 151 161 171 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 1 1 1 1 1 1 1 120 130 140 150 160 170 . 5 1 THRU 7 $ $ $ 150 151 160 161 GO ER E MTY 0.163 3.163 2.0 0.35 1 THRU 7 PLOAD2 60 .486 .60 1 THRU 7 PLOAD2 90 .315 .0 1.001 SET 2 S M O 1 BEGIN B L UK $ PARAMETERS PRM AA K6ROT 1.393 .479 6. 0.0 10.243 3.0 10.401 6.0 10.0 0.0 0.0 10.557 7. 111 111 121 131 141 151 161 .322 4.0 +C 1 CORD2C 1 +C1 1.I53 . 0. 0.PTITLEORTHOGRAPHIC VIEW Y BL PLOT STATIC 0 M A X I DEFO .503 .243 4.0 0.0 PRM AA LGDISP 1 NP R 1 6 AT LA M UO YES +NL1 N P R 2 5 SEMI LA M 2 YES +NL2 +NL2 5 +NL3 NP R 3 2 LA M AT UO YES +NL3 5 NP R 4 3 LA M A T YES +NL4 UO NP R 5 5 AT LA M UO YES +NL5 +NL5 5 NP R 6 1 AT LA M UO YES +NL6 $ B U D R CONDITIONS ONAY SPCl 200 12456 100 SPCl 200 246 110 120 130 140 SPCl 200 246 111 121 131 141 SPCl 200 123456 170 171 $ LOADING PLOAD2 35 .401 5.
6 .2533 1 .6 1.0 GRID 231 1 3.0 GRID 251 1 5.+6 PGAP 240 .+6 PGAP 230 .082 0. GRID 220 1 2. 1.I062 1 .0697 1 .0 GRID GRID 210 1 1.28 PSHELL 1 1 0.+6 PGAP 261 .0 GRID 211 1 1.0 0. 1062 1 .$ BACKUP O F G A P S 200 1 0.163 0.+5 .6 1.0 GRID 221 1 2.+6 PGAP 250 . 1.+6 PGAP 210 . 1.401 0. 1. 1. 7.5792 1.243 10. GRID 250 1 5.+6 PGAP 221 .163 10. 1.6 1. GRID 260 1 6.6 1.5792 1.7393 1.6 1.6 1.4212 1.180021 .0 GRID 241 1 4.2 1 1. 1.6 1.254 1 .6 1.8359 1.0833 1.322 10.479 10. 1.243 0.0 GRID 261 1 6.+6 $ P L O T DATA PLOTEL 11 200 210 PLOTEL 12 210 220 PLOTEL 13 220 230 PLOTEL 14 230 240 PLOTEL 15 240 250 PLOTEL 16 250 260 PLOTEL 17 260 170 ENDDATA . 1.+6 PGAP 211 .I862 1 . GRID 240 1 4.479 0.7393 1.2398 1.+6 PGAP 251 .082 10. 180021 . $ G A P S T O R E P R E S E N T BACKING P L A T E CGAP 100 200 200 100 CGAP 110 210 210 110 CGAP 111 210 211 11 1 CGAP 120 220 220 120 CGAP 121 220 221 121 CGAP 130 230 230 130 CGAP 131 230 23 1 131 CGAP 140 240 240 140 CGAP 141 240 241 141 CGAP 150 250 250 150 CGAP 151 250 251 151 CGAP 160 260 260 160 CGAP 161 260 26 1 161 $ PROPERTIES MAT1 1 2.6 1.+6 PGAP 241 .+6 PGAP 260 . GRID 230 1 3.6 1. 1.401 10.15 1 PGAP 200 .322 0.+6 PGAP 220 .0833 1.4212 1.2398 1. 1.6 1.0697 1 1.2533 1 .6804 1.I862 1 .6804 1.2354 1 . 1.6 1.7570 1.+6 PGAP 231 .2354 1 .
3 PLOTTER NAST PAPER SIZE 29. 6 PSI.200 DISPL = 1 SPCFO = 3 OO D = 3 LA SUBCASE 1 SUBTI = 0 .50 CN ED TITLE = SNAPTHROUGH ANALYSIS O A SPHERICAL D M F OE SUBTI = EDGE SIMPLY SUPPORTED L SEALL = A L EH C O = UNSORT SPC = 200 SET 1 = 100 110 120 130 140 150 160 SET 2 = 1 THRU 7 SET 3 = 170. X 21. 0 PSI (EDGE CONSTRAINT REMOVED) SET 5 = 100 110 120 130 140 150 160 170 DISPL = 5 = 300 SPC NLPAR = 7 OUTPUT (PLOT) CSCALE 1 .1.72 P S I .Table 11. 8 PSI NLPAR = 5 LA O D = 80 SUBCASE 6 SUBTI = 0 . GEL 12/10/86 TIME 30 SOL 66 DIAG 8. second snapthrough NLPAR = 4 LA O D = 72 SUBCASE 5 SUBTI = 0 .2 Partial Listing of the Input Data for the Simply Supported Case I D SNAPS. first snapthrough NLPAR = 2 LA O D = 60 SUBCASE 3 SUBTI = 0. 0 PSI. unloading NLPAR = 6 SUBCASE 7 SUBTI = 0 .7 PSI NLPAR = 3 LA O D = 70 SUBCASE 4 SUBTI = 0.V65 $ SHL 5/2/85. 5 PSI NLPAR = 1 LA O D = 50 SUBCASE 2 SUBTI = 0 . .
LABEL VIEW 0 . ORIGIN 1.SET 1 A L L L A SET 2 A L EXCEPT G P M A X I DEFO . .70 1 THRU 7 PLOAD2 72 .72 1 THRU 7 PLOAD2 80 . 0 . 0 .60 1 THRU 7 PLOAD2 70 . 1 *** *** .001 SET 2 S M O 1 BEGIN B L UK $ PARAMETERS PRM AA K6ROT 1.80 1 THRU 7 $ $ $ 2 10 140 141 150 151 160 161 140 141 150 151 160 161 GEOMETRY The r e s t i s identical t o Table 1 1 .001 FIND SCALE O R I G I N 1 SET 1 PTITLE=ISOMETRIC VIEW PLOT SET 1 . X . AXES MY . Z FIND SCALE O R I G I N 1 SET 2 PTITLE=ORTHOGRAPHIC VIEW Y BL PLOT STATIC 0 M A X I DEFO . 1 .0 PRM AA LGDISP 1 NP R 1 5 AT LA M UO YES +NL1 N P R 2 5 A T 2 YES +NL2 LA M UO +NL2 7 NP R 3 4 LA M AT UO YES +NL3 N P R 4 4 SEMI 2 LA M YES +NL4 +NL4 5 NP R 5 4 AT LA M UO YES +NL5 NP R 6 8 A T LA M UO YES +NL6 N P R 7 10 A T LA M UO YES +NL7 $ B U D R CONDITIONS ONAY SPCl 200 12456 100 SPCl 200 246 110 120 130 SPCl 200 246 111 121 131 SPCl 200 12346 170 171 SPCl 300 12456 100 SPCl 300 246 110 120 130 SPCl 300 246 111 121 131 SPCl 300 2346 170 171 $ LOADING PLOAD2 50 . .50 1 THRU 7 PLOAD2 60 .
the central area is lifted off.3. the snapback pressure is removed after the second snapthrough as shown. 11.98 psi. The first snapthrough can be considered as an interim snapthrough by flipping into a shape of double curvature.1.1.98 psi. the load history between points A and B can be conjectured by the dashed line in Fig. Closer examination of the analysis results reveals that the diaphragm comes into contact with the backing plate near the periphery after the first snapthrough. Fig. This SPC force (Q) upon full unloading is computed as 19.1. The first occurs between 0. is conjectured as shown by the dashed line in Fig.3 Analysis Results Clamped Case The loaddeflection curve at the center point is shown in Fig.6(a). 11. However.1. If it is unloaded before the second snapthrough. which would not have existed otherwise.1. and the snapback between 0.3. where the snapthrough and the snapback begin t o occur. snapthrough occurs twice as shown snapthrough occurs between 0.6(b)) in contact with the backing plate.1 psi.11. leaving the peripheral area of the diaphragm (point A in Fig.1. It is shown that the curvatures of the diaphragm surface are reversed upon snapthrough or snapback at the critical loads.700 psi diaphragm snaps back at does not occur when the As shown in Fig. The load history between two points A and B. The SPC force of .4 shows the deformed shapes of the diaphragm for various loading and unloading levels.45 psi and 0. The curve is constructed by combining loading and unloading paths. When the pressure is removed entirely. two steps of snapthrough assume distinct shapes.705 psi. 11.5. and the second and 0. This is because the springback force is trapped by SPC forces a t the periphery. In fact. 11. in Fig. It is attempted t o release this reaction force by removing radial constraints on the periphery. Since the center point comes into contact with the backing plate at p = 0. which is equivalent t o a reaction force of 15.5. The snapthrough occurs between 0. 11.54 psi and 0.) despite the increasing load. The central portion of the diaphragm comes into contact with the backing plate in addition t o the peripheral area upon the second snapthrough.9 lbs on each node a t the periphery.1. The central area remains in contact with the backing plate during unloading until the pressure is reduced t o 0. 11. the point does not translate beyond the gap distance (0.48 psi. Based on these observations.50 psi and 0. the a pressure between 0.96 psi and 0. Simply Supported Case In the simply supported case.1. such contacts stabilized the interim snapthrough mode. in the radial direction. It is interesting that the snapback does not occur upon unloading in the elastic problem. 11.56 psi.40 psi.088 lbs/in.757 in.
1.1. Major options that can be changed for better convergence are increment size and parameters KMETHOD. Fig.9 lbs is released in ten steps by Subcase 7. r The timing and the line search affect the base vector (displacements) for the matrix update.4 Solution Progression The convergence of the solution is the most important issue for the iterative and incremental solution processes of nonlinear problems. the solution was more easily converged by suppressing the line search. It was discovered in this study that the effect of the line search process is rather erratic and unpredictable around the critical loads. As shown in Fig. Sliding and Simply Supported Case As the pressure is increased. 11.9 lbs and 13. restarts are attempted from that step with different options in the NLPARM. Nonlinear analysis requires restarts in general. but on some occasions. If the solution has not converged at a certain step. Upon snapback. .184 psi and 0. The general strategy is t o start with default options and reasonable increments of the applied load. As shown. This behavior was scrutinized and the following observations were contemplated: r The convergence characteristics around the critical load (snapthrough or snapback) are very sensitive t o the timing of the stiffness matrix update. the diaphragm resumes the original shape as shown in the figure. The convergence characteristics are mainly influenced by the increment size and the iteration strategy. When the reaction force is removed entirely at the end.1. The increment size was also reduced in the vicinity of the snapthrough.6(b). These crucial data for controlling the incremental and iterative processes are specified in the Bulk Data NLPARM. The periphery also slides back and forth radially as the SPC force is diminished. and MAXLS.190 psi. MAXDIV.182 psi.9 lbs.188 psi and 0. The line search was helpful t o jump over the unstable region in most cases. 11. the snapback occurs at a pressure between 0. the central displacement overshoots due to the radial compression force remaining at the periphery. Fig. 11. The snapback occurs when the SPC force is between 15.1.19.8 shows the loaddeflection curve at the center point. 11. the diaphragm undergoes a slight snapthrough at a pressure between 0. It is recommended that the value of MAXDIV be increased (5 to 7) for the subcase where snapthrough or snapback is anticipated. the diaphragm snaps back as the reaction force is released.7 shows the vertical displacement at the center and the radial displacement at the periphery as functions of the SPC force.
and Subcases 4. the problem does not converge due t o the same causes. of L.96 psi 0.S. and/or line searches were required in these subcases t o obtain the solution than in other subcases without an unstable region. of K Update 1 12 1 2 4 0 No.90 1.5 0. Subcases 1. solutions were easily achieved by default options. MAXDIV t o 5. Clamped Case Table 11. 0 41 16 I 16 0 3 4 5* 6 Note: Snap occurs at subcases denoted by * .3 shows the solution strategy and the performance of each subcase when the solutions are obtained successfully. and the problem stops due t o TWO DIVERGING SOLUTIONS after 11 iterations with 3 stiffness updates. In the second and third trials. The snapthrough occurs in Subcase 2. and 6 represent the unloading phase. the program detects a negative term on the factor diagonal of [K].1. the problem stops before convergence due to the maximum iteration limit of 20. In subcases without an unstable region. of INC 6 5 2 3 5 1 Solution Strategy Default SEMI.00 No.  Table 11.1.OO 1. Table 11.3 Solution Strategy and Performance for Clamped Case Subcase ID 1 2* Load (psi) 0. 2. the problem converges after 16 iterations with 6 stiffness updates. Restart runs were made with smaller load increments and different iteration strategies until converged solutions were achieved. when KMETHOD is set to SEMI. in which snapthrough occurs.1.98 psi). It can be noticed that frequent stiffness updates at converged states (due t o KSTEP=2) expedite convergence. iterations. The solution has not converged with default options in the vicinity of the critical loads. and KSTEP t o 2. 5.60 0. In the first trial. of Iter 29 29 11 18 35 6 No. Finally. and the snapback occurs in Subcase 5.35 0. KSTEP=2 MAXDIV= 5 MAXDIV= 5 Default MAXDIV= 5 Default No.4 summarizes trialanderrors attempted for Subcase 21increment 4 (p = 0. It is noted that more stiffness updates. and 3 represent the loading phase.A slight difference in the base vector for the update could lead the solution to either side of divergence or convergence in the vicinity of the critical loads without due explanation. In the fourth trial.
and 5 denoted by analyzed the unloading path from 0. The first 3 trials in Table 11.S.1. uccessful solution is achieved at the final trial & Simply Supported Case Table 11. respectively.5 shows the solution strategy and the performance of each subcase upon successful solutions.6 were terminated due t o TWO DIVERGING SOLUTIONS. of L. After the pressure is removed entirely.4 Effects of NLPARM Parameters on Convergence for Subcase 2/Increment 4 of Clamped Case Trial Case 1 2 KMETHOD (AUTO) (AUTO) SEMI SEMI SEMI Parameters KSTEP MAXDIV (7) (7) (7) (7) 2 (2) 5 5 7 5 MAXLS (5) (5) (5) (5) (5) No. 4.Table 11. +. Subcase 3. . the radial constraint of the diaphragm is released. Another set.1.1.7 shows two trials terminated due t o the maximum iteration limit of 20. The solution requires strategies different from the default options for NLPARM in the vicinity of the critical loads.1. of K Update No. Tables 11.1. which is analyzed by Subcase 7.1. 3 4 5 3 5 2 5 6 19 6 20 16 36 48 6 45 41 ) denotes default option Note: 2. where the program detects a negative term on the factor diagonal of [K] as well as the GIMBAL ANGLE exceeding 90 degrees. Subcases 1 through 5 represent the loading phase and subcase 6 represents the unloading phase. of Iter 11 No.7 show the trialanderrors t o obtain solutions for the first (Subcase 2/increment 3) and the second snapthrough (Subcase 4/increment I).60 psi after the first snapthrough occurs. It is noted that variations of MAXDIV and MAXLS have significant effects on the performance while jumping over the unstable region for the snapthrough analysis.6 and 11. Table 11.
60 0. of K Update 1 2 2 KMETHOD (AUTO) (AUTO) (AUTO) (AUTO) MAXLS (5) (5) 2 2 No.80 0.00 0. of Iter 5 8 8 11 No. uccessful solution is achieved at the final trial k . denotes restart for snapback during unloading + Table 11. of L. Snap occurs at subcases denoted by * 2. of L.00 No.1.S.Table 11.S. of K Update 1 7 0 10 0 2 7 0 8 1 No.70 0. of Iter 24 25 17 25 8 30 75 12 27 24 No.5 Solution Strategy and Performance for Simply Supported Case Subcase ID 1 2* Load (psi) 0.1.40 0. of INC 5 5 4 4 4 8 10 4 5 5 Solution Strategy Default KSTEP=2 MAXDIV=7 MAXLS=2 Default SEMI KSTEP=2 MAXLS=10 Default Default Default No. 5 24 12 8 3 ) denotes default option Notes: 2.6 Effects of NLPARM Parameters on Convergence for Subcase 21Increment 3 of Simply Supported Case Trial Case 1 2 3 4 Parameters MAXDIV KSTEP (7) 2 2 2 (2) 5 5 7 No. 0 8 0 87 0 3 18 3 4* 5 6 7* 3+ 4+* MA~DIV=~ 5+ Default KSTEP=2 MAXDIV=7 Default 0 11 0 Notes: 1.00 0.50 0.50 0.72 0.
S. 12 20 0 87 2 2 (2) 7 7 5 3 6 6 7 ) denotes default option Note: 2.18 0. the load is incremented in very small steps around the critical load t o obtain a more descriptive loaddeflection curve. of K Update No.20 0.1 .S. of L.16 0.Table 11. and the snapback occurs in Subcase 8.00 No. Subcases 1 through 5 represent the loading phase.8.60 0.7 Effects of NLPARM Parameters on Convergence for Subcase 4/Increment 1 of Simply Supported Case Trial Case 1 2 3 4 KMETHOD (AUTO) (AUTO) (AUTO) SEMI Parameters KSTEP MAXDIV (7) 2 MAXLS (5) 2 0 10 No.18 0.8. 0 0 18 0 1 0 2 14 16 1 2 3* 4 5 6 7 8* 9 Note: Snap occurs a t subcases denoted by * 11. of K Update 1 0 4 0 2 0 1 3 2 4 No.19 0. uccessful solution is achieved at the final trial Sliding and Simply Supported Case As shown in Table 11. In fact. Since the snapthrough and the snapback in this case are moderate and shallow as shown in Fig. of Iter 12 20 20 19 No.16 . of L. of INC 4 1 5 1 4 1 1 5 2 Solution Strategy Default Default MAXDIV=5 Default Default Default Default Default Default No. nine subcases are used for this analysis. the coarser load increment rendered solutions by default options without giving details of snapthrough around the critical load.1.1.35 0. of Iter 24 5 32 4 25 6 9 29 8 No.1. 11. and Subcases 6 through 9 the unloading phase.8 Solution Strategy and Performance for Sliding and Simply Supported Case Subcase ID Load (psi) 0. Table 11. The snapthrough occurs in Subcase 3.1.19 0.
1.11. In general.5 Concluding Remarks While default options in NLPARM are effective as a general strategy for ordinary problems. some variations are required to overcome difficulties if divergence is encountered. . It is discovered that such variations are required to jump over the unstable region of the snapthrough and snapback problems. Helpful variations for snapthrough problems are SEMI rather than AUTO Smaller value for KSTEP to allow more frequent stiffness updates at convergence Larger value for MAXDIV to bypass the divergence criteria Larger value for MAXLS to exercise more line searches It is also observed that the solutions to the snapback are easier t o achieve than to the snapthrough. more conservative methods are more reliable but also expensive.
@ \ \ \ \ : : I .?'" . i : i0 r i i .."! Loading @ Unloading ..I 0 : : .j . ..?'. ..! . . * *1 .".. .. .""i" 0."" . 0. i i . \ \ \ ..1.00 7. . Load (psi) . "...3 Load vs..I i .00 &.. .25 .?'?'".I 5 .. .50.'i".7.  0.75 ."I.i"i"".00 .j" 7""r" .... !: 0.. \ \ \ 0....I i . .   M 0 0' B @ ....I .. Vertical Displacement at the Center (Clamped Case).i.. .".... : .Conjectured x @ A i i 1...5 TT.* . . 0.. ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 1 Displacement (in) Figure 11. ~ .i". : : ...I I I A g' \ $ 0 \ \ I x \ \ i .!.
' i I UNDEFORMED p = 0.75 p s i FULLY UNLOADED I / p = 0. Deformed Shapes of the Diaphragm (Clamped Case) (a) Loading (b) Unloading .45 p s i Figure 11.1.4.
.I .25 ! i i i ....? .."...I . ........": j ... ... 0 ... . .. .. ......1.00 Displacement (in) Figure 11."...."!x ? : ....5 Load vs...'.... .... ... ...: ..... .: p i i i j 0 \ \ \ \0 f~ 0.."".. . ............ : i 0... 0 j @ :....... . ...Conjectured x 9 j j A ?(X.....00 0.. @ i ? ! ? ? ......... .. x >i: \ \ \ \ \ \ 0 \ i \ \ Load (psi) I 0 \ \ \ \ \ \ \ i j j ........% @ .... ! I ....oo .... j i i i ...75 1............. :.. .100 ..... i \......I i : .Ao : : ..75 ..! i ! ' ! ! ! 0.I 0.5 0........ Vertical Displacement at the Center (Simply Supported Case)... "...... 8. : I 6 \ \ \ ... .. i Loading @ Unloading ...25 .."""..........
Q = 19. 0 psi..0 /FULLY UNLOADED ( p = 0 .._ .1. Deformed Shapes of the Diaphragm (Simply Supported Case) (a) Loading (b) Edge Constraint Removed after Unloading .p _ BACKINGPLATE = 0...6..72 psi (a) ' ...9 1 b s ) FINAL POSITION I I / Q = 11..9 lbs Figure 11..
0.40
, . .. . .,.,... 3~"""
.
0.20
&,,.~.""
*....,..*x"'"'
....* ...>c.""' '
.....',*.x...""'
 Q 
I I I
I
+....../K'.." .
I I I
,.' 0.oo 49@@@@
Displacement 0.20 (in)
I
!@"
I
Q..
+,
0.7
I I I I
I I I
Vertical Displacement (Center) @ Radial Displacement (Boundary)    Conjectured
x
I
0.40
I
I I I I I
0.60

k......,,'........
,
"'X. ...,,,
0.80 0
I
I
I
5
10
15
20
SPC Force (Q)
Figure 11.1.7 Vertical and Radial Displacements at the Center and Boundary Points vs. SPC Forces at the Periphery.
Load (psi)
Displacement (in)
Figure 11.1.8 Load vs. Vertical Displacement at the Center (Sliding and Simply Supported Case).
11.2
PARAMETRIC STUDY ON BFGS UPDATES AND LINE SEARCH METHOD IN SOL 66
11.2.1 Introduction
The quasiNewton update and the line search methods are employed as options to accelerate the convergence and improve the overall effectiveness of the modified Newton's iteration. When the BFGS method was implemented and tested in MSC/NASTRAN, our experience showed that the implementation scheme and adjustment of parameters had a significant impact on the program's performance [11.1].The line search method is an expedient, but not an expeditious method that could improve the computational efficiency. The line search method is modified and implemented t o optimize its usefulness for a general class of problems in conjunction with quasiNewton updates [11.2]. Parametric studies are performed to validate the effectiveness and efficiency of the algorithm. The study shows that the line search method is essential t o achieve convergence in some highly nonlinear problems, but has little impact on the efficiency. Various problems are examined to verify the algorithm's ability t o handle nonlinear problems involving softeninglhardening and buckling behavior with materialJgeornetric nonlinearities. The effectiveness of the solution process is studied for different values of the fields MAXQN, MAXLS, and LSTOL in the NLPARM Bulk Data entry. Three example problems are presented t o illustrate the convergence and efficiency characteristics of the quasiNewton update and the line search process. The analyses are performed using MSCINASTRAN Version 65. The iteration strategy is based on the hybrid Newton's method, by which the stiffness is updated only when it is necessary for efficiency and effectiveness [11.3]. Two cases are considered for each problem. In the first case, fields MAXLS and LSTOL are kept constant (equal t o 5 and 0.5, respectively, unless otherwise stated), and field MAXQN is varied. In the second case, field MAXQN is kept constant (equal t o 30), and fields MAXLS and LSTOL are varied.
11.2.2
The Cologne Challenge: a ZShaped Cantilever Beam
A Zshaped cantilever beam as shown in Fig. 11.2.1 was proposed as a benchmark problem for testing nonlinear capabilities of FEM codes [11.4]. The beam, when subjected t o the vertical load at the end, undergoes very large displacement and plastic deformation with von Mises yield criterion and isotropic hardening [11.5]. Two different models are used in the present analysis: a plate element model (ZPLATE) and a beam element model (ZBEAM). The analysis is difficult because of the highly nonlinear behavior resulting from the large rotation a t the plastic hinge as shown in Fig. 11.2.2. The loaddeflection curve a t the end point is shown in Fig. 11.2.3. Notice that the nonlinearity of the system in the loading phase of 01000 N is much more severe than that of 10002000 N. This makes the problem suitable for examining the performance of nonlinear iteration algorithms in both highly and slightly nonlinear regions.
plate thickness: 3.4 mm plate width : 20 mm tip load : 2000 N Figure 11.2.1. Cantilevered 2Shaped Beam (QUAD4 Model).
SHAPE
Figure 11.2.2. Deformed Shapes (QUAD4 model).
11.2  2
I700

ZPLATE
Figure 11.2.3. LoadDeflection Curve.
Case 1: Tables 11.2.l(a) and 11.2.2(a) summarize the numerical performance of the BFGS updates for ZPLATE and ZBEAM, respectively, by varying MAXQN from 0 to 30 with an interval of 5. The number of QN updates reflects the cumulative QN updates counting all the quasiNewton vector pairs involved in every iteration. By increasing the value of MAXQN, more QN updates are performed. When the QN update is suppressed (MAXQN=O), the beam model fails to converge, whereas the plate model requires an excessive number of line searches, stiffness updates and iterations for convergence. In terms of efficiency, both models seemed to show most favorable results with a MAXQN of 20.
Table 11.2.l(a). Numerical Performance as a function of MAXQN in ZPLATE
n
I
MAxQN COMPUTATION 15 10 0 5 EFFICIENCY 0 4191 4255 4995+ No. of QN updates . . 256 0  1000 1 No. of line searches 1 503 327 260 11 1130increments) I No.ofiterations i 1412 1326 1196 1195 140 147 225 148 No. of I< updates severe nonlinearity 2288 0 2329 1462 No. of QN updates 1 0 I 0 No. of line searches 8 1000  2000 436 (100 increments) No. of iterations 1 578 1 565 / 433 1 401 23 mild nonlinearitv I No. of K u ~ d a t e s 1 29 1 191 1 1.303 1 1.079 1 1.034 1 1.031 1 total CPU time*, normalized Notes: * VAX 11/780 with VAX/VMS V4.5, not accurate as performance index 15 QN vectors dragging at 870N for 21 iterations
H11
LOAD (N)
I I
I

I
I
20 4526 259 1212 138 3882 5 463 15 1.000
1
25 4476 252 1158 141 4971 9 507 12 1.013
I
1
1
I
30 4910 258 1196' 140 ' 5537 9 514 14 1.023
'
+
Table 11.2.2(a). Numerical Performance as a function of MAXQN in ZBEAM
Notes: VAX 111780 with VAX/VMS V4.5, not accurate as performance index + Diverging solution at load step 100200 N.  Reaching the iteration limit of 30 at load step 300400 N.
*
Case 2: Tables 11.2.l(b) and 11.2.2(b) summarize the numerical performance of the line search algorithm for ZPLATE and ZBEAM, respectively, by varying MAXLS from 0 to 6 with LSTOL set at 0.5. It is clear from the tables that the effectiveness of the line search varies with the extent of nonlinearity. Extensive line searches are performed in the highly nonlinear region: the number of line searches over the number of iterations is around 22% in the plate model and 2933% in the beam model. In the slightly nonlinear region, the line search is hardly required: the ratio of the number of line searches to iterations is around 1% the plate model in and zero in the beam model. By increasing the value of MAXLS, more line search operations are performed in the highly nonlinear region with more favorable results for MAXLS of 4 and 5. However, little difference is observed in the efficiency for RlAXLS between 3 and 6. In the less nonlinear region on the other hand, MAXLS of 2 or 3 shotvs more efficient results as illustrated in Table 11.2.l(b). When the line search is suppressed (MAXLS=O), both the plate and beam models fail to converge at the onset of plastic deformation around the load of 200 N.
Such failures illustrate the necessity of combining the line search operation in the nonlinear iteration algorithm. Table 11.2.l(b). Numerical Performance as a function of MAXLS in ZPLATE
*
+
Notes: VAX 11/780 with VAX/VMS V4.5 (not accurate as performance index) Diverging solution a t load step 200300 N.
Table 11.2.2(b). Numerical Performance as a function of MAXLS in ZBEAM
Notes: * VAX 111780 with VAXIVMS V4.5 (not accurate as performance index) Diverging solution at load step 100200 N.  Reaching the iteration limit of 30 at load step 500600 N.
+
11.2.3
SnapThrough and SnapBack: a Spherical Shell with Backing Plate
A spherical diaphragm resting on the rigid backing plate is subjected t o a uniform pressure as shown in Fig. 11.2.4. The analysis objective is t o find the critical pressure for snapthrough and
snapback under different boundary conditions: clamped, simply supported, and sliding/simply supported cases. Additional constraints are applied after the snapthrough by the backing plate. The material is linear elastic, and the deformation pattern is assumed to be axisymmetric because the diaphragm is connected t o a plunger which prevents the apex from rotating.
A 10degree sector of the diaphragm is modeled using shell elements with axisymmetric boundary conditions. The contact process between the diaphragm and the backing plate is simulated by gap elements as shown in Fig. 11.2.5. The flange portion is omitted in the model.
The main features of these problems are geometric nonlinearities due t o large displacements, follower forces, and changes in constraints due to the backing plate. The convergence characteristics are influenced by the load increment and the iteration strategy. The increment size is reduced in the vicinity of the snapthrough and snapback. The convergence behavior is rather erratic and unpredictable around the critical loads.
AFTER SNAPTHROUGH 7 SPHERICAL DIAPHRAGM (UNDEFORMED)
0'757 in
'f
I
L
RIGID BACKING P U T E
Side View
Isometric View
Figure 11.2.4. Spherical Diaphragm Subjected to a Uniform Pressure.
BOUNDARY CONDITIONS
G
Case 1
Case 2
Case 3
83 g3
n
8
GAP ELEMENT (1 3) A SECTION OF A SPHERICAL SHELL
z
RIGID BACKING PLATE
171
Figure 11.2.5. Spherical Diaphragm Model with Axisymmetric Boundary Condition and Constraints by Gap Elements.
i) Clainped Case
The loaddeflection curve a t the apex point for the clamped boundary conditions, constructed by combining loading and unloading paths, is shown in Fig. 11.2.6. The snapthrough occurs between 0.960.98 psi, and the snapback between 0.450.40 psi. Since the apex point comes into contact with the backing plate at 0.98 psi, the point does not translate beyond the gap distance (0.757 in.) despite the increasing pressure. The load history between two points A and B where the snapthrough and the snapback begin t o occur is conjectured as shown by the dashed line.
w
I
I

+
Loading Unloading
a

A
\
0
C)
Conjectured
.
\
\
\
\
<)
Load
\
(Psi)
\
\
\
\
\
\
'\
\
. .
1 1 1 1 1 1 1 1 1 1
1
1
1
1
1
1
Displacement (in)
Figure 11.2.6. Load vs. Vertical Displacement at the Center (Clamped Case).
Case 1: The parametric study is performed by varying MAXQN from 0 to 30 with an interval of 5. The results are summarized in Table 11.2.3(a) for loading and unloading phases separately. Convergence near the snapthrough point occurs only at a MAXQN of 10 and 20, which indicated the erratic nature of the problem. The unloading phase is analyzed by restarting from the converged solution with a MAXQN of 20. The fact that the unloading is easier than the loading for convergence is intriguing.
Table 11.2.3(a). Numerical Performance as a function of MAXQN in t h e spherical diaphragm with clamped boundary
loading (snapthrough)
Notes:
VAX 8700 with VAX/VMS V4.5, CPU time normaIized t Diverging solution a t snapthrough (0.960.98 psi) Case 2: The parametric study is performed by varying MAXLS from 0 to 8. The results are summarized in Table 11.2.3(b) for loading and unloading phases separately. As indicated in the table, the convergence near the snapthrough point requires MAXLS to be greater than 4. Computational efficiency appears t o be better with MAXLS of 5 or 7 than 6, which cannot be rationalized. Again, the convergence is easier for unloading than the loading.
*
Table 11.2.3(b). Numerical Performance as a function of MAXLS in t h e spherical diaphragm with clamped boundary
n
LOAD
I COMPUTATION I
hf AXLS
il
Notes: VAX 8700 with VAXIVMS V4.5 + Diverging solution a t the pressure 0.960.98 psi  ~ e a c l i n g iteration limit of 30 in the vicinity of the snapthrough pressure the
*
ii) Simply Supported Case
In the simply supported case, the snapthrough occurrs twice as shown in Fig. 11.2.7. The first snapthrough occurrs at 0.540.56 psi, and the second occurrs at 0.700.705 psi. The first snapthrough brings the diaphragm into an interim stable mode by flipping it into a shape of double curvature. The diaphragm comes into contact with the backing plate near the periphery immediately after the first snapthrough, stabilizing the interim position. The central area of the diaphragm comes into contact with the backing plate upon the second snapthrough. If the pressure is unloaded before the second snapthrough, the diaphragm snaps back at 0.500.48 psi. Once the second snapthrough occurs, however, the snapback does not occur upon unloading because the springback force is trapped by the constraint forces at the periphery. Instead, the central area remains in contact with the backing plate during unloading until the pressure is reduced to less than 0.1 psi (around point B). When the pressure is removed entirely, the central area is lifted off, leaving the peripheral area in contact with the backing plate. Based on these observations, the load history between points A and B can be conjectured by the dashed line in Fig. 11.2.7.
Case 1: Table 11.2.4 summarizes the numerical performance as a function of MAXQN for the first and second snapthrough phases separately. The convergence behavior dictates that the maximum number of line searches allowed for each iteration be set to 2 and 10 for the first and the second snapthrough, respectively. The first snapthrough requires MAXQN to be greater than 0 for convergence. The solution to the second snapthrough phase can only be obtained with a MAXQN of 15 and 20. Computational efficiency appeares t o be better with a MAXQN of 20.
Table 11.2.4. Numerical Performance as a function of h/IAXQN in the spherical diaphragm with simply supported boundary
Notes: VAX 8700 with VAX/VMS V4.5 + Diverging solution at first and second snapthrough  Reaching the iteration limit of 30 at the second snapthrough
*
Case 2: The first snapthrough requires MAXLS to be greater than 1 for convergence and the second snapthrough converges only with MAXLS of G and 10 for the MAXQN range of 0 to 10. The parametric study is not fruitful for this case due t o the erratic numerical behavior.
I
w
w
w
I
*
.A + ' , / .
3
c>
\
\
Load (psi)
f
\
\
\
\\
\
\
\
\
(>
\
\
\
\
()
   Conjectured
d .
0
Unloading
\
\ \
\
\ \
\
()
\
\
\
<>
\
\
\
)
\( \
B
l l l l l l l l l l l l l l L l " l l l
Displacement (in)
Figure 11.2.7. Load vs. Vertical Displacement at the Center (Simply Supported Case).
iii) Simply Supported and Sliding Case In the case where the periphery is allowed t o slide in the radial direction while it is simply supported. Vertical Displacement at the Center (Sliding and Simply Supported Case). The snapback occurrs at a pressure between 0.8. Since the snapthrough in this case is moderate and shallow. as shown in the figure.1840.190 psi. the pressure is incremented in very small steps around the critical load t o obtain a more descriptive loaddeflection curve. the diaphragm undergoes a slight snapthrough a t a pressure between 0. Load (psi) + Loading Unloading Conjectured .8.2.1880.182 psi as shown in Fig. . Load vs.ll.2. 0 Displacement (in) 'Figure 11.
000 1. The parametric study on the line search tolerance (LSTOL) with MAXLS of 5 is summarized in Table 11. It also indicates that the algorithm becomes inefficient if LSTOL has too large a value. of QN updates No. normalized + + 1. The choice of 0.5(b) summarizes the numerical performance as a function of MAXLS while LSTOL is fixed at 0. e.5(b) is the effectiveness of the line search. of QN updates + + 193 No.g.2. The computational efficiency improves slightly as MAXQN increases. Table 11.5(a) summarizes the numerical performance as a function of MAXQN for the loading and the unloading phases separately.. Notice that when LSTOL is small (0.108 LOAD (psi) I COMPUTATION I EFFICIENCY No. the algorithm keeps thrashing on line search operations until the solution diverges at the snapthrough or snapback point.0 20 + + (9 increments) No.2.Case 1: Table 11.6 . Numerical Performance as a function of MAXQN in t h e spherical diaphragm with sliding/simply supported boundary n Notes: * VAX 8700 with VAX/VMS V4.027 ' + Case 2: Table 11.5(c). No. + I + 1 311 0 . of iterations + + 61 unloading (snapback) No. This indicates that line search operations do help to overcome the difficulties near the singularity in the stiffness. of K updates 7 + + total CPU time*.5(a). The significance demonstrated in Table 11.3).7. the MAXQN values of 25 and 30 seem to be more efficient.0. of line searches 0 ( 5 1 10 MAXON II 1 15 1 20 1 25 1 30 I 367 1 608 1 603 1 653 L 21 19 19 19 90 91 90 90 7 6 5 5 171 231 245 183 21 21 U 23 26 61 59 55 59 7 6 5 5 1.5. This indicates that BFGS update operations do help to overcome the difficulties near the singularity in the stiffness. The table shows that the solution diverges around the snapthrough or snapback for MAXLS less than 3. and the efficiency is not sensitive to the value of MAXLS from 3 to 6.5 Diverging solution at'snapthrough or snapback point . 0.074 1.6 + +. (15 increments) 88 N o. The table shows that the solution diverges around the snapthrough or snapback point for a MAXQN less than 10.5 for LSTOL seems to be a good compromise for convergence and efficiency of the line search method.20 . of K updates loading (snapthrough) + + 10 No.2.2.115 1.2. .10. of line searches 0.
11.7]. of line searches 0. The problem is assumed to remain axisymmetric in geometry and loading throughout the deformation. Numerical Performance as a function of MAXLS in the spherical diaphragm with sliding/simply supported boundary RlAXLS COMPUTATION 6 1 4 5 0 2 3 EFFICIENCY (psi) 21 19 16 18 No.5(c).2. The external pressure is gradually increased until the shell collapses [11. 11. of line searches 0 . The details of the geometry and the material are shown in Fig. of iterations 90 I 91 91 90 + 87 5 ( 5 5 5 loading (snapthrough) No.5 (not accurate as performance index) Diverging solution at snapthrough or snapback point 11.6 .2. and the large displacement is assumed in the analysis.2.0 + + + 22 (9 increments) No.5 (not accurate as performance index) + Diverging solution at snapthrough or snapback point LOAD \~~ + ' I I I Table 11.2.4 PreBuckling Behavior: an Imperfect Spl~erical Cap A spherical shell with an initial imperfection is analyzed [11.008 1. The shell is subjected to an external uniform pressure. .10.2. of I updates 4 + + 22 22 22 No. of K updates 5 5 5 5 unloading (snapback) + + + total CPU time* (normalized) + + + 1. The material is elastoplastic with von Mises yield criterion and kinematic hardening. Numerical Performance as a function of LSTOL in the spherical diaphragm with sliding/simply supported boundary * Notes: + VAX 11/780 with VAXIVMS V4.010 1.0.11 represents a 10degree sector in solid elements with axisymmetric boundary conditions. of iterations + + + 5 9 62 62 62 No.003 1.Table 11.5(b). while the periphery is clamped.000 U U Notes: * VAX 111780 with VAXIVMS V4.6 0 + + (15 increments) No. The finite element model shown in Fig. The initial imperfection is introduced by making the radius of curvature near the apex greater than the shell radius.2.9. as shown in Fig.6]. 11.
a = 0. = 37. R = 0.8 x 1O6 psi u = 0.14328 in.8 x 10' psi Strain Hardening: Slope E.Geometry: h = 0. .32908 in. R. y = 26)' b = 0.2.3065' c = 0. Imperfect Spherical Shell.1506 in.3 uy = 7.267 in. = 1.8251 in.7612' Material: 7075T6 Aluminum E = 10.0251 in.1 x 1O6 psi Kinematic P Boundary Condition: Periphery Clamped Figure 11. = 1. a = 14.9.
10. 9 (psi) 3000 @ + buckling point by MSCiNASTRAN Finite Element by MSC/NASTRAN Finite Difference by Kao ) Experimental Buckling Point I 0 .) Figure 11.002 .014 .11.Pressure.010 .2.006 . W (in.16 . Deformed Shape upon Buckling with a Solid Model.012 .2.008 . Central Deflection).016 Central Deflection. ElasticPlastic Buckling of a Clamped Spherical Cap with Flat Spot (Load vs.004 . 11.2 . Figure 11.
5. not accurate as performance index + Diverging solution at load step 34003500 psi * Case 2: The results are summarized in Tables 11.2.2. The computational efficiency is most favorable with a MAXQN of 20.6(a) convergence is achieved for all values of MAXQN except for the second loading stage and MAXQN=O. which illustrate the numerical performance of the program as functions of MAXLS and LSTOL. . which illustrates the numerical performance of the program as a function of MAXQN. In this problem. Note that in Table 11. LSTOL=0. numerous analyses are repeated for various values of MAXQN. all the elements undergo plastic deformation. A quite natural consequence of unnecessary line searches is exemplified by decreasing efficiency with smaller LSTOL.2.6(c). However.2. The parametric study performed on LSTOL with MAXLS=5 is summarized in Table 11.6(a).6(b) and 11. When the line search operation is allowed (MAXLS > 0. Table 11.For the parametric studies.2. the loading can be divided into two stages: the plastic deformation developing stage (03000 psi) and the instability initiating stage (30003500 psi). Note that in Table 11. MAXLS and LSTOL with the pressure increased up to 3500 psi. more line searches are performed in the second stage where the geometric nonlinear effects become more prominent. Numerical Performance as a function of MAXQN in t h e imperfect spherical cap Notes: VAX 8700 with VAX/VhfS V4. Nonlinearity is fairly mild in both stages compared to other examples. In the second stage. where the geometric nonlinear effects become more prominent.6(c) does demonstrate that the accurate line search does not expedite convergence but rather deteriorates efficiency.2.6(c).6(b) the convergence is not only achieved when the line search is suppressed but also achieved most efficiently for MAXLS=O. Table 11.2. Case 1: The results are summarized in Table 11. This problem illustrates that the line search is unnecessary for moderately nonlinear problems where the divergence is not an issue.2.6(a).5).
Table 11. Although the line search method appears theoretically attractive.000 11.2. such as bypassing the unstable regime of the snapthrough . the solution requires more line searches.2.014 ( 1.020 7 1 7 1. The line search process can help t o converge the solution dramaticalIy in some cases.5 I 7 1. Notes: * VAX 11/780 with VAX/VMS V4.214 7 1.5 Concluding Remarks The BFGS update method is an expedient and effective procedure t o prevent divergence in complex nonlinear problems. of K updates total CPU time* (normalized) .6(b). but did not work on the unsymmetric matrix. Numerical Performance as a function of LSTOL in t h e imperfect spherical cap instability developing No.6(c)..5 Table 11.046 1 1 7 1. Preliminary tests showed that the current BFGS update scheme works well on the negativedefinite matrices.083 7 1.056 1 1 7 1. If divergence is not an issue. as demonstrated in the detailed parametric studies. It is particularly useful t o cope with difficulties when the stiffness approaches singularity. iterations and stiffness updates in the absence of BFGS updates. It is also an expeditious method that improves computational efficiency. the computational intensity involved in the search process often prohibits indiscriminate use of this procedure in nonlinear finite element analysis.2. Numerical Performance as a function of MAXLS in t h e imperfect spherical cap Notes: * VAX 8700 with VAX/VMS V4. The line search method is an effective procedure t o prevent divergence in complex nonlinear problems.
Tolerances and parameter values in MSCINASTRAN were tuned based on 50 nonlinear test problems [11. the line search diminishes the effectiveness and efficiency of the full NewtonRaphson iteration. Default values are intended to provide the best workable method for the wide spectrum of nonlinear problems without any insight or experience. the numerical behavior associated with the line search is rather erratic. They are not only problem dependent but also vary within a problem during the course of an incremental process. .8]. In fact. Optimal values of parameters and tolerances are extremely difficult to achieve. indiscriminate use of line searches causes adverse effects on efficiency.problem without the aid of a displacement control method or when the structure stiffens drastically as in the closing gap phenomenon. Even when the line search is valid. Moreover. However. but they are by no means optimal. the line search tends to deteriorate the efficiency in moderately nonlinear problems where divergence is not encountered.
.
03 inch The material has the following elastic properties: Young's Modulus 20 x lo6 psi Poisson's Ratio 0. The geometry of the pressure vessel. 11.3. A finite element model is constructed using 72 solid elements (4 PENTAs and 68 HEXAs) with 355 active degreeoffreedom.3 Model Description The modeling is simplified by axisymmetry and symmetry about the midplane.091 inch Fillet radius 0.1 The purpose of this example is to illustrate the creep solution of a typically encountered structure.2 Problem Description The creep behavior of a thickwalled pressure vessel with a flatend closure [11. respectively. as shown in Fig.25 inch Inside radius 0. is defined as follows: Overall length 0. The internal pressure is applied by PLOAD4 entries.89 inch Outside radius 0.3 and is assumed t o obey an empirical creep law in the form of where a and t are measured in psi and hours.3 CREEP ANALYSIS OF THICKWALLED PRESSURE VESSEL Introduction 11.159 inch Wall thickness (uniform) 0.3. Appropriate boundary conditions are imposed for symmetry (using SPC1. The structure contains a complex stress patteren requiring an axisymmetric threedimensional solution. 11.3. all the elements are treated as being made of elastoplasticcreep material by attaching MATS1 and CREEP entries t o the MAT1 Bulk Data entry.9] is analyzed under an internal pressure of 445 psi.3. SPCADD and GRDSET entries) in the cylindrical coordinate system (CORDIC). Although the applied load does not produce plastic deformation.1. 11. The material has the properties of an engineering metal subjected to high pressure. A local coordinate system (CORD2C) is used t o specify the geometry of a toroidal section. .11.
2 through 11. Notice that MSCINASTRAN shows a higher stress level at the inside surface (where the stress concentration occurs) throughout the analysis.1 shows a listing of the input data.4 Solution Procedure The creep analysis should be performed using SOL 66 of Version 63.3.2 and 11. this .5 Results and Discussion The entire analysis took six runs (five restarts) to obtain solutions up to five hours of creep. 11.e.1 0. Since the initial stress is similarly higher. 11.4 Time (Hour) 0. 0.5 in comparison with those of the Reference.0 3. Also shown is a model in the Reference [11.1) are set up as follows: Ending Creep Subcase ID 100 200 300 400 500 600 KMETHOD AUTO SEMIQN AUTOQN AUTOQN AUTOQN AUTOQN Increments 1 5 9 10 At (Hour) 0. 11. Then the analysis was continued by restarting from the last converged solution.3.3.1 shows an isometric view of the model. 11. 11. It can be seen that the steadystate stress is reached after approximately three hours of creep. Creep analyses are performed in the subsequent subcases by applying a nonzero time increment while the applied pressure is kept constant. Notice that the restart run requires three PARAMeter entries (i.3.3.4. The internal pressure of 445 psi is applied in the first subcase for the initial static solution.0 2.3. SUBID. Each creep solution took three to four iterations t o converge (average 10 CPU minutes on the VAX 111780).3.3.02 0. 11. Fig.3 show stress contours at t = 0 and 3. respectively..1 1.2 0.9] for comparison. 0. Six subcases (only two subcases are activated on the listing shown is Table 11.0 5.3. Table 11.0 5 5 The computer run was stopped after every subcase t o obtain the stress contour plot. The results of MSCINASTRAN analysis are presented in Fig.Fig.3. The stress history at two selected points (junction of the cylinder and end closure) are shown in Fig. LOADINC and LOOPID) to designate the solution from which the restart run starts.1 0.
It is not possible t o plot creep strain contours with the current version of MSCINASTRAN. 50 $. notice that the point with higher stress relaxes a little faster than might be expected.6 Input Data Listing for the Pressure Vessel ID TWCAS. Effective creep strains at five sampling points at t = 3 hours are shown on the creep strain contour plot in the Reference. 51 TIME 40 $ FOR VAX CEND TITLE = THICK WALLED CYLINDER SUBJECT TO INTERNAL PRESSURE SUBTITLE = AXISYMMETRIC ECHO = UNSORT DISP = ALL STRESS(PL0T) = ALL GPSTRESSALL STRFIELD=ALL SEALL = ALL SPC = 100 SUBCASE 1 0 0 LABEL = STATIC ANALYSIS LOAD = 100 NLPARM = 10 SUBCASE 200 LABEL = CREEP 0.20. 11.1 HOUR LOAD = 100 NLPARM = 20 SUBCASE 300 LABEL = CREEP 1 HOUR LOAD = 100 NLPARM = 30 SUBCASE 400 LABEL = CREEP 2 HOURS LOAD = 100 NLPARM = 40 SUBCASE 500 LABEL = CREEP 3 HOURS . More accurate solutions should be obtained with a more refined mesh around the fillet area.V65 $ SHL 6/21/83 SOL 66 $ DIAG 8. But the effective creep strains are printed with nonlinear stresses. The results are in good agreement with the Reference.error is actually a consequence of a coarse mesh. Also.3.
X.2 NLPARM 60 5 0. $ PROPERTIES MAT1 1 20.4 CORDlC 1 1000 1001 1002 GRID 1000 0 0.129 0. 0. GRID 1001 0 0. 0 . 0. .324 1. . 0.02 NLPARM 30 9 0. PTITLE=SIDE VIEW FIND SCALE ORIGIN 1 SET 1 PLOT SET 1 ORIGIN 1 LABEL ELEMENT PLOT STATIC SET 1 ORIGIN 1 OUTPUT(POST) SET 10 = ALL VOLUME 10 SET 10 SYSTEM CORD 1 BEGIN BULK $ GENERAL PARAM POST 0 NLPARM 10 1 AUTO NLPARM 20 5 0. 2 0 . 0 0 0 123456 123456 123456 0.Y VIEW 3 0 .0+6 0.129 0. 0.5 0.0+6 0.129 0.354 0.LOAD = 100 NLPARM = 50 SUBCASE 600 LABEL = CREEP 5 HOURS LOAD = 100 NLPARM = 60 OUTPUT(PLOT) CSCALE 1 . PLOTEL 1001 1000 1001 PLOTEL 1002 1000 1002 CORD2C 2 0.1 NLPARM 40 10 0. 0 . FIND SCALE ORIGIN 1 SET 1 PTITLE=ISOMETRIC VIEW PLOT SET 1 ORIGIN 1 LABEL POINT VIEW 0. 0. +COR2 1 1 . .1 NLPARM 50 5 0. 3 PLOTTER NAST SET 1 ALL AXES Z.324 +CUR2 0.5 GRID 1002 0 0.3 MATS1 I PLASTIC 2. . 0.
GRID 1101 0.CRLAW 1 +CRI 300 19. PLOAD4 200 104 1.159 1100 15. PLOAD4 200 308 1. PLOAD4 200 224 1. PLOAD4 200 316 1. PLOAD4 200 324 1. PLOAD4 200 216 1. PLOAD4 200 328 1. PLOAD4 200 108 1. PLOAD4 200 320 1.5953 0.025 CREEP $***** OTHER GRID CARDS ARE OMITTED.61 1 1 PSOLID 3 $ BOUNDARY CONDITIONS SPCADD 100 200 300 SPC1 200 23456 1301 +SPC 1361 1362 1381 SPC1 300 12456 1100 $ LOADING CONDITIONS LOAD 100 1. PLOAD4 200 112 1. PLOAD4 200 312 1.816 3. PLOAD4 200 304 1. PLOAD4 200 116 1.121 0. 2. PLOAD4 200 212 1. 1101 3 . PLOAD4 200 204 1. PLOAD4 200 220 1. PLOAD4 200 208 1. $ GRIDS R GRDSET 1 GRID 1100 0. 445. ***** 5. PLOAD4 200 120 1.324 1102 1120 1121 1122 GRID 1201 GRID 1396 $ ELEMENTS CPENTA 101 2 0.
$***** OTHER CONNECTIVITY CARDS ARE OMITTED. ***** 1375 1393 1394 +H328 CHEXA 328 +H328 1396 ENDDATA 3 1395 1373 1374 1376 .
Idealization of the pressure vessel with a flat end closure. Finite Element Model of Pressure Vessel. Figure 11.1. .3.
3. 11. Figure 11.Effective stress contour? at t i m e = 0 for the pressure vessel w i t h n flat end closure.2. Initial Stress Distribution.8 .3 .
....*... L e d C .............. L W J ...... Levd G ...... L w d E ....9 .................300.. 900.. 700..3 ... L w d 0 .. 200.3...... 800....EFFECTIVE STRESS AFTER 3 HOURS OF CREEP L W 8 ......... L W l . Levd H .. 500...  ..... Not@:Stre88 Unit8 in k d I i Figure 11. 1000.3........ 400.. Effective Stress Distribution after 3 hours of Creep 11... 600. L w d F ...0 hours for the pressure vessel with a flat end closure...Effective stress contours at time = 3.....
5.4.0  Figure 11. Effective Creep Strain ( x I o .3. Figure 11. 0.1 T W .001 0.  3 L . et.3.IJW / \ Im 1100 lOQJ  1 EFFKfNE STRESS A T INSIDE SURFACE \ \/' 4 / \ / \ / f m am. al.~ ) after 3 hours of Creep . \ 1'.W 1.01 I SUr*Q P U T E N D CLOSUIE I I E f ~ m r v S ~ U S CUI)IO( E AT x 10..007 176 ~. 700 > . 0.0 M S C / N ~ M 1W. \  X MSC/NASTPAN Greenbaum. Stress Relaxation at the Junction of the Cylinder and End Closure. L 600' WX) .O 0. I 400 XI0 2QO \ / IW.
Otherwise. The objective of this analysis is t o illustrate the applicability and effectiveness of the selfadaptive algorithm t o a nonlinear problem by examining an impulsively loaded cylindrical panel.4. Three different meshes are used t o solve this problem: 6x16 mesh .2 Finite Element Model This problem features material nonlinearity along with geometric nonlinearity due t o large displacements.l(a).4.11. The ends of the panel are simplysupported and the boundaries at the sides are fixed.205 inches and 60' in angle. as well as an adaptive stiffness matrix update strategy for efficiency. one element every 10 degrees .1 Problem Description A selfadaptive algorithm has been developed and implemented for the implicit time integration of the nonlinear transient response analysis (SOL 99 or 129) in MSCINASTRAN. Due t o symmetry. with an inside radius of 2. The panel is impulsively loaded by applying an initial velocity of vo = 5650 in.4.5 degrees .56 inches long. A 120' cylindrical panel is 12. as shown in Fig. 11. one element every 7. which is detected by the divergence criterion.9375 inches and a thickness of 0. dynamic buckling or snapthrough may induce a singular or indefinite dynamic stiffness matrix. the bisection will continue until the iteration process is stabilized and a converged solution is found. The iteration is bound t o diverge with a nearlysingular dynamic stiffness matrix. only one half of the cylindrical panel is modeled using shell elements.6 circumferential elements. Convergence difficulties induced by inadequate prediction of the time step size or the change in nonlinearities are tackled by the bisection method based on the divergence criterion. the time increment for the next time step is estimated based on the deformation pattern at the preceding step. For instance.8 circumferential elements. The iteration process for equilibrium employs expeditious methods such as quasiNewton updates and line searches. The dynamic stiffness is likely to be stable at this point and the difficulty in the snapthrough is coped with. The divergence criterion can detect and overcome dynamic buckling problems. In this algorithm.125 inches.4 TRANSIENT RESPONSE OF AN IMPULSIVELY LOADED CYLINDRICAL PANEL 11.16 elements along the panel's length r 8x16 mesh . respectively. 11. Then the bisection is activated and the dynamic stiffness is formed again./sec normal t o the region whose length and width are 10.
indicated by CORDlC. The automatic time step adjustment occurred eleven times in the analysis of the 16x32 model.5 psec. Subcase 1 contains the loading conditions (IC for initial velocity). 11. The remaining four subcases calculate the response to the impulsive loading. include the constraint of all six degreesoffreedom (DOF) for each GRID point at the fixed side of the panel. The initial step size was quadrupled a t the end of 10 psec. then reduced t o the original size a t 0. The initial step size was quadrupled to 8 psec a t the end of 0. 2. 11. The bisection was activated at 3 psec. The geometric nonlinearity is also taken into account by specifying PARAM.16 circumferential elements. The geometry is described in the cylindrical coordinate system. Material nonlinearity is signified by the inclusion of MATS1. with the time step sizes ranging from 8 psec t o 1 psec for the remainder of the analysis. in which points A and B (at z=6.75 degrees  32 elements along the panel's length The undeformed shape of the 8x16 model is shown in Fig. one element every 3. and 512 QUAD4 elements. The 6xl6. for the GRID points at the simplysupported ends of the panel are also constrained.4. The adjustment was activated again.42 in) indicate the nodal points where displacement responses are measured.LGDISP.4. namely components 1.4. the time step size was automatically adjusted nine times.8x16.3 Analysis Results and Discussion The analysis was performed for a duration of 1 msec with an initial time step size of 2 psec.04 msec.l.28 in and z=9.16 elements along the panel's length 16x32 mesh . A t was adjusted t o step sizes ranging from 4 psec t o 0. SOL 99 (in Version 67) is used for the nonlinear transient analysis of the impulsively loaded cylindrical panel. The material is assumed to be elasticperfectly plastic and is modeled using the von Mises yield criterion. specified using S P C l entries. Boundary conditions.22 msec. The analysis is performed in five subcases of equal duration (200 psec). specified by TIC in the Bulk Data. . and 6 of the GRID points along the side of the panel on the symmetric plane. The input data for the 8x16 model is listed in Section 11. then reduced t o 4 psec in the next time step when the bisection was activated. respectively. and 16x32 meshes used t o model the cylindrical panel contain 96. 4 . Symmetric boundary conditions are imposed by constraining components 2 .l(b). During the analysis of the 8x16 model.4. The solution strategy is controlled by the TSTEPNL data entry.. Translational DOFs. having the smallest value during peak displacements. and 3. the automatic time step adjustment was activated three times. reduced t o 4 psec at 0. 128. During the analysis of the 6x16 model. which specifies the parameter ADAPT indicating the adaptive strategy.31 msec. Thereafter.
4. Notably.78 0. [11. and the solution obtained by MSCIDYNA [11.NLAYERS. is given in Fig. Refinement of the mesh causes the cylindrical panel model t o become more flexible.87 0.3 in the Bulk Data. [11. Fig. 11. Contrary t o the observations in [11. This is attributed t o the fact that the dominant reaction force stems from membrane action and the material is perfectlyplastic. There is a significant improvement in the MSCINASTRAN results as the mesh is refined. The displacement response of the midspan of the cylindrical panel.2 for the three mesh sizes along with experimental results [11.12] 0.12] stated that the increased number of quadrature points through the thickness resulted in a stiffer behavior in the panel. Normalized Maximum Displacement at Midspan mesh size 6x16 8x16 16x32 MSCINASTRAN 0. 11. In order t o examine this point.13].l(c). thereby producing a more realistic displacement response curve. When the panel undergoes snapthrough. .12].3 compares the displacement response of the 8x16 model at points A and B using three and five Gauss points.69 0.4.4.10].4. the number of Gauss points did not have a significant effect on the displacement response.l(b).86 0. 11. Varying the number of Gauss points.11]. The table below lists the maximum displacement of the midspan normalized to the experimental results. and are compared with a solution using explicit integration [11. compared with the solutions for the same problem in Reference [11.12].The deformed shape of the 8x16 model at the peak displacement is shown in Fig.95 Kennedy et al. should not greatly affect the results of this problem. therefore.93 Kennedy et al. point A in Fig. 11. the stresses in the crown line reach the yield point and plastic yielding occurs in compression throughout the thickness. the 16x32 mesh solution is in very good agreement with the experimental results. MSCINASTRAN solutions were obtained using three Gauss points through the thickness by including PARAM.
5/27/87. XGRID LINES=YES YGRID LINES=YES YTITLE=DISP. MAXI DEFO 1.MZ.4.112 DISP=I SEALL=ALL SPC=IOO SUBCASE 1 LABEL=LINEAR ELASTIC REGION IC=I TSTEPNL=10 SUBCASE 2 LABEL=RESPONSE TO IMPACT TSTEPNL=lO SUBCASE 3 TSTEPNL=IO SUBCASE 4 TSTEPNL=10 SUBCASE 5 TSTEPNL=10 OUTPUT(PLOT) CSCALE=1. A T 1 0 8 % 1 1 2 ( T i ) XYPLOT DISP RESP/108(TI) .3 PLOTTER NAST SET I=ALL AXES X.Y VIEW 30.11.20.50 $ FOR VAX TIME 300 CEND TITLE=CYLINDRICAL PANEL UNDER IMPACT ECHO=BOTH SET 1=108.112(T1) . V66 $ SHL.E5 OUTPUT(XYPLOT) XTITLE= TIME IN SEC.0. DSNLTR SOL 99 $ NONLINEAR TRANSIENT ANALYSIS $ PRINT MATRIX TRAILERS AND ITERATIONS DIAG 8..4 Input Data Listing for the 8x16 Model ID CYROOF. PTITLE=ISOMETR..AIC VIEW FIND PLOT LABEL GRIDA PLOT TRANSIENT DEFORMATION 1 TIME 4.
1010 GRID 1 3. 1001 GRID 0. *1 == $ =11 $ TIC 1 301 1 *1 == $ =11 $ TIC 1 401 1 *1 == $ =11 $ TIC 1 501 1 *1 == $ =11 $ $ GEOMETRY A D CONNECTIVITY N CORDlC 1 1000 1001 1010 PLOTEL 1000 1000 1001 PLOTEL 1001 1000 1010 GRDSET. PARAM LGDISP 1 ADAPT TSTEPNL 10 100 2. 1. 1. GRID 1 3.785 101 1 3. 1.0 0. 100 GRID 0. GRID .0 0. 1000 GRID 0. 0. 0.5+6 0.14 104 1 3.BEGIN BULK PARAM K6ROT 100.0 0.355 103 1 3..0 0. 0.0 0. 0.57 102 GRID 2.6 1 $ MATERIAL AND GEOMETRIC PROPERTIES MAT1 1 10.125 1 $ BOUNDARY CONDITIONS SPCl 100 123456 900 THRU SPC1 100 246 100 THRU SPC1 100 123 100 200 +SPC 700 800 900 116 +SPC1 616 716 816 916 $ LOADING CONDITIONS TIC 1 101 1 *1 == $ =11 $ TIC 1 201 1 5650...33 MATS1 1 PLASTIC PSHELL 10 1 0. 0.l $ 0... GRID 3.
GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID GRID  =15 GRID  =15 GRID  GRID  =15 GRID  =15 GRID  =15 GRID  =15 GRID  CQUAD4 CQUAD4 CQUAD4 CQUAD4 CQUAD4 CQUAD4 CQUAD4 CQUAD4 CQUAD4 .
CQUAD4 CQUAD4 CQUAD4 CQUAD4 CQUAD4 CQUAD4 CQUAD4 CQUAD4  110 111 112 113 114 115 116 201 *1 = 14 CQUAD4 301 *1 = 14 CQUAD4 401 *1 =14 CQUAD4 501 *1 = 14 CQUAD4 601 *1 =14 CQUAD4 701 *1 = 14 CQUAD4 801 *1 = 14 ENDDATA .
349 msec Figure 11.sec2/in4 10.1 Impulsively Loaded Cylindrical Panel .5 x /b .4.5 x lo6 psi. Poisson's Ratio 4.2. Yield Stress 0 . Plastic Modulus 5650 i n l s e c .33. Young's Modulus 0. Initial Velocity (a) Problem Description (with 6 X 16 mesh) (b) 8 x 16 Model of Half the Panel (c) 8 x 16 Deformed Shape a t Time = 0.4 X lo4 psi.
2 Displacement Response at Midspan (z=6.Figure 11.4.28 in) .
3 Comparison of Gauss Integration Schemes Using 8x16 MSCINASTRAN Model .4.L Time Cmsec' Figure 11.
0 inch from the fixed end of the shaft and is free to slide along the shaft (this rigid body motion is restrained with a spring in the finite element model).0.5. K b . It is a nonlinear element available in both the nonlinear static and nonlinear transient solution sequences. K. which is typically very small and may be 0. and efficient reanalysis (restarting). Difficulties encountered with the use of the GAP include the selection of the GAP properties such as satisfactory open and closed stiffness values.e. provided that this does not make the global stiffness matrix singular. All analyses were performed on a VAX 8700. When compressive force exists in the GAP. A 20000 lb pointload on the . guidelines and principles suggested in this section is still valid for determining initial penalty values and the preload for the adaptive GAP element. respectively. i. the element coordinate system is not updated.0 inch long journal is 0. Although the same analysis can be performed more efficiently and effectively with much less userintervention by using the adaptive GAP elements in Version 67. This example examines these issues in SOL 66. it can carry any transverse shear load which is less than the coefficient of friction times the normal load. There is a 0.. which is a limitation t o be considered before using GAP elements.11. not defined by the distance between the GRID points attached t o the GAP element. The 9. Uo is the initial separation distance between the surfaces as defined by the PGAP entry. When the GAP is open it has an axial stiffness. 11. This implies that the orientation of the contact plane remains unchanged during deflection. While the GAP is open there is no lateral stiffness and no frictional forces. slip occurs. the GAP element does not take large displacements into account. When the shear load is greater than the friction force. It is completely fixed a t one end and free a t the other end.0 inch inside diameter and is 0.25 inches thick. The GAP is intended t o model pointtopoint contact (with a bilinear stiffness and a friction) and is applied t o the surfaces which may come into contact. Currently. the structure must deform so that the GRIDS attached t o the GAP deflect toward each other a distance Uo. when the GAP is closed and significant compressive and shear forces may be present. However. there is a large axial stiffness. For the GAP t o close..001 inch clearance between the shaft and the journal.5.0 inches long with a 2.5 inch thick flanges at each end. selection of the stiffness matrix update stategy. The journal is located 1.5 CONTACT ANALYSIS FOR THE SHAFTJOURNAL INTERFACE IN SOL 66 Introduction 11.2 Problem Description The model consists of a hollow steel shaft and a steel journal.10 inches thick with 0. SOL 66 and SOL 99.1 The GAP element is used to model the contact between surfaces. The shaft is 20. using MSCJNASTRAN Version 65.
The space between the shaft and the journal is 0.001.5. One element every 30 degrees is used in the circumferential direction. only half of it is modeled. The initial gap opening. These rigid body motions lead to singularities in the stiffness matrix. To avoid singularities. the journal and the gaps defining the contact surface. The rigid body motions are rotation about the yaxis and translation along the x and zaxes. The cylindrical coordinate system is used to define the xaxes of the GAPs t o be the TI or radial direction.0 inches from the free end.1 Finite Element Model Exposing GAP Locations The boundary conditions include constraining all 6 degrees of freedom (DOF) of each GRID point at the fixed end of the shaft. Rotational DOFs attached to the HEXA elements are also removed via SPCs. a soft spring is attached to the edge of the journal . Uo. With the GAPs open. By taking advantage of the plane of symmetry present in the structure. The constraints and boundary conditions are such that the journal is not restrained from the rigid body motion in three degrees of freedom. A cylindrical coordinate system is used to define 182 GRIDS for the 36 HEXA elements used to model the shaft and the 30 HEXA elements for the journal. The three fundamental components of the finite element model are the shaft. the stiffness between the shaft and journal is theoretically zero. one at each GRID location around the journal's flanges. Twentyeight GAP elements connect the shaft to the journal. This is a very coarse finite element mesh but adeqate for demonstration purposes.001 inches. Symmetric boundary conditions are specified along the edges on the symmetric plane for both the shaft and journal. is set t o 0. Locations Free EM X Figure 11.shaft acts radially inward and is located 5. Analyses are performed to determine the forces placed on the journal as the shaft deflects into the journal.
however.3 Selection of GAP Stiffness Kb must be large enough to prevent singularities in the initial stiffness matrix and small enough t o present realistic answers. 11. and Kb are: i Closed stiffness Open stiffness K. half of the 20000 lb force is applied t o GRID 10015 via a FORCE entry. Once the load is applied. the whole load is applied in 1 increment (INC = I on the NLPARM entry) in a single SUBCASE.2. The spring constant should be as small as possible t o avoid the singularity. Because there are no material or geometric nonlinearities in the model.68 x 10' lb/in. Due t o the symmetry condition. an increase in magnitude to 100. However. This CELAS2 scalar spring element elastically restrains translation in the zdirection.10 (Fo = Uo X K b ) .5.0 lb/in. When the GAPs closed and the stiffness matrix is updated with the larger stiffness value. Rc is the approximate Based on Eq. the preload. Fo on the PGAP entry. the stiffness of the flange portion of the journal in Fig. A K . K(shaft)] where K(journal) and K(shaft) are the local stiffnesses of the journal and shaft in the direction of the GAP axis at the GAP location. The radial stiffness of the cylinder subject t o the radial load can be estimated by where E is the elastic modulus. I is the area moment of inertia and radius t o the centroid. Using Roark and Young [11. the GAPs either open or close and applying the load in multiple increments has no affect on helping the solution to converge nor on the validity of the final solution.0 lb/in. To improve convergence. In estimating the local . K(shaft)] Kb < x min[K(journal). slower convergence or divergence. 11. value which is too small leads t o wrong answers.5.at GRID 30321 in the zdirection. the formulas for circular rings are used t o estimate the local structural stiffness.. K. > lo3 x max[K(journal). Experience shows that good choices for I. is satisfactory.14]. the CLOAD and LSEQ entries must be used to apply the loads t o upstream superelements. they remain singular until the contact occurs. K. A value of 10.37 x l o 5 lb/in. fails. thus a small nonzero Kb must be specified. the other two rigid body motions are taken care of. and for the shaft 1.5.2 is estimated to be 2.. A value that is too large may lead t o numerical illconditioning. 11. is set to 0. When superelements are used in SOL 66. Caution is also required in choosing the closed GAP stiffness.
34 x los and for the shaft is .5.2 Detail of Contact Region for Determining Local Stiffness Values Since two parallel rows of GAPs cover the contact region. F l rnge  R OD ~ournrl 'Shaft Figure 11.5. By using ASET entries for nonlinear DOFs. 11. Hence. the effective length of the shaft is assumed to extend 3 times the thickness of the shaft beyond each edge of the flange. and should be halved a t the plane of symmetry. A substantial reduction is made using static condensation via ASET entries. The estimate for the local journal stiffness is 1.1 Ii. Superelements are used for substructurillg in MSCINASTRAN. ASET entries are used for Guyan reduction. only stiffness terms from nonlinear .7 x lo5 lb/in.stiffness of the shaft. o x los while I{. half of the stiffness is attributed t o each row of GAPs and the estimates for the local stiffness values are divided by 2.5. the use of superelements (SEs) and ASET entries are implemented.4 Superelement and ASET To minimize computer time and turn around time for each analysis. The time t o perform the analysis is decreased by reducing the size of the stiffness matrix which is updated in SOL 66. from Eq. = 1 . 11.
The complement of the ASET entry is the OMIT entry. In the smaller model the ratio of nonlinear DOFs (Aset) t o linear DOFs (0set) is about 1:7. (Results for each model normalized t o 1. Normalized values of elapsed CPU time for analyses with and without ASETs and/or SUPERELEMENTS. only the residual structure is updated.5. The T 1 DOF from each GRID connected to each GAP are placed on ASET entries. when upstream superelements remain unchanged. t o restart with a new Ka value requires only to change the value of I<.) When restarts are performed with changes t o the GAP properties.1. specifically. The use of the ASET or OMIT entry is based upon convenience. The GAP'S axial DOFs are the only nonlinear variables in this problem. The nonlinear stiffness matrix is thereby reduced from 452 DOF t o 56 DOF.55 . not upstream SE 10. By placing the nonlinear elements in the residual structure (SE 0) and the linear elements in an upstream superelement (SE l o ) . The benefit from using superelements is realized in reanalysis. two models are compared.87 . a 66element model and a 900element model. the residual structure.89 .36 . DOFs t o be retained in the analysis set are placed on ASET entries and all others are automatically ommitted.1 compares run times of analyses done with and without static condensation via ASETs and with and without superelements. For instance.53 . Table 11.00 . In order t o emphasize the benefit that comes from reducing the size of the nonlinear stiffness matrix and using superelements.elements will be retained for nonlinear iteration. Table 11. In the big model the ratio of Aset to 0set DOFs is about 1:153.5. Elements 1. (The use of either the SESET entry and/or field 9 of the GRID entry t o place GRID points in superelements is a matter of convenience.) All DOFs in ASET by default 1.61 All DOFs in ASET by default Nonlinear GAP DOFs (T1)in ASET .59 66 Nonlinear GAP DOFs (T1)inASET .08 900 . on the PGAP entry and to perform an SEALL operation on SE 0.88 .06 No SEs GAP GRIDs in SE 0 GAP GRIDs in SE 0 (restart) No. only SE 0 requires reanalysis.OO . reducing substantially the time required for each iteration and stiffness matrix update.
T h e severe nonlinear effects (rapidly changing stiffnesses) make it difficult to obtain a converged solution. Inspection of the convergence factors and line search d a t a from . X. The results are compared in Fig.5 Analysis Procedure Several analyses are performed t o demonstrate how results vary depending on the value of li. Comparisons are done for two models: a 66element model with one element every 30 degrees ( a very coarse mesh). Variations of all three methods are used in several analyses and are tabulated in Table 115.5. Only axial G A P DOFs are in the nonlinear stiffness matrix via ASET entries.3. which control the stiffness matrix update strategy. a 132element model with one element every 15 degrees.5. = 10" and K b = 100.5. 2700 2600 2500 2400 u 66El m n t Model a  \ / I \ ow ' 2300 132El emant Model V I \ a  = aest p o ~ ~ i b s o l u t i o n le K recrma for efficiency 9 Ka X K. vs. axial G A P forces for 66element and 132element models T h e NLPARM entry provides for three types of nonlinear solution methods (ITER. T h e ITER method which updates the stiffness matrix after each iteration ( K S T E P = l ) is referred t o as full Newton's method. All cases use li. 11. K. for Best possible s o l u t i o n r e s u l t s tn "HI 4158 *ax r a t i o 1 7 Figure 11..2. It is considered that the best possible solution is achieved when the G A P forces become stationary with respect t o I<. SEMI and AUTO). which summarizes results from restarts of the residual structure. used.11.3.
At these locations the GAPS close and the stiffness changes from Kb to K.5. 12 13 14 15 16 ITER ITER ITER ITER ITER 2 2 1 1 0 0 stopped due to 2 diverging sols stopped due to MAXITER limit 0 stopped due to MAXITER limit converges in 63.6 Summary of Results The load deforms the shaft into an elliptical shape near the free end which then contacts the journal over a small area at a few locations shown in Fig.. . sec converges in 43.5.2 Variation of Iteration Strategy in Version 65 (Numbers in parenthesis or blank indicate default value.the nonlinear iteration output is required to make a judicious selection of NLPARM parameters for better convergence.5. Table 11.4. sec 11. 11.) Note that the performance is significantly improved in Version 67.
4 Sections from Fig. is as small as lo7.I.5. values increases as the mesh is refined. For the 132element model.5. which in this case is about 1 element every 10 degrees in the circumferential direction.1 uses 64% less CPU time when the nonlinear stiffness matrix is reduced via ASET entries to include only the essential nonlinear terms. = l o 7 is recommended for efficiency. = l o 7 gives a solution that is less than 1% different from the best possible solution. A value of I<. = 10".5. acceptable results are achieved with K . K .1 Indicating Which GAPS Close Fig. I(. The use of ASETs to reduce the problem size for the nonlinear iterations is strongly recommended if the model has high ratios of linear to nonlinear elements. . All three of the iteration methods determined identical solutions as shown in Table 11. = lo8 and the answers are less than than 5% off when K . Optimal iteration strategy varies with each individual problem. 11. The time to reach a converged solution was further minimized by restricting the number of line searches done. = 10'' while Ii. This trend continues until the "best" mesh is used. spanning six orders of magnitude provides results that differ by less than 1%. the full Newton's method is considered the most conservative and reliable method.3 shows that the 66element model converges to the best possible solution with I<. The large 900element model in Table 11. The best possible solution is achieved with K .5. However. The full Newton's method found the solution most efficiently (which is not the case in Version 67). Nevertheless. The best possible solution for the 132element model is achieved over a larger range of h'. Section AA Section 00 section Cc Section 35 @ = CLOSED Figure 11. hence recommended for efficiency. The range of acceptable K. = lo8 is considered the smallest value which provides a good solution.5. values. However. the default values on the NLPARM entry do not yield a converged solution. 11.2.
connected in series with the CELAS should be the predominant stiffness to resist the transverse loading. The analysis is of the same model but with two loading conditions. Friction coefficients of 0. SUBCASEs 2 and 3. is a line load along the edge of the journal acting in the +zdirection.5. K t .Superelements are especially effective when restarts are performed. lo4. At K t equal t o lo3. Before the full slip condition. However. as such. the entire analysis must be repeated. The journal slips completely a t 95 lbs with a total shear force equal t o about 1311 Ibs. As the remaining load increments are applied. a slightly higher load is required for slip. the shear force in each GAP equals K t (transverse stiffness from PGAP entry) times the shear displacement. the GAP'S shear DOFs (y and z) become nonlinear DOFs and. . With friction included. With Kt equal t o l o 4 and lo5. This is due t o the spring attached along the zaxis. the restart uses only 6% of the CPU as in the original analysis.7 Friction in the GAP The effects of friction in the GAPs are considered in a separate analysis. the same load produces slip. K t is an elastic stiffness (ideally rigid) used t o simulate the stationary condition under the critical load a t which slip occurs. l o 5 or higher. A parametric study using K t equal t o lo3. ratio of linear to nonlinear elements. iterations. additional load cases and NLPARM entries. however its stiffness (100 lb/in) provides an adequate stability t o the model. As such. At very large K t . slip occurs. These numbers will vary depending on factors such as: number of stiffness matrix updates. Without superelements.5. DOFs 2 and 3 for each GRID attached t o a GAP are placed on ASET entries (except where SPCed). need t o be included in the analysis set. All but one of the GAPs (no. The initial load closes the GAPs (SUBCASE I ) . 11. The load in SUBCASE 1 is the concentrated force. A 90 lb force is applied to each of the seven GRIDS along the edge of the journal. When the shear force is greater than the friction force. K t = lo4). In SUBCASE 3 the load is increased t o 97 lbs in 1 lb increments. when superelements are combined with ASETs for the large model. The second load. K t which is two orders of magnitude larger than the spring stiffness is proven sufficient (i.25 are used for the GAP y and zdirections. and lo5 found the minimum load at which full slip occurs. The data is nearly identical t o the data in Table 11. the second one loads the closed GAPs in shear (SUBCASE 2 and 3). the value of K t should be orders of magnitude larger than 100.e. the journal does not experience large displacements as would be expected in the full slip condition. The transverse stiffness..3 but with modifications to the PGAP and ASET entries.g. The CELAS spring used t o restrain the journal is not a structural member. e. For best efficiency. convergence is difficult t o achieve and caution is needed in selecting the loads and iteration parameters. 43) slip.
ID GAP. K UPDATE AFTER EACH ITERATION LABEL = ASETS AND SUPERELEMENTS $SET 111 = 0 $ FOR RESTARTS $SEALL = 111 $ FOR RESTARTS SEALL = ALL SUPER = ALL ECHO = BOTH SPC = 100 LOADSET = 950 $ ADDITIONAL OUTPUT REQUESTS $STRESS = ALL $ FOR UPSTREAM DATA RECOVERY $DISPLA = ALL $ELFORC = ALL $ FORCES IN GAP ELEMENTS SET 10 = 10 THRU 50 ELFORCE = 10 SUBCASE 1 CLOAD = 901 NLPARM = 2001 $ $ $ $ $ Remove ' I $ " from the following Case Control entries to include shear loads when friction effects are included in the Bulk Data as discussed in Section 7.5. CEND TITLE = KA = l. $SUBCASE 2 $ CLOAD = 902 $ NLPARM = 2002 $SUBCASE 3 $ CLOAD = 903 $ NLPARM = 2003 OUTPUT(PLOT) $ SHEAR LOADING FOR GAP ELEMENTS $ SHEAR LOADING FOR GAP ELEMENTS * * * BEGIN BULK $ $ USED TO DEFINE SHAFT. $ . KB = 100.3 MSCINASTRAN Input Data Listing for Case 16. JOURNAL AND THE RADIAL DIRECTION OF GAP AXES.. SUBTITLE = METHOD = ITER WITH 1 INC. APPNOTE $ BY BOB LOUWERS SOL 66 TIME 10 $ CONCENTRIC CYLINDERS CONNECTED BY GAPS.001.E+9.Table 11. SE MODEL. INITIAL GAP OPENING = 0.0.
0 0.+7 13 20511 30510 30520 20521 20611 30610 +MOO065 MAT1 13 PSOLID 12 . 2. 0 0.3 $ $ CGAPS DEFINE CONNECTIVITY $  USE COORD SYSTEM 1 T DEFINE GAP X AXES. 12 10110 10000 10001 10011 10010 10100 10101 +MOO000 CHEXA 100 +M00000 10111 CHEXA 405 +MOO065 30620 $ 12 20621 3. 0. 0.0 1 GRID $ $ $ 30621 1 2. 1 10 CHEXA ELEMENTS WHICH DEFINE SHAFT AND JOURNAL. 0. 10. GRID 10000 1 * * 0. $ $ GRID POINTS WHICH $ DEFINE SHAFT AND JOURNAL.75 180.10 $ $ SYMMETRIC BOUNDARY CONDITIONS $ SPCl 100 +MOO068 10006 246 10010 10000 10011 * SPCl 100 246 30600 30601 30610 30611 30620 30621 . 0. 0.CORD2C 1 +CORD 1.001 10614 PGAP 15 FO . O 20000 1 CGAP 10 15 10011 * CGAP $ $ 46 1 5 U O 0.
1000 975 CLOAD LSEQ $ FORCE $ $ $ $ $ 10000. 123456 10000 10310 10400 10010 10410 10100 10500 10110 10510 10200 10600 10210 10610 +MOO066 SPCl 100 +MOO06610300 $ $ $ PLACE AXIAL DOFS CONNECTED TO GAPS IN THE ANALYSIS SET. 0. 0. 901 950 1000 1.0 1. 975 10015 1. LSEQ REQUIRED FOR LOADING WHEN SUPERELEMENTS ARE USED.0 30321 3 CELAS2 76 CLOAD. 1 ASETl 10011 10012 * 10013 10014 * ASET1 $ $ $ 1 21100 21101 31100 31101 PLACE GRIDS ATTACHED TO GAPS IN THE RESIDUAL STRUCTURE. SPCl $ $ $ 100 456 10000 THRU 40000 BOUNDARY CONDITIONS FOR FIXED END OF SHAFT. NLPARM PROVIDES PARAMETERS FOR NONLINEAR ANALYSIS ITERATION STRATEGY. 100. INC 1 NLPARM 2001 KMETHOD KSTEP ITER 1 MAXQN MAXLS +NLP . 0 10011 10111 SESET * 10211 10311 10411 10511 10611 * SESET $ $ $ $ $ $ 0 30001 30101 30201 30301 30401 30501 30601 SOFT SPRING TO PREVENT RIGID BODY MOTION OF JOURNAL IN T3 DIRECTION.$ $ $ CONSTRAIN ROTATIONAL DOFS ON ALL SOLID HEXA ELEMENTS.
1. 0. 97. 0 1. ITER 0. O FORCE 1002 20010 90. 0. $ $ SUBCASE 2 L A S AND N P R CARD OD LA M $ $ CLOAD 902 1. 0. 1. 1. FORCE 1002 10015 10000. 1003 CLOAD LSEq FORCE FORCE 977 10000. 0. 977 10015 20010 1. O YES +NLP3 t h e following r e q u i r e d s h e a r DOFs when used i n SUBCASE 2 and 3 . * * * FORCE NP R LA M 1002 2002 20610 4 90. 0. 0. 1 0.0 1.0 1.+NLP 0 0 $ The following Bulk Data e n t r i e s d e f i n i n g t h e s h e a r loads and N P R e n t r i e s LA M $ f o r SUBCASE 2 and 3 a r e used when SUBCASE 2 and 3 a r e included. +NLP2 $ $ $ 0 0 SUBCASE 3 L A S AND N P R CARD OD LA M 903 950 1003 1003 1. 976 LSEq 950 976 1002 0. 0. 1 0 0.0 * * * FORCE 1003 20610 NP R L A M 2003 7 +NLP3 $ Remove t h e "$" t o include $ $ friction effects are $ $ASETI 3 10011 10012 $ASETI 23 10111 10112 97. ITER 0. 10013 10113 10014 10114 * * * $ASET1 $ASETl 23 3 20500 20600 20501 20601 30500 30600 30501 30601 .
10 $PGAP ENDDATA 15 Ka 1.E9 K b 1.25 muz 0.$ $ $ $ $ To include t r a n s v e r s e s t i f f n e s s and f r i c t i o n e f f e c t s f o r SUBCASE 2 and 3 .25 . Uo 0. r e p l a c e t h e above PGAP card with t h e following PGAP card.E2 Kt 1.001 Fo .E4 muy 0.
The velocity of particles at the end of the rod at the instant of impact (t = 0) is vo. Poisson's ratio p = 0. Determine the time history of stress and displacement at various cross sections of the rod during the impact. Young's Modulus v = 0. rod mass . Find the duration of the impact and the maximum stress in the rod for various values of mass ratio. Let vo be the initial velocity of the mass which is considered rigid. Rod mass = m L = 100.0 Kg. rod length A = 1. M / m .6.3.1 A rod with a fixed end is struck by a moving mass at the other end. mass density m = pAL = 10.0 m.0 m2.6 IMPACT ANALYSIS IN SOL 99: STRESS WAVE PROPAGATION IN AN ELASTIC ROD Problem Description 11.1 Kg/m3. crosssectional area of the rod E = lo3 N/m2.11.
we obtain [11. The general expression for the stress a t the fixed .2) Considering the force balance at the free end of the rod during the impact.6.2 ( m l M ) ( t l T ) Since the stress wave is reflected when it arrives at the fixed end of the rod. Let the time interval ( T )for the stress wave. be T= 2L C (11. where S o ( t ) = ooe .4) Solving the equilibrium equation for stress at the end struck in various time intervals. The initial compressive stress in the rod is 00 = 210 (11.6.6. traveling a complete cycle in the rod. the equilibrium eauation can be written as where A is the crosssectional area of the rod. As such.2 Theoretical Solution Let M be the mass of the moving body and m be the total mass of the rod. the stress in the wave is where c = determined by the ratio of the two velocities. Denoting the stress at the free end of the rod by a and the velocity of the body by v. we have is the velocity of sound in rod material.11.15] o = So(t) a = Sl(t)+So(tT) o=S2(t)+Sl(tT) for for for O<t<T T s t <2T 2T5t<3T etc. the stress value is twice the intensity of the incident wave.
4 Comparison of Results The instant when the stress a t the free end (or force in the GAP element) vanishes signifies the end of the impact.419 / 13. It was discovered. COUPMASS) is recommended for more stable responses.86 seconds in MSCINASTRAN) while the real departure of the mass body from the rod occurs at 6. A GAP element connects the CONM2 and the free end of the rod with a zero initial opening. A time step is chosen so that twenty vibration modes of the rod model can be adequately represented. that this appreciable discrepancy is caused by a limitation of the theoretical solution. The theoretical solution is valid only if the mass body stays in contact with the rod. 9 0 0 1 7.end during any time interval (12 t n)T 5 t < ( $ + n)T is therefore 11.01 % * The discrepancy is explained below. .7 seconds.6. however.0 % The MSCINASTRAN results given in the above table are obtained using the prescribed model of 20 CONRODs. MSCINASTRAN reveals that there is a momentary opening near the theoretical solution (5.4% at resonance) is introduced in the rod elements t o smooth out the undesired oscillations. A small damping (0. One CONM2 is used to represent the rigid mass.1 Duration of Impact (sec) I Theory % Difference 131168 F 7 0 8 r 5 .3 MSCINASTRAN Solution The rod is modeled with twenty CONRODs and a GAP element. The durations of impact for four distinct values of M/m are determined from MSCINASTRAN analyses and compared with the calculations of SaintVenant in the following table. Thus. The consistent mass matrix option (PARAM.6. we may conclude that MS.lm/sec is specified for both the CONM2 and the free end nodal point using TIC card.6. Table 11.C/NASTRAN predicts the true end of impact. and the mass value is varied for various ratios of m t o M. One end of the rod is constrained completely and the other is left free axially. With a more refined model. 11. 1 0. An initial velocity of vo = O.56 %* 0. They match the theoretical values closely except for M/m = 4.1 % / 0.
6.6. 11.6. SHL 5/28/87 $ TKT 5101985.5 Input Data Listing NASTRAN BUFFPOOL=37 NASTRAN SYSTEM(66)=112 INIT MASTER(S1 INIT SCRATCH(MEM=150) ID HAMMER. Illustrations in Fig. SHL 6/4/85 $ FOR VAX TIME 30 $ NONLINEAR TRANSIENT ANALYSIS SOL 99 $ PRINT MATRIX TRAILERS AND ITERATIONS DIAG 8. a stress vs.899 SET 3=99. TSTEPNL=20 . For each continuous and smooth curve segment. IC=I TSTEPNL=10 SUBCASE 2 $ UP TO 5 SEC. time in Fig.99 SET 2=101. for different values of Mlm. In general.2 are typical examples of the transient responses associated with the free end of the rod. element stresses are lower than the theoretical values due t o discretization error as expected. the MSCjNASTRAN result improves in agreement with the theoretical solution as it approaches the end of the segment.6. This shifting in time interval causes the discontinuity of vs. The maximum stress during impact always occurs at the fixed end. time plot for the element at the free end is given in comparison t o the theoretical one for the end point.3.Time histories of displacement at various grid points and stresses at various elements are obtained from MSC/NASTRAN analyses for several different values of M/m.V66 $ SSH 6/22/87.IMPACT SUBTITLE= BAR WITH FIXED END STRUCK BY A MOVING MASS AT THE OTHER END SET 1=21.6.50 CEND TITLE= TRANSIENT RESPONSE OF SHOCK WAVE IN BAR . 11. 11.1 for the mass point and the free end point clearly shows a good departure of the mass body. the element stress output can not reproduce the theoretical stress curve exactly at a grid point. It is shown that a refined model improves the results slightly.21 SEALL=ALL DISPLACEMENT=I VELOCITY=3 STRESS2 SUBCASE 1 $ UP TO 3 SEC. but in different time intervals. It should be noted that due to the stress gradients in both time and space. consistent with the theory. A plot of displacement vs. 11.120.1 and Fig. 11. In Fig.6. the slope of amax/ao M/m curve shown in Fig.2. 11.
0 0.0 0. 0.7.. 0. 0.. 0.0 0 .(GP99).0 0.0 0..0 GRID 7 30.0 0.0 GRID 9 40.0 0. 0.0 0.. 0. 0.0 0.3 XAXIS=YES YAXIS=YES XTITLE=TIME YTITLE=DISPLACEMENT TCURVE= DISP.2....0 10...0 GRID 18 85.0 0. GRID 11 50.... 0. 0... 0.0 0.0 GRID 12 GRID 13 60.. GRID 17 80...21(T1) YTITLE=STRESS TCURVE= STRESS AT FREE END (ELEMENT 120) XYPLOT STRESS /120(2) TCURVE= STRESS AT FIXED END (ELEMENT 101) XYPLOT STRESS /I01(2) TCURVE= FORCE IN THE GAP (ELEMENT 899) YMAX=2.FREE END(GP21) XYPLOT DISP /99(T1) .O GRID 10 45. 0.0 0 . 0..0 0.0 0.0 0. 0. 0.0 GRID 5 20..0 0.0 0.0 GRID 20 95.0 0....5.... GRID 1 0.0 0.0 GRID 16 75.4. XYPLOT STRESS /899(2) BEGIN BULK $.0 0.0 $ ROD ELEMENTS ....0 GRID 21 GRID 99 100..3. 0....$SUBCASE 3 $ TSTEPNL=30 OUTPUT(XYPLOT) CSCALE=1...0 GRID 15 70.0 0.0 55...0 0.0 100. 0.0 0.0 GRID 8 35.(T1) AT MASS PT.. 0...O GRID 19 90.0 0.6.... 0.0 GRID 2 5. 0.0 0.0 GRID 6 25.0 GRID 14 65.0 GRID 3 GRID 4 15.
025 TSTEPNL 30 75 0.010 ENDDATA AUTO AUTO AUTO .E+3 PARAM W4 1.025 TSTEPNL 20 80 0.CONROD 101 1 2 CONROD 102 2 3 CONROD 103 3 4 CONROD 104 4 5 CONROD 105 5 6 CONROD 106 6 7 CONROD 107 7 8 CONROD 108 8 9 CONROD 109 9 10 CONROD 110 10 11 CONROD ill 11 12 CONROD 112 12 13 CONROD 113 13 14 CONROD 114 14 15 CONROD 115 15 16 CONROD 116 16 17 CONROD 117 17 18 CONROD 118 18 19 CONROD 119 19 20 CONROD 120 20 21 $ GAP ELEMENT CGAP 899 90 21 PGAP 90 $ MASS ELEMENT CONM2 999 99 $ INITIAL CONDITION 1 99 1 TIC 1 21 1 TIC $ MATERIAL PROPERTY OF ROD MAT1 100 1 .570796 PARAM COUPMASS1 $ TIME STEP DEFINITION TSTEPNL 10 120 0.
1.6.Figure 11. . Displacement vs. Time for M l m = 1.
6.Figure 11.2 Comparison of Free End Stress (Time History for M / m = 1) .
6. Comparison of Impact Stresses..6 . 11. 4 .3.5. A 'Exact' Solution A 20 CONROO Yodel Refined Yodel Mass R a t i o M/m Figure 11.9 .
.
an artificial stiffness Ke. = rotation of a GRID from global displacement 0. 11. The differential stiffness develops as the deformation proceeds.7.16]. . is provided for the normal rotation in the QUAD4 and TRIA3 shell elements [11. but they are not equipped with nonlinear features. rotation measured in the element z where G is the shear modulus.1 Stiffness is not defined for the rotation normal to the surface of shell elements (QUAD4 and TRIA3). and J is the Jacobian. Unlike linear analysis. and the shell structure becomes progressively stiffer due t o the nonlinear geometric effect.2 Difficulties in the Thin Shell Model The thin shell structure is flexible in the lateral direction a t the beginning of the analysis.11. proportional to the userspecified parameter KGROT.7 EFFECTS OF KGROT ON GEOMETRIC STIFFENING OF THE THIN SHELL MODEL Theoretical Basis for Parameter KGROT 11. The pseudo stiffness Ke is added by the parameter K6ROT to the relative rotation of the element t o grid rotation as follows: I<* for with (a. Consider the stiffness of a (L x L x t ) quadrilateral element: kb 0: L2 Et3 and k cc E t . In order t o remove the singularity in the outofplane rotation (the sixth DOF). = 5 I (av .$). this degree of freedom can not be constrained in the geometric nonlinear analysis. t is the thickness. Insensitive to the mesh size. representing the area of the element. No effect on the rigid body rotation. .7. It is noted that QUADR and TRIAR elements have stiffness in this direction. The requirements for Ke are The unit should be in moments. Geometric stiffening of thin shells often poses difficulties in convergence during the initial loading stage. = * G * t * I JI * K6ROT 0. Pass the constant strain patch test. The difficulties arise from illconditioned stiffness due t o a very large ratio between bending and membrane stiffnesses. The value should be small enough t o maintain accuracy.8) .
)/ cos 8.3 Parametric Studies Four problems are examined t o study the effect of the parameter K6ROT on the iterative process of thin shell structures subjected t o geometric stiffening. Recommendations t o alleviate the convergence difficulties when the gimbal angle exceeds 90 degrees are: r If the physical problem specifies the rotation of 8. sin 8.57 radian (equivalent t o 90') for rotation. The large rotation could easily involve a larger displacement than 1. A8.cos 8. At). can favorably affect the iterative process. The gimbal angle for large rotations is applicable to the QUAD4 and TRIA3 elements. a smaller load increment with a proper value of KGROT should be used. and causes excessive normal rotation. sin 8. If l8.l 90' and the solution diverges. This limitation has been eliminated in Version 67 by introducing a set of auxiliary angles t o avoid division by cos 90'. the lateral pressure on the thin shell causes excessively large rotations due to bending of the initially soft stiffness. = (A8. = AO. / k b ) is dictated by ( L / t ) 2 . the user warning message "Gimbal Angle Exceeds 90 Degrees" is issued.A8. which provides a stiffness for the normal rotation. beyond 90°. 68. which causes singularity in the gimbal angle calculation. rotation of the thin shell structure propagates to the sixth DOF. As such. the coordinate system should be changed so that 8. sin 8. = A@. > The divergent process can be contained by using a proper value of KGROT. the gimbal angle is accumulated as 68. [(AO. This is because excessive 0. 68. even though there is no normal rotation in the physical problem.] + + + * sin 8. r If the problem diverges due t o numerical condition. 11. where A8 is the incremental rotation in the global coordinate system and S is the net increO mental gimbal angle.l 2 90°. The problems are r QUAD4 model of rectangular plate with line load r TRIA3 model of rectangular plate with line load r QUAD4 model of square plate with uniform pressure QUAD4 model of square plate with concentrated load . 5 90' (this limitation is removed in Version 67).)/ cos O. A mathematical singularity occurs when the incremental gimbal angle component 8. cos 8. KGROT. The problem is most likely aggravated beyond the occurance of (8. = zf90°. .7.Thus the membrane t o bending stiffness ratio ( k . This is usually caused by numerical illconditioning during the nonlinear iterations rather than by physical rotation exceeding 90'. cos 8. At each iteration.
2.7. iterations. 11.. Figure 11. The plate is modeled by 8 QUAD4 elements with linear elastic material. 11.e. The line load a t GRID 2. The simply supported boundary condition is imposed by constraining components 1. The performance (i. The effects of geometric nonlinearity are included in the analysis by specifying PARATVI. the number of stiffness updates. The convergence and efficiency characteristics of the solution process are observed in order t o determine the optimal range of values for KGROT. The plate is simply supported at the 200 mm sides and is subjected t o a line load of 400 N/mm acting in the 2 direction as shown in Fig. The input d a t a are given in Table 11. and line searches) of the successful executions is presented for the corresponding KGROT value.SOL 66 in Version 66 is used in the analysis of the problems. and 3 of the GRID points at the 200 mm sides of the plate. To obtain an even distribution of converging and diverging solutions for each problem.7.1 for K6ROT = lo4.5 mm. the value of K6ROT is varied from 1 to lo9.1. and 8 is specified using FORCE and LOAD Bulk D a t a entries. The main feature of these problems is geometric nonlinearity due t o geometric stiffening.7.1 QUAD4 Model of Rectangular P l a t e w i t h Line Load. LGDISP. .4 QUAD4 Model of Rectangular Plate with Line Load A rectangular plate has dimensions of 200 mm x 400 mm and a thickness of 0.l. 5.7. incremented by a factor of 10 in each run.
RECTANGULAR PLATE WITH 8 QUAD4 ELEMENTS AND LINE LOAD SEALL = ALL DISP = ALL OLOAD = ALL SUBCASE 1 NLPARM = 10 LOAD = 100 SUBCASE 2 NLPARM = 20 LOAD = 200 OUTPUT(PLOT) CSCALE 1. VIEW 30.S.Table 11. 20.E+4 PARAM LGDISP I $ NLPARM 10 NLPARM 20 $$ GEOMETRY 1 GRID GRID 2 GRID 3 GRID 4 GRID 5 GRID 6 7 GRID GRID 8 GRID 9 GRID 10 GRID 11 GRID 12 GRID 13 GRID 14 GRID 15 4 5 AUTO AUTO YES YES ... TLL 4/21/89 TIME 5 $ CPU MINUTES 66 $ NONLINEAR STATIC ANALYSIS SOL 50 DIAG CEND TITLE = S. SET 1 ORIGIN 1 SYMBOL 1 LABEL GRID BEGIN BULK $$ PARAMETERS PARAM KGROT l.1 Input Data Listing for QUAD4 Model with Line Load ID TKGROTI. 0. V66 $ SHL 3/16/90.7.3 PLOTTER NAST SET 1 = ALL MAXI DEFO = 1. FIND SCALE ORIGIN 1 SET 1 PTITLE = ISOMETRIC VIEW PLOT SET 1 ORIGIN 1 SYMBOL 1 LABEL GRID PLOT STATIC 0 MAXI DEFO 1.
07E+5 PSHELL I 10 $$ LOADING FORCE 100 2 FORCE 100 5 FORCE 100 8 LOAD 200 I. ENDDATA .$$ CONNECTIVITY CQUAD4 I I CqUAD4 2 I CQUAD4 3 I CqUAD4 4 I CQUAD4 5 1 CQUAD4 6 I CQUAD4 7 I CqUAD4 8 I $$ PROPERTIES MAT1 10 2. $ I 2 4 5 10 11 14 15 .5 10.
and lo3. For this problem.2. The loaddeflection curve of the plate at the center (GRID 8) is given in Fig. the solution diverges at the first load verges only for KGROT = lo4.2 Results of Parametric Study for KGROT: QUAD4 Model with Line Load ( K Updates/Iterations/Line Searches ) KGROT INC LOAD 1 10 lo2 lo3 104** u8 Notes : 1. ** indicates the most effective case. Discrepancies in U8 with different values of K6ROT is less than 0. For the remaining KGROT values of 1. 10. 3.7.0003% at *.2 summarizes the results of the KGROT parametric study. the optimum value of KGROT is lo4. . The analysis stops for KGROT = l o 2 because the maximum number of iterations is reached in the first increment of the second subcase. The solution conlo5. For K6ROT increment.7.7. Solution diverges a t the first increment beyond K 6 ~ 0 T = 1 0 ' . 2. > Table 11. the analysis stops due to a diverging solution. 11.Table 11.
~ " " ~ " " l " ' ' l l ' ' ' l l ' >L L " 1 " " 1 ' jL El 5 1D 15 CI d W  3U C)e f 1 ec t i or1 t 8 I IIIIII Figure 11. . .6.2 LoadDeflection at Center of QUAD4 Model with Line ~ o a d . 0 0 n J a CS na  m 0 " * 7 C (D _  6) m . a *. mm 63   63 L a\ E aEm m COW c.7. ~ .
5 mm. 5. Figure 11. because the solution did not converge with the original loading of 400 N/mm.z direction. The simply supported boundary condition a t the 200 mm sides of the plate is imposed by constraining components 1. and 8 is specified using FORCE and LOAD Bulk Data entries. T h e loading had t o be reduced. 2.3 shows the input d a t a for KGROT = lo2. The plate is modeled by 16 TRIA3 elements with linear elastic material. The line load a t GRID 2.5 TRIA3 Model of Rectangular Plate with Line Load This problem was chosen t o be identical to the previous one.3 TRIA3 Model of Rectangular Plate with Line Load. but with TRIA3 instead of QUAD4 elements in the model. The plate is simply supported a t the 200 mm sides and is subjected t o a line load of 140 N/mm acting in the .7. Table 11. however.7.11. . 11.3 shows the rectangular plate with dimensions of 200 mm x 400 mm and a thickness of 0.7. Fig. and 3 of the GRID points.7.
VIEW 30..E+2 PARAM LGDISP I $ NLPARM 10 NLPARM 20 $$ GEOMETRY I GRID GRID 2 GRID 3 4 GRID GRID 5 6 GRID GRID 7 GRID 8 GRID 9 GRID 10 I1 GRID GRID 12 GRID 13 GRID 14 GRID 15 4 5 AUTO AUTO I I 50 50 YES YES ..Table 11. 0.3 PLOTTER NAST SET I = ALL MAXI DEFO = 1.3 Input Data Listing for TRIA3 Model with Line Load ID TK6ROT2. RECTANGULAR PLATE WITH 16 TRIA3 ELEMENTS AND LINE LOAD SEALL = ALL DISP = ALL OLOAD = ALL SUBCASE I NLPARM = 10 LOAD = 100 SUBCASE 2 NLPARM = 20 LOAD = 200 OUTPUT (PLOT) CSCALE 1.7. SET I ORIGIN I SYMBOL I LABEL GRID BEGIN BULK $$ PARAMETERS PARAM K6ROT 1 . 20. FIND SCALE ORIGIN I SET I PTITLE = ISOMETRIC VIEW PLOT SET I ORIGIN I SYMBOL I LABEL GRID PLOT STATIC 0 MAXI DEFO I.S. V66 $ CSK 3/16/90 TIME 5 $ CPU MINUTES $ NONLINEAR STATIC ANALYSIS SOL 66 DIAG 50 CEND TITLE = S.
5 $$ LOADING FORCE 100 2 FORCE 100 5 8 FORCE 100 LOAD 200 I. 10. $ ENDDATA .$$ CONNECTIVITY CTRIA3 I I I CTRIA3 2 I I CTRIA3 3 I 10 CTRIA3 4 I 10 CTRIA3 5 I 2 CTRIA3 6 I 2 CTRIA3 7 I II CTRIA3 8 I II CTRIA3 9 I 13 CTRIA3 10 I 13 CTRIA3 I1 I 8 CTRIA3 12 I 8 CTRIA3 13 I 12 CTRIA3 14 I 12 CTRIA3 15 I 7 CTRIA3 16 I 7 $$ PROPERTIES MAT1 10 2.07E+5 PSHELL I 10 .
l o 2 . Discrepancies in U8 with different values of KGROT is less than 0. iterations. ** indicates the most effective case.011% at *. The loaddeflection curve of the plate a t the center (GRID 8) is given in Fig. 3.4 Results of Parametric Study for KGROT: T R I A 3 Model with Line Load ( K Updates/Iterations/Line Searches ) K6ROT INC LOAD 1 1o3 10 lo2** lo4 lo5 u 8 Notes : 1. lo3.4.The results of the KGROT parametric study are summarized in Table 11. the optimum value of KGROT is l o 2 because it is the most efficient in terms of the number of stiffness updates. 2. Solution diverges a t the first increment beyond K6ROT=106.7.7.4. In this problem. . The analysis stops due t o a diverging solution a t the first load increment for K6ROT lo6 and the remaining K6ROT values of 1 and lo4. and line searches. The solution converges for K6ROT values of 10. 11. and lo5.7. > Table 11.
Ce l f lect iutI i l l 8 k 111111 I Figure 11.7. .4 LoadDeflection at Center of TRIA3 Model with Line Load.
The plate is modeled by 16 QUAD4 elements with linear elastic material.5 QUAD4 Model of Square P l a t e with Uniform Pressure.5 mm. The plate is simply supported a t three of its sides and is subjected t o a uniform pressure loading of 1 N/mm2 applied normal t o its plane. and 3 of the G R I D points a t three sides of the plate. The simply supported boundary condition is imposed by constraining components 1.7. Figure 11.11.7.13 . The uniform pressure loading is specified using PLOAD2 and LOAD Bulk Data entries.7.7 . 11. 2.5 for K6ROT = lo3. The input d a t a are given in Table 11.6 QUAD4 Model of Square Plate with Uniform Pressure Fig.5 shows a square plate which has a length of 400 mm and a thickness of 0.7. 11.
S..Table 11.3 PLOTTER NAST SET I= ALL MAXI DEFO = 1. BEGIN BULK $$ PARAMETERS PARAM K6ROT I . 20.5 Input Data Listing for QUAD4 Model with Uniform Pressure ID TK6ROT3. V66 $ RA 3/7/90 TIME 5 $ CPU MINUTES 66 $ NONLINEAR STATIC ANALYSIS SOL DIAG 50 CEND TITLE = S..E+3 PARAM LGDISP I $ NLPARM 10 NLPARM 20 $$ GEOMETRY 1 GRID GRID 2 GRID 3 4 GRID GRID 5 GRID 6 7 GRID GRID 8 GRID 9 GRID 10 GRID 1 1 GRID 12 GRID 13 GRID 14 GRID 15 4 5 AUTO AUTO YES YES . 0. VIEW 30. FIND SCALE ORIGIN I SET I PTITLE = ISOMETRIC VIEW PLOT SET I ORIGIN ISYMBOL ILABEL GRID SET I ORIGIN 1 SYMBOL I LABEL GRID PLOT STATIC 0 MAXI DEFO I.7. SQUARE PLATE WITH 16 QUAD4 ELEMENTS AND UNIFORM PRESSURE SEALL = ALL DISP = ALL OLOAD = ALL SUBCASE I NLPARM = 10 LOAD = 100 SUBCASE 2 NLPARM = 20 LOAD = 200 OUTPUT(PLOT) CSCALE 1.
$$ CONNECTIVITY CQUAD4 I I I CQUAD4 2 I 2 CQUAD4 3 1 3 CQUAD4 4 I 4 CQUAD4 5 1 6 CQUAD4 6 I 7 CQUAD4 7 I 8 CQUAD4 8 I 9 CQUAD4 9 I II CQUAD4 10 I 12 CQUAD4 I1 1 13 CQUAD4 12 I 14 CQUAD4 13 1 16 CQUAD4 14 I 17 CQUAD4 15 I 18 CqUAD4 16 I 19 $$ PROPERTIES MAT1 10 2.I I LOAD 200 I. $ THRU 100 ENDDATA . GRID 17 100.5 $$ LOADING PLOAD2 100 .07E+5 PSHELL I 10 . GRID 20 400. 21 0. GRID 23 200. GRID GRID 22 100. 10.GRID 16 0. GRID 24 300. GRID 18 200. GRID 25 400. GRID 19 300.
Discrepancies in UZ3with different values of K6ROT is less than 0. 2. ** indicates the most effective case.7. Solution diverges at the first increment beyond K 6 ~ 0 T = 1 0 ~ .Table 11. 3.7.6 summarizes the results of the K6ROT parametric study.0087% at *.6 Results of Parametric Study for K6ROT: QUAD4 Model with Uniform Pressure ( K Updates/Iterations/Line Searches ) K6ROT INC LOAD 1 10 lo2 lo3** lo4 lo5 u23 Notes : 1. The solution diverges at the third load increment of the second subcase for K6ROT = lo5 and at the first load increment of the first subcase for K6ROT 2 lo6. the solution converges. 11. Table 11. The loaddeflection curve of the plate at GRID 23 is given in Fig.6.7. The optimum value of K6ROT is lo3 in this problem. For the remaining K6ROT values of 10 through lo4. . The analysis stops at the first load increment for K6ROT = 1 when the maximum number of iterations is reached.
Dei I E : C t I C I ~ I (IIITII 1 Figure 11. .7.6 LoadDeflection at Grid 23 of QUAD4 Model with Uniform Pressure.
The concentrated load a t GRID 19 is specified using FORCE and LOAD Bulk D a t a entries. but has a different loading condition.7 .7. and 3 of the GRID points at three sides of the plate. 11.7.11.7 QUAD4 Model of Square Plate with Concentrated Load This problem is similar to the previous one.7 QUAD4 Model o Square Plate with Concentrated Load f 11. The simply supported boundary condition is imposed by constraining components 1.4 mm.7. A square plate has a length of 400 mm and a thickness of 0.18 .200 N acting in the 2 direction as shown in Fig.7. T h e plate is simply supported on three sides and is subjected t o a concentrated load of 13. 2. The plate is modeled by 16 QUAD4 elements with linear elastic material. Table 11. Figure 11.7.7 shows the input d a t a for K6ROT = lo3.
SQUARE PLATE WITH 16 QUAD4 ELEMENTS AND CONCENTRATED LOAD SEALL = ALL DISP = ALL OLOAD = ALL SUBCASE 1 NLPARM = 10 LOAD = 100 SUBCASE 2 NLPARM = 20 LOAD = 200 OUTPUT(PLOT) CSCALE 1.E+3 PARAM LGDISP I $ NLPARM 10 NLPARM 20 $$ GEOMETRY I GRID GRID 2 GRID 3 GRID 4 GRID 5 GRID 6 GRID 7 8 GRID GRID 9 10 GRID GRID 11 GRID 12 GRID 13 GRID 14 GRID 15 4 5 AUTO AUTO YES YES . V66 $ RA 3/7/90 TIME 5 $ CPU MINUTES 66 $ NONLINEAR STATIC ANALYSIS SOL 50 DIAG CEND TITLE = S. FIND SCALE ORIGIN 1 SET I PTITLE = ISOMETRIC VIEW PLOT SET I ORIGIN I SYMBOL I LABEL GRID PLOT STATIC 0 MAXI DEE0 I. 0.Table 11. VIEW 30.3 PLOTTER NAST SET I = ALL MAXI DEFO = I. 20. SET I ORIGIN I SYMBOL I LABEL GRID BEGIN BULK $$ PARAMETERS PARAM K6ROT l.7.7 Input Data Listing for QUAD4 Model with Concentrated Load ID TK6ROT4...S.
GRID 23 200. GRID 21 0. GRID 19 300. $ ENDDATA . 6. GRID 17 100.4 $$ LOADING FORCE 100 19 LOAD 200 I. $$ CONNECTIVITY CQUAD4 I 1 I CQUAD4 2 1 2 CQUAD4 3 I 3 CQUAD4 4 I 4 CQUAD4 5 I 6 CQUAD4 6 I 7 CQUAD4 7 1 8 CQUAD4 8 I 9 CQUAD4 9 I II CQUAD4 10 I 12 CQUAD4 I1 I 13 CQUAD4 12 1 14 CQUAD4 13 1 16 CQUAD4 14 1 17 CQUAD4 15 I 18 CQUAD4 16 I 19 $$ PROPERTIES MAT1 10 2. GRID 22 100. GRID 25 400. GRID 18 200. GRID 24 300. GRID 20 400.07E+5 PSHELL I 10 .GRID 16 0.
7.The results of the K6ROT parametric study are summarized in Table 11.7.8.095 Notes : 1. *. Table 11.8 Results of Parametric Study for KGROT: QUAD4 Model with Concentrated Load ( K Updates/Iterations/Line Searches ) K6ROT  INC 11 12 LOAD 550 1100 1 MAXITER 10 lo2 lo3** 3/39/16 1/15/1 lo4 4/38/20 1/16/1 lo5 MAXITER u9 1 3/38/18 1/15/1 4/38/18 1/15/1 6. The analysis stops at the first load increment for K6ROT = 1 and K6ROT = l o 5 because the solution does not converge in MAXITER iterations.4118 8. and lo4. lo3. . The solution converges for K6ROT values of 10. 11.8. Discrepancies in U19 with different values of K6ROT is less than 0. The loaddeflection curve of the plate a t GRID 19 is shown in Fig. Solution diverges a t the first increment beyond K 6 R 0 ~ = 1 0 ~ . The optimum value of K6ROT is also lo3 in this problem. ** indicates the most effective case. For K6ROT 2 lo6. lo2. the analysis stops at the first load increment due to a diverging solution.7. 3. The results are similar t o those of the previous problem of a square plate with a uniform pressure.12% at 2.
.7.Figure 11.8 LoadDeflection at Grid 19 of QUAD4 Model with Concentrated Load.
however. Bending of the initially soft plate causes excessive rotation which propagates t o the sixth DOF. the plate becomes progressively stiffer due to the nonlinear geometric effect. the solution did not converge for KGROT = 1 in any of these test problems. . The parametric study thus indicates that the iterative process in geometric stiffening is highly sensitive to KGROT. Values of KGROT that are too large ( 2 lo6) also cause the solution to diverge in all of the problems. Since an excessively large value of K6ROT will lock the varying strain within an element by enforcing the rotation measured in the element (R. As the deformation increases. small values of KGROT (< lo2) make the solution convergence very difficult.) to conform t o the GRID point rotation (O. The optimal range of values is found to be lo2 2 KGROT 2 lo4. This excessive rotation in the sixth DOF is alleviated by a larger value of KGROT.7.8 0bservations At the beginning of the analysis. In fact.11. the plate is flexible.). the solution will be inaccurate even if it converges with a large KGROT value. whose proper value can contain the divergent process. Therefore.
.
The least square fit is described t o determine rheological parameters (Cs.11.Cp and Ii. 11. some kind of curve fitting is required t o find equivalent coefficients or parameter values corresponding t o the predefined empirical law (types 111 through 222) or the MaxwellKelvin rheological model.) at a constant stress ( a ) for the MaxwellKelvin rheological model represented by which should simulate the behavior of plastics expressed by the Findley's law. so that the creep strain is represented by If the user should model a material that is characterized by experimental data or by a creep law different from what is provided in MSCINASTRAN. Cp and K. users need t o define the creep characteristics either by empirical creep laws in the form of with or by defining the rheological creep parameters (C. An interactive program is developed t o perform this numerical analysis procedure.8.8.8 ADAPTATION OF CREEP ANALYSIS CAPABILITY TO GENERAL VISCOELASTIC MATERIALS 11.11.18].1 Introduction MSCINASTRAN's creep capability was developed based on the MaxwellKelvin rheological model.at E . which can be applied t o any viscoelastic materials [11. u cf = EO sinh 00 u u + &ttnsinh . This procedure should be followed in any similar application of the general viscoelastic or creep analysis... Cp and ICp) as functions of stresses. To use this capability..2 Formulation for Least Square Fit of Rheological Parameters Let us determine the parameter values (C.17. as functions of stresses) corresponding t o the Findley's law representing the creep behavior in plastics as an example.
the following condition has to be satisfied.e. a on the reference curve in Eq.z In order t o satisfy the minimum error in Eq. i....e.4) is obtained by a Taylor series with a substitution of Eq.4) should be linearlized with respect to C.8. i.8. i. the least square fit can be employed t o minimize the error. = at = 8500 psi E = 1. c.Cp and I<p apply an iterative and to incremental process..2). AC. sj = &$ .5).= 0. (11.. min. error = where x 3 6 .3). i..0011 a. (11.e.8. (11.&3f Eq.19 EO = 0. (11.0067 Et = 0. c. where 8eT at .8.. ac.8. .2 x lo6 psi n ) : Using the data points (tj.e. + The first order approximation of Eq. = c. (11.
The rheological parameters may be evaluated by the least square fit of the strainrate instead of the strain itself.8. (11. and A& as unknown variables. j h. namely. (11.lle(Kp/Cp)t c. and Air'. ACp + a&.air 11.( K ~ / c ~ ) t Ot c P 2 Computational Process The strain function in Eq. 8iT + PAC. Eqs.+ 0. (11.1).7). s (11.e .2b) (11. (11.1) and (11. which requires curve fitting. i.8.)] where the derivatives are defined in Eq.8. acp atT + A& ari.4). and E~ and ~f in Eqs. (11.8.1).7) should be replaced by the following equations: d = .2). (11.1b) (11. (11.2).6) and (11.When Eq. CS~ d A[ A t . By solving the simultaneous equations in Eq.8.8. ac. (11.8. respectively. has two variables.8. there are three linear equations with AC.8. (11. AC. (11.8.8.8..8.8.6) is substituted into Eq.8). which .e. these unknown parameters can be updated by Eq. (11. i A&] EK = 8" j C[(E~ ac.. (11.8..6b) (11.8. AC.4b) if = wttn' sinh dt 0 6 = iT if 6 ? ? iT i*+ PAC..7b) 0 ."CP a~r.9) for AC.  acs aiT . The experimental data are usually measured at constant stress levels.( K ~ l C ~ ) t e c c. ac.8.8.8. stress and time.5) for the next iteration. For this approach..3 acP aiT  aiT ac. .
(11. t = 10000 hours and a = 1000 psi .8. for C. and K. Newton's iteration starts to minimize the error in Eq. (11.9) are solved at every iteration and the iteration continues until the convergence is achieved.. the least square fit can be performed as formulated in the preceding section with data points in the operating time domain while the stress level is fixed. Cp cannot be evaluated by Eq. The least square fit is performed incrementally by sweeping along the entire operating stress range. Initial values should not be too far from the solution for a good convergence.12) because if approaches infinity at t = 0. C. i. e..3). Instead. (11.g.e. The simultaneous equations in Eq. Cp may be estimated by using lip. Therefore. Upon convergence.8. and K.526 x 1000 .1427 x lo5 psi However. N 1.8.. The initial values can be appropriately estimated by considering a physical interpretation of the creep strain curve in relation to the rheological model in Eq.982 x = 8.. (11.2. a set of rheological parameters (C. The iteration starts with initial values for Cs.become the basis for the empirical formula or the curve fitting.. C.1). and ~f a t some finite time span.) has been determined for a constant stress.8. The slope at a long time span ( t f ) approximates the secondary creep rate. from which The total primary creep strain a t a long time span can be approximated by from which  The slope at t=O can be related t o Cp by from which For a = 1000 psi and tf = 500000 hours Ii.
8. strain data (default = RHEOLOGY. Title t o appear in the output (no default). This FUNCTION has to be .5 show plots of rheological parameters saved in the file "RHEOLOGY. has been coded. Fig. Upon completion of the dialogue. Error tolerance for the convergence criterion (default = 4. which is an interactive program. as presented in the preceding sections. Maximum operating stress (default = 1500) and stress increments (default = 100) 6.8. 11.8.4 Analysis Procedure and Results The computation process. Output file names for time vs.PAM).8.8.8. The DISPLY program starts by prompting questions as shown in Table 11. The results of processing the file "RHEOLOGY. Table 11. 11.3.8.5 shows FUNCTION EPSF defining the Findley7s creep law in Eq.1.8.5 shows the source listing of this program. The program prompts six questions t o be answered as shown in Table 11.8.2 and 11.8.PAMV in Table 11.8.8.8. 11. Least square fit on the strain curve or strainrate curve (default = strain curve). The listing in Table 11. the iterative and incremental computation process proceeds with the iteration information printed on the terminal screen as shown in Table 11.2) and its time derivative. 3. Table 11.5 Recommendations The rheological parameters corresponding t o any empirical creep law may be obtained by using the program "RHEOLOGY". which can be converted into MSC/NASTRAN data specified on the CREEP and TABLES1 Bulk Data entries. The least square fit on the strainrate curve was attempted t o determine the rheological parameters.OUTV in Table 11. and 11. Upon termination of the job.2 are shown in Fig. This approach was not fruitful because the Hessian matrix became singular within the operating time domain.3.8. there will be two output files as shown in Tables 11.4.4.11.8.3.2. Cp and Kp) as functions of stress. 2. (11. 11. These questions pertain to: 1. rheological parameters (default = RHEOLOGY. Table 11.6 shows the source listing of the DISPLY program (to be linked with Tektronix PLOT10 package).8. Maximum time for creep (default = 5 x lo5) and the number of data points to be used for curve fitting (default = 200). The analysis starts by running the program "RHEOLOGY".. An interactive graphics program (DISPLY) is available t o plot curves for the data generated by the program RHEOLOGY. respectively.OUT) and stress vs.8.1.1 and 11.8.8. 5.3 shows rheological parameter values (C.
It is also possible t o use experimental data for the curve fitting in place of the empirical law with slight modifications t o the current version of "RHEOLOGY". The least square fit can also be applied to the creep laws defined by types 111 through 222 to be specified in the CREEP Bulk Data entry. This means that all the data points obtained by sweeping through the time domain. nested under the stress increments across the operating stress range should be included in the iterative process for the least square fit. In this case. . by the curve fitting. rather than the rheological parameters. the time and the stress must be treated as a unified variable. For those creep laws. coefficients "a" through "g" should be determined.redefined for a different type of empirical formula.
7227526E+11 3 4.000000E+02 ERROR 2.1 Run Progression of the Program "RHEOLOGY" $RUN RHEOLOGY * PERFORMS LEAST SQUARE FIT FOR RHEOLOGICAL PARAMETERS ** WRITTEN BY SANG H.OUT TYPE FILE NAME FOR STRESS VS. CS/CP/KP.078400iE06 1.7770010E09 ITERATION CS 0 1.3207905E09 4.1390349E+iO 8.6803444E+12 1.5875800E+iI 4 6.4180528E06 1.5539796E+ii 2 LEAST SQUARE FIT AT STRESS 2.E6) I. DEFAULT=RHEOLOGY.8. (DEFAULT= 1. STRAIN. 40 CHARACTERS).Table 11. LEAST SQUARE FIT ON FINDLEY'S CREEP LAW FOR RHEOLOGICAL MODEL.1678970E05 .4885375E+ii 2 2.55766203+11 8 8. TYPE A TITLE(MAX. TYPE IN AN ERROR TOLERANCE FOR ITERATION.8038159E07 2.68046073+12 I 1.20487013+10 9.4587280E08 5.7770992E09 4.7277285E+lI 5 8.2354318E08 5.5469646E+ii 7 8. OUTPUT FILE NAMES: TYPE FILE NAME FOR TIME VS.4883239E+ii 1. 3.1731938E+05 9.6204513E+05 2. DEFINE THE RANGE OF SECOND VARIABLE.E8 4. LEE. 2. DEFAULT=RHEOLOGY. STRESS. DEFINE AN XAXIS IN TIME.PAM I LEAST SQUARE FIT AT STRESS 1.8367003308 I 1.1780644E+ii 6 8.000000E+02 ITERATION CS CP KP ERROR 0 1. TYPE MAXIMUM TIME DOMA1N:TMAX (DEFAULT=5. TYPE IN THE MAXIMUM OPERATING STRESS: SMAX (DEFAULT=I500) 6.7783750E09 4.E5) HOW MANY POINTS WOULD YOU USE FOR CURVE FIT? (DEFAULT=200) 5.
317984E05 6.5488E+lI CP= 1.277523304 2.0044E+06 TIME 2.0000E+02 CS= 8.304991308 5.980861E04 1.830072E09 7.054546E04 4.832982E09 2.351199E08 1.841699E08 1.OUT File Content) I AT STRESS 1.180974E09 2.349165E09 7.018834E09 2.500000E+03 1.480685E09 6.000000E+04 1.500000E+04 STRAIN/EMPIRICALRHEOLOGICAL STRRATE/EMPIRICAL RHEOLOGICAL 4.698933E09 1.0042E+06 STRAIN/EMPIRICAL RHEOLOGICAL STRRATE/EMPIRICAL RHEOLOGICAL TIME 2.271788E05 2.215664304 7.600114E05 1.8.000000E+04 1.OOOOOOE+03 1.566844E08 1.000000E+03 7.949092309 1.961714E09 1.345837E09 7.414938E09 3.320230304 1.823994304 1.2 StrainIStrainRate vs.500000E+03 5.913807E05 4.063761304 3.5494E+lI CP= l.443429E09 7.227393E08 5.0108E+IO KP= 1.359683E09 RHEOLOGICAL 2.0040E+06 STRAIN/EMPIRICAL RHEOLOGICAL STRRATE/EMPIRICAL TIME 2.779724E09 6.974205E09 7.218916E05 5.786231E09 4.500000E+03 1.0000E+02 CS= 8.438558305 8.812336E09 5.596242E04 5.888488E05 2 AT STRESS 2.0103E+10 KP= 1.956276E05 6.625984E09 1.500000E+03 1.Table 11.077321E05 1.226731E05 1.500000E+03 1.500000E+03 1.187549E04 7.318766E05 1.560396E09 1.399461E04 1.362111309 5.444280E08 .592629E05 7.572710E09 9.0000E+02 CS= 8.OllOE+lO KP= 1.250000E+04 1.996244E05 7.000000E+03 7.445559304 2.5476E+ii CP= 1.250000E+04 3 AT STRESS 3.136065E09 3. Time (RHEOLOGY.660714305 1.463970E04 1.386654E05 6.582290304 6.
vs.PAM File Content) LEAST SqUARE FIT ON FINDLEY'S CREEP LAW FOR RHEOLOGICAL MODEL ./I(./C.Table 11. Stress (RHEOLOGY.8.3 C.
8. Y TYPE I. RHEOLOGY 7. Y2 . 10 CHARACTERS). Yl. NUMBER OF DATA POINTS? DEFAULT WILL INCLUDE ALL THE DATA POINTS IN THE FILE 3.Table 11.XMAX TYPE A LABEL(MAX. 2. TYPE IN BAUD RATE I. DUMMY. 3 I IF 2 IF 3 IF 4 IF I IST CURVE WILL BE A SOLID LINE. FINDLEY TYPE A LEGEND FOR DASHED LINE. OR 4 TO SPECIFY INPUT DATA FORMAT: INPUTS ARE X. Y2 INPUTS ARE X. CREEP STRAIN 5 . OVERLAY TWO CURVES? Y OR N. TIME IN HOUR 4 . Yi.4 Run Progression of the Program "DISPLY" $RUN DISPLY THIS IS A GENERAL PURPOSE PROGRAM FOR XYPLOT WRITTEN BY SANG H. 40 CHARACTERS). TYPE A TITLE(MAX. LINEAR OR LOGARITHMIC? TYPE LN OR LG WHICH DOMAIN DO YOU WANT TO DISPLAY? TYPE XMIN. XAXIS DEFINITION. X2. DEFAULT IS TO READ THE FIRST LINE FROM INPUT FILE 2. TYPE THE FILE NAME. LINEAR OR LOGARITHMIC? TYPE LN OR LG RANGE FOR YAXIS WILL BE SET AUTOMATICALLY. LEE. X2. GRID LINES? Y OR N. YAXIS DEFINITION. 2ND WILL BE DASHED TYPE A LEGEND FOR SOLID LINE. 6. Y2 INPUTS ARE Xi. 10 CHARACTERS). Yi. TYPE A LABEL(MAX. Y INPUTS ARE XI. DUMMY.
NUMBER OF DATA POINTS= 200 HIT RETURN TO CONTINUE .
.
Q 0 c ( 8 Q) w L: .u r n QO .
.
.
.
CREATE AN EXTERIAL FILE..) DSIGMA=IOO...O....EQ.DSIGMA.SMAX .GT. 5 .JSTRESS'...OUT~ IF(NAME~.0.STATUS=JNEW') OPEN(UNIT=8...E6 TMIN=O......1003) C C ..STATUS='NEWJ) INPUT TITLE WRITE(8...DTIME.O.EQ..... 11/6/87 THIS PROGRAM PROVIDES RHEOLOGICAL PARAMETERS KP.EQ. Source Listing for the Program "RHEOLOGY" C C C L PROGRAM RHEoLOGY(INPUT. IF(TMAX.Table 11S . TMAX .IBK*l DATA IBK/' ' / C INPUT DATA INTERACTIVELY C CALL INTERACT(NF .2OA4) 1002 FORMAT(/SX...THAX.. DTIME=(TMAXTMIN)/ANDP TINIT=DTIME*4.)SMAX=l500.O. 100 SIGMA=SIGMA+DSIGMA IF(SIGMA..... CORRESPONDING TO THE GIVEN EMPIRICAL CREEP LAW (FINDLEY'S) USING THE LEAST SQUARES FIT.I=1. LEE..ANDP .EQ. CP AND CS..JCS'14X'CP'14X'KP'~ C NUMBER OF LINES NL=O SIGMA=O.. C C C DEFAULT VALUES ARE DEFINED HERE...FILE=NAMEl.OUTPUT. C C C c C C COMMON / /TITLE(20) CHARACTER NAME1*20..O.IBK~NAME~=~RHEOLOGY..) ANDP=200.lOOi) (TITLE(1) .TINIT.EQ.. IF(ERROR..PAM~ OPEN(UNIT=7..1002) 1001 FORMAT(X ... NAME21 C c ..EQ.ERROR~ WRITE(7..SIGMA....NAMEl...NF..........)ERROR=l.NAME2*2O.TAPE~=OUTPUT) BY SANG H.EQ. IF(ANDP..FILE=NAME2....)TMAX=500000.. IF(NA~EI. IF(SMAX. C IF(DSIGMA.IBK)NAME~=~RHEOLOGY...20) WRITE(8.ERROR.10X.SMAX)GO TO 200 NL=NL+I CALL PROCESS(NL...
....STR................ C 1003 FORMAT( 1H ) GO TO 100 200 CONTINUE CLOSE(UNIT=7) CLOSE(UNIT=8) STOP END C FUNCTION EPSF(N.....CP..l9l.EQ.19*0.EQ..........19*0.l)THEN 1 EPSF=O...c .CS........O067*SINH(STR/8500.....l)THEN EPSR=STRESS*TIM/CS+STRESS*(l.......EQ....KP C REAL KP C C RHEOLOGICAL MODEL IF(N...) +O...............0011*TIM**(O.......191..) ENDIF C RETURN ENTRY EPSFDT(STR..TIM) IF(N...)*SINH(STR/8500.....TIM) COMMON/STHD/STRESS.2E6 ELSE EPSF=0...... TIM) EPSFDT=0..)*SINH(STR/8500.)STR/1......) RETURN END c .............00ll*TIM**(O..EXP(KP*TIM/CP))/KP ELSE EPSR=STRESS/CS+STRESS*EXP(KP*TIM/CP)/CP ENDIF RETURN DERIVATIVES ENTRY FUlF(N..........TIM) C C C FINDLEY'S CREEP LAW AND ITS TIME DERIVATIVE IF(N...l)THEN FUlF=STRESS*TIM/CS**2 ELSE FUlF=STRESS/CS**2 ENDIF RETURN C C .l9*SINH(STR/8500........... C FUNCTION EPSR(N....OOll*TIM**O.
IO13)IANS IF(IANs. Y OR B (DEFAULT) ' READ(S.0)GO TO 100 C INPUT TITLE 300 TYPE 1000 ACCEPT 1001.' 1 .ENTRY FU2F(N.EQ./KP+TIM/CP)/KPSTRESS/KP**~ ELSE FU~F=STRESS*TIM*EXP(KP*TIM/CP)/CP**2 ENDIF RETURN END C C SUBROUTINE INTERACT(NF.ANDP .SMAX .'B.998) 998 FORMAT(X.201 1000 FORMAT(X.)*ExP(KP*TIM/CP)/CP**~ ENDIF RETURN ENTRY FU~F(N.NAME2*2O COMMON / /TITLE(~O) DATA IY.'2. LEE. I / ) C C C LEAST SQUARE FIT ON THE STRAIN OR STRAIN RATE? NF=O 100 WRITE(6.EQ. TYPE 1004 .IN)NF=~ IF(NF. TYPE IN AFJ ERROR TOLERANCE FOR ITERATION.EQ.DSIG . LEAST SQUARE FIT ON THE STRAIN RATE?'.~)THEN FU~F=STRESS*TIM*EXP(KP*TIM/CP)/CP**2 ELSE FU~F=STRESS*(KP*TIM/CPI.'(DEFAULT= 1. ERROR.EQ.X.IN.)** WRITTEN BY SANG H.IBK/'Y'. TYPE A TITLE(MAX.(TITLE(1) .ERROR 1003 FORMAT(F20 .O) C XAXIS IS TIME.' ' / WRITE(6. 40 CHARACTERS).E6)') ACCEPT 1003.') 1001 FORMAT(20A4) C ERROR TOLERANCE TYPE 1002 1002 FORMAT(X. TMAX .999) 999 FORMAT(/X1* PERFORMS LEAST SQUARE FIT FOR RHEOLOGICAL PARAMETERS'/ 1 lOX.IBK)IANS=IN IF(IANS.NAME21 CHARACTER NAME1*20.l)THEN FU~F=STRESS*EXP(KP*TIM/CP)*(~. 1 2x.TIM) IF(N.NAME1 .IY)NF=2 IF(IANS.'N'.'I.Eq.EQ.I=1. TIM) 1F(N.
*TYPE IN THE MAXIMUM OPERATING STRESS: SMAX' 1 .'TYPE FILE NAME FOR STRESS VS.1013)NAME2 1012 FORMAT(5X.NF.') 20 TYPE 1005 1005 FORMAT(5XIJTYPEMAXIMUM TIME DOMA1N:TMAX (DEFAULT=5.CP.0) C DATA POINTS IN TIME TYPE 1007 1007 FORMAT(5X1'HOW MANY POINTS WOULD YOU USE FOR CURVE FIT?' 1 .' 1 X.'TYPE IN THE STRESS INCREMENTS SWEEPING STRESS RANGE' 1 .ERLMT) COMMON/STHD/STRESS. OUTPUT FILE NAMES:') TYPE 1011 READ(5.'4. STRESS TYPE 1014 1014 FORMAT(XIJ6. DEFINE THE RANGE OF SECOND VARIABLE.TMAX.2XJ(DEFAULT=1500)') ACCEPT 1003.TNAX)GO TO 200 FXI .' 1 X'DEFAULT=RHEOLOGY.Y(SOI) . DEFINE AN XAXIS IN TIME.'DEFAULT=RHEOLOGY.E5l1) ACCEPT 1003. STRAIN. TIME AND CS/CP/KP VS.CS.2X' (DEFAULT=200)' ) ACCEPT 1003. SMAX TYPE 1010 1010 FORMAT(SX.YRATE(901) REAL KP C STRESS=SIGMA I=O X(I)=O.DSIG C OUTPUT FILE NAMES FOR STRAIN VS. CS/CP/KP.2X'(DEFAULT=lOO)') ACCEPT 1003.OUTJ) TYPE 1012 READ(5.ANDP C SECOND VARIABLE IS STRESS TYPE 1008 1008 FORMAT(X.1013)NAMEl loll FORMAT(6XI1TYPEFILE NAME FOR TIME VS. C 10 I=I+l X(I)=X(Il)+DELTIME I ( ( ) .TINIT.PAMJ) 1013 FORMAT(A) RETURN END C C SUBROUTINE PROCESS(NL.DELTIME.GT.') TYPE 1009 1009 FORMAT(5X.TMAX 1006 FORMAT(2Fl5.'S.1004 FORMAT(X. STRESS.KP DIMENSION X(9Ol) .SIGMA.
STRAIN. CP AND KP FOR ITERATION.CP.1001)STRESSDCSDCPlKP C WRITE(7.IP STRAIN=EPSR(I.EQ. J IF(J.STRESS.X.EQ.C C C C NP=I Y(I)=EPSF(~.1)THEN CALL CURVFIT(NF.STRESS.999)NL.X.X(I) .STRATE X) ( 99 CONTINUE RETURN END C C C C SUBROUTINE GUESS(TIMEF.X(I)) WRITE(T.Eil.ERLMT.STRESSlCSDCPDKP WRITE(8.4. ( ) xI) STRATE=EPSR(2.NP) ENDIF WRITE(7. 1 2XD1STRRATE/EMPIRICALRHEOLOGICAL') C.'LEAST SQUARE FIT AT STRESS' .3RKP=. DO 99 I=i.CS.I5. COMMON/STHD/STRESS.TEMP) .NP) ELSE CALL CURVFIT(NF.KP REAL KP C cS=STRESS/EPSFDT(STRESS . TIMEI COMPUTE INITIAL VALUES OF CS.Y(I) .E11.6/) 1001 FORMAT(5(1PE16.YRATE(I) .1002) 999 FORMAT(I4.50XD 110) 200 CONTINUE WRITE(6.X.IPEII.X(I) YRATE(I)=EPsF(~. STRESS CALL GUESS(TMAX . 1 X.TIHEF) TEMP=EPSF(~.Eii.4/) 1000 FORMAT(lHO.4.STRESS.2X'AT STRESS'.ERLHT.3HCS=.iPEi5.Y.0.YRATE.Y(I).TIMEF) TEMP=TEMPSTRESS*TIMEF/CS KP=STRESS/TEMP TEMP=EPSF(I.X(I)) GO TO 10 READ 1001.1000)NL.3HCP=.TIMEI)STRESS*TIMEI/CS TEMP=TEMP*KP/STRESS TEMP=ALOG(I.X.~OO~) 1 .STRESS.3X.4.O)GO TO 10 NP=I 1001 FORMAT(2F10.TINIT) IF(NF.6)) 1002 FORMAT(8XD'TIME STRAIN/EMPIRICAL RHEOLOGICAL'.
NN Y(N)=YY(N)EPSR(NF.CS~CP~KP DIMENSION X(1) .30)GO TO 200 I=I+l ERROR=ERR CALL LSTSQ(NF.LT.ERLMT. M=3 I=O WRITE(6. ~ H K P . ' ) RETURN END C C C c SUBROUTINE LSTSQ (NF.c .NN.IOO~) I.YY( 1) COMMON/STHD/STRESS.ERLMT**~)CO TO 100 IF(ABS(ERROR/ERR~. I ~ . ~ ( I P E ~ ~ .X.*ERROR)GO TO 200 IF(I.A(20.CP.NN) DIMENSION X( 1) .N.KP.GT.CS.ERLMT.Y. ~ O X .B(20) . DO 20 N=I.Y .CP.MI COMMON/STHD/STRESS.C~M) CS=CS+C( 1) CP=CP+C(2) KP=KP+C(3) GO TO 5 0 200 WRITE(6.Y(I) . ~ H E R R O R ) 1001 FORMAT(2FlO.CS.Y(9OI) .X.ERLMT.).GT. 0) 1002 F O R M A T ( X .ERR CONVERGENCE TEST IF(ERR.~OOO.C(20) .C(20) .' ITERATION DOES NOT CONVERGE.CP=KP*TIMEI/TEMP C RETURN END C C SUBROUTINE CURVFIT(NF. ~ H C P .ERLMT)GO TO 100 IF(ERR.1003) 100 CONTINUE C 999 F O R M A T ( ~ X .20) .LT. ~ ) ) 1003 FORMAT (/ .F(20.X.KP REAL KP C ERROR= 1.X(N)) ERR=ERR+Y(N)**2 20 CONTINUE C C WRITE(~. ~ H C S .999) 50 ERR=O.901) .YY. ~ O X . ~ X . J I T E R A T I O N J ~ X .5X.
21) .J)A(I.K)*A(K. DO 2 J=l.N loo A(I.I)+F(I. J)=A(I.M DO 3 K=l. J) c DO 4 K=l.N.X(J)) C DO 3 I=l.X 1 BB DIMENSION AA(20.K)=A(K.NN.ID(20) NN=N+l DO 100 I=l.K.M.X(J)) ~ ( 2J)=FU2F(NFtX(J)) . .I)=o. DO 4 J=1.N.5 5 DO 10 I=l.ERLMT) CALL ELIM(A.NN)=BB(I) ID(I)=I DO 100 J=l.N 'IF(KI)41.3 3 KK=K+I DO 4 J=KK.A(20. J)=A(K.ID) 2 IF(A(K.B.l A(I.I)=A(K.1 4 B(K)=B(K)+Y(J)*F(K.2. J)=AA(I. J)=Fu~F(NF.999.IN A(K.C.C C C C C REAL KP DERIVATIVES FulF(T)=STRESS*T/CS**2 FU~F(T)=STRESS*T*EXP(KP*T/CP)/CP**2 FU~F(T)=STRESS*EXP(KP*T/CP)*(~.N 2 A(K.N C C .C) RETURN END SUBROUTINE ELIM(AA .K) DO 4 I=l.H B(K)=o. 3 A(I. J)/A(K.BB(20) .M. J) K=1 1 CALL EXCHI(A.41 41 A(I.J)=FuIF(NF.x(20) .B.K))3.4.I A(K.20) .I) C J)*F(K. 1 F(M.J) CALL GAUSID(A. J) 4 CONTINUE K=KK IF(KI)l.N F(I./KP+T/CP)/KPSTRESS/KP**~ C DO I J=l.
6.lOOO) 1000 FORMAT(5X.NCOL)=A(I.N IF(A(ID1))12. J))B)2.ID) NN DIMENSION ~ ( 2 0 ~ 2 1 ) .NN C=A(NROW.'NO UNIQUE SOLUTION.21 21 NROW=I NCOL=J B=ABS(A(I. . J) ) 2 CONTINUE IF(NR0UK)3.NN) 10 CONTINUE RETURN 999 WRITE(6.K)=C I=ID(NCOL) ID(NCOL)=ID(K) ID(K)=I 4 CONTINUE RETURN END C C SUBROUTINE GAUSID(A.K) ( 42 A(I.C(20.31 31 DO 32 J=K.N DO 2 J=l.3.N.N IF(ID(J)1)10.X(20) K=O NN=N+l DO 11 I=l.N.DO 10 J=i.N 1F(ABS(A(I.B(20) .K.12 12 X(I)=I. J)=A(K.6.ID(20) NROW=K NCOL=K B=ABS(A(K.K)) DO 2 I=l.N C=A(I .ERR) DIMENSION A(20.4. J) A(NROW.J) 32 A(K.') RETURN END C C SUBROUTINE EXCHI( A .J)=C 3 CONTINUE IF(NC0LK)4.X. .20) .20) .2.NCOL) AI .41 41 DO 42 I=l.B.10 6 X(I)=A(J.
J)=A(I.N 1 1 c(I.I) 1 CONTINUE E=O. C DO 3 I=l.N P=Pc(I.l WRITE(6.lOOO) RETURN FORMAT(5X.5 RETURN K=K+l IF(IO0K)6.4.C(I.I.J)/A(I.26H GAUSID DOES NOT CONVERGE.N P=C(I.NN) DO 2 J=l.J)*x(J) CONTINUE X(I)=X(I)+P E=E+ABS (PI CONTINUE IF(EERR)4.I) DO 11 J=i.) END 2 3 4 5 6 1000 .NN)=B(I)/A(I.
O)IBRATE=960 CALL INITT(IBRATE) CALL TERM(2.EQ.1024) C PROGRAM DISPLY 400 CALL ERASE CALL CHRSIZ(3) CALL ANMODE C INPUT TITLE TYPE 1000 ACCEPT 1001.EQ . ' / SX.IBK)IXA=O IF(IANS .LNsLGs1BK/'Y'.XLABEL(~) .wi(9oi) .'D1.6.EQ.20) 1000 FORMAT(XD1l. LEE.'LG'.IANS IF(IANS.'P'. 'DEFAULT IS TO READ THE FIRST LINE FROM INPUT FILE') 1 1001 FORMAT(2OA4) c DATA POINTS(NDP) TYPE 1002 ACCEPT lOO9.'WRITTBN BY SANG H.LG)IXA=I IF(IXA.DASHED (4) CHARACTER NAME*20 DATA 1Y81Ns1Ps1D. TYPE A TITLE(MAX.Table 11.xx2(901) .xS'TYPE IN BAUD RATE') READ(SD1009)BRATE IBRATE=BRATE/IO IF(IBRATE.S.10XS1DEFAULT C XAXIS TYPE 1004 IXA=2 10 TYPE 1005 ACCEPT lOO6.'LN'.998) 998 FORMAT(/. X X ~ (.I=1. 'THIS IS A GENERAL PURPOSE PROGRAM FOR XYPLOT' / 1 IOX.2)GO TO 20 TYPE 1007 GO TO 10 .999) 999 FORMAT(X.' '/ WFtITE(6.ANDP NDP=ANDP IF(NDP . (TITLE(1) .NE.LN) IXA=O IF(IANS. NUMBER OF DATA POINTS?'/ WILL INCLUDE ALL THE DATA POINTS IN THE FILE.O)NDP=SOI 1002 FORMAT(XS12.') w~ITE(6.') 1 .'N'.EQ.Eq.YLABEL(~)SOLID(4).~~2(901) ~O~) COMMON/TTL/TITLE(~O) . Source Listing for the Program "DISPLY" SHL 11/16/87 GENERALIZED PLOT PROGRAM C O M M O N / D T / N D P . 40 CHARACTERS).
EQ.E~. ')GO TO 21 1003 FORMAT(II) 1004 FORMAT(X.XMIN.'5.I=I.(YLABEL(I) .LG)IYA=I IF(IYA. YAXIS DEFINITION. ')GO TO 41 1011 FORMAT(X. ' C GRID OPTION 60 TYPE 1013 ACCEPT lOO6.'ANSWER NOT ACCEPTABLE. EQ .NE.IANS NG=2 IF(1ANS. ~ q . '1 1012 FORHAT(X .XMAX 21 TYPE 1010 ACCEPT 1001.3) IF(YLABEL(I) .EQ.'4.(XLABEL(I).IBK)IYA=O IF(IANS.IY)NG=I . GRID LINES? Y OR N.IBK)IANS=IN IF(IANS.4) IF(XLABEL(I) .EQ. '1 C YAXIS TYPE 1011 IYA=2 30 TYPE 1005 ACCEPT 1006.~)GO TO 40 TYPE 1007 GO TO 30 40 TYPE 1012 41 TYPE 1010 ACCEPT 1001. '1 C OVERLAY 60 TYPE 1014 IDASH=O ACCEPT 1006.I=l.O) 1010 FORMAT(~X.IANS IF(IANS. XAXIS DEFINITION. 'TYPE A LABEL(MAX. 10 CHARACTERS) .IN)NG=O IF(NG.'RANGE FOR YAXIS WILL BE SET AUTOMATICALLY.'UHICH DOMAIN DO YOU WANT TO DISPLAY? TYPE XMIN~XMAX') 1009 FORMAT(~FIO.~)GO TO 60 TYPE 1007 GO TO 50 1013 FORMAT(X.IANS IF(IANS. TRY AGAIN. IF(IANS.EQ.EQ.IBK)NG=O IF(IANS.'3.') 1005 FORMAT(5XD1LINEAR OR LOGARITHMIC? TYPE LN OR LG '1 1006 FORMAT(AZ) 1007 FORMAT(X.EQ.II)GO TO 200 TYPE 1007 .E~. ' ) 1008 FoRMAT(~X.IY)GO TO 100 IF(IANS.LN)IYA=O IF(IABS.NE. 20 TYPE 1008 ACCEPT 1009.EQ.
') C INPUT DATA FORMAT 100 IDASH=O TYPE 1016 ACCEPT 1003.(DASHED(I) .4) 1015 FORMAT(3X.XMAX.1=1. ' Y OR I ? ' //) 700 CLOSE(UNIT=INPUT) C C FOR MORE FILES TO PROCESS. IDASH. 2ND VILL BE DASHED') 1016 FORMAT(3X. 'TYPE 1.(SOLID(1). ~ q . X2.STATUS='OLD . X2. OVERLAY TWO CURVES? Y OR N.AI2. DUMMY.4)GO TO 100 C LEGEND TYPE 1015 TYPE 1017 ACCEPT 1001.READONLY.1 ) O TO 700 NG TYPE 1007 GO TO 450 1020 FORMAT(X 'MORE PLOTS TO PROCESS IN FILE ' .NG.IANS IF(1ANS. I=1. Y1. 1024) (TITLE(1).'6.IN)STOP J . 1024) (TITLE(I) . I=1120) IF(TITLE(I).ERR=~OO) . 'IST CURVE VILL BE A SOLID LINE.EQ.1) OPEN (UNIT=INPUT FILE=NAME.GT. 2.4) TYPE 1018 ACCEPT iooi. 3 OR 4 TO SPECIFY INPUT DATA FORMAT:' / 1 iOXIJ1 IF INPUTS ARE XI Y1.'3 IF INPUTS ARE Xi. I = W ) 500 CALL PREP(IXA. Eq .NAME./) 1022 FORMAT(A) c CALL ASSIGN(I. Yi.OR.Eq. 'TYPE A LEGEND FOR SOLID LINE.IDASB IF(IDASH.GO TO 60 1014 FORMAT(X. 350 TYPE 1023 ACCEPT IOO6. Y2'/ 2 1OXIJ2 IF INPUTS ARE XI.' ) 1018 FORMAT(5X1 'TYPE A LEGEND FOR DASHED LINE. GO TO 500 300 READ(INPUT.' ) C DATA INPUT 200 INPUT=I TYPE 1021 READ(5. Y ' / 3 iOX. ~2') 4: 1017 FoRMAT(~X.E~. C CALL BELL )READ(INPUT.'.IYA. TYPE THE FILE NAME.'4 IF INPUTS ARE XI DUMMY.1022)NAME iO21 FORMAT(XIJ7. Y2'/ 4 lOX.IY)GO TO 300 IF(IANS.0 . Eq.INPUT~IDASH) C 450 TYPE 1020.1 ) O TO 400 YG IF(IANS.IANS IF(IANS.XMIN.NAME ACCEPT 1006.
2.10. l)READ(INPUT.)THEN IF(YYl(1) . .NC.YYi(I) .IN/'YJ. E ~ .Eq.10.10..xx~(I).20.10:.**NSCALX XMAX=IHAXX*IO.IMINX.Eq.~)YY~(I)=YYI(I) IF(IDASH .E~..O..10.~~2(901) DIMENSION GINC(20).**NSCALX .~10.20. ~ q~)READ(INPUT.4)THEN READ(INPUT. //) END C C SUBROUTINE PREP(IXA.O. 4 ~ ~ 5 ~ ~ 5 ~ ~ 5 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 1 10.Eq.IMAXX.10.NDP C BOUNDARY CALL BOUND(XXI.IDASH~ COMMON/TTL/TITLE(~~) .O..20.YYi(I) ~ XX~~I IF(IDASH..YY~(I) IF(IDASH.XMIN.YYi(I) ..Eq.. 2 .)THEN IF(YY~(I).YY~(I) IF(IDASH.E~...~ 1 20..TYPE 1007 GO TO 350 1023 FORMAT(XJMORE PLOTS TO PROCESS? Y OR N'//) 1024 FORMAT(~OA~.DUMMY~DUHMY~YY~(I) . 2 . l)XX2(I)=XX1(1) IF(IDASH..5.~10.END=lO)XXi(I) .)THEN GO TO 10 ENDIF ENDIF ENDIF ENDIF GO TO 100 10 CONTINUE NDP=I1 TYPE 1008.END=l0)XXi(I) .NSCALX~ IF(XM1N.10..~10.. I F ( I D A S H . 1003.10.XMAX.10.O. YY2(1)=0.'NJ/ C INPUT DATA I=O 100 I=I+i XX2(1)=0..TINC(20) DATA T 1 N C / 1 ...10. 1003.E~.10.xx2(901) .IYA./ DATA IY./ DATA G1NC/2. .)THEN IF(XXZ(1) .YY2(1) XX2(I)=XXI(I) ENDIF IF(XXI(I).NDP. .~)READ(INPUT.YYi(I) IF(IDASH.XLABEL(4) JLABEL(4) . .5. 1003..END=10)XX1~1~ JXX2(1) .~i(9oi) .XX~(~O~) . .Eq. O ) R E A D ( I N P U T ~ ~ O O ~ ~ E N D = ~ O ). 1 .SOLID( ).20..DASHED(~) 4 COMMON/DT/NDP. 1OO3~END=iO)XXi~I) .5.E~.INPUT.20.XMAX)THEN XMIN=IMINX*lO.EQ..
YMAX .YMIN.E~.E~. i 350 TYPE 1005 ACCEPT lOO6.GINY D TYPE lOll.YYMAX)GO 300 TO YMIN=YYMII YMAX=YYMAX GO TO 300 1001 FORMAT(X.IDASH. **NSCALX TINY=TINC(ITY)*IO.**NSCAL1 YMAX=IMAXI*lO.**NSCALl C TICK MARK 200 ITX=(IMAXXIMINX)/IO+~ ITY=(IMAXIIMIN~)/~o+~ TINX=TINC(ITX)*lO.IN)RETURN GO TO 350 400 TYPE 1007 ACCEPT 1003.DUMMY 300 CALL GENPLT (IXA. NDP .XXMIN.XMIN.XMIN.GINX GIHY) .XMAX TYPE lOO2.E~.NSCALl C GRID INCREMENTS GINX=GINC(ITX)*lO.IDASH TYPE lOOS.ENDIF CALL BOUND(YYI .NDP.YMAX C CALL BELL TYPE 1004 ACCEPT 1006.NSCAL1)GO TO 40 IF(IMAX~.IANS IF(IANS.XXMAX)GO 500 TO XMIN=XXMII XMAX=XXMAX 500 TYPE 1012 ACCEPT lOO3.IMIN2. IYA XMAX .IYA.O)GO TO 50 CALL BOUND(YY2.XXMAX IF(XXMIN.NG.LT.IY)GO TO 400 IF(IANS.EQ.TINY.YMIN .**NSCALl D TYPE lOlO.GT.N S C A L ~ ) IF(NSCAL~.YYMAX IF(YYMIN. IMAXI .IMAX~)IMAX~=IMAX~ IF(IMIN~.IXA.NSCALI)GO 30 TO W(NSCAL2. .NG .TINY.ITY.GINX.lOA4) ' . IMINI.IMAX2.**NSCALX GINY=GINC(ITY)*IO.IHINI.IMIN~)IMIN~=IHIHZ GO TO 50 30 NSCALl=NSCAL2 IMAXl=IMAX2 40 IMINl=IMIN2 50 YMIN=IMIN1*10. LT .ITX.E~.TINX.TINX.YYMIN.IMAXl. **NSCALl D TYPE lO1l.NSCALI) IF(IDASB.GT.
ABS(YWIN) )NOSWAP=I IF(NOSWAP.)GO TO 210 IF(ABS(YMAX) .~10.NDP IF(YY(I) .IMAX. 10. X . GE. 'XMIN= ' .NDP.3) ' FORMAT(/.1)GO TO 200 C TYPE 1001.~~I~) FORMAT(X.X.IMIN.~EI~.LT.3) WHICH IS DOMINANT? YMAX OR YMIN C NOSWAP=I IF(ABS(YMAX).IS) FORMAT(//.0) FORMAT(//.E10.~) FORMAT(//. X.EQ. GE.XMAX' ) FORMAT(//.YMAX C TYPE 999.5F16.5X.NSCALE) DIMENSION YY (NDP) NSCALE=O C YMIN AND YMAX YMIN=YY (11 YMAX=YY 1 ( DO 100 I=2. 'NUMBER OF DATA POINTS=' . GO TO 200 . FORMAT(// .YMIN)GO TO 100 YMIN=YY(11 100 CONTINUE C TYPE 999.NOSWAP ciooi FORMAT(/IIO) YMAXX=YMAX YMAX=YMIP YMIN=YMAXX C IMAX AND PSCALE 200 IF (ABS(YMAX).' YMAX= .3. ' ) FORMAT(XDICHANGE DOMAIN? Y OR N1//) FORMAT(AZ) FORMAT(X.X.'YMIN= ' .IOO.E10. )GO TO 220 GO TO 300 210 NSCALE=NSCALE+I YMAX=YMAX/IO.'TYPE XMIN.YY C 999 FORMAT(/5E16.GE.YMIN. 'TYPE YMIN .LE. HIT RETURN TO CONTINUE.1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 C C .X.5XD XMAX= ' .X.YMAX)GO TO 90 YMAX=YY(I) GO TO 100 go IF(YY(I).YMAX1) END SUBROUTINE BOUND(YY. ~ 1 03) FORMAT(BN. GO TO 200 220 NSCALEzNSCALE1 YMAX=YMAX*lO.3.
001 CALL VWINDO(XMIN .TINX.NG.XMIN)GO TO 100 XTK=XTK+TINX GO TO 90 ioo IF(XTK.YY2(901) C~MM~N/TTL/TITLE(~O) XLABEL(4) . ~MIN.XMAX.WIN) CALL DRWREL(Os3) CALL MOVEA(XTK . ' .YHAX) CALL DRAWA(XMIN.YL) CALL MOVEA(XM1N .300 IMAX=YMAX IF(IMAX.YMAX) CALL DRAWA(XM1N .YMIN) C TICK MARKS IF(XMII.YMIN) CALL DRAWA(XMAX .xx2(901) . Eq .YMAX.840.SOLID( ).GE.YMAX)IMAX=IMAX+NOSWAP C IMIN YMIN=YMIN/IO.XMIN)GO TO 70 80 XTK=TIBX so IF(XTK.YMIN) CALL DRWREL(0.XMAX)G~ TO 200 CALL MOVEA (XTK .GT.GINX.O.XXi(901) sYY1(901) .YMIN)IMIN=IMINNOSWAP IF(NOSWAP.)GO TO 80 XTK=O 70 CALL MOVEA(XTK .YLABEL(4) .YMAX) CALL DRUREL(0.XL.GINY) COMMON/DT/NDP .NE.**NSCALE IMIN=YMIN IF(IMIN.B) CALL MOVEA(XTK.YMIN.IYA.150~600) XL=(XMAXXMINI*~.XMIN.YMIN) CALL DRAWA(XMAX .DASHED(4) 4 INTEGER IBUF(20) CHARACTER ISTR*80 CALL ERASE C FRAME CALL ~~1~~0(120. 3) XTK=XTKTIIX IF(XTK.GT.l)RETURN IMINX=IMIN IMIN=IMAX IMAX=IMINX RETURN END C C SUBROUTINE GENPLT(IXA.YMAX) CALL DRWREL(0 .TINY.3) I . 001 YL=(YMAXYMIN)*I.GT.IDASH.GT.
351 CALL MOVEA(XM1N .0) YTK=YTKTINY IF(YTK.O.EQ.GT.YMIN)GO TO 250 YTK=YTK+TINY GO TO 210 250 IF(YTK .YGD) CALL DRAMA (XMAX.GT.YMIN)GO TO 201 209 YTK=TINY 210 IF(YTK.GE .GT.YTK) CALL DRWREL(3.XMIN)GO TO 301 309 XGD=GINX 310 IF(XGD.YTK) CALL D R U R E L ( ~ .GT.YMIN) CALL DRAUA (XGD .XTK=XTK+TINX GO TO 100 200 IF(YMIN.O) CALL MoVEA (XMAX .YMIN)GO TO 370 YGD=YGD+GINY GO TO 360 370 1F(YGD. GT.GE. 201 CALL MOVEA(XM1N .YTK) CALL DRWREL(3 .YTK) CALL DRWREL(3 . 0) YTK=YTK+TINY GO TO 250 C GRID LINES 300 IF(NG.)GO TO 309 XGD=O.GT. GE.YGD) YGD=YGDGIHY IF(YGD.O)GO TO 400 IF(XMIN. XM1I)GO TO 320 XGD=XGD+GINX GO TO 310 320 IF(XGD.YMAX) XGD=XGD+GINX GO TO 320 350 IF(YMIN.YMAX) TO 300 GO CALL MOVEA (XMIN.O.YMAX) XGD=XGDGINX IF(XGD.GT. 301 CALL MOVEA(XGD .YMAX)GO TO 400 ' .YMIN)GO TO 351 355 YGD=GINY 360 IF(YGD .GT.GT.XMAX)GO TO 350 CALL MOVEA (XGD. ~ ) CALL MOVEA(XMAX.O.YMIN) CALL DRAWA(XGD .)GO TO 355 YGD=O.)GO TO 209 YTK=O.
4) CALL MOVABS (20.E~.341 C SCALE 500 CALL MOVEA(XMIN .YMIN) CALL MOVREL(50. YGD) YGD=YGD+GINY GO TO 370 C CURVES 400 CALL MOVEA(XX~(~) .YMAX) CALL M0VREL(110.YYZ(I) .30) CALL CHRSIZ(2) CALL AOUTST(80 .2) WRITE(ISTR. lOO2)YMIN CALL CHTOIN (ISTR .YMII CALL MOVEA(XM1H .I=I.~~~~)XMAX CALL CHTOIN (ISTR.IBUF) C TYPE 1002.YYi(I)) IF(IDASH.IBUF) TYPE 1001.IBUF.12) CALL AOUTST(12.(YLABEL(I1. 2) WRITE(ISTR.IBUF . I=ls3) C TITLE CALL MOVABS( 10.IOO~)YMAX CALL CHTOIN (ISTR.YYl(l)) DO 410 I=2.12) CALL AOUTST( 12.CALL MOVEA(XMIN.100) CALL AOUTST( 16.~OO~)XMIN CALL CHTOIN (ISTR.I B W .YGD) CALL DRAWA(XMAX.IBWs 12) CALL AOUTST( 12.O)GO 500 TO CALL MOVEA(XX~(I) .(XLABEL(1) .YLABEL) c TYPE 1003.YMII) CALL MOVREL(110.460) CALL AOUTST( 12.IBUF) TYPE lOO2. 20) WRITE(ISTR.XMAX CALL MOVEA (XMIN .XLABEL) C TYPE 1003.12) CALL AOUTST( 12.TITLE) CALL CHRSIZ(3) .XMIN CALL MOVEA(XMAX.YMAX C LABELING CALL MOVABS (500.YMIN) CALL MOVREL(20.IBUF) TYPE 1001.NDP 410 CALL DRAWA(XX~(I) .NDP 420 CALL DASHA(XX~(I) . 20) WRITE(ISTR.YY2(1)) DO 420 I=2.
Il.2.IBUF.lO) LEGEND IF(IDASH.675) CALL DSHREL(IOO.I.NC 11 = ICHAR(ICH(I:I) CALL CFS (IBUF.NC) INTEGER IBUF(1) CHARACTER*80 ICH C DO 100 I=l.(SOLID(I). 1OA4) END C C SUBROUTINE CHTOIN (ICH.O.I=l. SOLID) TYPE 1003.C C TYPE 1003.B~) CALL MOVREL(30.DASHED) 600 CONTINUE C CALL HDCOPY CALL HOME C CALL BELL CALL ANMODE RETURN 1001 FORMAT(1PE8.lD4X) 1002 FORMAT(IPEl0.I=1.O) CALL MOVREL(30.700) CALL DRWREL(IOO.O) CALL AoUTST(16.(TITLE(1) .O) 100 CONTINUE RETURN END .Eq.O)GO TO 600 CALL M0VABS(650.O) CALL AOUTST(l6.4) CALL MOVABS(650.4X) 1003 FORMAT(X.
.
[1. The Finite Element Method in Engineering Science.4] D.. PrenticeHall. L. Ed. 1982. Swami. "Advances in Nonlinear Analysis in MSC/NASTRAN. Luenberger. [2.BIBLIOGRAPHY [1. H. Second Edition. [1. J. Detroit MI. Swami." AUTOFAC 5 Conference Proceedings.3] S. S..3] B." ASME Symposium on Nonlinear Finite Element Analysis of Shells.4] K. The MacNealSchwendler Corporation. Irons and S.2] E. Structural Systems 1970.2] 0. F.1] R.5] D. March 1920. 1970. Herting. Theory and Practice. N. 4/118. [3. 1981. "The Reality of CostEffective Nonlinear Finite Element Analysis. Mathematical Theory of Elasticity. 1972. Lee. H. Introductory Computer Methods and Numerical Analysis. C. "Basic Considerations for Expedient Solution Strategies in Nonlinear Finite Element Analysis. K. Zienkiewicz. October 1988. Macmillan. [2. 1982. PrenticeHall. Pasadena CA.5] I." Proceedings. Finite Element Analysis Fundamentals. [1. K.2] D." Conference on Finite Element Methods and Technology. Blum. [3. Bibliography  1 . K. H. November 1520. Bathe. 1983. [3. Second Edition. [2. 1956. Herting. [2. August 1985. 1977. "Efficient Methods for Static Nonlinear Analysis of Large Order Problems.1] R. 1975. Dynamics and Stability. Los Angeles. "Architecture of Nonlinear Analysis.Gallagher.1]R.3] D. Herting.N. 1981. 1980. Techniques of Finite Elements. Pasadena CA." MSCINASTRAN User's Conference. November 1417. N. CASA Association of SME. AddisonWesley. McGrawHill. [1. Washington D. Sokolnikoff. MSCINASTRAN Version 64. McGrawHill. R. N. Wiley & Sons.4] D. March 1819.Pennington. Addison [3. Herting and R. [2. Second Edition.G. Numerical Analysis and Computation: Wesley.N.6] M. H. 1984. [2. Rubinstein. Third Edition. Ahmad. N.C. Linear and Nonlinear Programming. Harder. Finite Element Procedures in Engineering Analysis. PrenticeHall. ASME Region IX Computer Applications Symposium. pp. HANDBOOK FOR LINEAR ANALYSIS. MacNeal.  Statics. Karlsten and R.
J. [3. J. 593598. [3." Int. Davis.11] S. pp. San Francisco. of Applied Mechanics. 1981. 1972. [3.. Cimento.Lee. Vol.8. Numerical Methods in Engineering.33. 427444.20. P. 1981. No.13] P. Eng.2." Computers & Structures. Kao.7. Clough.5] C." Int. Vo1.18.32." Computer Meth.16] M. R. Strang. Num. [3. [3. "The Solution of Nonlinear Finite Element Equations. S. Crisfield. H. J. V01. Numerical Methods in Engineering. "Rudimentary Considerations for Effective Line Search Method in Nonlinear Finite Element Analysis. No. 1972.[3. Solids and Structures." Recent Advances in Nonlinear Computational Mechanics. 1984.10. "A Fast Incremental/Iterative Solution Procedure that Handles 'SnapThrough'. [3." J. pp. 1979.18] E. "Rudimentary Considerations for Effective QuasiNewton Updates in Non3 linear Finite Element Analysis. Lee. "A Modified Secant Newton Method for Nonlinear Problems. "Some Practical Procedures for the Solution of Nonlinear Finite Element Equations.7] M." Computers & Structures. Vo1. 12871301. H.A. [3. Vo1. "Line Search Methods for Extended Penalty Function Environments. Vo1.39. 1980. J . 463476.10] S. "Accelerating and Damping the Modified NewtonRaphson Method. Applied Mechanics and Engineering. 1980.Lee.3." Computers & Structures. 1988 ASME Int.6] M. Bathe and A.12] K. "Convergence Criteria for Iterative Processes.17] G.9] L.15. Ramm.19] R.2 .14. No. Vo1. Pineridge Press. Computers in Engineering Conference. Vo1. Vo1. "A Faster Modified NewtonRaphson Iteration. "The Application of Newton's Method to the Problem of Elastic Stability." Proceedings. [3.A." AIAA Journal. pp.15. 1971." Computers & Structures. [3. "The Riks/Wempner Approach . J. Vo1. 5 E. Vo1. pp.Riks. Vo1. Mech.17.13. "Discrete Approximations Related to Nonlinear Theories of Solids. J. 1989. Bibliography ." Int. 1982. Matthies and G.6. A. Crisfield. [3. Crisfield. G.8] H. August 1988.14] S. Wempner. "A Comparative Study on the ElasticPlastic Collapse Strength of Initially Imperfect Deep Spherical Shells. A. 1 .Bergan and R. [3. 1979. 1982. [3." Computers 6 Structures." Int. "Rudimentary Considerations for Effective Convergence Criteria in Nonlinear Finite Analysis. [3. W. Owen. 1989." Computer Meth. Applied Mechanics and Engineering. Vo1.22. Vo1.An Extension of the Displacement Control 311 Method in Nonlinear Analysis. Zhang and D. H.
.. Mech.. L. H. 151172. Chaudhary and K. and N. Meth. Curnier and W." Computers & Structures. 71108.. 1979. Landers and R." UCB/SESM85/09. N. Appl.. pp." Comp. Sackman.. Wriggers and R.10] H.Lee and S. of MSCINASTRAN Users Conference. Mech. Mech. [4.3] J. tact Problems. "A T. Bibliography  3 .." Comp. T." Comp. .2] Okabe. Eng. "Some Salient Considerations for Computational Procedure in Elastoplastic Material Model. Ju and R. . Simo. Oden. Bathe and A..21. Eng. [4. of Civil Eng. Kikuchi. Finite Element Method for a Class of Contact Problems. Lee and D. pp.. [4. ____ . [4. Appl. 1982. 1986. Taylor. 163180. 1980. 22. 19351937. L." Computer Methods in Applied Mechanics and Engineering.12]S. Deformation ElasticPlastic Formulations. . [4. C." Wallace Press. R. Taylor. 6." Comp. Yamada and I. Vol. Num. Cheng and N. Hsieh.G. W. "MSC/NASTRAN Primer. 1985. J. [4. Kikuchi.. Vo1. "Expedient Implicit Integration with Adaptive Time Stepping Algorithm for Nonlinear Transient Analysis. Mech. A. T . Vol. 49. Hughes. Meth. 1985.. [4. [4. 21.. Eng. in Appl. L. L. Mech. S. Seoul. Y. pp. Lee.20] H.. [4.[3.7]J . 8... Vol. "An analysis of Metal Forming Processes using Large J. 821845. Eng. pp. "Basis Transformation of Trial Function Space M... .11]S. Vol. "Comments on 'A Simple Approach to Bifurcation and S.1985.. [4. and Appl.Schaeffer. Korea. Dept. 249279. A..6]J . 1985..8]J.. Mech.1]H.5]A. pp. 1976. October 2223." Proc. "A Perturbed Lagrangian Formulation for the Finite Element Solution of Contact Problems. J.." Int. 1985. R. "An Augmented Lagrangian Formulation for the Finite Element Solution of Contact Problems. Numerical Methods Eng. Eng. 1990." Comp. "A Solution Method for Planar and Axisymmetric ConK. Bathe." J. Meth. "A Perturbed Lagrangian Formulation for the Finite Element Solution of Nonlinear Frictional Contact Problems. University of California. 50. 20. 1990...9]L. "A Solution Method for Static and Dynamic Analysis of Threedimensional Contact Problems with Friction. Herting. pp." Int. J. 6588. Meth. Vol. Campos. No. in Lagrange Interpolation. Appl. Berkeley. H. Kanoknukulchai. 7. Static and Normal Modes Analysis.. B. Appl. _. 34. 24. Limit Point Calculations'.. [4. Theo. 855873..4] J. pp. 81. "A Numerical Analysis of a Class of Contact Problems with Friction in Elastostatics. Nishiguchi.. Taylor. Eng. P. Meth. J. 1988. pp. [4. L... No... 8599. J. Meth. Taylor... pp. Chaudhary.
7] S. [5. Szilard. Bock.11.11. R. D. Raiten.1] R. Vibration Problems in Engineering. March 28. V. John Wiley & Sons. Haggenmacher. Likins. Herting and G. Los Angeles.14.6. N. [5. R." Int. J. H. [5. Timoshenko.2] S. H." Computer Methods in Applied Mechanics and Engineering. E.17.12] N. Solids Structures. 1978. McMeeking and J. PrenticeHall. C. [4. S.16] S. and A. 198199. Numerical Methods in Engineering. Lee and A.21. D. Zienkiewicz.. pp. C. [4.. Timoshenko and J.Lee." proceedings of 1990 MSC World Users Conference. [5. 1983.Kane. Vo1. [5. 409421.10] J. 1981.3] 0. PrenticeHall. 5/122.5] D." Int. Japan." Proceedings. H. 1979. MSCINASTRAN User's Conference. Vo1. W. J. H. Hibbitt. An Introduction to Mechanical Vibrations. [5.Hibbit. The Finite Element Method in Engineering Science. D. [5. pp. 1985. McGrawHill. Vo1." Proceedings of the MSCINASTRAN User's Conference. Vol. Third Edition. 1970. Steidel. Jr. P. Malvern. "Some Follower Forces and Load Stiffness. Tseng and G. [4. [5.11] J. pp. 1990. 1975. [5. Third Edition.8] H. Timoshenko and J . McGraw Hill. 1961.14] S. Nagtegaal. Met. Rice. Numerical Methods in Engineering. McGrawHill. McGrawHill. [5. Theory of Elastic Stability. "Incremental Formulation in Nonlinear Mechanics and Large Strain ElastoPlasticity: Natural Approach (Part 1 and 2). 1991. 15. Levinson. in Engineering. Lee. Solids Structures. New York. Los Angeles. F. "Inelastic Finite Strain Analysis of Structures Subjected to Nonproportional Loading. Rice. Hsieh and T. "Some Computational Aspects of ElasticPlastic Large Strain Analysis. 1974.[4. Theory of Elasticity. 1977.13] R. H. Jr. New York. Gere. [5. Marcal and J.15] S. P. Fourth Edition. Kleiber. Vo1. J. N. John Wiley & Sons. 1977 and Vo1.14. Vol. 1970. Introduction to the Mechanics of a Continuous Medium. R.9] R. M." Int. W. J." Int. New York. S. Theory and Analysis of Plates.6] T . 1987. P. "Adaptive ArcLength Methods in MSC/NASTRAN. October 1718. C. Bibliography  4 .4] H. "Pressure Follower Matrix for Geometric Nonlinear Finite Element Analysis. "Finite Element Formulations for Problems of Large ElasticPlastic Deformation. Argyris and M. 937941. [5. Young and W." Int. Tokyo. "A Finite Element Formulation for Problems of Large Strain and Large Displacement. T . Spacecraft Dynamics.13] L. Num. 1974. 1971. Weaver. 1969. L. J. [5. "Characterization of Geometric Nonlinearity with Some Salient Examples. Goodier.
." Structural Mechanics: Proceedings of the First Symposium on Naval Structural Mechanics. Finite Elements in Plasticity.13]G. J.4]S. Greenbaum and M. R. E." Proceedings. Seoul.98.. [6. Yang and S.12]M. pp. Mase. Computers in Engineering Conference. 1977 [6. pp. [6. 1965.16]W. The MacMillan Company.5]0. Hill.[5. Owen and E. Mechanics of Plastic Deformation in Metal Processing. The Finite Element Method. C. 169198. August 1822. 1976.11]Y. MSCINASTRAN User's Conference. 1980.7. "Convenient Form of Stress Invariants for Plasticity. 1974.9]H. ST4. Numerical Methods in Engineering. C. ASCE. J. Zienkiewicz.1960. "Effect of Timedependent Material Properties on Dynamic Response." Ph. 1991. Dissertation. ''A New Method t o Solve Nonlinear Creep Problems in Structural Mechanics Using Incompatible Isoparametric Finite Elements. Fung. Noor and H." Journal of the Structural Division. "Creep Analysis of Axisymmetric Bodies Using Finite Elements. New York. [6. ed. Mech." Proceedings. N.10. Introduction to Mechanics of Continua. Zienkiewicz. Bibliography  5 . Foundation of Solid Mechanics.2]E. Rubinstein. Pineridge Press. U. H. C. Third Edition. The Mathematical Theory of Plasticity. "Assumptions. 1991 ASME Int . Korea.N. [6. 161174. Vo1. K.15]G. G. F. Yamada. Thomsen. Continuum Mechanics. ('SelfAdaptive Algorithm for Efficient Computation of Elastoplastic Material Process. McGrawHill. T . McGrawHill. pp. [6. 1961 [6. Kobayashi. [5.D." Trends in Computerized Structural Analysis and Synthesis. October 2223. H. A. 403414. 8. Vo1.1] R. McComb. Hinton. G.8]Y. Badani. Rivlin. Sci. S. Lee." Nuclear Engineering and Design." Int. C. [6.14]Y. Models and Computational Methods for Plasticity. Calif. A. First Edition 1950. [6. Ginn and Company. C. [6. Berkeley. Pergamon Press. [5.3]S. [6. 1968. P. Sato. Sakurai. 343354. Lee. April 1972. C.6]G.Armen. Oxford a t the Clarendon Press. [6. Vol. Santa Clara." Int..7]D.1978.10]R. 1968. Fifth Edition.Prager. H. [6. 1990. Vol. pp. "Plastic Stress Strain Matrix and Its Application for the Solution of ElasticPlastic Problems by the Finite Element Method. J.Takabatake and T . Yamada and T. New York. Jr. PrenticeHall. "Some Topics in Finite Elasticity. Nayak and 0. 1967. "Some Salient Considerations for Computational Procedure in Elastoplastic Material Model. 1980.
15]A. Levy. 1990." Transactions. S." FWR27. 817.19]H." Int. 1967. H. "A Comparative Study on the ElasticPlastic Collapse Strength of Initially Imperfect Deep Spherical Shells". pp. [8. Levy. [6. MSC User's Conference.21] R.81. B. 1981. C." Computers and Structures.23] S.[6. Bibliography  6 . International Journal of Numerical Methods in Engineering. Innovative Methods for Nonlinear Problems. L. Foster Wheeler Corporation. "Application of the Rheological Model to the Creep Analysis Coupled with Plastic Deformation. [6.2]R. "A General Purpose Nonlinear Transient Integration System". Herting. pp.7." Oak Ridge National Laboratory. 1983. Juvinall.26." Int. 1988.17]S. John Wiley & Sons.3]Dinno and Gill. "Creep Analysis Capability in MSCINASTRAN. 1980. [7. Hsieh. ORNL5939. J. [7. "HighTemperature Inelastic Analysis. J . [7. J. H. M. August. T . 1965.Lee. "Generalized Viscoelastic Model for Creep Analysis Coupled with Plastic Deformation. March 1984. "ElevatedTemperature Deformation and Failure Testing and Analysis of NozzletoSpherical Shells: Specimens NS2 and NSI. [6. C. 22. Stress. Mesh. McGrawHill. 1984. January. Clinard et al. J. Pifko. Strain and Strength. Belytschko and K. Levy and A. [6. ORNL/Sub448513. and E. Vo1. 8592. Vo1. "An Experimental Investigation into the Plastic behavior of Flush Nozzles in Spherical Pressure Vessels. PrenticeHall." Proceedings. "Expedient Implicit Integration with Adaptive Time Stepping Algorithm for Nonlinear Transient Analysis". Kraus.1]J. 1966. Lee. [6. edited by W. 1985. Vo1. [6.. June 1980. "Intermediate Heat Exchanger for Fast Flux Test Facility: Evaluation of Inelastic Computer Programs.2]S.18]S.Lee and S. Numerical Methods in Engineering.3. Ripling. "On Computational Strategies for Problems Involving Plasticity and Creep. Vo1. K. Int. H. H. A. "HighTemperature Inelastic Analysis of a Pressure Vessel Nozzle Attachment. 151172.13. 1981..14]J . [6. Kao. Vo1. Vol.22]A. March 1972. H. p. Brussels.1] D. No. Chern. Strength and Structure of Engineering Materials. [8. Pineridge Press.17.17. Pasadena CA. John Wiley and Sons. Park. The 8th SMIRT Conference. Polakowski. 153165.16]A. [6. No. Creep Analysis. Liu. N. Lee. Computer Methods in Applied Mechanics and Engineering. 1981.20] N." International Journal for Numerical Methods in Engineering. pp." Grumman Aerospace Corporation. Vo1. [6. Sci.
No. Hsieh. "The Cologne Challenge. H . J.5] S. pp.7] S. P. Vo1. M. July. [10." Computers & Structures." Computer Methods in Applied Mechanics and Engineering. B. Vo1. No.2] M.." Int. Vo1. Numerical Methods in Engineering. J.17. [8. 1964.11.4] P. Lee and D." Computer 0 Structures." Int. Mondkar and G. " Int. [8. H. J. No. J. pp. Pergamon Press. Fujikake.. I. 1964. Lee. pp. G. Vo1. "An Automatic Time Stepping Algorithm for Dynamic Problems.3] N. Vo1. T.6. 1977. Bibliography  7 . [10. 1981. 427444. 12871301.2. "Adaptive Time Stepping Algorithm for Nonlinear Transient Analysis. 19351937.'' Proceedings. Mollestad.32.33. Vo1.1] S.49. 499520. Newmark. Structural Dynamics. Lee. J. pp. of Engineering Mechanics Division of ASCE. Lee.1] S. P." Air Force Flight Dynamic Laboratory. 299318. 6976.3] R. "Finite Element Analysis of Nonlinear Static and Dynamic Response.4. 463476. 1985. McGrawHill. H. 1988 MSC World Users Conference. A. [8. [8. Lee. July. [10. [8.8] J. Numerical Methods Eng. "A Method of Computation for Structural Dynamics. Vo1. 463476. H. Schulze Schwering." Computer Methods in Applied Mechanics and Engineering. "Rudimentary Considerations for Effective QuasiNewton Updates in Nonlinear Finite Element Analysis. pp. Powell. pp. 225251. Numerical Methods Eng. "Rudimentary Considerations for Effective Line Search in Nonlinear Finite Element Analysis. Belytschko. Tsay.10] T . No.11] D. [10.42. "A Comparative Study on the ElasticPlastic Collapse Strength of Initially Imperfect Deep Spherical Shells.9] H. Vo1. [8. Biggs. [8.[8. 1989. Bergan and E.2. Report FDPTDR64108. 185191. pp.21. Numerical Methods Eng. S. Herting. pp. N. Vo1. [8. "A Simple Approach to Bifurcation and Limit Point Calculations. Los Angeles.21. M.1984." J. 1989." Computers & Structures.6] S. [11. 1959. 1984. Lin and C.4] W. March 1988. A. pp." Int. Kao. "Rudimentary Considerations for Effective QuasiNewton Updates in Non linear Finite Element Analysis. H. "TheoreticalExperimental Correlation of Large Dy namic and Permanent Deformation of Impulsively Loaded Simple Structures. H. "Comment on 'A Simple Approach to Bifurcation and Limit Point Calculations.. Balmer and E. Bock and S. S. Ohio. L. Vol.2 (54). Witmer. "Explicit Algorithms for the Nonlinear Dynamics of Shells." Finite Element News. WrightPatterson AFB.33.' 1985. 1985. pp.
1988. [11. No.6] R. "A Comparative Study on the ElasticPlastic Collapse Strength of Initially Imperfect Deep Spherical Shells.2nd ed.4. Lin. Witmer. Vo1. Lee. 1968. T. Vo1. [11. 220. [11. Lee. [11. Kao. 4/118.[11.2] S. San Francisco. "Creep Analysis of Axisymmetric Bodies Using Finite Elements. [11. 1984. pp. J. No. Vo1.8 . "The Cologne Challenge. W. "Basic Considerations for Expedient Solution Strategies in Nonlinear Finite Element Analysis." Proceedings." Nuclear Engineering and Design. Vo1. [11. J. pp. Goodier. Fifth Edition. [11.3] S. Schulze Schwering.17.32. 1988 ASME Int. N. "Rudimentary Considerations for Effective Line Search Method in Nonlinear Finite Element Analysis. and E.38. Vo1. 19351937. 12871301. 593598. "An Improved Numerical Calculation Technique for Large ElasticPlastic Transient Deformations of Thin Shells: Part 2 Evaluation and Applications. Herting.11] C. [11. Tsay and R. pp.13] T . 429435. 1985. Vo1. the ASME Region IX Computer Applications Symposium. Keene.3. Vo1. Lee and D.4] W.1984. Greenbaum and M. New York. 1986. H. "Explicit Algorithms for the Nonlinear Dynamics of Shells. "Recent Developments in Explicit Finite Element Techniques and Their Application t o Reactor Structures. Formulas for Stress and Strain.5] S. S.8] S. p. pp. Kennedy.21. [11. 2425.6. 497504. F." ASME Journal of Applied Mechanics.141 Roark and Young.42. [11. 124. H. No.. 225251. Belytschko. 1971. [I 1. A. J .7." Finite Element News. Numerical Methods Eng.10] L. H. 1986. 1989." Int.5.97. Lee. Los Angeles. Lee. pp.12] J. N." Nuclear Engineering and Design. Lin.." Computer Methods in Applied Mechanics and Engineering." 1989.7] S. pp. pp. pp. Theory of Elasticity.9] G. H. P. [11. McGrawHill. pp." Computers 6 Structures. Timoshenko and J. and J. Bibliography . I. 1988. 1981. H. Belytschko. "The Cologne ChallengeThe Replies (Part IV).." Int.15] S. "Comments on 'A Simple Approach t o Bifurcation and Limit Point Calculations'. and C." Finite Element News. J. Rubinstein. A. Computers in Engineering Conference. 1951. Vo1. M. "Rudimentary Considerations for Effective Convergence Criteria in Nonlinear Finite Element Analysis." Proceedings. Leech. I. McGrawHill. pp. Numerical Methods Eng. 427444. "MSC/DYNA Verification and Demonstration Problem Manual. [11. Tsay. Morino.
8592. H. Vol. 1988.16] R. H. The 8th SMiRT Conference. Harder. 1. No. Lee. 153165." Int.[11. L. 1988. [11. "Application of Rheological Model t o Creep Analysis Coupled with Plastic Deformation. 28." Computers & Structures. August. pp.18] S. Vol.17] S. H. [11. 7584." Transactions. "Generalized Viscoelastic Model for Creep Analysis Coupled with Plastic Deformation. Vol. pp. MacNeal and R. Numerical Methods in Engineering. 26. Lee. pp. J. Bibliography  9 . "A Refined FourNoded Membrane Element with Rotational Degrees of Freedom. Brussels. 1985. L.
.
APPENDIX A Nonlinear Bulk Data Description .
.
if blank. Connected grid points at ends A and B (Integers GA. default 0 refers t o the basic coordinate system). 106.GB Xl. GO CID Remarks: 1. However. value of EID will be used). Grid point identification number t o optionally supply the orientation vector 5. 99. O Coordinate system identification number (Integer.CGAP BULK DATA ENTRY (V67) Bulk Data Entry: CGAP . GA # GB). Property identification number of a PGAP entry (Integer 2 0. Format: 1 2 3 4 5 6 X1.X2. but remains . The gap element is intended for the nonlinear solution sequences 66.GO 7 X2 8 X3 9 10 CGAP EID PID GA GB CID CGAP 17 2 110 112 13 Field EID PID Contents Element identification number (Integer > 0). and 129. See Remark 2. Components of the orientation vector v'. See Remark 2. Vector Cis from GA to G (Integer). See Remark 2.X3 > 0. See Remark 1. or blank. given in the global coordinate system of end A (Real). it will produce a linear stiffness matrix for the other solutions.Gap element connection Description: Defines a gap or frictional element of the structural model for nonlinear analysis.
the yaxis in the T2 direction..linear with the initial stiffness. F. which are requested with STRESS or ELSTRESS case control commands. Forces. The gap element coordinate system is defined by one of two methods given below: (a) If the grid points GA. etc. The element coordinate system does not rotate as a result of deflections. . 2. i. then line AB is the element xaxis and the orientation vector v'lies in the xy plane (like the beam element). coordinate system CID is used for the element.GB are coincident. are output in the element coordinate system. The stiffness used depends on the value for the initial gap opening (Uo field in the PGAP entry). the element xaxis is in the T 1 direction. Initial gap openings are defined on the PGAP entry and not by the GAGB separation distance. is positive for compression. (b) If the grid points GA.GB are not coincident.e.
Reference temperature (To)at which creep characteristics are defined (Real).0.444 CRLAW 7.985E6 l. Description: Defines creep characteristics based on the experimental data or known empirical creep law.OE9. 6. AH EXP Temperature dependent term. Format: 1 I 2 3 4 5 6 7 8 9 10 CREEP MID TYPE RT a EXP b FORM c TIDKP d TIDCP e TIDCS f THRESH g CREEP 8 121 0. See Remark 1.03254 0.CREEP BULK DATA ENTRY Bulk Data Entry: CREEP . Default = 1. e m .1072 6. in the creep rate expression (0.1479 3. Default = 0.0 1. TIDKP . See Remark 1. Identification number of a TABLES1 entry for specification of the creep model parameter li'p(a) (Integer > 0). See Remark 2. This entry will be activated if MATl with the same MID is being used and the NLPARM entry is prepared for creep analysis.E9 2.0). See Remark 3.Creep Characteristics. Field MID RT Contents Identification number of a MATl entry (Integer > 0). < Real 5 FORM Specifies the form of the input data defining creep characteristics: CRLAW (empirical creep law) or TABLE (tabular input data of creep model parameters).7339 0.
TEMPPI.0 < Real < 1. Specifies the identification number of the empirical creep law type (111 5 Integer 300). TIDCS THRESH TYPE < a?Z Coefficients of the empirical creep law specified in the TYPE field (Real).TIDCP Identification number of a TABLES1 entry for specification of the creep model parameter Cp(u) (Integer > 0). Identification number of a TABLES1 entry for specification of the creep model parameter C. Creep law coefficients (ag) are usually determined by least squares fit of experimental data obtained under a constant temperature.16 PARAM. TABS. From the thermodynamic consideration. Default = 1.(u) (Integer > 0). the creep rate . See Remark 3. The unit of the temperature input ( O F or OC) must be indicated in the PARAM entry TABS as follows: PARAM. The reference temperature (To) at which creep behavior is characterized must be specified in the RT field if the temperature of the structure is different from this reference temperature.98 cal/mole OK) T is the absolute temperature A is strain/sec per activation If the creep characteristics are defined at the temperature To (RT field).69 (if Celsius is used) (if Fahrenheit is used) When the correction for the temperature effect is required. 273.OE5. where E is the Young's Modulus. See Remark 4. and/or TEMPRB) which are selected by the Case Control command. the creep rate is expressed as where A H is the free energy of activation R is the gas constant (= 1. within the subcase. See Remark 4. See Remark 3. Threshold limit for creep process corresponding to the threshold strain (0.OE3). TABS. the temperature distribution must be defined in the Bulk Data entries (TEMP. Threshold stress under which creep does not occur is computed as THRESH*E. TEMP(L0AD) = SID. Remarks: 1. 459.
I<..g. Two different empirical creep laws are available at present. In the case of a linear viscoelastic material. The first class of creep law is Oak Ridge National Laboratory recommends the parameters A(a). Parameter A@) R(a) Type 1 aab cExp(da) e[sinh(f a)]g Type 2 aExp(bo) cod eExp( f a ) K(a) Each of the three digits (e. etc. or (a. and C. and C. and C. the CREEP continuation field should not be used.. the parameter (ICp.) represent parameters of the uniaxial rheological model shown in Figure 1. the value will be reset t o zero and a warning message (TABLE LOOKUP RESULTS IN NEGATIVE VALUE OF CREEP MODEL PARAMETER IN ELEMENT ID = ****) will be detected.at the temperature T is corrected by a factor is to be specified in the EXP field. If the table lookup results in a negative value. the creep law type and the coefficients (ag) must be specified in the CREEP continuation.) values are constant and two data points with an identical parameter value should be specified for each parameter.). Cp. The creep model parameters (K. The second class of creep law (TYPE = 300) is expressed as &'(a.0 .. C.) in the TYPE field selects type 1 or 2 for the parameters A(a). respectively. 3.0 < b < 8. where (e*) 2.) are t o be specified in the TABLES1 entries whose identification numbers appear in the TIDKP. 121.. The creep model parameters must have positive values as shown in Figures 24. 111.). TIDCP.. C.. Tabular entries (xi. the creep model parameters (ICp. If FORM = TABLE. and TIDCS fields. R(a).. and I i ( o ) in the form given in the following table. (o. . Cp. C and C. Cp. t ) = a ob t d where the values b and d are limited to the ranges 1. yi) in the TABLES1 entry will consist of (a. and K ( a ) . In this case.. R(o).) for the input of ICp. If FORM = CRLAW. respectively. 4.. respectively.
a(ksi) Figure 2. Uniaxial Rheological Model. Caution must be exercised t o make these units consistent with the rest of the input data.0 The coefficient g should be blank for creep law type xx2 material and coefficients c. The coefficients (ag) are dependent on the structural units.2 < d < 2. which can be obtained by a subcase with an NLPARM on which DT=O. as a Function of Stress a. + + + c. and g should be blank for creep law type 300. A 6 . f. Creep Model Parameter K . 4 Elastic Primary Creep K . (0) Secondary Creep + + +.and 0. Creep analysis requires an initial static solution a t t=O. 5. (0)  o(t> Figure 1. e.
Creep Model Parameter C. . as a Function of Stress u. Creep Model Parameter C.Figure 3. c. as a Function of Stress a. Ki hours ln3 Figure 4.
H=O. If H is given.0 1 1 2. See Remark 3. For elasticperfectly plastic cases. the stressstrain data must be supplied on a TABLES1 entry referenced by TID and this field must be blank. See Remarks 2 and 3.O. then this field should be blank. Work hardening slope (slope of stress vs. Yield function (Integer) is selected by one of the following values: 1 for von Mises yield criterion (Default) 2 for Tresca yield criterion 3 for MohrCoulomb yield criterion 4 for DruckerPrager yield criterion TYPE H YF . This entry will be activated if MATl with the same MID is being used in the nonlinear solution sequences (66 and 99). For more than a single slope in the plastic range. plastic strain) in units of stress (Real). Specifies the type of material nonlinearity: NLELAST (Nonlinear elastic) or PLASTIC (Elastoplastic). Identification number of a TABLES1 or TABLEST entry (Integer 2 0).MATSl BULK DATA ENTRY (V67) Bulk Data Entry: MATSl .+4 Field MID TID Contents Identification number of a MATl entry (Integer > 0).Stressdependent nonlinear material properties Description: Specifies material properties which are stressdependent for use in material nonlinearity applications. Format: MATS1 MID TID TYPE H YF HR LIMIT1 LIMIT2 1 MATS1 17 28 PLASTIC 0. See Remark 1.
HR. TID. Also a MATT1 entry should be supplied t o specify the temperature dependence of material properties. I the curve is defined only in the first quadrant. LIMITl. y2) must be at the initial yield point (Y1 or 2c) specified on the MATS1 entry. See Remark 4. yl = 0) and the compression properties will be assumed identical t o tension properties. may be specified but not both. it may not reference a TABLEST entry.) of stressstrain data ( E ~ . YF. The values H. LIMITl LIMIT2 Remarks: 1. must conform to Yk) the following rules (see Figure 1): (a) If TYPE is PLASTIC. If TYPE is NLELAST. Parameter representing the internal friction angle for the MohrCoulomb and DruckerPrager yield criteria. If TYPE is PLASTIC. If TID is given. If the table ID is omitted. 3. the full stressstrain curve (oo < x < oa) may be defined in the first and the third quadrant to accommodate different uniaxial compression f data. The slope of the line joining the origin to the yield stress must be equal t o the value of E on the MAT1 entry. either the table identification. or the work hardening slope. H. TABLES1 entries (xi. (0 5 Real < 45O). (c) If TYPE is PLASTIC and TID is given. should be specified in field 5 unless the material is perfectlyplastic. y. yl = 0) and the second point (x2. and LIMIT2 will not be used in this case. the curve must be defined in the first quadrant. 2. (b) If TYPE is NLELAST. For nonlinear elastic temperaturedependent material properties in SOL 66. The first point must be at the origin (xl = 0. TID must reference a TABLEST entry. the work hardening slope.HR Hardening rule (Integer) is selected by one of the following values: 1 for Isotropic hardening (Default) 2 for Kinematic hardening 3 for combined Isotropic and Kinematic hardening Parameter representing an initial yield point (Real). See Remark 4. the stressstrain data given in the TABLES1 entry will be used to determine the stress for a given value of strain. The plasticity modulus (H) is related to the tangential modulus (ET) by . then the curve must start at the origin (xl = 0. H.
See Figure 2. . LIMITl and LIMIT2 are parameters used in the yield function and must be defined as follows: Yield Funtion von Mises (1) or Tresca (2) MohrCoulomb (3) or DruckerPrager (4) LIMIT1 Yield stress in tension. 2c (in stress units) Angle of internal friction q5 (in degrees) Y (or a) I I I I I I E: = Effective Plastic Strain Figure 1. Y1 LIMIT2 Not used 2*Cohesion.where E is the elastic modulus and ET = curve in the plastic region. $$ is the slope of the uniaxial stressstrain 4. StressStrain Curve Definition with TABLES1 Entry.
StressStrain Curve Definition for a Single H. .Figure 2.
Default = 0.E7 AUTO 3 5 25 20. AUTO. 25 4 PW 0.5 Field ID NINC DT Contents Identification number (Integer Number of increments (0 > 0). SEMI or ITER for controlling stiffness updates. Incremental time interval for creep analysis (Real 2 0).2 20. no creep. See Remark 5 for AUTO and SEMI methods. See Remark 1.Parameters for Nonlinear Static Analysis Description: Defines a set of parameters for nonlinear analysis iteration strategy Format: 1 2 3 4 DT EPSW 5 6 7 8 9 10 NLPARM ID EPSU MAXBIS NINC EPSP KMETHOD MAXDIV KSTEP MAXQN MAXR MAXITER MAXLS CONV FSTRESS RTOLB INTOUT LSTOL NLPARM 15 1. KMETHOD KSTEP . See Remark 3. Default = AUTO. 1. < Integer < 1000). Number of iterations before the stiffness update in ITER option.NLPARM BULK DATA ENTRY (V67) Bulk Data Entry: NLPARM . See Remark 2. See Remark 4.E3 0. NO 0. Default = 5..E3 5 5 1.
Default = 0. P. See Remark 8.MAXITER Limit on number of iterations for each load increment (Integer See Remark 6. . See Remark 10. Default=20.2.0). Default = 1.0).OE3.0). Fraction of effective stress (F)used t o limit the subincrement size in the material routines (0.0). < MAXBIS MAXR Maximum ratio for the adjusted arclength increment relative t o the initial value (1. Default = 25. See Remark 9.01 5 LSTOL MAXQN MAXLS FSTRESS LSTOL MAXBIS < 0.). Line search tolerance (0.0). Default = 20. W or any combination) for convergence criteria. Default = MAXITER. Default = PW.OE7. Default = 1. Maximum number of QuasiNewton correction vectors t o be saved on the data base (Integer 2 0). See Remark 7. < RTOLB Maximum value of incremental rotation (in degrees) allowed per iteration to activate bisection (Real > 2. Default = NO. NO or ALL). > 0).9). Limit on the number of diverging conditions per iteration before the solution is assumed to diverge. See Remark 15. See Remark 14.5. Intermediate output Flag (YES. Default = 0. 5 MAXR 40. Default = 4. See Remark 12. > 0. Default = 5. See Remark 11. See Remark 13.OE3. Error tolerance for work (W) criterion (Real > 0.10). Default = 3. See Remark 11 Maximum number of bisections allowed for each load increment (10 < . < Real < 1. Default = 1. CONV Flags (U. (Integer # 0). Maximum number of line searches allowed for each iteration (Integer 2 0). Error tolerance for displacement (U) criterion (Real Error tolerance for load (P) criterion (Real INTOUT EPSU EPSP EPSW MAXDIV > 0.
and (iii) resumes the normal AUTO option.e.). if KSTEP > MAXITER. 2. 4. gravity loads.e. stiffness matrix is never updated. i. (a) If the AUTO option is selected. Applied loads. Stiffness is updated. The NLPARM Bulk Data entry is selected by the Case Control command NLPARM = ID. In case of creep analysis (DT > O. (ii) estimated time required for convergence with current stiffness exceeds the estimated time required for convergence with updated stiffness. NINC is the number of time step increments. It should be noted that NewtonRaphson iteration strategy is obtained by selecting the ITER option and KSTEP = 1. if (i) estimated number of iterations t o converge exceeds MAXITER. (b) If the SEMI option is selected. etc. NINC is used t o determine the initial arclength for the subcase. I the solution does f not converge in MAXITER iterations. (ii) updates the stiffness matrix. (c) If the ITER option is selected. the program updates the stiffness matrix a t every KSTEP iterations and on convergence if KSTEP 5 MAXITER. 6. temperature sets.. In case of static analysis (DT = 0.) using Newton methods. For AUTO and SEMI options. At each step the number of iterations required t o converge is estimated. NINC is the number of equal subdivisions of the load change defined for the SUBCASE. 5. the program for each load increment (i) performs a single iteration based upon the new load. The unit of D T must be consistent with the unit used on the CREEP entry that defines the creep characteristics. enforced displacements. In case of static analysis (DT = 0. and (iii) solution diverges (See Remarks 9 and 13 for diverging solutions). If the load increment can not be bisected (i.Remarks: 1. However. The number of iterations for a load increment is limited t o MAXITER. define the new loading conditions. the program automatically selects the most efficient strategy based on convergence rates. The stiffness update strategy is selected in the KMETHOD field. and Modified NewtonRaphson iteration strategy is obtained by selecting the ITER option and KSTEP = MAXITER. the stiffness matrix is updated on convergence if KSTEP is less than the number of iterations that were required for convergence with the current stiffness. The differences from the previous case are divided by NINC to define the incremental values. MAXBIS is reached or . 3. the load increment is bisected and the analysis is repeated. Each solution subcase requires an NLPARM command. and the number of load subdivisions will not be equal to NINC. Total creep time for the subcase is DT multiplied by the value in field NINC. DT*NINC..) using arclength methods.
since the computed load increment are always equal to the user specified load increment. i. if 1012 < E. are processed for every computed load increment. 9. Please refer t o Handbook for Nonlinear analysis for more details on convergence criteria. If MAXDIV is negative. I MAXDIV is negative.. MAXBIS is reached or MAXBIS is zero). the analysis is terminated on second divergence. EPSP and EPSW) define the convergence criteria.e. the best attainable solution is computed and the analysis is continued to the next load increment. If the solution diverges again in the same load increment while MAXDIV is positive.. If INTOUT = YES: output requests for displacements. The BFGS update with these QN vectors provide a secant modulus in the search direction. are processed only for the last load step of the subcase.e. NDIV = NDIV NDIV = NDIV + 2. The solution is assumed t o diverge when NDIV 2 JMAXDIVI. ALL: output requests for displacements. the best attainable solution is computed and the analysis is continued to the next load f increment. All the requested criteria (combination of U. The option ALL allows the user to obtain solutions at the desired intermediate load increments. element forces and stresses. The BFGS update is performed if MAXQN > 0. and W = Work error) and the tolerances (EPSU.e. < 1. If MAXQN is reached. NLPCI command is specified) the computed load increment in general are not going t o be equal to the user specified load increment. . 8. 7. etc... element forces and stresses. element forces and stresses etc. P and/or W) are satisfied upon convergence. if E' 1 1 or Ei < 1012 + 1. etc. Depending on the divergence rate. the stiffness is updated based on previous iteration and the analysis is continued. For Newton family of iteration methods (i. For arclength methods (i. and are not known in advance. NO: output requests for displacements. The test flags (U = Displacement error. the number of diverging iteration (NDIV) is incremented as follows: . no NLPCI command) the option ALL is equivalent to option YES. 10. If the solution diverges and the load increment can not be further bisected (i. the analysis is terminated. no additional QN vectors will ..MAXBIS = 0) and MAXDIV is positive. As many as MAXQN quasiNewton vectors can be accumulated. The ratio of energy errors before and after the iteration is defined as divergence rate (E". P = Load equilibrium error. are processed for every computed and user specified load increment.e.
Different actions are taken when the solution diverges depending on the sign of MAXBIS.5 MAXR .. FSTRESS is also used t o establish a tolerance for error correction in the elastoplastic material. The bisection is activated if the incremental rotation for any degreeoffreedom (AB.Alo Aln where Al. Number of bisections for a load incrementlarclength are limited to IMAXBISI. error in yield function < FSTRESS * 7 If the limit is exceeded at the converging state.be accumulated. 15. 1 MAXR < .. i. This bisection strategy based on the incremental rotation is controlled by MAXBIS field. If MAXBIS is negative the load is bisected every time the solution diverges until the limit on bisection is reached. Accumulated QN vectors are purged when the stiffness is updated and the accumulation is resumed. if MAXLS > 0. . The number of subincrements in the material routines (elastoplastic and creep) are determined such that the subincrement size is approximately FSTRESS * T (equivalent stress). the program will exit with a fatal error message. The line search is not performed if the absolute value of the relative energy error is less than the value specified in LSTOL. i. MAXR is used in the adaptive load incremntlarclength method to define the overall upper and lower bounds on the load incrementlarclength in the subcase.e. is the arclength at step n and Alo is the original arclength. The line search is performed as required. Ae. Otherwise the stress state is adjusted to the current yield surface. 11. the analysis is continued or terminated depending on the sign of MAXDIV (See Remark 9). Arclength method for load increments is selected by an NLPCI Bulk Data Entry. If MAXBIS is positive the stiffness is updated on first divergence and the load is bisected on second divergence. 12.e. If the solution does not converge after IMAXBISI bisections. or Ae. This entry must have the same ID as the NLPARM Bulk Data Entry.) exceeds the value specified for RTOLB. 13.. 14. In the line search the displacement increment is scaled t o minimize the energy error.
Default = 0. < Real 5 1. Default = CRIS. Minimum allowable arclength adjustment ratio between increments for the adaptive arclength method (0. > 0) of an associated NLPARM entry.25. Scale factor (w) for controlling the loading contribution in the arclength constraint (Real 2 0. Default = 0.0 0. 12 20 Field ID Contents Identification number (Integer 1.).Parameters for Arclength Methods in Nonlinear Static Analysis. See Remark 2.NLPCI BULK DATA ENTRY (V67) Bulk Data Entry: NLPCI . This entry is activated for a particular subcase if an NLPARM with the same ID is being used in the nonlinear solution sequence (SOL 66).25 4. Default = 4. Description: Defines a set of parameters for the arclength incremental solution strategies. Default = 12. RIKS. MAXALR SCALE DESITER .). See Remark 4.). Format: NLPCI ID TYPE MINALR MAXALR SCALE DESITER MXINC NLPCI 15 CRIS 0. or MRIKS). See Remark TYPE MINALR Constraint type (CRIS. Desired number of iterations for convergence to be used for adaptive arclength adjustment (Integer > 0). See Remark 3. See Remark 2.0. See Remark 3. Maximum allowable arclength adjustment ratio between increments for the adaptive arclength method (Real 2 1.
in case of Creep analysis (DT # 0.)~{U. the arclength method would be degenerated t o the conventional Newton's method. p. This field is useful in limiting the the number of increments computed for a collapse analysis. MINALR 5 MAXALR 4.p 0)2 . the constraint equation is gradually dominated by the load term. number of iterations required for convergence) and the change in stiffness. and the NLPCI Bulk Data entry is ignored if specified. 2. The scaling factor (w) is introduced as user input so that the user can make constraint equation unitindependent by a proper scaling of the load factor. The constraint equation has a disparity in the dimension by mixing the displacements with the load factor.U .MXINC Maximum number of controlled increment steps allowed within a subcase (Integer > 0). p is the load factor and A1 is the arclength.uO.~ . . In the limiting case of infinite w. .PO) = 0 : CRIS: {u. .) and the number of iterations desired for convergence in the current load increment ( I d ) . ..~ ) ~ { Uu:)~ + w 2 ~ p i A P = 0 1 MRIKS: {uk . as follows:  5. The available constraint types are as follows . The MINALR and MAXALR fields are used to limit the adjustment of the arclength from one load increment to the next by Alold The arclength adjustment is based on the convergence rate (i. use MINALR = MAXALR = 1. The NLPCI Bulk Data entry is selected by the Case Control command NLPARM = ID. For constant arclength during analysis. However. See Remark 5.u:) + w2(pi . As the value of w is increased.' ) ~ { U ~ . The MXINC field is used to limit the number of controlled increment steps in case the solution never reaches the specified load.. Remarks: 1.a12 . There must also be an NLPARM Bulk Data entry with the same ID. The arclength A1 for the variable arclength strategy is adjusted based on the number of iterations that were required for convergence in the previous load increment (I. 3. where w is the userspecified scaling factor (SCALE).U ~ .e.u) + w2api(piI .n RIKS: (21.. in NLPARM Bulk Data entry) the arclength methods can not be activated. Default = 20.
) Area moment of inertia in plane 1 about the neutral axis (Real 4. See Remark .Beam Property (Alternate Form of PBEAM) Description: Alternate format to PBEAM entry to define properties of a uniform crosssectional beam referenced by CBEAM entry. > 0. See Remark 1. > 0. > 0). See Remark 2 and 3.). This entry is also designed to specify lumped areas of the beam cross section for nonlinear analysis and/or composite analysis. Format: Field PID MID A I1 Contents Property identification number (Integer Material identification number (Integer Area of beam cross section (Real > 0).PBCOMP BULK DATA ENTRY (V67) Bulk Data Entry: PBCOMP .
Fraction of the total area for the ith lumped area (Real N1. default = 0). N2 SECTION Yi. > 0). default = 1. a maximum of 20 lumped areas may be input if SECTION=5.3. .e. K2 > O. A maximum of 21 continuations is allowed..z) coordinates of center of gravity of nonstructural mass (Real. (y.). (y. default = NO). The PID number must be unique with respect to other PBCOMP entries as well as PBEAM entries. If SECTION=l through 4. Symmetry option to input lumped areas for the beam crosssection (integer 0 to 5.I2 Area moment of inertia in plane 2 about the neutral axis (Real 4.).. the total number of areas input plus the total number generated by symmetry must not exceed 20. no default).. and Remark 6 . Section 1. default = O. If the user does not specify these. See Remark I12 Area product of inertia (Real.). default = 0. See Figure l a . Remarks: 1. See Remark 4. M2 (y. M I . i. Section 1.3.3. default = 0.). SOUTi Stress output required (YES) or not (NO) for the various lumped areas (Character. default = 0). default = 0). Zi Ci MIDi > O.. See Figure lg.. > 0. Section 1.z) coordinates of neutral axis (Real. The third continuation type may be repeated 18 more times. default = 0.0).) Shear stiffness factor K in (KAG) for plane 1 and plane 2 (Real See Remark 5. The PBCOMP entry creates equivalent PBEAM entries in a preface operation.z) coordinates of the lumped areas in the element coordinate system. See Figure l a . See Remark 1. 1112 Torsional stiffness parameter (Real > 0. See Re Material identification number for the ith integration point (Integer mark 3. (Real. > 0. See Remark 7. See Remark 4. J NSM K1. Nonstructural mass per unit length (Real > O.
the material should be perfectly plastic since the plastic hinge formulation is not valid for strain hardening. For material nonlinear analysis. the value will be that of MID on the parent entry. PBCOMP entries may reference only MAT1 material entries. MIDis will follow the same symmetry rules as Ci depending on the value of SECTION. MID. zi. and C. For heat transfer problems. fields 11.. The values Eo and Go are computed based on the value of MID on the parent entry. PBCOMP entries may reference only MAT4 and MAT5 material entries. These and other modified values will be calculated based on the input data (y. . C. If the lumped crosssectional areas are specified..) as follows: where n is the number of lumped crosssectional areas specified.. 12. values. z. Section 1. t o allow different stressstrain laws.the program defaults t o the elliptically distributed 8 nonlinear rods as usual. For structural problems. 4.. 2.3. and 112will be ignored. field on a continuation is blank. MID. See Figure lg. values may be input on continuations without the corresponding y. If MID. 3.
the transverse shear stiffness becomes rigid and the transverse shear flexibilities are set to 0.0. If one of the symmetry options is used and fewer than four lumped areas are input explicitly. As can be seen from Figure lg. the symmetry option will double the areas. The SOUTi field is a provision for features to be implemented later. he may specify the stressoutput location on continuation entries. Section 1. 6.I2 and II2 on the parent entry. 7. similar to the PBEAM entry. If the sixth field of this entry contains a string (such as "YESn or "NOn).5. If a value of 0. the sequence of output locations in the imaged quadrants is shown in Figure lg. Stress for the lumped areas is output regardless of the value of this field. If it is blank or real. For a doubly symmetric section (SECTION = I). In such cases. Stress output is provided at the first four lumped area locations only. and MID. For one specific example. values. The (y. it is recommended that users input the value of area as half of the actual value at point 2 to get the desired effect. . Blank fields for K1 and K2 are defaulted to 1. it is intended t o be for general beam data. if the user chooses to leave the SECTION field blank. For this particular case it is illegal to supply C.0 is used for K1 and K2. For the case where the user specifies 11.'s set to 0.3. output can be obtained at points 1 and 2 and in the image points 3 and 4. the entry is regarded as data for a beam station.0. if the lumped areas are specified on either axis. Its main function is to allow the program to determine if the first continuation is intended for general beam data (the data starting with K1) or data at a beam station (the data starting with Yl). For example for the section shown points 2 and 4 are coincident and so are points 6 and 8. the program defaults to the elliptically distributed 8 nonlinear rods.z) coordinates specified on these entries will serve as stress output locations with the corresponding (2. in the model shown in Remark 6.
. or without the continuation for general beam data ( K l . SECTION) but with the beam station data.. . . This practice is consistent with similar provisions on the PBEAM entry. K2. .Stated another way. the PBCOMP entry may be used without any continuations.
Gap Element Properties Description: Defines the properties of the Gap element Format: 1 2 3 4 5 6 Kb 7 8 PI 9 P2 10 PGAP PID TMAX u o MAR Fo TRMIN Ka Kt PGAP +PG 2 0.001 1..PGAP BULK DATA ENTRY (V67) Bulk Data Entry: P G A P . Default = 0. U .E2 1. 0.0.0. Default = pll'i. See Remark 2. U. . i. Kb Kt . > .Ub < Uo. 0. See Remark 1. 0. 0. See Remark 3. +PG Field PID Uo Contents Property identification number (Integer > 0). Transverse stiffness when the gap is in contact (Real 0. Axial stiffness for the open gap. Default = 0.0). (Real > 0. Initial gap opening (Real).Ub > Uo. (Real 2 0. F o K. loo. See Remark 1.e. Recommended range of K t is O.e..0).0).E6 l. Axial stiffness for the closed gap.OIKa 5 I'it 5 K. Default = 10'Ka. Preload (Real). i.E5 0.
Upper . for adaptive gap element p2 5 p1). 3.0.001. value may slow down convergence or cause divergence. Figures 1 through 3 show the GAP element and the forcedisplacement curves used in the stiffness and force computations for the element. See Remark 4. This value is adjusted as necessary if TMAX > 0. K . For most contact problems. The value of TMAX = 1. The recommended allowable penetration TMAX is about 10% of the element thickness for plates or equivalent thickness for other elements which are connected to the GAP. Default = 0. Maximum allowable penetration used in the adjustment of penalty values (Real). < Remarks: 1. Default = p l . See Remark 5. MAR < lo6). gap induced bisection. TMAX MAR ' A Maximum allowable adjustment ratio for adaptive penalty values . When the GAP is open. the adaptive GAP element is chosen. If TMAX 0. 2.0. while a much smaller K . There are two kinds of GAP elements: adaptive GAP and nonadaptive GAP. See Remark 6.) for the nonadaptive gap element (Real 2 0. Default = 0.0). > PP Coefficient of kinetic friction ( p k )for the adaptive gap element or coefficient of friction in z transverse direction (p. Positive value activates the penalty value adjustment. the GAP has the elastic stiffness ( K t )in the transverse direction until the friction force is exceeded and the slip starts to occur. > 5. and subincremental process). The maximum adjustment ratio MAR is used only for the adaptive GAP element. Default = 0. A much larger K .) for the adaptive gap element or coefficient of friction in y transverse direction (p.0. value may result in an inaccuracy. 4. defining the lower bound for the allowable penetration (0.0.0.) for the nonadaptive gap element (Real 0.PI Coefficient of static friction (p. When the GAP is closed and if there is friction.0 selects the nonadaptive (old) GAP element.0).0 5 Real 1. and Kt (1 < TRMIN Fraction of TMAX. Default = lo2. (penalty value) should be chosen to be three orders of magnitude higher than the stiffness of the neighboring grid points. penalty values will not be adjusted but other adaptive features will be active (the gap induced stiffness update. When TMAX = 0. there is no transverse stiffness.
or Kt. T h e penalty values are decreased if the penetration is below the lower bound.and lower bounds of the adjusted penalty are defined by Kinit MAR < K < Kinit MAR  where Kinit is the userspecified value of K . The GAP Element. . 6. The parameter TRMIN is used only for the penalty value adjustment in the adaptive GAP element. Figure 1. T h e lower bound for the allowable penetration is computed by TRMIN * TMAX.
Ub a : (compression) Figure 2. Shear Force for GAP Element. ForceDeflection Curve for Nonlinear Analysis. .F. (compression) t tension u o I + u . Nonlinear Shear Figure 3.
0 10 175. Temperature values must be listed in ascending order. An error is detected if any continuations follow the entry containing the endoftable flag ENDT. The end of the table is indicated by the existence of ENDT in either of the two fields following the last entry.Material Property TemperatureDependence Table Description: Specifies the material property tables for nonlinear elastic temperaturedependent materials Format: 1 TABLEST 2 3 blank I1 4 blank 5 6 blank T3 7 blank etc. 4.TABLEST BULK DATA ENTRY (V67) Bulk Data Entry: TABLEST . . ENDT Field TID Ti Contents Table identification number (Integer Temperature values (Real) Table identification numbers of TABLES1 entries (Integer > 0) Ii > 0) Remarks: 1. 2. 8 blank 9 10 TID T1 blank I2 T2  TABLEST 101 150. 3. I must be unique with respect to all TABLES1 and TABLEST table identification numi bers. This table is referenced only by MATS1 entries which define nonlinear elastic (TYPE'= NLELAST) materials.0 20 etc.
Form 1 Description: Defines a tabular function for stressdependent material properties such as the stressstrain curve and creep parameters Format: 1 TABLES1 2 ID xl 3 4 5 6 7 8 9 10 blank Yl blank x2 blank Y2 blank x3 blank Y3 blank x4 blank Y4 TABLES1 32 0. ENDT Field ID xi. 4.01 10000.0 0. . 3.TABLES1 BULK DATA ENTRY (V67) Bulk Data Entry: TABLESl .02 15000. Jumps between two points (xi = are not allowed. Any xy entry may be ignored by placing SKIP in either of the two fields used for that entry.Material Property Table.0 . but not both. . yi Contents Table identification number (Integer Tabular entries (Real) > 0) Remarks: 1. 2. The xi must be in either ascending or descending order. At least two entries must be present.
. An error is detected if any continuations follow the entry containing the endoftable flag ENDT. The table lookup yT(x). 6. The end of the table is indicated by the existence of ENDT in either of the two fields following the last entry. TABLES1 is used to input a curve in the form of where X is input to the table and Y is returned. x = X. is performed using linear interpolation within the table and linear extrapolation outside the table using the last two end points at the appropriate table end.5.
E6 0 2 0. AUTO. Time increment (Real 2 0.E2 5 1 ADAPT 10 16. See Remark 2. See Remark 1. (Integer > 0).1 10 0.TSTEPNL BULK DATA ENTRY (V67) Bulk Data Entry: TSTEPNL .E3 5 l. PW l. TSTEP. Format: 1 2 3 4 5 6 7 KSTEP MAXLS UTOL 8 9 10 TSTEPNL ID EPSU MAXBIS NDT EPSP ADJUST DT EPSW MSTEP NO MAXDIV RB METHOD MAXQN MAXR MAXITER FSTRESS RTOLB CONV TSTEPNL 250 l.75 Field ID NDT DT NO Contents Identification number (Integer > 0). Number of time steps of value DT (Integer 2 4). See Remark 4.0 2 2 0.0).2 20. Time step interval for output: every ~0~~ step solution will be saved for output. METHOD . Description: Defines a set of parameters for nonlinear dynamic solution strategy. or ADAPT for dynamic stiffness update and the direct time integration strategy. Default = 1. Default = ADAPT. See Remark 5.Parameters for Nonlinear Dynamic Analysis. See Remark 3.
75. Limit on number of iterations for each time step (Integer Remark 7. ADJUST Time step skip factor for automatic time step adjustment (Integer 5. See Remark 8. P. MAXITER # 0). See Remark 9. Error tolerance for displacement (U) criterion (Real Error tolerance for load (P) criterion (Real EPSU EPSP EPSW MAXDIV > 0.0). Error tolerance for work (W) criterion (Real > 0.1 5 Real 5 1). RB MAXR . Default = variable between 20 and 40. Default = 1. See Remark 10. Default = 2. Default = 0. Define bounds for maintaining the same time step for the stepping function in the automatic time step adjustment method (0.OE2. MAXQN MAXLS > 0). > 0.0). Default = 0.0). See Remark 6.KSTEP Time step interval in TSTEP method for stiffness update or stiffness update interval in bisection process for ADAPT method.0). See CONV Flags (U. Default = 2.2. See Remark 11. Default = 16. Default = MSTEP Desired number of time steps to obtain the dominant period response accurately (10 5 Integer 5 200). < Real < 1. Default = 10. Maximum number of quasiNewton correction vectors t o be saved on the data base (Integer 2 0). Default = PW.OE3. 2 0). See Remark 16. (Integer > 0). Default = 1. FSTRESS Fraction of effective stress (7) used to limit the subincrement size in the material (0. See Remark 14. Maximum ratio for the adjusted incremental time relative t o D T allowed for time step adjustment (1. Limit on the number of diverging conditions for a time step before the solution is assumed t o diverge (Integer > 0). W or any combination) for convergence criteria. routines Maximum number of bisections allowed for each time step (9 Default = 5. Default = 10. MAXBIS 5 MAXBIS 5 9). Real 5 32. See Remark 13. See Remark 15. See Remark 12.OE6. Default = 2.). Default = 1. Maximum number of line searches allowed per iteration (Integer See Remark 10.
the time increment (At) changes during the analysis in ADAPT option. Multiple subcases are assumed to occur sequentially in time. RTOLB Remarks: 1. the actual number of time steps. (b) If the TSTEP option is selected. 2. . Maximum value of incremental rotation (in degrees) allowed per iteration to activate bisection (Real > 2. in general. (c) If the ADAPT option is selected. 2N0. Since. Default=20.. For printing and plotting the solution. The KSTEP value is ignored. the program automatically adjusts the incremental time and uses bisection.e. The TSTEPNL Bulk Data entry is selected by the Case Control command TSTEPNL = ID. See Remark 18. The stiffness matrix is always updated for a new subcase or restart. 3. The time increment (At) remains constant during the analysis in AUTO and TSTEP options.001 < Real 5 1. . data will be output at steps = 0. the program automatically updates the stiffness matrix to improve convergence. 4. etc. NDT is the number of time steps in AUTO and TSTEP options (KMETHOD field). irrespective of the option selected. will not be equal to NDT.. See Remark 17. the program updates the stiffness matrix every KSTEP~~ increment of time. The stiffness update strategy as well as the direct time integration method is selected in the METHOD field. and the last converged step. (a) If the AUTO option is selected.. NDT*DT).. The Case Control command OTIME may also be used t o control the output points. The ADAPT method allows linear transient analysis.0).1. and is equal to DT. but AUTO or TSTEP will abort the run if the model does not have any data representing nonlinearity. So the initial conditions of each subcase are defined by the end conditions of the previous subcase. In ADAPT option DT is used as a initial value for At.0). In ADAPT option NDT is used to define the total duration for analysis by multiplying NDT with DT (i. Default = 0. 5. NO. Each subcase (residual superelement solutions only) requires a TSTEPNL entry.UTOL Tolerance on displacement increment below which there is no time step adjustment (0.
Otherwise. 10. the solution for the time step is repeated with a new stiffness based on the converged state at the beginning of the time step. the program computes the best solution and continues the analysis on second divergence during a time step. 7. the time step is bisected upon divergence. FSTRESS is also used t o establish a tolerance for error correction in the elastoplastic material. During the bisection process in ADAPT option. MAXDIV provides control over diverging solutions. If MAXBIS is negative and the solution does not converge in (MAXBISI bisection. All requested criteria (combination of U. The parameter ADJUST allows the user t o control the automatic time stepping in the ADAPT option. If the solution does not converge in MAXITER iterations. Nonzero values of MAXQN and MAXLS activate the quasiNewton update and the line search process. If NDIV reaches MAXDN again within the same time step. the program will exit with a fatal error message. error in yield function < FSTRESS * B If the limit is exceeded at the converging state. It should be noted that at least two iterations are necessary t o check the displacement convergence criterion. If ADJUST is positive. W = Work error test) and the error tolerances (EPSU. P = Load equilibrium error test. If MAXBIS is positive and the solution does not converge after MAXBIS bisections. stiffness is updated every KSTEP converged bisected solution to reduce computing cost. If MAXITER is positive. The number of subincrements in the material routines are determined such that the subincrement size is approximately FSTRESS * 5 (equivalent stress). and/or W) are satisfied upon convergence. i. Depending on the rate of divergence. the process is treated as a divergent process (See Remark 9) 8. 9. the analysis is terminated. The number of bisections for a time increment is limited to (MAXBISI. 11. P. The bisection process is activated when divergence occurs and MAXBIS # 0. the best solution is computed and the analysis is continued to the next time step. The number of iterations for a time step is limited to MAXITER. EPSP and EPSW) define the convergence criteria. The convergence test flags ( U = Displacement error test.e.6. the stress state is adjusted t o the current yield surface. A value of zero for ADJUST turns off the automatic adjustment completely. the analysis is terminated when the divergence condition is encountered twice during the same time step or the solution diverges for five consecutive time steps. 13. The solution is assumed t o diverge when NDIV reaches MAXDIV during the iteration. If MAXITER is negative. the time increment is continually adjusted for first few .. respectively. the analysis is terminated. the number of diverging solutions (NDIV) is incremented by 1 or 2. If the bisection option is used (allowed MAXBIS times) with the ADAPT method. 12. Otherwise.
Parameters MSTEP and RB are used t o adjust the time increment during analysis in the ADAPT option. A value of ADJUST a n order greater than NDT will turn off adjustements after the initial adjustement. This bisection strategy based on the incremental rotation is controlled by MAXBIS field../RB < > 16.25 for r < 0. where with f f f f f = 0. T h e time increment is adjusted as follows: At. In the default option. T h e bisection is activated if the incremental rotation for any degreeoffreedom (AB. 14. The ADJUST option should be suppressed for the duration of short pulse loading. start with D T which is much smaller than the pulse duration in order t o properly represent the loading pattern.. r < 3.0 for r 3. 5 s R B 5 r < RB = 1.) exceeds the value specified for RTOLB. = f ( r ) At.0 for 2. A larger value (say 40) is required for highly nonlinear problems. DT (2MAXBIS ' m)< At DT 5 MAXR * D T 17..steps till a good value of At is obtained.e. no time step adjustment is performed if 18.0 for RB 5 r < 2 = 2. After this initial adjustment. i.e. or AO./RB = 4.5 * RB = 0. MAXR is used t o define the upper and lower bounds for adjusted time step size. 15. the program automatically computes the value of MSTEP based on the changes in the stiffness.5 for 0 . it may be necessary t o limit the adjustable step size when the period of the forcing function is much shorter than the period of dominant response frequency of the structure. UTOL is used t o filter undesirable time step adjustment.+. the time increment is adjusted every ADJUST time steps only. It is user's responsibility t o ensure that the loading history is properly traced with ADJUST option. AB. The adjustment is based on the number of time steps desired t o capture the dominant frequency response accurately. . The recommmended value of MSTEP for nearly linear problems is 20. i. Since the automatic time step adjustment is based on the mode of response and not on the loading pattern. If unsure.
.
6.53 for arclength method.34 bisection algorithm. 7. 4.13. 2.14 BETA. 6.17.13.313 anisotropy. 1. 3.1 apparent Poisson's ratio. 5.11.55 boundary conditions.64 convective coordinates.2. 9.44.41 Almansi finite strain.1 BUCKLE. 3.18 absolute divergence.41 AUTOSPCR.48. 3. 11. 5.21 element. 8.42 bilateral stressstrain relation.22 Bauschinger Effect. 5. 1. 11. 4.11 bulk data. 3. 3. 4. 4. 9.16 associated flow rule. 3. 8. 3.31 convective.51 contact analysis.14 coordinate transformations.65. 7. 5.65 coordinate system. 5. 9.5.6.52. 3. 1. 4. 6.31 strain.4.18 criteria.1 material.61 information output. 7.83 1 .75.21 criteria. 3.610 constraint. 11.65 Coulomb friction law.212.48 BFGS update.71. 5.1 contraction factor.62 ALRATIO. 8.6. 6.11.1 convective rate of change.36 combined hardening. 3.13 Index  buckling. 5.31. 8. 4. 3.210 constitutive relations.44 GAP.41.32 penalty value adjustment.16 angular velocity tensor.42 basic coordinate system.4. 8. 5. 6.42. 6.33 adaptive arclength method. 3. 11. 3.65 convergence. 9.12 solution algorithm.1 absolute convergence.51 continuum mechanics. 7.65.6 BEND.12.37 COUPMASS.227 arclength method.33 CREEP.31. 6.11.311.46 case control. 5.22 BEAM. 5.11 Cauchy's deformation tensor.32 tolerance.2. 3. 5. 2. 1.22 corotational stress rate tensor. 1. 11. 6.5.17 rate.12.Index absolute convergence.15 creep analysis.48 auxiliary angle.51.11 biaxial loading. 5. 7. 3. 8. 8.22 isoparametric. 9. 7.14 automatic time stepping. 1. 2.2.63 CDAMPi. 3. 6. 7.74 bisection.81 strain rate. 5. 2.36 Newmark's method.31. 9. 5. 4. 4.76.11 time stepping. 4. 10. 6. 10. 6.44.65 anisotropic friction. 4.62 Cauchy stress tensor.
9.28. 4. 6.62. 8.21.23. 7.65 friction element.13 DrukerPrager yield criterion. 9. 3. 7.15  2 . 6. 3.63 error vector. 2. 5.13. 4.7.14 dynamic loads. 4. 7.219 element coordinate system.21 error functions.61.51 elastoplasticcreep relations. 5.24 SOL 99 and 129.37 modeling.31 finite element method.66. 3.3.41 effective plastic strain increment. 8.49. 1. 4. 3. 3.2.51. 8. 1. 4. 10.17.17. 1.17 equilibrium.31 G.Crisfield's method. 5. SOL 66 and 106. 9. 5.73. 1. 2.17 EPI. 9. 9.44 line search.17. 7.14 CVISC. 3. 9. 7. 5.33 frictional yield function.34 discretization.46 elastoplastic material.12 stiffness.214 effective strain for nonlinear elastic.12 displaced element coordinate system. 9. 8.52. 11.11 Eulerian formulation.61 EWI.31 doubling scheme. 5. 9.12 DBDRNL.43 time expiration.15 follower matrix.21.17 Eulerian strain.17 Executive System.11 discrete damping elements. 3. 5. 5. 6.36 cyclic loading. 6. 3.16 DFP method.61 divergence. 9. 9.23 critical buckling load.11 finite strain. 9. 7. 2. 3. 3.11 geometric stiffness. 7. 2. 7.12. 7.610 geometric nonlinearity. 3. 7.51.25.21. 1.32. 6. 2.16 criteria BFGS update.26 enforced motion. 2.34 numerical.17. 9.6.31.1.4. 6. 6.21 EUI.18 FBS. 5.31 EXP. 3.229 damping matrix.31.53 convergence. 7. 6.37 FSTRESS. 3. 3.16 FACTOR.51 differential stiffness.25 DT. 1. 3.44 DLMAG. 2. 9. 7. 7. 9. 1. 5.66.37 GAP element.18 EIGB. 1.15 functional module.44. 7.31 EFIRST.55 FIAT (File Allocation Table). 9.214 generalized logarithmic strain.23 displacement sets.45 ENIC.1 generalized effective plastic strain.18 Index EFIIL'AL. 3. 8.51).1 divergence criteria. 2.18 DMAP.23. 5. 8.52 diagnostic outputs (DIAG 50. 4.33 yield.12 CSHEAR. 5.36.15.3 DESITER.21. 3.32 frameindifference. 5.51 Gaussian elimination method. 4.12 nonlinear analysis.48 deformation gradient. 2.22 matrix. 5.31. 6. 3.15 direct time integration. 4. 4.62 follower force.32.
63 LGDISP. 7.31 threepoint method.11. 4.12.31. 3. 1.10 KMETHOD.13. 4.49.51 Jaumann stress.31 superelement.61 implicit integration.68.14 Index  Lagrange multiplier method. 5.1 isotropic hardening.62 Herzian problem. 8. 9.64.75. 11. 7.81 kinematic hardening.325 Hessian matrix.32 hyperelastic material. 3. 9.11 Kirchhoff stress. 7. 2.12 Newmark. 8. 6.43 iteration. 1.44. 5. 6.63.61 material coordinate system.12. 3. 7.52.10 material nonlinearity. 5.44 impact analysis.44.65. 11.2.17. 2.41.61 Lagrangian strain.11 material properties entries.31 Illinois algorithm. 5. 2. 10. 1.410 criteria.15 mass matrix.33 LOADINC.22.1 properties.42 GINO.51 hierarchy of nonlinear looping. 4.12 line search.26 implicit. 5.21 algorithm.61 input data structure. 6. 8. 7. 9. 6.2 initial anisotropy. 8. 3.49 loading and unloading. 6. 2.63 Lagrangian formulation. 6.3 Lagrangian strain rate. 3.3.5. 5.12.31.29 Kirchhoff & Love.11 hysteresis. 11. 5.12 line elements.10 KGROT.31 MAT2.3 large rotation.32 modeling.21 local coordinate system. 2. 2. 4. dynamic.49 LSEQ. 3. 1.411 linear interpolation. 6. 8.41 gimbal lock. 11. 5. 9.12 MAXALR.31 LSTOL. 5. plate theory.610 LAMBDA.1. 5. 7. 6.gimbal angles.31 global coordinate system.29 iteration. 9. 1.114 material derivative.31 logarithmic strain. 5.14 isoparametric elements.12 increment. 1. 3.11. 1. 5.3. 4. 3.43.62 Green's finite strain.61 MAT9. LIMIT2. 1.11 Gauss. 5. 5.229 IFP (Input File Processor). 5. 6.13 load.41. 5. 3. 9. 4.31 INTOUT.49. 2. 11. 5. 5. 6. 1. 9. 8.51 left stretch tensor.12 LIMIT1.31.41 static.10 LOOPID.41.62. 8.112 KSTEP.6.2.31 Green's deformation tensor.41 integration direct time.16 3 . 1.48.64.18 LANGLE. 5. 3.71 KelvinMaxwell model. 6. 4. 5.61 material stiffness.21 twopoint method.22 MATS1.76. 3. 7. 6.31 Lagrangian description. 8. 9. 3.51.
6. 7. 7.81 method of false position.3 MPL (Module Properties List). 3. 1. 8.21 NINC.14 MAXLP. 7.14 plotting output codes.114 penalty GAP element. 1. 9.12 NLPCI. 6. 4.4.4. 4.15 plane stress.2.31 MXINC.17.53 PiolaKirchhoff stress.15 MAXR. MaxwellKelvin model. 1.51 MohrCoulomb yield criterion.23 nonlinear modal analysis.31 modeling.35 plane strain.64 NDAMP. 7. 3. 3.15.3 penalty values adaptive adjustment.2.18 numerical damping.49 NLITER. 1. 7.12 NDIV.36. 7.31 output diagnostics.61 BFGS updates.21 input data block. 2. 7. 4.217. 1.11 nonlinear transient response analysis.2 output data block. 1.15. 7.5 MINALR.15 MAXDIV.11. 2. 1. 11. 3.21 penalty value in GAP.51.31 Newmark integration.27 NOGO flag. 4.23.15. 1.15 MAXQN. 9. 6. 4.15. 7. 9. 3.21. 8. 8.17.31 NewtonRaphson method. 1.63 PGAP.14. 3.3 parameters.72 line search.51 nonlinear elasticity.32 NMEP.21.18 plane strain element.317 static solution method.36 selection of Gap stiffness.417 4 . 7. 9. 9. 4. 9.15. 5. 11. 5.32.25 MPC.16 modal coordinate system. 3.13 NQNV. 8.42. 9. 3.42 NEWK. 8.12 properties.49. 7.48 parametric studies automatic time stepping. 4. 10.17 line search.410.24.11 nonproportional loading. 7. 10. 6. 1. 9.31 open/close boundary.51 NLTRD2 module. 2.17 natural strain. 7.313 nonlinear capabilities.112 MAXITER.21 nodal forces.3 NMATD.MAXBIS.52. 6. 4.1 normality principle. 9.48 parameters.51 interpretation for large rotation.49 MAXLS. 7. 9. 11.56. 7.4.12 onestep integration.41 nonadaptive GAP element.4.33. 6. 5. 9. 2. 11.15.43.74. 6.4 NLPARM. 4. 7.16 NLTRD module.75.44. 8. 3. 6.1 Index  nonlinear characteristics.44 neutral loading. 11. 7.4.311 as input parameters.13 NLAYERS. 9. 7.224 NMEVD.49.32. 8.41. 1.46 iteration process. 11. 1. 3. 8. 1. 11.52.22.36.21 KGROT. 7. 1.32 NMLOOP.55. 9. 7. 8.21 nonlinear static analysis.39 OSCAR.22.21 plate & shell elements. 11. 2.21. 6. 7.31 nonlinear element. 6.31. 3. 4. 9.33 NOLIN.218 plasticity.
4 static loads. 3. 1. 6. 7.23.81 right stretch tensor.14 slip yield surface. 1.18.81.11. 7. 11. 8.72 quadratic convergence.41 Index  5 .13 capabilities.12 PrandtlReuss stressstrain relation. 2.64 reduction of system equations.41.31 user interface. 2. 9.29 structual damping. 1.413 quasiNewton ( Q N ) update.17 SCALE.52 SOL 66 and 106.51 stress.21.15 element.11.51 user interface.38 stickslip boundary. 2. 1. . 7.71. 11. 5. 5.61 in SOL 99 and 129.2.3 tertiary creep. 11. 8. 1.413 SYSTEM(145). 3.62 postbuckling analysis. 6.23..17 SDATA. 3. 7. 1. 1.11 spatial derivative.41 shape function. 5. 4. 6.. 10.42.410 snapthrough. 1.21 RF63D89.31 STIME. 11. 2.9 strainrate increment.polar decomposition. 6. 5. 5. 9. 1. 3.3 shell element.4.2.22 secondary creep. 4.212 Preface Operation. 3.43 RT.4 spin tensor. 6.38 SLOOPID.14 solution method.411 stability analysis.34.39 stiffness differential. 11.16. 5.26 geometric.15 material.411 tangent material stiffness.51. 6.51.51 tangential stiffness. 4. 5.22.71.21. 6. 7. 5. 7. 5. 9.11 SOL 64. 1.3 rotation vector. 5.64 SPC. 9.61 static analysis.410. 2. 5. 3. 6. 11.21 QUAD4.41. 3. 1.14 updates. 4. 3. 6. 1.52 TABS. 3. 4. 3. 6.22 examples.411.51.31 stress stiffness.51.41.31. 3.33 program architecture.51.12. 2. 4.46.14 SYSTEM(141).512 TABLEST. 6..21. 5.22 TEMP.14 stretching tensor.31 proportional loading.21 SMALLDB.27 rheological model. 10.16 Ritz method.11 solid elements. 4.411 subincremental algorithm. 3. 4. 9.39 superelement. 4. 11. 5.22 quadratic interpolation.11 SOL 99 and 129. 4.64 stickslip behavior. 1. 7. 11.21 ROD element.72.33 surface elements.21 strain hardening. 1. 11.9. 4. 6.21 rate of deformation tensor. 9.1 probable divergence. 1.410. 3. 7. 9. 3. stress data recovery.31 primary creep.16 RTOLB.41.5 SUBID.21.41 restarts in SOL 66 and 106.64 SSG3. 6. 7.26 examples.63 Rik's method.
63. 9.1 uniaxial stressstrain.13 follower forces.61 large rotation. 2.41 plasticity.412. 10. 10.11. 2.51 thermal stress analyses.5 volume integration.37 creep. 1.53 required input. 6. 6. 8.7 von Mises yield criterion. 3.26 function derivatives. 11.81 viscous damping.64 W3. 6. 7. 4. 3. 4.25 TRIA3. 6. 9.7 transverse shear stiffness.1. 5. 4.23 static analysis in SOL 66.31 thermal strain. 5.11 nonlinear modal analysis. 7.412 weighted normalization.8 TSTATIC.TESTNEG. 10.21 nonproportional loading.413 enforced displacement. 8.218.21. 10. 10.22.17 nonlinear modal analysis.19 UPARTN.31 material nonlinearity.53. 6. 6. 5. 5.21 static analysis.34 thermoelasticity. 7. 6.71 static analysis in SOL 99. 4.66. 5.329 thermoelasticity.16 time expiration criteria.227 restart in SOL 66. 6.1 loads.412.51 threepoint method.34 geometric nonlinearity.215 stress. 8.3 function.16 buckling analysis. 1. 7. 9. 7.33 uniaxial stress.33 total Lagrangian. 8.325.61 viscoelastic.2.42 updated Lagrangian. 6.31.52 friction. 10.11 velocity gradient.51 nonlinear buckling analysis. 1. 6. 4. 6.4 yield yield yield yield yield criteria.51 transient analysis. 8.23 vorticity tensor. 4.79 buckling analysis.21 surface. 9. 3. 9. 1.314 initial anisotropy. 5. 6. 6.26. 6. 11.41 arclength.41.6. 6.31 TSTEPNL. 6. 6.12 creep analysis.15 GAP element. 8.61.3.7 user interface.36 Tresca yield criterion. 4.34. 7.53 transient response analysis. 10. 6.31 twopoint method. 5.32 surface contact.31. 10.72. 6. 6. 6.319 Index GAP.62 restart in SOL 99.61 arclength method. 7.412 W4.21  6 . 3.11 thermoelasticity. 1. 7.64 verification and validation problems adaptive time stepping. 6.49. 6.21 THRESH.21.
This action might not be possible to undo. Are you sure you want to continue?
Use one of your book credits to continue reading from where you left off, or restart the preview.