MD Nastran R3
Explicit Nonlinear (SOL 700) User’s Guide
Main Index
Corporate
MSC.Software Corporation 2 MacArthur Place Santa Ana, CA 92707 Telephone: (800) 3452078 FAX: (714) 7844056
Europe
MSC.Software GmbH Am Moosfeld 13 81829 Munich GERMANY Telephone: (49) (89) 43 19 87 0 Fax: (49) (89) 43 61 71 6
Asia Pacific
MSC.Software Japan Ltd. Shinjuku First West 8F 237 Nishi Shinjuku 1Chome, ShinjukuKu Tokyo 1600023, JAPAN Telephone: (81) (3)69111200 Fax: (81) (3)69111201
Worldwide Web
www.mscsoftware.com
User Documentation: Copyright © 2008 MSC.Software Corporation. Printed in U.S.A. All Rights Reserved. This document, and the software described in it, are furnished under license and may be used or copied only in accordance with the terms of such license. Any reproduction or distribution of this document, in whole or in part, without the prior written authorization of MSC.Software Corporation is strictly prohibited. MSC.Software Corporation reserves the right to make changes in specifications and other information contained in this document without prior notice. The concepts, methods, and examples presented in this document are for illustrative and educational purposes only and are not intended to be exhaustive or to apply to any particular engineering problem or design. THIS DOCUMENT IS PROVIDED ON AN “ASIS” BASIS AND ALL EXPRESS AND IMPLIED CONDITIONS, REPRESENTATIONS AND WARRANTIES, INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, ARE DISCLAIMED, EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD TO BE LEGALLY INVALID. MSC.Software logo, MSC, MSC., MSC/, MD Nastran, ADAMS, Dytran, MARC, Mentat, and Patran are trademarks or registered trademarks of MSC.Software Corporation or its subsidiaries in the United States and/or other countries. NASTRAN is a registered trademark of NASA. MSC.Nastran is an enhanced proprietary version developed and maintained by MSC.Software Corporation. Python is a trademark of the Python Software Foundation. LSDYNA is a trademark of Livermore Software Technology Corporation. All other trademarks are the property of their respective owners. This software may contain certain thirdparty software that is protected by copyright and licensed from MSC.Software suppliers. METIS is copyrighted by the regents of the University of Minnesota. NTMPICH is developed by Lehrstuhl für Betriebssysteme der RWTH Aachen. Copyright © 19922004 Lehrstuhl für Betriebssysteme der RWTH Aachen. PCGLSS 6.0 is licensed from Computational Applications and System Integration Inc. Use, duplication, or disclosure by the U.S. Government is subject to restrictions as set forth in FAR 12.212 (Commercial Computer Software) and DFARS 227.7202 (Commercial Computer Software and Commercial Computer Software Documentation), as applicable. Revision 0 04/11/2008 MDNA*R3*Z*ENON*Z*DCUSRPDF
Main Index
Contents
MD Nastran Explicit Nonlinear (SOL 700) User’s Guide
Contents
1
Introduction
MD Nastran Explicit Nonlinear (SOL 700) Defining the Model 16 Implicit and Explicit Methods Implicit Methods 17 Explicit Methods 18 Typical Applications 24 17 16
Principles of the Eulerian and Lagrangian Processors Feature List 26 27
24
How SOL 700 Solves Explicit Problems Input and Output 27 This User’s Guide 28 Additional Documentation for SOL 700 28
2
MD Nastran Data Files
The MD Nastran Input Data File 32 Input Conventions 33 Section Descriptions 33 Running Existing Models with SOL 700 34 Supported SOL 700 Entries 35 Supported Case Control Cards 35 Supported Bulk Data Entries 37 Supported Material Models in SOL 700 50 Unsupported Materials in MD Patran Preference Supported Parameters in SOL 700 50 Supported Parameters in MD Patran Preference Preprocessing with MD Patran 51 Generating the Bulk Data File 51 Editing the Bulk Data File 52
50 51
Main Index
4 MD Nastran Explicit Nonlinear (SOL 700) User’s Guide
Output Requests 52 Files Created by SOL 700
52 55
Postprocessing with MD Patran
3
Modeling
Coordinate Systems 58 Userdefined Coordinate System 58 Nodal Coordinate Systems 59 Element Coordinate Systems 59 Material Coordinate Systems 60 Nodes 61 Degrees of Freedom Elements 62 61
Modeling in MD Patran 63 Creating Geometry in MD Patran 63 Creating Finite Element Meshes in MD Patran Example using MD Patran 67 Description of the Problem 67 Solution Type 68 Specifying the Solution Type 68 References 68 Defining the Solution Type in MD Patran
65
69 79
Input and Output Files Created During the Simulation Simulation on Windows Platforms 79 Postprocessing 81 85 85 85
Running a Batch Job
How to Tell When the Analysis is Done
How to Tell if the Analysis Ran Successfully Running a Parallel Job 86
Main Index
Contents 5
4
Special Modeling Techniques
Artificial Viscosity Bulk Viscosity 90 Hourglass Damping 90 91
Mass Scaling 95 Problems Involving a Few Small Elements 95 Problems Involving a Few Severely Distorted Elements
95 98
Time Domain NVH 97 Time Domain NVH Example – Plate Subjected to a Pulse Loading
Prestress (Implicit to Explicit Sequential Simulation) 107 Comparative Study between Nastran and MD Nastran SOL 700 Implicit Element Formulations 107
5
Constraints and Loadings
Constraint Definition 116 SinglePoint Constraints 116 MultiPoint Constraints 116 Specifying Explicit MPCs 117 Contact in SOL 700 117 Contact Bodies 118 Rigid Walls 119 Contact Detection 119 Lagrangian Loading 120 Concentrated Loads and Moments 120 FORCE, FORCE2, or DAREA – FixedDirection Concentrated Loads Pressure Loads 121 Initial Conditions 122 Eulerian Loading and Constraints 123 Loading Definition 123 Flow Boundary 123 Rigid Wall 123 Initial Conditions 124 Detonation 129 Body Forces 129 Hydrostatic Preset 129 Speedup for 2D Axial Symmetric Models 129
120
Main Index
6 MD Nastran Explicit Nonlinear (SOL 700) User’s Guide
Boundary and Loading Conditions  Theoretical Background Pressure Boundary Conditions 130 Kinematic Boundary Conditions 132 Displacement Constraints 132 Prescribed Displacements, Velocities, and Accelerations 133 Body Force Loads 133
130
6
Elements
Elements Overview Element Types 136 136
Preliminaries 140 Governing Equations 141 CHEXA Solid Elements 144 Volume Integration 146 Hourglass Control 147 Fully Integrated Brick Elements and MidStep Strain Evaluation CTETRA  Four Node Tetrahedron Element 152 CPENTA  Six Node Pentahedron Element 153 CBEAM  Belytschko Beam 154 Corotational Technique 154 Belytschko Beam Element Formulation Calculation of Deformations 158 Calculation of Internal Forces 159
151
157
CBEAM  DYSHELFORM = 1, HughesLiu Beam Geometry 163 Fiber Coordinate System 167 Strains and Stress Update 169
163
CQUADA  DYSHELLFORM = 2, BelytschkoLinTsay Shell Corotational Coordinates 172 VelocityStrain Displacement Relations 174 Stress Resultants and Nodal Forces 176 Hourglass Control (BelytschkoLinTsay) 177 Hourglass Control (Englemann and Whirley) 178
172
CQUAD4 DYSHELLFORM = 10, BelytschkoWongChiang Improvements 181
Main Index
Contents 7
CTRIA3  DYSHELLFORM = 4, C0 Triangular Shell Corotational Coordinates 184 VelocityStrain Relations 185 Stress Resultants and Nodal Forces 188
184
CTRIA3  DYSHELLFORM = 3, MarchertasBelytschko Triangular Shell (BCIZ) 190 Element Coordinates 190 Displacement Interpolation 192 StrainDisplacement Relations 194 Nodal Force Calculations 195 CQUAD4  DYSHELL FORM = 1, HughesLiu Shell Geometry 197 Kinematics 200 Fiber Coordinate System 201 Lamina Coordinate System 202 Strains and Stress Update 204 197
CQUAD4, DYSHELLFORM = 6, 7, Fully Integrated HughesLiu Shells 207 Element Mass Matrix 210 Accounting for Thickness Changes 211 Transverse Shear Treatment for Layered Shell CROD, Truss Element 216 217 211
CQUAD4  DYSHELLFORM = 9, Membrane Element Corotational Coordinates 217 VelocityStrain Displacement Relations 217 Stress Resultants and Nodal Forces 218 Membrane Hourglass Control 218 CELAS1D, Discrete Elements and CONM2, Masses Orientation Vectors 220 Dynamic Magnification “Strain Rate” Effects 222 Deflection Limits in Tension and Compression 223 CDAMP2D, Linear Elastic or Linear Viscous Eulerian Elements 225 Element Definition 225 Solid Elements 225 224
219
Main Index
8 MD Nastran Explicit Nonlinear (SOL 700) User’s Guide
Graded Meshes 226 Requirements for Gluing Meshes 226 Gluing Meshes 227 Using Graded Meshes 228 Visualization with MD Patran 228
7
Materials
Lagrangian Material Models for SOL 700 232 Material Model 1: Elastic 238 Material Model 2: Orthotropic Elastic 239 Material Model 3: Elastic Plastic with Kinematic Hardening 240 Material Model 5: Soil and Crushable Foam 244 Material Model 6: Viscoelastic 245 Material Model 7: Continuum Rubber 246 Material Model 9: Null Material 246 Material Model 10: ElasticPlasticHydrodynamic 246 Material Model 12: Isotropic ElasticPlastic 249 Material Model 13: Isotropic ElasticPlastic with Failure 250 Material Model 14: Soil and Crushable Foam With Failure 250 Material Model 15: Johnson and Cook Plasticity Model 251 Material Type 18: Power Law Isotropic Plasticity 252 Material Type 19: Elastic Plastic Material Model with Strain Rate Dependent Yield 253 Material Type 20: Rigid 254 Material Model 22: ChangChang Composite Failure Model 255 Material Model 24: Piecewise Linear Isotropic Plasticity 256 Material Model 26: Crushable Foam 257 Material Model 27: Incompressible MooneyRivlin Rubber 261 Material Model 28: Resultant Plasticity 262 Material Model 29: FORCE LIMITED Resultant Formulation 263 Material Model 30: Shape Memory Alloy 270 Material Model 31: Slightly Compressible Rubber Model 274 Material Model 32: Laminated Glass Model 275 Material Model 34: Fabric 275 Material Model 36: Barlat’s 3Parameter Plasticity Model 276 Material Type 37: Transversely Anisotropic ElasticPlastic 278 Material Type 38: BlatzKo Compressible Foam 280 Material Model 39: Transversely Anisotropic ElasticPlastic With FLD 280 Material Model 53: Low Density Closed Cell Polyurethane Foam 281 Material Models 54 and 55: Enhanced Composite Damage Model 282 Material Model 57: Low Density Urethane Foam 284 Material Type 58: Laminated Composite Fabric 286
Main Index
Contents 9
Material Type 62: Viscous Foam 288 Material Type 63: Crushable Foam 289 Material Model 64: Strain Rate Sensitive PowerLaw Plasticity 290 Material Model 65: Modified Zerilli/Armstrong 290 Material Model 67: Nonlinear Stiffness/Viscous 3D Discrete Beam 291 Material Model 68: Nonlinear Plastic/Linear Viscous 3D Discrete Beam 292 Material Model 69: Side Impact Dummy Damper (SID Damper) 293 Material Model 70: Hydraulic/Gas Damper 295 Material Model 71: Cable 295 Material Model 73: Low Density Viscoelastic Foam 296 Material Model 74: Elastic Spring for the Discrete Beam 296 Material Model 76: General Viscoelastic 297 Material Model 77: Hyperviscoelastic Rubber 298 Material Model 78: Soil/Concrete 300 Material Model 79: Hysteretic Soil 302 Material Model 80: RambergOsgood Plasticity 302 Material Model 81 and 82: Plasticity with Damage and Orthotropic Option 303 Material Model 83: FuChang’s Foam With Rate Effects 306 Material Model 87: Cellular Rubber 308 Material Model 89: Plasticity Polymer 310 Material Model 94: Inelastic Spring Discrete Beam 310 Material Model 97: General Joint Discrete Beam 310 Material Model 98: Simplified Johnson Cook 311 Material Model 100: Spot Weld 311 Material Model 116: Composite Layup 313 Material Model 119: General Nonlinear Six Degrees of Freedom Discrete Beam 314 Material Model 124: TensionCompression Plasticity 315 Material Model 126: Modified Honeycomb 315 Material Model 127: ArrudaBoyce Hyperviscoelastic Rubber 318 Material Model 158: Rate Sensitive Composite Fabric 319 Material Model 181: Simplified Rubber Foam 319 Material Model 196: General Spring Discrete Beam 322 Grunseisen Equation of State (EOSGRUN) 323 Tabulated Compaction Equation of State (EOSTABC) 323 Tabulated Equation of State (EOSTAB) 324 Eulerian Material Models for SOL 700 Shear Models 325 Yield Models 326 Equations of State 336 Material Failure 339 Spallation Models 340 Material Viscosity 341 325
Main Index
10 MD Nastran Explicit Nonlinear (SOL 700) User’s Guide
8
Contact Impact Algorithm
Overview 344 SOL 600 Contact Capabilities not yet supported by SOL 700 344 Notes on Rigid Body Modeling in SOL 700: 345 SOL 600 Contact Limitations that do not apply to SOL 700 345 Penalty Methods Preliminaries Slave Search 346 347 351 352 355 356 346
Contact Force Calculation
Improvements to the Contact Searching Bucket Sorting 353 Bucket Sorting in Single Surface Contact Accounting For the Shell Thickness Initial Contact Penetrations Contact Energy Calculation Friction 359 357 358
9
Fluid Structure Interaction
General Coupling 362 Fluidstructure Interaction 362 Closed Volume 363 Porosity 364 Multiple Coupling Surfaces with Multiple Euler Domains Coupling Surface with Failure 368 Coupling Surfaces with Porous Holes 368 Flow Between Domains 369 Deactivation 369 Output 370 Using the Preference of Patran 370 Fluid and Gas Solver for the Euler Equations Modeling Fluid Filled Containers Hotfilling 373 372 371 367
Main Index
Contents 11
10 Eulerian Solvers
The Standard Euler Solver 376 Fluidstructure Interaction 377 The Numerical Scheme 378 The Time Step Criterion 380 Euler with Strength 381 The Multimaterial Solver 382 Approximate Riemann Euler Solver 386 Euler Equations of Motion 386 Numerical Approach 387 Entropy Fix for the Flux Difference Riemann Scheme Second Order Accuracy of the Scheme 390 Time Integration 391
390
11
Airbags and Occupant Safety
Introduction 394 Airbag Definition 395 Inflator Models in Airbags 397 Constant Volume Tank Tests 400 Porosity in Airbags 400 Initial Metric Method for Airbags 403 Heat Transfer in Airbags 404 Seatbelts 405 Seatbelt Pretensioner 406 Seatbelt Retractor 408 Seatbelt Sensor 411 Seatbelt Slipring 411 Occupant Dummy Models Pre and Postprocessing 413 414
Main Index
12 MD Nastran Explicit Nonlinear (SOL 700) User’s Guide
12 System Information and Parallel Processing
Introduction 416 General Information 416 Release Platforms 416 MPI for MD Nastran SOL 700 417 Hardware and Software Requirements Compatibility 418 Definition 418
417
User Notes 419 Specification of the Host file for Windows 419 Specification of the Host file for UNIX and LINUX 419 How to Run MD Nastran SOL 700 in Parallel 420 How to Run Fluid Structure Interaction (FSI) in Parallel 421 Additional information of Different Platforms Windows XP and 2000 425 Extra MPIch information 425 421
13 Examples
Crash 428 Pickup Truck Frontal Crash Trainbarrier Impact 437 428
Airbags and Occupant Safety 444 Simulation of Multicompartment Airbag Airbag with Dummy 450
444
Bird Strike and Fan Blade Out (FBO) 459 Bird Strike Simulation on Composite Glass Panel 459 Bird Strike on Rotating Fan Blades with Prestress 466 Multiple Birdstrikes on a Box Structure 479 Chained Analysis  Fan Blade Out to Rotor Dynamics 490 Drop Test 510 Drop Test Simulation of a Computer Package Wheel Drop Test 518 510
Main Index
Contents 13
Defense 524 Rod Penetration 524 Shaped Charge, using IG Model, Penetrating through Two Thick Plates Mine Blast 538 Blastwave Hitting a Bunker 548 Time Domain NVH Chassis 555 555
530
Prestress 564 Simulation of Prestress and Impact on Rotating Fan Blades Smooth Particle Hydrodynamics (SPH) Ball Penetration using SPH Method 576 576
564
Sheet Metal Forming 582 Square Cup Deep Drawing using Forming Limit Diagram 582 PART 2. Square Cup Deep Drawing using Implicit Spring Back Miscellaneous 596 Paper Feed 596
591
Main Index
14 MD Nastran Explicit Nonlinear (SOL 700) User’s Guide
Main Index
Chapter 1 Introduction MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
1
Introduction
MD Nastran Explicit Nonlinear (SOL 700) Implicit and Explicit Methods 17 24 16
Principles of the Eulerian and Lagrangian Processors Feature List 26 27
How SOL 700 Solves Explicit Problems This User’s Guide 28
Main Index
16 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
MD Nastran Explicit Nonlinear (SOL 700)
MD Nastran Explicit Nonlinear (SOL 700)
MD Nastran Explicit Nonlinear (SOL 700) offers a powerful explicit solution to analyze dynamic events of short duration with severe geometric and material nonlinearities. MD Nastran SOL 700 allows users to work within one common modeling environment using the same Bulk Data interface. The NVH, linear and nonlinear models can be used for explicit applications such as crash, crush, and drop test simulations. This dramatically reduces the time spent to build different models for implicit and explicit analysis and prevents you from making mistakes because of unfamiliarity between different programs. MD Nastran SOL 700 is being developed in phases. The first phase, available in MD Nastran 2005 r3 supports the explicit structural applications.
Defining the Model
A finite element model consists of a geometric description given by the elements, their nodes, and a set of properties associated with the elements describing their attributes. These properties include material definitions, crosssection definitions in the case of structural elements like beams and shells, and other parameters for contact bodies, springs, dashpots, etc. There may also be constraints that must be included in the model  RBE elements, multipoint constraints, or equations (linear or nonlinear equations involving several of the fundamental solution variables in the model), or simple boundary conditions that are to be imposed throughout the analysis. Nonzero initial conditions, such as initial displacements and velocities, may also be specified. The model is described and communicated to MD Nastran in the form of a text file (MD Nastran Input file). You can generate this file using a variety of preprocessor programs such as MD Patran or any text editor. It must adhere to MD Nastran conventions for the ordering and format of the model information. User Interface The user interface is the familiar MD Nastran card image interface comprised of executive control, case control, and bulk data cards. In most cases, they are identical to input formats defined for the MD Nastran Implicit Nonlinear module SOL 600 (and other solution sequences). The main differences involve the use of the SOL 700 materials and rigid joints (cylindrical, spherical, and revolute.) for various MD Nastran solution sequences. These materials and rigid joints may only be accessed using the SOL 700,129 executive control card. Additionally, in some specialized cases, some familiar MD Nastran cards may need small modifications and others might need to be added to specify items for defining new concepts such as moving coordinates. Using MD Patran with SOL 700 The amount of information that needs to be conveyed in the MD Nastran Input file is extensive for even a modest size model. The amount of information and the complexity of most models makes it virtually impossible to generate the MD Nastran Input file with a text editor alone. Typically, you benefit from
Main Index
Chapter 1 Introduction 17
Implicit and Explicit Methods
using a preprocessor such as MD Patran. MD Patran provides a graphical user interface, an extensive line of model building tools that you can use to construct and view your model, and generate a MD Nastran Input file. If you are using MD Patran as a preprocessor, you are required to specify an analysis code. Selecting MD Nastran Explicit Nonlinear (SOL 700) as the analysis code under the Analysis Preference menu, customizes MD Patran in five main areas:
• Material Library • Element Library • Loads and Boundary Conditions • MPCs • Analysis forms
The analysis preference also specifies that the model information be output in the MD Nastran Input File format.
Chapter 3: Modeling includes examples of forms and templates of the MD Patran Preference for
SOL 700.
Implicit and Explicit Methods
A detailed theory of explicit analysis is outside the scope of this section. However, it is important to understand the basics of the solution technique, since it is critical to many aspects of using MD Nastran, SOL 700. If you are already familiar with explicit methods and how they differ from implicit methods, then you may disregard this section.
Implicit Methods
Many finite element programs use implicit methods to carry out a transient solution. Normally, they use Newmark schemes to integrate in time. If the current time step is step n , a good estimate of the acceleration at the end of step n + 1 will satisfy the following equation of motion:
M a' n + 1 + Cv' n + 1 + Kd' n + 1 = F n + 1
ext
where
M C K Fn + 1 a' n + 1
ext
= mass matrix of the structure = damping matrix of the structure = stiffness matrix of the structure = vector of externally applied loads at step = estimate of acceleration at step
n+1 n+1
Main Index
18 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Implicit and Explicit Methods
v' n + 1 d' n + 1
= estimate of velocity at step
n+1 n+1
= estimate of displacement at step
and the prime denotes an estimated value. The estimates of displacement and velocity are given by:
d' n + 1 = d n + v n Δ t + ( ( 1 – 2β )a n Δ t ) ⁄ 2 + β a' n + 1 Δ t v' n + 1 = v n + ( 1 – γ )a n Δ t + γa' n + 1 Δ t
2 2
or
d' n + 1 = d * + βa' n + 1 Δ t n
* v' n + 1 = v n + γ a' n + 1 Δ t 2
where
Δt
is the time step and
* dn
β
and
γ
are constants.
The terms
and
* vn
are predictive and are based on values already calculated.
Substituting these values in the equation of motion results in
M a' n + 1 + C ( v * n + γa' n + 1 Δ t ) + K ( d * n + β a' n + 1 Δ t ) = F n + 1
2 ext
or
* * [ M + CγΔ t + K βΔ t ]a' n + 1 = F n + 1 – Cv n – K d n 2 ext
The equation of motion may then be defined as
M *a' n + 1 = F n + 1
residua l
The accelerations are obtained by inverting the
a' n + 1 = M * F n + 1
–1 resi dual
M*
matrix as follows:
This is analogous to decomposing the stiffness matrix in a linear static analysis. However, the dynamics mean that mass and damping terms are also present.
Explicit Methods
The equation of motion
M a n + Cv n + K d n = F n
ext
Main Index
Chapter 1 Introduction 19
Implicit and Explicit Methods
can be rewritten as
M an = Fn
–1 ext
– Fn
int
an = M Fn
resi dua l
where
Fn Fn
ext
= vector of externally applied loads = vector of internal loads (e.g., forces generated by the elements and hourglass forces) =
C v n + Kd n
int
F int M
= mass matrix
The acceleration can be found by inverting the mass matrix and multiplying it by the residual load vector. In SOL 700, like any explicit finite element code, the mass matrix is lumped which results in a diagonal mass matrix. Since M is diagonal, its inversion is trivial, and the matrix equation is a set of independent equations for each degree of freedom, as follows:
a ni = F ni
resi dual
⁄ Mi
The Leapfrog scheme is used to advance in time. The position, Forces, and accelerations are defined at time level n , while the velocities are defined at time level n + 1 ⁄ 2 . Graphically, this can be depicted as:
n–1 d, F , a
n–1⁄2
v
n d, F, a
n+1⁄2
v
n+1 d, F, a
time
vn + 1 ⁄ 2 = v n – 1 ⁄ 2 + a n ( Δ tn + 1 ⁄ 2 + Δ t n – 1 ⁄ 2 ) ⁄ 2 dn + 1 = dn + v n + 1 ⁄ 2 Δ tn + 1 ⁄ 2
The Leapfrog scheme results in a central difference approximation for the acceleration, and is secondorder accurate in Δ t . Explicit methods with a lumped mass matrix do not require matrix decompositions or matrix solutions. Instead, the loop is carried out for each time step as shown in the following diagram:
Main Index
20 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Implicit and Explicit Methods
GridPoint Accelerations Leapfrog Integration in Time GridPoint Velocities GridPoint Displacements
Element Formulation and Gradient Operator Element Stain Rates Constitutive Model and Integration Element Stresses Element Formulation and Divergence Operator Element Forces at GridPoints CONTACT, FluidStructure Interaction, Force/Pressure boundaries + External Forces at Grid Points
Explicit Time Step Implicit methods can be made unconditionally stable regardless of the size of the time step. However, for explicit codes to remain stable, the time step must subdivide the shortest natural period in the mesh. This means that the time step must be less than the time taken for a stress wave to cross the smallest element in the mesh. Typically, explicit time steps are 100 to 1000 times smaller than those used with implicit codes. However, since each iteration does not involve the costly formulation and decomposition of matrices, explicit techniques are very competitive with implicit methods. Because the smallest element in an explicit solution determines the time step, it is extremely important to avoid very small elements in the mesh. Courant Criterion Since it is impossible to do a complete eigenvalue analysis every cycle to calculate the timestep, an approximate method, known as the Courant Criterion, is used. This is based on the minimum time which is required for a stress wave to cross each element:
Δt = SL/ c
where
Δt S
= Timestep = Timestep scale factor (<1)
Main Index
Chapter 1 Introduction 21
Implicit and Explicit Methods
L c
= Smallest element dimension = Speed of sound in the element material
For 1D elements, the speed of sound is defined as:
c = E⁄ρ
where,
E ρ
= Young’s modulus = density
When to use Explicit Analysis The time step for implicit solutions can be much larger than is possible for explicit solutions. This makes implicit methods more attractive for transient events that occur over a long time period and are dominated by low frequency structural dynamics. Explicit solutions are better for short, transient events where the effects of stress waves are important. There is, of course, an area where either method is equally advantageous and may be used. Explicit solutions have a greater advantage over implicit solutions if the time step of the implicit solution has to be small for some reason. This may be necessary for problems that include:
• Material nonlinearity. A high degree of material nonlinearity may require a small time step
for accuracy.
• Large geometric nonlinearity. Contact and friction algorithms can introduce potential
instabilities, and a small time step may be needed for accuracy and stability.
• Those analyses where the physics of the problem demands a small time step (e.g. stress wave
effects as in crash, crush, and impact analyses).
• Material and geometric nonlinearity in combination with large displacements. Convergence
in implicit methods becomes more difficult to achieve as the amount of nonlinearity for all types increases. Explicit methods have increasing advantages over implicit methods as the model gets bigger. For models containing several hundred thousands of elements and including significant nonlinearity, SOL 700 may provide the most costeffective solution even for problems dominated by lowfrequency structural dynamics (see Figure 11).
Main Index
22 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Implicit and Explicit Methods
Cost (CPU Time)
Implicit Explicit
Problem Size
Cost (# of Matrix)
Implicit
Explicit Number/Extent of Nonlinearities
Figure 11
Efficiency and Cost
Another case where explicit solutions may have advantages is in extremely large linear or nonlinear problems where implicit solvers may be limited to the number or parallel domains. There is no limit to the number of SOL 700 explicit parallel domains. Therefore, if you have more than 256 or so processors available, the explicit analysis may be faster even for linear static analyses. SOL 700 should normally be used for structural components that may undergo large, sudden deformation, and for which the dimensions, deformed geometry, and residual stress state are of major importance. Table 11 summarizes the areas of overlap as well as the differences between the implicit and explicit analyses. Table 11 Implicit and Explicit Technology Comparison Implicit Explicit
Material Nonlinearity Linear isotropic elastic (metals) Nonlinear isotropic elastic (rubber materials) Linear orthotropic elastic (composites) Elasticperfectly plastic (limit analysis) Elastoplastic, strain hardening (metals) Viscoelastic (polymers)
Main Index
Chapter 1 Introduction 23
Implicit and Explicit Methods
Table 11
Implicit and Explicit Technology Comparison (continued)
Restricted orthotropic (metalforming) Damage accumulation and failure Tearing and failure Explosive detonation Deformation Nonlinearity Infinitesimal strains and rotations Infinitesimal strains and finite rotations Finite strains and rotations Large strains (100% plus) and large rotations (Multi) Material flow Contact Nonlinearity Small displacement gaps Gaps with friction Large displacement gaps Contact surfaces Single surface contact Fluidstructure interaction Motion Static (infinite) Quasistatic (noninertial) Vibration, fundamental modes Shock and vibration Stress wave propagation Shock wave propagation High Frequency Dynamics Detonation waves Implicit Explicit Implicit Explicit Implicit Explicit
Main Index
24 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Principles of the Eulerian and Lagrangian Processors
Typical Applications
Some of the typical structural applications which are well suited for the SOL 700 explicit analysis are:
• Automotive and Aircraft crash worthiness • • • • • • • • •
Crash/Crush simulations Drop testing Ship Collision Projectile penetration Bird Strike Simulation with structural Bird Metal forming, stamping, and deep drawing Jet engine blade containment Golf Club simulation Rollover events
Principles of the Eulerian and Lagrangian Processors
MD Nastran Nonlinear Explicit (SOL 700) features two solving techniques, Lagrangian and Eulerian. The code can use either one or both and can couple the two types to produce interaction. The Lagrangian method is the most common finite element solution technique for engineering applications. When the Lagrangian solver is used, grid points are defined that are fixed to locations on the body being analyzed. Elements of material are created by connecting the grid points together and the collection of elements produces a mesh. As the body deforms, the grid points move with the material and the elements distort. The Lagrangian solver is, therefore, calculating the motion of elements of constant mass.
Main Index
Chapter 1 Introduction 25
Principles of the Eulerian and Lagrangian Processors
The Eulerian solver is most frequently used for analyses of fluids or materials that undergo very large deformations. In the Eulerian solver, the grid points are fixed in space and the elements are simply partitions of the space defined by connected grid points. The Eulerian mesh is a “fixed frame of reference.” The material of a body under analysis moves through the Eulerian mesh; the mass, momentum, and energy of the material are transported from element to element. The Eulerian solver therefore calculates the motion of material through elements of constant volume. It is important to note that the Eulerian mesh is defined in exactly the same manner as a Lagrangian mesh. General connectivity is used so the Eulerian mesh can be of an arbitrary shape and have an arbitrary numbering system. This offers considerably more flexibility than the logical rectangular meshes used in other Eulerian codes.
However, you should remember that the use of an Eulerian mesh is different from that of the Lagrangian type. The most important aspect of modeling with the Eulerian technique is that the mesh must be large enough to contain the material after deformation. A basic Eulerian mesh acts like a container and, unless specifically defined, the material cannot leave the mesh. Stress wave reflections and pressure buildup can develop from an Eulerian mesh that is too small for the analysis. Eulerian and Lagrangian meshes can be used in the same calculation and can be coupled using a coupling surface. The surface acts as a boundary to the flow of material in the Eulerian mesh, while the stresses in the Eulerian material exerts forces on the surface causing the Lagrangian mesh to distort.
Main Index
26 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Feature List
Feature List
The complete features of MD Nastran Explicit Nonlinear (SOL 700) Phase I are: 1. MD Nastran Explicit Nonlinear (SOL 700) solves nonlinear (material, contact and/or geometric) static and transient dynamic structural finite element problems. 2. MD Nastran Explicit Nonlinear (SOL 700) supports the following elements/bodies:
• • • • • • • • • • • •
3noded triangular shell/membrane/plane stress/(generalized) plain strain elements. 4noded quadrilateral shell/membrane/plane stress/(generalized) plain strain elements. 4noded solid tetrahedral elements. 6noded solid wedge elements. 8noded solid hexahedral elements. 2noded beam element. 2noded bar element. spring elements. damper elements. Rigid and deformable contact bodies. Point Mass element. RBE elements and multipoint constraint equations are supported to tie specific nodes or degreesoffreedom to each other. Special MPC entities are supported, (e.g. rigid links) which can be used to tie two nodes together or equate the motion of two degrees of freedoms. Constrained nodal displacements (zero displacements at specified degrees of freedom). Enforced nodal velocities (nonzero velocities at specified degrees of freedom) Forces applied to nodes Pressures applied to element faces Inertial body forces. Linear acceleration and rotational velocity can be applied Contact between two bodies can be defined by selecting the contacting bodies and defining the contact interaction properties.
3. MD Nastran Explicit Nonlinear (SOL 700) supports the following loads and boundary conditions:
• • • • • •
4. MD Nastran Explicit Nonlinear (SOL 700) supports isotropic, orthotropic, and anisotropic material properties. Nonlinear elasticplastic materials can be defined by specifying piecewise linear stressstrain curves, material failure can be specified. 5. Physical properties can be associated with MD Nastran Explicit Nonlinear (SOL 700) elements such as the crosssectional properties of the beam element, the area of the beam and rod elements, the thickness of shell, plane stress, and membrane elements, spring parameters, and masses. 6. Laminated composite shell elements are supported in MD Nastran Explicit Nonlinear (SOL 700) through the PCOMP card of the materials capability. Each layer has its own material, thickness, and orientation and may represent linear or nonlinear material behavior. Failure index calculations are also supported. 7. Analysis jobs consisting of complex loading time histories are available. All loading must be applied in a single subcase. The applicable subcase for a particular analysis may be chosen from many subcases if so desired.
Main Index
Chapter 1 Introduction 27
How SOL 700 Solves Explicit Problems
8. MD Nastran Explicit Nonlinear (SOL 700) jobs are submitted using textbased input decks that may be generated manually with a text editor, or by a variety of pre/post processing programs such as MD Patran. The input file is read in and a number of output files, such as the .f06, .log, f04, .dytr.dat, .dytr.d3hsp, .dytr.d3plot files are generated. 9. Results can be requested in several output formats such as .dytr.d3plot, .dytr.d3thdt, .f06 files. These files are typically read back into the pre/postprocessing programs for the purpose of evaluating the results with plots such as deformed shape plots, contour stress/strain plots, or XY history plots. MD Nastran native output files such as .xdb, .op2, and punch will be supported in the MD Nastran 2006r2 release. 10. Nodal displacements, velocities and accelerations, element stresses, element strains, element plastic strains, nodal reaction forces, contact interface force values, and element strain energy are output when applicable. These may be visualized with results visualization tools such as MD Patran. Composite element results are returned for individual layers of the composite, as requested. 11. Crash, impact is supported by the Phase I release. Dynamic loading described by time histories as well as static loading is also supported.
How SOL 700 Solves Explicit Problems
SOL 700,ID solution resembles SOL 600,ID and uses the standard MD Nastran DMAP for solution ID. It is written so that there is not an effect on other solution sequences. The primary features are as follows: 1. Read MD Nastran Input File in IFP (input file processor) as in other MD Nastran solution sequences and translate executive control, case control, and bulk data to SOL 700 input formats. 2. Read SOL 700 input formats into explicit solver. 3. Run the model. 4. Write MD Nastran output result files.
Input and Output
Once the solver is finished, the native output files are written out. You can request the type of outputs desired on the OUTR portion of the SOL 700 executive control card and by various parameters. If the OUTR options are omitted, standard outputs are available unless a request is made to delete the files. In version 2006r2, the MD Nastran native output files .xdb, op2 and punch are not supported and will be implemented in future releases. Instead d3plot, d3thdt and d3hsp are generated. They consist of displacements, velocities, accelerations, stress tensors, strain tensors and plastic strain. The geometry blocks are generated as usual using PARAM,POST,1 or PARAM,POST,0.
Main Index
28 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
This User’s Guide
The example below shows the files that are generated in the working directory:
Input file:
filename.bdf filename.dytr.dat filename.dytr.str
Input deck for MD Nastran Intermediate internal files generated for checking input errors
Output file:
filename.dytr.d3hsp filename.dytr.out filename.dytr_prep.d3hsp filename.dytr_prep.out filename.f06
Contains model summary, calculation process, CPU time, etc. Contains time step summary and CPU timing information. Contains translation summary of the preparation phase. Nastran native output that contains summary of model. Contains complete model, used for plotting deformed shape and stress contour. Contains subset of the model, used for time history plots. Contains additional timehistory data.
Binary Result file:
filename.dytr.d3plot filename.dytr.d3thdt binout000
This User’s Guide
This manual provides a complete background to SOL 700 and fully describes using SOL 700 within the MD Nastran environment. The theoretical aspects of explicit analysis methods, types, and techniques are included as well as descriptions for explicit material models. Where appropriate, actual MD Patran forms and menus are shown so you can easily use SOL 700 from within the MD Patran environment.
Additional Documentation for SOL 700
MD Nastran Reference Manual  provides supporting information that relates to MD Nastran input formats, element libraries, and loads and boundary conditions. MD Nastran Quick Reference Guide (QRG)  contains a complete description of all the input entries for MD Nastran. Within each section, entries are organized alphabetically so they are easy to find. Each entry provides a description, formats, examples, details on options, and general remarks. Within the QRG, you will find the full descriptions for all of the SOL 700 input entries.
Main Index
Chapter 1 Introduction 29
This User’s Guide
The MD Patran library includes three key books which will be of assistance in running SOL 700:
• MD Patran User’s Guide  gives the essential information needed to immediately begin using
MD Patran for SOL 700 projects. Understanding and using the information in this guide requires no prior experience with CAE or finite element analysis.
• MD Patran Reference Manual  gives complete descriptions of basic functions in MD Patran,
geometry modeling, finite element modeling, material models, element properties, loads and boundary conditions, analysis, and results. It is the counterpart to the MD Nastran Reference Manual.
• MD Nastran Preference Guide  gives specific information that relates to using MD Patran with
MD Nastran as the intended analysis code. All application forms and required input are tailored to MD Nastran.
Main Index
30 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
This User’s Guide
Main Index
Chapter 2: MD Nastran Data Files MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
2
MD Nastran Data Files
The MD Nastran Input Data File Supported SOL 700 Entries Preprocessing with MD Patran Output Requests 52 35
32
51
Postprocessing with MD Patran
55
Main Index
32 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
The MD Nastran Input Data File
The MD Nastran Input Data File
The MD Nastran Input File, usually given the extension .dat or .bdf, is made up of three distinct sections: 1. Executive Control  describes the problem type and size. 2. Case Control  defines the load history. 3. Bulk Data  gives a detailed model description. Input data is organized in (optional) blocks. Key words identify the data for each optional block. This form of input enables you to specify only the data for the optional blocks that you need to define your problem. The various blocks of input are “optional” in the sense that many have builtin default values which are used in the absence of any explicit input from you. A typical input file setup for the MD Nastran program is shown below.
• Executive Control Statements
Terminated by a CEND parameter
• Case Control Commands
Terminated by the BEGIN BULK option
• Bulk Data Entries
Model data starting with the BEGIN BULK option and terminated by the ENDDATA option
Model Data  grids, elements, etc.
MD Nastran Explicit Nonlinear Complete Input Deck
Bulk Data
Element and Material Properties Fixed Displ, Etc.
Control Information
Case Control
Load Incrementation, Applied Loads, Applied Displacements Etc.
Executive Control
Title, Job Control, Solution Sequence, Etc.
Main Index
Chapter 2: MD Nastran Data Files 33
The MD Nastran Input Data File
Input Conventions
MD Nastran Explicit Nonlinear performs all data conversion internally so that the system does not abort because of data errors made by you. The program reads all input data options alphanumerically and converts them to integer, floating point, or keywords, as necessary. MD Nastran Explicit Nonlinear issues error messages and displays the illegal option image if it cannot interpret the option data field according to the specifications given in the manual. When such errors occur, the program attempts to scan the remainder of the data file and ends the run with a FATAL ERROR message. For SOL 700, certain entries generate Warning or Severe Warning messages and, in some cases, the run is terminated immediately after the message is written. Two input format conventions can be used: fixed and free format. You can mix fixed and free format options within a file. The syntax rules for fixed fields are as follows:
• Give floating point numbers with or without an exponent. If you give an exponent, it must be preceded by the character E or D and must be rightjustified (no embedded blanks). If data is double precision, a D must be used.
The syntax rules for free fields are as follows:
• Check that each option contains the same number of data items that it would contain under
standard fixedformat control. This syntax rule allows you to mix fixedfield and freefield options in the data file because the number of options you need to input any data list are the same in both cases.
• Separate data items on a option with a comma (,). The comma can be surrounded by any number
of blanks. Within the data item itself, no embedded blanks can appear.
• Give keywords exactly as they are written in the manual. • Enter data as uppercase or lowercase text. • Limit to 8 columns per field for small field format whether using fixedfield or freefield. Large
field is 16 columns; see the MD Nastran Quick Reference Guide for more details.
Section Descriptions
Executive Control This group of entries provides overall job control for the problem and sets up initial switches to control the flow of the program through the desired analysis. This set of input must be terminated with a CEND parameter. See “Executive Control Statements” in Chapter 3 of the MD Nastran Quick Reference Guide for additional descriptions on input formats.
Main Index
34 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
The MD Nastran Input Data File
Case Control This group of options provides the loads and constraints. Case Control options also include blocks which allow changes in the initial model specifications. Case Control options can also specify printout and postprocessing options. See “Case Control Commands” in Chapter 4 of the MD Nastran Quick Reference Guide for additional descriptions on input formats. Bulk Data Entries This set of data options enters the loading, geometry, and material data of the model and provides nodal point data, such as boundary conditions. This group of options must be terminated with the ENDDATA option. See “Bulk Data Entries” in Chapter 8 of the MD Nastran Quick Reference Guide for additional descriptions on input formats. Multiple BEGIN entries and superelements are not allowed.
Running Existing Models with SOL 700
MD Nastran SOL 700 supports the following solution sequence:
• SOL 129  Nonlinear Transient Dynamic Analysis
Some users may have existing models that have been developed and analyzed using MD Nastran Solution Sequences mentioned above. These models may be run through MD Nastran Explicit Nonlinear (SOL 700) by changing the SOLUTION procedure input to MD Nastran Explicit Nonlinear (SOL 700) input. The following is an example of the change required to run existing models through SOL 700. The first line shows an existing MD Nastran SOL 129 Executive Control Statement and the second shows its revision for MD Nastran Explicit Nonlinear (SOL 700): SOL 129 SOL 700,129 By changing the Solution Sequence, there is no guarantee that the job will run. A successful run depends on the nature of the problem and the extent that the input entries are supported for a SOL 700 analysis. SOL 700 supports only the input entries that are relevant for an explicit simulation (see Section 2 for supported entries for details). All other entries that are unique to a certain Solution Sequence, are ignored and warning messages will be printed out in the output files as the user information. You are advised to check your models for accuracy and completeness before starting the simulation. In many occasions, it is wise to check the model in the preprocessor for “poor modeling” to flag out the skewed or small elements. The CPU time in the explicit analysis is directly controlled by the smallest element dimensions in the model. Small or poor elements may not abort a dynamic analysis but they certainly cause an explicit simulation to run infinitely longer. It is recommended that proper material models with welldefined properties are prescribed in the model to accurately predict the material behavior as the model undergoes severe deformations and failure. For example, defining a suitable material failure criteria for the model, ensures that the distorted elements actually fail and are taken out of the calculation once they reach their failure limits. Otherwise, the smallest length in the distorted elements causes the time step to fall infinitely low, resulting in the premature termination of the job.
Main Index
Chapter 2: MD Nastran Data Files 35
Supported SOL 700 Entries
Supported SOL 700 Entries
Supported Case Control Cards
1. The following summarizes the Case Control Commands for SOL 700: (Only those which are supported or produce fatal errors are listed) Item $ ACCELERATION BCONTACT BEGIN BULK DEFAULT DISPLACEMENT DLOAD ECHO ELFORCE (see FORCE) ENDTIME FORCE & ELFORCE GROUNDCHECK IC INCLUDE LABEL LINE LOAD LOADSET MAXLINES MPC NLPARM Y Y Y Y (other BEGIN forms are not allowed) Y Y Y Y Y Y (new) Y (automatically produced in d3plot files; no user control) Y (Nastran f06 only) Y Y Y (Nastran f06 only) Y (Nastran f06 only) Y (for dynamic pseudostatics only) Y Y (Nastran f06 only) Y Y (pseudo static analysis only) Y Y Y Y Y Y Y Available in SOL 700 Patran Support Y Y Y Y
Main Index
36 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Supported SOL 700 Entries
Item NLSTRESS PAGE PARAM PRESSURE SET SET – OUTPUT(PLOT) SKIP SPC STRAIN STRESS SUBCASE Y N
Available in SOL 700 Y (changed to STRESS) Y (Nastran only) Y (only applicable parms are used)
Patran Support
Y
Y
Y (required if multiple subcases are present) Y Y Y Y Note: Only one subcase can be selected for a particular SOL 700 analysis. Many subcases may be entered in the input deck, but the one to be used must be selected using the SKIP ON and SKIP OFF Case Control commands. If the SKIP ON/OFF commands are not found or are in the wrong place, the first subcase encountered is used and the others are ignored. Y Y Y Y Y
SUBTITLE SURFACE TITLE TSTEP TSTEPNL VELOCITY WEIGHTCHECK
Y Y Y Y (same as TSTEPNL) Y Y Y (Nastran only)
Y Y Y Y
Main Index
Chapter 2: MD Nastran Data Files 37
Supported SOL 700 Entries
Supported Bulk Data Entries
Item ACCMETR ABINFL AIRBAG AXIC AXIF AXSLOT BARRIER BAROR BCBODY BCBOX BCGRID BCHANGE BCMATL BCONP BCPROP BCSEG BCTABLE BEAMOR BJOIN BLSEG BSURF CBAR CBEAM CBELT Y Y (new) Y N N N Y (new) Y Y Y Y N Y N Y Y Y (revised) Y Y N Y Y Y Y Y Y Y Y Y Y Y Y N Y N Y Y Y Y N Y Y Y Y Available in Sol 700 Fatal Error Patran Support
Main Index
38 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Supported SOL 700 Entries
Item CBEND CBUSH CBUTT CCONEAX CCRSFIL CDAMP1 CDAMP2 CDAMP1D CDAMP2D CELAS1 CELAS2 CELAS1D CELAS2D CFILLET CFLUID CGAP CHACAB CHACBR CHEXA CMARKB2 CMARKN1 COMBWLD CONM2 CONROD CONSPOT N N Y N Y Y Y Y Y Y Y Y Y Y N N N N
Available in Sol 700
Fatal Error Y Y
Patran Support
N Y N Y N Y N Y Y Y Y N Y Y Y Y Y
Y (8 nodes only) Y (new) Y (new) Y Y Y Y
N Y Y
Main Index
Chapter 2: MD Nastran Data Files 39
Supported SOL 700 Entries
Item CORD1C CORD1R CORD1RX CORD1S CORD2C CORD2R CORD2RX CORD2S CORD3G CORD3R CORD3RX COUOPT COUP1FL COUPINT COUPLE CYLINDR CPENTA CQUAD4 CQUAD8 CQUADR CQUADX CREEP CROD CSHEAR CSHP Y Y Y Y Y Y Y Y N Y Y Y (new) Y (new) Y (new) Y (new) Y (new)
Available in Sol 700
Fatal Error
Patran Support Y Y
Y Y Y
Y Y N
Y
Y Y Y Y Y Y Y Y Y Y Y
Y (6 nodes only) Y Y (4 nodes only) Y N N Y N Y (new)
Main Index
40 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Supported SOL 700 Entries
Item CSPOT CSPR CTETRA CTQUAD CTRIA3 CTRIA6 CTRIAR CTRIAX CTRIAX6 CTTRIA CTUBE CVISC CWELD D2R0000 D2RAUTO D2RINER DAMPGBL DAMPMAS DAMPSTF DAREA DBEXSSS DETSPH DLOAD DYCHANG DYDELEM Y Y
Available in Sol 700
Fatal Error
Patran Support Y Y Y
Y (4 or 10 nodes only) Y Y Y (3 nodes only) Y N N Y Y Y N Y Y Y Y Y Y Y Y Y (new) Y Y Y Y Y Y
Y Y Y
Y Y
Y Y Y Y Y Y
Y
Main Index
Chapter 2: MD Nastran Data Files 41
Supported SOL 700 Entries
Item DYPARAM DYRELAX DYRIGSW DYTERMT DYTIMHS ENDDATA ENDDYNA EOSGAM EOSGRUN EOSIG EOSJW EOSMG EOSPOL EOSTAB EOSTABC EOSTAIT FAILMPS FFCONTR FLOW FLOWDEF FLOWSPH FLOWT FORCE FORCE1 FORCE2 Y Y Y Y Y Y Y Y Y (new) Y (new) Y (new) Y (new) Y Y Y Y (new) Y (new) Y (new) Y (new) Y (new) Y (new) Y (new) Y N Y
Available in Sol 700
Fatal Error
Patran Support
N Y
Y Y Y Y
Y Y Y Y
Y Y
Y Y Y N
Main Index
42 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Supported SOL 700 Entries
Item FORCEAX GBAG GBAGCOU GENEL GRAV GRDSET GRIA GRID HEATLOS HGSUPPR HTRCONV HTRRAD HYDSTAT INCLUDE INFLCG INFLFRC INFLGAS INFLHB INFLTNK INFLTR INITGAS IPSTRAIN ISTRESS ISTRSBE ISTRSSH N Y Y (new) N Y Y Y Y Y (new) Y Y (new) Y (new) Y (new) Y Y (new) Y Y (new) Y (new) Y (new) Y (new) Y (new) N N Y Y
Available in Sol 700
Fatal Error Y
Patran Support
Y
Y Y N Y Y
Y
Y N
Y Y Y Y
Main Index
Chapter 2: MD Nastran Data Files 43
Supported SOL 700 Entries
Item ISTRSSO ISTRSTS LEAKAGE LOAD LSEQ MAT1 MAT2 MAT3 MAT8 MATDxxx MATDERO MATDEUL MATEP MATF MATG MATHE MATHED MATHP MATORT MATRIG MATS1 MATVE MATVORT MATVP MESH Y (new) Y Y Y Y Y Y Y
Available in Sol 700
Fatal Error
Patran Support Y Y Y Y Y Y Y Y Y Y Y Y
Y (see Supported Material Models in SOL 700 in this chapter) Y Y (new) N Y N N N Y N Y Y N N N Y (new) Y Y Y Y Y Y Y Y
Y
Y
Y
Main Index
44 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Supported SOL 700 Entries
Item MFLUID MOMAX MOMENT MOMENT2 MPC MPCAX NLPARM NLRGAP NOLINi NTHICK PANEL PBAR PBARL PBCOMP PBEAM PBEAM71 PBEAMD PBEAML PBELTD PBEND PBSPOT PBUSH PCOMP PCOMPA PCOMPG N N Y Y Y N
Available in Sol 700
Fatal Error Y Y
Patran Support
Y N Y Y Y Y Y Y Y Y Y Y Y
Y (for pseudo statics) N N N N Y Y N Y Y Y Y Y N Y N Y Y N Y Y Y
Y
Y
Main Index
Chapter 2: MD Nastran Data Files 45
Supported SOL 700 Entries
Item PDAMP PDAMP5 PELAS PELAS1 PELAST PERMEAB PERMGBG PEULER PEULER1 PGAP PHBDY PINTC PINTS PLOAD PLOAD1 PLOAD2 PLOAD4 PLOADX1 PLPLANE PLSOLID PMARKER PMINC PORFCPL PORFGBG PORFLOW Y N Y Y N Y (new) Y (new) Y (new) Y (new) N N N N Y N Y Y N Y Y Y (new) Y (new) Y (new) Y (new) Y (new)
Available in Sol 700
Fatal Error
Patran Support Y
Y Y
Y Y
Y Y Y Y Y Y N Y N Y Y
Y Y
Y
Main Index
46 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Supported SOL 700 Entries
Item PORFLWT PORHOLE PORHYDS PRESTRS PMASS PRESPT PROD PSHEAR PSHELL PSHELL1 PSHELLD PSOLID PSOLIDD PSPH PSPRMAT PTSHELL PTUBE PVISC RBAR RBE1 RBE2 RBE2A RBE2D RBE2F RBE3 Y (new) Y (new) Y (new) Y N N Y N Y Y Y Y Y Y (new) Y Y Y Y Y N Y Y Y Y Y
Available in Sol 700
Fatal Error
Patran Support Y Y
Y Y Y Y Y Y Y
Y Y Y
Y Y Y Y Y
Y
Main Index
Chapter 2: MD Nastran Data Files 47
Supported SOL 700 Entries
Item RBE3D RBJOINT RBJSTIFF RCONN RESTART RFORCE RLOADi RROD RSPLINE RTRPLT SBPRET SBRETR SBSENSR SBSLPR SLOAD SEQROUT SHREL SHRPOL SPC SPC1 SPCADD SPCAX SPCD SPCD2 SPHDEF Y Y Y Y Y
Available in Sol 700
Fatal Error
Patran Support
N N
Y (CID, METHOD, continuation line not supported) N N N N Y Y Y Y N Y (new) Y (new) Y (new) Y Y Y N Y Y Y (new) Y Y Y Y Y Y
Y
Y
Y Y Y
Y Y Y
Main Index
48 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Supported SOL 700 Entries
Item SPHERE SPHSYM SUPAX SURFINI SPWRS TABLED1 TABLED2 TABLED3 TABLEDR TABLES1 TEMP TEMPD TIC TICD TIC3 TICEL TICEUL1 TICREG TICVAL TIMNAT TIMNVH TIMSML TLRE1 TLOAD1 TLOAD2 Y (new) Y Y Y Y Y N N Y Y Y Y (new) Y (new) Y (new) Y (new) Y Y Y Y Y Y Y (new) Y (new) N
Available in Sol 700
Fatal Error
Patran Support Y
Y Y
Y N N Y Y Y Y Y
Y Y Y Y Y
Y N
Main Index
Chapter 2: MD Nastran Data Files 49
Supported SOL 700 Entries
Item TODYNA TSTEP TSTEPNL UGASC WALL WALLGEO YLDHY YLDJC YLDMC YLDMSS YLDPOL YLDRPL YLDSG YLDTM YLDVN YLDZA Y
Available in Sol 700
Fatal Error
Patran Support
Y (changed to TSTEPNL) Y Y (new) Y Y (new) Y (new) Y (new) Y (new) Y (new) Y (new) Y (new) Y (new) Y (new) Y (new) Y (new)
Y Y Y Y Y Y Y Y
Y Y Y Y Y Y
Main Index
50 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Supported SOL 700 Entries
Supported Material Models in SOL 700
In addition to the MD Nastran material models listed in the table above, the following material models are supported in SOL 700: MATD001, MATD2AN, MATD20M, MATD003, MATD005, MATD006, MATD007, MATD009, MATD010, MATD012, MATD013, MATD014, MATD015, MATD018, MATD019, MATD020, MATD022, MATD024, MATD026, MATD027, MATD028, MATD029, MATD030, MATD031, MATD032, MATD034, MATD040, MATD054, MATD055, MATD057, MATD058, MATD059, MATD062, MATD063, MATD064, MATD066, MATD067, MATD068, MATD069, MATD070, MATD071, MATD072, MATD72R, MATD073, MATD074, MATD076, MATD077, MATD080, MATD081, MATD083, MATD087, MATD089, MATD093, MATD094, MATD095, MATD097, MATD098, MATD099, MATD100, MATD112, MATD114, MATD119, MATD0121, MATD123, MATD126, MATD127, MATD158, MATD181, MATD196, MATDB01, MATDS01, MATDS02, MATDS03, MATDS04, MATDS05, MATDS06, MATDS07, MATDS08, MATDS13, MATDS14, MATDS15, MATDSW1, MATDSW2, MATDSW3, MATDSW4, MATDSW5
Unsupported Materials in MD Patran Preference
All the material models listed above are supported with the exception of the following: MATD018, MATD034, MATD040, MATD066, MATD072, MATD071, MATD72R, MATD087, MATD089, MATD093, MATD094, MATD095, MATD097, MATD112, MATD114, MATD119, MATD0121, MATD123, MATD158
Supported Parameters in SOL 700
PARAM,DYBEAMIP, PARAM,DYBLDTIM, PARAM,DYBULKL, PARAM,DYBULKQ1, PARAM,DYCMPFLG, PARAM*,DYCONSLSFAC, PARAM*,DYCONRWPNAL, PARAM*,DYCONPENOPT, PARAM*,DYCONTHKCHG, PARAM*,DYCONENMASS, PARAM*,DYCONECDT, PARAM*,DYCONIGNORE, PARAM*,DYCONORIEN, PARAM*,DYCONSKIPRWG, PARAM,DYDCOMP, PARAM,DYCOWPRD, PARAM,DYCOWPRP, PARAM, DYDEFAUL, PARAM,DYDTOUT, PARAM,DYENDTIM, PARAM*,DYENERGYHGEN, PARAM,DYENGFLG, PARAM,DYEPSFLG, PARAM,DYHRGIHQ, PARAM,DYHRGQH, PARAM,DYIEVERP, PARAM*,DYINISTEP, PARAM,DYLDKND, PARAM,DYMAXINT, PARAM*,DYMAXSTEP, PARAM*,DYMINSTEP, PARAM,DYNAMES, PARAM,DYNEIPH, PARAM,DYNEIPS, PARAM,DYNINTSL, PARAM,DYN3THDT, PARAM,DYRBE3, PARAM,DYRLTFLG, PARAM*,DYSHELLFORM, PARAM,DYSHGE, PARAM*,DYSHTHICK, PARAM,DYSIGFLG, PARAM,DYSTATIC, PARAM*,DYSTEPFCTL, PARAM,DYSTRFLG, PARAM,DYSTSSZ, PARAM*,DYTERMNDTMIN, PARAM*,DYTERMNENDMAS, PARAM*,DYTSTEPERODE,
Main Index
Chapter 2: MD Nastran Data Files 51
Preprocessing with MD Patran
PARAM*,DYTSTEPDT2MS, PARAM,DYXPENE, PARAM,S700NVH, PARAM,SBOLTZ, PARAM,SCALEMAS, PARAM,UGASC, PARAM,DYININT, PARAM,DYNREAL, PARAM,COPOR, PARAM,EULBND, PARAM,EULBULKL, PARAM,EULBULKQ, PARAM,EULBULKT, PARAM, EULSTRES, PARAM,FBLEND, PARAM,FMULTI, PARAM,GRADMESH,PARAM,MICRO, PARAM,RKSCHEME, PARAM,ROHYDRO, PARAM,ROMULTI, PARAM,ROSTR, PARAM,VELCUT, PARAM,S700NVH1, DYPARAM,AXIALSYM, DYPARAM,EULERCUB, DYPARAM, EULTRAN, DYPARAM,FASTCOUP, DYPARAM,FSIDMP, DYPARAM,HYDROBOD, DYPARAM,LIMITER, DYPARAM,VELMAX
Supported Parameters in MD Patran Preference
DYBEAMIP, DYBLDTIM, DYBULKL, DYCMPFLG, DYCONECDT, DYCONENMASS, DYCONIGNORE, DYCONPENOPT, DYCONRWPNAL, DYCONSLSFAC, DYCONSKIPRWG, DYCONTHKCHG, DYCOWPRD, DYCOWPRP, DYDCOMP, DYDTOUT, DYENERGYHGEN, DYENGFLG, DYEPSFLG, DYHRGIHQ, DYHRGQH, DYIEVERP, DYINISTEP, DYLDKND, DYMAXINT, DYMAXSTEP, DYMINSTEP, DYN3THDT, DYNEIPH, DYNEIPS, DYNINTSL, DYRLTFLG, DYSHELLFORM, DYSHGE, DYSHTHICK, DYSIGFLG, DYSTATIC, DYSTRFLG, DYSTEPFCTL, DYSTSSZ, DYTERMNENDMAS, DYTSTEPDT2MS, DYTSTEPERODE, PARAM,S700NVH1
Preprocessing with MD Patran
MD Patran offers a MD Nastran interface that provides a communication link between MD Patran and MD Nastran. It provides for the generation of the MD Nastran Input file and customization of certain features in MD Patran. The interface is a fully integrated part of the MD Patran system.
Generating the Bulk Data File
Selecting MD Nastran as the analysis code preference in MD Patran ensures that sufficient and appropriate data is generated for the MD Nastran. Specifically, the MD Patran forms in these main areas are modified:
• Materials • Element Properties • Finite Elements/MPCs and Meshing • Loads and Boundary Conditions • Analysis Forms
Main Index
52 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Output Requests
Using MD Patran, you can run a MD Nastran analysis or you may generate the MD Nastran Input File to run externally. For information on generating the MD Nastran Input file from within MD Patran, see “Analysis Form” in Chapter 3 of the MD Patran MD Nastran Preference Guide, Volume 1: Structural Analysis.
Editing the Bulk Data File
Once the bulk data file has been generated, you can edit the file directly from MD Patran. 1. Click the Analysis Application button to bring up the Analysis Application form. 2. On the Analysis form set the Action>Object>Method combination to Analyze>Existing Deck>Full Run and click Edit Input File. MD Patran finds the bulk data file with the current database name and displays the file for editing in a text editing window.
Output Requests
The SOL 700 default output file is the d3plot file which is a binary file including the displacements, stresses and strain output results. The file d3plot can be postprocessed with a number of commercially available tools such as MD Patran, ETA/Femb, Hypermesh, CEI/Ensight and LSPOST. In addition to d3plot, other output files are optionally available for output request.
Files Created by SOL 700
SOL 700 creates a number of files during the analysis. In the main sections of this manual, generic names are used when referring to a particular MD Nastran SOL 700 file. These generic reference names and the actual generic file names are given below: Generic Reference Name
dat or bdf d3hsp d3plot d3thdt ARC binout ssstat
File Input (ASCII) Output/d3hsp (ASCII) Result Output (Binary) Time History (Binary) Result Output (Binary) Result Output (Binary) Result Output (ASCII)
SOL 700 Name
Filename.dat (or .bdf) Filename.dytr.d3hsp Filename.dytr.d3plot Filename.dytr.d3thdt Filename.dytr.ARC Filename.dytr.binout Filename.dytr.ssstat
Main Index
Chapter 2: MD Nastran Data Files 53
Output Requests
File Restart Dump file (Binary) Execution Summary file (ASCII) Warnings and Errors (ASCII) MD Nastran Result file (Binary) MD Nastran Result file (Binary) MD Nastran Punch file (ASCII) Input File
Generic Reference Name
d3dump log f06 op2 xdb pch
SOL 700 Name
Filename.dytr.d3dump Filename.dytr.log Filename.f06 Filename.op2 Filename.xdb Filename.pch
The input file contains all the input data and must be present in order to run MD Nastran SOL 700. It is a text file with up to 80 characters on each line. The primary input file may reference one or more include files. Output (d3hsp) File The output (d3hsp) file is a text file suitable for printing purposes or viewing with a standard editor. It contains messages produced by explicit solver in SOL 700, which contains summary of model input, calculation status at every 100 cycles and error messages This file is normally also included in the f06 file. The quantities that are placed on this file are controlled by PARM statements. d3plot MD Nastran can create any number of output files containing results at times during the analysis. The d3plot files are binary files generated by explicit solver in SOL 700. They contain a complete description of the geometry and connectivity of the analysis model and the requested results. MD Patran can read d3plot files for postprocessing. ARC MD Nastran outputs the fluid structure interaction (FSI) results in the *ARC file. The ARC files are binary files and are consistant with Dytran output. To postporcess the FSI results by Patran, both d3plot and ARC files have to be read. binout The binout file is the file generated by SOL 700 and is consistent with the LSDYNA native binout file. It includes the time history results in binary format.
Main Index
54 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Output Requests
ssstat File The sstat file is generated by SOL 700 and contains the information of the subsystem. The DYPARAM,LSDYNA,DATABASE,SSTATM and DBEXSSS bulk data entries can be activated to generate the sstat file. The information of the sstat file is useful for FBORD chaining analysis. TimeHistory (d3thdt) Files SOL 700 can also create any number of timehistory files containing results for particular grid points and elements during the calculation. They are also binary files. Timehistory files only contain results, no geometry or connectivity. Restart Files SOL 700 supports the three kinds of restart. These are named simple, minor (or small) and full. Simple restarts are jobs that for some reason did not finish (for example the computer went down) and involve no changes in the original Nastran input deck. They can use either the running restart file (jid.dytr.runrsf) or the d3dump file (jid.dytr.d3dumpxx where xx is a numerical value such as 00, 01, 02 ...). Minor restarts involve simple changes to the Nastran input deck such as deleting contact surfaces, changing the run time or output time intervals, switching rigid bodies to deformable or visa versa. These require a Nastran input deck with the executive control, case control and only the changes as well as the jid.dytr.d3dumpxx file. The full restart normally involves massive changes to the Nastran deck where the complete executive control, case control and bulk data are input as well as restarting with the jid.dytr.d3dumpxx file. Error File SOL 700 produces an f06 file containing a summary of all warnings and errors issued during reading and subsequent data processing. SOL 700 has five levels of messages below: 1. Informative messages. An example of a Level 1 message is a message that indicates that a new processor has begun execution. These message provide job information. 2. Nonfatal warning message of something that could affect the results. An example of a Level 2 message is one indicating that the aspect ratio is greater than 15. This may or may not be a serious problem. 3. Serious warnings. An example of a Level 3 message is a warning about a highly distorted element. 4. Fatal error (all occurrences will be found before aborting). An example of a Level 4 message is the warning “undefined node used in rigid element.” 5. Immediately fatal errors. An example of a Level 5 message is “Unable to open file” message. The job is immediately aborted.
Main Index
Chapter 2: MD Nastran Data Files 55
Postprocessing with MD Patran
MD Nastran Native Result Files In addition to d3plot which is native result file, SOL 700 optionally outputs MD Nastran native result files such as op2, xdb, and pch files for postprocessing in version 2006r2 and subsequent versions.
Postprocessing with MD Patran
The results from an MD Nastran Explicit Nonlinear Analysis can be read into and postprocessed by MD Patran. Typically, you will get the most complete set of results if you use the d3plot results options (see Section 14.1 “Output from the Analysis” on how to select which output files will be created), but you can also postprocess using an .xdb or .op2 formatted file. The Results application in MD Patran provides the capabilities for creating, modifying, deleting, posting, unposting, and manipulating results visualization plots and viewing the finite element model. In addition, results can be derived, combined, scaled, interpolated, extrapolated, transformed, and averaged in a variety of ways which are controllable by you. Control is provided for manipulating the color/range assignment and other attributes for plot tools, and for controlling and creating animations of static and transient results. Results are selected from the database and assigned to plot tools using simple forms. Results transformations are provided to derive scalars from vectors and tensors and to derive vectors from tensors. This allows for a wide variety of visualization tools to be used with all of the available results. If the job was created within MD Patran such that a MD Patran jobname of the same name as the MD Nastran jobname exists, you only need to use the Results tools and MD Patran imports or attaches the jobname.xxx file without you having to select it. If you did not create the job in MD Patran, you can still import the model and results and postprocess. For more information, see MD Patran Reference Manual, Part 6: Results Postprocessing.
Main Index
56 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Postprocessing with MD Patran
Main Index
Chapter 3: Modeling MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
3
Modeling
Coordinate Systems Nodes Elements 61 62 63 67 79 58
Modeling in MD Patran
Example using MD Patran
Input and Output Files Created During the Simulation Postprocessing 81 85 85 85
Running a Batch Job
How to Tell When the Analysis is Done
How to Tell if the Analysis Ran Successfully
Main Index
58 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Coordinate Systems
Coordinate Systems
The basic coordinate system in MD Nastran Explicit Nonlinear is a righthanded rectangular coordinate system as shown in Figure 31.
Z
Y
X
Figure 31
Basic Coordinate System
Userdefined Coordinate System
MD Nastran provides six Bulk Data entry options for defining coordinate systems. The new coordinate system is directly or indirectly related to the basic coordinate system. The ten options are: CORD1R CORD2R CORD3R CORD1RX CORD2RX CORD3RX CORD1C CORD2C CORD1S CORD2S
Rectangular
Cylindrical Spherical
The CORD1R, CORD1C, and CORD1S entries define coordinate systems by referencing three defined grid points. The CORD2R, CORD2C, and CORD2S entries define coordinate systems by specifying the location of three points.
Main Index
Chapter 3: Modeling 59
Coordinate Systems
Nodal Coordinate Systems
Each grid point refers to two coordinate system. One system is used to locate the grid point (CP in field 3) and the other is used to establish the grid point’s displacement coordinate system (CD in field 7). The displacement coordinate system defines the direction of displacements, constraints and applied loads. The basic (default coordinate system) is indicated by a zero or blank in the CP and CD fields. CD and CP do not have to be the same coordinate system. The grid point output is with respect to basic coordinate system.
Element Coordinate Systems
The connectivity of the CQUAD4s defines the element coordinate system. It is a rectangular coordinate system and the direction of axes depends on the order of the grid points in the connectivity entry. The zaxis is normal to the shell as shown in the figure below. The xaxis is a vector from grid point1 to grid point 2. The yaxis is perpendicular to both the xaxis and the zaxis and is the direction defined by the righthand rule.
Zelem
G4 G3
Xelem
G1 G2
The counter clockwise node numbering determines the positive zdirection and is the top surface of the shell element and the bottom surface is the negative zdirection. By default, shell stresses/strains written to d3plots are in the basic coordinate system and shell stresses/strains written to elout (time history element output) are in the element local coordinate system. The local coordinate system of CBAR/CBEAM is explained in the Remarks of PBAR and PBEAM entries respectively in the quick reference guide. By default, stresses/resultants for beams are written to d3plot and elout (time history element output) in the element coordinate system. The output of solid elements in d3plot and elout (time history element output) is always with respect to the basic coordinate system.
Main Index
60 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Coordinate Systems
Material Coordinate Systems
For shell elements of anisotropic material, there are three options for defining the initial direction of the material axes, AOPT = 0, 2, and 3 (see MATD2AN, MATD022, MAT054, MATD059). During the solution, the element coordinate system rotates and transforms with the element, so the angle between the element system and material system can be assumed to remain constant. In other words, the material direction is constantly updated as the element rotates and deforms. It is, therefore, sufficient to define the material coordinate system in the undeformed geometry. For this discussion, the material coordinate system is called the abc system to be consistent with the user's manual. For shell elements, the caxis coincides with the element normal, the aaxis is in the plane of the shell, and the baxis is determined by the cross product, b = c x a. Actually, for warped elements, the aaxis is not exactly in the element plane, but is projected along the caxis such that it is orthogonal to c. This projection is also done for the local element system, so a simple 2D transformation between the two systems is possible. For reasons discussed in the last paragraph, the three available shell element options for defining the abc system boil down to defining the aaxis. For AOPT = 0, the aaxis is assumed to be equal to the local element system xaxis. For AOPT = 2, the aaxis is defined as the userdefined vector a , projected onto the surface of the element. Note that the userdefined vector d is not used at all. Also, note that the userdefined vector a , is not equivalent to the aaxis of the material unless a is orthogonal to the element normal. For AOPT = 3, the aaxis is defined by the cross product of a userdefined vector v with element normal; i.e., a = v x c. Given the same userdefined vector, AOPT = 3 defines a material coordinate system that is rotated exactly 90 degrees from coordinate system defined by AOPT. The material axes as defined by the AOPT option in the material input are rotated by the THETA element (in CQUAD4) angle to get the reference direction for the element. The material axes for the element integration points are then rotated by the integration point THETA angles in PCOMP option. In summary, AOPT, THETA in CQUAD4, and the integration point THETA angles in PCOMP go into defining the material directions at each integration point.
Main Index
Chapter 3: Modeling 61
Nodes
Nodes
Model geometry is defined in MD Nastran with grid points. A grid point is a point on or in the structural continuum which is used to define a finite element. A simple model may have only a handful of grid points; a complex model may have many hundreds of thousands. The structure’s grid points displace with the loaded structure. Each grid point of the structural model has six possible components of displacement: three translations (in the x, y, or zdirections) and three rotations (about the x, y, or zaxes). These components of displacement are called degrees of freedom (DOFs).
Degrees of Freedom
The degrees of freedom in MD Nastran Nonlinear are always referred to as: 1 2 3 4 5 6 xdisplacement ydisplacement zdisplacement Rotation about the xaxis Rotation about the yaxis Rotation about the zaxis
MD Nastran only activates those degrees of freedom needed at a node. Thus, some of the degrees of freedom listed above may not be used at all nodes in a model, because each element type only uses those degrees of freedom which are relevant. For example, twodimensional solid (continuum) stress/displacement elements only use degrees of freedom 1 and 2. The degrees of freedom actually used at any node are thus the envelope of those variables needed in each element that uses the node.
Main Index
62 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Elements
Elements
Once the geometry (grid points) of the structural model has been established, the grid points are used to define the finite elements. MD Nastran has an extensive library of finite elements covering a wide range of physical behavior. Some of these elements and their names are shown in figure below. The C in front of each element name stands for “connection.”
• Point Element (not a finite element, but can be included in the finite element model)
CONM2 (Concentrated mass)
• Spring Elements (they behave like simple extensional or rotational springs)
CELAS1, CELAS2, CELAS1D, CELAS2D
• Line Elements (they behave like rods, bars, or beams)
CROD, CBAR, CBEAM, CONROD
• Surface Elements (they behave like membranes or thin plates)
CTRIA3
CQUAD4
• Solid Elements (they behave like bricks or thick plates)
CHEXA
CPENTA
CTETRA
Main Index
Chapter 3: Modeling 63
Modeling in MD Patran
• Rigid Bar (infinitely stiff without causing numerical difficulties in the mathematical model)
RBAR Structural elements are defined on Bulk Data connection entries that identify the grid points to which the element is connected. The mnemonics for all such entries have a prefix of the letter C, followed by an indication of the type of element, such as CBAR and CROD. The order of the grid point identification defines the positive direction of the axis of a onedimensional element and the positive surface of a plate element. The connection entries include additional orientation information when required. Some elements allow for offsets between its connecting grid points and the reference plane of the element. The coordinate systems associated with element offsets are defined in terms of the grid point coordinate systems. For most elements, each connection entry references a property definition entry. If many elements have the same properties, this system of referencing eliminates a large number of duplicate entries. Details for each element type are described in Chapter 6: Elements.
Modeling in MD Patran
In MD Patran, geometric models are the foundation on which most finite element models are built. Geometric curves, surfaces, or solids provide the base for creating nodes, elements, and loads and boundary conditions; the geometric model also serves as the structure to which material properties, as well as element properties, may be assigned even before any mesh is actually generated.
Creating Geometry in MD Patran
Model geometry may be constructed in MD Patran, accessed directly from a CAD application, or imported in specially formatted translator files. Whatever the source of the geometry, a single geometric model will be maintained throughout all geometric and finite element operations. Geometric entities, even if obtained from external files, retain their original mathematical representation without any approximations or substitutions. Accessing the Geometry Application In MD Patran you can create, modify, and delete points, curves, surfaces, and solids. MD Patran assigns a default color to the display of all geometric entities. Pick the Geometry icon in the MD Patran Main Form to access the Geometry application.
Main Index
64 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Modeling in MD Patran
The Geometry form controls all processes in the Geometry application. The top portion of the form contains three keywords, Action, Object, and Method; these remain the same throughout all activities. The rest of the entries will vary depending on the requirements posed by the specified action, object, and method.
Action Object
Names the operation to be performed; for example Create, Edit, or Delete. Identifies the geometric entity upon which the action is performed, for example, Solid. In this case, if the Action is Create, then the command requests that a solid be created. Specifies the procedure which perform the action. Taking the above example one step further, if the Method is Surface, a solid is created by one of the techniques that utilize surfaces.
Method
There are hundreds of action, object, method combinations available for creating geometric entities in MD Patran. For complete descriptions on creating geometry models, see the MD Patran Reference Manual, Part 2: Geometry. Utilizing External Geometry (CAD) Files MD Patran can make use of geometry created in databases outside of MD Patran by either accessing geometric data directly from one of several CAD systems, or importing geometry using special files. Geometry access, performed through the unique Direct Geometry Access (DGA) feature, does not require any translation. MD Patran accesses the original geometry and uses the geometric definitions of all entities. On the other hand, when geometry is imported, MD Patran first evaluates the mathematical definition of entities in their originating CAD system, and then formulates the information to be appropriate for MD Patran operations. Imported geometry comes to MD Patran via IGES, Express Neutral files, or MD Patran Neutral files. IGES (Initial Graphic Exchange Specification) is an ANSI standard formatted file that makes it possible to exchange data among most commercial CAD systems. Express Neutral files are intermediate files created during a Unigraphics or CV CAD model access. MD Patran Neutral files are specially formatted for the purpose of providing a means of importing and exporting model data.
Main Index
Chapter 3: Modeling 65
Modeling in MD Patran
Geometry received into the database, whether through direct access or import, is treated as if it had been built in MD Patran; meshing, load and boundary condition assignments, element and material properties definitions are all performed as if on MD Patran’s own “native” geometry.
Creating Finite Element Meshes in MD Patran
Finite elements themselves are defined by both their topology (i.e., their shape) and their properties. For example, the elements used to create a mesh for a surface may be composed of quadrilaterals or triangles. Similarly, one element may be a steel plate modeling structural effects such as displacement and rotation, while another may represent an air mass in an acoustic analysis.MD Patran provides numerous ways to create a finite element mesh. At this stage of using MD Patran, where you are creating a finite element mesh using the Finite Elements application form, elements are defined purely in terms of their topology. Other properties such as materials, thickness and behavior types are then defined for these elements in subsequent applications, and discussed in Chapter 6: Elements of this guide. The most rudimentary method of creating a finite element mesh is to manually generate individual nodes, and then to create individual elements from previously defined nodes. Individual nodes can be either be generated from the geometry model or directly created using node creation tools that bypass the need for point definitions. A finite element model created manually supports the entire MD Patran element library and where applicable, MD Patran automatically generates midedge, midface and midbody nodes. MD Patran contains many capabilities to help you manually create the right kind of finite element mesh for your model, and capabilities that automate the process of finite element creation. MD Patran provides the following capabilities for finite element modeling (FEM):
• Mesh seeding tools to control specific mesh densities in specific areas of your geometry. • Several highly automated techniques for mesh generation. • Equivalencing capabilities for joining meshes in adjacent regions. • Tools to verify the quality and accuracy of your finite element model. • Capabilities for direct input and editing of finite element data.
Automatic Meshing Tools There are four basic mesh generation techniques available in MD Patran: IsoMesh, Paver Mesh, Auto TetMesh, and 21/2D Meshing. Selecting the right technique for a particular model must be based on geometry, model topology, analysis objectives, and engineering judgment. Isomesh Creates a traditional mapped mesh on regularly shaped geometry via simple subdivision. This method creates Quad and Tria elements on surfaces and brick elements on solids. The resulting mesh supports all element configurations in MD Patran.
Main Index
66 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Modeling in MD Patran
Paver The Paver is an automated surface meshing technique that you can use with any arbitrary surface region, including trimmed surfaces, composite surfaces, and irregular surface regions. Unlike the IsoMesh approach, the Paver technique creates a mesh by first subdividing the surface boundaries into mesh points, and then operates on these boundaries to construct interior elements. TetMesh Arbitrary solid mesher generates tetrahedral elements within MD Patran solids defined by an arbitrary number of faces or volumes formed by collection of triangular element shells. This method is based on MSC plastering technology. 21/2D Mesher Transforms a planar 2D mesh to produce a 3D mesh of solid elements, using sweep and extrude operations. Caution: In all finite element simulations, the size and shape of the elements is one aspect that controls the accuracy. In explicit simulations, small elements will also result in the need to reduce the time step or increase the mass scaling. See Chapter 4: Special Modeling Techniques.
Accessing the Finite Element Application All of MD Patran’s finite element modeling capabilities are available by selecting the Finite Element button on the main form.
Like the Geometry Application, the top portion of the Finite Element form contains three keywords, Action, Object, and Method; these remain the same throughout all activities. Finite Element (FE) Meshing, Node and Element Editing, Nodal Equivalencing, ID Optimization, Model Verification, FE Show, Modify and Delete, and ID Renumber, are all accessible by setting the Action/Object/Method combination on the Finite Elements form. For complete descriptions on creating geometry models, see the MD Patran Reference Manual, Part 3: Finite Element Modeling.
Main Index
Chapter 3: Modeling 67
Example using MD Patran
Example using MD Patran
The objective of the following example is to highlight the templates and forms that are used in MD Nastran SOL 700 Preference to set up the model, define the material, boundary conditions, type of output files generated and post processing.
Description of the Problem
The figure below is an example of a tapered beam which impact a rigid wall at 10,000 inches/seconds.
Steel Properties Density = 0.000783 Ibm/in3 Elastic Modulus = 21E9 psi Yield Stress = 1.4E7 psi Poisson’s Ratio = 0.3
Impact Velocity = 10,000in/sec
Rigid Plate = 6x6 in
Beam Tip = 2x2 in Beam End = 4x4 in Beam Length = 40 in
Y X Z
Impact of a Tapered Beam to a Rigid Wall
The beam is 40 inches in length and is made of steel with the following properties: Density = 0.000783 lbm/in3 Elastic Modulus = 2.1E9 psi Yield Stress = 1.4E7 psi Poisson’s ratio = 0.3 The example is designed to make the user familiar with the following objectives: 1 – How to define a Contact Body and Contact Table. 2 – How to define the initial velocity for the tapered beam. 3 – How to define elastoplastic material properties using MAT24 material model.
Main Index
68 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Example using MD Patran
Solution Type
MD Nastran can simulate many different types of structural response. In general an analysis type can be either static or dynamic. In a static analysis, loads and boundary conditions are applied to a model and the response is assumed to remain constant over time. In dynamic analysis the response changes over time. In MD Nastran, both static and dynamic analysis may simulate linear response or nonlinear response. SOL 700 incorporates the formulations and functionality to simulate nonlinear dynamic structural responses. The specific procedure MD Nastran uses is specified on the Executive Control Statement by the ID entry.
Specifying the Solution Type
MD Nastran Explicit Nonlinear (SOL 700) is designated with the following Executive Control Statement in the MD Nastran Bulk Data file, where the ID entry indicates which analysis procedure is to be run.
References
• “SOL 700,ID” in the MD Nastran Quick Reference Guide. In this release, ID = 129 indicates
a nonlinear dynamic analysis.
Main Index
Chapter 3: Modeling 69
Example using MD Patran
Defining the Solution Type in MD Patran
Prior to selecting a Solution Type, check to see that under Analysis Preferences the Analysis Code is set to MD Nastran, and then set the Analysis Type to Explicit Nonlinear. Open a new data base and name it tapered_beam: a. Open File and click New. b. Select MD Nastran for Analysis Code. c. Type tapered_beam under File name and click OK. d. Select Explicit Nonlinear for analysis and click OK.
a
c d
b
Main Index
70 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Example using MD Patran
Defining Material Model Define material properties using MAT24 constitutive material model: a. Materials: Action: Object: Create Isotropic (lsdyna) e. Enter Density Poisson Ratio Yield Strength f. Click OK g. Click Apply = 0.000783 = 0.3 = 1.4E7 Elastic Modulus = 2.1E9
Method: Manual Input b. Material Name: steel c. Click Input Properties d. Constitutive Model: Elastoplastic
Implementation: Curve Type Piecewise Linear (MAT24) Bilinear
a
d
e
b
c g
f
Main Index
Chapter 3: Modeling 71
Example using MD Patran
Defining the Initial Velocity Define a group called “beam” and proceed with creating the beam finite element model. Post group beam and define an impact velocity of 10000 in/sec in positive Z direction for the beam. a. Group: Post b. Select beam c. Click Apply and then Cancel d. Loads/Boundary Conditions Action: Object: Create Initial Velocity h. i. j. k. l. Click OK Click Select Application Region Click FEM Select Nodes: Select all beam nodes Click ADD Click Apply
m. Click OK n.
Method: Nodal e. Enter New Set Name initial_velocity f. Click Input Data g. Enter Trans Veloc = <0 0 10000>
a
d
b
g
j
e
l
k
c
f i n
m
Main Index
72 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Example using MD Patran
Defining the Rigid Plate Create a group called rigid_plate and proceed with constructing the finite element model of the rigid plate. Post group rigid_plate and define a constraint on the plate. a. Group: Post b. Select rigid_plate c. Click Apply and then Cancel d. Loads/Boundary Conditions h. Action: Object: Method: e. Enter New Set Name fix_plate f. Click Input Data Create Displacement Nodal i. j. k. l. g. Enter Translations: Rotations: Click OK Click Select Application Region Select Nodes: Select all the nodes on the plate. Click ADD Click OK <0 0 0> <0 0 0>
m. Click Apply
a
d
g
b
j k
e
h
l
c
m
f i
Main Index
Chapter 3: Modeling 73
Example using MD Patran
Defining the Contact Between the Beam and the Rigid Plate Post both group beam and rigid_plate and define a contact between the plate and the beam. a. Loads/Boundary Conditions: Action: Object: Type: Option: Create Contact Element Uniform Deformable Body f. g. h. i. j. k. l. Click OK Click Apply New Set Name: beam Target Element Type: 3D Click Select Application Region Choose beam elements and click ADD Click OK
b. New Set Name: rigid_plate c. Target Element Type: 2D d. Click Select Application Region e. Choose plate elements and click ADD
m. Click Apply
a
e
k
b c d g
h
f
i j m
l
Main Index
74 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Example using MD Patran
Defining the Simulation End Time and Contact Table Now, we are ready for analysis. a. Analysis: Action: Object: Analyze Entire Model d. Available Subcases: Default e. Available Load Cases: Default (Subcase Default is related to Load Case Default) f. Click Subcase Parameters g. Enter End Time 0.005
Method: Full Run b. Job Name: tapered_beam
c. Click Select Application Region h. Click Contact Table
a
d
e
b
f
c
Note:
Only when “Contact Table” in Solution Parameter is clicked in MD Patran is the Bulk Data Input BCTABLE written to the input deck. Otherwise, BCONTACT  ALLBODY is written in the Case Control section of the input file.
Main Index
Chapter 3: Modeling 75
Example using MD Patran
i. Global Contact Detection: First> Second l. j. Click OK k. Click OK
Click Apply
m. Click Cancel
l
m
Main Index
76 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Example using MD Patran
Defining the Time Step for Output Files Specify the time step for the output files. a. Analysis: Action: Object: Analyze Entire Model
Method: Full Run b. Enter in Direct Text Input Bulk Data Section: PARAM, DYDTOUT, 0.0001 c. Click OK.
b
c
a
Main Index
Chapter 3: Modeling 77
Example using MD Patran
Running the Model Now you are ready to write the input deck. a. Analysis: Action: Object: Analyze Entire Model
Method: Full Run b. Click Apply
a
b
Main Index
78 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Example using MD Patran
Examine the MD Nastran Input File
$ NASTRAN input file created by the MSC MSC.Nastran input file $ translator ( MSC.Patran 13.1.073 ) on February 14, 2005 at 15:15:28. $ Direct Text Input for Nastran System Cell Section $ Direct Text Input for File Management Section $ Explicit NonLinear Analysis SOL 700,NLTRAN STOP=1 $ Direct Text Input for Executive Control CEND TITLE = MSC.Nastran job created on 14Feb05 at 15:01:47 $ Direct Text Input for Global Case Control Data .. .. $ Direct Text Input for this Subcase BEGIN BULK PARAM POST 0 PARAM LGDISP 1 PARAM PRTMAXIM YES TSTEPNL 1 100 .00005 1 ADAPT 2 10 .. .. BCTABLE 1 3 SLAVE 4 0. 0. 0. 0. 0 0. 0 0 0 MASTERS 4 SLAVE 4 0. 0. 0. 0. 0 0. 0 0 0 MASTERS 3 SLAVE 3 0. 0. 0. 0. 0 0. 0 0 0 MASTERS 3 $ Direct Text Input for Bulk Data PARAM, DYDTOUT, 0.0001 $ Elements and Element Properties for region : steel_beam PSOLID 1 1 0 $ Pset: "steel_beam" will be imported as: "psolid.1" CHEXA 1 1 1 2 5 4 10 11
$ Referenced Material Records $ Material Record : steel $ Description of Material : Date: 14Feb05 MATD024 1 7.834 2.1+9 .3 1.4+7 1.
This section describes the Explicit Nonlinear Solution Sequence.
This TSTEPNL describes the End Time of the simulation. The actual number of Time Steps and Time Increment is determined by SOL 700 during the analysis.
This section of describes the contact table definition, as well as the master body and the slave body in each contact relation.
Time: 14:52:07
This PARAM entry describes the time interval of d3plot outputs. This section describes the ElasticPlastic Material definition.
$ Loads for Load Case : Default SPCADD 2 1 $ Initial Velocities of Load Set : initial_velocity TIC 1 1 3 10000. TIC 1 2 3 10000. TIC 1 3 3 10000. TIC 1 4 3 10000. TIC 1 5 3 10000. TIC 1 6 3 10000. .. .. TIC 1 368 3 10000. TIC 1 369 3 10000. $ Displacement Constraints of Load Set : fix_plate SPC1 1 123456 370 THRU 418
$ Deform $ Deform BCBODY BSURF Body Contact LBC set: contact_mid Body Contact LBC set: plate 3 3D DEFORM 3 0 3 161 162 163 164 168 169 170 171 172
This section describes the initial velocity to the tapered beam and fixed boundary condition for the plate.
165 173
166 174
167 175
.. .. $ Deform Body Contact LBC set: beam BCBODY 4 3D DEFORM 4 BSURF 4 1 2 3 8 9 10 11
This section describes the contact bodies of the tapered beam and the plate.
0 4 12
5 13
6 14
7 15
Main Index
Chapter 3: Modeling 79
Input and Output Files Created During the Simulation
Input and Output Files Created During the Simulation
Simulation on Windows Platforms
The following files are created in your working directory: Input file: tapered_beam.bdf tapered_beam.dytr.dat tapered_beam.dytr.str Output file: tapered_beam.dytr.d3hsp tapered_beam.dytr.out tapered_beam.dytr_prep.d3hsp tapered_beam.dytr_prep.out tapered_beam.f06 Binary Result file: tapered_beam.dytr.d3plot tapered_beam.dytr.d3thdt Contains complete model used for plotting deformed shape and stress contour. Contains subset of the model used for time history plotting of element and grid point data. Contains model summary, calculation process, CPU time, etc. Contains time step summary and CPU timing information Contains translation summary of the preparation phase MD Nastran analysis output that contains model summary Input deck for MD Nastran Intermediate input files
Main Index
80 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Input and Output Files Created During the Simulation
Examine the Output file tapered_beam.dytr.out for Errors and Debugging
1 t 0.0000E+00 dt 4.35E07 flush i/o buffers 1 t 0.0000E+00 dt 4.35E07 write d3plot file 239 t 9.9943E05 dt 4.11E07 write d3plot file 488 t 1.9964E04 dt 3.86E07 write d3plot file 740 t 2.9994E04 dt 3.86E07 write d3plot file : : 12088 t 4.7999E03 dt 3.85E07 write d3plot file 12341 t 4.9000E03 dt 3.85E07 write d3plot file 12593 t 5.0000E03 dt 4.09E07 write d3plot file Warning: Airbag data not written to full restart file ******** termination time reached ******** 12593 t 5.0004E03 dt 4.09E07 write d3plot file N o r m a l : : C P U t e r m i n a t i o n
This section (*.dytr.out) describes increment (740) and time step (3.86E07) at this increment.
This section describes successful completion of the analysis. This section describes CPU time and Elasped Time of the job.
T i m i n g
i n f o r m a
t i o n
Processor Hostname CPU(seconds) # 1 RWCDYTRAN 9.4844E+00 T o t a l s 9.4844E+00
Start time 02/14/2005 15:15:31 End time 02/14/2005 15:15:40 Elapsed time 9 seconds ( N o r m a l
0 hours
0 minutes
9 seconds)
t e r m i n a t i o n
Main Index
Chapter 3: Modeling 81
Postprocessing
Postprocessing
The MD Nastran output the results in the binary native file d3plot and d3thdt for XY plots. The MD Nastran native output files *.op2 and *xdb will be supported in the future releases. Read in the results by attaching the d3plot file where the analysis results are stored. a. Analysis: Action: Object: Access Results Attach d3plot d. Click OK e. Click Apply
Method: Result Entries b. Click Select Results File c. Click tapered_beam_dytr.d3plot
a
c
d
b e
Main Index
82 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Postprocessing
Let’s have a Quick Plot of the Deformed Shape of the last cycle. a. Results: Action: Create Object: Quick Plot b. Select Results Cases Click ...:Time 0.005 c. Select Fringe Result Click Stress Components d. Position ...((NONLAYERED)) Quantity: von Mises h. i. f. g. e. Select Deformation Result Select Displacement Click on Deform Attributes Scale Interpretation Click True Scale Deactivate Show Undeformed Click Apply
a f
b
g
c d e
h
i
Main Index
Chapter 3: Modeling 83
Postprocessing
a. Results: Action: Object: Create Deformation
f.
Scale Interpretation Click True Scale Deactivate Show Undeformed Click on Animation Options Animation Graphics Click 3D
g. h. i.
b. Select Results Cases Select all cases c. Select Deformation Result Select Displacement Components d. Click Animate e. Click Display Attributes
j. k.
Number of Frames 52 Click Apply
a e
h
b
c
g
f
d
Main Index
84 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Postprocessing
a. Results: Action: Object: Create Fringe
c. Position ...((NONLAYERED)) Quantity: von Mises d. Enable Animate e. Click Apply
b. Select Fringe Result Stress Components
a
b c d e
Main Index
Chapter 3: Modeling 85
Running a Batch Job
Running a Batch Job
After the generation of the input file is complete, it is submitted for execution as a batch process. Once the input file has been submitted, you have no additional interaction with MD Nastran until the job is complete except that you can terminate the job prior to completion if it becomes necessary and monitor several keys files such as: .f06, .out, etc. MD Nastran is executed with a command called nastran. (Your system manager may assign a different name to the command.) The nastran command permits the specification of keywords used to request options affecting MD Nastran job execution. The format of the nastran command is: nastran input_data_file [keyword1 = value1 keyword2 = value2 ...] where input_data_file is the name of the file containing the input data and keywordi=valuei is one or more optional keyword assignment arguments. For example, to run an a job using the data file example1.dat, enter the following command: nastran example1 See “The nastran Command” in MD Nastran Quick Reference Guide. The details of submitting an MD Nastran job are specific to your computer system — contact your computer system administrator or your MD Nastran Installation and Operations Guide for further information.
How to Tell When the Analysis is Done
If you submit the job from the MD Nastran icon (i.e., outside MD Patran), as long as the parent window the job was run from is active, the analysis is still running. If you submit the job from within MD Patran and use stdout when you execute MD Patran, you can look in the MD Patran parent window and it will tell you when it submits the Nastran job, and also when the Nastran job is completed. Of course, you can always use the Analysis Manager. Once the job is complete look in the parent window to see what files were generated.
How to Tell if the Analysis Ran Successfully
Look in the working directory and you will see the typical jobname.f06, jobname.out. If these files are there, you successfully submitted the Nastran job. If you submitted a job with SOL 700, there will also be some jobname.dytr.xxx files in the subdirectory. To see if the run was successful, open jobname.dytr.out and search for errors. If you see “Normal termination” at the end of the file, it means the run was successful. It is always a good practice to monitor the time step of the calculation. If you noticed that your job has taken a long time to be completed, check the time step and if it is too small (E08 to E09 seconds or smaller), you should investigate the model to determine the causes of the small time step. Post processing
Main Index
86 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
How to Tell if the Analysis Ran Successfully
the results can give you clues as to what is happening to the model during the simulation and which elements are controlling the timestep. See Chapter 4: Special Modeling Techniques to control small timesteps.
Running a Parallel Job
When Not Connected to the Network If you disconnect a PC system from the network and want to run a parallel job on that system, you will have to install the Microsoft Loopback Adapter. Follow these steps: Go to Control Panel, Add/Remove Hardware. Select the hardware task you want to perform: Add/Troubleshoot a device Choose a Hardware Device: Add a new device Do you want Microsoft Windows to search for your new hardware? No, I want to select the hardware from a list Select the type of hardware you want to install: Network adapters Select Network Adapter: Manufacturers: Microsoft Network Adapter: Microsoft Loopback Adapter It will now install the loopback adapter. You will have to enable/disable the loopback adapter as you remove/connect your machine to the network. On Windows XP System When Not a Member of a Domain If you will be running a parallel job on a Windows XP system that is not a member of a domain, you will have to modify a registry entry. Using regedt32, look for the following key: HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Lsa "forceguest" : REG_DWORD : 00000001 If you find this key, change the REG_DWORD value to 0. The name may also appear as ForceGuest. If you do not have this registry entry, your system will function properly.
Main Index
Chapter 3: Modeling 87
How to Tell if the Analysis Ran Successfully
On Windows XP SP2 After you install or upgrade to Windows XP SP2, the RPC protocol does not permit anonymous requests to the RPC Endpoint Mapper but requires client requests be authenticated. To workaround this problem, do the following: From a command prompt, run gpedit.msc. Select Computer Configuration, expand Administrative Templates, expand System, click Remote Procedure Call, double click RPC Endpoint Mapper Client Authentication. Change the value to Enabled.
Main Index
88 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
How to Tell if the Analysis Ran Successfully
Main Index
Chapter 4: Special Modeling Techniques MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
4
Special Modeling Techniques
Artificial Viscosity Hourglass Damping Mass Scaling 95 97 107 90 91
Time Domain NVH
Prestress (Implicit to Explicit Sequential Simulation)
Main Index
90 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Artificial Viscosity
Artificial Viscosity
Two types of artificial viscosity are used in SOL 700; bulk viscosity and hourglass viscosity are used to control the numerical process. The parameters for bulk viscosity are material parameters. The hourglass viscosity parameters are defined per property.
Bulk Viscosity
Artificial bulk viscosity is used to control the formation of shock waves. Shock waves are the propagation of discontinuities in velocity. The simplest example of a shock wave is a “square wave.” An ideal impact between two flat surfaces generates a square wave. Materials that stiffen upon deformation can produce a shock wave from a smooth wave profile. A finite element model of a continuous body cannot numerically represent this propagating discontinuity. When a time integration scheme without algorithmic damping (such as the explicit central difference method) is used to integrate the response, severe oscillations in amplitude trail the shock front. These oscillations can be traced to the limitations imposed by the finite frequency spectrum of the finite element mesh. To control the oscillations trailing the shock front, artificial bulk viscosity is introduced. Artificial bulk viscosity is designed to increase the pressure in the shock front as a function of the strain rate. The effect on the shock wave is to keep it smeared over approximately five elements. Reducing the coefficients in an attempt to steepen the wave front may result in undesirable oscillations trailing the shock, a condition sometimes referred to as “overshoot.” An artificial viscosity term “ Q ” is added to the pressure. An artificial viscosity term can be considered as a modification to the pressure p , which is replaced by:
p* = p + Q
The definition of the artificial viscosity term that modifies the pressure:
· · · V 2 2 V  Q = max ⎛ – ρ ⋅ C Q ⋅ d ⋅  ⋅ V , 0⎞ + max ⎛ – ρ ⋅ C L ⋅ d ⋅ c ⋅  , 0⎞ ⎝ ⎠ ⎝ ⎠ V V V
where
CQ CL d
= constant = 1.5 = constant = 0.06 is the characteristic element dimension and
c
is the material speed of sound.
The following parameters can be used in the SOL 700 to define the linear and quadratic bulk viscosity coefficients:
(Linear coefficient, Default =0.06) PARAM,DYBULKQ1, value (Quadratic coefficient, default = 1.5)
PARAM,DYBULKL, value
See How to control Hourglassing in SOL 700.
Main Index
Chapter 4: Special Modeling Techniques 91
Hourglass Damping
Hourglass Damping
The solid and shell elements in SOL 700 have only one integration point at the center of the element. This makes the program very efficient since each element requires relatively little processing, but it also introduces the problem of hourglassing. For simplicity, consider the twodimensional membrane action of a CQUAD4 element.
The element has four grid points, each with two degrees of freedom. There are, therefore, a total of eight degrees of freedom and eight modes of deformation. There are three rigid body modes, two translational modes, and one rotational mode.
With a single integration point, two direct and one shear stress are calculated at the center of the element. This means that only three modes of deformation have stiffness associated with them. Two modes of deformation remain, that correspond to the linear stress terms. With a single integration point, these have no stiffness associated with them and are called the zero energy or hourglass modes.
When no measures are taken to stop these modes from occurring, they rapidly spread through the mesh and degrade the accuracy of the calculation (see Figure 41), reduce the time step, and ultimately cause the analysis to abort when the length of the side of an element becomes zero.
Main Index
92 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Hourglass Damping
Similar zero energy modes exist for the bending deformation of CQUAD4 elements, in CHEXA and CPENTA elements. CTRIA3 and CTETRA elements do not suffer from hourglassing, since no zero energy modes exist in these elements.
Figure 41
Deformation of a Mesh Showing Hourglassing
How to control Hourglassing in SOL 700 There are two methods available in SOL 700 to control the hourglassing. Viscous and Stiffness damping. The viscous form damps out hourglass modes and is carefully tuned so that other modes of deformation are not affected. The stiffness form applies forces to restrict the hourglass deformation by controlling the nonlinear part of the strain field that produces hourglassing. Normally the viscous forms work well, but, in some instances, are not adequate. The stiffness form is more effective but tends to make the elements overly stiff, depending on the input parameters selected. Each of the hourglass forms has slightly different characteristics. The default model is efficient and recommended for general use. The hourglass type can be specified using the PARAM,DYHRGIHQ. The hourglass coefficients can be specified using the PARAM options DYHRGQH, DYBULKQ1, and DYBULKQW. The hourglass type and the hourglass coefficients can be defined for each individual property using the HGSUPPR bulk data entry. Note that the value specified on the HGSUPPR overwrite the default values defined by using the PARAM options. Careful modeling can help prevent the occurrence of hourglassing in a mesh. Try to avoid sharp concentrations of load and isolated constraints. Rather, try to spread the loading and constraint over as large an area as possible. Some examples of how to avoid hourglassing are shown in Figure 42. In the majority of cases, hourglassing does not cause any problem. In those instances where it does begin to occur, adjustment of the type of hourglass control and the hourglass viscosity should allow the analysis to be completed successfully. Extreme cases of hourglassing are normally caused by coarse meshes. The only solution is to refine the mesh. Increasing the hourglass coefficient helps prevent hourglassing. However, excessively large values can cause numerical problems. Start with the default value and only increase it if excessive hourglassing occurs.
Main Index
Chapter 4: Special Modeling Techniques 93
Hourglass Damping
Figure 42
Hourglass Promotion and Avoidance
The following parameters are used in MD Nastran SOL 700 to define viscous and stiffness methods of hourglassing: (Integer > 0, Default=1) Default hourglass viscosity type = 1: standard (default) = 2: FlanaganBelytschko integration = 3: FlanaganBelytschko with exact volume integration = 4: stiffness form of type 2 (FlanaganBelytschko) = 5: stiffness form of type 3 (FlanaganBelytschko) = 6: BelytschkoBindeman assumed strain corotational stiffness form for solid elements only. (Real >= 0.0)
PARAM,DYHRGIHQ
PARAM,DYHRGQH
Where QH is the default hourglass coefficient. Values of QH that exceed 0.15 may cause instabilities. The recommended default applies to all options except for BelytschkoBindeman (type 6) of hourglass control. The stiffness forms, however, can stiffen the response especially if deformations are large and therefore should be used with care. For the shell and membrane elements QH is taken as the membrane
Main Index
94 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Hourglass Damping
hourglass coefficient, the bending as QB, and warping as QW. These coefficients can be specified independently but in SOL 700 the default is set at QH=QB=QW which is adequate (see PARAM,DYBULKL and PARAM,DYBULKQ1 below). For type 6 solid element hourglass control, see note 4 below. Below are some guidelines to use the different formulations for the hourglass viscosity: 1. Viscous hourglass control is recommended for problems deforming with high velocities. Stiffness control is often preferable for lower velocities, especially if the number of time steps are large. For solid elements the exact volume integration provides some advantage for highly distorted elements. 2. For automotive crash, the stiffness form of the hourglass control with a coefficient of 0.05 is preferred by many users. 3. Bulk viscosity is necessary to propagate shock waves in solid materials and therefore applies only to solid elements. Generally, the default values are okay except in problems where pressures are very high, larger values may be desirable. In low density foams, it may be necessary to reduce the viscosity values since the viscous stress can be significant. It is not advisable to reduce it by more than an order of magnitude. 4. Type 6 hourglass control is for 2D and 3D solid elements only. Based on elastic constants and an assumed strain field, it produces accurate coarse mesh bending results for elastic material when hourglass coefficient QH=1.0. For plasticity models with a yield stress tangent modulus that is much smaller than the elastic modulus, a smaller value of QH (0.001 to 0.1) may produce better results. For any material, keep in mind that the stiffness is based on the elastic constants, so if the material softens, a QH value smaller than 1.0 may work better. 5. In part, the computational efficiency of the BelytschkoLinTsay and the under integrated HughesLiu shell elements are derived from their use of onepoint quadrature in the plane of the element. To suppress the hourglass deformation modes that accompany onepoint quadrature, hourglass viscous or stiffness based stresses are added to the physical stresses at the local element level.
PARAM,DYBULKL, value (Linear coefficient CL, Default = 0.06)
The linear bulk viscosity coefficient CL. The defaults are: CL=QB=QH. See remark 4 above.
PARAM,DYBULKQ1, value (Quadratic coefficient, default = 1.5) PARAM,DYBULKQW,value (Hourglass coefficient for warping). The default is QW=QB
Main Index
Chapter 4: Special Modeling Techniques 95
Mass Scaling
Mass Scaling
The explicit dynamics procedure of SOL 700 uses relatively small time steps dictated by the shortest natural period of the mesh: the analysis cost is in direct proportion to the size of the mesh. There are two types of problems where the cost effectiveness of the analysis can be increased:
• If a mesh consists of a few, very small (or stiff) elements, the smallest (or stiffest) element
determines the time step for all elements of the mesh.
• If a few severely distorted elements are obtained by the analysis, the most distorted element
determines the time step for all elements of the mesh. This may result in a very small stable time step. Speedup of those problems can be achieved by using mass scaling. Mass scaling is based on adding numerical mass to an element so that its time step never becomes less than the minimum allowable time step defined by the user. One can also employ mass scaling in a selective manner by artificially increasing material density of the parts needed for massscaling. This manual form of mass scaling is done independently of the automatic mass scaling invoked with PARAM,DYTSTEPDT2MS. Note that mass scaling can be risky in areas where either inertia effects are relevant or contact with other parts is expected to occur.
Problems Involving a Few Small Elements
It is common practice that meshing of reallife problems may involve some relatively small elements: elements frequently localized in transition region and meant to connect large structural parts to each other. Those elements determine the time step of the whole calculation although they might be present in the model to a very limited extent. Speedup can be realized by using mass scaling.
• Make a run for one cycle and retrieve the time step of all elements. • By using a postprocessing program, view which elements are determining the time step and filter
out the elements whose time steps exceed a userdefined minimum time step.
• See what the impact would be of specifying this new minimum time step. Select the value of
minimum time step such that hardly any elements would be scaled in the area of interest (for example, as much as possible outside the impact region in a crash simulation)
Problems Involving a Few Severely Distorted Elements
There are many applications where elements are distorted to such a high extent that a few of them determine the time step for all elements of the mesh. For example, crushing of a subfloor structure frequently involves failure modes associated with the occurrence of severely distorted elements. Modeling this kind of crushing behavior without including a failure mechanism result in a stable time step that is too small. Since those elements are often present in a relatively small region, the mass scaling method might be a good means to artificially speed up the calculation without losing the capability to model the global crushing behavior. Note that to prevent severely distorted elements, it is recommended that a proper failure mechanism be included, instead of coping with the distorted elements by making use of the mass scaling method. Some guidelines:
Main Index
96 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Mass Scaling
• Since you do not know in advance which elements will become too distorted, you should first
run the analysis as far as possible without mass scaling. You should gauge and monitor the time step of all elements.
• If the problem ends up with a too small stable time step, the analysis finishes prematurely. See
which elements are so severely distorted and decide what a reasonable minimum time step might be without affecting elements in the area of interest. See the guidelines of the previous section.
• Rerun the analysis specifying mass scaling if the region of highly distorted elements is relatively
small compared to the whole model.
• If there is too much mass added to the grid points of those elements, the model might show
significantly different inertia effects, and subsequently, different global structural response. In order to avoid this, no more mass is added if the numerically added mass exceeds a certain percentage (PARAM, DYTERMNENDMAS). How to define Mass Scaling in SOL 700 The following PARAM entries can be used to define mass scaling in SOL 700:
PARAM,DYTSTEPDT2MS,value (Default=0.0,
time step size for mass scaled solution) PARAM,DYSTEPFCTL,value (Default=0.9, scale factor for internally calculated time step) PARAM,DYTERMNENDMAS,value (Default=1.0e20, Percent change in total mass for termination of calculation).
Anytime you add nonphysical mass to increase the time step in a dynamic analysis, you affect the results (think of F = ma). Sometimes the effect is insignificant and in those cases adding nonphysical mass is justifiable. Examples of such cases may include the addition of mass to just a few small elements in a noncritical area or quasistatic simulations where the velocity is low and the kinetic energy is very small relative to the peak internal energy. In the end, it's up to the judgement of the analyst to gauge the affect of mass scaling. You may have to reduce or eliminate mass scaling in a second run to gauge the sensitivity of the results to the amount of mass added. When the value of DT2MS in PARAM,DYTSTEPDT2MS is input as a negative value, mass is added only to those elements whose time step would otherwise be less than Fact * DT2MS . The value of Fact is specified in PARAM,DYSTEPFCTL. By adding mass to these elements, their time step becomes equal to Fact *DT2MS. An infinite number of combinations of Fact and DT2MS will give the same product; i.e., time step but the added mass will be different for each of those combinations. The trend is that the smaller the time step factor Fact, the greater the added mass. In return, stability may improve as time step factor Fact is reduced (just as in nonmassscaled solutions). If stability is a problem with the default time step factor of 0.9, try 0.8 or 0.7. If you reduce Fact, you can increase DT2MS proportionally so that the product/time step is unchanged. The difference between using a positive or negative number for DT2MS is: Negative: Mass is added to only those elements whose time step would otherwise be less than Fact*abs(DT2MS). When mass scaling is appropriate, this is the recommended approach. Positive: Mass is added or taken away from elements so that the time step of every element is the same. There is no clear advantage in using this method over the negative DT2MS method.
Main Index
Chapter 4: Special Modeling Techniques 97
Time Domain NVH
Time Domain NVH
This is a new methodology in MD Nastran SOL 700 that is used to compute the natural frequencies of a structure similar to what is done in experimental modal identification using impact or other transient testing. The user applies an impact, simulates driving a vehicle over a rough road or some other type of transient loading to the structure such that it is likely to excite all the important modes. A standard SOL 700 analysis is run to output accelerations, velocities and/or displacements at selected grids using a reasonably fine output delta time. After the nonlinear analysis finishes, a postprocessing operation is used to transform the selected outputs from the time domain to the frequency domain using Fast Fourier Transforms. Various criteria are available to select which peaks will be approximately defined as modes (eigenvectors). This approximation is reasonably accurate if the damping is small. The amplitudes of all the selected degrees of freedom for each of the selected modal frequencies are normalized and become the eigenvectors and saved on a file. A "restart" capability is available to change the criteria selections. In addition, a user may pick modes from plots made from data saved during the first run and then compute the eigenvectors associated with the chosen natural frequencies. The advantages of Time Domain NVH include the ability to consider the material and structural nonlinearity, damping and contact interface between various components. In other words, instead of utilizing a linear implicit method, an explicit approach is under taken to first, predict the nonlinear behavior of the structure, and then compute the NVH characteristics by FFT method. The double precision version of MD Nastran Nonlinear Explicit (SOL 700) is used to ensure higher fidelity of the solution. The disadvantage of time domain NVH technique is the excessive CPU time since an analysis may be required to run for a few seconds, considered a long runtime for an explicit simulation, before an FFT is performed to compute frequencies and mode shapes. However, this problem is becoming less significant due to the advanced Distributed Memory Parallel (DMP) technology in SOL 700 and dramatic hardware performance improvements.
Time Domain Data
Frequency Domain Data
FFT Process
Figure 43
Time Domain NVH of vehicle running on proving grounds (courtesy of ETA)
Main Index
98 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Time Domain NVH
To define a time domain NVH simulation, refer to the new entries of TIMNVH, TIMSML, and TIMNAT in the QRG and see the chassis example in SOL 700 User Guide. A simple example is listed below.
Time Domain NVH Example – Plate Subjected to a Pulse Loading
SOL 700 entries included: SOL 700 TSTEPNL PARAM,S700NVH1 TIMNVH TIMNAT Description: This is an example of a virtual dynamic test. A plate was modeled and a impulse loading was applied at one of corner points. Time histories were obtained at six points and they were translated by Fast Fourier Transformation (FFT) method to a frequency domain to obtain modal frequencies and shapes.
Model To build a model, a plate was constructed with 231 grid points and 200 quadratic elements. A fixed boundary condition was applied along one end. All shell elements were BelytschkoWongChiang formulation. The impact loading at the corner of the plate was modeled by defining the load time history as shown in Figure 44. The simulation time is 1.024 seconds.
Main Index
Chapter 4: Special Modeling Techniques 99
Time Domain NVH
Figure 44
Impluse Load applied at One Corner of the Plate
Time NVH Scheme
MD Nastran SOL 700 (impluse loading) Time history  Displacement  Velocity  Acceleration (default)
Time domain > Frequency domain using FFT
Extract dynamic properties: Natural frequencies and Mode shapes
Input The loading was applied at node 231 by using FORCE and TABLE1 entries. Using TIMNVH, TIMNAT, and PARAM, S700NVH1, dynamic properties of the plate were obtained. There were three steps to acquire the dynamic properties as follows: Step 1: Find modal properties using TIMNVH entry in the first trial Step 2: Check the obtained modal properties and select required natural frequencies Step 3: Rerun with selected natural frequencies
Main Index
100 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Time Domain NVH
Input file timnvh.bdf SOL 700,NLTRAN path=3 stop=1 SOL 700 is a executive control card similar to SOL 600. It activates an explicit nonlinear transient analysis. Case control section is below. CEND TITLE = MD Nastran job created on 12Dec06 at 11:21:25 LOADSET = 1 $ Direct Text Input for Global Case Control Data SUBCASE 1 TITLE=This is a default subcase. TSTEPNL = 1 SPC = 2 DLOAD = 2 DISPLACEMENT(SORT1,REAL)=ALL SPCFORCES(SORT1,REAL)=ALL Bulk entry section starts. BEGIN BULK TSTEPNL 1 100 .01024 1 TIMNVH,1,,,1.0,500.,3,0.00005,2,+ +,0,3,1,0.015,0,3,13,.0030,+ +,11,21,116,126,221,231
ADAPT
2
10
TSTEPNL is a SOL 700 bulk data entry which describes the number of Time Steps (100) and Time Increment (10.24 ms) of the simulation. End time is the product of the two entries. Notice here the Time Increment is only for the first step. The actual number of Time Increments and the exact value of the Time Steps is determined by SOL 700 during the analysis. The time step is a function of the smallest element dimension during the simulation. TIMNVH entry is for Time NVH analysis.
Detail for TIMNVH entry: TIMNVH, 1, , , 1.0, 500., 3, 0.00005, 2,+
The range of natural frequencies to obtain is from 1.0 Hz to 500 Hz and only ztranslational degrees of freedom is considered (3). The sampling rate is 0.00005 seconds. The peaking criterion is two, which means a peak is selected if the number of increasing and decreasing amplitude around a peak is over 2. +, 0, 3, 1, 0.015, 0, 3, 13, .0030,+
Acceleration is selected for the response (0) and translational eigenvectors are only requested as ASCII format (3). Eignevalues are normalized by 1.0 (1) and 0.015 is selected as CLOSE value which means if there are two modes which distance is smaller than 0.015 Hz, it is assumed to be the same mode. ACII
Main Index
Chapter 4: Special Modeling Techniques 101
Time Domain NVH
file format of natural frequencies and eigenvalues are asked (0) and translational time histories of zdirection are requested (3). Frequencyamplitude data of zdirection are asked (13) and a peak which amplitude is less than 0.0030×the maximum amplitude is ignored (.0030) +,11,21,116,126,221,231 The grid points 11, 21, 116, 126, 221 and 231 are only considered for Time Domain NVH analysis. Bulk data entry that defines grid points and elements. CQUAD4 1 1 1 .. $ Nodes of the Entire Model GRID 1 0. .. Bulk data entry that defines boundary conditions. SPCADD 2 1 $ Displacement Constraints of Load Set : fixed SPC1 1 123456 1 22 43 64 127 148 169 190 211 Bulk data entry that defines material properties. $ Referenced Material Records $ Material Record : steel $ Description of Material : Date: 12Dec06 11:09:13 MAT1 is an isotropicelastic material card of SOL 700 bulk data entry. Bulk data entry that defines properties for shell elements with .001 thickenss. PSHELL 1 1 .001 1 1 2 0. 23 0. 22
85
106
Time:
Bulk data entry that defines applying forces. $ Loads for Load Case : Default TLOAD1 4 5 LSEQ 1 5 3 DLOAD 2 1. 1. 4 $ Nodal Forces of Load Set : impact FORCE 3 231 0 .01
1
0.
0.
1.
Main Index
102 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Time Domain NVH
Bulk data entry that defines tables. $ Dynamic Load Table : hammer TABLED1 1 10. 0. 0. 10. 0. ENDT .. ENDDATA Input file timnvh1.bdf This input is for the refinement of the selection of modal frequencies and mode shapes. Only different part is shown PARAM,S700NVH1,1 TSTEPNL 1 100 .01024 1 TIMNVH,1,,,1.0,500.,3,0.00005,2,+ +,0,3,1,0.015,0,3,13,.0030,+ +,11,21,116,126,221,231 TIMNAT,1,20.5,88.,129.,285.,360.
0.
.001
1.
.002
0.
ADAPT
2
10
PARAM, S700NVH is for the rerun of Time NVH analysis. Using 1 as option Time Domain NVH
analysis is carried out without rerunning SOL 700. The PEAK value (original: 2) in TIMNVH card is changed to 2 to use TIMNAT card.
TIMNAT card is for the control of the natural frequency selection. In this job, 20.5, 88, 129, 285, 360 Hz are selected to get the results.
Results There are three types of new results file from Time Domain NVH analysis. 1. mode.out: the natural frequencies and eigenvalues selected are restored. 2. amplfreq000001163.txt : amplitudefrequency output of DOF =3 at grid point 116. 3. timehist000001163.txt: time history output of DOF =3 at grid point 116. From the amplfreq*** files, the frequencyamplitude plots are obtained.
Main Index
Chapter 4: Special Modeling Techniques 103
Time Domain NVH
1.00E+04
3rd mode ≈ 129.
node 11
2
nd
mode ≈ 88.
node 21 node 116 node 126 node 221 node 231
1.00E+03
4th mode ≈ 285.
Amplitude (acceleration)
1.00E+02
1.00E+01
1.00E+00
1.00E01
1.00E02 0 100 200 Frequency (Hz) 300 400 500
Main Index
104 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Time Domain NVH
In mode.out file,
Total Number of Modes
MODES EIGV
1st Mode
EIGV
1 18 Natural Frequency 1 2.050821E+01 119.13892833E06 3.86241316E06 3.19331321E01 21 2.14690732E051.14613790E05 9.95118635E01 1169.93770531E06 2.13863912E06 3.30875181E01 126 7.27358928E062.94809676E06 1.00000000E+00 2211.23613298E053.06943202E06 3.20993777E01 231 1.35196599E05 6.87379272E06 9.99026322E01 2 8.789234E+01 115.69607103E06 1.65922185E066.09298304E01 21 5.33656071E067.67633319E069.90453563E01 1163.47585317E06 8.00428604E07 1.97687095E02 126 5.34868995E063.96739670E067.95414322E03 2214.07797686E062.98546031E06 5.90684541E01 231 3.21720171E06 7.30456344E06 1.00000000E+00
Node Number
Mode Shape of xdirection
Mode Shape of ydirection Mode Shape of zdirection
Comparison of natural frequencies between SOL 103 and SOL 700 (Hz) Mode 1 2 3 4 5 SOL103 2.0508E+01 8.7894E+01 1.2891E+02 2.8516E+02 3.6036E+02 SOL 700 2.0864E+01 8.8639E+01 1.2966E+02 2.8771E+02 3.6269E+02 Diff(%) 1.74 0.85 0.58 0.89 0.65
Comparison of mode shapes between SOL 103 and SOL 700
Main Index
Chapter 4: Special Modeling Techniques 105
Time Domain NVH
Mode 1
SOL 103
SOL 700
2
3
Main Index
106 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Time Domain NVH
Mode 4
SOL 103
SOL 700
5
6 points sampling
12 points sampling
Main Index
Chapter 4: Special Modeling Techniques 107
Prestress (Implicit to Explicit Sequential Simulation)
Prestress (Implicit to Explicit Sequential Simulation)
Many applications require a prestress analysis prior to transient analysis. For example in bird strike and blade out analysis, the blades of a running jet engine have residual stresses due to high rotational velocities. SOL 700 uses the implicit solver for prestress analysis to reduce the CPU time. Prestress calculations are performed by MD Nastran Nonlinear Explicit (SOL 700) implicit double precision version where initial state of the model is written into a file that can only be used for subsequent SOL 700 transient run. This is an automated process and there is no need for dynamic relaxation to eliminate the high frequency oscillations. The following entries may be used to define a prestressed state. PRESTRS ISTRSBE ISTRSSH ISTRSTS ISTRSSO Perform prestress run to calculate an initially stressed model and write out the initial state to a file that can be used for a subsequent explicit SOL 700 run Initialize stresses and plastic strains in the HughesLiu beam elements Initialize stresses, history variables and the effective plastic strain for shell elements Initialize stresses, history variables and the effective plastic strain for thick shell elements. Initialize stresses and plastic strains for solid elements
Caution must be exercised to use the appropriate element formulation to get consistent results during prestress analysis. Implicit and explicit element formulations might differ from each other resulting in different displacement or stress values. For this purpose, the following study is done to identify the best element formulations in MD Nastran Nonlinear Explicit (SOL 700) implicit solver that match closest to those in MD Nastran. It is therefore recommended that the users follow the guideline below for prestress simulation.
Comparative Study between Nastran and MD Nastran SOL 700 Implicit Element Formulations
Model description: Two simple models were constructed for the comparative study. The first model is a 1 x 1 square plate with a thickness of 0.01 and isotropic material. The element sizes are 0.2 x 0.2. The plate model was constructed using the standard MD Nastran CQUAD4, CTRIA3 shells and then were compared to MD Nastran Nonlinear Explicit (SOL 700) shell formulations under different loading conditions. Similarly, the second model is a 1 x 1 x 1 solid cube constructed with Nastran solid elements CHEXA, CPENTA, and CTETRA as well as MD Nastran Nonlinear Explicit (SOL 700) different solid element formulations. The size of elements are 0.2x0.2x0.2 with isotropic material. The solid model was subjected to different loading condition to determine the consistency of the results between MD Nastran and MD Nastran Nonlinear Explicit (SOL 700). Two sets of analyses were done to compare the results in linear and nonlinear domain. For linear analysis, only partial loading was used to remain in elastic range whereas the loading was ramped up beyond the yield point to compare the nonlinear behavior of the elements.
Main Index
108 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Prestress (Implicit to Explicit Sequential Simulation)
Model Material: Isotropic & elastic (E=1000., v=0.3 and density=1.) 1. Plate Model: Size: 1x1 Thickness: 0.01 Element size: 0.2x0.2 a. Boundary Condition – Fixed at all edges
b. loading condition: case 1. Z Translational Loading at the every grid point: 1.E2 case 2. XRotational Loading at the every grid point: 1.E2 case 3. Pressure at the top surface 1E2
Main Index
Chapter 4: Special Modeling Techniques 109
Prestress (Implicit to Explicit Sequential Simulation)
Case 1
Case 2
Case 3
c. CTRIA Loading and boundary conditions are the same as CQUAD4
d. CSHEAR Not tested yet. 2. Solid Model Cube Size: 1x1x1 a. CHEXA Element size: 0.2x0.2x0.2
Main Index
110 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Prestress (Implicit to Explicit Sequential Simulation)
Boundary Condition: Fixed at the bottom surface.
b. loading condition case 1. Z Translational Loading on the every grid point at the top: 1.E2 case 2. Pressure at the top surface 1E2
c. CPENTA The same loading and boundary conditions as the Hexa case
d. TETRA The same loading and boundary conditions as the Hexa case
Main Index
Chapter 4: Special Modeling Techniques 111
Prestress (Implicit to Explicit Sequential Simulation)
LSDYNA Model: The following entries were used for the linear analysis of the shells: *CONTROL_SHELL ,1 *CONTROL_IMPLICIT_AUTO 0 *CONTROL_IMPLICIT_GENERAL 1,0.1,,0,0,0,0 *CONTROL_IMPLICIT_SOLUTION 2,5,,0.01,0.01 *CONTROL_IMPLICIT_SOLVER , , , ,1,100. *CONTROL_TERMINATION 1.0000 Similarly for solid model: *CONTROL_SOLID 1,2 *CONTROL_IMPLICIT_AUTO 0 *CONTROL_IMPLICIT_GENERAL 1,0.1,,0,0,0,0 *CONTROL_IMPLICIT_SOLUTION 2,5,,0.01,0.01 *CONTROL_IMPLICIT_SOLVER , , , ,1,100. LSDYNA Element formulations used in the study: CQUAD4 : ELFORM= 1,2,6,10,16,18,20 CTRIA: ELFORM= 1,2,6,10,16,17,18,20,21 HEXA: ELFORM= 1,2,6,18 TETRA: ELFORM= 1,2,4,10,16,18 PENTA: ELFORM= 1,2,15,18 Results: The results for the linear and nonlinear static analyses are shown in Table 41. Even though many elements showed close results (within 10% differences), several cases had a large variance of more than 10%. These are highlighted in red.
Main Index
112 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Prestress (Implicit to Explicit Sequential Simulation)
Table 41
Nastran
Comparative Results Between MD Nastran and LSDYNA Implicit Elements
LSDYNA LOADING FORCE PRESSURE MOMENT FORCE FORCE FORCE FORCE FORCE FORCE FORCE PRESSURE MOMENT FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE FORCE PRESSURE FORCE FORCE FORCE FORCE PRESSURE FORCE FORCE FORCE FORCE FORCE FORCE PRESSURE FORCE FORCE FORCE DYNA (linear) DYNA (nonlinear) 1.008E+01 1.008E+01 4.031E01 4.03E01 3.559E+00 3.559E+00 1.019E+01 8.030E02 1.019E+01 8.030E02 1.002E+01 8.300E02 8.620E+00 8.539E02 1.003E+01 8.411E02 1.084E+01 7.781E02 1.040E+01 7.369E02 4.147E01 2.546E02 3.557E+00 Not converged 3.819E+00 7.382E02 3.819E+00 7.382E02 3.819E+00 7.382E02 3.819E+00 7.382E02 3.819E+00 7.382E02 9.778E+00 7.397E02 1.040E+01 7.369E02 1.040E+01 7.369E02 6.416E02 6.416E02 9.952E04 9.958E04 9.356E02 9.132E02 6.577E02 6.734E02 1.016E01 1.243E01 6.047E02 6.179E02 9.928E04 9.921E04 6.047E02 6.179E02 6.047E02 6.179E02 4.021E02 9.077E02 6.047E02 6.179E02 6.047E02 6.179E02 7.200E02 7.191E02 9.978E04 9.970E04 7.200E02 7.191E02 7.200E02 7.191E02 7.200E02 7.191E02 Output Displacement SOL101 SOL106 %Diff (linear and SOL101) %Diff (n 1.040E+01 8.400E02 3.08 4.140E01 2.680E02 2.63 3.400E+00 Not converged 4.68 1.040E+01 8.400E02 2.02 1.040E+01 8.400E02 2.02 1.040E+01 8.400E02 3.65 1.040E+01 8.400E02 17.12 1.040E+01 8.400E02 3.56 1.040E+01 8.400E02 4.23 9.660E+00 7.870E02 7.66 3.860E01 2.680E02 7.44 3.470E+00 Not converged 2.51 9.660E+00 7.870E02 60.47 9.660E+00 7.870E02 60.47 9.660E+00 7.870E02 60.47 9.660E+00 7.870E02 60.47 9.660E+00 7.870E02 60.47 9.660E+00 7.870E02 1.22 9.660E+00 7.870E02 7.66 9.660E+00 7.870E02 7.66 6.500E02 6.290E02 1.29 9.950E04 9.960E04 0.02 6.500E02 6.290E02 43.94 6.500E02 6.290E02 1.18 6.500E02 6.290E02 56.31 6.050E02 5.890E02 0.05 9.930E04 9.930E04 0.02 6.050E02 5.890E02 0.05 6.050E02 5.890E02 0.05 6.050E02 5.890E02 33.54 6.050E02 5.890E02 0.05 6.050E02 5.890E02 0.05 7.060E02 6.820E02 1.98 9.960E04 9.970E04 0.18 7.060E02 6.820E02 1.98 7.060E02 6.820E02 1.98 7.060E02 6.820E02 1.98
SECTION_SHELL with ELFORM=18 SECTION_SHELL with ELFORM=1 CQUAD4 SECTION_SHELL with ELFORM=2 SECTION_SHELL with ELFORM=6 SECTION_SHELL with ELFORM=10 SECTION_SHELL with ELFORM=16 SECTION_SHELL with ELFORM=20 SHELL SECTION_SHELL with ELFORM=18 SECTION_SHELL with ELFORM=1 SECTION_SHELL with ELFORM=2 SECTION_SHELL with ELFORM=6 SECTION_SHELL with ELFORM=10 SECTION_SHELL with ELFORM=16 SECTION_SHELL with ELFORM=17 SECTION_SHELL with ELFORM=20 SECTION_SHELL with ELFORM=21 SECTION_SOLID with ELFORM=18 CHEXA SECTION_SOLID with ELFORM=1 SECTION_SOLID with ELFORM=2 SECTION_SOLID with ELFORM=3 SECTION_SOLID with ELFORM=18 SECTION_SOLID with ELFORM=1 SOLID CTETRA SECTION_SOLID with ELFORM=2 SECTION_SOLID with ELFORM=4 SECTION_SOLID with ELFORM=10 SECTION_SOLID with ELFORM=16 SECTION_SOLID with ELFORM=18 CPENTA SECTION_SOLID with ELFORM=1 SECTION_SOLID with ELFORM=2 SECTION_SOLID with ELFORM=15
CTRIA
Description of LSDYNA element formulations: SECTION_SHELL ELFORM=1: HughesLiu, ELFORM=2: BelytschkoTsay (default), ELFORM=6: S/R HughesLiu, ELFORM=10: BelytschkoWongChiang, ELFORM=16: Fully integrated shell element, ELFORM=17: Fully integrated DKT, triangular shell element, ELFORM=18: Taylor 4node quadrilateral and 3node triangle (linear only) ELFORM=20: Wilson 3 & 4node DSE quadrilateral (linear only) ELFORM=21: Fully integrated linear assumed strain C0 shell (5 DOF). ELFORM=22: Linear shear panel element (3 DOF per node) SECTION_SOLID ELFORM=1: constant stress solid element, ELFORM=2: fully integrated S/R solid. See remark 5 below, ELFORM=3: fully integrated 8 node solid with rotational DOFs,
Main Index
Chapter 4: Special Modeling Techniques 113
Prestress (Implicit to Explicit Sequential Simulation)
ELFORM=4: fully integrated S/R 4 node tetrahedron with rotational DOFs, ELFORM=10: 1 point tetrahedron. ELFORM=15: 2 point pentahedron element. ELFORM=16: 5 point 10 noded tetrahedron ELFORM=18: 8 point enhanced strain solid element for linear statics only,
Main Index
114 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Prestress (Implicit to Explicit Sequential Simulation)
Main Index
Chapter 5: Constraints and Loadings MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
5
Constraints and Loadings
Constraint Definition Lagrangian Loading 116 120 123 130
Eulerian Loading and Constraints
Boundary and Loading Conditions  Theoretical Background
Main Index
116 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Constraint Definition
Constraint Definition
The motion of part or all of a mesh can be prescribed by application of constraints.
SinglePoint Constraints
A singlepoint constraint is used to prescribe the motion of a translational or rotational degree of freedom. The constraint is effective throughout the analysis and is used to specify boundary conditions or planes of symmetry. A singlepoint constraint is defined by an SPCn entry. The SPC entry defines the constraints on one grid point, while the SPC1 defines the constraints to be applied to a set of grid points. Several sets of SPC entries can be defined in the Bulk Data Section, but only those selected in the Case Control Section using the SPC = n command are incorporated in the analysis. Singlepoint constraints can also be defined using the GRIDt entry. These constraints are present for the entire analysis and do not need to be selected in Case Control. This is valid only for SPC and SPC1. Since Sol 700 is an explicit code, there is no matrix decomposition. Therefore, the problems of singular matrices that occur with some implicit codes do not exist. All, or part of the Lagrangian mesh can be entirely unconstrained and can undergo rigid body motion. SOL 700 correctly calculates the motion of the mesh. Similarly, the redundant degrees of freedom, such as the inplane rotation of shell elements, do not need to be constrained since they do not affect the solution. The only constraints that are needed are those representing the boundary conditions of the model and those necessary for any planes of symmetry. The following SPC types are supported in MD Nastran SOL 700.
• SPC • SPC1 • SPCADD • SPCD (at present, only velocity is available) • SPCD2
MultiPoint Constraints
MPCs are special element types which define a rigorous behavior between several specified nodes. The following MPC types which are supported for MD Nastran SOL 700:
• MPC • RBAR • RBE2 • RBE2A • RBE2D • RBE2F • RBE3
Main Index
Chapter 5: Constraints and Loadings 117
Constraint Definition
• RBE3D • RBJOINT • REJSTIFF • RCONN
Specifying Explicit MPCs
MPC’s may be created between a dependent degree of freedom and one or more independent degrees of freedom. The dependent term consists of a node ID and a degree of freedom; an independent term consists of a coefficient, a node ID, and a degree of freedom. An unlimited number of independent terms can be specified, but only one dependent term can be specified. The constant term is not allowed in MD Nastran. MPC Types
MPC
Description Defines a multipoint constraint equation.
References “MPC” in the MD Nastran Quick Reference Guide.
Contact in SOL 700
The detailed theory behind the implementation of contact in SOL 700 is discussed in Chapter 8: Contact Impact Algorithm. However, here we discuss the contact methodology and definition as implemented in MD Nastran Explicit Nonlinear. The simulation of many physical problems requires the ability to model the contact phenomena. This includes impact simulations, drop testing, component crush, crash, and manufacturing processes among others. The analysis of contact behavior is complex because of the requirement to accurately track the motion of multiple geometric bodies, and the motion due to the interaction of these bodies after contact occurs. This includes representing the friction between surfaces if required. The numerical objective is to detect the motion of the bodies, apply a constraint to avoid penetration, and apply appropriate boundary conditions to simulate the frictional behavior and heat transfer. Several procedures have been developed to treat these problems including the use of Perturbed or Augmented Lagrangian methods, penalty methods, and direct constraints. Furthermore, contact simulation has often required the use of special contact or gap elements. MD Nastran Explicit Nonlinear allows contact analysis to be performed automatically without the use of special contact elements. A robust numerical procedure to simulate these complex physical problems has been implemented in MD Nastran Explicit Nonlinear. Contact problems can be classified as one of the following types of contact.
• DeformableDeformable contact between two and threedimensional deformable bodies. • Rigid  Deformable contact between a deformable body and a rigid body, for two or
threedimensional cases.
Main Index
118 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Constraint Definition
Contact problems involve a variety of different geometric and kinematic situations. Some contact problems involve small relative sliding between the contacting surfaces, while others involve large sliding or penetration followed by perforation and failure. Some contact problems involve contact over large areas, while others involve contact between discrete points. The general Contact Body approach in SOL 600 is also adopted by MD Nastran Explicit Nonlinear (SOL 700) to model contact and can be used to handle most contact problem definitions.
Contact Bodies
There are two types of contact bodies in MD Nastran Explicit Nonlinear – deformable and rigid. Deformable bodies are simply a collection of finite elements as shown below.
Figure 51
Deformable Body
This body has three key aspects to it: 1. The elements which make up the body. 2. The nodes on the external surfaces which might contact another body or itself. These nodes are treated as potential contact nodes. 3. The edges (2D) or faces (3D) which describe the outer surface which a node on another body (or the same body) might contact. These edges/faces are treated as potential contact segments. Note that a body can be multiply connected (have holes in itself). It is also possible for a body to have a self contact where the entire surface folds on to itself. A contact may include 1D elements such as beams and rods, 2D elements such as shells and membranes, and 3D elements such as solids. Each element should be in, at most, one body. The elements in a body are defined using the BCBODY option. It is not necessary to identify the grid points on the exterior surfaces as this is done automatically. The algorithm used is based on the fact that grid points on the boundary are on element edges or faces that belong to only one element. Each node on the exterior surface is treated as a potential slave grid point. The second type of a contact body is called Rigid bodies. Rigid bodies are composed of (2D) or (3D) meshes and their most significant aspect is that they do not distort. Deformable bodies can contact rigid bodies and contact between rigid bodies is also allowed.
Main Index
Chapter 5: Constraints and Loadings 119
Constraint Definition
Rigid Walls
A rigid wall is a plane through which specified slave grid points cannot penetrate. The rigid wall provides a convenient way of defining rigid targets in impact analyses.
Any number of rigid walls can be specified using WALL entries. The orientation of each wall is defined by the coordinates of a point on the wall and a vector that is perpendicular to the wall and points towards the model. At each time step, a check is made to determine whether the slave grid points have penetrated the wall. These slave points are defined using a Case Control SET entry, and there can be any number of them. Since a check is made for every slave point at each time step, you should specify only those points as slave points that are expected to contact the wall in order to ensure the most efficient solution. If a slave point is found to have penetrated the wall, it is moved back towards the wall so that its momentum is conserved. If the slave point subsequently moves away from the wall, it is allowed to do so. Slave points cannot have any other constraint. They can, however, be part of other contact.
Contact Detection
During the incremental procedure, each potential contact node is first checked to see whether it is near a contact segment. The contact segments are either edges of other 2 D deformable bodies, faces of 3D deformable bodies, or segments from rigid bodies. By default, each node could contact any other segment including segments on the body that it belongs to. This allows a body to contact itself. To simplify the computation, it is possible to use the BCTABLE entry to indicate that a particular body will or will not contact another body. This is often used to indicate that a body will not contact itself. During the iteration process, the motion of the node is checked to see whether it has penetrated a surface by determining whether it has crossed a segment. For more detailed discussion on the Contact, refer to Chapter 8: Contact Impact Algorithm of this guide and see Chapter 12 in the SOL 600 User’s Guide.
Main Index
120 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Loading
Lagrangian Loading
This section covers the different ways that the analysis model can be loaded. The facilities available are:
• Concentrated Loads and Moments • FORCE, FORCE2, or DAREA – FixedDirection Concentrated Loads • Pressure Loads • Initial Conditions
Concentrated Loads and Moments
Concentrated loads and moments can be applied to any grid point using the DAREA, FORCE, FORCE2, entries in combination with a TLOADn entry. The types of concentrated load that can be applied are discussed in the following section.
FORCE, FORCE2, or DAREA – FixedDirection Concentrated Loads
The FORCE, FORCE2, and DAREA entries define fixed direction loads. In other words, the direction of the force is constant throughout the analysis and does not change as the structure moves.
FORCE, FORCE2, and DAREA entries have the same effect but define the loading in different ways. With the DAREA entry, you specify the grid point, the direction in the basic coordinate system in which the load acts, and the scale factor. With the FORCE entry, a concentrated force remains in the same direction for the entire problem. For FORCE2, the direction of the force follows the deformation and you define the grid point and the components of a vector giving the loading direction and the scale factor. In this case, the magnitude of the vector also acts as a scale factor, so the force in direction i is given by Pi = A Ni .
Main Index
Chapter 5: Constraints and Loadings 121
Lagrangian Loading
On a rigid body, the concentrated load or enforced motion is specified by defining the load at the rigid body center of gravity. To do so, set the TYPE field of the TLOAD1 or TLOAD2 entries to 13 and 12, respectively. The G field in the FORCE entry references the property number of the rigid body.
Pressure Loads
Pressure loads are applied to the faces of solid elements and to shell elements. Pressure loads are defined using the PLOAD or PLOAD4 entry in combination with a TLOADn entry.
TLOAD2 also references a set of PLOAD and/or PLOAD4 entries. Each entry selects the face of the element to be loaded by its grid points and defines the scale factor to be applied to the curve of pressure versus time. The actual pressure acting on the element p el is given as follows:
p el ( t ) = A p ( t )
where
A
is the scale factor.
The direction of positive pressure is calculated according to the righthand rule using the sequence of grid points on the PLOAD entry. For PLOAD4 entries, the pressure is inwards for solid elements and in the direction of the element normal vector for shell elements.
G3 G4 G3
G1 G2
G1 G2
The RFORCE entry defines enforced motion due to a centrifugal acceleration field. This motion affects all structural elements present in the problem. The GRAV defines an enforced motion due to a gravitational acceleration field. This motion affects all Lagrangian elements. Grid points with enforced motion cannot be:
• Attached to a rigid body. • A slave point for a rigid wall.
To specify the motion of a rigid body, the enforced motion of the rigidbody center of gravity must be defined. To do so, set the TYPE field of the TLOAD1 and TLOAD2 entries to 12. The G field on the DAREA, FORCE entry references the property number of the rigid body.
Main Index
122 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Loading
Initial Conditions
The initial velocity of grid points can be defined using TIC, TIC3, and TICD entries. This allows the initial state of the model to be set prior to running the analysis. It is important to recognize the difference between initial velocities and enforced velocities. Enforced velocities specify the motion of grid points throughout the transient analysis. Initial velocities, on the other hand, specify the velocity of grid points at the beginning of the analysis. Thereafter, the velocities are determined by the calculation. Where TIC and TIC3 set only the initial gridpoint velocity, the TICD entry can be used to set the initial value of any valid grid point variable. It can also refer to a local coordinate system by including the CID1 and/or CID2 entry in the list.
Main Index
Chapter 5: Constraints and Loadings 123
Eulerian Loading and Constraints
Eulerian Loading and Constraints
Loading Definition
The implementation of loading and constraints within Eulerian meshes is somewhat different than that in a Lagrangian mesh. Eulerian constraints apply to element faces within the mesh rather than to the grid points. The code allows you to set the initial conditions for material in Eulerian elements, constrain material with fixed barriers, apply gravitational body forces, apply pressure boundaries to element faces, apply flow boundaries where material enters or leaves the mesh, and couple the mesh so that the material interacts with the Lagrangian parts of the model. If an exterior or free face of an Eulerian mesh does not have a specific boundary condition, by default, it forms a barrier through which the material cannot flow. The default can be redefined by using a FLOWDEF entry.
Flow Boundary
A flow boundary defines the physical properties of material flowing in or out of Eulerian elements and the location of the flow. The FLOW entry is referenced by a TLOAD1. The TYPE field on the TLOAD1 must be set to 4. The FLOW entry references a set of segments, specified by BCSEG entries, through which the material flows. The subsequent fields allow you to specify the x, y, or z velocity, the pressure, and the density or specific internal energy of the flowing material. If only the pressure is defined, this gives a pressure boundary. Any variables not specified take the value in the element that the material that is flowing into or out of the flow.
Rigid Wall
The BARRIER entry defines a wall that is equivalent to a Lagrangian rigid wall. This is a barrier to material transport in an Eulerian mesh. The barrier is defined by a set of faces generated from a BCSEG entry through which no material can flow. This is the default condition for any exterior faces of the Eulerian mesh that do not have a FLOW boundary specified. However, the BARRIER entry can be used to specify rigid walls within an Eulerian mesh.
Main Index
124 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Eulerian Loading and Constraints
Eulerian Mesh
BARRIER Boundary
Initial Conditions
The initial conditions of Eulerian elements can be defined using the TICEL or TICEUL1 entry or by using the option “initeul=” in the sol700.pth file. This allows the initial state of the model to be set prior to running the analysis. It is important to recognize the difference between initial conditions and enforced conditions. Enforced conditions specify the loading and constraints of material throughout the transient analysis. Initial conditions, on the other hand, specify the state of the material only at the beginning of the analysis. Thereafter, the material state is determined by the calculation. TICEL The TICEL entry defines transient initial conditions for elements. Any valid element variable can be given an initial value. INITEUL The INITEUL entry imports an Euler archive into a followup run and maps it onto the defined Euler elements. If the followup run uses a coupling surface, then in the first run, this coupling surface can be left out provided that physics in the Euler have not reached the coupling surface at the end of the first run. In the followup run, a coarser mesh can be used to reduce CPU time. TICEUL1 The TICEUL1 entry defines transient initial conditions for geometrical regions in the Eulerian mesh using the TICREG entries. The TICEUL1 entry must be used together with the PEULER1 property definition.
Main Index
Chapter 5: Constraints and Loadings 125
Eulerian Loading and Constraints
With the TICREG entry, it is possible to generate initial conditions inside or outside multifaceted surfaces, analytical cylindrical or spherical geometry shapes and in sets of elements. Each geometrical region (multifaceted surface, cylinder, sphere, box or set of elements) has a level number. This allows the creation of regions of arbitrary shape by allowing the regions to overlap. An element that lies in two or more geometrical regions is assigned to the region that has the highest level number. Think of geometrical regions as shapes cut out of opaque paper. Position the region of the lowest level number on the mesh. Then, place the next higher region on top of the first and continue until all the regions are in place. When the last region is placed, you have a map indicating to which region each element in the problem is assigned. The following figure shows how three different geometrical regions can be used to create regions of arbitrary shape. The solid line represents the boundary of the mesh. Region one (LEVEL = 1) is the large dashed rectangle. Region two (LEVEL = 2) is the long narrow rectangle. Region three (LEVEL = 3) is a circular region. The numbers on the diagram indicate how the elements in different parts of the mesh are assigned to these three regions.
LEVEL = 1
1 2 1
3 3 3
1 2 1 LEVEL = 2
LEVEL = 3
If two or more regions with the same level number but different initial value sets or materials overlap, the regions are ambiguously defined resulting in an error. Multifaceted surfaces that are used in initial value generation must be closed and form a positive volume and are defined on the SURFINI entry. The SURFINI entry is referred to from the TICREG entry and together with a TICVAL entry the initial condition for the initialization surface is defined.
Main Index
126 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Eulerian Loading and Constraints
Example 1:
In the example below, a combination of two multifaceted surfaces, a cylindrical and a spherical shape together with a block of elements (all Eulerian elements) are used to define the initial conditions in an Eulerian rectangular mesh. The different shapes, initial value sets and levels used are shown below. All elements of the Eulerian mesh are defined as an ELEMENT shape with void and the lowest level 6 (see input file below). This means that the part of an element that doesn’t fall inside any of the shapes will be initialized as being void.7. Plot of the material inside the Eulerian mesh after initialization.
Input:
TICEUL1,101,101 TICREG,1,101,SURF,6,12,204,9. TICREG,1,101,SURF,7,12,205,10. TICREG,1,101,CYLINDER,8,12,206,7. TICREG,1,101,SPHERE,9,12,207,8. TICREG,1,101,ELEM,444,,,6. SET1,444,50000,THRU,53375 $ TICVAL,204,,DENSITY,1000.,XVEL,50. TICVAL,205,,DENSITY,1000.,XVEL,50.
Main Index
Chapter 5: Constraints and Loadings 127
Eulerian Loading and Constraints
TICVAL,206,,DENSITY,1000.,XVEL,50. TICVAL,207,,DENSITY,1000.,XVEL,50. $ BCPROP 116 SURFINI 6 6 Inside On $ BCPROP 117 SURFINI 7 7 Inside On $ CYLINDR,8,,1.0,.5,.5,2.0,.5,.5,+ +,.075 SPHERE,9,,1.75,.5,.5,.15
Example 2:
On On
In the example below a multifaceted surface is used with the OUTSIDE option. When the OUTSIDE option is used on the SURFINI entry, the parts of the Eulerian elements that fall outside the initialization surface are initialized.
Plot of the material inside the Eulerian mesh after initialization:
Main Index
128 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Eulerian Loading and Constraints
Input:
TICEUL1,102,102 TICREG,1,102,SURF,8,12,204,9. TICREG,2,102,,CYLINDER,7,12,206,7. TICREG,3,102,,ELEM,445,,,5. SET1,445,60000,THRU,63375 $ BCPROP 118 SURFINI 8 8 Outside On $ CYLINDR,7,,1.0,.20,.50,3.0,.40,.50,+ +,.1
On
Any initial condition that you define will act on the material as it is defined within the confines of the Eulerian mesh. The user can define a radial velocity field for the material in an Eulerian domain. The definition does not apply the standard element variables, but a sequence of four definitions that completely specify the radial velocity field. You need to define the center from where the radial is to emerge (XCENTER, YCENTER, and ZCENTER), the velocity in the direction of the radial (RVEL) and the decay coefficient (DECAY). Assume the element center at location ( x, y, z ) and the location of where the radial emerges as ( x, y, z ) . · With R , the velocity along the radial and the decay coefficient β , the velocity components for the element (mass) can be computed:
x – xr r = x – xr
and
β · v = r ⋅ R ⋅ ( x – xr )
The velocity components resulting from the radial field are added to the velocity components otherwise defined for the element. Note that the dimension of RVEL changes with the value of DECAY.
Example 3:
Assume the initialization is with sphere with origin at (0,0,0) and has a radius of 2. The DECAY coefficient is 3 and the velocity of air at the sphere boundary is 400. The value of RVEL is:
400 R – V E L =  = 50 3 2
Input:
PEULER1,100,,2ndOrder,101 $ TICEUL1,101,101 TICREG,1,101,SPHERE,1,1,4,2.0 TICREG,2,101,ELEM,2,1,5,1.0 $ SPHERE,1,,0.,0.,0.,2.0
Main Index
Chapter 5: Constraints and Loadings 129
Eulerian Loading and Constraints
TICVAL 4 DENSITY 1.14687SIE 3.204+8 ZVEL 20000. +CONT +CONT XCENTER0.0 YCENTER0.0 ZCENTER0.0 RVEL 50. +CONT +CONT DECAY 3.0 $ SET1,2,1,THRU,10000 TICVAL,5,,DENSITY,1.14687,SIE,3.204+8
Detonation
Eulerian elements that reference a JWL equation of state (EOSJWL) have to be detonated. A DETSPH entry must be present that defines a spherical detonation wave. You define the location of the detonation point, the time of detonation, and the speed of the detonation wave. The solver then calculates the time at which each explosive element detonates. Elements that do not have a JWL equation of state are unaffected.
Body Forces
If the GRAV entry is specified, the Eulerian material also has body forces acting on the material mass. The GRAV entry defines acceleration in any direction. All Eulerian material present in the problem is affected.
Hydrostatic Preset
With PARAM, HYDSTAT, the Euler element densities are initialized in accordance to a hydrostatic pressure profile. This PARAM requires the use of the GRAV entry. To impose matching boundary conditions FLOW,,,HYDSTAT and PORHYDS can be used. These two entries use the following boundary conditions:
• The pressure given by hydrostatic pressure profile. This is defined by the HYDSTAT entry • The velocities are set equal to their element values. • If fluid flows in, its density is derived from the hydrostatic pressure.
Speedup for 2D Axial Symmetric Models
To simulate a 2D axial symmetric model, a 3D pie model can be used. To get sufficient accuracy, the angle of the pie should be smaller than five degrees but not too small. The small angle gives a mesh size in circumferential direction that is much smaller than the mesh sizes in the other directions. This results in a small time step. In principle, the mesh size in the circumferential direction can be skipped for the time step computation. But often there are small errors in the circumferential normals and the circumferential direction has to be taken into account. With DYPARAM,AXIALSYM, these normals are automatically aligned. This allows a time step that is only based on the axial and radial directions, resulting in a significant larger time step. The larger time step is automatically computed when using DYPARAM,AXIALSYM.
Main Index
130 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Boundary and Loading Conditions  Theoretical Background
Boundary and Loading Conditions Theoretical Background
Pressure Boundary Conditions
Consider pressure loadings on boundary the integral:
∂b 1
in (51). To carry out the surface integration indicated by
∫∂ b1 N t t ds
(51)
a Gaussian quadrature rule is used. To locate any point of the surface under consideration, a position vector, γ , is defined:
r = f 1 ( ξ ,η )i 1 + f 2 ( ξ ,η )i 2 + f 3 ( ξ ,η ) i 3
(52)
where:
4
f i ( ξ ,η ) =
∑ φj x i
j= 1
j
(53) are unit vectors in the
x 1 , x2 ,
and
i1 , i2 ,
and
i3
and
x3
directions (see (52)).
Nodal quantities are interpolated over the fournode linear surface by the functions:
1 φ i =  ( 1 + ξξ i ) ( 1 + ηη i ) 4
(54)
ds
so that the differential surface area
d s = J dξdη
may be written in terms of the curvilinear coordinates as:
where
J
is the surface Jacobian defined by: (55)
∂r ∂r J =  ×  = ( E G – F 2 ) 1 ⁄ 2 ∂ξ ∂η
in which:
∂r ∂r E =  ⋅ ∂ξ ∂ξ ∂r ∂r F =  ⋅ ∂ξ ∂η ∂r ∂r G =  ⋅ ∂η ∂η
(56)
Main Index
Chapter 5: Constraints and Loadings 131
Boundary and Loading Conditions  Theoretical Background
A unit normal vector
n = J ∂r – 1 ⎛ ⎝ ∂ξ ∂r × ⎞ ∂η⎠
n
to the surface segment is given by: (57)
and the global components of the traction vector can now be written:
4
ti = ni
∑ φj pj
j =1
(58)
where
pj
is the applied pressure at the jth node. η
3 2
x3
x2 i3
i2
4 1
ξ
rξ,η
i1
x1
Figure 52
Parametric Representation of a Surface Segment
The surface integral for a segment is evaluated as:
∫–1 ⎨ ∫– 1 N t t J ⎬ ( dξ ) dη ⎩ ⎭
1
⎧
1
⎫
(59)
One such integral is computed for each surface segment on which a pressure loading acts. Note that the Jacobians cancel when (57) and (56) are put into (59). Equation (59) is evaluated with onepoint integration analogous to that employed in the volume integrals. The area of an element side is approximated by 4 J where J = J ( 0, 0 ) .
Main Index
132 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Boundary and Loading Conditions  Theoretical Background
Kinematic Boundary Conditions
In this subsection, the kinematic constraints are briefly reviewed. SOL 700 tracks reaction forces for each type of kinematic constraint and provides this information as output if requested. For the prescribed boundary conditions, the input energy is integrated and included in the external work.
Displacement Constraints
Translational and rotational boundary constraints are imposed either globally or locally by setting the constrained acceleration components to zero. If nodal single point constraints are employed, the constraints are imposed in a local system. The user defines the local system by specifying a vector u l in the direction of the local xaxis, x l , and a local inplane vector v l . After normalizing u l , the local x l , y l , and z l axes are given by:
ul x l = ul xl × vl z l = xl × vl yl = zl × xl
(510) (511) (512)
q
A transformation matrix
x lt q = y lt zl
t
is constructed to transform the acceleration components to the local system:
(513)
and the nodal translational and rotational acceleration vectors the local system:
a I = qa I
l
aI
and
· ωI ,
for node I are transformed to
(514) (515)
· · ω I = qω I
l
and the constrained components are zeroed. The modified vectors are then transformed back to the global system:
aI = q aI
t l
(516) (517)
t · · ωI = q ωI
l
Main Index
Chapter 5: Constraints and Loadings 133
Boundary and Loading Conditions  Theoretical Background
Prescribed Displacements, Velocities, and Accelerations
Prescribed displacements, velocities, and accelerations are treated in a nearly identical way to displacement constraints. After imposing the zero displacement constraints, the prescribed values are imposed as velocities at time, t n + 1 ⁄ 2 . The acceleration versus time curve is integrated or the displacement versus time curve is differentiated to generate the velocity versus time curve. The prescribed nodal components are then set.
Body Force Loads
Body force loads are used in many applications. For example, in structural analysis the base accelerations can be applied in the simulation of earthquake loadings, the gun firing of projectiles, and gravitational loads. The latter is often used with dynamic relaxation to initialize the internal forces before proceeding with the transient response calculation. In aircraft engine design the body forces are generated by the application of an angular velocity of the spinning structure. For base accelerations and gravity, we can fix the base and apply the loading as part of the body force loads element by element according to (518).
f ebody =
∫vm ρ N t N a base dυ
= m e a base me
(518) is the element (lumped) mass matrix.
where
a base
is the base acceleration and
Main Index
134 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Boundary and Loading Conditions  Theoretical Background
Main Index
Chapter 6: Elements MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
6
Elements
Elements Overview Preliminaries 140 144 154 163 172 136
CHEXA Solid Elements
CBEAM  Belytschko Beam
CBEAM  DYSHELFORM = 1, HughesLiu Beam
CQUADA  DYSHELLFORM = 2, BelytschkoLinTsay Shell
CQUAD4 DYSHELLFORM = 10, BelytschkoWongChiang Improvements 181 CTRIA3  DYSHELLFORM = 4, C0 Triangular Shell 184
CTRIA3  DYSHELLFORM = 3, MarchertasBelytschko Triangular Shell (BCIZ) 190 CQUAD4  DYSHELL FORM = 1, HughesLiu Shell 197
CQUAD4, DYSHELLFORM = 6, 7, Fully Integrated HughesLiu Shells 207 Transverse Shear Treatment for Layered Shell CROD, Truss Element Element 217 211
216CQUAD4  DYSHELLFORM = 9, Membrane 219
CELAS1D, Discrete Elements and CONM2, Masses CDAMP2D, Linear Elastic or Linear Viscous Eulerian Elements Graded Meshes 225 226 224
Main Index
136 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Elements Overview
Elements Overview
The heart of a finite element program lies in its element library which allows you to model a structure for analysis. MD Nastran has a very comprehensive element library which lets you model 1D, 2D, or 3D structures. This section gives some basic definitions of the element types available in MD Nastran Explicit Nonlinear. Please note, these elements differ from those used in other portions of MD Nastran.
Element Types
Each element has five definitive characteristics that determine its behavior:
• • • • •
Class Number of Nodes Interpolation Degrees of Freedom Integration Method
Class The type of geometric domain that an element represents determines the class of the element. Listed below are the classes of elements in the MD Nastran Explicit Nonlinear element library.
• Beam Elements  a 3D bar with axial, bending, and torsional stiffness. • Shell Elements  a curved, thin or thick structure with membrane/bending capabilities. • Plate Elements  a flat thin structure carrying inplane and outofplane loads. • Continuum Elements  Solid elements used to model thick sections. • Plane stress  a thin plate with inplane stresses only. All normal and shear stresses associated
with the outofplane direction are assumed to be zero. (All plane strain elements lie in the global xy plane.)
• Plane strain  a region where there is no outofplane motion and the normal and transverse
strains are zero.
• Generalized plane strain is the same as plane strain except that the normal zstrain can be a
prescribed constant or function of x and y.
• Axisymmetric elements are describe in 2D, but represent a full 3D structure where the
geometry and loading are both axisymmetric.
• 3D solid  a solid structure with only translational degrees of freedom for each node (linear or
quadratic interpolation functions).
• Truss Elements  a 3D rod with axial stiffness only (no bending). • Membrane Elements  a thin sheet with inplane stiffness only (no bending resistance). • Concentrated mass/Springs/Damper Elements • Rigid Constraints 
Main Index
Chapter 6: Elements 137
Elements Overview
Number of Nodes The number of nodes for an element define where the displacements are calculated in the analysis. Elements with only corner nodes are classified as first order elements and the calculation of displacements at locations within the element are made by linear interpolation. Elements that contain midside nodes are second order elements and quadratic interpolations are made for calculating displacements. MD Nastran SOL 700 do not support second order elements. In MD Nastran, the number of nodes is designated at the end of the element name. For example, a
CQUAD4 has 4 nodes.
Interpolation Interpolation (shape) function is an assumed function relating the displacements at a point inside an element to the displacements at the nodes of an element. In MD Nastran, three types of shape functions are used: linear, quadratic, and cubic. Degrees of Freedom Degrees of freedom is the number of unknowns at a node. In the general case, there are six degrees of freedom at a node in structural analysis (three translations, three rotations). For example there are three (translations) for 3D truss element; six (three translations, three rotations) for a 3D beam element and only three translations for the 3D solid elements. Integration Numerical integration is a method used for evaluating integrals over an element. Element quantities – such as stresses, strains, and temperatures – are calculated at each integration point of the element. Full integration (quadrature) requires, for every element, 2D integration points for linear interpolation and 3D points for quadratic interpolation, where scalar “d” is the number of geometric dimensions of an element (that is, d = 2 for a quad; d = 3 for a hexahedron). Reduced integration uses a lower number of integration points than necessary to integrate exactly. For example, for an 8node quadrilateral, the number of integration points is reduced from 9 to 4 and, for a 20node hexahedron, from 27 to 8. For some elements, an “hourglass” control method is used to insure an accurate solution.
Table 61 summarizes the elements available in MD Nastran SOL 700. This section describes the characteristics of the elements. For shell elements CTRIA3 and CQUAD4, shell element PARAM,DYNSHELLFORM is used to control the behavior.
Table 61 MD Nastran
CBAR CBEAM CBELT CBEND
Elements in MD Nastran SOL 700 Available in SOL 700 Y Y Y N Y Fatal Error
Main Index
138 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Elements Overview
Table 61 MD Nastran
CBUSH CBUTT CCONEAX CCRSFIL CDAMP1 CDAMP2 CDAMP1D CDAMP2D CELAS1 CELAS2 CELAS1D CELAS2D CFILLET CFLUID CGAP CHACAB CHACBR CHEXA COMBWLD CONM2 CONROD
Elements in MD Nastran SOL 700 Available in SOL 700 N Y N Y Y Y Y Y Y Y Y Y Y N N N N Y (8 Nodes only) Y Y Y Y Y (6 Nodes only) Y Y (4 Nodes only) Y N Y N Y Y Y Y Y Y Y Y Y Fatal Error Y
CONSPOT
CPENTA CQUAD4 CQUAD8 CQUADR CQUADX CROD CSHEAR CSPOT CSPR
Main Index
Chapter 6: Elements 139
Elements Overview
Table 61 MD Nastran
CTETRA CTRIA3 CTRIA6 CTRIA3R CTRIAX CTRIAX6
Elements in MD Nastran SOL 700 Available in SOL 700 Y (4 and 10 Nodes) Y Y (3 Nodes only) Y N N Y Y Y Y N Y Y Y Fatal Error
CTQUAD CTTRIA
CTUBE CVISC CWELD
Main Index
140 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Preliminaries
Preliminaries
Consider the body shown in Figure 61. We are interested in timedependent deformation in which a point in b initially at X α ( α = 1, 2, 3 ) in a fixed rectangular Cartesian coordinate system moves to a point x i ( i = 1, 2, 3 ) in the same coordinate system. Since a Lagrangian formulation is considered, the deformation can be expressed in terms of the convected coordinates X α , and time t
x i = x i ( X α, t )
(61) we have the initial conditions (62) (63)
At time
t = 0,
x i ( X α, 0 ) = X α · x i ( X α, 0 ) = V i ( X α )
where
x3
Vi
defines the initial velocities.
X3
n x2 X2
t = 0
b
∂b
∂B B0
x1
X1
Figure 61
Notation
Main Index
Chapter 6: Elements 141
Preliminaries
Governing Equations
We seek a solution to the momentum equation
·· σ i j, j + ρ f i = ρ x i
(64)
satisfying the traction boundary conditions
σi j nj = ti ( t )
(65)
∂b 1 ,
on boundary
the displacement boundary conditions (66) the contact discontinuity (67)
x i ( X α, t ) = D i ( t )
on boundary

∂b 2 ,
( σ i+j – σ i j )n i = 0
along an interior boundary ∂b 3 when X i+ = x i . Here σ i j is the Cauchy stress, ρ is the current density, f is ·· the body force density, x is acceleration, the comma denotes covariant differentiation, and n j is a unit outward normal to a boundary element of ∂b . Mass conservation is trivially stated
ρV = ρ 0
(68) is the relative volume, i.e., the determinant of the deformation gradient matrix,
F ij ,
where
V
∂x i F ij = ∂X j
(69)
and
ρ0
is the reference density. The energy equation (610)
· · · E = V s i j ε ij – ( p + q ) V
is integrated in time and is used for equation of state evaluations and a global energy balance. In Equation
(610), s i j and p represent the deviatoric stresses and pressure,
s i j = σ i j + ( p + q )δ ij 1 1 p = –  σ ij δ i j – q = –  σ kk – q 3 3
(611) (612)
δi j
respectively,
· εi j
q
is the bulk viscosity,
is the Kronecker delta ( δ i j
= 1
if
i = j;
otherwise
δi j = 0 )
and
is the strain rate tensor. The strain rates and bulk viscosity are discussed later.
Main Index
142 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Preliminaries
We can write:
·· ∫ ( ρ x i – σ i j, j – ρ fi )δx i dν + ∫
v
( σ i j n j – t i ) δx i ds

(613)
∂ b1
+
∂ b3
∫
( σ i+j – σ i j )n j δx i ds = 0 ∂b 2 ,
where δx i satisfies all boundary conditions on Application of the divergence theorem gives
and the integrations are over the current geometry.
υ
∫ ( σ ij δx i ), jd υ
=
∂b 1
∫
σ i j n j δx i ds +
∂b 3
∫
( σ i+ – σ i j ) n j δx i ds j

(614)
and noting that
( σ i j δx i ) ,j– σ i j, j δx i = σ ij δx i, j
(615)
leads to the weak form of the equilibrium equations:
δπ = ·· ∫ ρ xi δx i dυ + ∫ σ i j δxi, j dυ – ∫ ρf i δx i dυ – ∫
υ υ
t i δx i ds = 0
(616)
υ
∂b 1
a statement of the principle of virtual work. We superimpose a mesh of finite elements interconnected at nodal points on a reference configuration and track particles through time; i.e.,
k
x i ( X α ,t ) = x i ( X α ( ξ ,η ,ζ ) ,t ) =
∑ φ j ( ξ ,η ,ζ )x i ( t )
j j =1
(617)
( ξ ,η , ζ ) , k
where
φj
are shape (interpolation) functions of the parametric coordinates
x ij
is the number of
nodal points defining the element, and
is the nodal coordinate of the jth node in the ith direction.
δπ
Summing over the n elements we may approximate
n
with (618)
δπ =
∑
m=1
δπ m = 0
and write
⎧ ⎪ ·· m ∑ ⎨ ∫ ρ x i Φ i dυ + ⎪υ m = 1⎩ m
n
υm
∫
m σ ij Φ imj dυ ,
–
υm
∫
ρ f i Φ im dυ
–
∂ b1
∫
⎪ t i Φ im ds ⎬ ⎪ ⎭
⎫ = 0
(619)
Main Index
Chapter 6: Elements 143
Preliminaries
where
Φ im = ( φ 1 ,φ 2 ,… ,φ k ) im
(620)
In matrix notation (619) becomes:
⎧ ⎪ ∑ ⎨ ∫ ρN t N a dυ + ⎪ m = 1 ⎩ υm
n
υm
∫
B t σ dυ
–
υm
∫
ρN t b dυ
–
∂b 1
∫
N t t ds
⎫m ⎪ ⎬ = 0 ⎪ ⎭
(621)
where
N
is an interpolation matrix,
σ
is the stress vector (622)
σ t = ( σ xx, σ yy, σ zz, σ xy, σ yz, σ zx ) B
is the straindisplacement matrix,
ax
a
is the nodal acceleration vector
1 1
·· x1 ·· x
2
ay = N ay az
= Na
(623)
…
·· x3
k k
b
is the body force load vector, and
fx tx t = ty tz
t
are applied traction loads.
b =
fy fz
(624)
Main Index
144 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CHEXA Solid Elements
CHEXA Solid Elements
For a mesh of 8node hexahedron solid elements, (617) becomes:
8
x i ( X α , t ) = x i ( X α ( ξ , η , ζ ), t ) =
∑ φ j ( ξ ,η ,ζ )x i ( t )
j j =1
(625)
The shape function
φj
is defined for the 8node hexahedron as (626)
( ± 1, ± 1, ± 1 )
1 φ j =  ( 1 + ξξ j ) ( 1 + ηη j ) ( 1 + ζζ j ) 8
where ξ j ,η j ,ζ j take on their nodal values of the i th direction (see Figure 62).
ζ
and
x ij
is the nodal coordinate of the j th node in
5 8
6
7
η ξ
1
4
Node 1 2 3 4 5 6 7 8
ξ
η
ζ
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
2 3
Figure 62
Eightnode Solid Hexahedron Element
N
For a solid element,
is the 3 x 24 rectangular interpolation matrix given by
φ1 0 0 φ2 0 … 0 0 N ( ξ ,η ,ζ ) = 0 φ1 0 0 φ2 … φ8 0 0 0 φ1 0 0 … 0 φ 8
(627)
Main Index
Chapter 6: Elements 145
CHEXA Solid Elements
σ
is the stress vector (628)
σ t = ( σ xx, σ yy, σ zz, σ xy, σ yz, σ zx ) B
is the 6 x 24 straindisplacement matrix
∂  0 0 ∂x ∂ 0  0 ∂y ∂ 0 0 ∂z N ∂ ∂   0  ∂y ∂x ∂ ∂ 0  ∂z ∂y ∂ ∂  0 ∂x ∂z
B =
(629)
In order to achieve a diagonal mass matrix the rows are summed giving the k th diagonal term as
8
m kk =
υ
∫ ρ φk
∑ φ i dυ
i =1
=
∫ ρ φ k dυ
υ
(630)
since the basis functions sum to unity. Terms in the straindisplacement matrix are readily calculated. Note that
∂φ i ∂φ ∂x ∂φ ∂y ∂φ ∂ z     = i  + i  + i ∂ξ ∂x ∂ξ ∂y ∂ξ ∂z ∂ξ ∂φ ∂φ ∂x ∂φ ∂y ∂φ ∂z    i = i  + i  + i ∂η ∂x ∂η ∂y ∂η ∂ z ∂η ∂φ ∂x ∂φ ∂y ∂ φ ∂z ∂φ i = i  + i  + i    ∂ζ ∂x ∂ζ ∂y ∂ζ ∂z ∂ζ
(631)
which can be rewritten as
∂φ i ∂ξ ∂φ i ∂η ∂φ i ∂ζ ∂x ∂ξ ∂x ∂η ∂x ∂ζ ∂y ∂ξ ∂y ∂η ∂y ∂ζ ∂z ∂ξ ∂z ∂η ∂z ∂ζ ∂φ i ∂x ∂φ i ∂y ∂φ i ∂z ∂φ i ∂x = J ∂φi ∂y ∂φ i ∂z
=
(632)
Main Index
146 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CHEXA Solid Elements
Inverting the Jacobian matrix, J , we can solve for the desired terms
∂φ i ∂x ∂φ i ∂y ∂φ i ∂z ∂φ i ∂ξ ∂φ i ∂η ∂φ i ∂ζ
= J –1
(633)
Volume Integration
Volume integration is carried out with Gaussian quadrature. If volume, and n is the number of integration points, then
g
is some function defined over the
υ
∫ g dυ
=
∫–1 ∫–1 ∫– 1
1 1
1
g J dξ dη dζ
(634)
is approximated by
n n n
∑ ∑ ∑ g jkl Jj kl wj w k w l
j = 1k = 1l = 1
(635)
where
w j, w k , w l
are the weighting factors, (636)
g j kl = g ( ξ j, η k, ζ l )
and
J
is the determinant of the Jacobian matrix. For onepoint quadrature
n = 1 wi = wj = wk = 2 ξ1 = η1 = ζ1 = 0
(637)
and we can write
∫ g dv
= 8 g ( 0 ,0 ,0 ) J ( 0 ,0 ,0 ) 8 J ( 0 ,0 ,0 )
(638)
Note that
approximates the element volume.
Perhaps the biggest advantage to onepoint integration is a substantial savings in computer time. An antisymmetry property of the strain matrix
∂φ 1 ∂ φ7  = – ∂x i ∂x i ∂ φ8 ∂φ 2  = – ∂x i ∂x i ∂φ 3 ∂φ 5  = – ∂x i ∂x i ∂φ 4 ∂φ 6  = – ∂x i ∂x i
(639)
Main Index
Chapter 6: Elements 147
CHEXA Solid Elements
at ξ = η = ζ = 0 reduces the amount of effort required to compute this matrix by more than 25 times over an 8point integration. This cost savings extends to strain and element nodal force calculations where the number of multiplies is reduced by a factor of 16. Because only one constitutive evaluation is needed, the time spent determining stresses is reduced by a factor of eight. Operation counts for the constant stress hexahedron are given in Table 62. Included are counts for the Flanagan and Belytschko [1981] hexahedron and the hexahedron used by Wilkins [1974] in his integral finite difference method, which was also implemented [Hallquist 1979].
.
Table 62
Operation Counts for Constant Stress Hexahedron* FlanaganWilkins SOL 700 Belytschko [1981] 357 56 195 708 620 1328 270 1,113 680 1,793 FDM 843
Strain displacement matrix Strain rates Force Subtotal Hourglass control Total
94 87 117 298 130 428
*Includes adds, subtracts, multiplies, and divides in major subroutines, and is independent of vectorization. It may be noted that 8point integration has another disadvantage in addition to cost. Fully integrated elements used in the solution of plasticity problems and other problems where Poisson’s ratio approaches .5 lock up in the constant volume bending modes. To preclude locking, an average pressure must be used over the elements; consequently, the zero energy modes are resisted by the deviatoric stresses. If the deviatoric stresses are insignificant relative to the pressure or, even worse, if material failure causes loss of this stress state component, hourglassing will still occur, but with no means of resisting it. Sometimes, however, the cost of the fully integrated element may be justified by increased reliability and if used sparingly may actually increase the overall speed.
Hourglass Control
The biggest disadvantage to onepoint integration is the need to control the zero energy modes, which arise, called hourglassing modes. Undesirable hourglass modes tend to have periods that are typically much shorter than the periods of the structural response, and they are often observed to be oscillatory. However, hourglass modes that have periods that are comparable to the structural response periods may be a stable kinematic component of the global deformation modes and must be admissible. One way of resisting undesirable hourglassing is with a viscous damping or small elastic stiffness capable of stopping the formation of the anomalous modes but having a negligible affect on the stable global modes. Two of the early threedimensional algorithms for controlling the hourglass modes were developed by Kosloff and Frazier [1974] and Wilkins et al. [1974].
Main Index
148 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CHEXA Solid Elements
Since the hourglass deformation modes are orthogonal to the strain calculations, work done by the hourglass resistance is neglected in the energy equation. This may lead to a slight loss of energy; however, hourglass control is always recommended for the under integrated solid elements. The energy dissipated by the hourglass forces reacting against the formations of the hourglass modes is tracked and reported in the output files MATSUM and GLSTAT. This is only provided if the PARM,DYENERGYGE is included. It is easy to understand the reasons for the formation of the hourglass modes. Consider the following strain rate calculations for the 8node solid element
⎛ 8 ∂φ ∂φ k ⎞ 1 · k· · ε i j =  ⎜ ∑  x jk +  x ik⎟ ∂x ∂x j ⎟ 2⎜ ⎝k = 1 i ⎠
(640)
Whenever diagonally opposite nodes have identical velocities; i.e.,
· · · · · · · · x i1 = x i7, x i2 = x i8, x i3 = x i5, x i4 = x i6
(641)
the strain rates are identically zero:
· εi j = 0
(642)
due to the asymmetries in (639). It is easy to prove the orthogonality of the hourglass shape vectors, which are listed in Table 63 and shown in Figure 62 with the derivatives of the shape functions:
8
∑
k=1
∂φ k  Γ αk = 0 ∂x i
i = 1 ,2 ,3
α = 1 ,2 ,3 ,4
(643)
Table 63
Hourglass Base Vectors
α = 1 α = 2 α = 3 α = 4
Γ j1 Γ j2 Γ j3 Γ j4 Γ j5 Γ j6 Γ j7 Γ j8
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1
The hourglass modes of an 8node element with one integration point are shown [Flanagan and Belytschko 1981] (Figure 63). A total of twelve modes exist.
Main Index
Chapter 6: Elements 149
CHEXA Solid Elements
Γ1k
Γ2k
Γ3k
Figure 63 Hourglass Modes The product of the base vectors with the nodal velocities
hi α =
Γ4k
∑
k=1
8
· x ik Γ αk = 0
(644)
f ik α
are nonzero if hourglass modes are present. The 12 hourglassresisting force vectors,
f ik = a h h iα Γ αk α
are (645)
where
c 2 a h = Q hg ρν e ⁄ 3 4
(646)
in which ν e is the element volume, c is the material sound speed, and Q h g is a userdefined constant usually set to a value between .05 and .15. This is defined by DYHRGQH PARAM. The hourglass resisting forces of (645) are not orthogonal to rigid body rotations; however, the approach of Flanagan and Belytschko [1981] is orthogonal. This is controlled by using the DYHRGIHQ PARAM. Material subroutines add as little as 60 operations for the bilinear elasticplastic routine to ten times as much for multisurface plasticity and reactive flow models. Unvectorized material models will increase that share of the cost a factor of four or more.
Main Index
150 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CHEXA Solid Elements
Instead of resisting components of the bilinear velocity field that are orthogonal to the strain calculation, Flanagan and Belytschko resist components of the velocity field that are not part of a fully linear field. They call this field, defined below, the hourglass velocity field
· HG = x – x k LIN · · x ik i i
(647)
where
· · · LIN = x i + x i ,j ( x k – x ) x ik j j
8 8
(648)
1 · x i = 8
1 x i = 8
∑
k=1
x ik
∑ x· i
k=1
k
(649)
Flanagan and Belytschko construct geometrydependent hourglass shape vectors that are orthogonal to the fully linear velocity field and the rigid body field. With these vectors they resist the hourglass velocity deformations. Defining hourglass shape vectors in terms of the base vectors as
8
γ ak = Γ ak – φ k ,i
∑
n =1
x in Γ an
(650)
and setting
8
gi a =
∑
k= 1
· x ik γ αk = 0
(651)
the 12 resisting force vectors become
k f iα = a h g iα γ α k
(652)
where
ah
is a constant given in (645).
The hourglass forces given by (645) and (652) are identical if the, hexahedron element is a parallelepiped. The default hourglass control method for solid element is given by (645); however, we recommend the FlanaganBelytschko approach for problems that have large rigid body rotations since the default approach is not orthogonal to rigid body rotations. A cost comparison in Table 62 shows that the default hourglass viscosity requires approximately 130 adds or multiplies per hexahedron, compared to 620 and 680 for the algorithms of FlanaganBelytschko and Wilkins. Hourglass stabilization for the 3D hexahedral element is available. Based on material properties and element geometry, this stiffness type stabilization is developed by an assumed strain method [Belytschko and Bindeman 1993] such that the element does not lock with nearly incompressible material. This is activated by using the PARAM,DYHRGIHG,6. When the userdefined hourglass constant Q h g is set to 1.0, accurate coarse mesh bending stiffness is obtained for elastic material. For nonlinear material, a smaller value of Q hg is suggested and the default value is set to 0.1.
Main Index
Chapter 6: Elements 151
CHEXA Solid Elements
Fully Integrated Brick Elements and MidStep Strain Evaluation
To avoid locking in the fully integrated brick elements strain increments at a point in a constant pressure, solid element are defined by [see Nagtegaal, Parks, and Rice 1974]
∂Δ u =  + φ ∂x n + 1 ⁄ 2 ∂Δ u ∂Δν  + ∂ x n + 1 ⁄ 2 ∂y n + 1 ⁄ 2 = 2 ∂Δ w ∂Δ v  + ∂y n + 1 ⁄ 2 ∂z n + 1 ⁄ 2 = 2 ∂Δ u ∂ Δw  + ∂z n + 1 ⁄ 2 ∂x n + 1 ⁄ 2 = 2
Δ ε xx
Δ ε xy
Δ ε yy
∂Δ v =  + φ ∂y n + 1 ⁄ 2
Δ ε yz
Δ ε zz
∂Δ w =  + φ ∂z n + 1 ⁄ 2
Δ ε zx
(653)
where φ modifies the normal strains to ensure that the total volumetric strain increment at each integration point is identical
n Δ εv + 1 ⁄ 2 n+1⁄2
φ =
∂Δ u ∂Δν ∂Δ w  +  + ∂ x n + 1 ⁄ 2 ∂y n + 1 ⁄ 2 ∂z n + 1 ⁄ 2 – 3
(654)
and
Δ εν
is the average volumetric strain increment in the midpoint geometry
∂Δ u 1 ∂Δν ∂Δ w   ∫ ⎛ ∂x n + 1 ⁄ 2 + ∂y n + 1 ⁄ 2 + ∂z n + 1 ⁄ 2⎞ dv n + 1 ⁄ 2 ⎝ ⎠ 3 vn + 1 ⁄ 2 ∫ dv n + 1 ⁄ 2
vn + 1 ⁄ 2
(655)
Δ u, Δv ,
and
Δw
are displacement increments in the
x, y ,
and
z
directions, respectively, and (656) (657) (658)
(xn + xn + 1) x n + 1 ⁄ 2 = 2 (yn + yn + 1) y n + 1 ⁄ 2 = 2 (zn + zn + 1 ) z n + 1 ⁄ 2 = 2
To satisfy the condition that rigid body rotations cause zero straining, it is necessary to use the geometry at the midstep in the evaluation of the strain increments. The explicity formulation uses the geometry at step n + 1 to save operations; however, for calculations, which involve rotating parts, the midstep geometry should be used especially if the number of revolutions is large.
Main Index
152 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CHEXA Solid Elements
Since the bulk modulus is constant in the plastic and viscoelastic material models, constant pressure solid elements result. In the thermoelastoplastic material, a constant temperature is assumed over the element. In the soil and crushable foam material, an average relative volume is computed for the element at time step n + 1 , and the pressure and bulk modulus associated with this relative volume is used at each integration point. For equations of state, one pressure evaluation is done per element and is added to the deviatoric stress tensor at each integration point. The foregoing procedure requires that the straindisplacement matrix corresponding to equation (653) and consistent with a constant volumetric strain, B , be used in the nodal force calculations [Hughes 1980]. It is easy to show that:
F =
∫
vn + 1
B n + 1 σ n + 1 dv n + 1 =
t
∫
vn + 1
B n + 1 σ n + 1 dv n + 1
t
(659)
B.
and avoid the needless complexities of computing
CTETRA  Four Node Tetrahedron Element
The four node tetrahedron element with one point integration, shown in Figure 64, is a simple, fast, solid element that has proven to be very useful in modeling low density foams that have high compressibility. For most applications, however, this element is too stiff to give reliable results and is primarily used for transitions in meshes. The formulation follows the formulation for the one point solid element with the difference that there are no kinematic modes, so hourglass control is not needed. The basis functions are given by:
N 1 ( r, s , t ) = r N 2 ( r, s , t ) = s N 3 ( r, s , t ) = 1 – r – s – t N 4 ( r, s , t ) = t
(660)
t
4
3
1 2 s
r
Figure 64
Fournode tetrahedron
Main Index
Chapter 6: Elements 153
CHEXA Solid Elements
If a tetrahedron element is needed, this element should be used instead of the collapsed solid element since it is, in general, considerably more stable in addition to being much faster.
CPENTA  Six Node Pentahedron Element
The pentahedron element with two point Gauss integration along its length, shown in Figure 65, is a solid element that has proven to be very useful in modeling axisymmetric structures where wedge shaped elements are used along the axisofrevolution. The formulation follows the formulation for the one point solid element with the difference that, like the tetrahedron element, there are no kinematic modes, so hourglass control is not needed. The basis functions are given by:
1 N 1 ( r, s, t ) =  ( 1 – t )r 2 1 N 2 ( r, s, t ) =  ( 1 – t ) ( 1 – r – s ) 2 1 N 3 ( r, s, t ) =  ( 1 + t ) ( 1 – r – s ) 2 1 N 4 ( r, s, t ) =  ( 1 + t )r 2 1 N 5 ( r, s, t ) =  ( 1 – t )s 2 1 N 6 ( r, s, t ) =  ( 1 + t )s 2
(661)
If a pentahedron element is needed, this element should be used instead of the collapsed solid element since it is, in general, more stable and significantly faster. Selectivereduced integration is used to prevent volumetric locking; i.e., a constant pressure over the domain of the element is assumed.
t
4 6
r s 5 2 1
Figure 65
Six Node Pentahedron
Main Index
154 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CBEAM  Belytschko Beam
CBEAM  Belytschko Beam
The Belytschko beam element formulation [Belytschko et. al.1977] is part of a family of structural finite elements, by Belytschko and other researchers that employ a ‘corotational technique’ in the element formulation for treating large rotation. This section discusses the corotational formulation, since the formulation is most easily described for a beam element, and then describes the beam theory used to formulate the corotational beam element.
Corotational Technique
In any large displacement formulation, the goal is to separate the deformation displacements from the rigid body displacements, as only the deformation displacements give rise to strains and the associated generation of strain energy. This separation is usually accomplished by comparing the current configuration with a reference configuration. The current configuration is a complete description of the deformed body in its current spatial location and orientation, giving locations of all points (nodes) comprising the body. The reference configuration can be either the initial configuration of the body (i.e., nodal locations at time zero) or the configuration of the body at some other state (time). Often, the reference configuration is chosen to be the previous configuration, say at time t n = t n + 1 – Δ t . The choice of the reference configuration determines the type of deformations that will be computed: total deformations result from comparing the current configuration with the initial configuration, while incremental deformations result from comparing with the previous configuration. In most time stepping (numerical) Lagrangian formulations, incremental deformations are used because they result in significant simplifications of other algorithms, chiefly constitutive models. A direct comparison of the current configuration with the reference configuration does not result in a determination of the deformation, but rather provides the total (or incremental) displacements. We will use the unqualified term displacements to mean either the total displacements or the incremental displacements, depending on the choice of the reference configuration as the initial or the last state. This is perhaps most obvious if the reference configuration is the initial configuration. The direct comparison of the current configuration with the reference configuration yields displacements, which contain components due to deformations and rigid body motions. The task remains of separating the deformation and rigid body displacements. The deformations are usually found by subtracting from the displacements an estimate of the rigid body displacements. Exact rigid body displacements are usually only known for trivial cases where they are prescribed a priori as part of a displacement field. The corotational formulations provide one such estimate of the rigid body displacements. The corotational formulation uses two types of coordinate systems: one system associated with each element; i.e., element coordinates which deform with the element, and another associated with each node; i.e., body coordinates embedded in the nodes. (The term ‘body’ is used to avoid possible confusion from referring to these coordinates as ‘nodal’ coordinates. Also, in the more general formulation presented in [Belytschko et al., 1977], the nodes could optionally be attached to rigid bodies. Thus, the term ‘body coordinates’ refers to a system of coordinates in a rigid body, of which a node is a special case.) These two coordinate systems are shown in the upper portion of Figure 66.
Main Index
Chapter 6: Elements 155
CBEAM  Belytschko Beam
Y b2 e2
0
^ Y
e1
0
b1 I
X (a) Initial Configuration
^X
J
^ X J
Y
^Y
b2 e1 e2 X b1 (b) Rigid Rotation Configuration ^ Y b2 e2 e1
J
I
Y ^ X
e1 b1
0
X (c) Deformed Configuration
Figure 66
Corotational Coordinate System
Main Index
156 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CBEAM  Belytschko Beam
ˆ The element coordinate system is defined to have the local xaxis x originating at node I and terminating ˆ and, in three dimension, the local zaxis z , are constructed normal to x . The ˆ ˆ at node J ; the local yaxis y ˆ ˆ ˆ element coordinate system ( x, y, z ) and associated unit vector triad ( e 1, e 2, e 3 ) are updated at every time
step by the same technique used to construct the initial system; thus the unit vector element since it always points from node I to node J .
e1
deforms with the
The embedded body coordinate system is initially oriented along the principal inertial axes; either the assembled nodal mass or associated rigid body inertial tensor is used in determining the inertial principal values and directions. Although the initial orientation of the body axes is arbitrary, the selection of a principal inertia coordinate system simplifies the rotational equations of motion; i.e., no inertial cross product terms are present in the rotational equations of motion. Because the body coordinates are fixed in the node (or rigid body), they rotate and translate with the node and are updated by integrating the rotational equations of motion, as will be described subsequently. The unit vectors of the two coordinate systems define rotational transformations between the global coordinate system and each respective coordinate system. These transformations operate on vectors with global components A = ( A x, A y, A z ) , body coordinates components A = A x, A y, A z , and element coordinate
ˆ components A ⎧ ⎫ ⎪ Ax ⎪ ⎪ ⎪ A = ⎨ Ay ⎬ = ⎪ ⎪ ⎪ Az ⎪ ⎩ ⎭ ˆ ˆ ˆ = A x, A y, A z
which are defined as:
⎧ ⎫ b 1x b 2x b 3x ⎪ A x ⎪ ⎪ ⎪ b 1y b 2y b 3y ⎨ A y ⎬ = [ λ ] { A } ⎪ ⎪ b 1z b 2z b 3z ⎪ A ⎪ ⎩ z ⎭
(662)
where b i x, b i y, b iz are the global components of the body coordinate unit vectors. Similarly for the element coordinate system:
⎧ ⎫ ⎪ Ax ⎪ ⎪ ⎪ A = ⎨ Ay ⎬ = ⎪ ⎪ ⎪ Az ⎪ ⎩ ⎭ ⎧ e 1x e 2x e 3x ⎪ ⎪ e 1y e 2y e 3y ⎨ ⎪ e 1z e 2z e 3z ⎪ ⎩ ˆ Ax ˆ Ay ˆ Az ⎫ ⎪ ⎪ ˆ ⎬ = [μ]{A} ⎪ ⎪ ⎭
(663)
where e ix, e iy, e iz are the global components of the element coordinate unit vectors. The inverse transformations are defined by the matrix transpose: i.e.,
{A} = [λ] {A}
T ˆ {A} = [μ] {A } T
(664) (665)
since these are proper rotational transformations. The following two examples illustrate how the element and body coordinate system are used to separate the deformations and rigid body displacements from the displacements.
Main Index
Chapter 6: Elements 157
CBEAM  Belytschko Beam
Rigid Rotation First, consider a rigid body rotation of the beam element about node I , as shown in the center of Figure 66b; i.e., consider node I to be a pinned connection. Because the beam does not deform during the rigid rotation, the orientation of the unit vector e 1 in the initial and rotated configuration will be the same with respect to the body coordinates. If the body coordinate components of the initial element unit
0 vector e 1 were stored, they would be identical to the body coordinate components of the current element unit vector e 1 .
Deformation Rotation Next, consider node I to be constrained against rotation; i.e., a clamped connection. Now node J is moved, as shown in the lower portion of Figure 66, causing the beam element to deform. The updated element unit vector e 1 is constructed and its body coordinate components are compared to the body
0 coordinate components of the original element unit vector e 1 . Because the body coordinate system did not rotate, as node I was constrained, the original element unit vector and the current element unit vector are not colinear. Indeed, the angle between these two unit vectors is the amount of rotational deformation at node I ; i.e., 0 e1 × e1 = θ l e 3
(666)
Thus the corotational formulation separates the deformation and rigid body deformations by using:
• a coordinate system that deforms with the element; i.e., the element coordinates or • a coordinate system that rigidly rotates with the nodes; i.e., the body coordinates.
Then, it compares the current orientation of the element coordinate system with the initial element coordinate system, using the rigidly rotated body coordinate system, to determine the deformations.
Belytschko Beam Element Formulation
The deformation displacements used in the Belytschko beam element formulation are:
ˆ ˆ ˆ ˆ ˆ ˆT d = { δ IJ, θ xJI, θ yI, θ yJ, θ zI, θ zJ }
(667)
where
δ IJ ˆ θ xJI
= length change = torsional deformation = bending rotational deformations
ˆ
ˆ ˆ ˆ ˆ θ yI, θ yJ, θ zI, θ zJ
The superscript and
I
emphasizes that these quantities are defined in the local element coordinate system,
and
J
are the nodes at the ends of the beam.
Main Index
158 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CBEAM  Belytschko Beam
The beam deformations, defined in equation (662), are the usual small displacement beam deformations (see, for example, [Przemieniecki 1986]). Indeed, one advantage of the corotational formulation is the ease with which existing small displacement element formulations can be adapted to a large displacement formulation having small deformations in the element system. Small deformation theories can be easily accommodated because the definition of the local element coordinate system is independent of rigid body rotations and hence deformation displacement can be defined directly.
Calculation of Deformations
The elongation of the beam is calculated directly from the original nodal coordinates total displacements ( u xI, u yI, u zI ) :
1 2 2 2 δ IJ =  [ 2 ( X JI u xJI + Y JI u yJI + Z JI u zJI ) + u xJI + u yJI + u zJI ] l + lo ( X I, Y I, Z I )
and the
(668)
where
X JI = X J – X I u xJI = u xJ – u xI
(669) etc. (670)
The deformation rotations are calculated using the body coordinate components of the original element coordinate unit vector along the beam axis; i.e.,
0 e 1 , as outlined in the previous section. Because the body 0 coordinate components of initial unit vector e 1 rotate with the node, in the deformed configuration it indicates the direction of the beam’s axis if no deformations had occurred. Thus comparing the initial unit
vector
0 e1
with its current orientation
0 e1
e1
indicates the magnitude of deformation rotations. Forming the
vector cross product between
ˆ ˆ 0 e 1 × e 1 = θ y e 2 + θ z e3
and
e1 :
(671)
where
ˆ θy ˆ θz
is the incremental deformation about the local is the incremental deformation about the local
ˆ y ˆ z
axis axis
The calculation is most conveniently performed by transforming the body components of the initial element vector into the current element coordinate system:
⎧ ⎪ ⎪ ⎨ ⎪ ⎪ ⎩ ⎫ ⎧ 0 ˆ0 e 1x ⎪ ⎪ e 1x ⎪ ⎪ ˆ 0 ⎬ = [ μ ]T [ λ ] ⎨ e 0 e 1y ⎪ ⎪ 1y ˆ0 ⎪ e0 e 1z ⎪ ⎭ ⎩ 1z ⎫ ⎪ ⎪ ⎬ ⎪ ⎪ ⎭
(672)
Main Index
Chapter 6: Elements 159
CBEAM  Belytschko Beam
Substituting equation (672) into ((671))
e1 e2 e3
0 e 1 × e 1 = d et
1
0
0
ˆ ˆ ˆ0 ˆ0 = – e 1z e 2 + e 1 y e 3 = θ y e 2 + θ z e 3
(673)
ˆ0 ˆ0 ˆ0 e 1x e 1y e 1z
Thus,
ˆ ˆ0 θy = –e 1 z ˆ ˆ0 θ z = e 1y
(674) (675)
The torsional deformation rotation is calculated from the vector cross product of initial unit vectors, from
ˆ0 ˆ0 ˆ0 ˆ0 each node of the beam, that were normal to the axis of the beam, i.e., e 2 I and e 2J ; note that e 3 I and e 3J could also be used. The result from this vector cross product is then projected onto the current axis of the beam; i.e., e1 e2 e 3 ˆ ˆ 0 × e0 ) = e det ˆ 0 ˆ 0 ˆ 0 ˆ θ xJI = e 1 ⋅ ( e 2I e x2I e y2I e z 2I 2J 1 ˆ0 ˆ0 ˆ0 e x2J e y2 J e z2J
0 0 Note that the body components of e 2 I and e 2 J are transformed into the current element coordinate system before performing the indicated vector products.
ˆ0 ˆ0 ˆ0 ˆ0 = e y2 I e z2 J – e y2J e z2 I
(676)
Calculation of Internal Forces
There are two methods for computing the internal forces for the Belytschko beam element formulation:
• Functional forms relating the overall response of the beam e.g., momentcurvature relations, • Direct throughthethickness integration of the stress.
Currently only the former method, as explained subsequently, is implemented; the direct integration method is detailed in [Belytschko et al., 1977]. Axial Force The internal axial force is calculated from the elongation of the beam δ , as given by equation (668), and an axial stiffness:
ˆ f xJ = K a δ
(677)
where:
Ka = A E ⁄ l0 A
is the axial stiffness
is the cross sectional area of the beam
Main Index
160 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CBEAM  Belytschko Beam
E l0
is Young’s Modulus is the original length of the beam
Bending Moments The bending moments are related to the deformation rotations by
ˆ ˆ b K y 4 + φ y 2 – φ y ⎧ θ yI ⎫ ⎧ m yI ⎫ ⎨ ˆ ⎬ = ⎨ˆ ⎬ 1 + φy 2 – φ 4 + φ ⎩ ⎩ m yJ ⎭ y y θ yJ ⎭
b ˆ ˆ K z 4 + φ z 2 – φ z ⎧ θ zI ⎫ ⎧ m zI ⎫ = ⎨ˆ ⎬ ˆ ⎭ 1 + φz 2 – φ 4 + φ ⎨ θ ⎬ ⎩ m zJ ⎭ z z ⎩ zJ
(678)
(679)
ˆ ˆ x–z
where equation (678) is for bending in the plane. The bending constants are given by
EI yy b K y = l0 EI zz b K z = l0 I yy = I zz = ˆ ˆ ˆ ∫ ∫ z 2 dy dz ˆ ˆ ˆ ∫ ∫ y 2 dy dz 12E I zz φ z = GA s l 2 G
plane and equation (679) is for bending in the
ˆ ˆ x–y
(680) (681) (682) (683)
12 E I yy φ y = G As l 2
(684)
As
Hence
φ
is the shear factor,
the shear modulus, and
is the effective area in shear.
Torsional Moment The torsional moment is calculated from the torsional deformation rotation as
ˆ ˆ m xJ = K t θ xJI
(685)
where
GJ K t = l0 J = ˆˆ ˆ ˆ ∫ ∫ y z dy dz
(686) (687)
Main Index
Chapter 6: Elements 161
CBEAM  Belytschko Beam
The above forces are conjugate to the deformation displacements given previously in equation (667); i.e.,
ˆT ˆ ˆ ˆ ˆ ˆ d = { δ IJ, θ xJI, θ yI, θ yJ, θ zI, θ zJ }
(688)
where
i nt ˆ T ˆ {d } { f} = W
(689) (690)
ˆT ˆ ˆ ˆ ˆ ˆ ˆ f = { f xJ, m xJ, m yI, m yJ, m zI, m zJ }
The remaining internal force components are found from equilibrium:
ˆ ˆ f xI = – f xJ ˆ ˆ m yI + m yJ ˆ f zI = – l0 ˆ ˆ m zI + m zJ ˆ f yJ = – 0 l ˆ ˆ m xI = – m xJ ˆ ˆ f zI = – f zJ
(691)
ˆ ˆ f yI = – f yJ
Updating the Body Coordinate Unit Vectors The body coordinate unit vectors are updated using the Newmark β Method [Newmark 1959] with β = 0 , which is almost identical to the central difference method [Belytschko 1974]. In particular, the body component unit vectors are updated using the formula
j+1
bi
db i Δ t 2 d 2 b i j = b i + Δ t  +  dt 2 dt 2
j
j
(692)
where the superscripts refer to the time step and the subscripts refer to the three unit vectors comprising the body coordinate triad. The time derivatives in the above equation are replaced by their equivalent forms from vector analysis:
db i  = ω × b i dt d 2 bi  = ω × ( ω × b i ) + ( α i × b i ) dt 2
j j
(693)
(694)
where ω and α are vectors of angular velocity and acceleration, respectively, obtained from the rotational equations of motion. With the above relations substituted into equation (692), the update formula for the unit vectors becomes
bi
j+1 j Δ t2 = b i + Δ t ( ω × b i ) +  { [ ω × ( ω × b i ) + ( α i × b i ) ] } 2
(695)
Main Index
162 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CBEAM  Belytschko Beam
To obtain the formulation for the updated components of the unit vectors, the body coordinate system is temporarily considered to be fixed and then the dot product of equation (695) is formed with the unit vector to be updated. For example, to update the x component of b 3 , the dot product of equation (695), with i = 3 , is formed with b 1 , which can be simplified to the relation
b x3
j+1
= b 1 ⋅ b y3
j
j+1
j j Δ t2 j j = Δt ω y +  ( ω x ω z + α y ) 2
(696)
Similarly,
b y3
j+1
= b2 ⋅ b3
j
j
j+1
j j Δ t2 j j = Δt ω x +  ( ω y ω z + α x ) 2 j j Δ t2 j j = Δt ω z +  ( ω x ω y + α z ) 2 j+1 j+1
(697) (698) are found by using normality and orthogonality, where it is
b x2
j+1
= b1 ⋅ b2
j+1
The remaining components
b3
and
ω
b1
assumed that the angular velocities
are small during a time step so that the quadratic terms in the
b3
j+1
update relations can be ignored. Since
b z3
j+1
is a unit vector, normality provides the relation (699)
=
1 – ( b x3 ) – ( b y3 )
j+1 2
j+1 2
Next, if it is assumed that
b z1
j+1
b x1
j+1
≈1,
orthogonality yields
(6100) (6101)
b x3 + b y1 b y3 = – j+1 b z3 b x1
j+1
j+1
j+1 j+1
The component
b x1
j+1
is then found by enforcing normality:
j+1 2
=
1 – ( b y1 ) – ( b z1 )
j+1 2
(6102)
The updated components of b 1 and b 3 are defined relative to the body coordinates at time step j . To complete the update and define the transformation matrix, equation (662), at time step j + 1 , the updated unit vectors b 1 and b 3 are transformed to the global coordinate system, using equation (662) with [ λ ] defined at step j , and their vector cross product is used to form b 2 .
Main Index
Chapter 6: Elements 163
CBEAM  DYSHELFORM = 1, HughesLiu Beam
CBEAM  DYSHELFORM = 1, HughesLiu Beam
The HughesLiu beam element formulation, based on the shell [Hughes and Liu 1981a, 1981b] discussed later. It has several desirable qualities:
• It is incrementally objective (rigid body rotations do not generate strains), allowing for the
treatment of finite strains that occur in many practical applications.
• It is simple, which usually translates into computational efficiency and robustness. • It is compatible with the brick elements, because the element is based on a degenerated brick
element formulation.
• It includes finite transverse shear strains. The added computations needed to retain this strain
component, compare to those for the assumption of no transverse shear strain, are insignificant.
Geometry
The HughesLiu beam element is based on a degeneration of the isoparametric 8node solid element, an approach originated by Ahmad et al., [1970]. Recall the solid element isoparametric mapping of the biunit cube
x ( ξ ,η ,ζ ) = N a ( ξ ,η ,ζ )x a ( 1 + ξa ξ ) ( 1 + ηa η ) ( 1 + ζa ζ ) N a ( ξ ,η ,ζ ) = 8
(6103) (6104)
( ξ ,η , ζ )
where
x
is an arbitrary point in the element,
Na
are the parametric coordinates,
xa
are the global
ξ a ,η a ,ζ a
nodal coordinates of node a , and are ( ξ ,η ,ζ ) evaluated at node a .
are the element shape functions evaluated at node a , i.e.,
In the beam geometry, ξ determines the location along the axis of the beam and the coordinate pair ( η ,ζ ) defines a point on the cross section. To degenerate the 8node brick geometry into the 2node beam geometry, the four nodes at ξ = – 1 and at ξ = 1 are combined into a single node with three translational and three rotational degrees of freedom. Orthogonal, inextensible nodal fibers are defined at each node for treating the rotational degrees of freedom. Figure 67 shows a schematic of the biunit cube and the beam element. The mapping of the biunit cube into the beam element is separated into three parts:
x ( ξ ,η ,ζ ) = x ( ξ ) + X ( ξ ,η ,ζ ) = x ( ξ ) + X ζ ( ξ ,ζ ) + X η ( ξ ,η )
(6105)
where x denotes a position vector to a point on the reference axis of the beam, and X ζ and X η are position vectors at point x on the axis that define the fiber directions through that point. In particular,
x ( ξ ) = N a ( ξ )x a X η ( ξ, η ) = N a ( ξ ) X ηa ( η ) X ζ ( ξ, ζ ) = N a ( ξ ) X ζa ( ζ )
(6106) (6107) (6108)
Main Index
164 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CBEAM  DYSHELFORM = 1, HughesLiu Beam
ζ
Biunit Cube
ζ η
x
ξ
η
Beam Element
ξ
Nodal Fibers
Top Surface +1
zζ
+ +
Xζ
Xζ
ζ
ˆ Xζ
0
ˆ X
η
Xζ 1 Bottom Surface zζ
Figure 67
HughesLiu Beam Element
With this description, arbitrary points on the reference line x are interpolated by the one dimensional shape function N ( ξ ) operating on the global position of the two beam nodes that define the reference axis, i.e., x a . Points off the reference axis are further interpolated by using a onedimensional shape function along the fiber directions; i.e., X η a ( η ) and X ζa ( ζ ) where
ˆ X ηa ( η ) = z η ( η )X ηa
+ z η ( η ) = N + ( η )z ηa + N  ( η )z ηa
(6109) (6110)
Main Index
Chapter 6: Elements 165
CBEAM  DYSHELFORM = 1, HughesLiu Beam
(1 + η) N + ( η ) = 2 (1 – η) N  ( η ) = 2 ˆ X ζ a ( ζ ) = z ζ ( ζ )X ζ a
+ z ζ ( ζ ) = N + ( ζ ) z ζa + N  ( ζ ) z ζa
(6111) (6112) (6113) (6114) (6115) (6116)
zη ( η )
(1 + ζ) N + ( ζ ) = 2 (1 – ζ) N  ( ζ ) = 2
where
zζ ( ζ )
and
are “thickness functions”.
The HughesLiu beam formulation uses four position vectors, in addition to ξ , to locate the reference + axis and define the initial fiber directions. Consider the two position vectors x ζa and x ζa located on the top and bottom surfaces, respectively, at node a . Then
1 + – x ζ a =  [ ( 1 – ζ )x ηa + ( 1 + ζ ) ]x ζa 2
+ ( xζ a – xζ a ) ˆ X ζ a = + xζ a – xζ a
(6117) (6118) (6119) (6120) (6121) (6122) (6123) (6124)
1 + + z ζa =  ( 1 – ζ ) ⋅ x ζ a – x ζa 2 1 + z ζa = –  ( 1 + ζ ) ⋅ x ζa – x ζa 2 1 + – x ηa =  [ ( 1 – ζ )x ηa + ( 1 + ζ ) ] x ηa 2
+ ( x ηa – x ηa ) ˆ X ηa = + x ηa – x ηa
1 + + z ηa =  ( 1 – η ) ⋅ x ηa – x ηa 2 1 + z ηa = –  ( 1 + η ) ⋅ x ηa – x ηa 2 .
where is the Euclidean norm. The reference surface may be located at the midsurface of the beam or offset at the outer surfaces. This capability is useful in several practical situations involving contact surfaces, connection of beam elements to solid elements, and offsetting elements such as for beam stiffeners in stiffened shells. The reference surfaces are located within the beam element by specifying
Main Index
166 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CBEAM  DYSHELFORM = 1, HughesLiu Beam
the value of the parameters η and ζ , (see lower portion of Figure 67). When these parameters take on the values – 1 or +1 , the reference axis is located on the outer surfaces of the beam. If they are set to zero, the reference axis is at the center. The same parametric representation used to describe the geometry of the beam elements is used to interpolate the beam element displacements; i.e., an isoparametric representation. Again, the displacements are separated into the reference axis displacements and rotations associated with the fiber directions:
u ( ξ , η, ζ ) = u ( ξ ) + U ( ξ , η , ζ ) = u ( ξ ) + U ζ ( ξ , ζ ) + U η ( ξ , η ) u ( ξ ) = N a ( ξ )u a U η ( ξ, η ) = N a ( ξ ) U ηa ( η ) U ζ ( ξ, ζ ) = N a ( ξ )U ζa ( ζ ) ˆ U ηa ( η ) = z ηa ( η )U ηa ˆ U ζa ( ζ ) = z ζa ( ζ )U ζa
(6125) (6126) (6127) (6128) (6129) (6130)
where u is the displacement of a generic point, u is the displacement of a point on the reference surface, and U is the ‘fiber displacement’ rotations. The motion of the fibers can be interpreted as either displacements or rotations as will be discussed. Hughes and Liu introduced the notation that follows, and the associated schematic shown in Figure 68, to describe the current deformed configuration with respect to the reference configuration:
y = y+Y y = x+u ya = xa + ua Y = X+U Ya = Xa + Ua ˆ ˆ ˆ Y ηa = X ηa + U ηa ˆ ˆ ˆ Y ζa = X ζa + U ζ a
(6131) (6132) (6133) (6134) (6135) (6136) (6137)
In the above relations, and in Figure 68, the x quantities refer to the reference configuration, the y quantities refer to the updated (deformed) configuration and the u quantities are the displacements. The notation consistently uses a superscript bar ( ) to indicate reference surface quantities, a superscript caret to indicate unit vector quantities, lower case letter for translational displacements, and upper case letters for fiber displacements. Thus to update to the deformed configuration, two vector quantities are needed: the reference surface displacement u and the associated nodal fiber displacement U . The nodal fiber displacements are defined in the fiber coordinate system, described in the next subsection.
(ˆ )
Main Index
Chapter 6: Elements 167
CBEAM  DYSHELFORM = 1, HughesLiu Beam
u X Parallel Construction Reference axis in undeformed geometry u y
U
x
Y
x
y Deformed Configuration Reference Surface
Figure 68
Schematic of Deformed Configuration Displacements and Position Vectors
Fiber Coordinate System
For a beam element, the known quantities will be the displacements of the reference surface u obtained from the translational equations of motion and the rotational quantities at each node obtained from the rotational equations of motion. What remains to complete the kinematics is a relation between nodal rotations and fiber displacements U . The linearized relationships between the incremental components
ˆ ΔU
the incremental rotations are given by
⎫ ⎪ ⎪ ⎬ = ⎪ ⎪ ⎭ ⎫ ⎪ ⎪ ⎬ = ⎪ ⎪ ⎭ ˆ ˆ 0 Y η3 – Y η2 ˆ ˆ – Y η3 0 Y η1 ˆ ˆ Y η2 – Y η1 0 ⎧ ⎪ Δ θ1 ⎪ ⎨ Δ θ2 ⎪ ⎪ Δ θ3 ⎩ ⎧ ⎪ Δθ 1 ⎪ ⎨ Δθ 2 ⎪ ⎪ Δθ 3 ⎩ ⎫ ⎪ ⎪ ⎬ = hη Δ θ ⎪ ⎪ ⎭ ⎫ ⎪ ⎪ ⎬ = hζ Δ θ ⎪ ⎪ ⎭
⎧ ˆ ⎪ Δ U η1 ⎪ ˆ ⎨ Δ U η2 ⎪ ˆ ⎪ Δ U η3 ⎩ ⎧ ˆ ⎪ Δ Uζ 1 ⎪ ˆ ⎨ Δ Uζ 2 ⎪ ˆ ⎪ Δ Uζ 3 ⎩
(6138)
ˆ ˆ 0 Yζ 3 –Yζ 2 ˆ ˆ – Y ζ3 0 Y ζ1 ˆ ˆ Y ζ2 – Y ζ1 0
(6139)
Equations (6138) and (6139) are used to transform the incremental fiber tip displacements to rotational increments in the equations of motion. The secondorder accurate rotational update formulation due to Hughes and Winget [1980] is used to update the fiber vectors:
ˆn+1 ˆn Y ηi = R i j ( Δ θ )Y ηj ˆn ˆn+1 Y ζi = R i j ( Δ θ )Y ζi
(6140) (6141)
Main Index
168 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CBEAM  DYSHELFORM = 1, HughesLiu Beam
then
ˆ ˆn+1 ˆn Δ U ηa = Y ηa – Y ηa ˆ ˆn+1 ˆn Δ U ζa = Y ζa – Y ζ a
(6142) (6143)
where
( 2 δ ik + Δ S i k )Δ S j k R ij ( Δ θ ) = δ i j + 2D Δ S i j = e ikj Δ θ k 1 2 2 2 2D = 2 +  ( Δ θ 1 + Δ θ 2 + Δ θ 3 ) 2
(6144) (6145) (6146)
e ikj
Here
δi j
is the Kronecker delta and
is the permutation tensor.
Local Coordinate System In addition to the above described fiber coordinate system, a local coordinate system is needed to enforce ˆ the zero normal stress conditions transverse to the axis. The orthonormal basis with two directions e 2 and
ˆ e3
normal to the axis of the beam is constructed as follows: (6147) (6148)
y2 – y 1 ˆ e 1 = y2 – y 1 ˆ ˆ Y η1 + Y η2 ′ e 2 = ˆ ˆ Y η1 + Y η2
From the vector cross product of these local tangents.
ˆ ˆ ′ e3 = e1 × e2
(6149)
and to complete this orthonormal basis, the vector
ˆ ˆ ˆ e2 = e3 × e1
(6150)
is defined. This coordinate system rigidly rotates with the deformations of the element. The transformation of vectors from the global to the local coordinate system can now be defined in terms of the basis vectors as
⎧ ⎪ ⎪ ˆ A = ⎨ ⎪ ⎪ ⎩ ˆ Ax ˆ Ay ˆ Az ⎫ ⎪ ⎪ ⎬ = ⎪ ⎪ ⎭ e 1x e 1y e 1z e 2x e 2y e 2z e 3x e 3y e 3z
T⎧
⎫ ⎪ Ax ⎪ ⎪ ⎪ ⎨ Ay ⎬ = [ q ] { A } ⎪ ⎪ ⎪ Az ⎪ ⎩ ⎭
(6151)
Main Index
Chapter 6: Elements 169
CBEAM  DYSHELFORM = 1, HughesLiu Beam
where e ix, e iy, e iz are the global components of the local coordinate unit vectors, coordinates, and A is the same vector in the global coordinate system.
ˆ A
is a vector in the local
Strains and Stress Update
Incremental Strain and Spin Tensors The strain and spin increments are calculated from the incremental displacement gradient
∂Δ u i G i j = ∂y j
(6152)
where Δ u i are the incremental displacements and y j are the deformed coordinates. The incremental strain and spin tensors are defined as the symmetric and skewsymmetric parts, respectively, of G i j :
1 Δ ε i j =  ( G ij + G ji ) 2 1 Δ ω ij =  ( G i j – G j i ) 2
(6153) (6154)
The incremental spin tensor Δ ω ij is used as an approximation to the rotational contribution of the Jaumann rate of the stress tensor. Here the Jaumann rate update is approximated as
n σ i j = σ inj + σ inp Δ ω pj + σ jp Δ ω pi
(6155)
where the superscripts on the stress tensor refer to the updated ( n + 1 ) and reference ( n ) configurations. This update of the stress tensor is applied before the constitutive evaluation, and the stress and strain are stored in the global coordinate system. Stress Update To evaluate the constitutive relation, the stresses and strain increments are rotated from the global to the local coordinate system using the transformation defined previously in (6151).
σ i j = q ik σ kn q j n Δ ε i j = q i k Δ ε kn q jn
l ln
(6156) (6157)
where the superscript l indicates components in the local coordinate system. The stress is incrementally updated:
n+1 n 1 n + 2
l σi j
=
l σ ij
+
l Δ σi j
(6158)
Main Index
170 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CBEAM  DYSHELFORM = 1, HughesLiu Beam
and rotated back to the global system:
σi j
n+1
= q ki σ kn
l
n+1
q nj
(6159)
before computing the internal force vector. Incremental StrainDisplacement Relations After the constitutive evaluation is completed, the fully updated stresses are rotated back to the global coordinate system. These global stresses are then used to update the internal force vector
fa
int
=
∫ Ba σ dυ
T
(6160)
where f ain t are the internal forces at node a and B a is the straindisplacement matrix in the global coordinate system associated with the displacements at node a . The B matrix relates six global strain components to eighteen incremental displacements [three translational displacements per node and the six incremental fiber tip displacements of (6142)]. It is convenient to partition the B matrix:
B = [ B 1, B 2 ]
(6161)
Each B a sub matrix is further partitioned into a portion due to strain and spin with the following sub matrix definitions:
B1 0 Ba = 0 B2 0 B3 0 B2 0 B1 B3 0 0 0 B3 0 B2 B1 B4 0 0 B5 0 B6 0 B5 0 B4 B6 0 0 0 B6 0 B5 B4 B7 0 0 B8 0 B9 0 B8 0 B7 B9 0 0 0 B9 0 B8 B7
(6162)
where
∂N a N a, i = ∂y i Bi = ∂ ( N a z ηa ) ( N aηa ) ,( i – 3 ) = ∂y i – 3 ∂ ( N a z ηa ) ( N a z ζa ) ,( i – 6 ) = ∂y i – 6
for i = 1,2,3 for i = 4,5,6 for i = 7,8,9 (6163)
With respect to the straindisplacement relations, note that:
• The derivative of the shape functions are taken with respect to the global coordinates; • The B matrix is computed on the crosssection located at the midpoint of the axis; • The resulting B matrix is a 6 × 18 matrix.
Main Index
Chapter 6: Elements 171
CBEAM  DYSHELFORM = 1, HughesLiu Beam
The internal force, f , given by
i f ′ = T t f ant
(6164)
T
is assembled into the global right hand side internal force vector. (6138)):
I 0 0 0 hη hζ I
is defined as (also see equation
T =
(6165)
where
is a
3×3
identity matrix.
Main Index
172 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CQUADA  DYSHELLFORM = 2, BelytschkoLinTsay Shell
CQUADA  DYSHELLFORM = 2, BelytschkoLinTsay Shell
The BelytschkoLinTsay shell element ([Belytschko and Tsay 1981], [Belytschko et al., 1984a]) was implemented as a computationally efficient alternative to the HughesLiu shell element. For a shell element with five through thickness integration points, the BelytschkoLinTsay shell elements requires 725 mathematical operations compared to 4050 operations for the under integrated HughesLiu element. The selectively reduced integration formulation of the explicit HughesLiu element requires 35,350 mathematical operations. Because of its computational efficiency, the BelytschkoLinTsay shell element is usually the shell element formulation of choice. For this reason, it has become the default shell element formulation for explicit calculations. The BelytschkoLinTsay shell element is based on a combined corotational and velocitystrain formulation. The efficiency of the element is obtained from the mathematical simplifications that result from these two kinematical assumptions. The corotational portion of the formulation avoids the complexities of nonlinear mechanics by embedding a coordinate system in the element. The choice of velocitystrain or rateofdeformation in the formulation facilitates the constitutive evaluation, since the conjugate stress is the physical Cauchy stress. We closely follow the notation of Belytschko, Lin, and Tsay in the following development.
Corotational Coordinates
The midsurface of the quadrilateral shell element, or reference surface, is defined by the location of the element’s four corner nodes. An embedded element coordinate system (see Figure 69) that deforms with the element is defined in terms of these nodal coordinates. Then the procedure for constructing the corotational coordinate system begins by calculating a unit vector normal to the main diagonal of the element:
s3 ˆ e 3 = s3 s3 =
2 2 2 s 31 + s 32 + s 33
(6166) (6167) (6168)
(ˆ )
s 3 = r 31 × r 42
where the superscript caret
is used to indicate the local (element) coordinate system.
Main Index
Chapter 6: Elements 173
CQUADA  DYSHELLFORM = 2, BelytschkoLinTsay Shell
ˆ y
4
s3 3
ˆ e
r42
3
r31
ˆ e
2 1 s1
ˆ x
2
ˆ e
1
r21
Figure 69
Construction of Element Coordinate System
ˆ It is desired to establish the local x axis x approximately along the element edge between nodes 1 and 2. ˆ ˆ This definition is convenient for interpreting the element stresses, which are defined in the local x – y coordinate system. The procedure for constructing this unit vector is to define a vector s 1 that is nearly parallel to the vector r 21 . ˆ ˆ s 1 = r 21 – ( r 21 ⋅ e 3 ) e 3 s1 ˆ e 1 = s1
(6169) (6170)
The remaining unit vector is obtained from the vector cross product
ˆ ˆ ˆ e2 = e3 × e1
(6171)
ˆ e1
If the four nodes of the element are coplanar, then the unit vectors of the shell and
ˆ e3
and
ˆ e2
are tangent to the midplane
is in the fiber direction. As the element deforms, an angle may develop between the
ˆ e3 .
actual fiber direction and the unit normal
ˆ e3 ⋅ f – 1 < δ
The magnitude of this angle may be characterized as (6172)
where f is the unit vector in the fiber direction and the magnitude of δ depends on the magnitude of the strains. According to Belytschko et al., for most engineering applications, acceptable values of δ are on the order of 102 and if the condition presented in equation (6172) is met, then the difference between ˆ the rotation of the corotational coordinates e and the material rotation should be small.
Main Index
174 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CQUADA  DYSHELLFORM = 2, BelytschkoLinTsay Shell
The global components of this corotational triad define a transformation matrix between the global and local element coordinate systems. This transformation operates on vectors with global components
A = ( A x, A y, A z ) ⎧ ⎫ ⎪ Ax ⎪ ⎪ ⎪ { A } = ⎨ Ay ⎬ = ⎪ ⎪ ⎪ Az ⎪ ⎩ ⎭
and element coordinate components
⎧ e 1x e 2x e 3x ⎪ ⎪ e 1y e 2y e 3y ⎨ ⎪ e 1z e 2z e 3z ⎪ ⎩ ˆ Ax ˆ Ay ˆ Az
ˆ ˆ ˆ ˆ A = ( A x, A y, A z ) ,
and is defined as:
⎫ ⎪ ⎪ ˆ ˆ ⎬ = [ μ] { A } = [ q ]T { A } ⎪ ⎪ ⎭
(6173)
where e ix, e iy, e iz are the global components of the element coordinate unit vectors. The inverse transformation is defined by the matrix transpose; i.e.,
ˆ {A} = [μ]T{A}
(6174)
VelocityStrain Displacement Relations
The above small rotation condition, equation (6172), does not restrict the magnitude of the element’s rigid body rotations. Rather, the restriction is placed on the outofplane deformations, and, thus, on the element strain. Consistent with this restriction on the magnitude of the strains, the velocitystrain displacement relations used in the BelytschkoLinTsay shell are also restricted to small strains. As in the HughesLiu shell element, the displacement of any point in the shell is partitioned into a midsurface displacement (nodal translations) and a displacement associated with rotations of the element’s fibers (nodal rotations). The BelytschkoLinTsay shell element uses the Mindlin [1951] theory of plates and shells to partition the velocity of any point in the shell as:
ˆ v = vm – z e3 × θ
(6175)
ˆ where v m is the velocity of the midsurface, θ is the angular velocity vector, and z is the distance along the fiber direction (thickness) of the shell element. The corresponding corotational components of the velocity strain (rate of deformation) are given by ˆ ˆ ˆ 1 ⎛ ∂υ i ∂υ j⎞ d i j =  ⎜  +  ⎟ ˆ ˆ 2 ⎝ ∂x j ∂x i ⎠
(6176)
Substitution of equation (6175) into the above yields the following velocitystrain relations:
ˆ ˆm ∂v x ∂θ ˆ  ˆ y d x =  + z ˆ ˆ ∂x ∂x ˆm ˆ ∂υ y ˆ ∂θ x ˆ d y =  – z ˆ ˆ ∂y ∂y ˆm ˆ ˆm ˆ ∂υ x ∂υ y ˆ ∂θ y ∂θ x ˆ 2d xy =  +  + z ⎛  – ⎞ ˆ ˆ ˆ ˆ ⎝ ∂y ∂y ∂x ∂x ⎠
(6177)
(6178)
(6179)
Main Index
Chapter 6: Elements 175
CQUADA  DYSHELLFORM = 2, BelytschkoLinTsay Shell
ˆm ∂υ z ˆ ˆ 2d yz =  – θ x ˆ ∂y ˆm ∂υ z ˆ ˆ 2d xz =  + θ y ˆ ∂x
(6180)
(6181)
The above velocitystrain relations need to be evaluated at the quadrature points within the shell. Standard bilinear nodal interpolation is used to define the midsurface velocity, angular velocity, and the element’s coordinates (isoparametric representation). These interpolations relations are given by
ν m = N I ( ξ, η )ν I θ m = N I ( ξ, η )θ I x m = N I ( ξ, η )x I
(6182) (6183) (6184)
where the subscript I is summed over all the nodes of the element and the nodal velocities are · obtained by differentiating the nodal coordinates with respect to time, i.e., υ I = x I . The bilinear shape functions are
1 N 1 =  ( 1 – ξ ) ( 1 – η ) 4 1 N 2 =  ( 1 + ξ ) ( 1 – η ) 4 1 N 3 =  ( 1 + ξ ) ( 1 + η ) 4 1 N 4 =  ( 1 – ξ ) ( 1 + η ) 4
(6185) (6186) (6187) (6188)
The velocitystrains at the center of the element (i.e., at ξ = 0 , and η = 0 ) are obtained by substitution of the above relations into the previously defined velocitystrain displacement relations, equations (6177) through (6181). After some algebra, this yields
ˆ ˆ ˆ ˆ d x = B 1I υ xI + z B 1I θ yI ˆ ˆ ˆ ˆ d y = B 2I υ yI – z B 2I θ xI ˆ ˆ ˆ ˆ ˆ ˆ 2d xy = B 2I υ xI + B 1I υ yI + z ( B 2I θ yI – B 1 I θ xI ) ˆ ˆ ˆ 2d xz = B 1I υ zI + N I θ yI ˆ ˆ ˆ 2d yz = B 2I υ zI – N I θ xI
(6189) (6190) (6191) (6192) (6193)
Main Index
176 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CQUADA  DYSHELLFORM = 2, BelytschkoLinTsay Shell
where
∂N I B 1I = ˆ ∂x ∂N I B 2I = ˆ ∂y
(6194) (6195)
B aI
The shape function derivatives η = 0.
are also evaluated at the center of the element; i.e., at
ξ = 0,
and
Stress Resultants and Nodal Forces
After suitable constitutive evaluations using the above velocitystrains, the resulting stresses are integrated through the thickness of the shell to obtain local resultant forces and moments. The integration formula for the resultants are
ˆ f
R αβ
=
ˆ ∫ σ αβ dz
ˆ
(6196) (6197)
ˆR ˆˆ ˆ m αβ = – ∫ z σ αβ dz
where the superscript, R , indicates a resultant force or moment, and the Greek subscripts emphasize the limited range of the indices for plane stress plasticity. The above element midplane force and moment resultants are related to the local nodal forces and moments by invoking the principle of virtual power and integrating with a onepoint quadrature. The relations obtained in this manner are
ˆ ˆR ˆR f xI = A ⎛B 1I f xx + B 2I f xy ⎞ ⎝ ⎠ ˆ ˆR ˆR f yI = A ⎛B 2I f yy + B 1I f xy ⎞ ⎝ ⎠ ˆ ˆR ˆR f zI = A κ ⎛B 1I f xz + B 2I f yz⎞ ⎝ ⎠ ˆ ˆR ˆ R κ ˆ R⎞ m xI = A ⎛ B 2I m yy + B 1I m xy –  f yz ⎝ 4 ⎠ ˆR ˆ R κ ˆ R⎞ ˆ m yI = – A ⎛ B 1I m xx + B 2I m xy –  f xz ⎝ 4 ⎠ ˆ m zI = 0
(6198) (6199) (6200) (6201) (6202) (6203)
Main Index
Chapter 6: Elements 177
CQUADA  DYSHELLFORM = 2, BelytschkoLinTsay Shell
where A is the area of the element, and κ is the shear factor from the Mindlin theory. In the BelytschkoLinTsay formulation, κ is used as a penalty parameter to enforce the Kirchhoff normality condition as the shell becomes thin. The above local nodal forces and moments are then transformed to the global coordinate system using the transformation relations given previously as equation (6173). The global nodal forces and moments are then appropriately summed over all the nodes and the global equations of motion are solved for the next increment in nodal accelerations.
Hourglass Control (BelytschkoLinTsay)
In part, the computational efficiency of the BelytschkoLinTsay and the under integrated HughesLiu shell elements are derived from their use of onepoint quadrature in the plane of the element. To suppress the hourglass deformation modes that accompany onepoint quadrature, hourglass viscosity stresses are added to the physical stresses at the local element level. The discussion of the hourglass control that follows pertains to the HughesLiu and the membrane elements as well. The hourglass procedure is controlled by the DYHRGIHQ PARAM. The hourglass control used by Belytschko et al., extends an earlier derivation by Flanagan and Belytschko [1981], (see also Kosloff and Frazier [1978], Belytschko and Tsay [1983]). The hourglass shape vector, τ I , is defined as
ˆ τ I = h I – ( h J x aJ ) B aI
(6204)
where
+1 –1 +1 –1
h =
(6205)
is the basis vector that generates the deformation mode that is neglected by onepoint quadrature. In equation (6204) and the reminder of this subsection, the Greek subscripts have a range of 2; e.g., ˆ ˆ ˆ ˆ x aI = ( x 1I, x 2I ) = ( x I, y I ) . The hourglass shape vector then operates on the generalized displacements, in a manner similar to equations (6189) through (6193), to produce the generalized hourglass strain rates
ˆ ·B q α = τ I θ αI ˆ ·B q = τ I υ zI
3
(6206) (6207) (6208)
ˆ ·M q α = τ I υ αI
where the superscripts B and M denote bending and membrane modes, respectively. The corresponding hourglass stress rates are then given by
Main Index
178 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CQUADA  DYSHELLFORM = 2, BelytschkoLinTsay Shell
rθ E t3 A ·B ·B Q α =  B βI B βI q α 192 r w κ Gt 3 A ·B ·B Q =  B βI B βI q 3 3 12 rm E t A ·M ·M Q α =  B βI B βI q α 8
(6209) (6210) (6211)
r θ, r w ,
where t is the shell thickness and the parameters, 0.01 and 0.05.
and
rm
are generally assigned values between
Finally, the hourglass stresses, which are updated from the stress rates in the usual way; i.e.,
· Qn + 1 = Qn + Δ t Q
(6212)
and the hourglass resultant forces are then
B ˆH m αI = τ I Q α
(6213) (6214) (6215)
ˆ H = τ QB f 3I I 3 ˆ H = τ QM f αI I α
where the superscript H emphasizes that these are internal force contributions from the hourglass deformations. These hourglass forces are added directly to the previously determined local internal forces due to deformations equations (6198) through (6203). These force vectors are orthogonalized with respect to rigid body motion.
Hourglass Control (Englemann and Whirley)
Englemann and Whirley [1991] developed an alternative hourglass control, which they implemented in the framework of the Belytschko, Lin, and Tsay shell element. We will briefly highlight their procedure here that has proven to be cost effectiveonly twenty percent more expensive than the default control. In the hourglass procedure, the inplane strain field (subscript p ) is decomposed into the one point strain field plus the stabilization strain field:
· ·0 ·s εp = εp + ε p
(6216)
where the stabilization strain field, which is obtained from the assumed strain fields of Pian and Sumihara [1984], is given in terms of the hourglass velocity field as
·s · · ε p = W m q m + zW b q b
(6217)
Main Index
Chapter 6: Elements 179
CQUADA  DYSHELLFORM = 2, BelytschkoLinTsay Shell
Here,
Wm
and
p
Wb
play the role of stabilization strain velocity operators for membrane and bending:
f ( ξ, η ) f ( ξ, η ) f 6 ( ξ, η ) f 1 ( ξ, η ) f 2 ( ξ, η ) f 3 ( ξ, η )
p p p p 5p 4p p
f ( ξ, η ) Wm = f 2 ( ξ, η ) f 3 ( ξ, η ) – f 4 ( ξ, η ) Wb = – f 5 ( ξ, η ) – f 6 ( ξ, η )
p p p p 1 p
(6218)
(6219)
where the terms f i ( ξ, η ) i = 1, 2, …, 6 , are rather complicated and the reader is referred to the reference [Englemann and Whirley, 1991]. To obtain the transverse shear assumed strain field, the procedure given in [Bathe and Dvorkin, 1984] is used. The transverse shear strain field can again be decomposed into the one point strain field plus the stabilization field:
· ·0 ·s εs = εs + εs
(6220)
that is related to the hourglass velocities by
·s · εs = Ws qs
(6221)
Ws
where the transverse shear stabilization strainvelocity operator
f ( ξ, η ) – g ξ g η g ξ g η
1 s f 2 ( ξ, s s s s s
is given by (6222)
Ws =
η)
1 s g ξ 4
2 s g η 4
3 s –g ξ 2 s 1
3 s g η 1
Again, the coefficients
f ( ξ, η )
and
gs
1
are defined in the reference.
In their formulation, the hourglass forces are related to the hourglass velocity field through an incremental hourglass constitutive equation derived from an additive decomposition of the stress into a “onepoint stress,” plus a “stabilization stress.” The integration of the stabilization stress gives a resultant constitutive equation relating hourglass forces to hourglass velocities. The inplane and transverse stabilization stresses are updated according to:
· s s s τs , n + 1 = τ s , n + Δ t cs Cs εs
(6223)
where the tangent matrix is the product of a matrix C , which is constant within the shell domain, and a scalar c that is constant in the plane but may vary through the thickness.
Main Index
180 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CQUADA  DYSHELLFORM = 2, BelytschkoLinTsay Shell
The stabilization stresses can now be used to obtain the hourglass forces:
Qm =
∫ ∫ ∫ ∫ ∫ ∫
h T 2 s W m τ p dA dz h – 2A
Qb =
h T 2 s W b τ p dA dz h – 2 A
(6224)
Qs =
h T 2 s W s τ s dA dz h – 2A
Main Index
Chapter 6: Elements 181
CQUAD4 DYSHELLFORM = 10, BelytschkoWongChiang Improvements
CQUAD4 DYSHELLFORM = 10, BelytschkoWongChiang Improvements
Since the BelytschkoTsay element is based on a perfectly flat geometry, warpage is not considered. Although this generally poses no major difficulties and provides for an efficient element, incorrect results in the twisted beam problem, See Figure 610, are obtained where the nodal points of the elements used in the discretization are not coplanar. The HughesLiu shell element considers nonplanar geometry and gives good results on the twisted beam, but is relatively expensive. The effect of neglecting warpage in typical a application cannot be predicted beforehand and may lead to less than accurate results, but the latter is only speculation and is difficult to verify in practice. Obviously, it would be better to use shells that consider warpage if the added costs are reasonable and if this unknown effect is eliminated. In this section, we briefly describe the simple and computationally inexpensive modifications necessary in the BelytschkoTsay shell to include the warping stiffness. The improved transverse shear treatment is also described which is necessary for the element to pass the Kirchhoff patch test. Readers are directed to the references [Belytschko, Wong, and Chang 1989, 1992] for an in depth theoretical background.
Twisted Beam Problem 30 29 24 Y Displacement (104) L = 12 b = 1.1 t = .32 twist = 90 degrees E = 29 000 000 ν = .22 20 16 12 8 4 0 HughesLiu BelytschkoWongChiang BelystchknoTsay Displacementtime History
0
5
10 time (ms)
15
19
Figure 610
The Twisted Beam Problem Fails with the BelytschkoTsay Shell Element
In order to include warpage in the formulation it is convenient to define nodal fiber vectors as shown in Figure 611. The geometry is interpolated over the surface of the shell from:
x = x m + ζp = ( x I + ζp I )N I ( ξ, η )
(6225)
where:
ζh ζ = 2
and
ζ
is a parametric coordinate which varies between 1 to +1.
Main Index
182 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CQUAD4 DYSHELLFORM = 10, BelytschkoWongChiang Improvements
p3
p1 p2 h
Figure 611
Nodal Fiber Vectors
p 1, p 2 ,
and
p3
where
h
is the Thickness
The in plane strain components are given by:
· ˆ ˆ d xx = b xI v xI + ζ ( b xI v xI + b xI p xI ) · ˆ ˆ d yy = b yI v yI + ζ ( b yI v yI + b yI p yI )
c c 1 ˆ · · ˆ ˆ ˆ d xy =  b xI v yI + b yI v xI + ζ ( b xI v yI + b xI p yI + b yI v xI + b yI p xI ) 2 c c
(6226) (6227) (6228)
The coupling terms are come in through vectors as:
⎧ c ⎪ b xI ⎨ ⎪ bc ⎩ yI ⎫ ⎪ ⎬ = ⎪ ⎭ p y2 – py 4 ˆ ˆ p x2 – px 4 ˆ ˆ p y 3 – py 1 ˆ ˆ p x 3 – px 1 ˆ ˆ py 4 – py 2 ˆ ˆ px 4 – px 2 ˆ ˆ
bi I
c
which is defined in terms of the components of the fiber
py 1 – p y 3 ˆ ˆ px 1 – p x 3 ˆ ˆ
(6229)
For a flat geometry the normal vectors are identical and no coupling can occur. Two methods are used by Belytschko for computing b i I and the reader is referred to his papers for the details. Both methods have been tested and comparable results were obtained. The transverse shear strain components are given as
ˆ γ xz = – N I ( ξ ,η ) θ y I ˆ ˆ γ yz = – N I ( ξ ,η ) θ x I ˆ
c
(6230) (6231)
where the nodal rotational components are defined as:
θ x I = ( e n ⋅ e x ) θ n + ( e n ⋅ e x )θ n ˆ ˆ ˆ
I I K K
(6232)
Main Index
Chapter 6: Elements 183
CQUAD4 DYSHELLFORM = 10, BelytschkoWongChiang Improvements
θ y I = ( e n ⋅ e y ) θ n + ( e n ⋅ e y )θ n ˆ ˆ ˆ
I
I
K
K
(6233)
The rotation
θn
I
comes from the nodal projection (6234)
I I 1 I 1 ˆ ˆ θ n =  ( θ nI + θ nJ ) +  ( υ zJ – υ zJ ) 2 L IJ
where the subscript n refers to the normal component of side I as seen in Figure 612 and L IJ is the length of side IJ .
ˆ y
K r LK en k eˆ y
en
i J K
I
eˆ X
Figure 612
Vector and Edge Definitions for Computing the Transverse Shear Strain Components
Main Index
184 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CTRIA3  DYSHELLFORM = 4, C0 Triangular Shell
CTRIA3  DYSHELLFORM = 4, C0 Triangular Shell
The C 0 shell element due to Kennedy, Belytschko, and Lin [1986] has been implemented as a computationally efficient triangular element complement to the BelytschkoLinTsay quadrilateral shell element ([Belytschko and Tsay 1981], [Belytschko et al., 1984a]). For a shell element with five throughthethickness integration points, the element requires 649 mathematical operations (the BelytschkoLinTsay quadrilateral shell element requires 725 mathematical operations) compared to 1417 operations for the MarchertasBelytschko triangular shell [Marchertas and Belytschko 1974] (referred to as the BCIZ [Bazeley, Cheung, Irons, and Zienkiewicz 1965] triangular shell element). Triangular shell elements are offered as optional elements primarily for compatibility with local user grid generation and refinement software. Many computer aided design (CAD) and computer aided manufacturing (CAM) packages include finite element mesh generators, and most of these mesh generators use triangular elements in the discretization. Similarly, automatic mesh refinement algorithms are typically based on triangular element discretization. Also, triangular shell element formulations are not subject to zero energy modes inherent in quadrilateral element formulations. The triangular shell element’s origins are based on the work of Belytschko et al., [Belytschko, Stolarski, and Carpenter 1984b] where the linear performance of the shell was demonstrated. Because the triangular shell element formulations parallels closely the formulation of the BelytschkoLinTsay quadrilateral shell element presented in the previous section, the following discussion is limited to items related specifically to the triangular shell element.
Corotational Coordinates
The midsurface of the triangular shell element, or reference surface, is defined by the location of the element’s three nodes. An embedded element coordinate system (see Figure 613) that deforms with the element is defined in terms of these nodal coordinates. The procedure for constructing the corotational coordinate system is simpler than the corresponding procedure for the quadrilateral, because the three nodes of the triangular element are guaranteed coplanar.
ˆ z ˆ y
3
ˆ e3
ˆ e2 ˆ x
2
1
ˆ e1
Figure 613
Local Element Coordinate System for
C0
Shell Element
Main Index
Chapter 6: Elements 185
CTRIA3  DYSHELLFORM = 4, C0 Triangular Shell
ˆ ˆ The local xaxis, x , is directed from node 1 to 2. The element’s normal axis, z , is defined by the vector ˆ with a vector constructed from node 1 to node 3. The local yaxis, y , ˆ cross product of a vector along x ˆ ˆ ˆ is defined by a unit vector cross product of e 3 with e 1 , which are the unit vectors in the z directions, respectively. As in the case of the quadrilateral element, this triad of corotational unit vectors defines a transformation between the global and local element coordinate systems (see equations (6173) and (6174)).
VelocityStrain Relations
As in the BelytschkoLinTsay quadrilateral shell element, the displacement of any point in the shell is partitioned into a midsurface displacement (nodal translations) and a displacement associated with rotations of the element’s fibers (nodal rotations). The KennedyBelytschkoLin triangular shell element also uses the Mindlin [Mindlin 1951] theory of plates and shells to partition the velocity of any point in the shell (recall equation (6175)):
ˆ ν = v m – z e3 × θ
(6235)
ˆ where ν m is the velocity of the midsurface, θ is the angular velocity vector, and z is the distance along the fiber direction (thickness) of the shell element. The corresponding corotational components of the velocity strain (rate of deformation) were given previously in equations (6189) through (6193).
Standard linear nodal interpolation is used to define the midsurface velocity, angular velocity, and the element’s coordinates (isoparametric representation). These interpolation functions are the area coordinates used in triangular element formulations. Substitution of the nodally interpolated velocity fields into the velocitystrain relations (see Belytschko et al., for details), leads to strain ratevelocity relations of the form
ˆ ˆ d = Bv
(6236)
ˆ d
where
are the velocity strains (strain rates), the elements of B are derivatives of the nodal interpolation
ˆ ν
functions, and the
are the nodal velocities and angular velocities.
B
It is convenient to partition the velocity strains and the contributions. The membrane relations are given by
matrix into membrane and bending
⎧ ˆ ⎪ dx ⎪ ˆ ⎨ dy ⎪ ˆ ⎪ 2 d xy ⎩
⎫ ⎪ ⎪ ⎬ ⎪ ⎪ ⎭
M
1 = ˆ ˆ x2 y3
ˆ y3 0 ˆ ˆ x3 – x2
0 ˆ ˆ x3 – x2 ˆ –y
3
ˆ y3 0 ˆ –x
0 ˆ –x
0
3
3
ˆ y3
0 ˆ x2
⎧ ⎪ ⎪ ⎪ 0 ⎪ ⎪ ˆ x2 ⎨ ⎪ 0 ⎪ ⎪ ⎪ ⎪ ⎩
ˆ υ x1 ˆ υ
y1
ˆ υ x2 ˆ υ
y2
ˆ υ x3 ˆ υ
y3
⎫ ⎪ ⎪ ⎪ ⎪ ⎪ ⎬ ⎪ ⎪ ⎪ ⎪ ⎪ ⎭
(6237)
Main Index
186 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CTRIA3  DYSHELLFORM = 4, C0 Triangular Shell
or
ˆM ˆ d = BM ν
(6238)
The bending relations are given by
⎧ ⎪ ⎪ ⎪ 0 ⎪ ⎪ 0 ⎨ ⎪ ˆ x2 ⎪ ⎪ ⎪ ⎪ ⎩ ˆ θ x1 ˆ θ y1 ˆ θ x2 ˆ θ y2 ˆ θ x3 ˆ θ y3 ⎫ ⎪ ⎪ ⎪ ⎪ ⎪ ⎬ ⎪ ⎪ ⎪ ⎪ ⎪ ⎭
⎧ ˆ ⎪ κx ⎪ ˆ ⎨ κy ⎪ ˆ ⎪ 2κ xy ⎩
⎫ ⎪ 0 ⎪ –1 ˆ ˆ =  x 3 – x 2 ⎬ ˆ ˆ x2 y3 ⎪ ˆ ⎪ y3 ⎭
ˆ –y3 0 ˆ ˆ x3 – x2
0 ˆ x3 ˆ –y
ˆ y3 0 ˆ –x
0 ˆ –x 0
2
(6239)
3
3
or
ˆ def ˆ κ M = BM θ
(6240)
The local element velocity strains are then obtained by combining the above two relations:
⎧ ˆ ⎪ dx ⎪ ˆ ⎨ dy ⎪ ˆ ⎪ 2d xy ⎩ ⎫⎛ ⎧ ˆ ⎪⎜ ⎪ dx ⎪⎜ ⎪ ˆ ⎬⎜ ⎨ dy ⎪⎜ ⎪ ˆ ⎪ ⎜ ⎪ 2d xy ⎭⎝ ⎩ ⎫ ⎧ ˆ ⎪ ⎪ κx ⎪ ˆ⎪ ˆ – z ⎨ κy ⎬ ⎪ ⎪ ˆ ⎪ ⎪ 2κ xy ⎭ ⎩
M
⎫⎞ ⎪⎟ ⎪⎟ ˆM ˆ ˆ ⎬⎟ = d – z κ ⎪⎟ ⎪⎟ ⎭⎠
(6241)
The remaining two transverse shear strain rates are given by
⎧ ˆ ⎪ 2d xz ⎨ ˆ ⎪ 2d yz ⎩ ⎫ ⎪ 1 ⎬ = ˆ ˆ 6 x2 y 3 ⎪ ⎭
2
ˆ –y3
ˆ ˆ ˆ y 3 ( 2x 2 + x 3 ) ˆ2 ˆ2 x 2 – x3
ˆ2 y3 ˆ2 ˆ ˆ –y 3 ( x2 + x 3 )
ˆ ˆ ˆ y3 ( 3 x2 – x3 ) ˆ ˆ ˆ x3 ( x 3 – 2 x2 )
0 ˆ ˆ – 3x 2 y 3
ˆ ˆ x2 y3 ˆ ˆ ˆ x 2 ( 2x 3 – x 2 )
ˆ ˆ ˆ y 3 ( x2 – 2 x2 )
(6242)
⎧ ⎪ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎪ ⎩ ˆ θ x1 ˆ θ y1 ˆ θ x2 ˆ θ y2 ˆ θ x3 ˆ θ y3 ⎫ ⎪ ⎪ ⎪ ⎪ ⎪ ⎬ ⎪ ⎪ ⎪ ⎪ ⎪ ⎭
def
Main Index
Chapter 6: Elements 187
CTRIA3  DYSHELLFORM = 4, C0 Triangular Shell
or
ˆS ˆ def d = BS θ
(6243)
All of the above velocitystrain relations have been simplified by using onepoint quadrature. In the above relations, the angular velocities
ˆ θ ˆ def θ
are the deformation component of the angular velocity
obtained by subtracting the portion of the angular velocity due to rigid body rotation; i.e., (6244)
ˆ def ˆ ˆ ri g θ = θ–θ
The two components of the rigid body angular velocity are given by
ˆ ˆ υ z1 – υ z2 ˆ rig θ y = ˆ x2 ˆ ˆ ˆ ˆ ˆ ˆ ( υ z3 – υ z1 )x 2 – ( υ z2 – υ z1 )x 3 ˆ rig θ x = ˆ ˆ x2 y 3
(6245)
(6246)
The first of the above two relations is obtained by considering the angular velocity of the local xaxis about the local yaxis. Referring to Figure 614, by construction nodes 1 and 2 lie on the local xaxis and ˆ ˆ the distance between the nodes is x 2 ; i.e., the x distance from node 2 to the local coordinate origin at ˆ node 1. Thus, the difference in the nodal z velocities divided by the distance between the nodes is an average measure of the rigid body rotation rate about the local yaxis.
ˆ z ˆ y
3
1
2
ˆ x
ˆ z ˆ y
3
ˆ x
1 2
Figure 614
Element Configurations with Node 3 Aligned with Node 1 (left) and Node 3 Aligned with Node 2 (right)
Main Index
188 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CTRIA3  DYSHELLFORM = 4, C0 Triangular Shell
The second relation is conceptually identical, but is implemented in a slightly different manner due to the arbitrary location of node 3 in the local coordinate system. Consider the two local element configurations shown in Figure 614. For the leftmost configuration, where node 3 is the local yaxis, the rigid body rotation rate about the local xaxis is given by
ˆ ˆ υ z3 – υ z1 ˆ rig θ x – lef t = ˆ y3
(6247)
and for the rightmost configuration the same rotation rate is given by
ˆ ˆ υ z3 – υ z2 ˆ rig θ x – ri ght = ˆ y3
(6248)
Although both of these relations yield the average rigid body rotation rate, the selection of the correct relation depends on the configuration of the element; i.e., on the location of node 3. Since every element in the mesh could have a configuration that is different in general from either of the two configurations shown in Figure 614, a more robust relation is needed to determine the average rigid body rotation rate about the local xaxis. In most typical grids, node 3 will be located somewhere between the two configurations shown in Figure 614. Thus, a linear interpolation between these two rigid body rotation ˆ rates was devised using the distance x 3 as the interpolant:
ˆ ˆ x3 x3 ˆ rig ˆ rig ˆ rig θ x = θ x – left ⎛ 1 – ⎞ + θ x – right ⎛ ⎞ ˆ ⎠ ˆ ⎝ ⎝ x 2⎠ x2
(6249)
Substitution of equations (6247) and (6248) into equation (6249) and simplifying produces the relations given previously as equation (6246).
Stress Resultants and Nodal Forces
After suitable constitutive evaluation using the above velocity strains, the resulting local stresses are integrated through the thickness of the shell to obtain local resultant forces and moments. The integration formulae for the resultants are
ˆR f αβ = ˆ ∫ σ αβ dz ˆ
(6250) (6251)
ˆ ˆR ˆˆ m αβ = – ∫ z σ αβ dz
where the superscript R indicates a resultant force or moment and the Greek subscripts emphasize the limited range of the indices for plane stress plasticity. The above element midplane force and moment resultant are related to the local nodal forces and moments by invoking the principle of virtual power and performing a onepoint quadrature. The relations obtained in this manner are
Main Index
Chapter 6: Elements 189
CTRIA3  DYSHELLFORM = 4, C0 Triangular Shell
⎧ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎩ ⎧ ⎪ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎪ ⎩
ˆ f x1 ˆ f y1 ˆ f x2 ˆ f y2 ˆ f x3 ˆ f y3
⎫ ⎪ ⎪ ⎪ ⎧ ⎪ ⎪ ⎪ ⎪ T ⎪ ⎬ = A BM ⎨ ⎪ ⎪ ⎪ ⎪ ⎩ ⎪ ⎪ ⎪ ⎪ ⎭
ˆR f xx ˆR f yy ˆR f xy
⎫ ⎪ ⎪ ⎬ ⎪ ⎪ ⎭
(6252)
⎫ ˆ m x1 ⎪ ⎪ ˆ ⎧ ˆR m y1 ⎪ ⎪ m xx ⎪ ˆ m x2 ⎪ T⎪ ˆR ⎬ = AB M ⎨ m yy ˆ ⎪ m y2 ⎪ ˆ ⎪ mR ⎪ ˆ ⎩ xy m x3 ⎪ ⎪ ˆ m y3 ⎪ ⎭ A
⎫ ⎧ ⎪ ⎪ T⎪ ⎬ + AB S ⎨ ⎪ ⎪ ⎩ ⎪ ⎭
ˆR ⎫ f xz ⎪ ⎬ ˆR ⎪ f yz ⎭
(6253)
where
is the area of the element
ˆ ˆ ( 2 A = x2 y 3 ) . ˆ ˆ ˆ ( f z3, f z2, f z 1 ) ,
ˆ The remaining nodal forces, the z component of the force solving the following equilibration equations ˆ ˆ ˆ ˆ ˆ m x1 + m x2 + m x3 + y 3 f z3 = 0 ˆ ˆ ˆ ˆ ˆ ˆ ˆ m y1 + m y2 + m y3 – x 3 f z3 – x 2 f z2 = 0 ˆ f
z1
are determined by successively
(6254) (6255) (6256)
ˆ +f
z2
ˆ +f
z3
= 0
which represent moment equilibrium about the local xaxis, moment equilibrium about the local yaxis, and force equilibrium in the local zdirection, respectively.
Main Index
190 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CTRIA3  DYSHELLFORM = 3, MarchertasBelytschko Triangular Shell (BCIZ)
CTRIA3  DYSHELLFORM = 3, MarchertasBelytschko Triangular Shell (BCIZ)
The MarchertasBelytschko [1974] triangular shell element, or the BCIZ triangular shell element, was developed in the same time period as the Belytschko beam element [Belytschko, Schwer, and Klein, 1977], see CBEAM  Belytschko Beam, forming the first generation of corotational structural elements developed by Belytschko and coworkers. Although the MarchertasBelytschko shell element is relatively expensive (i.e., the C 0 triangular shell element with five throughthethickness integration points requires 649 mathematical operations compared to 1,417 operations for the MarchertasBelytschko triangular shell), it is maintained in SOL 700 for compatibility with earlier user models. However, as the user community moves to application of the more efficient shell element formulations, the use of the MarchertasBelytschko triangular shell element will decrease. As mentioned above, the MarchertasBelytschko triangular shell has a common corotational formulation origin with the Belytschko beam element. The interested reader is referred to the beam element description, see Corotational Technique for details on the corotational formulation. In the next subsection a discussion of how the local element coordinate system is identical for the triangular shell and beam elements. The remaining subsections discuss the triangular element’s displacement interpolants, the strain displacement relations, and calculations of the element nodal forces.
Element Coordinates
ˆ ˆ ˆ Figure 615a shows the element coordinate system, ( x, y, z ) originating at Node 1, for the Marchertas
Belytschko triangular shell. The element coordinate system is associated with a triad of unit vectors ( e 1, e 2, e 3 ) the components of which form a transformation matrix between the global and local coordinate systems for vector quantities. The nodal or body coordinate system unit vectors b 1, b 2, b 3 are defined at each node and are used to define the rotational deformations in the element, see Corotational Technique. The unit normal to the shell element
e 3 = l 21 × l 31 e3
is formed from the vector cross product (6257)
where l 21 and l 31 are unit vectors originating at Node 1 and pointing towards Nodes 2 and 3, respectively (see Figure 615b).
Main Index
Chapter 6: Elements 191
CTRIA3  DYSHELLFORM = 3, MarchertasBelytschko Triangular Shell (BCIZ)
ˆ y
3
ˆ z
e3
e2
ˆ x
e1 1 b2 b3
2 b1
(a) Element and Body Coordinates
ˆ y
3
ˆ z
e3 1
l 31
α
ˆ x
α⁄2
g
β
l 21
2 (b) Construction of Element Coordinates
Figure 615
Construction of Local Element Coordinate System
Main Index
192 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CTRIA3  DYSHELLFORM = 3, MarchertasBelytschko Triangular Shell (BCIZ)
Next a unit vector, g , see Figure 615b, is assumed to be in the plane of the triangular element with its origin at Node 1 and forming an angle β with the element side between Nodes 1 and 2, i.e., the vector l 21 . The direction cosines of this unit vector are represented by the symbols ( g x, g y, g z ) . Since g is the unit vector, its direction cosines will satisfy the equation:
2 2 2 gx + gy + gz = 1
(6258)
e3
Also, since
g
and
are orthogonal unit vectors, their vector dot product must satisfy the equation (6259)
g
e 3x g x + e 3y g y + e 3z g z = 0
In addition, the vector dot product of the coplanar unit vectors
I 21x g x + I 21 y g y + I 21z g z = cos β
and
l 21
satisfies the equation (6260)
where
( l 21x, + l 21y, + l 21z )
are the direction cosines of
l 21 .
Solving this system of three simultaneous equation; i.e., equations (6258), (6259), and (6260), for the direction cosines of the unit vector g yields
g x = l 21x cos β + ( e 3y l 21z – e 3z l 21y ) sin β g y = l 21y cos β + ( e 3z l 21x – e 3x l 21z ) sin β g z = l 21z cos β + ( e 3x l 21y – e 3y l 21x ) sin β
(6261)
These equations provide the direction cosines for any vector in the plane of the triangular element that is oriented at an angle β from the element side between Nodes 1 and 2. Thus the unit vector components of e 1 , and e 2 are obtained by setting β = α ⁄ 2 and β = ( π + α ) ⁄ 2 in (6261), respectively. The angle α is obtained from the vector dot product of the unit vectors l 21 and l 31 ,
cos α = l 21 ⋅ l 31
(6262)
Displacement Interpolation
As with the other large displacement and small deformation corotational element formulations, the nodal displacements are separated into rigid body and deformation displacements,
u = u ri gi d + u def
(6263)
where the rigid body displacements are defined by the motion of the local element coordinate system, i.e., the corotational coordinates, and the deformation displacement are defined with respect to the corotational coordinates. The deformation displacement are defined by
Main Index
Chapter 6: Elements 193
CTRIA3  DYSHELLFORM = 3, MarchertasBelytschko Triangular Shell (BCIZ)
⎧ ⎪ ⎪ ⎨ ⎪ ⎪ ⎩
ˆ ux ⎫ ⎪ ˆ uy ⎪ ⎬  ⎪ ˆ ⎪ uz ⎭
def
φx =
m
⎧ δ m φ y ⎪ ⎨  ⎪ ˆ ⎩ θ f φz
⎫ ⎪ ⎬ ⎪ ⎭
(6264)
where
{ δ } T = { δ 12 δ 23 δ 31 }
(6265)
are the edge elongations and
ˆ ˆ ˆ ˆ ˆ ˆ ˆ { θ } = { θ 1x θ 1y θ 2x θ 2y θ 3x θ 3y }
(6266)
are the local nodal rotation with respect to the corotational coordinates. The matrices
φx , φy
m m
and φ z are the membrane and flexural interpolation functions, respectively. The
f
element’s membrane deformation is defined in terms of the edge elongations. Marchertas and Belytschko adapted this idea from Argyris et. al., [1964], where incremental displacements are used, by modifying the relations for total displacements,
2 ( x j i u j ix + y ji u ji y + z j i u j iz ) + u j ix + u j iy + u j iz δ i j = 0 l i j + l ij
2 2 2
(6267)
where
x ji = x j – x i ,
etc.
f
The nonconforming shape functions used for interpolating the flexural deformations, φ z were originally derived by Bazeley, Cheung, Irons, and Zienkiewicz [1965], called the BCIZ element. Explicit expressions for
φz
f
are quite tedious and are not given here. The interested reader is referred to Appendix
G in the original work of Marchertas and Belytschko [1974]. The local nodal rotations, which are interpolated by these flexural shape functions, are defined in a manner similar to those used in the Belytschko beam element. The current components of the original element normal are obtained from the relation
e3 = μ T λ e3
0 0
(6268)
λ
where
μ
and
are the current transformations between the global coordinate system and the element
0 (local) and body coordinate system, respectively. The vector e 3 is the original element unit normal expressed in the body coordinate system. The vector cross product between this currentoriginal unit normal and the current unit normal,
ˆ ˆ 0 e3 × e3 = θ x e 1 + θ y e 2
(6269)
Main Index
194 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CTRIA3  DYSHELLFORM = 3, MarchertasBelytschko Triangular Shell (BCIZ)
define the local nodal rotations as
ˆ ˆ0 θ x = – e3 y ˆ ˆ0 θ y = e 3x
(6270) (6271)
λ
Note that at each node the corresponding
transformation matrix is used in equation (6268).
StrainDisplacement Relations
MarchertasBelytschko impose the usual Kirchhoff assumptions that normals to the midplane of the element remain straight and normal, to obtain
∂ 2 uz ∂u x e xx =  – z ∂x ∂x 2 ∂ 2 uz ∂u y e yy =  – z ∂y ∂y 2 ∂ 2 uz ∂u x ∂u y 2e xy =  +  – 2z ∂y ∂x ∂x ∂y
(6272) (6273) (6274)
where it is understood that all quantities refer to the local element coordinate system. Substitution of equation (6264) into the above straindisplacement relations yields
f ˆ { ε} = [Em] {δ } – z [E ] {θ }
(6275)
where
{ ε } t = { ε xx, ε yy, 2 ε xy }
(6276)
with
∂φ xi ∂x [Em] =
m ∂φ yi ∂y m m ∂φ xi ∂φ yi  + ∂y ∂x m
(6277)
and
Main Index
Chapter 6: Elements 195
CTRIA3  DYSHELLFORM = 3, MarchertasBelytschko Triangular Shell (BCIZ)
∂ 2 φ zi ∂x 2 [E ] =
f
f
∂ 2 φ zi ∂y 2 ∂ 2 φ zi 2 ∂x ∂y
f
f
(6278)
Nodal Force Calculations
The local element forces and moments are found by integrating the local element stresses through the thickness of the element. The local nodal forces are given by
ˆ = f
∫ [ E m ] t σ dV
ˆ
(6279)
where
ˆT f = { f 12 f 23 f 31 } ˆ σ T = { σ xx σ yy σ xy }
(6280) (6281)
where the side forces and stresses are understood to all be in the local convected coordinate system. Similarly, the local moments are given by
f ˆ ˆ m T = – ∫ z [ E ] t σ dv
(6282)
where
ˆ ˆ ˆ ˆ ˆ ˆ ˆ m T = { m 1x m 1y m 2x m 2y m 3x m 3y }
(6283)
The throughthethickness integration portions of the above local force and moment integrals are usually performed with a 3 or 5point trapezoidal integration. A threepoint inplane integration is also used; it is, in part, this threepoint inplane integration that increases the operation count for this element over the C 0 shell, which used onepoint inplane integration with hourglass stabilization. The remaining transverse nodal forces are obtained from element equilibrium considerations. Moment equilibrium requires
⎧ ⎪ ⎨ ⎪ ⎩ ˆ f 2z ˆ f 3z ⎫ ˆ ˆ ⎪ 1 – x 3 y 3 ⎬ = 2 A x –y ˆ ˆ ⎪ 2 2 ⎭ A ⎧ ˆ ˆ ˆ ⎪ m 1x + m 2x + m 3x ⎨ ˆ ˆ ˆ ⎪ m 1y + m 2y + m 3y ⎩ ⎫ ⎪ ⎬ ⎪ ⎭
(6284)
where
is the area of the element. Next transverse force equilibrium provides (6285)
ˆ ˆ ˆ f 1z = – f 2z – f 3z
Main Index
196 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CTRIA3  DYSHELLFORM = 3, MarchertasBelytschko Triangular Shell (BCIZ)
The corresponding global components of the nodal forces are obtained from the following transformation
⎧ ⎪ fi x ⎪ ⎨ fi y ⎪ ⎪ f iz ⎩ ⎫ ⎧ ⎪ ⎪ x + u ij x f ij ⎪ i j ⎪ =  ⎨ y i j + u ij y ⎬ l ij ⎪ ⎪ ⎪ ⎪ z ij + u i jz ⎭ ⎩ ⎫ ⎧ ⎪ ⎪ x ik + u i kx ⎪ f ik ⎪ +  ⎨ y ik + u i ky ⎬ l ⎪ ik ⎪ ⎪ ⎪ z i k + u ikz ⎭ ⎩ ⎫ ⎧ ⎪ ⎪ e3 x ⎪ ˆ ⎪ + fi z ⎨ e3 y ⎬ ⎪ ⎪ ⎪ ⎪ e 3z ⎭ ⎩ ⎫ ⎪ ⎪ ⎬ ⎪ ⎪ ⎭
(6286)
Finally, the local moments are transformed to the body coordinates using the relation
⎧ ⎪ m ix ⎪ ⎨ m iy ⎪ ⎪ mi z ⎩ ⎫ ⎧ ⎪ ⎪ ⎪ Tμ⎪ ⎬ = λ ⎨ ⎪ ⎪ ⎪ ⎪ ⎭ ⎩ ⎫ ˆ mi x ⎪ ⎪ ˆ mi y ⎬ ⎪ ˆ m iz ⎪ ⎭
(6287)
Main Index
Chapter 6: Elements 197
CQUAD4  DYSHELL FORM = 1, HughesLiu Shell
CQUAD4  DYSHELL FORM = 1, HughesLiu Shell
The HughesLiu shell element formulation ([Hughes and Liu 1981a, b], [Hughes et al., 1981], [Hallquist et al., 1985]) has several desirable qualities:
• It is incrementally objective (rigid body rotations do not generate strains), allowing for the
treatment of finite strains that occur in many practical applications;
• It is simple, which usually translates into computational efficiency and robustness; • It is compatible with brick elements, because the element is based on a degenerated brick
element formulation. This compatibility allows many of the efficient and effective techniques developed for brick elements to be used with this shell element;
• It includes finite transverse shear strains; • A throughthethickness thinning option (see [Hughes and Carnoy 1981]) is also available when
needed in some shell element applications. The remainder of this section reviews the HughesLiu shell element (referred to by Hughes and Liu as the U1 element) which is a fournode shell with uniformly reduced integration, and summarizes the modifications to their theory as it is implemented in SOL 700. A detailed discussion of these modifications are presented in an article by Hallquist and Benson [1986].
Geometry
The HughesLiu shell element is based on a degeneration of the standard 8node brick element formulation, an approach originated by Ahmad et al. [1970]. Recall from the discussion of the solid elements the isoparametric mapping of the biunit cube:
x ( ξ ,η ,ζ ) = N a ( ξ ,η ,ζ )x a ( 1 + ξa ξ ) ( 1 + ηa η ) ( 1 + ζa ζ ) N a ( ξ ,η ,ζ ) = 8
(6288) (6289)
where x is an arbitrary point in the element, ( ξ, η, ζ ) are the parametric coordinates, x a are the global nodal coordinates of node a , and N a are the element shape functions evaluated at node a , i.e., ( ξ a, η a, ζ a ) are ( ξ, η, ζ ) evaluated at node a . In the shell geometry, planes of constant ζ will define the lamina or layers of the shell and fibers are defined by throughthethickness lines when both ξ and η are constant (usually only defined at the nodes and thus referred to as ‘nodal fibers’). To degenerate the 8node brick geometry into the 4node shell geometry, the nodal pairs in the ζ direction (through the shell thickness) are combined into a single node, for the translation degrees of freedom, and an inextensible nodal fiber for the rotational degrees of freedom. Figure 616 shows a schematic of the biunit cube and the shell element. The mapping of the biunit cube into the shell element is separated into two parts
x ( ξ ,η ,ζ ) = x ( ξ ,η ) + X ( ξ ,η ,ζ )
(6290)
Main Index
198 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CQUAD4  DYSHELL FORM = 1, HughesLiu Shell
where x denotes a position vector to a point on the reference surface of the shell and X is a position vector, based at point x on the reference, that defines the fiber direction through that point. In particular, if we consider one of the four nodes which define the reference surface, then
x ( ξ ,η ) = N a ( ξ ,η )x a X ( ξ ,η ,ζ ) = N a ( ξ ,η )X a ( ζ ) ζ
(6291) (6292)
Biunit Cube
ζ η
x
ξ
η
Shell Element
ξ
Nodal Fiber z+ x+
ζ
Top Surface +1
ˆ x
0 Reference Surface
x
x 1 Bottom Surface z
Figure 616
Mapping of the Biunit Cube into the HughesLiu Shell Element and Nodal Fiber Nomenclature
Main Index
Chapter 6: Elements 199
CQUAD4  DYSHELL FORM = 1, HughesLiu Shell
With this description, arbitrary points on the reference surface x are interpolated by the twodimensional shape function N ( ξ ,η ) operating on the global position of the four shell nodes that define the reference surfaces; i.e., x a . Points off the reference surface are further interpolated by using a onedimensional shape function along the fiber direction, i.e., X a ( ζ ) , where
ˆ X a ( ζ ) = z a ( ζ )X a z a ( ζ ) = N + ( ζ )z a + N  ( ζ ) z a (1 + ζ) N + ( ζ ) = 2 (1 – ζ) N  ( ζ ) = 2
+ 
(6293) (6294) (6295) (6296)
ˆ As shown in the lower portion of Figure 616, X a is a unit vector in the fiber direction and z ( ζ ) is a thickness function. (Thickness changes (see [Hughes and Carnoy 1981]) are accounted for by explicitly adjusting the fiber lengths at the completion of a time step based on the amount of straining in the fiber direction. Updates of the fiber lengths always lag one time step behind other kinematical quantities.)
The reference surface may be located at the midsurface of the shell or at either of the shell’s outer surfaces. This capability is useful in several practical situations involving contact surfaces, connection of shell elements to solid elements, and offsetting elements such as stiffeners in stiffened shells. The reference surface is located within the shell element by specifying the value of the parameter ζ (see lower portion of Figure 616). When ζ = – 1, 0, +1 , the reference surface is located at the bottom, middle, and top surface of the shell, respectively. The HughesLiu formulation uses two position vectors, in addition to ζ , to locate the reference surface
+ and define the initial fiber direction. The two position vectors x a and x a are located on the top and bottom surfaces, respectively, at node a . From these data the following are obtained:
1 + x a =  [ ( 1 – ζ ) x a + ( 1 + ζ ) ] x a 2
+ ( xa – xa ) ˆ X a = ha
(6297)
(6298) (6299) (6300) (6301)
1 + z a =  ( 1 – ζ )h a 2 1 z a = –  ( 1 + ζ )h a 2 ha =
+ xa – xa
where
.
is the Euclidean norm.
Main Index
200 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CQUAD4  DYSHELL FORM = 1, HughesLiu Shell
Kinematics
The same parametric representation used to describe the geometry of the shell element (i.e., reference surface and fiber vector interpolation) are used to interpolate the shell element displacement; i.e., an isoparametric representation. Again, the displacements are separated into the reference surface displacements and rotations associated with the fiber direction:
u ( ξ ,η ,ζ ) = u ( ξ ,η ) + U ( ξ ,η ,ζ ) u ( ξ ,η ) = N a ( ξ ,η )u a U ( ξ ,η ,ζ ) = N a ( ξ ,η )U a ( ζ ) ˆ Ua ( ζ ) = za ( ζ ) Ua
(6302) (6303) (6304) (6305)
where u is the displacement of a generic point; u is the displacement of a point on the reference surface, and U is the ‘fiber displacement’ rotations; the motion of the fibers can be interpreted as either displacements or rotations as will be discussed. Hughes and Liu introduce the notation that follows, and the associated schematic shown in Figure 617, to describe the current deformed configuration with respect to the reference configuration:
y = y+Y y = x+u ya = xa + ua Y = X+U Ya = Xa + Ua ˆ ˆ ˆ Ya = Xa + Ua
(6306) (6307) (6308) (6309) (6310) (6311)
u X Parallel Construction Initial Configuration Reference Surface u y
U
x
Y
x
y Deformed Configuration Reference Surface
Figure 617
Schematic of Deformed Configuration Displacements and Position Vectors
Main Index
Chapter 6: Elements 201
CQUAD4  DYSHELL FORM = 1, HughesLiu Shell
In the above relations, and in Figure 617, the x quantities refer to the reference configuration, the y quantities refer to the updated (deformed) configuration and the u quantities are the displacements. The notation consistently uses a superscript bar ( ) to indicate reference surface quantities, a superscript caret to indicate unit vector quantities, lower case letters for translational displacements, and upper case letters indicating fiber displacements. To update to the deformed configuration, two vector quantities are needed: the reference surface displacement u and the associated nodal fiber displacement U . The nodal fiber displacements are defined in the fiber coordinate system, described in the next subsection.
(ˆ )
Fiber Coordinate System
For a shell element with four nodes, the known quantities will be the displacements of the reference surface u obtained from the translational equations of motion and some rotational quantities at each node obtained from the rotational equations of motion. To complete the kinematics, we now need a relation between nodal rotations and fiber displacements U . At each node a unique local Cartesian coordinate system is constructed that is used as the reference frame for the rotation increments. The relation presented by Hughes and Liu for the nodal fiber displacements (rotations) is an incremental relation; i.e., it relates the current configuration to the last state, not to the initial configuration. Figure 618 shows two triads of unit vectors:
( b 1 , b2 , b3 )
f f f
comprising the
f ˆ Y = b3
orthonormal fiber basis in the reference configuration (where the fiber unit vector is now
( b1 , b 2 , b 3 )
) and
indicating the incrementally updated current configuration of the fiber vectors. The reference
Δθ1
triad is updated by applying the incremental rotations,
f f
and
Δθ2 ,
obtained from the rotational
equations of motion, to the fiber vectors ( b 1 and b 2 ) as shown in Figure 618. The linearized relationship between the components of given by
⎧ ˆ f ⎪ ΔU 1 ⎪ ⎪ ˆ f ⎨ ΔU 2 ⎪ ⎪ ˆ f ⎪ ΔU 3 ⎩ ⎫ ⎪ ⎪ ⎪ ⎬ = ⎪ ⎪ ⎪ ⎭ ˆ ΔU
in the fiber system,
ˆ f ˆ f ˆ f Δ U1 , Δ U2 , Δ U3
, and the incremental rotations is
–1 0 0
0 –1 0
⎧ ⎪ Δ θ1 ⎨ ⎪ Δ θ2 ⎩
⎫ ⎪ ⎬ ⎪ ⎭
(6312)
Although the above HughesLiu relation for updating the fiber vector enables a reduction in the number of nodal degrees of freedom from six to five, it is not implemented in SOL 700 because it is not applicable to beam elements.
Main Index
202 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CQUAD4  DYSHELL FORM = 1, HughesLiu Shell
Fiber b3 f ˆ b3 = Y
Δθ
2
f b1 b1
b2 f b2
Δθ
1
Figure 618
Incremental Update of Fiber Vectors Using HughesLiu Incremental Rotations
In SOL 700, three rotational increments are used, defined with reference to the global coordinate axes:
⎧ ˆ ⎪ ΔU 1 ⎪ ˆ ⎨ ΔU 2 ⎪ ˆ ⎪ ΔU 3 ⎩ ⎫ ⎪ ⎪ ⎬ = ⎪ ⎪ ⎭ 0 ˆ –Y3 ˆ Y2 ˆ Y3 0 ˆ –Y1 ˆ –Y2 ˆ Y1 0 ⎧ ⎪ Δ θ1 ⎪ ⎨ Δ θ2 ⎪ ⎪ Δ θ3 ⎩ ⎫ ⎪ ⎪ ⎬ ⎪ ⎪ ⎭
(6313)
Equation (6313) is adequate for updating the stiffness matrix, but for finite rotations the error is significant. A more accurate secondorder technique is used in SOL 700 for updating the unit fiber vectors:
ˆn ˆ n+1 Yi = R ij ( Δ θ ) Y j
(6314)
where
1 ( 2 δ ik + Δ S i k ) Δ S j k R ij ( Δ θ ) = δ i j +  2 D Δ S i j = e ikj Δ θ k
2 2 2 1 2D = 2 +  ( Δ θ 1 + Δ θ 2 + Δ θ 3 ) 2
(6315) (6316) (6317)
Here, δ i j is the Kronecker delta and e ij k is the permutation tensor. This rotational update is often referred to as the HughesWinget formula [Hughes and Winget 1980]. An exact rotational update using Euler angles or Euler parameters could easily be substituted in equation (6314), but it is doubtful that the extra effort would be justified.
Lamina Coordinate System
In addition to the above described fiber coordinate system, a local lamina coordinate system is needed to enforce the zero normal stress condition; i.e., plane stress. Lamina are layers through the thickness of the shell that correspond to the locations and associated thicknesses of the throughthethickness shell
Main Index
Chapter 6: Elements 203
CQUAD4  DYSHELL FORM = 1, HughesLiu Shell
integration points; the analogy is that of lamina in a fibrous composite material. The orthonormal lamina ˆ basis (Figure 619), with one direction e 3 normal to the lamina of the shell, is constructed at every integration point in the shell.
η = con
ˆ e3
co ns tan t
stant
η
ˆ e2
ξ=
ˆ e1
ξ
Figure 619
Schematic of Lamina Coordinate Unit Vectors
The lamina basis is constructed by forming two unit vectors locally tangent to the lamina:
y ,ξ ˆ e 1 = y ,ξ y ,η ′ e 2 = y ,η
(6318)
(6319)
where, as before, y is the position vector in the current configuration. The normal to the lamina at the integration point is constructed from the vector cross product of these local tangents:
ˆ ˆ ′ e3 = e1 × e2
(6320)
To complete this orthonormal lamina basis, the vector
ˆ ˆ ˆ e2 = e3 × e1
(6321)
ˆ e2 ,
is defined, because to
ˆ e1
although tangent to both the lamina and lines of constant ξ , may not be normal
and
ˆ e3 .
The lamina coordinate system rotates rigidly with the element.
The transformation of vectors from the global to lamina coordinate system can now be defined in terms of the lamina basis vectors as
⎧ ⎪ ˆ = ⎪ A ⎨ ⎪ ⎪ ⎩ ˆ Ax ˆ Ay ˆ Az ⎫ ⎪ ⎪ ⎬ = ⎪ ⎪ ⎭ e 1x e 1y e 1z e 2x e 2y e 2z e 3x e 3y e 3z
T⎧
⎫ ⎪ Ax ⎪ ⎪ ⎪ ⎨ Ay ⎬ = [ q ] { A } ⎪ ⎪ ⎪ Az ⎪ ⎩ ⎭
(6322)
Main Index
204 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CQUAD4  DYSHELL FORM = 1, HughesLiu Shell
where e ix, e iy, e iz are the global components of the lamina coordinate unit vectors; lamina coordinates, and A is the same vector in the global coordinate system.
ˆ A
is a vector in the
Strains and Stress Update
Incremental Strain and Spin Tensors The strain and spin increments are calculated from the incremental displacement gradient
∂Δ u i G i j = ∂y j
(6323)
where Δu i are the incremental displacements and y j are the deformed coordinates. The incremental strain and spin tensors are defined as the symmetric and skewsymmetric parts, respectively, of G i j :
1 Δ ε i j =  ( G ij + G ji ) 2 1 Δ ω ij =  ( G i j – G j i ) 2
(6324) (6325)
The incremental spin tensor Δ ω ij is used as an approximation to the rotational contribution of the Jaumann rate of the stress tensor; SOL 700 implicit uses the more accurate HughesWinget transformation matrix equation (6314) with the incremental spin tensor for the rotational update. The Jaumann rate update is approximated as:
n σ inj + 1 = σ inj + σ inp Δ ω pj + σ jp Δ ω pi
(6326)
where the superscripts on the stress refer to the updated ( n + 1 ) and reference ( n ) configurations. The Jaumann rate update of the stress tensor is applied in the global configuration before the constitutive evaluation is performed. In the HughesLiu shell the stresses and history variables are stored in the global coordinate system. Stress Update To evaluate the constitutive relation, the stresses and strain increments are rotated from the global to the lamina coordinate system using the transformation defined previously in equation (6322).
σ ilj
n+1 n = q i k σ kn+ 1 q jn
(6327) (6328)
Δ ε ilj
n+1⁄2
n = q i k Δ ε kn+ 1 ⁄ 2 q j n
where the superscript
l
indicates components in the lamina (local) coordinate system.
Main Index
Chapter 6: Elements 205
CQUAD4  DYSHELL FORM = 1, HughesLiu Shell
The stress is updated incrementally:
σ ilj
n+1 l = σ ij n+1 l + Δ σ ij n+1⁄2
(6329)
and rotated back to the global system:
σ ij
n+1
= q ki σ kn
l
n+1
q nj
(6330)
before computing the internal force vector. Incremental StrainDisplacement Relations The global stresses are now used to update the internal force vector
fa
i nt
=
∫ T a Ba σ dυ
T
T
(6331)
int where f a are the internal forces at node a , B a is the straindisplacement matrix in the lamina coordinate system associated with the displacements at node a , and T a is the transformation matrix
relating the global and lamina components of the straindisplacement matrix. Because the B matrix relates six strain components to twentyfour displacements (six degrees of freedom at four nodes), it is convenient to partition the B matrix into four groups of six:
B = [ B 1, B 2, B 3, B 4 ]
(6332)
Each
Ba
submatrix is further partitioned into a portion due to strain and spin:
ε
Ba =
Ba Ba
ω
(6333)
with the following submatrix definitions:
B1 0 Ba =
ε
0 B2 B1 B3 0
0 0 0 B2 B1 0
B4 0 B5 0 B6
0 B5 B4 B6 0
0 0 0 B5 B4 – B4 B6 0 0 –B5 B4
B2 0 B3 B2 0 –B3
(6334)
Ba =
ω
–B1 B3 0
B5 0 –B6
–B2 B1
(6335)
Main Index
206 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CQUAD4  DYSHELL FORM = 1, HughesLiu Shell
where
⎧ ∂N a ⎪ ⎪ N a ,i = l ∂y i ⎪ Bi = ⎨ ∂ ( Na za ) ⎪ ⎪ ( N a z a ), i – 3 = l ⎪ ∂y i – 3 ⎩ for i = 1, 2, 3
(6336)
for i = 4, 5, 6
Notes on straindisplacement relations:
• The derivatives of the shape functions are taken with respect to the lamina coordinate system,
e.g.,
y l = qy .
( 0, 0, ζ ) .
The superscript bar indicates the B ’s are evaluated at the center of the lamina The straindisplacement matrix uses the ‘BBar’ ( B ) approach advocated by Hughes
[1980]. In the SOL 700 implementations, this entails replacing certain rows of the B matrix and the strain increments with their counterparts evaluated at the center of the element. In particular, the straindisplacement matrix is modified to produce constant shear and spin increments throughout the lamina.
• The resulting B matrix is a 8 x 24 matrix. Although there are six strain and three rotations
increments, the B matrix has been modified to account for the fact that integration of equation (6331).
σ 33
will be zero in the
Main Index
Chapter 6: Elements 207
CQUAD4, DYSHELLFORM = 6, 7, Fully Integrated HughesLiu Shells
CQUAD4, DYSHELLFORM = 6, 7, Fully Integrated HughesLiu Shells
It is well known that onepoint integration results in zero energy modes that must be resisted. The fournode under integrated shell with six degrees of freedom per node has nine zero energy modes, six rigid body modes, and four unconstrained drilling degrees of freedom. Deformations in the zero energy modes are always troublesome but usually not a serious problem except in regions where boundary conditions such as point loads are active. In areas where the zero energy modes are a problem, it is highly desirable to provide the option of using the original formulation of HughesLiu with selectively reduced integration. The major disadvantages of full integration are twofold:
• nearly four times as much data must be stored; • the operation count increases three to fourfold. The level 3 loop is added as shown in Figure 620.
However, these disadvantages can be more than offset by the increased reliability and accuracy.
η
ξ
Figure 620
Selectively Reduced Integration Rule Results in Four Inplane Points Being Used
There are two version of the HughesLiu shell with selectively reduced integration. The first closely follows the intent of the original paper, and therefore no assumptions are made to reduce costs, which are outlined in operation counts in Table 64. This is activated by using DYSHELLFORM = 6. These operation counts can be compared with those in Table 65 for the HughesLiu shell with uniformly reduced integration. The second formulation, which reduces the number of operation by more than a factor of two, is referred to as the corotational HughesLiu shell. This is activated by using DYSHELLFORM = 7. This shell is considerably cheaper due to the following simplifications:
• Strains rates are not centered. The strain displacement matrix is only computed at time n + 1
and not at time
n+1⁄2
.
Main Index
208 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CQUAD4, DYSHELLFORM = 6, 7, Fully Integrated HughesLiu Shells
• The stresses are stored in the local shell system following the BelytschkoTsay shell.
The transformations of the stresses between the local and global coordinate systems are thus avoided.
• The Jaumann rate rotation is not performed, thereby avoiding even more computations. This
does not necessarily preclude the use of the shell in large deformations. To study the effects of these simplifying assumptions, we can compare results with those obtained with the full HughesLiu shell. Thus far, we have been able to get comparable results.
LEVEL L1  Do over each element group, gather data, midstep geometry calculations LEVEL 2  For each thickness integration point center of element calculations for selective reduced integration
LEVEL 3  Do over four Gauss points stress update and force contributions
LEVEL 2  Completion LEVEL L1  Completion
Figure 621
An Inner Loop, Level 3, is added for the HughesLiu Shell with Selectively Reduced Integration Operation Counts for the HughesLiu Shell with Selectively Reduced Integration LEVEL L1  Once per element
Table 64
Midstep translation geometry, etc. Midstep calculation of
ˆ Y
204 318
LEVEL L2  For each integration point through thickness (NT points) Strain increment at
( 0, 0, ζ )
316 33 47
HughesWinget rotation matrix Square root of HughesWinget matrix
Main Index
Chapter 6: Elements 209
CQUAD4, DYSHELLFORM = 6, 7, Fully Integrated HughesLiu Shells
Table 64
Operation Counts for the HughesLiu Shell with Selectively Reduced Integration (continued) Rotate strain increments into lamina coordinates Calculate rows 3 through 8 of B matrix 66 919
LEVEL L3  For each integration point in lamina Rotate stress to
n+1⁄2
configuration
75 370 75 55 model dependent 69 75 358 75 245 522 +NT {1381 +4 * 1397}
Incremental displacement gradient matrix Rotate stress to lamina system Rotate strain increments to lamina system Constitutive model Rotate stress back to global system Rotate stress to Stresses in
n+1
configuration
Calculate rows 1 and 2 of B matrix
n+1
lamina system
Stress divergence TOTAL Table 65
Operation Counts for the SOL 700 Implementation of the Uniformly Reduced HughesLiu Shell LEVEL L1  Once per element Calculate displacement increments Element areas for time step Calculate
ˆ Y
24 53 238
LEVEL L2 and L3  Integration point through thickness (NT points) Incremental displacement gradient matrix Jaumann rotation for stress Rotate stress into lamina coordinates Rotate stain increments into lamina coordinates Constitutive model Rotate stress to
n+1
284 33 75 81 model dependent 69 125
global coordinates
Stress divergence
Main Index
210 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CQUAD4, DYSHELLFORM = 6, 7, Fully Integrated HughesLiu Shells
Table 65
Operation Counts for the SOL 700 Implementation of the Uniformly Reduced HughesLiu Shell (continued) LEVEL L1  Cleanup Finish stress divergence Hourglass control 60 356 731 +NT * 667
TOTAL
Element Mass Matrix
Hughes, Liu, and Levit [Hughes et al., 1981] describe the procedure used to form the shell element mass matrix in problems involving explicit transient dynamics. Their procedure, which scales the rotary mass terms, is used for all shell elements in SOL 700 including those formulated by Belytschko and his coworkers. This scaling permits large critical time step sizes without loss of stability. The consistent mass matrix is defined by
M =
υm
∫ ρ N t N dυ m
(6337)
but cannot be used effectively in explicit calculations where matrix inversions are not feasible. In SOL 700 only three and fournode shell elements are used with linear interpolation functions; consequently, we compute the translational masses from the consistent mass matrix by row summing, leading to the following mass at element node a :
M di sp
a
=
∫ ρ φ a dυ
v
(6338)
α:
The rotational masses are computed by scaling the translational mass at the node by the factor
M rot
a
= ∞M disp
a
(6339)
where
∞ = ma x { ∞ 1, ∞ 2 } 1∞ 1 = 〈 z a〉 2 +  [ z a ] 2 12 V∞ 2 = 8h
+ ( za + za ) 〈 z a〉 = 2 + [ za ] = za – za
(6340) (6341) (6342) (6343) (6344)
and
V
and
h
are the volume and the thickness of the element, respectively.
Main Index
Chapter 6: Elements 211
Transverse Shear Treatment for Layered Shell
Accounting for Thickness Changes
Hughes and Carnoy [1981] describe the procedure used to update the shell thickness due to large membrane stretching. Their procedure with any necessary modifications is used across all shell element types in SOL 700. One key to updating the thickness is an accurate calculation of the normal strain component Δ ε 33 . This strain component is easily obtained for elastic materials but can require an iterative algorithm for nonlinear material behavior. In SOL 700, we, therefore, default to an iterative plasticity update to accurately determine Δ ε 33 . Hughes and Carnoy integrate the strain tensor through the thickness of the shell in order to determine a mean value Δ ε i j :
1 1 Δ ε i j =  ∫ Δ ε ij dζ 2 –1
(6345)
and then project it to determine the straining in the fiber direction:
ε
f
ˆT ˆ = Y Δ εi j Y
(6346)
Using the interpolation functions through the integration points the strains in the fiber directions are extrapolated to the nodal points if 2 x 2 selectively reduced integration is employed. The nodal fiber lengths can now be updated:
ha
n+1
= ha ( 1 + εa )
n
f
(6347)
Transverse Shear Treatment for Layered Shell
The shell element formulations that include the transverse shear strain components are based on the first order shear deformation theory, which yield constant through thickness transverse shear strains. This violates the condition of zero traction on the top and bottom surfaces of the shell. Normally, this is corrected by the use of a shear correction factor. The shear correction factor is 5/6 for isotropic materials; however, this value is incorrect for sandwich and laminated shells. Not accounting for the correct transverse shear strain and stress could yield a very stiff behavior in sandwich and laminated shells. This problem is addressed here by the use of the equilibrium equations without gradient in the ydirection as described by what follows. Consider the stresses in a layered shell:
σ σ
(i) x (i) y ° ° ° ° = C 11 ( ε x + zχ x ) + C 12 ( ε y + zχ y ) = C 11 ε x + C 12 ε y + z ( C 11 χ x + C 12 χ y ) ° ° = C 12 ε x + C 22 ε y + z ( C 12 χ x + C 22 χ y ) (i) (i) (i) (i ) (i) (i) (i) (i) (i) (i)
(6348)
(i) τ xy
=
(i ) ° C 44 ( ε xy
+ zχ xy ) zx
(i )
Assume that the bending center
σ
(i) x (i ) (i)
is known. Then
(i)
= ( z – z x ) ( C 11 χ x + C 12 χ y ) + C 11 ε x ( z x ) + C 12 ε y ( z x )
(6349)
Main Index
212 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Transverse Shear Treatment for Layered Shell
The bending moment is given by the following equation:
⎛ NL ( i ) = χ x ⎜ ∑ C 11 ⎜ ⎝i = 1 z
zi
M xx
∫
i–1
⎞ ⎛ NL ( i ) z 2 dz⎟ + χ y ⎜ ∑ C 12 ⎟ ⎜ ⎠ ⎝i = 1 z
zi
∫
i–1
⎞ z 2 dz⎟ ⎟ ⎠
(6350)
or
NL NL (i) 3 C 11 ( z i
M xx
1 =  χ x 3
∑
i =1
–
3 zi – 1 )
+ χy
∑ C 12 ( z i
i =1
(i)
3
– zi – 1 )
3
(6351)
where “ N L ” is the number of layers in the material. Assume
εy = 0
and
σ x = Eχ εχ ,
then (6352)
z – zx ε x =  = ( z – z x )χ x ρ
and
(i) 3 1 3 M xx =  ( χ x ∑ E x ( z i – z i – 1 ) ) 3
(6353) (6354)
3M xx χ x = (i) 3 3 E x ( zi – z i – 1 ) ∑
Therefore, the stress becomes
σx
(i)
3M xx E x ( z – z x ) = NL
(i )
(6355)
∑
i =1
(i) 3 Ex ( zi
–
3 zi – 1 )
Now considering the first equilibrium equation, one can write the following:
∂τ xz ∂σ x 3 Q xz E x ( z – z x )  = –  = – NL ∂z ∂x (i) 3 3 ∑ E x ( zi – zi – 1 )
i= 1 (j) z2 3Q xz E x ⎛  – zz x⎞ ⎝2 ⎠ = –  + C j NL (j)
(6356)
(j) τ xz
(6357)
∑
i =1
(i) 3 E x ( zi
–
3 zi – 1 )
where Q xz is the shear force and C j is the constant of integration. This constant is obtained from the transverse shear stress continuity requirement at the interface of each layer.
Main Index
Chapter 6: Elements 213
Transverse Shear Treatment for Layered Shell
Let
NL
( E I )x =
∑  E x ( z i 3
i =1
1
(i)
3
– zi – 1 )
3
(6358)
Then
( i ) zi – 1 3 Q xz E x ⎛  – z i – 1 z x⎞ ⎝ 2 ⎠ (i – 1 C j =  + τ xz ) ( EI ) x 2
(6359)
and
τ xz = τ xz
(i) (i – 1)
Q xz E x z i – 1 z2 +   – z i – 1 z x –  + zz x 2 2 ( EI ) x
(i)
2
(6360)
For the first layer
(1 )
3 Q xz C 11 z – zo τ xz = –   – z x ( z – z o ) NL 2 (i) 3 3 ∑ C11 ( z i + z i – 1 )
i= 1
2
2
(6361)
for subsequent layers
τ xz = τ xz
(i – 1)
3 Q xz C 11 z – zo –   – z x ( z – z o ) ,z i – 1 ≤ z ≤ z i NL 2 (j) 3 3 ∑ C 11 ( z j – z j – 1 )
j =1
(i)
2
2
(6362)
(i Here τ xz – 1 ) is the stress in previous layer at the interface with the current layer. The shear stress can also be expressed as follows: (i ) 2
3 Q xz C 11 τ xz =  f NL
(i) x+
∑
j =1
(j) 3 C 11 ( z j
z 2 – zi – 1  – z x ( z – z i – 1 ) 2
(6363)
–
3 zj – 1 )
where
f
(i) x
zj + zj + 1 (j) 1 = ) ∑ C 11 h j  – z x (i 2 C 11 j = 1
NL
(6364)
and
hj = zj – zj – 1
Main Index
214 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Transverse Shear Treatment for Layered Shell
To find Q xz , the shear force, assume that the strain energy expressed through average shear modules, C 66 , is equal to the strain energy expressed through the derived expressions as follows:
1 τ xz 1 Q xz U =   =  ∫  dz 2 C 66 2 C 66 h C ( i) ( z – zi – 1 ) 9h 1  =  ∫  f x +  – z x ( z – z i – 1 )  11 2 C 2 NL C 66 66 (i) 3 3 C 11 ( z i – z i – 1 ) ∑
j =1 NL (i) zi 2 2 2 2 2
(6365)
dz
9h = 2
NL
∑
j =1
(i) 3 C 11 ( z i
– zi – 1 )
3
( C 11 ) 2 ∑ (i) C 66 z i =1
∫
i–1
f
(i) x+
z – zi – 1  – z x ( z – z i – 1 ) 2
2
2
2
dz
(6366)
NL
9h 1 =   ∑ 2 60 NL
(i ∑ C11) ( z i3 – z i3– 1 ) j =1 i
(i) ( C 11 ) 2 h  { f ( i ) [ 60f ( i ) x x (i C 66)
+ 20 h i ( z i + 2 z i – 1 – 3z x ) ] +
z x h i [ 20z x h i + 35z i – 1 – 10z i – 1 ( z i + z i – 1 ) – 15 z i ] 1 2 2 4 +z i ( z i + z i – 1 ) ( 3 z i – 7z i – 1 ) + 8 z i – 1 } = C 66
2
2
then
Q xz = τ xz h = C 66 γ xz h
to calculate
NL
zx
2
use
τ xz
for last layer at surface
z = 0
∑ C11
i =1
(i)
⎛ z i – zi – 1 ⎞ – z ( z – z x i i – 1) ⎝ ⎠ 2
2
= 0
(6367)
where
NL
∑ C 11 h i ( z i + z i + 1 )
z x = i = 1 NL 2
(i )
(6368)
∑
i =1
(i) C 11 h i
Algorithm:
Main Index
Chapter 6: Elements 215
Transverse Shear Treatment for Layered Shell
The following algorithm is used in the implementation of the transverse shear treatment. 1. Calculate 2. Calculate 3. Calculate
zx f (i) x 1 3
according to (6368) according to (6364)
(i) 3
NL
∑ C11 ( z i
i =1
– zi – 1 )
2
3
4. Calculate 5. Calculate 6. Calculate 7. Calculate
(i) 1 τ xz = – 3
⎛ NL ⎞ 1 (i h ⎜  ∑ C 11) ( z i3 – z i3– 1 )⎟ ⎜3 ⎟ ⎝ i=1 ⎠ C 66
according to equation (6366)
66
Q xz = C γ xz h τ xz
NL
∑
i =1
C 11 ( z i – z i – 1 ) Q xz C 11 f
(i )
3
3
(i)
(i) x+
z – zi – 1  – z x ( z – z i – 1 ) , z i – 1 ≤ z ≤ z i 2
2
2
Steps 15 are performed at the initialization stage. Step 6 is performed in the shell formulation subroutine, and Step 7 is performed in the stress calculation inside the constitutive subroutine.
Main Index
216 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CROD, Truss Element
CROD, Truss Element
Note: Unlike the CBEAM, this does not support torsion.
One of the simplest elements is the pinjointed truss element shown in Figure 622. This element has three degrees of freedom at each node and carries an axial force. The displacements and velocities measured in the local system are interpolated along the axis according to
x u = u 1 +  ( u 2 – u 1 ) L x · · · · u = u 1 +  ( u 2 – u 1 ) L
(6369) (6370) and at
x = L , u = u2 .
where at
N1
x = 0 , u = u1 ,
u1
A L
N2 u2
Figure 622
Truss Element
Incremental strains are found from
· · ( u2 – u1 ) Δ ε =  Δ t L
(6371)
and are computed using
· · 2 ( u2 – u1 ) Δ ε n + 1 ⁄ 2 =  Δ t n + 1 ⁄ 2 Ln + Ln + 1
n+1⁄2 n+1⁄2
(6372)
Et
The normal force
N
is then incrementally updated using a tangent modulus
according to (6373)
Nn + 1 = Nn A Et + Δ εn + 1 ⁄ 2
Two constitutive models are implemented for the truss element: elastic and elasticplastic with kinematic hardening.
Main Index
Chapter 6: Elements 217
CQUAD4  DYSHELLFORM = 9, Membrane Element
CQUAD4  DYSHELLFORM = 9, Membrane Element
The BelytschkoLinTsay shell element {Belytschko and Tsay [1981], Belytschko et al., [1984a]} is the basis for this very efficient membrane element. In this section we briefly outline the theory employed which, like the shell on which it is based, uses a combined corotational and velocitystrain formulation. The efficiency of the element is obtained from the mathematical simplifications that result from these two kinematical assumptions. The corotational portion of the formulation avoids the complexities of nonlinear mechanics by embedding a coordinate system in the element. The choice of velocity strain or rate of deformation in the formulation facilitates the constitutive evaluation, since the conjugate stress is the more familiar Cauchy stress. In membrane elements the rotational degrees of freedom at the nodal points may be constrained, so that only the translational degrees of freedom contribute to the straining of the membrane. A triangular membrane element may be obtained by collapsing adjacent nodes of the quadrilateral.
Corotational Coordinates
The midsurface of the quadrilateral membrane element is defined by the location of the element’s four corner nodes. An embedded element coordinate system (Figure 69) that deforms with the element is defined in terms of these nodal coordinates. The corotational coordinate system follows the development in equations (6166) to (6171).
VelocityStrain Displacement Relations
The corotational components of the velocity strain (rate of deformation) are given by:
ˆ ˆ 1 ⎛ ∂υ ∂υ ⎞ ˆ d i j =  ⎜ i + j⎟ ˆ ˆ 2 ⎝ ∂x j ∂x i ⎠
(6374)
The above velocitystrain relations are evaluated only at the center of the shell. Standard bilinear nodal interpolation is used to define the midsurface velocity, angular velocity, and the element’s coordinates (isoparametric representation). These interpolation relations are given by
v m = N I ( ξ, η )v I x m = N I ( ξ, η )x I
(6375) (6376)
where the subscript I is summed over all the element’s nodes and the nodal velocities are obtained by · differentiating the nodal coordinates with respect to time, i.e., υ I = x I . The bilinear shape functions are defined in (6185). The velocity strains at the center of the element, i.e., at ξ
Technique, giving:
ˆ ˆ d x = B 1I υ xI = 0 , and η = 0 , are obtained as in Corotational
(6377)
Main Index
218 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CQUAD4  DYSHELLFORM = 9, Membrane Element
ˆ ˆ d y = B 2I υ yI ˆ ˆ ˆ 2d xy = B 2I v xI + B 1I v yI
(6378) (6379)
where
∂N I B 1I = ˆ ∂x ∂N I B 2I = ˆ ∂y
(6380) (6381)
Stress Resultants and Nodal Forces
After suitable constitutive evaluations using the above velocity strains, the resulting stresses are multiplied by the thickness of the membrane, h , to obtain local resultant forces. Therefore,
ˆR ˆ f αβ = hσ αβ
(6382)
where the superscript R indicates a resultant force and the Greek subscripts emphasize the limited range of the indices for plane stress plasticity. The above element centered force resultants are related to the local nodal forces by
ˆ ˆR ˆR f xI = A ( B 1I f xx + B 2I f xy ) ˆ ˆR ˆR f yI = A ( B 2I f yy + B 1I f xy )
(6383) (6384)
where
A
is the area of the element.
The above local nodal forces are then transformed to the global coordinate system using the transformation relations given in equation (6173).
Membrane Hourglass Control
Hourglass deformations need to be resisted for the membrane element. The hourglass control for this element is discussed in Hourglass Control (BelytschkoLinTsay).
Main Index
Chapter 6: Elements 219
CELAS1D, Discrete Elements and CONM2, Masses
CELAS1D, Discrete Elements and CONM2, Masses
The discrete elements and masses in SOL 700 provide a capability for modeling simple springmass systems as well as the response of more complicated mechanisms. Occasionally, the response of complicated mechanisms or materials needs to be included in the models, e.g., energy absorbers used in passenger vehicle bumpers. These mechanisms are often experimentally characterized in terms of forcedisplacement curves. SOL 700 provides a selection of discrete elements that can be used individually or in combination to model complex forcedisplacement relations. The discrete elements are assumed to be massless. However, to solve the equations of motion at unconstrained discrete element nodes or nodes joining multiple discrete elements, nodal masses must be specified at these nodes. SOL 700 provides a direct method for specifying these nodal masses in the model input. All of the discrete elements are twonode elements, i.e., threedimensional springs or trusses. A discrete element may be attached to any of the other SOL 700 continuum, structural, or rigid body element. The force update for the discrete elements may be written as
ˆi+1 ˆi ˆ f = f + Δf
(6385)
i+1
where the superscript
indicates the time increment and the superposed caret
(ˆ)
indicates the force
in the local element coordinates; i.e., along the axis of the element. In the default case(i.e., no orientation vector is used), the global components of the discrete element force are obtained by using the element’s direction cosines:
⎧ ⎫ ⎧ ⎫ ⎧ ⎫ ⎪ Fx ⎪ ⎪ nx ⎪ ˆ ⎪ Δl x ⎪ ⎪ ⎪ ⎪ ⎪ ˆ⎪ ˆ f⎪ ⎨ F y ⎬ =  ⎨ Δl y ⎬ = f ⎨ n y ⎬ = f n l⎪ ˜ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ Fz ⎪ ⎪ Δ lz ⎪ ⎪ nz ⎪ ⎩ ⎭ ⎩ ⎭ ⎩ ⎭
(6386)
where:
⎧ ⎧ ⎫ ⎪ x2 – x1 ⎪ Δ ly ⎪ ⎪ ⎪ ⎪ Δ l = ⎨ Δ ly ⎬ = ⎨ y2 – y1 ˜ ⎪ ⎪ ⎪ ⎪ z2 – z1 ⎪ Δ lz ⎪ ⎩ ⎩ ⎭ l ⎫ ⎪ ⎪ ⎬ ⎪ ⎪ ⎭
(6387)
is the length
2 Δ l x + Δ l y + Δl z 2 2
l =
(6388)
and ( x i, y i, zi ) are the global coordinates of the nodes of the spring element. The forces in Equation (6386) are added to the first node and subtracted from the second node.
Main Index
220 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CELAS1D, Discrete Elements and CONM2, Masses
For a node tied to ground we use the same approach but for the ( x 2, y 2, z 2 ) coordinates in equation (6366), the initial coordinates of node 1, i.e., ( x 0, y 0, z 0 ) are used instead; therefore,
⎧ ⎧ ⎫ ⎪ Fx ⎪ ˆ ⎪ x0 – x1 ⎪ ⎪ f⎪ ⎨ Fy ⎬ =  ⎨ y0 – y1 l⎪ ⎪ ⎪ ⎪ Fz ⎪ ⎪ z0 – z1 ⎩ ⎩ ⎭ ⎫ ⎧ ⎫ ⎪ ⎪ nx ⎪ ⎪ ⎪ ˆ⎪ ⎬ = f ⎨ ny ⎬ ⎪ ⎪ ⎪ ⎪ ⎪ nz ⎪ ⎭ ⎩ ⎭
(6389)
The increment in the element force is determined from the user specified forcedisplacement relation. Currently, nine types of forcedisplacement/velocity relationships may be specified:
• Linear elastic • Linear viscous • Nonlinear elastic • Nonlinear viscous • Elastoplastic with isotropic hardening • General nonlinear • Linear viscoelastic • Inelastic tension and compression only • Muscle model
Orientation Vectors
An orientation vector,
⎧ ⎪ m1 ⎪ m = ⎨ m2 ˜ ⎪ ⎪ m3 ⎩ ⎫ ⎪ ⎪ ⎬ ⎪ ⎪ ⎭
(6390)
can be defined to control the direction the spring acts. If orientation vectors are used, it is strongly recommended that the nodes of the discrete element be coincident and remain approximately so throughout the calculation. If the spring or damper is of finite length, rotational constraints will appear in the model that can substantially affect the results. If finite length springs are needed with directional vectors, then the discrete beam elements, CROD, should be used with the coordinate system flagged for the finite length case. We will first consider the portion of the displacement that lies in the direction of the vector. The displacement of the spring is updated based on the change of length given by
Δ I = I – I0
(6391)
Main Index
Chapter 6: Elements 221
CELAS1D, Discrete Elements and CONM2, Masses
where I 0 is the initial length in the direction of the vector and node spring by
I = m1 ( x 2 – x 1 ) + m 2 ( y 2 – y 1 ) + m 3 ( z 2 – z 1 )
I
is the current length given for a node to
(6392)
and for a node to ground spring by
I = m1 ( x 0 – x 1 ) + m 2 ( y 0 – y 1 ) + m 3 ( z 0 – z 1 )
(6393)
The latter case is not intuitively obvious and can affect the sign of the force in unexpected ways if the user is not familiar with the relevant equations. The nodal forces are then given by
⎧ ⎫ ⎧ ⎪ Fx ⎪ ⎪ m1 ⎪ ⎪ ˆ⎪ ⎨ Fy ⎬ = f ⎨ m2 ⎪ ⎪ ⎪ ⎪ Fz ⎪ ⎪ m3 ⎩ ⎭ ⎩ ⎫ ⎪ ⎪ ⎬ ⎪ ⎪ ⎭
(6394)
The orientation vector can be either permanently fixed in space as defined in the input or acting in a direction determined by two moving nodes which must not be coincident but may be independent of the nodes of the spring. In the latter case, we recompute the direction every cycle according to:
⎧ ⎪ m1 ⎪ ⎨ m2 ⎪ ⎪ m3 ⎩ ⎧ n ⎫ n ⎪ x2 – x1 ⎪ ⎪ 1⎪ n n ⎬ =  ⎨ y 2 – y 1 ln ⎪ ⎪ ⎪ zn – zn ⎪ 1 ⎩ 2 ⎭ ⎫ ⎪ ⎪ ⎬ ⎪ ⎪ ⎭
n
(6395)
In equation (6395) the superscript,
, refers to the orientation nodes.
For the case where we consider motion in the plane perpendicular to the orientation vector we consider only the displacements in the plane, Δ l p , given by,
˜ Δl ˜
p
= Δl – m(m ⋅ Δl) ˜ ˜ ˜ ˜
(6396)
We update the displacement of the spring based on the change of length in the plane given by
Δl
p
= l – l0 l0
p
p
p
(6397)
where
is the initial length in the direction of the vector and l is the current length given for a node to
node spring by
l
p
= m1 ( x2 – x1 ) + m2 ( y2 – y1 ) + m3 ( z2 – z1 )
p
p
p
(6398)
Main Index
222 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CELAS1D, Discrete Elements and CONM2, Masses
and for a node to ground spring by
l
p
= m1 ( x0 – x1 ) + m2 ( y0 – y1 ) + m3 ( z0 – z1 )
p
p
p
(6399)
where
⎧ p ⎪ m1 ⎪ p ⎨ m2 ⎪ ⎪ p ⎩ m3 ⎧ p ⎫ ⎪ ⎪ ⎪ Δ lx ⎪ ⎪ 1 p ⎬ =  ⎨ Δ l 2 2 2⎪ y ⎪ p p p ⎪ p Δ lx + Δ ly + Δ lz ⎪ ⎭ ⎪ Δ lz ⎩ ⎫ ⎪ ⎪ ⎪ ⎬ ⎪ ⎪ ⎪ ⎭
(6400)
After computing the displacements, the nodal forces are then given by
⎧ ⎫ ⎧ p ⎪ Fx ⎪ ⎪ m1 ⎪ ⎪ ˆ⎪ Fy ⎬ = f ⎨ m p ⎨ ⎪ ⎪ ⎪ 2 ⎪ Fz ⎪ ⎪ p ⎩ ⎭ ⎩ m3 ⎫ ⎪ ⎪ ⎬ ⎪ ⎪ ⎭
(6401)
Dynamic Magnification “Strain Rate” Effects
To account for “strain rate” effects, we have a simple method of scaling the forces based to the relative velocities that applies to all springs. The forces computed from the spring elements are assumed to be the static values and are scaled by an amplification factor to obtain the dynamic value:
V F dynami c = ⎛ 1. + k d ⎞ F stati c ⎝ V 0⎠
(6402)
where
kd V V0
= a user defined input value = absolute relative velocity = dynamic test velocity
For example, if it is known that a component shows a dynamic crush force at 15m/s equal to 2.5 times the static crush force, use k d = 1.5 and V 0 = 15 .
Main Index
Chapter 6: Elements 223
CELAS1D, Discrete Elements and CONM2, Masses
Deflection Limits in Tension and Compression
The deflection limit in compression and tension is restricted in its application to no more than one spring per node subject to this limit, and to deformable bodies only. For example in the former case, if three spring are in series either the center spring or the two end springs may be subject to a limit but not all three. When the limiting deflection is reached momentum conservation calculations are performed and a common acceleration is computed:
ˆ ˆ f1 + f2 ˆ a common = m1 + m2
(6403)
An error termination will occur if a rigid body node is used in a spring definition where compression is limited.
Main Index
224 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
CDAMP2D, Linear Elastic or Linear Viscous
CDAMP2D, Linear Elastic or Linear Viscous
These discrete elements have the simplest forcedisplacement relations. The linear elastic discrete element has a forcedisplacement relation of the form
ˆ f = K Δl
(6404)
where K is the element’s stiffness and Δ l is the change in length of the element. The linear viscous element has a similar forcevelocity (rate of displacement) relation:
ˆ Δl f = C Δt
(6405) is a viscous damping parameter and
Δt
where
C
is the time step increment.
Main Index
Chapter 6: Elements 225
Eulerian Elements
Eulerian Elements
Element Definition
In the Eulerian solver, the mesh is defined by grid points and solid elements. The elements are specified as being (partially) filled with certain materials or with nothing (VOID), and initial conditions are defined. As the calculation proceeds, the material moves relative to the Eulerian mesh. The mass, momentum, and energy of the material is transported from element to element depending on the direction and velocity of the material flow. The solver then calculates the impulse and work done on each of the faces of every Eulerian element. Eulerian elements can only be solid but have a general connectivity and therefore are defined in exactly the same way as Lagrangian elements.
Solid Elements
There are three types of Euler elements, a sixsided CHEXA with eight grid points defining the corners, a CPENTA with six grid points, and a CTETRA with four grid points. The connectivity of the element is defined in exactly the same manner as a Lagrangian element, that is, with a CHEXA, CPENTA, or CTETRA entry. However, in order to differentiate between Lagrangian and Eulerian solid elements, the property entry for Euler is PEULERn rather than PSOLID. Unlike Lagrangian solid elements, the CPENTA and CTETRA elements perform just as well as the CHEXA element. They can be used, therefore, wherever meshing demands such use.
CHEXA
CPENTA
CTETRA
The PEULERn entry references a MATDEUL material entry that is used to define the material filling the elements at the start of the calculation. When no material entry is referenced (the field contains a zero), the element is initially void.
Main Index
226 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Graded Meshes
Graded Meshes
Nonuniform Euler meshes can easily be created by MD Patran. This is particularly useful when the elements do not need to be orthogonal. But in orthogonal meshes, the nonuniformity propagates to the boundaries of the mesh. Even at the boundary, there will be elements that have a small size in at least one direction. To allow for large element sizes in all three coordinate directions at the boundary, block structured meshing has to be used. This type of meshing is very effective when modeling the flow over bodies and is often used in CFD. Usually, the Euler elements are fine near the body and become coarser as the distance is increased away from the body. By using the graded mesh capability, a block of fine elements is used in the area of interest and a coarse block is used in other areas. To use this method, the blocks need to be glued together. This is done by adding PARAM,GRADMESH. Graded meshes are supported by all Eulerian solvers except by the single material Euler solver with Strength. Intersection of a coupling surface segment with the interface between a coarse and fine block is not allowed. Graded meshes are not supported by multiple Euler domains. Figure 1 shows an example of graded meshes. This mesh is be used in Example 4.14. Consider a blast wave simulation. Close to the ignition point elements need to be fine, but at some distance the wave blast becomes larger in radius and it becomes less steep allowing coarser mesh elements. To reduce the number of elements and the limit the problem size, part of the fine mesh can be replaced by a coarse mesh. For modeling, only part of the fine mesh is constructed and the coarse mesh is created such that it covers the whole problem domain. Next, the fine mesh is glued to the coarse mesh. This gluing is activated by PARAM,GRADEDMESH. The algorithm identifies the elements of the coarse mesh that are covered by the fine mesh. They are deactivated and removed from Euler archive output requests.
Requirements for Gluing Meshes
Connecting meshes with varying mesh sizes was already possible by using multiple Euler domains with porous coupling surfaces. Using this method, two connected meshes are surrounded by a coupling surface together with a fully porous subsurface that connects the two domains. The domains are exclusively setup by using the MESH entry. In this approach there are no requirements on the size and location of the meshes. In output requests, only one domain can be examined. Making plots of the whole domain is not possible in one Patran session. This makes the approach cumbersome. With the graded mesh functionality there is no longer any need to create coupling surfaces. An additional advantage is that there is no restriction on how the elements are created for the simulation. Any preprocessor may be used, defining the CHEXA elements in an input file, but also blocks of meshes can be created by means of the MESH entry. The interface between the fine and coarse mesh is identified in the solver when PARAM,GRADMESH is activated. However, one restriction applies to graded meshes. An Euler element of the coarse mesh has to be fully active or fully inactive. This means that the coarse element should not intersect elements of the fine mesh or it should be fully covered by the fine elements. Fine elements are not allowed to cover any part of the coarse elements. In practice, this means that the fine mesh has to fit nicely in the coarse mesh. As shown in the meshes in Figure 623, the four marked
Main Index
Chapter 6: Elements 227
Graded Meshes
locations a grid point of the fine mesh coincides with a grid point of the coarse mesh. This matching does not need to be exact, since the solver uses a tolerance to find the coinciding grid points. Visualization of the results of the complete Euler domain can be done in one session in a postprocessor like MD Patran.
Figure 623
Graded Mesh with Structure
Gluing Meshes
Gluing of fine and coarse meshes is activated by PARAM,GRADMESH. The algorithm removes coarse elements that are completely covered by fine elements. Here the criterion for removal is based on the element volume. The element with the largest volume will be removed. It is also possible to remove the elements with the lowest element numbers. These two approaches are activated by, respectively, the MINVOL and ELNUM option of the GRADMESH PARAM. The gluing algorithm performs the following steps:
• The Euler elements are sorted such that the connected elements are grouped together. • For each group of elements, the algorithm determines which elements have to be removed. As
mentioned the criterion for removal is based on the fact if these elements are completely covered by elements of another group that are smaller or have a larger element number.
• Next the groups of elements are connected at boundary faces and at locations where elements
have been removed. Special faces will be created that connect an element of one element group to an element of another group. If the fine mesh is fully surrounded by the coarse mesh, the interface between the two meshes consists of the boundary faces of the fine mesh. The geometry formed by these faces will be used to construct special faces that connect an element of the coarse mesh to an element of the fine mesh.
Main Index
228 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Graded Meshes
Using Graded Meshes
Blocks of Euler elements can be defined by either MD Patran or the MESH,BOX entry. It is important that at the interface a face of a coarse element can be matched to several parts of faces of fine elements. This matching does not need to be exact since a tolerance is used. One way to construct graded meshes is:
• Make the coarse mesh by using MESH,BOX. • Run the simulation and read in the Euler mesh into MD Patran. For a part of this mesh, the
coarse elements have to be replaced by finer elements. Select a part of this mesh by selecting two Eulerian grid points.
• Create the fine mesh by MESH,BOX, by using for the reference grid point as the first grid point.
The width of the box is given by subtracting the coordinates of the second grid point from the coordinates of the first. To finish the definition of the fine mesh, the number of elements in each direction has to be defined. This ensures that the fine mesh fits nicely in the coarse mesh.
• Add PARAM,GRADMESH which activates the algorithm that is described above.
If needed a part of fine mesh can be replaced by an even finer mesh, by iterating through the steps above. To construct graded meshes by MD Patran, a utility called “the break up of element” can be used.
Visualization with MD Patran
Elements that are fully covered by the fine mesh will not be included in the Euler archives. This allows for visualization of all euler elements in one MD Patran session. The fringe plots can lack smoothness at the interfaces. The reason is that Patran determines colors on the basis of grid point values. They are computed by averaging over the elements that are connected to the grid points. In graded meshes, there can be grid points that belong only to fine elements and not to coarse elements. These grid points are called hanging nodes. An example of a hanging node is shown in Figure 624. At the hanging node, the value only reflects the fine mesh. This results in loss of smoothness. An example is shown in Figure 625. If the results are postprocessed by using the element values this problem does not appear (see Figure 626).
Main Index
Chapter 6: Elements 229
Graded Meshes
Figure 624
Hanging Node
Figure 625
Lack of Smoothness
Figure 626
Using Element Values
Main Index
230 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Graded Meshes
Main Index
Chapter 7: Materials MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
7
Materials
Lagrangian Material Models for SOL 700 Eulerian Material Models for SOL 700 232 325
Main Index
232 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
Lagrangian Material Models for SOL 700
In addition to most materials available in MD Nastran Solutions 101  200, the following material models which are suitable for explicit applications are also made available for SOL 700. About ninety new material models are currently implemented in SOL 700. In this chapter, we will highlight the theoretical background of these material models. These are the material models that are most commonly used for typical structural applications. Additional materials will be added in future releases. 1 2 3 5 6 7 9 10 12 13 14 15 18 19 20 22 24 26 27 28 29 30 31 32 34 40 5455 Elastic Orthotropic Elastic Kinematic/Isotropic ElasticPlastic Soil and Crushable/Noncrushable Foam Viscoelastic BlatzKo Rubber Null Hydrodynamics IsotropicElasticPlasticHydrodynamic IsotropicElasticPlastic ElasticPlastic with Failure Model Soil and Crushable Foam with Failure Model Johnson/Cook Strain and Temperature Sensitive Plasticity Power Law Isotropic Plasticity Strain Rate Dependent Isotropic Plasticity Rigid Composite Damage Model Piecewise Linear Isotropic Plasticity Honeycomb Compressible MooneyRivlin Rubber Resultant Plasticity Forced Limited Resultant Fomulation ShapeMemory Superelastic Material Slightly Compressible Rubber Model Laminated Glass Model Fabric Nonlinear Orthotropic Composite Damage Model
Main Index
Chapter 7: Materials 233
Lagrangian Material Models for SOL 700
57 58 59 62 63 64 66 67 68 69 70 71 72 72R 73 74 76 77 80 81 83 87 89 93 94 95 97 98 99 100 112 114 119 121
Low Density Urethane Foam Laminated Composite Fabric Composite Failure Model  Plasticity Based Viscous foam Isotropic Crushable Foam Strain Rate Sensitive PowerLaw Plasticity Linear Elastic Discrete Beam Nonlinear Elastic Discrete Beam Nonlinear Plastic Discrete Beam SID Damper Discrete Beam Hydraulic Gas Damper Discrete Beam Cable Discrete Beam Concrete Damage Concrete Damage Release III Low Density Viscoelastic Foam Elastic Spring Discrete Beam General Viscoelastic Hyperviscoelastic Rubber RambergOsgood Plasticity Plastic with Damage FuChang’s Foam with Rate Effects Cellular Rubber Plastic Polymer Elastic Six Degrees of Freedom Spring Discrete Beam Inelastic Spring Discrete Beam Inelastic Spring Six Degrees of Freedom Discrete Beam General Joint Discrete Beam Simplified JohnsonCook Simplified JohnsonCook Orthotropic Damage Spot weld Finite Elastic Strain Plasticity Layered Linear Plasticity General Nonlinear Six Degrees of Freedom Discrete Beam General Nonlinear One Degree of Freedom Discrete Beam
Main Index
234 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
123 126 127 158 181 196 B01 S01 S02 S03 S04 S05 S06 S07 S08 S13 S14 S15 SW1 SW2 SW3 SW4 SW5
Modified Piecewise Linear Plasticity Modified Honeycomb ArrudaBoyce rubber Composite Fabric Simplified Rubber General Spring Descrete Beam Seatbelt Spring Elastic (Linear) Damper Viscous (Linear) Spring Elastoplastic (Isotropic) Spring Nonlinear Elastic Damper Nonlinear Viscous Spring General Nonlinear Spring Maxwell (Three parameter Viscoelastic) Spring Inelastic (Tension or Compression) Spring Trilinear Degrading Spring Squat Shearwall Spring Muscle Spot Weld (Simple DamageFailure) Spot Weld (Resultantbased Failure Criteria) Spot Weld (Stressbased Failure) Spot Weld (Rate Dependent Stressbased Failure) Spot Weld (Additional Failure)
In the table below, a list of the available material models and the applicable element types are given. Some materials include strain rate sensitivity, failure, equations of state, and thermal effects and this is also noted. General applicability of the materials to certain kinds of behavior is suggested in the last column.
Main Index
Chapter 7: Materials 235
Lagrangian Material Models for SOL 700
Material Title 1 Elastic 2 Orthotropic Elastic (Anisotropic  solids) 3 Plastic Kinematic/Isotropic 5 Soil and Foam 6 Linear Viscoelastic 7 BlatzKo Rubber 9 Null Material
Bricks Beams Thin Shells Thick Shells
Gn General Cm Composites Cr Ceramics Fl Fluids Fm Foam Gl Glass Hy Hydrodyn Mt Metal Pl Plastic Rb Rubber Sl Soil/Cone
Material Number
Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y
StrainRate Effects Failure EquationofState Thermal Effects
Gn, Fl Cm, Mt Cm, Mt, Pl Fm, Sl Y Rb Rb, Polyurethane Y Y Y Fl, Hy Hy, Mt Y Mt Mt Fm, Sl Mt, Pl Mt, Pl Cm Mt, Pl Cm, Fm, Sl Rb Mt Mt Rb Cm, Gl Y Cm
10 Elastic Plastic Hydro (dynamics) 12 Isotropic Elastic Plastic 13 Isotropic ElasticPlastic with Failure 14 Soil and Foam with Failure 15 Johnson/Cook Plasticity Model 18 Power Law Plasticity (Isotropic) 19 Strain Rate Dependent Rate Plasticity 20 Rigid 22 Composite Damage 24 Piecewise Linear Isotropic Plasticity 26 Honeycomb 27 MooneyRivlin Rubber 28 Resultant Plasticity 29 Forced Limited Resultant Formaultion 30 Shaped Memory Alloy 31 Slightly Compressible Rubber 32 Laminated Glass (Composite) 34 Fabric 40 Nonlinear Orthotropic
Y Y Y Y Hy, Mt
Y Y Y Y Y Y
Main Index
236 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
Material Title
54 Composite Damage with Chang Failure 55 Composite Damage with TsaiWu Failure 57 Low Density Urethane Foam 58 Laminated composite Fabric 59 Composite Failure  Plasticity Based 62 Viscous foam (Crash Dummy) 63 Isotropic Crushable Foam 64 Rate Sensitive PowerLaw Plasticity 66 Linear Elastic Discrete Beam 67 Nonlinear Elastic Discrete Beam 68 Nonlinear Plastic Discrete Beam 69 SID Damper Discrete Beam 70 Hydraulic Gas Damper Discrete Beam 71 Cable Discrete Beam 72 Concrete Damage 72R Concrete Damage Release III 73 Low Density Viscous Foam 74 Elastic Spring Discrete Beam 76 General Viscoelastic (Maxwell model) 77 Hyperelastic and Ogden Rubber 79 Hysteretic Soil (ElastoPerfectly Plastic) 80 RambergOsgood 81 Plastic with Damage (ElastoPlastic) 83 FuChang’s Foam 87 Cellular Rubber 89 Plastic Polymer Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y
Bricks Beams Thin Shells Thick Shells
Gn General Cm Composites Cr Ceramics Fl Fluids Fm Foam Gl Glass Hy Hydrodyn Mt Metal Pl Plastic Rb Rubber Sl Soil/Cone
Material Number
Y Y Y Y Y Y Y Y Y Y Y
StrainRate Effects Failure EquationofState Thermal Effects
Y Y Y Y Y
Cm Cm Fm Cm, Cr Fm Fm Mt
Y Y Y Y Y Y Y Y Y Y Y Y Y Y Sl Sl Fm Rb Rb Sl Mt, Pl Fm Rb
Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y
Main Index
Chapter 7: Materials 237
Lagrangian Material Models for SOL 700
Material Title Elastic Six Degrees of Freedom Spring Discrete Beam Inelastic Six Degrees of Freedom Spring Discrete Beam
93
94 Inelastic Spring Discrete Beam 95
97 General Joint Discrete Beam 98 Simplified JohnsonCook 99 Simplified JohnsonCook Orthotropic Damage
100 Spot weld 112 Finite Elastic Strain Plasticity 114 Layered Linear Plasticity 119 121 General Nonlinear Six Degrees of Freedom Discrete Beam General Nonlinear One Degree of Freedom Discrete Beam Y Y Y
123 Modified Piecewise Linear Plasticity 126 Modified Honeycomb 127 ArrudaBoyce rubber 158 Composite Fabric 181 Simplified Rubber 196 General Spring Discrete Beam B01 Seatbelt S01 Spring Elastic (Linear) S02 Damper Viscous (Linear) S03 Spring Elastoplastic (Isotropic) S04 Spring Nonlinear Elastic S05 Damper Nonlinear Viscous Y Y Y Y Y Y Y
Bricks Beams Thin Shells Thick Shells
Gn General Cm Composites Cr Ceramics Fl Fluids Fm Foam Gl Glass Hy Hydrodyn Mt Metal Pl Plastic Rb Rubber Sl Soil/Cone
Material Number
Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y
Y Y Y Y
StrainRate Effects Failure EquationofState Thermal Effects
Cm RB
Y Y Y
Main Index
238 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
Material Title
S06 Spring General Nonlinear Spring Maxwell (Three Parameter S07 Viscoelastic) S08 Spring Inelastic (Tension or Compression S13 Spring Trilinear Degrading S14 Spring Squat Shearwall S15 Spring Muscle SW1 Spot Weld (Simple DamageFailure) Spot Weld (Resultantbased Failure SW2 Criteria) SW3 Spot Weld (Stressbased Failure) SW4 Spot Weld (Rate Dependent Stressbased Failure)
Bricks Beams Thin Shells Thick Shells
Gn General Cm Composites Cr Ceramics Fl Fluids Fm Foam Gl Glass Hy Hydrodyn Mt Metal Pl Plastic Rb Rubber Sl Soil/Cone
Material Number
Y Y Y Y
Y Y Y Y Y
SW5 Spot Weld (Additional Failure)
Material Model 1: Elastic
In this elastic material we compute the corotational rate of the deviatoric Cauchy stress tensor as
s i∇ j
n+1⁄2
· n+1⁄2 = 2 Gε' i j
StrainRate Effects Failure EquationofState Thermal Effects
(71)
and pressure
p n + 1 = K ln V n + 1
(72)
V
where G and K are the elastic shear and bulk moduli, respectively, and ratio of the current volume to the initial volume.
is the relative volume; i.e., the
For standard MD Nastran solution sequence, this would be the same as using MAT1 to define a linear elastic material.
Main Index
Chapter 7: Materials 239
Lagrangian Material Models for SOL 700
Material Model 2: Orthotropic Elastic
The material law that relates second PiolaKirchhoff stress
S = C ⋅ E = T t Cl T ⋅ E S
to the GreenSt. Venant strain
E
is (73)
where
T
2 l1 2 l2
is the transformation matrix [Cook 1974].
2 m1 2 m2 2 m3 2 n1 2 n2 2 n3
l1 m1 l2 m2 l3 m3
m1 n1 m2 n2 m3 n3
n1 l1 n2 l2 n3 l3
T =
2 l3
2 l 1 l 2 2m 1 m 2 2n 1 n 2 ( l 1 m 2 + l 1 m 1 ) ( m 1 n 2 + m 2 n 1 ) ( n 1 l 2 + n 2 l 1 ) 2 l 2 l 3 2m 2 m 3 2n 2 n 3 ( l 2 m 3 + l 3 m 2 ) ( m 2 n 3 + m 3 n 2 ) ( n 2 l 3 + n 3 l 2 ) 2 l 3 l 1 2m 3 m 1 2n 3 n 1 ( l 3 m 1 + l 1 m 3 ) ( m 3 n 1 + m 1 n 3 ) ( n 3 l 1 + n 1 l 3 )
(74)
li , mi , n i
are the direction cosines for i = 1,2,3
Cl
x' i = l i x 1 + m i x 2 + n i x 3
(75) is defined in terms of the material axes as
and
x' i
denotes the material axes. The constitutive matrix
1  υ 21 υ 31  –  – E 11 E 22 E 33 υ 32 υ 12 1 –   – E 11 E 22 E 33
0 0 0 1 G 12 0 0
0 0 0 0 1 G 23 0
0 0 0 0 0 1 G 31
C l– 1 =
υ 13 υ 23 1 –  –  E 11 E 22 E 33 0 0 0 0 0 0 0 0 0
(76)
where the subscripts denote the material axes; i.e.,
υ i j = υ x'
i x' j
and
E ii = E x'
i
(77)
Since
Cl
is symmetric ect. (78)
υ 21 υ 12  =  , E 11 E 22
The vector of GreenSt. Venant strain components is
Et = E 11 ,E 22 ,E 33 ,E 12 ,E 23 ,E 31
(79)
Main Index
240 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
ρ ∂x i ∂x j  σ i j =    S k l ρ o ∂X k ∂X l
(710)
After computing S i j , we use Equation (710) to obtain the Cauchy stress. This model will predict realistic behavior for finite displacement and rotations as long as the strains are small. For standard MD Nastran solution sequences, this would be the same as using MAT9 to define a linear anisotropic material. For shell elements, you would have used the MAT2 or MAT8 option.
Material Model 3: Elastic Plastic with Kinematic Hardening
Isotropic, kinematic, or a combination of isotropic and kinematic hardening may be obtained by varying a parameter, called β between 0 and 1. For β equal to 0 and 1, respectively, kinematic and isotropic hardening are obtained as shown in Figure 71 where l 0 and l are the undeformed and deformed length of uniaxial tension specimen, respectively. Krieg and Key [1976] formulated this model and the implementation is based on their paper. In isotropic hardening, the center of the yield surface is fixed but the radius is a function of the plastic strain. In kinematic hardening, the radius of the yield surface is fixed but the center translates in the direction of the plastic strain. Thus the yield condition is
2 σy 1 φ =  ξ ij ξ i j –  = 0 2 3
(711)
where
ξ ij = S i j α ij
p σ y = σ 0 + βE p ε ef f
(712) (713)
αi j
The corotational rate of
2 ·p α i∇ = ( 1 – β )  E p ε i j j 3
is (714)
Hence,
n ∇ α inj + 1 = α ij + ( α ij n+1⁄2 n + α ink Ω ink + 1 ⁄ 2 + α jnk + 1 ⁄ 2 Ω ki + 1 ⁄ 2 )Δ t n + 1 ⁄ 2
(715)
Main Index
Chapter 7: Materials 241
Lagrangian Material Models for SOL 700
Et
E
Yield Stress
l ln ⎛ ⎞ ⎝ l 0⎠
β = 0 β = 1
Kinematic Hardening Isotropic Hardening
Figure 71
Elasticplastic Behavior with Isotropic and Kinematic Hardening
Strain rate is accounted for using the Cowper and Symonds [Jones 1983] model which scales the yield stress by a strain rate dependent factor
ρy = · ε 1⁄p p 1 + ⎛ ⎞ ( σ 0 + βE p ε eff ) ⎝ C⎠
(716)
· ε
where
· ε+
p
and
C
are userdefined input constants and
is the strain rate defined as: (717)
· · ε i j ε ij σy ,
The current radius of the yield surface,
p βE p ε eff
is the sum of the initial yield strength,
σ0 ,
plus the growth
, where
Ep
is the plastic hardening modulus (718)
Et E E p = E – Et
and
ε ef f
p
is the effective plastic strain
Main Index
242 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
t
ε ef f =
p
∫ ⎛  ε ij ε i j⎞ ⎝3 ⎠ 2·p ·p
0
1/2
dt
(719)
The plastic strain rate is the difference between the total and elastic (right superscript) strain rates:
·p · ·e εi j = εi j – εi j
(720)
In the implementation of this material model, the deviatoric stresses are updated elastically, as described for model 1, but repeated here for the sake of clarity:
σ i*j = σ i j + C ij kl Δε kl
n
(721)
where
σ i*j σ inj
is the trial stress tensor, is the stress tensor from the previous time step, is the elastic tangent modulus matrix, is the incremental strain tensor.
C ij kl Δ ε kl
and, if the yield function is satisfied, nothing else is done. If, however, the yield function is violated, an increment in plastic strain is computed, the stresses are scaled back to the yield surface, and the yield surface center is updated. Let
si j
*
represent the trial elastic deviatoric stress state at
n+1
* * 1 * s i j = σ i j –  σ kk σ ij 3
(722)
and
* * ξ ij = s ij – α i j
(723)
Define the yield function,
3 * 2 2 φ =  ξ ij ξ i*j – σ y = Λ 2 – σ y 2
< 0 for elastic or neutral loading > 0 for plastic harding
(724)
For plastic hardening then:
p ε ef f n+1
Λ – σy pn pn p = ε eff +  = ε ef f + Δ ε ef f 3 G + Ep
(725)
Scale back the stress deviators:
p 3 GΔ ε ef f σ inj + 1 = σ i*j –  ξ i*j Λ
(726)
Main Index
Chapter 7: Materials 243
Lagrangian Material Models for SOL 700
and update the center:
p ( 1 – β )E p Δ ε ef f n α inj + 1 = α ij + Λ
(727)
Plane Stress Plasticity The plane stress plasticity options apply to beams, shells, and thick shells. Since the stresses and strain increments are transformed to the lamina coordinate system for the constitutive evaluation, the stress and strain tensors are in the local coordinate system. The application of the Jaumann rate to update the stress tensor allows for the possibility that the normal stress, σ 33 , will not be zero. The first step in updating the stress tensor is to compute a trial plane stress update assuming that the incremental strains are elastic. In the above, the normal strain increment is replaced by the elastic strain increment
σ 33 + λ ( Δ ε 11 + Δ ε 22 ) Δ ε 33 = – λ + 2μ Δ ε 33
(728)
where
λ
and
γ
are Lamé’s constants.
When the trial stress is within the yield surface, the strain increment is elastic and the stress update is completed. Otherwise, for the plastic plane stress case, secant iteration is used to solve Equation (726) for the normal strain increment ( Δ ε 33 ) required to produce a zero normal stress:
p 3 GΔ ε ef f ξ 33 i * σ 33 = σ 33 – Λ i
(729) indicates the iteration number.
Δ ε 33
Here, the superscript
i
The secant iteration formula for
(the superscript
p
is dropped for clarity) is (730)
i i– Δ ε 33 – Δ ε 33 1 i i+ i–  – Δ ε 33 1 = Δ ε 33 1 –  σ 33 1 i i– σ 33 – σ 33 1
where the two starting values are obtained from the initial elastic estimate and by assuming a purely plastic increment; i.e.,
1 Δ ε 33 = – ( Δ ε 11 – Δ ε 22 )
(731)
These starting values should bound the actual values of the normal strain increment. The iteration procedure uses the updated normal stain increment to update first the deviatoric stress and then the other quantities needed to compute the next estimate of the normal stress in Equation (729). The iterations proceed until the normal stress convergence of the normal strains:
σ 33
i
is sufficiently small. The convergence criterion requires
Main Index
244 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
i i– Δ ε 33 – Δ ε 33 1  < 10 – 4 i+1 Δ ε 33
(732)
After convergence, the stress update is completed using the relationships given in Equations (725) and (727). For SOL SEQ 106, 129, or 600, this material model is the same as using MATS1, with the hardening rule (HR) set to either Isotropic or Kinematic. These models do not allow β between 0 and 1.
Material Model 5: Soil and Crushable Foam
This model, due to Krieg [1972], provides a simple model for foam and soils whose material properties are not well characterized. We believe the other foam models (such as material models 57, 62, and 63) in SOL 700 are superior in their performance and are recommended over this model which simulates the crushing through the volumetric deformations. If the yield stress is too low, this foam model gives nearly fluid like behavior. A pressuredependent flow rule governs the deviatoric behavior:
1 φ s =  s i j s ij – ( a 0 + a 1 p + a 2 p 2 ) 2
(733)
where a 0 , a 1 , and a 2 are userdefined constants. Volumetric yielding is determined by a tabulated curve of pressure versus volumetric strain. Elastic unloading from this curve is assumed to a tensile cutoff as illustrated in Figure 72. Implementation of this model is straightforward. One history variable, the maximum volumetric strain in compression, is stored. If the new compressive volumetric strain exceeds the stored value, loading is indicated. When the yield condition is violated, the updated trial stresses, s i*j , are scaled back using a simple radial return algorithm:
⎛ a + a p + a p 2⎞ 0 1 2 s inj + 1 = ⎜ ⎟ ⎜ ⎟ 1  s ij s i j ⎝ ⎠ 2
1⁄2 * s ij
(734)
If the hydrostatic tension exceeds the cutoff value, the pressure is set to the cutoff value and the deviatoric stress tensor is zeroed.
Main Index
Chapter 7: Materials 245
Lagrangian Material Models for SOL 700
Pressure
Loading and unloading follows the input curve if the volumetric crushing option is off (VCR = 1.0).
V ln ⎛ ⎞ ⎝ V 0⎠
The bulk unloading modulus is used if the volumetric crushing option is on (VCR = 0). Volumetric Strain (Compression) Tension Cutoff Value
Tension
Figure 72
Volumetric Strain Versus Pressure Curve for Soil and Crushable Foam Model
Material Model 6: Viscoelastic
In this model, linear viscoelasticity is assumed for the deviatoric stress tensor [Herrmann and Peterson 1968]:
∂ε' i j s i j = 2 ∫ φ ( t – τ )  dτ ∂τ
0 t
(735)
where
φ ( t ) = G ∞ + ( G 0 – G ∞ )e – βt
(736)
is the shear relaxation modulus. A recursion formula is used to compute the new value of the hereditary integral at time t n + 1 from its value at time t n . Elastic bulk behavior is assumed:
p = K ln V
(737)
where pressure is integrated incrementally.
Main Index
246 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
Material Model 7: Continuum Rubber
The hyperelastic continuum rubber model was studied by Blatz and Ko [1962]. In this model, the second PiolaKirchhoff stress is given by
Si j = G ⎛ V–1 Ci j – V ⎝
1 – 1 – 2υ δ ij⎠
⎞ V
(738) is the relative volume, υ is Poisson’s ratio, and C ij is the right
where G is the shear modulus, CauchyGreen strain:
∂x k ∂x k  C ij =  ∂X i ∂X j
(739)
Si j ,
after determining
ρ ∂x i ∂x j  σ i j =    S kl ρ 0 ∂X k ∂X l
it is transformed into the Cauchy stress tensor, σ i j : (740)
υ
where
ρ0
and
ρ
are the initial and current density, respectively. The default value of
is 0.463.
Material 27 and 31 better represent incompressible materials.
Material Model 9: Null Material
For solid elements equations of state can be called through this model to avoid deviatoric stress calculations. A pressure cutoff may be specified to set a lower bound on the pressure. This model has been very useful when combined with the reactive high explosive model where material strength is often neglected. The null material should not be used to delete solid elements. A optional viscous stress of the form
σ i j = μ ε′ ij
(741)
μ
is computed for nonzero
where
ε′ ij
is the deviatoric strain rate.
Sometimes it is advantageous to model contact surfaces via shell elements which are not part of the structure, but are necessary to define areas of contact within nodal rigid bodies or between nodal rigid bodies. Beams and shells that use this material type are completely bypassed in the element processing. The Young’s modulus and Poisson’s ratio are used only for setting the contact interface stiffnesses, and it is recommended that reasonable values be input.
Material Model 10: ElasticPlasticHydrodynamic
For completeness we give the entire derivation of this constitutive model based on radial return plasticity. The pressure, ρ ; deviatoric strain rate, defined in Equation (742):
· ε' i j ;
deviatoric stress rate,
· si j ;
and volumetric strain rate,
· εv ,
are
Main Index
Chapter 7: Materials 247
Lagrangian Material Models for SOL 700
1 p = –  σ i j δ ij 3 s i j = σ i j + pδ i j s i∇ j · · = 2με' i j = 2G ε' i j
1· · · ε' ij = ε ij –  ε v 3 · · ε v = ε i j δ ij
(742)
The Jaumann rate of the deviatoric stress,
· s i∇ = s i j – s ip Ω pj – s j p Ω pi j
s i∇ , j
is given by: (743)
First we update
*s n + 1 ij
s inj
to
s inj + 1
elastically
· · n n = s ij + s i p Ω pj + s j p Ω pi + 2Gε' i j dt = s ij + R ij + 2 Gε' ij d t
R s ij n
2 GΔ ε' i j
(744)
where the left superscript, *, denotes a trial stress value. The effective trial stress is defined by
1⁄2 3 s * = ⎛  *s inj + 1 *s inj + 1⎞ ⎝2 ⎠
(745)
y
and if
s*
exceeds yield stress
σ,
the von Mises flow rule: (746)
2 σy 1 φ =  s ij s i j –  ≤ 0 2 3
is violated and we scale the trial stresses back to the yield surface; i.e., a radial return
σy n s inj + 1 =  *s inj + 1 = m * s ij + 1 s*
(747)
The plastic strain increment can be found by subtracting the deviatoric part of the strain increment that is elastic,
n 1  ( s n + 1 – s iR ) , j 2 G ij
from the total deviatoric increment,
Δ ε' i j ,
i.e., (748)
R 1 n Δ ε ipj = Δ ε' i j –  ( s ij + 1 – s i j n ) 2G
Recalling that,
n+1 *s ij – s ij n Δ ε' i j = 2G R
(749)
and substituting Equation (749) into Equation (748) we obtain,
n n ( *s ij + 1 – s ij + 1 ) Δ ε ipj = 2G
(750)
Main Index
248 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
Substituting Equation (747)
n s inj + 1 = m * s ij + 1
(751)
into Equation (750) gives,
(1 – m) n 1–m n Δ ε ipj =  *s ij + 1 =  s inj + 1 = dλ s ij + 1 2G 2 Gm
(752)
By definition an increment in effective plastic strain is
2 p p Δ ε p = ⎛  Δ ε ij Δ ε ij ⎞ ⎝3 ⎠
1⁄2
(753)
Squaring both sides of Equation (752) leads to:
1–m 2 Δ ε ipj Δ ε ipj = ⎛ ⎞ *s inj + 1 *s inj + 1 ⎝ 2G ⎠
(754)
or from Equations (745) and (753):
3 p2 1 – m 2  Δ ε = ⎛ ⎞ 2 s *2 ⎝ 2G ⎠ 3 2
(755)
Hence,
s∗ – σ y 1–m ∴Δ ε p =  s∗ = 3G 3G
(756)
where we have substituted for m from Equation (747)
σy m = s∗
If isotropic hardening is assumed then:
n n σy + 1 = σy + E p Δ εp
(757)
and from Equation (756)
n n ( s∗ – σ y + 1 ) ( s∗ – σ y – E p Δ ε p ) Δ ε p =  = 3G 3G
(758)
Thus,
n ( 3G + E p ) Δ ε p = ( s∗ – σ y )
(759)
and solving for the incremental plastic strain gives
n ( s∗ – σ y ) Δ ε p = (3 G + Ep)
(760)
Main Index
Chapter 7: Materials 249
Lagrangian Material Models for SOL 700
The algorithm for plastic loading can now be outlined in five simple stress. If the effective trial stress exceeds the yield stress then 1. Solve for the plastic strain increment:
n ( s∗ – σ y ) Δ ε p = ( 3G + E p )
2. Update the plastic strain:
εp
n+1
= εp + Δ εp
n
3. Update the yield stress:
n n σy + 1 = σ y + E p Δ εp
4. Compute the scale factor using the yield strength at time
m = s∗
n σy + 1
n+1:
5. Radial return the deviatoric stresses to the yield surface:
n s inj + 1 = m *s ij + 1
Material Model 12: Isotropic ElasticPlastic
The von Mises yield condition is given by:
2 σy φ = J 2 – 3
(761)
J2 ,
where the second stress invariant,
1 J 2 =  s ij s i j 2
is defined in terms of the deviatoric stress components as (762)
p and the yield stress, σ y , is a function of the effective plastic strain, ε ef f , and the plastic hardening modulus, Ep : p σ y = σ 0 + E p ε eff
(763)
The effective plastic strain is defined as:
t p ε ef f =
∫ dε eff
p 0
(764)
where:
p dε eff =
2 p p  dε ij dε i j 3
(765)
Main Index
250 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
and the plastic tangent modulus is defined in terms of the input tangent modulus,
E Et E p = E – Et
Et ,
as: (766)
Pressure is given by the expression
1 p n + 1 = K ⎛  – 1⎞ ⎝ Vn + 1 ⎠
(767)
where K is the bulk modulus. This is perhaps the most cost effective plasticity model. Only one history p variable, ε ef f , is stored with this model. This model is not recommended for shell elements. In the plane stress implementation, a onestep radial return approach is used to scale the Cauchy stress tensor to if the state of stress exceeds the yield surface. This approach to plasticity leads to inaccurate shell thickness updates and stresses after yielding. This is the only model in SOL 700 for plane stress that does not default to an iterative approach. For MD Nastran SOL SEQ 106, 129, and 600, this material model is similar to using the MATS1 option to define an elasticplastic material.
Material Model 13: Isotropic ElasticPlastic with Failure
This highly simplistic failure model is occasionally useful. Material model 12 is called to update the stress tensor. Failure is initially assumed to occur if either
p n + 1 < p mi n
(768)
or
p p ε ef f > ε ma x
(769)
p where p mi n and ε ma x are userdefined parameters. Once failure has occurred, pressure may never be negative and the deviatoric components are set to zero:
si j = 0
(770)
for all time. The failed element can only carry loads in compression.
Material Model 14: Soil and Crushable Foam With Failure
This material model provides the same stress update as model 5. However, if pressure ever reaches its cutoff value, failure occurs and pressure can never again go negative. In material model 5, the pressure is limited to its cutoff value in tension.
Main Index
Chapter 7: Materials 251
Lagrangian Material Models for SOL 700
Material Model 15: Johnson and Cook Plasticity Model
Johnson and Cook express the flow stress as
n · σ y = ( A + B ε p ) ( 1 + C ln ε∗ ) ( 1 – T∗ m )
(771)
where
A, B, C, n,
and
m
are userdefined input constants, and:
ε p = effective plastic strain ·p ε · ε∗ = ·ε0
effective plastic strain rate for
· ε 0 = 1 s –1
T – T room T∗ = T mel t – T room
Constants for a variety of materials are provided in Johnson and Cook [1983]. Due to the nonlinearity in the dependence of flow stress on plastic strain, an accurate value of the flow stress requires iteration for the increment in plastic strain. However, by using a Taylor series expansion with linearization about the current time, we can solve for σ y with sufficient accuracy to avoid iteration. The strain at fracture is given by
ε f = [ D 1 + D 2 exp D 3 s∗ ] [ 1 + D 4 ln ε∗ ] [ 1 + D 5 T∗ ]
(772)
σ∗
where
D i , i = 1 ,… ,5 ,
are input constants and
is the ratio of pressure divided by effective stress: (773)
p σ∗ = σ eff
Fracture occurs when the damage parameter,
D =
∑ εf
Δ εp
(774)
reaches the value 1. A choice of three spall models is offered to represent material splitting, cracking, and failure under tensile loads. The pressure limit model limits the minimum hydrostatic pressure to the specified value, p ≥ p min . If pressures more tensile than this limit are calculated, the pressure is reset to p mi n . This option is not strictly a spall model since the deviatoric stresses are unaffected by the pressure reaching the tensile cutoff and the pressure cutoff value p mi n remains unchanged throughout the analysis. The maximum principal stress spall model detects spall if the maximum principal stress, σ ma x , exceeds the limiting value σ p . Once spall is detected with this model, the deviatoric stresses are reset to zero and no hydrostatic tension is permitted. If tensile pressures are calculated, they are reset to 0 in the spalled material. Thus, the spalled material behaves as rubble. The hydrostatic tension spall model detects spall
Main Index
252 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
if the pressure becomes more tensile than the specified limit, p mi n . Once spall is detected, the deviatoric stresses are set to zero and the pressure is required to be compressive. If hydrostatic tension is calculated then the pressure is reset to 0 for that element. In addition to the above failure criterion, this material model also supports a shell element deletion criterion based on the maximum stable time step size for the element, Δ t max . Generally, Δ t ma x goes down as the element becomes more distorted. To assure stability of time integration, the global time step is the minimum of the Δ t max values calculated for all elements in the model. Using this option allows the selective deletion of elements whose time step Δ t max has fallen below the specified minimum time step, Δ t cri t . Elements which are severely distorted often indicate that material has failed and supports little load, but these same elements may have very small time steps and therefore control the cost of the analysis. This option allows these highly distorted elements to be deleted from the calculation, and, therefore, the analysis can proceed at a larger time step, and, thus, at a reduced cost. Deleted elements do not carry any load, and are deleted from all applicable slide surface definitions. Clearly, this option must be judiciously used to obtain accurate results at a minimum cost. Material type 15 is applicable to the high rate deformation of many materials including most metals. Unlike the SteinbergGuinan model, the JohnsonCook model remains valid down to lower strain rates and even into the quasistatic regime. Typical applications include explosive metal forming, ballistic penetration, and impact. This material is similar to the use of the ISOTROPIC option with the JohnsonCook hardening rule.
Material Type 18: Power Law Isotropic Plasticity
Elastoplastic behavior with isotropic hardening is provided by this model. The yield stress, function of plastic strain and obeys the equation:
σ y = k ε n = k ( ε yp + ε p ) n σy ,
is a
(775)
εp
where
ε yp
is the elastic strain to yield and
is the effective plastic strain (logarithmic).
A parameter, SIGY, in the input governs how the strain to yield is identified. If SIGY is set to zero, the strain to yield if found by solving for the intersection of the linearly elastic loading equation with the strain hardening equation:
σ = Eε σ = lε n
(776)
which gives the elastic strain at yield as:
E ε yp = ⎛  ⎞ ⎝ k⎠
1 n–1
(777)
Main Index
Chapter 7: Materials 253
Lagrangian Material Models for SOL 700
If SIGY yield is nonzero and greater than 0.02 then:
σy ε yp ⎛ ⎞ ⎝ k⎠
1 n
(778)
Strain rate is accounted for using the Cowper and Symonds model which scales the yield stress with the factor
· ε 1/p 1 + ⎛ ⎞ ⎝ C⎠
(779)
· where ε is the strain rate. A fully viscoplastic formulation is optional with this model which incorporates the Cowper and Symonds formulation within the yield surface. An additional cost is incurred but the improvement is results can be dramatic.
This material model is a subset of what may be specified through the MATEP option for SOL600.
Material Type 19: Elastic Plastic Material Model with Strain Rate Dependent Yield
In this model, a TABLED1 is used to describe the yield strength
· ε σ0
as a function of effective strain rate
where (780)
· 2· · 1⁄2 ε = ⎛  ε' ij ε' ij⎞ ⎝3 ⎠
and the prime denotes the deviatoric component. The yield stress is defined as
· σy = σ0 ( ε ) + Ep εp
(781)
Ep
where ε p is the effective plastic strain and modulus by
E Et E p = E – Et
is given in terms of Young’s modulus and the tangent
(782)
Both Young's modulus and the tangent modulus may optionally be made functions of strain rate by specifying a TABLED1 ID giving their values as a function of strain rate. If these TABLED1 ID's are input as 0, then the constant values specified in the input are used. Note: All TABLED1s used to define quantities as a function of strain rate must have the same number of points at the same strain rate values. This requirement is used to allow vectorized interpolation to enhance the execution speed of this constitutive model.
Main Index
254 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
This model also contains a simple mechanism for modeling material failure. This option is activated by specifying a TABLED1 ID defining the effective stress at failure as a function of strain rate. For solid elements, once the effective stress exceeds the failure stress the element is deemed to have failed and is removed from the solution. For shell elements the entire shell element is deemed to have failed if all integration points through the thickness have an effective stress that exceeds the failure stress. After failure the shell element is removed from the solution. In addition to the above failure criterion, this material model also supports a shell element deletion criterion based on the maximum stable time step size for the element, Δ t max . Generally, Δ t ma x goes down as the element becomes more distorted. To assure stability of time integration, the global time step is the minimum of the Δ t max values calculated for all elements in the model. Using this option allows the selective deletion of elements whose time step Δ t max has fallen below the specified minimum time step, Δ t cri t . Elements which are severely distorted often indicate that material has failed and supports little load, but these same elements may have very small time steps and therefore control the cost of the analysis. This option allows these highly distorted elements to be deleted from the calculation, and, therefore, the analysis can proceed at a larger time step, and, thus, at a reduced cost. Deleted elements do not carry any load, and are deleted from all applicable slide surface definitions. Clearly, this option must be judiciously used to obtain accurate results at a minimum cost. This material model is a subset of what may be specified through the MATEP option for SOL600.
Material Type 20: Rigid
The rigid material type 20 provides a convenient way of turning one or more parts comprised of beams, shells, or solid elements into a rigid body. Approximating a deformable body as rigid is a preferred modeling technique in many real world applications. For example, in sheet metal forming problems the tooling can properly and accurately be treated as rigid. In the design of restraint systems the occupant can, for the purposes of early design studies, also be treated as rigid. Elements which are rigid are bypassed in the element processing and no storage is allocated for storing history variables; consequently, the rigid material type is very cost efficient. Two unique rigid part IDs may not share common nodes unless they are merged together using the rigid body merge option. A rigid body may be made up of disjoint finite element meshes, since this is a common practice in setting up tooling meshes in forming problems. All elements which reference a given part ID corresponding to the rigid material should be contiguous, but this is not a requirement. If two disjoint groups of elements on opposite sides of a model are modeled as rigid, separate part ID's should be created for each of the contiguous element groups if each group is to move independently. This requirement arises from the fact that SOL 700 internally computes the six rigid body degreesoffreedom for each rigid body (rigid material or set of merged materials), and if disjoint groups of rigid elements use the same part ID, the disjoint groups will move together as one rigid body.
Main Index
Chapter 7: Materials 255
Lagrangian Material Models for SOL 700
Inertial properties for rigid materials may be defined in either of two ways. By default, the inertial properties are calculated from the geometry of the constituent elements of the rigid material and the density specified for the part ID. Alternatively, the inertial properties and initial velocities for a rigid body may be directly defined, and this overrides data calculated from the material property definition and nodal initial velocity definitions. Young's modulus, E , and Poisson's ratio, υ , are used for determining sliding interface parameters if the rigid body interacts in a contact definition. Realistic values for these constants should be defined since unrealistic values may contribute to numerical problem in contact.
Material Model 22: ChangChang Composite Failure Model
Five material parameters are used in the three failure criteria based upon Chang and Chang 1987a, 1987b:
• S 1 , longitudinal tensile strength • , S 2 transverse tensile strength • , S 12 shear strength • C 2 , transverse compressive strength • α , nonlinear shear stress parameter.
and C 2 are obtained from material strength measurement. α is defined by material shear stressstrain measurements. In plane stress, the strain is given in terms of the stress as
1 ε 1 =  ( σ 1 – υ 1 σ 2 ) E1 1ε 2 =  ( σ 2 – υ 2 σ 1 ) E2 1 3 2ε 12 =  τ 12 + α τ 12 G 12
(783)
The third equation defines the nonlinear shear stress parameter A fiber matrix shearing term augments each damage mode:
2 τ 12 3  +  α τ 12   4 2G 12 4 τ = 2 S 12 3  +  αS 12   4 2G 12 4
α.
(784)
which is the ratio of the shear stress to the shear strength. The matrix cracking failure criteria is determined from
σ2 2 F matrix = ⎛ ⎞ + τ ⎝ S 2⎠
(785)
Main Index
256 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
where failure is assumed whenever υ 2 are set to zero.
F m atrix > 1 . If F m atrix > 1 , then the material
constants
E 2 , G 12 , υ , and
The compression failure criteria is given as
σ2 2 C2 2 σ2 F comp = ⎛  ⎞ + ⎛  ⎞ – 1  + τ ⎝ 2 S 12⎠ ⎝ 2 S 12⎠ C2
(786)
F comp > 1 . If F comp > 1 , then the material constants E 2 , υ 1 , and υ 2
where failure is assumed whenever set to zero.
are
The final failure mode is due to fiber breakage.
σ1 2 F fi ber = ⎛ ⎞ + τ ⎝ S 1⎠
(787)
F fi ber > 1 .
Failure is assumed whenever to zero.
If
F fi ber > 1 ,
then the constants
E 1 , E 2 , G 12 , υ 1 ,
and
υ2
are set
Material Model 24: Piecewise Linear Isotropic Plasticity
This plasticity treatment in this model is quite similar to Model 3 but only isotropic hardening occurs. Deviatoric stresses are determined that satisfy the yield function
2 σy 1 φ =  s ij s i j –  ≤ 0 2 3
(788)
where
p σ y = β [ σ 0 + f h ( ε eff ) ]
(789)
p f h ( ε eff )
where the hardening function hardening of the form
p p f h ( ε eff ) = E p ( ε ef f )
can be specified in tabular form as an option. Otherwise, linear
(790)
p is assumed where E p and ε ef f are given in Equations (716) and (717), respectively. The parameter β accounts for strain rate effects. For complete generality a table defining the yield stress versus plastic strain may be defined for various levels of effective strain rate.
In the implementation of this material model, the deviatoric stresses are updated elastically (see material model 1), the yield function is checked, and if it is satisfied the deviatoric stresses are accepted. If it is not, an increment in plastic strain is computed:
⎛ 3 s * s *⎞ – σy ⎝ 2 i j ij⎠ = 3G + E p
1⁄2
p Δ ε ef f
(791)
Main Index
Chapter 7: Materials 257
Lagrangian Material Models for SOL 700
is the shear modulus and is scaled back:
σy s inj + 1 =  s i*j 1⁄2 ⎛3 s* s*⎞  i j ij ⎝2 ⎠
Ep
is the current plastic hardening modulus. The trial deviatoric stress state
s i*j
(792)
For shell elements, the above equations apply, but with the addition of an iterative loop to solve for the normal strain increment, such that the stress component normal to the mid surface of the shell element approaches zero. Three options to account for strain rate effects are possible:
• Strain rate may be accounted for using the Cowper and Symonds model which scales the yield
stress with the factor
· ε 1⁄p β = 1 + ⎛ ⎞ ⎝ C⎠
(793)
where
· ε
is the strain rate.
• For complete generality a TABLED1, defining β , which scales the yield stress may be input
instead. In this curve the scale factor versus strain rate is defined.
• If different stress versus strain curves can be provided for various strain rates, the option using
the reference to a table definition can be used. A fully viscoplastic formulation is optional which incorporates the different options above within the yield surface. An additional cost is incurred over the simple scaling but the improvement is results can be dramatic. If a TABLE ID is specified a curve ID is given for each strain rate. Intermediate values are found by interpolating between curves.
Material Model 26: Crushable Foam
This orthotropic material model does the stress update in the local material system denoted by the subscripts, a , b , and c . The material model requires the following input parameters:
• E , Young’s modulus for the fully compacted material; • ν , Poisson’s ratio for the compacted material; • σ y , yield stress for fully compacted honeycomb; • LCA, TABLED1 number for sigmaaa versus either relative volume or volumetric strain (see Figure 73);
Note:
In Figure 73, the “yield stress” at a volumetric strain of zero is nonzero. In the TABLED1 definition, the “time” value is the volumetric strain and the “function” value is the yield stress.
Main Index
258 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
σi j Unloading and reloading path
0
Strain
–εi j Unloading is based on the interpolated Young’s moduli which must provide an unloading tangent that exceeds the loading tangent.
Curve extends into negative strain quadrant since SOL 700 extrapolates using the two end points. It is important that the extrapolation does not extend into the negative stress region.
Figure 73
Stress Quantity Versus Volumetric Strain
• LCB, TABLED1 number for sigmabb versus either relative volume or volumetric strain (default:
LCB = LCA);
• LCC, the TABLED1 number for sigmacc versus either relative volume or volumetric strain
(default: LCC = LCA);
• LCS, the TABLED1 number for shear stress versus either relative volume or volumetric strain
(default LCS = LCA);
• V f , relative volume at which the honeycomb is fully compacted; • E aa u , elastic modulus in the uncompressed configuration; • E bb u , elastic modulus in the uncompressed configuration; • E ccu , elastic modulus in the uncompressed configuration; • G abu , elastic shear modulus in the uncompressed configuration; • G bcu , elastic shear modulus in the uncompressed configuration; •
G cau ,
elastic shear modulus in the uncompressed configuration;
• LCAB, TABLED1 number for sigmaab versus either relative volume or volumetric strain
(default: LCAB = LCS);
• LCBC, TABLED1 number for sigmabc versus either relative volume or volumetric strain
default: LCBC = LCS);
Main Index
Chapter 7: Materials 259
Lagrangian Material Models for SOL 700
• LCCA, TABLED1 number for sigmaca versus either relative volume or volumetric strain
(default: LCCA = LCS);
• LCSR, optional TABLED1 number for strain rate effects.
The behavior before compaction is orthotropic where the components of the stress tensor are uncoupled; i.e., an a component of strain will generate resistance in the local a direction with no coupling to the local b and c directions. The elastic moduli vary linearly with the relative volume from their initial values to the fully compacted values:
E aa = E aau + β ( E – E aau ) E bb = E bbu + β ( E – E bbu ) E cc = E ccu + β ( E – E ccu ) G ab = G abu + β ( G – G abu ) G bc = G bcu + β ( G – G bcu ) G ca = G cau + β ( G – G cau )
(794)
where
1 – V min β = ma x min ⎛  , 1⎞ , 0 ⎝ 1 – Vf ⎠
(795)
and
G
is the elastic shear modulus for the fully compacted honeycomb material (796)
E G = 2(1 + ν)
The relative volume V is defined as the ratio of the current volume over the initial volume; typically, V = 1 at the beginning of a calculation. The relative volume, V min , is the minimum value reached during the calculation. The TABLED1s define the magnitude of the average stress as the material changes density (relative volume). Each curve related to this model must have the same number of points and the same abscissa values. There are two ways to define these curves: as a function of relative volume V , or as a function of volumetric strain defined as:
εV = 1 – V
(797)
In the former, the first value in the curve should correspond to a value of relative volume slightly less than the fully compacted value. In the latter, the first value in the curve should be less than or equal to zero corresponding to tension and should increase to full compaction. Note: When defining the curves, care should be taken that the extrapolated values do not lead to negative yield stresses.
Main Index
260 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
At the beginning of the stress update we transform each element’s stresses and strain rates into the local element coordinate system. For the uncompacted material, the trial stress components are updated using the elastic interpolated moduli according to:
σ aa σ bb σ cc
n+1 tri al
= σ aa + E aa Δ ε aa = σ bb + E bb Δ ε bb = σ cc + E cc Δ ε cc =
n σ ab n n n
n
n+1
tri al
n+1
tri al
n+1 σ ab
tri al
(798)
+ E ab Δ ε ab
σ bc σ ca
n+1
tri al
= σ bc + E bc Δ ε bc = σ ca + E ca Δ ε ca
n
n+1
tri al
Then we independently check each component of the updated stresses to ensure that they do not exceed the permissible values determined from the TABLED1s; e.g., if
σ inj + 1
t rial
> λ σ i j ( V min )
(799)
then
n λ σ ij + 1 σ inj + 1 = σ i j ( V min ) tri al n σ ij + 1 t rial
(7100)
The parameter λ is either unity or a value taken from the TABLED1 number, LCSR, that defines λ as a function of strain rate. Strain rate is defined here as the Euclidean norm of the deviatoric strain rate tensor. For fully compacted material we assume that the material behavior is elasticperfectly plastic and updated the stress components according to:
n dev s itjrial = s ij + 2G Δ ε ij n+1⁄2
(7101)
Where the deviatoric strain increment is defined as:
1 Δ ε idev = Δ ε ij –  Δ ε kk δ i j j 3
(7102)
We next check to see if the yield stress for the fully compacted material is exceeded by comparing:
3 tri tri 1 ⁄ 2 t rial s eff = ⎛  s ij al s ij al⎞ ⎝2 ⎠
(7103)
Main Index
Chapter 7: Materials 261
Lagrangian Material Models for SOL 700
the effective trial stress, to the yield stress σ y . If the effective trial stress exceeds the yield stress, we simply scale back the stress components to the yield surface:
σy s inj + 1 =  s itjrial t rial s eff
(7104)
K:
We can now update the pressure using the elastic bulk modulus,
n p n + 1 = p n – K Δ ε kk+ 1 ⁄ 2
E K = 3 (1 – 2 ν)
(7105)
and obtain the final value for the Cauchy stress:
n σ inj + 1 = s ij + 1 – p n + 1 δ ij
(7106)
After completing the stress update, we transform the stresses back to the global configuration.
Material Model 27: Incompressible MooneyRivlin Rubber
This material model, available for solid elements only, provides an alternative to the BlatzKo rubber model. The strain energy density function is defined as in terms of the input constants A , B ,, and υ as:
1 W ( I 1 ,I 2 ,I 2 ) = A ( I 1 – 3 ) + B ( I 2 – 3 ) + C ⎛  – 1⎞ + D ( I 3 – 1 ) 2 ⎝I2 ⎠ 3
(7107)
where
C = .5∗ A + B A ( 5υ – 2 ) + B ( 11υ – 5 ) D = 2 (1 – 2 υ) υ
(7108) (7109)
= Poisson’s ratio = shear modulus of linear elasticity
G = 2(A + B) I 1 ,I 2 ,I 3
2
= strain invariants in terms of the principal stretches:
2 2 2 2
I1 = λ 1 + λ 2 + λ3 I2 = λ 1 λ 2 + λ 2 λ 3 + λ 3 λ1 I3 = λ 1 λ 2 λ3
2 2 2 2 2 2 2
(7110)
Recommended values for Poisson’s ratio are between .490 and .495 or higher. Lower values may lead to instabilities. In the derivation of the constants C and D incompressibility is assumed.
Main Index
262 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
Material Model 28: Resultant Plasticity
This plasticity model, based on resultants as illustrated in Figure 74, is very cost effective but not as accurate as throughthickness integration. This model is available only with the C 0 triangular, BelytschkoTsay shell, and the Belytschko beam element since these elements, unlike the HughesLiu elements, lend themselves very cleanly to a resultant formulation. The elements are set by the SHELLFORM parameter.
Membrane
σy
h 2 σ dζ h y – 2
ny =
∫
= hσ y
Bending
σy
h 2
my =
∫– h σ y ζ dζ 2
h2 =  σ y 4
Figure 74
Full Section Yield using Resultant Plasticity
In applying this model to shell elements the resultants are updated incrementally using the midplane strains ε m and curvatures κ :
Δ n = Δ tC ε m h3 Δ m = Δ t  C κ 12
(7111) (7112)
E
where the plane stress constitutive matrix is given in terms of Young’s Modulus ν as:
2 2 2 m = m xx – m xx m yy + m yy + 3 m xy
and Poisson’s ratio (7113)
Defining
2 2 2 n = n xx – n xx n yy + n yy + 3n xy 2 2 2 m = m xx – m xx m yy + m yy + 3 m xy
(7114) (7115) (7116)
1 1 mn = m xx n xx –  m xx n yy –  n xx m yy + m y n y + 3m xy n xy 2 2
the Ilyushin yield function becomes
4 mn 16 m 2 2 f ( m, n ) = n +  +  ≤ n y = h 2 σ y h2 h 3
(7117)
Main Index
Chapter 7: Materials 263
Lagrangian Material Models for SOL 700
In our implementation we update the resultants elastically and check to see if the yield condition is violated:
2 f ( m, n ) > n y
(7118)
α:
If so, the resultants are scaled by the factor
α =
2 ny f ( m, n )
(7119)
We update the yield stress incrementally:
ef f n n σ y + 1 = σ y + E p Δ ε pl ast ic
(7120)
where
Ep
is the plastic hardening modulus which in incremental plastic strain is approximated by (7121)
f ( m, n ) – n y ef f Δ ε p la st ic = h ( 3 G + Ep )
Kennedy, et. al., report that this model predicts results that may be too stiff; users of this model should proceed cautiously. In applying this material model to the Belytschko beam, the flow rule changes to
ˆ2 ˆ2 4 my 4 mz 2 2 f ( m, n ) = ˆx +  +  ≤ n y = A 2 σ y f2 3 I yy 3 I zz
(7122)
have been updated elastically according to Equations (677) through (679). The yield condition is checked Equation (7118), and if it is violated, the resultants are scaled as described above. This model is frequently applied to beams with nonrectangular cross sections. The accuracy of the results obtained should be viewed with some healthy suspicion. No workhardening is available with this model.
Material Model 29: FORCE LIMITED Resultant Formulation
This material model is available for the Belytschko beam element only. Plastic hinges form at the ends of the beam when the moment reaches the plastic moment. The momentversus rotation relationship is specified by the user in the form of a TABLED1 and scale factor. The point pairs of the TABLED1 are (plastic rotation in radians, plastic moment). Both quantities should be positive for all points, with the first point pair being (zero, initial plastic moment). Within this constraint any form of characteristic may be used including flat or falling curves. Different TABLED1s and scale factors may be specified at each node and about each of the local s and t axes.
Main Index
264 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
Membrane
σy
h 2 σ dζ h y – 2
ny =
∫
= hσ y
Bending
σy
h 2 σ ζ dζ h y – 2
my =
∫
h2 =  σ y 4
Figure 75
Full Section Yield using Resultant Plasticity
Axial collapse occurs when the compressive axial load reaches the collapse load. The collapse loadversuscollapse deflection is specified in the form of a TABLED1. The points of the TABLED1 are (true strain, collapse force). Both quantities should be entered as positive for all points, and will be interpreted as compressive i.e., collapse does not occur in tension. The first point should be the pair (zero, initial collapse load). The collapse load may vary with end moment and with deflection. In this case, several loaddeflection curves are defined, each corresponding to a different end moment. Each TABLED1 should have the same number of point pairs and the same deflection values. The end moment is defined as the average of the absolute moments at each end of the beam, and is always positive. It is not possible to make the plastic moment vary with axial load.
M8 M7 M6 M5 M4 Force M3 M2 M1
Displacement
Figure 76
The Force Magnitude Limited Applied End Moment
Main Index
Chapter 7: Materials 265
Lagrangian Material Models for SOL 700
For an intermediate value of the end moment, MD Nastran SOL 700 interpolates between the curves to determine the allowable force. A corotational technique and momentcurvature relations are used to compute the internal forces. The corotational technique will not be treated here as we will focus solely on the internal force update and computing the tangent stiffness. For this, we use the notation:
E G A As ln ln + 1 I yy I zz J ei yI zI
= Young’s modulus = Shear modulus = Cross sectional area = Effective area in shear = Reference length of beam = Current length of beam = Second moment of inertia about y = Second moment of inertia about z = Polar moment of inertia = ith local base vector in the current configuration = nodal vector in y direction at node I in the current configuration = nodal vector in z direction at node I in the current configuration
We emphasize that the local y and z base vectors in the reference configuration always coincide with the corresponding nodal vectors. The nodal vectors in the current configuration are updated using the HughesWinget formula while the base vectors are computed from the current geometry of the element and the current nodal vectors. Internal Forces Elastic Update In the local system for a beam connected by nodes I and J, the axial force is updated as
fa = f a + K a δ
el n el
(7123)
where
EA el K a = ln δ = l n + 1 – l n ..
(7124) (7125)
The torsional moment is updated as
Main Index
266 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
mt
el
= mt + Kt θt
n
el
(7126)
where
Kt
el
GJ = ln
(7127) (7128)
1 T θ t =  e l ( y I × y J + z I × z J ) . 2
The bending moments are updated as
m y = m y + A y θy m z = m z + A z θz
el n el el n el
(7129) (7130)
where
1 E I ** ⎛ 4 + φ * 2 – φ * ⎞ el A * =   ⎜ ⎟ 1 + φ* ln ⎝ 2 – φ 4 + φ ⎠ * * 12E I ** φ * = GA * l n t n
T T θ y = e 3 ⎛ yI × zI y J × zJ ⎞ ⎝ ⎠
(7131)
(7132) (7133) (7134)
A*
el
T T θ z = e 2 ⎛ yI × zI y J × zJ ⎞ ⎝ ⎠
In the following, we refer to Plastic Correction
as the (elastic) momentrotation matrix.
After the elastic update the state of force is checked for yielding as follows. As a preliminary note we emphasize that whenever yielding does not occur the elastic stiffnesses and forces are taken as the new stiffnesses and forces. The yield moments in direction i at node I as functions of plastic rotations are denoted mYI ( θ iPI ) . This i function is given by the user but also depends on whether a plastic hinge has been created. Whenever the elastic moment exceeds the plastic moment, the plastic rotations are updated as
θ iI
P(n + 1)
= θi I
P(n)
m iI – m iI + Y ∂m iI ⎞ ⎛ el ma x ⎜ 0.001, A i ( I I ) +  ⎟ P ⎝ ∂θ iI ⎠
el
Y
(7135)
and the moment is reduced to the yield moment
Main Index
Chapter 7: Materials 267
Lagrangian Material Models for SOL 700
mi I
n+1
= m iI ( θ iI
Y
P (n + 1)
) sgn ( m iI )
el
(7136)
The corresponding diagonal component in the momentrotation matrix is reduced as
⎛ ⎞ ⎜ ⎟ el Ai(I I) ⎜ ⎟ el = A i ( II ) ⎜ 1 – α ⎟ Y ⎜ ∂m iI ⎞ ⎟ ⎛ el ⎜ ma x ⎜ 0.001, A i ( I I ) +  ⎟ ⎟ P ⎝ ⎝ ∂θ iI ⎠ ⎠ α≤1
A i ( II )
n+1
(7137)
where
is a parameter chosen such that the momentrotation matrix remains positive definite.
The yield moment in torsion is given by mYt ( θP ) and is provided by the user. If the elastic torsional t moment exceeds this value, the plastic torsional rotation is updated as
θt
P (n + 1)
= θt
P(n)
mt – mt + Y ∂m t ⎞ ⎛ el ma x ⎜ 0.0001, K t + ⎟ P ⎝ ∂θ t ⎠
el
Y
(7138)
and the moment is reduced to the yield moment
mt
n+1
= mt ( θt
Y
P(n + 1)
) sgn ( m t )
el
The torsional stiffness is modified as
⎛ ⎞ ⎜ ⎟ el Kt ⎟ el ⎜ = K t ⎜ 1 – α  ⎟ Y ⎜ ∂m t ⎟ el ⎜ K t + ⎟ P ⎝ ∂θ t ⎠ α≤1
Kt
n+1
(7139)
where again
is chosen so that the stiffness is positive.
Axial collapse is modeled by limiting the axial force by fY ( ε, m ) ; i.e., a function of the axial strains and a the magnitude of bending moments. If the axial elastic force exceeds this value it is reduced to yield
fa
n+1
= fa ( ε
Y
n+1
,m
n+1
) sgn ( f a )
el
(7140)
and the axial stiffness is given by
Ka
n+1 el ∂f a = ma x ⎛ 0.05 K a , ⎞ ⎝ ∂ε ⎠ Y
(7141)
We neglect the influence of change in bending moments when computing this parameter.
Main Index
268 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
Damping Damping is introduced by adding a viscous term to the internal force on the form
δ d θt f v = D dt θ y θz Ka D = γ
el
(7142)
Kt
el
Ay
el
(7143)
Az
el
where
γ
is a damping parameter.
Transformation The internal force vector in the global system is obtained through the transformation
fg
n+1
= Sf t
n+1
(7144)
where
– e1 0 – e3 ⁄ l n + 1 –e 3 ⁄ l n + 1 e2 ⁄ l n + 1 S = 0 – e1 e1 0 0 e1 fa fl
n+1 n+1 n+1 n+1 n+1
e2 ⁄ ln + 1 0 –e2 ⁄ l n + 1 e3
e2 e3 ⁄ ln + 1 0 e3 ⁄
0 ln + 1 e2
e3 – e2 ⁄ ln + 1 0
(7145)
=
mt
(7146)
my mz
Tangent Stiffness Derivation The tangent stiffness is derived from taking the variation of the internal force
δf g
n+1
= δSf l
n+1
+ Sδf l
n+1
(7147)
which can be written
δf g
n+1
= K geo δu + K mat δu
Main Index
Chapter 7: Materials 269
Lagrangian Material Models for SOL 700
where
δu = ⎛ δxT δωT δxT δωT ⎞ I I J j ⎠ ⎝
T
.
(7148)
There are two contributions to the tangent stiffness: geometrical and material. The geometrical contribution is given (approximately) by
K geo = R ( f l
n+1
1 ⊗ I )W –  T f ln + 1 L ln + 1 ln + 1
(7149)
where
R1 R = 0 – R1 0 0 R1 0 – R1 R3 ⁄ ln + 1 –R 2 –R 3 ⁄ ln + 1 0 R3 ⁄ l n + 1 –R 2 ⁄ ln + 1 – R2 ⁄ ( n + 1 ) 0 – R3 ⁄ ln + 1 –R 3 R2 ⁄ ln + 1 0 0 R2 ⁄ ln + 1 –R 3
(7150)
– R2
W = ⎛ – R 1 ⁄ l n + 1 e 1 eT ⁄ 2 R 1 ⁄ l n + 1 e 1 eT ⁄ 2 ⎞ 1 1 ⎝ ⎠ 0 0 – e3 – e3 e 2 T = 0 0 0 0 0 e3 0 0 0 e2
(7151)
0 0 0 e3 –e2 –e2 0 0 0
(7152)
L = ⎛ eT 0 eT 0 ⎞ 1 ⎝ 1 ⎠
(7153)
⊗
and I is the 3 by 3 identity matrix. We use
Ri v = ei × v .
as the outer matrix product and define (7154)
The material contribution can be written as
K mat = SK S T
(7155)
where
Ka K =
n+1
Kt
n+1
Ay
n+1
(7156)
Az
n+1
Main Index
270 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
Material Model 30: Shape Memory Alloy
This section presents the mathematical details of the shape memory alloy material in MD Nastran SOL 700. The description closely follows the one of Auricchio and Taylor [1997] with appropriate modifications for this particular implementation. Mathematical Description of the Material Model The Kirchhoff stress
τ = pi + t τ
in the shape memory alloy can be written (7157)
where i is the second order identity tensor and
p = K ( θ – 3 αξ S ε L ) t = 2G ( e – ξ S ε L n )
.
(7158)
Here K and G are bulk and shear modulii, θ and e are volumetric and shear logarithmic strains and and ε L are constant material parameters. There is an option to define the bulk and shear modulii as functions of the martensite fraction according to
K = KA + ξS ( K S – KA ) G = G A + ξ S ( G S – GK A )
α
(7159)
in case the stiffness of the martensite differs from that of the austenite. Furthermore, the unit vector n is defined as
n = e ⁄ ( e + 10 – 12 )
(7160)
and a loading function is introduced as
F = 2 G e + 3α Kθ – β ξ S
(7161)
where
β = ( 2G + 9α 2 K )ε L .
(7162)
Main Index
Chapter 7: Materials 271
Lagrangian Material Models for SOL 700
For the evolution of the martensite fraction
⎫ AS F – Rs > 0 ⎪ · ⎪ F · ⎬ ⇒ ξ S = – ( 1 – ξ S ) F>0 AS F – Rf ⎪ ξS < 1 ⎪ ⎭ ⎫ SA F – Rs < 0 ⎪ · ⎪ F · ⎬ ⇒ ξ S = ξ S F>0 SA F – Rf ⎪ ξS > 0 ⎪ ⎭
ξS
in the material, the following rule is adopted
.
(7163)
Here
Rs
AS
,
Rf
AS
,
Rs
SA
, and
Rf
SA
are constant material parameters. The Cauchy stress is finally obtained as (7164)
σ = τ⁄J
where J is the Jacobian of the deformation. Algorithmic Stress Update For the stress update, we assume that the martensite fraction ξ sn and the value of the loading function F n is known from time t n and the deformation gradient at time t n + 1 , F , is known. We form the left CauchyGreen tensor as B = F F T which is diagonalized to obtain the principal values and directions and Q . The volumetric and principal shear logarithmic strains are given by
θ = log ( J ) e i = log ( λ i ⁄ J 1 ⁄ 3 ) Λdi a g ( λ i )
where
J = λ1 λ 2 λ 3
n is the total Jacobian of the deformation. Using Equation (7161) with ξS = ξ S , a value F tri al of the loading function can be computed. The discrete counterpart of Equation (7163) becomes
⎫ AS F tri al – R s > 0 ⎪ AS AS F t ria l – β Δ ξ S – min ( ma x ( F n ,R s ) ,R f ) ⎪ n tri al – F n > 0 ⎬ ⇒ Δ ξ S = – ( 1 – ξ S – Δ ξ S ) F AS F tri al – β Δ ξ S – R f ⎪ n ξS < 1 ⎪ ⎭
(7165)
⎫ SA F tri al – R s > 0 ⎪ SA SA F tri al – βΔ ξ S – m in ( max ( F n ,R s ) ,R f ) ⎪ n F tri al – F n > 0 ⎬ ⇒ Δ ξ S = ( ξ S + Δ ξ S ) SA F tri a l – βΔ ξ S – R f ⎪ n ξS < 1 ⎪ ⎭
Main Index
272 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
n If none of the two conditions to the left are satisfied, set ξ S + 1
= ξS
n
, Fn + 1
= F t rial , and compute the stress
n using Equations (7157), (7158), (7159), (7164) and ξ S = ξ S . When phase transformation occurs according to a condition to the left, the corresponding equation to the right is solved for Δ ξ S . If the bulk
σn + 1
and shear modulii are constan,t this is an easy task. Otherwise, parameter and makes things a bit more tricky. We have that
ES – EA tri F tri al = F n al ⎛ 1 +  Δ ξ S⎞ ⎝ ⎠ En ES – EA β = β n ⎛ 1 +  Δξ S⎞ ⎝ ⎠ En
F tri al
as well as
β
depends on this
.
where
ES
and
EA
are Young’s modulii for martensite and austenite, respectively. The subscript n is
tn .
introduced for constant quantities evaluated at time relations are written
tri F tri al = F n al + ΔF t rial Δ ξ S
To simplify the upcoming expressions, these
(7166)
β = βn +
Δ β Δ ξS
respectively, where we have for simplicity set Inserting these expressions into Equation (7165) results in
f ( Δ ξS ) = Δ β ( 1 – ξ S ) Δ ξS + ( Rf
n 2 AS
˜n n – F AS + ( β n – Δ F tri al ) ( 1 – ξ S ) )Δ ξ S +
n ˜n t rial ( 1 – ξ S ) ( F AS – F n ) = 0
(7167)
and
˜n n 2 SA n f ( Δ ξ S ) = Δ βξ S Δ ξ S + ( F SA – R f + ( β n – Δ F t ria l )ξ S ) Δ ξ S +
n ˜n tr ial ξ S ( F SA – F n ) = 0
respectively, where we have for simplicity set
˜n AS AS F AS = m in ( max ( F n, R s ) ,R f ) ˜n SA SA F SA = m in ( max ( F n, R f ) ,R s )
.
The solutions to these equations are approximated with two Newton iterations starting in the point
Δ ξ S = 0 ..
Now set
ξS
n+1
= min ( 1 ,ma x ( 0 ,ξ S + Δ ξ S ) )
n
and compute
σn + 1
and
Fn + 1
according to Equations
n (7157), (7158), (7159), (7160), (7162) and ξ S = ξ S + 1 .
Main Index
Chapter 7: Materials 273
Lagrangian Material Models for SOL 700
Tangent Stiffness Matrix An algorithmic tangent stiffness matrix relating a change in true strain to a corresponding change in Kirchhoff stress is derived in the following. Taking the variation of Equation (7158) results in
δp = K ( δθ – 3αδξ S ε L ) + δK ( θ – 3αξ S ε L ) δt = 2G ( δe – δξ S ε L n – ξ S ε L δn ) + 2δG ( e – ξ S ε L n )
.
(7168)
The variation of the unit vector in Equation (7159) can be written
1 δn =  ( I – n ⊗ n )δ e e + 10 – 12
where I is the fourth order identity tensor. For the variation of martensite fraction, we introduce the indicator parameters H AS and H SA that should give information of the probability of phase transformation occurring in the next stress update. Set initially H AS = H SA = 0 and change them according to
⎫ AS F tri al – R s > 0 ⎪ ⎪ AS = 1 F tri al – F n > 0 ⎬ ⇒ H ⎪ n ξ S + Δξ S ≤ 1 ⎪ ⎭ ⎫ SA F tri al – R s > 0 ⎪ ⎪ SA = 1 F tri al – F n > 0 ⎬ ⇒ H ⎪ n ξ S + Δξ S ≥ 0 ⎪ ⎭
using the quantities computed in the previous stress update. For the variation of the martensite fraction we take variations of Equations(7167) and (7168) with
δF n
tri al
= 2 Gn: δe + 3 αKδθ
which results in
δξ S = γ ( 2 Gn: δe + 3 αKδθ )
where
( 1 – ξ S ) H AS ξ S H SA γ =  +  . ˜n ˜n AS t rial n SA t ria l n R f – F A S + ( β n – Δ F n ) ( 1 – ξ S ) F SA – R f + ( β n – Δ F n ) ξ S
n n
As can be seen, we use the value of γ obtained in the previous stress update since this is easier to implement and will probably give a good indication of the current value of this parameter.
Main Index
274 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
The variation of the material parameters K and G results in
δK = ( K S – K A ) δξ S δG = ( G S – G A ) δξ S
and, finally, using the identities
n: δe = n: δε δθ = i : δε δτ = iδp + δt
results in
ξS εL ⎧ δτ = ⎨ 2 G ⎛ 1 –  ⎞ I dev + K ( 1 – 9 α 2 K γε L + 3αγ ( K S – K A ) ( θ – 3 αξ S ε L ) )i ⊗ i + ⎝ e + 10 – 12⎠ ⎩ 2 γG ( K S – K A ) ( θ – 3αξ S ε L ) i ⊗ n + 6γ αK ( G S – G A ) ( e – ξ S ε L )n ⊗ i + ξS εL 2 G ⎛  – 2 Gγ ε L + 2 γ ( G S – G A ) ( e – ξ D ε L )⎞ ⎝ e + 1 – 10 – 12 ⎠ n ⊗ n – 6K Gαγ ε L ( i ⊗ n + n ⊗ i ) }δε
where I dev is the fourth order deviatoric identity tensor. In general this tangent is not symmetric because of the terms on the second line in the expression above. We simply use a symmetrization of the tangent stiffness above in the implementation. Furthermore, we transform the tangent to a tangent closer related to the one that should be used in the MD Nastran SOL 700 implementation,
ξS εL ⎧ C = J – 1 ⎨ 2G ⎛ 1 –  ⎞ I dev + K ( 1 – 9α 2 K γ ε L + 3 αγ ( K S – K A ) ( θ – 3αξ S ε L ) )i ⊗ i + ⎝ e + 10 – 12⎠ ⎩ ( γG ( K S – K A ) ( θ – 3α ξ S ε L ) + 3γ αK ( G S – G A ) ( e – ξ S ε L ) – 6 K Gαγ ε L ) ( i ⊗ n + n ⊗ i ) + ξS εL ⎫ 2G ⎛  – 2 Gγ ε L + 2 γ ( G S – G A ) ( e – ξ S ε L )⎞ n ⊗ n ⎬ δε ⎝ e + 10 – 12 ⎠ ⎭
Material Model 31: Slightly Compressible Rubber Model
This model implements a modified form of the hyperelastic constitutive law first described in [Kenchington 1988]. The strain energy functional,
U,
is defined in terms of the input constants as: (7169)
2 3 4 2 2 U = C 100 I 1 + C 200 I 1 + C 300 I 1 + C 400 I 1 + C 110 I 1 I 2 + C 210 I 1 I 2 + C 010 I 2 + C 020 I 2 + j ( J )
Main Index
Chapter 7: Materials 275
Lagrangian Material Models for SOL 700
where the strain invariants can be expressed in terms of the deformation gradient matrix, GreenSt. Venant strain tensor, E ij :
J = F ij I1 = Ei i I2 = 1 ij  δ pq E pi E qj 2 U
F ij ,
and the
.
(7170)
The derivative of
∂U S i j = ∂ E ij
with respect to a component of strain gives the corresponding component of stress (7171)
where, S i j , is the second PiolaKirchhoff stress tensor which is transformed into the Cauchy stress tensor:
ρ ∂x i ∂x j  σ i j =    S kl ρ o ∂X k ∂X l
(7172)
where
ρo
and
ρ
are the initial and current density, respectively.
Material Model 32: Laminated Glass Model
This model is available for modeling safety glass. Safety glass is a layered material of glass bonded to a polymer material which can undergo large strains. The glass layers are modeled by isotropic hardening plasticity with failure based on exceeding a specified level of plastic strain. Glass is quite brittle and cannot withstand large strains before failing. Plastic strain was chosen for failure since is increases monotonically and, therefore, is insensitive to spurious numberical noise in the solution. The material to which the glass is bonded is assumed to stretch plastically without failure. The user defined integration rule option must be used with this material. The user defined rule specifies the thickness of the layers making up the safety glass. Each integration point is flagged with a zero if the layer is glass and with a one if the layer is polymer. An iterative plane stress plasticity algorithm is used to enforce the plane stress condition.
Material Model 34: Fabric
The fabric model is a variation on the Layered Orthotropic Composite material model (Material 22) and is valid for only 3 and 4 node membrane elements. This material model is strongly recommended for modeling airbags and seatbelts. In addition to being a constitutive model, this model also invokes a special membrane element formulation that is better suited to the large deformations experienced by fabrics. For thin fabrics, buckling (wrinkling) can occur with the associated inability of the structure to
Main Index
276 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
support compressive stresses; a material parameter flag is included for this option. A linear elastic liner is also included which can be used to reduce the tendency for these material/elements to be crushed when the nocompression option is invoked. If the airbag material is to be approximated as an isotropic elastic material, then only one Young’s modulus and Poisson’s ratio should be defined. The elastic approximation is very efficient because the local transformations to the material coordinate system may be skipped. If orthotropic constants are defined, it is very important to consider the orientation of the local material system and use great care in setting up the finite element mesh. If the reference configuration of the airbag is taken as the folded configuration, the geometrical accuracy of the deployed bag will be affected by both the stretching and the compression of elements during the folding process. Such element distortions are very difficult to avoid in a folded bag. By reading in a reference configuration, such as the final unstretched configuration of a deployed bag, any distortions in the initial geometry of the folded bag will have no effect on the final geometry of the inflated bag. This is because the stresses depend only on the deformation gradient matrix:
∂x i F ij = ∂X j
(7173)
where the choice of X j may coincide with the folded or unfold configurations. It is this unfolded configuration which may be specified here. When the reference geometry is used, then the nocompression option should be active. With the reference geometry, it is possible to shrink the airbag and then perform the inflation. Although the elements in the shrunken bag are very small, the time step can be based on the reference geometry so a very reasonable time step size is obtained. The reference geometry based time step size is optional in the input. The parameters fabric leakage coefficient, FLC, fabric area coefficient, FAC, and effective leakage area, ELA, for the fabric in contact with the structure are optional for the WangNefske and hybrid inflation models. It is possible for the airbag to be constructed of multiple fabrics having different values of porosity and permeability. The gas leakage through the airbag fabric then requires an accurate determination of the areas by part ID available for leakage. The leakage area may change over time due to stretching of the airbag fabric or blockage when the outer surface of the bag is in contact with the structure. MD Nastran SOL 700 can check the interaction of the bag with the structure and split the areas into regions that are blocked and unblocked depending on whether the regions are in contact or not, respectively. Typically, the parameters, FLC and FAC, must be determined experimentally and their variation with time and pressure are optional inputs that allow for maximum modeling flexibility.
Material Model 36: Barlat’s 3Parameter Plasticity Model
This model was developed by Barlat and Lian [1989] for modeling sheets under plane stress conditions. The anisotopic yield criterion Φ for plane stress is defined as:
Φ = a K1 + K2 ∗ + a K 1 – K 2 ∗ + c 2 K
m
= 2σ Y
m
(7174)
Main Index
Chapter 7: Materials 277
Lagrangian Material Models for SOL 700
where
σY
is the yield stress and
K i = 1, 2
are given by:
σ x – hσ y K 1 = 2 K2 =
2 ⎛ σ x – h σ y⎞ + p 2 τ 2 xy ⎝ ⎠ 2
(7175)
The anisotropic material constants a, c, h, and p are obtained through
R 00 1 + R 90 a = 2 – 2  1 + R 00 R 90 c = 2–a h = R 00 1 + R 90  1 + R 00 R 90
R 00, R 45 ,
and
R 90 :
(7176)
The anisotropy parameter p is calculated implicitly. According to Barlat and Lian the thickness strain ratio, for any angle Φ can be calculated from:
2 mσ Y R φ =  – 1 ∂Φ ∂Φ ⎞ ⎛  +  σ ⎝ ∂σ x ∂σ y⎠ φ
m
R
value, width to
(7177)
where σ φ is the uniaxial tension in the Φ direction. This expression can be used to iteratively calculate the value of p . Let Φ = 45 and define a function g as
2 mσ Y g ( p ) =  – 1 – R 45 ∂Φ ∂Φ ⎛  +  ⎞ σ ⎝ ∂ σ x ∂σ y⎠ φ
m
(7178)
An iterative search is used to find the value of p . For facecenteredcubic (FCC) materials m=8 is recommended and for bodycenteredcubic (BCC) materials m=6 may be used. The yield strength of the material can be expressed in terms of k and n :
σ y = kε n = k ( ε yp + ε p )
n
(7179)
where ε yp is the elastic strain to yield and ε p is the effective plastic strain (logarithmic). If σ Y is set to zero, the strain to yield if found by solving for the intersection of the linearly elastic loading equation with the strain hardening equation:
σ = Eε σ = k εn
(7180)
which gives the elastic strain at yield as:
Main Index
278 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
ε yp
E = ⎛  ⎞ ⎝ k⎠
1 n–1
(7181)
If
σY
yield is nonzero and greater than 0.02 then:
1 n
ε yp
σy = ⎛ ⎞ ⎝ k⎠
(7182)
Material Type 37: Transversely Anisotropic ElasticPlastic
This fully iterative plasticity model is available only for shell elements. The input parameters for this model are: Young’s modulus E ; Poisson’s ratio υ ; the yield stress; the tangent modulus E t ; and the anisotropic hardening parameter R . Consider Cartesian reference axes which are parallel to the three symmetry planes of anisotropic behavior. Then the yield function suggested by Hill [1948] can be written
F ( σ 22 – σ 33 ) + G ( σ 33 – σ 11 ) + H ( σ 11 – σ 22 ) + 2L σ 23 + 2 Mσ 31 + 2 Nσ 12 – 1 = 0
2 2 2 2 2 2
(7183) are the shear yield stresses.
where σ y1, σ y2 , and σ y3 are the tensile yield stresses and σ 12, σ 23 , and The constants F, G, H, L, M , and N are related to the yield stress by
1 2L = 2 σ 23 1 2M = 2 σ y31 1 2N = 2 σ y12 1 1 1 2F =  +  – 2 2 2 σ y2 σ y3 σ y1 1 1 1 2G =  +  – 2 2 2 σ y3 σ y1 σ y2 1 1 1 2H =  +  – 2 2 2 σ y1 σ y2 σ y3
σ y31
.
(7184)
The isotropic case of von Mises plasticity can be recovered by setting
1 F = G = H = 2 2σ y
and
1L = M = N =  . 2 2σ y
Main Index
Chapter 7: Materials 279
Lagrangian Material Models for SOL 700
For the particular case of transverse anisotropy, where properties do not vary in the following relations hold:
1 2 F = 2 G = 2 σ y3 2 1 2H =  – 2 2 σ y σ y3 1 2 1  N =  –  – 2 2 σ y 2 σ y3
x1 – x2
plane, the
(7185)
where it has been assumed that Letting , F(σ) where
σy K =  , σ y3
σ y 1 = σy 2 = σ y .
the yield criterion can be written (7186)
= σ e = σy
1 2 2 2 2 2 2 2 2 2 2 F ( σ ) ≡ σ 11 + σ 22 + K σ 33 – K σ 33 ( σ 11 + σ 22 ) – ( 2 – K ) σ 11 σ 22 + 2 Lσ y ( σ 23 + σ 31 ) + 2 ⎛ 2 –  K ⎞ σ 12 ⎝ 2 ⎠
1⁄2
The rate of plastic strain is assumed to be normal to the yield surface so
∂F ·p ε y = λ  . ∂σ i j
·p εy
is found from (7187)
R
Now consider the case of plane stress, where σ 33 = 0 . Also, define the anisotropy input parameter the ratio of the inplane plastic strain rate to the outofplane plastic strain rate:
·p ε 22 R =  . ·p ε 33
as
(7188)
It then follows that
2 R =  – 1 . K2
(7189)
R,
Using the plane stress assumption and the definition of
F (σ ) = 2R + 1 2 2R 2 2 σ 11 + σ 22 –  σ 11 σ 22 + 2  σ 12 R+1 R+1
1⁄2
the yield function may now be written (7190)
Main Index
280 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
Material Type 38: BlatzKo Compressible Foam
μ I2  W ( I 1, I 2, I 3 ) =  ⎛  + 2 I 3 – 5⎞ ⎠ 2 ⎝ I3
(7191)
where μ is the shear modulus and I 1, I 2 , and I 3 are the strain invariants. Blatz and Ko [1962] suggested this form for a 47 percent volume polyurethane foam rubber with a Poisson’s ratio of 0.25. The second PiolaKirchhoff stresses are given as
I2 1 S i j = μ ( Iδ i j – G ij )  + ⎛ I 3 –  ⎞ G ij I3 ⎝ I 3⎠ ∂x k ∂x k  G i j =  ∂X i ∂X j ∂X i ∂X j  G i j =  ∂x k ∂x k
(7192)
where
after determining
ρ ∂x i ∂x  σ i j =   j S kl ρ o ∂X k ∂X l
Si j ,
it is transformed into the Cauchy stress tensor:
where
ρo
and
ρ
are the initial and current density, respectively.
Material Model 39: Transversely Anisotropic ElasticPlastic With FLD
See Material Model 37 for the similar model theoretical basis. The first history variable is the maximum strain ratio defined by:
ε major workpiece ε major
fld
(7193)
workpiece corresponding to . This history variable replaces the effective plastic strain in the output. Plastic strains can still be obtained but one additional history variable must be written into the D3PLOT database.
ε min or
The strains on which these calculations are based are integrated in time from the strain rates:
ε inj + 1 = ε inj + ε
1 1 ∇ n + n + 2 2Δt
(7194)
and are stored as history variables. The resulting strain measure is logarithmic.
Main Index
Chapter 7: Materials 281
Lagrangian Material Models for SOL 700
ε mnr = 0
Plane Strain
ε m jr
80 70 60 % MAJOR STRAIN 50 40 30 20
ε mjr ε mnr ε mj r ε mnr
10
Draw
Stretch
50
40
30
20
10
0
+10
+20
+30
+40
+50
% MINOR STRAIN
Figure 77
Flow Limit Diagram
Material Model 53: Low Density Closed Cell Polyurethane Foam
A rigid, low density, closed cell, polyurethane foam model developed at Sandia Laboratories [Neilsen et. al., 1987] has been recently implemented for modeling impact limiters in automotive applications. A number of such foams were tested at Sandia and reasonable fits to the experimental data were obtained. In some respects this model is similar to the crushable honeycomb model type 26 in that the components of the stress tensor are uncoupled until full volumetric compaction is achieved. However, unlike the honeycomb model this material possesses no directionality but includes the effects of confined air pressure in its overall response characteristics.
σ i j = σ i j – δ ij σ
sk ai r
(7195)
σ a ir
where
σi j
sk
is the skeletal stress and
is the air pressure computed from the equation: (7196)
p0 γ σ air = – 1+γ–φ
Main Index
282 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
where p 0 is the initial foam pressure usually taken as the atmospheric pressure and volumetric strain
γ = V – 1 + γ0
γ
defines the
(7197)
where V is the relative volume and γ 0 is the initial volumetric strain which is typically zero. The yield condition is applied to the principal skeletal stresses which are updated independently of the air pressure. We first obtain the skeletal stresses:
σ i j = σ i j + σ i j σ air
sk
(7198)
σi j
skt
and compute the trial stress,
σi j
skt sk · = σ ij + E ε ij Δ t
(7199)
where E is Young’s modulus. Since Poisson’s ratio is zero, the update of each stress component is uncoupled and 2G = E where G is the shear modulus. The yield condition is applied to the principal skeletal stresses such that if the magnitude of a principal trial stress component, stress, σ y , then . σ isk
sk σ i = min ( σ y, σ i ) skt σi skt
σi
skt
, exceeds the yield
(7200)
The yield stress is defined by
σy = a + b ( 1 + c γ )
(7201)
where a, b , and c are user defined input constants. After scaling the principal stresses they are transformed back into the global system and the final stress state is computed
σ i j = σ isk – δ ij σ ai r . j
Material Models 54 and 55: Enhanced Composite Damage Model
These models are very close in their formulations. Material 54 uses the Chang matrix failure criterion (as Material 22), and Material 55 uses the TsaiWu criterion for matrix failure. Arbitrary orthothropic materials, e.g., unidirectional layers in composite shell structures can be defined. Optionally, various types of failure can be specified following either the suggestions of [Chang and Chang, 1984] or [Tsai and Wu, 1981]. In addition special measures are taken for failure under compression. See [Matzenmiller and Schweizerhof, 1990]. This model is only valid for thin shell elements. The Chang/Chang criteria is given as follows:
Main Index
Chapter 7: Materials 283
Lagrangian Material Models for SOL 700
for the tensile fiber mode,
σ aa > 0
then
σ aa 2 σ ab e f2 = ⎛ ⎞ + β ⎛ ⎞ – 1 ⎝ Xt ⎠ ⎝ Sc ⎠
> 0 failed , < 0 elastic
(7202)
E a = E b = G ab = ν ba = ν ab = 0
for the compressive fiber mode,
σ aa > 0
then
σ aa 2 e c = ⎛ ⎞ – 1 ⎝ Xc ⎠
2
> 0 failed , < 0 elastic
(7203)
E a = ν ba = ν ab = 0
for the tensile matrix mode,
σ bb > 0
then
σ bb 2 σ ab 2 2 e m = ⎛ ⎞ + ⎛ ⎞ – 1 ⎝ Yt ⎠ ⎝ Sc ⎠
> 0 failed , < 0 elastic
(7204)
E b = ν ba = 0. → G ab = 0
and for the compressive matrix mode,
σ bb < 0
then
σ bb σ bb 2 Yc 2 σ ab 2 2 e d = ⎛ ⎞ + ⎛ ⎞ – 1  + ⎛ ⎞ – 1 ⎝ 2 S c⎠ ⎝ 2S t⎠ ⎝ Sc ⎠ Yc
> 0 failed < 0 elastic
(7205)
E b = ν ba = ν ab = 0. → G ab = 0 X c = 2Y c
for 50% fiber volume
In the Tsai/Wu criteria the tensile and compressive fiber modes are treated as in the Chang/Chang criteria. The failure criterion for the tensile and compressive matrix mode is given as:
2 σ bb σ ab 2 ( Y c – Y t )σ bb 2 e md =  + ⎛ ⎞ +  – 1 ⎝ Sc ⎠ Yc Y t Y c Yt
> 0 failed < 0 elastic
(7206)
For For
β = 1, β = 0,
we get the original criterion of Hashin [1980] in the tensile fiber mode. we get the maximum stress criterion which is found to compare better to experiments.
Failure can occur in any of four different ways: 1. If DFAILT is zero, failure occurs if the Chang/Chang failure criterion is satisfied in the tensile fiber mode. 2. If DFAILT is greater than zero, failure occurs if the tensile fiber strain is greater than DFAILT or less than DFAILC. 3. If EFS is greater than zero, failure occurs if the effective strain is greater than EFS. 4. If TFAIL is greater than zero, failure occurs according to the element time step as described in the definition of TFAIL above.
Main Index
284 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
When failure has occurred in all the composite layers (throughthickness integration points), the element is deleted. Elements which share nodes with the deleted element become “crashfront” elements and can have their strengths reduced by using the SOFT parameter with TFAIL greater than zero. Information about the status in each layer (integration point) and element can be plotted using additional integration point variables. The number of additional integration point variables for shells written to the database is input by the PARAM, DYNEIPS definition as variable NEIPS. For Models 54 and 55 these additional variables are tabulated below (i = shell integration point): History Variable 1. ef(i) 2. ec(i) 3. em(i) 4. ed(i) 5. efail 6. dam d3plot Component 81 1  elastic 0  failed 82 83 84 85 1  element intact 108  element in crashfront +1  element failed 86
Description tensile fiber mode compressive fiber mode tensile matrix mode compressive matrix mode max[ef(ip)] damage parameter
Value
These variables can be plotted in MD Patran as element components 81, 82, ..., 80+ NEIPS. The following components, defined by the sum of failure indicators over all throughthickness integration points, are stored as element component 7 instead of the effective plastic strain: Description
ni p
Integration Point 1
1 ni p
∑ ef(i)
i =1 ni p
1 ni p
∑ e c (i )
i =1 ni p
2
1 ni p
∑ cm(i)
i =1
3
Material Model 57: Low Density Urethane Foam
The urethane foam model is available to model highly compressible foams such as those used in seat cushions and as padding on the Side Impact Dummy (SID). The compressive behavior is illustrated in Figure 78 where hysteresis on unloading is shown. This behavior under uniaxial loading is assumed not to significantly couple in the transverse directions. In tension the material behaves in a linear fashion until tearing occurs. Although our implementation may be somewhat unusual, it was first motivated by
Main Index
Chapter 7: Materials 285
Lagrangian Material Models for SOL 700
Shkolnikov [1991] and a paper by Storakers [1986]. The recent additions necessary to model hysteretic unloading and rate effects are due to Chang, et. al., [1994]. These latter additions have greatly expanded the usefulness of this model.
σ
Typical unloading curves determined by the hysteric unloading factor. With the shape factor equal to unity.
σ
Typical unloading for a large shape factor; e.g., 5.8 and a small hysteric factor; e.g., .010.
Unloading Curves Strain Strain
Figure 78
Behavior of the Lowdensity Urethane Foam Model
The model uses tabulated input data for the loading curve where the nominal stresses are defined as a function of the elongations, ε i , which are defined in terms of the principal stretches, λ i , as:
εi = λi – 1
(7207)
The stretch ratios are found by solving for the eigenvalues of the left stretch tensor, V ij , which is obtained via a polar decomposition of the deformation gradient matrix, F ij :
F ij = R i k U kj = V i k R kj
(7208)
The update of V ij follows the numerically stable approach of Taylor and Flanagan [1989]. After solving for the principal stretches, the elongations are computed and, if the elongations are compressive, the corresponding values of the nominal stresses, τ i , are interpolated. If the elongations are tensile, the nominal stresses are given by
τi = E εi
(7209)
The Cauchy stresses in the principal system become
τi σ i = λi λ k
(7210)
The stresses are then transformed back into the global system for the nodal force calculations.
Main Index
286 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
When hysteretic unloading is used, the reloading will follow the unloading curve if the decay constant, β , is set to zero. If β is nonzero the decay to the original loading curve is governed by the expression:
1 – e – βt
(7211)
The bulk viscosity, which generates a rate dependent pressure, may cause an unexpected volumetric response and, consequently, it is optional with this model. Rate effects are accounted for through linear viscoelasticity by a convolution integral of the form
σi j = ∫0 g ij kl ( t – τ ) ∂τ
t
∂ε kl
(7212)
where
g i jkl ( t – τ ) σ ifj ;
is the relaxation function. The stress tensor, , σ irj augments the stresses determined from
σi j ,
the foam,
consequently, the final stress,
is taken as the summation of the two contributions: (7213)
σ i j = σ ifj + σ irj
Since we wish to include only simple rate effects, the relaxation function is represented by one term from the Prony series:
N
g ( t ) = α0 +
∑
m=1
α m e – βt
(7214)
given by,
g ( t ) = Ed e
–β1 t
(7215)
This model is effectively a Maxwell fluid which consists of a damper and spring in series. We characterize this in the input by a Young's modulus, E d , and decay constant, β 1 . The formulation is performed in the local system of principal stretches where only the principal values of stress are computed and triaxial coupling is avoided. Consequently, the onedimensional nature of this foam material is unaffected by this addition of rate effects. The addition of rate effects necessitates twelve additional history variables per integration point. The cost and memory overhead of this model comes primarily from the need to “remember” the local system of principal stretches.
Material Type 58: Laminated Composite Fabric
Parameters to control failure of an element layer are: ERODS, the maximum effective strain; i.e., maximum 1 = 100 % straining. The layer in the element is completely removed after the maximum effective strain (compression/tension including shear) is reached. The stress limits are factors used to limit the stress in the softening part to a given value,
σ min = SL IM xx ⋅ s tre ng t h
,
(7216)
Main Index
Chapter 7: Materials 287
Lagrangian Material Models for SOL 700
thus, the damage value is slightly modified such that elastoplastic like behavior is achieved with the threshold stress. As a factor for S LIM xx a number between 0.0 and 1.0 is possible. With a factor of 1.0, the stress remains at a maximum value identical to the strength, which is similar to ideal elastoplastic behavior. For tensile failure a small value for S LIM Tx is often reasonable; however, for compression S LIM Cx = 1.0 is preferred. This is also valid for the corresponding shear value. If S LIM xx is smaller than 1.0, then localization can be observed depending on the total behavior of the layup. If the user is intentionally using S LIM xx < 1.0 , it is generally recommended to avoid a drop to zero and set the value to something in between 0.05 and 0.10. Then elastoplastic behavior is achieved in the limit which often leads to less numerical problems. Defaults for S LIM xx = 1.0 E8 . The crashfrontalgorithm is started if and only if a value for TSIZE (time step size, with element elimination after the actual time step becomes smaller than TSIZE) is input The damage parameters can be written to the postprocessing database for each integration point as the first three additional element variables and can be visualized. Material models with FS=1 or FS=1 are favorable for complete laminates and fabrics, as all directions are treated in a similar fashion. For material model FS=1, an interaction between normal stresses and shear stresses is assumed for the evolution of damage in the a and bdirections. For the shear damage is always the maximum value of the damage from the criterion in a or b direction is taken. For material model FS=1, it is assumed that the damage evolution is independent of any of the other stresses. A coupling is present only via the elastic material parameters and the complete structure. In tensile and compression directions and in a as well as in b direction, different failure surfaces can be assumed. The damage values, however, increase only when the loading direction changes. Special Control of Shear Behavior of Fabrics For fabric materials a nonlinear stress strain curve for the shear part of failure surface FS=1 can be assumed as given below. This is not possible for other values of FS. The curve, shown in Figure 19.58.1, is defined by three points: a. a) the origin (0,0) is assumed, b. the limit of the first slightly nonlinear part (must be input), stress (TAU1) and strain (GAMMA1), see below. c. the shear strength at failure and shear strain at failure. In addition, a stress limiter can be used to keep the stress constant via the SLIMS parameter. This value must be less than or equal to 1.0 and positive, which leads to an elastoplastic behavior for the shear part. The default is 1.0E08, assuming almost brittle failure once the strength limit SC is reached.
Main Index
288 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
τ
SC TAU1 SLIMS*SC
GAMMA1
GMS
γ
Figure 79
Stressstrain Diagram for Shear
Material Type 62: Viscous Foam
This model was written to represent the energy absorbing foam found on certain crash dummies, i.e., the ‘Confor Foam’ covering the ribs of the Eurosid dummy. The model consists of a nonlinear elastic stiffness in parallel with a viscous damper. A schematic is shown in Figure 710. The elastic stiffness is intended to limit total crush while the viscous damper absorbs energy. The stiffness E 2 prevents timestep problems.
E1
V2
E2
Figure 710 Both
E1
Schematic of Material Model 62
V2 )
n2
and
–n1
are nonlinear with crush as follows:
t E1 = E1 ( V
(7217)
t V 2 = V 2 ( ab s ( 1 – V ) )
where V is the relative volume defined by the ratio of the current to initial volume. Typical values are (units of N ,m m, s )
Main Index
Chapter 7: Materials 289
Lagrangian Material Models for SOL 700
E 1 = 0.0036 n 1 = 4.0 V 2 = 0.0015 E 2 = 100.0 n 2 = 0.2 ν = 0.05
Material Type 63: Crushable Foam
The intent of this model is model crushable foams in side impact and other applications where cyclic behavior is unimportant. This isotropic foam model crushes onedimensionally with a Poisson’s ratio that is essentially zero. The stress versus strain behavior is depicted in Figure 711 where an example of unloading from point a to the tension cutoff stress at b then unloading to point c and finally reloading to point d is shown. At point the reloading will continue along the loading curve. It is important to use nonzero values for the tension cutoff to prevent the disintegration of the material under small tensile loads. For high values of tension cutoff the behavior of the material will be similar in tension and compression. In the implementation we assume that Young’s modulus is constant and update the stress assuming elastic behavior.
·n σ itjrial = σ inj + E ε ij + 1 ⁄ 2 Δ t n + 1 ⁄ 2
(7218)
σy ,
The magnitudes of the principal values, σ it ria l , i = 1 ,3 , are then checked to see if the yield stress, exceeded and if so they are scaled back to the yield surface: if
σ y < σ it ria l
is
then
σ it ria l σ in + 1 = σ y σ it ria l
(7219)
After the principal values are scaled, the stress tensor is transformed back into the global system. As seen in Figure 711, the yield stress is a function of the natural logarithm of the relative volume, V ; i.e., the volumetric strain.
Main Index
290 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
σ ij
a
d
E
c
b Volumetric Strain – ln V
Figure 711
Yield Stress Versus Volumetric Strain Curve for the Crushable Foam
Material Model 64: Strain Rate Sensitive PowerLaw Plasticity
This material model follows a constitutive relationship of the form:
· σ = kε m ε n
(7220)
· where σ is the yield stress, ε is the effective plastic strain, ε is the effective plastic strain rate, and the constants, k , m , and n can be expressed as functions of effective plastic strain or can be constant with respect to the plastic strain. The case of no strain hardening can be obtained by setting the exponent of the plastic strain equal to a very small positive value; i.e., 0.0001.
This model can be combined with the superplastic forming input to control the magnitude of the pressure in the pressure boundary conditions in order to limit the effective plastic strain rate so that it does not exceed a maximum value at any integration point within the model. A fully viscoplastic formulation is optional. An additional cost is incurred but the improvement is results can be dramatic.
Material Model 65: Modified Zerilli/Armstrong
The ArmstrongZerilli Material Model expresses the flow stress as follows. For fcc metals,
· ⎧ ⎫ μ(T) ( – C 3 + C 4 ln ( ε * ) ) τ σ = C1 + ⎨ C 2 ( ε p ) 1 ⁄ 2 e + C 3 ⎬ ⎛ ⎞ ⎝ ⎠ ⎩ ⎭ μ ( 293 )
(7221)
εp =
effective plastic strain
Main Index
Chapter 7: Materials 291
Lagrangian Material Models for SOL 700
· ε · ε * = ·ε0
effective plastic strain rate where
· ε0 = 1 ,
1e3, le6 for time units
of seconds, milliseconds, and microseconds, respectively. For bcc metals,
σ = C1 + C2 e
· ( – C 3 + c 4 lm ( ε * ) )τ
μ(T) + [ C 5 ( ε p ) n + C 6 ] ⎛ ⎞ ⎝ μ ( 293 )⎠
(7222)
where
μ(T)  = B 1 + B 2 T + B 3 T 2 μ ( 293 )
(7223)
The relationship between heat capacity (specific heat) and temperature may be characterized by a cubic polynomial equation as follows:
Cp = G1 + G2 T + G 3 T 2 + G4 T 3
(7224)
A fully viscoplastic formulation is optional. An additional cost is incurred but the improvement in results can be dramatic.
Material Model 67: Nonlinear Stiffness/Viscous 3D Discrete Beam
The formulation of the discrete beam (Type 6) assumes that the beam is of zero ngth and requires no orientation node. A small distance between the nodes joined by the beam is permitted. The local coordinate system which determines (r, s, t) is given by the coordinate ID in the crosssectional input where the global system is the default. The local coordinate system axes rotate with the average of the rotations of the two nodes that define the beam. For null TABLED1 IDs, no forces are computed. The force resultants are found from TABLED1s (see Figure 712) that are defined in terms of the force resultant versus the relative displacement in the local coordinate system for the discrete beam. The resultant forces and moments are determined by a table lookup, if the origin of the TABLED1 is at [0,0], as shown in Figure 712b, and the tension and compression responses are symmetric.
Main Index
292 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
R E S U L T A N T DISPLACEMENT
R E S U L T A N T
(a)
DISPLACEMENT (b)
Figure 712
Resultant Forces and Moments Determined by Table Lookup
Material Model 68: Nonlinear Plastic/Linear Viscous 3D Discrete Beam
The folmulation of the discrete beam (Type 6) assumes that the beam is of zero length and requires no orientation node. A small distance between the nodes joined by the beam is permitted. The local coordinate system, which determines (r, s, t) is given by the coordinate ID in the crosssectional input where the global system is the default. The local coordinate system axes rotate with the average of the rotations of the two nodes that define the beam. Each force resultant in the local system can have a limiting value defined as a function of plastic displacement by using a TABLED1 (see Figure 713). For the degrees of freedom where elastic behavior is desired, the TABLED1 ID is simply set to zero.
R E S U L T A N T
PLASTIC DISPLACEMENT
Figure 713
Resultant Forces and Moments Limited by the Yield Definition
Main Index
Chapter 7: Materials 293
Lagrangian Material Models for SOL 700
Catastrophic failure, based on force resultants, occurs if the following inequality is satisfied:
⎛ F r ⎞ 2 ⎛ F s ⎞ 2 ⎛ Ft ⎞ 2 ⎛ M r ⎞ 2 ⎛ Ms ⎞ 2 ⎛ Mt ⎞ 2 ⎜  ⎟ + ⎜ l⎟ + ⎜  ⎟ + ⎜ ⎟ + ⎜ l⎟ + ⎜ l⎟ – 1 ≥ 0 fail f fail f ⎝ Fr ⎠ ⎝ F s ai ⎠ ⎝ F tfail⎠ ⎝ Mr ⎠ ⎝ M s ai ⎠ ⎝ M tf ai ⎠
(7225)
Likewise, catastrohic failure based on displacement resultants occurs if:
⎛ ur ⎞ 2 ⎛ us ⎞ 2 ⎛ ut ⎞ 2 ⎛ θ r ⎞ 2 ⎛ θs ⎞ 2 ⎛ θ t ⎞ 2 ⎜ ⎟ + ⎜ ⎟ + ⎜ ⎟ + ⎜ ⎟ + ⎜ l⎟ + ⎜ ⎟ – 1 ≥ 0 f f fail f ⎝ u r ai l⎠ ⎝ u s ai l⎠ ⎝ u tf ai l⎠ ⎝ θr ⎠ ⎝ θ s ai ⎠ ⎝ θ tfa il⎠
(7226)
After failure, the discrete element is deleted. If failure is included, either one or both of the criteria may be used.
Material Model 69: Side Impact Dummy Damper (SID Damper)
The side impact dummy uses a damper that is not adequately treated by nonlinear force versus relative velocity curves, since the force characteristics are also dependent on the displacement of the piston. As the damper moves, the fluid flows through the open orifices to provide the necessary damping resistance. While moving as shown in Figure 714, the piston gradually blocks off and effectively closes the orifices. The number of orifices and the size of their openings control the damper resistance and performance. The damping force is computed from the equation.
⎛ Ap ⎞ 2 ⎧ ⎧ Cl ⎫ F = S F ⎨ K A p V p ⎨  + C 2 V p ρ fl ui d ⎜ ⎟ – 1 ⎬ – f ( s + s 0 ) + V p g ( s + s 0 ) } t t ⎝ C A0 ⎠ ⎩ ⎩ A0 ⎭
(7227)
where
Vp
K
is a user defined constant or a tabulated function of the absolute value of the relative velocity,
t is the piston's relative velocity, C is the discharge coefficient, A p is the piston area, A 0 is the total open areas of orifices at time t , ρ fl ui d is the fluid density, C 1 is the coefficient for the linear term, and C 2 is the coefficient for the quadratic term.
In the implementation, the orifices are assumed to be circular with partial covering by the orifice controller. As the piston closes, the closure of the orifice is gradual. This gradual closure is taken into account to insure a smooth response. If the piston stroke is exceeded, the stiffness value, k , limits further movement; i.e., if the damper bottoms out in tension or compression, the damper forces are calculated by replacing the damper by a bottoming out spring and damper, k and c , respectively. The piston stroke must exceed the initial length of the beam element. The time step calculation is based in part on the stiffness value of the bottoming out spring. A typical force versus displacement curve at constant relative velocity with only the linear velocity term active is shown in Figure 715. The factor, S F , which scales the force defaults to 1.0 and is analogous to the adjusting ring on the damper.
Main Index
294 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
Figure 714
Mathematical Model for Side Impact Dummy Damper
Linear loading after ofifices close
F O R C E
Last orfice closes
Force increases as orfice is gradually covered Displacement
Figure 715
Force Versus Displacement
Main Index
Chapter 7: Materials 295
Lagrangian Material Models for SOL 700
Material Model 70: Hydraulic/Gas Damper
This special purpose element represents a combined hydraulic and gasfilled damper which has a variable orifice coefficient. A schematic of the damper is shown in Figure 716. Dampers of this type are sometimes used on buffers at the end of railroad tracks and as aircraft undercarriage shock absorbers. This material can be used only as a discrete beam element.
Figure 716
Schematic of Hydraulic/Gas Damper
As the damper is compressed two actions contribute to the force that develops. First, the gas is adiabatically compressed into a smaller volume. Secondly, oil is forced through an orifice. A profiled pin may occupy some of the crosssectional area of the orifice; thus, the orifice area available for the oil varies with the stroke. The force is assumed proportional to the square of the velocity and inversely proportional to the available area. The equation for this element is:
C0 n ⎧ ⎫ V 2 F = S CLF ⋅ ⎨ K h ⎛  ⎞ + P 0 ⎛ ⎞ – P a ⋅ A p ⎬ ⎝ C 0 – S⎠ ⎝ a 0⎠ ⎩ ⎭
(7228)
where
S
is the element deflection and
V
is the relative velocity across the element.
Material Model 71: Cable
This material can be used only as a discrete beam element. The force, nonzero only if the cable is in tension. The force is given by:
F = K ⋅ m ax ( Δ L, 0. ) F,
generated by the cable is (7229)
where
ΔL
is the change in length (7230)
Δ L = c u rre nt le ng t h – ( i ni ti a l l en gt h – of fs et )
and the stiffness is defined as:
E ⋅ a rea K = ( i n it i al le ng th – o ffse t )
(7231)
Main Index
296 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
The area and offset are defined on either the cross section or element cards in the SOL 700 input. For a slack cable the offset should be input as a negative length. For an initial tensile force the offset should be positive. If a TABLED1 is specified, the Young’s modulus will be ignored and the TABLED1 will be used instead. The points on the TABLED1 are defined as engineering stress versus engineering strain; i.e., the change in length over the initial length. The unloading behavior follows the loading.
Material Model 73: Low Density Viscoelastic Foam
This viscoelastic foam model is available to model highly compressible viscous foams. The hyperelastic formulation of this model follows that of material 57. Rate effects are accounted for through linear viscoelasticity by a convolution integral of the form
σ irj = ∫0 g ij kl ( t – τ )  dτ ∂τ
t
∂ε kl
(7232)
σ irj , augments the stresses determined
where
g i jkl ( t – τ ) σ ifj
is the relaxation function. The stress tensor,
σi j ,
from
the foam,
; consequently, the final stress,
is taken as the summation of the two contributions: (7233)
σ i j = σ ifj + σ irj
Since we wish to include only simple rate effects, the relaxation function is represented by up to six terms of the Prony series:
N
g ( t ) = α0 +
∑
m=1
α m e – βt
(7234)
This model is effectively a Maxwell fluid which consists of a dampers and springs in series. The formulation is performed in the local system of principal stretches where only the principal values of stress are computed and triaxial coupling is avoided. Consequently, the onedimensional nature of this foam material is unaffected by this addition of rate effects. The addition of rate effects necessitates 42 additional history variables per integration point. The cost and memory overhead of this model comes primarily from the need to “remember” the local system of principal stretches and the evaluation of the viscous stress components
Material Model 74: Elastic Spring for the Discrete Beam
This model permits elastic springs with damping to be combined and represented with a discrete beam element type 6. Linear stiffness and damping coefficients can be defined and, for nonlinear behavior, a force versus deflection and force versus rate curves can be used. Displacement based failure and an initial force are optional If the linear spring stiffness is used, the force,
F = F 0 + F Δ L + DΔ L F,
is given by: (7235)
Main Index
Chapter 7: Materials 297
Lagrangian Material Models for SOL 700
where
K
is the stiffness constant, and
f(ΔL) .
D
is the viscous damping coefficient.
If the TABLED1 ID for given by:
is specified, nonlinear behavior is activated. For this case the force is
· ⎛ ⎧ Δ L ⎫⎞ · · F = F 0 + K f ( Δ L ) 1 + C 1 ⋅ Δ L + C 2 ⋅ sgn ( Δ L ) ln ⎜ ma x ⎨ 1. , ⎬⎟ ⎝ ⎩ D LE ⎭⎠ · · + DΔ L + g ( Δ L ) h ( Δ L )
(7236)
where C 1 and C2 are damping coefficients for nonlinear behavior, DL E is a factor to scale time units, and g ( Δ L ) . is an optional TABLED1 defining a scale factor versus deflection for TABLED1 ID, h ( Δ L ⁄ d t ) . In these equations,
ΔL
is the change in length
Δ L = currentlength – initiallength .
Failure can occur in either compression or tension based on displacement values of CDF and TDF, respectively. After failure no forces are carried. Compressive failure does not apply if the spring is initially zero length. The cross sectional area is defined on the beam property card for the discrete beam elements. The square root of this area is used as the contact thickness offset if these elements are included in the contact treatment.
Material Model 76: General Viscoelastic
Rate effects are taken into account through linear viscoelasticity by a convolution integral of the form:
σi j = ∫0 gij kl ( t – τ )  dτ ∂τ
t
∂ε kl
(7237)
where
g i jkl ( t – τ )
is the relaxation function.
If we wish to include only simple rate effects for the deviatoric stresses, the relaxation function is represented by six terms from the Prony series:
N
g(t) =
∑
m=1
Gm e
–βm t
(7238)
βi .
We characterize this in the input by shear modulii, G i , and decay constants, terms, up to 6, may be used when applying the viscoelastic model.
An arbitrary number of
Main Index
298 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
Figure 717 Note:
Relaxation Curve This curve defines stress versus time where time is defined on a logarithmic scale. For best results, the points defined in the TABLED1 should be equally spaced on the logarithmic scale. Furthermore, the TABLED1 should be smooth and defined in the positive quadrant. If nonphysical values are determined by least squares fit, SOL 700 will terminate with an error message after the initialization phase is completed. If the ramp time for loading is included, then the relaxation which occurs during the loading phase is taken into account. This effect may or may not be important.
For volumetric relaxation, the relaxation function is also represented by the Prony series in terms of bulk modulii:
N
k(t) =
∑
m= 1
Km e
–βk t m
(7239)
Material Model 77: Hyperviscoelastic Rubber
Rubber is generally considered to be fully incompressible since the bulk modulus greatly exceeds the shear modulus in magnitude. To model the rubber as an unconstrained material a hydrostatic work term, W H ( J ) , is included in the strain energy functional which is function of the relative volume, ( J ) , [Ogden, 1984]:
Main Index
Chapter 7: Materials 299
Lagrangian Material Models for SOL 700
n
W ( J 1 ,J 2 ,J ) = J1 =
– I1 I3 1 ⁄ 2
∑
p ,q = 0
C pq ( J 1 – 3 ) p ( J 2 – 3 ) q + W H ( J )
(7240)
– J2 = I2 I3 1 ⁄ 2
In order to prevent volumetric work from contributing to the hydrostatic work the first and second invariants are modified as shown. This procedure is described in more detail by Sussman and Bathe [1987]. For the Ogden model the energy equation is given as:
3 n
W∗ =
i = 1j = 1
∑ ∑ j ( λ i α
μj
*α 1
1 – 1 ) +  K ( J – 1 ) 2 2
(7241)
where the asterisk ( * ) indicates that the volumetric effects have be eliminated from the principal stretches, λ j* . See Ogden [1984] for more details. Rate effects are taken into account through linear viscoelasticity by a convolution integral of the form:
σi j = ∫0 gij kl ( t – τ )  dτ ∂τ Si j ,
t
∂ε kl
(7242) and Green's strain tensor,
E ij ,
or in terms of the second PiolaKirchhoff stress,
∂E kl t S i j ∫ G i j kl ( t – τ )  dτ 0 ∂τ
(7243)
where g i jkl ( t – τ ) and G i jkl ( t – τ ) are the relaxation functions for the different stress measures. This stress is added to the stress tensor determined from the strain energy functional. If we wish to include only simple rate effects, the relaxation function is represented by six terms from the Prony series:
N
g ( t ) = α0 +
∑
m= 1
α m e – βt
(7244)
given by,
n
g(t) =
∑ Gi e
j =1
–βi t
(7245)
This model is effectively a Maxwell fluid which consists of a dampers and springs in series. We characterize this in the input by shear moduli, G i , and decay constants, β i . The viscoelastic behavior is optional and an arbitrary number of terms may be used. The MooneyRivlin rubber model is obtained by specifying n + 2 . In spite of the differences in formulations with Model 27, we find that the results obtained with this model are nearly identical with those of Model 27 as long as large values of Poisson’s ratio are used.
Main Index
300 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
When viscoelastic terms are not included, this model is similar to the use of the Ogden model in solution 600, defined on the MATHE option.
Material Model 78: Soil/Concrete
ConcretePressure is positive in compression. Volumetric strain is defined as the natural log of the relative volume and is positive in compression where the relative volume, V , is the ratio of the current volume to the initial volume. The tabulated data should be given in order of increasing compression. If the pressure drops below the cutoff value specified, it is reset to that value and the deviatoric stress state is eliminated. If the TABLED1 ID is provided as a positive number, the deviatoric perfectly plastic pressure dependent yield function φ , is described in terms of the second invariant, J 2 , the pressure, p , and the table, F ( p ) , as
φ 3J 2 – F ( p ) = σ y – F ( p )
where
J2
is defined in terms of the deviatoric stress tensor as:
1 J 2 =  S i j S i j 2
assuming that. If the ID is given as negative, the yield function becomes:
φ = J2 – F p
being the deviatoric stress tensor. If cracking is invoked, the yield stress is multiplied by a factor according to a trilinear law as shown in Figure 718.
f
f which reduces with plastic stain
1.0
b
ε1
ε2
εp
Figure 718
Strength Reduction Factor
Main Index
Chapter 7: Materials 301
Lagrangian Material Models for SOL 700
b ε1 ε2 ε1
= residual strength factor = plastic stain at which cracking begins. = plastic stain at which residual strength is reached.
and ε 2 are tabulated functions of pressure that are defined by TABLED1s (see Figure 719). The values on the curves are pressure versus strain and should be entered in order of increasing pressure. The strain values should always increase monotonically with pressure. By properly defining the TABLED1s, it is possible to obtain the desired strength and ductility over a range of pressure. (see Figure 720).
ε ε2
ε2
P Figure 719 Cracking Strain Versus Pressure
Yield Stress
p3 p2 p1
Plastic Strain
Figure 720
Strength and Ductility Over a Range of Pressures
Main Index
302 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
Material Model 79: Hysteretic Soil
This model is a nested surface model with five superposed “layers” of elastoperfectly plastic material, each with its own elastic modulii and yield values. Nested surface models give hysteretic behavior, as the different “layers” yield at different stresses. The constants ( a 0 , a 1 , a 2 ) govern the pressure sensitivity of the yield stress. Only the ratios between these values are important  the absolute stress values are taken from the stressstrain curve. The stress strain pairs ( ( γ 1 ,τ1 ) , … ( γ 5 ,τ5 ) ) define a shear stress versus shear strain curve. The first point on the curve is assumed by default to be (0,0) and does not need to be entered. The slope of the curve must decrease with increasing γ . Not all five points need be to be defined. This curve applies at the reference pressure; at other pressures, the curve variesaccording to a 0 , a 1 , and a 2 as in the soil and crushable foam model, Material 5. The elastic moduli G and K are pressure sensitive.
G = G0 ( p – p0 )b K = K0 ( p – p0 ) b
where G 0 and K 0 are the input values, p is the current pressure, p o the cutoff or referencepressure (must be zero or negative). If p attempts to fall below p o (i.e., more tensile) the shear stresses are set to zero and the pressure is set to p o . Thus, the material has no stiffness orstrength in tension. The pressure in compression is calculated as follows:
p = [ – K 0 ln ( V ) ] 1 – b
1 
where
V
is the relative volume; i.e., the ratio between the original and current volume.
Material Model 80: RambergOsgood Plasticity
The RambergOsgood equation is an empirical constitutive relation to represent the onedimensional elasticplastic behavior of many materials, including soils. This model allows a simple rate independent representation of the hysteretic energy dissipation observed in soils subjected to cyclic shear deformation. For monotonic loading, the stressstrain relationship is given by:
τ γτ  =  + α τy γy τy τ γτ  =  – α τy γy τy
r
if γ ≥ 0
r
if γ < 0
(7246)
Main Index
Chapter 7: Materials 303
Lagrangian Material Models for SOL 700
where γ is the shear strain and τ is the shear stress. The model approaches perfect plasticity as the stress exponent r → ∞ . These equations must be augmented to correctly model unloading and reloading · material behavior. The first load reversal is detected by γ γ < 0 . After the first reversal, the stressstrain relationship is modified to
( γ – γ0) ( τ – τ0 ) ( τ – τ0 )  =  + α 2 γy 2 τy 2τ y ( γ – γ0) ( τ – τ0 ) ( τ – τ0 )  =  – α 2 γy 2 τy 2τ y
r
if γ ≥ 0
r
if γ < 0
(7247)
where
γ0
and
τ0
represent the values of strain and stress at the point of load reversal. Subsequent load
reversals are detected by
(γ − γ 0 ) γ& < 0 .
The RambergOsgood equations are inherently onedimensional and are assumed to apply to shear components. To generalize this theory to the multidimensional case, it is assumed that each component of the deviatoric stress and deviatoric tensorial strain is independently related by the onedimensional stressstrain equations. A projection is used to map the result back into deviatoric stress space if required. The volumetric behavior is elastic, and, therefore, the pressure p is found by
p = Kε v
(7248) is the volumetric strain.
where
εv
Material Model 81 and 82: Plasticity with Damage and Orthotropic Option
With this model an elastoviscoplastic material with an arbitrary stress versus strain curve and arbitrary strain rate dependency can be defined. Damage is considered before rupture occurs. Also, failure based on a plastic strain or a minimum time step size can be defined. An option in the keyword input, ORTHO, is available, which invokes an orthotropic damage model. This option, which is implemented only for shell elements with multiple integration points through thickness, is an extension to include orthotropic damage as a means of treating failure in aluminum panels. Directional damage begins after a defined failure strain is reached in tension and continues to evolve until a tensile rupture strain is reached in either one of the two orthogonal directions. The stress versus strain behavior may be treated by a bilinear stress strain curve by defining the tangent modulus, ETAN. Alternately, a curve similar to that shown in Figure 721 is expected to be defined by (EPS1,ES1)  (EPS8,ES8); however, an effective stress versus effective plastic strain curve (LCSS) may be input instead if eight points are insufficient. The cost is roughly the same for either approach. The most general approach is to use the table definition (LCSS) discussed here.
Main Index
304 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
Yield stress versus effective plastic strain for undamaged material
σ yield
Nominal stress after failure
Damage increases linearly with plastic strain after failure
0 Failure Begins
ω = 0
Rupture
ω = 1
ε eff
p
Figure 721
Stress Strain Behavior When Damage is Included.
Two options to account for strain rate effects are possible. Strain rate may be accounted for using the Cowper and Symonds model which scales the yield stress with the factor,
· ε 1⁄p 1 + ⎛ ⎞ ⎝ C⎠
(7249)
· ε = · ε i j ε ij . If the viscoplastic option is active, VP=1.0, and if SIGY is > 0 then
where
· ε
is the strain rate,
s p the dynamic yield stress is computed from the sum of the static stress, σ y ( ε eff ) , which is typically given by a TABLED1 ID, and the initial yield stress, SIGY, multiplied by the CowperSymonds rate term as follows:
·p ε ef f 1 ⁄ p p ·p s p σ y ( ε eff ,ε eff ) = σ y ( ε ef f ) + SI GY ⋅ ⎛  ⎞ ⎝ C⎠
(7250)
where the plastic strain rate is used. If SIGY=0, the following equation is used instead where the static s p stress, σ y ( ε eff ) , must be defined by a TABLED1:
·p ε ef f 1 ⁄ p p ·p s σ y ( ε eff ,ε ef f ) = σ y 1 + ⎛  ⎞ ⎝ C⎠
(7251)
This latter equation is always used if the viscoplastic option is off.
Main Index
Chapter 7: Materials 305
Lagrangian Material Models for SOL 700
For complete generality a TABLED1 (LCSR) to scale the yield stress may be input instead. In this curve the scale factor versus strain rate is defined. The constitutive properties for the damaged material are obtained from the undamaged material properties. The amount of damage evolved is represented by the constant, ω , which varies from zero if no damage has occurred to unity for complete rupture. For uniaxial loading, the nominal stress in the damaged material is given by:
P σ nominal = A
(7252)
where
P
is the applied load and A is the surface area. The true stress is given by: (7253)
P σ t rue = A – A l os s
where
A loss
is the void area. The damage variable can then be defined:
0≤ω≤1
A loss ω = A
(7254)
In this model damage is defined in terms of plastic strain after the failure strain is exceeded:
p ε ef f – ε fpai lu re ω = p ε rapt ur e – ε fpai lu re
if
p p ε fpai lu re ≤ ε eff ≤ ε ru pt ur e
(7255)
After exceeding the failure strain softening begins and continues until the rupture strain is reached. By default, deletion of the element occurs when all integration points in the shell have failed. Note in Figure 722 that the origin of the curve is at (0,0). It is permissible to input the failure strain, f s , as zero for this option. The nonlinear damage curve is useful for controlling the softening behavior after the failure strain is reached.
1 Damage
ε eff – f s
p
Failure
Figure 722
A Nonlinear Damage Curve (Optional)
Main Index
306 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
Material Model 83: FuChang’s Foam With Rate Effects
This model allows rate effects to be modeled in low and medium density foams, see Figure 723. Hysteretic unloading behavior in this model is a function of the rate sensitivity with the most rate sensitive foams providing the largest hysteresis and visa versa. The unified constitutive equations for foam materials by FuChang [1995] provide the basis for this model. This implementation incorporates the coding in the reference in modified form to ensure reasonable computational efficiency. The mathematical description given below is excerpted from the reference.
Figure 723
Rate Effects in Fu Chang's Foam Model
The strain is divided into two parts: a linear part and a nonlinear part of the strain
E ( t ) = EL ( t ) + EN ( t )
(7256)
and the strain rate become
· · · E ( t ) = EL ( t ) + EN ( t ) · EN
(7257)
EN .
is an expression for the past history of
∞
A postulated constitutive equation may be written as: (7258)
σ(t) =
∫
τ =0
[ E tN ( τ ) ,S ( t ) ] dτ
∞
where
S (t )
is the state variable and
∫ is a functional of all values of τ in T τ : 0 ≤ τ ≤ ∞ and
τ =0
E tN ( τ ) = E N ( t – τ )
(7259)
Main Index
Chapter 7: Materials 307
Lagrangian Material Models for SOL 700
where τ is the history parameter:
E tN ( τ = ∞ ) ⇔ the virgin material
(7260)
τ = 0.
It is assumed that the material remembers only its immediate past, i.e., a neighborhood about Therefore, an expansion of E tN ( τ ) in a Taylor series about τ = 0 yields:
∂E tN E tN ( τ ) = E N ( 0 ) +  ( 0 )d t ∂t
(7261)
Hence, the postulated constitutive equation becomes:
· σ ( t ) = σ∗ ( E N ( t ) , E N ( t ) , S ( t ) )
(7262) by
· EN ,
where we have replaced For a special case,
· σ ( t ) = σ∗ ( E N ( t ) , S ( t ) )
∂E tN ∂t
and
σ∗
is a function of its arguments.
(7263)
we may write
· E tN = f ( S ( t ), s ( t ) )
(7264)
which states that the nonlinear strain rate is the function of stress and a state variable which represents the history of loading. Therefore, the proposed kinetic equation for foam materials is:
σt r ( σ S ) 2n 0 · E N =  D 0 exp – c 0 ⎛  ⎞ ⎝ ( σ 2)⎠ σ
(7265)
S
where
c0 → ∞
and n 0 are material constants, and then the nonlinear strain rate vanishes.
D 0, c 0
is the overall state variable. If either
D0 = 0
or
n n · · S i j = [ c 1 ( a i j R – c 2 S ij ) P + c 3 W 1 ( W N ) 2 I i j ]R
(7266) (7267) (7268) (7269)
· n3 EN R = 1 + c 4 ⎛  – 1⎞ ⎝ c5 ⎠ P = W =
∫ tr ( σE
· N)
∫ t r ( σ dE )
c 1, c 2, c 3, c 4, c 5, n 1, n 2, n 3,
where
σ · E · EN
and
aij
are material constants and:
= ( σ ij σ i j ) 1 / 2 · · = ( E ij E i j ) 1 / 2 = ·N ·N ( Ei j Ei j )1 / 2
(7270)
Main Index
308 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
In the implementation by Fu Chang the model was simplified such that the input constants state variables S i j are scalars.
ai j
and the
Material Model 87: Cellular Rubber
This material model provides a cellular rubber model combined with linear viscoelasticity as outlined by Christensen [1980]. Rubber is generally considered to be fully incompressible since the bulk modulus greatly exceeds the shear modulus in magnitude. To model the rubber as an unconstrained material a hydrostatic work term, W H ( J ) , is included in the strain energy functional which is function of the relative volume, J , [Ogden, 1984]
n
W ( J 1 ,J 2 ,J ) =
–1 ⁄ 3 I1 I 3 –2 ⁄ 3
∑
p, q = 0
C pq ( J 1 – 3 ) p ( J 2 – 3 ) q + W H ( J )
(7271)
J1 =
J2 = I2 I 3
In order to prevent volumetric work from contributing to the hydrostatic work the first and second invariants are modified as shown. This procedure is described in more detail by Sussman and Bathe [1987]. The effects of confined air pressure in its overall response characteristics are included by augmenting the stress state within the element by the air pressure.
σ i j = σ isk – δ ij σ ai r j
(7272)
σ a ir
where
σ isk j
is the bulk skeletal stress and
is the air pressure computed from the equation: (7273)
γ
p0 γ σ air = – 1+γ–φ
where p 0 is the initial foam pressure usually taken as the atmospheric pressure and volumetric strain
γ = V – 1 + γ0
defines the
(7274)
where V is the relative volume of the voids and γ 0 is the initial volumetric strain which is typically zero. The rubber skeletal material is assumed to be incompressible. Rate effects are taken into account through linear viscoelasticity by a convolution integral of the form:
σi j = ∫0 g ij kl ( t – τ )  dt ∂τ Si j ,
t
∂ε kl
(7275) and Green's strain tensor, , E ij
or in terms of the second PiolaKirchhoff stress,
Main Index
Chapter 7: Materials 309
Lagrangian Material Models for SOL 700
Si j =
∫0 G ij kl ( t – τ )  dt ∂τ
t
∂E kl
(7276)
where g i jkl ( t – τ ) and G i jkl ( t – τ ) are the relaxation functions for the different stress measures. This stress is added to the stress tensor determined from the strain energy functional. Since we wish to include only simple rate effects, the relaxation function is represented by one term from the Prony series:
N
g ( t ) = a0 +
∑
m=1
α m + e – βt
(7277)
given by
g ( t ) = Ed e
– β1 t
.
(7278)
This model is effectively a Maxwell fluid which consists of a damper and spring in series. We characterize this in the input by a shear modulus, G , and decay constant, β 1 . The MooneyRivlin rubber model is obtained by specifying n = 1 . In spite of the differences in formulations with Model 27, we find that the results obtained with this model are nearly identical with those of 27 as long as large values of Poisson’s ratio are used. By setting the initial air pressure to zero, an open cell, cellular rubber can be simulated as shown in Figure 724.
Figure 724
Cellular Rubber with Entrapped Air
Main Index
310 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
Material Model 89: Plasticity Polymer
Unlike other MD Nastran SOL 700 material models, both the input stressstrain curve and the strain to failure are defined as total true strain, not plastic strain. The input can be defined from uniaxial tensile tests; nominal stress and nominal strain from the tests must be converted to true stress and true strain. The elastic component of strain must not be subtracted out. The stressstrain curve is permitted to have sections steeper (i.e. stiffer) than the elastic modulus. When these are encountered the elastic modulus is increased to prevent spurious energy generation.
Material Model 94: Inelastic Spring Discrete Beam
The yield force is taken from the TABLED1:
F Y = F y ( Δ L plasti c )
(7279)
where
L plasti c
is the plastic deflection. A trial force is computed as: (7280)
F:
· FT = Fn + K ⋅ Δ L ⋅ Δ t
and is checked against the yield force to determine
⎧ Y F = ⎨ F ⎩ FT if F T > F Y if F T ≤ F Y
(7281)
The final force, which includes rate effects and damping, is given by:
· ⎛ ⎧ Δ L ⎫⎞ · · · · F n + 1 = F ⋅ 1 + C1 ⋅ Δ L + C2 ⋅ sgn ( Δ L ) ln ⎜ ma x ⎨ 1. , ⎬⎟ + DΔ L + g ( Δ L ) h ( Δ L ) D LE ⎭⎠ ⎝ ⎩
(7282)
where
C 1 , C2
are damping coefficients,
DL E
is a factor to scale time units.
Unless the origin of the curve starts at (0,0), the negative part of the curve is used when the spring force is negative where the negative of the plastic displacement is used to interpolate, F y . The positive part of the curve is used whenever the force is positive. In these equations, Δ L is the change in length
Δ L = current length – initial length
Material Model 97: General Joint Discrete Beam
For explicit calculations, the additional stiffness due to this joint may require additional mass and inertia for stability. Mass and rotary inertia for this beam element is based on the defined mass density, the volume, and the mass moment of inertia defined in the beam property input. The penalty stiffness applies to explicit calculations. For implicit calculations, constraint equations are generated and imposed on the system equations; therefore, these constants, RPST and RPSR, are not used.
Main Index
Chapter 7: Materials 311
Lagrangian Material Models for SOL 700
Material Model 98: Simplified Johnson Cook
Johnson and Cook express the flow stress as
n · σ y = ( A + B ε p ) ( 1 + c ln ε * )
where A, B, C and n are input constants
εp · ε · ε * = ·ε0
ellective plastic strain effective strain rate for
· ε 0 = 1 s –1
The maximum stress is limited by sigmax and sigsat by:
n · σ y = min { min [ A + Bε p ,sig max ] ( 1 + c ln ε * ) ,sig sat }
Failure occurs when the effective plastic strain exceeds psfail. If the viscoplastic option is active, VP=1.0, the SIGMAX and SIGSAT parameters are ignored since these parameters make convergence of the viscoplastic strain iteration loopdifficult to achieve. The viscoplastic option replaces the plastic strain in the forgoing equationsby the viscoplastic strain and the strain rate by the viscoplastic strain rate. Numerical noise issubstantially reduced by the viscoplastic formulation.
Material Model 100: Spot Weld
This material model applies to beam element type 9 for spot welds. These beam elements may be placed between any two deformable shell surfaces, see Figure 725, and tied with type 7 constraint contact which eliminates the need to have adjacent nodes at spot weld locations. Beam spot welds may be placed between rigid bodies and rigid/deformable bodies by making the node on one end of the spot weld a rigid body node which can be an extra node for the rigid body. In the same way, rigid bodies may also be tied together with this spot weld option. It is advisable to include all spot welds which can be arbitrarily placed within the structure, which provide the slave nodes, and spot welded materials, which define the master segments, within a single type 7 tied interface. As a constraint method, multiple type 7 interfaces are treated independently which can lead to significant problems if such interfaces share common nodal points. The offset option, “o 7”, should not be used with spot welds. The weld material is modeled with isotropic hardening plasticity coupled to two failure models. The first model specifies a failure strain which fails each integration point in the spot weld independently. The second model fails the entire weld if the resultants are outside of the failure surface defined by:
⎛ N rr ⎞ 2 ⎛ N rs ⎞ 2 ⎜ ⎟ + ⎜ ⎟ + ⎝ N rr F⎠ ⎝ N rs F⎠ ⎛ N rt ⎞ 2 ⎜  ⎟ + ⎝ N rt F⎠ ⎛ M rr ⎞ 2 ⎜ ⎟ + ⎝ M rr F⎠ ⎛ M ss ⎞ 2 ⎛ T rr ⎞ 2 ⎜ ⎟ – 1 ⎜  ⎟ = 0 ⎝ M ss F⎠ ⎝ T rr F⎠
(7283)
Main Index
312 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
Figure 725
Deformable Spotwelds
where the numerators in the equation are the resultants calculated in the local coordinates of the cross section, and the denominators are the values specified in the input. If the user defined parameter, NF, which the number of force vectors stored for filtering, is nonzero the resultants are filtered before failure is checked. The default value is set to zero which is generally recommended unless oscillatory resultant forces are observed in the time history databases. Even though these welds should not oscillate significantly, this option was added for consistency with the other spot weld options. NF affects the storage since it is necessary to store the resultant forces as history variables. If the failure strain is set to zero, the failure strain model is not used. In a similar manner, when the value of a resultant at failure is set to zero, the corresponding term in the failure surface is ignored. For example, if only N rr is nonzero, the failure surface is reduced to N rr = N rr . None, either, or both of the failure
F F
models may be active depending on the specified input values. The inertias of the spot welds are scaled during the first time step so that their stable time step size is Δt. A strong compressive load on the spot weld at a later time may reduce the length of the spot weld so that stable time step size drops below Δ t . If the value of Δ t is zero, mass scaling is not performed, and the spot welds will probably limit the time step size. Under most circumstances, the inertias of the spot welds are small enough that scaling them will have a negligible effect on the structural response and the use of this option is encouraged. Spotweld force history data is written into the SWFORC ASCII file. In this database the resultant moments are not available, but they are in the binary time history database.
Main Index
Chapter 7: Materials 313
Lagrangian Material Models for SOL 700
The constitutive properties for the damaged material are obtained from the undamaged material properties. The amount of damage evolved is represented by the constant, ω , which varies from zero if no damage has occurred to unity for complete rupture. For uniaxial loading, the nominal stress in the damaged material is given by
P σ nominal = A
(7284)
A
where
P
is the applied load and
is the surface area. The true stress is given by: (7285)
P σ t rue = A – A l os s
where
A loss
is the void area. The damage variable can then be defined:
0≤ω≤1
A loss ω = A
(7286)
In this model damage is defined in terms of plastic strain after the failure strain is exceeded:
ε eff – ε fail ure p p p ω =  if ε fai lure ≤ ε eff ≤ ε rup tu re p p ε rupt ure – ε f ai lu re
p p
(7287)
After exceeding the failure strain softening begins and continues until the rupture strain is reached.
Material Model 116: Composite Layup
This material is for modeling the elastic responses of composite layups that have an arbitrary number of layers through the shell thickness. A preintegration is used to compute the extensional, bending, and coupling stiffness for use with the BelytschkoTsay resultant shell formulation. The angles of the local material axes are specified from layer to layer in the PCOMP or PSHELLD input. This material model must be used with the userdefined integration rule for shells, which allows the elastic constants to change from integration point to integration point. Since the stresses are not computed in the resultant formulation, the stress output to the binary databases for the resultant elements are zero. This material law is based on standard composite layup theory. The implementation, [See Jones 1975], allows the calculation of the force, N , and moment, M , stress resultants from:
⎧ ⎪ Nx ⎪ ⎨ Ny ⎪ ⎪ N xy ⎩ ⎧ ⎪ Mx ⎪ ⎨ My ⎪ ⎪ M xy ⎩ ⎫ ⎪ ⎪ ⎬ = ⎪ ⎪ ⎭ ⎫ ⎪ ⎪ ⎬ = ⎪ ⎪ ⎭ ⎧ 0 A 11 A 12 A 16 ⎪ ε x ⎪ 0 A 21 A 22 A 26 ⎨ ε y ⎪ 0 A 16 A 26 A 66 ⎪ ε z ⎩ ⎧ 0 B 11 B 12 B 16 ⎪ ε x ⎪ 0 B 21 B 22 B 26 ⎨ ε y ⎪ 0 B 16 B 26 B 66 ⎪ ε z ⎩ ⎫ ⎪ B 11 B 12 B 16 ⎪ + B 21 B 22 B 26 ⎬ ⎪ B 16 B 26 B 66 ⎪ ⎭ ⎫ ⎪ D 11 D 12 D 16 ⎪ ⎬ + D 21 D 22 D 26 ⎪ D 16 D 26 D 66 ⎪ ⎭ ⎧ ⎪ κx ⎪ ⎨ κy ⎪ ⎪ κ xy ⎩ ⎧ ⎪ κx ⎪ ⎨ κy ⎪ ⎪ κ xy ⎩ ⎫ ⎪ ⎪ ⎬ ⎪ ⎪ ⎭ ⎫ ⎪ ⎪ ⎬ ⎪ ⎪ ⎭
(7288)
Main Index
314 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
where
A ij
is the extensional stiffness,
Di j
is the bending stiffness, and
B ij
is the coupling stiffness, which
is a null matrix for symmetric layups. The midsurface strains and curvatures are denoted by ε i0j and κ ij , respectively. Since these stiffness matrices are symmetric, 18 terms are needed per shell element in addition to the shell resultants, which are integrated in time. This is considerably less storage than would typically be required with through thickness integration which requires a minimum of eight history variables per integration point; e.g., if 100 layers are used, 800 history variables would be stored. Not only is memory much less for this model, but the CPU time required is also considerably reduced.
Material Model 119: General Nonlinear Six Degrees of Freedom Discrete Beam
Catastrophic failure, which is based on displacement resultants, occurs if either of the following inequalities are satisfied:
⎛ ur ⎞ 2 ⎛ u s ⎞ 2 ⎛ ut ⎞ 2 ⎛ θt ⎞ 2 ⎛ θs ⎞ 2 ⎛ θt ⎞ 2 ⎜  ⎟ + ⎜ l⎟ + ⎜  ⎟ + ⎜  ⎟ + ⎜  ⎟ + ⎜  ⎟ – 1. ≥ 0 t tf t t ⎝ u r fa il⎠ ⎝ u s ai ⎠ ⎝ u tt fa il⎠ ⎝ θ r fa il⎠ ⎝ θ s fa il⎠ ⎝ θ tt fa il⎠ ⎛ ur ⎞ ⎛ us ⎞ ⎛ ut ⎞ ⎛ θt ⎞ ⎛ θs ⎞ ⎛ θt ⎞ ⎜ l⎟ + ⎜ l⎟ + ⎜  ⎟ + ⎜ l⎟ + ⎜ l⎟ + ⎜  ⎟ – 1. ≥ 0 cf cfai cf cf ⎝ u r ai ⎠ ⎝ us ⎠ ⎝ u tcfail⎠ ⎝ θ r ai ⎠ ⎝ θ s ai ⎠ ⎝ θ tcf a il⎠
2 2 2 2 2 2
(7289)
After failure the discrete element is deleted. If failure is included either the tension failure or the compression failure or both may be used.
R E S U L T A N T Loadingunloading Curve DISPLACEMENT DISPLACEMENT R E S U L T A N T Unloading Curve
Unload = 0
Unload = 1
Unloading Curve
R E S U L T A N T
Unloading Curve
R E S U L T A N T DISPLACEMENT
DISPLACEMENT
Umin
OFFSET x Umin
Unload = 2
Unload = 3
Figure 726
Load and Unloading Behavior
Main Index
Chapter 7: Materials 315
Lagrangian Material Models for SOL 700
Material Model 124: TensionCompression Plasticity
This is an isotropic elasticplastic material where a unique yield stress versus plastic strain curve can be defined for compression and tension. Failure can occur based on plastic strain or a minimum time step size. Rate effects are modeled by using the CowperSymonds strain rate model. The stressstrain behavior follows one curve in compression and another in tension. The sign of the mean stress determines the state where a positive mean stress (i.e., a negative pressure) is indicative of tension. Two TABLED1s, f t ( p ) and f c ( p ) , are defined, which give the yield stress, σ y , versus effective plastic strain for both the tension and compression regimes. The two pressure values, p t and p c , when exceeded, determine if the tension curve or the compressive curve is followed, respectively. f the pressure, p , falls between these two values, a weighted average of the two curves are used:
pc – p ⎧ ⎪ sc ale = p + p – pt ≤ p ≤ p c ⎨ c t ⎪ σ = sc ale ⋅ f ( p ) + ( 1 – sc ale ) ⋅ f ( p ) t c ⎩ y
if
Strain rate is accounted for using the Cowper and Symonds model, which scales the yield stress with the factor
· ε 1/p 1 + ⎛ ⎞ ⎝ C⎠
(7290)
· ε = · · ε i j ε ij .
where
· ε
is the strain rate
Material Model 126: Modified Honeycomb
For efficiency it is strongly recommended that the TABLED1 ID’s: LCA, LCB, LCC, LCS, LCAB, LCBC, and LCCA, contain exactly the same number of points with corresponding strain values on the abscissa. If this recommendation is followed the cost of the table lookup is insignificant. Conversely, the cost increases significantly if the abscissa strain values are not consistent between TABLED1s. The behavior before compaction is orthotropic where the components of the stress tensor are uncoupled; i.e., a component of strain will generate resistance in the local αdirection with no coupling to the local b and c directions. The elastic modulii vary from their initial values to the fully coaaumpacted values linearly with the relative volume:
E aa = E aau + β aa ( E – E aau ) E bb = E bbu + β bb ( E – E bbu ) E cc = E ccu + β cc ( E – E ccu ) G a b = G ab u + β ( G – G a b u ) G b c = G bcu + β ( G – G bcu ) G ca = G cau + β ( G – G cau )
(7291)
where
1–v B = ma x min ⎛ ⎛  ,1⎞ , 0⎞ ⎝ ⎝ 1 – vf ⎠ ⎠
(7292)
Main Index
316 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
and
G
is the elastic shear modulus for the fully compacted honeycomb material (7293)
E G = 2(1 + υ)
The relative volume, ν , is defined as the ratio of the current volume over the initial volume, and typically, V = 1 at the beginning of a calculation. The TABLED1s define the magnitude of the stress as the material undergoes deformation. The first value in the curve should be less than or equal to zero corresponding to tension and increase to full compaction. Care should be taken when defining the curves so the extrapolated values do not lead to negative yield stresses. At the beginning of the stress update we transform each element’s stresses and strain rates into the local element coordinate system. For the uncompacted material, the trial stress components are updated using the elastic interpolated modulii according to:
n σ aa+ 1 n σ bb+ 1 n σ cc+ 1 t rial t rial t rial n = σ aa + E aa Δ ε aa n = σ bb + E bb Δ ε bb n = σ cc + E cc Δ ε cc n σ a b+ 1 n σ b c+ 1 n σ ca+ 1 t ria l t ria l t ria l n = σ a b + 2 G a b Δ ε ab n = σb c + 2 Gb c Δ εb c n = σ ca + 2 G ca Δ ε ca
(7294)
We then independently check each component of the updated stresses to ensure that they do not exceed the permissible values determined from the TABLED1s, e.g., if
σ inj + 1
t rial
> λ σi j ( εi j )
(7295)
then
n λσ ij + 1 σ inj + 1 = σ i j ( ε ij ) t rial σ in + 1 tri al
(7296)
The components of σ i j ( ε ij ) are defined by TABLED1s. The parameter λ is either unity or a value taken from the TABLED1 number, LCSR, that defines λ as a function of strainrate. Strainrate is defined here as the Euclidean norm of the deviatoric strainrate tensor. For fully compacted material we assume that the material behavior is elasticperfectly plastic and updated the stress components according to:
n dev s itjrial = s ij + 2G Δ ε ij n+1⁄2
(7297)
where the deviatoric strain increment is defined as
1 Δ ε idev = Δ ε ij –  Δ ε kk δ i j j 3
(7298)
Main Index
Chapter 7: Materials 317
Lagrangian Material Models for SOL 700
We now check to see if the yield stress for the fully compacted material is exceeded by comparing
3 tri tri t rial s eff = ⎛  s ij al s ij al⎞ ⎝2 ⎠
1⁄2
(7299)
the effective trial stress to the yield stress, σ y . If the effective trial stress exceeds the yield stress, we simply scale back the stress components to the yield surface
σy s inj + 1 =  s itjrial t rial s eff
(7300)
K
We can now update the pressure using the elastic bulk modulus,
n p n + 1 = p n – K Δ ε kk+ 1 ⁄ 2
E K = 3 (1 – 2 υ)
(7301)
and obtain the final value for the Cauchy stress
n σ inj + 1 = s ij + 1 – p n + 1 δ ij
(7302)
After completing the stress update we transform the stresses back to the global configuration. In Figure 727, note that the "yield stress" at a strain of zero is nonzero. In the TABLED1 definition, the "time" value is the directional strain and the "function" value is the yield stress.
Curve extends into negative strain quadrant since SOL 700 will extrapolate using the two end points. It is important that the extrapolation does not extend into the negative stress region.
Unloading is based on the interpolated Young’s moduli which must provide an unloading tangent that exceeds the loading tangent.
Figure 727
Stress Quantity Versus Strain
Main Index
318 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
Material Model 127: ArrudaBoyce Hyperviscoelastic Rubber
This material model, described in the paper by Arruda and Boyce [1993], provides a rubber model that is optionally combined with linear viscoelasticity. Rubber is generally considered to be fully incompressible since the bulk modulus greatly exceeds the shear modulus in magnitude; therefore, to model the rubber as an unconstrained material, a hydrostatic work term, W H ( J ) , is included in the strain energy functional which is function of the relative volume, J :
1 1 11  2  3 W ( J 1, J 2, J ) = nk θ  ( J 1 – 3 ) +  ( J 1 – 9 ) +  ( J 1 – 27 ) 2 20 N 1050N 2 19 3 4 519  5 + nk θ ⎛  ⎞ ( J 1 – 81 ) +  ( J 1 – 243 ) + W H ( J ) ⎝ 7000N⎠ 673750 N 4 J1 = I1 J –1 ⁄ 3
(7303)
J2 = I2 J
The hydrostatic work term is expressed in terms of the bulk modulus,
K W H ( J ) =  ( H – 1 ) 2 2
K,
and J , as: (7304)
Rate effects are taken into account through linear viscoelasticity by a convolution integral of the form:
σi j = ∫0 g ij kl ( t – τ )  dτ ∂τ Si j ,
t
∂ε kl
(7305) and Green's strain tensor, Eij , (7306)
or in terms of the second PiolaKirchhoff stress,
Si j = ∫0 G ij kl ( t – τ )  dτ ∂τ
t
∂E kl
where g i jkl ( t – τ ) and G i jkl ( t – τ ) are the relaxation functions for the different stress measures. This stress is added to the stress tensor determined from the strain energy functional. If we wish to include only simple rate effects, the relaxation function is represented by six terms from the Prony series:
N
g ( t ) = α0 +
∑
m=1
α m e – βt
(7307)
given by,
N
g(t) =
∑ Gi e
i =1
–βi t
(7308)
This model is effectively a Maxwell fluid which consists of a dampers and springs in series. We characterize this in the input by shear moduli, G i , and decay constants, β i . The viscoelastic behavior is optional and an arbitrary number of terms may be used.
Main Index
Chapter 7: Materials 319
Lagrangian Material Models for SOL 700
When viscoelastic terms are not included, this model is similar to the use of the Arruda Boyce model in solution 600, defined in the MATHE option. When viscoelasticity is included, the formulation of these two models are different.
Material Model 158: Rate Sensitive Composite Fabric
See material type 58, Laminated Composite Fabric, for the treatment of the composite material. Rate effects are taken into account through a Maxwell model using linear viscoelasticity by a convolution integral of the form:
σi j = ∫0 gij kl ( t – τ )  dτ ∂τ
t
∂ε kl
(7309)
where g i jkl t – τ is the relaxation function for different stress measures. This stress is added to the stress tensor determined from the strain energy functional. Since we wish to include only simple rate effects, the relaxation function is represented by six terms from the Prony series:
N
g(t) =
∑
m=1
Gm e
–βm t
We characterize this in the input by the shear moduli, G i , and the decay constants, β i . An arbitrary number of terms, not exceeding 6, may be used when applying the viscoelastic model. The composite failure is not directly affected by the presence of the viscous stress tensor.
Material Model 181: Simplified Rubber Foam
Material type 181 in SOL 700is a simplified “quasi”hyperelastic rubber model defined by a single uniaxial TABLED1 or by a family of curves at discrete strain rates. The term “quasi” is used because there is really no strain energy function for determining the stresses used in this model. However, for deriving the tangent stiffness matrix we use the formulas as if a strain energy function were present. In addition, a frequency independent damping stress is added to model the energy dissipation commonly observed in rubbers. Hyperelasticity Using the Principal Stretch Ratios A hyperelastic constitutive law is determined by a strain energy function that here is expressed in terms of the principal stretches; i.e., ) , W = W ( λ 1 ,λ 2 ,λ 3 ) . To obtain the Cauchy stress σ i j , as well as the
TC constitutive tensor of interest, C ij kl , they are first calculated in the principal basis after which they are transformed back to the “base frame”, or standard basis. The complete set of formulas is given by Crisfield [1997] and is for the sake of completeness recapitulated here.
The principal Kirchhoff stress components are given by
∂W E τ ij = λ i ∂λ i ( no sum )
(7310)
Main Index
320 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
that are transformed to the standard basis using the standard formula
E τ i j = q ik q jl τ kl
(7311)
The q i j are the components of the orthogonal tensor containing the eigenvectors of the principal basis. The Cauchy stress is then given by
σ i j = J – 1 τij
(7312) is the relative volume change.
where
J = λ 1 λ 2 λ3
The constitutive tensor that relates the rate of deformation to the Truesdell (convected) rate of Kirchhoff stress can in the principal basis be expressed as
∂ iE i TKE C ii jj = λ j  – 2 τ iE δ ij i ∂λ j λ j2 τ iE – λ i2 τ jE i TKE C ij ij = j 2 λ 1 – λ j2
E λ i ∂τ ii ∂τ iE TKE C ij ij =  ⎛  – i ⎞ 2 ⎝ ∂λ i ∂λ j ⎠
i ≠ j ,( λ i ≠ λ j )
(no sum)
(7313)
i ≠ j ,λ i = λ j
These components are transformed to the standard basis according to
TKE TKE C ij kl = q i p q j g q kr q ls C pqrs
(7314)
and finally the constitutive tensor relating the rate of deformation to the Truesdell rate of Cauchy stress is obtained through
TC K C ij kl = J – 1 C iTjkl
(7315)
Stress and Tangent Stiffness The principal Kirchhoff stress is in material model 181 given by
3
1 τ iE = f ( λ i ) + K ( J – 1 ) – 3
∑ f ( λk )
k=1
(7316)
where f is a TABLED1 determined from uniaxial data (possible at different strain rates). Furthermore, K is the bulk modulus and J is the relative volume change of the material. This stress cannot be deduced from a strain energy function unless f ( λ ) = E ln λ for some constitutive parameter E . A consequence of this is that when using the formulas in the previous section the resulting tangent stiffness matrix is not necessarily symmetric. We remedy this by symmetrizing the formulas according to
Main Index
Chapter 7: Materials 321
Lagrangian Material Models for SOL 700
⎧ 2 ⎪ E E E  λ i f' ( λ i ) ∂τ ii ⎞ ∂τ ii ∂ τ jj ⎞ ⎪ 1⎛ 3 ⎛ λ =  ⎝ λ j  + λ i  ⎠ = KJ + ⎨ j ∂λ ⎠ ⎝ ∂λ j ∂λ i 2 j symn · ⎪ – 1 ( λ f ( λ ) + λ f' ( λ ) ) j j ⎪  i i 6 ⎩
if i = j
(7317)
otherwise
Two Remarks The function f introduced in the previous section depends not only on the stretches but for some choices of input also on the strain rate. Strain rate effects complicate things for an implicit analyst and here one also has to take into account whether the material is in tension/compression or in a loading/unloading stage. We believe that it is of little importance to take into account the strain rate effects when deriving the tangent stiffness matrix and therefore this influence has been disregarded. For the fully integrated brick element, we have used the approach in material model 77 to account for the constant pressure when deriving the tangent stiffness matrix. Experiments have shown that this is crucial to obtain a decent implicit performance for nearly incompressible materials. Modeling of the Frequency Independent Damping An elasticplastic stress
σd
is added to model the frequency independent damping properties of rubber.
G
This stress is deviatoric and determined by the shear modulus stress is updated incrementally as
˜ n + 1 = σ n + 2 GI dev Δ ε σd d
and the yield stress
σ Y . This part of the
(7318)
where Δ ε . is the strain increment. The trial stress is then radially scaled (if necessary) to the yield surface according to
σd
n+1
σY ˜n+1 = σ d m in ⎛ 1 , ⎞ ⎝ σ eff⎠ σ ef f
where
is the effective von Mises stress for the trial stress
˜n+1 σd
The elastic tangent stiffness contribution is given by
C d = 2G I dev
(7319)
and if yield has occurred in the last time step the elasticplastic tangent is used
3G C d = 2G I dev –  σ d ⊗ σ d 2 σY
(7320)
Here
I dev
is the deviatoric fourth order identity tensor.
Main Index
322 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
Material Model 196: General Spring Discrete Beam
If TYPE=0, elastic behavior is obtained. In this case, if the linear spring stiffness is used, the force, given by:
· F = F 0 + K Δ L + DΔ L F , is
(7321)
but if the TABLED1 ID is specified, the force is then given by:
· ⎛ ⎧ Δ L ⎫⎞ · · · · F = F 0 + K f ( ΔL ) 1 + C1 ⋅ Δ L + C2 ⋅ sgn ( Δ L ) ln ⎜ ma x ⎨ 1. , ⎬⎟ + DΔ L + g ( Δ L ) h ( Δ L ) D LE ⎭⎠ ⎝ ⎩
In these equations, .L is the change in length
Δ L = current length – initial length
If TYPE=1, inelastic behavior is obtained. In this case, the yield force is taken from the TABLED1:
F Y = F y ( Δ L plasti c )
(7322)
where
L plasti c
is the plastic deflection. A trial force is computed as: (7323)
F:
· FT = Fn + K Δ L Δ t
and is checked against the yield force to determine
⎧ Y F = ⎨ F ⎩ FT if F T > F Y if F T ≤ F Y
(7324)
The final force, which includes rate effects and damping, is given by:
· ⎛ ⎧ Δ L ⎫⎞ · · · · F n + 1 = F ⋅ 1 + C1 ⋅ Δ L + C2 ⋅ sgn ( Δ L ) ln ⎜ ma x ⎨ 1. , ⎬⎟ + DΔ L + g ( Δ L ) h ( Δ L ) ⎝ ⎩ D LE ⎭⎠
Unless the origin of the curve starts at (0,0), the negative part of the curve is used when the spring force is negative where the negative of the plastic displacement is used to interpolate, F y . The positive part of the curve is used whenever the force is positive. The crosssectional area is defined on the section card for the discrete beam elements. See *SECTION_BEAM. The square root of this area is used as the contract thickness offset if these elements are included in the contact treatment.
Main Index
Chapter 7: Materials 323
Lagrangian Material Models for SOL 700
Grunseisen Equation of State (EOSGRUN)
The Gruneisen equation of state with cubic shock velocityparticle velocity defines pressure for compressed material as
γ0 α ρ 0 C 2 μ 1 + ⎛ 1 – ⎞ μ –  μ 2 ⎝ 2⎠ p =  + ( γ 0 + αμ )E μ2 μ3 1 – ( S 1 – 1 )μ – s 2  – s 3 2 μ+1 (μ + 1)
(7325)
where E is the internal energy per initial volume, C is the intercept of the u s – u p curve, S 1, S 2 , and S 3 are the coefficients of the slope of the u s – u p curve, γ 0 is the Gruneisen gamma, and a is the first order volume correction to γ 0 . Constants, C, S 1, S 2, S 3, γ 0 , and α , are userdefined input parameters. The compression is defined in terms of the relative volume, V , as:
1 μ =  – 1 . V
For expanded materials as the pressure is defined by:
p = ρ 0 C 2 μ + ( γ 0 α μ )E
(7326)
Tabulated Compaction Equation of State (EOSTABC)
Pressure is positive in compression, and volumetric strain ε V is positive in tension. The tabulated compaction model is linear in internal energy per unit volume. Pressure is defined by
p = C ( ε V ) + γ T ( ε V )E
(7327)
during loading (compression). Unloading occurs at a slope corresponding to the bulk modulus at the peak (most compressive) volumetric strain, as shown in Figure 728. Reloading follows the unloading path to the point where unloading began, and then continues on the loading path described by Equation (7327)). In the compacted states, the bulk unloading modulus depends on the peak volumetric strain.
Main Index
324 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Lagrangian Material Models for SOL 700
C B Pressure p A
A K εv
B K εv
C K εv
A εv
B εv
C εv
Volumetric Strain ( – ε V )
Figure 728
Pressure vs Volumetric Strain Curve for EOSTABC
Tabulated Equation of State (EOSTAB)
The tabulated equation of state model is linear in internal energy. Pressure is defined by
p = C ( ε V ) + γ T ( ε V )E
(7328)
The volumetric strain ε V is given by the natural algorithm of the relative volume. Up to 10 points and as few as 2 may be used when defining the tabulated functions. The pressure is extrapolated if necessary. Loading and unloading are along the same curve unlike tabulated compaction equation of state (EOSTABC).
Main Index
Chapter 7: Materials 325
Eulerian Material Models for SOL 700
Eulerian Material Models for SOL 700
MATDEUL – Eulerian Material
The MATDEUL material entry is a general material definition and provides a high degree of flexibility in defining material behavior. The basis of the MATDEUL entry is the reference of a combination of material descriptions: equation of state, yield model, shear model, failure model, and spall model. The only material parameter defined on the MATDEUL entry is the reference density. The MATDEUL entry can be used to define all types of material behavior from materials with very simple linear equations of state to materials with complex yielding and shearing behavior and different failure criteria. The required input is the reference density, the number of an EOSxxx entry defining the equation of state, and the number of an SHRxxx entry defining the shear properties of the material. The equation of state defines the bulk behavior of the material. It may be a polynomial equation, a gamma law gas equation, or an explosive equation. A singleterm polynomial equation produces a linear elastic behavior. Further material property definitions are optional. A referenced YLDxxx entry selects one of the following: a hydrodynamic response (zero yield stress), a von Mises criterion that gives a bilinear elastoplastic behavior, or a JohnsonCook yield model where the yield stress is a function of plastic strain, strain rate, and temperature. If no YLDxxx model is referenced, the material is assumed to be fully elastic. A FAILxxx entry can be referenced to define a failure model for the material. This failure model can be based on a maximum plastic strain limit or Johnson Cook failure model. If no FAILxxx entry is referenced, the material has no failure criterion. A PMINxxx entry can be referenced to define the spall characteristics of the material. Currently, only the PMINC entry is available. The entry provides a constant spall limit for the material. When no PMINxxx entry is referenced, the material has a zero spall limit.
Shear Models
The shear model is referenced from a MATDEUL entry. It defines the shear behavior of the material. At present, an elastic shear model is available with a constant or polynomial shear modulus. SHREL – Constant Modulus Shear Model The SHREL entry defines a shear model with a constant shear modulus G. The model is referenced from a MATDEUL entry that defines the general material properties.
Main Index
326 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Eulerian Material Models for SOL 700
Shear Stress
G
Strain
Figure 729
Elastic Shear as Function of Strain.
SHRPOL – Polynomial Shear Model The SHRPOL model defines a polynomial shear model where the shear modulus is related to the effective plastic shear strain by a cubic equation.
G = G0 + G1 γ + G2 γ + G3 γ
2 3
where
γ
= effective plastic shear strain and
G 0, G 1, G 2
and
G3
are constants.
Yield Models
Yield models is referenced by MATDEUL entries. The yield models can be used to model elastic perfectly plastic behavior, bilinear elastoplastic behavior, piecewise linear behavior, or hydrodynamic behavior (zero yield stress). YLDHY – Hydrodynamic Yield Model The YLDHY entry defines a yield model with constant zero yield stress. This model should be used for fluids that have no shear strength and are, therefore, hydrodynamic. YLDVM – von Mises Yield Model The YLDVM entry defines a von Mises yield model. The yield stress and hardening modulus are defined by giving either a bilinear or piecewise linear stressstrain curve. Only an elastic perfectly plastic yield model can be used. The hardening modulus is not used.
Main Index
Chapter 7: Materials 327
Eulerian Material Models for SOL 700
Bilinear Representation
σ σ0 Eh
E ε
where the yield stress
E Eh σ y = σ 0 +  ε p E – Eh
σy
is given by
where
σ0 E Eh εp
= yield stress = Youngs modulus = hardening modulus = equivalent plastic strain
Piecewise Linear Representation
σ
ε
During every iteration, the stress s is determined from the current equivalent strain from the stressstrain table
σ = [ ( σi – σ i – 1 ) ( ε i – εi – 1 ) ⁄ ( ε i εi – 1 ) ] + σi – 1
ε
by interpolating
where σ j and ε j are the points in the table. The stressstrain characteristic used internally in MD Nastran Nonlinear Explicit (SOL 700) is in terms of true stress and equivalent plastic strain. However, for convenience, the stressstrain characteristic can be input in any of the following ways:
• True stress/true strain • Engineering stress/engineering strain • True stress/plastic strain • True stress/plastic modulus
Main Index
328 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Eulerian Material Models for SOL 700
True stress is defined as
F σ t rue = A
where
F
= current force,
ε pl
A
= current area.
Plastic strain
is
ε pl = ε true – ε el
where
ε t rue
= true strain,
ε el
= elastic strain
True strain is defined as
ε t ru e =
∫ l
dl
dl
where
= incremental change in length,
σ en g
l
= current length.
ε eng
By comparison, engineering stress
F σ eng = A0
and strain
are given by
where
A0
= original area = original length
( I – I0 ) ε eng = I0
where
I0
True stress/true strain and engineering strain are related by the following formulas:
σ t rue = σ eng ( 1 + ε eng ) ε t ru e = ln ( 1 + ε eng )
At small strains, there is little difference between true stressstrain and engineering stressstrain. However, at moderate and large strains there can be very large differences, and it is important that the correct stressstrain characteristic is input. When defining the material properties using Young’s modulus, yield stress, and hardening modulus, the hardening modulus must be estimated from a plot of true stress versus true strain. This estimate may well require a measured material characteristic to be replotted. Some simple examples follow:
Main Index
Chapter 7: Materials 329
Eulerian Material Models for SOL 700
True Stress Versus True Strain True Stress
σy
E
True Strain
Figure 730
True Stress vs True Strain Curve
The slope of the first segment of the curve gives the Young’s modulus for the material (when it is not defined explicitly) and the first nonzero stress point gives the yield stress σ y . The point corresponding to the origin can be omitted.
Engineering Stress Versus Engineering Strain Engineering Stress
σy
E
Engineering Strain
Figure 731
Engineering Stress vs Engineering Strain Curve
True Stress Versus Plastic Strain True Stress
σy
Plastic Strain
Figure 732
True Stress vs Plastic Strain Curve
Since the curve is defined in terms of the equivalent plastic strain, there is no elastic part in the curve. The first point must be the yield stress of the material at zero plastic strain. Young’s modulus is defined separately.
Main Index
330 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Eulerian Material Models for SOL 700
True Stress Versus Plastic Modulus True Stress
σ2 σ1 E h2 E h1
Plastic Strain
Figure 733
True Stress vs Plastic Strain Curve
This option is slightly different since the curve is specified as a series of pairs of stress and hardening moduli, rather than as a series of pairs of stress and strain. Young’s modulus and yield stress are defined explicitly so that the table consists of pairs of values with the hardening modulus (xaxis) and the true stress (yaxis) at the end of the segment. Yielding occurs when the von Mises stress
σ vm = [ ( σ1 – σ2 ) + ( σ2 – σ3 ) + ( σ3 – σ 1 ) 2 ] ⁄ 2 σy .
2 2
exceeds the yield stress
The principal stresses are
σ , σ2 ,
and
σ3 .
Isotropic hardening is assumed, which means that the yield surface increases in diameter as yielding occurs, but its center does not move. This yield model can be used with beam, shell, and solid elements. When used with shell or solid elements, strainrate sensitivity and failure can be included. Strainrate sensitivity can be defined in two ways: 1. You can specify a table giving the variation of a scale factor S with strainrate d ε ⁄ dt . The scale factor is multiplied by the stress found from the stressstrain characteristic to give the actual stress. The failure criterion is based on plastic strain. When the plastic strain exceeds the specified value, the element fails. All the stresses are set to zero, and the element can carry no load. (This failure criterion is referred to from the DMATEP or the DYMAT24 entry.)
· σd ⎧ε⎫  = 1 + ⎨  ⎬ 1 ⁄ P σy ⎩D ⎭
where
σd
is the dynamic yield stress,
σy
· is the static yield stress and is ε the equivalent strain rate.
YLDJC – JohnsonCook Yield Model The YLDJC entry defines a JohnsonCook yield model in which the yield stress is a function of the plastic strain, strain rate, and temperature
· · n σ y = ( A + B ε p ) ( 1 + C ln ( ε ⁄ ε o ) ) ( 1 – T * m )
Main Index
Chapter 7: Materials 331
Eulerian Material Models for SOL 700
where
T* εp · ε · ε0 T Tt Tm
=
(T – Tr ) ( Tm – Tr )
= effective plastic strain = effective strain rate = reference strain rate = temperature = room temperature = melt temperature and
m
A, B, n, C ,
are constants
YLDTM – TanimuraMimura Yield Model The YLDTM entry defines a TanimuraMimura yield model in which the yield stress is a function of the plastic strain, strain rate, and temperature
σy = · · A + B εp ε k ε A + B ε p + ( C + D ε p ) ⎛ 1 – ⎞ ln ⎛  ⎞ ( 1 – T * m ) + E ⎛ ⎞ ··⎝ ⎝ε ⎠ σ cr ⎠ ⎝ ε s⎠ 0 (T – Tr ) ( Tm – Tr )
where
T* T Tr Tm εp · ε · εs · ε0 σ cr
=
= Temperature = room temperature = melt temperature = effective plastic strain = effective strain rate = Quasistatic strain rate = reference strain rate = Critical yield stress and
k
A , B , C, D , m , E ,
are constants
This yield model is suitable for a wide range of strain rates, strains and temperatures.
Main Index
332 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Eulerian Material Models for SOL 700
YLDZA – ZerilliArmstrong Yield Model The YLDZA entry defines a ZerilliArmstrong yield model in which the yield stress is a function of the plastic strain, strain rate, and temperature
n σ y = ( A + B ε p )e · ⎛ ε⎞ – mT + CT ln ⎜ ⎟ · ε 0⎠ ⎝
for Fcc metals
σy = ( A +
n B ε p )D
+e
· ⎛ ε⎞ – mT + CT ln ⎜ ⎟ · ⎝ ε 0⎠
for Bcc metals
where
εp · ε · ε0 T
= effective plastic strain = effective strain rate = reference strain rate = Temperature and
D
A, B, n, C, m ,
are constants
This yield model can be used for both Fcc type of metals, like iron and steels, as well as Bcc type of metals, like aluminum and alloys. YLDRPL – Rate Power Law Yield Model The YLDRPL entry defines a rate power law yield model in which the yield stress is a function of the plastic strain and strain rate.
n· σ y = M AX ( C, A + B ε p ε m )
where
εp · ε
= effective plastic strain = effective strain rate and
C
A, B, n, m ,
are constants
YLDPOL – Polynomial Yield Model The YLDPOL entry defines a polynomial yield model in which the yield stress is a function of the plastic strain
2 3 4 5 σ y = M IN ( σ max, A + B ε p + C ε p + Dε p + E ε p + Fε p )
where
εp σ max
= effective plastic strain = Maximum yield stress and
F
A, B, C, D, E ,
are constants
Main Index
Chapter 7: Materials 333
Eulerian Material Models for SOL 700
YLDSG – SteinbergGuinan Yield Model The YLDSG entry defines a SteinbergGuinan yield model in which the yield stress is a function of the plastic strain, strain rate, and temperature.
AT = A 1 ( 1 + A 3 εp )
A4
ρ 3 σ y = m in ( A 2, A T ) 1 – H ( T – T r ) + B p ⎛  ⎞ , T < T m ⎝ ρ ref⎠ σ y = 0, T ≥ T m
1 
where
εp T Tr Tm p ρ
= effective plastic strain = temperature = room temperature = melt temperature = Pressure = density and
B
A, …, A 4, H ,
are constants
YLDMSS – Multisurface Yield Model for Snow Defines the yield model for snow material. This card must be used in combination with MATDEUL,
EOSPOL, and SHREL.
Snow is a very specific material that lies between water and ice. In micro term, the structure of snow looks like general porous materials where the degree of compaction can vary quite large. Therefore, from the constitutive equation point of view, snow belongs to the family of soil. One of plasticity models applied for snow is a multisurface one. The multisurface plasticity model for snow, see references 1 and 2 (there are misprinting in those papers) is characterized by two independent hardening (softening) mechanisms and a set of yield functions as shown in the following form:
f c ( I 1 , J 2, q c ( α c ) ) = cc 4 J 2 +  ( I 1 + q c ) – κ c I 1 + qc cc qc
3
= 0
f t ( I 1, q t ( α t ) ) = I 1 – q t = 0
3
I1
and
J2
are the first and second invariant of the stress tensor and
I 1 = T – I 1 . The material parameter T
is related to the cohesion of snow. q c ( q t ) is the hardening (softening) parameter associated with the yield surface f c ( f t ) . c c determines the shape of f c . Hence, it is a model parameter that may be set independently
Main Index
334 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Eulerian Material Models for SOL 700
from the specific type of snow. κ c is a material parameter related to the angle of friction. Figure 734 contains plots of the yield functions f c and f t in the meridian plane at different stages of the hardening process of
fc
and the softening process of f t , respectively.
J2
DruckerPrager failure surface
Hardening
Softeningg
Yield surfaces
I1
Tension cutoff planes
Figure 734
ft
Snow Model: Plots of Loading Functions in the Meridian Plane
defines a socalled “tensioncutoff”plane perpendicular to the hydrostatic axis. For simplicity, a linear softening law is adopted:
qt = ft t – Ds αt
where f tu is the hydrostatic tensile strength of snow, accumulated plastic volumetric tensile strain.
Ds
is the softening modulus and
αt
represents the
For the implementation in MD Nastran Nonlinear Explicit (SOL 700), the accumulated plastic strain is updated if the tensilepressure is bigger than the current q t . The incremental strain is calculated using the difference of the pressure divided by the bulk modulus. Then the new q t is updated to be used in the next cycle. Furthermore, the deviatoric stresses are brought to zero. constitutes a smooth yield function closed along the compressive and the tensile branch of the hydrostatic axis. Its shape in the stress space changes continuously in the course of hardening, see Figure 734. A specific hardening law, similar to the one used in the Cap Model (reference 3) was adopted for snow on the basis of results from hydrostatic compression tests:
fc αc 1 q c =  ln ⎛ 1 – ⎞ 2a c ⎝ bc⎠ αc – fc bc ln ( 1 – f c ) q c =  – 2 ( ac bc ( 1 – fc ) ) 2a c
if if
αc ≤ fc bc
αc ≤ fc bc
Main Index
Chapter 7: Materials 335
Eulerian Material Models for SOL 700
are parameters determined from hydrostatic compression tests. f c is a parameter that avoids singularity in above equations at α c = b c . It will be set to 0.99. As α c grows, the model obtains a shape similar to the DruckerPrager failure criterion, see Figure 734. The following relation obtains the correlation between the proposed model for snow and the DruckerPrager model.
κ c = α Dp
a c, b c
The plasticity evolution is done using an additive plasticity model and associative flow rule (with isotropic hardening law) as follows:
ε = εe + εp σ = C : (ε – εp) · · ∂f v ε = λ ∂σ
The incremental plastic strains can be derived as follows:
· ∂f c = Δ t λ ∂σ ∂f c ∂J 2D ∂f c ∂I 1  = Δλ   +  ∂J 2D ∂σ ∂I 1 ∂ σ ⎧ ⎫ cc 3 ⎪ ⎪ 4  ( I 1 + q c ) qc ⎪ ⎪ S = Δ λ  + ⎨ κ c –  ⎬ cc cc ⎪ 4⎪ 2 J 2 +  ( I 1 + q c ) ⎪ 2 J 2 +  ( I 1 + q c ) ⎪ qc qc ⎩ ⎭
Δ εp
n+1
n+1
They consist of deviatoric and volumetric plastic strains as follows:
S Δ e p = Δ λ cc 4 2 J 2 +  ( I 1 + q c ) qc cs 3 4  ( I 1 + q c ) qc = 3 Δ λ κ c – cc 4 2 J 2 +  ( I 1 + q c ) qc
p Δ εν
Δ λ is calculated according to the following procedure. First the trial stresses are updated using elastic assumption. σ n + 1 = Kt ra ce ε n + 1 + 2G e n + 1 = σ E – Kt ra ce Δ ε n + 1 – 2 Ge n + 1
e e p p
From the above formulation we can derive the following relation:
⎧ ⎫ cc 3 ⎪ ⎪ 4  ( I 1 + q c ) qc ⎪ ⎪ S = Δ λ – 3K ⎨ κ c –  ⎬ 1 – 2 G cc cc ⎪ 4⎪ 4 2 J 2 +  ( I 1 + q c ) ⎪ 2 J 2 +  ( I 1 + q c ) ⎪ qc qc ⎩ ⎭
σn + 1 – σ E
Main Index
336 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Eulerian Material Models for SOL 700
Using the Newton iteration scheme as follows:
∂f c f c ( σ n + 1 ) ≈ f c ( σ E ) + ∂σ : ( σn + 1 – σ E )
E J1
The following relation is obtained.
⎛ ⎞ ⎜ ⎟ S f c ( σ E ) + ⎜  ⎟ : Δ λ ( σ n + 1 – σ E ) = 0 ⎜ ⎟ cc ⎜ 2 J +  ( I + q ) 4⎟ 2 c ⎠ ⎝ qc 1
Therefore,
Δλ
can be calculated as follows:
fc ( σ E ) fc ( σE ) Δ λ =  = J 2D cc 3 ⎛ ⎞2 G 4  ( I 1 + q c ) ⎜ ⎟ J2 D qc cc 4 9K ⎜ κ c –  ⎟ + G J 2 +  ( I 1 + q c ) ⎜ ⎟ cc 4 qc cc 4 ⎜ J 2 +  ( I 1 + q c ) 2 J 2 +  ( I 1 + q c ) ⎟ qc ⎝ ⎠ qc
In this way the volumetric equivalent plastic strain can be updated with the consequence that the yield surface is growing. Therefore a few iterations are needed to bring the trial stresses, σ n + 1 , back to the updated yield surface with a chosen accuracy. Using this model an excellent agreement between simulation and the experiment results has been achieved as mentioned in reference 1 and 4.
References
1. “Friction Mechanisms of Tread Blocks on Snow Surfaces”, R. Mundl, G. Meschke and W. Liederer, Tire Science and Technology, TSTCA, vol. 25, no. 4, 1997, pp. 245264. 2. “A New Viscoplastic Model for Snow at Finite Strains”, G. Meschke, 4th International Conference on Computational Plasticity, Barcelona, April 36, 1995, Pineridge Press, pp. 22952306 3. “Material Models for Granular Soils”, F.L. DiMaggio and I.S. Sandler, Journal of Engineering Mechanics A.S.C.E., 1971, pp. 935950 4. “Implementation and Verification of Snow Material in Euler Scheme; Using the type10 element (hydrodynamic material with strength)”, M. Mahardika, MSC.Software, 2001, to be published.
Equations of State
Equations of state are referenced from the MATDEUL entry. The equation of state for a material is of the basic form Pressure = f (density, specific internal energy)
Main Index
Chapter 7: Materials 337
Eulerian Material Models for SOL 700
The simplest equation of state is the gamma law equation of state, defined by the EOSGAM entry. The only input required is the ratio of specific heats for an ideal gas. The EOSPOL entry defines a polynomial equation of state. The EOSTAIT entry defines an equation of state based on the Tait model in combination with a cavitation model. The EOSJWL entry defines an equation of state based on the JWL explosive model. It is used to calculate the pressure of the detonation products of high explosives. The JWL model is empirically based and requires the input of five constants. The EOSIG entry defines the properties of Ignition and Growth equation of state and the reaction rate equation used to model high explosives. EOSGAM – Gamma Law Equation of State The EOSGAM model defines a gamma law equation of state for gases where the pressure is a function of the density, the specific internal energy, and the ideal gas ratio of specific heats γ of an ideal gas
p = (γ – 1) ρ e
where
e ρ
= specific internal energy unit mass = overall material density
( Cp ⁄ C v )
γ = ratio of specific heats
The EOSGAM equation of state can also be used to model viscous gases. EOSPOL – Polynomial Equation of State The EOSPOL model defines a polynomial equation of state where the pressure is related to the relative volume and specific internal energy by a cubic equation. In compression
(μ > 0)
p = a 1 μ + a 2 μ 2 + a 3 μ 3 + ( b 0 + b 1 μ + b 2 μ 2 + b 3 μ 3 )ρ 0 e
In tension
(μ ≤ 0)
p = a1 μ + ( b0 ÷ b1 μ ) ρ0 e
where
μ η ρ
= =
η–1 ρ ⁄ ρ0
= overall material density
Main Index
338 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Eulerian Material Models for SOL 700
ρ0 e
= reference density = specific internal energy per unit mass
The EOSPOL equation of state can also be used to model viscous fluids. EOSTAIT – Tait Equation of State The EOSTAIT model defines a equation of state based on the Tait model in combination with a cavitation model where the pressure p is defined as follows: No cavitation
( ρ > ρc ) ,
p = a0 + a1 ( η γ – 1 )
Cavitation
p = pc
( ρ ≤ ρc ) ,
where
η ρ ρ0 ρc
=
ρ ⁄ ρ0
= overall material density = reference density = critical density which produces the cavitation pressure
pc
γ
The pressure can not fall below the cavitation pressure p c can continue to decrease below its critical value ρ c .
= a 0 + a 1 ( ( ( ρ c ) ⁄ ( ρ 0 ) ) – 1 ) , although the density
The EOSTAIT equation of state can als be used to model viscous fluids. EOSJWL – JWL Equation of State This equation of state can be used only with Eulerian elements. where
E ρ0 ρ P
= specific internal energy per unit mass = reference density = overall material density =
ω η –R ⁄ η ω η –R ⁄ η A ⎛ 1 – ⎞ e 1 + B ⎛ 1 – ⎞ e 2 + ω ηρ 0 e ⎝ ⎝ R1 ⎠ R2 ⎠ η = ρ ⁄ ρ0
A, B, ω, R 1 ,
and
R2
are constants.
These parameters are defined in Reference 3.
Main Index
Chapter 7: Materials 339
Eulerian Material Models for SOL 700
A DETSPH entry must be used to specify the detonation time, the location of the detonation point, and the velocity of a spherical detonation wave. When no DETSPH entry is present, all the material detonates immediately and completely. EOSMG  MieGruneisen Equation of State The MieGruneisen equation is useful in highstrain rate processes. The pressure is split in a part that only depends on density and a part that only depends on temperature.
p = pc + p T
The cold pressure is computed from the RankineHugenot equations and is given by
2 Γ0 η ρ 0 c0 η p c =  ⎛ 1 –  ⎞ 2⎝ 2 ⎠ ( 1 – sη )
Here
ρ0
is the reference density,
c0
is the speed of sound,
ρ0 η = 1 – ρ
is the volumetric compressive
strain. The defining equation for the parameter s is the linearization of the relationship between linear shock speed U s and particle velocity U p .
U s = c 0 + sU p
The thermal part of the pressure follows from thermodynamic considerations and reads
pT = Γ0 ρ0 e
where
e
is the specific internal energy and the parameter
Γ0
is given by
βK T Γ 0 ρ 0 = CV
where
KT
is the isothermal bulk modulus,
Γ0
CV
the specific heat at constant volume and
β
the volumetric
thermal expansion coefficient.
is the Gruneisen parameter at reference density. The Gruneisen
ρ0 Γ = Γ 0  . ρ
parameter at other densities is given by
Material Failure
One of the nonlinear features of a material's behavior is failure. When a certain criterion the failure criterion is met, the material fails and can no longer sustain its loading and breaks. In a finiteelement method this means that the element where the material reaches the failure limit, cannot carry any stresses anymore. The stress tensor is effectively zero. The element is flagged for failure and essentially is no longer part of the structure. Failure criteria can be defined for a range of materials and element types. The failure models are referenced from the material definition entries.
Main Index
340 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Eulerian Material Models for SOL 700
There are several different failure models available: FAILMPS FAILJC Constant, maximum plastic strain. The Johnsoncook failure model
FAILMPS – Maximum Plastic Strain Failure Model The most commonly used failure model is the one that is based on a maximum equivalent plastic strain. The material fails completely when the plastic strain reaches beyond the defined limit. FAILJC – JohnsonCook Failure Model Failure is determined from a damage model. Damage is an element variable and increments are given by the plastic strain increment divided by a fracture strain. In addition the damage variable is transported along with material as it move from one Euler element to the other. It is only available for the Multimaterial Euler solver with strength. The use of coupling surfaces is not supported.
Spallation Models
A spallation model defines the minimum pressure prior to spallation. At present there is only one spallation model, PMINC, that defines a constant spallation pressure. PMINC – Constant Minimum Pressure A constant minimum pressure must be defined that must be less than or equal to zero. Note that the pressure is positive in compression. If the pressure in an element falls below the minimum pressure, the element spall and the pressure and yield stress are set to zero. The material then behaves like a fluid. When the pressure subsequently becomes positive, the material will no longer be in a spalled state. The pressure can then decrease again to the specified minimum (the spall limit) before spallation occurs again.
Pressure
PMINC
Volume
Figure 735
Minimum Pressure Cutoff
Main Index
Chapter 7: Materials 341
Eulerian Material Models for SOL 700
Material Viscosity
Viscous fluid material models are available for the Roe fluid solver only. The viscous behavior is referenced from the entry to define the equation of state (EOSPOL or EOSTAIT). For these viscous materials, the stress tensor t ij is defined as:
t ij = p ⋅ δ i j + S ij dp Kdρ  =  dt ρ dt
and
d e idj S i j = 2μ ⋅ dt
d where K denotes the bulk modulus, ρ the density, S i j the deviatoric stress tensor, p the pressure, e ij the deviatoric strain tensor, and μ the coefficient of viscosity. The Roe solver computes the stresses directly from the velocity gradients.
Main Index
342 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Eulerian Material Models for SOL 700
Main Index
Chapter 8: Contact Impact Algorithm MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide Overview
8
Contact Impact Algorithm
Overview 344 346
Penalty Methods Preliminaries Slave Search 346
347 351 352
Contact Force Calculation
Improvements to the Contact Searching Bucket Sorting 353 356
Accounting For the Shell Thickness Initial Contact Penetrations Contact Energy Calculation Friction 359 357 358
Main Index
344 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide Overview
Overview
Overview
The input of contact data, Contact for Explicit Nonlinear (SOL 700), is defined similarly as for Implicit Nonlinear (SOL 600). It is instructive to first read Chapter 12 of the SOL 600 User’s Guide where an overview is given of the nonlinear contact methods. This section will summarize those parts of the contact methods applicable to SOL 700, and will describe the theory of the algorithms used. To activate contact in SOL 700, the Case Control entry BCONTACT must be given. There are three methods available to define contact: 1. All elements in the model in one contact definition, using default settings Case Control: BCONTACT=ALL This will result in automatic contact detection between all elements in the model. For SOL 700, it is advisable to use this method. 2. User defined Contact Bodies, using default or nondefault settings Case Control: BCONTACT = n Bulk Data:
BCBODY, BSURF, or BCBOX or BCPROP or BCMATL, and BCTABLE
It is possible to define contact bodies (BCBODY), and specify which contact bodies need to be checked for contact (BCTABLE). This method using BCBODY definitions provides extreme flexibility and is compatible with the Implicit Nonlinear solution (SOL 600). A contact body is defined by the Bulk Data entry BCBODY, which references a set of elements (BSURF), a set of elements inside a box (BCBOX), or certain property IDs (BCPROP) or with certain material IDs (BCMATL). Often used definitions related to contact methods are: Single Surface Contact: This refers to any contact definition where no master is defined. Master Slave Contact: This refers to any contact definition where a master is defined.
SOL 600 Contact Capabilities not yet supported by SOL 700
• 2D contact, since SOL 700 only applies to 3D • A rigid BCBODY defined by patches or geometric entities
(DIM (BEHAV=RIGID (ISTYP (CONTROL (IDSPL (ERROR (FNTOL
on BCBODY) on BCBODY) on BCBODY) on BCBODY) on BCBODY) on BCTABLE) on BCTABLE)
 Rigid BCBODY as a symmetry plane  Motion/Load controlled rigid BCBODY (See note below on rigid body modeling in SOL 700)
• Body smoothing • Userdefined distance below which a node is
considered touching
• Separation Force
Main Index
Chapter 8: Contact Impact Algorithm 345
Overview
• Interference closure •
(CINTERF (IGLUE &
JGLUE
on BCTABLE) on BCTABLE) on BCTABLE) on BCTABLE) on BCTABLE)
Glue options
• Contact heat transfer • Searching order • Initial node movement for stressfree contact (SOL 700 contact offers similar capability by means of the IGNORE option on the BCTABLE) • •
(HEATC (ISEARCH (ICOORD (BCHANGE entry) (BCMOVE entry)
Option to identify relevant nodes Initial approach, release, motion until contact
Notes on Rigid Body Modeling in SOL 700:
Currently, only BEHAV = DEFORM is supported in SOL 700 on the BCBODY option. Rigid body modeling is possible, however, by defining a rigid material (MATD020).
• Elements belonging to a MATD020 are properly treated in the contact calculations, and it is allowed to include them in a BCBODY with BEHAV = DEFORM. The contact calculations will
operate as if the material is rigid. The penalty based contact forces applied on the nodes are accumulated for the whole rigid body and applied as an external force and moment to the centerofgravity of the rigid body.
• Rigid body motion is allowed and properly simulated by SOL 700. • Initial velocities and boundary conditions acting on the nodes will be applied to the rigid body. • Rigid body must be assigned in “Masters” field in BCTABLE.
With these capabilities, a faceted rigid body, similar to a BCBODY with BEHAV=RIGID can be easily modeled. The BEHAV=RIGID logic will be implemented in the next release of SOL 700.
SOL 600 Contact Limitations that do not apply to SOL 700
To allow switching between SOL 700 and SOL 600, it is advisable to work within these limits. In SOL 600, each node and element should be in at most one contact body. When using the penalty method of SOL 700 (the default), this limitation does not apply. Forces as calculated by each contact are simply accumulated and applied as an external force vector. In SOL 600, only 1000 contact bodies are allowed. This limit does not apply to SOL 700. In SOL 600, it is important to properly define the order in which contact bodies are defined for deformabletodeformable contact. The order of contact body definition has no influence on the results in SOL 700.
Main Index
346 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide Overview
Penalty Methods
Penalty Methods
The treatment of sliding and impact along interfaces is an important capability in SOL 700. Internally, the interfaces are defined in three dimensions by listing in arbitrary order all triangular and quadrilateral segments that comprise each side of the interface. One side of the interface is designated as the slave side, and the other is designated as the master side. Nodes lying in those surfaces are referred to as slave and master nodes, respectively. When slave nodes penetrate the master, normal interface springs are placed between the penetrated nodes and the master surface. See Slave Search in this chapter for details on the contact force calculations.
Preliminaries
Consider the timedependent motion of two bodies occupying regions B 1 and B 2 in their undeformed configuration at time zero. Assume that the intersection
B1 ∩ B2 = φ
(81)
is satisfied. Let ∂B 1 and ∂B 2 denote the boundaries of B 1 and B 2 , respectively. At some later time, these bodies occupy regions b 1 and b 2 bounded by ∂b 1 and ∂b 2 as shown in Figure 81. Because the deformed configurations cannot penetrate,
( b 1 – ∂b 1 ) ∩ b 2 = φ
(82)
∂b
1
∂b
2
b
1
b
2
∂B
1
∂B
2
1 B0
2 B0
Figure 81 As long as
Reference and Deformed Configuration
∂ b 1 ∩ ∂b 2 = φ ,
the equations of motion remain uncoupled. In the foregoing and following
α ( = 1, 2 )
equations, the right superscript
denotes the body to which the quantity refers.
Main Index
Chapter 8: Contact Impact Algorithm 347
Slave Search
Before a detailed description of the theory is given, some additional statements should be made concerning the terminology. The surfaces ∂b 1 and ∂b 2 of the discretized bodies b 1 and b 2 become the master and slave surfaces respectively. Choice of the master and slave surfaces is arbitrary when the symmetric penalty treatment is employed. Otherwise, the more coarsely meshed surface should be chosen as the master surface, nodal points that define ∂b 1 are called master nodes and nodes that define
∂b 2
are called slave nodes. When
( ∂b 1 ∩ ∂b 2 ) ≠ φ ,
the constraints are imposed to prevent penetration.
Right superscripts are implied whenever a variable refers to either the master surface ∂b 1 , or slave surface, ∂b 2 ; consequently, these superscripts are dropped in the development which follows.
Slave Search
The slave search is common to all interface algorithms implemented in SOL 700. This search finds for each slave node its nearest point on the master surface. Lines drawn from a slave node to its nearest point will be perpendicular to the master surface, unless the point lies along the intersection of two master segments, where a segment is defined to be a 3 or 4node element of a surface. Consider a slave node, n s , sliding on a piecewise smooth master surface and assume that a search of the master surface has located the master node, m s , lying nearest to n s . Figure 82 depicts a portion of a master surface with nodes m s and n s labeled. If m s and n s do not coincide, n s can usually be shown to lie in a segment s s via the following tests:
( ci × s ) ⋅ ( ci × ci + 1 ) > 0 ( ci × s ) ⋅ ( s × ci + 1 ) > 0
(83)
where vector c i and c i + 1 are along edges of s 1 and point outward from m s . Vector s is the projection of the vector beginning at m s , ending at n s , and denoted by g , onto the plane being examined (see Figure 83).
Figure 82
Four Master Segments can Harbor Slave Node Master Node
ns
given that
ms
is the Nearest
Main Index
348 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide Overview
Slave Search
Figure 83
s = g – ( g ⋅ m )m
Projection of
g
onto Master Segment
s1
(84)
s1
where for segment
ci × ci + 1 m = ci × ci + 1
(85)
Since the sliding constraints keep n s close but not necessarily on the master surface and since n s may lie near or even on the intersection of two master segments, the inequalities of Equation (83) may be inconclusive; i.e., they may fail to be satisfied or more than one may give positive results. When this occurs n s is assumed to lie along the intersection which yields the maximum value for the quantity
g ⋅ ci ci i = 1, 2, 3, 4, …
(86)
When the contact surface is made up of badly shaped elements, the segment apparently identified as containing the slave node actually may not, as shown in Figure 84. Assume that a master segment has been located for slave node n s and that n s is not identified as lying on the intersection of two master segments. Then the identification of the contact point, defined as the point on the master segment which is nearest to n s , becomes nontrivial. Each master surface segment s 1 , is given the parametric representation:
r = f 1 ( ξ ,η )i 1 + f 2 ( ξ ,η )i 2 + f 3 ( ξ ,η ) i 3
(87)
where
4
f i ( ξ, η ) –
∑ φj xi
j =1
j
(88)
Main Index
Chapter 8: Contact Impact Algorithm 349
Slave Search
When the nearest node fails to contain the segment that harbors the slave node, segments numbered 18 are searched in the order shown in Figure 84.
1
2
8 7
3 5 4
6
Figure 84 Note that
∂r ∂r  ×  ≠ 0 ∂ξ ∂η r1
Search Order for Segments is at least once continuously differentiable and that (89)
Thus, r represents a master segment that has a unique normal whose direction depends continuously on the points of s 1 . Let
t
be a position vector drawn to slave node
ns
and assume that the master surface segment
( ξ c, η c )
s1
has
been identified with
n s . The contact point coordinates
on
s1
must satisfy (810) (811)
∂r  ( ξ c, η c ) ⋅ [ t – r ( ξ c, η c ) ] = 0 ∂ξ ∂r  ( ξ c, η c ) ⋅ [ t – r ( ξ c, η c ) ] = 0 ∂η
The physical problem is illustrated in Figure 85, which shows Equations (810) and (811) are readily solved for Equation (810) for in
ηc ξc ξc
ns
lying above the master surface.
and
ηc .
One way to accomplish this is to solve
in terms of
ηc
, and substitute the results into (811). This yields a cubic equation
.
Main Index
350 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide Overview
Slave Search
η
3 ns
∂r ∂η ∂r ∂ξ
2
ξ
x3
t
4
r x2 x1 1
Figure 85
Location of Contact Point when
ns
lies above Master Segment
The equations are solved numerically. When two nodes of a bilinear quadrilateral are collapsed into a single node for a triangle, the Jacobian of the minimization problem is singular at the collapsed node. Fortunately, there is an analytical solution for triangular segments since three points define a plane. NewtonRaphson iteration is a natural choice for solving these simple nonlinear equations. The method diverges with distorted elements unless the initial guess is accurate. An expanded search procedure as discussed in Improvements to the Contact Searching section is used. Three iterations with a leastsquares projection are used to generate an initial guess:
ξ 0 = 0, r ,ξ r ,η η 0 = 0, r ,ξ r ,η { r ( ξ i , η i ) – t },
⎧ ⎫ [ r ,ξ r ,η ] ⎨ Δ ξ ⎬ = ⎩ Δη ⎭
(812)
ξ i + 1 = ξ i + 1 + Δ ξ,
ηi + 1 = ηi + Δ η
followed by the NewtonRaphson iterations which are limited to ten iterations, but which usually converges in four or less.
⎧ r ,ξ ⎫ ⎧ Δξ ⎫ [H]⎨ ⎬ = – ⎨ ⎬ { r ( ξ i, η i ) – t } ⎩ r ,η ⎭ ⎩Δη ⎭ 0 r ⋅ r ,ξη ⎧ r ,χ ⎫ [ H ] ⎨ ⎬ [ r ,ξ r ,η ] + r ⋅ r ,ξη 0 ⎩ r ,η ⎭ ξi + 1 = ξi + Δ ξ ηi + 1 = ηi + Δ η
(813)
In concave regions, a slave node may have isoparametric coordinates that lie outside of the [ – 1 ,+1 ] range for all of the master segments, yet still have penetrated the surface. A simple strategy is used for handling this case, but it can fail. The contact segment for each node is saved every time step.
Main Index
Chapter 8: Contact Impact Algorithm 351
Contact Force Calculation
If the slave node contact point defined in terms of the isoparametric coordinates of the segment, is just outside of the segment, and the node penetrated the isoparametric surface, and no other segment associated with the nearest neighbor satisfies the inequality test, then the contact point is assumed to occur on the edge of the segment. In effect, the definition of the master segments is extended so that they overlap by a small amount. In the hydrocode literature, this approach is similar to the slide line extensions used in two dimensions. This simple procedure works well for most cases, but it can fail in situations involving sharp concave corners.
Contact Force Calculation
Each slave node is checked for penetration through the master surface. If the slave node does not penetrate, nothing is done. If it does penetrate, an interface force is applied between the slave node and its contact point. The magnitude of this force is proportional to the amount of penetration. This may be thought of as the addition of an interface spring. Penetration of the slave node indicated if
l = n i ( [ t – r ( ξ c, η c ) ] < 0 ) ns
through the master segment which contains its contact point is (814)
where
n i = n i ( ξ c, η c )
(815)
is normal to the master segment at the contact point. The amount of penetration is equal to the value of l. If slave node
fs = –l k i ni ns
has penetrated through master segment
si
, we add an interface force vector
fs :
if l < l < 0
ns
(816) and (817)
to the degrees of freedom corresponding to
i f m = φ 2 ( ξ c, η c )f s
if
l<0
to the four nodes ( i = 1, 2, 3, 4 ) that comprise master segment s i . By default, SOFT=1 on the BCTABLE entry, and the stiffness factor k i is given in terms of the nodal masses and the global timestep, d t as
m slave m mast er 1 k i = f si ⋅  ⋅ m slave + m mast er dt 2
(818)
With SOFT=1, it is mostly not needed to scale the contact stiffness, even if materials with very different stiffness properties come into contact. When SOFT=0 on the BCTABLE entry, the stiffness factor the bulk modulus
Ki , ki
for master segment
si
is given in terms of
si
the volume
Vi ,
and the face area
Ai
of the element that contains
as
Main Index
352 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide Overview
Improvements to the Contact Searching
f si K i A i2 k i = Vi
(819)
for brick elements and
f si K i A i k i = max ( shell diagonal )
(820)
for shell elements. When too much penetration is observed, the contact stiffness can be increased by the interface stiffness scale factor, f si . This scale factor is 0.1 by default, and can be defined by FACT on the BCTABLE entry. Larger values may cause instabilities unless the time step size is scaled back in the time step calculation.
Improvements to the Contact Searching
A number of recent changes have been made in the surfacetosurface contact including contact searching, accounting for thickness, and contact damping. Sometimes problems with the closest master node contact searching were found. The nearest node algorithm described above can break down since the nearest node is not always anywhere near the segment that harbors the slave node as is assumed in Figure 84 (see Figure 86). Such distorted elements are commonly used in rigid bodies in order to define the geometry accurately.
closest nodal point
slave node
Figure 86
Failure to find the Contact Segment can be caused by Poor Aspect Ratios in the Finite Element Mesh
To circumvent the problem caused by bad aspect ratios, an expanded searching procedure is used in which we attempt to locate the nearest segment rather than the nearest nodal point. Nodes 2 and 4 share segments with node 3. Therefore, the two nearest nodes are 1 and 5. The nearest contact segment is not considered since its nodes are not members of the nearest node set.
Main Index
Chapter 8: Contact Impact Algorithm 353
Bucket Sorting
The nearest contact segment to a given node, k , is defined to be the first segment encountered when moving in a direction normal to the surface away from k . A major deficiency with the nearest node search is depicted in Figure 87 where the nearest nodes are not even members of the nearest contact segment. Obviously, this would not be a problem for a more uniform mesh. To overcome this problem we have adopted segment based searching in both surface to surface and single surface contact.
1
2
3
4
5
Normal vector at node 3
Figure 87
Expanded Search
Bucket Sorting
Bucket sorting is now used extensively in the SOL 700 contact algorithms. The reasons for eliminating slave node tracking by incremental searching is illustrated in Figure 88 where surfaces are shown which cause the incremental searches to fail. With bucket sorting incremental searches may still be used but for reliability they are used after contact is achieved. As contact is lost, the bucket sorting for the affected nodal points must resume.
tied interface not yet supported
Figure 88
Examples of Models where Incremental Searching may Fail
Main Index
354 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide Overview
Bucket Sorting
In a direct search of a set of N nodes to determine the nearest node, the number of distance comparisons required is N – 1 . Since this comparison needs to be made for each node, the total number of comparisons is
N(N – 1)
, with each of these comparisons requiring a distance calculation (821)
1 2 = ( xi – x j ) 2 + ( y i – yj ) 2 + ( zi – zj ) 2
that uses eight mathematical operations. The cumulative effect of these mathematical operations for N ( N – 1 ) compares can dominate the solution cost at less than 100 elements. The idea behind a bucket sort is to perform some grouping of the nodes so that the sort operation need only calculate the distance of the nodes in the nearest groups. With this partitioning the nearest node will either reside in the same bucket or in one of the two adjoining buckets. The number of distance calculations is now given by
3N  – 1 α
(822)
where α is the number of buckets. The total number of distance comparisons for the entire onedimensional surface is
3N N ⎛  – 1⎞ ⎝ α ⎠
(823)
Thus, if the number of buckets is greater than 3, then the bucket sort will require fewer distance comparisons than a direct sort. It is easy to show that the corresponding number of distance comparisons for twodimensional and threedimensional bucket sorts are given by
9N N ⎛  – 1⎞ ⎝ αb ⎠ 27 N N ⎛  – 1⎞ ⎝ α bc ⎠
for 2D for 3D
c
(824) (825)
where
b
and
are the number of partitions along the additional dimension.
Incremental searching may fail on surfaces that are not simply connected. The contact algorithm in SOL 700 avoids incremental searching for nodal points that are not in contact and all these cases are considered (see Figure 86). The cost of the grouping operations, needed to form the buckets, is nearly linear with the number of nodes N . For typical SOL 700 applications, the bucket sort is 100 to 1000 times faster than the corresponding direct sort. However, the sort is still an expensive part of the contact algorithm, so that, to further minimize this cost, the sort is performed every ten or fifteen cycles and the nearest three nodes are stored. This can be specified by BSORT on the BCTABLE entry. Typically, three to five percent of the calculational costs will be absorbed in the bucket sorting when most surface segments are included in the contact definition.
Main Index
Chapter 8: Contact Impact Algorithm 355
Bucket Sorting
Bucket Sorting in Single Surface Contact
We set the number of buckets in the x, y, and z coordinate directions to NX, NY, and NZ, respectively. The product of the number of buckets in each direction always approaches NSN or 5000 whichever is smaller,
N X ⋅ N Y ⋅ N Z ≤ M IN ( NS N ,5000 )
(826)
where the coordinate pairs ( x mi n, x max ) , ( y mi n, y max ) , and ( z min, z ma x ) span the entire contact surface. In this procedure, we loop over the segments rather than the nodal points. For each segment we use a nested DO LOOP to loop through a subset of buckets from IM IN to IM A X , J MI N to J MA X , and K MI N to K MA X where:
IM IN = M IN ( P XI, P X 2, P X3, PX 4 ) IM A X = M A X ( P X 1, P X2, P X 3, P X 4 ) J MI N = MA X ( P Y1, P Y2, P Y3, P Y4 ) J MA X = M A X ( P Y1, P Y2, P Y3, P Y4 ) K MI N = MA X ( P Z1, P Z2, P Z3, P Z4 ) k MA X = M A X ( P Z1, P Z2, P Z3, P Z4 )
(827)
and P X k , P Yk , P Zk are the bucket pointers for the kth node. Figure 89 shows a segment passing through a volume that has been partitioned into buckets. The orthogonal distance of each slave node contained in the box from the segment is determined. The box is subdivided into sixty buckets.
Nodes in buckets shown are checked for contact with the segment
z
y
x
Figure 89
The Orthogonal Distance of each Slave Node
Main Index
356 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide Overview
Accounting For the Shell Thickness
We check the orthogonal distance of all nodes in the bucket subset from the segment. As each segment is processed, the minimum distance to a segment is determined for every node in the surface and the two nearest segments are stored. Therefore the required storage allocation is still deterministic. This would not be the case if we stored for each segment a list of nodes that could possibly contact the segment. We have now determined for each node, k , in the contact surface the two nearest segments for contact. Having located these segments we permanently store the node on these segments which is nearest to node k . When checking for interpenetrating nodes we check the segments surrounding the node including the nearest segment since during the steps between bucket searches it is likely that the nearest segment may change. It is possible to bypass nodes that are already in contact and save some computer time; however, if multiple contacts per node are admissible then bypassing the search may lead to unacceptable errors.
Accounting For the Shell Thickness
Shell thickness effects are important when shell elements are used to model sheet metal. Unless thickness is considered in the contact, the effect of thinning on frictional interface stresses due to membrane stretching will be difficult to treat. In the treatment of thickness we project both the slave and master surfaces based on the midsurface normal projection vectors as shown in Figure 810. The surfaces, therefore, must be offset by an amount equal to 1/2 their total thickness (Figure 811). This allows the program to check the node numbering of the segments automatically to ensure that the shells are properly oriented.
Length of projection vector is 1/2 the shell thickness Projected Contact Surface
Figure 810
Contact Surface Based Upon Midsurface Normal Projection Vectors
Main Index
Chapter 8: Contact Impact Algorithm 357
Initial Contact Penetrations
Figure 811
The Slave and Master Surfacess
Thickness changes in the contact are accounted for “if and only if” the shell thickness change option is flagged on the PARAM* DYCONTHKCHG. Each cycle, as the shell elements are processed, the nodal thicknesses are stored for use in the contact algorithms. The interface stiffness may change with thickness depending on the input options used. To account for the nodal thickness, the maximum shell thickness of any shell connected to the node is taken as the nodal thickness and is updated every cycle. The projection of the node is done normal to the contact surface:
Initial Contact Penetrations
The need to offset contact surfaces to account for the thickness of the shell elements contributes to initial contact penetrations. These penetrations can lead to severe numerical problems when execution begins so they should be corrected if SOL 700 is to run successfully. Often an early growth of negative contact energy is one sign that initial penetrations exist. Currently, warning messages are printed to the D3HSP file to report penetrations of nodes through contact segments and the modifications to the geometry made by SOL 700 to eliminate the penetrations. Sometimes such corrections simply move the problem elsewhere since it is very possible that the physical location of the shell midsurface and possibly the shell thickness are incorrect. In the single surface contact algorithms any nodes still penetrating on the second time step are removed from the contact with a warning message. In some geometry's, penetrations cannot be detected since the contact node penetrates completely through the surface at the beginning of the calculation. Such penetrations are frequently due to the use of coarse meshes. This is illustrated in Figure 812. Another case contributing to initial penetrations occurs when the edge of a shell element is on the surface of a solid material as seen in Figure 813. Currently, shell edges are rounded with a radius equal to onehalf the shell thickness.
Main Index
358 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide Overview
Contact Energy Calculation
Detected Penetration
Undetected Penetration
Figure 812
brick
Undetected Penetration
shell
Inner penetration if edge is too close
Figure 813
Undetected Penetration due to rounding the Edge of the Shell Element
To avoid problems with initial penetrations, the following recommendations should be considered:
• Adequately offset adjacent surfaces to account for part thickness during the mesh
generation phase.
• Use consistently refined meshes on adjacent parts which have significant curvatures. • Be very careful when defining thickness on shell and beam section definitions  especially for
rigid bodies.
• Scale back part thickness if necessary. Scaling a 1.5mm thickness to .75mm should not cause
problems but scaling to .075mm might. Alternatively, define a smaller contact thickness by part ID. Warning: if the part is too thin contact failure will probably occur
• Use spot welds instead of merged nodes to allow the shell mid surfaces to be offset.
Contact Energy Calculation
Contact energy,
E contact , is incrementally updated from time n
nsn n+1 n E contact = E contact + nmn
to time
n+1
for each contact interface as:
∑
i =1
Δ F isl ave × Δdi st isl ave +
∑
i =1
Δ F ima ster × Δ d ist imast er
1 n + 2
Main Index
Chapter 8: Contact Impact Algorithm 359
Friction
Where n sn is the number of slave nodes, n mn is the number of master nodes, between the ith slave node and the contact segment node and the contact segment,
Δ di st islave Δ di st ima st er Δ F ima st er
ΔF islave
is the interface force
is the interface force between the ith master
is the incremental distance the ith slave node has moved during
the current time step, and is the incremental distance the ith master node has moved during the current time step. In the absence of friction the slave and master side energies should be close in magnitude but opposite in sign. The sum, E contact , should equal the stored energy. Large negative contact energy is usually caused by undetected penetrations. Contact energies are reported in the GLSTAT file. In the presence of friction and damping discussed below the interface energy can take on a substantial positive value especially if there is, in the case of friction, substantial sliding.
Friction
Friction in SOL 700 is based on a Coulomb formulation. Let
n
f*
be the trial force,
fn
the normal force,
k the interface stiffness, μ the coefficient of friction, and f the frictional force at time n. The frictional algorithm, outlined below, uses the equivalent of an elastic plastic spring. The steps are as follows: 1. Compute the yield force,
Fy = μ fn Fy :
(828)
2. Compute the incremental movement of the slave node
n n n n Δ e = r n + 1 ( ξ c + 1, η c + 1 ) – r n + 1 ( ξ c , η c )
(829)
3. Update the interface force to a trial value:
f∗ = f n – kΔ e
(830)
4. Check the yield condition:
f n + 1 = f∗
if
f∗ ≤ F y
(831)
5. Scale the trial force if it is too large:
F y f∗ f n + 1 = f∗
if
f∗ > F y
(832)
An exponential interpolation function smooths the transition between the static, μ s , and dynamic, μ d , coefficients of friction where ν is the relative velocity between the slave node and the master segment:
μ = μd + ( μ s – μ d ) e–c ν
(833)
where
ΔeΔT ν = Δt
(834)
C
is the time step size, and
is a decay constant.
Main Index
360 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide Overview
Friction
Typical values of friction, see Table 81, can be found in Marks Engineering Handbook. Table 81 Typical Values of Coulomb Friction [Marks] STATIC 0.78 (dry) 0.74 (dry) 0.61 (dry) 1.05 (dry) 0.90 (dry) SLIDING .08 (greasy), .42 (dry) .10 (greasy), .57 (dry) .47 (dry) 1.4 (dry) .69(wet), .85(dry)
MATERIALS Hard steel on hard steel Mild steel on mild steel Aluminum on mild steel Aluminum on aluminum Tires on pavement (40psi)
Main Index
Chapter 9: Fluid Structure Interaction MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
9
Fluid Structure Interaction
General Coupling 362 367
Multiple Coupling Surfaces with Multiple Euler Domains Fluid and Gas Solver for the Euler Equations Modeling Fluid Filled Containers Hotfilling 373 372 371
Main Index
362 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
General Coupling
General Coupling
Fluidstructure Interaction
The objective of fluidstructure interaction using the coupling algorithm is to enable the material modeled in Eulerian and Lagrangian meshes to interact. Initially, the two solvers are entirely separate. Lagrangian elements that lie within an Eulerian mesh do not affect the flow of the Eulerian material and no forces are transferred from the Eulerian material back to the Lagrangian structure. The coupling algorithm computes the interaction between the two sets of elements. It thus enables complex fluidstructure interaction problems to be analyzed. The first task in coupling the Eulerian and Lagrangian sections of a model is to create a surface on the Lagrangian structure. This surface is used to transfer the forces between the two solver domains. The surface acts as a boundary to the flow of material in the Eulerian mesh. At the same time, the stresses in the Eulerian elements cause forces to act on the coupling surface, distorting the Lagrangian elements. By means of a BSURF, BCPROP, BCMATL, or BCSEG entry, you can define a multifaceted surface on the Lagrangian structure. A set of BCSEGs, element numbers, property numbers, material numbers, or any combination of these identify the element faces in this surface. The method of defining of the surface is therefore extremely flexible and can be adapted to individual modeling needs. The coupling algorithm is activated using the COUPLE entry. It specifies that the surface is used for EulerLagrange coupling. You can define whether the inside or the outside domain is covered by the coupling surface by setting the COVER field on the entry. This means that the Euler domain cannot contain material where it is covered by the outside or the inside of the Lagrangian structure. For problems where the Eulerian material is inside a Lagrangian structure (for example, an inflating air bag), COVER should be set to OUTSIDE since the Eulerian elements outside the coupling surface must be covered. For problems where the Eulerian material is outside the Lagrangian structure (for example a projectile penetrating soft material), the inside of the coupling surface must covered, and COVER should be set to INSIDE. The coupling surface must have a positive volume. This means that the normals of all the segments of the surface must point outwards. By default, the solver will check the direction of the normal vectors and automatically reverse them when necessary. However, if you wish to switch of the check to save some computational time in the generation of the problem, you can define this using the REVERSE field on the COUPLE entry. The coupling algorithm activated using the COUPLE entry is the most general interaction algorithm. It can handle any Euler mesh. There is an option, however, to switch to a faster algorithm by setting the parameter DYPARAM, FASTCOUP. This algorithm makes use of knowledge of the geometry of the Euler mesh. As a result, the requirement is that the Euler mesh must be aligned with the basic coordinate system axes.
Main Index
Chapter 9: Fluid Structure Interaction 363
General Coupling
Closed Volume
The coupling surface must form a closed volume. This requirement is fundamental to the way the coupling works. It means that there can be no holes in the surface and the surface must be closed. In order to create a closed volume, it may be necessary to artificially extend the coupling surface in some problems. In the example shown below, a plate modeled with shell elements is interacting with an Eulerian mesh. In order to form a closed coupling surface, dummy shell elements are added behind the plate. Dummy shell elements can be created using material MATD009. The shape of these dummy shell elements does not matter and it is best to use as few as possible to make the solution more efficient. The closed volume formed by the coupling surface must intersect at least one Euler element otherwise the coupling surface is not recognized by the Eulerian mesh.
Eulerian Domain
Lagrangian Shell Elements (Property 200)
Dummy Elements to Form a Closed Coupling Surface (Property 200)
COUPLE, 1, 10, INSIDE BCPROP, 10, 100, 200 PSHELL, 100, 100, 0.05 PSHELL, 200, 999, 1.0E6 MATD009, 999, 1.0E20
Care must be taken when doing so, however. The additional grid points created for the dummy elements do not move, since they are not connected to any structural elements. When the shell elements move so far that they pass beyond these stationary grid points, the coupling surface turns inside out and has a negative volume causing the solver to terminate.
Main Index
364 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
General Coupling
Porosity
There is a general purpose capability to model porosity of a coupling surface. Porosity allows material to flow from the Eulerian region through the coupling surface or viceversa. This method is addressed using the PORFLOW, PORFCPL entry. PORFLOW is used to model the interaction between an Eulerian region to the environment while PORFCPL is used for flow from an Eulerian region to another one. For air bags, a newer and better methodology for modeling porosity has been implemented. The coupling surface or parts of it can be made porous by referring to a LEAKAGE entry from the COUPLE entry. This will be further explained by means of two examples. The first example models an air bag with porous material and two holes using PORFLOW:
Dummy Elements with PID = 100
Membrane Elements with PID = 1
Dummy Elements with PID = 100
The required input is: PSHELL, 1, 100 , 1.E3 PSHELL, 100, 999, 1.E6 BCPROP, 11, 1, 100 BCPROP, 100, 1 COUPLE, 1, 11, OUTSIDE, , , 55 LEAKAGE, 1, 55, , PORFLOW, 42, CONSTANT, 0.009 LEAKAGE, 2, 55, 22, PORFLOW, 42, CONSTANT, 1.0 PORFLOW, 42, , MATERIAL, 33, PRESSURE, 1.E5, METHOD, PRESSURE The second example models two chambers divided by a membrane with a hole. Two sets of Euler elements must be defined in which each set belongs to each coupling surface (COUPLE). The interaction between the two sets of Euler elements is defined by using a hole. The hole is modeled by using LEAKAGE that refers to PORFCPL.
High Pressure Hole Low Pressure
Main Index
Chapter 9: Fluid Structure Interaction 365
General Coupling
The required input is: chamber 1 (low pressure) BSURF,1,..... BSURF,10,1,.... COUPLE,60,1,,,,70,,,+ +,,,,,,,,,+ +,,21 LEAKAGE,80,70,10,PORFCPL,50, ,<coeffv> PORFCPL,50,,,,1020 MESH,21,BOX,,,,,,,+ +,0.,0.,0.,1.,1.,1.,,,+ +,10,10,10,,,,EULER,500 PEULER1,500,,HYDRO,19 TICEUL1,19,1 (initialization to low pressure) chamber 2 (high pressure) BSURF,2,..... COUPLE,1020,2,,,,,,,+ +,,,,,,,,,+ +,,22 MESH,22,BOX,,,,,,,+ +,0.8,0.8,0.8,2.,2.,2.,,,+ +,10,10,10,,,,EULER,600 PEULER1,600,,HYDRO,20 TICEUL1,20,20 (initialization to high pressure) Note that the porosity characteristics need to be defined for chamber 1 only. The gas automatically flows from chamber 2 into 1 and vice versa. Two different algorithms are available to calculate the mass transport through the coupling surface. The desired method can be activated from the PORFLOW entry or by choosing PORFCPL with SIZE=LARGE for velocity method and PORFCPL with SIZE=SMALL for pressure method. 1. Velocity method This algorithm is activated by:
PORFLOW, 42, , MATERIAL, 33, PRESSURE, 1.E5, METHOD, VELOCITY
The transport of mass through the porous area is based on the velocity of the gas in the Eulerian elements, relative to the moving coupling surface.
.
Eulerian Element SXint Face of the coupling surface that intersects the Eulerian element
v
Coupling Surface
Main Index
366 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
General Coupling
The volume of the Eulerian material transported through the faces of the coupling surface that intersect an Eulerian element is equal to
V trans = d t ⋅ α ⋅ ( v ⋅ A )
where
V trans
= transported volume during one time step ( V tra n s > 0 for outflow; V trans < 0 for inflow) = time step = porosity coefficient = velocity vector of the gas in the Eulerian mesh
dt α v A
= area of the face of the coupling surface that intersects the Eulerian element A is equal to the area of the face that lies inside the Eulerian element.
The transported mass through the porous area is equal to the density of the gas times the transported volume. 2. Pressure Method This algorithm is activated by:
PORFLOW, 42, , MATERIAL, 33, PRESSURE, 1.E5, METHOD, PRESSURE.
The transport of mass through the porous area is based on the pressure difference between the gas in the Eulerian element and the outside pressure. The outside pressure is the pressure as specified on the PORFLOW entry.
Eulerian Element SXint Face of the coupling surface that intersects the Eulerian element
v
Coupling Surface
The volume of the Eulerian material transported through the faces of the coupling surface that intersect an Eulerian element is equal to:
V trans = d t ⋅ α ⋅ A ⋅ p exh γ p exh 2 p   ⋅ ⎛  ⎞ – ⎛  ⎞ ⎝ p ⎠ ⎝ p ⎠ γ – 1ρ
2 γ+1 γ
Main Index
Chapter 9: Fluid Structure Interaction 367
Multiple Coupling Surfaces with Multiple Euler Domains
where
V trans
= transported volume during one time step ( V tra n s > 0 for outflow; V trans < 0 for inflow) = time step = porosity coefficient = area of the face of the coupling surface that intersects the Eulerian element A is equal to the area of the face that lies inside the Eulerian element = pressure of the gas in the Eulerian element = density of the gas in the Eulerian element = adiabatic exponent = pressure at the face
= Cp ⁄ C v
dt α A
p ρ γ p exh
The pressure at the face is approximated by the onedimensional isentropic expansion of the gas to the critical pressure or the environmental pressure according to
⎧ (p > pc ) ⎪ p p = ⎨ env env ⎪ p c ( p env < p c ) ⎩
where
pc
is the critical pressure:
γ 
2  γ–1 p c = p ⋅ ⎛  ⎞ ⎝ γ + 1⎠
In case the outside pressure is greater than the pressure of the gas, inflow through the coupling surface will occur. This porosity model can only be used for ideal gases; i.e., materials modeled with the gamma law equation of state (EOSGAM).
Multiple Coupling Surfaces with Multiple Euler Domains
Multiple coupling surfaces are available for HYDRO, MMHYDRO, and MMSTREN Euler Solvers in combination with the fast coupling algorithm. It is not available for the Strength Euler solver. The fast coupling algorithm is activated by the DYPARAM,FASTCOUP entry. Each couple surface is associated with an Eulerian domain. An Eulerian domain is a mesh that is aligned with the basic coordinate axes. You can define such a domain using the MESHID on the COUPLE entry. Through a surface that is shared by two coupling surfaces, mass can flow from one coupling surface to the other. Such a surface is called a hole. A hole can be either a porous subsurface of a coupling surface or be part of a coupling surface with interactive failure.
Main Index
368 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Multiple Coupling Surfaces with Multiple Euler Domains
If all Euler domains are defined by MESH, BOX, or MESH, ADAPT, then there are no restrictions on the use multiple coupling surfaces. In that case, a simulation may both contain porous holes that connect one Euler domain to another as well as coupling surfaces with interactive failure. With the multimaterial solver, Multiple Euler domains are only supported if all Euler domains are defined by either MESH,BOX or MESH,ADAPT.
Coupling Surface with Failure
A coupling surface is always associated with a Lagrangian structure. When the material model used in the Lagrangian structure supports failure (for example, by defining a failure model for the material), the faces in the coupling surface can fail when the underlying material in the structure fails. You can define the failure mode for the coupling surface by specifying “DYPARAM,FASTCOUP, ,FAIL”. When a Lagrangian element fails and the element is shared by two coupling surfaces, mass from one Eulerian domain flows to the other Eulerian domain through the hole. The interaction between these Eulerian domains is defined through a COUPINT entry. When you do not define the interaction between the Eulerian domains but the coupling surface fails, default ambient values for the state variables are used to compute the in or outflow through the hole in the surface. The ambient values of the variables can be defined on the COUP1FL entry.
Coupling Surfaces with Porous Holes
The porous hole is a surface that is shared by two coupling surfaces and connects the two coupling surfaces to each other. By selecting either the porosity model PORFLCPL or PORFCP, flow is enabled from the Euler domain in one coupling surface to the Euler domain in the other coupling surface. The model PORFCPL with SIZE=LARGE uses the velocity method and is for general use whereas PORFCPL with SIZE=SMALL uses the pressure method and is only for small holes. The porous hole can be either partially or fully porous and can be a subsurface of the whole coupling surface. A major application is the flow inside multicompartment air bags. To activate flow between two coupling surfaces through a porous hole the following steps have to taken:
• Associate an Euler domain to each of the two coupling surfaces • Make a subsurface of the elements of the coupling surfaces that models the hole. The elements in
this subsurface should be shared by both coupling surfaces.
• Define a LEAKAGE entry for one of the coupling surfaces. This LEAKAGE references a PORFCPL entry. • Create a PORCPL entry. The other coupling surface has to be referenced by this PORFCPL entry.
Main Index
Chapter 9: Fluid Structure Interaction 369
Multiple Coupling Surfaces with Multiple Euler Domains
Flow Between Domains
The facets in the coupling surfaces that represent an open area are subdivided into smaller facets, that each connect exactly to one Euler element in the first Euler domain and to exactly one Euler element in the second Euler domain. Material flow takes place across these smaller, subdivided facets which are called POLPACKs. This is the most accurate method. This method has the following limitations:
• Flow faces and wallets are not supported. Note: flowdef is supported • Viscosity is not supported • All Euler domains have to be created by the mesh entry. Euler domains consisting of a set of
Euler elements are not supported. Euler meshes should have at least one element overlapping at the hole. When the meshes are created dynamically using MESH,ADAPT, this is taken care of automatically. When mesh sizes are comparable for two Euler meshes that are connected by a hole some reduction in costs is achieved by choosing the same mesh size and the same reference point for the two Euler meshes. In general holes should not be precisely on Euler element faces.
Deactivation
Deactivation is only supported by the Roe solver. In case you are using the multiple coupling surfaces functionality, it is also possible to deactivate a coupling surface and the associated Eulerian domain at a certain time using the TDEAC field on the COUPLE entry. The deactivation stops the calculation of the coupling algorithm and its associated Eulerian domain. The analysis of the Lagrangian structure continues. Activation of the coupling surface is not possible. Initialization To initialize Euler elements several PEULER1 and TICEUL1 entries are used as follows: PEULER1,6,,HYDRO,19 TICEUL1,19,19 TICREG,1,19,SPHERE,1,3,5,1.0 SPHERE,1,,0.0,0.0,0.0,500.0 TICVAL,5,,SIE,400000.,DENSITY,0.2 MESH,22,BOX,,,,,,,+ +,0.01001,0.01001,0.01001,0.14002,0.12002,0.12002,25,OUTSIDE,+ +,14,12,12,,,,EULER,6 $ PEULER1,7,,HYDRO,20 TICEUL1,20,20 TICREG,2,20,SPHERE,2,3,6,2.0 SPHERE,2,,0.0,0.0,0.0,500.0 TICVAL,6,,SIE,400000.,DENSITY,1.9 MESH,23,BOX,,,,,,,+ +,0.11499,0.00501,0.00501,0.1302,0.11002,0.11002,50,OUTSIDE,+ +,26,22,22,,,,EULER,7
Main Index
370 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Multiple Coupling Surfaces with Multiple Euler Domains
The MESH entry references a unique property number that is also used by the PEULER1 entry. So the PEULER1 entry provides the link between the MESH entry and a TICEUL1 entry. In this way, each Euler domain references a unique TICEUL1 entry. The level indicators that occur on a TICEUL1 entry only apply to the Euler mesh that is linked to this TICEUL1 entry. To initialize all meshes to one initial state, only one property set is used in combination with only one
PEULER1 and TICEUL1 entry. Also, the PEULER entry may be used in this case.
Output
Euler archive output is restricted when this option facet has not been set. The restrictions are:
• The entry ELEMENTS on an Euler archive output request is required to be ALLEULHYDRO, ALLMULTIEULHYDRO or ALLMULTIEULSTREN. Specifying element numbers is not supported. • For each mesh, a separate Euler archive file is created. Thus, one Euler archive output request
gives multiple archive files. Each of these archive files may contain more than one cycle. To distinguish the Euler archive files from each other, the Euler archive files have a tag that specifies to what mesh they belong. This tag has the form: FV_(MeshID). Here FV is an acronym for “Finite Volume”.
• When one of the Euler meshes is of TYPE ADAPT, all Euler archives will contain only one cycle.
Example: A mesh with ID=22 and a mesh with ID=23 have been defined. Then this output request would generate the archives ALLEULER_FV22_0 and ALLEULER_FV23_0.
Using the Preference of Patran
For an example, refer to workshop 11. The preference supports simulations with multiple coupling surfaces; each using a different Euler mesh entry. For Interactive failure, the preference also supports simulations with multiple coupling surfaces; each using a set of Euler elements. Interactive failure is activated by the entry COUPINT. Creating multiple coupling surfaces with multiple coupling Euler domains with Patran is done as follows: 1. Create geometry for the coupling surfaces and mesh these geometries. Use LoadsBCs/Coupling or LoadsBCs/Airbag to define each coupling surface. The air bag option allows for porosity definitions. 2. Make an arbitrary solid for each Euler mesh using geometry/Create/Solid. This solid only serves to associate an Euler Mesh to a 3D property set. The geometry of the solid will not be used and any solid will suffice. 3. Create a material. 4. Create a 3D Eulerian property for each of the dummy solids of step 2 using Properties/Create/3D/EulerianSolid. In defining the property set select the material created in step 3. 5. Create the Euler meshes with LoadsBCs/Create/MeshGenerator. Use the coupling or air bag bc’s from step 1 and the 3D properties from step 4.
Main Index
Chapter 9: Fluid Structure Interaction 371
Fluid and Gas Solver for the Euler Equations
To define a porous hole between two closed surfaces, the surfaces should not be defined as coupling surfaces but as air bags. The first air bag is defined in one go using all the elements in the first surface. To define the second air bag, two subsurfaces will be created from the second surface. The first subsurface is the hole. The second subsurface consists of all the elements of the second surface that are not part of the hole. The second air bag is then the combination of these two subsurfaces. 1. Create the first air bag using all elements in the first surface with LoadsBCs/Create/Airbag/Surface. 2. Define a porous hole with Airbag/Subsurface and choose as porosity model PORFCPL. 3. Create a subsurface consisting of all elements of the second surface that are not part of the subsurface created in step 2. 4. Create the second air bag by selecting the two subsurfaces in step 2 and 3.
Fluid and Gas Solver for the Euler Equations
For gases and fluids flow, a stateoftheart Eulerian solver is available that is based on the ideas of Professor Philip Roe. The fluid and gas Euler solver is based on the solution of socalled Riemann problems at the faces of the finitevolume elements. The mathematical procedure amounts to a decomposition of the problem into a discrete wave propagation problem. By including the physics of the local Riemann solution at the faces, a qualitatively better and physically sounder solution is obtained. The fluid and gas solver is also known as an approximate Riemann solver. The solver can be either first or second order accurate in space in the internal flow field. Second order spatial accuracy is obtained by applying a socalled MUSCL scheme in combination with a nonlinear limiter function. The MUSCL approach guarantees that no spurious oscillations near strong discontinuities in the flow field will occur. The scheme is total variation diminishing (TVD), meaning it does not produce new minima or maxima in the solution field. The original Roe solver can be activated by using the DYPARAM,LIMITER,ROE entry in the input file. Improvements have been made to arrive at a full second order scheme in the fluid and gas solver to further gain accuracy in the solution. All boundary conditions –that is, the flow boundary conditions, and the wall boundary conditions, are fully second order accurate in space. You can use the new and improved solver (either first or secondorder) by entering the keyword 2ndOrder or 1stOrder on the PEULER or PEULER1 entry. Furthermore, a socalled “entropy fix” has been added to avoid the sharp discontinuities in those areas where the eigenvalues of the local Riemann problem vanish. In effect, the entropy fix ensures that an expansion shock (although mathematically sound) is broken down into a correct expansion fan. The expansion shock would yield a physical impossibility of decreasing entropy in the system. That is the reason for the name “entropy fix”. The entropy fix brings a very, almost unnoticeable, form of locally necessary dissipation into the solution to improve the differentiability of the equations where the eigenvalues vanish. The time integration in the fluid and gas solver is performed by a multistage time integrator, also know as a RungeKutta type scheme. Higherorder temporal accuracy can be achieved by applying multiple stages in the time integration. The required number of stages is automatically selected when you select
Main Index
372 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Modeling Fluid Filled Containers
either a first order or a second order solution. A first order spatial accurate solution uses a onestage time integration scheme; a second order spatial accurate solution applies a threestage time integration scheme. Multiple coupling surfaces with failure can be requested if the fast coupling algorithm is used by setting the DYPARAM,FASTCOUP, ,FAIL entry. You can also define interaction between multiple coupling surfaces, like in cases where you wish to model “chambers” that after structural failure will show a “connection” between them through which fluid or gas could flow. A typical example is an explosion in the cargo space of an aircraft after which the floor may be ruptured and the highpressure gas can vent into the passenger cabin. The fluid solver allows you to introduce viscosity into the solution. You use Tait’s equation of state to model the fluid with additional viscosity terms. There are some limitations in the current implementation. Eulerian elements must be completely filled with materials, so void or partial void elements are not allowed. For fluid flows where voids sometimes may occur, we recommend that you use Tait’s equation of state. This equation of state is fully supported by the improved second order Euler solver, and allows you to define a socalled critical density. When the fluid’s density falls below the critical value, the fluid cavitates (i.e., the pressure retains the value associated with the critical density). The density can further drop, but the pressure remains constant. In this fashion, you avoid the creation of voids and still allow the fluid to cavitate. When you only have data available for the fluid that satisfies the simple bulk equation of state, you can still use Tait’s model by setting the A 0 parameter to zero, γ to one, and add the critical density value at which cavitation occurs. Especially for blast wave types of problems, the full second order solution is recommended because of the accuracy it inherently brings. The JWL equation of state is not supported. A major advantage of using the blast wave approach is the speed at which the analysis can be performed. Especially spherical wave propagation through a Cartesian mesh is much more accurate in a full second order solution than in first order, or second order with (cheaper) first order accurate boundary conditions.
Modeling Fluid Filled Containers
Containers, for example plastic bottles, are often subjected to axial loading. Axial loading occurs when the bottle is crushed, or for example, stacked. It may concern both empty and (partially) filled bottles. Fluid filled containers or bottles can be modeled using a full multimaterial Euler description. However, using full multimaterial Euler fluid dynamics solver is a quite expensive method to solve the quasistatic behavior of the fluid. An alternative way of modeling is available through the FFCONTR (Fluid Filled CONTaineR) option. Using this option removes the need for a full fluid dynamics solution. The FFCONTR option uses the uniform pressure algorithms to calculate the pressure increase due to the compression of the container. The pressure is uniformly distributed but may change in time due to volume changes that occur when the container deforms. You need to define a surface to indicate the boundary of the container. The volume enclosed by the surface then equals the volume of the container.
Main Index
Chapter 9: Fluid Structure Interaction 373
Hotfilling
The normals of the faces of the surface must point outwards in order to compute the correct (positive) volume. When the normals point inwards, they are automatically reversed such that the resulting volume is positive. The surface must be closed to ensure a correct volume calculation. You must define the amount of fluid in the container or bottle. The volume of the gas above the fluid then follows immediately from the difference between the volume of the container and the fluid volume in the container. Obviously, the fluid volume cannot exceed the volume enclosed by the surface. The fluid is assumed to be incompressible. Thus, any volume change directly translates to a volume change of the gas above the fluid. The gas above the fluid is assumed to behave as an ideal gas under isothermal conditions:
p⋅V = C
where
p
is the pressure,
V
is the volume, and
C
represents a constant.
To define the constant C , you have to specify the initial pressure of the gas above the fluid on the FFCONTR entry. The initial pressure is only used for the calculation of the pressure changes and does not have an effect on other boundary conditions that you may have applied. If an overpressure (for example, a carbonated soft drink) or an underpressure (for example, a hot filled container) is present, you must model this separately using a PLOAD definition. The values for the pressure on the PLOAD entry and the pressure generated by the FFCONTR are superimposed for the calculation.
Hotfilling
Filling bottles with hot liquid can cause large deformations during cooling. To simulate these deformations the Fluid filled functionality container option can be used. Since MD Nastran Nonlinear Explicit (SOL 700) has only a limited cooling functionality the temperature of the fluid has to be specified by the user. In addition, the volume of the fluid will depend on temperature. A temperature versus time table and water density versus temperature table are input options for the fluid filled container. If these are set the gas is no longer isothermal but satisfies
PV ⁄ T = C
Here V is the volume of the gas. This volume is computed as the difference between the total volume and the volume of fluid. The volume of the fluid is given by
M water V =  , ρ
Here the fluid density
ρ
depends on the temperature as specified by table entry.
Main Index
374 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Hotfilling
Main Index
Chapter 10: Eulerian Solvers MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
10
Eulerian Solvers
The Standard Euler Solver 376 386
Approximate Riemann Euler Solver
Main Index
376 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
The Standard Euler Solver
The Standard Euler Solver
In the Eulerian approach material is not attached to elements but can move from one Euler element to the other. Mass, momentum and energy are element averages and are defined in the centers of the elements. This property is called cellcentered. The equations solved are the conservation laws of mass, momentum and energy as given in equation (101). Here, ρ is the material density, u i are the velocity components, p is the pressure, q is the bulk viscosity, g is gravity and e is the specific total energy. V is a volume and A is its boundary.
d  ∫ ρ dV + ∫ ρ ( u ⋅ n ) dA = 0 dt
V A
d  ∫ ρ u i dV + ∫ ρ u i ( u ⋅ n ) dA = – ∫ ( p + q ) n i dA – ρ g e 3 V dt
V A A
(101)
d  ∫ ρ e dV + ∫ ρ e ( u ⋅ n ) dA = – ∫ u i pn i dA dt
V A A
For Eulerian materials with strength the pressure p is replaced by the stress tensor. The volume integrals represent the total mass, momentum and energy in the Volume V. The surface integrals on the left signify transport out of the volume through parts of the area A. The surface integrals on the right represent the momentum and energy increase caused by forces acting on the boundary of the volume. The numerical scheme is a finite volume method. It is obtained by applying equation (101) to the material inside an Euler element and by specifying how transport terms are computed. The first equation signifies that the decrease of mass in an element equals the loss of mass trough the element boundary. In transporting mass between elements mass should be conserved globally. This is achieved by looping across the element interfaces and adding the transported mass, momentum and energy to the acceptor element and subtracting it from the donor element. In this way the finite volume scheme conserves mass, momentum and energy. In applying equation (101), it is assumed that density, velocity, and specific energy are constant across an Euler element and only depend on time. In addition, they are constant within one time step. This is consistent with a firstorder approach. The evolved time at cycle n will be denoted by t n . Element density, velocity, and specific total energy inside an element at cycle n will be denoted by ρ n , u n , and e n , respectively.
Main Index
Chapter 10: Eulerian Solvers 377
The Standard Euler Solver
Mn + 1 – Mn = –
∫ ρ ( u ⋅ n ) dA Δt
A tn
Pn + 1 – Pn = –
∫ ρu i ( u ⋅ n ) dA
A tn
Δt –
∫ p ( ρ, s ) n i dA + ρ g e3 V
A tn + 1
Δt
(102)
En + 1 – En = –
∫ ρe ( u ⋅ n ) dA + ∫ ui p ( ρ, e ) ni dA
A A 3
En + 1 Mn + 1 1 ρ n + 1 =  ,e n + 1 =  ,s n + 1 = e n + 1 – 2 Vn + 1 Mn + 1
∑ uk
k=1
2
Applying firstorder time integration gives equation (102). The integration is from denotes the mass inside the element, P momentum, and E energy.
tn
to
tn + 1 .
Here,
M
For the surface integrals that represent transport, the forward Euler method is used. Consequently, surface integrals are evaluated at the beginning of the time step. The surface integral with the pressure terms is evaluated using the new density and specific total energy. The transport velocity u ⋅ n depends on both the donor as well as the acceptor element and is given by the average of the donor and acceptor velocity. Multiplying the transport velocity with surface area and time step yields the transport volume. This volume is filled up with mass of the donor element. Multiplying this volume with the density of the donor element gives the transported mass. Likewise the transport volume times the donor velocity gives the transported momentum. Since the fluidstructure interaction forms an integral part of the numerical scheme we first discuss fluidstructure interaction.
Fluidstructure Interaction
Material in an Euler mesh can interact with Lagrangian structures. Eulerian material can exert forces on a structure causing displacement and deformation. On the other hand structures provide a barrier to Eulerian material. That is Eulerian material cannot penetrate the structure and the structural surface determines which Euler element have the capacity to hold mass. Consider for example a tank shell surface. Euler Elements that are outside the tank surface cannot hold material and only elements that are partially or completely inside the surface have the capacity to contain mass. This surface defines the effective boundary of the Euler domain and will be called the coupling surface. In most cases the coupling surface will consist of Lagrangian Elements. But also the interaction of Eulerian material with a Lagrangian solid is possible. Then the coupling surface consists of surface elements that have no Lagrangian model attached but only serve to enable interaction. In the following we shall assume that the coupling surface is a Lagrangian shell surface. The coupling surface will also be referred to as the structural surface.
Main Index
378 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
The Standard Euler Solver
The coupling surface consists of shell elements that deform under pressure loads from material inside the Eulerian domain. An explicit finite element solver solves the shell dynamics, and an explicit Euler solver solves the fluid dynamics for the inside region of the coupling surface. The interaction between these two solvers takes place in two ways.
• The mass in the Euler elements exerts a pressure load on the lagrangian elements associated with
the structural surface. These loads constitute an additional set of boundary conditions for the finite element solver, resulting in new grid point accelerations and velocities for the structure. From the updated plastic strain or updated stresses of the shell elements it is determined which elements are failing. Finally the structural grid points are moved using the new velocities
• The structural grid points move and so the Euler mesh has a new effective boundary.
Consequently, the volume of mass in each element may change. Since density is mass divided by the volume of the mass, densities will also change, and so will the pressures. In the following we shall assume that the material is insides the coupling surface.
The Numerical Scheme
The Euler elements are integrated in time by applying a finite volume method directly to the physical domain, avoiding the use of coordinate transformations. Therefore the finite volume method is applied to the 3D object that consists of that part of the Euler element that is inside the coupling surface. This is in general not a cube but a multifaceted object. For the 2D case this is sketched in Figure 101.
Figure 101
The Boundary of an Euler Element
In this figure the square represents an Euler element that is intersected by the coupling surface. Only that part of the square that is inside the coupling surface can contain mass. Therefore this part is the effective volume of material in the element. The boundary of this effective volume consists of two types of surfaces:
• Euler element boundaries that connect two neighboring elements called ‘Euler faces’.
Main Index
Chapter 10: Eulerian Solvers 379
The Standard Euler Solver
• Parts of the coupling surface that are within the Euler element. They will be called ‘polpacks’,
which is short for “polyhedron packets”. The effective boundary of an Euler element consists of Euler faces and polpacks. A polpack is the intersection of a coupling surface shell element with an Euler element and is completely inside an Euler element and completely inside a coupling surface shell element. An algorithm is available that computes these polpacks for any given, closed 3D faceted surface, and any 3D Euler domain. Faces refer to two Euler elements, whereas polpacks refer to only one Euler element. For both faces and polpacks, areas and normals are computed. The finite volume method results from applying equation (102) to these 3D objects. The volume V n + 1 in equation (102) is the effective volume of the Euler element. Furthermore, the surface integrals are computed by summing over the faces and polpacks. A contribution of a polpack or a face to integrals signifying transport is called a flux. When there is more than one material present in the simulation the mass conservation law applies to each material separately. This means that for each material inside an Euler element the density has to be monitored. In applying the momentum law, it does not matter whether there are several materials since all materials inside an Element are assumed to have the same velocity. The energy equation is also applied to each material separately. First consider simulations with only one material present. In the mass conservation law, the mass flux across a face gives:
Δ M DONOR = – ρ DONOR V ⋅ A Δ t Δ M ACCEPTOR = +ρ DONOR V ⋅ A Δ t
Here M is the mass in the Euler element, V is the velocity vector, A denotes the area vector of the face, Δ t is the time step, DO NOR denotes the element supplying mass, and A C CE P TOR denotes the element receiving mass. In most cases, the coupling surface is not permeable and there is no transport across the polpacks. However, in case coupling surface shell elements have a porosity model assigned, the flux equations take that into account. The momentum in an element can increase by either transport of momentum or by a pressure load working on the polpacks and faces. The pressure load contribution to this momentum increase is the surface integral – ∫ pn i dA Δ t . The force contribution of a face to the momentum increase of the element left
A
to the face and right to the face reads:
Δ P Left = – p F ace A Δ t Δ P Right = p Face A Δ t
Main Index
380 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
The Standard Euler Solver
Here
P
is the momentum of an element,
p f ace
A
is the area vector pointing from the left element to the right
element and is a weighted average of the pressure in the two elements that are on the left and right of the face. These momentum updates clearly conserve the combined momentum of the left and right element. For polpacks, the contribution is likewise, but now the pressure at the polpack is given by the pressure in the Euler element that contains the polpack. To conserve momentum, the negative of this momentum contribution is put as a force on the coupling surface shell element that hosts the polpack. This is the way boundary conditions are imposed on the Lagrangian element constituting the coupling surface. In a similar way the energy equation is applied. The procedure for advancing the Euler domain with one time step is as follows: Do all finite element objects and contact. Move the finite element objects in accordance to their grid point velocities. Using the new position of the finite element structures compute new polpacks. Using polpacks and faces compute for all Euler elements the volume of the portion that is inside the coupling surface. Transport mass, momentum, and energy across all faces and permeable polpacks using the conservation laws. The flux velocity is the average of the left and right Euler element velocity. In case no right Euler element is available the flux velocity is determined from an inflow condition and in some cases the velocity of the Euler element. Examples are holes and parts of surfaces that enable flow into the inside region of the coupling surface as a means of filling the inside region of the coupling surface. At the end of this step, element masses are fully updated. For each Euler element, compute density from the new mass and volume and compute pressure from the equation of state using the new density. Compute the effect of Euler element pressures to both structure as well as other Euler elements by going over respectively polpacks and Euler faces. This effect will contribute to the Euler element momentum. The transport contribution to the momentum increase has already been computed in step 3. At the end of this step the element momentum and energy are fully updated. Advance the Lagrangian shell elements associated with the coupling surface with one time step using the internal shell element forces, contact forces, and external forces from the Euler domain and compute new velocities on the grid points. Compute a new stable time step based on the mesh size, speed of sound and velocity. The stability criterion used is the CFL condition and applies to both the tank surface as well as to the Euler elements.
The Time Step Criterion
To maintain stability of the explicit scheme the time step should not exceed:
Δx Δ t max = u+c
(103)
Main Index
Chapter 10: Eulerian Solvers 381
The Standard Euler Solver
Euler with Strength
Deviatoric stress is a property of mass and is transported along with mass. Deviatoric stress in an element changes because masses with different stresses can enter the element and because strain increments raise stresses. When moving along with a piece of material the change in deviatoric stress denoted by s is given by:
de i j ∂u i ∂u j 1 ∂u k ds ij  = 2μ  = μ ⎛  +  –   δ ij⎞  ⎝ ∂x i ∂x j 3 ∂x k ⎠ dt dt
dev
(104)
Here the derivative is along the path of the moving mass and ( u k ) denotes the velocity in the Euler element. Since the velocity of the moving mass equals the velocity in the Euler element the total derivative is given by
ds i j ∂s i ∂s i ∂s i ∂x k ∂s ij  = j + j  =  + u k j  dt ∂t ∂x k ∂t ∂t ∂x k
(105)
Therefore the change of deviatoric stress in an Euler element is given by
de ij ∂s ij ∂s ij  + u k  = 2μ ∂t ∂x k dt
dev
(106)
This equation is not in conservation form and using the equation as it is would require the additional computation of shear stress gradients. By putting the equation in conservation, gradient computations are not needed and the equation be solved using the divergence theorem. To enable further use for other quantities like plastic strain consider
∂φ∂φ  + u k  = D ∂ xk ∂t
(107)
By using the continuity equation it can be written in conservative form.
∂ ( ρu k ) ∂ ( ρ uk φ ) ∂φ ∂ ( ρφ )  = ∂ρ φ + ρ ∂ φ = –  φ + ρ ⎛ D – u k  ⎞ = –  + ρ D ⎝ ∂x k ∂x k ∂x k⎠ ∂t ∂t ∂t
(108)
This gives
) ∂ ( ρ φ ) ∂ ( ρ φu k  +  = ρD ∂x k ∂t ) 1 ) ∂ ( ρφ u k  ⎛ ∂ ( ρ φ  +  ⎞ = D  ∂x k ⎠ ρ ⎝ ∂t
(109)
For stresses this becomes:
1 ∂ ( ρ s i j ) ∂ ( ρ u k s ij ) dev  ⎛  + ⎞ = 2μ e ij ∂x k ⎠ ρ ⎝ ∂t
(1010)
Main Index
382 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
The Standard Euler Solver
In this way, transport of stresses can be computed in close analogy to mass and momentum by transporting mass times shear stress. In the same way transport of plastic strain is carried out. Strain rates are computed from velocity gradients. They are obtained by use of the divergence theorem as follows:
∂ ui V  = ∂x j
∫ j dV ∂x
∂u i
=
∫ div ( ej u i ) dV
=
∫ ui e j dS
=
∑
fa ces
ui
Face F ace Sj
(1011)
Pressure can be either computed from density or updated from volume strain rates. The first corresponds to splitting the stress tensor computation into a hydrostatic part and a deviatoric part. The second computes the stress tensor without any splitting and uses the isotropic Hooke’s law in terms of strain rates. We show that the two approaches are equivalent. Consider computing pressure from the volume strain rates. Differentiation of the isotropic Hooke’s law gives
∂u i dp  = – K dt ∂x i
(1012)
with
K
the bulk modulus. Using the continuity equation in the form (1013)
∂u dρ ∂ρ ∂ρ  =  + u i  = – ρ i dt ∂x i ∂x i ∂t
yields
dp  = K dρ  dt ρ dt dp d =  ( K ln ρ ) dt dt
(1014)
The pressure in an element can be traced back by using equation (1014).
ρ
p =
∫
ρ ref
ρ – ρ ref ρ – ρ ref ρ d ( K ln ρ ) = K ln ⎛  ⎞ = K ln ⎛ 1 +  ⎞ ≈ K ⎛ ⎞ ⎝ ρ ref⎠ ⎝ ⎝ ρ ref ⎠ ρ ref ⎠
(1015)
So to firstorder in density, Hooke’s law and the equation of state give the same pressure. Basing the pressure on the logarithm of the density ratio is expensive and the linearization is sufficiently accurate. Pressures are computed using the linearization. To account for rotation of material the Jaumann correction is applied.
The Multimaterial Solver
In simulations with multiple materials, it is important to keep track of the interfaces between materials. For example, in fuel tank sloshing simulations, there is an interface visible between regions filled with fuel and regions filled with air. To handle these interfaces, several extensions of the transport logic and pressure computation are necessary. The extended transport logic is known as preferential transport and
Main Index
Chapter 10: Eulerian Solvers 383
The Standard Euler Solver
tries to maintain interfaces between materials. Consider, for example, the case of a blast wave of air in water. It is important that the interface between water and air during the expansion of the blast wave is maintained and does not deteriorate by the unphysical mixing of water and air. To enable a multimaterial simulation, a certain amount of bookkeeping is needed. For every Euler element, the following information is available:
• The number of materials inside the Euler elements • For each material, the volume fraction, the material ID, the density, the mass, the specific energy,
the total energy, and volume strain rates are stored. The volume fraction of a material is defined as the fraction of Element volume that is filled with that material. The transport logic for multimaterial amounts to:
• Compute the volume to be transported. This is V ⋅ A Δ t and this volume flux gives rise to a mass
flux. If there is only one material, the mass flux is the density times the volume flux, but now the donor element has several materials and each material has a distinct density and, therefore, the mass flux is split into several mass fluxes. Each material in the donor element has a distinct mass flux and this material specific mass flux is easily converted into a volume flux by using the material density. Using this conversion, the mass fluxes should give rise to volume fluxes that add up to a total volume flux that equals V ⋅ A Δ t . Materials are transported out of the element until the prescribed total volume flux is reached. The only remaining issue is which materials are transported first.
• Determine for both donor element as well as acceptor element which materials are present in
the element.
• Look which materials are common to both elements. • First transport any material that is common to both elements. Transport these common materials
in proportion to their acceptor material fraction. A material is transported with the material density of the donor element and this material density translates a volume flux into a mass flux and vice versa. Subtract any mass that is transported from the flux volume. If there is sufficient mass of the common materials in the donor element the whole flux volume will be used to transport the common materials.
• If after transport of the common materials the flux volume is not fully used yet, transport
materials in ratio to their donor material fraction. To illustrate how this procedure aims at preserving material interfaces, consider two adjacent Euler elements and assume that flow is from the left element to the right element. The left element is filled with fuel and air and the right one is filed with only air. Since air is the only material common to both elements, it is transported first. If there is sufficient air, only air will be transported. If during transport there is no air left, transport of this common material is not able to use the full flux volume and fuel has to also be transported. In both cases, the interface between fuel and air is maintained. The pressure computation for Euler elements with only one material is straightforward: the pressure readily follows from the equation of state and the density. For elements with more than one material, each material has a distinct equation of state and a distinct density and this results in a distinct pressure for each material. The pressure computation for these elements is based on the thermodynamic principle of
Main Index
384 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
The Standard Euler Solver
pressure equilibrium. Since masses of materials in Euler elements are only changed by the transport computation, these masses are fixed during the pressure computation. The volume taken up by each material in an Euler element is not known but determines the pressure inside the material. By adjusting the volumes of the materials simultaneously, pressure equilibrium is achieved. Therefore, the pressure computation amounts to an iterative process that iterates on the volumes of the materials inside the Euler element. To understand the influence of the material volumes consider an element with fuel and air. Suppose that at the start of a cycle, there is pressure equilibrium and that during transport, air enters the element. Because of the surplus of air there is no longer pressure equilibrium. Physically, it is expected that the air very slightly compresses the fuel until pressure equilibrium is achieved. The compression of the air is just the adjustment of the material volumes of fuel and air. The material volume of air increases while the material volume of fuel decreases. Viscosity Viscous stresses only contribute to the momentum balance:
d ∫ ρ u i dV + ∫ ρ u i ( u ⋅ n ) dA = – ∫ pn i dA + ∫ s ij n j dA dt
V A A A
(1016)
Here, the deviator shear stress tensor
1 s i j = μ ⎛ e i j –  e kk⎞ ⎝ 3 ⎠ 1 ∂u i ∂u i e ij =  ⎛  + ⎞ 2 ⎝ ∂x j ∂x j⎠
si j
is given by
(1017)
The contribution of viscous dissipation to the energy balance law is small and is not taken into account. Velocity gradients are computed by Gauss’s law as given by equation (1018). For boundary contributions, the imposed velocity boundary condition is used. Material in one element exerts a viscous force on material in the adjacent elements and leads to changes in momentum. The momentum transferred across an Euler element face is given by
∫ s i j n j dA dt
∂u i ∂u n 2 = ⎛ –  μu k, k n i + μ  + μ ⎞ F AC E A R EA * Δ t ⎝ 3 ∂n ∂x i ⎠
(1018)
The second term is most significant and requires a special treatment. This term is proportional to the normal velocity derivative at the face. It can be obtained by averaging the normal derivative over the left and right Euler element:
∂u i ∂n 1 ∂ui =  ⎛ 2⎝ ∂n ∂u i +  ⎞ ∂n R⎠
Face
L
(1019)
This leads to decoupling. To show this, consider an Euler element and two of its opposing faces. For both opposing faces, a viscous flux that is proportional to equation (1018) is added to the Euler element. The net contribution is proportional to the difference of equation (1018) for the two faces. In this subtraction, the normal velocity derivative of the element itself drops out. As a result, the contribution of viscous
Main Index
Chapter 10: Eulerian Solvers 385
The Standard Euler Solver
fluxes to the momentum increase of the Euler elements is only weakly coupled to the velocity gradient in the Euler element. In practice, decoupling follows. To avoid this decoupling, the gradient is computed directly using the velocity difference between across the face, giving:
∂u i ∂n u i – ui = Δx
R L
Face
(1020)
Furthermore, walls exert viscous stress on material and, at the wall, a noslip condition is applied. Computing these in a specific local system enables a straightforward use of the noslip condition. In this local system, the xaxis is along the normal of the boundary. The noslip condition ensures that all tangential velocity derivatives are zero. In addition, normal derivatives are computed directly from the velocity difference between element and wall. This leads to shear stresses at the wall:
4 u wall – u el ement 4 ∂u l oc s xx =  μ  =  μ 3 x wall – x elemen t 3 ∂x v wall – v element ∂v l oc s xy = μ  = μ ∂x x wall – x element s zx
l oc
(1021)
∂w – = μ  = μ ∂x x wall – x element w wall w el ement
These shear stresses are added to the momentum balance. Fluidstructure Interaction with Interactive Failure Consider a box filled with gas. If a blast wave is initiated inside the box, some parts of the box may fail and gas can escape through ruptures. To simulate this flow, the gas inside the box is modeled by an Euler domain and the box surface by shell elements. These shell elements form the coupling surface for this Euler domain. Once shell elements of this box have failed, gas flows from the inner domain to an outer Euler domain that models the ambient. The shell surface also forms the coupling surface for this outer Euler domain and is, therefore, able to connect Euler elements inside the shell surface to elements outside the shell surface. Flow from one Euler domain to another is also possible through fully or partly porous segments.
Figure 102
The Overlapping Mesh
Main Index
386 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
Approximate Riemann Euler Solver
Flow Between Domains Flow from one Euler domain to another takes place through either shell elements that are porous or that have failed. Flow through a segment can only take place if it is inside both Euler domains. In the following, let us assume that both Euler domains are sufficiently large. In general, a segment can intersect several Euler elements of the first Euler domain and the same applies to the second domain. Therefore, the segment connects several Euler elements in the first Euler domain to several Euler elements in the second domain. For an accurate and straightforward computation of flow through the segment, it is partitioned into sub segments such that each sub segment is in exactly one Euler element of each Euler domain. To carry out this partitioning, an overlapping Euler mesh is created that is the union of both Euler meshes. Then, for each element in this overlapping domain, the intersection with the segment is determined. Each intersection gives one sub segment that refers to both the original segment and to the element in the overlap domain. Since an element in the overlap domain is in exactly one element of both domains, the sub segment connects exactly one element in the first Euler domain to exactly one element in the second domain. Transport across this sub segment is straightforward because it closely resembles transport that takes place between the Euler elements that are within an Euler domain. In computing transport across the sub segment, the velocity of the segment has to be taken into account. If the segment is moving with the same velocity as the material on either side, no material will flow through the segment.
Approximate Riemann Euler Solver
Euler Equations of Motion
The analysis of the physical behavior of fluids and gases is best solved using a Eulerian approach. The nature of the behavior of these types of materials is represented in a natural way using a finite volume description based on the Euler equations of motion. An accurate solver is available that allows you to analyze the behavior of fluids and gases, coupled to structures if necessary. The solution approach is based on a socalled Riemann solution at the element faces that defines the fluxes of mass, momentum and energy, the conserved problem quantities. This section gives a more detailed explanation of the theory behind the Riemannbased Euler solver, its boundary condition treatment, and accuracy in time and space. The inviscid flow of a fluid or a gas is fully governed by the Euler equations of motion. We will use the equations in their conservative form:
∂q ∂f ( q ) ∂g ( q ) ∂h ( q )  +  +  +  = 0 ∂x ∂y ∂z ∂t
(1022)
f ( q ), g ( q ) ,
where q is the state vector and They are defined as follows:
and
h(q)
represent the fluxes of the conserved state variables.
Main Index
Chapter 10: Eulerian Solvers 387
Approximate Riemann Euler Solver
⎛ ⎜ ⎜ q = ⎜ ⎜ ⎜ ⎜ ⎝
ρ ρu ρv ρw E
⎞ ⎟ ⎟ ⎟ ⎟ ⎟ ⎟ ⎠
⎛ ρu ⎜ ⎜ ρ u2 + p f( q) = ⎜ ρ uv ⎜ ⎜ ρuw ⎜ ⎝ ( E + p )u
⎞ ⎟ ⎟ ⎟ ⎟ ⎟ ⎟ ⎠
⎛ ρv ⎜ ρuv ⎜ g ( q ) = ⎜ ρ v2 + p ⎜ ⎜ ρv w ⎜ ⎝ ( E + p )v
⎞ ⎟ ⎟ ⎟ ⎟ ⎟ ⎟ ⎠
⎛ ρw ⎜ ρuw ⎜ h( q) = ⎜ ρ vw ⎜ ⎜ ρ w2 + p ⎜ ⎝ ( E + p )w
⎞ ⎟ ⎟ ⎟ ⎟ ⎟ ⎟ ⎠
(1023)
Equation (1022) describes the conservation of mass, momentum and energy. In equation (1023), ρ is the material density, u, v, w the velocity components, p the pressure, and E the total energy. For a gas, we can close the system (note that we have five equations, with six unknowns) by adding the equation of state for a calorically perfect gas (the “gamma law equation of state in MD Nastran Nonlinear Explicit (SOL 700)).
p = (γ – 1) ⋅ ρ ⋅ e
(1024)
In equation (1024), e denotes the specific internal energy of the gas and γ is the ratio of specific heats. There exist more equations of state for gases, but most gases can be described as calorically perfect gases, in which case equation (1024) applies. For a fluid in its simplest form, we may use a socalled “simple bulk” equation of state:
ρ p = K ⎛  – 1⎞ ⎝ ρ0 ⎠
(1025)
In equation (1024), K is the material bulk modulus and ρ 0 is the reference density at which the material has no pressure. Also, for fluids, there are more equations of state, like a full polynomial or Tait’s equation of state. Both are implemented in the approximate Riemann solver that MD Nastran Nonlinear Explicit (SOL 700) uses, but the method of implementation is similar to the simple bulk equation of state and is not described in detail here.
Numerical Approach
The conservation laws as described by equation (1023) are numerically solved by an upwind, cellcentered finite volume method on unstructured 3D meshes. We will briefly describe the solution method here. When the conservation laws are written in integral form, by integrating over an arbitrary volume, the finite volume (discretized) method becomes apparent when we consider each element in an Eulerian mesh as a finite volume on which we have to solve the conservation laws as described by equation (1023). The integral form of equation (1023)) when using Gauss’ integral theorem:
∂ q dV + ∂t ∫
V
∂v
∫ ( f ( q ) ⋅ nx + g ( q ) ⋅ ny + h ( q ) ⋅ nz ) ⋅ d S
= 0
(1026)
From equation (1026), it becomes apparent that the fluxes of mass, momentum, and energy have to be integrated normal to the boundary of the volume or its surface. When we use the rotational invariance of the Euler equations of motion, the integral form can be rewritten using the transformation matrix that describes the transformation of the state variables in a direction normal to the surface:
Main Index
388 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
Approximate Riemann Euler Solver
∂ ˜  ∫ q dV + ∂t
V
∂V
˜ ∫ f ( q ) ⋅ dS
= 0
(1027)
˜ where q denotes the state vector transformed to a coordinate system with the local xaxis in the direction of the normal to the surface. When we then make the step to a discretized form, by defining the volume as the volume of a finite element (an element of the Euler mesh), and the surface defined by the faces spanning the element, equation (1027) becomes a local onedimensional system of equations for each face of the element with the local xaxis in the direction of the normal to the element’s face. Note that the fluxes in the local y and zdirection do not contribute to the change of the state variable. The system of equations to solve for each element face thus becomes: ˜ ˜ ∂ q ∂f ( q )  +  = 0 ˜ ∂t ∂x
(1028)
˜ where x defines the xdirection normal to the element’s face. Considering the fact that each face has a left and a right element connected to it, we can view the state variables in the left and right connected element as initial conditions for the solution of the flux normal to the face: ˜ ˜ ⎧ qL x < 0 ˜ ˜ q ( x ,0 ) = ⎨ ˜ ˜ ⎩ qR x > 0
(1029)
Equations (1028) and (1029) describe a socalled Riemann problem. Thus, the solution for the fluxes at the element faces amounts to solving a local 1D Riemann problem for each of the faces of the element, considering the left and right state of the fluid or the gas. The contribution of the face fluxes result in the state change in the element as a function of time as denoted by the first term in equation (1028). The fluxes on the faces are determined using a flux function, f R ( q L, q R ) by which, using equations (1027), (1028), and (1029), the discretization becomes:
dq i 1 = – Vi dt
6
∑
n =1
f R ( q L, q R ) ⋅ A n
(1030)
Vi
In equation (1030), element, and
An
i
denotes the element number,
the element volume,
n
the face numbers of the
the associated face area.
Using a flux difference scheme, the flux function can be written as:
1 1 f R ( q L, q R ) =  { f ( q L ) + f ( q R ) } –  { Δ f + – Δ f  } 2 2
(1031)
The flux difference terms in equation (1031) are defined as:
Δ f + = f R ( q L, q R ) – f ( q L ) Δ f  = f ( q R ) – f R ( q L, q R )
(1032)
Main Index
Chapter 10: Eulerian Solvers 389
Approximate Riemann Euler Solver
When we use the eigenvectors, the eigenvalues, and the wave strengths that can be found from a diagonalization of the Jacobian matrix of the Euler equations, we arrive at a simple definition of the flux function. Note that the shape of the eigenvectors, eigenvalues, and the wave strengths depend on the type of equation of state the flux function is constructed for. In general terms, the numerical flux function used in the scheme is defined as:
⎧ 5 ⎫ 1 1⎪ ˜ ˜ ⎪ ˜ f R ( q L, q R ) =  { f ( q L ) + f ( q R ) } –  ⎨ ∑ α i ⋅ λ i ⋅ R i ⎬ 2 2⎪ ⎪ ⎩i = 1 ⎭
(1033)
After some rewriting of equation (1033), we find:
1 ˜ ˜ ˜ ˜ ˜ ˜ ˜ ˜ ˜ ˜ ˜ f R ( q L, q R ) =  { f ( q L ) + f ( q R ) – [ u ⋅ Δ q + ( u – a – u ) ⋅ α 1 ⋅ R 1 + ( u + a – u ) ⋅ α 2 ⋅ R 2 ] } 2
(1034)
Using the ideal gas equation of state (gammalaw equation of state), we find for the wave strengths:
˜ ˜ Δp – ρ ⋅ aΔu ˜ α 1 = ˜ 2 a2 ˜ ˜ Δp + ρ ⋅ aΔu ˜ α 2 = ˜ 2 a2
(1035)
And for the associated eigenvectors:
⎛ ⎞ ˜ R1 = ⎝ 1 u – a v w H – u a ⎠ ˜ ˜ ˜ ˜ ˜ ˜˜
T
(1036)
⎛ ⎞ ˜ R2 = ⎝ 1 u + a v w H + u a ⎠ ˜ ˜ ˜ ˜ ˜ ˜˜
T
In the above equations, the quantities denoted by a “tilde” are weighted quantities according to:
˜ φ = θ ⋅ φL + ( 1 – θ) ⋅ φR ρL θ = L + ρR ρ
(1037)
All quantities are averaged using the above definition, except for the density:
˜ ρ = ρL ⋅ ρR
(1038)
The above described flux evaluation scheme is called an approximate Riemann scheme due to the fact a linearization using the weighted quantities at the element faces is applied. As a result, the scheme exhibits an artifact, namely that it does not satisfy the entropy inequality. The entropy inequality states that the entropy of a system can only remain constant or increase. Due to the artifact, the scheme is able to also capture mathematically sound, but physically impossible discontinuities like expansion shocks. This is easily “repaired” by adding a socalled entropy fix to the scheme as described in the next section.
Main Index
390 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
Approximate Riemann Euler Solver
Entropy Fix for the Flux Difference Riemann Scheme
As described in the earlier section, a socalled entropy fix must be added to the scheme in order to have the scheme correctly decompose a expansion discontinuity into a physically correct expansion fan. The entropy fix in fact amounts to adding some numerical viscosity or dissipation to sonic points, shocks and contact discontinuities. The dissipation is added only at those points where any one of the system’s eigenvalues vanishes. The entropy fix can be written in terms of a simple function.
⎧ z ⎪ ψ ( z) = ⎨ ( z2 + δ 2 ) 1 ⎪ 2 δ1 ⎩ a ≥ δ1 a < δ1
(1039)
The function works automatically on the eigenvalues of the system, represented in equation (1039) by z , and is governed by a single parameter δ 1 that depends on the flow field:
˜ ˜ ˜ v w δ 1 = δ ⋅ ⎛ u +  +  ⎞ ⎝ ˜ ˜ ˜ a a a⎠
(1040)
Second Order Accuracy of the Scheme
When we consider the flux function as given in general by equation (1034), it does not say anything about the order of accuracy at which the face fluxes are computed. The accuracy is governed by the way in which the left and the right state variables are determined. A first order scheme results when the left and the right state variables are taken as the values the state variables have at the left and the right element center; a socalled first order extrapolation to the face. When we increase the stencil by which we determine the left and right state variable values at the face by including the leftleft and the rightright element, we arrive at a second order accurate scheme in space. A socalled nonlinear limiter that avoids the creation of new minimum or maximum values limits the second order left and right face values of the state variables. Such a scheme is called total variation diminishing, (TVD). Near sharp discontinuities the scheme reverts to locally first order as to introduce the necessary numerical viscosity to avoid oscillations in the solution near the discontinuity. The second order approximate Riemann solver in MD Nastran Nonlinear explicit (SOL 700) applies the Superbee limiter. The second order scheme can be formally written as:
1 q iL+ 1 ⁄ 2 = q i +  Ψ L ⋅ ( q i – q i – 1 ) 2
(1041)
for the left side of the face, with:
1 1 Ψ L =  ( 1 – κ ) ⋅ Φ ( r L ) + ( 1 + κ ) ⋅ r L ⋅ Φ ⎛ ⎞ ⎝ r L⎠ 2 qi + 1 – q i r L = qi – q i + 1
(1042)
Main Index
Chapter 10: Eulerian Solvers 391
Approximate Riemann Euler Solver
For the right side of the face, the second order approximation is defined by:
1 q iR+ 1 ⁄ 2 = q i + 1 –  Ψ R ⋅ ( q i + 2 – q i + 1 ) 2
(1043)
with:
1 1Ψ R =  ( 1 – κ ) ⋅ Φ ( r R ) + ( 1 + κ ) ⋅ r R ⋅ Φ ⎛  ⎞ ⎝ r R⎠ 2 rR qi + 1 – q i = q i + 2 – qi + 1 κ = –1
(1044)
The upwind scheme is defined for
Φ ( r ) = ma x [ 0 ,min ( 2 r ,1 ) ,min ( r ,2 ) ]
and the limiter function
Φ
is the Superbee limiter: (1045)
Time Integration
The set of equations is integrated in time using a multistage scheme. For the second order accurate solution, a threestage time integration scheme is used:
q(0) = qi
n
α1 ⋅ Δ t q ( 1 ) – q ( 0 ) = – ⎛  ⋅ F ( 0 )⎞ ⎝ Vi ⎠ α2 ⋅ Δ t q ( 2 ) – q ( 0 ) = – ⎛  ⋅ F ( 1 )⎞ ⎝ Vi ⎠ α3 ⋅ Δ t q ( 3 ) – q ( 0 ) = – ⎛  ⋅ F ( 2 )⎞ ⎝ Vi ⎠ q in + 1 = q ( 3 )
(1046)
In equation (1046),
q(k)
denotes the state variable value in the kth integration stage,
F(k)
αk
the stage
coefficients, and the flux contributions
6
are defined as (1047)
F(k) =
∑
n =1
f R ( q ( k ) ,q ( k ) ) ⋅ A n
L R
using the state variables at each stage of the integration. The final step gives the solution of the state variables at the new time level.
Main Index
392 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
Approximate Riemann Euler Solver
Main Index
Chapter 11: Airbags and Occupant Safety MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
11
Airbags and Occupant Safety
Introduction 394 395
Airbag Definition Seatbelts 405
Occupant Dummy Models Pre and Postprocessing
413 414
Main Index
394 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
Introduction
Introduction
The MD Nastran r2 release, SOL 700 includes a Fluid Structure Interaction (FSI) capability that is based on the advanced Finite Volume (Eulerian) and General Coupling Technology available in MSC.Dytran. The objective of fluidstructure interaction using the coupling algorithm is to enable the material modeled in Eulerian and Lagrangian meshes to interact. Initially, the two solvers are entirely separate. Lagrangian elements that lie within an Eulerian mesh do not affect the flow of the Eulerian material and no forces are transferred from the Eulerian material back to the Lagrangian structure. The coupling algorithm computes the interaction between the two sets of elements. It thus enables complex fluidstructure interaction problems to be analyzed. The first task in coupling the Eulerian and Lagrangian sections of a model is to create a surface on the Lagrangian structure. This surface is used to transfer the forces between the two solver domains. The surface acts as a boundary to the flow of material in the Eulerian mesh. At the same time, the stresses in the Eulerian elements cause forces to act on the coupling surface, distorting the Lagrangian elements. The method used for airbag simulation is full gas dynamics and is based on General Coupling with adaptive Euler. Unlike other techniques such as the ALE (Arbitrary Lagrange Euler) where the Eulerian mesh is fixed in space or GBAG method where the gas flow is modeled by applying a predetermined pressure profile to inflate the bag, in the General Coupling technique, the Eulerian mesh will “adapt” itself to the Lagrangian fabric model as the airbag is inflated. In other words, when the airbag is initially at the folded stage, there is a small Eulerian domain encapsulating the Lagrangian mesh. When the airbag is inflated, the Eulerian mesh expands as the gas jet flows through the airbag compartments and adapts itself to follow the airbag fabric. This technique is unique in MD Nastran r2 and is considered the most accurate method to predict the complex airbag behavior such as OutofPosition (OOP) simulation, as required by FMVSS 208, where the occupant is already leaning forward when the airbag is inflated. In addition, CFD deployment of multicompartmented airbags can easily be modeled with this technique by using multiple, fully automatic, adaptive Euler domains. The following capabilities are available:
• Analyze multiple compartments with the CFD approach. • Simulate flow from one CFD domain into another CFD domain. • The individual CFD domains are dynamic and adaptive. The user does not need to mesh the CFD
domains, nor does he have to worry about the size, since the CFD domains will automatically follow the deploying airbag compartments.
• Flow through both small and large holes is accurately calculated. • Inflator models
Main Index
Chapter 11: Airbags and Occupant Safety 395
Airbag Definition
(a) Shape of Side Curtain airbag at Start of Simulation
(b) Euler Meshes for all Six Regions
(c) Shape of Bag after 20 milliseconds
Figure 131
Multicompartment Side Curtain Airbag
Airbag Definition
The airbags are defined by using the AIRBAG entry in SOL 700. Airbags can be automatically inflated by defining a gas flow rate in Eulerian domain using full gas dynamics method. A second method, the conventional uniform pressure method, is also available but it is not as accurate as the full gas dynamics. All related airbag input definitions such as inflator and porosity models, environmental parameters, are grouped in entries that can be directly input in the AIRBAG entry. Every entity has certain parameters associated with it that has to be input immediately following the name of the entity. The order of which entity is entered first in the AIRBAG card is immaterial as long as the associated parameters are defined right after the name of the entity. The following entity groups are available: “CFD” Entries for this entity describe the properties of the Eulerian domain. Only one “CFD” section can be defined. If this section is not defined, the uniform pressure method is used.
“ENVIRONM” Entries for this entity describe the properties of the environmental conditions for the airbag. Only one “ENVIRONM” section can be defined. If not defined, the values for “INITIAL” will be used. “INITIAL” “INFLATOR” Entries for this entity describe the properties of the environmental conditions for the airbag. Only one “INITIAL” section can be defined and is required. Entries for this entity describe the properties of an inflator that is attached to the airbag. More than one inflator may be defined.
Main Index
396 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
Airbag Definition
“CGINFLTR” “SMALHOLE ”
Entries for this entity describe the properties of a cold gas inflator that is attached to the airbag. More than one inflator may be defined. Entries for this entity describe the properties of a small hole in the airbag. More than one SMALHOLE may be defined. A small hole should be used when the size of the hole is of the same order as the size of the elements of the Euler mesh.
“LARGHOLE” Entries for this entity describe the properties of a large hole in the airbag. More than one LARGHOLE may be defined. A large hole should be used when the size of the hole is of the larger than the size of the elements of the Euler mesh. “PERMEAB” “CONVECT” Entries for this entity describe the properties of the permeability of the airbag fabric. More than one PERMEAB may be defined. Entries for this entity describe the properties of the loss of energy of the gas in the airbag by means of convection through the airbag surface. More than one CONVECT may be defined. Entries for this entity describe the properties of the loss of energy of the gas in the airbag by means of radiation through the airbag surface. More than one RADIATE may be defined. When this option is used, the StephanBoltzmann constant must be defined by PARAM, SBOLTZ. Entries for this entity describe the properties of gases. More than 1 GAS may be defined and the gases defined in one AIRBAG entry can be referenced by other AIRBAG entries.
“RADIATE”
“GAS”
There are other entries in SOL 700 to define airbag properties. These are:
PARAM, UGASC PARAM, SBOLTZ PARAM, DYDEFAUL
defines a value for the universal gas constant. defines a value for the StephanBoltzmann constant controls the default setting of the simulation defines the hybrid inflator gas fraction Gamma Law Gas Equation of State grid point in airbag reference geometry
INFLFRC EOSGAM GRIA
Please see MD Nastran r2 QRG for more details.
Main Index
Chapter 11: Airbags and Occupant Safety 397
Airbag Definition
Inflator Models in Airbags
There are several methods available to define an inflator in airbag analyses. The most general inflator definitions are:
INFLTR INFLHB INFLCG
Standard inflator defined by mass flow rate and static temperature of a single inflowing gas. Hybrid inflator defined by mass flow rate and static temperature of multiple inflowing gasses. Coldgas inflator behaves like a reservoir filled with high pressure gas that flows into the airbag.
Figure 132
Airbag and Occupant Safety using SOL 700
For both the uniform pressure model and the full gas dynamics (CFD) method, the inflator location and area are defined by means of a subsurface created by a BSURF, BCPROP, BCMATL, or BCSEG parameters. These parameters are referenced by BFID field on the INFLTR entity. It can only reference shell elements that belong to the airbag surface, as defined by BFID. The characteristics of the inflator are specified on an INFLTR entity on AIRBAG card. This entry references tables for the mass flow rate and the temperature of the inflowing gas. A model can be defined containing both (CFD) and uniform pressure model for the airbag. These two options can be defined with identical inflator characteristics. When the CFD entity and its associated parameters are omitted from the AIRBAG card, the uniform pressure method will be used. In case the CFD entity is present, the airbag will use the Euler method from the start of the simulation. In case the value of SWITCH is nonzero, the airbag will switch from an Eulerian representation to a Uniform Pressure formulation. Hybrid Inflator Model The hybrid inflator supports the inflow of multiple gases through an inflator subsurface, as well as providing a type of thermally ideal gas for which the specific heat at constant pressure c p can be dependent on temperature. In addition, the properties of the gas contained in an air bag will be changed based on the gas composition and temperature. Updating of gas constants is available for use together with INFLTR, INFLHB, INFLTNK, and INFLCG inflator definitions, and with PORHOLE, PERMEAB, PERMGBG, PORFCPL, and PORFGBG porosity definitions.
Main Index
398 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
Airbag Definition
Ideal gas description
A thermally ideal gas is specified by the specific gas constant and the variation of specific heat at constant pressure with temperature. The specific gas constant for a gas is defined as:
R = R uni ⁄ M
where
R un i
is the universal gas constant and
M
the gas molar weight.
Using the specific heat as function of temperature, the specific internal energy of a gas as a function of temperature is found as:
T
e(T) =
∫
T ref
( c p ( T ) – R ) ⋅ dT + e ref
We can now define
e ( T ) = cv ( T ) ⋅ T
cv
so that:
Mixture of gas
A hybrid inflator is specifically meant to give an inflow of several gases with different properties. To account for the properties of these gases, it is necessary to keep track of the composition of a gas at a certain time, not only for the inflator but also for the gas mixture inside the air bag. For use with hybrid inflators, it is assumed that instantaneous mixing takes place. This means the gas composition are the same throughout the volume. For an inflator, gas fractions are given as user input. For gasbags and Eulerian, gas fractions are based on the total mass of each gas at a certain time. Gas fractions are defined as follows:
mass i ( t ) mfra c i ( t ) = m ass t ot ( t )
Properties of the gas mixture inside a surface are based upon the principle that a mixture of (thermally) ideal gases is itself an ideal gas. This yields for the properties of the mixture:
m
R* =
∑ mfrac i t ⋅ R i
i =1 m
* cv =
∑ mfrac i ( t ) ⋅ cvi ( T )
i =1
* * cp = cv + R* * cp γ * = * cv
Main Index
Chapter 11: Airbags and Occupant Safety 399
Airbag Definition
Here T is the temperature of the gas mixture. This may be the inflow temperature of the inflator, the temperature inside a constant pressure gasbag, or the average temperature of all Eulerian elements that are not covered by the coupling surface. The latter is found as:
∑ ecel l T Euler = cv
E ul er
Energy/work
A formulation for the change of energy in a closed volume can be found when c p and c v are a function of temperature and gas composition. This takes into account inflow (by hybrid inflators), outflow (by porosity), and energy loss (through convection and radiation). We know:
dQ = dU + pdV
where:
d Q = d q in – dq l o ss – d q out d U = m ⋅ de + e ⋅ dm
For inflow of a certain mixture, we find:
dq i n · *  = [ M i n ( t ) ⋅ c p ( T i n ) ] ⋅ T i n dt in
Similarly, for outflow:
dq out · *  = [ M out ( t ) ⋅ c p ( T out ) ] ⋅ T out dt out
For the work done by the gas mixture, the following holds:
· dV V * * p  = M ⋅ ( c p – c v ) ⋅ T ⋅ dt V
Given the expressions above for a thermally ideal gas and a general mixture of these gases, this finally yields for the rate of temperature change in an enclosed volume:
* ⎛ T dc v ⎞  ⎜ 1 +  ⎟ * c v dT ⎠ ⎝
· T 1 · · * * ⋅  =  ⋅ ( [ M in ⋅ c p ( t ,T i n ) ] ⋅ T i n – [ M o u t ⋅ c p ( t ,T ou t ) ] ⋅ T o ut – Q lo ss )+ M⋅e T in o ut · ∑ ( c vi ( T ) ⋅ m i )
· V i –  – ( γ * – 1 ) ⋅ * V M ⋅ cv
Main Index
400 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
Airbag Definition
Constant Volume Tank Tests
Constant volume tank tests are used to characterize inflators. The inflator is ignited within the tank and, as the propellant burns, gas is generated. The inflator temperature is assumed to be constant. From experimental measurements of the time history of the tank pressure it is straightforward to derive the · mass flow rate, m . From energy conservation, where respectively, we obtain:
· · · · cp m Ti = cv m Tt + cv m Tt Ti
and
Tt
are defined to be the temperature of the inflator and tank,
For a perfect gas under constant volume,
· · · p V = m R T t + m RT t
· V = 0,
hence,
and, finally, we obtain the desired mass flow rate:
· cv p V · m = cp R Tt
Porosity in Airbags
Porosity is defined as the flow of gas through the airbag surface. There are two ways to model this: 1. Holes: The airbag surface contains a discrete hole. 2. Permeability: The airbag surface is made from material that is not completely sealed. The same porosity models are available for both the uniform pressure airbag model as the Eulerian coupled (CFD) airbag model. The porous flow can be either to and from the environment or into and from another uniform pressure model. Holes Flow through holes as defined on the SMALHOLE entries is based on the theory of onedimensional gas flow through a small orifice. LARGHOLE entries define flow through a hole with the velocity method. The velocity method can only be active for Eulerian airbags. When the SMALHOLE is used on the AIRBAG card, the theory of onedimensional gas flow through a small orifice is applied. Velocity Method The transport of mass through the porous area is based on the velocity of the gas in the Eulerian elements, relative to the moving of coupling surface (airbag fabric).
Main Index
Chapter 11: Airbags and Occupant Safety 401
Airbag Definition
Eulerian Element
SX i nt
v
Face of the coupling surface that intersects the Eulerian element
Coupling Surface
The volume of the Eulerian materian transported through the faces of the coupling surface that intersect an Eulerian element is equal to
V trans = – d t ⋅ α ⋅ ( v ⋅ A )
where
V trans
= transported volume during one time step ( V trans > 0 for the outflow; the inflow). = time step. = porosity coefficient. = velocity vector of the gas in the Eulerian mesh
V trans < 0
for
dt α v A
= area of the face of the coupling surface that intersects the Eulerian element the area of the face that lies inside the Eulerian element.
A
is equal to
The transport mass through the porous area is equal to the density of the gas times the transported volume. Pressure Method The transport of mass through the porous area is based on the pressure difference between the gas in the Eulerian element and the outside pressure. The outside pressure is the pressure as specified on the ENVIRONM section in the AIRBAG entry.
Eulerian Element
S X i nt
v
Face of the coupling surface that intersects the Eulerian element
Coupling Surface
Main Index
402 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
Airbag Definition
The volume of the Eulerian material transported through the faces of the coupling surface that intersect an Eulerian element is equal to:
V trans = d t ⋅ α ⋅ ( A ⋅ A ) ⋅ 2 pρ γ p exh 2 / γ  ⎛ ⎞ – γ–1 ⎝ p ⎠ ⎛ p exh⎞ ⎝ p ⎠
γ+1 γ
where
V trans dt α v A p= ρ γ p exh
= transported volume during one time step ( V trans > 0 for outflow; = time step. = porosite coefficient. = velocity vector of the gas in the Eulerian mesh.
V trans < 0
for inflow).
= area of the face of the coupling surface that intersects the Eularian element the area of the face that lies inside the Eulerian element. = pressure of the gas in the Eulerian element. = density of the gas in the Eulerian element. = adiabatic exponent = = pressure of the face.
C p ⁄ Cv .
A
is equal to
The pressure at the face is approximated by the onedimensional isentropic expansion of the gas to the critical pressure or the environmental pressure ascending to
⎧ > pc ⎪ p p = ⎨ env p env < p c ⎪ ⎩
where
pc
is the critical pressure:
r
2 p c p ⋅ ⎛  ⎞ r – 1 ⎝ γ + 1⎠
In case the outside pressure is greater than the pressure of the gas, inflow through the coupling surface occurs. This porosity model can only be used for ideal gases; i. e., materials modeled with the gamma law equation of state (EOSGAM). Permeability Permeability is defined as the velocity of gas through a surface area depending on the pressure difference over that area.
Main Index
Chapter 11: Airbags and Occupant Safety 403
Airbag Definition
On the PERMEAB entity in AIRBAG card, permeability can be specified by either a coefficient or a pressure dependent table: 1. Coefficient:
Gas Velocity
δ ( massflow) ) coeff = δ ( pressdiff )
Massflow = coeff * pressure_difference
Coeff press_diff
2. Table
Gas Velocity
Pressure Dependent Table
press_diff
The velocity of the gas flow can never exceed the sonic speed:
V max – V sonic – γ R T crit T crit
where
γ
is the gas constant of in or outflowing gas and
is the critical temperature.
The critical temperature can be calculated as follows:
T crit 2  – T gas ( γ + 1 )
where
T gas
is the temperature of outflowing gas.
Initial Metric Method for Airbags
If the reference configuration of the airbag is taken as the folded configuration, the geometrical accuracy of the deployed bag will be affected by both the stretching and the compression of elements during the folding process. Such element distortions are very difficult to avoid in a folded bag. By reading in a reference configuration such as the final unstretched configuration of a deployed bag, any distortions in the initial geometry of the folded bag will have no effect on the final geometry of the inflated bag. This is because the stresses depend only on the deformation gradient matrix:
∂x i F ij = ∂X j
Main Index
404 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
Airbag Definition
where the choice of X j may coincide with the folded or unfold configurations. It is this unfolded configuration which may be specified here. Note that a reference geometry which is smaller than the initial airbag geometry will not induce initial tensile stresses. If a liner is included and the LNRC parameter set to 1 in MATD034, compression is disabled in the liner until the reference geometry is reached; i.e., the fabric element becomes tensile.
Heat Transfer in Airbags
For airbags with high temperature, energy is exchanged with the environment. There are two ways to define heat transfer in airbags, convection (CONVECT) and radiation (RADIATE). The heattransfer rates due to convection and radiation are defined by: 1. Convection:
q conv – h ( t ) A ( T – T env )
where h ( t ) is the timedependent heat transfer coefficient, A is the (sub)surface area for heat transfer, T is the temperature inside the airbag, and T env is the environment temperature. 2. Radiation:
A q rad = eA s [ T A – T env ]
where e is the gas emissivity, A is the (sub)surface area for heat transfer, inside the airbag, and T env is the environment temperature.
T
is the temperature
Main Index
Chapter 11: Airbags and Occupant Safety 405
Seatbelts
Seatbelts
Belt elements are single degree of freedom elements connecting two nodes. When the strain in an element is positive (i.e. the current length is greater then the unstretched length), a tension force is calculated from the material characteristics and is applied along the current axis of the element to oppose further stretching. The unstretched length of the belt is taken as the initial distance between the two nodes defining the position of the element plus the initial slack length. Seatbelt shell elements must be used with caution. The seatbelt shells distribute the loading on the surface of the dummy more realistically than the two node belt elements. For the seatbelt shells to work with sliprings and retractors it is necessary to use a logically regular mesh of quadrilateral elements.
Slipring
Retractor
Top View:
RN5 SN5
RE4
RN4
SRE14
SN4
SRE24 SRE23
RE3
RN3
SRE13
SN3
RE2
RN2
SRE12
SN2
SRE22 SRE21
RE1
SN1 RN1
SRE11
Figure 133
Seatbelt Shell Elements Definition
The ordering of the nodes and elements are important for seatbelt shells.
Main Index
406 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
Seatbelts
Seatbelt Pretensioner
Pretensioners allow modeling of five types of active devices which tighten the belt during the initial stages of a crash. Types 1 and 5 represent a pyrotechnic device which spins the spool of a retractor, causing the belt to be reeled in. The user defines a pullin versus time curve which applies once the pretensioner activates. Types 2 and 3 represent preloaded springs or torsion bars which move the buckle when released. The pretensioner is associated with any type of spring element including rotational. Note that when the preloaded spring, locking spring, and any restraints on the motion of the associated nodes are defined in the normal way; the action of the pretensioner is merely to cancel the force in one spring until (or after) it fires. With the second type, the force in the spring element is canceled out until the pretensioner is activated. In this case, the spring in question is normally a stiff, linear spring which acts as a locking mechanism, preventing motion of the seat belt buckle relative to the vehicle. A preloaded spring is defined in parallel with the locking spring. This type avoids the problem of the buckle being free to ‘drift’ before the pretensioner is activated. Type 4, a force type, is described below. To activate the pretensioner, the following sequence of events must occur: 1. Any one of up to four sensors must be triggered. 2. Then a userdefined time delay occurs. 3. Then the pretensioner acts. Type 1 pretensioner is intended to simulate a pyrotechnic retractor. Each retractor has a loading (and optional unloading) curve that describes the force on the belt element as a function of the amount of belt that has been pulled out of the retractor since the retractor locked. The type 1 pretensioner acts as a shift of this retractor load curve. An example will make this clear. Suppose at a particular time that 5mm of belt material has left the retractor. The retractor responds with a force corresponding to 5mm pullout on it's loading curve. But suppose this retractor has a type 1 pretensioner defined, and, at this instant of time, the pretensioner specifies a pullin of 20mm. The retractor then responds with a force that corresponds to (5mm + 20mm) on it's loading curve. This results in a much larger force. The effect can be that belt material will be pulled in, but there is no guarantee. The benefit of this implementation is that the force vs. pullin load curve for the retractor is followed and no unrealistic forces are generated. Still, it may be difficult to produce realistic models using this option, so two new types of pretensioners have been added. These are available in MD Nastran r2 and later versions. Types 2 and 3 are simple triggers for activating or deactivating springs, which then pull on the buckle. No changes have been made to these, and they are not discussed here. The type 4 pretensioner takes a force vs. time curve (see Figure 134). Each time step, the retractor computes the desired force without regard to the pretensioner. If the resulting force is less than that specified by the pretensioner load curve, then the pretensioner value is used instead. As time goes on, the pretensioner load curve should drop below the forces generated by the retractor, and the pretensioner is then essentially inactive. This provides for good control of the actual forces, so no unrealistic values are generated. The actual direction and amount of belt movement is unspecified, and depends on the other forces being exerted on the belt. This is suitable when the force the pretensioner exerts over time is known.
Main Index
Chapter 11: Airbags and Occupant Safety 407
Seatbelts
Force Retractor Pullout Force
Defined Force vs. Time Curve Retractor Lock Time
Time
Figure 134
Force versus Time Pretensioner. At the intersection, the retractor locks.
The type 5 pretensioner is essentially the same as the old type 1 pretensioner, but with the addition of a force limiting value. The pullin is given as a function of time, and the belt is drawn into the retractor exactly as desired. However, if at any point the forces generated in the belt exceed the pretensioner force limit, then the pretensioner is deactivated and the retractor takes over. In order to prevent a large discontinuity in the force at this point, the loading curve for the retractor is shifted (in the abscissa) by the amount required to put the current (pullout, force) on the load curve. For example, suppose the current force is 1000, and the current pullout is 10 (10mm of belt has been pulled in by the pretensioner). If the retractor would normally generate a force of 1000 after 25mm of belt had been pulled OUT, then the load curve is shifted to the left by 3, and remains that way for the duration of the calculation. So that at the current pullin of 10, it generates the force normally associated with a pull out of 25. If the belt reaches a pull out of 5, the force is as if it were pulled out 40 (5 + the shift of 35), and so on. This option is included for those who liked the general behavior of the old type 1 pretensioner, but has the added feature of the force limit to prevent unrealistic behavior. The type 6 pretensioner is a variation of the type 4 pretensioner, with features of the type 5 pretensioner. A force vs. time curve is input and the pretensioner force is computed each cycle. The retractor linked to this pretensioner should specify a positive value for PULL, which is the distance the belt pulls out before it locks. As the pretensioner pulls the belt into the retractor, the amount of pullin is tracked. As the pretensioner force decreases and drops below the belt tension, belt will begin to move back out of the retractor. Once PULL amount of belt has moved out of the retractor (relative to the maximum pull in encountered), the retractor will lock. At this time, the pretensioner is disabled, and the retractor force curve is shifted to match the current belt tension. This shifting is done just like the type 5 pretensioner. It is important that a positive value of PULL be specified to prevent premature retractor locking which could occur due to small outward belt movements generated by noise in the simulation.
Main Index
408 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
Seatbelts
Seatbelt Retractor
The unloading curve should start at zero tension and increase monotonically (i.e., no segments of negative or zero slope). Retractors allow belt material to be paid out into a belt element. Retractors operate in one of two regimes: unlocked when the belt material is paid out, or reeled in under constant tension and locked when a user defined forcepullout relationship applies. The retractor is initially unlocked, and the following sequence of events must occur for it to become locked: 1. Any one of up to four sensors must be triggered. (The sensors are described below.) 2. Then a userdefined time delay occurs. 3. Then a userdefined length of belt must be paid out (optional). 4. Then the retractor locks and once locked, it remains locked. In the unlocked regime, the retractor attempts to apply a constant tension to the belt. This feature allows an initial tightening of the belt and takes up any slack whenever it occurs. The tension value is taken from the first point on the forcepullout load curve. The maximum rate of pull out or pull in is given by 0.01 × fed length per time step. Because of this, the constant tension value is not always achieved. In the locked regime, a userdefined curve describes the relationship between the force in the attached element and the amount of belt material paid out. If the tension in the belt subsequently relaxes, a different userdefined curve applies for unloading. The unloading curve is followed until the minimum tension is reached. The curves are defined in terms of initial length of belt. For example, if a belt is marked at 10mm intervals and then wound onto a retractor, and the force required to make each mark emerge from the (locked) retractor is recorded, the curves used for input would be as follows: 0 Minimum tension (should be > zero)
10mm Force to emergence of first mark 20mm Force to emergence of second mark . . . . . .
Pyrotechnic pretensions may be defined which cause the retractor to pull in the belt at a predetermined rate. This overrides the retractor forcepullout relationship from the moment when the pretensioner activates. If desired, belt elements may be defined which are initially inside the retractor. These will emerge as belt material is paid out, and may return into the retractor if sufficient material is reeled in during unloading. Elements e2, e3 and e4 are initially inside the retractor, which is paying out material into element e1. When the retractor has fed L crit into e1, where
L crit = fed length – 1.1 × minimum length
Main Index
Chapter 11: Airbags and Occupant Safety 409
Seatbelts
(minimum length defined on belt material input) (fed length defined on retractor input) Element e2 emerges with an unstretched length of 1.1∞ minimum length ; the unstretched length of element e1 is reduced by the same amount. The force and strain in e1 are unchanged; in e2, they are set equal to those in e1. The retractor now pays out material into e2. If no elements are inside the retractor, e2 can continue to extend as more material is fed into it. As the retractor pulls in the belt (for example, during initial tightening), if the unstretched length of the mouth element becomes less than the minimum length, the element is taken into the retractor. To define a retractor, the user enters the retractor node, the ‘mouth’ element (into which belt material will be fed), e1 in Figure 135, up to 4 sensors which can trigger unlocking, a time delay, a payout delay (optional), load and unload curve numbers, and the fed length. The retractor node is typically part of the vehicle structure; belt elements should not be connected to this node directly, but any other feature can be attached including rigid bodies. The mouth element should have a node coincident with the retractor but should not be inside the retractor. The fed length would typically be set either to a typical element initial length, for the distance between painted marks on a real belt for comparisons with high speed film. The fed length should be at least three times the minimum length. If there are elements initially inside the retractor (e2, e3 and e4 in the Figure 135) they should not be referred to on the retractor input, but the retractor should be identified on the element input for these elements. Their nodes should all be coincident with the retractor node and should not be restrained or constrained. Initial slack will automatically be set to 1.1 × minimum length for these elements; this overrides any userdefined value. Weblockers can be included within the retractor representation simply by entering a ‘locking up’ characteristic in the force pullout curve (see Figure 136. The final section can be very steep (but must have a finite slope).
Main Index
410 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
Seatbelts
Element 1
Before
Element 2 Element 3 Element 2 Element 4 Element 1
After
Element 3 Element 4
All nodes within this area are coincident
Figure 135
Elements in a Retractor
with weblockers without weblockers
FORCE
PULLOUT
Figure 136
Retractor Force Pull Characteristics
Main Index
Chapter 11: Airbags and Occupant Safety 411
Seatbelts
Seatbelt Sensor
Sensors are used to trigger locking of retractors and activate pretensioners. Four types of sensors are available which trigger according to the following criteria: Type 1 – When the magnitude of x, y, or z acceleration of a given node has remained above a given level continuously for a given time, the sensor triggers. This does not work with nodes on rigid bodies. – When the rate of belt payout from a given retractor has remained above a given level continuously for a given time, the sensor triggers. – The sensor triggers at a given time. – The sensor triggers when the distance between two nodes exceeds a given maximum or becomes less than a given minimum. This type of sensor is intended for use with an explicit mass/spring representation of the sensor mechanism.
Type 2 Type 3 Type 4
By default, the sensors are inactive during dynamic relaxation. This allows initial tightening of the belt and positioning of the occupant on the seat without locking the retractor or firing any pretensioners. However, a flag can be set in the sensor input to make the sensors active during the dynamic relaxation phase.
Seatbelt Slipring
Sliprings allow continuous sliding of a belt through a sharp change of angle. Two elements (1 & 2 in Figure 137) meet at the slipring. Node B in the belt material remains attached to the slipring node, but belt material (in the form of unstretched length) is passed from element 1 to element 2 to achieve slip. The amount of slip at each time step is calculated from the ratio of forces in elements 1 and 2. The ratio of forces is determined by the relative angle between elements 1 and 2 and the coefficient of friction, μ . The tension in the belts are taken as T 1 and T 2 , where T 2 is on the high tension side and T1 is the force on the low tension side. Thus, if
T2
is sufficiently close to
T1 ,
no slip occurs; otherwise, slip is just
sufficient to reduce the ratio T 2 ⁄ T 1 to e μΘ . No slip occurs if both elements are slack. The outofbalance force at node B is reacted on the slipring node; the motion of node B follows that of slipring node. If, due to slip through the slipring, the unstretched length of an element becomes less than the minimum length (as entered on the belt material card), the belt is remeshed locally: the short element passes through the slipring and reappears on the other side (see Figure 137). The new unstretched length of e1 is 1.1 × minimum length . Force and strain in e2 and e3 are unchanged; force and strain in e1 are now equal to those in e2. Subsequent slip will pass material from e3 to e1. This process can continue with several elements passing in turn through the slipring. To define a slipring, the user identifies the two belt elements which meet at the slipring, the friction coefficient, and the slipring node. The two elements must have a common node coincident with the slipring node. No attempt should be made to restrain or constrain the common node for its motion will
Main Index
412 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
Seatbelts
automatically be constrained to follow the slipring node. Typically, the slipring node is part of the vehicle body structure and, therefore, belt elements should not be connected to this node directly, but any other feature can be attached, including rigid bodies.
Slipring B Element 2 Element 1 Element 1 Element 3 Element 2
Element 3
Before
Figure 137
After
Elements Passing Through Slipring
Main Index
Chapter 11: Airbags and Occupant Safety 413
Occupant Dummy Models
Occupant Dummy Models
The occupant dummy models, also known as ATDs (Anthropomorphic Test Devices) were introduced in the previous release of SOL 700 for those applications that airbag was not needed. Many applications such as sled test and aircraft seat design, or armored vehicle design where the occupant behavior is studied when a land mine is detonated do not require airbag simulation. SOL 700 supports occupant dummies that are readily available in LSDYNA *key file format. These include: 1. LSDYNA public domain dummies:
• 5th percentile deformable female dummy • 50th percentile deformable male dummy • 95th percentile deformable male dummy • 5th percentile rigid female dummy • 50th percentile rigid male dummy • 95th percentile rigid male dummy
2. ETA (Engineering Technology Associates) calibrated dummies. The calibrated dummies are similar to LSDYNA dummies except that they are validated against a set of standard tests. The calibration tests are conducted on all Hybrid III, SID and EUROSID models to assure model fidelity.
• 5th percentile deformable female dummy • 50th percentile deformable male dummy • 95th percentile deformable male dummy • 5th percentile rigid female dummy • 50th percentile rigid male dummy • 95th percentile rigid male dummy • EUROSID  1 (Euro Side Impact Dummy) • US DOT SID • EEVC Upper Legform • FMVSS 201 Headform
Figure 138
Typical Occupant Dummies in VPG (courtesy of ETA)
Main Index
414 MD Nastran R3 Explicit Nonlinear (SOL 700) User’s Guide
Pre and Postprocessing
There are many other ATD models available through ETA VPG (Virtual Proving Ground) prepost processor. In addition VPG supports the following barriers: Barriers
• FMVSS & ECE Side Impact Barriers • 0Degree Front Rigid Barrier • 30deg Rigid Front Barrier • Front Offset Deformable Barrier • Rear Impact Barrier
Impactors
• Rams • Pendulums • Head Forms
For more details, contact ETA in Troy, Michigan. 3. FTSS (First Technology Safety Systems) ATDs. The FTSS ATD’s are high fidelity dummies and are available with additional licensing. The following FTSS dummies are supported through SimX Crash or ETA – VPG:
• 5% Female dummy • SID II • Hybrid III  3% Child Dummy (W.I.P.) • Hybrid III  6% Child Dummy (W.I.P.) • BIOSID (W.I.P.)
Pre and Postprocessing
Even though MD Patran supports SOL 700, it lacks a versatile dummy positioner or airbag folder. Dummy positioning can be done using prepost processing tools such as MSC Software SimX Crash and ETAVPG, both of which support SOL 700.
Main Index
Chapter 12: System Information and Parallel Processing MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
12
System Information and Parallel Processing
Introduction 416 416
General Information User Notes 419
Additional information of Different Platforms
421
Main Index
416 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Introduction
Introduction
This chapter supports
• A general description about the hardware and software requirements and definitions. • How to use parallel operation for different platforms.
General Information
Release Platforms
MD Nastran SOL 700 was built and tested on the following hardware with the listed software installed as given in Table 121. Table 121 Supported Hardware Configuration Platform Intel (32bit) SGI ** R10K/R12K HPUX – PA RISC 2.0 ** HPUX Itanium2 Compaq Alpha Sun Sparc Solaris (**) IBM RS/6000 Linux Itanium2 IA64 *) Linux EM64T x86_64 Intel Linux ** Operating System Windows 2000SP4 Windows XPSP2 IRIX64 6.5.22m HPUX B.11.00 HPUX B.11.23 Tru64 UNIX v5.1/1885 Solaris 8, 64bit ( = SunOS 5.8 ) AIX 5.1 RedHat Advanced Server 3 RedHat Enterprise Linux WS 3 Red Hat Linux release 9
*For correct operation of the Intel Fortran compiler, MS Visual Studio .NET must be installed prior to installing the Intel 8.1 compiler. For SOL 700, these installs are not mandatory. **The SOL 700 in MD Nastran 2006r2 does not support the following configurations: SGI: R4k and R5k; HP: HPUX 10.20; SUN: Solaris 7; Linux: Redhat 7.3, Windows NT
Main Index
Chapter 12: System Information and Parallel Processing 417
General Information
MPI for MD Nastran SOL 700
SOL 700 requires MPI to be installed on every machine used. This is true even for single processor machines. SOL 700 expects hardwarespecific native MPI to have been installed at default locations. When MPI is not properly installed on your Unix/Linux machine or is not installed at the expected default location, a job submission will exit with an error message to this effect. To avoid problems of this nature, or problems caused by different versions of MPI, on several of the supported platforms the MPI version is now part of the release and will be installed at a defined location. See Table 122 for details.
Table 122
MPI Version and Expected Location Operating System Windows 2000SP4 Windows XPSP2 IRIX64 6.5.22m HPUX B.11.00 HPUX B.11.23 Tru64 Unix v5.1/1885 Solaris 8, 64bit ( = SunOS 5.8 ) AIX 5.1 RedHat Advanced Server 3 RedHat Enterprise Linux WS 3 Red Hat 9 MPI Version MPIch V.1.2.5 SGI MPT 1.2 HP MPI 1.08 HP MPI 1.08 Compaq dmpirun 1.9.6 SUN HPC mprun 5.0 POE 3.2 LAM 6.5.9 LAM 6.5.9 LAM 6.5.9 MPI Location 1 2 3 3
/usr/opt/MPI196 /opt/SUNWhpc/HPC5.0 /usr/lpp/ppe.poe
Platform Intel (32bit) SGI R10K/R12K HPUX – PA RISC 2.0 HPUX Itanium2 Compaq Alpha Sun Sparc Solaris IBM RS/6000 Linux Itanium2 IA64 Linux EM64T x86_64 Intel Linux
3 3 3
1. Exact location not important as long as it is installed. Install location will be picked up from the registry. 2. Proper install will provide soft links to /usr/bin, /usr/lib. 3. MPI is part of the release and automatically installed in $installdir/bin/exe/mpi
Hardware and Software Requirements
For the Linux platforms, the LAM 6.5.9 is required on all the machines that are clustered for the job submission. LAM 6.5.9 is made available and is installed with the Nastran SOL 700 installation.
Main Index
418 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
General Information
For HP (HPUX Risc2 and HPUX Itanium2), the HP MPI is required on all the machines that are clustered for the job submission. HP MPI 02.00.00.00 is made available and is installed with the MD Nastran SOL 700 installation. For all other Unix platforms, it is assumed that a native MPI program is preinstalled. For SGI IRIX64, "Array Services 3.3 MPT 1.3" should be available. For Sun/Solaris, it is "HPC 5.0". For Dec/Alpha "Compaq MPI 1.9.6" and for IBM AIX "POE 3.2". Although no specific hardware requirements exist for MD Nastran SOL 700 to run in distributed memory parallel mode, it is preferable to have fast network connections between the machines if more than one machine is used. It is recommended that the network should have a speed of at least 100 MBit per second. The appropriate licenses are required in all machines used. For windows, "MPMPICH" from University of Aachen, Lehrstuhl für Betriebssysteme is required on all the machines that are clustered for the job submission. MPICH is made available on the Nastran 2005 installation CD. Install MPIch version 1.2.5, by "default" and then the MPIch service will be installed. And, make sure you define MSC_LICENSE_FILE and/or LM_LICENSE_FILE environment variables correctly. If you make any changes make sure you always reboot your machine. The Windows Services manager that is running the MPIch service needs to get these new settings and that is why a reboot is necessary. One workaround is to make a registry entry setting for MSC_LICENSE_FILE in the Flexlm key location. This change is immediate and doesn’t need a reboot. The head node is used to check your licenses. So make sure the head node has correctly defined license settings. If only two machines are to be used, you can use a hub or a crossover cable to connect them. If more than two machines are to be used, a switch is preferable. TCP/IP is used for communications.
Compatibility
Version 2006r2 only support connection of homogeneous networks with machines of the same type. Two machines are compatible if they can both use the same executables. Some examples of compatible machines are: 1. Several machines with exactly the same processor type and O/S. 2. One SGI R8000/Irix 6.5 and one SGI R10000/Irix 6.5 machine. 3. One HP JClass/HPUX11.0 and one HP CClass/HPUX11.0.
Definition
1. Root machine: The machine on which the job is started. 2. Remote machine: Any machine other than the root machine that is part of a distributed parallel run on the network.
Main Index
Chapter 12: System Information and Parallel Processing 419
User Notes
User Notes
This section assumes that MD Nastran has been successfully installed on at least one of the machines that are to be used in a distributed analysis, and that the appropriate SOL 700 licenses are in order. Assume that host1 is the host name of the machine on which the job is to be started (the root machine). The host names of the other machines (remote machines) are host2, host3, etc. MD Nastran SOL 700 will always use one of the CPU's of the root machine. Additional slave nodes can be used in a single calculation by using a hostfile. The hostfiles between windows and Linux/Unix are different.
Specification of the Host file for Windows
See below for an example of a hostfile for windows:
segers 1 D:\MSC.Software\nastran\2006r2\dytranlsdyna\dytranlsdyna.exe weenix 2 C:\MSC.Software\nastran\2006r2\dytranlsdyna\dytranlsdyna.exe
Specification of the Host file for UNIX and LINUX
To use more than one platform, a host file is must be used. The host file has the following general format. hostname1 [number 1 of CPUs] [working directory] [executable location 1] hostname2 [number 2 of CPUs] [working directory] [executable location 2] hostname3 [number 3 of CPUs] [working directory] [executable location 3] ... hostnamei number i CPUs working directory = name of the machine in the network. Each new hostname must be on a new line. working directory on each machine. This is dummy input for this version.
= number of CPU’s to be used on each machine. Default is 1. =
executable location
location of executable on each machine. Default for executable location 1 is the script will figure out MSC = installation on hostname1. Default for other executable location is executable location 1.
For example, Start run on four CPU’s: two on the root machine, bari and two on the remote machine, pisa. Both machines have exactly the same location of executable. Then hostfile has the following lines: bari 2 pisa 2
Main Index
420 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
User Notes
And, if two CPU’s on bari and one CPU on pisa are used and pisa has different executable location, then the following lines must be specified in hostfile: bari 2 pisa 1 /tmp /app2/msc/nastran/msc2006r2/dyna (note the blank after /tmp)
How to Run MD Nastran SOL 700 in Parallel
This section explains how to run SOL 700 using mutiparallel processors. If test.dat file is submitted to analyze and host file name is hostlist.txt in UNIX, the command is: nastran test1.dat hlist=hostfile If more than one processor on one machine is used, sol700.pth file can be used using PATH=3. sol700.pth includes how many number of processors are used, the full path of DytranLSDYNA executable, amount of memory, etc. The details of this file are found in MD Nastran Quick Reference Guide. For example, sol700.pth includes the commands below if four processors, 20 Mb memories are used in this machine: /users/joe/sol700/run_dytran exe=/users/joe/sol700/dytranlsdyna nproc=4 memory=20m The name of script is run_dytran located to /users/joe/sol700/ and the executive dytranlsdyna located to /users/joe/sol700/dytranlsdyna. In this case, SOL 700 entry in the Executive Control Statement must have PATH=3. SOL700, 129, PATH=3 Or without sol700.pth, the command can control the number of processors: /users/joe/sol700/run_dyna exe=/users/joe/sol700/dytranlsdyna jid=abcd.dat nproc=4 memory=20m In the case of Windows, the sequence for multiparallel processing is similar to that of UNIX and Linux. For example, if PC’s named bari and pisa will use two and four processors respectively, sol700.pth must include following lines (additional options are shown in the MD Nastran Quick Reference Guide for the SOL 700 entry). E:\users\joe\sol700\ryn_dytran nproc=6 steps=2 memory=90m machine=bari#2+pisa#4 For SOL 700 parallel runs, PATH=3 must be set on the SOL 700 executive control entry and a file named sol700.pth similar to the above must be used. The option to use a separate hostfile for PC’s is not presently available.
Main Index
Chapter 12: System Information and Parallel Processing 421
Additional information of Different Platforms
How to Run Fluid Structure Interaction (FSI) in Parallel
See the following steps to run an FSI appplication on multiple processors: 1. In the sol700.pth file, add fsidmp=yes. 2. Add PARAM,FSIDMP,YES in the input deck. Parallel FSI is supported on Windows 32, Linux 32, and Linux x8664 platforms.
Additional information of Different Platforms
HPUX11 & HPUX IA64 The MD Nastran installation will create a full installation of HP MPI in the MSC installation directory. It is important to note that MD Nastran version 2006r2 is used on all the platforms in the cluster. Using different versions of MD Nastran for DMP calculations is not allowed. Linux X8664 & Linux i386 & Linux Itanium2 1. First check that rsh is installed by doing: rpm qa  grep rsh The answer should be: rshserver0.1714 rsh0.1714 2. If installed then check the following two files: /etc/xinetd.d/rsh /etc/xinetd.d/rlogin There is a line that should say:
disable = no
If it says yes, change it to no. 3. Check the following file:
/etc/nsswitch.conf
It should say: hosts: files dns nis
files should come before anything else
4. Restart after making changes: /etc/init.d/xinetd restart 5. Add in .rhosts file to your home directory that contains the following entries: {machine1} {user} {machine1 full name} {user} {machine2} {user} {machine2 full name} {user} +
Main Index
422 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Additional information of Different Platforms
For example: murano walter murano.adams.com walter usher walter usher.adams.com walter miller walter miller.adams.com walter + 6. Make sure that rsh does not show any echo's from the .cshrc file. You must modify your .cshrc file such with if (! $?prompt) that no echo's or clear's are done. You can check that by: rsh {remote system} date 7. Set the path such that hboot can be executed. This means that the path must be set correctly by the .cshrc on the remote machine. You can test this by: rsh {remote system} 'echo $path' rsh {remote system} 'which hboot' Do not omit the quotes from the command line. Without them, the actual path of the machine you are on will be echoed. The same working directory as used on the root machine does not have to exist on the remote machines. On Linux systems, if multiple jobs are to be run by a particular user simultaneously, a manual lamboot from the command line (or a script) must be done prior to running the first job. In addition, PATH=3 must be specifid on the SOL 700, ID entry and a line lamboot=no added. A user can determine the proper location of lam by running a small job, examining the f06 file and searching for the string “execut” The command shown in the example below should then be executed. A typical f06 file shows the following: dytranlsdyna will be executed using the command /usr/msc/programs/nastran/dytranlsdyna/run_dytran For this case, the following commands should be placed in a script and the script executed to establish the lamboot: #!/bin/csh setenv LAMHOME /usr/msc/programs/nastran/dytranlsdyna/mpi/bin set path = ($LAMHONE $path) $LAMHOME/lamboot To check if lam is running, the user should execute the following command: ps ef  grep lam As long as the lam processes is not “killed”, it will remain running unless the machine is rebooted. To terminate the process, either kill it from the command line or enter the following command: /usr/msc/programs/nastran/dytranlsdyna/mpi/bin/lamhalt
Main Index
Chapter 12: System Information and Parallel Processing 423
Additional information of Different Platforms
AIX The same working directory on the remote machines (the full path) must exist as the one used on the root machine. Alpha The same working directory as used on the root machine does not have to exist on the remote machines. Solaris The same working directory on the remote machines (the full path) must exist as the one used on the root machine. SGI Altix & Irix64 Before you can start the MPI on SGI, you first need to setup and start the array services. This array keeps track of which machines belong to the MPI Universe. Unfortunately you need to be root to define one: <> su root First check if the array services are enabled: # vi /usr/lib/array/arrayd.auth The variable AUTHENTICATION must be set from NOREMOTE to NONE. Create the array: # vi /usr/lib/array/arrayd.conf [add in the following lines:] array {array name} machine {machine1} machine {machine2} Add as many as desired. For instance: array irixaa machine zephyr machine tigra Restart (or start) the array: # /etc/init.d/array restart You also need to add a special login name: #vi /etc/passwd guest:x:81:99:array services guest:/dev/null:/bin/tcsh #vi /etc/group guest:x:99: Exit root login: # exit
Main Index
424 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Additional information of Different Platforms
To check which arrays are active you can do: <> /usr/sbin/ainfo array Arrays known to array services daemon ARRAY me IDENT 0x78ea ARRAY {array name} IDENT 0x78f2 The first array is the one for the machine itself (so you can DMP on the machine itself if it has more than 1 CPU). The second one is the one we defined ourselves and can be used for DMP with other machines. To see which machines are defined in the array: <> /usr/sbin/ainfo a {array name} machines For instance: <> /usr/sbin/ainfo a irixaa machines Machines in array "irixaa" per default server MACHINE zephyr HOSTNAME zephyr PORT 5434 IP_ADDR 192.168.22.196 IDENT unknown MACHINE tigra HOSTNAME tigra PORT 5434 IP_ADDR 192.168.22.194 IDENT unknown So this array has two machines available: zephyr and tigra. To check if the array was able to access the remote machines: <> /usr/sbin/array a {array name} who The return should be "who" from all machines. For instance: <> /usr/sbin/array a irixaa who User walter walter zephyr Machine tigra.dtw.macsch.com From 172.31.188.227 72.31.188.227 What /bin/tcsh array a irixaa who
The same working directory on the remote machines (the full path) must exist as the one used on the root machine. To start a SOL 700 calculation, you must specify the array to use: Add array=irixaa to sol700.pth file and use PATH=3 on the SOL 700 executive control entry
Main Index
Chapter 12: System Information and Parallel Processing 425
Additional information of Different Platforms
For Altix, the user has two options. He may use SGI MPI or LAM MPI. The LAM is default. To use the SGI MPI the user must use the option: sgimpi=yes. When using SGI MPI the user should not forget to specify also the array. For example: Add sgimpi=yes array=irixaa to sol700.pth file and use PATH=3 on the SOL 700 executive control entry
Windows XP and 2000
MPIch (service) needs to be running on all the machines you are planning to use. You should either start MPIch manually using “%MPI_ROOT%\mpd\bin\mpd install” or make sure the service runs by default. Automatically starting the MPIch daemon is not supported. Run MPIRegister to allow your account to be used to run MPIch. "%MPI_ROOT%\mpd\bin\MPIRegister.exe" Define your specific account settings DOMAIN\user. Supply your domain password. Option "Do you want this action to be persistent (y/n)?" >> y (not sure what happens once your password will be changed though) If ever needed, to clean your registry, use: "%MPI_ROOT%\mpd\bin\MPIRegister.exe remove" The most important thing is that you define two shares correctly. Share 1: %NASTRAN_INSTALLDIR%\%NASTRAN_VERSION%\dyna Preferably this share should have full read access for everyone. Share 2: %YOUR_RUN_DIR% This should be the directory where your input file is located. So it might be a good idea to define a single location to run all your jobs. Access should be rather open on this share. You will need to have write permissions and most likely the MPIch "service" too. All tests have been performed with Full Access for everyone.
Extra MPIch information
For more information please visit: http://wwwunix.mcs.anl.gov/mpi/mpich/mpichnt/mpich.nt.faq.html Why do I get this error "LaunchProcess failed, CreateProcessAsUser failed, No more connections can be made to this remote computer at this time because there are already as many connections as the computer can accept."? This error usually occurs when you try to launch an executable from a shared directory on WindowsNT Workstation, Windows 2000 Professional, or WindowsXP Professional. The professional versions of Windows as apposed to the server editions have limitations on the file sharing capabilities. Place the executable on a network share on a server machine or copy the executable to the local drive of each machine.
Main Index
426 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Additional information of Different Platforms
Why do I get this error, "Logon failure: unknown user name or bad password"? You must have the same account credentials on all the nodes participating in the mpich job. If your cluster is set up with a domain controller then you can use a domain account to launch an mpich job. If you do not have a domain controller then you must set up user accounts on all the nodes individually with the same credentials on each node. Each user can have whatever password they choose, but they must use the same password on all the nodes. In other words, UserAPasswordA must be the same on all the nodes and UserBPasswordB must be the same on all the nodes, etc. Why do I get this error, "LaunchProcess failed, CreateProcessAsUser failed, The system cannot find the file specified."? The executable used in an mpich job must be available to all the nodes participating in the job. The path to the executable must be valid on all the nodes. This can be accomplished by copying the executable to a common location on all the nodes or copying it to a shared location. For example, you could copy cpi.exe to c:\temp\cpi.exe on all the nodes and run "mpirun np 3 c:\temp\cpi.exe". Or you could copy cpi.exe to a shared directory \\myhost\myshare\cpi.exe and then execute.
Main Index
Chapter 13: Examples MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
13
Examples
Crash 428 444 459
Airbags and Occupant Safety
Bird Strike and Fan Blade Out (FBO) Drop Test Defense 510 524 555
Time Domain NVH Prestress 564
Smooth Particle Hydrodynamics (SPH) Sheet Metal Forming Miscellaneous 596 582
576
Main Index
428 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Crash
Crash
Pickup Truck Frontal Crash
Description Auto companies perform crash tests simulation to increase safety of the passengers and comply with government regulations such as those of FMVSS (Federal Motor Vehicle Safety Standards). This is an example of a pickup truck frontal crash at 15 m/s (34 m.p.h.) against a rigid wall. To model the simulation, contact was defined between the truck and the rigid wall to predict the stress and deformations of the structure. SOL 700 Entries Included SOL 700 TSTEPNL PARAM, DYDTOUT PARAM, DYSHELLFORM PARAM, DYSHINP BCTABLE BCBODY BCPROP BSURF MATD020 MATD024 CSPOT MATD20M Model
Main Index
Chapter 13: Examples 429
Crash
The model has a total of 62984 grid points and 55572 elements as follow:
•
346 Bars
• 1815 Trias • 49954 Quads • 3458 Hexas
The rigid wall and the ground are modeled by 500 and 25 shell elements, respectively. All shell elements are BelytschkoTsay formulation. The crash speed of the truck is modeled by defining an initial velocity of 15 m/s, applied on all the grid points of the truck in a horizontal direction towards the wall. Gravity load is included to take into account the mass of the truck. The simulation time is 0.09 seconds. The unit system is Newtons, seconds, and millimeters. Input All nodes of the truck have an initial velocity specified by the TIC entry. All nodes of the rigid wall and the ground have been constrained in all the degrees of freedom. Contact is defined between: 1. The truck and rigid wall surface 2. The truck tires and the ground surface 3. Self contact for the truck to avoid penetration among various components. Input file: SOL 700,NLTRAN path=3 stop=1
SOL 700 is a executive control that activates an explicit nonlinear transient analysis.
TIME 10000 CEND ECHO = NONE DISPLACEMENT(SORT1,print,PLOT) = ALL Stress(SORT1,PLOT) = ALL Strain(SORT1,PLOT) = ALL accel(print,plot)= ALL velocity(print,plot)= ALL echo=both SET 990009 = 105843 105655 .. .. SET 990619 = 74752 77110
Main Index
430 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Crash
SET is an executive control entry that defines a set that contains some grid points. The set will later be referenced by the CSPOT entry in the bulk entry section.
LOADSET = 1 SPC = 2 DLOAD = 2 IC=1 TSTEPNL = 20 BCONTACT = 1 weightcheck=yes page The bulk entry section starts… BEGIN BULK INCLUDE 1.truck.bdf INCLUDE rconn.bdf TSTEPNL 20 10 + + + + 0
9e3
1
5
10
TSTEPNL is a SOL 700 bulk data entry which describes the number of Time Steps (10) and Time Increment (9e3 seconds) of the simulation. The total time is the product of the two entries. Notice here the Time Increment is only for the first step. The actual number of Time Increments and the exact value of the Time Steps is determined by SOL 700 during the analysis. The time step is a function of the smallest element dimension during the simulation.
PARAM,OGEOM,NO PARAM,AUTOSPC,YES PARAM,GRDPNT,0 PARAM,dydtout,0.001
PARAM, DYDTOUT is a SOL 700 bulk data entry which describes the time interval of d3plot outputs.
PARAM*,DYSHELLFORM,2
PARAM*, DYSHELLFORM is a SOL 700 bulk data entry to define the shell formulation. If
DYSHELLFORM = 2, the default shell formulation is BelytschkoTsay. PARAM,DYSHNIP,2
PARAM,DYSHINP is a SOL 700 bulk data entry that specifies the number of integration points for SOL 700 shell elements.
Main Index
Chapter 13: Examples 431
Crash
Bulk data entry that defines gravity load effects: $  GRAVITATION LSEQ 1 5 4 TLOAD1 4 3 GRAV 4 0 9806. DLOAD 2 1. 1. TABLED1 100 0. 1. 1.
0. 4 1.
100 0. ENDT
1.
GRAV is a bulk data entry that defines acceleration vectors for gravity or other acceleration loading.
Bulk data entry that defines Contact relations and Contact bodies: BCTABLE 1 SLAVE MASTERS SLAVE MASTERS SLAVE 4 1 2 3 4 5
BCTABLE is a SOL 700 bulk data entry that is also used by SOL 600 and it is meant to define MasterSlave as well as self contact.
BCBODY BCBODY BCBODY BCBODY BCBODY
1 2 3 4 5
3 3 3 3 3
DEFORM DEFORM DEFORM DEFORM DEFORM
1 2 3 4 5
0 0 0 0 0
BCBODY is a bulk data entry that is used by MD Nastran Implicit Nonlinear (SOL 600) and SOL 700
only, which defines a flexible or rigid contact body in 2D or 3D. BCPROP,1,1,2,3,4,10,11,12,+ .. .. +,268,269,270,271,24,67,70,195 $ rigid wall BCPROP,2,266 $ tires BCPROP,3,168,169,185,187 $ ground BCPROP,4,272
Main Index
432 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Crash
BCPROP is a bulk data entry that is used by MD Nastran Implicit Nonlinear (SOL 600) and SOL 700
only, which defines a 3D contact region by element properties. All elements with the specified properties define a contact body. $ Single Surface Contact BSURF 5 5 6 .. ..
7
8
9
10
11
BSURF is a bulk data entry that is used by MD Nastran Implicit Nonlinear (SOL 600) and SOL 700 only,
which defines a contact surface or body by element IDs. All elements with the specified IDs define a contact body. Bulk data entry that defines properties for shell and beam elements: $ ========== PROPERTY SETS ========== PSHELL 1 63 3.137 63 .. .. PBEAM 224 208 452.4 16290. .. .. Bulk data entry that defines concentrated mass at grid points: $ $ CONM2 .. .. CONM2 * conm2 * 1990624 1990693 91344 983105 1e06 1e06
63 16290. 32570.
Main Index
Chapter 13: Examples 433
Crash
Bulk data entry that continues defining properties for elements, spring, damper, rigid wall and ground: $ * mpc158 * PBAR 226 158 .. .. $ * mpc408 * PBAR 263 224 $ $ * spring * PELAS 264 14.4 $ $ * damper * PVISC 265 2.935 $ $ * rigid * PSHELL 266 70 .. .. * ground_prop * PSHELL 272 71
.01
1e05
1e05
.01
1e05
1e05
2.
70
70
2.0
71
71
Bulk data entry that defines material properties: $ ========= MATERIAL DEFINITIONS ========== $ $ $  Material MATRIG.21 id =21 MATD020 217.89e09 210000. .3 .. ..
MATD020 is a SOL 700 bulk data entry that is used to model rigid materials.
MATD024 .. ..
637.89e09 210000. 1
.3
0.9
MATD024 is a SOL 700 bulk data entry. It is used to model an elastoplastic material with an yield stress
versus strain curve and arbitrary strain rate dependency. Failure can also be defined based on the plastic strain or a minimum time step size.
Main Index
434 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Crash
Bulk data entry that defines tables: $ ================ TABLES ================= $ $  TABLE 1: table270 TABLES1 1 +B000710 +B000710 0. 270. .04879 320.3 .09531 366.3 .1398 402.5+B000711 +B000711 .2231 438.8 .2624 448.5 2.398 449. ENDT .. .. Bulk data entry that defines rigid body elements: $  Nodal Rigid Body: NodalRigid_5 RBE2 5 104247 123456 104272 104614 105039+B000181 +B000181 105043 105044 .. ..
104615
105038
Bulk data entry that defines boundary conditions and initial velocity of the pickup truck: SPC1 SPCADD TIC .. .. TIC 1 2 1 1 123456 1 1 510439 990803 1 1 THRU 991384 15000. 15000.
TIC is a bulk data entry that defines values for the initial conditions of variables used in structural transient analysis. Both displacement and velocity values may be specified at independent degrees of freedom.
Bulk data entry that defines spotweld with failure: CSPOT,990009,990009 +,,,1e+08 .. .. CSPOT,990619,990619 +,,,1e+08
CSPOT is a SOL 700 bulk data entry and was used to define spotweld with failure. The number of a specific SET defined in the executive control section was referred in the entry.
Main Index
Chapter 13: Examples 435
Crash
Bulk data entry that merges and defines joint between rigid bodies: MATD20M,181,180,221,182,183 .. ..
MATD20M is a SOL 700 bulk data entry. It was used to merges MATD020 rigid bodies into one assembly
for SOL 700 only. RBJOINT,1,REVOLUTE,903453,903456,101294,903455 .. .. RBJOINT,18,UNIVERS,983102,983103,983104,983105 ENDDATA
RBJOINT is a supported bulk data entry in SOL 700, which defines a Joint between two rigid bodies.
End of input file.
Main Index
436 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Crash
Results
t = 0.0 seconds
t = 0.035 seconds
t = 0.09 seconds
Main Index
Chapter 13: Examples 437
Crash
Trainbarrier Impact
Description This is an example of a train frontal crash at 30.559 meter/second against two rigid barrier cylinders to represent an inability to stop at the end of the line. To model the simulation, Contact was defined between the train and the rigid barrier to predict the stress and deformations of the structure. SOL 700 Entries Included SOL 700 TSTEPNL PARAM,DYDTOUT PARAM*,DYTSTEPDT2MS BCTABLE BCBODY BCPROP MATD024 MATD001 MATD020 CSPOT Model
Because of symmetry, only half of the actual structures (train and barrier) were modeled in this example. Boundary conditions were applied along the center line of the structures to ensure symmetric behavior. The model has a total of 117820 grid points and 113770 elements as follow:
• 586 Bars • 1533 Trias • 107047 Quads
The train model contains 112530 elements in the form of shell and beam. The rigid barrier cylinder model contains 1240 shell elements. All shell elements are BelytschkoWongChiang formulation. The crash speed of the train is modeled by defining an initial velocity of 30.559 meter/second, applied on all the grid points of the train in a horizontal direction towards the barrier. The simulation time is 0.35 seconds. The unit system is Kilonewton, millimeters, and milliseconds.
Main Index
438 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Crash
Input All nodes of the train have an initial velocity specified by the TIC card. All nodes along the center line have a boundary condition that ensures symmetric behavior of the structure. Adaptive contact relation was defined as follows: 1. Contact between the train and the barrier 2. SelfContact of the train components Input file: SOL 700,NLTRAN path=3 stop=1
SOL 700 is a executive control that activates an explicit nonlinear transient analysis.
TIME 10000 CEND ECHO = NONE DISPLACEMENT(SORT1,print,PLOT) = ALL Stress(SORT1,PLOT) = ALL Strain(SORT1,PLOT) = ALL accel(print,plot)= ALL velocity(print,plot)= ALL echo=both SET 26148 = 92304 92319 92334 .. .. SET 84302 = 98670 103916
SET is an executive control entry that defines a set that contains some grid points. The set will later be referenced by the CSPOT entry in the Bulk Entry Section.
$
LOADSET = 1 SPC = 2 $ DLOAD = 2 IC=1 TSTEPNL = 20 BCONTACT = 1 weightcheck=yes page
Main Index
Chapter 13: Examples 439
Crash
The bulk entry section starts... BEGIN BULK INCLUDE modelini.dat INCLUDE cspot_new.dat INCLUDE cspot_old.dat $ TSTEPNL 20 10 + + 0
35.0
1
5
10
+ +
TSTEPNL is a SOL 700 bulk data entry which describes the number of Time Steps (10) and Time Increment (35 milliseconds) of the simulation. The total time is the product of the two entries. Notice here the Time Increment is only for the first step. The actual number of Time Increments and the exact value of the Time Steps is determined by SOL 700 during the analysis. The time step is a function of the smallest element dimension during the simulation.
PARAM,OGEOM,NO PARAM,AUTOSPC,YES PARAM,GRDPNT,0 PARAM,dydtout,5.0
PARAM, DYDTOUT is a SOL 700 bulk data entry which describes the time interval for d3plot outputs.
PARAM*,DYTSTEPDT2MS,5e3
PARAM*,DYTSTEPDT2MS is a SOL 700 bulk data entry card. When it is specified, mass scaling will be activated. The value given will be the minimum time step with which the model will run. If the stable time step of an element is smaller than this value, its density will be increased.
Bulk data entry that defines Contact relations and Contact bodies: BCTABLE 1 SLAVE 4 0. 0 full 1 0. 0
1 0 0.3
0. 0 0.0
0.7
0.
0
0.
+ 0 SLAVE 2 0 0.3 MASTERS 3 0. 0 0.0 0.7 0. 0 0.
Main Index
440 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Crash
BCTABLE is a SOL 700 bulk data entry that is also used by SOL 600 and it is meant to define MasterSlave as well as self contact.
BCBODY BCBODY BCBODY
1 2 3
3 3 3
DEFORM DEFORM DEFORM
1 2 3
0 0 0
BCBODY is a bulk data entry that is used by MD Nastran Implicit Nonlinear (SOL 600) and SOL 700
only, which defines a flexible or rigid contact body in 2D or 3D. $ Single Surface Contact BCPROP,1,585,772,586,587,634,635,636,+ .. .. BCPROP,2,585,731,1079,4915,4916,4364,4368,+ .. .. BCPROP,3,4381,4382
BCPROP is a bulk data entry that is used by MD Nastran Implicit Nonlinear (SOL 600) and SOL 700
only, which defines a 3D contact region by element properties. All elements with the specified properties define a contact body. Bulk data entry that defines grid points and elements: GRID 1 12750. 586.2983.919 .. .. GRID 117820 6002. 1145.5 950. .. .. $DBLOCK FS_AB_SIDEPOSTS $ CQUAD4 ELEMENTS IN PART  FS_AB_SIDEPOSTS (PID = 584) $ CQUAD4 1914 584 6594 6597 6544 6545 .. .. Bulk data entry that defines concentrated mass at grid points: $DBLOCK Added Masses $ CONM2 ELEMENTS IN PART  Added Masses (PID = 5242) CONM2 101538 6729 1.972 Ma .. .. CONM2 107566 104763 30000. Ma
Added
Added
Main Index
Chapter 13: Examples 441
Crash
Bulk data entry that defines boundary conditions along the center line: $ SPC BCS IN BCSET  Nodal SPC (SID = 2) SPC 2 11 246 0.0 .. .. SPC 2 104764 23456 0.0 Bulk data entry that defines material properties. MATD024,59,7.9006 ,200.000, 0.30, 0.340,,,,+ +,,,10 .. ..
MATD024 is a SOL 700 bulk data entry that is used to model an elastoplastic material with an yield stress
versus strain curve and arbitrary strain rate dependency. Failure can also be defined based on the plastic strain or a minimum time step size. MATD001, 61,7.8006 ,200.000, 0.30, .. ..
MATD001 is a SOL 700 bulk data entry that is used to model an isotropic elastic material available for
beam, shell, and solid elements. MATD020,62, 7.8006, 200.000, 0.30,,,,,+ +1.0,0,0 .. ..
MATD020 is a SOL 700 bulk data entry that is used to model rigid materials.
Bulk data entry that defines properties for bar, elastic and shell elements: PBAR .. .. PELAS Section_ .. .. PSHELL FS_FR_FR 687 5235 61 5.0 156. 24760. 24760.
5241
60
3.
60
60
Main Index
442 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Crash
Bulk data entry that defines tables: TABLES1,5,,,,,,,,+ +,0.0,3.494E1,5.00E02,4.58E01,2.0E01,5.749E1,3.02E01,6.119E1,+ +,4.0E01,6.33E01,ENDT .. .. Bulk data entry that defines initial velocities for train model: TIC,1,1,1,,30.559 .. .. TIC,1,117820,1,,30.559
TIC is a bulk data entry that defines values for the initial conditions of variables used in structural transient analysis. Both displacement and velocity values may be specified at independent degrees of freedom.
Bulk data entry that defines spotweld with failure: CSPOT,26148,26148,,,1e+20 ,0.5 .. .. CSPOT,48348,48348,,,1e+20 ,0.5
CSPOT is a SOL 700 bulk data entry and was used to define spotweld with failure. The number of a specific SET defined in the executive control section was referred in the entry.
Bulk data entry that defines spotweld with failure: CSPOT,1099,1099,,,40,100,1,1 .. .. CSPOT,84302,84302,,,40,100,1,1
CSPOT is a SOL 700 bulk data entry and was used to define spotweld with failure. The number of a specific SET defined in the executive control section was referred in the entry.
End of input file.
Main Index
Chapter 13: Examples 443
Crash
Results
t = 0.0 seconds
t = 0.24 seconds
t = 0.12 seconds
t = 0.35 seconds
Main Index
444 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Airbags and Occupant Safety
Airbags and Occupant Safety
Simulation of Multicompartment Airbag
Description The purpose of this example is to demonstrate the simulation of an multicompartment airbag in MD R2 Nastran SOL 700. Multicompartment airbag analysis is a new capability that is introduced in MD R2 SOL 700. New entries (AIRBAG, GRIA, EOSGAM) are added in Bulk Data entries to support the capability. SOL 700 Entries Included SOL 700 TSTEPNL PARAM, DYDTOUT AIRBAG EOSGAM GRIA Analysis Scheme
Airbag Model (BDF)
MD Nastran SOL 700
Obtain Binary Results • Deformation (AIRBAG) • CFD results (GAS)
Main Index
Chapter 13: Examples 445
Airbags and Occupant Safety
Model There are five compartments in the airbag as shown in the figure below. These compartments were folded, and each compartment was connected to the gas supply bag through a large hole. An inflator was modeled next to the gas supply bag. The gas jet is initiated from the inflator and running into the gas supply bag. Fixed boundary conditions were applied to the brackets attached to the gas supply bag. The simulation time was 0.02 seconds. Unit used for this analysis was kg for weight, meter for length, second for time, Kelvin for temperature.
Fix
Gas supply bag
Compartment
Inflator
Input
AIRBAG + + + + + + + + + , 1 ,CFD ,INITIAL ,INFLATOR , ,LARGHOLE ,LARGHOLE ,LARGHOLE ,LARGHOLE ,LARGHOLE , 25 , , , 3 , , 1.527 , 101325., 293. , 1.4 , 1001 , 1 , 350. , 1.557 , 243. , , 301 , 2 , , 302 , 3 , , 303 , 4 , , 304 , 5 , , 305 , 6 , , , ,0.009 ,0.009 , 294. , , , , , , 1.0 , , 1.0 , , 1.0 , , 1.0 , , 1.0 , ,0.009 , ,0.7 , , , , , , , , , , , , , , ,+ ,+ ,+ ,+ ,+ ,+ ,+ ,+ ,+
Detail for AIRBAG card was described below:
AIRBAG , 1 + ,CFD , 25 , 3 , , , , 1.527 , ,0.009 , ,0.009 , ,0.009 , , ,+ ,+
Within the CFD keyword line, CFD related data was defined. Gamma low equation of state was defined referring to the EOSGAM card as shown below.
+ ,INITIAL , 101325., 293. , 1.4 , 294. , , , ,+
Main Index
446 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Airbags and Occupant Safety
Within the INITIAL keyword line, initial conditions of gas property inside an airbag were defined. Initial pressure was 101,325 N ⁄ m 2 , initial temperature was 293 K, initial gamma gas constant was 1.4, initial R gas constant was 294m 2 ⁄ s 2 ⁄ K .
+ + ,INFLATOR , , 1001 , 1.557 , 1 , 243. , 350. , , , , , ,0.7 , , , ,+ ,+
Within the INFLATOR keyword line, gas property from an inflator was defined. Mass flow rate was defined referring a table data (TABLED1). Temperature of inflowing gas was 350 K, a scale factor of available inflow area was 0.7, and the gamma gas constant of the inflator gas was 1.557, and the R gas constant of the inflator gas was 243 m 2 ⁄ s 2 ⁄ K
+ ,LARGHOLE , 301 , 2 , , 1.0 , , , ,+
Within the LARGEHOLE keyword line, the airbag into which gas flowed was defined. Here, the airbag ID 2 was the airbag to which this airbag gave gas. A scale factor of available inflow area was 1.0 Each compartment airbag must be defined with AIRBAG card. The reference density must be the same for all AIRBAGs that were defined in one simulation.
AIRBAG AIRBAG + + , 1 , 2 ,CFD ,INITIAL , , , , 25 , , 35 , , 3 , , 1.527 101325., 293. , 1.4 , , , , ,0.011 ,0.011 , 294. , , , ,0.011 , , , , , ,+ ,+ ,+ ,
EOSGAM,3,1.517,226.4
The new card which describes gas state was provided. EOSGAM was the card which defined the gamma low gas equation of state where the pressure p was defined as:
p = (γ – 1) ⋅ ρ e
where
γ
was a constant,
e
was specific internal energy per unit mass,
226.4m 2 ⁄ s 2 ⁄ K
ρ
was overall material density.
A gamma constant was 1.517, R gas constant was
GRIA GRIA GRIA GRIA ... 1 2 61 62
in this model.
.0009375.626128 .0009375.626128 .0005000.414100 .0005000.339100
.230000 .220000 .450000 .450000
GRIA card is a SOL 700 Bulk Data entry which defines the final unstretched configuration of a deployed bag. All IDs of GRIA cards must exist in GRID cards and the same as the IDs of GRID cards.
Main Index
Chapter 13: Examples 447
Airbags and Occupant Safety
Result There are two types of results files, ARC and d3plot. The ARC file which is the original MSC.Dytran result file format is for the results of Euler elements (fluid) and d3plot file which is the native LSDYNA result file format. The deformation results between SOL 700 and MSC.Dytran are compared to be identical. These results were from the binary result files (ARC file for both cases). Deformation Result of Airbag (d3plot) Time 0.000 MD R2 Nastran SOL 700 MSC.Dytran 2005r3
0.002
0.004
Main Index
448 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Airbags and Occupant Safety
Time 0.006
MD R2 Nastran SOL 700
MSC.Dytran 2005r3
0.008
0.010
Main Index
Chapter 13: Examples 449
Airbags and Occupant Safety
Time 0.0148
MD R2 Nastran SOL 700
MSC.Dytran 2005r3
0.020
Euler Regions for Multicompartment (ARC) Time 0.020 MD R2 Nastran SOL 700 MSC.Dytran 2005r3
Main Index
450 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Airbags and Occupant Safety
Airbag with Dummy
Description Automotive companies perform crash simulations including airbags and dummies to predict the forces that would be exerted on the passenger. For people of average size the airbag can be computed using a uniform gas bag method. In other cases the airbag can hit the passenger before the airbag is fully deployed. Then the flow is not uniform yet and to get an accurate force prediction the flow inside the airbag has to be computed by a CFD approach. The following simulation demonstrates the CFD capability of the SOL 700 to predict the interaction of the airbag and the occupant model during a crash event. SOL 700 Entries Included SOL 700 TSTEPNL PARAM, DYDTOUT AIRBAG BCTABLE BCBODY BCGRID BCPROP BCSEG BSURF EOSGAM MATD001 MATD009 MATD020 SPC1 SPCADD SPCD2
Main Index
Chapter 13: Examples 451
Airbags and Occupant Safety
Model
The model has a total of 10868 grid points and 9594 elements as follows:
•
3 Bars
• 23 Beams • 2078 Quads • 3626 Trias • 3410 Hexas • 454 Pentas
The dummy is modeled by several types of elements and the floor, chair and window are modeled by shell elements. The airbag surface is meshed with membrane elements. To simulate the gas jet inside the airbag, no elements need to be defined because that is done by the adaptive hex mesher of the Eulerian solver of SOL 700. In other words, the adaptive Euler solver automatically creates the elements inside the bag. As the airbag is inflated, these elements are expanded and “adapt” to follow the fabric surface of the bag. This mesher provides an Euler mesh that is large enough to cover the whole airbag membrane but at the same time keeps the elements that are outside the airbag surface limited. Too many empty elements outside the airbag surface increases runtime without benefit. Input Several contacts are defined between airbag, dummy and parts of the car. Input file SOL 700,NLTRAN stop=1
Main Index
452 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Airbags and Occupant Safety
SOL 700 is an executive control entry similar to SOL 600. It activates an explicit nonlinear transient
analysis using MD Nastran solver. SUBCASE 1 TITLE=This is a default subcase. TSTEPNL = 1 SPC = 2 BCONTACT = 1 DISPLACEMENT(SORT1,REAL)=ALL SPCFORCES(SORT1,REAL)=ALL PARAM,DYDEFAUL,DYNA This parameter set defaults to original LSDyna values. The bulk entry section begins with... TSTEPNL 1 50 .0012 1 ADAPT 2 10
TSTEPNL is a SOL 700 bulk data entry which describes the number of Time Steps (50) and Time Increment (0.0012 sec) of the simulation. End time is the product of the two entries. Notice here the Time Increment is only for the first step. The actual number of Time Increments and the exact value of the Time Steps is determined by SOL 700 during the analysis. The time step is a function of the smallest element dimension during the simulation.
Bulk data entry that defines the airbag: AIRBAG 10 25 ON + CFD 1 1.02512 20.0 20.0 + + NONE 11. + INITIAL .101325 297. 1.4 226.45E6 + INFLATOR 82 100 102 + 1.4 286.E6 + $ SMALHOLE PERMEAB 81 OUT OUT 0.9 20 0.8 + 20.0 11. 11. + + + +
AIRBAG is a SOL 700 bulk data entry that instructs SOL 700 to create an airbag using either the CFD
solver or a uniform gas bag method. Here, the CFD solver will be used. Inflow of gas into the airbag is defined by the entries following the INFLATOR key word. Outflow is defined by adding SMALHOLE. Bulk data entries defining materials: MATD001,111,6.E10,6.E1,0.3
MATD001 is a SOL 700 bulk data entry that models an isotropic elastic material. It represents the membrane material.
Main Index
Chapter 13: Examples 453
Airbags and Occupant Safety
EOSGAM,1,1.4,226.45E6
EOSGAM is a SOL 700 bulk data entry that defines the ideal gas inside the airbag.
MATD009,222,1.E12 MATD020,444,7.8E9, 2.1E05,0.3 MATD020,185,7.8E9, 2.1E05,0.3
MATD009 is a SOL 700 bulk data entry that models materials without resistance to deviatoric stresses. MATD020 is a SOL 700 bulk data entry that models rigid objects.
All other MATD** entries model the various parts of the dummy. Bulk data entries entry that defines geometric properties of the various surfaces Surface Airbag membrane Inflator Holes in the airbag Elements 2000132132 2322523514 20655,20656,21461, 21462,22267,22268, 23073,23074 Car Window +frame 3186332102 Floor dummy 3168332132 15678 1 – 9015 2100 (CQUAD4) 2101 (CQUAD4) All other properties. Various element types Grid Points Properties
20001 32690 601613 651663 701713 751763 (CTRIA3) 605 (CTRIA3) 610,660,710,760 (CTRIA3)
Gas inside the airbag Elements created by adaptive Euler solver All properties that are not part of the dummy are defined by the PSHELL1 entry. The dummy is modeled by using many element types and joints: CPENTA, CHEXA , RBJOINT, RBJSTIFF, CBAR, CBEAM, HGSUPPR, CSPR, PSPRMAT, MAT1, and MATRIG.
Main Index
454 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Airbags and Occupant Safety
Bulk data entries that define Contact relations and Contact bodies: BCTABLE*1 * $ $ contact 1 : $ ID = 1 $ TITLE = CHEST TO RIB $ * SLAVE 1 * 0.9000000 * * * 0.9000000 * * 0.1000000 * * * 2 * * * 5.000000 * 1 * 7 * *
SS2WAY 1.000000 1 20.00000 5.000000 1.000000 1.0000000E+20 1.000000
YES
* * * * * * * * * * * * * * *
BCTABLE is a SOL 700 bulk data entry that is also used by SOL 600 and it is meant to define Master
Slave as well as self contact. $ Deform Body Contact LBC set: airbag BCBODY 21 26 BCPROP 3 26 601 602
0 2119
2120
2121
2122
BCBODY is a bulk data entry that is used by SOL 600 and SOL 700 only. It defines a flexible or rigid contact body in 2D or 3D. It could be specified with a BSURF, BCBOX, BCPROP, or BCMATL entry. BPROP is a SOL 700 bulk data entry that is used by SOL 600 and SOL 700 only, which defines 3D
contact regions by element properties. $ Deform Body Contact LBC set: plate BCBODY 4 3D DEFORM 4 BSURF 4 2801 2802 2803 .. ..
0 2804
2805
2806
2807
BSURF is a SOL 700 bulk data entry that is used by SOL 600 and SOL 700 only. It defines a contact
surface or body by element IDs. All elements with the specified IDs define a contact body.
Main Index
Chapter 13: Examples 455
Airbags and Occupant Safety
BCSEG* *
1 3400
1
2680
3399
*
BCSEG is a bulk data entry that is used by SOL 600 and SOL 700 only. It specifies grid points to be used
in a contact analyses. Using the BCTABLE and several BCBODY, BCSEG, and BCSURF entries the following contacts are defined Slave 1 Chest 2 neck 3 Rib 4 Rib 5 Neck 6 Shoulder belt 7 Lap 8 Airbag 9 Airbag 10 Dummy 11 Dummy shoes rib Self contact shoulder Jacket cable chest pelvic Self contact Dummy front Chair Ground Master
Bulk data entries that defines output at specific positions $ Number of Seatbelt elements = 6 $ $ to compare simulation results with actual measurements ACCMETR*1 9001 8316 8317 * 0 0 .. .. $ five more
*
Main Index
456 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Airbags and Occupant Safety
Bulk data entry that defines constraints: $ constraint for inflator SPCADD 2 1 2 5 6 $ Displacement Constraints of Load Set : spc1.1 SPC1 1 123456 20253 20254 20255 20256 .. .. Frame chair floor SPCD2, +, 0., SPCD2, +, 0., SPCD2, +, 0., SPCD2, +, 0., SPCD2, +, 0., SPCD2, +, 0., 2,RIGID,MR185,1,2,101, 10. 2,RIGID,MR185,2,2,101, 10. 2,RIGID,MR185,3,2,101, 10. 2,RIGID,MR185,5,2,101, 10. 2,RIGID,MR185,6,2,101, 10. 2,RIGID,MR185,7,2,101, 10. MR2 3923 8465 8297 8319 0.,,+ 0.,,+ 0.,,+ 0.,,+ 0.,,+ 0.,,+ * * *
20257
20258
RBE2A* 2 * * 3922 * 8463 End of input file.
Main Index
Chapter 13: Examples 457
Airbags and Occupant Safety
Results t = 0 sec
t = 0.02999 sec
Main Index
458 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Airbags and Occupant Safety
t = 0.0444 sec
t = 0.0444 sec
Main Index
Chapter 13: Examples 459
Bird Strike and Fan Blade Out (FBO)
Bird Strike and Fan Blade Out (FBO)
Bird Strike Simulation on Composite Glass Panel
Description Aerospace companies perform bird strike test simulation to predict the impactresistance properties of the aircraft structure. This is an example of a 3.8 lbs bird, impacting against the composite glass panel of an aircraft canopy. The bird’s velocity is 7874 in/s (447 m.p.h.). SOL 700 Entries Included SOL 700 TSTEPNL PARAM, DYDTOUT BCTABLE BCBODY BCPROP BSURF PCOMP MATD054 MATD010 PSOLIDD EOSPOL Model To model the bird, a cylinder with 2.42 inch radius and 6 inch length was built. Material model MATD010 was used to simulate the elasticplastic hydrodynamic properties of the bird with parameters as follows:
ρ = 9E5lbmass/in 3 σ y = 2.9 psi E h = 0.145 psi G = 145 psi
The relation between density and pressure ( P ) of the bird material was specified through an equation of linear polynomial:
P = a eμ 3 + ( b0 + b 1μ + b2 μ 2 + b3 μ 3 )ρ 0 E μ = ρ ⁄ ρ0 – 1 ρ 0 = reference density a1 = 0.0 a3 = 4.25 E6 psi ρ = overall material density E = specific internal energy per unit mass a2 = 0.0 b 1 = b2 = b3 = 0.0
Main Index
460 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
To model the composite glass canopy, a curved composite plate was built. The composite plate was modeled using PCOMP and MATD054 material. A picture of the model is shown below.
The model has a total of 7082 grid points and 6200 elements as follows:
• 2000 Quads • 4200 Hexas
The glass plate was modeled with composite shell elements. The bird was modeled with solid elements. The relative speed of the bird to the airplane is modeled by defining an initial velocity of 7874 in/s (447 m.p.h.), applied on all the grid points of the bird. The simulation time is 0.002 seconds. The unit system is inch, lbforce, second. Input All nodes of the bird have an initial velocity specified by the TIC entry. Nodes on the straight edges of the curved plate have translational constrains in all directions. Contacts are defined between the plate (master) and the bird (slave). Input File: SOL 700,NLTRAN path=3 stop=1
SOL 700 is a executive control that activates an explicit nonlinear transient analysis.
Main Index
Chapter 13: Examples 461
Bird Strike and Fan Blade Out (FBO)
$ Direct Text Input for Executive Control CEND TITLE = JOBNAME IS: BIRD_STRIKE2 SUBCASE 1 $ Subcase name : Default SUBTITLE=Default TSTEPNL = 1 BCONTACT = 1 SPC = 2 IC = 1 DISPLACEMENT(SORT1,REAL)=ALL SPCFORCES(SORT1,REAL)=ALL
The bulk entry section starts… $ BULK DATA SECTION BEGIN BULK INCLUDE final_scale2.bdf TSTEPNL 20 100 2e5 1
ADAPT
2
10
TSTEPNL is a SOL 700 bulk data entry which describes the number of Time Steps (100) and Time
Increment (2e5 seconds) of the simulation. The total time is the product of the two entries. Notice here the Time Increment is only for the first step. The actual number of Time Increments and the exact value of the Time Steps is determined by SOL 700 during the analysis. The time step is a function of the smallest element dimension during the simulation. PARAM,DYDTOUT,4e4
PARAM, DYDTOUT is a SOL 700 bulk data entry which describes the time interval of d3plot outputs.
Bulk data entry that defines Contact relations and Contact bodies: $ Define contact BCTABLE 1 SLAVE 3 0
0. 0
1 0. 0
0.
0.
0
0.
MASTERS 4
BCTABLE is a SOL 700 bulk data entry that is also used by SOL 600 and it is meant to define MasterSlave as well as selfcontact.
Main Index
462 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
$ Deform Body Contact LBC set: bird BCBODY 3 3D DEFORM 3 BCPROP 3 2
0
BCBODY is a bulk data entry that is used by SOL 600 and SOL 700 only and defines a flexible or rigid contact body in 2D or 3D. It could be specified with a BSURF, BCBOX, BCPROP or BCMATL entry. BCPROP is a bulk data entry that is used by SOL 600 and SOL 700 only. It defines a 3D contact region
by element properties. All elements with the specified properties define a contact body. $ Deform Body Contact LBC set: plate BCBODY 4 3D DEFORM 4 BSURF 4 1 2 3 .. ..
0 4
5
6
7
BSURF is a SOL 700 bulk data entry that is used by SOL 600 and SOL 700 only and defines a contact
surface or body by element IDs. All elements with the specified IDs define a contact body. Bulk data entry that defines material and element properties for composite glass plate: PCOMP 1 1 1 1 70. .012 .012 0. 0. 0. YES YES
.012 .012 .012
0. 0. 0.
YES YES YES
2 2
.. ..
PCOMP is a bulk data entry. It was used to define the properties of a 5ply composite material laminate. MATD054 is a SOL 700 bulk data entry that is used to model arbitrary orthotropic materials, e.g., unidirectional layers in composite shell structures. Various types of failure could be specified for the model. This model is only valid for thin shell elements.
Main Index
Chapter 13: Examples 463
Bird Strike and Fan Blade Out (FBO)
Bulk data entry that defines material and element properties for bird: $ Material Record : mat1.1 MATD054 1 2.334 1.16E7 0.223 4.74E6 4.74E6 4.74E6 ,,,,,,,,, ,,,,,,,,, ,,,,,,,,0.02, 42000. 42000. 42000. $ Material Record : mat1.2 $ Description of Material : MATD054 2 9.535 2812 0.48 950. 950. 950. ,,,,,,,,, ,,,,,,,,, ,,,,,,,,2, 281.2 281.2 281.2 .. ..
1.16E7
1.16E7 0
0.223
0.223
42000. 2812
1035. 2812 0 0.48 0.48
281.2
95
MATD010
3
9E5
145
2.9
0.145
MATD010 is a SOL 700 bulk data entry and is used to model elasticplastic hydrodynamic material.
$ Elements and Element Properties for region : bird_prop PSOLIDD 2 3 1 1 EOSPOL 1 0.0 0.0 4.25E6
PSOLIDD is a SOL 700 bulk data entry that defines properties for solid elements. It refers to entry MATD010 for material properties, and refers to entry EOSPOL for relation between the density and
pressure of the hydrodynamic material.
EOSPOL is a SOL 700 bulk data entry that defines the properties of a polynomial equation of state.
Main Index
464 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
Bulk data entry that defines geometric properties of the model: $ Pset: "plate_prop" will be imported as: "pshell.1" CQUAD4 1 1 1 2 13 12 .. .. $ Pset: "bird_prop" will be imported as: "psolid.2" CHEXA 2001 2 2115 2133 2092 2093 2253 22542255 .. .. $ Nodes of the Entire Model GRID 1 11.2583 4.725776. .. Bulk data entry that defines boundary conditions and initial velocity of the bird: $ Loads for Load Case : Default SPCADD 2 1 $ Initial Velocities of Load Set : bird_vel TIC 1 2092 1 4609.88 TIC 1 2092 3 3601.6 .. .. $ Displacement Constraints of Load Set : fixed_end SPC1 1 123456 1 THRU 11 .. .. $ Referenced Coordinate Frames CORD2R* 1 5.12207 * 4.0018 2.128675 0. 1.1203 0. 9.12389
0.
2256
0. 2.724585
TIC is a bulk data entry that defines values for the initial conditions of variables used in structural transient analysis. Both displacement and velocity values may be specified at independent degrees of freedom.
End of input file.
Main Index
Chapter 13: Examples 465
Bird Strike and Fan Blade Out (FBO)
Results
t = 0.0 seconds
t = 0.0008 seconds
t = 0.0016 seconds
t = 0.002 seconds
Main Index
466 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
Bird Strike on Rotating Fan Blades with Prestress
Description Aerospace companies perform bird strike test simulation to predict the impactresistance behavior of the aircraft engines. This is an example of a bird made by solid elements impacting against rotating fan blades. The fan blades and the rotor are initially prestressed due to the high rotational velocity by the implicit solver prior to the transient run that is simulated by the explicit solver. SOL 700 Entries Included SOL 700 TSTEPNL PARAM, DYDEFAUL, DYNA DYPARAM*, LSDYNA PRESTRS ISTRSSH SPCD2 TIC3 BCTABLE BCBODY BCGRID BCPROP EOSPOL HGSUPPR MATD010 MATD024 MATDERO PSOLIDD RFORCE TIC TLOAD1 Model The models for implicit and explicit runs are basically the same except that in the implicit run the bird model is not included. The hourglass control and definition of rotational velocities between implicit and explicit runs also differ. The rotor, hub and fan blades are modeled by shell elements while the bird is modeled by solid elements. The rotational speed of blades and rotor is 8000 rpm which is applied using RFORCE card (rotational static force) in the prestress run, and TIC3 card (rotational initial speed) and SPCD2 cards in bird strike run. The speed of the impacting bird to the fan blades is assigned by an initial velocity of 7692 inch/s (437 mph), applied on all the grid points of the bird. The simulation time is 0.004 seconds. The unit system is inch, lbf and second.
Main Index
Chapter 13: Examples 467
Bird Strike and Fan Blade Out (FBO)
The thickness of fan blades varies from 0.02659 to 0.40227 inch depending on the location of blade element. The radius of fan blades is 13.6 inch and that of rotor is 0.9 inch. The length and radius of bird model are 6.24 and 2.36 inches respectively.
8000 rpm
Fully fixed Fixed (x,y direction) (x,y rotation)
(a) Prestress Model (Implicit)
8000 rpm (initial speed)
8000 rpm (enforced speed)
Fixed (x,y,z direction) Fixed (x,y direction)
437 mph
(b) Bird Strike Model (Explicit)
Two material models are used in the analysis. All blades and rotors are elasticplastic material and modeled by MATD024. The material properties are:
ρ = 4.14e – 4 lbf/inch 3 – s 2 ⁄ inch Poisson’s
ratio = 0.35 modulus = 100000 psi
σy = 138000 psi
σ y = 138000 psi Tangent
Young’s modulus = 1.60E+7 psiPlastic strain failure limit 0.2 To model the bird, a cylinder with 2.42 inch radius and 6 inch length was built. Material model MATD010 was used to simulate the elasticplastic hydrodynamic properties of the bird with the following properties:
ρ = 9 E5 lbf/inch 3 – s 2 ⁄ inch G σ y = 2.9 psi Tangent
= 145 psi
modulus = 0.145 psi
Main Index
468 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
The relation between density and pressure (P) of the bird material was specified through an equation of linear polynomial:
P = a 1μ + a2 μ 2 + a3 μ 3 + ( b 0 + b1 μ + b2μ 2 + b 3μ 2 ) ρo E μ = ρ ⁄ ρo – 1 ρ = overall material density ρ o = reference density E a1 = 0.0 a2 = 0.0 a3 = 4.25 E6 psi b 1 = b 2 = b3 = 0.0
= specific internal energy per unit mass
The model used for bird strike analysis (explicit) has a total of 20724 grid points and 19055 elements as follows: NUMBER OF CHEXA ELEMENTS = 10752 NUMBER OF CQUAD4 ELEMENTS = 8227 NUMBER OF CTRIA3 ELEMENTS = 76 Input The simulation consists of two runs. The first run is a prestress analysis that computes the deformations and stresses due to rotation. This computation is essentially static using implicit solver. Conducting this computation in the impact run would require structural relaxation. To speed up the process, the static part of the computation is done by the double precision version of the implicit solver in the prestress run. Boundary conditions and initial conditions of the prestress run differ from the second run. In the prestress run the rear of the rotor is fixed and a force in circumferential direction is applied to the rotor and fan blades. In the impact run the rear of the rotor is given an angular velocity and the bird is given an initial velocity. Contacts are defined between the fan blades and the bird. SOL 700,NLTRAN STOP=1 PATH=3
SOL 700 is a executive control entry that initiates the explicit nonlinear solver in the MD Nastran. PATH is set to 3 because a different solver must be assigned for the implicit run (double precision solver.)
First, consider the prestress input deck. In this input deck, only the rotor and fan blades are included.
Main Index
Chapter 13: Examples 469
Bird Strike and Fan Blade Out (FBO)
The bulk entry section begins with LOADSET = 2 TITLE = MD Nastran job created on 18Jan07 at 10:58:50 $ Direct Text Input for Global Case Control Data SUBCASE 1 $ Subcase name : Default SUBTITLE=Default TSTEPNL = 1 DISPLACEMENT(SORT1,REAL)=ALL SPCFORCES(SORT1,REAL)=ALL DLOAD = 3
TSTEPNL $
1
5
1.5
1
ADAPT
2
10
TSTEPNL is a SOL 700 bulk data entry which describes the number of Time Steps (5) and Time Increment (1e5 sec) of the simulation. End time is the product of the two entries. Notice here the Time Increment is only for the first step. The actual number of Time Increments and the exact value of the Time Steps are determined by MD Nastran SOL 700 solver during the analysis. The time step is a function of the smallest element dimension during the simulation.
By using the SOL 700 bulk data entry PRESTRS a prestress analysis is carried out. Here the implicit SOL 700 solver is used. This solver requires the analysis to be run with double precision executable. Final deformations and stresses of elements are written to a text file named imput_file_name.nastin to provide initial conditions for rotor and fan blades of the impact run. Bulk data entries that define properties for shell elements. PSHELL1 135 3001 + 0.05440 KeyHoff Lobatto 5 Mid +
These elements model the fan blades and rotor. Bulk data entry that defines material properties for rotor and fan blade
MATD024 1 + 4.14E4 1.60E+7 0.35000 100000. 138000. 0.99000 3001 +
MATD024 is a SOL 700 bulk data entry that represents MD Nastran Material #24. It is used to model
elastoplastic material.
Main Index
470 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
Bulk data entry that defines geometric properties of the rotor and blades: CQUAD4 .. GRID .. 6062 1 301 300425 1.0762 300426 4.6577 300446 1.945 300445
Bulk data entry that defines boundary conditions and loads: TLOAD1 LSEQ DLOAD RFORCE TABLED1 SPC1 SPC1 SPC1 5 2 3 1 321 0. 1 1 1 6 6 1. 299999 321 1 1. 5 133.3330.0 ENDT 300443 0.0 1.
1. .001 1. 123456 300425 THRU 1245 400058 1245 400115
RFORCE is a bulk data entry that defines a static loading condition due to angular velocity.
The following parameters are needed to get optimal results. DYPARAM,LSDYNA,ACCURACY,OSU,1 DYPARAM,lSDYNA,ACCURACY,INN,2 $ $ control_ENERGY $ DYPARAM,LSDYNA,ENERGY,HGEN,2 DYPARAM,LSDYNA,ENERGY,RWEN,1 DYPARAM,LSDYNA,ENERGY,SLNTEN,1 DYPARAM,LSDYNA,ENERGY,RYLEN,1 $ $ control_HOURGLASS $ DYPARAM,LSDYNA,HOURGLASS,IHQ,1 DYPARAM,LSDYNA,HOURGLASS,QH,0.100000001 DYPARAM,LSDYNA,DATABASE,D3PLOT,0.1 PARAM,DYDEFAUL,DYNA
Main Index
Chapter 13: Examples 471
Bird Strike and Fan Blade Out (FBO)
Now consider the bird strike run. TITLE = MD Nastran job created on 18Jan07 at 10:58:50 $ Direct Text Input for Global Case Control Data SUBCASE 1 $ Subcase name : Default SUBTITLE=Default TSTEPNL = 1 DISPLACEMENT(SORT1,REAL)=ALL SPCFORCES(SORT1,REAL)=ALL IC = 1 SPC = 1 BCONTACT = 1 The bulk entry starts with: BEGIN BULK include rotor.dytr.nastin TSTEPNL 1 100 .4e4
1
ADAPT
2
10
TSTEPNL is a SOL 700 bulk data entry which describes the number of Time Steps (100) and Time Increment (.4e4 sec) of the simulation. End time is the product of the two entries. Notice here the Time Increment is only for the first step. The actual number of Time Increments and the exact value of the Time Steps are determined by MD Nastran solver during the analysis. The time step is a function of the smallest element dimension during the simulation.
The prestress results file is prestress_rotor.dytr.nastin. The name of this file was changed to rotor.dytr.nastin due to the long file name. It includes the information of grid points, elements and initial conditions. This makes the shell elements that are defined in the prestress run available to the impact run. Because the only properties of the rotor and fan blade that need to be defined in the impact run are material properties and boundary conditions, all other information like GRID, CQUAD, etc must be deleted in the explicit input deck.
ISTRSSH* * * * * * * * * 2275 0.000E+00 7.084E+03 0.000E+00 1.000E+05 1.000E+00 1.096E+04 0.000E+00 1 6.916E+03 3.908E+01 0.000E+00 0.000E+00 1.481E+04 5.499E+00 0.000E+00 5 7.371E+03 1.150E+02 0.000E+00 0.000E+00 1.149E+04 2.964E+03 0.000E+00 5* * 1.480E+02* 0.000E+00* 1.392E+05* 0.000E+00* 1.999E+03* 0.000E+00* 1.392E+05*
The file rotor.dytr.nastin contains the entry ISTRSSH. This entry specifies the prestress condition of the shell element. The information in the upper box are that element number 2275 has (1) inplane integration point and (5) throughthickness integration points followed by (5) additional history variables in the first row. The contents from next row to the end include the initial conditions of this element such as the initial
Main Index
472 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
stress in xx direction, yy direction and so on. These result variables of the prestress run are to be carried over to the impact run. If other elements types instead of shell are used, ISTRSBE, ISTRSTS, and ISTRSSO cards must be included in nastin file. Bulk data entries that define Contact relations and Contact bodies: BCTABLE 1 SLAVE + + + MASTERS 1001 SLAVE 1001 0. 0. 0.1 0. 0 2 0. 0
8001 0 0.1
0. 0
0.1 YES
0.
0
BCTABLE is a SOL 700 bulk data entry and it is meant to define MasterSlave as well as self contact. In
this analysis, adaptive contacts between the bird and the fan blades are used. BCBODY BCPROP 1001 1001 3D 1 DEFORM 2 1001 3 0 4
5
6
7
BCBODY is a bulk data entry that is used by SOL 600 and SOL 700 only, which defines a flexible or rigid contact body in 2D or 3D. It could be specified with a BSURF, BCBOX, BCPROP, or BCMATL entry. In this analysis, BCPROP cards are used because adaptive contact is applied.
Bulk data entries that define material and properties for the shell as same as the prestress run: MATD024 3001 4.14E4 1.60E+7 0.35000 100000. 138000. 0.20000 + + 3001 ... PSHELL1 1 3001 KeyHoff Lobatto 5 Mid + + 0.22768 ...
HGSUPPR 1 ...
SHELL
1
1
HGSUPPR card is for the definition of the hourglass suppression method, the corresponding hourglass
damping coefficients and sets for the bulk viscosity method and coefficients. Bulk data entry that defines boundary conditions and initial velocity: BCGRID 1 300425 THRU 300443
Main Index
Chapter 13: Examples 473
Bird Strike and Fan Blade Out (FBO)
BCGRID is also used to specify a set of nodes which are to be constrained with SPCD2 (angular velocity).
SPCD2 1 GRID 1 7 80 TABLED1 80 + 0.0 837.758 1. 837.758 ENDT $ Displacement Constraints of Load Set : Disp1 SPC1 1 3 21 THRU 31 .. .. $ Initial angular velocity for rotor +fan blade TIC3 1 299999 1. 837.758 1 THRU ... THRU 6384 300000 THRU 300018
1. +
300020
TIC3 defines the initial velocity as combination of translational part and a rotational part. SPCD2 defines imposed nodal motion on a node or a set of nodes.
In bird1.dat, bulk data that defines the bird properties PSOLIDD 3004 EOSPOL 3004 MATD010 3004 3004 0.0 9E5 1 0.0 145 3004 4.25E6 2.9
0.145
PSOLIDD is a SOL 700 bulk data entry that defines properties for solid elements. It refers to entry MATD010, EOSPOL for material properties. MATD010 is a SOL 700 bulk data entry that represents MD Nastran Material #10, which is used to model
elasticplastic hydrodynamic material.
EOSPOL is a SOL 700 bulk data entry that defines the properties of a polynomial equation of state. MATERO is used to assign a failure criterion on material property.
Bulk data entry that defines geometric properties of the bird: CHEXA 1000006 .. GRID .. TIC TIC .. .. 1000001 3004 1000001 1000002 1000003 1000004 1000005 1000007 1000008 1000001 1 1 2.35785 10.7693 17.238 1000001 3 1000002 3 7692. 7692.
Main Index
474 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
TIC is a bulk data entry that defines values for the initial conditions of variables used in structural transient analysis. Both displacement and velocity values may be specified at independent degreesoffreedom.
Results Prestress Run Using the TSTEPNL entry 5 results increments were archived. The results of all increments are essentially the same which indicates that the implicit calculations are stable. The results of the last increment were written to the file prestres_rotor.dytr.nastin. Result Increment 2
Main Index
Chapter 13: Examples 475
Bird Strike and Fan Blade Out (FBO)
Result Increment 5: written to the .nastin file
Main Index
476 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
Impact run The prestress result variables have been initialized at the begin of the analysis (Time = 0)
t = 0 seconds
t = 0.001 seconds
Main Index
Chapter 13: Examples 477
Bird Strike and Fan Blade Out (FBO)
t = 0.00152 seconds
t = 0.002 seconds
Main Index
478 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
t = 0.003 seconds
t = 0.004 seconds
Main Index
Chapter 13: Examples 479
Bird Strike and Fan Blade Out (FBO)
Multiple Birdstrikes on a Box Structure
Problem Description Bird strike on a box structure is a typical problem in aircraft industries. The box structure simulates the leading edge of lifting surfaces, e.g. wing, vertical, and horizontal stabilizers. The box can be simplified to consist of a curve leading edge panel and a front spar. The acceptable design criteria for bird strike are that the leading edge panel may fail but the front spar strength may not degrade to a certain level. In this example, two cylindrical panels are put in parallel. Two birds strike the upper panel. One bird strikes in horizontal direction and the second one vertically. The second bird will perforate the first panel and impact the second one. The birds are modeled as cylindrical slugs of jelly. The plate is constrained in such a way that the edges can only move in radial direction.
Figure 131
Initial Situations
The properties and initial conditions of the plate and birds are as follows: Plate Material Density (kg/m3) Bulk modulus (Pa) Poisson’s ratio Yield stress (Pa) Gamma Thickness (m) 0.0015 Titanium 4527 1.03e11 0.314 1.38e8 1.4 Ambience B Air 1.1848 Bird 1 Jelly 930 2.2e9 Bird 2 Jelly 930 2.2e9
Main Index
480 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
Plate Radius (m) Length (m) Mass (kg) Initial velocity (m/s) Fail (equiv. Plas. Strain) SOL 700 Model 0.1 0.25 0.25
Ambience B
Bird 1
Bird 2
0.36 150
0.285 200
Each curved plate is modeled using 33x16 BLTshells. The boundary conditions applied at the edges of the plate are defined within a cylindrical coordinate system, where the local zaxis is aligned with the length axis of the plate. The cylindrical system is defined using a CORD2C entry. To create a closed surface, required by COUPLING option, the two plates are connected with dummy quad elements. The two birds and air are modeled using Multi Material Eulerian (FV) elements, also known as MMHYDRO. The location of the bird in the Euler domain is defined using TICEUL option. The material for the birds and air are modeled using EOSPOL and EOSGAM, respectively. To allow the bird perforating the first plate and impact the second one, several modeling techniques can be used. One of them is using two Eulerian domains and two coupling surfaces. Both the Eulerian domains and the coupling surfaces have to be logically different. Each coupling surface associates with one Eulerian domain. In this model, the two coupling surfaces share the same physical space. By specifying that one domain is covered outside and the other inside, the Eulerian domain represents the correct physical space. The two Eulerian domains cannot interact with each other except through coupling surfaces. When coupling surfaces share the same shell elements with some or all shells failing, then the material can flow from one Eulerian domain into another one. The interaction between the Eulerian domains is activated using COUP1INT option and PARAM, FASTCOUP, INPLANE, FAIL. The rest of the Euler domain is filled with air. Please notice that when the effect of air is neglected, then the rest of the Eulerian domain should be filled with void. It will speed up the analysis. The first domain is associated with a coupling surface that is INSIDE covered. Therefore, it cannot be adaptive and is defined using MESH,, BOX option. The second domain is adaptive and defined using MESH,, ADAPT. The ADAPT option will let SOL 700 create and update the Eulerian domain to minimize memory allocation and consequently lowered CPU time. The default Eulerian boundary condition is set to that only outflow is allowed using FLOWDEF option. In this case, a bird that reaches the free face boundary will flow out of the domain. The initial velocity of the birds is defined using TICVAL option.
Main Index
Chapter 13: Examples 481
Bird Strike and Fan Blade Out (FBO)
The finite element model of the upper and lower plates, the Eulerian domains and the initialization of the birds are shown in the figure below. The dummy quad elements used to create closed coupling surfaces are not shown in this figure.
Input File:
SOL 700 is an executive control that activates an explicit nonlinear transient analysis:
SOL 700,NLTRAN stop=1
Case control cards for problem time, loads and initial conditions: $ Direct Text Input for Executive Control CEND TITLE = Multiple BIRD STRIKE on BOX Structure SUBCASE 1 $ Subcase name: Default SUBTITLE=Default TSTEPNL = 1 SPC = 1 IC = 1
Main Index
482 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
TSTEPNL is a SOL 700 bulk data entry which describes the number of Time Steps (10) and Time
Increment (0.0015 seconds) of the simulation. The total time is the product of the two entries. Notice here the Time Increment is only for the first step. The actual number of Time Increments and the exact value of the Time Steps is determined by SOL 700 during the analysis. The time step is a function of the smallest element dimension during the simulation. $ BULK DATA SECTION BEGIN BULK TSTEPNL 1 10 0.0015 1 Define the Initial, the Minimum and the Safety factor of the time step: PARAM*, DYINISTEP, 1e7 PARAM*, DYMINSTEP, 1e8 DYPARAM, STEPFCTL, 0.9 Define coupling surface that can fail and Multi material overflow array to store material data. In a problem where more than 10% of the elements have more than one material, the default value of FMULTI(0.1) must be increased. DYPARAM, FASTCOUP, INPLANE, FAIL DYPARAM, FMULTI, 0.2 Define Output results request for every 0.00015 s and time history output request for coupling surfaces: DYPARAM, LSDYNA, BINARY, D3PLOT, .00015 DYTIMHS,, .000001,,,,,,,+ +, CPLSOUT Euler domain 1: Define an Euler mesh with 50x28x44 elements reference to PEULER1 (=1): $ domain 1 $ MESH, 1, BOX,,,,,,,+ +,0.26,0.015,0.05,0.50,0.28,0.44,,,+ +, 50, 28, 44,,,, EULER, 1
Main Index
Chapter 13: Examples 483
Bird Strike and Fan Blade Out (FBO)
Define FSI coupling surface from elements listed in the BSURF entry (covering inside): $ COUPLING SURFACE 1 $ COUPLE , 1 , 1 , INSIDE , ON , ON , , , , + + , , , , , , , , , + + , , 1 $ BSURF , 1 , 7393 , THRU , 8448 , 13729 , THRU , 14048 , 14577 , + + , THRU , 15236 Define Eulerian element properties with reference to TICEUL1 (=11). PEULER1 , 1 , , MMHYDRO , 11
The initial conditions of these elements are defined in geometric regions. Define Regions with shapes, material, initial values and level indicators: $ Allocation of material to geometric regions. $ TICEUL1 , 11 , 11 TICREG , 1 , 11 , CYLINDER , 1 , 3 , 1 TICREG , 2 , 11 , CYLINDER , 2 , 5 , 2 TICREG , 3 , 11 , SPHERE , 4 , 4 , Define region shapes:
CYLINDR , 1 , , .13 , .125 , .2252 , .17 , .125 , .2944 , + + , .035 CYLINDR , 2 , , .1381 , .125 , .26 , .2381 , .125 , .26 , + .035 SPHERE , 4 , , .1381 , .125 , .26 , 1000
, , 5
3 2 ,
1
+
Define Initial values of the birds and the air: TICVAL, TICVAL, TICVAL, 1 2 5 , , , , , , XVEL , XVEL , SIE , 75 , ZVEL , 129.9 200 2.1388E5 , DENSITY , 1.1848
Main Index
484 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
Define Eulerian materials for the birds and the environment (air): $Material Bird MATDEUL , 3 , 930 , 3 EOSPOL , 3 , 2.2e9 MATDEUL , 5 , 930 , 5 EOSPOL , 5 , 2.2e9 $  Material Air id =4 MATDEUL , 4 , 1.1848 , 4 EOSG , 4 , 1.4
Euler domain 2: Define an adaptive Euler mesh reference to PEULER1 (=6): $Domain 2$ MESH , 2 , ADAPT , 0.01 , 0.01 , 0.01 , , , , + + , 0.26 , 0.015 , 0.05 , , , , , , + + , , , , , , , EULER , 6
Define FSI coupling surface from elements listed in the BSURF entry (covering outside):
$===Coupling Surface 2 $ COUPLE , 2 , 2 , OUTSIDE , , , , , , + + , , , , , , , , , + + , , 2 $ BSURF , 2 , 7393 , THRU , 8448 , 13729 , THRU ,
14048 ,
14577 , +
Domain 2 has only 1 region with air. TICEUL1,12,12 TICREG,11,12,SPHERE,7,4,5,1.0 SPHERE,7,,0.0,0.0,0.0,500.0 Interaction between the coupling surfaces 1 and 2: Define interaction between coupling surface 2 and 1: $ coupling interaction $ COUPINT,2,2,1
Main Index
Chapter 13: Examples 485
Bird Strike and Fan Blade Out (FBO)
Define default Eulerian flow boundary condition: $ Flow boundary $ FLOWDEF , 1 , , MMHYDRO , , , , , , + + , FLOW , OUT Define cylindrical coordinate system: $ CORD2C , 1 , , 0.0 , 0.0 0.0 , + + , 0.0 , 0.125 , 0.25 Define properties of the panels:
PSHELL1 , 2 , 2 , Blt , Gauss , 3 , + , .0015 $ MATD024 , 2 , 4527 , 1.150e11 , .314 , 1.38e8 , , 0.1 .83333 , Mid , , +
,
0.0
,
0.0
,
0.25
,
Define properties of dummy elements to close the coupling surfaces PSHELL,3,999,1.E3 PSHELL,4,999,1.E3 $ MATD009,999,1.E20 Results In this simulation, the time history of total zforce on the coupling surface is requested as shown in Figure 132. This force is the sum of all zforces on the nodes that belong to both the upper and the lower plate. From this figure, it is obvious that there are three large impact forces occurring on the plate. The first one is when the first bird impacts the upper plate, which is subject to a significant damage. The second one is when the second bird impacts the upper plate. The last peak is caused by the first bird impacting the lower plate. Snapshots of the motion of the two birds and the deformation of the plates are shown in Figure 133 at various time steps of the simulation. Figure 133a is the initial condition. Figure 133b is at the moment when the first bird penetrates the upper plate and second bird touches the plate. This corresponds with the first peak in the time history plot shown in Figure 132. Figure 133c is at the moment when the second bird penetrates the upper plate. It corresponds with the second peak of the time history plot. Figure 133d is at the moment when the second bird has left the plate and the first bird penetrates the lower plate. This corresponds with the third peak in the time history plot.
Main Index
486 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
Figure 132
Time History of Totoal Zforce on Coupling Surface
Figure 133
Deformation of Plates
Main Index
Chapter 13: Examples 487
Bird Strike and Fan Blade Out (FBO)
Files examp4_9_bs.dat examp4_9_bs.bdf examp4_9_bs.dytr.d3plot EXAMP4_9_BS.DYTR_EULER_FV1_*.ARC EXAMP4_9_BS.DYTR_EULER_FV2_*.ARC Abbreviated SOL 700 Input File SOL 700,NLTRAN STOP=1 CEND TITLE = Multiple bird strike using MultiMaterialFVSurfer IC = 1 SPC = 1 TSTEPNL=1 $ BEGIN BULK PARAM*,DYINISTEP,1e7 PARAM*,DYMINSTEP,1e8 DYPARAM,FASTCOUP,INPLANE,FAIL DYPARAM,FMULTI,0.2 Dyparam,stepfctl,0.9 DYPARAM,LSDYNA,BINARY,D3PLOT,.00015 DYTIMHS,,.000001,,,,,,,+ +,CPLSOUT $ TSTEPNL, 1, 10, .00015, 1 $ $ Include model + SPC INCLUDE examp4_9_bs.bdf $ $ domain 1 $ MESH,1,BOX,,,,,,,+ +,0.26,0.015,0.05,0.50,0.28,0.44,,,+ +,50,28,44,,,,EULER,1 $ $ COUPLING SURFACE 1 $ COUPLE, 1, INSIDE, ON, ON, , , , , + +, , , , , , , , , + +, 1 $ BSURF, 1, 7393, THRU, 8448, 13729, THRU, 14048, 14577, + +, THRU, 15236 $ $ Flow boundary, property, material and equation of state data. $ FLOWDEF, 1, MMHYDRO, , , , , , + SOL 700 Euler result file SOL 700 input files SOL 700 structure result file
Main Index
488 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
+, FLOW, OUT $ PEULER1, 1, , MMHYDRO, 11 PEULER1, 6, ,MMHYDRO, 12 EOSGAM,4,1.4 $ $Material Bird MATDEUL, 3, 930, 3 EOSPOL, 3, 2.2e9 MATDEUL, 5, 930, 5 EOSPOL, 5, 2.2e9 $ $============================================================= $ $ Allocation of material to geometric regions. $ TICEUL1 11 11 TICREG 1 11 CYLINDER1 3 1 3 TICREG 2 11 CYLINDER2 5 2 2 TICREG 3 11 SPHERE 4 4 5 1 $ CYLINDR 1 .13 .125 .2252 .17 .125 .2944 + .035 CYLINDR 2 .1381 .125 .26 .2381 .125 .26 + + .035 SPHERE,4,,.1381, .125, .26, 1000 $ $ Initial material data. $ TICVAL 1 XVEL 75 ZVEL 129.9 TICVAL 2 XVEL 200 $ $ LAGRANGE $ $ Property, material and yield model. $ PSHELL1 2 2 Blt Gauss 3 .83333 Mid + .0015 $ MATD024,2,4527,1.150e11,.314,1.38e8,,0.1 $ PSHELL,3,999,1.E3 PSHELL,4,999,1.E3 $ MATD009,999,1.E20 $ $ Boundary constrain. $ CORD2C 1 0.0 0.0 0.0 0.0 0.25 0.0 + 0.0 0.125 0.25 $ $  Material Air id =4 MATDEUL 4 1.1848 4
+
+
+
Main Index
Chapter 13: Examples 489
Bird Strike and Fan Blade Out (FBO)
$  $ > density $ $Domain 2$ TICEUL1,12,12 TICREG,11,12,SPHERE,7,4,5,1.0 SPHERE,7,,0.0,0.0,0.0,500.0 TICVAL,5,,SIE,2.1388E5,DENSITY,1.1848 $ $===Coupling Surface 2 $ COUPLE,2,2,OUTSIDE,,,,,,+ +,,,,,,,,,+ +,,2 $ BSURF 2 7393 THRU 8448 13729 THRU 14048 14577+ + THRU 15236 MESH,2,ADAPT,0.01,0.01,0.01,,,,+ +,0.26,0.015,0.05,,,,,,+ +,,,,,,,EULER,6 $ $ coupling interaction $ COUPINT,2,2,1 $ ENDDATA
Main Index
490 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
Chained Analysis  Fan Blade Out to Rotor Dynamics
Description This example presents a multidisciplinary, integrated implicitexplicitimplicit analysis process tailored for more accurate and efficient simulations of aero engine fan bladeout events using MD Nastran. A FBO event can be extremely nonlinear because of the heavy wide chord fan blades incorporated in the new generation of high bypass ratio jet engines. These new wide chord blades are used to meet airframe manufacturers’ demand for higher thrust engines with improved performance and optimum weight. Airframe and engine manufacturers use computerized analysis procedures to support the design of both the propulsion system and adjacent wing structures. However, the manufactures do not share their finite element models and traditionally resort to construction of a new model to suit their analysis objective. For example, typical FBO models are very detailed and can exceed two or three million elements whereby a rotordynamics models is much coarser and can be under 50,000 elements. So the challenge becomes as how to transfer the FBO loads computed by SOL 700 explicit solver based on a very fine meshed model to a coarse model for rotordynamics simulation, all within one common modeling environment. This example demonstrates the automated, multidisciplinary simulation capability in MD Nastran to streamline the FBO event simulation facilitated by SOL 700 and SOL 400 which normally consists of the following separate steps: 1. Prestress fan blades condition at the maximum rotating speed & static loads such as gravity with implicit solution (SOL 700 implicit solver or SOL 400). 2. Followed by an explicit solution for few cycles with release of a fan blade to simulate: damage to the trailing fan blade(s); fan rubs with the engine case; breakage or damage to the inlet or engine containment case; twisting and bending of the FAN shaft and/or other rotating shafts. 3. And converting to an implicit solution to continue the analysis more rapidly to reach the steady state windmilling speed. This is done by including realistic input forcing functions and damage incurred during the explicit solution. Techniques to reduce the loads for application to a coarser model are introduced to preserve the solution integrity. 4. This integrated SOL 700 explicit and MD Nastran SOL 400 implicit solution with the Nastran rotor dynamics capability is used to predict the engine unbalance and to extract the whirling diagrams and critical tolerances (Figure 134). This allows the engine manufacturers to share results of the explicit phase with other manufacturers of modern airframe/propulsion system components seamlessly and without compromising design secrets, thus achieving higher accuracy and improved productivity with fewer bottlenecks.
Main Index
Chapter 13: Examples 491
Bird Strike and Fan Blade Out (FBO)
Figure 134
Chained FBORD Analysis
All these steps are integrated in one common modeling environment by using MD Nastran. The MD Nastran MASTER database is tailored to include only the FBO loads and other relevant information required for rotordynamic simulation without compromising the confidentiality of model geometry and modeling details. The MASTER database can be shared by MD Natran users among different companies and organizations for followup analysis. For example, after the FBO analysis simulated by SOL 700, the MASTER database can be sent to airframe manufacturers to use the loads in their rotordynamics analysis facilitated by SOL 400. The following steps are completely automated in SOL 400 rotordynamics simulation: 1. Read the MASTER database generated by SOL 700 to use the FBO loads as a precondition to RD analysis. 2. Map the loads on the coarse finite element model in RD simulation. An advanced search technology is implemented in SOL 400 to identify the closest element to a given load. The load is then distributed on the corner nodes of the element.
Main Index
492 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
3. Synchronization of the explicit and implicit timesteps, The explicit timestep is much smaller than the implicit timestep, so the timesteps need to be “synched up” when the analysis is switched from FBO analysis to RD simulation. The synchronization is based on Fast Fourier Transformation (see Timestep Control on SOL 400 for details) It is believed that this process can result in much higher levels of accuracy and dramatically reduce the cost of analysis and design of the propulsion system and wing while still protecting any proprietary information. The example problem that is used in this paper is a representative finite element model of an engine mounted on a wing. SOL 700 Entries Included SOL 700 TSTEPNL DYPARAM,LSDYNA,DATABASE,SSSTATM DBEXSSS DYPARAM,LSDYNA,DATABASE,NCFORC CSPOT BLDOUT SOL400 ANALYSIS=NLTRAN ROTORG RSPINT CONM2 UNBALNC CBUSH PBUSH PBUSHT Loadings Types The dynamic loads on the engine after the FBO can be classified under two categories: a. Large amplitude transient impact loads generated inside the engine due to the released blade hitting the containment and the impact loads generated as a result of contact with the trailing blade(s). b. The socalled “Seizure Torque” being applied on the fan rotor due to unbalance caused by the missing blade. The seizure torque is as a result of rubbing loads caused by contact between the tip of the blades and the fan case. If the torque is large enough it could stall the rotor and the engine causing a “seizure” (see below for more details). The impact type transient loadings are calculated and stored by MD Nastran SOL 700. In this release, only three types of loadings are taken into account. 1. Impact loads between the broken blade and the case 2. Rub loads on fan case 3. Rub loads on blade tips
Main Index
Chapter 13: Examples 493
Bird Strike and Fan Blade Out (FBO)
Additionally, this release is limited to the analysis of only one released blade and assumes that there are no other failed trailing blades. In other words, only the released blade is considered for unbalance. The other types of loads and unbalances, such as impact loads between the broken blade and remaining blades unbalance generated by breaking some of the remaining blades due to the impact between the broken and remaining blades and so on, will be considered in future releases of MD Natran. The impact loads are generated due to released blade(s) “impacting” the containment structure or other components. These forces contain both a normal component (to the fan case) and tangential components which change with time as the blade hits various parts of the containment ring. The released blade, predetermined in the analysis and in testing, is the only blade which is actually released at the hub and impacts the fan case. In many cases, the trailing blade will impact the root of the released blade, causing the trailing blade to fail and break at a different location. As a result, one or more trailing blades will behave like shrapnel and will contribute significantly to the impact loads. These forces and their contact locations are stored in SOL 700 “binout” as well as the MD Nastran database MASTER file in the Nastran basic coordinate system. The new entry BLDOUT in MD Nastran defines blade out force output information and mapping criteria for a combined SOL 700 – SOL 400 Bladeout analysis (used both in the SOL 700 and subsequent SOL 400 analyses). Rub loads occur as a result of fan blade tips coming into contact with the inner surface of the containment ring and/or its trenchfiller materials. During the Fan bladeout event, as the unbalance forces
( M u rΩ )
2
on the rotor make it to go offcenter and the running tipclearance between the rotating blades and the stator structure gets consumed, the tips of the blades will rub against the enclosure. The rubbing loads are distinguished between those that are applied on the fan case and those that are located on the blade tips. There are equal and opposite sets of forces on the containment ring and on the blade tips. The primary difference is that the rubbing loads on the containment ring can be stationary whereas the rubbing loads on the blade tips are varying as the blades rotate and at any given instant different blades on the rotor continue to contact the stator structure in the same location. The rub loads ( F rub ) have radial and tangential components, with radial component
( Fr )
acting along the span of the blade and tangential
component ( F t ) opposite to the direction of motion of the rotor. Using the relationships for tip Coulomb damping with the coefficient of friction μ , the magnitude of the tangential component is computed as:
Ft = μ Fr
. Since the torque on the rotor produced by the frictional force
( Ft )
always opposes the
motion, its tendency would invariably be to slowdown the spin RPM of the rotorshaft. These loads, if severe enough, do not only slow down the rotor rather may even stop it, a phenomenon called “seizure torque”. Thus, rub forces have normal and tangential components ( F r ) and ( F t ) , respectively at the points of contact. Similar to impact forces, MD R3 SOL 700 will compute the contact location and magnitude of the rubbing loads and store them into “binout” as well as the “MASTER” file. The loads that have zero magnitude are filtered out and are not written to the database to save time and disk space.
Main Index
494 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
The unbalance force
( M u rΩ )
2
, which results from the mass
( Mu )
of the missing blade material, occurs
whether the remaining unbroken blades contact the containment ring or not. This force is output by SOL 700 in the Nastran basic coordinate system and saved in the SOL 700 “binout” file. In addition, the mass of the broken blade will be saved for use in the subsequent creation of UNBALNC entries for the SOL 400 rotor dynamics analysis. Time Step Control in SOL 400 The contact forces computed by SOL 700 are stored and transferred to SO 400. These forcing functions have very small time intervals and they may increase the analysis time. In order to increase the timestep and synchronize the explicit and implicit timesteps, a technique based on Fast Fourier Transformation (FFT) and Inverse Fast Fourier Transformation (IFFT) is used to eliminate the high frequencies of the data. First, the time histories from SOL 700 are changed by FFT from time domain to frequency domain. Next, the frequency domain histories are passed through a low pass filter where the low pass frequency can be selected by the user. Finally, the histories are changed by Inverse FFT from frequency domain back to time domain. Because the high frequencies are removed by the low pass filter, the final time histories do not include the high frequency vibration. Load Mapping Scheme
Main Index
Chapter 13: Examples 495
Bird Strike and Fan Blade Out (FBO)
Model
Figure 135
Implicit Prestress Blade and Rotor Model and Location of Bearings on the Rotor
Figure 136
Explicit FBO Engine Model and Location of Bearings on the Rotor
Main Index
496 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
Figure 137
Implicit Rotor Dynamics Model
A simplified generic engine model was provided by Boeing for the purpose of this study. The engine model was modified and enhanced by MSC to include realistic fan blades, rotor, three bearings and other components. Typically, full FBO models can easily have millions of elements and degrees of freedom to represent a realistic jet engine. However, for the purpose of this study, even though the FBO model was constructed with a much finer mesh density than the rotor dynamics model, it is not as elaborate as the full engine models that are used by manufacturers in their explicit simulation. The FBO model has 8864 nodes and 8256 shell elements and is deemed to be sufficiently detailed to capture the physics of the problem and compute the impact and rub loads. The fan blades were constructed by shell elements with various thicknesses across their width and length. The rotor was made of a hollow rod with varied cross sections across its length and a rotational velocity of 4500 rpm. The material for both rotor and fan blades is titanium grade with the following properties: Material: (Titanium) Blade and Rotor (SI Unit) ρ = 4.466 g/cm3 σy = 1009 MPa Young’s modulus = 117 GPa Poisson’s ratio = 0.35 Tangent modulus = 731 MPa Plastic strain failure limit 0.2
The bearings were modeled by constructing two concentric rings with predetermined stiffness properties that can contact each other. The flange on the bearings prevents the axial movement of the rotor during the fanbladeout. The bearing models and their properties are important design considerations to simulate the “fusing” during the FBO and rotor dynamics analysis. Fusing is an event where a bearing or other support structure fails as a result of high loads beyond the design strength of the fusing structure, and its stiffness is reduced to zero.
Main Index
Chapter 13: Examples 497
Bird Strike and Fan Blade Out (FBO)
Input The simulation consists of three runs. The first run is a prestress analysis that computes the deformations and stresses due to rotational velocities. This computation is essentially linear static and an implicit solver is selected for the purpose of computational efficiency.. Boundary conditions and initial conditions of the prestress run differ from the FBO run. In the prestress run the three bearing points are fixed and a force in circumferential direction is applied to the rotor and fan blades. Implicit Prestress Run Since the entries and details of the prestress input file are quite similar to that of “Bird Strike on rotating fan blades with prestress” example, explanation of the prestress input will be skipped. Explicit FBO Run Since the explicit FBO input is also similar to of the explicit input of “Bird Strike on rotating fan blades with prestress” example, only additional or different entries will be explained. SOL 700,NLTRAN STOP=1 PATH=3 BLADEOUT
BLADEOUT option activates the chaining simulation. All FBO forces assigned in BLDOUT entry will be stored in “MASTER” file after the simulation.
TSTEPNL
1
300
.1e3
1
ADAPT
2
10
TSTEPNL entry describes the number of Time Steps (300) and Time Increment (1.e4 seconds) of the
simulation. End time is the product of the two entries (30 ms). DBEXSSS 111 21 2 3 4 + 8 9 10 11 12 + 16 17 18 19 20 DYPARAM LSDYNA DATABASESSSTATM .00008 5 13 101 6 14 7 15 + +
DBEXSSS entry requests the statistics of subsystems. The subsystems are defined by BCPROP entries. DYPARAM, LSDYNA,DATABASE,SSSTATM parameter requests to store the mass, mass center and mass inertia tensor of the subsystems which are assigned by DBEXSSS entry. All information will be stored to
jid.dytr.ssstat ascii file at every 0.000008 seconds and will be used for unbalance input in the rotor dynamic simulation. $ $ ALL BLADES $ BCPROP,101,1011106,1011107,1011108,1011109,1011110,1011111,1011112,+
Main Index
498 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
BCPROP 101 includes the properties of all blades and BCPROP 21 includes only the broken blade properties. These two subsystem information will be used in UNBALANC and CONM2 entries of SOL 400 rotor dynamics simulation.
$ $ CONTACT ID $ 01 $ 02 $ 03 $ 04 $ 05 $ 06 $ 07 $ 08 $ 09 $ BCTABLE 1 SLAVE + + + SLAVE BODY 1 21 21 1 21 22 1001 1002 1003 1 0 0.1 0. 0 MASTER BODY 22 22 1 23 1004 1005 1006 9 0. 0 SS1WAY DESCRIPTION remaining all blades to direct contact case (recording) broken blade to direct contact case (recording) broken blade to remaining all blades remaining all blades (self contact) broken blade to nondirect contact case direct contact case (self contact) bearing point 1 : fuse at 6.0E3 seconds bearing point 2 bearing point 3 0.1 0. 0
1
1
In the example, 9 contacts are defined. In order to reduce the size of binout file which includes the contact forces, only the fan case is considered to capture the FBO loads.. Only the contact forces between the remaining blades and the fan case (contact 01) and the broken blade and the fan case (contact 02) are stored in the binout file. To store contact forces in binout files, two options in BCTABLE and one parameter are required. SPR and MPR options can store the contact forces on SLAVE and MASTER parts respectively. DYPARAM*,LSDYNA,DATABASE,NCFORC parameter controls the timestep of contact forces output which are defined in BCTABLE. To define the release mechanism, breakable joints (CSPOT) are used. These are elements that have coincident nodes on the hub and the blade roots but are distinct.
Figure 138
Adding Breakable Joints
Main Index
Chapter 13: Examples 499
Bird Strike and Fan Blade Out (FBO)
The breakable joints between the hub and the release blade are added using CSPOT. The joints will be released at 0.00001 seconds after the start of FBO simulation. CSPOT .. CSPOT 1111 101 0.00001 1126 116 10 10 1111 1126
CSPOT entry defines the complex or combined welds. This is used to connect two nodes which are defined by BCGRID entry and are released (broken) at 0.00001 seconds.
BLDOUT
1 0.09204 2 1
0 2.90E+1 4 99
1.0E6 0.0 0.244 12 1 3 3 3
0 1 99
0. 2 99
1. 2 99
0. 2 99
BLDOUT entry defines the contact force output information and mapping criteria for a sequential SOL 700 FBO and SOL 400 RD analyses. Using this entry, the all forces can be categorized and stored to MASTER file in the SOL 700 run. All slaves and maters in the BCTABLE must be assigned to BLDOUT entry using six different types of flags in ISLVi’s and IMASTi’s. In the example, nine ISLVi’s and nine IMASTi’s are required because there are nine contact definitions in BCTABLE. See MD Nastran Quick Reference Guide for other fields.
SPCD2 1 GRID 123 $ BCGRID 123 20003787THRU 2000394220003947+ + THRU 20004022
5
80
1.
2000386220003867THRU
The spin down event after the blade out can be defined by using a timedependent predetermined rotational speed of the turbine using SPCD2, BCGRID, and TABLED1 entries.
Main Index
500 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
Rotor Dynamics Run (SOL 400) The FBO loads computed in SOL 700 are read by SOL 400 by assigning the _FBO.MASTER file to DBSET in the File Management Section (FMS) of SOL 400 run. nastran buffsize=65536 nastran dbcfact=4 nastran system(151)=1 init scratch logi=(scratch(9999000)) assign dbloc1='impact_FBO.MASTER' dblocate datablk=(GEOM3K) logical=dbloc1 , where(projno>0 and version=* and wildcard)
SOL 400
SOL 400 executive control entry activates nonlinear static and transient analysis.
Case control commands are defined in the following box. analysis=nltran rigid=linear RGYRO= 100 DISPLACEMENT(print,plot,SORT1,REAL)=ALL STRESS(plot,SORT1,REAL)=ALL STRAIN(plot,SORT1,REAL)=ALL
ANALYSIS=NLTRAN actives “nonlinear transient analysis”.
Bulk data starts with BEGIN BULK. BEGIN BULK .. $MATD024 403153 4.14E4 1.60E+7 0.35000 2.5E5 $ MAT1 101 1.60E+7 0.35000 4.14E4 MATEP 101 SLOPE 2.5E5 1.38E5
1.38E5
0.25000
In order to use equivalent material properties in SOL 400, all MATD024 materials models used in SOL 700 are translated to MAT1 and MATEP with slope option. TSTEPNL 100 4000 1.0E2 0 1.0E4 1.0E2 2
Main Index
Chapter 13: Examples 501
Bird Strike and Fan Blade Out (FBO)
TSTEPNL entry of SOL 400 controls the convergence criteria and data for nonlinear transient analysis. ROTORG 10 $ RSPINT 10 $ TABLED1 1000 + 0.0 + 0.042 + 100. 10 11 4500.0 3689.1 2250.0 THRU 10 0.012 0.055 ENDT 21 RPM 4255.0 3605.1 1000 0.016 0.25 4096.6 2915.7 0.028 0.5 + 3834.2 + 2250.0 +
ROTORG entry defines the rotor which consists of GRID IDs from 10 to 21. RSPINT entry indicates the rotational direction which is assigned to the rotational axis from GRID 11 to GRID 10. The rotational speed is defined in TABLED1, 10 for describing the speed down at various time steps. Note that the magnitude of the rotational velocities defined in SOL 400 differ from SOL 700. This is because the unit of rotational velocity used in SOL 400 is RPM and is different to that used by SOL 700 (radian/seconds)
$ impact_FBO.dytr.ssstat subsystem: 1 total mass of subsystem xcoordinate of mass center ycoordinate of mass center zcoordinate of mass center .... subsystem: 21 = = = = 0.59591613E+01 0.16148860E+03 0.41383951E05 0.10000020E+03 0.2139E+01 0.1921E02 0.1748E+04 = 0.91899477E01 = 0.16037865E+03 =0.28772884E+02 = 0.10338639E+03
total mass of subsystem xcoordinate of mass center ycoordinate of mass center zcoordinate of mass center
inertia tensor in global coordinates row1= 0.2385E+04 0.5329E03 row2= 0.5329E03 0.1748E+04 row3= 0.2139E+01 0.1921E02 > translate UNBALNC,100,0.0919,12,0.,1.,0.,,+ +,29.00,180.0,1.10995 $ $ blade + hub $ CONM2, 2001,12, ,5.959,,,,,+ +, 0.2385E+04,0,0.1748E+04,0,0,0.1748E+04 $ GRID 12 161.488 0. 100.
Main Index
502 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
The mass, mass center and mass inertia tensor computed in SOL 700 are stored in the impact_FBO.dytr.ssstat file. These values are then used in SOL 400 to define mass unbalance by UNBALNC and CONM2 entries. As shown in the box above, the order of the subsystem id numbers in ssstat file is determined by the order of DBEXSSS as defined in SOL 700. For example, subsystem 1 represents the released blade while subsystem 21 represents all blades and hub information. The unbalance mass in the UNBALNC entry is the same value of total mass as defined in subsystem 1. ROFFSET and ZOFFSET of UNBALNC entry are calculated by the difference of the mass locations between subsystem 1 and 21. In the example, the xdirection in SOL 700 FBO simulation is coincident with the zdirection of the rotor in SOL 400 RD simulation. In addition, the mass inertia tensor of subsystem 21 is recorded to Iij fields of CONM2 entry. GRID 12 which describes the mass location of hub and blades is also set to the same center location of subsystem 21. In the gyroscopic nonlinear transient analysis, only the additional unbalance mass is considered as opposed to FBO simulation, where the unbalance mass results from losing mass due to blade out. Therefore, the additional mass must be added to the opposite side of the location where bladeout occured. To add the mass to the opposite side of the blade out, the unbalance is assigned at the location which is measured 180 degrees in the positive direction of the local unbalance coordinate system. BLDOUT,1, 1, 1.0E6, 0.0, 0, 0, 1.0E3, 1 +,0.0919,2.9000E+1, 1.10995, 12, 0, 0.,1.,0.
BLDOUT entry is also used in SOL400. BLDOUT entry in SOL 400 can control and apply the FBO forces
to the nonlinear transient analysis using different time steps. CBUSH PBUSH 101 101 101 K 1002 1012 1.0E7 0 1.0E7
Bearings in SOL 400 are modeled using CBUSH elements. PBUSHT controls the failure criteria. The CBUSH element is defined to fail at 1.65E5 lbf in radial (yz) direction.
Main Index
Chapter 13: Examples 503
Bird Strike and Fan Blade Out (FBO)
Results of Prestress Implicit Simulation
SOL 400 SOL 700
Figure 139
Displacement Contours on Fan Blades and Rotor – SOL 400 vs SOL 700
Main Index
504 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
SOL 400
SOL 700
Figure 1310 Table 131
Stress Contours on Fan Blades and Rotor – SOL 400 vs SOL 700 Comparison of SOL 400 vs SOL 700 Prestress Results SOL400 SOL 700 398 seconds 24.13 mm (0.950 inch) 696.4 MPa (101 ksi) Difference (refer to SOL 400 results) 300% 2.2 % 2.0 %
Analysis Time Maximum Resultant Displacement Maximum Equivalent Stress
135 seconds 24.66 mm (0.971 inch) 710.2 MPa (103 ksi)
The stress and deformation results between SOL 400 and SOL 700 were within 2% of each other, which is quite acceptable (see Table 131) . However, for this particular analysis, which took a few minutes to complete, SOL 400 ran the same model three times faster than the SOL 700 implicit solver. One reason for the speed up difference is because SOL 700 implicit solver is based on a double precision version.
Main Index
Chapter 13: Examples 505
Bird Strike and Fan Blade Out (FBO)
Results of Fan Blade Out (FBO) Explicit Simulation
0.0015 seconds
0.0037 seconds
0.0037 seconds at a different angle Figure 1311 Effective Stress Contour
Main Index
506 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
Figure 1312
Snapshot of Contact Between Released Blade and Containment
Plastic Strain
Effective Stresses
Figure 1313
Fan Case t = 3ms
Main Index
Chapter 13: Examples 507
Bird Strike and Fan Blade Out (FBO)
Node 3175 of Case
Impact Loads
Rubbing Load (Radial)
Rubbing Load (Tangential)
Node 1405440 of Blade Tip
Rubbing Load (Radial) Figure 1314 Loads
Rubbing Load (Tangential)
Main Index
508 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Bird Strike and Fan Blade Out (FBO)
Figure 1315
Node 31795 and 1405440 Locations on Case and Blade Tip
Results of Rotor Dynamics Implicit Simulation In this analysis, the orbit diagrams between SOL 400 and SOL 700 (see Figure 1316) demonstrate that the whirling characteristics are similar and deformation magnitudes of the whirl are very close in the zdirection but differ in the ydirection. The difference could be due to two contributing factors. One factor is that the contact forces between the trailing blades and the released blade were not taken into consideration in the rotordynamics simulation and only the impact and rubbing loads were included. The contact forces between the blades will have some contribution to the magnitude of the whirling deformation in the ydirection. The second reason is because the sectional properties of the rotor at center where hub is located are approximated in the stick model for the rotor dynamics simulation.
Main Index
Chapter 13: Examples 509
Bird Strike and Fan Blade Out (FBO)
Figure 1316
Comparison of Orbit at the Tip of Rotor
In the implicit rotordynamic analysis, the failure load for bearing 1 is set to 734 kN (1.6E5 lbf). A radial dependence is specified for the fuse option. Figure 1317 shows the time history for the force in this bearing. The bearing is found to fuse in less than a revolution after the FBO event. The timetofuse is then used to modify the explicit FBO analysis. In the FBO analysis, fusing is modeled by deactivating contact between the two rings of the bearing at the analysis time recorded in the implicit rotordynamic analysis.
Figure 1317
Loadings on the First Bearing and Fusing After 0.004 Seconds
Main Index
510 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Drop Test
Drop Test
Drop Test Simulation of a Computer Package
Description Many companies perform drop test simulation to predict the impactresistance properties of the package for their products. This is an example of a computer package which includes the CPU box, foam and the clipboard box which is dropped at a velocity of 3.81 m/s on a rigid floor. To model the simulation, contact relations were defined between the ground, the clipboard box, the foam and the CPU box itself. A picture of the model without the clipboard box is shown below. SOL 700 Entries Included SOL 700 TSTEPNL PARAM, DYDTOUT PARAM, DYSHELLFORM PARAM, DYSHINP BCTABLE BCBODY BSURF BCPROP MATD020 MATD014 Model
The model has a total of 33178 grid points and 127397 elements as follow:
• 13024 Trias • 1859 Quads • 112514 Tetras
Main Index
Chapter 13: Examples 511
Drop Test
The ground, clipboard box, and CPU box were modeled with shell elements. The foam was modeled with Tet elements. All shell elements are BelytschkoTsay formulation. The gravity is defined to take into account the mass of the CPU box and the bouncing effect after the drop. The drop speed of the CPU box is modeled by defining an initial velocity of 3.81m/s, applied on all the grid points of the CPU in a vertical direction towards the ground. The simulation time is 0.025 seconds. The unit system is Newton, meters, and seconds. Input All nodes of the CPU box have an initial velocity specified by the TIC entry. All nodes of the rigid ground have been constrained in all the degrees of freedom. Contacts are defined between: 1. The ground (master) and bottom of the clipboard box (slave). 2. The CPU box (master) and the foam (slave). 3. The clipboard box (master) and the foam (slave). Input File SOL 700,NLTRAN path=3 stop=1
SOL 700 is a executive control that activates an explicit nonlinear transient analysis.
TIME 10000 CEND LOADSET = 1 SPC = 2 DLOAD = 2 IC=1 TSTEPNL = 20 BCONTACT = 1 weightcheck=yes The bulk entry section begins... $ BULK DATA SECTION BEGIN BULK INCLUDE final_scale2.bdf TSTEPNL 20 100 2.5e4
1
ADAPT
2
10
TSTEPNL is a SOL 700 bulk data entry which describes the number of Time Steps (100) and Time
Increment (2.5e4 seconds) of the simulation. The total time is the product of the two entries. Notice here the Time Increment is only for the first step. The actual number of Time Increments and the exact value of the Time Steps is determined by SOL 700 during the analysis. The time step is a function of the smallest element dimension during the simulation.
Main Index
512 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Drop Test
PARAM,DYDTOUT,2.5e3
PARAM, DYDTOUT is a SOL 700 bulk data entry which describes the time interval of d3plot outputs.
PARAM*,DYSHELLFORM,2
PARAM*, DYSHELLFORM is a SOL 700 bulk data entry to define the shell formulation. If
DYSHELLFORM = 2, the default shell formulation is BelytschkoTsay. PARAM,DYSHNIP,2
PARAM,DYSHINP is a SOL 700 bulk data entry that specifies the number of integration points for SOL 700 shell elements.
Bulk data entry that defines gravity load effects: $  GRAVITATION $ TLOAD1 4 5 LSEQ 1 5 3 DLOAD 2 1. 1. GRAV 3 0 9.806 TABLED1 100
100 4 0. 0. 1.
GRAV is a bulk data entry that defines acceleration vectors for gravity or other acceleration loading.
Bulk data entry that defines Contact relations and Contact bodies: $ Define contact BCTABLE 1 SLAVE MASTERS SLAVE MASTERS SLAVE MASTERS
3 1 2 3 4 3 5
BCTABLE is a SOL 700 bulk data entry that is also used by SOL 600 and it is meant to define
MasterSlave as well as selfcontact.
Main Index
Chapter 13: Examples 513
Drop Test
$ bottom of cardboard as BCBODY 1 3 $ $ ground as master BCBODY 2 3 $ $ foam as slave BCBODY 3 3 $ $ CPU as master BCBODY 4 3 $ $ cardbox box as master BCBODY 5 3
slave DEFORM DEFORM DEFORM DEFORM DEFORM
1 2 3 4 5
0 0 0 0 0
BCBODY is a bulk data entry that is used by MD Nastran Implicit Nonlinear (SOL 600) and SOL 700
only, which defines a flexible or rigid contact body in 2D or 3D. BSURF .. .. 1 231364 THRU 231420
BSURF is a bulk data entry that is used by MD Nastran Implicit Nonlinear (SOL 600) and SOL 700 only,
which defines a contact surface or body by element IDs. All elements with the specified IDs define a contact body. $ define BCPROP $ $ define BCPROP $ $ define BCPROP $ $ define BCPROP ground using property 1 2 1 foam using property 4 3 4 CPU using property 3 4 3 cardboard box using property 2 5 2
BCPROP is a bulk data entry that is used by MD Nastran Implicit Nonlinear (SOL 600) and SOL 700 only, which defines a 3D contact region by element properties. All elements with the specified properties define a contact body.
Main Index
514 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Drop Test
Bulk data entry that defines properties for shell and solid elements: $ ========== $ PSHELL $ PSHELL $ PSHELL $ PSOLID PROPERTY SETS ========== * groung_prop * 1 1 .005 * cardboard_prop * 2 2 .005 * cpu_prop * 3 3 .005 * foam_prop * 4 4
Bulk data entry that defines material properties: $  Material ground id =1 $MATRIG 1 1e+20 1e+11 .3 $Define rigid material and apply constraint in all dof's MATD020 1 7850 1e11 0.3 1 7 7
MATD020 is a SOL 700 bulk data entry that is used to model rigid materials.
$  Material cardboard id =2 MAT1,2,2.5e9,,0.3,1.e2 $ $  Material cpu id =3 MATD020 3 2.16e3 2.1e11 0.3 $  Material foam id =4 MATD014,4,20,3e6,2e6,0.,0.,3.,1.e20,+ +,,,,,,,,,+ +,0.0513,0.1054,0.1984,0.5108,0.7985,1.0217,1.347,,+ +,,,,,,,,,+ +,60000,78000,90000,111000,150000,180000,240000,,+ +,,,,,,,,, .. ..
MATD014 is a SOL 700 bulk data entry that is used to model foam materials.
Bulk data entry that defines boundary conditions and initial velocity of the package: $  Initial Velocity BC nodal_vel $ Loads for Load Case : Default TIC 1 1 1 TIC 1 2 1 .. .. TIC 1 8228 1 TIC 1 8237 1 3.81 3.81 3.81 3.81
Main Index
Chapter 13: Examples 515
Drop Test
TIC is a bulk data entry that defines values for the initial conditions of variables used in structural
transient analysis. Both displacement and velocity values may be specified at independent degrees of freedom. End of input file.
Main Index
516 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Drop Test
Results
t = 0.0 seconds
t = 0.008 seconds
t = 0.0175 seconds
t = 0.0175 seconds
Main Index
Chapter 13: Examples 517
Drop Test
Drop Test Simulation of a Computer Package
Main Index
518 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Drop Test
Wheel Drop Test
Description This is an example of a wheel drop test as required in automotive industry to comply with government regulations. In this test a rigid block of 540 Kg is dropped at 13 degrees on a wheel. The drop velocity is 2052.8 mm/s. Several contacts are defined to predict the interaction between wheel, tire and the rigid block. SOL 700 Entries Included SOL700 TSTEPNL BCTABLE BCBODY BSURF TIC TLOAD1MATD027 PCOMP PLOAD4 HGSUPPR Model
230 mm
13°
Main Index
Chapter 13: Examples 519
Drop Test
The model has a total of 90428 grid points and 78153 elements as follow:
• 11040 Quads • 40077 Pentas • 27036 Hexas
The impact velocity of the rigid block modeled by 48 hexa elements is 2052.8 mm/s in the xdirection (vertical). The velocity is applied on all nodes of the block by using the TIC entry. Since the tire is inflated, PLOAD4 and Table entries are used to apply a pressure load of 0.2 MPa to internal surface of the tire as shown in the figure shown above. Gravity is also taken into account. The wheel is constrained in all directions at surface located around the bolt locations as marked in color.
This part is fixed
Four Contacts are defined between: 1. Rigid block and tire 2. Rigid block and wheel 3. Tire and wheel 4. Self contact of tire Total simulation time is 0.04 s. Input SOL 700,NLTRAN path=3 stop=1
SOL 700 is the command for the initiation of an explicit nonlinear transient analysis.
Main Index
520 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Drop Test
BEGIN BULK TSTEPNL 1
100
4.4
1
ADAPT
2
10
TSTEPNL is a SOL 700 bulk data entry which describes the number of Time Steps (100) and Time Increment (0.4 ms) of the simulation. End time is the product of the two entries. Notice here the Time Increment is only for the first step. The actual number of time increments and the exact value of the Time Steps are determined by SOL 700 during the analysis. The time step is a function of the smallest element dimension during the simulation.
BCTABLE
1 SLAVE
3 0
0. 0
7 0. 0
.3
0.
0
0.
MASTERS 3
BCTABLE is a SOL 700 bulk data entry to define MasterSlave contact as well as self contact.
BCBODY BSURF ...
3 3
3D 200001
DEFORM 200002
3 200003
0 .3 200004 200005
200006
BCBODY and BSURF are bulk data entries that are used by MD Nastran Implicit and Explicit Nonlinear only. BCBODY defines a flexible or rigid contact body in 2D or 3D and BSURF assigns a contact body
or surface defined by Element IDs. Bulk data entry that defines grid points and elements: GRID ... $ CHEXA 213813 ... CPENTA 211809 ... CQUAD4 ... 1 200001 213713 232641 310001 200 213700 230 310 201803 201204 201802 211204 201809 211703 211803 201703 211802 82.3 203800 144.119 138.485 203813 203713 203700 213800
Main Index
Chapter 13: Examples 521
Drop Test
Bulk data entry that defines boundary conditions: SPC1 ... SPC1 ... 1 3 123456 23 864 60001 874 THRU 875 60108 876 882 883
Bulk data entry that defines initial velocities for the rigid block model: TIC ... 2 60001 1 2052.8
TIC is a bulk data entry that defines values for the initial conditions of variables used in structural
transient analysis. Both displacement and velocity values may be specified at independent degreesoffreedom. SPCADD TLOAD1 2 6 1 7 3 1
TLOAD1 is a bulk data entry that defines a timedependent dynamic load or enforced motion for use in transient response analysis.
MATD027
200 0.
9.610
.49
0.6933
.03224
Five material models are used in this analysis. MATD027 is used to model the rubber of the tire, with the following properties: Density = 9.6E10 ton/mm3 Poisson’s ratio = 0.49 and Coefficients of strain energy equation are 0.6933 and 0.3224 (see MD Nastran Quick Reference Guide.) PSOLID ... PCOMP ... 200 360 301 303 200 .5 .6 90. 25. 0 0. YES YES 301 305 .5 .5 90. 0. 0. YES YES
Main Index
522 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Drop Test
The other material models are used to model the alloy wheel and the composite tire. PCOMP is used for the definition of a composite material laminate shell. The composite layers are shown in schematic above. It consists of a total of 4 material layers 301, 301, 303 and 305. The thicknesses and orientation angles are shown in the definition of PCOMP entry above. PLOAD4 210101 ... GRAV 4 5 232401 1. 2 9800. 1. 0. 0. 200105
PLOAD4 is a bulk data entry which describes a pressure load on a face of 2D or 3D elements.
TABLED1 ...
2 0.
1.
1000.
1.
ENDT
Several tables and coordinates are defined in the input file. HGSUPPR, 200, SOLID, 200, 1, , , , 0 ... , , , 0.040
HGSUPPR card is for the definition of the hourglass suppression method, the corresponding hourglass
damping coefficients and sets for the bulk viscosity method and coefficients.
Main Index
Chapter 13: Examples 523
Drop Test
Results
t = 0.000 second
t = 0.013 second
t = 0.027 second
t = 0.040 second
Main Index
524 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Defense
Defense
Rod Penetration
Description This is an example of a rod penetrating through a plate. The initial velocity of the rod is 124.6 m/s in horizontal direction, and 33.39 m/s in the vertical direction. To model the simulation, Adaptive Contact was defined between the rod and the plate to predict the stress and deformations of both parts after the penetration SOL 700 Entries Included SOL 700 TSTEPNL PARAM, DYDTOUT PARAM*,DYCONSLSFAC BCTABLE BCBODY BCPROP MATD024 TICD Model Because of symmetry, only half of the actual structures (rod and plate) were modeled in this example. Boundary conditions were applied along the center line of the structures to ensure symmetric behavior. The model has a total of 7668 grid points and 5664 solid elements. All the elements were hexahedrals (CHEXA). The initial velocity of the rod is 124.6 m/s in horizontal direction, and 33.39 m/s in the vertical direction toward the plate, applied on all the grid points. The simulation time is 0.11 seconds. The unit system is Kilonewton, meters, and milliseconds.
Main Index
Chapter 13: Examples 525
Defense
Input All nodes of the rod have an initial velocity specified by the TICD entry. All nodes along the center line have a boundary condition that ensures symmetric behavior of the structures. All nodes of the plate have been constrained in all the degrees of freedom. Adaptive Contact is defined between the rod and the plate. Input File: SOL 700,NLTRAN path=3 stop=1
SOL 700 is a executive control entry that activates an explicit nonlinear transient analysis.
TIME 10000 CEND param,marcbug,0 ECHO = NONE DISPLACEMENT(SORT1,print,PLOT) = ALL Stress(SORT1,PLOT) = ALL Strain(SORT1,PLOT) = ALL accel(print,plot)= ALL velocity(print,plot)= ALL echo=both $ LOADSET = 1 SPC = 2 $ DLOAD = 2 IC=1 TSTEPNL = 20 BCONTACT = 1 weightcheck=yes page The bulk entry section starts… BEGIN BULK $ $ $ TSTEPNL + + + +
20
10
11.
1
5
10
0
Main Index
526 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Defense
TSTEPNL is a SOL 700 bulk data entry which describes the number of Time Steps (10) and Time Increment (11 milliseconds) of the simulation. The total time is the product of the two entries. Notice here the Time Increment is only for the first step. The actual number of Time Increments and the exact value of the Time Steps is determined by SOL 700 during the analysis. The time step is a function of the smallest element dimension during the simulation.
PARAM,DYDTOUT,5
PARAM, DYDTOUT is a SOL 700 bulk data entry which describes the time interval of d3plot outputs.
PARAM*,DYCONSLSFAC,1.0
PARAM*,DYCONSLSFAC is a SOL 700 bulk data entry to define the default scale factor for contact
forces Bulk data entry that defines Adaptive Contact relations and Contact bodies: BCTABLE + YES MASTER 4 1 SLAVE 4 3
BCTABLE is a SOL 700 bulk data entry that is also used by SOL 600 and it is meant to define MasterSlave as well as self contact.
BCBODY BCBODY
3 4
3 3
DEFORM DEFORM
3 4
0 0
BCBODY is a bulk data entry that is used by MD Nastran Implicit Nonlinear (SOL 600) and SOL 700
only, which defines a flexible or rigid contact body in 2D or 3D. BCPROP BCPROP 3 4 2 1
BCPROP is a bulk data entry that is used by MD Nastran Implicit Nonlinear (SOL 600) and SOL 700
only, which defines a 3D contact region by element properties. All elements with the specified properties define a contact body.
Main Index
Chapter 13: Examples 527
Defense
Bulk data entry that defines properties for solid elements: $ ========== PROPERTY SETS ========== $ $ * projectile * $ PSOLID 1 1 $ $ * plate * $ PSOLID 2 2 Bulk data entry that defines material properties: $ ========= MATERIAL DEFINITIONS ========== $ $ $ $  Material MAT_PLASTIC_KINE.2 id =2 MATD024 1 18.62 1.17 .22 0.0179 $  Material MAT_PLASTIC_KINE.1 id =1 MATD024 2 7.896 2.1 .284 0.01
0.8 0.8
MATD024 is a SOL 700 bulk data entry. It is used to model an elastoplastic material with an yield stress
versus strain curve and arbitrary strain rate dependency. Failure can also be defined based on the plastic strain or a minimum time step size. Bulk data entry that defines initial velocity of the rod: $  Initial Velocity BC ini $ TICD 1 1 1 0.1246 TICD 1 1 3 0.03339
2586 2586
1 1
TICD is a bulk a SOL 700 bulk data entry that defines transient analysis initial conditions with
increment options.
Main Index
528 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Defense
Bulk data entry that defines grid points and elements. $  Define 7668 grid points $ GRID 1 9.241751.5e05.0513793 .. .. GRID 7668 23.0000 4.80000 .00000 $ $  Define 5664 elements $ CHEXA 1 1 1 2 5 11+A000001 +A000001 14 13 .. .. CHEXA 5664 2 7619 7620 7628 7660+A005664 +A005664 7668 7667
4
10
7627
7659
Bulk data entry that defines boundary conditions along the center line, and the fixed boundary conditions of the plate: SPC1 .. .. 2572 SPC1 2632 .. .. SPC1 SPCADD $ ENDDATA 2575 3 2576 123456 2579 2587 2580 2596 2583 2605 2584 2614 2623 1 2559 246 2560 1 2563 2 2564 3 2567 10 2568 11 2571 12
3 2
123456 1
7660 3
THRU
7668
End of input file.
Main Index
Chapter 13: Examples 529
Defense
Results
time = 0.0 seconds
time = 0.055 seconds
time = 0.025 seconds
time = 0.11 seconds
Main Index
530 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Defense
Shaped Charge, using IG Model, Penetrating through Two Thick Plates
Problem Description When a metal cone is explosively collapsed onto its axis, a highvelocity rod of molten metal, the jet, is ejected out of the open end of the cone. The cone is called a liner and is typically made of copper. The jet has a mass approximately 20 percent of the cone mass, and elongates rapidly due to its high velocity gradient. This molten rod is followed by the rest of the mass of the collapsed cone, the slug. Typical shaped charges have liner slope angles of less than 42 degrees ensuring the development of a jet; with jet velocities ranging from 3000 to 8000 m/s. A typical construction of a shaped charge is shown in Figure 1318.
Figure 1318
Typical Construction of Shaped Charge
Main Index
Chapter 13: Examples 531
Defense
An example simulation of shaped charge formation is carried out to demonstrate the ability of SOL 700 to perform such a simulation. A simplified axisymmetric model of explosives and a copper liner is created in a finite volume Euler mesh. Explosive are detonated starting from a point on the axis of symmetry at the end of the explosives. The simulation is carried out for 60 μs after detonation of the explosives. The jet is formed and penetrates two thick plates. See Figure 1319 for the model layout.
Figure 1319
SOL 700 Model Setup
Typical shaped charges are axisymmetric. However, aiming at higher velocity, 3D designs are targeted. 3D simulation of shaped charge formation would be necessary to avoid excessive experimental work. SOL 700 has full abilities to perform such a 3D simulation. SOL 700 Model The model is simplified as shown in Figure 1319. The aluminum casting is replaced with a rigid body. Detonation is assumed to start at a point on the axis at the rear end of the explosives. The liner shape is slightly simplified as shown in the figure. The retaining ring is assumed rigid and is modeled as a wall boundary for the Euler Mesh (BARRIER). SI units are used in this example.
Main Index
532 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Defense
A. Euler Mesh and Liner: A triangular prismatic Finite Volume Euler mesh is used with head angle of 5 degrees as shown in Figure 1320. A very fine mesh is used to accurately simulate the behavior of the extremely thin liner. The liner is placed in this Euler mesh. Symmetry conditions (closed boundary, default Euler boundary condition) are imposed on the two rectangular faces of the prism to create an axisymmetric behavior.
Figure 1320
Euler Mesh
The liner material pressure – density relationship is modeled with EOSPOL model. The liner is made of copper and the constants are taken as follows: a1 a2 a3 b1 b2 b3 1.43E11 0.839E11 2.16E9 0.0 0.0 0.0 N/m2 N/m2 N/m2
Material yield strength is modeled with a JohnsonCook yield model. The constants are taken as follows: A B C n m ε0 1.2E8 1.43E9 0.0 0.5 1.0 1.0 N/m2 N/m2
Main Index
Chapter 13: Examples 533
Defense
Tmelt Troom Cv
1356.0 293.0 399.0
K K J/kg
Other liner material properties of liner are as follows: Density Constant shear model Constant spallation model In the input deck: MATDEUL 701 8960. 711 712 713 714 EOSPOL, 711, 1.43+11, 0.839+11, 2.16+9 SHREL,712,0.477E11 $ JohnsonCook $ A B n C m EPS0 Cv YLDJC,713, 1.2E8, 1.43E9, 0.5, 0.0, 1.0, 1.0, 399.0,+ $ TMELT TROOM +, 1356.0, 293.0 $ PMINC,714,2.5E10 It is very easy to define the shape and position of the liner by using the method of geometrical regions when creating the initial conditions of the liner material. CYLINDR, 1,, 0.5391, 0.56, 0., 2.0, 0.4147, 0.,+ +,0.2958 CYLINDR, 2,, 0.5391, 0.56, 0., 2.0, 0.4147, 0.,+ +,0.2939 CYLINDR, 3,, 0.2, 2.0406, 0., 0.2047, 2.0406, 0.,+ +,2.0019 TICVAL,2,,DENSITY,8960. B. Casting and Retaining Ring: The casting is assumed to be rigid. It is modeled by the default Eulerian boundary condition (closed boundary). The retaining ring is also assumed to be rigid and is modeled by a barrier. C. Plates: Two thick plates are placed in this Euler mesh. Plate material is defined as steel: MATDEUL 801 7830. 811 812 813 814 EOSPOL, 811, 1.64E+11 SHREL,812,0.818E11 YLDVM,813,1.4E9 PMINC,814,3.8E9 The shapes and positions of the plates are defined by using the method of geometrical regions. 8960 0.477E11 2.5E10 Kg/m3 N/m2 N/m2
Main Index
534 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Defense
CYLINDR, 4,, 0.22, 2.0406, 0., 0.223, 2.0406, 0.,+ +,2.05 CYLINDR, 5,, 0.27, 2.0406, 0., 0.273, 2.0406, 0.,+ +,2.05 TICVAL,3,,DENSITY,7830. D. Explosive: The explosive is modeled by ignition and growth equation of state. The explosive is placed in this Euler mesh. EOSIG,100,,,,,,,,+ +,,,,,,,,,+ +,,,,,99,,MCOMPB,SI The explosive material is taken from the database that is build into SOL 700. To initialize the whole Euler mesh, a TICEUL card will be defined. TICEUL1 1 1 TICREG 1 1 ELEM 1 TICREG 2 1 CYLINDER1 TICREG 3 1 CYLINDER2 TICREG 4 1 CYLINDER3 TICREG 5 1 CYLINDER4 TICREG 6 1 CYLINDER5 $ SET1 1 1 THRU 15342 TICVAL,1,,DENSITY,1630.,SIE,4.29E6 Results
Figure 1321 shows the initial position of the copper liner and two thick plates at 0μs, snap shots of
100 701 701 801 801
1 2 2 3 3
1. 2. 3. 4. 5. 6.
liner collapse, jet formation and plates penetrated at 10 μs, 20 μs, 30 μs, 40 μs, 50 μs and 60 μs.
Main Index
Chapter 13: Examples 535
Defense
Figure 1321
Initial Position of the Copper Liner and Two Thick Plates, Snap Shots of Liner Collapse, Jet Formation and Plates Penetrated (Courtesy – Postprocessing by
CEI Ensight)
Figure 1322 shows the velocity field of explosive gases, liner, and jet at 20 μs. A jet velocity of about 6000 m/s is achieved
Figure 1322
Velocity Field of Explosive Gases, Liner, and Jet
Abbreviated SOL Input File SOL 700,NLTRAN STOP=1 CEND TITLE = SHAPED CHARGES TEST $ for QA purpose, run shorter time $ENDTIME = 1.E5 IC = 1 TSTEPNL=1 $$ $ BEGIN BULK
Main Index
536 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Defense
TSTEPNL 1 10 1.E06 1 PARAM*,DYINISTEP,1.E11 PARAM*,DYMINSTEP,1.E13 DYPARAM,VELMAX,20.0E+03 DYPARAM,LSDYNA,BINARY,D3PLOT,1.E5 $ INCLUDE model.bdf INCLUDE wall.dat $ EXPLOSIVE $ MATDEUL 100 1630. 100 101 102 $ EOSIG,100,,MCOMPB,SI,,,,,+ +,,,,,,,,,+ +,,,,,,,,,+ +,,,,,99 $ SHREL,101,3.E9 $ YLDVM,102,2.E8 $ $ COPPER $ MATDEUL 701 8960. 711 712 713 EOSPOL, 711, 1.43+11, 0.839+11, 2.16+9 SHREL,712,0.477E11 $ JohnsonCook $ A B n C m EPS0 Cv YLDJC,713, 1.2E8, 1.43E9, 0.5, 0.0, 1.0, 1.0, 399.0,+ $ TMELT TROOM +, 1356.0, 293.0 $ PMINC,714,2.5E10 $ $ STEEL $ MATDEUL 801 7830. 811 812 813 EOSPOL, 811, 1.64E+11 SHREL,812,0.818E11 YLDVM,813,1.4E9 PMINC,814,3.8E9 $ TICEUL1 1 1 TICREG 1 1 ELEM 1 100 1 TICREG 2 1 CYLINDER1 701 2 TICREG 3 1 CYLINDER2 TICREG 4 1 CYLINDER3 701 2 TICREG 5 1 CYLINDER4 801 3 TICREG 6 1 CYLINDER5 801 3 $ PEULER1, 1 ,, MMSTREN, 1 SET1 1 1 THRU 15342 CYLINDR, 1,, 0.5391, 0.56, 0., 2.0, 0.4147, 0.,+ +,0.2958
714
814
1. 2. 3. 4. 5. 6.
Main Index
Chapter 13: Examples 537
Defense
CYLINDR, 2,, 0.5391, 0.56, 0., 2.0, +,0.2939 CYLINDR, 3,, 0.2, 2.0406, 0., 0.2047, +,2.0019 CYLINDR, 4,, 0.22, 2.0406, 0., 0.223, +,2.05 CYLINDR, 5,, 0.27, 2.0406, 0., 0.273, +,2.05 $ TICVAL,1,,DENSITY,1630.,SIE,4.29E6 TICVAL,2,,DENSITY,8960. TICVAL,3,,DENSITY,7830. $ BARRIER,1,2 $ ENDDATA
0.4147, 0.,+ 2.0406, 0.,+ 2.0406, 0.,+ 2.0406, 0.,+
Main Index
538 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Defense
Mine Blast
Problem Description This is a simulation of an explosion under a vehicle. The vehicle has triggered a mine that is exploding underneath the bottom shield. In this example, the actual explosion of the mine is not modeled. Instead, the simulation is started moments after the mine explodes. This is called the blast wave approach. At the location of the mine, a high density and high specific energy is assumed in the shape of a small sphere. During the simulation, this region of high density, energy, and high pressure, expands rapidly. The blast wave interacts with the bottom shield and causes an acceleration of parts of the flexible body. The intent of this simulation is to find the location and the value of the maximum acceleration. SOL 700 Model An outline of the basic numerical model is shown in Figure 1323 below. It is composed of the following main components: a. Vehicle Structure b. Euler Domain 1  air outside vehicle and compressed air (explosive) c. Euler Domain 2  air inside vehicle d. Ground e. Fluid Structural Coupling
Figure 1323
Outline of Basic Numerical Model
Main Index
Chapter 13: Examples 539
Defense
A. The Vehicle: Vehicle structure is modeled by QUAD, TRIA shell elements and some BAR elements.
Figure 1324
Vehicle Structure
Material properties are taken as follows: Density Modulus of elasticity Poison ratio Yield stress 7.85E9 210000. 0.3 250. tonne/mm/s2 tonne/mm3 tonne/mm/s2
Assumed that there will be no failure of the structure. In a part of the structure, there is a hole through which air and pressure waves can freely flow. This hole will be modeled with dummy shell elements. B. Euler Domain 1: The first Euler domain is the air on the outside of the vehicle. The properties of air at rest are: Density Gamma Specific internal energy In the input deck: MATDEUL,230,1.29e12,203,,,,,,+ +,,1.01 TICVAL,5,,DENSITY,1.29E12,SIE,1.938e11 1.29E12 1.4 1.9385E8 tonnemm2/s2 tonne/mm3
Main Index
540 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Defense
At the location of the mine, a small region will be modeled with high density and specific internal energy equivalent to TNT of 7kg when the sphere has a radius of .25 meter: Density Specific Internal Energy The input deck will show: TICVAL,4,,DENSITY,107E12,SIE,3.9e12 SPHERE,400,,1797.5,0.,450.,250. The Euler region will be modeled by using the MESH card. The region will have to be large enough to contain the entire vehicle, including when the vehicle is in motion: MESH,1,BOX,,,,,,,+ +,2623.,1403.,903.,6100.,2800.,2150.,,,+ +,30,10,10,,,,EULER,201 For the most accurate blastwave simulations, it is advised to use the Secondorder Euler solver of SOL 700. This is activated by specifying the secondorder option on the Euler property card and specifying the parameter to use the secondorder Range Kutta integration method: PARAM,RKSCHEME,3 PEULER1,201,,2ndOrder,101 To initialize the whole first Euler mesh, a TICEUL card will be defined. To initialize the Euler domain, other than within the sphere of the explosion, a second large sphere is used. Because it has lower priority, the Euler elements within the mine blast are will still initialized with high density and energy: TICEUL1,101,11 TICREG,1,11,SPHERE,400,230,4,20. TICREG,2,11,SPHERE,501,230,5,1. SPHERE,400,,1797.5,0.,450.,250. SPHERE,501,,0.,0.,5000.,10000. The Euler domain has infinite boundaries. This can be achieved by defining a zero gradient flow boundary on the outside of the Euler mesh. Use an empty FLOWDEF card: FLOWDEF,202,,HYDRO,,,,,,+ +,FLOW,BOTH C. Euler Domain 2: The second Euler region represents the air inside the vehicle. Also for the second Euler region, a MESH card is used. The air is at rest again, so the same properties apply: PEULER1,202,,2ndOrder,102 TICEUL1,102,12 TICREG,3,12,SPHERE,502,230,5,5. SPHERE,502,,0.,0.,5000.,10000. 107E12 4.9E12 tonne/mm3 tonnemm2/s2
Main Index
Chapter 13: Examples 541
Defense
Many of the previous cards will be used to initialize the density and energy (TICVAL) and material (DMAT/EOSGAM) in this Euler region: TICVAL,4,,DENSITY,107E12,SIE,3.9e12 TICVAL,5,,DENSITY,1.29E12,SIE,1.938e11 MATDEUL,230,1.29e12,203,,,,,,+ +,,1.01 EOSGAM,203,1.4 D. The Ground: The ground is modeled as rigid body using dummy QUAD elements. It is used to close the Euler boundary under the vehicle so the blast wave will reflect on this boundary: PSHELL,999,999,1. MATRIG,999,,,,1.0E10,0.00,0.00,800.,+ +,1.E10,0.0,0.0,1.E10,0.0,1.E10,,,+ +,,,,,,,,,+ +,,,,1,7,7 E. Fluid Structure Interaction: In order to make fluid structure interaction possible, a closed volume needs to be defined. The car model itself is not closed, so a dummy boundary will be defined to close the volume. This extra surface consists of three parts: Part 1 resides on the back, Part 2 is the top cover, and Part 3 is the vent on the bottom of the vehicle. For all these parts, dummy shell elements are defined and hole definitions will be defined.
Figure 1325
Dummy Shell Elements Defined to Close the Volume
The input for dummy shell elements PSHELL,900,901,1. PSHELL,910,901,1. PSHELL,920,901,1. MATD009,901,1.E20
Main Index
542 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Defense
With this closed volume, the coupling surface can be defined. For each Euler domain, a separate surface is required. However, in this model, the interaction surface consists of the same elements, except for the extra ground elements (pid=999) for the outer Euler domain region 1. The surface definition will make use of the properties of the elements. The outer surface: BCPROP,97,60,61,62,110,135,150,900,+ +,910,920,999 The inner surface: BCPROP,98,60,61,62,110,135,150,900,+ +,910,920 Now the coupling surfaces can be defined. For the outer region, all elements inside the volume are not active. The covered option will, therefore, be set to INSIDE. Attached to this surface will be the first Euler MESH:
COUPLE,1,97,INSIDE,ON,ON,11,,,+
+,,,,,,,,,+ +,,1 The inner Euler domain is constrained by surface 2. For this volume, the outer Euler elements will be covered:
COUPLE,2,98,OUTSIDE,ON,ON,,,,+
+,,,,,,,,,+ +,,2 As discussed before, there are holes in the coupling surface. To this end, a flow definition is required for one of the coupling surfaces. In this example, the flow cards are referenced from the first coupling surface. The input to define flow between the regions is: LEAKAGE,1,11,1,PORFCPL,84,CONSTANT,1.0 BCPROP,1,900 Also, for each of the other two flow surfaces, these set of cards are repeated $ LEAKAGE,2,11,2,PORFCPL,84,CONSTANT,1.0 BCPROP,2,910 $ LEAKAGE,3,11,3,PORFCPL,84,CONSTANT,1.0 BCPROP,3,920 $ Finally, the flow definition itself prescribes that the Euler region from coupling surface 1 is interacting with the Euler region from coupling surface 2: PORFCPL,84,LARGE,,BOTH,2
Main Index
Chapter 13: Examples 543
Defense
F. Miscellaneous: a. Because this model uses the coupling surface interface, the time step safety factor for Eulerian elements has to be .6. However, the Lagrangian elements (the quadratic and triangular elements) determine the timestep, and it is beneficial to use a higher time step safety factor for the Lagrangian elements: PARAM,STEPFCTL,0.9 b. To show results every .0002 seconds the following output request was added: DYPARAM, LSDYNA, BINARY, D3PLOT,.0002 PARAM, CPLSARC,.0002 Results The Figure 1326 below shows the location, value, and time of the maximum acceleration. The stress distribution at this time is also in Figure 1327.
Figure 1326
Acceleration Plot
Main Index
544 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Defense
Figure 1327
Stress Distribution Plot
Abbreviated SOL 700 Input File SOL 700,NLTRAN STOP=1 CEND TITLE= Job name: mine blast (mm/tonne/s/K) IC=1 SPC=1 $ TSTEPNL=1 $ BULK DATA SECTION BEGIN BULK $ Parameter Section $ DYPARAM,RKSCHEME,3 DYPARAM,FASTCOUP DYPARAM,STEPFCTL,0.9 PARAM*,DYINISTEP,.5E7 PARAM*,DYMINSTEP,1.E13 $ $ DYPARAM,LSDYNA,BINARY,D3PLOT,.0002 PARAM,CPLSARC,.0002 $ MESH,1,BOX,,,,,,,+ +,2623.,1403.,903.,6100.,2800.,2150.,,,+ +,30,10,10,,,,EULER,201 $ MESH,2,BOX,,,,,,,+ +,2621.,1201.,251.,5900.,2400.,1250.,,,+ +,30,10,10,,,,EULER,202 $ PEULER1,201,,2ndOrder,101 $ TICEUL1,101,11
Main Index
Chapter 13: Examples 545
Defense
$ TICREG,1,11,SPHERE,400,230,4,20. TICREG,2,11,SPHERE,501,230,5,1. $ SPHERE,400,,1797.5,0.,450.,250. SPHERE,501,,0.,0.,5000.,10000. $ PEULER1,202,,2ndOrder,102 $ TICEUL1,102,12 $ TICREG,3,12,SPHERE,502,230,5,5. $ SPHERE,502,,0.,0.,5000.,10000. $ TICVAL,4,,DENSITY,107E12,SIE,3.9e12 TICVAL,5,,DENSITY,1.29E12,SIE,1.938e11 $ MATDEUL,230,1.29e12,203,,,,,,+ +,,1.01 $ EOSGAM,203,1.4 $ FLOWDEF,202,,HYDRO,,,,,,+ +,FLOW,BOTH $ COUPLE,1,97,INSIDE,ON,ON,11,,,+ +,,,,,,,,,+ +,,1 $ $ Define flow thru the holes $ LEAKAGE,1,11,1,PORFCPL,84,CONSTANT,1.0 BCPROP,1,900 $ LEAKAGE,2,11,2,PORFCPL,84,CONSTANT,1.0 BCPROP,2,910 $ LEAKAGE,3,11,3,PORFCPL,84,CONSTANT,1.0 BCPROP,3,920 $ PORFCPL,84,LARGE,,BOTH,2 $ COUPLE,2,98,OUTSIDE,ON,ON,,,,+ +,,,,,,,,,+ +,,2 $ BCPROP,97,60,61,62,110,135,150,900,+ +,910,920,999 $ BCPROP,98,60,61,62,110,135,150,900,+ +,910,920 $ $ ========== PROPERTY SETS ==========
Main Index
546 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Defense
$ $ * pbar.9988 * $ PBAR 9988 222 3600.1000000.1000000.2000000. $ $ * pbar.9989 * $ PBAR 9989 222 100000. 3.E+8 3.E+8 6.E+8 $ $ * pbar.9990 * $ PBAR 9990 222 3000. 200000.2500000.3000000. $ $ * pbar.9993 * $ PBAR,9993,111,459.96,25066.,55282.,16543. $ $ * pbar.9996 * $ PBAR,9996,111,895.52,309450.,55349.,48782. $ $ * pbar.9999 * $ PBAR,9999,111,736.,490275.,827555.,2095137. $ $ * pshell.30 * $ PSHELL 30 111 3 $ $ * pshell.40 * $ PSHELL 40 111 4 $ $ * pshell.50 * $ PSHELL 50 111 5 $ $ * pshell.60 * $ PSHELL 60 111 6 PSHELL 61 111 6 PSHELL 62 111 6 $ * pshell.80 * $ PSHELL 80 111 8 $ $ * pshell.110 * $ PSHELL 110 111 11 $ $ * pshell.120 * $ PSHELL 120 111 12 $
Main Index
Chapter 13: Examples 547
Defense
$ * pshell.135 * $ PSHELL 135 111 13.5 $ $ * pshell.150 * $ PSHELL 150 111 15 PSHELL 151 111 15 $ $ * pshell.200 * $ PSHELL 200 111 20 $ $ * pshell.450 * $ PSHELL 450 111 45 $ $ dummy elements for coupling surface $ hole PSHELL,900,901,1. $ top cover PSHELL,910,901,1. $ side cover PSHELL,920,901,1. $ MATD009,901,1.E20 $ $ ground PSHELL,999,999,1. $ MATRIG,999,,,,1.0E10,0.00,0.00,800.,+ +,1.E10,0.0,0.0,1.E10,0.0,1.E10,,,+ +,,,,,,,,,+ +,,,,1,7,7 $ $ * conm2 * $ CONM2,5000,1145,,1.5 CONM2,5001,1146,,1.7 $ $ ========= MATERIAL DEFINITIONS ========== $ MATD024,111,7.85e09,210000.,.3,250E10 $ MAT1,222,210000.,,.3,7.85e09 $ INCLUDE model.bdf INCLUDE ground.dat $ ENDDATA
Main Index
548 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Defense
Blastwave Hitting a Bunker Problem Description
The purpose is to demonstrate application of multiEuler domains to failing coupling surfaces. The problem simulates a bunker, located on the ground that is open at the sides and is surrounded by air. Gas can flow freely through the sides of the bunker. A blast wave is ignited close to the bunker and expands into the air. When by the impact of the blast wave, the bunker surface fails gas will flow trough the bunker surface.
SOL 700 Modeling
The bunker and the ground consist of cquad4 shell elements. The elements of the bunker are Lagrangian deformable shells and the ground is modeled as rigid, using a MATRIG. The explosive/air region is modeled by two Euler meshes. The first domain models the inside of the bunker, and the second one models the outside of the bunker. For the interaction between the bunker and an Euler domain, a unique coupling surface has to be used, therefore, two coupling surfaces are needed. The first coupling surface, for modeling the inside of the bunker, consists of the following facets:
•
The 180 degrees cylindrical surface and the two open sides of the bunker. The two open sides are represented by dummy shell elements. These are elements 1 to 2240. 2241 to 3280.
• The top of the ground that lies within the bunker. This is a square and is formed by elements
These facets make up a closed coupling surface, as shown in Figure 1328.
Main Index
Chapter 13: Examples 549
Defense
This coupling surface contains gas inside, and therefore Euler elements outside the coupling surface should not be processed and so the COVER is OUTSIDE.
Figure 1328
Coupling Surface 1
The second coupling surface consists of the following facets:
• The 180 degrees cylindrical surface and the two open sides of the bunker. These are elements 1
to 2240. The top of the ground inside the bunker is not part of the second COUPLE.
• The top of the ground that is outside the bunker and 5 dummy surfaces of the ground that are
used to close the coupling surfaces. These are formed by the elements 3413 to 4012, 4095 to 4340, 4505 to 4709, 4894 to 7904. These facets make up a closed coupling surface, as shown in Figure 1329.
Figure 1329
Coupling Surface 2
Main Index
550 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Defense
This coupling surface is used for simulating the gas outside the coupling surface. So Euler elements inside the coupling surface should not be processed and the COVER has to be set to INSIDE. The second coupling surface uses the second Euler mesh and serves as inner boundary surface for this Euler mesh. The outside boundary of this mesh is where the Euler domains ends and boundary conditions for this boundaries are provided by a FLOWDEF. The FLOWDEF is chosen as nonreflecting. Waves exit the Euler domain with only little reflection. To get an accurate expansion of the blast wave, the diffusion should be kept at a minimum, and therefore the Roe solver with secondorder is used. Interactive failure will be used for the bunker structure, while porosity will be used for the open sides:
• The bunker elements can fail and gas flows through the failed elements from outside the bunker into the bunker. All elements of the bunker are assigned to a BSURF, and occur in both coupling surfaces. They are able to fail interactively, using the COUP1FL entry. These parts are formed by elements 1 to 1600. The nodes of the failed elements are constrained in space by using PARAM, NZEROVEL, YES, to preserve the geometry of the coupling surfaces from severe distortion. •
Since gas can flow through the two sides without any obstruction, these two areas are modeled with BSURF entries, and are opened by using a PORFLCPL entry. These sides are modeled with dummy shell elements and consist of elements 1601 to 2400.
The couple cards refer to meshnumber. The first mesh for the Euler elements inside the bunker is created and initialized by:
PEULER1,301,,2ndOrder,111
MESH,2,BOX,,,,,,,+ +,430.,0.,1287.,837.,480.,1296.,,,+ +,24,16,30,,,,EULER,301 The value "2ndOrder" activates the Roe solver with secondorder accuracy. The property id is the link between the TICEUL1 card 101 and the mesh card. The second Euler mesh for the Euler elements outside the bunker is created and initialized by:
PEULER1,201,,2ndOrder,101
MESH,1,BOX,,,,,,,+ +,647.,0.,1293.,1057.,447.,1293.,,,+ +,33,23,37,,,,EULER,201
Main Index
Chapter 13: Examples 551
Defense
Results Figures 1330 and 1331 show a fringe plot and an isosurface. Figure 1331 has been created by Ensight.
Figure 1330
Deformed Effective Stress Plot of the Bunker
Figure 1331
Isosurfaces Created using SIE Variable for the Two Euler Domains
Abbreviated SOL 700 Input File
SOL 700,NLTRAN STOP=1 CEND TITLE= Job name is: bunker IC=1 SPC=1 TSTEPNL=1 $ BEGIN BULK $ BULK DATA SECTION $
Main Index
552 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Defense
$INCLUDE mesh.dat INCLUDE model.dat $ TSTEPNL 1 20 0.0005 1 $ Parameter Section DYPARAM,FASTCOUP,,FAIL PARAM*,DYINISTEP,1E7 PARAM*,DYMINSTEP,1E8 DYPARAM,LIMITER,ROE DYPARAM,RKSCHEME,3 DYPARAM,LSDYNA,BINARY,D3PLOT,0.002 $ $ ========== PROPERTY SETS ========== $ $ * steel prop * $ PSHELL 1 1 .150 $ $ * dummy_shell * $ PSHELL,2,2,1E3 MATD009,2,1E20 $ $ PSHELL 3 4 .100 $ $ ========= MATERIAL DEFINITIONS ========== $ $ $  Material steel id =1 MATD024,1,.000734,2.9e+07,.3,50000,,.21 $ $  Material AIR id =3 MATDEUL 3 1.2e07 3 EOSGAM 3 1.4 $ $  ground MATRIG 4 .000734 2.9e+07 .3 $ $ ======== Load Cases ======================== $ $  General Coupling: GENERAL $ $ COUPLE 7 1 INSIDE ON ON + + 1 1 $ BSURF 1 1 THRU 2240 3413 + THRU 4340 4505 THRU 4709 $ $ COUP1FL,1,1.2e07,3e+08 $
16
+ + 4012 THRU 4095+ 7904
THRU 4894
Main Index
Chapter 13: Examples 553
Defense
COUPLE 8 11 OUTSIDE ON ON + + 2 2 $ BSURF 11 1 THRU 3280 $ $ COUP1FL,2,1.2e07,3e+08 $ $ COUPINT,1,7,8 $ PORFCPL,81,LARGE,,BOTH,8 LEAKAGE,1,16,32,PORFCPL,81,,1.0 BSURF 32 1601 THRU 2240 $ $  Rigid Body Constraints $ SPCD2 1 RIGID MR4 1 0 1 SPCD2 1 RIGID MR4 2 0 1 SPCD2 1 RIGID MR4 3 0 1 SPCD2 1 RIGID MR4 5 0 1 SPCD2 1 RIGID MR4 6 0 1 SPCD2 1 RIGID MR4 7 0 1 TABLED1 1 + 0. 1. 1. 1. ENDT $ $Mesh.dat$ MESH,1,BOX,,,,,,,+ +,647.,0.,1293.,1057.,447.,1293.,,,+ +,33,23,37,,,,EULER,201 $ $ Inner Euler $ MESH,2,BOX,,,,,,,+ +,430.,0.,1287.,837.,480.,1296.,,,+ +,24,16,30,,,,EULER,301 $ PEULER1,201,,2ndOrder,101 PEULER1,301,,2ndOrder,111 $ $ TICEUL1 101 101 TICREG 1 101 SPHERE 8 3 TICREG 2 101 SPHERE 5 3 4 $ SPHERE 8 536.4 165 453.6 SPHERE 5 536.4 165 453.6 $ $ TICEUL1 111 111 TICREG 3 111 SPHERE 9 3 $
+ +
0. 0. 0. 0. 0. 0. +
5 1 85 10000
2
4
1
Main Index
554 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Defense
SPHERE 9 53.4 100 $ $ $  TICVAL BC AIRINI TICVAL 4 DENSITY 1.2e07 $ $  TICVAL BC EXPINI TICVAL 5 DENSITY3.84e06 $ $ FLOWDEF,202,,HYDRO,,,,,,+ +,FLOW,BOTH $ ENDDATA
673.6
10000
SIE SIE
3e+08 3e+09
Main Index
Chapter 13: Examples 555
Time Domain NVH
Time Domain NVH
Chassis
Description This is an example of a Time Domain NHV simulation on a pickup truck chassis. A chassis of a car was modeled and an impulse loading was applied at one of front corner points. Time histories were obtained at 12 points and they were translated to a frequency domain by applying Fast Fourier Transform (FFT) to obtain dynamic properties such as mode shapes and frequencies. SOL 700 Entries Included SOL 700 TSTEPNL TIMNVH PARAM,S700NVH1 TIMNAT
Main Index
556 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Time Domain NVH
Model The chassis consists of 1986 grid points and 1996 quadratic shell elements. An impact loading is applied at the end point of upper picture. The impact loading which is defined by FORCE and TABLE entries starts 0 at 0 seconds, goes up to 0.01 at 0.001 second and down to 0 at 0.002 second like lower figure. There is no boundary condition. The units are g, mm, and second.
Figure 1332
Impluse Load applied at One Corner of the Plate
Time Domain NVH Scheme
MD Nastran SOL 700 (impluse loading) Time history  Displacement  Velocity  Acceleration (default)
Time domain > Frequency domain using FFT
Extract dynamic properties: Natural frequencies and Mode shapes
Main Index
Chapter 13: Examples 557
Time Domain NVH
Input On node 902517, the impact loading is applied. This analysis is done by the following three steps. Step 1: Find modal properties using TIMNVH entry Step 2: Check the obtained modal properties and select required natural frequencies Step 3: Rerun with selected natural frequencies Input file timnvh.bdf SOL 700,NLTRAN path=3 stop=1 SOL 700 is the initiating command of an explicit nonlinear transient analysis. Case control section is below. CEND LOADSET = 1 SUBCASE 1 TSTEPNL = 1 SPC = 2 DLOAD = 2 Bulk entry section starts. BEGIN BULK TSTEPNL 1 100 .01 1 ADAPT 2 TIMNVH,1, , , 1.0, 1000., 3,.0005, 2,+ +, 0, 3, 1, 0.015, 0, 3, 13, .0030,+ +, 901581, 901641, 901697, 901865, 902061, 902097, , ,+ +, 902580, 902595, 902609, 902797, 902996, 903063
10
TSTEPNL is a SOL 700 bulk data entry which describes the number of Time Steps (100) and Time Increment (10 ms) of the simulation. End time is the product of the two entries. Notice here the Time Increment is only for the first step. The actual number of Time Increments and the exact value of the Time Steps is determined by MD Nastran during the analysis. The time step is a function of the smallest element dimension during the simulation. TIMNVH entry is for Time NVH analysis.
Detail for TIMNVH entry: TIMNVH, 1, , , 1.0, 500., 3, 0.00005, 2,+
Main Index
558 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Time Domain NVH
The range of natural frequencies to obtain is from 1.0 Hz to 500 Hz and only ztranslational degrees of freedom is considered (3). The sampling rate is 0.00005 seconds. The peaking criterion is two, which means a peak is selected if the number of increasing and decreasing amplitude around a peak is over 2. +, 0, 3, 1, 0.015, 0, 3, 13, .0030,+
Acceleration is selected for the response (0) and translational eigenvectors are only requested as ASCII format (3). Eignevalues are normalized by 1.0 (1) and 0.015 is selected as a tolerance value which means if there are two modes which distance is smaller than 0.015 Hz, it is assumed to be the same mode. ACII file format of natural frequencies and eigenvalues are requested (0) as well as translational time histories of zdirection (3). Frequencyamplitude data of zdirection are requested (13) and a peak which amplitude is less than 0.0030×the maximum amplitude is ignored (.0030) +, 901581, 901641, 901697, 901865, 902061, 902097, , ,+ +, 902580, 902595, 902609, 902797, 902996, 903063 The grid points (901581, 901641, 901697, 901865, 902061, 902097, 902580, 902595, 902609, 902797, 902996 and 903063) are only considered for Time NVH analysis. Bulk data entry that defines grid points and elements. CQUAD4 .. GRID .. 901092 64011 5 901572 901573 901705 901704 0. 0.
1711.76 595.649577.221
Bulk data entry that defines properties for shell elements with 3.5mm thickness:. PSHELL .. 5 63 3.5 63 63
Bulk data entry that defines applying forces. TLOAD1 LSEQ DLOAD FORCE 4 1 2 3 5 5 3 1. 1. 902517 0 1 4 .01 0. 0. 1.
Main Index
Chapter 13: Examples 559
Time Domain NVH
Bulk data entry that defines tables. TABLED1 .. ENDDATA $ Nodes of the Entire Model $ Loads for Load Case : Default SPCADD 2 1 $ Displacement Constraints of Load Set : fix SPC1 1 2 903448 903909 904202 Input file chassis.bdf: This input is for the refinement of the selection of modal frequencies and mode shapes. Only different part is shown PARAM,S700NVH1,1 TSTEPNL 1 100 .01024 1 ADAPT 2 TIMNVH,1, , , 1.0, 1000., 3,.0005, 2,+ +, 0, 3, 1, 0.015, 0, 3, 13, .0030,+ +, 901581, 901641, 901697, 901865, 902061, 902097, , ,+ +, 902580, 902595, 902609, 902797, 902996, 903063 TIMNAT,1,35.,43.,49.,101.,108. 1 10. 10.
0. 0.
0. ENDT
0.
.001
1.
.002
0.
904558
10
PARAM, S700NVH is for the rerun of Time Domain NVH analysis. Using 1 as option Time Domain NVH analysis is carried out without rerunning SOL 700.
The PEAK value (original: 2) in TIMNVH entry is changed to 2 to use TIMNAT entry.
TIMNAT entry is for the control of the natural frequency selection. In this job, 35, 43, 49, 101, 108 Hz are
selected to get the results. Results In this analysis, frequency resolution is just 1 Hz because end time of the analysis is defined by 1.0 second. As increasing an end time, the accuracy may increases. There are three types of new results file from Time NVH analysis. 1. mode.out: the natural frequencies and eigenvalues selected are restored. 2. amplfreq009018653.txt : amplitudefrequency output of DOF =3 at grid point 901865. 3. timehist009018653.txt: time history output of DOF =3 at grid point 901865.
Main Index
560 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Time Domain NVH
From the amplfreq*** files, the frequencyamplitude plots are obtained.
1.00E+02
4th mode ≈ 101. 1st mode ≈ 35.
1.00E+01
2nd mode ≈ 43. 3rd mode ≈ 49.
5th mode ≈ 108.
1.00E+00 node 901581 node 902996 node 902797 1.00E01 node 902609 node 902595 node 902580 node 902097 1.00E02 node 902061 node 901865 node 901697 node 901641 1.00E03 0 20 40 60 80 100 120 node 903063 140
Comparison of natural frequencies between SOL 103 and SOL 700 (Hz) Mode 1 2 3 4 5 6 7 SOL103 3.601695E+01 4.395227E+01 5.250646E+01 6.742810E+01 8.472197E+01 1.019688E+02 1.110159E+02 SOL 700 3.500018E+01 4.300022E+01 4.900025E+01 Small peak Small peak 1.010005E+02 1.080005E+02 0.95% 2.72% Diff(%) 2.82% 2.17% 6.68% Comparison Vertical motion dominant Vertical motion dominant Lateral motion dominant Lateral motion dominant Lateral motion dominant Vertical motion dominant Vertical motion dominant
Since an impact is applied to vertical direction and the peaks are detected by the vertical direction acceleration, the horizontal dominant modes are hard to catch.
Main Index
Chapter 13: Examples 561
Time Domain NVH
In mode.out file
1st Mode Natural Frequency
MODES EIGV
1st Mode
1 5 1 3.500018E+01 9015813.32998498E022.49243337E04 7.08997618E01 9016414.29914555E02 7.70991520E051.08571907E01 9016974.15069142E02 2.55256359E046.31611930E01 901865 4.37855265E021.51550001E044.18557096E01 902061 7.97601410E02 4.34427876E04 5.67705213E01 902097 8.68013598E02 8.02417982E03 1.00000000E+00 9025803.38588683E02 2.97715028E04 7.28400224E01 9025954.37831381E02 2.30181446E049.77437006E02 9026094.24521220E021.61168521E046.35288211E01 902797 4.11242103E023.00773060E044.29582120E01 902996 7.69986448E02 7.40153667E04 5.51699503E01 903063 8.41026922E023.47784987E03 9.82653769E01
..
Output Node Number xeigenvector yeigenvector zeigenvector
Comparison of mode shapes between SOL 103 and SOL 700 Mode SOL 103 SOL 700
1
2
Main Index
562 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Time Domain NVH
Mode
SOL 103
SOL 700
3
4
5
Main Index
Chapter 13: Examples 563
Time Domain NVH
Mode 6
SOL 103
SOL 700
7
Because the vertical mode shapes of mode 2 and 3 are very similar although the lateral mode shapes and the amplitudes of vertical mode are quite different, the mode shapes 2 and 3 obtained from SOL 700 are nearly similar.
Vertical Mode Shape of Mode 2
Vertical Mode Shape of Mode 3
Main Index
564 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Prestress
Prestress
Simulation of Prestress and Impact on Rotating Fan Blades
Description Aerospace companies perform bird strike test simulation to predict the impactresistance properties of the aircraft structure. This is an example of a solid object, impacting against rotating fan blades. The radius of the rotor and the fan blades are respectively 0.040 m and 0.195 m. The rotational speed of the rotor is 10 rev/s. The impactor is a cube of 0.0 4 x 0.0215 x 0.05 m and its velocity is 50 m/s (112 mph). Material model MATD024 was used to simulate the elasticplastic properties of the impactor with parameters as follows:
ρ = 1000kg ⁄ m 3 σ y = 2.E + 7 P a ETAN = 5 E + 7 P R = 0.3 E h = 2.1 E + 10P a
Plastic strain failure limit 0.15
To model the fan blades and the rotor MATD024 material was used as:
ρ = 7800kg ⁄ m 3 σ y = 2.E + 8 P a ETAN = 5 E + 8 P R = 0.3 E h = 2.1 E + 11P a
Plastic strain failure limit 0.15
SOL 700 Entries Included SOL 700 TSTEPNL PARAM, DYDEFAUL, DYNA DYPARAM*, LSDYNA PRESTRS ISTRSSH SPCD2 TIC3 BCTABLE BCBODY BCGRID BSURF DLOAD MATD024 RFORCE TIC TLOAD1
Main Index
Chapter 13: Examples 565
Prestress
Model
The model has a total of 3173 grid points and 2950 elements as follows:
• 2800 Quads • 150 Hexas
The rotor and fan blades are modeled by shell elements. The impactor is modeled with solid elements. The relative speed of the impactor to the rotor is modeled by defining an initial velocity of 50 m /s (112 mph), applied on all the grid points of the impactor. The simulation time is 0.01 seconds. The unit system is m, N, sec. Input The simulation consists of two runs. The first run is a presstress analysis that computes the deformations and stresses due to rotation. This computation is essentially static. Conducting this computation in the impact run would require structural relaxation. Since this would interfere with the impact process the static part of the computation is done by an implicit solver in a presstress run. Boundary conditions and initial conditions of the prestress run differ from the second run. In the prestress run the rear of the rotor is fixed and a force in circumferential direction is applied to the rotor and fan blades. In the impact run the rear of the rotor is given an angular velocity and the impactor is given an initial velocity. Contacts are defined between the fan blades and the impactor. Input file SOL 700,NLTRAN stop=1
Main Index
566 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Prestress
SOL 700 is a executive control entry similar to SOL 600. It activates an explicit nonlinear transient
analysis using SOL 700 solver. First consider the prestress input deck. In this input deck only the rotor and fan blades are included. LOADSET = 2 TITLE = MD Nastran job created on 07Dec06 at 16:50:17 $ Direct Text Input for Global Case Control Data SUBCASE 1 TITLE=This is a default subcase. TSTEPNL = 1 SPC = 1 DLOAD = 3 DISPLACEMENT(SORT1,REAL)=ALL SPCFORCES(SORT1,REAL)=ALL STRESS(SORT1,REAL,VONMISES,BILIN)=ALL The bulk entry section begins with TSTEPNL 1 5 1.5 1 ADAPT 2 10
TSTEPNL is a SOL 700 bulk data entry which describes the number of Time Steps (5) and Time Increment (1e5 sec) of the simulation. End time is the product of the two entries. Notice here the Time Increment is only for the first step. The actual number of Time Increments and the exact value of the Time Steps is determined by SOL 700 during the analysis. The time step is a function of the smallest element dimension during the simulation.
By using the SOL 700 bulk data entry PRESTRS, a prestress analysis is carried out. Here the implicit MD Nastran solver is used. This solver requires the analysis to be run with double precision executable. Final deformations and stresses of elements are written to a text file named imput_file_name.nastin to provide initial conditions for rotor and fan blades of the impact run. Bulk data entries that define properties for shell elements PSHELL1 1 + 0.0025 1 0.0025 BLT 0.0025 GAUSS 0.0025 2
These elements model the fan blades and rotor. Bulk data entry that defines material properties for rotor and fan blade $ Material Record : mat1.1 MATD024 1 7800. 2.1+11
.3
2.+8
5.+8
.15
MATD024 is a SOL 700 bulk data entry that models arbitrary elastoplastic material.
Main Index
Chapter 13: Examples 567
Prestress
Bulk data entry that defines geometric properties of the rotor and blades: $ Pset: "p1" will be imported as: "pshell.1" CQUAD4 1 1 77 78 6 .. .. $ Nodes of the Entire Model Bulk data entry that defines boundary conditions and loads $ Loads for Load Case : Default $ Displacement Constraints at rear of rotor SPC1 1 123456 21 THRU 31 .. .. $RFORCE applied to rotor + fan blades TLOAD1 5 6 1 LSEQ 2 6 1 DLOAD 3 1. 1. 5 TABLED1 1 ++00003D ++00003D 0. 1. .001 1. ENDT +00003E RFORCE 1 7000 10.00 0.
5
0.
1.
RFORCE is a bulk data entry that defines a static loading condition due to angular velocity.
The following parameters are needed to get optimal results. DYPARAM,LSDYNA,ACCURACY,OSU,1 DYPARAM,lSDYNA,ACCURACY,INN,2 $ $ control_ENERGY $ DYPARAM,LSDYNA,ENERGY,HGEN,2 DYPARAM,LSDYNA,ENERGY,RWEN,1 DYPARAM,LSDYNA,ENERGY,SLNTEN,1 DYPARAM,LSDYNA,ENERGY,RYLEN,1 $ $ control_HOURGLASS $ DYPARAM,LSDYNA,HOURGLASS,IHQ,1 DYPARAM,LSDYNA,HOURGLASS,QH,0.100000001 DYPARAM,LSDYNA,DATABASE,D3PLOT,0.1 PARAM,DYDEFAUL,DYNA
Main Index
568 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Prestress
Now consider the impact run. $ Direct Text Input for Executive Control CEND TITLE = JOBNAME IS: BIRD_STRIKE2 SUBCASE 1 $ Subcase name : Default SUBTITLE=Default TSTEPNL = 1 BCONTACT = 1 SPC = 2 IC = 1 DISPLACEMENT(SORT1,REAL)=ALL SPCFORCES(SORT1,REAL)=ALL The bulk entry starts with
$ BULK DATA SECTION BEGIN BULK TSTEPNL 1 100 1e4 1
ADAPT
2
10
TSTEPNL is a SOL 700 bulk data entry which describes the number of Time Steps (100) and Time Increment (1e4 sec) of the simulation. End time is the product of the two entries. Notice here the Time Increment is only for the first step. The actual number of Time Increments and the exact value of the Time Steps is determined by SOL 700 during the analysis. The time step is a function of the smallest element dimension during the simulation.
INCLUDE prestr10.dytr.nastin The inputdeck of the impact run includes the nastran file prestr10.dytr.nastin. This makes the shell elements that are defined in the prestress run available to the impact run. The only properties of the rotor and fan blade that need to be defined in the impact run are material properties and boundary conditions. The file prestr10.dytr.nastin contains the entry ISTRSSH. This entry specifies what result variables of the prestress run are to be carried over to the impact run. Bulk data entries that define Contact relations and Contact bodies: $ Define contact BCTABLE 1 SLAVE 3 0
0. 0
4 0. 0
0.
0.
0
0.
MASTERS 3 ..
Main Index
Chapter 13: Examples 569
Prestress
BCTABLE is a SOL 700 bulk data entry that is also used by SOL 600 and it is meant to define MasterSlave as well as self contact. In addition to contact between impactor and fan blade there is also self contact defined.
$ Deform Body Contact LBC set: impactor BCBODY 3 3D DEFORM 3 0 BSURF 3 2116 2117 2118 2119
2120
2121
2122
BCBODY is a bulk data entry that is used by SOL 600 and SOL 700 only, which defines a flexible or rigid contact body in 2D or 3D. It could be specified with a BSURF, BCBOX, BCPROP, or BCMATL entry.
$ Deform Body Contact LBC set: blade BCBODY 4 3D DEFORM 4 BSURF 4 2801 2802 2803 2807 .. ..
0 2804
2805
2806
BSURF is a SOL 700 bulk data entry that is used by SOL 600 and SOL 700 only, which defines a contact surface or body by element IDs. All elements with the specified IDs define a contact body.
Bulk data that defines the properties of elements: $ Elements and Element Properties for region : p2 PSOLID 2 2 0
PSOLID is a SOL 700 bulk data entry that defines properties for solid elements. It refers to entry MATD024 for material properties.
Bulk data entries that define material properties for the shell and solid elements: $ shell elements $ Material Record : mat1.1 MATD024 1 7800. 2.1+11 $ solid elements $ Material Record : mat1.2 MATD024 2 1000. 2.1+10
.3
2.+8
5.+8
.15
.3
2.+7
5.+7
.15
Bulk data entry that defines geometric properties of the impactor: $ Pset: "p2" will be imported as: "psolid.2" CHEXA 2801 2 2921 2922 2928 2927 2958 2964 2963 .. .. $ Nodes of the Entire Model GRID 2921 .146905 .0092747.023994 ..
2957
Main Index
570 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Prestress
The shell elements are defined through the Nastran initialization file prestr10.nastin. Bulk data entry that defines boundary conditions and initial velocity: BCGRID + + 1 THRU 21 2073 THRU 2741 31 1346 THRU THRU 2749 1355 2064
BCGRID is also used to specify a set of nodes which are to be constrained with SPCD2 (angular velocity).
$ Loads for Load Case : Default $ Initial velocvity of impactor $ Initial Velocities of Load Set : inivel_solid TIC 1 2921 3 50 TIC 1 2922 3 50 .. .. $ Displacement Constraints of Load Set : Disp1 SPC1 1 3 21 THRU 31 .. .. $ Initial angular velocity for rotor +fan blade TIC3 1 70000 1. + + 62.832 + 1 THRU 2920 $ $ Imposed angular velocity at the rear of the rotor SPCD2 1 GRID 1 7 80 1 .. .. and the referenced table TABLED1 80 + + 0.0 62.832 1. 62.832 ENDT $
TIC is a bulk data entry that defines values for the initial conditions of variables used in structural transient analysis. Both displacement and velocity values may be specified at independent degrees of freedom. TIC3 defines the initial velocity as combination of translational part and a rotational part. SPCD2 defines imposed nodal motion on a node or a set of nodes. INCLUDE inserts an external file into the input file.
End of input file.
Main Index
Chapter 13: Examples 571
Prestress
Results Prestress Run In accordance to the TSTEPNL entry 5 results increments were archived. The results of all increments are essentially the same which indicates that the implicit calculations are stable. The results of the last increment were written to the file prestres10.dytr.nastin. Result Increment 2
Main Index
572 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Prestress
Result Increment 5: written to the .nastin file.
Impact Run: The prestress result variables have been initialized at the begin of the analysis (Time = 0) t = 0 sec
Main Index
Chapter 13: Examples 573
Prestress
t = 0.0001 sec
t = 0.001 sec
Main Index
574 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Prestress
t = 0.002 sec
t = 0.005 sec
Main Index
Chapter 13: Examples 575
Prestress
t = 0.01 sec
Main Index
576 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Smooth Particle Hydrodynamics (SPH)
Smooth Particle Hydrodynamics (SPH)
Ball Penetration using SPH Method
Description This is an example of a ball penetrating through a plate using SPH method. The initial velocity of the ball is 6.18 km/s (0.618 cm/µs) in vertical direction. In the simulation, the center part of the plate and the ball projectile are modeled by SPH elements. SOL 700 Entries Included SOL 700 TSTEPNL DYPARAM,LSDYNA,BINARY,D3PLOT CSPH PSPH EOSGRUN SPHDEF TIC MATD010 PSOLIDD MATD003
Main Index
Chapter 13: Examples 577
Smooth Particle Hydrodynamics (SPH)
Model The circular section located at the center of plate was modeled using SPH elements and the remainder of the plate was modeled using structural material property (MATD003) in the example. The ball is modeled by SPH elements and is impacting the the center of plate at 6.18 km/s . The model has a total of 19479 grid points, 300 solid elements and 18759 SPH particle elements. All the structural elements were Hexas. The simulation time is 20.0 seconds. Input file SOL 700,NLTRAN stop=1
SOL 700 is an executive control card similar to SOL 600. It activates an explicit nonlinear
transient analysis. Case control section is below: DLOAD = 1 IC = 1 SPC = 1 BCONTACT = 1 TSTEPNL = 1 The bulk entry section starts: BEGIN BULK $ TSTEPNL*1 * $ $ DATABASE_BINARY
6000
3.3333334E03
*
TSTEPNL is a SOL 700 bulk data entry which describes the number of Time Steps (6000) and Time Increment (3.33 ms) of the simulation. End time is the product of the two entries. Notice here the Time Increment is only for the first step. The actual number of Time Increments and the exact value of the Time Steps are determined by SOL 700 during the analysis. The time step is a function of the smallest element dimension during the simulation. LSDYNA,BINARY,D3PLOT option of DYPARAM entry controls the output time steps of d3plot binary file.
The result plots at every 0.5 seconds are stored in d3plot binary file. $ $ Number of SPH = CSPH* * 10000001
18759 11 8.2517613E04 *
Main Index
578 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Smooth Particle Hydrodynamics (SPH)
CSPH entry defines a SPH particle. The SPH element number (10000001) must correspond to the
grid number which describes the SPH element location. The lumped mass (8.25E4) is applied on the SPH element. $ $ Part = material type #10 $ PSPH* 10 2 * * 1.200000 0.2000000 * 0.0 0.0
2 4.000000 0.0
* * *
PSPH entry defines the property of SPH particle. Both of the material property and equation of state are
set to 2. The smoothing length of the particles is set to 1.2. The scale factors for the minimum and maximum smoothing length are set to 0.2 and 4.0, respectively. $ $ Number of EOS = $ EOSGRUN*2 * 0.0 * 0.0 *
2 0.5328000 2.000000 1.339000 0.4800000 0.0 0.0 * * *
EOSGRUN entry defines a Gruneisen Equation of State.
The Gruneisen equation of state defines pressure for compression as
γ0 a ρ 0 C 2 μ 1 + ⎛ 1 –  μ –  μ 2⎞ ⎝ 2 2 ⎠ p =  + ( γ 0 + a μ )E 2 μ2 μ3 1 – ( S 1 – 1 )μ – S 2  – S 3 μ+1 ( μ + 1 )2
And for tension as
p = ρ 0 C 2 μ + ( γ 0 + aμ ) E
All fields are set for the coefficients of two equations above. Please see MD Nastran Quick Reference Guide for details. SPHDEF* 1 * 0 * * 0 0 0.0 0.0 * * *
0.0
Main Index
Chapter 13: Examples 579
Smooth Particle Hydrodynamics (SPH)
SPHDEF entry defines and controls the physics of SPH particles. All values of this example are assigned using default values. See MD Nastran Quick Reference Guide for details. $ MATD003*1 * 2.9000000E03 * 0.0 *
2.785000 7.0000000E02 0.0
0.7000000 1.000000 0.0
0.2690000 0
* * *
MATD003 entry defines an isotropic and kinematic hardening plastic material including rate effects. This
material is used to model the boundary structural plate in the example. The density is 2.785 kg/cm3 and Young’s modulus is 0.7 GPa. The Poisson’s ratio, the yield stress and the tangential modulus are set to 0.269, 0.0029 GPa and 0.07 GPa, respectively. The hardening parameter is set to 1, which describes the isotropic hardening only. $ $ *INITIAL_VELOCITY_NODE $ TIC* 1 10000001 * 0.0
…
1
*
$ $ *INITIAL_VELOCITY_NODE $ TIC* 1 100001 * 0.0 TIC* 1 100001 * 0.6180000
…
1 2
* *
MATD024 is a SOL 700 bulk data entry. It was used to model an elastoplastic material with an arbitrary
stress versus strain curve and arbitrary strain rate dependency. Failure can also be defined based on the plastic strain or a minimum time step size. Bulk data entries that defines initial velocity of the rod:
TIC entry defines a nodal initial condition. In the example, all SPH grids have initial velocity conditions.
Grid point 10000001 is located at the center of the plate and has zero velocity to all directions. Other nodes on the center of the plate have zero velocities similar to grid point 10000001. Grid point 100001 is located at the ball which has a velocity of 0.618 cm/s in y direction. All nodes on the ball have the same velocity as grid point 100001. $ MATD010 2 0.0 0.0 0.0 0.0
2.785 0.0 0.0 0.0 0.0
0.269 0.0 0.0 0.0 0.0
2.9E03 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
2.0E2 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0
Main Index
580 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Smooth Particle Hydrodynamics (SPH)
MATD010 entry defines an elasticplastic hydrodynamic material. This material is used to model SPH
elements. The density, shear modulus, yield stress and cutoff pressure are set to 2.785 kg/cm3, 0.269 GPa, 0.0029 GPa and 0.02 GPa, respectively. $ $ Part = material type # 3 $ PSOLIDD*1 1 *
0
*
PSOLIDD entry defines a solid element property using element formulation and equation of state. In the example, this entry can be changed to PSOLID because only material property is assigned.
Bulk data entries that define grid points and solid type elements.
$ $ Number of Nodes = 19479 $ GRID 1 0.3E+1 0.404 0.30E+1 .. GRID 101791 0.667E10.467 0.467 $ $ Number of Solid elements = 300 $2345678$2345678$2345678$2345678$2345678$2345678$2345678$2345678$2345678 CHEXA 1 1 13 14 20 19 1 2 8 7 .. CHEXA 300 1 714 715 720 719 704 705 710 709
Bulk data entries that define parameters. $ $ control_OUTPUT $ DYPARAM*ELDLTH * .. DYPARAM*LSDYNA * $ ENDDATA End of input file.
100 ENERGY RYLEN 1
* *
Main Index
Chapter 13: Examples 581
Smooth Particle Hydrodynamics (SPH)
Results
t = 0.00 s
t = 1.98 s
t = 3.98 s
t = 5.98 s
t = 8.99 s
Figure 1333 Ball Penetration at Various Increments
t = 20.0 s
Main Index
582 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Sheet Metal Forming
Sheet Metal Forming
Square Cup Deep Drawing using Forming Limit Diagram
Description This is a sheet metal forming example of a plate with anisotropic behavior that is drawn through a square hole by means of a punch. This particular example has experimental results as it was provided as a verification problem for participants of the 1993 NUMISHEET Conference held in Japan. The analysis involved results obtained at single punch depth (20 mm punch travel) for an aluminum alloy plate. The material is seen to be anisotropic in its planar directions; i.e., the material behavior is different in all directions in the plane of the sheet metal as well as in the out of plane direction. The data obtained from the NUMISHEET Conference were as follows: Aluminum Alloy Thickness = 0.81 mm Young’s modulus = 71 GPa Poisson’s ratio = 0.33 Density = 2700 kg/m3 Yield stress = 135.3 MPa Stress = 576.79 * (0.01658 + p)0.3593 MPa Lankford parameters: R0 = 0.71, R45 = 0.58, R90 = 0.70 Friction coefficient = 0.162 Size of the plate modeled was 0.15 x 0.15 (in meters). No strainrate dependency effects were included in the material data, so the metal sheet was analyzed without these effects. The dimensions of the plate, die, punch, and clamp are all given in Figure 1334. SOL 700 Entries Included SOL 700 TSTEPNL DYPARAM,LSDYNA,BINARY,D3PLOT CSPH PSPH EOSGRUN SPHDEF TIC MATD010 PSOLIDD MATD003
Main Index
Chapter 13: Examples 583
Sheet Metal Forming
Figure 1334
Dimensions of Plate, Die, Punch, and Clamp (in Millimeters)
Main Index
584 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Sheet Metal Forming
Model
Figure 1335
SOL 700 Model (Exploded View)
The SOL 700 model is shown in Figure 1335. The main parts in the finite element model are:
• sheet metal • punch • die • clamp
Sheet Metal The SOL 700 material model for sheet metals is a highly sophisticated model and includes full anisotropic behavior, strainrate effects, and customized output options that are dependent on material choice. Since not all of the materials can be derived from the simplified set given by the NUMISHEET organization, most participants in the conference used an isotropic material model. In reality, the process is definitely anisotropic and effect due to these differences can be seen in the transverse direction. For materials displaying inplane anisotropic behavior, the effect would be even more noticeable. The parameters on the MAT190 (refer to the MD Nastran Quick Reference Guide) specify planar anisotropic behavior and are as follows (for the aluminum sheet):
• MATD190 elastic material properties. • Isotropic behavior was assumed in the elastic range:
Exx = 71.0 GPa = 0.33
Main Index
Chapter 13: Examples 585
Sheet Metal Forming
• Planar anisotropic yielding and isotropic hardening were assumed in the plastic range:
A = Stress constant = 0.0 MPa B = Hardening modulus = 576.79 MPa C = Strain offset = 0.01658 n = Exponent for powerlaw hardening = 0.3593
• Lankford parameters:
R0 = 0.71 R45 = 0.58 R90 = 0.70 Punch, Die, and Clamp These three components provide the constraints and driving displacement for the analysis and are modeled as rigid bodies. Contact is then specified with the metal sheet using the friction coefficient values provided. The three contact types are specified as following:
• Contact between the punch and the sheet • Contact between the die and sheet • Contact between the clamp and sheet
Finally, the punch is given a scaled downward velocity providing the driving displacement for the analysis. Input File SOL 700,NLTRAN stop=1 SOL 700 is an executive control card and activates an explicit nonlinear transient analysis. Case control section is below: DLOAD = 1 IC = 1 SPC = 1 BCONTACT = 1 TSTEPNL = 1 The bulk entry section starts: BEGIN BULK $ TSTEPNL 1 $ DYPARAM LSDYNA
20 BINARY
2.0E3 D3PLOT 0.002
Main Index
586 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Sheet Metal Forming
TSTEPNL is a SOL 700 bulk data entry which describes the number of Time Steps (20) and Time Increment (2.00 ms) of the simulation. End time is the product of the two entries. Notice here the Time Increment is only for the first step. The actual number of Time Increments and the exact value of the Time Steps are determined by SOL 700 during the analysis. The time step is a function of the smallest element dimension during the simulation. LSDYNA,BINARY,D3PLOT option of DYPARAM entry controls the output time steps of d3plot binary file.
The result plots at every 0.002 seconds are stored in d3plot binary file. Bulk data entries that define properties for shell elements PSHELL1 1 + .81 1 BLT Gauss +
MATD020 2 1
1.0 4
210.E9 7
0.3
MATD020 entry defines the rigid material property. In the example, the clamp, die, and punch are modeled by the rigid materials.
MATD190 1 2.7E4 7.1E7 0.33 2.0 + 6.0 .71 .58 .70 + 2.0 77 + 1.0 0.0 + 0.0 1.0 TABLED1,77,,,,,,,,+ +,100.0,196.67,0.0,30.,30.,45.,40.,47.,+ +,50.,45.,ENDT
576.79E3.3593 .01658 0.0 0.0
0
+ + + +
MATD190 entry defines an anisotropic material developed by Barlat and Lian (1989) for modeling sheets under plane stress conditions and failure criteria of Forming Limit Diagram. This material allows the use of the Lankford parameters for the definition of the anisotropy.
In the model, Gosh’s hardening rule is used.
σY ( εp ) = k ( ε0 + εp ) – p
n
Forming limit diagram is defined in by TABLED1 as shown above. All fields are set for the coefficients of equations. See MD Nastran Quick Reference Guide for details. SPCD2,1,RIGID,MR2,3,0,100,1.0,,+ + TABLED1,100,,,,,,,,+ +,0.0,1000.,0.02,1000.,ENDT
Main Index
Chapter 13: Examples 587
Sheet Metal Forming
SPCD2 entry defines imposed nodal motion on a node, a set of nodes or nodes of a rigid body. The rigid punch is moving downward at 1000 mm/s from 0 to 0.02 seconds.
FORCE
9999
MR3
19.6E6
1.
FORCE entry defines a force on the grid point as well as rigids. Since the forces on the rigid body are not yet supported by Nastran input processor, TODYNA and ENDDYNA entries are used in conjunction with FORCE entry to bypass the IFP (Input File Processor) and directly access SOL 700.
BCTABLE
1 SLAVE
1 0 0.162
0. 0
3 0. 0 SS1WAY
0.162
0.
0
+ +
BCBODY entry defines a flexible or rigid contact body in 2D or 3D. Although SOL 700 only supports a flexible contact in BCTABLE, the rigid contact can be applied using the rigid material of contact bodies. In the example, all contacts have 0.162 static and kinetic friction coefficients. The surface to surface one way contact method is used on the all contact definition.
BCBODY .. $ BSURF ..
1 1 1
DEFORM THRU
1 1600
BCBODY entry defines a flexible or rigid contact body in 2D and 3D. BSURF entry defines a contact surface or body by element IDs. All elements with the specified IDs define a contact body.
$ GRID .. GRID $ CQUAD4 .. CQUAD4
1 4528 1 4468 1 63
75.
75.
0.0
8.3333337.006775.405 1 4527 2 4273 43 4274 42 4528
Main Index
588 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Sheet Metal Forming
Bulk data entries that define parameters. PARAM,DYDEFAUL,DYNA DYPARAM,LSDYNA,OUTPUT,NPOPT,1 .. $ ENDDATA End of input file. Results
Figure 1336
Comparison of Major Principal Strain Along Line OB (Numisheet and Dytran Results vs MD Nastran SOL 700)
Main Index
Chapter 13: Examples 589
Sheet Metal Forming
Figure 1337
Comparison of Minor Principal Strain Along Line OB (Numisheet and Dytran Results vs MD Nastran SOL 700)
Figure 1338
Forming Limit Diagram Along Line OB at 0.019 Seconds
Main Index
590 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Sheet Metal Forming
t = 0.000 s
t = 0.004 s
t = 0.008 s
t = 0.012 s
t = 0.016 s
t = 0.020 s
Figure 1339
Maximum Principal Strain Contour Plots at Mid Surface at Various Times
Note that the FLD diagram correctly predicts the failure of elements at t = 0.019 as shown in the stress fringe plots.
Main Index
Chapter 13: Examples 591
Sheet Metal Forming
PART 2. Square Cup Deep Drawing using Implicit Spring Back
SOL 700 Entries Included SOL 700 MATD036 SEQROUT SPRBCK Description The springback is refered to an event when the sheet metal starts vibrating after the punch is removed and might alter its final desired shape. Since it takes some time before the workpiece comes to a rest, the springback simulation is performed by the implicit solver to speed up the analysis. Using explicitimplicit switching available in SOL 700, the residual deformations after sheet metal forming are computed and used as a precondition for springback analysis. Because in this example, there was a failure at around 0.019 seconds in the sheet metal as shown in Part 1, the explicit simulation was terminated at 0.018 seconds. The initial condition including the final stresses and deformation and the element connectivity of the explicit run is transferred to the implicit run. The analysis scheme is described below. SOL 700 Explicit (Use SEQROUT Entry)
Generate jid.dytr.nastin
SOL 700 Implicit (Include jid.dytr.nastin) (Use SPRBCK Entry) Figure 1340 Model The model of explicit run is the same as Part 1. In the implicit run, only the sheet metal is used. Input File Explicit Input File BEGIN BULK $ TSTEPNL 1 Analysis Scheme
10
1.8E3
Main Index
592 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Sheet Metal Forming
As mentioned above, the end time of simulation is assigned to 0.018 seconds. SEQROUT 10 BCPROP 10
1
SEQROUT entry generates the jid.dytr.nastin file at the end of simulation. The nastin file includes the final deformations and stresses of the assigned part. The nastin file can be used for a subsequent explicit or implicit SOL 700 run. In the example, only the result for Part 10 which includes the sheet metal is written out to the nastin file.
Implicit Input File BEGIN BULK $ TSTEPNL 1
10
1.8E3
As mentioned above, the end time of simulation is assigned to 0.018 seconds. SEQROUT 10 BCPROP 10
1
SEQROUT entry generates the jid.dytr.nastin file at the end of simulation. The nastin file
includes the final deformations and stresses of the assigned part. The nastin file can be used for a subsequent explicit or implicit SOL 700 run. In the example, only the result for Part 10 which includes the sheet metal is written out to nastin file. Because all information of nodes and element connectivity is in jid.dytr.nastin file, Grid and CQUAD entries are removed in the implicit input. Only one point boundary condition at the center and SPRBCK entry are added in the input file. Since MATD190 is not available in the implicit analysis, MATD036 is used instead of MATD190. MATD036 and MATD190 are identical material models except that FLD is supported only in MATD190. MATD036 1 + 6.0 + 2.0 + + 2.7E4 7.1E7 .71 .58 0.33 .70 1.0 0.0 2.0 0.0 1.0 576.79E3.3593 .01658 0.0 0.0 0 + + + +
MATD036 is only different in the failure criteria using FLD. Others are the same as MATD190 in the
explicit simulations of Part 1 and 2. SPRBCK 1 + + 2 + 1 0.005 200 1 + + +
100
0.0 1.00E3 1.0E2 0.10 1
Main Index
Chapter 13: Examples 593
Sheet Metal Forming
SPRBCK activates the implicit spring back analysis. Nonlinear with BFGS updates solver type is used in
the example. See MD Nastran Quick Reference Guide for other fields. SPC1 1 123456 841
Only one point at the center of the sheet metal is fixed to prevent singular condition in the implicit simulation.
Main Index
594 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Sheet Metal Forming
Results Vertical (Zdirection) Displacement Contour Plot
:
Explicit Simulation
t = 0.000 s
t = 0.018 s (end of explicit run)
Implicit Simulation
Because the final results are applied as the initial condition for implicit simulation, the initial deformation of implicit simulation is set to zero.
Initial Condition of Implicit Run
Final Residual Stress of Implicit Run
Main Index
Chapter 13: Examples 595
Sheet Metal Forming
Figure 1341
Comparison of Vertical Displacements (zdirection) After Explicit and Springback Simulations Along Diagonal Line of Plate
Main Index
596 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Miscellaneous
Miscellaneous
Paper Feed
Description The objective of the paper feeding analysis is to predict the paper jamming and the capacity of the printer. In this example, angular velocity is applied on five rollers to feed the paper in the printer and 31are defined to simulate the paper feeding process. SOL 700 Entries Included SOL 700 TSTEPNL PARAM*,DYMINSTEP PARAM*,LDTSTEOFCTL BCTABLE CELASID CDAMPID TLOAD1 SPCD FORCE PELAS PDAMP CONM2 RBE2 Model
Main Index
Chapter 13: Examples 597
Miscellaneous
The model has a total of 6946 grid points, 6170 elements, and 20 rigid body connections as follows:
• 4972 Quads • 1188 Hexas • 10 Concentrated masses
The paper is modeled by 1830 quad elements. All shell elements are BelytschkoTsay formulation. The circumferential velocite is set to 1500 mm/s using different angular velocities of drives and pinches. Gravity is also taken into account. Total time of simulation is 0.4 seconds. Loading and Boundary Conditions The angular velocity of each drive and pinch is defined such that a 1500 mm/s circumferential velocity is created. The rotational velocities are applied sequentially at center node of the drive starting from drive 1 through drive 5 by defining Tables and SPCD. To close the gap between all the drives and the pinches, two opposite direction vertical forces are applied by using a combination of FORCE and Table entries. The magnitude of the load is predefined at each drive location.
entrance paper
drive_1
upper guide_3 pinch_4 upper guide_1 upper guide_4 pinch_5 upper guide_5 pinch_2 pinch_3 lower guide_5 drive_5 lower guide_4 drive_2 guide_2 drive_4 lower guide_3 drive_3
pinch_1 lower guide_1
Main Index
598 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Miscellaneous
RBE2 Damper Spring
Adding angular velocity to get 1500 mm/s circumferential velocity
Thirtyone contacts are defined between: 1. Self contact of a paper 2. Paper and drive_1 3. Drive_1 and pinch_1 (two ways) 4. Paper and drive_2 5. Drive_2 and pinch_2 (two ways) 6. Paper and drive_3 7. Drive_3 and pinch_3 (two ways) 8. Paper and drive_4 9. Drive_4 and pinch_4 (two ways) 10. Paper and drive_5 11. Drive_5 and pinch_5 (two ways) 12. Paper and entrance 13. Paper and lower guide_1 14. Paper and upper guide_1 15. Paper and guide_2 16. Paper and lower guide_3 17. Paper and upper guide_3 18. Paper and lower guide_4 19. Paper and upper guide_4 20. Paper and lower guide_5 21. Paper and upper guide_5
Main Index
Chapter 13: Examples 599
Miscellaneous
22. Paper and pinch_1 23. Paper and pinch_2 24. Paper and pinch_3 25. Paper and pinch_4 26. Paper and pinch_5 SOL 700,NLTRAN path=3 stop=1
SOL 700 initiates the explicit nonlinear transient analysis.
BEGIN BULK TSTEPNL 1
100
.004
1
ADAPT
2
10
TSTEPNL is a SOL 700 bulk data entry which describes the number of Time Steps (100) and Time Increment (0.004 s) of the simulation. End time is the product of the two entries. Notice here the Time Increment is only for the first step. The actual number of time increments and the exact value of the Time Steps are determined by SOL 700 during the analysis. The time step is a function of the smallest element dimension during the simulation.
BCTABLE
1 SLAVE
98 0
0. 0
31 0. 0
.3
0.
0
0.
MASTERS 98 ...
BCTABLE is a SOL 700 bulk data entry that is also used by SOL 600 and it is meant to define MasterSlave as well as self contact.
BCBODY BSURF 3007 ...
98 98
3D 3001
DEFORM 3002
98 3003
0 3004
3005
3006
BCBODY is a bulk data entry that is used by MD Nastran Implicit Nonlinear (SOL 600) and SOL 700
only, which defines a flexible or rigid contact body in 2D or 3D.
BSURF is a bulk data entry that is used by MD Nastran Implicit Nonlinear (SOL 600) and SOL 700 only, which defines a contact body or surface defined by Element IDs
Main Index
600 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Miscellaneous
Bulk data entry that defines grid points and elements: GRID ... $ CQUAD4 ... CHEXA ... CELAS1D ... CDAMP1D ... 1 1 325 406 21001 21002 1 2 405 18 19 12.5 1 361 21001 21001 20.05 2 362 2 2 0. 12 366 21002 21002 11 365 2 2 401 402
CELAS1D and CDAMP1D are for the scalar spring connection and scalar damper connection, respectively.
TLOAD1 LSEQ SPCD ... DLOAD ... FORCE ... GRAV
19 1 21 2 4 3
20 20 21001 1. 21001 0
VELO 21 6 1. 0 9800. 120. 19 9800. 0. 1.
1
22 0. 1. 0.
1. 0.
25
1.
TLOAD1 is a bulk data entry that defines a timedependent dynamic load or enforced motion for use in transient response analysis. SPCD is for the enforced motion value, velocity in this example, in dynamic analysis. Using FORCE card, a concentrated force at a grid point by specifying a vector is applied.
All materials are defined by isotropic elastic materials. MAT1 ... PELAS ... PDAMP ... CONM2 ... 6 3.+6 .3 8.47
18 19 51001
4.9 196. 21001 .001
PELAS and PDAMP are bulk data entries which describe properties of scalar spring and damper, respectively. CONM2 is for the definition of mass on a grid point.
Main Index
Chapter 13: Examples 601
Miscellaneous
Bulk data entries for element properties: RBE2 ... SPCADD ... SPC1 ... TABLED1 55003 21001 123456 1001 1002 1003 1004 1005
2 1
1 123456
3 21002
4
5
6
7
8
1 .12 4.4
0. 1.
0. .1201
0. 1.
1.4 .1205
.5 0.
2.4 1.2
.8 0.
RBE2 is a bulk data entry which defines a rigid body with independent degrees of freedom that are
specified at a single grid point and with dependent degrees of freedom that are specified at an arbitrary number of grid points. Several tables and coordinates are defined in the input file.
Main Index
602 MD Natran R3 Explicit Nonlinear (SOL 700) User’s Guide
Miscellaneous
Results
t = 0.00 second
t = 0.20 second
t = 0.40 second
Main Index