Bookmark this page for easy reference to all the resources you need to learn, find examples or solutions, interact with others, or receive support for LS-DYNA.
<p><span data-olk-copy-source="MessageBody">Hi,<br><br>I have Ansys LS-DYNA 2025 R2 installed on a Windows workstation. </span></p><div class="x_elementToProof" data-olk-copy-source="MessageBody">Currently, I am developing a custom phase-field fracture model in LS-DYNA, instead of using any inbuilt material models. Could anyone please help me with how to get (or whom to reach out at Ansys to get) the compatible/matching LS-DYNA 2025 R2 / v252 Windows x64 user-defined material and user-defined element object/static-linking package, including:</div><ul data-spread="false">dyn21.F / dyn21.f for user-defined materialsdyn21b.F / dyn21b.f for user-defined elementsinclude filesobject/static librariesMakefile or Windows build scriptsexample UMAT and user-element source filesexample keyword decks for user-defined elementsrecommended compiler and linker instructions<div class="x_elementToProof">Also, my current installation is using the SMP single-precision executable, but the phase-field fracture problems may become computationally heavy. Please also advise whether I should request or use:</div><ul data-spread="false">SMP single precision, SMP double precision, or MPP double precisionWindows MPP support, if availableMS-MPI or another MPI packageIntel oneAPI Fortran / Intel Fortran compiler versionMicrosoft Visual Studio / MSVC versionIntel MKL / BLAS / LAPACK support, if needed for solving the equations<div class="x_elementToProof">Please let me know which exact package is compatible with my LS-DYNA 2025 R2 / v252 installation and what additional software I need to install before compiling a custom executable.</div><div class="x_elementToProof">Also, if Ansys has any tutorials, documentation, examples, training material, or technical notes on using the LS-DYNA user-defined material and user-defined element packages, especially for compiling custom executables on Windows, I would appreciate it if you could share them.<br><br></div><div class="x_elementToProof">Thank you.</div><div class="x_elementToProof">Sincerely,<br><br>Umar<br>PhD student @GeorgiaTech</div>
<p>Hello,</p><p>Acoording to different Ansys recommendations sources, in case of using Hexa and Tetras for a solid mesh, it is recommended to use ESORT = 1. LSDYNA Recommendations for structural impact</p><p>This tranforms tetras to type 10 formulation, which adds a lot of stiffness according to Microsoft PowerPoint - solid_elements.pptx. </p><p>Why not ESORT = 11 to change to TET13, which is more accurate with a not that different rsimulation time? <br><br>Is there some reasons that I might not be seeing?<br><br>Best regards,</p><p>Roberto</p>
<p>Hello, I've used the S-ALE method to simulate a hull impact on the water.</p><p>I've definied the hull with SECTION_SHELL and the water/air with ALE_STRUCTURED_FSI.</p><p>The analysis works but I can't get the "correct" pressure on the shell. I've tryed with different approaches (DATABASE_FSI, DATABASE_FSI_SENSOR...) but I can't get the results: with the sensors there's a problem about the offset position and with database_fsi I get just the average pressure on the shell but I need the one on just a few shell elements.</p><p></p><p>Is there any keyword or method to measure the local pressure on a few shell elements?</p>
<p class="isSelectedEnd">Hello everyone,</p><p class="isSelectedEnd">I am using LS-DYNA ICFD for hemodynamic simulations.</p><p class="isSelectedEnd">I would like to implement a purely resistive outlet boundary condition (i.e., only R2, with R1 = C1 = L1 = 0). However, the solver diverges when I set C1 = 0 (division by 0 not allowed)</p><p class="isSelectedEnd">Is there a way to define a purely resistive model at the outlets using the available WTYPE circuit options in the card?</p><p class="isSelectedEnd">If not, could you clarify how WTYPE = 6 works, particularly regarding the use of the "icfd_usercircuit" in the file dyn21icfd.f?</p><p class="isSelectedEnd">Thank you in advance for your help.</p><p>Best regards,<br>Gred</p>
<div>Hi everyone,</div><div> </div><div>I am encountering a Signal 11 (Segmentation Violation) during MPP initialization that occurs only with my custom-built LS-DYNA R16.2 executable, while the standard Ansys-distributed binary runs the identical keyword file to normal termination without any errors.</div><div> </div><div>Environment & Executables</div><div>- Custom build: mppdyna.exe compiled from the R16.2 source makefile (Intel Fortran 19.0 + Intel MPI 2019.6.166, double precision I8R8, WINX64). I did not modify any source files or build flags — I simply ran the default makefile to verify that the custom library would execute without errors.</div><div>- Standard binary: lsdyna_mpp_sp_msmpi.exe (Ansys-distributed) — same keyword file, same machine, same MPI run command completes successfully with normal termination.</div><div>- Version: LS-DYNA R16.2-151-g276cb33db2</div><div> </div><div>Problem Summary</div><div>When I run the custom mppdyna.exe with thermal properties included, the solver crashes immediately after initialization:</div><div> </div><div></div><div>initialization completed</div><div>>>>>> Process 0 <<<<<</div><div>>>>>> Signal 11 : Segmentation Violation <<<<<</div><div></div><div> </div><div>The crash happens at Problem cycle = 0, right after MPP decomposition and initialization finish. No d3plot or binout data is written beyond the initial state.</div><div> </div><div>Critical Observation: Thermal Dependency</div><div>- With thermal cards active: The custom MPP executable segfaults every time (tested with -n 1 and -n 2).</div><div>- Without thermal cards: The custom MPP executable runs perfectly fine to normal termination.</div><div>- With thermal cards + standard Ansys binary: Runs perfectly fine to normal termination.</div><div> </div><div>This strongly suggests the issue is tied to how the custom-built executable handles thermal initialization in an MPP + S-ALE environment, rather than a keyword syntax error.</div><div> </div><div>Keyword Cards Used in the Model</div><div>The simulation is a 2D axisymmetric S-ALE hypervelocity impact problem using the following cards:</div><div> </div><div>- *KEYWORD, *TITLE</div><div>- *MAT_ELASTIC_PLASTIC_HYDRO_TITLE (HSLA 100 Steel, Tungsten Carbide)</div><div>- *MAT_ALE_VACUUM_TITLE</div><div>- *MAT_THERMAL_ISOTROPIC_PHASE_CHANGE_TITLE (Steel)</div><div>- *MAT_THERMAL_ISOTROPIC_TITLE (Tungsten Carbide, Vacuum)</div><div>- *EOS_GRUNEISEN_TITLE</div><div>- *NODE</div><div>- *ALE_STRUCTURED_MESH, *ALE_STRUCTURED_MESH_CONTROL_POINTS</div><div>- *ALE_MULTI-MATERIAL_GROUP</div><div>- *ALE_STRUCTURED_MESH_VOLUME_FILLING (ALL, BOXCOR, ELLIPSOID)</div><div>- *PART, *SECTION_ALE2D_TITLE</div><div>- *INITIAL_TEMPERATURE_SET</div><div>- *DEFINE_BOX_TITLE, *DEFINE_VECTOR_TITLE</div><div>- *BOUNDARY_SALE_MESH_FACE (SYM, NONREFL)</div><div>- *SET_PART_LIST_TITLE</div><div>- *CONTROL_ALE, *CONTROL_BULK_VISCOSITY, *CONTROL_ENERGY, *CONTROL_HOURGLASS</div><div>- *CONTROL_MPP_DECOMPOSITION_DISTRIBUTE_ALE_ELEMENTS</div><div>- *CONTROL_SOLUTION, *CONTROL_TERMINATION</div><div>- *CONTROL_THERMAL_NONLINEAR, *CONTROL_THERMAL_SOLVER, *CONTROL_THERMAL_TIMESTEP</div><div>- *CONTROL_TIMESTEP, *CONTROL_UNITS</div><div>- *DATABASE_ family (D3PLOT, D3THDT, GLSTAT, MATSUM, NODOUT, TPRINT, etc.)</div><div>- *DATABASE_EXTENT_BINARY</div><div>- *END</div><div> </div><div>Run Command (Custom — Fails)</div><div>bash</div><div>mpiexec -n 2 "...\mppdyna.exe" i="sale_shot10_umat44v_T09-fine.k" memory=200m</div><div></div><div> </div><div>Run Command (Standard — Succeeds)</div><div>bash</div><div>mpiexec -n 2 "...\lsdyna_mpp_sp_msmpi.exe" i="sale_shot10_umat44v_T09-fine.k" memory=200m</div><div></div><div> </div><div>Question</div><div>Since the standard Ansys MPP binary handles the identical thermal + S-ALE setup without issue, but the default, unmodified custom build crashes during thermal initialization, what could be causing this? Is there a known compiler/MPI flag, preprocessor directive, or library linkage issue in the R16.2 makefile that affects thermal solver initialization in MPP mode? Any guidance on resolving this would be greatly appreciated.</div><div> </div><div>Thanks,</div><div>Amrith</div>
<p>I am trying to extract contact force using contact force tracker (added by right clicking the 'Solution Information') scoped to a rigid body (tool) interacting with a flexible body (workpiece), when trying to simulate an orthogonal cutting process in ANSYS 2021R1 LS-DYNA. I only used 'body interaction' as a connection between these two bodies, with a frictional coffecient entered as 0.3, with no other contacts built betweeen these bodies (like bonded, etc). However, when I try to evaluate the results after solution has finished, the contact force tracker doesn't work, and generates me this error as I am sharing in the picture. I don't know how to add the body contact tracker, because I don't see any option to add it from anywhere (like it doesn't show up in the 'Solution Information'). Can you tell me how should I address this issue?<br><br></p>
<p class="isSelectedEnd">Hi there, </p><p class="isSelectedEnd">I have an explicit simulation that is running very slowly. One of the parts in the simulation is rigid, and according to the messag file, this part has the lowest timestep. Based on the LS-DYNA documentation, I decreased the young’s modulus of the rigid part (by assigning different material properties) in order to increase the timestep and reduce the overall runtime. I then ran the simulation again; however, it is now running even slower. Could anyone help me understand why reducing the young’s modulus of the rigid part would make the simulation slower instead of faster (although the document says it should make it faster)?</p><p class="isSelectedEnd">I also have another question. The other components of the simulation consist of uniformly distributed SPH particles. I checked the d3hsp file for the "smallest" timestep entries, and indicate a few SPH nodes. My question is: if the SPH particles are distributed uniformly, why are only a few of them controlling the smallest timestep? What could be causing this? How can I correct the issue??</p><p>I would appreciate any help with these issues.</p>
<p></p><p>In post-processing I need to plot the pressure on a few shell elements, I used the SALE method to simulate a shell impact on the water.</p><p>Using Post - History - Element/Integration points to plot the results for the same points I get different plots.</p><p>Also defining a FSI Sensor (same elements) the plot is different.</p><p>I don't know why and which is the correct (the sensor results are more attendible than the other ones).</p>
<p style="text-align: left;">I have to measure the mass of a shell set, using the measure mass control I get different results (20 times smaller):</p><p></p><p>In both measures I select the same body but using Part or Elements</p><p>Which measure is the correct one and why?</p><p>Thanks</p>
<p><span data-olk-copy-source="MessageBody">Hi,<br><br>I have Ansys LS-DYNA 2025 R2 installed on a Windows workstation. </span></p><div class="x_elementToProof" data-olk-copy-source="MessageBody">Currently, I am developing a custom phase-field fracture model in LS-DYNA, instead of using any inbuilt material models. Could anyone please help me with how to get (or whom to reach out at Ansys to get) the compatible/matching LS-DYNA 2025 R2 / v252 Windows x64 user-defined material and user-defined element object/static-linking package, including:</div><ul data-spread="false">dyn21.F / dyn21.f for user-defined materialsdyn21b.F / dyn21b.f for user-defined elementsinclude filesobject/static librariesMakefile or Windows build scriptsexample UMAT and user-element source filesexample keyword decks for user-defined elementsrecommended compiler and linker instructions<div class="x_elementToProof">Also, my current installation is using the SMP single-precision executable, but the phase-field fracture problems may become computationally heavy. Please also advise whether I should request or use:</div><ul data-spread="false">SMP single precision, SMP double precision, or MPP double precisionWindows MPP support, if availableMS-MPI or another MPI packageIntel oneAPI Fortran / Intel Fortran compiler versionMicrosoft Visual Studio / MSVC versionIntel MKL / BLAS / LAPACK support, if needed for solving the equations<div class="x_elementToProof">Please let me know which exact package is compatible with my LS-DYNA 2025 R2 / v252 installation and what additional software I need to install before compiling a custom executable.</div><div class="x_elementToProof">Also, if Ansys has any tutorials, documentation, examples, training material, or technical notes on using the LS-DYNA user-defined material and user-defined element packages, especially for compiling custom executables on Windows, I would appreciate it if you could share them.<br><br></div><div class="x_elementToProof">Thank you.</div><div class="x_elementToProof">Sincerely,<br><br>Umar<br>PhD student @GeorgiaTech</div>
<p>Hello,</p><p>Acoording to different Ansys recommendations sources, in case of using Hexa and Tetras for a solid mesh, it is recommended to use ESORT = 1. LSDYNA Recommendations for structural impact</p><p>This tranforms tetras to type 10 formulation, which adds a lot of stiffness according to Microsoft PowerPoint - solid_elements.pptx. </p><p>Why not ESORT = 11 to change to TET13, which is more accurate with a not that different rsimulation time? <br><br>Is there some reasons that I might not be seeing?<br><br>Best regards,</p><p>Roberto</p>
<p>Hello, I've used the S-ALE method to simulate a hull impact on the water.</p><p>I've definied the hull with SECTION_SHELL and the water/air with ALE_STRUCTURED_FSI.</p><p>The analysis works but I can't get the "correct" pressure on the shell. I've tryed with different approaches (DATABASE_FSI, DATABASE_FSI_SENSOR...) but I can't get the results: with the sensors there's a problem about the offset position and with database_fsi I get just the average pressure on the shell but I need the one on just a few shell elements.</p><p></p><p>Is there any keyword or method to measure the local pressure on a few shell elements?</p>
<p class="isSelectedEnd">Hello everyone,</p><p class="isSelectedEnd">I am using LS-DYNA ICFD for hemodynamic simulations.</p><p class="isSelectedEnd">I would like to implement a purely resistive outlet boundary condition (i.e., only R2, with R1 = C1 = L1 = 0). However, the solver diverges when I set C1 = 0 (division by 0 not allowed)</p><p class="isSelectedEnd">Is there a way to define a purely resistive model at the outlets using the available WTYPE circuit options in the card?</p><p class="isSelectedEnd">If not, could you clarify how WTYPE = 6 works, particularly regarding the use of the "icfd_usercircuit" in the file dyn21icfd.f?</p><p class="isSelectedEnd">Thank you in advance for your help.</p><p>Best regards,<br>Gred</p>
<div>Hi everyone,</div><div> </div><div>I am encountering a Signal 11 (Segmentation Violation) during MPP initialization that occurs only with my custom-built LS-DYNA R16.2 executable, while the standard Ansys-distributed binary runs the identical keyword file to normal termination without any errors.</div><div> </div><div>Environment & Executables</div><div>- Custom build: mppdyna.exe compiled from the R16.2 source makefile (Intel Fortran 19.0 + Intel MPI 2019.6.166, double precision I8R8, WINX64). I did not modify any source files or build flags — I simply ran the default makefile to verify that the custom library would execute without errors.</div><div>- Standard binary: lsdyna_mpp_sp_msmpi.exe (Ansys-distributed) — same keyword file, same machine, same MPI run command completes successfully with normal termination.</div><div>- Version: LS-DYNA R16.2-151-g276cb33db2</div><div> </div><div>Problem Summary</div><div>When I run the custom mppdyna.exe with thermal properties included, the solver crashes immediately after initialization:</div><div> </div><div></div><div>initialization completed</div><div>>>>>> Process 0 <<<<<</div><div>>>>>> Signal 11 : Segmentation Violation <<<<<</div><div></div><div> </div><div>The crash happens at Problem cycle = 0, right after MPP decomposition and initialization finish. No d3plot or binout data is written beyond the initial state.</div><div> </div><div>Critical Observation: Thermal Dependency</div><div>- With thermal cards active: The custom MPP executable segfaults every time (tested with -n 1 and -n 2).</div><div>- Without thermal cards: The custom MPP executable runs perfectly fine to normal termination.</div><div>- With thermal cards + standard Ansys binary: Runs perfectly fine to normal termination.</div><div> </div><div>This strongly suggests the issue is tied to how the custom-built executable handles thermal initialization in an MPP + S-ALE environment, rather than a keyword syntax error.</div><div> </div><div>Keyword Cards Used in the Model</div><div>The simulation is a 2D axisymmetric S-ALE hypervelocity impact problem using the following cards:</div><div> </div><div>- *KEYWORD, *TITLE</div><div>- *MAT_ELASTIC_PLASTIC_HYDRO_TITLE (HSLA 100 Steel, Tungsten Carbide)</div><div>- *MAT_ALE_VACUUM_TITLE</div><div>- *MAT_THERMAL_ISOTROPIC_PHASE_CHANGE_TITLE (Steel)</div><div>- *MAT_THERMAL_ISOTROPIC_TITLE (Tungsten Carbide, Vacuum)</div><div>- *EOS_GRUNEISEN_TITLE</div><div>- *NODE</div><div>- *ALE_STRUCTURED_MESH, *ALE_STRUCTURED_MESH_CONTROL_POINTS</div><div>- *ALE_MULTI-MATERIAL_GROUP</div><div>- *ALE_STRUCTURED_MESH_VOLUME_FILLING (ALL, BOXCOR, ELLIPSOID)</div><div>- *PART, *SECTION_ALE2D_TITLE</div><div>- *INITIAL_TEMPERATURE_SET</div><div>- *DEFINE_BOX_TITLE, *DEFINE_VECTOR_TITLE</div><div>- *BOUNDARY_SALE_MESH_FACE (SYM, NONREFL)</div><div>- *SET_PART_LIST_TITLE</div><div>- *CONTROL_ALE, *CONTROL_BULK_VISCOSITY, *CONTROL_ENERGY, *CONTROL_HOURGLASS</div><div>- *CONTROL_MPP_DECOMPOSITION_DISTRIBUTE_ALE_ELEMENTS</div><div>- *CONTROL_SOLUTION, *CONTROL_TERMINATION</div><div>- *CONTROL_THERMAL_NONLINEAR, *CONTROL_THERMAL_SOLVER, *CONTROL_THERMAL_TIMESTEP</div><div>- *CONTROL_TIMESTEP, *CONTROL_UNITS</div><div>- *DATABASE_ family (D3PLOT, D3THDT, GLSTAT, MATSUM, NODOUT, TPRINT, etc.)</div><div>- *DATABASE_EXTENT_BINARY</div><div>- *END</div><div> </div><div>Run Command (Custom — Fails)</div><div>bash</div><div>mpiexec -n 2 "...\mppdyna.exe" i="sale_shot10_umat44v_T09-fine.k" memory=200m</div><div></div><div> </div><div>Run Command (Standard — Succeeds)</div><div>bash</div><div>mpiexec -n 2 "...\lsdyna_mpp_sp_msmpi.exe" i="sale_shot10_umat44v_T09-fine.k" memory=200m</div><div></div><div> </div><div>Question</div><div>Since the standard Ansys MPP binary handles the identical thermal + S-ALE setup without issue, but the default, unmodified custom build crashes during thermal initialization, what could be causing this? Is there a known compiler/MPI flag, preprocessor directive, or library linkage issue in the R16.2 makefile that affects thermal solver initialization in MPP mode? Any guidance on resolving this would be greatly appreciated.</div><div> </div><div>Thanks,</div><div>Amrith</div>
<p>I am trying to extract contact force using contact force tracker (added by right clicking the 'Solution Information') scoped to a rigid body (tool) interacting with a flexible body (workpiece), when trying to simulate an orthogonal cutting process in ANSYS 2021R1 LS-DYNA. I only used 'body interaction' as a connection between these two bodies, with a frictional coffecient entered as 0.3, with no other contacts built betweeen these bodies (like bonded, etc). However, when I try to evaluate the results after solution has finished, the contact force tracker doesn't work, and generates me this error as I am sharing in the picture. I don't know how to add the body contact tracker, because I don't see any option to add it from anywhere (like it doesn't show up in the 'Solution Information'). Can you tell me how should I address this issue?<br><br></p>
<p class="isSelectedEnd">Hi there, </p><p class="isSelectedEnd">I have an explicit simulation that is running very slowly. One of the parts in the simulation is rigid, and according to the messag file, this part has the lowest timestep. Based on the LS-DYNA documentation, I decreased the young’s modulus of the rigid part (by assigning different material properties) in order to increase the timestep and reduce the overall runtime. I then ran the simulation again; however, it is now running even slower. Could anyone help me understand why reducing the young’s modulus of the rigid part would make the simulation slower instead of faster (although the document says it should make it faster)?</p><p class="isSelectedEnd">I also have another question. The other components of the simulation consist of uniformly distributed SPH particles. I checked the d3hsp file for the "smallest" timestep entries, and indicate a few SPH nodes. My question is: if the SPH particles are distributed uniformly, why are only a few of them controlling the smallest timestep? What could be causing this? How can I correct the issue??</p><p>I would appreciate any help with these issues.</p>
<p></p><p>In post-processing I need to plot the pressure on a few shell elements, I used the SALE method to simulate a shell impact on the water.</p><p>Using Post - History - Element/Integration points to plot the results for the same points I get different plots.</p><p>Also defining a FSI Sensor (same elements) the plot is different.</p><p>I don't know why and which is the correct (the sensor results are more attendible than the other ones).</p>
<p style="text-align: left;">I have to measure the mass of a shell set, using the measure mass control I get different results (20 times smaller):</p><p></p><p>In both measures I select the same body but using Part or Elements</p><p>Which measure is the correct one and why?</p><p>Thanks</p>
Community