Welcome to the LS-DYNA Product Space

<p>Dear community, <br><br>Is it possible to simulate a pre-strained material directly without any pre-simulation? We have conducted uniaxial prestraining on a material and would now like to run another simulation.<br><br>I have the major and minor strains from our previous operation. <br><br>Best regards <br><br> </p>

<p><p><p><p><p><p><p>I am wondering what controls how many states are packed into each d3plot file in LS-DYNA, and whether model complexity can affect this.</p><p>I'm running two simulations of similar brain models, one of which is standard and the other with embedded beam elements representing axons. Both have identical *DATABASE_BINARY_D3PLOT settings (dt = 0.3125 ms), identical termination times (50 ms), and identical *DATABASE_EXTENT_BINARY settings including BEAMIP=4 and IEVERP=0. Based on the d3hsp logs, both simulations wrote 161 states during the run and both completed normally.</p><p>However, on disk the simpler model has 41 d3plot files while the complex axon model has 161. Since the d3hsp confirms the same number of write events in both, it seems like LS-DYNA is packing more states per file in the simpler model (roughly 4 states per file vs 1). I haven't used any load curves to prescribe output intervals and the packing behavior doesn't seem to be controlled by anything obvious in the keywords I've checked. Ultimately, I would like to reduce the number of d3plot files in the complex model to match the simpler one in a controlled manner. Is there some keyword or setting that controls states-per-file packing that I might be missing?</p></p></p></p></p></p></p>

<div class="xdj266r x14z9mp xat24cr x1lziwak x1vvkbs"><div dir="auto">Dear colleagues,</div><div dir="auto">Below is what I read:</div><p class="MsoNormal" style="margin: 6.0pt 0in 6.0pt 0in;"><span style="color: #ee0000; mso-ansi-language: EN-US;">https://feassistant.com/forums/ls-dyna/how-to-give-displacement-to-constrained-nodal-rigid-body-body/</span></p><br><div dir="auto">I'm not sure if this is possible, but</div><div dir="auto">first you need to make sure that your reference node is the first node in the node set referenced by the constrained nodal rigid body (CNRB).<br><br></div><div dir="auto">The first node is the source point and represents the center of the CNRB.</div><div dir="auto"><br>Then you can try to apply *Boundary_Prescribed_Motion_Node to that reference node.</div></div><div class="x14z9mp xat24cr x1lziwak x1vvkbs xtlvy1s"><div dir="auto"><br>If Dyna will not accept this, then you will need a different approach.</div><div dir="auto"><br>You can create a simple element (1D, 2D, 3D etc.) which uses that reference node.</div><div dir="auto">Then make this part *Mat_rigid.</div><div dir="auto"><br>Then you will use *Constrained_extra_nodes_set and refer to the rigid part as the PID and the other nodes will need to be in a node set and be referred to in the NSID.</div><div dir="auto">You will then be able to apply motion to the rigid part using *Boundary_prescribed_motion_rigid.<br><br>==================================================================<br><div class="xdj266r x14z9mp xat24cr x1lziwak x1vvkbs"><div dir="auto">At the same time, in the given model:</div></div><div class="x14z9mp xat24cr x1lziwak x1vvkbs xtlvy1s"><div dir="auto">https://www.dynaexamples.com/.../Salzbur.../example_linear_3</div><div dir="auto">• 3.4_Connectors_CNRB_Interpolation.key.zip<br><br></div></div><div class="x14z9mp xat24cr x1lziwak x1vvkbs xtlvy1s"><div dir="auto">Two parameters are used:</div></div><div class="x14z9mp xat24cr x1lziwak x1vvkbs xtlvy1s"><div dir="auto"><br>NSID - if I understand correctly, this is a set of rigid body points + a "reference" point, all of which are fully described by Connectors CNRB.</div></div><div class="x14z9mp xat24cr x1lziwak x1vvkbs xtlvy1s"><div dir="auto"><br>At the same time, there is a parameter PID:=Part ID of the nodal rigid body.</div><div dir="auto">It is set to 100.</div><div dir="auto"><br>If I understand correctly, the number of the "new" part defined by the operator:</div><div dir="auto">Constrained Nodal Rigid Body (CNRB)</div><div dir="auto"><br>But there is no such part in the list of "parts."</div><div dir="auto"><br>Am I correct in understanding that this part number is an arbitrary number that the user must specify?!</div></div><div class="x14z9mp xat24cr x1lziwak x1vvkbs xtlvy1s"><div dir="auto"><br>And one more thing: <br>AFTER solving this problem, a new body, "10000000 Beam," appears in the list of "parts." <br>Before solving this problem, the "10000000 Beam" body wasn't listed as "parts."<br><br></div></div><div class="x14z9mp xat24cr x1lziwak x1vvkbs xtlvy1s"><div dir="auto">Can anyone comment on my questions?</div></div><br><br></div></div>

<p>I am a Third year engineering student working on an aluminium honeycomb Impact Attenuator for my FS/FSAE team</p><p> </p><p>When I first started working on it in Ls Prepost it didn't give any issue beyond a long loading time for it to appear in software up until I tried applying my mesh on it. Then it started to not respond and force me to close prepost.</p><p> </p><p>Then after that every subsequent time I tried to load the STEP file my Impact attenuator is in, prepost doesn't respond and I can't get beyond that point?</p><p> </p><p>Is there anyway to fix this?</p>

<p> </p><p>Hi!</p><p>I'm trying to set up an SSI analysis using *LOAD_SEISMIC_DECONV and *INTERFACE_SSI</p><p> </p><p>I have been trying to get it to work but I get the following error (this is the same error I get although node numbers might differ from mine as I got them from another example online). I don't understand why I'm getting this error. I use *INTERFACE_SSI_STATIC_ID in the static case and the same keyword without the STATIC in the transient case. I use the same sets for each. How can there be a node mismatch?</p><p> *** Error 30271 (INI+271)</p><p>     # of nodes 4418 in SSI contact interface id 1</p><p>     do not match stored # of nodes: 890217136</p>

<p>When running an impact analysis of termination time 0.25s, after the failure of a single shell element at 0.191s the job immediately terminates. The lsrun.out file records the following error message: </p><p class="MsoNormal">forrtl: severe (157): Program Exception - access violation</p><p class="MsoNormal">The version of LS-DYNA I am using is 2024, and the contact algorithm I am using for impact is *CONTACT_AUTOMATIC_NODES_TO_SURFACE. </p><p class="MsoNormal">Is it the way I have set up my contact or is there an internal LS-DYNA bug/issue with the 2024 version?</p>

<p>Hello gentlemen，</p><p>I want to write some program to extract some data from the d3plot and d3htd file, but I can't found any information about the script form the documentaion, also nothing from the ansys website. </p><p>Are there any help documention I can found to reference? </p><p>Thank you</p>

<p>Hello gentlemen，</p><p>I want to do a shock vibration with lsdyna explicit, shock condition: 18G, 6ms, half sine wave,Y direction</p><p>Question 1: I use the keyword *BOUNDARY_PRESCRIBED_MOTION_RIGID to apply the acceleration, but I found I have to release the freedom of Y on the base fix points so that the acceleration can be normal (18G). Unlike the random vibration and response spectrum analysis, I can fully fixed the base and apply the acceleration at the same time. Can you help me to explain it?</p><p>Question 2: When I was apply the acceleration, Do I need to apply a initial velocity on the oppsite direction of the acceleration? This question really confused me. I did not find a similar example from the dynaexample.</p><p>If any one have a K file model with similar condition, It will be really thanksful to share with me. Thank you!</p><div> </div><p> </p>

<p data-path-to-node="4">Hi everyone,</p><p data-path-to-node="5">I’m running a 2D plane-strain site response analysis of a soil basin. To handle the side boundaries, I meshed 1D columns (local periodic boundaries) on the far left and right edges, and connected them to the main basin using dashpots.</p><p data-path-to-node="5">During gravity initialization (Dynamic Relaxation), I use a <code data-path-to-node="7" data-index-in-node="73">*BOUNDARY_SPC_SET_BIRTH_DEATH to lock the basin edges horizontally so a gap doesn't form. DR converges perfectly, and the mesh sits completely flush.</p><p data-path-to-node="5">I am also applying <span class="math-inline" data-math="\sigma_{xx}" data-index-in-node="21">sigma_xx </span>and <span class="math-inline" data-math="\tau_{zx}" data-index-in-node="37">tau_zx  form elout to the inner sides of main model which are on the boundary between far side columns and main domain. <br><br>I am still not capturing uniform free field response on the surface. </span></p><p data-path-to-node="5"><span class="math-inline" data-math="\tau_{zx}" data-index-in-node="37">How can I mimic the free field case just like a 1D site response analysis? (I cant want to use *CONSTRAINED_NODE_SET as I will non-flat non-uniform geometry affects in future) </span></p><p data-path-to-node="5"><span class="math-inline" data-math="\tau_{zx}" data-index-in-node="37">Best,</span></p><p data-path-to-node="5"><span class="math-inline" data-math="\tau_{zx}" data-index-in-node="37">Muhsin</span></p><p data-path-to-node="5"> </p>

<p>Hi everyone,</p><p>I’m a student currently learning about LS-DYNA.<br>I would like to use simulation to analyze soft and flexible materials, I'm using thin structures so I would like to use shell (2D) elements.<br>Does anyone have suggestions for suitable material cards for this type of material?</p><p>Also, I currently only have tensile test data for the material and do not have compression test data.<br>Can tensile stress–strain data still be used as the Curve ID input even if the simulation involves compression behavior?</p><p>Thanks in advance!!!</p>