Welcome to the LS-DYNA Product Space

<p>Hello all, I am having a very peculiar problem attempting to do a simple ball impact on plate simulation on Workbench LS Dyna. I create the model in Spaceclaim, mesh it, select an initial velocity for the ball of 6000m/s, generate a fixed support for the plate and try to do the simulation. However the result is very strange- the ball moves towards the target plate at a very small velocity (maybe 5-10m/s), usually gently turning the other way before it comes near it. This happens if the velocity is 6, 60, 600, 6000, or 600000 m/s. I have tried several times changing different parameters, but still get the exact same result. The thing is this exact same simulation runs perfectly in Workbench Explicit Dynamics with virtually no changes. Below I have posted a video of the problem happening and the .k file of my LS Dyna setup. Any help would be appreciated, cheers.</p><p>https://youtu.be/3yIUbvH8_00</p><p>https://drive.google.com/file/d/1D-xqba6vQ8mapPSnN7IKga4mhqrZSGok/view?usp=sharing</p>

<p>Hello,</p><p>I am applying a UMAT to ELEFORM3 elements (Truss elements) in different PARTS to run my simulations. Although the application of the UMAT works fine for one of the PARTS, I am getting a message saying that the "default UMAT 42 is being used", which I believe that means that my material model is not being correctly applied to the elements that I desire.</p><p>It is important to notice that the UMAT work fine in my local computer (Windows), but I get this error when I run my UMAT in a cluster Linux machine.</p><p>Also, I am dynamically linking my shared objects by changing the PATH variable.</p><p>Please, could someone explain why is this happening?</p><p>Thank you very much.</p>

<p>我想知道lspp有几种爆炸模拟方法 具体是什么 有什么区别 最常用的是那个</p>

<p>Hi, I am trying to model strain rate dependent properties of fiber reniorced composite.</p><p>I am using Mat 54, LCID to input the strain rate dependent strength in each direction.</p><p> Wonder how the units work. </p><p>From what I've found (the link below), the strain rate seems to be a function of logarism.</p><p>      https://dyna-composites.netlify.app/mat_054</p><p>Is it different from Mat 162: ln (strain rate/reference strain rate) ?</p><p>Is "ln(strain rate)" correct for the x value of the load curve?</p><p> </p><p>I am currently modeling composites in solid elements with delamination modeled with contact_tiebreak. </p><p>I don't think my properties with strain rate dependent terms are modeled well, so I wonder if any one has used this option in Mat 54.</p><p>Thank you</p><p> </p><p> </p><p></p>

<p>Hello, I am modelling a soil column for a seismic analysis. First I run a gravity analysis for 10 sec incl. DR phase, which works fine. After the DR phase there are still some slight oscillations in the first 2sec, after that the model is perfectly still (in equilibrium). When I do a full restart however with the last d3dump, the model blows up. I do not change loading or boundary conditions in the restart deck, the only thing I change is I add the *STRESS_INITITIALIZATION card to tell Dyna its a full restart, extend the termination time to 20 sec, and in the *CONTROL_DR card I set idrflg to -999, since I should not need DR anymore as the equilibrated state should be imported (other than that the restart input is just a copy of the input file of the first run). However, very early on in the analysis the model distorts massively when viewing the d3plot output. The messag file of the restart run does complete, but says that the timestep got very large. The elout file only prints NaN for stress and strains... Hoping somebody can help me find out what is going wrong. (I know I could simply use the dynain file instead of restart, but then I always get these oscillations in the first ~2 sec. They are small, but anyway. I would like to know why the restart does not work...) Thanks</p>

<p>I am trying to implement the ALE method for an analysis of a section of a hull impacting the water.</p><p>I managed to create the model and run the analysis.</p><p>I don't understand why there is such a significant loss of energy during impact. In similar models using the SPH method, I had a maximum loss of 10% of the initial energy, but in this model it is almost halved.</p><p>I have already tried using both SOLID and SHELL elements with different materials, ELASTIC, RIGID, COMPOSITE... but the result is always similar.</p><p>Currently, in the model, I have defined a water tank with EOS_GRUNEISEN and the upper air with MAT_VOID.</p><p>Is it normal for such a high loss to occur, or is it necessary to implement other KEYWORDS?</p><p>Thank you in advance.</p><p> </p><p> </p><p> </p><p> </p><p></p>

<p>Hi all,</p><p>I am working on simulating an injection stretch blow molding (ISBM) process. The contact rod and the mold are modelled as rigid bodies. Contact rod is provided with the prescribed rigid body motion and assignes a velocity curve. Contact automatic surface to surface is used for all contacts. I am using MAT77-H to model the visco-hyperelastic behviour of Polyethylene Terephthalate (PET) preform. During the stretching part of the process i have large uniaxial strains. The geometry is modelled using shell elements (ELFORM 2 for the preform). I am using ihq = 1 for hourglass control. I tried several values for the hourglass coefficient from 0.1 upto 0.35 and i can observe least hourglassing at QH = 0.35. But such a high value is adding a lot of artificial stiffness at a few elements which are near the contact region between the contact rod and the preform. Even with this value some elements are still showing hourglass modes.</p><p></p><p>When i apply pressure after this step using load segment on the preform, it initially expands but then the elements at the contact region and a few neighbouring elements fail and are deleted. </p><p> </p>

<p>Hello friends,</p><p>I modeled a postensioned concrete column where I used hollow solid part to represent prestressing sheat/duct and I also used a T-shell element to create the same. For prestressing strand I used a beam/cable material then I applied beam in solid penalty contraint and also tried different contact types like autmatic surface to surface, single surface, and beam to surface. However, I could not prevent the penetration of cable through the duct wall when the column is loaded in lateral direction. At the same when I am looking at the contact between cable and column or duct and column it seems working because when both come in contact with the column they move and not penetrate into each other. Please give me your feedback how to make this contact between cable and hollow solid part.</p><p>Thanks</p>

<p>您好，我在模拟刚体受爆炸冲击波，为什么冲击波还没有到达，但是在刚体顶部出现了泄露，刚体网格和air网格都是10mm.我有好几次都出现了这种情况，请问该怎么办呢您好，我在模拟刚体受爆炸冲击波，为什么冲击波还没有到达，但是在刚体顶部出现了泄露，刚体网格和air网格都是10mm.我有好几次都出现了这种情况，请问该怎么办呢</p>

<p>Hello, </p><p>I recently started using LS-Dyna. In AnsysMechanical, I used to define boundary conditions in terms of displacement fields. Apparently, I cannot do the same thing in LS-dyna. It seems that the displacement command is ignored. </p><p>Nevertheless, I can define other boundary conditions in terms of initial velocity and load. But as I want to study the buckling problem in structure, load conditions can lead to non-stable solution.</p><p>I suppose I missed something, does anyone have a solution? Thank you !</p><p>The picture below is from a ball compression test. The lower support is fixed, while the displacement command is defined for the upper one.</p><p></p>