Jump to content

MartinaC

Members
  • Content Count

    58
  • Joined

  • Last visited

Everything posted by MartinaC

  1. Hi everyone, I asked for a new licence for educational version of Altair HyperWorks and I obtained it. I substitute it in the security directory of Altair but when I run HW the attached window appear. I also followed the operations indicated in the licensing mail. "Copy the file to a location that does not change (like a user directory) and set the environment variable ALTAIR_LICENSE_PATH to the location of the license file.To set an environment variable:In Windows 10, Go to Settings -> System -> About -> System Info -> Advanced System Settings.Click the Environment Variables button, and add a new User or System level environment variable with the New button. Add the variable name as ALTAIR_LICENSE_PATH, click Browse File…, browse to the location that the license is stored and select it." The same attached window appear. What can I do?
  2. Hi everyone, I am simulating a tensile test of a homogeneous isotropic material, according to the coupon bar example in "Introduction of Non Linear Finite Element Analysis using Optistruct". I imported a different material and the coupon bar geometry is a bit different. I do not understand why the numerical force -displacement curve decreases after the yielding point, while, for example in the ebook, it follows the input trend. I though the possible problems could be the definition of the true stress-strain curve, or the NLQS parameters NINC and NLOUT, the chosen finite elements (tetrahedral; I used this type because I have to calibrate the material to perform analyses on more complex geometries). Can anyone find the problem? I attach the .fem file (ModelB_prova_NLQS). In additiion I attach a model that behaves as expected (Model2_surf_prova10_NLQS ModelB_prova_NLQS.fem MODEL2_surfs_prova10_NLQS.fem
  3. Ok, thank you. There are differences in using the engineering or the true stress-strain curves, if correct defined according to the online guide? I mean, if I have analytically modeled the engineering stress-strain curve of the material, I can choose to import the whole curve (with TYPSTRN 0) or only the plastic phase correctly derived as explained in MATS1 description (with TYPSTRN 1), without any changes from the numerical results point of view. Is this correct? Bests Martina
  4. I have to simulate a tensile test and to compare the resultant stress-strain curve with experimental results in terms of engineering stress and strains. I should import the true or engineering stress-strain curve as input data for the material? Bests Martina
  5. I'm sorry. Is this the correct file? lines.hm
  6. Hi Rahul, here the .fem file of the lines set. Thank you! Lines.fem
  7. Hi everyone, I need to define a surface from a set of lines that represent level curves of a set of mountains. The creation of surfaces from lines does not work well. Is there a way to obtain good surfaces? Bests Martina
  8. Hi everyone, is it possible to define as input data the behaviour curve of a non linear elasto-plastic matierial in a MAT1 card image material with the NL parameters activated? I tried to import the curve, changing the TYPSTRN value, but the solver Optistruct does not start. The curve has non linear trend both in the elastic that in the plastic phase. Bests Martina
  9. There are other possibilities to simulate tensile tests of non homogeneous coupon bars and export the global stress-strain curve? Because I am able to export just the stress-strain curve of each element Best Martina
  10. Hi everyone, I am simulating a tensile test with Optistruct. The analysis is NLQS and the material is isotropic. I set the analysis as proposed in the eBook "Introduction of Non linear finite element analysis using Optistruct", section 3.1.5 Tutorial: Small-Strain and Large-Strain Plasticity in a Test Coupon. I would like to understand how to set the CELAS1 element in order to use it for the building of the force-displacement output curve. I created the element, but the output curve results flat, so I think the problem is the Set. Attached is the .fem file, is usefull. Bests Martina MODEL2_surfs_JarvO1_NLQS.fem
  11. Thank you. Yes, I have to perform SED analysis.
  12. Here the files, if you can find a different solution for my problem. The .h3d is too heavy to be uploaded here. MODEL2_surfs_O1_fy200_NLQS.fem MODEL2_surfs_O1_fy200_NLQS.out
  13. Thank you very much. I modified the stress-strain input curve and it worked!
  14. Of course. I attach elso the TABLES1. I imposed the strains between 0 and 1 because shown in the eBook. The real material has maximum total strain equal to 0.3.
  15. Hi everyone, I am simulating tensile tests with welded samples. I am using NLQS analysis in Optistruct, according to the eBook Introduction to Non Linear Finite Element Analysis using Optistruct. My problem is that the stress-displacement curve after the simulation shows stresses which value is higher than the ultimate stress imposed with the TABLES1. How is it possible? Why the stress-displacement curve does not become flat? Does it have any meaning? Bests Martina
  16. Hi everyone! Is there the possibility to mesh a surface with elements like in the attached image? Bests Martina
  17. Hi everyone, I am modeling a 3D hollow element made by more components in Hypermesh to conduct NLQS analysis with Optistruct. My system is symmetric and I want to know if there is the possibility to tell the software that the unmodeled part is the mirror of the modeled part, so that it considers the constraints due to the symmetry. I know the software allows to reflects elements, but I would like to not works with too many nodes. I attach pictures: 1 is the modeled/discretized part; 2 is the entire system (modeled, unmodeled parts) Thanks Martina
  18. Of course. Now I tried to add some other support to the element: the analysis is running. I will see if it converges when it will end. Simm2D_full compression_NLQS.out
  19. With my professor we thought the model was numerically too complex to be analyzed. So we simplified the model with a scheme and analyzed its behavior. Thanks for your support Martina
  20. Hi everyone, I am performing a non-linear quasi-static analysis on a system made by elements with different materials and geometry. I attach the files. The system has an SPC for dof 1, 2, 3 at the bottom part and compression pressure on the other 3 sides. The problem is that the analysis does not converge because of error 4965. Is this a problem of the load pattern or load type, or a problem of analysis parameter setting? Before this trial I performed on the same model a non-linear quasi-static analysis of simple tension along x axis imposing a displacement on the right side of the system and an SPC for dof 1, 2, 3 on the left side and the analysis converged. After this I imposed the displacement also along y axis, but it didn't converge, so I tried with forces, and then pressures, but the analysis didn't converge. If the problem is the load setting, the definition of an envelope of the vertical loads with the horizontal loads could give good results? Bests Martina Simm2D_full_compression.hm
  21. Hi everyone, I am modeling a system composed by solid elements that are welded together. I want to develop tetra mesh from 2D automesh at the contact surfaces, but when I create the 2D mesh on the adjacent surfaces they are different and this leads me to an incorrect 3D mesh because after tetra meshing I equivanlence the nodes on the adjacent surfaces and the mesh elements fail. I use "keep connectivity" when I create the 2D mesh, but sometimes the meshes are equal, sometimes they are quite different. Why? What can I do?
×
×
  • Create New...