Jump to content


  • Content Count

  • Joined

  • Last visited

About lanwujiu

  • Rank

Profile Information

  • Are you University user?

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. Thanks! That's exactly what I want. So according to the definition of Plan strain infinite elements, I should turn the orientation of Y to the supposed infinite direction? Besides, is there any advice for better aligning the direction when creating the mesh?
  2. The purpose is to define specific infinite directions for elements using Abaqus as a solver. From the Abaqus manual, the infinite elements should have an origin pole and the first-second nodes pointing indicates the infinite direction. If I would like to achieve infinite directions as follows, how should I modify the node numbers in Hypermesh? This is a 2D model and I am going to use CINPE4(4-nodes 2D elements). I referred to this method too but I am confused by what should be the order of numbering. https://connect.altair.com/CP/SA/training/self_paced/aero_v13/PDF/chapter11_2_demonstration.pdf
  3. Hi, Thank you for your instruction video. I also managed to do this by detaching the connected elements and generated new nodes.
  4. 3D model with solid elements, there are two surfaces in contact and friction interaction were set in Hypermesh. When .inp file was imported to ABAQUS and check data, there is a warning "the slave assembly has 1200 nodes in common with master surface". Should I separate the nodes on the interface and how to do it?
  5. Hi, thank you for your advice. The surface between the two parts are set as friction contact, after equivalence, Abaqus warns that the slave surf and master surf have all the nodes in common. How can I only select the inequivalence of the cross section?
  6. Thank you for your reply, I have deleted the 2d mesh before submitting into Abaqus, if I didn't do it the software would warn no material properties. To ease selection I associate the geometry to the elements, is that the case?
  7. The geometry was output from Solidworks. I did the mesh in Hypermesh by creating a vertical 2D element and spin 180 degrees and then reflect another half. No connectivity issue in check element tool. The .inp file was imported in Abaqus, a pressure was added to the upper surface of the inner part, and the results were as follows: the elements of the inner part overlapped each other. Since I have meshed model directly in Abaqus before, as for the same force they should be sinking down together instead of separated into two parts. Is there anything wrong with the mesh connectivity or any advice for meshing in Hypermesh? Attached the before and after loading, as well as the .inp exported from Abaqus. hpmtest.inp
  8. Hi, thank you for your advice. Your method works! But I am a little confused now. I was making 2d mesh but I selected the surfaces by geometry, which refer to the solid. So should I select the surface one by one now?
  9. I have created an assembly which is a rod inserting into the column in Solidworks and checked there is no interference. I output the file to .stp and import in Hypermesh, and check geometry for duplicated surfaces. 3 were found. But I cannot delete the duplicated surface and the message "3 selected were attached to solid; deletion ignored" Is it the problem of the original 3D model? How could I fix it?
  10. Hi, thank you. The problem is solved.
  11. Thanks, I merged those parts and tried again, the vertical line is from the geometry so I could not remove it. The lower part is not a part of a sphere but a flat bottom, maybe it is the case why it is still mappable. I am trying to mesh it in a quarter model.
  12. Thank you for your advice! I saw the other user in this line sharing a way of partitioning the sphere. Are you suggesting the same way? What does it mean by merging the existing partitions?
  13. I have a model with 5 parts and it is symmetric. I partitioned as much as I can, but still, have an issue in mappable solid as the picture below (this is the simples and most complex part in the model) Any advice on partitioning? Thanks!
  • Create New...