Jump to content

bsquared

Members
  • Content Count

    19
  • Joined

  • Last visited

About bsquared

  • Rank
    Beginner

Profile Information

  • Country
    Canada
  • Are you University user?
    Yes

Recent Profile Visitors

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

  1. @Prakash Pagadala Not sure what you meant by switching the animation mode to "transient". It is still only creating videos where each frame is an iteration of the analysis. I'd like it to be each frame is 0.01 increase in element density, at the last iteration.
  2. Okay I will try that. Thanks a bunch :).
  3. I have a results (.h3d) file created using the OptiStruct solver, and it's a topology optimization analysis of one geometry. However I'd like to create an AVI video (or some animation) with steadily increasing element density values as viewed in the ISO plots (for the optimized topology). I want to do this for a series of element densities and 100 or 200 images created manually can be time consuming. Does anyone have any tips on this?
  4. Hi tinh, Thanks for your response! I exported the file as a STEP file instead, and it turned out to work just fine, no cleanup needed. I read something online saying that the IGES format hasn't been maintained for the last 20 years, so I think that it was the problem. Thanks again!!
  5. Hi there, Created a fairly complicated part in Autodesk Inventor over the past couple of weeks. It looks this way as a result of a topology optimization run in Optistruct a while back. (attached is an image of the part, as taken in Inventor) However, now that I have drawn the part in Inventor, I am trying to bring it back into Hyperworks to analyze the part. However I cannot use the volume tetrameshing on this part. The imported IGS file seems to have gaps where there shouldn't be, and there are red lines appearing on the model when I try to do the tetrameshing. I should mention that I did do a lot of filleting of sharp edges prior to bringing the part into Hypermesh. Any thoughts or suggestions?
  6. This might be a half-dumb question but I am going to ask it... I recently noticed that there is a large step up in solution times using roughly 1 million elements on a component I am working with. Using tetra4 (tet) elements results in 1 FE run taking about 10 minutes However using a solid-mapped mesh with mostly hex8 (hex or brick) elements results in the same FE run taking over an hour. Is this simply due to the increased number of nodes per element? Also one other question: How do we switch first-order elements (i.e. tetra4 and hex8) into 2nd-order elements (i.e. tetra10 and hex20) without deleting the mesh and starting over?
  7. Hey Rahul, Thanks for the tip! I redid the 2D automesh and made sure it was quads only, so no tria elements were there the 2nd time around. I also deleted the 2D mesh completely so there was no confusion for the solver. Attaching the properties after that and running the analysis + later on the optimization worked fine. Thanks for the assist
  8. Recently been playing around with using 2D automeshes, and then mapping the resulting meshes to mappable solids. Before this, I was using volume tetrameshes on complete solids without having to split them up (required for mappability). I am trying to assign the property (containing material properties) to the elements now, and I have done this in two ways: 1) assign the property to the component containing the mesh 2) assign the property to the elements directly. It seems that the few CTRIA3 elements in the mesh did not like the property, however, and this is causing the analysis to crash before it even starts. Attached are the model and the .out file that is obtained from trying this analysis. Any thoughts? mm12_wider.hm mm12_wider_003.stat
  9. I had already separated the tube from the block using solid edit. I later just deleted the rod because it wasn't really helping the optimization (and I didn't want it removing material from the rod). I tried deleting the 2D mesh and hiding the original geometry, but was still having issues getting the properties to "attach" to the solid-map mesh properly I decided to go with an automatically generated tetramesh for now, as I could't get the mapped mesh to work as intended. Currently trying to understand the parameters of the opimization panel as I need them :P. Thanks for all the tips!
  10. Okay so automeshing works fine, and so does the analysis and optimization (to an extent) on an automeshed geometry. The issue is that when I start splitting up the geometry (as in my original post), and try to add relevant loads and boundary conditions that way, the solver exits with a series of error messages as follows: "CQUAD4 2205 0 4417 4418 4419 4420" This line was interpreted as: 451331:CQUAD4, 2205, 0, 4417, 4418, 4419, 4420 *** ERROR # 1000 *** in the input data: Incorrect data in field # 3. Expected INT > 0 or blank, found INTEGER (0). I am suspecting this error is caused by a property not being assigned to the elements created from solid mapping. The solid map component collector has the property assigned but I am still getting this error.
  11. Attached is the hm file. motorMount_att1.hm@Rahul R - I am still getting used to hyperworks but I'll try 3d tetra meshing. Does this require mappable volumes?
  12. Hi, I realize that this topic may make me seem like a noob with HyperMesh but yes that's pretty much what I am at this point. I'm trying to use 2D surface meshing and 3D mapping of the mesh on a geometry. I've been able to get fairly far but having issues getting some of the solid parts to become mappable. Moreover, there are a lot of shared edges which I can't seem to surpresss (image below) and I'm not sure how to surpress them. I am hoping to do this to take advantage of hyperworks far superior meshing abilities. My goal is to run a loading analysis followed by topology optimization of the geometry. Any help would be appreciated!
×
×
  • Create New...