Matthias, Thanks for pointing me to GiD. I have started experimenting with it. I see that it doesn't take .step files which complicates things for me. I can export the mesh of a .step file from gmsh and try Gid's "create geometry from mesh", but that crashes for complicated parts, as you know. What is your starting point with GiD? Do you read some kind of CAD file or mesh to start with? What do you then export to read by gmsh?
I can't tell if GiD will help or not. To experiment, I reproduce the simplest example of our problem with the attached cubes.geo. I mesh in 1D and then 2D with gmsh and export to .stl. The touching surfaces' meshes are clearly mismatched. I then import that to GiD and "create geometry from mesh" with no problem. I try to "collapse" the touching surfaces, but that has no effect - I feel like I am not using it correctly or misunderstanding what that should do. I can instead use the GiD GUI to select those surfaces, delete them and then redefine them without overlap, but that is just as well done in the gmsh GUI and will be intractable for real parts. Do you find that GiD can automatically detect and repair the contact patches, at least in simple geometries like my cubes? Hi Brent and Mark, at the moment, I am still working on my processor for the 2D mesh. But as soon as the problem I have with GiD (crash at geometry import in at least one case) is solved, I will stop that work and use GiD. It can collapse surfaces and is not expensive, so it may be worthwhile to try it wioth a one-month evaluation license. Sorry, gmsh team - gmsh is a very nice tool, and I like it, but it does not offer a real solution to this "3D geometries with double surfaces" problem. If I manage to develop my 2D mesh processor to a point where it is useful in any way, I will post it here. Matthias -- Brent A VanDevender, PhD Pacific Northwest National Laboratory Radiation Detection and Nuclear Sciences Group email:[email protected] phone:509-371-7982
cubes.geo
Description: cubes.geo
_______________________________________________ gmsh mailing list [email protected] http://www.geuz.org/mailman/listinfo/gmsh
