I usually create a 1D-2D mesh with NETGEN in “Mesh_1” with good surface control before importing them into “Mesh_2” for volume mesh. I have been using this method for testing various wing geometries. The script have been working well.
Now when I try to implement the same strategy for a full aircraft model, the “Mesh_2” throws an error saying that “the imported mesh does cover the entire geometry”. I have modified the script carefully and even tried redoing the entire procedure manually for the aircraft model, but the error still persists.
I have tried making the mesh finer, exporting Mesh_1 and then reimporting them. Is there any way to fix it?
Hello,
I’m not sure to understand why you need to import and export meshes. Why don’t you create mesh_1 then just use it in the same hdf for building mesh_2 ? How are you doing this export/import operation ? Could you provide us with a little bit more details ?
Sometimes exporting and importing introduces errors to coordinates. From experience, this creates gaps that render certain feature problematic. Not sure if this is the case with your issue but you can open your mesh file in notepad to check for small errors (you might see values ranging from E-7 to E-16 added to your coordinates).
To solve this I developed a python script to round it up to E-5. This solved it.
Also to note on another user’s reply. I believe you can surface mesh and volume mesh in the same “Mesh_1”. Like NETGEN 3D then NETGEN 1D-2D.
Even after having a well defined working surface mesh in Mesh_1, I tried to read it in the same tree for Mesh_2 (volume mesh) by choosing it from the hypothesis. But it throws the same error.
Sorry for the confusion. That is the strategy that I usually use.
When it did not work, I read somewhere to try and export the surface mesh into MED and then read it back before volume meshing.