Dear All,
What would you suggest as an open-source workflow when modeling with Blender?
The goal is to to electromagnetics... which requires conformal meshes.
Ive tried:
Blender>(engrid export script)>EnGrid>Elmer (failed: was never able to get EnGrid to mesh my models)
Blender>(Stl export)>Netgen>Elmer (fails: netgen will not recognize multiple bodies, so conformal meshes are impossible)
Any tips? Ive worked with hand-coded .geo files, which works, as netget can properly recognize the bodies and do conformal meshes, but this is very painful...
Example:
I would like to simulate the magnetic field and inductance resulting from a current in the spiral below.
Attachment 1: Blender 2.6 model.
Attachment 2: elmer mesh, after blender>netgen>elmer workflow (only one body!)
Conformal Mesh from Blender 2.6
-
- Posts: 7
- Joined: 12 Sep 2011, 16:09
- Antispam: Yes
- Location: Conifer, Colorado U.S.
- Contact:
Re: Conformal Mesh from Blender 2.6
If you haven't looked at e2aps (Sourceforge) you might give that a try.
-Gary
-Gary
Re: Conformal Mesh from Blender 2.6
Dear Gary,
Thanks for the reply. Interesting tool! What workflow did you have in mind using this tool?
1) Blender > ply export > e2Aps > gmsh export > gmsh > elmer?
2) Blender > ply export > e2Aps > inp export > elmer?
I tried these two methods, but get after going through either method (1) or (2) above, I get the following:
Blender 2.6 original (cube and corkscrew): Gmesh import, after ply>e2aps>msh conversion Elmer import, after ply>e2aps>inp conversion
..seems like e2aps doesnt convert correctly...
using all Windows apps by the way.
Mike
Thanks for the reply. Interesting tool! What workflow did you have in mind using this tool?
1) Blender > ply export > e2Aps > gmsh export > gmsh > elmer?
2) Blender > ply export > e2Aps > inp export > elmer?
I tried these two methods, but get after going through either method (1) or (2) above, I get the following:
Blender 2.6 original (cube and corkscrew): Gmesh import, after ply>e2aps>msh conversion Elmer import, after ply>e2aps>inp conversion
..seems like e2aps doesnt convert correctly...
using all Windows apps by the way.
Mike
-
- Posts: 7
- Joined: 12 Sep 2011, 16:09
- Antispam: Yes
- Location: Conifer, Colorado U.S.
- Contact:
Re: Conformal Mesh from Blender 2.6
Mike,
To me it looks like a Blender version thing. But that is not to say that e2aps will take you where you want to go.
Perhaps others have better suggestions.
regards
Gary
To me it looks like a Blender version thing. But that is not to say that e2aps will take you where you want to go.
Perhaps others have better suggestions.
regards
Gary
- Attachments
-
- gmsh-cscrew.jpg (52.06 KiB) Viewed 6748 times
-
- b-cscrew.jpg (62.63 KiB) Viewed 6748 times
Re: Conformal Mesh from Blender 2.6
Hi Gary,
I just tried it with Blender 2.49... still get the same result!
In any case, did you get two separete volumes after importing to gmesh, or only one??
Thanks,
Mike
I just tried it with Blender 2.49... still get the same result!
In any case, did you get two separete volumes after importing to gmesh, or only one??
Thanks,
Mike
-
- Posts: 7
- Joined: 12 Sep 2011, 16:09
- Antispam: Yes
- Location: Conifer, Colorado U.S.
- Contact:
Re: Conformal Mesh from Blender 2.6
Mike,
This is what I just did (on XP):
1. Downloaded e2aps 1.4, unzipped in its own folder.
2. Unzipped your screw2_elmer.zip in its own folder.
3. Ran e2aps, selected " Elmer to: .msh (Gmsh v) ".
4. Loaded the file mesh.nodes from the screw2_elmer folder.
5. Converted.
6. Saved.
7. Opened in Gmsh 2.5.
This takes Blender out of the equation for the purposes of testing, as well as variations in .ply format. The result was like the picture above, all structure visible, 2152 nodes, 10558 elements.
If you do an equivalent test, what happens?
regards
Gary
This is what I just did (on XP):
1. Downloaded e2aps 1.4, unzipped in its own folder.
2. Unzipped your screw2_elmer.zip in its own folder.
3. Ran e2aps, selected " Elmer to: .msh (Gmsh v) ".
4. Loaded the file mesh.nodes from the screw2_elmer folder.
5. Converted.
6. Saved.
7. Opened in Gmsh 2.5.
This takes Blender out of the equation for the purposes of testing, as well as variations in .ply format. The result was like the picture above, all structure visible, 2152 nodes, 10558 elements.
If you do an equivalent test, what happens?
regards
Gary