[Gmsh] Orientation of normals for compound surfaces

Rick rickgreen16 at protonmail.com
Tue Oct 1 14:49:55 CEST 2019

I am trying to mesh volumes from triangulated surface meshes using the "Compound Surface" construct to identify patches that should be parametrized and meshed together. I have ensured that the surface normals are pointing outwards according to the right hand rule but always seem to get warnings of the type

"Warning: Could not orient normal of surface ##"

If I permute the edge ordering for the "Curve Loops" I can suppress some of the warnings, but not all, and it is not clear why one permutation is allowed. If I remove the "Compound Surfaces" I don't have this issue, but also the re-parametrization is lost.

I have not been able to find any definitive documentation on this so I would be very appreciative if anyone could clear up how to get the normal orientations right for compound surfaces (I have attached a minimal unit cube geo file which illustrates the issue).
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://onelab.info/pipermail/gmsh/attachments/20191001/5edddbb5/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: unit_cube.geo
Type: application/vnd.dynageo
Size: 2233 bytes
Desc: not available
URL: <http://onelab.info/pipermail/gmsh/attachments/20191001/5edddbb5/attachment.geo>

More information about the gmsh mailing list