[Gmsh] Weird results in Transfinite Volume mesh with Transfinite AlternateLeft Surfaces

[Christophe Geuzaine]

Dear Hector,

Indeed, it's a limitation of the "Transfinite Volume" constraint implementation: conformity is guaranteed for recombined (hex) meshes; for tetrahedral meshes you need to specify the corners of the transfinite interpolation by had so that the boundary triangles match - and this is limited to the "non alternate" cases.

Christophe


> On 1 Sep 2018, at 00:06, Hector Gabriel Espinoza Román <hespinoza at utb.edu.co> wrote:
> 
> Hello:
> 
> I am meshing a cuboid without recombination in surfaces or volumes.
> 
> I use Transfinite AlternateLeft Surfaces with Transfinite Lines with odd number of nodes so I obtain a bi-symmetric surface mesh with all triangle elements without all nodes on the boundary of the surface. Then I define the Transfinite Volume and call <Mesh 3;>
> 
> The volume gets meshed but some surface elements do not coincide with the volume elements' faces. This gives me problems afterwards when I define boundary conditions over surfaces and the solver checks if that boundary element is the face of a volume element.
> 
> Attached is a simple script with many combinations which do not work as I expect. I would expect to get either all tets or all prisms with at least one surface with all triangles so I can join that to a fully unstructured pure tet volume. Only one method tested works but produces tets, hexas and pyramids.
> 
> Sincerely,
> 
> Ing. Héctor Espinoza, MSc, PhD
> Universidad Tecnológica de Bolívar
> <gmsh-3d-structured-transfinite-alternate.geo><gmsh-3d-structured-transfinite-alternate.geo.opt.back>_______________________________________________
> gmsh mailing list
> gmsh at onelab.info
> http://onelab.info/mailman/listinfo/gmsh

— 
Prof. Christophe Geuzaine
University of Liege, Electrical Engineering and Computer Science 
http://www.montefiore.ulg.ac.be/~geuzaine

Free software: http://gmsh.info | http://getdp.info | http://onelab.info