[Gmsh] Orientation of surface loops for defining a volume

[Christophe Geuzaine]

> On 19 Mar 2015, at 22:23, Sacconi, Andrea <a.sacconi11 at imperial.ac.uk> wrote:
> 
> Hi everyone,
> 
> I would like to ask you a very important question about the orientation of surfaces. It is driving me insane since I get contradictory outputs and I cannot understand where I am doing wrong.
> 
> The problem is the following: I need to generate a 3D mesh for a parallelepiped which has a spherical hole inside. Therefore, I have defined the 6 outer faces and the inner sphere. In order to correctly define the volume, I need to create two surface loops. I now have to pass the orientation: which is the correct rule? Inner normal or outer normal? In other words, clockwise orientation on the outer faces and anticlockwise on the inner ones or viceversa?

It actually does not matter for the 3D mesh generation algorithms. The one that requires outer orientation (Frontal 3D) reorients them automatically on-the-fly.

> I see that I experimentally get two different meshes if I switch from one to the other.
> Second question: for some of my simulations, I need to remesh my domain, since the hole drifts and the quality of the tetrahedra deteriorates. I create a surface loop with all the triangular faces the hole is composed of, but sometimes I get the weird message: "GMSH is not able to orient surface XXX. Normal computed with 5, 10, ....., 10000 points. Failed". How can I fix this? Does it depend on the resolution of my mesh?
> 
> I would really appreciate an answer, since I have a deadline for my PhD thesis, and so far I haven't managed to get these simulations work.
> Thank you a lot in advance !!!!!! 
> 

Which version of Gmsh do you use? Give the latest version (2.9.1) a try and let us know.

Christophe

> Andrea
> __________________________________________________________
> 
> Andrea Sacconi
> PhD student, Applied Mathematics
> AMMP Section, Department of Mathematics, Imperial College London,
> London SW7 2AZ, UK
> a.sacconi11 at imperial.ac.uk
> 
> _______________________________________________
> gmsh mailing list
> gmsh at geuz.org
> http://www.geuz.org/mailman/listinfo/gmsh

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