[Gmsh] Correct .geo file ... but self-intersections in surfaces for 3D mesh ?!?

[Simone GREMMO [531257]]

Hello,
Just add the keyword " Coherence;" at the end of your new .geo file and you should get the 3D mesh with no errors/warnings. 
I have not checked the definition of all the triangles on the cylinder surface, but probably you should check that (Coherence removes the duplicate elementary geometrical entities).

Regards,
Simone


-----Original Message-----
From: Sacconi, Andrea [mailto:a.sacconi11 at imperial.ac.uk] 
Sent: samedi 18 avril 2015 14:15
To: Simone GREMMO [531257]; gmsh at geuz.org
Subject: RE: [Gmsh] Correct .geo file ... but self-intersections in surfaces for 3D mesh ?!?

Hi all,

thanks for your help so far! Unfortunately I haven't managed to fix the issue yet. I think the two circular "lids", on the top and bottom of the cylinder, should be correct, even though I don't know how to check it in GMSH. The 2D mesh is obtained without warning.
Now the problem is: if I attempt 3D mesh generation, it seems it is meshing also the inside of the cylinder (which is weird) and it keeps telling me "couldn't compute the normal on face xxxx". I carefully tried to pass all the plane surfaces with consistent orientation, i.e. inner normal for all the triangular faces of the cylinder and the six faces of the outer cuboid. I tried to play with -1 in line loops, no luck. I also made sure to have just one "Surface Loop" which makes the final volume. What am I missing?? It's driving me nuts! The geo is attached.

Thanks again!
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

________________________________________
From: Simone GREMMO [531257] [Simone.GREMMO at umons.ac.be]
Sent: 17 April 2015 11:05
To: Sacconi, Andrea; gmsh at geuz.org
Subject: RE: [Gmsh] Correct .geo file ... but self-intersections in surfaces for 3D mesh ?!?

Hi Andrea,
If you try to create a 2D mesh and you look in detail on the top (and bottom as well) face of your model you can see some cavities appearing close to the circle boundary. Is this an expected behaviour? Maybe you selected the wrong "lines" when creating the Line Loop to define your plane surfaces.
I attach an image to try to clarify what these "cavities" are.

I hope this will help.
Regards,
Simone

-----Original Message-----
From: Sacconi, Andrea [mailto:a.sacconi11 at imperial.ac.uk]
Sent: vendredi 17 avril 2015 00:14
To: gmsh at geuz.org
Subject: [Gmsh] Correct .geo file ... but self-intersections in surfaces for 3D mesh ?!?

Hi GMSH users,

I have a problem with a .geo file (which is attached). The geometry is composed of a parallelepiped without a cylinder. As you can see, the cylinder is represented by small triangular faces, since they are the results of finite element calculations, during which the cylinder moves in time.
I can read the file without problem, but when I generate the 3D mesh, it keeps telling me that there are self-intersections in the surface mesh ... How come? All the surfaces are enumerated only once.
Could someone have a look at my file? I would really appreciate it!

Thanks in advance and congratulations for the amazing mesh generator :) 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