[Gmsh] Periodic elements

[demesy]

Bonjour Alexis, 

I had the same problem a few weeks ago (however, I am using Getdp). I
guess freefem++ also needs identical surface meshes when periodic
conditions are at stake. 

Basically, you have 2 options to generate identic meshes on periodic
surfaces. 

1- structured meshes : please see the .geo file attached to
http://www.geuz.org/pipermail/gmsh/2010/005259.html 

2- Extrude geometry + mesh and delete the aditionnal entries manually :
please refer to the attached.geo in
http://www.geuz.org/pipermail/gmsh/2010/005312.html for a minimal working
example. 

Depending on the physics you're dealing with, you can to use 1- or 2-
(BTW, what kind of problem are you trying to solve with freefem++?). I am
solving a vector Helmholtz equation with quasi-periodic conditions and 2-
is way more adapted (at least with getdp : I still don't understand why the
1- doesn't work at all in my case). 

Best, 

Guillaume Demsy 

On Wed, 17 Mar 2010 15:49:58 +0100, Alexis Blasselle  wrote: Dear
all,

I begin by thanking you for all the work on Gmsh, which is very
impressive.

I generate a 3d mesh with the attached file .geo. I can load it with a
Finite Element solver named FreeFem++, but do not succeed to define
periodic finite elements space on my weird cube: I have the following error
message

"the both number of face is not the same "

How can I fix this, and impose the same elements on the symmetric sides of
the cube ?

I thank you very much in advance for all piece of advise you could give
me,

Best regards,

Alex

-- 
http://www.ann.jussieu.fr/~blasselle/ [1]


Links:
------
[1] http://www.ann.jussieu.fr/~blasselle/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20100317/a7c9c7eb/attachment.html>