[Gmsh] Problem to define correctly triangular facets to make a volume + file

[Jerier Jean-François]

I send you my *.geo file (cube_gmsh.geo) to see my problem.
 I cannot define a volume, even if the plane surfaces are defined because two facets are not well oriented.


Question :

Hi all,

 I have a triangular mesh (facets) performed from the matlab function "tri".
I export this facets on Gmsh via my Matlab routine in order to define a volume.

It seems that some facets are not correctly defined or well oriented because I cannot make my volume.
I look at a simple geo file from Gmsh and I don't understand the principle or the logic to well oriented the facets.

Usually, the facet is defined in relation to its normal vector (i.e., clockwise or counter-clockwise). From the following figure, can you explain me how do I must define each facet or triangle ?

3------6------9
| \ T2 | \ T4  |
|   \     |   \     |
|     \   |     \   |
| T1 \ | T3 \ |
2------5------8
| \ T6 | \ T7  |
|   \     |   \     |
|     \   |     \   |
| T5 \ | T8 \ |
1------4------7

1 - 2 - ... - 9 = nodes
T1 - T2 - .... - T8 = triangles


In matlab, each facet is defined in the counter-clockwise. In Gmsh, I didn't manage to define a rule.

Please, can you give me more explanations on this subject ?


Reagrds
JF
_______________________________________________
gmsh mailing list
gmsh at geuz.org
http://www.geuz.org/mailman/listinfo/gmsh
-------------- next part --------------
A non-text attachment was scrubbed...
Name: cube_gmsh.geo
Type: application/octet-stream
Size: 2107 bytes
Desc: cube_gmsh.geo
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20110215/e5e4b067/attachment.geo>