# [Gmsh] question about the direction of normals on the surface

**Christophe Geuzaine**
geuzaine at acm.caltech.edu

*Tue May 21 00:05:51 CEST 2002*

Toru Takahashi wrote:
>*
*>* Hi.
*>*
*>* I'm a beginner of 'gmsh' and interested in structuring surface data rather than
*>* volume data or FE analysis.
*>*
*>* In my situation, it is required that all the normals of triangle elements on the surface,
*>* generated by the function of 2D-meshing, must be directed outward to the surface.
*>* Here, the surface is simply modeled with the some commands (Point, Line, Line Loop,
*>* Plane Surface or Extrude). And the method of constructing of normals using the gmsh's
*>* mesh data is determined uniquely i.e. if 72 th triangle element is defined as
*>* "72 2 1 1 3 45 48 50" in the mesh data under consideration, the direction of the normal
*>* is determined as same as that of the vector calculated by
*>* (position of 48th element - 45th ) x (50th - 45th),
*>* where 'x' means the vector product of two vectors in 3D.
*>*
*>* How should I satisfy the above requirement ?
*>*
*
All the triangles belonging to a given surface will have the same
orientation. You can choose the orientation by setting a + or - sign in
the definition of the Physical Surface. (Check the result by reloading
your mesh, and setting a nonzero value for the "Normals" option in
Options->Mesh->Visibility).
Christophe
--
Christophe Geuzaine
Tel: (626) 395-4552 http://www.geuz.org
Fax: (626) 578-0124 mailto:geuzaine at acm.caltech.edu