Hi Christophe,
thanks for your quick reply.
But I'm sorry there was a mistake in my question and I'm not sure if I
understand your answer. So I ask again.
So my question: does the meshGRegion operator requests any mesh vertex
reparametrization on its bounding faces of edges if using transfinite stuff?
I want to know:
Does the meshGRegion::operator() requests any mesh vertex reparametrization of
its bounding faces OR edges if using transfinite stuff?
Maybe an example: Assume there is a cube. 12 Edges are meshed with
GEdge::meshAttributes.method = MESH_TRANSFINITE
GEdge::meshAttributes.nbPointsTransfinite = 10;
GEdge::meshAttributes.typeTransfinite = 1;
GEdge::meshAttributes.coeffTransfinite = 1;
and 5 faces with
GFace::meshAttributes.method = MESH_UNSTRUCTURED;
and 1 face with
GFace::meshAttributes.method = MESH_TRANSFINITE;
Now I call the meshGRegion::operator() function. Question again: is there any
call to a reparametrization function of any bounding geometrical surface in the
meshGRegion::operator() function? Because the surface mesh is moved, but the
underlying geometrical surface definition doesn't change.
Best regards
Alex
On 03/31/2015 12:44 PM, Christophe Geuzaine wrote:
On 31 Mar 2015, at 09:35, Alexander Tismer
<[email protected]> wrote:
Hello,
I write a mesh operator that starts from a surface mesh and adds a boundary
layer to it (according to this paper
http://onlinelibrary.wiley.com/doi/10.1002/1097-0207%2820000910/20%2949:1/2%3C193::AID-NME929%3E3.0.CO;2-R/abstract).
Steps are:
1) create surface mesh
2) calculate normals
3) move old GFaces
4) create elements in boundary layer (green elements)
5) mesh internal GRegion (orange elements)
My algorithm works fine. But if I mesh some bounding surfaces using transfinite
edges and surfaces there occurs very very flat elements in the final internal
GRegion.
So my question: does the meshGRegion operator requests any mesh vertex
reparametrization on its bounding faces of edges if using transfinite stuff?
Hi Alex - No; if you change the definition of surfaces, you will indeed want to
reparametrize the mesh vertices so that the transfinite algorithm (which works
in parametric coordinates) can compute the new vertices correctly.
Please find attached a mesh cut of the final mesh.
Best regards
Alex
--
Alexander Tismer
Institut für Strömungsmechanik
und Hydraulische Strömungsmaschinen
Universität Stuttgart
Pfaffenwaldring 10
70550 Stuttgart
Tel.: +49 (0) 711 / 685 63007
Fax: +49 (0) 711 / 685 63255
eMail:
[email protected]
http://www.ihs.uni-stuttgart.de
<mesh.png>_______________________________________________
gmsh mailing list
[email protected]
http://www.geuz.org/mailman/listinfo/gmsh
--
Alexander Tismer
Institut für Strömungsmechanik
und Hydraulische Strömungsmaschinen
Universität Stuttgart
Pfaffenwaldring 10
70550 Stuttgart
Tel.: +49 (0) 711 / 685 63007
Fax: +49 (0) 711 / 685 63255
eMail: [email protected]
http://www.ihs.uni-stuttgart.de
_______________________________________________
gmsh mailing list
[email protected]
http://www.geuz.org/mailman/listinfo/gmsh
