[Gmsh] structured 3d mesh with unstructured center region

[David Colignon]

Hi Danny,

I don't see any "nice" way to handle your situation.
If a rough estimate of the cylinder is satisfactory, you could try to explicitly define "by hand" the disk as an 
octahedron, with lines instead of circles on the contour, and "forcing" the mesh with lines going from the center to 4 
of the 8 points on the contour, and lines going from the disk to the square in which it is inscribed.

Regards,

Dave

--
David Colignon, Ph.D.
Collaborateur Logistique du F.R.S.-FNRS
CÉCI - Consortium des Équipements de Calcul Intensif
ACE - Applied & Computational Electromagnetics
Institut Montefiore B28
Université de Liège
4000 Liège - BELGIQUE
Tél: +32 (0)4 366 37 32
Fax: +32 (0)4 366 29 10
WWW:    http://hpc.montefiore.ulg.ac.be/
Agenda: http://www.google.com/calendar/embed?src=david.colignon%40gmail.com

On 27/11/09 17:01, Danny Lathouwers - TNW wrote:
> Hello,
>
> We are trying to mesh a geometry which mainly consists of a three
> dimensional rectangular lattice divided in 14x14x4 (x,y,z)-regions. For
> this purpose I have used the extrusion-commands to generate a structured
> mesh of hexahedron-8 elements. At the moment i'm trying to replace the 4
> center (in x,y-plane) geometrical volumes by a circular 'beam' (cilinder).
>
> I have added a simplified version (6x6x1) of the model as an attachment.
> Here prims are generated in the center region, but our fem-solving
> software can't handle prisms at the moment.
>
> Is it possible to generate a mesh for this geometry which only consists
> of hexes (necessarily unstructured inside the center region), *without*
> having to refine the mesh outside of the center area? This is important
> for the time of computation. (so each side of an element (in the
> x,y-plane) has a length equal to that of the side of the smalles
> geometrical entity (8 cm in this case)).
> So only 8 nodal points (in the x,y-plane) are available at the boundary
> of the center region.
>
> Or is there another nice way to handle this situation and create a
> conforming mesh? A rough estimate of the cilinder is already
> satisfactory in my case.
> Or is the use of prisms the only really usable option?
>
> Greetings and I'm curious to hear some good ideas,
>
> (I would also like to know how I can prevent the two errors in the
> creation of the 1D-mesh due to deleting the original elementary entities
> in the center of the geometry. But this is not the most important
> problem at the moment.)
>
>
>
> _______________________________________________
> gmsh mailing list
> gmsh at geuz.org
> http://www.geuz.org/mailman/listinfo/gmsh