[Gmsh] Problem with 2D meshes

Christophe Geuzaine cgeuzaine at ulg.ac.be
Mon Feb 26 11:01:52 CET 2007


Min Yan wrote:
> Hi Christophe,
> 
> I used only second-order *line* and *triangle* elements in a 2D mesh. 
> And I have turned the "incomplete" option off, but that didn't help.
> 
> Now after reading the "Useless nodes" thread, I'm somehow convinced the 
> problem is caused by these extra nodes. My program pre-sets the global 
> sparse matrix dimensions according to the number of nodes. The filling 
> of the global matrices are done by examining each triangles. Now, the 
> existence of some extra nodes will leave the global matrix with 
> zero-valued diagonal elements! And that leads to NaN solutions. I wonder 
> if there is any way to discard, or to identify these extra nodes?

Just loop over all elements and tag the nodes?

(Maybe we should do it in our save routine; but it can actually be 
useful to have the center of circles, etc, in the mesh file even if they 
are not connected to any element...)


> 
> BTW, please also confirm in v2.0 if the nodes are re-ordered by default.
> 

Yes, nodes are now re-ordered by default. Note that nothing in the
.msh format actually dictates this, so assuming that the nodes are
always ordered is not completely general.

Christophe



> Thanks for the nice package.
> 
> Min Yan
> 
> On 2/12/07, *Christophe Geuzaine* <cgeuzaine at ulg.ac.be 
> <mailto:cgeuzaine at ulg.ac.be>> wrote:
> 
>     The default behavior in Gmsh 2.0 is to use "incomplete" second order
>     elements (8-node quads, 20-node hexas, etc.). Maybe that's causing some
>     trouble in your code?
> 
>     You can revert to using "complete" second order elements by unselecting
>     "Use imcomplete second order elements" in Options->Mesh->General.
> 
>  
> 
>     --
>     Prof. Christophe Geuzaine
>     University of Liege, Electrical Engineering and Computer Science
>     http://www.montefiore.ulg.ac.be/~geuzaine
>     <http://www.montefiore.ulg.ac.be/%7Egeuzaine>
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> gmsh mailing list
> gmsh at geuz.org
> http://www.geuz.org/mailman/listinfo/gmsh


-- 
Prof. Christophe Geuzaine
University of Liege, Electrical Engineering and Computer Science
http://www.montefiore.ulg.ac.be/~geuzaine