[Gmsh] Connecting structured to unstructured meshed volumes in 3D

Christophe Geuzaine cgeuzaine at uliege.be
Tue Mar 10 12:19:11 CET 2020



> On 10 Mar 2020, at 09:01, Octavio Castillo Reyes <ocastilloreyes at gmail.com> wrote:
> 
> Dear Prof. Ghristophhe, 
> 
> the error log (including gmsh version) is attached to this email. 
> 

The error 

   Error   : Singular value decomposition requires LAPACK

means that your version of Gmsh (that you recompiled from source?) does not include LAPACK. This is necessary for mesh generation.

Try the binary versions from the website. I also advise using a more recent version.

Christophe


> Best regards
> 
> O.
> 
> On Sun, Mar 8, 2020 at 8:40 PM Christophe Geuzaine <cgeuzaine at uliege.be> wrote:
> 
> 
> > On 5 Mar 2020, at 09:32, Octavio Castillo Reyes <ocastilloreyes at gmail.com> wrote:
> > 
> > Prof. Ghristophhe, thank you for your tip!
> > 
> > Right now I have tried it but I have obtained the following errors:
> > 
> 
> Which version of Gmsh? It should work as-is with recent versions.
> 
> Christophe
> 
> > Error   : No elements in region 2 
> > Info    : Done meshing 3D (0.005192 s)
> > Info    : Optimizing 3D mesh...
> > Info    : Optimizing volume 2
> > Info    : Done optimizing 3D mesh (0.00019 s)
> > Info    : 537 vertices 1236 elements
> > Error   : ------------------------------
> > Error   : Mesh generation error summary
> > Error   :   104 warnings
> > Error   :   105 errors
> > Error   : Check the full log for details
> > Error   : ------------------------------
> > 
> > In any case, this idea is very good. In spite of this, I would like to know if it is possible to replicate this case only with tetrahedra, in this way I can avoid using different base functions in my finite element code.
> > 
> > Best regards
> > 
> > O.
> > 
> > On Thu, Mar 5, 2020 at 9:19 AM Christophe Geuzaine <cgeuzaine at uliege.be> wrote:
> > 
> > 
> > > On 3 Mar 2020, at 13:45, Octavio Castillo Reyes <ocastilloreyes at gmail.com> wrote:
> > > 
> > > Dear gmsh developers, 
> > > 
> > > I would like to know if it is possible build a tetrahedral mesh with following features: 
> > > 
> > > - structured tetrahedral  elements on boundaries
> > > - unstructured tetrahedral elements inside the domain
> > > 
> > > Could you give some tips/ideas to build this kind of mesh (if possible).
> > > 
> > 
> > I will be done automatically (with a pyramid layer to connect both). Here's an example:
> > 
> > https://gitlab.onelab.info/gmsh/gmsh/-/blob/master/benchmarks/extrude/hybrid_pyramids.geo
> > 
> > Christophe
> > 
> > 
> > 
> > > Best regards
> > > 
> > > O.
> > > 
> > > _______________________________________________
> > > gmsh mailing list
> > > gmsh at onelab.info
> > > http://onelab.info/mailman/listinfo/gmsh
> > 
> > — 
> > Prof. Christophe Geuzaine
> > University of Liege, Electrical Engineering and Computer Science 
> > http://www.montefiore.ulg.ac.be/~geuzaine
> > 
> > 
> > 
> > _______________________________________________
> > gmsh mailing list
> > gmsh at onelab.info
> > http://onelab.info/mailman/listinfo/gmsh
> 
>> Prof. Christophe Geuzaine
> University of Liege, Electrical Engineering and Computer Science 
> http://www.montefiore.ulg.ac.be/~geuzaine
> 
> 
> 
> <hybrid_pyramids.err>

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






More information about the gmsh mailing list