[Gmsh] Surface Loops

John Doe ufnoise at gmail.com
Thu Oct 2 21:21:44 CEST 2008


Hi I just compiled from the latest source code release.  It doesn't
abort, but it only generates the triangulation of the volume surface.

Changing from:
8 corners of the box 1e-5 apart with a 5-e7 characteristic length

to:

8 corners of the box 1 apart with a 0.05 characteristic length

causes meshing to succeed.

The messages in the log, are much more numerous and are all similar to:
Info    : 5414 open elements
Error   : ERROR: Edge 2 - 28 multiple times in surface mesh
Error   : ERROR: Edge 2 - 30 multiple times in surface mesh
Error   : ERROR: Edge 14 - 28 multiple times in surface mesh
Error   : ERROR: Edge 8 - 46 multiple times in surface mesh
Error   : ERROR: Edge 30 - 31 multiple times in surface mesh
Error   : ERROR: Edge 31 - 32 multiple times in surface mesh
Error   : ERROR: Edge 32 - 33 multiple times in surface mesh
Error   : ERROR: Edge 33 - 34 multiple times in surface mesh
Error   : ERROR: Edge 34 - 35 multiple times in surface mesh
Error   : ERROR: Edge 35 - 36 multiple times in surface mesh
Error   : ERROR: Edge 36 - 37 multiple times in surface mesh

Regards,

Juan

On Thu, Oct 2, 2008 at 1:55 PM, Antoine Dechaume <boolegue at gmail.com> wrote:
> Hi Juan,
>
> your version of gmsh is quite old, you should give a try to a recent one.
>
> Antoine.
>
>
> On 2-Oct-08, at 12:52 PM, John Doe wrote:
>
>> I ended up drawing the volume in the gui and it still crash, but now
>> for a different reason.
>>
>> If I have the 8 corners of the box 1e-5 apart with a 5-e7
>> characteristic length the mesher crashes.
>>
>> If I have the 8 corners of the box 1 apart with a 0.05 characteristic
>> length the mesher works.
>>
>> This is gmsh 2.0.8 packaged for Ubuntu for a amd64 machine.
>>
>> Is there a way to make this work without crashing?
>>
>> Regards,
>>
>> Juan
>>
>> On Thu, Oct 2, 2008 at 12:56 PM, John Doe <ufnoise at gmail.com> wrote:
>>>
>>> I am learning to use gmsh, and I'd like to create a volume from the 6
>>> surfaces of a box.
>>>
>>> Right now gmsh is aborting, and I think that it may be because I don't
>>> understand how to order the indices in the surface loop command.  How
>>> are the surfaces supposed to be oriented, and when do I use a negative
>>> index?  At the end of this email are the errors I am  getting.
>>>
>>> Thank you,
>>>
>>> Juan
>>>
>>> Info    : Meshing 2D...
>>> Info    : Meshing surface 1 (Plane, MeshAdapt+Delaunay)
>>> Info    : Meshing surface 2 (Plane, MeshAdapt+Delaunay)
>>> Info    : Meshing surface 3 (Plane, MeshAdapt+Delaunay)
>>> Info    : Meshing surface 4 (Plane, MeshAdapt+Delaunay)
>>> Info    : Meshing surface 5 (Plane, MeshAdapt+Delaunay)
>>> Info    : Meshing surface 6 (Plane, MeshAdapt+Delaunay)
>>> Info    : Mesh 2D complete (1.28808 s)
>>> Info    : Mesh
>>> Info    : Meshing 3D...
>>> Info    : Meshing volume 1 (Netgen)
>>> Info    : Region 1 Face 1, 0 intersect
>>> Info    : Region 1 Face 3, 0 intersect
>>> Info    : Region 1 Face 5, 0 intersect
>>> Info    : Region 1 Face 6, 0 intersect
>>> Info    : Region 1 Face 4, 0 intersect
>>> Info    : Region 1 Face 2, 0 intersect
>>> Info    : CalcLocalH: 2638 Points 0 Elements 4840 Surface Elements
>>> Info    : Check subdomain 1 / 1
>>> Info    : 4840 open elements
>>> Fatal   : ERROR: Edge 1 - 9 multiple times in surface mesh
>>>
>>
>> _______________________________________________
>> gmsh mailing list
>> gmsh at geuz.org
>> http://www.geuz.org/mailman/listinfo/gmsh
>
>