[Gmsh] Compound surfaces and lines

[Christophe Geuzaine]

> On 17 Jul 2016, at 14:35, Martin Kraska <martin.kraska at th-brandenburg.de> wrote:
> 
> Hi Christophe,
> 
> Thanks for the hint. I can indeed create a mesh with compound surfaces in 2.13.1. I observed some restrictions:
> 
> - the midside nodes of the generated elements don't follow the compound surface (except on the edges of the surface).
> - Successful meshing relies on introduction of an upper element size limit
> - If I also combine the adjacent lines in order to get rid of the short edges (they still attract small elements), then 3D meshing fails. 
> 

With the current compound implementation you will also need to define the surfaces that touch a compound line as compound. Here's the corrected script:


Merge "part.step";

// Mesh control
Mesh.ElementOrder=2;
Mesh.CharacteristicLengthMax=10;
Mesh.Optimize=1;
// Display control
Mesh.SurfaceEdges = 1;
Mesh.SurfaceFaces = 0;
Mesh.VolumeEdges = 1;
Mesh.VolumeFaces = 0;

Compound Line(1000)={63,64};
Compound Line(1001)={65,66};
Compound Surface(30) = {23,25};
Compound Surface(31) = {9,24,26};
Compound Surface(32) = {21};
Compound Surface(33) = {22};

Mesh 3;
Physical Surface("support") = {5};
Physical Surface("load") = {17};
Physical Volume("part") = {1};
Mesh.SaveGroupsOfNodes = 1;
Save "gmsh.inp";



> The console complains about intersecting surfaces.
> 
> Also, switching on all geometry healing options and setting a high geometry tolerance does not remove the small edges.
> 
> Martin
> 
>> 
>>  1. Re: Gmsh crash (Christophe Geuzaine)
>> 
>> 
>> ----------------------------------------------------------------------
>> 
>> Message: 1
>> Date: Sat, 16 Jul 2016 22:44:23 +0200
>> From: Christophe Geuzaine <cgeuzaine at ulg.ac.be>
>> To: gmsh at onelab.info
>> Subject: Re: [Gmsh] Gmsh crash
>> Message-ID: <42CDB7BE-5537-4658-8DE2-B90A2D3C4B17 at ulg.ac.be>
>> Content-Type: text/plain; charset=us-ascii
>> 
>> 
>> Hi Martin,
>> 
>> Give 2.13.1 a try: the issue is that compounds cannot currently be meshed
>> high-order (this will be supported with the new compound implementation in Gmsh
>> 3.0); also, the error message was misleading.
>> 
>> Christophe
>> 
> <part.geo><part.step>_______________________________________________
> 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

Tetrahedron V, July 4-5 2016: http://tetrahedron.montefiore.ulg.ac.be
Free software: http://gmsh.info | http://getdp.info | http://onelab.info