Hi Marc,
 
what you do seems resonable to me.
When you save the mesh, you have the choice to ignore physical groups, so 
everything is saved, or to save just the physical groups. In the latter case, 
you have to define every surface and volume as physical if you want to have it 
saved. As far as I see, the physical groups appear as numbers in the mesh. I 
have never tried to name them, so someone else has to help you here.
 
To make a finer mesh, you can do one or more of the following things:
 
* Assign a characteristic length to the points. There is a possibility to do 
this with the GUI, but I don't remember it right now. You can still do it in 
the geo file.
* Open Tools->Options->Mesh->General and set the Max element size to something 
like 10 ini your case.
* Open Tools->Options->Mesh->Advanced and select "Compute element sizes from 
curvature"
 
You can have a look at http://www.geuz.org/gmsh/screencasts/ where the GUI is 
demonstrated.
(@gmsh team: It would be nice to add in the reference manual how the described 
commands can be executed from the GUI.)
 
HTH,
 
Matthias
 
---------------------------------------------------------------- 
ERBE Elektromedizin GmbH 
Dr.rer.nat. Matthias Zenker 
Dipl. Phys. 
Grundlagenentwicklung / Fundamental Development 
Waldhoernlestrasse 17 
72072 Tuebingen 
Germany 
Phone + 49 (0) 7071 755 - 226 
Fax + 49 (0) 7071 755 - 5226 
E-Mail: <mailto:[email protected]> 
URL: http://www.ERBE-med.com <http://www.erbe-med.com/>  
---------------------------------------------------------------- 
 

________________________________

Von: Marc Secca [mailto:[email protected]] 
Gesendet: Dienstag, 29. März 2011 14:06
An: Zenker, Dr. Matthias
Betreff: Re : AW: [Gmsh] Mesh STEP file



Hi Matthias,
 
Thanks for the reply!
I'm using the dolfyn cfd solver and it requires to define physical groups in 
gmsh in order to specify boundary conditions.  How can I define these physical 
groups for that particular case?
 
1. merge the step file into a .geo file
2. select geometry/physical groups/add/then the surfaces and volume as required
3. rename the physical groups ( inlet...etc)
4. mesh (2D first then 3D)/ optimize and save the .msh file.
 
Then when I try open the .msh file, my physical group names don't appear 
anywhere?
 
Can you tell me if i'm doing this right? Am I missing something?
Also If I select 3D when meshing, the mesh is really coarse. I then refine it 
but it doesn't converge and crashes as a result? Can you shed some light on 
this issue?
 
Thanks!
Marc
 
PS:I'm relatively new to Gmsh so please bare with me if I'm asking basic 
questions :-) 

________________________________

De : "Zenker, Dr. Matthias" <[email protected]>
À : Marc Secca <[email protected]>; [email protected]
Envoyé le : Mar 29 mars 2011, 12h 47min 21s
Objet : AW: [Gmsh] Mesh STEP file


Hi Marc,
 
when I mesh your geometry (without defining physical groups) and open it with 
Elmerfem (www.csc.fi/elmer), I can see all surfaces as boundaries. So they must 
be in the msh file. The problem may be the way your cfd software reads the msh 
file.
 
HTH,
 
Matthias
 

---------------------------------------------------------------- 
ERBE Elektromedizin GmbH 
Dr.rer.nat. Matthias Zenker 
Dipl. Phys. 
Grundlagenentwicklung / Fundamental Development 
Waldhoernlestrasse 17 
72072 Tuebingen 
Germany 
Phone + 49 (0) 7071 755 - 226 
Fax + 49 (0) 7071 755 - 5226 
E-Mail: <mailto:[email protected]> 
URL: http://www.ERBE-med.com <http://www.erbe-med.com/>  
---------------------------------------------------------------- 

 

________________________________

Von: Marc Secca [mailto:[email protected]] 
Gesendet: Dienstag, 29. März 2011 12:57
An: [email protected]
Betreff: [Gmsh] Mesh STEP file


Hello!
 
I am trying to mesh the attached STEP file in gmsh 2.4.2 and define boundaries 
for a cfd solver.  I can merge the file and select the relevant surfaces and 
volume as physical groups but after I mesh to geometry, I open the .msh file 
created and cannot see the boundaries I defined.  I tried saving the geometry 
as a .geo file but surfaces are missing at the pipe intersections.
 
Is there a better way to do this? Can I repair the missing surfaces?
 
Thanks for any help you can provide.
 
Marc
_______________________________________________
gmsh mailing list
[email protected]
http://www.geuz.org/mailman/listinfo/gmsh

Reply via email to