[CIG-SHORT] [CIG-HELP] GoCAD -> CUBIT

Brad Aagaard baagaard at usgs.gov
Wed Oct 27 08:53:22 PDT 2010


David-

The basic workflow for using SCEC CFM files in CUBIT is
(1) convert the Gocad tsurf file into a facet file (see the CUBIT 
documentation for information on the facet file format; this conversion 
is usually done with a shell script)
(2) import the facet file into CUBIT via the import command
(3) convert the triangulated surface into a NURB surface by creating a 
mapped mesh of the surface
(4) save the NURB surface to a file for later use

PyLith v1.5.x contains an example of this workflow for a DEM in 
examples/meshing/surface_nurbs/triangles.

Our experience is that some fault surfaces in the SCEC CFM require 
manual cleanup before they are suitable for mapped meshing. This is due 
to the fact that CUBIT needs a surface with four edges for mapped 
meshing and some fault surfaces have irregular boundaries. I have 
experimented with MeshLab (http://meshlab.sourceforge.net/) for manually 
cleaning up edges and merging multiple triangulated surfaces. Carl Gable 
has recently added some nice features to LaGriT (http://lagrit.lanl.gov) 
for more automated processing (e.g., smoothing and re-triangulating).

Brad


On 10/27/2010 08:01 AM, Walter Landry wrote:
> Hi David,
>
> I am copying this to the cig-short mailing list, where they can better
> help you.
>
> Cheers,
> Walter Landry
> walter at geodynamics.org
>
>
> david.oglesby at ucr.edu wrote:
>> To whom it may concern, I'm interested in using the SCEC CFM faults
>> to generate finite element meshes in CUBIT. It would appear based on
>> your workflow diagrams that this interface between the two codes has
>> already been accomplished. Could you put me in touch with someone at
>> your organization who has done this? I would appreciate some
>> tips.
>>
>> Thanks!
>> --David
> _______________________________________________
> CIG-SHORT mailing list
> CIG-SHORT at geodynamics.org
> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>



More information about the CIG-SHORT mailing list