[Gmsh] Geometric scaling of STP files

Jean-Francois Remacle jean-francois.remacle at uclouvain.be
Thu Apr 15 18:56:12 CEST 2010

Le 14-avr.-10 à 22:47, Bill Milewski a écrit :

> Hi,
> I recently started working with GMSH and have encountered odd  
> behaviors related to scaling geometry imported from Rhinoceros CAD  
> via STEP or IGES files.  I have not seen anything in the FAQ or Wiki  
> related to these issues.
> Where is the default geometric scale factor set?  (I am using the  
> GUI initially.)  The coordinates in the STEP file are consistent  
> with my CAD model.  When I save GEO and MSH files from gmsh the  
> coordinates all seem to be scaled by a factor of 1000 for one  
> model.  However, for a second model the scale is correct.

We use the OpenCascade STEP import. Maybe we missed something related  
to units. Can you give us examples
of such problematic behavior ?

> Does the Bspline algorithm handle non-uniform weights, repeated  
> control points and multiple internal knots correctly?  I have a body  
> of revolution, with singularities at the poles, that seems to behave  
> in a funny way.

I guess OpenCascade supports NURBS ;-)

Yet, singularities in mappings may result in problematic behaviors in  
meshers. Gmsh's meshers
make use of surface metrics that are typically singular at poles. We  
try to do our best, and some
enhancements in the meshers are scheduled for removing singularities  
(you can try compound
surfaces for that , and there is a wiki !!)


> Best regards,
> Bill
> Bill Milewski, Ph.D.
> _______________________________________________
> gmsh mailing list
> gmsh at geuz.org
> http://www.geuz.org/mailman/listinfo/gmsh

Prof. Jean-Francois Remacle
Universite catholique de Louvain (UCL)
Ecole Polytechnique de Louvain (EPL) - Louvain School of Engineering
Institute of Mechanics, Materials and Civil Engineering (iMMC)
Center for Systems Engineering and Applied Mechanics (CESAME)
Tel : +32-10-472352 -- Mobile : +32-473-909930

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20100415/8a8cb641/attachment.html>