[Gmsh] Avoiding duplicate surfaces with OpenCASCADE kernel

[Christophe Geuzaine]

> On 4 Mar 2019, at 19:34, Christophe Geuzaine <cgeuzaine at uliege.be> wrote:
> 
> 
> Hi Daniel,
> 
> It's one of the mysteries that we still haven't completely figured out with OpenCASCADE. I believe the issue is linked to our use of BRepBuilderAPI_Sewing to create shells ("surface loops"); but up to now we couldn't figure out how to create shells from surfaces without this, which indeed (sometimes) creates new surfaces...
> 
> Any insights from OpenCASCADE folks would be most welcome!
> 

PS : as a workaround you can add

BooleanFragments{ Volume{:}; Delete; }{}

at the end of the file, which will make everything conformal... but it's expensive.


> Christophe
> 
> 
>> On 4 Mar 2019, at 17:09, Ingraham, Daniel J. (GRC-LTV0) <daniel.j.ingraham at nasa.gov> wrote:
>> 
>> Hi Gmsh users,
>> 
>> I'm trying to mesh a geometry that has some rotational symmetry. In the
>> past, I've used the Rotate command to help with this, with good results.
>> With the built-in geometry kernel, everything works fine. With the
>> OpenCASCADE kernel, however, Gmsh appears to duplicate surfaces when I
>> define a new elementary volume after rotating the surfaces.
>> 
>> I've read on this mailing list that OpenCASCADE will duplicate surfaces
>> when they aren't oriented properly, and that this can be disabled if
>> `Geometry.OCCAutoFix` is set to `0`. I've tried that, but the duplicate
>> surfaces are still created.
>> 
>> With the built-in kernel, the orientation of a surface can be reversed
>> by negating its ID number in the `Surface Loop` call, but that doesn't
>> appear to be the case when using the OpenCASCADE kernel.
>> 
>> I've attached two example geo scripts that demonstrate the problem.
>> `builtin.geo` uses the built-in kernel, and `opencascade.geo` uses
>> OpenCASCADE.
>> 
>> Any ideas?
>> 
>> Thanks!
>> 
>> Daniel
>> 
>> <builtin.geo><opencascade.geo>_______________________________________________
>> 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
> 
> 
> 
> 
> _______________________________________________
> 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