Hi Christophe,further investigations are making me think it is ticking the opencascade "cut and merge faces" option, this gets rid of duplicate faces but the renumbering process seems to be random such that when you reload the file the surface numbers change (not checked lines & points) ?Now most CAD consisting of connecting multi bodies produce duplicate faces, if these can be removed before importing then this might then work OK?I tried loading the step file using the cut and merge option, saved the file out to a new step file then imported this new file but this made no difference.if I remove any duplicate entities in GiD and export as iges the problem goes away, but it is not an option if I only want to use a non commercial pre-processor ;-(RegardsMark From: Christophe Geuzaine <[email protected]> To: Mark Smith <[email protected]> Cc: "[email protected]" <[email protected]> Sent: Thursday, January 29, 2015 11:19 AM Subject: Re: [Gmsh] GMSH renumbers elemntaty tags on reload
> On 29 Jan 2015, at 10:45, Mark Smith <[email protected]> wrote: > > Hi Christophe, > Unfortunately it makes no difference ;-( > I used cadexchanger from http://www.cadexchanger.com/ to convert the step > file into a .brep file but this suffers with the same problem. > If is do file: save as and type say file2.geo and with the GEO options box > "save physical group lables" ONLY ticked I get a file with all the points, > lines line loops etc. When I lad this the element tag numbers stay fixed, ( > it is relatively simple geometry consisting of volumes bounded by plane > surfaces) but I cannot get the surfaces to be filled (rendered) but only > crossed faces even when I set Mark - Do not save your .brep/.step as a .geo! Gmsh is designed to work directly with the native files - "translating" them to .geo is only there for very specific uses or debugging purposes: the .geo CAD kernel is far too rudimentary to handle all the complexities of .step/.brep > Geometry.SurfaceType = 2 this makes it difficult to pick surfaces to put > into surface physical groups infact changing Geometry.SurfaceType to 0 1 or 2 > make no difference to the way the surfaces are displayed? > Strangely if I load in my cad geometry file Merge "file.STEP"; etc then the > surfaces displayed are as expected if one changes the setting of > Geometry.SurfaceType, any ideas? > > Thanks in advance. > Best regards > Mark > > > From: Christophe Geuzaine <[email protected]> > To: Mark Smith <[email protected]> > Cc: "[email protected]" <[email protected]> > Sent: Wednesday, January 28, 2015 7:11 PM > Subject: Re: [Gmsh] GMSH renumbers elemntaty tags on reload > > > > On 28 Jan 2015, at 15:36, Mark Smith <[email protected]> wrote: > > > > > > Gmsh version 2.8.5 > > > > Hi, I load a step file generated in solidworks saved in AP203 or AP214 > > format and load in into GMSH using a .geo file with the following lines > > > > // in gmsh options-geometry make sure the "cut and merge faces" is checked > > // to remove duplicate entities on import of STEP file > > Merge "file.STEP"; > > Geometry.AutoCoherence = 1; > > Coherence; > > > > in the options:geometry I get the model to display the surface labels with > > their elementary tags number, if I hit reload the tag numbers change this > > make defining physical surfaces impossible since the surfaces shown in the > > surface physical groups keep changing;-( > > I have not had this behaviour before and at this moment it is a show > > stopper, is there a way of preventing this behaviour? > > > > I don't see this behavior when I do the same thing over here (note that > "Geometry.AutoCoherence = 1; Coherence;" will have no effect on your STEP > data)... Does adding "Delete All;" before the "Merge" command help? > > > > > > > Thanks in advance > > Mark > > > > > > > _______________________________________________ > > gmsh mailing list > > [email protected] > > http://www.geuz.org/mailman/listinfo/gmsh > > -- > Prof. Christophe Geuzaine > University of Liege, Electrical Engineering and Computer Science > http://www.montefiore.ulg.ac.be/~geuzaine > > > > > -- Prof. Christophe Geuzaine University of Liege, Electrical Engineering and Computer Science http://www.montefiore.ulg.ac.be/~geuzaine
_______________________________________________ gmsh mailing list [email protected] http://www.geuz.org/mailman/listinfo/gmsh
