[Gmsh] Problems with compound curves

Maßmann, Jobst Jobst.Massmann at bgr.de
Thu Jan 22 13:45:00 CET 2015

Dear Gmsh developer,

first I would like to thank you for your great work. I really like to use your software and enjoy the very good mesh quality. Nevertheless, I have a problem with 2D meshing of compound surfaces/curves.

As far as I understood, compound curves are handled like one line in the meshing process. But in the attached example, compound curves are meshed differently from line elements.
It seems to be, that all compound curves are meshed by at least four nodes. This happens even if a very coarse density is choosen. If you please have a look at the attached example:
Compound Curve 141 {11 33} is meshed by  three lines (four nodes) but Line 2{0 18} by one line (two nodes) only. For all points, the same density has been choosen (lc=100), which should lead to a very coarse mesh.
How could this behaviour be avoided?

Attached, please find the input and output files.

I would appreciate very much to receive your answer to this problem.

Gmsh version 2.8.4
Build date: 20140207

Best regards

Jobst Maßmann

Dr.-Ing. Jobst Massmann
 Federal Institute for Geosciences and Natural Resources (BGR)
 - Geological-geotechnical Safety Analyses -
Stilleweg 2
30655 Hannover, Germany

Tel.:  +49 (0)511 643 2474
Email: Jobst.Massmann at bgr.de<mailto:Jobst.Massmann at bgr.de>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20150122/4e44476f/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 2015_01_compound_curve_problem.zip
Type: application/x-zip-compressed
Size: 4712 bytes
Desc: 2015_01_compound_curve_problem.zip
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20150122/4e44476f/attachment.zip>

More information about the gmsh mailing list