[Gmsh] BEM mesh

[Lehel Banjai]

Dear GMSH authors,

I've been using your excellent tool to construct some boundary element
meshes. I was wondering if you would mind having a look at a short code
I've used. I want to publish it in my next paper so that the reader may be
able to compare methods and want it to be as short, elegant, and correct
(!) as possible. One problem that I see in the mesh that I'm getting is
that some of the triangles are not well oriented (as in the current
experiments I do not need the normal derivative it did not make a
difference).

The whole paper is at :

http://www.mis.mpg.de/publications/preprints/2009/prepr2009-58.html

and the short code is here (followed by "mesh 2D" to construct the mesh):

refine = 0.05;
Point(1) = {-0.5,-1/2,0,refine}; Point(2) = {-0.5,-3/4,0,refine};
Point(3) = {-0.5,-1,0,refine}; Point(4) = {0.0,-1,0,refine};
Point(5) = {0.0,0,0,refine}; Point(6) = {0.0,-0.5,0,refine};
Point(7) = {1.0,0.0,0,refine}; Point(8) = {0.5,0.0,0,refine};
Point(9) = {-.75,-.75,0,refine}; Circle(1) = {1,2,9};
Circle(2) = {9,2,3}; Circle(3) = {6,5,8}; Circle(4) = {4,5,7};
Line(5) = {3,4}; Line(6) = {6,1};
Extrude {{1,0,0}, {0,0,0}, Pi} {
  Line{3,6,1,2,5,4};}
Extrude {{1,0,0}, {0,0,0}, Pi/2} {
  Line{26,22,18,14,10,7};}
Extrude {{1,0,0}, {0,0,0}, Pi/2} {
  Line{36,32,40,44,48,29};}

Thanks and best regards,

Lehel

-------------------------
Dr. Lehel Banjai
Max-Planck Institute for
Mathematics in the Sciences
Inselstrasse 22
D-04103 Leipzig
Germany
tel: + 49 (0)341 9959754
fax: + 49 (0)341 9959658