[Gmsh] 2D meshing pecularities

Pascal A. Dupuis Pascal.Dupuis at esat.kuleuven.ac.be
Mon Mar 4 16:51:19 CET 2002


Hello,

I am trying to nail down a 2D meshing problem; here is a minimal test case.
I have a geometry mostly done of half cyliders. To verify the design coherency,
I regulary generate a 2D meshing: in this case, most are 'ruled surface'
delimited by two lines and two half circles. As it appears, the intermediate
points generated on the ruled surface are incorrect and seems to fall outside
the surface. The following commands illustrate this behaviour:

lc=.05;

Point(1)={0,.5,0,lc};
Point(2)={0,-.5,0,lc};

Point(3)={0,0,0,lc};

Point(11)={1,.5,0,lc};
Point(12)={1,-.5,0,lc};
Point(13)={1,0,0,lc};


Line(1)={1, 11};
Line(2)={2, 12};

Circle(3)={1,3,2} Plane{ 1, 0, 0};
Circle(4)={11,13,12} Plane{ 1, 0, 0};

//Line(5)={1,2};
//Line(6)={11,12};

Line Loop(7) = {-4,-1,3,2};
Ruled Surface(8) = {7};

//Circle(9)={1,3,2} Plane{ -1, 0, 0};

Load this file in gmsh, look from the Y axis, and then press '2'

Uncommenting the def for lines 5 and 6 generate errors about missing edges,
and after adding the half circle 9, this circle which doesn't belong to the
ruled surface is also meshed, as it can be observed from a 'X' view.

Does these result from problems in the mesh engine or did a make some
mistake with the definitions ?

TIA

Pascal Dupuis

-- 
Dr. ir. Pascal Dupuis
K. U. Leuven, ESAT/ELEN  http://www.esat.kuleuven.ac.be/
Kasteelpark Arenberg, 10; B-3001 Leuven-Heverlee, Belgium
Tel. +32-16-32 10 30