[Gmsh] trying to mesh a sphere

[Christophe Geuzaine]

Adrin Gharakhani wrote:

> Hello,
> 
> I have just come across gmsh and was playing with it and I ended up with
> the following problem.
> 
> I'm trying to mesh the surface an ellipsoid (for the sake of simplicity
> let's just look at the sphere). For started I thought of meshing a
> hemisphere using the following:
> 
> mu=1;
> Point(1) = {0,0,0,.1};
> Point(2) = {1,0,0,.1};
> Point(3) = {1,1,0,.1};
> Circle(1) = {3,2,1};
> Extrude Line {1, {0,1,0}, {1,0,0}, mu*Pi};
> 
> 
> When mu=<1 the algorithm meshes the surface (though the quality is poor
> near the corners using no special options). However, when I use mu>1 the
> algorithm fails to recognize the correct surface and thus ends up giving
> the wrong mesh and worse, sometimes, fails to converge. Of course, the
> same problem exists with meshing of the full sphere (when we use two
> arcs - a semicircle - to extrude the surface).
> 
> Can you please tell me what is wrong with the above set of commands?

Please read the documentation and the FAQ before posting to the mailing
list. The case you describe is clearly explained, as well as the
solution (i.e., "subdivide the surface in smaller pieces").

C.

-- 
Christophe Geuzaine
Applied and Computational Mathematics, Caltech
geuzaine at acm.caltech.edu - http://geuz.org