[Gmsh] measuring surface deviation as a criteria of mesh quality?

Pierre JUILLARD pierre.juillard at gmail.com
Fri May 20 14:31:12 CEST 2011


Hi all,

(in the following, I only speak of surfacic elements such as quads and
trias)

When introducing transpatch remeshing in GMSH, some elements that...
- have nodes lying on original CAD surface
- but an element surface "far" from original CAD surface
...are likely to be generated between two CAD patches that may be
tangential, but are globally not in the same plane.

These elements may be considered of "bad" quality because not representing
correctly the CAD.

I would like to know if there is a way in GMSH to control the "surface
deviation" criteria that would be for an element, the max distance between
element surface and CAD surface.
A similar criteria could also be defined for edges: "edge deviation".

If this criteria can be evaluated in GMSH, is it possible to use it as a
constrain when running GMSH remeshing?
Such a criteria that would tend to generate small elements, would compete
with the criteria allowing to require for a homogeneous element size.

I thank you for any advices.
Have a good day.

Bests,

Pierre
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20110520/2fc5cea7/attachment.html>