|
[Sponsors] |
January 5, 2012, 12:05 |
Meshing thin, curved solids in Gmsh
|
#1 |
New Member
Tibi Chelcea
Join Date: May 2011
Posts: 3
Rep Power: 15 |
I've encountered a problem for a design I'm trying to mesh with Gmsh. Basically, the issue is in a solid that looks like a hollow cone; the wall cone is very thin (the thickness of the wall is 0.1, while the top diameter of the cone is 10, and the bottom is 6; height is 14). The issue seems to be with the 2D meshing of the outer and inner cone surfaces: some of the triangles from the outer surface intersect some triangles on the inner surface.
This design is supposed to be simulated a lot of times with slight variations, so I'd like to get a small mesh that's amenable to quick simulations. In my experiments so far a mesh of about 100K elements with a meshing time of 45s is good. I can change Mesh.MinimumCirclePoints in the .geo file, but if I set it high enough to avoid the issue, meshing takes 4 minutes and results in 800k mesh elements. I can also change the characteristic length (either for the entire design or just a box around this solid), but the result is again a large mesh and very slow meshing. Is there a way to avoid the issue of intersecting triangles while avoid a big penalty in speed/size? Thanks. Tibi |
|
Tags |
gmsh, thickness, thin walls |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
[Gmsh] GMSH error when meshing "airfoil_validation" githib's example | uzzi | OpenFOAM Meshing & Mesh Conversion | 2 | October 11, 2018 17:05 |
[Gmsh] Vertex numbering is dense | KateEisenhower | OpenFOAM Meshing & Mesh Conversion | 7 | August 3, 2015 11:49 |
thin extrusion meshing | guillaume | Main CFD Forum | 0 | October 27, 2008 19:36 |
Xperimental data on thin curved plates/airfoils | turb | Main CFD Forum | 0 | September 15, 2005 16:15 |
Meshing thin surfaces | Louwrens | CFX | 8 | August 18, 2003 18:49 |