Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.
Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
3D geometry problems
Posted 11 avr. 2013, 02:34 UTC−4 Geometry, Mesh Version 3.5a 1 Reply
Please login with a confirmed email address before reporting spam
Hello everyone,
I am trying to solve an electric field problem using a 3D geometry comprised of five subdomains. The difference between the biggest domain (a cube with the edge of 40 micrometers) and the smallest one (a cylinder with the height of 10 nanometers and the radius of 1 nanometer) is of 4 orders of magnitude.
The geometry is generated just fine. The problem occurs when I try to generate the mesh: I get the Error 4100 message (Failed to generate mesh).
Sometimes, after editing my geometry, but keeping the same 4 orders of magnitude difference between the biggest and the smallest subdomains, I get the Error 2035 message (Edges are too close in 2D domain) when accessing the Subdomain or Boundary Conditions (before trying to generate the mesh).
I'm aware that the problem is due to the great size differences between the computational domains. Could this problem be solved somehow? Did anyone encountered this and managed to override it?
I am trying to solve an electric field problem using a 3D geometry comprised of five subdomains. The difference between the biggest domain (a cube with the edge of 40 micrometers) and the smallest one (a cylinder with the height of 10 nanometers and the radius of 1 nanometer) is of 4 orders of magnitude.
The geometry is generated just fine. The problem occurs when I try to generate the mesh: I get the Error 4100 message (Failed to generate mesh).
Sometimes, after editing my geometry, but keeping the same 4 orders of magnitude difference between the biggest and the smallest subdomains, I get the Error 2035 message (Edges are too close in 2D domain) when accessing the Subdomain or Boundary Conditions (before trying to generate the mesh).
I'm aware that the problem is due to the great size differences between the computational domains. Could this problem be solved somehow? Did anyone encountered this and managed to override it?
1 Reply Last Post 11 avr. 2013, 02:52 UTC−4
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
what I do when I have such large differences in mesh is to "box" or "sphere" in the "void volume" around my small features so that I get roughly 2-3 orders of magnitude differences, this eases the meshing. This is rather model dependent and is not always as trivial just to add a simple box, but, often OK
Else apply manual meshing, start with the small features, then finish with the last large one (possibly by splitting it up)
--
Good luck
Ivar
what I do when I have such large differences in mesh is to "box" or "sphere" in the "void volume" around my small features so that I get roughly 2-3 orders of magnitude differences, this eases the meshing. This is rather model dependent and is not always as trivial just to add a simple box, but, often OK
Else apply manual meshing, start with the small features, then finish with the last large one (possibly by splitting it up)
--
Good luck
Ivar