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.
Mesh generation for a thin layer
Posted 14 avr. 2010, 03:37 UTC−4 3 Replies
Please login with a confirmed email address before reporting spam
Hi,
I am trying to generate meshes in a 2-D thin layer (1000 m by 0.01 m).
I tried many times using different predefined mesh size, but I failed.
How can I generate meshes?
Thanks.
I am trying to generate meshes in a 2-D thin layer (1000 m by 0.01 m).
I tried many times using different predefined mesh size, but I failed.
How can I generate meshes?
Thanks.
3 Replies Last Post 22 juin 2010, 17:04 UTC−4
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
use shell model or try extrusion from a 2D draw
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
try a search on the forum, there are a couple of earlier discussions in these items too
have fun Comsoling
Ivar
try a search on the forum, there are a couple of earlier discussions in these items too
have fun Comsoling
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi Ivar, hi Youngmin.
i searched in the forum but sadly no thread was about thin layers and big scale differnces in deatil. I hope it is okay to go on with the dicussion here ... :)
(Sadly) we have to handle similar problems. We have a 3D (swimming pool) model with a thin (spillway) outlet boundary. The shape ratio is1:6E3 (spillway layer to longest boundary). In our case solving Navier-Stokes laminar flow had problems to converge. The reason is of cource the mesh quality. First of our ideas was to swap the thin spillway layer and mesh the remaining with freemesher. With this method the mesh size (and mesh quality) has huge gradients in the transition region from swaped- to freemesh.
It seems defining a scale factor for the freemesher (for your (thin layer) height-direction) helps to generate usefull meshes. Even without swapping the thin layer.
What also seems to work is to divide the geometry in several subdomains by using internal boundarys.
Respecting both scale factor ond subdomain-division helps also to preventing "Failed to insert point at ..." . Actually we still have no propper mesh working for us.
BTW: found this in the www. Maybe it helps. cds.comsol.com/access/dl/papers/1704/Holzbecher_pres.pdf
Do you have some more meshing-ideas that help to handle big scale gradients in one model?
Regards,
Bertram
i searched in the forum but sadly no thread was about thin layers and big scale differnces in deatil. I hope it is okay to go on with the dicussion here ... :)
(Sadly) we have to handle similar problems. We have a 3D (swimming pool) model with a thin (spillway) outlet boundary. The shape ratio is1:6E3 (spillway layer to longest boundary). In our case solving Navier-Stokes laminar flow had problems to converge. The reason is of cource the mesh quality. First of our ideas was to swap the thin spillway layer and mesh the remaining with freemesher. With this method the mesh size (and mesh quality) has huge gradients in the transition region from swaped- to freemesh.
It seems defining a scale factor for the freemesher (for your (thin layer) height-direction) helps to generate usefull meshes. Even without swapping the thin layer.
What also seems to work is to divide the geometry in several subdomains by using internal boundarys.
Respecting both scale factor ond subdomain-division helps also to preventing "Failed to insert point at ..." . Actually we still have no propper mesh working for us.
BTW: found this in the www. Maybe it helps. cds.comsol.com/access/dl/papers/1704/Holzbecher_pres.pdf
Do you have some more meshing-ideas that help to handle big scale gradients in one model?
Regards,
Bertram