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.

High Aspect Geometery Meshing

Please login with a confirmed email address before reporting spam

I am modeling heat transfer through a stainless steel plate with dimensions 6.35mm x 6.35mm x .1mm . Does a meshing function exist which would allow me to change the mesh element size on the thickness of the stainless steel (.1mm) without changing the mesh element size in the other two dimensions?

Thank you,
David

13 Replies Last Post Feb 11, 2015, 4:10 p.m. EST

Please login with a confirmed email address before reporting spam

Posted: 10 years ago Feb 10, 2015, 11:33 a.m. EST
Sure. First mesh the plate surface, and then swept mesh across the plate. Changing the parameters in the swept mesh do not affect on the surfaces.

br
Lasse
Sure. First mesh the plate surface, and then swept mesh across the plate. Changing the parameters in the swept mesh do not affect on the surfaces. br Lasse

Please login with a confirmed email address before reporting spam

Posted: 10 years ago Feb 10, 2015, 1:39 p.m. EST
Thank you,

How do you mesh just the surface?
Thank you, How do you mesh just the surface?

Please login with a confirmed email address before reporting spam

Posted: 10 years ago Feb 10, 2015, 1:56 p.m. EST
Mesh --> More Operations --> Free Triangular/Free Quad/Mapped (with right click)

Mesh --> Boundary ---> etc (from the ribbon)

Then select the surface to be meshed.

br
Lasse
Mesh --> More Operations --> Free Triangular/Free Quad/Mapped (with right click) Mesh --> Boundary ---> etc (from the ribbon) Then select the surface to be meshed. br Lasse

Please login with a confirmed email address before reporting spam

Posted: 10 years ago Feb 10, 2015, 2:08 p.m. EST
Mesh attached.
Mesh attached.


Please login with a confirmed email address before reporting spam

Posted: 10 years ago Feb 10, 2015, 3:04 p.m. EST
Thank you, this was very helpful. I got my planar structures plate working using this method. However, I am trying to apply this to a hollow tube. However, the same method is not working. I get the error Unsupported topology of domain. Any suggestions for this?

Thank you for all the help,
David
Thank you, this was very helpful. I got my planar structures plate working using this method. However, I am trying to apply this to a hollow tube. However, the same method is not working. I get the error Unsupported topology of domain. Any suggestions for this? Thank you for all the help, David

Please login with a confirmed email address before reporting spam

Posted: 10 years ago Feb 10, 2015, 3:28 p.m. EST
No problem, if you first mesh the inner surface. Look at the attached image.

Lasse
No problem, if you first mesh the inner surface. Look at the attached image. Lasse


Please login with a confirmed email address before reporting spam

Posted: 10 years ago Feb 10, 2015, 3:31 p.m. EST
Thank you,

It appears my problem is occurring because the hallow tube is touching the plate. I am able to mesh the hollow tube when it is not in contact with the plate. I am not sure how to resolve this issue when the tube is touching the plate which is required in my model.
Thank you, It appears my problem is occurring because the hallow tube is touching the plate. I am able to mesh the hollow tube when it is not in contact with the plate. I am not sure how to resolve this issue when the tube is touching the plate which is required in my model.

Please login with a confirmed email address before reporting spam

Posted: 10 years ago Feb 10, 2015, 3:40 p.m. EST
The problem is that swept mesh creates rectangular elements on the cross section of the tube, while your plate has triangular ones, and they do not comply. I have to think about this for a while, but now I think that it is not possible to have a swept mesh both in the tube and the plate for the reason above.

br
Lasse
The problem is that swept mesh creates rectangular elements on the cross section of the tube, while your plate has triangular ones, and they do not comply. I have to think about this for a while, but now I think that it is not possible to have a swept mesh both in the tube and the plate for the reason above. br Lasse

Please login with a confirmed email address before reporting spam

Posted: 10 years ago Feb 10, 2015, 4:32 p.m. EST
I was wrong, it was rather straightforward. First I meshed the tube, then the remaining parts of the plate surface. Finally swept mesh across the plate.

br
lasse
I was wrong, it was rather straightforward. First I meshed the tube, then the remaining parts of the plate surface. Finally swept mesh across the plate. br lasse


Please login with a confirmed email address before reporting spam

Posted: 10 years ago Feb 10, 2015, 4:46 p.m. EST
Great. Thank you. This was very helpful.

David
Great. Thank you. This was very helpful. David

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 10 years ago Feb 11, 2015, 1:44 a.m. EST
Hi

A useful trick, once you have Swept the mesh of one Domain, and then if you want to continue in "Thet" for an adjacent Domain, you can select the common Boundary and do a "Mesh Convert" which will transform the 4 node Boundary swept mesh to a triangular mesh. Then its easy to continue in thets for the remaining Domains

Another point for diffusion equation (incl. thermal flow) in the time domain: do not forget to check the heat diffusivity versus mesh size versus time steps. There are numerous threads talking about this aspect, required to be able to correctly resolve the often steep gradients of the T dependent variable (or c in classical diffusion effects), particularly appearent with rough initial conditions

--
Good luck
Ivar
Hi A useful trick, once you have Swept the mesh of one Domain, and then if you want to continue in "Thet" for an adjacent Domain, you can select the common Boundary and do a "Mesh Convert" which will transform the 4 node Boundary swept mesh to a triangular mesh. Then its easy to continue in thets for the remaining Domains Another point for diffusion equation (incl. thermal flow) in the time domain: do not forget to check the heat diffusivity versus mesh size versus time steps. There are numerous threads talking about this aspect, required to be able to correctly resolve the often steep gradients of the T dependent variable (or c in classical diffusion effects), particularly appearent with rough initial conditions -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 10 years ago Feb 11, 2015, 1:46 a.m. EST
Thanks for a tip! What is a "thet"?

Lasse
Thanks for a tip! What is a "thet"? Lasse

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 10 years ago Feb 11, 2015, 4:10 p.m. EST
Hei Lasse

Sorry I'm a bit floppy in my language, and typing, actually "tet" (and not thet) is my shortcut for "tetrahedral" (mesh element).
And actually on a Boundary you are "Converting" a "quad" to a "tri" or triangular mesh element

--
Good luck
Ivar
Hei Lasse Sorry I'm a bit floppy in my language, and typing, actually "tet" (and not thet) is my shortcut for "tetrahedral" (mesh element). And actually on a Boundary you are "Converting" a "quad" to a "tri" or triangular mesh element -- Good luck Ivar

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.