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.
Meshing thin film geometry
Posted 19 nov. 2009, 16:24 UTC−5 3 Replies
Please login with a confirmed email address before reporting spam
Hi all.
I'm trying to mesh a very thin film geometry.
x = 5000
y = 6
I set the x scaling factor from 0.025 to 0.0001 and get completely different results...
I see that the meshing in the thin film is kind of "sliced", there isn't many nodes in the x direction...
The results (electric potential) doesn't look right either.
Any suggestion?
I'm trying to mesh a very thin film geometry.
x = 5000
y = 6
I set the x scaling factor from 0.025 to 0.0001 and get completely different results...
I see that the meshing in the thin film is kind of "sliced", there isn't many nodes in the x direction...
The results (electric potential) doesn't look right either.
Any suggestion?
Attachments:
3 Replies Last Post 24 mai 2011, 10:16 UTC−4
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
meshing thin films is a very tricky issue, therefore COMSOL proposes ways to approache thin films by special boundary conditions. You can find these for thermal conductance, electric conductance, and as "shells" in the structural models. However this means that you must apply more physics and equations to your layers to emulate the desired effects. These approaches are rather well described in the documentation.
So I would propose to well identify your physical phenomena desired then to check the docs and try to use these special boundary layer tabs, rather than meshing them as volumes. Especialy if the model becomes somewhat complex.
But you might desire something not yet implemented, then you might manage to build your own equations on top of the standard system.
When I have to mesh such layers, i fight mostly with the free mesh parameters tab and trying out different "advanced" scaling features.
Good luck
Ivar
meshing thin films is a very tricky issue, therefore COMSOL proposes ways to approache thin films by special boundary conditions. You can find these for thermal conductance, electric conductance, and as "shells" in the structural models. However this means that you must apply more physics and equations to your layers to emulate the desired effects. These approaches are rather well described in the documentation.
So I would propose to well identify your physical phenomena desired then to check the docs and try to use these special boundary layer tabs, rather than meshing them as volumes. Especialy if the model becomes somewhat complex.
But you might desire something not yet implemented, then you might manage to build your own equations on top of the standard system.
When I have to mesh such layers, i fight mostly with the free mesh parameters tab and trying out different "advanced" scaling features.
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi Ivar
Would you please tell me where the documents are. this is very helpful for my modellings.
regards,
mehrdad
Would you please tell me where the documents are. this is very helpful for my modellings.
regards,
mehrdad
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
You have the thin "film" layers described in the 3.5 and back agin in the 4.2, so it means it has take some time fro COMSOl to rewrite all their doc, but that I'm not astonished about, taking into account that I needed 4 years at the uni to cover all physics tackled by COMSOL ;)
--
Good luck
Ivar
You have the thin "film" layers described in the 3.5 and back agin in the 4.2, so it means it has take some time fro COMSOl to rewrite all their doc, but that I'm not astonished about, taking into account that I needed 4 years at the uni to cover all physics tackled by COMSOL ;)
--
Good luck
Ivar