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:
1 decade ago
Apr 10, 2013, 12:17 p.m. EDT
Hi
the easisest is to make for structural "solid" three domains in COMSOL (start in 2D, then once it runs reasonably go for full 3D): one for the dentine/enamel, one for you bone and one for your intermediate layer, wather that is.
Once the three geometry ojects are drawn, ensure that they have clean overlapping boundaries end finish in "Union mode"
Define your material properties for the 3 domains
Define a fixed boundary, (some of the external boundaries of the "bone" structure opposite to the tooth, then set a body load on part of the tooth boundary, set up a stationary solver study and go (meshing will be done automaticlly)
This should normally be enough to get first results, you might need to increase the boundary load (pressure or total force) until you see some reasonable results.
In anycase do not overdo your first geoemtry, keep it simple until you mater COMSOL, then only complexify, you will quickly be there with nice rainbow results, ut pls do not forget to check (=validate and verify) your results agains some hand calculations, it's very easy to miss a dimension a BC or something. Normally all FEM models are validated by experimentations
--
Good luck
Ivar
Hi
the easisest is to make for structural "solid" three domains in COMSOL (start in 2D, then once it runs reasonably go for full 3D): one for the dentine/enamel, one for you bone and one for your intermediate layer, wather that is.
Once the three geometry ojects are drawn, ensure that they have clean overlapping boundaries end finish in "Union mode"
Define your material properties for the 3 domains
Define a fixed boundary, (some of the external boundaries of the "bone" structure opposite to the tooth, then set a body load on part of the tooth boundary, set up a stationary solver study and go (meshing will be done automaticlly)
This should normally be enough to get first results, you might need to increase the boundary load (pressure or total force) until you see some reasonable results.
In anycase do not overdo your first geoemtry, keep it simple until you mater COMSOL, then only complexify, you will quickly be there with nice rainbow results, ut pls do not forget to check (=validate and verify) your results agains some hand calculations, it's very easy to miss a dimension a BC or something. Normally all FEM models are validated by experimentations
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Apr 13, 2013, 2:57 a.m. EDT
Thank you very very much.
I am starting with your advice. Actually I misunderstood 'roller on a plane' tutorial and tried to create pairs between different domains. But now I understand it is not required. I am doing with overlapping boundary. Could you please enlighten me, in the region of overlapping, both the material property should be given by duplicating geometry?
Thanks again.
Siva
Thank you very very much.
I am starting with your advice. Actually I misunderstood 'roller on a plane' tutorial and tried to create pairs between different domains. But now I understand it is not required. I am doing with overlapping boundary. Could you please enlighten me, in the region of overlapping, both the material property should be given by duplicating geometry?
Thanks again.
Siva
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:
1 decade ago
Apr 14, 2013, 5:31 a.m. EDT
Hi
use pairs only when you have specific physics between the two "adjacent domain" or want to have very differnt meshe densitites. Else use union mode and the default continuity.
But be aware in union mode no two domain may overlap (bewcause that would make 3 distinct domains, usefull in some cases) So one one material per domain (you might need to apply the material a SECOND time to BOUNDARIES if this is required (specific boundary physics). OK then if you have two boundaries of distinct materials, you have an issue => assembly ?I(I would to have to check and try it out fston a simple model)
--
Good luck
Ivar
Hi
use pairs only when you have specific physics between the two "adjacent domain" or want to have very differnt meshe densitites. Else use union mode and the default continuity.
But be aware in union mode no two domain may overlap (bewcause that would make 3 distinct domains, usefull in some cases) So one one material per domain (you might need to apply the material a SECOND time to BOUNDARIES if this is required (specific boundary physics). OK then if you have two boundaries of distinct materials, you have an issue => assembly ?I(I would to have to check and try it out fston a simple model)
--
Good luck
Ivar