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 The information provided may be out of date.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.

How to couple T variables in bluk and thin film

Please login with a confirmed email address before reporting spam

We are modeling heat transfer in solids that contain a thin film within the bulk. So we have "Heat Transfer in Solids" for the bulk, and "Thin Conductive Shell" for the thin film. The default variable for the former is T and the latter T2. We haven't found a way to specify "T=T2" (it is the same temperature we are trying to solve), except just change T2 to T, i.e. setting the dependent variable in both Physics to T. That seems to work, except we would get Reset History error, see this thread:

www.comsol.com/community/forums/general/thread/25804/

What is the correct way to tell COMSOL that T and T2 are the same temperature variable?

11 Replies Last Post 21 feb 2013, 20:58 GMT-5
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 2 feb 2012, 15:50 GMT-5
Hi

But why do you have a "T2" ?
Normally the thin film boundary physics should be in the same physics as the domain hence only one T and the T of the thin film should be the temperature of the domain when we tend tovards the boundary limit, no ?

--
Good luck
Ivar
Hi But why do you have a "T2" ? Normally the thin film boundary physics should be in the same physics as the domain hence only one T and the T of the thin film should be the temperature of the domain when we tend tovards the boundary limit, no ? -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 3 feb 2012, 09:25 GMT-5

Hi

But why do you have a "T2" ?
Normally the thin film boundary physics should be in the same physics as the domain hence only one T and the T of the thin film should be the temperature of the domain when we tend tovards the boundary limit, no ?

--
Good luck
Ivar


Thanks for the reply.

I was using two physics: Heat Transfer in Solids (ht) and Thin Conductive Shell (htsh). Are you saying that I should use only one physics: Heat Transfer in Solids (ht)? How do I implement the thin film in that case? I looked under Heat Transfer in Solids, the only sub-item under it is Thin Thermally Resistive Layer. But it does not have heat capacity in the equations. On the other hand the Thin Conductive Shell physics does have heat capacity in the equations, which should be more accurate. I still don't see a perfect solution yet.
[QUOTE] Hi But why do you have a "T2" ? Normally the thin film boundary physics should be in the same physics as the domain hence only one T and the T of the thin film should be the temperature of the domain when we tend tovards the boundary limit, no ? -- Good luck Ivar [/QUOTE] Thanks for the reply. I was using two physics: Heat Transfer in Solids (ht) and Thin Conductive Shell (htsh). Are you saying that I should use only one physics: Heat Transfer in Solids (ht)? How do I implement the thin film in that case? I looked under Heat Transfer in Solids, the only sub-item under it is Thin Thermally Resistive Layer. But it does not have heat capacity in the equations. On the other hand the Thin Conductive Shell physics does have heat capacity in the equations, which should be more accurate. I still don't see a perfect solution yet.

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 3 feb 2012, 15:22 GMT-5
Hi

I have in HT "Thin Conductive layer" for internal or external boundaries, and "Thin thermal layer" for internal boundaries only

No need for shell element hysics in addition, I would typically use a htsh with a structural "shell" model

--
Good luck
Ivar
Hi I have in HT "Thin Conductive layer" for internal or external boundaries, and "Thin thermal layer" for internal boundaries only No need for shell element hysics in addition, I would typically use a htsh with a structural "shell" model -- Good luck Ivar


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 6 feb 2012, 11:57 GMT-5
Hi, Thanks again! I looked in the menu and did find "Highly Conductive Layer" under Heat Transfer in Solids. It works now. :)
Hi, Thanks again! I looked in the menu and did find "Highly Conductive Layer" under Heat Transfer in Solids. It works now. :)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 1 mar 2012, 14:25 GMT-5
Hi Ivar,

Following up on your response, I am wondering if the "thin thermal layer" you mentioned is the same as "thin thermally resistive layers" in 4.2? Also, how is it different from "highly conductive layer"? I see from its equation that it has up and down direction which probably explains why it is for internal boundaries only. The doubt I have is whether or not I can make the boundaries (assigned to either "thin thermally resistive layer" or "highly conductive layer") either highly conductive or resistive by changing the value of thermal conductivity?

Thank you

Harn Chyi Lim
Hi Ivar, Following up on your response, I am wondering if the "thin thermal layer" you mentioned is the same as "thin thermally resistive layers" in 4.2? Also, how is it different from "highly conductive layer"? I see from its equation that it has up and down direction which probably explains why it is for internal boundaries only. The doubt I have is whether or not I can make the boundaries (assigned to either "thin thermally resistive layer" or "highly conductive layer") either highly conductive or resistive by changing the value of thermal conductivity? Thank you Harn Chyi Lim

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 1 mar 2012, 15:31 GMT-5
What I found was "thin resistive layer" does not have heat capacity and "highly conductive layer" does. So I stuck with the latter.
What I found was "thin resistive layer" does not have heat capacity and "highly conductive layer" does. So I stuck with the latter.

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 1 mar 2012, 17:40 GMT-5
Hi

its my typo I do not believe it has changed with the version (I have 4.2a .166)

The thin thermally resistive layer adds a temperature step along the normal of the boundary.
The thin highly conductive layer adds a thin layer that conduct along the boundary

careful not to add both to a common (union) boundary, my version fails with an error

If you are using assembly mode, then note that the pair continuity, pair thin thermally resistive layer, pair highly conductive layer can all be added but they interfer, and from my understanding only one should be enbled at each time

--
Good luck
Ivar
Hi its my typo I do not believe it has changed with the version (I have 4.2a .166) The thin thermally resistive layer adds a temperature step along the normal of the boundary. The thin highly conductive layer adds a thin layer that conduct along the boundary careful not to add both to a common (union) boundary, my version fails with an error If you are using assembly mode, then note that the pair continuity, pair thin thermally resistive layer, pair highly conductive layer can all be added but they interfer, and from my understanding only one should be enbled at each time -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 6 mar 2012, 17:50 GMT-5
Thanks for the replys.
Another question I have come up with is whether or not I can plot the heat flux for those boundaries assigned to highly conductive layers. I made a surface plots for that, but from the value, I think the results only display the values for the domain.
Thanks for the replys. Another question I have come up with is whether or not I can plot the heat flux for those boundaries assigned to highly conductive layers. I made a surface plots for that, but from the value, I think the results only display the values for the domain.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 24 ago 2012, 04:44 GMT-4
Hello
I am simulating cooking egg in COMSOL. do you think I could use thin thermal resistive layer to simulate yolk?
Because it seems layer have an upper suface and bottom surface. But for egg. The geometry confuse me. Thank you.
Regards,
Siqi
Hello I am simulating cooking egg in COMSOL. do you think I could use thin thermal resistive layer to simulate yolk? Because it seems layer have an upper suface and bottom surface. But for egg. The geometry confuse me. Thank you. Regards, Siqi

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 24 ago 2012, 14:51 GMT-4
I believe you can. As for geometry, you may refer to the link below if you run into any issue.

www.comsol.com/support/knowledgebase/1031/
I believe you can. As for geometry, you may refer to the link below if you run into any issue. http://www.comsol.com/support/knowledgebase/1031/

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 21 feb 2013, 20:58 GMT-5
Does anyone know if there is any way model a thermally resistive layer with a tangential conduction? I know the equation only as a normal component, and as Ivar said, it cannot be used in parallel with highly conductive layer. WIth only the normal component, it does not react to the gradient applied in the tangential direction.



Hi

its my typo I do not believe it has changed with the version (I have 4.2a .166)

The thin thermally resistive layer adds a temperature step along the normal of the boundary.
The thin highly conductive layer adds a thin layer that conduct along the boundary

careful not to add both to a common (union) boundary, my version fails with an error

If you are using assembly mode, then note that the pair continuity, pair thin thermally resistive layer, pair highly conductive layer can all be added but they interfer, and from my understanding only one should be enbled at each time

--
Good luck
Ivar


Does anyone know if there is any way model a thermally resistive layer with a tangential conduction? I know the equation only as a normal component, and as Ivar said, it cannot be used in parallel with highly conductive layer. WIth only the normal component, it does not react to the gradient applied in the tangential direction. [QUOTE] Hi its my typo I do not believe it has changed with the version (I have 4.2a .166) The thin thermally resistive layer adds a temperature step along the normal of the boundary. The thin highly conductive layer adds a thin layer that conduct along the boundary careful not to add both to a common (union) boundary, my version fails with an error If you are using assembly mode, then note that the pair continuity, pair thin thermally resistive layer, pair highly conductive layer can all be added but they interfer, and from my understanding only one should be enbled at each time -- Good luck Ivar [/QUOTE]

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.