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 The information provided may be out of date.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.

Boundary Interaction?

I am currently working on a model that uses the Heat Transfer in Fluids module in the lower domain to simulate the performance of a Metal Hydride that is absorbing heat by a given reaction.

The part of the model which only includes that portion works fine (Heat transfer in fluids and the 3 PDE expressions), the issue is Domain 2, the upper domain which is set to be air flowing over the Metal Hydride Bed and gain/lose heat based on the direction of the chemical reaction. The issue is getting the two domains to interact over the boundary where they interact. Right now the top domain is using a conjugate heat transfer function to simulate the air flowing over and heating up, but I can't seem to find the right boundary condition to get these two domains to interact.

Any suggestions?


5 Replies Last Post Dec 14, 2012, 3:54 PM EST
Posted: 5 years ago Dec 13, 2012, 1:44 AM EST
Hi

Are you sure you have selected the correct parts in the main physics and are not using the same physics on several domains ?

With COMSOL you cannot set BC applicable to external boundaries onto internal boundaries, as internal boundaries are unique and not de-doubled (normally two overlapping boundaries are made common and continuity (flux coherence) is implicitly imposed over the BC)

You need to use "assembly mode to dedouble internal boundaries, and then look after yourself the equations to couple these two overlapping boundaries

--
Good luck
Ivar
Hi Are you sure you have selected the correct parts in the main physics and are not using the same physics on several domains ? With COMSOL you cannot set BC applicable to external boundaries onto internal boundaries, as internal boundaries are unique and not de-doubled (normally two overlapping boundaries are made common and continuity (flux coherence) is implicitly imposed over the BC) You need to use "assembly mode to dedouble internal boundaries, and then look after yourself the equations to couple these two overlapping boundaries -- Good luck Ivar

Posted: 5 years ago Dec 13, 2012, 6:01 PM EST
I am quite sure that each portion is assigned to the proper domain. Where I am stuck on is the fact that the Thermal Insulation BC is automatically assigned to the common boundary. As I understood it is that this should automatically be assigned as an interior boundary but it has not worked that way.

I have no idea about how to go into "assembly mode" and manually manipulate the equations. Is there a BC that I can choose from the menu that will couple these two domains or do I have to delve into some menu to find the option to join them up?

I am quite sure that each portion is assigned to the proper domain. Where I am stuck on is the fact that the Thermal Insulation BC is automatically assigned to the common boundary. As I understood it is that this should automatically be assigned as an interior boundary but it has not worked that way. I have no idea about how to go into "assembly mode" and manually manipulate the equations. Is there a BC that I can choose from the menu that will couple these two domains or do I have to delve into some menu to find the option to join them up?

Posted: 5 years ago Dec 13, 2012, 6:13 PM EST
I am attaching the model to see if that helps at all.

The top domain (2) is air flowing over the bottom domain (1) which is a Metal Hydride that is absorbing hydrogen at the lowest boundary.
I am attaching the model to see if that helps at all. The top domain (2) is air flowing over the bottom domain (1) which is a Metal Hydride that is absorbing hydrogen at the lowest boundary.


Posted: 5 years ago Dec 14, 2012, 1:19 AM EST
Hi Kevin

Thermal insulation is applied by default to all external boundaries of your common domains, so its normal that the "internal boundary" is flagged by default "D" as thermal isolation. Normally if you define manually the link between your two physics you will add a given BC to your common boundary, then normally the Default Thermal isolation node will flag this boundary as "overriden", as for your boundary 2 in your HT case, this is the case if you select boundary 4 as "open Boundary" which is currently not applied to any boundary

--
Good luck
Ivar
Hi Kevin Thermal insulation is applied by default to all external boundaries of your common domains, so its normal that the "internal boundary" is flagged by default "D" as thermal isolation. Normally if you define manually the link between your two physics you will add a given BC to your common boundary, then normally the Default Thermal isolation node will flag this boundary as "overriden", as for your boundary 2 in your HT case, this is the case if you select boundary 4 as "open Boundary" which is currently not applied to any boundary -- Good luck Ivar

Posted: 5 years ago Dec 14, 2012, 3:54 PM EST
Thanks for helping me sort out that issue, but now the main error is back in force where duplicate velocities are keeping the solver from progressing. I think this may be due to the method used to program the Metal Hydride reaction, but it could also be because of the 2 N-S equations running in the separate domains. I tried separating the air portion from the metal hydride portion with different time dependent solvers, but without success. Attached is a picture of the error message I'm receiving, any advice on how to proceed?
Thanks for helping me sort out that issue, but now the main error is back in force where duplicate velocities are keeping the solver from progressing. I think this may be due to the method used to program the Metal Hydride reaction, but it could also be because of the 2 N-S equations running in the separate domains. I tried separating the air portion from the metal hydride portion with different time dependent solvers, but without success. Attached is a picture of the error message I'm receiving, any advice on how to proceed?

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.