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 The information provided may be out of date.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.

U-shaped Plate (Heating & Magnetic Fields)

Please login with a confirmed email address before reporting spam

Greetings!

As you can see in the uploaded file, I'm trying to simulate the heating and the magnetic field of an array of U-shaped plates. Each plate has an independent current source.

So my problem is that I have an idea of what I want to do but I can't implement this in Comsol. I've already gone through the AC/DC Module-Tutorial and I watched the Heat-Transfer Video-Tutorial (the Fuse).

So as you can see in the file I uploaded, I used the "Electric Current" Module and the "Heat Transfer in Solids" Module (as I've seen in the Fuse-Tutorial). The magnetic field simulation is not that important so my focus is on solving the heat problem.

I can't get any results with the settings I use at the moment. The solver ran for about 2 hours (1600 Iterations on the GMRES solver) before I stopped it. So I guess there are some problems with the settings in either the "Electric Currents" or the "Heat Transfer in Solids".



Secondary question to the AC/DC Module:

In the AC/DC Module Tutorial you build a hollow sphere around the model. May I ask why? I guess to simulate an open space but I don't really know how to implement this in my problem because I really don't know what boundaries to select for that purpose.



5 Replies Last Post 23 janv. 2012, 16:19 UTC−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 22 janv. 2012, 11:41 UTC−5
Hi

what about trying out the JH joule heating module, it's a pre-cooked HT + EC

--
Good luck
Ivar
Hi what about trying out the JH joule heating module, it's a pre-cooked HT + EC -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 23 janv. 2012, 09:21 UTC−5
Hello again!

Thanks for your answer.

I watched the Joule-Heating Video Tutorial again and found some mistakes I made.


Now I get quite realistic results but just for one Plate. As you've seen in the file I uploaded the "Array" in Geometry 1 was disabled. Now i wanted to simulate the situation for 3 or even more plates and enabled the Array again and tried to mesh them.

Unfortunately I get the following errors when I do so.

"Failed to generate mesh for domain.
-Domain 6
Failed to respect edge element on geometry edge.
-Edge: 121
some coordinates
Internal error in boundary respecting."

and

"Failed to generate mesh for domain.
-Domain 28
Failed to respect boundary element on geometry face.
-Boundary: 333
some coordinates
Internal error in boundary respecting."


I also tried to mesh every plate individually with its "own" tetrahedral mesh but I got the same error messages. I don't know how to solve this.

Is there anything like a Tutorial for (advanced) meshing or a way to solve this problem?

Thanks!
Hello again! Thanks for your answer. I watched the Joule-Heating Video Tutorial again and found some mistakes I made. Now I get quite realistic results but just for one Plate. As you've seen in the file I uploaded the "Array" in Geometry 1 was disabled. Now i wanted to simulate the situation for 3 or even more plates and enabled the Array again and tried to mesh them. Unfortunately I get the following errors when I do so. "Failed to generate mesh for domain. -Domain 6 Failed to respect edge element on geometry edge. -Edge: 121 some coordinates Internal error in boundary respecting." and "Failed to generate mesh for domain. -Domain 28 Failed to respect boundary element on geometry face. -Boundary: 333 some coordinates Internal error in boundary respecting." I also tried to mesh every plate individually with its "own" tetrahedral mesh but I got the same error messages. I don't know how to solve this. Is there anything like a Tutorial for (advanced) meshing or a way to solve this problem? Thanks!


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 23 janv. 2012, 10:57 UTC−5
Hi

you have a nasty geometry to mesh. You could try the new "copy" mesh for similar geometries. One of the difficulties, that destroys the symmetry is your round cables/rods. without you can mesh the lateral edge surface and sweep a structured mesh through the long direction.

It could be worth to forget/remove the round Cy cables and apply the GND voltage to the Al sides (at least to start with

--
Good luck
Ivar
Hi you have a nasty geometry to mesh. You could try the new "copy" mesh for similar geometries. One of the difficulties, that destroys the symmetry is your round cables/rods. without you can mesh the lateral edge surface and sweep a structured mesh through the long direction. It could be worth to forget/remove the round Cy cables and apply the GND voltage to the Al sides (at least to start with -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 23 janv. 2012, 15:17 UTC−5
Hello again!

I removed the rods/cables as you suggested (they would have been interesting for the magnetic field - to see how they influence the "main" magnetic field but as I said before my focus lies on solving the heat transfer problem.)

And the "copy" mesh function was the way to go! The mesh was generated in almost no time.

Thanks for the advice!
Hello again! I removed the rods/cables as you suggested (they would have been interesting for the magnetic field - to see how they influence the "main" magnetic field but as I said before my focus lies on solving the heat transfer problem.) And the "copy" mesh function was the way to go! The mesh was generated in almost no time. Thanks for the advice!

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 23 janv. 2012, 16:19 UTC−5
Hi

you can always put the rods back later, one way is to use assembly mode, then you can join uneven meshings over a common overlapping boundary, check the doc. But if you want to do some EC or MF or MEF analysis, do not forget that you need air/vacuum around your model to allow the fields to loop around, and that you have to mesh that too ;)

--
Good luck
Ivar
Hi you can always put the rods back later, one way is to use assembly mode, then you can join uneven meshings over a common overlapping boundary, check the doc. But if you want to do some EC or MF or MEF analysis, do not forget that you need air/vacuum around your model to allow the fields to loop around, and that you have to mesh that too ;) -- 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.