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.

Modelling Current Carrying Wires in COMSOL

Please login with a confirmed email address before reporting spam

Hello,

Can anyone tell me how I can build current carrying wires in COMSOL?

I am trying to build a rectangular loop of current in the x-y plane and calculate the resulting magnetic field. According to theory, in the x-y plane, the magnetic field should only have a z component (the sign of the magnetic field is determined by the direction of the current in the loop)

Is it possible to solve this problem using 2D space dimension in COMSOL, or is it necessary to use the 3D space dimension?

Thank you,
Kevin

7 Replies Last Post Jul 22, 2012, 3:03 a.m. EDT

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 28, 2010, 5:58 p.m. EDT
Can anyone help me out please?

Thanks,
Kevin
Can anyone help me out please? Thanks, Kevin

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 Oct 29, 2010, 1:26 a.m. EDT
Hi

it is well dscribed in the doc, and you must decide which physicsapplication mode you want to use, it all "depends" (and on version 3.5 or 4 ?)

--
Good luck
Ivar
Hi it is well dscribed in the doc, and you must decide which physicsapplication mode you want to use, it all "depends" (and on version 3.5 or 4 ?) -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 30, 2010, 3:09 a.m. EDT
Ivar,

I am using version 3.5 of COMSOL. When you say it is well described in the doc, I assume the doc that you were talking about is www.comsol.com/showroom/documentation/model/131/

While I have found the people on the forum very helpful, unfortunately I have to say the COMSOL help files are very incomplete and difficult to follow.

I have attached the model I am working on. In this simplistic model, I only have one finite segment of current. However COMSOL is not giving me the answer I would expect, so I assume I am doing something wrong. The z-component of the magnetic flux density should always be 0 since the current is in the z-direction.

If you could give me any advice on what I am doing wrong in my setup, I would appreciate it.

Thanks,
Kevin
Ivar, I am using version 3.5 of COMSOL. When you say it is well described in the doc, I assume the doc that you were talking about is http://www.comsol.com/showroom/documentation/model/131/ While I have found the people on the forum very helpful, unfortunately I have to say the COMSOL help files are very incomplete and difficult to follow. I have attached the model I am working on. In this simplistic model, I only have one finite segment of current. However COMSOL is not giving me the answer I would expect, so I assume I am doing something wrong. The z-component of the magnetic flux density should always be 0 since the current is in the z-direction. If you could give me any advice on what I am doing wrong in my setup, I would appreciate it. Thanks, Kevin


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 Oct 30, 2010, 4:28 a.m. EDT
Hi

I was also referring to the documentation in PDF, I can only suggest to read it and several times as there is too much in there to catch it all in one go, even after 5-6 re-readings I find new ideas, explanations that help me. And there are several exercices well described to follow. Just as in the model library, "l'exercice fait le maître"

--
Good luck
Ivar
Hi I was also referring to the documentation in PDF, I can only suggest to read it and several times as there is too much in there to catch it all in one go, even after 5-6 re-readings I find new ideas, explanations that help me. And there are several exercices well described to follow. Just as in the model library, "l'exercice fait le maître" -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jul 20, 2012, 2:01 p.m. EDT
Did you ever figure this out? I am having the same frustrating problem. I can't believe that comsol can't handle a loop current in the plane.
Did you ever figure this out? I am having the same frustrating problem. I can't believe that comsol can't handle a loop current in the plane.

Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jul 21, 2012, 7:23 p.m. EDT
Kevin,

First of all, your model is *unphysical*. Unphysical models tend to generate either nonsensical results or non-solvable computations. You simply cannot have a steady-state current in a discontinuous line segment like that. Instead, you should be able to make this model work if you put the full square current loop in your model.

Second, both the symmetry and dimensionality of a problem depend not only on the problem's geometry, but also on its solution. From that perspective, it becomes clear that this is an inherently 3D problem. So you should put in the full square current loop and solve it in 3D.
Kevin, First of all, your model is *unphysical*. Unphysical models tend to generate either nonsensical results or non-solvable computations. You simply cannot have a steady-state current in a discontinuous line segment like that. Instead, you should be able to make this model work if you put the full square current loop in your model. Second, both the symmetry and dimensionality of a problem depend not only on the problem's geometry, but also on its solution. From that perspective, it becomes clear that this is an inherently 3D problem. So you should put in the full square current loop and solve it in 3D.

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 Jul 22, 2012, 3:03 a.m. EDT
Hi

If I remember correctly, from my ACDC courses, the current return can be done efficiently by extending the loop wire to the external boundaries and have the boundaries loop the current around, this is mostly my default geometry construction when studyingthis cind of ACDC cases.
Depending on how you do it you might end up destroying a symmetry, so it needs some thinking for each geoemtry and depending on what you want to solve. Still a 2D-axi simplified reprsentation is far quicker to solve, but if you want to build something more complex with coupling, often 3D is required.

--
Good luck
Ivar
Hi If I remember correctly, from my ACDC courses, the current return can be done efficiently by extending the loop wire to the external boundaries and have the boundaries loop the current around, this is mostly my default geometry construction when studyingthis cind of ACDC cases. Depending on how you do it you might end up destroying a symmetry, so it needs some thinking for each geoemtry and depending on what you want to solve. Still a 2D-axi simplified reprsentation is far quicker to solve, but if you want to build something more complex with coupling, often 3D is required. -- 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.