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.

Electromagnet design

Please login with a confirmed email address before reporting spam

Hi,

I am trying to simulate an electromagnet, consists of two coils and cores, in order to test different core design and materials. In particular, I would like to test for the magnetic field between the cores, as a function of the current in the coils, taking into account the remanent magnetization in the core (if made from a ferromagnetic material).
Can anyone provide any tips, or point me to a similar model which I can investigate?

Thanks.

8 Replies Last Post 2013年3月6日 GMT-5 01:22
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 2010年1月2日 GMT-5 12:50
Hi

I would start with the ACDC manual, the ACDCmodref, and then take a look at the model exchange and model gallery of the web site
Good luck
Ivar
Hi I would start with the ACDC manual, the ACDCmodref, and then take a look at the model exchange and model gallery of the web site Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2010年1月8日 GMT-5 13:30
Thanks for the Reply.

I think I need a little more help:)

I am trying to simulate a scenario where the current is flowing in the coils, and then stopped (using one of the smoothed step functions).

Let's consider two coils, and air core. The magnetic field between the coils depends on the current in them. A 2D model of this scenario provides reasonable results ("2 coils, no core.mph").
When I add two soft iron cores to the model, I expect the cores to be magnetized when the current is turned on in the coils, and to remain magnetized when the current in the coils reduces to zero. This remanent magnetization should create a magnetic field between the cores. I cannot simulate that ("2 coils, iron core.mph").

Thanks.

Thanks for the Reply. I think I need a little more help:) I am trying to simulate a scenario where the current is flowing in the coils, and then stopped (using one of the smoothed step functions). Let's consider two coils, and air core. The magnetic field between the coils depends on the current in them. A 2D model of this scenario provides reasonable results ("2 coils, no core.mph"). When I add two soft iron cores to the model, I expect the cores to be magnetized when the current is turned on in the coils, and to remain magnetized when the current in the coils reduces to zero. This remanent magnetization should create a magnetic field between the cores. I cannot simulate that ("2 coils, iron core.mph"). Thanks.


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2010年1月11日 GMT-5 10:35
The iron is modeled as a non-linear material to account for saturation. However, the hysteresis is not represented and therefore, you cannot get any residual magnetization in the model you put together.
It is not trivial to take into account the hysteresis as the magnetic state of the iron would depend on the magnetization history of the material which could only be obtained through transient analysis.
Look at the different B(H) constitutive equations offered to model magnetized material; you can run simulations sequentially each one having a different B(H) equation representing each part of the magnetization curve and simulate the effect you are looking for.
Hope it helps.
Philippe.
The iron is modeled as a non-linear material to account for saturation. However, the hysteresis is not represented and therefore, you cannot get any residual magnetization in the model you put together. It is not trivial to take into account the hysteresis as the magnetic state of the iron would depend on the magnetization history of the material which could only be obtained through transient analysis. Look at the different B(H) constitutive equations offered to model magnetized material; you can run simulations sequentially each one having a different B(H) equation representing each part of the magnetization curve and simulate the effect you are looking for. Hope it helps. Philippe.

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 2010年1月11日 GMT-5 14:01
Hi

Indeed I hadnt fully understood your problem, I agree with Philippe, your iron M hysteresis is not simulated like that.

Finally you want to remain with a M after having done a B/H trajectory. Perhaps there is a possibility to have two similiar models, one where you calculate the magnetic fields over the iron, due to the current, and with whatever saturation, and the other where you set the M depending on a function related to your previously calculated magnetic field, and then solve the second model, without current.

Now how to define this function so that it represents the remanent magnetisation ? I'm not sure what to say there

Good luck
Ivar

Hi Indeed I hadnt fully understood your problem, I agree with Philippe, your iron M hysteresis is not simulated like that. Finally you want to remain with a M after having done a B/H trajectory. Perhaps there is a possibility to have two similiar models, one where you calculate the magnetic fields over the iron, due to the current, and with whatever saturation, and the other where you set the M depending on a function related to your previously calculated magnetic field, and then solve the second model, without current. Now how to define this function so that it represents the remanent magnetisation ? I'm not sure what to say there Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2010年1月18日 GMT-5 13:31
Thank you Philippe and Ivar,

Basically, you are saying that comsol cannot natively simulate hysteresis in ferromagnetic materials?

Ivar, thanks for the walk-around. I may try that, although, as you indicated, it is not trivial to write a function that outputs M for a given H, taking into account material properties and geometry.
I will also try to look for similar models, although i couldn't find any in the model exchange.
Please inform me if you have any new idea that can be helpful.

Thank you.



Thank you Philippe and Ivar, Basically, you are saying that comsol cannot natively simulate hysteresis in ferromagnetic materials? Ivar, thanks for the walk-around. I may try that, although, as you indicated, it is not trivial to write a function that outputs M for a given H, taking into account material properties and geometry. I will also try to look for similar models, although i couldn't find any in the model exchange. Please inform me if you have any new idea that can be helpful. Thank you.

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 2010年1月19日 GMT-5 14:36
Hi

Well I cannot state that COMSOL "CANNOT", but certanly that I havnt tried, nor to think it out, and it does not pop up as a trivial case for me,

If nobody alse out here can answer, then best way then is to try "support" they will certainly give you some interesting clues if its possible, and more definitively state if not possible

Good luck, and pls keep us informed, your problem is interesting indeed

Ivar
Hi Well I cannot state that COMSOL "CANNOT", but certanly that I havnt tried, nor to think it out, and it does not pop up as a trivial case for me, If nobody alse out here can answer, then best way then is to try "support" they will certainly give you some interesting clues if its possible, and more definitively state if not possible Good luck, and pls keep us informed, your problem is interesting indeed Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2013年3月5日 GMT-5 13:55
Hi Ivar, YB, Phillipe,

I am trying to simulate an electromagnet with Hiperco 50A alloy which has a non-linear magnetic properties.

From the discussion, I learned that you were talking about similar problem.

Could you please explain, if you were able to solve the problem?

Thanks in advance.

Cheers,
Dhruv
Hi Ivar, YB, Phillipe, I am trying to simulate an electromagnet with Hiperco 50A alloy which has a non-linear magnetic properties. From the discussion, I learned that you were talking about similar problem. Could you please explain, if you were able to solve the problem? Thanks in advance. Cheers, Dhruv

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 2013年3月6日 GMT-5 01:22
Hi

have you studied for example the "ecore_transformer" in the model library ? Or any of the others?

--
Good luck
Ivar
Hi have you studied for example the "ecore_transformer" in the model library ? Or any of the others? -- 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.