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.

Magnet - Magnetic and electric fields

Please login with a confirmed email address before reporting spam

Hi Jens,

I want to simulate this magnet (see attached file) and then see the effect of the induced magnetic field in a conducting disk that is situated under the magnet. I choose the magnetic and electric filds (mef) physics.
I have again the error message: singular matrix
Thanks in advance.
Have a nice day!

Regards.

John J


5 Replies Last Post Jul 17, 2012, 1:33 a.m. EDT
Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jul 12, 2012, 12:59 p.m. EDT
This appears to be a continuation (or at least very similar) to your earlier post. In that earlier case, you received advice from Jens Krause to the effect that you should not normally be specifying an external current if all you have is a moving disk and fixed magnets. Did you follow that suggestion?
This appears to be a continuation (or at least very similar) to your earlier post. In that earlier case, you received advice from Jens Krause to the effect that you should not normally be specifying an external current if all you have is a moving disk and fixed magnets. Did you follow that suggestion?

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jul 12, 2012, 1:37 p.m. EDT
Hi Robert,

Thank you for your response.

Yes, I followed his suggestion: In fact, I deleted the external current density and I get the same problem: singular matrix
My prblem is how to build in COMSOL the permanent magnets alternately magnatized (it is an NdFeB magnets) as shown in the attached file. it will be greatful if you help me.
Thank you!

Regards

John J
Hi Robert, Thank you for your response. Yes, I followed his suggestion: In fact, I deleted the external current density and I get the same problem: singular matrix My prblem is how to build in COMSOL the permanent magnets alternately magnatized (it is an NdFeB magnets) as shown in the attached file. it will be greatful if you help me. Thank you! Regards John J


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 15, 2012, 1:35 p.m. EDT
Hi

Why do you use MEF (which solves for the magnetic field and for V ?) isnt it enough with MF?.

As MF is far simpler to set up. in MEF you need to give some conduction to all materials (including air)and yo have far moe DoFs t solve. If you do not use any Eddy current analysis, use rather MF. For me you can set up the magnets by adding two amper law physics with magnetisation for the two pairs ofmagnets, such to polarise them correctly, or you can use a cylindrical coordinate and use a condition on the phi angle to get the correct sign of the magnet polarisation

--
Good luck
Ivar
Hi Why do you use MEF (which solves for the magnetic field and for V ?) isnt it enough with MF?. As MF is far simpler to set up. in MEF you need to give some conduction to all materials (including air)and yo have far moe DoFs t solve. If you do not use any Eddy current analysis, use rather MF. For me you can set up the magnets by adding two amper law physics with magnetisation for the two pairs ofmagnets, such to polarise them correctly, or you can use a cylindrical coordinate and use a condition on the phi angle to get the correct sign of the magnet polarisation -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jul 16, 2012, 8:35 p.m. EDT
Thank you for your response Ivar,

Because I am working in eddy current anaysis so I must use MEF.
As, you have said, it is possible to set up a cylindrical coordinate and use a condition in the phi angle for the magnets's polarisation. It works for me, but the only problem is that the streamlines of the magnetic flux density are not totally guided by the Yoke? (see fig1 and fig2).
Thank you again Ivar.

Regards.
Thank you for your response Ivar, Because I am working in eddy current anaysis so I must use MEF. As, you have said, it is possible to set up a cylindrical coordinate and use a condition in the phi angle for the magnets's polarisation. It works for me, but the only problem is that the streamlines of the magnetic flux density are not totally guided by the Yoke? (see fig1 and fig2). Thank you again Ivar. Regards.


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 17, 2012, 1:33 a.m. EDT
Hi

you will always get some field outside your iron, check the streamline settings and use another rule for magnitude, in your image it might be that 90% of the field is still in the yoke

--
Good luck
Ivar
Hi you will always get some field outside your iron, check the streamline settings and use another rule for magnitude, in your image it might be that 90% of the field is still in the yoke -- 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.