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.

Force and deformation caused by magnetic field?

Please login with a confirmed email address before reporting spam

Hello,

Using the Magnetic Field module and a time-dependent external current density (defined by ec.Jx, ec.Jy and ec.Jz), I simulated the magnetic field caused on a metal sheet by a 3D coil. Now I want to calculate the forces and deformations generated by this magnetic field on the metal sheet. I am using COMSOL V4.1 and I suppose I need Solid Mechanics for this.

However, adding the Solid Mechanics Physics for some reason makes the external current density "unknown" (orange letters for ec.Jx, ec.Jy and ec.Jz). Is some kind of coupling needed between the modules? Otherwise, could anybody explain to me what steps to take to calculate the deformation of the metal sheet from the (known) results of the magnetic field?

There does not seem to be an issue when I define the current solely by the external current density in Magnetic Fields. However, my coil geometry makes this rather complicated, so I defined potentials on the coil instead with Electric Currents and used the resulting current as external current density in Magnetic Fields, which seems to cause issues.


So my main question is: how do I calculate the deformation of the metal sheet when the magnetic field is known?

6 Replies Last Post Oct 20, 2011, 9:08 a.m. EDT

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 14, 2011, 1:54 a.m. EDT
Hello!

Please look carefully for the names of values: in my case the correct name is ec.JX, ecJY, ec.JZ with capital letters.
If you activate equation view, you can find the names for all values.

Ralf
Hello! Please look carefully for the names of values: in my case the correct name is ec.JX, ecJY, ec.JZ with capital letters. If you activate equation view, you can find the names for all values. Ralf

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 17, 2011, 5:47 a.m. EDT
Hi,

Can you attach the model for better understanding?

Thanks!

Jerry
Hi, Can you attach the model for better understanding? Thanks! Jerry

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 19, 2011, 10:03 a.m. EDT
Hi,

Sorry for responding so late. As can be seen in mf, ec.Jx, ec.Jy and ec.Jz were used for the external current density. I did this because I wanted to input electric potentials in the model and I thought the resulting current needed to be linked to mf using external current density.

As far as I know, the directions of the arrows seem to be as expected. The current induced on the metal sheet is directed opposite to the current in the coil. The shape of the Arrow Volume of the magnetic field caused by the coil also seems to be as expected. Up to this point, the letters of ec.Jx, ec.Jy and ec.Jz in the external current density were black.

However, I added Solid Mechanics afterwards to the model. The letters in the external current density immediately turned orange. Even when the Solid Mechanics Module is disabled, those letters remain orange. Therefore, any subsequent attempts at running the model will fail, even when the last module is disabled. Deleting the extra module returns the orange letters to black.

So, can anyone explain to me why the mere existence of Solid Mechanics has this effect on the external current density in mf? Other modules (such as Beam or even something unrelated as Laminar Flow) do not seem to have this effect. Am I missing some kind of coupling? I am not yet very familiar with COMSOL.


Thanks in advance.

P.S. I had to clear the solutions to make the file small enough for the attachment. Therefore, I made a screenshot of the arrows showing the J and H from mf.
Hi, Sorry for responding so late. As can be seen in mf, ec.Jx, ec.Jy and ec.Jz were used for the external current density. I did this because I wanted to input electric potentials in the model and I thought the resulting current needed to be linked to mf using external current density. As far as I know, the directions of the arrows seem to be as expected. The current induced on the metal sheet is directed opposite to the current in the coil. The shape of the Arrow Volume of the magnetic field caused by the coil also seems to be as expected. Up to this point, the letters of ec.Jx, ec.Jy and ec.Jz in the external current density were black. However, I added Solid Mechanics afterwards to the model. The letters in the external current density immediately turned orange. Even when the Solid Mechanics Module is disabled, those letters remain orange. Therefore, any subsequent attempts at running the model will fail, even when the last module is disabled. Deleting the extra module returns the orange letters to black. So, can anyone explain to me why the mere existence of Solid Mechanics has this effect on the external current density in mf? Other modules (such as Beam or even something unrelated as Laminar Flow) do not seem to have this effect. Am I missing some kind of coupling? I am not yet very familiar with COMSOL. Thanks in advance. P.S. I had to clear the solutions to make the file small enough for the attachment. Therefore, I made a screenshot of the arrows showing the J and H from mf.


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 19, 2011, 10:28 a.m. EDT
Hi

Pls note the difference between orange and red:
- orange is unit issues it's a warning,
- red is typing error hence failure to run,
while orange will run and give results, that might be correct too ;)

But if it happened when you added a new physics, it gives me the impression that you had hit a COMSOL internal name, and that the new physics has changed the units of your already defined variable, hence unit error, BUT then certainly also solver failure as you are mixing into a same variable 2 distinct physical meanings. This is only a guess, but it's my first idea when I read your issue (I do not have COMSOL accessible just now so I cannot look at the files)

--
Good luck
Ivar
Hi Pls note the difference between orange and red: - orange is unit issues it's a warning, - red is typing error hence failure to run, while orange will run and give results, that might be correct too ;) But if it happened when you added a new physics, it gives me the impression that you had hit a COMSOL internal name, and that the new physics has changed the units of your already defined variable, hence unit error, BUT then certainly also solver failure as you are mixing into a same variable 2 distinct physical meanings. This is only a guess, but it's my first idea when I read your issue (I do not have COMSOL accessible just now so I cannot look at the files) -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 20, 2011, 8:48 a.m. EDT
Hi!

In 3.5 there was a similar effect: variable names changes from small letters to capital letters.
If you change from ec.Jx to ec.JX, it will work.
Another problem is, that according to support this coupling of EC and MF includes not the complete equation; there is eps0*epsr*Att missing for transient case...
That's why the calculated EM-forces may be wrong for your solid deformation.

Ralf
Hi! In 3.5 there was a similar effect: variable names changes from small letters to capital letters. If you change from ec.Jx to ec.JX, it will work. Another problem is, that according to support this coupling of EC and MF includes not the complete equation; there is eps0*epsr*Att missing for transient case... That's why the calculated EM-forces may be wrong for your solid deformation. Ralf


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 20, 2011, 9:08 a.m. EDT
...wrong second image...


...wrong second image...

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.