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.

Magnetoelectric device in frequency domain

Please login with a confirmed email address before reporting spam

Hello everyone, I am trying to simulate a magnetoelectric device (with sandwich structure of magnetostricitive and piezoelectric materials) under the AC magnetic field. For that, I have made a Helmholtz coil which is driven by external electrical circuit of AC voltage. But the problem is that after simulation an error massage like "NaN or Inf found when solving linear system using SOR". The simulation absolutely fine with stationary domain. But I could not identified what is going wrong with frequency domain. I have attached the simulation file.

Can anyone help me in this regard? It will be a great help for me.

Looking forward for your kind reply. Thanking you Sincerely Sujoy



9 Replies Last Post May 20, 2019, 8:03 a.m. EDT
Edgar J. Kaiser Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 5 years ago Mar 12, 2019, 6:44 a.m. EDT

Sujoy,

for the dynamic analysis all domains need to have a small electrical conductivity in mf, say 1 S/m. Maybe it can be even lower, you need to try.

Good luck Edgar

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Sujoy, for the dynamic analysis all domains need to have a small electrical conductivity in mf, say 1 S/m. Maybe it can be even lower, you need to try. Good luck Edgar

Please login with a confirmed email address before reporting spam

Posted: 5 years ago Mar 13, 2019, 7:46 a.m. EDT

Hello Edger, I have tried with small conductivity but still not running the simulation. Could you please take a look to my simulation file which is attached ? Thanking you Sincerely Sujoy

Hello Edger, I have tried with small conductivity but still not running the simulation. Could you please take a look to my simulation file which is attached ? Thanking you Sincerely Sujoy

Edgar J. Kaiser Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 5 years ago Mar 13, 2019, 7:51 a.m. EDT

I did test it. I applied 1 S/m to air and to the piezoelectric domain and it solved.

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
I did test it. I applied 1 S/m to air and to the piezoelectric domain and it solved.

Please login with a confirmed email address before reporting spam

Posted: 5 years ago Mar 14, 2019, 2:02 a.m. EDT

Hello Edger, Now it works. Its a great help for me. I am absolutly thankful to you. Though, simulation is taken much more time. Is there any way to redue it?

                 Looking forward for your kind reply.
                 Thanking you
                 Sujoy
Hello Edger, Now it works. Its a great help for me. I am absolutly thankful to you. Though, simulation is taken much more time. Is there any way to redue it? Looking forward for your kind reply. Thanking you Sujoy

Edgar J. Kaiser Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 5 years ago Mar 14, 2019, 5:36 a.m. EDT
Updated: 5 years ago Mar 14, 2019, 5:36 a.m. EDT

There may be a couple of ways: * Reduce the size of the model

  • Find out if the mesh can be coarser

  • Use lower order discretization

  • Try different direct and iterative solvers

  • ...

Optimization can be tedious.

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
There may be a couple of ways: * Reduce the size of the model * Find out if the mesh can be coarser * Use lower order discretization * Try different direct and iterative solvers * ... Optimization can be tedious.

Please login with a confirmed email address before reporting spam

Posted: 5 years ago Mar 14, 2019, 6:00 a.m. EDT

Thanks a lot Edger.

Thanks a lot Edger.

Please login with a confirmed email address before reporting spam

Posted: 5 years ago Apr 3, 2019, 9:24 a.m. EDT

Hello Edgar, Hope everything is going well. Currently, I am facing a problem regarding the aforementioned problem. Though, it was solved for lower frequency region but when I am trying to solve with higher frequency region such as, 10 kHz, it is not solving and returned the following massages: Maximum number of segregated iterations reached. Returned solution is not converged. Not all parameter steps returned.

Could you please help me in this regard? Thanking you Sincerely Sujoy

Hello Edgar, Hope everything is going well. Currently, I am facing a problem regarding the aforementioned problem. Though, it was solved for lower frequency region but when I am trying to solve with higher frequency region such as, 10 kHz, it is not solving and returned the following massages: Maximum number of segregated iterations reached. Returned solution is not converged. Not all parameter steps returned. Could you please help me in this regard? Thanking you Sincerely Sujoy

Please login with a confirmed email address before reporting spam

Posted: 5 years ago May 20, 2019, 7:53 a.m. EDT

Hello Sujoy, Have you been able to sort out the above problem of not being able to simulate in higher frequency range...??

Hello Sujoy, Have you been able to sort out the above problem of not being able to simulate in higher frequency range...??

Please login with a confirmed email address before reporting spam

Posted: 5 years ago May 20, 2019, 8:03 a.m. EDT

Hello Sir/Madam, No, till now it is not solved.

Hello Sir/Madam, No, till now it is not solved.

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.