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.

Segregated? Or Fully coupled?

Please login with a confirmed email address before reporting spam

Hi, all,

I am trying to simulate S-parameters of my structure in RF module and I choose frequency domain. When I use segregated solver, there are always errors:

Failed to find a solution.
Maximum numbers of segregated iterations reached.
Returned solution is not converged.

Errors insist even after I increase maximum numbers of segregated iterations.
Then I use Fully coupled solver to simulate and there is no error. But I am not sure whether the results are correct or not. So if I have more than one variables in my model, is it a necessity to use segregated solver?

Thanks a lot for any help.


2 Replies Last Post 8 feb 2016, 15:10 GMT-5

Please login with a confirmed email address before reporting spam

Posted: 9 years ago 8 feb 2016, 14:49 GMT-5
Hi Tianjing,

Having the same problem with the electric + acoustic module...

Did you find a solution to the problem? How do you choose segregated or fully coupled models?

Best regards
Hi Tianjing, Having the same problem with the electric + acoustic module... Did you find a solution to the problem? How do you choose segregated or fully coupled models? Best regards

Please login with a confirmed email address before reporting spam

Posted: 9 years ago 8 feb 2016, 15:10 GMT-5
Hi, Tobias,

For my case, when I compute in different frequency range (low frequency), this error will disappear. Then I try to compare the results with segregated and fully coupled solver in low frequency and find they are identical. So I just use fully coupled solver to simulate my structure in a wide frequency range. I am not sure if it's right.

Maybe you can try to change maximum numbers of segregated iterations or tolerance factor to remove errors.

Good luck.
Hi, Tobias, For my case, when I compute in different frequency range (low frequency), this error will disappear. Then I try to compare the results with segregated and fully coupled solver in low frequency and find they are identical. So I just use fully coupled solver to simulate my structure in a wide frequency range. I am not sure if it's right. Maybe you can try to change maximum numbers of segregated iterations or tolerance factor to remove errors. Good luck.

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.