Menu

Discussion Forum

New Discussion
Filter

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.

Magnetic field around subsea cable

Posted Mar 31, 2017, 7:00 a.m. EDT 6 Replies

Aksel Kjetså
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Hi!

I am trying to model the magnetic field around an AC subsea cable. I've created a 2d model in the time domain consisting of a copper conductor, some insulation and a lead sheath.

The model will solve if i use the homogenized multi-turn mode. But if i use single conductor i get the error message:

"Nonlinear solver did not converge.
Maximum number of Newton iterations reached.
There was an error message from the linear solver.
The relative residual (0.17) is greater than the relative tolerance.
Time : 0
Last time step is not converged.
- Feature: Time-Dependent Solver 1 (sol1/t1)"

Any ideas why this might be? File attached

Attachments:

6 Replies Last Post Apr 3, 2017, 5:33 a.m. EDT
Edgar J. Kaiser Certified Consultant
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 8 years ago Mar 31, 2017, 9:25 a.m. EDT
Hi Aksel,

I played a little with your model and I am also surprised that the single conductor doesn't work. It does work when you change to voltage excitation. May be some Comsol support person could clarify this.

As your model seems to be all linear you could try frequency domain.

Cheers
Edgar

--
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Hi Aksel, I played a little with your model and I am also surprised that the single conductor doesn't work. It does work when you change to voltage excitation. May be some Comsol support person could clarify this. As your model seems to be all linear you could try frequency domain. Cheers Edgar -- Edgar J. Kaiser emPhys Physical Technology http://www.emphys.com
Aksel Kjetså
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 8 years ago Mar 31, 2017, 9:33 a.m. EDT
Thank you for looking into it!

I am using the frequency domain now. As it turns out the homogenized multi-turn is the better option as the conductor i'm modeling is in reality 91 small wires.

Cheers
Aksel
Thank you for looking into it! I am using the frequency domain now. As it turns out the homogenized multi-turn is the better option as the conductor i'm modeling is in reality 91 small wires. Cheers Aksel
Edgar J. Kaiser Certified Consultant
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 8 years ago Mar 31, 2017, 10:05 a.m. EDT
only if the 91 wires are insulated against each other!

--
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
only if the 91 wires are insulated against each other! -- Edgar J. Kaiser emPhys Physical Technology http://www.emphys.com
Aksel Kjetså
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 8 years ago Mar 31, 2017, 10:39 a.m. EDT
I did not know that! Thanks
I did not know that! Thanks
Edgar J. Kaiser Certified Consultant
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 8 years ago Mar 31, 2017, 12:10 p.m. EDT

Ok, fine. But still the initial issue is a little mysterious and I would encourage you to challenge the Comsol support. And please keep us updated if they find something.

Cheers
Edgar

--
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Ok, fine. But still the initial issue is a little mysterious and I would encourage you to challenge the Comsol support. And please keep us updated if they find something. Cheers Edgar -- Edgar J. Kaiser emPhys Physical Technology http://www.emphys.com
Aksel Kjetså
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 8 years ago Apr 3, 2017, 5:33 a.m. EDT
I got an answer now:

Dear Aksel,

The difference between the conductivity values for various region is of the order of 10e17. I would suggest you to try selecting the order of the conductivity difference that is less then the machine precision.

For acrylic, you could use a very low value (e.g. 1e-6[S/m], and for this you would need to increase the iteration number to say 25 under fully coupled solver node). However, the computation is much faster with 0.1[S/m] and the result difference compared with 1e-6[S/m] is within the relative tolerance.

Please see the image attached herewith this reply.

Hope this helps. Let us know if you need any further information on this.

Kind regards,
Prajakta Sabnis
COMSOL Support
I got an answer now: Dear Aksel, The difference between the conductivity values for various region is of the order of 10e17. I would suggest you to try selecting the order of the conductivity difference that is less then the machine precision. For acrylic, you could use a very low value (e.g. 1e-6[S/m], and for this you would need to increase the iteration number to say 25 under fully coupled solver node). However, the computation is much faster with 0.1[S/m] and the result difference compared with 1e-6[S/m] is within the relative tolerance. Please see the image attached herewith this reply. Hope this helps. Let us know if you need any further information on this. Kind regards, Prajakta Sabnis COMSOL Support

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.

Suggested Content