Menu

Discussion Forum

New Discussion
Filtra

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.

Single-turn coil gap feed problem

Posted 26 gen 2015, 16:35 GMT-5 Low-Frequency Electromagnetics, Studies & Solvers Version 5.0 11 Replies

TaeKen Kim
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Hello,

I am trying to model gradient coils for use in MRI. I created a saddle coil geometry and I tried to run a study with the current direction specified by gap feed, but it fails to find a solution. I've attached my model to the thread, I would appreciate any insight into this.

Thanks in advance!

Attachments:

11 Replies Last Post 14 mar 2015, 17:32 GMT-4
Edgar J. Kaiser Certified Consultant
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 26 gen 2015, 17:21 GMT-5

Hi,

you have to close the current path some way inside the modeling volume, or you have to extend the coil to the outer boundaries. Terminals and ports are an option too.
Current cannot just appear and disappear inside the model.

Cheers
Edgar

--
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Hi, you have to close the current path some way inside the modeling volume, or you have to extend the coil to the outer boundaries. Terminals and ports are an option too. Current cannot just appear and disappear inside the model. Cheers Edgar -- Edgar J. Kaiser emPhys Physical Technology http://www.emphys.com
TaeKen Kim
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2 feb 2015, 11:37 GMT-5

Hi,

you have to close the current path some way inside the modeling volume, or you have to extend the coil to the outer boundaries. Terminals and ports are an option too.
Current cannot just appear and disappear inside the model.

Cheers
Edgar

--
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com


Thanks for the reply. But I am looking at single_coil_gap_feed.mph at www.comsol.com/model/single-turn-and-multi-turn-coil-domains-in-3d-14067, and this model works without doing things you prescribed. My saddle coil is just a loop like this torus, so intuitively, I expect same modeling technique to work for both geometries. What am I missing?
[QUOTE] Hi, you have to close the current path some way inside the modeling volume, or you have to extend the coil to the outer boundaries. Terminals and ports are an option too. Current cannot just appear and disappear inside the model. Cheers Edgar -- Edgar J. Kaiser emPhys Physical Technology http://www.emphys.com [/QUOTE] Thanks for the reply. But I am looking at single_coil_gap_feed.mph at http://www.comsol.com/model/single-turn-and-multi-turn-coil-domains-in-3d-14067, and this model works without doing things you prescribed. My saddle coil is just a loop like this torus, so intuitively, I expect same modeling technique to work for both geometries. What am I missing?
Edgar J. Kaiser Certified Consultant
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2 feb 2015, 16:11 GMT-5
You may want to have another look into the example you mentioned. While the torus provides a perfect closed loop for the current I really cannot see which way your coil definition could be a closed loop.

The way I understand a saddle coil is a current that flows along the edges of the saddle or in a domain close to the edges. Something like this: www.ewh.ieee.org/tc/csc/News/Image12.jpg
In such a coil you can add a gap feed similar to the torus example. You may want to study the COMSOL documentation about coil domains and the way to excite them.

Cheers
Edgar

--
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
You may want to have another look into the example you mentioned. While the torus provides a perfect closed loop for the current I really cannot see which way your coil definition could be a closed loop. The way I understand a saddle coil is a current that flows along the edges of the saddle or in a domain close to the edges. Something like this: http://www.ewh.ieee.org/tc/csc/News/Image12.jpg In such a coil you can add a gap feed similar to the torus example. You may want to study the COMSOL documentation about coil domains and the way to excite them. Cheers Edgar -- Edgar J. Kaiser emPhys Physical Technology http://www.emphys.com
TaeKen Kim
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 5 feb 2015, 09:28 GMT-5
Here is a screenshot of the domains I selected for the coil in my model, and I can't figure out why I can't use gap feed to specify current direction.
Here is a screenshot of the domains I selected for the coil in my model, and I can't figure out why I can't use gap feed to specify current direction.

Attachments:

Edgar J. Kaiser Certified Consultant
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 5 feb 2015, 10:33 GMT-5
In the model you attached in the first post, the whole saddle is selected and the gap feed extends all across the saddle, which doesn't make sense.
I corrected your model and now it solves. Two more remarks. In mf the non-conductive domains mostly need a small conductivity and you have to add a 'Gauge Fixing for A-field' node to achieve convergence.
Also I would advise to assign a material or realistic conductivity to the coil domain to get meaningful coil parameters in the post processing.
I am attaching the rectified model.

Cheers
Edgar

--
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
In the model you attached in the first post, the whole saddle is selected and the gap feed extends all across the saddle, which doesn't make sense. I corrected your model and now it solves. Two more remarks. In mf the non-conductive domains mostly need a small conductivity and you have to add a 'Gauge Fixing for A-field' node to achieve convergence. Also I would advise to assign a material or realistic conductivity to the coil domain to get meaningful coil parameters in the post processing. I am attaching the rectified model. Cheers Edgar -- Edgar J. Kaiser emPhys Physical Technology http://www.emphys.com

Attachments:

TaeKen Kim
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 6 feb 2015, 16:24 GMT-5
Thanks for all your help, I didn't even think about Gauge Fixing for A-field. Unfortunately, my institution only has license to Comsol 4.4 and earlier so I can't open the file. Would it be possible for you to post the fixed model compatible with earlier versions of comsol? Thank you once again.
Thanks for all your help, I didn't even think about Gauge Fixing for A-field. Unfortunately, my institution only has license to Comsol 4.4 and earlier so I can't open the file. Would it be possible for you to post the fixed model compatible with earlier versions of comsol? Thank you once again.
Edgar J. Kaiser Certified Consultant
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 6 feb 2015, 18:06 GMT-5

Here we go. Enjoy!

--
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Here we go. Enjoy! -- Edgar J. Kaiser emPhys Physical Technology http://www.emphys.com

Attachments:

Francesco Coppari
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 13 mar 2015, 18:41 GMT-4
Actually I am trying to design a model of gradient coils for use in MRI. I really need to know how i can design a saddle coil geometry and run the study to get the magnetic flux .
thanks in advance
Actually I am trying to design a model of gradient coils for use in MRI. I really need to know how i can design a saddle coil geometry and run the study to get the magnetic flux . thanks in advance
Edgar J. Kaiser Certified Consultant
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 14 mar 2015, 04:50 GMT-4

Well, that's what the thread is about. Have you checked the postings above?

--
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Well, that's what the thread is about. Have you checked the postings above? -- Edgar J. Kaiser emPhys Physical Technology http://www.emphys.com
Francesco Coppari
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 14 mar 2015, 16:00 GMT-4
Well,i checked it , you really did a great job , actually i am beginner and i trying to design a model like the attachment photo , and how i can change the current in the coil to the opposite direction .
Best regards
Well,i checked it , you really did a great job , actually i am beginner and i trying to design a model like the attachment photo , and how i can change the current in the coil to the opposite direction . Best regards

Attachments:

Edgar J. Kaiser Certified Consultant
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 14 mar 2015, 17:32 GMT-4

Multiply with -1 ? This will also change the sign of the flux, nothing more.
--
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Multiply with -1 ? This will also change the sign of the flux, nothing more. -- Edgar J. Kaiser emPhys Physical Technology http://www.emphys.com

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