Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

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.

I have a singular matrix error on MEF simulation

Please login with a confirmed email address before reporting spam

I tried to make helmholtz coil. However, I have a problem of singular matrix.
What is the problem in my simulation?


5 Replies Last Post 26 apr 2013, 10:58 GMT-4
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 19 gen 2012, 04:23 GMT-5
Hi

to check your mesh, create a second view (Definition View) and add a sub node "Hide Geometrical entity" and select the air entity "1" then click the rebuild, I would start meshing your copper in a finer way, then mesh the air as very coarse

Ideally you should use a round cable/wires, as on the sharp edges you have singularities, but it also works more or less like that

Then I would not "ground the full surface but just the second wire connector, or at most the lower plate

Finally for MEF to work, you must have some conductance in all materials otherwise you cannot resolve "V", try 1[S/m] for air it should do

--
Good luck
Ivar
Hi to check your mesh, create a second view (Definition View) and add a sub node "Hide Geometrical entity" and select the air entity "1" then click the rebuild, I would start meshing your copper in a finer way, then mesh the air as very coarse Ideally you should use a round cable/wires, as on the sharp edges you have singularities, but it also works more or less like that Then I would not "ground the full surface but just the second wire connector, or at most the lower plate Finally for MEF to work, you must have some conductance in all materials otherwise you cannot resolve "V", try 1[S/m] for air it should do -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 27 gen 2012, 18:10 GMT-5
Thanks for your reply.
I will try that.
Thanks for your reply. I will try that.

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 27 gen 2012, 18:14 GMT-5
Hi

and also, in MEF, do not forget all materials must have some conductivity (even 1S/m sould do)

--
Good luck
Ivar
Hi and also, in MEF, do not forget all materials must have some conductivity (even 1S/m sould do) -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 26 apr 2013, 10:35 GMT-4
Thank you for pointing me into the right direction!

It took me almost half of a work day just to find out that the "singular matrix error" is solely related to some materials in my setup having zero conductivity...

How am I supposed to know that materials inserted from the Comsol Material Library have to be _altered_ in order to make the simulation work in some cases!?

Why doesn't Comsol simply _validate_ the materials used in a simulation in order to give _meaningful_ error message?

A check for conductivity=0 when the mef module is used shouldn't be too complicated to be implemented....
Thank you for pointing me into the right direction! It took me almost half of a work day just to find out that the "singular matrix error" is solely related to some materials in my setup having zero conductivity... How am I supposed to know that materials inserted from the Comsol Material Library have to be _altered_ in order to make the simulation work in some cases!? Why doesn't Comsol simply _validate_ the materials used in a simulation in order to give _meaningful_ error message? A check for conductivity=0 when the mef module is used shouldn't be too complicated to be implemented....

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 26 apr 2013, 10:58 GMT-4
Hi

its always our responsibility to check what we are doing. Anyhow this is a peculiarity of MEF, in MF a s=0 is OK. In fact in MEF too if you choose a "Domain Ampere law" node for the air.

It all depends on what you are modelling, and which hypothesis you need for a given physics, so there is no 1:1 way here

That is also why the training courses, or even the mini courses at the COMSOL conference are that useful, as they help you identify such simple traps, still remains the more subtle traps ;)

--
Good luck
Ivar
Hi its always our responsibility to check what we are doing. Anyhow this is a peculiarity of MEF, in MF a s=0 is OK. In fact in MEF too if you choose a "Domain Ampere law" node for the air. It all depends on what you are modelling, and which hypothesis you need for a given physics, so there is no 1:1 way here That is also why the training courses, or even the mini courses at the COMSOL conference are that useful, as they help you identify such simple traps, still remains the more subtle traps ;) -- Good luck Ivar

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.