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 The information provided may be out of date.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.

Saturation Current for a Power Inductor

Please login with a confirmed email address before reporting spam

Hi,
I am simulating a power inductor, but the program give me the simulation error "undefined value found"(with a extremely coarse mesh) or "out of memory during assembly" (with a coars mesh). Someone can give me some help??? Where is my error???
Raffaele


6 Replies Last Post 9 févr. 2011, 08:20 UTC−5
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 5 févr. 2011, 10:39 UTC−5
Hi

there are a few points in your model, among other you define a GND and a Voltage source internal boundary, but how can you write the Maxwell equations in a closed volume with suddenly you have an appearing/dissapearing current source ? and no loop ?

the best is probably to loop your coil all the way to the external boundary and then use the external surface/boundary as current closing loop.

Next is rather a remarque, the magnetic and electric field will tend to a spherical symmetry around your model, So I would have used a sphere for the air instead of a cube, as then your B field loops around nices without any "corner effects"


fnal comment, try to avoid right angles shapes for the domins, as a right angle corner is a singularity for MEF A part, you have rounded somecorners I see but not all, this is also of less importance, except if you are studying the filed in details

--
Good luck
Ivar
Hi there are a few points in your model, among other you define a GND and a Voltage source internal boundary, but how can you write the Maxwell equations in a closed volume with suddenly you have an appearing/dissapearing current source ? and no loop ? the best is probably to loop your coil all the way to the external boundary and then use the external surface/boundary as current closing loop. Next is rather a remarque, the magnetic and electric field will tend to a spherical symmetry around your model, So I would have used a sphere for the air instead of a cube, as then your B field loops around nices without any "corner effects" fnal comment, try to avoid right angles shapes for the domins, as a right angle corner is a singularity for MEF A part, you have rounded somecorners I see but not all, this is also of less importance, except if you are studying the filed in details -- Good luck Ivar

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 5 févr. 2011, 10:39 UTC−5
Hi

there are a few points in your model, among other you define a GND and a Voltage source internal boundary, but how can you write the Maxwell equations in a closed volume with suddenly you have an appearing/dissapearing current source ? and no loop ?

the best is probably to loop your coil all the way to the external boundary and then use the external surface/boundary as current closing loop.

Next is rather a remarque, the magnetic and electric field will tend to a spherical symmetry around your model, So I would have used a sphere for the air instead of a cube, as then your B field loops around nices without any "corner effects"


fnal comment, try to avoid right angles shapes for the domins, as a right angle corner is a singularity for MEF A part, you have rounded somecorners I see but not all, this is also of less importance, except if you are studying the filed in details

--
Good luck
Ivar
Hi there are a few points in your model, among other you define a GND and a Voltage source internal boundary, but how can you write the Maxwell equations in a closed volume with suddenly you have an appearing/dissapearing current source ? and no loop ? the best is probably to loop your coil all the way to the external boundary and then use the external surface/boundary as current closing loop. Next is rather a remarque, the magnetic and electric field will tend to a spherical symmetry around your model, So I would have used a sphere for the air instead of a cube, as then your B field loops around nices without any "corner effects" fnal comment, try to avoid right angles shapes for the domins, as a right angle corner is a singularity for MEF A part, you have rounded somecorners I see but not all, this is also of less importance, except if you are studying the filed in details -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 8 févr. 2011, 16:14 UTC−5
Thank you Ivar,
when we can get married? :D
Seriously your help saved my life, now i can simulate the inductor, i have rounded the corners, i can now see the B field...the real problem is the too big mesh, i need 26 turns in my inductor and with simulator i obtained "out of memory" with more of 3 turns!!!!
do you think is a good idea to work in 2D?I decided to draw the inductor as in the attached figure. But how to bring my results in 3d?


I have also used a BH curve for ferrite core material...now i am trying to find the saturation current (at least with 3 turns)
Thank you Ivar, when we can get married? :D Seriously your help saved my life, now i can simulate the inductor, i have rounded the corners, i can now see the B field...the real problem is the too big mesh, i need 26 turns in my inductor and with simulator i obtained "out of memory" with more of 3 turns!!!! do you think is a good idea to work in 2D?I decided to draw the inductor as in the attached figure. But how to bring my results in 3d? I have also used a BH curve for ferrite core material...now i am trying to find the saturation current (at least with 3 turns)


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 9 févr. 2011, 01:39 UTC−5
Hi

if you can work in 2D (i.e. a "simple" 2d-axi) its always much better. For multiturn coils you have the new BC's in V4.1, really nice, ut you must read carefully the doc, and look at the equations, these are subtil BC's adding new dependent variables to solve localy for V. But once you master them they will make life far easier for you

Concerning marriage, I'm already far engaged, with grown up kids, so I'm certainly not interesting, far better as physics teacher on the web ;)

--
Good luck
Ivar
Hi if you can work in 2D (i.e. a "simple" 2d-axi) its always much better. For multiturn coils you have the new BC's in V4.1, really nice, ut you must read carefully the doc, and look at the equations, these are subtil BC's adding new dependent variables to solve localy for V. But once you master them they will make life far easier for you Concerning marriage, I'm already far engaged, with grown up kids, so I'm certainly not interesting, far better as physics teacher on the web ;) -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 9 févr. 2011, 05:47 UTC−5
Hi Ivar,
i have only the 4.0a! It's the same? Or i need the 4.1?
Hi Ivar, i have only the 4.0a! It's the same? Or i need the 4.1?

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 9 févr. 2011, 08:20 UTC−5
Hi

v4.0 was the first issue of the new release, as with any software the "improvement curve" is strongest at the beginning ofa new version, that is why I believe one should never really work with a ".0" version, just use it to prepare onself to the new interface, and switch asap to the next release, today it's 4.1 + patch, and soon (a few months) I blieve the 4.2 will arive. With COMSOL as any with other heavy tool, it is really also worth to keep the maintenance, and I find their preices correct for the tool quality

--
Good luck
Ivar
Hi v4.0 was the first issue of the new release, as with any software the "improvement curve" is strongest at the beginning ofa new version, that is why I believe one should never really work with a ".0" version, just use it to prepare onself to the new interface, and switch asap to the next release, today it's 4.1 + patch, and soon (a few months) I blieve the 4.2 will arive. With COMSOL as any with other heavy tool, it is really also worth to keep the maintenance, and I find their preices correct for the tool quality -- 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.