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.

Induction Heating System and AC/DC Modules for Coil Simulation

Please login with a confirmed email address before reporting spam

Hello,
I am trying to simulate a 2d Asymetric Induction Heating system with coil and iron core setup.

I use AC/DC module to calculate de Magnetic Fields and know the correct amount of Current excitation and the Induction Heating module to understand the heat transfer at the core of the coil.

The error it gives is about Stationary Study which is strange because I am only doing a Frequency Domain study.

Thanks for your help!


1 Reply Last Post 26 sept. 2010, 14:36 UTC−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 26 sept. 2010, 14:36 UTC−4
Hi

first of all, check your material data, always recheck once you have finished the model and meshed it, as some of the demands for variables are really updated rather late and the "red flags" might appear after you have passed over the material properties. This comes from different selections you can do in the physics, that rather recalll some or other material data properties.

Then in the meshing you have defined some boundary layers, but these have not executed as the sub-node properties has not been set to apply to any border (choose the external borders, not the axis, but I'm not sure this is 100% needed, certainly not to start with, perhaps once you have the model up running clearly and you see where the skin effects might be.

Still I get an error, seems to be in some material definition, I suspect the mur of the iron. I prefer always to start with simple data, not the complex non-linear material data from the DB, as these couple several parameters, and introduces often non-linearities. I havent seen those error messages before ;)

Finally you solve both physics together, I'm not sure this is the best way, often using a segregated sover (sequence one physics after the other) improves convergence, but for that you must find out which physics is the "master" and which is the "slave" (if truely possible).
The confusion of solvers can also come from this, as an eigenfrequency often need to have a startionary case solved first to define the parameters to linearise around for the frequency analysis, obviously you are not getting that far.

Have you tried with only one physics at a time ?
this helps often in such complex multiphysics cases


--
Good luck
Ivar
Hi first of all, check your material data, always recheck once you have finished the model and meshed it, as some of the demands for variables are really updated rather late and the "red flags" might appear after you have passed over the material properties. This comes from different selections you can do in the physics, that rather recalll some or other material data properties. Then in the meshing you have defined some boundary layers, but these have not executed as the sub-node properties has not been set to apply to any border (choose the external borders, not the axis, but I'm not sure this is 100% needed, certainly not to start with, perhaps once you have the model up running clearly and you see where the skin effects might be. Still I get an error, seems to be in some material definition, I suspect the mur of the iron. I prefer always to start with simple data, not the complex non-linear material data from the DB, as these couple several parameters, and introduces often non-linearities. I havent seen those error messages before ;) Finally you solve both physics together, I'm not sure this is the best way, often using a segregated sover (sequence one physics after the other) improves convergence, but for that you must find out which physics is the "master" and which is the "slave" (if truely possible). The confusion of solvers can also come from this, as an eigenfrequency often need to have a startionary case solved first to define the parameters to linearise around for the frequency analysis, obviously you are not getting that far. Have you tried with only one physics at a time ? this helps often in such complex multiphysics cases -- 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.