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.
Inductive Heating Modeling in Comsol 4.0a
Posted 15 nov. 2010, 12:28 UTC−5 4 Replies
Please login with a confirmed email address before reporting spam
Dear All
In design of inductive heating system for copper stretching machine to quickly warm it up to certain temperature.
The copper will come through the coil and get heated up, as we all know, the question is with what current and at what frequency the current should be applied, given a certain geometry.
Is there any one who has made a simple three dimensional eddy current simulation?
I am trying magnetic and electric fields along with inductive heating in the same model to be able to provide magnetic flux potential as an input for the inductive heating model.
I am not yet getting converged and I am hugely under pressure to get the job done.
Is there any one who is willing to help me out from the ditch.
The model is attached for your kind reviews.
Reza Besharati Tabrizi
Reza Besharati Tabrizi
In design of inductive heating system for copper stretching machine to quickly warm it up to certain temperature.
The copper will come through the coil and get heated up, as we all know, the question is with what current and at what frequency the current should be applied, given a certain geometry.
Is there any one who has made a simple three dimensional eddy current simulation?
I am trying magnetic and electric fields along with inductive heating in the same model to be able to provide magnetic flux potential as an input for the inductive heating model.
I am not yet getting converged and I am hugely under pressure to get the job done.
Is there any one who is willing to help me out from the ditch.
The model is attached for your kind reviews.
Reza Besharati Tabrizi
Reza Besharati Tabrizi
4 Replies Last Post 1 avr. 2011, 13:08 UTC−4
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
If your model does not converge, why don't you start with a homogenous coil in 3D or even better, an axisymmetric homogenous in 2D. Also, do you know as to what temperature range are you are looking for ?
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Dear Venkat
I switched to 2D and it is quickly converging.
The temperature range is around 1200 K.
The application is to prepare the blocks of AISI 403Cb blocks of stainless steel for hot forging.
Now to switch back to 3D I will be using external current source since now I do not have the Multi-Coil domain in 3D anymore.
The frequencies are quite low and they will not reach 10 kHz.
--
Reza Besharati Tabrizi
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
first a warning inductive and eddy current models in 3D are rather tricky and require a powerfull PC to get solved, so you are using the best approach, start simple.
If you run the "MEF" physics then set some conduction to the air 1-1000 S/m, try a few values and lower until not solving anymore, then go back a step, this mostly helps the convergence, (check the air dissipation must still stay negligible if you want to understand your energy balance, as normally air dissipates 0W electrically
At 10kHz or below you have small skin effects so probably you should also check what is hapenning inside your conductors, even if you could start with line or shell surface coil as for a RF case, to continu simple before going to a full model.
If you are interested int the back-EMF too, check that your model is not fully overconstraining the current, as normally you have a voltage supply and the current is driven by the difference of the back-EMF and the supply volts, and in MEF you solve for "V" and not "I" !
it is worth to take the COMSOL ACDC course to get knowledge of many trick's, I learned a lot there, specially that you must think out your Physics FIRST and fully understand what is driving what and how these couple, before setting up your model
--
Good luck
Ivar
first a warning inductive and eddy current models in 3D are rather tricky and require a powerfull PC to get solved, so you are using the best approach, start simple.
If you run the "MEF" physics then set some conduction to the air 1-1000 S/m, try a few values and lower until not solving anymore, then go back a step, this mostly helps the convergence, (check the air dissipation must still stay negligible if you want to understand your energy balance, as normally air dissipates 0W electrically
At 10kHz or below you have small skin effects so probably you should also check what is hapenning inside your conductors, even if you could start with line or shell surface coil as for a RF case, to continu simple before going to a full model.
If you are interested int the back-EMF too, check that your model is not fully overconstraining the current, as normally you have a voltage supply and the current is driven by the difference of the back-EMF and the supply volts, and in MEF you solve for "V" and not "I" !
it is worth to take the COMSOL ACDC course to get knowledge of many trick's, I learned a lot there, specially that you must think out your Physics FIRST and fully understand what is driving what and how these couple, before setting up your model
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
I switched to 2D and it is quickly converging.
The temperature range is around 1200 K.
The application is to prepare the blocks of AISI 403Cb blocks of stainless steel for hot forging.
Now to switch back to 3D I will be using external current source since now I do not have the Multi-Coil domain in 3D anymore.
The frequencies are quite low and they will not reach 10 kHz.
--
Reza Besharati Tabrizi
I am developing the induction heating model . I also stared with the 2 d axis symetric geometry and got the results.. But now I am working on the 3d model of the same and the solution is not converging . Were you able to get the complete soultion for the 3d geometry. Can you Please mail me your simulation so that I can figure out what changes are required in my model .
Thanks in advance..
Harshad Tadas