Menu

Discussion Forum

New Discussion
Filtrer

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 The information provided may be out of date.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.

Joule heating simulation problem

Posted 23 déc. 2013, 19:43 UTC−5 Low-Frequency Electromagnetics, Heat Transfer & Phase Change Version 4.3b 4 Replies

Elham Khamsehashari
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Hi
I am doing joule heating simulation for a thin surface to generate high temperature. I applied 120V to a thin film but the obtained temperature was not high enough. I found that the temperature definition in boundary conditions is very important while without that the problem was not converged. Also applying the temperature to different surfaces causes different results. I am not sure how I should define the boundary conditions to get high enough temperature on the surface of thin film with the voltage of 120V.
Regarding the attached file, does anybody have any idea?

Regards,
Elham
4 Replies Last Post 3 janv. 2014, 13:42 UTC−5
Frank van Gool COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 27 déc. 2013, 03:20 UTC−5
Hi Elham,

In your case, the carbon has a very low conductivity, therefore 120V will only allow a current of 5e-6A, this won't heat up the object a lot.
What you need is a much higher current, say 1e-4A, that will need a voltage of around 2500 V. A voltage like that will heat it up about 100 degrees.

I noticed however two things that you can improve in your model:
1. Remove the PET from the current calculation. THis is done by adding a "heat transfer in solids" to your model, and assign it to the PET domain.
2. You conductivity values for silver and carbon are quite far apart. Very large differences are not a good idea in numerics. That you can do is lower the silver value to about 1e6 higher that the carbon. This will not change anything in your results, but is much more stable and faster.

Best regards,
Frank
Hi Elham, In your case, the carbon has a very low conductivity, therefore 120V will only allow a current of 5e-6A, this won't heat up the object a lot. What you need is a much higher current, say 1e-4A, that will need a voltage of around 2500 V. A voltage like that will heat it up about 100 degrees. I noticed however two things that you can improve in your model: 1. Remove the PET from the current calculation. THis is done by adding a "heat transfer in solids" to your model, and assign it to the PET domain. 2. You conductivity values for silver and carbon are quite far apart. Very large differences are not a good idea in numerics. That you can do is lower the silver value to about 1e6 higher that the carbon. This will not change anything in your results, but is much more stable and faster. Best regards, Frank
Elham Khamsehashari
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 27 déc. 2013, 11:38 UTC−5
Dear Frank
Thanks for your suggestions.
I change the electric potential into the normal current density with a value of 100 A/m^2 and got a difference in temperature of about 100 degree. However, this model must work with ordinary voltage/current applied to it. The silver is just used for electrical connection and it is the carbon film who must be heated up and radiated. First of all I tried to get the result, by convective heat flux and it seems that it works. You suggested to remove PET from calculation by adding a "heat transfer in solids" to its domain. Would you please explain what you mean, since PET is an thermal insulation?
Now, I changed the condition into radiation. However, the PET domain exists in "Electromagnetic Heat Source" since it is defined as electric and thermal insulation. You will find the file attached. Regarding the results I obtained, would you please let me know if the boundaries are well defined?

Best,
Elham
Dear Frank Thanks for your suggestions. I change the electric potential into the normal current density with a value of 100 A/m^2 and got a difference in temperature of about 100 degree. However, this model must work with ordinary voltage/current applied to it. The silver is just used for electrical connection and it is the carbon film who must be heated up and radiated. First of all I tried to get the result, by convective heat flux and it seems that it works. You suggested to remove PET from calculation by adding a "heat transfer in solids" to its domain. Would you please explain what you mean, since PET is an thermal insulation? Now, I changed the condition into radiation. However, the PET domain exists in "Electromagnetic Heat Source" since it is defined as electric and thermal insulation. You will find the file attached. Regarding the results I obtained, would you please let me know if the boundaries are well defined? Best, Elham
Frank van Gool COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 3 janv. 2014, 04:40 UTC−5
Dear Elham,

As you are not interested in the current flowing through the PET (there is hardly any), you can remove this from the computation. This is done by adding a "heat transfer in solids" condition and assigning it to the PET domain. It is then removed from the " joule heating" domain. It will still compute the temperature, but not the potential (V).

Regarding the current flowing, your total current is now in the order of 0.5 mA, that seems like an ordinary value. The fact that you need 12 kV for this is due to the isolating properties of your carbon.

Best regards,
Frank

note that switching to a direct solver improves convergence a lot.

Dear Elham, As you are not interested in the current flowing through the PET (there is hardly any), you can remove this from the computation. This is done by adding a "heat transfer in solids" condition and assigning it to the PET domain. It is then removed from the " joule heating" domain. It will still compute the temperature, but not the potential (V). Regarding the current flowing, your total current is now in the order of 0.5 mA, that seems like an ordinary value. The fact that you need 12 kV for this is due to the isolating properties of your carbon. Best regards, Frank note that switching to a direct solver improves convergence a lot.
Elham Khamsehashari
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 3 janv. 2014, 13:42 UTC−5
Dear Frank

Thanks for your kind help.
I applied "heat transfer in solids" to the PET domain and get the result. Now, I would like to measure the temperature gradient from radiation of the carbon film surface. For this purpose, I added a vacuum box over the heating element but I am not sure that it is a right configuration since I don't know how I should measure the temperature gradient! Would you please advise me if you have any idea? I have attached the file.

Best,
Elham
Dear Frank Thanks for your kind help. I applied "heat transfer in solids" to the PET domain and get the result. Now, I would like to measure the temperature gradient from radiation of the carbon film surface. For this purpose, I added a vacuum box over the heating element but I am not sure that it is a right configuration since I don't know how I should measure the temperature gradient! Would you please advise me if you have any idea? I have attached the file. Best, Elham

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