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.

Defining heat source correctly

Please login with a confirmed email address before reporting spam

I am busy modeling batteries that heat up during use, and an airflow over the batteries to cool them. Each battery produces 12W of heat, according to the manufacturer. I have simulated it as a 2D problem, where the batteries are 100 by 200mm. The batteries are also 50mm deep but this cannot be seen in 2D.

I then defined a heat source and put a total power of 12W on each battery. However, I then realised Comsol models 2D as 3D as well, and uses a standard depth of 1 meter. Then my simulation would be incorrect as the heat is generated for a battery only 50mm deep.

I then decided to use the General source option under heat source. There I can define W/m3. Dividing 12W/(0,1*0,05*0,2)m3 = 12000 W/m3. Would this be the correct way of modeling the batteries as a heat source? It seems that applying the 12000W/m3 gives very high temperature values, much higher than I anticipate, so I started to doubt if it is correct.

6 Replies Last Post Jan 7, 2016, 11:39 a.m. EST

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Nov 8, 2012, 7:06 a.m. EST
Hi,
I have been computing a similar problem and found the same issue. I managed to improve the results slightly by refining the cooling parameters. If these aren't correct then energy just keeps being inputted and cannot go anywhere other than heat the object.
Hi, I have been computing a similar problem and found the same issue. I managed to improve the results slightly by refining the cooling parameters. If these aren't correct then energy just keeps being inputted and cannot go anywhere other than heat the object.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Nov 8, 2012, 7:27 a.m. EST
I have defined cooling measures. The model consists of a box with three batteries in it. For a first simple analysis, I defined the outer boudaries of the box to be constant at ambient temperature, and you can see the temperature distribution. I have also made models were air flows past the box to provide cooling.

My question is more related to how I should define the heat source. How did you do this with your problem?
I have defined cooling measures. The model consists of a box with three batteries in it. For a first simple analysis, I defined the outer boudaries of the box to be constant at ambient temperature, and you can see the temperature distribution. I have also made models were air flows past the box to provide cooling. My question is more related to how I should define the heat source. How did you do this with your problem?

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Nov 8, 2012, 8:40 a.m. EST
I looked at the heat sink tutorial within comsol before modelling my problem, here they define a heat source with a total power, my model has this too. If you have the cooling measures I am unsure of why it isn't producing expected results.
The only other difference with your model that I can think of are the material properties. With batteries being composed of multiple materials, the heat capacities etc may differ from actual?
Sorry I cannot be of more help
I looked at the heat sink tutorial within comsol before modelling my problem, here they define a heat source with a total power, my model has this too. If you have the cooling measures I am unsure of why it isn't producing expected results. The only other difference with your model that I can think of are the material properties. With batteries being composed of multiple materials, the heat capacities etc may differ from actual? Sorry I cannot be of more help

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 Nov 8, 2012, 9:19 a.m. EST
Hi

COMSOL indeed simulates 2D as 3D with generally (if not always) 1 [m] depth, hence all fluxes are per m, and all heat densities are per meters.

This is OK if you respect the units in densities, some care if you use absolute values.

Try it out on a simple example: make a 2D 1m^2 square block add 1[MW/m^2] heat flux on the top, cte T temperature on the bottom, solve, this corresponds to an integrated value of 1MW input power for 1m depth and 1m aside, but only 1kW integrated power for a 1mm thin block

Then add anotherseparate square block of 1 m aside, add 1kW "total power" on top surface/edge and cte T on bottom, if you integrate now you have only 1kW/m^2 average top power density that is the value COMSOl averages out evenly, so a 1mm thick block would, with this intensity only consider 1W through the top edge

Basically if you think Flux it's OK, when passing on full total power you must each time remind yourself what is the true depth considered

Then do not mix up boundary sources and Volume sources, here it applies too, appart that your source is distributed evenly (if you use a scalar) over the full domain

Hope I'm clar

--
Good luck
Ivar
Hi COMSOL indeed simulates 2D as 3D with generally (if not always) 1 [m] depth, hence all fluxes are per m, and all heat densities are per meters. This is OK if you respect the units in densities, some care if you use absolute values. Try it out on a simple example: make a 2D 1m^2 square block add 1[MW/m^2] heat flux on the top, cte T temperature on the bottom, solve, this corresponds to an integrated value of 1MW input power for 1m depth and 1m aside, but only 1kW integrated power for a 1mm thin block Then add anotherseparate square block of 1 m aside, add 1kW "total power" on top surface/edge and cte T on bottom, if you integrate now you have only 1kW/m^2 average top power density that is the value COMSOl averages out evenly, so a 1mm thick block would, with this intensity only consider 1W through the top edge Basically if you think Flux it's OK, when passing on full total power you must each time remind yourself what is the true depth considered Then do not mix up boundary sources and Volume sources, here it applies too, appart that your source is distributed evenly (if you use a scalar) over the full domain Hope I'm clar -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 9 years ago Jan 5, 2016, 5:27 a.m. EST
Hi,
I am a beginner with COMSOL heat transfer module. I have a hollow cylinder which is being cooled by water on the inside and on the outside by air. I am supposed to model it as a steady state problem. Now the cylinder is supposed to be generating heat due to some reactions which I have to neglect. The amount of heat is 3.5 MW/m3. My problem is that my heat source is not being recognised. The temperature profiles look as though there is no heat generation. Any suggestions
Hi, I am a beginner with COMSOL heat transfer module. I have a hollow cylinder which is being cooled by water on the inside and on the outside by air. I am supposed to model it as a steady state problem. Now the cylinder is supposed to be generating heat due to some reactions which I have to neglect. The amount of heat is 3.5 MW/m3. My problem is that my heat source is not being recognised. The temperature profiles look as though there is no heat generation. Any suggestions

Please login with a confirmed email address before reporting spam

Posted: 9 years ago Jan 7, 2016, 11:39 a.m. EST
Hi,

I have the same problem as above, setting temp on boundary or overall heat transfer rate works, but not general heat source.
Thanks.
Hi, I have the same problem as above, setting temp on boundary or overall heat transfer rate works, but not general heat source. Thanks.

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.