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 The information provided may be out of date.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.

Doubled values in Total normal heat flux.... why?

Please login with a confirmed email address before reporting spam

Hi,

I have a little problem with an heat transfer problem.
I'm simulating a 2d problem modelled by a square box of air with a circular solid (silicon) heat source in the middle.
Using the "heat transfer in solid " physic I set a boundary temperature of the square box and a 10 W total fixed heat source for the circular solid.
The stationary problem converges...
I tried to evaluate the line integral for the total normal heat flux on the circular heat surface, and I obtained a correct 10 W integral, the same thing doing the line integral on the boundary of the air box.

Then I tried to insert in the "heat transfer" physic a "Heat transfer in fluids" node for the air box.
At this point, solving again the problem, I find the same exact solution for temperature, etc, but if I evaluate the normal heat flux line integral on the heat source boundary I find a doubled value of about 20W while on the square air box I still find 10W

The interesting thing is that on the heat source boundary the "Total Normal Heat Flux" is exactly the double of the "Total heat Flux Magnitude". How is it possible?

UPDATE:

I tried to change the physic: I used a "heat transfer in fluids" physic for both the domains, and I obtain the10W "normal heat flux line integral both for the heat source. As soon as i configure an "heat transfer in solids" for the source, the line integral doubles!!!

Is this a bug or there is something I can't understand?

thank you!

3 Replies Last Post 2012年5月29日 GMT-4 09:57
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 2012年5月29日 GMT-4 09:36
Hi

My first thought it's not a "bug" rather a feature from the multiphysics coupling, but I'm not sure:

Only COMSOL "support" can give you THE REPLY, my understanding are: when you link these two physics you split the comon boundary to an "up" and "down" part so if you integrate on the internal side or only on the outer side you will get your 10W, but it will double.

I did a quick test and noticed it concerns ONLY the "Total Normal Heat Flux" variable "ht.ntflux" which is defined as the sum of "ht.ndflux+ht.ntrlflux+ht.naflux" (check the equations view) hence the sum of the normal conductive, translational and convective heat fluxes. When you add the two boundaries (Fluid & Solid) these seem to add up

I agree in first view not logic, it's absolutely worth to ask "support" about a confirmation, and report it back here.
Send then also your file it will be easier for them to understand

My 4.2a file hereby slightly more complex show the same behaviour

--
Good luck
Ivar
Hi My first thought it's not a "bug" rather a feature from the multiphysics coupling, but I'm not sure: Only COMSOL "support" can give you THE REPLY, my understanding are: when you link these two physics you split the comon boundary to an "up" and "down" part so if you integrate on the internal side or only on the outer side you will get your 10W, but it will double. I did a quick test and noticed it concerns ONLY the "Total Normal Heat Flux" variable "ht.ntflux" which is defined as the sum of "ht.ndflux+ht.ntrlflux+ht.naflux" (check the equations view) hence the sum of the normal conductive, translational and convective heat fluxes. When you add the two boundaries (Fluid & Solid) these seem to add up I agree in first view not logic, it's absolutely worth to ask "support" about a confirmation, and report it back here. Send then also your file it will be easier for them to understand My 4.2a file hereby slightly more complex show the same behaviour -- Good luck Ivar


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012年5月29日 GMT-4 09:51
You are right,

the other interesting thing if you evaluate the integral of the sum of normal conductive, translational and convective heat fluxes you obtain the right 10W!
You are right, the other interesting thing if you evaluate the integral of the sum of normal conductive, translational and convective heat fluxes you obtain the right 10W!

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 2012年5月29日 GMT-4 09:57
Hi

Well, if you find nothing specific about this in the doc, send a mail to suport ;)
It's the safest way to get correctl replies, here we are all users

--
Good luck
Ivar
Hi Well, if you find nothing specific about this in the doc, send a mail to suport ;) It's the safest way to get correctl replies, here we are all users -- 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.