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.

Simulation of a floor heating

Please login with a confirmed email address before reporting spam

Hi all,

I want to simulate a floor heating in a 2D modell. Floor heating systems release about 90 % of heat in the form of radiation and about 10 % in the form of convection. So is it possible to take this situation on one boundary? I tried this with heat-flux and surface-to-surface radiation. But there i can't divided the heat transfer in 10 % heat flux and 90% radiation. I also don't understand the equation in surface-to-surface radiation, what does mean "G".

I hope somebody can help me.

Thanking you in anticipation.

Markus

10 Replies Last Post Dec 28, 2013, 3:27 p.m. EST
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 Dec 4, 2012, 3:47 a.m. EST
Hi

radiative transfer is somewhat tricky, surface to surface exchange depends on the "view factors" G the apparent sizes of the surfaces compared to the more or less isotropic emissivity into 2pi steradians of heat from the source surface

then your 10 - 90% rule is from the energy balance, if you simulate heat convection and radiations with the physics laws you need to adapt the emissivity, absorption respectively h convection coefficients such that you get your end net balance 10-90%. or you decide that that 10 % goes into the air as a bulk heat capacity and 90 distributed to the other surfaces. But radiation is bidirectional, the walls exchange also with the floor and deposits energy too, therefore one need really to think and sketch up the heat flows before one starts to define the simulation

--
Good luck
Ivar
Hi radiative transfer is somewhat tricky, surface to surface exchange depends on the "view factors" G the apparent sizes of the surfaces compared to the more or less isotropic emissivity into 2pi steradians of heat from the source surface then your 10 - 90% rule is from the energy balance, if you simulate heat convection and radiations with the physics laws you need to adapt the emissivity, absorption respectively h convection coefficients such that you get your end net balance 10-90%. or you decide that that 10 % goes into the air as a bulk heat capacity and 90 distributed to the other surfaces. But radiation is bidirectional, the walls exchange also with the floor and deposits energy too, therefore one need really to think and sketch up the heat flows before one starts to define the simulation -- Good luck Ivar

Nagi Elabbasi Facebook Reality Labs

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 4, 2012, 1:53 p.m. EST
Hi Markus,

Surface to surface radiation is tricky as Ivar mentioned. The best path for you in my opinion depends on what you want to get out of this model. If you know for sure that radiation is 9 times higher than convection you can just scale your heat transfer coefficient at the surface.

If however your objective is to validate that indeed 90% of the heat is lost due to radiation you need to accurately model both convection and radiation. You can model the convection heat loss to air either with a convection boundary condition or conjugate heat transfer (modeling the air as well that is). To model surface-to-surface radiation COMSOL solves for the Radiosity. You need that because, as Ivar pointed out, radiation is bi-directional and the walls also transfer heat to the floor by radiation. The COMSOL manuals describe the theory briefly. J is the radiosity or the thermal radiation leaving a point on the surface, G is the irradiation or radiation reaching a point from other surfaces, and F are the view factors. COMSOL solves for these quantities based on the geometry, emissivity and the temperatures.

Nagi Elabbasi
Veryst Engineering
Hi Markus, Surface to surface radiation is tricky as Ivar mentioned. The best path for you in my opinion depends on what you want to get out of this model. If you know for sure that radiation is 9 times higher than convection you can just scale your heat transfer coefficient at the surface. If however your objective is to validate that indeed 90% of the heat is lost due to radiation you need to accurately model both convection and radiation. You can model the convection heat loss to air either with a convection boundary condition or conjugate heat transfer (modeling the air as well that is). To model surface-to-surface radiation COMSOL solves for the Radiosity. You need that because, as Ivar pointed out, radiation is bi-directional and the walls also transfer heat to the floor by radiation. The COMSOL manuals describe the theory briefly. J is the radiosity or the thermal radiation leaving a point on the surface, G is the irradiation or radiation reaching a point from other surfaces, and F are the view factors. COMSOL solves for these quantities based on the geometry, emissivity and the temperatures. Nagi Elabbasi Veryst Engineering

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 10, 2012, 7:17 a.m. EST
Thanks for your help!!!!!

I calculated that 95% of heat transfer is radiation and 5% convection.

But now I have a problem to find the right solution.

I used the heat flux physics with the surface-to-surface-radiation. There I set the surface emissivity at every boundary. But which radiation direction is the right one? What does this mean?
I also don't know which boundary condition is the right one for the heated floor. I tried with a given temperatur, a general inward heat flux and with a inward heat flux. There are always different solutions. I think the model with the inward heat flux is the best one, because there I can define the surface temperature of the floor and the heat transfer coefficient. Is that right?

At last I have a general question: Is it possible to work with radiation in Comsol 3.5 or in a similarly way?

Excuse me for my bad english.

Thanks.

Markus
Thanks for your help!!!!! I calculated that 95% of heat transfer is radiation and 5% convection. But now I have a problem to find the right solution. I used the heat flux physics with the surface-to-surface-radiation. There I set the surface emissivity at every boundary. But which radiation direction is the right one? What does this mean? I also don't know which boundary condition is the right one for the heated floor. I tried with a given temperatur, a general inward heat flux and with a inward heat flux. There are always different solutions. I think the model with the inward heat flux is the best one, because there I can define the surface temperature of the floor and the heat transfer coefficient. Is that right? At last I have a general question: Is it possible to work with radiation in Comsol 3.5 or in a similarly way? Excuse me for my bad english. Thanks. Markus

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 18, 2012, 6:01 a.m. EST
Couldn't anybody help me?
Couldn't anybody help me?

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 Jan 7, 2013, 7:46 a.m. EST
Hi

I believe you do not get any good answer because it's not fully clear what you are doing or how.

If you draw a square, in 2D, set a given "hot" temperature to the 2 walls and to the "roof" and a lower fixed temperature to the "floor", fill the box with air and select HT you will get a heat conduction exchange (but no convecting as "Convection cooling" acts on the "external side" of a boundary).
Now if you add a surface to surface radiation exchange, set a material emissivity for the three wall and roof, and another for the "floor" Boundaries you can get the radiative heat exchange in addition to the air conduction.

Still to get true convection you need to define some Boyance forces and solve the CFD problem

--
Good luck
Ivar
Hi I believe you do not get any good answer because it's not fully clear what you are doing or how. If you draw a square, in 2D, set a given "hot" temperature to the 2 walls and to the "roof" and a lower fixed temperature to the "floor", fill the box with air and select HT you will get a heat conduction exchange (but no convecting as "Convection cooling" acts on the "external side" of a boundary). Now if you add a surface to surface radiation exchange, set a material emissivity for the three wall and roof, and another for the "floor" Boundaries you can get the radiative heat exchange in addition to the air conduction. Still to get true convection you need to define some Boyance forces and solve the CFD problem -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 7, 2013, 8:16 a.m. EST
Thank you very much for your answer!

When I set a temperature to the walls and the roof, they doesn´t change. But as a result of the radiation the wall and roof temperatur rise, because of my floor heating.So I set a heatflux from outside at the boundaries (with a higher temperature at the floor).

And when I add a surface to surface radiation, there isn´t a changing.

How can I define Boyance forces? Do you mean Volume forces?

Thanks.

Markus
Thank you very much for your answer! When I set a temperature to the walls and the roof, they doesn´t change. But as a result of the radiation the wall and roof temperatur rise, because of my floor heating.So I set a heatflux from outside at the boundaries (with a higher temperature at the floor). And when I add a surface to surface radiation, there isn´t a changing. How can I define Boyance forces? Do you mean Volume forces? Thanks. Markus

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 Jan 7, 2013, 9:00 a.m. EST
Hi

But if your walls are the outer boundaries and you have a fixed temperature, this T value will not change ?
but the heat flux to keep the temperature fixed might change.

Do you model also your walls as "solids" and apply constant T on the outside ?

Indeed Boyance forces is for me Volume forces from density changes due to T and gravity related effects starting a convective motion in a closed volume for a fluid

--
Good luck
Ivar
Hi But if your walls are the outer boundaries and you have a fixed temperature, this T value will not change ? but the heat flux to keep the temperature fixed might change. Do you model also your walls as "solids" and apply constant T on the outside ? Indeed Boyance forces is for me Volume forces from density changes due to T and gravity related effects starting a convective motion in a closed volume for a fluid -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 7, 2013, 11:19 a.m. EST
Yes, I model my walls as "solid" and I have given a fixed external temperatur of -15°C. The width of the walls is only 0,05 meter (consisting of metall, ply wood and plastic).

Why isn`t it possible to use "Convection cooling" on the "external side" of a boundary?
Supposed I heat the floor with an temperature over 100°C, then I think the surface-temperatur of the external wall isn´t -15°C near the floor, because of conduction, isn`t it?

When I apply "Convection cooling" I get generell a higher temperatur then applying a fifed temperatur on the external boundary.

Regards,
Markus
Yes, I model my walls as "solid" and I have given a fixed external temperatur of -15°C. The width of the walls is only 0,05 meter (consisting of metall, ply wood and plastic). Why isn`t it possible to use "Convection cooling" on the "external side" of a boundary? Supposed I heat the floor with an temperature over 100°C, then I think the surface-temperatur of the external wall isn´t -15°C near the floor, because of conduction, isn`t it? When I apply "Convection cooling" I get generell a higher temperatur then applying a fifed temperatur on the external boundary. Regards, Markus

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 Jan 8, 2013, 1:54 a.m. EST
Hi

sorry the convective cooling only works on the "external side" (towards where there is no material)
the convective "cooling" can also be convective heating, it all depends on the sign of the temperature gradient (Text - T boundary)

Check carefully the "override laws" and the differences between "Inflow heat flux", "Heat flux" and "Convective cooling", and where all three applies

--
Good luck
Ivar
Hi sorry the convective cooling only works on the "external side" (towards where there is no material) the convective "cooling" can also be convective heating, it all depends on the sign of the temperature gradient (Text - T boundary) Check carefully the "override laws" and the differences between "Inflow heat flux", "Heat flux" and "Convective cooling", and where all three applies -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 28, 2013, 3:27 p.m. EST
Hi Markus

I am simulating a thin film of carbon using joule heating module. This film will be heated up by connecting to a voltage/current with silver connections. I want to simulate heat convection as well as heat radiation emitting from this film. I found that your model is similar to mine. I started with a simple model as the attached file, but by defining boundary conditions of "convective heat flux" and "surface to ambient radiation" I dont know how I can separate the results of convection and radiation.
Regarding to your experience, would you please let me know your suggestions?

Thanks

Best,
Elham
Hi Markus I am simulating a thin film of carbon using joule heating module. This film will be heated up by connecting to a voltage/current with silver connections. I want to simulate heat convection as well as heat radiation emitting from this film. I found that your model is similar to mine. I started with a simple model as the attached file, but by defining boundary conditions of "convective heat flux" and "surface to ambient radiation" I dont know how I can separate the results of convection and radiation. Regarding to your experience, would you please let me know your suggestions? Thanks 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.