heat transfer on two-side walls
 

Hi,

I am trying to simulate the effect of solar radiation in a solar collector. The absorber is modeled as a wall between two fluids zones (air) and is set as two side wall by Fluent.

The problem is when I assigned a heat flux (solar radiation ) on one side only it seems like there is no transfer of heat. and if I do it on the two sides , it seems like I have doubled the heat flux value and the outlet temperature of air is tooooo high, so what should I do ?

Thanks to you in advance

Solaris

no answers ?!!!!!!!:confused:

Please , somebody has meet this problem ?

Is the absorber modelled as a thin wall or does it have a thickness with a mesh inside? If it's a thin/interior wall, did you thermally couple the wall with its shadow-wall?

Thank you for the reply,

the absorber is a thin wall (not meshed), and I have not coupled the two sides . I want to apply a constant heat flux on the absorber as solar radiation (W/m2)

What does your model look like? Do you have like a flat surface which receives solar radiation on one side (the constant heat flux) and you'd want this heat then transferred to a fluid on the other side? Perhaps show some images?

Either way, if you want heat transferred through a wall, you need to thermally couple the walls.

yes exactly, I'e attached a schematic representation of the solar collector. I'e tried to couple the two sides but then I need a heat generation rate (w/m3) and I'e have only solar radiation (W/m2) ? the example is in 2D !!

sorry , the picture was not clear , here is a new attachment

Thanks for the schematic. Basically, as I see it, there's two things you can do:

1: Prescribe the heat flux on the glazing, solve the energy equation in the fluid between the glazing and absorber, and thermally couple the absorber's two faces, using the right material (HTC) and thickness. This way you'll have transport of solar heat to your air flowing through your solar cell.

2: Remove the whole upper part of your geometry and prescribe the heat flux on the top face of the absorber. You can then model the absorber as a material that takes the thermal resistance of the fluid above the absorber into account.

But you're right that you can't prescribe a heat flux on an interface, simply because it's not a boundary. Therefore, if you want to prescribe heat flux on the absorber, you have to make it into a boundary, like in option 2.

Do note that you've got the option to model the absorber either as a thin-wall (only normal heat conduction) or using shell conduction, which also allows for in-plane heat transport.

You can find some more info here:

http://www.engr.uconn.edu/~barbertj/...20Modeling.pdf

Thanks a lot Kevin,

Well, I will try the first option ( the second one leads to another type of solar conllector without glazing and this is not my study case).

What do you mean by HTC ?

and thanks for the link I will read the document carefully

I will keep you informed

No problem.

HTC = heat transfer coefficient. What I meant to say is that the material you choose determines the heat conductivity of the wall. So make sure you know the heat conductivity of your absorber to model the conjugate heat transfer properly.

Hi,

Well, my problem is not resolved yet, that's what I've tried to do:

- couple the two sides of the absorber and prescribe a heat generation rate (volumetric) calculated from the heat flux (W/m2), but I get very high temperatures in the outlet :eek:

-prescribe heat flux on the glazing and no heat generation rate on the absorber, but I get nearly the same pb and beside that in the real case I have a mixed BC on the glazing (convection and radiation with the ambiance) and I can't do it when I use heat flux :confused:

I am using also a radiation model (S2S) , do you think it's a part of the problem ?

somebody told be it's hard to model heat transfer with fluent ,well I am ok with him :rolleyes:


Please help !!!!!!

Ok,

In your first attempt, where you coupled both sides of the absorber, did you remove all the parts above the absorber or did you keep them? And why replace the heat flux by a volumetric heat rate? And where did you prescribe that volumetric heat rate, in the absorber?

What do you mean with pb?

The S2S radiation model definitely complicates it, but I think it still should be doable. Where did you prescribe it? From the glazing to the absorber?

Hi,


Well, I did'nt remove anything hen I coupled the two sides of the absorber.

and when I used the "coupled " option in BC on the absorber , Fluent doesn't give me other possibilities than : heat generation rate (W/m3) (I have to specify also emissivity and thickness of the absorber)


For S2S , yes I use it between absorber and glazing

I don't know what to now?

PS: I meant Problem ith "pb", sorry for my english :o

You only use the heat generation rate when you're trying to model a heat source, which you don't have here. You'll automatically get a heat flux on the bottom wall of the absorber when you couple the upper and lower walls, as you'll allow for heat conduction to take place through the material. What you need to do is figure out the correct material/conductivity and thickness.

About the heat generation rate , I am totally OK with you , but that's what I get when I use the "coupled" option in thermal conditions (see the pic attached) , no way to prescribe a heat flux with the "coupled" option .

concerning the thickness and material conductivity , actually I have the correct values from an experimental study and I want to find the same outlet temperature to validate my own study.

The correct heat flux will be calculated by the model when you set the two walls as coupled. When coupling the walls, you set them indirectly as interior walls, no longer as boundaries. Hence you can't apply a boundary condition like a heat flux anymore on the walls. However, when you specify the correct thickness and set the correct material (i.e., the conductivity), Fluent will solve a 1D heat equation through the wall. This will allow heat to be transferred/conducted from the top wall op your absorber to the bottom wall of your absorber, which on it turn will allow the air passing through to be heated up. It's like you set a heat flux on the bottom face of the absorber, but instead Fluent will calculate it for you. So if you have the correct value, use those and thermally couple the walls. The rest, like emissivity you'd still need to add though.

Hi,

Thanks a lot again, but I think I was not clear about what I'm doing, I have tried two possibilities:


For this first option I didn't prescribe any heat flux on the glazing , that's why I needed to prescribe a heat generation rate on the "coupled" two side absorber (solar radiation)

Here ,I also coupled the two sides of the absorber but put the heat flux (solar radiation ) on the glazing. Now, my problem is:

first : on the glazing I have a mixed BC in the reality: convection and radiation that represents the loss to the surroundings), I can't put them when I use heat flux condition "

second: when I change the mass flow rate , the outlet temperature doesn't change, it seems like there is no transfer heat by convection between air and the absorber !!!!:confused:

Do you have any idea ?