[CFXer]

Hello.

I'm now calculating heat exchanger using porous domain in CFX.

I have to verify whether I use correct heat transfer coefficient in porous domain.

So I'm watching theory guide in ANSYS help.

But It shows weired correlation.

In equation (1-287) in figure, Q=hA(Ts-Tf)

h is overall heat transfer coefficient [W/m2K]
A is interfacial area density [1/m]

Those two parameters are input of porous heat transfer settings.

But.. following those parameters, Q is [W/m3] unit.

I cannot calculate actual heat transfer coefficient in CFD-Post because of that weird correlation.

Is that really [W/m3] unit? If so, how can I calculate actual heat transfer coefficient in Post-processing?

[ghorrocks]

From your previous post I understand you are using a porous model to model the flow resistance. You are not intending to use the porous model to affect the heat transfer, you have your own correlations you wish to use for that.

If this is the case then I would not use a porous model. I would impose the drag as a source term (set to the same equation as the porous model), and you are then free to do whatever you like with the heat transfer as a normal fluid.

If you use the porous model you are going to have to undo all the heat transfer model which is built in and not appropriate for your model. This is unnecessarily complex.

[CFXer]

Originally Posted by ghorrocks

From your previous post I understand you are using a porous model to model the flow resistance. You are not intending to use the porous model to affect the heat transfer, you have your own correlations you wish to use for that.

If this is the case then I would not use a porous model. I would impose the drag as a source term (set to the same equation as the porous model), and you are then free to do whatever you like with the heat transfer as a normal fluid.

If you use the porous model you are going to have to undo all the heat transfer model which is built in and not appropriate for your model. This is unnecessarily complex.

I'm sorry I cannot understand what you say.... (T_T)

In my case It is necessary to reduce computing nodes, so I selected porous media.

In pressure drop It shows very good performance, but only heat transfer I cannot understand how to calculate it.

So, I want to see how does it calculate heat transfer in full porous model, but it is very difficult to understand..

[ghorrocks]

The porous model is just a source term on the momentum equations. But it also puts source terms on the heat equation as well. You just want the momentum equation source terms, so just implement the source term yourself directly with just the momentum equation terms.

[1361012943]

Originally Posted by CFXer

I'm sorry I cannot understand what you say.... (T_T)

In my case It is necessary to reduce computing nodes, so I selected porous media.

In pressure drop It shows very good performance, but only heat transfer I cannot understand how to calculate it.

So, I want to see how does it calculate heat transfer in full porous model, but it is very difficult to understand..

Hi dear friend, I simulate porous media but it dosent have any effect on heat transfer. did you solved this problem?

[Opaque]

Originally Posted by 1361012943

Hi dear friend, I simulate porous media but it dosent have any effect on heat transfer. did you solved this problem?

What model did you use?
1 - Superficial velocity via Fluid Domain + Subdomain sources?
2 - True/Intrisic velocity via Porous Domain + Solid Material?

W/o such information it is nearly impossible to know why you are not getting what you expect.