Dear All,

I have just started using CFX 5.4.

My model is a half of an axisymmetric jet impingement flow with heat plate at the stagnation region. It is a Energy type heat transfer problem. I have specified heat flux as the heat tranfer boundary condition at the heat plate.

I've obtained a solution to this problem but I find that the CFX Visualise is not giving any value for heat transfer coefficient at walls with heat flux as the boundary condition. I turned to the manual but found nothing on How to obtain HTC at these walls ?

Anybody has a clue on how to get it ?

Any help will be highly appreciated.

Rushyen

When you load the file into CFX-Visualise there is a selector at the top or middle of the dialog for choosing the user level of the variables which Visualise reads.

I belive that heat transfer coeficient will appear in the "Expert" or "All" category. I'm don't remember where exactly the documentation describes how this heat transfer coefficient is calculated, but I would start with the section on the heat tranfer coefficient boundary condtion.

Dan.

Dear Dan,

Thanks for replying. I've actually tried what u mentioned, that is loading all the variables(heat transfer coeff. is one of it) that CFX Visualise can display from the result file. The thing is, I'm not getting any value at the walls that i specify heat flux as the boundary condition.

I also did look at the online help for CFX-5 Solver.

- Advice on flow modelling with CFX-5 - Boundary Condition Issues

- Using Walls and Heat Transfer Coeff.

- pg 2-149

The note says(or this is what i understand) the heat transfer coeff. is stored as zero, because no wall temperature is specified.

Could this be the reason why I'm not getting any heat transfer coeff. values ? If yes, how can i get the heat transfer coeff. then ? If no, what is my mistake. What do i do next ?

Thanks again,

Rushyen

Hi Rushyen, quick question, when processing your results are you using the Hybrid default (when setting which parameters to load in visualise), or the conservative values ? This does make a difference to the values you get out of the model. I used a sampling plane to extract mass flows from a model where I couldn't use a slice plane (I had to get the velocitys out and then calculate the mass flows - which is not 100% correct but was the only way possible at the time). The results I obtained using the hybrid values were different to those I obtained for the conservative setup, with the conservative setup giving the correct results. as I understand it, the conserative setup does not include any interpolation at the boundaries, and this may be what you need ? I'm not 100% on this but would definitely recomend you have a look at the manual on this consevative/hybrid option. Good luck Stuart

Sorry, I forgot to mention that you should also make sure you load hybrid values, otherwise visualise just sets boundary fields to zero.

I'm not completely sure off the top of my head but it could be that the heat transfer coefficent is actually only calculated for walls with a specified temperature, not heat flux specified. I'll have to check to make sure.

For Tspec boundaries, then the local heat transfer coefficient is computed using q=hA(T-Twall). Where T is the fluid temperature just off the wall. I guess with specified q walls, the question is what to use for temperatures in that case.

Dan.

Dear Dan and Stuart,

Thanks and also sorry, for disappearing for some time.

I've tried using both conservative and hybrid. Both didn't give any readings at heat flux boundaries. From what i can understand, CFX can't give readings for HT Coeff. at walls with Heat Flux boundaries. (Is this true ? Somebody help ???)

Probably it is the confusion as Dan stated,

Q = hA(T-Twall)

What is Twall? since it is not given right.

If this is the case, what do i do?

I can get the fluid temperature values at walls using a sampling plane (using it as a line). Can I somehow use this values to calculate the HT Coeff. myself?

I know Q, A, T(from the result file). What value can i use as the T wall? Any clue on getting this?

Rushyen

(1). I think, you need to know exactly how a heat transfer boundary condition is implemented in the code, when the flow is laminar or turbulent. (2). If the laminar model is used, then the heat transfer is computed from the conduction equation at the wall on the fluid side. In this case, the fluid conductivity is known, and the wall Q will be used to update the Twall, until the process is converged. In this case, Twall will be part of the result. And the fluid conductivity is given. (3). For turbulent flows, if a low Reynolds number model is used, then the same heat conduction model can be used at the wall as long as the fine mesh near the wall is used. (4). But if the wall function is used, then you need to find out how the code actually implement the heat transfer calculation through the wall function.