[mekanik]

Hi everyone.
I am modeling a cooling system(Radiator Panel) in CFX. I use the porous model approach for analysis.
First I did an analysis of a single fin structure. From this analysis I obtained values for the porous structure (heat transfer coefficient, porosity etc.).
The porous structure consists of 2 water columns, air inlet and outlet and porous (fin) construction. I put your photos below.
The water outlet temperature that I plugged in is 1.5-2 degrees above the expected. I have experimental data on my hand.
I thought that I got the heat transfer coefficient too high. So ı reduced it.
But this time it was something unexpected. As I dropped HTC, the output temperature of the water column began to increase.
That normally If the HTC should be reduced, the heat transfer is reduced and therefore the water outlet temperature must increase. Am I wrong here.
Model and porous structure values below.
Not:The first HTC I bought was 145W / m2K which I reduced it by 10 ares. The latest value is 10W / m2K and the water outlet temperature is 3-4 degrees above expected.(very strange) Mesh skewness max. about 0.5.
I could not find the problem. I need your help. Thank you in advance

[ghorrocks]

You will need to look at what the detailed temperature field looks like for both phases. Also see if the velocity field has changed.

If you can't work it out please post some images of your geometry, the temperature results and your mesh. Also include your output file.

[mekanik]

Originally Posted by ghorrocks

You will need to look at what the detailed temperature field looks like for both phases. Also see if the velocity field has changed.

If you can't work it out please post some images of your geometry, the temperature results and your mesh. Also include your output file.

Thanks to your interest.
I add some values. Changes are predicted(temperature contours)
İnlet water temp.=367.4K
Outlet water temp.=355K dT=12K
Total water column hight=1200mm
At first analaysis HTC=145W/m2K. I dropped it slowly.(135,125,100,...50...10W/m2K). But nothing change to water outlet temperature. I have working on it a few days. I am really stuck.Your ideas will be very helpful.
Where is the problem? ı cant figure out it.

[ghorrocks]

Have you done a mesh, time step (if transient) and convergence tolerance sensitivity check?

Also read FAQ: https://www.cfd-online.com/Wiki/Ansy..._inaccurate.3F

[mekanik]

I have already studied the work done on this subject (publications, private practice etc.)
I have taken into account the criteria you mentioned.
I made the studies you mentioned in the first analysis (mesh structure, convergence, turbulence models, etc.)

The analaysis 2. step(porous media approach) ı actually didnt study mesh.
I think skewness ,aspect ratio sufficient data.

Computional time 9-10h at least. so one simulation a day.(really time spending)
Maybe ı have to sub model working and consider that your points.

Well is it a problem in my mind from the interface definition?
Or I have a problem with the area that I added HTC settings.
I've added the settings below. I search at solver theory and solver modeling guide but couldnt find enough knowledge.

How bad is the mesh structure (which is not) needed to be different in the results in such large HTC changes?
Not:Steady state model.