hi i am doing the simulation for turbulent,free impinging jet for different H/D ratio(1,2,5,10,20)on flat plate i have good results for h/d 1 but when H/D ratio increases the simulaton results not matched the experimental results..... can any one tell me what the real problem is..... what i think that instead of increase H during modeling if we increases D may be results are good. but i also dont know how much dia should be increase for each H/D ratio......... if any one help me for that.i will be very thankful to him...

I advise you to study the grid influence on the solution for each case of H/D; you can use coarse, fine and very fine meshes. During the mesh generation process, care has to be taken in the choice of the first grid spacing near wall boundaries to obtain a proper resolution of the boundary layer for example when using the wall function approach to model the flow near the wall in the Y plus value has to be in the range of 20 to 100.

Hello; I advise you to study the grid influence on the solution for each case of H/D; you can use coarse, fine and very fine meshes. During the mesh generation process, care has to be taken in the choice of the first grid spacing near wall boundaries to obtain a proper resolution of the boundary layer for example when using the wall function approach to model the flow near the wall in the Y plus value has to be in the range of 20 to 100.

[fkhan7]

Hello,

I have been doing the same thing. and the velocity at the lip of the jet is the same as the experimental however when i get further it is higher then near the plate it get to 30% lower than the experimental. I have tried very fine grid and coarse but the result are still the same. I am using k-e model and k-w.

My inlet speed is 2.01 m/s Re= 16000. i am not sure if it is a grid related problem or turbulence model issue.

Thank you,

Fahad

[ghorrocks]

k-e is known to be too diffusive in free jet flows. It does not surprise me that results using that model are too low. Not sure about k-w, certainly some variants of k-w are better but I am not sure about the one implemeneted in CFX. Have you tried SST? That would be my guess.

But you need to do some research to find what turbulence model is going to have a good model of free jets before doing these simulations. Have a read of a turbulence textbook "Turbulence modelling for CFD" by Wilcox is a good one which discusses a lot of free jet work with various models.

[fkhan7]

Thank you very much for your input. I will surly read that book. Please feel free to recommend any other books that you think will be helpful in this matter. I do think it is a turbulence related issue more than it being just the mesh.

[elisun]

Hi. I am doing the simulation for turbulent,free impinging jet with Z/D=2 on flat plate. my Re number is 80000. the model i use is k-omega. the problem is that my result Nu number is so high ( approximately 2000), and i can not validate my results with any of simmilar problems in different papers. can any one tell me what the real problem may be..i really appriciate your help.
The fallowing is my geometry.

[ghorrocks]

How far off are you? Errors of 50% are not uncommon in heat transfer models.

[elisun]

ghorrocks, tnx for answering.
well,for Re = 124000 I found Nu to be maximum 600 in some papers. so my error is far above 50%.

[ghorrocks]

http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F