CFD Online Discussion Forums

CFD Online Discussion Forums (
-   OpenFOAM (
-   -   High nuSGS value in periodic hill case (

turbfoam March 4, 2013 19:42

High nuSGS value in periodic hill case
1 Attachment(s)
Hello Foamers,

I am work on the 2D periodic hill test case using channelFOam , validated by Ercoftac,

My problem is I get good profile for <U>,<V>,<u'u'>, <v'v'>,<w'w'> etc. but the magnitude of all these value are high i.e. the shape of the curve is good compared to the result in above link, but the numerical values overshoot.

the BIGGEST problem, is I get very high values of nuSgs during the simulation, and the profile does not look very reasonable too.

I have
1) USed a very fine grid - 4.7 million cells, - I know for sure that mesh is good since it was used in mnay previous journals with good result.
2) For initial condition: I built coarse mesh (450,000 cells) and averaged it for 200 flo through times, and interpolated the mean U field values to the fine mesh using mapFields
3) Attached the picture of my nuSGs at x/h = 4 axial distance. subgrid viscosity cannot be this high right?
the nuSgs value for the coarse (initial field) mesh was high like this too, but I thought that was due to poor mesh resolution near wall. but here its fine mesh, how is this hapening? I mapFields only the U file, to the fine mesh and not the others.
Please offer help... I have real no idea why this is happening.

Additional detial: Re = 10595
So based on Re and hill height of h = 28 mm, I calculate incoming bulk velocity as 7.56 m/s. In channelFoam transportProperties my Ubar is (7.56 0 0)

Thank you all... I thank your help very much.


All times are GMT -4. The time now is 16:37.