CFD Online Discussion Forums - Total Reynolds Stress and Turbulence Energy Spectrum of LES

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)

- OpenFOAM Post-Processing (https://www.cfd-online.com/Forums/openfoam-post-processing/)

- - Total Reynolds Stress and Turbulence Energy Spectrum of LES (https://www.cfd-online.com/Forums/openfoam-post-processing/120612-total-reynolds-stress-turbulence-energy-spectrum-les.html)

Total Reynolds Stress and Turbulence Energy Spectrum of LES

Hi,
I am doing LES with pisoFoam. My questions that may be useful as post-processing tools and/or as a utility for validation of model are:
1- How can we calculate the Total Reynolds Stress? Based on thread (http://www.cfd-online.com/Forums/ope...tml#post342243),
We just need to “UPrime2Mean + RMean” (respectively resolved + SGS parts) to get the full stresses. But, is this calculation valid for LES (pisoFoam)?
2- Is there any way to calculate “turbulence energy spectrum” of LES directly in openFoam ?
Thanks

Quote:

Originally Posted by sanysaly (Post 439053)

Dear sanysaly
The pisoFoam solver gives you the UPrime2Mean,referred to the "filtered scale", just like sad in the thread mentioned, so you have to add the contribution from the "SGS scale", just like:
(2.0/3.0)*I)*k_ - nuSGS_*twoSymm(fvc::grad(U_) ,
if you want to have the full Reynolds stress tensor.
For the second question, I suggest to calculate in an external software, just like twice the Fourier transform of the autocovariance.
Giulio S

Total Reynolds Stress in LES

Hi GiulioS,

This is an old post but i see you have suggested to calculate total Reynolds stress by considering
total Reynolds Stress = UPrime2Mean + Subgrid_ReynoldsStress.

But, my question is in the LES tutorial channel395, openFoam calculates reynolds stress only using UPrime2Mean, so which is correct?

postChannel utility.

Code:

     volScalarField Rxy(UPrime2Mean.component(symmTensor::XY));





     makeGraph(y, RxyValues, "uv", path, gFormat);

So which method is correct? I saw Eugene De Villiers also suggested your method. But I am confused now. :confused:

Thank you.

Hi there,

Did get an answer to your question?

Is it possible to use UPrime2Mean for the calculation of the energy spectrum ?

Thanks in advance

Quote:

Originally Posted by Uyan (Post 528906)

Code:

     volScalarField Rxy(UPrime2Mean.component(symmTensor::XY));





     makeGraph(y, RxyValues, "uv", path, gFormat);

So which method is correct? I saw Eugene De Villiers also suggested your method. But I am confused now. :confused:

Thank you.

Hi Mirage,

I figured out that for a well resolved LES the modelled part of Reynolds stress is of very little magnitude.
So if you look at this paper, it's clear why only the resolved part is used.

LARGE EDDY SIMULATION OF TURBULENT
CHANNEL FLOWS
C.Fureby, A.D.Gosman, G.Tabor, H.G.Weller,

I assume you meant "turbulent energy spectrum" , i have not done that.
I think it's a good idea to record velocity signals on some points and follow the approach suggested by GiulioS.