
[Sponsors] 
November 7, 2005, 06:11 
Sorry, but I can't find anywhe

#1 
New Member
Canelli Claudio
Join Date: Mar 2009
Location: Genoa, Italy
Posts: 11
Rep Power: 8 
Sorry, but I can't find anywhere a default set of values for the compressible realizable kepsilon model:
Cmu A0 C2 alphak alphaEps alphah can anybody help me? thanks 

November 11, 2005, 04:41 
From paper by Shih et al. (Com

#2 
Member
Niklas Wikstrom
Join Date: Mar 2009
Posts: 85
Rep Power: 8 
From paper by Shih et al. (Computers and Fluids Vol. 24, No. 3, pp. 227238, 1995) it seems to be
realizableKECoeffs { Cmu Cmu [0 0 0 0 0 0 0] 0.09; A0 C1 [0 0 0 0 0 0 0] 4.0; C2 C2 [0 0 0 0 0 0 0] 1.92; alphak alphaK [0 0 0 0 0 0 0] 1.0; alphaEps alphaEps [0 0 0 0 0 0 0] 1.2; } /N 

August 14, 2006, 11:49 
Seems like there are some miss

#3 
Member
stefan
Join Date: Mar 2009
Posts: 96
Rep Power: 8 
Seems like there are some missing entries like alphah etc.
Does anybody know the realiableKECoeff ? I cannot find any references in OF? 

August 14, 2006, 11:51 
realizableKECoeffs
{
Cmu

#4 
Senior Member
Hrvoje Jasak
Join Date: Mar 2009
Location: London, England
Posts: 1,765
Rep Power: 21 
realizableKECoeffs
{ Cmu Cmu [0 0 0 0 0 0 0] 0.09; A0 A0 [0 0 0 0 0 0 0] 4.04; C2 C2 [0 0 0 0 0 0 0] 1.92; alphak alphak [0 0 0 0 0 0 0] 1.0; alphaEps alphaEps [0 0 0 0 0 0 0] 0.76923; }
__________________
Hrvoje Jasak Providing commercial FOAM/OpenFOAM and CFD Consulting: http://wikki.co.uk 

August 14, 2006, 11:56 
thanks Hrv, but with this setu

#5 
Member
stefan
Join Date: Mar 2009
Posts: 96
Rep Power: 8 
thanks Hrv, but with this setup the following error occurs:
keyword alphah is undefined in dictionary " in turbulenceProperties::realizableKECoeffs" 

August 14, 2006, 12:40 
the coefficient alphah is used

#6 
Member
Pierre Le Fur
Join Date: Mar 2009
Location: UK
Posts: 60
Rep Power: 8 
the coefficient alphah is used to evaluate the effective turbulent thermal diffusivity. (see realizableKE.h)
just add alphah alphah [0 0 0 0 0 0 0] 1; as in realizableKECoeffs { Cmu Cmu [0 0 0 0 0 0 0] 0.09; A0 A0 [0 0 0 0 0 0 0] 4.04; C2 C2 [0 0 0 0 0 0 0] 1.92; alphak alphak [0 0 0 0 0 0 0] 1.0; alphaEps alphaEps [0 0 0 0 0 0 0] 0.76923; alphah alphah [0 0 0 0 0 0 0] 1; } 

August 29, 2007, 03:13 
I have a question about the re

#7 
Member
Roland
Join Date: Mar 2009
Location: Netherlands
Posts: 62
Rep Power: 8 
I have a question about the relationship between the value of sigmaEps given by Shih (1.2) and the value of alphaEps given by Jasak (0.76923).
With these values alphaEps = 1 / (sigmaEps + 0.1), but where does the 0.1 come from? I have studied the equations in the Shih paper and realizable.C, but there is too much mathematical wizardry going on for me to fully understand it. Thank you. Sylvester 

July 12, 2008, 12:13 
Dear Foamers,
to clarify th

#8 
Member
Stefan Radl
Join Date: Mar 2009
Location: Graz, Austria
Posts: 82
Rep Power: 9 
Dear Foamers,
to clarify the situation the following should be noted: OpenFOAM's implementation of the realizable ke model does not strictly follow the paper of Shih et al. and the coefficients proposed in it: sigmaEps is 1.3 in OF (and hence 1/sigmaEps = alphaEps = 0.76923), whereas Shih et al. [1] propose sigmaEps = 1.20 (see Eqn. 28 of their paper). Furthermore, OF uses C2=1.92 whereas Shih et al. [1] use 1.9 (see table 2 of their paper). These two deviations may be errors or may be introduced by the programmers/users to be consistent with the standard ke model. Also A0 is 4.0 in Shih et al. [1] and not 4.04 as proposed here. (However, the value of 4.04 is consistent with www.cfdonline.com/Wiki while the others are not). It seems that all equations are consistent with the paper of Shih et al. [1] except the fact, that OF uses (nut/sigmak+nu) and (nut/sigmaEps+nu) instead of (nut/sigmak) and (nut/sigmaEps) as in the paper of Shih et al. [1] respectively. This is of course a small effect and should not have any further consequences at high Renumber flows. I hope this clarifies the parameters of the OF realizableKE model and I would like to suggest to include a clarification in the next release of OF. Reference: [1]TsanHsing Shih, William W. Liou, Aamir Shabbir, Zhigang Tang and Jiang Zhu. 1995. A New kepsilon Eddy Viscosity Model for High Reynolds Number Turbulent Flows, Computers and Fluids 24. 227238. 

March 28, 2011, 16:09 

#9 
Senior Member
Join Date: Feb 2010
Posts: 213
Rep Power: 8 
Hi, foamers.
I have two question about realizable kepsilon turbulent model. I included in RASProperties file the following coefficients (I found them in realizableKE.H): Code:
realizableKECoeffs { Cmu 0.09; A0 4.0; C2 1.9; alphak 1.0; alphaEps 0.833333; alphah 1.0; // only for compressible } Anyway, I obtain the following error: Code:
Selecting incompressible transport model Newtonian Selecting RAS turbulence model realizableKE [0] [0] [0] LHS and RHS of + have different dimensions dimensions : [0 0 1 0 0 0 0] + [0 2 3 0 0 0 0] [0] [0] [0] From function operator+(const dimensionSet& ds1, const dimensionSet& ds2) [0] in file dimensionSet/dimensionSet.C at line 403. [0] FOAM parallel run aborting Thanks for your attention. 

March 31, 2011, 03:45 

#10 
Senior Member
Join Date: Feb 2010
Posts: 213
Rep Power: 8 
It means I'm an idiot. The realizableKE coefficients are correct, but I derived /0/epsilon from /0/omega (it was an 'old' case) and I forgot to edit the units too.


Thread Tools  
Display Modes  

