CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (http://www.cfd-online.com/Forums/openfoam-solving/)
-   -   Low-Re turbulence models (http://www.cfd-online.com/Forums/openfoam-solving/67946-low-re-turbulence-models.html)

grtabor September 1, 2009 11:34

Low-Re turbulence models
 
Hi. I'm trying to run simpleFoam using the low-Re turbulence models. I've run a preliminary calculation using the standard k-e model, then switched models. Each time on startup I get the following output;

Starting time loop

Time = 1001

DILUPBiCG: Solving for Ux, Initial residual = 9.45357e-05, Final residual = 9.97272e-08, No Iterations 2
DILUPBiCG: Solving for Uy, Initial residual = 0.0012608, Final residual = 2.48792e-06, No Iterations 2
DILUPBiCG: Solving for Uz, Initial residual = 0.000676913, Final residual = 7.49901e-07, No Iterations 2
DICPCG: Solving for p, Initial residual = 0.0957274, Final residual = 9.47486e-07, No Iterations 318
DICPCG: Solving for p, Initial residual = 0.00598807, Final residual = 9.84517e-07, No Iterations 207
time step continuity errors : sum local = 3.38296e-08, global = -5.26895e-10, cumulative = -5.26895e-10



request for volScalarField RASModel::G from objectRegistry region0 failed
available objects of type volScalarField are

14
(
pPrevIter
(1-(0.3*exp(-sqr((sqr(k)|(nu*epsilon))))))
k
((((Cmu*(1-exp((-Am*(((sqrt(k)*y)|nu)+1e-15)))))|((1-exp((-Aepsilon*(((sqrt(k)*y)|nu)+1e-15))))+1e-15))*sqr(k))|(epsilon+epsilonSmall))
nu
p
(symm(grad(U))&&grad(U))
((((Cmu*((1-exp((-Am*(((sqrt(k)*y)|nu)+1e-15))))|((1-exp((-Aepsilon*(((sqrt(k)*y)|nu)+1e-15))))+1e-15)))*sqr(k))|epsilon)*(symm(grad(U))&&grad(U)))
((1-exp((-Am*(((sqrt(k)*y)|nu)+1e-15))))|((1-exp((-Aepsilon*(((sqrt(k)*y)|nu)+1e-15))))+1e-15))
(sqr(k)|(nu*epsilon))
(((sqrt(k)*y)|nu)+1e-15)
(((C2*(1-(0.3*exp(-sqr((sqr(k)|(nu*epsilon)))))))*epsilon)|k)
epsilon
y
)

followed by a stack trace. What am I doing wrong? I've tried this with both the LaunderSharmaKE and the LienLeschzinerLowRe models with the same result. From an initial inspection of the code I can't see that G is being read in (in fact I think its calculated) and its not complaining about that.

Any suggestions gratefully received.

Gavin

gocarts September 1, 2009 19:20

Replace *WallFunction in k and epsilon fields
 
I came across the same problem - but it was my fault.

I hadn't replaced the epsilonWallFunction and kqRWallFunction conditions in the epsilon and k fields (case time directory) when I switched turbulence models high-Re to low-Re. Something to check.

ancsa February 15, 2010 10:52

Hi Richard,
by replacing you mean deleting it and simply adding a small value like in epsilon and k file e. g. :
SL6
{
value uniform 1e-10;
}
?

gocarts February 15, 2010 11:10

You have to provide meaningful epsilon and k values, for more see:
http://www.cfd-online.com/Wiki/Turbu...ary_conditions
but otherwise yes.

If you are not sure about the difference between high-Re and low-Re turbulence models then I'd suggest sticking with high-Re, i.e., turbulence models that use wall functions.

Hope this helps.

ancsa February 15, 2010 11:25

Thanks for fast reply. I will try to go on as you suggested, I know how to calculate the values.
I'm calculating my eps and k profile from measurement data, only I have no points near the wall, I was using wall functions hence, I just wanted to see the difference in the results if I change and I read in the forums to use small figures but not zero. I will read more about it.


All times are GMT -4. The time now is 02:23.