CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM (http://www.cfd-online.com/Forums/openfoam/)
-   -   K Epsilon convergance issue (http://www.cfd-online.com/Forums/openfoam/87339-k-epsilon-convergance-issue.html)

Ollie April 18, 2011 06:15

K Epsilon convergance issue
 
Hi OpenFoamers,

I am having some trouble with K Epsilon when using PisoFOAM, I am initialising in SimpleFOAM and taking the final time step obtained from SimpleFOAM to be used as the 0 file for PisoFOAM.

The flow type is high Reynolds number flow over a cylinder within a control volume.

The flow properly initialises in simpleFOAM and converges as expcted. However after 10,000 iterations in pisoFoam there is little difference if any in the flow structure, a steady-state results appears to be gained. I am using an Euler scheme and have ran for 10 seconds. There is sort of instantaneous convergance.

The expected flow structure should be completely turbulent, with eddies being shed alternately from the cylinder, however a pointed "tail" remains in the cylinder wake.

I have ran a similar model in LES with no complications, thus justifing the mesh is of a high standard.

Has anyone else had this problem, if so, how did you resolve it?


My initial/boundary conditions are below.

internal Field
U - (1.461, 0, 0)
p - 0
epsilon - 0.024
k - 0.020491
nut - 0

inlet
U - fixed value (1.461, 0, 0)
p - zerogradient
k - fixed value 0.020491
epsilon - fixed value 0.024
nut - calculated 0

Cylinder
U - fixed value - (0,0,0)
p - zerogradient
k - kqRWallFunction (value 0.020491)
epsilon - epsilonWallFunction (value 0.024)
nut- nutWallFunction (value 0)

outlet
U - fixedValue 1.461, 0, 0)
p - fixedVlaue 0
k - zerogradient
epsilon - zerogradinet
nut - calculated (value 0)

Cheers
Ollie

maddalena April 18, 2011 08:11

Quote:

Originally Posted by Ollie (Post 304016)

outlet
U - fixedValue 1.461, 0, 0)
p - fixedVlaue 0
k - zerogradient
epsilon - zerogradinet
nut - calculated (value 0)

Not sure if that can be the answer, however: you should not fix both pressure and velocity at the outlet! What happens if you set the following?
Code:

outlet
U - zeroGradient or inletOutlet
p - fixedVlaue 0
k - zerogradient
epsilon - zerogradinet
nut - calculated (value 0)

mad

Ollie April 18, 2011 08:28

Hi maddalena,

Firstly thanks for your quick reply. I have tried what you have suggested already, I have tried changing the boundary conditions around for a while, and the result is still the same.

Could it be the solver pisoFoam, its looks like pisoFoam is creating a solution which looks like a statistical average, instead of a transient solution ?


regards

Ollie


All times are GMT -4. The time now is 18:16.