[ivanyao]

hi
hi,I am running a cube case using standard k-epsilon. I have it running successfully on a coarse mesh (4000 cells) with a timestep which gives a maximum Courant number of around 0.6. I attempted to refine the mesh somewhat, taking it to 1100,000 cells, then mapping the existing solution across and restarting with a timestep 1/15th of the original. but, after about 20 timesteps (with max courant number of around 0.6) the solution blows up, Initially the k field seemed to be going unbounded, followed by the Courant number exploding;and then epsilon. I have also tried to use smaller timesteps, non-orthogonal correctors, and some changes to the differencing schemes used; without success. i can not find the solution on the froum.PS:when i use the smaller timestep ,it diverged at 100 timesteps.i am appreciate if someone could give a hand.

and then my schemes and checkMesh file:
Create polyMesh for time = constant

Time = constant

Mesh stats
points: 959472
edges: 2845851
faces: 2813580
internal faces: 2749620
cells: 927200
boundary patches: 12
point zones: 0
face zones: 0
cell zones: 0

Number of cells of each type:
hexahedra: 927200
prisms: 0
wedges: 0
pyramids: 0
tet wedges: 0
tetrahedra: 0
polyhedra: 0

Checking topology...
Boundary definition OK.
Point usage OK.
Upper triangular ordering OK.
Topological cell zip-up check OK.
Face vertices OK.
Face-face connectivity OK.
Number of regions: 1 (OK).

Checking patch topology for multiply connected surfaces ...
Patch Faces Points Surface
inlet 7200 7381 ok (not multiply connected)
outlet 7200 7381 ok (not multiply connected)
<front> 800 861 ok (not multiply connected)
<back> 800 861 ok (not multiply connected)
<left> 400 441 ok (not multiply connected)
<right> 400 441 ok (not multiply connected)
<top> 800 861 ok (not multiply connected)
<left1> 7860 8052 ok (not multiply connected)
<right1> 7860 8052 ok (not multiply connected)
<top1> 15720 15972 ok (not multiply connected)
<bottom1> 14920 15231 ok (not multiply connected)
<bottom> 0 0 ok (empty)

Checking geometry...
Domain bounding box: (-0.6513 -1.002 0) (2.5551 1.002 1.002)
Boundary openness (-2.04452e-15 4.55103e-16 1.62419e-15) OK.
Max cell openness = 2.19421e-16 OK.
Max aspect ratio = 21.943 OK.
Minumum face area = 1.14391e-05. Maximum face area = 0.0014593. Face area magnitudes OK.
Min volume = 1.7049e-07. Max volume = 3.59875e-05. Total volume = 6.43747. Cell volumes OK.
Mesh non-orthogonality Max: 8.53774e-07 average: 0
Non-orthogonality check OK.
Face pyramids OK.
Max skewness = 5.02743e-14 OK.
Min/max edge length = 0.002505 0.0549672 OK.
All angles in faces OK.
Face flatness (1 = flat, 0 = butterfly) : average = 1 min = 1
All face flatness OK.

Mesh OK.

End


ddtSchemes
{
default Euler;
}

gradSchemes
{
default Gauss linear;
grad(p) Gauss linear;
grad(U) Gauss linear;
}

divSchemes
{
default none;
div(phi,U) Gauss limitedLinearV 1;
div(phi,k) Gauss limitedLinear 1;
div(phi,epsilon) Gauss limitedLinear 1;
div(phi,R) Gauss limitedLinear 1;
div(R) Gauss linear;
div(phi,nuTilda) Gauss limitedLinear 1;
div((nuEff*dev(grad(U).T()))) Gauss linear;
}

laplacianSchemes
{
default none;
laplacian(nuEff,U) Gauss linear corrected;
laplacian((1|A(U)),p) Gauss linear corrected;
laplacian(DkEff,k) Gauss linear corrected;
laplacian(DepsilonEff,epsilon) Gauss linear corrected;
laplacian(DREff,R) Gauss linear corrected;
laplacian(DnuTildaEff,nuTilda) Gauss linear corrected;
}

interpolationSchemes
{
default linear;
interpolate(U) linear;
}

snGradSchemes
{
default corrected;
}

fluxRequired
{
default no;
p;
}

[santos]

Hi,

It would help if you could provide additional information on your setup, eg. Re and fvSolution file.

Moreover, starting with 2nd order schemes may lead to some problems on the first time-steps. Try to change your div and laplacian schemes to 1st order, and see if it helps. Later, change them to 2nd order for improving the accuracy of the results.

Regards,
Jose Santos

[ivanyao]

hi，
thank you for your attention。this is my fvsolution：

solvers
{
p PCG
{
tolerance 1e-06;
relTol 0;
preconditioner GAMG
{
tolerance 1e-06;
relTol 0;
smoother GaussSeidel;
nCellsInCoarsestLevel 20;
mergeLevels 1;
agglomerator faceAreaPair;
cacheAgglomeration off;
nPreSweeps 0;
nPostSweeps 2;
nFinestSweeps 2;
scaleCorrection yes;
directSolveCoarsest no;
};
};
U BICCG 1e-05 0;
k BICCG 1e-05 0;
epsilon BICCG 1e-05 0;
R smoothSolver
{
tolerance 1e-06;
relTol 0;
smoother GaussSeidel;
nSweeps 1;
};
nuTilda BICCG 1e-05 0;
}

PISO
{
momentumPredictor yes;
nCorrectors 2;
nNonOrthogonalCorrectors 0;
pRefCell 0;
pRefValue 0;
}

what i can not understand is why the finer mesh cause the problem?

[ivanyao]

sorry,i make a mistake.this is my fvsolution:
/*---------------------------------------------------------------------------*\
| ========= | |
| \ / F ield | OpenFOAM: The Open Source CFD Toolbox |
| \ / O peration | Version: 1.4 |
| \ / A nd | Web: http://www.openfoam.org |
| \/ M anipulation | |
\*---------------------------------------------------------------------------*/

FoamFile
{
version 2.0;
format ascii;

root "";
case "";
instance "";
local "";

class dictionary;
object fvSolution;
}

// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

solvers
{
p PCG
{
preconditioner DIC;
tolerance 1e-06;
relTol 0;
};
U PBiCG
{
preconditioner DILU;
tolerance 1e-05;
relTol 0;
};
k PBiCG
{
preconditioner DILU;
tolerance 1e-05;
relTol 0;
};
epsilon PBiCG
{
preconditioner DILU;
tolerance 1e-05;
relTol 0;
};
R PBiCG
{
preconditioner DILU;
tolerance 1e-05;
relTol 0;
};
nuTilda PBiCG
{
preconditioner DILU;
tolerance 1e-05;
relTol 0;
};
}

PISO
{
nCorrectors 2;
nNonOrthogonalCorrectors 0;
pRefCell 0;
pRefValue 0;
}


// ************************************************** *********************** //

[santos]

Your fvSolution file seems ok. Maybe with increasing resolution you capture relevant flow structures that were not captured by the coarser mesh. Is k-epsilon adequate for your problem?

Jose Santos

[ivanyao]

hi,
i think it is ok.it can calculate 200000 cells,( the courant number get smaller by the time step,but it can finish the work.)but when i am increase to 1000000 cells,it is fail.is the turbulent model not suit to my case?i use standard k-epsilon.
thank you.

[ivanyao]

hi,foamers:
i can not fix the problem,could someone give me a hand?