Grid-dependent pressure fluctuations
 

Hi all,

I am working on simulating a simple parallel plate rheometer flow. The geometry is a simple cylinder, and the upper plate rotates. The grid I am using is a fairly standard approach for discretizing a circle, as far as I know, you can see an image here: https://ibb.co/c3HRtc

Basically the issue I am having is that I am getting very nice convergence with simpleFoam, and the fluid velocity profile looks great, but the pressure shows some oscillation which is apparently due to the grid, you can see an image here: https://ibb.co/kZ6amx


The magnitude of this oscillation does decrease when I refine the grid, but I would prefer to know if there is a way to eliminate this artifact by changing some property of the solver for example. I am using fairly typical settings as far as I know. Here is my fvSolution:

solvers
{
p
{
solver GAMG;
tolerance 1e-16;
relTol 0.5;
smoother GaussSeidel;
cacheAgglomeration true;
nCellsInCoarsestLevel 10;
agglomerator faceAreaPair;
mergeLevels 1;
}

pFinal
{
$p;
}

U
{
solver smoothSolver;
smoother symGaussSeidel;
tolerance 1e-16;
relTol 0.5;
}

UFinal
{
$U;
}

C
{
solver PBiCGStab;
preconditioner DILU;
tolerance 1e-10;
relTol 0;
}
}

SIMPLE
{
nNonOrthogonalCorrectors 1;
pRefCell 0;
pRefValue 0;

residualControl
{
p 1e-16;
U 1e-16;
}
}

relaxationFactors
{
fields
{
p 0.3;
}
equations
{
U 0.7;
nuTilda 0.7;
}
}

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

and here is my fvSchemes:

ddtSchemes
{
default backward;
}

gradSchemes
{
default Gauss linear;
}

divSchemes
{
default none;
div(phi,U) Gauss linear;
div(phi,C) Gauss linear;
div((nuEff*dev2(T(grad(U))))) Gauss linear;
}

laplacianSchemes
{
default Gauss linear corrected;
}

interpolationSchemes
{
default linear;
}

snGradSchemes
{
default corrected;
}

fluxRequired
{
default no;
p ;
}


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

I would really appreciate any suggestions on how to eliminate this artifact in the pressure solution. Like I said, convergence is great according to all the residuals, etc...

Thanks!