[marcelgt87]

hello everyone,

I recently managed to run the ERCOFTAC Centrifugal Pump validation test case in a recent version of OpenFoam 1.5-dev with the help of this forum (thanks again!)

now I want to run this exact same case in OpenFoam 2.1.0 with AMI interfaces instead of GGI.

the turbulence model (k-epsilon) and all the boundary conditions are exactly the same as in the 1.5-dev case except for the pressure-outlet. in 1.5-dev the pressure outlet is set as "fixedmeanvalue", unfortunately this bc doesn't exist any more in 2.1.0, so i defined it as "fixedvalue".

the case converges, though not as good as in 1.5-dev.
the results for U make sense but still differ from those i got in 1.5-dev.
what puzzles me the most are the results for p. the pressure gradient is just inverted!

see for yourselves, i uploaded the pictures.

does anyone have an idea how this could happen or what i could do to get better results?

best regards,
Marcel

[paulo]

Hi,

Are the two cases rotating in the same direction?

Best Regards,

Paulo Rocha

[marcelgt87]

yes, they are. i just checked it again, to be safe.

[vonboett]

could you please post your initial U and p / p_rgh file and fvSolution, fvSchemes? I probably can't help put I'd like to take a look.

[vonboett]

...because I think your geometry is qite challanging for your turbulence model. Did you try with LES in combination with DMM subgrid scale model developed by the LTT Rostock and a SimpleGridFilter to allow backscatter? This would account better for the Influence of your geometry on the turbulence.

[wiedangel]

Hi marcelgt87,

I also wanted to do the same comparison you did using the ERCOFTAC validation case. This is because I also did not obtain the same P for a very simple case that I tried to compare using both approaches which you used ... It would be nice if you give some details about your schemes and solutions dictionaries.

Regards

[marcelgt87]

sorry, that it took so long. here are my files for OF 2.1.0.

U:

internalField   uniform (0 0 0);

boundaryField
{
    INLET
    {
        type            surfaceNormalFixedValue;
	refValue	uniform -11.4;
    }
    OUTLET
    {
        type            zeroGradient;
    }
    BLADE_ROT
    {
        type            fixedValue;
        value           uniform (0 0 0);    
    }
    BLADE_STAT
    {
        type            fixedValue;
        value           uniform (0 0 0);
    }
    HUB_SHROUD
    {
	type		empty;
    }
    AMI_INT
    {
	type		cyclicAMI;
	value		$internalField;
    }
    AMI_EXT
    {
	type		cyclicAMI;
	value		$internalField;
    }
}

p:

internalField   uniform 0;

boundaryField
{
    INLET
    {
        type            zeroGradient;
    }
    OUTLET
    {
        type            fixedValue;
	value		uniform 0;
    }
    BLADE_ROT
    {
        type            zeroGradient;
    }
    BLADE_STAT
    {
        type            zeroGradient;
    }
    HUB_SHROUD
    {
        type            empty;
    }
    AMI_INT
    {
	type		cyclicAMI;
	value		$internalField;
    }
    AMI_EXT
    {
	type		cyclicAMI;
	value		$internalField;
    }
}

fvSolutions:

solvers
{
    p
    {
        solver          GAMG;
        smoother        GaussSeidel;
        cacheAgglomeration true;
        nCellsInCoarsestLevel 20;
        agglomerator    faceAreaPair;
        mergeLevels     1;

        tolerance       1e-08;
        relTol          0.05;
    }

    pFinal
    {
        solver          GAMG;
        smoother        GaussSeidel;
        cacheAgglomeration true;
        nCellsInCoarsestLevel 10;
        agglomerator    faceAreaPair;
        mergeLevels     1;

        tolerance       1e-06;
        relTol          0;
    }

    "(U|k|epsilon)"
    {
        solver          smoothSolver;
        smoother        GaussSeidel;
        tolerance       1e-07;
        relTol          0.1;
    }

    "(U|k|epsilon)Final"
    {
        solver          PBiCG;
        preconditioner  DILU;

        tolerance       1e-05;
        relTol          0;
    }
}

PIMPLE
{
    nOuterCorrectors 4;
    nCorrectors     1;
    nNonOrthogonalCorrectors 0;
    pRefCell        0;
    pRefValue       0;
}

SIMPLE
{
    nNonOrthogonalCorrectors 0;
}

relaxationFactors
{
    fields
    {
        p               0.3;
    }
    equations
    {
        U               0.7;
        k               0.9;
        "epsilon.*"     0.9;
    }
}

cache
{
    grad(U);
}

fvSchemes:

ddtSchemes
{
    default         steadyState;
}

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(T(grad(U))))) 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               ;
}

unfortunately my background knowledge about all the settings in fvSchemes and fvSolutions is quite limited, so i copied those files from a tutorial case.

@wiedangel: did you succeed? did you get any results with that comparison? could you give me some more details about what exactly you did?

thank you!
marcel

[vonboett]

Hi Marcel, I would use OUTLET { type zeroGradient; } for p since I had some funny coordinate-dependent effects using fixedValue 0 at my outlets.

[marcelgt87]

i just tried to do what you said. i set the outlet to zerogradient. as a result openfoam tells me to provide a reference value for p. when i do that (refvalue 0 and refcell 0) the calculation doesn't converge and eventually crashes. i also tried setting the outlet as zerogradient and the inlet as fixedvalue 0, with the same result.

is there anything i overlooked?

regards,
marcel

[vonboett]

what does it look like when you use a high refvalue?

[marcelgt87]

quite the same

[kiddmax]

Hi Marcel,

Could you please send me a copy of testcase of ERCOFTAC Centrifugal Pump? I did not download it successfully. Thank you so much!

My email address: dlutyezhang@gmail.com or you can use this mail to share it by Dropbox.

Again thanks

Best regards,
Ye

[Lisandro Maders]

Hey guys, did you sort the pressure problem? If so, what was the problem and how do you manage to fix that?


Appreciate a lot your answer!

Lisandro