[Jingxue Wang]

I am calculating the flat building pressure in a wind tunnel.The model is just a cubic building inside a large cuboid wind tunnel. I used pisoFoam to solve and the LES Smagorinsky sub-grid model was adopted.The velocity of inlet boudary condition is turbulentInlet with fluctuation around 2m/s average wind speed.I improved the mesh for some times and the mesh quality is ok after I used checkMesh( shown in the followings).But the courant number becomes suddenly large and the caculation stoped at around 0.04s.I am really get caught in this and did not find the reason for this.The followings are the setups of my calculation.Could you please check them?Do you think the boundary conditions or discetisation schemes are reasonable?

I am really appreciate if somebody can give me any corrections or suggestions.Thanks!!!

0/p:

dimensions      [0 2 -2 0 0 0 0]; //*****Units: m2s-2   dynamic pressure=pressure/density*****//

internalField   uniform 0;

boundaryField
{
    INLET
    {
        type           zeroGradient;
    }
    OUTLET
    {
      //type           zeroGradient;
         type            fixedValue;
         value           uniform 0;
    }
    LEFT
    {
        // type           zeroGradient;
         type           symmetryPlane;
    }
    RIGHT
    {
        //  type           zeroGradient;
       type           symmetryPlane;
    }
    UP
    {
        // type           zeroGradient;
        type          symmetryPlane;
    }
    GROUND
    {
         type           zeroGradient;
    }
    ROOF
    {
         type           zeroGradient;
    }
    WINDWARD
    {
        type           zeroGradient;
    }
    BACKWARD
    {
         type           zeroGradient;
    }
    LEFTWARD
    {
         type           zeroGradient;
    }
    RIGHTWARD
    {
         type           zeroGradient;
    }
}

0/U

dimensions      [0 1 -1 0 0 0 0];//*****Units:m/s*****//

internalField   uniform (0 0 0);

boundaryField
{
    INLET
    {
        type            turbulentInlet;
        referenceField  uniform (2 0 0);
        fluctuationScale (0.02 0.01 0.01);
        value           uniform (2 0 0);
    }
    OUTLET
    {
         type           zeroGradient;
    }
    LEFT
    {
        type          symmetryPlane;
        //type            fixedValue;
        //value           uniform (0 0 0);
    }
    RIGHT
    {
        type           symmetryPlane;
        //type            fixedValue;
        //value           uniform (0 0 0);
    }
    UP
    {
        type           symmetryPlane;
        // type            fixedValue;
        // value           uniform (0 0 0);
    }
    GROUND
    {
        type            fixedValue;
        value           uniform (0 0 0);
    }
    ROOF
    {
        type            fixedValue;
        value           uniform (0 0 0);
    }
    WINDWARD
    {
         type            fixedValue;
        value           uniform (0 0 0);
    }
    BACKWARD
    {
         type            fixedValue;
        value           uniform (0 0 0);
    }
    LEFTWARD
    {
         type            fixedValue;
        value           uniform (0 0 0);
    }
    RIGHTWARD
    {
         type            fixedValue;
        value           uniform (0 0 0);
    }
}

fvschemes

ddtSchemes
{
    default         backward;
   //default         CrankNicolson 0.9;
}


gradSchemes
{
    default         leastSquares;
    //grad(p)         Gauss linear;
}

divSchemes
{
    default         none;
    div(phi,U)      Gauss LUST grad(U);
    div(phi,k)      Gauss limitedLinear 1;
    div((nuEff*dev2(T(grad(U))))) Gauss linear;
}

laplacianSchemes
{
    default         Gauss linear corrected;
}

interpolationSchemes
{
    default         linear;
}

snGradSchemes
{
    default         corrected;
}

fvSolutions

solvers
{
    p
    {
        solver          PCG;
        preconditioner  DIC;
        tolerance       1e-06;
        relTol          0;
    }

    U
    {
        solver          smoothSolver;
        smoother        symGaussSeidel;
        tolerance       1e-05;
        relTol          0;
    }
}

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

turbulenceProperty

simulationType  LES;

LES
{
    LESModel Smagorinsky;

    turbulence      on;
    printCoeffs     on;

    delta cubeRootVol;

    cubeRootVolCoeffs
    {
        deltaCoeff 1;
    }
}

[sheaker]

Hello. In my opinion courant number is too high.
Set Your deltaT in controlDict to 0.0005 to keep Courant Number around 1.

Also Your
"Mesh non-orthogonality Max: 63.7369"
And:
"nNonOrthogonalCorrectors 2;"
Could makes some troubles.

In my case I have non orthogonality max ~ 70 and:
"nNonOrthogonalCorrectors 6;"

At the beginning set Your parameters of piso loop on high values (fo example 5 nCorrectors and 10 nNonOrthogonalCorretors) and start simulation. Then try to lower them as much as possible until divergence occurs. You know the rest.

Have a nice day.
Sheaker

[Jingxue Wang]

Originally Posted by sheaker

Hello. In my opinion courant number is too high.
Set Your deltaT in controlDict to 0.0005 to keep Courant Number around 1.

Also Your
"Mesh non-orthogonality Max: 63.7369"
And:
"nNonOrthogonalCorrectors 2;"
Could makes some troubles.

In my case I have non orthogonality max ~ 70 and:
"nNonOrthogonalCorrectors 6;"

At the beginning set Your parameters of piso loop on high values (fo example 5 nCorrectors and 10 nNonOrthogonalCorretors) and start simulation. Then try to lower them as much as possible until divergence occurs. You know the rest.

Have a nice day.
Sheaker

Hi,sheaker
Thanks for your suggestion.

The minimum grid size is 0.01, and the inlet velocity is 2m/s. So to make the Co around 1, delta t=0.05 was adopted. The courant number is around 7 for the first steps but then increased suddenly. So more delta t should be adpoted?

Before, I tried to decrease the delta t to one order of magnitude(not for this case), it can continues more time steps but it also divergenced eventually.

I will increase nNonOrthogonalCorrectors and nCorrectors in PISO.

Many thanks for your help!I really appcreciate it.

[Jingxue Wang]

I decreased delta t to 0.0005 and increased nCorrectors and nNonOrthonalCorrectors to 6. But it stoped at the 3rd time step.

Maybe the boundary conditions are not reasonble? Do somebody has such experience?