[kaszt]

Hi,

I'm interested in simulating simple 2-D flow over a NACA0012 airfoil, starting with an AoA of 12 degrees. I'm using the k-omega SST turbulence model for closure, with the freestream velocity at 100 m/s, corresponding to a Re # of ~6.6 million. I've used a separate script to setup the mesh with OpenFOAM's blockMesh capabilities, and have been working on refining the mesh to a sufficient point without being too computationally intensive. However, throughout this effort, I have consistently been over-predicting the drag by several orders of magnitude. Comparing with a similar Re # from inviscid results obtained from XFLR5, Cd should be around 0.013, but I am getting around 0.17, an order-of-magnitude of 10 higher. The lift coefficient is a bit better, at 1.09 (XFLR5 is around 1.3). Without comparing to other results I know the drag is wrong. I've included the controlDict file here:

/*--------------------------------*- C++ -*----------------------------------*\
  =========                 |
  \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox
   \\    /   O peration     | Website:  https://openfoam.org
    \\  /    A nd           | Version:  8
     \\/     M anipulation  |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       dictionary;
    location    "system";
    object      controlDict;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

application     rhoPimpleFoam;

startFrom       latestTime;

startTime       0;

stopAt          endTime;

endTime         5;

deltaT          2e-5;

writeControl     adjustableRunTime;

    //div(phi,U)      bounded Gauss linear;
    //div(div(phi,U)) Gauss linear;

writeInterval   0.005;

purgeWrite      0;

writeFormat     ascii;

writePrecision  6;

writeCompression off;

timeFormat      general;

timePrecision   6;

runTimeModifiable true;

adjustTimeStep    yes;

maxCo        1.5;

functions
{
    #includeFunc residuals
    #includeFunc forces
    #includeFunc forceCoefficients
}



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

The fvSolution and fvScheme files can be seen below:

/*--------------------------------*- C++ -*----------------------------------*\
  =========                 |
  \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox
   \\    /   O peration     | Website:  https://openfoam.org
    \\  /    A nd           | Version:  10
     \\/     M anipulation  |
\*---------------------------------------------------------------------------*/
FoamFile
{
    format      ascii;
    class       dictionary;
    object      fvSolution;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

solvers
{
    p
    {
        solver          GAMG;
        smoother        GaussSeidel;
        tolerance       1e-6;
        relTol          0.01;
    }

    pFinal
    {
        $p;
        relTol          0;
    }

    "(rho|U|k|omega|e)"
    {
        solver          PBiCGStab;
        preconditioner  DILU;
        tolerance       1e-6;
        relTol          0.1;
    }

    "(rho|U|k|omega|e)Final"
    {
        $U;
        relTol          0;
    }
}

SIMPLE
{
    residualControl
    {
        p               1e-4;
        U               1e-4;
        "(k|omega|e)"   1e-4;
    }

    nNonOrthogonalCorrectors 0;
}

PIMPLE
{
    nCorrectors              2;
    nNonOrthogonalCorrectors 0;
    nOuterCorrectors         10;
}

relaxationFactors
{
    equations
    {
        ".*"     0.3;
    }
}

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

/*--------------------------------*- C++ -*----------------------------------*\
  =========                 |
  \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox
   \\    /   O peration     | Website:  https://openfoam.org
    \\  /    A nd           | Version:  10
     \\/     M anipulation  |
\*---------------------------------------------------------------------------*/
FoamFile
{
    format      ascii;
    class       dictionary;
    object      fvSchemes;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

ddtSchemes
{
    default         Euler;
}

gradSchemes
{
    default         Gauss linear;

    limited         cellLimited Gauss linear 1;
    grad(U)         $limited;
    grad(k)         $limited;
    grad(omega)     $limited;
}

divSchemes
{
    default             none;

    div(phi,U)          Gauss linearUpwind limited;

    turbulence          Gauss linearUpwind limited;
    energy              Gauss linearUpwind limited;

    div(phi,k)          $turbulence;
    div(phi,omega)      $turbulence;

    div(phi,e)          $energy;
    div(phi,K)          $energy;
    div(phi,Ekp)        $energy;
    div(phi,(p|rho))    $energy;

    div((phi|interpolate(rho)),p) Gauss upwind;

    div(((rho*nuEff)*dev2(T(grad(U))))) Gauss linear;
}

laplacianSchemes
{
    default         Gauss linear corrected;
}

interpolationSchemes
{
    default         linear;
}

snGradSchemes
{
    default         corrected;
}

wallDist
{
    method meshWave;
}

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

I wanted to ask the community what I might be missing. Has anyone run into problems with overestimating drag for airfoil simulations with OF? Any comments, feedback, or suggestions would be much appreciated. Thank you!

[Yann]

Hello,

How is defined your forceCoeffs function? Couldn't it be an issue with the reference frame used to compute the coefficients? (liftDir and dragDir?)

Yann

[kaszt]

It could be, I can definitely look into that. Here is the forceCoefficients and forces definitions:

/*--------------------------------*- C++ -*----------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  3.0.1                                 |
|   \\  /    A nd           | Web:      www.OpenFOAM.org                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       dictionary;
    location    "system";
    object      Force;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
forceCoefficients
    {
        type        forceCoeffs;
        functionObjectLibs  ("libforces.so");
        writeControl   timeStep;
        writeInterval  1;
        patches     ("surfaces");
        
        writeFields no;
        
        log         true;
        patches     ("surface");
        pName       p;
        UName       U;
        dragDir     (1 0 0);
        liftDir     (0 1 0);
        pitchAxis   (0 0 1);
        CofR        (-0.75 0 0);
        magUInf     100; // free stream velocity; inlet condition!!
        lRef        1;
        Aref        1; //0.79; // cross sectional area; should be this instead of 1
    
        rhoName    rhoInf; // of-v301       
        rho        rhoInf; // of-v1706
        rhoInf      1.205; 
        origin      (0 0 0);
    
        coordinateRotation
        {
            type        EulerRotation
            degrees     true;
            rotation    (0 0 0);
        }
        
        binData
        {
            nBin    90; //output data into 20 bins
            direction   (1 0 0); // bin direction
            format  gnuplot;
            cumulative  yes;
        }
    }

/*--------------------------------*- C++ -*----------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  3.0.1                                 |
|   \\  /    A nd           | Web:      www.OpenFOAM.org                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       dictionary;
    location    "system";
    object      Force;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
forces
    {
        type        forces;
        functionObjectLibs  ("libforces.so");
        writetControl   timeStep;
        writeInterval  1;
        patches     ("surface");
        pName       p;
        UName       U;
        rhoName    rhoInf; // of-v301       
        rho        rhoInf; // of-v1706
        rhoInf      1.205;
        log         true;
        CofR        (-0.75 0 0); // origin for moment calculation
        
        writeFields no;
        
        binData
        {
            nBin    90; //output data into 20 bins
            direction   (1 0 0); // bin direction
            format  gnuplot;
            cumulative  yes;
        }
    }


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

[Yann]

How is your reference frame defined?
Is it aligned with the inlet flow and you're moving the airfoil to simulate different aoa, or is it aligned with the airfoil and you're changing the inlet flow to change aoa?

To get proper force coefficients your dragDir and liftDir parameters have to be defined according to the inlet flow direction.

[kaszt]

Thanks for your comment - that makes sense. I'm changing the inlet flow to change the angle of attack, and so the reference frame is coded now to be aligned with the airfoil. I can try changing dragDir and liftDir to be aligned with the inlet velocity then and see if this makes a difference. By chance, have you tried this? Thanks!

[Yann]

Originally Posted by kaszt

By chance, have you tried this? Thanks!

Sure, AFAIK this is always how we compute aerodynamic coefficients on airfoils or aircraft.

[kaszt]

Thanks for your message - yes, that's correct, and I did update things to compute the forces in the body-reference frame for the airfoil. Lift is still reasonable but unfortunately drag is still high. I am curious if anyone else ran into this issue. Thanks!

[Yann]

Hello,

As far as I know on airfoils it is usually harder to get a good agreement on drag than it is on lift.
But of course it depends on how far you are from your reference values.

By the way, have you checked this?
https://doc.openfoam.com/2312/exampl...lent/naca0012/

[Puttanna pujar]

Hello
I'm doing simulation for co flow jet airfoil in which lift coefficient is negative.. can someone help me please!!