[L1011]

Hi Foamer!

I've been interested in implementing in openFoam the simple algorithm for the simulation of unsteady flows with a dual time stepping method. Before I start coding any stupid thing, I was wondering if:

1. adding a time derivative in the currant simpleFoam solver is enough to make it become unsteady
2. you could refer to me articles that discuss the dts implementation in incompressible solvers

It'll be interesting to see how an unsteady simple-based solver behave with respect to pisoFoam.

Great thanks!

L1011

[niklas]

...and pimpleFoam is not good enough because?

[nimasam]

i guess so, add fvm::ddt(U) to UEqn then
compare the result of simpleFoam for cavity with pisoFoam

[L1011]

Thanks for your responses!

@Niklas: you're right pimpleFoam already provides the implementations I intended to do. But what about the dual time stepping? In fact what I'm trying to do is, for each physical iteration:

1. move/deform a mesh
2. execute sub-iterations until convergence is reached
3. back to 1

and I'm wondering if the correction loops (nNonOrthogonalCorrectors, nOuterCorrectors, nCorrectors) can be considered as sub-iterations in the same meaning of the dual time step method?

L1011

[L1011]

Any guesses?

I used pimpleFoam to simulate the flow around a NACA0012 airfoil and the following settings:

/*--------------------------------*- C++ -*----------------------------------*\ 
| =========                 |                                                 |
| \      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \    /   O peration     | Version:  1.7.1                                 |
|   \  /    A nd           | Web:      www.OpenFOAM.com                      |
|    \/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       dictionary;
    location    "system";
    object      fvSchemes;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
ddtSchemes
{
    default         backward;
}
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 upwind;
    div(phi,omega)         Gauss upwind;
    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(DomegaEff,omega)          Gauss linear corrected;
}
interpolationSchemes
{
    default         linear;
    interpolate(U)           linear;
}
snGradSchemes
{
    default         corrected;
}
fluxRequired
{
    default         no;
    p       ;  
}


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

/*--------------------------------*- C++ -*----------------------------------*\n| =========                 |                                                 |
| \      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \    /   O peration     | Version:  1.7.1                                 |
|   \  /    A nd           | Web:      www.OpenFOAM.com                      |
|    \/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       dictionary;
    location    "system";
    object      fvSolution;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

solvers
{
    p
    {
        solver          GAMG;
        tolerance       1e-7;
        relTol          0;
     mergeLevels  1;
     smoother  GaussSeidel; 
     agglomerator faceAreaPair ;
     nCellsInCoarsestLevel  100;
     nPreSweeps 0; 
     nPostSweeps 2; 
     cacheAgglomeration true; 
    }

    pFinal
    {
        solver          GAMG;
        tolerance       1e-7;
        relTol          0;
     mergeLevels  1;
     smoother  GaussSeidel; 
     agglomerator faceAreaPair ;
     nCellsInCoarsestLevel  100;
     nPreSweeps 0; 
     nPostSweeps 2; 
     cacheAgglomeration true; 
    }

    U
    {
        solver          smoothSolver;
        tolerance       1e-7;
        relTol          0;
    nSweeps  2;
    smoother  GaussSeidel; 
    }
    UFinal
    {
        solver          smoothSolver;
        tolerance       1e-7;
        relTol          0;
    nSweeps  2;
    smoother  GaussSeidel; 
    }


    k
    {
        solver          smoothSolver;
    smoother GaussSeidel ;
    nSweeps 1 ;
        tolerance        1e-7;
        relTol           0;
    }

    omega
    {
        solver          smoothSolver;
    smoother GaussSeidel ; 
    nSweeps 1 ;
        tolerance        1e-7;
        relTol           0;
    };


    }
    PIMPLE
     {
        nNonOrthogonalCorrectors 4; 
       pRefValue 1;
       pRefCell 0;    
        nOuterCorrectors 10;
        nCorrectors     2;
      };
      relaxationFactors
      {
       default                  1;
        p                  0.5;
        U                  0.5;
        k                  0.5;
        omega                  0.5;
      };
    

}

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

However,I made two strange observations:

1) Even-tough the aerodynamic coefficients get in absolute close to zero, they change significantly from one iteration to the other (see attached file).

2) The velocity field converges from the first outer iteration while the pressure needs couple iteration to get physical.

Do anyone has already experienced something similar?

Thanks in advance!

L1011