[Sereff]

Hi FOAMers:

I am dealling with solving a fvScalarMatrix equation that is without fvm::laplacian() term, or have extremely low diffusivity. And this leads to "solution singularity" problem. On the other hand, change the fvScalarMatrix of TEqn in buoyanBoussinesqPimpleFoam to be without fvm::laplacian() term, the temperature field can be solved without any error. I chosed the same solver and divScheme for both of them, but the different result still remains. please help

fvSolution:

    "(T|s)"
    {
        solver          PBiCGStab;
        preconditioner  DILU;
        tolerance       1e-6;
        relTol          1e-3;
    }

    "(T|s)Final"
    {
        $T;
        relTol          0;
    }

fvSchemes:

divSchemes
{
    default         Gauss linear;

    div(phi,U)      Gauss LUST grad(U);
    div(phi,k)      Gauss limitedLinear 1;
    div(phi,epsilon) Gauss linear;

    div(phi,T)      Gauss upwind;
    div(phi,s)     Gauss upwind;

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