[Gesetzt]

Hi,

my old solver works like this:

 while (simple.correctNonOrthogonal())
        {
            fvScalarMatrix TEqn
            (
                        fvm::ddt(T)
                      + fvm::div(phi, T)
                      - fvm::laplacian(DM, T)
                      - T0/T1
                      +T/T1
                     ==
                        fvOptions(T)
            );

I wanted to treat the term T/T1 implicit, because T>0, and T1>0, so it would be negative on the RHS. As far as I have understand, this leads to better stability. Am I right?

So I changed my solver to this:

while (simple.correctNonOrthogonal())
        {
            fvScalarMatrix MzEqn
            (
                        fvm::ddt(T)
                      + fvm::div(phi, T)
                      - fvm::laplacian(DM, T)
                      - T0/T1
                      + fvm::Sp(1/T1,T)
                     ==
                        fvOptions(T)
            );

But I can't see any difference in the results. T1 is a dimensionedScalar and T a volScalarField. How can I implement it in the right way?

Edit: I would like to add, that the problem is, that my scalar T should be bounded. I have a pipe and at the inlet I have a constant value of T=1, but my solver tends to go to 1.1 in the beginning of the pipe, which is not realistic. My thought was, that part of the problem is my source term.

I found out that when I us "bounded Gauss linearUpwind grad(T)" there is no problem. But I think it's no good idea to use the bounded scheme for transient solution?

Best Regards,

Gesetzt

[Gesetzt]

Hi,

I want to give information about my fvSchemes file also:

/*--------------------------------*- C++ -*----------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  2.4.0                                 |
|   \\  /    A nd           | Web:      www.OpenFOAM.org                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       dictionary;
    location    "system";
    object      fvSchemes;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

ddtSchemes
{
    default         Euler;
}

gradSchemes
{
    default         cellLimited Gauss linear 1;
}

divSchemes
{
    default         none;
    div(phi,T)       Gauss linearUpwind grad(T);
   //div(phi,T)       Gauss banana;
   //div(phi,T)      Gauss QUICK;
    //div(phi,T)      Gauss limitedLinear 0.5;
}

laplacianSchemes
{
    default         none;
    laplacian(DM,T) Gauss linear corrected;
    //laplacian(DM,T) Gauss linear limited 1;
}

interpolationSchemes
{
    default         linear;
}

snGradSchemes
{
    default        corrected;
}

fluxRequired
{
    default         no;
    T               ;
}


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

The main problem is, that my bounded scalar is going out of bounds. I thought the problem is maybe the solver, but actually I am not sure about this.
I tried different schemes, but the problem continues. Can anybody give me advice?