[S03r3n]

Hello together,

I have the issue that my simulation (rhoSimpleFoam, steady-state) shows good convergence for velocities U and pressure p but poor convergence for energy e.
What can be the reason? Is it solver related?

I have attached the convergence plot.

Below you can find the thermoType I'm using and the chosen solver settings.

thermophysicalProperties

thermoType
{
    type            heRhoThermo;
    mixture         pureMixture;
    transport       const;
    thermo          hConst;
    equationOfState rhoConst;
    specie          specie;
    energy          sensibleInternalEnergy;
}

mixture
{
    specie
    {
        nMoles      1;
        molWeight   600000;
    }
    thermodynamics
    {
        Cp          1850;
        Hf          0;
    }
    transport
    {
        mu          40000;
        Pr          370000000;
    }
    equationOfState
    {
        rho         1154;
    } 
}

controlDict

application     rhoSimpleFoam;

startFrom       startTime;

startTime       0;

stopAt          endTime;

endTime         300;

deltaT          1;

writeControl    timeStep;

writeInterval   100;

purgeWrite      1;

writeFormat     ascii;

writePrecision  6;

writeCompression off;

timeFormat      general;

timePrecision   6;

runTimeModifiable true;

fvSolution

solvers
{
    p
    {
        solver          GAMG;
        tolerance       1e-10;
        relTol          0.1;
        smoother        GaussSeidel;
    }

    "(U|e)"
    {
        solver          smoothSolver;
        smoother        GaussSeidel;
        tolerance       1e-10;
        relTol          0.1;
    }
}

SIMPLE
{
    nNonOrthogonalCorrectors 2;
}

relaxationFactors
{
    fields
    {
        p               0.3;
        rho             0.3;   
    }
    equations
    {
        p               0.3;
        U               0.7;
        e               0.7;
    }
}

fvSchemes

ddtSchemes
{
    default         steadyState;
}

gradSchemes
{
    default         Gauss linear;
}

divSchemes
{
    default         none;
    div(phi,U)      bounded Gauss linearUpwind grad(U);
    div(((rho*nuEff)*dev2(T(grad(U)))))      Gauss linear; 
    
    div(phi,e)          bounded Gauss linearUpwind grad(e)

    div(phid,p)         Gauss upwind;
    div(phi,Ekp)        bounded Gauss upwind;
    div((phi|interpolate(rho)),p)  Gauss upwind;  
}

laplacianSchemes
{
    default         Gauss linear limited 0.33;
}

interpolationSchemes
{
    default         linear;
}

snGradSchemes
{
    default         limited 0.33;
}

[Alczem]

Hey!


Your thermophysicalProperties values seem weird to me, what kind of fluid has such molecular weight, viscosity and prandtl? Are you sure about these properties?


Apart from that, you could maybe add a limiter to the gradient, but everything looks ok to me

[S03r3n]

Originally Posted by Alczem

Hey!


Your thermophysicalProperties values seem weird to me, what kind of fluid has such molecular weight, viscosity and prandtl? Are you sure about these properties?


Apart from that, you could maybe add a limiter to the gradient, but everything looks ok to me

Thanks for your message.
I try to simulate uncured rubber.
Do you have some suggestions?

[Tobermory]

Just reiterating Alczem's earlier post ... lead has a molecular weight of 207 kg/kmol, and you are trying to apply 6x10^5 kg/kmol! Your density is approx the density of lead and your viscosity is off the scale. None of this seems very physical (although I appreciate that uncured rubber might be very viscous).

So again - as per Alczem's post, can I also suggest you check your inputs.

[S03r3n]

I found out that the convergence behavior of energy e highly depends on Prandtl number, thermal conductivity respectively.

I have changed the transport model to polynomial in order to specify the thermal conductivity (kappa).

However, I get bad e-convergence for this case:

    transport
    {
        muCoeffs<8>     ( 15000.0 0 0 0 0 0 0 0 );
        kappaCoeffs<8>  ( 0.15 0 0 0 0 0 0 0 );
    }

And good e-convergence for this case:

    transport
    {
        muCoeffs<8>     ( 15000.0 0 0 0 0 0 0 0 );
        kappaCoeffs<8>  ( 150 0 0 0 0 0 0 0 );
    }

I thought that the unit of kappa should be W/(m*K), so the bad case should have the correct value of 0.15 W/(m*K)?
Any suggestions?

Best regards