[mikejaeger]

Hi People,


I try to get along with the viewFactor Radiation model. I studied the Theory and than built up an OpenFOAM case. The Simulation runs good but I still have some open Questions which I can't answer with literature.

The viewfactor-model calculates the Parameter "qr", but when I turn radiation off, the temperature Distribution on a zeroGradient wall does not change. So the temperature field is always the same, it doens't matter if Radiation is turned on or off.

In General the results are reasonable in Terms of viewFactorfields, faceAgglomeration (and residuals etc...) except the fact that "T" does not Change with Radiation on.

I used the hotRadiationRoom-tutorial und modified it to be a viewFactor-case. Solver: buoyantSimpleFoam. I also tried buoyantBoussinesqSimpleFoam. Same results.

When I look in the ".H"-files of the solvers, I see that both solvers consider a "radiation-term" in the Energy equation. So I am confused why the temperature does not change.

Do I have to calculate the actual temperature considering Radiation within paraview?

Thank you in advance

[peterhess]

CHT +viewfactor model

[mikejaeger]

Thank you. I already read this thread. This is not exactly my case, because I wonder why the temperature is not changing at a wall with "zeroGradient" boundary condition.

If I use "fixedValue" that is clear, that the wall temperature will not change if radiation is turned on.

But at a "zeroGradient" wall it has to? Or am I wrong?

[peterhess]

Post the case please!

[mikejaeger]

I copied the "hotRadiationRoom"-tutorial, which is actually a fvDOM-radiation-tutorial and I changed it to a viewFactorcase (see CaseSetup.png).


First I execute "faceAgglomerate"

Create time

Create mesh for time = 0


Agglomerating patch : box

Agglomerating patch : ceiling

Agglomerating patch : floor

Agglomerating patch : fixedWalls

Writing facesAgglomeration
End

Then execute "viewFactorsGen"

Create time

Create mesh for time = 0


Total number of coarse faces: 70

Calculating view factors...
Writing view factor matrix...
End

And then "facesAgglomeration" and "viewFactorField" in the 0-directory is created (with reasonable values).

In the constant Directory radiationProperties, boundaryRadiationProperties and viewFactorsDict is located:

FoamFile
{
    version     2.0;
    format      ascii;
    class       dictionary;
    location    "constant";
    object      boundaryRadiationProperties;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

floor
{
    type            lookup;
    emissivity      0.95;
}

ceiling
{
    type            lookup;
    emissivity      0.95;
}


box
{
    type            lookup;
    emissivity      0.95;
}

fixedWalls
{
    type            lookup;
    emissivity      0.95;
}

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

FoamFile
{
    version     2.0;
    format      ascii;
    class       dictionary;
    location    "constant";
    object      radiationProperties;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

radiation       on;

radiationModel  viewFactor;

viewFactorCoeffs
{
    smoothing true; //Smooth view factor matrix (use when in a close surface
                    //to force Sum(Fij = 1)
    constantEmissivity true; //constant emissivity on surfaces.
}

// Number of flow iterations per radiation iteration
solverFreq 1;

absorptionEmissionModel constantAbsorptionEmission;

constantAbsorptionEmissionCoeffs
{
    absorptivity    absorptivity    [0 -1 0 0 0 0 0] 0.00;
    emissivity      emissivity      [0 -1 0 0 0 0 0] 0.00;
    E               E               [1 -1 -3 0 0 0 0] 0.00;
}

scatterModel    none;

sootModel       none;

transmissivityModel none;


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

FoamFile
{
    version     2.0;
    format      ascii;
    class       dictionary;
    object      viewFactorsDict;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

//maxDynListLength          1000000;
writeViewFactorMatrix     true;
writeFacesAgglomeration   true;
writePatchViewFactors     false;

box
{   
    nFacesInCoarsestLevel        20;
    featureAngle                 10; 
}

ceiling
{
 
    nFacesInCoarsestLevel     20;
    featureAngle              10;
}

floor
{
    
    nFacesInCoarsestLevel     20;
    featureAngle              10;
}

fixedWalls
{
    
    nFacesInCoarsestLevel     20;
    featureAngle              10;
}


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

When executing the simulation with buoyantSimpleFoam the residuals are in an acceptable range and there is the notice that the viewFactors-equations are solved

…
Time = 1000

DILUPBiCGStab:  Solving for Ux, Initial residual = 0.000665996, Final residual = 1.10894e-06, No Iterations 1
DILUPBiCGStab:  Solving for Uy, Initial residual = 0.00107227, Final residual = 2.26247e-06, No Iterations 1
DILUPBiCGStab:  Solving for Uz, Initial residual = 0.00145211, Final residual = 3.13363e-06, No Iterations 1
DILUPBiCGStab:  Solving for h, Initial residual = 0.000692169, Final residual = 1.15632e-06, No Iterations 1
Solving view factor equations for band : 0
DICPCG:  Solving for p_rgh, Initial residual = 0.00298525, Final residual = 2.88191e-05, No Iterations 51
time step continuity errors : sum local = 1.15563e-06, global = -1.19967e-19, cumulative = -5.61596e-18
rho min/max : 0.696266 1.16051
DILUPBiCGStab:  Solving for epsilon, Initial residual = 0.000605439, Final residual = 1.35487e-05, No Iterations 1
DILUPBiCGStab:  Solving for k, Initial residual = 0.00163685, Final residual = 1.98886e-05, No Iterations 1
ExecutionTime = 295.15 s  ClockTime = 295 s
...

The results after 1000 iterations is seen in Results.png. It doesn't matter if Radiation is turned on or off. The maximal temperature at the fixedWalls is Always 303.281 K (after 1000 iterations). In my point of view the temperature at the zeroGradient walls must change if radiation is turned on or off. Or am I wrong?

Regarding to the viewFactor-equations the heat flux "qr" at the wall is calculated. Obviously this heat flux is not considered in the energy equation of the solver. What I am doing wrong?

Maybe the boundary condition for the temperature T at the wall must be another and not "zeroGradient"? In that case I didn't find an appropriate so far.


Best regards, Mike

[YannGuyot]

Hello,
Did you check this thread :

Heated wall by radiation

[Chris T]

I know it is a pretty old post but I stumble upon it, as I myself, trying to figure out the view factor model in version 9 of OpenFoam.
In your radiationProperties file you have

constantAbsorptionEmissionCoeffs
{
    absorptivity    absorptivity    [0 -1 0 0 0 0 0] 0.00;
    emissivity      emissivity      [0 -1 0 0 0 0 0] 0.00;
    E               E               [1 -1 -3 0 0 0 0] 0.00;
}

maybe all those zeros are the reason your radiation model is ineffective. If I understand correctly with emissivity and absorptivity 0 your surfaces are perfect mirrors with no emission or absorption.

I see that you specify emissivity in boundaryRadiationProperties file. I my case I don't even have such a file and I specify emissivity in the 0/qr file.

Chris


Edit: Never-mind. The fields I mentioned above are only relevant for opaque regions.