[wc3speler]

Dear all,

I'm trying to verify the externalWallHeatFluxTemperature BC on a (geometrically) simple case using buoyantBoussinesqSimpleFoam in OF30x. The domain is a box of 2x2x2 m and heat is injected via the 'floor' boundary of T:

floor
{
    type            groovyBC;
    value           uniform 293;
    valueExpression "293";
    gradientExpression "gradT";
    fractionExpression "0";
    variables       
    (
        "heatFlux=1;"
        "rho=1.225;"
        "Cp=1000;"
        "myNu=1.0e-5;"
        "Pr=0.9;"
        "alpha=myNu/Pr;"
        "gradT=heatFlux/((alpha+alphat)*rho*Cp);"
    );
}

i.e. 1 W/m2 * 4m2 gives me 4 Watt (in controlDict this is monitored). At the top of the domain the temperature BC is:

ceiling
{
    type            externalWallHeatFluxTemperature;
    kappa           lookup;   
    Ta              uniform 273; // external air temperature 
    h               uniform 2; // external heat transfer coefficient
    // q uniform 1000;  
    //thicknessLayers (0.0); // thickness wall between in and outside
    //kappaLayers     (0.01); // equivalent heat conductivity wall
    kappaName       alphaEff;
    value           uniform 273; // initial wall temperature      
}

Without the kappaLayers it's like a zero-thickness wall with just heat transfer between the wall and the outside domain (temperature Ta). So:

Q = h*(T_wall - T_outside)*Area

Q = 4W, area = 4m2 and h = 2 W/(m2*K), so I would expect a 0.5 degrees temperaturejump between the wall and the outside, meaning T_wall=273.5 K. It turns out to be around 274.6 K, see the attached image.

It would be great if someone could comment on this discrepancy (I have attached the case) and/or share experience using this BC.

Thanks!

[wc3speler]

Woops, no. That's near the wall. The temperature of the wall itself is 273K. Hmmm...