[bbjolle]

Hello everybody,

as a preliminary work I am performing a validation of chtMultiRegionSimpleFoam for a simple nonlinear heat conduction problem. M. Danish et. al. developed an analytical solution for the 1D steady state heat conduction in a rod with linear varying heat conductivity.
http://www.iaeng.org/publication/WCE...p2550-2555.pdf

I face a problem that I need much more cells with the chtMRSF solver to predict the temperature profile and the wallHeatFlux as can be seen in this figure.
nonLinearHeatConduction.png

With ANSYS Fluent I am able to predict the wallHeatFlux quite good with even three cells. To get comparable results with chtMRSF I need 160 cells.


Are there any improvements possible in my settings?


I am using OF 5.0 and the case is based on the multiRegionHeaterRadiation tutorial. BottomAir is only a dummy, with no influence on the solid heater region. I have implemented the linear varying heat conductivity like this

thermoType
{
...
    transport       polynomial;
    thermo          hPolynomial;
...
}
mixture
{
...

    transport
    {
        kappaCoeffs<8>     (-58.8 0.2 0 0 0 0 0 0);
    }

    thermodynamics
    {
        Hf      0;
        Sf      0;
        CpCoeffs<8>     ( 1 0 0 0 0 0 0 0 );
    }
...

}

nonLinearHeatConduction.zip


regards


Björn