[sven82]

Hi Foamers,

I work with a cht-Solver based on chtMultiRegionSimpleFOAM
but with buoyant Boussinesq, of cource incompressible and kOmegaSST.
my boundarys are heatflux of a wall.
The heatflux himself is gradient()=q/(Cp*kappaEff) with the kappaEff
generate with nu/Pr+nut/Prt.
Thats all like buoyantBoussinesqSimpleFOAM

After finished the solver, I validate the fluid regions with a cavity case,
also the standart solver for cht and buoyantBoussinesqSimpleFOAM.
The conclusion is all works in the constraints well, in my eyes.

After that I will simulate a LED-System, with a LED-heatflux.
OpenFOAM works well, but the results are ...........

The temperature raise to 15000K ??? , thats a little bit too hot! with a concentration of high tempreature in the heatflux.patch region (200 Iterations but the temprature is raising).
Thats all fine, or not, but the development of the temperature field is really poor so thats the temprature a coupled finite volumes over the heatflux.patch is something like 310K !!!!!!!!!!
The velocity and all other fields are normal.

The heatflux is fine, because I got the same results with groovyBC.

The only problem with the case is that the mesh got no boundary layer,
not the best, but I got no option and CFX works fine with good results.

My question is how con I handle this build up situation of temprature in the heatflux region. At the moment I set nu=10e-3 for a lower heatflux with the
target of a stepwise raise of nu, maybe but .

Got anyone a same problem?
What can I do, ideas?
please post!

Thanks for all clues.

Sven