Some problems with Star CD

Micha

Hello, I'm simulation conjugate heat transfer in a modell of a fuel cell. The grid was generated with Icem-CFD and is quite comlicated.

I found some problems during working with this modell and Star CD. Maybe somebody has these problems,too.

1. Problems with conjugate heat transfer, radiation and load step definition as a temperature ramp at inlet of gas in a transient calculation: This doesn't work. Although all my boundaries are well defined, STAR doesn't start saying that I would have boundary regions defined to zero. With load step definition as step and exactly the same settings, STAR doesn't make problems. Probably a bug??!!!

2. Transient Calculation with conjugate heat transfer and radiation, load steps defined as steps because of 1.): I defined an adiabat wall with all radiation coefficients=0. Although the overall temperature solution field looks ok, there are hot unphysical spots in the cell row directly near that wall. I absolutely don't understand why. Maybe someone has an idea?!

3. Transient Calculation with conjugate heat transfer, no radiation, temperature dependant material properties defined in user subroutines conduc.f, porcon.f and specht.f (also for porous regions): I get unphysical hot spots (extrema) once again. Their location is in the porous media. I tried to take a smaller time step, it was even worse. My differencing scheme is SFCD. I also created a simple grid modell of the fuel cell in which exactly the same material definitions are defined. It's just not such a complicated grid. In this modell, the solution is absolutely fine. I can't understand this.

Maybe somebody of you has the same problems and maybe can even help me...

Thanks in advance. Michael