Hi,

I am solving a non-linear system of temperature and species equations (arrehinous terms for chemical reactions included as udfs). The system involves high temperature calculations. The soultion diverges even at under relaxation settings of 0.05 for temperature and species. Is it ok to use such low or even lower values of URs? Are there other ways of ensuring a converged soluton (I have checked denser grids as well).

Manan

Hi,

for species transport due to convection, URF should be 0.05. for heat transfer, it is advisable to use as small URF as possible for enthalpy. you can try even lower URF for enthalpy.

Swarup.

Hi Swarup,

Thanks for the advice. I have tried even lower URFs (0.005) but I still get divergence on reducing thermal conductivity values. Actually I am presently using higher than reality values of thermal conductivity to ensure convergence. But, on reducing K values, I get divergence which I think is due to high level of nonlinearity(exponential rate terms) in the my equations.

Can you or anyone else advice me on how to ensure further convergence.

Manan

Hi Manan,

the only way i can think as of now is that you should linearize your source term. it is likely that this has already been done for you in the FLUENT code. this will be an approximation for actual situation but may work.

How to do it i have little idea as of now and will be able to tell you later.

you may be knowing that a source term can be linearized using the approximation S=Sc+(phi)Sp where Sp should be -ve for stability purposes.

in my opinion, you should consult manual for reaction modeling or relaxation factors for discussion on coupling of enthalpy equation and species transport equation or even better can ask FLUENT support team itself.

Best of luck and regards, Swarup.