[ookalkan]

Hi everyone,

I am trying simulate the fluid flow and heat transfer in a convergent divergent nozzle in both Fluent and CFX. My aim is to validate the codes by comparing the results with an experiment.

I got satisfactory results by Fluent, however I have some problems with CFX. Here is the problem:

In this case, nozzle inlet total pressure is 75.2 psia, inlet total temperature is 843K and exit Mach number is 2.5. At that pressure, it is indicated by the experiment that there is a flow separation at the nozzle outlet. But, if I specify the outlet boundary condition as "Supersonic Outlet", CFX does not allow reversed flow, i.e. separation. Therefore, i have used "Opening" boundary condition at the outlet in order to let the flow turn back. This time I see quite weird temperature contours near the outlet. Temperature drops upstream of the throat,as supposed, but then it increases again near the outlet. I am suspicious that this is because of the "Opening Temperature" specification. But I am not sure.

So I wonder how can I simulate the separation near the nozzle outlet in CFX?
Which boundary condition should I use for outlet?

Best regards,
Kalkan.