[SubSupSonic]

Hi,



Could you please explain how to set up supersonic boundary conditions for p/T/U at the inlet while maintaining subsonic conditions at the outlet?

Thanks in advance

[ishan_ae]

I recently ran a supersoninc inviscid jet case using rhoCentralFoam with vanLeer flux limiter.

It depends upon the problem.

At the inlet I used totalPressure for p, totalTemperature for T and zeroGradient for U. At the outlet you want to impose a pressure difference and for my case I used fixedValue for p set to atmospheric pressure. For T and U at the outlet, I set them as zeroGradient as I was not expecting any kind of backflow.

If there is a subsonic outlet to be imposed, you can either control it using a specified pressure or velocity or even mass flow rate. Perhaps, a fixedValue type for p or U will be helpful.

Good luck.

[SubSupSonic]

Thank you Ishan for your reply.

I have done plenty of tests with a supersonic nozzle (subsonic in converging and supersonic in the diverging) and all my test converged. However, my efforts were without success till now for a supersonic diffuser. I have tried various combinations for the supersonic inlet and subsonic outlet so far. ( (fixedValue for p_inlet & T_inlet & U_inlet & p_outlet) OR (fixedValue for p_inlet & T_inlet & U_inlet & inletOutletTotalTemperature T_outlet) OR (fixedValue for p_inlet & T_inlet & U_inlet & U_outlet) ) OR (fixedValue for p_inlet & T_inlet & U_inlet & inletOutlet U_outlet) etc. ).

One common observation in all the (supersonic diffuser) test cases I have run is that the flow velocity becomes negative, which is problematic. Sometimes, the tendency is for it to go from very negative to 0, giving the impression that the solver "sees" the supersonic diffuser as a supersonic nozzle. Do you have by chance any suggestion?