[zoeburton1987]

I am using Fluent to simulate the flow structure in a steam turbine exhaust hood, but am struggling to achieve convergence when I drop the outlet pressure sufficiently to raise the mass flow rate to the 86kg/s I require. I believe the problem lies with the unusual contours I am finding at the pressure inlet boundary - but I do not fully understand how Fluent applies conditions at the boundary (despite reading the manual) to diagnose the problem.

At my pressure inlet plane I have applied representatively profiles of total temperature (absolute as my Operating Condition = 0), total temperature and axial, radial and tangential velocity using the cylindrical coordinate system. The initial/supersonic gauge pressure was lowered to 9362Pa as the inlet flow became supersonic (due to instant divergence from what was originally a stable solution).

At pressure outlet boundary plane the static pressure is determined by trial and improvement to achieve the desired mass flow rate, initially set at 9900Pa to give a subsonic flow for the early stages of the calculation.

The fluid is set as water vapor, and density as an ideal gas. I am using a pressure based solver and discretization is all 2nd order. After divergence issues, the pressure-velocity coupling was changed from SIMPLE to coupled, which enable me to drop the outlet pressure to 9300Pa and raise the mass flow rate to the desired 86kg/s.

However, the solution is unstable, with residuals oscillating (see attached figure) when I drop the outlet pressure after 5250 iterations. I believe this stems from the supersonic flow at inlet (see attached figures). The total pressure profile, is accurate to what I applied at inlet. I would expect supersonic flow on the underside of the inlet plane, but I don't understand the static pressure contour, and how this relate to the mach number contours.

I suspect this all stems from my lack of understanding of the pressure inlet boundary condition - can anyone suggest any ideas of where this might source from?

I can obviously provide more information, if needed.

Any help would be GREATLY appreciated. This has been many many week's work and I have hit a dead end.