[Boitshoko Baloyi]

Hi

I hope you are all well.

I am trying to do a CFD simulation of the flow inside a 2D external compression supersonic inlet that will be used on a ramjet. The simulation captures the oblique shock waves originating from the fore-body perfectly, however, it fails to simulate the normal shock wave that should occur inside the inlet's internal duct. I suspect it has something to do with the back pressure I have applied at the inlet exit. Even though the applied back pressure is high, the pressure of the flow in the inlet is never as high as the applied back pressure.

Does anyone know where I could be going wrong and how to fix this? Your help would be appreciated.

Kind Regards,
Boitshoko

[me3840]

It's been a while since I've done anything compressible, but shouldn't the pressure at that face be completely determined by the upstream conditions?

Is the flow supersonic all the way to your outlet then?

[Boitshoko Baloyi]

Currently, the flow is supersonic through to the outlet, however, the aim is to simulate the flow such that it is subsonic at the outlet.

[wlw1588]

Did you ever figure out what boundary conditions to apply? I'm having a similar issue.

[Tosbaga]

Could you share the related article please ?

[twolf59]

Alright, simulating a supersonic external inlet is not simple. The combination of supersonic external flow, with subsonic internal flow, boundary layer, and potentially highly separated flow means that depending on your flow conditions you may have a highly unsteady case. In my experience, initialization of the flow is extremely important to setting up the shock system correctly. Subsonic flow initialization in the duct is a must. Ideally, you initialize the external flow with some subsonic or transonic mach before running the supersonic case.

I have used both pressure- and density- based solvers. For me, pressure based worked better. Ramping up the CFL number is helpful in stabilizing the shock system. Pressure outlet fan face to start, switching to a mass flow condition only after flow is decently close to convergence.

If the normal shock is within the duct, you are pulling to much mass flow. The normal shock sits right at the entrance of the inlet at the critical mass flow ratio. Sizing your inlet ahead of time is extremely important!

[Alpha_Ashwin007]

Hello daniel, I am actually doing something very similar right now and am facing problems. If possible you could please tell me more about the initialization and the setup?
Thanks