[emab8]

Hi everybody!

I am simulating a 2D supersonic intake in Ansys FLUENT with two ramps for external compression (approximate values: delta1 = 7°, delta2 = 8.5°) and one ramp for internal compression (delta3 = 8°). The values were set using the Oswatisch criterion for maximum pressure recovery. For the subsequent section of the intake, I tested two main configurations: a divergent section at 2.5° and in the second case, a straight duct. I’m attaching images showing the analysis performed with a pressure far-field boundary condition on all external sides of the domain (M=2, p=0.26 atm, T=223 K), a wall boundary condition for the intake, and a pressure outlet (p depending on the duct, Ttot = 402 K). Pressure and temperature are calculated at 10000 meters height. In particular, the pressure outlet BCis entirely ineffective in my analyses, as the final section is in supersonic flow, meaning that information does not propagate upstream.

For the calculation, I configured Fluent with density-based, steady-state settings, energy equation active, inviscid flow; CFL is set to a base value of 5. For the solution, I’m using the implicit formulation, flux type: AUSM; spatial discretization: gradient Least Squares Cell Based; flow: Second Order Upwind. Initialization is performed on the far-field. I’m attaching images of the results for the two cases mentioned above.

I don’t know what isn’t working; the normal shock doesn’t position itself where it should. Increasing the number of iterations, the normal shock gradually shifts out of the intake.

I have already read quite a few posts here and many articles online, but aside from a few details, they haven’t helped me much in solving the problem. I also tried copying geometries from other articles, but they still didn’t give me the desired results. What can I do?