[Alpha_Ashwin007]

Hello,
I'm trying to simulate the flow in a 2-d duct under Supersonic Conditions.
Inlet -
M = 1.7
Total pressure =0.207Mpa
Total Temperature = 310K
Static pressure = 42Kpa
k-w SST model

I have tried using both density and pressure-based solvers. In both cases, the continuity residual is not converging, and I am getting the wrong solution(As compared to experimental results). I am not getting the shock train in the solution.

After a while, the residuals just oscillate slightly and never decrease.
Any advice on solving this would be great.
I have attached a picture of the mesh below.
Thanks

[LJA_Koers]

This is a relatively simple geometry. If you use hybrid initialization, chances are it is already quite close to the final and correct solution. The residuals are scaled to the initial solution (plus a few iterations), so if that initial solution is already quite good, the residual won't decrease much with further iterations, but that doesn't mean the solution is bad. The Fluent user manual discusses this in a section called 'convergence and stability'.

edit: forgot that your results don't match the experiments, I need to think about this some more.

[Alpha_Ashwin007]

Thank you for your reply, Even if the continuity residual doesn't converge, I find other properties and residuals keep slightly oscillating after a point. Should I just let it run for an even longer time?

[LuckyTran]

The boundary conditions don't sound right.

2 MPa total pressure and 42 kPa at an inlet is wayyyy more than a Mach number of 1.7. Is 42 kPa the outlet pressure? What is the supersonic static pressure at the inlet?


The mesh is cool and dense but I can already see a kink line in the grid at the ramp which conveniently is where the high speed flow meets the pressure jump at the outlet.

[Alpha_Ashwin007]

I'm sorry, it was a typo. The total Inlet pressure is 0.207 Mpa and the supersonic static pressure is 42Kpa.

Originally Posted by LuckyTran

The mesh is cool and dense but I can already see a kink line in the grid at the ramp which conveniently is where the high speed flow meets the pressure jump at the outlet.

Any ideas on how to fix this? Thanks

Edit: By playing around with the options, I found that by using the 3rd order MUSCL Scheme with the density based solver with k-w SST, the results seem to be close, but the residuals and properties have a slight oscillation in them. Are these problems due to the mesh?

I have attached some plots for your reference. Thank You