Supersonic diffuser flow - Shock system

cyln · September 23, 2019, 22:56

Hello,

I am simulating a steady 3D supersonic-diffuser flow with CFX. The supersonic diffuser has a test section upstream with M=4. Inlet BC is set to Ptot = 8.3 bars, Outlet is set to P_static = 1 bar. I expect to have a shock termination somewhere downstream of the diffuser throat.

k-omega SST turbulence model is used for turbulence closure.

When CFX starts to capture the shock system, there is flow separation downstream of the shock system and I am not sure whether it is real or not. the total pressure downstream of the shock system falls below 1 bar (about 0.5 bar). Since this is less that the outlet pressure, the solver places a wall at the outlet to avoid inflow from the outlet. This way, mass is not conserved and this leads to unstarting (The shock system moves upstream of the throat).

Is there anyone who has dealt with this problem? Which turbulence model is ideal for such a flow? Any other way to treat the outlet?

Cheers,
Cyln

Gert-Jan · September 24, 2019, 02:41

This is a very common phenomena. You have to place your outlet further downstream.

cyln · September 28, 2019, 18:35

I tried placing the outlet way downstream, however, it did not make any change. It unstarts itself (the shock moves upstream of the throat). I tried three different theoretically correct geometric settings, still same result.

Is there any paper/reference that modelled a supersonic diffuser using CFX? Or anybody who did that successfully?

Cheers

September 23, 2019, 22:56	Supersonic diffuser flow - Shock system	#1
cyln Senior Member cyln Join Date: Jul 2016 Posts: 102 Rep Power: 9	Hello, I am simulating a steady 3D supersonic-diffuser flow with CFX. The supersonic diffuser has a test section upstream with M=4. Inlet BC is set to Ptot = 8.3 bars, Outlet is set to P_static = 1 bar. I expect to have a shock termination somewhere downstream of the diffuser throat. k-omega SST turbulence model is used for turbulence closure. When CFX starts to capture the shock system, there is flow separation downstream of the shock system and I am not sure whether it is real or not. the total pressure downstream of the shock system falls below 1 bar (about 0.5 bar). Since this is less that the outlet pressure, the solver places a wall at the outlet to avoid inflow from the outlet. This way, mass is not conserved and this leads to unstarting (The shock system moves upstream of the throat). Is there anyone who has dealt with this problem? Which turbulence model is ideal for such a flow? Any other way to treat the outlet? Cheers, Cyln

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Supersonic Flow Modeling - internal	mahbood	FLUENT	0	October 25, 2016 17:59
[How to obtain supersonic flow inside a supersonic wind tunnel ?]	yx213	Siemens	1	September 17, 2014 13:52
Supersonic flow over an airfoil. FLUENT Lift don't match up with Shock Expansion Tech	enriccasas	FLUENT	1	August 21, 2013 06:43
Nozzle Flow with shock	jim clancy	FLUENT	1	December 18, 2001 04:19
Inviscid Drag at subsonic, subcritical Mach #	Axel Rohde	Main CFD Forum	1	November 19, 2001 12:19

September 24, 2019, 02:41		#2
Gert-Jan Senior Member Gert-Jan Join Date: Oct 2012 Location: Europe Posts: 1,827 Rep Power: 27	This is a very common phenomena. You have to place your outlet further downstream.

September 28, 2019, 18:35		#3
cyln Senior Member cyln Join Date: Jul 2016 Posts: 102 Rep Power: 9	I tried placing the outlet way downstream, however, it did not make any change. It unstarts itself (the shock moves upstream of the throat). I tried three different theoretically correct geometric settings, still same result. Is there any paper/reference that modelled a supersonic diffuser using CFX? Or anybody who did that successfully? Cheers