Simulation of supersonic flows
 

Hi guys,

my simulation is about a flow of about Mach 6 to 7 over a ramp oder other arbitrary geometries. It is a transient simulation. Inlet speed, pressure and temperatures vary as a function of time.

1. I already had problem to run a steady-state calculation of Mach 7. My pressure here is just about 10 Pa. Temperature about 230K. I finally managed that somehow.

2. I took the steady-state, changed it to a transient simulation but it just won't work. It calculates the first timestep with a wierd number at P-Mass and at second timestep it already breaks down showing F at P-Mass. Courant number is above 999 (I know it is not a parameter for transient simulations).

I checked tutorials and tried everything. Changing Advection Schemes, chaning inlet turbulence setting from low (1%) to the other one. I also tried an adaptive timestep setting holding the Courant Number below 0.5 but first my timestep gets very very small and second it fails anyway after few timesteps. My mesh also seems to be alright.

What am I doing wrong? Whats the basic settings for transient supersonic problems? I somehow have the feeling it has to do with the pressure level I use, but my problems environment has a low pressure level starting with 10 Pa going up to pressure at sea level at end of simulation time. What pressures values do I need to set to simulate this? Maybe a high value for relative pressures?

Thank you very much for taking your time and your response! Greetings, Abduri

Re: Simulation of supersonic flows
 

Dear Abduri,

I think the CFX do not handle the supersonic flow correctly.

Regards

MOP

Re: Simulation of supersonic flows
 

Dear Abduri,

I think the CFX do not handle the supersonic flow correctly.

Regards

MP

Re: Simulation of supersonic flows
 

Try tweaking with some solver options designed to deal with instabilities specifically associated with high speed flow. Turn on high speed numerics (switches on Carbuncle fix etc.), and increase the number of continuity loops performed (expert parameter "mass continuity loops = n". Furthermore try to get better initial conditions by increasing the velocity stepwise and restart with the results/

Re: Simulation of supersonic flows
 

Rubbish. CFX can do supersonic flows. They can be tricky to get right - Timon has some good suggestions.

Glenn Horrocks

Re: Simulation of supersonic flows
 

Hi Guys,

thank you for your answers. It helped. Changed some settings and especially the mesh and it all works now.

Cheers, Abduri