steady state compressible flow courant no

victoryv · February 1, 2013, 19:22

I am simulating flow through a wind tunnel. I see oscillations in my simulations after residuals are lowered to 10 ^-4.
I am using second order upwind scheme. I am also using kw SST model.I have tried reducing relaxation factors but it was no use. Also when I refined the grid(33M cells), the solution diverged after few iterations.

is the problem due to courant no or is there any other reason?

What should be the CFL no for steady state compressible flow ?
Do we have any general guidelines wrt to cell size and velocity ?

msaeedsadeghi · February 3, 2013, 05:24

You may should use a finer mesh or decrease URFs.

victoryv · February 3, 2013, 18:09

msaeedsadeghi ,

I have tried reducing URF's and refining the grid, there was no use.

msaeedsadeghi · February 4, 2013, 00:37

It should be investigated carefully.

victoryv · February 5, 2013, 18:09

When I used high CFL no and reduced the URF, the oscillations reduced but were still there. Then the switched from Least square method to Green Gauss method, the oscillations died but the residuals were same 10^-3.
I also monitored other parameters. These parameters were constant after certain iterations. (If that means did they reach steady state? ) .

I guess that for steady state compressible flow we need to monitor other variables other than residuals to see for convergence.

Correct me if I am wrong.

Also, can someone tell me , the best method (Least square v/s Green Gauss node V/S Green Gauss cell ) to use for compressible flow for modelling external flow?

victoryv · February 9, 2013, 21:57

Can someone please help me?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Compressible flow, no data at the outlet	mireis	FLUENT	6	September 3, 2015 03:10
Has anyone EVER managed to run: compressible steady state subsonic internal flow	mihaipruna	OpenFOAM Running, Solving & CFD	2	November 9, 2012 12:23
Steady state	Ruben	Main CFD Forum	43	May 7, 2011 04:32
Prism layer advice needed for steady state port flow case	MaxCFM	Main CFD Forum	1	October 1, 2009 10:26
Steady state transsonic turbulent flow	kvick	OpenFOAM Running, Solving & CFD	2	February 12, 2009 05:36

February 3, 2013, 05:24		#2
msaeedsadeghi Senior Member SSL Join Date: Oct 2012 Posts: 226 Rep Power: 15	You may should use a finer mesh or decrease URFs.

February 3, 2013, 18:09		#3
victoryv Member Join Date: Sep 2012 Location: FL Posts: 79 Rep Power: 14	msaeedsadeghi , I have tried reducing URF's and refining the grid, there was no use.

February 4, 2013, 00:37		#4
msaeedsadeghi Senior Member SSL Join Date: Oct 2012 Posts: 226 Rep Power: 15	It should be investigated carefully.

February 5, 2013, 18:09		#5
victoryv Member Join Date: Sep 2012 Location: FL Posts: 79 Rep Power: 14	When I used high CFL no and reduced the URF, the oscillations reduced but were still there. Then the switched from Least square method to Green Gauss method, the oscillations died but the residuals were same 10^-3. I also monitored other parameters. These parameters were constant after certain iterations. (If that means did they reach steady state? ) . I guess that for steady state compressible flow we need to monitor other variables other than residuals to see for convergence. Correct me if I am wrong. Also, can someone tell me , the best method (Least square v/s Green Gauss node V/S Green Gauss cell ) to use for compressible flow for modelling external flow?

February 9, 2013, 21:57		#6
victoryv Member Join Date: Sep 2012 Location: FL Posts: 79 Rep Power: 14	Can someone please help me?