In last years, I have developped CFD flowfield analysis software using JST scheme which is suitful for transonic speed , now I want to extend it to simulate the supersonic flows and I find it does not work well , so I want to implement AUSM scheme, but I found it is difficult to deal with the stagnent point condition and the iteration suddenly interruptted by Nan , So my question is : 1) Is Central Scheme suit for supersonic ( Ma = 4 ) flow calculation? 2) How about AUSM scheme (espacially about its robust)? 3)Is there any special treatment about the AUSM scheme near the stagnent point ?

Hi,

The inventor of AUSM Men-sing Lou (spelling might be incorrect) had posted a message here a few days back. You should send him email. I think there have been lots of improvements over the basic AUSM, by the Dr. Lou and also others.

ciao, chidu...

Thanks for your advice and I will send e-mail to Dr Liu

Boling,

It is not fair to say that the JST scheme works for the transonic range but not ofr supersonic flows. I know personally that the JST scheme works well even for hypersonic Mach numbers (perfect gas) with of course some tuning of the AD term coefficients. Please look at the literature carefull especially Radespiel & others.

The more likely problem in your case is the implememntation of boundary conditions in the supersonic case. Implementation of AUSM or AUSM+ is not very straightforward. This is not to discourage you from doing so, but if you have problems making the JST scheme work in cases it should work, you may find that AUSM experience might be similar.

Ravichandran

Hello

JST scheme is fine for supersonic flow. AUSM scheme is very nice at the stagnation point.

I think your failure may caused by another reason. Please check grid quality near the stagnation point.

If the fighly distorte mesh, or grid lines are crossed, then any scheme can produce unphysical value, such as negative density .. Then NAN(Not a number) will appear on screen.

Or

Print out intermediate values in your program, and find the place where physics break down.

T. Matsuzawa.

[aleem.engr]

Originally Posted by T. Matsuzawa
;11097

Hello

JST scheme is fine for supersonic flow. AUSM scheme is very nice at the stagnation point.

I think your failure may caused by another reason. Please check grid quality near the stagnation point.

If the fighly distorte mesh, or grid lines are crossed, then any scheme can produce unphysical value, such as negative density .. Then NAN(Not a number) will appear on screen.

Or

Print out intermediate values in your program, and find the place where physics break down.

T. Matsuzawa.

hi,
sir i am trying to develop 2d euler code for ausm scheme but i had a problem of convergance can u help me out where i am wrong

[yzeyue]

can you help me with my ausm code? Thanks very much.
http://www.cfd-online.com/Forums/mai...tml#post579860

[yzeyue]

Is your question solved? Can you see my code and tell me the mistake? Appreciate it very much.
http://www.cfd-online.com/Forums/mai...tml#post579860