CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Main CFD Forum (https://www.cfd-online.com/Forums/main/)
-   -   2d NS supersonic BC (https://www.cfd-online.com/Forums/main/1242-2d-ns-supersonic-bc.html)

thampy September 2, 1999 13:36

2d NS supersonic BC
 
hi,

Could anyone tell me as to what are the flow parameters that i need to specify as my exit BC when i'm simulating 2d NS supersonic flow ?

thanking you thampy

John C. Chien September 2, 1999 13:58

Re: 2d NS supersonic BC
 
(1). In supersonic flow, the local solution depends on the upstream condition or solution. It is controlled by the characteristics network. (2). So, you can give it a rough number at eixt B.C., it is not going to affect the interior points solution at all, as long as the flow is supersonic.

stephane baralon September 2, 1999 14:26

Re: 2d NS supersonic BC
 
If your flow is axially supersonic, then all characteristics goes out of the domain: you don't need to specify anything.

If your flow has a Y velocity component and also is not axially supersonic(M>1 but Mx<1) then you need to specify one information (one characteristic propagates inside the domain). Usually the pressure.

If your flow has a Y velocity component and also is axially supersonic, then again you don't need to specify anything.


Hongjun Li September 2, 1999 14:56

Re: 2d NS supersonic BC
 
Good point! To make it more general, you may say that if the Mach number component in the direction normal to the boundary surface is great than 1 (Mn > 1), then the exit BCs are not needed. The boundary values can be extrapolated from the upstream grid point. If Mn<1, although Mtotal is greater than 1, the exit boundary is still treated as subsonic. In that case, (if 2-D) three variables should be extrapolated from the upstream, and one variable must be specified at the exit.

clifford bradford September 3, 1999 11:44

Re: 2d NS supersonic BC
 
the above are correct but you have to be careful if you have a boundary layer or wake exiting your domain the flow in that region will/may be subsonicin which case you'll need one exit BC. in which case the pressure is usually appropriate


All times are GMT -4. The time now is 23:07.