Supersonic flow applied to a wedge!
Hi there!!, i am looking fo rthe governing equaitons for a supersonic flow applied to wedges. I have read J.Anderson`s ~Compressible fluid flow` and CFD. But i have doubts. Can you try to solve them.? it will be great. Thanx! for th intrest anyway.If you would like to discuss anything . mail, me. CYA!!:) manihkumarGP

Re: Supersonic flow applied to a wedge!
(1). Most gas dynamics books and fluid dynamics book or handbooks should have the transient, compressible NavierStokes equations listed. (2). For turbulent compressible flow, you will have to look for books on turbulence modeling to get the Reynolds averaged equations, which will have Reynolds stresses terms in it. (3). Then the equations can be written in different forms, in the Cartesian coordinates or in the general transformed coordinates. (4). Beyond that point, one can use FD, FV or FE methods to derive the algebraic governing equations.

Re: Supersonic flow applied to a wedge!
Dea john, thanks! i am trying to prepare a solver for 2d supersonic flow over awedge or for the expansion Prandtl_Meyer equations case, in c,c++or vc++. I have doubt if you can tell me somethign about it. HOw do i find the pbar(i+1,j+1) for the corrector step while introducing artificial viscosity inthe process.
Thanks for the same. I am ref. John Andersons CFD. BYE!!:) 
Re: Supersonic flow applied to a wedge!
(1). I don't have the book by John Anderson. So, it is a little bit hard to answer your questions. (2). The book normally don't include all the details. So, the best way to do is to: dig out related papers from Journals, or find some technical reports published by National laboratories. For example, a NASA report, or a Air Force Lab report would have more detailed steps included. (3). The other approach is to look for PhD dissertations. If it is CFD related, in most cases, the Fortran code is also included. (4). The artificial viscosity terms are included to smooth out the wiggles in the solutions. Sometimes, it is of fourth order, and sometimes it is of second order. For this reason, you need several more nodal points to handle the fourth order terms. (when it is expressed in finite difference form) (5). As a result, near the boundary, you will have to make some assumptions, because you are going to have fewer points there. What I am saying is, for predictorcorrector method, you will have to use say three nodal points. Since a fourthorder term will need more neighboring points, it will extend further out from the local nodal point. This is no problem, if you are several points away from the boundary. Near the boundary, you will have to assume that the artificial viscosity term are somewhat constant there, so you don't have to evaluate it explicitly for the near boundary points. (6). I think, most information are out there, so be patient and try to read more related technical papers.

All times are GMT 4. The time now is 15:25. 