Preconditioning in 3D

ganesh

Dear Friends,

I am trying out low speed preconditioning in 3D. I am using a preconditioned Roe scheme along with Weiss-Smith matrix for solving the problem. However, I am facing difficulties in convergence/oscillatory(badly!) contours of pressure/velocity. My test case is the inviscid flow past an ONERA M6 wing. The code with preconditioning goes through for higher Mach numbers, with good results, but problems begin at lower mach. The grid is relatively coarse. I believe that the problem is in definition of Uref. Can anyone with experience in 3D preconditioning throw some light on the issue? The 2d preconditioned solver derived from this 3d solver works absolutely fine for mach numbers as low as 1e-4.

All suggestions are welcome

Thanks in advance, Regards,

Ganesh