Performance of three-dimensional compressible Navier-Stokes codes at low Mach numbers

The low-Mach-number performance of two widely used three dimensional compressible Navier-Stokes solvers is examined. Both of the codes employ a finite volume formulation, but differ in their respective discretization and solution schemes. In the first code, a central-difference discretization is employed and solutions are obtained using an explicit Runge-Kutta time-stepping scheme, In the second code, an upwind-biased discretization is used, aad an implicit approximate factorization scheme is employed. The accuracy and efficiency of the codes are assessed by comparison with results from an incompressible Navier-Stokes solver and experimental data. The test eases examined include an untwisted reef angular wing and a wing-fuseIage configuration. The computed pressure distributions were nearly identical, and agreed quite well with experimental data. In addition, the predicted pressure distributions were nearly identical to those predicted by the incompressible how-solver. The influence of freestream Mach number on the efficiency of both compressible-flow solvers also is examined.