Vortical regime of the flow behind the bow shock wave (Q5950161)

A new non-stationary regime of the flow around a step and a cylinder is found to exist at high free stream Mach numbers for gas specific heat ratios \(\gamma\) below 1.2. The main features of the flow are strong vortices in the shock-compressed region with supersonic reversal velocities at the body face. It is also that the circulation induced temporary sources could survive in the flow upstream of the body. The vortex in the shock-compressed region influences the shock wave shape and standoff distance, and the flow field takes complex structures. The computations on the finest grid show that turbulence develops in the inviscid flow. This may be a cause of the absence of numerical convergence. The finest mesh resolutions used here is sufficient to capture the flow details, at least for viscous flows. The inviscid flows have general structures similar to the structures of viscous flows, and do not show the tendency of decreasing vorticity. Some numerical schemes, however, have the property of generating an admissible excessive vorticity, and numerical solutions with strong shock waves sometimes show an instability dubbed the ``carbuncle phenomenon''. The carbuncle instability appears as a re-circulation of the flow and a reformation of bow shock wave which is similar to the precursor obtained in the simulations. An exact Riemann solver employing Godunov schemes is used in the circulation and the undisturbed flow at stationary state. The solver does not show an evidence of instability for \(\gamma= 1.1\). The diffusive methods used satisfy Quirk's test, but there high numerical diffusion can suppress the vortical regime of the flow. An exact analytical investigation of the existence of vortical regime is hardly possible. But this regime could be demonstrated experimentally. Since the TVD scheme satisfies Quirk's test for odd-even decoupling, the stability of stationary solution suggests that the vortical regime is not an artifact caused by the carbuncle phenomenon. Relevant experiments can be conducted. Further studies are necessary to investigate the range of existence of vortices and conditions for initiation of vortical regime. The influence of viscosity on the development of vortical regimes could been studied by solving numerical Navier-Stokes equations.