Transition and Turbulence Modeling for Blunt-Body Wake Flows

摘要

Aerobraking has been proposed as an efficient means of decelerating spacecraft for planetary missions. Most current aerobrake designs feature a blunt forebody shielding the payload from the intense heat generated during atmospheric entry. Although this forebody will absorb the largest portion of the heat pulse, accurate prediction of heating in the near wake is of great importance, since large local heating values can occur at points of shear-layer impingement. In order to address the various issues associated with these blunt-body wake flowfields, the Advisory Group for Aerospace Research and Development (AGARD) formed Working Group 18 in 1992. One of the objectives of this activity was to examine real-gas effects in high-speed flow fields around a 70 deg. blunted cone. To date, many researchers have conducted experiments using this geometry in various facilities, such as the Large Energy National Shock (LENS) tunnel at Cubric/Calspan and the HEG shock tunnel at DLR-Goettingen. Several computational studies have also been conducted in concert with these tests. Many of the experimental results have indicated the possible presence of a transitional shear layer through a large increase in heat transfer downstream of the reattachment point. The presence of transition could in fact lead to much higher peak heating than if the separated flow is entirely laminar or turbulent. In the shock-tunnel tests, however, it is difficult to separate such viscous-flow phenomena from real-gas effects. In order to help make this distinction, Horvath et al. recently conducted a set of experiments in the NASA Langley 20-Inch Mach 6 Tunnel, and compared the results to laminar Navier-Stokes calculations. They found heat-transfer distributions similar to those obtained in the high-enthalpy facilities, with the measured peak heating along the sting support markedly greater than that predicted by the laminar computations. These trends point to the need to find transitional and turbulent computational solutions for these flowfields.