No abstract

Numerical simulations of the near wake of generic rocket con¦gurations are performed at transonic and supersonic freestream conditions to improve the understanding of the highly intricate near wake structures. The Reynolds number in both §ow regimes is 10 6 based on the main body diameter, i. e., speci¦c freestream conditions of ESA£s Ariane launcher trajectory. The geometry matches models used in experiments in the framework of the German Transregional Collaborative Research Center TRR40. Both axisymmetric wind tunnel models possess cylindrical sting supports, representing a nozzle to allow investigations of a less disturbed wake §ow. A zonal approach consisting of a Reynolds averaged NavierStokes (RANS) and a large-eddy simulation (LES) is applied. It is shown that the highly unsteady transonic wake §ow at Ma ∞ = 0.7 is characterized by the expanding separated shear layer, while the Mach 6.0 wake is de¦ned by a shock, expansion waves, and a recompression region. In both cases, an instantaneous view on the base characteristics reveals complex azimuthal §ow structures even for axisymmetric geometries. The §ow regimes are discussed by comparing the aerodynamic characteristics, such as the size of the recirculation region and the turbulent kinetic energy.

The turbulent wake of a generic space launcher at cold hypersonic freestream conditions is investigated experimentally and numerically to gain detailed insight into the intricate base §ow phenomena of space vehicles at upper stages of the §ight trajectory. The experiments are done at Ma ∞ = 6 and Re D = 1.7 · 10 6 m −1 by the German Aerospace Center (DLR) and the corresponding computations are performed by the Institute of Aerodynamics Aachen using a zonal Reynolds-averaged Navier Stokes / Large-Eddy Simulation (RANS/LES) approach. Two di¨erent aft-body geometries consisting of a blunt base and an attached cylindrical nozzle dummy are considered. It is found that the wind tunnel model support attached to the upper side of the main body has a nonnegligible impact on the wake along the whole circumference, albeit on the opposite side, the e¨ects are minimal compared to an axisymmetric con¦guration. In the blunt-base case, the turbulent supersonic boundary layer undergoes a strong aftexpansion on the model shoulder leading to the formation of a con¦ned low-pressure (p/p ∞ ≈ 0.2) recirculation region. Adding a nozzle dummy causes the shear layer to reattach on the its wall at x/D ∼ 0.6 and the base pressure level to increase (p/p ∞ ≈ 0.25) compared to the blunt-base case. For both con¦gurations, the pressure §uctuations on the base wall feature dominant frequencies at Sr D ≈ 0.05 and Sr D ≈ 0.2 0.27, but are of small amplitudes (p rms /p ∞ = 0.02 0.025) compared to the main body boundary layer. For the nozzle dummy con¦guration, when moving downstream along the nozzle extension, the wall pressure is increasingly in §uenced by the reattaching shear layer and the periodic low-frequency behavior

No abstract

Numerical investigations of the flow field of a supersonic single expansion ramp nozzle (SERN) and its interaction with a freestream of Ma=8 are conducted using Large-Eddy Simulations (LES) to allow accurate predictions of the high Reynolds number flow characterized by shocks and expansion waves, turbulent boundary and shear layers, and flow separation. The influence of selected parameters, i.e., boundary-layer thickness, wall temperature, gas composition, and flap angle on the thrust vector is determined to optimize the thrust vector. It will be shown that variations of boundary-layer thicknesses and wall cooling have inferior influence on the thrust compared to flap angle variations and change of the adiabatic exponent. A positive flap angle causes an increase of lift and drag, while a variation of the fluid, i.e., a gas at a lower adiabatic exponent than air, leads to a decrease of lift and drag.