We investigate a new type of flow-tagging velocimetry technique for hypersonic flows.

A theoretical model that determines the optimum excitation frequency for obtaining a fluorescence signal with a strong dependence on fuel mole-fraction is presented for supersonic fuel-air compressible mixing studies. The challenge associated with this is to maintain a high sensitivity to fuel mole-fraction with minimal sensitivity to temperature and pressure in a flow with large temperature variations and pressure gradients. The results of the model are applied to the mixing region behind various scramjet fuel injectors in a shock tunnel to measure fuel mole-fraction. Hydrogen fuel at a Mach number of 1.7 is injected into a mostly N2 free stream at Mach 4.8. Experimental fluorescence images are presented in streamwise and spanwise planes. Nomenclature cp Specific heat capacity at constant pressure, J.kg −1 cp,∞ Specific heat capacity of pure freestream, J.kg −1 c p,f uel Specific heat capacity of pure fuel stream, J.kg −1 f J Boltzmann fraction of the absorbing state g Spectral overlap integral, 1/cm −1 ga Absorption line shape, 1/cm −1 g l Spectral profile of laser, 1/cm −1 k Boltzmann's constant, J.kg −1 mNO Molecular mass of NO, kg mp Molecular mass of perturbing species, kg

Planar laser-induced fluorescence (PLIF) imaging was performed to visualize the fin bow shock, separation shock, viscous shear layer and recirculation region of the flowfield at the junction of a blunt fin and a flat plate. Making use of the temperature dependence of the PLIF technique, images were made sensitive to temperature to provide qualitative information on the flowfield. The PLIF technique was also used as the basis for a flow-tagging technique, making it possible to measure a velocity component and to demonstrate the reverse flow of the separated region. Flow visualisation of the plane of symmetry allowed determination of the point of boundary layer separation, the angle of the separation shock and the bow shock standoff distance. These parameters were compared with predictions made by computational fluid dynamic simulations of the flowfield. Good agreement between theory and experiment was achieved. Comparisons between theoretical and experimental velocity measurements showed good agreement.

No abstract

No abstract