Improving performance of high Mach number scramjets: fuelling strategies and combustor design

With prospective flight speeds exceeding Mach 12, airbreathing scramjets, incorporated in hybrid satellite launch vehicles, offer efficiency and operability benefits when compared to rocket-based systems. By capturing atmospheric air, the scramjet second stage negates the requirement for onboard oxidisers. Addressing the small satellite niche market, these hybrid systems provide a cost-effective solution to launching such payloads to orbit. At hypersonic speeds, however, conventional wing-mounted engine nacelles are not viable; the scramjet must be integrated to the airframe. Hence, the flight-candidate, Mach 12 Rectangular-to-Elliptical Shape-Transitioning (M12REST) scramjet blends a rectangular capture area (permitting parallel mounting of engine modules to the planar vehicle underside), with a structurally efficient elliptical combustor. As access-to-space scramjets suffer immense heating loads, and engine airflow residence times approach air-fuel reaction timescales, hydrogen fuel is utilised for its exceptional cooling capacity and rapid ignition characteristics. However, efficient fuelling techniques are required to avoid crippling the engine's performance through conservative combustor lengths. This thesis numerically and experimentally investigated whether the performance of such an airframe-integrated scramjet can be sufficiently enhanced through customisation of fuel injection and combustor geometry to achieve net thrust at Mach 12. Three-dimensional, chemically reacting Reynolds-averaged Navier-Stokes solutions enabled numerical analysis, while experimental validation was performed within The University of Queensland's, T4 Stalker Tube.Preliminary work investigated cascaded fuel injectors, targeting mixing and penetration improvements. Two streamwise-aligned jets were employed, with the upstream injector half the diameter of the rear, while the distance between each was varied. The upstream jet induced a low dynamic pressure region in its wake, shielding the downstream jet from the hypersonic crossflow. The leeward jet benefits from an increased local jet-to-freestream momentum ratio, while its larger diameter increases absolute penetration. Examining performance at M12REST mean combustor entrance conditions, unique optimal injector spacings existed for each performance metric, displaying improved penetration, spread, and mixing across the range of flight Mach numbers examined (6 lnM ln12). However, jet-to-jet spacings of 4-6 total jet diameters displayed universal performance enhancements over single fuel jets. With penetration improvements of 30-40%, and mixing improvements of 40-70%, the simple fixed geometry, passive technique is ideal for use within an accelerating access-to-space scramjet.While cascaded injectors improved performance within uniform flows, the flows ingested by airframe-integrated scramjets are far less homogeneous. Non-uniform compression fields combine with thick boundary layers developed over the vehicle forebody to deliver density stratified flow to the combustor. This thesis developed a method to exploit the interactions between this density stratified flow, and the vortices induced by a strategically positioned inlet injector to manipulate the flow and redistribute engine centreline oxygen in captured air to the more accessible combustor cowlside. When combined with four supplementary cowlside injectors (termed the MJ5c configuration), 100% mixing efficiency was reached prior to the combustor exit, while combustion efficiency exceeded 80% more than 3.6 combustor-heights further upstream than with previous fuelling configurations (which employed three cowlside, and two sidewall injectors - termed 3c2s). The 25% greater near-field heat addition rate delivered 7.6% more energy to the flow.With Mach 12 nearing the upper limit of scramjet-enabled access-to-space, internal components must be tuned to minimise heating and drag loads. The M12REST combustor length was hence targeted. Through numerical analysis, reductions in its constant area length of 75% suffered negligible performance losses prior to the nozzle entrance for the MJ5c fuelling configuration. Despite reductions in combustion efficiency, heat release and inviscid thrust of 6%, 18% and 26% respectively, net system improvements were achieved through the 22% depletion of viscous drag. Examining combustor geometry further, the M12REST combustor joined the isolator via a rear-facing step, originally included to facilitate film injection. However, removal of this legacy step promoted mixing limited combustion through increases in flow Damkohler number, with g80% combustion efficiency achieved 2.75 combustor heights further upstream (when injecting fuel via the 3c2s scheme). The combustor was hence shortened by 68%, and experimental analysis compared its performance to the original geometry. Improvements in gross thrust coefficients of 25% and 10% were achieved for combustor-only, and combined fuelling respectively. The latter achieved greater performance than that obtained through direct summation of the constituent inlet-, and combustor-only fuelled cases, indicating inlet-injected fuel pilots combustion of fuel injected further downstream.Maintenance of engine wall temperatures during flight may depend on fuel-based regenerative cooling to counter the immense heating loads experienced, and such operating conditions differ substantially to those within shock tunnels. Numerical analysis compared the M12REST scramjet's performance at each condition, employing the MJ5c fuel injection scheme with the shortened combustor. Room temperature walls and fuel modelled shock tunnel conditions. Flight was modelled through 800 K inlet walls, which transitioned to 1800 K by the combustor; fuel was injected at 1000 K. While flight conditions promoted rapid mixing, excessive combustor temperatures inhibited the completion of reaction pathways, with reactant dissociation reducing heat release by 16%. However, the hot walls incurred 28% less heat loss. Fuel injected at the inlet suffered from premature ignition, suggesting flight operation may benefit from these injectors being moved further downstream. When coupled with counteracting differences in heat release and loss to the walls, the optimal engine design for flight may differ considerably from that which provides the best performance in the tunnel.