An experimental investigation of an airframe integrated three-dimensional scramjet engine at a Mach 10 flight condition

Realisation of the dream of airbreathing access-to-space requires the development of a scramjet engine that produces sufficient net thrust to enable acceleration over a wide Mach number range. With engines that are highly integrated with the airframe, the net performance of a scramjet powered vehicle is closely coupled with the vehicle attitude and is difficult to determine only from component level studies. This work investigates the influence of airframe integration on the performance of an airframe integrated scramjet through the measurement of internal pressure distribution and the direct measurement of the net lift, thrust and pitching moment using a three-component stress wave force balance. The engine chosen as the basis for this study was the Mach 12 rectangular-to-elliptical shape-transition (M12REST) scramjet that was developed by Suraweera and Smart (2009) as a research engine for access-to-space applications. The inlet and combustor flowpath were integrated with a slender 6° wedge forebody, streamlined external geometry and three dimensional thrust nozzle. The scale of the engine was chosen so that the entire engine would fit within the core-flow diamond (bi-conic) produced by a Mach 10 facility nozzle. The Mach 10B facility nozzle was chosen because it is the largest nozzle current in use with the T4 Stalker Tube and because the off-design performance of a scramjet engine is of interest for access-to-space vehicles that must accelerate over a range of Mach numbers. Freejet experiments were conducted within the T4 Stalker Tube. Two true-flight Mach 10 test conditions were used: a high pressure test condition that replicated flight at a dynamic pressure of 48kPa and a low pressure test condition that replicated flight at a dynamic pressure of 28kPa. Scaling of the test conditions according to the established binary scaling law was not completed due to facility operational limits. The engine featured two fuel injection stations from which gaseous hydrogen was injected. The first injection station was partway along the length of the inlet while the second injection station was at the start of the combustor behind a rearward facing circumferential step. In addition to investigating inlet-only and step-only injection, a combined scheme where 68% of the fuel was injected from the step station and 32% from the inlet station was also investigated. To support the analysis of the experimental results, numerical simulations of the engine with no fuel injection were conducted using the NASA code VULCAN. Analysis of the simulations show that the mass capture ratio with respect to the projected inlet area is approximately 60% at each test condition. The simulations also show that spillage of flow from the slender forebody accounts for just 12% of the flow through the projected inlet area, a small but non-negligible fraction. By integrating the engine surface forces, the drag coefficient with respect to the projected frontal area of the engine is calculated to be 0.219 at the high pressure test condition and 0.243 at the low pressure test condition. A breakdown of the total drag shows that the internal and external drag are approximately equal and approximately double that of the forebody. With respect to the planform area of the engine, the lift coefficient is calculated to be 0.038 at both test conditions. The centre of force is located at 36% of the model length. Pressure measurements along the internal bodyside wall of the engine were used to assess inlet starting and the presence of combustion. The results show unequivocally that fuel injected from the inlet injection station acts as a pilot for fuel injected from the step injection station. For both the inlet and combined injection schemes significant combustion was obtained over a range of fuel equivalence ratios at each test condition. In comparison, negligible combustion-induced pressure rise was measured for the step injection scheme, a consequence of the reaction length being greater than the combustor length for this engine and test condition. Using a three-component force balance, the drag was successfully measured for both fuel-on and fuel-off tests. At the high pressure test condition the average fuel-off drag coefficient of the engine was measured to be 0.246±0.025, a value that is within 12% of numerical simulation. At the low pressure test condition the drag coefficient was measured to be 0.312±0.032, a value that is within 28% of numerical simulation. When gaseous hydrogen fuel was injected from the inlet injection station at an equivalence ratio of 0.75, the measured drag coefficient reduced to 0.218±0.062, corresponding to a specific impulse increment of 2180s and a specific thrust increment of 470Ns/kg. For the combined injection scheme, a drag coefficient of 0.118±0.034 was measured for a fuel equivalence ratio of 1.20, corresponding to a specific impulse increment of 2160s and a specific thrust increment of 740Ns/kg. Net positive thrust was not achieved, due in part to a low performance three-dimensional nozzle. Also, an interaction of the force balance shielding and facility nozzle was observed. This interaction adversely affected the size of the core-flow diamond and the measured lift and centre of force. This thesis represents the first time that force data have been measured for a hydrogen fuelled scramjet engine at true-flight Mach 10 test conditions. This work demonstrates that, although difficult, the direct measurement of the aerodynamic performance of a geometrically and mechanically complex, airframe integrated, fuelled scramjet engine module at a high Mach number flight condition is possible within the T4 Stalker Tube. Finally, airframe integration did not significantly alter the characteristics of the M12REST engine, indicating that this class of engine is suitable for use in an airframe integrated, airbreathing access-to-space system.