Rear fuselage flow studies on a modern transonic transport aircraft

Abstract

An experimental program was set up in the CERT/ONERA's T2 wind tunnel, which is transonic, pressurized, and has self-adaptive walls. A l/80th-scale model of a modern transonic transport aircraft was tested in two configurations: 1) without horizontal and 2) with horizontal stabilizer. Various measurements were performed: oil-flow visualizations, pressure distributions, boundary-layer surveys along the symmetry lines using a three-dimensional laser Doppler anemometry system, and wake surveys with both pressure and velocity measurements in a plane downstream of the base. Moreover, both inviscid and viscous computations were carried out separately under experimental conditions. This article reports a detailed analysis of the computations and experiments conducted on the model in both configurations; the objective of this study was to reach a better understanding of the flows developing along the rear part of a fuselage. ESIGNERS have given considerable attention to threedimensional flow separation on various parts of aircraft. Although rear-fuselage separation is less important on a commercial transport aircraft than on one with military-style, rearloading cargo doors, it is still worth investigating because it may be a significant source of drag. It should be emphasized that three-dimensional flow separation cannot be accurately detected by observing tuft stuff during flight tests or woollen threads in a wind tunnel. Generally speaking, because threedimensional flow separation is associated with the storage of fluid in the wall region, it causes an important vortex motion that develops near the lower symmetry line of the and, consequently, spreads downstream of the upswept base. The experiments reported here were conducted in the test section of CERT's T2 wind tunnel, which is transonic, pressurized, and has self-adaptive walls. Tests were performed at a stagnation pressure close to 2 bar, at ambient temperature and at a freestream Mach number of 0.82. The Reynolds number, based on the aerodynamical chord length of the model, was close to 2.5 x 10 6. That figure is comparable to Reynolds numbers reached for industrial wind-tunnel applications, but of course is less than the flight Reynolds number, which is close to 5 x 10 7. For each model configuration (fuselage without stabilizer and fuselage with stabilizer), a number of measurements were performed: oil-flow visualizations, pressure distribu