I T is shown in Ref. 1 that the presence of a fuselage significantly promotes the breakdown of delta wing leadingedge vortices. Applying the equivalent angle-of-attack concept only accounted for a fraction of the measured effect of the fuselage, predicting a 17% increase of the effective angle of attack compared to the measured 45% increase. The present comment describes a flow mechanism that can explain this discrepancy, i.e., the wing-camber effect generated by the fuselage-induced upwash along the leading edge of the delta wing. The effect of longitudinal camber on the breakdown of the leading-edge vortex on a slender delta wing is large (Fig. 1). For the same maximum local angle of attack on the delta wing max a positive camber of Aa/amax = 1 delays breakdown to occur downstream of the trailing edge, whereas a negative camber of the same magnitude, Aa/amax = — 1, causes burst to occur very close to the apex. Obviously, for a pitching delta wing the pitch-rate-induced camber will have similarly large effects on the breakdown of leading-edge vortices. It is described in Refs. 4 and 5 how the roll-rate-induced camber effect would be very similar to the pitch-rate-induced camber effect. In both cases, it is the motion-induced change of the local angle of attack at the leading edge that matters (Fig. 2). For the maximum reduced frequency and amplitude used in the roll oscillation test of a 65-deg, sharp-edged delta wing, the roll-rate-induced camber at the trailing edge was AaLE/a = 0.31. Following the suggestion in Ref. 7, static tests were performed with models deformed to produce the roll-rateinduced camber (Fig. 3). The results were as expected; i.e., the twisted-up side of the delta wing experienced later vortex breakdown than the opposite, twisted-down side, approximately at 70% chord compared to 45% chord (for zero roll angle <£ = 0). It should be noted that the variation of the induced angle of attack along the leading edge is linear, proportional to the local semispan, in the examples given in Figs. 1-3, whereas the variation is nonlinear, roughly inversely proportional to the local semispan, in the case of the body-induced wing camber. However, the results in Ref. 7 serve to illustrate how large the effect of camber is on vortex breakdown. Based upon these results one can expect that the bodyinduced negative wing camber along the delta wing leading edge is a very important parameter. Thus, a better approach than using any mean-alpha value for the delta wing, such as the equivalent angle of attack, is to use the body angle of attack a together with the body-induced wing camber at the trailing edge AaLE/a, to correlate measurements of the breakdown of delta wing leading-edge vortices in the presence of a fuselage.
Read full abstract