A winglet is a small wing-tip lifting surface, often inclined upward out of the main wing planform. By changing the three-dimensional loading and flow near the tip, it can reduce lift-induced drag. Its winglet cant angle, shape and attachment must be considered with the entire aircraft wing; it is not a universal bolt-on reduction in total drag.[ref]
A Design Approach and Selected Wind-Tunnel Results at High Subsonic Speeds for Wing-Tip Mounted Winglets by Richard T. Whitcomb. NTRS author abstract, geometry and measured aerodynamic effects
Richard T. Whitcomb's 1976 report compared measured forces, moments and loads on a representative transport wing with and without wing-tip devices. A particularly useful comparison was a tip extension giving approximately the same increase of bending moment at the wing–fuselage junction as the winglets. That comparison asks how much aerodynamic benefit is obtained for a similar structural penalty, rather than comparing only nominal span.[ref]
A Design Approach and Selected Wind-Tunnel Results at High Subsonic Speeds for Wing-Tip Mounted Winglets by Richard T. Whitcomb. Abstract, matched bending-moment increment for tip-extension comparison
An original coefficient budget illustrates why component and total benefits differ. Keep the same reference wing area, dynamic pressure and lift coefficient. Suppose initiallywhere the second term is the lift-induced drag contribution. If a proposed modification reduces that term by but adds to other drag coefficient contributions, thenThe total reduction is about , not . This is an invented budget, not a prediction or a measured winglet efficiency. Increased loading, additional wetted area and an off-design operating point can change the result.
A convincing wind tunnel comparison measures total drag at matched lift, records moments and loading, and documents how the mounting correction and reference area are handled. A photograph of a weaker-looking wingtip vortex is not a measurement of total drag. Conversely, reducing lift-induced drag does not require all three-dimensional trailing vorticity to disappear.
New to topics? Read the docs here!