Flat plate drag reduction by turbulence manipulation

Abstract

The major objective of the present paper is to delineate the conditions under which a turbulent boundary layer manipulated by the insertion of a passive object may lead to a lower overall drag than in the unmanipulated flow. It is pointed out that almost any device inserted in the boundary layer will lead to a lowerskin friction drag. Experimental evidence is presented to support this conclusion, which is most easily thought of as characterizing "wall wakes". However when the streamwise extent of the manipulator is not small, the no-slip condition forced on the manipulator boundary modifies the flow through what is here described as the "blade" effect. The presence of this effect may be inferred from experimental data which unambiguously show that a flat plate with a chordc of the order of the boundary layer thicknessd produces the same order of skin friction reduction as a cylindrical rod with a higher wake momentum thickness ϑ w . As in general both blade and wake effects are operating simultaneously, available data are analysed in the plane formed by the two variables c/ δ and ϑ _w /δ. This analysis shows that the net drag reduction, if it occurs at all, is quite small, and that its realization demands a blade chord (at zero incidence) larger than about 40 ϑ _w . Finally, it is pointed out that if, as often claimed, such manipulators effect a permanent decrease in boundary layer momentum thickness, then sufficiently far downstream where the boundary layer may be expected to have returned to equilibrium, the local skin friction coefficient at any station must be higher than in the unmanipulated flow at the same station because of the lower momentum thickness Reynolds number. It therefore follows that any possible reduction in drag due to the manipulator can only be achieved for certain limited downstream lengths behind the manipulator.