The wakes of cylindrical bluff bodies at low Reynolds number

Abstract

Experiments on the near wake of a cylinder will be discribed in an attempt to present a coherent picture of the events encountered as the Reynolds number increases from small values up to values of a few thousand. Much work on this subject has already been done, but there are gaps in our description of these flows as well as more fundamental deficiencies in our understanding of them. The subject has been reviewed several times and most recently by Berger & Wille (1972) whose paper covers much of the ground that will be discussed again here. The present work may be regarded as built upon this latest review. I remember with gratitude many helpful discussions with the late Rudolph Wille who contributed so much to this subject. The investigation has concentrated on circular cylinders, but the wakes of bluff cylinders of different cross sectional shapes have also been observed. Bluff cylinders in general are considered in §§4 and 5, together with the effect of splitter plates on circular cylinders in §9. The experiments concern, almost exclusively, flow visualization of the wakes by means of dye washed from the bodies. The patterns of dye observed are, therefore, filament line representations of the flow leaving the separation lines on the body. It must be stressed that the dye does not make visible the vorticity bearing fluid because at low Reynolds number, vorticity diffuses considerably more rapidly than does dye. The ratio of the molecular diffusivity of momentum to that of mass of dye is of the order of 100.