Abstract: | Further experiments on features of the vortex shedding from tapered flat plates normal to an airstream are described. The work extends that of Castro and Rogers (2002) and concentrates on the study of the effects of varying the spanwise aspect ratio for a fixed shape plate, by appropriate adjustment of end-plates, and of the nature of the shedding as the degree of taper becomes very large, so that the body is more like a triangular plate—e.g. an isosceles triangle—than a slightly tapered plate. With the taper ratio TR defined as the ratio of plate length to average cross-stream width, the paper concentrates on the range 0.58<TR<60. Reynolds numbers, based on the average plate width, exceed 104. It is confirmed that for a small enough taper ratio the geometrical three-dimensionality is sufficiently strong that all signs of periodic vortex shedding cease. For all other cases, however, the flow at different locations along the span can vary substantially, depending on taper. There appear to be at least four different regimes, each appropriate for a different range of taper ratio. These various regimes are described.List of symbols AR Spanwise aspect ratio, W/dav - d Local width of the plate - dav Average width of the plate between the end-plates (or end-plate and tip or wall and tip), m - d0 Base width of the plate (i.e. at z=0), m - f Shedding frequency, Hz - H Dimension of the tunnel cross-section normal to the plate symmetry axis, m - He Splitter plate height (see Fig. 1) - L Total length of plate (between base, where d=d0, and tip, where d=0), m - Le,Lf End plate dimensions (see Fig. 2) - Std Local Strouhal number, fd/U - St0 Strouhal number based on base width, fd0/U - TR Taper ratio, L/dav - U Measured free-stream velocity, m/s - W Spanwise distance between ends of plate - z distance along span, measured from base of plate - z =z/L - z Spanwise width of a constant-frequency cell |