Turbulence induced changes in vortex shedding from a circular cylinder

CER73-74KMB-JEC27.Includes bibliographical references (pages 61-66).Prepared under Office of Naval Research contract no. N00014-68-A-0493-0001; project no. NR 062-414/6-6-68 (Code 438).January 1974.Turbulence effects on vortex shedding from a circular cylinder were investigated in a wind tunnel. Most of the literature on vortex shedding reports investigations in air flows where special efforts had been made to keep the turbulence intensity very low (usually <.5%) in the free stream. Under such conditions, vortex shedding from a circular cylinder can be classified into four regimes with a discontinuity condition between each two succeeding regimes. These regimes are identified by the span of Reynolds numbers, Re, and Strouhal numbers, St, by the angle of boundary layer transition, α, and by the transition point from laminar to turbulent flow in the boundary layer or vortex. In this study, turbulence intensity, near the center of a wind tunnel, was increased from .3% to 25% by plates of similar geometry (many small holes) with blockage ratios from .50 to .97. The Re of all flows was kept near 4 x 104. Vortex shedding frequency, f, was observed simultaneously by a pressure transducer (with sensors on cylinder surface) and a hot wire anemometer (immediately downstream). Means of 50 to 150 cross-spectra defined f. Flow visualization, spectra and transducer output voltages with different α were utilized to estimate transition and separation. Four concepts were verified that are already in the literature. As the turbulence intensity of the upstream flow is increased: (a) the vortex shedding regimes occur at lower Re; (b) discontinuities lose any dominant f value that they might have had; (c) vortices are destroyed closer to the cylinder; and (d) the size of cylinders influence turbulence effects on vortex shedding. Six things were observed that have not been published, to the best of the author's knowledge: (a) analogous changes in vortex shedding processes can be obtained either by increasing the air speed (while holding the turbulence intensity very low) or by increasing the turbulence intensity (while holding the air speed approximately constant); (b) the frequency of the vortex shedding becomes relatively smaller, for the vortex shedding regime, as the turbulence intensity is increased; (c) an estimate of the turbulence induced changes in vortex shedding can be made from Re of the flow, mean air speed, rms value of longitudinal speed fluctuations, integral scale of turbulence and cylinder diameter; (d) the "super critical" regime , associated with highest α and St, can be disrupted by sufficient turbulence intensity; (e) the extent to which turbulence can induce changes in vortex shedding is limited. A model of vortex shedding is hypothesized which includes the concept that vortex shedding is continuous from Re near 40 until either the mean air speed and the turbulence intensity, or both, are increased until turbulence destroys individual vortices before they can be shed