Vortex Shedding from Smooth and Roughened Cylinders in Cross-Flow near a Plane Surface

Abstract

SummaryThe characteristics of the flow field around a circular cylinder in cross-flow placed at various distances from a plane, parallel both to the flow and to the cylinder axis, were analysed using a hot wire anemometer. Experiments were performed in a wind tunnel with Reynolds numbers ranging from 0.85×105 to 3×105. The spectra of the hot wire signals were obtained using a Fast Fourier Transform technique programmed on a PDP 11/40 computer. As regards a smooth cylinder, the main features of the vortex shedding mechanism in the subcritical regime remained unaltered for distances from the plane greater than approximately 0.4 diameters; in particular the Strouhal frequency did not show any significant variation relative to the typical value for an isolated cylinder. As for lower values of the distance from the plane, the regular vortex shedding disappeared and the hot wire spectra showed typical turbulent features. The possibility of obtaining supercritical conditions by roughening the cylinder surface was confirmed together with the importance of the Reynolds number based on the typical roughness size, Rk, in the evaluation of the flow regime around the cylinder. In the case of roughened cylinders, and with values of Rk below-350, the regular vortex shedding disappeared at a distance from the plane smaller than 0.3 diameters. This fact suggests that, at least in part of the supercritical regime, the influence of the plane can be smaller than in the subcritical regime.

Vortex shedding from a circular cylinder near a moving ground

“…They also measured velocity fluctuations in the near wake of the cylinder to study the effect of h / d on the Strouhal number of the vortex shedding, St, and reported that the critical gap ratio ͑h / d͒ c , at which the spectral peak of the local velocity fluctuations disappeared, was 0.2-0.3 ͑it should be noted, however, that the disappearance of the spectral peak of the local velocity fluctuations does not necessarily ensure the global cessation of the vortex shedding 5 ͒. Similar measurements of local velocity fluctuations and thereby of ͑h / d͒ c were also carried out at higher and lower Reynolds numbers, [6][7][8] whereas more details of the global cessation of the Kármán vortex shedding were later visualized but only at lower Reynolds numbers, 5,9,10 as summarized in Table I.…”

Flow Visualization Around a Circular Cylinder Near to a Plane Wall

“…Contrary to previous experiments, they found that for G/D(5)0 the gap ratio had a fairly strong e!ect on St. For example, at G/D"0)5 (the minimum value they tested) St was 10% greater than that at G/D"5)0. As discussed by Angrilli et al, a possible reason for the di!erence between their results and those of Buresti & Lanciotti (1979) and Bearman & Zdravkovich (1978) is the di!erence in Re. Grass et al (1984) investigated the vortex shedding response behind a cylinder with three di!erent boundary-layer thicknesses, /D+0)28, 2)6 and 6)0, for two di!erent Reynolds numbers, Re"1785 and 3570.…”

“…For G/D'0)6, regular alternate vortex shedding was observed. Buresti & Lanciotti (1979) also investigated the e!ect of G/D on the vortex shedding frequency, but in a higher range of Reynolds numbers (0)86;10}3)0;10) and for a much thinner boundary layer ( "0)1D). For a smooth cylinder, the critical gap below which vortex shedding did not occur was between 0)3 and 0)4D; however, for a roughened cylinder TABLE 1 Summary of previous work for #ow around a circular cylinder close to a plane wall.…”

Modélisation probabiliste des efforts de houle sur une structure offshore en présence de bio salissures

“…Ce constat est aussi établi par Buresti 9 qui démonte que pour des valeurs de Re k (nombre de Reynolds rugueux) supérieures à 200, il était possible d'atteindre un régime super-critique. Les bio-salissures marines modifient donc la fréquence tourbillonnaire autour des piles principales de la structure altérant le régime d'écoulement et les coefficients hydrodynamiques.…”

“…On effectue l'ajustement sur les fonctions de répartitions F Cd , fonctions monotones, par la méthode des moindres carrés. On note, sur la figure 10, une bonne corrélation (erreur quadratique totale inférieure au millième) entre la fonction de répartition issue de la modélisation discrète et celle de le loi de Gumbel de type I (loi de valeur extrême de type I) de fonction de distribution donnée en (9). Cet ajustement est particulièrement adéquat sur la queue de distribution à droite, fondamentale pour le calcul en fiabilité.…”