The Unsteady Flow and Wake Near an Oscillating Cylinder

Abstract

This paper reports on velocity, pressure, and correlation measurements in the turbulent wake and the adjacent unsteady potential flow of a circular cylinder. Particular attention is given to the effects that attend oscillation of the cylinder in the plane of the lift force. The study was undertaken for four reasons. First was the need to generalize measurement results on the lift force of an oscillating cylinder placed in a flow [1]. Second was the desire to clarify the fluidelastic nature of such forces. Third was the wish to compare gross features of real and computer generated flows. And fourth was the presumed possibility to create unsteady flow of controlled characteristics by using vibrating cylinders. The data presented here pertain to the velocity field close to the cylinder. Actually the records for locations further downstream suggest that prospects for control of flow characteristics using vibrating bluff cylinders do not look promising beyond, say, 15 dia. Measurements made close to the cylinder permit more definite conclusions. The fluidelastic nature of the flow associated with vortex-induced vibrations of cylinders has been demonstrated conclusively. Various existing notions about this question can now be segregated. Reynolds numbers of up to 105 were attained; this is sufficiently high to give results in the range of practical application. Revealing records about the flow field structure were acquired. They permit many conclusions by mere visual inspection because of the special attention given to the unsteady or potential flow region.

Suppression of three-dimensional flow instabilities in tube bundles

“…Since the amplitude of cylinder oscillation is A/D = 0.23 > 0.125, these results are consistent with the observations of Toebes (1969) that sufficiently large amplitude cylinder vibration amplitude should produce a high degree of spanwise correlation. We also find that the tight packing alone does not suppress three-dimensional instabilities.…”

“…He conjectured that similar effects might be observed in tube bundles. Blevins' investigations confirmed earlier work by Toebes (1969) who showed that a cylinder vibration amplitude of A/D ≥ 0.125 was required to achieve a high degree of spanwise correlation. In the case of tube bundles, it has also been suggested that it is the tight spacing of the bundle that suppresses three-dimensional instability.…”

Turbulent wake of an inclined cylinder with water running

“…Zdravkovich (1982) concluded that the main feature of vortex shedding synchronizing with an oscillating cylinder was to magnify greatly C D and C L , compared with a stationary cylinder. A number of factors contribute to this magnification associated with the lock-in phenomenon, including the reduced vortex formation length (Gerrard 1966;Griffin & Votaw 1972), the prolonged spanwise correlation length (Toebes 1969) and the increased period of vortex shedding (Sarpkaya 1979). The lock-in is in general characterized by (i) the synchronization of the cylinder oscillation with fluctuating lift and vortex shedding from the cylinder (Bishop & Hassan 1964;Govardhan & Williamson 2000), (ii) a short vortex formation region (Griffin 1971;Govardhan & Williamson 2001) (Hover, Tredt & Triantafyllou 1998), (vi) a considerable increase in the vortex strength (Gu, Chyu & Rockwell 1994) or fluctuating velocities (Govardhan & Williamson 2001), (vii) an increase in the oscillation of the separation points (Sarpkaya, 1979), and (viii) a low effective damping ratio, i.e.…”

A study of two-dimensional flow past an oscillating cylinder

“…all increase with increasing cylinder motion amplitude (e.g. Toebes 1969;Novak & Tanaka 1975) it is reasonable to suggest that the harmonic motion of a long circular cylinder seems to suppress three-dimensionality and produce flows that are more twodimensional than their fixed-cylinder counterparts, at least in the near-wake region. The stochastic nature of turbulent three-dimensional motions is obviously neglected in our model, but we do not believe that this is important in regard to the fundamental mechanisms driving the fluid-structure interaction, because these mechanisms are primarily two-dimensional.…”