Not available / text

published_or_final_version / Mechanical Engineering / Doctoral / Doctor of Philosophy

published_or_final_version / Civil Engineering / Doctoral / Doctor of Philosophy

The cross-flow past a pair of equal-diameter circular cylinders, arranged in a staggered configuration, was investigated experimentally in a closed-circuit water tunnel at Reynolds numbers, based on the mean-flow velocity and the cylinder diameter, within the lower subcritical range. The wake formation process was studied employing dye-injection flow visualization and hot-film measurements. The main emphasis was placed on acquiring a physical understanding of the mechanisms leading to vortex shedding, and particularly on the effect of a forced oscillation transverse to the flow direction of either of the two cylinders. For comparison purposes, investigations were also carried out with both cylinders stationary. / Experimental results showed that, for a reasonably large angle of incidence, the flow in the wake of a stationary cylinder pair could be characterized by two distinct periodicities, each of which was dominant on one side of the wake. Furthermore, for lower Reynolds numbers (Re < 1.0x10 4), there was an integral relationship between the two Strouhal numbers, but this integral relationship was no longer maintained for Re > 1.0x10 4. On the other hand, the flow around stationary cylinders for a small angle of incidence was characterized by a single Strouhal number, which remained approximately constant over the entire Reynolds number range. / For all the cylinder configurations investigated the wake flow patterns remained essentially the same as those of the corresponding static cases, when either of the two cylinders was forced to oscillate with a nondimensional forcing frequency less than approximately 0.10. However, beyond this value, the wake underwent considerable modification vis-a-vis when the cylinders were stationary, and the flow pattern within the wake was strongly dependent on the value of the forcing frequency. In particular, there were distinct regions of synchronization between the dominant wake periodicities and the cylinder oscillation; these synchronization regions involved sub- and superharmonics as well as fundamental synchronizations. With either upstream or downstream cylinder oscillation, the wake on the mean-flow side of the downstream cylinder synchronized with the shear layers separated from its outer surface, whereas synchronizations on the mean-flow side of the upstream cylinder were caused by the periodicities formed from the interaction of the other three shear layers. / The flow phenomena associated with the synchronizations were described in detail via flow visualization. The organization of the wake was strongly dependent on whether it was the upstream or downstream cylinder which was oscillating. The synchronized wake on the mean-flow side of the downstream cylinder at both lower and higher oscillation frequencies for upstream cylinder oscillation was observed to form either by the shedding of independent vortices or by the coalescence of two or more vortices. However, for downstream cylinder oscillation, although the synchronizations on this side of the wake at lower oscillation frequencies were caused by the shedding of independent vortices or by the coalescence of vortices, those at higher oscillation frequencies were the consequence of the coalescence of vortices only. For large incidence angles, the number of shear layers separated from the downstream cylinder which interacted with those separated from the upstream cylinder was critical in causing the synchronizations on the mean-flow side of the upstream cylinder. / In most cases, the flow for all the cylinder configurations traversed between the same patterns as those obtained when the cylinders were placed stationary at their minimum and maximum transverse spacings; but there were also some situations where the oscillation of either cylinder pushed the flow outside the regimes associated with the stationary configurations. The synchronization ranges obtained when the upstream or downstream cylinder was oscillating were different from each other, and these ranges were much wider than the corresponding synchronization ranges for a single oscillating cylinder. For two cylinders, an analysis of the fundamental synchronization showed that the frequency range over which this occurred was much broader for upstream cylinder oscillation than for downstream cylinder oscillation. Also, the fundamental synchronization ranges for downstream cylinder oscillation were closer to those for single cylinder oscillation in comparison to those for upstream cylinder oscillation.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.