Search

The flow-induced oscillation (FIO) of bluff bodies is commonly encountered in the fluid structure interaction (FSI) problems. In this study, we use an unstructured moving grid strategy and simulate the FIO of two rigid plates, which are elastically hinged at the two ends of a fixed flat plate in a cross-flow. We use a hybrid finite-element-volume (FEV) method in an arbitrary Lagrangian–Eulerian (ALE) framework to study FIO of the two hinged plates. The current simulations are carried out for wide ranges of flow Reynolds number (50–175), spring stiffness coefficient, and the two hinged plates' moment of inertia magnitudes. The influences of these parameters are investigated on the magnitudes of maximum deflection angle, the amplitude of oscillation, the total lift and drag coefficients, and so on. The study is also carried out in the transition period to describe the in-phase and out-of-phase angular oscillations occurring for the two elastically hinged plates with respect to each other. After the transition period, the two hinged plates eventually arrive to a similar periodic oscillation; however, with some phase lags. We find that the achieved phase lag is equal to the phase lag between the two pairs of flow vortices, which are alternatively shed into the flow from the upper and lower hinged plates. Similar to past FIO problems, the current model also exhibits two important lock-in and phase-switch FSI phenomena; however, in angular directions. There is a phase jump of approximately 170° between the aerodynamic lift coefficient and angular oscillations of hinged plates, which nearly occurs in the middle of lock-in region. Indeed, our literature review shows that this is the first time to report the phase-switch phenomenon in angular oscillations of three-element bluff bodies in a FSI problem. [ABSTRACT FROM AUTHOR]

Abstract: We have studied the effects of controlled damping on the amplitude and frequency response profiles of an elastically mounted cylinder in cross-flow. The dimensionless damping parameter, , which is closely related to the traditional “mass-damping” parameter, , was varied over a wide range of values through the use of a variable magnetic eddy current damping system. For low damping and sufficiently high Reynolds number we observe the previously described large-amplitude, three-branch (initial, upper, lower) response profile, and for high damping or low Reynolds number we observe the small-amplitude, two-branch (initial and lower) response profile. However we find that, because of the influence of Reynolds number, the traditional labels of “high mass-damping” and “low mass-damping” are incomplete with regard to predicting a large or small-amplitude response profile. In our experiments, as damping is systematically increased, we observe a transition between these two profiles characterized by a gradual “erosion” and eventual disappearance of the large-amplitude section (upper branch) and the scaling down of the lower branch region. We find that jumps from the upper to the initial branch originate on the 2S/2P boundary in the Williamson–Roshko plane. Another new finding is a hysteresis between the lower branch and the desynchronized region, which only appears at low Reynolds numbers. We also explore changes in the frequency response profile, which are connected with the changes in the amplitude profile, for our upper branch cases. We observe that analogous to the three amplitude branches, there are three distinct branches for the frequency response. [Copyright &y& Elsevier]