This Defense talk focuses on new developments in the physical understanding of Vortex-Induced Vibration (VIV). Classically, the flow-induced free oscillation of an elastic structure in cross-flow is believed to modify the shedding pattern such that the vortex shedding is tuned to or locked-in to the natural frequency of the structure with possible catastrophic consequences. It is also believed that the mass-damping parameter, namely the product of the structure-to-fluid mass ratio, m*, and the structural damping, zeta , can uniquely predict the maximum amplitude of oscillation achieved.

The experiments presented involve elastically-mounted cylinders in cross-flow performing one and two dimensional oscillations with an extended range of mass-ratios of 2<m*<200.

Vortex-induced vibration is studied experimentally for cylinder mass ratios, 2.1<m*<72.  For small mass ratios below 10, a new VIV mode is discovered which does not involve a lock-in behavior.  The oscillation and the shedding frequencies coalesce and deviate slightly from the nominal Strouhal frequency of St=0.2 to smaller values with increasing free stream velocity U.  With increasing mass ratio above 10 (m*>10), the frequency growth with free stream velocity U appears to approach the lock-in limit while the amplitude and the frequency range of oscillations diminish.  Additionally, a novel technique is employed to deduce the unsteady lift coefficient on the body using VIV time traces of the cylinder displacement and their numerical derivatives.

An analytical study of the dynamical equation shows that the oscillation amplitude (A/D) is inversely proportional to effective stiffness, k*eff=(m*/U2)(1-(f/fn)2), where U represents the non-dimensional flow speed and f/fn, the ratio of the oscillation to natural frequencies.  It is hence maintained that at high mass ratio cases studied previously (m*>100), lock-in behavior (f/fn~1 for U~1) is a prerequisite for nominal vibration amplitudes.  At low values of mass ratio, however, k*eff is minimized naturally without a need for lock-in.

Through a detailed study of a large number of cases with low to medium mass ratios in different experimental settings, it is additionally argued that lock-in is a sporadic phenomenon that appears at various mass ratios.  The few occurrences of lock-in at low mass ratios with nominal damping and the unexpected absence of any oscillations at medium mass ratios (m*~30), except for a few cases exhibiting lock-in tendencies indicate that lock-in is not as common as classically believed.