No abstract

No abstract

No abstract

For the investigation of the physical processes of human phonation, inhomogeneous synthetic vocal folds were developed to represent the full fluid-structure-acoustic coupling. They consisted of polyurethane rubber with a stiffness in the range of human vocal folds and were mounted in a channel, shaped like the vocal tract in the supraglottal region. This test facility permitted extensive observations of flow-induced vocal fold vibrations, the periodic flow field, and the acoustic signals in the far field of the channel. Detailed measurements were performed applying particle-image velocimetry, a laser-scanning vibrometer, a microphone, unsteady pressure sensors, and a hot-wire probe, with the aim of identifying the physical mechanisms in human phonation. The results support the existence of the Coanda effect during phonation, with the flow attaching to one vocal fold and separating from the other. This behavior is not linked to one vocal fold and changes stochastically from cycle to cycle. The oscillating flow field generates a tonal sound. The broadband noise is presumed to be caused by the interaction of the asymmetric flow with the downstream-facing surfaces of the vocal folds, analogous to trailing-edge noise.

We study the incipient motion of single spheres in steady shear flow on regular substrates at low particle Reynolds numbers. The substrate consists of a monolayer of regularly arranged fixed beads, in which the spacing between beads is varied resulting in different angles of repose and exposures of the particle to the main flow. The flow-induced forces and the level of flow penetration into the substrate are determined numerically. Since experiments in this range had shown that the critical Shields number is independent of inertia but strongly dependent on the substrate geometry, the particle Reynolds number was fixed to 0.01 in the numerical study. Numerics indicates that rolling motion is always preferred to sliding and that the flow penetration is linearly dependent on the spacing between the substrate particles. Besides, we propose an analytical model for the incipient motion. The model is an extension of Goldman’s classical result for a single sphere near a plain surface taking into account the angle of repose, flow orientation with respect to substrate topography and shielding of the sphere to the linear shear flow. The effective level of flow penetration is the only external parameter. The model, applied to triangular and quadratic arrangements with different spacings, is able to predict the dependence of the critical Shields number on the geometry and on the orientation of the substrate. The model is in very good agreement with numerical results. For well-exposed particles, we observed that the minimum critical Shields number for a certain angle of repose does not depend sensitively on the considered arrangement. At large angles of repose, as expected in fully armoured beds, the model is consistent with experimental results for erodible beds at saturated conditions.