Perpendicular blade vortex interaction and its implications for helicopter noise prediction: Wave-number frequency spectra in a trailing vortex for BWI noise prediction

Perpendicular blade vortex interactions are a common occurrence in helicopter rotor flows. Under certain conditions they produce a substantial proportion of the acoustic noise. However, the mechanism of noise generation is not well understood. Specifically, turbulence associated with the trailing vortices shed from the blade tips appears insufficient to account for the noise generated. The hypothesis that the first perpendicular interaction experienced by a trailing vortex alters its turbulence structure in such a way as to increase the acoustic noise generated by subsequent interactions is examined. To investigate this hypothesis a two-part investigation was carried out. In the first part, experiments were performed to examine the behavior of a streamwise vortex as it passed over and downstream of a spanwise blade in incompressible flow. Blade vortex separations between +/- one eighth chord were studied for at a chord Reynolds number of 200,000. Three-component velocity and turbulence measurements were made in the flow from 4 chord lengths upstream to 15 chordlengths downstream of the blade using miniature 4-sensor hot wire probes. These measurements show that the interaction of the vortex with the blade and its wake causes the vortex core to loose circulation and diffuse much more rapidly than it otherwise would. Core radius increases and peak tangential velocity decreases with distance downstream of the blade. True turbulence levels within the core are much larger downstream than upstream of the blade. The net result is a much larger and more intense region of turbulent flow than that presented by the original vortex and thus, by implication, a greater potential for generating acoustic noise. In the second part, the turbulence measurements described above were used to derive the necessary inputs to a Blade Wake Interaction (BWI) noise prediction scheme. This resulted in significantly improved agreement between measurements and calculations of the BWI noise spectrum especially for the spectral peak at low frequencies, which previously was poorly predicted.