Experimental and Computational Investigation of Rotor Noise in Hover

Abstract

The noise magnitude and directivity of a four-bladed 1.108-m-radius rotor in hover were measured at a tip Mach number of 0.42, representative of an electric vertical takeoff and landing vehicle rotor. The tonal noise was filtered from the acoustic spectra, allowing for analysis of the separate contributions of tonal and broadband noise. Accompanying computations of the same rotor and operating conditions were performed using the rotorcraft comprehensive analysis system (RCAS) with a viscous vortex particle method (VVPM) for aerodynamic calculations and PSU-WOPWOP with the semi-empirical Brooks, Pope, and Marcolini (BPM) broadband model for predicting acoustics. Measured broadband and tonal noise contributed similar magnitudes to the overall sound pressure level, with broadband always several decibels greater. This is in contrast to conventional helicopter rotors, which are dominated by tonal noise. Simulated trends of tonal noise matched the measurements, though the higher harmonic components of the acoustic spectra were underpredicted by up to 30 dB. The broadband noise was underpredicted by 10 dB at moderate rotor collectives, as the measured noise contained noise from laminar and turbulent boundary layers, while the model predicted predominantly laminar boundary layer–vortex shedding noise. Comparing the time histories of the measured and simulated rotor thrust revealed that the RCAS-VVPM method does not capture the amplitude of the 4/rev component of the thrust and lacks the higher frequency content present in the measured thrust, indicating a possible source of discrepancy in the tonal acoustic predictions. The results of this study show that methods to predict rotor tonal and broadband noise without the use of lengthy computational fluid dynamics computations must be further developed.