Date of Award

Summer 2014

Degree Type


Degree Name

Master of Science in Mechanical Engineering (MSME)


Mechanical Engineering

First Advisor

Patricia Davies

Committee Chair

Patricia Davies

Committee Member 1

Stuart J. Bolton

Committee Member 2

George T. Chiu

Committee Member 3

Hong Z. Tan


It is anticipated that the use of rotorcraft passenger vehicles for shorter journeys will increase because their use can reduce the time between boarding and take-off. The characteristics of rotorcraft noise are very different to that of fixed wing aircraft. There can be strong tonal components, fluctuations that can also make the noise sound impulsive, and future rotorcraft may produce proportionally more low frequency noise content. Most metrics that are used today to predict noise impact on communities around airports (e.g., Ldn) are just functions of A-weighted sound pressure level. To build a better noise annoyance model that can be applied to assess impact of future and current rotorcraft, it is important to understand the perceived sound attributes and how they influence annoyance. A series of psychoacoustic tests were designed and performed to further our understanding of how rotorcraft sound characteristics affect annoyance as well as evaluate the applicability of existing noise metrics as predictors of annoyance due to rotorcraft noise. The effect of the method used to reproduce sounds in the psychoacoustics tests was also investigated, and so tests were conducted in the NASA Langley Exterior Effects Room using loudspeaker arrays to simulate flyovers and in a double walled sound booth using earphones for playback. A semantic differential test was performed, and analysis of subject responses showed the presence of several independent perceptual factors relating to: loudness, sharpness, roughness, tonality, and impulsiveness. A simulation method was developed to alter tonal components in existing rotorcraft flyover recordings to change the impulsiveness and tonality of the sounds. Flyover recordings and simulations with varied attributes were used as stimuli in an annoyance test. Results showed that EPNL and SELA performed well as predictors of annoyance, but outliers to generate trends have tonal related characteristics that could be contributing to annoyance. General trends in results were similar for both test environments, though differences were greater for the annoyance tests than the semantic differential tests.