Development of sound quality metrics for gear rattle in diesel engines

Brandon M Sobecki, Purdue University

Abstract

Sound quality is an important factor in the design of competitive diesel engines. The noise produced by specific components and mechanisms in the engine can play a significant role in determining the perceived quality of the overall noise and in turn the perceived quality of the engine. The goal of the present research is to characterize the sounds produced by the phenomenon known as gear rattle and to develop a model that can be used to assess gear rattle levels in a way that connects directly with human perception of rattle. Most previous work in this area has been focused on rating the overall sound quality of diesel engines, but little has been done to develop models of perception of rattle noise in particular. Various engine recordings in which gear rattle was present were first compiled. These signals were analyzed to identify how sounds changed with increased levels of rattle. A method to simulate gear rattle noise and incorporate it into a no-gear-rattle (baseline) recording was developed. This procedure enabled controlled variation of rattle for use in a psychoacoustic test. The test was designed to quantify detectable levels, perception of growth, and increases in annoyance due to the presence of gear rattle noise. The responses of people who reported having experience with diesel engines were compared with those of a more general population. The subjects with diesel engine experience were found to be better at detecting gear rattle noise. The outcome of this research will be a method to predict how people perceive gear rattle noise so that component noise targets can be set that directly relate to human perception of gear rattle. When coupled with noise generation predictions, this approach can be used to optimize the quality of the component noise.

Degree

M.S.E.

Advisors

Bolton, Purdue University.

Subject Area

Automotive engineering|Mechanical engineering|Acoustics

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS