EVALUATION OF THE CONVENTIONAL AND ACOUSTIC INTENSITY METHODS OF NOISE SOURCE IDENTIFICATION FOR SMALL INTERNAL COMBUSTION ENGINES
Abstract
The conventional and acoustic intensity methods of noise source identification were evaluated in this study using both experimental and analytical means. As defined in this work, the conventional technique combined reverberation room sound power measurements, selective source wrapping and selective source operation. Using the conventional method, measurements were conducted for two eight horsepower engines, a Kohler Model K181 and Tecumseh Model HM80, to determine the primary noise sources at 12 combinations of engine speed and load. An extensive investigation was also conducted for the two-microphone acoustic intensity measurement technique. Several intensity measurements were conducted for sources ranging from simple loudspeakers to an eight horsepower engine. To further evaluate this technique, an investigation was performed to determine the errors introduced by instrumentation induced phase shifts, microphone misalignment, contamination, and the finite difference approximation in the intensity formulation. Comparing the two methods of engine source identification, the conventional technique provided the most accurate data. However, to obtain such data, extensive modifications were required to the engines for source isolation. Employing the intensity method, results similar to those obtained by the conventional method were possible in the frequency range from 500 to 4000 hertz. This limited accuracy was, however, compensated by the ease of the intensity measurements. Although not as accurate as the conventional method, the intensity measurement technique was shown to be potentially quite useful. The detailed error analysis performed for the intensity method indicated that inaccuracies as great as 60 decibels were possible for small phase errors. Also, the finite difference approximation and contamination were found to produce significant measurement errors. The results of this evaluation suggest that the intensity method can be accurately employed over a limited frequency range with care to avoid sources of error. It certainly is not, however, a universal measurement method.
Degree
Ph.D.
Subject Area
Mechanical engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.