Indirect combustion noise of auxiliary power units

Abstract

Recent advances in noise suppression technology have significantly reduced jet and fan noise from commercial jet engines. This leads many investigators in the aeroacoustics community to suggest that core noise could well be the next aircraft noise barrier. Core noise consists of turbine noise and combustion noise. There is direct combustion noise generated by the combustion processes, and there is indirect combustion noise generated by the passage of combustion hot spots, or entropy waves, through constrictions in an engine. The present work focuses on indirect combustion noise. Indirect combustion noise has now been found in laboratory experiments. The primary objective of this work is to investigate whether indirect combustion noise is also generated in jet and other engines. In a jet engine, there are numerous noise sources. This makes the identification of indirect combustion noise a formidable task. Here, our effort concentrates exclusively on auxiliary power units (APUs). This choice is motivated by the fact that APUs are relatively simple engines with only a few noise sources. It is, therefore, expected that the chance of success is higher. Accordingly, a theoretical model study of the generation of indirect combustion noise in an Auxiliary Power Unit (APU) is carried out. The cross-sectional areas of an APU from the combustor to the turbine exit are scaled off to form an equivalent nozzle. A principal function of a turbine in an APU is to extract mechanical energy from the flow stream through the exertion of a resistive force. Therefore, the turbine is modeled by adding a negative body force to the momentum equation. This model is used to predict the ranges of frequencies over which there is a high probability for indirect combustion noise generation. Experimental spectra of internal pressure fluctuations and far-field noise of an RE220 APU are examined to identify anomalous peaks. These peaks are possible indirection combustion noise. In the case of the APU RE220, such peaks are identified. The frequency ranges of these peaks are found to overlap those predicted by the model theory. Based on this agreement, a tentative conclusion is drawn that there is good reason to believe that APUs do generate measurable indirect combustion noise.This paper is dedicated to the memory of Prof. Phil Doak for his numerous contributions to Aeroacoustics and the Journal of Sound and Vibration.