Similarity spectra-based decomposition of noise from an installed F404 engine

Abstract

Similarity spectra analysis has long been a popular method for identifying both the fine-scale and large-scale components of jet noise. For an imperfectly expanded supersonic jet, addition of a broadband shock-associated noise (BSN) model allows spectra to be better represented. A similarity spectra analysis of T-7A-installed GE-F404 engine noise has been performed at conditions ranging from intermediate to full afterburner. The BSN contributions are significant at military and afterburner conditions. The combined model captures the noise reasonably well with some caveats. First, spatiospectral lobes present in the measured noise are not well represented. Second, the measured low-frequency spectral slope is steeper at higher engine conditions in the region of maximum radiation. Third, the measured high-frequency slope is shallower across most of the radiation angles. Fourth, the noise at large inlet angles, beyond the peak radiation lobe, are not well represented by the combined model. The successes (and failures of the model) for different spatial regions and frequencies will aid in developing improved models for noise radiation. [Work supported by ONR.]