Prediction of Fan Exhaust Noise Propagation Steven A. E. Miller * and Philip J. Morris The Pennsylvania State University, University Park, PA, 16803, USA Yuan Zhao RMT Inc., Madison, WI, 53717, USA A method is presented to predict the far-field sound pressure levels from the fan exhaust. The levels are found based on a known source distribution of the acoustic pressure and velocity perturbations inside the exhaust fan duct. This source distribution is propagated through the fan exhaust shear layer to a porous Ffowcs Williams-Hawkings surface using the linearized Euler equations. The linearized Euler equations are discretized in the frequency domain with the Streamline Upwind Petrov Galerkin method on an unstructured grid and are solved in parallel. This technique enables a stable numerical solution of the linearized Euler equations to be obtained. Results are shown for the Source Diagnostics Test which has a realistic engine geometry and a high speed fan-exit flow with fan tones at a relatively high frequency. Comparisons are made between the predicted and measured far-field sound pressure levels at twice the blade passage frequency. Introduction During approach or engine cutback flight conditions, the noise of a high bypass ratio aircraft engine is dominated by fan related noise sources. The dominant noise sources are associated with the fan alone as well as fan-exit guide vane interactions. Once generated, these noise sources propagate inside the bypass duct and either propagate in the forward direction through the engine inlet or through the fan exhaust to the far-field. The fan noise consists of a finite number of tones at different frequencies as well as a broadband component. These frequencies are often visible in the Sound Pressure Level (SPL) at many observer locations in the far-field as multiple discrete tones that are often many times more intense than the broadband component of noise from other noise sources. The research addresses the propagation of fan noise through the jet engine exhaust to far-field observers. Given a description of the source in terms of a modal decomposition in the annular fan duct and a realistic meanflow, the far-field noise in the downstream arc can be predicted. The problem is difficult because the interior of the duct and nozzle often have very complicated geometries that do not allow for simulations with standard finite difference schemes. Also, the mean flow from the duct exhaust is not irrotational and possesses strong velocity gradients. The linearized Euler equations (LEE) are a very good approximation for the description of acoustic prop- agation through non-uniform mean flows. The LEE are formed by taking the compressible Euler equations and replacing the field variables with the sum of their ensemble averages and a perturbation. These equations can be solved either in the frequency or time domain. One such time domain approach by Zhang, 1 solved the LEE to simulate fan exhaust propagation. The main advantage of a time domain approach is that it has the ability to simulate multiple frequencies simultaneously; essentially solving for all the frequency compo- nents of the fan tones simultaneously. Unfortunately, the time domain approach triggers a Kelvin-Helmholtz instability in the shear layer of the fan exhaust. This is a physical hydrodynamic instability present in all shear flows, which causes waves growing in amplitude in the downstream direction in both the jet primary and secondary flows. Physically, Kelvin-Helmholtz instabilities are limited in amplitude and their energy is transferred to smaller scales by nonlinear effects. In principle, the amplitude of any instabilities generated * Graduate Research Assistant, Department of Aerospace Engineering, 229 Hammond Building, AIAA Member Boeing/A. D. Welliver Professor, Department of Aerospace Engineering, 229 Hammond Building, AIAA Fellow Process Engineer, 744 Heartland Trail, AIAA Member 1 of 18 American Institute of Aeronautics and Astronautics 15th AIAA/CEAS Aeroacoustics Conference (30th AIAA Aeroacoustics Conference) 11 - 13 May 2009, Miami, Florida AIAA 2009-3145 Copyright © 2009 by Steven A. E. Miller, Philip J. Morris, and Yuan Zhao. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission.