Skip to main content
eScholarship
Open Access Publications from the University of California

UC Irvine

UC Irvine Electronic Theses and Dissertations bannerUC Irvine

Experimental and Theoretical Investigation of the Emission and Diffraction of Discrete Tone Noise Generated from the Exhaust of a Ducted Fan

Creative Commons 'BY-NC' version 4.0 license
Abstract

In modern turbofan engines powering commercial aircraft, the fan is becoming the dominant source of noise at takeoff and landing. Understanding and modeling of the fan noise source, and the interaction of the emitted sound with the airframe, are critical for the design of quiet aircraft. Of particular importance is discrete tone noise radiating from the fan exhaust whose propagation can be shielded by the airframe in advanced aircraft configurations. Current prediction methods for fan noise and its propagation require tremendous amount of computational resources and time. Experiments in large-scale facilities are extremely expensive. There is a need for efficient approaches to experimentally investigate and model fan noise, thus enabling parametric studies that can identify optimal configurations. This study combines novel small-scale experiments with low-order, physics-based modeling of the fan noise source towards achieving the aforementioned goal.

The experimental effort entailed the design and construction of a subscale ducted fan rig that includes all the relevant components of the turbofan engine and simulates accurately the sound emission generated by the fan of such engines. The ducted fan includes a nacelle, rotor, and stators, all fabricated using advanced stereolithographic or metal casting methods. It is powered by a high-performance DC motor and achieves rotor tip Mach number of around 0.61 and fan pressure ratio of 1.157, values compatible with the operation of high-bypass turbofan engines. Acoustic diagnostic was conducted inside an anechoic chamber using far-field and near-field phased arrays consisting of 23 microphones. Installation of a rectangular flat plate representing the airframe below the ducted fan recreates complex phenomenon such as scattering off the obstacle, and diffraction around the shield. Addition of the rectangular plate shield generated complex trends in the tonal content and demonstrates the large potential to reduce noise through shielding by the airframe. Tones below the shielding surface were well attenuated, while tones emitted in the aft direction were unchanged. Acquired acoustic data were used to formulate the wavepacket noise model and shielding simulations.

The theoretical effort comprises the following steps: (a) extraction of the harmonic content of the measured noise through use of the Vold-Kalman filter; (b) modeling of the aft-emitted fan noise source as a cylindrical wavepacket with azimuthal modes inferred from the Tyler-Sofrin theory; (c) determination of wavepacket shape parameters through least-squares matching of the measured cross-spectral density in the near field and the far field at a given frequency; and (d) propagation and diffraction of the sound from the modeled source using the Boundary Element Method (BEM), with comparisons to measured data.

Far-field source parameterization shows that the wavepacket can be modeled as a short cylindrical disturbance surrounding the exit of the nozzle. The resulting modeled sound field captures the most prominent experimental far-field and near-field magnitude and phase relations across numerous microphones. Near-field parameterization expressed similar values to the far-field parameter optimization. This consistency of the cross-spectra in the near and far-field justifies the wavepacket's applicability as an equivalent discrete tone noise source for sound generated by the exhaust of a ducted fan. Numerical shielding predictions using Boundary Element Method (BEM) exhibit complicated changes in far-field noise radiation. Reasonable agreement in average noise reductions is found with experimental data.

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View