Considerations of Aeroacoustics in Turbojet-Engine Testing

Air Force Research Laboratory

Monday, June 01 2009

Recommendations for preventing aeroacoustic damage to test cells have been made.

A study of the aeroacoustics of a turbojet-engine test cell has been performed as one step in the development of a computational aeroacoustics capability (CAA) that could provide guidance for the design and operation of such cells. Ground testing of turbojet engines in test cells necessarily involves very high acoustic amplitudes, often severe enough to cause damage to test-cell equipment and to engines under test. Heretofore, the acoustic responses of test cells containing energetic jets have been poorly understood and generally unpredictable. The CAA capability is intended to enable prediction of deleterious acoustic events, making it possible to design test cells and choose operating conditions to prevent damage and thereby avoid the costly interruption of test schedules.

The study consolidated what is known about the aeroacoustics of jets and of flows in ducts like those of turbojet-engine test cells. In broad terms, the concepts addressed in the study are summarized in what follows: A turbojet- engine test cell is either a closed or a partially closed system. The source of acoustic energy in the cell is the jet flow. The solid duct surfaces that bound the cell reflect incident acoustic waves. Thus, the cell creates a special acoustic environment that, depending on the cell geometry and engine design, may interact directly with the jet flow.

One important part of the noise in the cell is broadband noise that propagates directly from the jet and is essentially the same as the noise propagating from a jet in open air. Another important part of the noise is that of acoustic resonances that arise in coupling of instability waves of the jet flow with acoustic normal modes of the cell (see Figure 1). Usually, this coupling results in tones at the resonance frequencies. In a case in which the frequency of a duct normal mode coincides or nearly coincides with the frequency of maximum growth of the associated instability wave, then the resonance is denoted a super resonance. Super resonance is violent: once it begins, it must be suppressed immediately, or else the engine test must be interrupted immediately, to prevent structural damage to the cell.

As a potential superior alternative to suppression or interruption, it has been proposed to design and construct cells to disrupt the duct modes that participate in super resonance. In particular, in the case of a hot circular-cross-section jet, the most amplified instability waves are those characterized by azimuthal wave number ±1. These are helical or flapping-mode jet waves that interact with acoustic waves in the corresponding helical and flapping duct normal modes. In principle, it should be possible to prevent super resonance by blocking the helical and flapping duct modes. This could be done by, for example, inserting barriers into the duct as shown in Figure 2.

Another product of the study is a set of recommendations concerning the development of a CAA capability for a turbojet-engine-testing laboratory. Summarizing the recommendations, the CAA capability should be developed and maintained by an in-house engineer, perhaps augmented with an engineering assistant who preferably has some expertise in CAA and/or computational fluid dynamics (CFD). The engineer and/or the assistant (perhaps augmented with consulting engineers having CAA and/or CFD expertise) should develop a battery of special-purpose CAA computer codes, each optimized for a specific class of CAA problems.

This work was done by Christopher Tam of Florida State University for the Air Force Research Laboratory. AFRL-0057