The intense noise generated by high performance military fighter aircraft has a significant impact on communities near airbases as well as ground crews. This can impact basing decisions in addition to operations and training requirements. Prior to the initiation of this project, relatively little attention had been paid to the fundamental noise generation mechanisms of high performance supersonic military aircraft, or to methods for source noise reduction.

The objectives of this project were to:

  • Identify and test promising noise reduction concepts for military aircraft engines, in low cost, scale-model experiments.
  • Develop a methodology for using data obtained from testing at small and moderate scale, supported by computations, to reliably predict full scale engine noise.
  • Develop a constrained optimization design methods that minimizes shock-associated noise through nozzle internal shaping.
  • Assess installation effects on high performance military aircraft engine noise.
  • Gain a fundamental understanding of the source mechanisms in military aircraft engines.
  • Enhance an existing community noise prediction model.

Technical Approach

This program conducted an array of experiments of military style moderate (1/5) and small (1/25) scale model exhaust jets, supplemented with advanced numerical simulations. Aeroacoustic experiments and computations were conducted for a wide range of flow conditions and nozzle geometries directly relevant to high performance supersonic aircraft. The initial outcome of these studies has been the development of an improved fundamental understanding of the noise radiation mechanisms that have provided directions for future noise reduction methods.

A second major focus was the development of a scaling methodology to assist in the prediction of noise radiated by larger scale nozzle exhaust jets, including full-scale engines. The approach was to perform experiments at two scale sizes (1/5 and 1/25) and compare non-dimensional results for comparable geometries and flow conditions. The major result was to establish conditions under which the developed scaling methods work exceptionally well. In particular, the components of noise classified as large-scale structure noise, and broadband shock associated noise (BBSAN) were replicated well at the two experimental scales.

Two promising noise reduction techniques were explored at the two experimental scales. Chevrons with several geometric parameter variations and flow-field conditions were evaluated both experimentally and computationally. Since the flow in the vicinity of the chevrons is strongly Reynolds number dependent, the small scale experiments were not always capable of replicating the phenomena observed in the larger scale experiments. The reasons for these discrepancies were identified and this establishes a range under which small scale experiments should be interpreted with caution.


The small scale experiments were successful in replicating the results and noise benefit of jet nozzles with beveled exit planes. The experiments demonstrated significant noise reduction with little estimated thrust penalty. Further exploration of these nozzle concepts has formed a major position in future NASA experiments.

As a complement to the experimental studies, a numerical simulation methodology was developed. These large scale numerical computations were based on a solution of the short time-averaged equations of motion. The same nozzle geometries used in the experiments were replicated in the simulations. Comparisons between predictions for the baseline nozzles and experiment showed good agreement. In addition, a methodology to predict the effect of noise reduction devices – in particular chevrons – was developed. It is based on the Immersed Boundary Method. The results showed that this approach is capable of the efficient simulation of both the flow and noise from nozzles fitted with noise reduction devices.

An adjoint design method was also developed to automatically determine the optimum nozzle shape needed to achieve a desired pressure distribution inside the nozzle. Flow both with and without shocks were considered. This optimization approach has considerable potential for future use in aeroacoustic applications. 

The major results of the laboratory experiments on the baseline nozzles and the chevron nozzles were compiled into file formats convenient for incorporation into the community noise model being upgraded by Wyle laboratories. The upgraded version of this model, now called the Advanced Acoustic Model (AAM), has more realistic non-round nozzle capability as well as the capability to include twin jets and nozzles with chevrons in its predictions. This new capability will play an important role in most environmental impact studies associated with new air bases and/or deployment of new aircraft types into existing bases.


The understanding of the fundamental physical processes responsible for noise generation in high performance military aircraft, as well as the experimental and computational techniques developed in the research program, have provided directions for ongoing activities in jet noise reduction. In addition, the limitations observed in the noise predictions have resulted in the development of additional numerical techniques for their improvement.

An examination of existing noise reduction methods, some of which were examined as part of the research study, indicated that they had limited potential. In addition, other noise reduction methods, such the use of mechanical inserts into the diverging section of the nozzle, gave considerable performance penalties. This knowledge has shown a direction forward in jet noise reduction that is currently being pursued by the Principal Investigators under sponsorship from the US Navy and the Office of Naval Research. This work is also being pursued in cooperation with Lockheed-Martin Aeronautics Company and United Technologies Pratt & Whitney Division.

The basic idea is to use blowing in the divergent nozzle section to have an effect similar to the mechanical seal inserts, but with the ability to change the effective nozzle area ratio and shape fluidically. It is anticipated that this concept will reduce both broadband shock associated noise as well as the large-scale structure noise. The tools and methods developed in this project are being brought to bear to design, optimize, and demonstrate this noise reduction concept. Experiments are being conducted at small scale in the Penn State High Speed Jet Anechoic Facility. Numerical simulations are also being conducted to provide insight into the experimental observations. Finally, adjoint design methods for boundary control are being used to optimize the nozzle and blowing design. Preliminary results are promising with some noise reductions already achieved and flow simulations of the internal nozzle flow with blowing helping to interpret the experimental observations.