Richard Gaeta

Subtle Differences in Jet Noise Scaling with Narrow Band Spectra Compared to 1/3-Octave Band

The purpose of this paper is twofold: First, a set of narrowband noise spectra for an unheated round subsonic jet is produced that can be compared with other nozzles and other facilities as well as for evaluating computational or analytical methods. Second, the subtle differences in dealing with narrowband and 1/3-octave spectra when analyzing jet noise are highlighted and a general method is presented to adjust experimental data of any bandwidth. The spectral character of noise generated from an unheated round jet is examined for a range of subsonic jet velocities for a convergent nozzle of given size. Following established scaling guidelines, the 1/3-octave spectra at a polar angle of 90 0 (produced from the narrowband data) can be collapsed to a single curve based on a non-dimensional frequency and the eighth power law of the jet exit velocity. At angles normal to the jet axis, the peak non-dimensional frequency corresponds to a value of approximately unity. Performing the same analysis with the narrowband spectra reveals two things: (1) The peak non-dimensional frequency normal to the jet axis corresponds to a value of approximately 0.3 and (2) in order to achieve a good collapse of the spectra with velocity, the narrowband data must be normalized per unit non-dimensional frequency. The former point while intuitive and mostly understood, is important for those interested in prediction schemes (namely computational efforts) and to designers who want to use a universal jet noise spectra and determine the actual peak frequency of interest (at 90 0 ). The latter point is a nuance that results in the scaling of jet noise effectively as the seventh power of velocity for narrowband spectra. Finally, the present data is compared with well accepted subsonic jet noise data from the literature using a scaling method that can be used for both 1/3-octave data and narrowband data.

Noise Scaling for Unheated Low Aspect Ratio Rectangular Jets

A unique set of acoustic and fluid dynamic data was obtained for three converging rectangular jets of aspect ratios 1.5, 4.0, and 8.0 and a round nozzle. All four nozzles had the same exit area and were tested in the same facility for static and unheated conditions for a range of subsonic and slightly supersonic Mach numbers. Farfield noise measurements were made for polar angles ranging from 40° to 110° with respect to the jet axis and for azimuthal angles that corresponded to both a flyover and a sideline condition. On a spectral basis, it was found that the rectangular jets were typically louder in the high frequency range and quieter in the low frequency range. Attempts were made to collapse the acoustic data for all of the jets on a non-dimensional basis and thus determine the functional dependency of the noise upon the nozzle geometry. The process of collapsing the acoustic data mirrored a previously performed collapse for round jets. Upon discovering that existing methods for scaling round jet noise would not be sufficient to collapse the rectangular jet noise, additional scaling laws were adopted. Physical justification for the these scaling laws are provided.

Development of a Prediction Method for Jet Installation Noise: Reflection/Shielding

Aircraft noise has been a source of annoyance for communities near airports for decades, and noise regulations imposed by governments and airports have forced aircraft manufacturers to include noise as a key metric during design. Historically, much of the aircraft noise reduction has been achieved through increases in bypass ratio, but government and industry are increasingly investigating other avenues to reduce community noise. The primary approach is to place the engines above the airframe surfaces, expecting substantial noise reductions at observers on the ground through shielding, rather than attempting to reduce the level of the noise source itself. The prediction of jet installation noise during conceptual design is therefore very important to enable accurate predictions of the benefit obtained through installation effects. However, no basic prediction methods for reflection/shielding are commonly used which are appropriate for conceptual design but still can handle the distributed nature of jet noise. The available methods are even more limited with respect to other interactions between the jet and wing, such as scrubbing. This research lays the foundation for a prediction model for jet installation noise, starting with a first-order model for the prediction of the reflection/shielding of jet by a surface. Modelscale data has been obtained in the Anechoic Flight Simulation Facility at Georgia Tech Research Institute for a wide range of configurations and jet and tunnel velocities, enabling prediction for configurations beyond the conventional under-the-wing engine placement. The results presented in this paper identify the region in which reflection/shielding is the only acoustic installation effect and develop a methodology for a prediction method for shielding, supported by the data presented herein.

The Sound Absorbing Potential of Carbon-Graphite Foam

Carbon-Graphite (C-G) foam is a relatively new material developed by Oak Ridge National Laboratory that has three remarkable properties: 1) It is an excellent “heat spreader”, possessing a high specific thermal conductivity, 2) It is a very good EMI shield and 3) It is an excellent bulk sound absorber above 1 kHz. It is an enabling material that could radically improve the thermal efficiency of power producing machines. Moreover, its multi-functionality can be exploited to produce integrated thermal and acoustic management systems. The basic structure of the carbon-graphite foam is a series of cells that are between 100 and 500 mm in diameter which are interconnected with pores on the order of 50 mm. The highly aligned graphitized carbon atoms in between the cells possess a thermal conductivity similar to diamond. This paper focuses on characterizing the bulk acoustic properties of this unique material. Normal incidence impedance are measured and characteristic impedance are calculated from 300 Hz up to 6 kHz. The effects of density, metallic coating, and bulk temperature on the sound absorption performance are determined. Results indicate that above 1000 Hz, the C-G foam is an excellent absorber, having absorption coefficients from 80 to 90%.. The C-G foam compares very well with fiberglass, a known excellent bulk sound absorber. Indeed, attenuation rates of the C-G foam are shown to be higher than those of fiberglass. Finally, the role its high heat conductivity plays or doesn’t play in its ability to absorb sound is explored. Material temperatures from -35 F to +178 F produced and found not to significantly affect the sound absorption.

Design and Manufacturing of Composite Propellers for SUAS

This paper presents a method for creating multi-bladed composite propellers that are optimized to a specific vehicle’s thrust and noise requirements. The composite propellers are experimentally tested to verify their performance parameters. Using carbon fiber weave and various core materials, composite propellers were manufactured. The technique harnesses the use of wet lay-up composites and molds to create beautifully crafted propellers that are comparable with those commercially produced in industry. This method produces high quality propellers that are comparable to commercially available ones with the advantage of better vehicle performance. Testing results and comparison with theoretical predictions are presented.

Noise Measurements of Tactical UAVs

This paper describes several noise measurement activities of GTRI to measure and understand the noise produced by tactical UAVs. As UAVs in the tactical class have begun to see much more use in the battlefield, it has become apparent that the noise produced by these aircraft affects their utility in certain roles. GTRI has thus undertaken to measure the noise of several UAVs through a series of static ground measurements, flyover measurements, and measurements taken in an anechoic chamber. The results indicate that propeller noise and engine exhaust noise are generally of equal importance for typical UAVs. The difficulties in measuring flyover noise are also discussed as are methods for extracting noise buried in back ground noise. Finally some discussions on UAV noise relative to UAV detection is provided along with some concepts which would serve to reduce UAV noise and thus reduce detectability.

Design Considerations for Developing Composite Materials Providing Improved Acoustic Transmission Loss for UAVs

Research is currently underway at Oklahoma State University to develop light-weight composite materials providing high acoustic transmission loss for use in the fabrication on Unmanned Aircraft Systems (UAS). These materials, when used to fabricate UAS airframes, would provide acoustic shielding that limits breakout noise from the vehicle. Effort is currently given to the development of design strategies and test apparatus for comprehensive acoustic evaluation of component design and material construction, and will soon focus on the design of new hybrid composite materials and structures providing improvements in acoustic performance.

Development of a Noise Prediction Model for a Cruise Friendly Circulation Control Wing

New aircraft concepts that are to meet challenging cruise efficiency, short field takeoff, and noise goals will have to rely on innovative high lift technology. One such technology is the Circulation Control Wing (CCW) which leverages a thin high aspect ratio jet flowing over a curved flap surface to generate tremendous lift coefficients. Of concern is the impact that this technology will have on the overall noise of the aircraft. Some acoustic studies have reported noise characterization of such a CCW with extremely thick airfoils and trailing edges. More practical CCWs must be “cruise friendly”. It is the objective of the current paper to address the noise character of such a CCW configuration and to begin to develop a noise prediction model to assess new aircraft configurations using this technology. A current NASA Fundamental Aeronautics Subsonic Fixed Wing project being conducted by Georgia Tech Research Institute (GTRI) and teammate California Polytechnic State University involves the development of a CCW integrated with engines located over the wing (OTW) to improve the powered high-lift and cruise performance of Cruise Efficient STOL (CESTOL) aircraft. As part of this program, a series of aeroacoustic tests on a 2D model scale CCW are conducted. Far field acoustic data was acquired to quantify the far field noise as a function of aft flyover angles, free stream velocity, and slot jet velocity. It is found that an asymmetrical directivity pattern emerges where higher frequency noise is shielded to observers below the bottom of the flap chord. A 5 order velocity scaling collapses spectra very well for high frequencies and a 6 order works well for lower frequencies. Finally, these data are used as a basis for predicting the community noise of a CESTOL aircraft relative to a conventional take-off and landing aircraft. It was found that using a CCW in conjunction with a STOL trajectory has the potential for reducing overall aircraft noise, however a more complete set of acoustic data on a highly fidelity test article is needed to reduce the uncertainty of the prediction.

Optimal Propeller Design for Quiet Aircraft using Numerical Analysis

The present paper explores methods for reducing aircraft propeller noise with minimal losses in performance. A standard two blade propeller configuration was examined using numerical analysis. A number of blade design modifications that were investigated to reduce the noise signature including increasing the number of blades, adjusting the chord length, beta distribution, radius of the blade, airfoil shape, and operating RPM. In order to determine the optimal blade design, a baseline case is first developed and the design parameters are then varied to create a new propeller design that reduces the sound pressure level (SPL) while maintaining performance levels within a predetermined range of the original specifications. From the analysis, the most significant improvements observed in lowering the acoustic signature are dominated by operating rpm and blade radius. A three-, four-, and five-blade configuration was developed that reduced the SPL generated by the propeller during cruise flight conditions. The optimum configuration that produced the greatest SPL reduction was the five-blade configuration. The resulting sound pressure level was reduced from the original 77 dB at 1000’ ft above ground level (AGL), to 54 dB at 1000’ AGL while remaining within 1.4% of the original thrust and efficiency.

Validity of the Point Source Assumption of a Rotor for Farfield Acoustic Measurements with Shielding

Many theoretical models of shielding by hard surfaces assume the source to be a point monopole source. If one is interested in shielding the noise of a rotor system by interposing a hard surface between the rotor and the observer, can the rotor system really be considered to be a monopole? If rotating noise sources are under consideration what is the effect of configuration and de- sign parameters? Exploring the validity of point source assumption alluded to above for a rotor for far field acoustic measurements with and without shielding forms the backbone of the present work. This paper will investigate the validity of acoustic shielding models that are developed under the assumption that the source is a monopole, compared with the shielding of real rotor noise sources. To the authors knowledge, the present research is the first ever work that seeks to answer the question of whether rotor noise from a propeller or a fan can really be approximated by a point source (monopole) as is done in theoretical calculations when characterizing the effectiveness of acoustic shields. The effect of sound scattered and diffracted by the shield placed around or near the actual source of rotor noise is investigated and compared with data acquired using a point source. Shielding configurations consist of rectangular plates and ducts of various dimensions.