Ching-Wen Kuo

On the Scaling of Small, Heat Simulated Jet Noise Measurements to Moderate Size Exhaust Jets

Modern military aircraft jet engines are designed with variable geometry nozzles to provide optimum thrust in different operating conditions, depending on the flight envelope. However, the acoustic measurements for such nozzles are scarce, due to the cost involved in making full scale measurements and the lack of details about the exact geometry of these nozzles. Thus the present effort at The Pennsylvania State University and the NASA Glenn Research Center in partnership with GE Aviation is aiming to study and characterize the acoustic field produced by supersonic jets issuing from converging-diverging military style nozzles. An equally important objective is to validate methodology for using data obtained from small and moderate scale experiments to reliably predict the most important components of full scale engine noise. The experimental results presented show reasonable agreement between small scale and moderate scale jet acoustic data, as well as between heated jets and heat-simulated ones. Unresolved issues however are identified that are currently receiving our attention, in particular the effect of the small bypass ratio airflow. Future activities will identify and test promising noise reduction techniques in an effort to predict how well such concepts will work with full scale engines in flight conditions.

Experimental Comparison of Supersonic Jets Exhausting from Military Style Nozzles with Interior Corrugations and Fluidic Inserts

This paper describes the development and analysis of fluidic inserts for supersonic jet noise suppression. The study uses scale model experiments of jets that simulate the exhaust jets from typical lowbypass ratio military jet aircraft engines during takeoff. The fluidic inserts use distributed blowing in the divergent section of the nozzle to simulate mechanical corrugations, while having the advantage of being an active control method. Measurements with simulated forward flight are important early in the analysis of noise reduction methods. The current design of the delivery piping is not streamlined enough for forward flight measurements. Therefore, forward flight acoustic measurements were first conducted with hardwall corrugation nozzles. The noise reduction of the hardwall corrugation nozzles was not affected by simulated forward flight. Mean flow field measurements of the jet plume were conducted with both hardwall corrugation and fluidic insert nozzles. This led to the development of a newer fluidic insert nozzle fabrication technique. The new nozzles, along with supplying varying injection pressures to upstream and downstream injection ports, have produced the most noise reduction to date. Noise reduction has been optimized for two different overexpanded jet conditions and is near 5 dB in the peak noise emission direction at low polar angles and 3 dB off of the broadband shock associated noise in the sideline direction. These results were obtained with injection mass flow rates less than 4% of the core jet flow rate.

Effects of Jet Noise Source Distribution on Acoustic Far-Field Measurements

Jet noise production is well known to be of a distributed nature along the jet, with high frequency noise components radiating from locations close to the nozzle exit and low frequency noise being produced farther downstream, around the end of the potential core. Such a distributed source implies that measurements need to be made at a significant distance from the source in order to be in the true geometric (acoustic) far field. The current study presents measurements of fully expanded Mj = 1.5 jets operating with cold air and heat simulated at TTR = 2.2 made at various positions in the acoustic field, some short of the minimum distance required to be in the true geometric far field. A close look is taken at the details of the noise generation region in order to better understand the mismatch between spectra measured at various acoustic field radial locations. A processing method is presented to correct for near-field effects and efficiently compare near and far field spectra with unprecedented accuracy. This technique is then further used to clarify the nonlinear propagation effects that can be observed at the high frequency end of high speed jets noise.

Methods to Improve the Accuracy of Acoustic Measurements in Small Scale High Speed Jets

This paper focuses on methods to improve the accuracy of acoustic measurements in small scale high speed jet experiments. In such experiments the acoustic spectra contain significant energy up to 100 kHz which challenges the accuracy of present methods. Different physical effects alter the response of conventional 1/8” condenser microphones in this high frequency range. The electronic circuitry produces actuator variations in response with frequency. Refraction effects around the sensitive portion of the microphone produce free-field variations with frequency. Angulations of the microphone with the oncoming acoustic field also produce significant free-field variations. While examining these effects, an extensive series of measurements in the acoustic field of transonic and supersonic jets has demonstrated that for improved accuracy, the published free-field response of the microphones needs to be substantially altered to provide (significantly) improved accuracy of the measurements. This is required when the microphones are aligned for grazing incidence which can also provide some improvement in performance. The current study proposes to replace the published free-field response by an empirical free-field response experimentally determined. In addition the atmospheric attenuation correction is being simultaneously applied to all data. Application to measured data gives encouraging results validating the capability of the method to produce accurate measurements even at the highest response frequencies of the microphones.

Supersonic Jet Noise Reduction by Nozzle Fluidic Inserts with Simulated Forward Flight

The noise produced by supersonic, high temperature jets that exhaust from military aircraft is becoming more of a disturbance. Methods to reduce the noise produced from these jets in a realistic full-scale environment is difficult. This study describes the development and analysis of fluidic inserts for supersonic jet noise suppression. Distributed blowing within the divergent section of the military-style convergent divergent nozzle alters the shock structure of the jet in addition to creating streamwise vorticity for the reduction of mixing noise. Enhancements to the fluidic insert design have been performed along with experiments for a large number of injection parameters and core jet conditions. It has been shown that the noise reduction of the fluidic inserts is most heavily dependent on the momentum flux ratio between the injector and core jet. Maximum reductions of approximately 5.5 dB OASPL have been observed with practical mass flow rates and injection pressures. The first measurements with fluidic inserts in the presence of a forward flight stream have been performed. Optimal noise reduction occurs at similar injector parameters in the presence of forward flight. Fluidic inserts in the presence of a forward flight stream were observed to reduce the peak mixing noise by nearly 4 dB OASPL and the broadband shock-associated noise by nearly 3 dB OASPL.

Experimental Investigation of Near-Field Pressure Fluctuations Generated by Supersonic Jets

This paper describes an experimental investigation of near-field pressure fluctuations generated by round supersonic jets. Previous investigations have shown that there are two pressure components that exist in the near field of a subsonic jet. These are the hydrodynamic pressure fluctuations and the acoustic pressure perturbation components that form two distinct humps in the pressure-wavenumber spectrum (when the wavenumber is calculated directly from the radian frequency and the ambient acoustic velocity and is nondimensionalized by the radial distance from the middle of the jet shear layer). Microphone arrays have been used to measure the near-field pressure and auto and cross spectra between all microphone pairs. The double humped pressure-wavenumber spectra are not present in the near field of a highly supersonic jet where the acoustic Mach number is Ma = 1.85 and the spectra are dominated by the acoustic perturbations that radiate directly to the far field. The spectra in the intermediate subsonic range exhibit a reduced double humped behavior. The processing technique, Empirical Mode Decomposition, can assist in the identification of the components of the near field. Fourier mode decomposition is performed, including both + and – helical modes, including contributions from all 64 components of the cross spectral matrix obtained from the 8 microphone azimuthal array. This shows that the composition of the lower order modes did not change substantially between the subsonic and high supersonic jets. Proper Orthogonal Decomposition is also used to produce mode spectra only slightly different from those obtained with the Fourier azimuthal decomposition.

Acoustic assessment of small-scale military-style nozzles with chevrons

With the emergence of more powerful fighter aircraft, supersonic jet noise reduction devices are being intensively researched. Small-scale measurements are a crucial step in evaluating the potential of noise reduction concepts at an early stage in the design process. With this in mind, the present study provides an acoustic assessment of small-scale military-style nozzles with chevrons. Comparisons are made between the present measurements and those made by NASA at moderate scale. Measurements made with baseline nozzles (without chevrons) show excellent agreement with NASA data establishing the accuracy of the scaling methodology. The effect of chevrons on supersonic jets is then investigated for cold jets, highlighting the crucial role of the jet operating conditions on the effects of chevrons on the jet flow and subsequent acoustic benefits. At low Reynolds numbers (small scale) the penetration of the chevrons in the jet flow is the most important chevron parameter in reducing the generated noise. A small-scale heat simulated jet is investigated in the over-expanded condition and shows no substantial noise reduction from the chevrons. This is contradictory to moderate-scale measurements. The discrepancy is attributed to a Reynolds number low enough to sustain an annular laminar boundary layer in the nozzle that separates in the over-expanded flow condition. Transition of the boundary layer to turbulent flow is induced with inner roughness of the nozzle and results in more noise reduction with the chevrons. The resulting effect is comparable to results from NASA, seemingly validating the hypothesis made that jets of too low Reynolds number cannot be used directly to accurately measure chevron noise reduction. These results are important in assessing the limitations of small-scale measurements in this particular jet noise reduction method.

Advanced acoustic assessment of small-scale military-style nozzles with chevrons

With the emergence of more powerful fighter aircraft, supersonic jet noise reduction devices are being intensively researched. Small scale measurements are a crucial step in evaluating the potential of noise reduction concepts at an early stage in the design process. With this in mind, the present study provides an acoustic assessment of small-scale military-style nozzles with chevrons. Comparisons are made between the present measurements and those made by NASA at moderate scale. Measurements made with baseline nozzles (without chevrons) show excellent agreement with NASA data establishing the accuracy of the scaling methodology. The effect of chevrons on supersonic jets is then investigated for cold jets, highlighting the crucial role of the jet operating conditions on the effects of chevrons on the jet flow and subsequent acoustic benefits. At low Reynolds numbers (small scale) the penetration of the chevrons in the jet flow is the most important chevron parameter in reducing the generated noise. A small scale heat simulated jet is investigated in the overexpanded condition and shows no substantial noise reduction from the chevrons. This is contradictory to moderate scale measurements. The discrepancy is attributed to a Reynolds number low enough to sustain an annular laminar boundary layer in the nozzle that separates in the over-expanded flow condition. Transition of the boundary layer to turbulent flow is induced with inner roughness of the nozzle and results in more noise reduction with the chevrons. The resulting effect is comparable to results from NASA, seemingly validating the hypothesis made that jets of too low Reynolds number cannot be used directly to observe and measure chevron noise reduction. These results are important in assessing the limitations of small scale measurements in this particular jet noise reduction method.

Experiments on the Effect of Ground Reflections on Supersonic Jet Noise

*† ‡ The influence of ground reflections on the measurement of aircraft engine exhaust noise is examined in this paper. A series of experiments were performed in the UCI aeroacoustic facility of supersonic jet noise with and without a ground plane in close proximity to the jet. Thus the effect of the ground plane on the radiated noise was isolated. Additionally, a computational model for the phenomenon was developed, which included the determination of a distribution of noise sources within the jet column. This distribution was developed from phased microphone array measurements incorporating an advanced beamforming algorithm. The developed model did a reasonably good job of predicting the effect of the reflecting ground surface. The most important exception is in the additional noise source caused by aerodynamic “scrubbing” of the turbulent jet on the surface far downstream of the jet exit. Guidance on how to alter the proposed model to account for porosity of the ground in field experiments over grassy or sandy terrain, is also given.

Effects of Supersonic Jet Conditions on Broadband Shock-Associated Noise

This paper describes an experimental study of the effects of jet heating on broadband shockassociated noise (BBSAN). Far field noise measurements are made for three nozzle geometries – a convergent nozzle, and smooth convergent-divergent nozzles with design Mach numbers 1.5 and 1.76. The effect of jet heating is simulated with the use of helium-air mixtures. It is shown that the effects of simulated heating on the BBSAN give good agreement with measurements in jets using heated air. Measurements for unheated jets are presented and are consistent with previous BBSAN experiments. The noise measurements are supplemented with schlieren flow visualizations that show that the appearance of a Mach disk in the shock cell structure is consistent with the saturation of the BBSAN levels for both highly under- and over-expanded jets. The effects of jet heating are then described for the design Mach number 1.5 nozzle operating at four Mach numbers – two over-expanded (1 .2,1.4) j M = and two under-expanded (1 .7,1.9) j M = . Total temperatures in the range 1.0 to 2.2 are considered in increments of 0.2. These values are chosen to supplement existing measurements at higher total temperature ratios and to examine the effects of low levels of heating. It is shown that the peak BBSAN rapidly approaches a saturation value in all cases. In the under-expanded cases, the peak BBSAN level decreases with increasing total temperature ratio and the reverse is true for the over-expanded cases. However, these changes are always small, having maximum values of 4 ± dB. The Strouhal number of the primary BBSAN peak also varies only slightly, being within 20% of the unheated jet value. Again, there is a measureable difference between the under-and over-expanded cases with the peak frequency increasing in the former case and falling in the latter. It is also shown that the characteristic spectral shape of BBSAN is very similar for all the cases considered (with and without heating), and is described very well by the theory developed by Tam [13,14]. The variation of the peak frequency, peak sound pressure level and spectral width with jet Mach number and total temperature ratio are described. This is used as the basis for a proposed empirical prediction model for BBSAN.