Multirotor drones are becoming increasingly popular in both the civilian and military sectors of our society. These compact gadgets come in a variety of sizes with the smallest ones measuring less than two inches in diameter, while larger ones can be in excess of five feet. Surprisingly, very little is known about their acoustical footprint, which is becoming a topic of broad importance given that these vehicles most often operate in populated areas. Thus, the objective of this paper is to provide a first principles understanding of the acoustical characteristics of hovering drones. To accomplish this, a new test stand was constructed at the Applied Research Laboratories at The University of Texas at Austin for studying various multirotor drone configurations. The drone test stand is capable of powering up to eight DC electric motors with adjustable arms that allow different rotor diameters to be tested. Rotor diameters ranging from 8 in to 12 in are studied and with configurations comprised of an isolated rotor, a quadcopter configuration and a hexacopter configuration. A six degree-of-freedom load cell is used to assess the aerodynamic performance of each drone configuration. Meanwhile, an azimuthal array of 1/2-inch microphones is placed between 2 and 3 hub-center diameters from the drone center thereby allowing the acoustic near-field to be quantified. The analysis is performed using standard statistical metrics such as Sound Pressure Level and Overall Sound Pressure Level and is presented to demonstrate the relationship between the number of rotors, the drone rotor size and it’s aerodynamic performance (thrust) relative to the far-field noise.
The vibroacoustic loads that form during the startup of both rigid and compliant wallhigh area ratio nozzles is investigated. The rigid wall nozzle is fabricated from 6061 aluminum while the compliant wall nozzles are formed from urethane-based elastomers in orderto invoke aeroelastic coupling between the nozzle wall and the internal flow. Single pointmeasurements of the nozzle lip displacement are synchronized with a pressure field microphone located behind the nozzle where the base of a vehicle would reside. Particularattention is drawn to the sound field during transition from free-shock separated flow torestricted shock separated flow, as well as the end-effects regime loads. The findings revealthe sensitivity of the vibroacoustic loads to the aeroelasticity of the nozzle wall duringcritical stages in the startup process.
The use of helium-air mixtures to simulate the effects of elevated temperatures in aeroacoustics is plagued by the inability to match exactly the density and sound speed ratios between the jet flow and the ambient field, all the while maintaining the same gas dynamic Mach number and jet exit velocity. Real heated jet flows are typically achieved using either propane combustion in air or kerosene combustion in air, which results in the formation of carbon-dioxide and water vapor byproducts. In an effort to level the playing field between the heat simulated helium-air mixture system and the air breathing combustion system, a theoretical model is developed to isolate the effect of combustion byproducts on these aeroacoustic parameters to see if similar discrepancies arise. The motivation is to narrow the gap between laboratory and full-scale jet noise testing. Gas properties from the new combustion model are validated by laboratory measurements of a real propane combustion system as well as outputs from NASA’s Chemical Equilibrium with Applications code. The findings reveal how the additional combustion byproducts from propane combustion in air and kerosene combustion in air have a negligible effect on the parameters relevant to jet noise. Closer inspection of the helium-air mixture system demonstrates how variations in the Mach wave radiation angle at moderate pressure and temperature ratios of the nozzle is accurate to within a couple of degrees, relative to a pure heated air system. Similar accuracy is reported with the far-field intensity.
The unsteady wall pressure on the aft deck of a multi-stream, planar supersonic nozzle is studied over a range of nozzle operating conditions corresponding to independent changes to the core and bypass stream pressure ratios. The data are processed using time-frequency analysis and reveal various tones corresponding to transonic resonance as wellunsteady interactions of both separation and reflection shocks with the developing boundary layer. The position of the separation shock is shown to experience significant hysteresis effects, which subside at pressure ratios well above the design pressure ratio of the nozzle. Shadowgraphy images of the exhaust plume are also presented, which are then analyzed using the snapshot form of proper orthogonal decomposition. The findings from this low-dimensional analysis demonstrates how the first most energetic mode highlights the shock cell patterns whereas the second most energetic mode elucidates turbulence motions in the plume.
A nonintrusive measure of the exhaust plume and immediate sound field produced by a cluster of two thrust optimized parabolic contour nozzles is studied during two steady-state conditions. The first condition is at a nozzle pressure ratio of 25, at which point the flow is in a restricted-shock separated state. The second condition is at a nozzle pressure ratio of 37 and is when the flow and internal shock pattern transition rapidly between free-shock separated flow and the end-effects regime. These end-effects regime pulsations produce significant vibroacoustic loads due to the intermittent breathing of the last trapped annular separation bubble with the ambient. The exhaust plumes and surrounding sound field are first visualized by way of retroreflective shadowgraphy. Radon transforms of the spatially resolved shadowgraphy images are then used to characterize the statistical behavior of the acoustic wave fronts that reside within the hydrodynamic periphery of the nozzle flow. The findings reveal quantitative evidence of the sources of most intense vibroacoustic loads during the end-effects regime of clustered rockets.
The effect of stagger startup on the vibroacoustic loads that form during the end-effects regime of clustered rockets is studied using both full-scale (hot-gas) and laboratory-scale (cold-gas) data with vehicle geometry. Both configurations comprise three nozzles with thrust-optimized parabolic contours that undergo free-shock separated flow and restricted-shock separated flow as well as an end-effects regime before flowing full. Acoustic pressure waveforms recorded at the base of the nozzle cluster are analyzed using various statistical metrics as well as time-frequency analysis. The findings reveal a significant reduction in end-effects regime loads when engine startups are staggered. However, regardless of stagger, both the skewness and kurtosis of the acoustic pressure time derivative elevate to the same levels, thereby demonstrating the intermittence and impulsiveness of the acoustic waveforms during the end effects regime.
A method for calculating the effective Gol’dberg number for diverging waveforms is presented, which leveragesknown features of a high-speed jet and its associated sound field. The approach employs a ray tube situated along the Mach wave angle where the sound field is not only most intense, but advances from undergoing cylindrical decay to spherical decay. Unlike other efforts, a “piecewise-spreading regime” model is employed, which yields, separately, effective Gol’dberg numbers for the cylindrically and spherically spreading regions in the far field. The new approach is applied to a plethora of experimental databases, encompassing both laboratory- and full-scale jet noise studies. The findings demonstrate how cumulative nonlinear distortion is expected to form in the acoustic near field of laboratory scale round jets where pressure amplitudes decay cylindrically; waveform distortion is not expected in the acoustic far field where waveform amplitudes diverge spherically. On the other hand, where full-scale jet studies are concerned, effective Gol’dberg number calculations demonstrate how cumulative waveform distortion is significant in both the cylindrical- and spherical-spreading regimes. The laboratory-scale studies also reveal a pronounced sensitivity to humidity conditions, relative to the full-scale counterpart.
The plume produced by a cluster of two large area-ratio thrust-optimized paraboliccontour nozzles is visualized over a range of nozzle pressure ratios by way of retrore-flective shadowgraphy. Both nozzles exhibit free-shock separated flow, restricted-shockseparated flow and an end-effects regime prior to flowing full. Transient (startup) op-erations of the nozzles are studied with the primary focus being the pulsations thatform during the end-effects regime. This occurs at a pressure ratio of 37 for thesenozzles and is associated with elevated sound levels in the immediate vicinity of thenozzles and vehicle. The shadowgraphy images reveal the formation of turbulent largescale structures, on the order of the nozzle diameter, during the end-effects regime.These large scale structures are driven by the intermittent opening of the last trappedannular separation bubble to the ambient and grow rapidly within the first two nozzlediameters.
The spatial evolution of acoustic waveforms produced by a laboratory-scale Mach 3 jet are investigated using both 1∕4 in. and 1∕8 in. pressure field microphones located along rays emanating from the postpotential core where the peak sound emission is found to occur. The measurements are acquired in a fully anechoic chamber, where ground or other large surface reflections are minimal. Various statistical metrics are examined along the peak emission path, where they are shown to undergo rapid changes within 2m from the source region. An experimentally validated wave-packet model is then used to confirm the location where the pressure amplitude along the peak emission path transitions from cylindrical to spherical decay. Various source amplitudes, provided by the wave-packet model, are then used to estimate shock formation distance and Gol’dberg numbers for diverging waves. The findings suggest that cumulative nonlinear distortion is likely to occur at laboratory scale near the jet flow, where the waveform amplitude decays cylindrically, but less likely to occur farther from the jet flow, where the waveform amplitude decays spherically. Direct inspection of the raw time series reveals how steepened waveforms are generated by rogue like waves that form from the constructive interference of waves from neighboring sources as opposed to classical cumulative nonlinear distortion.
A theoretical combustion model is developed to simulate the influence of ideal gas effects on various aeroacoustic parameters over a range of equivalence ratios. The motivation is to narrow the gap between laboratory and full-scale jet noise testing. The combustion model is used to model propane combustion in air and kerosene combustion in air. Gas properties from the combustion model are compared to real lab data acquired at the National Center for Physical Acoustics at the University of Mississippi as well as outputs from NASA’s Chemical Equilibrium Analysis code. Different jet properties are then studied over a range of equivalence ratios and pressure ratios for propane combustion in air, kerosene combustion in air and heated air. The findings reveal negligible differences between the three constituents where the density and sound speed ratios are concerned. Albeit, the area ratio required for perfectly expanded flow is shown to be more sensitive to gas properties, relative to changes in the temperature ratio.
The effect of stagger startup on the vibro-acoustic loads that form during the end-effects regime of clustered rockets is studied using both full-scale (hot-gas) and laboratory scale (cold gas) data with vehicle geometry. Both configurations comprise three nozzles with thrust optimized parabolic contours that undergo free shock separated flow and restricted shock separated flow as well as an end-effects regime prior to flowing full. Acoustic pressure waveforms recorded at the base of the nozzle cluster are analyzed using various statistical metrics as well as time-frequency analysis. The findings reveal a significant reduction in end-effects regime loads when engine ignition is staggered. However, regardless of stagger, both the skewness and kurtosis of the acoustic pressure time derivative elevate to the same levels during the end-effects regime thereby demonstrating the intermittence and impulsiveness of the acoustic waveforms that form during engine startup.
The design, fabrication and calibration of a new thrust stand for conducting thrust andaeroacoustic measurements concurrently in a fully anechoic chamber is discussed. The new thrust stand employs the scale-force measurement technique and is designed to accommodate multi-stream nozzles (core and bypass flow streams). Each stream has a dedicated helium air mixture system thereby permitting a multitude of test conditions. The methodology for designing the thrust stand is described and uses a notch type flexure which demonstrates high repeatability over extended thrust ranges. Calibration is performed with elevated pressure inside the plenum to characterize the effect of increased pressure on the flexure performance. A further qualification of the thrust measurement accuracy is conducted using a small arsenal of nozzles comprising method of characteristics contours. Surveys of the far-field pressure are then conducted during various operating points along the startup curve of a Mach 1.71 rectangular supersonic nozzle.
Shock wave / boundary layer interaction is studied in a large area ratio axisymmetric nozzle comprising a design exit Mach number of 5.58. Shock motion unsteadiness is captured by way of the dynamic wall pressure and is evaluated during overexpanded operations up to a nozzle pressure ratio of 65. Stationary SWBLI is first considered at a nozzle pressure ratio of 28.7 such that the internal flow structure is in a restricted-shock separated state; the mean position of the annular separation shock resides at a fixed position. Conditional averages of the wall pressure fluctuations show how the motion of the incipient separation shock is out of phase with pressure fluctuations measured in the separated region downstream of the shock; pressure decreases when the shock moves downstream and vice versa. This is indicative of a long intermittent region, in terms of the boundary layer thickness, as the observed phenomena can be explained by translating the static wall pressure profile along with the shock motion. Non-stationary SWBLI is then considered by increasing the nozzle pressure ratio over time (transient startup). Under these conditions, the shock pattern varies in strength and structure as it sweeps through the nozzle. A time-frequency analyses of the fluctuating wall pressure during the non-stationary operations, and at the same location that the stationary unsteadiness is analysed, reveals a similar spectral footprint. However, for relatively slower start-ups, the amplitude of the unsteadiness is reduced by a factor of about seven. The findings demonstrate how the rate at which the nozzle pressure ratio increases can have a significant influence on the amplitude of the unsteady shock foot motion.
High area ratio rockets generate strong vibro-acoustic loads primarily during transient operations, like start-up and shut-down of the engine. These loads can adversely affect the launch vehicle and its payload. Thus, an accurate prediction of the loads produced during engine start-up is pertinent to the safety and reliability of the launch vehicle. The present work focuses on developing a robust framework for predicting these loads using laboratory scale rocket nozzles tested in the fully anechoic chamber at The University of Texas at Austin. This encompasses corrections for the observer position relative to the prominent source region, as well as scaling factors to correct for geometric factors. The test campaign encompasses single, two, three and four nozzle clusters, as well as differences in nozzle geometry and operating conditions (nozzle pressure ratio).
The plume produced by a cluster of two high area-ratio thrust optimized parabolic contour nozzles is visualized by way of retroreflective shadowgraphy. Both steady and transient operations of the nozzles (start-up and shut-down) were conducted in the anechoic chamber and high speed flow facility at The University of Texas at Austin. Both nozzles exhibit free shock separated flow, restricted shock separated flow and an end-effects regime prior to flowing full. Radon transforms of the shadowgraphy images are used to identify the locations in the flow where sound waves are being generated. During these off design operations of the nozzles, most sound waves are generated by turbulence interactions with the shock cells located in the supersonic annular plume. During the end-effects regime, this supersonic annular plume is shown to flap violently, thus providing a first principals understanding of the sources of most intense loads during engine ignition.
C. E. Tinney, Canchero, A., Rojo, R., Mack, G., Murray, N. E., and Ruf, J. H., “The Sound-field Produced by Clustered Rockets During Startup,” Whither Turbulence and Dig Data for the 21st Century. Symposium held at the Institute dEtudes Scientifques de Cargese, Corsica, France, April 20-24, (Springer Hardbound Volume, DOI: 10.1007/978-3-319-41217-7), 2015.Abstract
The vibroacoustic loads produced by a cluster of two large area-ratio thrust optimized parabolic contour nozzles are studied over a range of pressure ratios encompassing free-shock separated flow, restricted shock separated flow and the end-effects-regime. The rocket plume is visualized using a retroreflective shadowgraphy system while an experimentally validated RANS model provides insight into the internal flow and shock structure patterns. Pressure loads that form on the base of the vehicle (behind the nozzles) are then measured using a eighth-inch microphone, as most of these loads are caused by high intensity sound waves produced by the rocket nozzle flow. The objective of the study is to provide a direct link between the sources of most intense vibro-acoustic loads that form during the ignition of high area ratio rocket nozzle clusters.
The stability and turbulence characteristics of a vortex filament emanating from a single-bladed rotor in hover are investigated using proper orthogonal decomposition. The rotor is operated at a tip chord Reynolds number and tip Mach number of 218,000 and 0.23, respectively, and with a blade loading of CT /σ = 0.066. In-plane components of the velocity field (normal to the axis of the vortex filament) are captured by way of 2D particle image velocimetry with corrections for vortex wander being performed using the Γ1 method. The first POD mode alone is found to encompass nearly 75% of the energy for all vortex ages studied and is determined using a grid of sufficient resolution as to avoid numerical integration errors in the decomposition. The findings reveal an equal balance between the axisymmetric and helical modes during vortex roll-up which immediately transitions to helical mode dominance at all other vortex ages. This helical mode is one of the modes of the elliptic instability. The spatial eigenfunctions of the first few Fourierazimuthal modes associated with the most energetic POD mode is shown to be sensitive to the choice of the wander correction technique used. Higher Fourier-azimuthal modes are observed in the outer portions of the vortex and appeared not to be affected by the choice of the wander correction technique used.
The spatial evolution of acoustic waveforms produced by a Mach 3 jet are investigated using both 1/4 inch and 1/8 inch pressure field microphones located along rays emanating from the post potential core where the peak sound emission is found to occur. The measurements are acquired in a fully anechoic chamber where ground, or other large surface reflections are minimal. The calculation of the OASPL along an arc located at 95 jet diameters using 120 planar grid measurements are shown to collapse remarkably well when the arc array is centered on the post potential core region. Various statistical metrics, including the quadrature spectral density, number of zero crossings, the skewness of the pressure time derivative and the integral of the negative part of the quadrature spectral density, are exercised along the peak emission path. These metrics are shown to undergo rapid changes within 2 meters from the source regions of this laboratory scale jet. The sensitivity of these findings to both transducer size and humidity effects are discussed. A visual extrapolation of these nonlinear metrics toward the jet shear layer suggests that these waveforms are initially skewed at the source. An experimentally validated wave packet model is used to confirm the location where the pressure decay law transition from cylindrical to spherical. It is then used to estimate the source intensity which is required to predict the effective Gol’dberg number.
Low-dimensional characteristics of a helical vortex filament from a reduced-scale rotor are investigated using proper orthogonal decomposition (POD). Measurements are captured by way of particle image velocimetry. Experiments are performed on a 1.0 m diameter, single-bladed rotor in hover. The rotor is operated at 1500 RPM, which corresponds to a blade tip chord Reynolds number of 218,000 and a tip Mach number of 0.23. The blade is set to a collective pitch angle of 7.3◦, which resulted in a blade loading (CT /s) of 0.066. Classical and snapshot techniques of POD are applied to a helical vortex filament, both of which revealed similar characteristics of the dominant modes. Two different techniques (G1 and geometric center methods) of wander correction are applied to test the sensitivity of the low-dimensional characteristics using POD. Using the G1 method, POD revealed that an elliptic instability dominated the energy spectrum of the velocity fluctuations within the tip vortex. However, at early vortex ages an axisymmetric mode, which is found to perform vortex roll-up, is found to be equally dominant. Further, the spatial structures of the most energetic modes derived from POD are found to be sensitive to the choice of the centering technique used.
The plume and acoustic field produced by a cluster of two and four rocket nozzles is visualized by way of retroreflective shadowgraphy. Both steady state and transient operations of the nozzles (start-up and shut-down) were conducted in the fully-anechoic chamber and open jet facility of The University of Texas at Austin. The laboratory scale rocket nozzles comprise thrust-optimized parabolic (TOP) contours, which during start-up, experience free shock separated flow, restricted shock separated flow, and an “end-effects regime” prior to flowing full. Shadowgraphy images are first compared with several RANS simulations during steady operations. A proper orthogonal decomposition (POD) of various regions in the shadowgraphy images is then performed to elucidate the prominent features residing in the supersonic annular flow region, the acoustic near field and the interaction zone that resides between the nozzle plumes. Synchronized surveys of the acoustic loads produced in close vicinity to the rocket clusters are compared to the low-order shadowgraphy images in order to identify the various mechanisms within the near-field that are responsible for generating sound.