Engineering topics
Publications and source records attributed to Joshi, M. C..
A semi-empirical formula for the prediction of the broadband shock associated noise from rectangular supersonic jets in the fly-over plane has been developed. The predicted noise spectra compared very favorable with the measurement of Ponton, Manning and Seiner. Extension to side-line directions will be carried out.
The topics discussed include the following: atmospheric turbulence; BOOMFILE Database description; BOOMFILE flight conditions; XB-70 Database descriptions; analysis progression; extended database; prediction method; overpressure variability dependence on flight conditions; loudness variability on flight conditions; sonic boom variability in repeat flights; and statistical distributions.
Sonic boom signatures from two databases, the BOOMFILE and the XB-70, were analyzed in terms of C-weighted sound exposure level (CSEL), A-weighted sound exposure level (ASEL), and Stevens Mark VII perceived level (PLdB), as well as the more traditional peak positive overpressure and rise time. The variability of these parameters due to propagation through atmosphere was analyzed for different aircraft Mach number and altitude groups. The low Mach number/low altitude group had significantly greater variation in rise time, overpressure, and loudness level than the high Mach number/high altitude group. The loudness of measured booms were found to have a variation of up to 25 dB relative to the loudness of boom predicted for a non-turbulent atmosphere. This is due primarily to the steeper ray paths of the high Mach number/high altitude group and the corresponding shorter distances traveled by these rays through the lower atmosphere resulting in reduced refraction effects. The general trend of decreased overpressure and loudness level with increasing lateral distance was also seen. Sonic boom signatures from early morning flights had less variation in rise time and overpressure than afternoon flights because of reduced turbulence. Measures of asymmetry (difference between compression and expansion portion of the signature) showed that the variability in Delta loudness level was greater than the variability in Delta overpressure due to the large influence of turbulence on rise time. Lastly, analysis of data within 50 percent of lateral cutoff showed that the mean value for overpressure and loudness level was independent of time of day but that the frequency with which it occurred was greater in the morning. This is a clear indicator of increased turbulence in the afternoon.
Sonic boom signatures from two databases, the BOOMFILE and the XB-70, were analyzed in terms of C-weighted sound exposure level (CSEL), A-weighted sound exposure level (ASEL), and Stevens Mark VII perceived level (PLdB), as well as the more traditional peak positive overpressure and rise time. The variability of these parameters due to propagation through atmosphere was analyzed for different aircraft Mach number and altitude groups. The low Mach number/low altitude group had significantly greater variation in rise time, overpressure, and loudness level than the high Mach number/high altitude group. The loudness of measured booms were found to have a variation of up to 25 dB relative to the loudness of boom predicted for a non-turbulent atmosphere. This is due primarily to the steeper ray paths of the high Mach number/high altitude group and the corresponding shorter distances traveled by these rays through the lower atmosphere resulting in reduced refraction effects. The general trend of decreased overpressure and loudness level with increasing lateral distance was also seen. Sonic boom signatures from early morning flights had less variation in rise time and overpressure than afternoon flights because of reduced turbulence. Measures of asymmetry (difference between compression and expansion portion of the signature) showed that the variability in Delta loudness level was greater than the variability in Delta overpressure due to the large influence of turbulence on rise time. Lastly, analysis of data within 50 percent of lateral cutoff showed that the mean value for overpressure and loudness level was independent of time of day but that the frequency with which it occurred was greater in the morning. This is a clear indicator of increased turbulence in the afternoon.
An experimental investigation carried out to determine aerodynamic and acoustic characteristics of a low area ratio rectangular jet ejector is reported. A supersonic primary jet issuing from a rectangular convergent-divergent nozzle of aspect ratio 4, into a rectangular duct of area ratio 3, was used. Improved performance was found when the ejector screech tone is most intense and appears to match the most unstable Strouhal number of the free rectangular jet. When the primary jet was operating at over and ideally expanded conditions, significant noise reduction was obtained with the ejector as compared to a corresponding free jet. Application of particle image velocimetry to high speed ejector flows was demonstrated through the measurement of instantaneous two dimensional velocity fields.
Acoustic modal distributions were measured in a fan test model having an annular exhaust duct for comparison with theoretically predicted acoustic suppression values. This report contains the amplitude and phase data of the acoustic signals sensed by the transducers of the two mode probes employed in the measurement. Each mode probe consisted of an array of 12 transducers sensing the acoustic field at three axial positions and four radial positions.
Suppressions due to acoustic treatment in the annular exhaust duct of a model fan were theoretically predicted and compared with measured suppressions. The predictions are based on the modal analysis of sound propagation in a straight annular flow duct with segmented treatment. Modal distributions of the fan noise source (fan-stator interaction only) were measured using in-duct modal probes. The flow profiles were also measured in the vicinity of the modal probes. The acoustic impedance of the single degree of freedom treatment was measured in the presence of grazing flow. The measured values of mode distribution of the fan noise source, the flow velocity profile and the acoustic impedance of the treatment in the duct were used as input to the prediction program. The predicted suppressions, under the assumption of uniform flow in the duct, compared well with the suppressions measured in the duct for all test conditions. The interaction modes generated by the rotor-stator interaction spanned a cut-off ratio range from nearly 1 to 7.
A description of the structure of the Annular Flow Duct Program (AFDP) for the calculation of acoustic suppression due to treatment in a finite length annular duct carrying sheared flow is presented. Although most appropriate for engine exhaust ducts, this program can be used to study sound propagation in any duct that maintains annular geometry over a considerable length of the duct. The program is based on the modal analysis of sound propagation in ducts with axial segments of different wall impedances. For specified duct geometry, wall impedance, flow and acoustic conditions in the duct (including mode amplitude distribution of the source) and duct termination reflection characteristics, the program calculates the suppression due to the treatment in the duct. The presence of forward and backward traveling modes in the duct due to the reflection and redistribution of modes at segment interfaces and duct end terminations are taken into account in the calculations. The effects of thin wall boundary layers (with a linear or mean flow velocity profile) on the acoustic propagation are also included in the program. A functional description of the major subroutines is included and a sample run is provided with an explanation of the output.
Acoustic measurements made with in-duct modal probes show how individual acoustic modes, as generated by fan/stator interaction, interact with a well-characterized liner. This measured attenuation provides a critical test for predicted attenuation as developed from the theoretical analysis of sound propagation in annular ducts with treatment and sheared flow. The primary investigations took place in the exhaust duct of a model fan mounted in an anechoic chamber. The results presented include induct attenuation, sound power attenuation as measured in the far field, and in-situ acoustic impedance measurements of the treated wall sections. Good theory/experiment agreement was found; the differences of treatment suppression between theory and measured values differed at most by 2.7 dB.
Suppressions due to acoustic treatment in the annular exhaust duct of a model fan have been theoretically predicted and compared with measurements. The predictions are based on the modal analysis of sound propagation in a straight annular flow duct with segmented treatment. Measured values of the mode distribution of the fan noise source (fan-stator interaction source only) and the acoustic impedance of the treatment in the duct were used as input to the prediction program. The predicted suppressions (under the assumption of uniform flow in the duct) compared well with the measured in-duct suppression for all test conditions. The measurements were made in an anechoic chamber and the interaction modes generated in the fan spanned a cut-off ratio range from nearly one to seven.
An experimental investigation has been conducted to determine the physical process of the noise production from the trailing edge of an isolated two-dimensional airfoil embedded in a low turbulence uniform mean flow. The Reynolds number of the airfoil based on chord was greater than 1,000,000 and that the boundary layer was fully turbulent at the trailing edge. Smoke injection technique and spark shadowgraphy were used to study the structural features of the boundary layer. Two point joint statistical analyses were made on the surface pressure, the radiated sound and the relationship between the two quantities. Operating conditions included two free stream velocities and two angles of attack for both a naturally developed and an artificially tripped boundary layer. Flow visualizations revealed the existence of large scale coherent motions in the outer region of the boundary layer. The observed coherent motions had scales of the order of boundary layer thickness and a convection velocity near the free stream velocity. The production process of the airfoil trailing edge noise was determined to be the convection of the large scale coherent eddies over the trailing edge of the airfoil. The noise field so generated was found to be rather coherent and dipole-like.
The effect of mean flow modification on the noise production of upper surface blown flaps has been studied experimentally. Mean velocity profile at the nozzle exit was modified from the usual 'top-hat' shape to 'Gamma' and 'L'-shaped profiles. The 'L'-modification caused noise reduction around and above the peak frequency of the 'top-hat' spectrum when compared on an equal thrust per exit area basis. Modification to 'Gamma'-shaped profile resulted in a shift of the spectrum to lower frequencies and a lower overall noise reduction. These modifications alter the development of the large scale disturbances in the upper shear layer and trailing edge wake of the wall jet geometry.