Fig 9 - uploaded by Jayanta Panda
Content may be subject to copyright.
Power spectral density of (a) air density, (b) axial velocity, (c) radial velocity, (d) density*(axial velocity) 2 fluctuations; (e) density-axial velocity cross-spectrum, and (f) phase of cross-spectrum in Mach 1.4 jet at centerline & x/D = 10
Source publication
To locate noise sources in high-speed jets, the sound pressure fluctuations p/, measured at far field locations, were correlated with each of density p, axial velocity u, radial velocity v, puu and pvv fluctuations measured from various points in fully expanded, unheated plumes of Mach number 0.95, 1.4 and 1.8. The velocity and density fluctuations...
Citations
... Experimental sound source localizations [41] shown for instance that most of the downstream noise is generated just downstream of the potential core. Significant direct correlations have also been found between the downstream pressure and the flow fluctuations on the jet axis at the end of the potential core, experimentally by Schaffar [42] and Panda et al. [43,44] and from the present simulation results [45]. ...
The sound fields radiated by Mach number 0.6 and 0.9, circular jets with Reynolds numbers varying from 1.7×103 to 4×105 are investigated using Large Eddy Simulations. As the Reynolds number decreases, the properties of the sound radiation do
not change significantly in the downstream direction, whereas they are modified in the sideline direction. At low Reynolds
numbers, for large angles downstream from the jet axis, the acoustic levels are indeed remarkably lower and a large high-frequency
part of the sound spectra vanishes. For all Reynolds numbers, the downstream and the sideline sound spectra both appear to
scale in frequency with the Strouhal number. However their peak amplitudes vary following two different velocity exponents
according to the radiation direction. The present observations suggest the presence of two sound sources: a Reynolds number-dependent
source, predominant for large radiation angles, connected to the randomly-developing turbulence, and a deterministic source,
radiating downstream, related to a mechanism intrinsic to the jet geometry, which is still to be comprehensively described.
This view agrees well with the experimental results displaying two distinguishable components in turbulent mixing noise [1,
2].
... This dependency of the correlation coefficients on jet velocity was confirmed later from the Rayleigh scattering based measurements. 13,14 The ultimate success of the prior causality work appeared when Schaffar calculated the volume integral in equation (2) from a plume survey. For correlations measured from a fixed microphone, positioned at shallow 160° and 150° angles, time derivative and integration following equation (2) reproduced the microphone auto-spectrum ( fig. 2). ...
... Yet at certain other polar angles correlations were so high that a unit volume at the end of the potential core is found to produce more noise than the entire jet. 13 Therefore, the primary focus of the recent work was to find the nature of the correlation coefficients from different plume conditions and nozzle configurations without definite association with any theory. ...
... summarizes the operating conditions and nozzle configurations.Figure 3(a) shows a photograph of an unheated jet facility where one convergent nozzle, two C-D nozzles of design Mach number Mj = 1.4 & 1.8, one 16-lobed rectangular nozzle and one 4-tabbed circular nozzle were tested.13,14,16 ...
Significant advancement has been made in the last few years to identify noise sources in high speed jets via direct correlation measurements. In this technique turbulent fluctuations in the flow are correlated with far field acoustics signatures. In the 1970 s there was a surge of work using mostly intrusive probes, and a few using Laser Doppler Velocimetry, to measure turbulent fluctuations. The later experiments established "shear noise" as the primary source for the shallow angle noise. Various interpretations and criticisms from this time are described in the review. Recent progress in the molecular Rayleigh scattering based technique has provided a completely non-intrusive means of measuring density and velocity fluctuations. This has brought a renewed interest on correlation measurements. We have performed five different sets of experiments in single stream jets of different Mach number, temperature ratio and nozzle configurations. The present paper tries to summarize the correlation data from these works.
Nowadays, commercial aircrafts, invariably, use high-bypass-ratio dual-stream jets for propulsion. As yet, there is still an urgent need for an accurate physics-based noise prediction theory for jets of this configuration. Thus, an investigation is made to determine whether the Tam and Auriault theory (Tarn, C. K. W.,. and Auriault. L., "Jet Mixing Noise from Fine Scale Turbulence". AIAA Journal, Vol. 37, No. 2. 1999, pp. 145-153), originally developed for predicting file fine-scale turbulence noise of single-stream jets, is capable of predicting accurately the fine-scale turbulence noise of dual-stream jets from separate flow nozzles operating at various bypass ratios. The configuration of a separate How nozzle is fairly complex. Hence. the jet How and turbulence in the nozzle region and in the region immediately downstream are also fairly complex. However, these are also the most important noise source regions of the jet. To enable an accurate computation of the mean flow and turbulence level in these regions, a computational aeroacoustics marching algorithm for calculating file parabolized Reynolds averaged Navier-Stokes equations supplemented by the k-epsilon turbulence model is provided. It is shown that the computed mean How profiles are in good agreement with experiment. Extensive comparisions between computed noise spectra and measurements are reported. They include dual-stream jets front separate flow nozzles with and without an external plug. Jets operating at different combinations of primary and secondary,jet Much number and temperature ratio are considered. The bypass ratios range from 1.5 to 8.0. Effects of forward flight are also included in the comparisons. Good agreements are found not only in spectral levels but also in spectrum shapes and directivities over many sets of data.
Recent advancements in a molecular Rayleigh scattering based diagnostic technique allowed for simultaneous measurement of velocity and density fluctuations with high sampling rates. The technique was used to investigate unheated high subsonic and supersonic fully expanded free jets in the Mach number range from 0.8 to 1.8. The difference between the Favre-averaged and the Reynolds-averaged axial velocity and axial component of the turbulent kinetic energy is found to be small. On average, estimates based on Morkovin's strong Reynolds analogy are found to underpredict turbulent density fluctuations.
Experimental measurements indicate that the noise radiated-from a jet depends not just on the jet-exit velocity alone, but is significantly affected by the jet temperature. Now, there is evidence to support the proposition that jet mixing noise consists of two principal components. These are the noise from the large turbulence structures of the jet flow and the fine-scale turbulence. The prediction of fine-scale turbulence noise from hot jets is considered. Earlier Tam and Auriault (Tam, C. K. W., and Auriault, L., "Jet Mixing Noise from Fine-Scale Turbulence," AIAA Journal, Vol. 37, No. 2; 1999, pp. 145-153) developed a semi-empirical theory capable of predicting the fine-scale turbulence noise from cold to moderate temperature jets. In this work, their semi-empirical theory is extended to high-temperature jets, up to a temperature ratio above that of present day commercial engines. The density gradient present in hot jets promotes the growth of Kelvin-Helmholtz instability in the jet mixing layer. This causes a higher level of turbulent mixing and stronger turbulence fluctuations. In addition, recent experiments reveal that the two-point space-time correlation function of turbulent mixing for hot jets is substantially different from that for cold jets. The eddy decay time is shorter,, and the eddy size is slightly reduced. These changes have an appreciable impact on the noise radiated. In the present extended fine-scale turbulence theory, both effects are taken into account. Extensive comparisons between computed noise spectra and measurements for hot jets over the Mach-number range of 0:5-2.0 are reported here. Good agreements are found over inlet angle from 50 to 110 deg. This is the directivity for which fine-scale turbulence noise is dominant.
Measurements of space-time correlations of velocity, acquired in jets from acoustic Mach number 0.5 to 1.5, and static temperature ratios up to 2.7 are presented and analyzed. Previous reports of these experiments concentrated on the experimental technique and on validating the data. In the present paper the dataset is analyzed to address the question of how space-time correlations of velocity are different in cold and hot jets. The analysis shows that turbulent kinetic energy intensities, lengthscales, and timescales are impacted by the addition of heat, but by relatively small amounts. This contradicts the models and assumptions of recent aero acoustic theory trying to predict the noise of hot jets. Once the change in jet potential core length has been factored out, most one- and two-point statistics collapse for all hot and cold jets.
Ground testing of turbojet engines in test cells necessarily involves very high acoustic amplitudes, often enough and severe enough that testing is interrupted and facility hardware and test articles are damaged. The acoustic response of test cells containing energetic jets is poorly understood and generally unpredictable. Nevertheless, there is a clear need to be able to predict deleterious acoustic events in advance of facility entry. A predictive capability would permit evaluating possible fixes in advance of the entry to preclude interruption of testing and damage to hardware, both of which are costly and disruptive of weapons systems program schedules. To establish the needed predictive capability, the Arnold Engineering Development Center (AEDC) is implementing a computational aeroacoustics (CAA) capability. This report by C. K. Tam is one of several steps toward that goal. Here, Tam consolidates what is presently known about the aeroacoustics of jets and flowing ducts. The material presented includes analytical and semi-empirical models of various acoustic situations as well as test data. Also included is a proposal to ameliorate a particularly damaging acoustic event referred to as super resonance. A future report will present CAA technology appropriate for numerical solution of the flow equations as applied to jet cells.
Noise sources are investigated in subsonic jets at Mach numbers M = 0.6 and M = 0.9, with Reynolds numbers ReD = 1700 and ReD ≥ 10 5 , using data provided by Large Eddy Simulations (LES). The cross-correlations between signals of the radiated sound pressure and turbulence signals along jet axis and shear layer are in particular calculated. The normalized correlations are found to be significant, around 0.10, between the pressure radiated in the downstream direction and the centerline flow quantities. In the cases involving the sideline pressure or flow quantities along the shear layer, the correlations are much smaller. The maximum correlations are observed on the jet axis just at the end of the potential core, and fall at large emission angles. Furthermore the correlations appear to be lower as the Mach number is reduced, and to be enhanced as the Reynolds number is decreased. These correlations can be expected to be mostly due to the noise source radiating downstream, which may thus be located on the jet centerline at the end of the potential core. This flow region is moreover characterized by a dominant frequency over a large axial distance and by a high level of intermittency.
Recent advancement in the molecular Rayleigh scattering based technique allowed for simultaneous measurement of velocity and density fluctuations with high sampling rates. The technique was used to investigate unheated high subsonic and supersonic fully expanded free jets in the Mach number range of 0.8 to 1.8. The difference between the Favre averaged and Reynolds averaged axial velocity and axial component of the turbulent kinetic energy is found to be small. Estimates based on the Morkovin's "Strong Reynolds Analogy" were found to provide lower values of turbulent density fluctuations than the measured data.
