IJAV site directory

Published Articles

The Volume 1, No 1, March 1996

«« Back to the List of the Articles

Recent Advances in Interpreting Hearing Sensitivity

James Lighthill


The mammalian cochlea derives its fine sensitivity to very low levels of sound from an amplification process associated with a fast cellular motor located in outer hair cells. Recent discoveries about the nature of this motor, and about its regulation, are briefly sketched alongside new evidence for the motors effectiveness.

Display the first page:

Wavelet Analysis of Vibration Signals Part 1: Wavelet Properties

David E. Newland


Wavelet analysis allows the energy of a recorded signal to be decomposed into its contributions from different frequency bands and different locations. In contrast, Fourier analysis generates data which is the result of averaging over the duration of the record being studied; information about location can be obtained only by dividing a record into sections and Fourier analysing each section separately (which is what the short-time Fourier transform, STFT, does). For transient data analysis, the wavelet transform and the STFT produce similar results, but wavelet transforms have computational and other advantages which may be important in particular applications. Part 1 of this paper reviews wavelet theory and describes wavelet properties and computational algorithms for vibration signal analysis. Part 2 compares wavelet analysis with time-frequency analysis using the STFT and discusses particular advantages of the harmonic wavelet transform in this respect.

Display the first page:

Development of a Multiple Source Adaptive Active Noise and Vibration Control System

M. O. Tokhi, K. Mamour and M. A. Hossain


This paper presents an investigation into the development of an adaptive active control strategy for noise cancellation and vibration suppression using a multi-source configuration. A self-tuning control algorithm is developed on the basis of optimum cancellation of broadband noise/vibration at a set of observation points. The algorithm is implemented on a digital processor and its performance assessed in the cancellation of noise in a free-field medium and the suppression of vibration in a flexible beam structure.

Display the first page:

Vibrational Impact of an Elastic Link

Dan B. Marghitu and Malcolm J. Crocker


An experimental analysis was conducted to study the rebound velocities of a freely dropped bar on a large external surface. The low velocity collision occured in the presence of a frictional impact. A 3D motion analysis system was used to capture the kinematic data. Tests were performed for central and oblique collisions. A spatial theoretical model is presented for the impact with friction of a flexible body. This model consists of a system of nonlinear differential equations which considers the multiple collisions as well as frictional effects at the contacting end, and allows one to predict the rigid and elastic body motion after the impact. The model uses a dry coefficient of friction, a nonlinear contact force, and the sound radiation from the impact. Analytical and experimental results were compared to establish the accuracy of the model.

Display the first page:

Computation of Noise from Homogeneous Turbulence and from a Free Jet

Seungbae Lee, William C. Meecham


Acoustic radiation from turbulent fluids (aeroacoustics) has been a major research area since the pioneering work of Lighthill. We present here a new application of Large-Eddy Simulation (LES) techniques to the calculation of the acoustic far field produced by turbulent fluids. A direct numerical simulation is limited to mean flow Reynolds numbers of about 1500; see the heroic efforts of Mitchell et al. (1995) (clearly useful for many purposes, dealing with a Karman-vortex-street-like structure). This is a limitation set by both computer memory and computation time; there is little realistic prospect of relieving this handicap in the near future. The acoustic radiation problem consists of two steps; first one simulates the turbulent flow field using LES, and in the second step one generates the acoustic far field using Lighthill's procedure (the flow field as source). We first consider statistically homogeneous and isotropic turbulence found using LES employing three subgrid-scale models; the Smagorinsky model, a scale-similarity model, and a deductive model (based on a Taylor series expansion with no adjustable constants), and their combinations. A model which is a combination of the deductive and Smagorinsky model shows the most satisfactory features using as criteria: correlation, decay behavior, and energy spectra. The quadrupole directivity pattern from a few eddies is clearly seen. Of course the use of a time average would remove many of these details; we show first one member of the statistical ensemble. As a second example we treat the turbulent subsonic jet issuing from a square nozzle. A similar calculation for jet noise is performed. The combined subgrid model proposed here avoids a demanding integration process for filtering. This model is used in the simulation of the turbulent jet. In these aerosound calculations of the turbulent jet, the major source region is seen to be the end region of the potential core. The turbulence intensity (as found in experiments) is about 10%, independent of the initiating level. A rapidly changing quadrupole with a peak near 45 degrees from the axis of the jet is obtained. The sound pressures are, within the approximation of order of magnitude estimates, of the size of the Lighthill values, with a multiplicative constant set by the results of many experiments. The jet Reynolds number is 48,000, fully developed; there is no reason why the procedure cannot be extended well beyond such values. It is noted that the Reynolds number regime is far from that of vortex pairing and the like. Preliminary versions of this work have been presented in several symposia, the first being Lee and Meecham (1992).

Display the first page:

«« Back to the List of the Articles