Volume 33, Issue 6, June 1961
Index of content:
- PROGRAM OF THE SIXTY‐FIRST MEETING OF THE ACOUSTICAL SOCIETY OF AMERICA
- Session A. Physiological Acoustics
- Contributed Papers
33(1961); http://dx.doi.org/10.1121/1.1936773View Description Hide Description
For many years there have been various attempts to record cochlear potentials from the round window in a human being. Some earlier efforts have met with moderate success. During the past two years, approximately 30 cases have been attempted at the Johns Hopkins Hospital. A multichannel tape recorder has been used to simultaneously stimulate the human ear and record from the round window. The patients all had their tympanic membranes reflected and a small electrode was placed on the round window niche. In most cases, good cochlear potentials were recorded and measured. In addition, clicks were used to stimulate the human ear, and recordings have been made from the round window. The action potentials of the 8th nerve known as N1 and N2 were then recorded. Two groups of patients are now being studied. The first group consists of patients with otosclerosis. A recording is made before mobilization or a replacement of the stapes is done. A second recording is taken after the oval window is free, to show if there has been any objective gain in the cochlear potential. The second group of patients are children with congenital neurosensory hearing loss. Excellent cochlear potentials and 8th nerve action potentials have been recorded in this group. The recording of cochlear potential in man is being used as a diagnostic and exploratory tool in evaluation and understanding of hearing disorders. It is felt that this also gives a valuable tool in understanding some of the basic properties of hearing in the human being.
33(1961); http://dx.doi.org/10.1121/1.1936774View Description Hide Description
There has been some debate as to whether or not the action potential of the 8th nerve as viewed at the round window, and N1 and N2, originate within the cochlear or in the nervous system. Several experiments have been undertaken in cat in which the 8th nerve was sectioned completely, leaving the blood supply intact. By recording from the severed distal stump and the round window, the N1 and N2 were found to be preserved up to 24 hr after sectioning. By using careful time studies, several properties of the 8th nerve were also noted. The latency of the N1 as recorded at the round window with the first potential recorded on the 8th nerve was carefully measured. A discrepancy in conduction time was found. The conduction time was based upon the fiber diameter which has been measured, and also direct studies. The knowledge gained from these experiments will be of great use in the interpreting of the cochlear response to click as recorded in the human round window. The locus of both the N1 and N2 has been shown to be peripheral to the cochlear nucleus. Time studies indicate that the origin of the N1 and N2 is at some point peripheral to the ganglion cells in the modiolus.
33(1961); http://dx.doi.org/10.1121/1.1936775View Description Hide Description
Masking has been used to subtract groups of nerve impulses from the whole‐nerve action potential of the guinea pig. The form of that part of the whole‐nerve action potential which is eliminated by a narrow band of noise is approximately diphasic. This is the expected form of individual nerve impulses recorded from one electrode in the cochlea and the other on the neck. The stimulus variables investigated were intensity and rise time of the acoustic transient, and intensity, bandwidth, and center frequency of the masking noise. The whole‐nerve action potential to low‐pass transients is composed of impulses that arise from a large extent of the organ of Corti, with a distribution in time that parallels the space‐time pattern of the cochlear microphonic. (This research was supported by Bt‐366‐C2 and NIH‐B1726.)
33(1961); http://dx.doi.org/10.1121/1.1936776View Description Hide Description
Cats with electrodes permanently implanted on the round window of the cochlea and previously trained to respond to threshold and supra‐threshold tones in shuttle‐box conditioning were exposed to traumatic sound (135 db at 1 kc for 2 hr). Serial concomitant measurements of both behavioral and electrophysiologic (CM and AP) responses were obtained before and after exposure at several frequencies. Temporary threshold shifts of 65 to 95 db were produced in the middle frequency range. Associated cochlear microphonic changes were considerably less (5 to 25 db) and in most cases returned to pre‐exposure values long before behavioral indexes at varying degrees of permanent threshold shift.
33(1961); http://dx.doi.org/10.1121/1.1936777View Description Hide Description
Appreciable experimental data exist for the amplitude of basilar membrane displacement as a function of sound frequency. Data on the phase of displacement vs frequency are relatively meager. At low frequencies the latter do not relate well to the phase predicted by recent mathematical models for membrane displacement [Bell System Tech. J. 39, 1163–1191 (1960)]. To examine the difference, we programmed an IBM‐7090 computer to calculate: (a) minimum phase functions corresponding to the experimental amplitude vs frequency responses; (b) inverse Fourier transforms of the experimental amplitude and phase data; and (c) inverse transforms of the experimental amplitude and calculated minimum phase data. The results were: (a) At low frequencies the calculated phase differs from the experimental phase by about π/2 rad or more, with the membrane displacement leading that of the stapes. Constant delay (i.e., linear phase) does not account for the difference; (b) inverse transforms of the experimental data yield nonrealizable impulse responses, i.e., responses which are nonzero for negative times; and (c) inverse transforms of the experimental amplitude and calculated phase yield proper impulse responses. At low frequencies, therefore, the experimental amplitude and phase data do not seem compatible. Because of the great difficulties attending the experimental measurement, it is the phase data which are thought to need confirmation.
33(1961); http://dx.doi.org/10.1121/1.1936778View Description Hide Description
This paper describes an attempt to supplement the experimental approach to improving fenestration by circuit analysis. The fenestrated ear is considered an idealized acoustical system and its equivalent circuit is derived. The circuit consists of an input section (outer ear, middle ear, fenestrae), and seven sections representing the inner ear. These are found to be resonant at frequencies between 100 cps and 10 kc. The problem is simplified by assuming the partition between outer and middle ear to be rigid. The other significant parameters are then combined into a variable series capacitance. The resulting circuit is analyzed numerically (digital computer). Response curves are obtained for volume displacement across the basilar membrane as a function of frequency and distance from the input end, with input capacitance as parameter. The computed results and experimental findings (by Békésy) are similar. The startling result of the investigation is that a lowering of the combined window stiffness by a factor of 30 results, in the model, in a 30 db gain over most of the speech spectrum. This is in sharp contrast to controversial experiments in the 1930's by Hughson and Crowe who tried to improve fenestration by increasing the stiffness of the round window.
- Session B. Underwater Acoustics—Propagation Phenomena
- Invited Paper
33(1961); http://dx.doi.org/10.1121/1.1936779View Description Hide Description
The problems associated with acoustic reverberation in the ocean continue to be of prime importance in the use of active sonar. Even though a great amount of effort has been devoted to these problems, they are still not well understood and experimental evidence remains the prime source for an evaluation of the state‐of‐knowledge, in this field. This paper attempts to summarize the state‐of‐knowledge through the presentation of recognized fundamentals and results of pertinent experiments.
- Contributed Papers
Correlations and Frequency Spectra of Fluctuations in Sound Signals Transmitted over a Fixed Path in an Estuary33(1961); http://dx.doi.org/10.1121/1.1936780View Description Hide Description
Short pulses of 100‐kc sound have been transmitted from a fixed source to an array of fixed receivers at a range of 100 ft in the Severn River. A statistical analysis has been made of 7000 pulses transmitted directly across a sharp, changing velocity gradient, which produced large fluctuations, and of 14 000 pulses reflected from the surface when fluctuations in direct transmission were negligible. Autocorrelations and cross correlations with delays up to 300 sec have been computed for signals received simultaneously on hydrophones with various horizontal and vertical separations. For surface‐reflected signals, little correlation was found between successive pulses or between pulses received on separated hydrophones. For direct transmission, the decrease of autocorrelation with time and the decrease of cross correlation with both time and distance have been compared with various theoretical correlation functions. Phase differences between fluctuations on separated hydrophones have been found which are not simply related to the current. Frequency spectra of the fluctuations have been computed. In surface‐reflected signals there is fluctuation at frequencies up to 6 cps, while in direct transmission there is negligible fluctuation at frequencies above 0.2 cps. (This work was supported by the Office of Naval Research.)
33(1961); http://dx.doi.org/10.1121/1.1936781View Description Hide Description
Spectra of sound pressure amplitude versus frequency in the range 1.1–2.4 kc have been obtained at regular intervals in shallow (10 ft) water during two periods of extended observation. The acoustical data have been presented in the form of contour diagrams of sound pressure amplitude versus frequency and time as coordinates and compared with water conditions. The form of the contour diagrams permits differentiation between field structure variation and variation in attenuation, although differentiation between the associated processes has not been possible. Temporal variation in the vertical sound velocity profile produced only small changes in the gross features of the field structure—as inferred from the contour diagram—for up to the first three modes of propagation. For frequencies above 1.6 kc (i.e., when three or four modes were stimulated) the variations in field structure were prominent. Attempts have been made, with little success, to explain periods when the received signal was strongly attenuated in terms of vertical sound velocity structure.
33(1961); http://dx.doi.org/10.1121/1.1936782View Description Hide Description
The concept of coherence in the field of acoustics apparently means the same thing to all people, but only on an intuitive level, and a formal definition has not been generally agreed upon. The author suggests that the definition used in the theory of multidimensional stochastic processes be adopted by the acousticians. In the present paper, this definition is presented and, when Gaussianity is assumed, an interpretation is given in terms of the relationship between the phases of the signals. The proposed definition leads to the following interpretations: when the processes are highly coherent, the variability of the phase difference between the signals is small; when the processes are incoherent, the phase difference has a high degree of variability. This concept of coherence seems to agree with the intuitive notions currently used in acoustics. (This work was supported by the Office of Naval Research.)
33(1961); http://dx.doi.org/10.1121/1.1936783View Description Hide Description
Coherencemeasurements have been made relating sinusoidal signals simultaneously received at separated hydrophones. The definition of coherence advocated by Shapiro in the previous paper is used. The quantity measured is closely related to phase. Of interest is long term average phase as well as a measure of the fluctuations about this mean. Such measurements are useful as an aid in understanding the performance of complex hydrophone systems. The measurement is accomplished by cross multiplying and averaging stored five‐second segments of information from each of two channels. This computation is carried out for a range of delays comprising several periods of the sinusoidal signal. A facsimile type recorder provides a pictorial display of how phase wanders as a function of time. Long term averages are obtained numerically from samples of the continuous display. Recently collected quantitative data are presented for several hydrophone spacings in deep water. The results show violent phase fluctuations and dramatically illustrate the statistical nature of acoustic transmission in the ocean. (This work was supported by the Bureau of Ships.)
33(1961); http://dx.doi.org/10.1121/1.1936784View Description Hide Description
The complex shear modulus has been measured for samples of Hycar rubber under these conditions: (a) 0.25–2.0 cps at 22°C, with a torsion pendulum; (b) 40 cps–5 kc, between 0° and 41°C, with an electromagnetic transducer; (c) at 10 Mc and 22°C, by McSkimin's method. The “method of reduced variables” allows calculation of the shear modulus, from (b), from 5 cps to 1.2 Mc at 22°C. Measurements (a) and (c) are consistent. From the density and the real part of the modulus, the speed of transverse waves in the rubber has been calculated as a function of acoustic frequency. Results are compared with speeds inferred in a previous paper [R. K. Eby, A. O. Williams, R. P. Ryan, and P. Tamarkin, J. Acoust. Soc. Am. 32, 88 (1960)] that treated acoustic propagation in a model made up of a shallow water layer over a thick slab of this same rubber. The present speeds fall about 25% below those previously inferred, are undoubtedly more reliable, and are still reasonably successful in explaining attenuations within the water layer caused by transverse waves in the bottom, as hypothesized earlier. (A. O. W., Jr., assisted by the Office of Naval Research.)
33(1961); http://dx.doi.org/10.1121/1.1936785View Description Hide Description
Ewing and Press have discussed acoustic propagation in a liquid layer, over a solid with both compressional wave speed c 2 and transverse wave speed ct exceeding c 1 for the liquid. When, instead, c 2>c 1>ct , sound is only partially trapped in the liquid layer, with steady “leakage” into transverse waves. Such a case has been measured in a laboratory model (water over rubber) [R. K. Eby, A. O. Williams, R. P. Ryan, and P. Tamarkin, J. Acoust. Soc. Am. 32, 88 (1960)]; perhaps a compact sand bottom acts similarly. The problem is solved by satisfying acoustic boundary conditions at the interface, for a transverse and two compressional waves; when ct vanishes the equations reduce to the liquid‐liquid (Pekeris) case. The eigenvalue for each normal mode is now complex, betokening “leakage.” The real part is altered (relative to liquid‐liquid) as if the density of the bottom were decreased. The imaginary part gives an attenuation coefficient, for sound traveling into the liquid layer, roughly proportional to (ct /c 1)3, with a complicated dependence on other parameters. The calculation can be extended to include absorption of transverse waves in the bottom—probably the usual situation in a “flabby” material. (Work supported in part by the Office of Naval Research.)
- Session C. Waves and Vibration
33(1961); http://dx.doi.org/10.1121/1.1936786View Description Hide Description
The theory of a new method of determining the harmonic structure of a finite amplitude wave by light diffraction has recently been given [B. D. Cook, J. Acoust. Soc. Am. 32, 336 (1960)]. This method allows the computation of the amplitudes of the harmonic components of the instantaneous pressure from the measurements of the light intensities of all of the diffraction orders. Preliminary measurements and computations indicate that this method will be a valuable tool in the study of finite amplitude distortion. (This work was supported by the Office of Ordnance Research, U. S. Army.)
Diffraction of Wide and Narrow Light Beams by Distorted Finite Amplitude Progressive Ultrasonic Waves33(1961); http://dx.doi.org/10.1121/1.1936787View Description Hide Description
Results of an investigation of the diffraction of light passing through an ultrasonic wave of finite amplitude are given. A method is developed and used to determine the second harmonic component of the distorted ultrasonic wave at various distances from the transducer, while maintaining a constant local fundamental component. A wide light beam (giving discrete diffraction orders) is used for these determinations. It is then shown that the observed continuous light distribution (using a narrow light beam) is in good agreement with the theoretically predicted distribution. The effects of third and higher harmonics are neglected. These determinations of second harmonic are compared with values obtained from diffractioneffects observed after passing the second harmonic component through an acoustic filter plate. Measurements were made at 3.0 Mc in water with approximately of an atmosphere local ultrasonicpressure amplitude. (This work was supported by the Office of Ordnance Research, U. S. Army.)
33(1961); http://dx.doi.org/10.1121/1.1936788View Description Hide Description
This paper deals with the scattering of a plane wave incident on a prolate spheroidal shell along its axis of revolution. The procedure can be easily generalized to an arbitrary direction of incidence. Contrary to the case of spherical and cylindrical shells, the spheroidal wave functions in terms of which the incident and scatteredwaves are expressed, do not coincide with the normal modes of the shell. As a result, when the coefficients of the expression of the shell motion in the spheroidal functions are used as generalized coordinates, all the Lagrangianequations will be coupled. Expressions suitable for computer calculations are given, and an example is worked out in which the frequency of the incident wave is close to one of the natural frequencies of the shell. Under these conditions the scattering action of the shell differs strikingly from that of a rigid spheroid. (This paper is based on work sponsored by the Office of Naval Research.)
33(1961); http://dx.doi.org/10.1121/1.1936789View Description Hide Description
Explicit expressions for the driving point impedance of cylindrical shells subjected to concentrated fluctuating radial forces have been sorely lacking in the extensive literature on shell vibrations. This need was partly filled by a recent paper by P. A. Franken [J. Acoust. Soc. Am. 32, 473 (1960)], but his impedance expressions have been found to be valid only over a restricted frequency range. The purpose of the present paper is to supply convenient, more generally applicable expressions for cases of practical interest. The loads considered are sinusoidally varying, concentrated and locally distributed forces and moments. Impedance expressions are given for uniform shells, both infinite and finite. Impedance expressions are also given for the ring‐stiffened shell of finite length. (This paper is based on work sponsored by the Office of Naval Research.)
33(1961); http://dx.doi.org/10.1121/1.1936790View Description Hide Description
The driving point impedance of thin plates as implied by the work of Zener and given explicitly by D. A. Thomas [J. Acoust. Soc. Am 30, 220 (1958)] was determined by measuring with a pulse method the complex force and velocity at the driving point. The plate was terminated in wet sand to simulate infinite plate conditions. At low frequencies, the measurements were limited by the physical dimensions of the plate. At high frequencies, compliance in the mounting of the accelerometer and driving coil caused inaccuracies in the measurements.Measurements were made on one aluminum plate and one steel plate. The data obtained agreed well with theory in the frequency range 300 cps to 3 kc.
33(1961); http://dx.doi.org/10.1121/1.1936791View Description Hide Description
In any field testing involving horizontal propagation of sound, it is desirable to know how much is absorbed by the ground. Since the sound absorptionproperties of the ground are affected by many factors, it is usually necessary to make an estimate of this effect. This paper describes an experiment to measureabsorption coefficient of the soil by means of an impedance tube. Controlled laboratory tests are described which measured the effect of moisture content and particle size on the absorption coefficient.Effects of grass on absorption coefficient are also noted. (This work was sponsored by Project Michigan under a U. S. Army contract.)
- Session D. Architectural Acoustics—Transmission Loss
- Invited Papers
33(1961); http://dx.doi.org/10.1121/1.1936792View Description Hide Description
Walls are commonly delineated by a cross‐sectional drawing. Although a “one‐dimensional” description may be sufficient for building the wall, it does not generally convey enough information from which to calculate accurately the transmission loss. For single walls, the size and the detail at the edges are important [M. Heckl and K. Seifert, “Untersuchungen über den Einfluss der Eigenresonanzen der Messräume auf die Ergebnisse von Schalldāmmessungen,” Acustica 8 (1958); and I. Dyer, WADD Tech. Rept. (to be published)]. These two factors will be discussed in detail in the following paper. For walls separating reverberant rooms, the relative volumes, the sound absorption, and the symmetry of the rooms are important [M. Heckl, “Die Schalldämmung von homogenen Einfachwänden endlicher Fläche,” Acustica 10 (1960)]. For any wall, the angle of incidence of the sound wave is important [A. Eisenber, “Über die Schalldämmung von Glasscheiben und Fenstern,” pp. 34, 35 in Bauforschung im Wohnungsbau, Schwenk and Company, Frankfurt‐am‐Main, Germany]. The transmission loss of a wall is seen to be a function of many parameters, only one of which is the composition of the wall cross‐section. The role of these other parameters and their importance for the testing and use of walls will be discussed.