Volume 21, Issue 2, March 1949
Index of content:
21(1949); http://dx.doi.org/10.1121/1.1906483View Description Hide Description
From the equation for propagation of acoustic plane waves through a solid medium, expressions are derived for the vibrational amplitude, radiated power, and electrical admittance of a plane‐wave crystal transducer. The transducer consists of crystals, backing, and diaphragm, radiating into any medium. The crystals may be an assemblage of flat bars vibrating lengthwise, or there may be a single flat crystal plate, or mosaic of plates, in thickness vibration. The equations take account of attenuation in the diaphragm, but not of attenuation in crystals and backing, nor of the effects of cement layers and lateral stresses between crystals.
The general equations, which are valid at all frequencies and for all thicknesses of backing, crystal assemblage, and diaphragm, are specialized to the case in which the backing is air. Special attention is given to those frequencies at which the crystals, or diaphragm, or both, are in resonance. Some typical theoretical curves are shown, and a graphical method is described for the rapid plotting of curves.
21(1949); http://dx.doi.org/10.1121/1.1906484View Description Hide Description
In this paper the causes of noise generated by the conventional supersonic biasing and erasing fields used in magnetic recording systems are discussed. The data to be presented emphasize the manner in which this noise varies with the wave form and frequency of an applied supersonic field. The importance of these two factors is shown to be a function of the characteristics of the erase and recording heads and of the inherent frequency response and modulation noise characteristics of the medium. By proper control of the wave form and frequency of the supersonic signal, noise may be reduced to such an extent that the effective noise level of a given recording medium with bias and erase signals applied will be equal to or less than the noise level of a virgin sample of the medium. Under these conditions, values of saturated signal to noise exceeding 80 db may be obtained.
21(1949); http://dx.doi.org/10.1121/1.1906485View Description Hide Description
In listener tests of sound reproducing equipment there are many variable factors that contribute to the validity of the result. Among these factors are the program material, associated acoustical and electronic equipment, the size and nature of the audience, the method used in reporting audience response, and the statistical analysis of the audience response. This paper discusses these variables and proposes a definitive method for recording and presenting the data in a simple and thorough manner.
21(1949); http://dx.doi.org/10.1121/1.1906486View Description Hide Description
In this paper a method is described of measuring the acoustic impedance of a sample of material forming one boundary of a shallow cylindrical cavity, by determining the sound pressure produced when a known volume current is injected into the cavity from a high impedance source. The volume current is effectively determined by observing the pressure when the cavity is rigidly terminated. The use of a ring source suppresses the first radial mode of the cavity and extends the frequency limit set by uniformity of pressure by at least an octave.
Secondary effects due to the finite impedance of the source and microphone and heat losses at the walls have been evaluated. The method appears to be simple, rapid, and precise.
21(1949); http://dx.doi.org/10.1121/1.1906487View Description Hide Description
A high speed level recorder of the “potentiometer type” which is useful both for laboratory and field measurements is described. The recording system consists of a moving coil with translational motion, the speed of which can be varied with a maximum value of about 1000 db/sec. Recording is made by a steel stylus on waxed paper of the same kind as that in the usual Neumann recorder.
21(1949); http://dx.doi.org/10.1121/1.1906488View Description Hide Description
A precise experimental study of the reactive component of the complex impedance of small circular orifices has been made for a number of orifices varying in diameter from 2 cm down to 0.357 cm, with diameter/thickness ratios from 4 to 40, over the frequency range from 200 to 1000 c.p.s. The measurement of the impedance is performed with a precision impedance tube.
A theoretical correction for the influence of the tube walls is applied by considering the orifice to act as a plane piston and taking into account the influence of all possible higher order modes of the tube in the neighborhood of the orifice. Comparison between calculated and measured values of reactance shows that, as far as this quantity is concerned, the assumption that the orifice acts like a plane piston appears valid for radii equal to or greater than 1 cm for “thin” orifices within the range of variables specified above. For orifices of radii less than 1 cm, however, a modification of the radius is necessary to make the classical theoretical equation for orifice reactance fit the measured data. The observations and analyses reported here are restricted to wave‐lengths much greater than the hole diameter and to the low velocity region where reactance is independent of particle velocity.
21(1949); http://dx.doi.org/10.1121/1.1906489View Description Hide Description
21(1949); http://dx.doi.org/10.1121/1.1906472View Description Hide Description
The normal or common type of ultrasoniclight diffraction occurs when traveling sound waves act like a phase grating. The spacing and intensities of the respective orders of spectra are given by the Raman‐Nath theory. Experimentally and theoretically, this type of diffraction is obtained when the product of (width of sound beam) × (frequency)2 × (sound amplitude) is sufficiently low. When the light beam is maintained parallel to the sound wave fronts, the right and left spectra of the same order are of equal intensity.
An abnormal or uncommon type of ultrasoniclight diffraction occurs when the sound waves diffract the light in a manner satisfying the Bragg reflection law, well known in x‐rays. The spacing of the spectra is essentially the same as before but right and left spectra do not appear simultaneously. Strong intensity in a given portion of the spectra is obtained only when the glancing angles of incidence φ and of reflection θ are equal, as in mirror reflection, and when λ, Λ, and θ satisfy the Bragg law , (λ and Λ are the light and sound wave‐lengths, respectively, and n is the order of spectra). This type of diffraction is obtained only when the single product (width of sound beam) × (frequency)2 is sufficiently large.
Experiments and theory indicate the above criteria for these two types of light diffraction.Theory also indicates the possibility of other types of abnormaldiffraction, such as mirror reflection, diffraction satisfying ruled gratingtheory,, where φ does not necessarily equal θ, and others.
21(1949); http://dx.doi.org/10.1121/1.1906473View Description Hide Description
A procedure was devised for determining the air blast pressures of the A‐bomb at Bikini. In such a disturbance the air particles move with large and finite amplitudes; hence, the propagated waves do not obey the ordinary laws of acoustics. By measuring the ratio of the air shock wavevelocity to normal acoustic velocity, an indication may be obtained as to the peak pressure of the explosion.Blast waves in water, on the other hand, follow substantially the familiar acoustic laws and are propagated at a well‐known velocity.Measurements were made of transit times at identical positions for the air and the waterblast waves. These yielded immediately the desired air blast velocity in terms of the known velocity of sound in water. This procedure eliminated the need for simultaneous measurement of the continually shifting distances between buoys which carried observational equipments. However, careful positioning of the buoys was necessary in order to avoid their destruction and yet record the maximum possible air blast. These positions were calculated on the basis of an equivalent explosion from 20,000 tons of TNT. Since the measurements had to be made automatically and without human observers, special timing and recording systems were provided. All requirements for gathering the data were successfully met in the assembly of apparatus described herein.
21(1949); http://dx.doi.org/10.1121/1.1906474View Description Hide Description
A theory is outlined for the propagation constant in media containing numerous small spherical particles. Using expressions derived by Lamb for the zero and first‐order scattering coefficients of a particle free to move in a sound field, an expression for the complex propagation constant is derived whose real part yields a velocity which reduces to the homogeneous case for extremely small particles, and whose imaginary part yields an absorption coefficient identical with that derivable from the viscous‐drag theory outlined in a previous paper.
Using both an interferometer and a pulse‐reflection method, measurements of sound velocity and absorption at megacycle frequencies have been made on mercury‐in‐water and bromo‐form‐in‐water emulsions of non‐uniform particle size, up to a volume concentration of about 50 percent of emulsified liquid. These materials, though showing considerable deviation from a homogeneous behavior, are found to have a velocity and absorption in good agreement with the theory up to concentration of about 25 percent by volume.
21(1949); http://dx.doi.org/10.1121/1.1906475View Description Hide Description
In this paper a theoretical and experimental treatment is given of guided sound wave transmission along circular cylinders of ideal liquid with various non‐dissipative boundary conditions. The field patterns, phase velocities, and cut‐off frequencies are calculated for the natural modes of propagation in the following cases: (1) liquid cylinder with rigid walls; (2) liquid cylinder with pressure‐release walls; (3) liquid cylinder embedded in infinite liquid; (4) liquid cylinder with liquid walls; (5) liquid cylinder with thin solid walls. The problem of determining the field produced by an arbitrary excitation in terms of the natural modes is also discussed.
Experimental results are presented for cases (2) and (5) above. It is shown that the measured values of phase velocity are in agreement with those predicted by the theory.
21(1949); http://dx.doi.org/10.1121/1.1906476View Description Hide Description
The general form of the intensity function for the electrical responses of the cochlea is now well known. As we raise the intensity of a sound applied to the ear, the cochlear potentials rise at first in simple proportion to the intensity, and then at a high level, as the stimulus is elevated further, they cease to follow this simple relation and give a bending curve. From the first departure from linearity onward the responses rise more and more slowly in relation to the stimulus until at last they reach a maximum value, a value beyond which they actually diminish as the intensity is raised.
So much of the form of the function is established on the basis of numerous experiments (see reference 2), yet there still are many details not yet worked out with exactness. The present study seeks in a measure to supply this lack. We have been concerned especially with the initial departure from linearity and the maximum value of the response for various frequencies over the tonal range. This information, as we shall show, has a pertinent application to the problems of the locus and spread of response over the basilar membrane.
21(1949); http://dx.doi.org/10.1121/1.1906477View Description Hide Description
The most important results of the determinations previously published on the adaptation of the ear to sound stimuli are briefly recapitulated. Adaptation is defined as the elevation of the auditory threshold by a previous sound stimulus. It may be determined by means of a short tone impulse (testing impulse) which follows the sound stimulus (stimulating impulse) causing the adaptation. The testing impulse is adjusted to the threshold value. By altering the interval between the stimulating and testing impulses, the curve of return adaptation, i.e., the return to normal sensitivity, may be ascertained. When plotted on a logarithmic db scale, this curve is approximately a straight line.
New experiments have shown that the entire process of adaptation requires less than 0.4 sec. The return adaptation is complete within a few tenths of a second. After stimulation by a tone of 80 db above threshold intensity, return adaptation is complete after 250 μsec.
The increase in adaptation is approximately proportional to the intensity of the stimulating tone when both magnitudes are expressed in db. With a sound impulse of 80 db above the auditory threshold, the adaptation reaches a value of 40 to 50 db.
Binaural determination of adaptation, by leading the stimulating impulse to one ear and the testing impulse to the other, shows that adaptation is a monaural and therefore a peripheral process.
A pure tone induces an elevation of the auditory threshold not only for its own frequency but also for neighboring ones. The adaptation spreads chiefly to higher frequencies than that of the stimulating tone at which it reaches a maximum. The maximum is more pronounced at high frequencies of the stimulating tone than at low ones.
Comparative experiments between adaptation and masking demonstrate a far‐reaching similarity between the two processes. These show a similar behavior with respect to intensity and frequency. A quantitative comparison leads to the conclusion that masking depends chiefly upon the adaptation of the ear to sound stimuli.
- LETTERS TO THE EDITOR
21(1949); http://dx.doi.org/10.1121/1.1906481View Description Hide Description
21(1949); http://dx.doi.org/10.1121/1.1906482View Description Hide Description