Volume 28, Issue 6, November 1956
Index of content:
28(1956); http://dx.doi.org/10.1121/1.1908541View Description Hide Description
From the point of view of their use in supersonic wind tunnels in which it is desired to obtain a high Mach number without great expenditure of power, the substitution of the air by vapors having a low velocity of sound is of interest. The freons (halogen‐substituted methanes) suggest themselves for this. It is shown that these vapors have ultrasonic relaxation times in the neighborhood of 10−7 sec coupled with high absorption coefficients in the ultrasonic range, a fact which may limit their usefulness in wind tunnels. The relation between relaxation timesmeasured by other methods and by the present one are discussed.
28(1956); http://dx.doi.org/10.1121/1.1908543View Description Hide Description
In recent experiments the compositions and methods of preparation of various ferrites were varied with the aim of achieving optimum mechanical and piezomagnetic performance. Essential improvements were obtained by small cobalt substitutions in the chemical composition and by suitable modifications of the mechanical and thermal treatments.
In CW operation, appropriate ferrites can excite radiation intensities up to at least 10 w/cm2, with an electroacoustic efficiency of more than 75% up to 50 kc/sec. So they are highly promising for underwater transmission and processing, in spite of their brittleness (ultimate tensile strength up to about 6 kg/mm2). Wide‐band hydrophones with linear frequency response and high pressure‐sensitivity can also be built from such ferrites having magnetomechanical coupling coefficients at remanence up to almost 0.25, like Permendur.
The total variation of the mechanical resonant and antiresonant frequencies of ferrite filter elements in the temperature range from +20 to +50°C, being 0.1–0.25% for existing commercial ferrites, has been lowered to 0.03% and less. Since the mechanical quality factors are usually much better than 2000, such ferrites can be profitably applied to the construction of electrical and electromechanical band filters.
Effect of Vibration Amplitude, Frequency, and Composition of the Abrasive Slurry on the Rate of Ultrasonic Machining in Ketos Tool Steel28(1956); http://dx.doi.org/10.1121/1.1908545View Description Hide Description
Results of experiments conducted to determine the dependence of the ultrasonicmachining rate V in Ketos tool steel on the peak‐to‐peak amplitude 2ξ on the frequency of vibration f on the abrasive particle diameter d and on the ratio of the mass of abrasive to the mass of water used in compounding the slurry are given. It is shown that within the range under investigation the quotient V/2ξfd is a constant for a critical ratio of the mass of abrasive to that of water. It is also shown that events occurring within the volume of a cube having sides equal to d are of fundamental importance in the ultrasonicmachining process. A phenomenological equation is derived from which the most probable machining rate can be calculated for the range under investigation.
Backscattering of Sound from the Sea Surface: Its Measurement, Causes, and Application to the Prediction of Reverberation Levels28(1956); http://dx.doi.org/10.1121/1.1908547View Description Hide Description
In many applications of underwater sound, the “reverberation,” or backscattered return, from the sea surface is the limiting background in which the desired echo appears. Little has been known about sea‐surface backscattering in spite of its importance to the prediction of reverberation levels in practical applications.
Using two transducers—one for transmitting, one for receiving—mounted so as to be rotatable in a vertical plane and suspended from a buoy, measurements have been made of the backscattering of sound pulses from the surface of the sea as a function of grazing angle between 5 and 90 degrees, wind speed between 3 and 18 knots, and pulse length. When converted to a scattering coefficient per unit area, the results provide some hints as to the processes by which sound is scattered by the ocean surface, and give the basic data for computing the reverberation to be expected with transducers of arbitrary beam width and orientation.
28(1956); http://dx.doi.org/10.1121/1.1908549View Description Hide Description
The natural oscillations of membranes can be studied by reflection of light from an acoustically excited soap film. In order to obtain a uniform soap film, it proves advantageous to rotate the film so that its thickness is uniform. The oscillations may be obtained even when the films have become extremely thin after long rotation. In addition, one finds strange streaming phenomena with certain natural oscillations.
28(1956); http://dx.doi.org/10.1121/1.1908551View Description Hide Description
When a turbulent boundary layer is produced by air flow past a solid surface, the turbulence in the boundary layer can generate a sound field in the free stream and will also induce fluctuating loads on the solid surface. If the surface is flexible, this motion will generate an additional sound field on both sides of the surface.
In an initial investigation of the latter form of sound generation, suitable equipment has been developed to measure the fluctuating wall pressure in the turbulent boundary layer. The equipment includes a specially designed low noise and turbulence level wind tunnel and a small barium titanate transducer and preamplifier combination for frequencies up to 50 kc. The transducer and preamplifier may be useful for other applications.
Using this equipment, some of the properties of the wall pressurefluctuations in a turbulent boundary layer have been measured. It was found that the spectrum of the wall pressurefluctuations extends to 50 kc and that the root‐mean‐square wall pressure was a constant portion (0.0035) of the free stream dynamic pressure for and . A few typical spectra are given for different values of the Reynolds number and Math number.
28(1956); http://dx.doi.org/10.1121/1.1908553View Description Hide Description
Formulations of one‐dimensional sound propagation in a homogeneous, progressively moving medium are applied to calculations of recurrent filter characteristics of side branch tube attached to the side wall of an air duct. The complex propagation constants and characteristic impedance are functions of the flowMach number. Therefore, the insertion loss is dependent upon flow but relative changes with flow depend upon the terminating impedance levels as well as geometry. Neglecting dissipation the linear system is found nonreciprocal with respect to phase shift.
28(1956); http://dx.doi.org/10.1121/1.1908555View Description Hide Description
Studies of the insertion loss of a side branch resonant filter in moving air reveal that nonlinear excitation of the resonator and turbulence may influence the measured characteristic. These effects occur at relatively low Mach numbers for particular flow geometry patterns. The insertion loss is found to be destroyed for sufficiently high flow velocities. An interpretation of insertion loss values as averaged over a finite band of frequencies is also given some attention.
28(1956); http://dx.doi.org/10.1121/1.1908557View Description Hide Description
The origin of the magnetic driving forces which are responsible for magnetic noise and a method of calculating the mechanical response of the motor structure are reviewed in this paper. A general method is developed to calculate the sound pressure levels of magnetic noise at a specified point near the motor. The paper gives numerical results of calculations, and the corresponding measuredsound pressure levels of the magnetic noise for a particular experimental 4‐pole, 30‐hp, polyphase induction motor under five different operating conditions; calculations agree with measurements within 3 db.
28(1956); http://dx.doi.org/10.1121/1.1908559View Description Hide Description
Basic criteria are given for the incorporation of many reactive components in sound absorbers. Forms of absorbers, determined by these criteria, are derived. Design parameters are treated in relation to techniques for determining their optimum values. Many reactive components are shown to have the effect of enhancing uniformly high absorption over extended frequency ranges.
28(1956); http://dx.doi.org/10.1121/1.1908561View Description Hide Description
The important factors involved in the development of a phonetic typewriter are as follows. The particular form in which the words are typed. The means for analyzing the sounds of speech. The identification of the analyzed sounds. The encoding, coding, and decoding of the sounds for the operation of the actuating mechanism. The design of the mechanism for actuating the typewriter. A study has been made of these problems. As a result of this study a simplified model of a phonetic typewriter has been developed incorporating all of these aspects. This model serves to illustrate the principles involved and provides means for further study towards a complete system.
28(1956); http://dx.doi.org/10.1121/1.1908563View Description Hide Description
A successive‐binary‐selection system for automatic classification of spoken English into several groups of phonemes is described. The first step separates voiced from unvoiced by measuring the filtered output in the range of the pitch frequency. The next step separates turbulent (noise‐like) from nonturbulent by measuring the filtered output in the range of the first formant.
Nonturbulent sounds are classified into 6 groups of 2 or 3 phonemes each in successive selections based on spectrum measurements. For each of the two groups of phonemes being separated in a given step, a frequency band is chosen where phonemes of that group have large amounts of energy compared to those of the other group. The polarity of the weighted difference of the corresponding band‐pass‐filter outputs determines the classification.
Unvoiced turbulent sounds are separated into stops and fricatives by means of the amplitude information contained the onsets of these sounds. Two of the fricatives, |s| (see) and |∫| (she), are separated by means of their zero‐crossing‐density levels.
28(1956); http://dx.doi.org/10.1121/1.1908565View Description Hide Description
A device for processing speech for transmission over a narrow‐band communication channel is described. This device analyzes the speech wave form and extracts certain parameters continuously as a function of time. These parameters have been used to control a speechsynthesizer utilizing local excitation functions to simulate vocal and turbulent sources. Equipment to carry out the analysis‐synthesis procedure as it exists in its present state is discussed as well as consideration for its inclusion in a complete band‐width‐compression system.
28(1956); http://dx.doi.org/10.1121/1.1908567View Description Hide Description
The development of a speech‐band‐width compression system employing formant‐coding principles is described. The compression system codes the input speech in terms of seven electrical signals, representing the frequencies of the first three formants, the amplitudes of voicing and of friction, the fundamental vocal frequency, and the frequency of the spectral maximum of the fricative excitation. These signals occupy a total band width of the order of 60 cps and require signal‐to‐noise ratios of approximately 30 db for their transmission.
An evaluation of the intelligibility of speech transmitted by the compression system is described. Natural monosyllabic utterances are transmitted through the system and presented to listeners for identification. Vowel and consonant articulation scores are computed and sound confusion matrices are constructed and discussed. The results indicate that correct identification occurred for approximately 80% of the vowels and 25% of the consonants.
28(1956); http://dx.doi.org/10.1121/1.1908569View Description Hide Description
The first step in the construction of a speech writer must be concerned with the conversion of the flow of speech into discrete operations. Such operations have been brought within the realm of possibility by recent developments in the analysis and synthesis of speech in terms of frequency functions instead of pressure functions. The present paper deals primarily with the phonetic and graphic aspects of the response and discrimination required of the machine and the dictating person. Certain simplifications of design can be effected by consideration of the orthography of the language to be speech‐written.
28(1956); http://dx.doi.org/10.1121/1.1908571View Description Hide Description
The present American normal hearing reference zero is considered by many investigators to be too high. The data from two recent studies on normal hearing provide a possible explanation. One of these studies, which was a survey type study, confirmed the present reference zero. The second study, a laboratory type study, confirmed the results of other laboratory studies that had led to the opinion that the present zero is too high.
In our opinion, the difference between the results of these two studies and various other studies of the auditory threshold are due to the inherent difference in experimental technique in laboratory and survey type studies. In establishing normal hearing reference zero, it is no longer a question of resolving differences in various studies but a question of whether survey or laboratory data will be used.
28(1956); http://dx.doi.org/10.1121/1.1908573View Description Hide Description
Frequency range preferences of 210 college students for reproduced music and speech were determined by an A‐B‐A preference test. Two groups of subjects then listened to music reproduced over a restricted frequency range and a relatively unrestricted frequency range, respectively, for six and one‐half weeks. The results of a post‐frequency range preference test indicate that: (1) learning plays an important role in determining preferences for sound reproducing systems; (2) continued contact with a particular system produces shifts in preference for this system; (3) the average college student prefers music and speech reproduced over a restricted frequency range rather than an unrestricted frequency range; and (4) the frequency range preferences of college students are in part a function of the type of music to which they are listening.
28(1956); http://dx.doi.org/10.1121/1.1908575View Description Hide Description
The techniques which comprise the mobility analogy have been introduced into the classical impedance analogy. The mobility and impedance analogies are set forth in form for practical use. The completely dual character of these two analogies is rendered evident by placing the elements of the mobility analogy on a series of left‐hand pages and the dual elements of the impedance analogy on the facing right‐hand pages. In each analogy a set of mechanical and acoustical symbols is provided so that mechanical and acoustical schematic diagrams can be drawn without the necessity of drawing either of the analogous electrical schematic diagrams. These mechanical and acoustical symbols resemble somewhat their analogous electrical symbols so as to indicate to anyone familiar with circuit theory the algebraic operations which are to be performed in the analysis.
Whereas in an electrical schematic straight line represents an ideal wire, in a mobility schematic a straight line represents an ideal incompressible massless rod; in an impedance schematic a line represents a hydraulic tube filled with ideal massless incompressible liquid which can sustain either tension or compression.
Correct mobility schematics are obtained by inspection through considering each mass as having one terminal fixed in a frame of reference called the earth which has zero mobility and infinite mass and is analogous to electrical ground; correct impedance schematics are obtained by inspection through considering each spring as having one terminal fixed in a force of reference (P 0, etc.) called the sky which has zero impedance and infinite compliance (in both cases the analogous electrical element is a conducting sphere in free space, not a condenser). Although it is recommended that analogous electrical circuits not be used in routine analysis, they can be copied directly from a mobility schematic by considering mechanical earth analogous to electrical ground, or copied directly from an impedance schematic by considering mechanical sky analogous to electrical ground.
28(1956); http://dx.doi.org/10.1121/1.1908577View Description Hide Description
The equivalent circuit of a finite circular disk resonant at any frequency corresponding to the first three axially symmetric flexural modes is obtained from a function of two dynamic variables; (1) the kinetic energy in the disk, (2) velocity of the mechanical motion. This function, called the equivalent mass, is computed over a domain of arguments which is of practical interest. The results indicate fairly good agreement between theory and practice.
28(1956); http://dx.doi.org/10.1121/1.1908579View Description Hide Description
A mathematical analysis has been done according to Love's theory. The results are: (a) The coefficients of the dilatation and the rotation are equal in amount regardless of Poisson's ratio only for the modes whose circumferential order is one. (b) The first mode whose order is two has the lowest natural frequency. This mode is practically equivoluminal, while the second mode is dilatational. (c) Locations of nodes are considered. Circular nodal lines can exist only for the radial and the tangential modes. (d) Frequencies and displacements of the most important modes are calculated.
Applying the above results to bariumtitanate vibrators, we find that only the radial modes can be excited with full electrodes and the compound modes with split or partial electrodes, while the tangential modes can never be excited. Such an excitation of compound modes provides a convenient means for measuring Poisson's ratio of the material. Equivalent series inductances of the radial modes are calculated.
Vibrations of thick circular plates are also considered. It is shown that vibrations of certain compound modes which are practically equivoluminal can be approximated by the rotation only. Comparisons between these solutions and the above results are made to reveal that the accuracy of approximation is satisfactory.