Index of content:
Volume 25, Issue 4, July 1953
25(1953); http://dx.doi.org/10.1121/1.1907150View Description Hide Description
It is shown theoretically that plane longitudinal sound waves traveling in an electrically conducting plate will generate complex electric currents provided there is a magnetic induction perpendicular to the particle displacement. The currents react with the induction to produce a complex mechanical force which opposes the particle motion. The impedance associated with this force effectively modifies the impedance of the medium by adding a resistance and a stiffness reactance. The induced resistanceattenuates the waves while the induced stiffness increases Young's modulus and therefore increases the sound velocity.
Of all the elements, it appears that this effect is a maximum in lithium, i.e., at a given frequency and induction.
25(1953); http://dx.doi.org/10.1121/1.1907151View Description Hide Description
A new apparatus for investigation of sound propagation constants for large amplitude audio‐frequency vibrations in metals is described. One novel feature of the apparatus is the use of tuned resonant supports that are tilted with respect to the specimen. The tilting minimizes the tendency to chatter due to the Poisson effect. Because the supports are driven at their resonant frequency, large amplitudes may be achieved without the occurrence of an energy loss due to slipping at the contact surfaces between supports and test specimen. The error caused by energy dissipation due to the internal friction of the support is, in most cases, negligible. In those cases where it is not negligible, it can be closely estimated and corrected for. Auxiliary equipment permits determination of the resonant frequency of the specimen to better than 1/1000. The total error in the determination of the logarithmic decrement varies between 2 percent and 5 percent.
25(1953); http://dx.doi.org/10.1121/1.1907152View Description Hide Description
The frequencies of tones when metal rings are struck in radial and axial directions have been investigated in a variety of steel and bell‐metal rings.
The classical formulas relating frequency with ring dimensions and physical properties of the materials compare closely with actual measurements except for thick rings. The fundamental tone is predominant in the smaller rings but the fundamental fades and the overtones predominate as the rings increase in diameter and decrease in width and thickness.
Comments are given on musical properties of rings which may be promising but which have not been fully explored.
25(1953); http://dx.doi.org/10.1121/1.1907153View Description Hide Description
The theory of the transmittivity of immersed plates is extended to include the effects of losses associated with shear waves. The loss is computed for the transmission maxima which occur for five modes of vibration and is compared with observations made at a fixed frequency of . The general agreement is is good but not within the experimental error for all of the five modes. The value of the loss factor, which was determined to give agreement in a simple case, is difficult to appraise without data at other frequencies.
A Circular‐Orifice Number Describing Dependency of Primary Pfeifenton Frequency on Differential Pressure, Gas Density, and Orifice Geometry25(1953); http://dx.doi.org/10.1121/1.1907154View Description Hide Description
A nondimensional number for primary Pfeifentöne relating the dependence of differential pressure Δp across orifice to density of gas ρ, thickness t of orifice‐plate terminating pipe, and frequency f of primary Pfeifentöne is obtained by dimensional analysis. Correlation of experimental data presented on the basis of this number indicates that it is relatively constant in numerical value over range of variation of the parameters, pipe length, orifice diameter, differential pressure across orifice, and Pfeifenton frequency, studied.
25(1953); http://dx.doi.org/10.1121/1.1907155View Description Hide Description
Some numerical values have been computed for the functions which arise in connection with the scalar wave equation in rotational paraboloidal coordinates. Tables of these values are given with the aid of which a numerical analysis is carried out of the scattering of a plane sound wave by a rigid convex paraboloid of revolution. The surface of the paraboloid is defined by kξ0=0.25 and the direction of propagation of the plane wave makes an angle of 36.9° with the axis of symmetry of the paraboloid. The asymptotic expansions of the scatteredwaves are discussed and their amplitudes are tabulated. The magnitude and phase angle of the total pressure are evaluated at points on the surface of the paraboloid.
25(1953); http://dx.doi.org/10.1121/1.1907156View Description Hide Description
For low acoustic frequencies, a mixture (a porous medium or a suspension) is shown to have an effective density which differs slightly from the density given by Archimedes' principle. This effective density is computed from a physically elementary consideration of viscous, incompressible fluid flow. For higher frequencies, pore or particle size in the mixture becomes comparable with the wavelength of shear waves in the fluid, while still small compared with dilatational wavelength. The theory is extended to such frequencies through the known formula for the fluid's resistance to the oscillations of a rigid sphere. In both cases, the effective compressibility of the mixture is taken to be the volume‐average of the component compressibilities. From the effective density and compressibility, the acoustic properties of the mixture are predicted. Predictions are compared with previous theories and with experimental results.
25(1953); http://dx.doi.org/10.1121/1.1907157View Description Hide Description
Abnormal propagation of sound in the east‐to‐west and west‐to‐east directions has been studied throughout a period of a year. Experiments were carried out over the California‐Arizona desert using explosions of 1200 pounds of TNT. Returns of sound from altitudes of 30 to 50 km were consistently received to the east in winter and to the west in summer. Sounds which traveled to the high temperature region at 80 to 100 km altitude were received about one‐fifth of the time. Sounds were returned from the second region somewhat better to the east ha summer and to the west in winter. An early arriving abnormal wave was received at ranges greater than 400 km.
25(1953); http://dx.doi.org/10.1121/1.1907158View Description Hide Description
An extensive survey is made of the chemical effects of ultrasonicwaves. Increased understanding of the mechanism of these effects is sought through a consideration of the significance of certain experimental variables such as frequency, intensity, pressure, and temperature. Also considered is the role of cavitation.
25(1953); http://dx.doi.org/10.1121/1.1907159View Description Hide Description
The ultrasonicabsorption coefficients in water solutions of magnesium sulfate were measured at 30 Mc as a function of temperature and concentration. Temperatures from 0 to 30°C and concentrations from 0.05 to 2.00 molarity were used. Measured data were used to calculate Q, the absorption cross section per molecule of magnesium sulfate at several temperatures. Q is defined by (2α M /ν2)/N where N is the number of molecules of magnesium sulfate per cc, ν is the frequency, and α M is the amplitude absorption coefficient caused by the magnesium sulfate. These are constant for concentrations up to 1‐molarity but decrease by 50 percent for 2‐molarity concentration. The constant values of Q×1037, 25 at 25°C, 30 at 20°C and 37 at 15°C indicate an activation energy for the magnesium sulfate absorption process at 30 Mc of 6.8 kcal/mole°K.
25(1953); http://dx.doi.org/10.1121/1.1907160View Description Hide Description
In a dispersive medium the relative phases of the components of a complex sound wave change with path length. The present article describes a method of measuring dispersion in which the change in relative phases of a complex wave consisting of a fundamental and a selected harmonic are measured as a function of the path length. Sensitivities, in measuring difference in velocity, of one part in 105 with a probable error of 3 parts in 105 were obtained experimentally. Dispersion measurements made in water solutions of below one megacycle show linear dependence upon concentration up to 0.5 mole per liter and agree with those calculated from absorption data. The data indicate total dispersion of 13.6×10−4 per mole per liter, with a relaxation frequency of 160 kc at 24°C.
25(1953); http://dx.doi.org/10.1121/1.1907161View Description Hide Description
A 2.5‐Mc, barium‐titanate, spherically focusing radiator was used to produce cavitation in both degassed and aerated water entirely within the restricted, high intensity focal region, remote from the water boundaries. The sonic intensity rises to 1.8 kw/cm2, and the pressure amplitude to ±70 atmos at the focus. High‐intensity illumination and an unusual high speed photographic technique permit observation and timing of the step‐by‐step process of cavitation development.
Feather‐shaped cavitation bursts are sporadically produced, being initiated in the insignificant quill portion nearest the radiator, then abruptly expanding to form the catastrophic plume portion. The plume is believed to be formed by myriads of micro‐cavities, too small and close for individual observation. These two fundamental steps are identically produced, and with equal ease, both in degassed and aerated water. The whole action is over in several milliseconds, except that in the case of aerated water a third bubble step is produced. In aerated water, non‐collapsing gas bubbles are generated by and concurrently with, the catastrophic step. These bubbles remain after collapse of the burst, to be blown off down stream by the sonically induced liquid streaming.
The bubble step is not generated without the presence of the catastrophic step. The latter is generated only if the initiation step reaches a definite degree of development (not always attained). This requires sonic activation for increasing lengths of time for decreasingly smaller sonic intensities. Origination of the initiation step, and hence of the whole cavitation phenomena, is believed to occur whenever a stray nucleus (weak spot) streams into the high intensity sonic field.
25(1953); http://dx.doi.org/10.1121/1.1907162View Description Hide Description
When the mechanical damping of a transducer is large, as by acoustic radiation from one or both faces into a liquid or solid, the circular diagram that represents its characteristics requires special treatment. As a background for this treatment, the uses and limitations of the conventional circle for a resonator with small losses is first reviewed. The problem of the transducer with large losses is then considered with special reference to the equations and graphs for a thickness‐type transducer with unsymmetrical loading. For plane‐wave transducers the expressions are exact for all loads and at all frequencies, including harmonics. Either the voltage or the current may be constant. From the admittance or impedance diagrams the magnitude and phase of current, voltage, particle velocity, and vibrational amplitude at any frequency can be obtained immediately. Similar results would be found with plates in lengthwise vibration. A new type of diagram is developed for representing vibrational amplitudes. As an illustration, the case of a quartz plate radiating into three liquids of widely different acoustic properties is treated.
When the load is unsymmetrical, there is no true node anywhere in the crystal except when the load is zero or infinity. There is, however, a plane of minimal vibration, the amplitude and location of which are derived. The equations indicate certain peculiar effects when the specific acoustic resistance of the medium is just twice that of the crystal.
25(1953); http://dx.doi.org/10.1121/1.1907163View Description Hide Description
It is frequently essential for a transducer to have a smooth, wide‐beam directivity pattern with negligible side lobes. If, in addition, this transducer must operate in the fractional megacycle frequency region, then conventional designs such as small plane radiators or curved surface mosaics are not satisfactory. As a result of an extensive investigation, it was found that a bariumtitanate spherical shell sector properly designed will meet these difficult wide‐pattern and high frequency specifications. This paper outlines some of the design considerations required to adapt a ceramic shell into a transducer with the desired performance. The problems discussed are shell size, included angle of the spherical shell, inside diameter to thickness ratio, and the uniformity of the thickness dimension. Various baffle techniques are also examined detail. Shell arc lengths ranging from four to twenty wavelengths are considered as to their effects on the shape of the surface displacement distribution and on the directivity pattern. Though this paper presents a solution to a specific transducer problem, the design techniques discussed are applicable to transducers having other frequency and pattern requirements.
Electromechanical Response and Dielectric Loss of Prepolarized Barium Titanate under Maintained Electric Bias. Part I25(1953); http://dx.doi.org/10.1121/1.1907164View Description Hide Description
In transducer applications of ferroelectricceramics it is standard practice to rely for the electromechanical response on retained polarization. This is practicable only for moderate driving amplitudes and sufficiently low temperature. At higher driving fields dielectric losses increase inordinately and lead to eventual depolarization and loss of response. This can be remedied by application of a comparatively modest dc bias. The quantitative loss behavior of various ceramic bodies over extended ranges of temperature, field amplitude, and for various values of aiding bias is investigated. Results are also presented on various other effects obtained with bias operation, such as increase of electromechanical coupling and shifts of thermodynamic transition points. Work on other associated phenomena—electrostriction, secular relaxation, etc.—will be reported in part II.
25(1953); http://dx.doi.org/10.1121/1.1907165View Description Hide Description
Information about the velocity of sound in the earth is normally obtained by shooting dynamite at the surface and measuring traveltime to a detector suspended by cable within a deep well. In such measurements errors are sometimes introduced by noise transmitted down the cable. A piezoelectrichydrophone for use in deep wells was constructed and has been useful in discriminating against such noise.Hydrophones are also well suited for tandem use in wells to measure interval velocities. Their substantially aperiodic response facilitates resolution of shear waves, and their relatively constant coupling to the earth makes possible measurement of attenuation. Several types of hydrophone are described and the results of some field experiments are discussed.
25(1953); http://dx.doi.org/10.1121/1.1907166View Description Hide Description
An array of three pressure pick‐ups 100 ft apart covering a vertical range of 200 ft was used to record sound waves in two wells drilled to depths exceeding 12 000 ft. Dynamite charges were exploded in 80‐ft holes at a distance of 500 ft from the well‐head. A separate charge was fired with the array at each 500‐ft interval in the well. The recorded events were analyzed and found to belong to two categories. The first category includes the pressure pulse arriving at the pick‐up by a more or less direct route through the sedimentary rocks. The second consists of events set up in the mud column at points other than adjacent to the pick‐up by the compressional wave in the sedimentary rocks. These events then travel through the mud column and are recorded as they pass the pick‐up. Anomalies in the hole, such as the ends of casing strings and the bottom of the hole, cause events of larger amplitude than those set up in sections of the hole with no major anomalies.
25(1953); http://dx.doi.org/10.1121/1.1907167View Description Hide Description
Availability of a pocket‐sized miniature sound level meter meeting all ASA standards has created demand for a small, equally portable sound analyzer as a companion instrument. The analyzer described surpasses all applicable ASA standards. The separate high and low cut‐off filters may be adjusted independently, and the noise pass band may be locked in any band width which is an integral multiple of , from through the standard 1 octave band, etc. The range of the filters extends one octave lower and one octave higher than the ASA standards. These features permit considerable added precision and flexibility of analysis.
For protection and ease of carrying, the instrument is housed in a leather case similar to those used for amateur motion picture cameras. The case is equipped with a shoulder strap which allows the analyzer to be carried about easily while in use. It can also be used as a general purpose adjustable filter or as a wide‐range amplifier.
25(1953); http://dx.doi.org/10.1121/1.1907168View Description Hide Description
The travel time of sound across fixed gap between diaphragms of transmitting and receiving transducers is recorded each 1/500 second. From this the interdiaphragm velocity is computed. It is a function of both temperature and wind. Spurious travel time changes introduced by phase shifts in the receiver are minimized by using a 500‐cps intelligence signal amplitude modulating a 10‐kc carrier. Experimental errors when operating in air with interdiaphragm distances of 3 feet are less than 2 percent in the temperature range from 70° to 200°F, but the application of the instrument is not restricted to this range. A circuit for recording velocity directly on a paper tape is described. Although designed primarily for air, the unit is adaptable to other media. It should also be of special interest to the micrometeorologist.
25(1953); http://dx.doi.org/10.1121/1.1907169View Description Hide Description
The design and construction of an electrical analog of the human vocal tract is described. The vocal tract is viewed as an acoustic tube of varying cross‐sectional area, terminated by the vocal cords at one end and by the lip opening at the other. The analog is a lumped‐constant electrical transmission line consisting of thirty‐five pi‐sections. Each section represents a length of the vocal tract and is adjustable to simulate a range of cross‐sectional areas from 0.17 to 17 cm2. The electrical line can be excited by a periodic current source representing the vocal cord output or by a random voltage representing the noise from turbulent air flow at a constriction. The electrical analog can synthesize with good quality all English vowels and some consonants. The physical characteristics of the output of the analog for each sound are shown in terms of measured formant frequencies or by conventional spectrograms. For each sound, the vocal tract dimensions that the electrical network simulates are shown in graphical form. Applications of the speech synthesizer to linguistic and engineering research are discussed briefly.