Volume 30, Issue 4, April 1958
Index of content:
30(1958); http://dx.doi.org/10.1121/1.1909553View Description Hide Description
The difference limen for the azimuth of a source of pure tones was measured as a function of the frequency of the tone and the direction of the source. Tone pulses between 250 and 10 000 cps were sounded in the horizontal plane around the head of a subject seated in an anechoic chamber. The smallest angular separation that can be detected between the sources of two successive tone pulses (the minimum audible angle) was determined for each of three subjects. These threshold angles are analyzed in terms of the corresponding threshold changes in the phase, time, and intensity of the tone at the ears of the subject. A comparison of these thresholds with those reported for dichotic stimulation indicates that the resolution of the direction of a source is determined, at frequencies below about 1400 cps, by interaural differences in phase or time, and at higher frequencies by differences in intensity. When the conditions are optimal for temporal discrimination, the threshold for an interaural difference in time is about 10μsec, and when the conditions are optimal for intensity discrimination, the threshold for an interaural difference in intensity is about 0.5 db.
30(1958); http://dx.doi.org/10.1121/1.1909555View Description Hide Description
Oxygen tension of the endolymph in the scala media of the guinea pig cochlea was measured using a micro‐polarographic technique. Near the stria vascular is the oxygen tension was 55 to 70 mm Hg. Deeper in the scala media it decreased gradually to 16 to 25 mm Hg.Effects of sound, hypoxia, hyperoxia, and hypercapnea were studied. Oxygen tension decreased with hypoxia and sound, and it increased with hyperoxia and hypercapnea. A technique to check the integrity of a metallic‐plated microelectrode is described.
30(1958); http://dx.doi.org/10.1121/1.1909557View Description Hide Description
Under special conditions, numerous investigators have observed that, when duration of exposure is held constant, a greater temporary threshold shift (TTS) may occur after exposure to a lower sound level than after exposure to a higher sound level. However, at least after a short recovery interval, it would be reasonable to expect that TTS following exposure to higher sound levels would be an increasing function of the exposure level. In the present study TTS was measured continuously with a Békésy‐type audiometer at 4000 and 6000 cps for 10 min following a 3‐min exposure to a thermal noise. Five different noise levels, ranging in 5‐db steps from 108 to 128 db in SPL, were used. Among nine ears tested, twice on each condition, only one ear showed a pattern of decreasing TTS with an increased noise exposure level at both measuring frequencies.
30(1958); http://dx.doi.org/10.1121/1.1909559View Description Hide Description
The thresholds of audibility for 100 and for 1000 cps have been measured on five groups of originally naive listeners by various experimental techniques. All the experiments showed improvement of the threshold with practice. The improvement was greater at 100 cps than at 1000 cps. Pretraining at 1000 cps did not affect the threshold change at 100 cps. The improvement of the threshold with practice was enhanced considerably by reward and feedback.
30(1958); http://dx.doi.org/10.1121/1.1909561View Description Hide Description
The article first examines the various sources of error in absolute calibration of standard microphones in a pistonphone. A new differential pistonphone has been developed in the Czechoslovak Research Institute for Telecommunications which enables absolute calibration of microphones with the accuracy 0.1 db. The design and properties of the instrument are described.
30(1958); http://dx.doi.org/10.1121/1.1909563View Description Hide Description
The principal characteristic features of nasalization of vowels are (1) a dull resonance around 250 cps, (2) an antiresonance at about 500 cps, (3) comparatively weak and diffuse components which fill the valleys between the formants.
Features (1) and (3) are commonly found also during the period of oral closure of nasal constants, (3) in this case being influenced by the antiresonance of the oral cavity. The feature (3) affected by “antiformants” carries information about the tongue position during oral closure. If the nasal passage is stopped at the nostrils when nasalized vowels are pronounced, features (1) and (3) disappear, and the frequency of selective attenuation (2) is lowered. Some French and Japanese nasalized vowels have been examined and interpreted, applying the results obtained in the research.
Relative Intensities of Sounds at Various Anatomical Locations of the Head and Neck during Phonation of the Vowels30(1958); http://dx.doi.org/10.1121/1.1909565View Description Hide Description
Speech signals from 16 different anatomical locations were recorded as subjects intoned different vowels at a constant level. Power level analysis was made to determine relative intensity of the signals. It was found that significant difference in intensity exists among the anatomical locations. Some locations of lesser intensity were subjectively evaluated as providing more faithful signals.
30(1958); http://dx.doi.org/10.1121/1.1909567View Description Hide Description
Person‐to‐person tests of sentence intelligibility were carried out in low frequency and white noise at noise levels ranging from 85 to 118 db. Talkers attained shouting levels of vocal effort but the maximum tolerable noise levels for 90% sentence intelligibility and 1 m between talker and listener were estimated to be 95 db for white noise and 105 db for low‐frequency noise.
30(1958); http://dx.doi.org/10.1121/1.1909569View Description Hide Description
Successive word intelligibility tests were executed at a constant S/N ratio over a 13‐min period for a range of white noise levels between 45 and 130 db. Speech intelligibility declined over the test period as a result of continued exposure to noise levels of 115 db and higher. Interactions among different noise levels on successive tests, as evidenced by speech intelligibility scores in noise, was minimal when the noise levels were separated by less than 80 db.
30(1958); http://dx.doi.org/10.1121/1.1909571View Description Hide Description
A rating scale was added to the standard articulation test procedure in order to obtain independent information about a listener's criterion for message acceptance or rejection. We find that assignment of confidence ratings does not interfere with the accuracy of message reception. The form of the receiver operating characteristic—the relationship between correct confirmations and false alarms—yielded by the rating procedure is similar to that yielded by a binary decision of message acceptance or rejection. In addition, the confidence rating is directly related to the average accuracy of message reception. This relationship is relatively invariant over a range of speech‐to‐noise ratios.
30(1958); http://dx.doi.org/10.1121/1.1909573View Description Hide Description
The role of interaural effects upon speech intelligibility was examined at high noise levels. Over a wide range of conditions, a “mixed” listening condition (speech and noise to one ear and noise alone to the other ear) produced substantially lower intelligibility scores than monaural or binaural presentation of both speech and noise. The results are interpreted in terms of the interaction between the intelligibility of speech cross‐conducted to the other ear and the apparent localization in space of the speech and noise.
30(1958); http://dx.doi.org/10.1121/1.1909575View Description Hide Description
The velocity of ultrasonic waves in oxygen was measured at 40.6°C and 2 Mc/sec, with pressures ranging from one atmos to 0.35 cm Hg.Velocitydispersion was found at the lower pressures and is accounted for in part by rotational relaxation and at the lower pressures by the onset of the Stokes type of dispersion. The midpoint of the dispersion curve lies at 122 Mc/atmos, which corresponds to a relaxation time of 21.76×10−10 sec and to 13 as the average number of collisions necessary for the energy exchange. Consideration of the added translational dispersion effect establishes 12 as the average rotational collision number. A Hubbard‐type sonic interferometer was used together with a strip recorder, several different tuning techniques being employed at the lowest pressures.
30(1958); http://dx.doi.org/10.1121/1.1909577View Description Hide Description
Isotopic techniques involving oxygen‐18 have been used to study the processes involved in the sonochemical formation of hydrogen peroxide during cavitation in oxygen‐saturated solutions. Both tracer and equilibration experiments were carried out in an all glass system with converging ultrasonic fields providing maximum intensities of 103 w/cm2 at frequencies of 600 and 800 kc/sec. In oxygen‐saturated water, one‐third of the hydrogen peroxide oxygen was found to be derived from the oxygen gas while in solutions containing 30% (by volume) methanol, all of the peroxide oxygen was derived from the oxygen gas. In both cases, the oxygen molecule appears to be incorporated as a unit in one peroxide molecule without formal cleavage of the O ‐ O bond.
Conditions associated with ultrasonically produced cavitation appear sufficient to dissociate water into H and OH and to rupture the C ‐ H bond in organic molecules such as methanol but not sufficient to dissociate molecular oxygen. The oxygen molecule is involved in secondary reactions with hydrogen atoms, and hence, contributes to the formation of hydrogen peroxide. The experimental results can be explained most readily in terms of the large temperature rises which are presumed to occur within cavitation during compression but do not exclude electrical phenomena within the cavitation bubbles as a possible mechanism for the dissociation processes.
30(1958); http://dx.doi.org/10.1121/1.1909579View Description Hide Description
A new method has been developed for the accurate measurement of the velocity of compressional waves in solid specimens. This is based on the cancellation of a traveling wave train, which has undergone reflection at the ends of the specimen, by a second wave train launched from the same transducer. A description is given of the experimental system employed.
A correction must be made for the phase shift which occurs when the ultrasonic signal is reflected at the interface between specimen and transducer. A theoretical study of the effects of the transducer and coupling film shows that the correction is a function of the resonant frequency of the transducer and of the ratio of acoustic impedance of the transducer to that of the specimen.
The application of the correction enables the velocity of propagation to be determined to within 1 in 104.
30(1958); http://dx.doi.org/10.1121/1.1909581View Description Hide Description
The adiabatic elastic moduli of single crystalgermanium are reported as a function of hydrostatic pressure to 50 000 psi, and for a temperature range of 0–75°C. The variation with pressure of bulk modulus is found to compare favorably with isothermal values obtained by P. W. Bridgman. The techniques used for measurement are detailed.
30(1958); http://dx.doi.org/10.1121/1.1909583View Description Hide Description
For low Mach numberturbulence, Lighthill has given a result which expresses the acoustic radiation of turbulent fluid in terms of a fourth‐rank velocity correlation. Through the use of the Navier‐Stokes equation this relation can be put in terms of the space‐time pressure correlation. For very large Reynolds number turbulence (in the inertial subrange where Kolmogoroff's similarity principles are valid) one can write this pressure correlation in terms of a function of a single variable. Since the space‐time correlation of the pressure is needed one must take into account convective effects. This is done here by introducing a new Lagrangian type of correlation which is defined in such a way that a similarity argument can be applied without difficulty from the convective effects. Such convective effects then enter only through negligible Doppler shifts. Using this similarity result and Lighthill's formulation, one obtains the acoustic self‐noise power spectrum for the turbulence. The spectrum is proportional to ω−7/2 M 21/2 for high frequencies where ω is the acoustic angular frequency, M is the turbulenceMach number,c 0 is the velocity of sound in the fluid, and L is the size of the large‐scale turbulenteddies. The spectrum at the high‐frequency end is universal, that is it is independent of the details of the driving mechanism. Furthermore the spectrum is proportional to ω4 M 3 at the low‐frequency end and depends on the large scale eddies there. The similarity hypothesis made here for the special Lagrangian type of space‐time correlation is of interest in itself in turbulence theory. It is difficult to check this hypothesis directly; however, a measurement of the acoustic power spectrum offers an interesting indirect check on the hypothesis.
30(1958); http://dx.doi.org/10.1121/1.1909585View Description Hide Description
The problem of propagation of explosive sound in a layered liquid is formulated operationally, and an exact solution is derived based on ray theory. The contribution of each ray at a given time is expressed by a single definite integral taken over a fixed finite range. With the aid of an electronic computer the pressure‐field was evaluated both for the case of a slow‐speed bottom and for the case of a high‐speed bottom. It was found that for a Heaviside unit pulse and a high‐speed bottom the pressure becomes logarithmically infinite at the time of arrival of the totally reflected wave. For a high‐speed bottom, the pressure pulse at large ranges, as computed by the ray theory, exhibits the well‐known features, first observed by Ewing and Worzel and later deduced from the normal‐mode theory, of a long‐period ground‐wave followed by a high‐frequency dispersive water‐wave. We have found, furthermore, that at large ranges sufficient accuracy can be obtained by retaining only the contributions from a group of the last arriving quadruplets of rays. With this approximation it becomes feasible to use ray theory for the evaluation of the pressure field even at large ranges.
30(1958); http://dx.doi.org/10.1121/1.1909587View Description Hide Description
An approximate solution is developed for sonically‐induced steady flow near a fluid‐solid interface. The result is valid, subject to stated conditions, for the flow near any portion of surface in the vicinity of which the irrotational oscillatory velocity distribution u a is known. The principal condition on the validity is that the acoustic boundary layer parameter (where ω is the angular frequency and ν is the kinematicviscosity coefficient for the fluid) should be small compared to the scale of u a . Applications of the general result are made to special situations, one case of particular interest being that of a small source near a rigid plane. The conclusion is reached that small compressible bodies, and especially resonant gas bubbles, resting on boundaries, are likely sites of pronounced microstreaming a sound field.
30(1958); http://dx.doi.org/10.1121/1.1909589View Description Hide Description
The attenuation of high amplitude saw‐tooth waves of stable form in horns is investigated theoretically. The shock associated with each wave is assumed to be weak. An expression for the power loss for a generalized horn is obtained. Two quantities, the limiting particle velocity amplitude and the limiting power which is transmitted per refit throat area, occur in the solution. For long, gently tapering horns these are the limits to which the particle velocity and power tend as the input to the throat is increased. Uniform bore tubes, and exponential and conical horns, are discussed as particular examples of the general case.
30(1958); http://dx.doi.org/10.1121/1.1909590View Description Hide Description
Short sound pulses at 100 kc were transmitted over a 100‐ft range in a shallow estuary. Surface‐reflected signals were recorded separately from direct signals. Surface reflection was most pronounced in fall and winter when vertical sound‐speed distribution produces upward refraction in the upper ten feet of water. In spring and summer the opposite conditions prevail. Eight one‐minute records, consisting of about 1200 pulses each, were analyzed. When pulse amplitudes and their frequencies of occurrence were transformed into dimensionless units, it became feasible to combine all records into one frequency distribution diagram. Despite considerable scatter in the data, there is a marked resemblance to a Rayleigh distribution. This is to be expected on theoretical grounds.