Volume 27, Issue 5, September 1955
Index of content:
27(1955); http://dx.doi.org/10.1121/1.1908048View Description Hide Description
This paper reviews the available evidence (published and unpublished) on the relation between loudness and stimulus intensity. The evidence suggests that for the typical listener the loudnessL of a 1000‐cycle tone can be approximated by a power function of intensity I, of which the exponent is log102. The equation is: L = kI 0.3. Intensity here is assumed to be proportional to the square of the sound pressure.
In terms of sones, where 1 sone is the loudness produced by a tone at 40 db above the standard reference level, the equation for loudnessL as a function of the number of decibelsN becomes: .
Otherwise said, a loudness ratio of 2:1 is produced by a pair of stimuli that differ by 10 db, and this relation appears to hold over the entire range of audible intensities.
At low levels of intensity, the loudness of white noise grows more rapidly than the loudness of a 1000‐cycle tone, but above the level of approximately 50 db the two loudnesses remain more nearly proportional. The suggestion is made that for all levels greater than 50 db the loudness of continuous noises may be calculated from the equation: , where S is a spectrum parameter to be determined empirically.
27(1955); http://dx.doi.org/10.1121/1.1908050View Description Hide Description
The membrane of a dimensional model of the mechanical parts of the human cochlea (30 cm in over‐all length) showed under stroboscopic illumination traveling waves similar to those on the cochlea and a flat maximum of vibration which shifted from one end of the model to the other when the frequency was increased from 40 cps to 160 cps. When the skin of the arm is placed in close contact with the membrane, the vibrations are localized in a quite narrow section of the membrane, so that any change in frequency can be detected by a change in the locus of stimulation. The localization is as sharply defined for pure tone pulses of the length of two sine waves as for continuous tones. The localization of pure tones along the membrane is practically instantaneous.
The sharp localization in spite of the flat distribution of vibrations along the membrane is produced by the inhibitory effects of the skin. Their influence and magnitude are demonstrated by various experiments.
Some basic relations between nerve density on the cortex, threshold, and loudness increase on the skin are pointed out in the hope that they will later enable us to determine the differences between the organ of Corti and the simple skinmembrane. This would be helpful to our understanding of the nature and function of the complicated structure of the organ of Corti. On the whole, in the observation of pitch of pure tones, the simple skin‐model is as useful as the organ of Corti itself. The differences become apparent only in the observation of more complex sounds.
Two models were used to investigate the possibility of obtaining effects similar to the stereophonic effects in hearing.
27(1955); http://dx.doi.org/10.1121/1.1908052View Description Hide Description
Three experiments on the localization of air‐borne sound are described. They were conducted in an anechoic room and employed an acoustical “pointer” as the subject's method of indicating the direction of the stimulus tone. The pointer was a small loudspeaker carried on a boom which rotated about a vertical axis through the subject's head. This speaker presented a wide‐band noise which alternated with the tone to be localized. The switching was transientless and was performed by an electronic gate having a 100‐millisecond rise and decay time.
Three small loudspeakers in enclosures, one mounted directly in front of the subject and one 40° to each side, presented the stimulus tones to be localized. In the first experiment the speakers were employed singly. In the other two experiments they were used in pairs. In the second experiment the pairs of speakers were in phase; in the third, they were in phase opposition.
The stimulus conditions of Experiment 2 generate a “phantom source” which appears to lie between the two speakers employed. The predicted location is compared with the subjects' responses.
The stimulus conditions of Experiment 3 generate a phantom source which according to prediction, should lie toward the side opposite to the asymmetrically placed speaker and should move in direction with frequency. This prediction is borne out by the subjects' responses for frequencies where interaural time is the dominant basis for localization.
The results of the three experiments support the conclusion that the localization of tones below about 1500 cps is determined largely by interaural time differences. For frequencies above 1500 cps differences of intensity at the two ears must be the dominant factor, but neither probe microphone measurements nor audiograms for the subjects provided information from which satisfactory predictions could be made.
27(1955); http://dx.doi.org/10.1121/1.1908054View Description Hide Description
Measurements of the cochlear potentials as recorded from the round window were carried out on a series of guinea pig ears to ascertain the effects of injury by overstimulation with various tones. The results showed general impairments of the responses after such injury. With rare exceptions, the maximum responses to any tone are much less impaired than the sensitivity, and the maximums are more seriously affected for the low tones. In general, these changes are independent of the frequency of the tone producing the injury.
These observations confirm those of earlier studies on the patterns of action of tones in the cochlea, and indicate that all tones when raised to extreme levels cause widespread damage to the hair cells. It is suggested that certain hair cells, which are normally involved in the peak actions of all tones, are particularly susceptible to damage by overstimulation.
27(1955); http://dx.doi.org/10.1121/1.1908056View Description Hide Description
Various ways of creating artificial vowels are discussed and compared from the viewpoint of coding efficiency. A method which requires seven parameters to specify any vowel is then introduced and described in detail. It is based on the realization of the transfer impedance of a nonuniform transmission line by lumped elements.
27(1955); http://dx.doi.org/10.1121/1.1908059View Description Hide Description
An attempt was made to improve listeners' ability to respond to one of two or more simultaneous or overlapping voice messages by differentially filtering the competing message channels.
The limits of satisfactory intelligibility of two digits and two International Civil Aeronautics Organization (ICAO) phonetic alphabet words in a message also containing an address were determined under various conditions of high‐ and low‐pass filtering and speech‐to‐noise ratios (S/N). Within these limits, several amounts of filtering were tried in various combinations.
When performance on differentially filtered competing messages was compared with performance on unfiltered competing messages, it was found that differential filtering almost always improved performance. This was true regardless of whether it was the desired message that was filtered or whether it was one or more of the competing messages that was filtered, and the degree of filtering was relatively uncritical. High‐pass cutoffs of 565, 800, or 1130 cps were about equally effective and were considerably better than high‐pass cutoffs at or below 280 cps. Low‐pass cutoffs of 1600, 1130, and 800 cps also improved performance substantially.
Comparison of Hyponasality, Hypernasality, and Normal Voice Quality on the Intelligibility of Two‐Digit Numbers27(1955); http://dx.doi.org/10.1121/1.1908060View Description Hide Description
Six Eastern and General American speakers, using hyponasal, hypernasal, and normal voice qualities recorded random lists of two‐digit numbers for presentation in noise to 11 trained listeners, all of whom had 32 hours practice in listening and testing in noise. Results indicate that, while voice qualities acted in a relatively similar manner at both signal‐to‐noise ratios tested, hypernasality was destructive to intelligibility under both conditions, hyponasality was destructive at the milder noise level, and normal delivery was best throughout.
27(1955); http://dx.doi.org/10.1121/1.1908062View Description Hide Description
Test responses of listeners from 27 countries who were tested over a period of a year and a half on several versions of word spelling alphabets are examined to determine patterns of intraensemble confusion. On the word level, confusion seems to hinge on the accented vowel, and these confusions are stratified according to the articulatory position of the vowel. Greater discrimination for front vowels than for back is noted, as well as diminishing intraensemble confusion from high‐to‐low‐articulatory position. A method for computing confusion vectors is presented along with a rationale for increasing the phonemic stability of any particular set of words.
27(1955); http://dx.doi.org/10.1121/1.1908064View Description Hide Description
An auditory test to establish the adequacy of an idealized specification of vowel articulation is described. Articulatory configurations of the vocal tract are specified by means of three numbers: the position of the point of greatest constriction, the degree of constriction at that point, and the size of the mouth opening. The stimuli in the recorded test are the outputs of 279 such configurations of an electrical vocal‐tract analog. All sound samples are highly similar in duration, intensity, start‐stop characteristics, and fundamental frequency and inflection. The instructions to subjects are standardized, and are of two types: one restricts responses to one of nine vowel categories while the other permits a nonvowel response as well. Subjects were experienced in phonetic transcription.
The ranges of articulatory configurations identified as given vowels by the subjects are plotted as contours, and these data are compared to results of other investigators. Certain disagreements are pointed out and discussed. The results, in general, show good agreement with previous data, and demonstrate, therefore, the adequacy of the three‐number articulatory description.
27(1955); http://dx.doi.org/10.1121/1.1908066View Description Hide Description
The divergence factor, which is of importance in the reflection of waves from a convex spherical surface, is shown to have its counterpart in plane reflection and refraction in the presence of a velocity gradient. Within a small angle approximation, linear velocity gradient gives the ordinary spherical spreading of a pencil of rays; any discontinuity of gradient causes divergence or convergence of the pencil. Analytical expressions and tables of typical values are calculated for four different cases.
27(1955); http://dx.doi.org/10.1121/1.1908068View Description Hide Description
An apparatus is described which allows one to measure the normal and shearing component of the force of radiation pressure exerted by an ultrasonic beam through water on an intercepting disk. Results of measurements on an absorbing and on a reflecting disk are seen to confirm the theory of Brillouin that radiation pressure is due to a state of unidirectional stress in the region of a plane progressive wave and that it is properly represented by an anisotropic stress tensor. It also appears that the apparatus with the absorbing disk provides a very simple, and at the same time, accurate device for the absolute measurement of the power output of an acoustic transducer. Furthermore, it is shown how the measurement of radiation force components with this apparatus can be used for the determination of the average reflection‐, transmission‐, and absorption‐coefficients of any sample material as functions of angle of incidence. Results of measurements on cork, aluminium, and steel plates are presented.
27(1955); http://dx.doi.org/10.1121/1.1908070View Description Hide Description
In a previous article it was shown that if plane wave reflection coefficients were written as rotations, the characteristic equation for the normal modes could be expressed very simply by means of the arguments. The method is elaborated here for the case of multilayered media. Also, an exact and general solution the guided wave terms due to a simple harmonic point source in a duct is obtained. This is done by applying our characteristic equation to an integral solution due to Brekhovskih. A numerical example is given, illustrating quasi‐resonance and anti‐resonance phenomena associated with stationary values of the group‐velocity under simple‐harmonic conditions.
27(1955); http://dx.doi.org/10.1121/1.1908072View Description Hide Description
The ultrasonic shutter is a thickness‐expander plate of bariumtitanate which, when placed in a beam of ultra‐sound, will control the transmission of sound through it over a 40 db range near the frequency of half‐wave resonance. Control is exerted by means of a passive electrical network connected directly to the plate electrodes.
Theory of the “shutter effect” and supporting experimental data are presented.
27(1955); http://dx.doi.org/10.1121/1.1908074View Description Hide Description
A convenient and economical method for the production of cavitation damage is described. Cavitation is generated in the region of the maximum pressure amplitude of an acoustic field in a resonant cylindrical container. The specimen surface to be damaged is a stationary boundary in the system and is therefore not subject to accelerations such as occur in conventional magnetostrictiondevices for cavitation damage study. The absence of these accelerations eliminates any possibility of premature material removal from cavitation weakened surfaces. Mathematical analysis of the dynamics of cavitation should also be facilitated. It is shown that simple assumptions lead to an acoustical theory agreeing with experiment to within the accuracy required. Examples are given of the application of this technique to the study of both easily damaged and highly resistant materials. Relatively uniform damage is achieved which is particularly suitable for x‐ray diffraction study of the surface. Alternatively, perturbation of the acoustic field to permit concentrated damage is obtained by forming the specimen surface as the end of a small cylinder. Photomicrographic studies of damage to both monocrystalline and polycrystalline materials have been initiated with the aid of this apparatus. Cavitation experiments have been done in water with an air atmosphere and in toluene with a helium atmosphere. These experiments show that severe damage can result even if a chemically inert environment is employed.
27(1955); http://dx.doi.org/10.1121/1.1908076View Description Hide Description
The elastic constants of a single crystal of zircon have been determined using the ultrasonic wedge method. The values of the constants are (units of 1011 dynes/cm2) c 11 = 7.35, c 33 = 4.60, c 44 = 1.38, c 66 = 1.60, c 12 = 0.90 and c 13 = −0.54. The value of compressibility .
On the Lowest Flexural Resonant Frequency of a Circular Disk of Linearly Varying Thickness Driven at Its Center27(1955); http://dx.doi.org/10.1121/1.1908078View Description Hide Description
A solution for the lowest, flexural, resonant frequency of a center‐driven, circular disk of linearly varying thickness is obtained by application of the Rayleigh‐Ritz method. Experimental results obtained with a series of aluminum disks are given and compared with those predicted by the analysis.
27(1955); http://dx.doi.org/10.1121/1.1908080View Description Hide Description
The characteristic feature of B. Vonnegut's vortex whistle is the proportionality of pitch to flow rate [B. Vonnegut, J. Acoust. Soc. Am. 26, 18 (1954)], in contrast with more common types of whistles in which geometry determines the pitch. A formula for the performance of vortex whistles was proposed by its discoverer on the basis of tests; it is confirmed below on the basis of elementary physical considerations which show that high velocity and multi‐dimensionality of the jet are essential.
27(1955); http://dx.doi.org/10.1121/1.1908082View Description Hide Description
An instrument is described which combines the schlieren technique, a photomultiplier tube and an oscilloscope in a manner that enables one to measure the density distribution behind shock waves (in gases) produced in a shock tube. From these measured distributions one can determine the way in which the gas or gases relax to equilibrium and the time it takes to reach equilibrium after the gas's enthalpy is increased suddenly a calculable amount in passing through the shock wave.
The theory of the instrument is discussed and its predicted performance verified experimentally by measuring vibrational heat capacityrelaxation times behind shock waves in containing water vapor. The instrument in the reported tests demonstrated a sensitivity sufficient to record a change in atmospheric density of over 1‐mm distance and a space resolution of the density in the shock tube of 1/10 mm corresponding to times of the order of 1/10 μsec.
27(1955); http://dx.doi.org/10.1121/1.1908084View Description Hide Description
Impedance transformations of the “Hypex” horns are derived for all cases of mismatch for both increasing and decreasing characteristic impedance. A method of using a Smith Chart for these transformations is presented. The results are applicable to cases where losses are appreciable.
27(1955); http://dx.doi.org/10.1121/1.1908086View Description Hide Description
Square bars of polycrystallinebariumtitanateceramic have been prepolarized and their lowest longitudinal resonance excited piezoelectrically on remanent polarization. Bars were prepared as twins in every way, except that one twin was electroded on its ends, the other on a pair of opposite sides. The first polarization applied was that minimum (800 v/cm for 30 minutes at 27°C) which fixed any, electromechanical activity as detected by a sensitive ac bridge, and this was taken to approximate the unpolarized, isotropic state (s 11=s 33). Different pairs were then given different large scale polarizations (2.5, 5.0, and 10 kv/cm cooled from above the Curie point) and changes noted. All resonant frequencies rose with time, the decay constant being between 0.01 and 0.02 hr−1, and the total change being 1 to for “Ceramic A” (bariumtitanate), and for “Ceramic B” (bariumcalciumtitanate). For both materials the isotropic compliance was definitely much closer to than to , and averaged closer to than to . The ratios s 33:s 11 were 0.83 and 0.87 at constant D and 1.04, and 1.05 at constant E for the two materials after 10 kv/cm. Anisotropy varied somewhat with polarizing field strength.