Chronicle

Extended Meeting of the Commission on Acoustics of the USSR Academy of Sciences

From February 1 to 3 of this year, an extended meeting of the Commission on Acoustics of the USSR Academy of Sciences was held in Leningrad, devoted to questions of physical and measurement acoustics. The program of the meeting was very extensive. Twenty-six papers and communications were heard, partly at plenary sessions and partly at meetings of the sections on piezoelectricity and electroacoustic apparatus and measurements. More than 115 people took part in the meeting.

In his opening address, the chairman of the Acoustics Commission, Corresponding Member of the USSR Academy of Sciences N. N. Andreev, noted that the meeting was opening “with the fresh memory of the untimely death of the President of the USSR Academy of Sciences, Sergei Ivanovich Vavilov, a widely known scientist, public figure, and organizer of science. He reminded the participants of the meeting that Sergei Ivanovich attached great importance to acoustics and contributed considerably to its development. “It cannot be forgotten,” said N. N. Andreev, “that Sergei Ivanovich was the first president under whom acoustics began to develop widely in our country.” The meeting honored the memory of the outstanding physicist Sergei Ivanovich Vavilov by rising.

The first plenary session was devoted mainly to reports in the field of physical acoustics.

Corresponding Member of the Ukrainian Academy of Sciences A. A. Kharkevich, in his report “Spectra and Analysis,” gave a brief account of the contents of a book of the same title prepared for publication. The author noted the need to unify the theory of spectra and the theory of physical methods of harmonic analysis in connection with the development of a number of special branches of technology, and dwelt on some selected questions, namely: in the theory of spectra—on the relation between the duration of a signal and the width of its spectrum; in the theory of analysis—on resonance under a varying frequency of the driving force and on the analysis of single pulses by means of resonators.

The speaker noted, among other things, the erroneousness of the usual interpretation of a frequency-modulation altimeter, pointing out that the amplitude spectra of the transmitted and reflected signals are identical and that therefore it is impossible to obtain a smoothly varying difference frequency.

In the discussion of the report, G. S. Gorelik and others did not agree with the speaker’s opinion on the common interpretation of the principle of operation of an altimeter, noting the slowness of the frequency change in comparison with the change in the frequency deviation. A. A. Kharkevich objected,

indicating that, although there is no gross contradiction between the existing theory and experiment, the theory is nevertheless fundamentally incorrect. The radio altimeter gives not a continuous but a stepwise reading of frequency, which cannot be explained by the elementary theory.

In the paper “On the analysis of oscillations,” A. V. Rimsky-Korsakov raised the question of finding the characteristics of the spectral composition of nonperiodic oscillatory processes.

The speaker proposed introducing the concept of the so-called “current spectrum,” which is a spectral function of the oscillatory process from the beginning of the process up to the moment of observation, pointing out that in theory such a definition of the spectral function makes it possible to establish full correspondence between the Fourier, Bromwich, and Duhamel integrals. In practice, such a condition determines the limit of applicability of analyzers of various systems for the purposes of analyzing nonperiodic processes; thus, for example, an actual resonant system with losses registers only some fraction of the current spectrum corresponding to a certain part of the duration of the entire process. This registration time depends on the character of the process itself and can only be approximately estimated for a given type of resonant analyzer. The registration time of the spectrum can be made definite if a diffraction grating is used for the analysis. Ideally, this should be a relay grating with a number of periods proportional to the registration time. In doing so, however, other errors arise, associated with resolving power.

In the paper “Visualization of spatially modulated waves,” S. N. Rzhevkin set forth a theory of various types of spatially modulated waves, indicating an analogy between a spatially modulating wave in a tube with a rectangular cross-section and waves emitted by a plane diffraction grating. Results were then presented for the calculation of the distribution of the energy flux in the field of a diffraction grating, indicating the presence of focusing of energy in a number of planes spaced at certain distances from the grating, with an increase in intensity in these planes occurring successively on lines opposite first the impermeable and then the permeable parts of the grating. In the study of ultrasonic waves carried out as early as 1939 by the author and S. I. Kremer, photographic images of similar pictures of a diffraction grating were obtained. The positions of these images prove to be in agreement with the developed theory.

The report by L. M. Brekhovskikh, “On the theory of total internal reflection,” contained results obtained by the author concerning the displacement of sound (or light) beams under total internal reflection. In particular, the influence of ray displacement on the total internal reflection of a spherical wave was considered. It was explained that in this case a concentration of sound energy takes place along a certain surface (a caustic) in a spherical wave. Some new data were also reported concerning the reflection of sound waves from an inhomogeneous layer at grazing incidence. The speaker showed that in this case substantial additions must be made to the familiar picture of ray refraction.

In the discussion N. N. Andreev noted that the canonical treatment of the phenomena of refraction and reflection had always aroused doubts, but it was supported by tradition, and only now have works appeared that give a physically correct explanation of these phenomena.

L. A. Chernov spoke on “the scattering of sound by fluctuations,” pointing out that in pure media no anisotropic fluctuations of scattering occur, whereas isobaric fluctuations are the cause of sound scattering.

The speaker also presented results of the calculation of sound scattering by iso-

baric fluctuations of concentration in solutions near the critical mixing temperature of liquids and disperse systems. Scattering in pure media, solutions, and heavy suspensions is very similar: in all cases the scattered sound is a superposition of isotropic and dipole radiation. The intensity of the scattered sound is proportional to the fourth power of the wavelength. Deviations from Rayleigh’s law of scattering occur in pure media near the critical state, where they are due to the statistical dependence of the fluctuations. In heavy suspensions the deviations are due to relaxation processes. A particularly simple (isotropic) character is exhibited by scattering in the case where the dispersed substance and the filler differ strongly only in compressibility.

The physicochemical and biological action of ultrasounds was described by I. E. Elpiner, who gave a summary of the results of an experimental study of the effect of ultrasound on living substances. The speaker noted that ultrasonic waves cause instant rupturing of animal and plant cells and of microorganisms. In this process biologically active substances—enzymes, toxins, etc.—are extracted from the cells into the surrounding medium. However, under more prolonged sonication these substances themselves are also destroyed. Under the influence of ultrasonic waves large protein compounds undergo degradation and break down into individual amino acids entering into the composition of the protein molecule. Investigations (together with I. V. Zborskii) showed that amino acids of cyclic structure are predominantly destroyed. Transitions of some amino acids into others were observed (when histidine was sonicated in solution, aspartic acid was found), as well as depolymerization of nucleic acids, which play an important role in the vital activity of cells. The disintegration of porphyrin nuclei was discovered (together with Blumenfeld and Krasovitskii), as a result of which bimerubin, biliverdin, and individual pyrroles appeared in the sonicated solution. In the speaker’s opinion, ultrasonic waves cause oxidative and reductive processes; in the presence of oxygen, oxidation processes predominate, while when the sonicated solution is saturated with hydrogen, on the contrary, reduction processes predominate. The chemical action of ultrasonic waves is apparently due to the appearance in the solution of valence-unsaturated free radicals and atomic hydrogen—products of the splitting of water molecules. Splitting is caused by ionization of molecules by free charges arising in cavitated cavities.

At the meeting of the section on electroacoustic apparatus, nine papers and communications were heard.

M. A. Sapozhkov touched on questions of the “methodology for measuring the parameters and characteristics of telephones, microphones, and laryngophones,” pointing out that the existing assessment of the nonuniformity of frequency characteristics relative to a horizontal straight line is not indicative and that it is advisable to assess the frequency characteristic by its deviation from the tendency (a quadratic approximation of the first or second order). The speaker also proposed introducing the concept of an “optimal tendency,” to which one should strive to approximate the frequency characteristic, and noted that the usual method of assessing the linearity of an electroacoustic system from the amplitude characteristic is unsuitable and that it is better to determine it from the dependence of the slope of the amplitude characteristic on the input effect. The speaker recommended determining the mean values of “efficiency” on a logarithmic scale along both coordinate axes (frequency and intensity). In the discussion it was pointed out that the methodology of measurements must be regulated and that it was noted that the proposed method of recording amplitude characteristics was successful. The disputability of the proposal to determine the nonuniformity of the frequency characteristic relative to the tendency was noted.

I. M. Palkovskii spoke “on the application of an instrument with a thermistor for measuring the level of noises and speech,” noting that, since the human ear reacts to the effective value of sound pressure, it is advisable to have an indicator measuring the averaged magnitude of sound pressure over a comparatively long interval of time (on the order of several seconds).

The instrument developed by the speaker, using a thermistor, has a sufficiently large time constant, which made it possible to obtain frequency characteristics (with a band-pass filter) of certain noises and speech averaged over time. Such characteristics, in the speaker’s opinion, characterize the frequency composition of sound more reliably than those currently used.

In the report by A. G. Muratov, “On a universal measuring setup for testing electroacoustic apparatus,” an electrical circuit was described that makes it possible to obtain a direct reading of the average output of a loudspeaker in bars and of the irregularity of its frequency characteristic in decibels on the scale of a pointer indicator, and also makes possible observation of the frequency characteristics of the electroacoustic apparatus under test on the screen of a cathode-ray tube.

The apparatus makes it possible directly to measure the total electrical resistance, the average output, and the irregularity of the frequency characteristic within a specified range, as well as the frequencies of mechanical resonance of the moving system of the loudspeaker and the magnitude of the acoustic coefficient of nonlinear distortion.

I. N. Stoiko demonstrated the first experimental model of a noise meter powered from an alternating-current mains supply, having a dynamic range of 35–140 db and a frequency range of 100–10,000 cps, developed by the Central Scientific Research Laboratory of Piezoengineering.

The report by Yu. M. Sukharevskii contained a description of a method developed by him together with his collaborators “for measuring the modulus of elasticity and decrement of attenuation of materials.” This method uses the so-called “electromechanical Q-meter,” which makes it possible to determine these quantities for the material under test in the frequency range up to 100 kcps by electrical measurement of the capacitance and losses of a piezoquartz plate that is mechanically in contact with a specimen of the material under test or with a mechanical system including this specimen. From the position of the maximum electrical resistance of the piezoquartz plate and the magnitude of the minimum dip during tuning, the resonance frequency of the entire system, associated with the modulus of elasticity of the material under test, and the loss resistance, associated with the attenuation decrement of the material \((\delta)\), are determined.

At a frequency of 20 kcps, aluminum \((\delta \simeq 10^{-4})\), plexiglas \((\delta \simeq 5 \cdot 10^{-2},\) Young’s modulus \(7 \cdot 10^{10})\), and other materials were measured. Measurements were also made of the shear modulus \(G\) and of the attenuation decrement for the shear modulus \(\delta_q\) at different temperatures and pressures, and it was found that, as the temperature is lowered, \(G\) for rubber increases sharply, whereas a change in pressure within the range up to 10 atm does not produce a noticeable effect.

A. N. Poloskin spoke about the “method of automatic factory inspection of microphones and telephones,” developed by the Department of General Physics of Molotov State University on the basis of achievements in the technology of automatic microwave radiotelegraphy.

As the measuring process, giving the required quick and accurate answer concerning the suitability of the product under test, a comparison with a standard is carried out: the amplitude of the electrical signal at the output of an electroacoustic path, which includes the specimen under test, is compared with the amplitude of a standard signal applied to the input of this electroacoustic path. The comparison of the signals of the used

the sample and the standard is carried out in time, i.e., the signals from the input and output of the electroacoustic path continuously and alternately arrive, one after the other, at the same test amplification part of the instrument. The counting relay of the instrument gives the output on the reading scale of the tested sample after a certain number of individual readings, each of duration on the order of 0.2 sec.

I. I. Slavin read a report, “On Objective Noise Meters with a Scale of Natural Loudness,” in which he pointed out that the decibel scale existing in noise meters does not reflect the essence of the subjective sensation of loudness and is incomprehensible to the overwhelming majority of nonspecialist acousticians. The speaker proposed taking a level of 40 phons as the unit of the loudness scale. The report also showed that the presence in ordinary noise meters of only three frequency characteristics leads to substantial errors in measurements. The speaker described noise meters developed at the laboratory for the struggle against industrial noise of the Leningrad Institute of Labor Protection of the All-Union Central Council of Trade Unions, with a scale of natural loudness and frequency characteristics changing every 10 decibels. I. I. Slavin demonstrated a sample of the noise meter developed by LIOT, distinguished from foreign samples by its compactness and greater accuracy.

In the discussion of the report, Yu. S. Bykov questioned the advisability of changing over to the new loudness scale. However, the majority of those who spoke pointed to the desirability of adding such a scale, although it does not fully reflect the properties of hearing.

In the report “Problems of Measuring the Microphone Effect of Carbon Powders,” V. N. Fedorovich set forth original methods of measuring the parameters of carbon powder that directly determine its operation in microphones: electrical resistance, the coefficient of modulation of mechanical resistance, and its own noise. The author pointed to the dependence of the mechanical resistance of the powder on the amplitude of oscillations and to the influence of this dependence on the properties of microphones (a lower bend in the amplitude characteristic, a reduction in the resonant frequency, and an increase in sensitivity with increasing sound pressure). Preliminary studies of these parameters established their connection with the structural state of the powder, the degree of moisture content, aging, and other factors.

Thus, the possibility was shown of a direct experimental study of the factors determining the operation of carbon powder as an electroacoustic transducer. The report gave rise to a very animated discussion.

The section on piezoelectricity heard five reports. P. A. Ananyev spoke about the “piezoelectric measuring microphone” he had developed, with a crystal of monoammonium phosphate “PIEM-3” (piezoelectric measuring standard microphone—model 3). The microphone has an absolute sensitivity of 24 microvolts per bar with an unevenness in the range of 30–16,000 Hz in the axial direction of ±1 db. The diameter of the receiving part is 15 mm and therefore up to frequencies of 8–10 kHz the microphone can be regarded as a point receiver. The temperature range of measurements is from −30° to +50°.

The original so-called “vacuum” capsule of the microphone is moistureproof and ensures high constant insulation and stability of operation. The microphone has control plates which make it possible, electrically, with the aid of the “piezoelectric coupling coefficient,” to determine the sensitivity and unevenness of the microphone’s frequency response over the entire frequency range. Nonlinear distortions are practically absent, and one can measure sound pressures on the order of several millionths of a bar.

The speaker pointed out that the microphone “PIEM-3” can be regarded as a primary standard, the absolute sensitivity of which can be

Chronicle

is determined directly from the calculation of the piezoelectric constants of the crystal. A drawback of the developed microphone is its low sensitivity (0.3–0.5 bar) and the necessity of mounting the microphone together with an amplifier stage.

The report by V. A. Krasilnikov contained the results of “measurements of the Young’s moduli of Rochelle-salt bars in the dynamic regime,” carried out in the acoustics laboratory of NIIF MGU by the author jointly with L. A. Shuvalov and I. A. Sheinov by the compound-rod method. New data were obtained on the propagation velocity of longitudinal waves and on the values of the elastic constants of Rochelle salt. These data will be published in the near future.

In a communication concerning the “acoustic method of measuring vibrations,” V. N. Fedorovich described a method of measuring the vibration velocity by measuring with an acoustic probe the sound pressure caused by this vibration. In order to prevent the transmission of vibrations to the microphone, and also to ensure close contact with the surface of the vibrating body, a flexible tip is fitted onto the tube of the probe. An important advantage of this method of measurement is the possibility of using, for the determination of the sensitivity of an acoustic probe intended for measuring vibration, calibration by sound pressure.

The author presented frequency characteristics of the sensitivity of an acoustic probe-vibrometer that agree well with one another, one of which was obtained directly by a generator of mechanical oscillations, while the second was calculated on the basis of calibration of the acoustic probe by sound pressure. The frequency characteristic of the sensitivity of the “acoustic vibrometer” is found to be sufficiently uniform over a wide frequency range from 100 to 5000 cps.

In the report by M. A. Chelnysheva, “On the domain structure of Rochelle-salt crystals,” accompanied by a demonstration of a microfilm shot by the speaker, it was shown that Rochelle-salt crystals, when cooled below 24°C, change their symmetry, passing from rhombic to monoclinic, and that in this process the single crystal becomes a polysynthetic twin, the components of which are domains of two orientations. Under the influence of a concentrated load (pressure with the aid of a ball), either elastic or residual mechanical twins arise in Rochelle-salt crystals, depending on the magnitude of the applied force. The application of a constant electric field to a twinned crystal transforms it into a single crystal, the orientation of which depends on the sign of the field and corresponds, for different field directions, to the orientation of either one or the other component of the twin.

V. P. Konstantinova demonstrated “oscillations of plates made from piezoelectric textures” excited by a sound generator. At the resonance frequencies of the oscillating plate, distinct Chladni figures were formed from sand poured onto plates of quartz sand.

The report presented a method for preparing textured piezoelectric plates from Rochelle salt.

“The influence of ultrasound on the process of crystallization” was studied by A. P. Kapustin in numerous experiments. The speaker showed that, with the aid of ultrasound in a supercooled melt, a large number of new crystallization centers arise; upon reaching the “lower threshold” of ultrasound intensity, the rate of advance of the crystallization front increases hundreds of times.

An increase in intensity beyond the “upper threshold” does not affect any further increase in the crystallization rate. A material crystallizing in an ultrasonic field has a finer-grained structure and greater crushing strength.

At the beginning of the plenary session on February 3, G. A. Goldberg spoke, delivering a communication “On the works of the outstanding Soviet inven—

...the article by Doctor of Arts Evgenii Aleksandrovich Sholpo, who died in 1950. Sholpo’s idea, to the development of which he devoted 25 years of his life, consists in devising apparatus and a method that allow a musician to create a phonogram directly by synthetic means, without performing the music before a microphone. The synthetic recording of music offers an enormous, fundamentally unlimited, expansion of the means of musical expression, but at first it requires overcoming a number of serious difficulties. E. A. Sholpo’s apparatus—the “variophone”—is a camera for optical recording, within the limits of the sound track of a motion-picture film, of several narrowed tracks according to the number of voices in the musical work. On the path of the beam from a constant light source two diaphragms are placed. One of them, having the form of a screen with a triangular cutout, when moved causes a change in the intensity of the beam and is used for volume control. The other is made in the form of a toothed disk; its continuous rotation causes a periodic change in the length of the stroke and, accordingly, the transverse recording of sound vibrations. The frequency of the vibrations is determined by the speed of rotation of the disk, and the form of the vibrations by the pattern of the teeth. The speed of the film during recording is considerably (10–20 times) less than during reproduction, and therefore the operator, by slowly manipulating the various knobs of the apparatus, is able smoothly and subtly to control the properties of the vibrations being recorded, to interrupt and resume the recording instantly at any place. The apparatus is supplied with a number of additional devices: for modulating amplitude (the rise and decay of sound), frequency (vibration), etc.

E. A. Sholpo also carried out interesting work on the objective analysis of musical performance. In music one may distinguish macrostructure, created by the composer, and microstructure, created by the performer. Whereas the theory of composition has a long history of development and has been worked out in detail, the theory of performance is in its very infancy. The synthetic recording of music, being a recording of microstructure, is impossible without knowledge of the objective techniques of musical performance, and the lack of such knowledge substantially limits the possibility of creating artistic recordings at the present time. What is important is that the very method of synthetic recording is an invaluable means for the objective study of performance.

Of special interest is the synthetic recording made by E. A. Sholpo with the aid of the “variophone,” reconstructing a performance according to the data of analysis. This recording convincingly demonstrates the possibility of creating expressive music by synthetic means. The development of the work on the physical analysis of music begun by E. A. Sholpo will make it possible to fill a blank page in the science of music and may enrich music with new technical means of performance.

S. Ya. Lifshits considered the question of “standardizing measurements of the auditory threshold.” In connection with the author’s investigations of the discreteness of sensations, the phenomenon of accumulation, and others, he proposed standardizing the duration of the sound impulse, the interval between impulses, and certain other quantities in impulse measurements of the auditory threshold.

Proceeding from the phenomenon of fluctuation of the auditory threshold, the author proposed taking as the auditory threshold the level at which 20% of the impulses cease to be noticed.

In another communication S. Ya. Lifshits spoke about his investigations of the “discreteness of sensations in the field of hearing and of touch,” noting that the discreteness of sensations is determined by a second integral law of hearing, the phenomenon of accumulation, and fluctuations of the auditory threshold. At first these phenomena were attributed only to hearing, but further investigations showed that they also occur for touch and vision.

Applying to quantitative measurements the fluctuation of the threshold of sensitivity...

…using, in particular, the statistical method, one can obtain the value of the number of elements of sensation necessary for threshold sensory perception. This number proved to be equal to eight for the general senses.

In the report “Sound Transitional Layers,” B. D. Tartakovskii presented the results of a theoretical and experimental investigation of the propagation of plane waves, in particular sound waves, through plane-inhomogeneous media, and gave a calculation of devices ensuring the complete transmission of sound waves across the boundaries of media with different acoustic parameters.

Such devices were proposed by the author, for example, for increasing the effectiveness of ultrasonic flaw detectors, ultrasonic lenses, and other ultrasonic technical instruments.

In his report “On the Influence of Atmospheric Turbulence on the Propagation of Sound,” V. A. Krasilnikov discussed his new experiments on the study of internal structure in atmospheric turbulence and on determining the dependence on distance of phase fluctuations and fluctuations in signal level. The speaker noted that the observed experimental regularities are well explained by the statistical theory of turbulence developed by Kolmogorov and Obukhov.

In the report by G. S. Gorelik, “Some Problems of Statistical Acoustics,” the importance and fruitfulness of a statistical treatment of oscillatory processes, in particular acoustic ones, were demonstrated, both in their investigation and for the purpose of creating new high-efficiency devices. The speaker reported on a method for measuring very small phase-angle values of low-frequency oscillations, developed at the Physico-Technical Institute of Gorky University by I. L. Bernshtein, and pointed to the possibility and expediency of carrying out analogous experiments in acoustics.

The report provoked a lively discussion.

E. L. Feinberg noted that in an ordinary sound process, in essence, we are dealing with periodic features, and decomposition into a Fourier integral is, one might say, an artificial operation justified only by its use for the analysis of linear electrical systems, whereas the width of the frequency band of an ordinary sound is of the same order as the mean frequency of the band, and therefore a statistical approach in acoustics is almost necessary. There is a direct connection between frequency and statistical analysis, and from the correlation function one can obtain the spectral composition of oscillations. At the same time, as existing experience already shows, a systematic treatment, as the more adequate one, in a number of cases leads more rapidly to a result.

I. D. Myasnikov emphasized the importance of developing a statistical method of acoustic measurements and pointed out that, for measuring a weak signal that cannot be perceived because of the amplifier’s own noise, it is expedient to apply acoustic modulation.

I. E. Garon reported on the development of an instrument for measuring small acoustic noises by the modulation method.

S. G. Gershman, in a communication “Statistical Characteristics of Processes in Architectural Acoustics,” presented the results he had obtained from experimental determination of the autocorrelation coefficients of stationary (in the probabilistic sense) noises, and discussed the method he had developed for direct measurement of the acoustic ratio (the ratio of the intensity of direct sound to the intensity of sound reflected in a room) by measuring the correlation coefficient.

N. N. Andreev, summing up the discussion on questions of statistical acoustics, noted that the methods of statistical physics are still little used in acoustics and that it is very important to expand the field of their application, since they make it possible to carry out valuable experimen-

observations and obtain results important for engineering practice.

N. N. Andreev expressed the wish that future courses in acoustics include the results of work by Soviet acousticians in the field of statistical phenomena.

The meeting adopted a “resolution on questions of the creation of electroacoustic apparatus,” in which it noted the relevance of the development of the measuring microphone “PIZM-3,” carried out at Scientific Research Institute 100 of the Ministry of Communications, and the need for its investigation at the All-Union Scientific Research Institute of Metrology; it also noted the need for series factory production if the results of these tests prove positive. The meeting emphasized that, despite the increased demand for measuring electroacoustic apparatus on the part of research institutions and industrial enterprises, its production and release have still not been organized. Since certain research institutions have developed noise meters of various kinds, it is necessary that comparative tests of them be carried out at TsIRTA of the Ministry of Communications, with the aim of selecting the best model for introduction into production.

The meeting approved temporary technical requirements for noise meters for mass use:

a) an effective frequency passband from 60 to 8–10 thousand hertz;

b) simultaneous measurement of the frequency characteristic with shifting of the switch for the measured levels (in steps of up to 10 db);

c) a sound-pressure range from 35 to 150 db;

and pointed out the desirability of acoustic calibration, the inclusion of filters, and the presence in noise meters, alongside the decibel scale, of a loudness scale.

B. D. Tartakovskii