Pressure waveforms of pulsed ultrasonic waves by which cavitation is produced in aerated water are observed by a miniature barium titanate microphone. A laminated nickel transducer with a simple vibration pickup attached on its back surface is used at its resonant frequency of 24 KC to radiate the pulsed ultrasonic waves. At the pulse width of 0. 3ms, only a few cavitations are produced, so that individual shock waves caused by the collapse of cavities are observed as sharp impulses of about 0. 5 μs width during the decreasing phase of pressure waves. At the intensified pulse of 1. 5 ms width, the pressure waveform is no longer sinusoidal owing to the production of many cavitaitons, but is turned into saw-toothed form with large positive peaks instead. When the ultrasonic pulse is built-up, the first cavitation occurs at the decreasing phase subsequent to the second pressure maximum. The production of cavitation continues over several successive periods at nearly the same pressure phase as the first one, and the intensity of cavitation becoming less violent at every other period. Thus the subharmonic component appears on the pressure waveform, and, simultaneously, on the vibration of the transducer, too. As the generation of remarkably distorted wave is localized near the center of the radiating surface where the violent cavitation occurs, the saw-toothed waveform fades out rapidly as the distance increases from the transducer, and the waveform at a distance becomes similar to the velocity waveform of the transducer.
The periodicity pitch sensation is predicted to be in close correlation with the envelope time pattern of the complex tone from the Schonten's experiment on "residue", together with the duplex theory proposed by Licklider. The pitch of the unharmonic tones was determined by the pitch matching method simultaneously with the measurement of the time pattern of the periodicity of the envelope of complex tones. The values of the pitch measured were found to coincide with the frequencies determined from the periods of the envelopes. The unharmonic tones examined in this experiment were produced by shifting the spectral components of harmonic tones by Δ Fc/s on the frequency scale. The signals before being shifted were approximately the periodic damped waves, and were also the "output" of the resonance circuit driven by the periodic pulses. If it is assumed that the time pattern of the envelope is reserved in the signal traveling through the nerves, the result of this experiment may be explained by the duplex theory. The next experiment concerned with the frequency response of the periodicity pitch sensation. The complex tones examined had the clearly determined time pattern corresponding to the fundamental frequency but had not the frequency components under 1200 c/s. The result shows that the percentage score of the pitch perception to the fundamental frequency decreases as the frequency is increased higher, e. g. 75% at 300 c/s, 50% at 500 c/s. This result suggests that a filtering action exists in the circuit determining the periodicity pitch with the time pattern in the ear. These experiments seem to be useful for designing the pitch extracter from the vowels and also for composing the hearing theory.
This paper describes the fundamental aspects of an electrical analog method for the theoretical analysis of acoustic filters and its experimental verification carried out by means of an electric simulator. First, the differential equations of one-dimensional acoustical system are transformed into a finite difference equation with respect to a sound pressure ρ or a volume velocity U, which gives the T or π elementary network of a direct analog type equivalent circuit. It is noticed that each of the two representations has a different physical meaning. Since truncation errors would result under such approximation, the applicability of these analog models are discussed by designating the usable frequency range and the tolerable error. Next, the three-dimensional delay networks and the compound T elementary networks corresponding to the one-dimensional T are derived for the rectangular and cylindrical coordinates. The reduction of dimension is also studied. Finally, the details of the simulator composed of 18 (8 variable, 10 fixed) sets of L, C and R are explained and the comparison between the data obtained by the numerical calculation, acoustical experiment and simulator operation on the various types of filter elements reveals the practical usefulness of this analog method for the parameter change study of acoustic filters.
Generally, the sound absorbing materials are wasted appreciably by high-speed air current and high temperature inside a muffler, and that they become the causes of many troubles when they are used in the muffler. One of the solutions for this is to make the mass of the sound absorbing materials exchangeable. It is described in this paper how effective the small mass of absorbing material is, as the result of the experiments on such mufflers. The attenuation effect by means of a muffler varies with kinds and density of absorbing materials, though a conspicuous effect is always expectable in high frequency region. The absorbing materials inside the muffler may be easily exchanged as it is small. In designing a muffler, it is advisable, to analyze the exhaust noise of the engine, and then to design the overall shape and structure of the muffler in conformity with the theory of acoustic filter together with the method described in this paper.
Not to speak of the wellknown ripple tank method, various methods for model experiment on room acoustics have been designed to meet the purpose of respective investigations. Here, we propose a convenient method for the study of steady state sound field. A Two dimentional model of a room, of for example 40×50×5 cm, covered with a thick glass plate is made, and the air in the model is excited by pure tone with a loud-speaker mounted in its corner. When the frequency of sound from the loud speaker is equal to the resonant frequency of the model, scattered cork dust on the model floor move intensively and make a figure formed of groups of plaits. The figure varies its shape depending on the frequency, that is to say the mode of vibration of the air in the model. By observing the cork dust figure, one can visualize the sound pressure distribution of standing waves in the model, in another words the position of pressure maxima and pressure nodal lines, just as reading a map. As regards to the particle velocity of the air, one can compare its magnitude and direction at each point in the model, from the height and the direction of cork dust plaits. In this paper, some fundamental natures of dust figure and one of the applications of this method are reported. Relations between fineness of cork dust and space of dust plaits, height of plaits and sound particle velocity measured in a one dimentional model. Using two dimentional models, pressure contours are measured and compared with dust figures in various cases of frequency and different shapes of models. In the last section of this paper, the deformation of the pressure distribution and the frequency shift of normal modes from the rectangular case, when some types of projection are planted on one side of the rectangular model are reported. Mounting several loud-speakers apart, separate observations on each one of the degenerated modes of vibration were made possible. The dust figure method was very convenient for instant discrimination of vibration in the model room.
It is well-known that the irradiation of intensive ultrasonics into polymer solution causes cavitation and decreases the degree of polymerization. The mechanism of the ultrasonic degradation can be explained as follows. In the contracting stage of a cavitating bubble, j structual units of a polymer molecule take a straight line-configuration by the suction force of the bubble. Slipping occurs between solvent molecules and the polymer. The latter undergoes tensile force arising from friction between them, which is greatest at the middle carbon bond of the j units. If it is greater than a critical value, the bond is unable to absorb the work done by it, but the bond is torn off; i. e. , the polymer degrades. Under such considerations many experimental facts about ultrasonic degradation can be explained, and also a formula for final degree of polymerization after long-time irradiation is given here.