Reception of Sound

Measurement of the Amplitude of a Vibrating Body.— W. H. Bragg has described a very simple and ingenious method of measuring the amplitude of a vibrating surface (e.g., of a diaphragm). A small mass supported by a spring is brought into contact with the vibrating surface and the fixed end of the spring displaced through a distance A until there is no longer "chattering" between the mass and the diaphragm. At the point where chattering ceases, the maximum acceleration of the diaphragm is just equal to that of the mass attached to the spring. Thus if a and n and A and N represent the maximum amplitude and the frequency of the diaphragm and the spring respectively, we must have = AN' or a= ANV Con sequently if N is very small compared with n, A will be very large compared with a, in this way yielding a large "magnifica tion" of the amplitude to be measured. As an example, if n= ,000 p.p.s. N =I p.p.s. and A=1 cm., then a= The method permits of the measurement of very small amplitudes, the chattering being observed electrically. It is important of course to observe that the measuring device is sufficiently light so as not to interfere with the motion of the vibrating surface whose amplitude is required. Measurements of the amplitude of vibration of a diaphragm have also been made by making it one of the reflecting surfaces of a Michelson interferometer (Webster, Nat. Acad. Sci., 5, p. 179, 1919). The displacement of the interference fringes, photographed in vibration as a wavy line, gives a measure of the amplitude in terms of the wave length of light. Such a diaphragm then forms a standard source of sound. Known amplitudes of vibration may also be obtained by applying measured alternating voltages to piezo-electric crystals, provided resonance-frequencies (usually extremely high) are avoided. Rankine has described (Proc. Phys. Soc., Aug. 1919, and Feb. 192o) a method of recording sound vibrations by means of variations in the intensity of a beam of light, the resulting film, of varying transparency, forming a convenient means for reproducing the original sound, using selenium or other photoelectric cells. When a wave-form has been recorded by any of the methods indicated above, it may be analysed into a Fourier series (see p. 7) which gives the frequencies and relative amplitudes of the tones of which the sound is composed.

The past ten or fifteen years mark a period of considerable progress in the technical application of the principles of sound. It is proposed in what follows to deal with the more outstanding of these applications apart from those already mentioned in the foregoing sections. The exigencies of war stimulated the develop ment of apparatus for detecting, identifying, and locating sounds at long ranges. Numerous forms of directional sound-receivers, sound-ranging and sound-signalling devices were realised and applied to urgent problems on land and sea. Since the war, the growth of radio-telephony and "broadcasting" has resulted in improved methods of reproducing sounds of audible frequency. The phenomena of piezo-electricity, discovered by Curie in 188o, have in the hands of Cady and others developed into a means of standardising mechanical and electrical frequencies over a range extending up to millions of vibrations per second. In this respect also improvements in tuning fork design have played an important part. A further application of piezo-elec tricity, initiated by Langevin and developed by Boyle, employs the supersonic oscillations of quartz in depth sounding at sea and in the echo detection of icebergs. At the other end of the frequency scale, Constantinesco has developed a system of power transmission through water-filled pipes, employing generators, motors, transformers and transmission lines closely analogous to the corresponding electrical devices. Such alter nating mechanical systems are capable of dealing with large amounts of power in the form of low frequency pressure-waves. The elimination of objectionable resonance from gramophones and loud-speakers represents a marked degree of progress in the development of apparatus for reproducing speech and music. The study of the characteristics of speech and hearing, notably the work of the staff of the Bell Telephone Laboratory U.S.A., has greatly increased our knowledge of these subjects. The results have proved of great value in the improvement of tele phone apparatus, microphones, electro-magnetic receivers, trans mission lines, etc., and in the design of all forms of apparatus for recording or reproducing speech and music. In this connection

also the introduction of the conception of acoustic impedance, as the analogue of electrical impedance, in dealing with complex acoustical systems has proved of great value. In a recent dis course before the Royal Aeronautical Society, Tucker has dealt with the problem of noise-reduction in aircraft, the deafening roar in the cabins of civil aircraft being a serious hindrance to commercial development. The general question of reduction of traffic noise is becoming increasingly insistent as an urgent prac tical problem still awaiting solution.

Reproduction of Sound.

The telephone (q.v.), the gramo phone (q.v.) and the radio loud-speaker are among the most familiar types of reproducers of the sounds of speech and music. Any practicable device of this character for reproducing sound must necessarily have its limitations, for the faithful reproduc tion of even the simplest sounds is very difficult. As we have seen (p. 32), however, the ear is sufficiently accommodating to ignore fairly large defects in reproduction. Thus a io% error in intensity, or possibly more than this, may pass unnoticed; the accuracy of intensity reproduction therefore need not be very great to conform with such a standard. Frequency reproduction, however, must be much more accurate to satisfy the ear, and it is indeed fortunate that the practical difficulties in this case are relatively small. Reproduction of sounds involves reception, and possibly transmission and amplification, as a preliminary. The process is well illustrated in the gramophone with either mechanical or electrical recording and reproduction. The record ing system involves the direct mechanical action of sound-waves on a diaphragm, lever system, and engraving stylus, or alter natively on a system employing a microphone and a recorder operated electro-magnetically. In reproduction the process takes place in the reverse order, some form of horn, cone or diaphragm being utilised to couple the vibrations of the sound box or electrical "pick up " with the external body of air. The present practice in design is to regard the complex mechanical system, e.g., of a gramophone, as analogous to a corresponding electrical system. In this respect the analogy with electrical filter circuits, as stated by G. W. Stewart, has proved an extremely fruitful one. S. T. Williams and A. Whitaker have both ap plied this principle with considerable success in designing gramo phone recorders and reproducers. By suitably choosing the inertia, stiffness and damping of the various elements of the sound-box and horn, a moderately good response-curve is obtained free from pronounced resonance within the speech-music range 150-4,000 p.p.sec. Electrical methods of recording and re producing have recently yielded even better results. The desirable features in gramophones and loud-speakers are dealt with in a number of papers by various authorities in Proc. Inst. Electrical Engineers, Nov. 1923. The practical consid erations are far too numerous to mention here but the essence of the problem of faithful reproduction lies in the avoidance of predominant resonances in the electrical and mechanical sys tems. This result may be achieved in some degree (a) by arrang ing that the natural frequencies of all elements of the system are far removed from any frequency it is desired to reproduce, (b) by the use of heavy damping, or (c) by making use of multiple resonance, i.e., arranging the various resonance peaks to overlap in such a way as to give a fairly uniform response. (See Rothwell, Nature, Feb. 24, 1923, and Porter, Phys. Soc., Feb. Mention has already been made of the recording of sounds on a film by means of light of variable intensity (Rankine, loc. cit p. 33). When a beam of light, after passing through such a film moving at the recording rate, falls on a sensitive cell (selenium, thalofide or photoelectric) the original sounds may be reproduced in a telephone or loud-speaker connected in circuit with the cell and a suitable amplifier. "Speaking films" have been produced successfully on this principle, the speech record being made on the edge of the film simultaneously with the cinema pictures. (See SELENIUM CELL.) Numerous other optical methods have been employed, with more or less success, to modulate a beam of light as a means of sound-reproduction, e.g., use has been made of the Kerr electro-optical and the Faraday magneto-optical effects, and of the luminous electric discharge in rarefied gases.