July 7, 1923] 



NA TURE 



1 1 



Studies from a Wireless Laboratory.^ 



By Prof. W. H. Eccles, F.R.S. 



^T^HE studies pursued in a wireless laboratory are 

 J- mainly of two kinds : fiist^ those directed to the 

 solution of problems that have arisen in the develop- 

 m'ent or use of practical apparatus^ and^ secondly^ those 

 with which we are here concerned^ aiming at the appli- 

 cation of novel principles or novel physical phenomena 

 to the invention of new methods or apparatus-. Little 

 will be said of the methods of wireless communication 

 as they exist to-day; on the contrary, our attention 

 will be devoted to some possibilities of wireless tele- 

 graphy- — possibilities tested in the laboratory but not 

 yet tried on the large scale. In other words, no attempt 

 will be made to give a record of technical progress 

 accomplished to date but, rather, to discuss wireless 

 communication as it may be. 



The new methods to be first described are based upon 

 the phenomena, not yet fully known in detail, which 

 occur when one vibrating body is caused to influence 

 the vibrations of another. Consider the case of a 

 simple pendulum consisting of a weight tied to the 

 lower end of a string the upper end of which is held in 

 the hand, and suppose it is of such a length that it 

 would vibrate freely to and fro in a period of two 

 seconds, when the hand is held still. Then it is easily 



DYING OSCILLATIONS. 

 Fig. I. — Dying oscillations. 



seen that on moving the hand horizontally to and fro 

 with a complete period of, say, one second, the pendulum 

 will follow the hand and likewise vibrate with a period 

 of one second. Similarly, when the hand vibrates with 

 a period of, say, three seconds the pendulum will again 

 follow and take the new period. This experiment is 

 very familiar and is known to students of mechanics as 

 an example of the subject of " forced vibration." 



A pendulum forced in this manner may be said to 

 vibrate " in time with " the hand, but the experiment 

 shows that it is not " in step with " the hand. It 

 would not be correct to say that it is " in tune with " 

 the hand, since this term is reserved — in electrical 

 physics at any rate — to indicate that the natural period 

 of the free and unpropelled pendulum is the same as the 

 period of vibration of the hand. We may, however, 

 express the state of affairs by saying that the pendulum 

 is forced into accord with the hand and that it is then 

 in the " accordant state." A simple example of this 

 relationship between two alternating movements is seen 

 when a dog, for example, is walking along the road ; 

 his hind legs are in time l)ut not in step with his fore 

 legs. 



The vibrations of a simple pendulum left free to 

 vibrate with its own period gradually die down as indi- 



' Substance of a discourse delivered at the Royal Institution, Friday, 

 April 13. 



cated in Fig. i. The vibration is a dying oscillation, 

 and in such a case the theory of the forced vibrations is 

 easily understood. In a modern wireless laboratory, 

 however, we have to deal with growing and sustained 

 vibrations as in Fig. 2, and in such cases the theory of 

 the accordant state is rather different. This is to be 

 expected — for it is like comparing a living thing to a 

 dying one. Usually the vibrations are sustained by the 

 aid of the triode valves so well known, and the rates of 

 vibration are very high. In order to lead up to an 

 understanding of the accordant state at these high 

 frequencies it is best to study low frequencies first. 

 For the study of vibrations slow enough to be followed 



SUSTAINED OSCILLATIONS 



Fig. 2. — Growing and sustained oscillations. 



by the eye a new type of oscillator has been designed 

 and constructed and is here exhibited for the first time. 

 Fig. 3 is a diagrammatic plan of the apparatus. The 

 horizontal magnet has a horizontal ebonite rod fixed to 

 it at right angles and the whole is suspended from a 

 vertical torsion wire passing through the centre of 

 gravity. The poles of the magnet confront two hori- 

 zontal solenoidal coils connected in series with each 

 other and with a battery and diode valve, that is, a 

 thermionic valve of the type invented by Prof. Fleming 

 in 1904 and containing only two electrodes, namely, 

 a filament and a plate. Such a valve possesses the 



DIODE 

 OSCILLATOR 



Fig. 3. — Diode-sustained torsion pendulum (in plan). 



property that the electron current across the vacuum is 

 sensitive to outside electrical influences if the electrodes 

 have suitable relative positions — an ebonite rod charged 

 by rubbing causes a diminution of the electron current 

 when it approaches the diode and allows the current to 

 increase again when it recedes. The action of this 

 diode-sustained pendulum is now easily explained by 

 supposing it swinging, and noticing that the ebonite rod 

 as it moves to and from the diode causes an alternation 

 of magnitude of the currents in and magnetic fields of 

 the coils, which is automatically in correct time relation 



NO. 2801, VOL. I 12] 



