355. If ordinary Leyden jars are employed for producing the oscillations, it is found that the wave-length is still far too great, being from 150-300 feet with ordinary sized jars. A better arrangement for producing oscillations is that first employed by Hertz, and known as a Hertz oscillator. This consists simply of two brass spheres attached to the ends of two rods which are attached to the secondary of an induction coil, the spark passing between the two brass knobs. In order to give capacity to this, metal spheres or plates are sometimes attached to the other ends of the rods. 356. The following formula is given by Lodge1 for the calculation of L in the formula for the frequency of the oscillation : where /= length of the entire rod portion of the oscillator in cases where the oscillator consists of metal spheres mounted on the ends of brass rods, to the other ends of which are attached small plates to give capacity, and d = diameter of the rod. In measuring, it is best to include the knobs and spark-gap, and about a quarter the diameter of each end plate. The capacity K is practically half the static capacity of the plate or sphere at either end of the oscillator, since the two capacities are really in series. Thus for two oppositely charged spheres of radius ⁄ and distance s from centre to centre— 357. For isolated bodies, the following values of È are given by Lodge : 358. The following example will illustrate the calculation, and applies to an oscillator consisting of two rods each 6 cm. long and 1 cm. diameter, to which knobs 2 cm. diameter and plates 8 cm. diameter have been attached (see Fig. 149.) The spark-gap is o'8 cm. Rate of vibration n = 300 million per sec. 359. A still more convenient form of oscillator, originally due to Professor Lodge, consists of two small brass spheres connected to the secondary of an induction coil, and arranged so as to spark across the diameter of a larger brass sphere placed between them (see Fig. 150). This form has also been adopted by Bose.1 The small spheres in his case were 0°3 cm. diameter, and the large one 0'78 cm. diameter. The distance from centre to centre of the small spheres was I'2 cm. The waves produced by this oscillator were 184 cm. long. It is advisable to hinge the supports of the small spheres so that their distance apart may be varied. FIG. 150. 360. Another form of oscillator due to Lodge consists of two metal spheres sparking to the interior of a copper cylinder; with the dimensions in Fig. 151 waves of 3" long are produced. 361. A form of oscillator used by Marconi, and described by Preece, consists of two solid brass spheres fixed in an oil-tight 1 See Pro. Roy. Soc., vol. lx. p. 167. 2 Electrician, vol. xxxix. p. 217. insulating case filled with vaseline oil, two smaller spheres placed close to the outside of the others are connected to the induction coil, the spark passing across the three gaps. With the large spheres, 4" diameter, waves 1'2 metres long were obtained. 362. The great difficulty with all these oscillators is that when the sparking surfaces get roughened, as they do with use, the discharge ceases to be oscillatory; the brass knobs should therefore be made easily removable, to admit of their being burnished up. It is also necessary in all quantitative experiments to completely enclose the oscillating apparatus-battery, coil, and wires-inside a copper-lined box, leaving only the spark-gap outside, since oscillations may escape from other parts of the apparatus. For this reason the form of apparatus designed by Bose 1 is recommended. 1 363. The oscillator is fixed to the outside of a box containing the coil and battery; a copper tube open at both ends is slipped over the oscillator so as to direct the path of the waves. 1 See Phil. Mag., vol. xliii., Jan., 1897; also see Fro. Roy. Soc., vol. lix. p. 160. The box containing the battery and coil is encased inside two metal cases—an inner one of iron, which forms a magnetic shield round it, and an outer one of copper, which prevents stray radiation from escaping. Also, instead of having the usual vibrating make and break on the coil, an ordinary breakcircuit key is employed, the condenser being connected to it as in the vibrating break; the object of this is to enable single flashes of radiation to be obtained instead of a continuous stream, thus saving the sparking surfaces. The key can be manipulated from the outside of the box by a string. APPARATUS FOR DETECTING ELECTRO-MAGNETIC WAVES. 364. If the electro-magnetic waves are produced by a Leyden jar discharging through a wire circuit, the presence of the waves may be detected by placing a second similar jar, provided with a discharging circuit of the same dimensions, in the vicinity of the first. On the first jar being discharged, the oscillations produced set up electro-magnetic waves which, on arriving at the second jar, induce discharges in it, which can be detected by the presence of a spark at the discharging knobs. This effect, however, is only possible when the jars and their discharging circuits are identical with one another, the second jar having to be "tuned into resonance with the first before it will act as a detector. 365. A much more sensitive form of detector, and one which is easier to work with, is that known as a coherer. The action of the coherer was first demonstrated by Branly,1 who showed that a tube containing metallic filings, and which had a very high resistance, became a tolerably good conductor when placed near a sparking coil. This phenomena has been utilized by Lodge, Bose, Marconi, and others, as a means of detecting electro-magnetic waves. A simply made form of coherer consists of a glass tube about eight inches long and half-inch diameter, filled with small clippings of copper. Passing into the clippings through corks at either end of the tube are copper wires, which connect the 1 Branly, Comptes Rendus, vol. cxi. p. 785; vol. cxii. p. 90. Ꮓ coherer in series with a galvanometer and battery. The tube in its normal state is a very bad conductor, due to the number of bad contacts at the copper clippings, and no deflection will be obtained on the galvanometer. If a sparking coil is now set a-going in the vicinity of the tube, the galvanometer will be observed to deflect, this being due to the sudden acquisition of conducting power by the tube of clippings. On slightly tapping the tube, it returns at once to its badly conducting condition. This action, according to Lodge, is due to the breaking down of the insulating layer of oxide between the clippings by the electrical surgings set up in them by the electromagnetic waves. 366. A much more satisfactory form of coherer is that devised by Professor Bose,' in which the copper clippings are ww www www 1 Brass FIG. 152. side, has a small one side, 2 cm. Into this groove replaced by steel springs. The construction Two brass plates make contact with the upper and lower spirals respectively, the upper brass plate being attached to a screw, by means of which it can be made to compress the spirals. The upper and lower brass plates are connected in series with a battery and sensitive galvanometer. If the spirals are loose, no deflection will be obtained on the galvanometer. The screw is now turned so as to compress the spirals until a small deflection is obtained; the coherer is now in a sensitive condition, and if exposed to electro-magnetic radiation the galvanometer will deflect violently, indicating a great diminution of resistance at the spirals. In order to restore them 1 Phil. Mag., vol. xliii., January, 1897. |