Syntony or Tuning.— An important defect of wireless telegraphy in the simple form thus far described consists in the fact that but one message can be sent at one time, for the reason that if it is attempted to send two messages at once in the same vicinity, the signals will clash. Many inventors have striven to overcome this defect, among others Lodge, Marconi and Slahy. The plan followed by these workers has been that of employing a syntonic or tuning method; that is, a method by which the trans mitting and receiving circuits arc adjusted or attuned to a fundamental rate of electric oscil lations, to which rate of oscillations and no other the receiver so attuned will respond. This is done by taking advantage of the fact that the rate of oscillations, or the frequency of an electric circuit is proportional, as already noted, to the inductance, capacity and resist ance of the circuit. Hence by varying the ca pacity or inductance of the tuning circuits any desired rate of oscillation is obtainable. Tuned circuits are termed selective circuits, since by sending out a given rate of waves any desired tuned circuit within signaling distance may be selected as the receiving station.
Tuning also possesses the advantage that the benefits of resonance may be obtained by its use as follows: •It is known that a vertical wire grounded directly at its lower end is an excel lent radiator of electric waves, hut as it pos sesses very little capacity its oscillations are quickly damped as indicated in Fig. 11 and it is only the first few oscillations that are of sufficient strength to affect a receiver. When capacity and inductance are added to a circuit in certain proportions it may be made a persist ent vibrator (Fig. 12) and consequently a given amount of electrical energy expended at the transmitter in producing a succession of waves of more uniform amplitude will have a eumula jive or resonant effect upon a receiving circuit of equal capacity and inductance, and will even tually cause it to respond to the waves emitted by the transmitter; while an untuned receiving circuit containing a detector as sensitive as the first one, would probably not respond to those 'particular oscillations. In the case of tuned circuits, this is doubtless because the faintest oscillations, or electromotive forces, excited in the receiving circuit are resonantly amplified by the incoming waves of a selected frequency until they affect the detector.
As a rule, the 'spark-gap and the receiving 'apparatus of untuned circuits are connected directly to earth as indicated in Fig. 13, in which -A is the antenna; b b are the spark halls or knobs; s is the spark-gap; S is the secondary and P is the primary of an induction coil or transformer. In such an oscillating circuit (A, b, b), the oscillations are quickly damped. In the case of tuned circuits, or where, at least, the effects of resonance are desired, there is provided a closed oscillating circuit which is separated from the vertical wire by a trans former. Such an arrangement is outlined in
Fig. 14. The closed circuit in this case con sists of the spark-gap, condensers c c, and the primary of transformer T. In practice the con densers and the inductance coils are made ad justable, so that the capacity and the inductance may be varied at will. Best results arc obtained when the oscillating circuit and the vertical wire have equal fundamental oscillation periods ; or when the one is a multiple of the other. Oscillating circuits arranged as in Fig 13 are said to he tightly coupled. When arranged ar. an Fig. 14 they are said to be loosely coupled.
In addition to the fact that the oscillations in an ordinary vertical wire are quickly damped, there is the further fact to be considered that the oscillations set up by a spark-gap trans mitter are not continuous in the strict meaning of the word, since there is a perceptible time, while the oscillation circuit is being charged to the breaking down point of the air-gap, during which the oscillations tend to die out. Several inventors have striven to obtain a transmitter capable of supplying undamped or practically uniform oscillations to the oscillation circuit. Poulsen has developed some improvements in this direction by a modification of the Duddell singing arc (described in his British patent No. 21,629). (See TELEPHONY, WiamEss). Briefly, it consists of an arrangement iti which an arc lamp supplied by a direct current is placed in parallel with a capacity and an inductance. When the capacity and inductance are. suitably chosen, rapidly alternating currents are set up in this arc which produce a tone — hence the name singing arc. In Poulsen's modification of this arrangement the negative electrode of the arc is carbon, while the positive electrode is copper, the arc being enclosed in a box con taining hydrogen gas. The arc is shunted by a capacity and inductance and rapidly alternating currents are continuously set up which are thrown on to the vertical wire in the usual way. With the Duddell arrangement the frequency of oscillation was about 10,000 to 15,000 periods per second— too low for utilization in wireless telegraphy. A much higher frequency is ob tained by the Poulsen arrangement which is attributable to the atmosphere of hydrogen. To further the cooling of the electrodes (an essen tial first noted by Ulm Thomson, in United States Patent No. 500,630), the copper electrode is made in the shape of a hollow ring through which water is passed. Tests have demon strated that by this method of setting up un damped oscillations it is possible to signal to greater distances with much less electrical energy than is required with the spark-gap method.