Last September Mr. G. Marconi, the inventor of a practical system of wireless telegraph, brought over from England three sets of his apparatus for the use of the New York Herald in reporting the international yacht races.
The Navy Department detailed four officers to observe the working of the wireless telegraph during the races, and then, the reports of these officers being favorable to it, accepted Mr. Marconi's offer to set up his instruments on board of ships of the navy and there demonstrate their seagoing qualities and the adaptability of his system to naval signaling.
The three sets were installed on the flagship New York, the Massachusetts, and at the Navesink Highlands Light Station; and after a series of tests lasting over a week, three days of which the ships were at sea off Sandy Hook, the torpedo-boat Porter was fitted out with the apparatus from the New York, and spent a day steaming about the Massachusetts and off the Highlands for a further test.
Mr. Marconi then returned with his apparatus to England.
A board of three officers investigated the system during the navy tests, and in their report to the Chief of the Bureau of Equipment recommended that it be given a trial in the navy. During these tests the instruments were open to the inspection of the officers, except certain parts which were never dismounted; and their workings were explained in a general way, but the exact dimensions of the parts were not divulged.
The coherer, the principle of which was discovered some twenty years ago, is the only electrical instrument or device contained in the apparatus that is at all new.
Marconi's coherer, Fig. 1, is a glass tube about an inch and a half long, with a bore about one-tenth inch, into which are fitted two silver plugs, with leading wires attached. The space of about one-twentieth inch between the inner ends of the plugs is partly filled with a mixture of nickel and silver filings. The tube, after being exhausted to about one-thousandth of an atmosphere, is sealed. As shown in the figure, the coherer is seized to a glass rod for convenience in securing it in place in the receiver.
In their normal condition the "imperfect electrical contact" of these filings offers an infinite, or nearly infinite, resistance to a current of low electromotive force, but when the filings are "cohered" their resistance falls to a few hundred ohms, and so remains until they are "decohered," when they instantly resume their non-conducting state.
This coherence is brought about by certain phenomena, which are attendant upon the production of an electric spark and which act through space; the "decohering" is done by mechanically shaking or tapping the coherer.
In the Marconi system a series of sparks produced at the sending station "cohere" the coherer at the distant receiving station, which closes a local circuit containing the coherer, a single cell and a telegraph relay, which closes another circuit containing the tapper and the registering instrument. The instant the sparking ceases, the coherer is "decohered" by the tapper, and both local circuits are broken. Thus the dots and dashes of the telegraph code made in the spark-producing circuit at the sending station are reproduced as short and long closings of the local circuits at the receiving stations, and are registered as dots and dashes on the tape.
This transmission seems to be instantaneous.
Marconi discovered that the distance of transmission was increased enormously by using vertical insulated conductors, both for sending and for receiving; in sending, the vertical wire is connected to one pole of the spark-producing instrument and an earth wire is connected to its other pole; at the receiver, the vertical wire is connected to one leading wire of the coherer, while its other leading wire is connected to earth. This arrangement, shown in Fig. 2, has transmitted signals 25 miles.
Later it was discovered that by replacing the coherer by the primary of a transformer coil and putting the secondary of this coil in series with the coherer, the distance of transmission was again increased manifold.
The Marconi apparatus of to-day is illustrated in Fig. 3, and is diagrammatically represented in Fig. 4, which is reproduced by the permission of the Chief of the Bureau of Equipment from the diagram contained in the report of the Marconi board and which Mr. Marconi pronounced to be correct.
Referring to this diagram:
A is a sheet-iron box, 1/20-inch thick, about 24 inches long, 8 inches wide, and 8 inches high, containing the transformer MM2, coherer j, tapper p, relay n, relay cell g, and battery r, for the tapper and register circuits. There are usually two of these receiver boxes at each station; one adjusted for long range work, and the other for shorter ranges. They are connected to earth by the wire from E. The box shields the receiving apparatus from all outside electrical influences, except that brought in by the receiving wire.
a is the sending battery, giving a current through the primary of the sending coil, w, of about 6 or 8 amperes, at about 17 volts. The current may be taken either from a battery or from a dynamo circuit, direct or transformed, but usually a storage battery of 8 cells in series is employed on account of the steadiness of its current. This may be recharged from time to time, but continuous charging by a dry-cell battery of slightly higher E.M.F. has been found to give the best results. For this purpose a battery of 98 Obach dry cells, arranged in 7 rows in multiple and 14 in series, is joined in multiple with the storage battery.
b, the sending key, differs from an ordinary telegraph key in having a very long gap between its contacts, made necessary by the large sparks produced between them by the self-induction of the heavy current. These contact points are large and the rubber handle of the key is long. The back-stop of the key connects the vertical wire, H, to the receiver, so that after "calling" a station or sending a message by merely letting go of the key, the receiver is connected up ready to take down the answer or acknowledgment from the other station. This automatic connection is seldom used, however, it being preferable to shift the vertical wire by hand from the sender to the receiver.
The limit of speed in sending with this key, is about twelve words per minute; to send faster than this the vertical movement of the key must be shortened, and if shortened too much, the spark is not broken when the key is raised, which not only sends a false signal, but scores up the contact points until they either fail to make the next contact, or making it, they weld together and stick.
During the time that the circuit is closed at the key, the primary current is made and broken at the rate of about forty times per second by the spring circuit-breaker and sparks pass between the ball terminals, e, of the secondary of the coil, W.
This coil, known as App's lo-inch coil, is similar to the Ruhmkorff coil, and contains the condenser, Y, in its base. One of the terminal balls is connected to earth at E; on board ship this connection is made to the hull and must be well soldered. The vertical wire is connected to the other terminal.
The distance between these balls is varied from a half-inch to an inch; the range of transmission seems to increase with the length of the spark, but the sending must be done very slowly when the distance is greater than eight-tenths of an inch, otherwise many of the dots will fail to produce any spark at all. Any serious leakage from the coil or the vertical wire is indicated by the character of the spark; if a dead earth occurs no spark is produced at all, as the secondary circuit is then completed through the earth. A sharp crackling sound indicates the best spark. The sparks seem to pass always in the same direction, and one terminal becomes warm from sending, while the other does not change its temperature. The direction of the primary current through the coil is sometimes reversed in order to equalize the wear on the contact points of the vibrator.
The vertical or aerial wire was made up of seven copper wires, each 1/25-inch diameter and covered with rubber insulation. The upper end had an eye spliced in it, to which was secured a string of three or more round sticks of ebonite, each about 18 inches long and 1 inch thick, with holes in their ends for lashings. Sometimes the upper end of the wire was bighted up in several lengths of about two feet, and sometimes it was coiled up in a few turns of about one foot diameter; but no particular attention was paid to that part of it.
A temporary wooden mast was sent up and lashed to the topmast and fitted with hemp or manila rigging. A sprit about 18 feet long was hoisted to the truck of this mast by halliards bent on at one-third its length. The short end was steadied by a down-haul and the wire was triced up to the long end by halliards bent on to upper ebonite sticks, so that the upper end of the wire was above all metal masts, lightning conductors and wire rigging, and as high as it could be got. The wire was led away from the mast, keeping it clear of rigging, smokestacks, davits and the like, to a hatch or skylight; wherever it changed direction, guys of ebonite sticks were bent on to keep it in place, and where it went through a hatch, skylight, bulkhead or door, a rubber tube was used to lead it through. The matter of insulation of this wire is very important; in damp weather the number of ebonite sticks in the various strings must be increased. Sometimes these wires have been led down through the ventilators.
It is believed that the distance of transmission depends upon the vertical component of the aerial wire and varies as the square of the height of the upper end of the wire above the instruments, or above the hull of an iron ship when the instruments are placed below the upper deck.
During the navy trials this height on the New York was 130 feet, on the Massachusetts 140 feet, at Navesink 150 feet, and on the Porter 45 feet.
The height of a shore station above the sea-level or above the surrounding country does not seem to affect its range.
A person touching the wire when it is transmitting receives a severe but not dangerous shock.
The spark at the coil, or even one produced by a bad leak from the wire to earth, would set fire to an inflammable mixture of gas or other easily ignited matter.
There would seem to exist a danger of lightning being conducted below by the wire, but as it offers an unbroken metallic path to earth, possibly, by continuously discharging during an electrical atmospheric disturbance it may act as a protection from lightning. No casualties have been reported from lightning. Atmospheric electricity has been known to cohere the coherer, but the occurrence was never sufficiently frequent to prevent signals from being read.
The receiving wire takes up the impulses sent out by the sending wire and transmits them through the primary, M, of the transformer to earth.
The transformer is of the "step-up" type, the secondary coil being longer than the primary, so that the impulses induced in the secondary coil, M2 have an increased effect on the coherer. To prevent their passing through the relay coil into the battery and there being dissipated, the choking coils, K1K1, are placed in the circuit and an alternative path through the condenser, L, is provided. The choking coils are made of fine insulated wire wound on iron cores, and are known to impede an alternating current. The condenser, L, prevents the relay current from being short-circuited around the coherer, but seems to offer no resistance to the passage of the induced impulses.
The transformer and the condenser, L, were contained in a small wooden box about 7 inches long, 1 ½ inches wide, and 1 ½ inches high, fitted with from 5 to 8 binding posts. This box, known as the "jigger," was never opened. Sometimes a small additional condenser was connected up outside of the "jigger."
When the coherer is cohered the relay circuit is completed, and the armature of the relay is drawn over against its stop, and there completes the two parallel circuits from the battery, r, one through the tapper and the other through the Morse register or ink-writer, h.
The tapper has a spring armature like that of an electric bell, and its hammer strikes the coherer a series of rebounding taps. When the impulses cease the next stroke of the tapper decoheres the filings and breaks the relay circuit when the spring of the relay pulls its armature back and so breaks the tapper and register circuits.
The Morse register or ink-writer is of a commercial pattern and prints the dots and dashes on a tape, reeled off by clockwork.
Across the coils of the relay, tapper and register are the non-inductive shunts, q, p1 and h1. These consist of fine insulated wire wound "on the bight" on wooden bobbins, and their office is to short-circuit the high potential currents induced in the coils when the battery currents are broken and so prevent their affecting the transformer or the coherer. The non-inductive shunts, q1 and p2, across the contact points of the armatures of the relay and of the tapper, are to prevent the sparking which would otherwise take place there when their currents are broken.
The shunt, n1, containing the condenser, L1, is also placed across the contact points of the relay, but its action was not explained.
The leading wires from the back-stop of the key and from the ink-writer are the only ones that enter the receiving box, and to prevent them from transmitting the effects of the sending spark to the receiver at the same station, special devices are employed. The wire from the key is lead-covered, and the lead is soldered to the box where the wire enters. The ink-writer wire enters at B, which is an extension on the box containing a choking coil, each layer of which is covered with tin-foil in metallic contact with the box. The other leading wire is itself connected to the box.
The exact dimensions and the proper construction of all these devices have been arrived at by experiment, and the result is a complete and successful system of wireless telegraphy which has been thoroughly tested, during the past year, afloat and on shore and under various conditions of weather. The greatest distance yet covered by it is no miles, from Chelmsford, across the English Channel, to Boulogne. The greatest distance signaled across between two ships at sea was 72 miles, during
the British maneuvers. There are a number of permanent shore stations in England and the East Goodwin Lightship has been in communication with the South Foreland Lighthouse, 12 miles distant, since Christmas, 1898.
The instruments brought over to this country were not expected to work over about 25 miles during the races, but when put on board the New York and the Massachusetts, they kept up communication for 36 miles at sea, and the Massachusetts read the New York's signals at about 45 miles. The Porter read signals 7 miles from the Massachusetts, and the Massachusetts read the Porter's messages 8 ½ miles.
Each station receives all signals made within its range, and if two or more send at the same time, the receiving tape is illegible. So that if a ship got within range of an enemy's fleet she could not only read their signals, but, by working her transmitter continuously, could prevent the transmission of signals between the ships of the fleet.
Experiments looking towards the prevention of this interference have been made with partial success, but as it stands to-day interference cannot be prevented.
Another system, based on the use of metallic mirrors without vertical wires, has been developed, but its greatest range being about 1 ¾ miles it is not adapted to fleet signaling.
The newspapers report that the Marconi system has been adopted by the British army in South Africa and by the British navy; the hulk Hector is to be fitted up at Portsmouth for the instruction of signalmen of the navy.
The incorporation of an American company, controlling the Marconi patents in the United States is also reported.
If adopted by our navy, the care and operation of the apparatus would require about the same intelligence, education and length of special training as are now required for the electric plants on board ship, and a school of instruction should be established as soon as it is decided to adopt the system.