The telephone relay consists of a microphone C, Fig. 25, formed of the two pieces of osmium iridium alloy. The contact is separated to a minute degree partly by the action of the local current from F, which flows through it and also through the winding W of the two magnet coils. The local current from F assists in forming the microphone by rendering the space between the contacts conductive. The vibrating reed P is fastened to the metal frame (not shown) which carries a micrometer screw by which the distance between the contacts can be accurately regulated. It will be seen from Fig. 25 that the local circuit consists of a battery F (about 1.5 volts), the microphone contacts C, the windings W, milliampere meter B, and the terminals T, for connecting to the galvanometer or telephone, all in series. On the top of the magnet cores N, S is a smaller magnet D, wound with fine wire for a resistance of about 4935 ohms, the free ends of the coils being connected to the detector terminals. The working is as follows. Supposing the current from the detector flows through D in such a way that its magnetism is increased, the reed P will be attracted, the contacts opened, and their resistance increased. It will be seen that the current from F is passed through the coils W, in such a way as to increase the magnetism of the permanent magnet, so that any opening of the microphone contact increases their resistance, causes the current to fall, and weakens the magnets to such an extent that the reed P can spring back to its normal position. On the other hand, if the detector current flows through D in such a direction as to decrease the magnetism in the permanent magnets, the reed P will rise and make better contact owing to the removal of the force opposing the stiffness of the reed. Owing to the decrease in the resistance of the microphone, the strength of the local current will be increased, the magnets strengthened, and the reed P will be pulled back to its original position. This relay gives a greatly magnified current when properly adjusted, the current being easily increased from 10^-4 to 10^-2 amperes. It is also very sensitive, but needs careful adjustment in order that the best results may be obtained. A greater range of magnification can be obtained by placing two or more relays in series.

A very sensitive receiver designed by the writer is given in the figures 26 and 27. To the centre of a telephone diaphragm is fastened a light steel point P, and the movement of this point is communicated to the aluminium arm D, which is pivoted at C. As will be seen the telephone receiver is of special construction, it containing only one coil and therefore only one core; by this means the movement of the diaphragm is centralised. The coil is wound for a resistance of about 200 ohms, and the diaphragm should be fairly thin but very resillient.

To the free end of D is fastened the mirror T, made from thin diaphragm glass about 1¹/₂ centimetres diameter, and having a focal length of 40 inches. Light from the lamp L is transmitted by the lens N in a parallel beam to the mirror which concentrates it to a point upon a hole ¹/₁₀₀th of an inch in diameter in the screen J. As the telephone diaphragm vibrates under the influence of the received signals the arm, and consequently the mirror, vibrates also, and the hole in the screen J is constantly being covered and uncovered by the spot of light. It will be seen from Fig. 27 that the ratio between the centre of the mirror and the pivot C, and C and the steel point P is 10:1, so that if a movement of ¹/₂₀₀₀₀th of an inch is obtained at the centre of the diaphragm the mirror will move ¹/₂₀₀₀th of an inch; and as the focal length of the mirror is 40 inches a movement of ¹/₅₀th inch is given to the spot of light.

This receiver is capable of working at a fairly high speed, as the inertia of the moving parts is practically negligible; the weight of the arm and mirror being less than 20 grains. The hole in the screen is made slightly less in diameter than the traverse of the revolving cylinder, the slight distance between the cylinder and the screen allowing the light to disperse sufficiently to produce a line on the film of about the right thickness.

There are two other possible means of photographically receiving the picture that upon investigation may yield some results; but it is doubtful whether the current available, even that obtained from a telephone relay, will be sufficient to produce the desired magnetic effect, and the insertion of a second relay would detract greatly from the efficiency by decreasing the speed of working. If rays of monochromatic light from a lamp L, Fig. 28, pass through a Nicol prism P (polarising prism), then through a tube containing CS₂ (carbon bisulphide), afterwards passing through the second prism P' (analysing prism), and if the two Nicol prisms are set at the polarising angle, no light from L would reach the photographic film wrapped round the drum V of the machine. Upon the tube being subjected to a field produced by a current passing through the coil C, the refractive index of the liquid will be changed, and light from L will reach the photographic film.^[7]

The second method is rather more complicated, and is based upon the fact that the kathode rays in a Crookes' tube can be deflected from their course by means of a magnet. In Fig. 29 the kathode K of the X-ray tube sends a kathode ray discharge through an aperture in the anode A, through a small aperture in the ebonite screen J on to the drum V of the machine, round which is wrapped a photographic film; A and K being connected to suitable electrical apparatus. Upon the coil M being energised, the kathode-ray is deflected from its straight-line course, and the drum V is left in darkness.

The method which is now going to be described is very ingenious, as it makes use of what is known as an electrolytic receiver. This method of receiving has proved to be the most practical and simple of all the photo-telegraphic systems that have been devised.

The application of this system to wireless reception is as follows. The aerial A, and the earth E, are joined to the primary P of a transformer, the secondary S being connected to a Marconi valve receiver C. The valve receiver is connected to the battery B and silvered quartz thread K of an Einthoven galvanometer (already described). The thread is ¹/₁₂₀₀₀th of an inch in diameter, and will respond to currents as small as 10^-8 of an ampere. The light from M throws an enlarged shadow of the thread over a slit in the screen J, and as the thread moves to one side under the influence of a current, the slit in J is uncovered, and the light from M is thrown upon a small selenium cell R. In the dark the selenium cell has a very high resistance, and therefore no current can flow from the battery D to the relay F. When the string of the galvanometer moves to one side and uncovers the slit in the screen J, a certain amount of light is thrown upon the selenium cell lowering its resistance, allowing sufficient current to pass through to operate the relay.

Round the drum of the machine (shown in Fig. 7) is wrapped a sheet of paper that has been soaked in certain chemicals that are decomposed on the passage of an electric current through them. As soon as the local circuit of the relay is closed, the current from the battery Z (about 12 volts) flows through the paper and produces a coloured mark. The picture, therefore, is composed of long or short marks which correspond to the varying strips of conducting material on the single line print. In order to render the marks short and crisp, a small battery Y, and regulating resistance L, is placed across the drum and stylus. The diagram, Fig. 30, gives the connections for the complete receiver.

The paper used is soaked in a solution consisting of

The paper has to be very carefully chosen, as besides being absorbent enough to remain moist during the whole of the receiving, the surface must also remain fairly smooth, as with a rough paper the grain shows very distinctly, and if there is an excess of solution the electrolytic marks are inclined to spread and so cause a blurred image. The writer tried numerous specimens of paper before one could be found that gave really satisfactory results. It was also found that when working in a warm room the paper became nearly dry before the receiving was finished, and the resistance of the paper being greatly increased (this may be anything up to 1000 ohms), the marking became very faint. A sponge moistened with the solution and applied to the undecomposed portion of the paper, while still revolving, was found to help matters considerably.

Another experience which happened during the writer's early experiments, the cause of which I am still unable to explain, occurred in connection with the stylus. The stylus used consisted of a sharply pointed steel needle, and after working for about three minutes it was noticed that the lines were becoming gradually wider, finally running into each other. Upon examination it was found that the point of the needle had worn away considerably, becoming in fact, almost a chisel point. Almost every needle tried acted in a similar manner, and to overcome this difficulty the stylus shown in Fig. 31 was devised.

It will be seen that it consists of a holder A, somewhat resembling a drill chuck, fastened to the flat spring B in such a manner that the angle the stylus makes to the drum can be altered. The needle consists of a length of 36-gauge steel wire, and as this wears away slowly the jaws of the holder can be loosened and a fresh length pushed through. The wire should not project beyond the face of the holder more than ¹/₈th inch. The gauge of wire chosen would not suit every machine, the best gauge to use being found by trial, but in the writer's machine the pitch of the decomposition marks is much finer than of those made by the commercial machines, and this gauge, with the slight but unavoidable spreading of the marks, will produce a mark of just the right thickness. As already mentioned, no explanation of this peculiarity on the part of the stylus can be given, as there is nothing very corrosive in the solution used, and the pressure of the stylus upon the paper is so slight as to be almost negligible.

No special means are required for fastening the paper to the drum, the moist paper adhering quite firmly. Care should be taken, however, to fasten the paper—which should be long enough to allow for a lap of about ¹/₄ inch—in such a manner that when working the stylus draws away from the edge of the lap and not towards it.

The current required to produce electrolysis is very small, about one milliampere being sufficient. Providing that the voltage is sufficiently high, decomposition will take place with practically "no current," it being possible to decompose the solution with the discharge from a small induction coil. The quantity of an element liberated is by weight the product of time, current, and the electro-chemical equivalent of that element, and is given by the equation W = zct, where

The chemical action that takes place is therefore very small, as the intermittent current sent out from the transmitter in some cases only lasts from ¹/₅₀th to ¹/₁₀₀th a second.

The decomposed marks on the paper are blue, and, as photographers know, blue is reproduced in a photograph as a white, so that a photograph taken of our electrolytic picture, which will of course be a blue image upon a white ground, will be reproduced almost like a blank sheet of paper. If, however, a yellow contrast filter is placed in front of the camera lens, and an orthochromatic plate used, the blue will be reproduced in the photograph as a dead black.

There is one other point that requires attention. It will be noticed that the metal print used for transmitting is a positive, since it is prepared from a negative. The received picture will therefore be a negative, making the final reproduction, if it is to be used for newspaper work, a negative also. Obviously this is no good. The final reproduction must be a positive, therefore the received picture must be also a positive. To overcome this difficulty matters must be so arranged at the receiving station that in the cases of Figs. 17, 18, 22, and 24, the film is kept permanently illuminated while the stylus on the transmitter is tracing over an insulating strip, and in darkness when tracing over a conducting strip. In Fig. 30 the relay F should allow a continuous current from Z to flow through the electrolytic paper, and only broken when the resistance of the selenium cell is sufficiently reduced to allow the current from D to operate the relay.

The author has endeavoured to make direct positives on glass of the picture to be transmitted, so that a negative metal print could be prepared. The results obtained were not very satisfactory, but the method tried is given, as it may perhaps be of interest. The plate used in the camera has to be exposed three or four times longer than is required for an ordinary negative. The exposed plate is then placed in a solution of protoxalate of iron (ferrous oxalate) and left until the image shows plainly through the back of the plate. It is then washed in water and placed in a solution consisting of

After being in this bath for about fifteen minutes the plate is again well washed in water, and developed in the ordinary way. The first two operations should be performed in the dark room, but the remaining operations can be performed in daylight, once the plate has been placed in the bichromate bath. As already stated, the results obtained were not very satisfactory, and such a method is not now worth following up, as it is comparatively easy so to arrange matters at the receiving station that a positive or negative image can be received at will.

It is necessary to connect the stylus of the receiving machine to the positive pole of the battery Z, otherwise the marks will be made on the underside of the paper. The electrolytic receiver, owing to the absence of mechanical and electro-magnetic inertia, is capable of recording signals at a very high speed indeed.

"Atmospherics," which are such a serious nuisance in long-distance wireless telegraphy, will also prove a nuisance in wireless photography, but their effects will not be so serious in a photographic method of receiving as they would be in the electrolytic system. In a photographic receiver where the film is, under normal conditions, constantly illuminated, the received signals (both the transmitted signals and the atmospheric disturbances) will be recorded, after development, as transparent marks upon the film, the remainder of the film being, of course, perfectly opaque. By careful retouching the marks due to the disturbances can be eradicated, a print upon sensitised paper having been first obtained to act as a guide during the process.

CHAPTER IV

SYNCHRONISING AND DRIVING

Clockwork and electro-motors are the source of driving power that are most suitable for photo-telegraphic work, and each has its superior claims depending on the type of machine that is being used. For general experimental work, however, an electro-motor is perhaps the most convenient, as the speed can be regulated within very wide limits. For a constant and accurate drive a falling weight has no equal, but the apparatus required is very cumbersome and the work of winding both tedious and heavy. This method of driving was at one time universally employed with the Hughes printing telegraph, but it has now been discarded in favour of electro-motors, which are more compact, besides being cheaper to instal in the first instance.

Synchronising and isochronising the two machines are the most difficult problems that require solving in connection with wireless photography, and as previously mentioned, the synchronising of the two stations must be very nearly perfect in order to obtain intelligible results. The limit of error in synchronising must be about 1 in 500 in order to obtain results suitable for publication.

The electrolytic system is perhaps the easiest to isochronise, as the received picture is visible. On the metal print used for transmitting, and at the commencing edge a datum line is drawn across in insulating ink. The reproduction of this line is carefully observed by the operator in charge of the receiving instrument, and the speed of the motor is regulated until this line lies close against a line drawn across the electrolytic paper. Although this may seem an ideal method there are one or two considerations to be taken into account. Unless the decomposition marks are made the correct length and are properly spaced, however good the isochronising may be, the result will be a blurred image. Any one who has worked with a selenium cell, will know that it cannot change from its state of high resistance to that of low resistance with infinite rapidity, and the effects of this inertia, or "fatigue" as it has been called, are more pronounced when working at a high speed. In working, the effects of this inertia would be to increase the time of contact of the relay F (Fig. 30) as the current from D would flow for a slightly longer period through R to F than the period of illumination allowed by K. This, of course, would mean a lengthening of the marks on the paper; results would also differ greatly with different selenium cells. There is a method of compensation by which the inertia of a cell can almost entirely be overcome, but it would add greatly to the complicacy of the receiving apparatus.

In using an electro-motor with any optical method of receiving there are two methods available. The first is an arrangement similar to that used by Professor Korn in his early experiments with his selenium machines. The motor used for driving has several coils in the armature connected with slip rings, from which an alternating current may be tapped off; the motor acting partially as a generator, besides doing good work as a motor in driving the machine. This alternating current is conducted to a frequency meter, which consists of a powerful electro-magnet, over which are placed magnetised steel springs, having different natural periods of vibration. By means of a regulating resistance the motor is run until the spring which has the same period as the desired armature speed vibrates freely. The speed of the motors at both stations can thus be adjusted with a fair amount of accuracy. Another method is to make use of a governor similar to those employed in the Hughes printing telegraph system. A drawing of the governor is given in Fig. 32. It consists of a Fig. 32.Fig. 32. metal frame which supports an upright steel bar S, whose ends turn on pivots. This bar is rectangular in section. The gear-wheel G is fastened near the bottom of this rod and gears with a similar wheel on the shaft of the driving motor (not shown). Suspended from the broader sides of S are the two flexible arms D, each carrying a brass ball T. These balls are not fastened to the arms, but can slide up and down, being held in position by the wire springs M, one end of each spring being fastened to the screws C. These screws work in a slot cut in the upper part of S, and are connected to the adjusting screw E. When E is turned the screws are raised or lowered accordingly, and also the balls on the arms D.

Fastened to the arms are two brushes of tow B, and these revolve inside but just clearing the inner surface of the steel ring Z. Upon the motor speed increasing above the normal the arms D, and consequently the balls T, swing out, making a larger circle, causing the brushes B to press against the steel ring Z, setting up friction which, however, is reduced as soon as the motor regains its ordinary working speed. By careful adjustment the speed of the motors can be kept perfectly constant. The object of having the balls T adjustable on D, is to provide a means of altering the motor speed, as the lower the balls on D the slower the mechanism runs, and vice versa.

Fig. 33.Fig. 33.

A simple and effective speed regulator devised by the writer is given in drawings 33 and 34. It comprises two parts, A and B, the part A being connected to the driving motor, and the part B working independently. The independent portion B consists of an ordinary clock movement M, a steel spindle J being geared to one of the slower moving wheels, so that it makes just one revolution in two seconds. This spindle, which runs in two coned bearings, carries at its outer end a light Fig. 34.Fig. 34. pointer D, about two inches long, to the underside of which is fastened the thin brass contact spring S, which presses lightly upon the ebonite ring N. The portion A comprises a spindle, pointer, and contact spring similar to those employed in B, the spindle J' being geared to the driving motor by means of F, so that the pointer D' makes a little more than one revolution in two seconds. By means of a special form of brake on the driving motor, the speed is reduced, so that both pointers travel at the same rate, viz. one revolution in two seconds. By careful adjustment the two pointers can be made to revolve in synchronism,^[9] and when this is obtained the contact springs S, S', pass over the contacts C, C', completing the circuit of the battery B and lamp L. When working properly the lamp L lights up regularly once every second. This regulator is an excellent one to use for experimental work, although it depends a great deal upon the skill of the operator, but good adjustment should be obtained in about two minutes. It is a good plan to insert a clutch of some description between the driving motor and the machine, so that the regulator can be adjusted prior to the act of receiving or transmitting, the machine being prevented from revolving by means of a catch. The motor used should be powerful enough to take up the work of driving the machine without any reduction in speed. The clocks M can be regulated so that they only gain or lose a few seconds in twenty-four hours, which gives an accuracy in working sufficient for all practical purposes.

Connection is made with the contact springs S, S', by means of the springs T, T', which press against the spindles J, J'.

Another important point is the correct placing of the picture upon the receiving drum. It is necessary that the two machines besides revolving in perfect isochronism should synchronise as well, i.e. begin to transmit and record at exactly the same position on the cylinders, viz. at the edge of the lap, so that the component parts of the received image shall occupy the same position on the paper or film as they do on the metal print. If the receiving cylinder had, let us suppose, completed a quarter of a revolution before it started to reproduce, the reproduction when removed from the machine and opened out will be found to be incorrectly placed; the bottom portion of the picture being joined to the top portion, or vice versa, and this means that perhaps an important piece of the picture would be rendered useless even if the whole is not spoilt. It is evident, therefore, that some arrangement must be employed whereby synchronism, as well as isochronism of the two instruments can be maintained.

There are several methods of synchronising that are in constant use in high-speed telegraphy, in which the limit of error is reduced to a minimum, and some modification of these methods will perhaps solve the problem, but it must be remembered that synchronism is far easier to obtain where the two stations are connected by a length of line than where the two stations are running independently.

In one system of ordinary photo-telegraphy synchronism is obtained in the following manner. The receiving cylinder travels at a speed slightly in excess of the transmitting cylinder, and as its revolution is finished first is prevented from revolving by a check, and when in this position the receiving apparatus is thrown out of circuit and an electro-magnet which operates the check is switched in. When the transmitting cylinder has completed its revolution (about ¹/₁₀₀th of a second later) the transmitting apparatus, by means of a special arrangement, is thrown out of circuit for a period, just long enough for a powerful current to be sent through the line. This current actuates the electro-magnet. The check is withdrawn and the receiving cylinder commences a fresh revolution in perfect synchronism with the transmitting cylinder. As soon as the check is withdrawn the receiving apparatus is again placed in circuit until another revolution is completed. As the receiver cannot stop and start abruptly at the end of each revolution a spring clutch is inserted between the driving motor and the machine.

Although a method of synchronising similar to this may later on be devised for wireless photography, the writer, from the result of his own experiments, is led to believe that results good enough for all practical purposes can be obtained by fitting a synchronising device whereby the two machines are started work at the same instant, and relying upon the perfect regulation of the speed of the motors for correct working.

The method of isochronism must, however, be nearly perfect in its action, as it is easy to see that with only a very slight difference in the speed of either machine this error will, when multiplied by 40 or 50 revolutions, completely destroy the received picture for practical purposes.

From what has been written in this and in the preceding chapters it will be evident that the successful solution of transmitting photographs by wireless methods will necessitate the use of a great many pieces of apparatus all requiring delicate adjustment, and depending largely upon each other for efficient working. As previously stated, there is at present no real system of wireless photography, the whole science being in a purely experimental stage, but already Professor Korn has succeeded in transmitting photographs between Berlin and Paris, a distance of over 700 miles. If such a distance could be worked over successfully, there is no reason to doubt that before long we shall be able to receive pictures from America with as great reliability and precision as we now receive messages.

In nearly all wireless photographic systems devised up to the present the chief portion of the receiver consists of a very sensitive galvanometer, and although very good results have been obtained by their use they are more or less a nuisance, as the extreme delicacy of their construction renders them liable to a lot of unnecessary movement caused by external disturbances. A galvanometer of the De' Arsonval pattern, used by the writer, was constantly being disturbed by merely walking about the room, although placed upon a fairly substantial table; and for the same reason it was impossible to attempt to place the driving motor of the machine on the same table as the galvanometer. For ship-board work it will be evident that the use of such a sensitive instrument presents a great difficulty to successful working, and a good opening exists for some piece of apparatus—to take the place of the galvanometer—that will be as sensitive in its action but more robust in its construction.

CHAPTER V

THE "TELEPHOGRAPH"

In the present chapter it is proposed to give a brief description of a system of radio-photography devised by the author, and which includes a greatly improved method of transmitting and receiving, as well as an ingenious arrangement for synchronising the two stations; the whole being an attempt to produce a system that would be capable of working commercially over fairly long distances.

The system about to be described, and which I have designated the "telephograph," is the outcome of several years' original experimental work, many difficulties that were manifest in the working of the earlier systems having been overcome by apparatus that has been expressly designed for the purpose.

In any practical system of radio-photography the following points are of great importance: (1) the speed of transmission; (2) the quality of the received picture; (3) the method of synchronising the two machines so that transmission and reception begin simultaneously; (4) the correct regulation of the speed of the driving motors; (5) the simplicity and reliability of the entire arrangement. Points 1 and 2 are dependent upon several factors; the number of contacts made by the stylus per minute; the size of the metal print used; the number of lines per inch on the screen used in preparing the print; and the accurate and harmonious working of the various pieces of apparatus employed.

In the system under discussion the size of the metal print used is 5 inches by 7 inches, and a screen having 50 lines to the inch is used for preparing it. With the drum of the machine making one revolution in four seconds, the stylus makes 87 contacts per second, or 5220 a minute, the time for complete transmission being twenty-five minutes. By the use of ordinary relays not more than 2000 contacts a minute can be obtained, and in the present system it is only by means of a specially designed relay that such a high rate of working has been made possible. Similarly, too, with the receiving of such a large number of signals transmitted at such a high speed, a special instrument has been devised that can record this number of signals without any trouble, and could even record up to 8000 signals a minute, provided that a suitable transmitter could be designed.

In the present system the writer does not claim to have completely solved the problem of the wireless transmission of photographs, but it is a great advance on any system previously described, and the following advantages are put forward for recognition: (1) a greatly improved method of transmitting and receiving; (2) a simple method of regulating the speed of the driving motors and maintaining isochronism with a limit of error of less than 1 in 800; (3) an arrangement for synchronising the two machines whereby transmitting and receiving begin simultaneously; (4) the use of one machine only at each station.

Transmitting Apparatus

A diagrammatic representation of the apparatus required for a complete station, transmitting and receiving combined, is given in Fig. 35, the usual wireless equipment having been omitted from the diagram to avoid confusion.

The Machine.—This, as will be seen from Fig. 36, consists of a base-plate M, to which are attached the two bearings B and B'. The bearing B' is fitted with an internal thread to correspond with the threaded portion of the shaft D. The drum V is a brass casting, being fastened to the shaft by set screws. The shaft is threaded 75 to the inch. The bearings are preferably of the concentric type. The circuit breaker C is so arranged that when the drum has traversed the required distance, the end of the shaft pushes back the spring M, breaking the circuit of the driving gear and stopping the machine. The machine is connected to the driving gear by the flexible coupling A.

The drum measures 5 inches long by 2¹/₈ inches diameter, and this takes a metal print 5 inches by 7 inches, which allows for a lap of about ¹/₄ inch. In working, the print is wrapped tightly round the drum, being secured by means of a little seccotine smeared along one edge. Care must be taken that the edge of the lap draws away from the point of the stylus and not towards it. A margin of bare foil, about ¹/₈ inch wide, should be left on the print at the commencing edge, the purpose of which will be explained later.

The Stylus.—As the drum of the machine travels laterally, by reason of the threaded shaft and bearing, the stylus must necessarily be a fixture. It consists of a holder B, drilled to take a hardened steel point S, attached to the spring M. The spring is arranged to work in the guide F, which is provided with an adjusting screw W for regulating the pressure of the stylus upon the print; the pressure being sufficient to enable good contact to be made, but must not be heavy enough to scratch the soft foil. The needle should present an angle of about 60° to the surface of the print, as this angle has been found to give the best results in working.

To eliminate any sparking that may take place at the point of make and break, due to the self-induction of the relay coils, a condenser C, about 1 microfarad capacity, should be connected across the drum and stylus. The complete stylus is given in the drawings, Figs. 37, 37a, and also in the diagrams Figs. 8 and 9.

The Relay.—As will be seen from the diagram, Fig. 38, this consists of two electro-magnets having very soft iron cores, the magnet M being wound in the usual manner, while the magnet N is wound differentially. The armature A is made as light as possible, and is pivoted at P, and when there is no current flowing through any of the coils, is held midway between the magnet cores by the two spiral springs S and T, which are under slight but equal tension. The connections are as follows. The wires from the winding on M are connected directly to the relay terminals F and H, as are also the wires from one winding on N. The other winding on N is connected in series with the battery C, ammeter B, and regulating resistance R.

When the circuit of the battery C is completed, the coil of N, to which it is connected, is energised, and the armature A is attracted against the stop V. When in this position the tension of the spring S is released, while the tension of the spring T is increased. As soon as the circuit of the battery D is completed by means of the metal line print on the transmitting machine, the current divides at the terminals F and H, a portion flowing through the magnet coil M, and a portion through the remaining winding on N. The current which flows through the winding on N produces a magnetising effect equal to that caused by the other winding on N, but since the two windings are of equal length and resistance, and since the current flowing through the two windings is of equal strength but in opposite directions, the result is to neutralise the magnetising effects produced by each winding, and consequently no magnetism is produced in the cores.

The other portion of the current from D flows through the coil M, and it becomes magnetised at the same time that the coil N becomes demagnetised. The armature A is attracted by M against the stop X, and this attraction is assisted by the spring T, which was under increased tension. The conditions of the springs are now reversed, the spring S being under increased tension, while the tension of the spring T is released.

As soon as the current from D is broken, the magnetism disappears from M, the neutralising current in N ceases, and N once more becomes magnetised, owing to the current which still flows through one winding from C; the armature is therefore again attracted by N, assisted by the spring S. The current flowing through the two windings of N must be perfectly equal, and the regulating resistance R, and ammeters B and B', are inserted for purposes of adjustment. The current from C must flow in a direction opposite to that which flows from D.

The local circuit of the relay is completed by means of a copper dipper in mercury, somewhat resembling an ordinary mercury break, but modified to suit the present requirements. The arrangement will be seen from Fig. 39. The whole of the moving parts are made as light as possible, and for this reason the rod C and the dippers F, F' should be made as short as convenient. The containers H, H' are separate, of cast iron, and rectangular in shape. The dipper is of very thin copper tube—an advantage where alternating current is to be used—and is made adjustable for height on the suspending rod C. The leg F is of such a length that permanent contact is made with the mercury in the container H, while the leg F' clears the surface of the mercury by about ¹/₄ inch, when the armature of the relay is in its normal position. To prevent undue churning of the mercury, which would necessarily take place if the dipper entered and left the mercury at each movement of the armature, a pointed ebonite plug is inserted in the end of the tube. This will be found to give good results at a high speed, the mercury being practically undisturbed, and the production of "sludge" reduced to a minimum. To prevent oxidation of the mercury, and to prevent arcing, the surface is covered with paraffin oil. If this is not sufficient to prevent arcing a condenser should be shunted across the containers. The volume of mercury, and the area of the dippers, should be sufficient to carry the current used for a considerable period without heating up to any extent. An adjustable weight J is provided in order to balance the armature and dipping rod.

The remaining transmitting apparatus consists of the battery D² and the usual wireless apparatus. The double-pole two-way switch B' is to enable the photo-telegraphic set to be switched out and the hand key W switched in for ordinary signalling purposes. The battery D² should be about 12 volts.

Receiving Apparatus

The wireless portion of the receiver is similar to that given in Fig. 22, is of the usual syntonic type, and comprises an oscillation transformer, S being the secondary, and P the primary; C' is a block condenser, and C a variable condenser. The detector D is of the carborundum crystal or electrolytic pattern. A two-way switch B is provided so that the relay U can be switched out and the telephones J switched in for ordinary receiving purposes. The relay U is a Brown's telephone relay.