Wireless Telegraphy And Telephony

"Wireless Telegraphy And Telephony (see 26.529). - Wireless telegraphy and telephony (also called radiotelegraphy and radiotelephony) made enormous progress between 1910 and 1921. This was due chiefly to the improvements and advances, effected in three great inventions, viz.: the three-electrode thermionic tube developed out of the Fleming oscillation rectifying valve, the high frequency alternator, and the Poulsen arc generator of continuous waves. The first of these has given a means of detecting electric waves of immense sensitivity, and also a most effective and easily managed generator of continuous electric waves. The second has provided machines for creating high frequency electric currents, and therefore electric waves, of great power, enabling large long-distance radio stations to be equipped which can signal to any part of the world by day or night. The third has also given an alternative method of generating high-power continuous waves. These generating and receiving appliances quite revolutionized wireless telegraphy and made wireless telephony possible not merely as an experimental feat, but as a practically useful art. In addition to these inventions there have been others such as directive radiotelegraphy, and wire-guided high frequency telegraphy and telephony of immense utility. The application of the thermionic valve in ordinary wire telephony as a repeater is also bringing about improvements of very great importance. Contemporaneously with these achievements investigations have been made of a more scientific character arising out of the study of the nature of electric wave propagation round our globe and of the causes of atmospheric disturbances, called " strays," which have always been the great obstacle to practical radiotelegraphy.

We shall consider briefly the nature of these improvements in turn.

High Frequency Alternators

It had become clear by 1904 or 1905 that the use of continuous waves in radiotelegraphy would have marked advantages over the then employed damped wave trains produced by condenser discharges, and would be essential for the accomplishment of radiotelephony. The most obvious method of producing such continuous waves (C.W.) was by some form of high frequency alternator. At that time, when wave lengths of 300 to 3,000 metres or ',coo to 10,000 ft. were mostly in use, this meant the design of machines giving alternating currents having a frequency of 1,000,000 to ioo,000,000, and such frequencies seemed unattainable by any ordinary alternator construction as long as the revolving part of the alternator had to carry coils of wire. In low frequency alternating current dynamos, generating currents, reversed 50 to 200 times a second, there is an electromagnet which provides a constant magnetic field through which field coils of wire are moved so as to generate in the latter an alternating current. Either the field coils or the armature coils may be the rotating portion. In the case of alternators required to produce high frequency currents (20,000 to roo,000) it is impossible to rotate coil-wound armatures or fields at the necessary speed, and the most usual solution of the problem is to construct inductor alternators in which the moving part consists merely of a disk or drum of steel with teeth or ridges on its edge or surface, which serve to change the magnetic flux through stationary armature coils, the field coil also being fixed. We can then balance such a drum or disk and so fashion its edge that it can be rotated at a high speed safely. With the increase in capacity and wave length of the aerial wires or antennae requisite for long distance power stations, frequencies between 20,000 and ioo,000 came into use, and attention was again directed to the design of alternators giving such frequencies.

M. Latour has classified these machines into: (1) alternators in cascade, (2) internal cascade alternators, (3) homopolar or inductor alternators, (4) variable reluctance alternators, and (5) alternators with partial utilization of periphery. Although Bethenod in France constructed in 1912 a small machine of type (1), the first alternators of type (2) of ioo kilowatt output, or so, were constructed by R. Goldschmidt about 1912.

In machines of types (I) and (2) we start with the production of a single-phase alternating current of some moderate frequency, say 10,000, and multiply it up to much higher frequencies by taking advantage of a well-known principle called Fresnel's theorem. If there be two equal vectors represented by lines of equal length, which are pivoted to one point and revolve with equal angular velocities in opposite directions, their resultant is a line of constant direction but periodically varying magnitude with amplitude twice the size of that of the revolving vectors. Hence an alternating magnetic field of constant direction may be resolved into two fields of constant magnitude but rotating in opposite directions, each of half the maximum amplitude of the alternating field. If then we pass a direct electric current through the field coils of an alternator and induce in the revolving armature an alternating current say of io,000 frequency the field due to this armature current may be resolved into two oppositely rotating fields, one of which is stationary as regards the field magnets and the other cuts them with twice the angular velocity of the rotor. This gives rise to a current of twice the frequency in the field coils. The field due to this latter current can again be resolved into two oppositely rotating fields and these induce currents of still higher frequency in the rotor coils. We can so build up currents of frequencies in the ratio of I, 2, 4, etc. The currents of intermediate frequency can be taken up in circuits comprising capacity and inductance tuned to these frequencies respectively. The current of the highest frequency can be put into an aerial wire and employed to radiate long electric waves of corresponding wave length.

High frequency alternators of the above description were built for the radio station established at Tuckerton, N.J., in correspondence with a similarly equipped one at Eilvese, near Hanover in Germany, and used for trans-Atlantic transmission from about 1912 up to the time of the entrance of the United States into the World War.

The third type of high frequency (H.F.) alternator, called the homopolar inductor alternator, is represented by the machines of E.F.W. Alexanderson with disk rotor, and Bethenod-Latour and the Societe Frangaise Radio-Electrique with drum rotor.

The principle of these alternators is as follows: a fixed ringshaped frame or stator has inwardly projecting teeth of laminated iron, and a ringshaped magnetizing coil traversed by a direct exciting current causes magnetic flux to spring across from the teeth on one side (N. poles) to the teeth on the other side (S. poles). This flux traverses the air gap. Over these teeth is a zig-zag armature winding. Between the teeth in the Alexanderson machine revolves a steel disk with teeth cut in the periphery (see fig. i and fig. ia).

parallel, thus giving to the aerial 500 or Goo kilowatts of electric power. Also similar machines are employed in the large French radio station at Croix-d'Hins near Bordeaux, and are installed in the very large French radio station at St. Assise, near Melun, which began to be erected in 1921. Alexanderson alternators of 200 kilowatt power are installed at the American naval radio station at New Brunswick, N.J. This station communicates with Stavanger radio station in Norway 43,554 m.) with Lyons (3,845 m.) and Nauen, near Berlin (3,958 m.) (see fig. I).

The Arc Generator

We must next mention the improvements made in connexion with the Duddell-Poulsen arc generator. In this appliance an electric arc is formed with a direct current FIGS. I (left hand) and ia (right hand). - FIG. I, 200 kilowatt Alexanderson high frequency alternator, driven by an electric motor.

FIG. Ia. - Half section of the Alexanderson high frequency alternator; showing the inductor disk which revolves between the stator poles. (By permission of The Wireless Press.) The number of teeth in the rotor is half the number of teeth on each side of the stator. These rotor teeth as they pass between the stator teeth decrease the reluctivity of the air gap and increase the magnetic flux passing. Hence as the rotor revolves the magnetic flux linked with the' armature circuit is alternately increased and decreased. This creates an electromotive force and a high frequency current in the armature circuit. The only revolving part of the machine is thus a well-balanced rigid steel disk. The field exciting coil and the armature coil are both stationary. Such machines are called in English inductor alternators and in French homopolar alternators. In the Bethenod-Latour machines the inductor is in the form of a steel drum with exterior of laminated iron in which longitudinal grooves are ploughed out. The stator ring has inward radially pointing laminated iron teeth on both edges, and the magnetic flux, leaving one set of teeth (N. poles), passes down through the drum teeth and up again into the other set of stator teeth (S. poles). The ridges on the drum serve to increase and decrease the flux through the armature wire which is wound zig-zag on both sets of stator teeth.

The peculiarity of the French machine is that the rotor or drum has many more ridges or teeth on it than the stator ring. By this we can obtain the necessary high frequency without dangerous peripheral speed in the rotor, and yet leave plenty of space for the armature winding placed on the stator teeth.

On account of the few turns in the armature such machines give a small electromotive force, but this can be raised by suitable static transformers, one secondary terminal of which is connected to the radiating antenna and the other to the earth plates. The aerial wire is then tuned to the frequency of the alternator and the necessary wave length. A point of importance is the exact regulation of the speed of the machines which must be kept constant to within oi % of the normal. This is achieved by the use of very sensitive governors which control the speed of the direct coupled electric motor which drives the alternator. The signals are made in the case of the Alexanderson machines by varying the inductance of a tertiary coil on the transformer which transfers the energy from the aerial to a non-radiative circuit. By placing condensers in series with the H.F. alternators it has been found possible to run them in parallel, that is two or more together just as in the case of low frequency alternators in electric lighting and power stations.

High frequency homopolar alternators of the Bethenod-Latour type have been built by the Societe Francaise Radio-Electrique up to 220 kilowatts size for the French military radio station at La Doua, near Lyons. To avoid loss of power by air churning these machines are enclosed in an air-tight case in which a partial vacuum is maintained. The speed of the machines is controlled by a Thury governor, and signals are made by short circuiting the armature coils in sections. Two or more machines can be run in between a water-cooled copper electrode (the positive terminal) and a carbon negative electrode. The arc is inclosed in a chamber filled with coal-gas or kerosene or alcohol vapour, and a powerful transverse magnetic field is made across it (see fig. 2). If the FIG. 2.-25 kilowatt Poulsen arc generator of electric oscillations; showing the electromagnet and arrangement for dropping alcohol into the arc box. (By permission of Marconi's Wireless Telegraph Co., Ltd.) arc terminals are shunted by a circuit containing a condenser and an inductance coil, high frequency oscillatory currents will be set up in this latter circuit under certain conditions. Their energy can he inductively transferred to an aerial wire so as to radiate continuous electric waves from it. The arc must be in a certain active condition determined by its length and the strength of the magnetic field in order to produce oscillations. Many investigations have been made to elucidate the working of this oscillation-producing arc.

A fairly complete list of these papers is given by P. O. Pedersen in the Proceedings of the Institute of Radio Engineers (United States), vol. v, p. 255, Aug. 1917.

In order that the arc may be active, i.e. produce oscillations in the condenser circuit, it must be drawn out to a certain length, and the transverse magnetic field must have a certain optimum value, which depends upon the density of the gas in which it is immersed and on the frequency of the oscillations and the strength of the direct current feeding the arc.

Under best conditions the effective or root mean square value of the oscillatory current is1.2 = 0.7 of the strength of the direct feeding current. Thus, if the arc is fed with 100 amperes (D.C.), it should give 70 amperes (A.C.) in the oscillation circuit under best conditions: the possibility of this transformation is the result of the negative slope of the characteristic curve of the direct current arc, viz. that an increase in arc current is accompanied by a decrease in electric potential difference and vice versa. Also the necessity for maintaining round the arc a non-oxygenic atmosphere, or one consisting of hydrogen or carbon hydrides or oxides, is due to the fact that in these gases the arc characteristic has a steeper downward slope than in air (see W. L. Upson, Phil. Mag. July 1907). The transverse magnetic field is requisite suddenly to extinguish the arc at each oscillation, and so produce an electromotive force in the inductance coil which recharges the condenser in the reverse direction. Broadly speaking then, the operation which takes place is as follows: - if the arc is burning steadily and the condenser is shunted across the electrodes, the result is to rob the arc of some current. Hence the potential difference (P.D.) of the arc electrodes increases. This, however, continues the charging of the condenser in the same direction. Then the latter discharges back through the arc and this lowers the P.D. of the electrodes.

The study of the oscillatory arc by means of the oscillograph by H. Th. Simon, H. Barkhausen, A. Blondel, and P. O. Pedersen has shown clearly the nature of the operations taking place. If no magnetic field, or a weak one is employed, and if the arc is in air only, feeble oscillations are set up in the condenser circuit, and the current through the arc is a pulsatory unidirectional current. This is the case of the Duddell or musical arc which has no use in wireless telegraphy. If a stronger magnetic field is used and if the arc is in a hydrogen or coal gas atmosphere, then much more powerful oscillations are produced, and when the R. M. S. value of the condenser current is equal to, or greater than 70% of the direct current the arc current just falls to zero, or is extinguished at each oscillation. The function of the transverse magnetic field is then to blow out the arc by forcing the stream of electrons outward, and the effect of this sudden rupture is to create a strong adjuvant or assisting induced electromotive force in the inductance coil in the condenser circuit. This continues the arc current in the same direction, and the condenser thus becomes charged in the opposite direction. The process then repeats itself and we have powerful oscillations produced in the condenser circuit.

Although the condenser current is a sinusoidal current, and the arc current has the same form, yet owing to the shape of the dynamic characteristic curve the potential difference of the arc electrodes is an irregular curve with sharp peaks corresponding to the instants of cessation and recommencement of the arc current.

The practical construction of a Poulsen arc generator involves therefore a large electromagnet having poles which project into a box which can be kept full of alcohol, or kerosene vapour or else coal-gas. Into this box project also two electrodes, one of copper, through which water circulates to keep it cool, and one of hard carbon which is kept in slow rotation by a motor. The poles of the magnet are shaped bluntly conical so as to concentrate a powerful magnetic field transversely to the electric arc which springs from the copper (+) to the carbon (-) electrode. The arc is created by a direct current dynamo giving a voltage of 500 or more (see fig. 2). A separate shunt-wound dynamo is often employed to excite the electromagnet. In the circuit of the arc supply dynamo choking coils are inserted, and a circuit comprising a condenser of capacity C and an inductance (L) is connected as a shunt to the arc. If the capacity C is measured in farads and the inductance in henrys then the ratio ' L/1/ C is a function of the dimensions of a resistance reckoned in ohms, and should have some value of about 500 ohms or so.

Means must be provided for adjusting the magnetic field to its optimum value (Ho) which depends on the frequency (n) of the oscillations produced, where n is nearly equal to 1/2T1 / LC, or upon the radiated wave length X (in metres) where nX =300 million metres.

P. O. Pedersen states that Ho = a/X - b where a and b are certain constants (see " The Poulsen Arc and its theory " Proc. Insti- tute Radio Engineers, United States, vol. v, p. 255, Aug. 1917).

L. F. Fuller states that Ho = K where P is the power fur nished to the arc and K is a constant depending on the surrounding gas or vapour. For kerosene K =4-23; for ethylic alcohol K =8-5. For a power of 50 kilowatts and a wave length of 7,000 metres the arc in alcohol requires a field of 8,300 C.G.S. units and for a wave length of 20,000 metres and a power of 1,000 kilowatts the field must be 13,500 C.G.S. units. Hence as the air gap is large (generally several centimetres) extremely large magnets are required. For the 1,000 kilowatts arc plant the magnets weigh 80 tons and for the 500 kilowatts plant 65 tons. For smaller sizes of plant the magnet is of the open circuit type and for the larger of the closed circuit type. The arc chamber and magnets have to be cooled by water or oil circulation. From 100 kilowatts size and upwards this arc generator is very widely used in long distance radio stations. It labours, however, under the disadvantage that the signalling cannot be conducted by starting and stopping the arc but only by throwing the aerial out of tune or by deflecting the energy into a non-radioactive circuit. Hence power is equally consumed whether message signals are being sent out or only spacing waves.

FIG. 3. - Modern thermionic generating valve; showing the cylindrical anode and metal gauze grid, as made by the Marconi-Osram Valve Company. (By permission of Marconi's Wireless Telegraph Co., Ltd.) The Thernzionic Valve. - The third type of high frequency electric oscillation generator which has become of great importance in the last five years is the thermionic valve, which is a development of the Fleming valve (see 26.537).

The Fleming valve comprises a glass bulb, highly exhausted of its air, and contains a carbon or metal filament which can be rendered incandescent by an electric current. Around the filament is placed a metal cylinder carried on a wire sealed through the bulb. The peculiar property of it is that, when the filament is incandescent, the space between the filament and cylinder will conduct negative electricity from the filament to the cylinder but not in the opposite direction. Hence the name "valve" given to it. It can, therefore, be used to separate out the two constituents of a high frequency alternating current and " rectify " them into a direct current. This valve was extensively used as a detector of electric waves in wireless telegraphy from 1904 onward as described later. In 1907, Lee de Forest in the United States, after he had become acquainted with Fleming's invention of the valve and its use in wireless telegraphy, added t.o it an additional element in the form of a grid or zig-zag of wire placed between the cylinder and the filament but carried on a separate terminal. He thus made a so-called three-electrode thermionic valve, a name sometimes shortened into triode. In its modern form a thermionic valve of the latter type comprises a highly exhausted glass bulb having in it a filament of tungsten, or thoriated tungsten or of platinum wire coated with oxides of barium and strontium (see fig. 3). This is rendered incandescent by electric current from a storage battery. Around the filament is a spiral of nickel wire or else a cylinder of nickel wire gauze. This is technically called the grid. Around that again is a cylinder of sheet nickel called the plate. The plate and the grid are carried on separate wire stems sealed through the wall of the bulb. Although the three-electrode valve was originally devised as a detector of electric oscillations as described below yet about 1913, or before, it was found that both the two-electrode valve and also the threeelectrode valve can produce electric oscillations as well as rectify or detect them. When the filament is rendered incandescent torrents of electrons or particles of negative electricity are emitted from it. If the plate is given a positive potential relatively to the filament by means of a battery called a plate battery, these electrons are attracted to it, and this creates a movement of negative electricity called a thermionic current. If the bulb is highly exhausted and has a grid in it between the filament and plate, the electrons can only reach the latter by passing through the holes in the grid. If the grid is given a negative potential it reduces or stops the thermionic current. If it is given a positive potential it increases the current. The relation between thermionic current and grid potential can therefore be represented by a characteristic curve as shown in fig. 4.

Potential of Grid with respect to Filament FIG. 4. - Characteristic curve of a three electrode thermionic valve.

The three variables, viz. plate and grid potential (v p and v,) and the thermionic current (i,,), may be regarded as three rectangular coordinates which define a characteristic surface, and sections of this, parallel to the i,v g axes or i p v,, axes, delineate the principal characteristic curves. The central portion of this surface, corresponding to zero grid potential, is nearly a plane and has therefore the equation ip=av g +bv p where a and b are certain constants or coefficients.

If we pass a current by means of a high voltage battery from the filament to the plate and send this current also through the primary coil of an oscillation transformer, the secondary circuit of which is connected to the filament and grid, we have an arrangement by which continuous electric oscillations are produced and maintained. For, if properly connected, any variation of the grid potential will increase or decrease the plate current; and this acting through the transformer will in turn create the changes of grid potential which act to sustain the variations of plate current. The action is just like the well-known experiment of the singing telephone. If a magneto-telephone receiver is in series with a carbon transmitter and with a battery, then, when the transmitter is held near the receiver, the latter emits a shrill note. The sound given out by the receiver acts on the transmitter and this in turn actuates the receiver. The energy is drawn in both cases from the battery.

The three-electrode valve so used is called a transmitting valve, and the sustained electric oscillations it can produce, as above described, can be transferred to an aerial wire and cause it to radiate continuous electric waves.

Very large thermionic valves are now made with glass or silica bulbs, a foot or more in diameter, for use as transmitting valves, and numbers of these can be associated together to form a thermionic generator of large power. In this case the high voltage required to pass the plate current through the valve is obtained by the use of a battery of Fleming two-electrode valves which rectify a high tension alternating current of low frequency. A complete valve panel, as it is called, comprises the battery of rectifying valves, and threeelectrode valves and also the necessary transformers, induction coils and condensers (see fig. 5). Large valve panels are now constructed to transform electric power from 1 kilowatt to 50 kilowatts or more into high frequency electric oscillations of great energy.

Such valve generators are extensively used by Marconi's Wireless Telegraph Co. and others for the production of continuous waves, and are employed at Clifden Station in Ireland for the transmission of wireless messages across the Atlantic Ocean.

XXXII. -33.

Detectors

We must in the next place notice the improvements which have taken place in means for detecting continuous waves (C.W.) as used in wireless telegraphy. The reader may refer to the earlier article on Wireless Telegraphy (see 26.535) for an account of the principal appliances used in connexion with spark or damped wave telegraphy for the detection of electric oscillations, and especially to the section on the oscillation valve or two-electrode thermionic detector, from which other types of improved thermionic detector have been developed. Subsequently to the introduction of the two-electrode, but prior to the advent of the three-electrode thermionic detector, much use was made of crystal or rectifying detectors.

It will be remembered that the electric waves sent out from the transmitting aerial wire, which are identical in nature with light waves except for their much greater wave length, fall upon an aerial wire at the receiving station, and create in this latter extremely feeble, high frequency electric currents or oscillations which are a copy on a very reduced scale of the electric oscillations established in the transmitting aerial. The strength of these feebly received currents reckoned in amperes (I t) can be approximately computed from the strength of the currents in the sending aerial (I,) reckoned in amperes by means of an empirical formula valid up to about 2,000 m. due to L. W. Austin and L. Cohen which is as - follows: - I, 377 A d h* R hs e - 0.0045d/ N ix - I13 where h, and h r are the heights of the sending and receiving aerial wires in kilometres, d is the distance apart of the stations in kilometres, X is the wave length in metres, e the base of the Napierian logarithms, and R the total resistance of the receiving circuits in ohms.' The received currents may be something of the order of 5-10 microamperes more or less in the case of long distance working. To detect these feeble oscillations special appliances called detectors are in use. The so-called rectifying detectors do this by converting the received oscillations into feeble unidirectional currents, which in the case of damped waves are equivalent to short gushes of electricity in one direction corresponding in frequency to the condenser discharges in the transmitter. These can then be detected by a telephone, as they create in the latter a musical sound agreeing in pitch with the wave group frequency, and this, by the action of the key in the transmitter, is cut up into dot and dash audible Morse signals. One of the first rectifying crystal detectors was carborundum discovered in 1906 by H. H. C. Dunwoody in the United States. This material is a crystalline carbide of silicon produced in electric furnaces, and, in certain specimens, as shown by G. W. Pierce, has an electric conductivity 40 or 50 times greater in arc direction than in the opposite along one crystalline axis.

FIG. 5. - Valve panel for generating high frequency oscillations in the transmitters for wireless telegraphy, as made by Marconi's Wireless Telegraph Co., Ltd. (By permission.) The same properties are exhibited by hessite and anatase as well as by molybdenite and other native sulphides. Furthermore, it was found by L. W. Austin and G. W. Pickard that contacts between certain pairs of substances such as tellurium and aluminium, or zincite and chalcopyrite, also plumbago and galena, have the same kind of unilateral conductivity and can be employed for " rectify 1 The problem of predetermining the electric and magnetic force at any point on a conducting sphere due to a Hertzian oscillator at some point on it is a very difficult one. The reader will find references to the work of Macdonald, Nicholson, Love, Rybczynski, and others in The Principles of Electric Wave Telegraphy, Fleming, 4th ed. chap. ix., and also in a paper by Balth van der Pol in Phil. Mag., vol. xxxviii. (Sept. 1919). The final result is that diffraction alone will not account for long distance radiotelegraphy.

ing " high frequency oscillations. These crystal and rectifying detectors came at one time into great use in wireless telegraphy.' The limitation in the power of these crystal or rectifying detectors lay in the fact that the energy used in making the signal is only a portion of that captured by the receiving aerial from the incident waves. An immense improvement was therefore effected by the introduction of the three-electrode thermionic valve, which can act in the manner of a telegraphic relay and employs the received power merely to release a much larger amount of electric power from a local battery, which latter creates, the signal in the telephone or other instrument. Moreover, this type of detector is capable of being used in series so as to amplify or magnify enormously the signal-making power.

FIG. 6. - Thermionic amplifying and detecting valve of the type usually called the " French " valve with cylinder anode and spiral wire grid surrounding a straight filament.

FIG. 7. - Views of various types of three-electrode thermionic valves; (a) detecting and amplifying valve; (b) transmitting or generating valve; (c) amplifying valve of a type made by Marconi's Wireless Telegraph Co.; (d) small transmitting valve with gauze grid.

The modern hard or high vacuum thermionic valve as used for reception and amplification is now generally constructed as follows: - a small glass bulb or tube, a few inches in diameter, has sealed into it a filament which can be rendered brightly incandescent by current from a 2-to 3-cell storage battery (4-6 volts). This filament is of drawn tungsten wire, or else platinum coated with oxides of barium and strontium. The bulb is highly exhausted. Around the filament and close to it is coiled a spiral of nickel wire called the grid, and outside that a cylinder of nickel called the plate. The plate and grid are carried on wires sealed through the bulb, and connexions to the grid, plate and filament are brought to four terminal pins at the base fixed to a brass collar (see figs. 6 and 7). These pins fit into a suitable socket. A battery of 30 to 200 volts E.M.F. has its negative terminal connected to the filament and positive to the plate, and when the filament is incandescent a stream ' A fuller description of these rectifying detectors is given in The Principles of Electric Wave Telegraphy and Telephony by J. A. Fleming, 4th ed. chapter vi. (1919).

of electrons (atoms of negative electricity), called the plate current, flows from the filament to the plate through the apertures in the grid wires. This current completes its circuit through a coil of wire in the external plate filament circuit which may be one coil of a transformer. If the grid has a small negative charge given to it the plate current decreases, and, if a positive one, the plate current increases. The relation of plate current to grid potential can be delineated by a characteristic curve (see fig. 4). For a certain positive grid potential the plate current becomes saturated and ceases to increase. If the grid and filament are connected to the terminals of the receiving condenser in a wireless telegraph aerial, the incidence of electric waves on the aerial will create alternations of potential in the grid and alternations of plate current, and the amplitude of the plate potential may be five to ten times greater than that of the grid. The thermionic tube is then said to have an amplifying power of five to ten.

If the coil in the plate circuit forms the primary coil of a twocoil transformer the secondary circuit of the latter may be connected to the grid and filament of a second valve, and a second amplification of potential may take place. We can thus employ a series of valves in cascade and the total amplification increases as the nth power of the number of valves (a) in cascade. Thus if one valve amplifies potential ten times, three valves will amplify i,000 times and so on.

This .use of thermionic valves in cascade has given us detectors of enormous sensitivity. In order to detect damped oscillations we can adopt one of two methods. If we place a small condenser with a leak across its terminals in the grid circuit then the side of this condenser next the grid becomes negatively charged, and the plate circuit of the valve is reduced. This charge leaks away almost instantly and the plate current of the valve rises again. Hence if the incident waves are in " damped trains " a telephone receiver inserted in the plate circuit of the valve will give a sound of the pitch of the train frequency, and this can be cut up into signals. In this case the valve is used as a rectifier as in the case of the Fleming valve. The second mode of use depends upon the form of the characteristic curve. If we employ a small battery of cells to give the grid a certain positive potential we can operate the valve at a point on the curve near to the saturation point so that a small reduction in the potential of the grid lowers the plate current, but a small increment of potential does not increase it. Hence if the grid is connected to one terminal of the tuning condenser of a receiving aerial it will oscillate in potential when a train of electric waves falls on the aerial. This, however, will cause a drop in the plate current and hence a sound in a telephone receiver included in that circuit. If the incident waves are damped or intermittent trains the result is to make a steady musical sound in the telephone of the pitch of the train frequency. Accordingly, by interrupting the groups by the sending key, audible Morse signals can be received.

The above described methods of reception are however only applicable in the case of damped or intermittent trains of waves. If the electric waves are continuous as sent out by an alternator, arc, or valve transmitter, then we can only detect signals made with them by converting the continuous waves into the equivalent of a series of damped trains. This is done by generating in the receiving aerial electric oscillations by a local valve generator which have a frequency differing from that of the incident waves by about 300 to i,000. The result is to create in the receiving aerial resultant electrical oscillations which fluctuate periodically in amplitude just as " beats " in musical sounds are produced when two organ pipes slightly out of tune are sounded together. The number of beats per second is equal to the difference in the frequency of the two separate oscillations. In the aerial wire these electrical beats can then be detected by any of the types of detector and receiver used in spark wireless telegraphy. This method is therefore called " beat reception." The beats disappear when the signal bringing waves are interrupted at the sending station in making the spaces between the Morse code. signals.

One great use of the three-electrode valve, or triode, as it is sometimes called, is in amplifying feeble signals. It has been explained already that when the grid of the valve is electrified positively or negatively it increases or decreases the plate current and, therefore, the plate potential. The amplitude of the plate potential variations may however be five or ten times or more that of the grid potential variations. Hence the valve acts as a relay or magnifier of potential. Again we can interconnect a number of such triodes in series by induction coils so that the variations in plate current of one valve are made to vary the grid potential of the next. Hence by using a series of valves in cascade we can multiply potential variations of the grid of the first valve in a geometric progression and enormously magnify them. The remarkable achievements of modern long distance wireless telegraphy are chiefly due to the use of such cascade amplifiers. In fig. 8 is shown a view of such a detector made by Marconi's Wireless Telegraph Co. in which six valves are used as amplifiers and a seventh valve as a detector. So sensitive are these cascade receivers that it is not necessary to employ any long aerial wire to receive wireless signals from distant stations. It suffices to construct a large rectangle of a few dozen turns of insulated wire called a frame aerial and connect this in series with a condenser of suitable variable capacity and tune the arrangement ?

'?

?+iqh;:; exfEau t i ?ulb sp;rhi oryrid, pIa±e filames pf Pm terminals to the wave length of the wave to be received. This frame is then placed with its plane vertical and in the direction of the sending station. As the incident electric waves sweep over it they set up in the wire very feeble electric oscillations. If a cascade thermionic amplifier is then connected to the terminals of the receiving condenser and appropriate tuning carried out the signals will be heard in the receiving telephone. It is possible to make one small storage battery of three cells provide the electric current for incandescing the filaments, and one battery of 40-50 cells provide the plate currents for all the valves. Very compact and portable multiple valve receivers of this type have been constructed for use in aircraft and for reception of time signals from distant radio stations.

FIG. 8. - Marconi Co.'s type 55; thermionic amplifier with six amplifying valves and one detector as used in wireless telegraphy and telephony. (By permission of Marconi's Wireless Telegraph Co., Ltd.) Directional Wireless. - The frame aerial has the important quality of being directive; that is, it tells us the direction in which the incident waves are travelling. Hence if two receiving stations at a known distance apart are provided with directive aerials, and if they simultaneously observe the direction of the arriving waves from one transmitting station, which may be on an aircraft or ship, these observations laid down on a chart will enable them to fix the position of the source of the signals. In this manner the position of aeroplanes lost in the clouds or ships in the fog may be found and their exact position communicated to them. There was a considerable development of this directive radiotelephony during the World War of 1914-8.1 It has been found that there are peculiar difficulties in practising this direction finding at or about the times of sunrise and sunset.

In place of employing a movable frame aerial two fixed nearly closed circuit triangular aerials can be erected with their planes at right angles, and a resonant receiving circuit can be arranged to have a coil which is capable of rotation round a vertical axis but so as to be coupled inductively to both the fixed aerials by coils in the two aerial circuits. If electric waves fall on the aerials and if the movable coil of the receiving circuit is rotated into the azimuth in which it receives signals most loudly the direction of the plane of that coil will determine the line of direction of the transmitting station. It is possible by special arrangements to determine the direction along this line in which the electric waves are travelling. Many coast radio stations are now provided with direction finding aerials, and ships can call up these stations by wireless when in proximity, in case of fog, and have their bearings and exact position given to them. In another method of direction finding the coast station sends out a revolving beam of radiation which has a sharply marked point of zero radiation. The time of revolution of this beam is known, and also the instants when the zero radiation is in the true north and south direction at the sending station. Hence by observing the instants at which the zero radiation is observed at the ship, the ship's bearing with regard to the station can be determined. The station sends out time signals by which to correct the ship's chronometer.

1 See " Direction and Position Finding," by H. J. Round, Journal Inst. Elec. Eng. London 1920, vol. lviii, p. 224; also J. J. Bennett, Nature, May 19 1921, vol. cvii, p. 363.

Another ingenious application of radiotelegraphy has been made by Prof. J. Joly to enable ships to find their position in fogs and avoid collisions. For details the reader must be referred to his paper in the Proceedings of the Royal Society of London, vol. xcii., A, 1915-6, pp. 170, 176, 252, and also to a paper by H. C. Plumer on p. 377 in the same volume. See also J. Joly, Proc. Roy. Soc. Lond., vol. xciv, A, 1918, p. 547.

Wireless Telephony

The above described improvements in the production and detection of continuous electric waves had by 1921, within a few years, placed wireless telephony on a thoroughly practical basis. It is unnecessary to describe various experimental feats which had been achieved at intervals in this art of radiotelephony, in which the Poulsen arc or some modification was employed to generate the continuous waves (C.W.). All practical radiotelephony now involves the use of the thermionic valve both as a generator of C.W. and as detector.

The general principles of this method are as follows: - one or more three-electrode valves are employed in which the plate and grid circuits are inductively coupled so as to generate continuous oscillations. The plate circuit is also coupled inductively to a radiating aerial wire and continous waves sent off. The amplitude of these waves has then to be modulated in accordance with the wave form of a speech sound. This is done by means of another three-electrode valve called the control valve. The latter has the secondary circuit of an induction coil connected. between its grid and filament, and in the primary circuit is a microphone transmitter and a voltaic cell or two. Hence if speech is made to the transmitter the potential of the grid is varied in accordance with the wave form of the speech. The plate current of this control valve is caused to act upon the plate current or grid potential of the oscillating valve so as to modulate the resulting high frequency oscillations also in accordance with the wave form of the speech made. At the receiving end the received oscillations are amplified by a series of valves and then rectified and passed through a Bell magneto-telephone. The speech sounds are then reproduced by the receiving telephone. The advantage of this method is that only the ordinary standard telephone transmitter and receiver as used in telephony with wires are employed. To obtain the necessary high plate potentials in the oscillating valves we can either use voltaic batteries (dry cells) or else a small high tension direct current dynamo (1,000-2,000 volts), or else we can rectify a low frequency high tension alternating current by one or more Fleming valves. For aeroplane wireless telephony the plate voltage is supplied by a small dynamo driven by a wind screw which is set in action when the aeroplane flies. A large number of schemes for valve circuits for wireless telephone have been devised. During the war a great amount of ingenuity was expended in devising compact light weight sets of radiotelephone transmitters and receivers for use in aircraft and in the field (see figs. 9 and 9a). A problem of practical importance is that of two-way radiotelephony enabling two communicators to speak and hear simultaneously or to " cut in " or interrupt each other as can be done in ordinary telephony. If a single aerial wire has to be switched over from transmitter to receiver there is always risk of confusion owing to both operators trying to speak or listen at the same moment.

In the case of ground stations a practical solution is to use two wave lengths differing say by 5%. At each station there is a transmitter and a receiver say Ioo yd. apart. One transmitter is tuned to the distant receiver but the wave length of the home receiver, which is tuned to the distant transmitter, differs by 5%.. Each operator then speaks and listens on a different wave length and can " cut in " as he likes.

This method is, however, not applicable in the case of aeroplanes or ships for want of space. One suggested solution is that called the " quiescent aerial." The plate voltage of the oscillating valve is not supplied by a high voltage battery but at most by a few cells, and the remainder of the plate voltage is created by the rectification by the valve of the speech currents induced in the secondary circuit of the microphone transformer. In this case continuous waves are not thrown off from the aerial except in the act of speech to the microphone, and the receiver can then FIGS. 9 (Ieft hand) and 9a (right hand). - The wireless equipment of an aeroplane for wireless telegraphy and telephony. The generator is fixed to the outside of the hull of the aeroplane and driven by a wind screw. Fig. 9a shows the general interior arrangement. (By permission of Marconi's Wireless Telegraph Co., Ltd.) remain connected with the aerial. The arrangements will be understood from the diagram in fig. 9. The method is, however, not very successful and it cannot be said that two-way radiotelephony with a single aerial wire is a solved problem. On the other hand radio communication to and from aeroplanes up to 150 or 200 m. is now a thoroughly practical matter. The aerial is a wire about 230 ft. in length with a weight at the end which is unwound from a winch when the pilot wishes to communicate. The pilot generally wears a helmet with microphone transmitter opposite his mouth and two receivers over the ears. The transmission of speech from the aeroplane is more easy than reception owing to the great engine and propeller noises; nevertheless it is of immense use in connexion with air traffic as by it aircraft can be guided through the clouds to their destination, and the pilot informed of the conditions as regards fog or cloud at the landing station.

In May 1921 Marconi's Wireless Telegraph Co. carried out very successful demonstrations of practical radiotelephony on the two-wave system between Southwold in England and Zandvoort in Holland - a distance of 125 m. over the North Sea. The wave lengths used were 120 and 125 metres.

A very important use of the three-electrode valve is that of repeating speech currents from ordinary telephone wire circuits to wireless circuits and vice versa. Also this is perhaps the place to point out its extremely valuable qualities as a telephone repeater or relay for long wire circuits. Since the characteristic curve of the triode is nearly flat at the central part it follows that any irregular variations of grid potential are exactly copied by the corresponding variations of plate current. Hence if we connect the secondary terminals of a telephone transformer to the grid and filament of a valve and the primary terminals of another transformer to the plate and filament with plate battery inserted we shall have an arrangement called a thermionic repeater, which repeats and amplifies telephonic speech currents. If one or more such repeaters are inserted in a telephone line at intervals they will operate to neutralize the attenuation of the speech currents due to the resistance of the line and enable telephonic speech to be transmitted over larger distances without greater expenditure on the copper line conductor. Thus in the United States the long transcontinental telephone line from New York to San Francisco and Los Angeles has 13 such repeater stations in it, and speech is thus rendered possible over 4,000 m. of line.

In Oct. 1920 a remarkable feat was carried out in telephonic transmission by the aid of thermionic repeaters. A ship four hours out in the Atlantic spoke by wireless telephony to the mainland of the United States. The speech was then repeated into the transcontinental line from New York to Los Angeles and again repeated on to a radio circuit and delivered at Santa Catalina I. about 30 m. out in the Pacific. The speech transmission over this 4,000 m. was as perfect as over any exchange circuit in a large city. It is possible in this manner to speak to flying aeroplanes from the ordinary wire telephone of a town.

Long Distance Stations

The above improvements in generation and detection of electric waves have not only made radiotelegraphy from ship to ship and ship to shore a certain and indispensable aid to navigation, but have enabled a multitude of long distance radio stations to be established which can maintain communication over distances of several thousand miles. It is now generally agreed that this possibility is due not to true diffraction of these long electric waves round the earth but to the fact that the higher levels of the earth's atmosphere are in a state of permanent ionization due to sunlight or extra-terrestrial causes. This creates a high level reflecting layer which guides the wave round the earth. There are, however, peculiar difficulties and effects at times of sunrise and sunset. In the United Kingdom the Marconi Co. have a large station at Carnarvon, Wales, near Snowdon, which is in correspondence with another at Marion, N.J., United States, for transatlantic working. The British receiving station is at Towyn, about 60 m. from Carnarvon, to enable reception and transmission to be carried out simultaneously. The direct effect of the Carnarvon radiation on the Towyn receiving aerial is neutralized by a balancing aerial (see British Patent Specification No. 13020 of 1911 of G. Marconi). The aerial at Carnarvon is a Marconi directional one, 3,600 ft. in length and 400 ft. vertical height supported on io tubular steel masts. The wave length of the radiation is 14,000 metres. The system of wave generation is the so-called timed-spark of Marconi. A direct current high voltage dynamo keeps two sets of condensers charged, and by means of a pair of rotating wheels with studs on their peripheries these condensers are discharged alternately through the primary coil of a transformer, the secondary coil of which is inserted between the aerial and the earth. These two sets of oscillatory discharges are made to follow each other in step and in close sequence by means of a trigger disk discharger which times two discharges so as to constitute in effect a continuous oscillation. The signalling is conducted by switches worked by compressed air which are operated through a relay by electric currents from Towyn. The same company have also a radio station at Clifden in Ireland which corresponds with another at Glace Bay, Nova Scotia. The Clifden station employs large thermionic valves as generators of continuous waves.

The Imperial Wireless Telegraph Committee which reported to Parliament in June 1920 recommended the thermionic valve generator for the imperial stations of the British Empire on the ground that the capital outlay would be less than for arc or alternator stations. The high power radio stations in the United States comprise one at New Brunswick, N.J. which is equipped with Alexanderson alternators of 200 kilowatts capacity working on a wave length of 13,500 metres. The signals are made by means of a magnetic amplifier which is an independence coil, the impedance of which is varied by means of a small direct current which changes the per meability of the iron core (see Proc. of the Institute of Radio Engineers, United States, vol. iv, April 1916, p. ioi; a magnetic amplifier for radiotelephony).

The Radio Corporation of America began to build in 1921 a very large wireless station on Long Island which was to have 12 directive aerials, each 14 m. long, arranged radially around the station. The waves were to be generated by high frequency alternators. The station would cover an area of nearly 10 sq.m. and be the most powerful in the world (see The Engineering Supplement of The Times, Aug. 1920). Another large U. S. radio station is that at Tuckerton, N. J., containing a Goldschmidt 200 kilowatts high frequency alternator. The radio frequency machine is driven by an electric motor supplied from two direct current generators in Ward Leonard connexion. This station was erected to correspond with one near Hanover, Germany. In France there are four very large radio stations; one at Croix d'Hins near Bordeaux, which was erected by the U. S. army during the war to maintain communications with Washington, contained originally 400-500 kilowatts Poulsen arc generators but is now partly converted to an alternator station. The aerial is carried on 8 lattice towers 800 ft. high (see fig. to) FIG. to.-View of the large French radio station at Bordeaux erected by the American army during the World War for direct communication with the United States.

Another large French radio station at La Doua, near Lyons, with wave length of 12,000 metres is an arc station; a third exists at Nantes and a fourth is in Paris and employs the Eiffel tower to support its aerial wire.

The French Government began to erect in 1921 two large radio stations at St. Assise, near Paris, for European and world wide radiotelegraphy. These were to be equipped with Bethenod-Latour high frequency alternators and would have 1,500 kilowatts output.

In Germany there is a station at Nauen, near Berlin, which has a range of several thousand miles and wave length of 12,600 metres. This station like that at the Eiffel tower, Paris, sends out time signals at certain hours.

Broadly speaking, we may say that there were in 1921 about a dozen long distance radio stations in the world, which could signal to any part of the world by day or night, making use of wave lengths between 12,000 and 20,000 metres.

Bibliography. -FOr the full discussion the reader must be referred to special treatises on radiotelegraphy as mentioned below.

A. Fleming, The Principles of Electric Wave Telegraphy and Telephony (4th ed. 1919); R. Stanley, Text Book on Wireless Telegraphy (2nd ed. 2 vols. 1919); W. H. Eccles, Continuous Wave Wireless Telegraphy (1921); J. A. Fleming, The Thermionic Valve in Radiotelegraphy and Telephony (1918); W. H. Eccles, Wireless Telegraphy and Telephony (2nd ed. 1918); Bernard Leggett, Wireless Telegraphy, with special reference to the quenched-spark system (1921); J. A. Fleming, The Scientific Problems of Electric Wave Telegraphy, Cantor Lectures at the Royal Society of Arts (1919); also the following articles in the Proceedings of the Institute of Radio Engineers, United States, are authoritative and useful: vol. ii., 1914, 69, E. E. Mayer, " The Goldschmidt System of Radiotelegraphy "; vol. iii. 1915, 55, A. N. Goldsmith, " Radio Frequency Changers "; vol. iii., 1915, 215, E. H. Armstrong, " The Audion as Detector and Amplifier "; vol. iii., 1915, 261, I. Langmuir, " The Pure Electron Discharge "; vol. iv., 1916, 101, E. F. W. Alexanderson, " A Magnetic Amplifier for Radiotelephony "; vol. vii., 1919, 363, E. F. W. Alexanderson, " Simultaneous Sending and Receiving "; vol. viii., 1920, 3, 87, T. Johnson, " Naval Aircraft Radio "; vol. viii., 1920, 220, M. Latour, " Radio Frequency Alternators"; vol. viii., 1920, 263, E. F. W. Alexanderson, " Transoceanic Radio-communication "; vol. ix., 1921, 83, E. F. W. Alexanderson, " Central Stations for Radiocommunication "; the following are references to useful papers on the theory of the thermionic valve: The Physical Review, vol. xii., 1918, p. 171, H. J. Van der Bijl, " Theory of the Thermionic Amplifier "; Proc. Inst. Radio Engineers, United States, vol. vii., 1919, 97, 603, H. J. Van der Bijl on the theory and operating characteristics of the thermionic amplifier; Journal of the Institution of Electrical Engineers, London, vol. lviii., 1920, p. 65, C. L. Fortescue, " The Design of Multiple Stage Amplifiers using Three-electrode thermionic valves "; ibid., p. 670, B. S. Gossling, " Development of Thermionic Valves for Naval Use." For the discussion of the special difficulties introduced by the atmospheric electrical disturbances called "strays," which are vagrant electric waves produced by natural causes. the reader may be referred to a paper by Roy A. Weagant in the Proc. Inst. Radio Engineers, United States, 1919, vol. vii., 207, " Reception through Static and Interference." (J. A. F.)