5742 
 _B82e Ban gay 
 
 cop.l 
 The elementary 
 
 principles of 
 
 Southern Branch 
 of the 
 
 fniversity of California 
 
 Los Angeles 
 
 ae.
 
 J 
 
 This book is DUE on the last date stamped below 
 
 :C 1 1 1940 
 DEC 1 3 19*3 
 
 Form L-9-15m-8,'24 
 

 
 THE ELEMENTARY PRINCIPLES OF 
 WIRELESS TELEGRAPHY
 
 The 
 
 Elementary Principles 
 
 of 
 
 Wireless Telegraphy 
 
 by 
 
 R. D. Bangay 
 
 1 3 
 
 Wireless Press, Inc. 
 
 25 Elm Street 
 New Yor 
 
 7 'IS
 
 ALL RIGHTS RESERVED.
 
 Co 
 
 PREFACE TO FIRST EDITION 
 p 
 
 IN presenting this Handbook, the author has endeavoured 
 to explain, in the simplest possible manner, the theory 
 and practice of Wireless Telegraphy. 
 
 It has been his aim to make the subject intelligible to 
 persons who do not possess much technical knowledge, 
 and to be at the same time brief and accurate. 
 
 The book has been so arranged as to be useful as a 
 reference book on the subject for students and amateurs 
 in this special branch of electrical science. 
 
 Further and more complete explanations of the various 
 phenomena described can be obtained from the standard 
 scientific works on the subject, but it has been the object 
 of the author to deal with the subject clearly and simply 
 without going too deeply into the many highly technical 
 problems involved. 
 
 R. D. B.
 
 PBEFACE TO SECOND EDITION 
 
 WITH the object of increasing the usefulness of this 
 Handbook, the- author has extended-its scope without 
 going any more deeply into the technical side of the 
 subject. 
 
 Since the book has been used largely in the training 
 of Telegraphists who are frequently called upon to take 
 sole charge of complete Wireless Telegraph Installations, 
 the author has endeavoured to cover all parts of the 
 transmitting and receiving apparatus in such a way as 
 to give the student a sound working knowledge of the 
 apparatus entrusted to his care. 
 
 For the convenience 6f the student, the new edition 
 is divided into two parts. Part I. contains, in addition 
 to the matter published in the first edition, and now 
 revised, a good deal of further information regarding 
 Receivers and Aerials. In Part II. the component 
 parts of a Transmitter are explained separately, and 
 the theory of the condition of resonance under which 
 they can most effectively be combined, and to which 
 each part should be adjusted to form an efficient 
 transmitter, is fully discussed. 
 
 R. D B.
 
 
 CONTENTS 
 
 PAOE 
 
 MORSE CODE . . . . . ix 
 
 SYMBOLS USED IN DIAGRAMS . . . xi 
 
 ELECTRICITY AND MAGNETISM .... 3 
 
 Electro-statics Conductors and insulators Static in- 
 duction The Condenser Electro-dynamics Electric 
 circuits. 
 
 Units of electricity The Coulomb The Ampere The 
 Volt The Ohm The Henry The Farad The Joule 
 The Watt Ohm's law. 
 
 MAGNETISM ....... 24 
 
 Electro - magnetism Electro - magnetic induction 
 
 Mutual induction. 
 
 The construction of the induction coil Production of 
 
 electricity by chemical action Accumulators. 
 
 THE PRINCIPLES OF WAVE MOTION . . . .43 
 
 Properties of waves Communication by wave-motion 
 Measurements of waves. 
 
 PRESSURE WAVES ... 4 
 
 Aether waves Communication by means of aether 
 waves. 
 
 PRODUCTION OF WAVES 52 
 
 Production of height waves Production of pressure 
 waves Production of electric waves. 
 
 PRODUCTION OF HIGH-FREQUENCY OSCILLATIONS 65 
 
 Oscillatory circuits Energy and power in oscillatory 
 circuits Power in oscillatory circuit Open and closed 
 oscillatory circuits Variation of wave-lengths of open 
 oscillatory circuits To increase the wave-length of an 
 aerial To reduce the wave-length of an aerial Variation 
 of wave-lengths of closed oscillatory circuits. 
 
 PRODUCTION OF OSCILLATING CURRENTS IN AN AERIAL . 84 
 Direct excitation of the aerial.
 
 viii WIRELESS TELEGRAPHY 
 
 PAGE 
 
 COUPLED OSCILLATORY CIRCUITS . . . .89 
 
 Factors limiting the power in oscillatory circuits The 
 auto-jigger Reaction of secondary on primary Re- 
 sultant wave-lengths of coupled circuits Calculation of 
 the degree of coupling Methods of varying the coupling 
 between two oscillatory circuits. 
 
 THE WAVEMETER . . . . . .111 
 
 The oscillatory circuit of a wavemeter The" detector" 
 circuit of a waveraeter The use of crystals Construction 
 of an adjustable condenser. 
 
 WIRELESS TELEGRAPH RECEIVERS . . ... 118 
 
 Essentials of a receiver Methods of detecting the 
 oscillating currents The potentiometer Method of 
 applying the potentiometer to the" crystal The two- 
 circuit receiver Proportion of inductance and capacity 
 in secondary oscillatory circuit Characteristic curve of 
 crystal. 
 
 The telephone receiver High resistance telephones 
 Rectifying properties of carborundum Relation between 
 the spark frequency of the transmitter and sound pro- 
 duced in the telephones of receiver To tune a receiver. 
 
 THE TUNING BUZZER . . . . . . 146 
 
 THE ELECTROLYTIC DETECTOR . . . .151 
 
 THE MAGNETIC DETECTOR . . . . .159 
 
 " ATMOSPHERICS "...... 165 
 
 AERIALS 167 
 
 Shape of an aerial Size of an aerial Height of an aerial 
 
 The advantage of using aerials of a large capacity The 
 
 length of an aerial. 
 
 DISTRIBUTION OF POTENTIAL AND CURRENT ALONG AERIALS 177 
 
 Distribution of current in an aerial Effect on current 
 and voltage distribution of connecting an inductance or 
 capacity in series with an aerial Harmonics. 
 
 MASTS 189 
 
 Strain on masts Buckling of masts Mast stays. 
 THE INSULATION OF AERIALS . . . .199 
 
 Aerial insulators Earths. 
 
 INDEX . . 209
 
 THE EUROPEAN OR CONTINENTAL MORSE CODE 
 
 a 
 
 n 
 
 
 or a 
 
 n 
 
 Numi 
 
 b 
 
 
 
 i 
 
 c 
 
 6 
 
 2 
 
 ch 
 
 p 
 
 3 
 
 d 
 
 6 ~* 
 
 q 
 
 j- _ , i _ 
 
 4 
 
 
 s 
 
 5 " 
 6 
 
 f 
 
 t . 
 
 7 
 
 g 
 
 u 
 
 8 
 
 h 
 
 u 
 
 9 
 
 i 
 
 v 
 
 o 
 
 j 
 
 w 
 
 
 k 
 
 X 
 
 
 1 
 
 y 
 
 
 Punctuation Marks
 
 SYMBOLS USED IN DIAGRAMS OF WIRELESS 
 TELEGRAPHY CIRCUITS 
 
 3. 
 
 8. 
 
 -I'l' 
 
 6. 
 
 \. Conductor. 
 
 2. Conductors crossing. 
 
 3. Conductors connected. 
 
 4. Cell. 
 
 5. Battery. 
 
 6. Resistance coil. 
 
 7. Inductive winding. 
 
 8. Two coils having mutual 
 
 inductance. 
 
 9. Condenser. 
 10. Switch.
 
 SYMBOLS USED IN DIAGRAMS OF WIRELESS 
 TELEGRAPHY CIRCUITS 
 
 JZ. 
 
 16. 
 
 17. 
 
 IB. 
 
 o o 
 
 D 
 
 ^l/ 
 
 i 
 
 J5. 
 
 \\. Variable resistance. 
 
 12. Variable inductance. 
 
 13. Variable condenser. 
 
 14. Direct current dynamo. 
 
 15. Alternating current dynamo. 
 
 16. Manipulating key. 
 
 17. Spark-gap. 
 
 18. Aerial wire or antenna. 
 
 19. Earth connection. 
 20 Crystal detector. 
 21. Telephone, receiver.
 
 ELEMENTARY PRINCIPLES OF 
 WIRELESS TELEGRAPHY 
 
 THE object of this book is to instruct the reader in 
 the principles underlying the construction of modern 
 Wireless Telegraphy apparatus, more especially as 
 applied to small stations. 
 
 Although the same principles apply equally to 
 commercial stations having a range of communication 
 up to thousands of miles, it is obvious that the method 
 of applying these principles will depend to a certain 
 extent upon the size of the station. Thus factors which 
 require important consideration when dealing with 
 powerful plant will fall to comparative insignificance 
 when dealing with small stations. 
 
 Wireless Telegraphy is a special application of 
 electrical phenomena, therefore an elementary know- 
 ledge of the subject of electricity and magnetism is 
 absolutely essential before full advantage can. be taken 
 of a study of the principles of Wireless Telegraphy. 
 
 This book is written on the assumption that an 
 elementary knowledge of electricity and magnetism is 
 possessed by the reader, but in order to assist the un- 
 initiated, and for the purpose of reference, we have, 
 in the first part of the book, briefly described the 
 
 1 B
 
 2 WIRELESS TELEGRAPHY 
 
 various points of importance, but the information given 
 should be supplemented by a study of any of the 
 standard text-books on the subject 
 
 Electricity is the name given to that which causes 
 all electrical phenomena. 
 
 Electricity is invisible and intangible, although 
 both visible and tangible effects can be produced' by it. 
 Its exact nature, therefore, can only be imagined, 
 but its effect upon matter has been carefully studied, 
 and from the study and classification of these pheno- 
 mena the laws governing the effects of electricity 
 have been deduced. 
 
 Perhaps the best view - point to take is to 
 consider electricity as an agent or medium by which 
 work or energy can be transmitted from one point to 
 another. 
 
 For instance, let us take, as an example, an electric- 
 power installation In one building we have an engine 
 driving a dynamo, and in another building, which may 
 be far removed, we have an electric motor driving, say, 
 a circular saw used for cutting wood. 
 
 Energy is put into the boiler of the engine m the 
 form of Tieat by burning coal. The boiler converts this 
 heat energy into steam pressure carried along pipes to 
 the engine, where in turn it is converted by the engine 
 and dynamo into electrical energy. The electrical 
 energy is carried along wires to the motor in the other 
 building, and the motor converts the electrical energy 
 into mechanical energy by turning a circular saw, and 
 the energy is then used to cut through wood. 
 
 Now it is not the electricity which cuts the wood, but 
 the energy which is put into the boiler of the engine 
 Electricity is simply a convenient agent by which that
 
 WIRELESS TELEGRAPHY 3 
 
 energy can be transferred from the boiler-room to the 
 work-room. 
 
 It is on account of the fact that by this agency 
 energy can be transferred conveniently and cheaply 
 over a great or small distance, and can by means of 
 suitable apparatus be converted into almost any desired 
 form of energy such, for example, as light, heat, 
 motive power, etc. that electricity is so extensively 
 used for all purposes. 
 
 ELECTRICITY AND MAGNETISM 
 
 1. When a charge of electricity rests on the surface 
 of any substance, such as amber, glass, etc., the charge 
 of electricity is known as a static charge, and the 
 study ^of the effects of these charges is known as electro- 
 statics. 
 
 2. When a charge of electricity passes through a sub- 
 stance, such as copper, silver, etc., the charge is known 
 as an electric current, and the study of the effects of these 
 currents is known as electro-dynamics. 
 
 ELECTRO-STATICS 
 
 3. If we take a piece of amber and rub it with a piece 
 of silk, we find that the amber has acquired the property 
 of attracting very light objects, such as fragments of 
 paper, cork, cotton-wool, or pith balls, and that if these 
 objects actually touch the amber which is attracting 
 them, they are then repelled. 
 
 These attractions and repulsions are due to a static 
 charge of electricity, which has been generated by the
 
 \ 
 
 4 WIRELESS TELEGRAPHY 
 
 friction with the silk, and which is resting on the surface 
 of the amber. 
 
 4. If then, for convenience, we suspend a small pith 
 ball by a silk thread, as shown in Fig. 1, and approach 
 it with an electrified amber rod, we shall see that the pith 
 ball will first fly towards the rod, and that immediately 
 it touches the rod it will be repelled. 
 
 5. This is because by contact with the rod the pith 
 ball has itself become charged, and as long as both the 
 pith bail and the amber rod retain their charges, repulsion 
 
 will take place whenever they are 
 brought near each other. 
 
 6. If, now, instead of electrifying 
 an amber rod with a piece of silk, 
 we electrify a piece of sealing-wax 
 by rubbing it with a piece of flannel 
 or fur, and we approach the already 
 electrified pith ball with the electri- 
 fied sealing-wax, we find that instead 
 of repelling the ball as the electrified 
 FlQ - * amber rod does, it attracts it, al- 
 
 though as soon as they come in contact with each other 
 the pith ball is again repelled, and after being repelled 
 by the electrified sealing-wax will once more be attracted 
 by the electrified amber. 
 
 7. When this simple experiment is analysed, it is 
 found that there are two kinds of electrification one 
 produced by rubbing amber with silk, and the other 
 produced by rubbing sealing-wax with fur. 
 
 8. In order to distinguish between the two, that 
 produced by rubbing amber'with silk is called a positive 
 charge ( -f ), and that produced by rubbing sealing-wax 
 with fur is called a negative charge ( - ).
 
 WIRELESS TELEGRAPHY 5 
 
 9. By simple experiments it can be shown that 
 neither charge is ever produced alone, for when amber is 
 rubbed with silk, although a positive charge is produced 
 on the amber, an equal negative charge is produced at 
 the same time on the silk, and vice versa when sealing- 
 wax is rubbed with fur. 
 
 10. From these and other similar experiments the 
 following laws may be deduced : 
 
 (1) That when either a positive or a negative charge is 
 produced, an equal and opposite charge is also produced. 
 
 (2) That like charges repel one another, and unlike 
 charges attract one another. 
 
 (3) That when an electrified body touches an unelectrified 
 body, the latter becomes charged to the same " polarity " 
 as the former. 
 
 (4) That when an electrified body touches an oppositely 
 electrified body, if the two charges are equal their electri- 
 fication is destroyed, and they are then said to be discharged ; 
 but if the charge on one body is greater than that on the 
 other, their electrification is only partially destroyed, and 
 both bodies become charged to the same polarity as that 
 of the greater charge. 
 
 CONDUCTORS AND INSULATORS 
 
 11. The bodies which we have been electrifying do 
 not conduct electricity, but they resist or oppose the 
 passage of electricity through them, and it is for this 
 reason that the charge produced on them rests on their 
 surface. 
 
 12. When a charge of electricity is applied to a metal, 
 the electricity immediately flows through it, and for this 
 reason metals are called conductors of electricity.
 
 6 WIRELESS TELEGRAPHY 
 
 13. All metals are conductors, those most commonly 
 used in electrical apparatus for- this purpose being 
 copper, brass, aluminium, iron, etc. To a much lesser 
 extent the human body and water (except the purest 
 distilled water) are conductors. 
 
 14. The substances which will not conduct electricity 
 are called insulators, and for this purpose the chief 
 materials used in electrical apparatus are amber, sealing- 
 wax, glass, porcelain, ebonite, mica, silk, rubber, oils, 
 dry wood, string, and cotton. 
 
 15. An important point to bear in mind is that none 
 of the substances mentioned as conductors are perfect 
 conductors ; that is to say, none of them will carry a 
 current of electricity without some loss of energy due to 
 " friction " or " resistance " Some of them, however, 
 are better conductors than others ; for instance, copper 
 is a better conductor than iron, and for this reason there 
 is less loss of energy due to " resistance " in copper than 
 in iron. 
 
 16. Similarly no substance is a perfect non-con- 
 ductor, or insulator. There will always be some loss due 
 to leakage, although by using a suitable material this 
 leakage can be reduced to a negligible quantity. 
 
 STATIC INDUCTION 
 
 17. When an electrified conductor is brought near 
 another conductor which has not been electrified, an 
 electric charge will be induced in the latter. 
 
 This effect is known as Static Induction. 
 
 18. The charge which is induced in the non-electrified 
 conductor is not a permanent charge, but depends 
 entirely for its existence upon its proximity to the
 
 WIRELESS TELEGRAPHY 
 
 electrified conductor This is illustrated in Fig. 2, 
 where A is a plate df metal which has been permanently 
 charged by touching it with an electrified amber rod, and 
 B is a similar plate of metal which has not been charged 
 Both plates are supported by pillars of glass, or other 
 insulating material, to prevent their charges being con- 
 ducted to earth. 
 
 19. As A is brought nearer and nearer to B a stronger 
 and stronger charge is induced m the latter, and as A is 
 taken farther away from B the mduced charge in B 
 becomes weaker. All the time, 
 
 however, the plate A retains the 
 original charge which was given 
 to it. 
 
 20. The range of space over 
 which the electrified plate A has 
 the power of inducing a charge 
 in B is called the Electro-static 
 Field. 
 
 21. If the two plates are 
 brought together so that they 
 
 touch one another, then the permanent charge in A 
 flows into B, and the charge is divided equally between 
 the two plates. The plate B will then retain this charge, 
 even when taken away from the influence, or electro- 
 static field, of A. 
 
 22. We have said that the strength of the charge 
 mduced by an electrified conductor in another con- 
 ductor depends upon the distance between the two. 
 The strength of the charge also depends upon the nature 
 of the substance between the two bodies, which must be 
 a non-conductor. 
 
 This substance is called a dielectric.
 
 8 WIRELESS TELEGRAPHY 
 
 23 All dielectrics are* non-conductors. 
 
 24. The facility with which a dielectric allows static 
 induction to act through it is called its Inductive 
 Capacity. 
 
 If the space between the plates A and B is filled 
 by glass, it is found that a much stronger charge is 
 induced in B than when the same space is filled with air. 
 Therefore we may say that the inductive capacity of 
 glass is greater than the inductive capacity of air 
 
 25.' A simple mechanical analogy of these phenomena 
 
 -Leather 
 
 can be made by comparing the electrical inductive 
 capacity of a dielectric with the mechanical extensibility 
 of a material. 
 
 If we stretch two pieces of different material, such as 
 a strip of leather and a strip of rubber, each having an 
 equal thickness, between two fixed points, as shown in 
 Fig. 3, and we place on each of them a weight of, say, 
 1 lb., we find that the rubber stretches a great deal more 
 than the leather, and therefore we say that the exten- 
 sibility of rubber is greater than that of leather, just as 
 we said that the inductive capacity of glass was p ~ater 
 than that of air.
 
 WIRELESS TELEGRAPHY 9 
 
 26. The same analogy illustrates the effect of increas- 
 ing the thickness of the dielectric, for if we increase the 
 thickness of the rubber strip in the experiment just 
 described, although we are using a material of the same 
 extensibility as before, yet owing to the fact that it is 
 thicker, the same weight will not stretch the rubber to 
 the same extent as before. Similarly, if we increase the 
 distance between the two plates A and B in Fig. 2, or, 
 in other words,, increase the thickness of the dielectric, 
 the efiect of the static induction is reduced, although the 
 dielectric has the same Inductive Capacity as before. 
 
 THE CONDENSER 
 
 27. Static Induction is the principle underlying the 
 construction of a " condenser." A simple type of 
 condenser consists of a plate of glass, or some other 
 dielectric (vide paragraph 23), covered with tin-foil or 
 other conductor on either side. The tin-foil merely 
 acts as a means of distributing any applied electrical 
 pressure uniformly over the surface of the dielectric. 
 
 28. The property a condenser has of holding a large 
 or small charge of electricity is called its " Capacity." 
 
 This property of capacity can best be illustrated by 
 comparing it with the analogous mechanical property 
 of springiness or flexibility. 
 
 29. If a mechanical force is applied to a spiral spring, 
 the spring will be extended to a distance X, shown in 
 Fig. 4, until it exerts a force exactly equal and opposite 
 to the applied force. Similarly, if an electrical force 
 (vide paragraph 36) be applied to a condenser, the 
 dielectric of the condenser will be strained electrically 
 until the condenser exerts an electrical pressure exactly 
 equal and opposite to the force applied to it
 
 10 
 
 WIRELESS TELEGRAPHY 
 
 30 Taking the mechanical case of a spring, it will be 
 observed that a movement or an extension of the spring 
 must take place before it exerts an opposite force, and 
 further, that the amount of this movement. < t the 
 distance X in Fig 4, which takes place for a given 
 applied force, will depend upon the flexibility 01 
 springiness of the spring 
 
 31. Similarly, in the case of a condenser . owing to the 
 straining of the dielectric, we get a certain amount oi 
 electricity forced into the con- 
 denser when an electrical pres- 
 sure is applied to it A current 
 of electricity must flow into the 
 condenser before it exerts an 
 opposite force, and further, the 
 quantity of electricity which flows 
 into a condenser for a given ap- 
 plied force depends upon the 
 capacity of that condenser 
 
 32 An analogy which is per- 
 haps easier to understand is 
 shown in Fig 5, where a steel nozzle is shown, over 
 the end of which is fastened a thin india-rubber 
 cap, which under normal conditions will lie evenly 
 across the end of the nozzle If now a pressure be 
 applied to this india-rubber by connecting the nozzle 
 to a water-tank the rubber will be bulged out by the 
 pressure of the water, as shown in Fig. 6. until the 
 rubber exerts a pressure on the water equal and opposite 
 to the pressure exerted by the tank 
 
 Obviously the expansion or the bulging of the india- 
 rubber allows it to contain a certain quantity of water 
 when a force is applied to it so that we can say that, 
 
 FIG. 4
 
 WIRELESS TELEGRAPHY 
 
 11 
 
 owing to the stretching of the cap, a certain quantity 
 of water will be forced into it, and the quantity of water 
 which it will contain, for a given pressure, will depend 
 upon the flexibility of the cap. 
 
 33. Similarly, if an electrical force be applied to a con- 
 denser, owing to the electrical straining of the dielectric 
 of the condenser, a certain quantity of electricity will 
 be forced into it, until the condenser exerts an electrical 
 pressure equal and opposite to the pressure applied to it. 
 
 Fio. 5. 
 
 Fio. 6. 
 
 The quantity of electricity which will be forced, into a 
 condenser for a given applied pressure will depend upon 
 the capacity of that condenser. 
 
 Thus we may say that the capacity of a condenser 
 is analogous to the flexibility of the india-rubber cap, 
 or, in the first experiment described, to the flexibility 
 of the spring ; and we can take the flow of electricity 
 into a condenser as being analogous to the flow of water 
 into the cap, or, in the first experiment described, to 
 the movement of the spring. 
 
 34. Now just as the flexibility of the spring (or of 
 the cap, as the case may be) depends upon three things,
 
 If WIRELESS TELEGRAPHY 
 
 namely, (1) the sifce of the spring (or cap) to begin with, 
 i.e its length and diameter, (2) the thinness of the 
 material of which it is made, and (3) the mechanical 
 extensibility of that material, so does the capacity of a 
 condenser depend upon three analogous factors, namely, 
 (1) the area of the plate forming the condenser, (2) the 
 thinness of the dielectric, and (3) the inductive capacity 
 of the dielectric. 
 
 35 An important point to grasp is that the property 
 the condenser has of holding electricity is due to the 
 electrical straining of the dielectric and not to any 
 compression of the electricity, just as the property 
 an india-rubber cap has of holding water is due to the 
 straining of the rubber and not to any compression of the 
 water. In the case of the rubber cap, this straining has a 
 visible effect, inasmuch as it increases the size of the cap , 
 but in the case of an electrical condenser no such obvious 
 result can be seen, although electrical instruments will 
 indicate that when a pressure is applied to the condenser 
 a current of electricity flows into it until it is charged. 
 
 The effect of capacity in an electrical circuit is in 
 all respects similar to the effect of a spring in a mechanical 
 system, and the analogy is extremely helpful in studying 
 the effects of capacity in circuits such as those used in 
 Wireless Telegraph installations. 
 
 ELECTRO-DYNAMICS 
 
 36 An electric current is a flow of electricity 
 
 In order to produce an electric current, it is first 
 necessary to exert a difference of electrical pressure 
 between two bodies, or between two parts of tke same 
 body This difference of pressure is called the Eleotro- 
 motive Force.
 
 WIRELESS TELEGRAPHY 
 
 13 
 
 37. Ip order that an electric current will flow, how- 
 ever, it is necessary that the body, or bodies, across 
 which the difference of electrical pressure is 
 exerted is a conductor of electricity. 
 
 A simple mechanical analogy can be 
 made to illustrate this. 
 
 A long pipe, as shown in Fig. 7, is filled 
 with water. The pipe represents a con- 
 ductor, and the water illustrates the elec- 
 tricity in the conductor. Both ends of the 
 pipe are held upwards on a level 'with one 
 another, so that normally there is no difference in pres- 
 sure acting at each end of the tube, and therefore 
 the water will not flow through the tube. 
 
 If, however, we exert a pressure at one end of 
 the'pipe by blowing down it, or by increasing the height 
 
 FIG. 7. 
 
 FIG. 8. 
 
 of one end above the other, or, better still, by connecting 
 a tank of water to it which is situated at a higher level 
 than that on which the experiment is being carried out,
 
 14 WIRELESS TELEGRAPHY 
 
 as shown m Fig. 8, then the water will immediatelv flow 
 through the pipe. 
 
 38 By connecting the tank to one end only of the 
 pipe, we exert a difference of pressure on the two ends of 
 the pipe, but if we connect the tank simultaneously to 
 both ends of the pipe, then there is no difference of 
 pressure on the two ends of the pipe, and consequently 
 no water will flow through it. 
 
 39 As the water represents electricity, the flow of 
 water represents an electric current (vide paragraph 36). 
 
 40. The analogy of water flowing through a pipe, 
 although, useful in some cases, is not a good one to take 
 generally as representing the flow of electricity in a 
 conductor, and it is apt to be very misleading, more 
 especially when studying the effects of capacity and 
 inductance It is much better to take the mechanical 
 effect of " movement " to represent the flow of electricity, 
 and to take the nature of the body which moves to 
 represent the features of an electrical circuit. 
 
 Thus, if we take the movement of a shaft to represent 
 the flow of electricity, we can take the bearings in which 
 it moves to represent the conductor which carries ihe 
 electric current 
 
 In order to cause the shaft to move in its bearing 
 we must apply a force or pressure to the shaft just as 
 an electromotive force must be applied to a conductoi 
 in order to cause electricity to flow in it. 
 
 CIRCUITS 
 
 41. A circuit is a path composed of a conductor, or 
 conductors, through which an electric current flows 
 from one point in it around the conducting path back 
 to the point from which it started.
 
 WIRELESS TELEGRAPHY 
 
 15 
 
 An electric source, such as a battery, or dynamo, 
 is generally included in a circuit, the function of the 
 electric source being to produce a difference in pressure 
 or an electromotive force in the circuit. 
 
 Battery. 
 
 B 
 
 FIG. 9. 
 
 42. Different parts of a circuit can be connected in 
 parallel or in series. 
 
 B 
 
 43. Thus, when two conductors A and B are joined in 
 a circuit as shown in Fig, 9, they are said to be joined in 
 parallel, or if they are joined as shown in Fig. 10 t they 
 are said to be joined in series. 
 
 44. When conductors are joined in parallel, only part
 
 16 WIRELESS TELEGRAPHY 
 
 of the total current flows through each conductor. 
 When they are joined in series, the whole current 
 passes through each conductor successively. 
 
 45. When two cells (a cell is a source of electric 
 
 pressure, and is de- 
 scribed later) are con- 
 nected in a circuit, 
 as shown diagram- 
 matically in Fig. 11, 
 they are said to be 
 connected in parallel, 
 and only part of the 
 total current in the 
 Flo n circuit will flow 
 
 through each cell. 
 
 46. When they are connected, as shown in Fig. 12, 
 they are said to be connected in series, and the whole 
 
 FIG. 12. 
 
 current passing through the circuit will flow through 
 each cell. 
 
 UNITS OF ELECTRICITY 
 
 47. In order to measure and define the different 
 electrical factors of a circuit, certain practical standards, 
 or units, have been adopted. 
 
 It is not necessary for the purpose of this book to 
 explain how these units have been arrived at. It is 
 sufficient to describe the particular quality, or property, 
 which each represents and the relation which one bears 
 to another.
 
 WIRELESS TELEGRAPHY 17 
 
 48. The unit of quantity is one Coulomb. 
 
 current one Ampere. 
 ",, ,, electromotive force 
 
 or pressure one Volt. 
 
 ,, ,, resistance ,, one Ohm. 
 
 ,, inductance one Henry. 
 
 capacity one Farad. 
 
 energy ,, one Joule. 
 
 power one Watt. 
 
 THE COULOMB 
 
 50. The Coulomb is the electrical unit of quantity, 
 and can be compared with the water unit of quantity, 
 namely, " a gallon," or, better, with the mechanical unit 
 of rotary motion, namely, " one revolution " (vide para- 
 graph 40). 
 
 THE AMPERE 
 
 51. The Ampere is the electrical unit of current. 
 When a current of water flows through a pipe, the 
 
 amount of flow can be defined by stating how many 
 " gallons per second " are flowing. 
 
 Similarly in an electrical circuit, when a current of 
 electricity flows through a conductor, the rate of flow 
 can be defined by stating how many coulombs per 
 second are flowing. 
 
 Again, when a shaft rotates in a bearing the amount 
 of rotation or " speed " can be defined by stating how 
 many revolutions per second it is making, and in this 
 case one revolution per second is the unit of speed. 
 
 52. In an electrical circuit one " ampere " represents 
 a flow of one coulomb of electricity per second. 
 
 THE VOLT 
 
 53. The Volt -is the unit of electrical pressure,
 
 18 WIRELESS TELEGRAPHY 
 
 variously described as " difference of potential," or 
 " electromotive force " (E.M.F.). 
 
 54. It can be compared with the' practical unit of 
 mechanical force, namely, the pound. 
 
 The flow of water, that is, the number of gallons per 
 hour that will flow through a pipe of given length, size, 
 and shape, will depend upon the number of pounds of 
 pressure applied at one end of the pipe, or to put it 
 more correctly, upon the difference in the number of 
 pounds acting on the two ends of the pipe. 
 
 Again, the speed at which a shaft will rotate in a 
 given bearing will depend upon the twisting force or 
 " torque " applied to the shaft. 
 
 55. Similarly, the now of electricity, or the number 
 of amperes that will flow through a conductor of given 
 length, size, and shape, will depend upon the difference in 
 the number of volts acting at each end of the conductor. 
 
 THE OHM 
 
 56. The Ohm is the unit of resistance. 
 
 57. Resistance can be compared with the mechanical 
 property of friction, for example, with the friction 
 between the water and the inside of a pipe when the 
 water is flowing through the pipe, or the friction between 
 a shaft and its bearings. 
 
 58. Just as friction opposes the flow of water through 
 a pipe or the rotation of the shaft, so does resistance 
 oppose the flow of electricity through a conductor. 
 
 59. A conductor having a resistance of one ohm unll 
 require an electromotive force of one volt to force a 
 current of one ampere through it. 
 
 THE HENRY 
 
 60. The Henry is the unit of inductance.
 
 WIRELESS TELEGRAPHY 19 
 
 61. Inductance is that quality in a circuit which tends 
 to oppose any change in the flow of electricity. It must 
 not be confused with " resistance," which opposes the 
 flow of electricity. 
 
 62. It can best be described by comparison with the 
 mechanical property of mass, as its effect in an electrical 
 circuit is analogous to the effect of the inertia and 
 momentum of a heavy body in motion. 
 
 All bodies when stationary show a tendency to oppose 
 being put in motion, or if they are already in motion, 
 to oppose being accelerated. This tendency is called 
 " inertia." Similarly, all bodies when in motion show 
 a tendency to oppose being stopped, or having their 
 speed reduced. This tendency is called " momentum." 
 
 63. It is well known that it takes a considerable 
 time for an engine with a heavy fly-wheel to get up full 
 speed. This is due to the inertia of the fly-wheel. Also 
 an engine running at full speed takes a considerable 
 time to be brought to a standstill. This is due to the 
 momentum of the fly-wheel. 
 
 64. In the same way there is a tendency in a circuit 
 to oppose any increase, or decrease, in the current 
 flowing through it. This quality is called Inductance. 
 
 If an E M.F. be applied to a circuit possessing induct- 
 ance, the current flowing as a result of the E.M.F. will 
 only gradually grow, and the greater the inductance, the 
 slower the rate of growth. Again, if, when the current 
 is flowing through a circuit possessing inductance, the 
 E.M.P. which is making it flow be suddenly removed, 
 the current will only gradually stop flowing, unless of 
 course the circuit be " broken " or interrupted. 
 
 Tli us it will be seen that the effect of inductance in 
 a circriit on any current flowing through' it is exactly
 
 20 WIRELESS TELEGRAPHY 
 
 similar to the effect of inertia in a body on any 
 movement of that body. 
 
 65. Inductance is really due to the magnetic field 
 produced by the current in a circuit (see paragraph 85), 
 and the amount of the inductance depends upon the 
 strength of the magnetic field thus produced, just as the 
 amount of the inertia, or momentum, of a fly-wheel 
 depends upon the weight of that fly-wheel. 
 
 66. It will be shown later that the amount of magnetic 
 field produced by a circuit, and therefore the inductance 
 of a circuit, depends upon its form ; for instance, the in- 
 ductance of a given length of wire will be far greater 
 if that wire is wound into a coil than if it is stretched 
 o^it straight. 
 
 Similarly, the momentum of a fly-wheel depends upon 
 its shape as well as its weight . thus a fly-wheel two feet 
 in (pameter, weighing ten pounds, will have a very much 
 greater momentum or inertia than a fly-wheel one foot 
 in diameter, also weighing ten pounds 
 
 67. One great difference between the effect of resist- 
 ance and that of inductance in a circuit is that resistance 
 absorbs energy and dissipates it in the form of heat, 
 just as friction absorbs mechanical energy and dissipate? 
 it in the form of heat, whereas inductance only stores up 
 energy when the current is increasing, and gives its energj 
 back when the current is decreasing, just as a fly- wheel 
 stores up energy when its speed is increased and gives 
 back its energy when the speed is decreased. 
 
 When a circuit has an inductance of one henry, the 
 current flowing through that circuit will change by one 
 ampere when -a difference^/ potential of one volt has been 
 applied for one second 
 
 The microhenrv is sometimes used as a more con-
 
 WIRELESS TELEGRAPHY 21 
 
 venient unit when the circuits under consideration have 
 very small inductances. One microhenry is one- 
 millionth part of a henry. 
 
 THE FARAD 
 
 68. The Farad is the unit of capacity. 
 
 69. We have already described in paragraph 28 that 
 capacity is the property which a condenser has of holding 
 a cercain quantity of electricity, and we compared this 
 property with the mechanical flexibility of a spring 
 
 If we take the amount in inches that a spring will 
 extend when a given force of one pound is applied to it 
 as being a measure of its flexibility, we could specify 
 this quality of any spring by stating how many inches 
 or what fraction of an inch theoretically it would expand 
 for this given pressure. Thus by coining a word, "pound- 
 inch," we could say that if a spring was such that it 
 would expand one inch when one pound of pressure was 
 applied to it, it has a flexibility of one " pound-inch." 
 
 70. Similarly, the electrical unit of capacity is a 
 measure of the quantity of electricity which will flow into 
 a condenser when a pressure of one volt is applied to it. 
 Thus if a condenser be of such dimensions that it will hold 
 one coulomb of electricity when a pressure of one volt is 
 applied across it, it will have a capacity of one " Farad." 
 
 71. A condenser sufficiently large to hold a charge of 
 one coulomb of electricity at a pressure of one volt 
 would have to be of enormous dimensions, and therefore 
 a farad is too large a unit for practical convenience, and 
 the microfarad is therefore usually adopted in its plaro, 
 a microfarad being one-millionth part of a farad 
 
 72 An important point to grasp is that although 
 energy is expended in charging up a condenser, this
 
 22 WIRELESS TELEGRAPHY 
 
 energy is in reality only stored up by the condenser, and 
 is available for use by discharging that condenser through 
 a useful channel; just as in the case of a spring although 
 energy is expended in expanding or compressing a 
 spring, that energy is only stored up by the spring, and 
 is available for use by discharging the spring in a useful 
 way. For example, take the case of an air-gun energy 
 is stored in the gun by compressing a spring, and this 
 energy is then available for driving a shot against the 
 friction of fehe air when the spring is released 
 
 Later on in this book we shall show many examples 
 of how energy stored up in a condenser is made to do 
 useful work by discharging that condenser through 
 suitable circuits. 
 
 THE JOULE 
 
 73 V The Joule is the practical unit of electrical 
 energy or work 
 
 la order to cause a current of electricity to flow in a 
 circuit, energy or work must be expended. 
 
 The same rule applies to all matter For example, 
 m order to cause a body to move, energy or work must 
 be expended The mechanical unit of work is the 
 " foot-pound," and can be defined as follows 
 
 If a force of one pound is used to move a body, the 
 amount of work expended in moving that body a distance 
 of one foot is one foot-pound. For instance one foot- 
 pound of energy is expended in lifting a body weighing 
 one pound a foot off the ground, because the force of one 
 pound is being exerted on it throughout the 'distance it 
 is being lifted 
 
 The electrical unit of work is as we have already 
 stated, the joule, and can be denned as follows .
 
 WIRELESS TELEGRAPHY 23 
 
 74. // a force of one volt is used to cause an electric 
 current to flow through a circuit, one joule of work has been 
 expended when one coulomb of electricity has flowed. 
 
 From this definition of a joule in terms of quantity 
 and pressure, it follows from the fact that one ampere of 
 current is one coulomb per second, that a joule is also 
 the amount of energy expended during one second of 
 time in causing one ampere to flow through a resistance 
 of one ohm (vide paragraph 59). 
 
 THE WATT 
 
 75. The Watt is the electrical unit of power. Power 
 is the work done per unit time, or the rate of doing 
 work. One watt is the power required to do one joule 
 of work per second. 
 
 76. Now, since a flow of one coulomb per second is 
 one ampere, it follows from the definition of a joule that 
 one watt of power is expended when one volt is used 
 to cause a current of one ampere to flow. This can be 
 expressed as an equation : 
 
 Watts = Volts x Amperes. 
 
 For convenience the Kilo-watt (KW) is often used 
 as the unit of electrical power instead of the watt ; . one 
 kilo-watt equals 1000 watts. 
 
 Example. If a force of 10 volts causes a current oi 
 100 amperes to flow through a given circuit, the power 
 expended in this circuit : 
 
 -10 x 100 -1000 Watts or 1 KW. 
 
 OHM'S LAW 
 
 77. In every electrical circuit there are particularly 
 three factors, the true relation of which must be clearly 
 understood.
 
 24 WIRELESS TELEGRAPHY 
 
 These three factors are the force or pressure, the cur- 
 rent, and the resistance, and, as already explained, are 
 measured in terms of volts, amperes, and ohms, re- 
 spectively. 
 
 They bear a definite relation to one another, which 
 js expressed by Ohm's Law. 
 
 78. Ohm's Law. The strength of the current flawing 
 through any circuit is directly 'proportional to the pressure 
 acting across the circuit and inversely proportional to the 
 resistance of the circuit. 
 
 In other words, the current is equal to the pressure 
 divided by the resistance. 
 
 Using the units which are a measure of these factors, 
 the law can be stated as an equation thus : 
 
 Amperes- s 
 Ohms 
 
 and therefore by transposing, 
 
 Volts = Amperes x Ohms 
 
 Volts 
 Ohms = 
 
 Amperes. 
 
 Example. If a pressure of 6 volts be applied to &, 
 circuit whose resistance is 3 ohms, then the current 
 flowing through that circuit will be 2 amperes, thus : 
 Amperes = = 2. 
 
 MAGNETISM 
 
 79. Magnetism is the name given* to the power which 
 a magnet has of attracting iron or other magnetic sub- 
 stances. 
 
 The lodestone is a natural magnet, and if a piece 
 of hard steel is rubbed by it, or by another magnet, it will
 
 WIRELESS TELEGRAPHY 25 
 
 be found to act in the same way as the natural magnet 
 itself; that is to say, it will point north and south 
 when freely suspended, and will attract iron filings. 
 
 The piece of steel is then said to be magnetised, and 
 is known as a permanent magnet. One end of the magnet 
 is called the North (N.) Pole, and the other the South 
 (S.) Pole. 
 
 80. The range, or space, over which a magnet will 
 attract other magnetic substances is called the " mag- 
 netic field." 
 
 81. If the North Pole of one magnet is brought near 
 the South Pole of another magnet, the two will attract 
 one another, but if the two North Poles or two South 
 Poles are brought near one another, they repel each other. 
 
 It can be said, therefore, that like 'poles repel and 
 unlike poles attract one another. 
 
 82. Magnetic effects act in a definite direction along 
 imaginary lines, called " lines of force." 
 
 Every line of force passes out from the North 
 Pole round a com- 
 
 plete circuit, and re- / ' / ''" \ 
 
 turns into the South * ',*-" ~-- N \ .' 
 
 Pole, as shown in \ v \ t ,*---*--- > \ / y' 
 
 83. When a mag- < - ' - I"""""" " ~~s v " - * 
 netic substance is / , ( / ^ \ 
 
 brought into a mag- i \ ~ --_.__-' / 
 netic field, the sub- \ /' 
 
 stance becomes mag- F g 
 
 netised. 
 
 This effect is called " magnetic induction." 
 For the purpose of this book it is not necessary 
 to go fully into the factors controlling magnetic force,
 
 26 WIRELESS TELEGRAPHY 
 
 or the units by which these factors are measured, but 
 it is essential that the relation, or connection, between 
 electricity and magnetism is thoroughly understood. 
 
 84. The study of the relation between electricity and 
 magnetism is called " electro-magnetism." 
 
 E LECTRO- M AGNETIS M 
 
 85. A current of electricity passing through a con- 
 ductor produces a magnetic field round that conductor, 
 
 /' \^~^ / ^"'\^"^^ 
 
 r/;>N/VN/vx/: x > 
 
 i , , \/\ / r\i\* ." / x '\/-N * 
 
 FIG. 14 
 
 the Hues of force forming the magnetic field being a 
 number of concentric circles with the conductor as their 
 centre. 
 
 86. If the lines of force were visible, a side view of 
 the conductor would appear as shown in Fig. 14, and an 
 end view as shown in Fig. 15. 
 
 87. These lines of force have a definite direction de- 
 
 pending upon the direction in which 
 ' - N ^ \ the current is flowing (vide paragraph 
 / ' '^\ x v 82). 
 
 ,' / / 88. In Fig. 15 the direction of the 
 <' / lines is shown, assuming that the cur- 
 rent is flowing in the conductor upwards 
 towards the reader. If the current 
 were reversed, the direction of the 
 lines of force would be reversed, although they would 
 still remain as concentric circles.
 
 WIRELESS TELEGRAPHY 
 
 27 
 
 FIG. 16. 
 
 89. If the conductor is bent into a circle, as shown in 
 Fig. 16, and the current is passed through it in the 
 direction shown by 
 
 the large arrows, it 
 will be seen that the 
 magnetic lines are all 
 acting in an upward 
 direction on the inside 
 of the circle of wire, 
 and in a downward 1 
 direction on the out- 
 side of the circle. The 
 field thus produced is 
 exactly similar to that produced by a magnet, for it- 
 has polarity, 
 
 90. Since the lines of force come out of the upper side 
 
 _ _ of the circle and go 
 
 in at the under side of 
 the circle, the upper 
 side becomes the 
 North Pole and the 
 lower side becomes 
 the South Pole. 
 
 91. This effect is 
 still more marked if, 
 instead of making 
 only one turn of the 
 wire, we make a coil 
 of wire, as shown in 
 .Fig- 17. 
 
 92. In this case the lines of force produced by each 
 
 arn, instead of acting right round the conductor, can 
 
 be imagined to combine with those produced by the 
 
 FIG. 17.
 
 28 WIRELESS TELEGRAPHY 
 
 next turn, thus giving the resultant effect shown in 
 Fig. 17. 
 
 93. It may be said, then, that if a straight coil is 
 made by wrapping wire round a bobbin, and a current 
 of electricity from a battery is passed through the coil, 
 it will be found that the coil behaves exactly as if it were 
 a magnet. 
 
 94. If we insert a rod of hard steel into the coil and 
 pass the current as before, the steel rod will become a 
 permanent magnet. 
 
 95. If instead of the steel we insert a rod at solt iron 
 into the coil, it also becomes a magnet, but it is only 
 magnetic so long as the current lasts. This is called 
 an electro-magnet. 
 
 96. It is found that the strength of the magnetic 
 field produced depends upon three factors : (1) the 
 current passing round the coil, (2) the number of 
 turns in the coil, and (3) the " reluctance " (which is 
 the magnetic equivalent of electrical resistance) of the 
 "magnetic path" or "magnetic circuit." 
 
 The first two of these factors taken together constitute 
 the force-producing magnetism, which is called Magneto- 
 motive Force. 
 
 The unit of magnetomotive force is one ampere- turn. 
 
 97. If we place any magnetic substance, such as iron, 
 in the path of the magnetic lines of force, the reluctance 
 of the path of the lines of force is very much reduced 
 because the " permeability " (which is the magnetic 
 equivalent of electrical " conductivity ") of iron is very 
 much greater than that of air, with the result that the 
 strength of the magnetic field, or, in other words, the total 
 number of magnetic lines of force, produced by the same 
 current passing through the coil, is very much increased
 
 WIRELESS TELEGRAPHY 
 
 29 
 
 ELECTRO-MAGNETIC INDUCTION 
 
 98. As a magnet is made to enter a coil of wire, an 
 electromotive force is induced in the coil of wire, so that 
 if the electrical circuit be completed by connecting the 
 two ends of the coil together, a current of electricity 
 will flow through 
 the coil. 
 
 This effect 
 is known as 
 Electro-magnetic 
 Induction. 
 
 99. If a galva- 
 nometer or other 
 suitable measur- 
 ing instrument 
 be - connected 
 between the ends 
 of the coil, as 
 shown in Fig. 18, 
 so that any current flowing through the coil will flow 
 through the instrument, the deflection of the pointer will 
 indicate roughly the amount of current flowing through 
 the coil, and the direction in which it is flowing. 
 
 100. If now a magnet be thrust into the coil, the needle 
 of the galvanometer will be deflected from its normal 
 position, indicating that a current of electricity has been 
 generated in the coil. 
 
 101. If the magnet be left lying inside the coil, the 
 needle of the galvanometer will return to its normal 
 position, thus indicating that the current in the coil 
 has ceased. 
 
 102. We may say, then, that a current of electricity 
 
 FIG. 18.
 
 30 WIRELESS TELEGRAPHY* 
 
 will be induced in a coil of wire by a magnet so long as 
 there is a relative movement between the coil and the 
 magnetic field, or, in other words, when there is a change 
 in the number of lines of force passing through the coil. 
 
 103. If we continue the experiment and withdraw the 
 magnet from the coil, the needle of the galvanometer 
 will again be deflected, but this time in the opposite 
 direction, indicating that a current of electricity has been 
 generated in the coil in the opposite direction to that 
 produced by thrusting the magnet into the coil. 
 
 In effect, thrusting a magnet into a coil is equivalent 
 to increasing the number of lines of force passing through 
 the coil, and vice versa, withdrawing the magnet from the 
 coil is equivalent to decreasing the number of lines 
 passing through the coil. 
 
 104. We may say, then, that the direction of the 
 current induced in a coil by a relative movement between 
 it and a magnetic field depends upon whether the move- 
 ment tends to increase or to decrease the magnetic lines 
 of force passing through the coil. 
 
 105. By similar experiments it will be found that 
 the quicker we thrust the magnet into the coil, the greater 
 will be the current induced in the coil ; also that a stronger 
 magnet, that is to say, a magnet with a greater number 
 of lines of force, will induce a greater current in the coil 
 than a weak magnet, even though the two be thrust 
 into, or withdrawn from, the coil at the same speed. 
 
 106. We may say, then, that the current induced 
 in a coil depends upon the rate of change in the number 
 of magnetic lines of force passing through the coil. 
 
 In the above explanations we have taken the point 
 of view that currents were generated in the coil. It 
 must be remembered, however, that this is not, strictly
 
 WIRELESS TELEGRAPHY 
 
 31 
 
 speaking, accurate. It is really an electromotive force 
 that is induced in the coil, and the current only flows 
 as a result of this electromotive force when the circuit 
 through the coil is completed. 
 
 The word current is used merely to avoid complica- 
 tions. 
 
 107. Another variation can be made in these experi- 
 ments which greatly affects the amount of current 
 
 FIG. 19 
 
 induced in a coil, namely, the number of turns of wire 
 of which the coil is composed. 
 
 108. If we wind two separate coils, one with say 100 
 turns of wire and the other with 200 turns of wire, we 
 find that twice as much current is generated when we 
 thrust a magnet into the larger coil as when we thrust 
 the same magnet at the same speed into the smaller coil. 
 
 The best way to try this experiment is to connect 
 both coils in series with the galvanometer, as shown in 
 Fig. 19, and to introduce the magnet into one coil at a 
 time.
 
 32 WIRELESS TELEGRAPHY 
 
 By arranging it this way the resistance of the circuit 
 remains the same, whichever coil is used. 
 
 109. We may say, then, that the electromotive force 
 induced in a coil is proportional to the rate of change of 
 magnetic lines passing through the coil, and also to the 
 number of turns of wire in that coil, or, to put it as an 
 equation : 
 
 110. Electromotive force=rate of change of lines 
 x number of turns. 
 
 MUTUAL INDUCTION 
 
 111. The effects we have been considering up to the 
 present are those produced by " Electro-magnetic 
 Induction." 
 
 '20. 
 
 Referring to paragraph 89, \vc showed how a coil of 
 wire through which an electric current was passirig 
 produced a magnetic field similar to that produced by 
 a permanent magnet.
 
 WIRELESS TELEGRAPHY 
 
 33 
 
 112. It is obvious, then, that in the experiments 
 described in the last paragraphs we can produce exactly 
 the same results by replacing the magnet by a coil of 
 wire through which a current is kept flowing. 
 
 113. The effects then produced are known as those of 
 Mutual Induction, and an illustration of this is shown in 
 Fig. 20. 
 
 114. In the case of Mutual Induction, though, it is 
 not necessary to 
 
 move the first coil 
 in and out of the 
 second coil, for 
 we can produce 
 exactly the same 
 effect, namely, 
 that of changing 
 the number of 
 lines of force pass- 
 ing through the 
 second coil, by 
 leaving the first Flo .,, 
 
 coil permanently 
 
 inside the second coil, and making and breaking the 
 battery circuit through the first coil by means of a 
 switch, as shown in Fig 21 
 
 115 In these cases the coil P through which tbo cur- 
 rent is flowing :s called the Primary Coil, and the coil 
 S, in which the current is induced, is called the Secondary 
 Coil. 
 
 110 By relerring to paragraph ( J7, it is obvious that 
 the voltage induced in the secondary coil will be greatly 
 increased in the above experiments if a core of iron is 
 placed through the primary coil P. 
 
 D,
 
 34 WIRELESS TELEGRAPHY 
 
 117. If an iron core is used, the iron should be soft, 
 for the following reason. 
 
 Soft iron will retain only a very small amount of 
 the magnetism induced in it after the current passing 
 round it has been interrupted. Hard iron, or steel, 
 on the other hand, retains a very large portion of its 
 magnetism after the current passing round it has ceased 
 to flow. 
 
 The result, therefore, of using a steel core would l)e 
 that only a small change in the total number of magnetic 
 lines passing through the secondary would be obtained 
 by making and breaking the battery circuit. 
 
 THE CONSTRUCTION OF THE INDUCTION COIL 
 
 118. An Induction Coil is an instrument for producing 
 high-voltage impulses. 
 
 It is constructed on the principles of electro-magnetic 
 induction, which we briefly described in paragraphs 98 
 to 117. 
 
 119. In paragraph 114 we showed how a current of 
 electricity could be produced in a secondary coil by 
 making and breaking the battery circuit through a 
 primary coil. 
 
 In paragraph 109 we showed how the voltage, or 
 pressure, of the electricity induced in the secondary 
 coil depends upon two things : 
 
 (1) The rate of change in the number of magnetic 
 lines of force which pass through the secondary coil. 
 
 (2) The number of turns of wire with which the 
 secondary coil is wound. 
 
 120. The quicker the rate of change in the number 
 of magnetic lines of force the greater the resultant
 
 WIRELESS TELEGRAPHY 35 
 
 voltage across the secondary coil. Also, the greater the 
 number of turns in the secondary coil the greater the 
 resultant voltage across it. 
 
 121. By placing a core of soft iron in the primary coil 
 we very greatly increase the total number of magnetic 
 lines of force induced by the primary current (vide 
 paragraph 117), and therefore, when the primary circuit 
 is broken, we get a greater change in the number of 
 magnetic lines of force passing through the secondary 
 coil, and assuming that the time taken for the magnetism 
 to die down is the same as before, we get a greater 
 rate of change of magnetic lines passing through the 
 secondary, and theref ore _a ^higher voltage is induced 
 across it. 
 
 122. Further, by using a very fine wire we can wind 
 a very large number of turns on to the secondary coil, 
 thereby still further increasing the voltage induced 
 across the secondary. 
 
 123. By designing a coil on these principles, it is 
 possible to obtain voltages of 30,000 volts or more, 
 using only a small accumulator battery giving 4 or 6 
 volts in the primary circuit. 
 
 124. We may now describe how an induction coil is 
 actually made, and the means by which it can give auto- 
 matically a continuous stream of high- voltage impulses, 
 or sparks, when a low-voltage battery is applied to its 
 primary terminals. 
 
 The mechanical construction is shown in section 
 in Fig. 22, and the electrical connections are shown 
 diagrammatically in Fig. 23. 
 
 125. The secondary coil A is wound with about 5000 
 turns of fine wire on an ebonite bobbin B, the bobbin 
 having a hole through the middle sufficiently large to
 
 36 
 
 WIRELESS TELEGRAPHY 
 
 FIG. 22. 
 
 take the primary coil with its iron core, the two ends 
 
 of the secondary coil 
 *fj M are brought one to 
 each of the terminals 
 E, E, which are called 
 the " high-tension " 
 terminals of the in- 
 duction coil. 
 
 126. The iron core 
 C is made of a bundle 
 of soft iron wire, 
 bound together with 
 cotton tape, and 
 round this core is 
 wound the primary 
 
 winding D, consisting of about 50 turns of fairly thick 
 
 wire, through which the current from the primary battery 
 
 has to pass in order 
 
 to magnetise the iron 
 
 core. 
 
 127. One end of 
 
 this coil is taken 
 
 straight to the posi- 
 tive terminal of the 
 
 battery F, through 
 
 the manipulating key 
 
 G. The other end of 
 
 the coil, however, 
 
 instead of being con- 
 nected straight to the 
 
 negative terminal of 
 
 the battery F, is connected to the spring arm or trembler 
 
 blade H of the contact breaker K.
 
 WIRELESS TELEGRAPHY 37 
 
 128. This trembler blade carries on its side, which is 
 nearer to the core C, a small piece of soft iron L, and on 
 its other side a platinum contact M. Another platinum 
 contact N is carried on an adjusting screw by a 
 brass bracket P, m such a way that it comes immediately 
 opposite the contact M, the spring of the trembler blade 
 being adjusted so that normally the two contacts M 
 and N are making contact. The brass bracket is 
 connected to the negative side of the battery F. 
 
 129. The action of the coil can best be followed by 
 referring to the diagram of connections m Fig. 23. The 
 contacts M and N being in contact, if the arm of the 
 manipulating key G is depressed, the electrical circuit 
 through the primary coil is completed and a current 
 \\ ill flow from the positive side of the battery F, through 
 the manipulating key G, through the coil D, through the 
 trembler blade H, through the contacts M and N, through 
 the bracket P (Fig. 22), and back to the battery F. 
 
 130T~The effect of the current passing through the 
 coil D is to magnetise the iron core C, and the first 
 effect of this magnetisation is to induce a voltage across 
 the secondary coil of wire. This high voltage, however, 
 is only a momentary impulse, for it depends, as already 
 stated, upon the rate of change in the number of magnetic 
 lines of force passing through the secondary coil, so that 
 as soon as the iron core is fully magnetised by the 
 primary current, the change in the number of magnetic 
 lines ceases, and then-lore t he voltage across the secondary 
 falls to zero. 
 
 131. If, however, the primary current flowing round 
 the iron is interrupted, the iron core becomes demagnet- 
 ised, and there is again a rapid change in the number of 
 magnetic lines of force passing through the secondary
 
 38 WIRELESS TELEGRAPHY 
 
 coil, and we get a second high-voltage impulse across the 
 secondary coil. 
 
 132. Now this interruption of the primary circuit is 
 effected automatically by the contact breaker, for as 
 soon as the iron core becomes magnetised it attracts the 
 piece of iron L (Fig. 23), which, as already explained, is 
 fixed to the trembler blade, carrying the contact M, 
 thus separating the contact M from the contact N, 
 and interrupting the primary circuit. 
 
 133. As soon as the circuit is thus broken the iron 
 core C ceases to be a magnet, and therefore ceases to 
 attract the piece of iron L, allows it to fly back to its 
 original position, and the primary circuit is again com- 
 pleted through the contacts M and N coming together 
 again. 
 
 The same cycle of events repeats itself in rapid 
 succession so long as the manipulating key G is kept 
 
 134. The resulting effect in the secondary coil is 
 therefore a corresponding number of high - voltage 
 impulses across the coil, one impulse being induced when 
 the magnetism in the iron grows owing to the primary 
 current passing around it, and a second impulse being 
 induced in the opposite direction when the magnetism of 
 the iron collapses owing to the primary current ceasing 
 to pass around it. 
 
 135. As a matter of fact, the magnetism in the iron 
 grows comparatively slowly owing to the inductance of 
 the winding (vide paragraphs 61 and 66) as compared 
 with the rate at which the magnetism collapses on 
 breaking the circuit, and as the voltage across the 
 secondary coil is proportional to the rate of change of 
 magnetic lines of force, we get a very much bigger
 
 WIRELESS TELEGRAPHY 
 
 39 
 
 voltage during the collapse of the magnetism than 
 during the growth of the magnetism ; that is to say, 
 we get a higher voltage when the primary circuit is 
 interrupted than when it is completed. 
 
 136. Fig. 24 shows diagrammatically the voltage 
 induced across the secondary of an induction coil. The 
 upper part of the curve shows the voltage impulses due 
 
 FIG. 24. 
 
 to the making of the primary circuit, and the lower part 
 of the curve shows the voltage impulses due to the 
 breaking of the primary circuit. 
 
 137. From A to B the magnetism in the iron core is 
 growing comparatively slowly, and the voltage induced 
 across the secondary only rises to about 1000 volts. 
 
 At the point B the primary circuit is broken, and a 
 very high voltage, perhaps about 20,000 volts, is induced 
 across the secondary in the opposite direction, owing to 
 the very rapid collapse of the magnetism in the iron.
 
 40 WIRELESS TELEGRAPHY 
 
 138. Another important part of the induction coil is 
 the condenser, which is connected across the contact 
 breaker K, shown at Q in Figs. 22 and 23. 
 
 Owing to the inductance of the primary winding when 
 the current is suddenly interrupted at the contacts M 
 and N, a high voltage is generated in the primary coil 
 in the opposite direction to the applied voltage. This 
 voltage is sufficient to cause an arc to form between the 
 contacts M and N which interferes with the rapid inter- 
 ruption of the current necessary to produce a high- 
 voltage impulse in the secondary, and in addition causes 
 the contacts to fuse together and stick. 
 
 139. By placing a dondenser across these contacts the 
 energy liberated when the current is interrupted owing 
 to the inductance of the circuit (vide paragraph 62) ex- 
 pends itself in charging up the condenser, thus reducing 
 the arc to a minimnni and preventing the fusing of the 
 contacts. 
 
 By following the diagram of connections, Fig. 23, 
 it will be seen that the contacts, when closed, form a 
 short-circuit to the condenser, thus allowing the con- 
 denser to discharge itself through these contacts when 
 they fly together again, if it has not already discharged 
 itself through the primary winding of the induction coil. 
 
 PRODUCTION OF ELECTRICITY BY CHEMICAL ACTION 
 
 140. A Cell is an apparatus for producing a current of 
 electricity by chemical action. 
 
 141. A Battery consists of a number of cells joined 
 together either in parallel or series. 
 
 142. A cell usually consists of two dissimilar metals, 
 such as copper and zinc, immersed in a solution of acid 
 or salt, as shown in Fig. 25.
 
 WIRELESS TELEGRAPHY 
 
 41 
 
 Z//TC 
 
 Chemical action is set up by the acid attacking 
 the zinc, and the energy liberated by the dissolving 
 of the zinc appears in the form of electric potential 
 on the submerged surface of the zinc plate. 
 
 There is, however, practically no chemical action 
 set up on the copper plate, and therefore no electric 
 potential is produced on the surface of the copper plate, 
 with the result that the sub- 
 merged portion of the zinc plate 
 is at a higher electric potential 
 than the submerged portion of 
 the copper plate, and therefore 
 below the surface of the liquid 
 the zinc can be regarded as of 
 positive potential, and the copper 
 as negative. 
 
 143. When the two ends of 
 the plates which are above the 
 liquid are connected together by 
 
 a conductor, the current will flow from the zinc plate to 
 the copper plate under the liquid, and from the copper 
 plate to the zinc plate above the liquid. 
 
 144. For this reason the terminal which is joined to 
 the copper plate is called the positive terminal of the cell, 
 and the terminal which is joined to the zinc plate is 
 called the negative terminal of the cell. 
 
 145. Cells of this nature are called Primary Cells, and 
 a battery consisting of two or more of such cells properly 
 joined together is called a " Primary Battery." 
 
 ACCUMULATORS 
 
 146. An Accumulator is a cell in which the two 
 plates are made of such materials that when a current of 
 
 FIG. 25.
 
 42 WIRELESS TELEGRAPHY 
 
 electricity is passed through them from some outside 
 source in a certain direction, chemical actions are set up 
 between the plates and the electrolyte surrounding them, 
 thus altering the chemical composition of the materials 
 of which the plates were made. 
 
 147. This is known as " charging " the accumulator, 
 and the current which is passed through the accumulator 
 is known as the charging current. 
 
 148. On disconnecting the source of the charging 
 current, and connecting the two plates of the accumu- 
 lator together with a conductor, the cell will act in the 
 same way as a primary cell, the chemical composition of 
 the plates will start to return to its original state, and a 
 current of electricity will pass from the cell through the 
 conductor in the opposite direction to the charging 
 current. 
 
 149. Such cells are called " Secondary cells " or 
 " accumulators," and a battery, consisting of two or 
 more of such cells properly joined together, is called 
 an " Accumulator Battery," " Secondary Battery," or 
 " Storage Battery." 
 
 DIAGRAMS OF CONNECTIONS 
 
 150. For convenience in illustrating graphically the 
 connections of an electrical circuit, certain symbols are 
 used to denote the particular predominant property 
 which that part of the circuit possesses. When several 
 of these symbols are used to illustrate certain con- 
 nections, it is known as a " Diagram of Connections." 
 
 At the beginning of this book we give a number 
 of these symbols, and certain variations of them which 
 are most commonly met with.
 
 WIRELESS TELEGRAPHY 43 
 
 THE PRINCIPLES OF WAVE MOTION 
 
 151. Before showing how the principles of electricity 
 and magnetism are applied to wireless telegraphy, we 
 must first explain the principles of wave motion on 
 which the science of wireless telegraphy is founded. 
 
 152 Let us make a simple experiment to illustrate 
 these principles. 
 
 In a pool of water, and at opposite sides of it, two 
 pieces of wood are floating. If we strike one of these 
 pieces of wood with a hammer, or in any other way cause 
 it to disturb the water, it will be observed that a number 
 of ripples or waves are sent out in all directions. Follow 
 these waves until they reach the piece of wood at the far 
 side of the pool and it will be observed that this second 
 piece of wood is set in motion by the waves. 
 
 153. This is analogous to what occurs between two 
 wireless stations. The piece of wood that is struck with 
 a hammer corresponds to the transmitting station, the 
 water to the transmitting medium, and the piece of wood 
 at the far end of the pool to the receiving station. 
 
 154 The substance through which, or on the surface of 
 which, a wave travels is called the medium. 
 
 PROPERTIES OF WAVES 
 
 155. A wave has the property of propagating itself 
 radially from a given point. That is to say, once a wave 
 has been started it travels in all directions away from 
 the point at which it was started. 
 
 An illustration of this can be seen by dropping a stone 
 into the middle of a pool of water. The displacement 
 of the water by the stone starts a circle of ripples or 
 waves, which circle gets bigger and bigger until either the 
 waves die out or they reach the edge of the pool.
 
 44 WIRELESS TELEGRAPHY 
 
 156. A wave has also the property of producing at any 
 point in its path a disturbance in a body suspended in the 
 medium similar to the disturbance which started the wave. 
 
 Thus if we start a wave of water on the surface of 
 a pond by moving a stick in it, the wave will cause a 
 motion in another stick floating on the surface of the 
 pond at any point in the path of the wave. 
 
 COMMUNICATION BY WAVE-MOTION 
 
 157. We may say, then, that if we have a means of 
 starting waves at one point in a. medium, and a means 
 of detecting the passage of the waves at another point 
 in the same medium, we have a means of communicating 
 signals between these two points. 
 
 In order to communicate intelligence, however, we 
 must be able to distinguish between different signals, 
 and by means of the Morse Code (which is given in full 
 at the beginning of this book) the number of different 
 signals which it is necessary to produce at a receiving 
 station to communicate intelligence in the form of words 
 has been reduced to two, namely, the dot and the dash, 
 or " short and long." By different combinations of the 
 dot and dash we can represent every letter in the alpha- 
 bet, thus enabling us to spell out letter for letter any 
 word or sentence desired. 
 
 158. As an illustration, let us see how the Morse Code 
 could, for example, be applied to the method of com- 
 munication described in paragraph 153. 
 
 If we were to fix above the receiving piece of wood 
 a sheet of iron, or some other object, so that every time 
 a wave passed under it the piece of wood knocked 
 against this object, we should get a sound produced in 
 the form of a tap.
 
 WIRELESS TELEGRAPHY 45 
 
 A single blow from the hammer on the transmitting 
 piece of wood might produce two or three ripples, which 
 would be translated by the receiving piece of wood into 
 two or three taps. Several blows in rapid succession on 
 the transmitting piece of wood would send out perhaps 
 a dozen ripples following one another. These would be 
 translated by the receiving piece of wood by a similar 
 number of taps, thus giving us a ready means of producing 
 a short or long effect at the receiving end, and enabling us 
 to use the Morse Code for communicating intelligence. 
 
 159. An important point to understand is that 
 although a wave travels from one part of a medium to 
 another, the medium itself does not travel, and, except 
 for an up-and-down or to-and-fro motion while a wave 
 is passing, remains where it is. This can readily be 
 illustrated by placing something in a pond, such as a 
 fishing- float, which lies in the water with its top just 
 above the surface, and starting a wave some little 
 distance from the float. When the wave passes the float, 
 the latter will be seen to move up and down, but will 
 not be carried along with the wave. 
 
 MEASUREMENTS OF WAVES 
 
 160. In order to explain how the properties of waves, 
 more especially of 
 
 electric waves, can 
 be utilised for the 
 purpose of com- 
 munication, we must 
 know something 
 about the different 
 
 qualities or measurements of waves and the terms used 
 to describe them.
 
 46 WIRELESS TELEGRAPHY 
 
 161. The amplitude of a wave is the distance from the 
 highest point to the normal level, and is usually denoted 
 by the Greek letter a (alpha), as illustrated in Fig. 26. 
 
 162. The length of a wave is the distance from the 
 crest of one wave to the crest of the next, and is usually 
 denoted by the Greek letter X (lambda), as illustrated 
 in Fig. 26. 
 
 If we notice the surface of a pond over which a wave 
 is travelling, we see that only part of the wave is above 
 the normal level of the water, for there is a corresponding 
 depression between the crests. A complete wave con- 
 sists of the half which is above and the half which is 
 below the normal level. 
 
 163. The velocity of a wave or the speed of radiation 
 is the distance a wave will travel radially in one second. 
 
 Thus, if we start a wave on the surface of a pond, 
 and it takes one second from the time it was started for 
 the circle of ripples to reach a point on the pond 10 feet 
 away from the starting-point, the velocity of the wave 
 is 10 feet per second. 
 
 164. The frequency of a wave is the number of 
 complete waves that will pass a given point in one 
 'second, or, in other words, the rapidity with which one 
 wave follows another. 
 
 A good idea of what is meant by frequency can be 
 got by floating a cork on the surface of a pond, and after 
 starting a continuous succession of waves, or ripples, 
 some little distance away, counting the number of times 
 the cork bobs up and down. 
 
 The number of times it does this in a second is the 
 frequency of the wave. 
 
 165. Another definition of frequency can be made in 
 terms of the wave-length and the velocity. The explana-
 
 WIRELESS TELEGRAPHY 47 
 
 tion can be more readily followed by referring to Fig. 27. 
 Imagine a continuous succession of waves following each 
 other, as shown in Fig. 27 ; if we take two points A and 
 B to represent the distance that any one of these waves 
 will travel in one second, the total number of waves 
 included between the points A and B will be the fre- 
 quency of the wave, because all of these waves have to 
 pass the point A in one second. 
 
 166. Now the number of waves included between the 
 points A and B is equal to the distance from A to B 
 divided by the length of the wave. 
 
 s\ r \/\ /\ /\ 
 
 - toorxrr - 
 
 /\ ^ /^ 
 
 /\ /\ /\ 
 
 \/ \/ \/ 
 
 \ 
 
 \y v v \ 
 
 FIG. 27. 
 
 y \/ \/ \j 
 
 E 
 
 Take an example : Suppose the waves are travelling 
 at the rate of 100 feet per second, then the distance from 
 A to B is 100 feet ; and suppose there are ten waves 
 included between the points A and B, as shown in the 
 diagram, it follows that the length of each wave is 
 10 feet. 
 
 167. We may say, then, that 
 
 Velocity 
 
 Frequency = ^ , 
 Wave-length 
 
 and therefore by transposing, 
 
 Velocity = Frequency x Wave-length 
 
 or 
 
 . ., Velocity 
 Wave-length = 
 
 Frequency.
 
 48 WIRELESS TELEGRAPHY 
 
 PRESSURE WAVES 
 
 168. The waves we have been considering travel 
 along the surface of water, but another kind of wave can 
 be formed which travels through the body of a substance. 
 Such waves are called " Pressure Waves " and obey the 
 following laws. 
 
 169. (1) A pressure wave travels at a definite speed 
 depending on the substance or medium, and the speed 
 in a given medium remains -the same no matter how 
 big or how small the waves may be. 
 
 Generally speaking, the speed or velocity is greater 
 the greater the elasticity l of the substance. 
 
 Thus in air the velocity is about 1090 feet per second, 
 in water it is 4700 feet per second, and in steel it is 
 16,400 feet per second. 
 
 170. (2) The amplitude of the wave very rapidly gets 
 smaller as the wave gets farther from its starting-point, 
 until, if given sufficient room, it finally dies out alto- 
 gether ; in other words, the amplitude decreases as the 
 distance from the starting-point increases. 
 
 171. (3) The length of the wave remains the same no 
 matter how far it is from its starting-point ; in other 
 words, the wave-length, once started, remains constant, 
 and is quite independent of the amplitude. 
 
 AETHER WAVES 
 
 172. In order to explain the phenomena of light, 
 radiant heat, and electric waves, physicists have imagined 
 a substance or medium called the " aether," and waves 
 similar to the pressure waves we have just been con- 
 
 1 The elasticity of a substance is the force which must be applied 
 to a given length of unit cross-section in order to double this length 
 It must not be confused with the term " flexibility " used in par. 30.
 
 WIRELESS TELEGRAPHY 49 
 
 sidering produce rays having different properties, accord- 
 ing to the wave-length. 
 
 173. The shortest wave-lengths known produce 
 X-rays, which have the property of passing through 
 many bodies that are impervious to light rays, and of 
 causing chemical action on photographic plates. The 
 next in length produce actinic rays, causing chemical 
 action similar to those produced by X-rays. Then light 
 rays, which act on the eyes, producing the effect of 
 vision ; and heat rays which produce the effect of 
 warmth ; and, finally, " electric " rays, which will pro- 
 duce electric currents in conductors, and which are used 
 in Wireless Telegraphy. 
 
 The following is a table of some of these wave- 
 lengths 
 
 X-rays, about 2-5 millionths of an inch. 1 " 
 
 Actinic rays of maximum intensity, 10 millionths of 
 an inch. 
 
 Light rays, from 10 to 18 millionths of an inch. 
 
 Heat rays of maximum intensity, about 15 millionths 
 of an inch. 
 
 Electric rays, shortest measured, 0-24 inch ; used in 
 wireless telegraphy, 300 feet to 50,000 feet. 
 
 174. All these waves obey the laws stated for pressure 
 waves (paragraphs 169 to 171), and have the properties 
 explained in paragraphs 155 and 156, and, since they all 
 travel through the same medium, the velocity of all of 
 them is that of Light namely, 300,000,000 metres, or 
 about 1,000,000,000 feet per second, equal to 186,000 
 miles per second. 
 
 175. The velocity of aether waves is thus seen to be 
 
 1 The one-thousandth part of the thickness of a cigarette paper 
 is about one- millionth of an inch.
 
 50 WIRELESS TELEGRAPHY 
 
 far, greater than that of air waves. It is for this reason 
 that, as light travels in the form of aether waves, and 
 sound travels in the form of air waves, if we watch a 
 battleship from a distance firing guns, we see the flash of 
 5he gun long before we hear the report. 
 
 176. Caution must be exercised when the effects 
 produced by surface waves are used to explain the 
 phenomena of Wireless Telegraphy, Sound, or other 
 effects transmitted by pressure-wave radiation, because 
 surface waves do not follow altogether the laws governing 
 pressure waves. Thus they do not follow Law No. 1, 
 for the velocity of the surface waves on water depends on 
 the wave-length and amplitude, i.e. big waves travel 
 faster than small ones ; further, they do not follow 
 Law No. 2, for the amplitude of surface waves is not 
 independent of the wave-length, thus if a surface wave 
 of definite length be started, it will be found that its 
 length will increase as its amplitude decreases. 
 
 In general we may make the following deductions 
 regarding the effects produced by wave motion. 
 
 177. (1) The nature of the effect produced by a wave, 
 or a series of waves in a given medium, on the senses, 
 or on other matter, depends upon the frequency of the 
 waves. 
 
 Thus, if the frequency of waves travelling in aether 
 lies between about 1200 billions and 660 billions, they 
 will produce an effect on the eyes known as light, and 
 the various frequencies between those limits will produce 
 various colours. 
 
 178. (2) The strength of the effect produced by a wave 
 depends upon the amplitude of the wave, and since the 
 amplitude of the wave gets smaller as the wave gets 
 farther from its starting-point, the effect produced by a
 
 WIRELESS TELEGRAPHY 51 
 
 wave will be weaker as the distance from its starting-point 
 is increased. 
 
 Thus, taking a lighted candle as the starting-point 
 of a number of light waves, it will be observed that the 
 strength of the light it produces on, say, a sheet of paper 
 is very rapidly reduced as the distance between the 
 paper and the candle is increased. 
 
 COMMUNICATION BY MEANS OF AETHER WAVES 
 
 179. In paragraph 152 we showed how, by means of 
 waves on the surface of a pond, we could communicate 
 signals from one point to another, but the method there 
 described would be exceedingly slow, and would, for 
 many other reasons, be quite impracticable. 
 
 Since Aether Waves also possess the properties 
 mentioned in paragraphs 155 and 156, it is obvious that 
 these waves can be used in a similar manner for the 
 purpose of communicating signals from one point to 
 another (vide paragraph 157), with the great advantage 
 that, the speed of radiation being 186,000 miles per 
 second, communication from one point to another will 
 be practically instantaneous. 
 
 180. Communication has, for many years before 
 " Wireless Telegraphy " was thought of, been carried 
 out by means of Aether Waves in the form of light waves, 
 by using searchlights, heliographs, and similar apparatus, 
 but this method has the disadvantage that the range is 
 small and intervening objects interrupt communication. 
 
 The discovery leading to Marconi's great invention 
 of Wireless Telegraphy was made by Hertz. Hertz 
 first experimentally proved the existence of electric 
 waves and indicated how they could be produced by
 
 52 WIRELESS TELEGRAPHY 
 
 electrical means. For this reason they are sometimes 
 known as Hertzian waves. 
 
 It should be understood, though, that Hertz only 
 demonstrated the existence of these waves, and did not 
 in any way attempt to utilise them as a means of com- 
 munication. 
 
 181. The advantages which electric waves have over 
 light waves for communication can be briefly stated as 
 follows : 
 
 182. (1) They will pass through, or over, intervening 
 objects which are impervious to light waves, and therefore 
 these objects will not interrupt communication. 
 
 183. (2) They will follow the curvature of the earth, 
 and therefore the range of communication can be in- 
 creased beyond the limits of the horizon, whereas in the 
 case of light waves it is necessary that the point from 
 which a ray of light is being received is above the horizon. 
 
 PRODUCTION OF WAVES 
 
 184. We have already stated that waves formed on 
 the surface of a body do not follow exactly the same laws 
 as pressure waves, but the analogy of the wave produced 
 on the surface of a pool is useful in explaining how a 
 pressure wave is produced. 
 
 Let us first understand clearly what the difference 
 is between a wave produced on the surface of a body 
 and a wave produced through the substance of a body. 
 
 185. The wave on the surface of a pool depends for 
 its existence upon a difference in the height of adjacent 
 particles of water above or below the normal level of 
 the water. It may therefore be called a height wave. 
 
 186. The wave produced through the substance of a
 
 WIRELESS TELEGRAPHY 
 
 53 
 
 body, oil the other hand, depends for its existence upon 
 a difference in the pressure of adjacent particles of the 
 substance through which it is travelling above or below 
 the normal pressure of that substance. It is therefore 
 called a pressure wave. 
 
 PRODUCTION OF HEIGHT WAVES 
 
 187. It is not generally known why a height wave is^ 
 produced on the surface of a pool when a stone is dropped 
 into it, and therefore an explanation of it will be useful 
 before describing how a pres- 
 sure wave can be started. 
 
 188. It is evident that as 
 we fill up a pond we raise the 
 height of the surface of the 
 water in that pond. It does 
 not matter with what material 
 we fill the pond up, whether it 
 be water or stones or any 
 other matter, the effect is the 
 same, namely, the height of 
 
 the surface of the water is increased. 
 
 189. It follows, then, that if we drop only a single 
 stone into a pool, we increase the height of the surface of 
 that pool, even though it be ever so slightly. 
 
 Owing to the inertia of the water, however, the 
 height of the water is not immediately increased over 
 the whole surface of the pool, for the inertia of the water 
 tends to prevent it from rising. The result is, that when 
 the stone is plunged into a pool, the water that is dis- 
 placed by the stone forms in a heap all round the stone, 
 as shown in Fig. 28. A height wave is thus started on 
 the surface of the water, and travels radially like an ever- 
 
 Fio. 28.
 
 54 WIRELESS TELEGRAPHY 
 
 expanding circle with the point where the stone entered' 
 the water as its centre. 
 
 190. The size of the pool makes no difference to the 
 production of the wave, for it is just as easy to start a 
 wave in the middle of the ocean as it is in a pool of water, 
 for the effect does not depend upon the inertia of the 
 whole mass of water, the inertia of the water immediately 
 surrounding the stone being sufficient. 
 
 PRODUCTION OF PRESSURE WAVES 
 
 191. We may take a very similar experiment to ex- 
 plain the production of a pressure wave in the air, but it 
 
 must be remembered 
 in this experiment 
 that instead of start- 
 ing a circle of maxi- 
 mum height on the 
 surface of a body 
 we start a circle, or 
 rather a sphere of 
 maximum pressure, 
 in the substance of 
 the body. 
 
 Flo 29 Let us imagine 
 
 a closed chamber full 
 
 of air with an opening at the bottom through which 
 we can pump some water, as shown in Fig. 29. 
 
 192. It is well known that if we pump anything into 
 the chamber we increase the pressure on the air inside it. 
 It does not matter what we pump in, whether it be water 
 or air, the effect is the same, namely, the pressure is 
 increased. 
 
 193. We will suppose, for the purpose of explanation,
 
 WIRELESS TELEGRAPHY 55 
 
 that the chamber is full of air at normal pressure and 
 that we increase the pressure of this air by pumping 
 water into the chamber. 
 
 194. If, then, we suddenly force water into the 
 chamber, owing to the inertia of the air we momentarily 
 only increase the pressure in the air immediately above 
 the surface of the water, or in other words, the air that is 
 displaced by the water forms a pressure heap which 
 travels forward in the form of a pressure wave through 
 the substance of the air 
 
 195. It must be understood, though, that the air 
 itself does not travel, but only the pressure of the air 
 travels, just as tho water that is displaced by the stone 
 does not travel, but only the height of the water travels. 
 
 196 It must be further understood that the formation 
 of a pressure heap of air in the chamber does not depend 
 on the inertia of the whole of the air in that chamber, 
 the inertia of the air immediately above the water being 
 sufficient, and therefore just as the height wave can be 
 produced as easily in the middle of the ocean as in the 
 middle of a pond, so can a pressure wave of air be 
 produced in the open atmosphere just as easily as it can 
 be in a closed vessel, it being only necessary to displace 
 some of the air at a given point to start a wave. 
 
 197. Since the air is invisible it is impossible to see 
 these pressure waves, and they can only be imagined ; 
 but a simple experiment can be made in which a pressure 
 wave can be actually seen, and which compares very 
 closely to the experiment explained above. 
 
 198. Let us make up a very long spiral spring out of 
 fine steel wire, as shown at A in Fig. 30, and support it 
 at intervals along its whole length by threads, so as to 
 allow it a greater freedom of motion than if we laid it on
 
 56 WIRELESS TELEGRAPHY 
 
 a table where any motion would have to overcome /he 
 friction of the table. 
 
 199. If now we strike one end of this spiral a sharp 
 tap with a hammer, it will be observed that, at the 
 moment of impact, only that part of the spring' im- 
 mediately in front of the hammer will be compressed, 
 while the rest of the spring remains as it was. 
 
 200. This compression will be seen to travel along the 
 whole length of the spiral, like a pressure wave, leaving 
 that part of the spiral between it and the hammer in its 
 normal state after it has passed ; thus at the moment of 
 
 vJlw/Jww^^ A 
 
 <= /pOWPSWWWWM/^^ B 
 
 c ,_XpQM/VVVW\MA/WWVWW^ C 
 
 FIG. 30. 
 
 impact the spring will take the form as shown at B in 
 Fig. 30, and when the wave has travelled half the length 
 of the spiral, tb3 spring will take the form as shown at 
 C in Fig. 30. 
 
 201. To carry out this experiment successfully, the 
 inertia of the spring must be made big by using a very 
 long spiral, say 30 feet long ; also the wire forming the 
 spiral must be extremely fine, so as to allow it to com- 
 press easily without exerting too great a force against 
 the inertia of the whole spring, otherwise the effect will 
 be produced too rapidly for observation, and it will 
 appear that the spiral is only moved bodily by the 
 hammer. A suitable spiral would be a coil of fine steel
 
 WIRELESS TELEGRAPHY 
 
 wire, say No. 28 gauge, wound on a coil say 1 inch in 
 diameter and 20 feet or 30 feet long. 
 
 The spring should be suspended by threads as long 
 as possible and at intervals as frequent as possible. 
 
 202. In the above experiments we have considered 
 that a wave is produced by a sudden increase in the 
 pressure at a given point. This, however, is not strictly 
 accurate, for a single pressure pulse ddes not produce a 
 complete wave, but only one-quarter of a wave. 
 
 20.3. To produce- a complete wave, it is necessary that 
 the pressure be first in- 
 creased above normal, 
 then reduced to normal, 
 then reduced to below 
 normal, and finally 
 increased again to 
 normal. 
 
 204. An illustration 
 of this is given in Fig. 
 31, where a long india- 
 rubber tube is shown 
 full of water. The 
 size of the tube can be 
 increased or decreased 
 by connecting it 
 through a .pipe to a pump, "capable of pumping more 
 water into the tube and of sucking some of the water 
 out of it, thus increasing and decreasing the size of 
 the tube. 
 
 The full line in the illustration shows the normal 
 size of the tube, and the two dotted lines show the 
 maximum and minimum sizes of the tube. 
 
 205. When the tube has been expanded, reduced to 
 
 FIG. si.
 
 58 WIRELESS TELEGRAPHY 
 
 normal size, contracted and increased to normal size, it 
 is said to have passed through one complete cycle of 
 operations. 
 
 206. As the size of the tube is increased, the pressure 
 of the air surrounding it is increased above normal, 
 and vice versa as the size of the tube is decreased the 
 pressure of the air surrounding it is decreased below 
 normal. 
 
 Therefore, if we pass this tube through one cycle 
 of conditions, we shall produce in the air surrounding 
 it one complete pressure wave. 
 
 207. In Fig. 32 the same tube is shown in different 
 stages of the cycle with relation to time, assuming that 
 the time taken to inflate and deflate the tube through 
 one cycle is one second. Thus at the commencement of 
 operations the tube is at its normal size, as shown at A. 
 At the end of a quarter of a second the tube has been ex- 
 panded to its maximum size, as shown at B. At the end 
 of half a second the tube has been reduced again to its 
 normal size, as shown at C. At the end of three-quarters 
 of a second the tube has been contracted to its minimum 
 size, as shown at D, and at the end of one second the 
 tube has once more returned to its normal size, as 
 shown at E. 
 
 208. The effect on the pressure of the air surrounding 
 the tube, with relation to the time, can be illustrated 
 diagrammatically by the curve drawn below the illustra- 
 tion of the tube, where the distance of the curve above 
 or below the horizontal line represents the increase or 
 decrease in the pressure of the air, or what comes to 
 the same thing, the increase or decrease in the size of 
 the tube and the distance along the horizontal line 
 represents the progress of time
 
 WIRELESS TELEGRAPHY 
 
 59 
 
 It is evident that the frequency of the wave produced 
 depends entirely upon the frequency of the impulses 
 producing it. 
 
 I COMPLETE CYCLE 
 Fid. 32. 
 
 209. In paragraph 167 we defined the relation between 
 the length of the wave, the frequency, and the speed at 
 which it was radiated. This relation will be better
 
 60 WIRELESS TELEGRAPHY 
 
 understood by applying it to the foregoing illustration 
 of the production of a complete wave. 
 
 210. Since it took one second for the tube to pass 
 through a complete cycle of operations, we can say that 
 the frequency of these operations was one per second, 
 and since the cycle of operations only produced one wave, 
 we may say also that the frequency of the wave was one 
 per second, or in other words, the wave was not complete 
 
 -until one second after 
 
 f\ /\ /\ /\ /\ . its commencement. 
 
 211. Since the 
 
 speed of radiation of 
 
 pressure in air is roughly 1000 feet per second, it 
 follows that the beginning of the wave will have 
 reached a distance of 1000 feet from the starting- 
 point by the time the end of the wave has just left 
 the starting-point ; thus the length of the wave will 
 be 1000 feet. It will be seen we shall get the same 
 result by applying the formula wave-length = velocity 
 -T- frequency. 
 
 Up to the present we have considered only the 
 production of a single wave. 
 
 212. A group of waves may be defined as a natural 
 sequence of two or more waves immediately following 
 one another with- 
 out any interval 
 
 between them, as / \ / \ f\ r\ 
 
 illustrated in Fig. \ / \J 
 
 33. 
 
 213. If a group 
 
 of waves is produced in such 'a way that each successive 
 wave has the same amplitude, as shown in Fig. 33, the 
 waves are said to be " continuous."
 
 WIRELESS TELEGRAPHY 61 
 
 214. If a group of waves is produced in such a way 
 that the amplitude oi each successive wave is less than 
 its predecessor, as shown in Fig. 34, the waves are said 
 to be " damped." 
 
 215. A group of waves is produced by a series of 
 periodic displacements of the medium, which follow one 
 another without an interval. 
 
 216. It is obvious from the experiments described in 
 paragraphs 203 to 207 that if we repeat the cycle of 
 operations of expanding and contracting the rubber tube 
 periodically; vve shall produce a group of waves, and if 
 the extent of these expansions and contractions is 
 maintained, the result will be to produce a group of 
 continuous waves ; but if the extent of these expansions 
 and contractions gradually gets smaller and smaller till 
 the tube finally comes to rest at its normal size, the result 
 will be to produce a group of damped waves. 
 
 PRODUCTION OF ELECTRIC WAVES 
 
 217. In order to convey signals from one point to 
 another by means of wireless telegraphy it is necessary 
 to have an apparatus for producing electric waves at 
 one point, and an apparatus for detecting the presence 
 of such waves at the other point. 
 
 218. We have described how pressure waves in the 
 air are produced by the periodic displacement of the air 
 at a given point through a definite cycle, causing the 
 pressure in the air to vary from normal pressure to a 
 positive pressure, to normal pressure, to a negative 
 pressure, and finally to normal pressure. 
 
 219. Electric waves in the aether are produced by 
 the periodic displacement of the aether at a given point 
 through a similar cycle.
 
 C2 WIRELESS TELEGRAPHY 
 
 220. These displacements of the aether are produced by 
 electrical charges in what is known as the Aerial Wire. 
 
 For the purpose of explanation we may regard the 
 aether as a substance which exists in everything. 
 
 221. When we charge up a condenser, we put the 
 dielectric of the condenser in a state of strain. This 
 state of strain in the dielectric exerts a pressure on the 
 aether which exists in the dielectric, and the pressure 
 pulse thus produced will radiate through the aether 
 in a similar way to the radiation of the pressure pulse 
 in the air which we described previously. 
 
 222. This, however, only produces one pulse, and to 
 produce a complete wave in the aether we must pass the 
 condenser through a complete cycle of operations by first 
 charging it positively, tjien discharging it, then charging it 
 negatively, and again discharging it (vide paragraph 203). 
 
 223. To produce a group of waves we must pass the 
 condenser through a series of these cycles following each 
 other in periodic sequence. 
 
 224. If during each cycle the condenser is charged to 
 the same extent, i.e. to the same voltage, the group of 
 waves produced will be " continuous," but if each suc- 
 cessive charge of the condenser is weaker than the last, 
 the group of waves produced in the aether will be 
 " damped." 
 
 In this book we shall only describe the production 
 of damped waves* 
 
 225. Suppose we suspend a length of wire in the air 
 from a mast and insulate it from the earth, as shown in 
 Fig. 35, we may regard the wire, the air, and the earth 
 as forming a condenser, in which the wire acts as one 
 plate of the condenser, the air as the dielectric, and the 
 earth as the other plate.
 
 WIRELESS TELEGRAPHY 
 
 63 
 
 226, And suppose we have a way of charging 
 and discharging it in rapid succession by means 
 of a suitable electrical generator G, first charging it 
 positively and then negatively, each charge and dis- 
 charge produces a pressure pulse in the aether, and 
 the four pulses positive charge, discharge, negative 
 charge, discharge form a complete electric wave, which 
 starts travelling into space with the velocity of light, 
 namely 300,000,000 metres per second. 
 
 227. Such a wire is known 
 as the Aerial Wire, and is 
 given various shapes, as we 
 shall describe later. 
 
 228. The analogy of the 
 expansion and contraction of- 
 the india-rubber tube, de- 
 scribed in paragraph 204, 
 can be used to explain the 
 action of charging and dis- 
 charging the aerial wire. 
 
 229. Whilst being charged, 
 a current of electricity will 
 flow into the wire, just as a 
 
 current of water was made to flow into the tube to 
 expand it. The current will be large at first and 
 diminish as the aerial becomes charged, until it ceases 
 to flow when the aerial is fully charged. The instant 
 it has ceased to flow the current will start to flow in 
 the opposite direction, as the aerial discharges ; and so 
 on, the current will flow backwards and forwards, charg- 
 ing and discharging the aerial through successive cycles. 
 
 230. Such a current of electricity is called an oscillat- 
 ing current. 
 
 Fio. 35.
 
 64 WIRELESS TELEGRAPHY 
 
 231. It is obvious that the frequency of the wave 
 produced in the aether will depend entirely upon the 
 frequency of the oscillations in the aerial (vide para- 
 graph 208). 
 
 232. It follows, therefore, that by varying the fre- 
 quency of the oscillations in the aerial we can vary the 
 length of the wave radiated. 
 
 From the formula 
 
 Wave-length 
 
 x requency 
 
 it can be seen that the greater the frequency of the 
 oscillations the shorter the wave-length. 
 
 233. The wave-lengths usually employed for the pur- 
 pose of wireless telegraphy vary in length from 100 
 metres to 15,000 metres or more. Generally speaking, 
 the larger the power of the station the longer the 
 wave-length employed. 
 
 The wireless apparatus on ships and at the shore 
 stations with which the ships communicate is designed 
 to transmit wave-lengths of 300 metres and of 600 
 metres. Long-distance stations use wave-lengths vary- 
 ing from 1000 to 15,000 metres. 
 
 234. From the formula quoted above it will be 
 found that the number of oscillations per second required 
 to produce a wave-length of 15,000 metres is 20,000, 
 an$ the number per second required to produce a wave- 
 length of 100 metres 3,000,000. 
 
 235. Such oscillations are known as High-frequency 
 or oscillatory currents, to distinguish them from the 
 Low-frequency or alternating currents of between 25 
 and 1000 per second produced by ordinary alternating 
 current dynamos.
 
 WIRELESS TELEGRAPHY 65 
 
 PRODUCTION OF HIGH FREQUENCY 
 OSCILLATIONS 
 
 236. There are several ways of producing high- 
 frequency oscillatory currents. For the present we 
 will confine ourselves to the method known as the 
 " spark " method. 
 
 237. If a condenser is charged and then suddenly 
 discharged by connecting its two opposite plates together 
 with a conductor, not only does the current flow from 
 the positively charged plate to the negatively charged 
 plate until the plates are at the same potential, but the 
 current continues to flow in the same direction on account 
 of the inductance (vide paragraph 61) of the conductor, 
 causing that side of the condenser which before was 
 negatively charged to become positively charged, and 
 vice versa. 
 
 238. The reversed charge, however, does not charge 
 the condenser to the same extent, i.e. to the same voltage 
 as the original charge, because a certain amount of the 
 energy is frittered away by the resistance of the circuit, 
 and a further amount of energy is used up in the 
 production of pressure waves. 
 
 239. The action is then reversed with the same effect, 
 and so on, each time with less energy, until the whole 
 of the energy originally in the condenser is absorbed. 
 An oscillating current of gradually diminishing strength 
 is therefore produced. 
 
 240. The action can be illustrated by making the 
 experiment with the pendulum illustrated in Fig 36, 
 where the weight \V is shown suspended from a fixed 
 point A by a piece of string B. 
 
 241. If this weight be displaced to the position sbowr.
 
 66 
 
 WIRELESS TELEGRAPHY 
 
 by the dotted line marked W x and then released, it 
 will not immediately take up the position W, but owing 
 to the momentum of the weight will swing backwards 
 and forwards between the positions Vf l and W 2 . 
 
 242. Owing to the friction between the weight and the 
 air, and also to the fact that it gives up some of its 
 energy to the surrounding air in forming pressure waves, 
 
 B\ 
 
 FIG. 36. 
 
 each swing will be a little shorter than the last, and it 
 will finally come to rest at the position W, after making 
 a number of swings. 
 
 243. In this case the number of swings which take 
 place in a second that is, the frequency of the oscilla- 
 tionscan only be varied by varying the length of the 
 string. 
 
 244. A similar experiment can be made with a 
 vibrator, shown in Fig, 37, in which there is a flat steel
 
 WIRELESS TELEGRAPHY 
 
 67 
 
 spring B, fixed firmly at the point A, and carrying a 
 weight W at its other or free end. 
 
 If the weight be displaced and released, it will swing 
 or " oscillate " between the position Wj and W 2 . 
 
 245. In this case the number of swings per second 
 will depend upon the flexibility of the spring B and also 
 upon the inertia of the weight W, and we can therefore 
 vary the frequency 
 
 either by varying the 
 
 flexibility of the spring 
 
 or by varying the ill I . .A'WJ' ^8 
 
 weight. 
 
 246. It will be 
 found that the longer 
 or thinner and there- 
 fore the greater the 
 flexibility of the 
 spring, the less the 
 number of swings per 
 
 second, also the greater the weight W the less the 
 number of swings per second. 
 
 247. The distance between the position \V l and the 
 position W is called the maximum amplitude of the 
 swing, and generally half the distance between succes- 
 sive positions, such as W 3 , W 4 in Fig. 37, is called 
 the amplitude. 
 
 248. Owing to friction of the air and also to the energy 
 radiated in the form of pressure waves, the amplitude 
 will start at a maximum and gradually diminish until 
 the spring comes to rest in its normal position. The 
 rate at which the swing decreases is called the "damping." 
 
 249. The frequency will, however, remain constant 
 quite independently of. the amplitude ; that is to say, 
 
 FIG. 37
 
 68 WIRELESS TELEGRAPHY 
 
 in Fig. 37 the time taken for the weight to travel from 
 the position W x to W 2 and back again will be exactly 
 the same as the time it takes to travel from the position 
 W 3 to W 4 and back again. 
 
 250. To summarise, two matters have to be con- 
 sidered in connection with such a vibrator, viz. : 
 
 (1) The frequency, depending upon (1) the flexibility 
 of the spring and upon (2) the mass of the weight. 
 
 (2) The damping, depending upon the friction and 
 the rate at which the energy is radiated. 
 
 OSCILLATORY CIRCUITS 
 
 251. A circuit in which an oscillating current will 
 flow is called an oscillatory circuit, and must possess two 
 essential qualities, namely, Capacity and Inductance. 
 
 It may be compared with the vibrator described in 
 the last article. 
 
 252. The properties of the vibrator which decide 
 the frequency of the vibrator are its mass and its 
 springiness. 
 
 We have already explained in paragraphs 28 and 62 
 that the property of capacity is analogous to that of 
 springiness and the property of inductance is analogous 
 to that of mass. 
 
 Similarly, therefore, the properties of an oscillatory 
 circuit which decide the frequency of the oscillating 
 currents that will flow in it, and therefore the wave- 
 length it will produce, are its Inductance and its Capacity, 
 
 253. Also the property which tends to stop or " damp " 
 the vibrations of the vibrator is friction and radiation 
 of energy. Similarly the property which tends to damp 
 the oscillating current in an oscillatory circuit is the 
 Resistance of the circuit and radiation of energy.
 
 WIRELESS TELEGRAPHY 69 
 
 254. Obviously resistance is an undesirable property, 
 as it absorbs energy. In every oscillatory circuit, there- 
 fore, the resistance is reduced to a minimum quantity, 
 which is effected by increasing the size of the conductor 
 and reducing its length as much as possible ; there is, 
 however, always a certain amount of resistance left. 
 
 255. As oscillatory circuits are used for the production 
 f electric waves, it is important to know the relation 
 between the wa/ve-length and the capacity and inductance 
 of the circuit. 
 
 256. It will be found that as we increase either the 
 value of the capacity or of the inductance of the circuit, so 
 do we decrease the frequency of the circuit, just as the 
 frequency of a vibrator is decreased by increasing either 
 its springiness or its mass. 
 
 257. The frequency of an oscillatory circuit is inversely 
 proportional to the square root of the capacity and the 
 square root of the inductance. And since the wave- 
 length is inversely proportional to the frequency, it 
 follows that the wave-length produced is proportional 
 to the square root of the capacity and the inductance. 
 
 258. This law can be expressed as a formula, thus* 
 
 Wave-length = ^Capacity x v/Inductance, 
 
 or using the symbols by which these quantities are 
 known, 
 
 fcWCxL* 
 
 259. This formula does not define what units are used 
 to measure the different qualities. But if we measure 
 the wave-length in metres, the capacity in microfarads, 
 and the inductance in microhenries the formula becomes 
 
 \(m) = 1885 VQmf) x L(mh).
 
 70 WIRELESS TELEGRAPHY 
 
 ENERGY AND POWER IN OSCILLATORY CIRCUITS 
 
 260. In paragraph 178 we explained that the strength 
 of the effect produced by a wave depends upon the 
 amplitude of the wave, and since the amplitude of the 
 wave decreases very rapidly as the distance it has 
 travelled is increased, it follows that the effect produced 
 on a receiver gets rapidly weaker as the distance from 
 the transmitter is increased. 
 
 261. It is evident that to get a stronger effect at the 
 same distance, or to produce the same effect at a greater 
 distance, we must increase the amplitude of the wave 
 at the starting-point. 
 
 262. It is simpler to consider wave- motion as a means 
 of radiating energy ; also to consider the effect produced 
 by the wave on what we may call a receiver, as the 
 amount of energy received. 
 
 263. The amount of energy received at any point 
 must necessarily be very small compared with the total 
 amount of energy radiated, because in radiating energy 
 'we spread that energy out over a large space ; thus the 
 farther the energy has been radiated the greater the 
 space over which it is spread. 
 
 264. Take the case of a wave on the surface of a pond. 
 At the starting-point the whole of the energy in the 
 wave is concentrated in a very small space, and therefore 
 a receiver in the shape of a piece of wood of comparatively 
 small dimensions would receive the whole of the energy 
 in that wave. If, however, we took that same receiver 
 to a point 6 feet away from the starting-point, that is 
 to say, 6 feet away from the transmitter, by the time 
 the wave reached it, it would be spread over the circum- 
 ference of a circle 12 feet in diameter, and therefore only
 
 WIRELESS TELEGRAPHY 71 
 
 a very small part of the whole of the wave would affect 
 the receiving piece of wood, or, in other words, it would 
 only receive a small part of the energy in the wave. 
 
 265. It is evident that the- farther we get from the 
 starting-point the smaller the proportion of energy 
 which a given object will receive, although the energy 
 in the whole of the wave remains the same. 
 
 206. The energy radiated depends upon the energy 
 put into the oscillatory circuit producing the waves, 
 and it is therefore important to know on what factors 
 the energy in the oscillatory circuit depends. 
 
 267. In the method of exciting an oscillatory circuit 
 which we are now considering, namely, the spark method, 
 the amount of energy put into the circuit depends 
 upon the capacity of the condenser in the circuit, and the 
 voltage to which it is charged. 
 
 268. The vibrator described in paragraph 2 14 can be 
 energised by applying an initial pressure to the end of 
 the spring, thus bending the spring. 
 
 269. Similarly an oscillatory circuit is energised by 
 applying an initial pressure, or voltage, to the condenser, 
 thus charging it with electricity. 
 
 270. The amount of energy put into the vibrator 
 depends upon the flexibility of the spring, and the initial 
 pressure that is applied to it. Thus the greater the 
 flexibility of the spring the greater the energy put into 
 it by applying a given pressure. Also the greater the 
 pressure applied to it, the greater the energy put into a 
 spring of given stiffness. 
 
 271. Similarly the amount of energy put into an 
 oscillatory circuit depends upon the capacity of the 
 condenser in the circuit, and the initial pressure, or 
 voltage, to which it is charged.
 
 72 WIRELESS TELEGRAPHY 
 
 272. A little consideration, however, will show us 
 that the amount of energy is not directly proportional 
 to the voltage to which the condenser is charged, but is 
 proportional to the square of that voltage. 
 
 273. Taking again the analogous mechanical case of 
 the spring, it is evident that if we apply a force of, 
 say, 1 Ib. to a certain spring by hanging a 1 Ib. weight 
 to it, the spring will be stretched to a certain definite 
 extent, as described in paragraph 29. 
 
 274. Let us suppose, for the purpose of explanation, 
 that the spring is stretched 1 foot when a force of 1 Ib. 
 is applied to it, and let us for the moment suppose 
 that throughout the process of being stretched a force 
 of 1 Ibl is being exerted on the spring by the weight, 
 then it follows from the definition of the mechanical 
 unit of energy given in paragraph 73 that 1 foot- 
 pound of energy has been expended on the spring. 
 
 275. Now the amount which a given spring will 
 extend is directly proportional to the force applied 
 to it, thus in this case, if instead of applying a force 
 of 1 Ib. to the same spring we apply a force of 
 2 Ibs. to it, the spring in this case will be stretched 
 2 feet. 
 
 If for the moment we again suppose that throughout 
 the process of being stretched 2 feet a force of 2 Ibs. is 
 being exerted on the spring, then it follows that in this 
 case 4 foot-pounds of work have - been expended on 
 the spring, because a force of 2 Ibs. has been exerted 
 in moving a body 2 feet. 
 
 276. In other words, although we have only applied 
 twice the force to the spring, we have done four times as 
 much work on it. Now provided there are no losses ID 
 overcoming the friction, all the energy which is done or
 
 WIRELESS TELEGRAPHY 73 
 
 the spring is available for use when the spring is allowed 
 to contract. 
 
 277. We may say, therefore, that the energy stored 
 up in a spring is proportional to the square of the 
 force applied to it. 
 
 278. Now in paragraph 69 we denned the expression 
 " flexibility " as being the measure of .the amount by 
 which a given spring would be extended for a given 
 force ; it follows, therefore, that if in the experiments 
 described above we substitute a spring having twice 
 the flexibility, we shall stretch this more flexible spring 
 a distance of 2 feet by applying 1 Ib. of force to it, thus 
 storing in it 2 foot-pounds of energy. Further, we 
 shall stretch it a distance of 4 feet by applying 2 Ibs. 
 of force to it, thus storing 8 foot-pounds of energy 
 in it. We may say, therefore, that the energy stored 
 up in the spring is directly proportional to the flexibility 
 of the spring as well as being proportional to the square 
 of the force applied to it. 
 
 279. Similarly, if we apply an E.M.F. to a condenser, 
 we cause electricity to flow into the condenser, and the 
 quantity of electricity (which will correspond to the 
 amount by which we stretch a spring) will be directly 
 proportional to the electromotive force applied to the 
 condenser. 
 
 280. Thus, if the capacity of the condenser is one 
 farad and an E.M.F. of one volt is applied to it, one 
 coulomb of electricity will be forced into the condenser 
 (vide paragraph 70), and assuming for the moment that 
 this force is being exerted throughout the process of 
 charging, it follows that one joule of energy will be 
 stored in the condenser (vide paragraph 74). 
 
 281. Further, if we double the force applied to t!-e
 
 74 WIRELESS TELEGRAPHY 
 
 condenser by applying two volts instead of one, we shall 
 force two coulombs of electricity into the same con- 
 denser, and if again we assume that during the whole 
 process of charging, the full force of two volts is being 
 exerted, then in this case it follows that four joules of 
 energy will *be stored in the condenser, because we get 
 two coulombs of electricity upon which a force of two 
 volts has been exerted. 
 
 282. We may say, therefore, that the energy stored 
 up in a condenser is directly proportional to the capacity 
 of the condenser and proportional to the square of the 
 voltage applied to it. 
 
 283. Up to the present, however, we have considered 
 for the sake of simplicity that the force exerted is uniform 
 in the case of the spring throughout the process of 
 stretching and in the case of the condenser throughout 
 the process of 'charging. This, however, is in reality 
 not the case, as will be readily seen by analysing what 
 takes place during the time the spring is being stretched. 
 Taking, for instance, the case of the particular spring 
 which was stretched 2 feet by a force of 2 Ibs. This 
 same spring was stretched 1 foot when a force of 
 1 Ib. was exerted on it. Similarly it will be stretched 
 only 6 inches if a force of half a pound were exerted on 
 it, and so on. 
 
 284. Obviously, therefore, the total force exerted on 
 the spring in stretching it a distance of 2 feet is not 
 the maximum force of 2 Ibs. but the average of all 
 the forces from to 2 Ibs. 
 
 It will be found that this average force is always 
 half the maximum force, assuming that the force applied 
 at the commencement of the operation is zero. 
 
 285. Similarly with a condenser the average force
 
 WIRELESS TELEGRAPHY 75 
 
 applied in charging it from zero to a certain voltage will 
 be half that maximum voltage. Therefore, in order to 
 calculate the energy in the joules stored up in a con- 
 denser, we must take half of the product of the capacity 
 of the condenser and the square of the E.M.F. to which 
 it is charged in volts. This may be stated as an 
 equation, 
 
 E =| CV 2 , 
 
 where E = energy in joules, C =capacity of condenser 
 in farads and V = voltage to which the condenser is 
 charged. 
 
 POWER IN OSCILLATORY CIRCUIT 
 
 286. In paragraph 75 we explained that Power is 
 Energy expended per second. If, then, after charging a 
 condenser to a certain voltage, during which process we 
 expend so much energy upon it, we discharge that 
 condenser we shall be in a position to recharge it 
 expending the same amount of energy as before. It 
 follows, therefore, that the power used will be the 
 product of the energy expended during a single charge, 
 and the number of times per second that it is charged. 
 We may say, therefore, that 
 
 Power=-|-CV 2 xS, 
 
 where S is the number of times per second that the 
 condenser is charged. 
 
 OPEN AND CLOSED OSCILLATORY CIRCUITS 
 
 287. A simple oscillatory circuit (vide paragraph 251) 
 is shown diagrammatically in Fig. 38. Such a circuit 
 is called a closed oscillatory circuit. 
 
 288. Another form of osciPHory circuit is shown in 
 Fig. '39, which represents an aerial wire connected co.
 
 76 
 
 WIRELESS TELEGRAPHY 
 
 earth ; the aerial acts as one side of the condenser, the 
 earth acts as the other ; the wires forming the aerial 
 also form the inductance. 
 
 Such a circuit is called an open oscillatory circuit. 
 
 289 The chief difference in the properties of a closed 
 oscillatory circuit and those of an open oscillatory 
 circuit is that 
 
 An open oscillatory circuit will produce waves having 
 
 FIG. 38. 
 
 a very large amplitude, and is therefore a good radiator, 
 whereas a closed oscillatory circuit will only produce 
 waves of a very small amplitude, and is therefore a very 
 bad radiator. 
 
 290. The chief difference in the composition of a closed 
 oscillatory circuit and an open oscillatory circuit is that 
 
 In a closed oscillatory circuit the capacity and the 
 inductance are more or less seDarated from one another,
 
 WIRELESS TELEGRAPHY 
 
 77 
 
 all the capacity being grouped at one point of the circuit 
 and all the inductance at another ; whereas in an open 
 oscillatory circuit the capacity and inductance are, so 
 to speak, mixed up and distributed along the entire 
 length of the circuit. Thus the aerial wire itsel-f is 
 
 FIG 39. 
 
 :act ng as one plate of a condenser and at the same time 
 as an inductance. 
 
 VARIATION OF WAVE-LENGTHS OF OPEN 
 OSCILLATORY CIRCUITS 
 
 291. The frequency of the aerial must be adjusted 
 so that it produces electric waves of the desired length 
 (vide paragraphs 208 and 257), and this can only be
 
 78 WIRELESS TELEGRAPHY 
 
 done by altering the capacity or the inductance of the 
 aerial circuit. 
 
 TO INCREASE THE WAVE-LENGTH OF AN AERIAL 
 
 292. The capacity of an aerial can be increased by 
 increasing the number of the wires forming it, and the 
 inductance of the aerial can be increased by lengthening 
 the wires. Thus the larger the aerial the longer the 
 electric waves it will produce. 
 
 293. It would, however, be a tedious, and, in fact, 
 impracticable operation to alter the aerial every time it 
 was required to alter the wave-length, and therefore 
 another method is adopted to increase the wave-length 
 of an aerial. 
 
 294. If we connect an inductance in series with 
 another inductance, the total inductance of the circuit 
 is increased. Therefore, if we insert an inductance, 
 as shown in Fig. 40, in series with the aerial, we have 
 increased the total inductance of the circuit, and thereby 
 increased its wave-length. This method is adopted when 
 it is desired to increase the wave-length of the aerial. 
 
 This added inductance tends to make the open 
 oscillatory circuit of the aerial into a closed oscillatory 
 circuit (vide paragraph 290) ; the more inductance we 
 add the nearer do we approach a closed oscillatory circuit, 
 for although some of the total inductance of the aerial 
 is still, so to speak, mixed up with the capacity, a large 
 part of it is separate. 
 
 295. As already stated, a closed oscillatory circuit 
 does not radiate energy to any appreciable extent (vide 
 paragraph 289) ; we therefore reduce the radiating 
 properties of the aerial by adding inductance. There
 
 WIRELESS TELEGRAPHY 79 
 
 is therefore a limit to the amount of inductance that 
 can be so inserted into the aerial circuit without seriously 
 affecting its efficiency as a radiator. 
 
 In practice it is found that the natural wave- 
 . length of an aerial can be about doubled without 
 
 TO INCREASE A 
 
 FIG. 40 
 
 seriously interfering with the radiation , thus we have 
 a simple means of controlling the wave-length over a 
 comparatively large range. 
 
 TO REDUCE THE WAVE-LENGTH OF AN AERIAL 
 
 296. If we place a capacity in series with another 
 capacity, the total capacity, instead of being increased, 
 as might at first be imagined, is reduced.
 
 80 
 
 WIRELESS TELEGRAPHY 
 
 Therefore, if we insert a condenser, as shown in 
 Fig. 41, in series with the aerial, the total capacity is 
 reduced, and therefore the wave-length is also reduced. 
 
 297. The amount by which we decrease the capacity 
 depends upon the capacity of the condenser which is 
 
 r 
 
 TO DECREASE A 
 
 FIG. 41. 
 
 inserted in series, and it is important to remember that 
 the greater the capacity which is inserted in series with 
 another capacity, the less is the reduction of the total 
 capacity ; that is to say, by inserting a small capacity 
 in series with the aerial we reduce the wave-length of 
 that aerial far more than by inserting a large capacity 
 in series with it.
 
 WIRELESS TELEGRAPHY 
 
 81 
 
 298. Similarly, as in the case of adding .inductance, 
 the insertion of a capacity in series with the aerial 
 reduces the radiation of the aerial, but in practice it is 
 found that the natural wave-length' of the aerial can be 
 
 SHORT CIRCUITING 
 
 SWITCH FOR 
 CUTTIN&OUT THE 
 CONDENSER 
 
 FIG. 42. 
 
 about Halved by this means, without seriously interfering 
 with the efficiency of the aerial as a radiator. 
 
 299. It will be seen, therefore, that by connecting a 
 condenser or an inductance in series with the circuit 
 the length of the electric waves emitted by the aerial 
 can be varied from nearly one-half to double the 
 natural wave-length of the aerial without seriously 
 affecting its efficiency as a radiator.
 
 82 WIRELESS TELEGRAPHY 
 
 Fig. 42 shows an aerial with an adjustable con- 
 denser; and an adjustable inductance connected to it. 
 Such an aerial is capable of emitting waves of different 
 lengths within the practical limits mentioned above. 
 
 VARIATION OF WAVE-LENGTHS OF CLOSED 
 OSCILLATORY CIRCUITS 
 
 300. The wave-length of a closed oscillatory circuit 
 
 is varied in exactly the same way as that of an open 
 
 _ oscillatory circuit, namely, by in- 
 
 _L creasing or decreasing the capacity 
 
 I ^ g< and inductance of the circuit. 
 
 TS 301. The form of the circuit, 
 _ | however, lends itself more easily 
 _ to increasing the wave-length by 
 1 < L increasing the capacity of the 
 
 -^ -L. g, 5 circuit, as the capacity, instead of 
 __J ^p being distributed along the whole 
 
 I _ j circuit as in the case of an aerial, 
 , - I is almost entirely concentrated in 
 
 c-^3 the condenser. 
 
 ~3 C 302. if a capacity is connected 
 
 ^3 in parallel with another capacity, 
 
 L - WW^ - ' the total capacity is increased ; thus 
 
 by connecting an additional con- 
 
 denser in parallel with the exist- 
 
 ing condenser of a closed oscillatory circuit we increase 
 
 the wave-length of that circuit. 
 
 303. Fig. 43 shows different methods of increasing 
 the wave-length of a closed oscillatory circuit. 
 
 A represents the original circuit. B represents the 
 same circuit with an additional capacity' connected in
 
 WIRELESS TELEGRAPHY 
 
 83 
 
 parallel, thus increasing the total capacity and thereby 
 increasing the wave-length. C' represents the same 
 circuit with an additional inductance connected in series, 
 thus increasing the total inductance of the circuit and 
 thereby increasing the wave-length. 
 
 304. Fig. 44 represents different methods used for 
 reducing the wave-length of a closed oscillatory circuit. 
 
 1 
 
 T 
 
 T 
 T 
 
 X 
 T 
 
 
 . 1 
 
 ' f 1 
 
 
 T 
 
 r^ 
 
 A represents the original oscillatory circuit. B 
 represents the same circuit with an additional condenser 
 connected in series with it, thus reducing the total 
 capacity of the circuit and thereby reducing the wave- 
 length. C represents the same circuit with some of the 
 inductance cut out, thus reducing the total inductance 
 in the circuit and thereby reducing the wave-length. 
 
 305. The methods above described are only used 
 where a definite jump from one definite wave-length to 
 another is required. If intermediate wave-lengths are
 
 84 
 
 WIRELESS TELEGRAPHY 
 
 required, it is usual to make either the condenser or the 
 inductance adjustable, as shown in Fig. 45, where A 
 represents a circuit in which only the capacity is adjust- 
 able, and B represents a circuit in which only the 
 inductance is adjustable ; C represents a. circuit in which 
 both the capacity and the inductance are adjustable. 
 
 PRODUCTION OF OSCILLATING CURRENTS 
 IN AN AERIAL 
 
 306. The methods employed for causing an aerial to 
 
 FIG. 47. 
 
 oscillate, and thus radiate electric waves, fall under two 
 headings, viz. Direct Excitation and Indirect Excitation. 
 
 DIRECT EXCITATION OP THE AERIAL 
 
 307. We have already explained that an aerial con- 
 nected to earth is an oscillatory circuit, and therefore, 
 for convenience in explanation, we may consider it as a 
 condenser with its two plates connected by an inductance, 
 as shown in Fig. 46, the aerial wires A, Fig. 47, corre- 
 sponding to one plate of the condenser A, Fig. 46 ; the
 
 WIRELESS TELEGRAPHY 85 
 
 earth E, Fig. 47, corresponding to the other plate of the 
 condenser E, Fig. 4G ; and the connecting wire or " down- 
 lead " D, Fig. 47, corresponding to the inductance D, 
 Fig. 46. 
 
 308. It has already been explained (paragraph 237) 
 that if a condenser be charged up and then short- 
 circuited through an inductance, the charge of electricity 
 will not immediately come to rest, but the condenser will 
 over-discharge itself, and the current will oscillate back- 
 wards and forwards until, owing to the 
 resistance of the circuit and the radiation 
 
 of energy, the charge of electricity comes 
 to rest. 
 
 309. In order to excite an oscillatory cir- 
 cuit, such as is shown in Fig. 46, it is there- K K . 45 
 fore only necessary to give the condenser 
 
 an initial charge of electricity, by applying a " voltage " 
 or pressure of electricity across it, and then allow it to 
 discharge itself through an inductance. 
 
 Let us now see how this can best be accomplished. 
 
 310. We can charge up a condenser by connecting a 
 battery across it, as shown in Fig. 48, which will charge 
 the condenser up to the same voltage as the battery ; 
 but in applying a voltage in this way to a condenser, 
 whose two plates are connected together through an 
 inductance to form an oscillatory circuit, as shown in 
 Fig. 46, the electricity, instead of charging up the 
 condenser as desired, will simply flow through the 
 inductive winding D. 
 
 311. It is therefore obvious that during the time the 
 condenser is being charged, we must break the circuit 
 through the inductive winding, as shown in Fig. 49 at 
 the point marked S.
 
 86 WIRELESS TELEGRAPHY" 
 
 312. This, however, destroys the oscillatory circuit, 
 as it prevents the discharge of the condenser through 
 the circuit D, which discharge is required to produce 
 the oscillations. 
 
 313. In order, therefore, to get the conditions right, 
 both for charging up the condenser and for discharging 
 it through the circuit D, it would be necessary to devise 
 some form of mechanism for automatically breaking 
 the discharge circuit and connecting the battery to the 
 condenser at one moment, and then " making " the 
 discharge circuit at the next moment. 
 
 314. This method, however, is impracticable, as, 
 
 apart from the fact that it would 
 be somewhat complicated in oper- 
 ation, an additional drawback 
 arises inasmuch as a very large 
 battery would be necessary in 
 FlG 49 order to charge the condenser up 
 
 to a sufficiently high voltage to 
 
 store up the energy that is required to obtain a useful 
 
 range of transmission. 
 
 315. In paragraph 286 we showed that the power 
 put into a condenser depends upon three things, namely, 
 the capacity of the condenser, voltage to which it 
 "is charged, and the number of times per second that it 
 is charged. Obviously, therefore, the power in the 
 oscillatory circuit is also equal to CV 2 lxS, but in this 
 case S will be the number of times per second that the 
 condenser is discharged into the oscillatory circuit. 
 
 316. As already explained (paragraph 291), the 
 capacity of the aerial is limited by the wave-length it is 
 desired to produce. 
 
 Further, the number of times per second it can
 
 WIRELESS TSLEOBAPHY 8fi 
 
 be charged and discharged is limited by other practical 
 considerations, which will be dealt with later. 
 
 317. It follows that the only method we have of 
 increasing the power in the oscillatory circuit we are 
 considering, namely, an aerial, is by increasing the 
 voltage applied to the condenser. 
 
 Let us take, for example, a small " umbrella " aerial 
 supported by a mast 30 feet high, the length of the 
 radial wires forming the aerial being 70 feet long, as 
 shown in Fig. 47. The capacity of such an aerial 
 would be about '0005 microfarad. 
 
 318. Assuming that our automatic device for charging 
 and discharging the aerial is capable of doing it at the 
 rate of 100 times per second, it can be shown that the 
 initial voltage to which such an aerial would have to 
 be charged in order to radiate 10 watts of power would 
 be about 20,000 volts, assuming that all of the power is 
 expended in radiation and none lost in the resistance 
 of the aerial circuit. 
 
 The impracticability of the method described above 
 becomes obvious, as it would require a battery of 
 about 14,000 dry cells, or 10,000 accumulator cells, to 
 obtain this voltage. 
 
 319. A very much simpler method of exciting an 
 oscillatory circuit presents itself by making use of the 
 properties of a spark gap in conjunction with an 
 induction coil (described in paragraph 118 onwards) 
 
 320. Air in its normal state is nearly a perfect in- 
 sulator ; that is to say, for all practical purposes it will 
 not conduct electricity. If, however, a sufficiently high 
 voltage is applied across an air space the insulation of 
 the air is broken down, allowing the current to pass 
 through the air space, causing a spark to occur, and the
 
 88 WIRELESS TELEGRAPHY 
 
 effect is to make the air space momentarily into a 
 conductor. 
 
 Further, once the spark is formed it will be 
 maintained by a very small current, but as soon as the 
 succession of sparks ceases the air space returns to its 
 normal state of insulation. 
 
 321. By applying this phenomenon: to the oscillatory 
 circuit, as shown in Fig. 50, we get conditions such that 
 daring the time that the condenser is being charged the 
 path through the inductive winding is broken by the 
 air-gap, but as soon as the voltage across the condenser 
 rises to a certain maximum, depending upon the length of 
 the air-gap, the insulation of the air-gap is broken down, 
 a spark occurs across it, and for the moment the gap, 
 instead of being an insulator, becomes a conductor, and 
 allows the condenser to discharge itself through the 
 oscillatory circuit. 
 
 322. As already explained, the condenser not only 
 discharges itself, but over-discharges itself, and the 
 current oscillates backwards and forwards a number of 
 times, until, owing to the resistance of the circuit and 
 the radiation of the energy in the form of waves, the 
 oscillations die down, and the current flowing is not 
 sufficient to maintain the spark. The spark then goes out 
 and the air-gap assumes its normal insulating properties 
 until the next high-voltage impulse is applied to the con- 
 denser, when the same cycle of events takes place. 
 
 323. Such an arrangement is shown diagrammatic- 
 ally in Fig. 50, where A is the impulsive high-voltage 
 generator, B is the condenser, C is the inductance, and 
 D the spark-gap. 
 
 324. This method of excitation can be applied to a 
 closed oscillatory circuit, as already shown, or it can be
 
 WIRELESS TELEGRAPHY 89 
 
 applied to an aerial by connecting the spark-gap between 
 the aerial and earth, as shown in Fig. 51. 
 
 325. An aerial directly excited in this manner is 
 Dually called " plain aerial," and is extremely efficient 
 
 FIG. 50. 
 
 FIG. 51. 
 
 for obtaining a comparatively long range with the use 
 of a small power. 
 
 COUPLED OSCILLATORY CIRCUITS 
 
 326. We have seen how an oscillatory circuit can be 
 energised by charging up a condenser to. a high voltage 
 by means of an induction coil and allowing it to discharge 
 through an inductance and air-gap. Referring to Fig. 50, 
 we see that the right-hand part of the diagram is drawn 
 in thick lines. This is a convenient way of denoting 
 the oscillatory portion of the circuit, and as especially 
 in complicated diagrams it is of great importance to 
 distinguish between the oscillatory circuits and the 
 " low-frequency " circuits, such as the induction coil 
 windings and leads, the reader is advised to follow this 
 plan throughout. 
 
 327. Looking at Fig. 50 it may be asked, how can the
 
 90 WIRELESS TELEGRAPHY 
 
 thick lines be said to form a " circuit " at all since there 
 is a distinct gap D ? A little thought will show us that 
 the gap is only a break in the circuit while the condenser 
 is being charged up by the coil, during which time there 
 are no oscillations and the circuit is not oscillatory ; 
 but when the voltage of the condenser has reached the 
 value necessary to break down the insulation of the air 
 between the spark balls a spark takes place, and the 
 gap then becomes a conductor, and the circuit is truly 
 a closed one ; it is during this time only that the oscilla- 
 tions take place, so that the circuit is completed and 
 forms a " closed oscillatory circuit." Similarly, in the 
 next diagram, Fig. 51, when the spark takes place the 
 circuit is completed and forms an " open oscillatory 
 circuit." 
 
 FACTORS LIMITING THE POWER IN OSCILLATORY 
 CIRCUITS 
 
 328. Speaking in general, we may say that the ultimate 
 factor limiting the power we are able to supply to any 
 oscillatory circuit by any given method is the wave- 
 length to which the oscillatory circuit must be adjusted. 
 
 329. We have seen that the power which can be put 
 into any oscillatory circuit depends upon three factors, 
 namely : 
 
 (1) The number of times per second that the condenser 
 is charged ; 
 
 (2) The voltage to which the condenser is charged ; 
 and 
 
 (3) The total capacity of the condenser. 
 
 We find that all these factors are limited by practical 
 considerations. 
 
 330. Taking the first factor, namely, the number of
 
 WIRELESS TELEGRAPHY 91 
 
 times per second that the condenser is charged, we shall 
 show later that the note produced in the telephones of 
 the receiving station is the same frequency as the frequency 
 of the spark of the transmitting station. 
 
 331. This is called the spark frequency, and must not 
 be confused with the oscillation frequency (vide para- 
 graph 257), of which it is absolutely independent. 
 
 332. The human ear cannot hear a note whose 
 frequency exceeds a certain value, about 15,000 per 
 second, but before this limit is reached another practical 
 difficulty arises, namely, the mechanical construction of 
 a generator to produce such a high frequency. 
 
 333. Taking the second factor, namely, the voltage to 
 which we charge a condenser, we find we are limited 
 here in several directions. 
 
 334. In the first place, if the oscillatory circuit we are 
 energising is an aerial, the mere fact of charging it to a 
 very high potential is in itself bad, for at a certain voltage 
 the ends of the wires begin to " brush " and discharge 
 electricity to the surrounding air, causing a considerable 
 loss of energy. This phenomenon can sometimes be seen 
 at night at the ends of an aerial wire, which appear to 
 be surrounded by a bluish glow. 
 
 335. Secondly, the difficulty of maintaining a suf- 
 ficiently good insulation of the aerial to withstand such 
 a high voltage becomes very serious, especially in wet 
 weather. 
 
 336. Thirdly, to charge up the condenser in an 
 oscillatory circuit, to a high voltage necessitates using a 
 long air-gap in the oscillatoiy portion of that circuit so 
 that it may not break down until a high voltage is 
 reached. Although, as we have said, the air becomes 
 momentarily a conductor when the spark is passing, yet,
 
 92 WIRELESS TELEGRAPHY 
 
 like all conductors, it has a certain resistance, and this 
 resistance increases very rapidly with the length of the 
 gap. 
 
 337. As we have already shown, the introduction of 
 resistance in an oscillatory circuit causes a waste of 
 energy and a rapid dying away of the oscillations. For 
 this reason the use of extremely high voltages in oscilla- 
 tory circuits, necessitating, as it does, a long spark-gap, 
 leads to inefficiency. 
 
 338. Taking the third factor, namely, the total capacity 
 of the condenser, we find that an increase in the capacity 
 in any oscillatory circuit will necessarily increase the 
 length of the wave, unless a corresponding decrease 
 is made in the inductance of the circuit (vide para- 
 graph 256). 
 
 339. We find, however, that in a closed oscillatory 
 circuit we can reduce the inductance (and therefore 
 increase the capacity) to a far greater extent than we 
 can in an aerial. 
 
 340. In the case of an aerial we can increase the 
 capacity by bringing the aerial nearer the ground, and 
 thus reducing the thickness of the dielectric (vide para- 
 graph 19), but this decreases the range we can obtain. 
 
 341. We can also increase the capacity of an aerial 
 by increasing the length of the wires forming the aerial, 
 but this at the same time increases the inductance in 
 the wires, and therefore increases the wave-length. 
 
 342. The only other way of increasing the capacity 
 of an aerial is to increase the num'oer of wires forming it. 
 This, however, will not increase the capacity sufficiently 
 for our purpose, and, moreover, tends to make the aerial 
 costly and unwieldy. 
 
 343. As already pointed out, however, in the case of
 
 WIRELESS TELEGRAPHY 93 
 
 a closed oscillatory circuit, the proportion of the capacity 
 to the inductance of the circuit for a given wave-length 
 can be made far greater than in the case of an aerial, and 
 therefore we can make such a circuit capable of utilising 
 a larger amoujit of power for the same wave-length, the 
 same spark frequency, and the same voltage. 
 
 344. A closed oscillatory circuit, however, is not a 
 good radiator (vide paragraph 289), and is therefore not a 
 good substitute in this respect for the open oscillatory 
 circuit provided by the aerial. If, however, we can 
 combine the good energy-storing property of the closed 
 oscillatory circuit and the good energy- radiating property 
 of the aerial, we shall obtain the best results for a 
 limited wave-length. 
 
 345. This is the plan on which is based the " coupled- 
 circuit " transmitter now in general use, a diagram of 
 which is shown in Fig. 52. 
 
 346. The closed oscillatory 
 circuit X (Fig. 52) is excited in 
 the way described in paragraph 
 319. The oscillating currents 
 set up pass to and fro round the 
 circuit, which includes a coil 
 L, consisting of one or more 
 turns of wire. This coil is so 
 placed with respect to another 
 coil N connected in the aerial 
 circuit that the two coils exer- 
 cise mutual induction (see paragraph 114) on each other. 
 One end of this second coil is connected to the aerial A, 
 and the other end to earth E. The oscillating currents 
 flowing through L create, through the mutual inductance 
 of the two coils, oscillating currents in the aerial.
 
 94 WIRELESS TELEGRAPHY 
 
 347. If, however, the values of capacity and induct- 
 ance of the aerial circuit are arranged so that the aerial 
 has a frequency different .from that of the closed circuit, 
 the aerial will try to oscillate at its own frequency (vide 
 paragraph 252) in opposition to the oscillations put into 
 it by the closed circuit, with the result that one set of 
 oscillations will interfere with the other, and very little 
 energy will be transferred from the closed to the open 
 circuit. Under these conditions the two circuits are said 
 to be " out of tune." 
 
 348. To understand what happens when the circuits 
 are out of tune, we may take it that the first oscillation in 
 the closed circuit induces a wave in the aerial coil ; this 
 wave travels up the aerial, reaches the free insulated 
 end, turns back and. tries to return to earth ; but 
 on its way there it meets another wave coming up 
 the aerial, induced by the second oscillation in the 
 closed circuit, which is not " keeping time " properly 
 with the aerial, and these two waves partly destroy 
 one another. 
 
 349. But if the aerial circuit is so arranged as to have 
 the same frequency as the closed circuit, the first wave, 
 instead of meeting a contrary wave, will travel down to 
 earth unhindered, and as it swings back again it will 
 find the second wave, induced from the closed circuit, 
 ready to join it in its progress up the aerial and down 
 again to earth ; and this will go on, one wave adding on 
 to the others already in the aerial, until the condenser C 
 is discharged ; that is to say, until the energy originally 
 stored up in the closed oscillatory circuit is transferred to 
 the aerial. Under these conditions the two circuits are 
 said to be " in tune." 
 
 350. We may say, then, that in order to excite
 
 WIRELESS TELEGRAPHY 
 
 95 
 
 efficiently one oscillatory circuit from another in which 
 oscillating currents are flowing, it is necessary that the 
 two circuits have the same frequency. 
 
 351. A simple experiment can be made with pendu- 
 lums to illustrate this point. A piece of string is 
 stretched between two fixed points (Fig. 53), and two 
 pendulums, P! and P 2 , are hung from it a short distance 
 apart. 
 
 352. Now if these pendulums have the same time of 
 
 P: 
 
 FIG. 53. 
 
 swing, and therefore the same frequency, they may be 
 said to be in tune, and it will be found that if P x (which 
 may be taken to represent the closed oscillatory circuit) 
 be started swinging, it will, owing to its being coupled 
 to P 2 by the string, gradually start a similar swing 
 in P 2 . 
 
 353. The swing in P 2 will get greater and greater until 
 the energy that was originally put into P l is transferred 
 to P 2 , and Pj will have come to rest. 
 
 354. If, however, P 2 be made shorter or longer than
 
 96 WIRELESS TELEGRAPHY 
 
 Pj, so as to have a different frequency, the two pendu- 
 lums may be said to be out of tune, and it will be 
 found that, although a certain amount of swing will be 
 induced in P 2 , the two pendulums will interfere with 
 one another, and both will come to rest after erratically 
 jerking about. 
 
 355. The closed oscillatory circuit is spoken of as the 
 
 FIG. 54. 
 
 " primary " circuit ; the two coils L and N form 
 together an " oscillation transformer " or " jigger," 
 the coil L being the " jigger-primary," and the coil N 
 the " jigger-secondary." 
 
 356. In order to " tune " the primary circuit to the 
 aerial, it is usual to connect a "variable inductance" 
 both in the aerial circuit and in the primary circuit, as 
 shown in Fig. 54. Such an inductance in the primary 
 circuit is called the " primary tuning inductance," and 
 in the aerial circuit is called an " aerial tuning induct- 
 ance."
 
 WIRELESS TELEGRAPHY 
 
 97 
 
 THE AUTO-JIGGER 
 
 357. In the above method of indirect excitation we 
 had two entirely separate circuits, the primary circuit 
 and the aerial circuit, connected only by the mutual 
 induction of jigger-primary and jigger-secondary ; and 
 we saw that, provided each of these two circuits was 
 tuned to the same wave-length, the arrangement offered 
 us an excellent combination a good storer of energy 
 combined with a good radiator of energy. 
 
 358. There is another form of indirect excitation, 
 using what is called an " Auto-jigger," which at one time 
 was fairly extensively used, and is still popular among 
 amateurs owing to its simplicity. 
 
 359. In an auto-jigger we still have the two circuits 
 the primary circuit with its 
 
 condenser and j igger-primary , 
 and the aerial circuit with 
 its aerial, its tuning induct- 
 ance, its jigger -secondary, 
 and its earth, and these must 
 be tuned to the same wave- 
 length just as in the case of 
 the ordinary jigger, but in the 
 case of the auto- jigger the 
 primary circuit is in actual 
 metallic connection with the 
 aerial circuit ; in fact, the 
 jigger-primary is formed of a 
 certain number of turns of 
 the jigger-secondary itself. 
 
 360. Thus in Fig. 55, which illustrates the auto- 
 jigger, the aerial circuit consists of the aerial A, the
 
 98 WIRELESS TELEGRAPHY 
 
 aerial tuning inductance B, the jigger-secondary CD, and 
 the earth connection E; while the primary circuit consists 
 of the condenser F, the spark-gap G, and the jigger- 
 primary D, which is merely a certain number of turns 
 of the jigger-secondary CD. 
 
 361. With such an arrangement we have the same 
 advantage as with the ordinary jigger namely, a good 
 storer of energy (the closed oscillatory circuit containing 
 the large condenser F) transferring its energy, through 
 the action of the coil D, to the good radiator, the open 
 oscillatory circuit A, B, C, D, E, 
 
 REACTION OF SECONDARY ON PRIMARY 
 "362. The behaviour of a coupled-circuit transmitter, 
 whether an ordinary jigger or an auto- jigger, is less 
 simple than would appear at first sight. 
 
 363. It might seem natural to suppose that since the 
 primary circuit has the same wave-length as the second- 
 ary or aerial circuit, it would simply transfer its energy 
 (put into it by the induction coil) to the aerial in the 
 form of oscillations of the same frequency ; and that 
 the aerial would radiate out this energy in the form of 
 aether waves of a length corresponding to that frequency. 
 
 364. But we must remember that just as the currents 
 flowing to and fro in the jigger-primary induce currents 
 in the otherwise passive secondary, so do the currents 
 thus made to flow in the secondary act on the primary 
 and induce currents in it ; so that when the primary 
 circuit has given up its energy to the secondary the 
 latter starts giving back some of its energy to the prim- 
 ary, which returns it to the secondary, and so on. 
 
 365. This goes on till so much energy has been re- 
 moved from the circuits by losses in resistance and by
 
 WIRELESS TELEGRAPHY 99 
 
 radiation from the' aerial that the current in the primary 
 has no longer power to cross the spark-gap, when the 
 process stops until it is started again by the induction 
 coil charging up the condenser once more. 
 
 3CG. This will be more easily understood by referring 
 again to the pendulum experiment described in para- 
 graph 351. 
 
 367. In this paragraph we only followed the action 
 of the pendulums up to the moment when the driving 
 pendulum P l had transferred its energy to P 2 , but if 
 we watch their action still' further we see that P 2 now 
 becomes the driving pendulum, and its energy will 
 gradually be transferred back to P lf and this transfer 
 of energy goes on backwards and forwards until so much 
 energy has been lost in friction in the air and string that 
 both pendulums come to rest. 
 
 Now this rather complicated give-and-take process 
 has a peculiar effect on the wave set up in the aerial. 
 
 368. The result of coupling a closed oscillatory circuit 
 to an open oscillatory circuit, each of which is tuned to 
 the same wave-length, is the production of two wave- 
 lengths, one longer and the other shorter than the wave- 
 length to which both circuits have been tuned. These 
 two wave-lengths are known as the Resultant Wave- 
 lengths. 
 
 369. It is not an easy subject to understand, but it is 
 a very important one, and our readers are recommended 
 to take pains to master it. 
 
 RESULTANT WAVE-LENGTHS OF COUPLED CIRCUITS 
 
 370. The jigger-primary has a certain amount of 
 inductance (which has already been defined) due entirely 
 to itself its number of turns, its diameter, the spacing
 
 100 WIRELESS TELEGRAPHS 
 
 of its turns, etc. ; this is called the self-inductance of the 
 primary. 
 
 371. Similarly the jigger-secondary has a certain self- 
 inductance, due to its number of turns, diameter, spacing 
 of turns, etc. 
 
 372. But besides these two self-inductances, which 
 would remain unaltered if the primary were taken to the 
 Equator and the secondary kept at home, there is a 
 third inductance which affects both primary and second- 
 ary, and which is due to the proximity of the one coil to 
 bhe other. 
 
 373. This is called the mutual inductance; thus the 
 primary has, in addition to its self-inductance, the 
 mutual inductance due to the effect of the secondary, 
 and the secondary has, in addition to its self-inductance, 
 the mutual inductance due to the presence of the primary. 
 
 374. This mutual inductance depends on the position 
 of the primary with regard to the secondary, on their 
 distance apart, and on the number of turns acting on 
 each other. 
 
 375. The mutual inductance of two such coils, though 
 it is an abstract kind of thing which cannot be seen, is 
 nevertheless a definite quantity, and is very important, 
 as it is. .through the agency of the mutual inductance 
 that the primary circuit is able to transfer its energy to 
 the aerial circuit. 
 
 376. By making a simple experiment with two coils 
 of wire, we can demonstrate what is the effect of the 
 mutual inductance of the two coils on the total self- 
 inductance of the coils. 
 
 377. Let us suppose that we have two separate coils 
 A and B, each consisting of two turns as shown in 
 Fig. 56, placed at such a distance apart that none. of the
 
 WIRELESS TELEGRAPHY 
 
 101 
 
 magnetic lines of force produced by a current flowing 
 through A pass through the coil B, and, therefore, also 
 none of the lines of force produced by B pass through A. 
 If now we connect the two coils in series, as shown, and 
 pass a current through both of them, a certain number of 
 
 lines of force will be induced by the coil A, and an equal 
 number will also be induced by the coil B. 
 
 378. Now the inductance of a coil is proportional to 
 the number of lines which thread it, when unit current 
 flows through that coil, multiplied by the number of turns 
 in the coil (assuming that all the lines of force produced by 
 the coil thread all the turns) . Thus, if we assume that the 
 coils A and B each consist 
 of two turns, aud that when f 
 a current of one ampere flows 
 through them four lines of ,' 
 force are produced by each 
 coil, then the inductance of 
 each coil will be 4x2 = 8, 
 and under these conditions 
 the total inductance of the 
 two coils together will be 8+8 =16. 
 
 37 ( J. Let us now suppose that the two coils be placed
 
 102 WIRELESS TELEGRAPHY 
 
 rather closer to one another, as shown in Fig. 57, so 
 that half of the lines produced by coil A also thread 
 the coil B, and also half of the lines of the coil B thread 
 the coil A. Then it is evident that, when a current of 
 one ampere flows through the two coils, although each 
 coil individually produces the same number of lines 
 as before, the total number of lines threading each coil 
 will be increased by one half the lines in the other coil. 
 
 380. We have assumed that each coil produces 4 lines 
 of force ; therefore, in this case, a current of one ampere 
 will have iJie effect of causing 6 lines of force to thread 
 each coil, and the total inductance of the two coils 
 together will then be (6 x 2) + (6 x 2) -12 + 12 -24. 
 
 381. Now the difference between this total induct- 
 ance and the total inductance of the two coils, when 
 a great distance apart, is the mutual inductance of 
 the two coils. That is to say, in the case illustrated, in 
 Fig. 57, mutual inductance =24 - 16 =8. 
 
 382. Similarly, it will be found that if the two coils 
 are placed so close together that all the lines induced by 
 each coil individually thread the other coil, the total 
 inductance of the two coils together will then be 
 2 (turns) x 8 (lines) + 2 (turns) x 8 (lines) =16 + 16 =32, 
 and the mutual inductance in this case will be 32 - 16 =16. 
 
 383. These results will be obtained, provided the 
 direction of the magnetic lines is the same in both coils 
 (vide paragraph 82). 
 
 384. If, however, we connect up the two coils as 
 shown in Fig. 58, so that the current flowing in one coil 
 is in the opposite direction to that flowing through the 
 other coil, thus inducing magnetic lines in one coil in 
 the opposite direction to the magnetic lines in the other, 
 then different results will be obtained when the two
 
 WIRELESS TELEGRAPHY 
 
 103 
 
 coils are brought together, because the magnetic field 
 produced by one coil will tend to neutralise the magnetic 
 field produced by the other coil. 
 
 385. When the two coils are remotely separated, as 
 
 FIG. 58. 
 
 shown in Fig. 58, so that none of the lines of force in- 
 duced by A pass through B, and vice versa, then the total 
 inductance of the two coils will be the same as in the 
 case illustrated in Fig. 50, namely 16. But when the 
 two coils are placed closer to one another, as shown in 
 Fig. 59, so that two lines produced by each coil thread the 
 other, then these two lines - -^--,.,^. . , 
 
 being in the opposite direction 
 will neutralise two of the 
 lines induced by that coil, so 
 that in this case a current 
 of one ampere will have the 
 effect of causing only two 
 lines of force to thread each 
 coil, and, therefore, the 
 total inductance of the two coils together will then 
 be (2x2) + (2x2) 8. 
 
 386. Now the dill'ercnce between this total inductance 
 and the total inductance of the tv.o coils when a great 
 distance apart, as shown in Fig. 58, that is to say 8 - 16, 
 
 FIG.
 
 104 WIRELESS TELEGRAPHY 
 
 will again represent the mutual inductance of the two 
 coils, which it will be seen is the same value, but opposite 
 in sign to the value of the mutual inductance, when the 
 two magnetic fields were in the same direction. In that 
 case the mutual inductance, as we showed, was +8. 
 In this case the mutual inductance is - 8. 
 
 387. Similarly, it will be found that if the two coils 
 are placed so. close together that all the lines induced 
 by each coil individually thread the other coil, the total 
 inductance of the two coils together will be 0, and the 
 mutual inductance in this case will therefore be - 16. 
 
 388. Returning to the problem of coupled oscillatory 
 circuits, we can now follow the effect of the mutual 
 inductance on the resultant wave-length. 
 
 Let us suppose, for the sake of simplicity, that the 
 self-inductance of the primary circuit is equal to that 
 of the secondary circuit ; we know that the wave- 
 lengths of the two circuits are the same, but as a rule the 
 inductance of the primary is much less than that of the 
 secondary, so as to enable the primary condenser to be 
 of much larger capacity than that of the aerial ; there is 
 no reason, however, why we should not, for the sake of 
 argument, make the two capacities equal, and therefore 
 the two inductances also equal. Let each of these 
 inductances be L, and let the mutual inductance between 
 primary and secondary be M. 
 
 389. Now owing to the give-and-take process which 
 we described above, the relative directions of the oscilla- 
 tory currents flowing in the primary and secondary coils 
 are continually changing with this result : it makes the 
 mutual inductance M add itself to the self-inductance L 
 at one moment, and then, a fraction of a second later, it 
 makes M subtract itself from L. The result is that at the
 
 WIRELESS TELEGRAPHY 105 
 
 first moment each circuit behaves as if its total inductance 
 were L + M, and at the next moment as if it were L - M. 
 But these moments are so close together separated 
 only by such an infinitely small fraction of a second 
 that what happens is that the circuits appear to possess 
 these two values of inductance at the same time ; so that 
 they behave as if, instead of each having an inductance 
 L, they each had two different inductances, L + M and 
 L-M. 
 
 390. But if a circuit has two inductances and one 
 fixed capacity, it is clear that it will give two wave- 
 lengths ; and, as a matter of fact, the result of the 
 give-and-take action between primary and secondary is 
 that the aerial sends out two waves, one longer and one 
 shorter than the wave to which both the primary and 
 aerial circuits were tuned. 
 
 391. It is clear that. the production of these two 
 waves is governed by the size of M compared with L ; 
 if we make M very small compared with L by increasing 
 the distance between the primary and secondary of the 
 jigger, L + M will only be very slightly larger than L - M, 
 so that the two waves will be so nearly equal as to be 
 indistinguishable. 
 
 392. So if we move the jigger-primary farther and 
 farther away from the jigger-secondary, we can reduce 
 M and make the two waves approach nearer and nearer 
 to one another, till finally they merge into one wave- 
 length which will be of the same value as that of the 
 circuits taken by themselves. 
 
 393. We assumed, for the sake of simplicity, at the 
 beginning of paragraph 388, that the inductance of the 
 primary was equal to that of the secondary. If, as is 
 usual, these inductances arc different, the same thing
 
 106 WIRELESS TELEGRAPHY 
 
 holds good, except that the simple formula of L + M 
 and L - M becomes somewhat more complicated and 
 elaborate. 
 
 394. To summarise we may say that : 
 
 (1) Two oscillatory circuits can be coupled together 
 for the purpose of exciting one from the other. 
 
 (2) The two circuits must be both tuned to the same 
 wave-length. 
 
 (3) The result is the production of two distinct waves, 
 one longer and one shorter than the normal wave-length 
 of either circuit taken separately. 
 
 (4) The closer two circuits are coupled together the 
 greater the difference between the two resulting wave- 
 lengths. 
 
 CALCULATION OF THE DEGREE or COUPLING 
 
 395. For convenience the degree of coupling between 
 two oscillatory circuits is expressed as a percentage of 
 the full coupling. 
 
 A, large degree of coupling is known as a " close 
 coupling," and a small degree of coupling as a " loose 
 coupling." 
 
 396. If two oscillatory circuits were fully coupled 
 the two resulting waves would be so far apart that the 
 lower wave would be sensibly zero, and the only wave- 
 length left would be \/2 or 1'4 times the wave-length 
 of the two circuits taken separately. 
 
 397. In practice such conditions cannot be obtained, 
 because even if the primary and secondary coils were 
 so. close together that all the lines of force induced by 
 the primary coil threaded the secondary coil, and vice 
 versa, these are not the only lines of force induced by 
 each circuit.
 
 WIRELESS TELEGRAPHY 107 
 
 398. For instance, the currents flowing through the 
 conductors which connect the primary coil to the con- 
 denser and through the primary tuning inductance, 
 induce lines of force round those conductors which do 
 not thread any part of the secondary circuit ; similarly, 
 the currents flowing through the radiating portion of 
 the aerial, as well as through the aerial tuning inductance 
 coil, induce lines of force which do not thread' any part 
 of the primary circuit. 
 
 399. Since the degree of coupling is the proportion of 
 the number of lines which thread the other circuit to the 
 total number of, lines induced, it is obvious that a full 
 coupling between the primary circuit and the aerial is 
 impossible. These conditions will be more clearly under- 
 stood by referring to Fig. 60, which shows diagrammatic- 
 ally the distribution of the lines of force induced in the 
 two circuits. 
 
 400. It will also be seen that the greater the amount of 
 tuning inductance that is inserted in either the primary 
 or secondary circuit, the less will be the degree of 
 coupling between the two circuits, unless these induct- 
 ances be so arranged relatively to one another that the 
 lines of force induced by the currents flowing through 
 them thread the inductance coils of the other circuit. 
 
 401. It is evident that any " outside" inductance (i.e. 
 any inductance which is not mutually acting on the other 
 circuit) included in either the Primary or the Secondary 
 circuits is tending to weaken the coupling obtainable 
 between the two circuits. 
 
 It is not, however, necessarily an advantage to have 
 a close coupling between the two circuits, and in fact 
 too close a coupling has many disadvantages. 
 
 402. In practice, it is usual to allow for a maximum
 
 108 
 
 WIRELESS TELEGRAPHY 
 
 coupling of 15 per cent or 20 per cent between the 
 primary circuit of the transmitter and the aerial circuit, 
 and where very sharp teming is required this coupling 
 
 7F^<;,,;"-'y 
 
 '/! \-_y 
 
 
 Primary Tuning 
 
 ^~xS^ 
 
 FIG. 60. 
 
 is frequently reduced to 1 per cent or 2 per cent, by 
 methods described later. 
 
 403. As a matter of fact, with commercial stations, 
 a regulation has been laid down by the International
 
 WIRELESS TELEGRAPHY 109 
 
 Wireless Convention that no station is allowed to use a 
 closer coupling than 15 per cent. 
 
 404. Since the difference between the two resulting 
 wave-lengths of coupled circuits depends upon the degree 
 of coupling between the circuits, it follows that we can 
 calculate the coupling if we know the values of these 
 two waves, and the following formula, although not 
 exact, will give a very near approximation of the per- 
 centage of coupling. 
 
 405. If k = percentage of coupling between two 
 circuits, and Xj is the wave-length of the longer of the 
 two resulting waves, X 2 the wave-length of the shorter 
 of the two resulting waves, and X the wave-length of 
 each of the circuits taken separately, then 
 
 k = ? x 100 (approximately). 
 
 ^o 
 
 406. Let us apply this formula to a practical case. 
 On a certain vessel a wireless installation had Jbeen 
 
 fitted. The wave-length of the closed oscillatory circuit 
 was adjusted to 600 metres, and that of the aerial circuit 
 to the same. When the primary oscillatory circuit was 
 coupled to the aerial it was found that the resulting 
 wave-lengths were 570 metres and 630 metres respec- 
 tively. 
 
 From this it can be calculated that the coupling 
 between the two circuits was 10 per cent, for 
 
 -5xlOi 
 
 600 
 
 = 10 per ceni
 
 110 WIRELESS TELEGRAPHY 
 
 METHODS OF VARYING THE COUPLING BETWEEN 
 TWO OSCILLATORY CIRCUITS 
 
 407. Many different ways are employed for varying 
 the coupling between the inductive windings of two oscil- 
 latory circuits ; all of them, however, are based on the 
 principle of varying the proportion between the number 
 of lines induced by one coil, which thread the other coil, 
 and the total number of lines induced. 
 
 408. The method most commonly used to vary the 
 coupling between the primary circuit and the aerial 
 circuit of a transmitter is to slide the secondary winding 
 away from the primary winding. 
 
 This method is illustrated in Figs. 61 and 62, where 
 A is the inductive 
 winding of the open 
 oscillating circuit, that 
 
 B secondary' and B the 
 
 FTC, 61. Ffa.,6* inductive winding ^ of 
 
 the closed oscillating 
 circuit, that is to say, the jigger-primary. 
 
 409. When one of these two coils is immediately 
 above the other, as shown in Fig. 61, the coupling 
 between the two is at its maximum, but when the 
 secondary winding is moved until it occupies a position 
 near the edge of the primary winding, as shown in Fig. 
 (52, the coupling is at its minimum. 
 
 410. Another method of adjusting the coupling 
 between two circuits is to alter the relative angular 
 position between the axes of the two windings. 
 
 When these two axes are in line the coupling is 
 at its maximum, and when they are at right angles to
 
 WIRELESS TELEGRAPHY 111 
 
 one another the coupling is at its minimum. In this 
 case the lines of force induced by the one coil do not 
 thread the other coil, but pass along the conductors. 
 
 This method is illustrated in Figs. 63 and 64. Iu 
 Fig. 63 the axes of the two 
 coils are in line, and the 
 coupling is at its maximum, 
 whereas in Fig. 64 the axes 
 of the two coils are at 
 right angles to one another, 
 and the coupling is at its 
 minimum. FIG. 63. FIG. 64. 
 
 THE WAVEMETER 
 
 411. The Wavemeter is an instrument for measuring 
 the frequency and therefore the length of the wave or 
 waves emitted by oscillatory circuits. 
 
 412. Briefly, it consists of a closed oscillatory circuit 
 whose wave-length, or more strictly speaking, frequency, 
 it is possible to vary, connected to a detector, by means 
 of which it is possible to tell the comparative amount of 
 current flowing in the oscillatory circuit. 
 
 413. To measure the wave-length of an oscillatory 
 circuit, the instrument is brought sufficiently near some 
 part of that circuit, so that the oscillating currents 
 flowing in it will induce similar currents in the oscillatory 
 circuit of the Wavemeter. , v 
 
 414. In paragraph 346 we showed that when two 
 oscillatory circuits are coupled together, one of which 
 is set oscillating, similar oscillations are induced in the 
 second circuit, provided that the two circuits are in tune ; 
 that if they are out of tune, although a certain amount
 
 112 WIRELESS TELEGRAPHY 
 
 of current is still induced in the second circuit, this 
 current will be comparatively feeble and erratic, but will 
 rapidly rise as the circuits are brought nearer and nearer 
 into tune, reaching a maximum when the two circuits 
 are quite in tune. 
 
 415. When we bring the wavemeter near another 
 oscillatory circuit, we are in effect coupling the two 
 circuits together, and we shall obtain similar pheno- 
 mena. 
 
 416. By adjusting the frequency of the wavemeter 
 circuit, and at the same time noting, by means of the 
 detector, the comparative amount of current induced 
 into it, we can tell exactly when the wavemeter circuit 
 is in tune with the circuit we are measuring, for when the 
 circuits are in tune the current will be strongest. If we 
 know the value- of the wave-length, to which the wave- 
 meter circuit is adjusted, it follows that this wave-length 
 is also the wave-length of the circuit we are measuring. 
 
 THE OSCILLATORY CIRCUIT OP A WAVEMETER 
 
 417. In practice it is usual to vary only the capacity 
 of the circuit, keeping the inductance a constant value 
 throughout. This for various practical reasons is found 
 
 to be more convenient than ad- 
 justing the inductance. Fig. 65 
 shows such an oscillatory circuit, 
 where L is the fixed inductance 
 and C the variable condenser. 
 
 418. We have already learnt that the wave-length 
 of an oscillatory circuit depends upon the product of 
 the capacity and the inductance of that circuit : it 
 follows, therefore, that such a circuit can be " tuned up," 
 or, in other words, adjusted to the same frequency as
 
 WIRELESS TELEGRAPHY 113 
 
 that of the circuit whose wave-length it is required to 
 measure. 
 
 419. Practical considerations limit the maximum 
 and minimum values of the capacity to which the con- 
 denser can be adjusted, and therefore limit the maximum 
 and minimum wave-lengths to which the circuit can be 
 tuned. 
 
 420. An illustration of a variable condenser is shown 
 in Fig. 66. The principle on which it is constructed 
 will be described later, but for the present it is sufficient 
 to know that its capacity is varied 
 
 by turning the handle A. Fixed to 
 this handle is a pointer B, which 
 passes over a scale C. This scale is 
 carried half-way round the circum- 
 ference of the condenser, and is 
 divided into a number of equal divi- 
 sions which are marked from to 
 100. When the handle of the condenser is so turned 
 that the pointer indicates the figure 0, the capacity of 
 the condenser is at its minimum, and as the pointer 
 passes up the scale the capacity of the. condenser increases 
 until it arrives at its maximum capacity when the 
 pointer indicates the figure 100. 
 
 421. When this condenser forms part of an oscillatory 
 circuit, the inductance of which is fixed, as in the case 
 of the wavemeter, it follows that the wave-length of the 
 circuit will have a definite value for every definite position 
 of the condenser pointer 
 
 422. These wave-lengths are carefully and accurately 
 measured after the instrument is made (by methods 
 which for the purpose of this book it is unnecessary to 
 explain), and a list or chart is supplied with the instru- 
 
 I
 
 114 WIRELESS TELEGRAPHY 
 
 xnent giving the wave-lengths of the circuit corresponding 
 to each scale reading of the condenser. 
 
 423. By the use of this chart we can find out to what 
 wave-length the instrument has been adjusted by refer- 
 ring first to the condenser reading and then reading off 
 the chart the value of the wave-length corresponding 
 to that condenser reading. 
 
 THE " DETECTOR " CIRCUIT OF A WAVEMETER 
 
 424. In order to tell when the wavemeter circuit is 
 in tune with the other circuit, we must find a means of 
 measuring the current in the wavemeter circuit (vide 
 paragraph 416). 
 
 425. It is not necessary to know the actual value of the 
 current, but only its comparative value, so that a detector 
 Which will respond proportionally to the amount of 
 current passing through it will suit our purpose. 
 
 426. The telephone receiver is a very suitable instru- 
 ment for this purpose ; for one thing, it is extremely 
 sensitive to even the smallest current passing through it, 
 and for another thing, by judging the loudness of the 
 sound in the telephone we can judge the comparative 
 amount of current passing through it. 
 
 427. High-frequency Oscillating Currents, however, 
 will not affect the telephone receiver, as the alternations 
 are much too rapid for the diaphragm to follow. 
 
 428. So that, to enable us to detect the high-frequency 
 currents produced in the wavemeter, these currents, 
 or at all events that part of them which is made to pass 
 through the telephones, must be rectified, or, in other 
 words, converted into uni-dir$ctional currents.
 
 WIRELESS TELEGRAPHY 115 
 
 THE USE OF CRYSTALS 
 
 429. It is found that certain crystals, such as car- 
 borundum, have the property of rectifying high-frequency 
 oscillating currents. They really act as non-return 
 valves, allowing the current to pass through them in one 
 direction only, which is equivalent to converting the high- 
 frequency current into a uni-directional current. 
 
 430. These crystals, however, have an extremely 
 high resistance, and for this reason cannot be inserted 
 directly in the oscillatory circuit. A little thought, 
 however, will show us that it is not necessary to insert 
 either the crystal or the 
 
 telephones in the oscillatory 
 circuit. 
 
 431. The current in the 
 
 oscillatory circuit, as we FlG 67 
 
 know, charges up the con- 
 denser of that circuit to a certain voltage ; the greater 
 the current induced in the inductance coil of the wave- 
 meter the higher the voltage to which the condenser will 
 be charged. 
 
 432. If, therefore, we place our crystal in series with 
 the telephone across the condenser, as shown in Fig. 67, 
 we shall not in any way interfere with the oscillatory 
 properties of the oscillatory circuit, but we shall get a 
 certain current passing through the crystal and the 
 telephones, the amount of which will depend, upon 
 the voltage to which the condenser is charged, and 
 therefore will indicate the amount of current induced 
 in the oscillatory circuit. Moreover, the crystal will 
 rectify this current, so that in effect we shall get a 
 uni-directional current passing through our telephones.
 
 116 WIRELESS TELEGRAPHY 
 
 433. The current also will be an intermittent current, 
 the number of interruptions per second being the same as 
 the number of sparks per second in the oscillatory circuit 
 which is being measured (vide paragraphs 533 to 536). 
 We shall therefore get a buzz, or note, in the telephone 
 corresponding exactly to that produced by the spark of 
 the transmitter, and proportional in its loudness to the 
 amount of current induced in the oscillatory circuit of 
 the wavemeter. 
 
 434. It is clear, therefore, that if we vary the adjust- 
 able condenser of the wavemeter circuit, and at the same 
 time listen to the sound in the telephones, when this 
 sound is loudest the wavemeter circuit is in tune with 
 the oscillatory circuit, and by noting the position of the 
 condenser thus obtained, and referring this reading to 
 our chart, we find the value of the corresponding wave- 
 length, and therefore the length of the wave emitted 
 by the oscillatory circuit being measured. 
 
 CONSTRUCTION OF AN ADJUSTABLE CONDENSER 
 
 435. The construction of an adjustable condenser is 
 illustrated in Figs. 68, 69, and 70. A number of semi- 
 circular metal plates A are connected together, and held 
 rigidly parallel to one another and at a sufficient distance 
 apart to allow the second set of metal plates B to pass in 
 between them. Fixed to the upper sides of both the A 
 plates and the B plates are ebonite plates C of the same 
 shape. The second set of metal plates B are held 
 together on a spindle D, which can be rotated by the 
 handle E which is fixed to one end of the spindle. 
 
 The fixed plates A form one side of the condenser, 
 and the movable plates B form the other side of the 
 condenser, the dielectric of the ebonite being formed
 
 WIRELESS TELEGRAPHY 117
 
 118 WIRELESS TELEGRAPHY 
 
 by the ebonite plates C. When these plates are in the 
 positions shown in Fig. 68, the capacity of the condenser 
 is practically zero, but if the movable plates B are 
 rotated until they interleave themselves with the A 
 plates and occupy the position as shown in Fig. 69, the 
 capacity of the condenser is increased to half its maximum 
 capacity, as half the surface of the A plates is acting 
 through the dielectric on to half the surface of the B 
 plates. Finally, if the B plates be still further rotated 
 until they occupy positions entirely under the A plates, 
 as shown in Fig. 70, the capacity of the condenser is at 
 its maximum. It is only necessary to fix to the moving 
 plates a pointer F, which will pass across the scale G, 
 and thus denote the exact position of the plates. 
 
 WIRELESS TELEGRAPH RECEIVERS 
 
 436. A wave has the property of producing a dis- 
 turbance similar to the disturbance which started the 
 wave (vide paragraph 156). 
 
 437. We have shown that oscillating currents flowing 
 in an open oscillatory circuit, such as an aerial, will 
 produce electric waves. It follows, therefore, from the 
 above, that electric waves will produce oscillating currents 
 in an aerial. 
 
 438. The Receiver is that part of the apparatus of a 
 Wireless Telegraph Station which converts the oscillating 
 currents produced by electric waves in the aerial into 
 visible or audible signs. 
 
 439. The frequency of the oscillating current produced 
 in the aerial is the same as the frequency of the waves 
 which produce it. 
 
 440. By means of an aerial connected to a receiver,
 
 WIRELESS TELEGRAPHY 119 
 
 therefore, we can convert the electric waves which are 
 being radiated from a transmitting station into visible 
 or audible signs, thus enabling us to " read " the message 
 which is being transmitted. 
 
 ESSENTIALS OF A RECEIVER 
 
 441. We have already explained that an aerial forms 
 an " open " oscillatory circuit and has a natural fre- 
 quency of its own. We have also shown that an oscil- 
 lating current will not flow easily in a circuit unless the 
 frequency of that circuit is the same as that of the 
 oscillating current that is to say, in this case the aerial 
 circuit must be in tune with the wave which is to be 
 received. 
 
 The first essential of a receiver, therefore, is a variable 
 inductance and a variable condenser, which can be 
 connected in series with the 
 aerial by means of which 
 the latter can be tuned to the 
 desired wave-length. 
 
 442. Fig. 71 illustrates 
 these connections, where A 
 and E are the aerial and 
 earth terminals of the receiver, 
 
 I is the inductance more or , 
 less of which can be included 
 in the aerial circuit by means of the switch Sj and C 
 the variable condenser across which is fitted a short- 
 circuiting switch S 2 . 
 
 443. The inductance I is called the " Aerial Tuning 
 Inductance," and the condenser C the " Aerial Tuning 
 Condenser." 
 
 444. We know that by placing a condenser in series
 
 120 WIRELESS TELEGRAPHY 
 
 with the aerial we reduce the wave-length of the aerial, 
 and by placing an inductance in series with the aerial 
 we increase the wave-length of the aerial. 
 
 445. If, therefore, the wave-length which it is required 
 to receive is shorter than the natural or " fundamental " 
 wave-length of the aerial we must cut out all the in- 
 ductance in the circuit by means of the switch S 1} and we 
 must reduce the value of the adjustable condenser C 
 until the correct wave-length is obtained. The switch 
 S 2 will in this case be open, as shown in the diagram 
 {Fig. 71). 
 
 446. If, on the other hand, the wave-length which it 
 is desired to receive is longer than the fundamental wave- 
 length of the aerial, in order to bring the wave-length 
 of the aerial into tune we must first short-circuit the 
 condenser C by means of the switch S 2 , thus leaving 
 no capacity in series with the aerial, and we must increase 
 the inductance in the circuit by means of the switch S i 
 until the correct wave-length is obtained. 
 
 METHODS or DETECTING THE OSCILLATING 
 CURRENTS 
 
 447. The next essential of the receiver is some device 
 whereby the presence of the oscillating currents can be 
 
 ' detected. 
 
 448. In paragraph 432 we showed how this could be 
 done, in the case of a wavemeter, by placing across the 
 condenser of the oscillatory circuit a pair of telephones 
 in series with a crystal. The telephones in series with 
 a crystal constitute a detector. This method can be 
 adopted in the receiver by placing the detector across 
 the aerial tuning condenser, but it is not an efficient 
 method for the following reason.
 
 , WIRELESS TELEGRAPHY 121 
 
 449. The aerial tuning condenser forms only a part 
 of the capacity of the whole aerial circuit, so that 
 although the detector may be extremely sensitive, jt is 
 not being used to the best advantage. 
 
 450. Another method is to apply the detector across 
 tjie aerial tuning inductance, but this method has*also 
 the same disadvantage viz. that we are only applying 
 the detector to a portion of the whole inductance of 
 the aerial circuit. 
 
 451. If, however, we are receiving a wave very much 
 longer than the natural wave-length of the aerial, in 
 order to tune up the latter we naturally have to use a 
 large amount of inductance, and if this inductance forms 
 (as it may easily do) the greater part of the inductance 
 of the whole aerial circuit, 
 
 we may quite efficiently 
 apply the detector across 
 the inductance. 
 
 This makes one of the 
 simplest and cheapest 
 forms of wireless telegraph 
 receivers, and is shown 
 diagrammatically in Fig. 
 72, where A is the aerial, I FIO. 72. 
 
 the variable tuning induct- 
 ance, E the earth, D the crystal, and T the telephones. 
 
 452. Most of the amateur stations, more especially 
 those in towns, have very small aerials for obvious 
 reasons, and as they are chiefly used for " picking up " 
 signals from stations using long wave-lengths, this 
 form of receiver is particularly appropriate. With 
 such short aerials even the waves transmitted from ship 
 stations are sufficiently long to necessitate the use of
 
 122 
 
 WIRELESS TELEGRAPHY 
 
 a comparatively large inductance in series with the 
 aerial, so that the receiver may also be used fairly 
 efficiently for receiving signals from ships. 
 
 THE POTENTIOMETER 
 
 453. Some crystals, for example carborundum, 
 become more sensitive to minute currents when a slight 
 initial voltage is applied across them. This voltage must 
 
 Q 
 
 I 
 
 be regulated exactly to suit the particular crystal which is 
 being used, and this regulation is accomplished by means 
 of a potentiometer. 
 
 454. A potentiometer, shown in Fig. 73, consists of 
 a resistance coil R, connected across a battery Q, and 
 provided with a sliding contact S, by means of which a 
 lead can be connected to any point along the resistance. 
 
 455. The resistance of the coil should be kept suffi- 
 ciently high, so that the current passing through it from 
 the battery is not sufficient to discharge the battery
 
 WIRELESS TELEGRAPHY 123 
 
 rapidly. Too high a resistance becomes impracticable, 
 as either the resistance wire with which the coil is wound 
 would have to be so fine that it would easily become 
 broken or cut, or the resistance coil would have to be 
 of such a length that it would not be convenient on 
 account of its size. In practice suitable resistance coils 
 can be wound having a resistance of about 200 ohms, 
 and this connected across a battery of 4 volts will only 
 allow about one-fiftieth of an ampere to pass through it 
 (vide paragraph 78), so tliat a battery consisting of three 
 small dry cells would be sufficient to maintain its voltage 
 for many weeks with continuous working. 
 
 456. On referring to diagram (Fig. 73) and assuming 
 that the voltage of the battery is 4 volts, we have a differ- 
 ence of potential between the two ends of the resistance 
 coil, A and B, of 4 volts ; therefore, if we connect a wire 
 W 2 to the end of the resistance coil A, and another wire 
 W x to -the sliding contact S, and move the latter to the 
 far end of the coil shown in dotted lines and marked S 1} 
 the voltage between the two wires will be 4 volts. If, 
 however, we slide the contact towards the end of the coil 
 marked A, the voltage between the two wires diminishes 
 until the voltage becomes zero, when the slider occupies 
 the position S 2 . It is obvious that the voltage across the 
 two wires will be in proportion to the distance the slid- 
 ing contact is from the point A, and that by moving the 
 slider to any point between the two extreme ends of the 
 resistance we can regulate the voltage between the two 
 wires to any intermediate value between and 4 volts. 
 
 457. With most carborundum crystals, the voltage 
 which should be applied across them to bring them to 
 their most sensitive state is somewhere between 1 and 2 
 volts, so that by applying this potentiometer to our
 
 124 
 
 WIRELESS TELEGRAPHY 
 
 crystal, we have a simple means of bringing the latter to 
 its most sensitive state. 
 
 METHOD OF APPLYING THE POTENTIOMETER TO THE 
 CRYSTAL 
 
 458. The method of applying the voltage obtained 
 from the potentiometer to the crystal is not as straight- 
 forward as it might at first appear to be. 
 
 459. The most obvious way of doing it is shown in 
 Fig. 74, where the two wires from the potentiometer are 
 connected one to either side of the crystal. But it will 
 be immediately seen that this entirely neutralises the 
 
 I 
 
 FIG. 74. 
 
 FIG. 75. 
 
 value of our crystal as a rectifier, for the oscillating 
 currents, instead of trying to pass through the crystal 
 to the telephones, will pass through the resistance of the 
 potentiometer to the telephones. 
 
 460. We must therefore devise some means of 
 applying the voltage to the crystal without making a 
 bye-pass for the oscillating currents induced in the 
 inductance coil. 
 
 This can be accomplished by connecting up the circuit, 
 as shown in Fig. 75, where the junction of the battery
 
 WIRELESS TELEGRAPHY 125 
 
 and the resistance coil is connected to the earth side 
 of the crystal. One side of the telephone is then 
 connected to the earth terminal, and the other side to 
 the sliding contact of the potentiometer. 
 
 461. Assuming that the common junction of the 
 battery, resistance coil, and crystal J is the negative 
 side of the battery, the sliding contact S is the positive, 
 and this positive E.M.F. is conducted to the other side 
 of the crystal through the telephones and through the 
 aerial inductance, as indicated by the arrows. 
 
 Thus it will be seen that we have accomplished what 
 we desired, i.e. to apply an adjustable voltage across the 
 crystal without forming any short cut for the oscillating 
 currents, which must therefore pass through the crystal 
 and there be rectified before they reach the telephones. 
 
 THE TWO-CIRCUIT RECEIVER 
 
 462. As already explained, the single circuit receiver 
 just described is quite efficient for stations that are 
 receiving comparatively long wave - lengths on short 
 aerials, but it would be insensitive for stations which 
 might be required to receive messages on wave-lengths 
 as short as, or shorter than, the fundamental wave-length 
 of the aerial. 
 
 463. If we can cause all the energy in our aerial circuit 
 to be transferred to a secondary circuit and apply our 
 detector across the whole of the inductance and capacity 
 of this secondary circuit, it is obvious that the size of 
 the aerial will not limit us as to the value of the wave- 
 length for which such a receiver can be efficiently used. 
 
 464. Such a receiver has two distinct oscillatory 
 circuits, both of which must be in tune with the wave- 
 length which it is desired to receive. These two circuits
 
 126 WIRELESS TELEGRAPHY 
 
 are called respectively the primary circuit and the 
 secondary circuit. 
 
 465. The primary circuit as in the case of the single 
 circuit receiver will consist of the aerial, an adjustable 
 inductance, and an adjustable condenser for tuning up 
 tjie circuit, and a primary coil, by means of which the 
 oscillations can be induced into the secondary circuit, 
 thus transferring the energy from the primary to the 
 secondary circuit (vide paragraph 346). 
 
 466. The secondary circuit will consist of an induct- 
 ance coil with a variable condenser connected across it, 
 by means of which the wave-length of this circuit can be 
 adjusted so as to be in tune with the primary circuit 
 and at the same time with^he wave-length which it is 
 desired to receive. 
 
 467. The inductance coil of this secondary circuit 
 must be so placed relatively to the primary Coil that the 
 oscillating currents occurring in the latter will induce 
 similar oscillations in the former ; that is to say, the 
 axes of the two coils must be in line with one another, 
 and the two coils must be . sufficiently close together 
 (vide paragraph 407). 
 
 468. These circuits are shown diagrammatically in 
 Fig. 76, where the primary oscillatory circuit is formed 
 by A the aerial, I the aerial tuning inductance, C the 
 aerial tuning condenser, P the primary coil, and E 
 the earth. 
 
 The secondary oscillatory circuit is formed by S 
 the secondary coil and B the secondary tuning con- 
 denser ; the common axis of the primary coil and the 
 secondary coil being denoted by the dotted line XY. 
 
 469. The method of applying the potentiometer to 
 the crystal in this case is shown in Fig. 77.
 
 WIRELESS TELEGRAPHY 
 
 127 
 
 470. By applying our detector, as -shown in the 
 diagram, across the secondary inductance coil, we are 
 Bpplying it in the most efficient manner possible, 
 since, no matter to what wave-length the secondary 
 
 FIG. 77. 
 
 circuit is adjusted, the detector will be applied to the 
 whole of the inductance in that circuit. 
 
 PROPORTION OF INDUCTANCE AND CAPACITY IK 
 SECONDARY OSCILLATORY CIRCUIT 
 
 471. Another point which we have not yet touched 
 upon, affecting the efficiency of the crystal detector, is 
 the proportion of the inductance to the capacity of the 
 oscillatory circuit to which the detector is applied to 
 obtain maximum efficiency. In addition to the fact that 
 it is necessary that the detector be connected across the 
 whole of the inductance, it is found in practice that the 
 greater the inductance of that circuit compared with its 
 capacity, the more efficient will the crystal, as a detector, 
 become. 
 
 The reason for this is explained later in paragraph 488, 
 but for the time being we must take it as a fact and 
 develop our receiver accordingly.
 
 128 WIRELESS TELEGRAPHY 
 
 472. At first sight it would appear that there is no 
 limit to the amount by which we can increase the in- 
 ductance of the secondary coil S, so long as we reduce 
 the capacity of the condenser B in proportion. This, 
 howeveij-is not the case, for by increasing the number of 
 turns on the inductance S we not only increase the 
 inductance of the circuit but also the capacity. 
 
 473. Up to the present we have regarded coils of wire 
 as having only the quality of inductance. As a matter 
 of fact, however, every coil of wire has self-capacity, and 
 for this reason it is found that every coil of wire, even 
 without a condenser connected across it, forms an open 
 oseillatory circuit, and has all the essentials of an 
 oscillatory circuit that is to say, the two qualities of 
 inductance and capacity. This self-capacity then limits 
 the amount of inductance we can use, for in increasing 
 the inductance we cannot avoid increasing also the 
 capacity of the circuit. 
 
 474. The most efficient coil that we can design for the 
 secondary circuit of the crystal receiver is therefore 
 one whose wave-length by itself will be the required value 
 without the addition of any extra capacity. Our 
 adjustable condenser, however, is necessary in order to 
 enable us to increase the wave-length of the secondary 
 circuit, for a receiver only capable of receiving one length 
 of wave would be very inconvenient ; but again we are 
 limited to the extent to which we can vary it by the fact 
 that as we increase the capacity across the inductance, 
 so do we decrease the efficiency of our detector when 
 applied to that circuit (vide paragraph 471). 
 
 475. In practice it is found that, without materially 
 affecting the efficiency of the detector, we can connect 
 a sufficiently large capacity across the inductance to
 
 WIRELESS TELEGRAPHY 129 
 
 increase its wave-length to about three times its original 
 wave-length. If we go beyond this point the reduction 
 in the efficiency of the detector becomes noticeable. 
 
 476. We may say, then, that with a two - circuit 
 receiver in which a crystal is used as a detector, the 
 maximum efficiency is obtained when the capacity across 
 the secondary condenser is reduced to zero. Further, 
 we may say that the maximum wave-length to which it 
 can be efficiently tuned will be about three times the 
 value of its minimum wave-length. Thus, if the shortest 
 wave-length which a station is required to receive is 
 300 metres, the receiver would be designed so that the 
 minimum wave-length to which it can be adjusted will 
 be, 300 metres, and its maximum wave-length will 
 then be about 900 metres. 
 
 477. Where a longer range of wave-length than this 
 is required, special arrangements have to be made by 
 which the secondary inductance coil can be changed ; 
 thus if a receiver is required to receive wave-lengths of 
 any value between 300 and 1500 metres, it will probably 
 have two secondary inductance coils, one of which will 
 allow the receiver to be tuned up from 300 to 900 metres, 
 and the other from, say, 600 to 1800 metres. 
 
 CHARACTERISTIC CURVE OF CRYSTAL 
 
 478. Up to the present we have considered the action 
 of the crystal to be purely one of rectifying the oscil- 
 latory currents induced across it into uni-directional 
 currents. 
 
 479. The crystal can be better considered as a con^ 
 ductor offering a certain resistance to current passing 
 through it in one direction, and offering a very much 
 
 K
 
 130 
 
 WIRELESS TELEGRAPHY 
 
 larger resistance to current trying to pass through it in 
 the other direction. 
 
 480. Its value as a sensitive detector, however, de- 
 pends upon another property. Even in the direction of 
 conductivity, the crystal does not act in the same way as 
 an ordinary conductor. 
 
 481. With an ordinary conductor the current passing 
 through it increases directly as the voltage applied 
 
 across it increases (vide paragraph 78). Thus, if we draw 
 a " curve " illustrating the increase in the current 
 which would flow through an ordinary conductor as the 
 voltage across it is increased, this curve would take the 
 form of a straight line, as shown in Fig. 78. 
 
 482. If a curve be drawn illustrating the increase in 
 the current passing through a crystal as the voltage across 
 it is increased, it will take the form shown in Fig. 79. 
 
 483. In this case it will be noticed that when the 
 voltage is increased beyond the point A, the current
 
 WIRELESS TELEGRAPHY 
 
 131 
 
 passing through it rises very much more rapidly than 
 before in proportion to the increase in voltage across it. 
 
 484. This is due to the fact that the effective resistance 
 of the crystal does not remain constant, but starts to 
 decrease when the voltage across it is increased above a 
 certain value. 
 
 485. By referring to this curve it will be seen that 
 
 between the points and A a certain increase 
 in the voltage across the crystal produces a very small 
 increase in the current passing through .it, and thus 
 through the telephones, whereas beyond the point A 
 the same increase in the voltage across the crystal pro- 
 duces a larger increase in the current passing through it. 
 486. To produce a sound in the telephone it is neces- 
 sary that the current passing through it be increased, 
 and the strength or loudness of that sound will depend 
 upon the amount by which the current is increased.
 
 132 WIRELESS TELEGRAPHY 
 
 487. It is obvious, therefore, that the voltage pro- 
 duced across the secondary coil of the jigger by the 
 oscillating currents will cause a greater increase in the 
 current passing through the crystal and telephones if it 
 be applied after the point A is reached. It is for this 
 reason that a potentiometer is necessary in order to 
 bring the initial voltage across the crystal up to the 
 point A. 
 
 488. In paragraph 471 we mentioned the fact that 
 the efficiency of the receiver depended on keeping the 
 capacity of the secondary circuit as small as possible. 
 The reason for this is that a given amount of energy will 
 produce a greater increase in voltage across a small 
 condenser than across a large condenser (vide paragraph 
 272). 
 
 Now in our receiver the energy in the circuit is 
 a fixed quantity, depending upon the strength of the 
 oscillations produced in the aerial ; this energy is in 
 turn transferred to the secondary circuit. 
 
 It is therefore obvious that the only way to increase 
 the voltage across the condenser is to reduce the value 
 of that condenser. 
 
 489. In so reducing it we reduce also the wave-length 
 of that circuit, and as it is necessary to keep this in tune 
 with the wave-length which is being received, we must 
 counterbalance the effect of reducing the capacity by 
 increasing the inductance of the circuit. The extent to 
 which we can do this, as already explained, is limited by 
 the fact that every coil of wire has self-capacity, and as 
 we increase the coil to get a greater inductance, so, at 
 the same time, we increase its capacity. 
 
 490. The rate at which we increase this self-capacity, 
 however, can be controlled, to a large extent, by the
 
 WIRELESS TELEGRAPHY 133 
 
 design of the coil that is to say, by its diameter, its 
 length, and the size of the wire with which it is wound. 
 
 491. A question which will probably arise in the 
 minds of those studying this explanation will be that, 
 in describing the receiver, we said that the crystal was 
 placed across the inductance, whereas in explaining the 
 reason for keeping this inductance high, in proportion 
 to the capacity, we take the point of view that we wished 
 to increase the voltage across the condenser. This is 
 only because it is easier to understand how the voltage 
 must necessarily increase across the condenser if the 
 value of that condenser is decreased, and, since in an 
 oscillatory circuit the condenser is connected across the 
 inductance, it follows that the voltage across the in- 
 ductance is likewise increased. 
 
 THE TELEPHONE RECEIVER 
 
 492. So far we have not touched upon the construction 
 of the telephone receivers. 
 
 The function of the telephone receivers (usually 
 called " telephones " for short) is to convert electric 
 currents into an audible sound. 
 
 It is of course of as much importance for this part 
 of the apparatus to be efficient as any other, and in order 
 to be efficient it must be made suitable for the circuit 
 to which it is applied. 
 
 493. A telephone receiver consists essentially of an 
 electro-magnet and a diaphragm. 
 
 The diaphragm is a circular piece of very thin sheet 
 iron, supported all round the edge by the outer case, or 
 shell, of the " ear-piece," as close to the face of the 
 magnets as possible without actually touching.
 
 134 
 
 WIRELESS TELEGRAPHY 
 
 494. Fig. 80 shows diagrammatically a section of a 
 telephone ear-piece where A is the iron core of the 
 electro-magnet, B the coils of the electro-magnet, C the 
 case, or shell, and D the diaphragm. 
 
 495. Unlike an ordinary electro -magnet, the iron 
 core of the telephone receiver is to a certain extent 
 permanently magnetised. 
 
 496. It is evident, then, that the diaphragm will 
 
 normally be strained slightly towards the magnet, as 
 shown by the full line D, in Fig. 80. 
 
 As already mentioned, it is supported by the shell of 
 the ear-piece all round its edge, but, being thin and 
 springy, it will bulge in the middle towards the magnet. 
 
 497. The action of the telephone receiver is as follows : 
 If a current is sent through the coils in such a direction 
 that the lines of force set up by it assist those of the 
 permanent magnet, the strength of the magnet will be 
 increased, and the diaphragm will' be attracted still 
 closer to the magnet, thus taking the position shown 
 by the dotted line D x . 
 
 498. If, on the other hand, a current is sent through
 
 WIRELESS TELEGRAPHY 135 
 
 the coil in the opposite direction, thus setting up lines 
 of force opposing those of the permanent magnet, the 
 strength of the magnet will be decreased and the dia- 
 phragm will be allowed to spring farther away from the 
 pole and take up the position shown by the dotted line 
 D 2 , owing to the fact that it has already been displaced 
 out of its normal position due to the normal pull of the 
 permanent magnet. 
 
 499. Owing to the form of the diaphragm, it acts in 
 just the same way as the head of a drum, and will 
 produce a big sound with a comparatively small dis- 
 placement of its centre. 
 
 Just as the noise produced by a drum will depend 
 upon how hard it is hit by the drumstick, so will the 
 noise produced by the diaphragm depend upon the 
 amount of increase or decrease in the magnetisation of 
 the magnet. 
 
 HIGH RESISTANCE TELEPHONES 
 
 500. It is obvious that the increase or decrease in 
 the magnetism of the magnet will depend upon the 
 magnetisation force or " magneto-motive force " which 
 is applied to it. 
 
 The magneto-motive force depends upon two things : 
 (1) the number of turns of wire which are encircling the 
 magnet, and (2) the amount of current passing through 
 them (vide paragraph 96). 
 
 501. For a given size of magnet we have only a 
 definite space into which to get our turns of wire, so that 
 the only way of increasing the number of turns we can 
 wind on the magnet is to decrease the size of the wire. 
 The thinner the wire the greater the number of turns 
 which we can get into the space at our disposal.
 
 136 WIRELESS TELEGRAPHY 
 
 502. Unfortunately, however, as we reduce the size 
 of the wire, so do we increase the resistance per turn of 
 that wire, and therefore decrease the amount of current 
 which would pass through it for a given voltage. There- 
 fore, unless the current at our disposal is already limited 
 by some external resistances, we shall not gain anything 
 by increasing the number of turns if at the same time we 
 increase the resistance of the coil in proportion. 
 
 503. If, however, the telephone is in a circuit in which 
 there is already a high resistance, then the increase in 
 the resistance of the coil will not have so great an effect 
 on the total resistance of the circuit, and therefore on 
 the current which is passing through that circuit. 
 
 504. For an example, let us suppose that a coil wound 
 with 10 turns of a certain size wire will have a resistance 
 of 1 ohm, and let us suppose that the external resistance 
 of the circuit in which the coil is connected is 99 ohms. 
 The total resistance of the circuit is then 100 ohms. If 
 our voltage across this circuit is 1 volt, then it follows 
 that our current through this resistance will be a one- 
 hundredth part of an ampere, and consequently 
 
 Magneto-motive force = y^'x 10 turns = ^. 
 
 505. Now let us wind the same coil with wire y^th 
 the former cross-sectional area. It follows that we shall 
 get 10 times the number of turns that is to say, wetahall 
 get 100 turns of wire on to the coil, but our resistance 
 per turn will be increased ten times. The resistance per 
 turn in the first coil was y\jth of an ohm, so that our 
 resistance per turn will now be 1 ohm ; therefore the 
 resistance of the coil will be 100 ohms. 
 
 506. Adding this to our external resistance we get 
 a total resistance in the circuit of 199 ohms. Now
 
 WIRELESS TELEGRAPHY 137 
 
 for the same voltage, i.e. 1 volt across this circuit, 
 we shall get xd9 tn of an ampere, and therefore in 
 this case 
 
 Magneto-motive force = T -^ x 100 = approx. J. 
 
 507. It is obvious, therefore, that in this case we 
 have increased our magneto-motive force nearly five 
 times by winding the coils with a finer-sized wire. 
 
 508. On examining the diagrams of connections of 
 our wireless telegraph receiver, it will be seen that any 
 current passing through the telephones will have to pass 
 through the crystal. 
 
 509. The resistance of our crystal at its most sensitive 
 point is of the order of 10,000 ohms. It will there- 
 fore be obviously inefficient to wind the telephone 
 receiver with such a sized wire that its resistance is 
 only, say, 200 ohms, if a finer wire is available. 
 
 510. In practice special telephones are made suitable 
 for circuits with such external resistances. These 
 telephones are wound with the very finest wire which it 
 is possible to manufacture, in order to get the greatest 
 possible number oi turns on to the limited space of the 
 bobbins. 
 
 511. Such telephones are called High Resistance 
 Telephones, and have a resistance of approximately 
 3500 ohms per ear-piece, and two ear-pieces can be used, 
 connected in series, thus making a total resistance of a 
 pair of telephones about 7000 ohms. 
 
 512. The point which must be clearly understood is 
 that the object of using high resistance telephones is not 
 because they have a high resistance, but because they 
 are wound with a very much larger number of turns than 
 the low icsistance telephones, and therefore, owing to
 
 138 WIRELESS TELEGRAPHY 
 
 the high external resistance of the circuit, the magneto 
 motive force is increased to a greater extent than it is 
 decreased by the increase of resistance of' that circuit. 
 
 RECTIFYING PROPERTIES OF CARBORUNDUM 
 
 513. In paragraph 485 we explained why it is necessary 
 to adjust the initial voltage across the carborundum 
 crystal to a certain value in order that a given increase 
 in voltage will cause the greatest possible increase in 
 the current passing through it. We have not, however, 
 explained why it is necessary to adjust the initial volt- 
 age across the crystal to the exact point where the 
 effective resistance of the crystal commences to decrease 
 rapidly. 
 
 514. Referring again to the characteristic curve of a 
 carborundum crystal shown in Fig. 79, although it is 
 obvious that the crystal will be more sensitive when the 
 point A is reached, it is not quite so obvious why it is 
 necessary to adjust the initial voltage across the crystal 
 exactly to the point A, and not to any point beyond it, 
 such as the point B shown in Fig. 81. 
 
 515. As can be seen from this curve, a given increase 
 in voltage will cause practically the same increase in 
 current passing through the crystal whether this increase 
 be applied at the point B or the point A, but we must 
 remember that the extra voltage provided by the 
 oscillatory current in the secondary of the jigger is an 
 alternating current voltage, that is to say, a voltage 
 varying from a positive value at one instant to a negative 
 value at the next instant. 
 
 516. Since the initial voltage applied across the 
 crystal is a direct current voltage obtained from the
 
 WIRELESS TELEGRAPHY 139 
 
 potentiometer, it follows that the alternating current 
 voltage will at one instant be assisting the direct current 
 voltage, and at the next instant opposing it. 
 
 To facilitate explanation, let us put these voltages 
 into- figures. 
 
 517. Let us suppose that the initial voltage across 
 the crystal to bring it up to the ppint A is 2 volts, and 
 the voltage required to bring it up to the point B is 
 2J volts, these voltages being positive volts. 
 
 518. Let us also suppose that the value of the alter- 
 nating voltage provided by the oscillating current 
 varies from minus ^ a volt to plus \ a volt ; it is obvious, 
 then, that during the time that the oscillations are 
 being received the resulting voltage across the crystal, 
 if the initial voltage be adjusted to the point A, will vary\ 
 from 1| volts to 2| volts. Similarly, if the initial voltage 
 across the crystal be adjusted to the point B, the resulting 
 voltage will vary from If volts to 2f volts. 
 
 519. Now let us draw two separate curves, shown in 
 Figs. 82 and 83. showing the result of this variation in 
 voltage on the current passing through our telephones, 
 taking our figures from the curve shown in Fig. 81. 
 
 The curve in Fig. 82 shows the resulting current in 
 the telephones when the initial voltage of the crystal is 
 adjusted to point A. 
 
 520. At this point the value of the current passing 
 through the crystal and telephones before any oscillations 
 are produced in the secondary circuit is 1, therefore we 
 may draw a heavy line DD, representing the normal 
 value of the current. 
 
 521. When the negative part of the first oscillation is 
 applied across the crystal, the result, as already explained, 
 is to reduce the voltage to 1^ volts, thus the current
 
 140 
 
 WIRELESS TELEGRAPHY 
 
 passing through the telephones will be reduced, but, as 
 will be seen by referring to Fig. 81, owing to its being on 
 
 the flat part of the curve, the reduction in the amount 
 
 of current passing through the telephones is extremely 
 
 Via. 82. 
 
 small ; by reading from the curve we find it is reduced 
 to the value of f . 
 
 Therefore the curve representing the actual current
 
 WIRELESS TELEGRAPHY 
 
 ui 
 
 passing through the crystal and telephones when the 
 negative part of the first oscillation is applied, will dip 
 just below the line DD. 
 
 522. The next half of the oscillation is positive, and 
 therefore has the result of increasing the voltage across 
 the crystal to 2| volts. 
 
 By referring again to Fig. 81, it will be seen that the 
 effect on the value of the current is to increase it to 4. 
 
 We may therefore continue our current curve in 
 
 . 83. 
 
 Fig. 82, which will now show the current rising to the 
 value 4 above the normal line DD. 
 
 523. A similar cycle will take place for each oscillation, 
 with the result that we get a series of high peaks above 
 the normal current line, and a series of very shallow 
 dips below this line. 
 
 524. These oscillations are taking place at the rate 
 of perhaps millions per second, according to the length 
 of wave which is being received. 
 
 If the wave-length received is 100 feet, the number of 
 oscillations per second will be 10 millions per second 
 (vide paragraphs 167 and 174).
 
 142 WIRELESS TELEGRAPHY 
 
 525. These variations are infinitely too rapid for the 
 diaphragm of the telephone to follow, and it will therefore 
 be deflected to an extent corresponding to the average 
 value of the current passing through its coils. 
 
 526. Referring to Fig. 82 the average current passing 
 through the telephones when the oscillating voltage 
 is applied across the crystal is shown by the dotted 
 line AA, drawn approximately half-way between the 
 highest and lowest point on the curve. This value is. 
 somewhere about 2J, and the sound produced in the 
 telephones will be proportional to the difference between 
 the normal current passing through the telephones and 
 the increased current due to the oscillating voltage 
 applied, i.e. the difference between 1 and 2J. 
 
 527. Now let us see what takes place if we adjust the 
 crystal to the point B ; obtaining our values of current 
 as before from the curve in Fig. ,81, we find that the 
 result of the oscillating voltage is to vary the value of 
 the current from ^ to 5|, and we may therefore draw a 
 new curve as shown in Fig. 83, representing this variation 
 in the value of the current. 
 
 528. Again, these variations in current are too rapid 
 for the diaphragm of the telephone to follow, so that it 
 will again be deflected to an extent corresponding to 
 the average value of this current. 
 
 529. The average value of this current will be about 
 3, so that we may draw a dotted line BB, representing 
 the average value of the current passing through the 
 telephones when the oscillating voltage is applied across 
 the crystal. 
 
 530. But we have already increased the nonnaj Value 
 of the current passing through the telephones to the 
 value of 2$, as shown by the line DD, Fig. 83, this being
 
 WIRELESS TELEGRAPHY 143 
 
 the current which will pass through the crystal and 
 telephones when the initial voltage of 2J volts is applied 
 to bring it up to the point B, Fig. 81. 
 
 531. The strength of the sound produced in the 
 telephones will be proportional, not to the total current 
 passing through the telephones, but to the difference be- 
 tween the current passing through them when no oscilla- 
 tions are being received and the average current passing 
 through them when the oscillations are being received. 
 
 532. When the crystal was adjusted to the point A, 
 this difference in current was the difference between 
 1 and 2f , so that the strength of the sound was propor- 
 tional to 1| ; but when the crystal was adjusted to the 
 point B, the strength of the sound was proportional tc 
 the difference between 2| and 3, which is only . 
 
 RELATION BETWEEN THE SPARK FREQUENCY OF THE- 
 TRANSMITTER AND SOUND PRODUCED IN THE 
 TELEPHONES OF RECEIVER 
 
 533. In paragraph 433, describing the wavemeter 
 (which is in reality a simple form of tuned receiver), 
 we said that the current produced in the telephone 
 receiver would be an intermittent current, and the 
 number of interruptions per second would be the same 
 as the number of sparks per second in the oscillatory 
 circuit which is being measured. 
 
 The explanation of this is easy to follow if the fore- 
 going paragraphs are thoroughly understood. 
 
 It is obvious that the average current passing through 
 the telephone from any group of oscillations may be 
 regarded as a direct current flowing so long as the 
 oscillations arc maintained.
 
 144 WIRELESS TELEGRAPHY 
 
 534. If, then, the transmitting station were sending 
 out a stream of continuous waves (vide paragraph 213), 
 so long as the manipulating key were kept depressed 
 we should get a continuous current flowing through the 
 receiver, without interruption. 
 
 As a matter of fact, however, when we depress the 
 manipulating key we get a succession of short groups of 
 damped waves, one group each time the condenser is 
 charged by the induction coil and discharged through 
 the spark gap. 
 
 535. The uni-directional current, therefore, produced 
 in the telephone of the receiver will only be maintained 
 for the time that the group of waves lasts, with the result 
 that the diaphragm of the telephone is deflected for an 
 instant only, and returns to its normal position until 
 another group of waves is received ; thus a single click 
 will be produced by each group of oscillations. 
 
 536. As each spark in the transmitter produces a 
 group of waves, so does each group of waves in the 
 .-eceiver produce a click in the telephones. Thus the 
 sound produced in the telephones, or, in other words, the 
 frequency of the clicks in the telephones, will correspond 
 with the spark frequency of the transmitter. 
 
 To TUNE A RECEIVER 
 
 537. We will suppose that our receiver is of the 
 two-circuit type that is to say, that it has a primary 
 circuit and a secondary circuit, both of which 
 must be in tune with the wave-length it is desired to 
 receive. The only means we have of telling whether the 
 receiver is in tune is by the strength of the signals in the 
 telephones. If either circuit of the receiver is out of tune, 
 the signals are weakened, so that provided we have a
 
 WIRELESS TELEGRAPHY 145 
 
 variable inductance or condenser in each circuit, and 
 provided we can hear at least weak signals in the tele- 
 phones, it is a simple matter to tune up the receiver 
 by listening to the strength of the signals and adjusting 
 first one circuit and then the other circuit, until the 
 sound is at its loudest. 
 
 538. If, however, we are so much out of tune to begin 
 with that the signals are inaudible, the difficulty of 
 tuning up is increased enormously. 
 
 539. In the case of a single - circuit receiver the 
 difficulty is not so great, for we have only one circuit to 
 adjust, and therefore we can vary it slowly from its 
 maximum wave-length to its minimum wave-length, 
 and consequently we are bound to pass the point where 
 the receiver is in tune with, and therefore will respond 
 to, the signals. 
 
 540. In the case of a two-circuit receiver, however, 
 if signals are inaudible to begin with, we have no means 
 of telling which circuit is out of tune or when the two 
 circuits are in tune with each other. 
 
 541. If we know the wave-length of the signals we 
 wish to receive, and we have an instrument close to our 
 receiver which can be made to emit a similar wave-length, 
 the process of tuning up becomes quite simple. 
 
 Such an instrument is called a tuning buzzer. 
 
 542. Since our detector and telephones are actuated 
 by the secondary circuit of the receiver, we should first 
 cause the tuning buzzer to induce waves into the second- 
 ary circuit only, and we should then adjust this circuit 
 until the buzzer signals in the telephones are at their 
 loudest. 
 
 Having accomplished this, we should next move the 
 tuning buzzer to a point remote from the secondary 
 
 L
 
 146 WIRELESS TELEGRAPHY 
 
 circuit, but close to some part of the primary or aerial 
 circuit, so that no oscillations can be induced from it 
 directly into the secondary circuit, but only through 
 the primary circuit. 
 
 543. Now if the primary circuit is very much out of 
 tune with the wave-length emitted by the tuning buzaer, 
 it will not respond to the oscillations, and therefore no 
 oscillations will be induced in the secondary circuit, 
 but if we vary the wave-length of the primary circuit, 
 we shall reach a point when it is in tune with the wave 
 emitted by the tuning buzzer. Oscillations will then 
 be induced in the primary circuit, which will in turn 
 induce oscillations in the secondary circuit, as the 
 secondary circuit has already been tuned to the same 
 wave-length. Thus when by varying the adjustment 
 of the primary circuit we reach a point when the signals 
 in the telephones are again at their loudest, we know that 
 we have reached the point when the primary circuit is 
 in tune with the tuning buzzer, and therefore both 
 circuits are in tune with the wave-length emitted by 
 the tuning buzzer^ 
 
 THE TUNING BUZZER 
 
 544. The essentials of a tuning buzzer are, therefore, 
 
 (1) that it can be caused to emit feeble oscillations, and 
 
 (2) that the frequency of these oscillations can be adjusted 
 to any predetermined value. 
 
 545. To accomplish these desiderata, the tuning 
 buzzer has two circuits : firstly, an oscillatory circuit, 
 consisting of an inductance coil with an adjustable 
 condenser, and, secondly, a generating circuit, by which 
 the oscillatory circuit is excited.
 
 WIRELESS TELEGRAPHY 
 
 147 
 
 546. The construction of the oscillatory 'circuit of a 
 tuning buzzer is identical with that of the wavemeter, 
 which was described in paragraph 417. It consists of 
 a fixed inductance coil connected in series with an 
 adjustable condenser, the latter being provided with 
 a scale and pointer by means of which the value 
 of the wave-length to which that circuit is adjusted is 
 indicated. 
 
 547. There are several ways in which this circuit can 
 be excited. We can, of course, charge up the con- 
 denser by means of an induction c,oil and discharge 
 it through a spark gap in the oscillatory circuit, as 
 
 FIG. 84. 
 
 described in paragraph 319; but this would be an 
 expensive method and, moreover, would produce very 
 much stronger signals than are necessary. 
 
 548. The method most commonly used is shown 
 diagrarnmatically in its simplest form in Fig. 84, where 
 L is the inductance coil, C the condenser forming the 
 oscillatory circuit, and where B is a battery connected 
 across the inductance coil through the contact S. If the 
 contact S is depressed, thus completing the circuit from 
 the battery through the inductance coil, a continuous 
 current will flow through this coil. If this circuit is 
 broken by releasing the contact the current will be 
 instantaneously interrupted.
 
 148 WIRELESS TELEGRAPHY 
 
 549. As already described in paragraph 62, the 
 property of inductance is similar to the mechanical 
 .property of momentum, and therefore, when the current 
 is suddenly interrupted, the energy due to its momentum 
 is liberated, and is expended in the oscillatory circuit 
 of which this inductance forms a part. The result is 
 practically to give this circuit a kick, causing it to 
 oscillate to its own natural frequency ; thus every time 
 the battery circuit is broken we produce a group of 
 oscillations in the oscillatory circuit corresponding to the 
 wave-length to which that circuit is adjusted. 
 
 550. If a battery of only two or three volts be used 
 and the inductance included in the battery circuit be 
 a reasonable amount, the oscillations set up will be 
 sufficiently strong to affect our receiver circuit. 
 
 551. This, however, is not quite sufficient to enable 
 us conveniently to tune up the receiver, for each group 
 of waves will only give a single click in the telephones. 
 Some automatic arrangement must be used to make 
 and break the circuit rapidly in order to produce a con- 
 tinuous buzz or note in the telephones, it being very 
 much easier to distinguish when a buzz or note reaches 
 its maximum strength than if only a number of single 
 clicks were audible. 
 
 552. One method by which this rapid making and 
 breaking of the circuit can be accomplished is shown 
 diagrammatically in Fig. 85, where the battery circuit B, 
 through the inductance L, is made through a pair of 
 contacts S, S, one of which is mechanically connected to 
 the armature A of an ordinary electric buzzer D, so that 
 when this armature vibrates it causes the extra pair of 
 contacts S, S alternately to make and break the battery 
 Circuit through the inductance L.
 
 WIRELESS TELEGRAPHY 
 
 149 
 
 553. In this case two batteries are required, one for 
 working the buzzer and the other for the oscillatory 
 circuit. 
 
 554. There is no reason, however, why the ordinary 
 single contact buzzer cannot be used for exciting an 
 oscillatory circuit, for by connecting it in such a way 
 that the current passing through the coils of the buzzer 
 is made to pass also through the inductance of the 
 oscillatory circuit, as shown in Fig. 86, we have practi- 
 cally the same conditions as before. Energy will be 
 stored up in the inductance L, while the current is passing 
 
 i 
 
 FIG. 85. 
 
 through it, and will be liberated as soon as it is inter- 
 rupted at the contacts S, the energy thus liberated 
 giving the oscillatory circuit a kick, and thus causing it 
 to oscillate to its own natural frequency. 
 
 555. In this case, however, when the contacts S are 
 broken, we not only liberate the energy stored up in the 
 inductance L, but we also liberate the energy which is 
 stored up in the inductive coils of the buzzer itself. 
 
 The inductance of these coils is m#ny times greater 
 than the inductance in the oscillatory circuit, and 
 therefore a very much larger amount of energy will bo 
 liberated at this point when the circuit is interrupted. 
 
 556. If no path is provided in which this energy can
 
 150 
 
 WIRELESS TELEGRAPHY 
 
 dissipate itself, it will form a small arc at the contacts S, 
 and dissipate itself gradually in this manner. 
 
 557. Unfortunately,, this arc will also form a path 
 for the energy stored up in the inductance L, with the 
 result that the energy will be dissipated in the same 
 way without charging up the condenser ; so that under 
 these conditions the oscillatory circuit would not be 
 excited. 
 
 558. If, however, we connect a non-inductive resist- 
 ance, as shown by R, Fig. 86, across the coils, of the 
 
 buzzer, the energy liberated from the buzzer coils will 
 expend itself in the circuit formed by the coils of the 
 buzzer and the resistance coil R, instead of forming an 
 arc at the contact S. But the energy liberated from 
 the inductance L will not have this circuit in which to 
 expend itself, for the circuit is interrupted at the contact 
 S, and will therefore have to expend itself in charging 
 up the condenser C. 
 
 559. The connections shown in the figure are those" 
 usually adopted in an ordinary tuning buzzer, and it is 
 obvious that the instrument can be used either to buzz
 
 WIRELESS TELEGRAPHY 151 
 
 a calibrated closed oscillatory circuit, and so make it 
 emit waves of any desired length for the purpose of 
 testing receivers, etc., or it can be used to buzz any other 
 oscillatory circuit in order that the wave-length of that 
 circuit may be measured by means of a wavemeter. 
 
 560. We have now explained briefly the principles 
 of the design and application of crystal receivers. There 
 are, however, many other forms of receiver, some of which 
 make use of phenomena entirely different from those 
 already explained. 
 
 The purpose of this book will be served by describing 
 only two other forms of receiver, namely, the Electrolytic 
 Detector and the Magnetic Detector. 
 
 THE ELECTROLYTIC DETECTOR 
 
 561. In the Electrolytic Detector, use is made of the 
 depolarising effect which oscillatory currents produce 
 on a polarised electrolytic cell. 
 
 As some readers may be unacquainted with simple 
 chemical theory, we think it advisable first of all to 
 give a short account of the composition of water. 
 
 COMPOSITION OF WATER 
 
 562. Water is described chemically by the formula 
 H 2 0, which means that it consists of two parts by 
 volume of the gas hydrogen (H), and one part of the gas 
 oxygen (0), in chemical combination with one another. 
 
 563. Most readers will have noticed that if a gas-jet 
 is burned in a small room for any length of time, a 
 considerable amount of water is condensed on the 
 wmdow panes. This is really due to the fact that the
 
 152 WIRELESS TELEGRAPHY 
 
 oxygen in the surrounding atmosphere combines chemic- 
 ally with the hydrogen in the coal gas, and forms 
 water. This chemical action produces energy in the 
 form of heat which is apparent from the very high 
 temperature of the gas flame. 
 
 On account of this high temperature, the water 
 which results is produced hi the form of steam, and is 
 therefore invisible until it has had time to condense on 
 to some body which will conduct the heat away from it. 
 
 564.' Hydrogen and oxygen will only combine when 
 brought to a high temperature, so that if they be simply 
 mixed together at a normal temperature no chemical 
 action will take place, but as soon as a spark or flame 
 is applied to the mixture the gases immediately sur- 
 rounding the spark are brought to a high temperature, 
 with the result that chemical action is started. The 
 heat produced by this action is more than sufficient to 
 keep the process of combustion going, so that even if 
 the spark or flame which started the action be removed, 
 the gases continue to combine until one or the other 
 of them is exhausted. 
 
 565. Hydrogen and oxygen, like every other sub- 
 stance, consist of a very large number of minute units 
 to which scientists have given the general name of 
 " molecules." Every molecule is still further sub- 
 divided into units, known as " atoms," each molecule 
 consisting of a definite number of atoms. The molecule 
 of oxygen, for instance, contains two atoms of oxygen ; 
 the molecule of hydrogen contains two atoms of hydrogen; 
 the molecule of water, however, contains three atoms, 
 namely, two of hydrogen and one of oxygen. 
 
 The minuteness of a molecule can be gauged from the 
 fact that the smallest particle of water which onn bee.
 
 WIRELESS TELEGRAPHY 
 
 153 
 
 seen under the most powerful microscope is made up 
 of many millions of molecules. 
 
 566. Figs. 87 and 88, which show a simple graphical 
 illustration of the molecular theory, will enable the 
 reader to imagine more easily the difference between a 
 mixture and a combination. In these illustrations we 
 have represented molecules by a number of irregular 
 outlines, and atoms by small spheres contained inside 
 the molecule. For the purpose of distinguishing them 
 we have represented the hydrogen atoms as white dots, 
 thus, o ; and the oxygen atoms as black dots, thus, 
 
 FIG. 87. 
 
 Fro. 88. 
 
 567. Fig. 87 represents a number of hydrogen and 
 oxygen molecules mixed together, and as no chemical 
 action is yet supposed to have taken place, the hydrogen 
 and oxygen molecules still retain their individuality. 
 If this mixture be ignited the result of the chemical 
 combination of the hydrogen and oxygen is shown in 
 Fig. 88, where the hydrogen and oxygen no longer exist 
 as separate molecules, but have combined in the pro- 
 portion of two molecules of hydrogen to one of oxygen, 
 and form entirely new molecules, namely, molecules 
 of water.
 
 154 WIRELESS TELEGRAPHY 
 
 This brief description will perhaps assist the ivader 
 to understand the action of an electrolytic cell. 
 
 ELECTROLYTIC CELL 
 
 568. An Electrolytic Cell is an apparatus for splitting 
 up the molecules of water, or other compounds, into 
 their original elements. It is found that when a current 
 of electricity is made to pass through a conducting 
 solution of water, the water is decomposed, that is to 
 say, the molecule of water is split up into the three 
 atoms, two of which are hydrogen and one oxygen. 
 
 569. An electrolytic cell, therefore, consists of a 
 vessel usually of glass, containing a solution of sulphuric 
 or nitric acid in water, into which are dipped two con- 
 ducting rods ca'lled the electrodes. These rods are 
 usually made of platinum or some other non-corrosive 
 metal. 
 
 570. The object of the acid is to make the water 
 conducting, pure water being practically a non-con- 
 ductor of electricity. 
 
 571. Fig. 89 shows such an electrolytic cell connected 
 to a battery as the source of E.M.F. 
 
 The electrode to which the positive side of the battery 
 is connected is called the anode, and the electrode to 
 which the negative side of the battery is connected is 
 called the cathode. 
 
 572. The current flowing through the electrolytic 
 cell from the battery decomposes the water at a rate in 
 proportion to the strength of the current flowing through 
 the cell. .The gases thus formed are not liberated 
 evenly throughout the liquid, but collect at tlie electrodes 
 and then rise to the surface in bubbles, the oxygen 
 collecting at the anode and the hydrogen at the cathode.
 
 WIRELESS TELEGRAPHY 155 
 
 As water consists of two atoms of hydrogen and one 
 of oxygen, twice as much gas will collect at the cathode 
 as at the anode. 
 
 573. The amount of water decomposed, and therefore 
 tlie amount of gas given off at the electrodes, is pro- 
 portional to the current passing through the liquid, it 
 follows, therefore, that if the resistance of the electro- 
 lytic cell, as well as the resistance of the outside circuit 
 supplying the E.M.F. necessary to pass the current 
 
 FIG. 89. 
 
 through the cell, remained constant, the amount of gas 
 given ofE would be proportional to the ' voltage applied 
 across the cell. 
 
 574. This condition, however, is not obtained in 
 practice, for when only a small voltage is applied across 
 the cell the gas is not produced in sufficient quantity 
 to break away from the electrodes, instead it sticks to 
 the surface of the platinum in the form of tiny bubbles, 
 thus reducing the area of contact between the ele'ctrode 
 and liquid. This has the result of very considerably 
 increasing the resistance of the electrolytic cell. This 
 effect is known as polarisation, and when the gas is 
 removed either by mechanical, chemist or other means, 
 the effect is known as de-polarisation.
 
 156 WIRELESS TELEGRAPHY 
 
 1>75. In the case of electrodes being of a considerable 
 size, that is to say, having a large surface in contact 
 with the liquid, the increase in the resistance of the cell 
 is not so marked, and, moreover, a stronger mechanical 
 or chemical action is required to de-polarise it than if 
 the contact surface between the electrode and liquid 
 ;is very small. 
 
 576. Now, provided the electrode has a sufficiently 
 (small area of contact with the liquid to begin with, this 
 polarising action continues until the resistance of the 
 cell is increased to such an extent that practically no 
 
 j more current can pass through it at that voltage. 
 
 577. If the voltage across the cell be then increased, 
 current will again be forced through the cell with a 
 corresponding increase in the amount of gas formed, 
 until finally the gas will collect in sufficiently large 
 bubbles to break away from the electrodes and rise to 
 the surface of the liquid. When this happens the 
 liquid again comes into contact with the electrode, with 
 the result that the resistance of the cell drops ; this 
 effect in turn allows a larger current to flow through 
 the cell, and the gas then continues to be liberated in 
 sufficient volumes to bubble freely away from the electrode 
 until the current is cut off. 
 
 The electrolytic cell is most sensitive to the action 
 I of the high frequency currents when it is hi the 
 first stage of the process described above, that is, when 
 \the initial voltage applied across the cell is only sufficient 
 to form a cushion of gas round the electrode. If at this 
 stage a high-frequency voltage is applied to the cell, 
 the cushion of gas is apparently broken up and the 
 liquid again comes into contact with the electrode, 
 causing a sharp drop in the resistance of the cell. Thus
 
 WIRELESS TELEGRAPHY 
 
 157 
 
 we may say that the action of high frequency oscillating 
 currents on a polarised electrolytic cell is to de-polarise 
 the cell. 
 
 578. Figs. 90 and 91 show how the electrolytic de- 
 tector can be connected up to the receiver circuits. If 
 these diagrams be compared with Figs. 75 and 77, which 
 show the method of connecting a crystal detector of 
 the receiver circuits, it will be noticed that identically 
 the same circuits are used in both cases. 
 
 T 
 
 FIG. 91. 
 
 579. The action to the two detectors, however, is 
 fundamentally different, and it is essential that this 
 difference be thoroughly understood. 
 
 In paragraphs 429-433 we explained that the function 
 of a crystal is to rectify oscillating currents received, 
 thus converting them into uni-directional currents 
 which can be made to produce audible sounds in the 
 ear-piece of a telephone (vide also paragraph 525) 
 
 580. Now let us see what happens in the case of an 
 electrolytic detector ; by following the diagram in 
 Fig. 90 it will be seen that the initial voltage across 
 the detector D is provided from the battery B and 
 potentiometer P. By means of the potentiometer the
 
 158 WIRELESS TELEGRAPHY 
 
 voltage is adjusted to such a value that the resistance 
 of the cell is at its highest (vide paragraphs 576 and 577). 
 Since the telephones T are in series with the detector, 
 the current passing through the telephones will then be 
 practically zero. In any case what little current there 
 is will be of constant value, and will therefore produce 
 no sound in the telephones (vide paragraph 486). As 
 soon as a train of high frequency oscillations are induced 
 in the jigger secondary, the electrolytic detector becomes 
 de-polarised, and its resistance therefore drops, allowing 
 an increased current to flow from the battery through 
 the telephones and through the detector, this increase 
 in the current producing an audible click in the tele- 
 phones. 
 
 581. If the electrolytic cell is suitably made it will 
 recover its polarisation in time for the next group of 
 oscillations, so that each group of oscillations will 
 produce a click in the telephones ; therefore, as in the 
 case of the carborundum detector, the sound produced 
 in the telephones will correspond exactly to the spark 
 frequency of the transmitter. 
 
 582. In order to produce an electrolytic detector 
 which will in the first place be sensitive to very feeble 
 oscillations, and also which will recover its polarisation 
 rapidly enough for the highest practicable spark fre- 
 quency, the active electrode which is the anode must 
 be of extremely small dimensions. To this end the 
 anode usually consists of an extremely fine platinum 
 wire fused into a glass holder, a magnified sketch of 
 which is shown in Fig. 92. The end of the platinum 
 wire projects only a fraction of a millimetre beyond the 
 glass ; thus no matter how much the glass is immersed 
 in the electrolyte, only that part of the platinum wire
 
 WIRELESS TELEGRAPHY 
 
 150 
 
 which projects beyond the glass will come into contact 
 with the liquid. 
 
 583. So long as the anode is small the size of the 
 H 
 
 Fio. 92. 
 
 cathode is of no importance, and for this reason 
 usually consists of a comparatively thick platinum wire 
 thoroughly immersed in the liquid at any convenient 
 distance. 
 
 THE MAGNETIC DETECTOR 
 
 584. In the Magnetic Detector use is made of that 
 property in iron known as magnetic hysteresis. 
 
 MAGNETIC HYSTERESIS 
 
 If a piece of iron be brought near the pole of a 
 magnet, that part which is nearest the magnet becomes 
 magnetised to an opposite polarity by magnetic induc- 
 tion (see paragraph 83). 
 
 585. If it is then removed from this magnetising 
 force, it will still retain a certain amount of the magnetism 
 induced into it, by reason of its " retentivity. 
 
 586. In the case of soft iron this residual magnetism 
 is extremely unstable, and a very small mechanical 
 shock or twist is quite sufficient to destroy it. 
 
 In- other words, the magnetism in the iron does not
 
 160 WIRELESS TELEGRAPHY 
 
 follow exactly any change in the magnetising force, but 
 lags a little behind it This lagging behind is called 
 " Hysteresis." 
 
 587. In paragraph 98, under the heading of " Magnetic 
 Induction,"' we explained how an electric current could 
 be induced in a coil of wire by causing a change in the 
 strength of the magnetic field passing through the coil. 
 
 If, then, we wind a coil of wire round an iron 
 core, and after magnetising the latter, we subject it to 
 a mechanical shock, sufficient to destroy its residual 
 magnetism, a current will be induced in the coil of wire. 
 Moreover, if we connect a pair of telephones across the 
 coil, thus causing the current induced into it to pass 
 through the telephones, we shall get a click in the 
 telephones when the iron is de-magnetised. 
 
 Having de-magnetised the iron it must be re- 
 magnetised before a similar mechanical shock will 
 produce another current impulse in the coil. 
 
 588. It is obvious that the intensity of the current 
 induced in the coil of wire will depend on the difference 
 between the amount of magnetism in the iron before and 
 after it is subjected to the de-magnetising influence of 
 the mechanical shock. 
 
 589. A very feeble shock will only partially de- 
 magnetise the iron, with the result that a feeble sound 
 is produced -in the telephones ; but if the shock is suf- 
 ficiently strong to destroy all the magnetism in the iron, 
 we shall get a maximum sound in the telephones, and 
 any further increase in the strength of the shock cannot 
 further increase the sound produced in the telephones. 
 
 590. We have, however, a means of still further 
 increasing the strength of the current induced in the coil. 
 Owing to its hysteresis the iron will retain its residual
 
 WIRELESS TELEGRAPHY 161 
 
 magnetism, not only when the magnetising force has 
 been removed, but also when it has been reversed, 
 provided that this reversed magnetising force is not too 
 powerful. 
 
 In this case the effect of our subjecting the iron to 
 a mechanical shock is not merely to destroy its residual 
 magnetism, but to allow it to become magnetised in 
 the opposite direction by the influence of the reversed 
 magnetising force. 
 
 The intensity of the current induced in the coils will 
 then be proportional to the amount of residual magnetism 
 in the iron which is destroyed, plus the amount by which 
 it is magnetised in the opposite direction. 
 
 591. It was discovered that if a high frequency 
 oscillating current were passed through a coil of wire 
 round a piece of iron, it produced an effect on the iron 
 similar to that produced by a mechanical shock. 
 
 592. Let us now see how these principles are applied 
 in the magnetic detector. 
 
 An endless band B, Fig. 93, consisting of a number 
 of fine strands of iron wire, is passed over two pulleys 
 P, Pj, one of which is kept slowly rotating by means of 
 clockwork, thus keeping the band continuously moving 
 in the direction indicated by the arrow. The band is 
 made to pass through a small glass tube C, around which 
 is wound a single layer of insulated copper wire, the two 
 ends of which are connected, one to the aerial and the 
 other to the earth. 
 
 A second coil of wire D, consisting of a very much 
 larger number of turns of wire, is also wound around the 
 glass tube, and across this coil is connected a pair of 
 telephones T. A single horse-shoe magnet M is placed 
 in a position similar to that shown in Fig. 93, 'with one 
 
 M
 
 162 
 
 WIRELESS TELEGRAPHY 
 
 of its poles (in this case the north pole) close to the band 
 a short distance away from the windings, the other pole 
 a little distance away from the band near the middle of 
 the windings. 
 
 593. Let us now watch the progress of a particular 
 portion of the band* while it travels from the point X 
 to the point Y. 
 
 It first of all approaches the north pole of the magnet, 
 and thereby becomes magnetised as a south pole by 
 
 FIG. .93. 
 
 magnetic induction. As it proceeds farther on its course 
 it gets farther and farther away from the magnetising 
 influence of the north pole of the permanent magnet, 
 but, owing to its hysteresis, it will retain a certain amount 
 of magnetism. As it enters the glass tube it commences 
 to come under the weaker influence of the south pole of 
 the magnet, which is tending to make it into a north pole,
 
 WIRELESS TELEGRAPHY 163 
 
 but unless disturbed it will retain its original residual 
 magnetism as a south pole. 
 
 594. When an oscillating current is induced in the 
 aerial, this current will pass > round the single layer 
 winding on the glass tube and allow the magnetism in the 
 iron to be reversed, thus causing a sudden change in the 
 magnetic field, and thereby inducing a momentary 
 current in the secondary coil, to which the telephones 
 are connected. 
 
 595. If, on the other hand, no oscillations are received 
 m the aerial, the iron will pass through the primary tube 
 without having its magnetic polarity suddenly changed, 
 and therefore no sound will be produced in the 
 telephones. 
 
 596. It will be seen, then, that we have a continuous 
 supply of iron inside the primary tube in such a 
 condition that oscillating currents passing through 
 the coil of wire will cause it to change its polarity 
 suddenly. 
 
 597. Experience has shown that the magnetic detector 
 is quite the most reliable and robust form of receiver 
 which has yet been invented, but although extremely 
 sensitive, it is not as sensitive as the modern crystal 
 detectors. Its reliability, however, makes it a valuable 
 instrument as a stand-by, or in places where experienced 
 operators are not obtainable. 
 
 598. To tune up the magnetic detector to any desired 
 wave-length, an adjustable inductance and an adjustable 
 condenser are joined in series with the de-magnetising 
 or primary winding of the detector. 
 
 599. No tuning is required for the secondary or 
 telephone winding, for the currents induced in the 
 secondary are not oscillatory so that normally the
 
 164 
 
 WIRELESS TELEGRAPHY 
 
 magnetic detector can be regarded as a single-circuit 
 receiver. 
 
 600. As this arrangement does not give particularly 
 sharp tuning, an instrument was designed, known as 
 the Multiple Tuner, through which the oscillations have 
 to pass before, reaching the primary winding of the 
 magnetic detector. 
 
 601. The multiple tuner consists of three oscilla- 
 
 FIG. 94. 
 
 tory circuits acting inductively upon each other, each 
 of which is adjustable as regards its wave-length. A 
 diagram of connections of a simple form of this tuner 
 is shown in Fig. 94. 
 
 As will be seen from the diagram, there are three 
 distinct circuits, namely, the " aerial " circuit, the 
 " intermediate " circuit, and the " magnetic detector " 
 circuit. 
 
 Each of these circuits must be in tune with the 
 wave-length it is desired to receive.
 
 WIRELESS TELEGRAPHY 16P 
 
 602. The aerial circuit consists of an adjustable 
 inductance A, an inductive winding B, and an adjustable 
 condenser C, all connected in series with one another. 
 
 The inductive winding B is so placed that any 
 oscillations in it induce similar oscillations in the inter- 
 mediate circuit, which consists of two inductive windings 
 D and E, connected in parallel and across an adjustable 
 condenser F. 
 
 Any oscillations induced in the winding D will also 
 flow in the winding E, since the two windings are in 
 parallel, and the wave-length of the circuit is adjusted 
 by changing the capacity of the condenser F. 
 
 603. The third circuit the magnetic detector circuit 
 consists of an inductive winding G, and an adjustable 
 condenser H, and by itself is incomplete, but it is com- 
 pleted by connecting in series with it the primary 
 winding of the magnetic detector as shown by J. 
 
 The telephones are connected as described before, 
 across the secondary coil of the magnetic detector 
 
 " ATMOSPHERICS " 
 
 604. Electric disturbances in the atmosphere which 
 affect the receiving apparatus of wireless telegraph 
 stations are known by the name of " atmospherics." 
 
 They produce in the telephones noises which, if strong 
 enough, will drown the signals being received 
 
 Where small aerials are being used, these atmospherics 
 are not usually troublesome, but where large and high 
 aertals are employed, if measures were not taken to 
 reduce their effect, it would be impossible, sometimes for 
 days together, to communicate at all. 
 
 605. The difficulty in getting rid of " atmospherics "
 
 166 WIRELESS TELEGRAPHY 
 
 is that they have no particular tune, but will cause the 
 aerial circuit to oscillate to its own natural frequency, 
 
 so that, no matter to what wave-length the circuit is ad- 
 justed, " atmospherics " are still induced in the receiver. 
 
 606. One method of avoiding them can be described 
 briefly as follows : 
 
 Two receiving circuits are opposed to one another in 
 such a way that if equal effects are produced in each 
 circuit, these effects are neutralised, and therefore produce 
 no sound in the telephones. 
 
 If one of these circuits is in tune with the wave-length 
 being received, and the other circuit is out of tune with 
 this wave-length, unequal effects will be produced in the 
 two circuits by the waves, and the signals will be received 
 in the ordinary way. 
 
 The " atmospherics," however, as already stated, will 
 affect both circuits equally, so that the effect of the 
 " atmospherics " is neutralised, and they will produce no 
 sound in the telephones. 
 
 Although the principle of this method is simple, 
 the application of it to different receivers is extremely 
 complicated, and for this reason we have not described 
 the arrangement in detail. 
 
 607. There is another form of " atmospheric " called 
 " static," which is extremely troublesome should a 
 condenser be connected in series with the aerial for the 
 purpose of tuning. 
 
 The atmospherics continually charge up this con- 
 denser until either the condenser is broken down or the 
 charge sparks across the two sides of the condenser. 
 
 608. " Statics," however, can very easily be dealt 
 with by connecting a coil of wire, as shown by S, Fig. 94, 
 from the aerial side of the condenser to earth, which
 
 WIRELESS TELEGRAPHY 167 
 
 allows the current to pass through the coil of wire to 
 earth instead of charging up the condenser. 
 
 It is, however, necessary that this coil be highly 
 inductive, as otherwise not only would the current 
 caused by the " atmospherics " pass through it, but also 
 the oscillating currents, thereby interfering with the 
 tuning effect of the condenser C. 
 
 609. For this reason, in nearly all receivers that are 
 provided with aerial tuning condensers, a coil of wire, 
 known as an " inductive shunt," is connected from the 
 " aerial " side of the condenser to earth. 
 
 AERIALS 
 
 610. The function of an aerial, as we have already 
 shown, is twofold. 
 
 In the first place it is required to radiate energy in 
 the form of aether waves from the oscillating currents 
 flowing in it. In this case it may be said to act as a 
 radiator. 
 
 In the second place it has to pick up energy in the 
 form of oscillating currents from aether waves which 
 cross it. In this case it may be said to act as an ab- 
 sorber.. It is found that any oscillatory circuit which 
 is efficient as a radiator (vide paragraph 289) will also 
 act efficiently as an absorber, and for this reason in 
 nearly every case the same aerial is used both for the 
 purpose of transmitting and receiving. 
 
 SHAPE OF AN AERIAL 
 
 611. The shape any particular aerial takes will 
 depend upon many practical considerations.
 
 168 WIRELESS TELEGRAPHY 
 
 The shape of an aerial can be roughly classified under 
 one of four headings, namely, " Vertical " aerials, 
 
 UMBRELLA AERIAL 
 
 I 
 
 A\V\VS 
 
 T AERIAL 
 
 INVERTED L' OR "T AE.RIAU 
 
 FIG. 95. 
 
 " Umbrella " aerials, " T " aerials, and inverted " L " 
 aerials. Fig. 95 represents an example of each of these 
 
 612. For very small portable stations the u 1 jrella
 
 WIRELESS TELEGRAPHY 169 
 
 aerial is found to be convenient, principally because it 
 only requires one mast to support it, and the aerial, 
 instead of putting a side stress on the mast as in the 
 case of " T " or inverted " L " aerials, can be made to 
 act as a set of stays, thus assisting to support the 
 mast. 
 
 613. The " T " aerial, on the other hand, is used 
 largely on board ship, where in nearly every case two 
 masts are available and the wireless cabin is usually 
 amidships. 
 
 614. Sometimes, however, it is more convenient to use 
 the inverted " L " aerial, on account of the position of 
 the wireless cabin. The inverted " L " aerial is used 
 very largely where very long aerials are required, such 
 as in the case of long-distance stations. It is also used^ 
 largely for military stations of medium size, on account of 
 the fact that it is easy to erect in almost any position. It 
 can, for instance, be as easily erected in a road or street 
 as it can in an open field ; thus for military purposes 
 offering a great advantage over the umbrella aerial. 
 
 SIZE OF AN AERIAL 
 
 615. The next point to consider is the size of an aerial. 
 This depends chiefly upon the wave-length it is required 
 to transmit, which in turn depends to a large extent 
 upon the power that it is necessary to use. 
 
 Every aerial has its own natural wave-length, called 
 its fundamental wave-length, depending upon its own 
 capacity and its own inductance (vide paragraph 292 
 onwards). 
 
 616. If we increase the length of an aerial we in- 
 crei 5 both its capacity and its inductance, and thereby
 
 170 WIRELESS TELEGRAPHY 
 
 increase its fundamental wave-length. If we add on 
 to an aerial another parallel wire, we increase the 
 capacity of the aerial, because two capacities in parallel 
 result in a larger capacity, but at the same time we 
 decrease the inductance of the aerial, because two in- 
 ductances in parallel result in a lower total inductance. 
 Thus it is found that by adding another wire to an aerial 
 its fundamental wave-length remains more or less un- 
 altered. 
 
 617. If, .however, instead of keeping the additional 
 wire or wires of the aerial parallel with one another we 
 separate them out radially, as in the case of the umbrella 
 aerial, and if the extremities of the radial wires approach 
 the earth, as is usually the case in the umbrella aerial and 
 also in some forms of the " T " aerial, then the capacity 
 of the aerial as a whole is increased more rapidly than 
 the inductance is decreased, because the inductance of 
 the down lead is unaltered, with the result that 'the 
 fundamental wave-length is increased. 
 
 618. In practice it is found that with single- wire, aerials 
 ^r parallel-wire aerials, whose wires run either vertically 
 or horizontally, the wave-length is usually about four and 
 a quarter times the length of the aerial. With a " T " 
 aerial, the upper portion of which is kept horizontal, the 
 fundamental wave-length is about five times the length 
 of the aerial, but if the ends of the wires are brought down 
 so as to approach the earth, the wave-length will be still 
 further increased in proportion to the length of the wire. 
 
 619. With an " umbrella " aerial, the wave-length 
 may be as much as eight times the length of the aerial, 
 according to the number of radial wires forming it and 
 the height of their ends from the earth, and the height 
 of the mast.
 
 WIRELESS TELEGRAPHY 171 
 
 Thus, it will be seen that two or more aerials, both 
 having exactly the same fundamental wave-length, can 
 have different proportions of capacity and inductance. 
 
 620. We have already shown, in paragraph 294, 
 that we can increase the wave-length of the aerial by 
 connecting an inductance in series with it. Further, 
 we can decrease the wave-length of an aerial by con- 
 necting a condenser in series with it. 
 
 We are, however, limited by practical considerations 
 in the extent to which we can thus increase or decrease 
 the wave-length of an aerial from its fundamental value. 
 The reason for this is that the aerial is most efficient 
 as a radiator, and therefore also as an absorber, when 
 neither inductance nor capacity has been connected in 
 series with it (vide paragraph 295). 
 
 621. For thir reason it is usual to design an aerial 
 so that its fundamental wave-length is approximately 
 that to which it is to be used for transmitting or 
 receiving. 
 
 It is found in practice convenient for a station to 
 transmit on only a limited number of wave-lengths, but 
 it is essential that the same station be able to receive 
 over a very large range of wave-lengths, and therefore 
 it is usual to consider only the transmitting requirements 
 when designing the aerial. 
 
 622. The construction of an inductance coil, suitable 
 for doubling the wave-length of a given aerial, is far 
 cheaper and also more efficient than a condensed suitable 
 for halving the fundamental wave-length of an aerial, 
 and for this reason the aerial is usually designed to have 
 a fundamental wave-length equal to, or rather shorter 
 than, the shortest wave-length it is required to transmit. 
 There are, of course, exceptions to this rule where special
 
 172 WIRELESS TELEGRAPHY 
 
 conditions have to be fulfilled. For the purpose of this 
 book it is unnecessary to deal with them here. 
 
 HEIGHT OF AN AERIAL 
 
 623. The height of an aerial is a very important con- 
 sideration, because it is found that the range oJ a station 
 of a given power is directly proportional to the average 
 height of the aerial. Thus, if we double the average 
 height of the aerial of a given station, we double 
 the range of that station without increasing the power 
 we have to radiate. 
 
 624. It depends, of course, entirely on circumstances 
 whether it is cheaper, or for other reasons more con- 
 venient, to increase the height of an aerial or to increase 
 the power of the station in order to increase the range. 
 For portable stations it is obviously convenient to keep 
 the masts as low as possible, and to keep the aerial as 
 simple as possible, for tall masts are not only heavy for 
 carrying about, but take a considerable length of time 
 to erect. 
 
 It is. found in practice that for stations that are 
 going to be carried about by hand or on horseback, 
 30 feet is a very convenient height of mast, although, 
 where time taken to erect is not of primary importance, 
 masts 50 feet or even 70 feet high can be conveniently 
 used. 
 
 625. Further, the cost of a mast very rapidly increases 
 with its height, and it therefore becomes a question 
 on this account whether it is cheaper to increase the 
 power of the station or to increase the height of the 
 masts.
 
 WIRELESS TELEGRAPHY 
 
 173 
 
 THE ADVANTAGE OF USING AERIALS OF A LARGE 
 CAPACITY 
 
 626. The advantage of having a greater capacity iu 
 the aerial is very apparent when we try to increase 
 the wave-length by adding an inductance in series 
 with it. Adding an inductance to an aerial reduces 
 its efficiency, and also introduces difficulties of insulation 
 which will be better appreciated after reading para- 
 graph 632 onwards, so the less inductance we have to 
 add to obtain the required wave-length the better. 
 
 627. A large- capacity aerial requires less inductance in 
 series with it to increase its wave-length to a given value 
 
 B 
 
 \\\\\\\\\\\\\ 
 
 Fia. 
 
 than a small-capacity aerial, assuming, of course, that 
 the fundamental wave-lengths of the two aerials are the 
 same.
 
 174 WIRELESS TELEGRAPHY 
 
 It is quite easy to show this by the application of 
 the formula .given in paragraph 259, namely : 
 
 Xm = 1885 /C(mf) x L(mh) . 
 
 We will suppose that the fundamental wave-length 
 of each of the aerials' A and B, shown in Fig. 96, is 
 100 metres, but in the aerial B the capacity is larger, 
 and therefore the inductance is smaller than in the 
 aerial A. We will suppose that in the aerial A 
 
 C=-0002 microfarad 
 
 L = 14- 1 microhenries, 
 and in the Aerial B 
 
 C=-0004 microfarad 
 
 L = 7- 05 microhenries ; 
 then, without any extra inductance, 
 
 Xm of aerial A = 1885 N/-0002 x 14'1 
 
 =1885 x- 053 
 =100 metres, 
 and also 
 
 Xm of aerial B = 1885 */'0004 x 7 '05 
 = 1885 x -053 
 = 100 metres. 
 
 Now let us add on to each aerial an additional in- 
 ductance of 10 microhenries. Then we shall have 
 
 Xm of aerial A = 1885 N/'0002x24-l 
 =18&5x-0695 
 =131 metres (about) ;
 
 WIRELESS TELEGRAPHY 175 
 
 but 
 
 \m of aerial B = 1885<v/'0004 x 17'05 
 = 1885x0-826 
 =156 metres (nearly). 
 
 628. It will be seen that with the same additional 
 inductance we have increased the wave-length of the 
 aerial B from 100 metres to 156 metres, while we have 
 only increased the wave-length of the aerial A from 
 100 metres to 131 metres. 
 
 Thus, if we wished to increase the wave-length of 
 the aerials shown in Fig. 96 from 425 feet to, say, 600 
 feet, we should find that perhaps 10 turns of an in- 
 ductance coil would be required in the case of the 
 aerial A, while only about six turns would be required 
 in the case of the aerial B, as illustrated. 
 
 THE LENGTH OP AN AERIAL 
 
 629. The length of an aerial is not necessarily the 
 total length of wire, but is the length of wire from the 
 
 Fio. 97. 
 
 point where it is connected to the instruments to any 
 one of its extremities. Thus in the " umbrella " aerial
 
 176 
 
 WIRELESS TELEGRAPHY 
 
 shown in Fig. 97 the length of the aerial is 105 feet, 
 made up of 30 feet of " down-lead " and 75 feet of radial 
 wires. 
 
 630. The length of the " T " aerial shown in Fig. 
 98 is 150 feet, made up by 50 feet of " down-lead " and 
 100 feet of horizontal wire in a 200-foot span. 
 
 ^v 
 
 x/ 
 
 FIG. 98. 
 
 FIG. 99. 
 
 631. And the length of the " L " aerial shown in Fig. 
 99 is 200 feet, made up by 50 feet of " down-lead " 
 and 150 feet of horizontal wire.
 
 WIRELESS TELEGRAPHY 
 
 177 
 
 DISTRIBUTION OF POTENTIAL AND 
 CURRENT ALONG AERIALS 
 
 632. Before going into the question of the construc- 
 tion and insulation of aerials, let us first consider how 
 the voltage of the aerial is distributed when oscillating 
 currents are flowing in it. 
 
 633. Taking the case of an aerial directly energised 
 by an induction coil, i.e. " plain aerial," the maximum 
 initial voltage to which the aerial is charged will depend, 
 as we have shown, upon the voltage applied across c the 
 spark-gap by the induction coil. This voltage can be 
 regulated by increasing or decreasing the length of the 
 spark-gap. 
 
 634. (The voltage required to jump an air-gap where 
 points are used for the electrodes is about 
 
 12,000 volts for each centimetre length of air- 
 gap. If knobs or balls are used for elec- 
 trodes, the voltage required will be increased 
 to an extent depending upon the curvature of 
 the ball. Thus, using balls one inch in dia- 
 meter, it is found that a voltage of approxi- 
 mately 30,000 volts for each centimetre 
 length of air-gap is required.) 
 
 635. Assuming for the sake of explanation \ 30.000 
 that we are using ball electrodes set one centi- [^ ** C\ 
 metre apart, then at the instant immediately 
 
 before the gap is broken down the whole aerial 
 
 is charged up to a uniform potential of 30,000 
 
 volts. In this case the distribution of the 
 
 voltage along the aerial wire can be shown diagram- 
 
 matically, as in Fig. 100, by a dotted line drawn parallel 
 
 woo 
 
 FIG. 100.
 
 178 WIRELESS TELEGRAPHY 
 
 with the aerial at a distance from it representing 30,000 
 volts. 
 
 636. Now the energy stored up in a condenser can 
 be regarded as " potential " energy, as opposed to the 
 " kinetic " energy which is stored up in an inductance 
 through which a current is flowing ; just as the. mechani- 
 cal energy stored up in a compressed spring is in the 
 form of "potential" energy as opposed to the "kinetic" 
 energy which is stored up in a revolving fly-wheel or 
 in any moving body. 
 
 637. In all periodic oscillations, whether mechanical 
 or. electrical,' the energy is continually changing from 
 the one form to the other. Thus, for example, in 
 the case of a vibrator, as shown in Fig. 37, which is 
 made to vibrate between the positions W : and W 2 , when 
 it occupies one or other of these extreme positions, it 
 is for the moment at rest and all the energy is then 
 in the form of " potential " energy stored up in the 
 tension of the spring blade. On the other hand, when it 
 occupies the position W, there is for the moment no 
 tension in the spring, but at this moment the weight 
 W is travelling at its maximum speed, consequently 
 all the energy is then in the form of " kinetic " energy 
 stored up in the moving body W At intermediate 
 positions obviously part of the energy is stored up in 
 the tension of the spring and part of it in the moving 
 weight W 
 
 Similarly, in the case we are now considering, namely, 
 that of an oscillatory circuit, the energy at one instant is 
 all stored up in the condenser in the form of " potential " 
 energy, arid at this moment the electricity is at rest, 
 that is to say, there is no current flowing. At the next 
 instant when the condenser is fully discharged and
 
 WIRELESS TELEGRAPHY 179 
 
 before it com indices to be charged in the Opposite 
 direction, the current flowing in the circuit is at 
 its maximum, and all the energy is then in the form 
 of " kinetic " energy stored up in the current flowing 
 through the inductance of the circuit. Thus it will be 
 seen that the total energy in an aerial or other oscillatory 
 circuit at any moment is the sum of the " potential " 
 energy and the " kinetic " energy. 
 
 638.. In paragraph 285, we showed how the energy 
 stored up in a condenser at any moment could be 
 determined from the equation E=|CV 2 , where C is 
 the capacity of the condenser ; and V the E.M.F. to 
 which it is charged. 
 
 Taking the analogous mechanical case of energy 
 stored up in a compressed spring, it can be shown that 
 at any moment the energy E =|(F)T 2 , where (F) is the 
 flexibility l of the spring, and T the tension to which 
 it is stressed. 
 
 639. Now the energy stored up in a moving body, 
 i.e. the " kinetic " energy, depends upon the weight of 
 that body and the speed at which it is travelling. The 
 energy is directly proportional to the weight of the 
 moving body and proportional to the square of the speed 
 at which it is travelling. If M is taken to represent the 
 mass of the body and V to represent its velocity, then 
 it can be shown that energy stored up in it at any 
 moment E =|MV 2 . 
 
 640. Similarly, in an electrical circuit the energy 
 stored up in the inductance L of the circuit through 
 
 1 Note the term flexibility is used for the sake of simplicity, 
 and the equation will be found correct if F is taken equal to ^, 
 
 where L is the elongation of the spring, and E is the modulus of 
 elasticity.
 
 180 WIRELESS TELEGRAPHY 
 
 which a current is flowing, is directly proportional to 
 the inductance and proportional to the square of the 
 current flowing through that inductance. If J = energy 
 in joules, L ^inductance in henries, and I = current in 
 amperes, it can be shown that J = JLI 2 
 
 641. Returning to. the particular case we are consider- 
 ing, namely, that of an aerial which is uniformly charged 
 to a pressure of 30,000 volts, then it is evident- that at 
 the moment before the spark-gap is broken down, all 
 the energy is stored as " potential " energy, because at this 
 moment there is no current flowing ; therefore if C is 
 the capacity of the aerial, and V the voltage to which 
 it is charged, J =|t!V 2 . 
 
 At the next instant the gap is broken down, and 
 the " potential " energy in the charged aerial is gradually 
 transferred to " kinetic " energy in the form of a current 
 of electricity passing through the inductance of the aerial 
 to earth. 
 
 642. If L represents the inductance of the aerial, and 
 I represents the current flowing through that inductance, 
 and J again represents the energy in the aerial, then, 
 when the current is oscillating, the energy in the aerial 
 at any instant =|C(V) 2 + |L(I) 2 . 
 
 643. Taking the instant when one-quarter of an 
 oscillation has taken place, the Voltage of the aerial has 
 become zero, and therefore CV 2 =0. At this instant, 
 therefore, the whole of the energy is transferred to the 
 current flowing through the inductance and will then 
 equal |LI 2 , and consequently the current in the aerial 
 is at its maximum. 
 
 Similarly, taking the instant when one-half oscillation 
 has. taken place, the current Jhas become zero, and 
 therefore at this instant the whole of the energy is
 
 WIRELESS TELEGRAPHY 181 
 
 transferred to the charge in the capacity of the aerial, 
 and will then equal CV 2 , and consequently the voltage 
 of the aerial is at its maximum. 
 
 644. Assuming, for the moment, that none of the 
 energy is lost either in radiation or resistance, then the 
 energy in the aerial will be the same at the end of the first 
 oscillation as it was originally at the moment immediately 
 before a spark occurred, but since the spark-gap is now 
 broken down the bottom end of the aerial 
 
 must be considered as connected to earth, j- . 
 and therefore the voltage at this point \ 
 will remain at zero while the maximum \ 
 voltage will be found at the free end of \ 
 the aerial. Thus, it will be seen that the \ 
 distribution of voltage over the length \ 
 of the aerial will take a different form \ 
 from that shown in Fig. 100. It will take \ 
 the form of the curve shown in Fig. 101. \ 
 
 At first sight it would appear that the \ 
 
 maximum value of the voltage obtained v ^ - 
 
 at the free end of the aerial would be VoLTS 
 the voltage to which it was originally 
 charged. This, however, is not the case, FIG. 101. 
 for the following reason. 
 
 645. At the moment when the aerial is first charged, 
 the charge is uniformly distributed over the whole of 
 the aerial, as shown in Fig. 100, and since the capacity of 
 the aerial is also distributed over the whole of the aerial, 
 it follows that the whole of the capacity of the aerial is 
 charged to an equal voltage. 
 
 646. On the other hand, after the first oscillation 
 when the charge in the aerial is distributed, as shown 
 in Fig. 101, the whole of the capacity in the aerial
 
 182 
 
 WIEELESS TELEGRAPHY 
 
 is not charged to the same voltage, and therefore the 
 free end of the aerial must necessarily become charged 
 to a higher voltage than originally, in order that the 
 aerial may store up the same amount of energy as 
 before. 
 
 647. It can be shown mathematically th'at the. voltage 
 at the free end of the aerial at the end of a complete 
 oscillation will be -\/2 times, or approximately 1-414 times 
 the voltage to which it was originally charged, assuming 
 (1) that the voltage is then distributed in the form of a 
 sine curve, and (2) that no energy has 
 been lost or radiated during the oscilla- 
 tion. This will be readily understood by 
 referring to Figs. 102 and 103. 
 
 648. Fig. 102 shows the relative distri- 
 bution of the voltage along the aerial wire, 
 the dotted line showing the original charge 
 put into the aerial by the induction coil, 
 and the full line showing the charge in the 
 aerial at the end of the first oscillation. 
 
 649. Fig. 103 shows the variation of 
 voltage at the free end of the aerial wire 
 during the first oscillation. Up to the 
 point A the curve shows the compara- 
 tively slow rise of voltage, while the aerial 
 
 is being uniformly charged up by the induction coil to 
 a value of 30,000 volts. At the point A the. spark-gap 
 breaks down, and oscillations commence, and the voltage 
 at the end of the aerial first drops to zero at the moment 
 B when the aerial is discharged and then rises to a 
 value /v/2 times the original charge = about 42,400 
 volts. 
 
 650. We have said that the energy stored up in the 
 
 O VOLTS 
 
 FIG. 102.
 
 WIRELESS TELEGRAPHY 
 
 183 
 
 aerial when the whole of I he charge is in the aerial \( ' V 2 
 This is only true when the whole of the capacity of the 
 
 42.^00 
 30000 
 
 42.400 - 
 
 FIG. 103. 
 
 aerial is uniformly charged to the voltage V. When the 
 aerial is oscillating, the effective capacity of the aerial 
 
 must be taken as ~ 
 
 the true capacity. 
 
 DISTRIBUTION OF CURRENT IN AN AERIAL 
 
 651. We have already explained in paragraph 643, 
 that at the moment when the voltage at the end of the 
 aerial is at its maximum, the current flowing into the 
 aerial is at zero, and the distribution of the voltage along 
 the aerial when the latter is oscillating to its fundamental 
 wave-length is such that there is a node of potential 
 where the aerial is connected to earth, and an anti-node 
 of potential at the free end of the aerial. The distribu- 
 tion of the current is the reverse of this. 
 
 It is obvious that the maximum current will flow at 
 the earth end of the aerial, for all of the current which 
 flows into the aerial must necessarily pass this point,
 
 184 
 
 WIRELESS TELEGRAPHY 
 
 whereas, taking a point half-way up the aerial, only that 
 current which is required to charge the upper half of the 
 aerial will flow past this point, and taking the extreme 
 end of the aerial no current can flow through it. Thus the 
 distribution of the current in an aerial will also take the 
 form of a sine curve, but with its anti-node at the point 
 where the aerial is connected to earth, and its node at 
 the free end of the aerial ; for this reason the effective 
 
 inductance o! the aerial must be taken as 
 
 2 
 
 Zero RMPS 
 
 times the true inductance. 
 
 7T 
 
 652. We may draw a curve, as shown 
 by the dotted line in Fig. 104, representing 
 the distribution of the current flowing 
 along an aerial, where the distance between 
 the curve and the full line representing 
 the aerial wire represents the compara- 
 tive amount of current flowing. This 
 current will vary from a maximum value 
 when the potential of the aerial is at 
 zero, to zero when the potential of the 
 aerial is at its maximum, but the distri- 
 bution of the current along the aerial will 
 be always in the same proportion, so that although the 
 amplitude of the curve will vary, the form of the curve 
 will remain the same. 
 
 a 
 
 /Mnx. 
 
 .h- HMPS 
 
 Fia. 104. 
 
 INFECT ON CURRENT AND VOLTAGE DISTRIBUTION OF 
 CONNECTING AN INDUCTANCE OR CAPACITY IN 
 SERIES WITH AN AERIAL. 
 
 653. Jt is evident that whatever form the distribu- 
 tion of tfie voltage along an aerial takes, the distribution
 
 WIRELESS TELEGRAPHY 
 
 185 
 
 of tbo current will always take a relative form, but with 
 its node at the point where the voltage is at its maximum, 
 and vice versa, with its anti-node at the point wJiere there 
 is a voltage node. In describing further effects, therefore, 
 it will be sufficient only to indicate the distribution of 
 the voltage along different aerials. 
 
 654. When an inductance is connected in series with 
 an aerial, the distribu- 
 tion of voltage along 
 
 the aerial when oscil- 
 lating is similar to that 
 already described for 
 a simple aerial, except 
 that the inductance 
 must be regarded as 
 a continuation of the 
 aerial, and therefore 
 the voltage increases 
 along the inductance 
 as well as along the 
 aerial, as shown in 
 Fig. 105; thus the 
 greater the inductance 
 that is connected in 
 series with an aerial, 
 the higher will be the 
 voltage across that inductance when it is oscillating. 
 
 655. It is most important to bear this point in mind 
 when designing inductance coils to be used for transmit- 
 ting purposes, for the coil must be very highly insulated 
 from earth at the end which is connected to the aerial, 
 and, further, when there is a large amount of inductance 
 connected in series with the aerial, very high insulation 
 
 Fio. 105. 
 
 FIG. 106.
 
 186 
 
 WIRELESS TELEGRAPHY 
 
 must bo. provided where the aerial wire enters the, 
 building. 
 
 656. The effect on the distribution of the voltage^ of 
 connecting a condenser in series with an aerial, is to 
 create a node of potential in the aerial at some point 
 above the condenser, as shown in Fig. 106. 
 
 657. The exact position of the node will depend upon 
 the relative values of the capacity of the condenser, and 
 the capacity of the aerial. 
 
 Taking the two possible extreme values of capacity of 
 a condenser, we find that if 
 an infinitely large capacity 
 be connected in series with 
 an aerial, the node will be 
 found exactly at the junc- 
 tion of the aerial and the 
 condenser, as shown in Fig. 
 107, for an infinitely large 
 capacity is equivalent to 
 a direct connection to earth. 
 On the other hand, if we 
 connect an infinitely small 
 capacity in series with an 
 
 aerial, the node of potential wjll occur exactly half-way 
 
 up the aerial, as shown in Fig. 108. 
 
 658. Thus any intermediate capacity between the 
 values of infinity and will create a node somewhere 
 between the bottom of the aerial and half-way up the 
 aerial, according to the relative values of the capacity 
 of the aerial and the capacity of the condenser. 
 
 
 1 
 
 
 1 
 
 
 1 
 
 
 1 
 
 
 1 
 
 
 1 
 
 
 1 
 
 
 1 
 
 
 1 
 
 
 I 
 
 
 1 
 
 
 1 
 
 
 1 
 
 
 1 
 
 
 1 
 
 
 1 
 
 CRPKITY 
 
 l 
 
 I N _ 
 
 i 
 
 lMr \ 
 
 
 Fio. 107. 
 
 Fio. 108.
 
 WIRELESS TELEGRAPHY 
 
 187 
 
 HARMONICS 
 
 659. Any oscillatory circuit in which the capacity and 
 inductance are distributed, that is to say, any " open " 
 oscillatory circuit, will oscillate to harmonics of the 
 fundamental wave. 
 
 The first harmonic has a frequency of three times the 
 fundamental frequency, the second harmonic five times, 
 and the third harmonic seven times, and so on. Thus 
 the wave-length of the first harmonic will be one-third 
 the fundamental wave-length, that of the second har- 
 monic one-fifth, and that of the third one-seventh. 
 
 660. When the aerial is 
 oscillating to the first har- ' 
 mom'c the distribution of \ 
 the voltage along it will take \ 
 the form shown in Fig. 110, 
 from which it will be seen 
 that there are two points 
 on the aerial, of maximum 
 voltage, one at the end of 
 the aerial and the other a 
 
 third of the way up the aerial. Further, a node of 
 voltage is obtained at a point two-thirds of the way 
 up the aerial, as well as at the point where the aerial 
 is connected to earth. 
 
 The distribution of voltage along an aerial oscillating 
 to its fundamental, its first harmonic, second harmonic, 
 and third harmonic, is shown diagrammatically in 
 Figs. 109, 110, 111, and 112 respectively. 
 
 661. These harmonics can be distinctly detected with 
 a sensitive wavemeter when an aerial is excited as " plain 
 aerial." It will be noticed, however, that the funda- 
 
 \ 
 
 \ 
 
 \ 
 
 
 \ 
 
 ) f . 
 
 
 } / 
 
 '' \ 
 
 
 i \^ 
 
 
 F ^ 
 io. F 
 
 F "^ 
 [0. F 
 
 W 1 
 
 IG. Fl 
 
 09. 110. 111. 11
 
 188 WIRELESS TELEGRAPHY 
 
 mental wave-length gives by far the strongest effect in 
 the wavemeter. The first harmonic will be very much 
 stronger than the second, and "the second harmonic 
 stronger than the third, and so on. Difficulty will be 
 found in detecting any of the higher harmonics on 
 account of their weakness. 
 
 662. When an aerial is directly excited, as, for 
 instance, by means of an induction coil, the harmonics 
 are only feebly produced, and when an aerial is excited 
 indirectly by a coupled " closed " oscillatory circuit 
 which is tuned to the fundamental wave-length of 
 the aerial, the harmonic wave-lengths of the aerial 
 will be even more feebly produced, practically all of 
 the energy being radiated in the fundamental wave- 
 length. 
 
 663. If, however, the coupled " closed " oscillatory 
 circuit be tuned to one of the harmonics of the aerial, the 
 aerial will not oscillate to its fundamental wave-length, 
 and consequently only the harmonic to which the 
 " closed " oscillating circuit is tuned will be radiated. As 
 a matter of fact, an aerial excited to one of its harmonics 
 will radiate more rapidly than when excited to its funda- 
 mental wave-length. Use is therefore sometimes made 
 of this phenomenon to avoid the necessity of inserting 
 a condenser or a large inductance in series with the 
 aerial, thus reducing its efficiency as a radiator, when 
 a station is required to transmit a long wave-length and 
 a short wave-length on the same aerial, and where it 
 is possible to arrange that the short wave-length is a 
 harmonic of the long wave-length. 
 
 664. In the early days of wireless telegraphy, all 
 ships fitted with wireless could transmit on either of two 
 wave-lengths, which were called respectively " Tune A "
 
 WIRELESS TELEGRAPHY 189 
 
 and " Tune B." Tune A was a wave-length of 3GO feet, 
 and Tune B was a wave-length of 1080 feet. Thus it 
 was usually arranged that the wave-length of the aerial 
 was approximately 1080 feet, and that the primary 
 circuit of Tune A was tuned to the first harmonic of this 
 aerial, namely, 360 feet, and that of Tune B to the 
 fundamental. 
 
 It is, however, not possible to arrange this now, for 
 the International Convention of Radio-Telegraphy have 
 laid down that all ships must be able to transmit wave- 
 lengths of either 600 metres or 300 metres, and in. this 
 case the lower wave-length is not a harmonic of the 
 higher wave-length. 
 
 MASTS 
 
 665. In paragraph 623, we pointed out that the range 
 over which a given transmitter can communicate, is 
 approximately proportional to the height of the aerial. 
 It is obvious, therefore, that where range of communica- 
 tion is of paramount importance, the masts used for 
 supporting the aerial should be as high as practical 
 consideration will allow. 
 
 The masts of permanent Land Stations are usually 
 erected by experienced engineers and riggers, and skilled 
 men are usually available for keeping the masts in good 
 repair, but the masts used on portable stations are 
 frequently handled by men who have had no such 
 experience, and we therefore think that a few remarks 
 on the subject will be useful.
 
 190 WIRELESS TELEGRAPHY 
 
 STRAIN ON MASTS 
 t>66. A mast will withstand a very much greater 
 
 FIG. 113. 
 
 FIG. 114. 
 
 stress acting straight down its length than it will one 
 acting at right angles to its length. 
 
 FIG. 115. 
 
 A simple experiment with a stick about \ inch 
 diameter by 5 feet long will readily illustrate this point. 
 If we exert a pull of say 10 Ib. on one end of the stick
 
 WIRELESS TELEGRAPHY 
 
 191 
 
 in a direction at right angles to its length, as shown in 
 Fig. 113, the stick will probably break, or at all events 
 bend very sharply. .If, however, we exert the same 
 force on the end of the stick in a direction in line with 
 its length, as shown in Fig. 114, the stick will withstand 
 it easily. 
 
 667. In the case of a mast supporting an umbrella 
 aerial, as shown in Fig. 115, the result of all the forces 
 exerted by the wires is a force acting straight down the 
 length of the mast, and therefore the best advantage is 
 being made of the strength of the mast. 
 
 668. But in the case of a mast supporting a horizontal 
 aerial, as shown in Fig. ll6, 
 
 where the aerial is attached 
 to the top of the mast, the 
 force exerted by the aerial is 
 at right angles to the length of 
 the mast, and therefore the 
 strength of the mast is not 
 being used to the best advan- MOST 
 
 PULL OF PETRIPL 
 
 669. If, however, we attach 
 a stay to the top of the mast, 
 and connect the stay to a 
 point on the ground some dis- 
 tance from the foot of the -^ 
 mast, as shown in Fig. 117, x 
 the pull of the aerial will be 
 carried by the pull of the stay, and the resultant of the 
 two forces i.e. of the force exerted by the aerial in 
 one direction, and the force exerted by the stay in 
 another direction is a force acting straight down the 
 length of the mast. 
 
 FIG. 116.
 
 192 
 
 WIRELESS TELEGRAPHY 
 
 670. A very simple way of calculating the amount of 
 the force acting on the stay and that acting on the mast 
 is by drawing a parallelogram, as shown in Fig. 118: 
 
 Assuming that the aerial is exerting a horizontal 
 pull of 200 lb., and we take 1 inch to represent a pull 
 
 PULL or 
 
 Fio. 117. 
 
 of 100 lb., then we may set out a horizontal line, AB. 
 2 inches long, to represent the pull of the aerial. 
 
 671. The force exerted by the mast will be in a 
 vertical direction ; therefore we may set out a vertical 
 line, AC, of indefinite length, representing the direction of 
 the force exerted by the mast. Further, we may draw 
 a line, AD, representing the direction of the force exerted 
 by the stay, the angle oc the same as the angle formed 
 by the stay and the mast. 
 
 If now we draw from the point B a line parallel with 
 the line AD, this line will cut the line AC at the point 
 E, and the length of the line EA in inches will represent 
 the force exerted on the mast in hundreds of pounds.
 
 WIRELESS TELEGRAPHY 
 
 193 
 
 Further, if we draw from the point E a line parallel 
 with the line BA, this line will cut the line AD at 
 the point F, and the length of the line AF in inches 
 will likewise represent the force in hundreds of pounds 
 exerted by the stay. 
 
 672. Measuring these lines in the particular case 
 
 Fio. 119. 
 
 shown in Fig. 118, where the angle between the mast 
 and the stays is 30, we find that the line AE is about 
 3| inches long, and therefore the force required to be 
 exerted by the mast, or the pressure on the mast, is 
 350 lb., while the line AF is about 4 inches long, and 
 therefore the force required to be exerted by the stay, 
 or the pull on the stay, is 400 lb. 
 
 673. By making similar diagrams for various angles 
 between the mast and the stay, as shown in Fig. 119, it 
 will be seen that the greater the angle the less the strain 
 both on the mast and on the stay for a given aerial 
 pull.
 
 194 WIRELESS TELEGRAPHY 
 
 674. Obviously, then, it is an advantage to increase the 
 angle at which we stay a mast, more especially in the 
 case of portable masts, where anchor pegs have to be 
 used for attaching the stays to the ground, and in soft 
 ground a comparatively small pull would be required 
 to pull them out of the ground. 
 
 There are, of course, practical limitations to the 
 extent to which we can do this, for if >ve make the 
 angle too great, not only is a large open space required 
 in which to erect the mast, but also the necessary 
 length of the stays increases very rapidly after an angle 
 of about 30 is reached. 
 
 675. In practice it is usual to make the distance from 
 the foot of the mast to the anchor peg equal to half the 
 length of the mast. The angle between the mast and 
 the stay is then about 27. 
 
 BUCKLING OP MASTS 
 
 676. Let us now make a few further experiments with 
 jw the thin stick described in paragraph 
 
 666. If we take two such sticks of 
 exactly the same diameter and 
 length, one of which is perfectly 
 straight and the other very slightly 
 curved, as shown in Fig. 120, it 
 will be found that the straight 
 stick will carry a far greater weight 
 than the bent stick. If we increase 
 FIG 120 ~ *^ e we ih* on the bent stick gradu- 
 ally, and carefully watch the effect, 
 it will be seen that the bend increases gradually as the 
 weight is increased, until it reaches a certain critical
 
 WIRELESS TELEGRAPHY 195 
 
 bend, depending upon the nature of the wood of which 
 the stick is made. As soon as this critical point is 
 reached, a small increase in the weight will cause the 
 stick to collapse and break. 
 
 We will suppose, for the purpose of explanation, that 
 this critical point is reached when the weight applied is 
 20 lb., and that 21 Ib. is necessary to break the stick. 
 
 677. If now we apply a weight of 21 lb. to the straight 
 stick, it will be found that it carries the weight without 
 any sign of breaking. If, however, we apply a side 
 pressure in the middle sufficient to start a slight bend, 
 the stick will immediately collapse in exactly the same 
 way as the other stick. Only a very slight side pressure 
 is required to start the bend or " buckle." In the case 
 of a mast in the open, the pressure of the wind will be 
 found quite sufficient to start a buckle ; but in any case, 
 all masts made up of a number of loose sections will 
 have a slight bend in them to start with, owing to the 
 play between the plugs and sockets. 
 
 678. If now we take the same sticks and cut down 
 their length by one half, it will be found that exactly 
 the same effects will be produced by applying pressure 
 to the ends, except that it will now take four times the 
 weight to reach the critical point. 
 
 679. In practice it is found that tJie weight a given 
 stick or mast will carry is inversely proportional to the 
 square of its length. By staying the middle of a stick 
 or mast in such a way that the point of attachment 
 of the stays to the mast cannot move sideways, as shown 
 in Fig. 121, we have in effect converted the stick into 
 two sticks, each of half the length, one on top of the 
 other. Thus, by staying a stick or mast in the middle^ 
 we quadruple the weight or pressure it will carry
 
 196 WIRELESS TELEGRAPHY 
 
 MAST STAYS 
 
 680. The material of which the stays are made 
 depends entirely upon circumstances. For portable 
 masts the stays must be very flexible, as they have to 
 be coiled up on to drums when the mast is dismantled. 
 
 681. For masts up to 30 feet in height, rope stays are 
 the most suitable. For masts higher than 30 feet, how- 
 ever, it is better to use metal stays, because rope shrinks 
 
 FIG. 121. 
 
 badly when it is wet and stretches again when dry, the 
 result being that if a mast has been erected when every-- 
 thing is dry a shower of- rain will shrink a long stay 
 sufficiently to pull an anchor peg out of the ground 
 and allow the mast to fall. If, on the other hand, the 
 stays are adjusted when they are wet, they will stretch 
 as they get dry and allow the mast to buckle badly and 
 perhaps break. 
 
 682. For long stays, then, metal should always be
 
 WIRELESS TELEGRAPHY 
 
 197 
 
 ased, and for portable masts phosphor bronze is toimcl 
 to be the best metal for the purpose, although some- 
 what expensive, although it has not the same tensile 
 strength as steel, and will not corrode or rust when 
 exposed to the atmosphere Steel can, of course, be 
 galvanised to stop rusting, but this reduces its strength 
 very considerably, .more especially in the case 01 unely 
 
 stranded wires. In order to make the metal stays 
 flexible they are made up of many strands of fine wire 
 
 683. When vnelal slays are used they must be carefully 
 insulated from the earth, otherwise oscillatory currents 
 will be induced in them on account of their proximity 
 to the aerial, and they would thus absorb a large pro- 
 portion of the transmitted energy, and thereby reduce 
 the range of the station. 
 
 The insulation of stays does not, however, require-
 
 198 
 
 WIRELESS TELEGRAPHY 
 
 to be of a very high order, a short length of rope being 
 in most cases quite sufficient ; for even when wet its 
 resistance will be sufficiently high to stop any oscillatory 
 currents in the stays. Although in this case there will 
 be a certain amount of leakage to earth, the energy thus 
 absorbed would not be sufficient to affect the efficiency 
 of the station to any appreciable extent. 
 
 In tall masts of 200 feet upwards, where the stays are 
 necessarily long, it is usual to divide the stays into two or 
 more sections with rope lanyards, as shown in Fig. 122. 
 Special porcelain insulators are also used for this purpose. 
 684. In the case of portable masts which rarely exceed 
 70 feet or 100 feet in height 
 this division of stays is 
 quite unnecessary, and an 
 insulator consisting of a 
 piece of rope between the 
 stay and the anchor peg 
 is quite sufficient. 
 
 Some means of adjust- 
 ing the length of the stays 
 is necessary, and it is usual 
 to make this piece of rope 
 serve the two purposes of 
 insulating the stay and 
 providing a means of ad- 
 justing its length, as shown 
 in Fig. 123. 
 
 685. In the case of 
 wooden masts no insulation is necessary at the upper end 
 of the stay, but in the case of steel masts the insulation 
 of the stay from the mast is even more important than 
 (he insulation of the stay from the ground. The reason 
 
 ROPE 
 
 Fio. 123.
 
 WIRELESS TELEGRAPHY 
 
 199 
 
 FIG. 124 
 
 for this is obvious by glancing at Fig. 124, when it will 
 be seen that unless 
 
 the stays are insu- 
 lated at the points 
 marked " A," the 
 stays-, together with 
 the mast, form a 
 fair-sized umbrella 
 aerial, connected to 
 earth through the 
 mast; the stays 
 forming the radial 
 wires of the aerial 
 and the mast form- 
 ing the down-lead. This would absorb a very large 
 amount of the energy radiated from the aerial proper. 
 686. In the case of steel masts, therefore, it is 
 
 PLPTE 
 
 POPE ItHSULRTOf? 
 
 &' LOMi 
 
 Wif?E 
 
 FIG. 125. 
 
 necessary to insert insulators between the mast and the 
 stay. For portable masts a length of about 6 inches of 
 rope serves the purpose very well, as shown in Fig. 125. 
 
 THE INSULATION OF AERIALS 
 
 687. The insulation of the aerial is a matter of the 
 utmost importance.
 
 200 WIRELESS TELEGRAPHY 
 
 Bad insulation means a leakage of current and 
 therefore a loss of power. That is to say, instead of 
 radiating all of the energy in the form of ether waves, 
 part of the energy will be lost in leakage during each 
 oscillation. The effect on the oscillating current is to 
 increase the " damping," so that in addition to loss of 
 power we get flatter tuning, due to the more highly 
 damped waves produced. 
 
 688. It is obvious that the longer the energy remains 
 in a leaky aerial the greater will be the loss due to leakage. 
 It is also obvious that an aerial which is a good radiator 
 will not retain the energy put into it for so long as an 
 aerial which is a slow radiator. It follows, therefore, 
 that the slower an aerial radiates its energy the greater 
 the loss of energy due to bad insulation. 
 
 689. In addition to the rate at which an aerial radiates, 
 there is another point to be considered, namely, the rate 
 at which the energy is put into the aerial. Assuming 
 that the faulty insulator acts as a conductor with a 
 high resistance connected to earth, then the rate of 
 leakage from the aerial will be proportional to the 
 voltage of the aerial at the point of leakage, for the 
 higher the voltage the greater will be the current passing 
 through the resistance of the faulty insulator. 
 
 690. Taking the case of a transmitter in which a 
 primary circuit is loosely coupled to the aerial circuit, the 
 energy in the primary circuit is only slowly transferred to 
 the aerial, so that the aerial will not attain its maximum 
 voltage until, perhaps, the third or fourth oscillation. 
 But since, during these oscillations, the aerial is radiating 
 its energy, it follows that the maximum voltage it will 
 attain will not be so high as if the two circuits were so 
 closely coupled together that the whole of the energy
 
 WIRELESS TELEGRAPHY 201 
 
 were transferred to the aerial during the first oscillation. 
 Consequently the loss of energy due to leakage will be 
 less in the case of the loosely coupled circuits than in 
 the case of the closely coupled circuits. 
 
 691. Although faulty insulation is bad even in the 
 case of loosely coupled circuits, inasmuch as it results in 
 less power being radiated, it will not put the station 
 completely out of action, for oscillatory currents will still 
 be induced in the aerial, and therefore waves will be 
 radiated, though the range of communication may be 
 very much reduced. This is one of the reasons that give 
 coupled transmitters such a great advantage over 
 " plain aerial." In the case of a plain-aerial transmitter, 
 as the rise in voltage across the secondary of an in- 
 duction coil, when the primary circuit of the coil is 
 interrupted, takes an appreciable length of time, the 
 charge in the aerial may leak away through the faulty 
 insulators as fast as it is supplied by the induction coil, 
 with the result that it is impossible to get a spark across 
 the electrodes from aerial to earth. And, since the 
 current in the aerial is not oscillatory, until the spark 
 takes place, no oscillating currents are produced, and 
 therefore the aerial, under these conditions, does not 
 radiate at all. 
 
 AERIAL INSULATORS 
 
 '692. The first point to consider in connection witt 
 aerial insulators is the dielectric strength of the material 
 of which the insulator is made. 
 
 693. When an electric pressure is applied to an in- 
 sulating material or dielectric, a mechanical stress is set 
 up in the dielectric. Further, if the electric pressure is 
 increased beyond a certain limit (depending upon the
 
 WIRELESS TELEGRAPHY 
 
 thickness and nature of the material), the dielectric is 
 broken or punctured at its weakest point. If the di- 
 electric be a liquid or a gas, the puncture is only moment- 
 ary and heals up automatically as soon as the current 
 ceases to flow through the path thus made, but in the 
 case of solids the puncture remains, and the insulation is 
 permanently broken down at this point. The voltage 
 at which the puncture takes place for a given thickness of 
 material is called the dielectric strength of the insulating 
 material, and varies considerably with different materials. 
 694. The following table shows' the comparative 
 values of different substances in this respect. 
 
 Substance. 
 
 Voltage required to puncture 
 1 centimetre thickness of 
 material. 
 
 Air . . . . .' : 
 
 30,000 
 
 Oil . . . : . 
 
 60,000 to 80,000 
 
 Ebonite (best quality) . 
 
 600,000 
 
 Soft india-rubber 
 
 450,000 
 
 Mica . . . 
 
 1,000,000 
 
 Glass . . 
 
 250,000 
 
 Paraffin wax 
 
 170,000 
 
 Porcelain . . 
 
 100,000 
 
 The above figures are only approximate and vary 
 considerably with different samples of the same material. 
 In any case, in practice it is advisable to allow a factor of 
 safety of at least 3, usually more. 
 
 695. The second point to consider in connection with 
 aerial .insulators is the surface insulation. Even when 
 an insulator is perfectly dry, at high voltages electricity 
 will creep over the surface far more readily than it will 
 spark across an air-gap. Thus, if a pressure of 30,000 
 volts be applied across two metal discs separated by
 
 WIRELESS TELEGRAPHY 
 
 203 
 
 15 cms of an-, as shown in Fig 126, no discharge will 
 take place between them, but if the same space be filled 
 with ebonite as shown in Fig 127, although the ebonite 
 will not be punctured (see table of dielectric strengths 
 in paragraph 694), the electricity will run along the 
 surface of the ebonite between the two electrodes 
 
 FIG 126 
 
 696 In order to increase the length of the path along 
 the surface of the insulator without increasing the overall 
 length of the insulator, it is usual to make the surface 
 
 Fie; 127 
 
 corrugated, as shown in Figs. 128 and 129, thereby 
 doubling or trebling the length of the surface. 
 
 In the case of aerial insulators, however, the overall 
 length is not a matter of great importance, so that 
 corrugated insulators are not often used, the insulators 
 being made sufficiently long in themselves. 
 
 697 The actual length of surface insulation to allow is 
 a difficult matter to determine, as everything depends 
 upon the nature and condition of the surface. Where
 
 204 
 
 WIRELESS TELEGRAPHY 
 
 a dry, clean surface is assured at all times, it will be 
 quite safe to allow 4 cms of surface to every 30,000 volts 
 of potential. But aerial insulators are exposed to all 
 kinds of weather conditions, and if the surface of an 
 insulator be allowed to get coated .with a film of moisture 
 
 and dirt, almost any length will be useless for the purpose 
 of insulation. 
 
 698. In most cases dirt accumulates slowly, and 
 trouble from this source can be avoided by a periodic 
 inspection and cleaning of the insulators The chief diffi- 
 culty, therefore, is how to keep the surface of the insulator 
 dry. When the insulator occupies a more or less vertical 
 position this is easily accomplished by fitting a cone 
 over the insulator to act as a water-shed, as shown in
 
 WIRELESS TELEGRAPHY 
 
 205 
 
 Fig. 130 ; but when the insulator occupies a horizontal 
 position, as it might, for example, when supporting a 
 horizontal aerial, this method would obviously 
 be useless. In such cases it is usual to make 
 the insulator of ample length and to paint its 
 surface with a bitumastic varnish, so that 
 any moisture settling on it will detach itself 
 into separate drops instead of forming a con- 
 tinuous film of moisture over the whole of the 
 surface. 
 
 699. The surface of a porcelain insulator 
 has this property without being varnished, 
 but this material is, unfortunately, extremely 
 brittle and therefore unsuitable, at all events 
 for portable stations. 
 
 700. An important point to bear in mind when insulat- 
 
 FIG. 130. 
 
 FIG. 131. 
 
 ing an aerial is to use as few insulators in parallel as 
 possible, for each insulator thus used increases the total
 
 206 
 
 WIRELESS TELEGRAPHY 
 
 leakage from the aerial. Thus an aerial insulated as 
 illustrated in Fig. 131 will only have half the leakage as 
 
 FIG. 132 
 
 insulated as shown in Fig. 132. Moreover, only 
 half the number of insulators are required, and therefore 
 it is less costly. 
 
 EARTHS 
 
 701. A point of the utmost importance to the 
 efficiency of communication is a good " earth " at both 
 the transmitting and receiving stations. 
 
 " Earths " fall under one of two headings, namely, (1) 
 Direct Earths ; (2) Capacity Earths or " counterpoise." 
 
 We have already shown, in paragraph 651, that a 
 node of current exists at the free end of the aerial, and 
 that a maximum of current flows at the earthed end 
 of the aerial. This current must flow through the 
 earth connection from the aerial to the earth, and from
 
 WIRELESS TELEGRAPHY 207 
 
 the earth to the aerial. Therefore, any resistance in the 
 earth connection will cause loss of power and damping 
 of the oscillations. 
 
 702. In the case of ship stations the " earth " is a 
 simple matter, as salt water is an excellent conductor of 
 electricity, and it is therefore only necessary to attach 
 a conducting wire to the metal hull of the ship. In the 
 case of land stations, it is usual to bury a number of 
 plates of zinc, or other non-corrosive metal, at a sufficient 
 depth to ensure their being surrounded by damp earth. 
 
 703. It is found better to use a number of long strips 
 spreading radially under the soil than to use one large 
 plate, as the former gives less resistance than the latter. 
 The reason for this is that the effective resistance of a 
 buried " earth " is reduced in proportion to the capacity 
 of the earth plates. Thus, the greater the capacity of 
 the earth plates as a whole the less the effective resistance 
 of the " earth." The most efficient earth, where the 
 resistance of the soil is high, is made by burying 'a large 
 number of wires below the aerial or burying them 
 radially in all directions, like an underground umbrella 
 aerial. 
 
 704. It is not necessary, however, to have a direct 
 electrical connection with the earth: What are known 
 as " counterpoise " or " capacity earths " are frequently 
 used even with comparatively high-power stations. 
 Such an " earth " usually consists of a large number of 
 wires suspended above the surface of the ground and 
 carefully insulated from it. It is important in this 
 case that the capacity of the counterpoise should be at 
 least equal to that of the aerial. 
 
 705. For portable stations this form of " earth " is 
 frequently employed, as it takes less time to erect than
 
 208 WIRELESS TELEGRAPHY 
 
 it does to dig deep trenches. For such stations, perhaps 
 the most convenient form of earth is a number of long 
 narrow strips of wire-netting, which can be laid on the 
 ground star shaped, and can be rolled up into convenient 
 rolls for transport purposes. Under some conditions, 
 when the surface of the ground is wet and conducting, 
 the " earth " acts as a direct connection to earth, while 
 under other conditions, when the surface of the ground 
 is dry, the " earth " acts as a counterpoise. 
 
 706. This form of earth, although not as efficient 
 under some conditions as a true counterpoise, has the 
 great advantage of simplicity and ease of erection, and 
 moreover does not interfere with the approach to the 
 station. It will be readily understood that a number 
 of wires suspended a short distance from the earth over a 
 wide area round a station would be extremely incon- 
 venient in a military camp, where a stray horse might 
 easily become entangled with it at night, and perhaps do 
 serious damage both to itself and the station.
 
 INDEX 
 
 Accumulators, 41 
 Aerial insulators, 201 
 Aerials, distribution of current 
 in, 183-186 
 
 distribution of potential in, 
 177-183 
 
 effect of height of, 172 
 
 effective capacity of, 183*"* 
 
 excitation of, 84-96 
 
 functions of, 167 
 
 fundamental wave-length of, 
 170 
 
 harmonics in, 187 
 
 indirect excitation of, 93 
 
 insulation of, 199 
 
 length of, 175 
 
 shape of, 168 
 
 variation of wave-length of, 
 
 78-81 
 
 Aether, 48 
 Aether waves, 49 f 
 
 communication by, 51 
 Ampere, 17 
 Atmospherics, 165 
 Auto-jigger, 97 
 
 Battery, 40 
 Buzzer, tuning, 146 
 
 Capacities, in parallel, effect of, 82 
 
 in series, effect of, 79 
 Capacity, definition of. 9 
 
 inductive, 8 
 
 inductive, mechanical analogy 
 of, 8 
 
 Capacity, unit of, 21 
 Carborundum, rectifying pro- 
 perties of, 115, 138 
 Cell, electrolytic, 155 
 Circuits, electrical, 14 
 oscillatory, closed, 75 
 oscillatory, coupled, 89 
 oscillatory, coupled, mechani- 
 cal analogy of, 95 
 oscillatory, coupled, reaction of 
 
 secondary on primary, 98 
 oscillatory, coupled, resultant 
 
 wave-lengths of, 99 
 oscillatory, effect of resistance 
 
 in, 69 
 
 oscillatory, energy, 70-75 
 oscillatory, essential qualities 
 
 of, 68 
 oscillatory, mechanical analogy 
 
 of, 68 
 
 oscillatory, open, 76 
 oscillatory, power in, 75 
 oscillatory, variation of wave- 
 length of, 77-83 
 oscillatory, wave-length of, 
 
 69 
 
 receiver, proportions of, 127 
 Coil, induction, characteristic 
 oscillatory, voltage curve of, 
 39 
 
 induction, explanation of, 34-40 
 primary, definition of, 33 
 secondary, definition of, 33 
 Condenser, adjustable, construe- 
 tion of, 116 
 
 209
 
 21C 
 
 WIRELESS TELEGRAPHY 
 
 Condenser, energy stored up in, 75 
 
 mechanical analogy of, 9 
 Conductors, 5 
 Contents, vii, viii 
 Coulomb, 17 
 Coupling, calculation of, 109 
 
 effect of inductance on, 108 
 
 explanation of, 106 
 
 methods of varying, 110 
 Crystals, characteristic curve of, 
 131 
 
 rectifying properties of, 138 
 
 use of, 115 
 Current, electrical unit of, 17 
 
 oscillatory, detection of, 120 
 
 Damped waves, definition of, 60 
 Damping, definition of, 67 
 Detector, application of, in re- 
 ceivers, 120-121 
 
 crystal, use of potentiometer 
 for, 122 
 
 electrolytic, 151 
 
 electrolytic, circuits of, 157 
 
 in wavemeter, 114 
 
 magnetic, 159 
 
 Earths, 206 
 
 Electric waves, length of, 64 
 
 production of, 61^-64 
 Electricity, mechanical analogy 
 of, 14 
 
 production of, by chemical 
 action, 40 
 
 production of, by magnetism,29 
 Electro-dynamics, 12 
 
 definition of, 3 
 Electro-magnet, 28 
 Electro-magnetic field, 26 
 
 induction, 26-29 
 Electro -motive force, 12 
 Electro-static field, 7 
 
 induction, 6 
 Electro-statics, 3 
 Energy, electrical unit of, 22 
 
 in oscillatory circuits, 70-75 
 
 Energy, stored up in a condenser. 
 75 
 
 Farad, definition of, 21 
 Field, electro-static, 7 
 
 magnetic, 25 
 Force, electro-motive, 12 
 Frequency, in terms of wave 
 length and velocity, 47 
 
 of waves, 46 
 
 Harmonics in aerials, 187 
 Height waves, production of, 53 
 Henry, definition of, 18 
 High -resistance telephones, 135 
 
 Inductance, mutual, effect of in 
 coupled circuits, 104 
 
 unit of, 19 
 Induction coil, 34-40 
 
 electro -magnetic, 29 
 
 mutual, 32, 100 
 
 static, 6 
 
 Inductive capacity, 8 
 Insulation of mast stays, 197 
 Insulation, surface, 203 
 Insulators, 6 
 
 aerial, 201 
 
 corrugation of, 201 
 
 Jigger, 96 
 
 auto-, 97 
 Joule, 22 
 
 Length of wave, definition of, 46 
 in terms of frequency and 
 
 velocity, 47 
 Lines of Force, direction of, 25-26 
 
 Magnetic Detector, 159 
 
 Field, 25 
 Magnetism, 24 
 
 production of, by electricity, 27 
 Magneto-motive force, 28 
 Mast stays, insulation of, 197-199 
 
 strain on, 192
 
 WIRELESS TELEGRAPHY 
 
 211 
 
 Masts, buckling of, 194 
 
 staying of, 196 
 
 strain on, 190 
 Medium, definition of, 43 
 Microfarad, definition of, 21 
 Microhenry, definition of. 20 
 Morse Code, ix 
 Multiple Tuner, 164 
 Mutual Induction, 32 
 
 Inductance, effect of, in coupled 
 circuits, 104 
 
 Ohm, 18 
 
 Ohm's Law. 23 
 
 Oscillatory circuits, closed. 7f> 
 
 coupled, 89 
 
 coupled, mechanical analogy 
 of, 95 
 
 coupled, reaction of secondary 
 on primary, 98 
 
 effect of resistance m, 69 
 
 energy in. 70-75 
 
 essential qualities of. 68 
 
 mechanical analogy of, 68 
 
 open, 76 
 
 power in. 75 
 
 resultant wave lengths of, 
 99 
 
 variation of wave-length of. 
 77-83 
 
 wave-lengths of, 69 
 Oscillatory transformer. 96 
 
 Permeability. 28 
 Potentiometer, application of, 
 124 
 
 explanation of, 122 
 Power, unit of, 23 
 Pressure, electrical unit of. 17 
 Pressure waves, 48 
 
 mechanical analogy of, 56 
 
 production of, 54-64 
 Properties of waves, 43 
 
 Receivers, 118-164 
 circuits, proportions of, 127 
 
 Receivers, detectors of, 120 
 
 essentials of. 119 
 
 single circuit. 121 
 
 telephone, 133 
 
 telephone, sound produced in, 
 45 
 
 two-circuit, 125 
 
 tuning of, 144 
 Resistance, unit of. ' ? 
 Reluctance. 28 
 
 Spark-gap, function of, 87 
 
 Static induction, 6 
 
 Stays, mast, insulation of. 197 
 
 mast, strain on, 192 
 Symbols, xi, xn 
 
 Telephones, receiver. 134 
 receiver, high -resistance, 135 
 receiver, sound produced in, 
 
 143 
 Transmitters, coupled circuit, 
 
 93 
 
 plain aerial, 89 
 Tuning Buzzer, description of, 
 
 146 
 use of. 145. 
 
 Units, electrical, 16-23 
 
 Velocity of wave. 46 
 Volt. 17 
 
 Watt, 23 
 
 Wave, amplitude of, 46 
 frequency of. 46 
 Idngth of. 46 
 velocity of. 46 
 Wave-lengths, fundamental of 
 
 aerial, 169-172 
 
 relation to capacity and in- 
 ductance, 69 
 use of, in Wireless Telegraphy, 
 
 64 
 
 variation of, in aerial, 77 
 81
 
 212 
 
 WIRELESS TELEGRAPE[Y 
 
 Wave-lengths, variation of, In 
 closed oscillatory circuits, 
 82-84 
 Wavemeter, description of, 111- 
 
 114 
 Wave-motion, 43 
 
 communication by, 44 
 Waves, aether, effects produced 
 
 by, 49 
 
 continuous, definition of, 00 
 damped, definition of, 60 
 
 Waves, electric, length of, 64 
 electric, production of, 61-64 
 height, production of, 53 
 measurements of, 45 
 pressure, 48 
 pressure, laws of, 48 
 pressure, mechanical analogy 
 
 of, 56 
 pressure, production of, 64- 
 
 66 
 properties of, 43 
 
 THE END
 
 UC SOUTHERN REGIONAL LIBRARY FACILITY 
 
 STRATFORD & GREEN 
 
 Booksellers fc Slaters 
 44 South Main Street 
 f-os Angeles