INDUCTION COIL, an electrical instrument for producing high electromotive force, consisting of two coils of wire wound one over the other on a core formed of a bundle of thin iron wires or a number of thin iron sheets. An induction coil is thus essen tially an electromagnet with two windings called the primary and secondary coils. The action of induction coils depends upon the effect known as electromagnetic induction, discovered by Faraday in 1831. (See ELECTRICITY.) If an electric current in the primary coil is caused to vary rapidly, a current is induced in the secondary which, flowing to its terminals, produces temporarily an accumula tion of electricity on and near them and a high potential difference between them. The variation of the primary current is usually effected by "interrupting" this current, i.e., by breaking the con nection between the primary coil and the source of current supply. In some applications of induction coils, however, the primary coil is supplied with alternating current, and the action is then similar to that of a transformer (q.v.). Induction coils are largely used in X-ray work in which very high voltage is required to produce the discharge through the tube in which the X-rays are generated. A few years ago, in the early days of wireless telegraphy, induction coils were much used for actuating the transmitting apparatus. They are still employed in connection with the production of high frequency discharges for medical purposes. Small induction coils are used in very large numbers for producing the sparks required for ignition in internal combustion engines.
The history of the induction coil begins in Nov. 1831, when Faraday described an experiment (Phil. Trans. R.S., 183 a ; and Experimental Researches, vol. i.) in which two coils of wire were wound on an iron ring and insulated from each other. Faraday observed that if the terminals of one of the coils were brought near each other, short sparks appeared between them when a current was started or stopped in the other coil. Soon afterwards other experimenters began to develop this method of producing high potential effects, among them Joseph Henry, C. G. Page, N. J. Callan, and William Sturgeon. It was soon found that the effects could be obtained with coils wound on a straight iron core, and that a core made up of a bundle of thin iron wire was more effective than one consisting of a single bar of iron. Before 184o the automatic mercury, or platinum contact, interrupter, actuated by the intermittent magnetization of the core, had come into use. No further marked improvement was effected till 1853, when Fizeau added the condenser, connected to the terminals of the interrupter, which greatly increased the length of spark obtainable with induction coils. About the same time Ruhmkorff, in Paris, introduced several improvements, especially in connection with the insulation of the secondary wire. He also increased the number of turns in the secondary coil, and wound it on a glass tube to improve its insulation from the primary coil. Ruhmkorff also adopted in his later coils the method (previously suggested by E. and C. Bright) of winding the secondary coil in numerous flat sections instead of arranging it in layers extending over the whole length of the coil, the object being to keep as far apart as possible windings which are at a great difference of potential.
During the period 1867-190o several very large induction coils were made by different manufacturers for the purpose of produc ing sparks of great length. One of these, made by A. Apps, for the old Polytechnic Institution of London, had a core of iron wire 5 ft. long, and a secondary wire 15o m. in length. This coil gave sparks 3o in. long. A still larger coil, made by Apps for Spottis woode, and now preserved at the Royal Institution, London, had a secondary coil of 341,85o turns, the total length of the wire being 28o miles. With a battery of 3o Grove cells, sparks 42 in. long could be obtained with this coil. Very large coils have also been made by Klingelfuss, of Basle, who employed an improved method of sectional winding of the secondary coil. The wire is wound in one continuous length and, by the use of separating discs of stepped form, the thickness of the insulation between neighbouring turns of consecutive sections is graded in accordance with their difference of potential. Induction coils made by this firm are notable for their comparatively small number of secondary turns, one of them, with only 86,000 turns, producing sparks i metre long. A coil made by the same firm for the Paris Exhibition of 190o gave sparks 15o cm. (59 in.) long. During this period induction coils came much into use in experimental work, espec ially in connection with the discharge of electricity through gases, and with the production of electrical oscillations, and they were instrumental in such important discoveries as the Hertzian waves, the cathode rays, and the Röntgen rays.
In large modern induction coils, such as those used for the production of X-rays, the cores are usually built of long strips of thin transformer sheet insulated from one another, the length of the core being i o to 15 times its diameter. On the core, and well insulated from it, is wound the primary coil, the thickness of the copper wire depending upon the current to be used in operating the coil. The primary coil is sometimes arranged in three or four sections which, by means of a commutator, can be connected in series or in parallel. The core and primary coil are placed within a thick-walled tube of micanite or other material of great dielec tric strength, upon which is placed the secondary coil of no. 36 or no. 32 silk-covered copper wire. There is considerable diversity in the method of winding the secondary coil. Some makers arrange the coil in thin flat sections only one wire thick, separated from one another by discs of insulating material. Others prefer to wind the coil in sections, three or four wires thick. In either case the separating discs should be wide enough to extend well beyond the inner and outer boundaries of the sections, so that the wire can be well surrounded by insulating material, and the whole should be thoroughly impregnated with insulating wax.
Several modern coils are arranged in only two or four sections, each section wound in layers, on the grounds that with this con struction the windings are less likely to become displaced, and that the turns having the highest potential can be kept furthest away from the primary coil. As an alternative to impregnation with wax the plan of immersing the coil in insulating oil contained in a metal or porcelain vessel has come into use in recent years. Such oil-immersed coils are usually placed with their axes vertical. The dry type of coil is usually covered with a sheet of ebonite and mounted on a base board, or is cast in solid wax in a wooden case.
Improved forms of hammer break, with platinum contact pieces, are still used with small coils, but with large coils a motor driven mercury interrupter is much more suitable. The most effective form of mercury interrupter is that known as the "tur bine," or "jet" type, in which a pair of revolving jets of mercury, pumped up by the rotation from the bottom of an iron container, make contact during their rotation with two or more metal plates. The jets are diametrically opposite to each other, and the two plates with which they simultaneously make contact are connected to the source of current supply. A diagram of the circuits of an induction coil is shown in fig. i.
In early forms of mercury jet interrupters the jet and plate were immersed in oil or alcohol, but coal gas is now generally used as the insulating medium in which the interruption takes place. As many as 15o breaks per second may be produced with this particular type of ter. Sometimes the motor of a mercury jet interrupter is also used for driving a rectifier, or rotating commutator, connected in the secondary circuit for the purpose of eliminating the verse potential induced in the secondary when the primary cuit is "made." This reverse tential at "make" is usually very small in comparison with the direct potential at "break," but when the coil is operated by a battery or other source of high voltage, it may be sufficient to produce some discharge the wrong way through an X-ray tube.
If very high rates of interruption are required an electrolytic interrupter, devised by Wehnelt in 1899, may be employed. This consists of a plate of lead immersed in dilute sulphuric acid, and a: platinum wire protruding a short distance from a porcelain tube. When this is connected in series with the primary coil, with the platinum wire to the positive pole of the battery, the current becomes rapidly intermittent owing to the successive formation of gas bubbles, the frequency being sometimes over i,000 per second.
A very important application of the principle of the induction coil is found in the high tension magneto, which consists essentially of a very small induction coil mounted on an armature rotated in the field of a permanent magnet, the current in the primary coil being generated by the rotation, and being interrupted by a contact breaker at a suitable point in each revolution. Another form of induction coil, usually called an oscillation transformer, is the Tesla coil, which is used for generating currents of high fre quency and high voltage. It consists of a primary coil, having a few turns of thick copper wire, connected in series with a Leyden jar and a spark-gap, and a secondary coil having a large number of turns of finer wire. A succession of sparks is produced at the spark-gap by an ordinary induction coil, each spark giving rise to a group of high frequency oscillations which produce remarkable brush discharge and other effects in and near the terminals of the secondary coil. Similar effects are produced with the auto-trans f ormer, a single coil of bare copper wire wound on an ebonite frame and having a few turns at one end connected to the Leyden jar and spark-gap so as to form the primary circuit. The re mainder of the coil forms the secondary winding. In this arrange ment the primary and secondary influence each other not only by magnetic action but also in consequence of the electrical con nection between them.
Theory of the Induction Coil.—The theory is much simpli fied by the fact that the iron core usually forms a very "open" magnetic circuit, so that the magnetic flux in it is, over a wide range, approximately proportional to the current flowing in either of the coils. The self inductances, which we will denote by L, and may therefore be treated as constant. The mutual inductance may also be regarded as constant but, owing to the fact that the secondary current is not distributed uniformly along the secondary wire, the coefficient of inductance of the primary on the secondary is greater than that of the secondary (for unit current in its central winding) on the primary. These two coefficients will be denoted by L. and If the capacities in the primary and secondary circuits are C,, C2, the relation between the potentials Vi and in the two circuits is expressed by the equations where R,, are the resistances of the circuits and E is the battery electromotive force. If the resistances are neglected these two equations can be easily solved, and the result shows that the wave of potential in each circuit, after the interruption of the primary current, consists of two simple harmonic components dif fering in amplitude and frequency. Each of the two frequencies of the system depends upon the inductances and capacities of both circuits.
If the frequencies are represented by a,, being the greater) , the expression for the potential at the terminals of the secondary coil at any time t after the interruption of the primary current is This expression shows that the two components of the potential wave in the secondary circuit be gin from zero in the opposite phase, and that their amplitudes are inversely proportional to their frequencies. In fig. 2 are shown the potential waves for four different values of the fre quency ratio The upper curves show the two components separately, the lower curves the result of their super position, i.e., the actual wave of potential in the secondary circuit. It will be seen that when the frequency ratio is 3 or 7, maxima of the two component waves occur simultaneously, giving a specially high voltage at this mo ment. This also happens when has one of the values i r, 15, etc. When the frequency ratio is 5 or 9, a minimum of the more rapid component coincides with a maximum of the slower, result ing in a reduced value of the maximum voltage. In these latter cases there are, in fact, two equal peaks in the potential wave, one occurring before, the other after, the maximum of the slower com ponent. The curves in fig. 2 cover only one half-period of the slower component ; their continuation in the second half-period is a repetition with the ordinates changed in sign. Curves similar to the lower curves of fig. 2 can be obtained experimentally by the use of a suitable oscillograph. An example is given in fig. 3, a photograph of the wave of potential at the secondary terminals of an induction coil as indicated by an electrostatic oscillograph con nected directly with them.
The instrument was of the kind in which the deflection is pro portional to the square of the potential, consequently the deflections are in the same direction for both positive and negative parts of the wave. The form of the first half-wave of the curve shows that the ratio of the two frequencies was very nearly 7, as in the third of the lower curves in fig. 2. The wavy line below the oscillograph curve in fig. 3 represents the oscillations of a tuning fork, each wave in this line corresponding to second.
The foregoing expression for V2 enables us to calculate the effect of varying one or other of the "constants" of the circuits, for example, the capacity of the con denser connected to the interrupter. The result of such a calculation is shown in the full-line curve of fig. 4, in which the abscissa is proportional to the primary capacity—it represents, in fact, the ratio and the ordinate is proportional to the maximum secondary potential. In this example the coupling (defined as is It will be seen that the curve of secondary potential consists of portions of a series of arches, all lying within and all except one touching the broken line curve which represents the sum of the amplitudes of the two potential oscillations in the secondary cir cuit. As the primary capacity is increased from zero, the frequency ratio diminishes from infinity, and, at the four points of contact of the full-line and broken-line curves in fig. 4, has the values 19, 15, i r and 7. At the points of intersection of the arches the frequency ratios are i 7, 13, 9 and 5. One of the points of contact occurs at the summit of the broken line curve, and it therefore represents a very favourable adjustment of the circuits for producing high secondary voltage. Not only is the potential equal to the sum of the amplitudes of the two com ponents, but the sum of the am plitudes also has its maximum value in this adjustment. These favourable conditions do not oc cur at all values of the coupling, but only at certain values, four of which are 0•571, 0.835, 0.902 and 0.931. If the coupling has one of these values, the optimum primary capacity is that which makes L, C1= (r — and the maximum secondary voltage is then ioVL2i/L,2VL,/C2. The form of the curve showing the relation between primary capac ity and secondary potential de pends upon the coupling. In fig. 5 is shown a curve obtained by experiment with a coil the coupling of which was 0.767.
The ordinate of this curve represents the reciprocal of the least primary current the interruption of which causes a spark to appear at a spark-gap connected with the secondary terminals, and it is therefore proportional to the maximum secondary potential for a given primary current. The extreme arch on the right of this curve is more prominent than that of fig. 4, and at values of the coupling below 0.71 the right-hand arch is higher than all the others. It follows that as the coupling is reduced the frequency ratio neces sary for maximum secondary potential also becomes smaller, e.g., when the coupling is 0.57r, the primary capacity should be so adjusted that the frequency ratio is 3.
Damping of the Oscillations.—In this sketch of the theory we have neglected the resistances of the circuits and other causes of dissipation of energy. In practice, owing to the resistances and the leakage and core losses (see ELECTROMAGNET), the oscilla tions are subject to decay factors; they die away just as do the vibrations of a tuning fork after it is struck. This damping of the oscillations is clearly shown in fig. 3. In a good modern coil the maximum secondary voltage may be reduced by over 2 5 % by the losses, which are, however, not sufficient to affect to any great extent the frequencies of the oscillations or the conditions in which maximum voltage is produced.
When a spark or other form of discharge is allowed to pass between the secondary terminals, the theory hitherto described, with allowance for the damping of the oscillations, is applicable up to the moment at which the discharge begins. The ordinary spark discharge of an induction coil, in air at atmospheric pressure, generally consists of an initial true spark followed by an arc. In the initial spark, which represents the discharge of the electricity accumulated on the secondary coil, and which probably contains a train of high frequency oscillations, the potential falls very rapidly to a small value, and thereafter the secondary current flows as an arc in the conducting path prepared for it by the initial spark. In the arc the current usually pulsates with a period de pendent upon the constants of the primary circuit and the coupling, the approximate expression for the period being The total quantity of electricity passing in the ordinary spark discharge depends upon the length of the spark and upon the cur rent interrupted in the primary circuit. It increases with the cur rent, though not in proportion to it, and it diminishes as the length of the spark is increased. When the discharge takes place through a "soft" X-ray tube the secondary voltage again follows the course indicated by the foregoing theory up to the moment at which the discharge begins. During the discharge the potential falls, not with great rapidity as in the spark discharge, but much more gradually, and with fluctuation, to a smaller value at which the discharge ceases. After this the system continues to oscillate, with what energy it has left, in the two frequencies which it possesses when the secondary circuit is open. In the case of discharge through an ordinary high resistance, the discharge current consists of two damped oscillatory components if the resistance is suffi ciently high ; but if the resistance is below a certain value, one of the oscillations becomes replaced by aperiodic components.
Theory of the Tesla Coil.—This is somewhat similar to that of the induction coil, the chief difference arising from the fact that the oscillations are started in a different way. The potential in the secondary of a Tesla coil is the result of the superposition of two oscillations differing in frequency but having the same ampli tude. The two oscillations begin in opposite phase but not from zero, i.e., they begin at their maximum positive and negative values. In order to produce the highest potential at the secondary terminals of a given Tesla coil, operated by sparking across a given gap, the Leyden jar should have a capacity considerably greater than that which satisfies the resonance condition i.e., the condition which makes the periods of oscillation of the primary and secondary circuits equal when they are separated. In the ordinary induction coil worked by an interrupter the primary capacity which gives the highest secondary voltage is, as we have seen, much smaller than the value which makes equal to BIBLIOGRAPHY.-J. A. Fleming, "History of the Induction Coil," Bibliography.-J. A. Fleming, "History of the Induction Coil," The Alternate Current Transformer (19oo) and "Construction of Induction Coils and Theory of Coupled Circuits," Principles of Electric Wave Telegraphy (1919) ; Lord Rayleigh, "On the Induction Coil," Phil. Mag. (19o1) ; P. Drude, "Theory of the Tesla Coil," Ann. d. Phys. (1904) ; E. Ruhmer, Funkeninduktoren (19o4) ; H. Armagnat, La Bobine d'Induction (19o5) ; E. Taylor Jones, The Theory of the Induction Coil (1921). (E. T. J.)