Popular Science Monthly/Volume 3/September 1873/Magneto-Electric Illumination
By WILLIAM CROOKES, F. R. S.
THE progress made in electric illumination during its advance toward perfection has been several times recorded in the pages of this journal. In our first number, published nearly ten years ago, Dr. J. H. Gladstone gave a history of the early difficulties attending the introduction of the magneto-electric machine as a light-generator for light-house illumination. Two years subsequently, the present writer described Wilde's magneto-electric machine, and, after a further lapse of years, during which time no very important improvement in the industrial application of magneto-electricity has been recorded, another step in advance has been made which calls for detailed notice.
The chief difficulties in the employment of magneto-electric currents for industrial purposes have been their almost instantaneous character and the rapid alternation in their direction. The instrumental means necessary to seize hold of these rapidly-alternating waves, and convert them into a more or less continuous stream of force flowing in one direction, are necessarily of a delicate character, and are easily put out of adjustment. This is easily understood when it is remembered that, in the machine first tried by Mr. Holmes, the rubbing surfaces were worn away in ten or twenty minutes. The Berlioz machine required for its maximum of intensity 350 or 400 revolutions per minute, and the direction of the current is then reversed nearly 6,000 times per minute; here, however, the alternate currents are not brought into one. In the machine made by Mr. Wilde for the Commissioners of Northern Light-houses, the first armature is made to revolve about 2,500 times a minute, generating 5,000 waves of electricity. These alternate currents are converted into an intermittent current moving in one direction only by means of a commutator. The second armature revolves 1,800 times a minute, generating 3,600 alternately-opposed waves of electric force, which are picked up and sent in one direction by a commutator, as in the former case.
It is evident that when a good friction contact is to be kept between pieces of metal moving at these enormous velocities, the wear and tear is very great. For a long time, however, it was thought that these difficulties were inherent to the magneto-electric machine, until electricians found, first, that the almost instantaneous flash of the current could be considerably lengthened out, and then that the successive waves generated could be so produced as to flow in the same instead of in opposite directions.
These important desiderata are supplied in a magneto-electric machine of a novel form, invented by M. Gramme. The principle is not difficult to understand. Take a long bar of soft iron, E, E', Fig. 1,
round which is coiled an insulated copper wire; to this bar, forming an electromagnet, let a permanent magnet, S N, be presented, the south pole being nearest to the iron bar. Now move the permanent magnet in the direction of the arrow parallel with itself, with a uniform velocity, and always maintaining the same distance from the bar. The south pole of the permanent magnet will produce a north magnetic pole in the portion of the iron bar nearest to it; and the gradual displacement of this pole from one end to the other of the iron bar, caused by the motion of the magnet, will induce in the surrounding wire an electric current which may be rendered evident by the galvanometer, G. This current will not be instantaneous: it will continue to flow during the whole time the magnet is moving between the two ends, E E', of the iron bar, and its time of duration may therefore be varied at pleasure.
This experiment shows that it may be possible, by proper arrangements, to realize a machine which will furnish a continuous current of electricity for as long as may be desired. We have only to imagine the electro-magnet, instead of being the straight bar shown in Fig. 1, bent into a circular form as at E, E', E", E'", Fig. 2.
Submit this annular electro-magnet simultaneously to the influence of the two poles of the permanent horseshoe magnet, N S, and at the same time imagine it to revolve on its axis in the direction shown by the arrows.
The south pole, S, of the horseshoe magnet, will produce in that portion of the ring, E, which is near it, an electric current in a particular direction, as may be inferred from what we have said respecting the straight bar, Fig. 1. But the north pole, N, of the magnet, will likewise produce in the part of the ring which is in its neighborhood, E", an electric current flowing in the opposite direction; and it is easily conceived that, in the two portions of the ring, E', and E'", which are in what may be called the mean position, there is no current at all. If, therefore, we wish to collect the two contrary currents produced simultaneously in the wire surrounding the electro-magnet, we have only to connect the wires at the mean position to two conductors by friction-contacts, F F', when the current can be carried away to a galvanometer, G, and rendered sensible.
The principle of the arrangement being thus understood, the construction of the machine itself will be readily intelligible.
It consists of a permanent horseshoe magnet, S, O, N, Fig. 3, between the poles of which revolves an electro-magnet. This electro-magnet consists of a ring of soft iron, round which is wound an insulated conducting wire, presenting no solution of continuity. It may be conceived as being an ordinary straight electro-magnet bent round in a circle, and the two ends of the conducting wire soldered together to establish continuity.
In Figs. 4 and V the electro-magnet is represented at A in section, while in Figs. 3 and 5 it is shown at A with the covering wire on it. It revolves round its axis on an axle to which movement is communicated either by means of belting, or with toothed gearing, shown in Figs. 3 and 4, worked by a handle, M.
The current is generated and collected in the following way: The wire surrounding the electro-magnet is, as we have said, continuous, but it is disposed in 40 sections or elements, each consisting, say, of 100 turns. The outer end of the coil of one section forms the commencement of the first coil of the next section, and so on. The whole of the wire is therefore divided into 40 equal sections, being, however, continuous throughout.
To understand better how an uninterrupted current is produced, let us imagine a line to be drawn equatorially, or perpendicular to the lines of force between the poles of the horseshoe magnet, and dividing the ring armature into two parts; suppose, likewise, that to the two ends of one of the 40 coils two wires are soldered, the other ends of which are attached to a galvanometer. Now let the ring be intermittently revolved in one direction, so as to give to the said coil a
succession of movements of about 10 degrees, stopping each time to permit the galvanometer-needle to resume its normal position. It will then be seen that the whole time the coil is above the equatorial line the galvanometer-needle will be urged in the same direction, and the currents may be called positive. But, as soon as the said coil crosses the equatorial position, the currents generated in it will be negative, and in the opposite direction to what they were at the other half of the circle. This experiment shows that a reversal of the direction of movement carries with it a reversal of the direction of the current.
From this insight into what is produced in one of the sections, the general phenomena produced by the whole circle of coils are easily understood. The 20 sections which are on one side of the equatorial position are the source of positive currents; these may be of unequal intensity among themselves, but, for a uniform velocity of rotation, their sum is evidently constant; for, as one coil crosses the equatorial line from north to south, an opposite one comes up from south to north to take its place. On the other hand, the 20 sections which are on the other side of the equatorial line are the seat of negative currents, the sum of whose intensities is likewise constant, and equal to that of the positive currents. Thus the revolving armature presents two groups of coils, generating two equal but opposite streams of electric force. The wire being unbroken, the currents neutralize each other, and there is no circulation. The result may be likened to what would be produced by taking two batteries, each of 20 cells, and connecting them in opposition by joining similar poles.
The problem now is to pick up these dormant currents and utilize their force. Its solution is apparent from the comparison we have just made. To collect the electric current from two batteries which are connected together in opposition, it is only necessary to fasten conducting wires to the two points of contact of similar poles, when the whole force of the batteries will flow along these wires. They were hitherto opposed, they now flow together, quantity-wise. M. Gramme, in the second portion of his invention, has adopted this artifice in an ingenious manner.
The various sections of the continuous electro-magnet are connected with radial pieces of copper shown at R in Figs. 3, 4, and 7, insulated one from the other, but coming very close. The termination of one coil of wire and the commencement of the adjacent coil are soldered to the same radial connector, of which, therefore, there are as many as there are coils. These radial connectors, on approaching the centre, are bent at right angles, as shown at R, Figs. 4 and 7, and pass through to the other side, where their ends form an inner concentric circle, being still insulated one from the other.
The friction-pieces F (Figs. 4, 5, and 6), consisting of disks of copper, are pressed, by means of springs shown at r (Figs. 5 and 6), against the circle formed by the extremities of the conducting radii R, at two points which are accurately in the equatorial line; that is to say, at the place where the equal and opposed currents generated in the upper and lower halves of the ring neutralize each other. Consequently the currents are collected and flow together along conducting wires, which are fastened to the friction-pieces F.
The perfect continuity of the current so obtained is secured by causing the friction-pieces F to touch simultaneously several of the radial conductors R; consequently the metallic circuit is never broken.
The effects produced by these machines vary with the rapidity of rotation. Experience shows that the electro-motive force is sensibly in proportion to the velocity; but it is probable that this force tends toward a limit, corresponding to a particular velocity, beyond which the electromotive force would remain constant, or even diminish. Moreover, the electromotive force is greater in proportion to the number
of coils encircling the iron ring, but the relation between these two quantities has not yet been determined. The theoretical resistance of the machine should be one-fourth of the whole resistance of the wire wound round the ring armature; but the actual resistance is not so great, since each friction-disk always touches several radii, R, and the resistance of the coils thus embraced by the friction-disk has to be subtracted from the resistance of the circuit.
The possibility of augmenting the strength of the current by increasing the dimensions of the machine is too obvious to need more than a passing allusion. The effects may also be increased by connecting together several such machines, as galvanic piles are connected, either for intensity or quantity. The quality of the current likewise differs according to the kind of wire surrounding the armature, a short thick wire producing effects of quantity, and a long thin wire, of intensity. It is also easy to see that two horseshoe magnets, instead of one, may be made to act on one ring armature; that is to say, it may be actuated by four poles instead of two, or even by a greater number; always having a friction-disk between each pair of poles. Moreover, the permanent horseshoe magnet may be replaced by electro-magnets, which can be excited by a portion of the current derived from the machine itself, according to the now well-known method. At the beginning of rotation the residual magnetism of these electro-magnets will induce a feeble current in the ring; one-half of this passes round the electro-magnets, the four poles of which react on the armature. Of the four friction-pieces, two carry half the current to excite the electro-magnets, and the machine rapidly attains the maximum effect. From conducting wires attached to the other two friction-pieces a powerful current is available.
A machine of this kind, containing two horseshoe electro-magnets, one for exciting and the other for the exterior current, and having round each pole 7 kilos, of copper wire 3 m.m. diameter, when worked by hand, decomposes water, and fuses 26 centims. of iron wire 9-10ths m.m. in diameter. However slowly the armature is rotated, the needle of a large galvanometer having the wire only once round is deflected, and the effects increase in intensity as the velocity of rotation increases, up to a maximum of 700 or 800 turns a minute, a velocity which is easily obtained when steam is employed.
Such a machine, giving an absolute continuous current of electric force by the mere turning of a wheel, is of value outside the physical laboratory. It is available—(1) for medical purposes; (2) for telegraphy; (3) for electro-plating, gilding, etc.; (4) for military purposes, signalling, explosions, etc.; (5) for chemical decompositions; and (6) for electric illumination.
A large machine, which has lately been exhibited in London, driven by a 2½-horse-power engine, produced a light equal to 8,000 candles; a copper wire about 1¼ m.m. in thickness, suspended between the poles, became instantly red-hot with a revolution of little over 300 in a minute. Larger machines are being made that will probably give a light equal to 25,000 candles.
This machine has lately been examined by the French Société ď Encouragement, and, in accordance with therecommendation of the reporter, Count du Moncel, a prize of 3,000 francs has been awarded for it to M. Gramme; while the manager of the "Alliance Company," M. Joseph Van Malderen, who superintended its manufacture, has had awarded to him a gold medal. In his report, Count du Moncel says that a machine 1.25 metre in height, 0.8 metre long, and the same in width, driven by a 4-horse engine, gave a light equal to 900 carcel-lamps. It also heated to redness two juxtaposed copper wires 12 metres long and 0.7 m.m. diameter, and fused an iron wire 2.5 metres long and 1.3 m.m. thick.
The constancy of direction of the electric current generated by this machine is, however, not of so great an importance for the electric light as for other purposes for which it may be used. Indeed, the electric light is by many electricians thought to be superior when produced by a magneto-electric machine of the old form without any commutator. The alternate reversal of the currents of electricity produces no flickering or irregularity in the arc of light, as they occur far too quickly to be appreciated by the eye, while the rapid reversal of the direction causes the carbons to wear away with great regularity, thus enabling the point of light to be kept more easily in the focus.
For the electro-deposition of metals—copper, silver, etc.—constancy of direction of current is indispensable, and here the experiments show a marked superiority of the Gramme machine over other magneto-electric machines.
In the galvanoplastic works of M. Christofle, of Paris, where experiments have been going on for more than a year, it is found that the best machine hitherto known, when moved with a velocity of 2,400 revolutions per minute, only deposits 170 grammes of silver per hour; while a smaller Gramme machine moved with a velocity of 300 revolutions per minute deposits 200 grammes of silver per hour; the temperature of the annular armature not exceeding 50° C, with a velocity of 275 revolutions, no elevation of temperature is experienced. It will be easily comprehended how strongly this result, obtained with a speed of rotation eight times less than hitherto required, speaks in favor of M. Gramme's invention. Usually at M. Christofle's the circuits are arranged to deposit 600 grammes of silver per hour, and the manager of the factory finds that the deposition with this machine takes place with a regularity and constancy which leaves nothing to be desired, and which cannot be obtained by using any other source of electricity.
Recently, the electric light generated by a Gramme machine has been exhibited on the Victoria Tower of the Houses of Parliament. The machine is placed in the vaults of the House of Commons, near to the boilers, and is worked by a small engine, which was already there, and was convenient for the purpose. From the machine two copper wires, half an inch diameter, are led along the vaults to the base of the clock-tower, and thence upward to the signalling-point, a total length of nearly 900 feet, being about three times the distance that an electric current has ever before been conducted for a similar purpose. The signalling apparatus is placed in a lantern five feet high, four feet wide, and having a semicircular glazed front, which projects from the lantern of the belfry on the north side of the tower, or that overlooking the Victoria Embankment. It consists—first, of a fixed table, in which is inserted a flat brass ring 16 inches diameter and one inch broad, which serves as a roller-path for the apparatus carrying the lamp and reflector; next, there is a circular revolving table, having bearings on the roller-path, and which is moved around a central pivot projecting from the fixed table, being actuated by a worm wheel and screw. By means of this arrangement the light can be directed horizontally from side to side through an arc of 180°. It could, of course, be made to sweep the whole of the horizon, but the position of the lantern with regard to the clock-tower is such as to enable the light to be seen through the range of a semicircle only. Upon the revolving table, and hinged to it at the front, is the elevating table, which has a screw adjustment to the rear by which the light can be raised or depressed, being capable of vertical training through an arc of 25°. On the elevator is placed the lamp-table, upon which again is a sliding platform, on which the lamps themselves stand. There are two lamps, which are in use alternately, the carbon-points lasting but four hours, while the House frequently sits for ten.
The copper conductors terminate at the fixed part of the machine, and the method of carrying the current from them to the lamps is very ingenious, the moving parts of the apparatus forming in themselves conductors. The negative conductor is placed in metallic contact with one hinge of the elevator-table through the centre-pin on which the table revolves, and the positive conductor with the other hinge by means of the brass roller-path. The currents from those points are conducted to the lamp-table, and thence through the traversing platform to the lamps, metallic contact being obtained throughout the whole circuit by means of flat springs moving over flat surfaces. The changing of the lamps is effected, without any appreciable break of continuity in the light, by means of the traversing platform on which they stand, and which has a sliding motion from side to side. When the carbon-points in one lamp are nearly consumed, the traverser is quickly shifted from right to left, or vice versa, as may be necessary. The break of contact is but momentary, and only exists during the time required to move the traverser rapidly through a space of six inches. The light will not become extinct during that period, as there is not sufficient time to allow the incandescence of the carbon to entirely subside. The springs under the lamp thrown out of use are by this action removed from the metal plate in the lamp-table, and the springs under the fresh lamp are brought into contact, and the light is at once produced anew.
The intensifying apparatus at present in use is a holophole lent by Messrs. Chance, and through which the rays are sent in parallel lines. It is 21 inches in diameter, and is composed of lenses, surrounded by annular prisms, the centre part refracting the rays and the outer rings reflecting them. Should the electric light be adopted, a special lens will be constructed, by means of which the rays will be diffused through an arc of 180°, instead of being sent in one direction only. The cost of this electric light is at present estimated at 10d. per hour.
It may be of interest if we consider some matters of scientific interest in connection with this machine. In the first place, it possesses an enormous advantage over the voltaic battery in the absolute constancy of the current so long as the velocity of rotation is uniform. In an experiment carried on for eight hours with one of the first machines constructed, the deviation of the needle of a galvanometer was absolutely invariable. Again, a voltaic battery is a complicated piece of apparatus; for each element consists of four separate solid pieces (the outer cell, the porous cell, the positive and the negative element) and two liquids, while in most experiments a considerable number of batteries is required. From this multiplicity of parts a voltaic battery is subject to many accidental derangements, which are likely to weaken if not destroy its power. With the magneto-electric machine there is no complication. All the parts are solidly connected together, and no special care is required.
It must also be remembered that a powerful voltaic battery costs almost as much when it is at rest as when in action. The magneto-electric machine, on the contrary, costs nothing when it is not producing an external current. This may be understood in two senses. It is, of course, evident that, when no current is required, the rotation of the machine may be stopped; but it is a remarkable fact that, even when rotation of the armature is still going on, no mechanical force is expended except that necessary to overcome friction, provided the exterior current does not flow. To understand this, let us examine a little more closely into the working of the machine. In the first place, suppose the machine to be in rapid movement, and furnishing a current in an exterior circuit, it will be observed that the armature does not get hot; from this it may be concluded that all the mechanical force transmitted to the machine is converted into electricity, since none is changed to heat. In the next place, the machine continuing to revolve with the same speed, suppose the exterior circuit to be broken; still the machine does not rise in temperature, showing; that in this case there is neither production of heat nor electricity, and consequently no waste of mechanical force. From the way in which the currents in the armature are generated, when there is no exterior circuit along which they can flow, they neutralize one another, and keep in such perfect equilibrium that there is absolutely no circulation, and consequently no heating.
If the Gramme machine is set in motion by a force just sufficient to turn it with a definite velocity when the exterior current is flowing, and, if the outer circuit is suddenly broken, the machine is seen to acquire an increasing velocity, showing that the mechanical force applied to it, being no longer capable of going off as electricity, spends itself then in augmenting the velocity of the moving parts of the machine.
On the other hand, if the machine is kept at a certain speed of revolution while the outer circuit is broken, and the circuit is then suddenly closed, the speed instantly diminishes, showing that a portion of the force turning the machine changes into electricity.
These experiments show that, whether the machine be active or passive, there exists always a state of equilibrium between the expenditure of mechanical force and the production of electricity.—Quarterly Journal of Science.