LIGHTING, ELECTRIC. Artificial light is generally produced by raising sonic body to a high temperature. If the temperature of a solid body be greater than that of surrounding bodies it parts with some of its energy in the form of radiation. Whilst the temperature is low these radiations are not of a kind to which the eye is sensitive ; they are exclusively radiations less refrangible and of greater wavelength than red light, and may be called infra-red. As the temperature is increased the infra-red radiations increase, but presently there are added radiations which the eye perceives as red light. As the temperature is further increased, the red light increases, and yellow, green, and blue rays are successively thrown off. On pushing the temperature to a still higher point, radiations of a wave-length shorter even than violet light are produced, to which the eye is insensitive, but which act strongly on certain chemical substances ; these may be called ultra-violet rays. It is thus seen that a very hot body in general throws out rays of various wavelength, our eyes, it so happens, being only sensitive to certain of these, viz., those not very long and not very short, and that the hotter the body the more of every kind of radiation will it throw out, but the proportion of short waves to long waves becomes vastly greater as the temperature is increased. The problem of the artificial production of light with economy of energy is the same as that of raising some body to such a temperature that it shall give as large a proportion as possible of those rays which the eye happens to be capable of feeling. For practical purposes this temperature is the highest temperature we can produce. As an illustration of the luminous effect of the high temperature produced by converting other forms of energy into heat within a small space, consider the following statements. 120 cubic feet of 15 candle gas will, if burned in ordinary gas burners, give a light of 360 standard candles for one hour. The heat produced by the combustion is equivalent to about GO million foot-pounds. If this gas be burned in a gas-engine, about 8 million foot-pounds of useful work will be done outside the engine, or four horse-power for one hour. This is sufficient to drive an "A" Gramme machine for one hour ; the energy of the current will be about 6,400,000 foot-pounds per hour, about half of which, or only 3,200,000 foot-pounds, is converted into radiant energy in the electric arc, but this electric arc will radiate a light of 2000 candles when viewed horizontally, and two or three times as much when viewed from below. Hence 3 million foot-pounds changed to heat in the electric arc may be said roughly to affect our eyes six times as much as 60 million foot-pounds changed to heat in an ordinary gas burner.I Owing to the high temperature at which it remains solid, and to its great emissive power, the radiant body used for artificial illumination is nearly always some form of carbon.

The consideration of electric lighting naturally divides into two parts - the production of suitable electric currents, and the conversion of the energy of such currents into radiations. Although electric lights were first produced from currents generated by batteries, they have only attained commercial importance by the use of machines for converting mechanical energy into electric current.

Dynamo-Electric Machines. - In the widest sense a dynamo-electric machine may be defined as an apparatus for converting mechanical energy into the energy of electrostatic charge, or mechanical power into its equivalent electric current through a conductor. Under this definition would be included the electrophorus and all frictional machines ; but the term is used in a more restricted sense set forth here ; as they are fully explained in the article when the number of lines of force is increasing, it will be in the opposite direction when they are diminishing, it is clear that the current in each part of the circuit which passes through the magnetic field must be alternate in direction. Hence also the current in the wire outside the machine must also be alternate, unless something of the nature of a commutator be employed to reverse the conjust at the moment when the direction of the current would change.

The mathematical theory of alternate current machines is comparatively simple.3 Let T be the period of the machine, that is, the time taken to move the armature from one position to the next exactly similar position, e.g., in a Siemens alternate current machine of sixteen magnets on each side, one-eighth of the time of revolution ; let 7 be the coefficient of self-induction of the whole circuit, and R the resistance of the whole circuit ; and let t denote the time at any instant counting from any epoch as initial, and I the magnetic induction at time t multiplied by the number of convolutions. The electromotive force in the circuit at time t will be dt ; and the equation of the current will be t Rx = tit ' where x is the current. Now I may be expressed in the form - where A, and /, are constants for the macnine with given excitation of the fixed magnets. Hence 27rs t , , - t 7- + Ex- - A, cos 27rs - dt – - e-Ce 1-'5'=-A, T /(2irs-y tan 27. ST3 - 21-'Sy The term Ce 7 is unimportant except just after closing the circuit. In the Siemens machine N. Joubert states that the only important term is that of longest period. Hence, properly choosing the epoch, we write T x= 27rA 2,7)2 +112 tan - p T Hence we see the current is diminished either by increasing y or increasing B, also that the moment of reversal of current is not coincident with that of no electromotive force, but occurs after that time by an amount depending on the relative magnitudes of 7 and H. This explains in a general way what is known as the lead of the brushes in a continuous current machine. If we wished to apply a commutator to the Siemens alternate current machine for the purpose of producing an external current constant in direction, the change effected by the commutator should occur at an epoch after that of greatest electromotive force, an epoch which, with varying external resistance or varying speed, will depend on the resistance and speed.

The power of the current is Ilx2, and the energy in any considerable time, 0, is or,2,,A21 T2 / 2,7 ya + T ) which shows that most power will be required to drive the machine when In what precedes it has been assumed that the copper wires are the only conducting bodies moving in the magnetic field. In most cases the moving wire coils of these machines have iron cores, the iron being in some eases solid, in others more or less divided. It is found that if such machines are run on open circuit the iron becomes hot, very much hotter than when the circuit of the copper wire is closed; in some cases the phenomenon is so marked that the machine actually takes more to drive it when the circuit is quite open than when the machine is short-circuited. The explanation is that on open circuit currents are induced in the iron cores, but that when the copper coils are closed the current in the latter by its induction diminishes the current in the iron. The effect of currents in the iron cores is not alone' to waste energy and heat the machine ; the current produced is also actually less for a given intensity of field and speed of revolution. The cure of the evil is to subdivide the moving iron as much as possible in directions perpendicular to those in which the current tends to circulate.

Continuous or Direct Current Machines. - It has been shown that to produce a continuous current a commutator is needed. If there is but a single wire in the armature, or if there are more than one, but all are under maximum electromotive force at the same time, the current outside the machine, though always in the same direction, will be far from uniform. This irregularity may be reduced to any extent by multiplying the wires of the armature, giving each its own connexion to the outer circuit, and so placing them that the electromotive force attains a maximum successively in the several circuits. A practically uniform electric current was first commercially produced with the ring armature of Pacinotti as perfected by Gramme. Suppose a straight bar electromagnet surrounded by a coil of copper wire from end to end. Let the electromagnet be bent with the copper wire upon it until its ends meet and it forms an annulus or anchor ring. Let the two ends of the copper wire be connected, so that the iron core is surrounded by an endless copper wire, and you have the Pacinotti or Gramme ring. This ring rotates about its axis of figure between two diametrically opposed magnetic poles of opposite name. The ring may at any instant be supposed divided in halves by a diameter perpendicular to the diameter joining the centre of the poles. Equal and opposite electromotive forces act on the copper wire of the two halves, giving two opposite electric poles half way between the magnetic poles. If electric connexions could be maintained with these two points as the ring revolves, a continuous current would be drawn off. In practice this is only approximated to. The copper wire is divided into a series of equal sections, and at the point of junction of each section with its neighbour a connexion is made with a plate of a commutator, having as many divisions as there are divisions of the copper coil. Collecting, brushes bear upon the commutator plates, which are connected to the coil nearest to the point of maximum potential. Owing to the self-induction and mutual induction of the several coils of the armature, this point is displaced in the direction of rotation when a current is being drawn off, to an extent greater as the current is greater in relation to the strength of the magnetic field. The magnetic field in the Gramme and other continuous dynamo-electric machines may be produced in several ways.' Permanent magnets of steel may be used, as in the smaller machines now made, and in all the earlier machines ; these are frequently called magnetomachines.2 Electromagnets, excited by a current from a smaller dynamo-electric machine, were introduced by Wilde ; these may be described shortly as dynamos with separate exciters. The plan of using the whole current from the armature of the machine itself for exciting the magnets was proposed almost simultaneously by Siemens, Wheatstone, and S. A. Varley.3 For some purposes it is advantageous to divide the current from the armature, sending the greater part through the external circuit, and a smaller portion through the electromagnet, which is then of very much higher resistance, as the electromagnet is a shunt to the external circuit. Machines so arranged are sometimes called shunt dynamos.4 The last two arrangements depend on residual magnetism to initiate the current, and below a certain speed of rotation give no practically useful electromotive force.

In discussing the comparative efficiency of dynamo-machines there are two points to be examined - (l) how much of the power applied is converted into energy of current in the whole circuit, whether external or in the wires of the armature or of the electromagnets, and (2) how much of the power is available outside of the machine. The practical sources of loss are friction of bearings, and of the brushes on the commutator, electric currents induced in the iron of the machine, production of heat in the copper wire of the armature due to its resistance, and production of heat in the wire of the electromagnet due to its resistance. There is also a certain loss in sparks upon the commutator. The currents in the iron are reduced by dividing the iron by insulating surfaces perpendicular to the electromotive force tending to produce such currents. The loss by resistance of wire in armature and magnets greatly depends on the dimensions of the machine. For imagine two exactly similar dynamo-electric machines, the one being n times the dimensions of the other, we have the following relations between them, assuming the same magnetic field per square centimetre, and the same speed of rotation :- The electric resistances of the several parts are as 1 : n ; The electromotive force of the armature as n2 ; Current round magnets required to produce the field as a.

Thus the work wasted in heating the wire of the electromagnets varies as the linear dimensions of the machine. The current which the armature can carry with safety to the insulation will increase more rapidly than the linear dimensions of the machines, but less rapidly than the square of the linear dimensions. If the current vary as the linear dimensions n, the whole electric work done by the machine will vary as its weight n3, and the work wasted in the coils both of the electromagnets and of the armature will only vary as n, - showing a great theoretic advantage in favour of the larger machines.

Electric Lamps. Incandescent Lamp.' - The simplest way of obtaining light from an electric current is by passing it through a considerable resistance in such small compass that the conductor becomes intensely hot. It is of course necessary that the conductor shall be able to endure a very high temperature without injury. Iridium and platinum-iridium wire have been employed, but are too expensive for commercial use. Hitherto the only available substance is carbon, in the form of a thread or filament. This carbon must be protected from the air by enclosing it in a glass globe from which every trace of air has been removed. An electric current passing through a carbon filament obeys Ohm's law, as through a metallic wire. But in metals the resistance increases as the temperature rises, in carbon it diminishes.2 The filament or thread of carbon being enclosed in a vacuous space, the energy of current converted into heat in the filament only leaves it in the shape of radiations. To light economically, it is necessary to heat the filament to such a temperature that the greatest possible proportion of these radiations shall belong to that part of the spectrum to which the eye is sensitive, i.e., to the highest temperature the filament will stand. The fundamental problem of incandescent electric lighting is to produce a carbon thread the substance of which shall permanently stand the highest possible temperature, to make good electrical connexion between the ends of the filament and the conducting wires, and above all to secure that the thread shall be uniform throughout its length, for the current which can he safely used is limited by the weakest point of the filament. Several inventors have recently succeeded in meeting these conditions, but their relative merit and priority cannot be discussed here.3 Semi-incandescent Lamp. - The lamps of Werdermann, Reynier, and Joel are intermediate between arc lamps and incandescent lamps, and present the distinctive advantages of neither.4 Arc Lights. - Sir Humphry Davy discovered that if two pieces of carbon were placed in contact with each other, and the current from a battery of a sufficient number of elements were passed from one piece to the other, the current did not cease when the carbons were slightly parted, but that the current passed across the intervening space, causing an intensely high temperature and consequently brilliant light. The pieces of carbon gradually burned away, the positive carbon being consumed more rapidly than the negative. When an electric current passes through a conducting solid body maintained at a constant temperature, the difference of potential on the two sides of the body has a constant ratio to the current passing through; this constant ratio is known as the electric resistance of the body at its then temperature. No such constant ratio exists in the case of the electric arc. If you increase the current passing between two carbons at a small distance apart, you do not materially change the difference of potential at the two ends of the electric arc. It is, therefore, not strictly appropriate to speak of the resistance of the electric arc ; the appropriate constant, or approximate constant, for an electric arc is 'the difference of potential between the two sides of the arc.6 However near the carbons approach without touching, this does not fall below a certain minimum value, and as the carbons are separated its value increases. In ordinary practice with continuous currents the potential of the electric arc may be taken as ranging from 35 to -15 volts. If the current in amperes be multiplied by the difference of potential in volts, and the product be divided by -1-16, we have the power used in the are itself in horse-power, that is, the power effectively used in lighting. The mechanism of an electric lamp has two functions to perform, it has first to bring the carbons into contact and then part them, or simply part them if they are initially in contact when the light is started, or when it is accidentally extinguished (this is called striking the arc); it has also to bring the carbons together as they are consumed. The former function is always accomplished by an electromagnet or solenoid. In the electric candles, e.g., those of Jablochkoff, Rapieff, Wilde, or Siemens, the carbons are approximately parallel, and they burn down as does a candle, - the arc being forced to the ends of the carbons by the repulsions of the current in the carbons on the electric arc.6 In the ordinary arc lamps the carbons have their axes in the same line, and their approach or recession must be controlled by the current passing through, by the difference of potential, or by both combined. When the same current passes through a succession of lamps in series, it is clear that the regulation cannot be by the current alone, as this is the same for all the lamps, and might be maintained constant by the adjustment of any one only of the lamps. When lamps are burned in series, it is essential that the difference of potential shall be an clement in the control. This is done by using an electromagnet bound by fine wire so as to have a resistance of some hundreds of ohms, and connecting it to the two sides of the arc. In the Siemens differential lamp, and in some others, a potential or shunt coil and a current coil oppose each other ; as the arc lengthens the current becomes less, and the potential greater, each acting to cause the carbons to approach. It will be seen that the possible combinations of mechanisms and electromagnets for adjusting an electric arc are endless; and so also are the patents for such combinations.' When an alternate current is used for an electric arc, the phenomena are much more complicated, owing to the difference of potential being a discontinuous function of the time. The difference of potential will be (say) 40 volts in one direction for a certain fraction less than half of the periodic time of the current; the current then entirely ceases, generally for a finite time, and is then reversed with a sudden reversal of difference of potential.2 The work done in the arc is measured by the time integral of the product of difference of potential and current passing. A knowledge of neither the mean strength of the current, nor of the difference of potential, nor of both, gives the means of ascertaining the work done in an arc with alternate current. The only satisfactory electrical method is the quadrant electrometer suitably connected, and this is open to the objection that a considerable resistance must be introduced into the circuit.

Electric Light ifeasurements.-Under this head we content ourselves with a warning. A bare statement that an electric arc light is of so many candle power really conveys no accurate information at all. The light from an electric arc differs greatly in colour from that of a candle;3 a given arc light may have three thousand times as much red of a certain wave-length as a standard candle has of the same wave-length, but ten thousand times as much green light. Any one will admit that green light is not measurable in terms of red light; a mixture of red and green is not See for descriptions of various arc lamps :-BROCKIE : Engineering, xxxi. 93 ; Engineer, xlix. 268 ; Tel. Jour., viii. 114 ; Electrician, iv, 232. BRUSH : Engineering, xxxi. 55, 85, 123 ; Engineer, li. 15 ; Tel. Jour., vii. 21 ; Electrician, iii. 87 ; Fontaine, 45. CAECE Engineering, xxxiii. 30. cm: iF1Os : Engineering, xxxii. 205 ; Engineer, xlix. 323 ; Tel. Jour., viii. 131; Electrician, iv. 273, Jour., vii. 301, ix. 73 ; Electrician, iii. 201. IIICKLEY : Tel. Jour., vii. 371. JASPAR : Engineering, xxxii. 645 ; Fontaine, 40. KRUPP: Engineer, xlvii. 167 ; Tel. Jour., vii. 198 ; Electrician, ii. 255. LACASSAGNE and THIERS : Fontaine, 28. LoNTIN Shoolbred, 33 ; Fontaine, 59. MACKENZIE.: Engineering, xxxi. 38. Mnxiai : Jour., viii. 417, ix. 144 ; Fontaine, 69. 1110LERA and CEBRIAN : Tel.

vii. 2.31. ORNIE : Tel. Jour., vii. 184. PILSEN : Engineering, xxxi. 5l4, xxxiii. 152 ; Tel. Jour., viii. 419. PAFIEFF: Engineering, xxvii. 55 ; Tel. Jour., vii. 60 ; Fontaine, 22 ; Shoal-bred, 34 ; Report from the Select Committee on Electric Lighting, 239. SCRIBNER : Tel. Jour., viii. 379. SERRIN Shoolbred, 31 ; Fontaine, 53 ; Schellen, 218. SIEMENS : Engineering, xxxi. 276 ; Tel. Jour., vii. 318, 412, viii. 98 ; Electrician, ii. 52; Schellen, 227 ; Fontaine, 63 ; Shoolbred, 33. SOLEIL ; Engineering, xxxii. 453. STEWART : Tel. Jour., viii. 80, 115. Tuonsos and HOUSTAN: Engineer, xlvi. 295 ; Electrician-, i. 282 ; Fontaine, 67. TCIIIKOLEFF : E1ecerlcian, v. SO. WALLACE-FARMER : Engineer, xlvi. 295 ; Shoolbred. 36 ; Fontaine, 33 ; Report front the Select Committee on Electric Lighting, 246. WESTON : Engineering, xxxii. 42; Electrician, viii. 246.

Joubert, Journal de Physique, ix. 297.

measurable in terms o, another mixture in which the proportions of the colours are wholly different. Again, the intensity of the light obtained from an arc light depends greatly on the direction in which it is viewed.- Neither of these considerations applies in the same degree to incandescent lamps. (J. no.)