i84 



NATURE 



[June 25, 1908 



affect the introduction of metal filament lamps do not 

 appear to have been well founded* 



Beyond these two types of lamp the development in 

 incandescent lighting has been slight. The exceedingly 

 interesting discovery (from a scientific point of view) by 

 the General Electric Company in America of a method 

 for radically altering the nature of the carbon filament 

 unfortunately came too late to have any practical effect. 

 These so-called " metallised " filament lamps, had they 

 come four or five years ago, would have been welcomed 

 as a great step in advance, but coming as they have at 

 the same time as the metallic filaments, are practically 

 doomed to failure, since they possess the same dis- 

 advantages, and, in addition, can only be worked at an 

 efficiency of ■2\ watts per candle. The same may be 

 said of the Nernst lamp, which is almost bound to 

 give way in the sphere of incandescent lighting to metal 

 filament lamps. It is possible that the Nernst lamp will 

 find a sphere of its own for intermediate lighting where 

 light units of 100- to 200-candle-powcr are required, but 

 it is much more probable that it will be ousted also from 

 this field by the high candle-power metal filament lamps. 



Arc Lamps. 



The remarkable change in the prospect of incandescent 

 lighting which has been effected by the introduction of the 

 metal filament lamp has been paralleled by a similar 

 change in arc lighting by the introduction of flame lamps. 

 The gain in efficiency in a metal filament lamp over a 



Fig. I. — .\rrangement of Carbons in Flame Arc. 



carbon filament lamp is approximately three times, and 

 an almost similar gain is obtained in a flame lamp over 

 an ordinary open type arc lamp. The idea of introducing 

 chemicals into the arc in order to colour the flame is an 

 exceedingly old one, but the practical solution was not 

 obtained until Bremer brought out his mineralised flame 

 carbons. The carbons introduced by Bremer were 

 mineralised carbons in which the flame-producing material 

 was intimately mixed with the material of the carbon rod 

 itself. Alterations were later effected by the various 

 carbon manufacturers by which the flame-producing 

 material, instead of being introduced into the main body 

 of the carbon, was only introduced into the core. This 

 latter type of carbon is by far the most common, but 

 carbons of the Bremer type are still in use, and have been 

 considerably improved recentlv by M. Blondel, who claims 

 to have succeeded by their employment in producing a 

 far more efficient lamp than the ordinary flame lamp. 



Flame carbons are constructed to burn with the carbons 

 arranged vertically above one another, as in ordinarv 

 lamps, but the more general construction is to arrange 

 the carbons side by side inclined to one another at an 

 angle of about 15°, as shown in Fig. i. The arc, which 

 balances between the tips of the carbons, as shown in 

 the figure, is spread out into a fan shape, and kept down 

 at the tips by the use of a magnetic controlling field. 

 Immediatelv above the arc, both in the lamps for vertical 

 and inclined carbons, is placed an inverted cup, the object 



NO. 2017, VOL. 78] 



of which is to prevent the free upward currents of air 

 and maintain the carbons always in an atmosphere of 

 inert gases, thus considerably lengthening their burning 

 hours. This cup is consequently given the name of an 

 economiser. 



The arrangement of the carbons in the inclined lamps 

 causes a very large proportion of tlie light to be thrown 

 vertically downwards, which is certainly not a desirable 

 condition for the lighting of open spaces, involving as it 

 does hanging the lamps very high if even illumination is 

 to be obtained. The rich golden-yellow colour of the light 

 though useful for the purpose of display, is also to be 

 reckoned as a disadvantage of these lamps. The colour 

 is, however, not so very far different from that given by 

 the incandescent lamp, and has not been found a very 

 serious drawback. Flame arcs of other colours, for 

 example, white and pink, can be produced, but the volume 

 of light in these cases is greatly below that of the yellow 

 arc. The flame arc lamp has an efficiency of approxi- 

 mately 0-4-0-5 watt per mean spherical candle, which is 

 two or three times as good as that of the ordinary open- 

 type arc. The high cost of maintenance in carbons, which 

 are expensive to make and burn rapidly, is very much more 

 than compensated by the low cost of power for a given 

 amount of light. In the Blondel lamp the carbons which 

 are used are of larger diameters, and are arranged 

 vertically one above the other. This has the ad\'antage 

 of giving a better light distribution, and it is claimed that 

 the mean spherical candle-power is nearly double that of 

 the ordinary flame arc. 



Whilst the flame arc has been rapidly developed in 

 Europe, it has met with but little success in America on 

 account of the fact that in America the cost of labour 

 for frequent trimming counts much more than it docs in 

 this country. For this reason the use of enclosed arcs in 

 .■\merica has become almost universal in spite of the fact 

 that they only have about half the efficiency of the ordinary 

 open arc. To meet these somewhat special conditions, ex- 

 perimental work has been carried on during the past two 

 or three years by Messrs. Steinmetz and the General 

 Electric Company of America on the production of a long 

 burning flame lamp. The " magnetite " arc, as it is 

 called, which is the result of this work, utilises electrodes 

 composed chiefly of magnetic oxide of iron. In the latest 

 electrodes titanium oxide, which produces a more intense 

 light in the arc than iron oxide, is used, but iron oxide is 

 still employed to give the electrode conductivity when cold. 

 The actual mixture contains also oxide of chromium, but 

 this plays a purely secondary part in steadying the rate 

 at which the other o.xides are evaporated. 



The lamp is constructed with the magnetite electrode 

 as the negative, and a solid copper electrode as the positive. 

 In place of copper a special alloy is being introduced. 

 The arc has all the characteristics of a flame arc, but 

 possesses some peculiarities due to the fact that the flame- 

 producing material is contained only in the negative elec- 

 trode. Its efficiency, from figures which have been pub- 

 lished, appears to be in the neighbourhood of oS to i-o 

 watt per candle, from which it will be seen that it is 

 a little more efficient than the ordinary open-type arc 

 and twice as efficient as the ordinary enclosed arc, which 

 will be its greatest competitor in .'Xmerica. The electrodes, 

 when 12 inches long, are stated to have a life of 150 

 hours, which is as much or more than similar sized 

 electrodes in enclosed lamps. It will therefore be seen 

 that the magnetite arc is likely to prove a valuable advance 

 on the enclosed arc, and where economical conditions have 

 determined the use of this lamp it is likely to be super- 

 seded by the magnetite arc. It cannot, however, at present 

 compete against the ordinary flame lamp, where the cost 

 of trimming is not so important a factor. 



Vapour Lamps. 



.\ word may be said in relation to the development of 

 the mercury vapour lamp. This type of lamp, which has 

 been developed by Mr. Cooper-Hewitt in America, has not 

 met with any extended use on account of the exceedingly 

 unpleasant colour of the light which it gives. Though 

 in some circumstances this may not prevent the adoption 

 of this lamp, there is no doubt that it will always seriously 

 hamper its use in competition with pleasanter coloured 



