150 



THE CIVIL ENGINEER AND ARCHITECT'S JOURNAL. 



[Mat, 



A KEW SIGNAL LIGHT FOR RAILWAYS. 



By Alan Steveksox, LL.B., Civil Engineer, Edinburgh. 



(Riadht/ure ih Socit/ij vf Aits for Scotland, ■22vd Ftiniaii/, 1&4J.J 



The numerous accidents, attended with fatal consequences, which 

 have lately occurred on railways, have excited much alarm in the 

 public mind, iind the prevention of these casualties is unquestionably 

 a matter of great importance. The object of this communication is, 

 to point out one source of danger to which several of the late acci- 

 dents may be attributed, and to suggest the means of its removal ; 

 and from the personal inter<»st which all must have in the improve- 

 ment of railway travelling, both as regards its speed, and, what is of 

 much greater importance, its safety, 1 venture to hope that the follow- 

 ing observations, although limited to one part of the subject, will not 

 be found to have been unsuitably addressed to a society whose pro- 

 vince it is to improve the useful arts. 



One of the most imperfect parts of the railway system is undoubt- 

 edly the uncertainty of the night signals, and to tliis it is well known 

 many of the most fatal of the accidents which have occurred must be 

 traced. The great object of these signal lights is, to announce that 

 the train has reached a certain \>umt of its course, and to forewarn the 

 engineman of his approach to a station, or the junction of a branch 

 railway, so that the sjieed of the engine may be checked in proper 

 time tc prevent collision. The lights used for this purpose are gene- 

 rally exhibited at the jjlace the a|j|iroach to which they are intended 

 to announce : but the distance at which light projected horizontally, 

 may be seen by a person approaching in the line of its transmission is 

 very variable according to the state of the atmosphere, which in our 

 climate is subject to great and sudden changes, in regard to clearness 

 and fog. These variations in the visibility of lights of extensive 

 range are by no means confined within narrow limits, as experience 

 too amply demonstrates in the case of lighthouses, whose range has 

 been known to vary with the state of the atmosphere, from sixty miles 

 down to two or three miles ; and this evil is unhappily one of those 

 which, in the present state of chemical and optical science, must, we 

 fear, be pronounced irremediable. This defect, great as it is in regard 

 to lighthouses, is, in the case of railways, materially aggravated by 

 the excessive velocity of railway travelling. Any variation in the 

 distance at which a signal light is first seen, must lead to great mis- 

 conceptions as to the time of reaching a station, and all such miscon- 

 ceptions arc fraught vvitb the worst consequences, owing to the nume- 

 rous sources of danger from the crossings of branch lines, the meeting 

 of carriages on the rails, or the occurrence of other accidents, which 

 may render a railway impassable. It is therefore obviously indis- 

 pensable to safety that the signal-lights should be so constructed, that 

 in all states of the weather they shall be constantly visible at the same 

 point, and that this point shall be sufficiently distant from the station, 

 the approach to which the signal is intended to announce, so as to 

 allow ample time for checking the engine's speed before coming up 

 to it ; and upon no other grounds can the confidence of the public as 

 to their security be reasonably based. 



In the month of December last, it occurred to me in the course of 

 conversation with my friend Mr. Errington, civil engineer, that although 

 the variation in the visibility of lights of distant range must, according 

 to our present knowledge, be regarded as an evil without remedy, it 

 might still be possible, by means of some arrangement of the lights, 

 to render signals for railicaijs cvjinlanlhj risible at the same 2joint during 

 tvtry state oj the atmosphere. For this juirpose, all that seems to be 

 necessary is, to limit the range of the lights, and at the same time to 

 increase their intensity in such a manner that the combination of a 

 short range with great power may not merely render them capable of 

 penetrating any tog however dense, but of producing, at a certain 

 point, an eli'ect so brilliant and striking as forcibly to arrest the en- 

 gineman's attention. After considering the matter in various points 

 of view, I came to the conclusion that the object could be best attained 

 by placing the light considerably in advance of the station, the approach 

 to which it is intended to announce, and by giving the beam such an 

 inclination to the horizon, that its greatest power may fall upon the 

 esgineman's face, at so short a distance from the light itself, that it 

 could not fail to be always visible at that point, even in the thickest 

 fog. 



According to the present practice, a comparatively feeble light is 

 exhibited at the station whose position it is intended to point out, and 

 this light, which is permitted to pierce the gloom until its power is 

 greatly diluted by tlie united etVects of its own divergence, and the 

 length of its jiassage through a foggy medium, must necessarily be 

 subject to constant variation of visibility with every change of the 

 atmosphere. The change which I have to suggest, is to place a light 



of great power about a mile in advance of the station, and at the same 

 time to limit its range by the depression of the resultant beam within 

 such a distance as to ensure its being visible at all times. 



The arrangement I would propose for the attainment of this object 

 is remarkably simple, and consists in placing one of Fresnel's annular 

 lenses, illuminated by a gas or oil burner, as may be most convenient, 

 in a small chamber, glazed in front, and supported on a stage of car- 

 pentry of sufficient size to span the rails, and permit the train to pass 

 under it : but the purpose might perhaps be equally well served by 

 placing the stage at the side of the railway, and inclining the beam 

 obliquely to the line. In order to limit the range of the lens to a short 

 distance, and thereby to ensure the light being visible in all states of 

 the weather at the same point, I would incline the instrument, so that 

 the length of the trajectory from the lens to the oljserver's eye should 

 not exceed about 7uo feet, which f dis far short of the distance at 

 which the light of the lens would be obscured even in the thickest fog. 

 I may remark that the inclination of the lens is too small to require 

 any correction in the position of the flame; but this could be easily 

 accomplished if necessary, more especially when gas is employed. 

 In curved lines of raihvay the same effect might in certain cases be 

 produced by placing the lens on a level with the observer's eye, and 

 directing the refracted beam so ixs to cut the railway obliquely. In 

 this case the limitation of range would be produced without the ne- 

 cessity of inclining the lens ; but the principle of rendering the signal 

 at all times effective, by combining a short range and a powerful light, 

 is the same in both arrangements. 



The advantage of this arrangement I conceive to be great, for not 

 only would the light be at all times visible to the engineman on his 

 arrival at the same point which, as already mentioned, is really the 

 great object of signal lights; but it is obvious that his attention would 

 be most etfectually awakened by the contrast of suddenly passing from 

 darkness to receive the full effect of a powerful light viewed from a 

 short distance. One other advantage of the proposed signal light, I 

 must observe, lies in its being peculiarly susceptible of any modi- 

 fication of colour, whether of a temporary or permanent kind, which 

 the numerous and growing wants of an extended railway system may 

 require. The alphabet of nocturnal telegraphy, wherever a distant 

 range is required, is unhappily extremely scanty ; for the practice of 

 all Europe seems to have shown that, so far as colour is concerned, 

 red and lohtle are its aljjha and omega ; green and blue have been 

 frequently tried; but cautious inquirers have all agreed in pronouncing 

 them so equivocal when viewed from a distance, that they have been 

 almost universally abandoned. These colours, however, and even 

 much less marked varieties, although useless as distinctions for lights 

 of distant range, are perfectly effective when viewed from short dis- 

 tances, as the brilliant display of an apothecary's window sufficiently 

 proves. 



I shall no%v add a very few words regarding what appears to me 

 to be the chief arrangements which may, in practice, be found neces- 

 sary for signal-lights on these principles; but I would not be under- 

 stood as attempting to fix any thing permanently, for I am well aware 

 that various modifications may be suggested by experiment, which I 

 do not at present foresee in their full extent; in particular, it seems 

 probable that the range of visibility which I have adopted in the fol- 

 lowing view of the details, falls short of what will be found quite suffi- 

 cient in practice even during the thickest fog.s, when a light so power- 

 ful as that which may be derived from Fresnel's lens is brought into 

 play ; and should this expectation be realized, the duration of the 

 effect of the light, which depends on the range, might be increased 

 beyond what I have ventured to state. 



Referring to the above sketch, I would propose that the lens at L 

 should be elevated 24 feet above the rails R R, or about 15 feet above 



