246 



THE CIVIL ENGINEER AND ARCHITECTS JOURNAL. 



[Al'gvst, 



(lie explosion had niAnifeslly origiii;itpd, commencing in tlie older iron 

 wliicli remained in the hinder part of the boiler. The tubes were, for 

 the most part, still f.ist in the tube-sheets, but they were bent outwards 

 :it their middle, like the sta%-es of a barrel. The steam pipe, as it 

 passed tlirough the boiler, was coUapud, but not broken. The outer 

 shell of the boiler had been torn into fragments, and the rents had ex- 

 tended to the vertical part, the upper portion of which hid been en- 

 tirely torn away, so as to expos" the fire-box, which was sound, but 

 slightly caved in on the sides. The cylinders were unharmed by the 

 explosion, but had since been removed. Tlie qu dily of the iron ap- 

 jieared to the conimiltee to be uniformly good. 



There was, therefore, nothing about the engine to indicate that the 

 accident had occurred from defects in workmanship or material, nor, 

 indeed, did the tremendous power which was indicated, seem consis- 

 tent with the idea of an originally defective boiler. 



The evaporative power of these heavy engines is necessarily very 

 great. Mr. Nicolls assured the committee that the Neversink was 

 capible of drawing a train of 83 cars, weighing, load->d, 7i tons each, 

 (equal to G37 tons,) at a speed of 12 miles per hour, (lU5ti feet per 

 minute.) Allowing the traction to be 7i pounds per ton, (as experi- 

 ments upon this road show it to be,) this is equivalent to 153 horse 

 power — requiring an evaporation of 2-55 cubic feet of water per 

 minute. 



Now, by the peculiar construction of these engines, rendered ne- 

 cessary by the restricted space allowable for the boiler, when the 

 water-level stood two inches above the lower cock, the steam was 

 rontined exclusively to the hemisphericd dome above the fire-box, 

 the cubic content of which is rather less than 24 cubic feet, (23-8o6o 

 <uibic feet.) The cubic content of each cylinder (13-5X20) is 1-657 

 ( l5l cubic feet, and, as two cylinders are diawnat once, the ratio of the 

 cylindrical content is as 3-314 to 21, or more than one-eighth. When 

 the water-level is at the upper giuge-cock, the steam room is nine 

 cubic feet, and the ratio about one-third. Now, the most recent (and 

 apparently the best) authority upon the high pressure engine declares, 

 after nearly 30 years of practical experience, that " the steam space 

 should be at a minimum 20 times as great as the space to be filled 

 with steam in the cylinder. If it can be made greater, consistently 

 with the other arrangements of the boiler, so much the better." This 

 is, of course, inapplicable in locomotive engines. 



The reason, therefore, that these engines will throw water from the 

 safety-valve, and from the gauge-cocks, when the actual water-level 

 is dangerously low — and that, in the words of Mr. Kirk, they are tick- 

 lish in carrying their water, must be evident. The foaming in one of 

 these engines must be incessant, and the danger of priming verv great. 

 The gauge-cocks, which, under the most favourable circumstances, 

 are but indifferent indicators of the water-level, become, in this case, 

 useless, and the engine driver must rely upon his experience of the 

 engine and trust to incessant watchfulness alone, if he would avoid an 

 accident. 



A very remarkable fact about this explosion is, that the steam pipe 

 passing through the upper part of the boiler, from the throttle valve 

 to the cylinders, was co//a/)sei and unbroken, as is well seen in the 

 accompanying Daguerreotvpe portrait of the engine, taken after the 

 explosion by Mr. David Monday, of Reading, and kindly lent by him 

 to the committee. It is, indeed, possible that this may have been pro- 

 duced, during the explosion, by the sudden bending upwards of the 

 tube, otherwise it would seem to indicate that the engine was throttled 

 at the time of the explosion; an expedient which may have been re- 

 sorted to for the purpose of avoiding the dampness of the steam, or to 

 check the speed oftlie engine; but the fearful danger of which will 

 be seen when it is considered that, if the steam was shut off but one- 

 fourth, (the water being above the lower gauge-cock,) the pressure in 

 the boiler would double itself in about one minute. 



It seems useless to speculate upon the immediate cause of this ter- 

 rible accident, since the death of all upon the engine has removed the 

 direct testimony of the circumstances under which it occurred. How- 

 ever, it appears th.it tlie engine was under a very unusually heavy 

 pressure of steam, and sc.ircely less certain that the safety valve was 

 (dccidently or otherwise) fastened down. Mr. Nicolls and Mr. Kirk 

 both testify to the competency of the engine driver, who was in charge, 

 and every one bears witness to his character for sobriety. That he 

 may hai-e been deceived as to the height of water in the boiler is 

 possible from the char.icter of the engine, although it is diilicult to 

 imagine how an experienced hand could have neglected the indica- 

 tions given by the increased pressure, as shown by the rapid running 

 of the train and the sharpness of the exhaust. 



Upon the whole, it appears probable to the committee that the ex« 

 plosion of the Neversink occurred in this way: — 



That the engine w.is running under a heavy pressure of steam, and 

 that, owing to the defective indicatious of the gauge-cocks, the water 



in the boilers was permitted to get below the upper tubes, which thus 

 became unduly heated ; that the rapidly increasing pressure (assisted, 

 perhaps, by an injudicious partial closing of the throttle valve) caused 

 the starting of one or more of the lubes from the forward tube-sheet, 

 and this sudden relief of the pressure caused a foaming in the boiler, 

 by which the water was thrown over tl.e heated tubes, and being thus 

 rapidly evaporated, caused an instantaneous increase of tension, which 

 the additional openings were incompetent to relieve, and thus produced 

 the rupture of the outer shell of the boiler. This, however, is intended 

 only as a plausible suggestion, and by no means as a confident afTirma- 

 tion of the cause of the explosion. 



But whatever hypothesis may be adopted to explain this unfortu- 

 nate accident, its investigation has forcibly tailed the attention of the 

 committee to several matters which they believe to be of sutBcient 

 practical importance to deserve the attention of the Institute. 



First. The necessity of pr<ividing all steam engines with a second 

 safety-valve, of large dimensions, regulated to the maximum pressure 

 which the engine is intended to bear, and placed beyond the control 

 of the engine-man. It is true that this will entail upon the owners 

 the trouble of frequent examinition to maintain the efficiency of such 

 a valve, but this trouble will be more than compensated by the in- 

 creased s ifety which will be procured by its use. 



Secondly. The uncertainty of the ordinary gauge-cocks, as indica- 

 tors of the water-level uad-jr the m ist f ivour.ible circumstances, and 

 the deceptive character of their indications upon the modern locomo- 

 tivvj engines, where the amount of work to be done and the restricted 

 space which can be allowed to the boiler, necessarily confines the 

 water and steam room, and renders the evaporation more tumultuous 

 than in the larger boilers of stationary engines. 



Thirdly. The committee would suggest the inquiry whether it is 

 not feasible and advisable so to construct the locomotive engine that 

 explosions, if they occur at all, shall tike place in such a manner as to 

 bj less destructive to hum. in life than they at present are. One of the 

 great recommenditions of the tubular boiler, when first introduced 

 into use, was this very diminished liability to do injury, by allowing a 

 tubular flue, of comparatively smill siz-', to collapse, in place of tlw 

 large cylinders, by which the boiler was at once emptied of its con- 

 tents. 



REVIEWS. 



EartJimrk Ta'tUs. By Charles K. Sibley, C.E., and William 



RurEERFORD, F.R.A.S. London : Longm ui and Co., 1847. 



These tables are for the purpose of estimating the contents in cubic 

 yards of the e.irthwork of r. always : they are calculated, by the ordi- 

 nary prismoidal formula, for a central width of 33 feet at slopes of 

 1, 1^, and 2, to 1 ; heights from to 60 feet, at intervals of half-a- 

 foot. 



The advantages of the tables are, that ther.? is ni necessity for a 

 second calculation, as at one glance the cub.c contents of a chain in 

 length are seen by merely looking for the corresponding heights of 

 the respective ends of each chain's length in the table,— the heights 

 of one end being given at the bottom of the tible, and the other 

 height on the side, and at the intersection of the two lines the cubic 

 contents are given. Thus, for a cutting 5 chains in length, of the 

 respective heights of 



(0, 7-5, 13, US, 3i, and 0, the contents are read off 

 I 34S, 1090, 1355, 756, 151 :— total, 3700 cubic yards. 

 We believe these tables are the only ones that offer suc-fi a facility of 

 calculation; consequently, we strongly recommend them to the Pro- 

 fession. 



VVe must observe that there is another table, by which the content* 

 for any other width, from 23 to 43 feet, m.iy be fjunl ; for this pur- 

 pose, it will be requisite to have two inspections, but no multiplica- 

 tion. 



Archilecliiral Mixims and Theorems, and Lecture on the Education 

 and Ch'iracter of the Architect. By Thomas Leverto-V Dj.n'ald- 

 SON, M.l.R.A. London: John We.ile, 1S47. 



Professor Donaldson has laid the grouid-work for an excellent 

 book ; but in the present edition the M ixims are too concise, and are 

 not carried out sutBciently to render them of much service to the 

 student. M.inv of tlie M ixims require an expl.mition and a reason- 

 ing to prove that what is set forth is true. We feel assured that 

 Mr. Donaldson, if he can devote the time to the work, will be enabled 

 to enlarge it in such a m inner as to mik3 it a valuable work of 



