264 



THE CIVIL ENGINEER AND ARCHITECT'S JOURNAL. 



[August, 



licaring dirccl on cross timliers, without the intervention of chnirs, tlie wooJ 

 is cut to tlie requisite angle; or tlie inclination is some'.imes given to the 

 rails in i)assing tlirougli tlic rolls. M'ithout this precaution of inclining the 

 bearing stirl'ace of the rail to meet the cone of the wheel, the edge rapidly 

 wears, and the laminfe of iron peel oil' in strips, more or less, according to its 

 quality, and there is no more critical test of the perfection of rolled iron 

 rails than the manner in which the hutton edges go through this ordeal. 

 With the above precautions of "cone," " play." and elevation of the outer 

 rail, the res-istances opposed by curves to a single carriage may be considered 

 to be practically anniliibted ; hut when the trains become very lonp-, there 

 must, of necessity, be a considerable lateral action and grinding, from the 

 change of direction of the original drawing force through a number of 

 carriages', but Mr. Vignules stated that, although no conclusive experiments 

 had been made to show the e.'cact amount of resistance from this cause, his 

 oivn observations and experience led him to conclude that the degree of cur- 

 vature on railways might be safely extended further than they have hitherto 

 been laid down on principal lines. 



Mr. Vignoles referred to fonner observations of his, that the public would 

 be better accommodattd by more frequent departures of smaller trains, and 

 that with such trams the curves would be of still less importance, adding 

 that it could only be by the introduction of greater curvatures to save ex- 

 pense ; and, as he had re[)eatedly argued fur the same reason, by the 

 adoption of steeper gradients, that the benefits of railway communication 

 could be extended through many districts, and to the m.rc distant parts of 

 the country, as on the most economical principles of construction. Mr. 

 Vignoles referred to (he report of tlie Irish Railway Commissioners, and to 

 the works of Mr. \Vood, M. de Pambour, Lieut. Lecount, and other writers, 

 for further details on curves, observing, in conclusion of this part of the 

 subject, that where curves are so quick as to require it, especially in crossings, 

 the additional precaution of guard rails becomes expedient. 



In adding a few words on the subject of coupling carriages together in a 

 train, the Professor insisted strongly on the draw-boys being always in the 

 centre, and observed that, as a general rule, the connecting links should be 

 screwed up as stiff as possible consistent with the curves of the railway, as 

 otherwise the carriages are apt to swing. He mentioned that the best 

 coupling was that of Mr. Ilcnry Booth, the talented manager of the Liver- 

 pool and Manchester Railway frum its very first origin. But, as a general 

 form of combining the draw -boys and buffers on a central rod or tube with 

 spiral springs acting solely from the centre, Mr. Vignoles spoke in the 

 strongest terms of the apparatus of Mr. Thomas F. Bergin, the manager of 

 the Dublin and Kingstown Railway, on which line they had been used with 

 advantage for a number of years. 



I.ECTIKE IX.— ON TTNXELS. 



Ix proceeding to treat of the subject, which might be termed that of the 

 great works of art, to be introduced in the formation of roads or canals, but 

 particularly uf railways, Mr. Vignoles said that it would not be possible in 

 the lecture-room to go into the details of the constructions, but that he 

 must limit himself to general principles. The rules for consideration when 

 such works ought to be adopted were sufficiently simple ; for example, to 

 determine where tunnels should be substituted for open cuttings, or viaducts 

 for embankments. The French engineers, who are in general very much 

 better mathematicians than we are, and probably, from that very circum- 

 stance, more inclined to be theoretical, are much in the habit of introducing 

 formulx which, often very useful, are not always readily applied by the 

 practical men of this country. Supposing it to be required to determine aj 

 what point on a longitudinal sectiun (for road, railway, or canal,) it is 

 advisable to begin to tunnel, instead of continuing a simple excavation — that 

 is, the point where it becomes as cheap to tunnel as to cut open — for such a 

 case the following formula is given by an eminent French engineer : — Let 

 «= depth of cutting ; / = breadth of road, railway, or canal, on the travelling 

 surface; a — depth of bed, of road, or railway, to be first excavated, and 

 afterwards filled with road material or ballast ; or depth of canal below the 



water-line; ~ =slope of excavation ; p = price of the cutting per cubic 

 yard. The expense of excavation per yard forward of the sectional area of 

 any cutting will consequently he a pxp (Ix- = x; and, when this price ex- 

 ceeds the price per lineal yard forward of tunnelling, the latter is cheaper, 

 supposing the given prices to cover all risk and contingencies in each case. 

 But, as circumstances are continually varying, the English engineer so 

 repeatedly finds that he has to modify — and perhaps finally abandon — the 

 general theiiretical rule, and fall back on his own experience, and that of the 

 contractor he may be disposed to employ, that, although he may occasionally 

 resort to such a formula as an approximation, he ceases to employ it in 

 practice, and obtains the sectional area of the given cutting in superficial 

 yards, by simple mensuration, and multiplies it by the price. All the complex 

 conditions involved by slips, faults, water, and the ntimerous incidental 



occurrences in great works, to occasion unforeseen expenses, render prices 

 uncertain, and prevent any fixed general rule : and it is only when the 

 materials and probable contingencies are perfectly well known, that the 

 element of cost can be safely introduced Into the mathematical formula. For 

 dry indurated sands, gravel, sandstone rocks. &c., calculations may be made 

 within probable limits of error ; whereas, in many instances where the theo- 

 retical rule and general opinion, even of thosi sufficiently experienced, would 

 recommend tunnelling, it has been tried in vain, abandoned after great 

 expense in contending with water, and recourse had after all to open cutting. 

 The average cost of tunnelling upon the principal railway lines, as actually 

 executed, appears to be about 60/ per yard forward, in some instances as 

 much as 100?., especially when driven forward in reckless haste, and in 

 attempting to sink shafts or drive drifts, without due consideration as to the 

 quantities of water in the various strata, or the means of at once grappling 

 with the difficulties of drainage or pumping. With great facilities, favour- 

 able material, and not too much hurrying, the same area of tunnel has lieen 

 driven for so little as 20/. per yard forward. In round numbers, and on an 

 average, the sectional area of the ordinary tunnel for a double bne of rail- 

 way, to be worked by locomotive engines, may be called 50 superficial yards 

 when finished, or within the ring of brickwork or masonry, if lining were 

 required ; in this latter case, the sectional area of the opening to be excavate 1 

 may be assumed as about SO superficial yards. Mr. Vignoles observed that, 

 for future tunnel operations, with the benefit of past errors and experience, 

 by avoiding undue haste in execution, and with sufficient caution and activity. 

 40/. per yard forward for tunnelling may be taken as an average approximate 

 fair price. Now if it were wished to compare this expense of tunnelling 

 with the cost of open cutting, the Professor observed that, from his expe- 

 rience, which had been very considerable, in removing earth in large 

 quantities, he was not disposed to put a less price than that of Is. per cubic 

 yard for removing material for deep excavations, especially when this price 

 is to cover contingencies of slips. Sec. ; with such a price, then, an open 

 cutting 55 ft. deep, roadway 24 ft. wide, and soil requiring slopes of two 

 horizontal to one perpendicular, would give a sectional area of 800 yards— 

 that is, the expense (in estimate) would be the same as that of tunnelling. 

 But Mr. Vignoles observed that, in adlition, the future maintenance of the 

 tunnel should be taken into consideration, as well as whether the material 

 from the cutting could be disposed of with adv.antage ; the nature of the 

 soil, and a variety of other circumstances which he stated, all of which 

 would influence the decision. In sott rock, which would work with facility, 

 and \et stand nearly perpendicular, the depth might be very much greater 

 than 55 ft. before tunnelling would be cheaper. In such cases, a depth of 

 80 ft. and upwards had been resorted to. In chalk the proper slope to be 

 given, which was very variable, would greatly alter any elements of calcu- 

 lation, while, on the other hand, in forming tunnels through chalk, experience 

 had shown that water was the great enemy, and had entailed enormous 

 expenses. The Professor went into a great many other poiuts for comparing 

 excavations with turnelling, but they appeared too technical to be satisfac- 

 torily explained in a brief abstract such as this. 



On ViAPfCTS ..\Nn AouEDrcTS. — Mr. Vignoles next proceeded to the con- 

 sideration of viaducts and aqueducts, into w hich, he observed, a totally diffe- 

 rent set of conditions enter, the cost varying from 20/. per line.al yard to the 

 price for which no rule could possibly be laid down. Viaducts such as that 

 of which the L ndon and Greenwich Railway wholly consists, may, probably, 

 be executed for 20/. to 30/. per yard, including the foundations. Of course, 

 the foundation entered materially into tlie calculation, and where water had 

 to be crossed, largely increased the expense. In some peculiar instances, a 

 large river viaduct, or bridge, has cost as much as 200/. per lineal yard. The 

 Professor instanced a viaduct he had built over the river Kibble, at Preston, 

 for the Norih Union Railway. The length was 300 yards, the height about 

 45 feet above the water, .and the whole mass, including concrete foundations 

 (where the rock was not attained), comprised about 25,000 cubic yards ; 

 coller-dams were used for the piers, and fur one abutment. The bridge con- 

 sisied of five arches, each of 120 feet span,batiring on the face and spandrils 

 from the parapet to the impost course — roadway about 27 feet wide ; the 

 total cost, including all contingencies, was j£45,0C0, which is 150/. per yard 

 forward, or 36s. per cubic yard on the w bole solid contents ; this might be 

 considered a lo.v price, inasmuch as an ordinary brick bridge, of twenty to 

 thirty feet span only, and wiih facilities for construction, can seldom be built 

 for less than 20s. per cube yard. Where no water or expensive foundations 

 are to be encountered, and where the spans of the brick or stone arches do 

 not exceed about GO or 70 feet, Is. per cubic foot on the solid contents of the 

 viaducts may be put as a good covering price. Mr. Vignoles stated that, for 

 such viaducts, about GO/, to 70/. per lineal yard might probably be taken as 

 the average approximate cost, and the additional expense, from a consider- 

 able increase of height, does not become so very great, as it chiefly affects 

 the piers only. The Professor then enlarged much on adopting timber 

 arches, with piers of masonry, for viaducts of large span and great height, 

 and produced a number o! drawings of such bridges, some actually con- 



