On the 28 December 1879, the Tay Rail Bridge designed by Thomas Bouch collapsed in a terrible storm. The disaster claimed the lives of all 75 passengers (although only 60 bodies were found). Naturally the tragedy featured heavily in ‘The People’s Journal’ and large features on the inquests were regularly found in the paper throughout 1880. Amongst the reporting there were also letters from readers published about the events. In this letter John Roy, an engineer and architect in the United States gives his view on the reasons behind the bridge’s failure, and how it should be rebuilt.
Sir,—Pardon me for intruding upon you in the hour of your calamity. A friend sent me copies of your valuable Journal of the 10th, 17th, and 24th of January 1880, giving an account of the loss of life and destruction of the Tay Bridge at 7.16 P.M., December 28th, 1879.
The evidence shows the gale was unprecedented in that part of the world. Capt. Scott says, “In the gusts it came as high as 11, and at the time the wind was almost due west, directly at right angles to the high girders.” Eleven inches of water is equal to a pressure upon a square foot of 57 1/4 pounds, under which the wind travels 107 miles per hour. In this country the wind often blows so as a man cannot stand before it. On Mount Washington, in the State of New Hampshire, January 1878, the velocity of the wind was over 100 miles an hour during nine different days, but the highest registered is:—
|November 29th,||1875||6 P.M.||velocity,||170||miles per hour|
From the testimony of Captain Scott I infer that no long, high, and light iron bridge with a single track ought to be built at right angles to the prevailing winds and sea in an exposed position like that of the Tay Bridge. Had the Tay Bridge been built on a curve to the west, the girders forming a polygon, only one girder would be exposed at right angles to any wind, and the two adjoining girders would form a strut or tie to the strained girder; this would form an arch against the west wind and sea, and a suspension bridge to resist eastern storms. The strength of such a bridge would be in proportion to the length of the versed sign of the segment, the tensile and compressive force of the iron would be brought into action in a more favourable manner, and much of the cross or transverse strain avoided. A long train would add to its lateral strength as a brace and poise The train itself forming part of a curve, the leverage of its wheel base would be increased in proportion to the length of the versed sign of the segment.
But no amount of design can make a bridge stand with bad material and bad workmanship. I have seen many similar high, light iron bridges with only one track upon them (but generally resting upon stone piers) withstand great storms. In the State of Iowa, in the spring of 1875, the ice broke up in the river and lodged against the Keokuk Bridge piers to a height of about thirty feet (I think this bridge is about 3000 feet long, with a turngate or draw of 80 feet), sometimes a raft of timber with millions of feet in it would strike against these piers without damaging them. I attribute this strength to the weight of the stone piers.
Strange to say, the Dundee bridge was built of brittle cast-iron reduced to a minimum of weight, and the foundation was of brick, weighing only two-thirds of the beautiful stone in the neighbourhood. I remember when Dundee stone was shipped to London for the foundation of the Mint. And just to think the Dundee people do not know the value of their own materials!
I fully believe that the accident was caused by the unstaple or high girders not oscillating in time with the lower or staple girders. This derangement being continued subjected the cast-iron columns to a strain of tension greater than they could bear, and the concrete within the piers, having never been good, crumbled, and threw the track out of line, as was proved by the guard seeing the sparks of fire flying from the wheels of the cars that had passed before the ill-fated train entered the Bridge (the evidence shows this to be so with the lost train). Her wheels grinding upon the rails until some of the fish plates burst and the rail slipped, the engine left the track, came in contact with the guard rail, threw the train on the weak side of the Bridge, and most likely smashed the girder, as seen in your sketch, thus bursting the already weakened cast-iron columns, and throwing the train and unfortunate occupants into the river.
But the Bridge is broken, and I propose to show how it can be mended without reducing its height or mode or construction. I would reduce the number of high girders to three spans, and raise a strong, substantial pier of stone on the north and another on the south side of these high girders, carrying the masonry up to or a little above the upper chord of the high girders. Thus the vibration or wave of oscillation from the north and south would be brought to rest in the mass of masonry of the piers. The oscillation from the three centre girders would also terminate in the piers. This would save all that has been done on the low level Bridge, which has not only stood the storm, but arrested the force of the fallen portion of the Bridge, which has not only stood the storm, but arrested the force of the fallen portion of the Bridge. This would enable the traffic of the road to be resumed at an early day (and I would say to those that propose to lower the Bridge that it would involve a great deal of time); whereas if they consider that it is necessary to lower the Bridge for safety, the same end could be obtained, by filling in with rubble stone under and around the columns, commencing at Magdalen Green and running this embankment to the pier on the north side of the high level bridge, and the same on the south side, and when this is done it would be well to continue both embankments as far up the river as Newburgh. This would recover much valuable land, and instead of the city extending beyond Lochee for space on which to build factories, this recovered land would be available. Your Dutch neighbours have embanked out the sea, and the mighty Mississippi is embanked on both sides for nearly 1000 miles. I think your embankment could not exceed 12 miles.—I am, &c.,
John Roy, Engineer and Architect,
127 South Claiborne Street,
Orleans, Feb. 18, 1880.