Page:Forth Bridge (1890).djvu/20

Rh occur at the junctions of cantilevers with the central girders. The viaduct must therefore expand and contract and move in every way with the cantilevers, and these movements will be considered presently.

It remains to mention the various influences to which the structure or portions of it are likely to be exposed.

1. Expansion and contraction by changes of temperature, acting in the direction of the longitudinal axis of the bridge, and to some extent also transversely upon the circular masonry piers.

2. Influence of the sun's rays to one side or the other of the structure.

3. Wind pressure, acting at right angles or nearly so to the centre line of the bridge.

Provisions for the first are made in the sliding as distinguished from the fixed bedplates, in the joints between the ends of cantilevers on Inchgarvie Pier, and in the cantilever end piers at Fife and Queensferry.

Provisions for the second and third are made in the sliding bedplates of the two outer or fixed cantilevers, and in all the joints between ends of free cantilevers and central girders. All these movements are horizontal and are controlled and confined within specified limits.

The arrangements to meet these will be described in detail further on.

The vertical deflections due to dead load, live load, and wind pressure, whether acting singly or in combination, have already to some extent been described, and will be further considered later.

Expansion joints are also provided in the approach viaduct girders upon every second pier, two spans being made continuous, the intermediate fixed joints being, however, placed on sliding bedplates identical with the movable ones.

In the two large spans of the cantilever bridge, longitudinal movements are only possible at the Inchgarvie ends of the central girders, the Fife and Queensferry ends of the girders being fixed so far as this movement is concerned. These ends therefore move with, and in the same direction as, the cantilevers upon which they are resting.

It has already been stated that the south-east circular pier of Fife, the north-east on Inchgarvie, and the north-east on Queensferry, are the fixed points of the structure. The movements due to longitudinal expansion or contraction are controlled and limited by these fixed points, and extend from them as pivots to the various extremities of the cantilevers and central girders. The lengths affected are as follows. (See Fig. 19.)

1. The Fife central tower, 145 ft., plus outer or fixed cantilever, 680 ft., total, 825 ft. in length, the expansion of which must be provided for in the north cantilever end pier.

2. Fife south or free cantilever 680 ft., plus length of north central girder 350 ft., total 1030 ft., the expansion of which will go towards Inchgarvie, while the expansion of the Inchgarvie north cantilever, 680 ft., will go towards Fife. The total amount of movement to be provided for at the Inchgarvie end of the north central girder where the two movements overlap, will be that due to the expansion of 1710 ft. of girders.

3. Inchgarvie central tower, 260 ft., plus Inchgarvie south cantilever, 680 ft., total 940 ft., the expansion of which will go towards Queensferry, while the south central girder 350 ft., and the Queensferry north cantilever 680 ft., make up a total length of 1030 ft., the expansion of which will go towards Inchgarvie. The total movement at the Inchgarvie end of the south central girder where the two expansions overlap, is that due to a total length of 940 + 1030 = 1970 ft.

4. The movement between the north piers on Queensferry and the south cantilever end pier is due to the same length as on Fife, namely 825 ft.

So far as observations up to this time have gone it would appear that the expansion or contraction amounts to about $1/100$th of an inch for each degree of temperature for every 100 ft. of girder length. The changes in temperature have, however, been so slight, and so near mean temperature that the figures cannot probably be accepted as quite correct.

Assuming for the moment their correctness, the movements would be, for 70 deg. of full range:

These figures only give about 70 per cent. of the estimated amounts, and the provision made at the four points mentioned for longitudinal movement is more than double that given above.

Soon after the contract was signed in December, 1882, a start was made with the preparatory and temporary work.

Offices and stores as well as a workshop, hereafter known as No. 1 shed, had been erected by Mr. Arrol in connection with Sir Thomas Bouch's suspension bridge, and these were taken possession of and considerably added to from time to time.

To enable the contractors to make an early start with the permanent work, that is, the building of the masonry piers, it was necessary that the positions of these should be fixed without delay. A base line about 4000 ft. in length was laid down along the high ground on the south shore, starting from a point in the centre line of the bridge, passing along the North British Railway for some distance and terminating opposite the east breakwater at Port Edgar. Along the breakwater a timber gangway was erected, and near the far end of it, at a distance of about 3000 ft. from the base line, an observatory was built. Three points on the centre line of the bridge, one on the Queensferry shore, one on Inchgarvie, and one on the Fife shore were marked down, and their distances stated by the Ordnance Survey of Great Britain. Various other stations, about twenty in number, were laid down as required, and by means of these a most careful triangulation was made and the centres of the three main piers finally fixed. In order to verify, if possible, the correctness of the distances apart of the three principal stations on the centre line of the bridge, which had been obtained by calculation based upon the triangulation, a measurement of the north span of 1700 ft. from the centre of the north circular piers on Inchgarvie to the south circular piers on Fife was made in the summer of 1884.

In a straight portion of the North British Railway a distance of 1700 ft. had been carefully measured and marked and transferred to high posts at the side of the cutting. Upon these posts notched knife-edges were placed at the two extremities. A fine steel wire about $1⁄20$ in. in thickness was laid along the span and drawn over the knife-edges with a certain amount of stress put upon it, previously agreed upon. Thus drawn up the wire left a certain amount of sag in the centre, which was carefully measured by level and noted. Two narrow copper tags were then soldered on, to mark the end points. The wire was then coiled up and kept ready for use. The temperature also was noted.

On the two shores immediately under the piers which marked the stations, places had been prepared for levels, by means of which the amount of sag in the wire could be fixed. On a calm, cloudy day, with the temperature about the same, the wire was taken across the north channel and laid down upon the prepared knife-edges on the piers, and with the same amount of stress put upon it, and with the same amount of sag allowed, the two copper tags soldered on should have coincided with the notches in the knife-edges, provided the distance was correct.