Clifton bridge ===
Clifton suspension bridge spans the Avon gorge near Bristol. Designed by Isambard Kingdom Brunel, son of Marc Isambard Brunel, also an engineer, it was not completed in the designer's lifetime. Brunel wanted the towers to have an "Egyptian" appearance, but, as in the case of the George Washington bridge, the cladding was never added. The design of the towers is excellent as they stand. The 702 foot span complements the site magnificently, and is a fitting memorial to someone who was unusual among men, and unusual among engineers. See links at end of this page. The Avon gorge, which is around 250 feet/76 m deep, inspired numerous designs, some of which were fairly absurd.
The bridge has a set of three wrought iron chains on each side, the chains being displaced vertically. Each chain is made of eyebars, in numerous parallel rows, connected by bolts, from which the hangers reach down to the bridge. To avoid the near impossibility of correct load-sharing among three connected chains, these chains are in fact independent, apart from their connection to the deck. The deck is connected cyclically to the three chains.
Brunel planned for pairs of chains, not triples, but circumstances necessitated the use of chains from the dismantled Hungerford Foot-bridge across the Thames in London. That bridge had twin chains, but a different total length.
In 2004 a newspaper reported that the operators of the bridge have started the practice of closing it during large nearby festivals, because vast crowds of pedestrians have threatened the integrity of the structure.
Some details of Clifton suspension bridge
As you approach the bridge along the road from Clifton village, the dominant features are the massive wrought iron chains and the sturdy looking towers. It is not at first apparent, but there are three chains on each side of the road, arranged one above another. Each chain contains alternately ten or eleven links, in two groups of five, the eleventh being set alone. This alternation requires a special short link at the centre of the bridge, in order to preserve symmetry, though in fact there are at least three asymmetries elsewhere. The towers differ in two respects, and third asymmetry is the massive masonry support at the Leigh Wood end.
The the three sets of chains are not in phase, which allows the hangers to be relatively closely spaced, helping with the rigidity of the deck, while keeping the bars quite long. Short bars would mean more joints and thus extra weight. Here are some pictures of the chains. The cross section of the chains is very large compared with that of the high tensile steel cables that have been used in most later suspension bridges.
Side span Main span
Centre of main span
The chains used to look very neat, but they now carry a large number of lamps to illuminate the bridge at night. The bridge does not have side spans: the chains simply pass from the towers to the ground, carrying only their own weight, which produces a slight sag. In the main span, the angle between successive links is rather small, because the tension in the links is much greater than the tension in the hangers. The ratio of the tensions is very simply related to the angle, in fact. The pictures of the centre show the ends of the links, and another view is shown below.
The ends of the bars must be wider than the bars to maintain the area of cross section that withstands the tension. The other important feature is the smooth transition form the main cross section to the wide one. Sharp corners would have resulted in high stress concentrations. The tapered cable grip found on many electrical plugs are provided for the same reason. Some overlapped and glued joints even use adhesives with different elastic properties at the centre and ends of the joints.
Compare the Clifton bars with this connection on the early 20th century locomotive 6412 of the Swanage Railway. The shapes are quite subtle. Minimization of weight in these connecting rods was vital, as they swing around with the rotation of the wheels. With modern computer programs that can perform finite element analysis and function minimising, very efficient shapes can now be created. These are, of course, no better than the assumptions built into the programs, and older designs may well have included a greater safety factor simply because the designers, realising their inability to make exact calculations, probably played safe. The occasional things that broke showed when the theory or practice was inadueqate.
The chains are anchored deep in the rock in cavities that are narrower at the bridge end than at the remote end. These cavities were filled with brickwork after the cables were fitted, to provide a plug that cannot be pulled out. The pictures below remind us of the need to study the local geology, and the details of the rocks, to be sure that they are capable of doing the job.
The construction of the chains reminds us of the recurring problem of dealing with multiple components. In principle, ten or eleven bars are better than one bar, ans so they are in practice as well. But there are problems. Suppose the bars are not identical - what then? If one bar is longer than the rest, it may not take any of the tension. If one is shorter than the rest, it may take it all. The answer is obvious, make them all the same.
But that is not as easy as it looks. "The same" is a phrase that hides the problems. Equality obtainable in mathematics or digital computers, but not often in the material world. Since the only thing that matters in the case of the eyebars is the distance between the holes, a jig may be made to force the drilling to be done at a fixed distance apart. What if the drill slowly wears and changes in diameter? What if the temperature changes between drillings? This might not matter if the jig and the bars expand at the same rate, as long as they are at the same temperature at any moment.
Alternatively, a set of bars can be drilled together, using a long drill, ensuring that the drills at the two ends are parallel.
What is the acceptable limit on the variation in distance between the holes? The important point is the variation of stress with strain. If the material is very stiff, the stress will vary strongly with distance, whereas a set of rubber bands will share the load quite well.
As you approach either of the towers, you see a notice advising of the existence of the Samaritans, who listen to people in strict confidence. The reason for the notices is the use of the bridge by people who have decided to end their lives. Falling from the bridge is not guaranteed to be fatal: a few people survive, with terrible and incurable injuries. The bridge has been fitted with a high fence and five steel wires on each side, making exit very difficult. These were designed to have the smallest possible effect on the appearance of the bridge.
Returning to the subject of the chains, these are connected to the deck by means of vertical hangers. Each hanger is in two parts, joined by a turn-buckle, which has a left-hand thread at one end and a right-hand thread at the other. Lengthening or shortening of the hanger is a simple matter of rotating the turn-buckle. By this means, each hanger can be adjusted to the correct tension. When the tension is correct, the angles between successive links forms a smooth series. The pictures below depict some of the shortest hangers.
The second picture illustrates the principle of spreading loads as smoothly as possible, just as in the eyebars. The spreader is connected to one of the two main girders of the deck. The third and fourth pictures show spreaders in compression, holding up the roof of old buildings. The hangers are pivoted at top and bottom, allowing movement in the plane of all the hangers. Were they fixed, the effect of live loads at the fixed joints would be to induce bending stresses in the members: the hangers, being the thinnest parts, would deflect the most.
The deck is stiffened by a lattice truss on each side, somewhat obscured by the anti-jump fence. Stiffening is required both against the effects of live loads and against the effects of wind.
Leaving the bridge on the Clifton side, we can look across at the massive block of masonry that supports the Leigh Wood tower. The reason for this huge and very expensive construction is not now very obvious, but in 1829, when the first designs were put forward, the proposed span was considered quite daring, and the masonry support reduced the gap significantly. Choosing among the designs proved to be very contentious, and was marred by an unhappy episode of Thomas Telford, one of Britain's greatest engineers of all time. He was by then very old, and after a stupendously active and successful career, he was probably worn out. In the event, he submitted some designs that were the subject of ridicule. His only other failure (only a partial one) was at the Over bridge near Gloucester, where, against his own wishes, he had been forced to build a masonry bridge on unsuitable ground, instead of his standard iron design. Here is a picture of the western tower and support. A bonus of the design is that with a tower on the cliff top, the western end of the bridge would have been hidden in the trees.
Finally, here are a few general pictures of the bridge.
See also Clifton Suspension Bridge.
See "The History of the Clifton Suspension Bridge" by G W Barnes and Thomas Stevens, Copyright by Clifton Suspension Bridge Trust, and "Clifton Suspension Bridge", Pitkin Guides, ISBN 0 85372 758 9.