A tied-arch foot-bridge

The next diagrams are based on a foot-bridge over the A11 road, near its junction with the M11 motorway. The bridge consists of a tied arch, carrying the footway, with a strut at each end, carrying steps.  Which of the diagrams below do you prefer? Why? Or would you prefer an intermediate design?  Does it make any difference to the horizontal thrust at the abutments if the deck ties the arch or does not?

A concrete arch

To enter Evesham from the south, you have to cross the river Avon (tautology) over a concrete arch. Near the abutments, the arch curves sharply downwards, to join it smoothly to vertical piers. Do you think that this is better than continuing the curve to the abutments, which would have suggested the actual flow of the thrust? Don’t confuse this with a true elliptical arch, like Brunel’s at Maidenhead.

Note how the design of the deck in the arch is continued by the design of the viaduct over the flood-plain, though the connection is obscured by numerous trees. The bridge and viaduct system is common in both the Severn and Avon valleys, which are liable to flooding, especially in spring and winter.  

The profile of the deck is very slightly raised towards the middle. The shape is segmental, the central two bays being horizontal, and the outer ones sloping. This would have made shuttering much simpler than the use of a curved profile.  

The third picture shows one of the four pillars which adorn the ends of the arch. The style is more reminiscent of a stone monument than of a concrete structure. Evesham being a town with many very old buildings, perhaps there was pressure on the designers to make the bridge "fit in".

The next set of diagrams shows three versions of an arch. Do you have a preference?

The cables can be parallel or fanned from a point, or arranged in an intermediate pattern. They can be reduced to only two in number, or even one, per side. And instead of two planes of cables, a bridge can be furnished with a single set along the centre line. There are even examples where the plane of the wires is far from vertical.

If the cables fan from the same horizontal level, they must originate from a horizontal line. However short this is, it will affect the appearance from certain angles, because the cables will generally not be coplanar. In fact, in most cable-stayed bridges, the multiplicity of sloping cables is liable to lead to a disordered appearance unless great care is taken.  

One solution is to use only one plane of cables, requiring greater torsional stiffness in the deck. Another is to use a small number of cables, requiring greater flexural stiffness, and assembly in larger sections.

Solving the engineering problems may not be considered enough. In the left picture above, the view from the road shows a somewhat disordered appearance of the cables. This can be even worse if the cables are fanned out from a horizontal row of holes.

To achieve a vertical plane of cables, the second arrangement can be used. Another solution is to abandon the idea of a vertical plane and make an A-frame, which is very rigid.

A third way is to use only a single plane of cables, relying on the deck to provide stiffness against torsion.

Whereas the towers of suspension bridges are generally based on towers with vertical members, those of cable-stayed bridges have been given a variety of treatments.

Does it matter what bridges look like? Does it matter what anything looks like? Apart from the people, a town is made of all the buildings, roads, street furniture, bridges, and street furniture. Perhaps no one of these makes much difference, but the whole is definitely made of its parts, and must surely affect the way that some people feel and think. Nobody would build a town so that tourists would like it centuries later, but there is no doubt that many people visit cities and towns because they enjoy being in them.

What would happen if a town were built entirely of structures that had been designed to look good, and to go well with each other? Would we just get used to it and raise our expectations, or simply not notice it, or would we really find it pleasant?

If you drive from London to the other end of the M1 motorway you will see a summary of the history of motorways and motorway bridges in England. You can see very clearly the changes that have occurred in attitudes to design and construction since the early days.

Appearance of Arches

Arches are in principle relatively simple. Nevertheless they are capable of tremendous variation in style, from the heavy Roman type with wide piers, to thin deck-stiffened concrete types.

As in any other type, designers may be tempted to test the extremes, in flatness, slenderness, span, and so on. At the extremes it is perhaps more difficult to achieve good proportion than in a more average structure. How flat can an arch be made before the viewer feels a sense of strain? Is this feeling influenced by familiarity with multi-arch bridges for river crossings in towns?  How far can cornes de vache be taken without looking wrong? What about the proportions of arch, spandrel walls, and deck in a concrete arch?

Some large steel arches, like the Hell Gate bridge and the Sydney Harbour bridge, have massive masonry pylons at each end, marking the transition between the main span and the approach viaducts. The Bayonne bridge does perfectly well without them, but does it really look right? Don’t we need to see some visible terminator? Perhaps the problem with this type of bridge is that although the lower chord carries all the thrust at the abutments, the presence of the upper chord may create the illusion that the trust is distributed. Some steel arches, such as the Garabit bridge, narrow to hinges at the abutments, avoiding the visual ambiguity. The question here is whether the appearance of a structure should "look right" to everyone, or only to experts who understand it.

There is another possible point in the pylons. Medieval flying buttresses often carry small spires, which are supposedly there to push the funicular downward and keep it in the masonry. When someone is up a ladder, they often have someone at the bottom to increase the friction and therefore reduce the tendency to slide. Can it be that the pylons of arch bridges unconsciously remind us of these forces?

In fact, the combination of main spans with side spans is a difficult issue, because the engineering requirements are so different at different places in the structure, yet the whole structure should look as though all the parts belong to each other. Arches, trusses, cantilevers, suspension bridges and cable-stayed spans often have smaller spans on the approaches. The pylons mentioned above may act as "punctuation marks" or separators between segments that are essentially different in character.

And the problem isn’t confined to the type of span. The details of tje main span and approach spans can clash, as in the case of the Forth Road bridge.

Here is an arch built in 2002. What do you think?

Appearance of Beam Bridges

These pictures show a seaside pier that leads to a swimming pool and a cafe. The sections are all haunched beams, giving the appearance of shallow arches, which is clearly belied in the last picture. Do you think this termination looks good? The beams, of rectangular section, rest on piers of elliptical section. Does this work? See also footbridges.

Appearance of Cantilever Bridges

The design and appearance of a cantilever bridge is inevitably affected by the presence of the three essential parts – the two cantilevers and the suspended span. The Queensborough bridge has no suspended span, but it does not look better for it. The alternatives are to make the structure plain by giving the suspended span its optimal structure, and to disguise the structure by integrating the shape of the suspended span into the overall design. The second technique is usable with the smaller bridges, but in the larger sizes, it is impractical. This subject is discussed in more detail in the page about cantilevers, where we see that the presence of the two types of structure can even be used to produce unusual designs.

Here is an example where the shape of the suspended span has been integrated into the overall shape.

The designer has attempted something very difficult – the creation of a convincing single curve for the entire span. It is a matter of opinion as to whether a given example is successful, but it would be difficult to find a better example than this. Several bridges of this design are found in Gloucestershire and Wiltshire.

In this type of bridge the design of the towers and the deck are bound to dominate, though approach viaducts need careful design to integrate them into the whole. Aerodynamic decks can now be so narrow that the towers must not appear too massive. Although D B Steinman hinted at Gothic design in some bridge towers, later designs have generally been quite simple, though in the 1990s there has been in the UK some tendency to unusual designs in small foot-bridges.

Reaction to the Tacoma Narrows crash had a long lasting effect on deck design, which was reversed only when the aerodynamic deck was introduced. Even then, many designers persisted with trusses. Certainly, for a foot-bridge, not many people want them to move detectably.

Fritz Leonhardt’s book on bridges is well worth reading for his observations on the aesthetics of bridge design.

The truss has great potential for untidiness, or even ugliness, simply because of the large number of parts that point in many different directions. Trussed bridges usually take the form of beams, cantilevers and arches. These two bridges use square section tubes, welded together, creating a tidy appearance.

This ugly truss bridge was built in order to widen a road, and stands next to the original road bridge. It is impossible to view it properly, so its appearance hardly matters. When it was built, cost was probably a major consideration.

The best and simplest view of a truss is usually from the side at a large distance. The action of the structure then becomes plain to see. In this forty year old picture of the Jacques Cartier bridge, the viewpoint is perpendicular to the right hand end of the bridge, and we can clearly see the K-type frames, reminiscent of the Quebec bridge. But the view of the left hand end is far enough from a right angle that the picture is becoming confused. The Forth railway bridge is an excellent example of this idea – it looks very simple indeed when viewed exactly from the side. For trusses that are not too large, a good plan is to use members based on square section tubes, and to use as few as possible. These examples illustrate the idea well. 

In larger bridges, some of the main members are often trusses in themselves, almost as if the structure were starting out on the road to becoming a fractal, though the main compression members of the Forth railway bridge and the Royal Albert bridge are tubes.

This picture of the Quebec bridge hints at the complicated views that can be seen with trusses. Crossing a suspension bridge or a cable-stayed bridge, you have a good idea what it is doing: here you see only an apparent jumble of girders.

From the very big we turn to the very small. Sometimes it is easier to learn from small things: looking at big things, we may be so over-awed that we forget to notice the details.

This bridge illustrates the difficulty of achieving an orderly appearance when there are numerous piers. The problem arises because there are two rows of piers. What solutions can you think of?  The fences are inclined inwards for no obvious reason. Does this improve the bridge?

The Palisades Interstate Parkway was designed to blend as well as possible with the very attractive locality. The layout of the roadways, and the designs for the junctions and bridges was very carefully considered.

Whatever you think about bridges and their appearance, even a slight acquaintance with the subject will reveal that the engineer is not merely plugging numbers into formulae that generate beams, trusses, suspension spans, etc. He or she is creating a work which is fitted for the site and the purpose. True, some companies sell standard small bridges: true, Freyssinet produced a "limited edition" of the elegant Marne bridges. But many artists, for example, Henry Moore, have produced limited editions, some of which are quite large, and we don’t think less of Song dynasty porcelain because each artefact was worked on by as many as thirty people on a production line, each person performing his own task, or because large numbers of each product were made. Woodblock pictures by Japanese artists were often issued in large quantities, and they required the skills of woodblock cutters and printers as well as those of the artist whose names are usually the only ones we know. The drawings and the final prints of Hiroshige, for example, show how much work was done by these other men. Many modern artefacts, such as cars and consumer goods, are made in vast quantities, and by many workers, but they can still be regarded as attractive.

Bridge engineering is at one place on a spectrum that includes art and science, craft and music. There is plenty of room for human expression, so long as it is tempered to the needs of the users. Compare Telford, Brunel and Stephenson, or Steinman and Ammann. Look at the work of Calatrava and Muller, to mention just a few names.

Some bridges look very geometrical, but as with many natural forms, the apparent mathematics arises from the physical and engineering requirements. (See Nature’s Maths) As you can see in the page about Beams, even a beam, the simplest looking form of bridge, conceals a range of subtle stresses that cannot be imagined by looking at it. So the purist argument that structures must always show what they are doing, like any other argument, becomes absurd if taken beyond a certain point. Equally, actions like using pebble-dashed concrete, or trying to disguise completely the nature of a structure, probably represent undesirable excursions in the other direction.  

Finally, a determination to be original or to create the biggest or the longest, if it becomes too dominant, can over-ride the requirements of utility or even safety. There are a few big structures which carry only a small fraction of an economic or useful traffic load.

An unusual bridge

This is Peto’s bridge in Bristol. It opens by rolling on arcs and tracks with large teeth that prevent slipping. The two large objects that look a bit like funnels are the counterweights, which probably have drainage at the bottom to prevent rainwater changing their weight. Perhaps they are intended to look like lilies or ear-trumpets. The movement is controlled by a single hydraulic cylinder and piston. Note also the shapes of the piers.

The design of this bridge seems to be as much sculptural as functional – a bit of fun perhaps.

A bridge in Oxford

Here is a bridge in north Oxford. The bridge and its ramps are quite neatly designed, but the undulating decoration seems quite unrelated to rest of the bridge: it doesn’t even have a wavelength that matches the truss panels.  What does it add to the design?

A bridge at Ely

When you see a sign saying "To the River", it’s natural to go and take a look. Little children may hope to feed the ducks: adults may want to hire a boat or take a walk: dogs may want to swim. At Ely, you will find the view bounded by the trussed railway bridge on the right, and a concrete beam bridge on the left, shown in the two pictures. The bridge is not exactly ugly, but it is not very elegant either, though it does the job of providing access to some boat houses and marinas. Ely is a very attractive town, very popular with tourists, and not only for the beautiful cathedral. The builders of the old railway bridge could not have been expected to consider an as yet non-existent tourist industry.

But what about the present? Should everyone who builds be obliged to make their constructions "fit in", perhaps as Tower Bridge supposedly fits in with the Tower of London?  At Lechlade, on the Thames, a proposal for a footbridge using plastics was rejected as being "unsuitable" for the location.

Appearances are not always as planned 

Many a structure that looked so good in the artist’s impression, depicting the scene on a sunny spring day, with imaginary trees in leaf, later looks dowdy, as a result of fading or peeling paint, wrinkled or cracked surfaces, dirt, corrosion, creep, and other forms of decay.

The gasholder shown in the previous set of pictures has been painted since this page was begun, but the water used to seal the holder is already affecting the colour, as we see in the third picture.

Sometimes the changes in appearance are deliberate. The styles used for this type of addition appear to be similar in many different countries.

Cemeteries are good places to see the effects of weathering on stone. In this example we can see where the flow of water has affected the growth of lichens. But why is the flow apparently far from the vertical, as we see from the building behind.

Geological processes do not stop just because we have decided to use the rock. Here we see exfoliation making inroads into the stone.

More about the appearance of bridges can be found in Town Bridges and in Foot-Bridges

A Multi-span Beam

The A417/A419 from the M4 to the M5 provides a dual carriageway for most of the route. A tremendous effort has been made throughout in order to create an attractive route. Some farm land had to be purchased, and some old Cotswold walls had to come down.  Some were already old and crumbling. In return, the road builders have preserved a well-known wind break near Duntisbourne, they have planted large numbers of saplings, and they have built miles of new Cotswold walls.

The designs of the new bridges are as attractive as possible, given the technical requirements, and the whole route is a good example of modern construction.

Near Baunton in Gloucestershire, an interesting beam bridge crosses the valley of the river Churn, north of Cirencester, carrying the A417(T). This part of the route is a bypass of Cirencester, which relieved that town of through traffic, and took as much as fifteen minutes off typical journeys. The pictures below show this bridge from below. The line of the road is slightly curved in both dimensions, and the bridge is a continuous beam.

Crossing this valley, including the tiny river Churn, presented the problem of appearance. The actual construction tends to obscure the view of the valley from the road. Let’s look at this in some detail.

Firstly, the valley is barely visible from the road in most of the route, because of trees and hedgerows. That it is visible here may be partly due to the effects of construction work. Within a few years, trees will have grown up to hide the valley from the road. So it could be argued that the view doesn’t matter very much.

Secondly, what is the intention when building such a bridge? If we want to make the bridge as inconspicuous as possible, we can build it on very narrow piers. If this had been done in this case, replacing each pier by two narrow ones, making four at each position, there would have been a good view through the bridge, but the effect of all these piers would have been very untidy. Obtaining an ordered appearance from all directions is an almost insoluble problem with a two-dimensional array of piers. The other difficulty would have been the avoidance of a top-heavy appearance.

Another possible solution would have been to have one pier at each location, on the centre line of the road. The very wide beam would have had to be very stiff indeed to carry torsional forces to the abutments, increasing the cost of the structure. Alternatively, these single piers could have been splayed at the top to provide paths for the asymmetrical forces.

At the other extreme, solid walls would have produced a very heavy effect. The actual design balances the mass of the beam and the piers very well, and the hints of gothic curves echo those of the nearby Cirencester church and Gloucester cathedral. At the top of the piers, the outer faces are not quite vertical: evidently someone took great care with the appearance of the bridge.  See picture below.

The view across the valley is more or less blocked, but was not especially inspiring in any case.

The river Churn, which lies at the bottom of the valley, is only about two metres wide at this point. The picture shows it a few miles downstream as it reaches Cirencester. Fairly soon after Cirencester it is joined by the little river Thames. The names Churn and Cirencester probably echo the Roman Corinium, an important fortified town. The A417/A419, apart from the new bypasses of Cirencester and Latton, follows the old Roman road, later called Ermine Street, quite closely.  Comparing a map of Roman roads with a modern map shows that the routes of many Roman roads can still be traced via modern roads, tracks and footpaths. The influence of the Romans on our language is also enormous. On the road we may come across words such as bus (omnibus), car, engine, entry, exit, logistics, traffic, trailer, transport, and vehicle, all derived from Latin. The Corinium museum in Cirencester offers an excellent insight to Roman life in Britain.

Just south east of this bridge, a minor road, the Whiteway, crosses the bypass. At this point, as on the other side of the Churn valley, the A417 is in a cutting, to reduce the gradient down to the Churn bridge, and to reduce the height of that bridge. The typical bridge in this situation would have been a simple concrete beam or cantilever type, but what we see is apparently a flat arch, faced attractively with Cotswold stone. But as we pass underneath, we see that the arch is only a facade, and that the bridge is a concrete beam. In fact, in the picture, under the right hand side of the bridge, you can see a part of the sloping haunch.

Is it a fraud? Well, it looks a lot better in that location than a grey concrete bridge would have done. It avoids the concrete piers that would have blocked the view of the cutting, and it generates a lot less thrust than a real flat arch would have produced.

This picture, looking north from the small over bridge, shows the deck of the big beam in the distance. We can see the slight curvature of the bridge. On each side of the road, on both sides of the bridge, the Jurassic limestone is cut into steps, to improve the appearance of the cutting, and to provide footholds for the growth of vegetation, as the second picture shows. No doubt the fissures in the rocks will provide shelter for a wide range of invertebrate and vertebrate life. Kestrels are a common sight over verges, and the occasional buzzard is seen on the fences on days when thermals are not forthcoming.

These are just a few examples of the thinking that lies behind the A417/A419 link road.

A Beam Bridge in Leamington Spa

These two pictures show a beam bridge in Leamington Spa. It crosses the river Leam just above a weir, where the flow is slow and the water is shallow. It was therefore possible to use a large number of piers. Note the decoration on the side of the bridge, and the cutwaters at the base of the piers, which are rather pointless, given that the inner piers do not have them. Concrete is not very forgiving, especially on a small scale, seen from a short distance, and with expanses of bare flat surface. This bridge is quite old: modern treatments are usually more sensitive in terms of surface texture and colour.  

It illustrates the difficulty of breathing life into a structure when it is very low. The tall railway arches behind look far more interesting: they were needed because the railway was of course built at the level of the surrounding land. In any railway town that is in a valley, you are likely to find bridges like this.  

The arch bridge is an example of extreme post-tensioning or reinforcement; long bars have been passed through and secured with large end caps and nuts. This technique is not uncommon in older structures that have started to crack.

Why do you think that the piers of the beam bridge were tied together. Could it have been because the piers had to go deep in order to find good ground under the alluvium?  

If you have bridges in your area, it can be interesting to try to guess what the problems were, what other solutions might have been available, and why they were not chosen.

The leader or leaders of a country or an organization may wish to support the building of edifices which will impress the populace or the leaders of other countries or organizations. "Look on my works, ye mighty, and despair."

The designers may want to make their mark in their chosen field, and to add the proposed construction to their CVs.

The designers may wish to display "originality".  Such a person may be "inspired" by a Tang vase, and will seek justify the choice of the shape for a building by such ideas as putting the lifts in the "handles", and to use the rim for the track that supports the window cleaners’ cradles.

The designers may wish to design something that will be completely satisfactory to the users, with no thought of an outstanding appearance, or of their own fame.

When you are looking around a town or a city, do you ever look at buildings and other structures and wonder where their appearance came from?

Example – many of the wooden buildings in the pleasant and interesting town of Zakopane in Poland look as though they may have been there for centuries, having been built in a "traditional" style. The style was, in fact, created by Stanislaw Witkiewicz during the 19th century.

Appearance Part Two

500 Bridge Photographs

BridgeScape: The Art of Designing Bridges – Frederick Gottemoeller

Wiley Europe – ISBN 0-471-29296-6

ABCD Pittsburgh bridge aesthetics

Civil engineering education

HNTB bridge aesthetics

Iowa bridge aesthetics

Paint colour

Toronto University bridge aesthetics

Replacement of the Waldo-Hancock bridge

The three myths of bridge aesthetics – very interesting paper

Trinidad Goddard South bridge

Answers – not necessarily "right" –

Arched foot-bridge over motorway M42 – The actual bridge looks more like the lower one, which is based on circular arcs.  Perhaps this made the shuttering cheaper.  The upper diagram was drawn using two parabolas.

This is a different foot-bridge over the M42, further south.  It resembles the less curved of the two diagrams shown before.  Do you think that it is more nearly circular or parabolic?

Tied-arch foot-bridge over major road A11 – The actual bridge is like the one at the top, with straight legs . This is technically more correct, and it probably makes the steps simpler to design and assemble.

Evesham Abbey bridge – Given that the designers were probably trying to produce a design that would soften the effect of a concrete structure near a historic town, modifying the line of the arch and adding pillars may have seemed reasonable.  The line of the arch would only be noticed, if at all, by people in boats or on the river bank.

Box-girder beam – It is probably more expensive to make tapered piers.  It is not always a good idea to make the piers and the deck with the same thickness.  This can look boring and lifeless.  Even if the thicknesses come out the same from calculations, they can be varied because both deck and piers may be hollow, giving some freedom to vary the dimensions, while retaining stiffness and strength. The ratio of the widths of the piers in the two directions can also be varied, so that they do not have a square section.

High-deck arch – In the first diagram the stiffness is provided by the arch.  In the third it is provided by the deck, leaving the arch to take only the thrust.  In the middle diagram the appearance does not reveal the function.