Arches Two Part Two

Arch and Portal Frame

The diagram below takes an "arch" from a previous diagram, with three voussoirs, and adds a beam on top, to make a type of bridge with sloping piers which is quite often seen over  motorways.

Some examples of this type of bridge can be seen in the page about beams. Again there is some relationship between beam and arch. Not every bridge can be unambiguously placed into one of the basic categories.

The next picture shows a five-pinned arch being stabilised by a stiff deck. The central tie is not strictly necessary, but enables a thinner deck to be used.

WhitneyToll0273S.jpg (168546 bytes)This picture shows a part the toll bridge at Whitney-on-Wye. Would you describe this as a propped beam, a portal frame, or a four-hinge arch with stabilising beams on each side? It was built after previous spans had been washed away on several occasions.

.

Developing a  Beam into a Truss and an Arch

At the bottom of the picture the diagram represents a simple plate girder. In the next diagram some attempt has been made to shape it to suit the bending moments. In the third diagram this is taken further, and in the fourth picture the structure is greatly lightened by changing it into a truss. Finally, at the top, we see a tied arch or bowstring arch.

The point is to get the material as far from the neutral axis as possible in order to oppose the bending moment. Material near the neutral axis isn’t doing anything useful in this context. For a tension member, of course, you might as well use a wire as a tube, unless the member is very long and in danger of vibrating.

.

A  Very  Peculiar  Bridge

What is wrong with this design?  It is a three-pinned arch with a haunched beam on top.

.

A Millennium Arch

M42TiedW.jpg (29133 bytes)This beautifully simple tied arch bridge was completed in the year 2000. It takes an A road over the M42 motorway south of Birmingham. The span was built on land to the left of the picture, and was moved across to its final position during one night, minimising disruption to traffic. This possibility, together with the absence of outward thrust at the abutments, makes the tied arch very attractive for this type of location. This is one of the most elegant bridges depicted in this web-site.  

A tremendously useful feature of the tied arch is that there is no external horizontal thrust. This makes the construction possible in situations where a normal arch would not be a suitable design. It also means that the arch can be floated or slid into position in one piece. In fact, it acts as a beam. So if the tie replaces the ground, as a source of reaction to the thrust, what does this tell you about the stresses in the ground under an arch.

M42TiedV.jpg (82301 bytes)So here is a much larger copy of the  picture, unfortunately taken on a dull wet day. Note the great disparity between the thin hangers and the stout arch, as is usual for tension and compression members. This is a true millennium bridge.

EveshamA1.jpg (57013 bytes)EveshamA1V.jpg (51995 bytes)EveshamA1A.jpg (70280 bytes)Here is another tied arch, which you cross as you enter Evesham from the Cheltenham road. The Abbey bridge, across the Avon, is made of reinforced concrete. The soffits curve down sharply at the abutments. Do you think this is a good idea? Does the curve correspond to any forces?  

Many bridges in this region have approach viaducts across the flood plains. The design of these approach viaducts neatly echoes that of the main deck, though the effect is not easily apparent because of the many trees. But the pillars at the ends of the arch suggest an uneasy compromise with a type of design which would once have been used for a masonry arch.

The pictures below are of Evesham’s other Avon bridge, a fine structure dating from 1856.

EveshamB1.jpg (58569 bytes) EveshamB2.jpg (59247 bytes) EveshamB3.jpg (88633 bytes) EveshamB4.jpg (79231 bytes)

This bridge does not at first sight look especially noteworthy, being like many others in England. But look at the decoration and the balustrade – just enough to add interest without over-doing anything. And look at the piers – how narrow they are compared with the Roman ones. By this time, engineers had realised that all the horizontal thrust from any arch in a multi-span bridge need not be carried by its piers – it could be carried over through the other arches to the abutments. Of course, an excessive load on any one arch could in principle collapse the bridge.

The carrying through of loads is an elegant solution to many problems. Another type of through load is seen in mutlt-span beam bridges that rest on narrow piers under the centre-line. Torsional forces are carried through to the abutments by means of the stiffness of the deck. An excellent curved example is seen where the M42 motorway joins the M5 motorway south west of Birmingham.

EveshamC.jpg (83271 bytes)This little bridge in Evesham is noteworthy for the interesting treatment of the railings. Is it an arch, or is it a beam?

Here is a very pleasant footbridge built in Cheltenham in 2002.

.

Hollow Spandrels

This bridge is one of two at junction 18 of the M4, which connects that motorway with the A46.

It is not always necessary for the space between arch and deck to be filled in. One of Robert Maillart’s early reinforced concrete bridges developed cracks. Instead trying to make stronger designs he realised that the cracks were indicating that some of the material wasn’t doing what he thought it was doing, and in fact, wasn’t needed.

Like many other apparently lucky discoveries, this one came to a prepared mind. As a result he was able to design some beautiful bridges with a lot less concrete. Although the load is carried to the arch only around the centre of the arch, this is not reflected in the shape, which is a continuous curve. His designs have been very influential, and even today you can see bridges being built which owe much to him.

VintgarArch.jpg (432065 bytes)The next example, in the spectacular Vintgar gorge in Slovenija, is a high arch spanning the top of the deep gorge. A narrow, deep valley presents a superb opportunity for the bridge builder to produce a design that will complement the landscape. This picture exemplifies the difficulty and cost of providing transportation in mountainous terrain.

.

Stiffening Arches

The deck of a bridge like this must be stiff enough to carry the live loads between the spandrel walls. In fact, precisely because the deck is connected to the arch by rigid spandrel walls, we could imagine a bridge with a deck so stiff that the arch could be made quite thin. In such a deck-stiffened arch, the arch need only be thick enough to take the thrust without crumbling or buckling. So a bridge like this one could have a thick deck and a thin arch. The arch can then be quite light. It can also be made in straight sections. 

The deck can also be light, because it can be made in cellular form. It is obviously easier to design and build ribs or boxes for a straight deck than for a curved arch. The deck can be lighter than a simple beam, because its weight is taken by the arch. This is one of the many examples of separating the material of a structure and putting it where it is most useful.  (Lune Arch)

We also need to look at the appearance of an arch. If we make either the deck or the arch much thicker than the other part, it becomes obvious which is doing what. But if both look similar, there is confusion. The overall effect may also be rather uninteresting.

 

The way that the line of thrust behaves in a simple deck-stiffened arch can be seen by looking at this download. (Choose Run from Current Location.) It simulates loads with random weights moving with random speeds. In contrast with the normal self-stiffened arch the arch has no ability to withstand the bending moment: this is absorbed by the beam at the top.

To avoid any hint of tension in an arch, the line of thrust should lie within the middle third of the section at all times.

We also see why the suspension bridge is so difficult to build for railways. The cable has no stiffness, and so the deck must provide it. To ask that of a 1000 metre deck is asking a great deal; it would have to be very heavy.

The deck-stiffened arch is a beautiful example of the benefits that can accrue from separating different functions into different components. The opposite is often true, as in the wing box used as a fuel tank. But nobody has made much progress with aircraft which are all wing and no fuselage. Nor has the animal world, apart from gliding snakes.

If we compare the deck-stiffened arch with the beam we can see how the benefits arise, (considering only static loads).

Straight beam

No horizontal thrust at supports

Very strong horizontal tension and compression within beam

Strong bending moment within beam

Simple arch

Strong horizontal thrust at supports

No tension within arch

No bending moment within arch

Arch must be thick enough to resist buckling

Deck-stiffened arch

Strong horizontal thrust at supports

No tension within arch

No bending moment within arch

Arch need only be thick enough to withstand compression

Only small forces in short deck sections

No overall bending moment in deck

.

Loyn bridge over the river Lune

LoynBig.jpg (257053 bytes)LoynBigX.jpg (363867 bytes)This fine bridge spans the river Lune in three arches. The simple design gives a monumental effect. The piers are extended by pointed cutwaters, to steer the water around them. The line of the cutwaters reaches the top of the bridge, where they provide refuges for pedestrians when vehicles cross the bridge.  

Like many bridges in this area, the Loyn bridge has a heavy stone pavement around the piers, to counteract the possibility of scouring in time of spate, when the flow of water down from the fells is very powerful. Remember that a cubic metre of water weighs a tonne, and at even 15 km per hour, it can exert strong forces. The effect of turbulence is to cause local variations in pressure and velocity, causing vibrations in immersed structures. These variations, added to the aerofoil effect over curved stones, and the effect of Archimedes principle, enables rushing water to move and lift large objects. The lifting power increases as a large power of the speed of the water.  The next pictures show the bridge in time of flood, though nowhere near the highest that can occur.

LoynNose.jpg (101461 bytes)LoynCutwaterA2.jpg (109740 bytes)LoynFlowA.jpg (137495 bytes)LoynArchLP2.jpg (177342 bytes)The first picture shows the turbulence induced by the nose of a cutwater of the Loyn bridge, in water that is already swirling at high speed. The second picture shows the flow of the water as it enters the central arch. Since the piers of the arches reduce the width available for the flow, you might expect that the water would pile up.  But no, it slopes down along the nose of the cutwater, and it drops again as it enters the arch. Hydrodynamics, like aerodynamics, is not a subject that you can work out using "common sense".  The third picture follows the flow right under a span, and out the other side, while the fourth picture shows the view from the downstream side. Note the turbulence downstream of the bridge.  

Note also the double arrays of voussoirs and the slight angle at the crown. This may have been produced by spreading of the central arch, which is off the picture to the right. This bridge is thought to be of late 17th century construction. Like many old bridges, this one has refuges which are continuations of the cutwaters. These allow pedestrians to avoid vehicles on the bridge, which is very narrow.

Natural events affect not only people and their works, but every form of life. Events can occur on time-scales from milliseconds (an exploding meteorite) to hundreds of millions of years (motion of continents). Slow changes can stimulate evolution, medium ones can cause discomfort, fast ones can cause catastrophe. These particular floods can have a disastrous effect on the breeding of sand martins, oystercatchers, and some other birds that nest near the river.  

MartinA.jpg (96121 bytes)MartinB.jpg (149622 bytes)MartinC.jpg (142512 bytes)The pictures here show one of four sand martins – Riparia riparia – which had climbed on to the river bank to escape their flooded tunnel. They were coaxed into a tube to allow them to dry out, and when the rain had stopped, and the water level had started to fall, the tube was buried in the bank, to make an imitation tunnel.

LoynMartinsBigSX.jpg (182411 bytes)The position of the person at the right of the picture shows roughly the extent of the flood  on the previous day. Note the piling up of the water at the two piers.

LunePrC.jpg (123938 bytes)LoynPrB.jpg (184356 bytes)LunePrA.jpg (105073 bytes)GWNest2.jpg (107952 bytes)Not far from the Loyn bridge stands a private bridge of the early 19th century, also over the Lune. Its two main arches are ribbed, and a nice touch is the use of a slightly different colour for the stone above the voussoirs. In a small niche in one wall of the arch – last picture – a grey wagtail – Motacilla cineria – had built a nest. Judging by the quantity of material, the niche had probably been used before. The niche harbours at least one growing plant, which helps to disguise the nest.

StoneMarkA2.jpg (58298 bytes)StoneMarkB.jpg (89550 bytes)Not only do you sometimes get free plants and birds with bridges – you sometimes get free abstract art. But it isn’t art until an artist defines it as art. Or is it?

Bruges

Bruges.jpg (35076 bytes) BrugesQT.jpg (75313 bytes) BrugesTG.jpg (103576 bytes) BrugesTH.jpg (82715 bytes) BrugesSF.jpg (94021 bytes)

Bruges has many bridges and arches in buildings.

Les Ponts de la Caille (Quail)

CailleBigHF.jpg (225080 bytes) CailleArch.jpg (68913 bytes) UsseX.jpg (102989 bytes)

At Cruseille, in Haute-Savoie, between Annecy and Geneva, two bridges cross a deep limestone gorge in which flows the River Usse. The arch bridge was built between 1925 to 1932.  It has a span of 140 metres. See also Caille Bridges.

 

Firenze

Vecchio1.jpg (68645 bytes)  Vecchio2.jpg (53328 bytes)  BruneDome.jpg (60698 bytes)

The Ponte Vecchio needs no comment.  Brunelleschi’s dome is a work of genius. A dome is like an arch rotated about a vertical axis. Unlike an arch, it can in principle be built without centring, like an igloo. The reason is that whereas an arch is made of parallel sections, a dome comprises tapered sectors. In a globe of the earth these would be called gores. Brunelleschi’s dome, like some others, actually comprises an inner dome and an outer shell. The thickness of the inner dome is such that at every level the octagon contains a complete circle.

Roma

The "ancient Romans" were skilled, though conservative, in the construction of arches. Their piers were so wide that most of their bridges could survive the loss of a span or two by flood, scouring, or act of war. To have made great bridges and aqueducts that have survived for up to 2000 years is a tremendous achievement.  

Here is a picture of the Ponte Sant’Angelo, spanning the Tiber after more than 1800 years. The balustrade and the statues were added in the 17th century by Bernini. The original name, when the bridge was built in about 135 AD, was Pons Aelius.

Note the large platforms for the piers. The Roman builders were well aware of the dangers of scouring, and took their piling and piers down to good ground. They had a form of concrete that could set under water.

Roma2.jpg (41093 bytes)Another fine Roman bridge.

London

StampWestminsterSmall.jpg (140097 bytes)Westminster bridge in London, a type of bridge that is familiar in many a city that is divided by a river.

Karluv Most – Praha – Charles Bridge – Prague

KarluvMost.jpg (27653 bytes)  MK2.jpg (18796 bytes)

This beautiful bridge was begun in 1357 under Charles IV, who had founded the new town in 1348. Statues were added from 1706 to 1714.

Glasgow

GlasArch95.jpg (120353 bytes)LjubljanaBeamA.jpg (103161 bytes)Glasgow is a good place to see many bridges in a short distance. The Glaswegians obviously did not let the Clyde be a barrier to transport. This picture is over-exposed to show the arches in the piers. Note also the attention to detail in the ornamentation. The second picture shows a bridge in Ljubljana with similar treatment of the piers.

Bayonne Bridge

BayonneBr.jpg (37648 bytes)Completed in 1931 by Othmar Amman, the Bayonne bridge is, at 50 feet15 m shorter than the New River Gorge bridge, the second longest arch span. The span is slightly longer than that of Sydney harbour bridge. It connects New Jersey with Staten Island. This is an example of truss construction, in which most of the volume is empty space, the forces being channelled along struts and ties.

To avoid the construction of massive and expensive falsework, which would obstruct the channel, such arches are often built as cantilevers, the halves being pulled back by temporary cables. The Sydney Harbour bridge was built in this way. Just before closure, the centre of the span might look like this –

The diagram is not intended to show correctly the details of an actual bridge. After joining, the result would look like this –

Joining like this gives no control over the relative stress in the top and bottom chords. By employing jacks top and bottom, and joining the halves positively when the everything is right, the correct stresses may be obtained. The hydraulic pressure in the jacks gives an indication of the forces.

The diagrams below show another style of trussed arch.

The next diagram shows the same design with redundant members removed. See also Indeterminacy.

All these main spans with through connection to the side spans can in principle be constructed by cantilevering, keeping the navigation channel clear at all times. The next diagram shows the two sides at different stages of construction.

Now let’s colour some of the members in red for compression and blue for tension.

But what happens when the two halves meet, and we complete the top and bottom chords?  We could arrange over-size rivet holes and join the two parts, retaining the existing forces, but that is not the usual way. By jacking the bottom chords apart at the crown, the span is turned into an arch, relieving the tension at the two outer piers, and creating outward thrust at the main abutments. But there is still the top chord. That, too, can be jacked if required, until it is in compression and not tension.

Can we then confidently assign blue and red to the vertical and sloping members? What we learn from this thought experiment is that we cannot always, just by looking at a structure, know even qualitatively what all the parts are doing. In fact, it is quite possible to construct in such a way that we cannot fully understand even by calculation.  

This type of bridge can be constructed as a cantilever bridge, so that there is no arch action at the main supports. Can you see advantages or disadvantages for the arch construction or the cantilever construction?

.===

Continue