Article - Issue 18, February/March 2004
I.K. Brunel: Some thoughts on his engineering
Dr Jim Shipway
In the competition organised by the BBC last year for the title ‘The Greatest Briton’, Churchill led the field of ten entries, and the runner-up was Isambard Kingdom Brunel (1806–59), a brilliant polymath who engineered the Great Western Railway and numerous other works, including the design and construction of ships. It was astonishing that this man, little-known to the public and from a profession little understood by it, gained second place in a countrywide competition involving such names as Shakespeare, Nelson, and the Princess of Wales. How did this happen?
Brunel had a striking personality, but he was a civil engineer first and foremost, with a keen inquiring mind, immense energy, and a supreme ability to lead, inspire and persuade others; he was a brilliant innovator who at times it seemed, had an aversion to following any other man’s lead. What was his real contribution to engineering? Several books have been written about Brunel, his life and times, and in the scope of a single short article assessing his achievements, it is possible to do justice only to a very few of his works. Since he was best known as a railway engineer and a builder of bridges before he became a designer of ships, two of his later railway bridges have been chosen for comment. Brunel’s chief rival as a railway engineer was Robert Stephenson (1803–59), also a great engineer and designer of the Conwy and Menai tubular bridges, but his work lacked the flair displayed by Brunel, and he did not have the same genius for calculation.
A note on units: we generally use SI units in Ingenia, but in this historical article we have retained the original Imperial units
The Great Western Railway
Brunel was appointed engineer to the GWR in 1833 when he was 27 years old, and it was the first of his railway works, extending 118 miles between London and Bristol. It was not constructed without controversy, since it embodied the startling 7 ft gauge (50% wider than the standard gauge) and the equally startling two miles long Box tunnel, as well as the Maidenhead Bridge, which had the longest and flattest semi-elliptical arches ever constructed in brickwork. The fall of this bridge on completion was confidently predicted, but it continues in use to this day. These features of the GWR give the flavour of Brunel’s approach to engineering and of his achievements in other fields.
The development of the girder bridge in railway work took place in the UK mainly between the years 1830–1860, and included lattice girders, plate girders, tubular girders and hybrids. These years almost exactly paralleled the career of Brunel, and it was natural that he was involved in bridge development, making his own unique contribution. The plate girder bridge flourished from the 1840s onwards, but had weakness, in that the top flange was vulnerable to buckling. Brunel’s skill in design manifested itself in flanges with curved plates and web branches in several forms, stiffening the flange and the girder web together.
Large-span bridge development
In the 1840s the plated-web box girder made its appearance on a grand scale with the design of the 400 ft span Conwy tubular bridge and the Menai tubular bridge with two main spans of 460 ft, both by Robert Stephenson. These bridges, in which the trains run inside the tube, were Stephenson’s answer to the then seemingly impossible problem of bridging gaps of that magnitude. His solution was innovative and worked well (Conwy is still in use today), but the bridges were heavy and uneconomical in the use of materials, mainly due to the closed-in plated webs. Only five such bridges were built, and none after 1856.
Meanwhile, the open-web or trussed girder, with panels, was gaining ground in America, where timber suitable for bridge construction was plentiful and railway construction was booming. These girder bridges were used mainly for spans up to 150 ft, and were gradually replaced by all-iron bridges of the same types. The most popular form was the Pratt, or N-truss, and later the Warren girder. Early Pratt trusses from 1845 onwards had iron ties for diagonal members and timber verticals forming the panels, before the advent of iron. The all-iron Warren girder followed in the 1850s, with the construction of the Newark Dyke Bridge in England in 1853.
Chepstow Bridge, 1852
In 1852 Brunel was engineer to the South Wales Railway, and had to bridge the river Wye at Chepstow. The Wye had a tidal range of 40 ft, the second highest in the world, and was swift-flowing. The Admiralty insisted on a 300 ft clear width for shipping and 50 ft headroom at high tide. They further insisted that the navigation channel be closed only for the duration of one tide, i.e. for 12 hours only. Stephenson had no such problem at Conwy or Menai.
Brunel had a strong artistic streak, and most of his work, though functional, had much of elegance: for example the roofs at Paddington Station; Clifton Suspension Bridge; tunnel portals at Box and elsewhere; and his elliptical arch work at Maidenhead, Hanwell viaduct and other masonry bridges. The Chepstow Bridge, however, was a plain answer to an engineering problem, and was ungainly and lacking in charm and elegance. Nevertheless it embodied engineering of the highest order.
The site is remarkable for the contrast between the two sides of the river. On the east side there is a limestone cliff 120 ft high, and on the west there is low-lying alluvial material requiring caisson construction for the river piers. This resulted in an asymmetrical bridge of which one half was three 100 ft plate girder spans, and the other half was a 300 ft span of quite different design. The twin tracks were of the 7 ft gauge, which imposed loading 50% greater than the standard gauge.
The greatest obstacle was the maximum of 12 hours allowed for the possession of the river channel. Brunel decided on a separate bridge for each track, making for smaller pieces to be lifted into place. Each bridge was to be of three panels, each of approximately 100 ft, making a gigantic N-truss. The top chord of the truss was a 9 ft diameter tube built up from wrought-iron plating, and the web members were chain links chosen for lightness and easy handling. The vertical members of the truss were A–frames, and joined the tube to the plate girders supporting the deck. The heaviest part of each bridge was therefore the 300 ft long tube, which weighed 161 tons and spanned the entire navigation channel. Brunel’s plan was to launch the tube on barges across the channel, then lift it in stages by jacks to the required clearance in the course of a single day. The scheme worked brilliantly, and the first of the two bridges was completed in three months between April and July 1852. The second bridge was completed the following year.
But the real genius of the design lay in its structural form, in the fact that it was virtually the first open-web girder in Europe. Until that time all major iron railway bridges in Britain had closed, plated web members joining the flanges, as seen in the Conwy and Menai bridges, and Fairbairn’s patent box girder designs. But there were stirrings that the heaviness of these bridges was uneconomic, and that a framed structure, if it could be designed, would be an improvement. Thus the Warren girder with its triangular open web arrived at Newark Dyke in 1853 and slightly later at Crumlin. In the USA, advances in design, mainly in timber, had been made in the 1840s and later by Pratt, Howe, Whipple and others who developed the concept of panels in framed structures for modest spans.
But Brunel led the field in Europe. At Chepstow his gigantic, open, 100 ft panels, 50 ft high, crossed by chain diagonals, were an impressive advance, and the N-truss became established as a cheap and easily constructed structural form. Nor was the economy of the bridge compromised by high stresses. An analysis of the forces by the author in the members under the heavy 7 ft gauge loading showed stresses easily compatible with those accepted for the wrought iron construction of the time, and Brunel test-loaded a span to 65% above the designed loading with satisfactory results.
The 300 ft spans at Chepstow lasted 110 years until 1962, when they were replaced owing to the increase in loading from heavier traffic. Thus passed into history these remarkable iron spans, the first in Europe to have open-web N-truss panels. There were no others in iron outside America until J.H. Latham’s railway bridge at Allahabad, India, in 1859–65, which had spans of 210 ft, a long way short of Chepstow.
The Royal Albert Bridge, Saltash, 1859
Saltash Bridge is frequently described as the last and greatest of Brunel’s railway works, since its completion in 1859 coincided with the year of his death. If Chepstow Bridge lacked charm, the same cannot be said of Saltash, although its two main spans form what is generally avoided in architecture; an unresolved duality. This occurs when two bridge spans, or building features, are identical and side by side. The eye tends to move restlessly between them, generating a sense of unease. Whereas if a bridge has, say, three spans with the centre one slightly larger, the eye tends to rest on the centre, and is at peace.
Saltash carried the Cornwall Railway’s single line across the River Tamar in 17 approach spans and two main spans of 455 ft each, resting on a single river pier and on each bank. This arrangement was dictated by ground conditions and the depth of water in the Tamar. The foundation construction took three years and the pier was founded on rock 16 ft below the bed of the river, which is 70 ft deep. The Admiralty demanded a clear width of 400 ft and headroom of 100 ft above high water.
In 1859 the only railway bridge of similar span in Britain was Robert Stephenson’s Menai Bridge, which had two spans of 460 ft. Each span carried a single track, and weighed in the region of 1587 tons. The solid, plated, web sides accounted for about 40% of this weight.
It is unlikely that Brunel was tempted to emulate this design, and instead he created a unique bowstring truss in which a tubular arch acts in conjunction with chain ties to form a remarkably rigid braced structure. The rise of the arched tube is a shallow parabola matched exactly by the sag of the chains, the two being braced apart by vertical members with light cross-bracing between them. The bridge deck is suspended equally from the tube and the chains. The weight of this arrangement (designed to carry the heavier 7 ft gauge loading) was approximately 1069 tons, or 67% of the weight of the same Menai span. The chain ties could equally well have been rigid tension members, and indeed the use of chains led to intricacies in construction, which must have been a mixed blessing. The only other bridge of the Saltash type known to the author is a 102 ft span minor road bridge at Auchindrean, near Ullapool, in Wester Ross, and it has rigid tie members in place of chains.
The elegance of the original design of the Saltash spans arises from the shallow parabolic curves and the absence of fuss in the details of the bracing. Since they were built, however, the applied loading has increased and led to additions to the bracing both laterally and in the elevations; this has detracted somewhat from the original graceful appearance. The lightness of the trusses, incidentally, was not attained at the expense of high stresses or inadequacy in the strength. An analysis by the author of the stress in the major components of the structure (the tube and chains) shows values easily compatible with the recommended stresses of the day for wrought iron, both in compression and under tension. This remarkable bridge is still in use today, more than 140 years after its completion.
Yet, apart from the minor bridge at Auchindrean, the form of the Saltash spans has never been repeated, (the Pauli truss is a different form), and it remains a remarkable solution to a particular problem. Why is this? Structurally the bridge was a significant advance on the Menai Bridge, but in comparison the construction of the Saltash structure was complex and demanding. The arched tube was elliptical in cross-section and parabolic in elevation, and had to be built up laboriously from wrought iron plates. The chains had to be manufactured and each link tested individually. The details of the attachment of the verticals to the tube and chains involved extremely careful workmanship for lack of fit to be avoided. The temporary works, i.e. scaffolding and centring for the arch and chains, were also complicated, and had to be substantial. The connection at the ends of both tube and chains by massive pins demanded more than ordinary skills in construction. Curved work is always more expensive than straight-line fabrication, and the Saltash trusses had more than their share. Finally, the truss form did not lend itself to easy erection methods such as rolling-out or cantilevering. All these factors added to the time and cost of construction. The panel truss girder had already appeared over the horizon and offered greater simplicity and economy in building.
The Saltash design was unique and a masterpiece, but it was a blind alley and did not advance the development of the girder bridge. It remains a unique tribute to Brunel’s outstanding genius.
Brunel’s contribution to engineering
It would take several articles of this length to cover adequately Brunel’s achievements and his contribution to engineering, but some conclusions can be drawn from the limited examples given above.
At the time of his death he was a vice-president of the Institution of Civil Engineers, and he attended its meetings throughout his professional life whenever he was in London and it was possible for him to do so. He believed in sharing his knowledge and experience, and he was against the patent laws, which he felt restricted the spread of knowledge and invention.
Brunel’s greatest gift was originality of thought, seen for example in the adoption of the 7 ft gauge, and the brilliant solutions to the separate problems of bridging at Chepstow and Saltash, as well as in other fields such as ship design and the development of the screw propeller. This originality was always fully directed to improving solutions to the problems in hand. He was constrained by circumstance, and the properties of materials then available, and by the limitations of construction, but he did not allow these constraints to limit his ideas. It sometimes seemed he cared little for the expense incurred or the concern of the shareholders and directors, but he was a man who got things done, superbly.
Next to his originality of thought was his remarkable insight into the behaviour of structures. This may have been strongly intuitive, but it was a perception which was backed by wide experience and observation, and by calculation. In reading of the construction of Conwy and Menai bridges, it often seems that the calculations came after construction, which was at best an informed guess. But it was not so with Brunel. Whether he was designing the complex structure of the Saltash Bridge, or the hull of the Great Britain, or the elegant arches of the roofs at Paddington, or the timber viaducts of the Cornwall Railway, his analyses of the forces and stresses were based on a deep understanding of structural behaviour. Always, his designs were bold and imaginative, often daring. He had a remarkable confidence that his structures would behave as he predicted, shown, for example, in his design for the Maidenhead Bridge.
Yet Brunel left nothing to chance. In a letter on design to one of his assistants in 1854 he wrote, memorably, ‘always put rather an excess of material in quantity’. His works were built to last, and there was no short-cutting or skimping, or lack of care in the detailing. The sequence of erection procedures was planned in meticulous detail, with careful forethought for every eventuality. His remarkable insight resulted in few mistakes. His immense energy allowed him continuous travelling for long periods, in which he was able to supervise distant site works and so maintain the highest standards of construction.
Finally, in a private letter to his eldest son in 1858, the year before his death, Brunel lifted a corner of the veil over his soul, and wrote:
‘Finally, let me impress upon you the advantage of prayer…of this I can assure you, that I have ever, in my difficulties, prayed fervently, and that – in the end – my prayers have been, or have appeared to me, to be granted, and I have received great comfort.’
Perhaps this was a clue to Brunel’s inner life, about which he never spoke: he believed in God and trusted in Him. Isambard Kingdom Brunel may have been a runner-up in the BBC competition, but he has surely emerged as the greatest of the world’s engineers.
Brunel, I.: Life of Isambard Kingdom Brunel, civil engineer. Published origina lly 1870, Longman Green & Company, London.
Shipway, J.S.: Some aspects of the development of the girder bridge, 1820–1890. Ph.D. thesis, Heriot Watt University, Edinburgh, 2002 (unpublished).
Dr Jim Shipway
Dr Jim Shipway is a Fellow of the Institution of Civil Engineers and is a member of its Panel for Historical Engineering Works. He retired in 1998 and recently completed a PhD at Herriot-Watt University on some aspects of the development of the girder bridge. Much of his career was spent in civil engineering consultancy, including some years with Ove Arup & Partners where he led the Glasgow office and later was with the firm in Nigeria. He has had three technical papers published by the ICE, mainly engineering history and is the recipient of two awards.