Structures- Or Why Things Don't Fall Down

by J E Gordon

  What actually happened was that, during the ‘proof loading’ the vital welds which held the pressurized heating jacket together were considerably distorted – although nobody noticed it at the time. In reality these welds were so near to failure that a few reversals of stress, resulting from a much lower pressure in the jacket, sufficed to cause a fatigue failure, with disastrous consequences. This possibility should have been spotted by a competent trained engineer. In law, and perhaps in equity, the major blame lay with the people who had made the vessel; but I cannot help thinking that the danger should have been foreseen by a competent firm of chemical engineers. When I went to see these people the managing director took me out to lunch. By way of making conversation I said’ How many graduate engineers do you have in your organization, Mr—V

  ‘None, thank God!’

  On cutting holes in things

  Although it is generally rash to cut holes in an existing structure some people seem unable to resist the temptation to do so. A case in point occurred with the Master aircraft. This aeroplane was built as an advanced trainer for the R.A.F. just before the war. It had some of the performance, and many of the handling qualities, of the Hurricane and the Spitfire. In the emergency of 1940 some of the Masters were converted into operational fighters by installing six machine guns in the wings. The original trainer version of the machine had wire-operated control surfaces which, though they were perfectly satisfactory, gave a slightly ‘softer’ response than those of a real fighter. Somebody therefore decided to change over from wire to rod control linkages in the fighter version of the Master. To make room for the rods which operated the rudder and elevator, suitable slots were cut in the rear bulkhead of the fuselage.

  Before long we were faced with a series of three fatal accidents. In each case the tail had come off in flight. When we got the fuselage on the test-frame we found that its strength had been reduced to only 45 per cent of the fully factored load. The moral is, I suppose, to leave well alone.

  A much better-known accident of this type, in which a great many lives were lost, occurred with the troopship Birkenhead. This iron steamship had started life as a warship in 1846 with adequate strength and well supplied with continuous water-tight bulkheads. When she was converted into a troopship, however, the War Office insisted that very large openings should be cut in the transverse water-tight bulkheads* so as to give more light and air and more apparent space for the troops.

  In 1852 the Birkenhead was dispatched to India, by way of the Cape, with 648 persons on board, including twenty women and children. By an error of pilotage, the ship struck an isolated rock about four miles off the South African coast. The vessel was badly holed forward, and, since the bulkheads had been cut away, all the troop-decks in the forward part of the ship were flooded so quickly that many of the troops were drowned as they lay in their hammocks (the time being 2 a.m.).

  Under the weight of the incoming water the flooded fore-part of the ship broke off and sank almost immediately, leaving the survivors crowded into the after-part, which sank more slowly. It was dark, the sea was full of sharks and the life-boats were inadequate. The troops behaved with great courage and discipline, smartly fallen-in on the after-deck, while the women and children were sent ashore in such boats as there were. All of the women and children were saved but only 173 men survived: the rest were drowned or eaten by sharks.

  The most obvious effect of cutting holes in the bulkheads was, of course, that the various compartments in the ship flooded very rapidly, and this was undoubtedly the prime cause of the ship’s loss. Fewer lives might have been lost, however, if the ship had not broken in two, and this must be attributed, at least in part, to the weakening of the hull as a whole by cutting away the bulkheads on which its strength depended.

  The loss of the Birkenhead immediately became famous as an example of discipline and heroism – and deservedly so. When the news reached Berlin, the King of Prussia ordered the story to be read aloud to all the units of his army, specially paraded for the purpose. But perhaps it would have been better still if he had instructed his War Office not to interfere with the structure of ships, a matter which soldiers do not always understand.

  According to Mr K. C. Barnaby, a distinguished naval architect, the idea that open space was more important than safety in troopships lasted for many years. He says that, as late as 1882, shipowners were complaining that, when they fitted additional bulkheads as urged by the Admiralty, the trooping authorities would not accept the ships on the ground that the spaces between the bulkheads were too small.*

  On being overweight

  Almost every structure has a tendency to turn out heavier than its designer intended. This is partly due to over-optimistic estimating in the weights office, but it is also due to a tendency on the part of almost everybody to ‘play safe’ by making each part just that much thicker and heavier than is really necessary. In many people’s eyes this is a sort of virtue – a sign of honesty and integrity – and we talk of things being ‘heavily built’ as a term of praise, while ‘lightly built’ is almost synonymous with ‘flimsy’ or’shoddy’.

  Sometimes this does not matter, but there are cases where it matters very much indeed. With aircraft the weight is tending to increase all the time, from the drawing-board onwards. Extra weight naturally restricts the fuel capacity or the payload of the aeroplane, but, besides this increase in gross weight, the centre of gravity of an aeroplane somehow always manages to work its way too far aft. In other words the weight of the tail tends to increase out of proportion to that of the rest of the machine. This can be a serjous matter. If the C.G. gets too far aft, the aircraft will acquire dangerous flying characteristics. It may have a tendency to go into a spin from which it is unable to recover. For this reason a surprising number of aircraft – including some very famous ones – have gone around all their lives carrying massive lead weights permanently bolted into their noses; this is necessary in order to keep the C.G. in a tolerably safe position. It need hardly be said that this is a bad thing.

  The effects of overweight are just as bad, perhaps worse, with ships. Not only do all ship hulls tend to be overweight absolutely, but the C.G. tends, in this case, to creep, not backwards but upwards – ineluctably upwards. Now the stability of a ship, that is, her tendency to float right-side up, instead of upside-down or on her side, depends upon something called her ‘metacentric height’. This is the vertical distance between a mystic but important point called the ‘metacentre’ and her centre of gravity. For excellent reasons the metacentric height of even a large ship is likely to be quite a small distance – in fact in the region of one or two feet, perhaps less. Thus the position of the C.G. has only to rise by a matter of a few inches to reduce the metacentric height by a very significant fraction which may well imperil the safety of the ship. Various ships have capsized on launching for this reason, and no doubt the yard foremen, or whoever were responsible for the extra top-weight, considered that they were in no way to blame.

  We mentioned the loss of H.M.S. Captain in Chapter 11. The whole story of the Captain was intensely political and controversial at the time; I suppose few accidents can have had such far-reaching historical consequences. The Captain represented one turning point in the evolution of the steam battleship and perhaps in the modern concept of world power. The Admiralty have often been criticized by historians who know very little about ships for their slowness in changing from sail to steam. These are sometimes just the historians who are most critical of ‘imperialist expansion’ and so forth.

  It has to be borne in mind that, until comparatively recently, the unreliable engines, the high coal consumption and the short range of steam warships made them dependent upon bases and coaling stations and ‘colonies’ as soon as they ventured beyond home waters. The exercise of world power by steam navies is a very different sort of thing from the strategy and logistics of eighteenth-century sailing fleets. It was basically for such reasons that the British Admiralty insisted u
pon the retention of full sail power, in addition to engines, in most of their battleships almost to within living memory.

  The technical difficulty of combining sail with steam propulsion lay less in the nature of engines and sails than in the developments which took place during the nineteenth century in guns and armour. Turret guns require a wide angle of fire, besides being very heavy. The necessary protective armour was even heavier. To combine the required fields of fire, and also adequate stability, with full sail propulsion constituted a very difficult problem in naval architecture. In the 1860s the Admiralty were understandably inclined to proceed cautiously. If they had been allowed to continue to do so, all might have been well and history might have been considerably different.

  This applecart was upset by a certain Captain Cowper Coles. Coles was one of those clever men with an exceptional talent for controversy and publicity. Having invented a new sort of gun-turret, he set himself to persuade the Admiralty to build a battleship around it with full sailing rig and therefore unlimited range. Coles managed to involve, not only the Admiralty, but also both Houses of Parliament, the Royal Family, the Editor of The Times and practically the whole of the Establishment in what became one of the greatest publicity exercises of its kind.

  Tiring eventually of being called ‘reactionary’ by half the newspapers and more than half the politicians in the country, the Admiralty gave way. They did what they had never done before, and will certainly never do again; they allowed a serving naval officer with no qualifications in naval architecture to design his own private battleship and have her built at the public expense.

  The ship was built by Lairds at Birkenhead as Coles’s responsibility and with none of the usual checks on design. She was, moreover, built in a blaze of vituperation and controversy. For much of the time Coles himself was ill and unable to leave his home in the Isle of Wight to attend the yard. As a result of all this muddling, the ship ended up about 15 per cent overweight. If this had not been the case it is at least possible that the ship would have been a success and comparatively safe.

  As it was, the Captain was much too deep in the water and her C.G. was much too high up. Subsequent calculations showed that the ship would capsize if allowed to heel beyond an angle of 21°. However, the ship was commissioned in 1869 with much publicity. She made two deep-water cruises to the great satisfaction of The Times and of the First Lord of the Admiralty, who had his own midshipman son transferred into her. It looked as if the problems of world power, without the encumbrance and potential embarrassment of world bases, were going to continue to be soluble.

  On her third voyage, returning from Gibraltar in 1870 in company with the rest of the Channel Fleet, H.M.S. Captain suddenly capsized in a rather moderate squall in the Bay of Biscay. 472 lives were lost – more than the total British dead at Trafalgar. Both Cowper Coles himself and the First Lord’s son were drowned. Only seventeen men and one officer were saved.

  Though not, of course, the sole factor, the loss of the Captain had a powerful effect in accelerating the change from sail to steam, or rather on the abolition of the full sailing rigs in big battleships. Whatever the technical consequences, the political ones were extensive. It will be remembered that the Suez Canal, which was opened just before the Captain was launched, originally belonged effectively to France. Disraeli bought the Suez Canal shares for the British government in 1874, and the acquisition of a worldwide chain of coaling stations became a political necessity. The whole story of the Captain disaster is complicated, but the immediate technical cause was undoubtedly the determination to ensure that the masts and hull of the ship should have really adequate strength – regardless of weight. It was one of many structural accidents in which nothing actually broke, but the causes were just as’ structural’ as if they had.

  Aeroelasticity -or a reed shaken by the wind

  When a fluid, such as air or water, flows past an obstruction, which might be a tree or a rope, eddies of fluid are formed behind it. Quite often, if you observe a reed or a bulrush growing in a fairly slow-moving river, you will see that the eddies in the sliding water are formed first on one side, then on the other, alternately. The result is a rhythmic variation of fluid pressure, from one flank of the obstruction to the other. Such a succession or ‘street’ of eddies is called a ‘Karman strasse’, after the aerodynamicist von Karman, who first described it. It is often quite easy to see eddies on the surface of smooth water though eddies in air are invisible unless they are shown up by smoke or dead leaves or some similar indicator. In fact, however, just the same Karman strasse of eddies happens when air blows past a flag or a tree or a wire. The result of these alternate eddies, acting first on one side then the other, is that the flag flaps, the tree sways and the telegraph wires sing and hum in the wind. Thus a sail will flap as soon as the sheet is eased and may very well split itself or injure somebody. I remember seeing a man knocked out by a flogging sheet-block; there is a lot of energy involved. When a big ship is tacking in a breeze, the noise is as loud as gunfire and much more impressive.

  If the frequency of the aerodynamic stimulus provided by the eddies happens to coincide with one of the natural periods of vibration of the obstruction, then the amplitude of the movement may increase until something breaks. It is this sort of thing, rather than steady wind pressure, which usually accounts for trees being blown down. In a somewhat more sophisticated way this is also what is rather too apt to happen with aeroplanes and suspension bridges. It can be prevented by making the structure adequately stiff, especially in torsion. As we have already remarked, it is the torsional stiffness requirements which generally govern the design, and the structure weight, of modern aircraft.

  Although Telford’s Menai suspension bridge was quite badly damaged by wind-induced oscillations not long after it was built, it took about a century for the reality of this danger to register properly with bridge designers. The classic catastrophe was that of the Tacoma Narrows bridge in America in 1940. This bridge, which had a span of 2,800 feet (840 metres), was built without adequate torsional stiffness. As a result it swayed in even a moderate breeze to such an extent that the locals immediately christened it ‘Galloping Gertie’. Quite soon after it was built it swayed and wriggled itself into a dramatic collapse in a wind of only 42 m.p.h. Fortunately somebody happened to be present with a film in their cine-camera. The camera worked and the price of the film must have turned out to be a good investment, since it has been shown repeatedly in practically every engineering school in the world ever since (Plate 20).

  In consequence modern suspension bridges are built with adequate stiffness, especially torsional stiffness. As in aircraft, the stiffness requirements account for a good proportion of the weight of the bridge. In the case of the Severn road bridge (Plate 12), for instance, the decking is made from a very large steel tube of flattish six-sided section, built up from mild steel plates. During construction this tube was floated out in sections, which were hoisted into place and then welded into a continuous structure.

  Engineering design as applied theology

  In nearly all accidents we need to distinguish two different levels of causation. The first is the immediate technical or mechanical reason for the accident; the second is the underlying human reason. It is quite true that design is not a very precise business, that unexpected things happen, that genuine mistakes are made and so forth; but much more often the ‘real’ reason for an accident is preventable human error.

  It is rather fashionable at present to assume that error is one of those tilings for which it is not really fair to blame people, who, after all were ‘doing their best’ or are the victims of their upbringing and environment, or the social system – and so on and so on. But error shades off into what it is now very unpopular to call ‘sin’. In the course of a long professional life spent, or misspent, in the study of the strength of materials and structures I have had cause to examine a lot of accidents, many of them fatal. I have been forced to the conclusion that
very few accidents just ‘happen’ in a morally neutral way. Nine out of ten accidents are caused, not by more or less abstruse technical effects, but by old-fashioned human sin – often verging on plain wickedness.

  Of course I do not mean the more gilded and juicy sins like deliberate murder, large-scale fraud or Sex. It is squalid sins like carelessness, idleness, won’t-learn-and-don’t-need-to-ask, you-can’t-tell-me-anything-about-my-job, pride, jealousy and greed that kill people. Though some engineering firms have splendid design teams, far too many firms in this country are technically incompetent – often to a criminal extent. Many of these people have risen from the shop floor, and, out of a mixture of pride and meanness, they intensely resent any suggestion that they should seek proper advice or employ qualified staff.

  It is my experience that far more accidents occur every week than ever get into the papers; generally they are caused by lack of proper care and professional competence. I very much doubt if the remedy lies in the imposition of yet more regulations. It seems to me that what is wanted is the creation of more public awareness and a climate of opinion which regards such ‘mistakes’ as morally culpable. The man who drilled a hole in the wrong place in the wing-spar of a wooden aeroplane, plugged the hole, and said nothing, was acquitted. Presumably the jury thought that the moral blame was negligible.

  What is wanted is much more publicity; the difficulty lies in the law of libel. In most cases, if the real causes of an accident are made public, somebody’s face will be very red, and it is likely that their business or professional reputation will suffer. Most practising engineers are acutely aware of this and have to keep quiet or risk heavy damages. In my opinion there should be some way round this, for it is in the public interest that accidents and blunders should be publicized.