Britannia Bridge - Pushing the Limits: New Adventures in Engineering - page 42

generally like an I-beam, but Stephenson knew that there most likely would be unknown differ-
ences. The tubes that Stephenson envisioned could conceivably behave in ways that he could not
imagine without doing some experiments.
When Stephenson consulted with William Fairbairn, who was then well-known for his investig-
ations into the strength of cast iron, the bridge engineer was reassured that the idea of iron tubes
would work. Fairbairn related stories, which Stephenson had also heard from a ship-builder on the
railway board, of iron ships hundreds of feet long that had survived being tossed about in the sea
and even being accidentally supported with one end raised on a wharf. A great hollow beam sup-
ported in the air only by its ends would not be unlike a ship in such a predicament. However, ships
were also not exactly like the tubes Stephenson had in mind, and so Fairbairn could not tell exactly
how stiff or strong the bridge tubes should be.
Fairbairn became engaged to assist Stephenson in an “experimental inquiry.” Although there was
growing confidence among the engineers that the several sections of tubes would be strong enough
to support themselves, in order to get a bill authorizing the construction of the bridge through com-
mittee in the House of Commons, Stephenson had to “leave the impression upon the minds of the
Committee that at all events the chains might be left as auxiliaries to the tube if necessary.” Thus,
the masonry towers, which would be built before the tubes could be jacked into place, had to be tall
enough to receive suspension chains, if necessary.
In the meantime, Fairbairn tested models of tubes of various shapes, supported by their ends and
loaded in the middle. The experiments showed clearly how and under what weight a riveted tube
would break, and they led Stephenson to realize that rectangular tubes were best for the purpose.
In failing, the models also identified the weak spots in the structure. These were beefed up in later
experiments, and so the final configuration of iron in the tube could be established. Stephenson and
Fairbairn did not seem especially inclined to determine which parts of their broken model tubes
were overly strong, and so they made no systematic attempts to trim excess material where it was
not needed. Had they done so, the final design of the Britannia Bridge might have been more eco-
nomical. But such economy would have been gained at the expense of time and so might have been
false economy to the board of directors, who wanted to see the rail link completed as soon as pos-
sible. As it turned out, the largest tubes Stephenson used were 460 feet long and thirty feet deep at
midspan, a ratio of length to depth of about fifteen to one. Engineers today would characterize the
tubes as slender structures.
Even with Fairbairn's methodical experiments, it would have been risky to extrapolate from the
experiments alone the exact dimensions of the cross section of a full-size tube. A quantitative the-
ory of the strength of tubes was necessary, and when Fairbairn suggested that Eaton Hodgkinson
might help, Stephenson immediately consented to engaging him, already “being familiar with the
valuable contributions of this gentleman to engineering science.” What Hodgkinson did was to ex-
press analytically the strength of a wrought-iron tube in terms of its dimensions and an empirical
factor determined from the results of Fairbairn's experiments. With the complementary support of
theory and experiment, then, the detailed design of the tubes could be determined and construction
could commence. The combined experimental, empirical, and analytical approach to engineering
remains to this day a paradigm for designing large and complex systems.