Some random thoughts on bridges

Brian Micklethwait made a post on his blog about the new, magnificent Millau viaduct over the Tarn River in southern France. Various people made further comments about it (which can be read if you scroll down the post). In particular Michael Blowhard made an observation that amounts to the fact that he doesn't like modern cable stayed bridges compared to more classical suspension bridges. I started writing a comment which got a little out of control, so I ended up deciding to post it here. I recommend you read the original post/comments first. It would also probably be better with pictures, but I don't have time to draw them.

There is actually nothing impractical in terms of physics with what Michael Blowhard suggests. You could build a viaduct with multiple suspension spans, just as the Millau viaduct has been built with multiple cable stayed spans. Each end tower would be anchored to the ground on one side, and would have one end of the cable catenary (not a parabola) from which the deck is suspended hanging from the other side. The non-end towers would have a catenary hanging from each side. This is in fact probably the way the bridge would have been built prior to the invention about 20 years ago of the materials that made long span cable stayed bridges practical. However, it wasn't done because it would have been too expensive. (There are also perhaps issues with safety. A catastrophic failure of a multiple span suspension bridge would be more likely to cascade from one span to another than is the case with a multiple span suspension structure).

Or perhaps Michael is saying that he would prefer a single span suspension bridge. Once again there is absolutely no physical reason why this couldn't be done across the Tarn at Millau, other than the cost of it. One possibility would be to actually build the two towers on the banks of the valley - that is have a bridge which does not touch the floor of the valley anywhere. This would lead to an extremely long span (2.5km) which would be the longest in the world, but not by a huge margin (the largest is presently 1.991 km long). Or you could anchor the towers on the floor of the valley and have a shorter main span. This would require pretty immensely thick towers, but again there is no reason why it couldn't be done. Other than the expense.

The reason why we now have lots of cable stayed bridges nowadays is of course simply cost. (cable stayed = the cables connect the tower to the deck directly but at a non-vertical angle. suspension bridge = there are catenaries between the two towers to which the deck is connected by vertical cables). In around 1980 new materials were invented that were strong enough to allow large cable stayed bridges to be constructed. As these dispense with an entire aspect of the design of suspension bridges (the suspended catenaries) the total mass of such bridges is much smaller, and the total cost of much lower.

But, this only hold up to span lengths up to about 1km. For lengths above this the lateral stresses on the deck are so great the the modern materials cannot cope with them, and a classical suspension bridge is still the only possible way to go. So, simultaneously with the construction of a great many cable stayed bridges around the world (This one in Normandy being one of the largest) the last decade has also been a great time for the advancement of the art of the classical suspension bridge. However, this is perhaps not noticeable to the casual observer, as while there have been a huge number of new cable stayed bridges build with span lengths of (say) 500m to 1000m, there have been fewer than ten new and immense suspension bridges. (List here). But, these small number of bridges have been particularly great. The longest is this great structure connecting Kobe in Honshu to the Island of Shikoku in Japan. (Unlike the Millau viaduct, this one is pretty genuinely a white elephant).

But all that have been built so far pale in comparison next to the bridge the Italians are planning on building across the Straits of Messina connecting mainland Italy and Sicily. This will have a main span of 3.3km, which will be by far the longest span ever built, massively further than the distance across the Tarn Valley for instance.

Unlike the Millau viaduct, this one will indeed be a colossal white elephant (and a lot of the money to pay for it will end up in the hands of dubious people, as happens in southern Italy). But somehow I just want to see it, and to walk across it. For white elephant or not it will certainly be magnificent.