26 June 2014

Australian Bridges: 3. Sydney Harbour Bridge

What can I possibly add to the volumes that have already been written about the Sydney Harbour Bridge?

Well, I have selected nine photos from when I visited, so here are nine thoughts (in no particular order, and not especially related to each photo!)

Completed in 1932, spanning 503m, Sydney Harbour Bridge was never the longest arch bridge in the world, but it was by far the widest and heaviest. It was pipped to the record by the Bayonne Bridge, opened in 1931, and which spans 25 inches longer. However, Sydney's bridge is probably still the most iconic of the world's long span arch bridges, situated in a beautiful harbour, and forming a key landmark visible from much of the surrounding city.

Sydney's chief engineer, J.J.C. Bradfield, had originally proposed a cantilever bridge for the site. A tour of bridges around the world convinced him that a steel arch might be more efficient, and he prepared alternative plans based on the Hell Gate Bridge in New York. Both bridges are two-pinned trussed arches, with monolithic and essentially non-structural pylons abutting the steel arch.

There are subtle differences. The Hell Gate Bridge has a more visible reverse curvature towards the upper ends of the arch, and its top chord disappears within the bridge pylons, rather than stopping short as is the case in Sydney. The upper chord carries no significant force at its ends, so the Sydney Bridge is more honest in this respect, but I think it looks quite odd.

A further difference between Hell Gate Bridge and the Sydney Harbour Bridge is in the approach spans. Hell Gate has girder spans of similar depth to the main span, with a continuity of road deck line which the pylons don't greatly interrupt. Sydney's bridge has very deep deck trusses for the approach spans, and its pylons serve a useful purpose in distracting the eye from the very different structural forms on either side.

The bridge arch is two-pinned, quite literally supported on a pin at each end. The entire weight of the bridge and the road, rail and pedestrian traffic that it carries is passed through the bottom chord into these pins. What then, is the point of the upper chord of the arch truss? The deep trusses are needed to stiffen the arch against buckling, particularly from unevenly distributed loading. However, they could converge to a point at the ends, a "crescent arch", as was the case for the much smaller Tyne Bridge, built by the same contractor, Dorman Long, in 1928.

I think the reason for the difference lies in the method of construction. Sydney Harbour Bridge was built by cantilevering from each bank of the harbour, with the steelwork tied back by massive arrays of temporary cables anchored into bedrock. 

In its temporary state, the bridge doesn't behave as an arch at all, but as a cantilever bridge of the type that Bradfield eventually rejected. As such, it relies on its strength in bending to stand up, and this is why the arch truss is so deep towards its ends. It had to resist enormous bending during construction, even if this meant that the top chord at the ends became largely redundant in service. While a crescent arch might be strong enough to be built by cantilevering at smaller spans, it would simply not be deep enough and strong enough for a span of this magnitude.

The entire form of the iconic Sydney Harbour Bridge is therefore essentially a relic of its process of making.

The debate on the type of arch which should have been used was rehearsed by other engineers at the time, and I'd thoroughly recommend the discussions of the papers presented to the ICE. These cover a wider variety of procurement and technical issues which remain relevant today.

Cantilever construction of large arch bridges was a common method at the time (and since), although now it would be recognised as highly inefficient (a cable-stay bridge uses its temporary erection cables as part of its permanent structure, so is almost always more efficient, as is a suspension bridge at longer spans). Indeed, other engineers commented on this at the time. David Steinman, who had submitted a design for a losing tenderer, considered that a suspension bridge should have been cheaper, and noted that Dorman Long were reported to have lost a million dollars on the contract, suggesting the arch option was under-priced.

Steinman also commented that the heavy temporary cables would have been sufficient to build a suspension bridge. Ironically, some of the temporary cables were indeed used for precisely that purpose, to build the Walter Taylor Bridge at Indooroopilly in Brisbane in 1936. Again ironically, this was one of the very few bridges in the world to copy Steinman's "Florianopolis" bridge design, an unusual trussed suspension bridge which never gained wide favour.

The design of the Sydney Harbour Bridge was controversial. A huge disagreement arose between J.J.C. Bradfield, who had conceived the bridge, and Dorman Long's designer Sir Ralph Freeman, who was responsible for the detailed design and for all the construction drawings. Bradfield claimed to be the designer, and went so far as to entirely omit any mention of Freeman in various articles, including his technical paper for the Proceedings of the ICE. Freeman, in turn, campaigned in the Sydney press for his role to be recognised, with the full support of Dorman Long. The contractor even threatened to sue the New South Wales government if Freeman's role wasn't properly credited on a plaque to be placed on the bridge.

The entire controversy is discussed in great detail in Peter Lalor's excellent book, The Bridge. In the discussion in the ICE Proceedings on the various technical papers by Bradfield and Freeman, Oscar Faber is recorded making the point that Freeman appeared to have received too little credit. Faber and several others also spoke against the design of the bridge pylons, which were regarded by many as an unnecessary extravagance.

The bridge must be one of few large bridges in the world where the public can gain access to climb the structure, albeit for a hefty fee. BridgeClimb is one of the most popular tourist highlights in Sydney, and I took advantage of it when I visited. It must make considerable profit for both the bridge owner and for BridgeClimb, as when I was there, at the height of the Australian summer, groups of visitors could be seen walking up the bridge on a very regular basis.

The BridgeClimb experience is very well organised: visitors are thoroughly briefed on what to expect, and even get to climb a trial set of staircases before being allowed onto the bridge itself. The safety arrangements are impressive, with "climbers" hooked onto a static line constantly from the beginning of their experience to the end. The scariest part of the experience is passing across the approach spans towards the pylon, where the proximity of the roadway directly below the catwalk is a little disconcerting. Once past the main pylons and onto the main span, the climb is straightforward, and not at all terrifying. The large width of the main arch chords is such that you never feel close enough to an edge to feel at all worried.

For anyone wanting to know more about the bridge, I can recommend Peter Lalor's book (see below), which is from the perspective of a journalist, concentrating on the human-interest stories, and Peter Spearritt's (also see below), from the perspective of a historian, very well illustrated and placing the bridge narrative in more of a social context.

Further information:


Anonymous said...

Thanks HP, it is very interesting to read how the design is so influenced by its mode of construction, which is not obviously apparent to the architectural eye! The difference between the temporary and permanent conditions introduces an interesting dilemma for those who like the design of bridges to reflect their structural 'truth'. Perhaps it is easier to zoom out from this level of concern and simply enjoy the presence of the bridge in its setting. It really is wonderful - iconic in the proper sense - and I am very jealous that you managed to visit it!


The Happy Pontist said...

Here's an image of the bridge during erection which makes clear why the trusses are so deep at their ends: http://4.bp.blogspot.com/-5xyVr8bnZHo/UnIDtvaaAbI/AAAAAAAAIzk/a_pLvCgK3QI/s1600/Sydney_Harbour_Bridge_Construction.jpg

A cantilever bridge of any kind, including a cable-stayed bridge, is more honest to both its temporary and permanent states, but I think either would have been a terrible choice in Sydney.

Eric said...

Great write-up, I'm always interested in how construction transforms the design of a structure.

Not sure if you've posted about it before but the New River Gorge Bridge in West Virginia USA has a similar program for walking the catwalk of the deck truss.

I'm one of the lucky ones that actually gets paid to climb them for inspections.