23 December 2019

London Bridges: 54. Chiswick Park Footbridge

It has been a while since I've visited and reported on a recently-built bridge (May, since you ask, and before that, August 2018), so this visit to see Chiswick Park footbridge in London felt long overdue.

According to the Ian Visits blog, the bridge itself is also long overdue, with planning applications dating back to 2003. It forms part of a walking route connecting Chiswick business park to Chiswick Park tube station. The bridge finally opened to the public in January this year. It was designed by Expedition Engineering and Useful Studio, with the steelwork constructed by Severfield.

I believe it's only the second network arch bridge to be built in the UK, and the first such pedestrian bridge. The design didn't start life as a network arch structure, and has been through a lengthy evolution to get to what was eventually built.

Originally, proposals were for a truss bridge of some form, as per the 2003 and 2006 planning applications shown here:

The bridge's three spans were largely determined by clearances to road and rail routes below, and in these early designs it was indicated that the truss span over the railway would be fully enclosed to prevent risks from vandalism.

In the 2012 application (top diagram in the image above), the design had become three steel bowstring arches, each of increasing span and height from west to east, with vertical hangers supporting the deck. Tall mesh parapets were indicated above the railway, eliminating the need for full enclosure.

The overall form of the bridge changed very little thereafter. The shape of the arches was the result of a form-finding exercise, to maximise the visual slenderness of the arch. I'm not clear how that will have worked, as the critical bending arrangement for an arch of this sort is usually with only half the span loaded with pedestrians.

That scenario is often more onerous in design than the full span loading which produces the greatest axial load in the arch. In a conventional bowstring arch, the stiffness of either the arch or the deck (or both) is required to resist this half-span bending.

In any event, during design development the bowstring arch was found to perform badly under dynamic pedestrian loading, and was amended to a network arch in the 2015 planning application (middle diagram above).

This is many times stiffer than the previous design, raising the bridge's vulnerable natural frequencies, and eliminating or mitigating the dynamic problems. Bending moments in the arch due to asymmetrical loading arrangements are also greatly reduced.

I believe the previous design incorporated a concrete deck - adoption of the network arch also allowed a lighter all-steel deck to be used, minimising the weight required and making craneage of the spans into place easier.

The final change is indicated in the 2017 planning application (bottom diagram above), and indicates that the design team had failed in their desire to persuade Network Rail that a mesh parapet would be sufficient above their railway line.

The railway authority is never noted for its flexibility when there is a rulebook that can consulted, so the final introduction of a solid (imperforate) parapet screen above the railway tracks is unsurprising. It is at least largely disguised by being hidden behind the facing mesh.

The bridge's spans are 37.0m, 40.7m and 44.4m, totalling 122.1m. The arches and deck are connected integrally to the two intermediate piers, with bearings allowing thermal articulation at each end. All the structural steelwork is weathering steel, with stainless steel parapets and hanger cables. The decking is floored in timber planks.

The articulation is interesting, as conventional wisdom would be that arches of this type should sit on bearings at all points, allowing the tie girder connnecting the ends of each arch span to expand freely. This allows it to take up a full tension balancing the compression in the arch, and allowing the hanger network to interact efficiently with the main steelwork.

In this instance, the V-shaped piers are sufficiently flexible longitudinally that they will offer only limited restraint to the arch thrust, and the network on the central arch will still work although I'd guess with slightly reduced effectiveness.

The most striking aspect of the bridge is the effort that has been expended to make its main elements slender, with cruciform sections for the arch and piers, and a simple stiffened steel plate for the deck. In addition to being slender, all parts are visible for inspection and maintenance, unlikely the closed box sections often seen in footbridges.

What is also very much apparent is an impressive attention to detail. Wherever possible, connections are kept simple, with welding used extensively instead of bolting. The edges of the deck are made clear and sharp in profile, and the hanger and parapet connections are well-detailed and as minimal as possible.

It's the sort of bridge that any designer would be proud of, and especially impressive given that this is predominantly a structure used to get rapidly from A to B, rather than a destination in its own right. it has been shortlisted for an IStructE award, been a finalist in the CE Awards, and won two ICE awards. I'm a little surprised that it hasn't been more widely rewarded, to be honest.

The bridge is not completely without its flaws. Site constraints mean that although at its western end the bridge connects directly with a podium deck level in the business park, at the eastern end the approach is via steps and a lift.

I imagine mobility-impaired users are crossing their fingers in the hope that the lift will be better maintained than is often the case. It is at least attractively detailed in keeping with the rest of the bridge.

Also on the eastern approach I noticed a sign that recommends no cycling, perhaps inevitable given the steps and lift, but also advises users to "Walk With Care" due to gaps in the decking.

The timber decking is visually attractive, but perhaps some users with high heels have found it a problem. In any event, I suspect problems with the decking won't end there.

The timber slats are raised above the bridge deck, the upper surface of which is a flat steel plate. Rainwater drains through the slats, and flows along the deck (which I am told is waterproofed), before spilling straight to the ground at the ends via cut-outs in the deck plate. The whole arrangement is an inspection and maintenance liability - nobody will lift the decking to properly inspect underneath, and it's easy to imagine dirt and detritus leading to trapped water over time.

Despite these oddities, the Chiswick Park Footbridge is a very impressive feat of design and construction and well worth a visit.

Further information: