Kent Bridges: 5. Eureka Skyway

This footbridge is a very marked contrast to the one at Swanscombe Cutting that I covered recently.

First, it has a silly name, the "Eureka Skyway". I was just going to call it the M20 Ashford Footbridge, as that's what it crosses and where it is, but I've decided to use its official name for the blog post title. It links two retail parks either side of the motorway, and also acts as a gateway to Ashford itself, an indication to motorway users that perhaps something of significance can be found here.

The £8m bridge was designed by Nicol Russell Studios with Jacobs, and built by BAM Nuttall. It was installed in May 2011 and opened in September 2011. There were rumours before it opened that it suffered from "wobble", but I could feel no evidence of that when I visited earlier this year.

Three immediate precedents come to mind when viewing the Ashford bridge: Lancaster's Lune Millennium Bridge (2001), and Newport's City Footbridge (2006) are two. Closer to Ashford, Maidstone's Lockmeadow Footbridge (1999) has a similar form. All four bridges share a resemblance to a giant crane, with twin masts tied together with cables, and a deck supported from cable stays. There are some structural advantages to this arrangement, chiefly that the angle of the cables supporting the deck is steeper, and hence they provide a stiffer and more efficient support. A significant disadvantage is that maintenance is more difficult, as cables are required with no low-level termination, making it more difficult to adjust or replace them in the future.

I saw preliminary visuals for this bridge some time ago, and you can find them on the architect's website. I thought it looked quite nice, appealing in its height and slenderness. In real life, I found the sheer scale of the bridge to be much more difficult to accept. The bridge deck is roughly 100m long, with a 67m clear span. The masts are 38m high. It's nowhere near as big as the Newport bridge (70m tall, 145m main span), but it's still a very large structure to span a motorway. The span length is driven by the presence of two motorway slip roads, and a watercourse, forcing the bridge to be much larger than is normally required on motorways.

The bridge looks attractive enough when viewed from the motorway, but as a pedestrian it is quite overpowering. The masts tower far above you, and their angle of inclination makes them loom in a way that I don't recall experiencing with vertical masts.

What seems like a simple enough cable system when viewed in elevation becomes not only complex but positively confusing from most other perspectives. This is true of the forestays, but doubly so for the backstays. As well as tying the masts back to anchorage foundations, these also support a secondary deck, which curves below the main deck and provides part of a shallow gradient ramp access to the bridge's north end. The end result is a web of cables with little apparent visual order.

In the main span, the confusion is partly caused by the presence of a set of tie-down cables, two on each side of the deck, which anchor the foremast to the foundations. These are an odd presence on an asymmetric cable-stayed bridge, where normally the anchor cables are only required at the "rear end" of the bridge. Here, I guess the main deck is insufficiently heavy to hold the main masts in place on its own, or they are needed because the masts are offset to the side of the deck, or they minimise movement and vibration in the masts. Compare the Swansea Sail Bridge as an example of a cable-stayed bridge with an offset mast which doesn't need to be tied down in this way (and where the mast has been more artfully shaped).

Below deck, the oddest feature is a Y-shaped strut which holds the deck in place, presumably to restrain either vertical or lateral movement, or both. It has a sort of "tacked-on" feeling. My initial thought was that a corbel from the masts would have been better, but it's not clear which mast you would add the corbel too, and what the effect on structural behaviour would have been. The view from below also highlights the somewhat rudimentary details where the cables are anchored to the deck.

The approach ramps at each end of the bridge are also very heavy, largely because of their height above the surrounding ground. There's a stone-walled ramp at the south end, and an earth mound at the north end, into which stairs and a spiral ramp are cut. Growth of landscaping over time will help soften these, but they weigh the bridge down rather than allowing its slenderness to float free.

Overall, it's a bridge which had the potential to be great, but which is let down by the awkward resolution of many of the details. Notwithstanding the funder's desire for a gateway structure, I found the sheer scale to be oppressive. The bridge at Swanscombe spanned the motorway with delicacy and modesty - the one at Ashford has neither.

Further information:

• Google maps / Bing maps (both views pre-date construction of the bridge)
• Construction video

Kent Bridges: 4. Swanscombe Cutting Footbridge

I visited Kent earlier this year, and took in a couple of bridges while there. Both are pedestrian bridges spanning over a major highway, but very different in conception.

The first dates from 1965, and spans the A2 highway, which follows the route of the ancient Watling Street. Approached from the east, the road rises up across the landscape before slicing through the hillside in the Swanscombe Cutting, and it's not dissimilar approached from the west. The cutting is therefore at the high point of the road, and the bridge that spans it is a key marker on the route. As a note on the Concrete Bridge Development Group website states, "the scene is set for both drama and beauty".

The bridge was designed under the direction of H Bowdler, County Surveyor of Kent, and J A Bergg, County Bridge Engineer. It was built by Leonard Fairclough Construction at a cost of £17,000. In Bergg's paper covering a number of bridges built in Kent in the 1960s, he reports the cost to equate to £116 per square metre of deck.

The structure spans 48.77m, being 2.3m wide between railings, and 89.5m long in total. The arch rises 7.83m. Triangular abutments are buried in the cutting slopes, approximately 8m tall, which provide combined support to both the arch and the deck arms.

The bridge was Listed Grade II in 1998. I'd be interested to know whether any other bridges built this recently in the UK have been given similar protected status! It certainly seems remarkable, but then, it is clearly a remarkable bridge.

Structurally, the bridge is a descendant of Robert Maillart's 1906 Tavanasa Bridge, a three-pinned concrete arch, with propped cantilevers for the bridge deck. All parts of the bridge taper towards a hinge - the arch members to hinges at their base and crown, and the deck to the support at their ends. At the thinnest points the bridge is a mere 0.36m thick.

The bridge at Swanscombe is far more slender than even Maillart could achieve. It represents what Maillart could have done had he collaborated in later life with his contemporary, Eugène Freyssinet. Maillart worked only with reinforced concrete, but it's Freyssinet's pioneering of prestressed concrete which ultimately allowed the bridge at Swanscombe to be so slender. The deck and arch at Swanscombe were both prestressed using the CCL system.

The main deck has a roughly 1m wide central prestressed spine beam, with the deck slab cantilevering either side. The deck is so thin at the arch crown that it entirely disappears into shadow below the deck cantilevers, an effect which is visually a little disconcerting as the visual continuity of the arch soffit can be lost.

It's not an easy bridge to photograph, thanks to the extensive tree growth along the cutting slopes. There's a photo in Bridges of Britain which shows the bridge with bare slopes, which really shows the structure at its best. I can't help but think that if this bridge spanned across a Swiss gorge rather than an English dual carriageway, it would be far better known, and properly recognised in the bridges literature. As it is, even Henry and Jerome's Modern British Bridges, published in 1965 and intended to recognise the best modern bridges of its time, fails to include it.

Perhaps the only criticism I would offer of this bridge is a personal one. The parapets provide an effectively minimalist solution, largely eliminating their visual impact on the bridge's elevation. However, they did little to calm my sense of vertigo and exposure while crossing the bridge. How much of this was the normal experience of crossing a busy main road, and how much due to the bridge's setting, I couldn't tell.

Swanscombe Cutting Footbridge is far from the only ultra-slender concrete footbridge built in Kent under Bergg's direction, and hopefully I will get a chance to visit some of the others in the future.

Further information:

• Google maps / Google street view / Bing maps
• British Listed Buildings
• Eighty Highway Bridges in Kent (Bergg, ICE Proceedings, 1973)
• Bridges of Britain (de Maré, 1975)

Kent Bridges: 3. Whatman's Field Downstream Bridge

The last of the trio of millennial footbridges in Maidstone goes by the somewhat cumbersome name of the Whatman's Field Downstream Bridge, so called simply because it is further downstream on the River Medway than the Kent Messenger Millennium Bridge.

Like the Messenger bridge, the Downstream Bridge was built by Balfour Beatty (steelwork fabricated by Fairfield Mabey), and designed by Flint and Neill with Studio Bednarski. It also opened in July 2001, but it's a very different bridge.

On Bednarski's website, it's called the "Blue Art Beam Bridge", and that seems apposite. It consists of a three-span, 75m long single steel box girder, in a very unusual "top hat", "spine beam", or "finback" arrangement. The river could only be closed for 4 hours for the main span lift. In most box girder bridges, the wider flange is at the top, to allow the full width of bridge deck to be carried while allowing the main structural section to be narrower and hence more economic (as well as visually appealing).

There are a few cases where the box girder has been inverted, with a recent example being Gifford's Central Park bridge in Manchester (illustrated in Masterpieces: Bridge Architecture + Design and the subject of a technical paper). This allows the construction depth (i.e. depth from top of pavement or rails to underside of bridge) to be minimised while retaining the advantages of the box girder form, and is a rarely used alternative to a through-truss design. The major disadvantage is that such bridges are often bulky and ungainly in appearance.

So, for the Downstream Bridge to have the wider part of the box section at the bottom seems initially a perverse choice. But the reason may be apparent from the two photographs above. Only the bottom part of the cross-section is highlighted by painting it blue. The upper part is recessed behind the parapet mesh, and a mesh hides the detail from above as well. The intent is to make the bridge deck appear far thinner than it really is, and I think it works well.

The other prominent feature of the bridge deck design is the presence of numerous circular cut-outs in the girder's bottom flange. These are in part a response to the technical approval authority's refusal to accept a fully enclosed box girder (presumably on the grounds that it would be too shallow for internal inspection). They allow for inspection access to all parts of the girder, but are protected against unauthorised entry by the use of grilles. Clearly, they will have made the design more challenging, with the girder effectively having two bottom flanges connected by a series of lateral crossbeams, Vierendeel style, providing the required torsional stiffness.

The bridge certainly needs plenty of torsional stiffness, as its two V-shaped supports are aligned along the bridge centreline, providing no resistance to torsion at all. All the torsion in the bridge is resisted at the end abutments.

The supports are made from tapered tubes (presumably circular tubes welded to flat plates or fabricated boxes), supported by a simple stiffener detail. The choice of V-shaped piers on the bridge centreline is shared with Arup's Gatwick Air Bridge [PDF] and makes me feel a little queasy: the lack of wider supports at the intermediate piers just looks somehow wrong to me, as if the whole thing could topple in the wind.

The bridge parapets use the same stainless steel "Medway" mesh as on the Kent Millennium Messenger Bridge, and like that bridge, it results in a simple and minimalist line to the structure. The photo also shows the grille panels which hide the bottom flange (while unfortunately also trapping debris).

You can also see a slight kink in the vertical alignment, which is unfortunate. I wonder if it's a relic of the need to change the bridge height during the design period. This unhappy tale is told in detail in an audit report [PDF], commissioned to explain the substantial overspend on Maidstone's Millennium River Park project. This noted a £200k overspend on the bridges (relatively small beer), attributable to high tender costs, a major flood which occurred during construction, and the need to raise the planned bridge height by 1.5m in order to mollify river users unhappy with the effect on their navigation rights.

It's nice to see a bridge which is colourful, although blue is perhaps too often the choice on steel bridges. I think the attempt to make the deck look slender has been very successful, and the circular openings below deck help break up the mass of the bridge as well as being functional. As with the other Maidstone footbridges, it's an idiosyncractic design and not one likely to be repeated often elsewhere (if at all). Nonetheless, it's always good to see designers stepping into the unknown, and it's a worthy member of a very interesting trio.

Further information:

• Structurae
• Google maps
• Flint and Neill
• Studio Bednarski
• Steel Design Award

Kent Bridges: 2. Kent Messenger Millennium Bridge

Walk north in Maidstone from Lockmeadow Footbridge along the river Medway, and you'll come to the first of the town's two other millennial footbridges. This pair provide entry across the River Medway into the Whatman Field Park, and are named the Kent Messenger Millennium Bridge and the Whatman Field Downstream Bridge. These other bridges were both designed and built by a single team, as part of the Millennium Commission funded development of Whatman Park, one of the main sections of Maidstone's Millennium Riverside Park. This post will cover the Kent Messenger bridge, with the Downstream bridge to follow later.

In a paper on the two bridges, their architect Cezary Bednarski notes that they were the only millennium bridges in Britain not to be the result of a design contest. I'm not convinced about that, but Maidstone Borough Council certainly got good value, with both bridges being interesting and innovative structures, and both going on to win awards.

At the time the bridges were designed, Bednarski was a director of Studio E Architects, although he started a new firm Studio Bednarski a few months after the Kent Messenger bridge opened in July 2001. His engineering partners for the design were Strasky, Husty and Partners, and Flint and Neill. The main contractor was Balfour Beatty.

I'm not sure what the Kent Messenger bridge cost: Structurae and fib's Guidelines for the design of footbridges both report a figure of £1.76m, which works out at about £5.6k per square metre of deck, but Bednarski's paper states the cost to have been £2.9k per square metre.

The bridge is a very unusual structural form - the world's first cranked stress ribbon bridge. Perhaps it's still the only one. A stress ribbon bridge is essentially a catenary structure which has been stiffened by prestressing against itself. Depending on the specific geometry, it can be thought of as a reverse arch, where the prestress provides a broadly uniform upwards force along the structure's length, locking the structure in place. I've covered one such structure here before, the Punt da Suransuns, in Switzerland.

The stress ribbon bridge's generally low load-bearing capacity, together with the sagging geometry, mean that it's a structural form generally limited to footbridges. Even so, it's rarely an optimal form. What is gained in material economy in the deck (Kent Messenger bridge's 3.1m wide deck is only 0.29m deep) is generally lost in the cost of massive foundations, and the foundations for the Kent Messenger bridge are massive indeed, as can be seen in a diagram at Strasky's website. The sagging geometry is also often hard to reconcile with the need to provide a gentle slope, suitable for access by the mobility-impaired. Keeping the slope shallow enough can result in very high horizontal forces, and still larger foundations.

So, what does "cranked" mean, and why is it so unusual? Essentially, it means that the bridge is kinked in plan. The Kent Messenger bridge is 101.5m long in total, with two spans of 37.5m and 49.5m respectively. At the central support, the deck changes direction in plan by 25°. To make that possible, a large horizontal restraint is called for, as the tension forces in the two spans don't balance.

The nearby Downstream bridge (which I'll cover in the last post in this short series) was also originally proposed as a cranked stressed ribbon, with the resultant force at the "crank" restrained by a tie. It was eventually built in a much different form, but the Kent Messenger bridge retained the crank, with the resultant force restrained by a strut. The strut is angled downwards according to the vector sum of horizontal and vertical reactions, and conveniently doubles as a staircase, a very effective solution.

Unfortunately, it doesn't quite work perfectly under all load cases, and a very slender stainless steel member had to be incorporated below the support position, acting in many cases as a tie to prevent the deck rising, and in some cases as a second strut. It could only have been omitted by providing the staircase with much enhanced bending capacity, but I think it's a shame it was included, as the bridge would be far more impressive without it.

The bridge deck itself consists of 3m long precast concrete segments composite with an in-situ infill deck slab. The segments are supported on a set of four bearing and four prestressing cables cast into the infill.

Drainage grilles are located along the centreline of the bridge (and can be seen on the image on the right), breaking up the structure's monolithic appearance and allowing the river to be seen through them. These alternate with lights embedded in the deck (see below left). These are the same kind of lights you would find on an airport taxiway, a rather idiosyncratic choice.

The bridge parapet incorporates a custom-designed stainless steel mesh, which is now marketed as the "Medway" mesh by Potter and Soar.

As with the Lockmeadow footbridge, the Kent Messenger Millennium Bridge wasn't shown to its best on a dull, overcast day. Given that stress ribbon bridges are often chosen for their lightweight, slender appearance, I was surprised to find this bridge offering an almost brutalist statement. The cranked geometry was clearly an attempt to push at the boundaries of structural design - there are several other forms which would have served as well, perhaps incorporating some of the length eastern approach ramp into the bridge structure. The staircase strut is a great design feature, although it's a shame about the presence of the central tie, as the bridge would have had a far more startling visual tension without it.

The overall geometry aside, it's a plain, functional, rather minimalist structure, and none the worse for any of that. Compared to the tangled cat's cradles of two of its contemporary Millennium Bridges (at Bankside and Hungerford), it deploys its structural ambition to a more self-effacing end. The views off the structure are kept free of cables and struts, and there's little more to see below. I can't say I found it at all an exciting bridge to visit, but I guess that's part of the intention, and it's good to see a strong design that doesn't take its lead from Calatrava.

Further information:

• Structurae
• Google maps
• Kent Messenger Millennium Bridge, Maidstone, UK (C Bednarski, Footbridge 2002)
• Kent Messenger Millennium Bridge, Maidstone (Concrete magazine, Nov/Dec 2003)
• Stress ribbon and cable supported pedestrian bridges
• Footbridges
• Masterpieces: Bridge architecture + design

Kent Bridges: 1. Lockmeadow Footbridge

A recent trip to Kent, in southern England, gave me the opportunity for a short side-trip to the town of Maidstone. This is the home of no less than three modern landmark footbridges, all built around the turn of the millennium. They were promoted by Maidstone Borough Council, with external support from funders such as the Rochester Bridge Trust.

It wasn't the brightest of days, so none of the bridges were shown to their best effect. The first is Lockmeadow Footbridge, spanning the River Medway close to the 14th century Archbishop's Palace. This is a historically sensitive site, and it's apparent that the bridge designers, Flint and Neill with Chris Wilkinson Architects (now Wilkinson Eyre), sought to create as lightweight a structure as possible in an attempt to minimise its visual intrusion.

They won the bridge design contract following a design competition in 1997, and the bridge was opened in October 1999. It was built by Christiani & Nielsen for £650,000 (£3.9k per square metre of deck), with steelwork by D&B Darke Ltd.

The bridge is 80m long, with a 45m main span, but the deck is only 300mm deep. That gives a remarkably slender span-to-depth ratio of 150. The span arrangements are driven by the need not only to bridge the river, but also to clear the adjacent floodplain. The choice of a large central support was driven by construction constraints, with a weak river bank wall and limited access available on the east river bank. The resulting bridge deck is a little awkward, as in dry conditions it has a very low clearance underneath (see main photo at the top of this post). However, the overall structural form, with twin pylons each carrying twin cables, is very striking, particularly in elevation.

To achieve the desired deck slenderness while minimising weight, a cellular aluminium deck has been adopted. A series of extruded profiles, supplied by Dutch firm Nedal, are placed side by side, and locked together by transverse stainless steel prestressing bars at 1.6m intervals. A non-slip texture is obtained by sawcutting the upper surface of the extrusions. Ian Firth's paper in SEI (see link below) gives much more detail on how the deck was conceptualised, designed, and built. The aluminium segments still look in very good condition a decade after the bridge opened in October 1999. The main question that strikes me is why, if this system is so economic and durable, it hasn't received much wider use?

The use of cellular aluminium extrusions in bridge decks isn't unique, with patented systems such as those from Alumadeck and Sapa Front (see paper by Tomasz Siwowski [PDF] for details). But the arrangement used at Lockmeadow was highly innovative, and only one of several aspects of the bridge which are unusual.

The skeletal steel masts are a design which Chris Wilkinson Architects had already used for the entrance canopies to the Dyson Headquarters at Malmesbury (completed just a few months before the Lockmeadow bridge). They remind me a little of the Skylon, which was similarly cigar-shaped, but also evoke the stylings of high-tech architecture. The use of twin plates for each diaphragm is a nice feature.

The parapet posts are resin-filled, curved FRP forms, supporting stainless steel handrails and stainless steel mesh infill. These are surprisingly transparent, but still add to the visual depth of the structure.

The bridge seems to me to be most successful visually when viewed in elevation, where the unusual cable-stay arrangement is most effective. However, it's almost impossible to see the whole bridge from any viewpoint to the north or south. It therefore seems unnecessarily gigantic - the field of view struggles to contain its scale and it's therefore perhaps more domineering than was intended.

Viewed along the deck, the visual complexity of the pylons and cable arrangements again give the impression that bridge is larger than it really is. It seems to be trying a little too hard, grabbing attention rather than relinquishing it.

That was all something of a surprise to me, as having only seen photographs of the bridge before, I'd been very much looking forward to visiting it. Perhaps my expectations were too high; perhaps the overcast day would have made even a structure of filigree glass seem heavy and awkward rather than graceful.

Given the bridge's unusual slenderness, I did my best to get it moving under an oscillating pedestrian load, but without success. Tie-downs secure both ends of the bridge, and I guess these are balanced with the tension in the locked-coil cable stays to limit vertical movement of the deck.

As a piece of structural engineering, it's a tour de force, with a distinctive visual form, several innovative elements, and having been well-engineered to prevent its slenderness leading to vibration problems. Architecturally, I left it with mixed feelings. It's unexpectedly ostentatious, giving the distinct impression that it might work better in a different visual context. But I'd like to visit it again on a sunny summer day, as I suspect that would lead to a more positive appraisal.

Further information:

• Structurae
• Google maps
• Lockmeadow Footbridge, Maidstone, UK (paper in SEI by I Firth)
• The Design and Construction of Lockmeadow Footbridge, Maidstone (paper from IABSE conference 1999, abstract only online)
• LUSAS case study
• Bridge Builders
• Stress Ribbon and Cable-supported Pedestrian Bridges
• Wilkinson Eyre: Bridging Art and Science