The new bridge is being built to relieve pressure on the existing 1936 crossing, which will need substantial refurbishment in the near future. In the current arrangement, the bridge acts as a diversionary route for high-sided vehicles when the Forth Road Bridge is closed, and has to endure even greater pressure on its limited capacity. A high proportion of the bridge's regular traffic is heavy goods vehicles, all of which have to drive through the centre of Kincardine causing congestion and environmental damage.

The new bridge will complement the existing crossing rather than replacing it, supplying additional capacity and in combination with a bypass, enabling all through-traffic to be diverted away from the village. The entire scheme extends across some 6.4km, with a 1.2km-long single-carriageway viaduct forming the main crossing of the river. The scheme was developed over a number of years by the Scottish Executive, which employed Jacobs as consultant for the outline design. The contract itself went out to tender in September 2005, with the 29-month contract being let in March last year. Before the scheme went out to tender, Jacobs had developed the outline design which had been submitted to the relevant environmental agencies for approval. As a result, the location of the piers and the depth of the deck in the outline design were prescriptive; the piers in order to minimise the impact on the river bed and hydraulics, and the slender deck depth in order to minimise the visual impact of the bridge.

A key part of the works, as Morgan Vinci project director John Osborne explains, is the construction of the new viaduct over the Forth, although as a proportion of the works it runs to only about 50% of the total cost. "We had been tracking the progress of the project since 2003," says Osborne, "and when it came out to tender we looked for ways that we could develop the design further. We got our designer on board within the first few weeks of the tender prequalification period."

One of the main constraints on the construction of the new bridge was the fact that the salt marshes along the side of the river serve as nesting areas for migrating birds. The alignment and 45 degree skew of the bridge is aimed at avoiding these areas as much as possible, but there was also the danger of construction noise causing disturbance to the birds and ultimately driving them away from the site. Research on previous projects has shown that as long as such disturbance only occurs during two consecutive winters, the birds are likely to return - but if it extends into three winters, the birds may abandon the site. This major constraint dictated the overall programme - construction work on site began last June, and works within the designated area may not continue into the winter of 2008.

But this was not the only constraint driving the construction process; the client also stated that the main construction work for the bridge could only take place on the north side of the river. Although there are salt marshes on both banks, there is also a disused power station on the north side, and this is where the majority of the site work is taking place.

Although a small area of salt marsh and mudflat will be lost to construction on the south bank, a larger replacement area will be created as part of the works. These include part of the old power station site which will be returned to the estuary by moving the existing sea wall.

The outline design for the bridge was aimed at a launched system, but as Osborne explains, the environmental constraints prevent the contractor from working from both sides of the river using the traditional incremental-launching procedure. Instead, the entire 1,200m-long bridge deck will be launched from the north side - and consequently the whole launching procedure will be on the critical path of the project. The team believes this is the second-longest incrementally-launched bridge of its type in the world.

When Bd&e visited the site in late February, just two launching cycles had been completed, the second of which had taken 20 days. Osborne explained that the plan was to reduce the cycle time further in order to meet the schedule; by the time of the fourth launch in late March, this had been reduced to 13 days and the team only had to shave another day off the cycle in order to meet the required cycle time.

Some modifications were made to the outline design, as Benaim managing director Simon Bourne explains. The client's designers originally specified a straight, single pier shaft with no flare or widening at the top - this was initially for aesthetic reasons, but made it difficult to design the structure for the proposed launching technique. All bidding contractors requested additional width at the top of the columns to provide space for the temporary bearings, for jacking equipment and so on, and the modified design includes a flared section at the top of each pier. Another aesthetic consideration was the depth of the deck - because of its length and the fact that it is skewed across the estuary the illustrative design was for a deck just 2m deep. But according to Bourne, it is difficult to carry out the launching process with such a slender deck - and this depth was eventually relaxed to 2.8m.

Additionally, the designers and contractors have had to contend with the doubly-curved soffit of the concrete deck - a challenge in terms of construction and also analysis. The cross-section shape was kept to its original design, says Bourne, which although it was done for aesthetic reasons, results in more expensive construction as it is a three-cell box rather than a single-cell. However the continuously-curved soffit was adapted to generate a horizontal section at the base, producing a flat area useful for the launching process.

One thing that was firmly fixed was the overall layout of the spans and position of the piers - the environmental impact assessment that was carried out for the project had been based on this layout, and if any changes were to be made, the new layout would have to be subjected to a revised EIA. In this case, says Bourne, there was no reason to change the pier positions as 45m spans are standard for the launching method. There is a 36m span at each end of the bridge, and two 53m spans, one each side of the 'notional' navigation span of 65m. Whether or not the navigation span is strictly necessary at the present moment is a matter for debate, but the client wishes to leave future access open. There are port facilities upstream of the bridge, but they are not widely used.

The only alteration to pier positions is the addition of three temporary piers which will be used for the launching process. One has been installed at the centre of the main span, and one in each of the adjacent spans, next to the piers that flank the shipping span. Use of these intermediate piers, explains Bourne, will allow the 45m launching span to be kept constant across the whole launch.

The diameter of the pier shafts was originally set at 3.75m across the entire bridge to maintain a visual constant, but this was modified during the value engineering period with the diameter being downsized to just 2.5m except on the four central piers on the larger spans, which have to resist ship impact. These will have 3.85m-diameter piles which extend into 3m-diameter pier shafts. The remainder of the substructure will have 3m-diameter piles extending up into 2.5m-diameter pier shafts. At the tender stage, says Bourne, Benaim was pushing to go for monopiles rather than clusters of piles with a pile cap, but was not sure whether any contractor could deliver this size of pile with a rock socket.

The tender design specified driven steel tubes for the foundations, arranged in the traditional manner of clusters of piles with pile caps. It soon became clear from the ground conditions that bored piles would be a much better solution. When the team started talking to specialist subcontractor Seacore, and realised that it was capable of meeting these demands, the team argued successfully that a bored solution, using single, large-diameter piles, would be more reliable in terms of the programme. These could be extended above the water level to form insitu 'cofferdams' within which the single-shaft piers could then be built. This meant savings of time and money for the contractor in eliminating the need to construct pilecaps, and omitting the use of cofferdams on the majority of the piers. Five piers of the 25 in total still had to be built in cofferdams – four on the south side and one on the north – these are in the salt marsh area and not accessible to Excalibur.

"It makes it a very tight squeeze to carry out the concreting of the pier shaft inside the pile casing," says Bourne, "but the contractor has designed a very clever spring-loaded shutter that can be used inside the casing, and is flexible enough to make some adjustments as required."

With the contract awarded, Seacore brought in its marine jack-barge Excalibur to carry out the foundation works. The barge, explains Seacore project manager Phil Wilkinson, was originally built to service the growing market for offshore wind farms. It is a 60m by 32m barge with eight 1.8m-diameter legs on which it can jack itself up in order to carry out installation of large diameter piles. The barge carries all the necessary drilling and down-hole equipment, it has a 280t-capacity crane with 54m boom and 12m fly boom, and it also incorporates all the facilities necessary to allow the crew to live on board permanently while carrying out the work.

The barge must be moved by a tug, and this was kept on hire during the work on site at Kincardine. A GPS survey system on board the barge is linked with the site system and used to position the rig with an accuracy of a few centimetres. The piling gate which carries out the final positioning and levelling of the pile before installation, operates in what is called the 'moonpool'; a cut-out of the barge which gives access directly to the water. Once the rig is in position, the casing is delivered - it is floated out on inflated airbags from the shore - and lifted on board, raised upright and presented to the piling gate. The gate grips the casing, manoeuvres it into position and a theodolite is used for the final positioning of the pile. The tolerance at this stage is about 75mm, says Wilkinson, which can be quite tricky to achieve on a pile that is nearly 4m in diameter. The casing is either pushed, or a vibro hammer is used to vibrate it into position, after which the down-hole drilling equipment is lowered into place. "We use our own 12 inch internal Seacore drill pipe reverse-circulation drill," says Wilkinson, "and once we have cleared out the softer material, we get the drill bit into the casing toe, after which the casing toe and the casing start to follow the drill. Each time the casing moves, we check its verticality again."

Ground conditions are extremely varied on this project, and one essential part of the site investigation was to take boreholes at each and every pier location. The process was invaluable, says Bourne, and there have only been a couple of occasions where the pile installation uncovered a different story, requiring changes to the pile design. "The geology is probably the most horrendously variable and complex that you could imagine. It is very faulted and variable, with some 20m of overburden ranging from soft marine clays through boulder clays, gravel and so on. The rock is a mixture of weak mudstone and extremely strong sandstone - all of which is mixed together at various locations and depths." Rock sockets vary from three or four metres to nine metres, but it is impossible to predict what will be found at any location. The design pile length was determined based on the boreholes, but this was continuously checked and upgraded while drilling was being carried out.

Once the final depth is reached, the rock socket is cleaned out by running the reverse circulation process for about half an hour. "We monitor the water that comes out, checking its specific gravity and mud content; once we are sure that the sandstone is clean, we remove the equipment and install the rebar cages which can be up to 70t in weight," says Wilkinson. Concrete is delivered from the batching plant on shore, via a gangway that the contractor has designed to attach to the tops of the pile casings as they are installed, which also provides access for construction of the pier shafts. The average volume required for each of the 3m-diameter piles is about 180m3, with the larger piles taking anything up to 300.

Wilkinson says the original subcontract programme allowed a four and a half day cycle for each of the 3m piles - although they managed to install one of the piles in just three days, it averaged at about the rate expected, he says. The larger piles were programmed to take seven days each, but the crew got this down to just five days. Although the pile work started a little later than expected, in October last year, they caught up on the original schedule and finished on time at the beginning of March. The delay was caused by the fact that access to the site for the Excalibur rig is very tight - it has to squeeze under the existing Kincardine Bridge with very little space to spare, and in order to do this, it had to undergo extensive alterations. "Lots of the equipment has to be removed from the deck, and we have to break down the crane," says Wilkinson. The same process must be carried out when the rig leaves the site, and meeting the schedule is all-important given that there is only a short window each month when the water level is low enough to allow Excalibur under the bridge.

With the foundation contract almost complete when Bd&e visited the site, attention had turned to the superstructure, which was starting to edge slowly out over the river. According to Bourne, the structural system of the deck will be effectively 'partially prestressed' which is now allowed under UK codes as long as all the cables are external. Bourne explains how the designers have created a partially prestressed deck which enables the contractor to reduce the tonnage of prestressing by almost half. Normally in this type of structure built using incremental launching, the prestressing cables used for the launch are small and are internal, and the continuity prestressing is larger and usually external. In the design of the bridge for the  Upper Forth Crossing, both launching and continuity cables will be external, and reinforcement will be used to provide some of the additional launching and long-term capacity.

"We originally planned on a traditional, fully prestressed bridge which was classically designed in terms of prestressing," says Bourne. "But after the tender we rethought it and came up with this alternative. Under the permanent loads, the bridge acts as a fully-prestressed bridge, but under live loads it effectively acts as a partially-prestressed bridge, and as with a reinforced concrete structure, we monitor it by checking the crack widths." A modest amount of additional reinforcement has been added, but the amount of prestressing required has been substantially reduced.

Because the bridge is being launched from one side of the river only, the final push will be a 1,200m length of deck. This will equate to some 32,000t which will require a jacking force of about 1,800t to move, although the contractor will provide a capacity of 2,000t just in case. At the back of the launching area, the joint venture has installed a set of jacks with a capacity of 1,200t; the same jacks that were employed by the joint venture on the Thurrock Viaduct in east London. A further 500t capacity will be provided by Eberspacher jacks on the north abutment, which will lift and push, as well as providing the overall stability for the launching procedure. The scheme is also designed to enable the installation of another set of Eberspacher jacks on pier 14 in the middle of the river, which can provide 300t capacity if required, although the team hopes not to have to install this set. “We have provision for a further 300t near midspan if eventually required,” says Osborne, “but we hope not. The pressure increases incrementally as we cast and launch more deck. We went to great lengths to install the casting beams and the remainder of the launch infrastructure to very precise tolerances. This is clearly paying off in that the pressures we are experiencing so far are well below those calculated.”

The 1,200t jacks react against casting beams 110m long, which are extensively piled to resist the pushing force during the launch and are the foundation upon which deck spans are cast insitu on the line of the launch and the bridge when complete. Each span is cast and incrementally-launched in its entirety; the casting and launching cycle taking about two weeks.

The standard 45m deck span is cast in seven pours, while the 12m-long section of each span that will eventually come to rest over the pier, is cast in one. This is where all the prestressing is applied, hence it has a much more complex layout and construction and it is important that the concrete has time to reach full strength before the launch. To keep it off the deck critical path the pier section is constructed on-line but well behind the main deck span under construction. The remaining 33m of each deck span is cast in three intermediate 11m sections, with base, walls and top-slab cast in two pours. The pier section is jacked forward into position in readiness for the casting of the last of the three intermediate sections.

Printer friendly version

Email this article to a friend