Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to the newest version of your browser.

Your browser appears to have cookies disabled. For the best experience of Construction News, please enable cookies in your browser.

Welcome to the Construction News site. As we have relaunched, you will have to sign in once now and agree for us to use cookies, so you won't need to log in each time you visit our site.
Learn more

How Eurostar wil cross the marsh

PILING & GROUND ENGINEERING

The team that solved the wobble on Arup's Millennium Bridge over the Thames used the same diagnostic software to put Eurostar trains on a sure footing on the Channel Tunnel Rail Link's 7 km piled slab across the Essex Marshes. Paul Wheeler reports

IT IS NOVEMBER 2000 and CTRL lets a complicated test pile contract to assess the behaviour of several types of piles in the Essex Marshes to see how they might behave under the cyclic loading generated by the high-speed Eurostar train traffic.

With more than 7 km of piled slab needed on the 12 km stretch across the marshes between Thurrock and Dagenham, the client's engineer, RLE, believes there might be considerable savings to be made by arming the main contract bidders with detailed pile test data.

A year and a half down the line this trial has without doubt reaped significant benefits for the project - but not quite those that had originally been anticipated.

'The analysis suggested we would need extraordinarily long foundations to deal with the long-term settlement caused by the repetitive cyclic loading of the passing trains, ' explains RLE's geotechnical manager Nick O'Riordan.

But the team also now knew that if it had a good dynamic analysis model it could refine the design and make it less conservative. Fortunately for RLE, Mr O'Riordan is a director with Arup and some of his colleagues were having dynamic response problems of a different nature - the Millennium Bridge.

He realised that the dynamic finite element programme, LS Dyna, used to back-analyse the bridge's wobble, might be just what the Essex Marshes pile designers needed.

Mr O'Riordan says: 'Suddenly we had a state-of-the-art tool.

As a by-product of the Millennium Bridge, Arup had developed new expertise in the analysis of things that move.'

'Arup's Advanced Technology Group used the programme to investigate the transient response of the structure at various speeds, ' explains Rod Allwright, RLE's geotechnical design representative and field engineer.

'This provided assurance that the structure would not be subject to resonance and allowed a rational assessment of dynamic and quasi-static pile loads to be established as a basis for the pile design.'

The software even modelled the suspension mechanism of the Eurostar trains. This made it possible to predict accelerations in the coaches and even how to design the foundations so that passengers' coffee would not spill.

Arguably of greater importance, the analysis showed dynamic loads were less than those predicted by conventional dynamic load factors, and that a more economic design could be applied safely.

On a project of the scale of CTRL contract 310, this will lead to impressive savings. As a rule of thumb, Mr O'Riordan estimates that reducing pile length by 1 m will lead to savings of a million pounds.

The piled slab typically comprises a 10 m-wide, 450 mm-deep reinforced concrete slab built in 60 m lengths between movement joints. It will be supported by at least 6,000 piles, typically in rows of four and six at 5 m centres.

Each 60 m slab section will have a combination of piles to suit the soil conditions, with layouts and depths tweaked to provide the most economical design. This is a long way from the 'one size fits all' approach that had been assumed at the outset.

'Conceptually, we thought the project was quite simple but that was before the dynamic analysis, ' says Mr O'Riordan.

With hindsight, with variable ground conditions and the large number of utilities crossing the route, a standardised foundation design was never going to be achievable.

It's hardly surprising, then, that determining the most economic foundation layouts has taken greater design input than had been anticipated. But, as Mr O'Riordan says: 'Design costs are miniscule compared with construction costs.'

GROUND conditions in the Essex Marshes are up to 10m of alluvium, underlain by floodplain sands and gravels, Tertiary clays and sands, then chalk.

RLE's original outline proposals included either 600 mmsquare driven pre-cast piles founded in the flood plain deposits or longer 600 mm-diameter CFA piles. The CFA piles are needed either in proximity to houses or services where noise and vibration preclude the use of driven piles or simply where the floodplain deposits are thin to the extent that they do not provide an acceptable founding layer.

The advance pile testing contract in November 2000 showed that the longer CFA piles working in friction had to be designed to a higher ultimate capacity than the end-bearing precast driven piles.

Although driven piles offered by far the best technical and most economic solution, working restrictions mean they cannot be used on at least half of the 6,000 piles.

Given the comparatively poor performance of the CFA piles, EuroLink is currently promoting the use of the Belgium screw pile system as an alternative. (See below, left) In December last year, it instigated a second phase of trial cyclic pile load testing (see below) to assess the performance of screw piles compared with the benchmark dynamic tests from the earlier trial. This demonstrated that the screw piles performed considerably better than CFA, but not as well as driven precast.

The conclusion was that, compared with CFA piles, screw piles could be typically 5 m shorter. As a result the project team, says O'Riordan: 'is gravitating away from CFA to screw piles wherever possible'.

EuroLink is currently talking to a number of contractors - not all of whom are Belgian - and an announcement of the successful bidder is expected imminently.

CTRL client Union Rail awarded the £178 million contract 310 in October last year to EuroLink, a joint venture comprising Morgan Est (then called Miller Civil Engineering) and French firm Vinci. Contract 310 was the largest single contract on CTRL phase 2 until earlier this month, when Union Rail merged the St Pancras extension and refurbishment.

EuroLink is presently gearing up to start foundation work on site with a further series of test screw piles next month. Contract completion is scheduled for December 2004.

Screw pile technique THE SCREW pile is not a new piling system. The original Atlas screw pile has been used in Belgium for decades, but the few attempts to bring it to the UK in the past have had little impact.

It is a reinforced cast-in-situ concrete pile formed using a steel casing with a single helix auger attached at its base. The resulting pile has a characteristic corrugated shaft so that it resembles a screw, hence its name. During installation the casing, complete with sacrificial steel pile tip, is screwed in to the soil while applying a vertical pushing force. After the required depth has been reached, reinforcing steel is inserted and concrete pumped into the shaft.

Next the casing is removed by reversing the rotation, which uncouples the casing from the sacrificial tip. As the casing is extracted, concrete flows into the helix formed by the upward rotation of the auger. This creates a regular helical concrete fin around the pile shaft, increasing both its effective base diameter and shaft area.

The steel tip removes concerns over soft debris at the pile base. Because it is a 'slow' displacement method, it can be used where noise and vibration are restricted. Its chief drawback is that casings are only available up to 560 mm in diameter. This results in relatively slender piles, which restricts the amount of steel reinforcement and can be an issue where lateral resistance is needed.

The screw pile is particularly suited to Belgium's deep and loose sand deposits but so far the technique has not travelled well owing to lack of performance data in other soil conditions. 'This is a pity, ' says Mr O'Riordan: 'Because it is fundamentally a better product than CFA.'

The Belgium piling industry, appreciating that the technique might have good export potential, has sponsored trials to provide data to support the tried and tested, but unscientifically proven, design basis. This should significantly help the spread of the technique. It's almost certain use on CTRL will be a big boost.

Cyclic pile trial THE ORIGINAL cyclic pile trial included tests on 600 mm square precast sections. The sections were specially manufactured and installed by Aarsleff, and are the largest precast pile sections ever installed in Europe. The sections were selected because it was thought that using fewer higher capacity piles could lead to the most economic pile layout.

The dynamic finite element analysis showed that selecting these large and relatively massive sections had unexpected benefits on the slab's behaviour under working load. If conventional precast sections had been used, the behaviour of the slab under dynamic loads would have been more extreme, says O'Riordan.

It all comes down to the mass and the ability of the train generated dynamic wave to accelerate the structure: 'The bigger piles are a significant part of the structure's overall mass and its ability to resist shakedown.'