Crossrail sets new challenges for tunnellers

There may be signs of an underground revival, but tunnelling beneath busy cities is a tricky business. Jon Excell reports

It is one of the largest and most ambitious engineering projects in Europe, but once complete, many thousands of Londoners will go about their business largely oblivious to its presence.

It is, of course, Crossrail, the 120km Berkshire-to-Essex railway, which will pass beneath London’s busy streets along 42km of underground tunnels.

However, despite the superlatives, the engineers currently puzzling over how to thread Crossrail’s tunnels though London’s tangled subterranean infrastructure are not alone.

Last year, 140km of tunnels were built beneath Shanghai. In Rome, engineers are pondering a new metro line that will pass beneath the Coliseum. In Amsterdam, meanwhile, the North-South metro line, which will connect the central districts with the suburbs north of the river, will require 6km of new tunnels. Some would say that, against a backdrop of increased urbanisation, congestion and pollution, tunnelling is undergoing a worldwide renaissance, but the challenges of tunnelling beneath a city are huge. Varying geology, an array of sewers, tunnels and pipes, and the foundations of buildings make it one of the most delicate and complicated balancing acts the world of civil engineering has to offer.

One person who knows more about this balancing act than most is Prof Robert Mair, head of civil and environmental engineering at Cambridge University and director of geotechnical consultancy GCG, which lists the channel tunnel rail link (CTRL), the Jubilee Line extension and now Crossrail among its recent projects.

Talking recently during a public lecture at the Royal Society, Mair said that tunnelling beneath cities raises a number of key questions: ‘First, what are the properties of the ground, how feasible is it to construct a tunnel and can its stability be ensured? Can we predict the ground movements and subsidence likely to be caused? What will be the potential effects on the buildings and how can the buildings be successfully protected?’

Fortunately, the London Clay, through which much of Crossrail’s tunnels will pass, is well suited to tunnelling. Indeed, its strength is one of the reasons that the London Underground was built relatively early. However, the clay itself does vary and one of the first challenges is characterising these tiny fluctuations in its properties by analysing the load response of samples recovered from boreholes. ‘The microstructure of soils is… very important,’ explained Mair. ‘There is an abundance of diatoms — tiny shells — that has a very significant effect on the way the clay behaves both in terms of its strength and how compressible it is.’

However, although London’s geology is better suited to tunnelling than Shanghai — where the clay has the texture of toothpaste — it is not entirely uniform. ‘Geology varies in the London area,’ he said. ‘The 7km tunnel for the CTRL between Kings Cross and Stratford traverses a number of geologies: Thanet sand, a mixture of clays, sands and gravels and London clay.’

These variations will have a bearing on the type of tunnelling machines used to excavate Crossrail’s tunnels and it is likely that construction methods will vary according to conditions in an area.

Crossrail is expected to use around eight tunnel boring machines to build the 6m-diameter tunnels through which the trains will run. Excavating an estimated eight million cubic metres of soil, these enormous machines will burrow beneath the streets of London for 24 hours a day, erecting pre-cast concrete tunnel sections in their wake.

In some areas, such as larger platform tunnels and shorter passenger circulation tunnels where the geology permits, boring machines will not be used at all. Instead, holes will be cut with other excavation machinery and lined with sprayed concrete. Mair explained that this approach was used beneath Waterloo station during the construction of the Jubilee Line extension.

Perhaps the biggest challenge posed by any tunnelling project, however, lies in predicting and mitigating the effect it has on surface structures. For instance, the Crossrail team has to carefully consider the effect of its actions on the 4,500 buildings — many of them listed — that lie directly above the route.

According to Mair, it is inevitable that tunnelling will have some effect on the surface, but modern techniques can generally prevent disaster. ‘However well engineered a tunnel may be, there will be some settlement,’ he said. ‘The question is how to reduce the magnitude of that to acceptably small values.’

The first way to reduce movements is by increasing support to the ground from within the tunnel itself. This can be achieved using tunnel boring machines that can control the pressure on the tunnel face and limit the disturbance to the surrounding ground and the surface.

Ground movements can be further reduced by a relatively new technique that involves injecting liquid cement into the ground between the tunnel and the foundations of surface structures.

This approach, known as compensation grouting, was pioneered by Mair and used for the first time during the construction of the Jubilee Line extension.

It was particularly useful for controlling the movement of the clock tower housing Big Ben, which was potentially threatened by the excavation for Westminster station, explained Mair: ‘Big Ben [sic] was built in 1860 and has very shallow foundations that are only about 34m from the edge of this 40m-deep station… There was the potential for [it] to lean or tilt towards all this new work and away from the palace of Westminster.’

During this work, a shaft was sunk next to the tower in Bridge Street. A drilling rig was then lowered down the shaft and 60m-long steel tubes were drilled horizontally beneath the foundations of the tower.

When the excavations began, the tower itself was heavily instrumented and linked directly to computers that were able to tell precisely how much it was moving. Liquid cement was pumped into the steel tubes beneath the tower in direct response to these readings. ‘It was determined that Big Ben could tolerate up to 15mm of tilt before action would be needed,’ said Mair. ‘For the first year or so, it was tracked very carefully and when it reached the 15mm point grouting started.’

The fact that, several years on, tourists are not referring to the leaning tower of Westminster is pretty strong evidence that the technique worked. Since then, compensation grouting has been used to protect King’s Cross Station during tunnelling works for Eurostar and Mair believes it is likely that the approach will be used in Rome to protect the Coliseum from settlements caused by excavations for a new metro line in the city.

Sometimes, however, given the complicated nature of the subterranean world, there’s simply no alternative but to go back to the laboratory.

‘Underground is very congested,’ said Mair. ‘The reality is that, in every modern city in the world, there is an awful lot down there — there are existing tunnels, pipes and lots and lots of pile foundations — and that is very important when thinking about new tunnelling projects.’ A good example is Amsterdam, where engineers working on the North-South line must be careful not to disturb the timber piles that support many of the city’s buildings.

One intriguing method of gaining a greater understanding of the complex forces at work is to build scale models and then put them in a centrifuge. He explained that, by spinning a scale mode at speeds up 200mph (322km/h), it is subjected to around 75 times the force of gravity and acquires similar stresses as the full-scale tunnel it represents. ‘If you get the scaling right, it reproduces what actually happens at full size,’ he said.