London river crossing preliminary design 2 Dec 2014

Preliminary design for the proposed £700 million twin-bore Silvertown road tunnel under the River Thames in East London calls for a single 12.5m diameter TBM, likely to be an EPBM rather than a slurry machine.

Fig 1. Silvertown and Blackwall tunnel alignments

A second round of public consultation on a preferred alignment between Greenwich Peninsula on the south side of the river, and Silvertown near the Royal Victoria Docks on the north side (Fig 1), ends in two weeks time. The project, which has been in development by Transport for London (TfL) since 2009, is on the UK Government’s list of 105 Nationally Significant Infrastructure Projects (NSIPs). A planning application is expected in the last quarter of 2015, to be followed by a Public Planning Inquiry process of up to 18 months.

According to TfL’s latest project timetable, construction procurement for the 2 x 2-lane tunnel could begin in 2016-17. Contract award is scheduled for 2018, with project completion set for 2021-22. It is proposed that introducing tolls for the nearby Blackwall Tunnel (which is situated only a few hundred metres away on the other side of the peninsula and is currently free to use) will fund the construction of the new Silvertown Tunnel. The tolls would also be applied to the new infrastructure when it is completed.

The original bore of the Blackwall Tunnel was constructed between 1892-97, and now serves northbound traffic under 4m high. A second eastern bore for southbound traffic was completed in 1967, but both tunnels are already operating at full capacity and neither is particularly suitable for the demands of modern-day traffic requirements. Both are freqently blocked as a result of road traffic incidents, especially at peak hours, and there are huge congestion problems in the area.

Blackwall Tunnel cannot cope with traffic demand

Preliminary tunnel design for the Silvertown Tunnel, which would relieve pressure on the Blackwall tunnels, is being carried out by Mott MacDonald, with Atkins, Hyder and London Bridge Associates also involved as part of the engineering design team. Earlier studies examined TBM and immersed tube methodologies, but TBM excavation has been identified as the lower cost option and is currently the favoured solution.

Cut-and-cover considerations

The Silvertown Tunnel is currently designed along a curved alignment under the River Thames (Fig 2). It is proposed that the TBM will launch from a chamber within a 250m-long cut-and-cover section on the Silvertown side of the river. This will be accessed via a further 250m of open cut excavation, a configuration that will be repeated on the opposite side of the alignment for connection with the existing road infrastructure.

Fig 2. Open cut, cut-and-cover and bored alignment design

On the Silvertown side the cut-and-cover section will cross under the Docklands Light Railway (DLR) viaduct, although provision was made for a future “Blackwall Third Crossing” under span 2 when the viaduct was originally built. Clearance under here is just 6m, a factor that will limit the use of traditional piling equipment that will be employed along the alignment’s other cut-and-cover sections. Additionally, two 1.8m diameter rising mains that form part of the Royal Victoria Dock drainage discharge into the Thames will need to be diverted.

Geological risk

One of the principal features of the preliminary design is the low level of cover above the tunnel crown. At the TBM launch site this will be just 10m, reducing to as little as 6m as it passes under the relatively shallow river bed (16m), and to within 2-3m of the foundations of the south river and north river retaining walls (Fig 3). Preliminary design drawings (October 2014) show that at its deepest point the tunnel invert is only 30m below ground level.

Fig 3. Cross sectional alignment of Silvertown Tunnel

Although further geotechnical scoping will be required, especially in relation to the design of the cross passages, historical geological data shows that a predominantly mixed and “challenging” geology will feature a layer of stiff London Clay overlying the dense sands of the Lambeth Group and the dense gravels of the Harwich Formation. However, large sections of both the open cut and cut-and-cover sections on both sides of the river, as well as in significant reaches of the TBM alignment at the tunnel crown, are expected to run through an overlying strata of River Terrace Deposits, characterised by silty, sandy gravels.

This, coupled with the shallowness of the alignment compared to similar river crossing projects, presents a project risk, and will be “significant in TBM selection and specification as it reduces the range of face pressures within which the TBM can be operated safely and efficiently.”

TBM selection and specification

Early studies suggested a TBM of 14m diameter, but this has now been reduced to 12.5m. Although a final decision will need to be based on further geotechnical information, the preference at this stage – given the fact that the TBM bored section is only 1,000m long – is for a single EPBM which would either be turned around on the Greenwich (south) side, or transported back to the original launch site at Silvertown.

12.5m o.d. preliminary Silvertown bored tunnel design

Mott MacDonald’s preliminary tunnel engineering report concludes: “The bored tunnel face will be mixed throughout the length of the drive. The River Terrace deposits are likely to be water-bearing and there is a likelihood of water-bearing sand and gravel lenses or channels in the Lambeth beds. In the building of the Jubilee Line Extension tunnels in this area, it is worthy of note that EPB tunnelling machines were employed and tunnelling from North Greenwich to Canary Wharf was executed in closed mode and with difficulty.”

The report continues: “The mixed ground conditions, the likelihood of encountering water-bearing strata beneath the river, and the experience on the Jubilee Line Extension indicate that an EPB type machine could be successfully employed. However, the nearby DLR extension to Lewisham [constructed shortly after the Jubilee Line] was constructed using a slurry machine in similar ground, albeit slightly lower in the geological succession than the proposed Silvertown crossing.”

The report concludes by saying there is a “high risk” of losing face support pressure through slurry loss along an alignment that features such a low level of cover along its whole length. This is considered a “critical” factor in TBM selection. Furthermore, selecting a convertible machine or a slurry machine would require the construction of a spoil separation plant, which is considered a “significant disadvantage” when compared to the advantages of an EPBM.

Cross passage design

Cross passages

Following consultations with the London Fire Brigade, and after completing assessments of the benefits of installing a fixed fire suppression system as part of the overall design, cross passages are to be spaced at a maximum interval of 350m. It was found that for this type of tunnel, installation of a fixed fire suppression system would result in safety levels that might be even higher than for a design which incorporated cross passages at 100m intervals alone. In all there will be three cross passages within the bored section and one within the TBM reception chamber at Greenwich. They are currently designed at 14m long x 4.5m wide.

Either segmental cast iron linings or SCL can be used effectively in London Clay without need for prior ground treatment, but it is expected that compensation grouting or localised dewatering methods will be adequate for those areas where there is a risk of encountering River Terrace Deposits or other water-bearing soils at the crown. In any case, expensive ground freezing is not considered likely to be necessary.