Innovative geotechnical solutions, including the first use of a US drilling polymer in the UK and dry soil mixing, are being used to cope with difficult ground conditions and restricted working areas on Contract 440 of the Channel Tunnel Rail Link.
A PARTNERING approach is being used to tackle geotechnical issues in the design and construction of Contract 440 of the Channel Tunnel Rail Link, from the tunnel portal to Ashford in Kent.
Foundation contractor Stent is working with main contractor Balfour Beatty, client Union Railways (South) and project manager Rail Link Engineering (RLE) to formulate geotechnical solutions to deal with the highly varied ground conditions on this section of the link.
Geology ranges from dense sands and sandstone to over-consolidated clays and peats. Drilling techniques have had to be tailored not only to geotechnical conditions but to restrictions imposed by the route of the new railway, which was chosen to minimise environmental impact.
Because some of the soils are water bearing, support was needed during piling to keep boreholes open.
However, because some of the piles are up to 32m deep, casing could not be used. Bentonite was the obvious solution to prevent collapse, but created its own problems.
Space restrictions on some sites led to predictions that setting up the bentonite plant would be almost impossible, explains Stent project manager John Spence.
To overcome this problem, Stent proposed the first UK use of a polymer-based drilling fluid from the US.This consists of an inert polymer which when mixed with water acts in a similar way to bentonite.'However, the polymer needs none of the bulky and elaborate equipment that bentonite needs to mix it, clean it and store it, ' says Spence.
The polymer, supplied by US firm KB technologies, is introduced into the pile (or diaphragm wall) whenever soils are liable to flow or water ingress is likely to compromise excavation stability.After drilling is completed, chemicals in the polymer cause solids to drop out of suspension relatively quickly and alter viscosity, bringing the drilling fluid parameters into specification. Solids collect on the pile base where they can be removed using a conventional cleaning bucket. The pile can then be concreted in the normal way.
Spence explains that using these chemicals means that the drilling fluid does not need to be replaced, as is the case with bentonite, so storage space for drilling fluid can be considerably reduced.
Stent developed an extensive testing programme with RLE to ensure the polymer had no adverse effect on reinforcement bond strength compared with other drilling fluids.
Drilling fluid was needed on 19 locations on contract 440, many of which only required a handful of piles. The compact nature of the polymer set-up meant moving from site to site was relatively easy, whereas to mobilise a full bentonite set-up for each location may have eaten into valuable programme time.
The downside, Spence says, is that the raw polymer is considerably more expensive than bentonite. However, he believes there are indirect but unquantifiable savings as well as a direct cost savings in equipment that make the use of polymer economically viable overall.
Elsewhere on the contract, Stent has encountered soft clays and peats, which has led to the use of another innovative technique.
At the approaches to Sandling tunnel, a layer of peat up to 6m thick affects 300m of the new line.The dynamic loading of a high-speed train travelling at nearly 290km/h meant the peat had to be replaced with 'something more substantial' However, here the new line runs parallel to and only 5m from the existing Ashford to London railway.Removing the peat would have required installation of substantial retaining walls (either temporary or permanent) extremely close to the live railway used by Eurostar and other mainline trains.
'The walls would need to retain some 8m with railway surcharge and there was the problem of groundwater, ' says Spence.
This initial solution seemed extremely costly in both time and money. Instead, working in joint venture with Swedish geotechnical firm Hercules Grundlaggning, Stent proposed the Limix method of dry deep mixing, to produce soil/cement columns.This improved the stiffness and shear strength of the ground, enabling it to cope with the dynamic and static loading of the high-speed trains.
Field and laboratory investigations and laboratory trial mixing were carried out before installation and testing of trial columns for final design. The proposed method was approved after a cost analysis review.
Dry deep mixing involves disaggregation of soil insitu and its mixing with dry binders and fillers to form the strengthened soil; at CTRL 440 the most suitable binder was cement.
Mixing is typically carried out using rotating single mixing tools.The tool is first inserted into the ground to the desired depth. The binder is fed through a nozzle in the mixing tool with compressed air from a separate binder tank and then mixed with the soil during retraction of the mixing tool. A computer controls the rotation speed and withdrawal rate of the tool.
These two elements, together with the shape of the tool, control the mixing energy generated. The weight of added cement is measured by weighing units located on each tank and is updated constantly.The withdrawal rate is continuously monitored and altered to ensure even amounts of cement are mixed into the soil. The relevant installation parameters are recorded and stored on PC and presented to the client.
On Contract 440, columns are installed from a 1m thick working platform and within four hours of installation 1.5m of surcharge is placed.This helps consolidate the soil mixed peat by squeezing out water and entrained air as the cement starts to set, 'locking in' the improved shear strength.Some 7500 columns will be installed on the contract.
Column testing using standard column penetration test has shown soil shear strengths to improve from 10kPa to more than 100kPa, says Spence.
'Measurements from settlement plates show that after about 15 days the primary settlement starts to even out and after a total of 40 days there is no further creep settlement.'
Seismic cone penetration testing performed in the stabilised area shows that a shear wave velocity of 250m/s to 320m/s was obtained.