Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

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

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

An earthmoving experience

EARTHWORKS

A huge amount of work has gone into earthwork construction and providing a stable trackbed for the high-speed trains on CTRL, all in an environmentally responsible way. Alan Phear and Andrew Lord report.

Like all high-speed railways, CTRL has to meet very tight tolerances on track alignment to ensure safety and passenger comfort. Also, the financial success of the link depends on its availability to take revenue earning traffic, which means downtime for maintenance must be minimised.

A very stable trackbed and earthworks were required to achieve these aims. The other main objective of the design was to enable the earthworks to be built to a satisfactory standard of design and longevity at minimum cost. This required the maximum re-use of on-site or locally available materials. Waste minimisation and sustainability were also important considerations.

Most of the CTRL, except in the tunnels, is on ballasted track. The trackbed comprises three layers of high quality, imported granular materials over the earthworks subgrade. These granular layers are:

lA minimum of 300mm of ballast under the sleepers;

lA 200mm thick layer of subballast;

lA 350 to 700mm thick layer of prepared subgrade (similar to capping on highway projects), depending on the ground and groundwater conditions.

The combination of high-speed and repetitive high axle loadings (170kN for Eurostar trains but heavy freight can be up to 250kN) from trains running at up to 300km/h results in much higher dynamic loading than would occur, for example, under a typical motorway pavement.

The CTRL railway earthworks and trackbed were designed using the International Union of Railways (UIC) Code 719R. The recommendations for high-speed lines are largely based on extrapolating to higher speeds an empirical approach developed from experience on conventional railways (ie up to 200km/h).

It was found that the soils classification on which the code relies did not take account of local knowledge and experience. The recommendations were improved in places by French TGV best practice, and thus the design and construction of the earthworks and trackbed involved much cooperative teamwork with French colleagues, in particular, Alain Gonon, Alain Hocke and Pascal Babin from Systra.

A key aspect of trackbed design is trackside drainage. This efficiently removes surface water and, in 'wet' cuttings, maintains groundwater level below the base of the prepared subgrade. This minimises the amount of water in the trackbed and reduces any tendency for 'pumping' due to cyclic loading from trains.

One of the aims of the design was to achieve a trackbed with a reasonably consistent stiffness.

During the design process, RLE and SNCF (French Railways) carried out a study of the maintenance tamping necessary to maintain the ride quality on the TGV Nord (Paris-Calais) highspeed line in northern France.

One of the most striking conclusions was that by far the most frequent tamping was required at the transitions between underbridges and embankments - ie where there were relatively sudden changes in trackbed stiffness from less stiff embankments to more stiff structures.

As a result of the TGV Nord experience, backfill transitions at underbridges on more recent TGV lines and CTRL were carefully engineered and include several blocks and layers of cement stabilised and well-graded granular material to give a gradual transition in stiffness.

Prepared subgrade stiffness was an important acceptance test in the CTRL's construction. Plate bearing tests were carried out by loading a 600mm diameter plate to a maximum pressure of 250kN/m 2in the first load cycle and then reloading it to 200kN/m 2in a second load cycle. The test used standard French equipment and was undertaken at 40m staggered centres along the trace. The requirement was to achieve a reload elastic modulus (E v2 ) at the top of the prepared subgrade of between 80MN/m 2and 500MN/m 2.Values at the upper end of the range were obtained over insitu and remoulded chalk subformations, and here, the purpose of the granular layers was to reduce the stiffness of the underlying geology. However, further east along the route, values at the lower end of the range were obtained for prepared subgrade over the clays and fine grained granular materials there. The main purpose of the granular layers here was to stiffen up the trackbed.

One of the main differences between the CTRL and most previous schemes is the amount of landscaping or mitigation fill:

There is almost as much mitigation fill as fill for railway and highway embankments, and considerable effort has been put into blending the mitigation fill areas into the Kent landscape. On Section 1, state of the art GPScontrolled earthmoving equipment was used by several contractors to achieve the designed profiles for the mitigation areas.

As a result of good practice in thorough desk studies, ground investigation and specialist geotechnical interpretation design, few problems and/or surprises were encountered with the earthworks materials during construction of Section 1. Baseline monitoring, mainly of groundwater levels, was carried out for nearly five years along the full length of the route. Discussions were held with local materials engineers in an attempt to understand the idiosyncrasies of the chalk and fine grained granular soils in Kent.

Comprehensive interpretative reports were prepared for each material and cut.

To achieve the minimum maintenance requirement, no overconsolidated clays were allowed to be used in construction of the railway embankments. The main railway embankment fill materials were Upper Chalk (35% of Section 1 route) and Folkestone Beds sand (65%). End product compaction specifications were adopted - with a maximum of 10% air voids for the chalk, and a minimum of 95% of target dry density for the sand. Head materials, overconsolidated clays and the silts, sands and clays of the Sandgate and Hythe Beds were generally used for the mitigation earthworks.

The chalk, despite its difficult reputation, proved to be an extremely good and forgiving fill material as long as it was carefully handled. The mild winters during the construction of the CTRL chalk earthworks helped this.

The techniques adopted for the successful handling of this material have been described in the earthworks section of the forthcoming CIRIA report on the Engineering Properties of Chalk.

The Folkestone Beds sand is generally a uniform or very uniform material. Its maximum dry density varied and was very sensitive to subtle changes in its grading. Despite being dense or very dense insitu, it exhibited a reduction in volume when recompacted in embankments. It was a good fill when confined and compacted, but extremely erodible, and the nearly completed earthworks suffered many washouts during the exceptional rainfall in Kent in the autumn and winter of 2000/01.

Washouts were also a problem in the Folkestone Beds sands cuttings before vegetation had established. This was a rare example of a conflict in the earthworks design between the construction requirement for rapid establishment of vegetation to reduce short term surface erosion and the future railway operator's requirement for slow growing and thus low maintenance vegetation.

The design of the cuttings followed typical British earthworks practice. During the construction of the interchange between the A289 Wainscott bypass and the M2 near Strood in 1997, the top of the chalk was found to be extremely irregular and almost karstic.

This was also observed in the CTRL cuttings in the same area - if anything to an even greater extent, with the chalk rockhead varying by up to 6m over a typical 10m length. Secure yet cost-effective methods for addressing this required extensive investigation and deliberations. These resulted in special trackbed and trackside drainage details along nearly 3km of the route, affected by solution features and the use of soil nailing to stabilise solution feature infill material in the cutting side slopes.

Achieving acceptable sub-formations in cuts in the Sandgate and Hythe Beds silts, sands and clays was sometimes difficult, even with temporary drainage, because of the heavy compactive effort required on the prepared subgrade. Acceptable sub-formations were produced using various methods including dig out and replacement with granular materials and lime cement treatment.

Alan Phear is RLE lead earthworks specialist and Andrew Lord is RLE geotechnical design reviewer.

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Please note comments made online may also be published in the print edition of New Civil Engineer. Links may be included in your comments but HTML is not permitted.