No-dig alternative for busy London streets Apr 1992

Shani Wallis, TunnelTalk

It was to be open trenches that would make way for new electricity cables through the streets of London until the tunnel option was investigated. This exploded the myth that tunnelling would be outrageously more expensive and troublesome. The tunnelled options now being built are proving highly acceptable not only to the client and the tunnelling industry but also to the businesses, traffic and crowds of tourists and pedestrians who are thankfully oblivious of the suffering they have been spared.

Shallow services such as these exposed at the start of shaft sinking at Sherwood Street would have plagued an open trench operation along its entire length
Shallow services such as these exposed at the start of shaft sinking at Sherwood Street would have plagued an open trench operation along its entire length

Like all urban public utilities, the installation and maintenance of electricity cables is an on-going programme. Traditionally, open trench has been the cheapest method of working on such shallow underground services. A principal reason is that the social cost of open trench work is not included in the feasibility or contract cost estimates.

The situation is changing, however. The substantial cost of traffic delays and commercial inconvenience caused is now being studied. As such, both legislators and clients are investigating the no-dig alternative. It is for this reason that London Electricity (LE), the now privatised electricity supplier for the London area, abandoned original plans to route a new high-voltage cable across London in an open trench and has chosen instead a cost effective no-dig alternative.

The new 132,000 volt cable is needed to link a supergrid sub-station at St John's Wood with a new sub-station concealed beneath the gardens of Leicester Square, London's tourist and theatre heartland. Work on completing the link began in 1988 when a contract was let to lay 3km of cable in trench from the St John's Wood sub-station to Duke Street sub-station. To facilitate cable jointing, up to 500m of roadway was open at any one time. This required a 5m-wide lane possession to accommodate a 3 x 3m trench.

It was after work had started on the open-cut contract but before award of the next leg of the link, from Duke Street to Leicester Square, that a tunnel alternative was examined. "Having suggested a tunnel and given LE's existing schedule, we had just six weeks in the early summer of 1990 to conduct a feasibility study and find a tunnel route through what must be the most congested underground space in London," said Maurice Gooderham, Head of Tunnelling with consulting engineer Howard Humphreys. "To avoid possible protracted negotiations and agreements with private property owners and others along a more direct route, the tunnel follows the centre line of the streets. Even so, it was necessary to investigate the location and types of building foundations as well as the route of many existing underground services, including London Underground's railway tunnels and stations, water and sewer tunnels, gas mains and other cabling."

Figs 1 and 2. A tunnelled alternative to the planned open trench laying of a new electricity cable has proved highly advantageous both environmentally and economically

These studies identified a zigzag 2km­long route some 15m below the surface (Figs 1 and 2). The optimum route through Piccadilly Circus could not be used as the underground space was already densely occupied. At this depth in the centre of London and north of the River Thames, the tunnel runs through London Clay, a highly favourable tunnelling medium. The extensive network of tunnels under London, particularly the Underground system, exemplifies more than a hundred years' experience of tunnelling through this material and verifies its suitability for application of simple, cost effective tunnelling methods.

Based on open-faced, hand mining methods and lined with precast concrete segments, Howard Humphreys estimated that the tunnel would cost about £1500/m. Under current market conditions in London, this is similar to estimates for open trench work when comparing direct costs.

The tunnelled route has many advantages. It is about 700m shorter than the planned open-cut route. This results in a corresponding saving of 700m of 132 000 volt cable, which costs some £7000/m. It also saves in transmission costs. With a design life of 125 years, the 2m i.d. man-entry tunnel provides better protection of the cable as well as space for other utility cables and room to carry out future maintenance and repairs.

Tunnelling under the protection of the shield using clay spades
Tunnelling under the protection of the shield using clay spades

LE accepted the tunnelled alternative and in January 1991 a £2.9 million contract was awarded to J Murphy & Sons as the lowest of 11 invited bids. Howard Humphreys was retained to provide construc­tion supervision. To satisfy itself that the tunnel was cost effective, LE also called for three open-cut tenders with one contractor bidding both. Murphy's tunnel bid of about £1500/m was competitive with the lowest open-cut bid of £1270/m.

This comparison is made without taking into account the social chaos or service diversion costs of an open trench operation nor the extra benefits of the tunnel. Howard Humphreys' design calls for four permanent manhole access shafts and 2km of 2m i.d. tunnel lined with bolted precast concrete segments. Several tight curves of 12m radius are required to keep the tunnel alignment under the streets. Larger 3.05m-diameter chambers are required adjacent to each access shafts to facilitate cable jointing.

At Maddox Street, the principal working shaft (shaft B) has been established in a narrow roadway beside St George's Church. A section of Sherwood Street has been closed to provide space for shaft C near Piccadilly Circus and, similarly, a section of Whitcomb Street near Leicester Square has been reduced in width for shaft D and its worksite. Shaft A is established on a traffic island adjacent to Grosvenor Square.

To complete the 2km of tunnel in the required time, Murphy chose to use three 2.3m o.d. open faced, hand mining shields manufactured by Stelmo of Kent. As an accepted cost-saving alternative to the bolted segmental tender design, Murphy is installing an unreinforced precast concrete wedge block lining for most of the tunnel drives. The nine segments and the key of the 750mm-wide ring are erected by hand directly against the surface of the excavated ground and the expanding wedge key is driven home by a shove ram on the shield. In the cable jointing chambers and in areas close to other sensitive services, a bolted segmental lining is specified. Segments for both the wedge block and the concrete rings are manufactured and supplied by C V Buchan.

Elevation of Sherwood Street shaft: discovery of old storage chambers belonging to a theatre give Sherwood Street shaft its unusual stepped shape
Elevation of Sherwood Street shaft: discovery of old storage chambers belonging to a theatre give Sherwood Street shaft its unusual stepped shape

Murphy arrived on site in March 1991 and began work by sinking the four shafts required. After a criss-cross maze of underground services lying just beneath the road surface was diverted, shaft excavation was carried out using the underpinning method. The work progressed normally except at Sherwood Street where the presence of old storage chambers belong ing to a theatre extend out under the road. To accommodate the structure, the top 9m of the shaft passes down through the chamber in 2.3m-diameter concrete rings. Once clear and for the next 14m, the diameter increases to the 3.7m needed for break out of the tunnel. Shaft B also has an anomaly in that a sewer cuts across side of the shaft. Instead of being rerouted, the sewer has been encased in concrete and will remain within the shaft.

In June '91, Murphy lowered the first of its three 2.3m o.d. hand shields down Whitcomb Street shaft to drive toward shaft C in Sherwood Street. In October '91, when Tunnels& Tunnellingvisited the site, the short 50m run from shaft D to the Leicester Square sub-station was com­plete and the shield was about 300m into the drive from shaft D to shaft C. The short 226m hand-built drive from shaft A to the Duke Street sub-station was also finished. At Maddox Street, the second shield was advancing towards the Grosvenor Square shaft and the third was about 150m into its drive to Sherwood Street.

To accommodate two consecutive drives from the Maddox Street shaft, Murphy engaged Blackwell Mining & Mineral Engineering of Derby to design, build and erect a temporary structure capable of accommodating all necessary equipment as well as providing space for materials storage and spoil bunker. Given the limited 5m-wide x 28m-long compound between St George's Church on the one side and a high-rise building on the other, the resulting facility is a 13m­high, triple-decked structure with a bunker capacity of 150m . A 5 tonne hoist attached to the central roof beam raises and lowers the muck skips through the shaft. It travels the length of the beam to discharge spoil at various positions into the bunker situated on the first deck. Two reversible chain conveyors, independently controlled and hydraulically driven, discharge the spoil from the bunker through an opening into lorries for disposal off-site. A guillotine door helps control the muck flow into the lorry beneath.

A service area on the first deck accommodates the power packs, electrical control equipment and compressors, while the second floor houses battery charging facilities. A 3 tonne crane, attached to and travelling the full length of the under side of the bunker, handles segments and materials to and from the stockpiles and unloads them from delivery trucks. Finite element analysis was used in the design to calculate stress values and ensure stability of the structure. "Because of the close proximity of the public walkways, extra and continuous safety awareness was needed during erection of the structure", said Frank Price of Blackwell Mining. So far, all drives have been, as predicted, in London Clay. In areas where the tunnel passes close to sewers and water mains, the specified bolted precast concrete lining must be installed from within the tail skin of the shield. A crew takes about one 10h shift to fit the tail skin to the 2.4m-long shield and convert from the wedge block to the bolted segmental lining. It takes about half a shift to convert back to the wedge block system.

All tunnel excavation was scheduled to be complete by March 1992, working one 10h shift/day, 5 days/week. Once complete, there will be little evidence, if any, of the operation which removed 20 000 tonne of spoil and installed some 2700 segments.

More of a good idea

Having realised the cost effectiveness of the tunnelled option, London Electricity accelerated future cable laying plans and included them into the current new works budget.

A drive of 350m to link with a sub­station in Carnaby Street was added to Murphy's contract. This 1.7m-diameter tunnel passes under Regent Street and is being excavated concurrently with the two other drives from the Maddox Street shaft. As a result, the Maddox Street installation must handle some 50m3 of spoil/day and stockpile three different sizes of segments. Provision of an adequate muck and materials handling set-up at the start of the contract at the Maddox Street working site ensured that Murphy could service all three drives and complete the Carnaby Street drive within the original contract period.

Taylor Woodrow's contract: the 1.2km Central Road electricity cable tunnel was driven in opposite directions from a single shaft
Taylor Woodrow's contract: the 1.2km Central Road electricity cable tunnel was driven in opposite directions from a single shaft

In addition, LE engaged Howard Humphreys to design and supervise the construction of a similar l.2km-long x 2.lm-i.d. cable conduit tunnel along Central Street in the London Borough of Islington to link the City Road sub-station with one at Beech Street. Feasibility was completed in January 1991, a detailed design was completed in early April 1991 and tenders were invited from six prequalified contractors on April 13. On June 10, a £1.2 million contract was awarded to Taylor Woodrow, the lowest bidder. Work started on site on August 5, 1991.

The Taylor Woodrow tender was based on completing two headings from a central 20m-deep x 3.7-i.d. working shaft out towards the sub-stations at either end. Again, the geology comprises London Clay, with the tunnel running close to its horizon with the Woolwich and Reading Beds. These were encountered at the bottom of the central shaft which, at 20m, is the deepest point of the alignment. Howard Humphreys again designed the tunnel as a hand-mined bolted segmentally lined tunnel to suit the equipment and expertise widely available in the industry. However, Taylor Woodrow also submitted a segmental wedge block lining alternative of its own design. "With a tight contract period of 43 weeks, this contract all about production", said Ross MacKenzie, director of Taylor Woodrow Civil Engineering. "We have therefore designed and are producing our own wedge block rings, which are wider and have fewer segments per ring than usual." Each 2.1m-diameter ring of the Taylor Woodrow design comprises five 750mm­wide x 120mm-thick unreinforced segments and a key. The segments, at 300kg, are too heavy for man handling. A simple erector system has therefore been added to the two new hand shields bought for the project. These were designed and supplied by Gordon Ince Ltd and manufactured by CDC (Manchester).

Along the alignment, the tunnel passes 4m over the running tunnels of London Underground's Northern Line, and under four major brick-lined sewers. In these areas, as well as in the 3.05m­diameter jointing chambers and where required by poorer ground conditions, bolted segments must be erected.

In Islington, Taylor Woodrow is able to work two 11h shifts every weekday with a maintenance shift on Saturdays. It has established the central working shaft on the footpath adjacent to a sports facility and has provided a walkway around the site. In late November 1991, the first shield was lowered down the shaft and started its 650m drive toward Beech Street in the Barbican complex. The second shield was launched a week later for the 450m drive in the opposite direction. Both tunnels were due to be completed by March 1992.

At either end, the tunnels must break into the basements of each substation. At Beech Street, excavation of any material at the site is restricted. As a result, a small 12m-deep pilot heading, raised by hand from within the tunnel, is subsequently enlarged to the correct diameter from the top down. Special tapered lining segments from CV Buchan are used to build the 'lobster back' or smooth curve from the vertical to the horizontal. Because the tunnelling shields will not actually breakthrough, they will be stripped of their components and the skins will remain buried behind the lining.

Once both projects are completed, LE will have two modern and valuable extensions to its current extensive network of cabling throughout the city.

Project: electricity cable tunnels Location: central London
Client: London Electricity plc
Consulting engineer: Howard Humphreys & Partners Ltd
Contractors: Contract A - J Murphy & Sons; Contract B - Taylor Woodrow Ltd
Contract data (Contract A): Contract A - 2.4km of 2.05m i.d. wedge block tunnel between sub-stations at Duke Street and Leicester Square with a 350m-long x 1.7m-i.d. spur to a sub­station in Carnaby Street;
Contract data (Contract B): 1.2km of 2.0lm i.d. wedge block tunnel between substations at City Road and Beech Street
Contract value (Contract A): £2.9m plus £0.5m for the Carnaby Street extension
Contract value (Contract B): £1.2m Contract period (Contract A): 67 weeks from March 1991 to May 1992
Contract period (Contract B): Contract B - 43 weeks from August 1991 to June 1992
Tunnel access: via working shafts of up to 22m deep

Taylor Woodrow’s segment operation

As a member of the Channel Tunnel's TML consortium, Taylor Woodrow was directly involved in the design and manufacture of the expanded concrete segmental lining used on the British side. These were produced by TML at a modern casting factory on the Isle of Grain. With eight steam cured carousel production lines, the most modern of computer controlled concrete batching plant equipment and precise steel casting moulds from Sacma of Italy, the Isle of Grain produced half a million segments (a maximum of 1000 segments/day).

With the tunnelling on the Channel Tunnel now complete, the Taylor Woodrow team decided to make use of its expertise in segment casting. It purchased two of the Isle of Grain production lines from TML and moved them to its London headquarters in Southall. It then clinched one of the major contracts in the 1991 UK tunnelling industry when it landed the £10m order for precast segments for the remaining 33km of the London Water Ring Main for Thames Water.

"We won the order after responding designing a new tunnel lining system which would increase previous tunnelling rates and make up for lost time on the project as a whole", said Ross MacKenzie, director of Taylor Woodrow Civil Engineering, "Thames also wanted to improve the integrity of the lining system while reducing the cost".

In an unconventional, further hands-on approach, Thames Water, the client, has bought three Lovat tunnel machines from Canada for each of the last three Ring Main contracts.

As part of this 'global' approach, Taylor Woodrow worked closely with Thames Water to perfect a cost effective, wedge block lining and smooth bore bolted lining which would meet the rigorous specifications for these pressurised water tunnels. It also worked with Lovat Tunnelling which developed an efficient handling and mechanical erector system for faster ring-building rates.

The Thames Water segments are markedly different from the industry's standard '100 inch' diameter wedge block rings which were developed in the 1950s and comprise 12 segments per 600mm­wide ring. They are wider, at 750mm, and there are fewer (five) segments per ring plus the key. They also have a specified minimum strength of 60N/mm2 (as opposed to the usual 40-50N/mm2 of the traditional 100 inch wedge block segments) and are subject to rigid quality assurance procedures.

Rings of different specifications have been designed to meet different ground requirements. At normal depth in London Clay the segments are unreinforced and 180mm thick. To allow their use at up to 70m deep without increasing thickness, reinforcing is introduced. Bolted variations of the system have been designed for use in poor ground, and for the initial lengths of each drive. These are all compatible with the expanded ring segmentation and also with the Lovat TBM's handling system.

Segments for the Ring Main are being cast at the Southall factory where two steam curing production carousels are producing segments at a rate of 500 rings/wk in a 12h/day, 5 day/wk. Each cast can be filled and struck in 4h.

When Taylor Woodrow won the LE cable tunnel contract, it proved more cost effective to design and produce its own segments than have an established manufacturer gear up to produce the non­standard segments. These segments are being cast at the Isle of Grain, where Taylor Woodrow is using one of the remaining TML production lines to enhance its Thames Water production. Here, the segments for 330 Thames Water rings are produced per week working 24h/day, 5 days/wk. The LE segments are cast in static mode as an adjunct to the main production line and steam cured, which allows each mould to be used twice in a 24h period. High-precision steel moulds from Stelmo (UK) are being used for Thames Water and LE segments.

Mixing the ingredients

For the Channel Tunnel segments, Taylor Woodrow designed the concrete mix at its own testing laboratories at Southall. The mix had to give a strength of 80N/mm2 after 28 days and produce segments to within 0.1mm tolerance.

With this full-scale loading test rig at the Southall works, Taylor Woodrow combines design, testing and manufacturing of precast concrete segments as a complete package
With this full-scale loading test rig at the Southall works, Taylor Woodrow combines design, testing and manufacturing of precast concrete segments as a complete package

"The strength and quality of the cast concrete depends largely on the quality of the ingredients, their mix and the precision of the moulds", said Sam Simons, Precast Manager for Taylor Woodrow. "The challenge is to arrive at a mix which will yield a high early strength for early strike and segment handling as well as high temperature tolerances for rapid curing as well as liquidity for easy fill of the moulds," he said. This is a tall order, since the methods of achieving each objective are in contradiction to each other.

Like the Channel Tunnel operation, Taylor Woodrow is using top quality granite aggregate from Foster Yoeman from the Glensanda quarry in Scotland and Blue Circle cement. A 30% pulverised fuel ash (pfa) content removes the need for sulphate resistant cement. "The pfa substitute avoids development of cracks in the segments from the excessive heat generated by sulphate resistant cement during hydration", said Simons.

All ingredients, including the necessary additives, are mixed in a twin bin, automatically computer-controlled batching plants which have an output capacity of 60m3/h. On striking the segment from the moulds, the concrete has a strength of 15N/mm2. This continues to increase with time to about 80N/mm2. "Some Channel Tunnel segments have tested to 100N/mm2", said Simons. The Thames Water rings also under­went full-scale loading tests on a rig at the Southall works, the largest and most sophisticated such testing rig in the UK. With 44 jacks, the rig simulates various horizontal and vertical loading combinations likely to be exerted by different types of ground. "This 'real ground' modelling with finite element analysis avoids over design while ensuring optimum quality at the most cost effective price", said Simons.

This is not the first segment casting operation for Taylor Woodrow. "We cast our own segments for, among others, the Heathrow Cargo Tunnel project in 1967; the Hartlepool and Lowestoft outfall projects in 1970; and the Piccadilly Line extension of the London Underground to Heathrow Airport Terminal 4 in 1982", explained MacKenzie. "We are not in competition with the established suppliers in the short lead time, readily available, standard precast market, with whom we have good relations. Our expertise comes into its own on high production, high value operations with a pronounced technological content such as demanding tolerances, design and testing requirements."


  • Visit our Archive and use our TunnelTalk library search facility for a wealth of information about the use of steel fibre reinforced precast concrete segments in projects around the world.

Add your comment

Thank you for taking the time to share your thoughts and comments. You share in the wider tunnelling community, so please keep your comments smart and civil. Don't attack other readers personally, and keep your language professional.
In case of an error submitting Feedback, copy and send the text to
Name :

Date :

Email :

Phone No :

   Security Image Refresh
Enter the security code :
No spaces, case-sensitive