Planning and Engineering of Heathrow Express

David Williams opened the presentation, commenting that it was very surprising that Heathrow did not have a dedicated city centre rail link. However, the signing of the Joint Venture Agreement between BAA and the British Rail Board in early August 1993 had allowed the delayed scheme to proceed.
An NATM trial had been carried out in 1992; enabling works started in the late summer of 1993; the cut-and-cover contract was let in October'93, and tunnelling works will be let early in 1994. The new rail service, financed predominantly by private investment, will be in operation by late autumn 1997.

The predicted ridership is 14 per cent of passenger and attendant traffic and staff. There will be four trains an hour with up to eight car trains and a 16 minute journey time to Paddington. The total construction cost, at 1992 prices, is £260m, split roughly 80/20 between BAA and BR.
The route distance from Paddington to the Central Terminal Area (CTA) and on to Terminal 4 (T4) is 28km, with 20km on the existing main line. BAA and BR are responsible for the design, construction and management costs of their section of the works. BR is responsible for all the works on the main line and the grade separated junction at the start of the spur. BAA will look after all works to the south, which include the cut-and-cover works, the NATM works at Shepiston Lane, 3km of twin bored tunnels of 5.7m i.d. to the CTA and three intermediate escape shaft and ventilation draft relief shafts.
The CTA will have two elliptical station tunnels, 8.2m, wide by 6.4m high, 200m long, connecting through cross passages to a central concourse tunnel of similar cross section, with an extended link directly to Terminal 1. The main 19.5m i.d. interchange shaft at the west of the station will connect into the CTA subway system for access to Terminals 2 and 3.

West of the CTA, the twin tunnels converge to a step plate junction and continue in 5.7m id bored tunnel 2.8km long towards Terminal 4. Stub tunnels will be provided west of the CTA for the future link for the planned Terminal S. Two draft relief and escape shafts will be provided on the T4 link, one of which has already been constructed during the NATM trial tunnel works. The single bored tunnel will connect in a step plate junction and lead into twin station tunnels at T4.
High standards of safety are demanded in the design, construction and operation of the Heathrow Express.
Alan Finch proceeded to discuss the detailed design after an initial review of rail and metro links to Heathrow. Only one site in the CTA had been identified as a reasonably sized working area: the fuel depot between Terminal 3 and the Terminal 1A car park.

The safeguarded location for the CTA station was adjacent to the London Underground (LUL) station. The studies showed that an east/west station crossing beneath the LUL station was feasible with the access from the north and an extension to the south to T4. This location enabled direct access to Terminals 1 and 3 and provided a future link to the planned Terminal 5 to the west of the airport.
At the CTA, good access to the terminals will be provided by escalators and lifts and draught relief and ventilation shafts at each end of the stations and emergency escape shafts. At T4, the stations connect into an 80m concourse tunnel with direct access to the terminal.
The geology under Heathrow consists of made ground for up to 2m, over 2-4m of Taplow gravels. 50m of London Clay above 20m of Woolwich and Reading Beds. There are two main aquifers. These are in the upper superficial deposits and gravels and in the chalk below the Woolwich and Reading Beds.

Cut-and-cover section

The cut-and-cover section of reinforced concrete twin cell box in a landfill site will be 535m long, 12m wide and 7m high. The depth to rail level increases from 6.5m at the north end to 19.5m at the south. The running tunnels to the north and to the south will be within the London Clay. The specified rings will be 1m wide and 220mm thick with nine segments and a key. The ring is tapered; it has an invert segment upon which the track slab will be cast. The initial planning of the station tunnels at the CTA and T4 was based upon conventional segmental linings. As part of the constructability and cost review the use of NATM was investigated. The station lining will consist of a primary shotcrete lining or ground support, designed by the contractor, a PVC waterproof membrane and a 250mm thick unreinforced concrete lining which will be painted with a calcium silicate-based paint.
Settlement caused by conventional tunnelling methods is well documented but there was no precedent for the use of NATM. Thirty-two structures were identified for which measures would be required to monitor settlement caused by tunnelling. The contract documents specified performance criteria for each building and structure. Where appropriate, compensation grouting as a settlement alleviation measure was identified. The design for safety aspects of stations was based upon: sufficient exit routes for the evacuation of all passengers from a fully occupied platform in 4 mins; evacuation from the remotest point on the platform to a place of safety within 6 mins; one excavation route being out of service. The concrete trackbed has been designed within the tunnels for a maximum speed of 130km/h.

Tony Deane then discussed the NATM trial tunnel. There was considerable concern about the settlement likely to occur from constructing the station using NATM in London Clay below airport terminals, car parks and the LUL Piccadilly Line. The cover to the tunnels is typically 18m.
Back analysis of data for previous projects, together with extensive finite element analysis, suggested that acceptable settlement could be achieved. It was decided to construct a trial tunnel at a ventilation and emergency escape shaft on the south side of the airport where a station sized tunnel could be constructed along the line of the single bored tunnel. A secondary objective of the trial was to confirm the viability of NATM in London Clay and thus to assist tendering for the main station contract. The trial tunnel, at a depth of 20m, consisted of three different construction sequences, each over a length of 30m. To limit the number of variations, the specification for the thickness of shotcrete, mesh and lattice girder spacing was maintained. The first and most conservative sequence was to excavate two side headings followed by the central core. The second sequence was the excavation of one side of the tunnel and an enlargement to complete the tunnel. The third sequence was a top heading and bench sequence with the base of the heading supported on inverted concrete arches. The excavation cycle was lm and a relatively thick 250mm of shotcrete was applied.
The trial tunnel contract was awarded to Kier/Lilley/Kunz in January '92 at a tender price of about £1.2m. The work commenced in February '92 and was completed in June. The 10.7m i.d. segmentally lined shaft was constructed using caisson techniques in the superficial deposits and by underpinning methods in the London Clay. The bottom section with two eyes was constructed using shotcrete.

Dr Ing Gerhard Sauer discussed construction of the NATM tunnel with a selection of slides covering the sequence for each section of the tunnel. The ground loss for the first sequence was 1.2 per cent. The second sequence which was not affected so much by the learning curve had a reduced ground loss of 1.1 per cent. The third sequence suffered from a weak temporary ring support and had a ground loss of 1.5 per cent. The measurements of ground loadings and shotcrete stressed confirmed the short term ground loading of 60 to 70 per cent of the full overburden. A study of the ground movements shoved that they levelled off after the completion of excavation and the closure of the ring for each sequence of construction, confirming the importance of the early closure of the ring.

Williams then described the procurement and contracting strategy. All the works will be competitively tendered strictly in accordance with the EC Utilities Directive. The original time scale for the project dictated a 'fast track' programme, which led to the construction management procurement route. All the designers and the contractors are separately and directly appointed by Heathrow Airport Ltd.

In the autumn of 1959 Taylor Woodrow Management was appointed programme manager with a likely 37 to 40 packages. For mainly political and financial reasons the design and preconstruction phase has been prolonged. The detailed design is substantially complete and the project is now ‘fast build’.
Two strategies were considered: construction management with eight packages or a single package. In January '93 the BAA Project Board agreed that the 'eight package scenario' under a construction manager should be adopted. Major construction will be completed by the end of 1996 or early 1997, with the rail link in operation from late 1997 following commissioning. test trials and running. The first main construction contract for the cut-and-cover tunnel works began in Oct '93. Tenders for the main tunnelling works were to be let in Jan/Feb '94. The main tunnelling works will be performed under the New Engineering Form of Contract (NEC).

Discussion

B New of TRL opened the discussion with comments on the design and installation of the instrumentation for the trial tunnel which was carried out by TRL. The trial was important also for the Jubilee Line Extension and CrossRail. The main differences from the predicted movements were those close to the tunnel and the horizontal movements. which were under predicted. The use of a ground movement model and iteration would reduce construction costs. D R Basset, Mott MacDonald, had designed an electro level used for monitoring of settlements. He said that the earlier closure of the Type 3 ring would have reduced the settlement. He discussed the settlement contours which showed a bias to the left side caused by the construction methods, and different side support. For the Type 3 excavation the suggested face losses were 1.5 per cent for the top open face excavation. 1.3 per cent for the rest of the crown and 0.8 per cent for the lower section of excavation. He discussed the results from the piezometers where there were strong suction pressures as the tunnel approached and very strong horizontal pore pressures at the tunnel walls.

J Moss, Mott MacDonald, queried why a waterproofing membrane was provided in the NATM tunnels. The speaker said that as concreting in the crown was difficult, an unreinforced waterproofed lining was chosen. J Thompson said that a membrane between the shotcrete and the cast in situ concrete permitted the shrinkage of the concrete to be controlled, otherwise the lining would have had to be reinforced.
W Ward said that he was involved in experimental tunnels in mudstone in the early 1970s and that one of the measurements taken was on the dust levels. He asked what was the attitude to dry shotcreting for the trial tunnel and what precautions were taken. For the trial tunnel, the HSE took particular care and was very helpful. The HSE was satisfied with the measures for the short length of tunnel which could be cleared by ventilation. Ward did not understand why a pilot tunnel had not been used to reduce the face loss as on the DLR to Bank tunnel in London. A central pilot had been reviewed for the larger tunnels, particularly as this had been used successfully at Frankfurt with a 20-25 per cent reduction in settlement, but it did increase the project cost.
E Snowden, Kier, was responsible for the construction of the trial tunnel and supported the brave step forward to provide the tunnel and its use for the acceptance of the method.
C N P Mackenzie, AMEC and BTS chairman, was surprised that the face loss figure for an expanded lining was higher than for NATM. J R King, Pemberwell Services, said that one reason was the distance from the face to the ring support with a shield tunnel. He was, however, concerned about safety with large faces in London Clay.
The speakers said that safety during the construction was down to the contractor. The trial tunnel showed good London Clay but one face was battered back. There would have to be more engineering control and supervision. Mackenzie said that, although London Clay could be excellent, it could also be blocky and extremely dangerous when blocks fell out. These movements could not be predicted.
W Anderson, Sandberg, asked about the New Engineering Contract. BAA has had experience with NEC with a building contract. It has a working party which is continuously reviewing contracts and it was considered the form of contract most suitable for this scheme.
TW Mellors asked about dosage rates of additives and long term effects on strength and durability. Dr Sauer said that they had moved from powder to liquid accelerators with 4-5 per cent to weight of cement. Tests on long term strengths of shotcrete had been carried out.
J Buchanan asked about the acceptance of method statements. For the project the methods had to comply with the overall specification and had to be technically sound. The lowest tender would not always be accepted. The contractor had to provide a safety plan with its tender.