Unique is an overused word and it’s often applied erroneously. Whilst all rail projects are different – due to the location and scheme design – there is always some commonality between, for example, every bridge replacement or track renewal. But the same cannot be said of the events at Farnworth last year where a 295-yard tunnel was filled with foam concrete and a new one driven. This was genuinely unique: no reference could be made to a dusty blueprint hiding in the archives.
Rail Engineer looked forward to this £20.8 million project in May 2015 (issue 127), charting the tunnel’s history, as well as describing the theoretical methodology and the open-faced shield which would protect the miners. Yes, such people do still exist. This time we’ll look retrospectively at how things actually went.
If you don’t have last May’s issue to hand, here’s some context. Farnworth’s 1838 two- track tunnel, engineered by John Hawkshaw, formed part of the line connecting Manchester with Bolton. The introduction of out-of-gauge Pullman carriages on services to Carlisle brought the need for a second bore to host Down trains, opening in 1880. The first was then repaired, significantly reducing its structure gauge; thereafter only the Manchester-bound Up line passed through it.
Now fast forward to the twenty-first century and enabling works for the North West Electrification Programme. Neither tunnel could accommodate two tracks and overhead line equipment so the decision was made to bore a larger one on a similar alignment to the original.
Network Rail appointed Buckingham Group as principal contractor in November 2014, including works to reconstruct Farnworth’s adjacent station, an overbridge and retaining wall. J Murphy & Sons Ltd – we’ll call them Murphy hereafter – undertook the tunnelling to a design by OTB Engineering.
Site mobilisation got underway at the end of February 2015, with a generous compound being established on leased land over the eastern half of the tunnel. This was separated from a smaller compound on the west side by the busy A666 linking Bolton to the nearby motorway network. Construction traffic could enter the site via a slip road built off the southbound carriageway. A few yards away was a full-width ventilation shaft reaching the midpoint of the Up tunnel.
An immediate priority was to strengthen the single-track Down bore through which a reduced-frequency train service would operate whilst the new tunnel was being mined. Over 11 weekends, 54-hour possessions were secured for completion of this work, requiring a 200mm concrete lining to be sprayed over two layers of welded mesh fabric reinforcement.
There was insufficient clearance to adopt this solution through an area of distortion so here the brickwork was removed and steel ribs fitted. Murphy Surveys installed an optical deformation monitoring system which would provide real- time alerts as progress was made with the new tunnel. At their closest point, the intrados of the two bores are just 1.5m apart.
Bi-directional working was introduced through the Down tunnel on 2 May, allowing the core works to proceed. An early focus for Murphy was to prepare for the tunnelling shield’s arrival by excavating a launch pit in front of the east portal – 3.6 metres deep, 10.5 metres wide and separated from the adjacent open line by a secant-piled wall. As the pit was being dug out, a 15mm movement of the piles was recorded, triggering a monitoring alarm and hasty repacking of the track late one Sunday night. Also assembled was a steel frame to act as a thrust block for the shield. Two new portals were constructed as part of the concrete works, together with a slab above the eastern entrance where the lack of cover could have resulted in the ground and existing parapet being uplifted. Working from the Up tunnel, Keller Group consolidated the ground and filled voids between the two bores by injecting a high-strength grout through 1,150 pipes at two metre centres. With shutters erected at both ends, the tunnel and 13 cross passages to the Down bore were then filled with 7,800m3 of foam concrete in 1.5 metre layers, delivering a dry density of 1,050kg/m3 with 1.5N/mm2 compressive strength. This was sufficient to prevent a collapse during tunnelling but proved soft enough for easy excavation. The material was produced using on-site batching facilities and then pumped down the shaft before it too was filled. The process took 17 days, led by GS Foam Concrete.
Late in July, the shield arrived from Oldham where it had been manufactured by Tunnel Engineering Services. It came in six main pieces – the heaviest weighing 64 tonnes – with 12 lorry loads of ancillary equipment such as trailing gantries and the rotary segment erector for the precast concrete lining. A 500-tonne crane, supplied by Ainscough, assisted in its two-week assembly. Testing and commissioning took place prior to launch on 1 August; the first lining ring was completed on the 10th after overcoming the very stiff resistance offered by Hawkshaw’s 177-year-old stone portal.
Extensive ground investigations by Aspin Group had been ongoing for two years as a dozen possible options for accommodating an electrified railway through Farnworth were whittled down to a final design. Two dozen bore holes were sunk – extending to 10 metres below invert level – as well as eight window samples and more than 1,000 cores through the lining either side of the tunnel. All this painted a picture of the conditions that would be encountered as the mining work moved forward, and a long-section representation was pinned to the office wall where Rail Engineer spoke with Network Rail’s Rhiannon Price and Murphy’s project manager Mick Boyle.
It showed the original tunnel to have been driven mostly through glacial till, with laminated clay above for two-thirds of its length from the west end. A narrow layer of sand extended inwards from the east portal for a distance of 50 metres, becoming deeper after 35 metres. It’s worth making the point that the maximum ground cover above the tunnel was only seven metres at the shaft, which raises the question as to why Hawkshaw did not specify an open cutting here. No-one seems clear as to the answer.
The team then embarked on a venture into the unknown: they had expectations as to what they would face, but no certainty. As the machinery was bespoke and effectively prototype, a number of early breakdowns occurred, keeping the fitters busy swapping pumps and hydraulic rams. Consequently, the anticipated daily advance of three rings – or 4.2 metres – was reduced to just one.
On 14 August, as the shield moved beyond the concrete slab above the portal, came the first intervention by running sands as loose, wet material poured into the excavation. Looking up, daylight could be seen. Around 35 tonnes of Carbofill resin was needed to plug the largest of several substantial voids. Orica UK dealt with these, thereafter remaining on site
until the end. Two weeks later, ring No.31 had been reached when a second fall inundated the working face with 100 tonnes of sand which had to be removed by hand and fed back through the machinery, the front of the shield having been buried.
With settlement at the surface, it was decided to take the load off the shield by excavating a 15 metre long pit, supported by sheet piles and braced with Megashor propping frames. Having removed the spoil, this allowed a section of tunnel to be constructed without any further threat of collapse, first grouting the haunches and then strapping the rings together as they were installed by the segment erector.
During this period, the Down tunnel monitoring issued a red alert, a movement of 14mm having been recorded – a 4mm exceedance of the trigger threshold. Trains were stopped for an hour whilst engineers examined the lining, although no cracking or water ingress was apparent. Signs of distress did however develop in the Megashor frames, underlining the need to backfill the pit as quickly as practical after the rings had been grouted and a concrete slab poured above them.
Curiously, this extra work brought some relief from the tension felt by the project team. The original reopening date of 5 October was now clearly unachievable, and everyone could see that. So, having paused, they took the opportunity to reassess and adopt a more cautious approach.
As the GI had indicated, ground conditions were much more benign by the time the ventilation shaft was reached. To quote Rhiannon, “we’d started to fly”. But with the next critical landmark approaching – the drive under the A666 – four days out were taken to service the machinery, recognising that any failure at this stage would have a significant reputational impact, not to mention the disruption it might cause for many thousands of motorists.
Whilst a typical surface settlement of 100mm was predicted before work commenced, the actual figure proved closer to 300mm. In light of this, contingent mitigation measures were agreed with Bolton Council when excavating within the road’s zone of influence, including increased monitoring and closure of the lane immediately above the working face. The lane was changed each night as progress was made from east to west. A standby highways gang was called on to carry out resurfacing works on four occasions, the greatest settlement being a very noticeable 191mm in the central reservation. The reality though is that the section under the A666 advanced unremarkably, apart from the frustrating need to slow the work rate to align with the lane closure regime.
Breakthrough came at 12:31 on 25 October – less than 12 weeks after re-boring had got off to an uncertain start. A hydraulic hammer forced a hole in the foam concrete; then there was light. It was a proper team effort – around 120 people had played their part in this ambitious adventure. On cue, a train sounded its horn to acknowledge their achievement.
This was, however, a milestone, not journey’s end. Two possible hand-back dates emerged (based on the availability of specialist resources and considering passenger impacts): mid- December or February, but the latter was deemed unacceptable given the overrun. It was therefore decided that a full timetable would be restored on 14 December, with trains using the new tunnel under a speed restriction. The project team had seven weeks to reinstate the railway.
Having arrived in the reception pit, the job of dismantling the shield and its machinery got underway immediately, a crane lifting the pieces into the small west compound where they were cut up and dispatched on low-loaders. Following discussions with Bolton Council, this element of the work was restricted to the day shift to minimise the impact of noise and lighting on local residents.
The original design called for a concrete invert to be poured through the tunnel, on which ballasted track would be laid. This approach was reviewed in light of the time constraints, stone fill being used instead.
Unfortunately, persistent rain and the quantity of fines within the supplied material resulted in sludge forming, so it all had to be removed and the process restarted using ballast. As a consequence, the programme fell behind by four days.
Despite the project’s pressure gauge edging upwards, history will show that the first train did pass through the tunnel, as planned, early on Monday 14 December. The last two weekends of January saw removal of the temporary crossovers used for bi-directional working, allowing the speed restrictions to be lifted. Now redundant, a concrete invert will be installed through the Down tunnel; it will then serve as an access route.
Successful delivery of any project relies on a collection of like-minds pulling towards their shared goal. Beyond that truism, what made Farnworth exceptional was the nature of the challenges that conspired to test Network Rail, its contractors and the train operating companies. This closely-knit team had ups and downs to contend with – such is life – but everyone has learned lessons to take away with them. In that respect, they’ll reflect on Farnworth as a uniquely positive experience for many years to come.
Photos courtesy of Four by Three