It’s fair to presume that Thursday 9 June 1892 started much like any other for Archibald McLay, a ‘servant’ of the Midland Railway with 24 years driving experience. Well rested, his 10½-hour turn began in Bradford and thereafter followed an uneventful course until he departed Leeds with the 3:07pm service to Ilkley, his fourth trip of the day. Alongside him on the footplate was Walter Bolton, his regular fireman.
Approaching Apperley Junction, the Home signal cleared, allowing him to cross onto the Ilkley branch. As he passed the box, the signalman took off the Starting signal and displayed a green flag indicating that the section ahead was clear but Esholt Junction, two miles and 18 chains away, was blocked. McLay acknowledged by raising his hand.
Tackling the stiff 1:60 rising gradient towards Esholt demanded some effort, speed being only slightly slackened as the next Distant signal came into view. As expected, it was at Caution. The train then entered Springs Tunnel; half way through it, McLay caught first sight of the junction’s two Home signals.
The one controlling the converging line from Bradford was off for a train that had just passed through the junction; his own signal – still at Danger – then disappeared behind a bush. When he looked up again, all he saw was a Clear signal and proceeded accordingly.
By the time McLay had realised his error and applied the brake, a Bradford-bound service was crossing the line ahead of him. Still travelling at 15mph, his engine struck the third coach from the rear which was pushed over and disintegrated. Bolton was jettisoned from the locomotive as it too turned onto its side. Four passengers lost their lives, 26 were injured. McLay told Major General Hutchinson, chairman of the inquiry, that “the only reason I can assign for my mistake was my being dazzled by the bright light after leaving the tunnel.”
Springs Tunnel is a modest one, driven through shale and extending on a slight curve for just 77 yards. It would have been a cutting had it not proved necessary to appease William Stanfield, a local landowner. Partly covered way, it was constructed in the early 1860s for the Otley & Ilkley Extension Line, a joint enterprise by the Midland and
North Eastern railways. John Crossley acted as engineer for both companies, whilst Rennie & Co successfully tendered to build it.
Work on the tunnel was noteworthy only for the death of 30-year-old James Rolls who was grafting in a wagon when, for reasons unknown, one end of it was violently forced upwards, causing his head to smash against the brickwork above. He survived for less than an hour.
Fast forward 150 years and we find ourselves in Control Period 5. Under its civils framework contract, a minor works team from AMCO was dispatched to carry out routine stitching close to the north portal, responding to an examination report which identified hollow and spalled brickwork, ring separation and open joints. The work didn’t happen as, once on site, it was felt that such an intervention might make matters worse. Instead, Network Rail asked AMCO to develop plans for a partial reconstruction encompassing the northernmost 12 metres of arch. A nine-day blockade would be arranged in which to deliver it.
Gently does it
Site investigations and historical drawings suggested the arch was formed of at least seven rings of brickwork, sufficient for COWI – the project designer – to propose the replacement of the inner three rings with 250mm of sprayed concrete reinforced with Ram-Arch, a system of interlocking steel mesh panels. This brought with it the welcome side effect of a 50mm improvement in structure gauge.
It was recognised at an early stage that using conventional peckers to break out the affected brickwork would almost certainly prompt a collapse, being too brutal and intrusive. Milling offered better prospects, with much greater control over the rate and depth of cut. Network Rail’s project team bought into this approach. However, the methodology is largely untested on the railway, not least because of the lack of appropriate machinery.
Done to a turn
To overcome this issue, AMCO went to its longstanding ally Chris Scott at Foulstone Forge. who fabricates solutions to problems in the ‘too difficult’ pile, using steel and ingenuity. Rail Engineer readers have met him before – in the Severn Tunnel (issue 146, December 2016), where his drilling rigs punched 7,000 holes into the high haunch to mount electrification equipment, and on Derbyshire’s Ecclesbourne Valley Railway (issue 149, March 2017) during a demonstration of a ‘tunnel screen’, developed in partnership with AMCO to enable a train service to be maintained over one line whilst work takes place safely on the other. Yes, you can have your cake and eat it.
On this occasion, Chris had to manage expectations of what could realistically be achieved, in part because timescales were tight. He used an existing machine as a base – developed for an upcoming debris removal scheme – and mounted on it a newly-designed milling head, giving a 300mm wide cut, 125mm deep.
The system was put through its paces in Kirton Tunnel and a disused bore at Alfreton, the latter presenting the challenge of extremely durable bricks. Each trial resulted in refinements to improve performance. By the time of the blockade, all parties had confidence that the machine would do what was required of it.
Weighing 7.5 tonnes, it sits on a T8 trailer and can turn through 360º, with 250º of rotation around the horizontal axis. Powering the milling head are two Hydroma bent-axis hydraulic piston motors, collectively generating 150HP. These turn 40 diamond-tipped blades at 1,700 revolutions per minute, set in a cassette which also comprises a shaft and two bearings. When eventually worn out or damaged, the cassette can be changed in half-an-hour.
Provided for dust extraction is a fan pulling 4.2cubic metres of air over the blades every second. This has
a significant cooling effect, bringing the added benefit of increasing the blades’ longevity. As each cassette costs around £25,000, this is obviously good news commercially.
AMCO arrived on site in September – six weeks out – setting up a compound in a field adjacent to the railway, half-a-mile south of the tunnel. The immediate priority was to create a ramp up the east side of the embankment to gain vehicular access onto the track, with a scaffold staircase installed alongside it. Thereafter, manpower could be shuttled to and fro using a gator from Total Rail Solutions.
A series of midweek and Saturday night possessions were booked in which to progress general remedial works, improving the condition of the tunnel lining. This entailed the injection of resin-based products from Minova, specifically Geoflex to strengthen the brickwork, Wilkit into the sidewalls and CarboStop E filling the voids above the arch.
The blockade got underway at 00:01 on Saturday 21 October. Planned to coincide with half-term week, disruption was further limited by the ability to run Leeds-Ilkley trains via Shipley – using the single line from Esholt Junction to Dockfield Junction – although the suspension of Bradford-Ilkley services meant some passengers faced a change of trains.
The first requirement was to remove the overhead line equipment, a task subcontracted to SPL Rail Resourcing and lasting six hours. Once complete, a mobilisation of manpower and plant into the tunnel was followed by the laying of boards to protect the ballast and trackwork. Then the fun began.
It was early afternoon by the time the machine had been prepared for milling, initially positioned on the Up line from where it could reach across to the Down side. The intention was to take out metre-wide channels around the arch – starting from the low haunch – into which one setting of Ram-Arch would be inserted; then the cycle would repeat. However…
“I’ve never known brickwork as hard,” recalls AMCO’s contracts manager Dave Thomas. “The diamond disks were glowing red and there were sparks coming off. It was like that for eight hours.” Periodically, the blades were checked for damage. There was nothing; not much heat either. On they went. The experience clearly tested the machine operators as they refined their technique in real-time, an inescapable reality whenever you introduce something new. But it worked, cleanly and effectively, until…
Later that evening, a problem was reported. The four outermost blades in the cassette had sustained damage. The head was removed and returned to Foulstone Forge for investigation. The conclusion: not all the blades had been sitting in the cut due to an operational maladjustment. Lesson learned, you move on. But then…
Overnight it was decided to open a window in the brickwork to confirm its condition. What was found changed the complexion of the project. Although the inner ring was good, the next two were poor, with cracked bricks and soft mortar; the fourth had disintegrated. And that was it: four rings – not the seven or more that were expected. Whatever had been seen during the investigations was an aberration, much like the one which misled Archibald McLay when he passed through the tunnel. If three rings of brickwork had been taken out as planned, there would have been practically nothing left.
On the Sunday morning, engineers from Network Rail and AMCO assembled in the tunnel to consider the problem. They found every brick in the window was loose enough to be moved by hand. It was clear therefore that milling was no longer a viable option.
Working with COWI, a revised plan was formulated to break out longitudinal channels of brickwork, immediately above the stone sidewalls, 1.5 metres in height. On the Down side, just the inner ring was removed but, on the Up, two rings had to go due to gauging constraints. These channels created sufficient space from which to spring the Ram-Arch, with the panels pinned back using 400mm masonry pins in resin. 250mm of sprayed concrete would still be applied.
COWI staff worked through the night to complete the calculations and produce drawings, confirming there would be no interference with gauge in the critical zone below 4.2 metres. In the meantime, a start was made on the physical works.
Spray that again
Whilst not so aesthetically pleasing, the new design fulfilled the remit and involved less work. This meant that, by Wednesday, the programme was a day ahead despite a day-and-a-half having been lost.
Thursday lunchtime brought the first consignment of concrete from Cemex – a C40/50 mix including microsilica, a superplasticiser and stabiliser. Waiting at the top of the ramp was a remixer – sitting on another T8 trailer – into which the material was pumped. It was then propelled up the line to the tunnel for discharge into the main delivery pump. Gunform, the sprayed concrete contractor, applied two layers 125mm in thickness, a BASF SA160 accelerator being added to speed up the curing process. The work was finished later that evening.
The measure of a contractor is not how it performs when all is running smoothly; more critical is its response to fate inserting a spanner, as it does from time to time. Springs Tunnel’s brickwork revelation was beyond AMCO’s control, yet the job was completed safely and successfully – albeit involving a hasty and pragmatic redesign – and commuters found their train service restored on Monday morning. Those are three key variables: tick – tick – tick.
Despite this not being the encouraging demonstration of milling that AMCO had hoped for, the firm remains convinced that it’s a method with great potential, increasing efficiency and producing cleaner repairs. It might even prove useful for calcite removal. Combine it with the aforementioned tunnel screen and milling could offer a means of programming extensive tunnel repairs without the need for a blockade.
On a strained network with increasingly high passenger expectations, that’s a prospect we’d be foolish to ignore.
This article was written by Graeme Bickerdike. All photos are courtesy of Four by Three.
Read more: Moscow’s new ring