Channel Tunnel: The 20th century’s most expensive construction project — but worth every penny
No list of railway wonders would be complete without including the great marvel of modern civil engineering that is the Channel Tunnel… GARETH DENNIS looks at the story of its inception, construction and operation.
A version of this article also appeared in Issue 878 (8th May 2019) of RAIL magazine as part of their Seven Railway Wonders series.
It is one of the few train journeys left in the UK that can inspire the giddy joy of the early days of steam, and yet passenger travel is only a very small part of its purpose.
Without it, the UK couldn’t function. It is as simple as that. Countless vital products too time-critical to ship and too numerous to fly arrive onto our little island every day via this portal to the continent.
Of course, I’m talking about the Channel Tunnel — Britain’s railway conduit to Europe.
Back in 1994, the American Society of Civil Engineers named it as one of the seven modern Wonders of the World and, in spite of Silicon Valley trying to reinvent tunnelling and transport more generally, this miracle of modern engineering remains one of the great testaments to tunnel engineering and the railway’s ability to connect people.
However, during its construction the Channel Tunnel acutely exposed the challenging link between project funding and project delivery as well as the UK’s antipathy with civil engineering megaprojects.
Stepping away from the engineering, the UK’s attempts to extricate itself from the European Union remind us of our confused political relationship with mainland Europe and, at several points in recent weeks, the threat of a “no deal” departure risked ceasing Eurotunnel operations altogether.
[Edit: How time flies… We’re in mid-2020 and “no deal” is now an inevitability, with all the horror that entails (including risking the entire Eurotunnel operation).]
2019 brings the Channel Tunnel to its twenty-fifth birthday, and rather than trying to cut through all of this noise, I’m going to try and tell its story through each of these lenses, considering how the project was influenced by the very particular circumstances during its inception, and how, in turn, it influences us today.
The Channel crossing is still sensitive to disruption, with most people in Kent now familiar with Operation Stack (parking lorries along the M20) but, before the Chunnel was built, things were far worse. Inclement weather prevented ferries from making the crossing, and this had major impacts on the supply of goods across the UK.
Even when the ferry was running to time, the journey from London to Paris still took most of a working day. The 1980s had made business and industry an increasingly global affair, and such an unreliable connection to the continent was becoming a problem for the UK economy.
The French had been keen to construct a physical link to Britain for decades and, even after the UK government had cancelled the 1970s version of the project, still were. The trouble was, despite wishing to revive the scheme, Thatcher wasn’t willing to stump up a penny of public cash. However, she had just spent a lot of political effort untethering the banks from strict regulation and — ignoring the later consequences of doing so — this provided her with the perfect tool to pay for the new megaproject: private cash.
At that point, the idea of privately financing a major project using guarantees secured on the basis of an asset that didn’t exist yet was a new and novel idea that the City of London was still chewing over. It was also a significant opportunity.
The newly-empowered finance industry was ready to flex its muscles, and flex them it did. Equity, institutional fundraising, share offers, syndicated bank loans, letters of credit… In all, £2.6 billion (in 1985 prices) was raised for the project. Tunnelling was back on the menu.
Given Britain’s current fixation with the letters “E” and “U”, it is unsurprising that the idea of a permanent link with Europe didn’t please everybody though, and the project garnered a significant protest movement from the public by the mid-1980s.
In an episode that should embarrass all of us, an angry horde harassed Francois Mitterrand as he witnessed the signing of the Treaty of Canterbury in 1986. Surrounded by placards emblazoned unironically with messages like “KEEP BRITAIN AN ISLAND” and “BATTLE FOR BRITAIN” and shouts of “Froggies out!”, France’s then-president’s car was pelted with eggs. There were similar scenes of anger in London, too. Given how little mobilised opposition there has been to High Speed 2 by comparison, you might think that Brits are generally in favour of Britain’s latest megaproject.
In any case, the noise of the public protests was matched by the anger within Parliament about the project, and over the closer ties with Europe the new tunnel risked forging. Despite Thatcher’s enthusiasm for the scheme, support from her ministers was far from unanimous, and Eurosceptic MPs from the governing party continued to brief against the project until well after construction had started. Labour was also split over whether to support it.
This isn’t hugely surprising. Administrations of both stripes had created formal plans to use a nuclear detonation to demolish the 1960/70s iteration of the Channel Tunnel (yes, really), so the project had form in bringing out the worst in ostensibly rational public servants.
Eventually, MPs were brought around to supporting the project through promises about direct regional connections to Europe that have yet to materialise (though HS2 will help).
In signing the Treaty of Canterbury, France and the UK had agreed once and for all that they were going to build a fixed connection between mainland Europe and its biggest island. Now someone had the unenviable task of actually building it.
Let’s skip forward to the end of 2018. Elon Musk’s Boring Company has just driven a rather rudimentary single-bore tunnel underneath Hawthorne in California, claiming to have heralded a new era in transportation as they ferried press back and forth in a lone Tesla Model S at a teeth-clenching 40mph (no sarcasm intended — the bumpy ride gave one reporter a headache).
Putting the ridiculousness of their overall proposals to one side, one key claim the Boring Company makes is that they can reduce the costs of tunnelling by “a factor of more than ten”. There is probably no better way to describe the challenges they will face than by looking at the gargantuan human effort involved in driving the three bored tunnels under la Manche.
Whilst there are several examples of significant tunnels built in the times of antiquity, British engineers were the first to develop modern tunnelling techniques during the construction of the canal network in the late 1700s. Fritchley Tunnel in Derbyshire (engineered by canal specialist Benjamin Outram) was the first railway tunnel in the world when it was opened in 1793, and Marc Brunel’s Thames Tunnel was the first under a navigable river when it eventually opened in 1843.
By 1880, tunnel engineers were successfully utilising three major aids: the tunnelling shield to strengthen and support the excavation face; pre-constructed segments to line the tunnel wall (initially using cast iron); and the use of compressed air to prevent the ingress of water in saturated ground.
However, a new tool had been patented by Royal Engineer Frederick Beaumont that would revolutionise tunnel construction — the tunnel boring machine — and the first viable attempt at permanently connecting Britain and Europe would be where it was trialled.
Edward Watkin, the visionary railway strategist (and a hero of mine), adapted Beaumont’s design and by 1882 had used two of these mechanical marvels to push a tunnel nearly 3km out under the English Channel, at which point the British establishment chickened-out and told Watkin to stop work in the interests of “national security”. Apparently, Queen Victoria wasn’t a fan of the idea.
Roll forwards one hundred years, and the structures from the 1880s attempt were abruptly rediscovered when someone dropped a cement mixer through the roof of one the original drainage adits.
The escapades of construction plant notwithstanding, the contracts for the project were awarded in 1986. TransManche Link (TML) had been formed from the British (Translink Contractors) and French (Transmanche Construction) construction consortia, and were tasked with turning this wildly ambitious idea into a reality in only seven years.
Before the physical works started though, a spectacular volume of detailed survey and design needed to be completed, and this threw up unique challenges.
Surveying long, thin shapes like the route of a railway is troublesome thanks to the curvature of the earth. Accounting for this is hard enough within one national boundary but, in the case of the Channel Tunnel, survey data had to be normalised based on the types of mapping projection used in both the UK and France.
On this side of the Channel, we (mostly) use Mercator projection, whereas France’s mapping uses the Lambert system. Even our sea levels were different, with the UK’s Ordnance Datum in Newlyn giving a value 300mm different to sea level in France. When you are designing and building a tunnel under the sea, agreeing where sea level sits is very important!
Though these special survey grids are commonplace in the UK today (we call them Snakegrids, and every major railway corridor has one), this was exciting stuff for geomatics nerds at the time, particularly in combination with early use of commercial GPS to enhance accuracy, and the new grid — glamorously named “RTM87” — was used to coordinate all survey and design information. Once tunnelling started in earnest, this data had to be good enough to guide the tunnellers in lieu of intermediate shafts as would normally be used when tunnelling under solid ground.
Assessments of the geotechnical (subsurface) conditions along the route of the tunnel had started in 1875 in readiness for Watkin’s original attempt, but the first serious investigative work for today’s tunnel started as far back as 1956, when surveyors started collecting data about the marine environment and geology under the Channel. Once the contracts were let to TML, hundreds of boreholes were driven into the seafloor, and hundreds of kilometres of geophysical traverse lines (using technology like seismography) were made, all helping engineers to build up a three-dimensional picture of the terrain.
It then fell to the designers to plot a course through this 3d maze of geology, keeping the tunnels largely within the confines of a layer of mudstone known as the Chalk Marl (though several significant fault lines meant diversions through several other rock layers on the French side of the alignment).
On the 4th January 1988, tunnelling started on both sides of the Channel, at Folkestone and Sangatte.
11 tunnel boring machines (TBMs) were used in total: six reasonably conventional TBMs were launched from the UK side, and five more complex TBMs (to deal with the more challenging tunnelling conditions that were expected) were launched from the French side.
The teams constructing the tunnels only had a measly tolerance of 150mm to play with. This consisted of 55mm for steering the TBM, 55mm for the manufacturing and placement of the lining segments and 40mm of topography allowance. In other words, the tunnellers were guiding a 1000 tonne TBM towards a moving target not much bigger than the palm of your hand.
The service tunnel pushed ahead of the main rail tunnels, allowing their alignment to be refined during construction. Continuous ground investigation helped inform the modifications to the design as the six tunnel-ends progressed towards each other.
Safety not the first priority
One major challenge of the project being privately funded was that there was an immense pressure on the scheme to be delivered as quickly as possible to enable a return on the initial investment. This had a direct influence on the health and wellbeing of staff working on it, and the safety record of the tunnel was abysmal during the initial months of tunnelling, exacerbated by productivity-based remuneration.
Eleven workers were killed during the construction of the Channel Tunnel, most in its early months, most on the UK side of the dig, and all avoidably.
Perhaps unexpectedly, the workers found an ally in the Conservative MP Michael Howard, who, as Employment Secretary, intervened following a spate of worker deaths, mandating that all staff should complete health and safety training and that far more rigid safe systems of working should be followed.
Thankfully, this made a significant difference, and the project’s safety record greatly improved after his intervention.
Unfortunately, the UK contractors hadn’t prepared for encountering significant pockets of trapped water within the Chalk Marl, and only a significant injection of cementitious grout into the area surrounding the tunnel face allowed tunnelling to progress. Initially, there were fears that the huge ingress of water was coming from the Channel above, but quick thinking by engineering geologist (and former British Rail alumni) Helen Nattrass — a lick of a finger to indicate fresh rather than salty water — avoided someone slamming a big red button, abandonment of the tunnel and a hefty delay.
Even with these early setbacks, the rate of tunnelling was staggering. Rates of over 250m a week were being regularly reached. For comparison, Crossrail tunnelling averaged a rate of 38m per week. This meant a huge volume of material being shipped out of the tunnels, and lining materials being shipped in.
On the UK side alone, 442755 reinforced concrete tunnel lining segments were produced, transported, and slotted into place behind the TBM. Each of these was meticulously designed with a life of 120 years, and had to be both watertight and durable. A further 38000 tonnes of spheroidal graphite cast iron linings were used in the connecting tunnels and passages.
As soon as the precast tunnel lining segments were laid, they were quickly followed by water pipes, ventilation ducts, power and communication cables, a double-track railway, walkways, lighting, emergency equipment… And all of that was just the temporary stuff allowing onwards tunnelling.
This rate of construction also required an army of workers (as many as 15000 at its peak), enough to justify a temporary village being constructed at Farthingloe. The stepped ground that the prefab cabins were constructed upon is still visible today.
On the 1st December 1990, the UK and France were reunited for the first time in 8000 years when Philippe Cozette and Graham Fagg pushed through a wee hole and shook hands following the breakthrough of the service tunnels. And what an iconic image… It was civil engineering doing what it does best — connecting people, rendering geography as inconsequential, borders as unimportant, and nations as neighbours.
Watching the news footage today still brings a lump to my throat; I can’t imagine what it’s like for the people who worked on the project.
I’ll briefly point out that everyone on this side of the Channel seems to make a great fuss about the fact that “the English side tunnelled the greater distance”. This is hardly something to gloat about. The French side had to deal with a significantly more challenging tunnelling environment, with saturated rock strata and several fault lines to negotiate, and it was planned well in advance that they would require longer to tunnel a reduced distance. Both sides can be equally proud of the hugely impressive pace of excavation.
On the 28th June 1991, less than three and a half years after digging first commenced, tunnelling was completed as the final breakthrough of a French TBM occurred.
The unusual benefits of tunnelling megaprojects
An oft-forgotten element of civil engineering megaprojects is that they can give us a chance to improve the landscape for the better, undoing some of the shorter-term construction impacts and other past planning mistakes.
In the case of major tunnelling projects, a significant benefit is that it is usefully convenient to do some rather creative stuff with the rock and soil exhumed by them.
In the mid-1960s, Notre Dame Island was created from the material dug out from underneath Montreal during the construction of its metro system, and is now home to the various major sporting events and acres of parkland. More recently, 2400 loads of excavated clay were shipped out from the centre of London during Crossrail’s tunnelling phases to expand the Wallasea Wetlands in Essex.
The Channel Tunnel was no different. Samphire Hoe Country Park was created using 4 million cubic metres of mudstone that was hauled out from under the English Channel, and is now home to various rare plant and bird species. Wandering through the lovely, wildflower-strewn greenery and looking out to sea, you’d be forgiven for not realising that you were essentially walking across a waste repository.
Calling the final product the Channel “Tunnel” was rather underselling it… Firstly, the Chunnel isn’t really one tunnel at all: it’s three, with the two 7.6m-diameter rail tunnels 30m apart straddling a 4.8m-diameter service tunnel, each 50.5km long. 38km of those are under the English Channel; this gives the Chunnel the longest section of undersea tunnel in the world (it isn’t the world’s longest undersea tunnel as is often quoted — this title belongs to the Seikan Tunnel in Japan).
These aren’t the only holes that the Channel Tunnel includes, either. There are many kilometres of interconnecting ducts, shafts and passages throughout the length of the link, not to mention additional service tunnels and connections at each portal.
For example, there are nearly 200 2m-diameter piston relief ducts connecting the two rail tunnels at 250m intervals, regulating air flow and pressure as trains pass. There are also 270 3.3m-diameter cross-passages connecting the three operational tunnels at 375m intervals, providing maintenance access and additional ventilation, not to mention an escape route in case of emergencies.
Perhaps most impressive are the two colossal crossover caverns, upwards of 150m long, 10m high and 18m wide, enclosing the volume of eleven Olympic swimming pools and large enough to house 240 London buses (to use the standard comparisons). These enable trains to move between the single bore tunnels, and at the time of their construction they were the largest undersea excavations ever undertaken.
Though fitting out the tunnels was complex, and delays at this stage would push back the opening by more than a year, it received its royal inauguration on 6th May 1994. Unlike her great-great-grandmother, Queen Elizabeth II appeared quite pleased about it all, describing the Channel Tunnel as “a monument to the joint efforts and talents of our engineers, technicians and construction workers, who faced and overcame many difficult and unexpected problems.”
In the end, costs had overrun by 80%, giving a price tag of nearly £14bn in today’s money. Per kilometre, that’s twice the cost of Crossrail and will be the best part of four times the cost of HS2.
Not that many of the 50000 average daily passengers pay any attention to the cost of its construction these days, perhaps as a result of what the Chunnel achieves today…
For starters, Eurostar currently provides an hourly service in each direction, and seeing as High Speed 1 is only operating at 50% of its potential capacity, this could double to nearly 1800 seats per hour in each direction if growth justified it.
The greatest passenger capacity, though, is provided by the Eurotunnel Shuttle: with each passenger shuttle being able to carry 12 coaches and 120 cars, and operating at a peak frequency of five trains per hour, up to 5400 passengers can be transported through the tunnel in each direction. That’s impressive, considering that each train carries a ferry-load of vehicles in a quarter of the time it takes for them to negotiate the sea crossing. This fleet of trains is due for renewal in 2025, too.
However, despite 21 million of them travelling through the Chunnel a year, passengers are only a small part of the story.
In one year, through trains carry a whopping 1.2 million tonnes of freight to destinations across the UK and Europe, and a whopping 23.1 million tonnes of freight use the Eurotunnel Shuttle. At peak frequency, eight freight shuttle trains per hour head through the tunnel, carrying a total of 256 HGVs (or 11260 tonnes). That’s an immense carrying capacity.
In all, an average of 400 trains use the tunnel daily, but passengers and freight aren’t the only commodity that the tunnel will transport by 2020. The owners of the tunnel are in the process of adding a €580m, 1000MW electrical interconnector through the Chunnel, increasing the exchange capacity between the UK and mainland Europe by 50%.
It’s not all been plain sailing — three massive fires in 1996, 2006 and 2012 required closure of the tunnel for repairs — but the tunnel is now an invaluable element of infrastructure in the daily operation of our economy. Yet most of us take its existence for granted.
Whilst there are those above water trying to unpick decades of collaboration with our continental friends and family, the Channel Tunnel and all those who worked on it cemented our connection with Europe like nothing else could.
Considering its strategic importance and the scale of the task to create it, it’s not an exaggeration to describe it as one of Europe’s greatest engineering achievements.