When the Gotthard tunnel was initially divided into construction sections, the shortest length was allocated to Sedrun towards the middle of the project. It would take the greatest effort.
Apart from the Piora problem further south, this was where the really difficult ground was known to exist, with rock so cracked and fragmented it could be pulled from the face by hand. Under more than 1,000m of overburden, at this point it would squeeze and deform.
Making a large bore would be hard going. “If it was possible at all,” says Heinz Ehrbar, AlpTransit’s chief engineer. “There were some very expert geologists who said the rock would be just ‘toothpaste’ and it could not be done.” Even core drills had become stuck.
Others believed this was exaggerated but “no-one really knew because there was no experience that deep in the massifs,” says Ehrbar. There was, however, deep experience in coal mines in Germany, at about 1km down, and also in the famous gold mines of Johannesburg. Technology that might help was there.
The problem was how to handle very big deformations of the rock that would be inevitable under high pressures. That could not be done with conventional anchors, steel arches and shotcrete.
“You just cannot put rigid support close to the excavation front because the forces are so great it will be destroyed immediately, even the strongest of profiles,” says Ehrbar. In his office he keeps a piece of steel section from an unexpected area of squeezing ground. It is thicker than the strongest miner’s bicep and is almost completely twisted back on itself, proving the point graphically.
US Tariffs are shifting - will you react or anticipate?
Don’t let policy changes catch you off guard. Stay proactive with real-time data and expert analysis.
By GlobalDataBut the tunnel face must be supported or it could collapse.
The way out of the quandary is to let the rock deform to some extent and then gradually start to resist it. At a certain point the diminishing force of the rock and the increasing strength of the support will intersect and “you have a point of equilibrium,“ he explains. Stability is achieved and normal rigid support can be added.
“You can describe it as ‘calming down’ the rock,” he says. “You have not to be afraid of deformation, but know it is under control.”
How to do that was imported from the German coal mines where a system of telescopic, sliding arch supports had been in use for several decades. They used overlapping sections of curved, profiled steel which were connected by specially designed bolted friction joints. These would “give” as pressure increased allowing some deformation of the cut.
“They were initially devised for building the hulls of submarines, which face the same difficulties as they dive deeper,” he says. They are made with a very special high strength steel he adds, with particularly low brittleness. “The profile of the sections is also important and you could say that the whole experience of the mining industry is embedded in their form.”
Although they worked in the mines, they had not been used in tunnelling. More importantly they had not been used on the scale required for the train tunnels. “The mine passages are 20m2 in section and we needed 100m2, with a face more than 10m high,” says Ehrbar.
Calculations and tests, including detailed studies on the rock at Zurich’s ETH technical university, and on the telescopic arches, suggested that the system could be made to work. It would mean slow going, perhaps as little as 0.4m a day in places.
Additionally the excavation would have to be much larger in diameter than elsewhere, up to 13m, to allow both for the squeezing and for the eventual inner concrete lining to the tunnels, which must give the running tunnels long-term stability for the next 100 years. In much of the Gotthard, the lining needed is around 300mm thick but for Sedrun between 600mm and 900mm thickness of concrete was used.
“We even had a design option of 1.2m thickness, though in the event that was not needed,” says Ehrbar.
This extra excavation was somewhat paradoxical since removing more of the ground only added to the effect of the squeezing.
Other complications beset the section. To begin with, there was no easy access from high on the Gotthard pass and that meant the tunnel would have to be reached by a vertical shaft close to the Alpine village of Sedrun. By starting inside a mountain, via a 1km long horizontal tunnel, this shaft could be less than the full height of the peaks, though it still needed to go 800m deep.
Like the Faido access tunnel further south, the shaft would have a permanent function as part of the emergency and ventilation facilities for the operating tunnel around a multi-function station. So the contract also included the excavation of the additional caverns, crossover tunnels, side and ventilation tunnels making up the MFS at the shaft base. These were not in the most squeezing ground however, which was in the 2km of rock to the north of the access. Another 5km of excavation, would take the tunnels southwards through some faulted but mostly good ground.
So time-consuming was this work expected to be in total that the 1km long access tunnel was begun very early, in 1996. A South African company from the goldmines, appropriately called Shaft Sinkers, had the particular experience to win the lead role in the contract for the blind excavation of the 8m diameter shaft down through the rock, begun in 1998. It worked with local contractors Murer, Locher, Marti, Zschokke and CSC, using a platform-mounted Atlas Copco multi-boom jumbo for the drill and blast.
A construction camp with semi-permanent buildings was also set up in a small valley of the Rhine below the village, where the access tunnel portal emerged and a small rack-and-pinion rail link for later spoil disposal.
The shaft was envisaged initially as the main access for the construction and later for the MFS. But one of the bidding groups for the main contract, the Transco-Sedrun consortium of Swiss contractor Implenia leading Frutiger, Germany’s Bilfinger Berger and Italian Pizzarotti, was concerned about the safety and logistical implications of using only one shaft.
“We were convinced the project would have major safety issues that way” says Jakob Lehner, a supervising engineer for the contractor and currently with responsibility for safety and a number of other roles. AlpTransit had added a second shaft to the design during the long tender but it was to be 4.5m.
“We further suggested it should be 7m, at our own cost, because we thought it should have space for an elevator as well as that in the first shaft,” says Lehner. He believes that the suggestion helped put the bid into a good light with the client. A CHF 1.6bn, before VAT, (USD 1.6bn) contract was signed in April 2002.
Early work therefore included the drilling of a second shaft, this time by a raise bore operation subcontracted to Thyssen Schachtbau in JV with Ostu Stettin and Murray & Roberts RUC. Thyssen used a Wirth Vee-Mole for the shaft, which is 35m distant from the first.
For the main excavations the contractor decided on a Rowa suspended platform installation for both drives, north and south.
As for the contracts at Amsteg, the steel frame platform is mounted on rails suspended from ceiling rock bolts and can keep virtually all the static equipment, including rock crushers, water treatment, conveyors and desanding units in the top half of the tunnel, leaving clear space below for drill rigs, loaders, excavators, railway wagons, and other mobile equipment to pass freely.
Most importantly, on front end of the platform was a special handling machine for the movement and installation of the sliding arch units. The 50t unit was developed and made specially for the project by German manufacturer GTA based on smaller “Streckenausbaumaschine” or “seam support handlers” from the coal mines.
“It has arms with pantograph-like extenders which hold up the first sections of girders,” explains Lehner. “There is another segment manipulator and then there are two man-basket arms as well and from these the installers can put in the intermediate sections and the special clamps.”
The multi-section arches, made by supplier Heinzmann, are bolted to a specific torsion according to the friction needed.
“They slide quite suddenly as the stress is relieved and that can sound like a gunshot,” says Ehrbar. He made sure that the full-scale trials for the system were done in an in situ section at the shaft base early on, with some of the working miners present “so that they could get used to the noises, because the natural inclination for a miner would be to run.”
Lehner says that the noise is more of a pistol shot than the “cannon fire sound or explosion that you get with rock burst or collapse” and that the miners quickly became accustomed. Though it was thought that squeezing might take up to six months to stop, mostly it was over within the first month, he says.
For the northern drive into the most difficult ground the contractor decided to use a single-track rail for muck out. It would be laid on a temporary bed built up from spoil.
“The arch support needs a circular excavation. But we did not dare build up the invert with a concrete slab because the amount of convergence to come was not known. In places this could mean re-excavation if it went too far and so the less there was to replace the better,” says Lehner
In the event of re-profiling, just 200m to 300m was needed only on the south section. But the rail track proved difficult as there was some heave and it made the ride on the trucks erratic.
“That makes them slower and causes a lot of track wear,” says Lehner.
The face was cleared with Toro low profile loaders into the rail cars for the northern drives. This was the arrangement used in the south until the harder rock was entered and drill and blast began. At this point the trucks tipped into a crusher on the Rowa platform, feeding a newly installed 600m long conveyor to the back where the railcars were loaded.
“But we realised this was inefficient and finally installed a fixed tunnel conveyor running back to a loading point near the exit,” says Lehner. “That freed the rail for other purposes.”
Full trucks ran to a two-level fast lift in the main shaft, one 12m3 car per level, making a 50t load. At the shaft top these could run to a muck handling facility outside the portal where they were tipped out with a rotating drum for disposal, which was handled by a separate contractor.
The contractor also moved to cast concrete track bed in the southern tunnels once it was past the first kilometre of potentially difficult rock in the faulted Urseren-Garvera zone.
“After that we were using drill and blast anyway in hard rock,” says Lehner.
In the first, difficult ground excavation was carried out by “pecking” the face with an excavator-mounted hydraulic hammer, as it was for the entire northern drive, which took until 2008 to complete its two crumbled 2km bores.
Here the rock had to be assessed constantly and was “consistent only in its inconsistency,” says Lehner. Progress varied from over 1m to just 400mm daily and even on occasion demanded partial face work in 2sqm to 3sqm sections when there was “clayey powdery material,” he explains. Arches varied in spacing from three rings per metre, to one every 1.5m.
The face was reinforced using steel rods. “The contractor had the option of fibre glass,” explains Ehrbar, but since there was no roadheader work, the more quickly installed self-driving steel was better. It was thought some 270 insertions would be needed; the anchors up to 18m long placed in an overlapping sequence of 90 new rods every 6m.
Mostly however the full anticipated reinforcement was not used. “We mainly managed with 12m lengths and only 80 or so rods,” says Ehrbar. A Tamrock T12 was ordered for the face drilling but a smaller T11 proved more nimble and was used more often. The same machine worked on the drill and blast in the south and the cross-passages.
“We might have had more difficulties at the face if there had been a lot of water,” says Lehner, but nothing more than 70lt/sec was encountered and only in the south.
The excavations had to be prepared for up to 1,000litre/sec inflows with corresponding safety arrangements. Particularly until the first breakthroughs were made, and a drainage route was established, this would have flooded the tunnels and it was demanded that shafts could evacuate everyone in one lift.
Huge pumps were necessary at the bottom capable of lifting such volumes through an 800m shaft. Eight Sulzer pumps from ABB were used until breakthrough.
Like the Amsteg tunnels, Sedrun also had to install massive supplies of cold water to service tunnel cooler units. Overburden as much as 2,300m meant rock temperature could be up to 45 degrees centigrade in the excavation area at least.
Once again ventilation could not perform the cooling duties, even during winter when the outside air temperature drops well below zero. “In fact we had to heat the incoming air to prevent the formation of icicles in the shafts,” says Lehner. “If one had broken off and fallen it would have been fatal.”
A complex ventilation system was used, with 250m3/sec entering via the main shaft to feed a 200m inner ring of tunnel in the multi-function area. From here fans drew supplies for the various drives. Cross passages are sealed with airlocks to make a circuit along one tunnel and back in the other. Exhaust went out of the secondary shaft.
Though the north was not easy, it never proved impossible and was eventually completed nine months ahead of the schedule. The south meanwhile went much better than expected. The first 600m was also anticipated to be relatively difficult rock similar to, if not as bad, as the north. Although there were some faulted sections, progress was very fast.
“With in six months we were a year ahead of schedule,” says Lehner, adding that the contractor had geared up to move at 1m a day and ended up doing 6m. “We actually struggled because we were too prepared.”
Complexities and changes were also met in the multifunction station where additional ventilation tubes were requested by the government for the side passages, to ensure smoke would be exhausted quickly from the passenger emergency refuges if there was a fire. Organising the spaghetti of tubes and vents was logistically complicated.
Remaining work now includes some re-profiling in the south and much of the lining work. The contractor has a sophisticated, hydraulically-adjusted travelling formwork for this, which can allow a variable internal diameter to be created, ensuring no more concrete is used than is required.
Progress has continued to be good in the harder rock, leading to a contract option for an additional kilometre drive to be activated. Later as difficulties became more apparent on drives from Faido, an option for another 1.5km was added in a contract revision, eventually bringing the original 4km southern Sedrun sections to 6,404m at the final breakthrough in October.
Gotthard tunnel, Sedrun north section. Using the man baskets from the front of the hanging platform to install mesh on a partial face excavation in squeezing ground. Picture by Maurice Schobinger courtesy of Transco Sedrun Tunnel section Sedrun The Sedrun workcamp is sited outside the portal for the shaft tunnel in a small valley 150m below the village. A small rack and pinion railway was built to carry materials in and spoil is removed by conveyor. Picture Alptransit Gotthard Installing sliding arch supports on the Sedrun squeezing ground section Jakob Lehner from, Transco A roadheader is used to re-profile an area of tunnel where convergence has squeezed the diameter at Sedrun. Picture couresy of AlpTransit Gotthard Lining work on the multi function station area where tunnels diverge for the cross over Installing face anchors for the squeezing ground with a Tambrock T11 at Sedrun. Picture by Maurice Schobinger courtesy of Transco Sedrun
