Shani Wallis, TunnelTalk

A new TBM to save a project that had fallen seriously behind schedule is not a decision taken lightly. But such was the situation at Cleuson-Dixence in Switzerland where a new Lovat TBM was commissioned to drive a steep inclined drive to rescue a contract that had fallen more than a year behind schedule and was under critical time pressure.

Viability of the new SF1.1 billion ($US786 million (April 1997)) Cleuson-Dixence hydroelectric project in Switzerland was based on a five-year construction period starting in mid-1993. Excavation of more than 26km of 4.8m to 5.8m o.d. tunnelling and construction of a 1,200MW underground power station at Bieudron was programmed to be completed ready for on-line power production by October 1998 to meet peak electricity demand. For EOS, the regional western electricity generating authority in Switzerland and owner of the project, missing a full season of production would mean a serious financial loss in capital interest charges. Also Cleuson-Dixence's production will equal 20% of Switzerland's peak hydropower production. It will meet demand in western Switzerland and provide surpluses for export into Europe.

The Lovat TBM built in a hydraulically operated cradle that simulated the 70% gradient

To meet the target, the civil works were divided into four main contracts and five hard rock TBMs were employed to excavate the majority of the project's 26km of tunnelling. Once launched these operations progressed well - except for the machine on the steep Lot D penstock tunnels that suffered serious delays and mechanical problems and was having to cope with difficult geological conditions.

At the end of 1995, the Lot D TBM - and therefore the project - was running a year late. The first season of power production was in jeopardy and EOS was forced to examine all possibilities of completing Lot D in time to save the first 1998/99 winter of production.

After much discussion and analysis, the project management team came to the conclusion that a second TBM would be needed to complete Lot D on schedule. In addition to dealing with equally difficult geology, the second TBM would have to meet three other fundamental requirements if the proposed solution was to succeed.

1. Speedy delivery. The remaining 2.9km of TBM tunnelling had to be completed by December 1996. That is in less than 12 months. A second TBM would have to be built in weeks not months.

2. Swift and trouble free assembly and launch. The working site for the new TBM, its launch chamber and all back-up services would have to be prepared ahead of time. TBM assembly to a tried-and-tested formula, with a minimum number of components and a rapid start-up, would be essential.

3. The back-up system would have to be simple. There would be no time for a prolonged learning curve and steady sustained progress would be needed once into the routine.

Finally, given the critical status of the situation, the second TBM manufacturer would have to be selected and would have to work as a team player with EOS and the Lot D contractor if the plan was to work.

In such risk management situations, decision making and rapid implementation of works are major challenges - challenges which require excellent collaboration and true partnership between client, contractor, machine manufacturer and system suppliers.

The Lovat solution
Following further core drilling and geotechnical surveys, Jacques Botte, tunnelling project manager for EOS, contacted Lovat. Botte, an ex-tunnel contracting engineer from France, was familiar with the Lovat concept and after discussions with Lovat and other contractors operating Lovat TBMs, Botte believed that a Lovat TBM would meet the specifications.

The TBM in its launch position

To deal with the friable material known to exist on the tunnel line, Lovat and EOS discussed two machine concepts, one based on a screw conveyor system and another based on Lovat's patented pressure relieving gates (pRGs). After reviewing the geotechnical data and a visit to the project, Lovat recommended the PRG system. Lovat's cutterhead flood doors, the machine's bulkhead, the muck ring and the PRGs would provide greater flexibility in dealing with varying conditions. The mechanical control of the face offered by the concept, particularly on a steep 70% incline, would avoid over excavation of friable mylonite material. The system also offered quick, easy and safe access to the face for cutter maintenance.

In early 1996 Lovat was selected as the recovery TBM manufacturer and an order for an RMP174SE TBM was placed on 18 March 1996.

Having adopted the recovery TBM option, a supervision executive committee, comprising two representatives of the Lot D contractor and two of the owner's project management team was created to supervise the operation while an operations committee was set up to

handle technical details. This operations committee included Botte, the project's tunnelling manager and Michel Zermatten, an engineer with more than 30 years' experience in tunnelling who was selected by EOS to run the second TBM solution. To bypass contractual and financial concerns EOS, with the co-operation of the contractor, took full responsibility for the proposed solution. It agreed to buy the machine and the back-up, to finance another working site at intermediate adit F5, and to manage the entire recovery proposal.

Steep undertaking
Lot D was a tough task from the outset. The 4.1km long penstock on its steep 64%-70% (32-35°) incline drops some 1,800m in elevation and constitutes the highest head of any hydroelectric power station in the world at present.

Breakthrough after completing the steep 400m drive in just 14 weeks

In addition, the geology on the north side of the Rhone valley is known for its complex, glacial composition of highly weathered layers of white granular gypsum and crushed and squeezing mylonite which intersperse sediments, carboniferous and metamorphic rocks and highly abrasive quartz.

The TBM commissioned to excavate these steeply inclined tunnels was a Robbins double shielded machine which was manufactured in Italy and incorporated several modifications designed by the JV partners. Behind the TBM the JV installed bolted, reinforced precast concrete segments as the necessary primary support. An inner 3m to 3.4m i.d. steel lining installed under separate contract will contain the internal pressures of up to 200 bar exerted by the 1,800m hydrostatic head. Intermediate adits to assist steel lining installation also provide access for tunnel excavation.

On being launched in July 1994, the Robbins TBM was expected to complete the 3.7km length of TBM tunnelling for Lot D to hand over to the steel pipe installation contractor in February 1996. But the machine fell a year behind schedule as a result of various problems. There had been delays launching the TBM on its 68% incline; it had taken time to install the gantries of the 108m-long trailing back-up and to master the operation of the TBM on the steep incline; and over excavation through the open, dome-shaped cutterhead was a major problem in the crushed mylonite. Large quantities of polyurethane resin grout was injected to provide sufficiently stable conditions for effective TBM advance. In early 1995 the TBM suffered a five month delay when broken teeth on the bull ring of the cutterhead required complete replacement of the component in-situ.

The Robbins TBM did finally breakthrough at adit F6 in April 1996 but the time needed to complete the remaining 2.9km of Lot D machine tunnelling had been lost. The second TBM recovery plan was based on working both the Robbins and Lovat TBMs at the same time. The existing machine would carry on from Intermediate adit F6 to F& and the new TBM would start at Intermediate adit F5 and excavate the last 400m of TBM tunnelling at the top.

But a new machine and set up for only 400m?

"Yes. Not an easy decision to make," confessed Botte. "Discussions were intense but progressive."

Critical schedule
Given the projected programme rate of 4-5m/day, 18h/day, 7 days/week, tunnelling with the Lovat TBM would have to start no later than early September to finish the 400m drive by December 1996. On August 5, just 21 weeks after the official order date the Lovat TBM was ready for despatch. As part of the agreed tight delivery schedule, the shield of the machine, weighing 110 tonne, was shipped from Canada to Geneva, Switzerland by air in Russia's enormous Antonov AN-124. This was the second Lovat TBM to take a ride in the world's largest cargo plane. The shield was built in two semi-circular sections especially to accommodate the Antonov's cargo bay, an exercise which further complicated the engineering design, manufacture and assembly of the shield. The machine's other associated components, weighing 62 tonne, were air freighted via commercial aircraft.

From Geneva airport, the TBM was trucked up the steep alpine roads to the Intermediate adit F5 site where it was backed down the 300m long adit and unloaded into the working chamber,both of which had been enlarged

A ride in the Russian Antonov

for the purpose. Seven days later, the TBM was on its way with the full back-up system.

The 172 tonne Lovat machine was actually built in a special hydraulically operated cradle which was purpose-built by Lovat to simulate the steep 70% (35°) slope on which the TBM would operate from launch to breakthrough. Operating a TBM on a 70% (35°) angle is far from normal and the special simulation cradle assisted the Lovat technicians and design engineers in taking in all the considerations of the working application. Manufacture of the machine on the working slope was also instrumental in ensuring a successful and speedy launch.

In addition to building the machine, Lovat also provided two qualified operators and two experienced technicians to maintain round-the-clock TBM operation. It also supplied a consignment of emergency spare parts and wear consumables to prevent TBM downtime and had interfaced with the design of the trailing back-up which was manufactured in parallel by Giovanolla in Switzerland.

Recovery plan
In setting up the second TBM heading, it was decided to keep the systems fundamental. As Zermatten said, engineers, in their desire to develop technology, can forget that the standard solution has equal compensation as the high-tech option. "We didn't have the time to make a complex system work. From mucking out, lining design, back-grouting and back-up supply lines, we needed a straight forward concept that would keep going through all types of comprising layers of finely ground and with absolute minimum downtimes."

Looking back down the steep penstock

In describing the TBM itself, Zermatten explained that the machine was designed for powerful and consistent progress - "not as a high speed racer". The machine plus all the associated systems were calibrated for an optimum advance of 14m/18h day and a programme rate of between 4-5m/day working 18h/day, 7 days/week.

The Lovat TBM was built with a variable-speed hydraulic drive powered by three 225kW water-cooled electro-hydraulic motors. The cutterhead rotated at 1.6-3.0 rev/min and maximum torque was 410tm. The 4.41m diameter cutterhead was able to turn in either direction and was equipped with mono block single disc Palmieri cutters and Lovat scraper teeth. Flood doors across the openings limited the size of blocks entering the chamber.

For efficient mucking out, the material was allowed to fall freely from a muck hopper on the back-up and down a covered chute in the invert of the 70% inclined tunnel. From the Lovat belt conveyor, cut material passed into the hopper where water was introduced to flush the material down the chute and into a large primary screening tank at the bottom.

From here separated solid material was loaded by conveyor into dump trucks for disposal while the slurry was pumped out to a purification system on the surface. Here water was purified to pumpable industrial quality and recycled, initially through the TBM's cooling system and finally in the flushing system. The cooling system needed a circulation of 6 litre/sec but all pumps in the water recycling system were calibrated at 20 litre/sec.

Behind the TBM, Lovat and the operational team agreed a decision to erect 1m wide rings of bolted, liner plate type steel segments as the necessary primary lining. These are much lighter at 2 tonne/1m wide ring than

Light steel segments were easy to transport and erect on the 70% incline

the 10 tonne rings of 1.2m wide x 30cm thick precast concrete segments erected behind the Robbins machine. The fabrication was easier to set up; they were faster to produce with no moulds involved; they were easier to transport from the factory with more rings carried per 15 tonne load than for the concrete segments; and they were easier to transport up the steep incline to the advancing TBM. Once there, they were easier to handle and erect using the proven Lovat erector arm system.

The trailing back-up built by Giovanolla and designed by the operations committee, was much shorter at 35m-long than the 108m-long system behind the Robbins machine and was also much lighter at 50 tonne as compared to 500 tonne. It comprised four trailers and a self-propelling advance. Rather than being pulled by the TBM, it had four sets of clamps on hydraulic cylinders that gripped the rails and worked in tandem to hold and push the system forward.

A special two-cage shuttle ferried men and materials to the TBM (Fig 1). This was powered by a winch set up in the working chamber and was operated by remote control from within the shuttle. Fully loaded with four men and a ring of liner plates the shuttle weighed about 3.5 tonne. An 82kW winch was sufficient to work the Lovat system shuttle. By contrast a winch of 1,000kW power was needed to operate service trains behind the Robbins machine, which also included muck-hauling skips.

Fig 1. The Lovat TBM and its back-up systems designed for a steady 5m/day up the steep 70% incline

At the Lovat TBM back-up, the steel segments were off loaded onto a self motorised trolley which carried them to the ring-build area in the tail skin. The Lovat machine had no grippers. It pushed off the lining exerting a maximum 2,280 tonne thrust via the 24 shove rams. During the 1.67m stroke, the lining was grouted up using a special grout mix and a tail-skin grouting system. The grout was mixed at a station in the working chamber and pumped up the maximum 240m head to the advancing TBM. Accelerator was added at a small mixer on the back-up and from here grout was pumped through the wire brush tail seal.

TBM performance Following a fast start, Lovat's two operators and two technicians had the TBM on a steady routine within days of the first shove. Then on November 28, within 14 weeks of starting and some four weeks ahead of schedule, the TBM broke through. It had advanced at a mining average of about 6.5m/day and had reached peaks of 13.25m/day. During the drive, the machine had operated mostly in the open mode but, in describing the ground, Zermatten said that the geology had presented no gifts. The machine had passed through several bands of mylonite ranging from large boulders to sandy crushed fines and on through a 270m stretch of very hard and abrasive quartzite of up to 120 MPa in compressive strength. Probe drilling and grouting ahead of the face was possible via four ports and a factory-fitted drill rig but these facilities were not actually needed. In adverse conditions the PRGs (pressure relieving gates) successfully controlled the volume of excavation and the cutterhead flood doors helped moderate the muck intake. Under these conditions the PRGs were set at approximately 2 bar pressure. Water was injected into the excavation chamber to lubricate the material and reduce wear in the quartz while a liquid soap polymer was injected in zones of carboniferous shale and softer materials to limit torque. The cutterhead had been coat-welded with half-inch tungsten plate to protect against wear. Cutter consumption was much as expected with 13 discs replaced over the 400m drive. The longest hold up during the drive, according to Zermatten, was a planned downtime in which the TBM and the various support systems were calibrated and balanced for optimum advance.

Table 1. Principal statistics of the Lovat TBM Ordered March 18, 1996 Exit works August 5, 1996 Total fabrication time 23 weeks Delivery by Antonov air freight 3 days Transport to site and assembly 16 days total Started excavation August 24, 1996 Breakthrough November 28, 1996 Average daily advance 4m/18h day (two 9h shifts/day) Best advance 13m/18h day Total excavation for 400m drive 14 weeks Total from order to completion 37 weeks TBM diameter 4.413m nominal Connected power 848kW Cutterhead power 3 x 225kW electric water-cooled) motors coupled to10 variable displacement hydraulic drive motors Main bearing Cross roller type with pressurised lubrication system Cutterhead design 27 x 15 back-loading disc cutters 40 Lovat scraper type teeth Pressure relieving gates Set at 2 bar pressure Thrust cylinders 24 x 95 tonne each Total maximum thrust 2,280 tonne Stroke 1,670mm Tail seal Double row of wire brushes Foam injection system 1,000 litre/min capacity

In parallel
Meanwhile, the Robbins TBM had also finished its second 754m drive. A new, flat-shaped cutterhead had been fitted to the TBM and the contractor had undertaken substantial modifications to simplify the back-up operation. These modifications had taken some three months to complete and, even though the machine did perform much better than expected on its second drive, the time needed for it to complete the top 400m of the drive had been lost.The second TBM was essential and, in the final analysis, the Lovat operation proved a much less expensive undertaking. According to Botte, the TBM and its support services consumed much less energy and was operated with only eight people/shift as opposed to 17 needed to operate the Robbins TBM and its back-up.

"Now, at the end of tunnel excavation, we can say that we are very pleased with the result," said Philippe Mean, Project Director for EOS. "We gained what we set out to achieve. The project is now on-line to start producing electricity by October 1998. The expenditure on a second TBM to save the first winter season of power generation was a gamble but the alternative was unacceptable and we had tightened the odds through detailed preplanning, tight hands-on control and an integrated working team."

Close co-operation between owner, operator, machine manufacturer and materials suppliers had ensured success from what was a serious project situation.

Search

TunnelTalk Articles TunnelCast Videos Archive Feedback Current Diary Dates World News Briefs Career Moves Accolades & Tributes Stay Connected Procurement Job Opportunities Equipment Sales DIRECTORY Marketplace About Us Advertising Contact Us Legal Notices Home Office     Latest Alert Alert Signup Alert