Shani Wallis, TunnelTalk

Things did not go according to plan in Delhi. TBM excavation at one point was inching along and was off programme by months. Nevertheless, work is now on targetfor a ROD (revenue opening date) some seven months ahead of contract. TunnelTalk travelled to Delhi to find out how this could possibly be the case.

IMCC celebration of final TBM breakthrough

On 3 September 2004 celebration of the last TBM breakthrough on the IMCC MC1 B contract for the Delhi Metro Line 2 project carried a great sense of relief. The 6.5m diameter TBM had taken 369 days of 24/7 shift work to complete 780m of tunnel and had left behind a cutter consumption rate in excess of one per meter.

"We knew conditions on the proposed rock TBM drives would be difficult," said Klaus Münz, of Dywidag International and project leader for the IMCC JV, with partners Samsung (Korea), Larsen & Toubro (India), Shimizu (Japan), and Ircon (India), "but nothing prepared us for what we met. The heavily tectonically disturbed bedrock of the Himalayan foothills region combined with the heavy weathering of Delhi's Yumana River flood plane has created a geology of mixed hard, tough quartzite of up to 250MPa and near vertical bands of highly weathered, very weak mica schist, all hosting a high ground water table. For TBM tunnelling this presented on occasion full faces of extremely abrasive quartzite but more often the extremes of both hard rock and soft ground conditions in one face. When production was down to less than one ring or 1.2m/day and the programme was becoming critical, we consulted various experts and manufacturers to advise us on a solution but none were able to guarantee that the alternative would do any better. It has been the most difficult set of TBM excavation conditions I have experienced or heard of and I think the manufacturing industry agrees."

Plan A
When TunnelTalk visited the project for the first time, the plan was to use one Herrenknecht rock EPB TBM to excavated the 1.6km length of twin tube running tunnel from Delhi Main to New Delhi stations in three drives - mining through Chawri Bazaar Station on the first drive and being pulled through the excavated station on the second (Fig 1).

At the same time, two refurbished Herrenknecht soft ground EPB TBMs would complete six drives through favourable Delhi Silt. First from Patel Chowk to Connaught Place stations from New Delhi back to Connaught Place at the south end (being pulled through the excavated Connaught Place Station) and from Delhi Main to the ISBT (Interstate Bus Terminal) stations at the contract's north end for a total of 5km.

Fig 1. Planned and actual excavation sequence for contract MC1B

Five of the contracts six stations are open cut boxes with the sixth, under the old, incredibly dense and crowded Chawri Bazaar district of Old Delhi designed as a mined NATM excavation between small surface accesses at either end.

Things began well, although activity within and underneath the Chawri Bazaar was always going to be a logistical tangle with the added challenge of rnaintaining the integrity of the buildings - many of which, either side of narrow alleys, are hundreds of years old, on very rudimentary foundations, of poor construction quality and practically ramsnackle in condition.

Dewatering in this vincity was strictly controlled and limited by the contract documents and stringent surface settlement criteria were set. Under the lump-sum, fixed-price, design-price, design-build contract, the contractor was responsible for all geological and geotechnical risk, all third party liability, and all necessary liability insurances. The stakes were high for the IMCC (International Metro Construction Consortium) JV.

The two refurbished EPB machines were first on site and got off to a good start from the Patel Chowk in May and June 2002 respectively. "Progress through the Delhi Silt was actually quite good and we achieved an average of +6.5 rings/day for the first two drives and of more than 10 rings/day for the first two drives and of more than 10 rings/day on the second two," said Steven Meredith, IMCC tunnel manager. "The problem was with the rock machine drives."

The new Herrenknecth EPB rock TBM set off on its first 100m long drive from the Delhi Main station box on 23 October 2002. Seventeen months later on 23 March 2004, it broke into the excavated Chawri Bazaar mined station platform tunnel after travelling just 926m in 518 24hr working days, six days/week.

"The time was spent perhaps equally between grinding through very hard, tough quartzite and struggling through mixed face conditions, passing from material banded mica schist of no positive strength to very hard quartzite, and in maintenance downtime to change worn, broken, pitted and flat spotted cutters. Changes were needed often, after every shove, replacing on occasion up to five or more of the cutterhead's 44 discs per 1.2m ring," said Meredith. "We also had two long mid-Drive downtimes of eight weeks and four weeks where we undertook, in free-air, major structural repairs of abrasive ware to the machine and modification of the cutterhead, to attach grizzly bars to prevent blocks of rock entering the chamber, and to reduce the size of the openings to prevent softer wet weathered mica schist running from the face and threatening settlement of the buildings above."

The solution
As reported in the March 2002 Delhi Metro article, the JV was prepared to order a second rock EPB TBM if, as predicted, the rock drives proved slower going than programmed, but as Münz explained; "there was just no time to procure and launch a second rock TBM. Also a new TBM investment would not provide the increased production needed to salvage the programme. What we decided to do was to take one of the refurbished EPB TBMs after breakthrough of its second drive in June 2003 and convert it to a rock EPB machine by fitting a new rock cutterhead."

Working again with Herrenknecht and with the experience of the first drive, setting on the design and incorporating upgrades and modifications from the original cutterhead was not straightforward. "it was a classic conflict between those who approach the task from a hard rock TBM background and those with soft ground TBM expertise," said Münz. "The result was a compromise built on an available soft ground EPB centre piece used to ensure a more speedy cutterhead delivery."

The new cutterhead arrived and was fitted in time for launch of the converted machine in the eastern B2 rock tunnel drive on 28 August 2003, long before the TBM in the first western drive ended its slow going in March 2004. The second TBM, as predicated, had the same struggle. "In such mixed conditions, it was practically impossible to operate in genuine full-plenum EPB mode," said Meredith. "We operated more regularly with the chamber about half full."

Cutter consumption also remained exceptionally high. To control this cost, the policy of the JV was to research supply of the least expensive, best performing cutters. "These we sourced from Korea," said Meredith. "The supplier Dong Yang invested a great deal of effort to modify the discs (lower the ring profile, decrease the space between the rings on double cutters, and improve the hub and bearing seal designs) to get us a better cutter consumption rate on the second drive - though still very high."

Progress was also, as predicted, little better. The second TBM achieved an average of 1.76 rings/day just slightly better than the 1.45 rings/day on the first drive. More was needed to save the programme.

NATM to the rescue too
This came in the form of converting the short 387m B3 running tunnels between Chawri Bazaar and New Delhi stations from rock EPB TBM to NATM excavation. A third NATM running tunnel heading was also advanced for some 71 m towards the second rock TBM heading in the zero rock cover B2 tunnel section to recapture further time (Fig 2).

Fig 2. Challenging conditions for 71m of NATM for the B2 drive

Chawri Bazaar Station beneath Old Delhi was already planned, designed and started as a NATM excavation but changing the running tunnels to NATM was not straightforward. A degree of ground water drainage was permitted for the station excavations but the General Consultants (GC) representing the client on the two design-build Line 2 metro contracts (a JV led by Pacific Consultants International of Japan with Parsons Brinckerhoff International, JARTS and Tonichi of Japan and RITES of India) had precluded any form of dewatering along the route of the running tunnels.

"Changing to NATM did involve risk," said Tony Burchell, project director of GC, and even for excavation of the station, a 2.2m x 2.75m pilot tunnel was advanced and various test undertaken to determine the behaviour of the ground under different methods and techniques before the contractor's full NATM proposal was approved. For converting the B3 running tunnels to NATM and given the NATM station experience, we permitted and insisted upon dewatering of the ground but via long drainage holes drilled ahead of excavation rather than deep hole vacuum pumps from the surface."

Lead consultant to the IMCC design-build Jv, Mott MacDonald, advanced the early stages of the NATM works but as excavation approached, the JV engaged Geoconsult of Austria to provide specialist NATM services to refine design; specify, install and monitor the necessary instrumentation; monitor the works; and provide on-site engineering during excavation. "Careful and steady was the approach," explained MI Singh Saini, IMCC's deputy project leader (NATM).

Extremes of mixed face conditions got the better of the rock EPBM

From the initial NATM shaft excavation an 80m long drain hole was drilled parallel to, and at the same elevation as, the platform tunnels to lower the ground water. In the running tunnels, gravity drainage comprised three 6m-12m long weep holes into the faces and 21m long lateral drain holes into the outside tunnel walls. In addition, a very tight NATM excavation cycle was adopted. Top heading and bench plus invert cycles were progressed consecutively on maximum 1.5m cycles with the top heading a minimum 5m ahead of the bench for an effective stagger, and the invert closed within 30m of the top-heading face or earlier in weak strata.

Further to this, 6m long forepoling on close centres and with long overlaps were installed where necessary. A dumpling left in each top-heading provided stability as well as crown access for installing the wire mesh and lattice girder support, and shotcrete was used to seal suspected unreliable faces during the support cycles.

While these measures are regularly associated with softer ground NATM excavation, not all in the Delhi tunnels was soft. Blasting was needed to excavate the hard quartzite in extreme mixed face conditions.

Through all, three full time Indian geologists mapped every face of advance for review by the two on-site NATM engineers from Geoconsult. Real time readings of installed instrumentation, including horizontal inclinometers, pressure cells, strain gauges, electrolevels etc., were also recorded, with the computer server sending out an instantaneous mobile phone message to 25 senior engineers and project managers - should readings exceed the 25mm emergency trigger. This night-shift facility was never activated but cumulative settlement over the course of the two B3 drives did reach between 30mm and 40mm maximum.

Careful NATM work applied to excavate Chawri Bazaar station

This was due, according to Saini, almost equally to the affects of draining the water table and to excavation deformation. "There were a few instances of weak material dropping from the crown but never a major event of crown or face fallout." A significant achievement on the contract here in India is that there was no serious injury or fatality caused as a direct result of excavation activity.

The only event of mishap was on the rock TBM drive when the machine is suspected of having encountered an old unidentified water well. "Along with unidentified wells, there are also unrecorded basements and extensions to basements," said Meredith. "The NATM work encountered one of these unidentified wells but it was backfilled and the forepoling avoided any kind of ground loss. On the TBM there was no forewarning. Once surface settlement readings and TBM operation gave indications of what was happening, residents of a home directly above were temporarily relocated." Unfortunately the old home structure was damaged beyond reasonable repair and a new home has been built for the family on the same spot - at the contractor's expense.

Mr Sreedharan, Director of DMRC, shakes hands with Tony Burchell, Director of General Consultants (left), and Klaus Münz, Director of the IMCC JV (right), at the final TBM breakthrough

Positive ending
When TunnelTalk joined celebrations of the last TBM breakthrough into the short section of B2 NATM running tunnel, there was relief all around. "With this breakthrough, this is the end of excavation on our 7.9km long contract. We are very happy that it is over, even at a much higher cost for the rock tunnels than originally estimated," said Münz.

Breakthrough marked the end also of all excavation for Line 2. IMCC and its counterparts, the KSHI JV on the adjoining 4km of cut and cover subsurface alignment, are now working towards completing construction of the full 11.9km long, 10 station Line ready for an earlier revenue opening date (ROD) than the planned January 2006 and March 2005 dates respectively. Despite facing difficulties of its own on the cut and cover works, the 50-month KSHI (Kumagai/ Skanska/Hindustan/ltochu) JV's MC1A contract will meet an ROD five months earlier than planned in November 2004 while IMCC is working towards having its 60-month MC1 B contract open to the public seven months earlier than planned in June 2005.

When asked how could this possibly be achieved, DMRC (Delhi Metro Corporation) managing director Mr E Sreedharan said: "It is all a matter of working as a team and of being fair and reasonable. We have been sympathetic to the problems faced by the contractors and we, with our General Consultants, have made every effort to help were we can. The contractors have then responded positively in return." The fact that these are lump-sum fixed-price design-build contracts and have hefty liquidated damage clauses for late completion might also have a substantial bearing on their progress through major programme delaying problems.

Once the 70km network of Lines 1, 2 and 3 is fully opened by the end of 2005, another 75km to extend these lines, mostly elevated and at grade, is planned to be in operation by 2010 for when Delhi hosts the international Commonwealth Games.

Prior to that, IMCC is finishing its MC1B contract. This includes erecting rings of the proposed rock EPB bolted and gasketed precast concrete segmental g to finish the NATM running tunnels. Using a purpose built stand-alone segment erector, designed and fabricated by a supplier in Mumbai and mounted on the back of a truck, and invert segments standing freely until the annular grout and proof grouting secures them as an integral part of the finished tunnel.

Other points discussed during the visit included the following:

Rock bolts were not used as part of the NATM support regime. "Their use would have been limited in such variable mixed conditions," explained Saini. "Only Swellex might have provided some benefit but the benefit against the cost of installing them would also have been limited."

Shotcrete was not steel-fibre reinforced. "Steel fibre was not needed to provide that early tensile strength advantage," said Saini. "We installed wire mesh as a better form of support in this mixed geology. The Indian Crews were also more familiar with wire mesh and have confidence in its application."

Compressed air as an aid to NATM excavation or as the controlling mechanism for a slurry TBM alternative was not considered viable. "The risk of blowout beneath the shallow cover and escape of air through old unidentified wells would be too great," explained Meredith. "As it was, we had escape of rock EPBM conditioning foam into a deep basement and up to the surface excavation on three occasions and had to monitor carefully the working pressures through the mixed face conditions."

The ability to pull the cutterhead back from the face by a certain degree is a facility that Meredith and Münz agree would have been a useful addition on the rock EPB machines.

A stand alone erector builds segments to line the NATM running tunnels

Rotating disc cutters were said to be practically useless in such extreme mixed conditions. They were useless in the soft weathered mica schist and did not perform adequately in the hard quartzite, grinding across the abrasive surface rather than chipping me rock face. "We would have been better off fitting blocks of hard, toughened steel, but for the fact the housing were designed only to accept disc cutters," said Münz. "Such blocks would rip through weaker and stronger material with feed through the screw conveyor being the first line of defence against over excavation, face loss, and surface settlement."

Natural drainage of the groundwater was said to be key to the success of the NATM and EPB TBM drives, despite the fact that it was limited or precluded in the contract. "Natural drainage of the material at tunnel horizon via earlier adjacent excavation and through the drainage and weep holes was essential to the success," said Tony Burchell of GC, "and did not cause the surface settlement and damage to buildings initially feared." Methods and the affects of lowering the water table or reducing the water content in different geologic compositions it seems needs re-examination.

References

Delhi Metro emerging - TunnelTalk, March 2002

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