The catastrophic inundation at the face of the Ring Main by saturated Thanet Sand at 4bar pressure as the tunnel was being driven deep beneath Tooting Bec in November 1998 totally changed delivery of projects for Thames Water.
'It triggered the revision to the whole process,' says water services project director David Glendinning. Previously contracts had been let on a traditional and piecemeal basis. The tunnel failure drove Thames into partnering, long before that buzzword had been coined.
The shield at Tooting Bec had unexpectedly run into a lens of highly pressurised wet and very fine Thanet Sand within the Woolwich & Reading beds, which are generally clayey material with some gravels and sands. Recovery of the tunnel took months and cost millions. In the end a massive freezing operation had to be resorted to stabilise a huge block of ground and enable tunnelling to restart.
There were still many kilometres of large diameter tunnel to drive on other sections of the main, all of which had to be completed to turn the installation into a useful operational tool.
'We brought in the tunnel machine manufacturers and contractors,' says Glendinning. Previously each contractor had ordered tunnelling machines that were reckoned to be good enough for the job in hand but with not too much spent on an allowance to cover unknown risks. The nature and risk balance in the contracting business drove the specification to the minimum that would do the job. Manual erection of small lining plates rather than fully mechanised assembly of heavier sections with less joints was the norm. If a tunnel boring machine had to pay for itself on a short drive there was no margin for any refinement.
'With the machine manufacturers we redesigned the machines, and the linings,' adds Glendinning. Thames commissioned the machines and then supplied them to the successful bidders that would drive the tunnels. Partnering contracts were introduced so that difficulties and unknowns became things that were jointly owned, and overcome, by contractor and client.
The second phase of the Ring Main went smoothly from start to finish and was a revelation after the delays and overruns of the first contracts. Lessons learned were fed into the rest of Thames' capital projects.
Pressure on leaks
Asset maintenance is at the heart of any utility business and Thames is responsible for about as large and complex a mass of pipes as any organisation in the world.
It has an excellent heritage in central London's sewers. The oldest include the massive and well built Fleet conduit. Constructed from good quality bricks in 1780 it is still in good condition. So is Joseph Bazalgette's network of trunk sewers which transformed London in the Victorian era.
Problems come with some of the newer installations and the fact that in a heavily built up area there is damage from high levels of road traffic together with almost continual excavation and activity on the part of other utilities adding new infrastructure.
Water mains leakage became an issue with the 1995 drought. Universal metering of individual offices and households is not regarded as cost effective. Unravelling complex plumbing systems in multiple occupancy buildings could cost as much as £1,000/meter installed, the company estimates. But Thames has been prompted to install district metering linked to new technology with real time telephone and telemetry links.
'We can target leaks using the technology. We've taken advantage of our late entry into the field,' says water services project director David Glendinning. Pressures as well as flows are being measured and the monitoring network will be used to manage the pressure in the whole system.
Now Glendinning has a team of 26 dedicated to network modelling of the individual cells in the distribution system. 'The problem is London is so complex,' he says. Typical past interventions in the system are massive fire mains installed during the Second World War to bring large quantities of water direct to the centre for fighting fires during the Blitz. Since then these pipes have been incorporated into the network irrespective of whether there was any engineering logic to them.
Leakage was estimated as 770M litres/d up to last March against a target of 778M litres/d. The target for the year to next March is 665M litres/d.
Investigation and repairs following Thames' most spectacular leak for years - the flooding of a North Circular Road underpass - is still ongoing. Glendinning has one of the rusted bolts from the failed hatch on his desk at Reading.
The main, built in 1953, feeds north London from Ashford Common and runs at up to 12bar pressure. Investigation so far has found that bitumen encasing some of the hatches had not ensured protection against corrosion as was intended nearly 50 years ago. 'The (corrosion) problem is not necessarily ground conditions,' says Glendinning.
The difficulty is that the risks of prompting a failure during investigation mean that 'you can't just dig down. There is a planned sequence of events to examine them.'
It's a Health & Safety issue, says Glendinning: 'The event happened. It's our job to make sure it doesn't happen again.'
The old Metropolitan Water Board which then became Thames Water Authority always had an international consultancy profile, a consequence of its being the water and wastewater undertaker to one of the world's largest and best served conurbations.
But privatisation and the resulting jump in environmental standards has given Thames Water an edge in the competitive business of working with new megacities to install water and sewerage systems. 'We are working or doing business development in 20 countries and have 4,000 people working around the world - predominantly local people.' says business services manager Peter Hemmings.
'British privatisation is leading. We were first and that has been very useful. Governments are very interested in the UK model.' Hemmings sees this from the perspective of someone with a career in London's water since the 1960s. 'It's a wonderful industry to work in!'
Hemmings says the company is 'doing well in South East Asia and we want to balance the portfolio in Europe and America.' There is already a strong presence in each area.
In the Americas, Thames is working in Puerto Rico with Dick Corporation of Pittsburgh. Thames has also been bidding in Chile and in Brazil.
New European activity includes appointing a French country manager this year and a hard look at impending privatisation in Italy. In Adelaide, Australia, Thames is already working with French company CGE on research and development of greywater use for industrial applications. The recently publicised system for the Millennium Dome at Greenwich, using rainwater and partially treated washing water for toilet flushing, is an exemplar for this work.
Asian joint ventures include the recently completed Da Chang water treatment works and a project in Jakarta to install a drinking water system in the eastern half of Indonesia's capital.
Problems for which this joint venture is finding solutions include reducing leakage and sharpening up on 'non-revenue water' - overhauling the customer billing system and dealing with illegal mains connections.
A project in Bangkok sums up the Thames philosophy of working closely with local partners: 'It's no use just going and saying 'we know how to do it',' emphasises Hemmings.
Thames has taken a relatively low profile as far as its Izmit bulk water supply project is concerned. This sits on top of the epicentre of Turkey's recent catastrophic earthquake. Put to this ultimate test it performed far better than the failed multistorey structures.
'But we don't want to bark about it,' says Hemmings.
The main consequences of the large earthquake were a modest 150mm settlement of the dam embankment and a major buckle in the aqueduct where the steel pipe crossed a fault with a throw of some 2.5m. But apart from some small leakage at the creased section of pipe, the system remained in good order.
The London company owns a substantial share of Izmit which is believed to be the largest build, operate and transfer water supply project in the world, and a model for doing business and realising other such schemes. BOT finance rescued the once moribund Izmit scheme which had been started by the Turkish in the 1980s. The dam is Turkish designed and the treatment works and pipeline largely British. It will be operated by Thames for 15 years.
Water is impounded by a 100m high rockfill dam then treated before being fed through a large diameter piped aqueduct to 13 municipalities along the shore of the sea of Marmara.
Circle of supply
London's Ring Main is summarised by Thames economic regulation manager Richard Oake as 'the piece of engineering which enabled us to beat the three year drought.'
It is an 80km long tunnel linking the capital's main water supply reservoirs in the Thames Valley to the west and the Lea Valley to the north with the principal water mains within London.
Credit for the idea of the Ring Main goes to those working on London's water system before privatisation. They began a project that was to be completed in the mid-1990s. 'No other water company has done such a thing,' says Oake.
Water in the main, between about 30m and 40m below ground, runs at low pressure. Flow can be directed in either direction according to demand and the availability of water from the treatment plants. Up to 1,300M litres/d can be moved around the system. Where demand is predicted to be particularly high, due to weather conditions or, say, a major sporting event, appropriate service reservoirs can be filled up overnight ready to cope.
Since water can be moved around in high volumes and lifted out at one of the many shafts interconnecting the old mains system, the pressure in existing pipes has been reduced, increasing security against bursts. Flexibility of the system allows costs to be reduced. Thames has concentrated development of upgraded treatment at major works such as Coppermills, Kempton Park and Hampton while closing smaller ones.
'If we have a problem with say, algal blooms at one works we can source water from another and cut the costs of treatment,' says water services project director David Glendinning. 'It enabled us to rationalise works such as Hampton where there were three separate companies at one time. We could look at an overall integrated water supply.'
Thames still needs a new large storage reservoir remote from the capital but small local facilities at Barnes and Fulham were rendered obsolete by the Ring Main. That gave the opportunity to create 100ha of wetland at Barnes while selling off 13ha for housing development to pay for this new wildfowl centre which is to be officially opened next May.
Water treatment was given more than usual attention by Thames during the first five years after privatisation for a very good reason.
'We keep recycling,' says environment and quality manager Dr Peter Spillett.
Direct abstraction from the River Thames inevitably involves re-use of treated wastewater from upstream. 'We've got to have higher standards - probably the highest in the UK and Europe,' says Spillett. As a result a huge R&D effort was put into optimising equipment for Thames' Advanced Water Treatment process involving granular activated carbon (GAC) and ozone treatment.
Pushed by the Regulator to upgrade its treatment works Thames' R&D department was devising radically revised ways of using GAC while projects were under construction. 'It was quite an exciting time. There was a hell of a lot going on,' says water services project director David Glendinning.
Originally GAC was contained in large concrete box filters. The research work was concentrating on whether the carbon could be incorporated in less costly manner as part of a sand filter bed. The questions to be resolved were: 'Would it work?' and 'How can the large quantities of GAC involved be recovered for reprocessing?'
Reuse of the carbon was vital: 'We used to say it cost £5 a bucketful,' says Glendinning.
The filter bed technique was perfected. GAC is now included as a very precisely placed layer sandwiched within the bed. Life of each bed is about 18 months. Every 28 days or so the upper layer of sand is skimmed off to remove algae. It is then replaced with cleaned sand. Taking out the GAC when it needs reprocessing is a delicate and skilled job involving wheeled loaders fitted with laser levels to guide the operators.
Judging when the GAC needs replacing is critical because as it absorbs material in the filter bed the carbon's density increases from 2.4 to nearer the 2.5 of the sand. Separation of the carbon from the sand is done in a hydrocyclone which would not be effective if the two materials had the same density.
The separated carbon is transported by tanker to a special plant at Kempton Park where it is reactivated by heating with steam in a furnace to 1,000degreesC. Peak period for conversion to the new system was 1995-97 and the final large plant was commissioned at Hampton last Christmas.