These days I despair for our tunnelling industry in the UK, which appears to have been hijacked by armchair experts, computer geeks and regulation fanatics, no more so than in sprayed concrete lined tunnels. Whilst I can see the need for some input from such specialists, when did we lose sight of the simplicity, elegance and practicality of the sprayed concrete method and replace it with the monster of their devising we now call SCL?
In any debate about the subject we must inevitably refer back to the New Austrian Tunnelling Method, which was not originally devised for soft ground but was adapted from its hard rock, high stress version by a collaboration between eminent Austrian and German engineers in the 1960s for use in constructing the Frankfurt Metro in a similar clay to that found in London. Perhaps it was a mistake to retain the NATM acronym for the soft ground version but the rather catchy name stuck and the method quickly caught on in Europe and then Worldwide for the very good reasons that it was simple, adaptive, cost effective and above all it worked.
Unfortunately, it met with massive resistance in the UK throughout the 1970s and 80s on several fronts, especially contractual, as it was difficult to incorporate the flexibility of the method within the rigid confines of the forms of contract in use at the time. Also, the traditionalist tunnellers vigorously expressed their scepticism by asserting that shotcrete was poor quality concrete that would not stick to clay or chalk, which was subsequently demonstrated to be erroneous on all counts. Perhaps the biggest anti-NATM cry was that it was neither new nor Austrian, which may well have been true, certainly German engineers had a big input and shotcrete had been used very successfully elsewhere, notably in Norway. However, I suspect that the main reason was that it had the word Austrian in it and not British. Incidentally, the acronym was originally adopted simply to distinguish it from the old Austrian tunnelling method and not for any overt nationalistic motives.
Eventually sanity prevailed and NATM began in the UK, not in civil engineering but in mining. However the inadequacy of traditional tunnelling methods to construct the massive underground complexes required for London’s new transport infrastructure forced the UK tunnelling industry to adopt NATM and all went along swimmingly until the Heathrow debacle in 1994. What cannot be denied is that it was a monumental cock-up and the companies responsible were duly prosecuted. What then followed was a lengthy process of collective navel gazing from which emerged the SCL method.
What seems to have been forgotten was that all tunnelling methods can potentially fail given a particular circumstance or combination of circumstances such as poor design, inferior materials, substandard construction, poor supervision or unforeseen ground, but above all there is only one mechanism that causes a tunnel to collapse and that is gravity. Tunnelling in whatever context is in the end a contest against the force of gravity with gravity holding most of the aces.
Pick and mix
The mistake was that NATM was turned into cookbook of the engineering codes and standards beloved of the British engineer and has since evolved into a methodology that would make Delia Smith [a British cooking author] proud. The fundamental but true criticism of the NATM is that it was never clearly defined as to what it is and is attempting to achieve. Unfortunately the same can be equally applied to SCL. When this question is raised, one commonly held belief is that NATM is essentially an observational
method whereas SCL is not; i.e. SCL is designed to cope with all the anticipated ground conditions and performance requirements. If this is so then what is the purpose of monitoring it? One would reasonably assume that it is to check whether the design is working correctly. If the monitoring shows that it is not, what then, change the design? If so then it is, by definition, an observational method. By way of confirmation an SCL design often incorporates some ‘pick-and-mix toolbox’ support supplements (rather like NATM).
The SCL acronym itself is somewhat misleading as it refers only to the finished product, when the fundamental key to its successful application is the sequential nature of the excavation and support and above all the most important design consideration is to close the invert as quickly as possible otherwise gravity (remember that irritating component) will have its evil way. So the best approach must surely be the simplest and quickest method in achieving ring closure and, to their credit, British contractors have looked to new technology to achieve this, such as semi-automated laser controlled excavation and robotic shotcreting. I would contend that this was the true and fundamental definition and aim of NATM that was developed using the best technology of the day to achieve rapid ring closure and should be the basis of SCL.
Spurious focus
Unfortunately, SCL development appears to have concentrated mainly on the sprayed concrete lining itself. For whatever reason there has been a fundamental shift towards very high strength mix designs and the exclusive use of steel and/or plastic fibres. On largely spurious health and safety grounds the use of lattice girders and steel mesh has been ditched and as a result we now have SCL linings double the thickness and twice the concrete strength, totally reliant on chemical additives and quality control of the mix and application for holding it up in the roof. Consequently, in a modern SCL tunnel you are at more risk of injury from falling shotcrete than collapsing ground (that damned gravity again) and I await with trepidation the first casualty in the UK – there has already been a fatality in the USA. Let me state my heretical view that fibres and high-strength shotcrete, whilst being very beneficial from a durability point of view, are no effective structural substitute for steel reinforcement both in the temporary and permanent condition and is no quicker to install, given the increased lining thickness and excavation volume required.
I recently had sight of an SCL design of 4m diameter, 12m deep shaft in London Clay. It has a lining thickness of up to 575mm applied in four specified layers. Not only that, on top of an under-reamed and domed, heavily reinforced base is placed a 0.5m thick reinforced concrete base slab below a further 1.5m thick mass concrete plug (you can’t be too careful). From the shaft bottom a 3.5m wide cable tunnel is driven, which has a 425mm SCL lining applied again in four layers. The tunnel is (for no apparent reason) ovoid in profile requiring no less than seven setting out points and, just to be sure, has an 85mm internal profile tolerance, oh, and a further 75mm allowance for a drip-shield (you really can’t be too careful). Amazingly this ‘heroic’ design has been constructed and I expect the contractor enjoyed himself hugely spraying up the world’s most expensive textured rendering job.
Finally, let me air another heretical view that London Clay and the underlying Lambeth Group is arguably some of the best tunnelling ground in the World and, rather than a soil, can often be classified as a weak rock. Perhaps it would do our armchair SCL designers some good to turn off the finite element program for an hour or two and venture underground to view it in its native habitat, then decide whether the computer output reflects reality or not.
