Recent work at Harrods department store in London, UK, required a significant amount of handwork using vibrating tools which carried the inherent risk of hand arm vibration syndrome (HAVS). At the start of October’s BTS meeting, Rob James posed the question to the audience: “Is there a future for handwork tunnelling in the UK?” which he surmised would be answered in the presentations that followed.
The talk was split into two sections, with Phil Ball tackling the ground engineering, the compensation grouting and other settlement reducing measures.In the second section Rob James described the tunnelling works and the management of HAVS.
The contract, known as the Harrods X5 lightwell project, was to provide access into the department store from an adjacent plot of land previously occupied by Knightsbridge Crown Court (KCC). This access was to be via a tunnel through the KCC deep basement retaining wall, across Basil Street and to connect with a new lift shaft in the lightwell in the centre of the existing store. Through this new tunnel and shaft would pass the store’s entire range of goods, from Armani suits to fish boxes. The benefits of the new arrangement were security and the ability to service the store without disruption to shoppers.
The design for the new basement to the KCC building concentrated on restricting horizontal ground movements and horizontal strains from the deep diaphragm walls by the use of “settlement reducing piles” and “plunge column piles”. However the focus of the work at Harrods was the control of vertical settlements from the tunnelling works. Many of the techniques for monitoring settlement were borrowed from successful work on the Jubilee Line Extension, particularly at King’s Cross. The area was of great sensitivity and the risk of settlement damage was to be kept to an absolute minimum.
There were three main aspects to the X5 lightwell project: the new lift shaft access with three lifts and a stairwell; the access tunnel to the KCC basement; and the increase in retail space gained by infilling the existing lightwell area. Access to all the works was through the existing building, which was severely restricted because of the need to keep the store open for business and to keep the traffic flowing around the access point in a very busy part of Knightsbridge.
To mitigate vertical settlements from the construction of the new shaft at the heart of the building, 112, 29m deep contiguous shaft piles were sunk prior to shaft sinking. In addition, spine wall piles were sunk in groups to control settlement from the tunnel construction. New bases were also to be constructed for the very heavy sub-basement turbine hall and the heavily loaded, 7m staging area. These were to be built on 38m deep piles, slip coated to tunnel invert level to counteract the effects of the negative skin friction caused by the tunnelling.
The main process, however, for settlement control was the use of compensation grouting. This would employ a tube à manchette (TAM) grouting system from horizontal arrays drilled from the lift shaft into areas above the proposed tunnelling work. Ground monitoring instruments installed above the arrays would measure the effects of settlement and provide information on the effects of grout injection.
To counteract downward loading of the tunnel from the upward action of compensation grouting, a system of exclusion zones was introduced to coincide with the advancement of tunnelling. This meant that, generally, compensation grouting was excluded from an area 5m in front and 3m behind the advancing tunnel face and from one diameter above the tunnel. “This can be different on other sites and may be time dependent,” said Ball. “Some interesting papers on research by Soga/Bolton in Cambridge into compensation grouting has been published,” he added.
The effect of “jacking down on the tunnel” is dependent on the level of the TAMs. Closer to the tunnel, width of the exclusion zone narrows but the effect on the tunnel is more acute and the direct effect on the foundation footings above is much less. “In all cases, the loads imparted to the soil can only be marginally above full overburden once hydofractures have developed. However, close to the point of injection, pressure concentrations may, in the short term, be very high.”
There were many points to consider. Small volume overlapping points of injection while the grout is still fluid can apply pressure over a wider area than anticipated. This can be either good if deliberate or bad if unexpected. Equally, large injection volumes at one spot may encourage grout flow to the tunnel face emphasising the desirability to use many small injections. It was found that short, sharp bursts of grouting were more effective than longer, sustained bouts.
Sequencing of the grouting was significant because:
The construction sequencing and grouting was assisted by the fact that the advancing faces were always close to shaft for much of the time and within easy reach of grouting arrays.
The principles of compensation grouting were:
In practice it was found that the key elements were:
TAMs were set at different levels to maximise flexibility. At high level, inclined TAMs were used with strict control over the use of injection ports in the critical areas.
For the initial pilot tunnels the concerns for deformations caused by grouting were less acute but some deformations measured were attributed to grouting.
As the pilot tunnel was enlarged and returned towards the main building, overlapping exclusion zones seriously curtailed the grouting. Only the high level inclined TAMs could be used.
In conclusion the solution in this case was to:
The causes of failure to grout properly were listed by Ball as:
These causes were not necessarily applicable to the works at Harrods but should be considered in any situation where compensation grouting was to be applied to safeguard structures, said Ball.
Face losses were also a major factor in the effectiveness of the grouting, he said, because it meant increased grout volumes to be injected in the same period of time. It also added a complication to the cycle lag between excavation and the time required to equalise the soil pressures. These, combined with the increased individual injections, caused a loss of control and therefore potentially higher settlements. The control of volume loss was the most critical factor and this should be done at the tunnel face to prevent a vicious circle of movement, more grout, more deformation, and more restriction, resulting in even more movement.
To summarise, Ball identified the developments and improvements that had taken place over the contract with regard to ground treatment as:
As a final word of caution, Ball said that in the UK London clay had been host to most of the compensation schemes and was generally very tolerant to this technique. Other areas and soils may not be as tolerant and success may be limited. He also said that vertical settlements may not always be the most appropriate for potential structural damage. Regular injection of water helps maintain the TAM sleeves and this ensures the most flexibility in grout injection, if all sleeves are available, but probably the most important aspect was full co-operation between the tunnelling and grouting gangs.
Ball then handed over to Rob James to deal with the construction and vibration issues at Harrods.
James said that with the high profile nature of working for one of the best known shopkeepers in the land, it needed to be a first class job with no shoddy or chipped goods. The work had also to be absolutely secure.
Access was very restricted, with the goods received area being the only entrance to the store available to the contractor. A site gantry was established in tight conditions at the very busy corner of Basil Street and Hans Road. Traffic management was of the utmost importance, with deliveries and muck away closely controlled.
The hoarding around the gantry was required to be both noise and dust proof to avoid nuisance, and a satellite site for burning, cutting and welding operations was set up some 32km away, to minimise these operations on site.
Access to the main works was via the central shafts, which had been sunk under an earlier contract by Gallagher. Of the four shafts, only one was available to the tunnelling operations, the remaining three being for the compensation grouting teams. The available shaft had an area of 2.7m² through which all equipment, workers, muck and segments had to pass. Access to the shaft top was by rail with monorail loading and unloading facilities.
From the shaft bottom, a heading was driven with a break-up into the 3.66m internal diameter bolted segmental pilot tunnel that was to form the parallel connection tunnel between the bottom of the shafts and the access tunnel. This tunnel would be later enlarged to 7.1m internal diameter in cast iron to form the loading area before distribution of goods up the shafts to the store above. All this initial work was completed using hand-mining techniques.
Mucking out and segment supply to the pilot tunnels and enlargements was by Bobcat loaders as the sharp changes in gradients ruled out rails and skips.
From the end of the straight 3.66m diameter pilot tunnel, a second 2.74m internal diameter pilot tunnel was driven around a 94° bend using special tapered inserts between the straight rings to form the very sharp curve. From the end of the curve, the 2.74m diameter tunnel was enlarged up to 5m diameter to the lift shafts and then the 3.66m diameter original pilot tunnel was enlarged to its final size of 7.1m internal diameter in cast iron using traditional roller bolts and winches.
The connections to the other lift shaft were then completed using “Prince of Wales feathers” in cast iron as supports prior to the breakouts. Standard lintel and jamb supports provided the permanent opening supports.
After completion of the 7.1m enlargement, the 2.74m pilot tunnel was resumed towards the KCC building, with final enlargement to 5m diameter following on behind using the very successful, purpose designed, built and tested erector nicknamed the London Eye. James remarked that he was glad they had chosen that name and not the Dome!
The tunnels were finished off with a concrete invert, a final sheeted floor and painted walls. The M&E fit out was followed by crash barriers along each side of the tunnels to assist any less than expert fork-lift truck drivers, James commented.
Moving on to the problems associated with the use of hand tunnelling and James’ question: “Is there a future for handwork tunnelling in the UK?”
The real problem was HAVS, caused by exposure to excessive vibration, the most common form being vibration white finger (VWF). Recent compensation claims had seen £500M paid out to sufferers, particularly in the mining industry.
The guidance note HS(G)88 gives the following procedures for the control of HAVS:
The A(8) exposure level was set by the HSE after research into HAVS, calculated on the basis that regular eight-hour exposure gave a 10% risk of developing HAVS in a 10 year period. The higher the regular vibration exposure level, the lower the exposure period has to be, to keep to within the 10% risk level.
The choice of lower vibration tools was of prime importance but what was the correct level of vibration applicable to the use in tunnelling? The manufacturers published figures under the standard ISO 8662-5 test but this is a vertical test in a workshop with ball bearings which does not represent what happens in practice.
James said that at Harrods the strategy was to select low vibration tools, ignore the effects of anti-vibration gloves, make provision for job rotation and provide a regular maintenance schedule for the tools. He also said that training and education for management supervisors and operatives was implemented together with proper health surveillance. To minimise the use of hand tools, an excavator with a rotary cutting head was supplied.
The health surveillance programme involved a pre-employment medical and regular six week check up. But a pre-employment VWF questionnaire relied very much on the honesty of the miners who were well aware of what the employment implications were of being diagnosed with VWF.
As to how the exposure time was measured, James said that two members of each gang were selected and timed with stopwatches for their daily period of “trigger time”. Vibration levels were measured and each machine given its own “vibration signature”. These were fed into exposure tables and a daily record of vibration exposure built up for each of the workers selected. An average daily exposure was then established for the gang.
A full time HAVS site engineer was employed on site equipped with a EPM VIS-015 vibration meter and a stopwatch for the trigger times.
In summary, it was found that that the use of the excavator was very limited with a serious man/machine interface problem in the handwork. The machine was not compatible with the handwork support system used, introduced fumes to the works, created a fire risk and its size did not lend itself to the variation in section size from 2.7m to 7.1m diameter. As the machine had to be almost totally dismantled to get it down the shaft it was not practicable to keep changing. The conclusion was that it created potentially greater risks.
The operator exposure data was illustrated on slides by James and the method of calculating individual exposure levels from trigger times and machine. From this data, prediction sheets could be made up for the next tunnelling section.
The illustration given was that for the 5m diameter enlargement where a prediction of a vibration exposure level was given as 4.45m/s² against the actual recorded average of 4.31m/s². This illustrated that it was possible to manage the HAVS issue and still perform handwork.
James, however, considered that tool manufacturers should do a lot more. The machine vibration signature should reflect the life-cycle of the machine and its usage. More research should be given to developing a fit for purpose, low vibration tool and addressing the reliability issues. The necessary cost of increased levels of maintenance should also be recognised.
In conclusion, James said that job rotation was essential, operative training and a health passport would be essential in the future and buying the best machines and maintaining them was imperative. With regard to designers, he said they had to recognise the consequences of HAVS and design the works accordingly. Contract data had to be compiled and published so that everyone was aware of what had been done and where the problems were. Industry needed to develop guidelines and good working practices. There was no easy solution to the HAVS problem.
