Twenty five years ago, driving a second Dartford road tunnel under the River Thames was one of Britain's most difficult jobs, with engineers having to cope with huge inflows of water. So it might be thought water would be the main challenge facing rail engineers as they begin work on a twin bore crossing just downstream, for the second phase of the high speed Channel Tunnel Rail Link.
But despite a maximum expected water head over the tunnel line of 37m at low tide and groundwater pressures anticipated to be at least 4bar, this is not the main concern for Rail Link Engineering (RLE), CTRL's engineer and project manager, nor for its contracting partner, an Anglo-German joint venture of Hochtief and Murphy.
The problem is flint.
Times have moved on since tunnellers battled in compressed air on the M25 motorway bore and the modern soft ground TBM should now allow high pressures to be contained at the face of the machine. But that brings new problems in dealing with the ground, which is mainly chalk for almost the entire 2,500m long drive .
The chalk which underlies 10m of thick and sulphurously smelly Thames estuary mud, and some 5m to 10m of waterlogged gravels beneath that, would normally be a very good medium for tunnelling. 'It is barely a rock at all, ' says Tony Bermingham, RLE project manager at the site - 'more a hard ground with a strength of 7MPa'.
Unfortunately, the geological reports suggest it has unusually high levels of flint inclusion, perhaps up to 25% in banded layers.
And flint is very hard indeed.
'It is the second hardest naturally occurring substance known, ' says Bermingham 'up to 700MPa plus - just behind diamond.' While this may have been good for Stone Age man and his axe-making, it could wreak havoc with 21st century man and his tools - in this case the rotating cutterhead on a 8.15m diameter Herrenknecht slurry TBM.
It is not getting the football sized flint chunks out of the tunnel face that is the major problem - though this may cause some problems for the picks - but dealing with them afterwards. As it happens, flint is brittle and shatters easily, as any amateur archaeologist can demonstrate.
Pieces small enough to pump through the bentonite spoil removal system should result from pick impact and roller cutter action - the machine has both - but the effect inside the machine, or at the surface spoil separators, is unknown.
'Flint will score steel and so we don't know what will happen to pipework and particularly to the centrifugal pumps, ' says Bermingham.
Past experience is limited, he adds, and a literature search on the problem turned up very little, though there are stories of machine heads worn 'paper thin' by cobbles.
However, fresh from dealing with the 16km of Irlahull and Euerwang tunnels - a hugely difficult job on Germany's high speed rail system - project director Andreas Tauschinger from Hochtief is relatively sanguine.
He believes the thick chalkbentonite slurry mix will cushion much of the abrasion;
discussions with pump manufacturer Warman have led to the most resistant impellers being used.
If wear is heavy, the big Australian made slurry pumps can be replaced. The pipework is equally maintainable.
Even so, at ú30,000 apiece, Tauschinger does not want to replace too many pumps, and the downtime needed to remove the 6t, 2m high units would also prove costly.
There could also be difficulties with the cutterhead tools.
Even though these have automated monitoring, they may need physical inspection or changing.
The TBM has a built-in decompression chamber in case divers have to enter the slurry head chamber.
But Bermingham adds that it is not intended that divers would be used at the full pressure. 'The chalk fissures should be closed up at the deepest point, ' he adds.
Tauschinger may be confident but he is far from complacent. 'I don't need to lose too much time on an already tight contract, ' he says. Hochtief/Murphy has just 18 months - until June 2003 - to make the two tunnel bores. A ú38,000 ($55,000) daily overrun penalty figure is 'burned into my brain', he adds.
Hochtief/Murphy has one big advantage. Despite the relatively short drive length, the CTRL contract specified that two machines would be used - one for each of the the twin bores.
More accurately, two machine heads will be used with a single back-up train brought back through the tunnel when the first drive is completed and used again.
First drive begins in the summer on the south side where a 100m long starter pit is being excavated between heavily instrumented diaphragm walls, as part of a 700m long cut and cover extension and approach to the tunnel. Machine delivery and assembly begins this month.
A 400m long cut and cover is also to be made on the north side as part of the overall $190M contract. Amec/Spie has a $30M subcontract for all the diaphragm walling.