As i say, i worked on the high level section.  The low level was
built - and the train ran - proof of concept - but the civil
engineering of the high level was a disaster.

The track consisted of precast concrete beams about 6' deep by 3'
wide by 60' long (again - excuse me if the dimensions arent right -
its a long time ago! - the beams may have weighed 60 tons and been
some other length!). The supports - at 60' intervals - were narrow
columns about 1' wide and some 10' at the base tapering to 3' at the
top.  Two beams were designed to rest on these columns on two
bearing pads each, about 4" wide by 1' long.  The beam was hollow
and a clean run from end to end, the concrete being around 4" think
all the way round (though it was profiled for the metal plates of
the track on top).  These beams were to be installed by a launching
gantry which had its nose resting on the next pillar and picked a
beam up, put it in place and moved on.

When I came on site (at Sutton Gault) as one of the Contractor's
engineers, the place was at a standstill, since after installing the
first two beams (with a crane) and the launching gantry, and
installing the third beam, as the gantry moved onto it, the end of
the third beam broke and the whole thing crashed (30') to the
ground.

As I remember and was told (and bear in mind I was only a junior and
didnt get to high level meetings!) the designers insisted that the
beam was strong enough, but instructed us to install massive
concrete diaphragms in each end anyway: these were about a foot
thick, with just enough hole to crawl though in the middle, and had
enough reinforcement to anchor a battleship!  

Of course - we were told that this massive reinforcement was not
really needed - but strangely enough when we started putting up
beams again, they didnt fall down - at least not for that reason!

One of the requirements was that the line and level of the track had
to be extremely accurate: as I remember a tolerance of 1/8" in line
and 1/16" in level was required on each 60' beam.  Fine - not really
a problem - we has special shim plates that would take up minor
inaccuracies at the bearings, and the pillars were in any case
mounted on two huge bolts set into the base block so they could be
very accurately levelled.  These tolerances were no more or less
than i'd been used to when tunnelling, and so didnt present much of
a problem, except for trying to work with  a theodolite on a 3' wide
concrete beam nearly 40' above the fens in mid winter with snow
coming horizontally at 40mph....  Anyway we managed to get a number
of beams in place OK....

Until... we checked the line and level of beams which had been
placed previously, and found that they were inches out of line and
level!

It soon became clear that the bases (the pile-caps), which were
mounted on two concrete piles sunk deep into the fens, were tilting
all over the place under load.  The /design/ was for the whole
structure from the base of the piles to the top of the pile-cap to
be a rigid element, with the pillars and beams (which were anchored
at the free end) placed on top on the screws which formed a pin
joint allowing for thermal expnsion and contraction.  The original
ground surveys had apparantly identified a layer of hard sand at
some depth (60' is in mind but I could well be wrong) or sufficient
quality and depth to bear the load.

What exactly was happening at the bottom of the piles was never
established I believe - although the indications were that either
the sand wasnt strong enough, or it was too thin, or at a different
depth - or plain not there in places - either way, the piles (of
which more later!) started to sink at different rates and everything
tilted alarmingly in all directions!

At which point our managing director called a halt (for the second
time) - much to our relief - since lots of tons of concrete waving
gently in the breeze above ones head makes for a less than perfect
workplace!

Lots of meetings, discussions and recriminations later it was
decided to test each pile cap with massive loads of concrete to see
what happened.  We did, and with monotonous regularity, they failed
by sinkng a lot mor than was tolerable, and tilting at the same
time.

While the bosses were working out what to do next, we carried on
testing.  Now about some way west of Sutton Gault the embankment on
the southern Bedford River has been breached at some time, and there
is a great horseshoe shaped dent in the emebankment, which we had to
cross.  There wee three pile caps in there I think - it was very
muddy and squishy underfoot, and not an easy place to work.   We
moved the test loads from one pile to another with a crane that sat
in the between two pile caps and swung the loading blocks around
from one to the next.  

One Sunday we'd loaded a pile cap, and it had duly sunk a bit and
twisted a bit - nothing new, anyway - and on monday morning I was
standing on the bank taking some measurements while my 3 man gang
was preparing to move the loading blocks onto the next pile.  I'd
turned my back for a minute - and heard a sort of ruxhing sound and
a large economy-sized splash, and looked round to see no more pile
of loading blocks, and three very frightened men who happened to
have been working on the side it *didnt* fall over onto... a very
lucky escape for us all (because I'd have been the one in court if
they'd been hurt).

about an hour later our MD posed the simple question to the
designers - "Can you guarantee the safety of my men?".... and the
job stopped - never to resume.  

We spent much of the spring and summer trying all sorts of remedial
measures, none of which were very successful, and the project was
finally cut off from government funding since it became obvious that
only a complete redesign and reconstruction would serve.

Among other things we extracted on of the piles which had failed in
the collapse - with some interesting results.  Not only had it
broken in the middle, some 30' dowm into the fen, but we discovered
that there was no reinforcement in part of it, and that the upper
and lower sections were not joined properly.  The company who'd done
the piling had (I think) gone bust by then - I'd better not name
them - but it was wryly amusing that part of their name was
"economic" - with quality control as it appeared...

And so the whole thing foundered on problems with the design and
implementation of the Civil Engineering, work - and Eric sold the
system overseas after having proved the concept and as usual the
Brits missed out because they didnt have enough courage to
acknowledge a mistake and put in sufficient money to rectify it and
capture a serious world market.

An interesting and happy year in my life - it could have been a
marvellous thing to be associated with - but as it was it ended as a
single line in my CV.

The first time I've told the tale in full - I've seen no references
in the technical literature - and most of the parties concerned are
now out of business anyway - buried and forgotton in the mud of the
fens.