Return to Archive -by date - by topic - 2001 Archive.

INFORMED SOURCES April 2001

Great Heck highlights double standards

Can differing attitudes to road and rail safety be reconciled?

Let there be no dubiety. The fatal collision at Great Heck was not the ‘freak accident' claimed by the Association of Train Operating Companies. Even less was it, as various commentators asserted claimed, an act of God.

As in most accidents there were precursors. In the last three years there have been 30 cases of road vehicles ending up on railway tracks. But because it was road vehicles going out of control or having accidents, and because the road vehicle had come off worse when collision with a train ensued, this was not seen as a major risk, compared with, say, SPADs and not seen as a risk at all by those responsible for road safety .

If Great Heck was a ‘freak accident' then the term should apply equally to train drivers who, statistically, will SPAD once every 17 years). But we know it doesn't. The railway has zero tolerance of SPADs and the causes of SPADs.

Contrast this with the restrained media treatment of the Land Rover driver at Great Heck, whose remorse one cannot start to imagine. One senses the unspoken ‘there but for the grace of God go all of us motorists'.

Complacency

Contrast it even more with the Highways Agency which has 589 bridges crossing railways of which 230 are motorway bridges. When I asked the Agency, a week after the accident, what they had done about the risk of cars leaving the motorway at similar over bridges I was told ‘We are waiting for the full report from the Transport Research Laboratory (TRL)'.

But surely there had been some form of Quantified Risk Analysis (QRA), to see whether any other bridges would give a similar clear run to a railway line if a vehicle left the road before the crash barrier?

And the spokesman didn't see where I was coming from. So I explained about Ladbroke Grove, where all high risk signals were subsequently risk assessed, or Hatfield, where speed restrictions were imposed.

Still no engagement. No recognition of risk. No action until the full TRL report was received in two weeks time.

Not so much complacency as incomprehension. So I gave up.

But it's a fair bet, that if the railways were responsible for motorway bridges over main lines, the inside lane would be coned off before each bridge and a speed limit imposed until a QRA had been completed.

Road conditions

Tiny details highlight the differences inroad and rail safety. The Highways Agency had the motorway bridge and its approaches checked by a TRL investigation team which didn't get to site until 1430 hours. The derailment happened at 06.12 and the first HMRI inspector was on site at 09.24.

TRL concluded that the that the road surface was salty at the time the car left the road and that there was no ice or snow present on the motorway. While nearly a week after the accident the ‘surfacing history' at this site was still unknown, visual inspections indicated that the motorway surface was in generally good condition and this was confirmed by skid resistance and other tests. The TRL believes that road surface was probably not a factor.

Barrier length

Since the car went round the safety fence the main issue is the length of fence before the bridge parapet.

Under the current safety fence standard (TD19/85) there should be at least 30 metres of full height safety fence in advance of an obstruction, in this case the bridge parapet.

British Standard BS6779 covers parapet design and Part 3 the particular type of parapet on the M65 bridge at Little Heck. However, this standard does not refer to lengths of safety fencing in advance of the parapet.

Another document, BE5, is the technical memorandum referring to parapet requirements. This document is consistent with TD19/85 in that it also refers to a minimum requirement of a 30 metre length of safety fence in advance of the parapet.

Contrast this mélange with Railway Group Standards.

In fact, the M62 bridge over the ECML has 33.5 metres of full height safety fence in advance of the parapet, plus an additional 9.2 metres where the fence slopes down to its anchorage point.

So the safety fence was to specification as far as length was concerned. But for the whole of the front of the Land Rover to have engaged with full height safety fencing at the point where it left the motorway, the length of the barrier would have had to be more than doubled.

Apart from which, the TRL notes that the combination of a Land Rover County towing a double axle trailer unit carrying a Renault Savanna estate car is outside the specification for this type of safety fence. So, concludes the TRL, which writes in weirdly stilted prose. ‘we cannot be certain at this point in our investigations that the safety fence would have restrained the vehicle, even if it had fully engaged with it'.

But, ‘notwithstanding the above, our preliminary opinion is that any engagement between Land Rover and safety fence would have been likely to prevent the vehicle reaching the railway line'.

Well, of course it would, because the Land Rover would have hit the barrier at a gentle angle, rather than head on, scrubbing off speed as it graunched along the metal barrier.

But the fact that, the barrier could not contain a consist weighing under 5 tonnes when 44 tonne lorries are now legal has to be of concern.

Note, no mention of vehicle speeds, tyre marks, or the point where it left the motorway. To be fair this is probably down to the police, but it does make rail accident investigation look more rigorous.

Unforeseen

What is really being said is that no one considered the possibility that a vehicle could leave the motorway at a slight angle and drive for over 100 yards in a straight line along an embankment before dropping onto the railway track. I see a strong parallel with Ladbroke Grove where the track layout assumed that any driver passing signal SN109 at danger would realise the error in the 800 years before joining the main line.

As for the HMRI Interim Report it is was thorough as usual. It established quite clearly that there was nothing the railway could have done. But one comment rang alarm bells. ‘Whilst the coal train had in fact departed early, this early departure was permitted under the operating rules' .

What is the point of this? It is relevant only if the freight train had been pathed to avoid passing trains coming the other way in case they derailed. In an accident which was unambiguously due to external factors, this hints at possible blame.

Finally, a brief note on precursors. Over the last three years (1998/99 to 2000/01) there were 30 incidents of motorists losing control of their vehicles and landing on the railway, obstructing the line. If you include vandalism, the number effectively doubles.

As table 2 shows, in very few cases did a train strike the obstruction, but the potential for disaster was clear.

Table 1 Incidents of road vehicles obstructing the line not at a level crossing

Table 2 Incidents of road vehicles obstructing the line not at a level crossing and struck by a train (train running into obstruction )

Loss of control from road over rail bridge (Table 1) were:

The three incidents of collisions (Table 2)

Continues.........