Trouble in the Heathrow Tunnel
Whilst the thought of the phrase ‘explosion on Crossrail’ appearing in print probably gave Crossrail’s engineers (or at least their Press Office) a few sleepless nights, the problem really worth watching seems to have been emerging further down the line for TfL.
For months, Howard Smith has been telling the TfL Board that the biggest concern is the signalling in the Heathrow tunnel. Part of the concern was the lack of engineering possessions available for testing the tunnel-based European Train Control System (ETCS) signalling equipment.
ETCS is a European signalling specification for the safe operation of trains. In fact it is used throughout the world, especially in China. The current National Rail standard, Train Protection and Warning System (TPWS), is very safe. TPWS is largely responsible for the incredible safety record achieved in recent years. It is not, however, 100% fail-safe in all circumstances and does not always protect against human error. For example, it won’t always prevent a buffer stop collision at a terminal – not the sort of thing you want to happen on a terminal platform in an underground tunnel at, say, Terminal 4.
The problem is that whilst very reliable once working properly, ETCS is known to be a fickle beast. Like all electronic signalling systems, it is very susceptible to hostile electrical environments. Unfortunately, it seems that a hostile electrical environment is
exactly the sort of thing one finds in a round bored tunnel with a 25kV, 50Hz AC catenary running alongside it, with the added complication of a similar but different signalling system already present. Which is an accurate description of the Heathrow tunnels.
To make matters worse, the ETCS system installed at Heathrow is not exactly the ‘out-of-the-box’ version typically installed on a brand new railway today. Heathrow has the bolt-on edition tacked onto an existing signalling system (ATP) and ETCS was not around when it (ATP) was first conceived. The new track to train signalling being put in place for Crossrail also still needs to work in conjunction with existing trains that know nothing about ETCS and which were built in the days when regulations concerning electromagnetic interference were a world away from today’s tight, rigorously-enforced standards.
It seems this has indeed caused issues. Howard Smith’s comment to the TfL Board perhaps sums the current situation perfectly:
"It performed as expected as in functionally it works, but the reliability needs a lot of work before we’d be happy with it as a passenger service."
Worryingly, the impression given is that things now work when adverse factors (such as other trains) are isolated out, but they just don’t in the real-world environment. These means that even if these initial tests had worked perfectly, there would have still been a long way to go.
