On Wed, 8 Jun 2005, Clive D. W. Feather wrote:

In article , Tom Anderson
writes

Also, i'm not sure what was happening with trains further back than
Lancaster Gate, but it seems to me that it would have made sense to
bring the last train in the queue back to the station as soon as the
problem was detected.

Reversing a train isn't trivial. Apart from the issue of the driver
having to walk through a crowded train,

Aha, i'd not thought of that!

you lose signal protection because the line isn't set up for
bidirectional working. So you need to *ensure* that the next train isn't
going to move even if it gets a green signal.

In the case of the Central, reversing also means driving in Restricted
Manual, because there are no codes for that direction. Stopping the next
train is also harder, because there may be no signal between it and the
station.

Overall, it could easily take 15 to 20 minutes to get a train reversed
to the previous station.

I see. Are you saying that they actually did what i suggested, and it's
just that this took a long time to do, or that what i suggested is a bad
idea?

If the latter, i don't (yet!) agree with you:if the line ahead is blocked,
then the train would otherwise only spend those 15 to 20 minutes sitting
in a tunnel, so you might as well start pulling it back.

How long does it take to switch from reversing to going forward again? If
it's a long time, that would matter, since deciding to reverse the train
would mean a significant delay if the blockage was cleared quickly. If
it's not a long time, then you might as well start reversing the train,
then send it forward again if the chance arises.

Interesting that both this and the H&C problem we were going over a week
ago are situations where the impact of failures was amplified by
shortcomings in signalling. Not that the signalling is broken, but if it
had been better - if there had been overlap protection on the H&C, and if
the Central line was signalled for reversible use - the effect of the
mechanical failure could have been contained far better. Of course, there
are powerful historical and economic factors shaping LU's signalling - i'm
not accusing anyone of incompetence - but it drives home how important
this stuff is to running a reliable, high-frequency railway. I hope LU
have a serious plan for transitioning the entire network to a signalling
system that isn't basically out of the 19th century, and that can handle
this sort of thing more smoothly.

tom

--
SAWING CHASING CRUNCHING ROBOTIC DEMOLITION