33 trains an hour – but why not 40?

I recently made my acquaintance with a 1910 working timetable for the Charing Cross, Euston and Hampstead Railway, now part of the Northern Line. You may recall that the line then ran from Charing Cross to Camden Town where it split into a branch to Golders Green and another to Highgate (now Archway). 1910 was a great year to study as it coincided with the first operation of some non-stop trains while also delivering a 40 trains per hour (tph) service along the trunk section, a level of service that is probably unachievable today.

If you wanted to make a timetable ‘interesting’ then a good way of doing it is to have trains of lots of different lengths and then couple some (but only some) trains together at busy times. Another way to spice things up is to specify different train lengths on each branch service so that trains on one service cannot work on the other. So that’s what they did.

Trains of 2, 3, 4 and 5 cars were in operation at various times. On the whole, all off-peak trains were 2-cars (using 20 trains) and this produced a 4-minute service. In the peaks, the plan tended towards a 3-minute service on each branch, providing a 1½ minute service along the core (an intensity of 40 trains an hour). Generally the Highgate service employed 5-car trains and the Golders Green service used 3-car trains. Interesting: in later years the Highgate branch was said to be the lighter loaded, but that was probably after the Edgware extension opened.

This straightforward peak hour arrangement was disrupted by the desire to run some non stop trains on the Golders Green route (which were 4-car sets). This was far from simple. The running time from Golders Green to Charing Cross was just 19½ minutes for an ordinary train, but the non stops omitted stations Hampstead to Euston (inclusive) and Goodge Street, so accomplished the journey in 16 minutes. The problem was that this acceleration was greater than two train headways, but trains cannot overtake. That means a path has to be found in the service for this train to pass by uninterrupted. At Golders Green it was despatched just two minutes in front of an ordinary train so it ran into a gap, and at Camden Town the Highgate service was slightly held back so it continued to run into a gap, arriving at Charing Cross just behind a train from Highgate. Whether the dislocation of a regular service was worth a 3½-minute faster journey time on the lighter loaded branch is anyone’s guess, but these trains operated for some years.

The service intensity hinged entirely around slickness of operation at Charing Cross, where there were only two platforms and turn around time was often only a minute. For the timetable to work, trains must not be delayed outside Charing Cross, so, ideally, the route will be set and the home signal clear before the approaching driver sees it. The train must then run in and clear the pointwork. Only then can the train already in the other platform pull out, after all the gatemen have signalled that the gates are shut and it is safe to proceed. The train must then pull out and not until the rear car has cleared the crossover can the route be set up for another train to arrive. The crossovers were subject to speed limits, which didn’t help. Meanwhile the train had to unload passengers through narrow end gates, and reload. The driver had to fully apply the brake, ‘shut down’ and alight. Then the same (or a different driver) would board the other end and ‘open up’, recharging the air brake (which took a few seconds) and be ready to move, all in a minute. Anecdotally, it appears all this was achieved, and that was in 1910. 40 trains an hour.

It is in that light that I have been taking a very keen interest in the Victoria Line and the ambition to run 33 trains an hour, something found quite impossible to achieve on the Central Line, though that was the plan.

The Victoria Line that I knew and for quite a while worked on was a pig of a thing. It was being pushed well beyond the expectations set for it when it was built and was running a 2 – 2½ minute service, or about 27 trains an hour, in theory. This was more than it was designed for and only the infiltration of 1972-stock cars off the Northern Line allowed the fleet to be topped up to provide sufficient trains (24 trains an hour was the most originally contemplated). The signalling and automatic train control systems did not work well with such intensity. The slightest irregularity would cause a train following too closely to just catch a 270 (restricted speed) code instead of a full speed code, and condemn it to run at controlled speed into following platform, perhaps losing it 10-20 seconds, depending on location. This cost capacity and reduced effective throughput. The track layout was hugely inflexible, especially at Seven Sisters, and any significant delay would cause mayhem with the service as the controllers and crew managers battled to sort things out. The railway was working at its limits, so when something went wrong it had a disproportionately big impact. The service either ran well, or was dreadful. It was hard to be in between.

In that light I was suspicious of the claim that a 33 tph train service had been modelled with the new trains and signalling. My suspicions were around crew management and slickness of operation at Brixton, which has the same station layout now as Charing Cross had in 1910. Any delay would check an approaching train, and that would cause checking back and reduce the headway – the test of this would be late running after the peak services which would be the sum of all the headways not met.

The new 33 tph timetable came into use in January, quite properly without too much of a fanfare (in case it didn’t work). I have spotted some late running after the evening peak (only about seven minutes on a ‘quiet’ day, at worst) and decided to investigate further. Last week, one day at 18:00, I boarded a train at Victoria southbound, and observed it to be 1½ minutes late, which is immaterial. We had a clear run into Brixton (a good sign), the train was ready to depart within one minute, and actually departed in under two (the instant the next train berthed in adjacent platform), and it departed having made up a few seconds time. The crews were where they should have been, and I saw no additional supervision providing ‘encouragement’, as was once the case. I returned to Victoria and noted a few more trains and all appeared to be pretty much all on time (and I didn’t spot any gaps in the scheduled service, which is a tell tale for a timetable not working).

It is perfectly clear that the high rate of acceleration and braking, and (as far as I can see) the ability of trains to operate at near full speed over the Brixton crossovers (not, I think, possible under the old system), have much contributed to the slickness. This is aided by the new signalling that reduces the safety distances that have to be provided because the trains work more ‘real time’, compared with the old arrangements. The timetable is thoughtfully arranged too, with some additional time booked at Stockwell and Vauxhall northbound so that the effects of slightly ropey working at Brixton are discreetly masked whilst maintaining good regularity at Victoria. The new train control system works to better than a quarter minute and can be applied at any station unlike the old system with was half a minute accurate and only applied at a few locations and was by comparison very crude indeed.

My conclusion is that with everything working perfectly, a full 33tph service is operable. I remain suspicious that the shortcomings of the line’s design may mean that if there are delays then a 33tph service may create more problems than it solves, but if technical reliability can be maintained then I reckon that by the summer holidays everyone will have made up their mind. Personally, I hope it all works out.

The Railway Engineer for May 2013 has a useful note on all this. 33 tph means a headway of 109 seconds, so a train must arrive, turn round, depart and clear the station area in 218 seconds, including alteration in lie of the points. It is stated that in practice this can be done in no less than 203 seconds, leaving some seconds in hand, but not all that many. 34 tph might be worth going for at this site but it sounds as though getting more would be expensive. It should be possible at some other locations though. I wonder if we’ll get to 40 again, like we did 103 years ago. If not, it might be worth investigating closely why not. We must keep pushing.