None of today’s Underground lines is a good example of optimal track layout for the train services required today. The Victoria line was close for a while. This was designed as a complete entity to deliver a particular train service, but, even here, a half-century after it was built, train service patterns have changed in a way that was not expected.
Other lines are in a far worse position with track layouts sometimes dating to the steam age. Modern requirements are also frustrated by the removal of flexibility a few decades ago, at a time when traffic levels were falling. The result today is a largely arbitrary collection of reversing points, crossovers and sidings that no competent planner or operator would dream of installing if they were building an equivalent line today.
What one really wants is something along the following:
- Sufficient capacity at terminal points to reverse the service required and provide a suitable amount of recovery time.
- Intermediate terminating points where full capacity is not required to the end of the line and arranged so not to delay through trains. These can also be used to reduce late running and relieve congestion by turning trains short.
- Emergency reversing points in order (1) to enable emergencies to be handled and avoid trains with passengers on them getting trapped, and (2) to provide reversing points during engineering work to minimise the length of track where a service must be withdrawn.
The Metropolitan Line affords an example of the problem. It also provides an example of a problem that has been recognised over many years and has got so near to resolution without anything happening!
The problem is the railway between Edgware Road and Aldgate where today two factors conspired to make the life of the operator difficult. The first is the succession of busy flat junctions where a train passing through one route conflicts with an alternative route. Perfect timetabling would schedule what is known as parallel pathing at flat junctions, where conflicts and hence delays are avoided. This is easy where there are only odd junctions but it is unachievable where there is a succession. Lots of stand time and an optimistic outlook on behalf service operator works better than might be expected, but as service levels increase the inappropriate layouts become more obvious and ultimately limit capacity. The second factor is that once the train leaves Baker Street eastbound it is effectively committed to go all the way to Aldgate and back, as turning short Moorgate (now the only place available) is usually an act of desperation as it interrupts service in the other direction and ultimately saves so little time that it isn’t worth the bother. This makes it more difficult to recover from delays. Let us look at how the world might very nearly have been.
The problems at Baker street are threefold. First there is a very awkward flat junction, with the turnout on a sharp curve, so train speeds are low (and therefore trains occupy the junction for a long time and further reduce capacity). Secondly it is not feasible to reverse H&C or District trains east to west here, requiring trains to turn at Edgware Road when passengers really want to get to Baker Street (and there is only a 5-minute service between the two when better intervals are really needed). Thirdly, about a third of the Metropolitan main line service must turn round at Baker Street, but with the terminating platforms at either side of the through lines reversing trains obstruct other movements.
The Met Railway became perfectly aware how unsatisfactory the arrangements were as soon as the main line rush-hour services were extended through to the City in 1909/12. When the station was entirely reconstructed between 1910 and 1913 provision was made for a low-level platform in its own tunnel, more or less following the line of platform 2, but underneath. At the south end, the line was aimed so as to continue in a curve beneath the Circle Line and rise to the surface a few hundred yards further east where it would diverge from the westbound line. At the north end it would rise to the surface in the covered way. By this means the conflicting movements would be avoided. Most of this tunnel (more correctly a covered way) was actually constructed, though the route to the upper Bakerloo escalators installed in 1939 now pierce the route.
London Transport was quick to decide that the Baker Street junction was not fit for purpose and obtained powers in their 1935 Act to rearrange things. A flyunder was to be built diving under the Circle Line but it would have a low level platform on the site of the existing Platform 1, using some of the 1912 works. However the widening under the Marylebone Road would be much more extensive and would provide width for a reversing siding, enabling trains from Edgware Road to turn round without fouling movements to and from platforms 2 and 3. This would allow the District’s Edgware Road reversing trains to proceed to Baker Street though I do not have the precise service details. In addition, platform 4 would become the main through platform for City trains, reducing fouling movements north of the station. This work would have been immensely useful and would probably allow today’s proposals for 32 trains an hour around the top of the Circle to be improved upon. Unfortunately nothing was done.
London Transport’s view was that traffic between Baker Street and the important interchange at Kings Cross deserved off peak trains but the City (to and from which traffic was dead outside the peaks) did not. As the existing station was not well sited, a relocated station with reversing facilities appeared to be called for and the present station location was arranged with two widely-spaced side platforms and a central reversing bay, with a platform either side, for Kings Cross reversers and long enough for 8-car trains. War broke out and although the track was rearranged and the central bay was built, and track laid, the signalling had not been installed when heavy bombing destroyed much of the old station and the new one was brought into use early. The signalling never was finished and after the war priorities changed. For some years there was just this dead track in place, but in 1957 the east end was nibbled off to enlarge the concourse, later much more enlarged and now there is no trace of the original plan. For many years it would have been a useful point to have a short working terminal available but traffic is so busy now that it would be rather a luxury and the space is better used as a concourse. Be it noted that a new crossover west of the station means it is now a viable emergency reversing point, but not for turning round a late running service.
Moorgate station has an unusually complicated history. Before 1925 the only reversing facility available to Metropolitan trains was a bay road and siding on the north side of the through lines. Arrangements were then made to electrify part of the Widened Lines (today’s Thameslink tracks) so that City-bound Metropolitan trains could use the Widened Lines to Moorgate to reverse and return on the Met tracks at either Moorgate or Farringdon. To facilitate this, the station was rearranged to give Met trains access to two platforms on the south side of the through lines. In later years this pair of platforms was disconnected from the widened lines, resulting in the two Metropolitan reversing bays still there today. These are theoretically useful for short-reversing trains but the signalling requires incoming trains to be brought to a stand at the home signals before the route can be set, after which trains can creep into the platforms at 10mph, meanwhile blocking the westbound line. This so much reduces the advantages of using these bays at all that they are seldom used for reversing except in the direst emergency.
In 1989 London Underground investigated a proposal to install a short connection at the east end of the station that would convert platform 3 into a through line, leaving platform 2 available for reversing trains without any conflicting movements. This would have provided a lasting solution to the need for a turning point west of Moorgate. It was not, however, pursued and the construction of the new Crossrail ticket hall would seem to make such a link in the future impossible. It would have been easy and cheap to do this during the 1960s reconstruction, but with service levels falling I don’t suppose anyone thought this important.
The working at Aldgate is so intensive that the slightest hitch means that trains can block back to Kings Cross, or even further west, and some kind of safety valve where trains can turn would be useful but hard for the financial wizards to justify against competing schemes to provide new facilities. Nevertheless, if along the north side of the Circle an opportunity arose for an intermediate reversing point, it would be useful to install one.
For many years Liverpool Street had a bay road on the south side, by tradition used by Amersham and Chesham trains. Again, the problem here was that eastbound arrivals interfered with the service on the westbound line. During the early 1980s, Metropolitan Line management felt that extending these trains to and from Aldgate should be tried in order to improve the running. This was found satisfactory and the bay road fell out of use and is now beyond putting back.
What is still possible
The new crossover west of Kings Cross is useful for reducing impact of engineering work but cannot realistically be used for short-working trains in a long queue. The only space left now appears to be the largely open area between Farringdon and Barbican, some of which are earmarked for use as siding space. A possible arrangement is shown below and involves re-routing the westbound line along the old widened lines route from east of Barbican (through one of the screen walls) and along the old ‘down’ track. In view of the Crossrail works it is preferable to use a widened central platform rather than the narrow platform on the south. There is space across the old siding area to reconnect with the existing line east of Farringdon station. The old route would be used to form a new central bay road. There is space for stabling sidings either side of the rerouted line (and into the old Moorgate platforms), if required. Easy to do now. Very difficult if any more development takes place.