Crossrail: Timetable for Success?

354 comments

Sufficient information is now available, unofficially, so that we can be fairly sure of the exact service pattern now proposed for Crossrail in December 2019. Furthermore, we can have a good guess at how it may develop in the coming years. The good news is that the service pattern is better than we were generally led to expect. The bad news is that… well… incredibly, for once, there doesn’t seem to be any bad news.

Getting it right, early

In a sane world, any plan for a significant new timetable for a complex railway system is in the detailed planning stage well over a year before its introduction. If, for one group of services, a clockface interval is required then that really needs to be in place first so that any other services can work around it. The specification of the new service must be clear and consistent from the outset – even a minor ‘tweak’ later on may well have ramifications. To that end, it is vital that the calling pattern of the services is decided well in advance and known to be workable. In the case of Crossrail’s December 2019 timetable, less than 16 months away, we are really already at the point where the service pattern needs to be finalised.

It is important to emphasise from the outset that previous suggestions as to the proposed calling pattern were never really intended to be the final plan. At all times there was an indicative calling pattern – a plan that was believed to be workable, but also one that it was hoped could be improved upon.

The peak: mostly the same

To be clear, there is nothing really dramatic here. For the most part, nothing has changed. The proposed December Crossrail 2019 service is unaltered east of Paddington. The peak service will consist of 12tph (trains per hour) between Paddington and Shenfield and also 12tph between Paddington and Abbey Wood. All the peak Abbey Wood terminators will start from, or continue to, destinations west of Paddington, and none of the services to or from Shenfield will go west beyond Paddington. There will be an additional 4tph service between Liverpool St (high level) and Gidea Park in the peak direction only.

Hanwell: the big Crossrail beneficiary

Airport services (shown in blue) will all call at Hanwell

Hanwell having only 2tph has always been a bit of an anomaly. Without a competing Tube service, or any other station nearby for that matter, it is strange that the service from this station has been so poor. Even today it still has only 2tph despite now being served by TfL Rail rather than Great Western Railway. It was promised 4tph once Crossrail was fully opened but will actually have 6tph from December 2019. Not only will all trains to and from Terminal 4 call there, those to and from Terminal 5 will now call there as well.

Whilst this is a bit unexpected, with hindsight it shouldn’t have been. TfL (and the future Elizabeth line’s operational director, Howard Smith) strongly believe in turn-up-and-go in the London metro area. 4tph isn’t considered good enough these days. It should be at least 6tph and ideally there should be no need to wait for longer than 10 minutes.

Note that, unfortunately, 6tph at Hanwell does not mean a train every 10 minutes and a worse case scenario of just missing a train could mean a 15 minute wait. Nevertheless, this is a significant improvement.

Whilst 6tph, effectively an even 4tph with an additional two trains, might not seem that much of a further improvement, one has to look to the west not just to the east. Hanwell will have a direct service to Terminal 5, the biggest terminal at Heathrow. Whilst this will be of limited benefit to the occasional Hanwell local wishing to fly out from Heathrow, it could make a significant difference to airport workers living in the area.

West Ealing is the key?

Most Elizabeth line trains will call at West Ealing

Hanwell may see a minor improvement compared to what was originally planned on the airport routes but where it gets more interesting is at West Ealing which will have 10tph in the peak whereas it was expected to have just 4tph.

We can only speculate on why there has been such a dramatically improved service proposed at West Ealing but we think we have a good idea. Similarly to Hanwell, there was probably already a desire for at least 6tph at West Ealing in the peak timetable.

TfL could probably also see development going on around almost all Crossrail stations and the increase in demand that the new construction would create. This is happening at many locations in the suburbs on Crossrail’s western arm. Less so at Hanwell which retains its 1930s suburban character, but more so at most other stations which already have a proposed 10tph service.

What we believe, admittedly without any evidence, was the absolute clincher for going to 10tph at West Ealing is that once you go to 6tph you hit a new problem. The previous proposals dealt with the issue of fitting so many trains into (effectively) a two track section of railway by having trains call at Hanwell or West Ealing or neither but never both. The trouble is with 10tph (ignoring the peak-only Reading semi-fasts for the moment, we will look at those later) and 6tph calling at each station, you have to have some trains calling at both stations. And once you have trains calling at both stations it makes a lot of sense to have all the trains calling at West Ealing to even out the service pattern.

Acton Main Line: the poor relation

Still only 4tph for Acton Main Line

The improvements at Hanwell are not matched at Acton Main Line. Nevertheless, Acton Main Line will see an improvement as its service will increase from a paltry 2tph to 4tph. It will still be the only Elizabeth line station in the London area (excluding the special cases of Heathrow Terminal 4 and Terminal 5) that will not have at least 6tph. One imagines that TfL just couldn’t bring themselves to offer a better service at Acton Main Line knowing that the station would never be popular owing to the number of Central line Tube stations nearby.

(Almost) All-stations airport services

Only airport services are shown

All previous plans suggested a complex mix of stopping and non-stopping airport services on the Elizabeth line and this was partly down to train pathing and partly down to some stations having less than 6tph. Now the rule is fairly simple: all services from Terminal 4 and Terminal 5 stop at all stations except that trains to or from Terminal 5 do not stop at Acton Main Line. This is much more in keeping with TfL’s desire to keep things easily understood by passengers.

According to the Crossrail website, the Elizabeth line will take 30 minutes to get to either Terminal 4 or Terminal 5 from Paddington. In the case of Terminal 4, that is a five minute improvement on today’s timing even though there will be an additional stop at Acton Main Line. Not only will the frequency improve, the perception will be one of a fast, frequent service to central London – at least compared to what was on offer in the past.

The Reading semi-fasts

Peak services to Maidenhead and Reading

We now look at the Reading semi-fast peak services which were always going to be the most interesting, and least known, aspect of any final plans.

Just to recap, the plan in the last couple of years was always to have 4tph to and from Reading in the peaks and an additional 2tph terminating at Maidenhead. The trains starting at Maidenhead were all stations except for Hanwell and Acton Main Line. The trains from Reading (4tph) were to be two pairs of complementary services that both called at all the major stations (such as Slough) and shared a combination of stops at lightly used stations beyond London and busier stations in the London suburbs.

The latest proposal, which is expected to make it to the final timetable, is for the Maidenhead terminators to be unaltered but that the 4tph Reading services now are paired, with one pair being all stations (except Hanwell and Acton Main Line) and the other one being a limited-stop service calling only at Twyford, Maidenhead, Slough, West Drayton and Ealing Broadway.

This new semi-fast service will now provide a real incentive for passengers from Reading to Crossrail stations in central London to travel direct using a single Elizabeth line train rather than change at Paddington. The extra time (around 20 minutes) is offset by avoiding the need to change at Paddington and having an almost guaranteed seat for the entire journey. This must play very well with TfL’s finances.

A further real benefit of the Reading semi-fast service is that those on the Henley-on-Thames and Marlow branch lines who lost their occasional direct service to Paddington will have a better comparable service (admittedly always having to change).

It might be desirable to reduce the number of stops further on the semi-fasts but it is hard to see how it is possible. They need to stop at Twyford to provide a minimal 4tph service, Maidenhead and Slough are too big to ignore and they need to stop at West Drayton for that station to get its promised 6tph service. One could argue that it really doesn’t need to stop at Ealing Broadway but the suspicion is that if it did not stop, it would not fit into the timetable and so either depart from Reading so late that it delays the train behind it (from Terminal 4) or arrive too early at Royal Oak portal for its slot through the central Crossrail section.

Off peak services

Off-peak services west of Paddington

Off peak services will be very similar to peak services on the Elizabeth line. The service to Shenfield and Abbey Wood goes down slightly from 12tph to 10tph (giving 20tph in the central section). The Reading fast service won’t run. We believe the calling pattern will be identical to peak hours. Again, TfL’s philosophy of keeping things simple for the passenger would suggest that this is the case.

We presume there will be minor timing adjustments between Maidenhead and West Drayton to even out the service but this may be limited due to the need to keep freight paths available. If the Reading and Maidenhead trains have the same calling pattern then obviously with a 10tph off-peak service west of Paddington one cannot achieve an even-interval 15 minute 4tph service.

A consequence of at 20tph off-peak service through central London is that, simplistically, in the off-peak trains from a given originating station west of Paddington will alternate between serving Shenfield and Abbey Wood unlike in the peak 24tph period. In fact it is believed that some ‘tweaking’ is possible so that off-peak more trains that originate west of Paddington serve Abbey Wood than Shenfield. In particular it is believed that the 2tph from Terminal 5 will both go to Abbey Wood.

The off-peak GWR anomaly

2tph GWR service – off-peak only

A very strange anomaly with the new proposed Elizabeth line service is that in the off-peak the Elizabeth line loses its Reading semi-fast service, but this will be replaced by a 2tph GWR service calling at, wait for it, Reading, Twyford, Maidenhead, Slough, Hayes & Harlington and Ealing Broadway before terminating at Paddington. In other words, if it called at West Drayton instead of Hayes & Harlington, an identical calling pattern to the Crossrail peak semi-fasts.

Presumably the omission of West Drayton is because 6tph off-peak is thought less important than providing at connection at Hayes & Harlington for Heathrow. This though doesn’t explain why it is a GWR service.

There are three plausible reasons for the apparent anomaly and it is probably a combination of these that will lead to this oddity:

  • Removing the semi-fast service from Crossrail (so 10tph west of Paddington) makes it easier for the Elizabeth line to have a consistent 20tph in the centre of London
  • It is better for GWR which can make better use of its electric rolling stock. As such this may have been a sweetener to aid GWR’s co-operation of losing long distance commuting traffic between stations east of Reading and London
  • It is better for some passengers west of Reading because the semi-fast GWR service will actually start from Didcot

A less rational reason could be that the Secretary of State could not quite bring himself to hand over all intermediate services on the main line (as opposed to branches) between Reading and Paddington to TfL.

Downsides?

Unlike most timetable patterns, it is hard to see much desire to tinker with the December 2019 Crossrail timetable. Possibly, for consistency, it might be nice to have 6tph at Acton Main Line.

Having only 4tph at West Drayton in the off-peak seems a little unsatisfactory and contrary to TfL’s desire for at least 6tph yet it is hard to see how this can be resolved. One could have the GWR trains call at West Drayton instead of, or as well as, Hayes & Harlington but this creates a further issue because there is a freight loop between Iver and West Drayton. So it is desirable for passenger trains not to stop at West Drayton so that they can overtake a slow moving freight train.

Equally it is a little unsatisfactory that a peak-hour journey between Reading and Heathrow involves either travelling via Ealing Broadway (or Paddington) or taking the 2tph all stations train to Hayes & Harlington.

What next for the timetable pattern?

The next big thing for the Elizabeth line is to try to get 4tph to Terminal 5. This is known to be an aspiration and would make a lot of sense. For one thing Terminal 5 (2tph Elizabeth line) is much busier than Terminal 4 (4tph). For another, it would give an even interval service of 8tph for airport services at all stations from Terminals 2&3 to Ealing Broadway.

A 4tph service to Terminal 5 would also make better use of 2tph by avoiding having them terminating at Westbourne Park sidings, which involves the rather unsatisfactory situation of ensuring a westbound train at Paddington has not got any overcarried passengers.

The challenge with 4tph to Terminal 5 would not be getting the trains to Terminal 5. You could have a nice consistent pattern of Terminal 4/ Terminal 5/ Reading (or Maidenhead) almost-all-stations train repeating itself four times an hour. This would be almost essential because Terminal 5 will be shared with Heathrow Express which will also be running 4tph, at 15 minute intervals, to Terminal 5.

The challenge would be what to do with the Reading semi-fasts which have to somehow squeeze in between two trains 5 minutes apart at Paddington, so that they can take over the path of a Shenfield train starting at Westbourne Park. With no obvious opportunity for any additional loop at any station east of Hayes & Harlington, it looks like something would have to give. If something does have to give then introducing 4tph to Terminal 5 could involve options such as:

  • Extend 2tph from Shenfield that are planned to terminate at Paddington to Terminal 5. This would lead to slightly inconsistent intervals in the timetable to Terminal 5 which could cause problems
  • Add London suburban stations stops to the semi-fast Reading trains to even out the pathing of Crossrail trains. This would enable the Reading semi-fasts, which would therefore in future go to Shenfield, to be squeezed in between the regular 12tph to Abbey Wood. This would significantly reduce benefits to westernmost part of Crossrail
  • Increase the number of Crossrail trains per hour in the central section to 28tph which would mean that there would be 14tph west of Paddington. Unfortunately, it would be hard to match up 4tph even-interval services west of Paddington with 28tph in the centre – but presumably could be done with very judicious use of station calling patterns and scheduling additional running time for certain trains

The importance of planning

The problem of running 4tph to Terminal 5 in future does highlight why it is really important to plan a timetable years in advance. If 4tph to Terminal 5 were the intention from the outset then, no doubt, provision would have been made for suitable infrastructure on the ground to make such a service possible. Even if it made the project unaffordable, passive provision could have been made. The omission is understandable though with Terminal 5 not built when Crossrail was planned. So having Crossrail serving Terminal 5 was not even on the speculation list.

Old Oak Common to the rescue?

Proposed Old Oak Common services

It is clear that Crossrail will already have more services west of Paddington than originally planned and there are aspirations for at least an additional 2tph. One can envisage further simple enhancements in the future if the capacity and necessary reversing sidings or platforms become available. But, the more services there are, the more challenging it will be for them to arrive at the right time at the Royal Oak portal outside Paddington. The opening of Old Oak Common Elizabeth Line platforms should help considerably with train regulation, as both the up and down GWR relief lines will be allocated an island platform each (similar to the Charing Cross platforms 6-9 at London Bridge).

A further benefit for some future Elizabeth line users is that it is planned that all GWR fast services will stop at Old Oak Common. This will eliminate an inconvenient change at Paddington for passengers from the west country who want to use the Elizabeth line to continue their journey. Although in the reverse direction, if they have sufficient time, they may wish to continue to change at Paddington to increase their chances of getting a seat.

There are clearly benefits to the operation of the Elizabeth line if there are platforms available at Old Oak Common and the obvious question that it raises is whether these could open in advance of the HS2 station. Less obvious is that, for developers of the site, it is the Elizabeth line not HS2 that they see as the desirable investment influence that the station brings.

If pressure were brought to bear to ensure the Elizabeth line platforms opened early at Old Oak Common (and independently of the opening of HS2) then the benefits would not only be local as, rather like a capacitor in an electrical circuit, Old Oak Common platforms would help ensure an even flow of trains to London. One potential problem in such a proposal is that the platforms are funded by HS2 as they are an essential part of the project, but HS2 itself has nothing to gain by opening them early. Indeed, it will add to the complexity of construction work.

Still signalling issues

There is still uncertainty as to whether the legacy signalling issues in the Heathrow tunnels will ever be resolved so it might well be December 2019 before we see any proper Elizabeth line (class 345) trains at Heathrow, due to the need to remove all existing Heathrow Express trains that rely on the obsolescent GW-ATP signalling in the Heathrow tunnels. In fact even this December 2019 date could be in doubt.

If the existing signalling issues cannot be resolved then the legacy signalling needs to be removed. This cannot be done until the current Heathrow Express rolling stock is replaced by refurbished class 387 trains. This, very conveniently, is due to happen in December 2019, according to a First Group press release, but the class 332 trains are becoming problematic and expensive to maintain so, one suspects, that Heathrow Express Ltd will shed no tears if plans can be brought forward.

Given the soul-searching caused by the recent May timetable change one does wonder if the December 2019 timetable change will become the January 2020 timetable, change which would give Network Rail a bit of extra time to sort out the tunnel signalling – assuming of course that the Heathrow Express rolling stock can be replaced within this timescale.

More questions

The December 2019 Crossrail timetable is becoming clearer but there are still more questions. Is it still planned to run late night/early morning services between Heathrow and Paddington (high level)? What exactly is meant by the peak-period? Are we talking about a high-peak period of an hour or a full three hour peak service? Will there really be such a good service on the western arm of Crossrail on a Sunday morning or will engineering work considerations mean that a reduced service will operate? Indications and expectations are that Sunday morning Elizabeth Line travellers will be disappointed with the service frequency on offer.

We are getting closer to really understand what exactly a full service on the Elizabeth line will mean but there are still a lot of unknowns, and the fare structure west of West Drayton has not yet been revealed.

Complete December 2019 peak service. For completeness the off-peak service can be found here

.

Thanks to ngh for all the information provided and for creating the master diagrams. Also to Jonathan Roberts and Moosealot.

Written by Pedantic of Purley