Longer trains for Crossrail?

Any mention of overcrowding on the Elizabeth line tends to elicit either the suggestion that TfL should run more trains or that they lengthen the trains which are currently nine carriages long. If the latter, some people are quick to point out that the Crossrail platforms on the newly built central operating section (Paddington to Abbey Wood) were designed to be able to accommodate 11 carriage trains. Hypothetical 11-car trains would not currently be able to use the stations below ground level because the platform edge doors currently installed only cater for 9-car trains. Moreover, at some underground stations, a portion of the platform tunnel is not currently in passenger use and is blocked off. The blocked off portion has not been fitted out for passenger use and may even be used for something else.

The very wise policy invoked during the planning of Crossrail to provide for longer trains on the central operating section did not extent to the rest of the project. It was argued, with a lot of justification, that catering from the outset for longer trains over the whole network did not make financial sense because there was little or no ultimate cost savings to be made by doing so. It was claimed that ‘passive provision’ was made where this was practical but that didn’t necessarily mean there wouldn’t be challenges.

A question we would like to consider is whether lengthening trains likely to happen any time soon. We do not know the answer but attempt to provide as much background information as we can so that readers can have an informed idea as to whether this is something that will happen – or not.

More Trains v Longer Trains

The issue of whether it is better to run more trains or longer trains to cater for an increase in demand is a complex one with no easy answer. It also depends on whether you ask Network Rail, TfL, or a train operating company.

As a very rough guideline, TfL’s policy is generally to run more trains when possible, rather than lengthen trains. This also tended to apply to TfL’s predecessors and has historically been largely down to the cost of retrospectively lengthening platforms at underground stations. An exception to this rough and ready guideline tends to be when introducing new rolling stock as this is the ideal time to consider extending the length of trains.

A good example of using the introduction of new stock to take advantage of the opportunity to make the trains longer is the introduction of ‘S’ stock on the sub-surfaces lines where, by means of maximising the length available for public access to platforms and using selective door operation (SDO), major tunnel reconstruction was avoided. Maximising the effective platform length was largely achieved by elimination of platform-based monitors used by the driver to check that it was safe to close the passenger doors.

Similarly, almost unnoticed by passengers, the replacement Victoria line stock was slightly longer than its predecessor. This was to take advantage of a situation where the platforms were built unusually long for 8-car trains because the technology available at the time of building the Victoria line could not accurately stop an automatically driven train in the platforms. Hence the platforms were slightly longer than they would otherwise needed to have been for a new equivalent length train.

In a situation more analogous to Crossrail, Jubilee line trains were extended from 6-cars to 7-cars in December 2005. Despite the obvious parallel, this is not a clear-cut example. An extra carriage was ordered for the 59 existing trains and, additionally, four new 7-car trains were ordered (so 28 additional carriages). Consequently, almost one-third of the order was actually for the new trains.

The Crossrail Case

We have no idea of what costs are involved to lengthen trains on Crossrail but it should be borne in mind that one should also take into account the timescale of the benefit when considering lengthening trains. A newly-purchased (or leased) extra carriage typically lasts as long as the current rolling stock is in use. However, a lengthened platform is lengthened forever.

One way of looking at potential benefits of lengthening trains on Crossrail is to look at the ‘carriage benefit’ achieved. Each new train provides an extra nine carriages if the train length remains unchanged. Each new carriage on lengthened trains will provide a benefit of 80 carriages if one takes into account the 10 new trains on order as well as the existing 70 trains. Of course, a number of carriages will be out of use for maintenance but the portion should affect both figures roughly equally.

Eighty carriages are roughly the equivalent of nine 9-car trains. So, each extra carriage added provides the rough equivalent of nine new trains. A factor to be taken into account is that purchase of extra carriages does not involve purchase of carriages with driving cabs. Driving cabs nowadays tend to be much more expensive because of the onboard technology (including signalling technology) needed to support them – and each train has two driving cabs. Furthermore, trains require drivers. If you have longer trains, not more trains, you don’t need to worry about the costs and logistics of employing extra drivers.

A factor for preferring more trains in preference to longer trains is that off-peak services can be better tailored to requirements and extra carriages are not needlessly in service on trains. This is particularly true where the peak/off-peak split of passengers is reported to be around 56% / 44% on Crossrail rather than the 50% / 50% split more typical of Underground lines.

The option of increasing the number of trains is irrelevant if you are already running the maximum number of trains that can be realistically run. Longer trains achieved by increasing the train length by one carriage might then come out as more expensive but, in the case of Crossrail, one could look at it as providing an 11% increase of capacity to a railway that cost around £22 billion to build (excluding rolling stock) for a cost that should be a small fraction of that original infrastructure cost.

Running Out of Options

In many ways, it doesn’t matter what one’s preference is. When you have two options and have proceeded as far as you can with one option, then the other option is the only one left. The central operating section of Crossrail is signalled for 31tph and that is probably as far as modern technology can currently take us when running main line trains on what is basically a Tube service underground. If Package East is fully implemented, as we expect it to be by the end of 2026, that will account for 28tph in the central operating section and there has to be some flexibility in the system in order to recover from delays.

Theoretically, 30tph is possible on the central operation section of Crossrail and has been talked about by TfL but it gives the operator precious little flexibility to recover from delays. Also, from a timetabling point of view, Crossrail runs on 15-minute cycles meaning a tph figure that is a multiple of four is desirable. Currently the peak service is 24tph and the off-peak service is 16tph with early Sunday morning and late Sunday evening being an 8tph service.

A further consideration that most train operators do not need to consider is that in the Crossrail tunnels there is a rule enforced by the signalling system that states that only a maximum of two trains can be in a ventilation section. Obviously, more trains put a strain on this restriction, but longer trains don’t.

There is a third option of reducing demand by building a new railway line to take pressure off the Elizabeth line. In an era where money is tight and even a relatively short extension of the Bakerloo line to Lewisham cannot be afforded, this does not seem a feasible option. Nor is it one that can be achieved in any reasonable timescale.

Finally, a potential fourth option is to suppress demand by using the fare structure to discourage passengers. Such a policy would be an anathema to the current mayor and probably any successors.

Going for 10-Car Trains rather than 11

Although some provision has been made for 11-car trains on Crossrail, we would argue it makes sense to initially go for 10-car trains if one is to lengthen the trains at all. There are three main reasons for this – all of them largely financial.

With a railway carriage costing in the order of £2m these days, extending 80 trains by just one carriage would cost in the order of £160m just for the extra carriages which is a lot of money for a cash-strapped organisation. The finance people would probably not want money paid up front for an 11^th carriage or the additional infrastructure consequences, until such time as one could see a return on investment in the form of bums on seats – or, more realistically in the case of the Elizabeth line, the extra carriage being comfortably busy with a high proportion of standing passengers.

The second financial reason is that, unlike building underground stations with longer platforms from the outset to cater for future growth, there is very little cost saving, if any, in building the above ground infrastructure in place for possible future use on an existing railway.

There is a third reason to go for 10-car trains rather than 11-car trains. At many locations, as we shall see, it looks quite easy to cater for 10-car trains but difficult, if not impossible, to cater for 11-car trains. So, it does seem to make sense to initially introduce 10-car trains and then, after this has been done, consider the implications of introducing 11-car trains should there be a demand to justify considering them.

A 9-car train on Crossrail is 205 metres long. A 10-car trains would be 227 metres long and an 11-car train would be 249 metres long. As there is no need for the rear of the train beyond the final door to be in the platform, the required platform length is slightly less. However, it appears that on the Shenfield branch, at any rate, platforms were generally extended to 211 metres where that was possible prior to Crossrail being introduced.

Can and Should it be Done?

The rest of this article looks at what is involved in extending Elizabeth line trains to 10-cars. To some extent, when it comes to platform lengthening, we can only speculate but we will try and present to you enough information for you to form your own opinion.

We cannot give an opinion on whether lengthening trains is worth doing because we do not have the necessary financial information. We can only suggest that we think it is technically possible.

We will look at the issues we believe to be involved. The longest section looks at platform lengthening outside the central operating section but this doesn’t necessarily mean it is the most critical issue.

Extending the Depot for Longer Trains

When considering lengthening trains, it is often hard to know where to start but often the biggest issue is that of overnight storage at the depot. London Overground found it challenging when extending trains on the East London line from 4-cars to 5-cars. The stabling sidings at their only depot were a little over 12 carriages long so previously one siding could take three trains but with 5-car trains it could only accommodate two. Consequently, a site had to be found to accommodate one third of the fleet and it wasn’t easy to find one though Silwood sidings was eventually made available.

SouthEastern Railway has wanted to more run 12-car trains for many years but is prevented from doing so largely because of the lack of depot space for the longer trains and the impossibility of obtaining land to expand Slade Green depot for anything like a reasonable cost.

We mentioned earlier replacing the original Victoria line trains with new longer rolling stock. The trains could probably have been even slightly longer still if it wasn’t for the restricted space available at the Victoria line’s Northumberland Park depot.

The main depot for Crossrail is at Old Oak Common. This consists of a covered shed for nine sidings and uncovered sidings for around 34 trains. If the covered shed needs lengthening there looks as if there is space to do so though it might involve the diversion of a minor road. This might be one of few occasions where it could make sense to extend the covered depot to house 11-car trains from the outset. Extra covered space in a depot is generally welcome even if it is not initially needed for longer trains.

It looks as if most of the exposed sidings have enough space at the ‘throat’ end to accommodate a 10-car train. If needed, a road could be diverted at the buffer stop end to provide the necessary extra length of siding.

Ability to procure the Extra Carriages

Probably the next consideration is ensuring that the extra carriages can be supplied for the longer trains. This would not appear to be a problem. The Alstom factory in Derby (previously the Bombardier factory) still exists. Depending how complex an extra carriage needs to be, it may be possible for other manufacturers to supply them.

An important feature would be whether the bogies (wheelsets) need to be motored on the extra carriage. Currently eight of the nine carriages are motored. If the extra carriage could be a trailer (unmotored) car then this would make the carriage cheaper and easier to build. With an additional trailer car, the maximum draw on power would be only slightly increased but there might be a small reduction in rate of acceleration achievable by the train.

It is presumed that the extra carriages can be supplied at a reasonable cost.

An Adequate Power Supply

Regardless of whether extending trains or providing more of them, the ability to supply electricity to the overhead line (or rigid overhead power rail) or third rail can be a critical factor when considering a service improvement. Fortunately, in the case of Crossrail, over most of the route the power supply is either new, has been recently replaced, or is due for renewal. It would be very disappointing if there wasn’t a significant margin for growth built into the present power supply and the extra demand created by an extra carriage should be only minimal.

Signalling

A big unknown is signalling. Presumably the automatic train operation (ATO) software in the central operating section will need to assume all trains are 10 carriages long. The stopping location at the nine through stations in the ATO area would need to be changed on one of the platforms at each of these nine stations.

There don’t appear to be any critical junctions or other places where a stationary 10-car would foul another route. On the Network Rail portion of the Elizabeth line, signalling generally caters for any length of train so, hopefully, there isn’t an issue there.

Platform Edge Doors

There are platform edge doors at eight stations. Clearly extra doors will be needed for the extra carriage. Implementation would be an issue to investigate. It is probably impractical to implement a ‘big bang’ approach to converting 9-car trains to 10-car trains on a railway timetabled to require 80 trains (including trains undergoing maintenance). A means of phased implementation would be necessary. This means that there would have to be a fail-safe way of ensuring that platform edge doors did not open if there were not a carriage on the other side.

Sidings

It would be critical that any sidings would need be long enough to either take 10-car trains or be capable of extension to 10-car trains. An overhead view using Google Maps suggests sidings at Plumstead and Westbourne Park already long enough. It looks as if the little-used Chadwell Heath turnback siding and the sidings at Shenfield could be extended to accommodate the extra carriage.

The important depot at Gidea Park has sidings that can currently accommodate two trains each and it looks as if they could each accommodate one extra carriage – which clearly isn’t quite enough. It therefore looks unlikely they would be able to accommodate two 11-car trains without moving the pointwork at the eastern end further to the east. This appears to be feasible. The train wash at the country end of the sidings may not be long enough once track modifications are made but it appears that it would be easy to extend this under the A127 Southend Arterial Road which is on an embankment at this location. This may involve slight repositioning to avoid any critical support structures to the adjacent footbridge as well as punching a hole through the road embankment.

In the case of Maidenhead, it looks like use of a small plot of land that appears to currently be part of the station car park would be sufficient to extend these sidings.

Probably, the most challenging sidings are those at Ilford. This is a much-in-demand site used by Greater Anglia, London Overground and Crossrail as well as housing a depot run by Alstom, the train manufacturer, used for maintenance and refurbishment projects. The site was remodelled in 2016 in order to accommodate the class 345 train for Crossrail. It looks very much as if it was remodelled with 9-car Crossrail trains in mind.

It appears that all the Crossrail sidings at Ilford could be extended for the loss of one London Overground siding. It may just be possible to reconfigure the Crossrail portion of the site to squeeze in 10-car trains without resorting to losing space elsewhere. More probably, a rethink may be required as to the best way to cater for longer trains, should this happen. London Overground would probably welcome extra sidings there and, maybe, Greater Anglia could be persuaded to store more of its trains out in Essex. Alternatively or additionally, TfL could look for an extra site for sidings for Crossrail rather than continually attempting to squeeze more trains into the sites they already have.

Stations

The factor which most people seem to think is the main issue, but probably is the one of the least important issues, is the ability to extend station platforms to accommodate the longer trains. Of course, on Crossrail we have a huge advantage here because the critical and busy underground platforms in central London are already long enough. Indeed, at the three truly double-ended underground stations (Bond Street, Tottenham Court Road, and Liverpool Street) longer trains may actually assist passenger flow within the station.

In addition to the underground stations on the new part of the railway, there are above-ground stations at Custom House and Abbey Wood. At Custom House the platforms are only 9-cars long but there is clearly space to extend for another two carriages in the direction of the Connaught Tunnel whilst at Abbey Wood the Crossrail platforms have already been built to a full 11-car length.

We cannot generalise on the problems involved with lengthening platforms, or choosing not to, so they really have to be considered individually. The newly-built underground stations are known to have long enough platforms for 11-car trains. Those at Heathrow which existed before there were plans to serve them with Crossrail trains are more of an issue.

Above ground, Google satellite view can be very helpful in determining how easy it would be to extend the platforms – especially if one can view a train in the station. Unfortunately, some of the satellite views are quite old so it is important to check that the train is a 9-car train and not a 7-car one. In the following text, when discussing future possible platform extensions, there are numerous links to the Google Satellite view so the user can have an indication of the practicality and issues involved in extending platforms at a particular station.

It should be possible to determine existing platform lengths because they are listed in documents known as the Timetable Planning Rules on the Network Rail website. Each document is a large PDF. The one for Anglia Region (AR) appears to be up-to-date but the one for Western and Wales (WW) is clearly hopelessly out-of-date in this respect and individual entries cannot be relied on.

Existing Short Platform Issues

Before looking at potential short platform issues on the Elizabeth line, it makes sense to look at existing issues. Fortunately, a 2022 Freedom of Information (FOI) Request has a response which has details of all the current short platforms (except one) on the Elizabeth line and the reason for them. We have assumed all the information in this FOI is still as accurate as it was when published. This appears to be the case although there are minor discrepancies when compared to a current poster showing the restrictions and one station appears to have been omitted.

STATION	SHORT PLATFORM	REAR DOORS LOCKED OUT
Twyford	4	4
Maidenhead	1	4
Taplow	ALL	3
Burnham	BOTH	3
Langley	1 and 2	5
Iver	ALL	4
Hayes & Harlington	5	1
Hanwell	ALL	8
Paddington (National Rail)	NONE	6 (centre doors only due to curve)
Maryland	ALL	6
Forest Gate	ALL	5
Seven Kings	ALL	5

The GPS Issue with the SDO System

Possibly the most significant part of the FOI response is the following sentence:

As the SDO system works off GPS, one short platform at a station means that the doors are locked out on all of the platforms at the station.

The statement could have been worded better. It would be more accurate to state that one short platform that Elizabeth line trains are capable of calling at … Fairly obviously, the short terminating platform at Slough for the shuttle train to Windsor & Eton Central is no reason to lock doors on Elizabeth line trains calling at Slough. And there is bound to be a short platform somewhere at Stratford station.

Of course, there wouldn’t be a problem if one could guarantee that Elizabeth line trains would never call at platforms other than the designated Elizabeth line platforms. However, there is always the rare occurrence of the need to divert to other platforms due to either ‘operating difficulties’ or planned engineering work. Consequently, any platform where there was any possibility of an Elizabeth line train stopping at needs to be taken into account.

This above sentence from the FOI response has significant consequences and not all are immediately obvious. The first major consequence of this is that at stations where it was known from the outset that one of the platforms could not be lengthened, the decision was taken not to lengthen any of the platforms. This means that platforms that could have easily have been lengthened (and undoubtably would otherwise have been) were not lengthened.

The above situation is even worse when it is considered that generally it was the London-bound platform that caused the need to have SDO. Having a platform not long enough for London-bound passengers isn’t that critical. You can’t board a train from a platform that isn’t there. However, on the return journey, passengers could find themselves sitting in a carriage which doesn’t open at the station. Apart from inconveniencing passengers and possibly causing mild panic, this potentially delays the train.

Another consequence of relying on GPS-dependent SDO is that where there are platforms on both fast (main) and slow lines you really need to ensure that all the platforms are extended to the required length – even those platforms on the fast lines will be rarely used.

On the line from Paddington to Reading (exclusive) all stations except Burnham, Hanwell, West Ealing, and Acton Main Line have at least four platforms. It appears that platforms west of West Drayton have not been lengthened as part of Crossrail construction except for the two Crossrail platforms at Slough where the platforms on the fast lines were already long enough.

East of Stratford a policy of extending all platforms, where necessary, was generally undertaken. However, if extension of one platform at a station proved problematic then none of the platforms were extended.

SDO being Needlessly GPS Constrained

To add to the frustration, this GPS issue would seem to be completely avoidable and doesn’t happen with other train operating companies. It seems incredible that a train with so much software running it cannot determine which platform it is at either from the signalling system or from having additional data fed into the Passenger Information System, which is used to update the passenger information displays present in every carriage.

The solution to achieving SDO down to individual platforms used by other train operating companies is to have a balise in the track that ‘strikes in’ each set of doors as the train enters the platform. On exiting the platform there is another balise that ‘strikes out’ each set of doors as it continues beyond the platform. That way only the correct doors are opened even if the train does not stop in exactly the correct position. Regardless of whether trains were to be made longer or not, it would appear to be a good solution to implement on the Elizabeth line which would enable more platforms to be able to fully accommodate existing Elizabeth line trains.

There is an obvious caveat to the seemingly simple solution of not having a totally GPS dependent SDO system. If it were that easy to do then, surely, wouldn’t TfL have implemented it by now? This seems to suggest that either there is a technical issue or its implementation would be much more costly than one would imagine.

For the purpose of the remainder of the article we will assume that this GPS issue will not be fixed and so the shortest platform at a station determines the number of doors that can be opened on all platforms. We will also presume that it is essential that all doors can be opened at the Heathrow underground stations.

At present there is the situation at Maidenhead where a short platform on the down main line (rarely used by any trains and normally only by Elizabeth trains in the event of engineering works) means that the rear four doors are locked out on the two platforms that Elizabeth line trains regularly use and also on the platform that these trains occasionally use (platform 5).

At Hayes & Harlington it is the terminating platform (platform 5) that has a restriction and the rear set of doors is unavailable due to the platform length restriction. This is presumably the rear set on arrival of a terminating train. It is baffling why the new platform isn’t quite long enough, given it could have been, and the consequences for all Elizabeth line trains using this station.

It should be noted that in each situation where short platforms are present, it is the rear doors that are not opened. This is different from London Underground and DLR policy where, if appropriate, a combination of front and rear doors are locked out which reduces the distance passengers on the train may be required to walk to reach an opening door. Such a policy on Network Rail might be unpopular because it is the usual practice to site station starter signals at the end of the platform.

A Look at Stations with Existing Short Platforms

It is instructive to see why there are currently short platforms on Crossrail. This list is taken from the Freedom of Information response previously referenced. The relevant stations are listed in west-to-east order.

Reading Branch

Twyford has a short platform on the up relief line. It cannot be extended at the London end due to a large arched brick road-over-rail bridge which would cost millions of pounds to replace. It cannot be extended at the country (Reading) end due to the points leading to the branch line being located there. The points could be replaced by a west facing connection to the branch line through the car park but it would cost a lot of money and it would be a tight curve. Passenger usage of Twyford is relatively light and many of those passengers would use a semi-fast GWR train which does fit in the platform.

As already stated, Maidenhead has a restriction due to the limitations of the GPS aspect of the SDO system. This is confirmed by the declared lengths of platforms in the Timetable Planning Rules. However the Google satellite image, which is a few years old, appears to contradict this.

Taplow, Burnham, Langley and Iver are all lightly used stations on Crossrail which are outside London. There is space to extend the platforms but it seems that an economic decision was made not to do so. At Langley the FOI document claims that only platforms 1 and 2 are short platforms whereas both the timetable planning rules and Google satellite view suggest that they are all the same length.

Hayes & Harlington, as already stated, just seems to be due to an aberration when designing the station track layout leading to the dreaded GPS issue kicking in.

Hanwell is an unusual case. The station does not have operational platforms on the fast (main) lines. Originally the two operational platforms were going to be lengthened but, during the passage of the Crossrail bill through the House of Lords, local resident petitioned against the scheme because they feared passengers would look into their back gardens from the stations. At the time automatic failsafe SDO was in its infancy. Once Crossrail decided to adopt SDO they agreed to provide ‘Letters of Comfort’ to the residents assuring them that the platforms would not be extended so that the neighbours would drop their petition. One suspects that the saving in costs at what was then a very lightly used station may have been a factor in this.

Shenfield Branch

Maryland had been problematic from the outset. A road-over-rail bridge at each end of the platforms makes a platform extension impossible without major reconstruction involving at least one of these bridges.

Earlier Crossrail schemes specifically excluded Maryland from Crossrail due to its short platforms which would be unable to accommodate the proposed Crossrail trains. It was supposedly largely down to the protests against Maryland not planned to be on Crossrail that the Crossrail team decided to adopt SDO in order to overcome this problem.

Forest Gate and Seven Kings appear to have an issue with the need to acquire land for platform lengthening in the London direction. As these stations were in the original Crossrail plans, it seems that once SDO was adopted it was decided to not to bother with obtaining the previously required land.

Also having short platforms, but omitted from the FOI list, is Manor Park where the up slow (‘electric’) platform could not have been extended due to lifts at the country end and a freight loop line behind the platform preventing extension of the platform at the London end. As well as that the FOI response states that at Forest Gate and Seven Kings the rear five doors will not open, whereas the more recent poster shows six rear doors not opening.

Lengthening The Platforms – Heathrow Branch

If there is one branch likely to be problematic when in comes to lengthening trains on Crossrail then it is the Heathrow one. Not only would platform extensions involve underground tunnelling, it would involve doing it under an airport – and airports get extremely sensitive about people tunnelling under runways.

Within Heathrow, the most potentially problematic station is Terminal 4. The Timetable Planning Rules show a usable length of the platforms of 195 metres which is sufficient for the current 9-car trains. Unfortunately, the line between Terminal 4 and Terminals 2&3 is a single tunnel with points located immediately at the London end of the two platforms at Terminal 4 station.

Fortunately, the buffer stops are quite a way beyond the end of the platform and the platform has been blocked off to give it its current usable length. So, it appears that by knocking down a wall and finishing off the extended platform to match the existing platform, an extra carriage could be accommodated. It appears doubtful if an 11^th carriage could be accommodated so this is possibly the biggest reason why lengthening to 11 carriages may not be eventually possible.

On the bright side for the long-term possible future, it does appear from Carto Metro that the Terminal 4 station is well clear of any runway or taxiway above it so it may be possible to extend the platforms, if necessary, at the buffer end without too much protest from Heathrow Airport should that extension be desirable. Moreover, potentially this work could be carried out without interrupting passenger service.

Terminal 5 already has a recognised usable platform length of 217 metres which is sufficient for a 10-car Crossrail train. It looks as if the platform tunnel is sufficiently long for the platform to be extended further if equipment located there could be relocated elsewhere.

Terminals 2&3 station, like Terminal 4 station, has a usable length of 195 metres. Like Terminal 5 there does appear to be an opportunity to extend the platform if necessary. However, unlike the Terminal 4 station, SDO could be used if necessary if some non-GPS method were available to implement it.

Contrary to what many people think, Terminals 2&3 station is not that busy. For example, Slough is busier although there is the issue that alighting passengers at Heathrow tend to have a lot of luggage. Despite that, SDO should be able to easily cope at this station should it be needed.

Lengthening The Platforms – Reading Branch

Reading, a recently rebuilt station, has platforms that are more than ample length for Crossrail.

At Twyford, as already mentioned, there is no opportunity to length the London (up relief) platform. The Reading (down relief) platform could be extended if the GPS issue were fixed.

Maidenhead appears to have the space to extend the Crossrail platforms if the GPS issue were fixed. However, although busy by Reading branch standards, the station has only slightly higher entrance and exit figures than Maryland (the least used station on the Shenfield branch). It is probable that a large portion of users of this station catch a GWR train which is short enough to fit in the platform

At Maidenhead, one could question if platform lengthening would really be needed. Arguably, a better use of the money would be to raise platform heights where necessary to comply with current standards.

As Taplow, Burnham, Langley, and Iver are all lightly used stations which currently have short platforms, it could be argued that SDO is all that is necessary. Platform extensions should be easily possible if desired but these would not necessarily be sufficient to cater for 10-car trains. An argument could be made that the situation shouldn’t be worse than the current situation so maybe the platforms should be extended by at least one carriage length for 10-car trains.

Of these four stations, Burnham is by far the busiest, which isn’t actually saying much. It also only has one island platform that could easily be extended at the country end so maybe a case could be made extending the platform here is as it would be relatively cheap and not problematic to do so.

Slough would probably need to have platform extensions on the Elizabeth line platforms. There is space to do this. The other through platforms (platforms 2 and 3) are already long enough.

At West Drayton platform extensions appear to be possible though it could get tricky if extending to 11 carriages.

Lengthening The Platforms – Suburban West London

Hayes & Harlington – The point at the London end of platforms 4 (London direction) and platform 5 (terminating platform) should ideally be moved slightly closer to London. This would provide space for the necessary platform extension to platform 5. There is space to move the point but a long little-used freight siding might need to be slightly shortened. Why this point wasn’t located fractionally further from the existing platforms in the first place when platform 5 was built as part of Crossrail is a mystery. Platforms on the fast line could be extended to take an extra carriage. Platform 4, surprisingly, is already long enough for 10-car trains.

Southall – There appears to be space for lengthened platforms on the relief (Elizabeth) lines. Platform 1 (down fast) is constrained because of a turnout leading to the sidings there.

Hanwell – Currently, eight pairs of doors are locked out at this station – the most on Crossrail. One suspects that the Office of Rail and Road (ORR) would be unsympathetic to a further three doors being locked out though worse examples exist elsewhere – even within London.

The previously mentioned ‘letters of comfort’ are not generally legally enforceable and are usually used in accountancy – not rail planning. A critical crossover on a freight route at the London end precludes extending in that direction. In principle the London platform at the country end could be lengthened but the railway is on an embankment at this location so it could be expensive. It should be possible to screen off the gardens of disgruntled residents. The foliage is quite tall and thick here so any visual screening should not be unduly intrusive. The Heathrow platform could also be extended at the country end.

West Ealing – a platform extension here should be easy to do at the country end. There is a crossover close to the west end of the down platform which doesn’t prevent the platform being extended but could cause signalling issues. Other options are to extend the down platform at the London end (although the platform would be of substandard width under the bridge), accept SDO at this location and or, as an extreme solution, rebuild the portion of the road bridge over the Crossrail (relief) lines.

Ealing Broadway – if a platform extension were needed, some railway signalling cabinets would need repositioning but there are no apparent major issues in extending at the London end. Google Satellite view appears to show a now withdrawn class 360 (Heathrow Connect) train at the station so the satellite view may not be up-to-date.

When GWR needs to divert trains away from Paddington such as during the construction of Old Oak Common, some GWR trains are expected to terminate there and others call there on their way to Euston via the relief lines and Acton Wells so one would expect the platforms here to be sufficiently long for all eventualities.

Acton Main Line – it appears that an overhead line mast would need moving on the London- bound line and that and extended down platform would be narrower than desirable but it does look feasible.

Old Oak Common (under construction) – one really hopes that the platform will be long enough. As the Elizabeth line platforms would be used for GWR trains during engineering works and possibly in times of disruption one could be fairly confident these will be sufficiently long for even 11-car trains.

Paddington (National Rail) – regular Elizabeth line usage at this station is generally confined to early morning, late evening and Sunday mornings until around 09.15. In addition, platforms 11 and 12 here are used on most days to curtail a late running train to Abbey Wood so that it can pick up its path in the return direction.

All platforms currently in use except platform 14 appear to be stated to be long enough for even 11-car trains but platforms 6 & 7 are contractually reserved for Heathrow Express and class 345 trains (Crossrail trains) are specifically banned from them. It is highly undesirable to use platform 1 as there are no ticket gates there and platform 14 appears to be infrequently used for passenger trains as it only long enough for conventional 8-car trains (7-car Crossrail trains which are no longer in use). There is no platform 13.

For a long while, platforms 11 & 12 have been shortened at the buffer end to allow for repair work to the roof. This work must finish one day and so the platforms would easily be able to handle 10-car Crossrail trains. Use of the full platform would also mean more carriages at the buffer end would be on straight section of platform so obviating the need to close so many centre doors on carriages due to platform curvature.

Lengthening the Platforms – Shenfield Branch

Current platform lengths on the Shenfield Branch

Station	Platform 1	Platform 2	Platform 3	Platform 4
Maryland	168	169	168	167
Forest Gate	173	178	190	198
Manor Park	168	185	194	163
Ilford	226	221	233	227
Seven Kings	180	180	187	187
Goodmayes	212	212	212	212
Chadwell Heath	211	213	214	213
Romford*	211	211	211	211
Gidea Park	211	211	227	228
Harold Wood	209	211	209	211
Brentwood	211	211	211	211

To complete the picture, Liverpool Street (National Rail) platforms 16 and 17 are shown as being 205 metres long in the Train Planning Rules. Stratford platform 5 is 253 metres long and platform 8 is 250 metres long. At Shenfield where platforms 5 and 6 are used for Crossrail terminating trains platform 5 is 245 metres long and platform 6 just 209 metres long.

Liverpool Street (National Rail). This is used in a very similar way to Paddington (High Level). Originally there were four platforms (15 to 18) available to the former 7-car Crossrail trains. Platform 18 was sacrificed to enable platform 15 to take full length trains up to 12 conventional length carriages long and for platforms 16 and 17 to take 9-car Crossrail trains.

The plan was always to make platform 15 available for Greater Anglia trains which currently are 236 metres long – just slightly short of a conventional 12-car train of 240 metres. Currently a few Greater Angl;ia trains use platform 15 on a regular scheduled basis. Although usually unoccupied and therefore available for Crossrail trains, none are scheduled to use it and it is rare to see a Crossrail train at this platform – though it does happen.

Typically, Crossrail trains at Liverpool Street use platform 17 and one is positioned at platform 16 during the day as a ‘hot spare’. In principle the ‘hot spare’ could be positioned elsewhere with plenty of derelict railway land between Liverpool Street and the abandoned platforms at Bethnal Green station.

It seems that platforms 16 and 17 were rebuilt with 9-Car Crossrail trains specifically in mind. The platforms cannot be extended at the country end due to the width restrictions on the approach to Liverpool Street station constrained by retaining walls that probably support the offices built over the station.

However, it is probable that platforms 16 and 17 could be extended by one carriage length into the concourse, subject to what is located underneath the concourse, as the platforms in question are tucked away in a little used part of the main concourse.

Using the concourse would involve introducing a slight curve to the platform 17 in order to avoid critical supporting pillars. There would also need to be a pedestrian route created for rail staff who need access to the far side of platform 17 (the former platform 18).

Alternatively, it could be questioned as to whether Liverpool Street really needs 17 platforms these days and would 16 be sufficient. If so, platform 17 could be sacrificed so that platform 16 can be extended at the country end. This may seem like an unfortunate loss of a further platform but on many lines, including the Great Eastern Main Line, we are approaching the situation where platforms considerably longer than 200 metres are highly desirable.

Stratford station platforms 5 & 8 were built to take 12-car trains as part of the Shenfield Electrification/Central line eastward extension scheme that opened post World War II. All the platforms at Stratford that are capable of being used by Elizabeth line trains are sufficiently long to cater for future lengthened trains.

Maryland is a really trickly station when it comes to extending platforms. The station was built on a very constrained site in a cutting constrained by road-over-rail bridges at each end. Platforms are only long enough for conventional 8-car trains (7-car Crossrail trains) whereas today’s Crossrail trains are 9 carriages long.

The station has seen a dramatic rise in passenger numbers since Covid and, in all probability, that will continue making the station completely unfit for purpose. One of the roads constricting it is Maryland Point which is not that busy and has wide pavements with a small garden area that has seating.

It appears that the bridge could be rebuilt with a single span over all four railway tracks with space for platforms at the side but it would be an expensive job. If done in two halves traffic could be kept flowing. Additionally, a small amount of land would need to be compulsorily purchased from commercial premises to provide space for the London bound platform.

Although rebuilding a bridge just to lengthen the platforms may sound like an extreme solution it is certainly not without precedent. An example from just over a decade ago is Streatham Common station.

Forest Gate could have extended platforms but this would involve compulsory purchase of the rear of the premises of a church (not the church itself) plus the ends of the long gardens of two or three private gardens.

Manor Park platforms could be extended but this would probably require elimination of a freight loop located behind the London bound platform. The freight loop appears to be disused but is still regarded as operational. It sees limited use these days as nowadays freight on this route runs on the main lines not the ‘electric’ lines and, in any case, the freight trains can generally keep up with the passenger trains so a freight loop would serve little purpose. Under the original Crossrail proposals (before SDO was to be a feature of the plan) it was proposed to replace this loop with one between Goodmayes and Seven Kings, but despite groundwork being carried out, this was never implemented.

Ilford already has three platforms sufficiently long for 10-car Crossrail trains and one that may be long enough and, if not, is currently only short by a couple of metres or so.

Short platforms currently at Seven Kings is surprising. The only apparent reason SDO is currently used appears to be the need to acquire a small patch of land currently used as a car wash. Moreover, it only needs a small portion of the car wash site. The acquisition of the car wash site might not be enough to completely eliminate SDO with 10-car trains but it would reduce it to one or two sets of doors at most.

At Goodmayes there doesn’t appear to be an issue with extending the platforms at the country end.

At Chadwell Heath there doesn’t appear to be a problem with extending the platforms at the London end.

At Romford there doesn’t appear to be a problem with extending the platforms at the London end other than repositioning one or two supports for portal frame for the overhead line.

Currently Greater Anglia has a 2tph fast off-peak service calling at just Stratford and Liverpool Street so extending the platforms on all four through lines has an additional benefit as these trains are 236 metres long.

Gidea Park is unusual in that the Crossrail platforms were lengthened at the country end as part of the Crossrail project even though they were already 9-cars long. It seems that for some reason the end of the platform at the London end is not to be used. It looks as if there is space to further lengthen all the platforms at the country end.

At Harold Wood there does not appear to be an issue at the country end on the Shenfield bound platform but it might be a bit tricky on the London bound platform due to restricted space between the up ‘electric’ line and the down main line.

At Brentwood there does not appear to be any issue with extending the platforms at the country end.

On detailed examination, Shenfield may turn out to be a little problematic but it looks as if there is just sufficient space available where required for this location not to present a significant issue. Before Crossrail there were five platforms and all of them could take a conventional 12-car train – and still can. For Crossrail a sixth platform was built and currently this can only take 9-car Crossrail trains. It looks as if there is space to extend the platform by a single carriage at the London end. A probable worst-case scenario is a need to reposition the platform signal a carriage length nearer to London and use SDO for the final rear carriage when entering the platform.

If there is still not quite enough space to clear the points then it may be possible to slightly extend the platform at the country end as the buffer stops are set back from the platform.

Will It Happen?

In these uncertain times it is unlikely that rail schemes that cannot show sufficient financial benefit will be progressed. On the other hand, if lack of capacity was constraining patronage of the Elizabeth line and a solution produced a net financial benefit then it would be hard to understand why such a scheme would not go ahead when the time is right.

We can be fairly sure nothing will happen until TfL has introduced more trains and seen the effect of that. That would suggest, if such a scheme were to be approved it would not be before 2028 at the earliest. More realistically, little can be expected before HS2 to Old Oak Common has been open for a year to see what additional impact that has if any. This suggests that there will be no decision until around 2032.

On the other hand, if numbers on the Elizabeth line dramatically rose further with demand not satisfied by more trains then, just maybe, something may happen earlier.

Whilst, realistically, nothing will happen on the ground for a long time, the time to be thinking about such a scheme for the future and its feasibility is now. A rough estimate of costs would need to be made to give some idea if such a scheme is worth pursuing. If so then there should be planning to include it in some future budget.

Assuming it is feasible, small but critical portions of land ought to be safeguarded or bought on the open market if they become available at a reasonable price. This especially applies when there is a current benefit in extending platforms. Such actions would not be wasted if longer trains were never to happen.

We would regard TfL predictions for numbers on the Elizabeth line as somewhat cautious and don’t appear to take into account of considerable residential developments along the line of route. Adding an extra carriage to Elizabeth line trains could well be the best value for money TfL will have in the next few years. In this coming era of expected austerity it would be a pity if this potential opportunity were missed through lack of advanced planning.