Notions for a GWML metro

[mr_jrt]

Over on London Reconnections I've been discussing an idea I had for implementing a metro service along the GWML. Obviously it's a bit blue-sky, but I thought I'd get some thoughts from posters over here about it.

Basically, the premise is this: When Crossrail opens, the short platforms (13/14 - etc) at Paddington used for metro-like services won't be needed any more as the service will either be run onto Crossrail or cut back (i.e. Greenford to West Ealing). Also, when the OOC redevelopments go ahead, there's a distinct chance the Crossrail depot will be moved away from OOC to permit more redevelopment. The former permits the restoration of 4 H&C platforms, the latter frees up the depot lines from Royal Oak to OOC to use as a segregated 6-track alignment.

This would enable the H&C to gain a branch to OOC very, very easily, and would permit platforms at all of the locations along the GWML clamouring for them that can't be justified on Crossrail (so Royal Oak, Westbourne Park, Ladbroke Grove/Kensal Gasworks, OOC).

Once you're at OOC, you have options. You could take over the Central Line service up the NNML, you could take over the freight lines down past Acton yard and boot the Central or the District out of Ealing Broadway and terminate there (so adding Acton Mainline, Noel Road (OSI with West Acton) and Ealing Broadway), or you could use the former GWR goods alignment along the NNML to Perivale, then run down the Greenford branch and then 6-track along to Slough/Airport Junction to bypass the Ealing Broadway to West Ealing bottleneck that prevents 6 tracking.

Crossrail could then just focus on Hayes and Harlington, Ealing Broadway, OOC and Paddington, and you could even perhaps argue against Ealing Broadway once OOC opens.

Basically, you start treating the H&C a lot more like the Metropolitan Line rather than the Circle Line, and Paddington Bishop's Road a lot more like Baker Street. If the Met can operate that many services into 4 platforms with conflicting terminal movements, then the H&C should be able to easily handle at least that many, if not more, and the GWML gets a high-frequency metro that can handle the suppressed demand along that axis far better than Crossrail's proposed jack-of-all-trades skip stop timetabling can.

 

[Tobbes]

I'm slightly confused by this, Mr_jrt - is there evidence of suppressed demand on this axis, and what would happen to the existing H&C services, now that they've been tea-potted into the not-quite-circle line?

Interesting, though.

Tobbes

 

[mr_jrt]

I'm slightly confused by this, Mr_jrt - is there evidence of suppressed demand on this axis

I think so, yes. The GWML metro services are rammed to the gills (with a 5-car 4tph service it's somewhat to be expected), and the roads are heavily congested. Just look at the heat maps produced to justify Crossrail. The problem with Crossrail is that it's got to fill too many roles - Paddington to Heathrow, Paddington to Maidenhead, in all likelihood, Paddington to Reading as well...and there's not enough line capacity to serve the stations within the M25 properly - just witness the horrible proposed skip-stopping pattern that leaves some stations with a measly PEAK service of 2tph! This simply isn't good enough on one of the mainline from London. Oh, and don't forget the freight Crossrail will have to share with on the relief lines! All these factors make running a proper metro service with a regular service interval largely impossible.

Building this metro would enable Crossrail to focus on suburban services, more akin to Thameslink. I'd imagine the only Crossrail stops being Hayes & Harlington, Ealing Broadway and when it opens, OOC replacing Ealing Broadway as a stop. Maybe. Rather than a pathetic 2tph, the stations along this axis would get something more reasonable like 8tph or some such. Smaller trains than Crossrail, but most importantly - far more frequent. The freight would also find it far easier to path between Crossrail services that keep moving and stop infrequently than all-stations services.

and what would happen to the existing H&C services, now that they've been tea-potted into the not-quite-circle line?

That's where the options come in.

Option a is to leave it alone, and have a flat junction west of Royal Oak operating as a branch. You could turn these services back at Paddington, or have them run through. It's as flexible a set of options as the Met has at Baker Street. Personally, I'd want to take advantage of the fact that you could build the new route to handle S8s, and Paddington and Royal Oak should be easy to extend to handle them, then you're largely good to get to Upminster with S8s. So with that in mind, I'd probably terminate the current H&C services at Paddington and run longer S8s though to the Circle from further afield along the GWML to make the most of capacity.

Option b is to find another service to hook it up to, and in all likelihood that would be the Bakerloo extended from Edgware Road. Cut it back to Queens Park, reinforce the LO service to replace it, then divert the majority of the Bakerloo service over the new link to Hammersmith. Maintains the links, but makes better use of capacity on both.

Option a is obviously dramatically cheaper than option b, and I'd probably go with that unless capacity between Royal Oak and Paddington became too much of a problem, in which case you progress on to option b at a much later date.

 

[Tobbes]

Interesting. But your proposal demands that Crossrail's OOC maintenance base be moved, which I can't see happening as

(i) it would significantly increase delivery risk to do it now with the bids for Class 345 based around an OOC depot;

(ii) where would you put it? And how long would it take to get planning permission, EIAs ahead of a 2017 delivery of the 345s?

Which, whilst I accept the point in the LR series (Pt 2; Pt 3) about the Crossrail Depot being in the way of the OOC masterplan, there's no obvious call for it to be moved.

 

[swt_passenger]

It's a big assumption that all the short platforms at Paddington will become spare.

I thought there was a plan to combine 12 and 13 to provide an extra intercity length platform, for use by all the extra LDHS paths resulting from the Reading re-modelling. The DfT refer recently to 4 tph additional - which needs more platforms overall, I would expect.

 

[LexyBoy]

It's a big assumption that all the short platforms at Paddington will become spare.

That was my first thought as well. There will be more IC services, and Crossrail will not take out all the suburban services - only HeC and Greenfords (4 tph), possibly the Reading stoppers if extended (2 tph) and hopefully HeX eventually. The Oxford and Newbury services will remain, and there's been a Paddington-Basingstoke service mooted too though I don't know what's going on with this.

I would agree to the need for a proper inner-suburban service though, the GWML is very poorly provided for. My instinct is that this should be part of Crossrail though; having to change at OOC or Paddington would make it less attractive.

 

[HSTEd]

There are so many trains being turned back East at Paddington on crossrail that providing additional services would not be hard if you have track to use on the GWML.

I assume that there will be a lot of trains extended to OOC to link with HS2 as well....

 

[jopsuk]

Problem is that there's still going to be quite a few GWML "local" services sharing tracks with Crossrail. Ideally CR would have a thried "pair" of tracks. As I understand it, this will be less of a problem on the GEML side, where most trains beyond Shenfield will be able to use the "main" tracks?

 

[anthony263]

I assume that there will be a lot of trains extended to OOC to link with HS2 as well....

That is the plan put forward by TFL when they published a document looking at the redevelopment of the Old Oak Common area which also mentioned a station on the West London Line

 

[swt_passenger]

Problem is that there's still going to be quite a few GWML "local" services sharing tracks with Crossrail. Ideally CR would have a thried "pair" of tracks. As I understand it, this will be less of a problem on the GEML side, where most trains beyond Shenfield will be able to use the "main" tracks?

There is a significant difference on the GEML. The other few (6 tph non-Crossrail) peak extra services from Shenfield heading into Liverpool St are only interworking with a single Crossrail branch, so superimposed on Crossrail's 12 tph, and they have space to provide a properly grade separated junction between Stratford and the Pudding Lane Portal, so inbound trains off the 'Up Electric' (slow lines) are non-conflicting with Crossrail.

(The residual 6 tph Liverpool St high level service off the Up electric is as shown in the London and SE RUS, section D2 Bow Jn remodelling.)

At the Paddington end, all 24 tph Crossrail are leaving the same portal, but 14 tph are heading for the turn back. But there is in this case no grade separation where the Crossrail pair joins the relief pair, so anything going into Paddington off the up relief has to cross the down relief somewhere west of the Westbourne Park reversing sidings. So if only two more Crossrail trains than the currently intended 10 tph (6 to the GW and 4 to Heathrow) are extended further west, then the situation regarding flat crossings is immediately worse than the equivalent GEML set up.

It remains to be seen if any layout changes could or would be made. If more (or all) Crossrail services went further west, clearly the Westbourne Park turnback is starting to become redundant, but I don't know if there's room there to insert an up relief avoiding route over the down Crossrail.

 

[jopsuk]

Bascially, if it was possible the GWML should be six track to Maidenhead, used:
2 tracks for Crossrail only
2 tracks for HEX (yes, I know this is supposed to be replaced by CR, would be better if just Connect was), GWML "locals" to the branches and Reading.
2 tracks for HSS.

(similarly, the ECML should be six tracks to Welywn and four to Hertford North, the GEML six to Shenfield & four to Ipswich, but that's another matter)

 

[HSTEd]

Why was one of the tunnel bores not extended to permit a fully grade seperated junction at the western end of the tunnel core?

 

[swt_passenger]

Why was one of the tunnel bores not extended to permit a fully grade seperated junction at the western end of the tunnel core?

Probably because the only way of doing that would be to have the 14 tph turnback happening somewhere else? Once the decision to have what is effectively a 'dummy' terminus station between the up and down Crossrail running lines was made, (including narrow platforms to allow for emergency evacuation or over carrying passengers, and with all the necessary S&C), it all gets more and more complex...

If the original idea had been to have had two full (ie 12 tph) western branches, I think the split would have had to be much further west (eg at Old Oak Common). Original decisions to have a completely mis-balanced network have come home to roost...

 

[jimm]

Problem is that there's still going to be quite a few GWML "local" services sharing tracks with Crossrail. Ideally CR would have a thried "pair" of tracks. As I understand it, this will be less of a problem on the GEML side, where most trains beyond Shenfield will be able to use the "main" tracks?

No, there aren't going to be GW services sharing tracks with Crossrail. The current plan is that east of Maidenhead, the only things running on the relief lines will be Crossrail and freight. Hence the ever-increasing speeds being proposed for electric GMWL outer-suburban stock, so it can keep up with IEPs on the fasts.

And there is no possibility of providing another pair of tracks. There are places that loops can be reinstated, etc, but that's all.

Crossrail should go to Reading. The fundamental reason it doesn't (for now) was its backers didn't want to get landed with a bill for improvements and electrification at Reading, but so drawn out has the development of Crossrail been that Network Rail have taken care of the issue of Reading, so the sooner someone sees sense and Crossrail goes to its logical western finishing point, the better. Saying that, services from the likes of Didcot or Oxford into the Crossrail tunnel would be no bad thing. Most people doing the commute don't finish their journeys in the Paddington area.

 

[swt_passenger]

No, there aren't going to be GW services sharing tracks with Crossrail. The current plan is that east of Maidenhead, the only things running on the relief lines will be Crossrail and freight.

Someone ought to tell Crossrail then. There are definitely still GW services shown on their website's version of the timetable, including one through service each day from each of the branches:

When full service commences in 2019, Crossrail will provide ten trains per hour on the Great Western main line at peak. This will include four trains per hour to Maidenhead, four trains per hour to Heathrow and two trains per hour to West Drayton.

Other operators’ services are assumed to be:

2 trains per hour from Reading calling at Twyford, Maidenhead, Slough, Hayes & Harlington and Ealing Broadway to Paddington;
2 trains per hour from Reading calling at Twyford, Maidenhead, Taplow, Burnham and terminating at Slough;
1 through train from Henley on Thames to Paddington; and
1 through train from Bourne End to Paddington

http://www.crossrail.co.uk/route/surface/western-section/

 

[Rational Plan]

There plans floating around for a fifth track between Langley and Southall. Not sure if it got anywhere though. I think the idea was to separate the freight for as much as possible.

Realistically that is all that is going to happen along with the new western access to Heathrow.

At some point reality will intrude and crossrail will arrive in Reading. It is likely that rail services will altered to something similar to the RUS plan and we will see more crossrail trains along the GWML. with the majority going to the airport. I think it was something like 8 or 10 an hour, with 2 potentially heading to Staines (at some future point) and 2/4 rejoining the Great Western towards Reading.

Crossrail is more than capable as acting as a Metro service for the stations between Reading and London.

If the idea of new high speed lines takes off, then you could see an extra pair of tracks. The most obvious option would be a new high speed line between Old Oak Common and just East of Reading, the best route to take would be along the M4 with a big chunk in tunnel. (obviously we are talking the 2040's here)

The problem then is the lack of platforms, and the only solution would be to put some the trains to Oxford and Newbury into the crossrail tunnels.

 

[mr_jrt]

Crossrail is more than capable as acting as a Metro service for the stations between Reading and London.

Indeed, but once in London it's pathetic until you get to the core, which is kinda my point.

If the idea of new high speed lines takes off, then you could see an extra pair of tracks. The most obvious option would be a new high speed line between Old Oak Common and just East of Reading, the best route to take would be along the M4 with a big chunk in tunnel. (obviously we are talking the 2040's here)

Indeed, I think this the most likely outcome actually, I'm just musing that you might be able to get a quicker turnaround by building this metro incrementally on the surface to get something a lot quicker.

The problem then is the lack of platforms, and the only solution would be to put some the trains to Oxford and Newbury into the Crossrail tunnels.

Again, I agree with this, which is why I don't see having Crossrail serving all-stations (including the additional stations I propose) Paddington to Slough is the best use of capacity, serving Oxford/Didcot/Newbury makes a lot more sense (i.e. those are the outer suburban services all stations to Reading, then semi-fast to the core, and the inners are those that turn back at Reading and are all-stations to the core). Having a separate metro service permits the Reading terminators to run semi-fast from Slough/Hayes & Harlington, etc.

 

[Rational Plan]

Indeed, but once in London it's pathetic until you get to the core, which is kinda my point.

Indeed, I think this the most likely outcome actually, I'm just musing that you might be able to get a quicker turnaround by building this metro incrementally on the surface to get something a lot quicker.

Again, I agree with this, which is why I don't see having Crossrail serving all-stations (including the additional stations I propose) Paddington to Slough is the best use of capacity, serving Oxford/Didcot/Newbury makes a lot more sense (i.e. those are the outer suburban services all stations to Reading, then semi-fast to the core, and the inners are those that turn back at Reading and are all-stations to the core). Having a separate metro service permits the Reading terminators to run semi-fast from Slough/Hayes & Harlington, etc.

Crossrail will provide plenty of capacity for the GWML well into the 2020's and considering political realities there are plenty of other corridors that will want billions spending on it.

So in reality I don't think anyone will even a whisper a thing till till the end of the 2020's.

I think the key date is once phase one of HS2 is open, as long as it has not gone terribly wrong, I think everyone will be screaming for their own line.

The most obvious option is speeding up construction of phase 2, with some possible extensions further North. (Lancaster or Carlisle say in the West or Darlington in the East).

 

[jimm]

Someone ought to tell Crossrail then. There are definitely still GW services shown on their website's version of the timetable, including one through service each day from each of the branches:

Fair enough, but what they are suggesting, such as Reading-Slough trains, illustrates full well why Crossrail should go out to Reading in the first place, so that 110mph (or more) emus are not pottering along amid 90mph Crossrail services.

The original poster complains about suppressed demand with five-car dmus now. Crossrail Class 345s will be 200m long with a capacity of 1,500 people. Whatever frequency they are operating at, that's a hefty capacity increase.

As for

that leaves some stations with a measly PEAK service of 2tph!

Some stations would, according to the indicative service table swt_passenger linked to, appear to be one station - Hanwell - presumably that 2tph is based on the current service and demand and an assessment of future demand, which says it isn't worth more.

Even if you do have six tracks available between Paddington and Old Oak, the idea of then placing crossrail stations at Royal Oak and Ladbroke Grove just has bottleneck written all over it - just at the point where trains are entering and leaving the tunnel, which doesn't sound like much a recipe for good train regulation with a 24tph service either.

Crossrail is more than capable as acting as a Metro service for the stations between Reading and London.

Indeed, but once in London it's pathetic until you get to the core, which is kinda my point.

And kinda the point of Crossrail was not to provide yet another Underground-style service in west London, it was to provide a relief to the Tube lines from Paddington and Liverpool Street by moving people travelling in from further out and leaving trains at those stations on to direct services taking them closer to their final destinations in London without a change or, if they are coming off longer-distance services at Paddington in particular, to avoid the less than fantastic onward Tube links, the narrow H&C platforms, etc

 

[Zoe]

Some stations would, according to the indicative service table swt_passenger linked to, appear to be one station - Hanwell - presumably that 2tph is based on the current service and demand and an assessment of future demand, which says it isn't worth more.

Looking at that service table, it seems that the Off Peak service at West Drayton will be reduced to from 4 to 2 tph. Is it the case that there is only demand for 2 tph off peak and if so then why provide 4 tph currently? Also is there any reason why places like Manor Park, Forest Gate and Maryland have demand for 6 tph off peak into Central London but Hanwell only has demand for 2 tph?

 

[mr_jrt]

The original poster complains about suppressed demand with five-car dmus now. Crossrail Class 345s will be 200m long with a capacity of 1,500 people. Whatever frequency they are operating at, that's a hefty capacity increase.

Though it's a accepted fact that unless you have at least 4tph people won't switch mode as once you factor in the waiting time, their journeys still take longer. Remember, with a 2tph service, worst-case scenario for someone turning up to use the service unplanned is that they spend 29 minutes waiting on top of however long the Crossrail train takes to reach their destination. Odds are that's longer than the time it will take them to zone 1. Hardly a wonder usage isn't high!

Even if you do have six tracks available between Paddington and Old Oak, the idea of then placing crossrail stations at Royal Oak and Ladbroke Grove just has bottleneck written all over it - just at the point where trains are entering and leaving the tunnel, which doesn't sound like much a recipe for good train regulation with a 24tph service either.

Quite, which is why I propose linking it to the H&C at Royal Oak. The Underground operates on different principles to NR (i.e. not timetable-bound at junctions) that are much better suited to high-frequency metros. If you're going to have a flat junction on a metro system, you're better off with a Underground one. That said, Crossrail is going to have to merge at Airport Junction and the turnback at OOC. If you were to operate this metro as part of Crossrail, then with platforms just west of the merge at least the trains can wait in platforms for their slots at the junction.

And kinda the point of Crossrail was not to provide yet another Underground-style service in west London, it was to provide a relief to the Tube lines from Paddington and Liverpool Street by moving people travelling in from further out and leaving trains at those stations on to direct services taking them closer to their final destinations in London without a change or, if they are coming off longer-distance services at Paddington in particular, to avoid the less than fantastic onward Tube links, the narrow H&C platforms, etc

I agree, again, why I think projecting the H&C would be a great idea. More passengers from the GWML axis will be able to stay on their H&C train all the way through to the city without clogging up Crossrail.

Looking at that service table, it seems that the Off Peak service at West Drayton will be reduced to from 4 to 2 tph. Is it the case that there is only demand for 2 tph off peak and if so then why provide 4 tph currently? Also is there any reason why places like Manor Park, Forest Gate and Maryland have demand for 6 tph off peak into Central London but Hanwell only has demand for 2 tph?

I agree, I think catering for current demand with new infrastructure is a fool's errand. There is huge suppressed demand - just look at the line on Google maps and witness the amount of housing along it along with the road congestion! It's the self-same discussion had over and over regarding the Uxbridge Road corridor as well.

In fact...one more extravagant option might be for the Central Line to dive down at Ealing Broadway and continue to Uxbridge along the Uxbridge Road, with the H&C taking over from North Acton to West Ruislip and via the NNML and Greenford branch to West Drayton/Slough. It's a 6-track formation though North Acton, so there's plenty of room, and it avoids the issues people have had elsewhere with my proposal's impact on Acton Yard and freight.

I should probably draw some maps

 

[HSTEd]

High speed line that goes from Paddington, to a station under or adjacent to Heathrow and then proceeds at least as far as Cholsey with a trajectory that permits easy extension to Bristol and beyond.

That frees the fast tracks from everything but the Reading-London service that will replace the current station calls by long distance trains.
That allows everything to pack down and Crossrail to have complete control of the relief lines.

 

[tbtc]

Oxford to London is a confusing route. Some enthusiasts want to run it with toiletless trains that only have 45 seats per carriage... yet some complain that the Intercity IEPs designed to run it may not have a buffet.

Given the money being spent on Crossrail and the large number of trains from Abbey Wood/ Shenfield that'll be terminating at Paddington, its hard to buy the argument that we need to spent further money on this corridor - its going to take quite a while for the 200m 345s (with capacity for 1,500, mainly standing) to fill up compared to the 165s.

Maybe after a few years there'll be a follow on EMU order to replace HEx (once the ability to take a through train to "central" London at Oyster prices decimates much of the market currently prepared to pay £15 just to get as far as Paddington) which would allow for additional units to extend beyond Maidenhead to Reading. Apart from that... pie in the sky, sorry.

 

[jimm]

Oxford to London is a confusing route. Some enthusiasts want to run it with toiletless trains that only have 45 seats per carriage... yet some complain that the Intercity IEPs designed to run it may not have a buffet.

Given the money being spent on Crossrail and the large number of trains from Abbey Wood/ Shenfield that'll be terminating at Paddington, its hard to buy the argument that we need to spent further money on this corridor - its going to take quite a while for the 200m 345s (with capacity for 1,500, mainly standing) to fill up compared to the 165s.

Maybe after a few years there'll be a follow on EMU order to replace HEx (once the ability to take a through train to "central" London at Oyster prices decimates much of the market currently prepared to pay £15 just to get as far as Paddington) which would allow for additional units to extend beyond Maidenhead to Reading. Apart from that... pie in the sky, sorry.

I wouldn't advocate operating an Oxford service with toiletless trains with only 45 seats per carriage. Just as there is not reason that IEPs cannot have different interior configurations, there's nothing to say you can't have an alternative interior in a Crossrail set. Even if you put in more seats, a 200m train would still have plenty of carrying capacity.

In any case, the Oxford service is a mixture of fast and semi-fast services, with broadly two types of rolling stock in use, something which will remain the case after electrification and crossrail.

And if you want to go beyond Maidenhead to Reading, you order more Crossrail-type stock and less of the GWML semi-fast type emus, for which there would be a reduced requirement as a result.

While the Superlink proposal to make Crossrail into far more of a regional service, with several branches, was rejected by DfT, TfL and Ken Livingstone, there are successful examples of mixed services on such routes, like the east-west railway tunnel under the centre of Oslo. Yes, it's a rather shorter affair in a smaller city, with just one underground central station, but it carries long-distance, regional and suburban services and is capable of handling 24 tph with the current signalling, with the potential for more - increasingly likely to be needed given the demand for more trains to operate. There is also talk of an additional tunnel under the city, possibly built for high-speed operation.

I suppose London might cotton on eventually but given that Crossrail's origins date back to the 1940s and we have years of wrangling over Crossrail2 to look forward to, who knows how long it will take to work out that main line services could also pass under the centre of London.

 

[Rational Plan]

The Great Western is lucky to get the Heathrow link as well as crossrail, as time goes on it will only be about adding more services to Crossrail , got to use those terminating services for something after all!