AHC\WI\PC: Joint London Central Station?

[Alternate History Geek]

As the title says, basically; how can we make it so that the 20ish terminal stations scattered in a ring around London, instead of staying separate to the present day, are instead eventually combined into a single unified London Central station (plus, of course, some smaller suburban stations on the lines radiating out from downtown to serve the areas formerly served by the old ring of terminal stations)? What would be the knock-on effects if this came to pass? And how plausible is it for such a unification to happen, anyway?

Based on some back-of-the-envelope calculations, I'd say we can nail some things down for certain right off the bat; London Central would be an utterly vast station, with somewhere in the vicinity of 225 platforms (over five times as many as the OTL record-holder), and would certainly need to be underground due to its massive size (it would take up a large proportion of the total area of the City of London, assuming a location about where City Thameslink station is IOTL).

Discuss, please!

 

[Alternate History Geek]

Bump .

 

[Lindseyman]

Well it was finally finished in 1997. However it's opening was rather submerged in the euphoria caused by the victory over the Treens!

Sorry that's rather below the belt.

However a London Central station is in ASB territory before WWII due to the competing companies. The negotiations over running rights OTL were bad enough. Imagine the furore here!

After WWII British Railways certainly didn't have the money to attempt it even if they had wanted to.

British Rail in the 70's to 90's likewise.

After privatisation money is the issue (it wouldn't make a profit for YEARS), it would have to be Government funded and run by Network Rail (so even if it was stated in 1997 it still wouldn't be halfway built today!)

There is an exceedingly small window of opportunity when Wilson is PM the first time (white heat of technology and all that) but it is still unlikely.

The availability of land and the costs involved. (Plus where will it be north or south of the Thames? There are problems which ever you choose).

However if London was totally flattened by the Luftwaffe then it might happen.

 

[Alternate History Geek]

Well it was finally finished in 1997. However it's opening was rather submerged in the euphoria caused by the victory over the Treens!

Sorry that's rather below the belt.

However a London Central station is in ASB territory before WWII due to the competing companies. The negotiations over running rights OTL were bad enough. Imagine the furore here!

After WWII British Railways certainly didn't have the money to attempt it even if they had wanted to.

British Rail in the 70's to 90's likewise.

After privatisation money is the issue (it wouldn't make a profit for YEARS), it would have to be Government funded and run by Network Rail (so even if it was stated in 1997 it still wouldn't be halfway built today!)

There is an exceedingly small window of opportunity when Wilson is PM the first time (white heat of technology and all that) but it is still unlikely.

The availability of land and the costs involved. (Plus where will it be north or south of the Thames? There are problems which ever you choose).

However if London was totally flattened by the Luftwaffe then it might happen.

Hmm, now I'm thinking that it might be more plausible with a communist or fascist Britain (probably one that's gotten rid of the monarchy and isn't as likely to hesitate in other aspects of overturning the status quo)...

EDIT: The land-availability argument is why I assumed that the station would have to be mostly underground (like Grand Central Station x5+; there's quite an impressive building on the surface, but it's utterly dwarfed by the vast underground caverns holding the actual guts of the station), or even totally so (given the opposition one would likely face in proposing to tear down anything significant in downtown London to build a train station).

And looking at a map, three possible locations strike me as useable:

- In the City itself, in approximately the same location as City Thameslink station IOTL;
- In Soho at\near the intersection of Oxford Street and Tottenham Court Road;
- In Southwark at\near the rail junction at Blackfriars and Union.

 

[pdf27]

One big issue - the area served by the current ring of stations is rather larger than would be served by one central station: Central London is far bigger than just the Square Mile of the City. If you work around London Bridge, for instance, you can live in Sussex and commute in while for Euston you're looking at the northern suburbs and Cotswolds.
What might work is a far earlier Crossrail/Thameslink-esque scheme with several of the central stations being served by the same train service. That gets you many of the same effects with far less expense or knock-on effects.

 

[pdf27]

EDIT: The land-availability argument is why I assumed that the station would have to be mostly underground (like Grand Central Station x5+; there's quite an impressive building on the surface, but it's utterly dwarfed by the vast underground caverns holding the actual guts of the station), or even totally so (given the opposition one would likely face in proposing to tear down anything significant in downtown London to build a train station).

Actually, that probably isn't an issue. Plenty of bombed out sites during and after the war which you could redevelop.

- In the City itself, in approximately the same location as City Thameslink station IOTL;
- In Soho at\near the intersection of Oxford Street and Tottenham Court Road;
- In Southwark at\near the rail junction at Blackfriars and Union.

You might want to look at a map (I find cycle.travel really, really good for this sort of thing in the UK, far clearer than google maps) here. Those stations are a long way away from the present system: King's Cross to Blackfriars is 2.4 miles, while Paddington to London Bridge is almost 6 miles. A lot of commuters work near the stations they come in to, and for the rest you've already got an excellent system in the Tube. Running every single London commuter through a single Tube station is a recipe for chaos, if nothing else because you're very space-limited underground. Some stations (Camden Town springs to mind) are regularly exit- or entry-only simply because they're so busy: you've got 700,000 coming in by surface rail every day, which you're trying to get through one station and then onto the tube.

 

[Alternate History Geek]

One big issue - the area served by the current ring of stations is rather larger than would be served by one central station: Central London is far bigger than just the Square Mile of the City. If you work around London Bridge, for instance, you can live in Sussex and commute in while for Euston you're looking at the northern suburbs and Cotswolds.
What might work is a far earlier Crossrail/Thameslink-esque scheme with several of the central stations being served by the same train service. That gets you many of the same effects with far less expense or knock-on effects.

There'd still be commuter stations in outer London, to address precisely this issue without giving up the benefits of a consolidated London Central Station - see the OP:

(plus, of course, some smaller suburban stations on the lines radiating out from downtown to serve the areas formerly served by the old ring of terminal stations)

 

[Alternate History Geek]

Actually, that probably isn't an issue. Plenty of bombed out sites during and after the war which you could redevelop.

Even so, it'd almost certainly still have to be underground (the concourse and platform levels at least) due to sheer size; imagine taking the underground portion of New York's Grand Central Terminal (the world's largest train station IOTL), multiplying that by five, and then trying to fit that aboveground, in central London, even accounting for bomb damage. You'd need Berlin 1945 levels of bomb damage in order to be able to make room for that at surface level - and even that might not be sufficient.

You might want to look at a map (I find cycle.travel really, really good for this sort of thing in the UK, far clearer than google maps) here. Those stations are a long way away from the present system: King's Cross to Blackfriars is 2.4 miles, while Paddington to London Bridge is almost 6 miles. A lot of commuters work near the stations they come in to, and for the rest you've already got an excellent system in the Tube. Running every single London commuter through a single Tube station is a recipe for chaos, if nothing else because you're very space-limited underground. Some stations (Camden Town springs to mind) are regularly exit- or entry-only simply because they're so busy: you've got 700,000 coming in by surface rail every day, which you're trying to get through one station and then onto the tube.

See my previous post (you ninjad me, damn you!)

 

[pdf27]

There'd still be commuter stations in outer London, to address precisely this issue without giving up the benefits of a consolidated London Central Station - see the OP:

There are a whole bunch of commuter stations already, e.g. Denmark Hill - the problem is that offices and the like have relocated over time to be near the London termini, and people have moved house to be near the stations using them. The whole point of the OP was to close these termini (Waterloo, Euston, etc.) and build one big central station instead: if you keep them open as "commuter stations" then what'll happen is that most of the traffic will still go into them and "London Central" will never be much bigger than say Blackfriars or Farringdon (Crossrail, Thameslink and 3 tube lines).

 

[Opunium]

Maybe connecting all Terminus stations? Kings Cross, St Pancras and Euston could follow south, join tracks and than continue a cross the Thames to connect to Waterloo. London Bridge and Charring cross can be connected to Waterloo as well, although just demolishing the latter makes more sense if the new Waterloo rail line goes north. Cannon and Liverpool can be connected too, maybe even with Fenchurch, but joing those tracks with the Liverpool tracks earlier makes more sense. Then join Paddington with Victoria. All these tracks would be underground of course with Grand Central style stations along them. Waterloo will be probably the biggest, with trains from a least 5 old terminus stations passing/terminating at it.

A scenario could be a worse post WW1 crisis in which the rail companies go bankrupt, with the government deciding to nationalise them.

If you want a real grand Central Station you could extent all lines into the city centre, to the British Museum, converting the BM into a Grand Central station. That basically requires a communist or fascist Britain.

 

[Dunois]

You wouldn't need 225 platforms since through stations requires less platform space than terminal stations.

The best that one can hope for here is some form of Crossrail for long distance trains, as studied by BR during the early 1980s under the Crosslink label.
http://www.londonreconnections.com/2011/what-if-crossrail-1980/

Such a scheme would free up a lot of platform space in Victoria and Euston/King's Cross.

 

[Alternate History Geek]

You wouldn't need 225 platforms since through stations requires less platform space than terminal stations.

How much less, proportionately?

 

[Devvy]

How much less, proportionately?

At terminus stations, you usually manage around 3-4 trains per hour for shorter distance trains, and around 2-3 trains per hour for longer distance trains per platform. This is due to driver swaps, train cleaning, time for passengers to vacate the platform (as the whole train is emptying) before advertising the train departure for an onrush of people leaving.

For through stations, you can normally get around 12 metro trains per hour with little difficulty - higher then this, and you need trains with more standing room, more doors, and less seats to improve circulation within the train so it can stop and start at each station quicker. Up to 20 or so isn't difficult if you remember those design considerations and invest in signalling. Crossrail style with 30-32 is difficult and need full grade separation everywhere, really good signalling and mitigate risks for delay propogation (ie. delays from a late train on a different network, affecting your train on the fringes, which then cause it to be late in the central section, which causes havoc with your other services).

For longer distance trains, people like comfort, better seats, so it's difficult to design trains for quick stops - 8-10 trains per hour is reasonable, 12 if you have everything going for you.

In answer to the question, you'd probably need 100-120 platforms for the amount of trains that serve London nowadays. An absolutely insanely colossal amount, that is going to need to be subdivided in to small sub stations in order for passenger circulation. So, I don't think a single "Grand Central" station is feasible in any shape or form sadly. Where do these people go after getting off the train?

Bear in mind Waterloo - only one of the major stations, matches Penn Central station in New York in annual usage, and despite the availability of Northern, Jubilee, Bakerloo and W&City tube lines, still has to shut entrances to the Underground network not infrequently during rush hour due to crowding downstairs. Also, London is a large city; one station isn't going to serve everyone anyway. People like one-seat rides, and will complain if their normal service is going to change.

What is possible though, is a network of longer distance Crossrails using infrastructure that London Overground now use for orbital metro services. You can probably get 3 or so north-south links and maybe 2 east-west links which each interface with each other and major interchange points, so 3-4 stops as each passes through London. Maybe something in the 1950s when BR tries to modernise? But it would be an enormous amount of money to ask a Government that is dealing with a shrinking country and decolonisation to pay.

Completely theoretical crayonisting (sorry for size)
http://www.braithwaites.org/aht/lgc.jpg

 

[DaveB]

I think that, due to the influence of dwell time, it makes more sense to look at the number of lines running into terminal stations rather than their number of platforms.

Doing this, north of the Thames you probably need four lines each for the traffic into Paddington, Euston, Kings Cross and St. Pancras, perhaps six for Liverpool Street, and a couple each for Fenchurch Steet and Marylebone/Baker Street, possibly two more for Broad Street, total about 26-28 lines.

South of the Thames, eight each for Waterloo and Victoria, perhaps two each for Charing Cross, Cannon Street and Blackfriars, and perhaps four more for the terminating platforms at London Bridge, so again about 26 lines.

Fortuitously that matches up quite well, doesn't it

So theoretically, if you can get all your trains to run through, that's a minimum of only 26 platforms, though you probably need a few more to allow for looping etc. No need to make provision for trains running through non-stop or for through freight traffic though, which will be routed elsewhere.

In practice some trains would probably need to terminate. In a totally different universe, I once idly imagined a railway system for Middle Earth, in which Minas Tirith station had one huge island platform long enough for two full length trains, with full length bay platforms at each end for terminating trains. For London Central, about eight of these might be adequate . The concourse will have to be placed on the level above the platforms, and everything will have to be underground.

As to how and when you build it, I think you have to start quite early, and create some kind of joint company to run the station and its approach lines, which will obviously all run in tunnels and have to be electrified, probably by third rail. I see this happening around the turn of the century or a little later. Long distance trains will need to change to steam haulage at some point, probably 10-20 miles out.

This all sounds a bit like a model I once envisaged for a 1950/60's Thameslink, which would necessarily of been third rail, and which I thought of as 'VEPs to Bedford', though some through trains between Brighton and Sheffield/Leeds might have been loco-hauled, changing from 71s to Jubilees/Scots/Britannias at Bedford

 

[Devvy]

Fortuitously that matches up quite well, doesn't it

So theoretically, if you can get all your trains to run through, that's a minimum of only 26 platforms, though you probably need a few more to allow for looping etc. No need to make provision for trains running through non-stop or for through freight traffic though, which will be routed elsewhere.

You'd probably need 2 platforms for each direction/line (trains alternating between the two platforms); trains, especially during peak, would be coming along too rapidly. Considering it would be one single London station, we're talking about the vast majority of the train getting off, and a huge number getting on to each train. Probably looking at 4-5 minutes dwell time per train.

Also, because each platform will have a huge number of destinations, and that pesky habit of passengers to arrive on the platform around 5-10 minutes before the train is due for departure, you're going to have significant amounts of passengers standing around on a cramped underground platform (loitering whilst waiting for their train), in addition to the huge numbers of passengers attempting to get off, or get on the train.

Then multiply all those problems by 10 for the Underground connections. Basically, this "Grand Central" station is going to be exit-only on the Underground at peak times, and probably for a majority of the day.

EDIT: Eventually found some nice data to use as well:

On a typical autumn weekday in 2012, 536,000 passengers arrived into central London (Zone
1 of the TfL travelcard area) during the three hour morning peak (07:00 to 09:59). This was a
0.6 per cent increase from 2011. Across the whole day 981,000 passengers arrived into
central London by rail, a 0.9 per cent increase from 2011.

So, during 2012, on a weekday, around 180,000 people arrived in to central London per hour over the 3 hour AM peak. If all these people are arriving at one station, it's an unmitigated disaster on a completely unfathomable level. Even halving this, to allow for some other suburban stations, considering most people will be transiting and connecting via this station, almost 100,000 per hour will be arriving per morning hour in the morning. If this station is underground, it's still an absolute disaster of crowding and probably spending longer getting out of the station then the train took to arrive at the station.

PPS: When I watch my local footy team (the gloriously average Reading FC), the stadium is about 24,000 capacity. It takes a good 15-20 minutes for people to filter out of the stadium, which is around 75,000-100,000 per hour. And that is literal one-way traffic as everyone disappears; no people waiting around for subsequent matches (trains), people attempting to come in to the stadium for the next matches, in built above ground with huge areas just outside the stadium for people to circulate in and make their different ways home via wide roads that are de-facto pedestrian only during match times.

 

[Alternate History Geek]

As to how and when you build it, I think you have to start quite early, and create some kind of joint company to run the station and its approach lines, which will obviously all run in tunnels and have to be electrified, probably by third rail. I see this happening around the turn of the century or a little later. Long distance trains will need to change to steam haulage at some point, probably 10-20 miles out.

Why would it have to be third rail, necessarily? If you use overhead catenary, you could (eventually) not need to change locos at all, assuming that the network gradually moves over time to total electrification. Plus, using catenary allows the lines (or at least some of them) to be converted to high-speed rail at some point; third rail is speed-limited due to the stress of the contact shoes hitting the joints in the rail.

(Oh, and it's also much safer than third rail if someone accidentally falls into the track pit or wanders onto the aboveground portions of the rights-of-way. With catenary, you don't run the risk of someone getting electrocuted if they step on the tracks wrong. If you must use third rail, the problem can be mitigated by using bottom-contact TR, but then the ends of the third rail have to curve up at grade crossings to allow the contact shoes to reliably reengage after crossing the road, which leads to situations like this.)

 

[pdf27]

Overhead lines and tunnels aren't a great mix - you can't use 25kV AC without a lot more clearance, and even the stepped down voltage (12.5kV? 6.25kV?) used for tunnels and bridges is problematic and causes performance issues. The big issue is simply that you need a much bigger tunnel than you do for a 600V DC third rail.

 

[Alternate History Geek]

Overhead lines and tunnels aren't a great mix - you can't use 25kV AC without a lot more clearance, and even the stepped down voltage (12.5kV? 6.25kV?) used for tunnels and bridges is problematic and causes performance issues. The big issue is simply that you need a much bigger tunnel than you do for a 600V DC third rail.

Why is that, precisely? And if that's the case, how come subway trains using catenary don't have these kinds of problems?

(Although I suppose one could use dual-mode trains, running on third rail underground and then lifting the shoes and going catenary once they leave the tunnels...)

 

[DaveB]

You'd probably need 2 platforms for each direction/line (trains alternating between the two platforms); trains, especially during peak, would be coming along too rapidly. Considering it would be one single London station, we're talking about the vast majority of the train getting off, and a huge number getting on to each train. Probably looking at 4-5 minutes dwell time per train.

Also, because each platform will have a huge number of destinations, and that pesky habit of passengers to arrive on the platform around 5-10 minutes before the train is due for departure, you're going to have significant amounts of passengers standing around on a cramped underground platform (loitering whilst waiting for their train), in addition to the huge numbers of passengers attempting to get off, or get on the train.

Then multiply all those problems by 10 for the Underground connections. Basically, this "Grand Central" station is going to be exit-only on the Underground at peak times, and probably for a majority of the day.

Can't disagree with most of that.

Thinking about it, I think many peak time commuter services would have to be routed into the traditional terminuses, and any Central station would have to deal mainly with longer distance services. That would at least allow passengers easier interchange than OTL.

Thinking back to the 70's, off peak main line services weren't that intense, and were probably even less in earlier times. On the WCML after electrification IIRC there were flights of six or seven fast trains every hour (2xBrum, Liverpool, Manchester, Glasgow, North Wales/Blackpool/Barrow/Carlisle semi-fast?), on the MML only two, on the ECML perhaps three or four, WR four or five (West Country, Bristol, South Wales, Hereford or Brum via Oxford), Liverpool St. just a couple. That's only about twenty expresses an hour, and even with some local trains, perhaps no more than 40-50 an hour. South of the river there was e.g. on the Brighton main line just one hourly Brighton non-stop plus an Eastbourne via Lewes, a Bognor (or Littlehampton?) via Hove and another via Horsham. Service frequencies to Kent and the Solent were similar.

Suddenly doesn't seem all that difficult. Might be interesting to work out a rough timetable, though a bit too much work

Edit: meant to add that 3rd rail was probably the most likely option in a turn-of-the-century scenario, though low voltage OH was a possibility - 1500 v DC, or whatever the other southern system (ex LSWR?) that was abandoned in favour of 3rd rail was. If it was done more recently it would obviously be 25 kV AC or dual voltage.

 

[Alon]

Question: does something like Chatelet-Les Halles, a central station with access to several commuter rail lines but no intercity lines, count? Because if so, then it could have easily happened if planning had gone just a little differently in the 1960s.

Even a central station with some intercity rail access could have possibly happened, with a POD in the 1950s or 60s, but that would be at the edge of plausibility. It requires a TL in which the UK heavily invested in a few mainlines, e.g. the ECML, WCML, and GWML, possibly turning them into high-speed rail. I can sort of see a single central station in Central London with connections to both the ECML/WCML and the GWML. In similar vein, I can sort of see a mainline rail tunnel in Paris connecting Gare de Lyon and Gare du Nord, offering TGV through-service that also serves central Paris rather than swerving around it as in OTL. Berlin built such a tunnel, the 4-track Tiergarten tunnels, when it built the new Berlin Hauptbahnhof.

For the record, my Anglo-French TL assumes central stations in the major cities, including Paris and London, as above. New Amsterdam's was built in the early 20c and is nicknamed Cartel Station, since it was a union station of competing railroads; London and Paris's were built in the 1960s alongside investments in high-speed rail and commuter rail upgrades.

Another alternative: back when it knocked down Euston, London might well have built a joint King's Cross-St. Pancras-Euston joint station. They're all a spit's distance from each other anyway.