Quick Note: High Speed 2 and Euston
There was reporting in the Sun, since officially denied, that Britain is planning to cut Euston from its high-speed rail project and run trains only as far a Old Oak Common, a future development site west of Central London. I assume given the source and lack of any other confirmation that the plans are to run to Euston as planned. But what if the story is not completely fake news, and there are plans to cut on construction at Euston? I can see a cut being positive value engineering, using space at the station more efficiently.
What’s the issue with High Speed 2?
High Speed 2 is an extremely expensive line. Among proper intercity high-speed rail lines (as opposed to suburban lines running at medium speed), it is the most expensive in our database per kilometer. The projected cost as of 2019-20 is about the same as that of all lines built to date in Germany and France combined; Germany has about 1,000 km of newly-built high-speed lines and France 2,500, whereas HS2 is planned to be 530 km.
The high costs are related to some massive scope creep, including tunnels in relatively flat terrain through the Chilterns, dug essentially because the area has rich NIMBYs and the British state decided not to fight them. Those are already in advanced enough construction that I don’t think descoping them and building the line at-grade with compulsory purchase of land is viable. However, some of the scope is new stations, which British defenders of the system insist are necessary. Birmingham is to get an entirely new station at Curzon Street, and London Euston is to get a substantial increase in size, with additional tracks and approaches. This is said to be necessary for capacity reasons.
Are new stations necessary for capacity?
No. Euston today has 16 platform tracks; it had 18 before HS2 construction started. The S-Bahn-quality Watford DC line can use two; the remaining slow services at the station amount to around 10-12 trains per hour, which S-Bahn-quality terminals like Saint-Lazare on the RER E and Catalunya on the Barcelona Rodalies network can comfortably turn on four tracks; those two comparisons turn 16 and 24 trains per hour on four tracks, respectively. The services out of Euston branch more than the RER and Rodalies, but this is mostly a mix of stopping patterns, largely on the same legacy line.
Then there’s HS2 itself. The line is expected to get very high ridership, justifiably: all cities along the line are larger than their comparison cases on the LGV Sud-Est, often substantially, and the projection is that very high capacity is required, on the order of 16 trains per hour. This stretches high-speed rail to its limit: the Shinkansen, which mixes local and express trains on double track, peaks around 14-15 trains per hour, and the complexly branched TGV around 12. HS2 expects to do better perhaps through better signals but also through having a simpler stopping pattern on its most congested section, between London and the bifurcation at Birmingham Interchange, on which trains are to run nonstop.
However, 16 trains per hour can still turn on about six platform tracks. This is not easy: the Tohoku Shinkansen turns 14 on four tracks, but this is a limit case, famous not just in rail media but also in business media as successful optimization of infrastructure. Nonetheless, given how constrained the site is, it’s useful to learn from the best and not the average. If it’s possible to descope the plans to add new tracks to Euston, this should be done; present plans for Euston cost billions.
Is this happening?
Maybe. Britain is aware of the situation at Tokyo Station, although it seems more interested in looking for reasons not to learn than in learning. Perhaps very high costs are leading to a reevaluation, in which Euston can be made smaller than in current plans and trains can turn more efficiently.
But again, the ultimate source on this said nothing of this sort, and is unreliable. So who knows?
Pedestrian Observations 
Does the number of trains turnable scale with platform tracks? Maybe the need to cross tracks to get to the opposite direction makes things more complicated?
Route conflicts, where departing trains block arrivals or vice versa, are indeed a separate issue from platform occupancy and the time it takes to reverse trains.
When I last remember having had the stomach or enthusiasm to look at HS2 Euston plans (a decade ago?) there was a super cool and (of course) massively expensive two-level underground bifurcation of the approach tracks which was clearly designed to minimize arrival and departure route conflicts. Is this still the case, or was it “value engineered” out at some stage?
Anyway, somewhat tangentially …
What surprises me (and amazes me, because it’s so much contrary to what one sees in exemplary major station track configurations in Europe, with their arrays of parallel “ladder tracks”) is just how little this terminal station flat junction throat throughput seems to matter in operations at the global boss HSR terminal stations in Tokyo, both JR East and JR Central.
Three of the four JR East Shinkansen platform tracks at Tokyo’s main station (20, 21, 22) fan out from just the inbound (southbound) track, and there are no crossovers that allow other simultaneous arrival/departures, so aside from a departure from platform 23 simultaneous with arrival on 20/21/22, every single other train movement into or out of the station involves what is effectively a single-track bottleneck for the 375m between the ~400m station platforms and the scissors crossover that is the limit of the station throat. (Not dissimilar at the adjacent and disconnected JR Central Tōkaidō platforms 14/15/16/17/18/19, where 14/15/16 and 17/18/19 respectively fan out from a single approach track, with no “parallel move” crossovers between them.)
Watching online videos of trains traversing this bottleneck one sees that it takes over 60 seconds for a 400m train to traverse the station throat turnout zone (and they’re very nice expensive higher-speed turnouts, perhaps 70kmh or even 80kmh diverging speed), meaning that at 14tph, almost half the time the station throat bottleneck is physically occupied by a train! That’s incredible. (OK, many of the trains are less than 400m long, and departing trains are faster than arrivals, but still — really impressive!) Schedule adherence “elektronik” before concrete before … well, everything.
The COMTRAC traffic control system used by JR Central is the mechanism behind the Tokaido Shinkansen throat track operation- it is currently in its ninth generation, which allows point changes taking 8 seconds, shortened by 4 seconds from the previous generation- thus allowing earlier departures. In October this year the tenth generation is scheduled to go into operation, featuring more flexibility in the case of schedule disruptions- when previously a late running arrival would be given priority to enter the crossovers, the new system will prioritize the departing trains- holding the late train while getting the departing train(s) out on time, thus minimizing cascading delays.
The problem of terminal stations is that trains have to cross a large number of tracks either when they enter or when they leave, and that crossing movement is the biggest limitation on capacity. Zurich’s solution was to build a flyover (Letzigrabenbrücke) from one edge of the station throat to the other. Though Alon might have some comments about the cost (CHF 300 million for two flyovers).
Yes, I believed so too, and this is nearly always the case, but Shinkansen terminal operations in Tokyo are mind-blowing.
Zürich HB or Luzern HB or Stuttgart HB just aren’t in the same league.
Actually, it was first to build the Hirschengrabentunnel and the underground four track through platforms of Museumstrasse.
Then the flyover near Killwangen-Spreitenbach, so the Fernverkehr tracks out of Zürich HB terminal tracks run “wrong way” (right hand traffic) to reduce terminal path conflicts. (George Raymond’s neat old 2001 web site http://www.railweb.ch is still up after all these decades.)
Then Weinbergtunnel and the second four track through platform underground platforms of Löwenstrasse.
Then the over-the-top (hah hah!) Letzigrabenbrücke (second phase of Löwenstrasse)
Lots and lots of concrete, and all well targeted and well spent.
But still, Japan, man. Japan. Whoa. Excess crossovers are for those with weak minds and weak discipline!
I’m very much ignorant about all of the operations aspects of this stuff. Is it really just the way it’s done and some signaling software that makes the difference? If so, besides dumb reasons like it being “different from how it’s done here” or unfounded national pride, why don’t other places just copy wholesale?
Because there’s a hierarchy of who looks down on who. Northern Europe refuses to admit Southern Europe does anything well other than food and music. And Anglophones have a really hard time learning from the rest of the world, because the sort of generalist US/UK elites who would lead any such mission outearn their counterparts abroad and therefore assume the rest of the world is too poor to learn from, except maybe Singapore, Hong Kong, Dubai, and other such places that overpay US/UK elites out of cultural cringe.
Regardless of looking-down hierarchies, Shinkansen operations are a special case, and deliberately engineered to be so, fundamentally by not interacting with other traffic. This non-interaction is famously so exactly at the point we’re discussing, in Tokyo, where two different stub-end Shinkasen terminals lie directly adjacent to each other, without any sort of track connection, let alone shared platform tracks, let alone through-running.
While direct transfer of JR Central/JR East Shinkansen operating procedures might not be possible to the very mixed-traffic infrastructure of most of Europe (aside from Spain, gauge changing trains and all), UK HS2, for all its accomodation of “classic” UK trains vs “captive” HS2-only shuttles, is very much segregated traffic.
Once a train enters the HS2 infrastructure, it’s headed to Euston (or maybe Birmingham), with no interaction with the “classic” network, and it absolutely must play by HS2’s rules. Could those rules and procedures be more Shinkansen-like? Seems worth considering! Could those rules and procedures squeeze more passenger throughput out of of given amount of nose-bleed-expensive CBD real estate and insanely-expensive Anglo civil engineering? Sounds not impossible.
PS It’s not as if the Shinkansen operations people must relish being forced to be world-besting despite constraints. I have saved somewhere a JR peer review of the proposed San Francisco California mainline stub-end rail terminal where the reviewers said that it would be desirable if feasible to engineer a grade-separated approach throat to allow more conflict-free arrival/departure train reversals and higher throughput. The right “concrete” in the correctly chosen places can indeed be very helpful, even if heroic levels of “electronics” can work around many things.
While they might not have done so in terms of transport Britain is definitely in general prepared to learn from Scandinavia for example.
And while parts of the Westminster publicly educated “blob” might not be prepared to learn as much from other countries I think plenty of other people aren’t.
The shadow transport secretary doesn’t exactly scream English establishment in her official portrait – https://en.m.wikipedia.org/wiki/Louise_Haigh – and has some interesting opinions that show she isn’t stupid (or innumerate). And the shadow rail secretary is a Sikh.
Well Tokyo Station kind of gets by because they’ve got a depot just past Tabata to stage trains starting up for the day from on the Tohoku, Joetsu, and Hokuriku. Likewise, there’s big depot in Yashio where trains stage out of for the Tokaido. I watch them all day long going in and out parallel to the Tokyo Monorail from my office window.
2 platforms for the DC lines, 4 for the semi fast local trains and 4 for the long distance classic trains that will still use Euston – including the two Scottish sleeper services. So that only leaves 6 platforms for HS2 and everything is overall pretty tight. So you probably need some extra platforms on top of the 16 they already have.
The cheap solution would be to terminate in Paddington as then you could move the semi-fast trains to Oxford and Bedwyn as well as the Heathrow express into the Crossrail tunnels as Crossrail trains.
Then you’d only have the Hereford, Cheltenham, two Cardiff, two Bristol and two south west trains needing to terminate in the classic Paddington station. So if you could use 6 platforms for the 8 long distance trains per hour that’d free up 7 platforms for HS2.
And you could probably build a few more beyond platform 1 with relatively little fuss.
Terminating HS2 in Paddington? Then you would lose the walking connection to the continental trains at Kings Cross. It’s a bit of a hike, but certainly less than at your average airport.
On the other hand you gain a direct connection to Crossrail – and you shouldn’t need any more tunnelling. It’s a trade off 🙂.
You already have a direct connection at Old Oak Common though, so this doesn’t add anything
I’ve been convinced HS2 was never going to be completed its too expensive. Honestly though I though that they would ditch the north-of-birmingham bits first!
Seriously Alon can you do some interviews starting Dennis and Elledge! I know your busy and TERF island is terrible but somebody need to actually tell people a lot of this is just incompetence and there are ways to solve at least bits of it.
Seriously it tells you how stupid the government is they wasted all that money on a Chilterns TBM tunnel when they could have done a Euston-St Pancras one.
Curzon street is totally cursed, that area is all not-especially valuable industrial. Dig a trench that sell the space above for high rises instead they had to go back to the future with a ridiculous Neo-Victorian roof (which is something we have plenty of in England).
I’ve debated costs with Dennis on Twitter and he won’t accept it. He still supports northern powerhouse rail even though he accepts it is highly unlikely to have a positive cost-benefit ratio.
And doing northern powerhouse rail is unlikely to bring any swing districts along for the ride given its city centre to city centre with few other stations.
Oh god! Finally somebody else who agrees with me NPR is stupid. Electronics before concrete etc. Yeah improving service on the Northern city peripheries is the no.1 thing that rail could do in swing districts. (Trafford especially since its not only trending Labour but is Yimby unlike Stockport and Bury)
Dennis is bad, but he gets on the BBC etc a lot to talk HS2 because he has a youtube channel.
On the comparison between HS2 and the Tohoku Shinkansen, I wonder how differences in service culture, operational norms, and rolling stock design impact turnaround times.
1. Food service on the Shinkansen is limited by European standards: none at all on many departures, a barebones trolley on certain fast services, and Gran Class refreshments on Hayabusa and Kagayaki trains, the expectation being that people who want food and drink will just get ekiben/takeaway at the station. As such, there’s no kitchen or buffet car to clean or restock, merely a quick exchange of empty and pre-loaded trolleys.
2. Handovers between Japanese train crews are rapid: personnel for the outbound service show up on the platform before the inbound service arrives, and the switch occurs while the passenger cabins are being prepped; no need for drivers or conductors to walk the length of the trainset or wait a while longer to be relieved.
3. I’ve often seen individual seat-by-seat litter bins on European passenger trains. Those are presumably emptied during at least some turns, and the number suggests a time-consuming task. On Shinkansen stock, the bins are consolidated in the vestibules, and there’s a culture of taking trash with you, which probably simplifies matters.
4. Regarding onboard toilets, even if the proper cleaning takes place before and after hours at the depot, I assume some freshening-up might happen at the platform during turns. Shinkansen trains tend to have one cluster of washing facilities for every two cars. European passenger trains seem to have a higher ratio, typically one or two WCs per car, which again increases the workload.
1. No need for food service on a 71-minute London-Manchester trip.
2. This practice can be learned
3/4. With 16tph, you can hire a lot of workers to parallelize the cleaning, and there are so many trains that they won’t be sitting idle in between trains.
The real barrier are the two sleeper trains which realistically need a platform each.
So that means the minimum number of platforms assuming 6 for high speed trains would be 12 if you need 4 for the other trains on the fast tracks, 4 for the other trains on the slow tracks and 2 for the DC line.
This means 18 platforms would probably have been sufficient – but given they only have 16 now that probably isn’t.
Why are the sleeper trains even going to keep running after HS2 opens? The trip times to Scotland are going to be so short that only day trains are viable. But more fundamentally, you shouldn’t build an extra platform at a constrained location for a daily night train.
I want to know why sleeper trains need extra platforms, That sit there all day long including two rush hours, doing nothing.
I don’t get why the sleepers need to spend the whole day at the platform. In most stations in Europe sleepers are taken to the yard after unloading, and are serviced there. Anyway, could the sleepers maybe move to Paddington? Crossrail should have freed up some capacity there. And you’d have all three UK sleepers in the same station, which might make servicing more efficient as well…
They don’t sit in the platform all day but do sit there for about 50 minutes. Plus you can’t use those trains for daytime service so you need to try pretty hard to get everyone off the train.
And they might be asleep or having sex so you have to be a bit careful and can’t do it instantly. Perhaps you could speed it up a bit – especially after HS2 comes in and there will be more capacity out of Euston – but there are no magic wands.
HS2 without the Goulburn Link will probably allow you to get to Edinburgh or Glasgow in about 4 hours. It’s not an enormous improvement on the status quo – especially for the highland sleeper which is logistically the more annoying one. And those times assume you can change the stopping pattern north of Crewe which may or may not be possible without more passing tracks for freight.
Besides there are three groups who like the sleeper – none of whom the government wants to annoy.
* Business people who live in the Highlands – because the sleeper allows you to come to London for business with a day elapsed.
* Scottish MPs
* Upper middle class environmentalists
You can do this at the yard. Build a public platform there for the oversleepers, and you don’t have to wait at the prime downtown location.
Maybe you can even build separate sleeper platforms just outside of the main station canopy like the old Scarborough Londesborough Road station or the current tracks 1 and 2 east of Roma Termini (binari 1 e 2 est)
Although the four northbound tracks in Tokyo Station can turn 14 trains per hour, it rarely but regularly happens that there are 15 or more trains coming from Omiya. What happens then?
Easy. JR East just turns the additional trains at Ueno.
Similarly, if you can’t turn all the trains in Euston, then you turn them at Old Oak Commons. Maybe you will need some additional side tracks or stub tracks but it will be much cheaper than building them at Euston.
Also there’s 6 platforms at Omiya giving further turning capacity.
So to provide the same turning capacity as Greater Tokyo you’re looking at needing a minimum of 6 at Euston and 6 at Old Oak Common.
Whereas Ueno is actively used to turn trains (albeit only two per day outside of peak season), Omiya’s turning capacity is mostly a historical holdover and for contingencies (even during the recent New Year rush, there were zero services scheduled to turn there), so full emulation isn’t really necessary to get equivalent capacity.
But contingencies are important too. You can’t terminate trains in Birmingham if you have a medical emergency at Euston.
By their very nature, contingency setups don’t have to be optimal, simply the minimum required. Emergency provisions can even be less expensive than a fully furnished station platform, like a set of pocket tracks with catwalks. I think your estimate of 12 actual platform tracks (6 Euston, 6 Old Oak Common) is probably sufficient – indeed, it already exceeds the capacity on the other side of Tokyo (the Tokaido Shinkansen’s 6 at Tokyo and 4 at Shinagawa).
Also, I don’t see how a medical emergency would cause a major snarl. I’ve seen it happen with some regularity on the Shinkansen, and it usually incurs a recoverable delay for one or two trains.
So they’re using “medical emergency” there too, huh, to drive immense amounts of waste? You don’t say!
It’s almost like there’s some trans-national (and mono-lingual) conspiracy or consultancy or something.
Anyway, you can’t be too careful where medicine (or fires, another favourite ask-no-questions cost driver) are concerned. Best to massively over-build and under-perform, just to be on the safe side. It’s what the doctors ordered, after all.
I mean to be honest I don’t think 18 (or even 23 as in the HS2 plan) platforms when 16 is the absolute minimum is massively overbuilding. Especially as you’ve got to sell it to the politicians.
Could someone well connected sell 10 platforms for normal trains as per the above? Sure. Could they sell 6+2 for the high speed trains plus the sleepers, perhaps in two independent sets of 3+1? Sure.
And that would mean that while your railway operations would need to improve the police and medical workers could continue to handle issues at their own pace – and you’d have some flexibility to deal with railway problems too. And you’d be “working towards” Japanese style performance.
The politicians don’t know anything about platform efficiency but do care about keeping the budget down.
The politicians have to get stuff past the allergic to risk civil servants.
And the civil servants will (rightly IMO) argue that Britain can’t manage Japanese style logistics in full.
“Medical emergency”: two, related problems.
1) Usually called “the Copenhagen interpretation of ethics”: If you can be proven to have observed a problem, any bad outcome is your fault. The best way out is to pretend to be (or just be) hilariously incompetent so that you can plausibly say “we never thought that X could happen”. Failing that, “we did everything we could” to a degree that your accusers can’t point to a believable (though stupid) way that *might* have prevented the bad outcome.
2) One death is a tragedy, a million deaths is a statistic. It appears that a lot of people tend to think of stuff like this by, in effect, imagining one representative example and ignoring numbers that overflow their ability to compare to other things. Thus if you inform them that saving one heart attack patient by changing what operating patterns the physical plant enables would cost a a zillion passenger-hours (and/or pounds) they blank on the latter and basically truncate it to the largest value they can keep in their minds, and then say “oh, absolutely, I would gladly wait for ten hours and pay a hundred pounds to save a (friend’s/relative’s?) life, and you are selfish to the point of evilness if you wouldn’t”.
That remind me Taiwan rejected through running high speed train to central station of Kaohsiung despite they started building a new tunnel linking to the station there for various reasons
Isn’t Barcelona’s Cataluyna 32 TPH (12 TPH on S1, 12 TPH on S2, 8 TPH on L7) on 4 tracks? I still don’t know how this is possible.
https://pedestrianobservations.com/2021/09/24/paris-world-capital-of-expensive-regional-trains/
https://www.fgc.cat/nou-servei-barcelona-valles/
https://www.trenscat.com/metrovalles/linial7_ct.html
I can’t tell whether this project actually increased capacity-(they slightly extended tracks and platforms, and started having automated reversals) (https://www.naciodigital.cat/latorredelpalau/noticia/76687/video-aixi-remodelat-estacio-fgc-placa-catalunya-incrementar-frequencia-trens, https://www.diaridesabadell.com/2020/01/16/obres-placa-catalunya-sabadell-frequencies-fgc-ferrocarrils-obres-millora/) they simplified service patterns so S1/S2 trains were all-stop in December 2022. The S5/S6/S7 were eliminated (see the GIF at the bottom-https://www.trenscat.com/metrovalles/linies_ct.html)
There had previously been a proposal to add relay tracks to get to 40 TPH (https://web.archive.org/web/20180723212310/http://premsa.gencat.cat/pres_fsvp/AppJava/territori-sostenibilitat/notapremsavw/84977/ca/dptop-sotmet-informacio-publica-nova-cua-maniobres-dfgc-placa-catalunya.do, http://www.elpuntavui.cat/article/1-territori/12-infraestructures/446196-la-cua-de-maniobres-de-fgc-a-placa-catalunya-no–fara.html)
I thought it was still 24 but with plans to upgrade to 32? But I haven’t seen timetables, I rely on what Roger Senserrich/Egocrata tells me.
The timetables show 5 minute headways for both the S1 and S2, and then you have 7.5 minute headways for the L7 (https://www.fgc.cat/nou-servei-barcelona-valles/) The S1/S2 timetables are here.
I am not familiar with rails in Spain, but aren’t those lines rapid transit? What’s impressive about rapid transit turning 32 tph over 4 tracks?
It is a commuter operation that has become more of rapid transit. The L7 is listed as a metro service, while the S1/S2 are listed as suburban/commuter services.
HS2 phase 1is vastly over engineered, for two reasons. Rich Nimbies in the South, who aren’t getting a station and don’t live near the the main lines that are getting congestion relief. So not just excessive tunnelling but arge scale embankments. Some have joked you wont be able to see any countryside through your window between London and Birmingham. The department of transports strange desire to transfer all risk to the contractors. We are talking about less than few cm soil movement over 40 year periods. The obvious reaction is gold plate that sucker so no possible liability can ever occur. They finally realised there mistakes on this when it comes to phase 2 onwards but it is too late for phase 1, the most expensive phase.
The seems to affect all British infrastructure. The lower Thames crossing a new 21km motorway and tunnel is expected to cost £8 billion. The environmental and design review process has now run to 150,000 pages and cost £289 million. The planning inspection is only now about to start.
This article has some inside ball..its paywalled though.
https://www.thetimes.co.uk/article/how-much-hs2-cost-money-spent-uk-2023-wvz9hdnz3
In Mainland China, intercity lines like this often benefit from having multiple terminals to avoid capacity constraints, such as the Nanjing-Shanghai Intercity Railway, Shanghai to Huzhou High-speed Railway and so on. The fastest trains on HS2 could terminate at Euston, occupying 4 to 6 tracks, and the remaining slower ones could be redirected to Paddington, thus eliminating the need for huge reconstruction at Euston. I agree with the notion that you can shove everything not-too-long-distance into Crossrail at Paddington to free capacity, since it would be time for a major rolling stock and signal upgrade when HS2, in its entirety, opens.
Shanghai Hongqiao has T H I R T Y through tracks, so capacity is not an issue.
The problem in China is that many newly built HSR terminals (like Hongqiao) are quite far from city centre. That means shorter-distance HSR won’t have that much of an advantage when we consider city centre to city centre trip time. The reasonable thing to do, and China is indeed doing, is to reroute these short haul trips to the old central stations in the city centre.
It’s Shanghai Railway Station that has a capacity problem. Some locomotive-hauled services that originally terminated there were even cut down to Kunshan to free up the platforms and allow intercity and high speed trains from Nantong, Nanjing and Beijing to get to the central station.
Previously the capacity problem could be solved by rerouting some trains to Hongqiao, but with the increase in demand of trains that terminate in Shanghai, the only solution is to remove the slow, unprofitable services.
Shanghai Railway Station doesn’t have a capacity problem. It’s CR’s obsolete operating practices, inherited from the steam era, that created these sorts of problems. Since JR central managed to turn back 16 high-speed EMUs per hour with just 6 terminal tracks, there’s basically no technical hindrance to operate more of them into Shanghai Railway Station.
Even if Shanghai Railway Station was at capacity, additional MUs / local-hauled trains could still run through the station, then turn back at Shanghai Coach Yard.
Shanghai Ralway Station isn’t even THAT centrally located. Maybe they should extend the line to a more central location (People’s Square or Lujiazui) and from there to Shanghai South – a through running RER type project. More capacity as well as better service.