Riding London’s unexpectedly fantastic Elizabeth line
newyorker.com
newyorker.com
The Elizabeth line (it was called Cross Rail during development) is the first train line that combines both the national rail infrastructure and the London Underground. Combining the signalling systems has be an incredible complex problem. There is quite a bit online to read about it:
https://www.wired.co.uk/article/crossrail-delay-logistics
https://www.crossrail.co.uk/project/our-plan-to-complete-the...
So the trains will need to seamlessly transition between TPWS on the national railway lines, then CBTC in the London underground, and ETCS when going to Heathrow airport.
I think this is what they call integration hell, and is responsible for delays of several years. When you consider the complexity, but also the safety-critical nature of these train signalling systems, it's no wonder it has overrun.
So the "integration hell" is trying to integrate between services with completely different service strategies.
Both have planned where every trains should be at any moment. The difference is the amount of leeway you have to fix operational issues.
You can think of the metro as a “timetabled” system where the density is so high you have no margin. When something happen you have to shift and cancel. Train lines start having exactly the same issues when they operate near capacity.
Both systems have the same constraints. Trains can’t overtake. You can only have one per block. The integration difficulty is that they don’t manage blocks in the same way and use different signalling. You see the same issue on lines which are shared between SNCF and RATP in Paris. The A line used to require a driver change at the junction.
There actually is a timetable for all of London Underground, it's needed to roster drivers. But the only services ordinary passengers actually care about are the first ones and the last ones, Amersham to Baker Street last train is 23:22, if you're out of the city too late for that, you're not getting a tube home. When/ where there is Night Tube you probably don't care, as most people taking Night Tube aren't exactly on a tight schedule.
However, the Underground is not run at capacity as you imply. There is deliberately wiggle room for "recovery" from events even at 30tph. This is one reason why the Circle Line isn't a circle any more, it makes it possible to recover more easily on that line.
Part of the underground is definitely run so close to that you can think of it as run at capacity. There is some wiggle room especially where it’s manually driven and during the slowest hours but nothing comparable to a typical train service. I am admittedly a lot more familiar with the French situation - I worked for the French national infrastructure manager on modernising traffic management and for TGV on their planning system - but some lines in Paris the block sec is so low even the slightest delay in doors closing will lead to cancellation and trains can’t be controlled manually.
EDIT: I guess this is the literal definition of "Civil Engineer"
(I know it’s a forum post but that old website is a wealth of information about the District Line).
No, it isn't. Historically the boundary between the Underground and the rest of the "mainline" railway network used to be even fuzzier with lots of through-running, but even today a few stretches of lines served by both Underground and mainline trains remain (the Metropolitan Line north of Harrow-on-the-Hill, the outer ends of the District Line to both Richmond and Wimbledon, the Bakerloo Line north of Queen's Park).
What's true though is that due to its capacity requirements, Crossrail is using a different signalling system than the rest of the railway network, which means lovely interface issues at the transition points…
The New Yorker sub-editors are notoriously anal about their house style, but you'd think the one who dealt with this article could have left it as "the edge of the City" rather than completely changing the meaning.
https://www.scmp.com/news/world/europe/article/3179455/londo...
Sometimes there is nothing wrong with this, but equally we have lost vast amounts of skills within our borders basically because privatisation is aimed at the wrong kind of efficiency.
Now you also have the option of doing Reading > Hayes > Heathrow entirely on the Elizabeth Line.
My biggest gripe is that despite the high price, you spend the entire journey being force-fed live news on the TV and PR fluff about how it's the best rated line in the UK (no surprise, it goes between two stations).
I never used these and I kind of wish I had, because after 9/11 they closed and never reopened.
[edit: Wikipedia tells me the check-in service ran until 2003 and "was withdrawn due to low usage and high cost of operation", nothing to do with 9/11. My mind must have filled in the more dramatic memory]
Then a few years later the Heathrow Connect service opened, a slightly slower, much cheaper alternative to the Express (it changed names a few times but I believe is the same service as now constitutes the Elizabeth Line on that stretch) - so nobody who was coming from Paddington, planning ahead, and paying their own fare had much reason to take the Express any more.
I expect Heathrow Express to be quietly abandoned in a few years' time. Network Rail and GWR would quite like the trains off the GWML fast lines.
There are ads and ticket sellers points of the Heathrow Express all over Heathrow's terminals. It's easy to accidentally pay for an overpriced train if you don't know there is an alternative that costs half the money and arrived in almost the same amount of time.
The last time I was in Heathrow I befriended a poor couple of Italians who bought a £25 Heathrow Express ticket because they thought they needed it to ride the £2.50 Picadilly Line.
The Heathrow Express was one of the worst legs of the system; having just regurgitated my airline breakfast, I was delighted to be sitting in a coach with no air conditioning in August.
I agree that the alternatives are woefully underdocumented. I recall the choice being framed as "well, you can go to the taxi stand, spend 75GBP to go the whole way to the hotel, or use the rail option and spend 20GBP on a taxi from Paddington.
In retrospect, what I should have done is taken the Underground from near Paddington to where I was staying (100 metres from Euston, so I think it involves about 8 train changes within 75m) and saved the difference.
It is a bit annoying to take from terminal 3 — they really push you to the express. But easy enough and I’ve done it a couple times when traveling on my own dime.
Total price ends up around £4-5.
There's a cohort of us that have to put our fingers in our ears on the approach to some stations as the grinding of wheels against tracks reaches an incredible volume. Pretty sure it must be illegal on some checklist somewhere.
Not to mention the raggedy state of most of the other lines. I'll stop before this becomes a full blown rant.
on a tube train there's nowhere to expel the heat other than the tunnel, which doesn't really solve the problem
(plus the trains are small enough as it is, without large AC units)
It's a lack of funding and vision, not technology ... which returns us to my original point that I wish the money had goone on the existing network.
which essentially isn't possible, so that would require building an entire new London Underground
this would cost probably a trillion pounds
Northern line had a much needed extension along to battersea
Picadilly line stations have been seeing a lot of elevator and escalator replacements.
Circle and district line new trains are magnificent.
Add in the tfl rail connections, the outer circle line of sorts that creates
The London Underground is effective, but exceedingly average. The Liz line looks excellent on first impression. Some of it is decent. Some of it is terrible. I'd prefer a more even distribution than a spanking new line ... or both. But I'm sure we sprayed the money on the overspend ...
Its like painting the Sydney harbour bridge, you start at one end and work your way to the other.
By the time you get to the other side you have to go back and start again.
I always found it bizarre that most of the press around HS2 never really mentions the capacity issues of the West Coast Main Line.
Most people will ask "why spend £100 bn to shave 30 mins off London - Birmingham, can't you just upgrade the existing lines?", which IMO is a fair question to ask. They need to do a better job of explaining why it's not practical to upgrade the WCML & why building a new line is the better long term solution.
[0] https://en.wikipedia.org/wiki/InterCity_250#Project_scope
[1] https://en.wikipedia.org/wiki/West_Coast_Main_Line#Modernist...
[2] https://en.wikipedia.org/wiki/West_Coast_Main_Line#Increased...
Trains on HS1 have TVM or equivalent, in-cab signalling. https://en.wikipedia.org/wiki/Transmission_Voie-Machine#Cab_... - you can see at a glance that the computer says you're going 269kph, the limit here is still 270kph, but the signals indicate the limit ahead will be 220kph so you should be braking.
But being realistic.... 90% of business, social, and civic activity is in London so that's where I need to get quickly.
And that imbalance is what we really should be trying to fix. Will HS2 make any difference? I don't know...
[edit to add: speaking as one who lives less than 60 miles from central London.]
It should be noted that 4 years ago the BBC were running documentaries about how successful the Crossrail project had been after the tunnels were finished and most of the stations were almost done. It seems that nobody expected the software and signalling to take an additional 4 years of time. Somebody somewhere really messed up that part of the project. Everything else - the main civil engineering project - did go extremely smoothly and was even ahead of schedule at times.
I frequently see one answer fail - we just need to plan it more thoroughly, don’t just estimate, go through exactly what you’ll need, identify dependencies now not during coding. These efforts always fall for the same reason, when it comes time to implement, the plan is hopelessly out of date, incomplete and just plain naïve.
It’s easy to deliver software on time up to around 10kLOC (i.e. single developer short project). Beyond that, the variables cascade and avalanche exponentially. The problem with this is all the interesting problems need >1dev and certainly more than 10kLoc!
On balance I’m more persuaded by XP than by Royce but honestly i don’t think either approach makes or breaks, both can work well.
I think the issue plaguing estimation of projects is that a lot of software development involves running into problems and trying to solve them. Sometimes figuring out solutions to these issues is difficult and solutions appear as if by magic (in this was the art analogy makes sense in terms of inspiration). I’m not sure how we can ever estimate large projects accurately while we often need these moments of inspiration in order to solve problems.
How do you do that though? If we’re only talking about nailing down requirements (so lets ignore the problem of dependencies for now), then a 300 page BRD buys you enough spec to gainfully employ 4 devs for maybe a month-ish? I’m going off fin reg style work, i have no idea how that might translate to civil engineering so my numbers might be well off?
The BRD wasn’t free to produce, it contains 3-9 errors (assuming human rate of error is 1 to 3 in 100), and it’s being updated to v2 as you’re working.
How is this getting us all to the overall goal with predictable timing?
The problem is that in real-world testing, conditions around the changeover points produced behaviour that was not addressed by the spec. It turns out that dealing with radio-based systems in the middle of a busy rail network in the middle of a city is really hard. Many of the problems were caused by radio interference in the area. It's definitely the sort of problem practically nobody on this forum works on.
I understand software and signaling took extra time, but those drove the costs too? Compared to the rest, they don't seem expensive.
Indeed. "Quite good" is usually considered the absolute pinnacle of fawning adulation.
(The former can vary with tone sure, but typically I think quite slating, as parent describes. Somehow saying it's 'really' so though turns it around completely, making it impressive.)
The second this is coming from Newyorker.
On the NYC DC stretch, scheduled “high speed” service is only 20-30 minutes shorter than the much cheaper regular service that stops at every station.
Wikipedia says 150, but I guess if you live there you probably know better.
135 to 150mph. There are plenty of YouTube videos of people independently verifying the speed either from the train using a phone speedometer app) or using a speed measurement device from alongside the track.
Here is one (go to timestamp 6:13):
The Acela schedules for nyc-dc last week were pretty damning. Shaving 30 minutes (at best!) from a 3 hour makes-every-stop trip clearly doesn’t mean significant high speed travel.
At earlier stages of my life I was a regular traveler on tgv/Thales/Eurostar, ICE, and Shinkansen. I know what a high speed train feels like.
In my experience the Acela is rather absurd. Comfy, sure; I do take it on the BOS-NYC route but in my experience it’s not worth the price for NY-DC. Lookimg right now at Amtrak app nyc-was, we have the regional 57 Vermonter taking 3:27 and the Acela 2259 3:04. I don’t think that’s worth paying extra for.
in the UK it's the opposite, freight trains get pushed into sidings so passenger trains can overtake
(but then they don't get charged for causing delays)
I take commuter rail every now and then which mostly doesn't compete a lot with freight and delays are, if not super-common, not rare.
Amtrak is very possibly worse than other systems in this regard but it's not hard to see how multi-day train trips could accumulate delays when sharing tracks with miles-long freight trains traveling at sometimes very slow speeds.
Other than that, having some padding and a schedule should get you two nines.
I imagine you have to strech the schedule or skip maintenance to get that sort of cascading failures.
We do have a huge freight network that moves a lot of tonnage, but I wouldn't call it modern or world-class in track infrastructure, speeds, signaling, train control automation, or electrification.
Leeds? Not so much. Daventry? Nothing. Whitby? One slow train every two hours.
There is a reason its the MTR (of Hong Kong) Crossrail Line...
Give them a train network that's almost as old as London's to run (example: Melbourne), and they are just as bad as any other operator.
My experience in many other Asian and European modern cities has been far better.
It's the same principle as the Parisian RER network ( suburban lines outside the city, new tunnels under the city center with few key stops) which started in the 60s, just with a newer and more modern ( platform screen doors, most notably) execution.
So there’s no signs to other tube lines. Just signs on how to get to the tube which feels weird because you thought you were already on the tube.
It would be better to explicitly spell out directions to what tube lines are available.