Comment by habosa
1 day ago
Comments so far are not discussing what makes this light rail “very” light, so here’s an excerpt. The project claims to cost half of “normal” light rail.
> The vehicle is battery-powered, eliminating the need for overhead wires. It features an innovative turning system, enabling it to handle 15m radius curves. This allows for installation in tight corners within the existing highway. The Council intends for it to operate at a high frequency, providing a turn-up-and-go service. The vehicle has a capacity of 56, is comfortable and has low floors to enable passengers to embark and disembark easily. The vehicle has been developed to allow autonomous operation in future.
> The new track is laid just 30cm within the road’s surface, minimising the need to relocate pipes and cables, which is time-consuming and expensive. It achieves this by leveraging cutting-edge materials science, while still utilising standard rail parts to ensure ease of manufacture.
A 15 meter turning radius is tight, but not revolutionary. San Francisco's MUNI system's tightest turns are 45 feet, or 13.7 meters. The newer vehicles are designed for that, and the old PCC cars had to be modified to allow the trucks some extra rotation. The turning loop on Embarcadero near Market is that tight. There is much wheel screeching when a PCC car goes through that loop, because wheels have to slip to turn that tight. But it works.[1]
Battery powered trams have real potential, now that batteries with 5 to 7 minute charging and large numbers of charging cycles are a thing. That's compatible with typical end of line holding times. Steel wheel on steel rail is low friction, and you get most of the energy used to go uphill back when you go downhill. This could work out.
[1] https://youtu.be/XKN0MTCUSV0?t=265
Overhead power is more efficient and sustainable though. No batteries to chug along, no batteries to replace or manufacture. And I wonder if the cost over time really makes up for hanging a few overhead lines.
Several North American older systems go down to 10-11 meters, although perhaps the tightest curves are getting upgraded over time:
LRT systems on which the existing minimum curve radius falls below 15 meters include:
• Boston—10 m (33 ft) and 13 m (43 ft) for the Green and Mattapan lines, respectively;
• Newark—10 m (33 ft);
• San Francisco—13 m (43 ft); and
• Toronto—11 m (36 ft).
https://onlinepubs.trb.org/Onlinepubs/tcrp/tcrp_rpt_02.pdf (1995)
> Steel wheel on steel rail is low friction, and you get most of the energy used to go uphill back when you go downhill.
If you were going up and down hills, would you still use steel wheel in steel rail unless you had some sort of cable to work with? I always thought the Muni did relatively level routes for that reason? The Lausanne m2 for example uses rubber (well, ideally you’d be able to just balance the train going up with the train going down, but that only works for simple inclines with limited stops). Actually, a battery powered rubber wheeled tram service on some sort of steep incline like SF’s cable car routes could get some wicked regen going down.
Even if level, they could still get some regen from making stops.
Modern speed control technology has expanded the incline range for steel-wheeled trains quite a bit. Inclines that would have historically pointed towards rubber-tired or non-traction systems are usually within the range of steel wheels with solid-state motor control. Basically the control of torque is much finer than in old resistance-box parallel/series speed controllers, so you can avoid slippage much more easily.
> would you still use steel wheel in steel rail unless you had some sort of cable to work with?
A rack rail is also an option, though tends towards the noisy and slow.
But yeah usually light rail keeps under 5%, and can’t really go above 10 on pure adhesion.
The J Church line on Muni is still a train in part because back when they were converting lines to buses, the hill on Church St was too steep for buses.
1 reply →
muni has "relatively level routes" because the routes that were preserved were ones with tunnels that buses couldn't fit through (or narrow ROW in the case of the J), and given sf geography those tunnels invariably go through hills. muni, and especially the J, is one of the steeper adhesion railways in the world
Problem with rubber wheels in metros is absolutely atrocious air quality. I avoid them like the plague.
11 replies →
Maintaining a quiet operation is listed as one of the advantages of the turning system for this system.
> A 15 meter turning radius is tight, but not revolutionary. San Francisco's MUNI system's tightest turns are 45 feet, or 13.7 meters.
Revolutionary as the turning circle will be used at speed with passengers to traverse roundabouts in-lane. CVLR doesn't need turning loops, as the vehicles can be driven from either end.
It’s fairly obviously designed to avoid the issues which almost caused the cancellation of the new Edinburgh tram — spiralling costs caused by the need to move existing utilities under the deep track base. That crisis was probably as much to do with a badly formed set of contracts as with the technical issues themselves, but it’s still worth designing out.
Avoiding relocating utilities is only really a stopgap until the utilities reach end of life at which point you would be ripping up the newly installed trackbed during the middle of its life.
At least in the US utility relocation also generally involves moving what was underneath to the side, so it can be accessed without disrupting the new transit line.
You can leave what's there and run new ones.
Contracts that lock in a waterfall process.
At last they were not trying to use agile!
I do wonder how much of the backlash against agile is driven by people who never experienced waterfall
4 replies →
If you want a real tram that is compatible with existing street geometries, the really nice Škoda ForCity Smart Artic tram was developed to meet the requirements in Helsinki, which wanted a smooth modern tram even though there are 15m radius curves and 8% inclines in the old parts of the network. https://en.wikipedia.org/wiki/%C5%A0koda_Artic
Heavy and Light Rail in the UK is also a safety distinction: Light Rail is limited in speed, drives on sight (exceptions apply) and a few other restrictions (https://www.orr.gov.uk/about/who-we-work-with/railway-networ...).
Trams are one example of Light Rail, but so is the Docklands Light Railway in London (an exception to "on sight" - it's automatic), as opposed the the Tube (underground) which is Heavy. But the Welsh "metro" project's "tram-trains" (Stadler Citylink) are also Light, even though their tracks into the valleys are very much not urban in the usual sense - the full valleys journeys are over one hour with the current trains and go through mostly rural areas with small towns.
These things are tiny! I've traveled in larger airport shuttles.
It feels like that's putting this into a really awkward place in the tradeoff space. Trams work because they can scale higher than buses. That scale comes at the cost of more up-front infrastructure, much less flexibility, and needing dedicated lanes. So cities don't have trams everywhere, but they're only installed on routes that can support the scale.
For these you still have the up-front investment (just less of it) and inflexibility, but don't get the efficiencies of scale due to how small the capacity is.
Is this really just a bet that they can get autonomous tram-driving on city street approved a decade+ sooner than autonomous buses?
Trans don’t need dedicated lanes, not sure where you got that idea from. Trains do.
Buses however are slow (in London about the same as walking) and (outside London) prone to vanishing on timetable changes. Closing a rail link is tricky, you can be confident that if you live near a tram stop it will be there in 10 years. 60% of our local (say 10 mile radius) buses have been removed in the last decade, removing entire villages from service.
A rail solution allows you to read, a bus throws you around everywhere and makes you sick.
Buses are considerably faster than walking, no? Eg 8 to 9mph or about 13kph in London on average[0].
I also observe that this is an average speed, which night be useful for statistical summarization but is not as useful as knowing whether the portion of the route that you want to take is in the faster part or the slower part of the data readings.
For example, if I took a bus from Aldwych up Holborn to Euston I might expect that the first mile would indeed be walking pace but the second mile I would be zipping along. It's important as a bus rider to not let the slow parts color your perception of the whole ride.
[0] https://www.london.gov.uk/who-we-are/what-london-assembly-do...
2 replies →
Buses also put a lot of weight on the road surface. Even more if you fill the bus with batteries. If you can reduce road surface wear at a cost of an upfront investment in installing these rails that could be a good trade off.
Trams can be removed too. Bristol used to have trams. I doubt it is alone.
I definitely feel like trams are a weird technical solution to a policy/perception problem. On a technical level I don't think there's that much to recommend them over buses with bus lanes. It's just that governments never put bus lanes the whole way like you are forced to do with a tram.
1 reply →
>A rail solution allows you to read, a bus throws you around everywhere and makes you sick.
I got exactly the opposite impression the first time I rode a tram in my life. The tram is really really shaky and the connection with the overhead line is flaky, leading to all sorts of strange noises.
1 reply →
I think the up front investment is quite literally the problem this is solving for. If it weren't, you would just use light rail.
The ability for the tracks to be laid so shallow is in my view, the entire innovation and cost is the reason for this approach.
Tiny might make sense if they are running every 2 minutes and thus getting their capacity via frequency. However there is no reason to think they will do that. (if they were running anywhere near that frequent overhead wire would be a lot cheaper than a battery on every tram)
Yeah but you could do that with a bus today without miltiions in infrastructure spending
16 replies →
I didn't really mean that they needed higher capacity. If they had the passenger volume to justify such high intervals, they'd already have real trams.
But rather, this is giving up the benefit trams have over buses, without gaining any new edge to replace it. So why is it a good tradeoff? And why now, not 20 years ago?
The autonomous driving angle is the only idea I have.
7 replies →
That is in the article. The intention is a frequent, arrive and go service. Maybe every 2, 5, 10 minutes, whatever the actual details will be, that is the goal.
Trams can share lanes with normal car traffic. There's still a massive cost in terms of infrastructure (especially the overhead lines, utilities that need to get out of the way for the rail base, that sort of thing), but this project has a detailed description of why those aren't a problem for this project.
The tram they show in the animation also very much has a driver in the front.
If they can deliver on what they show in their demos, I don't see why the size of the trams or the infrastructure should be a problem. All the expensive stuff has been thought about, the system barely takes up any extra space, and the system is capable of scaling up by just sending more vehicles into service.
Generally I'm in favour of this sort of project, but having lived in Coventry (albeit a while ago) I'm a little sceptical: it's basically just adding a lot of infrastructure cost to what were low frequency suburban bus routes (the actual centre of Coventry is compact and walkable). You can run regular buses with similar capacity on batteries too, and divert them more easily.
1 reply →
Trams don't actually scale higher than busses; the highest ridership BRTs have far more ridership than the highest ridership light rail. The key thing that makes it work is having a dedicated right of way. I expect busses get a bad rap as a scaled transit solution mostly because they have to share the roads so often. But it's indeed an advantage of light rail that it's a lot harder to make that mistake with it.
Those BRTs will have lower passenger/operator ratios though since trams tend to be bigger than articulated buses and are frequently coupled together for busier routes.
I definitely agree that the dedicated right of way is the main thing. It's why some of San Francisco's trams are so slow outside of the city centre (where they run in a tunnel) and why Manchester's trams are so slow through the city centre (where they run at surface level sharing the street with pedestrians.)
3 replies →
Buses have slightly lower capex but much higher opex from the perspective of the community. City street gets ripped up by buses, cyclists get caught under the bus and dragged for miles, children get asthma from the tire dust. It being easier to give trams priority at traffic lights and easier to make the tram autonomous are just added bonuses.
Lower flexibility is actually a feature when it comes to mass transit: People will build density along rail lines because they assume the town won't rip them up, making the rail line more valuable over time. A bus route can be cancelled the day after a disruptive mayor is voted into office.
I also don't see why you can't scale up the tram with additional cars, as long as you keep the lbs/sqft the same. 3 car trams are fine, 3 car busses are... not
Also higher OPEX from the operator's point of view - trams can carry a lot more passengers per driver, they're more efficient energy-wise, and the replacement cost for bus tyres is large compared to steel wheels that last way longer.
If higher capacity is needed you can always link multiple units together
Linking multiple units means making the stations longer. A more likely scenario is to deploy more units and deal with staffing by migrating to autonomous operation.
In CVLR, you actually cannot as the extreme bogey angles mean they can only operate individually. You also can't order longer cars.
You can run them at high frequency though.
3 replies →
I'd assumed the point of them was that you can take the several cars of a tram and split them up to have more frequent services. Though I suppose this would compound with the cost of having a driver on each car, potentially cancelling some of the gains from cheap installation. As for the point of automation, I think the tram can probably be a lot easier than the bus because of the human factor. It seems safer, so legislators will be more willing to legalise it and residents less likely to complain. Also, you've got rails in the road that clearly mark the route of the tram which make it more visible than an automated bus. Most of these automated taxi companies still have a human supervising the process, and I imagine that could be employed here to good effect and with fewer or faster manual interventions than would otherwise be needed.
Even if all that falls through, I'm not gonna complain about it. We sorely need more public infrastructure in the UK. Even if an experiment like this fails, at least you actually get a tram line and experience out of it. Much better than a project which sucks up million then gets cancelled. (Cough cough HS2.)
Buses on dedicated lanes are OK indeed. However, buses are simply not as comfortable as trams: roads unless in tiptop shape are not as smooth as rail, and bus drivers always take corners too fast.
Having to hold on to something discards it from my preferred list of solutions.
Have recently read Gareth Dennis’ How Railways will fix the Future…
It’s a worthwhile read BTW
I suspect these are too small to carry a significant number of passengers per hour
They’ll also probably never be autonomous as the challenge with autonomous is less the driving and more with passengers getting on and off, getting trapped etc
This is such a typical American sneer at public transport.
It's tiny, how it possibly carry all those 2x4s, powertools and sheets of plywood when I'm out doing manly things. I'd better go buy that monster truck so I can look like a real man.
This is a) an unnecessary counter-sneer at two whole continents and b) dismissive of something that would be a real problem in a city bigger than Coventry.
Those teeny tiny little carriages have a capacity a quarter of what the trams in my city provide. If one of them pulled up in peak hour here, I imagine it would fill up after two stops and be a nuisance from there on.
I guess you don’t need much space to dip your baguette in a cuppa tea
1 reply →
A normal public transit bus has twice the seats and doesn't need rails.
2 replies →
In many non-North American cities one needs tiny. Big stuff just does not fit.
Eliminating "the need" for overhead wires seems like a terrible trade off when you're adding "the need" for expensive batteries and charging infrastructure.
Overhead wires are not trivial. Charging is.
You're very wrong about that.
Definitely the high points of the story. And to emphasize — 15m corner radius!
Toronto has 11m loops. Newark 10m. Boston had 10m until a few years ago. This used to not be a problem anywhere.
Why is it easier for cities to develop obsolete stuff than to eg, run their own Uber?
What's the highest capacity vehicle Uber has ever operated? Because Tranist can move thousands of people with a single operator
In the whole universe of intellectually honest, valuable benchmarks for transportation, do you think ride sharing wins on zero?
I’ll give you an important one as an example: door to door journey times. I support RTO, which is the best way to improve that metric for the average person, which is to say, I was hoping this would be a discussion for out of the box thinking. Or really, what do you invest millions of infrastructure bucks into? All of Uber was only a little more expensive than a single high speed rail line. Why can’t cities run ride share? Why would they run ride share worse than a train service?
3 replies →
Uber and mass transit are an apples and oranges comparison.