In the GD thread on HSR, it’s noted (correctly) that stops are a problem. Lots of people may want to embark or stop along the route, but stops waste both time and energy.
So how about a scheme where cars separate from the train? Instead of actually stopping, the train slows (from, say, 250kph to 100), and the rearmost car separates, is shunted onto a separate track, and coasts into the station. The main train continues past the station, at which point another car that has recently left the station is accelerated and joins the rear of the train. Repeat as often as necessary along the route.
It’s clear that it would be challenging to work out the details of accelerating the joining car. I can imagine it being done roller-coaster style, or perhaps with a large flywheel.
The fact that the main train has to slow down (since joining at full speed is probably impractical) represents some time and energy inefficiency. But this would be much better than full stops, and could give HSR benefits to any town along the route.
This came up in one of the other high-speed rail threads, and I think there were links to a couple proposals, but I don’t think it’s ever been in operational use.
For one thing, all the HSR trainsets I’ve seen have specific cars on each end, and the whole train is blended together pretty well for streamlining. So you can’t just drop any car you want. And there are probably mechanical, electrical, communications and pneumatic conections that run between cars, so you’d need an automatic method for making and breaking those connections.
Second, you’ll have to be schlepping through the train all the time. You get on at one station, the car shoots out of the station and hooks up with the end of a train. Now you’ve got to get out of that car and move forward, while people who want the next stop are trying to come back before that car disengages again. and if the disengage/reengage is always at the back of the train, either the lead cars will hardly be used, or some people will have carry all their stuff the full length of the train, twice.
What happens at a popular stop when more people want to get off than can fit in one car? If two cars detach, will that station have two replacements available? During morning and evening commutes, you’d have cars stacking up in some stations and not used again all day.
And most of all, the stops just aren’t that big a problem. Everybody knows when their stop is coming, and makes their way to the doors. Once the train stops, there are doors along the whole length, so you can get the old passengers off and the new ones on in about a minute. (It’s not like boarding an airliner.) And electric motors have oodles of torque, so you get back up to speed pretty quickly.
In short, you’re proposing a very complicated solution to a problem that barely exists.
“Slip coaches” were used in Britain in the steam days, but never widely.
An express train (traditional consist of cars with a locomotive up front) would have the “slip coach” on the back. The coach had specially-designed couplers and a separate, isolated braking system (normally, if a train not designed for this separates in motion, both halves of the train automatically go into full emergency braking because of the lost air pressure). At a certain distance from the station, a brakeman riding in the slip coach would uncouple the thing from the train, and manually slow the car down into the station, while the express train continued on its high speed way.
This system was never a roaring success, mainly because it was expensive. You needed to pay the wages of the extra brakeman, you needed to engineer the car specially, then you needed to pay for a locomotive and crew to come and rescue the stationary car from the main line. If this took longer than expected, other trains would be delayed. Then the car had to be returned to origin.
So this side of things is possible, but not worth the hassle (and probably wouldn’t pass the modern safety regulations anyway). It would be virtually impossible to manage it with the push-pull configuration of modern high speed rail.
As for accelerating a lone car up to line speed to join a train, forget it. Again, it would only be possible with a traditional loco-hauled train, rather than push-pull, but also the safety authorities would laugh you out of their office. It breaks a hundred and eleventy-three basic tenets of “safeworking” (railway safe operating procedures). It would involve a “train” (the single car) occupying the same section of line as another train, which is probably the most fundamental no-no. The driver of the accelerating car would then be facing all signals at danger (red) because of the train in front, and wouldn’t know if this was because of that train, or because of yet another in front of that one, in which case, the main part of the train is going to be stopping quickly, and it’d be faceplant city for our passengers in the joining car. Also, it would be extremely difficult to do smoothly, even if foolhardy enough to try.
Picture it: the car is powering up through 100mph or more and is closing in on the rear of the train (say 200 yards back). The driver can’t see much ahead because the train is blocking his view. Meanwhile, the driver of the train up ahead sees a semi truck blocking a crossing. He slams the brakes into emergency… not pretty.
Obviously, such a scheme can’t work smoothly or safely without a high level of automation (as I strongly suspect is required even for normal HSR operations). It would be absurd to even think about doing this with separate manual control of the joining trains.
The braking of the two trains would certainly have to be tightly integrated - just as if they were one.
And doesn’t HSR aim to have no at-grade crossings? Could the driver of a 250 kph train realistically expect to spot an obstruction, apply brakes, and get stopped in time?
It’s certainly possible to do this. But I really do believe it would be spectacularly dangerous, whether under automatic or manual control. Train brakes are a weird and wonderful thing. Even with modern trains, there is no escaping the fact that they are very heavy vehicles, and the braking characteristics of each train (within the same class or type) can be quite different from one another. It’s a common thing for events like platform overruns to occur when a driver is at the start of his or her shift, because of a “new” train. But there are other reasons, such as the signalling ones I outlined above, that would put the er… brakes on this idea before it got off the drawing board.
HSR should have no at-grade crossings, yes. However, there are fallen trees, cows, broken rails, sabotage, vandalism, and any of a number of things that might force a driver to apply hard braking. For the system proposed to work, you’d have to basically say, “OK, for the couple of minutes where we’re doing this, use of the brakes is not possible.” And I don’t think the general public would like travelling in any vehicle like that. I wouldn’t.
I guess an analogy would be air-to-air refuelling. It’s done for military types who live dangerously anyway, so it is certainly possible, but most folks wouldn’t like to fly on an airliner that has another plane almost touching it, and the aviation authorities wouldn’t countenance it for a moment.
Trains often can’t stop within the distance they can see. The signalling systems are geared to compensate for this. This is why they have more lights on each pole than road traffic lights - to tell the driver not only what this signal is saying, but also what the next one is going to show. But that’s built around a hundred and fifty years of a generally excellent rail safety system that is very good but would be pushed probably beyond its limits by this scheme. A train might not be able to stop within the distance the driver can see, but if he can shed a few mph before impact, then all the better.
Scary video of a freight train accident. The train was accidentally switched to an occupied siding. From the moment the driver goes to full emergency brake (when you hear the “ssshhh” of air) until impact, you don’t see much in the way of speed decrease. Train brakes not only have to deal with a heavy train, but they can be sluggish in response to the controls - which would also make coupling-at-speed schemes a bit awkward.
And don’t get me wrong, I love weird and wonderful rail schemes…
Surely existing HSR systems rely on something better than sharp-eyed drivers to deal with these things? What about at night?
Or, far better, while the separated trains are on the same track, their braking needs to be tightly coordinated - whatever the lead train does is automatically and exactly copied by the other.
There are safety measures in place. So that tree might not fall on the line because it wouldn’t be allowed to grow there, or that vandal might not do anything because of fences, etc. Some things can show up on computer screens and be dealt with. But even in this day and age, an awful lot relies of human intervention at the pointy end of the train.
That’s where the problem lies. It’d be hard enough to do that with two Toyota Camrys. It’s harder again with trains. First, you might have 600 tons of high speed train versus 50 tons of single car. Then you’ve got the fact that the train is longer, and braking response can take time to propagate down the length of the train (not so much of an issue with passenger trains, but it’s still there).
The scheme is possible, but I’d be very surprised to ever live to see it. And I wouldn’t ride on one (well OK, actually I might, because I’m a rail nut, but a lot of folks wouldn’t). But it does truly need an understanding of how train brakes (and acceleration for that matter) work to see that this isn’t as simple as it looks.
For dealing with obstacles such as fallen trees, you can put me down as skeptical that “see and avoid” can be effective at anything like the speeds of HSR.
I’m suggesting that this scheme will need to be computer-controlled; the computers will certainly need to know the mass and braking response of the trains involved. It should not be especially challenging to ensure that the smaller following train is always capable of stopping at least as quickly as the leader.
I wouldn’t argue that it looks at all simple - it’s obviously a complicated enhancement.
Speed costs money. No HSR system is cheap; the truly fast ones are hideously expensive. So a technique that boosts the achieved speed without increasing the maximum speed could be quite valuable.