A radically different type of rolling stock

[DJL]

So I had this idea a number of years ago and thought I'd throw it out there for criticism.

The basic idea is that all trains on a given route could become all-stations while remaining fast resulting in high frequencies at all stations resulting in shorter journey times across the board.

All trains would start at the maximum length possible at the most major termii on the route.
e.g. Charing Cross, Waterloo East, Cannon Street and London Bridge in the case of the SEML.

A number of the carriages near the rear of the train would be designated for the first station (call this the rear section) and, while the train is still moving at speed, would separate from the main train some distance before the station is reached and begin to decelerate.

At around about the same time a number of carriages (call this the new section) would depart from the first station and accelerate up to a speed slightly below that of the bulk of the train.

The bulk of the train would couple with the new section at speed.

Meanwhile the rear section comes to a stand in the station and passengers disembark.

This process would repeat for each station - with some stations having larger rear sections than others. The number of carriages in rear/new sections would not necessarily have to match - meaning the overall train length could change during the course of the journey.

There would probably be some overlap at some stations where 2 rear sections are separated before new sections are attached depending on station proximity and speed.

There are disadvantages of course:

* There are probably a lot of things that would need to change other than just the rolling stock - the signalling for starters.

* Each and every carriage would need to be able to power itself.

* After each separation / coupling some passengers would need to move through the train in order to be in the next rear section for the next station (or not)

* Timing would have to perfect. This would probably mean "doors close 30 seconds before departure" would have to change to "doors close 2 minutes before departure" to give time to iron out any kinks (read: idiots trapped in doors even though there will be another train in just a few minutes time) and/or instruct the bulk of the train to slow down!

* There would probably have to be some kind of "separation safety officer" who moves through the train to seal the doors on the section to be detached.

* It would probably become impossible, or at least very very hard, to travel just one stop on any given route (though I question why anyone would do that)

* Probably a great many other things that I haven't thought of.



Assuming the technical challenges could be overcome and an implementation of this idea was at least possible (probably on some brand new line rather than converting any existing line) - would people like it?
Would the benefit of a fast frequent service outweigh the pita of having to constantly move through the train? Or would people prefer to sit in a comfy seat while the train repeatedly speeds up and slows down?

 

[Moonshot]

Why not just do it the old fashioned way in the manner of travelling post offices from years ago....ie stick passengers out of windows in sacks and then catch them in nets at the end of the platform? . Would actually mean trains dont have to stop at all !!

 

[edwin_m]

In steam days there were sometimes "slip coaches" that were uncoupled on the move from a passing express, with a guard on board who braked them to a stand in the station using the vacuum stored in the on-board reservoirs. And I believe the Danes intended to use coupling and uncoupling on the move with their IC3 stock (the ones with the big rubber doughnut round the cab - possibly this was to provide some cushioning during the coupling?) but I don't know if they ever did.

Coupling on the move in particular would be fraught, as it would need some kind of vital control system to match the speeds and make sure that the rear part didn't collide violently if the front part had to slow down suddenly. There has been at least one fatal collision where a banked train gained traction and pulled away from the uncoupled banking engine, and then re-collided when the train slowed down again.

 

[jopsuk]

You're far from the first person to have had this idea! Indeed, in the steam days this was half implemented- there were services that ran with what were known as "slip carriages". They had a coupling that could be safely undone in motion, and a set of braking equipment operated by the guard. THe carriages would be "slipped" as the express approached the station and would coast to the platform- I think sometimes with the signalman even diverting the slip coaches off the mainline?

So, from there, there have been many proposals over the years for how you could go about reversing that process.

 

[matchmaker]

Good vintage article here:

http://mikes.railhistory.railfan.net/r134.html


"SLIP-COACHES were invented comparatively early in the history of British railways. Early in 1858 the then London, Brighton and South Coast Railway made the first experiments on record with slip-coaches, dropping a Lewes portion from the rear of a Brighton express at Hayward's Heath. By December, 1858, the G.W.R. had also begun, and the first G.W.R. slips were detached at Slough and Banbury. In 1869, on the Bristol and Exeter Railway—now part of the Great Western—the "Flying Dutchman" began the practice of slipping a coach at Bridgwater. Since then the Great Western Railway has always been partial to the device. Before the war of 1914-18 some of the other important British railways had more or less numerous slip-coach services ; but in recent years they have almost all disappeared except on the Great Western, and one or two on the Great Central and Great Eastern sections of the L.N.E.R. Even on the Great Western their popularity is less than it was."

 

[Mojo]

This idea has sprung up on here, and other websites a few times before.

In addition to the technical issues, on a more practical basis I can think of why it would not work.

1. How long would the train be? At say a maximum of 12 cars if I wanted to go from one station to the next one along the line it is unlikely I would be able to walk the entire length of the train to alight at the next stop.
2. If the last car would 'slip' off at every intermediate stop then quite a lot of customers would always have to move cars. Let's say that I was going from London to Dover Priory then there are potentially around 24 stations that would lose a car so I'd have to board at the very front of the train and then continually keep hopping between cars after 11 stops.
3. It doesn't take into account that some stations are busier than others. For instance I'd imagine that Orpington or Sevenoaks would require more than 1 car to convey all their customers, yet a station like Pluckley probably doesn't need even 1 car off-peak.
4. How would it work with First class accommodation?

 

[broadgage]

The idea has some merit, but the practical problems of ensuring that it can be done safely are substantial.
Detaching coaches at speed is relatively safe, as posted above it has been done in the past.
Attaching is potentialy more dangerous, and AFAIK has never been done on a regular basis.

I cant see it being viable for very frequent station stops for the reasons given previously that passengers travelling a short distance might not have time to be in the correct portion.

One could PERHAPS see the merit of say a West of England service leaving Paddington, and detaching say 4 vehicles for Reading without stopping.
Then West of Reading, another 4 vehicles could be attached at line speed, these vehicles having left Paddington earlier and called at all stations.
That could provide a faster service, and also a through service from say Slough to the West.
Similar arguments could apply each side of Plymouth, drop say 4 vehicles for those alighting at Plymouth, and then West of Plymouth, attach at line speed 4 vehicles that started perhaps from Taunton and picked up at all stations.

Sounds as though a great deal of expense and complication would result, and in present circumstances I would prefer that money be spent on longer trains, and on minor speed enhancements.

 

[Moonshot]

How would this work with absolute block signalling? Wouldnt 2 trains be detected ?

 

[WestCountry]

Someone proposed a similar solution for Tube trains a few years ago - except their concept split longitudinally down the middle!

One side keeps moving constantly, the other side splits off, stops, then joins the next non-stop side to go past.

Keeping the two sections together when joined (esp. on curves) and reliably separating them would be a pain though. What if there's something stuck in the intermediate doors just before they're meant to split?

Link to New Scientist article (sadly semi-paywalled).

 

[ainsworth74]

A suggestion so radical it was first thought of 150 years ago

 

[matchmaker]

A suggestion so radical it was first thought of 150 years ago

Indeed! The article I linked to above itself dates back to 1935

 

[DJL]

3. It doesn't take into account that some stations are busier than others. For instance I'd imagine that Orpington or Sevenoaks would require more than 1 car to convey all their customers, yet a station like Pluckley probably doesn't need even 1 car off-peak.

Actually I had thought of that...

This process would repeat for each station - with some stations having larger rear sections than others. The number of carriages in rear/new sections would not necessarily have to match - meaning the overall train length could change during the course of the journey.

4. How would it work with First class accommodation?

I Didn't think of that though

Sounds as though a great deal of expense and complication would result, and in present circumstances I would prefer that money be spent on longer trains, and on minor speed enhancements.

I would actually agree with that - certainly in the near term.
Although consider that with all trains calling at all stations you would naturally get an increase in capacity at the busy stations with trains every few mins (or maybe even more if multiple platforms are available) and these train sections needn't necessarily be short - they would probably be a % length of the overall train based on the expected % patronage for that station relative to the entire route.

However I agree that there are far cheaper ways to increase capacity in the near term.

How would this work with absolute block signalling? Wouldnt 2 trains be detected ?

Yep - Everything would have to be resignalled as both block and modern* systems would not work with this type of train (maybe the detach but certainly not the attach)

* sorry - I am assuming that all block systems are old ones that have yet to be replaced and I don't know the name for the newer systems. Apologies if this is wrong.


I like the longitudinal idea as it reduces the problems of moving through the train - however with many many more doors to open and close I can see problems occurring...


Sorry for the long reply to everyone in a single post (I'm new here - am I better off splitting and doing several posts in a row?).

 

[D841 Roebuck]

Back in the 1830s, the London and Blackwall Railway had a system rather like this. It was rope powered, and each vehicle had a guard who would release the rope at the appropriate station.

It didn't work very well.

 

[DownSouth]

And I believe the Danes intended to use coupling and uncoupling on the move with their IC3 stock (the ones with the big rubber doughnut round the cab - possibly this was to provide some cushioning during the coupling?) but I don't know if they ever did.

The "rubber doughnut" is part of the end corridor connection system that has absolutely nothing to do with dampening bumps, it is purely for sealing the connection. The system allows for both a generous-width connection when coupled and a full-width driving cab which swings into place when not coupled.