Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 7, 2011 18:46:43 GMT
I've been wondering for a while, what is the maximum amount of trains you could fit through one platform at a non-terminus station? (a bit random I know).
Would 60tph be possible? Or is 40tph only just managable?
|
|
|
|
Post by superteacher on Jun 7, 2011 19:08:11 GMT
I've been wondering for a while, what is the maximum amount of trains you could fit through one platform at a non-terminus station? (a bit random I know). Would 60tph be possible? Or is 40tph only just managable? The Central line could probably manage 33 - 34tph if all went smoothly. Have heard that the new Jubilee signalling will allow 36tph. New Victoria line signalling 33 - 34tph. Obviously, these being the ATO lines will have a higher frequency. Non ATO lines will be somwhat less than the above. 40tph would be very difficult with todays passenger loadings and poor passenger habits like holding open doors etc. I read in Underground News a few years ago that the Charing Cross, Euston and Hampstead raliway (when it opened) ran 44tph. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 7, 2011 19:12:38 GMT
Great, thanks for the info.
But let's say you had a perfect railway. Large doors to get people on and off quicker as well as reducing station dwell times, ATO, a guard at the back of the train and fast acceleration to clear the station quickly.
What would the absolute maximum, pushing the railway to the limit be?
|
|
|
|
Post by edwardfox on Jun 7, 2011 19:25:01 GMT
Perhaps LU should consider a poster campaign
"Passenger habits that delay trains"
Passengers on platforms
Standing directly in front of the doors of an arriving train, making it difficult for people to leave the train
Attempting to board when the doors have already started closing
Attempting to board an already packed train
Passengers on trains
Passengers intending to leave the train waiting until the train has actually stopped at their station before getting up from their seat
Passengers holding the doors open for late arriving passengers to board
Passengers leaning against the doors while the train is moving
|
|
cso
Posts: 1,043 
|
Post by cso on Jun 7, 2011 19:33:26 GMT
Passengers intending to leave the train waiting until the train has actually stopped at their station before getting up from their seat I know a number of passengers that do this for a very good reason - they are disabled, and unable to stand for long periods of time... so I'm not sure I'd agree with that one. |
|
|
|
Post by norbitonflyer on Jun 7, 2011 20:00:43 GMT
Train length must come into it - a long train takes longer to clear a section than a short one. However, a few long trains can move more people per hour than a lot of short trains.
|
|
|
|
Post by andypurk on Jun 7, 2011 20:01:55 GMT
Another important consideration is the length of the trains, as a longer train take more seconds to pass through the pointwork etc. at reversal points. Arrangements other than the scissors crossovers seen at many LU locations, for instance loops like Kennington (and on the Paris Metro), would lead to an increase in maximum capacity.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 7, 2011 20:12:47 GMT
Ok then. I know this would never happen because a lot of passengers are extremely irritating and insist on doing the things above, but lets say we had a fairly typical 4 car train (so not that long) with perfect passengers on board.
What would the maximum platform capacity be then?
|
|
Chris M
Global Moderator
Forum Quizmaster
Always happy to receive quiz ideas and pictures by email or PM
Posts: 19,429
|
Post by Chris M on Jun 7, 2011 21:17:27 GMT
If you wanted maximum throughput on a line then you could have an island platform per direction with trains alternating, such that as a train is leaving the station the next one is already approaching the stopping mark (if not already arrived).
|
|
|
|
Post by plasmid on Jun 7, 2011 21:42:02 GMT
The Jubilee Line can supposedly manage 36TPH as the signalling system is designed for but this is primarily because the trains are allowed uber close to each other.
Also if a train is travelling faster then then it has to stay further behind. A higher throughput would come from trains travelling at a slower speed with greater acceleration.
Trains on the Helsinki Metro appear to have great acceleration as well as some of the new French lines.
Though running more TPH isn't always the best idea...running longer trains is a better idea.
There is a Chinese/Japanese railway system (can't remember the name) which runs only 30TPH however the trains are 12 cars long. If you take the Central Line as an example where the trains are 8 Cars long you could say that this Chinese/Japanese line is operating 45TPH when it is only operating 30TPH given that the trains have 50% more cars.
More cars and greater acceleration would be the way forward without daring to exceed 33TPH at best.
|
|
Ben
fotopic... whats that?
Posts: 4,282
|
Post by Ben on Jun 7, 2011 21:56:57 GMT
Thats a bit of a paradox; the lower the demand the higher the potential capacity possible.
I wonder if theres a graph somewhere of loadings versus tph?
|
|
slugabed
Zu lang am schnuller.
Posts: 1,480
|
Post by slugabed on Jun 8, 2011 7:08:36 GMT
Thats a bit of a paradox; the lower the demand the higher the potential capacity possible. I have been told many times that the railways would work much more efficiently if they didn't have to deal with passengers..... |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 8, 2011 7:20:08 GMT
Thats a bit of a paradox; the lower the demand the higher the potential capacity possible. I have been told many times that the railways would work much more efficiently if they didn't have to deal with passengers..... They don't deal with passenges anymore they are now customers!  Xerces Fobe |
|
|
|
Post by tubeprune on Jun 8, 2011 7:54:57 GMT
I think we have debated this before. But, there are several factors that affect throughput at one point:
Approach speed, run in time; dwell time; run out time and signalling design.
Unless someone wants the full details on all these, I can summarise it all by saying that the ideal approach speed is about 27mph and the best throughput you could get with this and a 40s dwell would be 38.8 trains per hour (tph). To get this, the whole line must be restricted to a maximum of 27mph. The best you will get on the Victoria Line is about 34.1 tph because of the 50mph approach speed.
Of course, each line has its pinch points and these sort of throughputs are therefore impossible.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 8, 2011 11:58:18 GMT
I think we have debated this before. But, there are several factors that affect throughput at one point: Approach speed, run in time; dwell time; run out time and signalling design. Unless someone wants the full details on all these, I can summarise it all by saying that the ideal approach speed is about 27mph and the best throughput you could get with this and a 40s dwell would be 38.8 trains per hour (tph). To get this, the whole line must be restricted to a maximum of 27mph. The best you will get on the Victoria Line is about 34.1 tph because of the 50mph approach speed. Are those figures based on existing (LU) design and practice, or what an 'ideal' set up might be? In particular, do they consider using 'in station' signals and having the second train running in before the previous has completely run out? (New York certainly does this) Given that I have read that the District (I think it was) had 90 sec headways (i.e. 40 tph); and getting the impression that the LPTB expectation was 30+ tph in the central area, I am wondering what has changed to make these figures now seem so difficult. On a slightly different theme, is there any view on the point at which, other things being equal, lengthening trains becomes a better option to running more of them? |
|
|
|
Post by tubeprune on Jun 8, 2011 17:59:58 GMT
etr220, it's a complex issue but here are some pointers:
Short trains clear sections more quickly so the following train can run in more closely;
Older signalling had smaller safety allowances built in so trains were allowed closer together;
Nowadays, everything is recorded and no one goes faster than they should; we used to sometimes;
OPO adds at least 5s to every stop and often even more;
Under the 40 tph timetables, at the end of the peaks, the service was invariably at least 10 minutes late;
Longer trains need more infrastructure and this means changing the signalling berths, OPO equipment, CSDE, stabling sidings etc. Just look at what's happening now on the SSL because it was thought to be a good idea to make the new trains 2.2m longer than the old ones. It's costing squillions.
|
|
|
|
Post by edwin on Jun 9, 2011 2:02:38 GMT
Lines in Paris and Moscow can and do run 36-38tph in the peaks. They're very different beasts...
Paris has close together stations, cramped stations and short trains (under 100m).
Moscow has far apart stations, spacious well designed stations and long trains (160m).
Oh and Moscow doesn't even have ATO I believe.
They differ from LU in that they both have well designed terminus that allow a high throughput. In Moscow every terminus is like Bank on the DLR, but with two sidings rather than one. In Paris I believe many of them are similar to Morden in layout. With better designed termini (and without the foolish tipping out laws) I don't see why LU couldn't achieve these frequencies on some of their lines.
|
|
|
|
Post by norbitonflyer on Jun 9, 2011 7:19:24 GMT
In Paris I believe many of them are similar to Morden in layout. . Paris has a great variety of terminal layouts, with loops (either beyond the terminus, Kennington-style, or with one or more stations (T4 style), or reversing in the station or beyond it (as at the two ends of the Drain). Indeed, some sterminals can be operated in more than one way, for instance having both a terminal loop and the capability to reverse in or beyond the platforms - see lines 2, 4 and 6 on this link to the full track plan |
|
|
|
Post by edwardfox on Jun 9, 2011 9:10:04 GMT
Tipping out laws?
|
|
|
|
Post by tubeprune on Jun 9, 2011 17:06:49 GMT
Yes, Edwin means that there is a rule for LU that says every car on every train must be checked to see that it's empty of passengers before it goes into the reversing siding. It takes too long to do this so you lose time at places like Willesden Green. This restricts the headway. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 13, 2011 21:13:21 GMT
Moscow manages to run somewhere around 39 tph, using a coded track circuit ATP system and no ATO, and driver-only operation on most lines. ATO is not used as until recently, it wasn't terribly good at stopping the train in the right spot, and would take something like an extra second of coasting to the proper stopping point, which is fine until you consider that that's 0.5 tph of capacity on a system as busy as Moscow's.
The key to this high level of service is a combination of carefully designed signalling, having relatively powerful trains that can accelerate out of a station quickly, and keeping dwell time consistent and low. This is something that is affected by the design of the stations and trains, and the behavior of the passengers. Trains have plenty of room for passenger circulation and plenty of relatively wide doors. Trains stop in the same spots fairly consistently, so passengers waiting to board are already away from the doors, letting alighting passengers get out easily. Stations are spacious so that crowding on the platform itself is not an issue (though there are often significant crowds queueing at escalators to get out of the station). And the doors close quickly, giving passengers little opportunity to try to hold them, and they're powerful enough that you wouldn't really want to try anyway. The drivers even have a habit of holding down the door open button so the train doors open as soon as the zero speed detection trips, which is usually just a bit before the train actually stops. This all saves a few seconds, but multiplied by 39 tph, that's at least another 1 tph just from maximising the efficiency of the doors.
There are also a couple of features in the Metro that allow them to run this level of service reliably. Each station headwall has a headway clock that displays how long it's been since the previous train, which allows very precise management of headways by the drivers. Also, the timetable is written in such a way that the trains are not run at 100% speed under normal circumstances, so there's still some ability to "catch up" when there are minor delays.
|
|
