Thursday, October 25, 2007
Train Splits
Nick over at 295bus asks, why can't we split trains? Well it seems as if the Sprinter in Oceanside will do it. I've heard a lot of people marvel at how trains split to go in different directions once they get to a certain station, mostly in Europe, but often wondered why we couldn't do that here.
Subscribe to:
Post Comments (Atom)
6 comments:
It would require too many drivers (i.e. labor costs)?
How is this functionally different from running multiple lines on the same tracks? What's the advantage?
Suppose you're on Y shaped system, where the base of the Y is A, and trans can go to two terminals, the branches of the Y, B and D.
If you're going somewhere on the stem of the Y, it doesn't make any difference if they run alternate trains to A and B, or if they run combined trains that split at the junction.
But if you're going to A or B, it makes a big difference--running combined trains that split means trains that actually go where you want to come by twice as often.
For people with non-flexible work schedules, service frequency is at least as important, maybe more important, than speed.
Sure, it may require more operators, but it's better service.
There's always going to be trade-offs in this sort of thing. For a given staffing level, an agency might hypothetically be able to run trains from A every 20 minutes with an alternating destinations schedule, or every 30 minutes with a merge/split-trains schedule. But note that with the alternating destinations approach, if you want to B or C, there's only a train for *you* every 40 mins.
On the other hand, the increases in staffing needed to do split/merge operations should be offset by reduced wear and tear on equipment (i.e, run long trains in the core of the system where there's lots of passengers, then break them up into shorter trains for outer branches, where there's less ridership, instead of running long mostly empty trains).
I think the real reason transit agencies are loath to use train splitting is that it requires good planning and high reliability (it's easy to see how small problems could snowball and clog up the whole system)--which are definite problem areas for most US transit systems.
Amtrak at least used to do this on the Lake Shore Limited at Albany (actually Rensallier). They may still do it, but my experience dates from the late 90's. But they took such a long break to do the split, and the train was so frequently late, that this isn't really a very good example.
Amtrak Empire Builder does this. A train leaves Chicago, splits at Spokane, and ends up in both Portland and Seattle.
The LSL limited stopped splitting a year or two ago. Passengers going to/from Boston have to transfer at Albany.
A really prominent precedent in the United States: MUNI Metro light rail trains used to split at West Portal, where the K-L-M lines join into the Downtown tunnel. That suited the system -- which can handle 2-car trains on the street-running branches but has platforms easily long enough for 4 or 5 car trains in the subway -- very well. But METRO's consultants sold them on a new train control system that allegedly supported shorter headways, and the practice was abandoned. The operators stayed on the trains for the full length of a run -- in the subway, you'd see a guy in the cab of the 2nd or 3rd car of a train, just waiting for when the train would split.
Post a Comment