Center City Commuter Connection - Line Operations

I asked Matthew Mitchell of the Delaware Valley Association of Rail Passengers, for more information the Commuter Tunnel operations and the Philadelphia regional commuter rail system. Here is what he said and gave me permission to publish (thanks!). He provided these in October 2001 and he reviewed them for currency in October 2003.

I wrote, "I rode the Broad Street Subway (BSS) at times in the early- and mid-1970s, and the cars were very old with slow acceleration". Matthew wrote back (blue text), More importantly, they were unreliable (fires were not uncommon), and the attrition from those fires and other mishaps was such there weren't enough cars to "make the line" (deliver the scheduled service). The Kawasaki cars were very successful--perhaps the best car procurement SEPTA has ever done (though that's not saying much).

Matthew had the following to say about the Commuter Tunnel (blue text):

The original tunnel operating plan was developed by Vukan Vuchic of the University of Pennsylvania. A copy of the plan can be found in Penn's engineering library -- it is a 'must-read' for those interested in the history of the railroad and the peculiar mindset that infects many of the SEPTA people that plan and run the railroad. Vuchic is an unabashed fan of the 'S-bahn' metropolitan rail services in Germany, and in many ways modeled his plan after them. The R-numbers used to designate the commuter rail lines are analogous to the numbering of the S-bahn lines, and were part of Vuchic's plan (which we should note was never fully implemented).

Now as for numbers, one of the most frequently asked questions about the railroad is how the numbers were assigned, and why there is no R4. First of all, the pairing of the lines was done by ridership, so the most-traveled line on the PRR side (Paoli) was paired with the busiest line on the Reading (Lansdale-Doylestown), and on down. While some might think it would be better to pair lines by what through-tunnel trips would be the most popular, that through-tunnel (from an outlying station on the PRR to an outlying station on the Reading) ridership is very, very small, and to some extent dominated by weekend and college travel to/from Trenton. The benefits of pairing that way would be pretty small. On the other hand, pairing by ridership minimizes consist mismatches. Heavy ridership on one side of the tunnel and light on the other would cause you to have to operate a lot of empty car miles--though there's only so much you can do because PRR-side ridership is 50% or so more than Reading-side ridership.

That brings us to the second point about the pairing: it assumed that the city project called the "Swampoodle Connector" would be done. The ex-PRR Chestnut Hill line and the ex-Reading Norristown branch run within a few hundred yards of each other near their respective main line junctions, in a North Philadelphia neighborhood called Swampoodle (the area between the two lines is mostly vacant and isolated land). The plan was to build a connecting track between the two so Chestunt Hill West trains would be connected to the Reading side via the Norristown branch. The plan would have a lot of benefits:
-- Better balance of PRR-side and Reading-side ridership (CHW was second or third in PRR-side ridership)
-- Balances the number of lines and trains on the PRR and Reading sides (Airport and Paoli get roughly double the service of the other lines)
-- Gets CHW trains off the Amtrak Northeast Corridor (reduces expensive trackage rights payments) (improved schedule reliability)
-- Gets more CHW passengers to their destinations sooner (Market East ridership is higher than 30th St. except on Sundays)

So assuming Swampoodle, Vuchic stacked up the lines by ridership and numbered them clockwise according to their PRR endpoints.

Matthew provided the following data, and I put it in table format:

Line PRR endpoint Reading endpoint
R1 Airport West Trenton
R2 Marcus Hook (a) Warminster
R3 Media-West Chester (b) Chestnut Hill West
R4 Bryn Mawr (c) Fox Chase
R5 Paoli Lansdale-Doylestown
R6 Ivy Ridge (d) Norristown
R7 Trenton Chestnut Hill East

a--Service into Delaware suspended c. 1983 until Delaware DOT started paying SEPTA. Service to Wilmington restored c. 1990, followed by reopening of Claymont station, extension to Newark, and opening of Churchmans Crossing station

b--Service to West Chester suspended c. 1986, ostensibly because of  track conditions (SEPTA failed to do routine maintenance on tracks, and cannibalized signal and grade crossing equipment). Perhaps the shortage of train crews following SEPTA's takeover of the railroad from Conrail also had an effect--the tracks beyond Elwyn were routinely used for training purposes. Later, Chester County officials accepted SEPTA's de facto abandonment of the West Chester service in exchange for extending R5 trains to Downingtown and Parkesburg. In c. 1982-83, service was suspended beyond Swarthmore because the Crum Creek bridge was condemned.

c--Vuchic's plan called for dividing Paoli line service into separate Bryn Mawr and Paoli runs (both locals, IIRC).

d--The ex-PRR bridge over the Schuylkill was also condemned c. 1985-86, and service was truncated at Cynwyd. SEPTA tried to abandon the Cynwyd service on several occasions, but was rebuffed by the community, which pointed out that the decline of the line and poor ridership was mostly SEPTA's own fault for running such poor service. A signal bungalow was then struck by a very suspicious fire, and SEPTA once again tried to abandon the line, but was unable to.

But since Swampoodle wasn't even at the construction stage when the tunnel opened, an interim operating plan and line pairing scheme had to be implemented. Chestnut Hill West was paired with Fox Chase and given the temporary designator R8. Rather than split up into R4 Bryn Mawr and R5 Paoli, all R5 service operated to/from Paoli, with alternate trains turning at Lansdale and Doylestown. With CHW still on the Pennsy side, Media-West Chester was temporarily paired with West Trenton, which isolated the Airport line from the rest of the system--this was considered a good thing since it reduced the likelihood that delays on the Reading side would affect the  Airport trains, whose on-time performance was critically important. Those trains turned at Roberts Yard. For many years, West Trenton retained its R1 designator even though it was paired with R3 Media.

Acknowledging the permanence of the Media-West Trenton pairing and the fact that Swampoodle wasn't going to be built was the first change to the tunnel operating plan. I don't have the date of this, but it could easily be ascertained from a look at the West Trenton timetables in my collection. By that time, the old Reading-side power supply had been replaced and service was much more reliable. Again I don't know the year, but perhaps it was after RailWorks(NSM) that SEPTA decided to routinely pair Airport trains with Warminster. Alternate Airport trains turned at Glenside, giving that junction point 15-minute headways all day on weekdays. (hourly trains to Lansdale, Warminster, Doylestown, and Glenside)

After much prodding by DVARP, which pointed out that slavish adherence to the fixed line pairings was a significant source of inefficiency, SEPTA finally started loosening up some more (there had always been a few trains --particularly peak-hour Bryn Mawr locals that started or ended on an unpaired line). That brings us to our current situation where some rush-hour R2 Warminster trains are paired with R2 Wilmington again instead of Airport, at other times, Wilmington is paired with R6 Norristown (R6 Cynwyd has peak-hour service only), and on weekends Airport trains alternate between West Trenton and Warminster (bringing back the original R1 pairing). Weekend Media trains usually terminate at Market East, as do alternate Saturday Paoli trains (instead of Lansdale). Recently, there have also been changes in Trenton pairings, corresponding to cuts in service on Fox Chase and Chestnut Hill East. Further cuts were proposed, but rejected after SEPTA and the City of Philadelphia pointed out ways to reduce the number of trains necessary to run the weekend service without cutting service available to passengers -- the situation is still somewhat in flux, and to their credit SEPTA has been better about introducing flexibility into the system.

The 'mispairing' also is a good place to introduce SEPTA's train nomenclature, which is also connected with the tunnel.

SEPTA's basic train numbers are three-digit numbers, with the line designator in the hundreds place (so 501 would be an R5 train). Bryn Mawr locals are specially designated with a zero in the hundreds place (e.g. 001). To avoid conflicts with Amtrak train numbers, numbers in the 100, 200, and 600 series (400 would be this way too if SEPTA had an R4) have a zero prefix (e.g. 0101 for an Airport train instead of 101, which is also a Metroliner number). Numbers are then sequential through the daily schedule, with odd numbers refering to trains running timetable south on the main line (from Reading-side to PRR-side) and even numbers running timetable north (30th Street to Chestnut Hill West is officially a southward train).

Prefixes in the thousands place are then assigned as needed:
1XXX--Saturday-only train
2XXX--Sunday-only train
3XXX--not used
4XXX--trains not fitting the standard line pairing (*)
5XXX--trains originating/terminating at Fern Rock (#)
6XXX--trains originating/terminating at Powelton Yard (30th St)
7XXX--trains originating/terminating Suburban Station
8XXX--trains originating/terminating Market East
9XXX--trains originating/terminating Roberts Yard (North Phila./Temple U).

*--In this situation, the hundreds digit is used to refer to the train's scheduled destination rather than the origin, so dispatchers know where to route a train from its number. For example, a Warminster to Airport train might be 4115 while the return Airport-Warminster run might be 4220

#--Used primarily during RailWorks(NSM)

Numbers pertaining to origin/destination (4-9) take precendence over  weekend designators, so a Saturday only train from Media to Market East is going to be an 8314 instead of a 1314.

RailWorks is a NoServiceMark of SEPTA.

[end of Matthew's information]

Lead article - Center City Commuter Connection (Commuter Tunnel)

Copyright 2003 by Scott Kozel. All rights reserved. Reproduction, reuse, or distribution without permission is prohibited.

Back to top

By Scott M. Kozel, PENNWAYS

(Created 10-5-2003)