100 express service

During the 3 other seasons there is considerable demand at the intermediate stops between Bryn Mawr and 69th Street. It's certainly enough to justify dedicated local service - local cars are often SRO by the time they get to Penfield. Also the stations on the low end of the line are quite close together, so letting the expresses skip them actually does contribute to a shorter ride for passengers coming from the more distant stations.

One other advantage to having separate locals is that it reduces the need to have double cars running the full route. There are only a few trips that justify having the extra capacity north of Bryn Mawr. The remainder have the second car operating empty on the upper end of the line, all the while drinking electrons and carrying an operator who has nothing to do. (Note - the drivers I've talked to don't care for that, even though it means they are clocking hours for just sitting. They'd rather be working.)

The sad part about the running times cited is that they would have been unacceptable as late as the 1970s. End-to-end times were in the 22-25 minute range on express runs, with some limiteds making the trip in 18 or so. Current express timings look more like the locals of that era.

I agree that having a second car is wasteful if you only need them between Bryn Mawr and 69th St.

Instead of operating a two-car train, they could run more frequent service. The problem with having a two-car trains is that it costs about as much to run as two separate cars. Instead of a two-car train leaving at 4:00pm, for example, they could have one trolley car leave at 4:00 and another at 4:04.

In retrospect, they should have designed the N-5s to accomodate more people per car. Is there any reason why the cars have to be so short (and boxy)?

As for speed, it seems most of the extra time for peak vs. off-peak schedules is padding. Off peak local trains do the run in 25-28 minutes. The limiteds still do 20 minutes end-to-end. I think the increased popularity of the line is causing it to slow down, since loading times at stations goes up. I was at Radnor station last time and there were about 20 people getting on.

Lucius Kwok wrote:Instead of operating a two-car train, they could run more frequent service. The problem with having a two-car trains is that it costs about as much to run as two separate cars. Instead of a two-car train leaving at 4:00pm, for example, they could have one trolley car leave at 4:00 and another at 4:04.

That schedule might be thwarted by the sheep-like way people behave when 2 vehicles run close together on the same route. Watch what happens when buses bunch up on Market Street, for example. Two 17's will pull up together, and everyone tries to pile onto the first one while the second is essentially empty. I also saw the same thing in the days when the 125 and 95 buses ran 2 minutes apart between K of P and Gulph Mills. Everyone headed for G.M. would stuff the first bus, and the second had only stragglers and those riders headed for destinations beyond the 100 interchange.

If you did the same on the 100, not enough people would show up in that 4 or 5-minute interval to make the loads balance. The first car would be slowed by the extra boarding (see below) and after a few stops they'd be running together anyway.

In retrospect, they should have designed the N-5s to accomodate more people per car. Is there any reason why the cars have to be so short (and boxy)?

I'm not sure of all of the twists and turns in the design process but I know that SEPTA did a lot of the work w/o consulting the operators or getting passenger focus groups involved (shades of the Silverliner V fiasco). The federal gov't also had its hand in the pie. Finally there was the screwup with the seat pitch. You can see that the seats don't match up with the window spacing. They were supposed to, but after the bodies were under construction one of the senior Red Arrow managers double-checked and found that the average rider's knees would be smack up against the back of the next seat. They had to increase the pitch which knocked two seats from each side of the car. (At least I think it was 2 - Matt, do you remember?)

The car length may have to do with the line's geometry. Again I don't have the exact figures, but on the old board someone did an analysis of the line's curves vs. carbody length. Some of the stations are not on straightaways and the cars have to be able to clear the platforms at speed if they're not stopping.

I think the increased popularity of the line is causing it to slow down, since loading times at stations goes up. I was at Radnor station last time and there were about 20 people getting on.

Yes, but the line was even more popular back in the Bullet days and running times were consistently shorter. The Bullets had better acceleration and a higher top speed.

Dwell times are an issue for many SEPTA routes. The primitive fare collection system on the transit lines is one factor(*) in slow boarding. If everyone could use a farecard it might go faster, but since SEPTA's zoned Trailpasses cost more than paying fares using tokens and change you have a lot of people feeding tokens and coins into the fareboxes, at least north of Bryn Mawr. Dropping coins is slower than swiping a card, the operator has to make sure the coins were read correctly, and lastly they have to give the passenger a transfer. Yes, that amounts to maybe only an extra 5 or 10 seconds per rider, but it adds up. If half of those 20 people at Radnor didn't use a farecard you're talking about an extra couple of minutes right there.


(*)Over on the RRD, low platforms are another bog point.

RE: Bunching of cars: If they would reinstate express service, it shouldn't be a problem. The schedule I propose would be:
Essentially, this is increasing frequency instead of running two-car trains.

I think a farecard is a good idea. Most people boarding that time in Radnor used a token or transpass and paid the zone charge. It did take a couple minutes for everyone to board.

Couldn't have they designed the cars to be articulated in the middle? Then you'd have one-person operation with 100-seat capacity, and still run on the same tracks and platform clearances. Every other system in the country has purchased articulated trolleys for the past couple of decades.

Interesting to hear about the design process for the N-5. There's an old photo of the interior of the Bullet and it looks like it seats only 52. (N-5 seats 56.) I would guess that the N-5 is heavier than the Bullets due to crashworthiness requirements, and that lowers it acceleration rate and top speed.

I misunderstood. I thought you had proposed running two expresses back-to-back. In any case, the schedule you proposed is nearly the same as that run on the line about 4 or 5 years ago. Expresses left every 12 minutes. At first the locals were 6 minutes behind but the next express tended to catch up between Wynnewood Road and Haverford, so they moved the locals to 3 minutes (?) behind each express. It made for uneven departure times but much smoother flow.

I did ask a number of people why the N-5's couldn't have been designed more like RRD cars with some kind of vestibule and bellows to allow pass-thru, but never got an answer. Artics might work but then you're still running a huge car even for light-duty times.

There are lots of questions around the design and construction of the N-5's. I do know that the cars came in way overweight (about 6000 kg, I hear) and there were serious issues with body flex. A couple of cars had windows pop out when the carbody twisted.

Yes, a farecard would be great but SEPTA can't/won't revise its current crazy-quilt. Nothing like charging RRD fares for transit riders beyond the supposed Zone 1 boundary. You have to ride 35 days a month to break even, but people still buy the cards out of convenience or ignorance.

Interesting that you mention flex, because rectangular sections like the N-5 design is structurally weaker than the round sections used in the PCC and Bullet cars. It's like twisting a cardboard box versus a cardboard tube.

I assume the zone 2 trailpass be used between Bryn Mawr and Norristown. I haven't been able to find this actually written on the web site or schedule.

The strange thing about fares is that it is cheaper to use tokens than to use a transpass (if you commute 22 days a month, or 5 days a week) unless you need to transfer. Tokens are $1.30 while the pass is $1.59 per trip, but token plus transfer is $1.90. The zone 2 trail pass works out to $2.41 per trip while a token and zone charge is $1.80, and add in a transfer and it's still only $2.40. Buying a transpass and paying the zone charge works out to $2.19 per trip. I'm not even going to mention the Cross-County Pass. I think someone needs to explain why the fare structure is so convoluted.

Lucius Kwok wrote:The strange thing about fares is that it is cheaper to use tokens than to use a transpass I think someone needs to explain why the fare structure is so convoluted.

At the risk of derailing the original topic, SEPTA's fare structure is a sore point with me and a lot of other people. It is arguably the most complex of any transit system in the country, though possibly tied with Denver.

SEPTA has initiated at least 4 studies over the last decade, each one of which concluded that the current mess is a significant impediment to making the system usable and attractive. SEPTA rejected each one. In fact the last study was halted before it was even completed.

The scary part is that there appear to be people in the upper echelons at 1234 who do not want to simplify the fare system. Two different employees, one at an info center and one a mid-level manager, both told me there is a faction (if you will) in management who have tried to stymie efforts at rationalizing fares. They have correctly concluded that confusion about payment options causes a certain percentage of riders to regularly pay a higher fare than necessary. They are a source of "found money" for a system that is perpetually in financial crisis. SEPTA's own figures put the fraction at about 15% but I saw some numbers from UMTA that indicated overpayment occurred in excess of 20% of boardings.

Most schedules do have fairly accurate information but a few lines for some reason do not and never have. The 100 is unfortunately one of them. The 124/125 buses also have very misleading info but years of complaints to SEPTA have gone nowhere except into the can.

I assume the zone 2 trailpass be used between Bryn Mawr and Norristown. I haven't been able to find this actually written on the web site or schedule.

Using a trailpass on a transit route is almost always a guaranteed way to overpay. SEPTA makes a big deal about how "convenient" it is to have intermodal privileges but does nothing about the fact that you are paying the full whack for an RRD pass to get that one extra transit zone. The trailpasses are even promoted as the "best way" to ride transit lines, at least on the prior incarnation of the SEPTA website. Unfortunately a lot of people don't do the math and spend an extra $10 or $15 a month for that "privilege".

You have to work out everything on a case-by-case basis. For me, it was cheapest to buy a CTD transpass and pay the zone fare on a long month or when I knew I was going to travel on weekends (no zone charges then). A short month was often cheaper using tokens and paying for the zone and transfer. Regardless, I had to run through some Algebra I each month to make my decision. No big deal (I used to teach college math ...) but the fact is, IT SHOULDN'T BE THAT COMPLICATED! The people who can least afford to overpay tend also the ones who don't have the analytical skills to build a little model each month, just to ride a train line. I encountered too many riders who were either using trailpasses or even paying full cash, simply because comparing the alternatives was too complicated a task.

Lucius Kwok wrote:I assume the zone 2 trailpass be used between Bryn Mawr and Norristown. I haven't been able to find this actually written on the web site or schedule.

Actually, the Cross-County pass would be appropriate here.

As mentioned above, the N-5 order was snakebit from the beginning. But it would have been a tremendous challenge even for the most competent transit agency and car builder.

The key thing to understand about this equipment is it had to have a horsepower/weight ratio that exceeded everything else in the industry. That was necessary to meet the fast schedules on a line with stops so closely spaced. It was tough enough to do with the Brills, but the added weight from air conditioning made it even harder.

Among other things that happened to the project, there were issues with the construction and strength of the bodyshells, problems stemming from this being the first order of US transit cars with AC drive, and the builder (originally Asea Corp) went through a merger (becoming ABB) in the process. And just to add insult to injury, there was a period when they couldn't ship the bodyshells here from Portugal because all the available freighters were pressed into Gulf War service.

Matthew Mitchell wrote: Actually, the Cross-County pass would be appropriate here.

Yes, you're right. The problem is that it doesn't work on the El, does it? What happens if you present a pass at the El station booth? Do you have to pay some kind of transfer or zone charge, or full fare? I mean, it costs more than a monthly transpass, but you can't use it in the city, no?

Lucius Kwok wrote:Yes, you're right. The problem is that it doesn't work on the El, does it? What happens if you present a pass at the El station booth? Do you have to pay some kind of transfer or zone charge, or full fare? I mean, it costs more than a monthly transpass, but you can't use it in the city, no?

The Cross-County pass is intended only for riders not travelling into the city. You can use it on the El but yes, there is an additional charge of 60c. The higher base price of the C-C pass means it is generally not cost-effective for someone only going to Zone 2 on a suburban transit line and then tranferring to the El. Part of the reason it costs more is that it is valid on the RRD as well. The C-C pass is intended more for people using a large number of zones on suburban lines, or making long RRD trips outside of the city; e.g. Frazer to Haverford on the R5.

If you are planning to commute regularly on the 100 and MFSE, the only 2 viable options you have are a monthly Transpass plus the zone fare or a token plus the zone fare plus the transfer charge. You have to work out which is best, given a guess at what your commuting pattern will be for the upcoming month. For me, a Transpass was generally best but there was the month I was out for 6 or 7 days with the flu, and another when I was sent on a short-notice business trip. All you can do is eat the unused trips.

One thing people often don't take into account is that a Transpass is good for one boarding anywhere on a transit line. If you make a switch between 2 lines you don't have to pay extra for the transfer. So, for example if you took the MFSE to 69th Street, the 100 to Gulph Mills and finally switched to the 125 to go to the malls, the only extra charge incurred would be 50c for the second zone on the 100. Of course it also results in some anomalies: when I'd have to stop in Bryn Mawr to see my doctor, the trip from King Manor to Bryn Mawr was one zone / one boarding, as was the trip from Bryn Mawr to 69th. But if I had stayed on the 100 all the way in, I would have made a 2-zone trip and had to pay the extra half a buck.

The other goodie with a Transpass is that it has "anywhere" privileges on weekends, including the RRD, without any zone charges. If you have to ride then it can be a real savings.

Matthew Mitchell wrote:As mentioned above, the N-5 order was snakebit from the beginning.

Yes, there are no white hats here. In addition to the difficult specifications, there was a lot of politics and fumbling involved. The domestic-content regs forced final assembly of the cars to be done here. Initially Amtrak was chosen but Beech Grove didn't really have any expertise in light-rail work. Eventually you had the whole Asea, ABB and Morrison-Knudsen mess to contend with.

But to be blunt, nothing about the cars was helped by SEPTA's go-it-alone stance on design and construction. They didn't get input from operators or riders. More importantly, they turned down an offer to share information and maybe even hardware from PATCO, which is the one other transit line in the country that comes anywhere close to the 100 in operating philosophy(*). And nothing excuses Homer Simpson-like oversights such as delivering the cars without sanders, so they had to be pulled from service the first time there was bad weather! Just think about it - "Marge? Lisa? Where are you? D'oh!! Philly, Phoenix ... Both places have funny names starting with "PH". Can't ever keep them straight. I know one's really hot and doesn't get a lot of rain, so they probably don't need those stupid sanders anyway!"


(*)When PATCO was designed, a number of features of the 100 were incorporated and taken to the next level. There was even talk at one point of connecting the two lines via the old Cardington branch and an extended tunnel to West Philly.

Didn't the 100 have three zones? When did they go to the two zone system, and why?

The zoned trailpass system makes sense when compared to railroad fares, with monthlies about a 36% discount over peak fares. But for transit, the deep discount for tokens skews my chart. It is like a hole where revenue falls out. If tokens were $1.60 to $1.80, everything would make more sense.

The 100 had a third zone north of Gulph Mills for decades. About a year ago SEPTA flattened the fare structure a bit and absorbed the 3 northernmost stations into Zone 2. For anyone in the old zone 3 the change negated all but a nickel of the fare increase that went into effect about the same time.

Flattening zones is a favorite practice at SEPTA, presumably to reduce complexity but it actually creates a lot of distortions. The elimination of the old Zone 1 RRD fares actually reduced revenue. It also had the counterproductive effect of creating a big disparity between the central zone costs and what became the adjacent zone 2. They had evidence of people driving longer distances to take advantage of the cheaper central zone fares. Nothing like a transit agency encouraging people to drive ...

Also in the case of many suburban lines, fares paid for intermediate boardings (i.e. not starting or ending on a zone boundary) are determined purely by which zone you enter in and leave in, rather than the number of stations in between. Thus Norristown to Bryn Mawr is one zone, but shift the trip one station south (Bridgeport to Haverford, roughly the same distance) and you cross a zone boundary and have to pay an extra 50 cents. In fact, according to the schedule, Wynnewood Road to Rosemont is a 2-zone trip but Ardmore Jct. to Norristown is twice as far but only one zone!

Sanity, thy name is not SEPTA.

Lucius Kwok wrote:

Matthew Mitchell wrote: Actually, the Cross-County pass would be appropriate here.

Yes, you're right. The problem is that it doesn't work on the El, does it?

Sorry, I read your query as a trip from Norristown to Bryn Mawr, rather than a trip from one of those stations to the city.

What happens if you present a pass at the El station booth? Do you have to pay some kind of transfer or zone charge, or full fare?

You pay a zone charge: 60 cents.

I mean, it costs more than a monthly transpass, but you can't use it in the city, no?

Right. I think its price was pegged to the zone 1 pass when it was created.