From time to time, readers mention in comments the common delays at Spadina Station caused by the manner in which streetcars unload and load at the platform.
Before the introduction of the new Flexitys, cars would enter the station, unload at the east end of the platform, then pull ahead to load at the west end. There was even enough room that three cars could be on the platform at once with one ready to move forward into the loading zone as soon as room was available.
However, with the longer Flexitys, two cars will not quite fit on the platform, and although there are still separate unloading and loading stops, in practice only one car can be on the platform at a time. When this is compounded with delays for crew changes and with the siestas some cars take while loading, delays to passengers waiting to get off of arriving cars are chronic and lengthy.
I asked the TTC’s Brad Ross why cars did not make better use of the space.
An option that comes up is for the loading car to pull slightly beyond the platform and not open its front-most door. This would allow the following car to come fully onto the platform and unload. The front of the loading car would not be in the tunnel itself, but adjacent to an unused area of the platform at the west end.
The reply from the TTC is:
- It is not possible to isolate the front doors of the LFLRV. If the streetcar were to go past the glass barrier to allow a second streetcar to access the platform, we would not be able to open the remaining 3 doors, while keeping the first door closed. All doors would closed but enabled, and would require our customers to push the door button to gain access to the streetcar. In addition, we would not be able to prevent anyone from opening the first set of doors from the inside of the streetcar (or even from outside, should a customer go around the glass wall and push the button). This is a huge safety issue, as the front doors are not flush with the platform at that position, and customers would have to step down to the rail bed.
- One obvious solution would be to extend the platform so that it does sit flush with the front doors, but that is not an option at this point. The issue actually lies with the structural pillars that support the station; they are too close to the streetcar to allow proper egress. This is where I am seeking clarification from the Construction group. I am unsure of which pillar is in the way and what the actual requirement is, but have heard that it would cost millions to relocate the pillar to allow us to extend the platform. That is why that section of the platform is not being used and the glass partition is in place.
This begs the rather obvious question of why it is not possible for an operator to selectively open doors on a car. It is not unknown for vehicles to be in locations where a physical barrier would prevent use of all doors. In the specific case of loading at Spadina Station, the loss of the comparatively narrow first door would be a good tradeoff for simply getting a car far enough into the loading area that its follower could unload behind it.
I await further feedback on the matter of the cost and practicality of modifying the station, but in the meantime, it is useful to look at this problem in the manner we normally see for evaluations of expensive rapid transit projects: the value of riders’ time. Billions in spending on rapid transit has been justified by the premise that people will move more quickly and thereby save time, time that has a value against which the capital investment can be offset.
(This is a dubious proposition because the public investment is “balanced” against a private saving in “money” that can never be recaptured, but stay with me for the purpose of the exercise.)
- There are 15 cars per hour attempting to serve Spadina Station. If we assume that each car is delayed by an average of 2 minutes, and that it has an average of 50 passengers, this translates to 1,500 minutes of passenger delay per hour.
- If this condition persists on average for 8 hours per day, that means there are 12,000 passenger minutes of delay, or 200 passenger hours.
- The value of riders’ time is often quoted at about $30/hour, and this means a value of lost time of $6,000 per day. Scaling up to a year with a 300:1 factor (counting weekends as one day) gives us an annual lost time value of $1,800,000.
It is self evident that any of the variables used here can be tweaked up or down, but this gives the general idea of how the calculation would come out. Of course the City cannot “spend” that $1.8 million to offset reconstruction at the station because it is not real money, as I’m sure we would be told by the financial boffins.
In addition to any notional saving that riders might obtain, there is the real cost of, probably, one extra car on the Spadina route thanks to the extra running time needed to serve Spadina Station Loop.
There is a real need for the TTC to sort out operations at Spadina Station to minimize delays. This should include both figuring out how to use the loading area with a car projecting slightly beyond the platform and ensuring that crew changes happen as briskly as possible.
I will update this article when/if more information becomes available.
Updated Aug. 14, 2017 at 4:05 pm
Based on Twitter feedback, a few comments are in order.
In this article, I did not refer to explicit methods of crewing, but used the term “drop back” on Twitter. Some took umbrage saying that the term is “step back”. In fact both terms have been used over the years. It is the mechanism that counts, not the name. The idea is that operators get a break without the car having to sit there while they take it. To do this, there are more operators than cars, and an operator from car “n” “steps back” to car “n+2” (or whatever) so that they get a break of, in theory, two headways (the number can vary depending on how long a break is desired). This works fine as long as the operator who is supposed to take over an incoming car is actually available when it shows up.
At a location like Spadina Station, there is no stacking space to accommodate late crew changes whether they are part of a step back system or a regular shift change, and the problem can be compounded when the home division for the line is a long way away, and operators have to travel to pick up their cars in service.
Delays of cars getting on to the platform because of the loading techniques just make this worse by holding cars in the tunnel.
My aim in writing this article was twofold:
- Everyone involved – TTC management, planners, line supervisors and operators – need to work together to find a way to improve operations at this busy station which have deteriorated noticeably since the new cars were introduced.
- The principle of “value of riders’ time” is often used to justify big ticket capital projects, but it does not have the same clout in day to day operations.
What about the feasibility of extending the east end of the platform?
Steve: That is probably trickier because it is right at the curve. At the west end, it’s a case of making use of the space that’s there, or simply not opening the front-most door if the space cannot be freed up.
Adding to the delay, fare inspectors usually hang out on the streetcar platform and create a bottleneck for passengers getting off the train. When the Metropass disappears I’m guessing this will be even slower since a tap will be required to verify a fare instead of flashing a pass. It’s a shame that the location of the fare inspectors is so predictable that fare evaders know to get off at the stop before the station to get a free ride.
How common is this problem at other stations? I think you’ve mentioned that the platform length at Union is a problem? Bathurst seems a little stressed by the length of the Flexitys too. It is hard to fathom how the TTC could fail to consider platform capacity when choosing the design.
I guess individual door controls becomes impractical when there are too many doors, but only middle (aka the two double width doors) would have been a sensible setting. No doubt Bombardier is going to tell us that the doors aren’t individually addressable, so it can’t be changed in software? Maybe just post a guard to stop people from using the wrong door during peak hours? It could be a fare inspector drawing a rope across even (surely there is a door override/lockout that a fare inspector could engage)?
Steve: Spadina is a special problem because of the frequency of service and the shape of the platform. Every inch is needed, but it’s not available because of the way the service is operated combined with the constraint on selective door opening.
Why not use a TTC employee to stand near the front of the car in order to prevent people from boarding at the front of the car…at least till they figure out a proper solution…Allowing only one car on the platform to reload is not a good enough solution…
What if the second car only opened it’s front door, this, I understand, can be isolated from the others. Allowing passengers to disembark through the front only, while the car ahead loads. This would be respectful of the passengers time. This may not work with a crush load as the time it takes for all to exit , the car ahead may have moved off.
Steve: When a car has over 100 people on it, letting them off only through the narrow front door is not workable. It is one thing to have a crowd martialled into rows corresponding to doors that will actually open on a car loading, and quite another to stem the tide of people trying to get off a car coming into the station. Far better to work on moving the loading cars up so that unloading cars can get fully onto the platform.
I’m sure bombardier would charge a pretty penny for a software upgrade to be able to isolate the front door, something that should have been standard for all the doors.
Oh so, only Downtown folks’ time is important whereas people in Scarborough are forced with unnecessary transfers because us downtrodden Scarborough folks’ time is not important.
Steve: It’s always about Scarborough isn’t it. My reference was not only to the Scarborough subway but to many large capital projects where one of the major “benefits” is time saving for riders. Indeed, one of the arguments against having more stations on GO (which would serve Scarborough among other places) is the delays they would cause for travellers originating further out.
My purpose was not to dis Scarborough, but to comment on the whole financial model on which so much of our decision making has been based.
Why not build a fake wall at the front of the platform…so they can open all the doors but you can’t get out…paint a DO NOT EXIT THIS WAY sign on it…or even better…an advertisement!
Pretty terrible software if it can’t be patched to add an extra option. Is there not enough memory left in the onboard computer? Sounds like bull to me.
I say go hog wild. Dig up the loop, shore up the station, and extend the loop over to Madison so that 3 cars could fit into the new expanded platform.
Steve: The east end of the loop is already under Madison.
Personally I prefer the drainage pits be place there. First step’s a doozy?
All joking aside I am surprised nobody thought about including door isolation in the specs. You would think that would have been a no-brainer for security, safety and overall common sense reasons.
If they drop one run from the route, would that space them out so that they don’t bunch up at the station?
How about taking one section out of the Flexibles and putting it in the trains on Line 1 on Yonge Street? After all the trouble that was taken to design cars for all the special situations that only exist in Toronto…
Steve: There is a slight difference between subway cars and streetcars.
It doesn’t matter how easy it is to patch the software, it it wasn’t in the specs, they can charge whatever they like as a change fee.
Talking about changing software, I’ve noticed on the Flexities still have the abandoned stops like Adelaide and Bloor in the stop announcements.
Steve: The TTC takes a while getting around to updating the stop announcements. It can be annoying for riders and operators when the car doesn’t stop at an announced location.
How about something old fashioned like a relay that disables the front door mechanism when a button is pushed that would reset after all the doors were closed.
I remember my family friend working for one of the transit agencies in Poland, driving Ikarus articulated buses. The door controls were 5 individual controls, 4 for each of the passenger doors (including an alarm to ring to warn of the closing doors), 1 for the driver’s door at the front. Perhaps I’m not seeing something, but if 1980’s Iron Curtain buses can do individual doors, why can’t a modern streetcar in 2017?
On the case of Spadina station streetcar loop versus Scarborough subway, of course if the loop cost $3.8B to fix we wouldn’t do that, no matter what the time savings. And if the Scarborough subway extension cost a few million, I’d enthusiastically say to go ahead.
But the bottom line for Scarborough is all those billions. You won’t get them down to millions. Spadina station loop, though, might only be a few million, and it won’t be billions.
I don’t know what a similar “time saved” calculation might be for present-day riders of the SRT, but I doubt that a replacement subway would save billions of dollars a year of “lost time”.
Exactly my thoughts – quick, easy and inexpensive to implement. However, it doesn’t address the potential problem of a customer who has boarded from pressing the interior button for the front doors. The software has to be modified so that individual doors, or at least the front door, can be isolated. I’m floored that this is not possible on a 21st century streetcar.
This is going to be a major problem at Broadview, as well. Streetcars already get backed up there, and that’s with the current CLRVs: the platforms can currently only hold one Flexity. Because the 505 and 504 share the same tracks up to the station, a backup of either line causes a backup on the other. Now that they’re eliminating the northbound stop at Danforth, the option to get out and walk isn’t going to be available anymore. This will be “interesting.”
I really think they should remove the occasional fare inspection at Spadina and install a set of fare gates. Put a couple of fare vending machines for those who somehow “forgot” to pay. Vancouver has something similar where people who didn’t buy the right amount of zones had to buy a new fare to tap out of the system. This would become systematic behaviour where everyone would tap into the station and reduces a change of a bottleneck. Everyone heading to the subway will tap instead of trying to cheat the system when no inspector is present. The fare inspectors can be relocated to the surface where actual fare evasion are more likely to happen.
Steve: The problem with this idea is that many people will have paper transfers or stamped tickets that are not machine readable. These are not going away any time soon, even with Presto.
Steve, in other words, I should add this to the list of issues with the new streetcars? I take it that the TTC never considered this when they ordered the new streetcars. It would have seemed like an obvious thing to consider. Is there any chance that the TTC could expand either end of the platform (or both ends) in order to make it easier? For example, if they need an extra two feet, why not extend the platform by two feet – either put the two feet at one end or the other, or extend each end by one foot, or by one and a half feet at one end and six inches at the other.
Steve: Read the TTC’s reply: there is column in the way at the west end. As for the east end, the geometry is different because the tunnel begins immediately at the end of the platform and the track curves south there.
The new cars can open and close individual doors. That function can’t be accessed from the driver’s main console for a bunch of different reasons, however.
“The tunnel begins immediately at the end of the platform and the track curves south there.”
As long as the last set of wheels is passed the curve, the back door will lined up strait with the rest of the platform.
Steve: My comment refers to extending the platform into what is now tunnel. In any event, I believe the focus should be on the loading end of the platform, in addition to ensuring that operational issues like crew breaks/changes are handled as efficiently as possible.
The driver may open only the frontmost door, as I’ve witnessed many times (going all the way back to test runs at Russell). But that’s the opposite of the problem we’re trying to solve here (open Doors 2 through 5, keep Door 1 closed).
The rearmost door can be opened from the outside, via a method that will be obvious in retrospect if somebody points it out to you in person. This too ain’t what we’re looking for.
Would it be possible to knock out the glass panels and open the front door in the space between two of the columns? The TTC already does this at Broadview.
Steve: Please explain. I live at Broadview and don’t understand what you are proposing.
Or just go back to the old system which worked well for decades: just let people off the streetcar and head for the subway. This whole new idea of requiring people go through another fare gate, or through an inspection, only works if there are zones. With a one zone system like what the TTC operates with makes it a waste of time (for passengers) and money (for the TTC) to make the change in my opinion.
Steve: Fare inspection on board is just about impossible because the cars are packed, and it’s convenient to check everyone coming into the station. However, the way it’s done can be a bit disorganized especially with no Presto reader on the platform where people can pay up when caught.
In other words because of a bad design by Bombardier. Common sense would have made the function possible from the driver’s console. If the TTC realized this early enough in the process, they could have insisted on things being done better.
Steve: Be careful in where you assign blame. Decisions on what door controls were available could just as well have been a TTC decision as a Bombardier one.
I have a feeling there is going to be loading/unloading problems at St. Clair West Station also.
I’m not sure how much of a problem St Clair West will be as it has separate boarding and unload areas already. The current load are will just become one rather than two and they won’t be able to load a car on the passing track it will have to loop around before it can load if a flex is there.
Steve: The issue will be that cars unloading/loading at the current “unloading” zone on the west side of the loop will have to get in/out quickly with little break time because this would block following cars. However as you say this happens already on the loading platform and so there’s not that much of a change. Conversely, cars can sit on the north platform and be bypassed via the spare track.
Regarding the new streetcars, why is the white-colour band at the front of some new streetcars discontinuous and inconsistent? Did Bombardier start manufacturing parts in the wrong colour?
Steve: Looking at pictures of car 4411 (the one illustrated in the linked Toronto Life article) on Google, it had the correct colour scheme when it was delivered. I am willing to bet that it was in a collision, and the TTC replaced that section of skirting with the wrongly coloured spare part. This is rather like the old days of PCCs with mismatched sets of doors.
It would be amusing to track down 4411 to see if this has been fixed. In any event, it’s not a Bombardier problem.
I saw the Bathurst Station platform handle two Flexity streetcars at once. The first two doors of the first Flexity were at a narrow platform beside the small building at the north end of the platform area. There was a supervisor and a cone to block boarding at the narrow front door. This narrow platform seems roughly the same width as the lost platform space behind the glass wall at Spadina Station but is paved and has no pillars. The rear door of the second Flexity car at Bathurst was opposite a fence used to discourage shortcuts from the street to the fare-paid area; however, there was sufficient width between the car and fence for passengers to reach the platform.
With Flexity streetcars, the Gunns Loop exhibits some of the problems of the streetcar loop at Spadina Station. Today (Sunday), all streetcars on 512 St Clair were Flexity cars – 12 of them according to NextBus. The platform at Gunns Loop is long enough for one Flexity. The second Flexity arriving lets passengers off in the park area on the north side of the loop where there is less than 2 feet of paving against the outside rail to act as a platform. If a third Flexity arrives, it must wait on St Clair Avenue until the first Flexity leaves in order to discharge its passengers. Today, I saw three at Gunns, and the third had to wait a minute or two.
St Clair Station on the other hand has three slots each long enough for a Flexity: arrival, departure and the curve in between. On Sunday, I once saw all three slots occupied by a Flexity for a few minutes.
Assuming that all streetcars at St Clair Station go to Gunns Loop, then there appears to be a mismatch of capacity at the two termini. The narrow “platform” for the second Flexity at Gunns Loop may be a problem on a snowy winter day.
Steve: There also appears to be excessive running time in the schedule.
Something to keep in mind with 512 until it is completely all Flexity cars all the time without any CLRVs scheduled then they will have the headways still set as being for a CLRV. They will most likely be looking at changing up the headways like they did on 510 in a few weeks or so.
Steve: That change is planned for early 2018. See this article and the linked briefing note.
A TTC video gives a view of the alcove in the tunnel behind the glass wall. Even with the pillars, it seems as wide as some other areas where I have seen doors opened for passengers. Note that in the video, the Flexity streetcar has very little overhang going around the curve. Perhaps, one could extend the platform into the alcove with a yellow strip over most of its width, and install a swing gate at the alcove entrance with a do-not-enter sign on the outer side, and a push-to-exit sign on the inner side.
I really hate to beat a dead horse here as I’ve done a few times before I jut simply can’t, for the life of me, comprehend why the TTC ordered such long cars in the first place. Surely it ought to strike every sentient creature who sees this as more than obvious that that the sheer length of these beasts is always going to be a problem unless major changes in loops aren’t implemented.