The Scarborough RT Derailment

On Monday, July 24 at 6:45, a southbound SRT train was leaving Ellesmere Station when its rear car derailed. This was not simply a case of slipping off of the tracks, but a more complex incident in which:

  • The rear car of the train broke away from the other three cars.
  • The truck (or bogie, the undercarriage holding the wheels and motor) at the rear of this car was detached from the car.
  • A section of the reaction rail which is part of the propulsion system broke away from the guideway.

The car came to rest with one end leaning against the guideway’s fence and a low sidewall. Fortunately, this is a location where a derailed car could not fall any distance.

Five passengers were taken to hospital with minor injuries.

The line remains closed with shuttle buses ferrying riders, and the TTC has announced that the line will remain closed for at least three weeks pending an investigation of the cause of this accident. That will take us until at least mid-August.

It is not certain if the line, which was planned to shut down permanently in mid-November, will ever re-open depending on the cause of the derailment, the amount and cost of work to remediate it, and the limited time during which this expense would have any benefit.

Updated July 27 at 10:20pm: The proposed schedule for the SRT right-of-way conversion to a busway has been added to the article to clarify that “Winter 2025” really means the end, not the beginning, of 2025.

For the convenience of readers, especially those out of town who might not have followed this event in detail, here are links to many articles which include a wealth of photos.

City TV

CP24

Global News

Toronto Star

Spacing

  • Technical Difficulties: The Scarborough RT accident was absolutely predictable as policy-makers and riders have been aware of it’s deterioration for almost 20 years

CBC

State of Good Repair on the TTC

The mounting backlog of TTC capital needs has been a concern for transit watchers for some time. TTC and City of Toronto practice was, for decades, to show only the “funded” projects in the 10-year capital budget, plus a small number on the then-current wish list. This understated the actual size of the funding gap by over $20 billion, a problem that finally came to light when the TTC published its first 15-year capital plan in 2019.

That gap triggered calls for better funding, but what has materialized in the intervening years is only a small part of the total needs. Even with Ontario taking responsibility for major transit projects, much remains to be funded primarily to keep the system we already have in operation. A recent effect of this situation has been the withdrawal of a Request for Proposals for new subway cars for Line 2 Bloor-Danforth.

With political attention focused on the capital backlog and, more recently, with service cuts due to pandemic ridership losses, an important area has escaped scrutiny: the day-to-day maintenance of the transit system. We can see service cuts while waiting for a bus, streetcar or subway train to appear with, we hope, room for us to ride comfortably. What we do not see is any gradual deterioration of vehicles and infrastructure, something that tends to show up only when it fails.

Toronto has been down this path before in the early 1990s when deferred maintenance coupled with poor operator training led to the worst crash in TTC history. Those of us with long memories will recall how TTC management assured the Board that despite operating subsidy constraints, the system would be properly maintained. It was not.

There is no published reason yet on the cause of the SRT derailment, but a few of the photos are revealing. They show a missing section of Linear Induction Motor (LIM) reaction rail, and a section bent back.

Photo: Screen grab from CP24

The aluminum plate that is between the rails is the top of the reaction rail, part of the LIM that SRT cars use to propel themselves. The other part of the motor is a coil under each truck that is, in effect, an unfolded winding like one would find on a rotary electric motor (“linear” because the motor is flattened out). Alternating current flowing through the coil creates a magnetic field that induces a corresponding field in the reaction rail. Because the current is AC, the field moves, and the car “pulls” itself along the track. There is a small air gap between the car-mounted coil and the reaction rail.

An important part of regular maintenance is to ensure that this gap is constant, and that the coil does not strike the reaction rail. Because of the way the motor works, there is an upward force pulling on that rail, and it is key that the rail mounts keep the rail in place. Also, as is clear in the photo, the aluminum rail comes in sections, and it would be possible for the edge of a section to snag a car rather than simply rubbing along the face of the motor coil.

Whether that has happened here, or if there is some other cause, we do not yet know.

(For those who might be wondering, the cables hanging on either side of the reaction rail are the antenna for the train control system.)

What is essential is that the investigation be full and open so that everyone can trust its conclusions, unlike a near-collision in the subway that was kept under wraps for a year after it happened in June 2020.

If this is even in part due to a reduced standard of inspection and maintenance, that has implications for the system as a whole, not just for the SRT.

The Shuttle Bus Service

There have been conflicting numbers cited about the replacement service. TTC states that it is running 20-25 buses off-peak, 35-40 peak, to replace the RT service.

In preparation for the SRT shutdown, the City has been planning reserved bus lanes on Ellesmere, Kennedy (northbound), Midland (southbound) and Eglinton. See details here.

The target frequency for service is 70 buses/hour under this plan, although whether this is actually achieved or not is another matter.

If the SRT does not reopen, the need for these lanes and associated transit priority signals, plus completion of loading areas at terminal stations like Kennedy, goes from a “we need it soon” to “we need it yesterday” priority. Whether this will actually happen remains to be seen.

In parallel, the TTC and City intend to repurpose the existing SRT right-of-way between Ellesmere and Eglinton for a bus-only roadway. This project almost ran aground for lack of funding even to complete the design work, and it is still uncertain whether actual construction will be mired in a finger-pointing exercise between City Hall and Queen’s Park. Such is the nature of real concern for transit riders that a key project can be mired in political posturing.

The TTC projects that the busway will not be open until early Winter 2025, over two years from the projected shutdown date for the RT.

Updated: Here is the project plan for the busway project as presented in early 2022. This clarifies that “Winter 2025” really is two years away.

There is a severe need to light fires under this project and reduce the timelines as much as possible so that riders can have the benefit of a faster trip, even if by bus, as soon as possible. The history of construction in Toronto is littered with projects that seem to amble to completion with little regard for the inconvenience caused by delays. At least this should be a TTC-only project without the need to co-ordinate work by other utilities like Hydro, Bell, Toronto Water and others that have bedeviled recent jobs.

There is no worry about having enough buses as the TTC at peak only uses about two-thirds of its fleet, a very generous spare ratio well above industry standards, and an order of new buses is now arriving that will replace the oldest and least reliable part of the fleet.

The challenge will be for City Council to get on with the job, even if the RT makes a brief return to service for a few months, as quickly as possible.

22 thoughts on “The Scarborough RT Derailment

  1. “The TTC projects that the busway will not be open until early Winter 2025, over two years from the projected shutdown date for the RT.”

    Isn’t that more like a year and a half, from 2023 November to 2025 January or February (or, cynically, May)?

    I ask mostly for understanding, not because I think such a small error counts for anything when stacked up against the great work you do.

    Steve: If you look at the project schedule on page 14 of the Line 3 Bus Replacement Study, you will see a chart that clearly shows that construction ends in late 2025, not early in that year. So my comment about two years remains. I should pull that chart into the main article for reference.

    Like

  2. A couple of thoughts about the age of the SRT fleet (37-40 years).
    How many of you are driving cars that old? (In daily, extensive use, not historical vehicles)

    There is a vehicle in the TTC’s historical collection that is younger than these cars. Also at the museum in Rockwood. (ALRVs, plus some work equipment.)
    We had a couple of motor homes that were 20 years old when we bought them. We didn’t keep them long.

    Steve: People seem to forget that the original subway trains on the Yonge line from 1954 have been through two rounds of replacement, and on Bloor-Danforth (1966) there has been one complete fleet replacement with another one in the RFP stage until it was cancelled recently. Rail vehicles are designed for a lifespan of 25-30 years. The SRT has been kept running through major rebuilds.

    Liked by 1 person

  3. Now I see, thanks. I think of Winter as being at the beginning of the year but I suppose I should know by now to take the farthest in the future possible interpretation of any completion time description coming from a transit agency.

    Liked by 2 people

  4. Cut it up on site and start the bus shuttle. But for goodness sake, get a head start on the conversion.

    Politicians LOVE to show up at the ribbon cutting for the Shiny New Thing but they will *always* underfund maintenance. Always.

    Liked by 1 person

  5. I keep hearing that RT users can use GO at the TTC fare (to go….somewhere). I wonder how long that protocol will last?

    It’s also interesting to contrast the TTC’s fleet replacement cycle with Montréal’s Métro. They have only retired the very first generation of cars. The second gen, from the mid-70s, is still very much in base service on all but the orange line, and they only have had three generations of trains from the mid-’60s, during which time Toronto went through H1 through H6, and T1 and TR fleets.

    Steve: Montreal’s fleet has the advantage of being completely underground where it is less affected by weather. I suspect also that less salt makes it down to platform level from the street than in Toronto. It would also be interesting to compare the duty cycle of trains in Montreal in terms of grades, passenger loads and hours of service.

    As for car series, you cannot cite the H1-6 plus T1 and TR as if they were all replacements for each other. We began with the G trains on the Yonge line, and then added the M1s for the University extension. The H1s were bought for the Bloor-Danforth subway, and then there was the BD extensions, North Yonge, Spadina, Sheppard and Vaughan. The H2s were purchased for the North Yonge extension. The H3 class was a trial of chopper controls and these are really part of the H2 order. The H4s were purchased for expanded service, and the H5s for Spadina.

    Many H6 trains replaced the original G cars on Yonge, but the H6s were a problem fleet and they did not live out their 30 year lifespan. Eventually, they were replaced by the TRs. The T1s replaced the original fleet for the BD line about 30 years after it opened.

    The equipment assignments are a bit more complex than this potted history (for more see TransitToronto’s site) but the main point is that many of these fleets co-existed as part of gradual service and network expansion, not as replacements for older cars.

    Liked by 1 person

  6. Yesterday a work car derailed at Greenwood.

    I wonder if there is a systemic issue at the TTC rather than a mechanical fault. I mean if train cars keep derailing, there has to be a reason.

    Maybe their SOGR is lacking or perhaps it is poor training overall but in any case this is too close together not to make someone wonder.

    Things like this should be an isolated incident not a weekly occurrence.

    Steve: A wider issue is that the TTC routinely describes a wide variety of problems as a “mechanical” or “operational” issue without specifying exactly what is happening. This makes following potential trends difficult. And, of course, incidents in yards do not even make it into the delay notices.

    Liked by 1 person

  7. Original Montréal Métro cars were the Vickers-built MR-63 trains (M-T-M triplet “elements” formed into 9-car trains). MR-73 cars were basically the same design with updated propulsion and 1970s interiors built by Bombardier. They worked anyway likely because they didn’t design them. These were necessitated by the Olympic-era system expansion.

    MPM-10s are a completely new design with a VERY long gestation period – so long that they were supposed to have been MPM-08s. Now that the tunnels have all been modified because (*someone* didn’t read the specs) they are in service. Finally. These cars are open gangway cars, so a little more capacity – MR-63 and MR-73 cars had emergency doors between the cab cars and the intermediate cars, but you couldn’t get through the full-width cabs and you weren’t supposed to open the doors between the end cars (M) and the middle cars (T).

    As noted, Montréal cars never go outside (there is some above-ground track at the Youville shops). PLENTY of salt in the stations but most stations are pretty deep so the escalators seem to eat most of it. 🙂

    Grades would be steeper, I would think, than Toronto. Notwithstanding the last 3.5 years, I would think loads would be similar? Though I’m pretty sure the individual cars are smaller in the Métro. AFAIK, the span is still 0600-0100 pretty much seven days/week (shorter on Sunday, maybe?)

    Liked by 1 person

  8. The fact that some other cities run their rolling stock past 50 years is about as irrelevant to the TTC today as it was when they retired every previous generation in the 30–40 year range, and is not a valid argument in favor of the TTC running the existing fleets that long. I’m also going to go on a limb and speculate that if the SRT was able to last nearly 40 years despite all of its problems, the H6s were almost certainly capable of lasting that long as well.

    Like

  9. Meanwhile, the O-Train in Ottawa is offline again, and any attempt to restart has yet again been delayed. Shades of Scarborough RT. What will Ottawa’s next 40 years be like…

    Like

  10. Montréal had a collision and fire on the Métro back in 1971. However I have not found any stories of derailments, let alone fatal accidents to compare with the TTC’s Russell Hill disaster.

    The fact that Montréal can can continue to (apparently) safely operate its system using rolling stock dating from the mid-’70s, where Toronto retired similar vintage cars a decade ago, may or may not suggest differences in maintenance and repair philosophies. So I don’t think it’s necessarily irrelevant to compare the two systems.

    I rode the Métro extensively this spring when I visited Montréal. The new Azur trains have, to my eye, beautifully-designed interiors, kind of like walking into the cosmetics area of a high-end department store. The trains move right along too. While some of this is due to the tunnel profile of raised stations and deeper tunnels, the trains don’t particularly seem to mind the climbs, certainly the yellow line has long sustained grades, crossing as it does underneath the St. Lawrence, River and Seaway both. And while I tried to avoid rush hours, ridership is quite decent, and standing room only is not unusual.

    One of the nice touches in the system: large screens in the station show not only train arrival times, but passenger loading in each car. At least on the orange and I think green lines.

    Another nice feature (for tourists) is the variety of fare options, including short-term passes. And no $6 charge for a card.

    There certainly are things Toronto and the TTC could learn from a visit to Montréal.

    Steve: An important difference between the systems is that the primary suspension is on the rubber tires, and the steel wheels and rails are used only for junctions. “Derailment” would have a completely different meaning in the Montreal context.

    Like

  11. The TTC should procure new trains for the SRT line even if only to serve until the Ford Extension finally brings subway service to Scarborough.

    Steve: There would be at least a two year lead time to get new trains, and the line could not use an “off the shelf” model such as current Vancouver models because they will not fit on the SRT infrastructure. This would be an extremely expensive fix.

    Like

  12. The Métro’s Michelins run on concrete rollways outside the standard gauge steel rails. But there are also guideways at right angles to the rollways with more tires that run along those as well to keep the train on the rollways (steel rails notwithstanding). It’s weird, but it has worked OK for 57 years. Now, does any of it make sense? Do the tires and rollways last longer than real wheels and rails? No idea. Is the system quieter than real wheels and rails? Yes .. but only if you have never ridden a real subway. 🙂 I’m not even convinced that the trains can climb steeper grades than steel/steel. Power-wise, I can’t imagine that the rubber tires are better than real train wheels given the flexing. But it would be cost-prohibitive to change at this point and there is a certain amount of “we do it the right way and those other cities are *fools*” to contend with, too.

    REM is conventional in propulsion (standard gauge, as Mr Stephenson decreed) and in power (1500 V dc catenary). It has other issues (too short trains of too small cars and no easy way to expand) but they didn’t do anything crazy otherwise.

    Steve: It is telling that after Montréal “copied” the Paris Métro and its rubber tires, that Paris stopped converting lines to that technology. As for the steel rails, there is little or no contact with them unless a tire deflates, and at switches where all of that business with vertical and lateral tire guidance would be prohibitively complex. Car size is an issue due to weight limitations, just as it was on the SRT and the original “Mark I” Skytrain cars.

    Like

  13. Steve: the line could not use an “off the shelf” model such as current Vancouver models because they will not fit on the SRT infrastructure. This would be an extremely expensive fix.

    I like to take every opportunity to remind people about this fact in a bit more detail. The province deliberately designed the tunnel north of Ellesmere station to fit the ICTS Mark I vehicle to prevent the TTC from easily converting the RT over to any other technology. Turned out they designed it so well that Mark II & Mark III cars can’t fit through this tunnel either! That’s why the TTC can’t just order new rolling stock for this line.

    Like

  14. If only the Downtown Liberal Elites had shown care and concern for the people of Scarborough and created, years ago, some sort of plan for transit for the city–a “Transit City” plan, if you will–that would have had a replacement for the SRT up and running by now. It surely would have been embraced by Scarberians and in revenue service and not killed by some cokehead mountebank they elected.

    Liked by 1 person

  15. On the one hand I cannot imagine the SRT ever runs again.

    On the other hand, spending exorbitant amounts of money to get it up and running again by November for some 2 weeks of service, just to decommission it, is exactly the sort of thing I’ve come to expect from our politicians…

    Liked by 2 people

  16. Every option for the replacement/future of the SRT (new trains, LRT conversion, subway extension, etc), and the feasibility of each, has been debunked a long time ago, which unfortunately leaves us where we are now (replaced with nothing but shuttle buses, since it’ll be nearly a decade at best before the SSE is ready, at which point it’ll be a bit too late to say it “replaced” the SRT).

    One thing I don’t get is why weren’t the successors to the Mark I model (the Mark II and onward) designed to the same specs (specifically length) as the Mark I to begin with?

    Steve: The Mark I’s were small because of weight limitations on the early version of LIM propulsion. This affected car size, the amount of sound insulation (minimal) and the power supply system (designed to minimize electrical insulation needs).

    Once this was overcome, cars could be bigger and that’s why Vancouver cars (and Bombardier’s general offering) were larger. There was no technical reason to build small cars, and the larger cars reduced the cost vs capacity. The SRT was marooned from the outset with a bad spec.

    Like

  17. I like to take every opportunity to remind people about this fact in a bit more detail. The province deliberately designed the tunnel north of Ellesmere station to fit the ICTS Mark I vehicle to prevent the TTC from easily converting the RT over to any other technology. Turned out they designed it so well that Mark II & Mark III cars can’t fit through this tunnel either! That’s why the TTC can’t just order new rolling stock for this line.

    Isn’t it funny that the same face involved with the Scarborough RT is shilling for new “advanced”, “light”, and “automated” trains for the new Ontario RT line coincidentally called for specifications which are ever so smaller than standard TTC subway specs? They couldn’t even run 4-car trains like Sheppard if they could. On that note, I’m surprised he hasn’t restarted agitating for the North York RT line.

    Like

  18. IIRC, from over 50 years ago, the reaction rail was laminated sheets of aluminum to reduce eddy currents caused by the induced current from the induction motor. The current was greatest during braking and starting so there were more thinner laminates in those sections, just before, in and after stations. These laminates used to loosen over time and hum when there was an induced current in them. I don’t know if the laminates could have loosened enough to cause one to fly off with a motor over it.

    Also, during testing at Kingston, a LIM (Linear Induction Motor) “exploded” when the control circuit fed pure DC into the AC LIM. Actually the magnetic forces in the motor were so strong that the wires flew apart and destroyed the motor. Fortunately thee is a plate above the coil to keep the bits from entering the car but the effects below the car were spectacular.

    Like

  19. “On that note, I’m surprised he hasn’t restarted agitating for the North York RT line.”

    I suspect that the lack of anything happening on Sheppard and trying to not draw attention to that is the reason why.

    Like

  20. We had the ribbon cutting ceremony when the SRT opened. Will we have a ribbon bow joining ceremony when the SRT is “officially” closed?

    Like

  21. Steve: It is telling that after Montréal “copied” the Paris Métro and its rubber tires, that Paris stopped converting lines to that technology. As for the steel rails, there is little or no contact with them unless a tire deflates, and at switches where all of that business with vertical and lateral tire guidance would be prohibitively complex. Car size is an issue due to weight limitations, just as it was on the SRT and the original “Mark I” Skytrain cars.

    I’m assuming it’s only the flanges making contact with the rails at switches (assuming no flat tires)? Also, are the rubber tires firmly affixed to the same axle as the steel wheels, so as to spin as a whole, or can they spin independently of the steel wheels (do the steel wheels spin at all when not making contact with the rail)?

    Steve: They would have to spin independently given their different radii.

    Like

Comments are closed.