The TTC’s Low-Floor LRV Presentation of August 27, 2008 (Update 2)

The TTC has not published the presentation on their website, and in the interest of having the material in view, I am transcribing it here.  Few pages involve diagrams, and so for the most part I will simply transcribe the text.

Where I comment on the material, I will do so in italics to distinguish my words from the TTC’s.

The pages with diagrams are linked to scanned images.

Update 1 (August 30):  In the post below, I originally said that the chronology did not mention the change from a 70% low floor spec to 100%.  A quibble has been raised on this point.

The original spec had both 70% and 100%, and the change was to remove the 70% option.  The net effect was the same:  any 70% low floor car that might have been proposed was eliminated from consideration.  It remains to be seen whether, in fact, any builder can adapt their designs to Toronto’s track geometry.

Update 2 (September 16):  The full presentation is now available on the TTC’s website.  The links in the post below take you to scanned images on my site.  If you want better resolution, use the TTC copies.

Pages 1-2: August 27, 2008 Commission Meeting

LF LRV Procurement Project
Cancellation of RFP & Way Forward

Stephen Lam, P. Eng; Superintendent, LRV Engineering; Rail Cars & Shops
Sandy MacDonald; Manager; Materials & Procurement

The Toronto Transit Commission operates one of the most unique streetcar systems in the world.

Page 3: Unique Toronto Operating Environment

Pages 4-5: Importance of Appreciating an Operating Authority’s Environment

Derailment of Boston’s No. 8 Cars (100-car order signed May 1995)

Main Causes:

  • Carbuilder was permitted to use “San Francisco track geometry” for vehicle design due to lack of complete inhouse engineering data
  • Carbuilder was permitted to simulate derailment behaviour not based on vehicle being proposed


  • 4 derailments between April and July 2000; fleet grounded
  • Service resumed April 2001
  • 3 derailments between June and August 2001; fleet grounded
  • Changed wheel profile, upgraded track; service resumed in 2003
  • Vehicle derailed onto station platform in August 2004
  • Contract truncated at delivery of Car No. 85 at YE 2006

Massachusetts’ State Auditors’ Report Publicly Released February 16, 2007:

Audit Results:

  1. The MBTA’s inability to ensure that the No. 8 low-floor Green Line cars were properly designed for the Green Line infrastructure will cost the Authority approximately $101 million in additional contract and track maintenance costs.
  2. The MBTA did not adequately plan andtest the No. 8 low-floor cars, thereby contributing to a derailment flaw that caused a significant delay servicing its disabled patrons.

Steve comments:  The Type 8 cars were manufactured by Breda.  The MBTA has a long history of purchasing less than successful cars from a succession of vendors.  The TTC’s own experience with the CLRVs was far from perfect.  Those cars also had problems with derailments and accelerated track wear due both to the choice of wheels and the excessive vehicle weight compared to their predecessors, the PCCs.

Pages 6-7: TTC’s Procurement Process to Meet Challenges

  1. Analyzed technical risks & identified best practices
  2. August 15, 2006 — Advertised & issued Request for Information (RFI) to known carbuilders — 7 responded
  3. Summer 2007 — Public consultation
  4. On-going discussions with industry and internal stakeholders
  5. TTC and its consultants conducted:
    a) 3-D track geometry mapping to ensure compatibility of LRV with TTC infrastructure — data subsequently included in RFP
    b) Simulated LFLRV behaviour — ground borne vibration, overhead power capacity
  6. May to June 2007 — In-depth technical discussions with various interested carbuilders
    a) Safety Against Derailment — single point track switch & tight radius curves. Preliminary carbuilders’ analyses suggested requirements could be met with appropriate vehicle customization
    b) Gradeability — all wheels poweredc) 100% low floord) Ground borne vibration
  7. August 14, 2007 — Released pre-RFP specification for industry comments.  Suggested changes were incorporated while encouraging competition and protection the Commission’s interests
  8. December 2007 — Canadian Content requirements finalized
  9. January 17, 2008 — Issued Request for Proposal (RFP)
    a) 13 addenda issued
    b) no question was received during the 5 ½ month RFP period re: safety against derailment, wheel profile, coefficient of friction or allowable wheel climb
  10. February 11, 2008 — Pre-Bid Meeting
  11. June 30, 2008 — RFP Closed

Steve comments:  Not mentioned in the chronology above was the decision to move from a 70% Low Floor to a 100% Low Floor car in the middle of the process.  Although both 70% and 100% designs were originally acceptable, the 70% option was dropped. This changed the type of car vendors might bid.

Page 8: 5-Step RFP Evaluation Process


  1. Compliance of Proposals (Commercial)
  2. Pass/Fail Technical Requirements
  3. 25% Canadian Content Plan
  4. Qualitative Technical Evaluation Score
  5. Pricing including Life Cycle Cost

> Negotiate with preferred Proponent


  • Proponent must meet all requirements of steps 1, 2 and 3 above to have its proposal qualitatively technically evaluated (Appendix C — Instructions to proponents, Part B)
  • Evaluation process was structured, fair & transparent — all bidders evaluated on the same basis
  • Evaluation process was overseen by Fairness Monitor

Page 9: Step 1.  Compliance of Proposal

  • Form of Proposal
  • Bid bond
  • Proposal submission requirements

Page 10: Step 2.  Pass/Fail Technical Requirements:

  • 13 criteria in 5 general categories
  • Proponent must pass ALL of the 13 technical criteria

Page 11: RFP Submissions & Evaluation Summary

Two Submissions Received*:

  • TRAM Power Ltd.
  • Bombardier Transportation Canada Inc.

* Alstom Transportation Inc. and Siemens Canada Limited submitted “no-bid” letters prior to closing

Evaluation Team:

  • TTC Staff
  • External Consulting Engineers provided advice

Evaluation Summary:

  • TRAM Power’s Proposal failed at Step 1 — Compliance of Proposals (Commercial)
    — Did not submit required documentation including Form of Proposal and Bid Bond
  • Bombardier’s Proposal failed at Step 2 — Pass/Fail Technical Requirements
    — Failed Safety Against Derailment analysis

Steve Comments:  Note that the process has never progressed to Step 4, the detailed technical evaluation.

Pages 12-14: Simplified Derailment Theory

Steve comments:  These diagrams show the geometry of the wheel-rail interaction.  The TTC’s specification (see below) allows an angle up to 70° although the existing system runs at a lower value.  Street railway systems and the cars that were designed for them, notably the PCC, assume that track and wheels will not be in perfect condition and the trucks are designed to work in poor-maintenance settings.  The TTC’s track is in generally good shape thanks to years of reconstruction, but it is still challenging.  Moreover, special work at intersections is replaced on a fairly long cycle based on the assumption that existing and future fleets can handle the track as it ages.

Pages 15-16:  Step 2:  Technical Evaluation

TTC Specification

Input data for Safety Against Derailment simulation analysis requirements:

  • TTC wheel profile and existing TTC track data
  • Use Coefficient of Friction = 0.5
  • No wheel climb

Wheel Profile

  • Flange angle = 70° front and back, tapered bottom

11m Curves

  • 10 km/h operation
  • 70° front and back flange angle

14m End Loops

  • 10 km/h operation
  • 70° front and back flange angle

Single Point Trackswitch

  • 10 km/h operation
  • 70° front and back flange angle

High lateral force in the loops and curves causes:

  1. Increased risk of derailment
  2. Accelerated wheel flange wear and rail gauge wear (reducing wheel and rail life)

Conclusion:  Modify existing system or modify a standard LFLRV

Pages 17-19:  Modification to TTC Surface Rail Network

Infrastructure changes to expand approximately 90 tight radius curves and loops to 20m radius, including property acquisition, is not feasible.

In addition:

  • Cost prohibitive
  • Service disruption
  • Must have all curves and loops done to mitigate risk
  • Existing carhouses present major problem with physical constraints to expand curvature
  • Time line — 10 years to complete change

Steve comments:  The track map above can be a bit confusing because it shows two sets of tight curves marked with a check (somewhat tight) or an X (very tight).  The degree of the problem is obvious from the amount of the map flagged here.

The intersection diagram (King and Dufferin) shows how right turns built to a wider radius would overhand the sidewalk and would in some cases require building demolition.

Page 20:  Adoption “Standard” LRV Wheel Profile

Impact of adopting a “standard” wheel profile without expanding curve and loop radii:

  • Requires change of existing streetcar wheel flange angles
  • Requires frequent wheel machining to maintain steeper wheel flange angles
  • Still requires grinding of special track work
  • Not practical; expensive and presents unacceptable risk of derailment

Page 21:  Conclusions

The existing network or streetcars cannot be modified.

“Standard” LF LRV must be modified to be compatible with the TTC network.

Pages 22-23:  RFP Cancellation Activities

Fairness Monitor reviewed and supported the decision

  • “I am satisfied that both submissions were evaluated fairly and in a manner consistent with the RFP.”
  • “The RFP establishes the rules of the game, and in that sense its provisions must control the evaluation process.  In fairness to the two proponents who responded to the RFP, and to those who may have chosen not to respond to it, it would be wrong to treat the RFP as an invitation to negotiate.  Rather the RFP is a document that triggers a competitive process where proposal submissions are subject to evaluation as prescribed with clarity in the RFP.”

Cancellation of RFP letters issued to TRAM Power and Bombardier — July 17, 2008

Pages 24-26:  Developing Way Forward

TTC contacted carbuilders who responded to RFI or purchased a copy of RFP and asked a series of 5 questions

Results of 5 questions

4 Companies Considered:

  • TRAM Power does not have proven 100% LFLRV
  • 3 other proven Carbuilders have 100% LFLRV

Alstom, Bombardier, Siemens clearly indicated that they could provide 100% LFLRV that can operate on all of unique requirements of the TTC’s track system

Proceed with 3 proven carbuilders

Steve comments:  At this point, Škoda does not have a 100% low floor car in production, but is about to unveil one at a transport fair in Berlin.  The vehicle is designed for the Prague system.

Pages 27-29: Recommended Way Forward

Structured Multi-Phase Bid Process:

Phase 1 — Introduction

  • Invite Alstom, Bombardier and Siemens to participate based on proven experience in manufacturing 100% LFLRVs
  • Develop preliminary timeline
  • Commitment to participate

Phase 2 — Technical

  • Carbuilders to demonstrate ability to meet Pass/Fail requirements
  • Carbuilders to demonstrat ability to meet other technical requirements

Phase 3 — Commercial

  • Negotiate acceptable commercial conditions

Phase 4 — Competitive Bidding

  • Formal process for submitting pricing and Canadian Content plan

Phase 5 — Commission Approval / Award


  • Process structured & competitive
  • 3 proven carbuilders — 100% Low Floor LRV
  • Bidders engaged throughout process
  • Address questions/concens (Technical/Commerial)
  • Encourage participation/competitive bids
  • Formal process: pricing & Canadian Content
  • More likely to result in compliant bids


It is recommended that the Commission authorize staff to proceed with the Multi-Phase Bid Process to procure 100% Low Floor Light Rail Vehicles.

27 thoughts on “The TTC’s Low-Floor LRV Presentation of August 27, 2008 (Update 2)

  1. I’m amazed at how many curves are less then 15M. I’m assuming, however, that most of these have similar radii, and I’m curious how many are below say 12, or 13, and if that’d make any difference at all


  2. Steve, what other tram networks in the world use 11 m turns? Do you know if there are any?

    Steve: Although this type of curve would have been quite common when North America had many street railway systems operating within the confines of narrow street geometry, tight curves are now quite rare. Even in Boston which had all its problems with various cars, the minimum curve radius spec for the Type 8 is about 13.5 m.

    A big problem for low floor cars is the smaller wheel diameter and the changes this brings in the wheel-rail interface. Technical papers about this subject go on in excruciating detail about the interplay of various parameters and I won’t attempt to describe the situation here beyond saying that the tight Toronto curves lie well outside of the range of values for which “light rail” as opposed to “streetcar” vehicles are designed.

    The root of the problem is that two forces operate on the wheel: the vertical force from the weight of the car and the horizontal force from the track pressing on the flange. On a curve, the lateral forces are higher and a car is, in effect, constantly “falling” into the curve because the vertical force outweighs, literally, the horizontal one. The wheel tries to climb up the rail, but cannot because the vertical force wins out. On some worn curves, you can see corrugations created by this repeated slip, and of course once they start to form, following vehicles follow the same pattern and reinforce it.

    This is affected by the wheel and track geometry, and any proposal that requires a substantial increase in the standards of track maintenance puts the TTC in the position where a derailment could be blamed on track (or wheel) condition, not on an inherent design flaw in the vehicle. That’s why they are being so stubborn about the specification

    However, the fact remains that three companies claimed they could supply a vehicle that would operate in the Toronto environment. Now let’s see what they can achieve.


  3. If Siemens or Alstom come to the party they will have qualifications just like Bombardier. No one is coming to the party without a qualification . TTC will have to fork out some money to fix tracks which may vary from supplier to supplier.

    Steve: The question is just what modifications each vendor’s product requires. Bombardier makes a claim that their design will only cost the TTC $10-million without specifying the period. In the press conference, they seemed to have assumed that all of the existing vehicles have a very limited lifespan (sat 5 years) and therefore any work to make the two fleets co-exist is time-limited. This is a totally invalid assumption given that the 204 car order is not enough to replace the existing fleet plus additions to bring service quality back up, handle growing ridership and new lines.


  4. I really think some of the blame lies with the TTC on this because they refuse to grind the streetcar track, and grinding the rails is how the correct track profile is maintained – and one half of the wheel/rail interface that’s at issue here is invariably the track profile.

    The tight curve radii in many places in Toronto, as the TTC report correctly points out, can’t realistically be changed and only 100% low floor streetcars are not being considered for this purchase to the exclusion of everything else. This means that the range of technical factors involved with the streetcars that can be adjusted to accommodate them, regardless of whoever happens to build them, is going to be more narrow.

    That means that the TTC will have to eliminate conflicting variables elsewhere, which means the track is going to have to be maintained to a higher standard than it is now. This probably means the end of building track and then forgetting about it for the next 30-40 years and letting streetcars screech and rumble around tight, corrugated curves sending shockwaves through nearby buildings or letting broken rails on the bridges deteriorate to the point of safety related work refusals (Spadina bridge, Dundas bridge over Don River) or emergency repairs to make it last until rebuild (Queen bridge over Don River, Fleet St.) and grinding the track in locations where uneven rail wear results in non-standard profile that could cause tracking problems.

    On another subject, I’ve given the comparisons between the proposed new cars and the PCCs some thought and I’m beginning to believe that it isn’t a very fair comparison. The PCC cars ability to navigate bad track hinges on two key points: the design of the Clarke B2 and B2-B (as well as the B3 on Pittsburgh’s interurban cars) trucks and a lightweight car body that remains within the weight limit the trucks are designed to carry even when fully loaded and also reduces the amount of wear on the track. The low floor cars aren’t a fair comparison because they immediately rule out the ability to use the B2 trucks and their derivatives, which were the key to the whole PCC car. The President’s Conference Committee didn’t have the challenge of making a low floor streetcar that tracks as well as a ordinary PCC with the B2 truck placed off limits so it’s unfair to say whether or not they’d have done any better or worse than the people of today.

    Ironically, as a side note to the whole issue of lightweight car bodies, a lightweight car body can also be considered a disadvantage in some ways. One of your comments about the vertical force of the streetcar rolling along the track overwhelms the lateral force experienced by the streetcar as it rolls through curves; reducing the car’s weight also reduces that vertical force that is needed to keep the car on the tracks – often some parameters can work for you as well as against you at the same time.

    Steve: Yes, the PCC designers didn’t have to deal with low floors, but my point is that we have a system that was built around this technology and its track geometry. Maintenance will address the worst of the problems, but it won’t eliminate the general concern of whether cars are safe to drive on our network.

    Over the years, some of the decline in streetcar speeds came about because operators were worried about being able to stop quickly, especially with the ALRVs, and they tended to drive them less aggressively. Also, the TTC instituted a stop-and-proceed rule at facing point switches because they would prefer to leave a track switching system that is unreliable in operation for decades. There has been a capital budget project for a few years to replace them, but is has never surfaced as an actual current work project. Indeed, this may be waiting on a decision on new cars to ensure compatibility.


  5. I hope I’m stating the obvious, but: the new Transit City lines will be designed to avoid tight curves and other challenges, right? It’d be silly to create a brand-new network that would drastically narrow the field of potential LRVs to run on it.

    Steve: Yes, Transit City will be designed to “Light Rail” rather than to “Streetcar” standards.


  6. Sorry, I’d like to open my question to anybody who might have this information. Melbourne Australia has an extensive tram network from the same era as Toronto’s, yet they run plenty of Siemens and Alstrom low-floor vehicles. I am searching for what the narrowest turn radius is on that network, and I have been unable to find it yet. Thank you.

    Steve: I’ve been digging through many websites, but eventually I wound up on the Siemens site and located a document describing their products in various cities. Select the third document called “Ref.Light Rail_GB”.

    In this document, they state that the standard minimum curve for the Combino is 15m but that for Basel they went down to 11.8m. This document is a few years old and more recent information may be available elsewhere, but all of the Combino Plus info on the Siemens site is silent on minimum curve radii. Whether the value cited for the original Combino still applies I don’t know.


  7. luke: I don’t know the specific details, but I do recall hearing that Melbourne needed to do some track work for the Citadis to operate. Also, though their network is of a similar age, they’ve had various forms of modernisation along the way that never really hit Toronto.


  8. “the track is going to have to be maintained to a higher standard than it is now.”

    A very prescient comment in my opinion. Here’s my prescription:

    1. Overhaul as many CLRV/ALRV vehicles (hereafter collectively “CLRV”) as reasonably possible with an Out Of Service Date prior to 31.12.2024 (per the Ontarians with Disabilities Act) but capable of running right up to that date as the TTC are certain to order less than they will need until they are finally forced to withdraw them (likely with an appeal to delay to 2034).

    2. Don’t make Dublin’s mistake (DART overhaul) by making the first cars to be overhauled the ones that have been rotting as parts hulks which means the overhauler has to spend months trying to figure out what a working one looks like.

    3. Instead of ordering a CLRV replacement immediately, look hard at what lines could reasonably accommodate Transit City cars using a limited number of crossovers and tailtracks on routes like 501, 503 (using King turnback rather than Wellington St.), 508, 512 and Cherry-King, WWLRT too maybe but that would involve a terminus prior to Union Loop. Basically we’re talking about routes that require <15m radius turns on the route rather than stations, loops and yards. 512 might have to lose the paid-zone loops in this scenario though, and access maintenance at the Eglinton TC yard with 501, 503, 508 and Cherry doing so at the Portlands facility. 512 also makes sense to make low floor early since the ROW was built with accessible features (at page 8, 2.4.1) and problems introducing the cars would not cascade into interlining/connecting CLRV routes as a derailment downtown would. TC car introduction would essentially jump the queue in this scenario which might be no bad thing given that those LRVs would require fewer modifications from existing models (gauge, power, internal fit, signalling, passenger announcements, next gen CIS, etc.)

    4. As a lowfloor tight radius car comes available over a longer term to serve routes like 504, 509, 511 etc., perhaps developed in co-operation with similarly affected US systems with tight radii, these could replace the CLRVs and ultimately the “downtown” TC cars which would in turn be transferred to the suburban TC routes as they expand post 2025 – possibly with the addition of extra segments to become more “light rail” than streetcar just as Dublin has expanded its Citadis fleet from 30m to 40m with yard fitted modules.

    The danger in this scenario is that TTC might attempt to close routes that can’t be served by the TC cars under any circumstances (where a tight route turn can’t be avoided) rather than follow through with the CLRV replacement.


  9. A number of observations:

    1. The number of intersections and loops with the tight radius problem is far greater than I remember reading before.

    2. The force balance equations/analysis for this issue are very complex EVEN FOR PERFECT CONDITIONS. I think the TTC is right to be conservative in its assessment of the track changes that might be needed to accomodate the Bombardier cars as proposed.

    3. Grinding track – as I understand it – would have some drawbacks (noise, disruption of service and potentially shortenning track lifespan.)


  10. If the “force balance equations/analysis for this issue are very complex EVEN FOR PERFECT CONDITIONS”, the TTC has to make track modifications (hopefully limited) and has to maintain the tracks and car wheels. If grinding is not ‘friendly’, then pull out the tracks and install new ones.

    TTC has to be conservative in their evaluation of costs for track changes/maintenance.

    Cost for new steet cars equals

    1. Cost for initial track changes
    2. Cost for steetcars
    3. Track/wheel Maintenance costs

    Anyway, the streetcars shown on the Siemens web site look great.

    Steve: The TTC used to do rail grinding religiously using a pair of elderly passenger cars converted to rail grinders, then later a PCC rail grinding train. Eventually this stopped probably due to budget cutbacks and the difficulty of maintaining a pair of PCCs just as work cars. The purpose of the grinding was mainly to eliminate corrugations and associated wheel noise.

    Before we start talking about wholesale changes to the infrastructure, we need to know what the various proponents can offer and whether it is technically feasible to stick with the 100% low floor specification at least for the “legacy” system of streetcar lines. As many others have pointed out, we seem content to run a fleet of buses with mixed floor heights and are asking the streetcar system to perform well beyond the level we expect of the buses.

    Transit City requires a large enough fleet that a 100% low floor car can be procured for that environment. The remaining constraint would be if these cars need to access the existing main shops at Hillcrest, then trackage on the connecting route would have to be compatible with the specs for that fleet.

    The TTC can paint itself into a corner where the cost of retaining streetcars is excessive because of an unattainable spec, or they can compromise. We need to balance the capabilities of the carbuilders, the cost of a 100% low floor car for Toronto geometry and the cost of revamping the entire streetcar track structure to suit the specifications of available cars.


  11. Serious question: why would it be necessary for Transit City cars to access Hillcrest? Since we’re more than doubling the fleet, why not have a TC major yard?

    Steve: An equally valid point is that since the network will be built bit by bit, there will be a long time before it is sufficiently connected to get cars from, say, Sheppard East, to Hillcrest.


  12. I’ve heard from drivers that only recently is the TTC close to completing upgrades and renos to the existing carhouses that the introduction of the CLRVs and ALRVs neccessitated 30 years ago. If this is the case, let’s hope they’re more on the ball this time around. The drivers who have told me this tidbit didn’t seem inclined to hold their collective breath.


  13. Firstly, are the new vehicles not supposed to be bi-directional? How many of the 90 “tight turn” locations can be eliminated because the vehicles won’t need a loop to turn around or short-turn?

    And secondly, why hasn’t the TTC examined the route structures to rank the importance of each loop and tight turn? Are there more critical turns that are used every day, versus very rare turns that could be simply removed from the system?

    I can’t imagine that locations like the Kipling Loop, or the tracks along Wellington, are as important as the locations along the College/Carlton route, where the very functioning of the route is jeopardized by tight turns?

    Why put red X’s on ALL the locations without reviewing which ones are most important?

    Steve: One big problem with replacing loops is that the amount of on-street real-estate for crossovers plus layover areas is restricted. For example, at Neville Loop, there would have to be room for at least one car to sit in the middle of Queen Street awaiting its return trip. Whether this could be accommodate by a short extension onto the lands of the Water Works I don’t know, but not every terminal has this capability.

    Typically, any right turn at a standard Toronto intersection where two four-lane streets cross at 90 degrees will be challenge (there is a diagram of King & Dufferin in the presentation). There are many places like this that are integral to regular operations (Broadview at Queen and at Dundas are two close to my home base, but there are many others). This is not the sort of argument that can be dismissed by pointing to streets like Wellington because much of the track used for diversions around special events is on 4-lane streets.


  14. Hi Steve:-

    Just wondering how TTC’s two and three door steel trailers and the inherited self propelled single truck ‘Birnies’, all with small wheels stayed on the track? These cars, certainly when new, would have run over the scary private company’s ribbons of worn and bent rust for a period before that track was renewed to TTC’s early 20’s standards (virtually what we have now as far as curve radii goes). These vehicles ran for another 20 to 30 years before being retired or sold off for continued use elsewhere, seemingly without the constant fear of derailing during all those years in Toronto service.

    Yonge Street trailer trains operated over some of the most horrendous deteriorated track conditions (conditions that rivaled the crumbling track structure assumed from the old TR) on a 1-1/2 minute headway when proper maintenance and renewal ceased with the Yonge car’s imminent replacement by subway and trolley coach in sight in the early 50s.

    Has street railway technology not kept up with vehicles that can perform under less than optimum conditions, as the old clunkers that used to run in almost every city in North America once did daily? And too, these trailers and ‘Birnie’ cars didn’t have the luxury of being equipped with the extremely flexible Clark truck which carried the PCCs. Are we being too fussy in our demands of what a streetcar should be able to do? The worst of the old street railway technology is still a quantum leap better than a diseasel bus! And I’ll ask again. Why is it essential that streetcars be 100% low floor when a 100s strong fleet of so called low floor ‘free-wheelers’ are a far cry from even being close to 70% low floor let alone the difficult to achieve 100%? Ever notice that the low floor bus doesn’t have small wheels to minimize the floor space sapping intrusion of wheel wells? Is this a double standard demand or what?? Am I pointing out the Emperor’s new clothes?

    Does the dream world requirements of a streetcar bid for what some day may be easily achieved, rather than with great cost and difficulty it appears at present, need to be inflexibly adhered to just to save face? Could we not satisfy the legislated needs with a vehicle that will be less than 100% low floor for now and then when the industry catches up to the pie in the sky desires of the ultimate wishes to be included in a transit vehicle, then start introducing them into the equation, what, maybe another 20 years down the pike? Meanwhile we would be gaining operating experience with a compliant streetcar, one that is only partly radically new and be providing the reliable service that streetcars are capable of!!!

    Remember when the TTC was amongst the best transit properties in North America; how they resisted jumping on the new band wagon just for the sake of it being new, but waited until the bugs were worked out first before committing? Surely to goodness there is a streetcar on the market now that is tried and true and that will fill the legislated bill? Look back to the good old days everyone, because erring on the side of conservatism (not politically mind; technologically speaking) once was and can still be an important way to think!

    And yes, until the new cars are on the property, let’s get some (most) of the old CLRVs rebuilt. Once new cars are indeed a reality (which they aren’t anywhere near to yet), we will at least have a fleet of cars that we can rely on to handle existing conditions (better the devil you know, you know). It doesn’t have to be a rebuild to like new condition, but more a determined mechanical and electrical consideration so that break downs and out of service time will be minimized.

    Even, heaven forbid this thought; could a low floor centre section be slung between newly motorised trucks on a TTC ALRV; somewhat like Dallas’ recent and continuing capacity building purchase of new centre sections? Just a thought!

    Dennis Rankin


  15. When we talk about derailment theory, are we simply talking about a streetcar sliding out of it’s routing (which would undoubtedly cut the vehicle’s power) or are we actually talking about a vehicle flipping on it’s side (giving way to physics)?

    Steve: Streetcars fall over only when they are driven into curves at a speed much higher than prudent operating speed. A derailment simply winds up with the streetcar (likely only one truck) sitting on the pavement.


  16. Dannis Rankin makes several good points. The key amongst them is the desire for politicians and their Boards to meet “feel good” requirements that are not entirely practical. I challenge anyone, nothwithstanding the boarding/egress issue, to tell me how a wheelchair could fit on a five PM streetcar westbound at King and Bay. It is difficult for all the ambulatory individuals to fit on these cars. We have purchased a whole fleet of inadequate buses based on the desire to be “accessible” but they have grabby brakes that makes it difficult for even the marginally disabled standee to cope. In order to meet that standard, the number of seats have been compromised. Lately, we have wasted 33% of our bus purchase budget on “hybrids” that get little or no improvement in diesel consumption. Politicians and their Boards keep steering us in directions that not only are unfriendly to transit, but also, in the reality of the world, end up not achieving the “noble” goals of these people. In fact, in the end the noble goals are further from achievement.

    At the very least, faced with a deteriorating fleet of streetcars, the CLRVs and ALRVs should be aggressively rebuilt. There may or may not be 100% low floor “streetcars” suitable for Toronto service in the future and that “may or may not” actually be a desireable goal. In the meantime we need to have reliable streetcar service.

    I am fully ambulatory, but if I have a bag or two I fear the possibility that I cannot get to the door to get out on the dysfunctional Orions and the overcrowded CLRVs. It is beyond me how the frequency of service and crowding conditions are ever going to allow for effective “mobility” for those confined to a wheelchair, even if the access/egress is theoretically possible.

    Society does owe “mobility” to all its citizens. That is something we should pay for without complaint. It is less obvious that the mobility in question can be provided by an accessible but inefficient mass transit system.


  17. It would seem reasonable to pick a 70% low floor, reliable design for the old streetcar network, and pick a 100% low floor car for Transit City. Over the next 30 years as they rebuild track, try to widen out those tight radius curves to meet the design expectations for the 100% LF cars. In 30 years when the 70% LF cars are ready for the scrap heap, you have fixed the issue, and just pick an off the shelf car, to replace both fleets.


  18. OK – it would be nice if buses were 100% low floor – but operationally, there is little advantage because passengers are almost always boarding at the front. For the LRV, the 100% low floor would seem to have an advantage of better balancing loading along the vehicle – which is important here because loading time is currently the biggest thing slowing down the service.

    Now – I’m stil not clear how someone in a wheelchair is going to board from street level. Even low-floor means a big step up from street level. (The low floor buses kneel at the curb.)

    Steve: Yes, this is a major issue. Where roads are wide enough to have loading platforms (Spadina, St. Clair), there can be ramps up to platform height for level loading. There are proposals for sidewalk widening in selected locations to narrow the road to one lane at streetcar stops, but this runs headlong into objections from cyclists.


  19. A question about switches that has been bugging me for some time: why are the single points of the single point switches on the side where they need to “pull” the car into the curve? If they were on the other side, it seems that they would “push” the car into the curve, or “pull” it straight (but pulling straight isn’t very hard, so should avoid the problems you’ve discussed with “pulling”).

    In a two-point switch, does the point on the “pulling” side actually do any work?

    Of course, the above only makes sense in switches which choose between going straight and turning, but that must be almost all of the switches in the system.

    Steve: The “pulling” side is not just at the switch itself but also on the curve (including curves that are not at junctions). The wear point is on the “inner” rail of the curve. The situation is different on railway/subway rails where the curves are not as tight, and the outer “pushing” rail does the work.


  20. As I understand it the back side of the inner wheel is used to guide the vehicle through the turn because it offers far more surface area. Street-based rail vehicles use shallower flanges than railways or subways. Combine this with lighter vehicle weight and you get a much higher risk of derailment without a reliance on a wing-rail on the inside of all sharp turns.

    Subways actually use a number of guard-rails for extra guidance that you won’t find on a railroad because they must navigate much sharper turns at higher speeds than a railroad could ever be capable of.

    A double-blade switch in a streetcar system still provides flange guidance on the inside of the turn if necessary. However the switchblades don’t have to anywhere near as strong as a single blade and generally flex rather than pivot to select the route.


  21. Much thanks to Mark for the Skoda link! I found that Google’s Czech-to-English page translator did a remarkable job of converting the page to English and I would highly recommend viewing the page this way. Here’s a direct link to the translation:

    The Skoda 15T really does appear to be the only ready-to-roll product that will suit Toronto’s needs. (And heck, it’s nearly in TTC colours already!) While it is unfortunate that the passageways through the articulations are so narrow, this is a small price to pay for real pivoting trucks and a far better solution than jacking up the seating over the wheels. The prototype is only 13cm narrower than the CLRV/ALRV so there isn’t likely much interior width to be gained in a production vehicle for Toronto, but our wider track guage may also help.

    Every wheel is individually powered and independently rotating. I suspect 16 separately-tuned motors can handle any traction needs we might throw at it. I have to think that the technical aspects of this vehicle (and any future double-ended version) would make it an excellent candidate for both the legacy system and Transit City lines. It would be awfully dumb to buy vehicles for Transit City that would not be able to enter the lagacy system at all even just for non-revenue moves.


  22. Hi Steve and Isaac Morland:-

    All of TTC’s street railway switches allow a car to be directed straight or on a right or left hand curve. None will allow for the option of turning right or left (in railway terms a wye switch) without the straight. The switch casting (with its movable switch point) is always on the inside of the curve and the other matching non-movable side casting (the outside of the curve) is called the mate. Since street railway curves are sharp in relation to other railway practices, little extra protecting rail design is needed when the straight option is chosen, but when curving extra design comes into play.

    The switch point is not flat across the top along its length. It is lower at its tip than the casting in which it nests. There is a gradual elliptical transitional rise in the height of this piece of track hardware so that it becomes higher than the casting it sits in from near its tip to the heel (the pivot point or hinge point of the point). This height is necessary so that when the car direction desired is to take the curve, the back of the wheel, wanting to go straight of its own natural volition and at the same time clawing into the steel and trying to climb over the switch point (which has now become a check or more commonly a guard rail) pushes back against those forces (remember, for every action there is an equal and opposite one) against the back of the wheel and forces it to go in the direction chosen.

    Why choose to have the back of a wheel engage a check rail versus the flange of the outer wheel as the guide? Safety is the primary reason. Flanges are designed ideally to not contact the rail at all. Ideally the cone of the wheel tread geometry should see that this never happens, but if it should then the shape of the wheel profile is such that the flange will engage as a last resort and due to the slope of the contours it will quickly drop back down due to car weight. On sharp curves, this ideal is overcome by the centrifugal forces of the car being forced off its straight path thus then guard or check rails are introduced into the special track layout. Consider too, why this is indeed safer, for as the back of the wheel, flange and all, is touching against a checking face that is higher than the running surface of the rail head, there is much more wheel being persuaded to turn when it does not want to!

    Since the two wheels on the same axle are for all intents and purposes for this discussion one solid piece, pushing on the back of the wheel now negotiating the inside of the cure, will cause the outer wheel tread to be dragged across the rail head on the mate casting and therefore cause it to travel down the curved path itself. As long as the wheels are in the curve, once off of the two parallel castings, they will travel along ordinary girder rail. On the inside of the curve this rail will be girder guard or in TTC parlance, check rail. The outer rail will be the same kind of rail but will be called reverse check since it is bent lip side in; the lip being the raised portion of the rail forming the flangeway (the path where the wheel flange travels) and thus not the riding surface (horizontal), but the reacting surface (vertical).

    It has proven to be an extremely simple and safe way to divert a streetcar’s direction and this design of switching tracks has been used in almost all of North America’s streetcar systems for over 100 years. Simple, since only one point moves thus minimizing the quantity and mass of hardware required to toggle or lock it into place (when a manual switch, which all trailing switches are) (little used facing switches will remain manual too). Too, when electrified and thrown by a big honking solenoid under a street surface cover plate, there needs only be one adjustment instead of two which would be necessary to get the second point to move in concert and proper alignment with the first one. And one needs only purchase one switch point and casting for each switch. A mate is a far less costly bit than the switch.

    Earlier I mentioned the reverse check. Seems odd doesn’t it that when the centrifugal force of a streetcar being asked to do what it naturally doesn’t want to, i.e., curve instead of travel in a straight line, why would one want to guard against a wheel wanting to go to the centre of the circle when the streetcar wants to go away from curving? The lead axle attests to this thought but the trailing axle of each truck wants to go to the centre of the curve, so the reverse check is checking against that possibility. Next time you get an opportunity to watch a streetcar curving, watch the lead axle, it wants to go straight, but so does the trailing axle. It wants to help the lead axle guide the car’s truck into a straight alignment with the carbody by moving to the centre. After the car has gone by, if you’re not putting yourself into harms way by playing in traffic, look at the wear patterns. If an older installation, one should see the evidence of those trailing wheels riding against the lip of the reverse check rail. The forces on the reverse check are exponentially less than guarding for the lead wheels though.

    There are quite a number of other factors that went into the design of TTCs switch and mate castings and all have an interesting story on their own. I am not aware of any other street railways using exactly TTC’s designs, but the whole thought process is the same. There are two different lengths of switch and mate castings in present use in Toronto. They are 12 feet and 15 feet. They are designed to be used when curves are relatively speaking, broader or sharper.

    As to the track switches used in the subway, the design is more like regular railway use, in as much as they have two switch points, but the subway design uses much more hardware to make them safer than ordinary railway track, for subway switches have an extra piece called a kicker plate which works in concert with a restraining rail (yet another name for guard rail). The diverging side switch point also has a check rail feature like its counterpart on the surface which contacts the back of the wheel after rolling past the kicker plate. This kicker plate, restraining rail and switch point combination cause the back of the wheel to come into play on the diverging direction so that they can do exactly what the checking face on the back side of a street car switch point does. They work in concert so that the wheel flange on the outside of the curve does not come into contact with the opposite side switch point, thus minimizing wear on both the wheel flanges and the points. Of course, once kicker plates wear after many months of use, the flange will contact the outer point and one now has an indicator that maintenance and or replacement should be looked into. I believe this overall design if not in detail but in intent was adopted from the NYC subway from their 1904 designs.

    A lot of trail and error went into designing street railways once upon a time and some of what has gone on in recent years is an evolution of those past practices. Sometimes too, some of the trial and error which eventually proved a reliable and safe way to run streetcars has been forgotten or discounted as being old and there fore should be left to antiquity, not to be taken seriously in the space age. That has and can prove to be folly. Don’t discount the past out of hand!

    Dennis Rankin


  23. Hi, I’ve been following the Toronto light rail story with interest from here in Sydney, Australia. We have 5 cities that either have light rail/tram or are making moves to get or expand it. As you probably know, Melbourne has a big old system, probably with many issues in common with Toronto in terms of quality of some old trackwork etc. With new low floor trams like Citadis/Combino/Variotram I find we have that issue of unsuitability for city streets with sharp corners where these trams (with their virtually fixed, non-pivoting bogies, as I understand) are rough on bends as the entire body sections try to track the curves. Even a problem in Sydney with new modern trackwork. Adelaide has gone for normal-bogied Bombadier Flexis but pays the price with part high-floor.

    Here too we’ve noticed the advent of the Skoda ForCity(15T) and the way it addresses this issue by brilliantly achieving pivoting bogies whilst remaining 100% low floor – a major breakthrough that seems to put all the other trams in a previous generation. (Certainly huge wheelboxes in the main cabins and their effect on seating arrangements was never a good solution.) Kristian above seems to have similar thoughts.

    What puzzles me then is why Skoda-Inekon has been excluded from the tender process in Toronto? I know that, earlier, Skoda requested consideration of 70% low floor – presumably as the 15T wasn’t ready then. But now they have the new tram why shouldn’t it be considered? They are hardly an inexperienced tram manufacturer and designer (Czechs were the world’s largest tram-manufacturing country for most of the second half of the 20th century!). Would be silly to exclude what could be a perfect design and many years to pay for such a mistake.

    It seems to me that TTC is taking a bit of a chance with a whole lot of manufacturers, none of whom seem able to guarantee that their product will work in Toronto’s environment. It almost seems like one of the best (if a little costly) answers would be to get a demonstrator from each manufacturer and do real-world trials.

    Would be interested to hear any thoughts because the Skoda would be a very strong contender in principle for next generation trams here in Australia. It’s power and pivoting bogies are certainly very relevant for when Sydney’s system is expanded (we are hilly in places!).


  24. Here’s a novel idea, as quoted from a news item on the Skoda website:

    According to Radovan Šteiner, City Councilman for transportation matters at the City Council of the Capital City of Prague, two test prototypes of the ForCity trams will start appearing on the tracks in Prague as early as the first half of 2009. “The prototypes will differ from each other due to their various components, so the best configuration of equipment for series production will be selected for Prague based on experience from operation. The first vehicles will be put in regular operation by the end of 2009,” added Radovan Šteiner.

    Testing various configurations in real life on the real system BEFORE committing to a final production design? Why would we ever want to try THAT in Toronto?


  25. I’ve found out that Skoda-Inekon didn’t have the ForCity in consideration for Toronto – it was an Inekon design. However I don’t think you can get the ForCity anyway because its turning radius is same as T3, 18m in service, 15m in depot. I don’t know that anybody could design a 100% low floor that will do Toronto’s 11 metres, you might have to ease the curves like Melbourne did. But, at 20m minimum curves, we’re fine in Sydney thanks! Will be interested to see how you go there, good luck.


  26. Adam Giambrone keeps telling me that the new LRT system in Hong Kong (named “MTR Light Rail”, operating in the “New Territories” since the 1980’s) was designed based on the Toronto streetcar system. Do you know what he’s talking about?

    Steve: Nope.


Comments are closed.