Preliminary Specs for New Streetcars

Tony Turrittin of Transport 2000 sent along a copy of a note explaining to potential bidders the kind of requirements that would appear in the specifications for new streetcars.  I am publishing it here to save on answering ongoing queries about what the TTC plans to do.  Note that this is a shortened version of the original text to retain the core information about the Commission’s technical requirements.


Stephen Lam, P.Eng.
Superintendent – Vehicle Engineering
Vehicle Engineering Department
Toronto Transit Commission

I am sending this note to all who have expressed an interest in our Low Floor Light Rail Vehicle Procurement project through the submission against our RFI, which closed last November.

I want to let you know of the Commission’s position concerning floor configuration of the new vehicles, as a result of my meeting with our senior managers and the Chair and Vice Chair of the Commission.

The direction was that the new vehicle shall have at least a 70% continuous low floor area between the end trucks, AND with all axles powered.  From an engineering perspective, we recognize that the combination could present a challenge for some of you.  We believe, however, this arrangement is necessary to ensure that the low floor vehicles would have the best chance of success in our stringent operating environment, while providing accessibility and good passenger flow within the vehicle.  I would like to qualify this requirement further with the following principal requirements:

  1. car length can be between 27m and 29m (approximate), meeting TTC’s maximum dynamic envelope;
  2. modular design to offer flexibility for possible shorter vehicle length for subsequent order/option;
  3. all wheels powered – must meet 17% adhesion limit, 8% grade-ability, and be able to push a disabled vehicle of the same type up an 8% grade from stationary in an emergency recovery operation;
  4. minimum aisle width to be confirmed;
  5. Documentations are required to address the following issues.

More rigorous analyses and simulations are expected if powered stub axles and “independent rotating wheels” are offered.

  1. safety against derailment;
  2. minimum wheel unloading that can be accomplished;
  3. compatibility with the Commission’s single-point trackswitch;
  4. “sinusoidal” running on tangent track for maximum stability and minimum flange bearing with 1:20 wheel tread conicity;
  5. track parameters and special trackwork (warp, crossfall, superelevation) standards required;
  6. suspension system, unsprung mass, wheel radial and axial stiffness;
  7. ground borne vibration.

As a general program note, I would like to advise that the procurement will likely be divided into two phases – 50 cars for Phase 1 and 154 cars for Phase 2.  The absolute quantity of the phase 2 purchase may change depending on:

  • number of shorter vehicles required, if any;
  • passenger carrying capacity of the vehicle selected; and,
  • headway requirement calculations at the time of exercising the option.

This total quantity of 204 cars may increase in future as it does not include vehicle requirements to meet the City’s ridership growth strategy and expansion plans.

A technical meeting is planned for mid-June for interested bidders, and the Request for Proposal should go out in November.

The splitting of the order into 50 + 154 cars is a budgetary fig-leaf to keep the bean counters happy.  If the initial commitment is only for 50 cars, the amount of future dollars “above the line” in the Capital Budget is smaller.  “Below the line” projects don’t have committed funding.

27 thoughts on “Preliminary Specs for New Streetcars

  1. These low-floor streetcars sound great. I have a couple of questions though; are these new cars for the existing streetcar routes or are they for the new Right-Of-Way routes planned in the Transit City proposal? How many passengers do they carry as opposed to our current streetcars as well?

    Steve: At this point, the primary role of the new cars will be on the existing system. However, expansion into new routes may take place either with the same cars, or another car from the same “family”. It’s far too early to tell on that issue. As for capacity, one new car will be roughly double the capacity of a CLRV or 1/3 more than an ALRV.

    A related issue will be the presence of a low-floor section and all-door loading with some form of proof-of-payment fare system. This will make a big difference in dwell times at busy stops as well as distribution of passengers throughout the cars. Low floors are handy not just for those with mobility problems, but they avoid the delays needed for people to climb up and down steps.


  2. Is there a specific reason for specifying a car length of 27m-29m? I notice that this is longer than the ALRV, which is 23m. Is this the maximum length that certain loops, platforms or yards can accommodate? Would it be possible to run streetcars longer than this on certain routes, with only minor modifications to the platforms and loops? This probably won’t be needed in the near future, but ridership growth might make this necessary eventually.

    Steve: The length specification comes from various factors all of which are constraints on the existing system. The length of safety islands and layover positions at loops is one obvious one, but there are also clearance issues if the truck spacing gets too long — the car swingout on tight curves won’t clear physical obstructions and vertical curves can be an obstacle.


  3. These specs seem as though they were taken from the Bombardier Flexity Swift Car. They are used in Croydon England, Minneapolis St. Paul Minnesota, Cologne and Frankfurt Germany, Istanbul Turkey, Porto Portugal, Stockholm Sweden and Rotterdam Netherlands.

    I have ridden these cars in Croydon and they give a nice ride, load fast and have the advantage that there is only one low floor truck. The outer two trucks are a more normal design. The high end sections reduce the number of articulations by two and get rid of two low floor trucks which are a source of derailing problems.

    If you go to the Bombardier site and select English then Transportation, Bogies, Light Rail Vehicles and look at the S1100 and the BM1000 trucks (bogies) and I believe that you will see the middle and end trucks for Flexity Swift cars. Note the the centre truck has four, 45 kw motors and the end trucks each have two, 100 kw motors for a total 580 kw of power. This is about 770 horse power so it should have plenty of acceleration and power to push a dead vehicle as lon as the wheel do not slip.

    If you go into Transportation, Rail Vehicles, Light Rail Vehicles, Flexity Swift, you can get the spec for all of these cars that they run in the world. Their length is about 28 m. The Minneapolis cars have a minimum curve of 20 m (66′).


  4. I don’t see any mention of whether the vehicles are to be bi-directional, like most of the world’s modern trams, or continue the TTC’s tradition of building them like buses, with only one cab and doors on the right side only. Has there been any indication of what their plans are, or will this be left to the bidders?

    Steve: This is unclear, although given the intended use on new lines, we may see bi-directional cars.


  5. If a CLRV runs every 5 minutes, then if they put these double-length cars on, won’t they run them once every 10 minutes?

    It’s better to have smaller cars run more often than longer cars running less often. Order smaller cars and couple them together when needed. I don’t understand this at all. If this goes through, I can just hear everybody complaining about a streetcar that comes once every 45 minutes when a gap happens.

    Steve: This is a point that I and others made during early stages of planning the new order. Note that the quantity is for about 200 new cars each of which would be equivalent to two CLRVs. If we take the existing fleet as 196 CLRVs plus 78 CLRV-equivalents (52 ALRVs times 1.5), we get 274. This means that the new fleet would be much larger than the old one if all of the cars were used on existing routes, and significant service improvements would be possible.

    I agree that running “new” cars on even wider headways than some of the existing services, especially off-peak, will only drive riding away, not attract it to the system.


  6. Andrew MacKinnon wrote, “Is there a specific reason for specifying a car length of 27m-29m?”

    In addition to the TTC’s own reasons for specifying this, this is a pretty industry standard length range for LRVs (close to 90′ for those thinking in that mode).

    One of the exceptions to this is the SLRV developed for Dallas where they take a short low-flow section and add it to the middle of their existing LRV (either new or retrofit) making a 38m (124′) car.


  7. With the new Cars, will they be Bi-Directional or the standard stupid One Sided Streetcars we have today?

    I think we need more crossovers and less loops in our streetcar system which:

    A) Makes it look more like a subway, a move to start implementing if we are going to sell LRT as a proper alternate.

    B) Reduce the amount of track necessary, which means less track to maintain, which means less costs.

    Steve: My suspicion is that the new cars would be bidirectional, certainly any fleet that is bought for the new Transit City lines. The elimination of the need for loops cuts down on curves, but they are replaced by crossovers. Also, this type of operation is only practical if cars can take terminal layovers in the middle of the street. Crossovers are ideal for emergency short turns, although I’m not sure I want to encourage any more of those than necessary.


  8. Is this order considered to be a replacement for the entire streetcar fleet? I was under the impression that the ALRVs had a good number of years left in them.

    I would like to see hybrid vehicles that can run off battery for short distances. I certainly wouldn’t miss the ghastly electrical “spider webs” we see above some intersections.

    Steve: Yes, the intention is to completely replace the CLRV and ALRV fleets so that we would have a 100% accessible streetcar network. As for batteries and overhead, that’s the least of your problems. There is the small matter of track without which you don’t go anywhere. Battery power for a vehicle as large as a 28m streetcar would add substantially to its weight and be of limited use.


  9. “Also, this type of operation is only practical if cars can take terminal layovers in the middle of the street.”

    Yes, you’re correct about that. I was thinking in terms of the terminals of the lines. Example the Humber LOOP can instead be the Humber terminal where the streetcars can just use a crossover and change directions. The Neville Park Loop can be eliminated, the land can be sold, and I’m sure, in the literal dead end of Queen, they can just take out 2 lanes for a KM so that they can put a crossover. Spadina Station is an exception because it’s the end of the line same with the current and future Harbourfront Station.

    However, yes reducing the amount of “Loops” will also reduce the amount of curves it has to handle. For St. Clair, why don’t they take an extra block of median so they can make a crossover? Doesn’t that make more sense?

    Also, for Spadina, isn’t a crossover better then having a streetcar leave its ROW loop around and rejoin?

    Toronto’s LRT system still has hope, small changes like this certainly will go along way. And doors on both sides are vital also to making it more “subway like”.

    Steve: I think that the good folks in the Beach might object to losing so much of Queen Street to a replacement for Neville Loop. If we are going to embrace in-street turnbacks, they belong on new lines where there is room. Selling Neville Loop would give a very short-term capital gain and have a major impact on the street forever.


  10. Luke, just some information here. The Bombardier MITRAC technology does allow a tram to lower its pantograph for up to 1000 meters. It captures the energy lost in braking and sends it up to capacitors on the top of the tram. The capacitors can release large amount of energy for a short time much more effectively than batteries.

    It is a good idea. Imagine running the Queen tram with lowered pantograph in front of Old City Hall. The tourists should love it.


  11. I don’t find overhead wire to be any problem at all..

    If you take a look, you will see there is much more overhead for residential and commercial then there is for transit services.

    My one question about these cars is whether we will be keeping the trolley pole or finally making the switch to pantograph?

    Steve: We don’t know yet, but if there is a changeover, it will likely be gradual as the existing fleet is phased out.


  12. I think that we should have both crossover and loops at the same time co-existing so that the current vehicles could still use the tracks and make the same emergency short turns like the new ones. If we switched to crossovers only, the old ones would not have the same benifits as the new vehicles and would be useless on the roads. This would cause a shortage of vehicles, leading to congestion and delays.


  13. Do you know whether the TTC is planning to implement a requirement in the tender that the new streetcars, or a certain percentage thereof, be made in Canada?

    Steve: Yes.


  14. This would mean streetcars between 90 and 95 feet long, which is longer then the current ALRVs, but not quite as long as two CLRVs. IMO this is a great idea for busy routes, but some streetcars run out to Kingston road on 12 minute headways or more, even up to 20 minutes. That 20 minute headway could turn into 30 when the TTC starts adding and subtracting “per-seat” ratios. IMO they’d also need a fleet of smaller streetcars, or they should replace those specific routes with articulated busses.

    Steve: The real issue is that the minimum headway on major routes should not be as wide as 20 minutes. It’s odd that the Coxwell bus runs more frequently than this in the evenings and on weekends, but the streetcars come once in a blue moon during the week.


  15. I think we should be careful about putting 90 feet long streetcars in mixed traffic on Queen and King. Experience suggests that length of vehicle will cause a slowdown in service, as drivers drive extra (perhaps over) cautiously in traffic to account for the additional stopping distance required from the larger vehicle. And this doesn’t even take into account additional dwell time at stops as more people get on at each stop, assuming a slight reduction in service to account for the additional seats.

    I don’t mind a minimum headway on major routes of 20 minutes after 10 PM, as long as it was accompanied by a 10 minute minimum in the base and say 5 in the peaks, as long as the service was reliable.

    Steve: We ran 92-foot long trains of PCCs on Bloor Street until 1966, and trains of PCCs on Queen until the service was cut back and shorter headways of single cars improved the quality of service. Stopping distance is a question of the quality of the brakes, not the length of the train. Once operators learn that they can trust the new cars, they will drive accordingly. As for dwell times, the move to all-door loading will probably make stops shorter rather than longer because everyone won’t be getting on single-file past the operator.

    But, yes, all of these things have to come true or we could have a disaster on our hands for service quality.


  16. The whole idea behind streetcars is frequent service on routes too heavy for buses. If the TTC wants to run the longer cars at the same service levels as the shorter ones, that’s one thing, but this is something else. Is it that they don’t make short streetcars anymore? What’s the deal here? This is just wrong. I can’t see how anyone can say this is better than using buses or trolley buses on the same route.

    Steve: The length of the car is partly due to having a long low-floor central section between high-floor end sections. Running larger cars on an unchanged headway gives more capacity for surge loads that can bedevil off-peak services. The point here is not to run the same service, but to run more service.


  17. I have been reading the debate of single versus double ended cars and loops versus crossovers and I have a few points to make:

    1 At the ends of lines loops are faster than crossovers as the driver does not have to change ends and change trolley poles (hopefully they will switch to pantograph but this is the TTC) and crossovers in the middle of 4 lane streets do take up road space and reduce traffic flow. I have seen this in Melbourne. Chicago Transit Authority uses loops on many of its rapid transit lines as it allows for shorter headways than you can achieve with crossovers. If you wish to have scheduled short turns then you should have a centre track on a PROW or a reversing track like they used to have on the side of the Road at Keele St. north of Dundas to reverse the cars or you will hold up traffic and LRV’s. Trailing point crossovers are ok for emergency short turns.

    2 If you are going to order a bunch or cars for the existing network and not put in any form of reversing track then double ended and double sided cars just add to the cost of the car and remove seating capacity. I fear that the TTC will order double end cars and use them like they were single ended. The amount of land required for a double crossover with two tail tracks like on the subway is much less than is required for a loop but you cannot hold as many cars and keep them in order after a gap unless to put in some extra switches and increase the length of the tail tracks.

    3 While the advantages of double end cars are high if the TTC takes advantage of them they also increase cost and decrease capacity. Going double ended because every one else does is a waste of money and capacity unless you make use of them.


  18. Another question, could any of the cars on the market be equipped with 3rd rail?

    I believe at one time the New Haven ran electrics that used overhead in open country and 3rd rail in tunnels. Although that was a heavy rail service, 3rd rail reduces tunnel heights.

    Steve: That is theoretically possible, but there are major safety issues operating cars with third rail shoes in locations where people can touch them. Although in theory the shoes are electrically isolated when the pantograph is in use, this is not something you want to bet on.

    I can’t see the TTC removing any existing loops to reverse in the street. Humber Loop I could see replaced if double-ended cars are used. Although one loacation doesn’t seem to justify the expense of double ended cars.

    As to pantographs, isn’t it typical for pantographs to use overhead that is staggered to reduce wear on the collector? All of the existing wiring is set-up for trolley poles. The catch wires at intersections might interfere with pantographs.

    Steve: Mixed operation is common in systems that are migrating from poles to pans. Skates are used around the special work to allow the pantographs to ride below the level of the overhead used by the poles. However, this will be complicated at our major intersections.


  19. That 20 minute headway could turn into 30 when the TTC starts adding and subtracting “per-seat” ratios.

    As Steve says, we need to get past some of that thinking. The TTC went a long way to stabilizing its ridership when they standardized subway service so that, regardless of the time of day, you never had to wait more than five minutes for a train to show up (of course, service was more frequent during rush hours, but this made the subway convenient to take during the midday, evening and weekend periods).

    I’ve said this before, but the TTC should operate on a 5-10-20 principle. If you’re going to bother to run a route anywhere, then passengers along that route should never have to wait more than 5 minutes for a subway train, 10 minutes for a streetcar, or 20 minutes for a bus.


  20. Mimmo asked if short (i.e. PCC- or CLRV-length) streetcars are no longer being made.

    There is at least one contemporary example, the Vario LF made in the Czech Republic. It has a low-floor central section, but it is only 36% of the total floor area, encompassing the area between the bogeys – the sections above the bogeys and further to the front and rear are high-floor. So it would have the advantage of being directly comparable geometrically with our existing fleet, but it wouldn’t meet the 70% low-floor threshold.

    There are also replica streetcars that are being manufactured and used on systems in the States, but they wouldn’t offer low-floor operations (they’re installed with lifts instead to be ADA-compliant). I’m also not sure how many are being used in high-demand routes (as opposed to lower-demand systems like, say, Tampa or Little Rock) and how they would hold up in a more demanding environment.


  21. Brent said …

    “There are also replica streetcars that are being manufactured and used on systems in the States, but they wouldn’t offer low-floor operations (they’re installed with lifts instead to be ADA-compliant). I’m also not sure how many are being used in high-demand routes (as opposed to lower-demand systems like, say, Tampa or Little Rock) and how they would hold up in a more demanding environment.”

    Don’t forget New Orleans and Canal Street. They have brand new 1920’s street cars with the lift except they need all new electrics because of Katrina.



  22. From what I’ve read in various places, the Flexity Swift cars have unpowered center trucks. Hence they don’t meet the stated requirement that all axles be powered. I assume the TTC has concluded that the center trucks (bogies) must be powered in order for the vehicles to negogiate the existing track network’s extremely tight turns without derailing.

    In terms of pushing power is less an issue on grades than ‘adhesion’. Actual adhesion can vary greatly based on track condition, temperature and whether the track is straight or curved.


  23. Having read all of the posts concerning the proposed specifications for TTC’s new streetcars, I notice that all of those advocating a double-ended (bi-directional) configuration because of its “flexibility” advantages have neglected to mention its significant disadvantages compared with single-end cars (ALRVs, CLRVs, PCCs and even Toronto’s venerable “Peter Witt” cars of yore.

    These disadvantages of double-ended cars are:

    1) A major loss of seating capacity due to have doors on both sides and cabs at both ends. Yes, more standing room is gained in the off-side doorways but this often is a theoretical plus, especially when street railway lines are configured to have some side platforms and some centre (island) platforms requiring that doorway areas on both sides of the cars be kept clear for alighting and boarding passengers. The result is that double-ended cars usually have not more “real” carrying capacity that that of single-ended cars of the same lenght and width. Compared to single-ended cars operating on the same headways, then tend to discourage public transport riding because more people have to stand during peak periods. Let us remember that no automobiles are built with provisions for standees.

    2) More components, i.e. doors, door engines, duplicate cabs and all of their inherent equipment, add to both the capital and maintenace cost of each car.

    Steve: These are valid observations up to a point. The real question is whether the operational flexibility of being able to use crossovers rather than loops warrants the extra cost, etc., of double-ended equipment. Issues of platform placement are system and route specific.

    As for systems being designed for standees, may I offer the Yonge subway as Exhibit Number 1. People are used to this sort of arrangement, although obviously we don’t want to force more to stand than is necessary.


  24. Having looked at pictures of most of the LRVs being made today, it would make the most sense to embrace proof of payment once these new vehicles enter operation. Articulated MU LRV trains operating along all routes (esp Queen, King and Spadina -think old PCC and CLRV MUs) with alerting chimes, POP as aforementioned and with the same frequency of existing service if not moreso, would definately send a strong message to Torontonians that the TTC is a competitive alternative. Granted this is just one person’s dream and yes the safety island platforms would have to be lengthened, but still I feel that I must share it with someone like yourself Mr. Munro.


  25. The difficulty with converting to pantographs is that the entire catinary has to be relaid. They experimented with pantographs on the existing system a few years ago and found that the wire wears a groove in the pantograph unless it is staggered. However, I hope they’ll make this change. Toronto pays a high price for having a non-standard system, and I hope that new lines will bear this in mind.

    Steve: All I can say is that if anyone had to actually perform an experiment to learn that wire strung for trolley poles wears a groove in a pantograph, well, they don’t get out as much as they should. This is one of those “Transit 101” design issues. Trolley poles will tolerate a small almount of slew, but not as much as would be typical for pan operations.


  26. The point is that pantographs and trolleys can’t co-exist. The whole catenary has to be re-layed before pantographs can be used.

    Steve: It is physically possible (Boston did it for years), but tricky. Also, Boston didn’t have complex intersections like Toronto’s. A bigger problem here is that we continue to erect new overhead with hangars that will only work for trolley poles because the span wires are at the same height as the contact wire. This will all have to be redone for pantographs as will locations with drooping, untensioned and, in some cases, structurally unnecessary span wires.


  27. ‘ “sinusoidal” running on tangent track for maximum stability and minimum flange bearing with 1:20 wheel tread conicity; ‘

    What does that mean? That streetcars should be able to follow laterally oscillating tracks?

    If my guess is correct, I guess that this requirement is mainly because of track paths in streetcar yards?

    Steve: Wheels on rail vehicles vehicles are not flat, in the sense that the surface is not level with the horizontal. The surface is slightly tilted (1:20) upward away from the flange. The purpose of this is to mate with the rail head which itself is not level, but rounded. The combined effect of the two tracks and the wheels on each side of the car is that the track pushes inward slightly to keep the car going in a straight line. If the wheels and track were level, this work would all have to be done by the flange, and the ride would be much rougher.

    The sinusoidal bit refers to the fact that as a car travels along the rails, the wheels will always hunt from side to side due to minor changes in track alignment and to the load of the car itself (even a gust of wind can cause lateral movement). As long as this motion is damped out in the suspension, riders should not notice the effect. It is far simpler to design a system that reacts to and damps out the routine unevenness in the roadbed than to try building one that is absolutely perfect and rock solid. Railways have been doing this pretty much since they were invented because track is never, ever perfectly straight.

    The TTC had a big problem with some of its early H-series cars whose trucks had a natural frequency of swinging (think of it almost as a pendulum) at certain speeds. This tended to wear the track in a pattern matching the hunting frequency, and it becomes a self-sustaining problem. The north Yonge subway had extremely rough rides for H-1 cars at high speed.


Comments are closed.