Trams vs Skytrain: A view from Vancouver

Today’s Globe & Mail includes an op ed article Rethinking the Need for Speed reporting on a recent study comparing the cost of transportation modes.  The study and the article conclude that trams (streetcars) are the best choice, and that Skytrain (also known as the “RT” in Toronto) is a distant choice.

Those who know me well know that any chance to give the RT/ICTS/Skytrain advocates a black eye is more than welcome, but in this case I have to put a bit of context on the discussion.

The Skytrain vs LRT debate has consumed Vancouver transit advocates, planners and policitians for decades.  The original Skytrain was a combined product of a premier who didn’t like streetcars and of lobbying by the Ontario government to get its then-new ICTS showcased for Expo in Vancouver.  Certain characteristics of the original Skytrain route including the availability of a tunnel under downtown that could handle stacked Skytrains, but not LRT, an available right-of-way that kept down elevated construction costs, and the operational advantage of close headways of short trains tipped the balance in Skytrain’s favour.

Having said that, I must also observe that the technology was used to its maximum during Expo with a far more sophisticated operating model than anything the TTC has ever implemented on any line.  This was automated transit really shining, but only for a brief moment.  Probably the most important thing about the Vancouver system is that the people running it really wanted to make it work.  From the day it opened, they analysed operations (including automatically produced charts such as those you see in my TTC route studies) looking for ways to handle demands and unusual events better.  The idea of throwing up your hands in resignation, the TTC’s approach to line management, was totally foreign.

Skytrain works not just because of the technology, but because the people running the system care to make it run well.

All the same, the love affair with Skytrain wedded Vancouver to high-cost system expansion, and a route design skewed to handling commuters more than local trips.  Indeed, most of the original Skytrain line does not follow city streets, and it depends on local redevelopment, walk-in trade and bus feeders for passengers.

The LRT vs Skytrain debate heated up recently with a proposed east-west line along Broadway, a major bus and trolleybus corridor.  This is a street with much local development and Skytrain foes look to LRT as a way of achieving better local access and support for the community throiugh which the line will pass.  Elevated construction is out of the question, and a Broadway Skytrain will almost certainly be underground adding considerably to its cost.

This is the political background to the Skytrain vs Trams study, and it’s important to read the study in context.  The study itself does not address specific corridors, but simply looks at the operating and capital costs of each mode, as well as the environmental effects.  When the numbers are combined, trams come out on top (or more accurately on the bottom with the lowest cost and carbon impact).  Skytrain is much higher, primarily due to capital cost.

The basic debate in all of this is one of philosophy:  should new transit lines be built to serve long trips where speed between stations is paramount, or should lines serve shorter trips and local demands with easily accessible stations?  In the ongoing debate here, Transit City comes under fire because the lines won’t be fast enough for long trips.  Should that be their purpose?  What role does GO have as a regional carrier within the 416? 

Some Transit City proposals call out for redesign, especially regarding the Sheppard/Finch transfer and the dubious nature of surface proposals for the south ends of the Don Mills and Jane routes.  Work on new proposals is already underway as a spinoff of the Metrolinx studies, but the old plans still get lots of play including the TTC’s own Transit City campaign all over the system.  The TTC needs to update the proposals to remove the less credible options and to indicate that they are not just drawing lines on maps.

Finally, I hope to see the Metrolinx study of options for the Scarborough RT published soon.  This is an ideal chance to convert the line to LRT, and even the TTC’s own recommendation to upgrade with Mark II cars only, barely, made sense if the line would never be extended.

We now know that the “SRT” will run north into Malvern and possibly north of Steeles Avenue.  The cost comparison between LRT and Skytrain should spell the end of the RT as we know it.

26 thoughts on “Trams vs Skytrain: A view from Vancouver

  1. An interesting source article from my alma mater UBC* and the G&M follow-up story! *Motto: Tuum Est~It’s up to you!

    I was particularly impressed by the favourable ratings of electric trolley buses (the “Red Rockets” of my youth in Vancouver/Burnaby/UBC) on many dimensions, virtually tied with Modern Trams/LRT (but not on capacity). N.B. TTC Staff are currently working on a report responding to TTC Chair Adam GIambrone’s request to reevaluate trolley buses for future deployment.

    I hope all those who delight in slagging Steve on this this, his blog, for being “biased” in favour of LRT over subways will pause, read the articles, take a deep breath and acknowledge all along Steve’s been a proponent of building/running the most appropriate mode for the current and future demand on a particular route, rather than deciding transit purely where politicians want to build/run transit to reward their local ward/riding voters or party.

    As we head into uncertain economic times this is the only sane economic path, regardless of future infrastructure spending announcements. It’s high time politicians are held accountable by the public for the efficacy of their long term transit capital and short term operating spending, just as in the private sector, given the magnitude of the proposed (Metrolinx RTP) expenditures.

    Do we want one or limited express subway/skytrain rapid transit line(s) or a network of lesser local modes (eg. Local/Express combo, 20-minute headways, Transit City) that cater to where people actually live and work?

    It really IS up to US to make sure it happens!

    Bob (UBC BASc ‘74/MBA ‘76)


  2. As maybe the only person around here who regularly drives, I think the Transit City LRT advocates really need a major reality check.

    I’ve driven all over and I don’t really see any *severe* congestion on local roads in the AM and PM rushes that can’t be avoided. Instead, I see massive congestion on the 400 series highways, the DVP, the Gardiner, and on a few streets downtown. To me, this says that the demand and traffic is on long haul, not short … so where do the informed advocates say we should spend our money, on local light-rail services of course!

    Steve, your attitude would change if you didn’t have such a peachy commute … Bloor subway east and then RT (and always against the flow). Let’s take away the Bloor subway and the RT and replace them with local LRT services. Your commute to STC would double, maybe triple. How would you feel then?

    Try coming in from the west, or the north-west. Better yet, try the DVP or the Gardiner in the morning and then come back here and tell me with a straight face we need better local LRT services.

    Steve: You are misrepresenting my position. The purpose of Transit City is NOT to relieve congestion on the 401 or DVP or any other exressway, it is to improve local transit on many routes that cannot handle the demand with buses (at least without completely taking over the streets).

    I have always argued that better GO services are needed to bring people into downtown, and have often observed that almost nothing in the Metrolinx RTP addresses congestion on the road network in the 416 or 905. Indeed, the only benefits in the RTP is to trim future congestion by providing an alternative way for some types of travel to grow, much as GO Transit did for the core area.

    An LRT from Kennedy to STC will take exactly the same length of time as the SRT does today because the stops spacing and vehicle performance characteristics will be identical. I would never put an LRT where the Danforth subway is now because the demand on that line far exceeds what LRT can carry.

    It’s easy to make my “position” look foolish when you contrive an alternative universe to suit your argument.


  3. The competition between fast trains and local service somehow doesn’t seem to be as much of an issue in Vancouver as it is in Toronto. The Millennium Line actually has more stations than the express buses it replaced, and the plans I’ve seen for the Broadway extension likewise show enough stations that any point along the street would be within reasonable walking distance of a station. It’s certainly a far cry from the North Yonge line.


  4. It helps that Vancouver is so much smaller than Toronto. A light rail on Broadway makes about as much sense as light rail down the middle of Eglinton Ave. I don’t understand why everyone makes the mode out to be black and white. Sheppard Ave E is suburban sprawl; which, by definition, means there’s nothing there, right? So doesn’t it make more sense to plan for longer distance travel, so you can get somewhere that is meaningful? Eglinton Ave, on the other hand, is dense central city Toronto with a lot of local travel because there are a lot of things to go to. If Vancouver can have subway stations closer than 1 km apart, then Toronto should be able to do so as well, even if it means the Eglinton subway has to be built cut-and-cover with the traffic nightmare that will result.


  5. Well to Chris’ comment I know there are some open sections on Eglinton for stations to be built by cut and cover method. There is a huge section of open land on Eglinton and Kipling, west side of Islington by Eglinton, a small space on Royal York….but that’s all I know of. In these parts, traffic diversions shouldn’t happen if a “subway” of what ever kind was built. The TTC has to take advantage of this moment. I know this land will be used eventually for some sort of condo developpment which is was the mayor wants this city to become. I think the TTC should purcahse this land if they can for future developpment. If condos go up there, I’m afraid a subways can never be developed in this region.


  6. “Turkey” is the word that former TTC general manager Al Savage used to describe the Scarborough Rapid Transit (SRT) line. Now the question has come as to what do you do with a turkey? The Scarborough RT line is in need of a major overhaul. In fact the wrong system was installed from day one. Extending the Bloor Danforth subway would have been the correct choice. Let’s wait and see if they blow this one again.


  7. Steve, is it just me, or have the authors of the report allowed a simple matter of arithmetic to throw off their calculations?

    Figure 11 does a pretty good job of identifying transit vehicle design loads, but without access to the report’s appendices, I can’t rightly tell whether they’ve accounted for how the vehicles operate in the field. As a case in point, I understand from my own research that SkyTrain Mk IIs operate in married pairs with provisions for single, double, or triple paired configurations, which means that at a single-vehicle design load of 90(ish) passengers, a Mk II consist would carry either 180, 360, or 540 passengers at any given time. Conversely, the 160(ish)-passenger design load the report identifies for a typical C-Train LRV is notable only to the extent that the vehicles are intendeded to run in consists ranging from three to five cars, meaning that one C-Train would carry 480, 640, or 800 passengers at design load. For that matter, last I checked Transit City was supposed to provide for two-tram consists, thus doubling the single-vehicle design load of 150(ish) to 300. (The viability of operating muliple-vehicle consists of trolleybuses, diesel buses, Toyota Priuses, and Ford Explorers is an exercise I will leave for the reader. 😉 )

    What I am not seeing from the report is the degree to which the authors of the study have accounted for multiple-vehicle operation in their determination of operating and energy costs. Again, it may simply be that Auld Nick is buried in the details in the report’s unpublished appendices, but surely it would have occurred to someone at UBC’s Design Centre for Sustainability that if indeed “one tram driver is more than twice as
    productive per hour than is a diesel bus driver,” one light rail operator is anywhere from two to five times more productive per hour than is an uncoupled tram driver. I’m not about to dispute the good work trams do in central city settings, but if UBC’s Design Centre for Sustainability fully intends to argue that the tram, rather than the SkyTrain, is the right tool for the job of transporting Vancouverites, they must substantiate in their response to this report the current and the maximum feasible operating conditions for each of these transportation modes.

    Steve: Yes, I agree that the report is thin on operational details and has all of the problems we often find with this type of document whether it come from tram, busway or car advocates. Basic points such as those you mention undermine a report’s credibility to a less-than-receptive audience.


  8. It’s another indicator that reliability of service has a greater impact than the speed of service, and also that adding stops to a line can make it more attractive, as even “the fast-lane operator” GO Transit has been learning this past decade or so.


  9. I agree with much this article and Steve has said over the years.

    Though it should be added one problem with the SRT not addressed through vehicle type is its absolutely terrible location/routing.

    Stations under bridges, following a rail corridor, rather than a major road, all while missing the biggest trip generator between its 2 key points (Scarborough General) is a very significant problem, beyond the technology choice.

    Having said that, LRT’ing the line is better that what it is now; pending a more sensibly aligned B-D extension, in the year 2045!

    Steve: You may be interested to know that the original alignment was to follow the same old railway right of way that the subway uses approaching Kennedy from the southwest and go diagonally up through Scarborough. This was changed to follow the rail/hydro corridor in response to objections from people living along the right-of-way about noisy streetcars. Little did the locals know how noisy the RT would be in its time.

    The bridges at Lawrence and Ellesmere were also add-ons built to grade-separate the railway and make room for the RT. The original LRT line was to cross at grade with traffic lights.

    The line parallel to Progress was originally going to be at grade, but the Scarborough Councillors were bamboozled by the TTC into thinking that this would cut their precious town centre in half. You may have noticed that the bus station and RT elevated did a pretty neat job of that anyhow.


  10. Technically speaking, the trams used to replaced the RT will not have the same performance characteristics as the ICTS MkII. Most trams top out at about 80 km/h versus the 110km/h for the ICTS. In addition, trams cannot accelerate as fast due to lack of traction. There is a wider surface area on the metallic strip for the linear induction motors to propel the vehicle.

    Steve: For the station spacing typical of Transit City lines, the higher speed of the Mark II cars is not needed. As a matter of comparison, the subway tops out at about 80km/h. (For those old farts like me who remember miles, 110km/h is 70 m/h, and no transit line except GO runs at that speed.)

    I do not understand why there is a double standard. Trams running down the street also produces eyesore. Do people like to power poles every where? ICTS guideways are always demonized. However, if one build a rapid transit, intensification will occur. This means that buildings can be built on top of elevated station which makes look a little better.

    Steve: Unless the TTC is designing them, the poles holding up the overhead would be the same ones as holding up the street lighting and, possibly, the hydro distribution system. Also, the elevated structure of an RT line is a LOT bigger than a pole.

    The Globe article also does not point out that speed is very important in attracting riders. ICTS or heavy rail is expensive, however, there is a speed advantage. (i.e. 60 seconds headway which trams cannot do). People pay more to ship packages with FedEX. Why? Because it is faster. The TTC should not be just a service for the poor, it should also be a premium service.

    If I have it my way, we should build a ring in Toronto. ICTS trains would run from Scarborough Center to Finch to Airport to Downtown and back to Scarborough Center. A city’s success is measured in how fast it can move people. The faster it can move people, the bigger pool of talent are available to employers.

    Steve: Nice idea, but there are many, many more places people want and need to get to within Toronto and beyond than would be served by your proposal. The ring looks good, but only if both your origin and destination lie on or very near it. Otherwise, the dominant factor in your trip will be access and waiting times for feeder/distributor services.


  11. Steve,

    Thanks for the interesting info. I knew about the technology imposition, but not the change in routing, or the dispute about the elevated nonsense.. (who could ever have said with a a straight face that this was less divisive than a ground line?)

    Question about the route that didn’t happen.

    I see the corridor clearly on Google, where the single track still exists, and its continuation to where it crossed Eglinton.

    But after that it seems to vanish. Though, I can clearly identify a corridor on the same trajectory, just north of the 401&Progress.

    Is the entire previous corridor gone?

    Is what I see north of the 401 part of the same original corridor, or is it another one?

    Thanks for your insight.

    Steve: Yes, I believe that the piece north of 401 is the same line, but much of it has been lost to development.


  12. Benny Cheung SAYS

    “Technically speaking, the trams used to replaced the RT will not have the same performance characteristics as the ICTS MkII. Most trams top out at about 80 km/h versus the 110km/h for the ICTS. In addition, trams cannot accelerate as fast due to lack of traction. There is a wider surface area on the metallic strip for the linear induction motors to propel the vehicle.”

    Technically speaking this is not true. The accepted value for adhesion of steel wheel on steel rail is 0,25g, or about 2.5 m/s/s. Anything over 1.0 m/s/s is considered to be uncomfortable, especially by standing passengers. The efficiency of a standard chopper powered rotary traction motor is at least 85% while a linear induction motor, LIM, is lucky to get to 30%. The maximum speed is set as a function of station spacing and ride comfort. I grant that building an elevated concrete guide way will probably provide you with a smoother ride than one that is on the ground but at what cost? One of the drawbacks of ICTS in the article was its energy costs. If you are going to use LIMs then you energy costs per passenger are almost tripled because of reduced efficiency of the motor design. If this mode were so good then how come there aren’t more systems in the world? It is because of construction cost for the guide way and the poor operating efficiencies of LIMs.

    Steve: Also worth noting: The new Canada line to the airport uses rotary motors.


  13. “If you are going to use LIMs then you energy costs per passenger are almost tripled”

    What’s the attraction to Bombardier to continue using them then?

    Steve: Proprietary technology. Once you’re hooked, you stuck with extending lines using their vehicles. Vancouver build the Canada line completely separate from the Skytrain network, but the Evergreen line will be a Skytrain extension.


  14. Mr. Wightman, you are absolutely correct that Linear Induction Motors are not as energy efficient. However, motor energy consumption is not the only measure. One would have to look at the entire picture. ICTS vehicles have less drag since it is narrower and shorter. Both the Honda CRV and the Accord use the same engine. The Accord gets better mileage because there is less drag.

    Second, trams tend to have much more glass areas which require more air conditioning. More air conditioning can easily wipe out any gains at the motor. Third, most tram applications have more stop and go situations than ICTS. Think of the Spadina tram line and how many times the tram starts and stops for traffic light.

    While fast acceleration is not always desirable, more traction is always desirable. Since a steel wheel has limited traction, trams cannot navigate steep inclines. ICTS can handle a steeper approach angle than a tram. This makes it useful when building transit in built up areas. Instead of bulldozing, it can go around or over obstacles.

    Building guideways is still much cheaper than building a tunnel. Yes building trams (like Spadina) is cheaper, however, the tradeoff is the lack of speed. Let’s throw a question out to everyone. If ICTS can take you from Scarborough Center to Union in 20 minutes for $10 per ride (hypothetically speaking) and a tram that covers the same route in 30 minutes while costing say $5 per ride, which one will you choose?

    Finally, Linear Induction technology has found a home in many places. On the top of my head, I recall that the Beijing Airport Line uses ICTS technology. There is also a presence in Seoul. The Tokyo Toei Oedo Line and the Yokohama Green Line also uses Linear Induction.

    Steve: You have set up a false comparison. If the desire is to go from STC to Union, and there really is enough demand to have a line dedicated to that sort of trip, then this suggests a largely grade-separated right-of-way with few stops (akin to an airport to downtown line). If we were building it with trams, we would look for a right-of-way or create one. The Broadway line in Vancouver, the subject of current trams versus Skytrain debates, is a local route with fine-grained demand. It is not an express service between its end points.

    The difference with a tram/LRT is that we would not have to build every inch of the network with complete grade separation where it was not needed.

    Your comments about drag and the amount of glass speak to the size of the vehicle. We could build teensy-weensy trams too if we wanted to. The reason ICTS vehicles are so much smaller than trams or subway cars is that the LIM couldn’t handle the power requirements to move a larger, heavier vehicle. The majority of heat/cooling load on a transit car comes from losses when the doors are opened. If you build a line with infrequent stops, then you have less opportunity for loss.

    Your whole argument is apples and oranges.


  15. Benny Cheung Says:

    “Second, trams tend to have much more glass areas which require more air conditioning. More air conditioning can easily wipe out any gains at the motor. Third, most tram applications have more stop and go situations than ICTS. Think of the Spadina tram line and how many times the tram starts and stops for traffic light.

    “While fast acceleration is not always desirable, more traction is always desirable. Since a steel wheel has limited traction, trams cannot navigate steep inclines. ICTS can handle a steeper approach angle than a tram. This makes it useful when building transit in built up areas. Instead of bulldozing, it can go around or over obstacles.”

    Your first point is a result of a choice in operating characteristics, not in technology. One can build narrow windowless trams that operate on a “light flimsy” guide way at 110 km/h at a 2 km spacing. The question is “is that what is best?” If you ran an ICTS line with the same stop spacing as Spadina you would still use more power per passenger. Steel wheel on steel rail can handle any grade that I have seen on ICTS. The problem in Toronto is vehicle design and mix of vehicles, ALRV and CLRV, not the steel wheel steel rail interface. I can name more LRT systems in North America than you can name LIM systems in the world. The systems that you name were built for a reason other than the superior technology of the LIM.


  16. Hi Steve:-

    In response to Benny Cheung and his comment about ICTS performance vs. real transit vehicles, ie:- regular motored trolley cars. Benny, please read the book about the history of the development of the PCC streetcar published by Interurban Press about 1980. There is a section there about the testing and experimenting with acceleration rates and deceleration rates. The test bed PCC cars initially were set up at higher rates than what was ultimately adopted. In theory, ICTS can be set up to run at higher rates than they presently use, but this is an argument that holds no water, for the Electric Railway President’s Conference Committee technical crew proved, in 1933, that a PCC car could be set up as to be totally uncomfortable for a passenger too, (and all that without a computer, merely olde fashioned electrics, no electronics nor LIM even) just as one could do with ICTS. Is it wise to do so? Don’t think so!

    As to grade climbing, well the Pittsburgh Railway used PCC cars and their low floor conventional 1900’s streetcar technology predecessors on 14% grades. That’s really steep. Think too about the grey, sloppy and snowy weather that Pittsburgh gets. I’m not sure what the TTC’s steepest was for streetcars, but Witt motor cars used to pull fully loaded dead weight trailers (themselves fully loaded) up Yonge Street south of St. Clair and also on Avenue Road north of the CPR Tracks.

    As to the advantages of automatic train control Steve, please do not confuse this with the necessity of using linear induction solely to achieve automatic control. Not true, not true. Linear induction is the inappropriate portion of the equation in all that is ICTS. The Washington subway does not use linear induction and this is one of the smoothest, quietest subway systems ever built, and it has automagic train control to boot. I’m sure that given the same desire in D.C. as in Vancouver to supply a service of short turns, dense train traffic and alternate routings, it could be achieved there too with their computer system and without the LIM.

    I know that in Vancouver one of the technological triumphs was in the double decking of the former railway tunnel without increasing its dimensions. If this was the only reason to adopt UTDC’s linear induction, I would suggest that if in reality that tunnel could not have been widened and/or heightened (I’d be surprised if it couldn’t have been one or the other) to accomodate conventional LRT vehicles, (thinking at the time of construction, the widely accepted North American version of the Duwag car) then it would really surprise me beyond all sci-fi disbelief, that a small bodied trolley car with outside third rail pickups could not have been constructed to fit the same loading gauge as a Mark II car. I would hazard to bet that a car of these dimensions with conventional electrcal equipment could perform as well as the ‘bullet cars’ on the Norristown high speed line from West Philly, north. No LIM there! And boy-oh-boy, what a smooth, quick ride, and that on old fashioned bolted and spiked track.

    ICTS’s linear induction motor technology is the most innappropriate use of rail transit resources one could imagine. The worst LRT installation is a quantum leap ahead of the best linear induction merely because of its flexibility of being able to be designed into any right-of-way need. Total separation is not a necessity, desirable yes and use it where you can, but where you can’t the options are almost limitless when LRT is considered. Scarborough’s ICTS should be relegated to a museum of misfits ASAP. LRT technology that can be integrated into the whole of the Transit City System should replace the ‘turkey’ as quickly as humanly possible, otherwise Scarborough will forever remain the laughing stock of the transit industry.

    Oh and what does one do with a turkey? Slaughter it, for in this instance it’s already done like dinner.



  17. James asked about the abandoned rail corridor that was to be the right of way for the LRT that became the SRT…

    This was the ROW of the Canadian Northern Railway. This section (from Todmorden in the valley to Greenburn) was abandoned in 1926. From Kennedy Station at Eglinton, the alignment can be clearly seen as street patterns tended to parallel its diagonal alignment. For instance, the ROW runs between Fitzgibbon Ave and Khartoum Ave just northeast of Kennedy Station. Back yards on each of these streets didn’t touch each other as the ROW was between them (this may still be the case, satellite views show trees between the yards). Where streets cross the ROW (such as Lord Roberts Drive and Wainfleet Road for this example), there are now homes built on the ROW, though some are fairly recent.

    Though you may have to look closely for it, about the only point where it totally disappears is where it crosses the hydro corridor at Brimley and Lawrence. It continues through Thompson Park where I believe there is still a culvert once used by the line (there are some historical pictures on displays in the park showing the line). The driveway into the McCowan Yard for the SRT is on the alignment, and so is the eastern tail track of the yard. From here, the ROW is fairly clear, except for the portion crossing the 401. The line paralleled the CPR east of Finch/Morningside, where there are still bridge abutments next to the CPR bridges over Old Finch, Sewells, and Meadowvale Roads.

    Now a question that I have: The CNoR is on the south side of the CPR in Scarborough, but out in Pickering, the bridge piers over Duffins Creek are on the north side of the CPR’s bridge. Where did the two cross? I seem to believe it was just west of Fairport Road. Does anybody know if it was a grade crossing or not?


  18. Steve, thank you very much for publishing the study and this discussion. Here in Kuala Lumpur, Malaysia tenders for two “LRT line” extensions and new LRT line (really all metros of small capacity) are going to be public in Q1 if 2009.

    Never mind that the public consultation has not taken place (and likely never will)

    It will be interesting to see if Bombardier will lobby for the new line to use the LIM technology…

    The existing Kelana Jaya line extension will be LIM technology but I have been pushing for a “LRT” (north american term) instead, in order to build a longer line serving more communities.

    I hope to get some updates on what is happening over here. I did have a conversation with the director of rail operations for RapidKL who said that the LIM system did not fulfil the promises.

    Regards, Moaz Yusuf Ahmad

    Kuala Lumpur, Malaysia


  19. I really think the ‘long commutes between stations’ argument should be dropped since Skytrain has pretty good spacing between stations.
    In fact even a full-fledged subway can have short distance tween stations.
    I habitually walk between stations on the Bloor-Danforth line.
    And I get irritated at buses and LRTs which seem to stop every 100 metres.. nobody needs that! It’s stupid.


  20. When you think about it, there is no advantage of trams over electric buses. In fact, since it cost $8.5million just to upgrade the short existing stretch of track being used for the Olympic “flexity” demonstration, the cost of laying track on Broadway would be very high, indeed. Buses need drivers; so do trams. That’s one of the most expensive aspects of their operation. Buses can be rerouted fairly easily; trams cannot. I really think an off-grade computer-run system like SkyTrain is the answer.

    Steve: The question is one of relative capacity. There is a point beyond which buses, especially those running on a busy arterial like Broadway, cannot carry any more people. At that point larger capacity is needed and this can only be provided with rail vehicles in the street, or a grade-separated system. From a design point of view, one must also look at whether grade separation is needed over the entire line, or only part of it. With Skytrain, you don’t have the option.


  21. Having travelled on the Millennium, Olympic and Canada Lines during the Olympics, I would at least advocate for the current Millennium line which terminates at VCC-Clark to extend to intercept Canada Line (2.5-3km west) at either Olympic Village or Broadway-City Hall as ICTS. VCC-Clark is probably an important stop for the college’s students but doesn’t seem helpfully located for other people, especially those coming north from Richmond/YVR and going to the east.


  22. “With Skytrain you don’t have the option”?? Is the buried portion of the Canada line just an optical illusion? What’s the agenda here?

    Steve: First off, let’s get the entire quote for context.

    The question is one of relative capacity. There is a point beyond which buses, especially those running on a busy arterial like Broadway, cannot carry any more people. At that point larger capacity is needed and this can only be provided with rail vehicles in the street, or a grade-separated system. From a design point of view, one must also look at whether grade separation is needed over the entire line, or only part of it. With Skytrain, you don’t have the option.

    The last time I looked, “grade separated” includes tunnels and, therefore, the Canada Line. What I argued was that depending on demand, a grade-separated option may not be necessary (either underground or elevated), but the Skytrain and Canada Line technologies cannot operate down the middle of a street, or even use a surface right-of-way with grade crossings.

    There’s no hidden agenda at all.


  23. Why do Scarborough RT cars seem to pause in their acceleration every time they depart a station? Vancouver SkyTrain cars don’t seem to do this.

    Steve: It’s a TTC thing. The maximum speed is held low until the train clears the platform as a safety precaution. Stragely, they don’t do this in the subway, although who knows what bizarre schemes they will come up with when they finally get to convert the Yonge line to ATO.


  24. Here’s another question I’ve always wanted to ask, but nobody’s ever been able to come up with an answer (it may not be possible to answer this question anyways). Seeing how your explanations in transit has even encompassed scientific details of operation, here goes:

    What causes that 3-part acceleration noise on Mark I ICTS vehicles, and why isn’t that same rich-sweet sound replicated in other LIM models?

    And what causes that long dragged-out tone when the Mark I ICTS comes to a full stop? Sometimes, after that long dragged tone, the ICTS cars give a quick higher-pitched “sigh” before the doors open (18 seconds into this video).

    I’m aware of how the motor itself works; I believe that a 3-phase AC current produces a U-shaped magnetic field that travels in the direction of vehicle travel, inducing secondary electric currents in the induction plate, which in turn produces a secondary magnetic field to oppose the original U-shaped magnetic field, pushing the train forward (also known as Lenz’s law, which states that any induced current would produce a magnetic field to oppose the change that the inducing magnetic field from the motor was trying to invoke; in this case, the change is the traversing motion of the U-magnetic field relative to the stationary reaction plates). Is this explanation correct?

    Steve: I will leave it to Robert Wightman to go into the fine details, but the frequency of the various tones will be determined by whatever acceleration and deceleration program is built into the LIM controller. The Mark I’s scheme could be different from that used on other equipment.


  25. LRT and ICTS style systems have very different roles in the transportation system of a city and I agree that the way SRT uses ICTS is ineffective. Driverless operation would be very useful however the TTC dropped the ball on that one. Steve has some great comments about how LRT has local benefits and integrates into a livable community. Where LRT shines is in moving people short distances conveniently at street level with a smooth comfortable ride (over a bus). If you want to move lots of people along a major corridor you need a subway, if you need to need to move people long distances commuter rail is what is needed, if you have an intermediate load or need some of the benefits listed below then ICTS has its place. I simply wish to clear up the misconceptions that are rampant in this and other posts on this site. I fully agree that the way the SRT is used today has no benefits over an LRT vehicle running the same line. Alternatively I simply wish to explain how ICTS is supposed to work and where it should be used. If used properly there are actually many parts of Toronto that could use it if the residents of the city were not so soured by the SRT experience. Why the SRT is how it is remains a further and much more confusing discussion.

    To clarify; the top design speed is 100 km/h however no speed over 90 km/h is ever used. 80 km/h is the speed sought by the train control in Vancouver when cruising and 90 km/h is typically used to catch up if time is lost in the schedule due to held doors or other delays. These speeds are typically the same for an LRV with the exception of the top speed being determined by the driver.

    For energy consumption: The LIMs are not as electrically efficient as a typical rotary motor however the design benefits that the LIM allows for makes the system as a whole more energy efficient. The comment above to the contrary is inherently bogus. The design of the ICTS concept cannot easily be broken into parts and examined separately but rather as a whole system. Remember also that the SRT is dramatically different then the ICTS concept called for which is due to decisions from TTC not its designers. Vancouver is a far better representation of how the concept is supposed to work.

    The Key features of the concept are the steerable trucks, the linear motor, the exceptionally low weight, and fully driverless operation. These features are closely related and are intended to deliver a system that allows for very flexible operation, very flexible alignment choices and delivers that at the lowest cost possible.

    The LIM eliminates gearboxes which reduces the rotational inertia of the bogies drastically. This in turn reduces the brake requirements drastically and the wearing thickness of the wheels and their required size drastically. To modify the ICTS bogies to rotary technology would almost double their weight. The steerable truck (assuming the rails are ever ground and maintained… I do not think SRT has been ground in a very long time rendering the steerable trucks a liability since they now adversely steer almost everywhere in the system) reduces flange pressure which reduces rail wear as well as wheel wear… again this reduces the needed wear allowance on the wheel and therefore the weight of the bogie. The steerable truck also reduces railhead creep which slows the formation of corrugation which along with the reduced flange pressure should reduce noise (for a demo of this you have to ride a Vancouver train… sorry to say the SRT seems to be totally corrugated and deprofiled like most of the TTC trackwork). The weight of the carbody is made very light as well partly due to the reduced need for sound deadening since if maintained the system is very quiet. This also helps keep the weight of the bogie needed to support it light and the size of the LIM down. As for power of the LIM it is sized for the vehicle rather then the other way around. The MK I LIMs are smaller then the MK II because the MK I trains are smaller then the MK II trains. These features also lower the maintenance costs assuming the right maintenance is done when needed. For example not grinding the rails causes wheel and rail wear, makes noise, and vibrates all of the equipment.

    The LIM is also part of reducing the dependence on adhesion for stopping however this does not set the minimum safe headway (the track and disk brakes do this). It simply allows the tight headway to be kept without risking the use of the emergency brakes for normal deceleration which effectively sets the operating or useful headway.

    There are a number of reasons these features are useful, none of which significantly benefit the SRT. First the short service headway potential (65 seconds, much shorter then used on the SRT or Vancouver) allows for a high line capacity without large trains which minimizes the station length and cost. It also allows low capacities to be served at off peak times without either very long headways or empty trains. The driverless system allows off peak times to be served at reasonable headways without incurring exorbitant cost for drivers. This is really a service quality issue since a system with drivers simply stops providing service when the demand is too low. Additionally when an anticipated demand peak occurs like the end of a game at a stadium the trains can stack up on pocket tracks or on the lines in preparation for clearing the platforms quickly when the game ends without having drivers sitting idle or requiring special staff scheduling. The light weight is a major part of the concept which is facilitated by the LIM and the steerable truck as well as the carbody structure. The amount of dead weight worth of train per passenger when loaded is far less then any alternative (eg Canada line trains in Vancouver have twice the dead weight per passenger and LRVs while light do not carry the same number of people) This is partly about energy consumption which is less then a typical LRV contrary to the comment above about inefficient LIMs. The other more significant reason is the cost of an elevated structure which is how the system was supposed to be used. The thinking is that fast service needs grade separation so either elevated or tunneled… and elevated is far cheaper then tunneled. The structures used for the expo line in Vancouver are very light and slender and much of the skytrain is elevated. SRT makes very little use of this benefit. Also not used on SRT is the ability to make and break trains without a driver to meet demand quickly and provide quality service.

    So in conclusion remember that this is not a discussion about what is inherently superior but what is right for the application at hand. I hope I have been able to clarify the technology a bit.


  26. Rationally, the competition should be split into two categories. Grade separated and at-grade. In this case, Skytrain and buses or streetcars.

    Skytrain in Vancouver is one of the portions that can pay for itself. All other nitpicking about costs or energy efficiencies should be secondary to that. The same goes for usability, as evidenced by ridership and the development patterns upon its implementation without any enticements by the municipality.

    I’ve seen many different people engage in petty debates about metros vs streetcars, and in my estimation the reasons for arguing come down to petitioning for one’s own self-interest selfishly and uncooperatively, or it’s because of a more general emotional discomfort that cannot be resolved through pure reason.

    I think any successful transit implementation ends up being a combination of modes, resulting in several layers that complement each other. This provides a robustness that is important equally for reliability and efficiency. Even on a bad day where there is a problem in one part of the network, it’s possible to adapt with minimal interruption. There dichotomy is eliminated between station spacing in terms of the local pedestrian’s needs and the travel time of the cross-town traveller.

    In the Vancouver context, trolley buses largely supersede trams if for no other reason than hills. There is room for LRT routes using old railway rights-of-way or perhaps some routes along popular pedestrian/tourist areas like near the waterfront or Gastown/Robson/Granville where speed/congestion is less of a concern than capacity/loading times/proximity/distance between stops.

    On Broadway today, double-length buses run so frequently there are commonly several within one block and all are at crush-load capacity. That’s a problem that a “cheaper” streetcar cannot solve, because in order to be cheaper it has to be at-grade. The only way to get more capacity is to go faster, which requires grade separation. Partial separation is not an option due to congestion on the road as-is, and in addition there are many major crossroads that require left turns be allowed (eg Burrard, Granville, Cambie for the bridges, not to mention Heather or some others important for Ambulances going to the hospital) or even the minor issue of not severing the bike lane network running on secondary streets with priority signalling. So full grade separation is required for a Broadway route, and that leads to an underground system.

    Once you have the parameters of underground network and desired station placements, the other considerations are obvious from speed to driverless. The only choice to be made is which bidder offers the best deal. Buses will still run along Broadway to serve local traffic, but many fewer will be needed even as more people rearrange their travel plans to take advantage of the faster metro speeds.

    We need solutions that can pay for themselves, and can infectiously spread their network to the extent that is economic.

    I don’t see the grounds for conflict, where do you think it comes from?

    Steve: There are several factors. The example you have on Broadway is not unlike what we have on Eglinton where the LRT line will be underground. The LRT advantage here is that on the outer parts of the line where space is available, the LRT can come to grade, an option not available to SkyTrain nor to full subway. However, there is a strong resistance to the loss of road space and intersection capacity and the larger question is whether this can or should be preserved forever. It is ironic that the existing and planned densities that will drive up transit demand beyond the level where buses are practical will also generate lots of auto traffic and demands for more, not less, road space.

    Streetcars/LRT occupy that range in transit technology where tradeoffs are essential. The kind of improvements typically made for buses (queue jump lanes, loading bays, even peak period reserved lanes) are in many ways benefits to motorists as they (a) get the buses out of the way and (b) provide extra capacity. LRT, by contrast, transfers capacity from motorists to transit. Motorists do not think of transit riders as “us” and don’t factor rider benefits into their mental calculations. North America does not have a tradition of strong public transit systems with political support for giving priority in space and spending.

    Add to this a long-standing anti-streetcar bias among some in transit management, and the natural love of the engineering and construction community for the most expensive transit implementations possible, and you have a political mix that leaves streetcars/LRT with few natural allies. Oddly enough, the whole idea that one might attempt to provide transit at a better balance of cost and effectiveness for corridors where buses are reaching their limits but subways would be overkill is a very “conservative” approach. However, that community tends to be less favourable to transit of any kind.

    We have been through a few rounds in Toronto where transit management and the planning establishment have grudgingly accepted that subways can’t do everything. Then someone wants to get elected, and they announce plans for a network on a grand scale (David Peterson, 1990). When the tap is turned off (Harris), everything stops. Somewhat later we reached Transit City (Miller) which, while not perfect, did not presume subways as a starting point. It took a few years of bitter wrangling for Queen’s Park (via Metrolinx) to accept that LRT was viable, and they are now advocating this mode in several locations. All it will take is one anti-transit election to set this all back a decade. This sort of see-saw debate is possible only because there is no broad public support for transit or understanding of what various options can give us.


Comments are closed.