Streetcar Track Replacement Plan 2010-2014

For those who are purists about their transit advocacy, listing all of the planned track replacements may seem just the sort of thing a railfan might do.  If that’s your attitude, you’re in the wrong article, probably the wrong site.

The streetcar track repairs planned for the next five years mark an important milestone in Toronto’s streetcar network.  For two decades after the TTC decided to keep streetcars in operation, they continued to build poor track, and the quality of construction actually deteriorated as time went on.  Some bright spark thought that unwelded rail, sitting on untreated wood ties, all poured in a slab of concrete, was a wonderfully modern way to built track.  It wasn’t, and we saw streets fall apart quickly, in some cases within a decade of construction.

Things changed in the early 1990s, and over the years the TTC has moved to use continuously welded rail (a practice abandoned in the late 1960s), steel ties and rubber insulation to keep the track from vibrating the roadbed like a gigantic sounding board.  At intersections, the change started later, but recent installations such as Church at Queen and at Adelaide are built to the same standards as the regular tangent track.  This is track built to last.


The photo above shows Church and Adelaide.  Visible here are:

  • steel ties where they can be used, otherwise treated wooden ties
  • welded rail
  • rubber matting around the rails, and rubber moulded around the special work
  • the machine used to move pre-assembled track panels into location

Within five years, presuming the gods of budget cutting don’t intervene, all of the track used by main routes in Toronto will have been rebuilt.  Intersections will take longer because there are so many, and the TTC settled on its current standard of construction for them only recently.

This list may be amended to stretch out spending and relieve pressure on the City’s capital budget.  However, even with some delays, this will see the streetcar track network back in first class condition over the coming decade.


  • The Church Street reconstruction north of Dundas, originally expected to occur in 2010, has been moved forward to 2009 because good progress has been made on the section from Dundas south to King.


  • The major project for 2010 will be the west end of the King route.  Track will be reinstalled in the reconfigured Roncesvalles Avenue, and King will be rebuilt from Queen east to Close.  This will include the intersections of Roncesvalles with Howard Park and with Dundas.
  • There is a project to reconfigure The Queensway so that the streetcar right-of-way continues east to Roncesvalles, but this does not appear to include track construction, only lane restrictions.  The intersection at Roncesvalles will be rearranged by eliminating the eastbound bypass lane from The Queensway to King and replacing it with a dedicated left turn lane for autos.  The eastbound stop will be moved farside (I am not making this up) allowing convenient access to the restaurant “Easy” directly from the Queen car.  It is unclear where eastbound 508 Lake Shore cars will stop.
  • Parliament Street from Gerrard to King will be rebuilt.  Although this is not a streetcar route, per se, the track sees much use from diversions and short turns, and the road is in bad shape.  Slow orders here are common.  The intersections of Parliament with Queen and with King will be replaced.
  • Queen Street will be rebuilt from Connaught to Coxwell including the intersection at Connaught, the eastern access from Russell Carhouse onto Connaught, and the northern accesses onto Queen.


  • A project to rebuild track on Kingston Road from Queen to Bingham, including Bingham Loop, is planned to begin in 2011 and complete in 2012.
  • Track over the Gerrard Street Bridge will be replaced.  Some of this track was part of an early experiment in mechanically isolating track from the roadbed (1991).  This was not entirely successful because, in part, the TTC was still using non-welded rails.
  • McCaul Street from Queen to College including the loop is scheduled, but I doubt they will actually go more than a short distance past McCaul Loop as this track is rarely used and in reasonable condition.
  • Shaw Street from Queen to King.
  • Dufferin Street from Queen to and including Dufferin Loop, but not the intersection at King.
  • Loops at Long Branch and Wolseley.
  • King and Bathurst intersection (one of Toronto’s 3 grand unions).
  • Various yard tracks at Roncesvalles Carhouse (two year program continuing in 2012)


  • Kingston Road reconstruction continues.
  • Roncesvalles Carhouse yard tracks
  • Harbourfront tunnel
  • Queen’s Quay from Spadina to the portal at Bay Street including Queen’s Quay Loop (this is part of the Queen’s Quay revitalization project).
  • Spadina from Sussex to the portal at Spadina Station, around Spadina Circle and from King to Queen’s Quay.
  • Spadina intersections at:
    • King, Adelaide, Queen, Dundas (including Toronto’s other 2 grand unions)
  • Ossington from Dundas to College.  (I have my doubts about the need for this project and it may be in the list simply by virtue of the timing of the last reconstruction, 1992.)
  • Richmond from Yonge to York
  • York from King to Queen including intersections at Adelaide, Richmond and Queen
  • Adelaide from Charlotte to Bay, and from Yonge to Church.  Given the work done as part of the Church and Adelaide intersection recently, I suspect this will actually stop at Victoria.  This will re-establish Adelaide Street as a bypass for eastbound diversions of King and Queen cars after a very long absence due to construction and general deterioration of the track.

The work planned for Spadina below King and on Queen’s Quay will get rid of the “thunder track” installed here for the Harbourfront route’s opening in 1990.  Its noise has been a source of much ill-will to LRT plans in the central waterfront, and, more recently, many slow orders thanks to the disintegration of the track and roadbed.


  • Wellington Street from Church to York
  • Victoria Street from Dundas to Adelaide
  • Intersections at
    • Queen and Victoria
    • Bathurst and Dundas
    • Queen’s Quay and Spadina (it is unclear why this is not part of the 2012 program to reconstruct the Harbourfront route)
    • Charlotte and Adelaide
    • Charlotte and King

The work on Wellington will remove the last of the “wrong way” track on streets that became one-way six decades earlier.


  • Kipling Loop
  • Intersections at
    • Spadina and College
    • Dundas and Parliament
    • Dundas and Victoria
    • Dundas Square and Victoria

34 thoughts on “Streetcar Track Replacement Plan 2010-2014

  1. Now if only they’d start re-using grooved rail as well. They stopped doing this many moons ago and offered the public reason that such track was no longer commercially available. Tell that to the Europeans and you’ll be laughed at. The real reason of course was that the rail was so much cheaper to buy. Methinks the new low-floor cars aren’t going to like T-rail set in concrete but perhaps I’ll be proven wrong. We can hope.

    It would also be nice to see some right of way (such as Queen’s Quay as rebuilt) set in grass rather than concrete. Regardless of the base underneath, such trackage is MUCH quieter in operation, proven many times over in Europe.

    Hopefully the QQ rebuilt won’t also see center poles installed. St Clair and Fleet are about as ugly as you can get, although St Clair around stops with only single poles and extended cross-arms works well. I have problems imagining why this was simply not done over the entire length of the line.

    Steve: The TTC has an amazing supply of what I can only describe as BS to describe why they have to have centre poles. It relates to the claim that if they use hydro’s poles, their overhead will be torn down all the time as people hit the hydro poles. This does not explain why they are content to sling overhead to hydro poles where it’s necessary, not to mention everywhere else in the city. Moreover the centre poles make it harder for buses and other vehicles to drive down the right-of-way.

    The real reason St. Clair got centre poles was a planned change in the street lighting that would have moved the hydro poles further apart. Alas, the new lighting didn’t work as well as planned, and the hydro poles wound up more or less where the old ones had been. The TTC could have used span wires, saved the extra space needed for centre poles, and given us a cleaner-looking street. They were too busy defending a bad decision to change the design.


  2. What did you mean by King and Bathurst being one of Toronto’s 3 grand unions?

    Steve: All eight possible turns are present at this intersection. That’s why it is called a “grand union“.


  3. I see that the Parliament/Queen intersection is being rebuilt in 2010. Do you think they will – or should! – put in a north to west curve to increase options for short-turns?

    I notice that the Wellington track has been postponed until 2013. The track west of Yonge looks in fairly good condition but the track from Church to Yonge is not – I am not an engineer but I doubt it will last until 2013 without major, and expensive, temporary repairs.

    The junction of King/Church – noted in the 2013 list – was replaced very recently. I think 2004 or 2005. The very early replacement may be explained as this is one of the junctions where the automatic switching does not work, there is often a TTC person there to do it with a crow-bar.

    Steve: Sorry — King/Church was a mistake, and I don’t know how it got in there. In any event, the pointman was there for the Queen diversion which used a manual switch west to north.


  4. When they do the Bay St Portal work, that will coincide with the East Bayfront service construction, right?

    Steve: Until we know definitively about funding for East Bayfront, I’m not holding my breath. However, it would make sense to co-ordinate the projects. There are no explanations about project timings in the budget papers, just a list of locations.

    I am surprised TTC is going ahead with Long Branch Loop… I don’t think this should be done until a decision is made about improving the loop’s relationship with GO Transit. They were talking about that during the WWLRT consultations, although where that project is going now is anybody’s guess. Regardless of the WWLRT, the current Long Branch Loop location is poor in relation to GO connections, and I’d rather see the loop redesigned before they replace the track. It should be of interest to both GO and TTC with the coming electrification of Lakeshore and 15 minute service all day long.

    Steve: TTC think about GO Transit? Surely you jest!

    “Grand Unions” are the omni-directional intersections, I take it? There’s only three?!

    Steve: Yes, and we have the largest collection in the western hemisphere, all in walking distance of each other.

    It’s funny that they’re about to remove the remaining “wrong way” trackage at the same time as they are talking about restoring Richmond and Adelaide to two-way streets west of University Avenue. That would only apply to Adelaide since Richmond trackage stops at York, but a curious coincidence nonetheless.


  5. Ossington must have been rebuilt much more recently than 1972, or at least a significant portion of it was. I recall riding on a PCC charter on the then-new track on Ossington shortly after it re-opened sometime in the late 1990s. If I had the time to dig up the photos I’m sure I could find the PCC sitting on the fresh concrete on a photo stop. The rail is definitely the new style, and not the old girder type such as found on Church St.

    Steve: Another oops! That’s a typo. Should be 1992. The intersections are even more recent and were done as part of work on College and Dundas in 2004 and 2007 respectively.


  6. Granted, it’s not replacing track. However, I wonder when the TTC plans to install tracks on Cherry Street.

    Steve: This is tied in with the West Don Lands work by Waterfront Toronto. As you may have noticed, the billboards are now up for the first phase of housing construction, and the reconfiguration of Cherry will occur over the next few years.


  7. If the SHAW loop of the 501 service remains after the trial period, the track on Shaw will need some work. It’s in poor shape now and I didn’t see it on your list.

    Steve: Oops! It’s in 2011. I have updated the main post.


  8. Any news on the repair/replacement of electric switches which seem to be failing everywhere?

    When intersections are repaired or reconstructed in the schedule provided, should we expect manual switches to be replaced? Manual switches may have been acceptable in the 19th century, but I’m amazed no streetcar driver has been struck by a vehicle while jumping from the car to manually set a switch.

    Steve: This project has been in the Capital Budget for some time, and the problem was flagged in 2001(!!!). We have reached the point where the TTC expects to demonstrate a replacement product in 2010, design and procure for the system as a whole in 2011, and install starting in 2012. For a system that cites “safety” so frequently, this is not exactly a speedy response to a well-known problem.


  9. Is it possible that stretch of Ossington is being included because they plan to, somehow, include switches at both ends to allow cars to turn in either direction. Going from east Dundas to north Ossington, and vice versa, would be mighty hard.

    Steve: The intersections are not part of the planned work. They are both less than 10 years old.


  10. Not to be TOO picky… BUT you say (or they say):

    “» Adelaide from Charlotte to Bay, and from Yonge to Church. ”

    They may want to look at – or for – the ‘track’ between Yonge and Bay. There is a totally ‘trackless’ portion – ca 50 feet – here – where they excavated to build the, soon-to-be-opened, PATH connection at the Bay-Adelaide Centre. Possibly it will be rebuilt with $$ from the B-A contractors and not TTC funds but it does need to be there!


  11. 2011-2012: Roncesvalles Carhouse Yard Tracks

    How much foresight has the TTC put into this? There’s rumours that Roncy will close shop once the CLRVs are put to pasture. I thought it is necessary to build a new facility for the Bombardier streetcars, or can tracks at Roncy be reconfigured to accommodate the new vehicles, hence their reconstruction? I understand that there will be an overlap of CLRVs in service with the Bombardier streetcars, but does this justify construction that would keep the carhouse in service for a few more decades?

    Steve: The new carhouse in the port lands will not hold the entire 204-car Flexity fleet. Operationally, it would make more sense to retain Roncesvalles than Russell to distribute the fleet on both sides of the city.


  12. Hi Steve.
    Do you know if the rebuilding of the intersections of King and York and York and Adelaide will included any extra curves? An east to north at King and York and a north to east at York and Adelaide would be handy to have for King and Queen diversions.

    Steve: King and York is not being rebuilt. The TTC was too cheap a few years ago to add the east to north curve, and I say “cheap” because the answer, when asked, as to why it wasn’t done was “budget”. They missed the chance to put a useful curve at a key location.


  13. A “pointman” has been assigned for years to manually throw the switch at King and Church Streets for the left turn movement of #503 York street service via westbound on King St. to southbound on Church St. The electric (NA) switch has simply been inoperable and unable to be made functional.

    The #501 Queen westbound diversion required a “pointman” outside of just the rush hours of #503 tripper service.

    Steve: Considering that the cost of covering this on an ongoing basis is the better part of one operator’s salary, one might think that simply reinstalling the switch equipment might be a lot cheaper.


  14. 2014 for Dundas Square and Victoria…

    I havn’t been to the intersection lately, but when Rogers was under construction, they eliminated the north to east access through the entrance way.

    Have they re-opened it?

    And will it be taken away since Church is so close?

    Steve: It’s not open yet, although the street looks as if it will be passable again soon. It would be a shame to lose this quaint bit of track, but I don’t know its long-term status.


  15. Regarding electric switches, do you know if all of them originally installed on Spadina were ever actually officially activated? I remember some of them weren’t for many years. It would be quite rediculous if some remained un-commissioned all the way through to the up-coming intersection replacements! Also, was there ever any ‘real’ resolution to the problem with the control circuit not staying locked while cars passed over the switch? How many accidents occurred as a result of this issue, and were they due to excessive speed or equipment malfunction?

    It will be interesting to see if any replacement for the track switch control system will allow more flexibility than simply requesting to “diverge here”. There should be a ‘diverge left’ and ‘diverge right’ button at the very least, with something like ‘1st right/2nd right/3rd right’ for locations like the Roncesvalles Yard access. I’ve heard of at least one existing system that works this way. It would be sensible to have either a cab signal or other lighted indication of the route selected. I saw pole-mounted signals in Oslo that did this.

    Steve: I don’t know any specifics on accident stats. Yes, it’s quite ridiculous that many switches on Spadina were never energized. This hooks in with the long-standing request for staff to explain the shortcomings of traffic signal priority.

    The idea of a multi-way route selection is an obvious question, although it would require a change in TTC operating procedure. Today, the selection point for a left turn can be well beyond the right turn, possibly even past the right turn’s switch. If a common route selection is used, then a following car would have to hang back far enough to use a common pickup antenna for either direction. Alternately, the intersection would have to be able to detect and remember the route selection for an incoming car while its leader was enroute through the switches. This requires more sophistication in vehicle tracking than we have on the TTC.


  16. As for Roncesvalles – one must also remember that the new streetcars will not all arrive on the same day, so a carhouse will be required to maintain the CLRV and ALRV fleets while the new streetcars are delivered.

    I find it funny that Kipling loop will not be fixed until 2014, as this picture clearly shows that the track along Kipling Ave is in need for repair sooner rather than later (the tracks turn onto Kipling Ave off Lake Shore Blvd, and then right into the loop).

    The picture was taken in back in August and the concrete around the tracks is shot!


  17. Is the work on Kingston Road likely to be part of one of these newfangled “do it all at once” City projects? If there is rebuild work continuing on east of Victoria Park, I wish they’d consider dropping in streetcar tracks, even potentially unusable deadend tracks to as far as that phase of the rebuild went, with the ultimate goal of getting streetcars to run as far as Danforth & Kingston, where there could be a smooth interchange to the new BRT service.

    As it stands, Bingham is such a middle-of-nowhere point for streetcar service to suddenly stop (was the old municipal boundary part of it?). Alternately, they could run up Victoria Park Ave to Victoria Park subway station, but I guess the train has already left the station on piggybacking on that rebuild.

    Steve: I don’t know what is planned for Kingston Road, but an extension beyond Victoria Park (the boundary between Scarborough and Toronto) is highly unlikely. Streetcar service north on Victoria Park is also a non-starter because the TTC would insist on widening the street and that’s a non-starter for hte residents.


  18. Wow downtown is going to be a complete disaster with work the TTC plans to do on its grand unions. And to add to this, those are very busy intersections which can be a complete nightmare when doing tiny projects such as replacing a hydro wire which shuts down part of a lane for a couple hours … but imagine months. The TTC bettter do this smoothly or there will be some big revolts. O and by the way Steve, how long do the TTC’s grand unions generally last for?

    Steve: The TTC has already rebuilt many intersections that are a substantial subset of a grand union, although in cases such as Church and Queen there was no rush to complete one direction as it had no scheduled service. This work was also done in co-ordination with watermain work in the area, and the street was torn up for a much longer period than needed just for the trackwork.

    A full grand-union will likely take the better part of five to six days, depending on the weather. The issue for the old ones is that they must be torn up right down to the underlying concrete base because many old intersections were not built to as good standards in the first place. The real saving comes in 25 years when the work must be repeated, but only from track level up, not a complete reconstruction of the road.

    The Spadina car began service in 1997, and the intersections will only be 15 years old when they are replaced. This does seem a bit premature, and I wouldn’t be surprised to see this work deferred a bit. Bathurst & King dates from 1995.

    The new construction should be much more robust. In “the old days”, the intersections were assembled on street, bit by bit, sometimes while service continued over the track. Many parts of the intersection were not welded together, and there was no rubber around the rails or castings to isolate the track from the concrete which didn’t take long to start falling apart. Today, intersections are pre-assembled at Hillcrest and broken into panels for delivery to the work site. This simplifies installation and greatly reduces the amount of onsite welding that is needed.

    One constraint on re-opening any intersection is the time needed for concrete to cure. If service resumes too quickly, the concrete is not as strong as it might be otherwise.


  19. I thought I saw you poking around Adelaide/Church awhile back! I have a direct view of the intersection from my window so have had the chance to see the entire project.

    It really is amazing how much care is taken with the special trackwork at intersections. I gather that most of the work (excavation, concrete pouring, etc.) is done by private firms, but the TTC construction staff seemingly spent many hours placing each piece of special track, even working through the night to get the job done quickly.

    Steve: Yes, the road construction is done by private contractors (the work is actually tendered by the City), but the TTC looks after the trackwork.


  20. “Today, intersections are pre-assembled at Hillcrest and broken into panels for delivery to the work site. This simplifies installation and greatly reduces the amount of onsite welding that is needed.”

    There’s plenty of photographic evidence in the City of Toronto Archives that this was done back in the good ol’ days (i.e. before WWII) as well.

    Steve: I’m not sure you are correct. Could you give some citations of photos from the archives showing pre-assembled track sections?


  21. This is the one I could find with a quick search. I have seen a couple of others for other intersections from approx. the same time period, but I can’t remember which intersections.

    Steve: The special work was always preassembled at Hillcrest to make sure that everything fit, but it was then completely disassembled, shipped to the site, and re-assembled in the street. I remember an amusing problem at Queen and Roncesvalles where the track didn’t line up. I believe that, after much contemplation, they discovered a few parts were in the wrong place.


  22. “I believe that, after much contemplation, they discovered a few parts were in the wrong place.”

    LMAO! Was this AFTER it was put into service? 😉

    Steve: No, it was partly assembled, and things were clearly out of whack, so they paused for a bit.

    Well, if most of us get a few pieces mixed up putting together something from Ikea, then I guess it’s possible to mix up pieces for such a complicated intersection as Queen/King/Roncesvalles.


  23. About a “pointman” at King and Church, Steve wrote, “Considering that the cost of covering this on an ongoing basis is the better part of one operator’s salary, one might think that simply reinstalling the switch equipment might be a lot cheaper.”

    We always hear about how capital and operating budgets are totally separate entities (not just here, but in many other circles). Once in awhile, the idea of “capitalizing expenses” comes up, as is the case when transit vehicles are retired early (using capital funds to purchase new equipment early instead of the operational budget to maintain what is there now).

    Given that ongoing operational funding is the one that requires arm twisting (no photo opportunities for politicians), why on earth would the TTC opt to convert what should be a one-time expenditure to an ongoing one?

    I was tempted to call this a case of “operationalizing” a capital expense, but track switch replacement is one of those things that may fall into either a maintenance budget or a capital budget. Case in point: railways may only expense so many replaced ties per kilometre per year and anything above that is a “capital cost”. This is because for tax purposes, 100% of maintenance can be deducted from income in the year, while capital costs have the depreciation deducted over several years.


  24. With the Dufferin jog soon to be but a memory, is there any thought into laying some track and modifying the Queen/Dufferin intersection for a possible run up Dufferin? Even if the tracks aren’t immediately used it’d save on having to rip up the roadbed in the underpass later on. Wasn’t Dufferin south of Bloor one of the potential routes the TTC was once looking at for streetcar upgrades (fleet availability aside)? Even if the line only went as far as Dundas or College for short turn diversions you then have the possibility of upgrading the 193 Exhibition Rocket to streetcar service!

    Steve: The Dufferin underpass has been designed to take streetcar tracks in the future if they are ever needed. As for the 193, it used to be the 522 Dundas Exhibition car and became a bus because the TTC didn’t have enough cars or operators to run it (or at least that was the official story).


  25. As of today, Church Street from Gerrard and Dundas is almost all ripped up. Rotten wood, non-welded rails and a small amount of ties could be seen before placed into the dump trucks. In comparison to the newly constructed tracks south of Dundas, the old tracks were built to a lower standard.

    This also reminds me that track reconstruction could be done in a month whereas on St. Clair, in years. Funny thing they recycled the sidewalk fences from St. Clair, still baring the signs “Shop on St. Clair” that can be seen all over Church Street.

    Steve: Problems on St. Clair related less to track construction itself than to the rebuilding of the entire street including utilities. The actual trackwork barreled down the street once the other work got out of the way.

    The 193 was a better service than 522. Buses have the ability to detour on to narrow streets instead of being stuck in traffic. This is the case for the union parade where the 193 detoured onto Block Avenue while Dufferin was closed off for the parade. The bus squeezed though one lane street with cars park along the side. A little carelessness would of cause a disaster. In the meantime, the 522 would be stuck behind the 505. Sorry, it’s tied down to the tracks.


  26. In anticipation of the new LRVs, will the TTC be installing double/single-tongue points in the rebuilt intersections?

    Also, I thought Kipling loop had been sold for development.

    Steve: There are no plans to move to double-bladed switches for the “legacy” system, only on Transit City routes.


  27. Two things:

    1) What is the difference between a single blade and a double bladed switch? I am not a technical person in this regard.

    2) I never heard that anything about Kipling Loop being sold before now. Unless the 501 is truly fixed (for example through the reinstatement of the 507 car), Kipling would not really be required. However, it better to have “just in case.”

    Steve: A double blade switch is the sort used on the subway and on railways. There is a moving point for both rails. On the streetcar system, there is a moving point only on the “inside” rail of a curve. This difference affects the way that forces are transmitted between the track and the wheels of streetcars as compared to subway and railway trains.


  28. Are you sure that the Kingston Road work won’t begin until 2011? I live on Kingston Road and it looks like they’re getting ready to do some serious work. They’ve painted markings every 5 metres along the roadway. I’ve seen a lot of Toronto Transportation trucks and some TTC workers along the street over the past month.

    Steve: My list comes from the TTC Capital Budget background materials. This list gets shuffled around from time to time to accommodate other plans, notably Toronto Water’s projects. The other possibility is that they are just doing the detailed survey work now in preparation. I will try to find out what the plans actually are.


  29. I’ve noticed that sometimes the Streetcar rumbles along certain sections in the network (eg. King, south of Queen and Roncesvalles). But why would rumbling occur on the St Clair Avenue ROW around Deer Park? Is rumbling caused by the half-assed track installation? If not, then what causes rumbling?

    Steve: The track on King east from Roncesvalles to somewhere around Jameson is “old” track that was not installed with rubber around the track, among other structural changes dating from the mid 90s to improve the noise, vibration and lifespan characteristics of the track. That stretch of King is scheduled for replacement this year as part of the job to reinstall track on Roncesvalles, and this is one of the last major piece of “old” track on the main streetcar routes. (There are still a number of lesser used tracks such as Parliament that have not been redone.) The Spadina route is “new” track, but the original Harbourfront line (King and Spadina to Union) predates the new construction methods and is very noisy.

    St. Clair is “new” track and should not rumble. I will have to go to Deer Park to listen to exactly what you’re talking about.

    There are two principal sources of noise. One is that the old style track does not damp vibrations and the whole road acts as a big resonator. This also accelerates the disintegration of the road itself. The other source of noise even on “new” track can be corrugations. This is the same effect as you see on a dirt road, except at a much smaller scale, and it causes wheel roar when a streetcar passes over corrugated track at speed. It is much harder to develop these on the “new” style track but not impossible. Another related factor is ground water, but that’s more involved in transmitting vibrations from underground structures.


  30. How long should this new track last?

    On a somewhat related note, how often have subway tracks been replaced?

    Steve: It is important to remember that there is more than “track” in the street. There is a foundation slab, and sitting on top of that is the slab containing the steel ties. Finally there is the layer with the track itself.

    The topmost layer would be torn out when the track is worn, typically in 20-25 years depending on the level of service and local conditions. Rails at carstops wear out faster and are replaced in short sections, but this does not require the underlying structures to be rebuilt. Even when we reach the point of full track replacement, the foundation and the ties will remain and only the top layer will be demolished.

    Intersections are a different situation. Only very recently have they been built to standards comparable to the tangent track with fully welded sections and rubber insulation between concrete and track to minimize concrete breakup. We are comparatively early in the cycle of getting major intersections on a long term footing.

    The Spadina intersections from Adelaide north date from about 1996 on average (the line opened in 1997), and they are due to be replaced possibly starting in 2012. They have not completely collapsed, and doing them all at once may be as much about scheduling convenience as the condition of individual locations. The track from King south, and east on Queen’s Quay to Bay is the original Harbourfront line that opened in 1990. It’s 20 years old and in rough shape. That track was not built to the new standards, and many local fixups have been needed over recent years to correct problems with collapsed concrete.

    Subway track is replaced on roughly the same cycle, although this is shorter for places of high wear (stations, curves and frequently used switches). Usually you don’t see this work except as a slow order where a gradual change from old to new track is taking place overnight. A new crossover is being installed south of College Station (the original one was removed many years ago) as part of the YUS signalling project. There is a slow order because the ballast has been dug away from the ties in preparation for swapping in the special work.


  31. My understanding of a road structure is that it consists of the following layers (from top to bottom):

    Asphalt surface (or concrete surface, where there are streetcar tracks)
    Concrete surface
    Base (granular)
    Subbase (also granular)

    The Transportation Services website states that resurfacing of asphalt takes place every 18-25 years for normal roads. Is that how long the concrete underneath the asphalt lasts too?

    Here’s their website.

    Since streetcar tracks last for 25 years, is it true that the presence of tracks wouldn’t make a difference in how often a road is torn-up?

    This is an argument commonly used by anti-LRT people, that streetcar infrastructure doesn’t last long and causes streets to be torn up frequently (creating more havoc for neighbourhoods). I’m trying to refute that. You’ve also mentioned that subway tracks and streetcar tracks have the same lifespan too.

    Final question: Do you mind if people comment on your really old blog posts? Does it create any more trouble?

    Steve: You can comment on any post you like. Only people who are using RSS to see recent comments are likely to read the discussion.

    Streetcar track construction methods have varied over the years, and the generation of track built between roughly the mid 70s and the early 90s was not the best in many ways. The TTC had cut back on track quality in anticipation of system abandonment that would have started in 1972, and they didn’t really get back to building “good” track for two decades. The very high rate of track replacement over the past decade has been the combined effect of the bad track wearing out faster and the older, better track coming to its normal end of life. There are other factors too including the use of unwelded rail, the absence of mechanical isolation of the track from the roadbed, and the use of Bochum wheels on the CLRVs as delivered. Those were eventually replaced by the SAB wheel which is similar to a PCC wheel and which is much more suited to track laid in concrete because of its vibration characteristics. The CLRVs did pound the track to bits until those wheels were replaced because of the amount of low frequency vibration the Bochum wheels transmitted to the concrete through the tracks.

    Current track construction has three layers of concrete sitting on top of the road base. The bottom layer is the foundation pad, and it is intended to last for a long time. It was intriguing to see some of the older track didn’t have this pad when the roads were torn up, and without it, the whole track structure can deteriorate faster as it settles into the ground. The next layer up (poured as a separate step in the construction) holds the steel ties. Again, this layer is intended to have a very long life. The top layer contains the track itself, and that track is encased in a rubber sleeve to isolate it from the concrete. The track is welded into strings so that the “click clack” of passing wheels doesn’t cause vibration to break up the concrete. When track wears to the point it must be replaced, only the topmost layer is removed and this is much faster and cheaper than digging all the way down to the foundation layer.

    We are now at the point where much of the system has been rebuilt to this new standard. King will be completed this year (work starts east from Ronces to Close on May 30), the south end of Spadina and Harbourfront are scheduled for the fall, and even the non-revenue track is well on its way to completion. A chunk of Queen East near Russell Carhouse was to be done this year, but the project was put off to 2012 thanks to the Mayor’s intervention. Kingston Road awaits completion of a watermain project starting this summer which will cause buses to replace streetcars.

    Intersection construction methods have also changed, but more recently and so there is a longer backlog of “old” track. New installations now include a completely new base, and track is laid in pre-welded panels and encased in rubber. This makes intersections much quieter and extends the life of the concrete. It’s a bit too early to say how much this will add to the lifespan as the technique is relatively recent. Dufferin and King was the first major intersection rebuilt this way. By contrast, the high-traffic intersections on Spadina are falling apart even though they are less than 15 years old, and there are slow orders everywhere. They are scheduled for replacement in 2012.

    Anti-LRT advocates are correct in stating that the track wears out too fast if they use the “old” TTC style construction and vehicle technology as their reference point. It’s almost as if the TTC set up the system to be expensive to maintain, but I prefer to think of this as a problem of lack of understanding or incompetence rather than Machiavellian plotting. Properly built LRT track not only lasts longer, it is cheaper to maintain because it does not have to be forever patched back into safe condition.

    As a footnote, last year, I documented the reconstruction of Roncesvalles Avenue and Parliament Street in considerable detail. This work went extremely quickly from a track construction point of view. The delays on Ronces were all traceable to a combination of the contractor and cock-ups with the utilities. Both of these projects involved “from the ground up” construction at a scale that will not be required the next time these tracks have to be replaced.


  32. “The bottom layer is the foundation pad, and it is intended to last for a long time. … The next layer up (poured as a separate step in the construction) holds the steel ties. Again, this layer is intended to have a very long life.”

    Just checking, those concrete layers last as long as 50 years? (All other roads without streetcar tracks have a 50 year lifespan).

    Steve: None of the current batch has been in place long enough to test the 50 year theory, and many of us won’t be around to tell. However, there is no reason to believe that they won’t last a long time independently of the top “track” layer of the street which may need replacement/repair.


  33. The concrete around the Carlton tracks at Church St. is crumbling and is frequently patched with asphalt. Is this an older section of track scheduled for rebuild, or perhaps bad concrete?

    Did I recently read of a plan to install a NE curve here?

    Steve: The tangent rails on Carlton date from 2004 when the left turn lane was added between the tracks in front of Maple Leaf Gardens. The intersection itself from 1999. This may be bad concrete, but I would have to look at it to be sure. The section at the stop rails would wear out faster.

    As for a NE curve, that’s possible, but not until the whole intersection is up for renewal. There was a report at the TTC arising from a suggestion from an operator to Adam Giambrone that talked about various possible new curves. I commented on it at the time, but the whole issue of adding anything to the streetcar system is in limbo right now with Rob Ford’s anti-streetcar attitude.


  34. Does the new TTC welded rail involve the use of “Breather switches”? Wouldn’t having a joint also lead to more vibrations anyways, than if there were no joints?

    Steve: The only actual expansion joints in the streetcar system are on bridges. Yes, having fewer joints reduces vibrations and this improves the longevity of the track structure.


Comments are closed.