Tom Jurenka sent in the following note, and it raises questions that deserve a debate.
As a non-native Torontonian (grew up in Winnipeg, but have lived in Toronto for 24 years now) I have always been puzzled — and often infuriated — by streetcars (and the absolutely terrible traffic light timing in Toronto, but that is another story).
My question is an honest one — WHY? All I can see is the negatives of streetcars:
- they tear up streets (I’ve lived through Queen Street E, Gerrard, now St. Clair, being torn up utterly to undo the damage of streetcars pounding the rails)
- they are slow as molasses (as a bicyclist, I routinely pass 5 or 6 streetcars on Queen Street heading from AC Harris to downtown)
- because of their slowness and immobility they delay traffic all the time, causing snarls and the attendant idling pollution
- they are super expensive (witness the recent funding mess)
So I’m really curious why streetcars are a better alternative to trolley buses or just plain old buses, which move fast, are mobile, and are less expensive per unit to buy. Would you be able to point me at some links/articles/studies/whatever to help me understand this issue?
Thank you for your time.
This is a far more complex question than just the list above, but I will use this as a jumping off point.
Without question, track construction is a major pain in the ass for affected neighbourhoods. We are now nearing the end of a long program to correct the combined effect of short-sighted TTC practices in track construction and design flaws in the CLRV fleet. Once that’s done, the frequency of track construction, especially on the grand scale we have seen for the past eight years or so, will diminish and along with it the associated disruption and capital cost.
This is an object lesson in the perils of bad design and the long-range effects of poor choices.
Until the late 1960s, TTC track was built from continuously welded sections of rail, and this was installed in the street in a manner that made it fairly easy to dig up and make repairs. The welded rail holds together much longer and does not produce vibrations at the joints that lead to breakup of the pavement. In 1968, after a derailment accident with one of the two crane cars used to perform track installation, this practice stopped. At that time, the TTC’s policy was that streetcars would be gone by 1980, and there was no point in building track that would last for decades.
This decision, however, was compounded by a change in road paving standards imposed by the city primarily to increase roads’ structural capacity for trucking. All pavement around streetcar tracks was built of concrete slabs with no mechanical isolation from the track. The vibration of passing streetcars and the effect of the unwelded joints led to concrete deterioration around the rails. Salt water seeped into the trackbed, and the common freeze-thaw problem further broke up the pavement.
When the TTC had decided to keep streetcars, track construction standards did not revert back to the old, more robust welded arrangement, and it was claimed that the concrete would hold the rails in place. This short-sighted stance was compounded when the new streetcars arrived. They were about 25% heavier than the PCC cars, mainly because they had been designed for high-speed suburban operation that they would never actually see. Moreover, the wheels on the cars were particularly good at transmitting low frequency vibrations into the ground, and this accelerated the demise of the track. That problem has since been fixed with the use of new wheels, but we are stuck with the weight.
Fast forward to the early 1990s. By this time, the track infrastructure was badly deteriorated through inferior construction and vibration from the newer fleet, and the TTC had to roughly double the rate at which it replaced track. Roadbeds that should have lasted 25 years were wearing out in about 10. They are now using a construction technique with welded rail and mechanical isolation of the track from the roadbed. Moreover, the substructure uses steel ties, rather than the untreated wood used since sometime in the 1970s. This means that the track bed will not disintegrate as the ties rot underneath it. Finally, all recent construction has gone right down to the base slab, and the lower layers of the structure should last a very long time with future track replacement limited to the upper part of the structure.
Just as we reach that point, the City has begun a program to rebuild its antique watermain system and this has complicated and lengthened the period during which streets are under construction (Roncesvalles and Church as examples). On St. Clair, leaving aside basic design issues, the decision to rebuild just about everything — water, hydro, streetcar track, roads — sounded like a good idea. Do it once and get it over with. However, in practice there were many problems with co-ordination of the projects and arbitrary changes by individual agencies that cascaded through the overall plan. The design and tendering process was done in such a way that work did not begin as promtly as it might in some areas, and jobs that should have finished in one construction season dragged into two.
The streetcar, as a vehicle, is taking the rap for many atrocious decisions of past TTC management and poor contract co-ordination by the City. On top of this, the St. Clair project suffered from one major problem: many competing interests wanted their priorities reflected in the design, but there simply isn’t enough room on St. Clair to fit everything in. Some design decisions resulted in precious space being lost, and political decisions favoured priorities for motorists over pedestrians.
In brief, the TTC decided to retain streetcars in 1972, but its track construction, if anything, deteriorated over the following decades. Couple that with heavier cars and a bad initial choice of wheels and you have a recipe for a self-destructing system. Stir in a decision to perform “co-ordinated” repairs by many agencies and you have never-ending construction projects.
Slow Streetcars on Queen
Some aspects of streetcar operation are going to be inherently slower than road traffic because transit vehicles must stop to load passengers. However, this is compounded by several TTC operating practices.
- Queen runs with larger cars, but at most stops all loading is done through the front doors. This underutilizes the space within the car and substantially increases the time spent at stops. This problem will not be corrected until the new cars are in operation with all-door loading and self-service fare collection.
- At all intersections with switches, the TTC now has a stop-and-proceed policy for facing point switches. This prevents cars from quickly pulling away from stops and, on occasion, even results in cars getting just far enough to trip the sensor for the “transit priority” signal and turn the light red against the streetcar (which is assumed to have crossed the intersection). Why stop-and-proceed? When the ALRVs (the long cars on Queen) were delivered, the TTC had to change the way in which electric switches were controlled to a system that would work for any vehicle length. This new system has never worked properly, and to guard against derailments, operators must approach any switch prepared to stop in case it leaps open in front of them. This also happens at manual switches where only a transit poltergeist could force the switch to move. There is a capital project to replace the switching systems, but it has not actually started yet.
- On Queen, one of the ways that the TTC has attempted to deal with short turns and service reliability is to pad the schedule with recovery time. However, at some times of the day the running times are grossly excessive and cars must kill time to avoid getting ahead of schedule. The TTC has a policy of fining operators for running “hot” even when it is impossible to avoid this because the schedule deliberately has extra time. This is a classic case of conflicting priorities. Streetcars are quite capable of sprightly operation. A move from schedule-based line management to headway-based management would help a lot in this area. Cars would be able to move at whatever the prevailing traffic speed is provided that they maintain a regular spacing from each other.
You mentioned traffic signal timings. Toronto has a self-image as a world leader in transit priority signalling. If this is true, then the rest of the world must have little postcards of Toronto intersections sitting on altars for worship by frustrated traffic engineers. I think not.
There are a number of problems with “transit priority signals” in Toronto including:
- There is no mechanism for interaction between operators and the signals. A car can arrive at an intersection where a large crowd is waiting to board, but the traffic light will hold green for the streetcar to speed it away. There is a good chance that this process will time out, and the signal will turn red against the streetcar just when it needs a green. This is time that could have been used for the cross-street. Once an operator signals that departure is imminent (the crowd is nearly all boarded), the signals should clear for the cross-street(s) as far as the next stop (see next point).
- Many intersections do not work ideally for transit vehicles because detection for an oncoming car is too close to the signal. This can happen where signalized cross-streets are close together and some of them don’t have transit stops. The system should control minor streets based on the progression of green time for streetcars from locations where there are stops, not on a block-by-block basis.
- At farside stops (Spadina, Harbourfront, St. Clair, etc.), streetcars can be forced to stop twice because they arrive at an intersection on a red cycle and must wait for the following left turn phase before finally crossing to the transit island. On Harbourfront, there is a separate transit cycle and streetcars cannot use the green time for through traffic. The cycle is only long enough for one streetcar to cross even though pairs of cars are common during heavy service periods.
- At a few locations, notably crossings of Lake Shore Boulevard, the green time for east-west traffic is exceptionally long even though there are times when no traffic can be seen. This is an example of a system that is incapable of analysing actual conditions and adjusting priorities accordingly.
All of these problems would be visited on buses were they to replace streetcars. The problem is with traffic engineering that is not transit focussed, putting it generously. The priority is for maximising the green time available for cars together with a “trickle down” claim that if cars move faster, so will transit. The TTC complains about this all of the time, but refuses to publicly advance alternatives during design discussions such as St. Clair. Effectively, the TTC is complicit by its lack of open advocacy for better signalling practices.
Cost of Vehicles
Vehicle costs need to be compared in light of capacity, lifespan and operating expenses.
The new streetcars have a base price just under $5-million. For this we get a car that should last at least 30 years and be rebuildable for another 10 to 15. The vehicle capacity is claimed to be 240 by Bombardier, but I remain skeptical that this can be achieved under normal operating conditions. By contrast, the design capacity for scheduling purposes of an ALRV is 108 compared with a crush capacity of around 150. The new cars, with all-door loading will achieve a better use of space, but I suspect their capacity for planning purposes will be around 150. By contrast, the capacity of a bus is about 50 for scheduling purposes. For a subway car, the design capacity is about 167 (1,000 per train) although the crush load is about 200.
Diesel buses cost the TTC about $500K each based on an order for delivery in 2010, and hybrids cost about $932K each based on the current order. I will not comment on the relative merits of propulsion technologies or political choices of one over the other as that is outside of my scope here. We can hope that hybrids will become relatively cheaper and of course there are some savings, but not as much as expected, in fuel costs to offset the higher capital cost.
As a matter of interest, back when Streetcars For Toronto argued for streetcar retention back in 1972, the relative cost of a new streetcar and a new bus was much lower than it is today, even allowing for the much larger size of the new cars. Today, a new streetcar represents roughly 3 buses on a planned capacity basis. The cost per “planning” space on a streetcar is $33,300, on a hybrid bus is $18,600 and on a diesel bus is $10,000. However, the streetcar will last at least twice as long as a bus and will have lower operating costs per passenger. The operator driving a bus is one of the most expensive components of that transit service.
The figures above are quite rough and are not corrected for inflationary effects of deliveries in different time frames. They are meant to give a general indication only, not a definitive answer.
When the City decided to keep its streetcars in 1972, demand on the streetcar lines was better than today in most cases. Part of the change can be traced to changes in demographics, land use and travel patterns. However, much blame rests with the TTC.
Starting in 1980, “excess” capacity was tuned out of the system — “tailoring service to meet demand” was the catchphrase from a former TTC Planning Manager. This is the best sort of accounting exercise that ignores the real effect of trimming the so-called fat.
Transit routes encounter all sorts of upheavals, and it is impractical to schedule service based on every bus or streetcar having a full load. Even within a peak hour, there will be a super-peak which the service tries to accommodate. The less “excess”, the more likely any overload from traffic congestion or variations in demand (something as simple as whether or not vehicles meet at heavy transfer points) will affect service leading to delays, short turns and overcrowding.
The link between service levels and riding on the streetcar routes is a chicken-and-egg question to some extent, but one thing is clear. Where service dropped markedly through the combined effect of service cuts and wider headways for ALRVs (Bathurst and Queen routes), riding fell. Where service stayed roughly the same (King) riding held.
In the future, the downtown lines will have to cope with growing riding if only there is service good enough to attract it. The population along streetcar routes will rise and, with it, demand on those routes.
I have written previously on the decline in service on major routes both as an update to Transit’s Lost Decade, a review of service and ridership since 1976 and a look at that old TTC slogan Always A Car In Sight taking things right back to the mid-50s when the Yonge Subway opened.
Recent changes in the Service Standards, part of the Ridership Growth Strategy, have lowered the design targets for vehicles resulting in more service. However, there are two big caveats: there must be enough vehicles and operators to actually field the service, and the budget must have enough funding to pay for it. Streetcars are in short supply, and as readers will see when I review the September 2009 service changes, so are buses. We have riders, we even have the political will to pay for more service, up to a point, but we have no more vehicles.
Taking King as an example, the AM peak headway gets down to 2’00″. The design capacity is, roughly, 23 CLRVs at 74 plus 7 ALRVs at 108 for a total of 2,464. Providing this with buses would require almost 50 vehicles per hour, a service frequency low enough that it would add considerably to congestion due to platooning. This frequency is possible on some suburban routes only because the streets are wider and there is mixed local and express operation.
By comparison, the Dufferin bus operates on an old-style 4-lane city street, and provides under 1,200 passengers per hour of design capacity. Buses there commonly run in packs and the capacity is not evenly utilized.
New neighbourhoods in the Waterfront are designed with the premise that most commuters will use the TTC, and projected demands are at and above those now seen on the heaviest parts of the streetcar system. These cannot be met with buses even with a move to articulated vehicles, assuming reliable, long-lived versions of these can be found.
The streetcar routes need more capacity, but operation with buses would limit what the TTC could provide. As lands along these routes redevelop, transit capacity must also rise or we will have the absurd situation of forcing people to drive cars in the very part of Toronto where it should be easiest for them to use transit. The absence of good service has a cost too, although it may not appear on the TTC’s balance sheet.
This article is not intended as a definitive argument for streetcars, but an overview of major issues. In 1972, Toronto saved its streetcar system, but then did little to reinforce that decision with better service and system expansion. We waited almost two decades just for the Harbourfront shuttle, and a quarter-century for the restoration of service on Spadina. Suburban expansion was completely off of the table.
Times change. I won’t really believe that Transit City exists until I can ride the lines and see new cars brimming with happy riders, but it’s a goal the city finally has pursued.