The following comment was submitted by “Dave” in the thread on King Car analysis, but it deserves a post of its own. It deals with the planned redesign of Roncesvalles Avenue, the St. Clair experience, and where we might be headed on road design in general for streetcar routes.
An important point to consider is that the Roncesvalles tracks, and in fact the entire street, are going to be rebuilt very soon, including alterations to improve the public right-of-way (ie, wider sidewalks and boulevards with sreet trees, intersection bump-outs, etc).
An example of this can be seen on the Roncesvalles Village site.
The TTC is apparently not pleased with the proposal because they think that the more narrow sections of the street will result in left-AND right-turning traffic blocking the streetcar.
They counter-propose a different cross-section.
I haven’t seen anything regarding whether Roncesvalles will be getting transit signal priority installed, so perhaps these alterations will make no difference either way, since the streetcars will often get held up at the traffic lights anyway.
To me, this speaks to the much larger issue in Toronto of how street cross-sections are arranged when streetcars run on them. Street design affects everything that is within the PUBLIC (yes, we all “own” the space) right-of-way, including transit service, trees, benches, garbage cans, car traffic, sidewalks, street lights, etc.
It seems that Roncesvalles could become the St. Clair debate all over again…there just doesn’t ever seem to be much agreement between the City, the TTC, residents, and/or businesses, on what are the “optimal” cross-section treatments for streets with streetcars on them.
For example, St. Clair ended up giving away a lot of right-of-way space over to traffic, at the expense of sidewalks, in some areas. Is the St. Clair treatment the “best” option?
I understand that balancing the space given to traffic, pedestrians, and transit is extremely difficult, especially with narrow right-of-way widths for certain streets.
But in Europe, they seem to be able to create very lively and vibrant streets that are extremely pedestrian-friendly (with trees, benches, cobble-stones, etc) but that also accommodate good transit service (trams) and even occasional automobile traffic.
One could argue that these narrow, European, pedestrian-friendly streets result in less traffic choosing to drive and park on it, resulting in more walking trips, more transit riders, AND better transit level of service.
This raises the question of what IS the “optimal” cross-section for “streets with streetcars”?
Should space be given over based on the percentage of people trips? For example, if 50% of people walk or take transit, give them 50% of the street right-of-way space? (Just a random percentage I chose…I have no data to back that up.)
It seems we’re just randomly experimenting with each streetcar line individually (Spadina, St. Clair, future waterfront lines, future Transit City lines…), instead of borrowing cross-sections that actually work from other successful European cities with streetcars and trams.
I see this being of upmost importantance when Transit City EA’s finally begin and we have to start taking right-of-way space from each street user on the new routes. I’d hate to see an S.O.S. St. Clair fight on every route!
One big difference on Roncesvalles is that the businesses seem to be embracing the concept of the street as a neighbourhood rather than as an arterial road. This changes the priorities they have as compared with the folks on St. Clair. The latter, of course, is also a wider street and this changes any cross-section exercise that will be undertaken there.
Other streets like Queen’s Quay and Cherry are also the subject of “cross section” studies and in both cases the results will be very site-specific. For example, Queen’s Quay doesn’t have much south of it (except the lake) to generate traffic and this allows a cross-section from north to south of street, transit, pedestrian realm. Cherry will probably wind up with the transit offset on one side as well. These schemes also allow the sidewalk to double as the transit loading island in one direction.
That sort of approach won’t work on a street like King with strong two-way traffic and major traffic generators on both sides. Out in the suburbs on arterials where Transit City lines are planned, the design must again be different reflecting the nature of these streets.
A one-size-fits-all “solution” won’t work.
I hope that the TTC installs a decent signal priority system here, and that the Roads Department doesn’t get in its way. Obviously, signal priority will never be as effective here as it should be on Spadina and St. Clair, given the presence of mixed traffic, but if well designed it should perform reasonably well.
Steve: Look on the bright side: there are not too many traffic lights on Roncesvalles, and this limits the negative impact that “transit priority” might have on the service.
Like most large organizations, the TTC does seem to have a passion for standardization though it’s clear, as Dave notes, that for transit – and, many other facets of life – one size does not fit all. An example is the TTCs fairly new insistence on only building streetcar lines in their own separate rights of way. This is certainly going to provide faster and more reliable service on long streetcar lines but the line being proposed in the West Don Lands Project down Cherry Street from King to Queen’s Quay is a very short distance. This will presumably be operated as a branch of the 504 King line , and a very short one at that. A streetcar right of way may be better in many circumstances but for a section of a longer line which is not in its own ROW (and never going to be) having a separated ROW for the length of Cherry is unlikely to measurably improve either speed or reliability.
The discussion on the Cherry Street line was based on a separate ROW being essential and this clearly makes it more difficult to bring this line to the south side of the railway embankment. At the last public meeting on the West Don Lands Transit I was told by a consultant that, though they were pushing the TTC to put the turning circle underneath the Gardener, the TTC were resisting due to the cost of punching a new tunnel under the railway. Sooner or later the Cherry line will need to link with the Queen’s Quay East line – it seems better to me to “get it over with” and build a loop under the Gardner which could be used as the terminus (and possible transfer point) for both the Cherry Street line (which is supposed to be built in 2008/09) and the Queen’s Quay East Line (supposed to be built in 2009/10).
Steve: The consultants really can’t get their stories straight. At the consultative meetings, we tried to get a straight answer on linking the Queen’s Quay and Cherry tracks repeatedly, and were told that this couldn’t be settled until the design for the mouth of the Don was completed. Well, now it is. Meanwhile, it finally dawned on some of the folks at the meeting that with half of Cherry north of the railway as a private right of way, the volume of traffic through the underpass would not require that the transit line have its own space. We’re still hopeful of a decent design using the existing spans.
As for loops, the larger issue is the status of an extension down Cherry south of Queen’s Quay to serve developments planned for that area. I’m waiting for the consultative meetings to start up again in the fall to get a sense of where things stand.
I have to agree with Steve. Searching for an “optimal cross-section” is a fruitless exercise. Ronces is such a different street from St. Clair in so many ways. A good design will always be hugely dependent on context.
While context is plainly important, what I think what the OP was getting at is that there should be a set of guiding principles as opposed to a rigid LRT form which must be imposed on both arterial and neighbourhood routes. While every street is somewhat different, it seems that the City decides to completely reinvent the wheel every time they take a swing at new or upgraded alignments, rather than using an evolution from existing designs. This then makes LRT harder to sell to areas of the city which don’t have it, since how are you to tell people what it might look like from an example of a similar existing street when TTC and Roads make it up as they are going along.
I just discovered Steve’s site last month and love it! I live in the Ronces hood and I too hope we have good, fast streetcar design as well as pedestrian friendly sidewalks. I think many people in the hood who drive (I drive too – but not much) would agree with me. We all want a more people-friendly neighbourhood – it’s part of the scene!
The fact that Roncesvalles has a nearby parallel auto-oriented thoroughfare (Parkside Dr) suggests that accommodating car traffic isn’t as important as it was on St. Clair, which is really the only thorough east-west street for quite a distance. Given the confusing nature of lanes ending and becoming left turn lanes on the old St. Clair, the new private right of way I think is an improvement both for transit and for the car. As for Roncesvalles, I don’t remember ever seeing lots of traffic on it. Is there a lot of car traffic on the street?
Steve: Ronces has less traffic than St. Clair, although there is a move underway to divert some of the existing traffic now on Sunnyside Dr. onto Roncesvalles by adding an east-to-north left turn lane (with the usual transit-friendly signalling, no doubt) at Queen & Roncesvalles.
Both proposals have merits and the collaborative tone on the Roncesvalles Village blog is very encouraging. I wonder if the right solution is a mix of both: the TTC’s proposal for major stops, and the BIA’s proposal for more minor stops.
City-wide consistency is too much given the huge range of street types, but it would be awfully nice to get to the point where a successful design for certain streets could be used as a starting point for similar ones. Right now people show up at these meetings with nightmares about St. Clair or Spadina; if they came with dreams of their neighbourhood looking like Roncesvalles or Cherry, that’s major progress.
The TTC wants more than just transit-friendly signalling at Queensway/Roncesvalles.
The left turn onto Roncesvalles from the Queensway apparently will involve moving the TTC island from the west to the east side, along with the creation of a separate left turn lane, located south of the exclusive streetcar lane. This will also require the creation of a through lane on the east side where the sidewalk currently lies.
As a result, the southeast sidewalk will have to be narrowed by the width of a car lane, making it barely legal according to Vibrant Streets. This, at one of the most important pedestrian corners in the west end, serving as a gateway to the western beaches, Parkdale and the foot of Roncesvalles Village!
Can you think of a reason why the TTC and left-turning car traffic can’t share a lane, just like they do all over downtown? Alternatively, why can’t TTC drivers trigger an advance green and flush any left-turning traffic out of the intersection? Why should Parkdale pedestrians lose half their sidewalk?
When it comes to moving their streetcars through the downtown, it seems the TTC can’t think of anything more creative than creating a complicated maze of exclusive lanes and traffic restrictions that are likely to cause more problems than they solve!
Steve: Yes, I was hoping to avoid that discussion until a separate post, but we might as well include it here.
I am appalled by the proposed configuration of the intersection. Actually, the business of moving the stop farside is NOT a TTC initiative, but is part of a scheme to improve the ability of traffic to use Roncesvalles rather than Sunnyside to get up to Bloor.
The curb cut proposed for the southeast corner is ludicrous.
Equally odd is the proposed relocation of the westbound stop to a farside location (shown in drawings for the Waterfront West LRT) which would place the island in a constrained area between the intersection and the carhouse entrance.
This proposed design must not be approved.
Steve – with the possible arrival of double ended, double sided LRTs, is it possible we will see island streetcar platforms? Are there any configurations where this might be an advantage?
Steve: I expect that this would only apply to underground sections of the Transit City lines or on wider streets. One big issue with a consolidated platform would be the amount of traffic on it at busy stops, not to mention shelter placement and openings.
I will repeat my past comments regarding shared centre-island platforms.
Single shared islands can reduce the amount of road space taken away where two side islands would have otherwise been typically used. This is mainly because there only needs to be one common support structure for the shelters. Also in most locations there would always be an imbalance in the volume of passengers in each direction such that overflow from one side of the platform can be handled by vacant space on the other side. I use this principle to my advantage frequently at St. George station by walking along the opposite side of the platform rather than trying to squeeze my way through all the folks waiting on the busy side. Common electronic signs and fare machines would be another cost-saving bonus.
I fully expect the TTC to go the lower-cost route of single-sided LRVs even on the new Transit City lines. As such I would again strongly support the idea of left-hand running in the underground sections and on any other long stretch with centre-islands. It is a very simple matter to flip the tracks as required and signallize these ‘diamonds’. This arrangement allows for a more practical conversion to double-sided cars if future demand justifies the additional cost. At that point the crossing tracks could be converted to complete, switched double-crossovers to enable turn-backs. I really hope someone will see the logic in this scheme and put it forward to the powers that be for study (unless double-sided cars are on the table from day one that is).
Steve: My main point about shelters, just to clarify, was that on a “single sided” platform, a shelter can have a closed “back” side to guard against the elements. On double-sided platforms, there is no “back” and this decreases the shelter’s ability to do its job. Moreover, clearance must be left on both sides of the shelter for people to walk past it. Yes, there is a saving, but not as much as it might appear. I do agree with the left-hand running for tunnels if we have single-ended cars. Stations cost a lot of money, and building them with centre platforms saves a lot not just for space, but for accessibility via escalators and elevators.
As it is now, Roncesvalles is probably one of the best street car lines in the city. There is rarely much heavy traffic, and there are very few traffic lights. The real problem, as Steve has alluded to in the past, is that Roncesvalles is just one leg of the Broadview-King-Roncesvalles line. Cars are often short turned on King, reducing service on Roncesvalles itself. I think there are many other streets that need service improvements such as signal priority/override/preemption first!
Reading through all of the blogs here and on Steve’s site which great, seem to address almost all matters and concerns.
I haven’t seen the DWG’s for Roncy but have seen some of the renderings and I’m not sure whether the streetcar tracks will be raised and have curbs which don’t allow crossing over with a bicycle or a car etc.
Does someone know whether the raising of the tracks is happening ?
Also, haven’t seen too many improvements to the lower part of Roncy just about where the Credit Union is downwards, will there be any curbing done in this area and towards Marion St. and onwards ?
Otherwise if construction is planned properly not to disturb business and pedestrian flow of traffic, this will be a wonderful street to live by and enjoy its quaint little businesses.
Steve: No raised curbs for the streetcars. There is still quite a debate about the design of bike lanes at carstops.