Toronto Eglinton Line 5 Crosstown West Extension | ?m | ?s | Metrolinx

[ARG1]

You realize that there exists a facility at Laird to turn back trains without them reaching the surface, right?

In fact, the original service plans call for every second train to be short-turned there. And all current messaging implies that as well.

Dan

I was specifically addressing the points he was making. While yes I do know that trains will short turn at Laird, the question is still can you run trains every 90s in the central section, and the possibility is dubious at best. The trains coming in from Sunnybrooke Park have to be precisely on point - coming in exactly 90s after the previous train left the station, with maybe a tolerance of +5 to 10s, if you choose to modulate the speed at which the train travels between Sunnybrooke Park and Laird. That is still something that requires a significant amount of precision in how you run the trains that would require some insane super TSP - something that wouldn't be feasible at 3 minute headways. Again, according to sound transit, an at grade LRT alignment has a headway limit of every 5 minutes, so okay, let's move the requirement and have even more trains short turn at Laird. First, in order to get above that 5 minute minimum, you can only send 1/4 trains beyond laird, so immediately you're creating a situation where you're going to have people just standing on the platforms waiting for the right 1/4 train to arrive which... isn't great. Second, you still have to assume that the total amount of disruption that occurs to a train doesn't exceed a time loss of more than 15s - which is almost impossible to assure when you have a train running out in the open that has to deal with idiot pedestrians and drivers. You're going to have cars crossing the street during traffic that drove ahead and is now stuck in the middle of the road, pedestrians that decided its a genius move to jaywalk while there is an oncoming train, old ladies that are taking their sweet time crossing to reach the station even after the signal has turned red, pedestrians that decided to walk along the track for some inexplicable reason (believe it or not this isn't that uncommon of a thing), and much more. When you have an open system like this, guaranteeing precision at a specific part of the line is basically impossible. You're going to have trains that are late or early by at least 30s on the regular - and this is enough to completely break any notion of 90s. Something like 2-2.5 minute headways might be possible, in fact I'd wager on seeing at most 2.5 minutes before major upgrades are done in the eastern section, but certainly not 90s.

 

[afransen]

There is no room on the surface for rail transit where the city decided to put it underground,

Huh? All the way to Mt Dennis?

 

[DirectionNorth]

But my point is that we are not going to be at that state to require 1.5 minute headways until decades in the future. At the current state, at 6,000 pphpd for 2031 we will need an LRT train once every 5 minutes in the morning peak.

10 years is "long-term planning" is so Toronto ...

When we get to 3 minute or lower headways in 20-30 years, hopefully, we have more level-headed folks in charge who will allow the implementation of full TSP to allow for a much better at-grade section.

Cross streets need to get green lights at some point, if only to let buses pass. Full TSP at 3 minute headways will never work, because people are stupid and drive their cars onto the tracks, pedestrians jaywalking, other traffic that actually needs to cross Eglinton ...

If you choose to ignore the published city documentation stating the expected ridership then I can't really help that. Nothing I say will change what you've decided in your mind.

City documents go to 2031. Hardly planning for the Toronto of 2040, or 2050, or 2100.

If your reasoning for stating that the city's estimates for ridership is too low because other transit lines had the ridership exceed expectation, that is anecdotal evidence. I can bring up 100s of proposals from all over the world where they have to showcase ridership estimates 10x the actual just to get shovels in the ground. These numbers will only be met 20 years after the subway lines open. Again, anecdotal evidence does nothing here.
If you don't want to go far, then see what happened with the Union Pearson link, reducing price from $25/$19 to $12/$9 (I think, I haven't paid attention) in order to increase ridership.

We're not talking about the rest of the world. It's great to talk about what the rest of the world gets right, but reality exists, and ridership projections are made by the CITY OF TORONTO and not Osaka, or Los Angeles, or Charleroi.

You have no evidence of overstating ridership projections, and I have no evidence of overstating projections. Let's agree to disagree here. Either way, my opinion is based on my longer-term (than the 2031 projections) ridership guesses that we're debating.

 

[fanoftoronto]

I was specifically addressing the points he was making. While yes I do know that trains will short turn at Laird, the question is still can you run trains every 90s in the central section, and the possibility is dubious at best. The trains coming in from Sunnybrooke Park have to be precisely on point - coming in exactly 90s after the previous train left the station, with maybe a tolerance of +5 to 10s, if you choose to modulate the speed at which the train travels between Sunnybrooke Park and Laird. That is still something that requires a significant amount of precision in how you run the trains that would require some insane super TSP - something that wouldn't be feasible at 3 minute headways. Again, according to sound transit, an at grade LRT alignment has a headway limit of every 5 minutes, so okay, let's move the requirement and have even more trains short turn at Laird. First, in order to get above that 5 minute minimum, you can only send 1/4 trains beyond laird, so immediately you're creating a situation where you're going to have people just standing on the platforms waiting for the right 1/4 train to arrive which... isn't great. Second, you still have to assume that the total amount of disruption that occurs to a train doesn't exceed a time loss of more than 15s - which is almost impossible to assure when you have a train running out in the open that has to deal with idiot pedestrians and drivers. You're going to have cars crossing the street during traffic that drove ahead and is now stuck in the middle of the road, pedestrians that decided its a genius move to jaywalk while there is an oncoming train, old ladies that are taking their sweet time crossing to reach the station even after the signal has turned red, pedestrians that decided to walk along the track for some inexplicable reason (believe it or not this isn't that uncommon of a thing), and much more. When you have an open system like this, guaranteeing precision at a specific part of the line is basically impossible. You're going to have trains that are late or early by at least 30s on the regular - and this is enough to completely break any notion of 90s. Something like 2-2.5 minute headways might be possible, in fact I'd wager on seeing at most 2.5 minutes before major upgrades are done in the eastern section, but certainly not 90s.

So 2.5 minutes or 2 minutes gives a pphpd of 12,000 to 15,000. Much higher than the 2031 estimate of 6,000 pphpd. Sufficient for a good long time. Maybe 2041 or 2051, but we don't have that information.

 

[ARG1]

So 2.5 minutes or 2 minutes gives a pphpd of 12,000 to 15,000. Much higher than the 2031 estimate of 6,000 pphpd. Sufficient for a good long time. Maybe 2041 or 2051, but we don't have that information.

That was never in contention. You have a serious problem with reading what's actually said - the point that was argued that you can build a light metro that is faster, more frequent, and more reliable, for a similar price and a similar capacity. You have just said yourself that the capacity is indeed similar.

 

[Voltz]

Huh? All the way to Mt Dennis?

To just past Keele, before that it is tight.

 

[TRONto]

You know, its very difficult to argue with someone who doesn't read what he's trying to argue against. Virtually every single point you make in your post I literally addressed in what you're replying.

Yes, and since its a downtown core and thus "the destination", its totally acceptable. Plus I do also think that the at grade section downtown is the weakest part of the C-Train, so I don't know why you're using it as a crutch.

And?

Yes, they put an LRT line in the middle of the parkway. It allowed them to save money by not building massive elevated viaducts, while still providing extremely rapid service.

Let me uh... give you the opportunity to reread what I said... this time CAREFULLY. Let me bold the areas of importance


You know that there is a middle ground between good and bad right? Every project has some good elements to it, but also some bad elements. Line 2 overall is a great line with many good things going for it. However, one of its major drawbacks is that it has very tight stop spacing - and a lot of stations that probably shouldn't exist that overall decrease the speed of the line making it less efficient as transport.

In fact, same with the C-Train. It is by no means a perfect system - there are a lot of odd quirks with it that absolutely bring it down. When I brought up the C-Train, it was meant to be an example of how you do suburban LRT in a way that makes sense for its environment - it wasn't an endorsement that C-Train is the jesus of all LRTs and we should bow down to its might will.

Unfortunately with you it seems like something is either awful, or its fantastic. If someone praises something, that means that this person thinks that thing can do no wrong. Likewise if that person thinks something is flawed, then its an unsalvagable trash heap.

You do realize that 28km/h encompasses the entire stretch from Science Center to Kennedy right?

Its going to be something like 35-40km/h in the tunneled section, and something like 20km/h on the surface section. 28km/h is simply the average speed over the entire line.

Yes. Fun fact, most people in these suburban areas do not live next to the street - and have to walk like 10-15 minutes to reach the street. Adding an extra minute of travel time isn't going to do much. As for development, well that's why you focus development around where the stations are built.

And you can always run complementary bus service, like we currently do on Line 1, and like we are going to do for the entirety of Line 5 (newsflash, Line 5 won't be replacing busses anymore).

Ok? Factoring all of that in, that gives you a travel time that is roughly the same as LRT given my generous estimates, although once again your points are very circumstantial. Fun fact, most of Line 1's ridership comes from transfers from busses and other modes, not from walk in traffic. So if you're going to ride Line 5 after transferring from say a Warden bus (like most people who will be riding the line), then most of the timesaves here that you pointed out are irrelevant. Second, the benefits of Light Metro are much more than speed, its reliability.

Look at downtown toronto, and tell me how often you see stories of cars crashing into streetcars, idiots running red lights, drivers driving into the Queens Quay tunnel, etc. Now imagine that instead of a small local streetcar route, you are now dealing with a 40+km line that stretches from the Airport, all the way to UTSC and beyond. What this means is you now have a line where because someone decided to have a nap in front of the streetcar tracks, or again there was a collision at an intersection, some guy waiting at Martin Grove station (a fully underground part of the line) will now have a massive delay. I can keep going.

I went to the office on TTC today. First time since COVID started. Took streetcars and Line 1 +2. Only delays were on (drum roll please....), Line 1.

 

[44 North]

There is no limit - if you have no problem with bunching and inconsistent headways. If you want trams that are just constantly backed up and waiting for the tram in front of them to move on, you can run them as often as you want. No amount of TSP is going to help you here - you need to have a green light on cross traffic at some point, which means if you're running at such tight headways, you are guaranteed to feel the full blunt of some red lights. A quick rule of TSP is the less trains you run, the more room you have to have proper TSP that is likely to let trains through. The rule of thumb according to Sound Transit is that if you want consistent headways on an LRT, you can run a train at most, every 5 minutes.

Getting trains to run consistently every 90s isn't easy, you have to have very precise scheduling, and you have to minimize how many disruptions you have. For that you need to have a line that is basically fully automated. If you want to integrate a manual surface section into a 90s tunneled section, you have to find a way of guaranteeing that the train will enter the tunnel at every exact 90s interval - which is physically impossible. Worst case scenerio, you have a deluge of trains back to back to back, and you just have people waiting for up to 5 minutes until they get passed the first tunnel station.

So no, if you want to have an actually USABLE line, you simply cannot run 90s frequencies on the Eglinton Line. Period. End of Story. If we want to reach your theoretical 20 or 30k pphpd that we can reach with 1.5s headways, the only way to get that is to grade separate the entire line.

Hold on a sec. Weren't you arguing that lines like Crosstown should be like Calgary - with a central surface street-running portion in the highestest-density area? Also where the lanes are most reduced, pedestrian volumes are the highest, intersections are very close together. How well would scheduling and frequencies work with such a backward design.

 

[ARG1]

Hold on a sec. Weren't you arguing that lines like Crosstown should be like Calgary - with a central surface street-running portion in the highestest-density area? Also where the lanes are most reduced, pedestrian volumes are the highest, intersections are very close together. How well would scheduling and frequencies work with such a backward design.

I'm not going to repeat myself, please go back 2 pages and reread what I said.

 

[ARG1]

I went to the office on TTC today. First time since COVID started. Took streetcars and Line 1 +2. Only delays were on (drum roll please....), Line 1.

Congrats on your anecdotes, I have mine as well. Last Friday I took Line 1 from Finch to King, then took the streetcar from King to Distillery. Line 1 had no issues, however It took 10 minutes for the streetcar to show up, while 2 streetcars went the other direction.

 

[TRONto]

Congrats on your anecdotes, I have mine as well. Last Friday I took Line 1 from Finch to King, then took the streetcar from King to Distillery. Line 1 had no issues, however It took 10 minutes for the streetcar to show up, while 2 streetcars went the other direction.

Thank you. Rudeness noted.

I was referring to "Look at downtown toronto, and tell me how often you see stories of cars crashing into streetcars, idiots running red lights, drivers driving into the Queens Quay tunnel, etc" the description from last Friday's anecdote refers to bad line management. Today's Line 1 incident was not a line management issue. Different things.

 

[ARG1]

Thank you. Rudeness noted.

I was referring to "Look at downtown toronto, and tell me how often you see stories of cars crashing into streetcars, idiots running red lights, drivers driving into the Queens Quay tunnel, etc" the description from last Friday's anecdote refers to bad line management. Today's Line 1 incident was not a line management issue. Different things.

Hypocrisy noted, although how does the 504 refer to bad line management? Cars crashing into barriers, running into Queens Quay, its no more bad line management than someone jumping onto the tracks, running into the tunnels, or holding the doors open on Line 1...

 

[syn]

People don't use line 2 end to end because it is slow. There new plan seems to be to loop Ontario line up and over the 401 to Richmond Hill, the airport, and on to MCC maybe. Honestly would have been more logical to extend line 5 but the choice of technology makes it unreasonable. Lots of people would use a fast connection from MCC or STC to line 1 or Ontario Line or the airport.

We seem to be stubbornly refusing to build transit that creates network effects.

Which is exactly why they wouldn't use this line to travel from Scarborough to Mississauga. That's not what it was ever intended for.

It's a local Toronto line, very much like Bloor-Danforth.

 

[fanoftoronto]

That was never in contention. You have a serious problem with reading what's actually said - the point that was argued that you can build a light metro that is faster, more frequent, and more reliable, for a similar price and a similar capacity. You have just said yourself that the capacity is indeed similar.

You stated that you can save money on the stations by building a 40 or 50m long station and elevate the eastern portion with the savings. I asked for a source on how you can say this with confidence. I asked you direct questions on cost estimates. How much would you save by shortening the station? How much would it cost to elevate the eastern portion? No answer from you.

You stated that there's enough demand for 20,000 pphpd on Eglinton. I asked for a source. You said anecdotally "the congestion on the 401" and provided no actual source or proof. No real answer or source from you.

You stated a Canada Line style system has higher capacity. I've already proved that Canada Line has an upper limit of 13,360 pphpd as the trains can hold 334 people and max of 1.5 minute headways. This is lower than the LRT by a long shot. If your imaginary number of 20,000pphpd actually shows up then your Canada Line train is packed beyond capacity from day 1. Capacity is higher on the LRT Crosstown at 2 minute headways at 15,000 pphpd. 15,000 >13,360. Not to mention push comes to shove, we can make use of the 90m stations and add a 3rd car to each train effectively increasing pphpd to 22,500, which is far higher than the 13,360 limit on the Canada Line. Also, how would you increase capacity on a Canada Line style system? You've locked yourself into a 13,360 pphpd maximum limit with zero opportunity to expand without shutting down the system. You tried to insult my ability read but no other reply from you.

You stated that Canada Line trains will be faster. For reference the Canada Line from Richmond–Brighouse station to Waterfront station takes 26 minutes, 12 stations, and is 14.5 km long. Average speed is 33.5 km/h, a far cry from the 40 km/h average speed you're touting. The LRT in the underground section is stated to go at 32km/h while following all design norms of the Canada Line. Fully grade separated, full ATC, 1km minimum station spacing. The time saving would be 4 minutes by grade separating the Eastern portion and making the average speed 32 km/h there as well. I asked for a source on how a light metro would be 40 km/h when the LRT is achieving only 32km/h and the Canada Line is achieving only 33.5 km/h. No reply from you or a source for the increased speed.

I've stated that when they changed the western portion from at-grade to grade separated as you wanted, the ridership numbers went down! I argued that the same could happen with the eastern portion as well. You said you don't trust the city numbers while providing no sources or numbers to back up your claim.

I'm walking into this conversation with research and evidence and sources. You're walking in with anecdotes and gut feelings. Show me the numbers to back up your claim and I'll be happy to read through them.

Edit: Edited for typos and added the Canada Line average speed and reworded that paragraph. Added a note in the capacity paragraph to ask how to expand capacity on a Canada Line style system when demand exceeds 13,360 pphpd and added a question on how to expand capacity on Canada Line style system. Replaced all 13,000 pphpd to the more accurate 13,360 pphpd.

 

[fanoftoronto]

10 years is "long-term planning" is so Toronto ...

Cross streets need to get green lights at some point, if only to let buses pass. Full TSP at 3 minute headways will never work, because people are stupid and drive their cars onto the tracks, pedestrians jaywalking, other traffic that actually needs to cross Eglinton ...

City documents go to 2031. Hardly planning for the Toronto of 2040, or 2050, or 2100.

We're not talking about the rest of the world. It's great to talk about what the rest of the world gets right, but reality exists, and ridership projections are made by the CITY OF TORONTO and not Osaka, or Los Angeles, or Charleroi.

You have no evidence of overstating ridership projections, and I have no evidence of overstating projections. Let's agree to disagree here. Either way, my opinion is based on my longer-term (than the 2031 projections) ridership guesses that we're debating.

2031 is what the city documentation goes up until.

Though the capacity limits for the Crosstown is 15,000 pphpd, so even with a modest 500 yearly increase from 6,000, we're good for 18 years before we need to do anything to cope with the added demand.