Canada Line races toward capacity

[Migrant_Coconut]

Originally Posted by Reecemartin

Adding a station at Capstan way is going to increase people on the system not reduce it, and the current Expo Line capacity is supposed to be somewhere around 16k, which is far higher than the likely max of the Cl at about 13k (maybe 14k if we improve flow in key stations).

A) More stations means more trainsets, in order to retain speed and frequency.

B) The capacity may be 16k, but actual ridership is, again, 13,250. Before anybody forgets, that's all the way from Waterfront to Surrey, including Commercial-Broadway and Metrotown and New West - so roughly what we can expect once 33rd/Oakridge/57th/YVR/Richmond et all are built out - and the most crowded it gets is 13k. Well under the "14k" ceiling.

C) Exactly where are you getting your numbers from? We're all quoting official TransLink/Vancouver/Surrey studies; you're contesting them with... hunches?

Originally Posted by Reecemartin

Saying longitudinal seating wastes space is probably the most off kilter comment I've seen in a long long time... I don't think I've ever been in ANY packed train with tons of (or really any) open seats even in cities that have longitudinal seating. Typically the seat is filled just as soon as someone gets up. To the contrary seats like on the Canada Line lead to FAR more waste.

You've got anecdotes, I've got anecdotes: when I take the Canada at rush hour, all the seats are full. Whereas on the bus, the older models with longitudinal seats usually have a passageways of standing riders and one or two empty seats. People just don't like to sit next to other people - much less between two people.

Personally, I like the New York arrangement best: 2-3 seat benches, then a front/back block, then another bench.

Originally Posted by Reecemartin

And with regard to Arbutus while they may very well be planning it, iirc even though its on the row, its going to be a streetcar running right next to the pedestrian greenway. It will come nowhere close to the Surrey LRT in capacity (fully segregated and given signal priority) if its not grade separated. Last time I checked Skytrain on Arbutus was not happening.

You seem to have completely missed the point behind grade separation - to limit track intrusions on routes with many points of conflict, like the middle of a busy street. BUT if the corridor is already segregated so that most track intrusions can't happen, then by all means, save the money!

I know you were present for this discussion, so I'll make this quick: the Arbutus Greenway, being part of the former interurban route, is by definition already fully segregated. Not to mention the planners are willing to grade-separate the major crossings, close the minor ones, and put lights and barriers on the rest. So no need for signals or grade-separation; all it needs in order to out-LRT Surrey's actual LRTs... is a simple shoulder-height fence.

Originally Posted by Reecemartin

And of course Surrey should have a large rt system, the LRT will certainly not be the last thing that goes into the core if the growth is like was mentioned. AND if traffic etc is as bad as some predict then the system will very likely get some grade separation.

Well, why can't they grade separate it to begin with (besides Hepner and friends wanting to have their cake and eat it too)? Come on, if the Canada is "underbuilt," then Newton-Guildford and Langley are "BOTTOM-built!"

Nname beat me to it - Surrey's not going to stay tiny forever. If it grows into an urban core like its City Council hopes for, it needs an urban metro system to match. Can't dream big and plan small.

And Metro's got a point: you seem to undersell the Canada Line and oversell the LRT plan. I'm not saying "double standard" just yet, but it seems odd.

[Sheba]

Originally Posted by Migrant_Coconut

You've got anecdotes, I've got anecdotes: when I take the Canada at rush hour, all the seats are full. Whereas on the bus, the older models with longitudinal seats usually have a passageways of standing riders and one or two empty seats. People just don't like to sit next to other people - much less between two people.

TransLink also heard that people here don't like longitudinal seats - that's why all the newer (aka after Mark 1) trains don't have them.

[Migrant_Coconut]

^ Thanks - I figured as much, but always nice to have proof.

[idunno]

I had thought that the end-to-end trip time for the Canada Line was to remain stable, even with the addition of 33rd, 57th, and Capstan Way. As in, the operating speeds right now are effectively slowed down in certain sections to make up for the future increase in stations.

So Waterfront -> Brighouse will remain 26 minutes end-to-end even after Capstan Way is built.

[Reecemartin]

[Deleted]

[Reecemartin]

[Deleted]

[Reecemartin]

[Deleted]

[deasine]

Originally Posted by Reecemartin

With respect to Longitudinal seating, sure its less comfortable that's a common claim. But being sardined in, or having to wholesale remove seats just to reduce crowding is really not saying much for the current arrangement IMO. And again most systems with significant crowding do use bench/ longitudinal seating so do with that what you will. I know I for one would much rather be squished and have a seat than be squished and standing.

In the latest design of the MK II.5 and III, it was determined longitudinal seating did not necessarily give the capacity for the reasons stated earlier: in a 1-1 seating layout, one must consider the knees as well as the slight tilt in the seat itself, which brings it up to 1.5-1.5, which is then no different than a 2-1 seating configuration.

Admittedly, the Canada Line's wider carriages at 3m may give a bit more room, but as with many metros, passengers form generally maximum of 2 rows of standees. Living in London and a frequent commuter of the London Overground, despite there being room for 3 rows of standees done, I only see 2 rows no matter how packed it is. And trust me, the Canada Line is no where close to being as packed as some of the London Overground lines (not that this is a good benchmark to compare against).

Originally Posted by Reecemartin

IIRC the tail tracks at Waterfront are not long enough to hold a full train, otherwise they wouldn't have one sitting on the other platform at all times.
(
I'm not sure about Brighouse, but looking on Google Maps it doesn't look to me like they could hold a train on it, I could be wrong though.

Nonetheless not having tail tracks on the main terminus is the kind of poor planning that makes me annoyed with the Canada Line.

There seems to be some misinformation here - Waterfront has been built to its full 50m length already (as with Vancouver City Centre and Yaletown-Roundhouse). The standby train has nothing to do with platform lengths as trains regularly alternate between the two platforms. During the Olympics, both platforms were also in use.

Achieving a higher frequency will obtain a higher capacity throughput as some has already pointed out. Bearing in mind that the highest potential capacity on the Canada Line is assumed to run at 30 trains per hour (2 minute frequencies) when the signaling infrastructure on the Canada Line can realistically achieve 40 trains per hour (90 second frequencies) in the combined track sections, there is good evidence to suggest a capacity even higher than 15,000 pphd in the combined sections. As you rightfully have pointed out, yes certain stations will struggle to handle this, e.g. Vancouver City Centre and Broadway-City Hall. I think the biggest flaw of the Canada Line is under estimating the utilisation of some of the busy stations not designing for egress at a 85% load factor.

As for your point on challenges in reaching potential capacity, it is common for systems to begin to struggle when average capacity reaches 85% of potential capacity. The London Underground, Docklands Light Railway (perhaps one of the closest comparisons of Vancouver actually), Hong Kong MTR currently, etc. The challenges in station throughput on the London Underground are immense compared to the Canada Line, and even there the majority of its investments are focused on additional lines to relieve the existing London Underground network (e.g. Crossrail).

Unfortunately, what will restrict your aspirations in expanding the Canada Line's train length beyond 50m is the location of the switches in proximity to the station itself. For instance at Waterfront, the crossover is right next to the station and trains must be short enough to fully clear the switch. But unlike the Expo Line, there are plenty of other parallel corridors to offset the Canada Line (e.g. Arbutus). Even early in the ALRT expansion plans from the GVRD identified Arbutus as another potential corridor. If the Arbutus Corridor was built exactly like the Canada Line, a 30,000 pphd is quite high for Vancouver.


Source: GVRD

I do not think anyone in this forum doubts that the Canada Line could have been designed with higher capacity, especially as it has achieved massive success. Passenger forecasts are challenging and with a private partner, it is especially important to not be overly optimistic (consequences of this can be found right now in UK's East Coast rail franchise, which yet again will need to be subsidized and bailed out by the government). And of course with such unexpected success leads to unanticipated operational challenges, which is what we are all seeing on the Canada Line today. The best course of action we can take today is see how we can better manage our (very good) challenges and how we can better learn from the lessons of the Canada Line, and I hope all of us on this forum take this on in our discussions to come.

[Colin4567]

Originally Posted by Reecemartin

And with regard to the Surrey LRT, not only is it a completely different project but, there's a seriously difference in the ability to EXPAND capacity. The Surrey LRT is planned to start with two car LRT trains ....

In the case of the Surrey LRT there is no real barrier to making train at least 4 cars long in the future (just the cost of slightly lengthening the inexpensive platform)

It's only planned to start with 1 car - do you have insider information saying otherwise?

Secondly, as far as I know the distance between intersections is a factor in determining maximum car length. What's the minimum distance between intersections on the route?

[nname]

Originally Posted by Reecemartin

IIRC the tail tracks at Waterfront are not long enough to hold a full train, otherwise they wouldn't have one sitting on the other platform at all times.

I'm not sure about Brighouse, but looking on Google Maps it doesn't look to me like they could hold a train on it, I could be wrong though.

Nonetheless not having tail tracks on the main terminus is the kind of poor planning that makes me annoyed with the Canada Line.

All tail tracks are long enough, but why travel the extra distance to/from the tail track if there's an empty platform available?

Originally Posted by Colin4567

It's only planned to start with 1 car - do you have insider information saying otherwise?

Secondly, as far as I know the distance between intersections is a factor in determining maximum car length. What's the minimum distance between intersections on the route?

Intersection doesn't really matter. If they have to, they can close any intersection they want to extend the platform as long as the track is going straight.

The main constraint is probably at Surrey Central though. From the diagram, the current plan is to have a 60m platform. It can be extended to at most 100-120m before both end reaches a curve. However, by doing so, they would need to move and relocate the entrance and part of the bus loop.

[Migrant_Coconut]

Deasine raises a good point: if 15k assumes trains will run every 120 seconds instead of 90, then the Canada could actually go as high as 20k. Even then, that’s 340 people/minute on average, which is perfectly doable for most stations after expanding the platforms and adding more faregates and another elevator.

Originally Posted by Reecemartin

And with regard to the Surrey LRT, not only is it a completely different project but, there's a seriously difference in the ability to EXPAND capacity. The Surrey LRT is planned to start with two car LRT trains which have a similar capacity to the Canada Lines trains, however it would be much easier to go and add a third car in the future (or expand capacity in general) for any surface level transit (LRT, BRT what have you). The reason the Canada Line is so appalling to me and some others I suppose is that it was underbuilt with the designers well knowing that expanding capacity beyond 15k per hour would be totally impossible. In the case of the Surrey LRT there is no real barrier to making train at least 4 cars long in the future (just the cost of slightly lengthening the inexpensive platform) at which point even at a 5 minute frequency with no grade separation you can get to nearly 11,000 ppdph (900 person train * 12 per hour). You mention the second round of station improvements obviously needed in any case to increase the capacity of either line but, seem to ignore the point that those improvements will cost far far more on the Canada Line because of poor planning from the outset.

ONE car, expanded to two cars; 120 per car x 20 trains/hr x 85% efficiency = 2.4k and 4.1k respectively. That’s the LRT boosters’ own numbers.
Keep in mind that four cars is about 100m, and would require a second upgrade (two to four). Assuming Surrey’s willing to permanently close minor intersections - making traffic even worse - to accommodate the trains and platforms, it still only only gets you 8.2k due to speed/frequency limits imposed by traffic constraints and lack of drivers; with SkyTrain, you just buy more trains.

If the Canada Line won’t reach 15-20k in practice, then likewise, Surrey’s only going to get 85% of that super-optimistic 11k figure... more like 9k. That won’t cut it for a downtown core, even a secondary one.

At the risk of sounding like a broken record, you get what you paid for with transit. Low cost means low returns – and not even that “low” a cost, because the same money could get you a region-wide grid of BRTs which would provide the exact same usefulness on many corridors, or a SkyTrain that’s twice as useful on that one corridor.

Originally Posted by Reecemartin

The Arbutus Greenway indeed allows trains to run mostly separated from cars but, my point is if they are running right next to pedestrians it isn't really any better, because they still won't be able to run trains all that fast.

One word: "fence" - pretty sure I said it last post. And since there’s no cross-traffic or signalling to worry about, the only limit to capacity is really how many trains and drivers they can afford.

Originally Posted by Reecemartin

And being critical of the Canada Line doesn't mean I need to be just as critical of the Surrey LRT. The Canada Line is running and built right now for one, and while of course Surrey will grow, I think nobody would deny that it will be a long time til it has the type of density and throughput of Vancouver.

Like how it’ll be a long time before the Canada gets fully developed? Broadway to Marine Drive is still mostly suburbia, Richmond's still a work in progress, and YVR is still thinking about how to expand. Perhaps consider that you have an unconscious bias...

Originally Posted by Reecemartin

I know I for one would much rather be squished and have a seat than be squished and standing.

Depends on how much a person’s culture emphasizes personal space. Most Asians are used to being crowded in on a bus/metro and will readily take the last two square feet, but many Caucasians will decide “nah, I’ll wait 5-20 minutes for the next one.”

Same principle behind seats – “am I tired enough to want to intrude on him/her? I don’t like it when people do it to me, after all.”

[dleung]

I don't know why people are surprised by the Canada line ridership. The initial projections are far too low. The corridor is as urban as the Expo line, and far more urban than the M-Line. Richmond has been built for rapid transit for a long time - they should've designed for New Westminster or Metrotown volumes. Other than Lougheed, the next busiest station on the M-Line (Brentwood) has less traffic than all but 4 of the stations on the Canada Line.

https://www.translink.ca/-/media/Documen...-TSPR-Appendix-E-Rail-Line-Summaries.pdf

[nname]

Originally Posted by dleung

I don't know why people are surprised by the Canada line ridership. The initial projections are far too low.

The initial projection was bang on. It's just that most people, including may in this forum, dismiss it for being way too high.

If you still have the RTM document produced in 2007, the projection was:

2010 - 116,300
2014 - 126,100
2021 - 143,100

The 2021 projection is probably a bit low though

[WarrenC12]

Originally Posted by nname

The initial projection was bang on. It's just that most people, including may in this forum, dismiss it for being way too high.

If you still have the RTM document produced in 2007, the projection was:

2010 - 116,300
2014 - 126,100
2021 - 143,100

The 2021 projection is probably a bit low though

True, and we have been going through an economic and real estate boom since shortly after it opened, so the stars have aligned in that sense.

[Reecemartin]

[Deleted]

[Reecemartin]

[Deleted]

[Migrant_Coconut]

Originally Posted by Reecemartin

In terms of LRT your numbers are way off, as we discussed not so long ago, the operating assumption for the line is typical 30m LRV's (to address Colins point I was referring to the built in 2 car capacity, I also don't think any intersections would make 4 cars not viable, correct me if Im wrong) nonetheless that means a 4 car train would be 120m and each LRV is rated for about 250 passengers so your estimate is only about half of the true capacity.

The maximum passenger capacities, in the standard seating layouts, are 135 and 251, for the three and five-module configurations respectively. When run in the five-module configuration, with train-sets of four vehicles, the maximum capacity of a light rail line is 30,000 passengers per peak hour. (fyi 5 module is the 30m one, and the 30,000 ppdph figure would be fully grade segregated using atc similar to skytrain but still the capacity is much higher than what you suggested)

No. Nada. Non. Nein. Nyet. Nee. Nahi. These are from TransLink and Surrey's own studies. 120 people per car. 1-2 cars per trainset. 1 trainset every 3-5 minutes, x 85% efficiency. That gets you 2,040 pphpd for one car, 4,080 for two cars; by extension, four cars is 9,600 theoretical and 8,160 "actual."

By continuing to say those numbers are wrong, you again insist that the people responsible for studying, designing and promoting Surrey's LRTs are deliberately downplaying and underselling their own pet project for some unknown reason. Feel free to arrange a meeting or attend an open house with TransLink and/or Surrey's engineers, and let them know that their years of education and experience in city/transit planning are worthless, since you can do their job better than they can just by looking at other cities' numbers off the internet. They may have missed something, but definitely not something as big as capacity.

As for grade-separated LRT... well, that's basically what an ALRT is. Since it's going to cost SkyTrain levels of money to disrupt service and bury the LRT - meaning that Metro Van's basically paying the capital cost twice (once for construction, twice for tunneling or cut-n-cover) - it's incredibly penny wise, pound foolish to not just build a SkyTrain or buried LRT from the get-go.

Originally Posted by Reecemartin

Wrt the Canada Line the development is definitely greater and the density higher than the L. Of course some major centres like SFU Guildford and the Surrey Memorial Hospital are connected to the L, CL has all of Richmond, an International Airport, several fairly major malls, several significant planned redevelopments and downtown Vancouver + an eventual link to the Broadway Subway.

Again, you're downplaying Surrey's growth potential and hyping Cambie's.

One's got downtown Van, Cambie, Oakridge, Richmond, YVR, and a future connection to the Millennium. The other's got Newton, Whalley, Guildford, probably White Rock and Coquitlam, and a future connection to the Expo. Po-tay-to, po-tah-to.

Originally Posted by Reecemartin

And again sustaining 90s headways is going to be a pretty major challenge. If trains are crowded that headway is going to decay very quickly with longer dwell times. Just lengthening the platform doesn't do much to help dwell times and adding faregates is quite limited, in order to increase capacity we will need to add more entrances and exits and that is not as simple as adding one or two faregates per station (though potentially adding faster narrower faregates could have a limited effect on improving throughput).

Or just keep dwell times the same and have latecomers get the next train - Jiminy H. Cricket, it's only a minute and a half.

Another entrance/exit for major stations would be good. On the other hand, crowding at the faregates is a fairly common occurrence no matter where you go.

Originally Posted by Reecemartin

This is a pretty silly point, I really don't think your sentiment that Asian people are more likely to be comfortable on a crowded bus or train is in the least based on any evidence. Sure Asian transit systems are known to be crowded but are you really suggesting you can generally just expect one race is more accepting of crowding?

It's a general trend based off observation, used as an example of cultural differences - like how Americans hoot and holler for a good movie, but Japanese will be deathly quiet out of respect.

Obviously, not every Asian is going to squeeze in like a sardine every time, and I never said they'd be "comfortable" with it, but since the population is big and the city/country small, less room is a fact of life. Nor will every Caucasian wait patiently for the next train, but they're used to a lot more personal space and room to move around; extreme example - Shaughnessy residents recently compared highrises to slave ships. Saying that you like to sit rather than stand is a mix of personal preferences and a few cultural ones.

Originally Posted by Reecemartin

With respect to seats this seems pretty anecdotal, I remember the TTC was considering removing seats in the past to free up some space and they had a study that showed that seats were generally 90+% filled before a vehicle started to have significant numbers of standees (i.e. if there is an open seat people usually take it). In my experience this isn't really any different from Metro Van.

Yup, six of one, half-dozen of another - I prefer Chicago seating, but I don't think either arrangement is significantly better at packing a train.

However, you DID say this:

Originally Posted by Reecemartin

To that point part of the reason the Canada Line feels so full is the hilariously poor planning that has seats all facing forward, wasting tons of space and creating a narrow aisle.

Neither poor planning, nor a waste of space, is what I'm saying.

[aberdeen5698]

Originally Posted by Migrant_Coconut

a general trend based off observation, used as an example of cultural differences - like how Americans hoot and holler for a good movie, but Japanese will be deathly quiet out of respect.

Different cultures absolutely have different tolerances for personal space. Watch an Italian talking to an Englishman in an open setting and the Italian will be advancing while the Englishman backs away from him.

[Reecemartin]

[Deleted]

[Migrant_Coconut]

Originally Posted by Reecemartin

If Translink listed the capacity as 120 per car then I guess they are as bad as some of the more extreme people has said, thats completely false.

Because of course you're more of an expert than the actual experts.

Fair enough though, looks like I remembered wrong: 200 passengers/vehicle x 12 cars/hour x 85%. Not 250, because neither Bombardier nor Siemens nor Alstom has 30m vehicles that can bend dimensional time and space.

That brings up another problem - traffic constraints apparently limit frequency to every five minutes, not every three. The "ultimate" capacity remains the same: two cars get you 4,080, and so four cars still only get you 8,160... again, higher frequency/capacity requires a second round of capital costs, making each LRT line just as expensive as the SkyTrain option (if not more).

Originally Posted by Reecemartin

And I think it's pretty clear that Cambie has more potential (certainly in the next 20 years) than Surrey, thats not some hidden bias against Surrey.

Meaning that Surrey's development plans (displayed above) aren't going to happen any time soon, and therefore there isn't any urgency or even a reason for LRT in the first place? Can't have the argument both ways.