ION - Waterloo Region's Light Rail Transit

[ijmorlan]

I just remembered I wanted to post an observation. I have noticed in recent weeks that southbound LRTs slow to a crawl roughly in front of the Perimeter Institute. Not just slowing a bit ahead of the Erb/Caroline intersection, but slowing down to probably a slow jogging pace way back from the intersection. Anyway, I was on the LRT yesterday, and it seems the slowing is ahead of the facing-points freight crossover. We actually sped up slightly once past the switch before arriving at the Erb/Caroline intersection.

Now I need to watch to see if northbound trains slow ahead of the switch, although I don’t recall seeing that. Has anybody else observed crawling ahead of facing-points switches elsewhere? I believe the only other locations with facing-points switches would be the double crossovers at the ends of the line and a couple near the OMSF.

I hope this is a temporary condition related to some issue with that particular switch. That being said, it seems weird — the switch only moves for freight traffic, so it shouldn’t be moving at all between the freight train leaving in the early morning and the next one coming late in the evening. So they ought to be able to verify that it is positioned properly and then lock it in that position for the day.

[JoeKW]

I don't think the system, as designed, can accommodate 5 minute headways. Unfortunately, I don't have any details but I remember reading it... probably here.

[ijmorlan]

I don't think the system, as designed, can accommodate 5 minute headways.  Unfortunately, I don't have any details but I remember reading it... probably here.

I thought 5 minutes was the design headway. 5 minutes is a long time — it shouldn’t be a problem. If you think of how far away a train is that is 5 minutes ahead and compare that to things like stopping distance, it should be OK. On the road segments, if buses can follow each other so closely that they meet at traffic lights and stops then so can LRTs. Personally I think the system itself should be able to handle 3 minute service, although at that point the benefits of saving operator salaries by running in pairs become hard to ignore and certain crossings would be engaged too much of the time under current practices.

[timc]

The thing about five minute headways is that it starts to get nasty for traffic at crossings. At a five minute headway, there is an average of 2.5 minutes between trains.

[danbrotherston]

The thing about five minute headways is that it starts to get nasty for traffic at crossings. At a five minute headway, there is an average of 2.5 minutes between trains.

I always hear this, even here, and I don't get it.  If we have the ridership to justify a rain every 5 minutes (or every 3 minutes)...that's 250 people per train, then that justifies keeping those crossings down...if it was an intersecting road, with cars on it, we would keep the traffic signals green for far longer to clear the traffic jam...and there wouldn't even be discussion about that, but somehow because it's people on a train, that's a reason to reduce service on the intersecting road?

[JoeKW]

I think it's more that the crossing became barely functional and you need to start looking at shutting them down.

[ijmorlan]

The thing about five minute headways is that it starts to get nasty for traffic at crossings. At a five minute headway, there is an average of 2.5 minutes between trains.

I always hear this, even here, and I don't get it.  If we have the ridership to justify a rain every 5 minutes (or every 3 minutes)...that's 250 people per train, then that justifies keeping those crossings down...if it was an intersecting road, with cars on it, we would keep the traffic signals green for far longer to clear the traffic jam...and there wouldn't even be discussion about that, but somehow because it's people on a train, that's a reason to reduce service on the intersecting road?

I have a few comments, but no firm conclusion.

It’s no problem to have one bus per cycle of the traffic lights. I believe this is typically around 2 minutes or so. Even more is possible, but especially if there is a stop at the intersection you can start to have problems. So based on this, there should not be a problem at regular intersections.

On the other hand, crossings work differently and some of them block traffic for much longer than necessary. For example, northbound trains leaving Seagram stop University traffic before the train even leaves. This isn’t so bad, especially since pedestrians can use the time, but it does mean the train takes more time from the crossing than it really needs. Much worse, however, northbound trains leaving the public square stop shut down Erb and Caroline before they even close their doors. On top of that, they’re closing down an intersection: nobody can move (well, in reality pedestrians can do some crossings of the intersection that don’t cross the tracks, but officially they can’t). In effect, it’s really the intersection of 3 routes: two roads and a train line, each of which needs its green time.

Note too that traffic lights go on their own schedule, which can be optimized but isn’t pre-empted by anything except emergency services. One LRT per 2 minutes in each direction could pretty much shut down a railway crossing permanently whereas one LRT per minute arriving at a traffic light in an on-road segment should be OK, just as it is OK for buses (but note the warning above about stopping), because the LRTs in each direction would have to wait and go together when their direction got its green.

The comment about the number of people on a train is very on-point: if the traffic being carried by the train line is similar to that being carried by the crossing street, it’s perfectly reasonable for the train to take half the time, even if that results in the street being backed up. It’s not the fault of the LRT that it’s efficient.

So overall I think that from the point of view of “snarling” (to use the word always used by LRT denialists) traffic, frequencies down to about every 5 minutes should be OK with the system as it is. Going above that would require at a minimum re-timing some of the crossings and possibly re-design of some areas.

Even without such an increase in frequency, it would be prudent to allow Caroline St. traffic to proceed when only the southbound LRT track is in use. There is no conflict between that track and Caroline St. other than turns onto westbound Erb St. At higher frequency however this improvement would be necessary unless road traffic is significantly reduced from how much we have now.

[jgsz]

Pedestrian fatally struck by LRT

[clasher]

Ugh, rode by train with the tarp on the side of it and had a sick feeling about it. Seemed pretty far from the Columbia crossing.

[bgb_ca]

Pedestrian fatally struck by LRT

https://www.therecord.com/news-story/979...ion-train/

The man was walking on the Ion tracks about 125 metres north of Columbia Street at about 4:50 a.m. when he was struck by the southbound train, say police.

So it sounds like someone made the poor choice to walk down the tracks instead of the trail next to the tracks, I'm guessing with headphones on, and by the time the driver saw him it was too late (it would have still been dark that early).

[jeffster]

Sad news. I feel for the mans family, but also the driver of the Ion. You don't ever want to be the first one to be involved in a fatality. Hopefully Grandlinq can offer some good counselling, as the driver will need this.

[ijmorlan]

So it sounds like someone made the poor choice to walk down the tracks instead of the trail next to the tracks, I'm guessing with headphones on, and by the time the driver saw him it was too late (it would have still been dark that early).[/size][/color]

I wonder if it’s more than that. In that location it’s not just a poor or questionable choice — it makes no sense at all. I looked at the location in Google Maps aerial view, and there is no conceivable trip that is more convenient or shorter by walking along the tracks. There is a path immediately parallel to the tracks on the west, and the parking lots on the east are all linked up. Conceivably I could imagine somebody trying to cut some time off by crossing the tracks, but that would involve climbing two fences and crossing two ditches (in addition to the tracks themselves) and would only save about 500m at the most, for a very specific pair of endpoints.

So I wonder about being suicidal or otherwise not of sound mind. I hope we’ll find out something. It’s quite clear to me that this is most likely not an LRT safety problem in any way, based on the limited information available.

[ijmorlan]

I wanted to add some basic analysis based on the published schedule. At 04:50, no stop service has taken place on a Saturday. The schedule shows 2 southbound trips starting at R&T park, with the first of these at 05:06. Looking at the Fairway schedule, it seems obvious that these become the 2nd and 3rd northbound trips of the day. Based on this, I believe the LRV was deadheading to Fairway to become the 1st northbound trip, leaving at 05:30.

[Spokes]

Oh man. So sad. Dammit

[ijmorlan]

I just noticed that the CTV item on this incident contains (or passes along) incorrect information:

https://kitchener.ctvnews.ca/pedestrian-...-1.4753011

It says that “Waterloo Regional Police say this is the first road fatality of the year in Waterloo Region”. This is not a road fatality — it is a railway fatality.

If it occurred on an on-road segment then there would actually be an interesting discussion to be had as to whether it would be a road fatality, but since it occurred many metres from a road on a non-road segment of track, it cannot reasonably be considered a road fatality.

Sad either way, but I would hope that the official statistics will be maintained correctly.