ION - Waterloo Region's Light Rail Transit

[ijmorlan]

3/3 - Saturday, August 12, 2017
As a reminder, these arrays will trigger a derail device further down if a freight train is too high or too wide to safely pass along the shared LRT corridor.

So is there a red light or something before the de-railer?  I only ask say, if someone put a bucket on a boxcar, and set off the detector, derailing a train seems extreme...

Coke

I don’t know about the north end, but the south end one has a signal right at the derailer. I’m not aware of any other signals but I haven’t gone down the line further to check. Normally that wouldn’t give time to stop but at the speeds they go in this section they would be able to stop before reaching the signal.

If the detector works like a hotbox detector, it will also advertise its result after the train goes by over the radio. Those detectors advertise their location, how many axles were counted, and whether or not there were faults. But I don’t know if that is how overgauge detectors operate.

[Canard]

I think that's what transporter was alluding to - that it will automatically call out the fault over the radio. Still, I too worry about false triggers.

[trainspotter139]

False positives happen with any train sensors and there are regulated procedures for handling detector alarms.

[Canard]

I think what we're saying here though is that anyone on the Spur Line Trail can just walk over and block the beam of the right-hand overwidth through beam sensor while a train is going through. That's super easy to defeat, with potentially very bad concequences.

[jamincan]

Perhaps it relies on their being a small intersection between the sort of people who know what it does, and the sort of people who would deliberately derail a train. In any case, I don't think the procedure for the trains using the Spur Line would change with the new LRT - they are still going to go very slowly through Waterloo. Perhaps their stopping distance is short enough that they would have time to stop the train before hitting the derailer.

[Canard]

I guess it would result in more of a nuisance-type problem, if it calls out the fault over the radio and the train just stops. The crew would have to deal with whatever procedures are in place to reset the derailler and proceed - probably tying them up for a couple of hours.

(I'm kind of having weird pangs about discussing this, now - kind of wondering if I should delete my last couple of posts about these devices as I don't really want there to be anything online that might give troublemakers any ideas in the future... what do you guys think?)

[timio]

There's probably some level of fault protection in these systems, otherwise a bird flying through could trigger the device.

Information can be readily found on the Internet about almost anything. I wouldn't worry about being the sole source of information for people wanting to cause trouble.

[timc]

Do you know how overheight detection works? Because I don't. I may be naive, but I assume it is not easily spoofed.

[ijmorlan]

Perhaps it relies on their being a small intersection between the sort of people who know what it does, and the sort of people who would deliberately derail a train. In any case, I don't think the procedure for the trains using the Spur Line would change with the new LRT - they are still going to go very slowly through Waterloo. Perhaps their stopping distance is short enough that they would have time to stop the train before hitting the derailer.

If the crew see the red as soon as it is visible, the train will be able to stop before reaching the derailer. Source: I witnessed a train-auto collision at that location. The train stopped before the engines had even fully cleared the crossing. By contrast the signal is visible from well down the track, probably almost as soon as the train has started crossing Allen St.

This is related to why I think a lot of the safety stuff around the freight interacting with the LRT is a bit much. At the speeds they go in this area, line-of-sight is a perfectly valid way of navigating, or backup to an automatic signalling system. Not at all the same as large fast mainline freights. I’ve seen videos of trains going into emergency stop. It takes a significant fraction of a minute before you can even tell that anything is different — eventually you realize the train isn’t going quite as fast, then after a longer time it gradually slows down and eventually screeches to a halt. Meanwhile, whatever they hit has been shoved a kilometre or so down the track. Around uptown, by contrast, a train stops much like a transport truck does: they hit the brakes, and they stop.

[EdM]

This is related to why I think a lot of the safety stuff around the freight interacting with the LRT is a bit much.

I'd argue that doesn't apply to the overheight / overwidth detectors though - regardless of how slow they're moving it seems feasible a crew could not know they're too wide / high.

[Canard]

Do you know how overheight detection works? Because I don't. I may be naive, but I assume it is not easily spoofed.

Yes, a few pages ago I posted some photos close-up of the one installed at the North end.

There are 3 pairs of through-beam photosensors - standard industry stuff. One is a transmitter, one is a receiver. If something blocks the beam, the state of the sensor changes (depending if it's Light-ON or Dark-ON).

The sensors itself are very easily spoofed. Put your hand in front of the ones down low at waist height. You'll block it - I guarantee it. The robustness of the system that others are discussing comes from stuff like debounce timers and so on to give it some level of stability against false triggers.

[Canard]

I can't stop laughing at this:

<blockquote class="twitter-tweet" data-conversation="none" data-lang="en"><p lang="und" dir="ltr"> <a href="https://t.co/OWlJeEB4mm">pic.twitter.com/OWlJeEB4mm</a></p>&mdash; Jesse Onland ?‍☠️ (@thefringthing) <a href="https://twitter.com/thefringthing/status/897511774051807234">August 15, 2017</a></blockquote>

[ijmorlan]

I can't stop laughing at this:

<blockquote class="twitter-tweet" data-conversation="none" data-lang="en"><p lang="und" dir="ltr"> <a href="https://t.co/OWlJeEB4mm">pic.twitter.com/OWlJeEB4mm</a></p>&mdash; Jesse Onland ?‍☠️ (@thefringthing) <a href="https://twitter.com/thefringthing/status/897511774051807234">August 15, 2017</a></blockquote>

That would make a good jigsaw puzzle.

[timc]

If we can put that much density of lighting (the original, not the 'shop) on King Street, then is there really much justification for not putting a crossing light where the IHT crosses Park Street? From what I remember, the proximity to the intersection at John Street and the potential for driver confusion was the reason to not do it.

[ijmorlan]

If we can put that much density of lighting (the original, not the 'shop) on King Street, then is there really much justification for not putting a crossing light where the IHT crosses Park Street? From what I remember, the proximity to the intersection at John Street and the potential for driver confusion was the reason to not do it.

I think the photo has been doctored

Having said that, a lot of the excuses given for why cycling infrastructure can’t be improved are just that — excuses, as far as I am concerned. Anyway, if a second stoplight would be too confusing, just put in a stop sign and give the trail priority (only semi-kidding).