News   Mar 06, 2026
 1.3K     6 
News   Mar 06, 2026
 304     0 
News   Mar 06, 2026
 1.8K     0 

TTC: Other Items (catch all)

Now I'm curious how the St. Petersburg subway manages to have PSDs Ultra Pro Max with old manual 81-717 trains with 1970s technology (definitely not driverless or ATC'd).
PSDs don't actually require moving block CBTC i.e. modern ATC systems or even older fixed block ATC. Case in point metros that had PSDs prior to ATC retrofit. There are regional and intercity HSR-esque lines that have PSDs in China without the ATC we're all talking about, much less CBTC.

The earliest ATC / ATO was fixed block, and later systems gradually improved to a sort of halfway point between traditional fixed block and moving block, before communication based train control became widespread. Technically you don't need CBTC to get moving block, but in practice, CBTC is a prerequisite for moving block. ATP and ATO are parts under the umbrella of ATC. I assume the ATC we're talking in this thread would imply ATO/driverless capability or at least semi-automatic / GoA2 like on the TTC's Line 1.
 
PSDs don't actually require moving block CBTC i.e. modern ATC systems or even older fixed block ATC.
But they do require the train stopping at the exact same place (down to the cm) every single time, something that just isn't doable with human drivers, which aren't exactly perfect robotic machines.
 
But they do require the train stopping at the exact same place (down to the cm) every single time, something that just isn't doable with human drivers, which aren't exactly perfect robotic machines.
Not sure where this idea even comes from. Yes, a computer driven machine will be able to stop in the same place all the time much more reliably, but there are workarounds.

Nothing says that PSDs have to have the exact same dimensions as the train doors, if you make the PSDs wider then you suddenly have a lot more room for error (see also: the gap fillers New York used/uses at the old South Ferry and 14th St - Union Square stations).

14479789430_d1f84935bb_z.jpg


New York's IRT stations tend to have no extra room at the end of the platform, either, and they get on fine, too.


Sure, it's not as convenient, but if they truly felt that the danger was so prominent that PSDs were expeditiously required, they could do it.
 
But they do require the train stopping at the exact same place (down to the cm) every single time, something that just isn't doable with human drivers, which aren't exactly perfect robotic machines.
It's possible with manually driven trains and some driver-aids active when trains are alongside platforms, but none that rise to the level of the ATC we seem to be talking about. Also, they don't have to be down to the centimeter, PSDs are usually much wider than train doors (buffer).

I think the buffer is probably smaller, but not by much compared to what I've read can cause a TTC subway operator to get moved permanently to other vehicles. If an operator misses their mark by too much, 3 times, they apparently get booted from subways onto streetcars or buses? (UT or reddit post)
 
Last edited:
Nothing says that PSDs have to have the exact same dimensions as the train doors, if you make the PSDs wider then you suddenly have a lot more room for error
Sure, but there's a limit to how much wider the PSDs can be vs. the train doors, and it's basically this. Might add a few feet for error. And those doors in St. P don't appear that much wider than the train doors.
(see also: the gap fillers New York used/uses at the old South Ferry and 14th St - Union Square stations)
Bit of a different situation here, since, as you're very well aware, these trains run on a fixed path that allows them to pass within a few cm of the platform without scraping it, unlike a bus where the lanes are wider to allow room for error.
It's a total shame none of the 5's or 6's were preserved. Not a single one :/
That's why I always keep saying it loud & clear: the T1/TRs better never be saved either, not a single one. That's the only way to make it fair, "not repeating past mistakes" would just make it unfair×2.
 
Last edited:
Sure, but there's a limit to how much wider the PSDs can be vs. the train doors, and it's basically this. Might add a few feet for error. And those doors in St. P don't appear that much wider than the train doors.


Let's assume a scenario where the fixed structure of the PSD is placed on the middle of the yellow safety line, and the sliding doors themselves jut out of the fixed structure, in the same way the SRT doors did (they are not flush with the fixed structure in the way most subway car doors are).

If you alternate the positions of the actual sliding door portion (one door on the inner (passenger facing) side, one door on the outer (track facing) side), you can have a gigantic amount of space.

Bit of a different situation here, since, as you're very well aware, these trains run on a fixed path that allows them to pass within a few cm of the platform without scraping it, unlike a bus where the lanes are wider to allow room for error.
The point of the example was to show how they don't have to match the width of the doors. Not sure where I mentioned buses.
 
Let's assume a scenario where the fixed structure of the PSD is placed on the middle of the yellow safety line, and the sliding doors themselves jut out of the fixed structure, in the same way the SRT doors did (they are not flush with the fixed structure in the way most subway car doors are).
The SRT-type doors are still constrained by the same limitations as sliding subway doors with door pockets, to avoid hitting each other when they're fully open.
If you alternate the positions of the actual sliding door portion (one door on the inner (passenger facing) side, one door on the outer (track facing) side), you can have a gigantic amount of space.
I suppose, but I've never seen such a door configuration anywhere.
The point of the example was to show how they don't have to match the width of the doors. Not sure where I mentioned buses.
Got it.
 
The SRT-type doors are still constrained by the same limitations, to avoid hitting each other when they're fully open.
I suppose, but I've never seen such a door configuration anywhere.
Well, that is why I specified they would have to alternate. Of course, if they were all on the same side, it wouldn't work.

As for the fact that you haven't seen any anywhere... I'm offering possible solutions that show that new trains are not inherently a prerequisite for PSDs, if sufficient engineering creativity is shown. Just because no property thus far has deemed such a configuration to be necessary, doesn't mean it couldn't work if the TTC did do it. But the arrival of new trains, even if they are catastrophically delayed, will still predate the installation of PSDs by many years, so the point is moot. It's just a thought exercise.
 
Last edited:
The point of the example was to show how they don't have to match the width of the doors. Not sure where I mentioned buses.
Also a different situation, because they don't slide in the same direction as doors do, and thus are not constrained by width whatsoever, they could have a single long gap filler spanning the length of the entire platform (aside from it being curved, that is) if they wanted to.
Just because no property thus far has deemed such a configuration to be unnecessary

But the arrival of new trains, even if they are catastrophically delayed, will still predate the installation of PSDs by many years
It should also predate the installation of ATC by at least a couple years, and ATC is due in 2036.
Frankly, the T1s were an outdated design for something built between 1995-2001 anyways.
Good reality check for those who think they're still top-of-the-line modern in 2025.
 
Among others. Montreal's MR-63 cars started delivery in 1965 and were retired in 2018. 53 years. Montreal's MR-73s started delivery in around 1974, and are still in service 52-years later. They are planned to be replaced in 2036 - at the age of 60 years.

NYC are still retiring their R44s, with deliveries starting in 1971, and probably all gone by the end of 2026 - 55 years service. The R42s did 50 years. The R32s from 1964 were retired in 2022 - 58 years.

The 1972 London tube stock are likely to be replaced in the late 2030s or early 2040s. Well over 60-years old, might hit 70! However the 1973 stock should be gone in 2030 - only 57 years old. The final 1938 stock was withdrawn in 2021 - 80+ years. Though the final 30 or so years was by British Rail (and successors) on the Isle of Wight.

The subway in Buenos Aires is still using some some subway cars from 1954 - over 70 years old. Paris's MP 59 (very similar to Montreal's MR-63) trains were in service for over 60 years from 1963 to 2024 while their MF 67s entered service in 1968 and won't be fully retired for 2034 - over 65 years of service.

The 1972 stock is hardly an outlier. Yeah, it requires more work as it gets older. But it's 100% an option; but harder to fund maintenance than capital costs in Ontario.
You forgot to mention that SEPTA is still running the Kawasaki B-IV subway cars from the early 1980s on the Broad Street line. They want to run them to 50 years I think, something TTC never did on the H5s back then.
 
It’s not quite a subway, but it’s crazy thing Boston and I think Philadelphia still run PCC’s

Also does anyone know what’s going on at downsview (Sheppard west)

IMG_6025.jpeg
 

Back
Top