CBTC for the T

[Lurker]

US has a higher standard for crash safety on almost every form of transportation than the rest of the world.

Freight also frequently shares the same tracks as passenger cars in this country, as the US still has the best freight railroads in the world, and that directly impacts safety concerns for all rail vehicles.

 

[F-Line to Dudley]

I have one question, why are all the subway cars in North America made of stainless steel, while the rest of the world uses aluminum? Is it some sort of safety thing?

Corrosion and graffiti prevention. Less rust, easy cleanup when they get tagged. NYC aggressively deployed them when it was trying to change the subway's image a graffiti-covered toilet. By the point a few huge systems went all-in on stainless steel the manufacturing side reached an economy of scale that it's pretty much all that's offered on carbodies now. As for why Europe still prefers steel: maintenance. U.S. systems go a lot longer on average between car orders and rehabs than other worldwide transit agencies that do much more aggressive maintenance, so that little problem called rust is not something they have to worry about as much eating their vehicles alive.


As for U.S. vs. Euro safety standards, varies by country but they aren't really any different for Metro systems. Metros are self-contained and not on the RR network, where the U.S. has the overkill FRA crashworthiness standards for freight intermixing. Same subway car manufacturers for the most part since there's hardly any left in North America, so the worldwide makes end up being very similar. Light rail too, but that's more because the rest of the world copied the PCC design to the hilt and made the same evolutionary improvements.

 

[whighlander]

F-Line --- I never understood why all the emphasis on cables -- pick the right frequencies and you can do the who thing wireless with an occasional redundant repeater

Such a system can in effect create an "indoor GPS" for when you are in the tunnel and use a real GPS for when you are on the Longfellow, etc.

 

[F-Line to Dudley]

F-Line --- I never understood why all the emphasis on cables -- pick the right frequencies and you can do the who thing wireless with an occasional redundant repeater

Such a system can in effect create an "indoor GPS" for when you are in the tunnel and use a real GPS for when you are on the Longfellow, etc.

Reliability/redundancy. GPS does have dead spots out in the wild, and in the tight linear confines of a tunnel all it takes is one transponder on the fritz to lose a signal around a curve. Train is instructed to auto-stop during a signal loss, so that's really a drawback. Whereas if you've got track-mounted hardware there's other layered systems that can better cover the loss at restricted speed.

Plus on an electrified line you've got tons of pre-existing cable plant to begin with hanging all over the walls, ceiling, and at track level. Even moreso in an century-old subway tunnel where there's a wide assortment of various-era stuff strung up. Total path of least resistance to go with fixed infrastructure over something newfangled for newfangled's sake. All that ancient cable doesn't last forever, either, so they're tardy replacing a spaghetti mess of crumbling and electrical-short prone copper from God knows what decade with fiber consolidating X many cables into 1.


The GPS thing is what freight RR's are using as their PTC system on their mainlines, and commuter rail systems in the Midwest and West that travel on freight-owned track. But freights can deal with a signal outage around a mountain pass much better than some traffic-choked northeast transit line. The Long Island Railroad did a trial with GPS-based PTC on one of its mains that was such a glitchy disaster they quickly scrapped it for Amtrak's battle-tested ACSES. Fixed infrastructure's always got a lower failure rate than remote, which is why nearly every grade-separated (non-streetcar) rapid transit system in the world uses something hard-wired, be it track circuit/cabs or mechanical trip-stop.

 

[NJBostonFan]

Corrosion and graffiti prevention. Less rust, easy cleanup when they get tagged. NYC aggressively deployed them when it was trying to change the subway's image a graffiti-covered toilet. By the point a few huge systems went all-in on stainless steel the manufacturing side reached an economy of scale that it's pretty much all that's offered on carbodies now. As for why Europe still prefers steel: maintenance. U.S. systems go a lot longer on average between car orders and rehabs than other worldwide transit agencies that do much more aggressive maintenance, so that little problem called rust is not something they have to worry about as much eating their vehicles alive.


As for U.S. vs. Euro safety standards, varies by country but they aren't really any different for Metro systems. Metros are self-contained and not on the RR network, where the U.S. has the overkill FRA crashworthiness standards for freight intermixing. Same subway car manufacturers for the most part since there's hardly any left in North America, so the worldwide makes end up being very similar. Light rail too, but that's more because the rest of the world copied the PCC design to the hilt and made the same evolutionary improvements.

Okay, I get it now. That means we can order an off-the-shelf Bombardier Movia to operate in America without any mods. Kinda like the ones in Toronto!

 

[F-Line to Dudley]

Okay, I get it now. That means we can order an off-the-shelf Bombardier Movia to operate in America without any mods. Kinda like the ones in Toronto!

Yep. The same exact LRV model they supply to the Minneapolis Hiawatha Line is used in 5 other countries. And Toronto has long used subway cars derived from our Orange Line makes, which Bombardier absorbed from from the old Hawker-Siddeley, Inc. Look familiar?. . .

.

Nigeria just bought a bunch of rehabbed Toronto units last year, so next time you're riding the subway in Lagos you'll feel right at home with the faux wood paneling.

 

[NJBostonFan]

Yep. The same exact LRV model they supply to the Minneapolis Hiawatha Line is used in 5 other countries. And Toronto has long used subway cars derived from our Orange Line makes, which Bombardier absorbed from from the old Hawker-Siddeley, Inc. Look familiar?. . .

.

Nigeria just bought a bunch of rehabbed Toronto units last year, so next time you're riding the subway in Lagos you'll feel right at home with the faux wood paneling.

No way I'm ridin on those anymore! I do think the Bombardier LRV they use on the Hiawatha Line is heavier than the one's in other countries. Want to see a Citadis Tram in Boston, with curved shapes and all that.

 

[NJBostonFan]

I also suppose a Movia would be a logical replacement. How many cars are the Orange Line trains now? 8? What about the rest of the system?

 

[F-Line to Dudley]

I also suppose a Movia would be a logical replacement. How many cars are the Orange Line trains now? 8? What about the rest of the system?

6. We don't have the platforms long enough for octuplets. It took 20 years of tedious excavation at the downtown stations to get all the century-old platforms from 4 to 6.

 

[NJBostonFan]

The best thing about articulated trains, is that you can walk from one car to another.

 

[DigitalSciGuy]

Super super late to this party - found it recently on a tangent looking for more information to summarise the need for some level of ATO for the Green Line...

...Plus on an electrified line you've got tons of pre-existing cable plant to begin with hanging all over the walls, ceiling, and at track level. Even moreso in an century-old subway tunnel where there's a wide assortment of various-era stuff strung up. Total path of least resistance to go with fixed infrastructure over something newfangled for newfangled's sake. All that ancient cable doesn't last forever, either, so they're tardy replacing a spaghetti mess of crumbling and electrical-short prone copper from God knows what decade with fiber consolidating X many cables into 1...

To add to this, that fiber backbone also provides greater capacity for other IP devices in the system, like digital signage, cellular, WiFi (both public and MBTA-protected networks), and surveillance, pretty much like what's going on in this really great Thales video that pretty much shows the full potential of modern IT-based transit operations:Thales Video

The capital cost for tunnel infrastructure reconstruction is high, but in the long run, unlocks for the MBTA a whole (growing) market of open or multi-standard IP-based rail devices that is becoming increasingly cheaper, reliable, and can be made inexpensively redundant because of economies of scale and Moore's Law. This also makes the MBTA more flexible in its operations by significantly reducing the manual labour to maintain or tweak the system. Doing something as basic as updating passenger information announcements can be remotely flashed to a balise in a tunnel or updating the firmware on trains resting in the yards at night.

The IT-ifying of transit also has significant implications for the workforce needed to maintain the system's electronics. In the long run, there will be less need to machine and rebuild out-of-warranty mechanical systems. With solid-state electronic devices becoming smaller, cheaper, and more powerful, you can build them in housing that protects them better from the elements and makes them much more reliable than the early railroad electronics that has left many railroad veterans with a bad impression. There will also need to be a mental mode-shift to maintenance practices; modularity of components goes up at the same time as complexity of electronics. This doesn't necessarily mean that we'll have to ship out all maintenance duties to the OEMs of the equipment, but it certainly reduces the level of hands-on, I-can-repair-that-myself maintenance that the guys at the Everett shops are excessively proud of.

 

[F-Line to Dudley]

Yep. And that's exactly what NYC is finding in its post-Sandy assessment of subway infrastructure. The shuttle lines that had solid-state CBTC fared much better in the storm than the others. Fewer rusted-out mechanical parts, less shorted-out copper, fewer cables and relays total to repair/replace, and much better-documented cables after such a recent rebuild. Get waterlogged...it's easier to replace some far-spaced relays that are nothing but printed circuit boards connected to un-rusting plastic fiber than it is to replace miles and miles of copper, switches, switch heaters + cables, and signal lights. It is one of the justifications the MTA is using for spending >$10B to roll out CBTC conversions systemwide over the course of the next 15+ years. Their couple years of trials have proved successful at debugging the system and testing its operating efficiency. Now Sandy has inadvertently triggered an urgent rush to get it out on a mass scale because of the maint vulnerabilities it solves. The latter decidedly unsexy consideration ended up being way more important to the bottom line than all the gee-whiz new traffic management technology it brings to their table.


This is why the Blue Line is the logical first choice for getting it. It's the only one left using the NYC-style mechanical trip arms. According to the specs the T lists in the Cap Improvements Financials, the shortest of the 3 heavy rail lines has: 145 trip arms with 2 switch heaters each, and 74 signal heads over 12 miles. All of that needs a shitload of its own electricity and own electric cables, and has moving parts, heating elements, and incandescent light bulbs that routinely fail during any given service week and need daily spot inspection/spot replacement. The Red Line--45 miles of track, 4 times Blue's, enormously more complex overall--only has 2 mechanical trip arms x 2 heaters each on the entire line (don't know where...probably some yard where they spare no precaution keeping a signal failure from sending a train through a wall or fence), and 12 signal approach lights (only used for protecting crossovers). The solid-state ATO cab signals do all the work of those masses of mechanical switches with simple analog radio pulses through the running rails. And that system still uses a lot more copper wire and track-mounted hardware than CBTC, which consolidates it further and uses software instead of hardware to mark the signal blocks.

All it takes is a wet Nor'easter from hell to flood the surface portion of the Blue Line for these decidedly unsexy improvements to sound awfully, awfully sexy to North Shore commuters. Modernizing is the difference between getting it back up to full speed 3 days after a natural disaster vs. a month of it having to run at max 10 MPH line-of-sight with no working signals, trains once every 15 minutes at peak, and pauses for schedule adjustment at every single station while an army of inspectors with clipboards dispatch it by hand. Much like the Kenmore end of the Green Line had to for months after the Great '96 flood.

 

[DigitalSciGuy]

Hopefully the origins of the '96 flood will be mitigated with the daylighting and restoration of parts of the Muddy River and improving its natural capacity to absorb and withstand storm surges. That said, you're absolutely right. I had almost completely glossed over the increasing threat of storm surge.

Having another justification for fortifying transit infrastructure, namely that of flooding mitigation, is an even better way to pitch it to those who insist on maintaining a copper-based system or spending the capital investment elsewhere just because of the sticker price.

 

[whighlander]

Yep. And that's exactly what NYC is finding in its post-Sandy assessment of subway infrastructure. The shuttle lines that had solid-state CBTC fared much better in the storm than the others. Fewer rusted-out mechanical parts, less shorted-out copper, fewer cables and relays total to repair/replace, and much better-documented cables after such a recent rebuild. Get waterlogged...it's easier to replace some far-spaced relays that are nothing but printed circuit boards connected to un-rusting plastic fiber than it is to replace miles and miles of copper, switches, switch heaters + cables, and signal lights.

Up to a point:

First the Fiber might look plastic on the outside but its Glass inside

2nd when real power is involved the Digital Stuff still depends on Relays with mechanical contacts -- right now you are not turning on big currents purely with solidstate although in the near future with Silicon Carbide devices you will

But the real keys:

1) Control modules [outside of the remaining ones with power relays and contactors] can be potted in epoxy -- no dismantling -- just plug and play -- and recycle the boards

2) Much of the wiring can be reduced to much less volume of Glass Fiber

3) Most importantly -- all of the stuff becomes remotely programmable from HQ -- almost no need for manual labor in the tunnels

 

[winstonoboogie]

Looks like the installation of CBTC on NYC's second line to have it installed (the L has it, the 7 is getting it) has been delayed, in part because of Sandy

http://www.railwayage.com/index.php...Deadline+extended+for+Flushing+Line+CBTC+work

 

[DigitalSciGuy]

...First the Fiber might look plastic on the outside but its Glass inside...

But the real keys:

1) Control modules [outside of the remaining ones with power relays and contactors] can be potted in epoxy -- no dismantling -- just plug and play -- and recycle the boards

2) Much of the wiring can be reduced to much less volume of Glass Fiber

3) Most importantly -- all of the stuff becomes remotely programmable from HQ -- almost no need for manual labor in the tunnels

Thankfully, glass also doesn't rust and the epoxy coating on the outside had better not be rubbing up against other cables otherwise you've got a whole mess of other things to worry about. Suffice it to say, fiber optics make for a pretty robust, 21st-Century upgrade from miles and miles of thick copper transmission cables.

Yes, ultimately the industry is moving toward plug-and-play, commoditized, open-standard, IP-based devices so that the expense of bespoke devices from yesteryear largely disappears and you are liberated to spend that money elsewhere (software and developers) while still providing even better data resolution of every moving (and not moving) piece of your system. Ideally, if the MBTA ever goes full deployment of CBTC on the Green Line, they take the opportunity to rip out all the non-power transmission copper in the tunnel, get some fiber in there, and really pushes for a proven CBTC system that's modular and can be supplied with open standard parts.

Also, as part of the joyous announcement about being the first transit system in the US to fully deploy next train information on all their heavy rail lines, the MBTA also alludes to whatever work they're slowly doing on the Green Line to get a higher data resolution of trains to facilitate next train information by either next year or end of this year. Anyone know what that's about?

Looks like the installation of CBTC on NYC's second line to have it installed (the L has it, the 7 is getting it) has been delayed, in part because of Sandy

Yep, I was in New York for over a week earlier this month and part of my commute I could take the 7 train (I almost literally grew up on the 7 train). Looks like they're well into CBTC work already and took several full weekends with no 7 trains into Manhattan to not only do repair to the Steinway tunnel that connects Manhattan and Queens, but to also do general track work and signal upgrades. Indulge yourself with a little bit of train/construction/equipment porn here and here.

 

[winstonoboogie]

Also, as part of the joyous announcement about being the first transit system in the US to fully deploy next train information on all their heavy rail lines, the MBTA also alludes to whatever work they're slowly doing on the Green Line to get a higher data resolution of trains to facilitate next train information by either next year or end of this year. Anyone know what that's about?
[/IMG]

They are installing GPS systems on the cars, first the Type 8s and then the Type 7s, to provide location data on the surface. They will also be installing more Automatic Vehicle Identification (AVI) transponders in the subway to get better location data in the subway. There is an AVI system now in place since the 1990s, but there are very long gaps between data collection locations. They will be updating this system to get the underground location data that the GPS system can't provide. The data from the two systems will be combined to provide location data both to the public and better information to dispatchers than provided by the present AVI system.

 

[Matthew]

I'm just going to put this here for future reference.

http://forums.philadelphiatransitvehicles.info/topic/7053108/1/

SEPTA's collision-avoidance program delaying trolleys
By Paul Nussbaum

Inquirer Staff Writer

A recently activated collision-avoidance system for SEPTA subway-surface trolleys has created chronic rush-hour delays for riders, despite $25 million and much of a decade spent trying to work out the flaws.
The "communications-based train control" system was designed to prevent rear-end collisions by automatically controlling trolleys as they travel underground on their way to and from West Philadelphia.

But the system has slowed the cars too often and sometimes has backed up trolleys five or six deep at the West Philadelphia entrances to the 2 1/2-mile tunnel.

The heart of the problem is that the automatic control system cannot keep up with the rapid arrival and departure of trolleys during peak hours, when a trolley arrives about every 50 seconds.

"Look at that. It's brake and go, brake and go," a frustrated trolley driver said last week, pointing to a red light on his instrument panel. Each time the light came on, he had two seconds to slow the trolley or an automatic brake would kick in.

Operators complain that trolleys are routinely 30 minutes to an hour behind schedule during peak periods, and they blame the control system. They say they made much better time with the old-fashioned wayside signal lights and manual braking.

Five Green Line subway-surface routes carry about 90,000 riders a day, converging on the tunnel at two portals in West Philadelphia. The five routes use 112 electric trolleys, which run at street level under manual control after they leave the tunnel.

After about four years of design and installation, SEPTA put the automatic system into part-time operation in the tunnel in May 2005. Because of continuing rush-hour problems, though, SEPTA kept using the manual system during peak hours for three years.

That ended after 11 people suffered minor injuries July 22 when a trolley ran into the back of a stationary trolley at the 13th Street station and knocked it into a third stopped trolley.

It was the kind of accident the automatic system was created to avoid, and SEPTA officials decided to put the system into full-time operation.

 

[George_Apley]

That sounds like a problem for software engineers to fix.

 

[bigeman312]

That sounds like a problem for software engineers to fix.

Not as I understand it. Sadly, there is a limit to the possible frequency when using ATC and that limit is well under manual braking/line-of-sight. If the MBTA were to implement ATC on the Green Line it would quite likely cause decreased peak frequency and there is nothing we can do about it. All in the name of safety.

Is this correct, experts?