Green Line Extension to Medford & Union Sq

jklo said:
Missing the forest for the trees here. If GLX is falling apart because of substandard materials, that's not good.
Click to expand...

Says the person who injected the xenophobic aspect into the conversation. Grow up.
 
Stlin said:
Err so the GL track geometry LRV and 2 more LRV based work trains were non operational and scrapped without direct replacement in 2020. https://www.universalhub.com/2020/imagine-getting-these-trains-under-christmas-tree

I'm not actually sure what the MBTA has been using since for its geometry testing, but probably a hi-rail of some sort or contractor equipment?
Click to expand...
I believe they now have an in-house sensor rig they can swap to any trolley in the revenue fleet to do the geometry testing, and don't have to stock or lease any specialized work equipment. It's just a bunch of laser guides...nothing all that bulky. The old Boeing geo work car wasn't exactly a hyper-custom beast...it just had a permanent rather than portable sensor rig so they didn't have to vulture revenue cars for the inspection shift. The 3 HRT lines have long used portable sensor rigs in lieu of any specialty geo cars. Commuter Rail's rented Amtrak Corridor Clipper geo car is likewise a pretty basic setup; the NEC-member states' CR agencies simply coordinate their inspection schedules so it's in constant rotation around the country rather than each agency bothering to outfit one of their own coaches with sensors.

Highwayguy said:
That’s a good point that opens up a whole bunch of new questions; although l wonder if the wheel profile between the modes could be a partial factor? Although my recollection is that the LRV profile was more forgiving than mainline.
Click to expand...
Source rail is all pretty generic and bulk-supplied. The only differences is that RR rail comes in 1/4 mile ribbons to be welded together, while rapid transit rail comes in 800 ft. (or .15 mile) ribbons to be welded together...the shorter lengths so that the rail drop can navigate tighter rapid transit turning radii. Commuter rail generally uses 136 lb./yd. weight rail (sometimes as low as 100 lb./yd. on turnouts, but the T sticks with the 136 lb. mainline & freight 'superset' weight for bulk ordering purposes). Rapid transit standardizes on 100 lb. rail pretty much everywhere worldwide. Bulk rail usually ships at a RR wheel profile (because that's the majority of worldwide volume for rail, and thus fetches the best bulk rate), and for rapid transit installs they simply run a rail grinder over it to change the profile to rapid transit wheels. That ends up easier than buying specialty-ground bulk rail, so most rapid transit systems apply the re-grinding post-install. The specialized Loram rail grinders were very busy on the GLX corridor in 2021 and early-'22 working over the track before the first test trains went on the corridor.

As mentioned, the geometric defect detected was that the tracks were spaced too closely together, which is an alignment problem not a wheel profile (or wheel profile wear) problem. It's not clear what's causing that, but you can rule out a problem with the profile/rail grinding.
 
F-Line to Dudley said:
As mentioned, the geometric defect detected was that the tracks were spaced too closely together, which is an alignment problem not a wheel profile (or wheel profile wear) problem. It's not clear what's causing that, but you can rule out a problem with the profile/rail grinding.
Click to expand...
I mentioned the wheel profile in the sense of tolerance to gauge deviations. Per the headline of most recent Globe article on this:

“Green Line extension track, which T says ‘has always been narrow’”

AKA the gauge has been ‘too narrow’ for a year with 0 issues running at line speed. So either for the past 12 months every green line train had a significant risk of flying off the rails and the T put riders at an unacceptable risk of being in a derailment, or the 3 MPH slow zones are overcautious overkill. Take your pick which is worse.
 
Slow zones seemed less slow from Gilman to North Station (and vice versa) today, though still present.
 
I rode the green line between GLX stops and North Station today, and noticed something interesting.

In previous closures between North Station and Government Center for garage demolition, when going towards North Station, the train would cross over from the inbound track to the outbound track in the North Station yard before entering the station, then enter North Station backwards on the outbound track, and leave the way it came on that track.

However, this time, they didn't do that. The inbound train stayed on the inbound track into North Station, let everyone off, and then continued on the inbound track towards Haymarket. Later, when I was heading back, a train heading outbound came from the Haymarket direction to pick us up to go outbound.

Before, I figured that they had been switching tracks in the North Station yard because any crossover (if one exists) between North Station and Haymarket was too close to the garage demolition to be safe. However, today, it seems like they must consider it safe enough (at least for their employees)
 
Highwayguy said:
I mentioned the wheel profile in the sense of tolerance to gauge deviations. Per the headline of most recent Globe article on this:

“Green Line extension track, which T says ‘has always been narrow’”

AKA the gauge has been ‘too narrow’ for a year with 0 issues running at line speed. So either for the past 12 months every green line train had a significant risk of flying off the rails and the T put riders at an unacceptable risk of being in a derailment, or the 3 MPH slow zones are overcautious overkill. Take your pick which is worse.
Click to expand...
Right. I said this in the other thread but the T owes the public an explanation of the risk management and tradeoffs with slow zones in general here.

If the trains have been running for months on defective track at higher speeds, in some cases significantly higher, what risk of derailment were they running? If that was very low, why slow them down so much and so widely? The answer may be "because we don't have any idea what we're really doing with regard to track maintenance" but if so then why would anyone trust them to fix it or even run any trains at all?

If the answer is "because we decided six months ago we don't want to risk even a .0001% chance of a derailment" that seems like a bad way to run a mass transit system honestly, and I would argue that the long term harm to ridership to say nothing of the short term pain for riders make that a bad choice. Especially when there have been numerous safety incidents recently that have nothing to do with the track and prompted few similarly dramatic changes.

If the risk of derailment is relatively high, but the defect can only be detected with the geometry train, why do they only run the geometry train infrequently?

If we ran a geometry train on BART or Baltimore light rail or whatever, would we find similar defects? If so, do they have a less strict standard for slow zones? Why? If not, and the MBTA is singularly bad at this, why aren't we firing everyone and hiring whoever can fix rail as soon as possible? There's just no good logic here.
 
More problems with the MBTA!!!

commonwealthmagazine.org

Skepticism, exhaustion at state of the T - CommonWealth Magazine

SUBWAY FIXES on the MBTA hardly feel like they’re on rails. As Red Line riders gear up for a 16-day October closure of a southern spoke, the shiny new Green Line extension was revealed to have such serious track defects that trains are crawling along at 3 miles per hour. Over the past month...
commonwealthmagazine.org commonwealthmagazine.org
 
They should close the line & have repairs done immediately! We wait all this time to find THIS out. Ridiculous!! :eek:
 
I know and I am sure this is wayyyyyyyyy more complicated than just a simple close it and fix it, but like, what if it could be hypothectically possible to just shut the whole damn thing down if they can't fix a defect within 2 weeks of identification? Can't fix a track defect within two weeks of identification? Sorry, but the tracks must be closed down until it can be fixed. The MBTA shouldn't be allowed to run trains on tracks where a defect more than 2 weeks old have been identified.


The MBTA can't have lone workers on tracks where trains are running due to FTA orders. The FTA should just go a step further and just ban the MBTA from running trains on tracks where they have defects more than 2 weeks old identified.

It's just so ridiculous at this point. Dragging this out for ages. There's like months and months of tracks shutdowns to do at this point. If they ever get around to finish re-training all the workers, after that, they should just get them out there and get the months and months of track repairs done. 24/7 repairs around the clock, system-wide.

They can fix I-95 in Phily in 10 days. At this rate, the city of Boston would like need another hard shutdown like that of COVID to get this done anywhere from 5 weeks to 2-3 months; or just drag this out 5 more years of post-pandemic-half-recovered-barely-funcutional city. Which poison is worse and which is the least bad poison? I have no idea.
 
Last edited:
Mother Brook said:
Right. I said this in the other thread but the T owes the public an explanation of the risk management and tradeoffs with slow zones in general here.

If the trains have been running for months on defective track at higher speeds, in some cases significantly higher, what risk of derailment were they running? If that was very low, why slow them down so much and so widely? The answer may be "because we don't have any idea what we're really doing with regard to track maintenance" but if so then why would anyone trust them to fix it or even run any trains at all?

If the answer is "because we decided six months ago we don't want to risk even a .0001% chance of a derailment" that seems like a bad way to run a mass transit system honestly, and I would argue that the long term harm to ridership to say nothing of the short term pain for riders make that a bad choice. Especially when there have been numerous safety incidents recently that have nothing to do with the track and prompted few similarly dramatic changes.

If the risk of derailment is relatively high, but the defect can only be detected with the geometry train, why do they only run the geometry train infrequently?

If we ran a geometry train on BART or Baltimore light rail or whatever, would we find similar defects? If so, do they have a less strict standard for slow zones? Why? If not, and the MBTA is singularly bad at this, why aren't we firing everyone and hiring whoever can fix rail as soon as possible? There's just no good logic here.
Click to expand...
This is what I don't get. . .

First...On the rail install there's a geometric check immediately. The contractor uses one of their rented track critter machines to do a geo check right when the track is tamped and aligned, and that should've caught an alignment issue unless said alignment issue is only something that manifests after wear (which the T has said specifically is not the case). That's so they can fix the alignment problem within hours/days while the track laying/tamping track critter machines are still readily available on-site. It's theoretically possible that the contractor's geo-checking track critters had faulty sensors and failed to catch something. It's less possible but still theoretically possible that they didn't catch it on a re-check. The T's own geo car rigs wouldn't have caught it at the installation stage because the overhead wires weren't installed when the track was being laid and the track was not yet connected to the greater Green Line because Lechmere Viaduct was still severed and being worked on. So okay...could be contractor fault that it escaped notice, but there still should've been fail-safes given the multiple rounds of geo checks.

Second...the very first thing the T did when the rails went live for testing was doing clearance and geo tests. Those were the first crawling-speed rides on the extension, and they used a few borrowed revenue trolleys of each Type 7/8/9 class fitted with foam blocks for the side clearance tests, the geo rig for the rail alignment, and all kinds of other sensors and doohickeys to gather telemetry. And only after weeks of doing those slow-crawl tests over and over again did they proceed to higher-speed tests, and finally "simulated-run" full-speed tests. Each graduation in that sequence would've required sign-off that the geometry was up-to-spec. And, finally, stacking up under simulated daily-service loading. It was far, far from a "one-and-done" pass. There were quite likely several dozen geo tests. And the FTA would've needed to see receipts on those tests before final sign-off was given to initiate revenue service. So it's implausible at best that something could be missed here, but I guess incompetent testing x20 could theoretically do it.

Third...as above, they're sure as sure can be that this wasn't a "track settling" thing after a year of simulated-and-actual revenue service. Sure enough to tell the public that definitively, and plausible enough since the experts have already chimed in that track spreading is the most common wear profile and that track shrinking is almost unheard of in that case. While geo checks would've resorted to a "normal" schedule after testing, they aren't zero. The geo testing cars aren't brought out only once a year; they do at least an intermittently regular churn across the system, and GLX would've been under increased scrutiny vs. the rest of the GL because its wear profile was an unknown. It's certainly more possible that testing was lax at this regular-service stage, because there's a firehose of evidence coming out of the agency that inspections across-the-board are slipping. But our infamous derail-on-an-askew-human-hair Bredas have had zero trouble running at full speed on all that trackage all this time, so the canary-in-a-coal-mine gimps in the fleet didn't catch anything amiss either. Maaaybe they fell down hard on this stage (and can expect to be sanctioned by the FTA for it when they can't show the receipts), but it doesn't explain how they got through Stages 1 & 2 clean first. Again, it's like geo test x20+ that would've missed it...before very suddenly in the last week they caught it over 4 widespread miles of track.


Holy moly, is that a lot of blown stop signs and some acrobatic leaps-of-logic it takes to even hazard a guess as to what happened here. It's bizarre in the extreme. The accounting on this needs to be immaculate and detailed, both for the riding public and the feds. Because there seems to be no good explanation for how something this basic and pervasive got through the various...and numerous...checkpoints from 2022 rail installation to today. You can't fault "lazy slacking workers" here without going full-on tinfoil hat with mass coordinated global agency-wide conspiracies, because of the sheer number of different shifts that would've laid eyes on testing telemetry taken at dozens of different times over the course of a year for us to even get to a point where regular revenue service was operating ho-hum daily.
 
Last edited:
My personal take from what I've read and heard is that while the initial geo tests were probably on the narrow side of the ± baseline measurement, to allow for expected wear based widening, but it was inside that tolerance even if closer to the - limit at the time from all indications. It's how I would expect a wear component to be engineered for maximum service life. The Globe article noted that the T claims that a geometry scan in March was clean, a scan in June resulted in some defects that were promptly repaired, but new scans in Sept say they've come back and there are more of them. That's if we assume that the geometry data is clean and proper, which has question marks all over it given the Ts acknowledged shortcomings in that area. But if it is true that there is something progressive, that has resulted in the track gauge narrowing and/or continuing to do so in defiance of normal wear logic, that is a problem I don't understand, and from what the T has said publicly doesn't either. I'm willing to say, in the absence of evidence to the contrary that either abnormal track wear and or poor testing protocols allowing out of spec track to pass are possible, but I'd agree that the T needs to be very upfront about which is is and why.
 
Last edited:
But if they were to shut the whole thing down, commuters would be stranded from trying to get to or from work, or to or from doctor appointments. To stay healthy is just as important as being able to work or just to go about life. There would be massive tie-ups at every line as commuters try to get to or from work on shuttle buses. Outside of that, there's always night work. Like they sometimes do with the D Line. When the last train leaves & heads to town, close that line down & get to work to fix the track problem. I can understand closing down a line for track work, such as what was done for the Orange Line last summer, but to shut down a line for a month, then reopen it & still have the same track problems & then slow zones still, that is utterly insane & ridiculous! Now, practically ALL lines have those annoying slow zones, & I don't see this work being done with just a snap of the fingers!!

The FTA should have really LEANED on the T & made sure that everything was fixed & well taken care of. New job in Somerville? How on earth would you get there in time on a trolley crawling along at only 3 mph?!! This is slowest of the slow! The MBTA should & could fix the lines. They're just going to have to get the manpower in here to do the work. Enough is enough! Spend the money & have the work done right, instead of Mickey Mousing things along. This is their problem. They seemingly refuse to get up off the money! Even if it means that they are short on the funds, they HAVE to do this, or some type or disaster might happen & then they face possible lawsuits!! :eek: :eek::eek:
 
Last edited:
Additional consideration (I think): In the initial speed restrictions implementation in March, third-party inspectors and internal ROW/maintenence inspectors were out inspecting all track and verifying each other's tests. Those slow zones spread across GLX, which presumably indicates both an independent party and internal personnel did not see this issue.

Unless the third party inspectors were blindly following the MBTA's orders and lack of SOP to improperly inspect track, I'm not sure how this couldn't have been caught then. Presumably the firm they contracted out to would have some liability to report these defects if they were apparent.
 
RandomWalk said:
Tin foil thought: This is yet more fallout from Baker’s rush to get the GLX grand openings done during his term.
Click to expand...
Also, his rush to get the new Red & Orange Line trains in. So much of a rush it was, that proper inspections of the new cars weren't done right & properly. Things kept going wrong with them & they were seemingly being yanked out of service for every little thing! :eek:
 
Without being an expert, this reads more to me like an inspection/data problem than in infrastructure problem. An agency trying very hard to be cautious, new track probably laid toward the narrow end to anticipate widening over time, and maybe some mis-calibrated equipment. As F-Line said, there were zero problems running the most persnickety trains we have over these tracks. The only issue so far seems to be on paper.
 
The GLX slow zone is getting Eng's attention and focus:
www.bostonglobe.com

T chief trying to figure out how Green Line extension tracks got so bad, so quickly - The Boston Globe

“I need to find out exactly how this came about,” MBTA general manager Phillip Eng said at a meeting of the T's oversight board on Thursday.
www.bostonglobe.com www.bostonglobe.com

Full text, thanks to Reddit:
MBTA general manager Phillip Eng said he’s focusing on fixing the problems on the Green Line extension that are forcing trains to travel at just three miles per hour along the transit agency’s newest piece of subway.



“I ride that Green Line extension every day, as well. I talk to the riders. I fully understand their frustration,” Eng said at a MBTA board of directors meeting Thursday. “These types of occurrences are unacceptable, and that’s why we’re working very hard in terms of capital delivery and working through these issues earlier. What we are doing now though is we are going to make sure we address those recent speed restrictions promptly, timely and focus on safety.”


The Green Line extension branches to Union Square and Medford/Tufts, which opened in March 2022 and last December, respectively, are in such bad condition that the agency is now forcing trains to travel at a walking pace over sections that add up to more than one mile long, making rides unbearably slow less than a year after the full project opened.



The MBTA has said the rails are now too close together in many areas, making full speed train travel too dangerous. The T said the Green Line extension tracks were built “narrow,” but have narrowed further to an unsafe degree since the project opened last year.


The agency has not explained why the project was built with narrow tracks or why such serious problems are happening so soon after it opened, or who is responsible.
Click to expand...
 
I've been asking myself this question for the past couple of months - since all of these crazy slow zones and restrictions started. Who is actually making the call on where these slow zones are and how fast (or slow) the trains are allowed to go? Because it's starting to feel like the lawyers are the ones ultimately making these decisions and not the "train" staff.
 
Is there a chance the concrete ties were laid "green" / not fully cured and shrank a bit? What's the tolerance on the distance between rails??
 
dhawkins said:
Is there a chance the concrete ties were laid "green" / not fully cured and shrank a bit? What's the tolerance on the distance between rails??
Click to expand...
No chance that it's anything tie-related. Concrete is superduper rigid from the factory, and wood...while more flexible...doesn't move enough to cause alignment issues until the ties are at end-of-life (which GLX's obviously aren't).


I dunno about how the FTA regulates the tolerance for rapid transit, but here's the specs tome from FRA land: https://railroads.dot.gov/sites/fra.dot.gov/files/2020-08/2008_Track_Safety_Standards (1).pdf. Track geometry starts on p. 108 of the PDF.

A 50 MPH-design rapid transit line would correspond to track Class 3 on a RR. On a Class 3 RR the gauge cannot vary more than 1-3/4 inches between 4'8" and 4'9-3/4". Whereas standard gauge is 4'8-1/2". So that wiggle room is permissive by 1-1/4 inches in the spreading direction, but only 1/2 inch in the contracting direction. All track classes have the same 4'8" minimum, but the maxima go from 4'9-1/2" for high-speed track to 4'10" for passenger-minimum 10 MPH Class 1 track and 4'10-1/2" for freight-only Excepted track.
 
Last edited:
You must log in or register to reply here.

Back
Top