Green Line Extension to Medford & Union Sq

I need to check what I've added. The problem is I don't have exact locations for switches and I'm not 100% on what they are doing with the new yard. If anyone can get me that info I'll add it to the map.

Pretty sure they've had track layouts final for quite some time, so it's probably in the docs archive of the project site.

No such luck with the carhouse. Funding for that dragging on so long, it ended up being downsized, and it'll be the very last thing constructed because it's not a prerequisite for service starts. There aren't publicly available schematics of all the revisions they're doing because it's lower-priority project deliverables. All we really know is acreage: the 2 buildings south of 3rd Ave. bordering the tracks are going to get blown up, and 2 of the 4 thru tracks wrapping behind BET (plus one freight siding attached to one of the buildings) are going away.
 
I believe three tracks will remain between BET and the GLX Car house: CR east, westbounds and the 4th Iron. The Willey Track and the portion of the 3rd Iron which went to MS Walker are retired.

The original designs took no pity or had a clue on commuter operations, eliminated all but one through track.
 
Will likely never happen, but would be nice if the extended D would skip over low-ridership Science Park (which will only get served by the E, kind of like how Savin Hill is Ashmont-only). It'd save only a minute or so since the D would still have to slow down to platform speeds. But it would help the Tufts-to-Government Center travel times stay competitive with driving. GC/Haymarket/NS/Science Park are four stops in roughly a mile of track.
 
Will likely never happen, but would be nice if the extended D would skip over low-ridership Science Park (which will only get served by the E, kind of like how Savin Hill is Ashmont-only). It'd save only a minute or so since the D would still have to slow down to platform speeds. But it would help the Tufts-to-Government Center travel times stay competitive with driving. GC/Haymarket/NS/Science Park are four stops in roughly a mile of track.


Green Line isn't set up for regular expressing. It's only done when there's a headway to make up after a delay, where a train simply plows through a bunch of empty blocks to catch up with pace of traffic. There's no regular overtakes on rapid transit like there is commuter rail. Savin Hill isn't skipped; it's literally on a separate branch's tracks.


Science Park, while a low ridership outlier, has seen significant increases in recent years from all the nearby development. The gap used to be much wider. It would be one thing if it were stagnant, but it's actually growing at a healthy clip and should see that kick up a further notch with the multiple branch schedules coming via GLX.
 
Agreed speed is critical, but rebuilding the Lechemere viaduct (in the CIP, currently in design) and redoing signals/track in the Central Subway (in design) will help in that area without needing to skip any stops.
 
Will likely never happen, but would be nice if the extended D would skip over low-ridership Science Park (which will only get served by the E, kind of like how Savin Hill is Ashmont-only). It'd save only a minute or so since the D would still have to slow down to platform speeds. But it would help the Tufts-to-Government Center travel times stay competitive with driving. GC/Haymarket/NS/Science Park are four stops in roughly a mile of track.

Science Park is the closest GL stop to MGH, but it doesn't make sense for GL riders coming from the West to ride all the way through Downtown when getting off at Park and walking or switching to the Red Line would be faster. Once there's larger-than-token ridership GL coming from the North, I'd expect to see Science Park ridership increase for MGH access.
 
I recall way back in the day, the plan was to do Union Sq first and get that running. Is that still the case?
 
I recall way back in the day, the plan was to do Union Sq first and get that running. Is that still the case?

No. I think it’s sill scheduled to be completed at the sa,e time now, late 2021
 
No. I think it’s sill scheduled to be completed at the sa,e time now, late 2021

I think it was more than Union would finish before Medford because it's shorter - it wasn't a dependency. Union indeed is scheduled for a 2021 opening.
 
Quoting from what I wrote a while back on the Wikipedia article:

"In August 2012, the City of Somerville, MassDOT, and the MBTA reached a memorandum of agreement about the station. Through the Somerville Redevelopment Authority, the City will acquire $8 million worth of land for the station and grant the MBTA a permanent easement, while retaining the rights for transit-oriented development overhead. In return, the MBTA and MassDOT will pay for cleanup costs at the site, begin construction by the spring of 2014, and open the station no later than "late 2016-early 2017"

In October 2012, the Somerville Board of Aldermen approved the Union Square Redevelopment Plan and authorized an $8 million bond, including $6 million to purchase the land and $2 million for cleanup and station planning. In May 2013, the Board of Aldermen announced that the North Prospect block - a mostly industrial area bordered by the railroad tracks, Prospect Street, Somerville Avenue, and the rear of residential properties on Allen Street - had been acquired by the city via eminent domain for $4.5 million. The properties were to be vacated by August. The city also received a $1 million EPA grant to clean up one of the properties."

So yes, for a couple years the plan was for the Union Square Branch to be on an accelerated timeline. The city did their part, but the state did not.
 
w00t! Signs of life at New Lechmere. That looks like a viaduct piling-to-be at the split between new and old.
 
[Responding to Stick N Move's image posting in the Union Sq thread]
It isn't the architect's job to sweat the transit details in their render, and they obviously didn't (the tracks pop in and out of alignment and change in number as they go)

So I'm trying to remember if they are lining things up so that if you redo the bridge, the tracks can extend without re-doing the substation.

Or whether going onward to Porter Sq will require both a new bridge and a new substation to get out of the stub-ended station here?

The most recent PDF on the GLX site looks like an extension to Porter would interfere (overlap) by about 8' the substation. (https://www.mass.gov/files/documents/2019/02/05/glx_Union_Sq_Station_1218.pdf) Not quite as dire as the image below makes it look (this "non-technical" render makes the outbound GLX track seem pointed straight at the heart of the substation) but any "hit" is still just as costly.

01_Brick_Aerial_LandscapeRoof__5___1_.jpg
 
Last edited:
The tracks stub out at a portion of bridge wall where they can punch through the existing wall with an archway. Substation impact was surveyed and equipment-shifting would do the trick.


There might be 1 more bridge upstream needing modifications, but that's it until Beacon St. where the station tunnel incline starts.
 
So I'm reading the latest MassDOT construction update (https://www.mass.gov/info-details/glx-construction-activities-and-traffic-updates) and this jumped out at me:
Near Inner Belt Road workers completed welding together 6 miles of 60-foot-long sections of light-rail (1,700 tons total) that will be placed along the alignment next spring. In the near term we will be placing an additional 175 tons of commuter grade rail line this fall along the east side of the corridor.
There must be a good reason why they didn't just order a trainload or two of 1/4-mile sections of CWR and drop it along the alignment and weld it together in place, but I can't begin to think of what it might be.
 
So I'm reading the latest MassDOT construction update (https://www.mass.gov/info-details/glx-construction-activities-and-traffic-updates) and this jumped out at me:

There must be a good reason why they didn't just order a trainload or two of 1/4-mile sections of CWR and drop it along the alignment and weld it together in place, but I can't begin to think of what it might be.

It's the difference in wear that more frequent rapid transit exerts on the rails, plus HRT/LRT not being able to physically handle the kind of braun the RR's can at staging the equipment for a quarter-mile welded rail installation.

It's physically much harder to get an extreme-length ribbon of welded rail onto and across across the rapid transit system. The fact that the fence hasn't been erected yet between the Lowell Line and GLX ROW doesn't mean they'd go super-long today...because when the extension is open it's going to be hard to do the first rail replacement when the initial install is super-long rail.

  • You have to have an open connection to the RR network to get that rail it delivered. Only Green @ Riverside and Orange @ Wellington have that at the moment. A welded rail train wouldn't even fit into Riverside off the Worcester Line. And Orange's connection would require a backup move from the (9 years out-of-service) Medford Branch, with the quarter-mile rail train blocking the Middlesex Ave. grade crossing for 10 minutes while it lurches almost to Fellsway so the back of the train can clear the switch to Wellington for the reverse.

  • Subway tunnels don't have the clearance for the special-handling cars that make thousands of feet of rail transportable. And the tighter curves would make the acutal rail drop require extra human assistance to fold it out of the way of the active track.

  • Weight-restricted places such as Lechmere Viaduct or portions of any of the subways that may be anchored in soft soil won't be able to take the weight of the rail-handling cars or some of the heavy equipment for manipulating such enormous lengths. The Blue Line tunnel's under-Harbor crossing, for instance, is actually designed to flex back and forth below the water; that definitely cannot take overweighted work cars.

  • The B/C/E simply don't have lengths of track between grade crossings long enough to hold even one of the quarter-mile lengths. On commuter rail they drop it a ways back of any crossing clusters on the stretch of track they're eventually going to replace, but no such option exists on the surface Green because the crossings are so close there's no "off to the side" to be had.

  • Portions of the system have a density of switches and turnouts that also preclude super-long welded rail because the crossover pairs, track circuit insulation sections, etc. end up chopping the rail much sooner than the same structures would on commuter rail. It's a function of how much more extreme the headways are.

  • With traffic levels so much more extreme than RR, rail wears out a lot faster on the rapid transit div. Installing one of those quarter-mile ribbons on RR takes a big production, which is why the rail can be dropped on the ground up to 6 months before crews actually install it. Rapid transit is kept busy enough with rail replacements that they have to be able to make incremental progress across a swath of track on a regular night shift's staffing...since replacements are needed too much more often than the crew & equipment extravaganza the RR guys put on when they're doing a replacement. It's the difference between wear replacement being a regular chore vs. a special event. Rapid transit is organized such that they can make more hay being able to change out several 60-foot segments per night, while RR it's a considerably rarer task but one they get done with a tactical nuclear strike's worth of manpower and machinery.
 
You have to have an open connection to the RR network to get that rail it delivered.
I was assuming they'd run the rail train up the northbound Lowell Line track and drag the rails off to the GLX side.
Rapid transit is organized such that they can make more hay being able to change out several 60-foot segments per night, while RR it's a considerably rarer task but one they get done with a tactical nuclear strike's worth of manpower and machinery.
They're welding the segments together anyway (or so it sounds like), and I assume replacement involves cutting out the old segment and welding in a new one. Can you not do that with CWR?
 

Back
Top