MassDOT Rail: Springfield Hub (East-West, NNERI, Berkshires, CT-Valley-VT-Quebec)

Copying over the ongoing discussion about E-W rail and capacity from the SCR thread, which roughly starts here.

At which point does it become a good idea to add bypass tracks at local stations like the Wellesleys, so that Amtrak and Worcester express trains aren't necessarily held back by local trains? I believe the ROW and bridges have space for 4 tracks.
 
Teban54 said:
Copying over the ongoing discussion about E-W rail and capacity from the SCR thread, which roughly starts here.

At which point does it become a good idea to add bypass tracks at local stations like the Wellesleys, so that Amtrak and Worcester express trains aren't necessarily held back by local trains? I believe the ROW and bridges have space for 4 tracks.
Click to expand...
They're already doing that. Wellesley Farms to Framingham Jct. is getting a contiguous third center track, so expresses can skip all the intermediates.
 
I have a bad habit of watching cab-view videos from places like Switzerland, and a thing that's really struck me is that every station, even in tiny little burgs out in the middle of nowhere on a little single-track (but still electrified) line, has a siding for the platform.
(It's also really impressive how they do timed meets on single-track lines. Almost never have I seen a train wait more than 30ish seconds for the oncoming train to show up.)
 
Teban54 said:
A one-minute, not-serious proposal that's partially inspired by my comment on expanding rapid transit boundaries:

Extend the GL D branch from Riverside to Wellesley College via Wellesley Farms, Wellesley Hills and Wellesley Square. This allows Regional Rail to Framingham/Worcester to skip these stations and speed up their trips.

Or is a future Blue Line extension (via tunnel beneath Worcester Line to also serve the Newton stations) a better idea?
Click to expand...

Responding in this thread because it gets at something I've been thinking about.

Has anyone on this board mocked up a workable schedule that includes everything that's being thrown at the Worcester line?
  • New stations at West and Newton Corner
  • 15 minute headways inside 128
  • 30 minute headways from Wellesleys to Worcester
  • Functional service to Springfield
I was messing around with this in Excel (starting with the skip-stop generator @Riverside built). Given that from South Station to Riverside is mostly eternally limited to 2 tracks (unless I'm mistaken) it seems to require a tightly choreographed operation of at least 5 trains per hour on various skip-stop patterns, and probably 7 at peak hours.
 
737900er said:
Responding in this thread because it gets at something I've been thinking about.

Has anyone on this board mocked up a workable schedule that includes everything that's being thrown at the Worcester line?
  • New stations at West and Newton Corner
  • 15 minute headways inside 128
  • 30 minute headways from Wellesleys to Worcester
  • Functional service to Springfield
I was messing around with this in Excel (starting with the skip-stop generator @Riverside built). Given that from South Station to Riverside is mostly eternally limited to 2 tracks (unless I'm mistaken) it seems to require a tightly choreographed operation of at least 5 trains per hour on various skip-stop patterns, and probably 7 at peak hours.
Click to expand...
I haven't mocked up a schedule, but my thinking is that a lot of those frequencies will come from stacking services.

For example, something like this would satisfy your criteria:
  • 1 tph to Riverside (all stops)
  • 1 tph to Framingham (all stops)
  • 1 tph to Marlboro (all stops)
  • 1 tph to Worcester (skip Wellesleys and Naticks)
  • 1 tph "CommPass" service to Springfield and beyond, stopping at Framingham and Worcester

1700694363954.png


(This, incidentally, is why I think a Regional Rail/Indigo Line station at Riverside doesn't make a lot of sense. Even in this base level of service -- hourly all-stops to Marlboro, hourly semi-express to Worcester, hourly all-stops to Framingham -- gets you close to 4 tph to the Newtons. The 1 extra tph you need to reach that goal could probably be put to better use on the outer sections of the line.)

I tried mocking up this schedule, and yes it does require at least one overtake, I am pretty sure. In this mock-up, I've done the overtake in Wellesley -- don't know if you strictly speaking need the third track to do it.

1700699256134.png


That is indeed tight, but I don't see a huge reason why it has to be more than 5 tph. (Yes, Ashland <> Grafton get a slightly slower journey, by about 6 minutes. If there's huge concern, that Riverside short-turn could be extended out to Worcester to give them half-hour frequencies as consolation.)

But yeah: if you imagine the Framingham/Worcester Line as a pair of branches that diverge at 128 -- not unlike Providence/Stoughton or Newburyport/Rockport -- then 15 min freqs inside of 128 is just a matter of 2 tph on each "branch". Then the only question becomes how to slot in a CommPass train that skips the Newtons. (Although it would be interesting to try to imagine a schedule that sees a Springfield train stopping in Newton for a timed transfer to a bus to Waltham and other 128 employment hubs.) Slotting in the CommPass train is not trivial, but shouldn't be earth-shattering.
 
Riverside said:
I haven't mocked up a schedule, but my thinking is that a lot of those frequencies will come from stacking services.

For example, something like this would satisfy your criteria:
  • 1 tph to Riverside (all stops)
  • 1 tph to Framingham (all stops)
  • 1 tph to Marlboro (all stops)
  • 1 tph to Worcester (skip Wellesleys and Naticks)
  • 1 tph "CommPass" service to Springfield and beyond, stopping at Framingham and Worcester

View attachment 44880
Click to expand...
I know this is only a quick mock-up schedule, but the only issue I have is that local stations between Framingham and Worcester still only see 1-hour frequencies, while Worcester only gets 30-min frequencies because of CommPass that's likely operated by Amtrak, which I'm not sure if everyone would see as a substitute for MBTA Regional Rail due to higher fares.

But in a world without the Marlborough branch, it should be pretty easy to have 2 tph on the Wellesleys-Framingham local service and 2 tph on the Framingham-Worcester semi-express service (or, as you once suggested, a skip-stop pattern). You can probably extend one of the 2 local Framingham trips to Marlborough.

Of course, this requires doing away with Riverside, which I think is reasonable and even beneficial. If connectivity to the D branch is desired (questionable), I think your Norumbega Junction proposal makes more sense (assuming you can deal with Weston NIMBYs).

Riverside said:
(Although it would be interesting to try to imagine a schedule that sees a Springfield train stopping in Newton for a timed transfer to a bus to Waltham and other 128 employment hubs.)
Click to expand...
That may be what West Station can be used for. (Let's admit it, with its current plans, West Station is nothing more than an ordinary station like Boston Landing.)
 
Riverside said:
I haven't mocked up a schedule, but my thinking is that a lot of those frequencies will come from stacking services.

For example, something like this would satisfy your criteria:
  • 1 tph to Riverside (all stops)
  • 1 tph to Framingham (all stops)
  • 1 tph to Marlboro (all stops)
  • 1 tph to Worcester (skip Wellesleys and Naticks)
  • 1 tph "CommPass" service to Springfield and beyond, stopping at Framingham and Worcester

View attachment 44880

(This, incidentally, is why I think a Regional Rail/Indigo Line station at Riverside doesn't make a lot of sense. Even in this base level of service -- hourly all-stops to Marlboro, hourly semi-express to Worcester, hourly all-stops to Framingham -- gets you close to 4 tph to the Newtons. The 1 extra tph you need to reach that goal could probably be put to better use on the outer sections of the line.)

I tried mocking up this schedule, and yes it does require at least one overtake, I am pretty sure. In this mock-up, I've done the overtake in Wellesley -- don't know if you strictly speaking need the third track to do it.

View attachment 44884

That is indeed tight, but I don't see a huge reason why it has to be more than 5 tph. (Yes, Ashland <> Grafton get a slightly slower journey, by about 6 minutes. If there's huge concern, that Riverside short-turn could be extended out to Worcester to give them half-hour frequencies as consolation.)

But yeah: if you imagine the Framingham/Worcester Line as a pair of branches that diverge at 128 -- not unlike Providence/Stoughton or Newburyport/Rockport -- then 15 min freqs inside of 128 is just a matter of 2 tph on each "branch". Then the only question becomes how to slot in a CommPass train that skips the Newtons. (Although it would be interesting to try to imagine a schedule that sees a Springfield train stopping in Newton for a timed transfer to a bus to Waltham and other 128 employment hubs.) Slotting in the CommPass train is not trivial, but shouldn't be earth-shattering.
Click to expand...

Under the service you’ve provided, wouldn’t GLX to Auburndale provide far better service and connectivity than CRX to Riverside? Rather than 1 tph connecting Green and Common Rail service, you’d only be limited by the number of D-Branch trains per hour. Additionally, it frees up another CR/Amtrak slot.

For example, something like:
  • 10 tph to Auburndale via D-Branch of Green Line
  • 1 tph to Riverside (all stops)
  • 2 tph to Framingham (all stops)
  • 1 tph to Worcester (all stops)
  • 1 tph to Worcester (skip Wellesleys and Naticks)
  • 1 tph service to Springfield and beyond:
    • 80% to New Haven (or NYP), stopping at Back Bay, Landsdowne, Framingham, and Worcester.
    • 10% to Chicago, stopping at Back Bay and Worcester.
    • 10% to Toronto, stopping at Back Bay, Landsdowne, Framingham, and Worcester.
EDITED to strike through and accidental Riverside inclusion.
 
Last edited:
bigeman312 said:
Under the service you’ve provided, wouldn’t GLX to Auburndale provide far better service and connectivity than CRX to Riverside?
Click to expand...

Riverside was dying even because the CR is faster to where people want to go/commute. I believe that's what led the MBTA to want to develop the parking lots there (and the Hotel Indigo). People that wanted Green perhaps just go to Waban instead.

What you could do is something like this:
2 tph Worcester Line trains as is
2 tph Framingham Secondary trains that pick up the Wellesley stops but skip the Newton ones
2 tph Riverside trains that include the Newton stops
H2H (if you could somehow had room for it)

I don't think there is enough room to do this... let alone also include some amount of Springfield service.
 
jklo said:
I don't think there is enough room to do this... let alone also include some amount of Springfield service.
Click to expand...

That’s the problem. If you’re going to bump something, it would be the Riverside service, IMO.

Which leaves you with two options:
  1. Status quo of the Riverside terminus of the D-Branch and no direct transfer between the Green Line and Worcester Line, but plenty of indirect options:
    • 53, Auburndale <—> Woodland
    • MWRTA, Natick/Wellesley <—> Woodland
    • 59, Newtonville <—> Newton Highlands
    • Walking transfer between Landsdowne and Kenmore/Fenway
    • Waking transfer between Back Bay and Copley
  2. GLX to Auburndale
Both are fine. Obviously status quo is cheaper. Both options are better than CRX to Riverside hoarding a slot that prevents service from being expanded to points west, all while not providing any more frequency to inner stations than if the branch didn’t exist.
 
Last edited:
Teban54 said:
But in a world without the Marlborough branch, it should be pretty easy to have 2 tph on the Wellesleys-Framingham local service and 2 tph on the Framingham-Worcester semi-express service (or, as you once suggested, a skip-stop pattern). You can probably extend one of the 2 local Framingham trips to Marlborough.
Click to expand...
bigeman312 said:
Additionally, it frees up another CR/Amtrak slot.
Click to expand...
Right, yeah, apologies for not being clearer -- my point was that even at these (too) low levels of service, you're very close to 15 min headways within 128. I agree, an ultimately better service pattern would be something like this:

1700747738233.png


(I mean, this is the actually sorta remarkable thing about the Regional Rail network: except for Fitchburg, and assuming you use Wildcat, all of the trunks run at "double" service levels before splitting around 128. Meaning, 30 min headways to each of Rockport, Newburyport, Haverhill, Lowell, Framingham/Marlboro, Worcester, Franklin, Foxboro, Providence, Stoughton, and the Old Colony Lines, and you're at 15 minute headways within 128 almost everywhere. The short-turn services are supplemental.)

I tried sketching up the schedule, and yeah the Springfield service would require an additional overtake, and that's a headache to plan. But also seems like it could be readily doable, especially if we are okay with the reverse peak schedule getting slightly disrupted from a clockfacing perspective.
jklo said:
What you could do is something like this:
2 tph Worcester Line trains as is
2 tph Framingham Secondary trains that pick up the Wellesley stops but skip the Newton ones
2 tph Riverside trains that include the Newton stops
H2H (if you could somehow had room for it)

I don't think there is enough room to do this... let alone also include some amount of Springfield service.
Click to expand...
I don't see why Marlboro trains would need to skip the Newton stops?

I assume that a H2H service would get consolidated into a Springfield service. I also think there's some possibility of a CommPass service stopping at Grafton <> Ashland on one or two of its morning runs; there's a weird "anti-sweet spot" where the train will have to leave Springfield at such an ungodly hour that the ~10-minute penalty from those stops isn't really going to make such a huge difference. For example, Springfield trains scheduled to arrive in Boston at 7am and 8am will need to leave Springfield at 5am and 6am respectively; moving those departures back to 4:50 and 5:50 won't really change anyone's mind about taking the train at that hour, compared to later commutes intended to leave Springfield in the more reasonable 7am and 8am hours where I suspect there will be more sensitivity to travel time.
Teban54 said:
while Worcester only gets 30-min frequencies because of CommPass that's likely operated by Amtrak, which I'm not sure if everyone would see as a substitute for MBTA Regional Rail due to higher fares.
Click to expand...
I would not take that for granted -- Amtrak's pricing is weird these days because they adjust prices based on demand, but you can pretty regularly get a cheaper ticket on Amtrak between Boston and Providence than you can on the T (which is just absolutely bonkers, but I digress).
Teban54 said:
If connectivity to the D branch is desired (questionable), I think your Norumbega Junction proposal makes more sense (assuming you can deal with Weston NIMBYs).
Click to expand...
Appreciate the positive words, but I admit that at this point I think it makes more sense just to extend the D Line to Auburndale -- I think there is space there and it would require less construction and avoids Weston altogether literally and figuratively.

A couple of crazy pitches that are fun to imagine on the map:
  • a resurrected Newton Lower Falls shuttle, with transfers both to the D Line at Riverside and to Regional Rail at Auburndale
  • build an Aldgate Junction at Cook St (why is it called Cook St? Am I misremembering?) and run a shuttle from Auburndale to Needham (or just to TV Place/128)
bigeman312 said:
Under the service you’ve provided, wouldn’t GLX to Auburndale provide far better service and connectivity than CRX to Riverside? Rather than 1 tph connecting Green and Common Rail service, you’d only be limited by the number of D-Branch trains per hour. Additionally, it frees up another CR/Amtrak slot.
Click to expand...
Yeah, this was what I concluded when thinking about Norumbega Junction (linked by Teban54 above) -- it's better to bring the Green Line to Regional Rail rather than the reverse. I think Riverside could be useful as a short-turn location, but like I said above, I don't see it as a service with any high frequency.
Teban54 said:
That may be what West Station can be used for. (Let's admit it, with its current plans, West Station is nothing more than an ordinary station like Boston Landing.)
Click to expand...
Maybe? But like West Station is almost as far from 128 as, like West Natick is? So I would think that Auburndale would make for a better transfer location.
 
Riverside said:
Right, yeah, apologies for not being clearer -- my point was that even at these (too) low levels of service, you're very close to 15 min headways within 128. I agree, an ultimately better service pattern would be something like this:

View attachment 44892

(I mean, this is the actually sorta remarkable thing about the Regional Rail network: except for Fitchburg, and assuming you use Wildcat, all of the trunks run at "double" service levels before splitting around 128. Meaning, 30 min headways to each of Rockport, Newburyport, Haverhill, Lowell, Framingham/Marlboro, Worcester, Franklin, Foxboro, Providence, Stoughton, and the Old Colony Lines, and you're at 15 minute headways within 128 almost everywhere. The short-turn services are supplemental.)

I tried sketching up the schedule, and yeah the Springfield service would require an additional overtake, and that's a headache to plan. But also seems like it could be readily doable, especially if we are okay with the reverse peak schedule getting slightly disrupted from a clockfacing perspective.

I don't see why Marlboro trains would need to skip the Newton stops?

I assume that a H2H service would get consolidated into a Springfield service. I also think there's some possibility of a CommPass service stopping at Grafton <> Ashland on one or two of its morning runs; there's a weird "anti-sweet spot" where the train will have to leave Springfield at such an ungodly hour that the ~10-minute penalty from those stops isn't really going to make such a huge difference. For example, Springfield trains scheduled to arrive in Boston at 7am and 8am will need to leave Springfield at 5am and 6am respectively; moving those departures back to 4:50 and 5:50 won't really change anyone's mind about taking the train at that hour, compared to later commutes intended to leave Springfield in the more reasonable 7am and 8am hours where I suspect there will be more sensitivity to travel time.

I would not take that for granted -- Amtrak's pricing is weird these days because they adjust prices based on demand, but you can pretty regularly get a cheaper ticket on Amtrak between Boston and Providence than you can on the T (which is just absolutely bonkers, but I digress).

Appreciate the positive words, but I admit that at this point I think it makes more sense just to extend the D Line to Auburndale -- I think there is space there and it would require less construction and avoids Weston altogether literally and figuratively.

A couple of crazy pitches that are fun to imagine on the map:
  • a resurrected Newton Lower Falls shuttle, with transfers both to the D Line at Riverside and to Regional Rail at Auburndale
  • build an Aldgate Junction at Cook St (why is it called Cook St? Am I misremembering?) and run a shuttle from Auburndale to Needham (or just to TV Place/128)

Yeah, this was what I concluded when thinking about Norumbega Junction (linked by Teban54 above) -- it's better to bring the Green Line to Regional Rail rather than the reverse. I think Riverside could be useful as a short-turn location, but like I said above, I don't see it as a service with any high frequency.

Maybe? But like West Station is almost as far from 128 as, like West Natick is? So I would think that Auburndale would make for a better transfer location.
Click to expand...

Exactly! “It’s better to bring the Green Line to Regional Rail rather than the reverse.”

That being said, and in the interest of this not being a ‘Crazy Transit Pitch’ thread, what does a full-build service pattern without the inclusion of a Northborough Branch look like? The best I came up with was:
  • 10 tph to Auburndale (or Riverside) via D-Branch of Green Line
  • 2 tph to Framingham (all stops)
  • 1 tph to Worcester (all stops)
  • 1 tph to Worcester (skip Wellesleys and Naticks)
  • 1 tph service to Springfield and beyond:
    • 80% to New Haven (or NYP), stopping at Back Bay, Landsdowne, Framingham, and Worcester.
    • 10% to Chicago, stopping at Back Bay and Worcester.
    • 10% to Toronto, stopping at Back Bay, Landsdowne, Framingham, and Worcester.
Is that realistic and/or feasible?
 
Riverside said:
I don't see why Marlboro trains would need to skip the Newton stops?
Click to expand...
The 2002 feasibility study said travel times between Framingham and Northborough/I-290 would be 25 minutes making stops at Framingham Center/FSU, Route 9, Marlborough/I-495, and Northborough/I-290 (2 fewer stops than your spider map). On top of 50 minutes Boston-Framingham making all stops, so 1:15 total. That obviously gets faster with EMU's, so a target of 1:05 is probably appropriate.
 
bigeman312 said:
Exactly! “It’s better to bring the Green Line to Regional Rail rather than the reverse.”

That being said, and in the interest of this not being a ‘Crazy Transit Pitch’ thread, what does a full-build service pattern without the inclusion of a Northborough Branch look like? The best I came up with was:
  • 10 tph to Auburndale (or Riverside) via D-Branch of Green Line
  • 2 tph to Framingham (all stops)
  • 1 tph to Worcester (all stops)
  • 1 tph to Worcester (skip Wellesleys and Naticks)
  • 1 tph service to Springfield and beyond:
    • 80% to New Haven (or NYP), stopping at Back Bay, Landsdowne, Framingham, and Worcester.
    • 10% to Chicago, stopping at Back Bay and Worcester.
    • 10% to Toronto, stopping at Back Bay, Landsdowne, Framingham, and Worcester.
Is that realistic and/or feasible?
Click to expand...
This is what I got, playing around with this:

1700759909264.png


(I don't think it makes a huge difference to the scheduling whether the Springfield service stops at Lansdowne and Framingham, so I didn't fuss with creating the one-off service patterns.)

FWIW, the hard part is reconciling the overtakes and the clockfacing. I tried, for example, to have even 20-minute headways from Framingham inbound, but the problem then becomes ensuring there is sufficient padding between trains occupying the platforms at Worcester and Framingham. The 8-minute gap between the Springfield and Worcester Local trains departing Worcester seemed doable to me based on the 2nd platform being free while the Worcester Local turns, but I'd need to build out the whole outbound timetable to validate that there'd be a free platform at Framingham to handle close departures.

Basically the problem is that the express trains start in one clockfacing "slot" but gradually move into a separate one. This was actually part of how I came to propose the skip-stop model as a (short-term) adjustment to increase frequencies without needing extra equipment -- it's the homogeneity of the schedule that makes things work.
F-Line to Dudley said:
The 2002 feasibility study said travel times between Framingham and Northborough/I-290 would be 25 minutes making stops at Framingham Center/FSU, Route 9, Marlborough/I-495, and Northborough/I-290 (2 fewer stops than your spider map). On top of 50 minutes Boston-Framingham making all stops, so 1:15 total. That obviously gets faster with EMU's, so a target of 1:05 is probably appropriate.
Click to expand...
Yeah that makes sense. From what I can see, skipping the Newtons only saves 5 minutes, and I don't quite see 1h15m as so egregious to merit that savings at the cost of near turn-up-and-go freqs to the Newtons.
 
Riverside said:
Yeah that makes sense. From what I can see, skipping the Newtons only saves 5 minutes, and I don't quite see 1h15m as so egregious to merit that savings at the cost of near turn-up-and-go freqs to the Newtons.
Click to expand...
The EMU's will make all the difference there. Which is why you probably don't want to attempt Fitchburg Secondary RER until the main is fully electrified. The better acceleration out of station stops will shorten the time on the main lots, and it'll be crucial for treating all of the numerous curves on the Fitchburg Sec. with better re-acceleration times off of curve speed penalties. 1:05 to 290 might even be a little conservative. Even hour with a couple more intermediate stops seems possible given how many curves would have better accel recovery on the branch.
 
Riverside said:
This is what I got, playing around with this:

View attachment 44894

(I don't think it makes a huge difference to the scheduling whether the Springfield service stops at Lansdowne and Framingham, so I didn't fuss with creating the one-off service patterns.)

FWIW, the hard part is reconciling the overtakes and the clockfacing. I tried, for example, to have even 20-minute headways from Framingham inbound, but the problem then becomes ensuring there is sufficient padding between trains occupying the platforms at Worcester and Framingham. The 8-minute gap between the Springfield and Worcester Local trains departing Worcester seemed doable to me based on the 2nd platform being free while the Worcester Local turns, but I'd need to build out the whole outbound timetable to validate that there'd be a free platform at Framingham to handle close departures.

Basically the problem is that the express trains start in one clockfacing "slot" but gradually move into a separate one. This was actually part of how I came to propose the skip-stop model as a (short-term) adjustment to increase frequencies without needing extra equipment -- it's the homogeneity of the schedule that makes things work.

Yeah that makes sense. From what I can see, skipping the Newtons only saves 5 minutes, and I don't quite see 1h15m as so egregious to merit that savings at the cost of near turn-up-and-go freqs to the Newtons.
Click to expand...

This is great! Thank you so much.
 
MassDOT gets a $500K federal grant to begin service development planning on a new Boston-Albany Amtrak route:
mass.streetsblog.org

MassDOT Wins More Federal Funds to Plan Increased Boston-Albany Amtrak Service - Streetsblog Massachusetts

Last week, the Biden administration’s Federal Railroad Administration awarded a $500,000 grant to MassDOT to plan for additional Amtrak service between Boston and Albany. “The proposed Corridor would connect Boston, MA, and Albany, NY, via Springfield, MA. The proposed Corridor would provide up...
mass.streetsblog.org mass.streetsblog.org
 
StreetsblogMASS said:
MassDOT gets a $500K federal grant to begin service development planning on a new Boston-Albany Amtrak route:
mass.streetsblog.org

MassDOT Wins More Federal Funds to Plan Increased Boston-Albany Amtrak Service - Streetsblog Massachusetts

Last week, the Biden administration’s Federal Railroad Administration awarded a $500,000 grant to MassDOT to plan for additional Amtrak service between Boston and Albany. “The proposed Corridor would connect Boston, MA, and Albany, NY, via Springfield, MA. The proposed Corridor would provide up...
mass.streetsblog.org mass.streetsblog.org
Click to expand...

In theory we can walk and chew gum at the same time, but shouldn't there be an ungodly number of Inland Route frequencies before we consider anything close to eight daily round trips between Boston and Albany?
 
You must log in or register to reply here.

Back
Top