Design a Better Boston Back Bay Station

[Commuting Boston Student]

Boston Back Bay is not a bad station, strictly speaking. While I never actually use Back Bay, except for the Orange Line, I think it's an excellent station in dire need of some improvements.

I envision the future Back Bay as 'Boston West Station' - while a North-South Rail Link will enable through running north and south, there's no way to through run east and west without fundamentally changing South Station's track alignment or bypassing it entirely. So, the future Back Bay would be Boston's only stop along a Pittsfield-Springfield-Worcester-Boston-Hyannis rail line. (The restored Clamdigger.)

One of the most glaring flaws I think is present in Back Bay, even above the actively hazardous air quality on the Commuter Rail platform, is the track arrangement. This came up in another thread recently, but the Orange Line nicely cuts down the middle of Back Bay's five rail tracks, forcibly separating the Worcester Line and the NEC and preventing trains from crossing over west of the station. This is a huge problem for the artificial capacity constraints it creates - there are likely four to five times as many trains using the NEC platforms than there are Worcester Line platforms. Now, my proposal to fix this is to dig out another level beneath the existing platforms, and sink the Orange Line down to that level. This has the added benefit, besides creating room for two more Commuter Rail and Amtrak tracks, of creating a lower level subway platform that can be used or widened out for a future HRT line that parallels the Pike corridor.

After boosting up the available non-RT trackage to 7, and adding switches for trains to move between the two ROWs west of the station, Back Bay will load balance out much better with South Station, increasing breathing room against imaginary 'capacity constraints.'

I'm sure there are other fixes we can and should make, so I want to hear from you guys.

 

[HenryAlan]

I don't think the placement of the Orange Line is an issue. There could be no crossover west of the station because the NEC immediately turns Southward while the Worcester line continues Westward.

 

[Commuting Boston Student]

I don't think the placement of the Orange Line is an issue. There could be no crossover west of the station because the NEC immediately turns Southward while the Worcester line continues Westward.

No, I think you could have the center platforms formerly occupied by the Orange Line served by either Worcester, NEC or any of the other three commuter lines through Back Bay. The turn is not so immediate that there would not be room to install a track switch series to allow for moves from Worcester to the NEC and vice versa.

I think there's significant benefit to being able to send any train to any platform.

 

[Matthew]

Two tracks are more than sufficient to handle 30+ tph in each direction, given proper signalling, rolling stock, operations etc. The latter is what should be worked on.

 

[F-Line to Dudley]

No, I think you could have the center platforms formerly occupied by the Orange Line served by either Worcester, NEC or any of the other three commuter lines through Back Bay. The turn is not so immediate that there would not be room to install a track switch series to allow for moves from Worcester to the NEC and vice versa.

I think there's significant benefit to being able to send any train to any platform.

What is the cost/benefit ratio to doing this, though? You realize the Pike and the Orange Line portal were constructed more or less simultaneously...'65 and '67. The station itself was held up by the SW Expressway's fate, but the current layout was the provisioned Plan B if the Expressway was a no-go. Air rights, locations where building foundations could be driven...it was all provisionally integrated into the same mid-60's design. So it's not like the area is a hodgepodge of different-era development. Fundamentally altering the layout means taking on a lot more sunk cost in maneuvering around the integrated structural dependencies of that layout. This is why the N-S Link design spits its portal out on the Washington-Harrison block in the middle of the NEC/Worcester track gap. You'd be talking a couple $B more if the thing added platforms or otherwise blew up and rearranged stuff in the station area.

EXTREME cost. To what end? A minor simplification of the layout and pedestrian convenience? A minor capacity increase (but not nearly enough for new platforms)? I think you've got to draw the line somewhere on these minor fussing-around jobs when they start getting counted in $B's. Back Bay can't fundamentally be remade without a $B. Back Bay curve can't fundamentally be remade without a $B.

And intercity trains efficiently turn back at terminals every hour of the day, every day of the week. Northeast Regionals continue south of D.C. several trips a day on branches to Newport News and Lynchburg. They reverse directions AND pull an engine swap from electric to diesel on the platform at D.C. Union Station with no undue schedule penalty. The relocated Vermonter will pull a reverse (like it used to pre-1989) at Springfield Union Station because the station is about 200 feet down the B&A from the junction with the Springfield/Conn River Line. They've been doing that for 100 years. Just because North Station and South Station dispatch don't have to do that right now and are well out-of-practice on needing to...doesn't mean it can't be done quite efficiently. It was done every day, many times a day, back when those stations were bigger (and, yes, I realize SS expansion has to happen before there'd ever be enough slots to try it). If Amtrak had a need for it...say, hypothetical thru-routing a Downeaster or two to Springfield or Hartford via the Grand Junction...they know how to dispatch that so it doesn't screw things up. It's not a foreign concept at all. Not everything has to be on linear track in every direction at $B's in retrofit to work. The art of efficient dispatching goes way deeper than the simplistic one-size-fits-all the T practices on its relatively idiot-proof linear routes. They don't even bother shooting for an effective sweet spot of short-turns on the schedule.


I think if they want to make Back Bay better there needs to be some understanding reached about what trains stop there. The T's fixated on every southside line physically capable of reaching it having a direct one-seat there, and its N-S Link plans of course push the untenable EVERYTHING thru-routing scheme. BB matters to Amtrak because of the number of business travelers getting off in the Back Bay are a significant factor on the Acela and Regionals ridership, so they aren't going to budge (and they dispatch the NEC--not the T--so they hold the cards here).

Ultimately, the Franklin Line is going to have to compromise. The Foxboro proposal thru-routes everything over the Fairmount. I think regular Franklin commuters are also going to have to accept that if they want more frequent service levels they're going to have to give up their one-seat to BB and zip down the Fairmount. Maybe half the trains at first, but eventually it's going to make sense to pull all of them off the NEC. And the T has to figure out what it's going to do about the Needham Line. It's a drag on NEC congestion because it stops at BB, Ruggles, and Forest Hills like a rapid transit line would, but hardly anyone uses those 3 stops. It was supposed to be rapid-transit converted before they pulled a cut-and-run on Rozzie and W. Rox in the 80's. It isn't well-suited to "Fairmounting" because it'll never get the necessary schedule slots on the NEC. Ultimately they're going to have to move forward with the rapid transit conversion or it's going to be a real fly in the ointment by 2030.

This wouldn't be bad...Providence, Worcester, Stoughton/South Coast, and Amtrak. Hopefully Providence and Worcester both electrified by then. *Judicious* thru-routing of prime northside routes if the Link gets built. Blow up the useless Ruggles CR platform so there's less penalty coming in/out of the curve. And then work on getting a real Silver Line Phase III or NS-SS rapid transit connection to lessen the need for the god-given right for every line to stop for the direct Orange transfer. That's what's going to do it. Not spending $B's fussing with the not-stratospherically-improvable BB + curve infrastructure. Bang-for-buck...it just doesn't rate vs. the whole arsenal of other things big and little they could do.

 

[Commuting Boston Student]

I'm going to admit I hadn't really considered kicking Franklin out and dropping Needham CR, though I really should have.

How DO direction reversals happen anyway? I can't imagine it's as simple as stopping the train and throwing it into reverse.

 

[F-Line to Dudley]

I'm going to admit I hadn't really considered kicking Franklin out and dropping Needham CR, though I really should have.

How DO direction reversals happen anyway? I can't imagine it's as simple as stopping the train and throwing it into reverse.

Exactly the same way the rapid-transit trains do...engineer changes ends at the terminal and gets in the cab at the opposite end. Get out, lock the cab, walk down the platform, unlock the cab, get in. On all outbound MBCR trips the locomotive is the operating cab and locomotive is in pull mode. On all inbound MBCR trips the rear control car is the operating cab and the locomotive is in push mode (that way the diesel exhaust is always facing out at NS and SS). Same exact controls on either end, even though one of the cabs is in a locomotive and one of the cabs is in a crew compartment at the end of a coach. So it would actually be extremely easy to do a multiple-destination trip on the commuter rail because the train is always pointed the same direction at the terminal and the engineer is always making the same cab switch after pulling into the platform.

With Amtrak they can do it a couple ways. Long distance trips often go with locomotives at both ends of the train, rear one usually turned off (saves a refueling if only one engine is working one direction, and saves a breakdown if an engine quits hundreds of miles away from a replacement). They can go without a rear control coach in the consist if there's locos at both ends. They do that on the Vermonter most of the time as well so the awkward reverse move in Palmer is as fast as possible, and probably will continue doing that on the new route in winter because the forward-facing loco can plow a lot more snow off the track than a control coach can. On the mixed-diesel/electric Regionals they sometimes leave the diesel attached but turned off in electric territory, then fire it up if they hit the end of the wires in New Haven on the Springfield Line or stick it on the opposite end for the reverse in D.C. to/from Virginia. Whichever mode is unused stays off. And, yes, you can drive the diesel from the electric's cab and vice versa just like a control coach would, so it doesn't truly matter which direction the active engine is facing. And, yes, you can do an on-the-fly engine switch where one button switches engine control, drops/raises pantograph, turns on/off the diesel, etc.

Occasionally when locos are at a premium they'll do a more delicate maneuver where the electric always faces in at Union Station, they fetch the diesel and back it in from the yard while passengers are loading/unloading, and then they couple it after the doors close and pull out. Either leaving the electric behind at the platform or dragging it out, stopping for 30 seconds in the yard, and jettisoning it there. It's a shift-change point anyway, so the engineer stays with his/her respective engine instead of switching. They need an extra control car to pull that off because the 2nd cab is needed for redundancy.


They're good at this because steam locomotives were ONLY unidirectional and had to get yanked out and either spun around on a turntable to the reverse direction or hot-swapped on the platform. At every terminal. Even at the last stop in some podunk town. Dispatchers have been coordinating these same types of moves for 150 years. It's more than old hat.

 

[Commuting Boston Student]

Oh.

So, wait, doesn't that mean all the seats are facing the wrong way once the train is reversed? (Minor quibble, I know, but...)

 

[HenryAlan]

Oh.

So, wait, doesn't that mean all the seats are facing the wrong way once the train is reversed? (Minor quibble, I know, but...)

On most trains the coaches are configured so that all seats face the middle of the car, meaning that half are always facing backwards.

 

[Nexis4jersey]

I would push for a North - South Tunnel which would mean you wouldn't have to expand South Station. I would electrify the whole MBTA Commuter Rail network except the Fitchberg line , high level platform every station similar to the MTA's Network and buy EMU's which would mean the air quality at BB would improve. Next I would remove the current station completely and reconstruct it. It would be open to the Elements similar to Newark Penn station with some protection in the form of Platform canopies with a slits in the canopy to block out most of the rain yet still vent diesel fumes and allow in Fresh air. The Waiting area would be redone into and allow in more light....

 

[Commuting Boston Student]

I would push for a North - South Tunnel which would mean you wouldn't have to expand South Station. I would electrify the whole MBTA Commuter Rail network except the Fitchberg line , high level platform every station similar to the MTA's Network and buy EMU's which would mean the air quality at BB would improve. Next I would remove the current station completely and reconstruct it. It would be open to the Elements similar to Newark Penn station with some protection in the form of Platform canopies with a slits in the canopy to block out most of the rain yet still vent diesel fumes and allow in Fresh air. The Waiting area would be redone into and allow in more light....

Why electrify everything EXCEPT for Fitchburg?

 

[Nexis4jersey]

Why electrify everything EXCEPT for Fitchburg?

Actually since the Knowledge Corridor will eventually be Electrified , it would make sense to Electrify the Fitchburg line if they extended it to Greenfield.

 

[F-Line to Dudley]

Why electrify everything EXCEPT for Fitchburg?

Consolidated infrastructure. You can get to Fitchburg equally fast on an electric by zipping up the (HSR-supportable) Lowell Line, then cutting down Pan Am's Stony Brook Branch from North Chelmsford to Ayer and points west. In fact, I bet if HSR to Montreal ever happens that's exactly where it's gonna go because the route through NH on the Feds' HSR vision map has far too many engineering challenges to tolerable speeds and goes through a whole lot of nowhere past Concord. Leaves Waltham in a transit lurch, but that's solvable with a thicker net of other solutions. The combined Belmont-Littleton run isn't going to return ridership worth the $1/3B cost of another Link portal dig, and the wide station spacing between 128 (after the 3 Weston stops get consolidated) and Ayer lessens the performance difference between electrics and (good) diesels.


I don't think we truly need a 100% electric commuter rail to begin with when you see how much the traffic skews to only a few corridors. And a 100% EMU commuter rail is physically impossible because we have freight clearance routes that require mini-high platforms because there's no room for passing tracks. And all existing EMU models are high floor, with the T needing to gorge itself on another dubiously expensive Frankenstein car design if it wanted EMU's with low-platform stairs.

-- Fairmount Line: of course..."rapid transit-like" service, T and Amtrak needing to make a lot of non-revenue moves with electrics between Southampton Yard and Readville without fouling the NEC, and safety valve for service disruptions on the NEC. I wouldn't build that stupid Link portal, though. If anything here has to go through the tunnel, relocate the line behind South Bay plaza and combine it into the Old Colony portal.
-- Worcester Line: of course...intercity, 2nd-highest projected CR ridership and schedule density on the system, potential HSR candidate. Tri-trackable the whole length outside 128 so the new-construction stations Framingham-Grafton can be modified with high platforms (at some $$$) while allowing passing tracks for wide freights.
-- Needham: no. Or rather, electrify it as Orange and Green Line halves, because it's needed to be rapid transit for decades and has no extra gear it can reach for with CR service levels given the thorny NEC congestion issues.
-- Franklin: no. Station spacing is wide except for the pair of downtown Norwood stations that need consolidation...gap between diesel and electric performance is smaller here vs. closer-packed stops. Freight clearance route from Walpole to Readville (and Walpole-Foxboro if that's a new passenger branch), with no room to tri-track...platforms have to stay mini-high from Endicott-Windsor Gardens + Foxboro, so EMU's can't run. Too many potential branches to electrify with Foxboro, Milford, Woonsocket in the mix on timeframes the N-S Link would be built, and the branches split way further out of town (Walpole and Franklin) than the 128-area splits on the Old Colony and Eastern Route. Can--and should--run exclusively via Fairmount to save NEC congestion and keep the fumes out of the tunnel. And if they want to use the Link: dual-mode push-pull locos instead of EMU's.
-- Lowell/Haverhill via Anderson: yes. Both are freight mains bound by mini-high platforms, so they might have to use push-pull electrics instead of EMU's, but major intercity corridors and 3rd densest ridership/schedule growth on CR after Providence and Worcester. If West Medford can eventually be flipped to Green Line, everything Wedgemere-Mishawum and Wilmington-North Billerica can support 3rd passing tracks for platform raisings and EMU's (albeit for more station mod $$$ on those older stops vs. the modern-construction outer Worcester stations). Haverhill can't support 3 tracks so may have to remain push-pull, but still a definite yes for electrics because of the Downeaster.
-- Reading: no. Orange Line it like it was supposed to all along. Too much redundant electrification if CR gets that treatment, it's a slower route than the NH Main, and EMU's could only be used to the Reading short-turn because of the unexpandable mini-highs further north.
-- Eastern Route: yes. Currently a non-clearance route for freight because of the high Lynn platforms, and no freight whatsoever north of Salem. Heavy, dense ridership and growth potential. Dense station spacing tunnel-Beverly and on Rockport branch. Diesel fumes a factor with all the downtown "bedroom communities" and sensitive environmental land it passes through. Relatively short branches to electrify, entire main + Peabody lined by high-tension power lines to feed substations.
-- Fitchburg: no. Because as noted that extra N-S Link portal is stupid expensive for the limited upside, and everything past Littleton is reachable by spurring off the end of the Lowell electrification. Also can't use EMU's at Shirley or North Leominster (well, could, but bang-for-buck would be questionable) because of the freight, unexpandable mini-highs, lack of room for passing tracks.
-- Old Colony: push/lean-no. I don't think that extra Link portal is worth building for a long time extra because the Cape is reachable via Stoughton/Taunton, Greenbush is a short stub that will never be all it was cracked up to be, and Kingston/Plymouth's an unexpandable stub. At least put it off till last here.
-- Stoughton: push/partial-only. Station spacing on South Coast Rail is long, diesels would perform well. Ditto if Cape service is thru-routed here (negates need to build Old Colony Link portal). Service density would only pass the tipping point for electrics on the main to Taunton, gets diffuse after branches split (but you could do it to Taunton and use push-pull dual-modes for thru service and EMU's for short-turns). Branches split outside of 495, furthest out of any line. Like with most things SCR, bang-for-buck far less than awesome (so of course it's the very first one they're rushing to electrify!)

 

[Commuting Boston Student]

Consolidated infrastructure. You can get to Fitchburg equally fast on an electric by zipping up the (HSR-supportable) Lowell Line, then cutting down Pan Am's Stony Brook Branch from North Chelmsford to Ayer and points west. In fact, I bet if HSR to Montreal ever happens that's exactly where it's gonna go because the route through NH on the Feds' HSR vision map has far too many engineering challenges to tolerable speeds and goes through a whole lot of nowhere past Concord. Leaves Waltham in a transit lurch, but that's solvable with a thicker net of other solutions. The combined Belmont-Littleton run isn't going to return ridership worth the $1/3B cost of another Link portal dig, and the wide station spacing between 128 (after the 3 Weston stops get consolidated) and Ayer lessens the performance difference between electrics and (good) diesels.

I don't think HSR to Montreal can happen any other way than through New Hampshire because the force of the potential Boston-Manchester-Concord market is not strong enough to justify the NH Capitol Corridor without a non-zero level of support for an eventual extension to Montreal. So, emphatically killing Boston-Concord-Montreal (at least before Boston-Concord is fully operational) can't happen or Capitol Corridor is dead in the water.

But let's go ahead and talk about this some more. What are the engineering challenges for Concord-Montreal? Why wouldn't cutting a new ROW west to 89 from just north of Concord, then following I-89 up through stops at Lebanon, Montpelier, Burlington VT, St. Albans? The only real problem area I can see is in the vicinity of the actual border crossing.

 

[Nexis4jersey]

The Terrain between Boston and Montreal effectively cancels out a HSL , unlike the proposed NY to Montreal line which deals with relativity flat land.

 

[Commuting Boston Student]

The Terrain between Boston and Montreal effectively cancels out a HSL , unlike the proposed NY to Montreal line which deals with relativity flat land.

So, reconfigure the terrain.

 

[Nexis4jersey]

So, reconfigure the terrain.

You would need to punch through it , which can be expensive.... For a Boston-Montreal High Speed line 150mph+ you'll need at least 40 tunnels some over 7 miles long. You'll need Urban and Suburban tunnels to keep the NIMBYs happy those range from 1000ft to 1 mi....while the Rural Mountainous tunnels can go up to 7 miles.

 

[Commuting Boston Student]

You would need to punch through it , which can be expensive.... For a Boston-Montreal High Speed line 150mph+ you'll need at least 40 tunnels some over 7 miles long. You'll need Urban and Suburban tunnels to keep the NIMBYs happy those range from 1000ft to 1 mi....while the Rural Mountainous tunnels can go up to 7 miles.

Or you could use the Interstate 89 ROW, which features pre-blasted mountains and no need to cater to NIMBYs.

 

[F-Line to Dudley]

I don't think HSR to Montreal can happen any other way than through New Hampshire because the force of the potential Boston-Manchester-Concord market is not strong enough to justify the NH Capitol Corridor without a non-zero level of support for an eventual extension to Montreal. So, emphatically killing Boston-Concord-Montreal (at least before Boston-Concord is fully operational) can't happen or Capitol Corridor is dead in the water.

But let's go ahead and talk about this some more. What are the engineering challenges for Concord-Montreal? Why wouldn't cutting a new ROW west to 89 from just north of Concord, then following I-89 up through stops at Lebanon, Montpelier, Burlington VT, St. Albans? The only real problem area I can see is in the vicinity of the actual border crossing.

The ex-B&M Northern Route has over 100 grade crossings and clusters crossing state highways and residential streets at odd angles. It was an awful route when it last saw passenger service, and many of those simply can't be bridged because of proximity to abutters and those odd angles. They had a thread on this on RR.net comparing BOS-MTL travel times on the multiple old routes that used to run, and Fitchburg Line to Greenfield was more or less a match for the Northern Route back in the day. And that part of NH truly is in the middle of nowhere, with no potential for any local passenger traffic or any new freight. Total, utter nonstarter.

I-89 is reachable from Concord by an abandoned ROW due west from downtown to Hopkinton. HOWEVER, somebody has to successfully demonstrate that HSR can be grafted onto an interstate median without a cost premium into the billions. It's not necessarily open land. There are emergency vehicle crossovers every 1-2 miles on an up-to-spec interstate, and no good solutions for eliminating those without having lots of grade crossings between high-speed car traffic and high-speed rail traffic. There are regs about buffer space that would have to be heavily adjusted. And rail can't exceed 1% grades without speed restrictions, vs. 6% for interstate standards in hilly terrain. You would have to re-grade the ROW with superelevation or de-elevation in many places on I-89, with tunnels required in some parts. This is the same reason why Amtrak's fantasy map showing HSR to Boston along I-84 in Tolland County will never happen...there's at least 1 mile-long tunnel required to handle the grades.

This type of construction can work on new interstate construction, which is why the long-desired I-384 extension in Connecticut may work with better-bang-for-buck as a shotgun rail+interstate marriage on its proposed 1000 ft. greenway median. But where does this graduate from fantasy to buildable reality on existing interstates? The fed HSR map has all kinds of pretty PowerPoints showing lines that will do this...Inland NEC on I-684 and I-84, relocated NEC on I-95. But no one--ever--has done an engineering feasibility study about the actual costs and complications of grafting one onto a pre-existing interstate. There are many experts who say it'll be even MORE expensive than taking private land far away from the interstate. And there's great suspicion that the reason they haven't studied is that they know what the assessment is going to say: it'll cost twice as much.


Also, we don't live in this perfect universe where fully grade-separated, freight-separated, perfectly curving HSR can be plunked down in heavily built-up areas or old towns. We have different private property rights than Europe or Southeast Asia. The most cost-effective places to do this are existing corridors. And yes, that means putting up with freight. The Patriot Corridor has lots of areas that can be straightened out. It's got about a third the number of grade crossings as the Northern. It's got the multi-party stakeholders and 24/7 utilization. It's mostly out in the woods and doesn't have many NIMBY's or the physically awkward abutter problems of the available NH routes. Portions of it can be tri-tracked for passing the freights. And chances are if they were diligent about where they applied the money to speed it up the travel times are--like they used to be--going to be comparable to the best a NH route could achieve...for several $B less.

If NH were more proactive about rail, maybe this would be a different story. But their rail network is so chewed up from abandonments they have 4 regions of the state with absolutely no line interconnections, they are doubling-down on the stupid Tea right now which if it doesn't reverse VERY SOON will set them back an entire generation over more progressive MA/VT/ME on transportation options, and they don't have the tax base to contribute themselves to the build or lure other stakeholders (reason why it's largely a freight black hole compared to all its neighbors). They can't even pay for the add-a-lane Kool-aid they're drinking and have to raid the bridge repair fund to deficit-spend on I-93.

Sorry...they're not gonna get gifted this. The numbers have to kinda sorta make sense, and they don't. NH's got to halt its slide into third-world New England to support it; building it isn't going to pull create enough momentum to pull them out of third-world New England by itself. Amtrak will happily spur a Regional to Concord...that's an excellent corridor and desirable market from NYC. But I think NH residents will be taking a commuter shuttle down to Lowell to transfer to the Montreal train via Greenfield, with consolation prize of faster/more frequent service into MA. Otherwise it's just a fantasy line on a PowerPoint. If NH-or-bust is truly the preferred alternative, somebody's got to start coming up with empirical evidence that it can graduate from fantasy to feasibility. Nobody has. And nobody's even trying to...they're just printing more PPT's with that perfectly curving fantasy line.

 

[Nexis4jersey]

Actually the New Amtrak NEC Plans call for the line to hug US-6 from Manchester to Providence then rejoin the NEC and 4 track the NEC to Boston. The line will share ROW which extends 400ft out from the Highway on both I-684/84 & US-6. The Sticking parts that aren't answered with New England are what happens in Hartford and how do you connect the 684 segment to the current NEC. As for using I-89 its still to curvy and I don't see demand for a line like this. Commuter Rail up to Portsmouth , Concord and a Branch to Rochester is enough for New Hampshire at this point.