Acela & Amtrak NEC (HSR BOS-NYP-WAS and branches only)

[F-Line to Dudley]

I feel like the complexity here is exactly why these agencies need to get together and standardize. We should at least be able to agree on things like signalization and electrification scheme and a dynamic envelope. It doesn’t necessarily have to be “one EMU to rule them”. There can be a finite number of variants (e.g. a tight-clearance 3rd rail capable set of requirements vs a mainline express set of requirements).
But getting to a point where companies can put together off-the shelf NEC-spec designs (for rolling stock, stations, tunnels, etc.) and pitch them to multiple agencies would be huge. Getting to the point where agencies along the NEC could borrow or sell rolling stock to one another would be huge. Getting to the point where capital costs are easier to estimate, alternatives are easier to compare, and systems are easier for mathematicians to optimize would be HUGE. You’d see fewer studies, faster timelines, more vendor competition, and more confident investors. All good things.

Except "one EMU to rule them all" is exactly what they're pitching in lieu of touching any incumbent electrification or signaling schemes. The Effective Transit Alliance, which is cross-pollinated with many of the same stakeholders as the Transit Costs Project, puts the M10 uni-design as the centerpiece of its grand pan-NYC electrification scheme. They aren't looking at targeted segmentation or eliminating lineside equipment fragmentation. They're putting all their eggs in the basket of one unicorn vehicle make while saying out the other sides of their mouths "but it can't be overcustomized".

It won’t happen overnight, but if everything was written down and agreed-upon, everyone could start working towards it.

Real change happens through pressure and time, not by clicking one's heels in a single moment of clarity. So why is it that these advocacy-driven reform proposals never plot their reforms against time? It's always like "you should just fire all your Not-Invented-Here management failsons and replace them instantaneously with their superior foreign counterparts who I'm sure would be ecstatic to culturally uproot and reassimilate, adopt all world best practices overnight, and have one capital and one procurement plan to accomplish it all at once or else you've failed." We see that here with the incoherent and self-contradictory "one EMU to rule them all" cudgel. These proposals aren't evolving completely enough from transpo blogosphere comment section Internet arguments to realistic, feasible phased proposals that can have their success plotted on change vs. time and are accomplishable by re-teaching and evolving our locally-bred transit management ranks that we're practically stuck with. It absolutely sucks that we're stuck with so many backward and structurally ingrained transit mismanagement practices in this country, but we're not exactly being fed actionable blueprints for how to get caught up through time with proposals like this and the Effective Transit Alliance. Even TransitMatters gets caught up in the "click your heels and change all the things everywhere at once" tendency, though they tend to be much more consistently grounded about it than their counterparts in NYC.

 

[sneijder]

Alon Levy wrote about the 16-car trains a few months ago, and went into deep detail on the required station renos: https://pedestrianobservations.com/2025/02/19/16-car-trains-on-the-northeast-corridor/

You're absolutely correct: the math doesn't add up for a legacy-system retrofit. They're talking about things like radical re-segmentation of platform assignments from local commuter rail agencies to segregate the longest platforms to HSR conflicts-be-damned; outright skipping hard-to-modify stops like Stamford, Trenton, and Back Bay; a major do-over of New Haven because of a conjoined proposal to do a major curve bypass east of the station; ugly hacks like not opening up all doors on the trains at some stops despite that being problematic with reserved seating; and on and on. The station mod cost blowouts are literally going to be in the billions. And that's without even touching the fact that 16-car trains put a tremendous individual strain on the power supply, which likely needs to be upgraded in multiple spots if they're to be run frequently. The thoroughly-mongrel NEC absolutely isn't comparable at all to post-WWII greenfield HSR systems that overrepresent the long-consist HSR ranks, and there's been no effort whatsoever to right-size the associated costs.

I mean, we run 12-car Regionals every day of the week (absorbing some boarding inconvenience at a few too-short stations), which is 33% more capacity than the 8-car Acela 2's. We could get halfway to target with minimal cost and disruption. But no...somebody in the world did it under completely different circumstances, therefore we've failed by not shooting for absolute perfection damn the logistics.

But to be fair, if we really are able to achieve 2 hours from NYC <> BOS and NYC <> DC (which is a BIG IF), there would be a ton of induced demand and modeshare shift. We'd expect tons of people to stop driving and the flight schedules can be pared down to be for connecting flights only. I think expecting demand to quadruple is not unreasonable given the expected growth in the next 20 years + quick math on reducing utilization of I-95 capacity.

I know parts of this report are still a pipe dream but it does seem far more achievable than the Ronkonkoma > Hartford re-routing of HSR, or finding any other new ROW for that matter.

 

[ritchiew]

Alon Levy wrote about the 16-car trains a few months ago, and went into deep detail on the required station renos: https://pedestrianobservations.com/2025/02/19/16-car-trains-on-the-northeast-corridor/

You're absolutely correct: the math doesn't add up for a legacy-system retrofit. They're talking about things like radical re-segmentation of platform assignments from local commuter rail agencies to segregate the longest platforms to HSR conflicts-be-damned; outright skipping hard-to-modify stops like Stamford, Trenton, and Back Bay; a major do-over of New Haven because of a conjoined proposal to do a major curve bypass east of the station; ugly hacks like not opening up all doors on the trains at some stops despite that being problematic with reserved seating; and on and on. The station mod cost blowouts are literally going to be in the billions. And that's without even touching the fact that 16-car trains put a tremendous individual strain on the power supply, which likely needs to be upgraded in multiple spots if they're to be run frequently. The thoroughly-mongrel NEC absolutely isn't comparable at all to post-WWII greenfield HSR systems that overrepresent the long-consist HSR ranks, and there's been no effort whatsoever to right-size the associated costs.

I mean, we run 12-car Regionals every day of the week (absorbing some boarding inconvenience at a few too-short stations), which is 33% more capacity than the 8-car Acela 2's. We could get halfway to target with minimal cost and disruption. But no...somebody in the world did it under completely different circumstances, therefore we've failed by not shooting for absolute perfection damn the logistics.

Thanks for that link. That adds the further logic that "China does it," which, you're right, not especially compelling.

That blog post does fill in a couple of numbers I was looking for, if anyone else is interested. I think this is all correct:

• For the HSR the report plans, there would be 10 stations, plus two maybes. Already today, those stations all accommodate 10-car trains, which is bigger than Acela's 8-car trains. They could just target that, and no platform expansions needed.
• They could get 12-car trains by fixing one station: New Haven. That doesn't seem to be one of the harder fixes. That could also just be the one stop where they selectively open doors (though that's a busy station and maybe not a good candidate). So, they could get 12-car trains with between 0 and 1 station expansions. That would likely cost in the $10s of millions.
• If they target 12-car trains, they could add in some other stations they say are "too short." That would at least include Back Bay.
• They instead target 16-car trains. That requires platform expansion at 10 stations. At least an order of magnitude more cost for a marginal capacity improvement. I don't get it.

 

[Jahvon09]

Is the Avelia Liberty delayed again? Why in the world does this train just keep on being put off from going into service again, & again, & again?!!!!!!!

 

[themissinglink]

https://www.bloomberg.com/news/articles/2025-05-07/amtrak-to-cut-20-of-top-level-employees-amid-trump-uncertainty

Amtrak to Cut 20% of Top-Level Employees Amid Trump Uncertainty​

Amtrak is cutting roughly 20% of its top-level management staff in an effort to reduce costs amid uncertainty about President Donald Trump’s plans to invest in infrastructure, according to people familiar with the matter.
[...]
Among the areas affected by this week’s layoffs are an expanded unit within the railroad that was intended to shepherd some of its largest capital investment projects, including new multi-billion-dollar rail tunnels in New York City and Baltimore, a replacement for the Susquehanna River Bridge and other major construction projects along the busy Northeast corridor.
The tunnel and bridge projects are continuing ahead, so far, but some of the future funding Amtrak was counting on is now in doubt – including the expected fifth year of investment from former President Joe Biden’s infrastructure bill, this person said.

 

[F-Line to Dudley]

• They instead target 16-car trains. That requires platform expansion at 10 stations. At least an order of magnitude more cost for a marginal capacity improvement. I don't get it.

Requires all-new rolling stock too to replace the all-new rolling stock they just ordered and which hasn't entered service yet. The Aveila power cars are rated for a maximum of 9 cars, only 1 more than the sets they ordered. So they could theoretically buy a supplemental order of carriages and easily add +1 to each set for slight capacity expansion, but they're not getting 10+ cars per set until the rolling stock order-after-next which doesn't happen for 25 years. The similar but slightly different Aveila Pendolino's can do up to 11 cars, so it's *possible* an uprating-cum-midlife overhaul of the Aveila Liberty power cars could squeeze a couple more...but that's still 5 fewer cars than they're asking for with this drunken bar bet of a proposal. 16 cars pretty much requires an EMU-configuration HSR set instead of power car -bookended sets...and that runs headfirst into the thorny power draw issues I cited in the last post. A 16-car set of EMU's rated for 300 km/h at HSR acceleration is insanely power-hungry. Even SCNF didn't want to go there with its TGV electrical infrastructure, nevermind Amtrak with what passes for the NEC's electrical backbone.

Totally not a serious real-world proposal, though it's being presented as a deadly serious.

 

[Jahvon09]

There' a bridge somewhere along the Northeast Corriddor that turns to let boats trough, similar to the one in Boston that closed. It is being replaced with a new & higher one that lets the traffic on the water go under it, but it's not completed yet. Supposedly, it's not ready yet.

 

[HenryAlan]

Totally not a serious real-world proposal, though it's being presented as a deadly serious.

I'm convinced on the unsuitability of 16 car trains, but what do you think about the rest of the proposal? Although more knowledgeable than the average man on the street, I definitely don't have the knowledge base to critically evaluate the report. That said, from my semi-informed layman's point of view, a lot of the document's suggestions seem reasonable and possible.

 

[F-Line to Dudley]

I'm convinced on the unsuitability of 16 car trains, but what do you think about the rest of the proposal? Although more knowledgeable than the average man on the street, I definitely don't have the knowledge base to critically evaluate the proposal. That said, from my semi-informed layman's point of view, a lot of the document's suggestions seem reasonable and possible.

Most of the rest of it is solid, straightforward, and recommended from many sides. I don't think the bypasses are practically going to happen with the opposition in very built-up parts of Connecticut and think they're hilariously underestimating the costs in those same built-up parts, but those are not particularly essential pieces for netting a world-class NEC so I give them somewhat of a pass for aiming high on that one. It's not load-bearing for their overall goals, so I kind of gloss over the overall impracticality of it. I also think the politics of cutting stations on the HSR runs is going to get in the way of some of their timetabling. You can't be omitting very high ridership stations like Stamford and Back Bay from Acela schedules, and good luck getting political cooperation with a state capitol like Trenton omitted. But those likewise aren't showstoppers for netting a world-class NEC, so they're aiming high and the more-stops compromise won't be inferior. The costs are a little bit lowballed; like my critique of TransitMatters' opposition for SSX I don't think the net gain in platforms at South Station can functionally be zero/nada/nothing and accommodate everybody's growth and everybody's schedule perfection with adequate safeguards for chaos effects, but it also doesn't have to be +7 platforms like the studies spec either if we were truly focused on studying the right balance for ops reform. There is a middle ground somewhere in there if Regional Rail advocates would just stop infighting with other Regional Rail advocates and take a serious look. But those are relatively small nits to pick that drill down into individual commuter rail systems and will be decided by state-level politics not master planners.

The biggest overall whopper is their full-throated backing of the Effective Transit Alliance's "everybody runs-thru everywhere" in NY/NJ to get out of touching Penn capacity which is naive to the nth power about the politics involved and so very disingenuously glosses over the technical challenges. They may not have invented that one--it was the ETA's baby--but they're doing themselves a great disservice by not being the least bit questioning of its practicality and swallowing it in-full. It's not practically going to happen, so a lot of load-bearing things about Penn and NYC traffic need to change in their plans to contour to that reality. And the 16-car trains is just such a random thing to be doubling-down on and driving other recommendations like station omissions. I can't even guess at what they're substantively going for there, other than "so-and-so did it, so you've failed if you don't match".


I think you can fashion an authoritative thrust for improvements out of most of the recommendations, and from the overall themes. So it's a big net-positive in spite of some small quibbles and those couple of big whoppers that require major retooling.

 

[bigeman312]

Most of the rest of it is solid, straightforward, and recommended from many sides. I don't think the bypasses are practically going to happen with the opposition in very built-up parts of Connecticut and think they're hilariously underestimating the costs in those same built-up parts, but those are not particularly essential pieces for netting a world-class NEC so I give them somewhat of a pass for aiming high on that one. It's not load-bearing for their overall goals, so I kind of gloss over the overall impracticality of it. I also think the politics of cutting stations on the HSR runs is going to get in the way of some of their timetabling. You can't be omitting very high ridership stations like Stamford and Back Bay from Acela schedules, and good luck getting political cooperation with a state capitol like Trenton omitted. But those likewise aren't showstoppers for netting a world-class NEC, so they're aiming high and the more-stops compromise won't be inferior. The costs are a little bit lowballed; like my critique of TransitMatters' opposition for SSX I don't think the net gain in platforms at South Station can functionally be zero/nada/nothing and accommodate everybody's growth and everybody's schedule perfection with adequate safeguards for chaos effects, but it also doesn't have to be +7 platforms like the studies spec either if we were truly focused on studying the right balance for ops reform. There is a middle ground somewhere in there if Regional Rail advocates would just stop infighting with other Regional Rail advocates and take a serious look. But those are relatively small nits to pick that drill down into individual commuter rail systems and will be decided by state-level politics not master planners.

The biggest overall whopper is their full-throated backing of the Effective Transit Alliance's "everybody runs-thru everywhere" in NY/NJ to get out of touching Penn capacity which is naive to the nth power about the politics involved and so very disingenuously glosses over the technical challenges. They may not have invented that one--it was the ETA's baby--but they're doing themselves a great disservice by not being the least bit questioning of its practicality and swallowing it in-full. It's not practically going to happen, so a lot of load-bearing things about Penn and NYC traffic need to change in their plans to contour to that reality. And the 16-car trains is just such a random thing to be doubling-down on and driving other recommendations like station omissions. I can't even guess at what they're substantively going for there, other than "so-and-so did it, so you've failed if you don't match".


I think you can fashion an authoritative thrust for improvements out of most of the recommendations, and from the overall themes. So it's a big net-positive in spite of some small quibbles and those couple of big whoppers that require major retooling.

What do you estimate is the feasibility/cost of their proposed Kingston-New Haven bypass? They estimate $5b. To my untrained eyes, this is the most intriguing component of the proposal. A 75-mile bypass that saves 32 minutes on express trips is a sexy proposition. How realistic is it?

 

[HenryAlan]

I think you can fashion an authoritative thrust for improvements out of most of the recommendations, and from the overall themes. So it's a big net-positive in spite of some small quibbles and those couple of big whoppers that require major retooling.

Thanks, this is really helpful. I definitely found the part about time table construction to be quite illuminating, along with quite a bit of the other pieces. Overall, it seems like this can push the conversation closer to a consensus plan and therefore better activisim.

 

[F-Line to Dudley]

What do you estimate is the feasibility/cost of their proposed Kingston-New Haven bypass? They estimate $5b. To my untrained eyes, this is the most intriguing component of the proposal. A 75-mile bypass that saves 32 minutes on express trips is a sexy proposition. How realistic is it?

Given that it's along I-95 through challenging geography (lots of hills and trap-rock cliffs that make even simpler projects like add-a-laneing 95 to a consistent six lanes multi-billion dollar propositions) and a lot of development that's grown up in the last 70 years around 95, not very realistic. The locals and their politicians were violently against it when the NEC FUTURE commission proposed it over their heads, and the State of CT itself prefers a further-inland Hartford-Providence bypass. Its inclusion is a bit of "dictatorial master planner ruling like the fist of a God over the local rabble" delusion of grandeur. It was spat back at the federal government already, and will be spat back at a ragtag group of advocates just as forcefully.

The relative "sexiness" of a bypass comes after you do all the nonsexy things like in-situ infrastructure improvements, ops reform, and revamped timetabling. Where you get a long way towards a better NEC but then are staring at an immovable object towards getting the best NEC. Then people will think more seriously about it. Demanding it up-front just isn't serious, though.


EDIT: Complete widening of I-95 to 6 lanes throughout CT comes with a batshit $11.2B price tag. So, yeah, $5B for a longer-than-that HSR bypass on a similar ROW is low by at least half if not more.

 

[themissinglink]

https://www.nytimes.com/2025/05/12/technology/elon-musk-boring-company-amtrak.html

Elon Musk’s Boring Company Is in Talks With Government Over Amtrak Project​

Federal Railroad Administration officials have talked with employees at the Boring Company about assessing the costs and progress of the Frederick Douglass Tunnel program, a new tunnel along a busy Amtrak stretch connecting Baltimore to Washington and Virginia. Amtrak initially expected the development to cost $6 billion, but now estimates it could cost as much as $8.5 billion.

As part of the talks, officials with the Department of Transportation, which oversees the Federal Railroad Administration, met with employees from the Boring Company last month and were told that the firm could find ways to build the tunnel more cheaply and efficiently, according to two of the people familiar with the discussions.

A Transportation Department spokesman, Nathaniel Sizemore, confirmed that the Boring Company was one of several firms being consulted for the purposes of awarding a new engineering contract. He declined to name the other companies.

 

[HenryAlan]

Musk knows absolutely nothing about building multi track train scale tunnels. And the conflict of interest is obviously significant.

 

[Longfellow]

I remember hearing a long time ago that the boring company's innovation was using sewer tunneling technology which is more common and therefore cheaper to reduce costs compared to the big TBMs that everyone has always used for trains. If anyone has seen videos of their dinky tunnel in Vegas you'll know what I mean -- it's only barely wider than a single lane of traffic. However, the problem is that these tunnels are actually a much smaller diameter than train tunnels. So not sure what to make of this.

 

[kingofsheeba]

Musk knows absolutely nothing about building multi track train scale tunnels. And the conflict of interest is obviously significant.

He enrages me to a level that I haven’t felt as an adult

 

[ceo]

Not that I'm any kind of expert here, but I've never had the sense that that tunnel boring is the kind of enterprise that's ripe for the kind of disruptive innovation that Tesla and SpaceX achieved. (How much credit Musk deserves for those innovations is, of course, debatable.)

 

[RandomWalk]

The whole point was to sandbag mass transit in favor of private transportation (personal vehicles, personal people movers, etc)

 

[themissinglink]

Warehouse Demolition Planned After Fire Disrupts MD Trains

Loud booms from a MD warehouse fire are under investigation. The building, which neighbors heavily delayed train lines, will be demolished.

patch.com

The ATF will investigate what caused loud booms at the still-smoldering warehouse fire that broke out Monday evening in West Baltimore.

The warehouse lies feet from the railway shared by Amtrak and MARC. The blaze temporarily halted rail operations Monday evening through early Tuesday morning. The Amtrak and MARC are both back in service, but delays are still ongoing.

Officials are concerned about the stability of the warehouse and its proximity to the tracks.

 

[RandomWalk]

I rode the Acela this week. The trains are really showing their age. The trays are very temperamental and the bathrooms are really worn. However, I would still spring for first class Acela over any of the flights down to NYC.

I’m hoping that the Avelias don’t get caught up in politics and my next trip will be on them.