Other People's Rail: Amtrak, commuter rail, rapid transit news & views outside New England

Yeah, I've got mixed feelings on the topic. I do ultimately feel that Providence needs more express service to Boston, and that any sort of rail service to South County needs to have some sort of express component to Boston (whether thru service or cross-platform transfers). I'm less convinced than I was a year ago that Amtrak cross-honoring is the best way to achieve that, though.

Their strategy paper is well-argued, but now is in dire need of an update. The issue of gaps in off-peak service has been almost completely eliminated by the new MBTA schedules. Frequency of course can still be improved, but frequency can always be improved, and we are already at 2tph peak/1tph off-peak, which is light-years better than what we had pre-pandemic (for the most part).

Which pretty much leaves us with "faster service" as the main draw here, and the irony is that Amtrak isn't much faster than the T under certain conditions. For example, those super-expresses that run non-stop after Mansfield can run PVD-BOS in 62 minutes. According to the Jan 2020 timetable, the northbound Regionals clocked in at 50-60 minutes, while the southbounds did a bit better, usually at about 45 minutes.

62 vs 45 minutes isn't trivial, but it also isn't mind-blowing. With electrified rolling stock and full-highs at Attleboro and Mansfield, an MBTA super-express would likely be competitive. (Particularly since the MBTA would be able to blow through Route 128, which Amtrak never will.)

Finally, it's worth noting that Amtrak is running a fundamentally different kind of service than the T, which is reflected in its on-time performance; most trains serving BOS-PVD terminate/originate at Washington or points further south, which creates a lot more opportunities for delay, but also means that, in general, they are dealing with journeys where, say, a 15 minute delay is going to be seen differently than a 15 minute delay on a rush hour MBTA train. According to the Bureau of Transportation Statistics, the Northeast Corridor's OTP has pretty much been 80% ± 5% for several years now, which is borne out by data from the ASMAD (link will take long time to load); my eyeballing of the MBTA's OTP puts it closer to 85% ± 5% in recent years.

According to that ASMAD data, about 17% of Amtrak trains passing through PVD during 2017-2020 (pre-pandemic) were 15 minutes late or more; if we use the MBTA's definition of "on-time" (no more than 5 min late), that number jumps to 38%. That 15 minute delay wipes out the travel time savings, which statistically would happen to the average commuter once a week. That's lousy, straight-up. I'm also unsure how tolerant the riding public would be of Amtrak trains running at 62% OTP by MBTA standards.

So... like I said, I have mixed feelings about it. If they can hash out a negotiated deal to cross-honor on that 5pm southbound train -- which legitimately causes real problems for Boston commuters -- that would be worthwhile. And maybe likewise on that 7am northbound train.

But overall, I'm less confident than I wish I were that cross-honoring fares would make as much of a difference as one might think. Oh, I think it's a good idea, and I support it (and would definitely take advantage of it). But overall I think it's not a slamdunk, so in some ways I'm encouraged by the possibility that the focus might be narrow -- I'd prefer the advocacy capital be spent in measures commensurate with their impact.
 
Careful...there was only one media report from a not-very-reputable outlet that said that, and when pressed the reporter refused to elaborate. This needs additional confirmation that trade-in was even offered as a possibility by Siemens in the one-vendor contract, much less that Amtrak would ever consider pursuing that. It would directly contradict just about every piece of published fleet maint documentation from them until very recently, and doesn't make sense on an enormous number of levels over how many units worth of dual-modes they'd have to buy to track with service increases. The RR.net thread on this is a total dumpster fire of wild unhinged speculation right now, and the AMTK employee sources who usually confirm/deny what Mechanical/Procurement's aims are haven't been around to contain it. Until there's a second credible confirming source, treat this Sprinter replacement rumor as very highly suspect.

Fair enough. I thought it was pretty strange and wasn't quite sure what to make of some of the articles - going from pure electric to hybrid seemed a downgrade. So, just for my own edification: probably keeps the ACS-64s, and will be (as far as we know) a replacement of the passenger cars/amfleet. Just rather confused on the various articles as to what this actually is.
 
Two questions primarily for @F-Line to Dudley (but anyone else can certainly chime in!), and I’m sure he’s already written a short novel on these exact matters, so just a link to any such previous answers is more than enough:

- What are the easiest ways the Acela could shave time off its runs?
- What are the easiest ways Amtrak, in general, could avoid being delayed by freight trains?
 
- What are the easiest ways the Acela could shave time off its runs?
For the Acela, there's now a good list at the Connect NEC 2035 project site:
Where they seem to say there's about an hour's worth of time to be saved between Boston and NYC, (about half between BOS & NHV and the other half between NHV and NYP)
 
Two questions primarily for @F-Line to Dudley (but anyone else can certainly chime in!), and I’m sure he’s already written a short novel on these exact matters, so just a link to any such previous answers is more than enough:

- What are the easiest ways the Acela could shave time off its runs?
- What are the easiest ways Amtrak, in general, could avoid being delayed by freight trains?

There has been mention that the speed through curves has been increased in some areas. The NEC north of NYP has plenty of them. I guess part of this is that Amtrak has raised the cant (which was set artificially low) and increased the top speed to 165. I believe a lot of this is because the new Acela II is much lighter and more nimble.
 
Two questions primarily for @F-Line to Dudley (but anyone else can certainly chime in!), and I’m sure he’s already written a short novel on these exact matters, so just a link to any such previous answers is more than enough:

- What are the easiest ways the Acela could shave time off its runs?
- What are the easiest ways Amtrak, in general, could avoid being delayed by freight trains?
1) Curve straightening. Specifically whacking a handful of bad ones where a single-point curve or S-curve induces a punitive speed restriction. The docs Arlington linked to bullseye exactly which ones they are, and the most savings are rolled up into a relative handful of spots. Other things include stuff they're doing like renewing the overhead for greater contact reliability where quality of wire contact slows things down. Much of the infrastructure penalty exerts itself not so much in raw clock time but over-padding on the schedules, where if you winnowed down the variability where things could run a few minutes late you tighten up the whole works enormously. Way tauter timetables, and more slots because more trains can plug a lot of the gaps that are infringed on today by contingency padding.

2) There's extreme-little freight overlap on the NEC. That's really a total non-factor. Especially north-of-NYC where each line segment barely has 1-2 freights per day and a lot of that is already graveyard-shift when almost nothing passenger is out to begin with. NYC-DC the freights are largely segregated onto paralleling freight-only lines and only interact with the NEC for very short distances (Port of Baltimore access, couple touches on the Jersey coast) before turning out onto coastal branchlines. It really doesn't rate at all. Not even for the weight, as honestly running fast with over-heavy passenger equipment (see: Acela 1) beats up the infrastructure way more than the piddling freight schedule that constrains no known slots. Commuter rail interference ends up an enormously bigger factor. You need to get the T (proposed), Shore Line East (planned), and MARC (planned) fully electric to get rid of the slovenly diesel acceleration. You have to keep pushing the New Haven Line towards full state-of-repair so Metro North's enormous service layer cake runs as padding-free as can be so Amtrak can run as padding-free as can be. And you need to get the level-boarding laggards (T, MARC, and--above all others--SEPTA) to start better-controlling their boarding dwells with high-level platform buildouts. Amongst other optimizations.
 
Last edited:
There has been mention that the speed through curves has been increased in some areas. The NEC north of NYP has plenty of them. I guess part of this is that Amtrak has raised the cant (which was set artificially low) and increased the top speed to 165. I believe a lot of this is because the new Acela II is much lighter and more nimble.
Largely the Aveilas being more conventionally lightweight. The A1's are so uniquely world-heavy that they absolutely murder curve infrastructure at speed. Wear out curve rail more than more numerous commuter trains, more than more numerous Regionals, more than super-long LD's with their pretty heavy specialty cars, and more than the freight trains. The combo of force and speed exerted under all that excess weight lays waste to the track in record time. And the A1's tilt infrastructure is so bizarrely overcustom it doesn't behave conventionally with forces exerted around curves to make the endemic weight problem even worse at points of maximum forces. The Aveilas aren't lightweight by any stretch (mainly because the French TGV tech they're imported from isn't light by world standards), but they at least rate as "conventional" HSR rolling stock weight without the senseless extremes of their predecessors.

Second is the wire contact interface being massively improved on the Aveilas. The A1's run front-pantographs-up as a rule...the opposite of world best practice rear-pans-up...because their power cars ride so rough the pantograph doesn't apply even pressure on the overhead on all curves leading to problematic power spikes/dips. The leading pantographs--usually an icy-weather only ops hedge on worldwide HSR--go up full-time because they can maintain more solid contact than the rear pans...but at a price of much higher rates of friction shortening infrastructure lifespans. The Aveilas are world-conventional rear-pans-up, which is a lot less punishing to both the overhead and the trainset. And they're able to do that because the power cars are ride-smooth and don't jerk all over the place on their suspension like the unruly A1's. The A1's also can't maintain even enough contact on bouncy old variable-tension overhead, and have to run both pans up in some remaining sections (e.g. parts of New Jersey where the center tracks are constant-tension but the outer tracks are still variable-tension, and slotting mismatches sometimes force the Acela onto the non-preferred outer tracks)...which really really beats up the electronics to a whole other level. A whole slew of wire-related spot restrictions that were capped at 130 MPH for the A1's around now round up to 150. This nets a ton of shed padding, as right now the schedules have to be predicated on traffic or maintenance forcing the Acela onto the non-preferred variable-tension at a lot of built-in timekeeping flab.
 
At least they call out some of the grade crossings for elimination. It’s nutty that they exist on a HSR right of way.
 
Interesting. When I was taking the Acela a couple times a year (about a decade ago), I was struck by how rough the ride was in the 150mph stretches in MA. I assumed at the time it was lousy track maintenance, but was it more a problem with the A1's suspensions?
 
At least they call out some of the grade crossings for elimination. It’s nutty that they exist on a HSR right of way.

The crossings are all pretty much on slower sections of track. Certainly nowhere near the 110 (or 125 mph with barrier) limits allowed.
 
1) Curve straightening. Specifically whacking a handful of bad ones where a single-point curve or S-curve induces a punitive speed restriction. The docs Arlington linked to bullseye exactly which ones they are, and the most savings are rolled up into a relative handful of spots.

There's a lot of info there and I did not see any specific curve information. Can you please tell us where this is exactly. Thanks,
 
1) Curve straightening. Specifically whacking a handful of bad ones where a single-point curve or S-curve induces a punitive speed restriction. The docs Arlington linked to bullseye exactly which ones they are, and the most savings are rolled up into a relative handful of spots. Other things include stuff they're doing like renewing the overhead for greater contact reliability where quality of wire contact slows things down. Much of the infrastructure penalty exerts itself not so much in raw clock time but over-padding on the schedules, where if you winnowed down the variability where things could run a few minutes late you tighten up the whole works enormously. Way tauter timetables, and more slots because more trains can plug a lot of the gaps that are infringed on today by contingency padding.

2) There's extreme-little freight overlap on the NEC. That's really a total non-factor. Especially north-of-NYC where each line segment barely has 1-2 freights per day and a lot of that is already graveyard-shift when almost nothing passenger is out to begin with. NYC-DC the freights are largely segregated onto paralleling freight-only lines and only interact with the NEC for very short distances (Port of Baltimore access, couple touches on the Jersey coast) before turning out onto coastal branchlines. It really doesn't rate at all. Not even for the weight, as honestly running fast with over-heavy passenger equipment (see: Acela 1) beats up the infrastructure way more than the piddling freight schedule that constrains no known slots. Commuter rail interference ends up an enormously bigger factor. You need to get the T (proposed), Shore Line East (planned), and MARC (planned) fully electric to get rid of the slovenly diesel acceleration. You have to keep pushing the New Haven Line towards full state-of-repair so Metro North's enormous service layer cake runs as padding-free as can be so Amtrak can run as padding-free as can be. And you need to get the level-boarding laggards (T, MARC, and--above all others--SEPTA) to start better-controlling their boarding dwells with high-level platform buildouts. Amongst other optimizations.

Thanks (as well to @Arlington and @BostonTrainGuy ). For a clarification regarding freight (or commuter!), I meant across Amtrak, not just on the NEC.
 
Interesting. When I was taking the Acela a couple times a year (about a decade ago), I was struck by how rough the ride was in the 150mph stretches in MA. I assumed at the time it was lousy track maintenance, but was it more a problem with the A1's suspensions?
The power cars have a major 'hunting' issue with rocking back and forth laterally on their suspensions at top speed. Engineers who run them have said on RR.net that the cab ride can be a real barf-bag. That's a large share of their wire contact problems right there...the fact that they upchuck such an excess amount of lateral movement at high speed that the power draw has trouble sustaining itself. And also one of the reasons why they have equipment-induced speed restrictions around curves and aren't allowed to go above 150 even though they're technologically rated for 170, because they throw themselves against the railhead like a ragdoll. The retired HHP-8 locomotives, which are loosely based around the same Alstom truck design as the A1 power packs, had an even more pronounced variant of the same rocking problem...such that you can find YouTube videos of them twerking around like crazy at track speed pulling Regionals. You'd probably feel some of that roughness if you were in the first or last carriage of the set abutting the power cars. The active-tilt mechanisms damp down a lot of that movement, but in their advanced age those Frankenstein-design carriage suspensions probably don't do nearly as good a job of it as they once did and it's unlikely that a ride today would be as smooth as it was when they were still relatively new.

So you've got overweight cars with all-around very poorly-designed and hideously overcomplicated suspension throwing a lot of forces around, and everything has to be in utter perfect alignment to cancel those forces out. Not a recipe for lasting success, and the main reason why the A1's are currently the most expensive passenger trains in the world to operate. They forces they throw beats the living snot out of the equipment itself, giving them their career-long status as shop queens...and they beat the living snot out of the rail and overhead inducing more MOW cost to keep running than any other trainset on the planet.
 
I love "there's a word for that" terms from maritime and railroad domains, like Hunting.
I've added Wikipedia's gif for Hunting: below so you can have a mental image and re-read F-line, above, when he says the A1s
"throw themselves against the railhead like a ragdoll," which "beats the living snot out of the equipment itself [...].and they beat the living snot out of the rail and overhead" [as the pantographs danced under the wire]
Hunting_oscillation_section_ja_150px.gif

Video of an empty gondola car with its front truck hunting: (starts at 2:33)
(seems to induce plenty of wobble in the car in front and damped hunting in the trucks ahead and behind)
 
Last edited:
I love "there's a word for that" terms from maritime and railroad domains, like Hunting.
I've added Wikipedia's gif for Hunting: below so you can have a mental image and re-read F-line, above, when he says the A1s
"throw themselves against the railhead like a ragdoll," which "beats the living snot out of the equipment itself [...].and they beat the living snot out of the rail and overhead" [as the pantographs danced under the wire]
Hunting_oscillation_section_ja_150px.gif

Video of an empty gondola car with its front truck hunting: (starts at 2:33)
(seems to induce plenty of wobble in the car in front and damped hunting in the trucks ahead and behind)

The local example of 'hunting' in action on the T is the Green Line Type 7's. All-around excellent cars over 35 years of service, but their one design Achilles heel is that on the very fastest parts of the D Line on the Newton straightaways the vehicles do 'hunt' quite a bit. Function of them being somewhat top-heavy with all the HVAC equipment on the roof. You may have noticed that the ride gets a little uncomfortable at 45+ MPH, and can be a white-knuckle experience as a standee with how much the G's are throwing you side-to-side. That's the Kinkis' top-heaviness inducing some 'hunting' at-speed. Supposed to be a little bit better post-rebuild since the new roof HVAC units are a little bit lighter and lower-profile than what was there pre-rebuild. The Boeings that came before them and the Type 8's/9's that came after them did/do not have the same suspension issues, and so the ride in any not-7 car will be noticeably smoother on the outer D. You don't notice any hunting at lower speeds. It is primarily a thing that'll manifest itself near the vehicle's top speed range.

For a Type 7, that's >40 MPH and very noticeable degradation in ride quality at 45+. For an HHP-8 locomotive, the rockin'-and-rollin' was primarily at >100 MPH. And for an Acela 1, it's 150 MPH. Not inherently dangerous if the track is maintained well...it's just that with the Acela it's costing Amtrak an arm-and-a-leg to maintain track because the morbidly obese Acela unleashes such unprecedented punishment on the railhead, wire, trainset axles, pantographs, etc. at top speed. One reason out of many why they can't wait to get those pieces of crap retired and replaced by the much lighter and ride-smoother Aveilas.
 
You may have noticed that the ride gets a little uncomfortable at 45+ MPH, and can be a white-knuckle experience as a standee with how much the G's are throwing you side-to-side. That's the Kinkis' top-heaviness inducing some 'hunting' at-speed.
For a Type 7, that's >40 MPH and very noticeable degradation in ride quality at 45+.

No Greenline Type 7 trains go over 40 mph. IFAIK the old Boeing LRVs were the last trains to run at 50 mph on the Highland Branch.
 
I've noticed the Type 7s truck hunting on the Lechmere viaduct going well under 40.
 

Back
Top