Amtrak’s Rolling Stock Order: Followup

A year ago, based on a leak from Senator Charles Schumer’s office, I attacked Amtrak for paying double for its new high-speed trains – $2.5 billion for 28 trainsets, about $11 million per car. Amtrak at the time denied the press release, saying it was still in the process of selecting a bidder. However, last week Amtrak announced the new order, confirming Schumer’s leak. The trainsets are to cost $2 billion, or $9 million per car, with an additional $500 million spent on other infrastructure. The vendor is Alstom, which is branding all of its export products under the umbrella name Avelia; this train is the Avelia Liberty.

You can see a short promotional video for the trains here and read Alstom’s press release here. Together, they make it obvious why the cost is so high – about twice as high per car as that of Eurostar’s Velaro order, and three times as high as that of the shorter-lived N700 Shinkansen. The Avelia Liberty is a bespoke train, combining features that have not been seen before. Technical specs can also be seen in Alstom’s press kit. The Avelia Liberty will,

  • Have a top speed of 300 km/h.
  • Have articulated bogies.
  • Be capable of 7 degrees of tilt, using the same system as in Alstom’s Pendolino trainset.

In particular, the combination of high speed and high degree of tilt, while technically feasible, does not exist in any production train today. It existed in prototype form, as a tilting TGV, but never made it to mass production. The Pendolino has a top speed of 250 km/h, and the ICE-T has a top speed of 240 km/h. Faster tilting trains do not tilt as much: Talgo claims the Talgo 350 is capable of lateral acceleration of 1.2 m/s^2 in the plane of the train, which corresponds to 180 mm of cant deficiency, achievable with 2-3 degrees of tilt; the tilting Shinkansen have moderate tilting as well, which the JRs call active suspension: the N700 tilts 1 degree, and appears capable of 137 mm of cant deficiency (270 km/h on 2.5 km curves with 200 mm cant), whereas the E5 and E6 tilt 2 degrees, and appear capable of 175 mm (in tests they were supposed to do 360 km/h on 4 km curves with 200 mm cant, but only run at 320 km/h for reasons unrelated to track geometry).

I have argued before, primarily in comments, that a train capable of both high speed and high degree of tilt would be useful on the Northeast Corridor, but not at any price. Moreover, the train is not even planned to run at its advertised top speed, but stay limited to 257 km/h (160 mph), which will only be achievable on short segments in Massachusetts, Rhode Island, and New Jersey. Amtrak has no funded plan to raise the top speed further: the plans for constant-tension catenary in New Jersey are the only funded item increasing top speed. There is no near-term plan on the horizon to obtain such funding – on the contrary, Amtrak’s main priority right now is the Gateway tunnel, providing extra capacity and perhaps avoiding a station throat slowdown, but not raising top speed.

Running trains at 300 km/h on the segments that allow the highest speeds today, or are planned to after the speedup in New Jersey, would save very little time (75 seconds in New Jersey, minus acceleration and deceleration penalties). Making full use of high top speed requires sustaining it over long distances, which means fixing curves in New Jersey that are not on the agenda, installing constant-tension catenary on the entire New York-Washington segment and not just over 40 km of track in New Jersey to eliminate the present-day 215 km/h limit, and building a bypass of the entire segment in southeastern Connecticut along I-95. None of these is on the immediate agenda, and only constant-tension catenary is on the medium-term agenda. Hoping for future funding to materialize is not a valid strategy: the trains would be well past the midpoint of their service lives, and spend many years depreciating before their top speed could be used.

What’s more, if substantial bypasses are built, the value of tilting decreases. In advance of the opening of the Gotthard Base Tunnel, Swiss Federal Railways (SBB) ordered 29 trainsets, without tilting, replacing the tilting Pendolino trains that go through the older tunnel. SBB said tilting would only offer minimal time reduction. The eventual cost of this order: about $36 million per trainset as long as 8 US cars. On the entire Northeast Corridor, the place where tilting does the most to reduce travel time is in Connecticut, and if the eastern half of the tracks in the state are bypassed on I-95, tilting loses value. West of New Haven, tilting is not permitted at all, because of Metro-North’s rules for trains using its tracks; on that segment, tilting will always be valuable, because of the difficulty of finding good rights-of-way for bypasses not involving long tunnels, but to my knowledge Amtrak has not made any move to lift the restriction on tilting. Even with the restriction lifted, a 300+ km/h train with moderate tilting, like the N700 or E5/6 or the Talgo AVRIL, could achieve very fast trip times, with only a few minutes of difference from a hypothetical train with the same top speed and power-to-weight ratio and 7 degrees of tilt. It may still be worth it to develop a train with both high speed and a high degree of tilt, but again, not at any cost, and certainly not as the first trainset to use the line.

Another issue is reliability. The Pendolino tilt system is high-maintenance and unreliable, and this especially affects the heavier Acela. SBB’s rejection of tilting trains was probably in part due to the reliability issues of previous Pendolino service across the Alps, leading to long delays. Poor reliability requires more schedule padding to compensate, and this reduces the advantage gained from faster speed on curves. While tilting trains are overall a net positive on curvy routes like the Connecticut segment of the Northeast Corridor, they are probably not useful in any situation in which 300 km/h top speeds are achievable for a meaningful length of time. This goes double for the Avelia Liberty, which is not a proven Pendolino but a new trainset, sold in a captive market that cannot easily replace it if there are maintenance issues.

In my post a year ago, I complained that Amtrak’s specs were conservative, and did not justify the high cost. I stand behind that assessment: the required trip times are only moderate improvements over the current schedule. At least between New York and Boston, the improvement (9 minutes plus stop penalty at New London) is less than the extent of end-of-line schedule padding, which is at least 10 minutes from Providence to Boston for northbound trains. However, to achieve these small trip time improvements, Amtrak elected to demand exacting specs from the trainsets, leading to high equipment costs.

In 2013, I expounded on this very decision by borrowing a Swiss term: the triangle of rolling stock, infrastructure, and timetable. Planning for all three should be integrated. For example, plans for increases in capacity through infrastructure improvements should be integrated with plans for running more trains, with publicly circulated sample schedules. In this case, the integration involves rolling stock and infrastructure: at low infrastructure investment, as is the case today, there is no need for 300 km/h trainsets, whereas at high investment, high top speed is required but 7-degree tilt is of limited benefit. Instead of planning appropriately based on its expectations of near-term funding, Amtrak chose to waste about a billion dollars paying double for trainsets to replace the Acela.


  1. Marc

    The $2 billion is the total of two contracts, one for the trains themselves, one for “long term” technical support and parts. None of the articles or press releases mention the costs of the contracts separately. Does the Eurostar Velaro price also include technical support/parts for a comparable time period, or are you (and others) comparing apples to oranges?

    • Benjamin Turon

      Good point… I forget that, ran across that being reference only one I think, some money in the loan is also going to station improvements on the NEC.

    • Alon Levy

      Long-term maintenance is a standard component for European orders nowadays. It’s not always spelled out, but see e.g. here, with breakdown of manufacturing vs. maintenance costs here. Without the maintenance, those trains (250 km/h non-tilting Pendolini) are just $25 million per set, with 7 cars and about the same number of seats as the Avelia Liberty (400) since it’s an EMU.

      • Nathanael

        In practice, I doubt that either the tilt or the high top speed increased the price much. The bizarre FRA requirements probably jacked the price up so high the other features didn’t matter much. 😦

        • Nathanael

          …and this is Amtrak’s rather unpleasant bind. Their opportunities to order rolling stock are random, unpredictable, and unreliable; when they order rolling stock, they must therefore order *all-rounder* rolling stock which can be redeployed. This is not ideal but given their recurring funding situation it is the only rational thing to do. If they ever got steadier funding, they would be able to make more sensible procurements.

          • Adirondacker12800

            The new Acelas are not going to be leaving the NEC before 2040, 2045. Unless a lot of money drops out of the sky.

          • Alon Levy

            Amtrak’s unpleasant bind is that it thinks it’s rational to respond to uncertain funding by spending $2 billion on rolling stock, instead of drawing up where it’s best to spend $2 billion, which probably means $800 million on rolling stock and $1.2 billion on priority infra upgrades (superelevating the tracks at Zoo and on the South Station approach, New Rochelle rail-on-rail grade sep, constant tension catenary, Elizabeth curve fix).

          • Patrice

            @Alon: regarding the 1.2 billion of priority infra upgrades, how much would only the constant tension catenary be (NYP-WAS except for the in progress NJ stretch)? Would that be a multi billion price tag?

          • Alon Levy

            I don’t know. In theory, it should be doable NY-DC for maybe $900 million – that’s if anything more than how much electrification normally costs over that distance (New Haven-Boston was $450 million in 2016 dollars). But the price tag of constant tension in Jersey is much higher, partly for reasons that are not Amtrak’s fault, like having active transmission lines over the overhead wires (see list of problems here), and partly for reasons that boil down to “Amtrak overbuilds everything.” But I want to emphasize that constant-tension is only useful in conjunction with both better rolling stock for better acceleration and some curve fixes like Elizabeth; just being able to go from 135 mph to 160 mph or whatever on short segments is kind of useless.

          • Adirondacker12800

            It’s 80 years old and needs to be replaced. They are designing it so that someday it can be converted to 25kV 60Hz. As long as they are ripping everything out they might as well do constant tension in the places it makes sense.

  2. Benjamin Turon

    I would agree that designing a new high-speed train for the NEC with a commercial speed higher than 160-mph makes little in sense, since there is no infrastructure plans that are planned, approved, designed, or funded that would create 186-mph track on the NEC. If this significantly increases the the cost of the train-sets, then it should not be done. I was a surprised by the new train, I and many others were under the impression that Amtrak would order modified Pendolinos from Alstom.

    I wonder if the “potential” higher speed of the Avelia Liberty is a move by Alstom to make this the go to train for other “potential” high speed systems in the USA, like California? Although with a Sacramento Plant I would think Siemens would politically be the leading contender for that system, which is the only one to make it from press announcements to actual construction. All Aboard Florida of course is getting a Siemens version of the Intercity 125 HST and Texas Central is backed by Japan and will of course buy from Japan, if it ever manages to actually get built. Overall the best high speed train (or HrSR) for America is a train like the HST, for corridors in the Midwest and the Northeast.

    I also wonder if the high price is due to building the train in America (in Hornell, NY in the Southern Tier) instead of just ordering it from Alstom’s Italian plant like Virgin Trains and everyone else I think does? “Buy America” is expensive, and while OK I think for conventional rolling-stock like the Viewliner IIs or Siemen’s Brightline train-sets because America has a long history and existing plants and work forces to build such trains, and potential orders for more. Its not so for true high speed trains where there is such a limited market (once every 20 years?) so why go thru all that trouble and higher expense of creating a new production line for just 26 train-sets? That’s not done by foreign airlines for Boeing or Airbus jets! Its patriotism gone to far, and does seem to threaten to be a repeat of the Acela, which being a unique train design could not be expanded by additional coaches like Virgins Train’s Pendolinos. As a American I would have no problem riding a European built train if it was well built and saved millions or billions that could be spent on other improvements.

    Still I’m very excited about the new train, which still does promise I think to be a significant improvement over the existing Acela fleet whose greatest flaw is being to short! Why Amtrak ordered a short train (money…) when similar corridors overseas utilized train-sets 7,8,9,10, or even 16 coaches long! Why be shorter than the Intercity 125 or 225??? I do think the Acela’s are very nice looking high speed trains, in the last 12 months for the first time I got to seem them up close, both at the station (BWI) and at speed on the mainline (Rocky Neck State Park, Conn). Maybe by rail advocacy connections might get me a tour of the plant! Can’t ever ride them… toooo expensive!

    • Alon Levy

      Why would Alstom want to market this train in California? California doesn’t need tilting trains, so it’s much simpler for Alstom to market the AGV; in fact, the California HSR technical documents use the AGV as one of four example trains for purposes of figuring out schedules and infrastructure requirements.

      In theory, it’s possible to link two 200-meters trains together to form a longer train. It’s standard in France and Germany – the TGV Duplex sets are short, and run in double on the busier lines. I don’t know if the Avelia Liberty will have this capability, but I suspect it will. As it is, most of the stations from New York south have platforms that either already are long enough or can be lengthened; north of New York, more work is needed, but this is still possible if Amtrak is willing to sacrifice the New London stop. Unfortunately, Amtrak has no interest in lengthening trains and platforms – it would rather spend billions on a new tunnel to Penn Station than run longer trains. (And before you ask: no, lengthening platforms isn’t billions; installing high platforms from scratch is a couple million per platform face on the MBTA, which is not a frugal agency.)

      The problem with the foot-in-the-door theory is that it compels Alstom to build a top-notch train, which may not be possible given the specs. If there’s a groundswell of HSR construction in the next 20 years, and Alstom is remembered for building a high-maintenance lemon, it’s unlikely to get orders for a while. The risk here is greater than the reward.

      • Benjamin Turon

        I was thinking that Alstom would sell future orders of the Avelia Liberty without the tilt component, just like it as done with the Pendolino, for example to Poland and China. California wants a top speed of 200-mph or more, and not 186-mph as the Alstom press release says their new train will be capable of in the future. There really is not much high speed rail market in the USA outside Amtrak with its NEC and California, if the Texas Central gets built it will be the Shinkansen, and who knows about other HSR projects like the Las Vegas-LA XpressWest or the Twin Cities-Rochester proposals in Minnesota.

        HSR takes big money from

        • Alon Levy

          California wants EMUs; it’s spelled in the design docs.

          Also, I think you hit submit before your comment was finished, judging by the sentence fragment at the end.

      • michael.r.james

        Alon wrote:

        The problem with the foot-in-the-door theory is that it compels Alstom to build a top-notch train, which may not be possible given the specs. If there’s a groundswell of HSR construction in the next 20 years

        Benjamin Turon wrote:
        blockquote>French, Chinese, or Japanese think America is ready to go big on HSR, after all look at 50 years of failure after failure

        Guilty as charged. Seriously, at some point Americans and transport planners have to catch up, don’t they? Has anyone looked at the projections of air travel and the horrendous and expensive airport expansion required, decades into the future? HSR is proven to reduce enormously the air traffic between city-pairs of about 500km separation. And how happy with air travel are we, before it gets even worse?
        Finally, most Americans who experience proper HSR become converts. Yes, all old arguments but at some point all this has to turn the tide.

        Also, re Alstom and this kind of tilt-train, isn’t it consistent with the strategy of running HSR onto standard (non-straightened) track to serve more of a rail network with direct trains, ie. trains that for most of the journey run at full speed on special HST but travel some of the terminal journey at slower speed on regular track (but still faster than regular trains)? This was partly why Alstom purchased Pendolino. And Alstom has a lot more experience and resources (and highly-branched domestic rail network) to improve Pendolino (to make it more reliable etc). Many of the rail networks in Europe, especially the bigger countries like France and Germany can benefit from such an approach. I would have thought the same would apply to parts of the US.

        • Alon Levy

          Re Alstom and this tilt-train:

          1. Evidently, very few people are buying the tilting Pendolino nowadays. In the last few years, SBB exercised options on an order from 2004, which were cheap, and NTV bought new trains, at $65 million per trainset (so, almost as bad as the Amtrak order).

          2. There’s a very good reason not to run high-speed trains on legacy track at low speed too much: high-speed trainsets are more expensive, so it’s best if they spend as much time as possible running at high speed. Of course, this must be balanced against the desirability of direct service. But evidently, France is building LGVs on the most popular classic lines used by TGVs.

          3. There aren’t a lot of lines in the world where high-speed tilting trains (not counting Talgos or active suspension) are useful. In France, the only one is Paris-Nice, which requires high speed north of Marseille and tilting east of Marseille. But the Marseille-Nice line is already passenger-primary, with a lot of superelevation, and regulations that permit around 175 mm of cant deficiency even without tilting; tilting doesn’t speed up trains too much in that environment. SNCF isn’t buying these trains for one peripheral line. The NEC could use tilting at high speed (although it’s not a huge game changer, just a nice to have), but the only other first-world line for which this is even semi-useful is the West Coast Main Line, which won’t do it because of capacity problems. If Amtrak had the in-house expertise of the JRs, it could develop such a train. But it clearly doesn’t, and paying $9 million per car is frankly not worth it.

          • Adirondacker12800

            A tilting Kodama from Harrisburg can go through the tunnel and make better speed on the squiggly bits between North Broad and Yardley and all the squiggly bits between New Rochelle and New Haven. And squeeze a few minutes between New Haven and Springfield. Or the squiggly bits between New Haven and Westerley. The Nozomi from Cleveland can bypass Philadelphia and use the NEC, without needing tilt. And the bypass all the squiggly bits in Connecticut.

      • EJ

        Surely at less important stations where platforms can’t be lengthened they could simply use selective door opening? They already do at some of the smaller stations served by the NE Regional – I was just on one the other day where they made an announcement to the effect of “if you are detraining at Aberdeen, MD, you need to be in cars 1, 2, or 3, as only those doors will open.”

        • Adirondacker12800

          The new trains won’t be stopping at those stations. Just like the current Acelas don’t stop at those stations.

    • michael.r.james

      “Buy America” is expensive …

      That is not so simple a calculation. Especially if the ultimate payer is the government, which I am going to assume is the case for Amtrak. The point of course is that at the most superficial level, about 30% of expenditures are rapidly returned to the government in taxes, and quite quickly. Then there is the effect on a nation’s trade balance. Then there is the saving in unemployment benefits etc. Then at a deeper level there is the building of industrial capacity: heavy industrial manufacture like this will stimulate a whole supply-chain of smaller manufacturing ecosystem (some of whom will be thus enabled to make products for other customers etc.) All of which can also ultimately lead to exports.
      If any of that scenario sounds familiar it should because it is exactly the way all industrial powerhouse nations got built, including the USA, Japan and the NICs (Korea, Taiwan) and of course today China. No accident that China did technical transfer deals, and largely local manufacture with all the major HSR manufacturers.

      • Alon Levy

        In Korea and Japan, the focus was always on exports rather than on building a captive domestic market.

        Amtrak is not doing any tech transfer, and wouldn’t be able to get a tech transfer deal from anyone. The broader issue is that very little of the extra expense ends up paying line workers in the US. The trains are manufactured over a few years, at a factory employing 750 people. That’s many hundreds of thousands of dollars per job-year; those jobs aren’t going to pay that much – the balance is going to go to materials, and to consultants. Lots and lots of consultants in France who’d love to oversee the new train’s design and engineering. To say nothing of net profit for Alstom, which can highball bids because there isn’t that much competition (who else has a factory in the US that can get its foot in the door? Siemens, maybe Bombardier, but none of the smaller vendors).

        • Adirondacker12800

          How many Shinkansen have the Japanese exported? Compared to their domestic market.
          Buy American or Buy America rules aren’t unlimited pots of money. If the vendors can steeply discount an import, the import is allowed. The parts they source from plants in the US are filled with workers being paid a wage that gets taxed. So do their suppliers. All the way back to the miners who grub the raw materials out of the ground. The people who made the paper towels in the lunchroom, the truck driver who delivered the fuel….
          The consultants also get paid and pay taxes on that. The consultants and lawyers for the inevitable lawsuits are going to be there no matter where something is made.

          • Benjamin Turon

            The Japanese have not exported much “pure-Shinkansens” but more conventional equipment is different. The only Shinkansen overseas so far is in Taiwan but look at all the commuter trains they have exported. And of course you have the new high speed intercity trains that Japan has sold the British, the high speed Javelin on HS-1 and the Class 800/801 Super Express Trains. But most important to the Japanese rail manufacturers is that they have a huge captive domestic market, and the Shinkansen train-sets at the end of there decade long lifespans seem to be usually replace with new trains instead of being rebuilt. Compare the generations of Bullet Trains to the TGV! That also adds to orders I would think.

          • Alon Levy

            It’s not really correct to call the domestic market captive. There’s extensive competition within Japan, between Kawasaki, Hitachi, Kinki Sharyo, etc.; the domestic market is vast, numbering in thousands of cars purchased per year. The market is closed to foreign companies, but the difference between calling Japan a captive domestic market and calling the first world writ large a captive domestic market that exports to the third world is one of degree, not kind.

  3. Benjamin Turon

    Sorry, my computer glitched on me… what I was going to say I find in funny how foreigners be they French, Chinese, or Japanese think America is ready to go big on HSR, after all look at 50 years of failure after failure (California Bullet, Texas TGV, Florida’s FOX). And right now this is the land of Donald Trump, Koch Brothers, and the Tea Party. Even among pro-infrastructure its widely suspected that their support for HSR is broad but shallow.

    And very, very, few in government or the media actually know anything about HSR, or railroading in general. I had an exchange with a editor at the Wall Street Journal who utterly failed to understand that like Califorina the TGV operates over a “Blended System”. He told me that he rode the Eurostar and it was all new tracks with no freight trains. I told him wasn’t the Gare de Nord a Victorian Station full of commuter trains? Even HSR boosters are pretty clueless… claiming that no government money is needed and that the construction costs can all be paid for by the private sector.

    And some current efforts are counterproductive, like Northeast Maglev for example. And of course there is the hype and misinformation of Hyperloop and Self-Driving Cars, “who needs HSR” is a common attitude in the media. The USA in terms of geography economics, and population could be a big market for HSR, but the past and current circumstances would seem to suggest that its not going to happen anytime soon. I have been promoting intercity rail for a decade in Upstate NY, and have little to show for it, our state’s Empire Corridor HSR Study (well.. higher speed rail) is years overdue, having completely disappear into a black hole at DOT. I did help NYSDOT w ith their study by providing them sources from credible researchers that for intercity rail corridors where ridership will be below 10 million annually (like NYC-Albany-Buffalo) that new build HSR is not in general economic feasible.

    It would be a pretty good bet that Alstom will build this “Acela 2.0” train at Hornell… and that will be the end of HSR production at that facility. Much better to focus on commuter, subway, and light-rail vehicles, but with the Chinese and Koreans building US plants joining the French, Germans, Japanese, or Spanish… that market may have too many competitors.

    • Benjamin Turon

      I thought that Amtrak’s specs for the Acela replacement also called for an EMU, and for a “off-the-shelf” train thus the Pendolino… that was a pretty strong consensus I thought and I was very surprised by the end result. Did Amtrak change the specifications? I read a news report that Bombardier dropped out because Amtrak did change the specifications when it dropped the idea of a joint order with the CHSRA. The Avelia Liberty actually reminds be a lot of the APT, it includes a lot of the same features, power cars, articulation, and tilt. I hope that is not a bad omen!

    • Alon Levy

      Wait, did your WSJ correspondent not notice the trains going slower in the Chunnel than in the open? The Chunnel has overgenerous dimensions and trains can go at any speed there, but they only do 160 km/h to avoid conflicting with slower freight trains tooooo much.

      I do think California will end up building something. The problem with opposition to HSR is that it’s too partisan. The highway lobby invested in hackhouses like Demographia and Reason’s transportation division, and effectively ended up polarizing US transportation policy, breaking the consensus (to which APTA is still tethered) of tons of highway spending and a little bit of transit spending. The result is that in California, opposition to HSR is too concentrated among Republicans, who are a permanent minority party. HSR is a flashy project that lets state politicians show that they’re doing something; they don’t care enough to do it right, but evidently they are directing cap-and-trade funds there. If California HSR succeeds, which I think it will despite Pacheco and Palmdale, then other states will copy it; polarization won’t last forever.

      Is New York-Buffalo really a sub-10 million corridor? Obviously the ridership today is a rounding error, but, on a gravity model calibrated to Shinkansen, TGV, and AVE ridership, ridership assuming there’s HSR is pretty close to 10 million.

      • Adirondacker12800

        Albany to Buffalo is the same corridor that would get people to Cleveland and Toronto. And all of New England could use it. Albany to NYC is the corridor Philadelphians and very likely Washingtonians would be using to get to Montreal. And all of New England except for Vermont.
        ….. Conventional speed from Burlington VT to Whitehall NY and then 3.5, 4 hours to Washington is quite good. Conventional speed from Burlington to Whitehall and then 2 hours to Boston is quite good too.

      • Benjamin Turon

        “Is New York-Buffalo really a sub-10 million corridor? Obviously the ridership today is a rounding error, but, on a gravity model calibrated to Shinkansen, TGV, and AVE ridership, ridership assuming there’s HSR is pretty close to 10 million.”

        The population of Upstate NY is not that big anymore, several decades of deindustrialization, urban decline, and population loss. And you don’t have strong central business districts (extensive suburbanization) and extensive high quality transit systems, only Buffalo has rail transit and its one line of a few miles along Main Street, I use to ride it while at university (UB). Buffalo and Rochester have metro populations of just over a million and Syracuse over 600,000 and Utica just under 300,000. The distance of Buffalo from New York City is about 375 miles via the Southern Tier or 435 miles along the NYS Thruway via Albany, NY. According to a recent (incomplete study) 160-220-mph HSR service would have a ridership of about 5-million annually on this corridor, NYC-Buffalo, and that it would require a operating subsidy.

        Maybe if this was Europe or East Asia things would be different, but unlike Houston-Dallas or Las Vegas-LA or LA-Bay Area this doesn’t seem to be a good place to build a new HSR line in America costing many tens of billions of dollars in my opinion. The state is not going to spend that money and it isn’t coming from the Feds. That’s a shame, because at university I enter rail advocacy with the idea of a Upstate “Shinkansen”… literally transit-oriented development writ large thru HSR. After a year or two of research I downsized to the Intercity 125 as a model… and now I’m at 110-mph south Albany with hourly service as the most realistic goal… basically “All Aboard Florida”.

        The strongest rail market is NYC-Albany (Capital District) which is 141 miles away (Penn Station-Rensselaer) with a metro population of 1,170,483 (Albany-Schenectady-Troy-Saratoga), it’s the state capital, it’s got the “Tech Valley” technology initiative, and is the only part of Upstate NY with economic growth. NYC-Albany is the busiest single city pair outside the NEC and Albany-Rensselaer is the 9th busiest Amtrak station. The ridership of the “Empire Service” in 2015 was 1,152,536 NYC-Albany and 403,985 Albany-Toronto.

        There is lot of work (from our humble standpoint) being done thanks to the ARRA of 2009 and TIGER grants (with state matching funds) including signaling and grade crossing improvements south of Albany, the long awaited 4th platform track at Rensselaer, the long awaited 2nd track Albany-Schenectady, and new stations at Schenectady, Rochester, and Niagara Falls. The next big projects to get the GE dual-modes replace, they are all going over their 20-year life expectancy and reliability is a growing issue.

        The federally mandated and funded Empire Corridor HSR EIS study started in 2010 has vanished without a trace, the big issue is NYS fighting with CSX which owns the formerly four-track former-New York Central mainline from Schenectady to Buffalo. The state’s assumed preferred option calls for a new dedicated third track for passenger trains at 110 and CSX said no. The right plan would call for three shared tracks at 90, it would reduce travel time almost as much as the 110 plan and costs a bit less, and CSX would be far more likely to say “yes”… but this New York State so logic and pragmatism doesn’t apply.

        High Speed Rail Empire Corridor Public Outreach Video

        I had a lot of issues with the above study, they rejected bidirectional sets because they told me people won’t ride backwards, and were mystified when I told them that you or the train crew can easily turn seats using foot petals are a button at the end of the coach… like in Japan. They rejected the 160 alternative but studied a nearly identical 125 alternative that would build a new electric high speed railway from Albany to Buffalo, but run on diesel power down the Hudson Valley. The rejected my suggestion to electrify the Hudson Line as too complex and expensive or forgoing electrification and using Bombardier’s 160-mph JetTrain. And forget about Britain’s new Class 800 bi-modal trains from Hitachi. Overall both NYSDOT and the consultants seem to have glaring gaps in their knowledge of modern intercity services overseas. I think the effects of this issue of lack of experience or knowledge about HSR in the USA can be seen at the CHSRA, Amtrak, and USDOT/FRA.

        Well enough on my lamenting on the Empire Corridor when the subject is the Avelia Liberty. Overall I remain optimistic because gloom will get you nowhere, besides I like trains and when a train rushes past at even “only” 110 or 135 mph it’s pretty cool!

        • Benjamin Turon

          P.S. The Empire Corridor study had a rejected $1.5 billion alternative (79A) which made minor improvements west of Schenectady but major (relatively) south of Albany reducing travel time to 2:10 from 2:30 and increased frequency from 13 to 16 round-trips a day to New York City from Albany. Ridership was predicted to increase by about half million to over 2 million total annually, and there would be a annual operating profit instead of a loss. This alternative was rejected early in the DEIS process.

          • grendelkhan

            I’m the opposite of an expert here, but “this passenger rail service would turn an operating profit” seems like headline news, especially outside of the Northeast Corridor. How solid are those ridership estimates? And why on earth was a plan that would increase frequency, decrease travel times, increase ridership *and* run at a profit rejected? Am I missing something?

          • Benjamin Turon


            It was part of the initial EIS study for the corridor, one of two dozen options of which only I think 5 including the “Base Alternative” was chosen for further study. Without making a big investment west of Albany while increasing the frequency south of Albany while also reducing the travel time NYC-Albany would maximize revenue south of Albany while not increasing costs west of Albany, thus the new ridership and revenue would be enough to outweigh the loss making service west of Albany to Niagara Falls. However this would be politically unacceptable because a lot of folks (politicians and voters) want HSR out there too in Central and Western NY, so this plan was not followed up upon for further study or consideration. In reality nothing really changes NYC-Albany in the other alternatives, only that additional train frequencies west of Albany increases costs higher that new revenues so the entire state wide service has a year-to-year operating deficit.

        • Nathanael

          Ben, do you work with NARP and ESPA? I’ll try to connect with you. There seems to be a consensus among advocates that we want 90 mph track *geometrically upgradable to 110 when CSX stops having panic attacks*. There are so many things wrong with NYSDOT’s proposals that it’s not funny, however; their designs are suspiciously poorly designed for both freight and passenger purposes. Maybe what we really need is to present a proposal which has a track engineer behind it who has designed something, for free, which is more suitable for both CSX and passengers.

          • Benjamin Turon

            Yes, I’m the “Social Media Coordinator” for ESPA, in other words I do the posting on our Facebook page. Three shared mainline tracks at 90-mph Hoffmans-Buffalo is the best idea opposed to the third “dedicated” track for exclusive passenger use at 90 or 110 that the EIS laid-out. It will allow for a increase in passenger train frequency, a big increase in on-time performance, and a significant reduction in travel times, cutting about an hour off the current 5 hr 30 min schedule. It also benefits freight too since the increase capacity would be shared with CSX. A big problem on the Chicago Line is with only two tracks, annual summer maintenance regularly reduces segments to single tracks on what is CSX’s busiest mainline.

            My understanding is that if or when the HSR EIS for the Empire Corridor is ever completed and released to the public by NYSDOT, it may be a more pragmatic and realistic plan then the DEIS alternatives presented at public hearings several years ago. Any final alternative needs the blessing of CSX, or its a no go. The freight railroad which owns the ROW allegedly would support three shared mainline tracks (there was once four,, but the track centers are wider now) at 90-mph. That is my understanding of the current situation.


          • Adirondacker12800

            All the trains move at the speed of the freight trains on busy freight lines.

  4. Demetrius

    Hello Alon,
    Amtrak did make a rather decisive business move that shifted some positivity their way with the hiring of Wic Moorman who recently retired from Norfolk Southern after turning the company around during his tenure. I’m still not thrilled about the price of these new trainsets, especially considering the fact that they will be running on infrastructure that is more than a century old in some places, and only achieve marginally faster speeds. Not long ago, the Feds were looking into raising the top speed for Acela service to 160mph in the high speed section, and that costs nothing. All in all, this is just a political power move by Amtrak.
    I also invite you to see more into my thoughts along with another correspondent from the High Speed Rail America Club ( . I will be sharing this blog post on the Facebook group as well, I thoroughly enjoyed your insight in your previous post about the NEC 90% cheaper.
    Best regards,
    Demetrius Villa

  5. crzwdjk

    Here’s a theory: maybe there’s actually a reason behind the seemingly crazy specs and Amtrak has a good reason for asking for these trains. Specifically, the combination of 300 km/h and tilting sounds like it would be actually impossible to fit into an FRA compliant package, and there are no current regulations that cover rolling stock designed for those speeds. So maybe the point of all of this is to force the FRA’s hand and make the long-awaited reform actually happen.

    • Jon

      If you look at the new NPRM from the FRA for Tier III compliance, I think Amtrak had to buy a 300 km/hr train set to get around the tier II requirements used for the original Acela. Tier III graduates you to Euro style aluminum trains; mostly off the shelf UK designs should work, although the loading gauge is narrower, so some aisle and seat width may be lost importing an off-the-shelf design. The only thing the FRA admits to being a problem is seat attachments, and they give a little on that even from their typical 8g acceleration resistance requirements.

      The new bins are:
      Tier I: 0-125 mph; historic FRA compliance standards
      Tier II: 125-160 mph; Acela express standards (take a Tier I set and make it heavier; changed top speed from 150 to 160)
      new Tier III: greater than 160 mph; use Euro norms and some UK standards with better seat attachments to the trainset

      Tier I alternative: 0-125 mph; comply with Tier III crash standards

      Much of the testing can now be done piecewise and with FEA computer modeling, instead of full-scale testing that was destructive if failed.

  6. Patrice

    I’m overall very happy with the end result, except for the 5 year wait to ride the trains. The trains are cleverly financed, plenty fast, look good, modular and carry 40% more fare-payers. Passengers will be thoroughly comfortable in their seats. Look for Amtrak to increase its already dominant air-rail market share. I was really hoping the choice would be between Alstom, which has decades of successful TGV experience in France & abroad, and Siemens, whose Sacramento factory is steadily cranking out good products on time! I suspect that the 186+ mph requirement was to try to weed out some of the second-tier suppliers of HSR equipment, no names mentioned. As for tilting tech, it does help on curvy track (12 complete circles between NYC-BOS) and lord knows track straightening & takings just aren’t in the cards in the near future. Let’s focus on expanding electrification and seeing these trains someday in Albany, Harrisburg, Richmond, etc. And why not buy a faster train even if the tracks aren’t up to true HSR standards? The media can continue to write articles to remind/educate people about how slow American trains are compared with the rest of the world (i.e. shame Congress). The US is a country after all with lots of highly-overpowered vehicles driving on pothole-filled roads and rusty bridges. Roads and tracks just aren’t sexy. Only true train nerds like us see the condition of rail, ties and catenary.

    • Ryan

      “…track straightening & takings just aren’t in the cards in the near future”

      Maybe not in leafy southern Connecticut, where the trains already run – but where the tracks don’t exist, or are in such poor shape as to be completely unworkable, then takings are absolutely on the table. Actually, in those cases, it is impossible to do anything without takings – takings to connect a ‘bypass’ to the main line on either end of the segment it is meant to bypass, and probably substantial takings to jam a square peg into a round hole as well.

      In general, there’s no way to do anything substantial without expanding the amount of land we have available to do it in. I like the purchase of the new trainsets, as long as it is part of a comprehensive plan that includes some amount of route modification.

      However, it is important to acknowledge that route modification is unavoidable and therefore so are takings.

  7. Max Wyss

    A few comments about the SBB and their tilting train experience:

    The serious issues were with the ETR 470 Pendolino trains. The tilting system was actually among the most reliable components. They could correlate the reliability with the level and quality of maintenance, and there, it seemed as if the Italians did a particularly bad job with the ETR 470 (the ETR 460, without 15 kV/16.7 Hz equipment seem to have a much better reliability). It also appears as if the Gotthard line is extremely hard on the trains, as there were fewer issues when they were running over the Lötschberg (pass) – Simplon route.

    The second generation Pendolino (ETR 610 / Class 503) are more reliable, and the main maintenance base is in Switzerland. However, there are some issues, mainly in the “comfort” area (aka restaurant), such as dishwasher not working, or espresso machine not working, or problems with the doors of the accessible toilets).

    When the new trains (Giruno) are in service on the Gotthard line, they will be concentrated on the Genève- / Basel-Bern-(Lötschberg base tunnel) /-Brig -Simplon-Milano services. A factoid: These trains are running at 250 km/h through the Lötschberg Base Tunnel (using ETCS L2).

    Then, the SBB has 50 tilting trainsets for domestic services (ICN, Intercity Neigezug). Their reliability is quite good, and they are also used on the Gotthard (besides their main field on the Genève – Neuchâtel – Olten – Zürich – St.Gallen route. Technically, they are a bit different, as they are using the SIG tilting mechanism (which is electro-mechanical, as opposed original FIAT’s hydraulic mechanisms). This is also what the Virgin Pendolini in the UK are using. The essentially Swiss-built ICN are good workhorses, and quite reliable.

    That said, the SBB is probably the biggest, but if not, then the second biggest operator of tilting trains. That they did order a non-tilting train for the Gotthard route has also the reason that those trains are supposed to be certified not only for Italy, but also for Germany and Austria (towards their end of life, they may even make it to München…). The planned use in Germany is also the main reason why they have a few doors at 760 mm above rail head.

  8. Oreg Meyer

    So why does Amtrak make such questionable decisions? Cui bono? Are there any political games involved? Is it really just utter incompetence?
    On the other hand, why do the Swiss always seem to do exactly the right thing? Are they simply the most competent train experts in the world, or is there a system of incentives that they got right?

  9. Benjamin Turon

    The Railway Gazette had this article a few months ago… a friend e-mailed it to me… “Acela influences Alstom’s HS2 concept train”

    “While the double-deck train on display in London is still a conceptual design, Railway Gazette understands that the single-deck proposal [‘classic-compatible’ HS2 train] borrows heavily from the fleet Alstom is to supply to Amtrak for its premium Acela Express service linking Boston with New York and Washington DC. Alstom expects final confirmation of this order ‘within the next couple of months’. “

  10. Ryan

    Allow me to object to the notion that tilting, and especially high-speed tilting, have limited value at very high levels of infrastructure investment.

    The technology is unproven, yes, however there is enormous potential especially in regions such as Upstate New York, Northern New England, and Western Pennsylvania to completely change the calculus on how to best utilize very high levels of infrastructure investment. Tunneling as a rule will never be less expensive per km than straight surface construction, and a new technology that could open up a lot of these highly curved mountainous regions to surface construction with less or even no landscaping required (where landscaping here means “blasting,” which in turn means even more environmental whining than we already get any time anybody wants to build anything) is game-changing. A handful of ~1~2 km tunnels to get past the most completely unworkable parts of the Allegheny Mountains and mostly surface construction the rest of the way is a much easier sell than 40, 50, 60 km tunnels would be, yet PIT – PHL would still absolutely fall under very high levels of infrastructure investment. Ditto BOS – MTL, which is back on the table with high-speed tilting (neither the NH/VT route nor the MA/NY route are really doable at high speeds without tilt, and both require substantial infrastructure investments either way), and a number of other corridors that might have otherwise been relegated permanently to a second-tier status as “Amtrak+ routes.”

    • Adirondacker12800

      Google maps says it’s 386 miles between Boston and Montreal via Burlington and 387 via Plattsburg. Spend a whole lot of money going through the Green Mountains and it connects New England to Montreal. Spend a whole lot of money going through the Berkshires and it connects New England to Montreal. And Upstate New York. And Toronto. And Cleveland.

      • Ryan

        And either way you’re still going through an awful lot of mountains.

        You’ll note I acknowledged both alternatives.

        • Adirondacker12800

          Going through an awful lot of mountains is unavoidable. The alternatives are more airport and highway. Which aren’t cheap either.

    • Alon Levy

      I have not tried different alignments across the Appalachians with different assumptions on curve radius. I doubt there will be a big difference, because of an important feature of high-speed lines: they’re already highly superelevated. The Pendolino was a game changer on legacy lines with limited superelevation because of mixed traffic with freight and slow local trains. It sped up trains by 30% on curves; this can be done by going from 130 mm to 270 mm cant deficiency on a line with 70 mm cant (340/200 = 1.7 ~ 1.3^2). But if the line has no freight trains, and the regional trains that use it are pretty fast, then 180 mm superelevation is trivial to install, and the difference drops from 30% to 20%.

      This is especially true for high-speed trains. First, the maximum possible cant on lines that will never host slower trains is even higher, 200 mm on the Shinkansen. Second, the research done at KTH here and here suggests that at ~350 km/h, the highest possible cant deficiency is less than what Pendolini achieve at <200 km/h speed. The former study finds that 250 mm is okay with adequate safety margin, the latter study assumes even worse. So at best you're getting up to 250 mm cant deficiency, where existing proven technology goes up to 175 on the Shinkansen and 180 using the Talgo system. So now this extra canting gets you 10% more speed on curves, or, alternatives, 1/6 reduction in minimum curve radius. Nothing to sneeze at, but also nothing Earth-shattering.

      While I haven't looked much into Philly-Pittsburgh, I did recently draw a possible alignment for high-speed rail between New York and Albany, through the Hudson Highlands. It requires 6 km of tunnel assuming 360 km/h top speed, 200 cant, 175 mm cant deficiency, and 3.5% grades. If you instead assume 250 mm cant deficiency, the required tunnel length is the same, because both tunnels on the line go under hills that are unavoidable without very tight curves. If you allow 5% grades, which Shinkansen are capable of climbing, then fewer viaducts are required, but the required tunnel length still stays the same, because the hills are so steep that small changes in ruling grade matter very little. (In contrast, when Clem Tillier came up with alignments across Tejon Pass, he did find significant reductions going from 3.5% to 5%, but again nothing Earth-shattering.)

  11. Benjamin Turon

    Tilting for high speed trains on new lines I imagine should be limited to the passive 2-degree air-suspension system you see on the newest Bullet Trains on the Tokaido Shinkansen. Its relatively inexpensive to install and maintain from what I have read, in fact I think such low-cost systems would be worthwhile for many “Higher Speed” conventional trains like new train-sets for the Empire Service, an extra 5 to 10 to 15 mph around curves in the Hudson and Mohawk Valley would eliminated the need for expensive civil engineering to reduce curvature. However for a new high speed railway, you just go thru hills and over valleys in tunnels and on viaducts, look at photos of the Taiwan Shankansen… in one aerial photo its tunnel-bridge-tunnel-bridge-tunnel-bridge in a one or two mile section, as it passes thru ridge lines. Mountain railroad will always be expensive (and urban too!) and tilting is more useful for upgrading service on existing lines to reduce the costs of the civil engineering to increase speeds (average and top), and to avoid NIMBY outcry from taking land to straighten tracks.

    • Luke Richardson

      The same is definitely true in China as well. Mountainous HSR in China is either in tunnel or on viaduct almost exclusively, although there’s lot a lot of grade level HSR in China to begin with. Even in flat areas HSR is almost exclusively on viaduct, espcially for the highest speed lines. Only lines rated at 200km/h or less have any at-grade sections.

      • Benjamin Turon

        The Japanese and Chinese seem to build their new HSRwys like New York and Chicago built their elevated railways, one long viaduct stretching to the horizon (see early pictures of Queens where the rapid transit lines expanded into undeveloped greenfield areas)… you see that with the Taiwan Shinkansen. The current Chinese HSR proposal in Minnesota (Twin Cities-Rochester) and the Japanese HSR proposal in Texas (Dallas-Houston) would run almost exclusively on viaducts over the rural prairie lands.

        • michael.r.james

          Benjamin Turon wrote:

          The Japanese and Chinese seem to build their new HSRwys like New York and Chicago built their elevated railways, one long viaduct stretching to the horizon …

          Given the density of roads, canals and rivers, and not least existing rail (and the creeping temptation to share lines with slower traffic including freight), in these densely inhabited parts of the world (China’s east coast strip, Japan and the NEC), it probably makes a lot of sense, in engineering, political and economic terms. Most of the Beijing-Shanghai 1318-kilometre route is elevated and seems to hold every kind of record for train infrastructure:

          There are 244 bridges along the line. The 164-km long Danyang–Kunshan Grand Bridge is the longest bridge in the world,[15] the 114-km long viaduct bridge between Langfang and Qingxian is the second longest in the world, and the viaduct between Beijing’s 4th Ring Road and Langfang is the fifth longest. The line also includes 22 tunnels, totaling 16.1 km. 1,268 km of the length is ballastless.

          Incidentally one reason for building it elevated was to relieve the heritage rail infrastructure which is increasingly congested freight and thus a major economic impact.

  12. Benjamin Turon

    NEWS LINK – US Senator Blumenthal D-Conn, And Others Threaten to Tie Selves to Tracks in Response to FRA’s Old Lyme NEC Future High-Speed Rail Proposal for a Rail Route Through the Center of Old Lyme

    EXCERPT: “The [Federal Railroad Association’s] half-baked, harebrained proposals to reroute Amtrak through Old Lyme fail to reflect the priorities and needs of Old Lyme residents and those who work and live in surrounding shoreline communities. The FRA’s resources would be better spent on making critical rail improvements in the region, not on unrealistic proposals that are met with strong, well-merited opposition,” said Blumenthal. I hope that the FRA will fully listen to the community’s concerns and drop this misguided plan.” A high-speed Amtrak bypass line connecting Old Saybrook and Rhode Island and cutting through historic Old Lyme has been proposed by the FRA, Fox 61 reported. It is considered by the agency as a better way to connect New York City, Boston and Washington D.C. Blumenthal was adamant that he would block the proposal as it stands by fighting as long and hard as possible to do so, CT Mirror reported. Many residents and officials made clear they are not opposed to high–speed rail, but would rather see the existing infrastructure improved.


    • Alon Levy

      To be fair to Blumenthal, one of the two impact areas that Old Lyme is annoyed at, the transition from the Shore Line to I-95 in Old Saybrook, is in a kind of lolzy location. The Shore Line and I-95 are very close to each other a few km farther west and there’s nothing in between, which is why I advocated using it as a potential phasing break here (see under the phasing subheading, toward the end). The NEC Future Alt 1 proposal transitions closer to the river, so the transition is longer and more developed.

  13. Adirondacker12800

    It’s going to ruin the bucolic charm of the immediate vicinity of the Connecticut Turnpike!

    • Benjamin Turon

      NEWS LINK – Feds’ undisclosed ‘preferred route’ for NEC Future rail lines sparks outrage in Connecticut

      While the federal government was spending months soliciting feedback from the public on several alternatives for overhauling the railroads in the Northeast corridor, it had already identified a preferred plan that would dramatically change rail travel in Connecticut. The FRA’s selection of the favored hybrid route, even as the agency spent months collecting more than 3,000 public comments and other input on three other alternatives, has sparked outrage.


      • Benjamin Turon

        I see little need to build a new railway next to a existing railway that we have already spent a lot of money on… plus its very scenic! I would prefer a new “inland route” that would better serve other markets while also adding to capacity and cutting travel times NYC-Boston. In my written comments to NEC Future I suggested further upgrades to the existing Amtrak line to Hartford, and then building a new line to Boston from there.

  14. Pingback: Obwohl Schienen nicht passen: Amis kaufen Milliarden-Zug | Aktuelle Nachrichten
  15. Pingback: Amtrak’s Gold-Plated Trains | The Antiplanner
  16. Pingback: Are Amtrak’s New High-Speed Trains Too Good for Amtrak? | KCJones
  17. Pingback: Why Infrastructure Costs May Rise Thanks to Trump’s Cronyism – Streetsblog USA
  18. Pingback: Why Northeast Corridor Privatization is Doomed to Failure | Pedestrian Observations
  19. Pingback: Un viaje en Acela » Politikon

Leave a Reply to Alon Levy Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.