Quick Note: Good News Week

Via Systemic Failure, I learn that the FRA is finally reforming its train safety regulations on its own. This is an amazing development, partial as it is. This appears to derive from the FRA’s previous research into crash energy management, which concluded that buff strength alone did poorly at protecting train occupants. This development is especially good for the MBTA and Metra, as agencies that could make large orders, especially of EMUs if they electrify (and both have good reason to); this will allow them to obtain better EMUs, for example measured by weight, than currently run in New York and Philadelphia.

Unfortunately, the reforms are partial, and lack two elements. First, they start from past crash tests, rather than from good rolling stock, and may still require imports to undergo substantial modifications; this is not a problem for large orders, but tends to raise the unit cost for small orders. That said, the rules are being developed in consultation with representatives from many rolling stock vendors, not only the large ones as with Caltrain’s waiver application but also smaller ones such as Nippon Sharyo and Stadler. Second, they do nothing about operating rules as opposed to procurement rules; these include brake tests, cant deficiency rules (only partially reformed), and so on. Still, count this as a positive development for the FRA.

The other good transit news: the Florida East Coast Railway, a Class II railroad primarily carrying intermodal traffic between Jacksonville and Miami, is announcing a privately-funded $1 billion project to build a medium-speed line from its mainline to Orlando and run passenger trains between Orlando and Miami, making the trip in 3 hours. This corresponds to an average speed of about 80 mph, just under 130 km/h, or in other words the same as that achieved by the supposedly high-speed Acela between New York and Washington.


  1. Steven Vance

    I tell people that Metra electrifying would be good, but I’m not confident in knowing the reasons why. Could you or another reader help me list the benefits of electrification?
    1. Non-point source pollution keeps emissions away from stations
    2. Because of 1, trains don’t have to be backed into downtown terminals (not sure if this is really a benefit)
    3. Electric trains accelerate faster. Is this the case for EMUs and electric locomotives? I see Metra switching to electric locomotives before EMUs because Metra could still use all of its existing wagons.
    4. Electric trains are more efficient than diesel in terms of energy transferred per mile travel

    I’d love it if someone would write a short article about this for http://gridchicago.com, my Chicago-based sustainable transportation blog.

    • Alon Levy

      It’s mostly explanation #3, with a smattering of #1. Electric locomotives don’t perform very well, but EMUs of the modern kind do. Metra runs very short interstations in Chicago’s inner suburbs, shorter than those of many subway systems. It matters that a diesel locomotive accelerating to 60 mph loses about 70 seconds accelerating, and an EMU loses about 13; the EMU can also decelerate faster, and using better door configuration (which can be built into unpowered coaches, but would require new trains anyway) and level boarding dwell times could be significantly reduced as well.

      I don’t actually know if Chicago has anything as bad as Back Bay, but Union Station and Ogilvie are extremely noisy and presumably quite polluted, though I’m told that the trains are hooked to electricity for HEP while idling.

      Explanation #4 is not terribly important. A good DMU gets 4-5 car-mpg. In Chicago’s case, it’s probably 4 or a bit less because of all the stops. You need 12 people on the 4 mpg car to do better than a single-occupant Prius. Metra’s average occupancy is 37, according to the NTD. EMUs in principle do better, though possibly not with Chicago’s coal-fired power, but the priority is to move people from cars to trains, regardless of whether those trains get 60 or 160 passenger-mpg.

      • ant6n

        Isn’t electric operation cheaper than diesel, making it cheaper to run all-day frequent service?
        Also, noise can be an issue — you could run electric equipment much later near residential areas.

        • Danny

          It is cheaper, but nowhere near the amount necessary for it to be a deciding factor. Staffing levels are far more important.

      • Beta Magellan

        It’s unbelievable horrible at Union Station, especially their south platforms:



        I only use the north platforms, and the diesel stench there is sometimes hard to take.

        I’ve heard part of the issue is that the ventilation spaces were originally designed to clear steam from the platforms, but are fairly useless when it comes to diesel fumes and soot.

      • Nathanael

        #3 is the most critical in “urban” applications, because time is money.

        #4 is very true, but dependent largely on the difference in cost between diesel and however you generate your electricity. If you’re getting hydro from Niagara, it’s significant, if you’re getting electricity from imported oil, it’s roughly the same price. The price difference isn’t huge right now, though it will go up.

        #1 is true as well, and is a reason why people prefer to live next to electric trains rather than diesel or steam trains.

        The missing #5: capital costs. The cost of an electric engine is less than that of a diesel engine and it lasts longer, but the cost of the electric infrastructure is very large.

        The breakeven point where electricity is better comes when you have to get lots and lots and lots of trains and run them lots and lots; in all cases, electrical power is better for higher volumes of travel.

        • Alon Levy

          Is it actually true that electric locos cost less than diesel locos? I thought the cost was actually higher, though rechecking the recent monthly rolling stock market reports on Railway Gazette gives several $1.7 million/car EMU orders, and I can’t find diesel loco costs quickly.

          Either way, the maintenance cost is I believe lower (diesel engines are maintenance-intensive), and the trains last longer.

          • Max Wyss

            Diesel engines (for the US market) are cheaper because of the bigger number produced. Of course, companies like Bombardier don’t publish their prices, but it would be interesting to do a comparison among the various TRAXX models, but you may expect something between 3 and 6 M$ per unit. The price varies considerably on the options and customizations ordered, and in many cases, it also includes services and spares.

            Operation cost is definitely lower, because of longer maintenance intervals, longer lifespan (although you may have to plan for a complete replacement of the power and control electronics with a mid-life overhaul), lower energy cost thanks to regenerative braking, etc. etc.).

          • Adirondacker12800

            Shhh, dirty little secret is that the world class locomotive manufacturers don’t make their own diesel engines. They buy them from the same few vendors.

          • Nathanael

            Max Wyss is correct, of course; but if you’re not in the bizarre US market, electric engines are cheaper because of the bigger number produced. :roll eyes:

      • Steven Vance

        We have two coal-fired power plants that are getting shut down soon. But the power from there isn’t used in Chicago. Like most places, a majority of our power still comes from coal but Illinois also has a lot of nuclear power.

        I didn’t know about Metra’s average occupancy. Does NTD separate weekdays from weekends?

        • Nathanael

          Averages are crazy misleading. Metra has some very-well-performing routes, and some very-poorly-performing routes; it has some very busy time periods and some very non-busy time periods.

          You want to electrify the fullest routes, obviously. At the moment, with Metra Electric already electrified, that means BNSF first, then UP-N, then UP-NW, then Rock Island. UP is likely to be very uncooperative. Rock Island has multiple branches. Honestly the best benefit would be from electrifying the BNSF line, if an arrangement could be made with BNSF to do so.

          • Alon Levy

            If they electrify UP-N and UP-NW, then they might as well also do UP-W and have a fully electric system. The cost of electrifying the last line in a system is much less than the cost of electrifying the first line.

          • Adirondacker12800

            They electrified the first line back in the 20s. I suspect the Illinois Central fully amortized it.

    • mulad

      I’d add the fact that electrics are quieter than diesels (at least in areas where they’re not blowing horns all the time). I feel electric power brings a great benefit to people who live near rail lines — particularly near areas where trains are accelerating. They’re not silent, but they’re still quite a bit better.

    • Beta Magellan

      In Chicago’s case, there’d be significant air quality improvements on the Union Station platforms, which are currently very poorly ventilated (although this issue can conceivably be solved without electrifying a good chunk of the network, you wouldn’t get all the other advantages listed above). Decent air quality on Union Station platforms opens the door towards further improvements that could make it possible for passengers to wait on platforms (as in Europe) rather than inside CUS’s labyrinth of waiting rooms, though I’ve never taken Metra from Union Station and don’t know if Metra riders there queue up like Amtrak riders (I know at Millennium there’s sometimes a bit of a rush to get from waiting room to the platforms, which, while clean, aren’t terribly comfortable for the passenger, plus a trickle of people who board the trains minutes before they leave.)

      • Nathanael

        On these grounds, again it becomes clear that the BNSF line should be electrified first. It gives the largest reduction in sheer number of diesel engines going through and sitting in the station, and they’re on the western southern part of the station, which is the worst ventilated. If funding can be found and BNSF can be made agreeable it could work.

        Further arrangements with Amtrak and BNSF could conceivably allow for the eight Amtrak trains a day which pass along this line (currently planned to increase to ten) to change to electric engines in Aurora, giving even more benefits.

    • Andre Lot

      There is other additional advantage: electric trains are incredibly much safer to operate on tunnels, and the ventilating system they require are less complicated as there are not toxic fumes being exhausted.

      • Max Wyss

        Actually, electrified railroad tunnels do not need any ventilation at all; almost independent of their length.

        • ant6n

          You’re referring to the piston effect, right? Because that would assume separate tunnels for both directions.

          • Max Wyss

            No, there is no need for ventilation at all when you have electric operation. Just as an example, the Gotthard tunnel is one bore with a double track, and there is at least one train in the tunnel all the time.

          • Nathanael

            If your tunnel is long enough and narrow enough you need fresh air for the passengers, but that would be longer than any tunnel which exists. Normally the trains move enough air through the tunnels for the people.

            In practice the Gotthard tunnel has two intermediate ventilation points, but this is in case of emergency, if there are fires in the tunnels, not for regular use.

          • Max Wyss

            The tunnel with the two ventilation shafts is the Gotthard _Base_ tunnel, 57 km long. The current Gotthard rail tunnel (in operation since 1882; electrified since some time around 1918) has no ventilation.

          • Nathanael

            Thanks for the clarification. Lots of Gotthard tunnels. 🙂

          • Max Wyss

            Well, I forgot to mention the Gotthard Road tunnel, also 15 km long. This one has heavy ventilation…

        • Andre Lot

          @Max Wyss: some ventilation is ALWAYS needed, otherwise chances for dangerous accumulations of carbon dioxide could happen. Workers need also to have proper ventilation, and trains are not exactly airtight vessels that could handle passengers for long without ventilation.

          Of course, the ventilation systems for electric rail are far less complex and smaller, but they do still exist.

          • Max Wyss

            With the exception of very long, modern tunnels, all the ventilation you have is the natural ventilation plus the earlier mentioned pistion effect. New very long tunnels do have ventilation systems, usually in combination with evacuation tunnels etc. (such as in the Channel Tunnel, or the Gotthard Base Tunnel or Lötschberg Base Tunnel). But tunnels like the Gotthard tunnel, or even the Simplon do not have any ventilation. Neither does the Vereina tunnel or the Furka Base tunnel.

            Googling for “Vereina Tunnel Ventilation” brought up this: http://www.takesteps.org/pdf/VereinaTunnelClarification.pdf which is communication between an organization wanting to build car shuttles through a rail tunnel and the Rhaetian Railways.

            For construction/renovation work of older tunnels, they are usually working with mobile ventilation equipment.

    • Ryan Richter

      The Regional Transportation Authority, which is the parent agency of the CTA, Metra and Pace did an electrification workshop about a year ago which focused on commuter rails: http://www.rtachicago.com/index.php?Itemid=327

      I can tell you a couple of reasons why Metra is against electrification, broadly speaking:
      1. Cost. Both operating at capital, although I believe Metra could achieve economies of scale on the operating side, not to mention huge fuel savings. However, the Union Pacific, which owns three of Metra’s 11 rail lines, estimate costs between $5-7 billion to install systemwide.
      2. Intermix of freight and passenger operations raises questions about catenary structure in the ROW, particularly under bridges. Clearance height thus becomes a major issue systemwide, particularly with freight double stacks and bridges, and likely in Union Station as well.

      • Beta Magellan

        Clearance is the major reason why Metra Electric still uses DC, so it’s not surprising there’d be major issues with clearance.

        Does that $5-7 billion include rolling stock?

          • Nathanael

            The Rock Island grade separations are almost invariably OVER the other roads or railroads it meets; the same is true of the BNSF. I don’t think overhead clearance is a big issue on either of those two lines. True that the Milwaukee lines have a lot of trenching.

          • Beta Magellan

            The catenary alone isn’t the issue—bridge clearance is. Chicago’s full of grade separations which can’t necessarily accommodate both.

          • Nathanael

            Beta: For the hell of it, I counted the bridges which pass over the
            BNSF line. I ignored bridges north of the Canal St./16th St. wye because only passenger trains run north of the wye.

            (1) Canal St.
            (2) Dan Ryan Expressway
            (3) Pink Line CTA
            (4) Belt Railway of Chicago
            (5) Manufacturer’s Junction Railway (right next to the Belt)
            (6) Laramie Avenue
            (7) CN’s Iowa Central line
            (8) North Oak Street, Hinsdale
            (9) I-355
            (10) Yackley Avenue, Lisle
            (11) Columbia St., Naperville
            (12) CN’s former EJE line
            (13) N Eola Road
            (14) N Farmsworth Ave
            (15) N Ohio St
            (16) Wood St
            (17) High St

            This is really a pretty small number of overgrade bridges for this length of line.

            Obviously, some of these bridges have sufficient clearance and do not need to be raised. I would also lay bets that several of the 5 railway bridges are life-expired and need to be replaced anyway. It seems possible that track grade could be lowered (rather than bridges being raised) f necessary in many of the more rural areas far from Chicago. The largest cluster of overgrade bridges is between Naperville and Aurora (6 of 17 total, and 6 of 12 road bridges).

            Looking at it more closely, and having ridden on this line, I think it’s really just the Dan Ryan Expressway bridge which poses problems; everything else either has substantial clearance or is easy to fix. I might simply suggest an electrification gap underneath it; coast across.

        • Max Wyss

          With rail catenary, as produced by Furrer & Frey, for example, you gain at least 15 cm of clearance, which means you can increase the voltage. On the other hand, there is nothing against 3kV DC; successful networks like the Italian, the Spanish, or part of the Russian network are 3 kV DC.

          I agree that the price tag mentioned appears to be very high. How many track miles would that include?

    • Max Wyss

      To your points:

      1) This is correct; and they say that it is easier (and economically more worthwile) to produce cleaner in a big plant, as opposed to a small power plant. OTOH, modern diesel engines do get pretty good, and for automobiles, a modern diesel engine powered car is about as “clean” as a gasoline powered hybrid.

      2) It is definitely an operational advantage, but it would be even better if the network would consist of transversal lines, and not end at a downtown terminal. Modern electric is in any case push-pull, but it does not matter that much where a “locomotive” is located.

      3) Accelleration depends on the overall tractive force and the total weight of the train. Having the Swiss DPZ bi-level train for the S-Bahn Zürich in mind, we have a locomotive rated at 3200 kW continuous (that’s about 4000hp), coupled to two cars and a cab car. Tractive force is 240 kN, which is actually rated down; at the beginning, they got higher forces, but the squeal in the micro-slip state upset the residents along the line. But accelleration is kind of decent; about 0.9 m/s2.

      The 20+ year younger KISS by Stadler comes along as 4-car EMU with 8 driving axles. It is rated at 4000 kW continuous and 6000 kW short time, providing 400 kN tractive force and an accelleration of about 1.1 m/s2.

      So, EMUs can accellerate faster, and they have the advantage that the tractive force can be higher when the vehicle is full (those 10 tons or so of passengers have their advantage…). Opposed to that, a locomotive hauled train gets lower accelleration when the train is full.

      4) This is absolutely the case, particularly with modern trains and regenerative braking. This can feed back to the grid something around 30% of the energy needed to accellerate.

  2. Zmapper

    Good for the FEC! If they succeed, and they likely will considering the lack of decent non-auto transportation options between Orlando and Miami, what will this mean for the rest of the country? Could this spell the slow death of Amtrak?

    • ant6n

      Well, to what extend would private passenger rail operators have access to existing track? Afaik, Amtrak gets a special deal to access existing freight track.
      Also, would it be possible for a publicly funded train to not be operated by Amtrak?

      Amtrak owns the North-East corridor, and the operate all those federally mandated/subsidized long distance routes. It would seem that as long as that’s true, Amtrak won’t die.

      • Alon Levy

        In this particular case, it’s a railroad operating on its own tracks, so it’s not a problem. If this leads to Amtrak’s death as an operator, it could still live as the owner of most of the NEC, offering trackage rights to whoever runs trains there.

        Publicly-funded commuter lines aren’t always operated by Amtrak. There’s competition between Amtrak and foreign private operators on some lines that are contracted out, such as Metrolink. Presumably the same could be true of intercity state-supported lines.

      • Jiim

        There’s a commenter on the CAHSR blog who goes by the handle jimsf. He works for Amtrak on the San Joaquins. He says they have been told by management that there’s a possibility that BNSF will bid to operate the San Joaquins.

        Three factors:

        1. CalTrans owns much of the rolling stock that Amtrak-California uses and could make it available to another operator.

        2. BNSF owns much of the track the San Joaquins operate over and has reciprocal trackage agreements with UP which owns the rest. If UP screws with BNSF, BNSF can retaliate.

        3. BNSF already operates some western commuter routes (Sounder, at least). So it has some nearby passenger operations expertise it can call on.

        This is sort of the ideal situation for another operator to take over an Amtrak route. It’s unlikely that all these will come together for many other routes. It’s hard to get operators even for commuter routes. The last time MARC sought an operator for its Brunswick and Camden lines it got no bids. CSX agreed to continue to operate another year until MARC can issue another RFP.

    • Beta Magellan

      It’s worth noting that the FEC doesn’t carry slow, bulk traffic but rather quicker intermodal trains, so passenger rail is easier to slip into their existing operating paradigm. It’s hard to replicate was FECRR’s doing in places like the midwest, where there’s a lot more bulk freight and its more expensive for a private operator (both in terms of upgrading infrastructure to accommodate higher speeds and opportunity cost from taking up slots meant for high-profit-margin slow bulk freight).

  3. Paulus Magnus

    In partial mitigation of my earlier time calculating, it does hinge on where they mean by South Florida. It could be as slow as a 60 mile per hour average speed, although this is still only 5 mph slower than the NEC Regionals and significantly faster than current corridor services.

  4. Adirondacker12800

    and may still require imports to undergo substantial modifications; this is not a problem for large orders, but tends to raise the unit cost for small orders.

    There’s gonna be one standard for everything east of the Rockies, “Northeast Corridor”.
    Who has 85 foot long cars that are 10’6″-ish over the thresholds and have 51 inch floors in their portfolio? Anything that anyone comes up with is going to significantly modified to conform to that.

    The small orders can piggyback on the MTA and NJTransit orders. Metra orders, if Metra gets the urge to make a big order. Rotem will have “Silverliner VI” in it’s portfolio. Which could be Arrow IVs or M9/M10s Someone will have Comet VIs.

    …east of the Rockies and Salt Lake City. Salt Lake City went with “NEC” for Frontrunner….
    …wouldn’t be the awfullest thing if California decided “NEC is good enough” and then it would be everything in North America.

      • Zmapper

        48″ platforms, while acceptable for regular box cars and other standard equipment, do not allow for the passage of high-and-wide equipment easily. For example, almost 12′ wide Boeing 737 fuselages are shipped on the tracks near me. If there were high platforms, they would have to find another way to move from Kansas to Washington, possibly by oversize truck.


        Instead, 21″ should be pursued as the non-NEC platform height. Level boarding designs for this height do exist, and it matches the existing height of bi-level equipment here. Freight traffic of all sizes can pass, because the oversize part is almost always >2′ high. The only downside is that steps exist for passengers to walk over the bogies, which are closer to 3.5′ high.

        • Jiim

          The European standard is 755mm (a bit under 26 inches). If we’re going to adopt a new standard then that would be better than one that no-one else uses.

          • Adirondacker12800

            The people getting on the high speed trains from Buffalo and Pittsburgh will want to get off the train in Philadelphia, New York and Washington DC. Level boarding in Washington DC., Philadelphia and New York is going to be 48″ forever and ever. It would cost too much to convert the suburban stations around New York, not to mention convert Penn Station ( any of them ) Grand Central, 30th Street, Union Station and South Station. They also going to want to go to Cleveland,. Toledo, Detroit and Chicago. People in Toledo are going to want to go to Syracuse. Unless you want everybody to do cross platform transfers in Buffalo and Pittsburgh everything all the way out to Chicago is going to be 48″ platforms. Including Cleveland, Columbus, Cincinnati. If you are going to Cincinnati it’s not that far to Louisville. If you gotten to Louisville it’s not that far to Memphis… Memphis means New Orleans. Which means the trains running from Houston to Atlanta and Jacksonville will have 48″ platforms. Since Chicago will have 48″ platforms so will St. Louis which is handy if Columbus has them too. And Pittsburgh. Milwaukee will want to have the same platform height as Chicago which means Minneapolis will too. Or the people in Raleigh will want to go to Atlanta in addition to going to Washington DC. And the people in Charlotte will want to go to Atlanta and Birmingham in addition to Washington DC..
            Once you connect all the places less than 400 miles apart you have a network that stretches from Portland Maine to San Antonio Texas and from Minneapolis Minnesota to Miami Florida. You can throw in Toronto and Montreal too which then means most of Canada. Pretty big market.

          • Jiim

            There are currently low platforms in Blatimore and Washington. There are spare platforms at Philadelphia. Any currently four-tracked station can have two diferent platform heights. NY Penn would be a problem, as would Wilmington and Back Bay.

            48″ is not as embedded as you think. It’s not like the Bell System.

          • Alon Levy

            The problem is that those stations then force everything in the network to adapt – at a minimum, everything in the New York and Boston regions, and probably also Philadelphia, because of Market East (the intercity NEC trains would have to be New York-compatible anyway). On top of that, the Northeast is littered with commuter rail platforms that are high or partially high.

          • Adirondacker12800

            … there’s a train blocking the HSR only platform in Trenton during the evening rush hours because it’s disabled or has some other sort of emergency. How badly does that snafu everything between Sunnyside and Wilmington? How much of that snafu propagates out to the Reading side of SEPTA, the LIRR and Metro North (if Metro North is serving Penn Station?) The commuters who were expecting to have the express whisk them home are delayed, do we give them your cell phone number or let them complain to SEPTA or NJTransit and the local news outlets about being stuck behind the local? It happens during the morning rush, how’s your boss gonna feel about all the emails you are sending out to their bosses about the delay?

            [insert some Midwestern city not Chicago or Cleveland here} decides that commuter service to the suburbs makes some sense because there’s that underutilized HSR track leading into the station in the middle of their CBD. Do they build more platforms or do they use the HSR platforms? The suburban HSR station on either side of the city, where their commuter trains are going to terminate, do they build more platforms out there or use the HSR platforms? 48 inch platforms may not be the exactly perfect fit for that city but there’s nothing extraordinarily bad about them either.

          • Alon Levy

            Yes, this is true. It’s a matter of gradation, though. Are there problems with passengers illegally crossing tracks in 76-cm territory in the Netherlands? Or, do you (or others) know of problems in 55-cm territory in Switzerland and France? Clearly there are problems with American 8″ platforms, but practically none apart from suicides or people having seizures and falling on subways or on proper high-platform (about a meter or more) commuter lines…

          • Andre Lot

            Well, track crossing is extremely rare around here… But I guess that goes with other measures, such as virtually 100% track fencing or isolation (often by water ditches, but they serve as fences for all purposes), trains that don’t whistle unless on emergencies, virtually absence of un-gated at-grade crossings etc.

            In this scenario, people don’t cross tracks elsewhere, don’t even get access to tracks and therefore don’t get tempted to jump 80cm down and up.

        • Max Wyss

          21′ is about 55 cm, which is one of the European standards. Low floor units can have the floor at about that height, so level boarding is possible. This is also possible with bi-levels, even within the (compared to the US) smaller UIC loading gauge. If the gap is considered too big, it is common practice to have a step which gets pushed out towards the platform, and which still creates an almost level boarding (consider a step of 2 cm or so).

        • Alon Levy

          My thinking is that the US should split the difference and decree 48″ as the standard on lines that could plausibly be connected to the NEC someday, and 21″/55 cm on other lines. The question is what to do with California – should it be high-platform territory because all high-speed trains are high-floor, or low-platform? I know others (e.g. Richard) disagree, but I tend toward high platforms for California, but that’s because the commuter lines either have trivial freight traffic (Caltrain) or space for passenger and freight track separation (some Metrolink).

          Another solution is to have retractable platforms, as on the single car-length high platforms on the Worcester Line, which are necessary for combining ADA compliance on the high-platform commuter trains and some CSX freight traffic. It works well for commuter lines, though perhaps not for low-traffic regional lines, if the retractable platforms add cost too much.

          • Richard Mlynarik

            The reason for low(er) platform, everywhere, is bi-level trains with acceptable station dwell times.

            There’s no reason for California HSR not to be uniformly bi-level from the start and forever.

            It’s the high platforms which are, to a great extent, a relic of historical vehicle design, and there’s no overwhelming reason to go with such a compromise in a green-field new start (which is what all mainline passenger rail outside the immediate NYC region is.)

          • Alon Levy

            There are advantages to single-level, high-floor designs. The legacy advantage coming from Penn Station, Market East, and Back Bay/South Station is very large, though it’s more relevant to small Northeastern operations than to Midwestern ones. Besides that, high-floor trains can have seating continuously through the train’s body, and this reduces the capacity advantage of bilevels, and eliminates it when you look at train weight. The Class 423 beats the KISS on both seats and standing space per ton, and that’s with 3 doors rather than 2 per car. The E233 Series comes close on seats per ton, with subway-style seating and 4 doors per car; it beats the KISS on standing space per ton by a small margin, but Japan computes standing capacity based on a standard that works out to 3 per m^2 rather than 4, and if you use 4/m^2, it beats the KISS by a lot. (Both claims for the E233 remain true if you pretend every car is a motor, i.e. heavier, in order to make the power-to-weight ratio closer to that of the KISS and Class 423.)

            Outside San Francisco, there just isn’t the special set of circumstances that makes the added capacity of low-floor bilevels so important. The capacity limit is for the most part the capacity limit of running tracks with stations rather than that of the throat, and this means short dwell times are crucial. Transbay is an exception and has special needs (pun intended), and Penn is too but can’t use low-floor bilevels, but the rest aren’t in that situation. Market East is four-track on a four-track line, Boston has ample throat space and will have as many station tracks as running tunnel tracks (2 or 4) if it builds the North-South Rail Link, Chicago is going to need all the boarding speed it can get if it ever builds a station under Clinton Street. If those cities ever start having real capacity problems, they’ll be like Berlin and Munich, rather than like Zurich.

            The question then is whether it’s really worth it to design all of California around low-platform bilevels just because of Transbay throat problems. I think it’s worth looking into the cost of blowing up 301 Mission and doing Transbay right and comparing it with the cost of a marginally usable terminal.

            The main reason to stick with low platforms is that it costs a lot of money to raise platforms. But this isn’t as big of a problem in a country where both 550 mm and 1,220 mm platforms would require extra construction.

          • Adirondacker12800

            Or gauntlet tracks like Roselle Park and Union NJ. Apparently they have them on the South Shore line too.

          • Adirondacker12800

            Yes but sometimes that’s cheaper than making the station three or four tracks wide. The videos I’ve seen of Union and Roselle Park the freight is on the east side of the island platform so they may have decided that one gauntlet is enough.

          • Zmapper

            Gauntlet tracks don’t allow for oversize loads to pass easily; they just allow through trains to pass through at a higher speed.

          • Richard Mlynarik


            This has nothing to do with Transbay.

            I said nothing about the NEC,

            I explicitly said I was talking about greenfield construction: ie the entire US outside the NEC.

            Double deck intercity trains are desirable everywhere for network capacity maximization reasons. Most of the world is stuck, for historical reasons, with either the inability to run double-deckers at all or with non-optimal train designs (dictated by ~1m platforms, meaning triple-level trains and end doors, and please don’t bring up irrelevant RATP MI2N.)

            There’s no reason for California, as a particular example of the rest of the US, to not go for a no-compromise solution, and there’s no reason for greenfield construction to go with platforms up above 800mm.

          • Max Wyss

            To Richard: Lower level entrance in bi-level cars is not really the best for passenger flow and distribution to the two levels. Studies/tests proved that for two-door bi-levels, triple-wide doors over the bogies provide the best passenger flow, even if there are steps involved. Of course, there is now the darned accessibility legislation requiring level boarding at least somewhere in the train. However, passenger flow is more a concern for commuter services, and less so for long distance. Of course two double-wide doors per carbody (assumed that they are not short ones à la Talgo) are still a very good idea.

          • Adirondacker12800

            I explicitly said I was talking about greenfield construction: ie the entire US outside the NEC.

            Where’s the break between NEC platforms and UIC platforms happen? Pittsburgh, Buffalo and Washington DC?
            That means passengers between Cleveland and Rochester, Syracuse or Toronto have to change trains in Buffalo. Cleveland is close enough to Philadelphia that Cleveland-Philadelphia is concievable. They have to change trains in Pittsburgh? Run a branch from the Pittsburgh-Harrisburg line to Hagerstown MD and it makes Cleveland-Baltimore and Cleveland-Washington DC concievable and Pittsburgh-Baltimore, DC and maybe Richmond reasonable. Make Cleveland-Harriburg fast enough it makes Cleveland-NY an option. You can leverage that to get Columbus-Philadelphia and Cincinatti-Pittburgh into that. Make things between Harrisburg and suburban Chicago fast enough and you can make Columbus-New York, Columbus-Baltimore, Columbus-DC concievable. The branch to Cincinatti is short. You then have Cincinatti-Cleveland, Cincinatti-Pittsburg, Cinncinatti-Philadelphia and for people who don’t want to deal with DC or NYC Airports, DC and NYC.

            Move the break to Cleveland and people who want to go from Toledo to Buffalo have to change trains in Cleveland as do people who want to go from Columbus to Buffalo. Or Columbus to Toronto or Rochester.

            Chicago to Indianapolis makes sense. St Louis to Indianapolis, meh. But connect Indianapolis to Dayton you get Dayton-Chicago, Columbus-Chicago and Cincinatti-Chicago. Connect St. Louis to Indianapolis with a connection to Dayton and you get St Louis-Dayton, Columbus and probably Cincinatti. Throw in Indianapolis-Cleveland, Indianapolis-Pittsburgh and Indianapolis-Milwaukee. Put the break in Milwaukee and people going between Madison and Chicago or Chicago and Minneapolis have to change trains in Milwaukee.

            People in Richmond want to go places along the NEC so having the break at Washington DC means they have to change trains in DC. So move it to Richmond. The the people who want to go from Charlotte to DC have to change trains in Richmond. Move it to Charlotte and people who want to go from Atlanta to Raleigh have to change trains in Charlotte. Move it to Atlanta and the people who want to go from Charlotte to Birmingham have to change trains in Atlanta.

            The Canadians are going to get around to Toronto-Montreal someday. Which then makes Quebec City-Montreal and Toronto-Windsor reasonable extensions. Toronto-Detroit is cheap-ish which gets you Toronto-Toledo. It’s real easy to get Toronto-Chicago then. And the people in Toronto and Montreal are going to want to go to other places in the US. Move Customs and Immigration onto the train and with a 4 hour trip between Montreal and New York Montreal-Philadephia looks good. Montreal-DC if they get NY-DC under 2 hours. Sub four hours NY-Montreal is possible, the study NY commissioned assumed Albany-NY would be 1:45 and they’d be running 150 MPH trains.

            Play this game long enough and you end up in Miami, Minneapolis, San Antonio, or Portland Maine, though the people in Maine would like it to be Augusta or even Bangor.

            please don’t bring up irrelevant RATP MI2N.

            The platforms in Philadelphia, New York and Washington DC are relevant to someone in Cleveland, as are the platforms in Chicago. The people from Indianapolis get a less than excruiciatingly optimal ride on their way to St. Louis because of 48 inch platforms. It’s balanced by not having to rip out thousands of commuter platforms and one seat rides from St Louis to Columbus and Indianapolis to Pittsburgh.

          • Gauephat

            Adirondacker, there’s a whole bunch of different platform heights in Europe. They manage just fine. Many of the busiest train lines in the United States and Canada have different platform heights along their lengths. Most of Amtrak and VIA’s rolling stock has a 1220 mm floor, with steps down to 200 mm. It would just as simple to build a high-speed trainset with a 1220 mm floor, and steps down to a 560 mm height.

          • Alon Levy

            They manage because the regulators understand that there’s so much legacy compatibility going on that demanding full accessibility now is not reasonable. But putting in steps and inaccessibility on a greenfield rail network is pointless.

            By analogy, large legacy rapid transit systems are often wheelchair-inaccessible: see maps for New York and London. But new builds have to be accessible, even in New York and London: the newer London Underground stations as well as the extensions under construction in New York are accessible, and newer systems like Hong Kong and Singapore (or systems that bothered to do station rebuilds, like Boston) are 100% accessible or close to it.

          • Adirondacker12800

            It would just as simple to build a high-speed trainset with a 1220 mm floor, and steps down to a 560 mm height.

            Stairs aren’t ADA ( Americans with Disabilities Act) compliant. If you are building a new station or rehabilitating an old one, passengers have to be able to roll their wheelchair on and off the train.

          • snogglethorpe

            So, can you give an example of what you would consider an “optimal” train design using low-floor bi-levels?

            [It surely can’t be the TGV duplex, as that seems to offer no great capacity advantage over single-level designs on other railways, and seems to have pretty much exactly the same “issues” you ascribe to high-platform bi-levels (lots of wasted space for stairwells, very low door capacity). As far as I can tell, the only reason they went with a duplex design at all is that their single-level design is greatly constrained for historical compatibility reasons—and obviously that’s not an issue for the “green field” cases you’re talking about.]

          • Richard Mlynarik

            [I hope this thread correctly; WordPress isn’t helpful]

            So, can you give an example of what you would consider an “optimal” train design using low-floor bi-levels?

            I think the configuration with two wide low-level entry doors at the “quarter points” of each ~25m car together with high-level gangways between cars has a great number of advantages for inter-city travel.
            Unpowered Amtrak California passenger cars and unpowered Swiss IC2000 cars have this overall shape. (Yes, something sort of OK in California!)
            Bombardier’s new EMU design for SBB (contract awarded last year, first deliveries in 2013) is extremely promising. (It’s what made me a believer, in fact. Check out the “Typenskizze” train design sketch links towards the end of that article.)

            * Doors are evenly spaced along the train length.
            * Good total door width for acceptable dwell at major stops. (Unlike TGV-Duplex. Academic at wretchedly low US passenger flows and with wretchedly slow US schedules, but one can dream that this might be an issue!)
            * Inter-car intra-train circulation doesn’t interfere with entry/exit.
            * Extremely accessible equipment bays (traction, HVAC, etc) just above ground level at car ends. (“Tri-level” designs require roof mounting, which, beyond accessibility, is still problematic for heavy transformers and inverters.)
            * Adequate technical equipment space for EMU equipment. Locomotives (TGV-Duplex) not required.

            I’m very impressed by what can be done within this configuration.

            A tangential note is that the floor heights of the new SBB intercity trains are ~440mm and that the lower level is a step down from the 550mm standard platform in order to squeeze the two floors in. With larger US loading gauge (and with greenfield CHSR in particular), there’s no reason not to increase overall vehicle height enough to put the lower floor at ~550-650mm.

            Then there’s the Talgo-22 vapourware (interesting design, but didn’t win contracts, and died) with connections at both upper and lower levels and the usual Talgo articulation, lack of axles, short car lengths, and compact non-distributed traction in the end cars.) It would have been interesting to have seen that developed.

          • Nathanael

            Richard, even that design is still a prima facie ADA violation for trains with bistro cars, cafe cars, dining cars, lounge cars, etc.

            And they still have those in Europe, and they’re going to still have those in the US, and they’re going to be stopping at the same platforms as all the other trains.

            It’s a prima facie violation because passengers in wheelchairs can’t roll between cars, but everyone else can. In contrast, you can make single-level trainsets completely roll-through.

            It’s not clear that bilevels are really such a great idea, unless you can do them with the bottom level above the bogies, or have width between separate-axled wheels to roll through (which I believe is Talgo’s scheme).

          • Max Wyss

            @Nathaniel and @Richard:
            > Richard, even that design is still a prima facie
            > ADA violation for trains with bistro cars, cafe cars,
            > dining cars, lounge cars, etc.

            Interesting… and then, there are some Swiss Disabilities organitations dragging the SBB to court because their accessible toilet and general purpose space for wheelchairs is in the restaurant car, where the lower level is served, of the new TWINDEXX EMUs. This may cause a more serious delay than they already have.

    • anonymouse

      Salt Lake City went with Bombardier bilevels for Frontrunner, matching the standard used pretty much everywhere south/west of the DC-Chicago line. Denver is the exception you’re looking for: they went with NEC standards for their upcoming electric commuter rail system, but that’s an operation that’s not going to be connected to the mainline network except at Union Station, and will be generally much more metro-like in nature, which makes the high floor single level approach that much more appropriate.

      • Zmapper

        Salt Lake City even has level boarding with their Bombardier sets, as this video shows:

        The biggest reason for the second crew member is to deploy the manual wheelchair ramp in ‘traditional’ Bombardier cars. When wheelchairs can simply roll right over the gap without a device to bridge it, the need for the second crew member goes away. Sure, it may be nice to have a second crew member to assist the passengers, but it isn’t a necessity if you solve the wheelchair problem.

        IIRC, they were going to only have a train engineer with no conductor, but the FRA said no.

      • Nathanael

        Denver is far enough away from everyone else with passenger rail — look at the distance from there to Provo — that its only connection to the rest of the nation’s passenger railways will be at a single point, Union Station, for the foreseeable future — and it can keep separate platforms for that purpose. It therefore has the luxury of setting its own standards.

  5. Pingback: Today’s Headlines | Streetsblog Capitol Hill
  6. Zmapper

    I mapped out all existing platforms used by Amtrak outside the NEC, and what surprised me is there aren’t as many high-level platforms as I thought, and no high-level platforms west of I-81 at all (other than regional rail in Chicago).

    Intercity trains from Boston could use a separate 21″ platform with two tracks, with no further modification needed west of there except at Worcester. Convert two tracks and a platform to 21″ in Philadelphia and they are set. Washington already has low platforms, so raise those platforms from 8″ to 21″.

    New York isn’t quite as easy as the others, though could still be done. Penn Station is overcrowded and should focus on regional and NEC trains only. Long distance trains could be redirected to Grand Central Terminal for trains headed north, with plenty of extra capacity, and to Hoboken for trains headed west.

    The Empire Corridor between NYC and Albany should remain at 48″ for a few reasons. Because it shares tracks with Metro-North for most of the way, 48″ allows for platform compatibility. Furthermore, 48″ allows the Empire Corridor to continue to access Penn Station without setting aside a separate platform.

    The Keystone Corridor could go both ways, but if a 30 minute layover is already required in Philadelphia, you might as well have them change trains there. If 48″ is chosen, then long distance trains from Chicago could run non-stop between Harrisburg and Philadelphia, with passengers transferring to the Keystone Corridor for local service.

    Outside the NEC, 48″ platforms aren’t as ingrained as some would like to think.

    • Adirondacker12800

      Instead of any platform for any train in Springfield MA. you have two sets. Someone who wants to go from Hartford to anyplace other than Springfield has to change trains in Springfield. Anyone in Boston who wants to go to Hartford needs to change trains. Someone in Worcester who wants to go to New Haven or New York has to change trains. If you want to go from Hartford to Syracuse you have to change trains in Springfield and then change trains again in Albany. The Vermonters are going have NEC platforms. Change trains in Springfield to get to Boston or Worcester? Then I guess someone in Hartford can get to Amherst without changing trains but not to Pittsfield.

      Instead of laying over in 30th Street the train from Harrisburg could do like the expresses between Harrisburg ( the trains coming in from Pittsburgh, Chicago, Cleveland and St Louis ) and New York did in days of yore and use Philadelphia’s New York Subway. There’s a direct connection between the Main Line and the New York Branch west of Zoo interlocking. Cut a half hour off the fastest trains and you are up to an average speed of 80 miles an hour….

    • Alon Levy

      The Keystone trains have variable dwell times in Philadelphia, ranging from 10 to 36 minutes (on weekdays, one train dwells for 36 minutes, and the rest for 10-15). The limit is most likely set by brake test regulations; 4 minutes is technically achievable given better operating rules, e.g. in Germany. The Pennsylvanian dwells half an hour, because of the engine change. That said, SEPTA should really be looking at implementing systemwide level boarding, and because of Market East, legacy high-floor EMUs that are good enough despite their weight, and total lack of in-vehicle capacity constraints, this means high-floor, single-level vehicles. So for the same reasons Empire should be high-floor, even Keystone trains that do not run through to the NEC should be high-floor.

      The issue here is that non-NEC traffic connecting to Northeastern cities is really one of three classes: Empire, Keystone, and long-distance. The first two need to be high-platform, for compatibility with part of the route they run on. The last one is a fairly small component of traffic anyway, and would be the last to be modernized. The train that’s most ready for modernization, the Carolinian, is also the one where there’s the most nontrivial chance of through-HSR to the NEC in the next 30 years.

      • Adirondacker12800

        You wanna run Talgos between Cleveland and Cincinnati, go right ahead and repave the patch of asphalt by the side of the tracks that’s been there since 1930.

        The issue is what happens when Buffalo-Albany is two hours and Pittsburgh-Philadelphia is two hours. And Washington-NY is two hours. That would make Pittsburgh-NY three hours. At the same time Pittsburgh-NY becomes three hours Pittsburgh-Chicago it likely to become three or four hours. That puts Cleveland-Philadelphia in the three hour range. But not if you have to change trains in Pittsburgh. So you extend the NY-Pittsburgh express to Cleveland to capture the Cleveland-Philadelphia market you screw over the Toledo-Pittsburgh and the Detroit-Pittsburgh market because they have to change trains in Cleveland. I think the Midwest HSR booster proposal to make Chicago-Cleveland two hours is a bit optimistic, it’s an average speed of 170 MPH, but three hours is conceivable. Which makes Buffalo-Toledo reasonable. Keep playing with city pairs and when you start in Boston or Washington DC you are eventually in Minneapolis or Kansas City.

        Atlanta-Washington DC will never be popular as HSR unless jet fuel gets very very expensive or airports get very very annoying. But Atlanta-Charlotte makes sense. As does Atlanta-Raleigh. Charlotte-Richmond makes sense and Richmond-Baltimore or Philadelphia. Play the same game with city pairs and Atlanta has 48 inch platforms.

        As for the engine changes in Philadelphia and Washington DC. slap an ALP45 on them until it’s route is electrified. The engine change in Philadelphia probably has more to with Amtrak’s equipment inventory than it being a good idea to change engines in Philadelphia… and ALP45 means the Pennsylvanian can go 125 between Philadelphia and Harrisburg…

        • Jiim

          Actual 220mph HSR won’t use existing stations anyway. The Amtrak Vision assumed new stations in Baltimore, Wilmington, Philadelphia, New York and Boston (the last two collocates with existIng stations). The SNCF expressions of interest assumed new stations on new greenfield tracks. So existing platform heights are irrelevant.

      • Nathanael

        Current plans are for Raleigh, NC to get high platforms. So much for the Carolinian….

        People should face it: anything connecting to New York City is gonna be high platform sooner or later.

    • Jiim

      Even inside the NEC, 48″ isn’t as ingrained as people think. Stations that are served by only a single passenger operator can have their platforms reconstructed to a new height if and when that operator buys new rolling stock for the route that includes that station. LIRR stays at 48″ for ever and it doesn’t bother anyone. It’s the stations that are shared by two or more operators that there’s a problem. If one operator shifts to a new floor height before the other, then there’s a period where the station needs to accommodate two platform heights. But there aren’t many such stations. MARC shares with Amtrak Washington Union, New Carrollton, BWI, Baltimore, Edgewater and Aberdeen. Edgewater and Aberdeen could be dropped, BWI is planned to be reconstructed anyway, Washington and Baltimore both have existing 8″ platforms. Apart from 30th St, SEPTA shares two stations on the Wilmington line and one on the Trenton line with Amtrak (and NJT). It’s only Keystone East where there’s many shared stations. Even NJT only shares Trenton, Metropark, Newark Airport, Newark and New York with Amtrak. (Though Secaucus would create a problem for NJT rolling out new floor heights on one route at a time.) Metro-North shares just two stations on the Hudson Line and four on the New Haven Line with Amtrak. And so on.

      If (and it’s a big if) we wanted to adopt a single platform height standard across the US, it wouldn’t be a major job to implement it.

      • Adirondacker12800

        What’s wrong with adopting 48″ high platforms as the standard? You save tens of billions of dollars reconfiguring the stations along the NEC and it’s branches.

        Amtrak’s vision plan is more like Amtrak’s hallucination. Spend billions of dollars building a high speed bypass in Philadelphia so that every train can slow down and come to a stop? It takes a long time and distance to slow down a high speed train. North Philadelphia is 5.5 miles from Suburban. Where does the high speed train start to slow down if it’s stopping at 30th Street? Wilmington’s station is perfectly good for trains that stop in Wilmington. The ones that don’t stop in Wilmington can bypass it on the available ROW to the east. What does building a few billion dollars worth of track in Baltimore buy you? And a few billion dollars excavating a station cavern to build a station?

        • Zmapper

          48″ platforms should be limited as much as possible, in order to allow for oversize freight trains to pass. Those “wide load” trucks sure do drive slow and block traffic from passing easily. The goal should be to get as many trucks of the road as possible, because of the intrinsic advantage of freight rail.

          While 48″ works okay for small stuff that fits within clearance limits, large loads cannot pass them. 21″ allows for large loads to pass, because the bottom of the load is almost always above 2′.

        • Adirondacker12800

          There’s not going to be a whole lot of freight on the 200 MPH routes. There’s not going to a whole lot of freight on the 126 mph or higher routes either.

          • Zmapper

            It is not the high speed routes I am worried about, it is the branch and freight main lines. High Speed Rail can use 48″ platforms all they want, because little freight will be on the tracks. As I stated before, oversize freight trains can’t pass 48″ platforms, meaning there is no way the freight companies will agree to 48″ platforms. 21″ platforms are more reasonable, as the freight companies won’t be impaired much.

          • Adirondacker12800

            and there’s not going to be a whole lot of passengers either.

          • orulz

            A single gauntlet track past the platform on each station that (1) carries oversized loads and (2) has high platforms would be sufficient. This is not free but not that expensive – probably less expensive than rebuilding platforms and all the vertical circulation that go with them.

      • ant6n

        Btw, in Germany, for example, they use 550/760mm for regional/long distance travel, and 960mm for some S-Bahn services (for example berlin). Since those are separated anyways, the difference doesn’t matter. You could have an analoguous situation for example in New York, separating long distance and commuter rail.

        Another possible compromise solution would be to have split platforms, the first half at low height, the remanining that high height. Then at least one car whould be accessible via level boarding, making it ADA compliant.

        • Jiim

          ADA doesn’t necessarily require level boarding. It requires accessibility. Ramps or lifts are acceptable. The problem is that 48″ to 8″ is hard to bridge with a ramp or lift.

          • Alon Levy

            Ramps or lifts are unacceptable for any good service metric, except for very long-distance intercity services, for which dwell times are not important.

          • Adirondacker12800

            Alon, they don’t understand that when a 12 car standing room only Trenton Express opens it’s doors in Penn Station that 2,000 people get off. And that have 2,000 people getting off in Penn Station they have to get on somewhere. And that other lines have the same sort of peak travel.

          • Jiim

            Some routes require level boarding because dwell times are critical. The commuter services serving New York City are examples. Other routes don’t. The FEC route mentioned in the OP is one. It’s going to run 240 miles between Miami and Orlando with just two intermediate stops. Dwell times at those stops are not going to be performance critical.

            The argument for a single standard floor height is that it creates a single more competitive market for rolling stock. The argument against is that it’s too difficult to change. If one accepts that there should be a single standard floor height, then it can’t be 48″. The oversize freight issue rules 48″ out. Any particular station can have 48″ platforms. One can always work around single cases. The NEC can have 48″ platforms. Freights don’t like to use the NEC, they’ve alternate routes for most cases, and where they have to use the NEC (NS reaching the Port of Wilmington or the Delmarva Secondary) the stations are on at least triple-tracked segments. But you can’t have 48″ platforms everywhere.

            I am sympathetic to the argument that transitioning from 48″ to something lower is simply too hard. Reducing the height of a platform takes it out of service for a period. In some cases it may be that too large a rolling stock buy would have to be made to enable the transition. But I don’t believe anyone’s quantified how hard the transition might be.

          • ant6n

            “The argument for a single standard floor height is that it creates a single more competitive market for rolling stock.”
            And of couse if it it’s some UIC floor height, then with the hinted FRA changes, that market could utilize ‘off-the-shelf’ equipment from many vendors.

          • Adirondacker12800

            Almost a million people get on and off the commuter railroads that serve Manhattan – on weekdays. Ya need pretty big fleet to achieve that…. One of the world’s most commonly manufactured railroad cars is the MTA’s M7/M8. Pretty much off the shelf. No reason why with some minor tweaks, like not installing third rail shoes, M8s couldn’t be used by SEPTA. SEPTA got a better price from Rotem for the Silverliners.so they went with Rotem. No reason why Metra Electric couldn’t use them with some minor tweaks. Or with some minor tweaks MARC or the MBTA couldn’t use them. Or AMT. I think the only tweak the MBTA might need is “no third rail shoes” Though once Amtrak approves them for east-of-New Haven, there’s no reason why they couldn’t go all the way to South Station as they are.
            Which narrow stubby UIC standard do we promulgate? There’s many. A standard North American passenger car is 10’6″ over the thresholds and 85′ long. They can go anywhere and do, which is why Metrolink in Los Angeles leased cars from Frontrunner in Utah which were used NJTransit Comets. Metra gave Frontrunner bilevels, which they decided were too far gone to be worth rebuilding, They kept them to salvage spare parts for the Comets.

    • Nathanael

      Grand Central is high-platform, too, Zmapper. Even if it weren’t, you can’t direct anything from the south into it unless Alt G is built.

      Every train entering New York City has to be able to platform at high platforms.

      • Alon Levy

        Grand Central has an enormous surplus of tracks, is the point. That’s why Zmapper and Jim are focusing on the smaller stations, like Back Bay and Wilmington, which do not have any extra tracks and can’t segregate low and high platforms. (Though, South Station has to be added to the list because the MBTA should run high-platform operation, for other reasons.)

        • Nathanael

          Grand Central also has third rail (not that it *should*, but that’s another matter). Third rail and low platforms… are not mixed anywhere I’ve ever been. Even if it is underrunning third rail.

          • Max Wyss

            Probably the lowest platforms with third rail can be found between St. Gervais and Martigny, where they are about 35 cm above rails. The Swiss section (Martigny – Vallorcine) is now mainly overhead, but the French section is all third rail. Of course, most of the line is very rural. Similar platform heights may also be found on the “Métro des Pyrénées” (Villefranche – La Tour de Carol). Both lines are “old” lines.

            I am not sure whether the third rail was laid through the stations on the Mont Cenis line (Chambery – Modane), but if so, the platforms would have been low-level. Third rail operation continued into the 1970s, but got gradually replaced with overhead catenary.

            Besides of subways, third rail installations are no longer built new (unless small extensions of an existing system).

          • Adirondacker12800

            Clearances in Grand Central and the Park Avenue tunnel may not be big enough for high voltage catenary. While you wouldn’t go out and design it that way today, it’s possible to run trains on high voltage catenary in the high speed territory and third rail in Grand Central, Metro North and it predecessors have been doing it for a century.

          • Walter

            Suburban stations on the New York Central (and I assume on the LIRR) were low-level with third rail until the mass platform-raising of the late 60s to prepare for the M1s. The third rail would always be opposite the platform (it’s still this way today).

            There still exist grade crossings in close proximity to the third rail, especially on Long Island and in a few spots on the Upper Harlem Line, but I’ve never heard of anyone getting zapped.

          • Adirondacker12800

            There was a grade crossing on the Carnarsie line until the 70s or 80s. The L in Chicago still has grade crossings. Extending the electrification to Wassaic was an extension. It would be very difficult to get approval for a new system that uses third rail and that has grade crossings.

  7. Eric

    Alon: Do you think transit blogs played a role in the FRA decision, by creating public discussion of issues that were previously visible only to small groups of in-house engineers?

    • anonymouse

      It’s possible, but it’s probably not a very big role. I think it had more to do with the increasing number of transit authorities asking for waivers, as well as the likes of Metro North realizing, once they bought the M-7s and M-8s, that the current regulations were rather suboptimal for their sort of operation. I suspect Caltrain’s waiver and Metrolink’s experiments with CEM played a role as well.

  8. Richard Mlynarik

    I’ve not actually made the effort (it would be too much depressing work, with too many idiotic US unit conversions and too much language translation, amd because I can just guess the outcome) to verify that this really is any sort of “good news”, or just some inconsequential pig lipstickery.

    The people who know, of course, are the non-Alstom non-Bombardier rolling stock vendors. My guess is that a couple Alstom USA-authored changes to a just a couple of the thousands of FRA/PRESS regulations and standards will have minimal effect on competitiveness of procurement, weight and energy efficiency of equipment, or capital or operating costs.

    I’m not exactly expecting Kawasaki, Stadler, Talgo, CAF, Mitsubishi, Siemens, Hitachi … products to start popping out of existing product production lines any decade soon, in other words.

    But I’m just going on without data, other than very suggestive and consitent historical data. Somebody, somebody, do the work, do the investigations and report back! I so want to be pleasantly surprised, just for once.

    • Adirondacker12800

      Kawasaki stuff pops out of US production lines at a fairly steady rate and has been for a long time. Siemens is busy upgrading their California plant so they can manufacture electric locomotives. Talgo was going to build a plant in Wisconsin but then changed their mind and moved it to Illinois. I suspect Rotem is hoping that everybody just loves them there Silverliner Vs and starts ordering some up, once they are done with the SEPTA order. They would be just the thing for MARC, MBTA or NJTransit… Rotem, the company that built the stuff for Metrolink that incorporates crash energy management…

    • Alon Levy

      I wanted to say the same, but the FRA has consulted a lot of different vendors on these changes – not just Alstom, Bombardier, and Kawasaki, but also the smaller ones.

  9. Pingback: The Magic Triangle: Infrastructure-Timetable-Rolling Stock | Pedestrian Observations
  10. Pingback: FRA Reform is Here! | Pedestrian Observations

Leave a Reply to Nathanael Cancel reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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.