Can HSR Connect Hartford and Providence?

If new high-speed rail construction has to largely follow Interstate corridors, then a new line from New Haven to Boston can serve either Providence on I-95 or Hartford on I-91/84/90, but not both. However, there’s still the possibility of building a completely greenfield alignment between Hartford and Providence; the FRA is investigating this as option 13 of NEC Future and Amtrak is proposing this in the latest update of its Vision. Since the terrain is hillier than on the coast, it requires some investigation as to whether it’s possible to connect Hartford and Providence without excessive tunneling. The answer turns out to be yes, but only at the cost of slowdowns both north and south of Hartford that impose real costs relative to following I-95: construction is likely to be more expensive and travel time including a Hartford stop is 9 minutes longer.

I believe the alignments depicted in this map are near-optimal for New Haven-Providence via I-95 and via Hartford. The New Haven-Hartford alignment is similar to that of Penn Design with two major differences: Penn Design diverges to cut off some curves near Hartford, but to guarantee sufficient curve radius it has to slice a significant chunk of downtown New Britain; and Penn Design also straightens the route in New Haven with a tunnel, which is unnecessary as the time savings do not justify the expense. Amtrak prefers getting to Hartford from Danbury, but to get there from New Rochelle requires long suburban tunnels, which my alignment avoids. I have not seen a detailed Hartford-Providence alignment, and I drew a line based on Google Earth elevation with an eye toward avoiding tunneling, which means there may be some further optimization, for example a rigorous cost-benefit analysis of viaducts versus curve avoidance.

The Hartford-Providence greenfield route has no tunnels except in Providence itself, where the line tunnels under Olneyville for about 2 kilometers. In Connecticut the route has many viaducts, but does not need to tunnel through the inland hills. Rather than giving detailed cost estimates, which are possible but not with sufficient reliability or precision, I am going to qualitatively describe construction challenges for each route and then the differences in travel time, which favor not serving Hartford. The final decision should boil down to the question, what cost is it acceptable to impose on New York-Boston travelers to allow for service to Hartford?


The I-95 route is zero-tunnel. The Hartford route has no tunnels in Connecticut, but requires a tunnel of 1.5-2 km in Providence. There exists an old railroad alignment going around the river and connecting Providence to the west without a tunnel, but the right of way was given away and to restore it would require some urban building demolitions as well as configuring a flying junction under Route 6 while also slowing down trains further.

River crossings

The I-95 route has significant challenges in river crossings, since it is close to the coast. Three difficult crossings are needed, of the Quinnipiac, the Connecticut, and the Thames. The Connecticut only requires a span parallel to I-95. The Quinnipiac requires a new span parallel to US 1 and I-95 and a new approach from Union Station; there is space for this approach, and the curve radius can be kept to at least 500 meters, but it requires work on active track. The Quinnipiac span can be avoided by using the existing route around the bay, which crosses the river at a much narrower point, but this adds several slow kilometers to the route. Recent construction costs for parallel bridges are $125 million for the four-lane US 1 bridge and $554 million for a signature ten-lane I-95 span; I believe the lower cost is more indicative of the infrastructure required for a two-track rail bridge.

The Thames is the hardest, since the route of I-95 and the terrain make it hard to cross anywhere except downtown New London, a constrained urban location. There is just enough space for a station between the decks, and the alignment may impose further constraints on curve radius. There is more space north of both decks, or alternatively Connecticut could build a third I-95 deck and repurpose one of the existing decks for rail.

The Hartford route has one significant water challenge: crossing the Connecticut in downtown Hartford. There is an existing bridge, but it is single-track and would require a completely new span to be used by high-speed rail. It is also used by freight, but only by a short branch line that could be bought out.

The Hartford route also needs to cross the Scituate Reservoir, adding about 3 kilometers of viaduct. However, there is a choice of where exactly to cross it and not much development on its banks, making construction easier than on I-95 or across the Connecticut in Hartford.

Terrain and viaducts

I-95 is substantially flatter than the inland route. Only two short segments require significant overland viaducts and earthworks: the transition in southern Rhode Island from I-95 to the Shore Line, and the curve west of New London cutting off curvier parts of the Interstate. The transition is in total 16 km long but only about the western 10 km of it are difficult (of which about half require viaducts and half can be done cut-and-fill), and west of New London there are 6 difficult km requiring a viaduct north of the Niantic.

In contrast, the inland route needs to be on viaduct for a significant portion of the Hartford-Providence section. Of particular note is the Quinebaug River valley, about 13 kilometers of route of which most requires extensive grading and viaducts, as well as some takings in the built-up areas of the towns of Brooklyn and Killingly. The Willimantic River-Mansfield Hollow Lake-Natchaug River complex adds another 16 kilometers, some hard and some less hard; the Willimantic itself is in a deep valley requiring a tall viaduct of about 3 km, and the total viaduct length required appears to be about 8 km. The following 12 km, on the crest heading to I-84, require some earthworks, but probably no significant viaducts.

Urban construction challenges

I-95 has an existing route into Providence. Some curve modifications from East Greenwich northward are helpful for keeping speeds up, but the grade-separated route already exists. The main challenge is fitting regional trains if Rhode Island desires to run them: the right-of-way has room for four tracks but only if track centers are narrowed so much as to preclude tilting, reducing cant deficiency to about 125 mm. At the New Haven end, the main challenge is crossing the Quinnipiac, but once the tracks are east of the harbor, suburban development intensity drops rapidly, requiring only occasional grade separations with roads crossing I-95. Conversely, if intercity trains are all routed through Hartford then no new construction is required for any Rhode Island regional rail.

The major problem then is New London. The entire complex of crossing the city and the Thames is the biggest difficulty in the route, as outlined above in the water crossing section. In addition to geometric difficulties, there are also noise abatement issues, since the track geometry still allows very high speeds (the curve drawn above just west of New London looks like it can be eased to about 3 km, allowing 310 km/h). This is what favors putting the tracks between the two I-95 bridge decks instead of to the north.

The inland route has far greater difficulties. First, it needs to carve a partially new route into Providence, hence the Olneyville tunnel; however, it also leaves the Providence built-up area much faster, within about 6 kilometers vs. 24 for the Shore Line. In New Haven and Hartford it can for the most part transition between legacy rail routes and expressway corridors, but a substantial portion of the route is in the suburbs of those two cities, which requires more grade separations and makes curve modification harder. There are also noise abatement issues, though Shinkansen trains skip some urban stations at 300 km/h, so those issues are more about cost than about speed limits.

There are several alignment choices north of New Haven. The one I used on the map follows the Providence and Worcester’s Middletown Branch right-of-way and thence I-91, but it is equally feasible to take a more westerly route via the Amtrak line transitioning either to I-91 or Route 15; both options involve grade crossings and extensive suburban construction. In all cases, the trains are almost continuously in built-up area from New Haven until 19 kilometers east of Hartford. Grade separations have the full cost of urban or dense-suburban construction, and moreover, the transition to I-384 east of Hartford requires some additional takings.

Total new construction

This is the primary advantage of I-95, cost-wise: the track already exists from Kingston north and requires only minor facelifts. The New Haven-Kingston construction is just 124 km, whereas between the splits with the legacy Northeast Corridor in New Haven and Providence the Hartford route is 167 km.


With this in mind, nearly the entire I-95/Shore Line segment between East Greenwich and East Haven can be eased to a curve radius of 4 km. New London, where noise abatement prevents running at full speed anyway, can accommodate slightly lower radius, about 3 km on the western approach. At the New Haven end, the transition to the Quinnipiac bridge right next to the station has radius 500 meters, but the speed restriction is minor since it is so close to the station.

Hartford-Providence can also be eased to quite high curve radius. In Rhode Island, once out of the Providence built-up area, the tracks can maintain a 4-km standard, and until the transition to I-384, the worst radius is 3.1 km around Mansfield. However, from I-384 west, things become far worse: the transition to the east has a radius of 1.2 km and seems impossible to increase further, the transition to the west has a radius of at most about 1 km, and the curve west of the Connecticut bridge is 500 m and is slightly farther away from Hartford than the Quinnipiac bridge curve is from New Haven.

It is south of Hartford that things deteriorate. The worst curves on the legacy lines are in Meriden and can be bypassed, but there is a 1.3-km curve in New Britain, on an S with a 2.3-km curve just south in Kensington that makes it unfixable. At the New Haven end there’s a curve on the legacy line, bypassed on I-95 by the Quinnipiac bridge, with radius about 450 m about 2.5 km out of the station.

Overall travel time

The explicit assumptions on trains are aggressive, based as always on the need to keep speeds up in big cities and on the only partially fixable New York-New Haven segment. Trains accelerate like the N700-I (26.74 kW/t, more than any high-speed train that currently exists except the Talgo AVRIL), cant deficiency is 175 mm as on the E5/E6 and on the AVRIL, cant is 200 mm as on the Tokaido Shinkansen, and initial acceleration is 0.89 m/s^2 as on the N700-I. With these performance specs, the minimum curve radius required for a full speed of 360 km/h is 4 kilometers; the Tohoku Shinkansen has such radius and JR East intended to run trains on it at 360 km/h before deciding to reduce speeds to 320 for reasons that are not track geometry.

For simplicity of computation I’m going to ignore grades. Since the I-95 route is flat, with very few grades higher than about 1%, this is justifiable there; it’s a little less justifiable through Hartford because a few segments have 3% grades, but they are also quite limited.

Without any schedule padding, we can set the following speed zones for I-95, measuring from 0 km point in Providence and going southbound:

0-0.6 km: 90 km/h (curve around Providence Station)
0.6-4.5 km: 120 km/h (two 450-m curves)
4.5-7.5 km: 180 km/h (Mashapaug Pond curve is too close to 120 km/h to matter, curve into Cranston is about 1 km)
7.5-17 km: 250 km/h (no curves, trains can achieve 270 in between curves but this would only save 5 seconds)
17-22 km: 220 km/h (curves have radius about 1.4 km and the controlling curve at km-point 17 can be eased a bit)
22-92 km: 360 km/h (full speed to New London)
92-103 km: 310 km/h (speed restriction in New London and the curve north of the Niantic River)
103-162 km: 360 km/h (full speed to East Haven)
162-167 km: 250 km/h (curve around an East Haven hill, though trains can barely accelerate fast enough for it to matter going eastbound)
167-168 km: 100 km/h (New Haven approach)

The time taken to transition between speed zones is the average of acceleration and deceleration time penalty. This gives a technical travel time of 33:40 for nonstop trains. If trains have a top speed of 300 km/h, this raises the technical travel time to 37:28.

Now, let us set speed zones for the Hartford route:

0-0.6 km: 90 km/h (curve around Providence Station)
0.6-4.5 km: 180 km/h (curve north of Hartford)
4.5-6.5 km: 200 km/h (curve into Johnston)
6.5-10 km: 240 km/h (curve west of I-295)
10-57 km: 360 km/h (full speed to the Hampton-Mansfield area)
57-86 km: 310 km/h (Hampton and Mansfield impose a 310 km/h restriction to km-point 67, and trains going eastbound can’t accelerate to 360 before they have to slow down again anyway)
86-88 km: 220 km/h (gentler curve in the transition to I-384)
88-101 km: 200 km/h (transition curve to I-384, further curves on I-384 making speedup between transition curves pointless)
101-103 km: 160 km/h (transition curve)
103-109 km: 200 km/h (minor opportunity to make up time, saves 20 seconds over 160 km/h)
109-110 km: 130 km/h (curve on eastern approach to bridge)
110-112 km: 110 km/h (curve on western approach)
Hartford Station: all trains stop since curves limit time savings from not stopping, as at New Haven and Providence
112-127 km: 250 km/h (New Britain curve, speed increase to 270 km/h in between is possible but saves only about 8 seconds)
127-153.5 km: 270 km/h (Kensington and Berlin curves preclude higher speed)
153.5-155 km: 210 km/h (S-curve precludes easy straightening, and significant speed boost requires significant residential takings)
155-169 km: 250 km/h (this requires straightening the kink around and north of the I-91 underpass, otherwise 210 km/h to km-point 162, 160 km/h to km-point 164, and 200 km/h farther south)
169-172 km: 120 km/h (New Haven approach, legacy line curve)

The travel time is 25:30 for nonstop trains from Providence to Hartford and 16:10 from Hartford to New Haven. With a minute of dwell time at Hartford, this is exactly 9 minutes longer than I-95.

Compatibility with other plans

Although I-95 requires less construction overall than Hartford and the construction difficulties are about comparable, Hartford is more compatible with other intercity rail plans for New England, which reduces the advantage of I-95. Under an I-95 option, it is still useful to serve Hartford (and Springfield), which means the Amtrak Shuttle line needs to be electrified, double-tracked, and partially curve-modified anyway. Under the Hartford option this is not required except to provide regional service to Wallingford and Meriden, so the bypassed parts of the legacy line could be built to lower standards.

That said, 60 km of 160-200 km/h electrified track is still a lot cheaper than 60 km of 250-270 km/h track, which means that this reduces the cost advantage of I-95 but does not eliminate it. Of course 60 km of 250-270 km/h track is cheaper than 60 km of 360 km/h track, but I-95 still involves much less overall greenfield track construction.Hartford is also more compatible with any plans Rhode Island might make for southward commuter rail service. The current plans are too low-ridership to bother accommodating, but future plans might involve higher service levels.

Conversely, I-95 is useful for Shore Line East service, since regional trains could use the Quinnipiac bridge as a shortcut. The tracks cross in East Haven and a track connection could be built; it is likely that there will always be enough capacity for 5 km of track-sharing between intercity and regional trains. I-95 is also useful for the New London connection in case anyone wants to build a New London-Norwich regional train serving Mohegan Sun on the way.


Neither route is particularly expensive by the standards of what both Penn Design and Amtrak think are appropriate budgets. At French construction costs, 124 km of high-speed track with no tunnels, few viaducts, and a mostly preexisting Interstate right-of-way should be about $2.5 billion. Likewise, the cost of 167 km with only 2 km of tunnel and a fair number of viaducts should be less than $4 billion, possibly down to $3.5 billion.

However, in case there’s only enough money for part of the route, construction has to be phased. The Hartford route has no track connections to usable passenger railroads between Hartford and Providence, so the only useful partial construction there is the entire Hartford-Providence segment at once plus electrification of New Haven-Hartford(-Springfield). The I-95 route comes sufficiently close to the legacy track in East Haven and Old Saybrook, giving three segments each of which can be built separately: across the Quinnipiac, from East Haven to Old Saybrook, and from Old Saybrook to Kingston.

Station-skipping decisions

Every possible train station on a route deserves an answer to two questions: what is the time advantage gained by skipping it?, and who is served by it?. Stations very close to urban terminals, for example Back Bay, have a very low stop penalty because of low approach speeds, but don’t add much service since people can just ride to the urban terminal. Suburban stations such as Route 128 and even Stamford given necessary track upgrades impose high enough a cost that they should also be skipped by express trains even if there’s a fair number of people who’d use them on the local trains.

Between New York and Boston, there are three stations where the answers to both questions favor express stops: New Haven, Hartford, Providence. With New Haven and Providence, the time cost of serving them is so low given urban curves that the only way to skip them at speed is to build new urban tunnels, which cost a lot of money relative to how much time they save. With Hartford, the situation is the same if all trains go via the inland route that serves it.

However, on some level, the time cost of serving Hartford is 9 minutes, compared with about 2 for Providence. But this is not really comparable, so we can’t just say “9 minutes is too much,” as it would be if a station on a running line imposed a 9-minute stop penalty. If we skip an intermediate station that imposes a time penalty of 4 minutes, the express trains gain 4 minutes but there are still local trains serving it. In contrast, if we go via I-95 we save 9 minutes but have no way of serving Hartford on local trains; trains can branch off north of New Haven and serve Hartford and Springfield at lower speed, but this only connects Hartford to New Haven and points south rather than to Providence or Boston. So we lose something more fundamental than stopping train frequency.

So it’s not enough to say that Hartford should be skipped because it saves the trains 9 minutes. That cost-benefit calculation depends on how important serving Hartford is to people. It’s up to the potential users of Northeast Corridor HSR and the politicians providing the funding to decide whether it’s worth it to connect Hartford with Providence and Boston.


  1. Ryan

    WordPress started eating your line breaks around half way through this piece, FYI.

    I’ll have a more detailed commentary for you later.

  2. Adirondacker12800

    No but that’s just after reading the title. Yaphank-New Haven-Hartford connects far more people and solves problems other than getting people from Providence to Hartford. Providence-Worcester gets the people in Providence a slightly faster trip to places other than Hartford and a slightly longer trip to Hartford. Providence-Hartford is an icky drive and people will put up with a longer trip to avoid the icky drive.

      • Adirondacker12800

        Why? Hartford to New Haven has between 7 and 9 trains a day last time I looked. Electrify it – it’s 60 miles from New Haven to Springfield and they can use the HSR track from New Haven to get south/west of New Haven and the HSR from Springfield to Boston to get to Boston. Providence is on the Shore LIne Bypass and they don’t care. For that matter if all you build is Boston to New Haven via Springfield to leverage Boston-Albany they have have halfway decent service on the existing Shore Line.

        • letsgola

          A Boston-NYC route that includes Hartford and Providence seems more practical to me than a route that depends on a Boston-Springfield HSR line, which would probably have to be a greenfield route too, right? MA politicians are frequently required to make overtures to the western part of the state to prove they know the commonwealth doesn’t end at the 495 freeway (example, here’s Matt Amorello having the turnpike study a monorail to Springfield), but the fact is there are just not that many people there. People in MA view Boston on par with NY, Phila, and DC. I don’t think anyone in MA cares one way or the other about an HSR connection to Hartford.

          CT and RI, on the other hand, have a lot to gain from a Hartford-Providence routing. CT because it puts Hartford on an HSR main to Boston and NYC, RI because it reduces the probability of having Boston-NYC service bypass the state. If you’re RI, you’re trying to increase your connectivity to Boston and NYC. I’m not sure what Worcester gets you if you’re Providence.

          • Adirondacker12800

            but the fact is there are just not that many people there.

            There are in Albany Syracuse Rochester and Buffalo. It’s 640 miles to Cleveland from Boston. Toronto, Montreal and Ottawa are closer. An HSR line to New York doesn’t get people in Boston to any of those places. And you only have to build as far as Albany to get to any of them.

          • Adirondacker12800

            but the fact is there are just not that many people there.… part II

            Legislators representing Hartford don’t vote on Massachusetts legislation. Legislators representing Springfield do.
            WIkipedia says metro Springfield is 600,000 people. Metro Hartford is 1.2 million.

  3. Ryan

    From the “more detailed commentary” department:

    Your Hartford – Providence greenfield isn’t necessarily bad, but you (and in fairness, apparently Amtrak and the FRA as well) have failed to consider the Washington Secondary / Coventry Greenway alignment, which gets you as far as Moosup, CT (actually a little less if you want room to maneuver around Moosup), is (to my knowledge) landbanked from top to bottom, provides a logical Phase 1 for phasing construction (Rhode Island Regional Rail out to West Warwick, which potentially serves Warwick and Cranston much better than the NEC while also providing a higher-speed approach into Providence, but at the expense of a connection to T.F. Green) and totally eliminates the engineering challenges of an Olneyville tunnel and Scituate Reservoir crossing in exchange for one problematic curve in West Warwick that would need to be bypassed somehow, as well as the problem of connecting back to your proposed inland route.

    It’s not hassle-free by any means – and it possibly comes at the direct expense of the existing trail – but I maintain that it’s the most equitable solution available to us and quite possibly a significant cost savings versus the Scituate/Olneyville alignment.

    • Joey

      You could avoid the curves in West Warwick by transitioning to I-95 roughly parallel to SR-118, then back again along I-295.

  4. Michael Noda

    I think a lot of the attraction of an inland greenfield HSR route is a realization that New London is just intractable; the geometry doesn’t allow for an alignment that includes a sane Thames crossing and a station on tangent rails in downtown New London. The one available alignment is the existing I-95, as you identified, but any serious work on revitalizing New London itself has to start with ripping out most or all of I-95 and its associated mess just north of downtown, to reconnect the USCG Academy and Connecticut College (both stranded on the far side of an I-95 spaghetti junction) with the urban core. Right now, the schools are as accessible to downtown as Naval Submarine Base New London, which is on the Groton side of the river; they should be a 30 minute walk from Union Station.

    The big win for Connecticut in a Hartford-Providence HSR alignment (apart from the obvious benefit to Hartford) has to be access to the University of Connecticut. Storrs has to be the least accessible college town in New England for the carless, and that is starting to be a recurring theme in the college decision discussions I overhear. A key tell is that the men’s and women’s basketball teams are both playing more of their games at the XL Arena in Hartford (steps away from Union Station) than in the Gampel Pavillion on campus, and that discussions of a new arena center on Hartford (current capacity 16K) and not Storrs (current capacity 10K). Think what you want about the American phenomenon of sports franchise tails wagging the higher education dog; Connecticut is very, very good at that particular game. CTDOT will happily carve out a nice ROW to move the station from 2.5 miles away to within walking distance of the UConn campus.

  5. Michael Noda

    Also, it’s heresy, but does a 186+ mph BOS-PVD-HFD HSR alignment, in combination with a 79mph EMU shuttle HFD-SPG, cut the travel time Boston-Springfield to be properly competitive with the MassPike? If so, under what assumptions? It’s not as good for western Massachusetts as a direct BOS-SPG HSR, but it saves construction costs over any total system configuration other than Empire Cross variants. It also folds Boston traffic into the main travel axis of the region, which is up and down the Conn River valley, which would also enable (more heresy approaching) the station to be moved off of CSX to the Conn River Line, which would put an end to all the time-wasting switching and backing up that the current Union Station requires to run through to points north.

  6. Adirondacker12800

    Invert the question. How many HSR lines is Boston going to have someday? The first one has to be Boston to New York. Where does it want it’s second one? Where does New England want it’s second one? Build Providence to New Haven via Hartford to get to New York it wants Boston to Albany. Build Providence to New Haven more or less directly and it wants Boston to Albany.

    The Vermont Agency of Transportation studied Boston to Montreal. Conventional rail isn’t fast enough to attract a lot of people. “Higher speed” rail doesn’t cost that much more and attracts enough people that it can make money. High Speed rail attracts more people but costs so much that it doesn’t make money….. They would have to spend a lot of money in New Hampshire with any option. And have to convince New Hampshire that railroads aren’t the work of the devil that will take away their Sarah Palin books. It’s 310 road miles from Boston to Montreal via the most direct route. It’s 390 via Albany.

    Montreal to New York… costs 5 billion dollars, if I remember correctly, according to the NYSDOT study on I-87 congestion. They were doing conservative estimates. 150 MPH diesel trains with Albany to New York improved enough for 1:45 gets you 4 hour travel time. With blurb that electrification and higher speeds between Albany and New York were outside of the scope of the study. When the Federal Government is thinking about funding this Vermont can flutter it’s eyelashes and point out that going from Whitehall New York to Montreal via Burlington avoids all those nasty issues in that lovely pristine park on the New York side of Lake Champlain. And that there are more people in metro Burlington than there are in Plattsburgh. And that going through Vermont doesn’t do anything dastardly to that lovely park on the New York side. And mention that you don’t have to go through a park if you go through Vermont.

    By the time Montreal to New York is built, on the eastern side of Lake Champlain because the Vermonters mentioned to the people in Western New and New York CIty that building in Vermont means their park is preserved, NYC to Philadelphia is much faster trip. And so is NYC to Albany. That gets you Montreal to Philadelphia. With customs and immigration on the train it’s faster than flying. . With customs and immigration on the train it’s attractive to people in Baltimore and Washington. Build Boston to Albany and Bostonians can get to Montreal. Vermonters who want to get to Boston can get on the Boston-Montreal train like they can do to get to New York. If they want to get to Harford they can change in Springfield or get on the once an hour train that goes to New York via Hartford because that train gets the Hartford-Montreal, New Haven-Montreal and Stamford-Montreal traffic.

    Do that kind of thing with Maine. Or Providence for that matter. They can get on the once an hour Providence-Buffalo train that uses the MBTA-RI route between Worcester and Providence to get to Albany and change to a train to Montreal… If they are doing customs and immigration on the train they can change in Albany for the train to Toronto too. Or change trains in New Haven.

  7. Adirondacker12800

    Invert the answer to the question, which two HSR lines does New England want.?
    We can build Providence to Hartford across the rolling hills of Connecticut or we can build across the rolling hills of Massachesetts to get from Framingham to Springfield. Hartford to Albany is 120-ish. 90 miles of it is in Springfield to Albany.
    Build 100-ish miles of HSR from Providence to Springfield via Hartford or build 60 miles of HSR-ish from New Haven to Springfield and 40 miles of HSR-ish track from Providence to Worcester.

    • Ryan

      Who cares?

      No, I’m being absolutely serious right now. Who cares?

      Montreal, and Toronto, and Buffalo, and Vermont are all so far outside of the scope of the Boston – New York line that they might as well be located in Greece. Trying to somehow tie the hypothetical traffic between either city and mid-state Vermont is already a huge failure in logic – as is the assertion that there’s any amount of theoretical traffic – even international traffic (and, news flash, international crossings are going to be just as much of a hassle at Rouses Point than they are in CYUL or KJFK, sorry to break it to you) – that could possibly have enough of an impact on mainline travel between New York and Boston to cause the route to be changed to passing through Albany.

      Not that it matters because New England is going to have two HSR lines. The first is Boston – Providence – Hartford – New York City, and the second is Boston – Springfield – Albany – Montreal. New York, which is not part of New England, is going to have one more HSR corridor: New York City – Albany – Buffalo – Toronto. So, the grand total is three.

      SURPRISE! They are NOT IN CONFLICT. In fact, these three corridors are so far away from being in conflict with each other that somehow trying to forcibly couple them into any number of corridors less than three gets you absolutely batshit routings that don’t make any sense and leave you worse off than you would have been otherwise. Routings like Providence to Springfield, or Hartford to Albany.

      • Adirondacker12800

        Why do people who have origins or destinations other than Hartford or Providence care about travelers between Hartford and Providence? In a future where cars are banned there’s enough demand for three HSR ROWs between Boston and New York. You still don’t come up with “lets connect Hartford to Providence”

        • Ryan

          Except there’s only two lines between Boston and New York. The one that goes to NYC, and the one that goes to Albany and Montreal. The third line is New York and Canada and doesn’t touch Massachusetts (or Rhode Island, or Connecticut).

          And yes, you do come up with connecting Hartford and Providence, as both of those are state capitals, both metro regions are of comparable size, and unlike virtually every other city pair you can think of in New England, there’s no interstate connection. You said it yourself: the drive between the two is obnoxious (and might I add – still actively hazardous in places, especially re: trucks), and for that reason, this connection is getting built one way or the other.

          The other way, by the way, is Interstate 82. Are you so dead set against this routing that you’d prefer the highway be completed instead? Because I-82 is gaining steam again and will almost certainly be built if the rail line isn’t.

          • Adirondacker12800

            They’ll have connections taking the train from Providence to New Haven and changing trains for Hartford or vice versa. Though taking the train to Worcester and changing to the train for Hartford or vice versa would probably be faster. Or the train from Providence to Buffalo and changing in Springfield. If there’s a lot of demand they can get on the train that goes to the other via Springfield and Worcester or via New Haven.

        • Peter Brassard

          The three most disconnected regions within the Northeast by both rail and roadway are Long Island, Hartford, and Providence.

          The three largest metropolitan areas by population and economy between New York and Boston are Long Island, Hartford, and Providence.

          No other combination of regions between New York and Boston, even if Albany is added, has as large a population as does Long Island, Hartford, and Providence, which is 5.59-million or almost a million more than Boston itself.

          Not including New York and Boston, the top four regions for interregional travel (of all travel modes) are Long Island, New Haven, Hartford, and Providence. Springfield is somewhere in the middle and Worcester is toward the bottom.

          Even though the expense would be enormous, Long Island should to be added to any analysis for a proposed second NEC-HSR route in southern New England.

          Long Island, Hartford, Providence, and New Haven already have the necessary population and interregional travel to justify a second NEC cross-sound/inland route. Does Springfield and Worcester?

          • Alon Levy

            Is it really justifiable to build such a long tunnel, though? Penn Design’s proposed Sunnel is 33 km long. At the Chunnel’s per-km cost that’s about $11 billion. If you tunnel under a narrower part of the Sound, you either go too far east and miss New Haven and Hartford, or go too far west and still have to deal with Bridgeport-area alignment pain.

            And that’s without discussing approaches: although the track geometry in Long Island is much easier than in Connecticut, everything else there is more difficult. There is a lot more development next to any possible alignment in Long Island than in Fairfield County, and noise regulations alone could prevent trains from doing more than 200 km/h. There is more demand for commuter service – the LIRR has twice the ridership of the New Haven Line – and there are critical capacity problems on the Main Line that mean any capacity expansion should go to commuter rail first. The transition between the Main Line ROW and the power line leading to the Sunnel is ghastly in terms of both curve radius and takings.

            On a broader level, I don’t think any of the intermediate cities is large enough to justify a second HSR line. A first HSR line, sure, and whoever doesn’t get it should get high-quality branches. This means 200 km/h from New Haven to Springfield if adding Hartford to the mainline is judged not worth the slowdown, 160-200 km/h limited-stop trains running frequently all day from Ronkonkoma to Penn Station (the Main Line has to be four-tracked anyway), fast regional trains from Boston to the Nashua-Concord corridor (with the North-South Rail Link), etc.

          • Peter Brassard

            I take your point. A Port Jefferson to Bridgeport sound crossing would be roughly half the distance of crossing from the island to New Haven, would using the CT Route 8 corridor make a sound crossing more viable? NEC Future’s Alternative 15, suggested a Nassau to Stamford crossing, but that would involve either using the circuitous Danbury route or awkwardly dealing with the 37km/23-mile segment between Stamford and Bridgeport. If money is the big hurdle, would upgrades to the existing NEC route be the only viable alternative?

          • Adirondacker12800

            The market between Springfield and Worcester isn’t big enough for HSR, Neither is the market between Hartford and Providence, not across rolling hills. All four of them get HSR because they are along the way between bigger markets that do make sense for HSR.
            If we are only ever going to build one HSR ROW across the Connecticut River Hartford-Providence is worth examining. If there are going to be two, Hartford-Providence is the one I think should be the third.

          • Michael Noda

            Third? Ouch. Although I suppose that means you have a solution to New London that elevates the Shore Line to a clearly superior position over Hartford. Would you mind sharing?

            I’m not absolutely convinced of Hartford’s superiority over the Shore Line or the Empire Cross, nor am I absolutely convinced BOS-PVD-HFD-NYC is superior to BOS-SPG-HFD-NYC, but a strong case has been made, and I won’t dismiss it without due consideration.

          • Adirondacker12800

            Is it really justifiable to build such a long tunnel, though?

            It doesn’t need to be tunnel the whole way. The Chesapeake Bay Bridge Tunnel is 23 miles long and there’s a lot more people on the Long Island and Connecticut sides of a crossing than there are in Newport News and the, relatively speaking, rural Delmarva peninsula. It has tunnels because there is only one outlet from Newport News to the ocean. There are two from anyplace in the Sound. You might not even need tunnels.

            There is a lot more development next to any possible alignment in Long Island than in Fairfield County

            Go all the way out to Yaphank and use the median of the Robert Moses extravaganza of the William Floyd Parkway. I’m sure it’s a lovely parkway. No one lives near it because it’s a noisy parkway and there’s scads of undeveloped land out there that isn’t next to a noisy parkway. The shore is in the middle of where the Shoreham Nuclear plant was supposed to be. Who knows who owns it since LILCO went belly up and became LIPA which is being bought by PSE&G or maybe it’s National Grid…

            and there are critical capacity problems on the Main Line that mean any capacity expansion should go to commuter rail first.

            The MTA offered to mitigate them. The NIMBY’s objected. The MTA had better things to do with the money. Like complete East Side Access. Which is why they were offering grade separations etc. The NIMBY’s made their bed. Let them lie in it until the honking from the traffic jams a busier Main Line having ESA trains being wedged onto it makes them beg and plead with the MTA for grade separations. If the MTA has ulterior motives I’m sure that was one of them. Make the grade separation project four tracks wide designed to have a second deck added in 2075. Hicksville and east is wide enough for four tracks and four tracks is good enough until there are capacity problems into Manhattan again. Shifting LIRR traffic to Wall Street would be one of the options. Spend the money that would have been spent on a deep cavern under Penn Station on a deep cavern under Wall Street. Trains from east of Babylon don’t have to go through Hicksville. Neither do the ones east of Stewart Manor, the could go through West Hempstead. Nobody would be happy about that but they could. You’d still problems at Jamaica but a tunnel from Floral Park to Wall Street could help.. but then if you are double decking they could use the second deck to get to Jamaica and Brooklyn.

            Go play in traffic for a while and see how attractive a really high toll to New Haven is if you are going to New Haven from the Island. Or Bridgeport. Even if you are in Hauppaug and you want to get to Norwalk. No contest if you want to get anyplace east of New Haven.

            What does New Haven to NYC have to do with the merits of Hartford to Providence?
            … Google maps will tell you to get from Stamford to New Brunswick over the George Washington Bridge – use I-95 though no one in metro New York calls that I-95 anything with numbers. Only at 3 AM. During what can be described as “day” or “evening” it’s faster to take the Tappan Zee and go all the way out to the Garden State Parkway….

          • Charles Smith

            Peter: Your focus on demand is dead on. And, you’re gut is right. – From a demand point of view, Springfield and Worcester have less demand than a simple population analysis would suggest, because demand is a function of income as much as it is of headcount, and Springfield and Worcester are poor cities. If one is doing a simple gravity demand model, one calculates demand between two points as proportional to

            [(population of a*avg income of a)**0.8] * [(population of b*avg income of b)**0.8]

            This clearly reduces the amount of demand that these cities will produce today below the levels suggested by simple population analysis. It also suggests that Hartford will generate less revenue than simple population analysis would predict. Long Island, on the other hand, is not poor, and desperately needs connections to the North and East.

            From an economic development point of view, giving poor cities good passenger transportation tends over time to significantly improve the average income, but this takes years (and years). So, a public policy argument can be made for providing otherwise less economically sound service. In the case of New York and New England, however, an even stronger public policy argument can be made for improving freight service – the financial roi is real; the societal roi is extraordinary, particularly for Long Island, Connecticut, RI, and Massachusetts.

          • Ryan

            I think that’s the crux of the issue right there. There’s a huge unmet demand for moving both people and freight between Providence and Hartford – a demand which I am absolutely confident is going to lead to something being done eventually.

            As far as I’m aware, the I-384 extension to Willimantic is still on the books, and I believe RIDOT similarly has plans to connect 295 to 395 in Connecticut. These are both subject to being shot down, of course (and I-384 has been on again/off again for multiple decades now) but the fact that these are still regarded as serious proposals at all should speak to the importance of connecting Hartford and Providence. And with each successive portion of disconnected expressway that gets built to Interstate standards, the amount of work that would need to be done to connect them all grows smaller and smaller. I’ll say it again – I’m absolutely confident that Interstate 82 is a thing that we could see opened in our lifetimes, no matter if it’s called I-82, 384, US-6, the Intercapital Connector or anything else you can think of.

            But – and this is a pretty big but – none of the prerequisite road projects have made it to construction, let alone completion – yet. That means two very important things:

            #1) The huge inherent demand for both passenger and freight travel between Hartford and Providence can be applied towards driving the opening of a rail line instead of the completion of a freeway, and

            #2) Unlike HSR on the B&A Main Line, the Shore Line, the New Haven – Springfield Line, or actually any other permutation of existing lines we could rattle off as potential HSR alignments, the Hartford – Providence greenfield alignment does not have to compete with an interstate for demand.

            The combination of both of these things means while the cost savings may be relative, the demand/benefits side of this equation comes down heavily enough in favor of Hartford and Providence that if I’m right (and I maintain that I am) about the completion of some kind of link between Hartford and Providence being a certainty, then the cost-benefit analysis of any other feasible routing must be adjusted to account for the cost of a new freeway that most likely lives or dies based entirely on whether or not a rail line capable of carrying both passengers and freight is approved and constructed before the I-384 extension and/or enough further improvements to US-6 are.

          • Adirondacker12800

            Third? Ouch.

            Yes third. There’s gonna be trains running from New York to Springfield and then onto Boston. One or two an hour. If the “higher speed” section is Springfield to New Haven thats 60 miles in nice round numbers. In even rounder numbers that means Hartford is 30 miles from either
            At an average speed of 60 it takes 30 minutes from Hartford to either
            At an average speed of 90, which they could probably do with minor upgrades and grade crossing eliminations on the current line, it takes 20.
            At an average speed of 120 which means you are building lots of expensive greenfield HSR through suburban Connecticut it takes15.
            At an average speed of 150, which would be very difficult to do with greenfield HSR, a station stop and the merge into the line at either end, it takes 12. You save a whole 8 minutes by building HSR up and down the Valley.
            It’s not worth it to save 8 minutes to or from Hartford or 16 minutes to or from New Haven and Springfield. People aren’t going to drive to New Haven or Springfield over 10 minutes. Anyway at an average speed of 90 they are leave Hartford on the train and are departing New Haven or Springfield at the same time they would be pulling off the highway. Faster than driving is good enough for Hartford. Spend the money that would have been spent tunneling through the rolling hills to viaducts across the gentle valleys of Connecticut to do the same in Massachusetts. That gets all of New England Albany and anything beyond Albany for 75 miles more tunnels and viaducts. Which are going to be built anyway.

          • Adirondacker12800

            Hartford – Providence greenfield alignment does not have to compete with an interstate for demand.

            If there was a lot of demand they would have built an interstate grade highway. They haven’t. Which means there isn’t a lot of demand.
            No you can’t compare it to what’s going on in Virginia. Those people are getting on the train to go to Washington DC. Baltimore and Philadelphia. Many of them all the way to New York. The train is faster than driving for all those destinations. Which all have decent mass transit. And 30 million people in their combined metro areas.

          • Peter Brassard

            Charles: You raise an interesting point regarding the impact of wealth in how it affects demand. There may be other factors that effect demand as well.

            There are certainly a fair number of wealth pockets in the Providence area. However, when the Providence area is compared with other regions, it usually appears to be a poorer place. Also, why does Springfield, which is poorer than the Worcester area, have a higher rate of interregional travel? University enrollment might be part of the reason.

            Western Massachusetts (Springfiled area): 129,000 students
            Providence proper: 40,000+ students; with all of Rhode Island: 63,000 students
            Worcester County: 36,000 students

            Why is Kingston such a successful suburban/rural train station? Besides much of South County being an affluent area, the answer is also URI with its 17,000 students a little over a mile from the station. Besides Hartford proper and the region, an inland Hartford-Providence route would have a built in ridership with the 18,000 UConn students at Storrs, which is only a two or three miles from where Alon proposed his inland route, within the same town, not to mention other central Connecticut colleges.

            Students themselves may be classified as poor in statistics, but their parents who live elsewhere who pay expenses may not be and students who are from less advantaged households usually get grants and/or loans, which doesn’t show as income. This may mask an area’s wealth, making it appear to be poorer, when actually it’s richer due in part to its student population.

          • Ryan

            If there was a lot of demand they would have built an interstate grade highway. They haven’t. Which means there isn’t a lot of demand.

            Wow! You’re absolutely right! Boston’s Southwest Expressway, Hartford’s Beltway, Providence’s Other Beltway , the rest of the Long Island Expressway, the Long Island Crossing, and countless other interstate-grade highway projects have all been dropped from the books because there’s just no demand for people to get between any of the places that those highways would have traveled, not because of any sort of wide-spread protest or action by ordinary citizens en masse.

            Gosh, it’s a good thing you pointed that out – otherwise, I might have gone on thinking that the highway revolts were actually an example of successful community organization on a national level and a cautionary tale for future politicians and engineers about angering the people of this country and pushing governmental prerogatives such as eminent domain too far! Man, it’s good to know that really, citizens have no power and the highway revolts actually appearing to have an impact was just one big coincidence.

          • Ryan

            Yes third. There’s gonna be trains running from New York to Springfield and then onto Boston. One or two an hour.

            Amtrak’s Vision, which I have open and in front of me right now, explicitly refers to all service to Springfield as lower-speed “branch” service. A cursory search of the document reveals the word “Inland” was used exactly zero times.

            Amtrak’s Infrastructure Master Plan, which is actually grounded in the tenets of reality and sanity, does mention the Inland Route as a primarily state project and sets a goal of “five trains” operating over it, which under the most liberal interpretation possible means five round-trips by Northeast Regionals. Amtrak’s Infrastructure Master Plan also, however, calls for hourly Acela service and “near-hourly” Regional service between Boston and New York.

            Hourly or better service from New York to Springfield to Boston appears to be entirely outside the scope of what Amtrak wants to provide. Even if you assume they want to provide more service than they’re willing to put to paper, you need to assume they want to provide at least four times the amount of service relative to what their stated goals are before you get to hourly Inland Regional service, and eight times the amount of service to get to half-hourly headways.

            I will agree with you at least as far as there being nowhere near enough demand for three distinct HSR alignments in New England. However, I’ll disagree with your assessment as to what the “third” line is if we only get two: it’s going to be the Shore Line.

          • Adirondacker12800

            Amtrak’s Vision,

            Is for the NEC. Anything that doesn’t involve going through Manhattan is outside of the scope of the study.

          • Charles Smith

            Your argument for a Long Island connection is strengthened, if one were to believe that the only route which would permit a true 400kph/250mph alignment between New Haven and NYC is via Long Island. The problems of fixing the ShoreLine or even of going up through Westchester may just not be sufficiently susceptible to solution no matter how many dollars one wished to put in. I have been driven to accepting the Long Island routing, by trying to make the alternatives work. One can argue about the length of the tunnel, but if it gives Long Island passengers a connection to the North and permits a non-stop New Haven to Penn Station NY time of 27 min or less, the Long Island routing is a huge win. (The UPenn Design folks deserve a lot of credit for getting our heads enough out of the box to put Long Island on the map.)

            Yes, I know that my trains accelerate at .1g and do 240mph, but even if they were Alon’s “slowpokes,” going through Long Island may be the only way to obtain true speed between New York City and New Haven. And, the Connecticut/New York shore west of New Haven generate enough traffic to justify most, if not all, of the improvements that Alon would suggest in his comment in addition to paying even for the “long, long tunnel.” (I suspect we all would learn a lot from a separate piece by Alon on all of what he would propose between NYC and New Haven, not just interim steps, but that is not central to this discussion.)

  8. Charles Smith

    Elaboration on one detail. I know that what follows is teaching many of my grandmothers here to suck eggs, but the conclusion changed my thinking.

    A slight generalization and the introduction of one additional factoid significantly strengthen your analytic framework for deciding whether to include intermediate stops, but may change the specific conclusions:

    1. The generalization: Time cost of an intermediate stop = {(Time lost in deceleration) + (Resting Time) + (Time lost in acceleration)}.

    Definition: Time lost in acceleration or deceleration = {(Time to traverse distance while accelerating or decelerating) minus (Time to traverse distance traveling at full speed)}

    Simple analytic trick 1: Assume acceleration and deceleration while having opposite vectors are of the same magnitude. Then, the time lost in acceleration and deceleration is twice the time lost either to decelerate or accelerate.

    Simple analytic trick 2: Assume that acceleration is constant or approximately so, that the vehicle is traveling at max speed, and that the vehicle decelerates at the max permitted acceleration. Then,
    Stopping Time = Max Speed/Max Acceleration
    Time Lost due to Acceleration and Deceleration = Stopping Time

    For analytic purposes, approximate 100 Kph by 60 mph or 1 mile per minute.
    Also, for analytic purposes, approximate the top speed of High Speed Rail at 400 kph or 4 miles per minute. (They will get there one of these days.)

    Extend this approximation so that
    an acceleration of 0.1g will be approximated for analytic purposes by
    200 kph per minute
    120 mph per minute or
    2 miles per minute per minute.

    Thus an acceleration of 0.05g is approximately
    1 mile per minute per minute

    And, an acceleration of 0.25g is approximately
    0.5 miles per minute per minute.

    This says that the time lost due to acceleration AND deceleration is approximately
    2 minutes for an acceleration of 0.1 g
    4 minutes for an acceleration of 0.05g
    8 minutes for an acceleration of 0.25g

    The SNCF and most European railroads schedule a resting time of 5 minutes at major intermediate cities and of 2 minutes at lesser intermediate cities. Historically, at significantly smaller cities, some inter-city rail services (and, I believe, some Shinkansen) have scheduled resting times of only one minute.

    As a result, one can estimate
    time for a stop at a lesser city as
    2 minutes plus (time for acceleration AND deceleration)
    and at a much lesser city as
    1 minute plus (time for acceleration AND deceleration).

    Given an acceleration of 0.025 g, then a stop costs 10 minutes for a lesser city and 9 minutes for a much lesser one.

    Given an acceleration of 0.05g, then a stop costs 6 minutes for a lesser city and 2 minutes for a much lesser one.

    Given an acceleration of 0.1g, then a stop costs 4 minutes for a lesser city and 3 minutes for a much lesser one.

    Thus, the merely by shifting the rate of acceleration and taking into account city size, one can vary the time cost of a station stop by more than a factor of THREE (3).

    2. The factoid (with a bit of discussion). Currently, high speed trains are grossly under-powered. As a result, they have difficulty maintaining full speed on even an 0.5% grade, and more important for purposes of this discussion accelerate less than an average rate of 0.025 g. (Note, this is much less acceleration than a 1954 VW or even a Trabant can manage.)
    Further, the low rate of acceleration at high speeds doubles, quadruples, and sometimes octuples the time to recover from slowdowns, thus dramatically increasing overall running times.

    The European spec for HSR limits routine acceleration/deceleration in the direction of travel to 0.05 g, even though the same spec limits lateral acceleration to 0.15g, three times as much. After some prodding, two explanations for this specification emerged, both of which could have been better treated by controlling the actual variable that needs to be controlled. If anyone out there has the references that put quantitative clothes on these issues, I would be immensely appreciative.

    The first explanation was a need to limit the amount of force being exerted at any given wheel. Clearly, HSR rolling stock is powering more and more wheels in the consist, and not just a few wheels in the motor, so putting the force constants down explicitly and any variation by speed regime should clarify this issue.

    The second explanation was a need to limit the jerkiness experienced by passengers so they wouldn’t be knocked down or spill their coffee. For purposes of analysis, we ignore the possibility that the engineer operating the vehicle might actually be able to control the jerk himself. Instead, we observe that controlling the jerk requires controlling the rate of change of acceleration. The reason that higher levels of lateral acceleration are accepted is that the rate of onset and cessation of such acceleration is constrained by track geometry to be much more gradual than that for acceleration in the direction of travel. Note that there are specifications for third derivative issues in lateral acceleration.

    What is needed is either DIRECT or PROGRAMMED third derivative control of motion in the direction of travel and limitations on the third and fourth derivatives in the specification. (This is NOT a new idea. I first encountered it when riding in a Danish or Swedish diesel-hydraulic railcar in the 1960’s. I was astonished at the elegant acceleration and braking, and fascinated when the crew was kind enough to show me the control system and how it worked.) With third derivative control, average accelerations of .12g to .15g all the way from zero to 400kph/250mph, can be implemented with no risks to passenger comfort or safety. Further, third derivative control expand the envelope of forces the vehicle can safely and effectively exert upon a rail, thus reducing the wheel-rail interaction problem as well.

    One can roughly double the power and acceleration of an AVG or a Velaro by powering every wheel. One can increase the low speed acceleration by using third derivative control and raising the max acceleration to 0.12g until power becomes inadequate. After that, one needs to slightly more than double the power per wheel/axle, which is also not impossible. Assuming rolling stock engineered for routine, fully-loaded operation at 400kph with average acceleration of 0.1g changes timings for non-stop services significantly, and just as importantly dramatically reduces the costs of stopping and serving the “in-between”. Losing 3 minutes to stop in Bristol, makes the stop a lot more attractive than losing 10 minutes doing so.

    • Alon Levy

      Independently of European specs for HSR, there seems to be an adhesion limit. High-performance regional trains, like the FLIRT and Coradia Continental, already accelerate at 1.2 m/s^2 initially. Subway trains do that routinely, too. There’s even a mainline FRA-compliant train running in the US that accelerates at what seems to be that rate from videos (the Silverliner V). HSR acceleration rates are universally lower, though the example train I use in calculations, the N700-I, has initial acceleration of 0.89 m/s^2. My understanding is that there’s a tradeoff between initial acceleration and top speed and that it comes from motor characteristics like gear ratios.

      But the transitions between speed zones in my example timetables are at too high a speed to be governed by initial acceleration. At 210 km/h, the example train would accelerate at 0.46 m/s^2 in a vacuum without friction and actually accelerates at 0.38 m/s^2. And this is a train with all wheels powered, lower weight per unit length than anything that runs in Europe that’s not manufactured by Talgo, and aerodynamics designed around stringent noise limits. It’s possible to increase P/W ratio much farther than that: the N700-I has a power-to-weight ratio of 26.7 kW/t and the Talgo AVRIL has a ratio of about 30, but some experimental Shinkansen sets could get up to about 50. But what’s the point? It makes the trains a lot more expensive to buy and maintain, there are still noise issues in built-up areas, and there are diminishing returns to being able to transition from 210 to 270 more quickly.

      If the point is to use experimental technology to improve performance even further, then it’s much more useful to improve tilting. It is possible to build trains that can do 250 mm of cant deficiency at 360 km/h; so far the upper limit is 180 on the Talgo and about 175 on the E5/E6, but the technology for going farther up exists elsewhere and has just not been used together with high P/W ratios and high top speeds yet. One of the example trains I tested a while ago was a mongrel New Pendolino with all axles powered to give about 20 kW/t, which turned out to outperform the faster-accelerating N700-I on New York-New Haven. The problem: nobody needs such a train outside the Northeast Corridor. Most legacy rail lines have sharp speed limits coming from heavy low-speed traffic and sharp curves, and old high-speed lines are built to high enough standards that active suspension is enough. The NEC’s status as a partially legacy, partially high-speed line is unique.

      The problem with doing that is the same as with raising power, though: it makes rolling stock procurement much harder. Discarding the Velaro, Zefiro, and even AGV in favor of the AVRIL and N700-I/E5 means less competitive bids and more expensive trains, but saving the same amount of time with infrastructure upgrades is more expensive. But once you hit the upper limit of existing production trains, it’s cheaper to build longer tunnels, fully four-track the Providence Line, exercise more eminent domain in Connecticut and New Jersey, build the more expensive Baltimore alignment, and so on.

      • Adirondacker12800

        All this twacking of slide rules and calculating charts is, I’m sure, a lot of fun. What’s the time difference between a Nozomi and a Kodama divided by the number of extra stops? That’s real tracks with real stations and real trains being used by real people.

        Some of the time lost on a Regional between NY and DC is the speed difference between 125 and 135. Some of it is that for the Regional to stop it’s on the Class 5 or 6 local tracks instead of the Class 7 express tracks. Lots of it is in the extra stops the Regionals make. The long distance trains take forever but thats because they dwell forever, can’t go faster than 100 and spend a lot of time on the local track. Shinkansen are all the same trains on the same track.

        The report has disappeared into the mists of the NYSDOT’s website. The one about using diesels between Albany and Montreal. With a 150 MPH maximum speed on the hypothetical diesel, stopping ate up 4 minutes not including dwell time.

        BTW modern trains using variable frequency inverters don’t behave the way your slide rule expects them to. The nice three phase motors behave differently at 15Hz and the same same voltage and different frequency. Or at the different voltage and a different frequency. That’s one of the reasons they have sophisticated software driving the inverter instead of having the engineer watch the ammeter. And they can be overdriven safely for short periods. On the order of tens of seconds fairly frequently. And slightly overdriven for longer periods.

        • Alon Levy

          The Kodama dwells for a long time at overtake stations. One commenter, I think Swing Hanger, said he could run and buy something from a station while his Kodama was making a stop. JR Central claims the N700 does 0-270 km/h in 3 minutes, which imposes an upper bound of 90 seconds on each of acceleration and deceleration time penalties. This assumes constant-rate acceleration; in reality the train accelerates faster at lower speed than at higher speed, so it spends more of those 3 minutes at higher speed than at lower speed, reducing the time penalty to less than 50% the acceleration time. Given 50-second Nozomi dwells this is an upper bound of 3:50. Actual timetabled speed difference is 7 minutes, but this includes overtake dwells.

          For what it’s worth, the Acelas that stop at Trenton take 5 minutes longer than the Acelas that don’t. The Regionals are a bit slower for the reasons you mentioned plus worse acceleration profiles (less power, more weight, lower percentage of the train’s weight on drivers, worse aerodynamics at higher speed) and probably more schedule padding for trains coming north from Virginia.

          Modern trains behave slightly differently at different voltages, but it’s not a big difference. They lose a bit of power and accelerate more slowly at less than 25 kV, but the difference between 25 and 15 kV is not big enough to worry about; Alstom supplies AGV acceleration profile information for multiple voltages. The acceleration curves for all of those trains are pretty smooth, the acceleration profiles of medium-speed regional trains on YouTube also look smooth. Voltages and frequencies are a big deal for the engineers designing the motors, but not so much for people trying to figure out performance specs.

          • Charles Smith

            Thank you for the excellent Acela example. The max speed for Acela’s around Trenton today is on the racetrack through Princeton Junction and Monmouth Junction. The Acela was built to accelerate at approx. 0.07 g until power limited, and appears to be actually starting out from Trenton accelerating at about 0.05g these days. Thus, given the builder’s acceleration curves, time lost due to the stop should be 2 min 40 sec plus rest time of about 2 min plus safety time of 20 sec. for a total schedule difference with the non-stop of 5 minutes. Obviously, if the Acela were going even 160 (as it actually may one of these years), much less 240, the stopping penalty would be larger.

          • Adirondacker12800

            They lose a bit of power and accelerate more slowly at less than 25 kV, but the difference between 25 and 15 kV is not big enough to worry about

            Different voltages and frequencies on the motor side of the power system. The wonder of IGBT power supplies is that the software can control them and you can get almost any voltage and frequency combination you want on the output side. The motors don’t care whats on the pantograph side of the power system these days.
            Until you bump into volts x amps = watts.
            If you can suck down 6 megawatts at 25,000 volts that means you are using 240 amps. put in a nice 300 amp pantograph and you’ll almost never overload it. Running the same train at 15,000 volts to get 6 megawatts you need 400 amps. Which makes your 300 amp pantograph very very unhappy. Put in a 500 amp pantograph and you are blowing the circuit breakers back at the substation when it draws 400 amps. Because the wires are only good for 375, which is all they need for the regional trains that only use 5 megawatts. Or melting the transformer. Or both.

      • Charles Smith

        I love your examples. You probably should add the Frecciarossa 1000, aka the Zefiro V300 to your list, built by Bombardier and AnsaldoBreda. It has a fully loaded, operational top speed of 360kmh AND tilting, not perfect but tilting all the same. In spite of the unique spec, the initial order of 50 trainsets came in about 12-12% lower than bids for competing standard trainsets.

      • Charles Smith

        I love your super-Pendolino to New Haven. How can one disagree with a man who see the virtue in tilting and increasing cant deficiency. Of course, it’s right, and it makes a significant difference almost anywhere.

        Combining higher cant deficiency with higher acceleration is not necessarily a diminishing returns game. The power that enables higher acceleration at high speeds has three important effects:

        1. It can have a significant but rarely dramatic effect on non-stop times.
        2. It has a dramatically effect on times for stopping trains. Applying that difference to an upgraded NEC, demand increases by at least 12 million passengers per year originating from smaller intermediate cities, almost all of them un-served or grossly under-served by timetables built around low acceleration rolling stock. A look at the PennDesign folks demand v service matrix shows you precisely where these passengers come from.
        3. The same power that permits acceleration also permits one to manage grades without slowing down. Think how much different the design and cost of new alignments would be if one had confidence that the rolling stock could handle 4% (or even 2%) grades without slowing down between Harrisburg, Pittsburgh, and Steubenville; between Palm Springs and Phoenix; between Bedminster and Easton; or between Burbank and Bakersfield (between Burbank and Bakersfield Junction, if you’re running the express alignment via the Tejon pass and up I-5 so as to be fast enough to be economically sound and finance-able.

        4. 400kph, 0.1g, and 4% grades look like a natural sweet spot. The power required to do a 4% grade at 400kph without slowing is approximately the amount of power required to have a residual acceleration capacity of 0.1g at 400kph. Also, the basic rail-wheel interaction even under degraded conditions is still sufficient to support this rate of acceleration/deceleration at 400kph.

        • Joey

          The power required to maintain a steady speed uphill is not why speed is limited on grades. What you are actually limited by is (a) being able to start uphill from a full stop and (b) being able to break in a reasonable distance downhill.

          • Charles Smith

            I have hesitated to reply to Joey’s comments on acceleration, because I am not sure which of several points he is trying to make. The following text tries to go through the possible variations of his argument, and address them. I think we agree on the importance of braking; I hope we agree on the fundamentals of dynamics and power engineering, and that that agreement on the physics clears the field to see why residual acceleration at full speed is critical to the performance of trains.

            Let’s take the second of the two comments first. I suspect we can be in violent agreement there. With any vehicle, control requires that you be able to decelerate better than you can accelerate. For trains this can be a particularly tricky issue, in large part because trains are dramatically underpowered.

            Let us assume a fixed mass operating in a friction free atmosphere.
            1. The energy that has to be dissipated slowing down from a given velocity ON THE LEVEL is proportional to the square of the velocity.
            2. If one is going downward, then the energy that has to be dissipated is increased proportional to the rate of change in altitude.

            In stopping a train, there are two traditional means of braking
            a. air brakes which use disk brakes attached to the axle, and which also depend on there being sufficient adhesion between wheel and track to accept the braking force. Steel wheel on steel rail adhesion on even modest slopes can be problematic.
            b. regenerative dynamic braking, literally turning the traction motors into electric generators – given diesel-electric engines, regenerative braking is limited by the amount of heat the engine can dissipate, To support the heavy loads of emergency braking, diesel locomotives have resistance grids into which regenerative braking energy is dumped. There is a finite limit to the amount of energy that an diesel-electric can absorb and it can easily be less than required to recover from a given speed on the level or a given rate of descent. –
            If, instead one is propelling the train with an electric “motor” (doesn’t matter whether it’s a GG1 or distributed power in emu’s) the train draws electrical energy from catenary or a third rail. And, the train dumps the dynamic braking energy back into the power distribution system which, assuming not totally pinch-penny electrical engineering, can be safely thought of as an infinite sink. Dynamic braking can usually take up about 85% of the total braking energy assuming the connection to the electric power grid is maintained and the motors have adequate cooling for the relatively small amounts of heat due to inefficiency and friction in their operation.

            To deal with the adhesion issues of emergency stops, high speed rail rolling stock providers have been developing supplemental stopping solutions, most notably electromagnetic and eddy current brakes. There are those who argue that electromagnetic brakes and eddy-current brakes are too new to be trusted; they should examine Swiss history of braking on slopes.

            Speed on any given alignment is regulated by the ability of the alignment to deal with emergency contingencies, so downhill speeds are necessarily limited to what the brakes can handle or the backup brakes or the backup, backup brakes can handle. And, rolling stock performance is similarly specified. For example if CAHSR persists in going through the Tehachapi, they will be speed limited in their descent going south because the curvature will not sustain the speeds on that slope that could eventuate in the event of a braking failure at full speed.

            So, yes, the brakes matter, on trains as much as on your Mercedes coupe. High speed trains tend to be lighter, and brakes are getting better, so let us assume for the moment that the train has adequate braking to go down a 2 or even a 4 % grade safely at full speed. Could you have gotten up the other side. Remember that ascent needs energy input proportional to the rate of change in altitude, IN ADDITION TO the energy required to MAINTAIN SPEED or ACCELERATE on the flat.

            Taking the AGV as an example, it has so little residual acceleration at full speed that operating loaded at 350KPH , it can not sustain speed on even a 1/2% grade. To be able to maintain speed on a 2% grade it needs an increment of more than 170% in power – doable but you have to power every wheel, instead of a little more than half of them, and you then have to upgrade the motors by 40%, doable with the advances over the last few years. (Note that dynamic breaking capacity is increased proportional to power here.) That same additional power also would give the AVG enough residual acceleration to be able to accelerate at .05g all the way to 350kph. (1 mph per sec or 120 sec to 120mph, 240 sec to 240 mph).

            If you want to operate at high speeds and make stops, then you probably want to operate with a sustained acceleration/deceleration to/from full speed of 0.1g to 0.12g (0.1g is approximately 2 miles per hour per second, worse than a Trabant.) which can be quite comfortable given 3rd derivative shaping at the beginning and the end of acceleration. Note that the AVG appears to average about .023g on its way from zero to 360 kph. This acceleration get you to 120mph in only 60 secs and to 240 mph in 120 secs. Assuming 2 minutes resting in station, a stop from 240 mph would add only 4 minutes to the schedule,versus the 10 minutes it would add using an AVG or Velaro. Just as important, the additional power enables the train to slow and recover much more rapidly for slowdowns. Just dialing up the rolling stock power on a BOS, NYC, PHL, DC run would improve the run time by 25 mins over the best time projected by Amtrak for high speed, with no increase in maximum speed and no further improvement in alignment and build out over what they assume. This much improvement in the power to weight ratio requires some serious engineering but is not impossible.

            In all of the above, please note that it is residual acceleration at speed that matters, NOT JUST initial acceleration. This spec is often published but usually ignored, because most people assume that the acceleration at 10mph will be the acceleration at 60 mph or at 240 mph; Sustained acceleration to max speed (not initial acceleration) determines your effective operating times, and the residual acceleration at max speed, tells you what the shape of the acceleration curve must be.

            Now, back to your first point. I can’t tell whether you’re trying to address issues of close coupling, of stiction, of stiction, or of adequacy of torque.

            1. As is implicit in the standard high school physics problem, trains are minimally powered, so starting up a loosely coupled train lets you build momentum as you start pulling the cars, one car at a time, and takes less power than starting a close coupled train smoothly. Even freights now act close coupled because some of the power units are spread through the consist so as to reduce the drawbar load and MORE IMPORTANT to provide electronically controlled brake reservoirs through the train and thus dramatically shorten the time for brake signals to propagate. (Braking is even safer and faster with electronically controlled braking on every wheel, which should be standard on modern passenger equipment, although it doesn’t yet seem to be so. It is an immense safety improvement, reducing the skill required of an engineer in an emergency to something a human being can manage; and the distance traveled while the brakes come into play by miles for long freight consists, for example for unit trains going 50 mph.

            2. Stiction is always interesting and does have to be paid attention to, but is not in the first significant digit for HSR.

            3. Adhesion can be a bitch, and having more power wheels reduces adhesion problems. Required adhesion is proportional to energy required DIVIDED BY NUMBER OF WHEELS. A favorite example of adhesion being addressed by increasing the number of wheels was the re-opening of passenger service into Atlantic City, where a consist being pulled by a diesel-electric locomotive couldn’t get over the brief 4 degree slope onto the bridge even with some momentum behind it, but emu’s could do so effortlessly with less total power starting from a dead stop.

            Third derivative shaping of power output can also help address initial adhesion problems; combining third derivative shaping of power output with feedback at the wheels enables both positive traction control and anti-lock braking.

            4. Low speed torque – the reason that the engine builders have been going to fewer powered axles but with ac vice dc traction motors is that the ac motors can provide much better torque at low speeds for the same amount of power. The electronic power controls on HSR equipment provides as good or better low speed torque and power management as the ac traction motors do for freight engines.

            I hope this addresses your concerns. We agree totally on the need for braking. I hope you agree on the importance of electronic breaking to provide uniform braking and eliminate brake initiation delay, of power dispersion to enhance adhesion, and of 3rd derivative power shaping and feedback with individual wheel control to provide positive traction control and anti-lock braking.

            With all of that agreed, I hope that it is clearer how a rational analyst might argue that high residual acceleration at speed is critical to high speed rail succeeding in hilly or tunneled environments, or on alignments with localized speed reductions like the NEC, or for stopping services, or just to reduce non-stop run time.

  9. Charles Smith

    1. There appears to be an implicit assumption that non-stop and stopping services should share common alignments throughout. Is this really a necessary condition or merely the result of “thinking poor” rather than working through the economics. If one looks at LA-SF, or Chicago to Minneapolis, the economics are unequivocal – the overall system can be profitable, quite possibly so profitable as to be financeable, only if the non-stop alignment and the stopping alignments differ significantly in places.

    I haven’t worked enough of the details to be able to make that argument here. However, it is clear that there is a huge knee in the curve around a BOS-NYC non-stop time of 1h15m or less, that we are all ignoring, because it is radical and not easy to achieve. Examining alignments with that constraint for the non-stop route in mind might be a fruitful exercise.

    Also, there are many circumstances when system completeness will more than pay for itself. For example, if one believes even the most conservative demand analysis, extending CAHSR from SF to Oakland by tunnel significantly reduces station costs in SF, and more than pays for itself in increased traffic over the system. I believe a similar case can be made for the North-South connection in Boston. I can see system completeness benefits coming into play in designing other stopping service and tributary alignments as well.

  10. Adirondacker12800

    The major problem then is New London. The entire complex of crossing the city and the Thames is the biggest difficulty in the route, as outlined above in the water crossing section.

    Inversion 3.
    The Shinkansen Kodama can’t go to the old station. The Acela II or Acela III cars can go to the old station. Is it worth it to worry about dragging the new road ( Shin-kansen ) to the stations that get local ( Kodama ) HSR level service?

    Is it worth it to worry about any of the stops that are between Providence and New Haven? The old tracks can carry the HSR local and the new tracks can carry the HSR expresses…. the bypass around Frenso or Bakersfield for southern New England. Or the bypasses around Italian cities for New England. The old tracks are always going to be there for commuter agency trains.

    Once an hour the HSR trains toddles along the old Shore line from Boston and when it gets to New Haven it expresses to New York. If you want something more local take the SLE train. If you want a Hikari stop between New Haven and DC change to the Hikari in New Haven, New York or Philadelphia. Have it stop in Stamford and Metropark and you don’t need Hikari between New Haven and Trenton for people between New Haven and Providence.

    Stopping the Kodama at the HSR station means you have to coordinate the HSR train over the whole route. Running the Kodama to the SLE station you have to coordinate the HSR trains between New York and New Haven and Providence and Boston. No stations to be built on the new line versus building new stations on the new line. Though it would be coordination from someplace between Kingston and Providence and Boston. Don’t have to worry about three or four tracks through Kingston that way either.

    ….don’t worry about where the Kodama are going to stop between New Haven and Providence they already have a place to stop, the SLE/MBTA station … and the ones in Connecticut are going to have level boarding before anything for HSR gets built. They need level boarding so M8s can serve them…

    • Adirondacker12800

      because that’s where the tracks and the station already are? Zero cost to build either. If they decide not to zero cost to abandon it because the MBTA and the less express trains will still be using it.

      • Ryan

        Except it isn’t zero cost. There’s a pretty sizable time penalty inherent in stopping at Route 128 which is otherwise smack-dab in the middle of nothing but Class 9 track between Canton and Roxbury once Readville Junction is dealt with (happening anyway.)

        There’s also the minor issue of having it as a super-express stop meaning that your super-express trains are making just five stops from one end of the line to the other: DC, Philadelphia, New York City, Boston and… Route 128.

        One of these things is not like the others, one of these things just doesn’t belong…

        • Adirondacker12800

          Just because there’s a platform with an Amtrak shingle hanging over it doesn’t mean all Amtrak trains have to stop there.
          There’s going to be more than one train a day. There’s going to be more than two trains an hour, there are two trains an hour now. One of them can leave Boston and never stop until it gets to New Haven, two can stop in Providence, like they do now. One can stop at Rt.128, Providence, New London and Old Saybrook. When it gets to be four an hour all of them can stop at Providence and New Haven. Two an hour can stop at Rt. 128 and Stamford. One an hour can stop at Rt 128, Providence, TF Green, Kingston, New London, Old Saybrook, New Haven, Bridgeport, Stamford and New Rochelle. If you want to get from Boston to New York you get on the one that only stops in Providence, If you want to get from some station other than Providence you get on one of the trains that does stop at your station. Tweak the fares to keep the seats full.

          • Joey

            I wasn’t asking why there is a stop at Route 128. I was asking why amtrak shows the Super Express Service, which otherwise only serves the main CBDs (DC, Philadelphia, New York, Boston – it even skips Baltimore and Newark) stopping at Route 128.

          • Adirondacker12800

            I dunno why they stop at Rt128, the guy who drew the map has family in Westwood? The guy who drew the map has heard rumors that it’s the happenin’ suburban station and didn’t look at current ridership numbers? The station and tracks are there anyway. If it doesn’t work out they don’t stop there anymore. Didn’t cost them anything. If they don’t stop there it didn’t save them any money either. The person designing schedules in 2050 isn’t going to be using 40 year old maps for advice.

      • Charles Smith

        A stop at Rt 128 will pick up about 20% additional passengers from what you get with just South Station. Were there a N-S connection and a stop or two to the North side that would pick up another 20+%. The demand data is very strong.

        As I still think of Marblehead as home, I’ve resented stopping at 128 all my life.

        Having looked at a fair amount of demand data, I’m beginning to think a good model for how to run high speed rail, is to go like hell between major metropolitan areas, and stop a lot within them. (assuming decent acceleration, of course, he said smiling). Following this model for the Cleveland, Akron, Youngstown area will a bit more than triple demand over treating railroads like airlines, as the SNCF and MWRRI both do in their forecasts for Cleveland.

        • Adirondacker12800

          ,,,but not all of them have to stop at all the suburban stations. There can be an express from Chicago to New York that only stops in Cleveland and Albany and the only reason it stops in Albany is that you can change to the local to New York or Boston there.. Not because Albany is a great place to stop…. That train doesn’t have to stop in Syracuse or Schenectady because the people in Cleveland who want to go to either can get on a “local” and the people in Syracuse and Schenectady can get on a local that originated in Buffalo for New York or Boston. There’s an limited that stopped in suburban Chicago western suburbs of Cleveland, Cleveland and eastern suburbs of Cleveland that dropped off Chicagoans in Cleveland that were replaced by suburban Clevelanders.That then stops in Erie, Buffalo, Rochester, Syracuse and Schenectady ……. People who want to go from Chicago to downtown Cleveland or downtown Cleveland to New York can use the express….

  11. Christopher Parker

    There is an old railroad alignment between Hartford and Providence that runs through Willimantic and Plainfield. I’ve been told that’s the route Amtrak is thinking of and I’m also told that the route is decently aligned for high speed trains. (The TGV to Lyon used some parts of old railroad branch lines).

    • Adirondacker12800

      The old railroad route through Willamantic was built because everybody said what’s now the Shore Line couldn’t be built. Then the Shore Line opened. It was faster so everyone not going to Willamantic started using the Shore Line. It was so slow that it lost through service to New York in the 30s and when the hurricanes of 1955 severed it, it had so little traffic that they didn’t bother to fix it. It never had more than one track.
      … The LIRR railroad was built because everybody said what that was, couldn’t be built and what’s now the Shore Line was even more difficult.
      It’s a rail trail now. Google maps displays parts of it as a nice green line wandering all over the place. Or what I think was part of it. There are official-ish maps on the park’s website.

    • Charles Smith

      The principal advantage of the old Willimantic line is that it avoids the Providence to Boston MBTA chokepoint by using the old Woonsocket – Franklin alignment, also an MBTA local now, and cursed with some really interesting environmental issues between Franklin and Woonsocket. It isn’t clear how much of this can be brought up to 400 kph standards, but it actually increases capacity where an increase is needed AT THE PRICE OF SKIPPING PROVIDENCE. Apparently, Amtrak found this price too high politically so it added 10 minutes to its NY to Boston time and moved their proposed route from Woonsocket to Providence between 2011 and 2012.

      How far West of Woonsocket one should follow the old Willimantic line is less clear; however, one should note that routing through Willimantic is rather roundabout, and that serving Storrs is an admirable goal.

      • Adirondacker12800

        There is no “old Williamantic line” the old Willamantic line is curvier than the Shore line and it’s got hills to tunnel through and valleys to cross with viaducts.

      • Adirondacker12800

        And if there is a “bottleneck” approaching Boston there’s a doozy in Fairfield County that’s going to be the constraint.

      • Alon Levy

        On the contrary: avoiding Providence-Boston is a major disadvantage. The Providence Line is already built to HSR standards for the most part, so using it saves several dozen kilometers of construction cost. The station spacing is wider than on the Franklin Line, which means that it’s possible to share tracks with a smaller speed difference and fewer overtakes, and there’s already a preexisting overtake at Attleboro that can be used. There’s less development next to the ROW, which is part of the reason the station spacing is wider, but which also means that it’s cheaper to widen to four tracks when it becomes necessary; the most development next to the ROW is in Attleboro, which already has four tracks.

        For some schedules showing how track-sharing on the Providence Line is feasible, see and The tl;dr version is that if the MBTA speeds up trains to modern electrified regional rail standards then 4 tph each of Providence regional service and HSR can be accommodated with a single overtake between Route 128 and Readville in addition to Attleboro, or with a single overtake at Sharon without using Attleboro. It’s not said in the post, but 6 tph each (about the upper limit that can be done without extra trans-Hudson capacity) requires using all three overtake positions – Attleboro, Sharon, Route 128-Readville – and at this stage they might as well also four-track Attleboro-Sharon, which is the least ROW-constrained and most schedule-constrained segment between overtakes.

        • Adirondacker12800

          Just using Google for approximate mileages, a T shaped HSR system – New Haven to Springfield with cross in the T at Hartford to Providence means there’s 165 miles of HSR track. It cuts time out of Hartford-Providence and Springfield-Providence trips but adds time to any other trip.
          A triangle shaped system with the leg between Hartford and Springfield upgraded to “higher speed” is 150 miles of HSR track. That adds less than ten minutes to any trip to Hartford. The 165 mile system is 60 miles of Connecticut River Valley construction costs and 60 miles of Connecticut River Valley property acquisition costs. 75 miles of construction costs across the rolling hills of Connecticut. Versus 75 miles across the rolling hills of Massachusetts. Trade the higher construction costs in the Connecticut River Valley along with it’s medium cost property aqquistion costs for low cost construction from Westerly to New Haven with medium cost property acquisition. Might have enough leftover to do Providence Worcester at “higher” speed. Overall faster than a bus in the direction of Worcester is more important to Providence than a slightly faster trip to Hartford. Faster than a bus to New Haven and Springfield is more important to Hartford than slightly faster to Providence…. Build Westerly to New Haven first. Needham to Springfield second and get higher speeds in the Connecticut River valley and service to Worcester for Providence.
          …..once the capacity problems across the Hudson are solved the next place capacity problems appear are into Manhattan again. Providence to Boston can be four tracked cheaply and easily and unless cars are banned or the population of New England quadruples it’s not going to have capacity problems…. How many people want to get to Boston if there are 6 intercity trains, 4 MBTA expresses and 10 MBTA locals at peak?

          • Ryan

            “Overall faster than a bus” is an incredibly low standard that can be met by the commuter rail improvements which are already happening anyway, independent of new HSR construction. So, once you take those off the table, the next three important city pairs you’re left with are Providence and Boston, Hartford and New York, and… Hartford and Providence, which simply is not possible without a new Hartford – Providence line.

            No matter what rationalizing you use to try and justify proposing routings that require a tremendous amount of double-backing and/or connections via lower grade services through lower tier cities – routings like “Providence to Worcester to Hartford” or “Providence to New Haven to Hartford” – the fact of the matter is you didn’t add any amount of time to the trip between Providence and Hartford, you eliminated it. You keep drumming out the “only if cars are banned” line but then you claim that people are going to be more than happy putting up with a shitty, third-tier rail connection instead of dealing with a shitty drive over third-tier roads? No, they’re going to demand something better and we end up building the rail line anyway… or far more likely, we complete the freeway.

            Don’t build Westerly to New Haven first – don’t build Westerly to New Haven at all. The Shore Line isn’t usable as HSR but the demand for traffic between Westerly, New London, Old Saybrook and New Haven skews so heavily towards local traffic anyway that it doesn’t matter that we can’t run HSR service in the area. And because cars won’t be banned, the prospective HSR user in South County or Southeastern Connecticut isn’t going to notice they “lost” service, since the New York/Boston express train wasn’t going to stop anywhere on the Shore Line anyway.

            And that’s the crux of the issue here. You’re suggesting the Shore Line because we don’t have to (and wouldn’t) build stations for the Shore Line HSR – in fact, that was the entire point. Zero new stations means zero stops, but all that actually means is ultimately less people will be served relative to if we had built the new line that will serve new communities by virtue of going somewhere that rail doesn’t currently. Higher cost, yes. But also higher benefit.

          • Alon Levy

            Well, there is an implicit intermediate local-only stop on I-95 HSR, in New London. I’d peg service to New London and Norwich as about equally useful as service to U Conn (and Willimantic). Of course the important New Haven-Providence stop on the inland route is Hartford rather than Storrs-Willimantic… but at any rate, the major cost advantage of the shore route is less overall construction rather than lack of stations. Done right, stations aren’t expensive; an intermediate HSR station, even in a city as big as Providence or Hartford, is 2 high platforms with shelters, signage, a few access points to street level, a bus loop, and some retail.

          • Adirondacker12800

            “Overall faster than a bus” is an incredibly low standard

            How much better is HSR compared to an express bus over 30 miles? How much better is a train that is faster than driving for those 30 miles compared to HSR?
            An average speed of 60 between New Haven and Springfield it would take an hour.
            An average speed of 90 between New Haven and Springfield it would take 40 minutes.
            An average speed of 120 between New Haven and Springfield it would take 30 minutes.
            It’s not worth full blown HSR to save ten minutes.

          • Adirondacker12800

            Well, there is an implicit intermediate local-only stop on I-95 HSR
            Kingston to New Haven right now is around an hour and half. Kingston to Boston is going to get faster if this gets built. So will New Haven to New York. Not by much but having a local toddle through on the tracks that are going to be there anyway for CDOT and RIDOT locals is good enough. You save having to build a 100 million dollar station. Have the HSR “local” stop at TF Green Parking Garages and you get the people who want to get to the airport and people who find getting to TF Green easier than going into downtown Providence.

            Done right, stations aren’t expensive; an intermediate HSR station, even in a city as big as Providence or Hartford, is 2 high platforms with shelters, signage, a few access points to street level, a bus loop, and some retail.

            Providence already has one of those. So do New Haven and Hartford. Hartford needs level boarding, if it doesn’t already have it, which is really really cheap in the greater scheme of things. These trains, whatever they turn out to be, are going to be nominally 10’6″ wide and have floors that give you level boarding at 48″ high platforms.
            Ya wanna make Rye an HSR stop? All it costs is hanging out a shingle with the Amtrak logo on it. Works that way if you want to make Rahway or Cornwells Heights or Aberdeen or….. too.
            Springfield will be interesting but it could use the same solution that the New York-Pittsburgh subway solves in Philadelphia,. You wanna go to Springfield or Worcester, get on the train to Boston or get on Metro North, or whatever Connecticut decides to call the trains shuttling between Stamford and Springfield. You wanna go to Albany get on the train that goes to Albany, that doesn’t stop in Springfield. You cut frequency but you save a half billion dollars “fixing” Springfield. And the half billion or so we’d have to spend getting trains to make a 90 degree turn in downtown Hartford.

          • Ryan

            Well, there is an implicit intermediate local-only stop on I-95 HSR, in New London. I’d peg service to New London and Norwich as about equally useful as service to U Conn (and Willimantic). Of course the important New Haven-Providence stop on the inland route is Hartford rather than Storrs-Willimantic… but at any rate, the major cost advantage of the shore route is less overall construction rather than lack of stations. Done right, stations aren’t expensive; an intermediate HSR station, even in a city as big as Providence or Hartford, is 2 high platforms with shelters, signage, a few access points to street level, a bus loop, and some retail.

            New London Union Station exists already and perfectly serviceable tracks to Norwich (and then on to Worcester) already go there – they just need to be rehabilitated to commuter rail standards and pressed back into service. Everything has the added benefit in this case of being on the correct side of the Thames River to not become embroiled in a political fight over the movable bridge there, as well.

            As far as I’m aware, P&W is in control of that track and far more open to passenger services than certain other freight operators that shall remain nameless. It’s absolutely usable and useful trackage and doesn’t need to wait for action on the NEC corridor to go forward at least as far as Norwich. Cursory Google Maps inspection suggests you could conceivably then run the line from Norwich on out to Willimantic and Storrs, meaning it could link up to Hartford – Providence HSR there anyway while still existing independent of HSR, as well as connecting Conn College and the USCG Academy to UConn. That may be straying too far from the point, though.

            How much better is HSR compared to an express bus over 30 miles? How much better is a train that is faster than driving for those 30 miles compared to HSR?
            An average speed of 60 between New Haven and Springfield it would take an hour.
            An average speed of 90 between New Haven and Springfield it would take 40 minutes.
            An average speed of 120 between New Haven and Springfield it would take 30 minutes.
            It’s not worth full blown HSR to save ten minutes.

            Right – which is why there’s no point in running “inland route” service as HSR. At an average speed of 75 MPH, which is doable with the minimum level of investment required to run regular commuter rail service on the New Haven – Springfield Line, it’s 24 minutes from Hartford to Springfield or New Haven, 48 between New Haven and Springfield. There’s no way that an express bus could ever get close to those travel times, so for all intents and purposes “faster than an express bus” is a solved problem – no HSR required.

            It is, however, worth full-blown HSR to connect two regions that aren’t connected at present, especially where there’s no existing connection – road, rail, or otherwise.

          • Ryan

            “It is, however, worth full-blown HSR to connect two regions that aren’t connected at present, especially where there’s no existing connection – road, rail, or otherwise.”

            The word whatsoever should be inserted right before the hyphen.

          • Adirondacker12800

            It is, however, worth full-blown HSR to connect two regions that aren’t connected at present, especially where there’s no existing connection – road, rail, or otherwise.

            75 MPH service between Springfield and Boston doesn’t get you to Buffalo in 4 hours. If you are in Hartford and you want to get to Providence you take the train to New Haven and change trains or you take the train to Worcester and change trains. The market demand between Hartford and Providence isn’t worth 75 miles of full blown HSR across the rolling hills and valleys of central Connecticut. 75 miles of full blown HSR across the rolling hills and valleys of central Massachusetts gets the people of New England faster trips to anyplace not Hartford.

        • Adirondacker12800

          …. And using the Shore Line saves several dozen between Providence and Westerly-ish.

  12. Adirondacker12800

    In nice round numbers 25% of the population of Connecticut lives in Fairfield County, 25% in New Haven County and 25% in Hartford County. The other 25% of the state’s population lives in the other counties. Which one gets the faster ride to Providence and Boston, The 50% of the population that lives in Fairfield and New Haven counties or the 25% that lives in Hartford County?

    • Ryan

      The 25% that lives in Hartford County gets the “faster” ride, because this is a trick question.

      What you failed to mention when asking this question, and the key point that I have been stressing this entire time, is that by connecting Hartford and Providence with HSR, you provide a fast ride between two state capitals that is otherwise not possible. The 50% in Fairfield and New Haven Counties will have a “slower” ride as a result, by about 10 minutes, but actually, most of the residents of those counties who will be using this line aren’t going to notice or care that their ride is “slower” than it could have potentially been by a factor of about 10 minutes. It still gets them there in fantastic time.

      However, if you don’t build HSR between Hartford and Providence, the 50% of the population that benefits from a ride that’s “faster” by about 10 minutes isn’t really going to notice that, either. The fast train still gets them there in fantastic time and beats driving, it just beats driving 10 minutes better now. The 25% in Hartford County, however, is going to notice that they don’t have a ride to Providence at all.

      This is the point: Joe Q. Hartford, trying to get to Providence, isn’t going to look at the rail map in New England and see “oh, well, I just ride on down to New Haven, switch trains, then ride on out to Providence. Or, wait, maybe I just take the Boston train out to Worcester and switch to a train for Providence out there instead. Man, it’s good to have options!” No, Joe Q. Hartford is going to look at the map, see that there’s no line to Providence, and say “I can’t (won’t) take the train to Providence. There’s no way to do it without going out of the way and possibly getting stuck waiting for a transfer.” Then Joe Q. Hartford is going to get in his car and drive out instead, or just not the take the trip at all. Both cities suffer as a result, and people continue to agitate for something better.

      • Adirondacker12800

        What you failed to mention when asking this question,

        What you fail to realize is that the people in Hartford travel to places other than Providence and people in Providence travel to places other than Hartford. Giving people in Hartford upgraded legacy tracks to the HSR tracks in New Haven and Springfield gives them better transportation overall than HSR tracks to Providence and giving people in Providence upgraded legacy tracks to Worcester and Westerley gives them better travel options than HSR tracks to Hartford.

        • Ryan

          Upgraded legacy tracks for the New Haven – Springfield commuter line are happening anyway, regardless of what the outcome is on Boston-New York City HSR. Upgraded legacy tracks for the Providence-Worcester line are happening anyway, regardless of the outcome for HSR between Boston and New York City.

          Even the other HSR line that New England is bound to get – the one between Boston and Albany – is going to happen totally independent of the HSR line between Boston and New York City.

          The only legitimate “loss” to transportation options that occurs if we go with Hartford – Providence is upgrades to the legacy Shore Line, but it can’t be upgraded much farther than it already has been and a full-on bypass doesn’t serve the needs of South County or New London County, which both need far better regional/local service (and sending the HSR traffic out to Hartford on the Hartford-Providence line frees up enough capacity that improvements aren’t necessary to give Shore Line East and whatever RIDOT calls its South County/Providence service full schedules seven days a week).

          The question doesn’t boil down to “do we want Hartford – Providence OR [Other Rail Projects]?” – it’s “do we want Hartford – Providence AND [Other Rail Projects]?” It’s quite literally, something or nothing?

          • Adirondacker12800

            It’s one HSR corridor running east-west across New England, two HSR corridors or three HSR corridors. I pick Boston-Albany and Boston-Providence-New Haven because if you have those two you don’t need the third.

            Building 75 miles of HSR across the rolling hills of Connecticut between Hartford and Providence and 40 miles of HSR in the sububanized valleys between Hartford and New Haven is going to cost more and add 10 minutes to every trip that is not Hartford-Providence, than 75 miles of HSR across the less diffucult terrain of coastal Connecticut and 25 miles of upgrading the existing. Either way you face the same capacity problems between Providence and Boston and New Haven and New York.

            Claiming that there will be capacity problems between Providence and New Haven… makes me giggle.. At worst it’s capacity problems between Providence and where ever the new HSR row splits off the legacy ROW and whatever that is will be cheap and easy to do.

          • Ryan

            I didn’t claim that there would be any capacity problems between New Haven and Providence. Just the opposite. There’s plenty of room for regional and local services on that line already, so why spend the money to build a bypass? That money is better spent on the 115 miles of HSR between Providence, Hartford and New Haven than it would have been on the 100 miles of HSR between Providence, New London and New Haven.

            Providence – New Haven is one of the least traveled segments of the HSR line that we’ve got today, and absolutely nothing suggests that making the trip faster is suddenly going to drive more demand for people going between Providence and New Haven. More total trains per day and faster trains is certainly going to drive the demand for travel between Providence and Kingston and Westerly – but that’s not a market that HSR can or should address. It’ll certainly drive demand between Westerly and New London and Old Saybrook, but again, that’s not a market that HSR should be addressing.

            HSR cares about the market between Providence and Hartford. HSR cares about the market between Providence and New Haven. 15 additional miles of net construction, 10 additional minutes of net travel time for New York and Boston commuters (and, let me reiterate: Joe Train Rider isn’t going to notice those 10 minutes – let alone make travel decisions based on them), and presumably higher costs per mile of construction are all acceptable prices to pay to connect Providence and Hartford, and I am making that assertion because I’m absolutely confident in it. By all means, I’d love to see this studied in an official capacity because I believe the studies will ultimately agree with my assessment, not yours.

          • Adirondacker12800

            You don’t have to build 100 miles of HSR between Providence and New Haven. You have to build 75 miles of comparatively easy and cheap to build HSR between New Haven and the state line and 25 miles of comparatively easy and cheap to build upgrades to the existing between the state line and Providence. There would be four tracks of capacity between the state line and New Haven, that’s more than enough until some dystopian future where there are no cars or the population of the US is 1.5 billion.

            When you compare the market for Boston-New Haven, Boston-New York, Boston-anything all the way down to Richmond Virginia and the market between Providence and Hartford the question that gets asked is “what market?” They can change trains in New Haven or Worcester. The three people who tell pollsters that they are going to drive because they have to change trains are telling them that because they don’t want you to know they don’t like trains and would find some other reason to drive if there was a direct train.

          • Ryan

            When you compare the market for Boston-New Haven, Boston-New York, Boston-anything all the way down to Richmond Virginia

            Boston-New Haven HSR riders aren’t going to notice 10 minutes, Boston-New York City HSR riders aren’t going to notice 10 minutes, people coming from as far flung places as Richmond Virginia are sure as hell not going to notice the 10 minutes.

            Hartford – Providence is not in direct competition with any of those markets. It can quite easily coexist with the main line NEC.

            They can change trains in New Haven or Worcester. The three people who tell pollsters that they are going to drive because they have to change trains are telling them that because they don’t want you to know they don’t like trains and would find some other reason to drive if there was a direct train.

            This just in: going 30 or 40 miles out of your way and changing trains actually has a substantial impact on total travel time (easily half an hour, potentially longer) in addition to being a lot more immediately and viscerally obvious than 10 minutes “out of sight, out of mind” generated by cutting a slightly longer route between Boston and New York City.

            In a related breaking newsflash, tacking a 30+ minute, 1-transfer detour out to Worcester or New Haven is a real deal breaker for real people who really would take the train otherwise. The vast majority of the people who cite the lack of a direct option as the reason why they would continue to drive between Hartford and Providence are not, in fact, anti-transit NIMBYs trying to veil their true opinion.

          • Alon Levy

            Just one comment on this: the technical time for New York-Boston, assuming a bunch of bypasses of the worst parts of the New Haven Line, is 1:27. 9 minutes are 10% of the travel time.

          • Ryan

            10% is still a very small margin, and I genuinely have trouble believing that we’re ever going to see Boston to New York City in anything less than 2 hours even.

            Even so, adding 10% of the trip time to 80% of the people is, in my opinion, both better and more equitable than adding 40%, 50% or more to the trip times of 20% of the people. (The numbers are almost certainly not that even, in fairness. My point remains unchanged.)

          • Adirondacker12800

            going 30 or 40 miles out of your way and changing trains actually has a substantial impact on total travel time (easily half an hour, potentially longer) in addition to being a lot more immediately and viscerally obvious than 10 minutes “out of sight, out of mind” generated by cutting a slightly longer route between Boston and New York City.

            That’s just too too bad. If everybody gets express service everywhere everybody gets local service everywhere, Some compromises have to be made. People all over New England that have to use something other than their feet to get to the HSR station are going to have two seat rides. The amount of people who can walk to the Hartford station and get to their destination by walking from the Providence station or vice versa isn’t worth worrying about. They can change trains.

            You have three choices,. one HSR corridor, two HSR corridors and worrying about giving people in Ohio train service, any train service or two HSR corridors and giving people Scranton train service . Or the people in Pennsylvania that aren’t in metro Philadlephia or Harrisburg. Or the people in Indiana. There isn’t going to be three. Pick two that gives everybody in New England the better compromise. People all over New England that have to use something other than their feet to get to the HSR station are going to have two seat rides. The amount of people who can walk to the Hartford station and get to their destination by walking from the Providence station or vice versa isn’t worth worrying about. They can change trains.

          • Ryan

            That’s just too too bad. If everybody gets express service everywhere everybody gets local service everywhere, Some compromises have to be made. People all over New England that have to use something other than their feet to get to the HSR station are going to have two seat rides. The amount of people who can walk to the Hartford station and get to their destination by walking from the Providence station or vice versa isn’t worth worrying about. They can change trains.

            You’re right, some compromises do have to be made, which is why running HSR between Boston and New York City via Hartford and Providence means no HSR running on the Shore Line – no Hikari stops, no Kodama stops. And that’s an acceptable compromise, because the Shore Line that we have today is perfectly serviceable for the needs of the New Haven Line, the Shore Line East, and whatever South County Commuter Rail ends up being into perpetuity. The only thing that could cause a legitimate capacity concern is pumping HSR through the Shore Line, which is why the bypass would have been needed.

            But sacrificing the bypass in favor of running a line where trains don’t run today is an acceptable compromise. It means spending more money, yes, but ultimately it also means serving more people, and it doesn’t cause an undue burden on the users of the rest of the line. Sure, another 9 minutes, 10 minutes, some additional percentage worth of travel time means that the trip is not the absolute fastest it could possibly be – but you know what? That’s just too damn bad for the three people who are going to claim that 9 or 10 more minutes between Boston and New York City is the reason why they’re not going to take HSR. Those are your real anti-transit NIMBYs, right there.

            You have three choices,. one HSR corridor, two HSR corridors and worrying about giving people in Ohio train service, any train service or two HSR corridors and giving people Scranton train service . Or the people in Pennsylvania that aren’t in metro Philadlephia or Harrisburg. Or the people in Indiana. There isn’t going to be three. Pick two that gives everybody in New England the better compromise.

            If you’re going to argue that two HSR corridors in New England should be weighed against the needs of places like Ohio, Florida, or Texas, than the correct answer is “zero HSR corridors for anyone” because they are entirely distinct regions and it’s impossible to fairly weight the decision to invest in one region over another. The choice to build HSR in New England OR the South isn’t a choice that can actually be made.

            New England has the metrics necessary to support two HSR lines. Not one, not three, not one-and-a-half – exactly two. And we can build exactly two HSR lines without unduly impacting plans to provide train service to Ohio, or Allegheny County, or Florida, or anywhere else you can possibly think of. So the question is, which two rail lines should we build?

            And the answer is, we should build the two rail lines that serve the maximum number of people, especially those who aren’t served already. That disqualifies the Shore Line, because a Shore Line HSR bypass replicates existing service. Yes, it costs less. But it serves less people, and so it’s not worth the cost savings.

          • Adirondacker12800

            You’re right, some compromises do have to be made, which is why running HSR between Boston and New York City via Hartford and Providence means no HSR running on the Shore Line – no Hikari stops, no Kodama stops.

            So then it’s not going to be stopping in Willamantic?

            Why can’t one of the four five or six trains an hour between Boston and New York take the slow route between Providence and New Haven while the rest of them just blast through? This isn’t BART or the Boston Subway all of the trains don’t have to make all of the stops and all of the trains don’t have to use the same route. The Northeast has enough people to be doing that.

            That gives people who can get to Old Saybrook, New London and Kingston one seat rides to Boston and makes possible one seat rides all the way down to Richmond. One of them can stop in New Rochelle, Bridgeport, Hartford, Springfield and Worcester. That gives all those people one seat rides all the way to Richmond. Someday one or two of the four trains an hour that leave from Harrisburg will be extending to Boston. The 6th or 7th train an hour between Boston and New York can run local-ish all the way to Harrisburg. You can tweak this many other ways.

            If you’re going to argue that two HSR corridors in New England should be weighed against the needs of places like Ohio, Florida, or Texas

            Yes I’m going to argue that. Because the people of New England have to convince Senators from Ohio, Florida and Texas that building a HSR route across New England is worthy of Federal support. Unless you are suggesting that New England does this without Federal support. And the ones from New York, New Jersey, Pennsylvania Maryland Delaware and Virginia who don’t give a flying one how long it takes to get from Hartford to Providence. The ones from Massachusetts for that matter.

            Go for the cheap one that cuts the most travel time from the big market and when that is successful it makes it much easier to get the money for the second one. If New Haven to the state line gets built first it costs just as much to build someplace in suburban Boston to Springfield as it does to build Providence to Hartford and you don’t have to build HSR between New Haven and Springfield. The comparatively small market between Hartford and Providence gets two seat rides like everybody else that’s not at an HSR station. Everybody else gets faster rides. Compromises have to be made.

          • Ryan

            No HSR trains will ever be stopping in Willimantic, or Plainfield, or Coventry, or West Warwick. Very few HSR trains are liable to stop even in Storrs, and that’s only because UConn is there.

            By blazing the path with a zero (or at worst, one) station HSR line running between Hartford and Providence, however, we open up the option to later go back and add local tracks and infill stops in the communities through which the line passes – communities that otherwise don’t get served because if we’re having so much trouble justifying the expense on greenfield HSR, there’s no way in hell that commuter/regional lines would ever be approved. Or, looking at it from another angle, the states in question aren’t liable to get serious about regional rail plans unless the Fed money to back those plans is ready for use, and that money is pretty much tied to HSR lines – which means if the HSR line doesn’t run through first, the regional line isn’t going to ever start running.

            And since you did mention the fact that we’re largely dependent on Fed funding to get anything done, let me just say this: if you build a brand new line following the I-95 corridor, then for all intents and purposes in the eyes of the politicians responsible for approving it, you’ve built a replica of something that exists already. It doesn’t matter if the technology is different or if it’s HSR exclusive trackage or whatever explanation you provide – the politician looks at the I-95 routing and says “so you want $100 billion to create a copy of perfectly good tracks that trains run on already?”

            We’re far more likely to get approval for funding on the Hartford – Providence alignment choice, even though it is going to cost more, because politicians aren’t going to look at it and say “but there’s already a line running there. Why do you want a second one?” Similarly speaking, we’re far more likely to get funding for brand-new alignments running greenfield out in the wilds of the Deep South if and when we can move past general anti-spending sentiment in that region because even though they cost more, it’s not upgrading tracks that are there already or building new tracks in the “same” place the old tracks are.

          • Alon Levy

            New Haven-Kingston isn’t $100 billion. At $10 billion it’d be a boondoggle, and at $5 billion it should still raise questions about what the hell they’re doing. The point of doing things as cheaply as possible is that it’s feasible to go to politicians and say “for $10-15 billion we’re going to knock Boston-Washington down to 3:30 while also allowing a 1,000-seat train every 10 minutes and keeping prices reasonably affordable.”

            If it’s impossible to get approval for a route roughly paralleling the existing one, there are far bigger problems than Hartford. I don’t think the Hartford route is bad, just that it’s somewhat worse than the I-95 route. But once you go south of New Haven, the penchant for avoiding existing infrastructure leads to some really bad proposals, like a Long Island Sound crossing, tunnels under Westchester County suburbs, and the Market East tunnel.

          • Ryan

            I’m arguing for Hartford – Providence and focusing my argument on specifically the routing choice.

            On a broader scale, I agree with you that our penchant for not just avoiding existing infrastructure but actively spitting on it in some cases is a huge problem, even if in certain cases (like Hartford, but also greenfield alignments outside of New England in general) I think it’s inoffensive enough that I’m comfortable standing behind the proposition to build more.

            But do I think that we as a country have far larger problems than where the NEC route ultimately goes? Yes. Do I believe for a single second that this thing gets done for anything less than $50 billion and without all sorts of top-to-bottom nonsense like Moynihan Station or Market East/Philadelphia Airport tunneling? No, I don’t. We just don’t operate that way.

            That’s really a conversation for an entirely different post, though.

          • Adirondacker12800

            No HSR trains will ever be stopping in Willimantic, or Plainfield, or Coventry, or West Warwick. Very few HSR trains are liable to stop even in Storrs, and that’s only because UConn is there.

            If everybody has express service everybody gets local service.

            It’s never going to stop in any of them, the only reason to drag the tracks through them is if you are going to build a station there. It’s cheaper to go around them. At best they might, in some car free future consider a beetfield station between Storrs and Willamantic. Probably not, you can’t just throw down a patch of asphalt by the side of the tracks and plop a bus shelter on it. You need 100 million dollars to build the miles of station siding to serve the station, a station with HSR train length level boarding and a way to get from one platform to the other without crossing the tracks. There aren’t enough people out there to do that. There aren’t enough people out there for a trolley line. That’s what happens when you build a campus on the cheap land in Storrs or State College.

            New Haven-Kingston isn’t $100 billion. At $10 billion it’d be a boondoggle, and at $5 billion it should still raise questions about what the hell they’re doing.

            If the price is inflated for one it’s going to be equally inflated for the other. The 100 miles from Providence to New Haven is going to be cheaper than the 75 from Providence to Hartford. It’s too bad that the relatively few people in Hartford who want to get to Providence and vice versa have to change trains but compromises have to be made. Just like the people in Norwalk or Attleboro will have to change trains.

            But once you go south of New Haven, the penchant for avoiding existing infrastructure..

            Chuo Shinkansen.

            Does the existing infrastructure have the capacity in 2050? There is a limit to how many cubicals can be stuffed into Manhattan. Is that less than the capacity or more than the capacity? It has scads of capacity south of Rahway and north of Stamford. It’s Rahway to Stamford that gets the odd options to crawl up onto the drawing board. Tunnel from White Plains to Linden has it’s charms. A short run to the Hudson gets the Empire Service trains to the deep cavern station with the center of it’s platform under 50 W 38 Street with escalators and funiculars for wheelchair access to Grand Central, Times Square and Penn Station. ( oriented from roughly 36th and Sixth to 40th and Fifth. ) A few commuter trains can be wedged into the HSR tunnels. The Metro North express from Hartford or the NJTransit express from Atlantic City for instance…. A tunnel from Brooklyn to Jersey City gets people out of Penn Station which then frees up capacity for the people who want to get to Penn Station. It makes Hoboken obsolete but that may be a good thing. Frees up capacity on PATH for the people who use it for local travel.

      • Charles Smith


        In addition to maximizing service to populations, there are a large number of specific criteria by which one should judge passenger rail proposals.

        1. connection with major population centers within and without the state

        (In Connecticut’s case, this is unusually difficult, because a. almost the entire state as far north as Hartford and as far east as New London is included in the New York City major combined statistical area and b. the Connecticut population is unusually diffuse. Connecticut has a total population of approximately 3.6 million spread across only eight (8) counties and NO large cities.
        a. Largest County – Fairfield pop 916,829 – cities – #1 Bridgeport pop 144,229, #4 Stamford pop 122,643, #6 Norwalk pop 85,603, #7 Danbury pop 80,893
        b. 2nd Largest County – Hartford pop 894,014 – cities – #3 Hartford pop 124,775, #8 New Britain pop 71,254, #9 Bristol pop 61,353
        c. 3rd Largest County – New Haven pop 862,477 – cities – #2 New Haven pop 129,779, #5 Waterbury pop 110,366, #10 Meriden pop 59,653, #11 Milford pop 52,759, #12 West Haven pop 52,721 )

        2. connection of the state capital with major population centers within the state

        3. connection of the state capital and population centers with major state and national universities (certainly all those with 10K students in residence or more, and arguably those with more than 5k) (In Connecticut’s case, these would be New Haven (Yale and Southern Connecticut), Storrs (University of Connecticut), New London (USCGA plus UConn), Willimantic (Eastern Connecticut), Danbury (Western Connecticut), and New Britain (Central Connecticut).)

        4. connection of the state capital with command, training, and mobilization points of emergency services including the national guard (In Connecticut, this would include connections between Hartford, the ANG base at Bradley Field, and Stones Ranch Military Reservation in the Lymes),

        5. connection of the state capital with capitals of adjacent states (In Connecticut’s case, this would mean connections with Albany, with Boston, and with Providence),

        6. intermodal connections with major airports (Bradley Field and, for Western Connecticut, the 3 major New York airports),

        7. service to military bases (In Connecticut’s case, Subase Groton and the associated State Piers in New London).

        I am surprised by how many of the above criteria we have all ignored in the discussion thus far.

        But, with respect to your argument about connecting two state capitals, I am driven to ask
        Should there be totally direct alignment from Hartford to Boston and to Albany as well?

        • Ryan

          Well, to your specific criteria:

          1 & 2. Hartford – Providence HSR, if it proceeds onto New Haven and rejoins the existing New Haven – New York line, successfully connects the four largest cities in Connecticut (per your numbers) to each other as well as to Providence and the Providence metro region. The Shore Line bypass would maintain New Haven’s connection to Bridgeport and Stamford, but fail to connect Hartford; the inland route or the I-84 corridor options both connect all four cities but only connect them to Springfield and Worcester. I think the Hartford – Providence alignment is superior here.

          3. The Hartford – Providence HSR line would pass close enough to Storrs for a UConn stop. The Shore Line runs through New London, and a legacy line exists to connect New London to Conn College and the USCG Academy – but that same line runs through Norwich and winds up in Storrs anyway, and the entire thing could probably be reactivated independent of whatever the final HSR alignment is. It’s probably a wash.

          4. This is also likely a wash, but I’m not intimately familiar with all of the state’s emergency services locations.

          5. The Hartford – Providence line connects Hartford to Providence to Boston. By bypassing Providence, you could shave some time off the Hartford – Boston trip, but in my opinion that’s being extremely penny wise and pound foolish.

          6. The Shore Line is actually superior for this because it connects to T.F. Green, which the Hartford – Providence alignment doesn’t do.

          7. The Shore Line also wins this metric, although I’m not intimately familiar with the placement of military installations throughout Connecticut, either.

          I’m not arguing that the Shore Line is meritless. I’m arguing that the Shore Line doesn’t need a full-bore HSR bypass constructed because the legacy line we have today is more than sufficient to meet the needs of the region in every respect except for HSR service – but the money spent providing for HSR on the Shore Line is better invested in providing HSR between Hartford and Providence.

          As for Albany… honestly, Albany’s got a pretty raw deal. It’s totally overshadowed by New York City and the line that you can draw between Boston, Providence, and Hartford extends to New York City a hell of a lot better than it does Albany. So, no, I don’t think it’s worth running a direct rail line out to Albany, as that would require an extremely significant deviation.

          I do believe that there should be a direct connection between Boston and Albany, and that that connection should be built independent of the connection between Boston and New York City.

  13. Joey

    Does anyone have decisive data on the difference in ecological impact between the shore route and the Hartford route? To my knowledge not having to cut a new path through wilderness is an advantage, since linear corridors can have a significant impact on the areas around them. But the shore route also involves longer water crossings at the mouths of various rivers, which might impact those areas more.

    • Alon Levy

      I wish. So far nobody’s studying I-95 that I know of; Penn Design claims it’s not constructable (in a way that tunneling near I-84 is…), Amtrak’s ignored it, and I’m not even sure that the relevant NEC Future option is I-95.

      My guess is that the differences even out and the cost per km is the same for both, but I haven’t even tried to get a second-filter estimate.

  14. Charles Smith

    1. The surge in column contributions suggests the term is over. Hope that you’re enjoying the break.
    2. Thank you for showing how good an alignment one can achieve using I-95 through Connecticut and Rhode Island. It has a remarkable benefit to cost ratio.
    3. Given the work you’ve already done, how well can an I-95 based alignment be made to work when one is seriously trying to fly through as close to 400kph as possible?
    (I am NOT quite monomaniacally fixated on the higher speed, but there are some really large knees in the demand curve around 1h50 non-stop and 1h15 non-stop BOS-NYC. It is not clear that capital investment in the corridor is self-financing with end-to-end non-stop times above two hours. One should still push ahead because of the overall public benefit, and to generate more demand for the higher performance, truly financially sound option. BUT, Meeting the one hour 50 minute knee, makes an average investment (excluding stations and rolling stock) per route mile over the entire route of $100 million financially sound using a government-guaranteed, inflation-expectant, Amtrak-operated, full financing model. Beating the one hour 15 min knee would justify an average investment per route mile of at least $170 million and, given appropriate systemic embedding, as much as $235 million per route mile.)
    4. When discussing the “New London” station, a key goal is to enable the “shoreline services both toward NH and toward Westerly to have either cross platform transfers or just continue through on the I95 line, a sort of exurban S-bahn. It would also be desirable to have the lower speed service through Mohegan territory, Norwich, Willimantic, Storrs… able to get into the same station. Why not put the station on the GROTON side, east of Fairview Ave, between I-95 and Bridge Street. If you can make the track heights compatible, I think the Groton location solves the other problems, although it does suggest that the Norwich service should come down the West bank of the Thames and across the bridge to Groton. One would really like the station to be served by a streetcar to EB and Subase Groton that goes across the bridge to State piers and downtown New London, and perhaps up river at least to USCGA and Conn College.
    5. I believe but have not heard you verify that your I-95 solution does not suffer from service frequency limits due to environmental agreements around the use of movable bridges as the Shore Line alignment does. Merely eliminating these will permit enough additional service frequency to very significantly increase demand, probably close to double it.
    6. As your prior work so clearly shows, current fares on this route are uneconomically high (likely to limit demand to what the environmental agreements will permit). If one were to lower expected average revenue per passenger mile to around $.22 (i.e. an unrestricted coach fare of $.16), which seems to cross optimize public benefit and financial performance, then demand would MUCH more than double again. Even if one were merely to lower the expected average revenue per passenger mile to somewhere between $.28 and $.34 (the SNCF pricing approach which assumes near-obscene operator profit), ridership would likely double.
    7. As the FRA has at least implicitly accepted (Look closely at the service diagrams in their proposed routes report that you cited.), the Boston North-South connection will increase traffic more than enough to pay for itself at the demand level of the 1 hour 45 min knee, and you can lengthen the route by a mile and some, and add another $15 million per mile over the entire run, to more than pay for the N-S connection.
    8. There has been some poor mouthing in comments here and elsewhere about how many people might want to go between the 3.9 million person Boston statistical area and the 20+ million person New York statistical area. If one merely notes the effects discussed above, the numbers are clearly much larger than one might project if one incorrectly were to assume present demand is unconstrained by pricing, environmental agreements, and omissions of alignment necessary for system completeness.
    9. It doesn’t matter how one routes high speed service across Connecticut, it is hard to imagine not having well justified high speed service connecting NH and Hartford, and in fact linking NH, Hartford, Bradley Field, and Springfield. A possibly more aggressive alignment through Hartford, may be justified for this service..
    Like all stopping services, the one through New Haven, North Haven, Wallingford, Meriden, New Britain, Hartford, Windsor Locks, ?, Springfield, Northampton, Holyoke, and Amherst, while electrified and grade-separated, and possibly cross-connected to hs lines at NH, Hartford, and Springfield is a regional service, and should not be expected to self-finance its capital cost.
    10. It is highly desirable that any heavily traveled route have multiple emergency alternatives, a BOS-Worcester and perhaps Springfield alignment in addition to I-95 might be an excellent precaution against the every 20 year or more often hurricane, and the all too frequent other outages now being experienced on the NEC.
    11. BOS-Worcester-Springfield-Pittsfield-Albany gives Boston an important connection to the West. It is also from a Massachusetts point of view very significant in pulling the state together, and providing a larger housing pool for the metropolis to draw from.
    12. If I were the governor of Connecticut I would want the Bradley connection and a station stop in the Lymes for public safety reasons.
    13. Completeness of system coverage really does matter, and a lower speed East-West connection Brewster-Danbury-Waterbury-Bristol-New Britain-Hartford-East Hartford (?Pratt and Whitney?-Manchester-Storrs-Willimantic-… should be part of any plan, for both passenger AND freight. It will have interesting feeder properties for high speed service.

    Thank you all for your patience with my pedantically tying up these loose ends.

    • Adirondacker12800

      If everybody has express service everybody gets local service.

      When discussing the “New London” station, a key goal is to enable the “shoreline services both toward NH and toward Westerly to have either cross platform transfers or just continue through on the I95 line, a sort of exurban S-bahn.


      People boarding or alighting at New London don’t experience all of the slowness between New Haven and Providence. They only experience the slowness to New Haven or the slowness to Providence.

      Amtrak takes 40 minutes to get to New Haven to New London on Acela. Over an hour on some Regionals. Since they are at rush hour I’m going to assume it’s because SLE is slowing them down. Speed up SLE with electric trains and level boarding 45 minutes on the Amtrak train that diverted off the high speed line in New Haven is good enough. Same thing on the New London Providence side.

      People at the minor Amtrak stations between New York and Philadelphia got over the loss of Amtrak service. Some sort of coordinated limited stop service between Providence and New Haven might be good enough.

      Merely eliminating these will permit enough additional service frequency to very significantly increase demand, probably close to double it.

      If you want the people to get off in New Rochelle and change to Metro North for Grand Central. But then by the time this gets built East Side Access will have opened…. inducing enough demand that you face the same problem…

      It doesn’t matter how one routes high speed service across Connecticut, it is hard to imagine not having well justified high speed service connecting NH and Hartford, and in fact linking NH, Hartford, Bradley Field, and Springfield. A possibly more aggressive alignment through Hartford, may be justified for this service..

      If you link New Haven, Hartford, the airport and Springfield you no longer have high speed service. It takes time to accelerate and decelerate. If you want HSR between Springfield and New Haven you get one stop at most. Otherwise you are slowing it down so much you might as well have gone with the lower speed option that’s much cheaper and gets more service, on local trains, to more people.

      It’s 60 miles from New Haven to Springfield. At an average speed of 90 that takes 40 minutes. Average of 120 it takes 30 and average of 180 which would need tunnel most of the way, it takes 20. How much do we spend to save 20 minutes? It’s even less if the trip is Hartford to New Haven. 40 miles at 90 is 27 minutes. At 120 it’s 20 minutes and at 180 it’s 14 minutes. Is it worth spend the enormous amount of money it would take to make that possible, to save 13 minutes? Once the train gets to New Haven or Springfield it can continue on the high speed line at high speeds.
      Boston, Worcester, Springfield, Albany, Utica ( but maybe not ), Syracuse, Rochester, Buffalo is probably the line that should be built after California and Florida. Maybe Chicago-St Louis. Get to Buffalo and it connects to the line from Chicago to Cleveland cheaply and easily.

      Brewster-Danbury-Waterbury-Bristol-New Britain-Hartford-East Hartford (?Pratt and Whitney?-Manchester-Storrs-Willimantic

      High speed trains don’t stop that frequently. There aren’t enough people out there to be building HSR to them anyway. If you aren’t building HSR taking the bus to the HSR station is faster and since it’s not hauling around 500 seats, more frequent.

      • Ryan

        If everybody has express service everybody gets local service.

        No. If everyone has express service, everyone gets… express service. Cutting a new ROW through the center of Connecticut doesn’t automatically cause a commuter rail line and/or a dozen stations to appear. In places like Storrs, the ROW might be deviated a little so that commuter rail can be attached to the line later. In places like West Warwick, the HSR line is being built because nobody is going to try and take the bike path and spend the money to re-lay the track for podunk little commuter trains, but if HSR already runs there, dropping down six two- or three-car platforms for the six places where a local shuttle train ought to be stopping between the West End and Coventry is a trivial proposition. These things can be taken as one cohesive project and phased accordingly, but more likely is that if the money is there to knock out a full HSR line in one go, that gets done and commuter rail on that line isn’t revisited for another 20 years.

        But these aren’t HSR “local” stops. These wouldn’t even be built with HSR rolling stock in mind. You’re absolutely right – that would be ridiculous, and wouldn’t be high speed. You might even be right in that Hartford – Storrs – Willimantic would never see commuter service no matter what we do. Of course, if you’re right about that, then New London isn’t getting another station and Norwich service is never happening either.

        Two-car trains. Three cars only if gas jumps up another dollar or two and commuter rail goes off like gangbusters as a result. These aren’t projects that get done ever unless HSR was used to drive the initial track work – but they’re not HSR projects, either.

        • Adirondacker12800

          is a trivial proposition.

          Until you want to run two HSR trains an hour over the track. This wouldn’t be build so that two traIns an hour can run between Boston and New York, it’d be built so that there can be four trains an hour between Boston and New York and if that’s being done you can’t have a commuter train toddling along the HSR tracks for very long.

          • Ryan

            Even a complete third track (if you don’t think passing sidings every two miles, coincidentally the same distance on which the stops would be spaced, would be enough) is still much, much cheaper to add to a line that exists already than the alternative, which is 18 miles of brand-new track along a ROW currently occupied by a bike path for something as provincial as Regional Rail in the state of Rhode Island.

          • Adirondacker12800

            The commuter trains between Providence and anywhere are going use something other than existing railroad ROW?

          • Joey

            There might be justification for local service to West Warwick/Coventry but south of the Providence/Kent county line that ROW is unsuitable for anything but local trains.

          • Ryan

            The commuter trains between Providence and anywhere are going use something other than existing railroad ROW?

            They would use the Washington Secondary, currently a bike path, to connect Coventry and West Warwick to each other, to Warwick, and to Providence.

            There might be justification for local service to West Warwick/Coventry but south of the Providence/Kent county line that ROW is unsuitable for anything but local trains.

            The Washington Secondary has an infinitely better approach into Providence than the existing Shore Line / NEC does, only one problematic curve in West Warwick that could be (would be) dealt with, and the only grade crossings that would be particularly challenging to axe happen to be co-located with the aforementioned curve in West Warwick.

            On what grounds, exactly, are you declaring it “unsuitable?”

          • Adirondacker12800

            They would use the Washington Secondary, currently a bike path, to connect Coventry and West Warwick to each other, to Warwick, and to Providence.

            That’s not the HSR line to New Haven or the HSR line to Hartford is it? If it is, you can’t plop a patch of asphalt by the side of the tracks, dump a bus shelter on it and call it a station. Not if you expect to run more than one HSR train an hour.

          • Joey

            There are several curves in the vicinity of West Warwick which would necessitate either massive property takes or massive speed restrictions. Hence the word “unusable” for HSR.

    • Alon Levy

      1. I’m teaching a compressed summer semester. It’s just a culmination of a few things I’d been working on for a while.

      2. I hope it gets more traction outside blogs. Penn Design unfairly passed it over, and Amtrak seems to believe it’s impossible to bypass the Shore Line except inland.

      3. New London is 310 km/h territory, and raising this requires more extensive takings. In general, the biggest travel time gains come from speeding up the slowest segments rather than raising top speed. That’s why I pointed out in the post that 360 km/h only saves 4 minutes versus 300 km/h. 400 km/h saves even less. North of Providence there isn’t enough space between Boston Switch and the Canton Viaduct to get up to 360 anyway; south of New Haven the route is a series of slow zones, compromises, and tradeoffs between saving small amounts of time and not pissing off yet another rich suburb with NIMBYs.

      4. It’s really not possible to make the track heights compatible. HSR has to use a high bridge to clear ship traffic; but once HSR diverts intercity traffic from the movable bridges, there’s no need to replace them, since the capacity the Coast Guard permits is adequate for regional traffic, if the local traffic is kept away from the bridges. The point of a New London stop intersecting the branch line to Norwich is that it is close enough to downtown New London to be a good city station, while passengers can still transfer to a branch line serving a bunch of New London neighborhoods but also Mohegan Sun and Norwich. A Groton stop could connect to a line serving Norwich better but it would not serve Mohegan Sun or Connecticut College.

      5. All bridges are assumed to be high enough to clear ship traffic. Bridge technology has advanced a lot since the 1880s and so has the need for high reliability, and I don’t know of any movable bridges on new-build HSR lines.

      6. Yes! I want to say that variable costs can be kept to about 9-10 cents per passenger-km (15-16 cents per passenger-mile) assuming an unimpressive load factor of 50-60%, and then an average fare of about 14-15 cents per passenger-km (22-24 cents per passenger-mile) will cover expenses and provide enough profits to pay off construction and give Amtrak a large cushion to fund additional lines out of. That’s about in line with TGV practice once you take discounts into account. But about half of that cost is a black box of operating costs that aren’t maintenance or on-board labor or energy. There’s also a hidden assumption of constant frequency throughout the day, which guarantees very high rolling stock utilization; this is partly seen in the assumption of worse-than-Acela/TGV/KTX load factors, but also partly justified based on the distances between the cities, which make the line a combination of long-range commuter service and intercity service, with many different peaks such that it’s worth it to run the same frequency all day.

      7. The NSRL is primarily a commuter project. There’s a huge dropoff between demand for intercity rail south of Boston and demand north of Boston, since north of Boston the only major destinations are Portland and the Nashua-Manchester-Concord corridor. The main advantages of the NSRL are through-running on the MBTA network and a connection from the North Side lines to the CBD, which doesn’t go as far north as North Station. Of course it should be pursued, but Massachusetts is legally required to build it as an environmental mitigation for the extra traffic created by the Big Dig, and the fact that it’s currently breaking the law is not a good reason for the federal government to kick in extra money. At most, federal funding should go to smaller improvements, such as adding more electrification on the South Side lines to encourage the MBTA to obtain high-performance EMUs that can keep up with HSR on the Providence Line without too many overtakes.

      8. My first-filter traffic estimates basically ignore current train traffic and look at traffic on various HSR city pairs in Spain, France, and Japan. See here.

      9. It’s fine to have separate trains providing local and intercity service. Bullet trains should branch out to serve Hartford and Springfield because that’s intercity service and the speed in between is high enough that their better performance at medium and high speed is a real advantage, but there’s no need to drag those trains to Northampton to provide local functions. There are a lot of people in Meriden, Wallingford, etc. who want to travel to Hartford, Springfield, and New Haven without really caring about through-service to New York. So it’s fine to have, eventually, 2 tph of HSR going to Springfield and 2-4 tph of regional service.

      10. The Tokaido Shinkansen has no redundancy at all. While most of the route is a few tens of meters above sea level, some parts are in low-lying coastal areas, in a region with far more tropical cyclonic storms than the Northeastern US. A region whose largest city is in a flood zone, protected only by high latitude reducing the frequency of hurricanes, has bigger storm preparedness priorities than HSR.

      11. Of course. If New York-Montreal service is desired (personally I’m unconvinced) it can also automatically provide Boston-Montreal service without needing another route through the Adirondacks/Northern New England mountains.

      12. I get Bradley (though honestly once the route is in place people are likely to ride to Newark or JFK, possibly with commuter rail connections at the end), but why Lymes? Anyway, if I were a Connecticut power broker, I’d push the Hartford route hard and offer some state funding.

      13. The problem with this is that HSR compresses travel distance in the direction of travel. The same is true of CBD-focused rapid transit. What this means is that with all the other upgrades, Danbury can be served by an awkward alignment through Norwalk and New Haven east; it’s slower than a direct route but not by much, and it doesn’t require new construction. Secondary cities don’t usually merit train service on their own – instead they piggyback on larger cities or on nearby secondary cities, since trains go in a line and can make intermediate stops without losing as much time as planes or highway buses.

  15. Josh

    I thought I had some hare-brained schemes but I stumbled over a whole new level of crazy. Cars are never going to be banned, their use will fade as oil gets scarcer. There will never be a bridge or tunnel between LI and CT. Nor did I read much about air travel, because that’s the HSR demographic. Amtrak and CR are plenty fast for those not driving.

    So I’ll just say this: I think of HSR as a hub and spoke system. And I see it like the Interstates- they didn’t just duplicate existing roads (yes, yes, in places that’s exactly what happened. Please don’t give me examples.)
    I would travel by whatever means to one of a handful of stations. Outside of the termini there is no reason to go downtown anywhere. If I can get it together to get to Logan, I can do the same for a remote station where the Mass Pike, 495, the Boston-Worcester line come together. Bradley and Newark rate stations, in fact I can see Bradley being as much an entry to eastern NE as Logan, PVD and MHT. If your terminus isn’t New York, bypass the city altogether. Build a transfer station on the way to Croton- Harmon, the next at Newark Airport. I suppose I’m saying skip Hartford, New Haven, and Providence. Nor do I see a downtown HSR station as a vital element of urban design. HSR is for long distance travel- urban revival is based on transit. Both may have steel rails, but that’s it.

    • Alon Levy

      HSR is not an air travel substitute. Contra what you say, Amtrak speeds have been enough to take away mode share from air – more people ride rail than fly from New York to Washington or Boston, and air travel in the Northeast lives off of Boston-Washington. It’s highways that are absolutely killing Amtrak. Overall NEC mode shares for intercity travel are 5% rail, 5% air, and 90% highway, of which buses are in the single digits. In contrast, actual HSR can have mode shares higher than 50% in such places as France, South Korea, and Japan. You may be able to get it together to get to Logan, but if the train station is that inconvenient most Bostonians will keep driving to New York. If it’s too hard to get into city center, as is the case in Bakersfield, then it’s fine to have a greenfield station near a secondary city, but in New Haven, Hartford, and Providence, the states of the tracks entering the station and of the sprawl around the cities are all such that it’s much easier to serve the preexisting station than to build bypasses.

      The New York bypass idea looks good only until you look at the topography and the legacy infrastructure. Going across Westchester requires multiple tunnels, and crossing the Hudson is also difficult. If a new Hudson crossing must be built, it should be near Penn Station to allow more commuter rail capacity at the same time. Track sharing near major cities is normal in the standard-gauge world: the LGV Sud-Est used to share the last 30 kilometers into Paris with regional trains.

      • Adirondacker12800

        If a new Hudson crossing must be built, it should be near Penn Station to allow more commuter rail capacity at the same time.

        The 5th and 6th tunnels should be to Wall Street. Shift the commuters out of Penn Station you give the commuters shorter rides to their jobs downtown and free up capacity in Midtown for the intercity trains. The 7th and 8th tunnels – in 2050 or 2060 should be from Linden to New Rochelle. Or Floral Park to Linden to connect to the bridge to New Haven.

  16. Bbbut

    Could you give an estimate on the travel times of a line that actively avoids the city centers?

    For example, looking at the new Chinese HSR lines in Google Earth I see almost none of them going into central stations (except Beijing and Shanghai). We are talking about big 10 million plus cities getting the ‘suburb is close enough’ treatment there.

    So, if going through the central station makes a line drop the speed below 150 km/h, my instinct would be to bypass the ‘problem section’ altogether! But maybe I am wrong and it is not worth the hassle for just a few minutes gained…

    Have you ever done rough estimates on such versions?

  17. Pingback: NEC Future: Moving Sideways | Pedestrian Observations
  18. Pingback: When There’s Nothing Left To Burn, You Have To Set Money On Fire | Pedestrian Observations

Leave a Reply

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

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

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s