Eliminate Penn Station

Note: this is a somewhat trollish proposal, but I do think it should be considered.

New York Penn Station is a mess. Its platforms are infamously narrow, with only enough room for single-direction escalators, leading to overcrowding during peak hours, as passengers scramble to find an up escalator or a staircase. Its two concourses are confusing and cramped, and have claustrophobic low ceilings. Trains’ track assignments are only announced minutes in advance (as at other major US stations), leading to last-minute passenger scrambles to get onto the platforms. Everyone with an opinion, from the city’s architect community to the Regional Plan Association to Amtrak, wants to build an alternative. Let me propose something simpler and cheaper, if uglier: eliminate all above-ground structures, and reduce Penn Station to a hole in the ground.

Most of the preexisting plans for Penn Station do not do anything about the track level. It’s assumed that the tracks will remain narrow, that trains will not run reliably enough for consistent track assignments, and that dwell times will remain high. The architects’ proposals involve a nice station headhouse to make passengers feel important. Amtrak wants to decamp to a nice headhouse at Moynihan Station, again to make its passengers feel important, and add a few extra tracks without fixing the existing ones. The RPA proposal is heavy on redevelopment but says nothing about moving trains in and out more efficiently. Only Penn Design’s proposal says anything about consolidating platforms, in addition to constructing a headhouse, but the need to maintain a pretty headhouse places constraints on the ability to move tracks and platforms.

Eliminating the headhouse moves the focus from making passengers feel important to getting passengers in and out as fast as possible. Most importantly, it means there’s no need for girders and columns all over the track level; they support the buildings above the station, including the headhouse, and would not be needed if the station were a simple open cut. Those girders make it hard to move the tracks and platforms – the only reasonable option if they are kept is to pave over pairs of tracks between platforms to create very wide platforms, which would not be well-aligned with the approach tracks.

In the hole in the ground scenario, the two blocks from 7th to 8th Avenue, from 31st to 33rd Streets, would have no above-ground infrastructure. This requires demolishing Two Penn Plaza and Madison Square Garden. Two Penn Plaza is a building of 140,000 m^2, in a city where the private sector builds office towers of such size for about $750 million (at least when they’re not above active railyards); the city has been making noises about moving Madison Square Garden, although in 2013 it extended its lease by ten years. The tracks and platforms would thus be in the open air, and even from the depth of the platforms, passengers could see the surrounding buildings, just as they can in the open cut west of 9th Avenue, just before trains head into the North River Tunnels.

The two-block compound would be trisected by a pair of wide walkways, as wide as a Manhattan street, parallel to 7th and 8th Avenues. Each of the two walkways would have an access point in each direction toward each platform; with the current narrow platforms this means single-direction escalators, but as tracks would be moved and platforms widened, this would be a pair of wide single-direction escalators flanking a wide staircase. There would be an additional access point heading west out of 7th Avenue and one heading east out of 8th, for a total of six per platform. This is an improvement over the current situation, in which the number of access points ranges from four to six, excluding the LIRR’s West End Concourse, which is west of 8th and thus excluded from this discussion; see diagram here. Penn Station’s tracks are about 14 meters below street level; with 30-degree escalator angles, this means that the escalators would be 24 meters long plus short approaches, say 28 meters total, and this provides adequate separation between access points on the platforms as well as on the two walkways, although unfortunately the spacing on the platform would not be even. For disabled access, elevators would be provided at 7th and 8th Avenues and on both walkways.

The main functions of a train station would be devolved to the surrounding streets and the two walkways. Large clocks, mounted on the high-rise towers next to the station, would show the time. Screens posted over the entire compound would show train departure and arrival times and track assignments. The walkways, and the sides of 7th and 8th Avenues facing the compound, would have ticket-vending machines, selling tickets for all railroads using the station; if the platforms were widened, then there would be room for TVMs and some retail on the platforms themselves. There might even be room for some kiosks on the walkways and food trucks on the streets and avenues. Large ticket offices are not required, and small ones can fit either on the walkways or in a building storefront on the perimeter of the compound.

The technological advances of the last half-century or so have largely made station headhouses obsolete. Train stations used to have telegraph operators; they no longer do. They used to have mail sorting space; mail is now carried by air and road, or electronically. TVMs allow passengers to obtain tickets without buying them at ticket offices, and nowadays e-tickets are making TVMs somewhat obsolete as well. Checked baggage is largely a thing of the past. Transportation companies that aim at low costs, including low-cost airlines and intercity express buses, barely have stations at all: intercity buses pick up at curbs, while low-cost airlines often prefer budget terminals with reduced infrastructure. As far as possible, this is the way forward for train stations as well. Recall that my proposal for a Fulton Street regional rail station followed the same logic, using the street as its mezzanine. This is the way forward for Penn Station, too.


  1. ofsevit

    Only a bit trollish!

    I’ve always thought that thru-running trains (especially for Amtrak) should follow the Spanish Solution and have platforms on both sides of each track, since at peak times for a large train set 500 people can be alighting and boarding the same train. Instead of a train pulling in, its passengers getting off and going up and then passengers massed above coming down, a train would pull in, passengers would get off on one side (platforms would alternate as exits and entrances) and then the others would already be on the platform ready to board. It would get rid of the scrums in the station when a train is called and five hundred people have to squeeze down the escalators. Right now, with the single escalators, it is akin to boarding the entire train through one door, or, put another way, a jumbo jet through a single jetway. It would also mean that dwell times would be reduced from 15-20 minutes to 2-4.

    Is there any other city in the world that has similar service where trains thru run through such a large city between relatively large cities in their own right (Boston, Phila, DC)?

    Almost as trollish: knock down Penn Station to a hole in the ground, rework the trackage as best as you can, and then rebuild the old Penn Station headhouse as close as you can to the original. Maybe do a $5 billion kickstarter for that? I’d kick in $100.

    • Alon Levy

      I used to think the Spanish solution was better. But eventually I figured that it’d be better to have fewer, wider platforms. The reason is that each platform has a minimum width that needs to be clear to allow passengers to circulate on the platform (around 1.5 meters on each side), plus some staircases. This means that if the platforms are wider, then there’s more space for escalators. Even if everything scales, you’d rather have 1 platform with 4 escalators than 2 with 2 each, because you can run a bank of 4 escalators 3-and-1 at the peak, whereas pairs of escalators have to be 1-and-1. So peak capacity from platform to street is higher if you have have fewer, wider platforms.

      Concretely: Penn is about 140 meters wide, which is enough for about 12 tracks and either 6 16-meter island platforms, or 11 8-meter Spanish island platforms and 2 6.5-meter side platforms for the northernmost and southernmost tracks. Subtract 3 meters from each island platform and 1.5 from each side platform; you get 13 meters of usable width for staircases and escalators with regular platforms, and 5 with Spanish platforms. 5 gives you a pair of escalators and a narrow staircase; 13 gives you a bank of six escalators (running four-and-two, say) and a wide staircase, or, more realistically, four escalators running three-and one and a wide staircase, with a lot more extra unobstructed space.

      Under the Spanish solution, you get 13 up escalators per access point (12 effective), i.e. 72 effective total, i.e. 12 per eventual inbound approach track (Empire Connection, North River Tunnels, new Hudson Tunnels, 2*East River Tunnels, Penn-GCT link), which is 72,000 passengers an hour. That’s 30 12-car tph, with the same number of passengers per unit of train length as the rush-hour Lex Express or RER A. It’s very crowded, but conceivable that future regional rail usage from inner suburbs would lead to similar crowding levels; to squeeze more capacity, people would have to use unmentioned access points to the subway, cram into elevators, or climb 4.5 flights of stairs. Under the conventional solution, you get 18 up escalators per access point, i.e. 108,000 passengers an hour, readily expandable to 24 up escalators, i.e. 144,000 passengers an hour, which at 30 12-car tph is a hair more than Tokyo’s single most overcrowded passenger rail line.

      You’re also not getting much benefit out of the Spanish Solution’s use of separate platforms for boarding and alighting, not at a peaky CBD station where pretty much everyone is headed up in the morning and down in the afternoon. There are transfer passengers, but Penn Station isn’t Park Street; O&D passengers will always predominate.

      Regular island platforms also work better with the girders underpinning Madison Square Garden. If you pave over alternating tracks to create Spanish platforms, one side of the train will face the girders; the girders are not placed at intervals in a simple ratio with a single car length, and when I looked a few years ago, about one door of an 8-car New Jersey Transit train faced a girder. If you pave over pairs of tracks to create super-wide platforms, the girders will obstruct the platforms, but the trains will at least be able to open their doors to an unobstructed platform. It’s also friendlier to the current escalator setting: you’re not widening separate platforms, but conjoining pairs of platforms, so one platform in each pair gets the down escalators and the other gets the up escalators, and voilà, paired escalators.

      I thought about reducing Penn to a hole in the ground, reconfiguring, and then building a nice station. That’s probably the easiest sequencing for any rebuild. The problem: the hole in the ground would be a perfectly functional station. It could even be architecturally nice – the walkways could have signature spans (e.g. a series of arches, or a through-arch), which would to some extent be necessary anyway to ensure platforms below could be shifted without messing with bridge supports. Meanwhile, all the functions that Penn hosts that a station doesn’t need, for example various Amtrak back offices, would have to be moved anyway. Once that’s in place, the usefulness of building a stationhouse drops to zero. Literally every ancillary function of a train station – back offices, retail, TOD – is easier to provide in the adjoining blocks.

      Re through-running between other large cities: there aren’t that many first-world cities as large as or larger than Boston, Philadelphia, and Washington. But in the one place where cities of this size class do occur linearly, the Tokaido, we do see through-running in the middle (at Nagoya).

      • Fbfree

        For the Spanish solution, your assuming that the each platform has to have the same width. I had though it would be possible to have platforms of unequal width that switch roles between peak periods. It took me a minute to remember that the platform between the two tracks on the same branch and direction always has to be a departing platform.

        I’d also want to either shave about 1.5 meters off each island platform and sacrifice one escalator, to add an extra bi-directional platform and track to the North Hudson-East River route. I wouldn’t want to plan on platform access being the limiting capacity constraint from day 1. There will be trains, especially intercity trains, that will need a little extra dwell time. Providing a spare platform or two would be useful.

        • Alon Levy

          What do you mean by the North Hudson-East River route? There are two routes that could match the description: the existing Hudson tunnels paired with the southern East River Tunnels, which I call Line 1 here, and the Empire Connection (and thence the Hudson Line) paired with the northern East River Tunnels, which I call Line 3.

          About providing a spare platform, I did think about it, but I think it’s the wrong choice. Using conventional, non-Spanish platforms, there’s room for 6 16-meter platforms, or 8 10-meter platforms. Under the latter option you could dedicate 4 platforms to intercity trains, so each tunnel access track would branch into four platform tracks, whereas for regional trains each access track would branch into two tracks flanking the same platform. The problem: 10-meter platforms have enough room for a wide staircase and two escalators per access point, or a narrow staircase and three escalators running two-and-one per access point; it leads to the same potential platform access capacity limit as the Spanish solution. Making NEC trains and regional trains that use the same access tunnels share the same four platform tracks sounds restrictive, but it means 4-5 minutes between successive trains on the same platform track, and this accommodates dwells in the 2-3 minute range (even 2.5-3.5, if error margins are cut to driverless metro levels). At Shin-Osaka, Shinkansen through-trains dwell for 1 minute.

          • Fbfree

            I did mean line 3: North River tunnels to East River Tunnels. This route will host the largest variety of trains.

            I agree that a pair of tracks and a full width platform would take up too much space. i do wonder if one track and a narrower platform would be useful. Nominally, any consist with a long dwell time (sleeper services, holiday specials, etc…) that would need to use the extra platform does not need the same number of escalators as the main platforms. Then again, these would be scheduled off-peak, when the regional services do not need 4 platform tracks each.

          • Alon Levy

            …I think I explained it poorly – North River Tunnels to East River Tunnels is Line 1; Line 3 stays east of the Hudson, pairing the East River Tunnels (northern tracks) with the Empire Connection and Hudson Line.

            Line 1 indeed would have a variety of trains. The simplest solution to this is “don’t run sleeper service at rush hour.” An even simpler but more contentious solution is “don’t run anything that’s not high-speed rail or regional rail”; in the longer run there’s room for some high-speed sleeper service, but the problem is that in the long run, US urban geography’s not very amenable to high-speed sleepers. Sleeper service is most competitive when it takes 8-10 hours: board at 10 pm, sleep, arrive at 7 am. Miami is 7-8 hours from New York if there’s HSR all the way, but 7 hours is dicey, and it’s a leisure market more than a business market, which means people fly midday rather than try to time themselves for a rush hour arrival. Anything else… there’s New York-Houston, or maybe New York-Dallas, but neither is such a big travel market. Keep the Western sleepers in order to get Interior West Senators to shut up (Empire Builder losses are trivial relative to potential NEC HSR profits), but run daytime trains on the Eastern network rather than unpopular night trains.

          • Adirondacker12800

            They clog up the station and they clog up the system along the line. The land cruise aficionados can go land cruising west of the Mississippi. The people who take the train because they can’t or won’t fly can blend in with the HSR passengers who are travelling shorter segments.

      • Nathanael

        For what it’s worth, the “Moynihan” plan in the Farley building would basically move all of Amtrak’s longer-distance facilities over there: baggage handling, waiting room, ClubAcela, ticketing, and similar stuff. (This stuff isn’t going away. If anything there will be more a-few-a-day trains with longer consists, such as Vermont’s planned second train to Montreal or train to Burlington, or additional trains to Pittsburgh or Chicago, and people will arrive well in advance for them. The Amtrak waiting room right now is overflowing on a regular basis.)

        Having moved all that stuff across the street, you *really* wouldn’t have any ancillary facilities to put in the main part of Penn Station. So at that point, ripping it out as you propose would be perfect.

        • fantomex9

          Even better, build a seperate terminal for the land cruise trains can pick up land cruise passengers without bugging everybody else who take trains to get someplace and not to land cruise.

    • Adirondacker12800

      It would get rid of the scrums in the station

      Use the stairs on the LIRR level.

  2. Aaron M. Renn

    In Chicago, they seem to manage to have tracks posted well in advance of train departures. At Northwestern Station, which I used to commute through regularly, such track posting was basically irrelevant anyway because your train was pretty much on the same track every day. Of course, they had enough capacity to just park the trains and let people board at their leisure. I think there’s a roughly similar setup at GCT, though I’m not aware of whether or not the track assignment is consistent.

    Boston South Station and Penn always kill me where everybody is just standing their looking at the board, the a mad rush all at once to the train. Amtrak enforces a similar queuing mechanism at Chicago Union Station for some reason, even though trains like the Hiawatha are just sitting on the track and passengers could easily board at their leisure.

    • Brendan Dawe

      Hell, they do it at Seattle or Portland where there is at most two passenger trains in the station at any one time. Just annoying Amtrak customs

  3. Pingback: Today’s Headlines | Streetsblog New York City
  4. snogglethorpe

    …using the street as its mezzanine…

    C’mon, just build a lightweight extra layer, using whatever configuration of (roofed-over) bridges/platforms/whatever to function as a real mezzanine, and avoid every passenger using your name as a curse-word…

  5. caelestor

    The air rights are too valuable. I’d expect to see a high-ceiling mezzanine built, with a tall mall / hotel / office building on top.


    “Trains’ track assignments are only announced minutes in advance (as at other major US stations),”
    Not at all. Just a short hop away, the track assignments at Newark are announced ages in advance.
    There’s the occasional short-notice change, but its almost always a track 4 train going to track 3 (same platform).

    In your Streetsblog reference, people are obsessing over shelter. Thats where domes come in

  7. Dan H

    It is quite a practical proposal, there needs to be wider platforms and we have seen this done in China. Narrow platforms kill throughput and to start by taking out a few platforms at a time, rebuild them with more light coming down and build up from there would make sense.

    Even with less tracks, would they be able to fully utilize the capacity of 4 track tunnels to either end of the Hudson would be my only question. If that’s the case, function over form, leave some air rights potential for later.

    • Joey

      Two platform tracks per tunnel track should be plenty to push the tunnels at capacity.

  8. Joey

    So I’ve been looking at maps of Penn Station a bit, and I think that Alon’s proposed configuration – 6 wide island platforms – might be possible without too many structural modifications. This map in particular shows the locations of the support columns (This one suggests that they’re mostly still in the same locations even with Madison Square Gardens on top).

    The interesting part is this – the columns do not constrain the exact positions of the tracks, but rather, there’s a row of columns between each pair of tracks and at the center of each platform. So if you alternately shift pairs of tracks and pave over the others, then add one track each in the unoccupied space under 31st and 33rd, you get 6 wide island platforms serving 12 platform tracks.

    Now, there are a couple of complications to this. Firstly, the rows of columns on the platforms split at the staircases, so those areas would have to be modified (assuming you don’t start with demolishing MSG). Secondly, a few columns in the station throat areas would probably have to be moved. Thirdly, my recollection is that the mezzanine areas have to dip under the subways at 7th and 8th (this also provides access to the subway platforms which are right below grade). But the subway access areas could be provided in the middle of the (amply wide) platforms anyway, and this simplifies mainline-subway transfers too.

  9. apocalypsepony

    Most functions in the station today could be moved or removed, provided the railroads are willing to let go of some of the olde timey railroading attitudes.

    But there are certain back of house functions you can’t dismiss: mostly crew facilities for engineers and conductors between and before shifts. Maybe these don’t /need/ to be located in the station, but it certainly makes staffing easier. And moving will be a union fight…

  10. TheEconomist

    This is certainly a worthwhile proposal if this was a new station and there was nothing around it. How do you propose to keep full service during construction? You MUST keep service during construction — I am not old enough (for the record, I am not young!) to recall the last time a station in the US serving more than 2,000 people per direction per day being completely closed for reconstruction. Demolition/construction in stages does not keep full service and this being the busiest station in the US, full current service needs to be available during construction. Politically reducing the current service for the duration of construction is impossible. Especially given that in the US this will take optimistically 5 years, realistically double that much. That is why everything being discussed by the “decision makers” now involves addition of infrastructure first before modifying existing infrastructure if such modification is contemplated at all. Before anyone can mess with the current setup, new capacity is needed, so that it replaces the capacity taken out of service while the reconstruction takes place.

    What do people think is the best (and cheapest) additional capacity that can be added before this reconstruction plan can proceed?

    • Nathanael

      (1) Move the Amtrak waiting room and similar facilities to Farley. Farley’s historic and can’t be knocked down, so we’ll have to work around it in any reconfiguration plan anyway. This removes the most space-occupying operations in the station while moving them into a currently vacant building.
      (2) Demo MSG, which causes nothing but trouble.
      (3) Work south to north, or north to south, on the track/platform renovations.

  11. Pingback: Penn Station Elimination Followup | Pedestrian Observations
  12. oinonio

    I actually like this idea. Efficient. Great views in all directions, and demolition of 1 Penn Plaza and that eyesore MSG. What’s not to love? Several stations around the world are practically in the open air anyways.

  13. SomeGuy32

    ” the city has been making noises about moving Madison Square Garden, although in 2013 it extended its lease by ten years.”

    Why are we still perpetuating this myth that the City owns MSG? There is no lease. MSG owns MSG. Here’s their most recent property tax bill:
    (yes, they are tax exempt because some lawyer screwed up back in the 80s when Gulf+Western asked for a tax break to renovate – the tax break was basically supposed to last only 10 years but it wasn’t worded properly.)
    (look up block 781, lot 9001 on ACRIS – you’ll find the deeds going from the Penn Railroad to MSG)

    The City controls the occupancy zoning regulations (and they’ll lose in court if they try to not renew the 10 year variance – if they’re going to grant a variance a few blocks away in what is currently a residential zone (and the fact that the City approved the $1B renovations in the first place – one would reasonably assume they were meant to last more than 10 years) then they have nothing to work with)

    And we’re ignoring the true roadblock. Amtrak owns Penn – not the City…. who’s paying to buy out and move MSG? (I have no confidence in the City, the States (NJ and the MTA) and Feds working together – they can’t even get the current Penn Station mezzanine space running efficiently).

    This “moving MSG” fantasy needs to end and we need to focus on more fiscally responsible ideas.

    • Spendmor Wastemor

      Where there’s enough will, there’s a way. MSG’s owners are looking only at the numbers, they don’t give 2 hoots about the building as long as their equity and income pencil out.
      Penn is a sewer and needs to be fixed at some point. If I had to drag through that rathole twice a day, 210 days per year I’d consider moving. I’m sure it’s one of the straws that chases some number of talented people out of the area once they have a choice of locale.

      • SomeGuy32

        “Where there’s enough will, there’s a way. MSG’s owners are looking only at the numbers, they don’t give 2 hoots about the building as long as their equity and income pencil out.”

        Sure… I repeat – who’s paying for it?

        • Ryan

          The city has a prevailing interest in seeing Madison Square Garden razed to the ground, and could initiate eviction proceedings any time it wanted to – if not under a zoning variance then through eminent domain. Sure, it’s going to be an exceedingly messy decades long process and that’s before they said “you’ve got 10 more years and THEN we’re going to kick you out” – it’s the city’s land. Doesn’t matter who owns the buildings above it or under it, the city has the prevailing interest as far as the dirt itself is concerned.

        • Spendmor Wastemor

          Taxpayers and riders, obviously.

          A big chunk of what we’re paying for now is self-dealing and vote-buying by New Tammany, and redistribution schemes targeted to burden those struggling on the way up to favor those scamming the system. The excess costs for, say, the new Path station and 2nd ave subway are probably enough to cover most of this project. Add the spend from the mysterious cost multiple for building in NYC as opposed to anywhere else and there’s enough mystery money (eg theft and union appeasement) to do the whole project.
          A decent Penn Station, whether this proposal or one with a structure on top, is good for 100+ years. Once the bonds are paid, the asset stays and keeps producing a functional dividend. Think long, not short.

        • Alon Levy

          If MSG has legal grounds to stay if it wants to, the city has an eminent domain case against properties right next to Penn Station (e.g. in the block immediately to the south): “it’s in the public interest to have a train station with wider platforms and more access capacity.” I’d prefer it if the TOD next to the station were commercial high-rises rather than sports stadiums, but the best candidates for TOD next to such a station are things that generate very high traffic density at the peak, since rail transit serves those trips better than any other mode. A stadium might be a correct use of TOD in this context.

          Again, all this assumes the city can’t just choose to tell the MSG Company to pound sand and not renew the variance.

          • Doctor Memory

            If there has been one consistent, predictable feature of multi-use sports/entertainment facilities in contemporary american cities, it is this: “free” has never been good enough for the owners, and they will sooner chew off their own fingers than absorb their own capital improvement costs. Sooner or later, and my bet is on the near side of 2027, Dolan is going to look at his aging skyboxes and dwindling revenues vs Barclays Center, and start reaching his hand in the direction of the city’s coffers. It won’t be necessary to tell him to go away, merely to tell him that no assistance will be forthcoming. He’ll threaten to decamp to New Jersey; our job will be to hold firm and say Bon Voyage.

          • Alex B.

            The city can indeed sieze MSG via eminent domain, but they’ll pay through the nose to do so. They own that particular piece of real estate.

            There’s a reason the old plans involved MSG moving voluntarily – MSG was aging and their owners were interested in a new arena. But the talks dragged on, MSG decided to renovate the existing arena with their own money. Now they have very little incentive to move voluntarily; the city’s cost to buy them out just went up.

          • SomeGuy32

            Again…. who’s paying?

            Feds don’t seem interested – and they’re the ones who own Penn Station.

            The City doesn’t even control any of the rail agencies that operate in Penn Station.

            So who’s paying to move MSG? Who’s paying for the new station?

          • Ryan

            Again, it doesn’t matter that the city has no control over the station or the agencies that operate within it. They control the land upon which the stadium squats and beneath which the trains run. They can, at any time, initiate eviction proceedings – without having another stadium location lined up, without having a new train station lined up. They don’t have to pay for MSG 2.0, they don’t have to pay for the new station, and they don’t need the support of the states or the feds to do it.

            The only thing the city has to pay for is the expensive, messy, and lengthy carnival of lawyers needed to see this thing through. It’ll take decades, during which the new stadium and the new train station will both work themselves out.

          • Nathanael

            The city can throw out MSG by eminent domain, and pay to do so. The city can’t throw Amtrak out, but I expect Amtrak would cooperate because Amtrak *always* cooperates with city station projects.

    • fantomex9

      People need to stop making sports arenas sacred places that can’t be torn down, ever; if there needs to be a new sports arena in New York as big as MSG, let the sports agencies/companies raise the money themselves and build a new stadium the same size as MSG else place in New York.

  14. Spendmor Wastemor

    Penn station is a hole in the ground, a dark, filthy, overcrowded hole. it’s telling that even a troll-ish proposal is vastly better than the Penn Station which exists today.

    Serious proposal: Open air terminal as above, and sink pilings to allow the air rights to be used later. Require whatever structure above the station to have about sixty feet of clearance above ground and some carved-out shape so as not to simple repeat the current Penn misery. That kills most of the overhead light, but it allows enough space to design something both workable and pleasant.

    I disagree on the unimportance of an elegant terminal station. Letting people feel good, or at least less bad, for a few minutes times four hundred trips per year has economic value. Check the property values for residences with ocean views but having no dock or beach access. The ocean view is thus “useless”, yet the market signals its value.

  15. Evans

    Caltrain SF terminal need samething. Too many (12) tracks while running only 5 train/h,

    • Richard Mlynarik

      Caltrain SF is getting a new terminal that is everything wrong with NY Penn.

      Massive, completely unnecessary construction over buried train tracks.

      Totally unnecessary underground mezzanine level above tracks, blocking all ready access and light from the street level.

      Totally unnecessary above ground mezzanine level, serving less than no purpose.

      Criminally stupid and unnecessary four-levels-up massive heavy park level, driving incredible structural demands and killing the entire transportation utility of the edifice.

      Hugely operationally inefficient for train movements, with few non-conflicting moves possible; spatially extended, mis-configured and slow approach tracks and interlocking.

      Structural columns placed exactly where they will most impede passenger circulation. Totally unnecessary columns! Large diameter columns. Everywhere. Everywhere.

      Inadequate platform open space, nearly all space filled with structural columns (due to unnecessary structures above); fire escape stairwells (due to unnecessary and risk-increasing buried-warren levels-deep platforms) and series of insane single bank single flow narrow escalators.

      Hugely inadequate vertical circulation (escalators, stairs) from the platform levels, due to (1) the unnecessary structural columns (2) the “designers” and “architects” of this mess simply not giving a fuck. The columns are so wide that there is no room for even pairs of (contra-direction) escalators on the platforms, let alone the universally employed solution a pair of wide escalators flanking a wide stairway. The columns aren’t even positioned to align with the centres of the platforms, so the narrow single escalators are offset towards the platform edges.

      This is what they’re building. On a brownfield site with constraints not of their own design. No joke:

      Contrast with https://www.flickr.com/photos/28888420@N07/3277880381/ or with any photo of any functional passenger transportation facility.

      Oh, and the insane edifice above the (permanently sub-sub-functional) train level has gone so wildly over budget ($2.6 billion for a small aerial bus station, and increasing by the day, that the train station has been completely defunded: no tracks, no power, no systems, no connecting tunnels.

      It is everything bad that you can possibly imagine, and much than you would never have imagined in your worst fantasies about the limitless incompetence and total corruption of America’s Finest Transportation Planning Professionals.


      Simply unbelievable.

  16. Pingback: Little Things That Matter: Vertical Circulation | Pedestrian Observations
  17. Pingback: Penn Station Tracks | Pedestrian Observations

Leave a Reply to Spendmor Wastemor Cancel reply

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

WordPress.com Logo

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

Facebook photo

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

Connecting to %s

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