15 June 2009

The Peninsula Corridor Investment Strategy

The Metropolitan Transportation Commission (MTC) recently published a Peninsula Corridor Investment Strategy, shepherding together various competing interests (CHSRA, TJPA, Caltrain, SFCTA, San Jose, VTA) to ensure that the Bay Area presents a unified front in its pursuit of a $1.8 billion slice of the $8.0 billion of ARRA stimulus funding available for high speed rail.

The MTC report contains a few surprises. But first, here are the projects contained in the $1.8 billion peninsula funding request. Also refer to pie chart at right.
  • $400M for the San Francisco Transbay Transit Center "train box"
  • $516M for Caltrain electrification
  • $230M for Positive Train Control (automatic signal enforcement) for the peninsula (*)
  • $212M for Caltrain's San Bruno grade separation project
  • $205M for a newly revealed TTC train box "Rail Platform Extension" (*)
  • $149M for reconfiguration / HSR modernization of Caltrain's Diridon station in SJ (*)
  • $98M for reconfiguration / HSR modernization of Caltrain's 4th & King station in SF (*)
  • $52M for DTX (SF downtown rail extension) design (*)
Projects with an asterisk (*) are funded nearly entirely out of the stimulus. Those without an asterisk have additional sources of funding already allocated.

A Two-Phase Approach

ARRA stimulus funds are targeted for so-called "shovel-ready" projects that can create short term jobs. That puts the California High Speed Rail Authority's San Francisco to San Jose HST project out of the running; the high speed rail project on the peninsula is in the early stages of an environmental clearance process that promises to be highly contentious, to be completed in 2011 at the earliest. While design work is already well underway, the controversial quadruple tracking, the numerous grade separations, and the wholesale reconstruction of nearly every Caltrain station planned by the CHSRA are not even close to shovel-ready.

That evidently pushed the MTC to split the construction of high speed rail on the peninsula into two phases. The first phase (as listed above) consists of a patchwork of projects that were already in the planning / environmental clearance pipeline before Proposition 1A was passed and HSR became a realistic prospect. It's definitely a pragmatic approach, with some caveats discussed below.

A Sudden Caltrain Bounty

Caltrain, with its appointed three-county board of directors, has historically held very little clout in getting its major capital improvements prioritized by the MTC. As recently as a few months ago, the MTC re-allocated $91M of Dumbarton rail (a.k.a. Caltrain) funding over to BART. Caltrain's biggest capital improvement project, planned for decades but with an ever-receding construction start date, is the electrification of the peninsula corridor with 25 kV overhead wires to provide quicker and more modern service. (artist's concept at left).

In the Peninsula Corridor Investment Strategy, Caltrain projects are suddenly front and center, thanks to their relatively advanced state of shovel-readiness and some new-found political support. This is a welcome change from the usual BART ΓΌber alles funding strategy.

Caltrain electrification was always considered an unwelcome fly in the BART-around-the-Bay ointment. Now that billions of HSR funding have materialized, the Bay Area transportation-industrial complex is seemingly more attracted to HSR than to BART, if that is even possible. This changes everything: far from being an impediment to BART, electrification is now a "foot in the door" for bringing high speed trains into San Francisco and undertaking some very juicy construction projects during Phase 2--including several billion dollars of grade separations.

For thousands of Caltrain commuters, it should be welcome news that electrification has gone from problem to solution. Nevertheless, one question comes immediately to mind...

Is The Cart Before The Horse?

Does it make sense to electrify Caltrain in Phase 1, only to rebuild the corridor once the HSR project is environmentally approved and construction begins for Phase 2?

  • Commuter improvements are not beholden to HSR Phase 2 schedule, which may be delayed pending environmental review and inevitable litigation
  • Electifying in Phase 1 conveniently coincides with the timing of Caltrain's unavoidable need to replace their obsolete train fleet
  • Electrical substations and other lineside equipment can remain as-is for Phase 2 (Caltrain's planned 25 kV system is world-standard and fully compatible with HSR)
  • The overhead contact system (basically anything above the rails) is made of modular components, easily reused or reconfigured
  • Electrifying in Phase 1 allows testing / commissioning and possibly even limited high speed train service prior to Phase 2 completion
  • Building temporary "shoo-fly" detour tracks during grade separation construction becomes marginally more complicated
  • Since vertical and horizontal alignment of tracks will be altered in many places, pole footings will need to be reconfigured. This is probably just a footnote on the demolition orders, considering that nearly all existing Caltrain station platforms and most existing grade separations will be entirely rebuilt in Phase 2.
Not that any of this matters, since the agenda is more political than technical--but thankfully, the pros do seem to outweigh the cons on a purely technical basis.

Bang for Buck: High

Go-Go Gadget Train Box!

A few months ago, the Transbay Joint Powers Authority (the agency charged with building the new Transbay Transit Center) got into a tiff with the CHSRA over the adequacy of their train platform layout. Entrenched in conflicting positions, the TJPA designed a 150-meter radius curve (extremely and impractically sharp) into one end of their platforms, while the CHSRA adamantly insisted on perfectly straight 400-meter platforms. We didn't hear any more about this little mind-the-gap issue, until now.

The Peninsula Corridor Investment Strategy asks $205 million for a "Rail Platform Extension" which would extend the train box north-eastward from Beale St. to Main St., as shown in the figure at right, to enable the sacrosanct straight platforms.

Sadly, it's another design by committee. The CHSRA's straight platforms are not a valid technical requirement. The underlying requirement concerns the admissible gap between the platform and a train's doors... and any reasonable value of that gap will still allow for some amount of curvature, however slight. For example, a 1 km radius will produce an additional gap of 5 cm (2 inches) at most. As if proof were necessary, high speed trains stop at curved platforms in nearly every county that has them.

Extending the train box has enormous potential for the Transbay Transit Center, for other reasons than making the platforms perfectly straight. That potential would be fully exploited by a proper design of the train box, to be discussed another time. And why does a 250-foot extension cost another 50% more than the basic train box?

Bang for Buck: Medium (for reasons unrelated to platform curvature)

San Bruno Grade Separation

A worthy project provided they do it right and straighten the darn curve.

Bang for Buck:
Zero (if built as Caltrain designed it)
Medium (if they straighten the curve)
High (if they straighten the curve and build an island platform)

Fourth and King Terminal

Caltrain had unfunded plans to modernize and expand the 4th & King terminal in San Francisco, as part of their Project 2025 plan. They even have a track layout in mind. For this, a design totally removed from system-level optimization of San Francisco rail operations, they are asking $98 million.

If somebody cared enough to configure the Transbay Terminal correctly, perhaps the 4th & King station wouldn't need to make up for Transbay's glaring lack of capacity.

Bang for Buck: Low (if only they fixed Transbay instead!)

Positive Train Control

The Investment Strategy calls for fully $230 million (!) to implement positive train control on the peninsula corridor. While PTC would finally bring the corridor out of the 19th century, why is Caltrain in the driver's seat to define the system for all of California? And why are they re-inventing the wheel by "defining the requirements" for a "new signal system" when this technology is a solved problem, sold as a commodity off-the-shelf item by countries that already have modern railways? Buy a fully debugged ETCS Level 2 system from your choice of several established vendors, and be done with it already! This is sure to become another disastrous R&D project.

Bang for Buck: Low

Diridon Pan-Galactic

San Jose's naked ambition to become the Grand Central of the West is on full display, with a $150 million grab to build a new station whose grandiose architectural configuration is rooted more firmly in a civic inferiority complex than any realistic transportation need.

Bang for Buck: Very Low

In closing, can we really expect $1.8 billion dollars (more than 20% of the nationwide high speed rail stimulus) to land in the Bay Area? Probably not. Whatever the Bay Area may score in Washington, the Peninsula Corridor Investment Strategy reveals a great deal about the region's evolving funding priorities.


  1. Looks like they're too eager to spend money before getting all of their ducks in a row. Better wait till 2011 and get the terminal design and the plan for the whole corridor straightened out.

  2. Even better perhaps we should be considering advanced rail technology from a interstate rather than an intrastate perspective. I have been reading up on ALT, a group that proposes a comprehensive national rail system along the interstate that combines energy technology to keep it running. I think they may have something.

  3. And why are they re-inventing the wheel by "defining the requirements" for a "new signal system" when this technology is a solved problem, sold as a commodity off-the-shelf item by countries that already have modern railways

    Look at Muni: they didn't "define their requirements" properly, and now they have a signal system that doesn't really match their operating realities very well. Trains have to stop before entering the tunnel. The system doesn't support double-berthing so the huge stations on Market St are mostly wasted space. And ETCS Level 2 is still in the R&D phase, although getting close to being finally debugged. ETCS Level 1 is already debugged though, and is pretty much a commodity but doesn't really meet Caltrain's requirements in terms of improving performance (though it would make an excellent replacement for Metrolink's ATS).

  4. I'm looking at it and thinking ambit claim. Apply for more than you think you can get, to frame getting what you thought you were going to get as "partial funding" so people don't complain about, say, 10% of the national HSR funding ending up in the Bay.

  5. I give up. "Arcady" has just got to be on the payroll of some defense contractor scamming for a juicy pork contract to "design" a train control system for "Unique American Conditions".

    "ETCS Level 1 is already debugged though, and is pretty much a commodity but doesn't really meet Caltrain's requirements in terms of improving performance."

    Where on earth do you get this stuff from? Switzerland is planning to deploy ETCS-LS (a small delta about ETCS-1) nation wide on the most intensively used rail network on the planet. Today's Swiss network would collapse immediately if ETCS-2 didn't function day and night on the most critical parts of the network. (Hundreds of trains a day, 200kmh, 2 minute headways.) But no: Caltrain has unique performance needs. Maybe furrin signal systems can't deal with trains that run infrequently, slowly, late and accelerate like molasses?

    Anyway, you'll get your wish. Caltrain's immensely qualified in-house team of global signalling experts are designing their own unique system. Somebody will make a fortune out of failing to deliver it. And we'll be screwed.

  6. I love it. This completely and utterly blows the CHSRA's negotiating power out of the water. Caltrain gets all spruced up without HSR - imagine that. hahahahahahahahaha

    And Impact analysis for the project level EIRS? In the next round CHSR will ahve to compare itself to the new and improved version of Caltrain - in other words; NO incremental benefits derived by installing HSR, only net incremental negatives for environemtnal impacts. Love it. Love it Love it. Lets see how they try to worm their way out of this one.

  7. Caltrain seems to already know what it wants to do here to I guess that's why chsr is pushing to to do the la ana and sj sf portions first so they can get in and work with them. Caltrains seems to be moving ahead with their plans.

  8. Richard: first off, there's a difference between "requirements" and "design". "Requirements" is figuring out what you need in the first place, whereas "design" consists of writing a specification for a product to meet those requirements. Who knows, maybe some form of ETCS will meet those requirements, or maybe the requirements of ETCS as set by bureaucrats in Brussels are not ideal for the requirements of a commuter rail system in California. It would also be a shame if Caltrain installed ETCS Level 2 now, before most of the rest of Europe has, and the standard were revised leaving Caltrain with an incompatible and unsopported legacy installation.

    As for Switzerland, I believe on the approach to Zurich Hauptbahnhof they don't have any kind of ETCS at all, only good old fashioned lineside signals supplemented by an intermittent inductive automatic train stop system (a more useful version of Metrolink's ATS). And the busy section into Zurich Hbf has four tracks for the 20 tph, while the line from Kearny to New York Penn has 24 tph on only two tracks, and the peak-direction flows on the LIRR and Metro North are even greater at the terminals. Switzerland's rail network is a model of precision in its operations, but I'm not sure it's a great model even for Caltrain given that the biggest city (Zurich) has a population of 300,000.

    And really the big concern with ETCS is the GSM-R portion of it, specifically whether the FCC will allocate spectrum for it, and whether it will be the same spectrum allocation as in Europe. In case you haven't noticed, plain old GSM is not compatible between the US and Europe because of the difference in spectrum allocation, and you need tri-band or quad band phones to deal with that.

    And I'd like to refer you back to the Muni example. They bought a "proven" system (Seltrac), "off the shelf", from a french company, which had a pretty good track record up to that point. And it proved to be a disaster, though by now its status is more of "annoyance". Because the system is inflexible, and its design makes assumptions that are simply not true in Muni's case.

  9. whether it will be the same spectrum allocation as in Europe.... you need tri-band or quad band phones to deal with that

    California's trains, of whatever variety, aren't going to be operating all that much in Italy. So whether or not they use the same slice of RF spectrum is a minor problem. I see quad band phones on Ebay for well under $100 even if they were to install all the possible frequencies used all over the world it's going to be a minor expense.

  10. Except GSM is a very mass-produced technology, even in the US variant. GSM-R is not, and US-Band GSM-R is going to be a niche product, on the order of double digits of train radios and single digits of base stations, which will probably have to have entirely custom parts in the RF interface for the incompatible frequency band. And of course even getting GSM-R going would require dealing getting appropriate regulations from the FCC, which is a whole other regulatory risk on top of the huge one of the FRA waiver.

  11. Clem, I'd argue that Diridon bang for buck is actually hugely negative - nothing but a liability when the nation looks to california for accountability on how federal stimulus was spend (several years from now), Grand Diridon Station will make the whole thing look like a huge porky boondogle. Especially when there will be actually zero HSR to show for these funds at the end of the day. Funding Diridon with HSR Federal Stimulus funding risks shutting down CHSR completely.

  12. Anonymous makes a good point about the "zero HSR to show for it" actually. Because even if HSRA builds everything they want to on the SF-SJ and LA-Anaheim segments, there will still be exactly zero miles of HSR in California. And even if the Central Valley get their test track, there will still not be HSR service to any of California's biggest cities.

  13. If there both ends and the CV track are in place, then if there is a cost blow-out in one of the tunnel projects, which is a risk intrinsic to any major tunnel project, they can still start services operating across the other one and start issuing revenue bonds to cover the cost blow-out.

    However, in any event, speeding up the Caltrain will lead to increased service frequency, and so to increased delays at level crossing, and so to increased demands for grade separations. And anti-HSR corridor NIMBY's faced with trying to make the case in favor of traffic congestion, the issue that they are at present ducking by recourse to scare stories.

  14. Is there funding for new trainsets? Otherwise, it defeats the purpose of funding electrification and signaling.

    And please don't tell me they will just replace the locomotives with electrics. The 2025 plan calls for EMUs.

  15. Clem,

    The hell with San Jose being the 10th largest city in the U.S. and the largest city in the Bay Area! You obviously have a huge bias against the great city of the South Bay; "Diridon build out isn't needed, San Jose has an inferiority complex, blah blah blah." Simple question for you bro...why the bias?

    Be bias all you want. I'm glad to see we'll be getting our due with HSR and a grand "St. Pancras International" West at Diridon. Even Anon 18:20 will learn to enjoy the new Diridon station with HSR/Caltrain/BART.

  16. Municipal boundaries are pretty arbitrary, so it really doesn't make sense to compare populations of cities to determine which ones are "significant". San Jose managed to get lots of annexations while it was growing, San Francisco almost none. Despite the larger population, San Jose is still mostly a vast low-density sprawling suburb with a fairly small downtown core. It is, however, a major node in the rail network, and its station does need a bit of improvement. But I think the current Caltrain planto build two more island platforms and reconfigure the approach from the north is enough for the foreseeable future, even with HSR. There's also the minor issue of Rod Diridon using our money to build a monument to himself, which is why I now make a point of calling it Cahill Street Station.

  17. Bruce McF - you just made a case for putting Caltrain grade separations in the funding request - but none is there - instead Diridon Grand central. Why? Wouldn't grade separations for Caltrain between SF and SJ be the higher priority once electrication is assumed funded? All the more reason that Diridon Grand Central doesn't make sense to be on this list. It appears to be robbing from higher priorty grade separation projects. It just looks like a good ol' boy favor to Mr. Diridon

  18. @Arcady, perhaps I sounded too flippant about PTC. I do realize implementation / integration is never easy, even with standard technologies. But we shouldn't be in the business of inventing a new made-in-USA standard unless the operating environment is completely different in the USA-- and it isn't, for these high speed passenger trains! ERTMS / ETCS was developed to unify a patchwork of ancient signal systems, which is exactly what we have today. Substitute "operator XYZ" for "nation XYZ" of the European model. A gradual overlay capability leading to wholesale replacement is exactly what we need, and you can buy the equipment (and the integration expertise and experience) right now. Equipment suppliers know the standard is evolving, and a lot of that will happen through firmware upgrades-- not outright obsolescence.

    Re: GSM-R, I understand that Amtrak is already playing with it and that some freight operators are interested. Frankly, overcoming whatever FCC spectrum allocation issues there may be is trivial compared to the cost of a not-invented-here attitude.

    @Tony D, I'm not slagging San Jose, only their megalomaniac station plans. They will need two, at most four tracks for HSR (by the very nature of being an intermediate stop for every single train, with ~3 minute dwells). Caltrain needs maybe another three or four, which can be shared with the old-tyme FRA iron. The bottom line is that a massive 14-track double-decker station with complex flying approaches is overkill, serves no pressing transportation need, and is a massive waste of taxpayer money. It's too much station for not enough train traffic. Caltrain's south terminal plan (9 platform tracks), or some variation thereof, will be an amply adequate build-out even in 2030. They're already building this as we speak.

    As for bias, tough noogies, it's my blog!! ;-)

  19. they can still start services operating across the other one and start issuing revenue bonds to cover the cost blow-out.

    There's not going to be enough HSR demand within the Bay Area and within the LA Basin to generate significant revenue. SF to Gilroy, or LA to Bakersfield, maybe, but not SF to SJ or LA to Anaheim.

  20. single digits of base stations

    The only way they'd be able to get away with 10 base stations is if the base stations looked - and used power - like TV stations or FM stations. The thing the conductor carries would have to be tethered to a similarly sized transmitter on board the train. They'd have to site the base stations so that trains are always in line of sight. None of it would work in tunnels. There's gonna be a few tunnels in the system.

    From Wikipedia: GSM-R is typically implemented using dedicated base station towers close to the railway. The distance between the base stations is 3-4 km.....In Germany, Italy and France the GSM-R network has between 3000 and 4000 base stations.

    The initial system, SF to Anaheim is 465 miles or 744 kilometers long. Even at 5 kilometer spacing there are going to be 150-ish base stations. I suspect quite a few more since the train has to be able to "see" at least one base station all the time. If it works like cell phones, for the network to make decisions about which tower to switch to next at least two towers have to be "visible"

    which will probably have to have entirely custom parts in the RF interface for the incompatible frequency band

    I did some rummaging around. I suspect all it would be is a software change in the parts of the code that relate channel numbers to a frequency. Something they would have to test out in the Central Valley on conventional trains or even trucks on Route 99.

    Reading between the lines in some of the vendor's sale brochures they are doing just that. Changing the software on standard GSM phones and "modems". Maybe changing the antenna length to get a better gain.

  21. Interesting comments about signaling systems. I'd like to add that Hitachi also has a proven and reliable system (digital ATP) that is used on the Shinkansen system as well as the Keihin Tohoku commuter line in Tokyo, with its 2 minute headways and 26 trains/hour.
    Hitachi digital ATP

  22. Alon Levy said...
    "There's not going to be enough HSR demand within the Bay Area and within the LA Basin to generate significant revenue. SF to Gilroy, or LA to Bakersfield, maybe, but not SF to SJ or LA to Anaheim."

    If its the southern tunnel with a cost blowout, it would be Bakersfield, Fresno, Gilroy to SJ/SF, if its the northern its Fresno, Bakersfield, Palmdale. Plenty of observations to get SF/SJ / LA Basin ridership projections for revenue bonds.

    Anony-mouse: "Bruce McF - you just made a case for putting Caltrain grade separations in the funding request - but none is there - instead Diridon Grand central. Why? Wouldn't grade separations for Caltrain between SF and SJ be the higher priority once electrication is assumed funded?"

    The argument doesn't say anything about the sequence in applying for Federal Funding, since both can be completed well before the tunneling is done. If the new station is part of supporting a higher frequency of Caltrain services when the electrification is done, which will increase trip speed, that would in turn increase the Benefit Cost Ratios of the grade separations ... the more train movements over a crossing, the stronger the benefit of the separation ... so its a plausible funding sequence. Whether its the perfect funding sequence ... I don't see any reason to worry about that.

  23. What if there's a cost overrun in the LA-Anaheim and SF-SJ sections? Construction in built up areas can be almost as problematic as in the mountains, given that you have to be much more careful about not damaging surrounding structures, poorly documented utilities, and so on. And if they're building the two tunnels simultaneously and after the SF-SJ and LA-Anaheim sections are done, what happens if they run out of money then? The likely construction priorities given what politicians have been saying is SF-SJ, LA-Anaheim, Bakersfield-Merced, with the two actually important bits being left for last. Why not start with the most important bits first instead?

    Clem: I think you continue to misunderstand the difference between "defining the requirements" and "defining a standard". Defining the requirements is simply figuring out what any potential system will need to do, regardless of whether that system will be built to an existing standard or not. In Caltrain's case, I think they don't like the idea of a fixed block system, which ETCS L2 is. The other benefit is GSM-R, but Caltrain (and most US railroads) mostly have a working radio system, whereas railroads in Europe sometimes do not.

  24. I think they don't like the idea of a fixed block system, which ETCS L2 is

    Do they need a moving block system when they start? ... has anyone built an HSR one yet? Has anyone built one of any sort?

  25. Re Diridon "not yet dead" Memorial Intergalactic and Clem's comment: "They will need two, at most four tracks for HSR (by the very nature of being an intermediate stop for every single train, with ~3 minute dwells).

    Correct. Well, at most 2 platform tracks for HSR (we're only talking 4tph in real, non-fraud-promoting life), and remembering that all trains are going to stop at San Jose Capital of the Entire Universe (an obvious fact nobody mentions is that speeds through the curves approaching the station are so low they might as well be stopped anyway), and that HS trains aren't going to be overtaking each other here (or anywhere along the Caltrain line.)

    But we also need to deal with a smattering of Caltrain to Gilroy or wherever and Caltrain turnbacks (aka deadheads) to Tamien, which might well get overtaken here. So your 4 through platforms tracks is plenty for more than "HSR".

    "Caltrain needs maybe another three or four, which can be shared with the old-tyme FRA iron.".

    Arggggggghhhhhhhhh!!!! Noooooooooooooooooooooo! Keep the FRA steam train horrow completely segregated from the post-19th century (Caltrain/HSR) operations or we're utterly doomed.

    Fortunately, that it's easy to do. In fact, it's hard not to do.

    The steam trains (two through tracks, 8" platforms or whatever crap Amtrak and UPRR want) stay on the eastern side of the ROW all the way from the "Coast" junction in Santa Clara down through San Jose and onto Gilroy or wherever. Two tracks from Santa Clara to SJ Cahill, one track south of there.

    The UIC-or-whatever non steam trains want at least four platform tracks -- two for terminating trains, two for through trains. We assume all trains are stopping or low speed so it's hard to justify non-platform express bypass tracks.

    Four platform tracks is a minimum, and perhaps only works with some tail track action. But fortunately, we have automatically already have that, in the form of trains that continue to Tamien (say 25% of them) and Gilroy (maybe 12% -- one TPH.) Given 8tph Caltrain at full system build-out, terminating 4 to 6tph of them on a pair of platform tracks is operationally trivial.

    Five platform tracks works better, and the physical reality is that really means six tracks (three island platforms). It's a bit of a squeeze fitting these in together with the steam trains, but it can be done. I've done the CAD track layout work to prove it.

    The biggest problem at Cahill Street is the FRA Vasona Line branch that sends coal (no, I'm not being hyperbolic!) to a cement plant. In a rational world, we'd just bag it. In our crazy legislative world, we almost certainly have to make space for it. In a big squeeze, this is just possible, using an expensive flyover to the north so that the coal trains never touch the UIC tracks, and putting a single steam train line on the east side of the station. Outrageously uneconomic, socially unjustifiable, but probably unavoidable.

    [... continued ...]

  26. [... continued ...]

    So Cahill Street Station ends up [west to east] looking like
    * VTA Toonerville Trolley
    * VTA Toonerville Trolley
    * Coal train branch track (no joke!)
    * UIC (used a generic for "non-FRA" ...
    * UIC ... no anti-Japanese regulation bias implied.)
    * UIC
    * UIC
    * UIC
    * UIC
    * FRA
    * FRA

    North of the station the coal trains flyover and we end up with
    * [CEMOF Hell, plus potential HSR yard, which should of course have gone in at Newhall Yard where the Guardino Memorial BART megaporkplex is slated for erection]
    * UIC
    * UIC
    * UIC
    * UIC
    * FRA
    * FRA
    as far as Santa Clara, then the usual
    * UIC fast
    * UIC slow
    * UIC slow
    * UIC fast
    (where four tracks are operationally needed, that is) northwards.

    South of the station
    * UIC
    * UIC
    * FRA

    Two UIC tracks suffice, through a third central local track as far as Tamien (for Tamien terminators, and to allow a bit of extra HSR overtake flexibility) wouldn't hurt.
    * UIC fast
    * UIC slow
    * UIC fast
    * FRA
    just as far as Tamien.

    "The bottom line is that a massive 14-track double-decker station with complex flying approaches is overkill, serves no pressing transportation need, and is a massive waste of taxpayer money."

    Exactly right.

    Oh BTW having most Caltrain terminate here (or even at Tamien) while HS and other trains run through to the south only works well with a "FSSF" configuration. Just think about it! Otherwise the terminating trains have to cross over all the running tracks and interfere with the through services. Caltrain stays on the inside tracks, and doesn't interfere with the through trains.

    To do otherwise would be completely stupid unless of course you expect to score a couple hundred million dollars of completely unnecessary pork to build underground flying junctions and multi-level stations to do this very basic job of segregating two traffic classes.

    (My coal train flyover is unfortunate, but is what I believe is the absolute minimum we can get away with and still have very good, meaning complete segregated, non-FRA train operations.)

  27. Isn't the reason they are planning on a multi-level station is because they want to elevate the HSR and Caltrain tracks near SJ Diridon because of the awful curves? According to the CAHSRA website, it shows an aerial structure south of Diridon. I agree that a double-decker mega station at Diridon is idiotic but if it improves operations of the vital lines then so be it.

  28. Yes, I strongly suggest we do what Richard "Internet Blowhard" Mlynarik says. Cause disparaging everything existing left and right is clearly the way to win friends. And anyway, how exactly do you propose we transport coal? The way they do it in Europe, on plentiful rivers like the Rhine and Danube? Anyhow, if an FRA-compliant spur can cross the VTA line, why not just cross the Caltrain tracks at grade, or have the Caltrain/HSR tracks fly over south of the station? With that, you can get 6 tracks for Caltrain+HSR, and three for ACE, Capitol Corridor, and through UP and Amtrak trains. Which, unlike HSR and electric Caltrain, actually EXIST and you can go and buy an actual ticket for them TODAY, not a potential ticket in 10 years.

  29. @Arcady: I see the distinction quite well. By "defining the requirements" and over-specifying the job, they will ensure that no existing standard fully meets their spec, and guarantee that $200 million is blown to develop or tailor & debug a system that isn't quite like any other and may not even be suitable for 220 mph operation and may have to be redone for compatibility with HSR. In my humble opinion, what they should be doing is developing their spec around existing standards, or better yet, specify broad capabilities and let the vendor do the rest. Caltrain (let alone HSR) isn't some unique, special, unusual or challenging case, so they have no business micro-managing the detailed requirements. All they should be doing is to write a top-level specification (X mph, Y tph, fixed block, 25kV compatible, MTBF, etc.) and stop. right. there.

    @Alex M: the Bay Area - Central Valley EIR/EIS shows the tracks following the Caltrain corridor exactly, including just south of Diridon. All curves are retained.

    @Anon: if you're going to criticize someone, at least have the spine to sign your name to it. No class.

  30. How many total trains per day? 48 for caltrain ( so far), 32 (eventually) for capital corridor, 50 or so hsr trains, 8 or more ace trains plus trains 11 and 14 plus extended surfliner and or coast daylight trains plus freight so you need a station that can take a couple hundred trains a day, more than have of which will need dwell space.

  31. @ Richard Mlynarik -

    afaik, the Vasona line is no longer used to haul coal to the Hanson Permanente cement plant in Cupertino because coal contains mercury which ends up in people's lungs and SF Bay fish. I believe the Cupertino plant has switched to natural gas, though I'm not 100% certain of that.

    EPA is working on limits on mercury emissions from cement kilns. California ARB is targeting cement production - an extremely energy-intensive industrial process - for aggressive reductions in the state's CO2 emissions.



    However, UPRR apparently still runs the occasional freight train up the Vasona line, presumably to collect finished cement.


  32. Clem, what about the Muni example? They didn't define the requirements, and had to spend lots of time and piles of money debugging the system, and it's still not really working well in several respects. And there is similar potential for little annoying incompatibilities with Caltrain. For example, in Switzerland, they can't run 100 car freight trains. In fact, train length is limited to 64 vehicles or less because of... integer overflow in the axle counters used for train detection. Clearly such a solution will not work in the Caltrain corridor given that longer freight trains can and do run there (and will be even more of a necessity once freight is only allowed in short nighttime windows). With a system as complicated as any railroad is, you have to be careful in how you design things, and be very aware of the costs of being nonstandard versus the costs of not having a system designed exactly for what you need. Both blindly following others and inventing a new system entirely on your own have great potential for disaster.

  33. Yo Clem,
    I never said anything about NOT being bias toward SJ, I was just asking why it exists. You're right, it's your blog and you can do/say whatever the hell you want. It's completely your right to worship the future SF transbay terminal with your posts but completely blast SJ's future Diridon.

    Considering Diridon will one day host HSR, updated Caltrain, direct BART, VTA light-rail, Altamont commuter HSR, and Amtrak Capitol corridor, all the while SF transbay will only host HSR, Caltrain and BART (via long walk)...well, perhaps your bias in grand station design is focused on the wrong project. Just my opinion.

  34. "Considering Diridon will one day host HSR, updated Caltrain, direct BART, VTA light-rail, Altamont commuter HSR, and Amtrak Capitol corridor [...]"

    Consider this riddle: Diridon "not yet dead" Memorial BART station will surely be the most importantest BART station in the entire system, what with being the most intermodaloriffic and trainiest and being sited and in the most important and largest and most populous city in the region, and being the Southern Gateway to the Capital of Silicon Valley. It is planned to be served by a train every six minutes in both directions (the EIR tells us so, and such documents never ever lie about ridership or budget) and yet somehow this crucial, vital, absolutely key, critical, world class station will only have two tracks.

    Surely such an important BART station with trains whizzing about left and right day and night really ought to have have at least 10 tracks and at least two levels of platforms. It would be an insult, a slap in the face to the whole of Santa Clara County if Macarthur BART station in Oakland -- Oakland, for God's sake! -- had more platforms. There just has be be some sort of reasoning failure somewhere in the long chain of premise and deduction somewhere, but it's so hard to put one's finger on just where that might be. A puzzler, indeed.

    Besides which, I'm rather missing the point about how the
    * Capitol Corridor (1tph and remember: it duplicates BART, and BART makes other rail redundant and obsolete, as CCJPA head Gene "foamer's hero" Skoropowski will testify as often as he's told to);
    * ACE (3 trains per direction per day and no hope of ever having more than 2tph given the PBQD/SLVG/VTA/MTC/CHSRA Pacheco scam);
    * Amtrak (one train per direction per day, and guaranteed to disappear if ever HSR shows up -- sorry, railfans)
    * UPRR (circa 0.5 trains per direction per hour); and
    * VTA toonerville trolley (single track to bustling happening Campbell, 4tph on segregated tracks)

    together or in any sort of combination mean that the mainline train station half a mile from downtown San Jose needs 14 tracks on two levels.

    Does not compute. Does not even add.

    (Caltrain+HSR = 5. ACE+Amtrak+UPRR < 2. 5+2 < 14.)

  35. Is it correct that trains will be able to stop at 4th & King, or Transbay, but not both? Because that would seem to be problematic.

  36. "Is it correct that trains will be able to stop at 4th & King, or Transbay, but not both?"

    The current official professional highly competent rail engineering plan appears to be the following:

    * 2 Caltrain tracks through the existing SF tunnels.

    * New 2 track tunnels parallel to and east of it. (Problem number one: this can't be justified on a line capacity or any other basis.)

    Both sets of tracks cross Sixteenth Street in SF at grade, and there's a cheesy set of crossovers that allow trains to cross from one pair of tracks to the other before the four tracks cross Common Street at grade.

    * The eastern pair of tracks continues (screeeeeeeeccch) on the surface to a completely over-the-top surface rail yard (yard, not station, since parking trains rather than running them seems to be what we're all about here) that occupies the entire existing Fourth and Townsend site. This rail yard features 8 platform tracks (lots of parking!), 6 of them 200m long finger islands and 2 of them (lying on the outside extremities, not adjacent!) theoretically allowing 400m long trains of some type.

    * The western pair of tracks dives, but only after cleverly crossing Common Street at grade, into a trench as they round the corner from Seventh to Townsend. The depressed tracks enter a tunnel underneath Townsend Street. The two tracks split into three just before a simply gorgeous 2 platform track (OUTSIDE platforms, which are always the wrong thing, central bypass track) station underneath Townsend, underneath a lovely mezzanine level.

    The platforms at this station are 200m long and appear to be designed with low level platforms (21 inches ATOR, incompatible with every railway standard on the planet) and hence no HS trains will be physically able to stop at this delightful location.

    From the station three (count 'em! three!) tracks continue to the magnificent Transbay Terminal rail station.

    Some people at Caltrain appear to be of the belief that not only is the Transbay Terminal so fatally flawed that they can't run most of their train service there (true), but that this can't be fixed, and so a $100 million do-over of 4th&Townsend (independent of whatever the Transbay clowns do) will give them a nice yard in the middle of nowhere where they can park all the trains they like.

    Pity about the passengers.

    Some minority of Caltrain services would continue to Transbay instead. Perhaps. And perhaps some of these would stop at the underground Mission Bay station. Perhaps. (Doing so outbound will be especially problematic, since only 1 of 3 tracks can be the designated outbound platform, which doesn't square well with the sort of bidirectional running that's needed to keep bottlenecks like the TTT approach from exploding.)

    "Because that would seem to be problematic."

    You have no idea...

  37. Wait wait wait... 21 inches, that's 533.4 millimeters, 16.6mm short of 550. Is anyone at all going to notice the 1.5 cm difference between Caltrain's "incompetent and incompatible with the world" platforms, and Switzerland's "best thing since cheese with holes in it" standard platforms? The rest of your criticism is still valid, mind you, because there really is a lot wrong with the plans that Caltrain/HSRA have regarding the whole North Terminal area. Among other things, the signal system can already support 10 tph between Bayshore and 4th/King, and with minor modifications, there should be no problem fitting in 12 tph, even if some of them stop at 22nd Street.

  38. It's completely your right to worship the future SF transbay terminal ...

    Tony, worship? You ought to try reading a little more closely. SF Transbay is a clear and present design emergency because they are about to cast major errors in concrete. That being said, I do admit to coverage bias. I haven't written about SJ yet, which is why some of my comments may come off sounding a little grumpy... the fuller picture will be painted in due time.

  39. SF Transbay is a clear and present design emergency because they are about to cast major errors in concrete.

    Especially when they look around for a place to put the commuter and regional trains coming in from Oakland.

  40. Wow I thought I was cranky but that Richard guy is a real piece of work. Have you read his other blog posts? Hey, dude it's just a train, being a passionate hsr foamer is one thing but don't get your blood pressure up over it. There will all kinds of trains to go around for everyone. Quite frankly, I hope San Jose becomes the intergalactic hub of the universe because then the next big incoming population clusterfuck of yuppie asswipes and third world society degrading newcomers will go there instead of coming to sf. Put it all down there. keep it out of here. send us up a couple trains a day and san jose can knock itself out with the rest. ( by the way, skoropowski has taken capitol corridor from failure to one of the top three most successful routes in the nation.) What do you know how to do besides bitch? Are you someone important or just a blogger?

  41. Re Transbay :"Especially when they look around for a place to put the commuter and regional trains coming in from Oakland."

    Not going to happen.

    MTC (in the persons of Steven Heminger and Larry Dahms) explicitly decided that in 1999 when it deliberately excluded the possibility of rail service from it (now $5 billion over budget, and counting) new Bay Bridge East Span.

    TJPA explicitly put the nail in the coffin in its astonishingly incompetent (subsequently privately confessed as such by agency staff, far too late) locally preferred alternative "selection".

    Game over.

    (A number of Clem's readership were there for all of it and watched it unfold. Disaster fully and accurately predicted. again.)

    It's geometrically infeasible to operate a meaningful train service along the lines (500' radius curves, only two tracks through the station and platforms = dwell time bottlenecks = high headways = bus service over the bridge has higher capacity, especially when $15 billion cheaper) that the back-of-envelope TJPA "engineering" sketches show for connection to an imaginary tube.

    It's all a fantasy at this point.
    Let it go.

    Besides, where exactly is the developed East Bay rail system that would be filling this $15 billion fantasy tunnel with trains and passengers? (Ah, BART.)

    We have BART. It works well transbay, though there's not a lot of capacity left to squeeze. We have transbay buses. There is a massive amount of bus-based capacity available and feasible across and connecting to the Bay Bridge. (Perhaps you've noticed that something other than trains are supposed to arrive at Transbay? The bus part of the program is actually well designed, because people who cared and had basic professional skill in their domain made it so. Unlike rail.)

    Of course you have to look at things as a transportation problem to be solved, not just as an exercise in building large scale model train sets.

    Last point: through-running trains consume less station capacity than terminating trains. (D'oh!). So even if somebody waved a magic wand and magically connected the other end of the station to a pipe of trains, we don't start running out of platforms compared to the existing situation.

    The problem is with throat capacity, and, beyond the obvious mitigation, it can't really be fixed, unfortunately. (Basic deal: terminating all trains on one side of the station with through tracks on the other means every terminating trains conflicts with what is supposed to be a constant, close-headway stream of through traffic. It's the FSSF problem all over. If somehow the outer tracks (outer 2 or 4, or 6!) at TTT ran through and turnbacks happened in the middle we'd be fine. But even more counter-factual.)

    In short: not going to happen. Could have happened, should have happened, but we've been screwed again by our MTC Overlords. So deal with the world as it is. No magic Caltrain tunnel to Oakland. No Caltrain to Sacramento (from either direction, thanks to Pacheco).

    And there's nothing wrong with buses.

  42. As for the TBT approach, I still think they should have had single track tunnels leading into and out of the terminal, so that the trains don't need to be turned around and eliminating the bottleneck mess. A thru station is best in this case

  43. BART does want another tube - one with four tracks (2 for conventional rail) and I wouldn't dout their ability to get what they want. They are thinking 30 years from now and 30 years ago, no one thought BART would get the the airport or San Jose either. Alameda to the tbt tail tracks... It just wouldn't surprise me if they pull it of eventually. Especially with folks from BART Capitol Corridor and HSR already working together.

  44. Re: "Wait wait wait... 21 inches, that's 533.4 millimeters, 16.6mm short of 550. Is anyone at all going to notice the 1.5 cm difference between Caltrain's "incompetent and incompatible with the world" platforms, and Switzerland's "best thing since cheese with holes in it" standard platforms?"

    That was a typo (likely, a subconscious unit translation, or a subconscious insanity correction...)

    The crazy Caltrain+TJPA+PTG number is 1'6" (18", 457mm) Above Top of Rail.

    My apologies for the misinformation.

    As for accusations of Swiss worship, thralldom to blue helmeted communist one world government ETCS bureaucrats from Brussels or similar aspersions, I try look at what works and what doesn't, and I consider at historical records of success or failure in project delivery both as indicators of credibility and as rough guides (though excellent in practice as it turns out) to future performance. In addition, the engineering principles of laziness (not reinventing wheels, profiting from other's pain, etc) and the that of not borrowing trouble (avoiding risk) generally pay off handsomely.

    These are good and tested approaches to practical engineering in many different fields, including my own.

    In the context of rail engineering and of success in providing transportation service, it would be an act of willful blindness to avoid comparisons with or attempt to learn lessons from the most intensively operated and most successful rail network on the planet.

  45. Re service to San Jose Cahill, it's interesting to consider that a far larger and far more grandiose station would have been required if the interests of BART contractors hadn't pre-determined the outcome.

    2 to 4 HSR trains per hour to the LA terminating in San Jose (= 6-15 minute platform dwells and some serious turnback infrastructure, not 2-3 minute run-though stops), 4+ tph commuter trains San Jose-SJ Airport-Milpitas-Fremont-(Livermore/Oakland), 2tph HSR trains to Sacramento terminating in San Jose, in addition to 8tph Caltrain terminating or running through the station, and in addition to FRA passenger trains from Gilroy passing through (they'd turn back in Santa Clara or at Great America for station capacity reasons.)

    We'd have been talking over 20tph in the peaks, with lots of reversing.

    Some serious and seriously impressive infrastructure would have been justified.

    Which again makes it clear that it is rent-seeking economic interests (ie BART-SJ contractors') rather than the political window dressing oh-woe-to-poor-San-Jose-largest-city-in-California-Capital-of-Silicon-Valley which are driving and have always driven the agenda.

    To make up for a one-level 7-platform station at Cahill Street which is all that's needed given Pacheco, $8 to 10 billion will be disappearing without trace into holes in the ground for BART stations in places like the SJ Flea Market.

    Follow the money!

  46. I think the "historical records of success or failure in project delivery" are not really a function of engineering, but more of a function of project management, and successful project management for the most part only comes with institutional experience. Look at the Los Angeles MTA: the construction of the first subway line was a disaster, with dodgy contracting, substandard materials, and even a sinkhole on Hollywood Boulevard. But the last section of the Red Line went much more smoothly, and the Eastside Gold Line was a very well managed project that finished ahead of schedule with no major surprises. So the reason that the Transbay Terminal is that the TJPA has never built a railroad station before, and has no idea how to do it right. Likewise, HSRA has never built anything at all, ever, and thus have no idea who even knows how to build a high speed line properly. I think in such a case, the safest way to go, really, is to take the most stable, proven technologies from the rest of the world until you can learn how to do it yourself. The US did this with 25 kV electrification in the 80s, and with electric locomotives (both the AEM-7 and ALP-46 are variants of European designs), and the ACSES system is based on French balises. ETCS Level 2 isn't at that stage yet, though it probably will be in five years at most. Of course, in some cases, FRA regulations pose a bit of a problem, and antiquated CPUC rules pose an even bigger problem, but the thing to do is lobby to reform or repeal rules that don't make sense or are not based on evidence.

    As for platform heights: the situation in Europe is the product of a long process of historical development, backward compatibility, and the like.If we're going to be throwing out our own historical quirks and backward compatibility, it doesn't seem right to just import Europe's. If we're going to wipe the slate clean, let's pick a standard that makes sense.

  47. On platform heights, once its "high enough", what makes sense is compatibility. As long as its 2 1/2 feet or more, the precise height is not all that critical. Just as long as its high enough to discourage passengers walking across the tracks.

  48. Why not use barts unused milbrae-sfo row and convert it so some hsr trains can terminate at the international terminal. one per hour to meet inbound international flights directly so they can have a direct connection to points south and inland on thru fare code share tickets. I mean I'm sure one could come up with a million reasons why it "can't" be done. But actually it could be done. IT would be a good example of filling yet another possibility of hsr - the international say paris/london/hong kong to fresno/san jose and even la connections. just adds some versatility and convenience. hsr can spare a train or two per hour to get right in there.

  49. not to mention the convenience it would be for valley passengers who lack airport access, to able to take those international vacations by zipping over from fresno and up to sfo for a step off the train step on the plane connection. while a people mover from millbrae works, the direct international terminal service is a very nice thing especially with the international terminal now serving some domestic flights as well.

  50. Game over.

    Thanks for the link. I just want to make sure I have the timeline correct. People look at transportation needs in the Bay Area and come up with "We are going to need trains from the East Bay. We could carry them over the new bridge. There's a good place to put a combined train station and bus terminal between Main and Beale." they then turn around and put an initiative on the ballot, which passes, to cram it into the existing site?

  51. The crazy Caltrain+TJPA+PTG number is 1'6" (18", 457mm) Above Top of Rail ...

    Weird, I once saw some drawings with 2'1"... 639 mm = height of Bombardier floor. Where did you get 1'6" ?

  52. The problem with "compatibility" with Europe is that they have different requirements from the US, especially in terms of accessibility. You either need level boarding or a conductor to operate a wheelchair lift or bridgeplate, while in Europe there is no such requirement. Thus, for Caltrain level boarding is essential, and for HSR it is very desirable to minimize dwell time. So choice of platform height is restricted to the choice of floor height for the rolling stock. For HSR, the lowest floor train is the Talgo 350, with a floor height of 755 mm. But of course this whole discussion is purely academic until the antiquated and counterproductive CPUC clearance requirements are repealed and platforms can be built higher than 8 inches while being closer than 7.5 feet from the track centerline.

  53. one per hour to meet inbound international flights directly so they can have a direct connection to points south and inland on thru fare code share tickets.

    Because there's not going to be enough traffic to justify running a whole train. They aren't flying 747s between Fresno and San Francisco. They are flying Bombardier regional jets and Embraer turboprops. . . less than one HSR car can carry.

    Why not use barts unused milbrae-sfo row and convert it so some hsr trains can terminate at the international terminal

    You have to remember that BART is very extra special.

    HSR and BART have different track gauges. You could reguage the BART ROW but then HSR could get very close to but not in the station. Or BART could get very close but not in the station.

    I suppose you could do something with dual guage track in the station. HSR would have to be much narrower than BART. So for one station you would create nice narrow cars all over the HSR system. (Assuming only three running rails in the dual gauge station )

    HSR and BART probably won't have the same platform height....unless you want the whole HSR system to have BART's platform height.

    BART uses turnstiles. I guess you could make HSR tickets that open BART turnstiles...

    They probably won't do the sensible thing - abandon BART between SFO and Millbrae - and build a intermodal out where the wye for BART is, the one just outside the airport. So the next best thing is to extend the people mover to Millbrae.

  54. @Adirondacker- I see your point about possible lack of demand. and yes bart is very power hungry, bu twhat I was thinking was keep bart into the sfo station from the north just as it is, then take the aerial structure of the millbrae sfo - that bart really isn't using, rip out those tracks and third rail and just use the structure and bart's unused platform. with mods, it would create a space where say one out of X number of hsr trains could terminate. I mean if you think about it now, anyone who flies who lives in the areas between palmdale and palo alto has to get to the nearest airport somehow. It would just add flexibility as to which airport to use. If I live in fresno or even bakersfield and Im going to London, I can have the choice - say maybe british air has a BFD London ticket with a connection at sfo. Also certain airlines have nonstop flights and different prices - LAX versus SFO. so one could shop around. it woulds also add an additional terminus (tbt-4th-sfo) in which to distribute hsr volume. I'm not saying its a very important idea, just trying to think outside the box. that way at least some people could get from plane to train with no intermediate modes. Say there really are 12 tph. 8 at tbt 2 at 4th and 2 at sfo. something like that? it would certainly be good for folks in san jose (to sfo) gilroy/monterey, merced, fresno, and i think that if I lived in bakersfield and had a choice of going to london on hsr via sfo versus, the schlepp by whatever mode to LAX - sfo is the more pleasant choice. no?

  55. Because there's not going to be enough traffic to justify running a whole train. They aren't flying 747s between Fresno and San Francisco.

    They aren't flying 747s between Los Angeles and San Francisco, either. Most passengers even between LA and SF will be induced demand or diverted from cars, and that's a market that has cheap air service. FAT-SFO is more expensive and less convenient.

  56. had a choice of going to london on hsr via sfo versus, the schlepp by whatever mode to LAX .

    There's not train loads of people going from the Central Valley to international destinations. I doubt there's more than a trainload or two an hour of people going from all of California to international destinations.

    IF everybody who goes to SFO from the Central Valley decided to get there on HSR you still wouldn't fill a train an hour. People will still fly to LAX. While there's lots of overlap between service from LAX or SFO there's going to be destinations that make sense via LAX - or the other airports that Central Valley airports connect to.

    Extended people mover to the HSR tracks adds a few minutes to the total trip. Anybody not destined for the international terminal will probably be on the people mover anyway. It means passengers can use any train that stops at their station and SFO instead of waiting for the train that goes to SFO. If the timing gets snafu'd for some reason they can take a train that doesn't stop at SFO and transfer at San Jose. . . you aren't suggesting the HSR take up BART tracks and also extend the people mover to an HSR station on the mainline?

  57. I was suggesting that because I didn't know for sure that they are indeed going to bring air train to the millbrae station. If that's the case then that's way it has to be. I just wanted to do something with that unused bart structure - all assuming tha bart will even give it up. They still have sights set on continuing south from millbrae ( they never give up,never they will ring the bay come hell or highwater someday)

    but whatever, Ill continue using civic center-sfo direct anyway so it doesn't matter to me. i was thinking this

  58. Main line trains don't fit in BART sized holes.

    This is not an incredibly subtle and nuanced deduction based on immensely difficult and painstaking observations and exhaustive precision train metrology. It ought to be blindingly obvious, even if you've never visited the Bay Area in your life.

    Stop the insanity!

  59. No, but hsr trains could:

    and here

    there may be no need, but it could be done.

  60. jim: look closely at the BART train and the HSR train. You'll notice that the roofline of the BART train starts almost directly above the doors, and the roof is almost flat. They don't even have room for roof-mounted air conditioners up there. The HSR train's roof is higher and probably a bit more curved too. And of course you need room for the wire above, at least a foot above the roof of the train, and maybe 8 inches from there to the crown of the tunnel. That's why BART uses third rail.

  61. YEs I saw that - but again my origianl thought was to term a couple hsr train in the sfo terminal - avoiding any tunnels ( by just using the aerial structure from from where it leaves the caltrain row - into the terminal where there is plenty of clearance - and to do so for a few trains a day to give some people added convenience, and to make use of an other wise abandoned aerial structure. But if bart is going to give up that aerial structure for air train use then good enough. or maybe bart is going to use its own train to shuttle people between millbrae and sfo. what exactly IS the plan anyway? My only point was if there were a desire to do it - you get hsr into sfo by modifying the existing structure. apparently no on thinks there's a need. I'm also thinking of ways to term some of th trains in areas other than tbt since there are capacity concerns. but 4th can probably handle it anyway. seriously though is there a plan yet pertaining to air train/millbrae/hsr?

  62. ( btw - I think the muni metro tunnels are the same size as the bart tunnels or smaller and they have room for taller trainsets plus catenary) but its a moot point.

  63. Wrt Transbay, FWIW, Eric at Transbay Blog claims:

    We are at a preliminary stage in all of this [TBT design]. Designs have been sketched, but now is the time when we dig into the details. TJPA is requesting DTX design funds, and I’m also told the throat design will be reevaluated as part of the train box design development. That will continue throughout this year, before construction commences. Hopefully the layout will be redesigned, since TJPA is aware that the current layout is not ideal, but we can’t say just right now what the delta will be from the current sketch.

  64. Rishard said

    "The current official professional highly competent rail engineering plan appears to be the following:


    Both sets of tracks cross Sixteenth Street in SF at grade..."

    This seems like a bad idea. What happened with the plan to run all of these tracks below grade?

  65. Updated with the final MTC document. The post previously had only a draft; the only major change is a $200M increase in the request for Caltrain electrification.