02 December 2008

Focus on: Millbrae

Millbrae, by virtue of its proximity to SFO and its connection with BART, is planned as an intermediate HSR stop. The existing BART / Caltrain station is to be expanded, and is naturally the focus of the HSR project in Millbrae.

Millbrae Station

Plans for the Millbrae HSR station, described in the CHSRA's environmental impact documents, call for four tracks to pass through, with two outside platforms. This is shown as Option A in Figure 1 below.

Unfortunately, the incompetent design of the existing Millbrae station (opened in 2003 and billed as the largest multi-model station west of the Mississippi) has squandered much of the space available within the Caltrain right of way. Massive support columns for the expansive station mezzanine have been placed right where you'd want to run extra tracks, and a third platform track at the south end of the station (presumptuously built to terminate some Caltrain services, assuming everyone would want to ride BART into San Francisco) sits unused, with no possibility of continuing north through the station. The photo at right, taken in 2000 during construction, illustrates how a forest of concrete pillars now constrains the right of way.

The total lack of foresight in this design is breathtaking.

To make room for four tracks, the existing southbound platform and part of the mezzanine that it supports (including the ill-placed pillars) may have to be demolished and rebuilt a few dozen feet further to the west.

View Larger Map

Millbrae is an important stop for Caltrain, ranking 5th in ridership, and even higher when counting BART. As a measure of its importance, every Baby Bullet stops there. Assuming one wanted to make Millbrae an important interchange point with HSR, it would make sense to maximize the opportunities for cross-platform transfers. Cross-platform transfers require an island platform between two tracks, allowing passengers to quickly and easily change trains on the same platform. For example, a much-touted cross-platform transfer is already possible between Caltrain and BART in the northbound direction.

With four tracks through the station, a solution that maximizes cross-platform transfers would consist of two island platforms flanking the center pair of express tracks. The existing northbound island platform shared with BART would be preserved. Advantages:
  • 4 trains can occupy the station at once (not counting BART)
  • Cross-platform transfers between HSR and Caltrain
  • Cross-platform transfers between Caltrain local and Caltrain express
  • Flexible routing through station under perturbed schedule, which minimizes cascading delays
There are also some disadvantages. Non-stop HSR trains pass platforms at 125 mph, likely requiring a physical safety barrier. Other disadvantages depend on how this configuration is built. Two possibilities are labeled Option B and Option D, in Figure 1 below.

Figure 1: Millbrae station configuration options

Option B, to construct new island platforms and preserve the existing BART cross-platform access, increases the station footprint the most. The westernmost track is shifted ~60 feet to the west (vs. ~30 feet for thru configuration), with possible impact to private residences to the north and Millbrae Ave overpass to the south. As in the CHSRA's Option A, most of the station mezzanine and support columns must be reconfigured.

Option D (raised in the very thoughtful comments to this post) calls for converting the westernmost BART track to use by Caltrain / HSR. The way BART is operated, with half the trains terminating at SFO, the Millbrae station is underused. The eastern BART island platform is usually closed to service. With the expected HSR and Caltrain improvements, BART has no chance of being extended past Millbrae. All of these factors support the notion of cutting the BART portion of the Millbrae station from three platform tracks to a more reasonable two.

The biggest advantage of this approach is to preserve the existing mezzanine and platform configuration, without the need for large-scale demolition. The cross-platform transfer to BART (the cherry on the cake!) would be lost, in favor of better cross-platform options between HSR, Caltrain express and Caltrain local services. Option D also has the least encroachment on land to the west of the station, which may ultimately seal the deal.

Constructing Option D would require some reconfiguration of the existing station, such as removing the eastern Caltrain siding track, removing the western BART tail track, reworking the BART subway portal, and probably constructing a physical barrier to separate BART from Caltrain/HSR on the adjacent track.

Another possibility that preserves the existing mezzanine structure is Option C, although it only provides three platform faces for HSR and Caltrain. This option involves routing two new tracks to the west of the existing station, with some of the disadvantages mentioned above.

None of the options discussed here is perfect; all of them have drawbacks because of the suboptimal way the station was originally built. If the city of Millbrae, Caltrain, and the CHSRA are serious about making the Millbrae station live up to its intermodal billing, they should opt for the lesser of all evils (Figure 1 Option D), and not the design currently described in the EIS/EIR documents. While the BART cross-platform transfer (the station's political selling point) would be lost, other advantages far outweigh this loss.

Station Area Development

Whatever the final station configuration, adding two tracks to the Millbrae station will expand its footprint to the west, encroaching on an area locally known as 'Site One,' where the city of Millbrae and private developers desire to build a Transit Oriented Development. Ironically, the developers have claimed that HSR threatens to derail their TOD plans, quite shamelessly undermining their claim to be "Transit Oriented." The people who run Caltrain appear to have capitulated prematurely by promising the developers that HSR can be accommodated within the existing station footprint, an option that even the CHSRA doesn't try to pass off as feasible.

The CHSRA has also penciled in another parking structure for 230-400 cars to the west of the station on Site One, apparently without regard to the existing under-utilized 3,000 space BART parking structure, which has plenty of surrounding land for expansion.

Grade Separations

Most road crossings in Millbrae are already grade separated. A separation at Center St, at the entrance to Marina Vista, remains to be built. This is a constrained location where the BART tunnel box runs immediately to the east of the crossing, which prevents lowering the road.

BART Tracks

The BART to SFO project (which, as built, is much better described as BART to Millbrae) was brought to you by Quentin Kopp, who is now chairman of the High Speed Rail Authority. While Kopp has clearly established a track record in bringing large transportation projects to fruition, his BART extension shows all the signs of having satisfied not just a transportation need, but the raw urge to pour way more concrete than strictly necessary.

A lot of this concrete was poured on, under, and around the Caltrain right of way, constraining it in various unpleasant ways that will come back to bite us (the taxpayer) as HSR is built. The botched Millbrae station discussed above is only the tip of the iceberg. The construction companies won't mind: there's real money to be made in jackhammering old concrete to make room for new concrete.

According to the CHSRA's environmental impact report Appendix 2D for the Caltrain Corridor, the HSR tracks will run through Millbrae at grade, right next to the BART tracks. But after BART was built so expansively into the Caltrain right of way, there isn't a lot of room left to build another pair of tracks. For example, take a look here and here, and think about how to fit in the planned 100-foot, 4-track right of way as shown at left.

NOTE: This post will be updated continuously, as warranted by additional information or new events relating to Millbrae. Last updated 09 December 2008.


  1. Clem -

    Robert Cruickshank over at the California High Speed Rail blog alerted me to yours. It's great to see someone delving into the nuts and bolts of actually getting HSR implemented in the SF peninsula. There is no shortage of problems, Millbrae arguably being one of the thornier ones.

    IMHO, there are two basic approaches to getting this done. The one that will likely yield the best solution is to widen the entire Caltrain corridor to 100 feet, enough for 4 tracks side-by side, all of them fully grade separated.

    Since not every HSR train will stop at Millbrae and, FRA hasn't ruled on mixed traffic in the corridor yet, the project-level EIR/EIS assumption should be that HSR alone will need 2 express plus two station tracks at Millbrae. That implies an aerial structure flying over Millbrae Ave to platforms high above grade. That would mean demolishing much of the existing station.

    If FRA does allow Caltrain to use UIC-compliant EMU equipment (Siemens Desiro), it could share track with HSR, which will also use UIC-compliant trainsets. In that case, the existing Caltrain tracks could be shared. In that case, only two new express bypass tracks would be needed at that station. Without them, it could be a bottleneck for HSR operations since Caltrain equipment will be limited to 90mph.

    Note that HSR trains require raised platforms and, that those trains can be up to 1320 feet long. If Millbrae station is to be an effective HSR feeder point for SFO, full-length trains will have to be accommodated. It will take more than one pedestrian overpass to handle the large number of passengers that may board or alight an HSR train at Millbrae.


    The quick-and-dirty approach to HSR implementation is to run an aerial structure above the existing Caltrain tracks using gantries. This would avoid eminent domain proceedings for widening the ROW, secure the required gradient for HSR, grade separate HSR and, minimize disruption to Caltrain service. The VTA tracks in Mountain View might not have to be moved.

    However, the Caltrain tracks would retain grade crossings, albeit upgraded to quiet zones. This would retain all existing cross-roads, greatly reducing the overall implementation cost. The overhead catenaries for Caltrain would be attached directly to the gantries and, those long HSR platforms would be immediately above Caltrain's. Express bypass tracks for HSR would be no problem, since they would only be needed in places where the ROW is 100 feet wide anyhow - with the notable exception of Millbrae, see above.

    Track stacking would likely create some vertical clearance conflicts with existing overpasses, so there may be a case for a hybrid solution involving stacking only where it makes the most sense. HSR trains can handle gradients of up to 3.5%.


    Note that CHSRA has stated that HSR network operations will be put out to tender and, that multiple train operators may be permitted. Separating the monopoly infrastructure operator (on a long-term franchise) from competing train operators is the model already in effect in the UK and will be all over the EU as of 2010.

    Quite possibly, Caltrain would have an opportunity to operate strictly regional HSR trains between Gilroy and SF, on the HSR tracks, in addition to local trains. Caltrain would then have to decide if it wanted to retain the "baby bullet" service level as well.

    The rolling stock for regional HSR trains would not need to support top speeds in excess of those used by the long-distance bullet trains in the Caltrain corridor, i.e. 125-150mph. It's quite likely that Pacheco Pass will be among the last sections of the HSR starter line to be finished.

    It's also possible - though not currently planned - that the standard gauge network throughout Northern California will be upgraded in coming decades - especially if FRA defines a regulatory path toward mixed traffic that I call rapid rail.

    This multi-page map shows how that could include express service up the East Bay, across the Bay Bridge (perhaps), over to Concord and the CV and up to Sacramento/Napa Valley/Santa Rosa via Vallejo. Separately, there could one day (perhaps) be a new Richmond - San Rafael rail bridge. There is no reason why Caltrain should be confined to the SF peninsula forever.

  2. Clem - I just wanted to say that this is a fantastic blog performing a very useful public service. I had no idea that the San Bruno curve was so sharp (I'd assumed the worst curve was at Sierra Pt. I'm sure you'll get more traffic now that the CHSR blog has linked to you.

  3. This is a fantastic resource. May I humbly make a suggestion? This site would be better served by a Wiki instead of a blog. Since you state your intention to continually update each topic, the Wiki format would be perfect. It is also possible for readers to subscribe to an automatic RSS feed of changes to a Wiki, which would be extremely convenient for your readers.

    Keep up the good work.

  4. Forget Millbrae. Bypass it and go directly to SFO.

    HSR bypass to SFO

    This would allow a bunch of problematic stations to be skipped. A bunch of slow curves to be avoid. And lastly and most importantly -- a direct connection with the airport.

  5. addendum to my earlier comment. There is now more detail about the HSR bypass over on my blog

    But Clem.. glad you are doing this!

  6. Pat - No offense, but your proposed alignment would probably cost at least $2-3 billion. It looks like it would have to be underground almost the entire way, and it couldn't just be a cheap cut-and-cover job either. So you're proposing spending the majority of the SF-SJ section budget just on this one bypass.

  7. @Rafael - Millbrae basically boils down to which option will require the least demolition. As you point out, an aerial structure would require a lot of demolition.

    I suspect that limited eminent domain takings will be far cheaper than bending over backwards to build an aerial structure.

    The FRA is very likely to allow Caltrain to go "non-compliant"; Caltrain appears to be doing a good job selling the safety case. It is conceivable that Caltrain equipment could be spec'd to 100 - 110 mph; such speeds are already practiced by New Jersey Transit. The smaller the speed differential with HSR, the easier it will be to mix traffic.

    Good point about the platforms; that is probably a good subject for a future post. The California PUC will need to relax its 19th century clearance requirements.

    @jwb - I thought about that... and concluded it would be better to abuse a blog as a wiki than to abuse a wiki as a blog.

    @Pat - I think we members of the public need to make sure the KISS principle is religiously applied. (Keep It Simple, Stupid). The alternative you suggest is far more complicated and expensive than straightening a curve and reconfiguring some stations. For just those reasons, I'm sure the construction companies would love it! Anyway, I agree with mike.

    Thank you all for visiting.

  8. @Mike -- $2B! not a chance at least not for alternative #1.

    I haven't really looked at alternative #2.

    Figure $100 million/mile of tunneling. This is the number BART-2-SJ is using. Also the entire length is not tunneled some of it would be in a cut but not tunneled.

    To the north of the airport I think it is possible to make the ROW be above ground more.

    But to be realistic you have to look at the alternative of following the existing Caltrain ROW. From just south of San Mateo to just south of Bayshore is probably the worse section of the Caltrain ROW to try to squeeze in HSR.

    Also consider the benefits of a SFO terminal.... what are the costs of not having that in terms of ridership? Specifically this could make a direct SFO-SJC shuttle possible. Additionally, a direct SFO would make it easily for a airline to code-share with a HSR train so a traveler would "transfer" to a HSR flight to Modesto/Fresno, etc. If the HSR terminal is way off-site like Millbrae is.. then a traveler can not have seamless transferring of luggage.

  9. @Clem -- and demolishing people's houses is KISS? Not a chance! The route I choose impacts only businesses. Much easier to eminent domain on. Also it removes the impact of HSR from the center of the cities and avoids fights around the Millbrae station.

    I don't mind fights when necessary, but why fight when a direct SFO/HSR station would result in an infinitely better system?

    What you are talking about is nibbling around the edges and involves a lot of engineering in its own regards. Rather than try to "fix" the problem spots .. just avoid them!

    (P.S. I have tweaked the map a litle bit)

  10. ...(continued) (internet keeps dropping). Also this bypass avoids fights with 4 cities (South San Francisco, San Bruno, Millbrae, Burlingame) about how HSR will impact their cities. This alone will save millions in escalating construction costs.

    Thus reduce the possibility of another Quentin-Kopp-engineering marvel like the Millbrae BART Y occurring

  11. Hey Clem,

    Are you taking requests? How about Focus on: Burlingame. I grew up there and my dad still lives next to the tracks between Burlingame and Broadway stations. They have been putting off grade separation for years and I am curious what the plan will be. They just completed a big remodel of the Burlingame station which I am sure is out the window now. My dad's big question is what will happen to the Eucalyptus grove that he walks his dog through. We're big train fans and are looking forward to this.


  12. @Pat - eminent domain is not as hard or scary as it sounds; the process exists for a good reason. I'm not sure I see why it would be harder on private residences than on businesses. There was plenty of eminent domain (nearly $200M worth) for BART to Millbrae.

    It would be nice to improve the connection with SFO. Since they didn't put the intermodal station in San Bruno (where it would have made more sense), the least awful remaining option is probably to convert the south leg of the BART wye for use by the AirTrain... BART will no longer need it because Millbrae will be forever the end of the line.

    I wouldn't oversell the importance of the SFO-HSR link. I don't believe that CDG, AMS, etc. are busy HSR stations compared to their corresponding city centers.

    @luceandrick - Focus on: Burlingame is in my plans. There's a bottomless well of interesting topics at every milepost along the peninsula! Stay tuned.

  13. @ clem -

    there is one obvious eminent domain alternative: kick BART out of Millbrae!

    Convert their broad gauge tracks to standard gauge, including those on the southern access ramp into SFO. Change the tracks there such that there are three rails so BART and electrified Caltrain can both directly serve SFO. This would mean any given northbound Caltrain would terminate at either the TTC or 4th & King or SFO, but I suspect those are trade-offs peninsula residents could live with.

    At Millbrae station, the tracks Caltrain uses today would either be dedicated to HSR or, both Caltrain and HSR would use the platforms currently reserved for BART and designate the existing Caltrain tracks at the station a bypass for HSR express trains.

    BART would get to keep its tracks between its San Bruno station and San Diego Ave for overnight storage of trains. Early morning service from SFO would involve shunting trains via the BART San Bruno station. Caltrain would use the third track at Millbrae station to park a train or two overnight.

    Caltrain tracks would pass the BART tail tracks to the east at San Diego and return to its own ROW at Lions Field Park.

    For clarity, please look at this Millbrae HSR implementation map.

    Anyone wanting to transfer between Caltrain and BART would do so at the SFO station. If sharing tracks there is not possible, then the southern ramp over to SFO should be abandoned altogether. Instead, the Caltrain San Bruno station should be moved north to Scott St., one block east of the BART station. In addition, SamTrans shuttle bus service between Millbrae and SFO should then be restored.

  14. @ Rafael

    I don't know that kicking BART out of Millbrae is a good idea, but given the developments since opening of the station, I think it could be readdressed. Specifically, bring the people mover into Millbrae using the south BART track as a previous comment stated and/or use the third BART berth for HSR/Caltrain purposes. This doesn't solve the issue - one berth isn't sufficient for needs, but I bet Caltrain/HSR could make better use of it now that BART isn't running into SFO from here.

    I also agree that further BART progression south-bound is a non-starter given poor showing to SFO and the limited ROW through Peninsula communities, so we should look at what that means to BART in Millbrae...

  15. @ Rafael, Clem -

    All of that work sounds awfully complicated. Also, considering Millbrae BART is Kopp's baby, I'm not sure he would even consider ending service there. What is wrong with the proposed Millbrae station concept that was presented in the EIS/EIR documents? Because Caltrans will be sharing tracks with HSR, it means local HSR/Caltrans trains will go up the coast on the right track (looking at picture) and offer cross platform transfers to BART going into the bay. On the very left track, local HSR/Caltrans going down the coast will have walkways to transfer if needed. The center two tracks between these local tracks will allow for the baby bullet (if still running) and HSR to run at full speed without the need for barriers to protect passengers on the platforms because they are farther away from the platforms. It seems to me that Clem, your proposal makes through express trains difficult because they are right next to platforms. I'm pretty sure that in other countries with developed HSR networks all choose a setup similar to the one in the EIS/EIR documents. What is wrong with this proposal exactly?

  16. @ Alex

    I cannot find anything in the library docs specifically about Millbrae, can you point me in the right direction?


  17. Yes, I have read your proposal thoroughly Clem, I just don't understand why Express trains would stop at Millbrae, isn't the idea that the express trains will do a TTC-SJ Diridon-one central valley stop-Union Station/Anaheim run? I understand if someone lived in San Bruno and was coming from LA and was on the express train (obviously they wouldn't take a local train from that far away), they would want the train to stop at Millbrae so they could transfer easily but wouldn't it be easier to implement the proposal in the EIS/EIR documents, meaning this person would have to go the TTC and transfer and come back down. Is the cost of implementing your plan worth the cost versus making the trip slightly longer and inconvenient for some passengers? Is it even possible to run an express train at 125 mph next to platform legally?

    @ timote - I'm referring to the pictures Clem put together in his blog, I don't know where it is specifically discussed in the EIS/EIR documents, sorry.

  18. @ Alex-

    Oh, ok. Well I presumed that I missed something and found these links in the library:



    They are not overly useful as they are pretty generic. They are just saying what this post is saying - that they're going to have to reconstruct the Millbrae station...

  19. @ timote -

    Great! Thanks for finding those.

  20. I find it hard for myself to blame the construction of the Millbrae station for it's lack of compatibility;

    When consider that the millbrae station was designed when HSR was very uncertain, it is very understandable that it was not designed with a 4 track solution in mind, especially when you consider that most engineering projects are designed with the problem and the budget for the current project and not uncertain upgrades.

    I have faith that when HSR comes, the station can be effectively modified without major demolition or very expensive bypasses. HSR Engineers should be able to meet the needs of HSR without incurring huge costs.

    (i am just talking about the HSR component, not the other issues with the station)

  21. @Alex - yes, I meant that it ought to be possible to transfer between Caltrain expresses and locals (not HSR expresses). If this is low-hanging fruit, it ought to be picked, although I agree that it remains to be seen exactly what the costs might be.

    I agree that BART shouldn't be touched... that's another can of worms, politically and technically.

  22. @Ben - I'd love to see a concept that doesn't require demolition or eminent domain and can be built at reasonable cost. What do you have in mind?

  23. @ Clem -

    Is it completely out of the question to convert one of the existing BART tracks to standard gauge because it is not used very often and run express trains in either direction? This seems like the only possible solution that wouldn't require extensive modification of the station. Local HSR/Caltrans could run on the existing tracks. I doubt they could run express trains through the station in both directions, sounds way too unsafe.

  24. Your right in that eminent domain will be required to expand the station,

    and there will need some demolition, but complete demolition of the entire station (i may have read that wrong, but i believe i read that a complete demolition was required).

    I dropped out of civil engineering in college, but it seams feasible to me that just the left side that is over the cal train row would need to be demolished, which can be demolished without significant disruption in service.

    This station is a problem, but i don't see it as a major obstacle to construction

  25. @Ben - nobody's talking about complete demolition. Only the portion of the mezzanine over the Caltrain tracks.

    You're right, in the grand scheme of things it's not a big deal, but still frustrating that Caltrain didn't have the foresight to better preserve their most valuable asset: their right of way.

  26. Ahh i guess i misunderstood your plan then, sorry about that ;)

  27. Looking at the map on Google of the Millbrae station, I don't see how it is possible to add in anymore space to the station given the space restrictions. How could you add in two more tracks without leaving the train ROW and entering onto that parking lot to the north and later that building? Is the proposal to quickly branch out of the two existing tracks to the north, enter the station, then exit onto the 3 tracks that already exist to the south + 1 new track added to the west of the existing tracks?

  28. All I can say after studying the aerial photos is that Kopp and BART must have been smoking something to decide to take up so much of the Caltrain's ROW. They made the development of HSR infinitesimally more difficult and expensive. Hopefully Kopp realized he screwed up when he comes to work on the actual engineering plans.

  29. Does anyone actually use BART service to Millbrae for anything other than transferring to/from Caltrain because it doesn't go to downtown SF?

    With the DTX tunnel, Caltrain will go downtown and so will HSR. Transfers between the services would still be possible at the Embarcadero/TTC (2 block walk), San Bruno station (on block walk if Caltrain station is moved) and SFO (no walk at all).

    As for the argument that BART was Kopp's baby, keep in mind that his original job description was to take BART to SFO. The whole subway-under-San-Bruno thing and the intermodal at Millbrae was something he claims San Mateo county insisted on, hence the cost escalations.

    Besides, his new baby is HSR so if kicking BART out of Millbrae completely is the easiest way to get HSR its express bypass tracks, so be it. Simply sharing Caltrain's existing tracks at Millbrae - assuming FRA permits it - would create a bottleneck for both services while the adjacent BART platforms remain almost completely unused.

  30. Rafael - I agree completely that a two-track solution is not feasible. I need to find out exactly what Samtrans (Caltrain) has promised the city and developer for Site One.

  31. Btw, there is a middle ground option. If BART agrees to vacate Millbrae entirely so it can become a station with five standard gauge tracks, Caltrain could agree to equip some number of its new Siemens Desiro EMU trainsets with variable gauge trucks. It's a retrofit solution, certified up to 150mph if the base equipment was designed for standard gauge. It is in commercial service in Spain.

    In my kick-BART-out-of-Millbrae scenario, a single gauge change station would be installed on the tracks heading up the ramp from Millbrae to the SFO station. Caltrain equipment would switch gauge on the way up and back again on the way down. This would avoid any disruption to BART service into SFO or Caltrain service to SF while Millbrae station is remodeled.

    The problem of sharing tracks at SFO would reduce to the installation of overhead catenaries there, plus signaling and dispatch modifications to those trainsets Caltrain chooses to run between SJ/Gilroy and SFO.

  32. I think the best solution is to simply end BART service to Millbrae and terminate the service at SFO. Rafael- You make a great point saying that the whole point of Millbrae is to allow transfer to Caltrains. That job will be filled by the TTC, therefore there is no reason for Millbrae. There could easily be a shuttle of some kind from Millbrae to SFO for local Caltrains passengers. Is it possible if BART was removed past SFO, that the BART tracks that were put in the Caltrains ROW could be removed? It sounds like an awful waste to destroy something you just build but it could mean 4 tracks could easily be fit going down the ROW and make life for the planners of HSR a whole lot easier.

  33. @ Alex -

    I was just talking about ending BART service to Millbrae, not to SFO. BART still needs somewhere to park a number of trains overnight. See this map for details, the green pin denotes where the BART tail tracks would terminate.

    The HSR express bypass tracks would run west of them and the tracks shared by Caltrain and HSR to serve Millbrae station routed just east of them through Lions Field Park, where the BART tracks run underground.

    The new set of standard gauge tracks woul run on top of the covered BART trench for almost 3000ft. Grade separation of the standard gauge tracks will probably require an aerial structure for that short section because the cover of that BART trench almost certainly wasn't built to support the weight of a berm, tracks plus a train. Once the alignments diverge just past I-380, CHSRA/Caltrain have more options for implementing grade separation.

    Operationally, BART could choose to run its northbound early morning trains from San Bruno or, reverse up the ramp into SFO. Once the first trains of the day arrive from Pittsburg/Bay Point and either Dublin/Pleasanton or Richmond at SFO, normal service is delivered. Similarly, the last train or two at night would either not go to SFO or reverse to San Bruno before cutting over to the tail tracks for the night.

    Caltrain would use the fifth track at Millbrae for overnight parking of the special trainsets deployed on the new Gilroy/SJ - SFO service that I'm proposing. The first trains of each morning could head directly to SJ or, backtrack to SFO and run to SJ via the regular tracks in the middle of Millbrae station.

  34. Those are interesting ideas, but I think you guys need to keep it simple and practical, both technically and politically. It's easy to dream up all kinds of ideas, but in the end the best ideas are usually the simplest.

    Call me unambitious, but I'm not in favor of anything that complicates HSR or uses it as an opportunity to re-optimize another part of the pathetic configuration in Millbrae... just get the minimum for effective HSR and Caltrain, and move on.

    Sorry, didn't mean to be a party pooper! Just my two cents.

  35. @ Clem -

    I completely agree. It sounds like there are many different options for serving Millbrae effectively, but many will require more funds than necessary. I hope Caltrain can simply implement the proposed plans in the EIS/EIR documents and move on like you said. It sounds to me like too much aerial work would become prohibitively expensive because lots of concrete=$$$.

    @ Rafael -

    I still don't understand how someone with the wealth of knowledge about this project that you possess isn't part of the CHSRA. I could never look at a map and come up with as many imaginative and well thought out proposals as you have.

  36. Place two tracks west of the existing western Caltrain platform, seal the "cross-platform" platform, and let express HSR and Caltrain run on the outside and have stopping trains stop on the central platform.

    The platform will need to be extended for HSR, so add a second mezzanine and more escalators and elevators. Assume/hope for a SFO Airtrain extension to Millbrae, stopping north of the existing mezzanine.

    This Airtrain connection could support a very nice airport hotel in Millbrae, connected to the station. Also, the existing mezzanine can be extended through future development to El Camino.

    South of the station, the historic depot needs to be moved out of the way, but it was moved once before, from right next to Millbrae Avenue. A move some feet to the west shouldn't be a problem.

  37. @anon, I like it a lot.

    - no major demolition at Millbrae, no messing with BART
    - reasonably compact
    - no high speed movements on platform tracks

    - fewer cross-platform opportunities
    - need for flyovers to turn the express tracks "inside out" at some point... although the outside config (with island stations) may work well in other locations as well... e.g. Belmont

    I'll try to draw it up at some point. With 400 m platforms, would it be feasible / desirable to have two Caltrains occupy the same platform at the same time, one behind each other? This might make up for the loss of cross-platform opportunities.

  38. @ Clem -

    Yes please, make the map, it sounds like it could work. What do you mean by "have two Caltrains occupy the same platform at the same time, one behind each other?" What are the trade-offs between partially demolishing half of the Millbrae station and rebuilding it as opposed to making flyovers about a quarter of a mile past the station, i.e. which costs more?

    And on another slightly relevant note, where do the freight tracks split away from the caltrain's tracks? Because you couldn't run a freight train on a flyover I'm pretty sure if that became the solution.

  39. @ Clem -

    what I proposed keeps all tracks at the Millbrae station at grade and within the existing ROW. There would be no need for modifications, other than elevating platforms for level boarding (which Caltrain is planning to do anyway) and increasing their length to support full-length HSR trains.

    No demolition of the mezzanine superstructure would be required at all. No eminent domain against nearby residents would be needed at all.

    The only new land required would be about 1000 feet south from Lions Gate Park, where there are no private properties anyhow.

    Plus, you'd have Caltrain service directly into SFO. Extending the AirTrain to Millbrae would be much harder>: it's a level above the BART tracks at SFO, so you'd need to construct something for the AirTrain to reach the existing ramp across 101. Also, the AirTrain uses concrete guideways, so you'd have to rip up the broad gauge tracks on that ramp. Finally, the ramp tracks are the middle two of four on the approach to Millbrae station. Crossing AirTrain and heavy rail tracks is point blank not possible, so that's the deal killer right there.

    My objective is not to optimize the layout of Millbrae station, it's to minimize eminent domain, minimize station modifications and maximize utility for both Caltrain and HSR operators and customers.

    As for "political" complications, puhleeze. Who do you think has more clout right now, CHSRA or BART?

  40. If you've every used BART to SFO, most people end up transferring to the AirTrain. By building an elevated Airtrain extension all the way to Millbrae, you get a connection to all terminals without a second transfer. If Millbrae's redevelopment at the station includes a hotel, you get that connection, as well as a connection to all the samTrans buses that serve Millbrae station.

    Maintaining a dual-guage fleet will be an onerous task. Spain does it because they have to. To implement your (Rafael) solution, why not just re-gauge the 3/4 mile of tracks that make up the south wye for BART? You'll have to erect OCS anyway, or do you want to specify dual-guage, dual-voltage trains to save that cost?

    Beyond that, the clearances into the SFO station would not permit standard rolling stock, let alone overhead electrification. I doubt the BART structure could carry a standard EMU either.

    The Millbrae station and connection to SFO isn't a place to be penny wise and pound foolish.

  41. @ Rafael-

    I think eliminating BART from Millbrae is a non-starter. It has too many riders (granted, not near what they expected) and too many parking spots. If Millbrae were eliminated and the passengers moved up to San Bruno, it would triple that station, but that won't happen cause the San Bruno parking lot fills at 8 am.

    I'm not a fan of eliminating services that work, even if imperfectly. I'm down with reclaiming a berth that BART uses as parking and possibly reclaiming the Millbrae -> SFO track for AirTrain/Caltrain/whatever, but kicking BART out of Millbrae I'm not in favor of.

    Also, I don't understand the AirTrain/heavy rail argument. From what I see on google maps, it appears that the center tracks that go to SFO are grade-separated from the outside tracks, that they fly over the outside track. If that's the case, then you could tear up the rails and AirTrain could own that center area of the tracks.

    On the airport side, yes - the interconnects up to the existing AirTrain is definitely non-trivial. But for the benefit it would cause, I'm sure that they can figure it out...

  42. Errr I shouldn't have made the San Bruno that definitive. It would likely cause big issues with parking, let's put it that way, and Tanforan would not be happy to become BART's Millbrae parking lot...

  43. @ anon @ 14:33, timote -

    equipment with variable gauge trucks can be used on both standard and broad gauge tracks, that's the whole point. Ergo, Caltrain would be free to use it for Gilroy/SJ - SF (4th & King or TTC) as well. It's just that operationally, it probably would exercise that option very frequently.

    Converting tracks at the SFO station to three rails (+ BART power rail) would allow Caltrain to avoid messing around with variable gauge equipment. However, BART tracks are laid on top of concrete sleepers that you cannot safely drill holes into after the event. Ripping up tracks inside the SFO station would result in massive disruption of BART service, so IMHO that's not an option.

    The vertical clearance at the SFO station is definitely something to keep in mind. It's possible that bi-level EMUs would not fit. Worst-case, there's always the option of using standard-gauge subway equipment and slapping a pantograph on it. The point is that direct Caltrain service from Gilroy/SJ into SFO would be valuable.

    As for the overhead catenaries, the pantographs are spring-loaded. At the low vehicle speeds that would apply in this case, the usual wire system could be replaced with a rigid overhead power rail running fairly close to the roof of the vehicle.

    I'm not suggesting that Caltrain equip any of its trains with a third rail pickup unless there is both interest and opportunity to run Caltrain service through the transbay tube as a "guest" of BART's. That would give peninsula commuters direct service into/out of downtown SF and downtown Oakland long before the DTX tunnel is completed. I don't see that happening, though.

    Wrt BART parking, there might be a need for a new parking structure at San Bruno if service to Millbrae were discontinued. E.g. at the Southern edge of Tanforan, next to I-380. They'd have to start charging for parking. Shoppers would get an hour or two of free parking if they spend enough money.

  44. Rafael, I think you underestimate the difficulty of integrating differences in track gauge, loading gauge, weights, signal systems, traction systems, etc.

    All for what? Preserving the idiotic two-change system to get from the peninsula rails to your airport terminal?

    "Rupture de charge" is what we are trying to avoid, and running Caltrain into the SFO BART station won't do that. Running Airtrain to Millbrae will.

  45. Millbrae: BART doesn't need 3 platforms.

    15 minute headway can easily be operated to a single BART platform (the furthest east) IF -- if! -- there is any use in operating BART to Millbrae at all.

    The most logical plan is simply to abandon BART south of the the SFIA station. (Thanks for the $200+ million wasted, Quentin Kopp!)

    The existing, extraordinarily expensive BART guideway from SFIA on the southern leg of the wye would be reused by the airport people mover, itself extended to the Caltrain/HSR Millbrae station (preferably at an elevated level, to minimize stupid ups-and-downs for transfers). It probably makes sense for this to run at and above grade -- this is in an active surface rail ROW after all -- rather than perpetuate use of the BART cost-plus-plus-plus-plus tunnel.

    If technology-inappropriate, wasteful urban subway BART service to Millbrae is to be retained for some non-economic, non-operational, purely political reason, that can be done via the existing (100% unnecessary) tunnels and to a SINGLE platform at grade on the eastern side of Millbrae station. People mover from SFIA above, Caltrain/HSR (four tracks, two island platforms) to the west.

    BTW, Here's what an airport-external truly multi-modal station looks like when certified morons don't design it: El Prat

    A few highlights:

    P9. The mode split of "l'estaciĆ³ intermodal del Baix Llobregat" (rolls off the tongue just like "Millbrae Kopp Memorial Pan-Galactic Intermodal", no?)
    is 30% rail, 30% taxi, 9% bus, 24% private automobile, 9% kiss+ride.

    p 119. 642 parking spaces (possible future expansion to 912 -- compare 3000!! at Millbrae.)

    p 130. Nice overview map of Barcelona airport (El Prat), El Prat intermodal station, central Barcelona, future L1 metro line extension, future L9 metro, existing R2/R10 commuter rail.

    p 133. Commuter/HSR level cross-section.
    Four tracks, four platforms faces for commuter.
    Four tracks, two platform faces, two express tracks for HSR.

    p 134. Commuter/HSR/metro interconnections plan.
    p 138. Plan detail.
    p 141. Section.
    Compare and contrast with ultra-convenient and ultra-fast Millbrae Parking Fortress intermodality.

    p 147. Future roadway (+ metro line) overview

    p 152. Taxi(!!!!!!!) + kiss&ride + coach level.

    It's all pretty straightforward to do this stuff right of anybody but grotesquely corrupt sub-morons have any design input. Just hire some Spaniards. Or Germans, Or Swiss. Or French. Or Swedes. Or Japanese. Or Norwegians. Or Koreans. Or ...

    Best of luck!

  46. Millbrae is a mess, isn't it?

    The most straightforward solution is to reduce BART's track count at Millbrae. It can run out of one track (easternmost) or two tracks (easternmost and second-easternmost). At least the westernmost BART track should be converted to Caltrain. Add a track to the west of the westernmost Caltrain platform and you have four Caltrain tracks.

    HSR needs more work because it needs straighter tracks.

    However, again this is a case where running the express tracks on the west side makes sense. It makes sense here, it makes sense in the entire San Carlos area, and I think it's probably worth considering on a broader basis.

  47. I think a good solution would be to take the western most track back from BART so BART would have the two eastern ones. Then another track would be added on the western part of the station. This would allow the express HSR trains to continue straight through the station without having to curve. The local trains' tracks would branch out before the station but that wouldn't matter because they were stopping anyways. Also, using the existing Caltrain tracks as the center express tracks would allow for Express to Local transfers at the station if the express trains did stop. I just don't think there is enough room for an additional 2 tracks to the west of the station. Adding in one on either side allows the express tracks to remain in the center (as it will be the rest of the route) without having to build expensive aerial structures.

  48. At SwitchingModes there is an article that discusses why going to Millbrae is not right for HSR. Instead, HSR should have a terminal in SFO.

  49. Some posters may have already stated the importance of continuing to have BART serve the Millbrae station. HSR is far more apt to succeed if there are good mass transit connections at its stations. Millbrae BART provides high speed passengers with an easy connection to northern San Mateo County and southern San Francisco (to stations and neighborhoods not served by Caltrain). As faulty as the original planning was, since Millbrae BART is in place it should be used. (Although a direct AirTrain connection to Millbrae, using BART's ROW if possible, would be the missing link to better connectivity.)

  50. https://hsr.ca.gov/docs/programs/san_francisco_san_jose/Draft_EIRS_FJ_V3-05_Alternative_A_Book_A3_Stations.pdf

    CAHSRA's latest station plan from DEIR. The tunnel track/platform in the previous plans is gone, instead it seems the station west of the Caltrain/BART platform will be rebuilt, adding one island platform and one side platform with 4 platform faces for Caltrain/HSR.