12 December 2015

Optimizing the Midline Overtake

The most important piece of infrastructure required for "blending" HSR with Caltrain on the peninsula rail corridor is an overtake facility, basically a several-mile long stretch of up to four tracks that will enable faster trains to overtake slower trains.  Caltrain studies have shown that the best performance (measured by robustness to cascading delays) can be achieved with a midline overtake  from San Mateo all the way through Redwood City.  Preliminary engineering and environmental clearance for this infrastructure is now resuming, with the recent award of a $36 million contract by the CHSRA to engineering firm HNTB.

The baseline configuration for this overtake is described in Caltrain's blended operations analysis, and assumes rebuilt stations at Hayward Park, Hillsdale, Belmont, San Carlos and Redwood City with four tracks and outside platforms in the tried and true style of the 1930s Pennsylvania Railroad. High-speed trains would use the center pair of tracks to overtake slower trains on the outside pair of tracks, as shown below (click figure to enlarge):

For comparison, the figure also shows a better solution that ensures the highest level of punctuality for all trains using the corridor.  This optimized overtake configuration differs from the baseline configuration as follows:
  • The slow trains run in the middle, so that disruptions to local commuter service (for example, when an incident blocks a track for hours) do not disrupt high-speed service when commuter trains are re-routed around the incident location.  This track configuration is known as Fast-Slow-Slow-Fast (FSSF) as opposed to the traditional SFFS.  Real world examples of FSSF can be seen in train cab videos from Sweden and Australia.
  • All commuter stations are built with central island platforms.  This allows commuter trains to use either platform face without confusing passengers, and requires only one set of station amenities (shelters, elevators, escalators, stairs, ticket vending machines, PA systems, train arrival screens, lighting, benches, etc.) because there is only one platform.
  • A major new interchange station at Redwood City, with four platform tracks and additional train storage sidings.  This station (described below) would serve as a transfer point for HSR, Caltrain and future Dumbarton trains, as well as non-rail transportation modes.
  • A carefully planned future-proofed high-speed rail junction where the Dumbarton rail corridor meets the peninsula rail corridor, preparing for the inevitable arrival of passenger rail service across the Bay.
Here's how it would ideally play out.

Short Term: the San Mateo Grade Separation

Preliminary rendering of new
Hillsdale station with island platform
The next step in the decadal process of grade separating the peninsula rail corridor will soon begin in San Mateo.  A new $180 million grade separation project is in the final stages of planning for 25th Avenue (currently a grade crossing) as well as 28th and 31st Avenues (currently not connected).  Concurrently with this project, the busy Hillsdale station will be moved a bit north of its current location and turned into an island platform.

This project is caught in an interesting political bind.  There is on one hand a rush to complete it by 2019 before Caltrain's electrification project, to minimize disruptions to Caltrain service.  On the other hand, due to its strategic location, this project will form a key building block of the blended system with HSR, which still needs to be environmentally cleared.  It is a near certainty that this portion of the corridor will require four tracks to enable trains to overtake each other, but any attempt to design and build it as such is likely to run afoul of HSR opponents who will accuse the CHSRA of advancing their project through CEQA piece-mealing.

The southern San Mateo grade separation design will have to be very carefully considered to preserve the ability to add two additional tracks with as little disruption as possible.  Road underpass profiles and bridge abutments should be designed for four tracks, as should the elevated structure that will support the tracks.

The choice of an island platform configuration for the new Hillsdale station is either a sneaky way to build a wide four-track embankment in preparation for yet another new station with SFFS outside platforms, or is an excellent choice for FSSF because it allows future tracks to be added without rebuilding the station for a second time.  One hopes the station access (stairs, ramps, etc.) will be designed to allow the platform height to be raised easily from 8" to 50".  An intelligently designed San Mateo grade separation would atone for the terrible failures of the San Bruno grade separation, designed with great hostility towards higher speeds or additional tracks.

Medium Term: Redwood City HSR Station

Amsterdam Bijlmer (photo by tataAnne)
could just as well be the future
Redwood City train station.
While the CHSRA's plans for a mid-peninsula stop have been shrouded with ambiguity for several years, Redwood City stands out as a more optimal location for a new HSR station than Palo Alto or Mountain View, the other two locations in the running.  Unlike its neighbors to the south, Redwood City favors strong urban growth, has a large amount of railroad land available, and is reasonably well-connected to the existing road and transit network.  Redeveloping the antiquated but popular Sequoia Station shopping center would enable the construction of an elevated four-track station with plenty of capacity to support not just HSR but also Caltrain cross-platform connections and future Dumbarton service.

The station complex would feature two shared (HSR or Caltrain) 400-meter island platforms centered between Broadway and Brewster, easily accessible from both streets.  Bus connections would be conveniently located under the station. To the north of the platforms, a pocket turnback track would allow Dumbarton trains to reverse without fouling other traffic.  Similarly, to the south of the platforms, another pocket turnback track would allow southbound Caltrain locals to terminate in Redwood City before turning northwards again to serve the densely-spaced stations of San Mateo County, allowing Caltrain to serve more passengers with fewer trains and crews.  Thanks to the FSSF configuration, all this to-and-fro by commuter trains would stay well out of the way of HSR.

This would be a large train station and quite a tight fit (if you're curious about exactly how large and how tight, download this KML file into Google Earth to view the station footprint and track layout).  It would be a big change for Redwood City, but with a huge payoff: the tracks would no longer form a barrier through town, and the Sequoia Station shopping center would be merged with the station to form a gateway and a destination in its own right that is connected to downtown.  HSR service could make the city a very desirable location for business.  The new station could become the centerpiece of the ambitious revitalization strategy described in Redwood City's downtown precise plan.  But this idea is not without pitfalls, as the size of the station could be compared to plonking a couple of Nimitz-class aircraft carriers in the middle of town.  To use the tired slogan, it needs to be done right.

Medium Term: a New Fair Oaks Station

Approximate location of new Fair Oaks
station island platform, view to northeast.
Overtake would extend just beyond
platform to the right (south).

The key to reliable overtaking on a multiple-track railroad is to ensure that the average speed of the slower train being overtaken is sufficiently slower than the average speed of the faster overtaking train.  One of the ways of ensuring a good speed differential is to have the slower train make station stops that the faster train doesn't; each station stop is worth about 2.5 minutes.  Therefore, locating stations on the four-track overtake section is helpful.

This brings us to a lemon of a station immediately south of the midline overtake: Atherton.  Located in an area of very sparse population and jobs density, Atherton should be permanently closed.  This closure would come not only as a show of appreciation commensurate with the town's support of Caltrain modernization, but especially because census data shows clearly that Atherton is precisely where you would never place a train station.

To replace Atherton, a new Fair Oaks station should be built just 0.6 miles to the north, at the 5th Avenue grade separation.  The overtake section would be extended a bit southwards, just beyond the station, enabling locals to be passed while stopped at the central island platform that can be accessed from either side of 5th Avenue.  The new Fair Oaks stop would be equidistant from Redwood City and Menlo Park, and located in an area with very high population density that could support thriving ridership, in contrast to Atherton.

Longer Term: a Seamless Dumbarton Connection

Dumbarton rail has been an uncertain prospect for decades, with some political backing but insufficient funding.  While it may take another few decades for the money and the will to finally materialize, large concentrations of employment and the need for additional transbay corridor capacity make some form of passenger rail service inevitable.  The Dumbarton corridor also happens to be ideally suited for high-speed rail.

The key node is Dumbarton Junction, which should be reconfigured in such a way that trains can enter and leave the peninsula rail corridor swiftly and seamlessly.  This will likely involve a flyover track, enabling southbound trains to enter the Dumbarton corridor without crossing (and therefore blocking) any of the northbound tracks.  To minimize the altitude of the flyover, the Rte 84 / Woodside Road overpass would be turned into an underpass.  As for the Redwood City station, the fit would be quite tight with a 90-foot corridor width where the flyover track begins.

While the flyover may seem like an expensive solution to a problem we don't yet have, planning for it now (if not actually building it) will save money in the long run when passenger rail service grows.

Design Values

No matter how the midline overtake is ultimately configured, it must reflect design values that are clearly articulated.  One of these values should be compatibility between Caltrain and HSR.  It's not enough to talk about the "blended system" without actually taking the steps to make the two systems seamlessly interoperable, allowing any train to use any track to serve any platform.  This means no tracks can be dedicated to one operator at the exclusion of another.  Everything must be shared, including the platforms at the new mid-peninsula station.  This sharing contributes to another important value, robustness to service disruptions.  The fast-slow-slow-fast track layout is the key to ensuring that a commuter train delayed in Belmont won't create a statewide domino effect that eventually makes a train late in Los Angeles.  A third important value is future-proofing.  Infrastructure like the midline overtake will define what is possible (and not) for generations to come.  It would be short-sighted not to plan for a fast and seamless connection to the Dumbarton corridor, even if its future use isn't well-defined today.

The midline overtake is the key to an effective blended system.  When evaluating its design, ask yourself: is it compatible?  Is it robust?  Is it future-proof?

13 November 2015

The U-Shaped Grade Separation

While some cities and towns on the peninsula are still holding out for trenches or tunnels to bury the railroad tracks out of sight, the astronomical cost and difficulty of constructing such structures below the water table in seismically unstable soils makes it likely that above-ground solutions will ultimately prevail, anywhere rail traffic needs to be separated from road traffic.  An attractive above-ground solution is the U-shaped grade separation.

What is a U-shaped grade separation?

U-shape bridge cross section, showing the benefits of
reduced track elevation
A U-shaped grade separation is a type of railroad bridge used to elevate the tracks above road traffic with as few community impacts as possible; there are no property takes and all road turning movements are preserved.  The bridge structure consists of sections made from two pre-stressed concrete side beams, forming the two sides of a U shape, connected by a flat slab forming the bottom of the U, on which the tracks are laid.  The side beams bear the bending loads from the weight of the bridge and the trains that it carries.  This is not a typical railroad bridge design; it is a specialized configuration used to quickly and efficiently build elevated urban metros in cities where these systems are being built from scratch in a densely built environment.  The concept is further explained in a paper and a patent.

While the peninsula rail corridor is not a new metro system, these U-shaped structures could still prove useful in a major push to grade-separate the 40 grade crossings that remain, enabling higher speeds and more train traffic while relieving road congestion and improving east-west access across the tracks.

What are the advantages of U-shaped grade separations?

U-shaped grade separations combine several attractive features that make them ideally suited for developed areas along the peninsula rail corridor, and certainly much better than the massive hollow core concrete box girder bridges considered standard issue by the HSR project as shown in the graphical comparison above.
  • Lower track elevation.  The U shape minimizes the depth of the structure (measured from the underside of the bridge span to the top of the rails) to 3 feet or less.  This allows the standard 16-foot road clearance to be provided by raising the tracks just 19 feet above the road surface, about 8 feet less than the large elevated concrete box-girder viaducts that were proposed during the 2010 Analysis of Alternatives for peninsula HSR.  The rails are lowered thanks to the U shape, which places the structural support of the bridge to the sides, rather than under the trains.
  • Lower visual impacts.  When the tracks don't need to rise as much, the rail approaches to a grade separation become correspondingly shorter and less obtrusive, impacting fewer views. The structures above rail level, such as overhead electrification poles, are also lowered.  This reduces the so-called "Berlin Wall" effect of a grade separation structure.
  • Lower train noise.  The side beams function as natural sound walls, trapping rail noise before it has a chance to escape into adjacent neighborhoods.  They are especially effective because they are thick and quite close to the train.  This obviates the need to add sound walls on top of the bridge, making the finished structure less visually obtrusive.
  • Better earthquake resistance.  The lower profile of the bridge structure reduces bending moments applied to the piers and foundations, whether by earthquake forces or train braking and acceleration or wind loads.  This makes the bridge piers less massive and integrates them better into the built environment.
  • Better station integration.  Where stations must be located on an elevated section, structures are simplified thanks to the lower profile of the track, which reduces the reach of stairs, ramps, escalators or elevators, making for a more passenger-friendly environment.  The side beams of a U-shaped viaduct have their top flange at the same height as the train floor and form the actual platform interface, 50 inches above the rail and 72 inches from the track center line, allowing the U-shaped structure to continue uninterrupted through the station.
  • Better safety in case of derailment.  The side beams are close to the train.  In case of a derailment, train cars will be guided by the structure and will not topple off the bridge.  This feature is known as "derailment containment."
  • Lower construction cost.  U-shaped elements can be prefabricated off-site and assembled with minimal disruption compared to traditional cast-in-place construction methods.  Using standardized elements throughout the corridor, in dozens of locations, provides economies of scale.  The decreased profile changes for both rail and road (whether the U-shaped bridge is elevated or at-grade with the road sunk underneath) require less excavation or fill.
The U-shaped design can minimize property takes, preserve turning movements for cars and trucks, cost much less to build than below-grade solutions, and tread more lightly through built-up neighborhoods than a conventional (box beam) viaduct or split-grade separation.  U-shaped bridges are ideal for grade separation in dense areas like the peninsula.

26 August 2015

Level Boarding: It's Official

A montage of what a Caltrain EMU might
look like, before platforms are raised.
Based on a photo by Yevgeny Gromov
Caltrain just released the final Request For Proposals for their new electric train fleet.  Train manufacturers will now prepare detailed bid packages, and in early 2016 a winner will be selected to build an initial fleet of 15  electric trains that will enter into commuter service in 2021.

Several changes were made to this document after the draft RFP was circulated for industry review.  The single most remarkable change is that level boarding and platform sharing with high-speed rail is now a firm requirement, instead of an option suggested by stakeholders.  These are the words from section
CHSRA trains will run over the same alignment and stop at some of the same stations as JPB trains. The bi-level EMU must therefore have the same interface with the infrastructure as the future High Speed Rail cars, including clearance envelope, and platform boarding height.

JPB plans to raise platform heights to approximately 50.5-50.75” ATOR (to interface with a vehicle threshold height of 51” ATOR), initially at San Francisco, Millbrae, and San Jose stations. Other station platforms on the JPB system may ultimately be raised to the same level.
This is not only an endorsement of level boarding.  It is an endorsement of complete integration with high-speed rail including not just shared tracks but also shared stations.  It is a major step forward for riders and taxpayers, because it will increase the speed, efficiency and usability of Caltrain at the same time as it makes high-speed rail more affordable.  It will help bring to California what Europeans take for granted.

The New Platform Interface

Section 3.3.3 of the RFP details Caltrain's new high platform interface:
  • Platform height: 50.5 - 50.75 inches
  • Platform side clearance: 72 inches from track center line
  • Maximum boarding gap: < 1.5 inches horizontally, < 5/8 inches vertically
Caltrain has entirely dropped the previous plan to implement level boarding at a height of 25 inches, which would not have been compatible with high-speed rail and would have created significant complications in the station infrastructure served by both systems.

Dual Height Doors

The new EMUs will have two sets of doors, not just as an option but as a non-negotiable requirement.  The RFP describes the configuration in section 12:
Each vehicle shall have eight door openings, four on each side of the vehicle, directly across from each other. One set of four shall be located just inboard of the trucks and the other four above the trucks. The set located inboard of the trucks (the low level set) shall be compatible with JPB's existing platform height and existing mini-highs. The set located above the trucks (the high level set) shall be compatible with JPB's future high level platforms.
A large number of bikes (at an 8:1 ratio of seats to bikes) will be stored on the lower level of two cars per train.  They will access the high doors using wheel ramps built into the stairs between the lower bike level and mid level vestibule of the train.

While this is a rather unique configuration, no other train operator worldwide has had to plan for a system-wide platform height transition of more than four feet of vertical change.  For such a large height transition, it makes perfect sense to use the vehicles as a tool to enable the flexible and independent reconfiguration of each individual platform, without imposing system-wide construction schedule or funding constraints.  It is an unusual but quite logical solution to an unusual problem.

The upper set of doors, which will provide level boarding at new high platforms, will feature retractable door threshold extenders, to bridge the gap between the train and the platform.  These are described in section 12.2.12 of the RFP.

Looking Ahead to a Well-Blended System

Caltrain has come a long way on the issue of level boarding and blending with high-speed rail.  A key architectural decision has now been made that will ensure the future success of the blended system.  In the 2020s, Caltrain passengers of all abilities won't give a second thought to the seamless experience of boarding a train, and will take for granted the brevity and punctuality of station stops.  Meanwhile, a few train nerds will photograph the platform interface.

In the meantime, three cheers for compatibility!

08 August 2015

Peninsula HSR, Take Two

Environmental clearance of high-speed rail in the peninsula rail corridor was initiated right after the Proposition 1A bond passed in 2008.  The development of engineering and environmental documentation for a four-track alignment connecting San Francisco to San Jose was in full swing during the years 2009 and 2010.  The Peninsula Rail Program, as it became known, was an ambitious yet awkward collaboration between the CHSRA and Caltrain, with the engineering consulting firm HNTB doing most of the heavy lifting.  The decision to concentrate HSR resources in the Central Valley, combined with fierce community opposition on the peninsula, brought the process nearly to a halt in early 2011.  By that time, thousands of pages of documents had been drafted, hundreds of stakeholder meetings held, and $45 million spent for preliminary engineering and environmental clearance.

Then came a long pause during which two major developments took place.  First, as a result of a political compromise, the idea of a four-track high-speed railroad was dropped in favor of a "blended system" where Caltrain and HSR would share the peninsula corridor primarily on two tracks, with less impact to surrounding communities.  Second, the Caltrain electrification project came closer to being realized, passing key milestones of environmental clearance, funding, and procurement.  Throughout this pause, plans for peninsula HSR became somewhat nebulous, both in their scope and timing.  The media spotlight turned away.

Fast forward to the August 2015 meeting of the California High-Speed Rail Authority (YouTube video).  As reported by the Fresno Bee (and without a peep from the Bay Area press) we have the first hints of what lies ahead, in the form of a Request For Qualifications issued by the CHSRA to re-start the environmental clearance process for the peninsula.  This RFQ lays out a new timeline and a $36 million budget allocated over a term of three years, shared between two sections: San Francisco to San Jose, and San Jose across Pacheco Pass to the Central Valley wye.  The following schedule milestones are envisioned:
  • Consultant contract award - November 2015
  • Project scoping - March 2016
  • Preliminary design for project definition - May 2016
  • Technical reports - June 2016
  • Administrative draft EIR/EIS - August 2016
  • Draft EIR/EIS release - November 2016
  • Preliminary design of preferred alternative - April 2017
  • Final EIR/EIS certification - November/December 2017
The timeline for actual construction is not specified, but it rarely begins immediately after EIR certification, since final design and the inevitable CEQA lawsuits take time.

Highlights from the RFQ

The RFQ deliberately does not reveal the scope or exact nature of the alternatives to be studied, but it does contain some interesting nuggets:
  1. Work for restarting the peninsula HSR process has already started, as noted on PDF page 24: "Work on some of the tasks listed in Exhibit A of Attachment C has commenced and is currently being performed by Authority and Rail Delivery Partner staff."  What the blended system will look like is already being hammered out.
  2. The RFQ emphasizes that the proposed EIR is separate from Caltrain's electrification EIR, on PDF page 28: "On January 8, 2015, the JPB certified the PCEP Final EIR and is currently in the process of procuring a design/build contract to implement the project. While the PCEP will not include all infrastructure necessary to implement HSR service in the SF-CVY Corridor (such as HSR maintenance facilities, station platform improvements, track straightening, or passing tracks), the electrification infrastructure (such as overhead wire systems), along with additional infrastructure improvements, will accommodate future coordinated service and will not preclude HSR."  This point is the subject of a CEQA lawsuit against Caltrain, claiming that electrification is an inseparable component of the HSR project.
  3. Phased implementation is described on PDF page 57: "The Consultant shall develop an incremental plan as directed by the Authority to construct the project over a phased implementation schedule, dependent on funding. The Consultant shall recommend appropriate construction elements for each increment of implementation. This plan shall identify operable project segments or elements of the HSR infrastructure (such as grade separations) that could be constructed early and bring near-term project benefits to existing freight rail and conventional passenger rail services, as well as other increments of construction to build out the full set of improvements over a phased implementation plan."  The peninsula corridor is uniquely suited to a number of construction packages to be built independently from each other.
  4. San Jose is no longer an artificial boundary between two project sections.  This has been a weakness in the past, with insufficient coordination to optimize the configuration of the station and its approaches because each end was being handled by a different consultant.  With the same consultant handling both ends of San Jose, sanity may finally prevail with a shared at-grade solution.
  5. Level boarding planned for Caltrain, on PDF page 60: "Platform design for level boarding at all Caltrain stations will be required."  Even if not at the same height as selected for HSR, level boarding is a prerequisite for the blended system, to improve the average speed and punctuality of Caltrain.
  6. A temporary San Francisco terminal is planned at 4th and King.  The mere idea of it illustrates the frosty relationship between the CHSRA and San Francisco's TJPA, but also helps to satisfy the requirement for a 30-minute trip from San Francisco to San Jose, a threshold of great legal significance that is embedded in the Proposition 1A bond act.  Starting from 4th and King, rather than from the Transbay Transit Center, running at no more than 110 mph, and counting only pure run time (with no timetable margin), the 30-minute run becomes feasible.
EIR Cost Magnitudes

Environmental Impact Reports are extremely complex and voluminous documents designed to clear a project under the California Environmental Quality Act, ensuring that impacts are properly disclosed and mitigated.  It takes a large team of engineers, environmental specialists, writers and lawyers to concurrently design a project and pull together an EIR that can pass legal muster without incurring years of litigation.  To understand exactly where the process currently stands for peninsula HSR, it helps to remember that the published record for the San Francisco to San Jose project section forms only the tip of the iceberg.  The vast majority of the material assembled by HNTB in 2009 and 2010 remains unpublished, to be continued by this new contract.

How much EIR preparation did the $45 million spent so far buy?  We can establish an extremely crude metric for the cost of one EIR page by taking the ratio of the cumulative cost incurred for the preliminary design and environmental clearance of a project, as of the time of EIR certification, divided by the total number of pages in the resulting EIR.  Here are some examples:

Project Certification Cost Incurred Page Count Cost Per EIR Page
Merced - Fresno HSR May 2012 $45M 13,000 $3500
Fresno - Bakersfield HSR May 2014 $120M 20,000 $6000
Caltrain Electrification Jan 2015 $14M 5,400 $2600
Peninsula HSR SF - SJ Dec 2017 $65M* 13,000** $5000***

*cost basis $45M expended to date + $20M of the new $36M contract
**estimated based on cost per page
***estimated based on past history and biased high for scope change from full build to blended

Given that the new consultant won't be starting from scratch, it's conceivable that there will be sufficient budget in the new contract to produce a full EIR for the blended system on the compressed two-year timeline envisioned in the RFQ.

What the Blended System Might Look Like

The CHSRA and Caltrain take great pains to remind everyone that we won't know what the blended system for the 50+ mile peninsula corridor will look like, nor what the blended service plan will be, that is, until the Alternatives Analysis is released next year.  The specific discussions regarding the scope of the blended system are underway behind closed doors.  Taking into account the phased and incremental nature of the project, one can engage in some informed (wishful?) speculation, listed from north to south:
  • 4th and King shared station modifications.
  • Brisbane HSR maintenance facility.
  • Millbrae shared station modifications, hopefully with an affordable shared at-grade solution.
  • San Mateo County grade separation Phase II at Linden Ave in San Bruno, Center St in Millbrae, Broadway in Burlingame, and 25th / 28th / 31st in San Mateo.  The latter are likely to happen sooner than the other projects to enable the mid-line overtake.
  • Grade separation through highly constrained downtown San Mateo.
  • Four-track 110 mph mid-line overtake facility (from San Mateo 9th St, through Belmont and San Carlos, initially to Whipple in Redwood City).
  • Redwood City grade separation Phase IV, extending the four-track mid-line overtake through downtown, possibly with a new HSR station replacing the Sequoia Shopping Center, if the city and CHSRA agree to add this to the project scope.
  • PAMPA (Palo Alto Menlo Park Atherton) grade separations, likely to happen later than the other projects.
  • Santa Clara County grade separation Phase III at Charleston and Meadow in Palo Alto, Rengstorff and Castro in Mountain View, Mary and Sunnyvale Ave in Sunnyvale, creating a continuous 14-mile stretch of grade-separated track good for 110 mph from Palo Alto to San Jose.
  • San Jose approach realignment and a shared ground-level station.
  • A three-track at-grade alignment through San Jose's Gardiner neighborhood, along the existing right-of-way, avoiding a slow and expensive viaduct above the 87/280 interchange.
  • Curve flattening throughout the peninsula, except (unfortunately) in San Bruno
  • Level boarding across the entire Caltrain system, a key blending ingredient that ensures commuter trains can clear the shared tracks quickly and reliably in front of high-speed trains.
The next formal step in the process will be a new Notice of Preparation (NOP) to be published by the FRA in the Federal Register, an action that could come in the coming months.  Then we'll party like it's 2009.

18 July 2015

News Roundup, July 2015

Cost of Dual Height Boarding: with its industry review of the EMU draft RFP, Caltrain sought feedback from vehicle manufacturers regarding the cost and feasibility of delivering vehicles with dual boarding height capability.  According to the latest EMU procurement update, the feedback received indicates that vehicle cost would increase by just 3 to 5 percent.  This small premium (roughly $20 million) all but ensures that the HSR project will be able to pay for this important compatibility feature.

New Rules for Electrification: the California Public Utilities Commission has released General Order 176, the Rules for Overhead 25 kV Railroad Electrification Systems for a High-Speed Rail System.  The new GO, effective as of 26 March 2015, will also serve as the regulatory framework for Caltrain's electrification project, despite the peninsula corridor not qualifying as "high-speed rail" as narrowly defined in the document.  Three major issues remain to be hammered out with regulators and freight railroads: (1) grade crossing warning systems, (2) vertical clearances, and (3) freight personnel safety and training for operating freight trains in electrified territory.  These three issues are minor and unlikely to require a new rule-making process.  The release of GO 176 is timely for Caltrain's electrification project.

Cap and Trade Maneuvering: at least two lines of attack are being pursued by opponents of Caltrain electrification and the high-speed rail project.  First, there is or will be legal action that seeks to deny the use of Proposition 1A HSR bond funding to pay for Caltrain electrification, based on the (quite defensible) argument that electrification isn't high-speed rail and won't meet the legal restrictions of the bond measure. Prop 1A high-speed rail funding accounts for the lion's share of the funding package for electrification, a contribution of $600 million.  The underlying calculus is that denying this funding would kill the electrification project. Second, there is or will be legal action that seeks to deny the use of Cap and Trade funding to pay for high-speed rail, based on the (quite defensible) argument that the greenhouse gas reductions from HSR will only occur far in the future, well beyond the time frame required by CnT legislation.  Both of these legal challenges can be neutralized in one fell swoop, by substituting HSR CnT funds for the HSR Prop 1A funds. According to Caltrain's EIR, electrification will reduce greenhouse gas emissions by 80,000 metric tons per year initially, increasing to 190,000 metric tons by 2040, largely by cutting automobile traffic.  It doesn't get much more short-term than that, and CnT funds come with far fewer strings attached than Prop 1A funds.  Look for the funding swap to occur this fall, when the nine-party MOU is revised to reflect the growing electrification budget.

HSR Business Plan Machinations: due to the lack of funding to build the extremely expensive mountain crossings, there are indications that the 2016 business plan for HSR will call for service to begin in the SF and LA areas several years before the Central Valley is linked to anything.  This should be of some concern to Caltrain because it would put HSR in direct competition with Caltrain for affluent tech commuters, no matter what they say.  If Caltrain is elbowed out of the lucrative express market, the loss of revenue will be entirely HSR's gain.  While this Bay Area mini-HSR might show an operating profit (as required by law), it could only do so with a hidden subsidy, provided in the form of extremely scarce and valuable rush-hour track capacity.  Competition is great, but the market mechanisms for sharing the peninsula rail corridor "fairly" (whatever "fair" means to each stakeholder) would need to be carefully developed.  Then again, this silly idea might just wither on the vine, since the letters 'H' and 'S' would be absent from HSR.

14 June 2015


Caltrain recently published their Electric Multiple Unit (EMU) Request for Proposals (RFP) in draft form, to obtain feedback from potential proposers prior to the formal issue of this document in August 2015.  Here are some initial impressions:

What's Pleasantly Surprising:
  • Level boarding is a serious consideration throughout the RFP.  Caltrain appears to recognize the importance of this issue: it's the Next Big Thing after electrification, and cannot wait until the next round of vehicle replacement in the 2050s.  The RFP shows that Caltrain is starting to walk the walk, not just talk the talk.
  • Platform sharing and full blending with high-speed rail is firmly on the agenda with "Option B," a dual boarding height train that enables an eventual transition to level boarding at 48 - 51" height.  This solution has been described and advocated on this blog as the best one available given the constraints of the problem.  Most people who reject this solution ultimately take issue with one of the constraints, but once you accept these constraints, "Option B" starts to make more sense.  Proposers are asked in the draft RFP to further evaluate its feasibility and cost impact.
  • In section 3.3.3, future capability for level boarding is described as ADA-compliant with a 3-inch maximum horizontal gap and a 5/8-inch maximum height mismatch.  Section 4.2.3 specifies a pneumatic leveling suspension to meet the vertical tolerance. It looks like Caltrain  is going the extra inch after all, which is commendable.
  • The vehicle static envelope in Appendix C is a full 3.4 meters wide, enabling the use of extra-wide trains, wider than the conventional AAR plates.  While Caltrain doesn't seem keen on 3+2 seating (something about the "middle seat" situation), extra width that takes full advantage of the static envelope is good for all uses besides seating.  Let's hope the proposers aren't timid about this.
  • While it isn't immediately obvious from looking at the raw numbers, the run times required in section are fairly aggressive and will require some sporty power-to-weight ratios.  This high level of acceleration and braking performance is good for the blended system and will ensure that the most can be made of shared tracks.
  • In section, door controls are required to be installed in the driving cab, offering the possibility of consolidating and automating a task currently performed by conductors.
  • In section 14.4, an automatic station announcement system is specified to automate a task currently performed by conductors.  This isn't exactly surprising for a modern train, but Caltrain and its labor practices sometimes seem stuck back in the 20th century.
  • In section 14.9, the trains are required to be pre-wired (if not yet equipped) for passenger Wi-Fi.  For a rail system that serves Silicon Valley, that's overdue by easily a decade.
  • No trap doors!  That always seemed like a horrible way to achieve level boarding.
What's Disappointing:
  • The RFP is highly prescriptive.  The cost of preparing such a voluminous requirements specification, and then to formally verify such a large number of requirements, will easily run into the millions of dollars.  So much for off-the-shelf procurement.
  • In section 2, no allowance is made for articulated trains (e.g. Bombardier Omneo) or for trains with mixed bi-level and single level arrangements (e.g. Siemens Desiro HC).  The desired EMU is prescribed as a set of traditional bilevel cars each about 85 feet long.  A better approach would be to define a minimum passenger capacity per unit length, leaving more room for creative and unconventional interior layouts.
  • Still no detailed thought appears to have gone into how to transition to level boarding under "Option A".  The height of 24 or 25 inches is intended to match Caltrain's existing Bombardier bilevel cars, which is a bit mystifying since there is no feasible transition to level boarding using these cars.  Section 3.3.3 basically asks for proposers to figure it out for Caltrain: "Heights below 24 inches will be considered for future level boarding if the Contractor can demonstrate conclusively that the height is advantageous for JPB’s envisioned service and compatible during the transition to that envisioned service."
  • Section 12 requires all doors of the train to open at every stop, as they do today.  To reduce wear and tear on the door mechanisms, modern vehicles often come equipped with interior and exterior push buttons for passengers to initiate door opening.  Instead of opening the doors by default, the crew-operated door control station should de-inhibit the doors to be opened only as requested by passengers at each individual door.
  • Section 2 envisions that trains would be lengthened from 6 cars (~150 m) to 8 cars (~200 m) at some future date.  This is far too timid a capacity expansion.  A more flexible and future-proof approach would be to order more 150 m trains and double them up-- after the necessary platform extensions are constructed at stations with the highest ridership.
What's Weird:
  • Section 2 requires shorter 6-car trains to have their performance de-rated (by software) to the same performance as a longer 8-car train.
  • Section 23.2.2 requires the delivery of 200 (two hundred!) 1/50th scale models of each car type for "internal and external distribution"... That's some serious schwag!
What's Missing:
  • "Option B" with dual boarding levels requires level boarding at 48 to 51 inches ATOR but fails to describe the basic platform interface dimensions, including height above rail, offset from track center, and tolerances thereon.  Proposers are unlikely to be able to design against such a critical interface when it hasn't even been defined.  This data should be agreed upon with the California HSR Authority, after some technical decision making that may have to occur sooner than they would like.
  • In section 3.3.3, Caltrain requires that "the entire platform interface system must also be usable during the transition from the current platform height to the level boarding platform height," a stealth requirement that makes "Option A" trains at least as mechanically complex as the "Option B" trains, by requiring boarding capability at two different heights (8-inch legacy and 25 inches with ADA-compliant level boarding).  A requirement of such great importance and design impact ought to be made more explicit, saying what it actually means and using the word "shall".
  • The door control system (section 12) envisions 100% manual operation of the doors by train crews.  This may not properly address the challenge of operating during a platform height transition, when each individual platform may need to be raised in successive construction phases to avoid closing the station or doubling the construction footprint for temporary platforms.  With manual door operation, the risk of human error resulting in opening a door at the wrong platform height will likely be unacceptable, particularly to regulators such as the CPUC.  It may be warranted, at the cost of some additional complexity, to require a platform sensing system that automatically inhibits door opening when the incorrect height is sensed, preventing crew errors and potential passenger injury.
Please use the comment section to add your own review of this document.

24 May 2015

Going the Extra Inch

Assisted level boarding on
Amtrak's Northeast Corridor
To enable wheelchair users to board a train without assistance, the Americans with Disabilities Act (ADA) requires level boarding platforms to have a maximum 3-inch horizontal gap with the train floor, and a maximum height mismatch of 5/8 inch (see 49 CFR 38.93).  When this specification is met, wheelchair users can safely and quickly board a train by simply rolling across the narrow gap between the platform and the train, giving it no more thought than when using an elevator.

Today, few commuter rail systems in the United States offer this level of accessibility.  The NCTD Sprinter (see video) may be the only one, and is often classified as "light" rail.  U.S. systems with level boarding (such as in the Northeast) have a gap of six inches or more between the platform and the train, requiring the use of bridge plates for a wheelchair user to board.  As shown in the photo above, train crew members must assist with the process of deploying the bridge plate, monitoring the wheelchair, and re-stowing the bridge plate.  Assisted boarding can easily extend station dwell times and cause delays, even on a rail system with level boarding.  These delays are allowed for by padding the time between successive trains, to prevent a delay from cascading to multiple trains.

So, just build the platforms close enough to the tracks and we're done, right?  It's not quite that easy.

Dynamic Vehicle Envelope

Dynamic envelope as defined
for Caltrain electrification
When trains are moving at speed, they can sway from side to side.  Suspension failures or shifting loads could even cause them to sag or lean to one side, requiring additional clearance between the track and nearby obstacles.  Station platforms form one such obstacle.  Caltrain's electrification RFP defines a dynamic vehicle envelope (shown at left, from page 95 of this PDF document) that encompasses the range of motion that can be expected from Caltrain's existing diesel fleet, future high-speed trains, and freight trains that use the peninsula corridor.  The dimensions of the dynamic envelope constrain how far from the track center line any future level boarding platforms would have to be set back, in order to prevent what is known as a "platform strike" from a train passing at speed.
  • 8" platforms (existing) are 64" from track center
  • 25" platforms would have to be 67" from track center
  • 30" platforms would have to be 68" from track center
  • 50" platforms would have to be 70" from track center
Meeting the ADA Gap

Unassisted level boarding in Zurich
(Siemens photo)

The vehicle envelope, because it is dynamic, forces a clearance between platforms and trains that is wider than the 3-inch ADA maximum for unassisted boarding.  To provide unassisted boarding without bridge plates, the gap can be bridged automatically by a moving step that extends from the train, a moment before the door opens.  As shown in the photo at right, this step is the key to ADA-compliant unassisted boarding for wheelchair users, and provides a more comfortable boarding interface for bicycles, strollers, luggage, and anything else with wheels.  The step retracts after the doors close, a moment before the train departs.

These gap-filling steps are quite common outside the confines of U.S. commuter rail, and all major vehicle manufacturers worldwide can provide them if the customer asks.  Video examples:
Another common and useful train feature is an automatically leveling suspension, to control the plus or minus 5/8" vertical alignment between the train floor and the platform regardless of passenger load or wheel wear.  A nice bonus of such a system is that it can measure passenger loads in real time.  This too can be provided by vehicle manufacturers if the customer asks.

Caltrain's Approach: What Gap?

Caltrain is now taking a "not to preclude" approach to level boarding, attempting to future-proof the new EMU fleet for any future decision regarding level boarding, pending the outcome of additional planning for the Caltrain / high-speed rail blended system.  This approach is largely a result of not having seriously thought about or planned for level boarding until quite recently.

[Update 6/14/2015: turns out that Caltrain's EMU RFP does require the ADA gap specs for unassisted level boarding!  My source had it wrong.]
As it turns out, Caltrain has no intention to comply with the ADA gap requirement.  Never mind the gap.  As will be apparent in the upcoming vehicle RFP, the new EMU fleet will comply with the ADA using crew-assisted boarding with bridge plates, even after level boarding platforms are built and regardless of the selected platform height.

If Caltrain fails to specify gap-filling steps and leveling suspensions for their new EMU fleet, then wheelchair users will still need crew assistance to board or alight, resulting in random and unpredictable impacts on station dwell times.  Such a failure would preclude reliable and punctual operation of the blended system, increase the amount of timetable padding between trains, and limit the capacity that can be extracted from the peninsula rail corridor before expensive and controversial infrastructure upgrades become unavoidable.  Gap-filling steps and leveling suspensions are perfect examples of small off-the-shelf features that pay off in the long run.

In order "not to preclude" an efficient blended system that extracts the highest capacity from limited infrastructure, Caltrain should require that the new EMU fleet be equipped for ADA-compliant unassisted wheelchair boarding, once new level boarding platforms become available.  Because the new fleet will be in service until the year 2050, this capability cannot be an afterthought and must be engineered into the new trains from the outset.

17 May 2015

CBOSS Headed for Trouble?

The button we may soon have
to press. Photograph by
Sander van der Vel
Caltrain's new Positive Train Control (PTC) system, known as CBOSS, appears to be running into serious technical difficulties just as the program enters its most challenging phase: testing and commissioning.

The system was originally proposed to be built on top of GE Transportation's Interoperable Incremental Train Control System (I-ITCS) technology, an approach that was touted as advantageous for being "off the shelf."  There are subtle signs that things aren't going so well:
  • This month's CBOSS project status update states rather cryptically that the top challenge for the project is GE software release delays.  These delays could be related to the recent purchase of GE's Transportation arm by the French firm Alstom, which already offers a range of PTC technologies that could make the ITCS product line redundant.
  • The technical requirements for the electrification RFP state (see PDF p. 251 of 2,840) that CBOSS is built on Wabtec's Interoperable Electronic Train Management System (I-ETMS), the main competitor to GE's I-ITCS.  Switching from ITCS to ETMS would be like changing the foundation of a house after the roof is completed.
  • Section 4.13.1 of the electrification RFP document describes the electrification project scope, stating that "The Contractor shall provide the signal, train control and grade crossing systems"... a definition of scope that overlaps significantly with the CBOSS project.
  • Section of the electrification RFP document further describes items that are in the scope of the electrification program, including:
    • System-wide track circuit replacement
    • Manufacturing and assembly of signal enclosures, including installation and wiring
    • Installation of signal enclosures, wayside signals, cables, and cable infrastructure
    • Field testing of the signal system and integrated testing with the electrification, EMU, CBOSS/PTC and other interdependent systems
    • All work associated with the modification of the signal system required for the Project, including the CBOSS/PTC system as necessary and as required by all regulatory agencies, including the FRA, MUTCD and CPUC. The Contractor shall ensure that its wayside CBOSS/PTC systems are 100% compatible with the existing CBOSS/PTC systems that its systems will interface with.
  • The electrification RFP document continues for dozens of pages, describing how an almost entirely new signaling system will have to be installed as part of the electrification project.  What CBOSS was supposed to "overlay" will be largely replaced.
Is there a major architectural change in the CBOSS project that Caltrain staff failed to disclose to the board?  The entire CBOSS budget of $231 million (an astronomical sum for just 50 route-miles of railroad) having already been appropriated and mostly spent, are we about to see large cost overruns get squirreled away in the small print of the electrification RFP?  Is the respective scope of the CBOSS and electrification projects sufficiently well delineated to preclude spending money twice on the same item under two separate contracts?

A faint odor of fish wafts over the whole affair.

25 April 2015

The Blue Doors Will Open

Blue doors open at old 8" platforms (shown here)
Yellow doors open at HSR height
There are noises that Caltrain's new EMU fleet might sport a double set of doors, to enable boarding at two different platform heights.  While this isn't necessarily ideal, it is a reasonable solution given the constraints of the problem.  And yet, the prospect of trains with double the usual number of doors elicits gasps of horror from some transit advocates and industry insiders.  Let's go through some frequently asked questions to explore the roots of this choice, and feel free to ask more in the comment section.

Q. Do you intend for every Caltrain platform to be rebuilt?

A. Yes!  Today, the number of Caltrain platforms that support level boarding is zero.  In order to achieve the short and predictable station dwell times necessary to operate the blended system with acceptable reliability, level boarding is an operational imperative for Caltrain. The most important thing to realize is that Caltrain will eventually have to rebuild every single platform system-wide -- if not to the same height as HSR, then to some other significantly greater height for level boarding than the current 8-inch standard.  Level boarding is not just an option; it is a necessary expense without which blending Caltrain and HSR will fail.  Every platform must be rebuilt no matter what; this is the premise from which the rest of the discussion must start.

Q. But why insist on compatibility with HSR platforms?  Caltrain and HSR are separate systems serving separate markets, so why is this compatibility thing such a big deal?

A. Compatibility with high-speed rail is important for two major reasons.

First and most importantly, the San Francisco Transbay Transit Center will be a system-wide bottleneck for both Caltrain and HSR, with just six platform tracks.  This cramped but critically important terminus will be even more constrained if the two operators are forced to use segregated platforms.  In a segregated world, opposing flows of arriving Caltrain and departing HSR could conflict in the station approach tracks, triggering cascading delays should even one train fall behind schedule.  With platform compatibility, any arriving train can be routed to any available platform, minimizing the domino-effect of delays.  The Transbay designers know this issue is the Achilles' heel of the entire design, which is why they are pushing Caltrain and HSR towards compatibility.  The risk of an occasional equipment failure or medical emergency causing a system-wide meltdown depends on the probability of such an incident, combined with the underlying resiliency and flexibility of the infrastructure.  A segregated Transbay design is asking for trouble when things don't quite go according to plan.

Secondly, compatibility has enormous cost advantages for sharing station infrastructure, as will be seen below.  The savings from sharing station infrastructure at just four locations along the peninsula (Transbay, Millbrae, Redwood City and San Jose) could easily exceed the combined cost of converting Caltrain to high platforms system-wide.

Compatible platforms if operated carefully will not interfere with HSR security or fare collection methods.  They are solely a means to maximize the utility and robustness of the Transbay Transit Center and to reduce the capital costs of building California's HSR system by about a billion dollars (yes, with a 'B').

Q. Why can't HSR just select a train design with low floors?

A. It's not that easy.  The CHSRA has expressed an understandable preference for service-proven designs, to draw from the widest range of suppliers worldwide.  Very-high-speed trains (VHST) capable of speeds greater than 200 mph typically do not have low floors.  Nearly all high-speed train designs from Europe, Japan and China for the past several decades have featured high floors, with  few exceptions.  To achieve level boarding as mandated by the ADA while still drawing from the greatest possible selection of vendors, high platforms are almost a necessity for California's HSR system.  The only 200+ mph train with a "low" 30-inch floor is the Talgo AVRIL prototype, still in development.  It does not have distributed traction, which will be important in California's mountainous terrain.

The three foregoing questions allow Caltrain's entire range of possible solutions to be encapsulated in one simple flow chart:

Notice that ALL the level boarding solutions require dual level boarding, at the 8" legacy platform height and at whatever new level boarding height is selected.  Dual level boarding is not an easy problem to solve and usually involves some degree of awkward and clunky mechanisms, be they deploying steps, wheelchair lifts, automatic trap doors, or double sets of doors.  Even the supposedly "simple" level boarding scenario at 25" suffers from this complexity, a fact that is either glossed over or completely misunderstood by most advocates of this solution.

Q. So who cares if Transbay is so constrained?  Can't Caltrain just terminate whatever overflow traffic doesn't fit at the 4th and King terminal?

A. Emphatically, No!  Transbay is a key destination that every Caltrain must serve, especially at rush hour.  75% of Caltrain riders are commuting to work, and there are more jobs located within a 1/2 mile radius of the Transbay Transit Center than within a 1/2 mile radius of all other Caltrain stops from 4th and King to Gilroy, combined!  This enormous concentration of jobs in the heart of San Francisco will only increase with the many new office towers going up today.  Terminating even one Caltrain short of this gold mine of ridership would be quite simply counter-productive, a waste of taxpayer money and a failure to meet obvious demand.

Q. Do you understand the enormous effort and cost to do this?

A. It is a large expense, but also a necessary expense.  The cost of raising platforms is not strongly sensitive to height: rebuilding to ~48" is only slightly more expensive than rebuilding to 25", 30", or any other level boarding height.  Rebuilding to the same boarding height as HSR creates an opportunity to have the high-speed rail project defray some of Caltrain's expense for the conversion to level boarding.

The cost of totally rebuilding every platform is generously $10 million per platform, placing Caltrain's system-wide level boarding tab at (very roughly) $10 million/platform x 2 platforms/station x 32 stations = $640 million, less than half of the cost of the modernization project. On the basis of cost per minute of trip time saved, level boarding beats electrification.

Q. Doesn't this whole issue just boil down to a San Francisco Transbay problem that should be solved at San Francisco Transbay?

A. No, this is not just a Transbay issue.  Rebuilding to the same interface as HSR also enables savings of more than a billion dollars where station infrastructure can be shared elsewhere:
  • The massive dual-level elevated station, the six-mile approach viaducts, and the "iconic bridge" in San Jose would no longer be required, with platforms shared at ground level within the existing footprint of the Diridon station.
  • Squeezing a fourth track under the Millbrae station, requiring hundreds of millions of dollars of tunneling expense, would no longer be required.
  • A Redwood City HSR station, configured to provide HSR service to the booming northern end of Silicon Valley, to enable Caltrain to make cross-platform transfers from locals to expresses, and to tie in future Dumbarton Corridor service, would have a much reduced footprint and would more easily fit in the available site.
Factoring in these infrastructure savings, the cost of converting Caltrain to ~48" is actually negative.  It would be unwise not to do it.

Q. Won't rebuilding all the platforms take years and be an operational nightmare?

A. It will take years, but it can happen with relatively little disruption.  Over the past 15 years, Caltrain has rebuilt 37 platforms from the ground up.  Caltrain has a demonstrated track record for planning, funding and executing platform reconstruction projects.  You'll be hard pressed to find anyone who remembers this causing major disruptions.  EMUs with dual height boarding would considerably simplify the logistics of rebuilding platforms, since each station could be rebuilt independently as funding becomes available and as the planning process progresses in each community.

Q. Will Caltrain end up with a mix of high- and low-level boarding platforms, perpetuating this strange dual boarding height situation forever?

A.  No.  The lower set of doors has only one temporary purpose: to enable boarding from 8" platforms during the transition.  All platforms would be rebuilt to ~48", and boarding from the lower level would ultimately be discontinued.  Any remaining 8" platforms would become an impediment to the blended system, because they would introduce longer station dwells with a significant probability of unplanned delay when boarding or alighting persons of reduced mobility.  Making a mess of the timetable will no longer be tolerable, so there will be a strong incentive to finish the job even at minor Caltrain stops to ensure the highest level of punctuality and system reliability.  This makes Caltrain very different from MUNI or some East Coast commuter railroads, where a mix of boarding heights has persisted for decades because there is no operational imperative for 100% level boarding.

Q. Until all platforms are raised, wouldn't accessibility and dwell time be worse than they are today?

A. No.  EMUs with dual sets of doors would board from 8" platforms with the same efficiency as Bombardier cars, with just two steps up from the platform into the lower level of the train.  (Note that trap door designs would not fare nearly as well in this respect!)

Q. Won't dual sets of doors cause passenger confusion and long dwells?

A. Boarding the train will be obvious, based on which doors open.  Alighting requires queuing at the correct door, which can be facilitated by color-coding of the doors, LED displays, and audible messages.  "Next stop, Menlo Park.  The blue doors will open.  (...)  Next stop, Palo Alto.  The yellow doors will open."  Blue and yellow are good contrasting colors that can be distinguished by color-blind passengers.  People aren't stupid, and should someone get confused, the different doors would be within a few steps and within sight of each other so any mistakes would not lead to significant delays.

Q. Won't all these extra doors displace seating areas and reduce Caltrain's seating capacity?

After conversion to level boarding,
blue doors are plugged and replaced with seats
A. Not necessarily.  Caltrain has estimated that dual sets of doors would displace between 78 and 188 seats per train, or roughly 15 to 25% of a train's seating capacity.  This is a temporary situation during the transition to level boarding, and can be mitigated by procuring extra-wide trains with 5-abreast seating and longer 8-car trains to preserve overall seating capacity.  The space lost to extra doors can be used by standees, who do not have many good options on today's Caltrain fleet.  When the platforms are all converted to level boarding, the lower doors can be removed and additional seating can be installed, especially if this feature is designed into the new EMUs from the outset.

In the meantime, to minimize the loss of seating capacity, it would make sense for Caltrain to make seating capacity a selection criterion in the vehicle procurement process.

Q. Wouldn't this create an accessibility problem, in terms of ADA compliance?

A. No.  Just like today, lifts or bridge plates would be required to board persons of reduced mobility from an 8" platform.  There would need to be an in-vehicle wheelchair lift to change levels inside the vehicle, to allow wheelchair users to board and alight at stations with different height platforms, or to avail themselves of an accessible bathroom on the lower level.  This is not a new technology; these off-the-shelf mechanisms are no more complicated than the exterior lifts used on Caltrain's gallery cars.  An example of such a lift can be seen in this video.

Q. Will bicyclists have to navigate interior vehicle steps, potentially while the train is moving?

A. Yes.  These steps could be made wider and shallower (greater tread depth) than anything in today's Caltrain fleet.  The three steps from a 25" lower level to a 48" mid-level floor could be fitted with wheel gutters to allow bicycles to easily roll up or down along the stairs.  This would make the steps far easier to navigate than the four steep steps up from an 8" platform into a 45" gallery car, turning the corner around a pole through a crowd of Giants fans--the scenario that Caltrain bicycle riders are forced to contend with today.  And riders commonly lift, turn and sort their bicycles by destination while the train is moving, so a bit of jostling isn't exactly a new thing for the bike crowd.

Q. Wouldn't trap doors resolve this whole situation with dual doors?

A. No.  Trap doors have numerous flaws, including one fatal flaw: they would preclude Caltrain passing high platforms at speed while still maintaining an ADA-compliant 3-inch gap when stopped at a high platform.  Trains sway from side to side when running at speed, and the alignment between the track and the platform edge isn't perfect; that means the space between the train and the platform needs to be wide enough to prevent platform strikes but narrow enough to comply with ADA rules.  There is currently no rail system in the United States that can do both: there is either a speed limit when passing platforms (e.g. BART) or the gap when stopped is greater than 3 inches (e.g. Northeast Corridor).  Satisfying both constraints (< 3" gap and 100+ mph past platforms) requires a small bridge step to extend from the train when stopped.  This sort of gap filler mechanism is unlikely to be compatible with a trap door configuration.  Trap doors have other disadvantages, such as increased dwell times while the trap door mechanism is moving, sensitivity to damage from dirt buildup and foreign objects commonly found on train floors, and too many steps up from a low platform during the transition period.

The Takeaway
  • Level boarding is not just an option; it is an operational imperative for the blended system.  The blended system will not work reliably without it.
  • The new EMU vehicles must enable Caltrain's transition to level boarding, or the chance to convert to level boarding will be lost for another 30+ years, the life span of the new train fleet.
  • It is appropriate for Caltrain to adopt the same platform height and width as HSR, in exchange for the funding to achieve the transition to level boarding.
  • Transitioning to level boarding is complicated regardless of the chosen platform height; there is no easy solution.
  • Dual doors are the path to level boarding with the fewest flaws, under the imposed constraint of high-platform HSR.

08 April 2015

Find Out Your Pole Placement

Plan showing where electrification poles
will be placed in the vicinity of the
historic El Palo Alto redwood
(see RFP Volume 3 page 1193)
Caltrain has spent over $15 million on its electrification project so far, primarily for environmental clearance and preparations for procurement.  In late February, we found out where a lot of this money went: the Request For Proposals (RFP) for the electrification project was released.

This RFP is an incredibly prescriptive document that tells prospective bidders precisely what the project should look like, down to the last bolt.  Volume 3 of the RFP (download the 2840 page, 214 MB PDF file) includes layout plans of the overhead electrification system that dictate the exact placement of every single pole foundation.  The prospective contractor is admonished that pole locations cannot be changed without first submitting a formal design variance request to Caltrain.

This procurement is being carried out as a "Design-Build" where the winning bidder will be tasked with "designing" the project, which in this case will amount to a connect-the-dots exercise to duplicate Caltrain's highly prescriptive preliminary engineering drawings into final construction-ready drawings.  What little room is left for creativity and efficiency is stifled by an onerous variance process that requires the "designer" to submit extensive paperwork to Caltrain for approval of the slightest change to the design prescribed in the contract.  One can easily imagine how the goal of Design-Build contracting, namely to reduce risk and cost by consolidating decision-making under a single entity, would be lost under the hyper-prescriptive approach that Caltrain has chosen.

The thousands of pages of the RFP highlight the cozy symbiotic relationship that exists between government agencies, their in-house consultants, and private contractors.  Without an ounce of nefarious intent on the part of any of its participants, this self-reinforcing triangle, hardly unique to Caltrain, brings together hollowed-out government agencies with rubber-stamp boards run by politicians, permanent in-house consultants whose primary motivation is to justify their existence through highly prescriptive decisions that increase scope at their whim, a profit-hungry coterie of construction companies ticking all the boxes for shareholders and labor interests, and a byzantine system of contracting regulations and reporting requirements, quite ironically intended to prevent taxpayers from being defrauded.  The results of this firmly-entrenched Transportation Industrial Complex are projects that deliver less and cost more, typically three times the going rate in other first-world countries where government agencies are centralized, smart, and employ an experienced staff of technocrats whose first interest lies in serving the public with better transit at lower cost.  What can be done about this system?  Not a whole lot.  It is the logical byproduct of our decentralized system of government and of our free markets, pursuing their respective enlightened self-interests.  These self-interests include neither low cost to the taxpayer nor excellent transit service to the user.

The recently-completed modernization of Auckland, New Zealand's commuter rail network, of quite similar technical scope, is an instructive benchmark against which to evaluate Caltrain's modernization efforts.

15 March 2015

News Roundup, March 2015

Regulations for 25 kV electrification.  After two years in the making, CPUC proceeding R1303009 appears to have produced the new rule book for stringing up high-voltage rail electrification in California, although the final document is still to be formally adopted by the Public Utilities Commission.  The final draft hammered out after numerous meetings by the high-speed rail consultants, freight railroads, utility companies and other interested parties thankfully bears little resemblance to the original draft proposed by the high-speed rail consultant, which was a mess.  While the new rules apply only to dedicated high-speed rail corridors without grade crossings or freight trains, the language is flexible and there appears to be room allowed for parties such as Caltrain and UPRR to agree on those particular items on a case-by-case basis, without resorting to an entire new CPUC rulemaking process for the peninsula rail corridor.  Most importantly, the new rules provide a clear framework that allows detailed engineering design of Caltrain's electrification project to proceed with a very low risk of future regulatory surprises.

Grade separations.   An unfortunate series of accidents and suicides have re-ignited the debate over grade separations in PAMPA (Palo Alto - Menlo Park - Atherton).  This portion of the corridor abuts some of the most expensive real estate on the peninsula and is home to the most contentious and litigious environment for local decision-making (or lack thereof, as the case may be), and it is gradually dawning on these communities that Something Must Be Done.  Grade separation is often misunderstood as an all-or-nothing proposition, when in reality it is a process that has been underway for decades.  The peninsula rail corridor is already 62% grade-separated today.  Communities up and down the peninsula are starting to talk more about finishing the remaining 38% of the job.

Quiet zones. Palo Alto recently conducted a study session on quiet zones, which would stop routine horn-blowing at grade crossings.  Curiously, the systematic use of train horns before every protected grade crossing is an American practice not usually seen in other countries, where crossing gates combined with visual and audible warnings are considered to provide a sufficient level of public safety.  But then again, those other countries don't have lawyers like ours.  The staff report contains a nice overview of the process for establishing a quiet zone; the main impediment seems to be the local community's assumption of legal liability for grade crossing collisions.

Peninsula HSR reboot.  After spending $45 million on environmental clearance and design of the original and much-reviled four-tracks-all-the-way peninsula HSR project, the California High-Speed Rail Authority shelved the plans with nothing delivered.  The Authority recently submitted a Project Update Report to the legislature, stating that a new environmental clearance for the peninsula segment is scheduled for completion in 2017, with construction complete in 2028.  [UDPATE 3/21]  Furthermore, an obscure Ridership Technical Advisory Panel Review of the California High-Speed Rail Ridership and Revenue Forecasting Process dated December 2014 includes the map graphic at right, and states that the Authority "has requested a stand-alone analysis of two individual segments of the overall system, to assess the viability of each segment operating independently for a handful of years prior to the completion of the system. (...) The Bay Area segment would require improvements to existing rail infrastructure. HSR service would be blended with existing Caltrain commuter rail."  The peninsula schedule may be pulling to the left.  To clear an EIR that fast, for a project so much more controversial than mere electrification, the process would have to be re-started very soon.  This may or may not be related to the choice of...

New Caltrain CEO.  Jim Hartnett was selected as Caltrain's new chief.  He is a consummate local political insider with no background as a transportation agency executive, other than chairing the Caltrain board of directors and vice-chairing the board of the CHSRA.  While he will no doubt be well equipped to navigate the choppy waters of inter- and extra-agency politics, he is less likely to shake up the Caltrain organization or to bring fresh outside-the-box thinking to the development of the "blended system."  While Caltrain has plenty of political and organizational problems to work on, the technical aspects of blending will determine the success of the rail corridor for decades to come.  Will Hartnett do anything about the top ten problems facing Caltrain, or will he just hold the current course?