Muni Double-Berthing Still Delayed Pending CPUC Approval

The ongoing delays for double-berthing in Muni Metro stations continue, as Muni waits for its training plan to be approved by the CA Public Utilities Commission.

Photo: RobVSF/Flickr

After Muni officials demonstrated double-berthing for the CPUC in December, expecting the green light, Muni Operations Director John Haley told Streetsblog that CPUC needed to sign off its training plan, which Muni officials apparently didn’t anticipate. In early February, Haley told us he was in talks with CPUC and hoped to have approvals in place within two to four weeks.

But CPUC spokesperson Terrie Prosper said it was only on March 27, last Friday, that the agency received all of the documentation needed.

“We had already done the training,” said Muni spokesperson Paul Rose, “but the CPUC requested that we get signatures from each operator. We have done that and are awaiting a response.”

“It is too early to provide a date for launch,” he added.

Prosper told us on Wednesday that CPUC would send a written reply “in the coming days” to “allow SFMTA to place the system in service.” Rose said Muni hasn’t received it yet.

The SF Transit Riders Union “has been eagerly awaiting double berthing for quite a long time, and we’re very happy that the SFMTA is ready to move forward,” said spokesperson Reed Martin. SFTRU “urges the CPUC to move quickly and approve the plan, allowing Muni Metro riders to finally experience double berthing in action!”

We also have a few more details on some of the limitations of double-berthing, also known as double-train loading. According to Muni Deputy Director of Operations Jim Kelly, double-berthing will only be possible at Civic Center, Powell, and Montgomery Stations because the platforms at the rest of the stations are too short for two trains to load safely. Each train which loads behind another train will also stop a second time at the front of the platform.

  • Jeffrey Baker

    Bureaucrats gunna ‘crat.

  • njudah

    oh whatever. more bibble babble and talkytalktalk because that’s all anyone is capable of doing in San Francisco. No Getting Things Done. No Responsibility.

  • mx

    Years worth of nonsense and delays at who knows what expense for a change that will result in basically minimal improvement to Metro service? That sounds like the Muni way. Why are we so optimistic about the new “rapid” service changes again when they can’t even pull this one off?

    Thinking about it, there’s a good case to be made that double berthing could slow down operations. While it would be moderately useful for disembarking passengers to be able to get off sooner during “heavy traffic” conditions, delays will only be exacerbated by the need for trains to stop twice at each station. Why are SFTRU such big supporters of this?

    A far better approach would be to reduce the backup trains experience headed into Embarcadero by putting a stop to the long dwell times that occur there. This can be done without developing new technology or any CPUC approval; just have empowered supervisors on the platform who ensure that operators are in position before trains arrive and that everybody keeps moving quickly.

  • Andy Chow


  • reedm

    Agreed. To be clear, none of us are sure how double-berthing will play out, particularly in the “random” model that appears to be how the SFMTA will roll it out. The SF Transit Riders support moving the concept forward to see if there are improvements, and generally support streamlining SFMTA’s ability to trial concepts that have the opportunity to improve service.

    My personal guess is that this will not lead to a noticeable improvement in Metro service, and could increase confusion, but the SFMTA has indicated opportunity, and I’m curious to see how it plays out.

  • Chris

    Too bad double-berthing won’t occur for inbound trains at Embarcadero. That’s where it makes the most sense because the inbound J, L and M do not pick up passengers there. I don’t buy the explanation that the platform is too short. I think it’s because they don’t want to open up the portion of the platform that’s currently off-limits to the public.

  • datbeezy

    I don’t understand the double-berthing fetish. You can still only go as fast as the train in the front.

  • theqin

    It seems to me like if double berthing is not allowed at embardadero, then the trains will be timed a certain distance apart and so there will be no double berthing advantage for outbound trains, only inbound ones.

  • Haven’t you ever been stuck in a train that has just pulled into a station but is waiting for the train in front to move just so they can open the train doors? It’s so frustrating and pointlessly stupid that it’s taken so long for these people to figure out how they can open the train doors while the trains is parked somewhere other than the very front of the platform. Double berthing will save time and frustration for passengers.

  • rickbynight

    Agreed, this is the potentially the best outcome of double-berthing IMO.

    That said, it remains to be seen whether having everyone who is waiting at the end of a platform have to move down the platform to board will end up holding the train up for longer, increasing backup in the tunnel.

    In cases of prolonged delay, Muni drivers today do have that ability to manually open the doors at another location on the platform, though the train is still expected to stop in its normal location once traffic resumes.

  • jonobate

    When this was first proposed there was a regular issue with trains waiting at the back of the platform for several minutes while the train at the front of the platform just sat there, either because that train had a mechanical or ATCS issue preventing the train from leaving, or because another train in the subway had such an issue and all the trains were being told to hold at their station.

    This delay could sometimes be 10 minutes or more, and it was very frustrating not to be able to just get off the train and walk to your destination when you could see the platform right there. I even saw someone pull the emergency brake to open the doors and allow him to leave. (The driver was pissed when he realized what hat happened.)

    This is why double birthing was proposed, but it’s really a band-aid solution to these mechanical and ATCS issues, which are the real problem. In my highly subjective experience, these mechanical and ATCS issues are less of an issue now than they were when double birthing was first proposed, and getting stuck behind the train in front is less common. Hopefully the new trains will be more reliable and these issues will largely be eliminated.

  • mx

    Well said. In the case of serious delays, operators could simply be instructed to put the train into cutout mode and open the doors manually (and they’ll do this sometimes anyway). Such an instruction would have required a single memo to accomplish the same purpose, rather than years worth of inaction.

    Double-berthing is only useful if the delay will be more than a minute or two, at which point manual door operation by the operator is seemingly a perfectly valid solution. In the case where a second train is briefly stopped waiting for the forward train to clear the station, opening the doors and having two load/unload cycles at the same station will just cause further delays to following trains.

  • twinpeaks_sf

    Anyone know if the 3-car trains were a good deal (or do we just not have enough trains)?

  • SFnative74

    I’d like to see subway trains be able to open their doors before they come to a complete stop. In most countries, the doors start opening when the train has slowed to 2-3 mph. By the time the door is open enough to allow someone through, the train has stopped. Instead, our trains have to stop then it’s another 2-4 seconds before they open. It may not seem like a big deal, but if you instead open the doors early, that could save up to 10 seconds per stop, which could add up to over a minute between the Embarcadero and West Portal. If we can get over the hyper-safe restrictions we suffer in the US mindset, it’s basically free travel time savings. In cities like NYC with a large number of stops per line, this travel time reduction can really add up.

  • omaryak

    The “limitations” are terrible news … those platforms deemed “too short” to board “safely” were obviously designed to accommodate multiple trains. Half the space on them is being completely wasted. And stopping a second time in the same station completely defeats the purpose of double berthing. Accurate signs would let people know where to board in advance. This is a sad day for transit in San Francisco. Very sad.

  • omaryak

    In theory you can load twice as many trains in the same amount of time … signs would let know riders where to board in advance. Unfortunately if the CPUC is making trains stop a second time in the same station then you are right, it is pointless.

  • omaryak

    And apparently they think riders can’t read signs if they are going to make trains stop twice.

  • From a slightly less subjective experience sitting on the SFMTA’s advisory Engineering, Maintenance, and Safety Committee that’s a pretty good summary and status report.

    Ideally trains would indeed be more evenly spaced so each train is able to pull right up to the first berth because it’s had the time to clear. It’s taken about a decade of work replacing everything from the overhead wire to the major computer system components that are all dependent upon one another before it could even start to be felt, but it’s finally gotten to the point where riders indeed are starting to see the pay off.

    I view double-berthing as a bonus feature primarily for unloading that came included with the new software. Three-car trains (which prevent double-berthing) seems to me like a more important capability in the years to come. Especially if the M-line/Parkmerced/19th Avenue Subway is constructed as it seems likely with three-car platforms.

    One particular issue with the Muni Metro subway I think is worth pointing out because it’s starting to really effect BART now is track maintenance. As track wears out, trains have to move over it more and more slowly to prevent derailment or further damage from the strain and shaking which snags service. Being able to travel at a faster top speed allows control computer more flexibility to speed up and slow down trains to even out train spacing.

  • jonobate

    Agreed about the three-car trains. I haven’t seen one in a long time, so I guess they’ve been discontinued.

    It was pretty cool to watch them show up and completely clear a crowded platform, but as shuttle trains they took up valuable slots in the subway and were often underutilized. They will be more useful once one of the surface lines are upgraded to handle three-car trains.

    The M subway project you mention would accomplish this, but I think a higher priority should be upgrading to N to handle three-car trains, at least for short line runs turing around at UCSF or 9th Ave. This could be accomplished by lengthening the surface platforms, or by extending the Sunset Tunnel west as a subway, emerging in the center of Judah west of 9th Ave.

  • The three-car shuttles were launched as a six-month trial, but I never saw any results other than everyone liking them.

    Any shuttle though takes trains away from the surface sections of the lines and having proved they work it might just be a matter of having enough trains once the new ones start arriving.

    By ridership alone the N-Judah does make sense for three-car service, but there’s a unique opportunity open right now with the M-line: The city, CalTrans, SF State, the developers or Parkmerced and Stonestown are all lined up to share the costs of undergrounding the M along 19th Avenue and St. Francis Circle.

  • aj

    RE: SF Examiner article:
    Muni given go-ahead for double berthing at downtown stations

    “As welcome a change as double berthing is, a 2013 SFMTA report revealed it would result in zero minutes of travel-time savings. Still, passengers said, standing up while stuck on a train when you could easily off-board can be maddening. Double berthing should alleviate the problem.”

    This was upper management BS!

    Although I haven’t seen the referenced 2013 report, it exemplifies Upper Management disconnect from reality: So you get to your destination in X minutes. But apparently what the report did not see as being important was the fact that even though the train had arrived at a destination platform, passengers could be trapped for a long time until the train reached the sole forward “target point” where the doors could open in “AUTO.”

    Double, triple, and even quadruple berthing (depending on train-consist) was done as a matter of routine when METRO trains were controlled by Operators, from around 1980 to late 1990’s/early 2000’s (?).

    When MUNI moved to the Automatic Train Control System (ATCS) in the late 1990’s, the programming of this system bought from Alcatel only allowed, if I recall correctly (not sure–it was a long time ago), only one train on the platform. And I think it was only later that it allowed a second train to enter a station platform.

    When the second train was brought onto the platform, the ATCS system would not open its doors until the first train had departed and the second train got to the target point. When there was a delay ahead, this would cause passengers to be prisoners on the trains–justifiably angry prisoners!

    That it took MUNI 15 years to finally correct this problem is pathetic.

    The reality is that Operators early on had taken it upon themselves to make up for this ATCS system design shortcoming by opening the doors under Operators’ control to lessen the problems of the ATCS programming that only allowed door-opening at the single target point at each station.

    However due to Upper Management changes/new sheriff-in- town scenario that took place somewhere around 2006-2008, this practice of Operator-initiated door openings pretty much was put to a stop.

    This was a time of “Don’t think for yourselves; just follow orders and don’t do anything without authorization.” I may be biased, but I think this coincided with one of the periods of the worst MUNI Metro service.

  • The Bredas have a nasty habit of losing communications to a third car, and uncoupling, so we’ve stuck with 2-car consists.

  • twinpeaks_sf

    I’d heard that. And nothing different since the pilot last year?

  • mx

    Something different? You must be new in town 😉

  • I haven’t heard anything since that pilot.

    Bigger picture, Breda won the contract in 1991, you’d think they’d have figured out a way to recable things sometime in the last 24 years. Maybe pick up some Ethernet switches at Radio Shack’s going-out-of-business sale?


Muni “Double Berthing” on Hold Two to Four Weeks, Pending CPUC Approval

Muni may finally launch “double berthing” in its Metro stations in two to four weeks, “barring anything we can’t think of,” according to Muni Operations Director John Haley. Muni riders were first promised simultaneous loading of two trains in underground stations in October 2013. More recently, the launch was scheduled for December, then delayed again. Muni demonstrated a live test on […]

Muni Plans to Launch “Double Berthing” in October, a Year Behind Schedule

Muni still hasn’t launched double loading of trains in its metro stations, a.k.a. “double berthing,” but says it could finally happen in October — a full year after the original launch date. “The project is still in testing, to debug issues with the platform signs and trains,” said Muni Operations Director John Haley. “According to […]

Muni Metro to Launch Double-Train Loading, Three-Car Trains in October

Muni is poised to begin double-train loading and running three-car trains at underground metro stations to speed up boarding and increase capacity on the system’s busiest stretch. As we reported, Muni began testing double-train loading, a.k.a. “double berthing,” in July, but Transit Director John Haley said the agency hasn’t run tests with actual trains yet. […]