Siemens and Kinkisharyo vehicles in Bellevue last weekend (image by author)

The Sound Transit System Expansion Committee received an update yesterday on the availability of new Siemens vehicles to support the beginning of service to Northgate this Fall. Significant risks remain, but progress has been made since a previous report in November. CEO Peter Rogoff expressed confidence that Northgate Link will open on the intended schedule.

A minimum of 30 cars must be available to open Northgate service with the full complement of four car trains at eight minute headways.

Sound Transit had baselined the Northgate project with 40 cars being available by end of 2020. That plan was knocked off track by a series of early supply chain holdups and later testing issues, and further complicated by COVID. Manufacturing is now on schedule and on budget. Commissioning, the lengthy series of steps between manufacturing and putting cars in service, remains behind schedule. When we last reported on this in November, Sound Transit was pursing a ‘recovery schedule’ to get the needed minimum of 30 cars ready by the opening of Northgate in September 2021.

Are they on track? 37 cars are now in Seattle (vs. 26 in November), with 20 ready for mainline testing (vs. 12 in November). The largest single problem remaining in November, a signals interference issue between on-board subsystems, has been resolved.

If all goes as scheduled, the first six cars will be available by the end of March. That’s enough to put the first train into service (with two vehicles as spares). The significance of the ‘qualification vehicle’ goes beyond just having one train available. In important ways, it verifies the design is performing as expected, and prepares the way to move the following vehicles through burn-in to verify reliability.

Staff characterized the path forward as a reasonably achievable schedule with 42 days of float (at the time of the November report, there was still two months of float).

Whether all the needed LRVs are ready by September seems to hinge on whether there are any more adverse surprises, with the greatest unknowns in the next several weeks. There will be another update to the Committee in Spring 2021.

39 Replies to “Northgate LRVs on schedule, but risks remain”

    1. They have been so far, but there’s no reason for them to keep doing so. The era where a single-car train could suffice in any circumstance has long passed.

      ST should switch to single-cab cars for all purchases in the near future, until the number of single-cabs in service equals the number of dual-cabs. Once that happens, they can start purchasing zero-cab cars in addition to single-cab, with the end result being four-car trains with only two cabs in total.

      1. This will never happen. They will always use daily cab trains. There is no way to turned the trains around in single cab trains.

      2. Also, for day to day operations, it makes since for us to stick with double cabs so that LRVs are interchangeable in a moments notice.

    2. It was clear in the mid 2010s when ST switched to 2-car trains in the evenings that it would never again be able to run 1-car trains. All the extensions will generate more overlapping trips, meaning more people on a train simultaneously, not fewer. Right now from Northgate to downtown you take the 41, so the 67, Link, and the 49 are irrelevant. But with Northgate Link you’ll take Link, thus overlapping between UW and downtown. From Lynnwood to Roosevelt you’ll take the 512+Link, but after Lynnwood opens you’ll take Link all the way, meaning more people will take transit, so they’ll overlap between Northgate and Roosevelt.

      We raised this in the mid 2000s, that the ST2 order should include cars with open gangways. ST refused because it said having all cars double-cabbed makes all the cars interchangeable, which makes maintenance easier. So ST turned down the change to increase capacity 20% “for free”, so that it could swap out any car for any other. ST doesn’t change its mind very often, so it’s not that likely it will switch from double-cabbed cars in the future.

      1. I meant 2010s. ST2 was approved in 2008. I don’t remember exactly when the Northgate Link train order was but it must have been somewhere between 2012 and 2016. ST stopped using 1-car trains evenings sometime around 2012. The train order specs were refined over several months, and that’s when I and others sent ST feedback saying the order should have open gangways or single- and zero-cabbed cars. I don’t remember exactly when ST placed the orders for UW, Northgate, and Lynnwood, but at least two of them probably occurred after ST stopped using 1-car trains.

  1. Do they really need four car trains this year? By late summer/fall I expect more people to have vaccinations, but I don’t expect ridership to increase all that much.

    More vaccinated people means the need to spread out on a train so much goes down, and low ridership means you don’t need four car train capacity.

    Oh sure ridership will eventually rebound, but I doubt it will rebound that much this year.

    so….. what am I missing?

    1. so….. what am I missing?

      Not much. I was going to get on here, and make a snarky comment about how maybe the shortage of trains will get them to improve frequency (and how we should all be routing for a delay). But that isn’t realistic. They will simply have fewer trains. It will likely take ridership a while to get back to pre-pandemic levels.

      1. so, why not six-minute headway and three-car trains to reduce waiting? To integrate with CT, ST, and Metro bus networks, short waits are more important than capacity.

      2. “so, why not six-minute headway and three-car trains to reduce waiting?”


        Did the voter approved measure say anything about 4 car trains? If it did maybe that is it…..

      3. Shortage of operators/expense of paying more? I don’t know about ST but I remember a month or two ago metro was missing a lot of runs because so many of their drivers were sick or in vulnerable groups.

      4. 3-car trains every six minutes only reduces fleet requirements by about 3 cars, compared to 4-car trains every eight minutes. It yields the same number of cars per hour (30) so it is capacity-neutral; the lower car requirement comes from only needing 3-car “gap” trains plus some rounding on the maintenance/out of service side. However it would increase operating costs by about a fair amount (you need 33% more operators at 10 trains per hour instead of 7.5). That’s significant. So you’re trading a slightly smaller fleet and minor maintenance savings for a large increase in operating cost. No doubt shorter headways are better for users, but ST’s budget is far from unlimited so operating costs matter.

    2. If we plan for today, we will not be prepared for the future ridership.
      It is a catch 22.
      l think it is better to have excess capacity and not need it, rather than asking why we didn’t purchase the equipment required for future ridership.

  2. I think we’re good to get trains ready for Northgate Link in September. Very excited for what’s next.

  3. How does the East OMF and East Link scheduling/ opening sync with the vehicle delivery and need for storage? Northgate is a much smaller impact to service hours and new vehicles needed than Line 2 (East Link) will be.

    1. OMF east doesn’t impact the Northgate schedule, the 30 trains needed to service Northgate presumably all fit in OMF-central.

      But OMF-E it is needed to hold the K trains so the OMF-C can store more Siemens trains for testing. The Siemens trains needs to get ‘burned in’ on the in service Link segments prior to East Link opening to ensure there is a sufficient fleet available to run East Link. My understanding is that once the OMF-C is full, ST will need to truck some K trains to OMF-E for storage to make room for continued delivery & testing of Siemens. This is why OMF-E is finished before East Link. There will be trains parked in OMF-E before ST is able to drive a train between the two OMFs.

      1. I never said transit should not be subsidized. I said if expected revenues — general fund or farebox — drop then you need to increase tax RATES or fares, or reduce expenditures, to equal the same revenue. . That’s all HB 1304 is, except I doubt it will be enough to complete ST 3 in N. King Co.., if it passes and a levy passes.

        And I don’t think a 40% farebox recovery target, which is part of estimated revenue and is fare X riders, is insignificant.

        It is meaningless to have a funding discussion about “Sound Transit” because there are five different Sound Transits, with five different project lists and five different future revenue estimates.

        It is also pointless to claim the N. King Co. subarea has sufficient revenue until 2029 — if you believe Rogoff who will be long gone and is pleading for a HB 1304 levy — if the major ST 3 projects begin and end after 2029.

        Of course all expected future ST 3 revenue is allocated to projects. That is how east King Co. got a $4.5 billion line from Issaquah to Kirkland when 98% of the time those residents will drive that route, even with a rail line, since they have to drive to a park and ride anyway. If it were not for the higher general fund revenues for the Eastside subarea due to unanticipated economic growth I would imagine the likely much lower ridership on East Link — which equals farebox recovery — would threaten that crazy idea.

        The reality is based on regional economic activity, and the fact the N. King Co. subarea had to fund rail to Snohomish Co. and S. King Co. and pay 1/2 the second transit tunnel the subareas would have been drawn differently.

        Rogoff is admitting now to underestimating costs for ST 3 projects in N. King Co. because his expected future general fund revenue for decades looks like it will be significantly lower, when he always anticipated higher because he always knew the ST 3 projects were estimated low, and Seattle’s economy had grown so fast over the last decade, especially the tunnel which still has not been explicitly repriced but ain’t $2.2 billion, except four other subareas have budgeted for a $2.2 billion cost.

        The spine will be completed. Will ridership and farebox recovery come close to estimates for each subarea? Who knows, although my guess is east King Co. will have the biggest gap between estimated and actual ridership, although I don’t know what effect that will have on ST 3 projects in that subarea. I
        doubt Issaquah residents would be as heartbroken if their rail line never got built as West Seattle and Ballard residents, especially if traffic congestion post pandemic is manageable on the Eastside and ST continues direct express buses for that area.

        I still think the real issue, at least in the short run, is first/last mile access, which means buses, and Metro. ST can’t add a transfer and seat to most rides for tens of billions of dollars and have the ride take longer than before.

        Metro looks like an unreliable and politically driven agency, in part due to the same funding issues ST is facing with a 20% farebox recovery goal. If first/last mile access and frequency don’t make total trips shorter with rail ST will have been a colossal bust, but right now ST’s success doesn’t seem to be a priority for Metro. We can manufacture reasons for transit all day long, but if rail and bus truncation is slow it will be a very costly failure.

  4. The new Siemens design is just awful, and ignores ideal configuration data from the London Underground and NYC subway system. Is there any way we can trade them for more Kinkisharyos?

    1. How are the Siemens cars awful?

      I think the seating layout of the Kinkisharyos is far from ideal. The entire lower floor sections should only have seats on the perimeter, and maybe only on one side.

      1. 2013 was the last year a major subway design study was released. It recommended more front/back facing seats and less side facing seats. The Siemens cars are a move in the opposite direction from the science driven data.

      2. Okay looks like the seating layout is nearly the same. What a missed opportunity to create more room for those standing or those who need extra space and to improve passenger circulation space.

  5. Curious, is the plan to stick to 8-minute headways after Northgate Link opens? Is it scheduled to get more frequent at some point? I assume when East link opens the frequency should bump up on the downtown – Northgate segment at least.

    1. The pre-ST3 environmental and Federal grant documents said that the peak trains on lines 1 and 2 would be at 8 minutes, with 10 minutes at other times. That would be 4 minutes peak and 5 minutes non-peak between Northgate and ID.

      ST3 promised 6/10 minutes for each line or 3/5 minutes when combined. It’s unclear when that would happen — but probably after more extensions and OMFs for more trains open, and more riders use the system.

  6. Jack, the existential answer to your question is who is paying. If transit were not so heavily subsidized the answer is obvious: ridership drives frequency.

    But in reality fare-box revenue (20% bus, 40% rail) and general fund revenue determine frequency, along with equity politics, which only means transit frequency is limited because subsidies are limited.

    Despite politics, in the end general fund subsidies will determine frequency. This depends directly on the economic growth of your subarea. If farebox recovery is 20% on Metro and ideally 40% on rail obviously general fund subsidies determine frequency.

    Metro and ST believe reduced ridership fare recovery and reduced general fund subsidies — especially in the N. King Co. subarea — will reduce frequency, although many transit advocates believe that approach is backwards, if money is no option.

    1. Fare recovery never pays for transit. The reason fares exist is to demonstrate transit value, not generate a profit. In this context, as long as you have drivers, transit should continue to run even if coaches and trains are empty. This doesn’t look correct when viewed through the lens of capitalism which is why transit is terrible in this country.

      1. Fare revenue is an important part of transit funding for most US transit agencies. ST is unique is that its long term capital plan is fully funded without fare revenue, which makes fare revenue look inconsequential in the context of the overall financial plan, but if Link achieves 40% farebox recovery that is still going to be a significant chunk of change

    2. You are ignoring the dedicated taxes, which will continue after construction ends. If ST3 builds only to Paine Field, Tacoma Dome and Redmond Downtown because your prediction of downtown Seattle’s collapse is correct, the agency will have a FLOOD of revenues from sales and car tab taxes, which will continue whether any bonds have been sold or not.

      There will be no need for “General Fund subsidies”. The only problem ST faces is that the “hump” in indebtedness limits its authority to issue new bonds for the seven to ten year period after then. If there is no need for WSBLE, Everett and Kirkland-Issaquah, there will be no “hump”.

    3. You are ignoring the dedicated taxes, which will continue after construction ends. If ST3 builds only to Paine Field, Tacoma Dome and Redmond Downtown because your prediction of downtown Seattle’s collapse is correct, the agency will have a FLOOD of revenues from sales and car tab taxes, which will continue whether any bonds have been sold or not.

      There will be no need for “General Fund subsidies”. The only problem ST faces is that the “hump” in indebtedness beginning in 2029 limits its authority to issue new bonds for the seven to ten year period after then. If there is no need for WSBLE, Everett and Kirkland-Issaquah, there will be no “hump”.

    4. “If transit were not so heavily subsidized the answer is obvious: ridership drives frequency. But in reality fare-box revenue (20% bus, 40% rail) and general fund revenue determine frequency, along with equity politics, which only means transit frequency is limited because subsidies are limited. ”

      Frequency and fares are independent political decisions. Of course frequency can’t go above the ceiling of available revenue, but within it they are arbitrary decisions, hopefully made based on sound values. Daniel mentions two factors justifying frequency: ridership and equity. There’s also a third factor: convenience. Five- or ten-minute headways mean you never have to wait long and you can fit a maximum number of activities in a day. It provides rider satisfaction, allows people to make their optimal number of trips, and is competitive with driving. Cities and all residents have an intrinsic benefit in people having the maximum ability to travel to work, shopping, personal activities, and health maintenance. It increases the GDP, personal self-sufficiency, social cohesion, and health. And it’s more energy-efficient, space-efficient, and equitable than driving. Transit should be available and convenient for as large a cross-section of people and trips as possible, and driving as a second choice.

      This must be balanced with the reality of vast swathes of low-density residential-only areas, which are difficult to serve with frequent fixed-route transit and we don’t have enough political consensus to provide it anyway. We should never have allowed those areas to grow like that, but now they’re there and hundreds of thousands of people live in them. Still, even if we can only provide a partial solution to those areas in the medium term, we should provide a full solution in the easier-to-serve areas, which at minimum includes most of Seattle, most of Bellevue-Redmond-Kirkland-Bothell, and between downtown Renton, Kent, and Auburn and their eastern residential areas where the bulk of their population lives.

      Ideally transit should be a government-funded service paid by everybody’s taxes (free fares) like the libraries, parks, courts, 911 service, etc. Historically most cities charge fares as a token user fee, to pay a percent of operating/maintenance costs, and to deter inappropriate activity. That’s OK as long as the fare isn’t unaffordable and there’s a sliding scale for the poor. The US also has a particular problem with wide inequality and forcing poor people to sleep outside and not meeting their basic needs. This causes unnecessary stress and resentment and untreated mental conditions, which all lead to inappropriate behavior on transit and people using them as shelters if there are no fares. So charging fares is less than ideal, but can be justified or necessary. King County’s policy is to keep fares with a 20-30%-ish window of operating costs. Link has higher fare-recovery targets. But because Link is more efficient than buses, its fare went below Metro’s several years ago for trips up to Westlake-Rainier Beach distance. There are arguments on both sides of flat or distance-based fares, exacerbated by the long distances of ST3 Link, Sounder, and ultra-long express buses.

      So fares often pay a percent of operating costs, but the decision of whether it should be 20%, 50%, 100%, or 0% is a political one.

    5. Another thing I didn’t fully explain, people benefit from transit even if they don’t ride it that day, because the fact that it runs every day means it will be there when they need it. And the fact that it’s there affects their choices on how many cars to own, how many miles to drive, and where to live. These people aren’t counted in ridership numbers because they’re not currently in a bus/train seat, but the fact that transit is running and they have the choice to ride it benefits them.

  7. Does anyone know of a document which describes the LRV commissioning process? I’m curious what tasks that encompasses.

Comments are closed.