.@SoundTransit Light Rail Pulling Into Westlake Station

photo by Joe Kunzler / flickr

Commenter “Alex” offered the following tip for finding out when 2-car and 3-car Link trains are coming:

“Actually, if you check the vehicle identifier at the front of the train, if the number is 1 through 40, it’s a 3-car consist. 41 and above are 2-car “trippers.” Do note that once the 3-cars begin to cycle out after the weekday evening peak, these numbers no longer are clear indicators. The vehicle id is also accessible in OneBusAway.”

On weekends and during the weekday mid-day period from 9 am to 3 pm, all Link trains are 3-car consists.

I have been giving the advice to wait near the third car, which is pretty dependably the one with the most available space, during peak hours and on weekday evenings. The rest of the time, board the third car.

ST spokesperson Kimberly Reason had this response to Alex’s tip:

– It’s true that morning trains are 40 series numbers and afternoon trains are 50 series numbers. Trains are typically 2-car consists, while all day trains are 3-car consists.

– However, this information isn’t fool-proof. During service disruptions, it is not uncommon for Operations to renumber trains to get the system back on schedule. Often when this happens, 3-car and 2-car trains can be assigned a train number.

– Therefore, attempts to rely solely on train ID to determine consist length would produce inconsistent outcomes.

– Also, train length doesn’t determine dwell times. Dwell times are set at an industry standard of 20 seconds, and signals are set in synchrony to this standard. Even when trains load faster than 20 seconds, dwell time remains constant.

Given that dwell time is often significantly longer than 20 seconds, the tip may still help out, and every little bit of time savings will help during Carpocalypse, even if Carpocalypse itself degrades the utility of the tip.

If you are able to access OneBusAway from your platform, please be so kind as to let Link passengers know what length of train is coming, and migrate accordingly.

Counting Available Light Rail Vehicles (LRVs)

Alex’s tip is consistent with having thirteen 3-car trains running all day, and one LRV totaled. The most trains that can run at 6-minute headway, with 48-minute travel time each way, and no more than 11 minutes layover at either end (since only two trains can be in a station at a time) is nineteen trains.

Assume all six peak-only trains are 2-car trains. If “seat slides” (in which operators drive a different train back than the one they drove to the station, after taking a break) were to be enacted at one end, so that only one train maximum is laying over at that station, then those two LRVs from the unused trainset would enable two other peak trains to become 3-car trainsets. Do the same at the other end, and the peak fleet in service would consist of 17 trains, all 3-car trainsets. In other words, from a certain perspective, the presence of 2-car trainsets in the peak rotation could be seen as a result of policy priorities more than lack of available vehicles.

A word about seat slides: Link operators do seat slides when they take over a train already in service, stopping briefly by the maintenance base. Most of you have been on one of those trains at least once, if not many times, so you know they happen. The policy question is then not one of whether seat slides are allowed, but whether they are allowed at Angle Lake Station and UW Station.

54 Replies to “How to Predict 2-Car Link Trains Coming”

  1. I am aware of the so-called seat slides happening at the OMF (the train yard), Beacon Hill, and Angle Lake. Not sure about UW.

    Question, what’s the benefit of knowing whether it will be a 2 or 3 car train that’s coming?

    1. At stations where trains are more crowded it can make a big difference in how much room you have, or chance of getting a seat if it’s less crowded. Enough people are doing the trick of waiting at the tail end of where a 2 car will end then jumping to the third if it’s there, that when it actually is a 2 car train the back end is a lot more crowded then the first car.

    2. It must be going downtown at rush hour, or going toward the stadiums on game day. I usually ride off-peak or reverrse-peak and don’t have to do any of it.

    3. Traveling southbound from Westlake at 5:30 pm, the difference between a 2 and 3 car train is a miserable standing ride until Rainier Station and actually getting to sit down.

      I actually wait at the very back of the platform and don’t bother trying to get onto a 2 car train. Another train will be by in a few minutes, and the extra wait time is worth it.

  2. I’ve always kind of been curious. Does anyone know: What is the cost savings for running a 2 car train vs a 3 car train?

    And why not just run 3 car trains all day and night?

      1. Shouldn’t “bite” on this one, Sam, but it’s New Years’. Obviously, if the train is fully packed, every single waiting passenger can not get on.

        But until everybody gets used to the new conditions, station security will help space waiting passengers for even orderly boarding. Remember that it’s important to leave deboarding passengers a clear way out.

        When it’s finally trains-only, there’ll no longer be any conflict between train riders and bus passengers, whose standing space has always had to be more defined. Anybody that knows: can those sweet yellow labs with their green vests be trained to help out here?

        Until everybody gets on page, the uncooperative can be gently tail-whacked into correct behavior, can’t they?

        Mark

      1. Exactly. The cost difference is minimal. That is why they run 3 car trains outside of rush hour. Sure, they could probably save a tiny bit of money if they ran 2 car trains then, but you might as well allow people to stretch out. But during rush hour, when they are running lots of train, they simply don’t have enough cars to do that.

      2. The text of the post says they can run 17 three car trains and thereby run two fewer trains in the peak with the same capacity and fleet: (17×3=51. 13×3=39 + 6×2=12; 39+12=51).

        Is it worth going at it with the union to get seat slides made policy until the new LRV’s arrive? I don’t know how vigorously the union would push back. But having every train be three cars in length would certainly quickly result in people spreading out more efficiently. That would improve reliability and reduce instances that a train “overstays” its dwell time.

        That seems worthwhile to me.

    1. In the early days of U-Link (a few months after March 2016), they ran 2-car trains normally. It was roughly the opposite of today, with 2-car all-day trains, and 3-car peak trains inserted, and cost of operation was the rationale, so there must be some significant cost difference. Weekends were only 2-car trains. During huge events, they would run “inverted” trains (and run all 3-car trains with extra 2-car trains at peak, just like today), and at some point they switched permanently to “inverted” trains once they deemed ridership was high enough.

      I’ve also heard from some comment on STB that in the very early days of Link (probably in 2009 before the opening of the Tukwila to SeaTac segment) they ran 1-car trains at night to save money;

      ST has an order in for trains that will begin to arrive this year, 122 trains for Northgate (2021) and East Link (2023), and another order for 30 trains within 18 months after for Federal Way and Lynnwood (2024).

      I think when the next batch of trains arrives late this year, they’ll have to gradually put them into service because the cars don’t arrive at once, they trickle in. I would imagine they start with all 3-car trains, peak and off-peak, see how full they are, and changing a few peak trains to 4-car once they are available. When Northgate opens, they will run all 4-car trains all the time.

      1. The first year it was one-car evenings and maybe weekends. Not “night”, it started around 8pm. They’d put an extra car on for event days, but there got to be so many event days that there was something almost every weekend and Link got caught short on a few of them and was overcrowded, so they they started running two cars every weekend and evening, and now three cars.

        Yes, we’ve been waiting for the trains to arrive this year.

  3. Will ST be adding both cars and trains for the Northgate opening? With a doubling of the number of riders north of Westlake in 2021, just adding one car (or two cars to an occasional two-car train) could easily not be enough.

    The problem most disappears when the second East Link trains run in 2023 and frequencies go from 6 to 4 minutes (eventually 3 minutes per ST3). The next North Line issue could somewhat reappear once Lynnwood Link opens in 2024/5 though.

    Solutions for 2021-2023 are easy:

    – an occasional “shuttle” reliever train at the peak of the peak (Muni Metro has a “Castro Shuttle”)

    – running a shorter version of the East Link line early (testing will already require that out-of-service operations will start on the East Link segment by 2022 anyway)

    Will ST be proactive and implement an easy solution like those above, or just ignore the situation and instead plead surprise and ignorance when it happens in 2021?

    1. Yes, all 4-car trains in 2021, at the same 6/10 headway’s as today. 33% more off-peak capacity and 54% more peak capacity. The new cars that will enable the longer trains begin arriving this year but won’t be fully in service until 2021.

      1. I thought they were switching to 3 minute headways with Northgate Link. Maybe that is only when East Link opens (that way it is easy switching, and they don’t have to keep dumping trains at SoDo).

      2. Oh, and why would the train run only every ten minutes in the middle of the day. We have buses that run more often than that. It seems like it would run every six minutes all day long and maybe taper to ten late at night. The corridor Northgate Link will be service seems stronger than, say, the 3/4 — shouldn’t it have better frequency?

      3. Ross, they really shouldn’t run three- minute trains down the MLK median. Six minutes is about the best they can do. Pedestrians need time to cross MLK to get to Link (and 106 bus stops). That’s on top of cars and buses crossing MLK at Alaska, Othello and Henderson.

        The EIRs for East Link and Lynnwood Link as well as their FTA FFGAs show a peak of 8 minutes and 4 minutes respectively. Federal Way Link shows 8 minutes on their FTA application description. The peak 3 minute/ 6 minute operation was merely an ST3 promise with an unclear start date, and is not part of any FTA grant or loan to date.

        Since FTA helps pay for both vehicles and yards, this directly affects the capacity and headways of the system.

      4. Forty cars per hour (10 trains at six minute headways) may be sufficient for Northgate Link, but sixty per hour (15 at four minute headways) probably won’t be for Lynnwood. Yes, that’s a 50% increase, but the Lynnwood extension will intercept over 80 bus runs that pass the county line during the 6:30 to 7:30 AM one hour period.

      5. Six-minute peaks is a temporary measure between U-Link and Northgate Link. When Northgate Link starts and 4-car trains are running it will go back to 8-minute peaks as it originally was. I forgot about that too sometimes. Then when East Link starts in 2024 the northern frequency will double to 4 minutes.

        Of course, if ridership is higher than projected, ST may have to revise its plan, and bring back 6-minute peaks. But will we again run into a ceiling of trains?

      6. I recall the earlier peak frequency being “7-8 minutes,” which in actuality was 7.5 minutes (this was also before there was a schedule, but even a schedule wouldn’t show half a minute). That matters for a 15 minute bus connection. It sounds like the new frequency will be a solid 8 minutes per line, which means not only shifting relative bus timings, but also different timings per hour. One thing I like about Link is its regularity (like at Capitol Hill in the evening, I know Link comes “on the 5s,” like 9:45pm, 9:55, 10:05…) which is more important of the headways increase.

        But the problem with doing it the other way is that the combined headway would be… 3.75 minutes, which is even weirder. Personally, I’d have no problem with dropping some trains down to 3 cars if frequency can be increased to 6 minute headways each line. Maybe if ST had decided we can never have a center platform at the existing IDS, this is what we can ask for in return.

      7. @Al — I know the trains are limited to every six minutes down Rainier Valley. What I was thinking was running every three minutes north of SoDo. Half the runs go down to Rainier Valley, the other half end at SoDo. Like I said, though, maybe they don’t want to hassle with that, and are instead just going to wait until East Link.

      8. Six-minute peaks is a temporary measure between U-Link and Northgate Link. When Northgate Link starts and 4-car trains are running it will go back to 8-minute peaks as it originally was.

        So when Northgate Link opens. headways will get worse?!! Why do we always have to take a step backwards when we take two steps forward.

      9. Forty cars per hour (10 trains at six minute headways) may be sufficient for Northgate Link, but sixty per hour (15 at four minute headways) probably won’t be for Lynnwood. Yes, that’s a 50% increase, but the Lynnwood extension will intercept over 80 bus runs that pass the county line during the 6:30 to 7:30 AM one hour period.

        A lot also depends on when the various agencies decide to truncate their buses. Sound Transit has already said they will truncate the 522 with Northgate Link (at Northgate or Roosevelt). Metro will undoubtedly truncate every downtown express that runs through Northgate or places south (41, 309, 312, 74, 76. etc.). Metro only runs a handful of buses north of there and some of them would likely be truncated as well.

        Sound Transit runs about 15 buses an hour. But the big one is Community Transit, with somewhere around 45 to downtown (and some to the U-District). I don’t think the U-District ones matter, as that is not where the crowding will occur. For every rider that gets off southbound at the U-District, there will be two that board there or at Capitol Hill. I think it really depends on what CT and ST wants to do with those express buses. A lot of those are Lynnwood to downtown. If one agency truncates at Northgate, the other will likely follow. In my opinion, there would be a lot to gain if those agencies truncated, but many are very conservative, and would rather just muddle along, instead of immediately taking advantage of one of the biggest improvements in mass transit the state has every seen.

      10. “So when Northgate Link opens. headways will get worse?!! Why do we always have to take a step backwards when we take two steps forward.”

        I’ve gotten used to 6-minute peaks and it’s gone for so long that I’m not the only one who forgot that it was a temporary expediency to manage U-Link’s demand because ST didn’t want to increase train length. It has to do with the ratio of spare trains available and some kind of hour-based warranty. If it runs trains in operation before the extension they were intended for, it uses up warranty hours that won’t be available after the extension opens. However, Link became more popular than expected and so ST has backed into both increasing frequency and train length.

        Now that it has happened multiple times, it could happen again. The Northgate and I think Lynnwood Link train orders were placed before this phenomenon became clear. And I’m not the only one who has gotten used to 6-minute peaks. When I look at the sign and see 6 minutes I think, “Soon it’s going to be 4 minutes and will just get better”, not “Soon it will go down to 8 minutes and I’ll have to get used to waiting almost 10 minutes peak hours again?” So there’s a large passenger expectation now of 6 minutes minimum, and ST may find it harder than it thinks to go back to 8 now, especially if ridership keeps increasing.

      11. A short Link train that you recommend is one of my suggestions in the original post, Ross! I think we agree on that.

        The reason that I called it a “shorter” East Link line is because that option seems to make it easier to reverse trains with the two new tracks available to switch train directions and provide a buffer in time slots. Using Stadium third track or the light rail yard to reverse trains seems doable too — but the logistics could be more challenging.

        My bigger point is that 2021 is here in two years. A solution requires systems planning and driver and car scheduling and that probably needs to be in process now! I’m not sure if any capacity enhancement besides a fourth car is on ST’s work program.

        If ST does Northgate operations badly and ignores potential capacity problems in 2021, they should not be able to get away with that mistake and call it as a “surprise” and an unanticipated problem.

    2. As a non-engineer, I hate to argue with the ST engineering department, but hey, they got it wrong on what the most valuable use of the center of ID/C Station is, so it is possible they could miss other important details.

      The fact that the mock-up bridge used to test Link’s impact on floating bridges was tested using only one LRV (and full of enough weight to simulate a full passenger load?) gives me trepidation about how quickly Link operations will wear and tear on the I-90 center floating bridge.

      But more importantly, won’t 4-car trains running every 8 minutes exert more pressure than 3-car trains running every 6 minutes? The impact of a 4-car train is 16/9 times the impact of a 3-car train in certain factors, is it not?

      If the problem is the bridge can only carry the weight of one train at a time, that may present a problem if live operational testing reveals the trains will have to run more slowly across the bridge. But then, wouldn’t distributing the weight with shorter trains work better?

    3. I can’t imagine CT truncating at Northgate, unless there are bus lanes provided on First NE. Even then the solid line of buses from the west on Northgate Way would be horrid. Plus the buses would have to cross four lanes of traffic from the HOV. They’ll keep going downtown in the express lanes.

      1. If CT or Metro were going to truncate at Northgate they would have started planning it long before now. I expected ST2 to only reach Northgate and was surprised when it included Lynnwood. I think that was the turning point. In deciding to extend Link to Lynnwood two years after Northgate, they decided not to truncate buses at Northgate. (It has now slipped to three years.) Truncating hundreds of buses at Northgate would require more layover space and new bus ramps to I-5; otherwise they would get caught in the already at-capacity traffic between the northern freeway exits and the transit center. There’s no reason to spend capital money for ramps and transit center space that will only be used for three years. Better to leave the buses on the freeway now and finish Lynnwood and the stations down to 145th sooner.

      2. There are only so many express Metro bus routes north of the ship canal, 13 of them by my count: 41, 63, 64, 74, 76, 77, 301, 303, 304, 308, 312, 316, and 355.

        Expect route 63 to go away after Northgate Link opens, with increased service on routes 8 and 70, among other local routes. Expect route 76 to also go away.

        Expect route 74 to terminate at U-District Station.

        Expect routes 64, 77, 309, 312 (along with ST Express 522), and 355 to terminate at Roosevelt Station, at least until 522 BRT rolls out.

        Expect routes 41 and 303 (since it already serves Northgate TC) to terminate at Northgate Station.

        That leaves routes 301, 304, 308, and 316 as the only north-end Metro routes likely to keep expressing downtown. 301 and 316 already serve the U-District, so diverting them to U-District Station seems an obvious non-controversial way to save service hours until Lynnwood Link opens.

        The later Metro starts the restructure planning, the fewer rounds the one-seat-ride lobby has to undo the restructure. But I don’t foresee much of a lobby for keeping the other routes with obvious Northgate Link connection points duplicate-heading all the way downtown.

        For CT routes, getting off I-5 and going a few blocks to U-District Station is a much more palatable restructure than coming within a few blocks of UW Station. Buses that continue downtown pretty much have to finish their morning and lay over at a CT-leased lot in the SODO. CT buses going to U-District Station could conceivably go back to Snohomish County in service, picking up reverse commuters, and then do another peak-direction run.

      3. The situation seems quite similar to what happened when we built U-Link. Just about everyone assumed that the 71/72/73 would continue to connect the U-District to downtown. After all, truncating at UW station was cumbersome. The station is not designed as a bus intercept. Buses would have to spend a lot of extra time getting to the station, and riders would have to walk a long way to make the transfer. And the area might be really crowded with buses. It just made sense to wait a couple years, and make the obvious truncations when we build the U-District station. If people really wanted to truncate those buses, they would probably talk about that years ahead of time, not right before U-Link is complete.

        Yet that is exactly what happened. They truncated the hell out of the buses. In many cases, riders did lose a lot of time. They were forced into transfers that are way worse than any possible with a truncation at Northgate. The transfer is bad at UW Station, and the time the bus spend serving it is bad as well. I’m sure there are times when the slog through the U-District or UW is more time consuming than the trip downtown, what with no special bus lanes most of the way, and the bus spending time picking up or dropping off people along the way. I’m sure there are lots of riders that hate this. But others love it. They love it because it has enabled the kind of frequency in the region that changes the way you look at transit.

        It is a trade-off, and one that makes *more* sense for Community Transit than it did for Metro. A bus in the U-District is very close to the freeway, and very close to downtown. A bus at Northgate may be on the freeway, but it is still a lot farther away from downtown. The savings from truncating there would be huge, especially for deadhead buses (which make up the bulk of the buses there). Getting to downtown in the evening — just to pick up riders and send them back — is horrible, and typically starts right around Northgate. By around 3:00 in the afternoon, traffic is often moving at a crawl. The savings would be enormous.

        Yes, Tom, a bus would have to change lanes four times. Lots of buses do that every day. The 512 does that ever 15 minutes between 145th and Mountlake Terrace and all the 800 series buses (serving the UW) do that as well. Traffic around that part of Northgate is usually not terrible, and when it is, it is much worse on the freeway. For a bus headed to the station it wouldn’t be bad when it mattered: in the morning. There aren’t that many people who use that exit, nor are there that many people headed eastbound on Northgate Way (those headed south on the freeway would use the 107th street ramp). The only significant congestion occurs when an express is dead-heading (going to Northgate in the evening) and that is still a lot better than slogging through downtown.

        At the same time, the benefit would be huge. Yes, some riders, on low traffic days, would lose time. But they would lose way less than riders who used to take the 71/72/73. Meanwhile, they get the added benefit of five additional stops. Someone who rode the 71/72/73 already had a ride to the U-District — the new buses didn’t really get them anything except a good connection to Capitol Hill (something many had already). But someone riding a truncated 400 series bus would get a direct ride to Northgate, and an easy connection to Roosevelt, U-District, UW, Capitol Hill and downtown. As it is, Community Transit runs direct buses to the UW right now. The UW is already enough of a destination that they run buses to both locations. Do you think it makes sense to do that, when it will take only five minutes to get from Northgate to the U-District, and the train will run every six minutes?

        Look, I get it. On low traffic times an express — any express — is faster. When Northgate Link is built, the 41 would be faster for a lot of riders to downtown. When Lynnwood Link is built, the 402 is still faster for some than Link. Same with Ash Way and every stop to the north. But that is just the nature of trains. They make stops. and those stops take time. Transfers take time. But if you have an area with *lots of stops* that people are actually using, then the truncation makes sense. Lots of people go to the UW. Lots of people go to Capitol Hill (which also happens to be close to First Hill where even more people are headed).

        I have no idea if this happen. It is quite possible that CT will chicken out, play it safe, and do what everyone assumed that Metro would do. What I do know is that it would make a lot of sense for CT (and ST and Metro) to truncate at Northgate.

      4. The U-District restructure truncated *LOCAL* routes, not peak expresses, while you’re expecting CT to do the opposite. Yes, the 71/72/73 were nominally expresses, but the ones that couldn’t use the express lanes were barely distinct from “locals” and a lot like the sham-express 66. Also, Community Transit is not Metro, and Snohomish County ratepayers aren’t like King County. When CT proposed a heavily in-county oriented restructure during the recession cuts, the majority feedback was “No! The expresses to Seattle are the best value for our tax dollars!” Because there’s nothing equivalent to N 45th Street or 35th Ave NE or 15th Ave NE in Snohomish County, where just a bit more bus service can make the difference in making a lot of destinations accessible by transit, because there are no such concentrations of destinations. Sure, you’ve got a route to the Lynnwood Fred Meyer, but you can’t do everything at Fred Meyer, and some people want Safeway or Costco or a great mom n pop shop. So they come to Seattle for things Snohomish County doesn’t have. I forgot about the CT-UW routes; truncating just them and the 512 would be a lot easier than truncating the much more numerous downtown routes. CT can’t just “decide” to truncate them; it would have to work with WSDOT and ST on capital improvements at the Northgate exit and at the transit center. The Northgate traffic is already at capacity with the existing cars and buses; otherwise we wouldn’t have objected to the 67 and 75 and other routes getting caught in that traffic. If CT tried to truncate the routes there, SDOT might not even allow it because of this, the way Metro has to petition SDOT to use a new street and sometimes doesn’t get the narrow ones.

      5. @Brent — I agree with you about Metro. Just about everything will be truncated.

        Sending CT buses to the U-District (instead of Northgate) is another possibility. Since they already send buses there, it would be simple (no surprises). It would mean serving the second most popular destination with a direction connection, and make for an easy, fast transfer to downtown. I still think it makes more sense to truncate at Northgate, just because the service savings would be bigger. Getting to the Northgate station from the freeway wouldn’t take that long, and getting back on the freeway would be extremely quick. It would be easier than going directly to the UW, even though the UW is obviously a much bigger destination. Both sound better than pretending that Northgate Link doesn’t exist.

        ST is actually a bigger player than Metro (although tiny compared to CT). As you said, just about all of the Metro truncations are easy. But ST has the same decision as CT does — continue to run the express, or ask people to transfer. The big difference between CT and ST is that Community Transit only runs commuter buses into Seattle (peak direction buses). Sound Transit has the 512, which runs all day. This bus should also be truncated (in my opinion) even though traffic is less of an issue, and the train runs less frequently in the middle of the day. There is still some traffic once in a while, but the tipping point is service to the UW. The 512 makes only a freeway stop at 45th, which is a long ways from the UW or the U-District. A Northgate truncation, or a direct loop in the UW (like the CT 800 series buses) would be much better.

      6. One of the deciding factors on where CT will send its commuter armada after Northgate Link opens is where the double-talls can go. After Lynnwood Link, I have no idea what CT will do with all those double-talls.

      7. @Brent, I suppose they will terminate at either Lynnwood or a destinations on 405. CT still has a large service area both north and east of Lynnwood, including cities outside of the ST service area (Marysville, Lake Stevens, Monroe, etc), so plenty of route options to put those buses to work.

  4. well .. if you’re interested in knowing if the coming train is 2 or 3 cars long, just pay attention to the wind it generates. As vehicles travel through the tunnel they push a column of air in front and the longer the train the more air gets pushed. So, the stronger the wind the longer the train :-)

    1. Or just look at the conductor. If he is driving a three car train, he just has a certain swagger about him. With a two car train, he is sheepish, almost ashamed, like why bother.

    2. Ah, no, this is not true. A piston in a tube will push the same amount of air regardless of how long the piston is. An LRV obviously isn’t a perfect piston, but side flow and entrainment effects are both higher order terms and have no easily discernible effect.

      Now if 2-car trains travelled faster than 3-car trains you could feel a difference, but they don’t. 2 and 3 car trains travel at the same speed.

      1. no quite .. since this isn’t a closed system. Unlike a piston, the train has an air gap between the train and the walls allowing SOME of the air to escape. But the longer the train, the slower the air can escape allowing for more air to be pushed forward. Think “back pressure”.

    3. There has got to be an app out there that detects the oncoming force of the train as it approaches the platform.

  5. “If “seat slides” (in which operators drive a different train back than the one they drove to the station, after taking a break) were to be enacted at one end, so that only one train maximum is laying over at that station…”

    No, no no. It’s seriously tempting, but that will kill Link reliability at peak. Trains are super reliable and run much less late compared to buses, but they’re not magic. It’s very common to see a train at peak run 1-3 minutes late, and it is because they are packed to the gills, and it often takes a long time for a passenger in the middle of a crowd to get to the doors. To me, I’d say a 3-minute late train is as surprising as a 10-15 minute late bus, which is very good but not perfect.

    With layover times of 6-11 minutes, a train running 1-5 minutes late is totally recoverable, but if we reduce that to, say, 1-6 minutes, then a delayed train can eat the entire layover time. It will take a minute or so longer to get the train running again due to the seat slide, and the unreliability will accumulate over many runs after that. Train bunching will occur, even with spacing out the trains. It’s easy to imagine a 10-12 minute wait for a train that fell behind, followed by a 6-minute train that was intentionally delayed to keep 6-minute headways. Maybe it will be easier to do this after we have 4-car trains, which will help reliability not just because the capacity is higher, but also because there will be 16 (I think?) doors that can all be used in parallel, as opposed to the 8-12 on today’s trains.

    1. One time I was waiting for a CTA train headed for O’Hare leaving downtown Chicago at 4 pm. There had been some sort of minor train delay and the train cars were stuffed. The train was getting further delayed because it took so long for riders to squeeze on and off at each station.

      The operator then made this announcement: “After the next stop, this train will be an express train to Jefferson Park. If you are heading to a closer station, please exit and board the train right behind this one.”

      Boom! About 30 percent of the train exited, and the train was able to gain several minutes on the delay.

      At first I felt badly for the local riders — but quickly I appreciated the systems benefit!

  6. It’s unfortunate that ST can’t use the electronic and audio announcements to indicate what length train is arriving. BART does, and hopefully when the displays are upgraded ST can as well.

    1. ST will be running all 3-car trains by the end of this year. Then, each 3-car Siemens train that replaces a Kinkysharyo train in the service rotation will have enough space to carry almost as many riders as a Siemens 4-car train would. When Northgate Link opens for service, all will be 4-car trains, forevermore.

      So, it is almost too late to bother with train-length announcement technology.

    2. It is amazing how the changeable message signs say “Welcome to …” when the station name is already plastered all over the station already, as well as say “Sound Transit Link Light Rail” even though that’s painfully obvious and announced everywhere else too.

      It’s as if the signs aren’t being put to the best use most of the time. Then, when non-destination messages appear, they are scrolled as a complete sentence so a rider can miss part or all of the message — when all they have to communicate can be a succinct “Next Train: 2 Min” and not be scrolled.

      A UW systems engineering case study team could have a field day critiquing them!

    3. It’s like the trains saying “Sound Transit” when they’re not in service. Is ST trying to plaster its brand in everybody’s face like it were Coke or Pepsi?

      1. I wish it were a black sign with yellow letters instead of a cartoonish TV/smartphone-like picture. I see enough blue TV screens in a day.

  7. What a disappointment for the fanboys!! ST is already hedging on better frequencies for segments that are years away. They do however have the money to send me a mailer that Light Rail is coming to my neighborhood of Belltown. Link is just one bad joke after another..

    1. It must be a miserable existence having buses pass through your neighborhood every minute all day.

  8. I hope STB poses the question raised by RossB above to ST. off-peak Link headway and waits should be shorter.

    Also, STB could ask about bus headway. off-peak express bus headway and waits could be shorter on routes 512, 535, 522, 545, 550, and 574. ST has buses available. the portion of the budget spent on operations is very small.

Comments are closed.