Northgate Station on a recent, not-raining day

It’s been a few months since our last check-in with Northgate Link, and things have changed dramatically around the three stations. Holiday shoppers no longer throng Northgate Mall, which is now split in two and without several of its longtime tenants. Roosevelt has gained the first of two cross-streets and welcomed a few new apartment buildings. U District Station now has furnished entrances and lighter fencing.

There is just about two years left until Sound Transit’s due date for Northgate Link, which is set for September 2021, but the project is currently sitting on enough float time to open months earlier. Sound Transit says that, as of this week, overall construction on the project is 95 percent complete and should be turned over for systems installation soon next year.

Plenty of pictures after the jump.

Northgate Station

Northgate Station as seen from the mall parking garage

Let’s begin with Northgate Station, the soon-to-be gateway for shoppers and hockey players alike. The platform is fully furnished with glass windows, signposts, and escalators all the way down to street level.

Looking down towards the express lane entrance

The low-slung Northgate Station parking garage has been open for a full year now, but was only half-full at the time of this check (on a Wednesday at noon). The top floor remains closed for future work on connecting it to the mezzanine level of the station.

South side of the Northgate Station site
Bus bays and entrance at Northgate Station

The south side of the station has an exciting development: bus bays. The relocated transit center will be tucked right under the station, with a clear path up the escalators or stairs to the mezzanine before turning around to reach the platform. Dozens of bus routes, perhaps including a few from Snohomish County, are set to terminate here, and their passengers will find standard Sound Transit-style shelters that are wider and roomier than Metro’s offerings.

Northgate Mall demolition in progress

There is also plenty of activity in the periphery of the station, especially at the mall. The old Bon Marché store, temporarily part of the Macy’s brand, has lost its facade and is being clawed away from the outside. One can no longer walk through the entire length of the mall, but are directed towards the sea of parking lots to continue a shopping trip. It’s a strange coma-like state for our region’s first suburban mall (and one of the first in the country), but as Charles R. Cross said to Crosscut, there is a boundless emptiness to the mall without its stores or people.

Roosevelt Station

Hard-hatted media at the north entrance plaza

To celebrate substantial completion on the construction contract for Roosevelt Station, Sound Transit invited members of the media on a journey down to the platform.

Looking towards the escalators and stairs from the south entrance
The preserved Standard Radio sign

We started from the south entrance at 65th Street, which will serve as an effective transfer to nearby buses and the protected bike lane network of the neighborhood. The ticketing hall is bright and well-lit thanks to the windows in the roof, and comes installed with a small piece of local history: the preserved facade and neon signage of the Standard Radio building, a local landmark that once stood here.

Bus shelter at the south entrance

Before we go further inside, a quick look back at the new bus shelter for Routes 45 and 62 on 65th Street. The station entrance is five steps above the sidewalk on 65th Street due to the slope of the site, so there is also an accessible ramp to the sheltered bus stop.

Escalators from the mezzanine to the south entrance

While we weren’t allowed on the new, heavy-duty escalators, Sound Transit was happy to show off a new feature: slow mode. When pedestrians aren’t detected on the escalator or approaching from either end, these escalators are able to conserve power and prevent unnecessary wear and tear by running slower. With peak train frequency of four minutes once East Link comes online, don’t expect to see slow mode all that often.

The small mezzanine, where you have your choice of stairs or escalators for onward trips
Looking down the mezzanine towards the north entrance stairs/escalators

In case the escalators are stuck in the “off” position, there’s a set of parallel stairs. In fact, there are stairs down to the mezzanine from both entrances and another pair down to the platforms. The two escalator banks from the platform do meet at the center of the mezzanine, which is much better suited for a station of this size than UW Station.

The lower set of escalators also have a fun bit of lighting that creates “waves” on the steps

Sadly, there is no option to slide down to quickly catch your train. We’ll hold out hope for U District Station. The walk down to the platform on the stairs does, however, include a bonus mezzanine that can be skipped on the escalator or elevators.

Looking south down the northbound track, with a cameo from a future real-time arrivals sign
Low angle view from the southbound side of the platform
The stairs and lower mezzanine from platform level
Mid-mezzanine view of the platform

While the stairs felt a bit claustrophobic, with little room to pass, the platform is just the opposite. Much like Capitol Hill Station, there’s plenty of room to look up into the rest of the station, but there’s plenty of columns to break up and subdivide the space.

Platform level artwork from Luca Buvoli

The highway ramp-like columns are a bit bare at the moment, but the underside of the stairs and escalators are painted in an almost-Husky shade of purple that creates a good contrast. There is also a nice pop of color coming from the art on the platform, a series of pieces by sculptor Luca Buvoli that are inspired by the motion of runners and cyclists, a good fit for the closest station to Green Lake.

A surprise guest at platform level
Looking down the southbound tube of Northgate Link

While Roosevelt Station has now reached substantial completion on the construction side, there is still plenty of work left to do. Systems installation will be the most obvious, as the station lacks overhead catenary to power Link trains and real-time signage. One of the only forms of transport in the Northgate Link tunnel at the moment is a hi-rail diesel truck, which made a surprise appearance during the media tour.

The bike plaza, tucked under the north entrance at 66th Street

The media tour ended in the plaza level at 66th Street, which will lead up to the new transit-oriented development on the former QFC parking lot to the west of the station. The 66th Street plaza is also home to a bicycle center that will have a cage with space for 20 bicycles and 10 on-demand lockers (with room for 20 more), along with 44 exterior bicycle racks.

U District Station

The southernmost of the new stops was also the last to be reached by the boring machines and is a bit further behind. The station entrances now have most of their structural elements in place, with bare openings and steel left to be covered.

The north entrance at U District Station
Looking south on Brooklyn Avenue

The north entrance at NE 45th Street has been bolted to the back of the Neptune Theater. Street restoration on Brooklyn Avenue should begin soon, as it is already graded with rough pavement.

The south entrance at U District Station
Looking through the fence at the south entrance

The south entrance at 43rd Street is much further along, with pavement for the new bus/bike corridor in place and the first windows waiting to be installed.

That’s it for now, but expect another pair of project photo updates in the near future. Unfortunately, the pace of construction for East Link has often leapfrogged my slow photo editing, so an end-of-the-year roundup is all that I can promise.

81 Replies to “Finishing touches for Northgate Link as work continues below”

    1. It’s not a binary on-off; it’s a gradual loss of usefulness. Bellevue Station is a half block and across the street from the bus bays, and further from Bellevue Square or 106th (a closer geographic center). Columbia City Station is three blocks from the center of Columbia City. Ballard Station is threatening to be at 14th, which is effectively 8 blocks east of the center at 20th (halfway between Swedish Ballard and Ballard Ave, with dense housing north of it). These aren’t fatal but they make the line less convenient for the bulk of the travelers. Each block or intersection you add deters some people, and it adds up cumulatively. People have an expectation of shorter walking in denser areas. If you add to that, it looks substandard, not what New York or London or Munich would do, but what a car-centric government would do.

      Downtown Bellevue, Seattle, Capitol Hill, the U-District, and Ballard have tried to make themselves significantly walkable and pleasant to linger in, like a northeastern city center. In Northgate you can walk from the station to Northgate North or the apartments east of it, but it still feels like you’re walkng in car-land across superblocks. That already depresses ridership, so the distance of the station is a smaller factor. Whereas in Ballard, people want to go to Ballard because it’s Ballard and a prewar-style neighborhood, so the distance of the station is a more primary factor in why they’d drive or skip going to Ballard.

      1. Mike, you keep on talking about Ballard as if it’s this static thing that will never change and evolve. That the way it is is the way it will always be. Look at all the projects in the pipeline near Bellevue Station. They are plentiful and they are large. And they are happening, in part, because a new light rail line and station is coming. If the station is built on 15th in Ballard, eventually, that will be the bulk of people’s destination.

      2. The problem is not the size and shape of the buildings but the design details of the buildings and lots. Some people prefer modern architecture and layout, but on the whole it makes a place that people “have to be at” rather than “want to be at”. More critical are the walking paths, relationship of pedestrians to cars, and small blocks and streets.

        Prewar storefronts are usually narrow and vertically-oriented and come right up to the sidewalk. That gives a choice of more businesses per block or within a 5-minute walk, and their close front door makes it easy to step into them. Postwar and recent buildings are usually wider and are designed for the convenience of drivers rather than pedestrians, and have gratuitous extra space you have to walk through or past. Even when the building is vertical, has no setback, has parking unobtrusively underground, it’s usually not as convenient or intimately-scaled as prewar buildings. It’s grossly scaled to look nice from a car.

        Developers can build better buildings but they refuse to. The best ones I’ve seen keep a prewar first-story facade and put a modern building behind it and above it. The most important part is how it interfaces with the sidewalk, and at least then they get that part right. But when they design completely new buildings, they don’t.

        North of Market in Ballard has new apartments with all the problems I’ve described, but at least the street grid is small so that makes it more walkable and pleasant and the buildings’ negatives are limited.

      3. I don’t live in Ballard, but I have walked and ridden (a bike) across it multiple times and I don’t see it actually being that hard to get eight blocks in Ballard. Or two blocks, since 20th (very arguably the Apex of Ballard) is only actually two longish blocks away. Even the Norse Museum. Eight “blocks” from 15th is actually all the way to the locks. Conversely, if the station was built West of 15th, it would mean tearing down some of that marvelous prewar architecture (I may not agree with you that newer sensibilities are worse, but we can agree that Ballard character is important and shouldn’t be torn out).

        Ideally the station would sit adjacent to 15th on the East, with entrances on the West allowing people to skip crossing 15th before hopping on the train. I’m sure concerns over homeless occupation will prevent this, but it would be pretty logical positioning without pushing into the most desirable property in Ballard.

      4. You are a physically able-bodied individual. If you were not, those 8 blocks are longer and more difficult.

      5. The station’s success depends on what Americans actually do, not on what we think they should do. Transit planners have a lot of studies on this, and we can see how people feel about the metro networks in various cities, whether they think it’s convenient and would be their first choice, and whether the majority of the population chooses non-car modes. That’s the goal, getting driving rate below 50% or as close as we can, as many cities are now and Seattle was before 1940. If you really want most people to take transit, it needs to be very convenient and plentiful.

        My impression for Ballard is 20th would be ideal (underground to avoid disrupting the historic core), 15th is borderline, and 14th is substandard. That’s a judgment based on my own amateur impressions, and you can argue it’s really slightly longer or I’m weighing the factors wrong, etc. The issue is what most people would consider a high-quality location or the station, and at what point less than that do people’s satisfaction become significantly less.

        ST says 14th would not lose riders, but it has also said I-5 is as good as Aurora, 130th will not gain riders, First Hill Station is not a priority, etc. Those may be accurate according to ST’s metrics, but it suggests ST’s metrics are incomplete. And the issue is not just how many butts are on the train, but how well it serves the neighborhood and the city. ST’s metrics seems to be missing some intangible factors, or factors that aren’t prioritized enough here.

        Another factor is how many people come from outside the neighborhood. Residential areas generate ridership only from the residents and the few people visiting them. Having a few businesses of common chain stores doesn’t add much: people don’t go to a particular Subway or Bartell’s, they go to the closest one. But having unique draws like a large number/variety of businesses, an exceptional chef or yoga teacher, a store that has something no one else does or high-quality staff, or unique non-commercial amenities, creates bidirectional all-day ridership, sometimes twice as high as the unidirectional case.

        Roosevelt Way has grown but is still not as popular as University Way, and I predict east of 16th Ave NW will be the same. These generic modernist buildings attract generic chain stores. There are exceptions, but the overall direction is this. This is partly because chains look for a wide storefront, ideally on two sides, to maximize visibility.

        Downtown Bellevue has mixed results. It certainly draws a lot of people and is successful, but I contend it would draw more and they’d be more satisfied if its post-1990 development had been designed differently, putting pedestrians and traditional aesthetics first.

      6. A kind of random but related thought regarding Ballard and West Seattle Link. With Ballard, one way to look at it is the 15th Ave. location has the greater potential for growth, some of which is already happening. Which is a pretty solid argument especially when you add on the potential to extend north along 15th (even if I’d rather get off the station a half mile to the west). Now, for West Seattle, there are apartments recently built and going up along the curve of Fauntleroy, a few blocks east of the Alaska Junction, but still a relatively easy walk. And yet, I haven’t seen the same argument made for having the station in the area of the potential growth in West Seattle, which (admittedly based on just visiting a few times) seems to be more towards Fauntleroy and Alaska than California and Alaska. The bonus: building elevated following Fauntleroy doesn’t require demolishing homes, though you might need to take out a surface parking lot or two for construction staging. And you end up with the tracks pointing on the “right” direction: Southwards. It looks like there is an addition station or two along Fauntleroy that would have some medium density, at least right along the main road.

      7. @Phyzzi — Different people have different tolerances for walking, and these depend on the circumstance. It is important to keep in mind that not that many people live right on Market. So if you are at 57th and 24th (where lots of big apartments are: then it is quite a schlep to 15th and Market, and much worse for 14th. Studies have shown that there is a limited tolerance for such walking. We can also see that poor station placement can make all the difference in the world. On paper, a station close to Franklin High School, at the convergence of some of our most popular buses, seems likely to gain a lot of riders. But because the station placement is awful, it lags just about every station in our system. Riders simply take the bus, as if the station wasn’t there.

        My impression for Ballard is 20th would be ideal (underground to avoid disrupting the historic core), 15th is borderline, and 14th is substandard.

        Yes, that is it exactly, and you aren’t alone. Lots and lots of people have said the exact same thing.

        With Ballard, one way to look at it is the 15th Ave. location has the greater potential for growth, some of which is already happening.

        My grandkid is now three feet tall. She has tremendous potential for growth. Somehow though, I don’t think she will be as tall as Kareem Abdul-Jabbar, no matter how much milk she drinks.

        There is just as much growth on 24th as 15th. Very little of it is happening east of 15th. As this diagram shows, a station at 20th is at the center of the growth. The arguments for that station are as strong as ever — whether we can afford it instead of a barely acceptable, “borderline” 15th is a different matter.

    2. Not to quibble too much, but the shopping center at Rainier and Andover probably has more activity and more commercial square footage than the Columbia City business district has. Cute storefronts and trendy restaurants are great, but I would rather see a neutral perspective based on data rather than character.

      1. Rainier and Andover? Dude, just look at Rainier and Andover ( It has a big parking lot that takes up over half the retail area. Across Andover is the Darigold Warehouse. On the other side of the street is a strip of green and some houses on big lots. Overall, it has low population and low employment density. In contrast, the area Mike is talking about has a lot more shop fronts, and a lot less in the way of parking lots. Oh, there are some ( but it is a more vibrant, pedestrian friendly area, with a lot more people, day and night.

        Neither are like the heart of Ballard, which has way more people. But the contrast is similar. While 15th and Market has a lot more in the way of retail than in the past, the land use is dominated by big grocery stores with huge parking lots: These are not places that people go to via a train (whose nearest stop is well over a mile away). Again, these are not pedestrian friendly areas, but areas designed for drivers. 14th is worse.

        Of course things could change, but you have to pretty naive to assume that historical architecture and character are meaningless when it comes to attracting people to the area. That’s like saying that the UW station will suddenly become fantastic, because the UW will get ride of the stadium (which sits empty 350 days a year) and the parking lot surrounding it. Instead they will build high rise buildings there, making the choice of location seem like inspired genius. Likewise, the area surrounding Mount Baker station will suddenly, magically, become similar to Capitol Hill, and ridership there will rival it.

        Sorry, that isn’t gonna happen. The flaws in station design will continue to be flaws. Some are worse than others, but we are pretty much stuck with them.

      2. Oh, the Safeway plaza. I was thinking of the other shopping plaza, Rainier Market? I haven’t paid much attention to those so I have no comment on them. So a line emphasizing those would move Columbia City Station a dozen blocks north. Where is the other plaza, I can’t find it on the map but I’ve seen it on the 7; it’s facing south rather than west. Genesee? Anyway, to serve the Safeway plaza you’d need a walkway on probably Bradford Street. Neither Andover nor Charlestown streets seem to exist much between MLK and Rainier. There’s also two hillsides in between. And the distance from MLK to Rainier is long, maybe as long as Rainier Beach. So unless you put Link on Rainier (which I preferred), it doesn’t seem feasible.

        Columbia City was chosen to emphasize the inherited neighborhood centers. As in Ballard, prewar centers have more potential than new creations. Columbia City had been revitalized starting in the 90s. The Safeway plaza, while it could theoretically surpass it, would take a lot of work. You’d have to move the buildings in front of the parking lot or put it underground, and catalyze a wider variety of uses. Neither the city nor the neighborhood is ready to champion that.

        The Safeway itself may draw more people than Columbia City center, but it’s not a very good Link station target. People going to that Safeway will drive or take the 7. Everyone else will go to their nearest Safeway, or if they can’t get to it on transit easily, they’ll go to the one adjacent to Othello Station. Walking from MLK to the Andover Safeway is not even feasible: the distance is far, there’s the two hillsides, and it looks like you’d have to extend a street or punch through a business.

      3. Rainier and Andover has a shopping center over 100,000 square feet!

        Dude, look at the map. It is mostly parking lot. As in, over 50%. Of course you have people shopping — but it is essentially a suburban strip mall surrounded by parking, and yet that is the area with the best land use (the surrounding areas are even worse). The front is a huge parking lot, while the back is this: Here is the side: (Wow, look at all that retail). It is not the least bit pedestrian friendly. Much of that square footage is essentially warehouse, as opposed to store front. The biggest client is a grocery store, for crying out loud. It just doesn’t employ or attract that many people per square foot of land.

        In contrast, look at the area that Mike is talking about: This is a real city. On the map, you can literally see people sitting and enjoying lunch on the street (you didn’t see that at the mall). This is the type of place that employs a lot of people, and attracts a lot of people. The point is, it is quite likely that someone from outside the area will want to go to an area like that, unlike the mall. Someone might want to go to the Royal Room to take in a show. At that point, they will definitely be in Columbia City. Even though Link has a station called that, the walk from the club to the train station is just too far.

        I’m not saying ST did anything wrong. I’m just saying that pretending that Link serves the cultural heart of Columbia City is silly. To suggest that more people are headed to a strip mall is nuts.

      4. I think it’s more an illustration of how small the Columbia City business district really is. It’s great — but small. There is even no longer a Starbucks (although the nearby shopping center has one).

        The redevelopment of the station area has been mostly residential with some modest one-story commercial along Rainier or MLK but not in-between. Had the redevelopment of Alaska between the two streets been done with the intent of having a new east-west shopping and restaurant street, things would be entirely different.

        I think that mistake won’t be repeated in Ballard or West Seattle if their stations aren’t in their commercial core. Market and Alaska Streets are already designed as shopping and restaurant streets in those cases. It would be like putting the West Seattle station at California and Dakota, or the Ballard station at NW 60th St NW and 24th Ave NW.

  1. I notice both up and down escalators at Roosevelt. The down escalators are features coming at only the stations in the nicest neighborhoods.

    1. Moreover, they are only to be found at non-at-grade stations!

      Up/down elevator parity is a worse outcome than not having no stairwells. The larger problem is not having sufficient pathways to always be able to clear the station platform faster than people can get down to it.

      Opening up some formerly emergency stairwells should work as long as they are either down from the platform or only have signage pointing to them to get up to the street level.

      The largest nuisances remain the lack of accessibility that frequently happens, and for weeks in duration, at SeaTac Airport Station, Tukwila International Boulevard Station, and Mt. Baker Station, because each elevator is a single point of failure. Based on ST’s old, poorly-thought-out, operational standards, each elevator was allowed to be out of service up to 5% of the time. There was no standard for having each pathway accessible much better than 95% of the time.

      SeaTac Airport Station needs an additional elevator between the east side of Pacific Highway and the eastern pedestrian bridge, or, better yet, an elevator between the west side of Pacific Highway and the pedestrian bridge. Mt. Baker needs two more elevators, one for each platform. TIBS needs another elevator between the mezzanine and the ground floor and one elevator each between the platforms and the mezzanine. Judging by the frequency of announcements, the top priority is an extra elevator at SAS.

  2. Looking good! Everything about Roosevelt Station seems to be right. There are stairs right from the beginning, so opening up the emergency exit stairs won’t ever be needed except in an emergency. And we’ll have to see how the escalators perform. Seems weird to have slow mode when they could have them stop when no one is using them, but that would be an issue if the escalator doesn’t detect someone until that person steps on it, and it moves backwards. At the very least, it looks like this time, they probably didn’t cheap out on the escalators.

    1. It’s just really unfortunate how much space on the surface Roosevelt station is taking. Adding insult to injury is how they won’t even have apartments on top of the entrances. Just two blocks of blank space.

      1. If this comment ( is correct and Roosevelt is in fact ventilating the entire Northgate Link tunnel, then it’s probably cost-prohibitive to build above the ventilation structures at Roosevelt, whereas Brooklyn station won’t have them.

        I took a quick street-view look around NYC’s Second Avenue Subway stations, and their ventilation buildings don’t have housing (or anything) above them either.

      2. Sure, Pat but the Second Ave Subway buildings are also a hell of a lot smaller. They don’t place two blocks of blank wall anywhere, they might take up a corner.

    2. Not “everything”. There are no underpasses to the east side of 12th, the west side of Roosevelt and the south side of 65th.

      No, none of those streets are like Montlake at HSS, but they will inevitably end in injuries to transfers running for a bus they see coming.

      1. Your faith in your friends at the Metro planning department is misplaced. I’m afraid when your bus fleet arrives, this station’s loop-de-loops will be quite operational.

      2. What does your reply mean? Are you seriously proposing that EVERY bus that serves Roosevelt Station will terminate there? That the 45 won’t either continue to the U District as it does now or head out 65th toward Wedgewood?

        I think you are ignoring the difficulty of a bus heading eastbound on 65th as the 45 and the 62 Rapid Ride successor would have getting to such a lay-over on 66th. They will have to turn left from 65th to 12th and then again from 12th to 66th. Buses coming from the south on 12th will have to turn left either at 65th to hot loop or 66th if they’re going to lay over.

        Similarly, any bus coming from the north on Roosevelt not continuing south will turn left at 66th and then return north on 12th. That turn to 66th from either direction is easy from a traffic standpoint; Roosevelt and 12th are each one-way so a left turn crosses no opposing traffic. But it’s going to be occurring in an environment with many pedestrians crossing parallel to the buses’ direction on the approach. So adding traffic signals at each end won’t help separate them. Buses will will regularly have to wait for a “break” in the flow of pedestrians, especially during the critical peak hours.

        This is a pretty inflexible plan, dependent on “good behavior” from pedestrians with earbuds and staring at their phones crossing 66th northbound on 12th and southbound on Roosevelt giving way to buses, and it ignores the reality that 65th will certainly have some bus crossing Roosevelt and 12th even if it’s not the 45. Passengers transferring between any such bus eastbound and the train will have to cross 65th, assuming the bus stops between Roosevelt and 12th or both 65th and Roosevelt or 12th if the stop is nearside Roosevelt or farside 12th. That nice new bus stop with the ramp pictured will of course be fine for westbound passengers.

        They should have included an underground passage under 65th for eastbound transfers. I will admit, though, that the plaza at 66th does eliminate the need for underpasses below 12th and Roosevelt, though it will probably be plagued by bus/pedestrian conflicts.

      3. I didn’t understand Brent’s comment but I think he means the buses would detour to the station entrance like the F at TIB or the 145th routes at Shoreline South. I don’t think so. SDOT just renovated that part of NE 65th Street and moved bus stops in anticipation of Link. The 45, 62, and 67’s successors will surely stick close to the current routing. There’s also Roosevelt RapidRide, which will terminate at the station somehow.

      4. If Roosevelt RR is going to terminate there, and not just hot loop, it will be on 66th. That street is the only practical layover for many blocks in any direction.

      5. I only meant the route would end there. I don’t know how it would turn around or where it would layover.

      6. The most likely layover is at Green Lake Park and Ride.

        I didn’t understand Brent’s statement either. Maybe it was a cynical criticism of Metro for rerouting a bus so that riders don’t have to walk a few feet (as in “Don’t worry, Metro will have the bus loop around and drop people off at the right side of the street before continuing.”). Obviously that won’t happen.

        Tom has a point, people will have to cross the street to get to the station. They just had bad luck with the street layout. It is too bad Roosevelt is southbound and 12th northbound. If it was reverse, then you wouldn’t have anyone crossing there. Oh well. They could have spent a bunch of money on pedestrian tunnels, but ST seems to overspend on mezzanines, and underspend on pedestrian access. Speaking of which, I don’t understand the point of the mezzanine. There is no opportunity to turn around or transfer to another train — it isn’t needed. Anyway, the other drawback of building pedestrian tunnels is that it would tie up the streets for a while. The city probably didn’t want that.

        I’m not too worried about pedestrian accidents though. By no mean is this the safest street, but it is quite likely that the streets will be “calmed” quite a bit before it opens. Other stations have similar issues, and it hasn’t resulted in a lot of accidents.

  3. Could somebody explain why Roosevelt and Northgate are almost finished, but completion isn’t expected until fall of 2021? I’m not being contentious; I’d just like someone with some expertise about the project’s construction to explain it. Is it because the U-District station isn’t as far along as those further north?

    1. Once the stations are complete, the systems contractor moves in and begins work on traction power, signal, etc. most of the infrastructure needs to be complete before this work can start.

      1. As someone who lives near Roosevelt, I am kind of happy the station is staking out some ground territory. This will be space that 1) can’t block the view from the high school as easily 2) will not be overrun by people trying to do non-transit things and 3) will be space that might even eventually have bike parking available. And don’t worry, the half mile wall of apartments going in on 65th will more than make up for any lost density from not building right on top of the station.

      2. Phyzzi: the housing potential is lost. You can upzone a block away to replace the theoretical housing above the station, but you could upzone the other block and build the station housing and then even more people could live in Roosevelt and use the station. There are only a few places where you can live within walking distance of a Link station, and two or three times more people who would like to live near a station with business destinations right there. Everyone should have that opportunity, and before WWII they did. The region should have located all stations in such neighborhoods or built such neighborhoods around them, but absent that, we need to maximize the amount of housing around stations like Roosevelt. Vancouver, I’ll remind you, has highrise clusters around mid-neighborhood and suburban stations. We could have had that in Roosevelt too. And forget the view from the high school. People need a place to live and walkable neighborhoods more than they need a view from the high school. We could provide those amenities elsewhere and not in Roosevelt, but we’re only doing a little bit at non-Roosevelt stations so Roosevelt needs to take up some of the burden.

      3. “The region should have located all stations in such neighborhoods”

        Except maybe one or two P&R stations like South Bellevue. If we must have P&Rs, they should be outside city centers, and if that means adding a station for them, so be it.

    2. No OCS, no signalling system, no SCADA, and rumor has it that the tracks themselves do not meet standards. (Should be within 1/16″, but is in some places up to ½” out of spec.)

      1. rumor has it…

        You must not know much about rail construction. There is always a track gauge applied before final tightening of the anchors. And this is panel track we’re talking about. The ties come with holes cast in them from the precision molds.

    3. A year is needed for systems testing. There are so many systems — evacuation and fire safety, signal and train control, tracks and caternaries, escalators and elevators, tunnel and station electrical, plumbing and lots more.

      Recently, BART to East San Jose has been delayed many months because of problems found in systems testing. It’s a big thing capable of taking lots of time!

      1. This bit has honestly always confused me. When we build a new road, how long after construction has finished do we need to spend testing the pavement surface and the stoplights and the crosswalk throughput and the signage and the utilities running underneath? It sure doesn’t seem like anything approaching a year.

        What makes a train track so different? How many times do you have to run a train down a set of tracks before we are convinced it has been built properly? What’s stopping that number from being small enough that it can be accomplished in a matter of weeks rather than months?

        Also do they need volunteers to practice evacuating a full train in or between these new stations? If so, where do I sign up?

      2. They’re testing the trains too. If your car loses a wheel that’s your problem. If a train loses a wheel and harms passengers that’s ST’s problem, and the government also says ST must do the testing. Your car is also self-powered, not part of the road project, while trains are powered by the newly-installed catenary.

      3. If I recall, previous Link extensions had several months of test trains running back and forth, but it wasn’t anywhere near a full year. Did the rules get more stringent in the past 5 years, or am I remembering wrong?

      4. If your car loses a wheel that’s your problem. If a train loses a wheel and harms passengers that’s ST’s problem…

        If my brand-new car loses a wheel that’s most definitely the manufacturer’s problem too. Major safety-related defects in a particular model would make the national news, and the car companies have every financial incentive to make sure that never happens. Is that not the case with trains?

        Imagine if the standard, government-mandated requirement for a new car purchaser was to hire a professional to drive the thing around on a closed course for the better part of a year because you can’t really be sure about its build quality until you do.

        Also while I’m sure there are a lot of different things that need to be tested, can’t most of them be tested concurrently? In Al’s post above he brings up escalator testing. Surely that can be done at the same time as the train testing, and by different people. Does testing the escalator take longer than testing the train? Seems unlikely, so why even bring it up if it’s not on the critical path for scheduling?

      5. “If my brand-new car loses a wheel that’s most definitely the manufacturer’s problem too. ”

        But it’s not WSDOT’s responsibility, the agency that built the road. If a train or track or signaling system breaks and injures passengers, it’s ST’s responsibility. ST may be able to recover from the manufacturers but that’s a secondary third-party issue. An agency’s responsibility to the public is more critical than a manufacturer’s responsibility to its customers, and that’s why agencies have additional regulations. WSDOT is responsible for the smoothness of the road and the functioning of the traffic lights; it’s not responsible for the cars. ST is responsible for the trains because it’s the one moving them and hosting passengers on them.

        I don’t know what the threshold is for requiring a whole year. One elevated station is a few months. U-Link was probably in between. Northgate Link is long enough that maybe it’s treated as a new line. Or maybe we’re just overestimating. The actual requirement may be 6-9 months and we may be rounding it off to a year.

        In any case, ST’s schedule is to start testing in 2020, and to open in September 2021, and maybe as early as April 2021. The exciting thing is September is just 1.66 years away, and April is 1.66 years. We’ve spent three years dealing with the gap between UW Station and the U-District and Stevens Way buses, and it’s not that much longer.

      6. I think the Cascades derailment in DuPont showed us what happens when you don’t do enough testing…

    4. Although initially behind because of some tunneling issues and the late inclusion of an additional staircase, U District is trending with the other stations to be substantially complete by end of 1st quarter next year and in the meantime, as noted below, systems contractors take over, following which testing for 6 months begins (Federal requirement for new segments). Overall project is still running about 5 months ahead, based on their “float days” figure so unless something dramatic occurs during testing, it is conceivable ST could announce a late April/early May 2021 opening sometime next year at the conclusion of testing. ST is also taking delivery of the new additional rail vehicles to support Northgate (and East/Lynnwood) LINK extension(s) so having those vehicles in hand and tested is also necessary for opening and that process is also ahead of schedule. Metro/ST/CT also need some planning room for schedule adjustments of the bus routes and budget adjustments accordingly, particularly if an early opening is planned.

      1. It could be more advantageous to do an early “soft opening” and change bus routes later in September 2021:

        – Capacity for more riders diverted from buses can be better estimated.

        – Public eagerness for connecting to Link will be stronger, so cutting routes to Downtown will be less controversial.

        – Adjustments to station circulation like drop-off zones, bus movements and stops, and neighborhood parking restrictions and pricing will be less controversial.

        Being reactive rather than proactive is a PR risk, but given That North Seattle has many people who fight change, it may be preferable in this situation.

  4. They have to install track and systems. Plus a full year of testing before it opens. Why a full year? Who knows?

      1. I’m not sure where people are getting a year of testing from. That’s not what Sound Transit’s Link Progress Report seems to say.

        If I understand correctly, testing of indivuaal components and systems has been ongoing for a few year years. The current schedule has construction finishing up on the power/communication/signaling systems on January 7, 2021. with various testing work continuing into March 2021.

        Pre-revenue service is scheduled to start on January 7, 2021 and run through April 25, 2021. There’s also 5 months of float time in case there are issues during construction or testing. Assuming that the testing turns out OK, the time from that construction finishes until the time Northgate Link opens is scheduled to as little as 3.5 months

    1. I would love to see it open quicker (I commute from the north!), but OTOH It does seem like a year is an appropriate “breaking in” period for something that tens of thousands of people will ride daily for decades to come, especially considering that much of it is custom built for our transit system.

    2. Let’s not forget that this is going to mostly operate in a subway. A derailment in a subway with passengers can be pretty awful.

    3. Let’s not forget that the U-Link escalators had a year warranty – but it began a year before it opened for riders because of the testing period. When riders started using it, it was already beyond warranty. I think ST learned to not do that again.

  5. Thanks for all of the great pics! It’s really appreciated.

    Frankly I’m underwhelmed by the level of finish work at the Roosevelt Station given its $160+M price tag. I guess we really need to start cherishing the finish work at the original Metro bus tunnel stations completed in the late 1980s (despite their more recent vertical conveyance issues which is more of a “state of good repair” lack of funding priority type of matter).

    Btw, how long has the area around the Brooklyn U-District Station entrance been closed off? It seems like it’s been a very, very long time.

    1. I am convinced that as time goes on, the Westlake and Pioneer Square Station architecture will be regarded as exemplary and timeless among subway stations worldwide. The new stations not so much.

      Seattle architecture in general parallels current trends of 95% plain buildings (“sleek” = plain) with a handful of buildings with post-modern curves that are mostly facades.

      We are in an uninspired era.

    2. Who needs finish work when you can save a theater that ruins the bus transfer and the signage from a record store that isn’t actually saved?

      1. Exactly this. It’s much more palatable to neighborhoods and trolls to spend nothing on finishing work and just repurpose what was previously on the lot. And so for 100 years we get to look at concrete and context-free neon art.

      2. In hindsight, there should have been an entrance at the UW Tower plaza (or even across the street at the KeyBank), but the north entrance isn’t terribly far from 45th Street.

      3. My “in hindsight” would have been for a 520/Montlake Station, a UW station a tad just north of the arena (not as deep as the current UW Station) and a 90-degree turn to the U-District Station running east-west under 45th or 44th (entrances near 15th and 12th) before turning north-south again.

    3. I wonder how much of that price was to acquire the land? The cost of land and the rents extracted from it are at the root of so many of issues Seattle faces — the rising cost of living is driven by rising rents and any renter can tell you that you don’t suddenly get more of anything when your rent goes up. You just have to do with less of something else. In the case of these stations or the cookie-cutter apartments we see going up, what we get less of is design and finish work. The developer has so much $ sunk into acquiring the land, there’s not as much to spend the building. Seattle was a very reasonably-priced place to live in the 80s and 90s when a lot of the public infrastructure mentioned was built. Bring in 200,000 people who can afford to pay a lot and watch your land prices spike.

      1. Nobody “b[rought] in 200,000” people. They came of their own volition because there were jobs, the winter climate has moderated, summer is now wonderful, and Puget Sound is still there. There was no “conspiracy” to destroy your Seattle-in-Amber.

      2. The rising cost of land is the biggest factor in housing projects and transit projects in general, but I don’t know how exactly that applies to Roosevelt Station. ST2 was passed in 2008, and in 2012 prices started rising faster than the every had before, surpassing even the previous peak in the 2000s. That threw ST’s budgets off-balance, and has most affected Lynnwood Link. I haven’t heard how much it specifically affected Roosevelt. Underground stations require only a couple blocks around it, while elevated/surface stations and track may require more land if it’s not already publicly owned. Land use may also make a difference. Large urban centers may cost more than small urban villages because of the potential profit a highrise building or mall can generate.

      1. I love how none of the drop in sessions are accessible to anyone working 9-5:30 downtown (so most of the people interested in the project)

      2. FWIW, the last pedestrian bridge at SeaTac Airport Station was installed mere weeks before the station opened. Of course, it was much shorter than the one that will be needed to transit I-5.

    1. Didn’t ST2 contribute some money to it? Or at least the station area budget? I think ST preferred to contribute a fixed amount as matching funds rather than being responsible for the entire cost. It also had to fit into the total of what North King wanted relative to the chosen tax rate.

  6. Think of how much more we could have built in a shorter amount of time if these stations weren’t so absurdly huge and over built. Two mezzanines? Are we insane?

    1. Perhaps the longer the escalators get, the more often an escalator will fail? And the longer the station box needs to be?

      I hope they allow the floor space to be used well.

    2. FWIW, after the Northgate Link stations open, I don’t think ST has added a mezzanine to another station slated to open before 2030. That’s 19 more new stations without mezzanines by my count.

    1. It does. has some recent articles on US rail construction costs vs other countries. Even Canada is lower. But it’s a different culture, different regulations, different unions and contractors, and a willingness to adequately prioritize and fund transit right so it can be built the most efficient and cost-effective way.

      1. While Canada is better, they still have their fair share of issues regarding building or funding. If you want a good example of how bad transit planning in Canada can be. The decades long quibble over building TTCs “Downtown relief subway” (subway intended to make line 1 more manageable during rush among other reasons) in its many many iterations that have been put forth over the decades. The debate over the TTC relief line makes the Seattle process look quaint in comparison.

Comments are closed.