
Update: Normal service on the 1 Line and 2 Line resumed shortly after 2pm on Sunday.
Late last night, an empty Link train derailed while turning around using the pocket track at International District/Chinatown station. The train was on its way to Stadium station to help ease the crowds of Ed Sheeran fans waiting for a train home. It appears the train derailed at the switch and may have damaged the overhead wire. As of this morning, the train has been removed from the station and crews are inspecting the tracks and overhead wires. Sound Transit has not responded to our request for comment.
Shuttle buses have been deployed to cover 1 Line and 2 Line service downtown. Trains are still running between Capitol Hill – Lynnwood, SODO – Federal Way, and Judkins Park – Downtown Redmond. Expect delays across the whole system today and use regular bus service whenever possible. Passengers can find alternative bus routes on Sound Transit’s website.

This is a developing story. The article will be updated with more details when we know more.

I hope the operator is placed on leave pending an investigation. If operator is at fault, they should not be allowed to return to work until remedial training is completed
I think honestly the way you wrote your statement it seems that you are rushing to judgement on this.
Sound transit and king county metro are undoubtedly taking steps to ensure this never happens again.
With incompetent transit agencies in WA, you never know what is going to happen. Their steps are rendered useless because they won’t address the root problem and continue to build outdated systems along with random virtue signaling operations that have no bearing on the quality of the transit itself.
I got stuck in the door 3 weeks ago and passenger had to free me. If passengers are on the bus and the train slams on their brakes. There would be 100 people injured.
It would be interesting to know the concensus on this idea –
Separate Operations from the construction side
Operations needs to be its own entity and not affected by the whims of any political decisions the board might make.
Sound Transit is a mature enough agency so that should be the top priority now.
So… this separate operations team wouldn’t have a board of public or elected officials? Or you’d expect this operations agency to have a different board of public officials, which would somehow not exhibit “whims of any political decisions”?
What principles would guide this supposedly apolitical operations team, and who would have the power to change those principles?
“It would be interesting to know the concensus on this idea –
Separate Operations from the construction side
Operations needs to be its own entity and not affected by the whims of any political decisions the board might make.”
When ST started out, it was contracted bus routes on public roads and train routes on private railroad tracks. More than 90% of the riders were served this way until 2009. It worked ok.
Now it has a light rail system with exclusive right of way with tracks and stations with vital maintenance needs are carrying over 80% of daily riders.
It’s way past time for ST to mature into running their own operations with their own staff. Contracting almost everything including operations is just bad public accountability at this point. And senior management should more know how to actually run a light rail system through disruptions rather than know how to contract out.
Operations doesn’t need its own board.
Sound Transit staff needs to have more control over the day-to-day operations without having those decisions micromanaged by elected officials (or be overly concerned with the ramifications of that when making them).
Effectively, make the Sound Transit board the “Expansion Committee”.
The operations entity can be given the set budget and expectations, and it would be up to them to decide how to get there (efficient and effective operations)
Having politics intrude on those operational decisions has very negative consequences.
Jim, can you provide an example of a situation which you feel the Sound Transit Board “micromanaged” operations in a way that would be prevented in this setup?
No, I can’t.
There is no specific dramatic example as far as Link goes (although our own general discussion of whether DSTT2 would have affected the overall effects of this service disruption are a non-sequitur in my opinion)
I do have opinions about how “management by committee” in general has caused problems, but I’m not going to set fire to that pile.
That’s what made Joni Earl so effective, she kept the board out of the staff’s hair, and gave them the room they needed to solve problems. What I loved and respected about her style was how she gave credit to the staff for their efforts.
My employment history has presented numerous examples, especially in IT, where upper management involvement has succeeded in only hamstringing the project.
My OPINION is – give the operations people the freedom and authority to make the proper decisions to achieve the results, without having to worry about the board.
The “culture” needs to grow up.
I appreciate the honest answer. Readers should understand that Sound Transit’s operations are already minimally managed by the Board. Recently, and with good reason, the Board has focused almost exclusively on expansion and associated long-range budget issues. They rarely provide substantive input into operations, and I think that’s largely because the only Board member who seems to even remotely understand rider concerns is Claudia Balducci. Which is why she’s on the Rider Experience and Operations Committee.
The only issue the Board might mettle in now is in deciding how to reduce service to improve reliability. The next big issue operations issue will be how to implement fare gates. But the Board doesn’t bother with staffing or scheduling. The actual nitty-gritty of daily operations is beneath them, which is probably why we’ve seen no senior management engagement in Sunday’s derailment.
The are numerous reasons and experiences that contribute to my point of view on this.
The most glaring issue (concerning ST) of what went wrong culture-wise was how Peter Rogoff threw staff “under the bus” after the Apple Cup incident.
First major Link incident, and his response wasn’t to stand behind decisions made in the chaos of the moment, but to apologize. Operations was doing the best they could under extreme (but not deadly by fault of staff’s actions) circumstances.
Maybe the best outcome was a greater presence of security now to keep ‘the hordes’ contained.
This culture focus for me was pretty much formed from my retail experience, and how important it is to have an empowered Frontline.
And here’s the rub –
Building Good Tribal Knowledge is going to take some time, and it only comes from experience handling situations like these.
The backing of ‘on the ground’ decisions by management is what will get them there.
(Hence, my ‘don’t micromanage’ viewpoint)
“Building Good Tribal Knowledge is going to take some time, and it only comes from experience handling situations like these.”
I’m sorry. This is not a good approach. Learning solutions from scratch while moving 160K people a day is crazy-making.
ST should have been hiring experienced people who have worked in other high demand systems with multiple lines all along. They should not hire people above the middle levels that do not have years of light or heavy rail operations experience through disruptions and crises . It’s been like hiring a pro football coach who is hired because of their TV camera experience and their ability to be controlled by elected officials.
You’re correct, but you reference an issue stemming from poor senior staff management, not the Board. Yes, the CEO was selected by the Board, but Operations are now managed by Deputy CEO with the title of Chief Transit Operations Manager (https://www.soundtransit.org/get-to-know-us/executive-leadership).
ST has already adopted your solution, which was recommended by the Technical Advisory Group following the 2021 Realignment. The problem is leadership culture at large firms (public or private) is very difficult to change unless you remove every experienced employee who was trained under that culture.
You’re misunderstanding my point, Al. I’m referring to those on the front lines, of which there is no limitless pool of light rail operators.
Railroad operations and train movements have myriad rules governing how and when that movement occurs.
There have been times where it was just the experienced eye of the engineer that avoided a situation.
If the investigation doesn’t find mechanical issues, then the operators (while obviously not trying to channel Casey Jones) following all the rules, can be the last line in averting a bad situation. That’s what experience brings – that feeling of “Something doesn’t look right‽” or “It’s the last run of the night, so I’ll take the extra time and be extra cautious!”
That’s where the Tribal Knowledge comes in.
Interviewee material!
People could have been killed.
“People could have been killed.”
Seriously?
Not at the speeds at which an LR train enters a single-ended pocket track, which is to say, “about 5 mph”.
Seattle Times article by Mike Lindblom.
Note that Link may still be single-tracked north of SODO, so frequency may be every 15-20ish minutes.
“The incident illustrates how Sound Transit would benefit from adding a downtown crossover track — as recommended by outside experts, either there or at Pioneer Square Station. A crossover area would have allowed the downtown stations to be open Sunday morning.” — Times article. “There” seems to refer to CID, but the suggestions I’ve heard are at Symphony or Pioneer Square.
There is a turnback facility at Stadium. Why aren’t the 1 Line trains running to Stadium?
Maybe the issue is where they can cut power. Isn’t there also a crossover just east of Westlake? That could have allowed the northern segment to operate to Westlake. But maybe the Westlake to Stadium all lose power when they want to disconnect ID.
CID is possible simply by adding a “wye” to the north end of the pocket and a matching turnout to each of the main tracks.
I would put an “active frog” in each of the main tracks to minimize the possibility of derailment of in-service trains “running through” the turnouts.
The technical term for that is “swing nose frog” btw. Translink uses them and it allows for faster turnout ops.
Thanks. They’re SOP on any trackage faster than 110 mph, way faster than Link, but I think that they’re a wise idea when a turnout is rarely opened on any high-frequency passenger line.
When the 99+% of through moves occur, there’s no frog at all; the flange experiences smooth, uninterrupted rail contact identical to an ordinary stretch of rail.
It’s hard to visualize how the train could end up at the northbound side of the station if it was backing into the southbound platform, unless it was going way too fast. Even if the switch were aligned for the pocket track when the first bogie went over the switch, and the switch was then thrown for the platform track.
Figure the train was at least 3 feet off the track by the time the operator realized there was a problem, then another 6 feet or so of coasting after the brakes were applied.
The front truck sits behind the cab, so it will feel like it’s derailed any time the car goes into a sharp curve. So, the response time will be slower than you might think it should be.
It’s not an automobile, where the steering wheels are in front and there’s an immediate visual indication of what direction they are going.
I’m assuming that braking on Link is primarily done through regen back into the line. With the cab wheels off the rails and maybe contact lost with the overhead how is braking affected? I’d also assume that the trains have an air emergency brake such that when the air pressure is released mechanical brakes clamp on hard (after the operator pushes the “Panic Button”.
Is there no guard rail in the switch areas that can keep a train from wandering so far; maybe not possible or extremely difficult to do in a switch.
I’ve been told there is automatic speed limiting software on Link so that if an operator was going faster than protocol it would have automatically applied the brakes?
There’s no air brakes, but the brakes do go on when the power gets shut off.
It’s far more dangerous for passengers for a train to hit a solid object than to gradually come to a stop.
So power must be required to release the mechanical brakes? I’m not clear on whether this train actually lost power or not. The wheels of the lead car look to have “derailed” such that there would be no return path to ground. But the other car(s) were most likely still in at least partial contact with the rails. Do the light rail vehicles “share” power or does each one rely on getting power independently?
I still don’t understand what the derailment was. I imagined a train completely off the tracks and maybe lying sideways on the platform or ground. But the photo shows it straddling two tracks as if turning onto the left one. There is a turn track around there somewhere. How close is it to the turn track, and is it going in a direction compatible with it? Was the train trying to turn? How did it get off the rails?
A derail is anytime a train’s wheels leave the rails. The train doesn’t have to be completely off the tracks or turned over.
Trains aren’t like cars and can’t turn on their own (usually). Their wheels are guided by the rails on the track and switches are used direct trains onto the correct track when tracks intersect.
The track arrangement at Chinatown-International District Station is rendered fairly well on OpenStreetMap, here: https://www.openstreetmap.org/#map=20/47.5976502/-122.3281570&layers=T
Looking at the photo at the top of the article: the view is looking south from the central mezzanine of ID/C Station, so the northbound track and platform are on the left, the pocket track is in the center, and the southbound track is on the right. If you look closely, you can see the pocket track splits into a northbound and southbound track right where the train is derailed.
It appears the train was on a southbound track moving northward and was switched toward the pocket track. When the train entered the switch at the south end of the pocket track, the first set of wheels (located under the raised section of passenger seats behind the operator cabin) were somehow forced up out of the track slot and onto the surrounding concrete surface. The train continued for several feet toward the northbound platform, with the first set of wheels derailed and the remainder of the train apparently still on-rail.
Thanks for providing the Open Streets map. I’ve never seen that resource before.
Between Weller St. and the CID platforms I see 4 switches or crossovers that a train moving northbound on the southbound track would need to navigate through. That’s plenty of opportunities for something to be misaligned.
How would a train successfully make that move? Would it first head northbound into the pocket track, then reverse direction onto the southbound track and then reverse direction again onto the southbound track? That’s 3 movements and lots of switches.
Minor correction: with some digging, I was able to find this plan showing the actual design for the pocket track: https://storage.ghost.io/c/19/31/1931222a-d1c5-42ac-8932-2ee3ac1f8927/content/images/old-theurbanist-org/wp-content/uploads/2019/12/screenshot-2019-12-31-at-07-50-59.png
Anyways, I assume the pocket track is usually only ever entered from the northbound track and exited onto the southbound track. Conversely, I assume entering from the southbound track (requiring northward movement against the track direction) is highly unusual. Like most highly unusual events, this derailment appears to have resulted from a combination of multiple unusual conditions. Unfortunate!
There’s a double switch just south of SB 1093+07.76 that leads to the turnback track. From the photos, it looks like one of those switches is the point of derailment. Is it possible that the train unexpectedly entered the turnback track because of a misaligned switch and then derailed at the 2nd switch which might have not been aligned for a northbound train running on the southbound track?
Based on Sam’s comment below (indicating a northbound train being asked to switch to southbound operations using the pocket track), I think this is exactly what happened.
That’s the thing about using the term “derailed”, which covers anything from dropping 1 wheel off the rail (a switching error, such as this), all the way up to the Dupont tragedy.
I thought it was interesting that the King5 article about this had ‘derailment’ in quotes.
Thankfully sound transit sent the single tunnel idea out the airlock. It is already a mess when a simple wire get damaged and now we’re escalating to derailment.
Once again no notification at the station (Northgate) as to what alternatives are to points east and south – just a “trains are running every 20 minutes” and the fact that the next train is to Cap Hill (now, that I’m on it, they’re sending it to Westlake). This agency needs gutting and rebuilding.
“this agency needs gutting and rebuilding”
It is okay to be annoyed and frustrated. But blowing it all up without a plan isn’t the answer and won’t solve these issues
lol. There is a laundry list of things that are best practice throughout the world that this agency can’t manage. I’ve used public transit in 60+ countries and in many ways this one falls far, far short of places with far fewer resources. Mike’s list below is part of it. Take political posturing out of every decision, reorganize the agency so that it is run by transit experts with minimal input from politicians that of necessity spend their time on many other things and therefore have little interest or knowledge in anything beyond the minimum required to take a vote (there are a handful of exceptions, but on the whole it’s a bloated board with parochial interests, often making technical decisions).
I’ve been caught up in several of these “malfunctions” and the minimum requirements of communicating with the public are not met. How do you get from the north end to the east? No idea, there’s a 1 Line bus bridge and a 2 Line bus bridge, but nothing about that on the signs, nor where you would have to transfer (thankfully our driver described it, but that’s not helpful on the platform trying to figure it out). I waited at Westlake specifically to help people going to the airport as the trains had started running through again, but no notice on the station signs, no info on how often they were running – the first through train literally still said “last stop” at Westlake despite it continuing on to Federal Way. This lack of information happened during the last major incident as well, as I found out trying to get home from Bellevue. No way to run a railroad. as they say. Backup plans should be seamless and well described to everyone from a frequent user to a visitor. ST falls woefully short.
Not sure “gutted and rebuilt” is the right approach, but agreed that the communication to riders when stuff like this happens is beyond unacceptable. We need to have a little patience, as it probably takes a little time for them to know what happened and how to deal with it, but in my experience, most of the time by the time you get the information it is already obsolete, if it was ever actually comprehendible or correct in the first place.
Horribly, Sound Transit has decided to dig a new tunnel in a way where it can’t provide any help when one tunnel gets damaged. If we had the second tunnel open, the Lynnwood and Seatac lines would still have to turn back in exactly the way we’re doing it now.
But you would be able to transfer at Westlake (where trains have been getting sent) and SODO to bypass the issue of trains. Contrast that to now where you need a shuttle bus for all the people on trains from Judkins and SODO. In a two tunnel scenario you only need to get people from IDS2 or PSS2 or whatever option gets chosen (remember, these aren’t final so anything could still be the final design) to JPS.
Of course there are scenarios where two tunnels would help. But the added redundancy is relatively minor. The second tunnel will not only be very close to the existing tunnel but it will be very close to the spine that runs through downtown. Imagine trying to get from Northgate to CID and the trains are turning back at Westlake. No big deal. Just take the bus the last mile. Not a special shuttle, but one of the many buses that run north-south through downtown every few seconds.
In contrast, imagine they are turning the trains back at Capitol Hill. Now you have to get from the UW to downtown. How exactly do you do that? Take a shuttle I guess. What if it is rush hour? It sounds like a complete mess, even if you are just going to Westlake. The point being that added redundancy is great. But if you were really focused on that, you would add a second line from the U-District to Westlake, where Link is doing is heaviest lifting.
Then there is the fact you have two tunnels very close to each other. It is bizarre, really. Very few other cities do this — if any. They would have sent the new tunnel to First Hill. It isn’t clear it would have cost much more and yet it would completely change the nature of the second tunnel. You would have added redundancy but more importantly you would have a connection worth transferring to (from either direction).
But then there is the idea of a second tunnel in the first place. Do we really need a second tunnel from Westlake to CID given the spine and the existing tunnel? Not really. The main reason they want the second tunnel is to serve West Seattle. For no good reason ST decides to focus on building West Seattle Link. Now, of course, the agency defends this decision even when Ballard Link will be built decades from now (if ever). Now the reasoning is simply bureaucratic laziness more than anything. The board doesn’t want to do the extra paperwork involved with changing the order of things. (And some people wonder why America is the absolute worst when it comes to building transit.)
There is an easy way out. Without the train to West Seattle you don’t need a second tunnel. The Ballard line can just end in Westlake or have only one station in First Hill. Then eventually it could go to West Seattle (although it would probably make sense to go somewhere else given the terrible cost-benefit of running trains to West Seattle). But again, that is what a sensible, transit-focused transit agency would do. Sound Transit is not that agency.
Oh, and then of course there is the money which could go into maintenance which in turn would avoid many of these outages in the first place. We are spending billions on expansion while the maintenance budget is stretched so thin that they are calling for cutbacks in service. No one dare consider hiring more workers as that would push out the various projects even farther.
One blunt truth is that it would be faster to ride a bus from Westlake to IDS than to use the other rail transit tunnel. There’s a shorter walking distance and fewer vertical devices to use.
And if the northside train can’t reach as far south as Westlake it still would require a shuttle bus.
And Judkins Park as a 2 Line stop is not anywhere close to the planned second tunnel.
A better redundancy strategy would have been to connect Ballard Link to a Metro 8 subway that ends at Mt Baker Station (transfers maybe at Capitol Hill). But only if the vertical distances aren’t punitive. That was never considered.
DSTT2 to me looks more like a real estate game and not a redundancy game or even a disruption solution game. If it was a disruption solution game, ST would be exploring building the missing crossovers under Third Avenue.
Westlake 2 is about 12 floors and 9 escalators below the current Westlake.
On July 20, I was on one of the trains that was delayed due to the mess near U District. I went to the surface at Westlake and got a bus back to King Street Station. The process of exiting the current Westlake station took me almost as long as the surface bus ride. The deep Westlake 2 will be much worse in terms of getting to the surface.
But trains weren’t going to Westlake here! They were getting turned back at Capitol Hill. The second tunnel would be useless.
The second tunnel would be useless.
In this particular instance that is true. It is also true in the vast majority of outages. But in certain cases it would be useful to have the second tunnel. My point is that those instances are not only rare but there is an excellent alternative (surface buses). Al is correct. If you are trying to add redundancy than a Metro 8 Subway is a lot more useful.
The same thing would happen to the north and east corridors even with the second tunnel. Only the south corridor in DSTT2 would be unaffected. The only places you could go to on an intact line would be toward Tacoma or Ballard. If you’re near the future 2 or 3 Lines, say at Capitol Hill or U-District or Judkins Park station, you’d have to take a bus to a future 1 Line station in order to use the intact line.
ST could alleviate this by having a crossover between the tunnels, so that it could reroute some trains. But the current plans don’t include this. We’ve asked for it repeatedly. But that has succeeded just as well as asking for (A) open-gangway trains for ST2, (B) the Ballard-Westlake starter line alternative (“Build Ballard First”), (C) closer transfers between the tunnels, (D) a crossover track in Symphony or Pioneer Square station, (E) keeping down escalators. In other words, it has not succeeded at all.
A second tunnel does not solve the issue. Under ST3, there is no connection between the Ballard line and the DSTT; there is no connection between the North line and the DSTT2. There can still be messy situations.
The day-to-day operation and service will be much worse with the ST3 two tunnels: transfers between the east and south lines will take about 10 minutes; the DSTT2 will be deep and transfers will take more minutes; all south line riders will have to transfer between the tunnels.
Of course you’re gonna get flamed for this here because this place is full of egoists who can’t help but parrot everything they see to appear smarter than they are.
Build DSTT2.
I guess since DSTT2 removes the stadium station from Line 1, so that many of these passengers at this event would be stuck with surface buses, or take Line 3 north to Westlake and go up 9 flights of escalators, DSTT2 does solve the problem of ST wanting to provide extra service to the stadium station. They won’t be able to move trains from Line 2 to Line 1 in this manner after that.
Please explain how DSTT2, as planned, would have helped here? I’m not seeing it.
lol
Of course you’re gonna get flamed for this here because this place is full of egoists who can’t help but parrot everything they see to appear smarter than they are.
So you are argument is basically “I am right, all of you are wrong, nyah, nyah, nyah”. Wow, what a solid argument. You must really impress your Kindergarten colleagues.
“Chicken Nuggets” appears to think he’s named for “food” rather than its results.
The track arrangement in SODO as planned will not allow for a seamless track switch. That’s because ST has every track running in the opposite direction next to it. Other rail systems in the US and worldwide with four parallel tracks generally put tracks running in the same direction next to each other.
So switching to another track headed the same way will require pauses in service on three tracks. That will be disruptive in many ways and will limit how many trains can actually do this.
And it should be pointed out that if ST had built a full wye where 1 and 2 Lines merge, the incident would not have happened.
Driver on this SB 1 Line train entering Westlake (run 023) has been great – just announced that the train is continuing to Fed Way (but didn’t change the head sign from “last stop” so everyone is confused. Station signs also say “Station closed” still, so…. Cluster.
I was at the Mariners game today. I took the bus there. When the game was over, I checked the Sound Transit web site, and it still said at that Link was still out of service downtown. But the Link station was on the way to the bus stop, and it was obviously back up and running, so I took Link home instead.
This afternoon at 1:30pm both the shuttles were running and Westlake station was open. The alert had said some trains may turn back at Judkins Park or SODO, so perhaps the shuttles were to mitigate the lower frequency, or ST keeps running them for a buffer time after stations open.
Anyone know why they can’t operate the 1 line to Lynnwood-Westlake using the crossover tracks east of the station? Seems like that would be much more efficient. Capitol Hill was crazy earlier today.
For some reason ever since they built the Cap Hill/UW tunnel ST has been absolutely allergic to using the Westlake crossover tracks. Not entirely clear why.
Is the crossover a full scissors? With five minute headways they are essentially mandatory.
1 and 2 Lines have now resumed regular service, per a notification email from ST at 2;13 pm
Unlike lots on here i will not claim yo have answer butter currentincident situation and response structure is definitely 7nacceptable for the money being spent. I don’t see any credible argument that dstt2 would have helped this so kill it voila, money for ballard!kill dstt2 before it drahs the whole system down.
Ross Bleakney is absolutely correct. A 2nd tunnel paralleling the first is unnecessary. Ballard Link should be built before West Seattle due to its greater benefits to the Region. The ST board needs to stop saying the ridiculous notion that West Seattle is “shovel ready.” Repeatedly saying it ain’t gonna make it so.
I checked the online Link scanner to see when the derailment happened, and the first call came in, seemingly from a station security worker, at 11:57:57. The Link Operator called in about a minute later. He kept it short. “Train 2 Operator, uh, I have a derail.” Poor guy was asked minutes before if he was ok doing a 2 Line trip. Before the derailment, the Link Control Center explained everything he needed to do to get the northbound train to the southbound IDS platform. After the derailment, obviously, it was chaos, but everyone was professional and knew just what to do.
Interesting. So maybe the failure resulted from the switch on the southbound mainline being in an incorrect position, so the train went back toward the pocket track unintentially?
The southbound main is connected to the pocket by a trailing point turnout. That is, a train entering the turnout from either the southbound platform or the pocket passes through the frog first and runs through the points “outbound”. That can “break the switch machine” [tear the points from the crossbars that move them] but often simply ends up bending the points enough to act like a “spring switch”.
That would not end up with a weird derailment like that pictured.
To be completely off the rails at low speed a train usually has to run over a derailer or some other obstruction that raises the flange above the rail.
Or, a switch can be thrown under a passing car which then has one truck on one track and another on a different, diverging track. Eventually the two tracks will be far enough apart to pull one end orcthe other off the rails.
That can only happen when a train enters a turnout (switch) from the FACING-point side, and to do that a train would have to approach the station running against traffic on the southbound main. If the point was to get to the southbound platform, running through the turnout in normal position would be the right thing.
That’s what Sam reported was happening. The train was running out-of-direction from somewhere south of Stadium northward. So for some reason the now facing-point turnout linking the southbound main and the pocket was not in the “through” position for the length of the trains. Truly, nothing needed to happen at CID except the operator would have to reverse direction which means changing cabs. The turnout just to the south of the platform should have been inert throughout the movement.
Yet it would appear that it was reversed while the train was passing over it.
OR some large obstruction fell across one or both the rails.
Cause for concern.
Apologies. It should say “If a train runs through such a turnout that is set to the other pathway from the frog side, it cat can “break the switch machine” [tear the points from the crossbars that move them] but often simply ends up bending the points enough to act like a “spring switch”. I was unclear.
Of course, if it just bends the connecting hardware, then anything is possible.
That’s a good theory. As far as the radio calls, I didn’t listen to all the calls, but of the one’s I heard, this is what preceded the derailment.
They asked the northbound 1 Line Operator at Columbia City if he was still willing to do a concert trip. He said yes. They said they were going to have him reverse run into the southbound platform. He called in again at Mount Baker saying he was out of service. He called in again at SODO. They said he was to diverge left and go north in the southbound track, presumably at the crossover north of Holgate. I think the last call before the derailment was when he was approaching Royal Brougham, and they were telling him how to get the barrier to go down, and saying something about in the tunnel, “to keep the directionality, you will have to bypass.” A few minutes later it derailed.
So it does seem like he was on the southbound track before entering the tunnel. But I’m not sure from that point if he, or the control center, had to do anything further for the train to approach the southbound platform. And after entering the tunnel, and was on the southbound track, did the train somehow get switched over to the pocket track, and it derailed changing tracks, or in the pocket track curve? I don’t see how it could derail from the southbound track.
This just seems like a systemic failure. WAY to much manual input that has to be correct and they were trying to walk someone through it blind. This isn’t a moon shot that should require Neil Armstrong at the controls. This really is a strong case in point for extensive automation. Unfortunately, the agency in charge can’t even run what they have. It also speaks to why ST won’t do the obvious when there is a special demand or a system fault; they can’t.
So, their plan was to change direction on the southbound CID platform?
That means the whole pocket track arrangement thing at CID is unnecessary for its supposed required purpose.
Isn’t the pocket track to store an extra train that has to wait a few minutes before it starts its run?
No, Glenn. The pocket is used to take trains out only service from the 2 Line and send them to Forest Street.
At least, that’s what they said when they built it.
I guess they could something similar by using the pocket just west of Judkins Park to get the train onto the southbound track at CID. But both movements against traffic are pretty intrusive during normal service periods. They block the “outbound” tracks south and east of CID while the out-of-direction train approaches and reverses at the platform. The pocket allows a train making the reversal to slot in between normal northbound in-service trains on the approach, slip into the pocket, complete the reversal walk, and lay for a gap in the southbounds.
Glenn, the train had already changed tracks about 3/4 a mile south of there, so it was already traveling north in the southbound tracks before it reached the tunnel. That part I’m sure of. What I’m not completely sure of, is the following. After entering the tunnel, then beginning to make the slight curve to the platform, it didn’t take the main track to the platform like it was supposed to, it instead seemed to start to follow the pocket track, then somehow derailed.
And for anyone who doesn’t know why it was going in the wrong direction to the platform. It was an out of service 1 Line train that was coming from the south that was going to do a concert crowd 2 Line trip from International District Chinatown Station to Redmond.
Looking at the track chart, the only place a northbound 1 Line train can easily switch to the southbound track is just south of Massachusetts Street (which is just south of Stadium Station). It would also be possible for a northbound train to switch to the pocket track at CID, then reverse direction onto the southbound track, then reverse direction again and enter the southbound CID southbound platform track; but that maneuver would require 3 separate train movements and realigning the switches 3 times.
Oh great, another derail. And I thought Sound Transit learned their lesson 17 years ago with that other person.
15+ years without a derailment is…bad, now?
ಠ_ಠ
15+ years of only having to run trains one way. East Link was 8 years (at least) late in opening. Now “challenged” with running two trains there’s an operational breakdown. The result will be… don’t even try to adjust to demand. Just do the only thing we sort of know and leave it at that. If something unexpected happens then too bad; the system is down for however long it take for ST to get things back to the “normal” they sort of know how to operate.
The CEO knows nothing about train operations; but a lot about pulling in political favors. The SPD Chief just got fired (with benefits) because of one incident that he had no control over. In fact, the ACAB position of the establishment probably made it inevitable. The main reason given was a failure to provide timely information to the public ST can’t even manage that for planned closures.
Can anyone fire the CEO of ST? Or is it like an appointment to the Supreme Court? Yeah, I know the board can but the current CEO owns the board so that’s not going to happen. Maybe the Governor but that’s not going to happen either.
You really just have to compare ST to peer transit projects/operations and it’s obvious what a half ass system we have for 10X the cost. And it’s now to the point that the taxes are just going to eat money with no deliverables for the foreseeable future. But I predict executive salaries will continue to grow.
“East Link was 8 years (at least) late in opening.”
6 years. The schedule was supposed to be:
2020 Northgate. (Opened 2022, 2 years late.)
2021 2 Line to Redmond Tech. (Opened 2026, 5 years late.)
2022 Lynnwood. (Opened 2024, 2 years late.)
2023 Federal Way to 272nd. (Opened 2026, 3 years late.)
Downtown Redmond and Federal Way Downtown were not in ST2 so they weren’t on the schedule. In the 2008 recession, Federal Way was truncated to Angle Lake, then re-extended to KDM. When ST3 passed on 2016, Federal Way was postponed to 2024, Star Lake was re-added, and Federal Way Downtown was added. Marymoor Village and Downtown Redmond were also added but I don’t remember their initial timeline.
The 2 Line lost 1 year because Bellevue insisted on a half dozen more south Bellevue alternatives and it and Kemper Freeman obstructed the project. The concrete workers’ strike added 9 months to Northgate and anything else under construction. The pandemic disruptions and supply-chain delays added maybe a year; we’d have to say what caused what specific delay to disentangle that further.
Northgate opened in 2021 one year late actually.
it’s not great that this happened, but I’m almost more concerned about how long it took to fix, esp since they had the overnight hours to work on it
How long did it take your railroad company to fix a similar problem?
google suggests that major Asian cities do it 2-6h, European ones in 6-15h
Seattle’s downtown tunnels are relatively close to the maintenance facility, and they were designed for buses, so getting cranes into position should be relatively easy compared to tighter century-old systems.
Since Seattle uses overhead power, I would expect less damage to the power system than in a 3rd rail system, but maybe ground-level repairs are easier to implement so it’s a wash.
They fixed it like 2 days. Does the rail go to tawilla?
From the time of the incident to the restoration of full 1 & 2 line service was around 14 hours.
Asian and European systems are a lot bigger and busier.
This, for a first time derailment on a system this size, falls within those parameters Google displays.
Recovery times improve the more “practice” the crews get on real situations.
I’d prefer they don’t practice that often.
If we want light rail to grow “bigger and busier”, we need ST to focus on service uptime, whether that’s the trains, the escalators, the elevators, the electrical system …
ST repair crews should conduct derailment drills. Do they have a derailment simulator at one the maintenance facilities to practice using the equipment? Can they use the overnight closure times to simulate the difficulties of getting crews and equipment to various points in the system?
Derailments and other disruptions will always be occurring. ST seems to have had more than usual lately — and several are because of staff not paying better attention. I do wonder if the designs and/or the installations were faulty in some of these instances. And correcting things can be expensive so where is the money coming from?
The more relevant issue to me that should be answered is how well ST reacts. They should know what to tell riders quickly, and send emergency staff to the scene to get trains back up and running ASAP. They should have pre-prepared contingency operations plans for every segment of track and every junction that may fail. They should have status electronic signs that activate when there is a disruption that tells riders what to do and how long it will last. Finally, they should have funds to fix systemic problems in the system available so that they can be fixed without having to wait for another referendum.
And what’s the timeline for how long it should take to implement a plan?
10 minutes?
20 minutes?
1 hour?
Derailments aren’t occurring at a frequency that warrants detailed immediate recovery instructions.
It can take up to an hour sometimes just to get a handle on what happened and, based on the severity, roughly how long it will take to rectify.
If you’re someone who’s waiting for a resolution/instructions, it feels like it’s taking FOREVER.
If you’re the one responsible for the resolution you’re feeling like everything is on fire, before the full picture begins to gel. That’s something only experience can help with.
One thing that’s worse than information that takes a long time to get out the public is getting information out quickly, that’s also WRONG.
If I were going to have an “emergency kit” to hand to Link passengers, it would be a quick guide of the major bus routes that are an option, which then could (hopefully) be later supplemented by bus bridges (assuming equipment and driver availability).
These situations can be very fluid in the initial stages.
I was referring to a contingency plan for any unexpected disruption — not just derailments. It would not be an emergency kit for riders, but would be a play notebook for staff to guide them on what to say and how to take action.
There are two types of emergency information: the train status and the alternate transit path.
Even OBA discloses when a vehicle is late and ST stations disclose the wait until the next train already — so the status part looks easy and quick. The only major thing actually missing is an estimate of when train service will resume. A rider can then determine when to return to the station if they do not want to wait around an hour or two.
Alternate transit like shuttle buses obviously take time to arrange. But even then, ST could have pre-prepared diagrams about where the bus will arrive. When a decision to use shuttle buses is made, the sign could let riders know when they are anticipated to arrive. It’s important because the rider may be able to use a regular bus rather than wait for a shuttle bus.
Again I ask –
What time frame do you expect things to happen?
Immediately?
Within 10 minutes?
½ hour?
An hour?
It’s fun to take the train. Mayor Wilson needs to bring business back to Seattle so they can fund the train. We gotta keep business moving.
Begone, bot