Monthly Archives: August 2011

(Not quite) open door policy

I am totally doing this to taunt you. That’s how I roll.

Last week, Dr Jeff (fellow blogger and also a super-commuter – as in, his morning commute is Gateway to Hillsboro TC on MAX, then a transfer to the 57 to Forest Grove) encountered the all-too-familiar situation where he sprinted to a train at Gateway and hit the yellow door button, only to watch the train pull out of the platform without him. He blogged in a rather entertaining fashion about what happened, including the ensuing Twitter exchanges with TriMet.

And of course, me being me, I admittedly winced at the role played by the door buttons. Actually, when I die, someone needs to make sure that goes on my tombstone: “Fought the uphill battle of explaining how the MAX door buttons work”  All joking aside though, it sounds like everything in this situation worked the way it should have, including the actions of the operator, even though that meant Dr Jeff missed his train.

Question: Why don’t trains wait?

Some of this will be specific to Dr Jeff’s situation, other parts more general because I know the issue comes up a lot. I get it that it’s extremely frustrating to be on the wrong side of closing doors. I get it that it sucks to bolt for a train only to make eye contact with the operator from a few feet away and then watch the train leave. But there are a lot of factors as to why a train most likely isn’t going to wait for a late runner, and they revolve around how a rail system works and are not about spiting the customer. I mean, it’s not like there’s a “How many late runners did I screw over today by closing the doors as they ran to my train?” club of operators who regularly meet, gloating  “Excellent…” and tenting their fingers as members tell each other stories of stranding passengers. (Though now that I think about it, that sounds kind of fun. Memo to myself: start this club.)

I’ve previously written about why trains on the mall won’t wait for you (short version – the pre-empts cascade which will delay buses if a train does not go when its signal is up).  There are also other areas of the alignment where tight schedule timing means you really can’t expect a train to wait for you, and Gateway is one of them, especially during rush hour.

TriMet TV had an episode about getting left behind, but it didn’t go into a lot of detail about why trains, even more than buses, can’t wait for you. It is more complicated for a train than it is for a bus to wait for a late runner. For one thing, as narrator Shirley mentions, a train waiting for people can make the trains behind it late. As an example, check out the schedule westbound at Gateway in the early morning:

Minutes apart.

Those “Blue Line to Gateway TC” trains listed in the above schedule are actually Green Line trains that are coming into Gateway from Ruby. They go up the auxiliary track into the pocket track, swap cabs, and then go out to Clackamas.

Green waiting in the auxiliary track for a Red coming off the fishhook. Both will be using the pocket track, scheduled about a minute apart with the Red Line going through westbound first, then the Green Line pulling in to swap cabs

I’m not positive which train Dr Jeff tried to get on, but based on the time I am guessing it was 43, which is one that comes into Gateway from the airport as a Blue Line to Hillsboro in the morning. If that’s the case, then that train would’ve come up into the pocket track at Gateway, rather than the westbound mainline.

Gateway Review: The pocket track is the red one in the middle used (mainly) by westbound Red Line trains.

If the train he wanted was in the pocket track, that could be one explanation why it didn’t wait. Trains in the westbound mainline track use signal 72, and that signal does not time out. In other words, I can come into Gateway in the westbound main track, call my signal and sit there, and it will stay permissive for me. However, trains in the pocket track use signal 74, which *will* time out after 90 seconds. So if I come into the pocket track, call my signal and get a green or yellow, but then sit there, after 90 seconds it’ll time out and go back to a red.

Granted I wasn’t there and I don’t know all the circumstances surrounding Dr Jeff’s train leaving that morning, but I’m just throwing it out there that it’s a possibility that if his train was westbound in the pocket track, they did not have time to wait for late runners if they had a permissive aspect on signal 74 because it times out. And given how many trains are moving through Gateway at that hour, waiting to recall it can mess up the schedules of a lot of other trains.

Gateway isn’t the only part of the alignment that can be a time sink if trains aren’t running on schedule. Rose Quarter is another place where you can be locked out by a train making a conflicting move.

Parallel move through Rose Quarter

At Rose Quarter, Yellow Line trains can only make parallel moves with other Yellow Line trains (this is a conflicting move for everyone else) so if a Red, Blue, or Green westbound train gets to Rose Quarter off-schedule, they may have to wait for a Yellow Line to finish going through the interlocking before them. Then on the western side of the bridge is another area for conflicting moves – you may have noticed this eastbound if you are on a train sitting at Old Town/Chinatown or 3rd & Glisan for a while, or westbound if you get stopped on the bridge span at signal 14. Yellows and Greens can make parallel moves on the bridge, and Reds and Blues can make parallel moves, but Yellows and Greens can’t parallel with Reds or Blues. And with a train on the bridge about every 56 minutes of the hour, that’s a lot of opportunities for lost time waiting for a train to finish a conflicting move if the trains are not on schedule.

In other words…

Stop thinking of the trains as isolated vehicles.

They don’t function in isolation, and a delay for a train in one part of the system can ultimately affect other trains in other parts of the system (the whole butterfly effect thing). I realize that the interconnectedness of the trains may not be readily visible to passengers who don’t understand why a train can’t wait just a few more seconds, but that doesn’t make it any less the main reason why trains don’t wait for you. And on a related note, I am not a fan of the “blame the operator” reactions from people who have no concept of what it’s like to operate a train. Obviously, most people haven’t operated trains, and so when people don’t understand how something works at rail and they ask questions, that’s perfectly fine. I am happy to answer questions as are many others. But it does rub me the wrong way when someone who has never operated a train makes assumptions about what will/won’t affect a train’s schedule, and then concludes that the operator must have acted out of spite. From the comments on Dr Jeff’s post:

It *might* have put the driver a fraction of a second behind schedule. It is not that they are cutting off hundreds of extra riders, it is that they are cutting off that last guy who isn’t really late.

 YES they get their jollies about leaving behind a few and quite often its more than one rider running.

But most of the time our transit people here are dicks, and I’m going to bet that driver was too. And yeah, I get having to maintain a schedule, but seriously stopping for someone who is right at the door will not throw them off.

As I said above, it’s really just more in the interest of not making multiple trains late that an operator often will not reopen doors for a late runner. It’s not because I’m out to get you (I haven’t started that club yet). Also keep in mind that there are a number of things outside an operator’s control that can make their train run behind schedule:

DOOR HOLDERS (“here I’ll just hold the door open for you while you buy your tickets from the machine/walk from a block away/finish your conversation/etc”). Yes, you can get on the PA and ask people to stop holding the doors, but many door holders won’t. And you can’t just leave the cab easily to deal with it in person – leaving the cab requires getting permission from Control, keying out, exiting the cab, locking the door, walking back to whatever the problem is, dealing with the problem, keeping Control informed of the situation/outcome, walking back to the cab, etc.

– Slow orders (due to weather, track condition, etc) where you’ll be running below the normal speed limit

– People who don’t realize they’re leaning against the bridgeplate button and therefore the ramp comes out at every. single. stop. Again, you can get on the PA and ask people to check if they’ve got something leaning against the button, but if it’s in the trailing car and the person doing it is obliviously rocking out on their headphones, they’re not going to hear you.

Workers/walking inspections in the ROW, where in high speed areas you drop your speed to 35mph until you pass them. This is a safety issue and you do not violate it.

– Mechanical issues with your train or one in front of you


Please do not be this kind of driver. That train can’t move until the Tahoe gets out of the way. Why the Tahoe driver thought it’s a good idea to be hanging out over railroad tracks like that is anyone’s guess, but this is hardly an isolated incident.

This happens all the time downtown – cars fouling the ROW because they just HAVE to get through the intersection before a train does. Problem is, trains take up more space than just that area between the rails. If the back of your car is hanging over the different color concrete, you’re in the way.

All of these things can slow a train by seconds or even minutes, and those delays can really add up from one end of the line to the other, and there’s not much you can do as an operator to control it. But the one thing that an operator *can* control to reduce delays is not reopening doors for late runners. So even if it looks like it might just cost an operator a few seconds to reopen the doors for you, you don’t know what other delays they’re going to be facing (or have already passed) which is why they really can’t afford to wait for you.

Oh, and as for why an operator will sometimes make eye contact with you as you run to the train but still not open the doors? Glancing in the direction of movement is an instinctive response for someone who has made driving/vehicle operation their profession! And building on that, if the operator sees you doing something extremely stupid to run for their train (e.g. running in front of an oncoming train in the other direction or running up the ROW toward the train you want), don’t expect your stupid behavior to be rewarded with a ride.

I’m just truthfully sayin’.

Portland, stand up. Yeah.

Someone put Chillaxin’ on the MAX on Reddit a few days ago. This is the second time MAX FAQs ended up there (the first, strangely enough, being my post about dynamic braking – who knew the internet would find that so interesting?), so for everyone still wandering in from Reddit, hi.

Anyway, some good news: we now have the video PSA for Chillaxin’ on the MAX.

Can we bring these guys back for the Orange Line opening? I think I like them better than the Green Line dance team.

MAX accessibility

One disadvantage of spending a lot of time working with/thinking about/talking about the trains is that it becomes easy to forget that just because something is obvious to me does not mean that it’s also going to be obvious to tourists or other people who don’t ride the trains often (why no, most people do not have the motto “Eat… Sleep… Run trains”).

Case in point: I was downtown last week on one of my days off, walking east on Morrison near the Pioneer Courthouse and Pioneer Place Mall. I crossed 5th Ave and saw a woman in a wheelchair on the westbound platform waiting up near where the lead cab of the train would stop. Looking further down Morrison, I saw the train that would be coming in had a Type 1 as the lead car.

Type 1 with doors open at BTC – not wheelchair accessible because of stairs, no bridgeplates (and wheelchair lifts are no longer used)

So to save her (and the operator of that train) some time, I mentioned to the woman that the front car of the incoming train would have stairs and no ramp, so she would have an easier time boarding if she went further down the platform to where the low-floor would stop. She thanked me for letting her know and asked “How were you able to tell? I’ve never known where to wait to get on.”

Type 1, Type 3, Type 2, Type 4. Elmo Yard

I pointed out the windshield of the train: Type 1s, which are the only cars in the fleet with stairs and no bridgeplates, are also the only ones with a bevelled cab where the cab windows flank the windshield, making it look like the windshield is in three pieces. You can tell a Type 1 at a distance because all of the rest have a single pane windshield and the cab windows are on the sides. If you see a train with a 3-piece windshield coming in, then the front car of that train has stairs and the trailing car will be a low-floor with bridgeplates (this keeps every train ADA accessible).

Picture courtesy of a friend at Millikan Way – approaching eastbound train with Type 1 lead car

Elmonica eastbound platform – that’s a low-floor as the lead car

I also said that since the cars with stairs are always coupled to a low-floor car with bridgeplates, the best place for people waiting to board with a mobility device is about halfway down the platform. That way, it doesn’t matter whether the leading or trailing car is the low-floor, the passengers will have at most half a car length to go to get to a door with bridgeplates. This is mentioned on TriMet’s webpage, but I don’t think it’s very well-known because I see a lot of people who will need bridgeplates waiting at one end of the platform instead of the middle (in which case they’ll need to go almost a car-length and a half, or roughly 140 feet, to get to a bridgeplate door if the end they’re waiting at gets a Type 1). Not a small distance to cover in a hurry!

Related to this, I went back and dug up a post I remembered of Al M‘s from a while back showing video of a TriMet board meeting where a panel gave a presentation on serving the needs of an aging society. One of the comments brought up during their talk (at about 40min 30sec in the video, I looked it up so you don’t have to) was that they wanted to see the low-floor car always in the same location on every train – by which I’m assuming they mean something like the trailing car will always be a low-floor so people with disabilities know where to wait for easier boarding. They thought this had “been addressed” but “apparently that’s not the case, it still moves around and makes it hard for people with disabilities to know where to be on the platform.”

My Twitter background. On the left, Train 9 had a trailing Type 1 coming in westbound to Hatfield; it will have a leading Type 1 going east to Cleveland

Looking at the video of the meeting, I don’t think anyone responded to this at the time, but the answer is that it does not work, practically speaking, to always have the low-floor car be in the same place. The alignment isn’t a loop on which the trains run in a circle which would allow the same car to always be in the front. At the ends of the lines, operators switch cabs – what had been the trailing car coming in is now the leading car going out. It’s not going to happen that a pit crew will be on hand to uncouple the cars, shuffle them, and recouple so that the cars will be in the same position going out as they were coming in (though that would be pretty cool, now that I think about it..) There really isn’t anything to “address” with this, aside from TriMet proactively educating people with disabilities on where to wait to board the trains. More outreach needed maybe?

Bridgeplate, Type 2

One more thing about the bridgeplates that often isn’t known by people who aren’t regular MAX commuters is that the doors must be closed for the bridgeplates to deploy and retract. Many passengers who aren’t used to how the bridgeplates work will see the doors start to close and force them to reopen because they think they’re going to be left behind. This is how the doors are designed to work though, and the doors will reopen once the bridgeplate is extended. If you need the bridgeplate and the operator sees you, they can be deployed from inside the cab. This is faster than hitting the blue bridgeplate release button outside the train since you’d have to wait for the doors to close first. However, if the operator does not see you and deploy the bridgeplates for you, you can use that button yourself as long as the doors are open or on release. The doors will close, the bridgeplate will extend, and then the doors will reopen so that you can board the train.

This is a photoshopped train to Portland Airport

I’m probably late to the game on this because I never pick up maps/schedules for the trains (given that I have no need of them..) But anyway, this is the map of the Red Line that’s currently available. Do you see what I see?

Hey wait a minute, that’s not really a Red Line train!

Those of you who read the train number post might have recognized 61 as a Yellow/Green Line train number. Or you might have noticed the double white line of the transit mall (where Red Line trains don’t run) in that picture. The really sharp-eyed readers who use TriMet’s website might have realized this is the same picture of a Green Line train used on the main page for MAX light rail service. It looks like a red square and “City Center” were photoshopped over a green square with “PSU” as the destination sign.

Which.. well, that’s just odd. Why not use a picture of an actual Red Line train for a Red Line map? I’ve got plenty of them. I would’ve donated one for the cause!

Here’s one, though it’s of a Type 1 (so the bottom part of the map where it says “Roll your luggage on board” doesn’t really apply since Type 1s are high-floors) but at least it’s a real Red Line train.

Or if we want to stick with the Type 4 theme – same car even! – here is another picture I took, but unfortunately the destination sign would still have to be photoshopped due to the self test error (which is admittedly why I took the picture in the first place).

And I have this Type 4 at night…

Also, the full uncropped version of my old header image…

Obviously this is not a post of any real consequence… just a little thing I noticed that seemed strange. I’m curious to see what will be on the cover of the new map in September (when that $2.35 fare is no longer accurate)