Automatic Train Stop

A pause in the signal series, because it was getting too difficult to write about the last type without having explained ATS first.

ATS, or Automatic Train Stop, or “What keeps the trains from crashing into each other?”

Link to Wikipedia article, for those that like that sort of explanation

ATS in t-railATS magnet.  I don’t remember where I took this pic, but I think it’s Beaverton Transit Center, westbound platform

ATS in girder railATS in girder rail, Lloyd Center westbound platform

These little yellow rectangles are found all throughout the alignment, much to the delight and happiness of rail operators.  These are the ATS (automatic train stop) magnets, and given their name it’s pretty easy to figure out what they do. If a train goes over one while the magnet is active, the train automatically comes to a stop.

They are associated with every signal capable of displaying a red aspect, and will be active as long as that signal is red.  So, by default, a train physically cannot run a red light – attempting to do so will bring the train to an irretrievable stop.

ATS magnet and red signal aspect, Sunset TC westbound.

That magnet is currently active. Stopping the train and selecting for a permissive signal will turn the magnet off once that signal displays something other than a red aspect, which allows the operator to move the train forward again. If the operator had tried to keep going, her train would have been brought to a stop (and she’d have some explaining to do and paperwork to fill out!)  Again, if you scan the radio, that’s what’s called a “trip”, sometimes also referred to as “popping a red.”

ATS magnets are also located in areas where it is really unsafe for a train to speed – for example, coming into Gateway TC from any direction because of how busy it is, or the single track that goes into PDX International Airport.  These “speed trip” magnets have a pickup a set distance away from the magnet depending what the speed limit is, which activates the magnet for as long as it would take a train going faster than the posted speed limit to reach. So with a 15mph magnet, for example, if a train is doing 16mph when it goes over the pickup and doesn’t slow down, when it goes over the magnet it will trip it and come to a stop.  And again, the operator will have paperwork to fill out and explaining to do, because any type of ATS trip – from running a red or speeding – is a rule violation.

How a speed trip worksMore or less how ATS works for speed – let’s say that distance takes 30 seconds to cross if you’re going 15mph. If you operate a train through it and it only takes you 25 seconds, you’re speeding, and that magnet will stop you.

Although new operators especially don’t like the magnets since it’s hard to remember the speed limits of all parts of the alignment at first (and therefore easy to get tripped!), they are an extremely important safety feature and it’s a very good thing that they’re there – they prevent collisions in ABS territory (which covers all high speed areas) because a train will come to a stop at the red light, long before getting close enough to the train in front of it to hit it. And they prevent derailments or other accidents in areas where speeding would be extremely dangerous.

19 responses to “Automatic Train Stop

  1. Do you know if the ABS system uses overlaps? That is, is it possible for you to encounter a non-red signal followed red signal, then a stopped train immediately ahead? The ATS won’t really be much use if that were the case, but with overlaps, there would be some distance ahead of the red signal guaranteed to be clear, so a train that is tripped by ATS would almost certainly stop before hitting another train.

    • EDIT: I incorrectly guessed no to this question originally, but signal designer below posted that overlaps are in fact used. Scroll down for more info.

      • Yeah, that sounds about right from what I’ve seen of the Portland system (which admittedly is not much). One other possibility is that the overlap is actually the length of a full block, but that sort of thing generally only happens in places like the NYC subway where blocks are fairly short. What about home signals (where there are diverging or especially converging switches ahead), are those also placed right at the point of convergence, or is there actually room for a tripped train to stop before it hits another train moving on a conflicting route? That tends not to be standard practice on mainlines in the US, but in the UK for example, they like to leave 200 yards from a signal to a converging junction for the overlap.

        • For the most part, no, diverging signals are not placed immediately at the switch points, so if you trip one you’ll stop before you get to the switches.

          At first I was originally going to say all of the diverging signals are far enough away from the points that a conflicting route wouldn’t be an issue, but then I thought of this signal at Elmonica (one of the rail yards). The picture was taken facing east, and so an eastbound train here would be stopped at the platform behind me to select that signal, either to diverge into the yard or to continue on the mainline. So to that extent, if they were stopped at the platform and tried to move on a red, they’d be tripped almost as soon as they started. However, I suppose it’s mechanically possible for a train to disregard the platform and attempt to continue east without slowing or stopping – yes, they would be tripped at the red, but if they were coming at it with enough speed, I don’t know that they would be stopped before those switch points.

          Aside from a situation like that, of all the possible diverging signals I can think of on the mainline where a train wouldn’t expected to be at a full stop prior to the signal, yes, they’re set well ahead of the switch points that a train tripped at one would be stopped before reaching the switch.

  2. If you trip the ATS magnet and the train stops, what’s involved in getting it moving again?

    • You call Control and with their permission reset the train. If you tripped a signal they’ll probably have you pull back since bypassing a red means running restricted speed, stopping at every set of switch points, etc, so it’s faster to pull back behind the signal and call it (or wait for it to turn yellow if it’s a non-selectable signal).

      • Is this the same procedure for an overspeed?
        Also, I thought you weren’t allowed to back up on the mainline. Does this mean you actually have to move to the rear cab?

        • No, an overspeed happens when a train goes over about 58-59 mph. That brings the train to an MSB (maximum service brake) stop. Overspeeds happen because of a governor in the train, not because of anything in the trackway like an ATS magnet. The magnets will be associated with either an ABS or ABS/combination signal or a speed trip pickup, and ATS magnets will only stop a train at that magnet. Overspeeds will stop a train wherever they are if they go above the internal governed speed.
          When you get an overspeed it’s usually because you were coasting down a hill when you thought you were in a speed maintain mode to hold the train at 55 mph. If you get an overspeed, you pretty much just let the train come to a stop, ring the bell twice, and keep going. Overspeeds are not rule violations and don’t need to be called into Control like ATS stops.

          And yes, to pull a train back, you have to walk back to the trailing cab to pull it back far enough, then key out of that cab and walk back up to the lead cab.

          • This seems like it could be a real pain on a 4xx train that has had the second car taken out of service due to a problem.

            • How so? Type 4s are never run as single cars because of their 1-cab design.

              • So there’s not even an emergency provision for single car operation on the 4xx trains? That seems like yet another deficiency with that design, up there with making impossible-to-use seats in the center section and sacrificing a regular cab just to come to the same seating capacity as the original 1xx cars…

                • Not really, no… They can be operated from the B-end, but are limited to 5mph and so that’s only done in the yards. I’m really hoping that Mary Fetsch’s comment about the Type 5s that “We’re taking lessons from the Type 4 and applying it to the Type 5” includes an operational cab at both ends. Right now if there’s something wrong with a cab of a 1, 2, or 3, that can be buried in the coupled end and still function as a train, but that flexibility is not available with the 4s.

                  • I hope they learn the lessons of the 2’s, 3’s and 4’s and bring back some of the amenities of the 1’s including quieter ride, greatest seating capacity, and bicycle facilities with seats. Really seems like it would have been better to raise the platforms and redesign the existing car entries on the 1’s than go to the low floor design on the rest…if they can’t come up with a quiet low floor with plenty of seats with six bike facilities (four with seating), on a single car, then I’m willing to declare the entire low-floor concept inherently flawed and a tree Portland should stop barking up.

                    • For passengers, I think the layout of the 1s is the best arrangement of seating, but that doesn’t translate well to a low-floor car. I think the layout of the seats in the 4s is awful (and I remember thinking the same thing when they were first delivered at Ruby Junction – people are NOT going to like how cramped these seats are!) The 2s and 3s are fine as far as seating goes though I know the lack of bicycle space translates to full racks and other bicycles blocking the aisles. The upper deck design in those cars is much better than the 4s as that aisle in the 4s is an awful tripping hazard.

                    • Oddly enough, I think I like the upper deck design in the 4’s better than the 2s and 3s because it’s harder to lose your footing, as the step spacing is substantially longer than the rather shallow, tall steps on the older cars that make them a real hazard on wet/icy days especially if you’re wearing leather soles…

  3. Might want to mention that the magnets in the ATS Gleismagnets are permanent ones and are therefore failsafe. The magnet, less any outside influence, is always creating a magnetic field and detectable by the pickup under the car. Only when there is an electric current going to the ATS magnet (on anything other than a red signal) does it cancel out the permanent magnet field and allow the train to pass over it as if it wasn’t there.

    At Ruby a while ago (and still may be there) on the westbound track, the ATS magnet protecting a eastbound move to the switch into the throat had no wires connected to it at all.

    If the signaling system went out, all the magnets in the ABS areas would need to be bypassed, a real headache!

  4. FYI- The first question in this thread you did not answer correctly. There are indeed overlaps on the Tri-Met system. I have always reffered to them as block and a half spacing. In other words you can’t get a clear signal until the train in front of you is braking distance (at 59 MPH, unless in speed zone) beyond the red signal that is also in front of you. This will indeed ensure that you stop prior to running a red signal before finding the rear of another train.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s