Switching troubles halt north Red Line for two hours

A Red Line train approaching Granville “took the wrong route” early Monday afternoon and ended up straddling two set of rails, says a Tribune report. The CTA corrected their own original assessment of a derailment in the incident in which no passengers were injured.

Before it all was over, the Chicago Fire Department sent five ambulances to the scene, thinking a derailment had occurred. But the CTA was able to uncouple the rear car and move the rest of the train back to Loyola station, where passengers exited safely.

Service between Howard and Belmont was suspended soon after the incident occurred at about 12:45 p.m. The CTA was able to restore full service by 2:30.

In April of 2011, a similar incident occurred on the Brown Line, with the CTA insisting it wasn’t a “derailment” because “all of the wheels stayed on the track.”



Leave a comment
  • The real questions are:

    ...how does the operator run over an open switch? Aren't there supposed to be "red over red, stop and stay" signals there?

    ...how did this get switched? Given that there were traffic reports that the Yellow Line was also suspended, is the new Howard signal system fouling up stuff 2 miles south?

    For that matter, did it have anything to do with the Red and Dead project?

  • In reply to jack:

    The NB Red Line was a mess just before that happened.
    I was on the next to the last NB train to get through & got off at Loyola.
    By the time I got downstairs, the follower was arriving at Loyola, maybe 4 minutes after the one I was on.
    The train I was on went slow as can be, even through areas without major slow zones, but the motorman kept it really slow, far after passing the Argyle construction site.
    This was another time that the way the train is operated affects the entire system. [I forget, is it a dwarf or wayside signal there?]
    But I have to wonder, how did the SB motorman not see the red signal for the crossover or that the switchpoints were turned?
    And were the points for the switch on track 3 also thrown?
    There was a serious possibility of a head on crash if that had happened.

  • Jack-

    I have been asking myself the same question since the incident occurred. The basis of the Granville interlocking-the switches, signals, crossover tracks, etc.-is that a signal that governs a route over any type of movable switch cannot clear beyond a STOP indication unless the switch is in the correct position. Contacts in power-operated switches provide feedback to the train control system to positively indicate that the machine and the switch points are in the position that the system called for. In hand-operated switches, contacts in those devices tell the signal system that the points are in the normal position (set for "straight" route) and that the machine is electrically and mechanically locked against movement.

    I was under the impression that the switch that was set incorrectly was a hand-operated machine. Again however, the system should have detected that the machine was not in the correct position and the governing interlocking signal should have indeed been displaying a Red over Red/Stop and Stay aspect.

    The CTA, like many heavy rail transit systems uses "train stop" devices to provide red-signal enforcement. When the signal is red, a tripper arm (usually painted white and installed in close vicinity to the signal) is raised above the running rail. If a train attempts to pass it in this position, the head of the trip makes contact with a switch on the train which causes the train to stop immediately (no operator action required). When a signal is called to clear, the trip is driven down below the rail head where it cannot contact the trip switch on the train.

    Given what I've said then, why did the train not stop? There are a few things that could have attributed to this incident:

    1. The contacts in the machine did not follow the actual position of the switch points. If they did not open (as they should have if the switch was open) then the system *may have* allowed the governing signal to clear.

    2. If the signal was indeed displaying a Red/Red aspect (STOP) then the trip should have been in the stop position and the train should have been stopped automatically if the operator passed the signal. It is possible however that the head of the trip arm did not make contact with the trip switch on the train.

    3. In addition to the wayside signals that the CTA uses, they also use a cab-signal system. Equipment in wayside houses transmits coded signals in the rails which are picked up by coils on the trains. The received information is used to display a maximum permitted speed in the operators cab. On-board electronics compare the actual speed of the train to the allowable speed and in the event that the operator exceeds the permitted speed, an alarm is sounded. If the operator does not comply with the alarm by slowing the train, the system automatically stops the train.

    The permitted speed information that is transmitted to a train is usually affected by the condition of the wayside signals in approach.If the signal is cleared for the straight route, the cab signal display in the operators cab will typically display the maximum permitted speed. If the signal is displaying a stop-aspect or a clear aspect for a lower-speed route, the allowable speed is typically adjusted accordingly. So, if the signal was displaying a stop-aspect, the allowable speed in approach to it should have been less than that compared for normal conditions. This usually alerts the operator that something is not right and we would *assume* that he/she was looking carefully at the road ahead to determine what is not right.

    In the end, these are all just hunches. On the other hand, derailment or not, getting part of a train on a different track is a serious incident.


    The CTA does not use dwarf signals. The preference is that a wayside signal be of a "high" configuration whenever possible. However, there must be enough room next to the tracks so that the signal can be installed far enough from the train so that it cannot strike an operator who may have his/her head out the window. Where clearance is limited, they will install a "low" configuration. Usually (although not always), signals installed between tracks will be low due because the tracks are simply too close.

    If this crossover was controlled by power-operated switch machines then typically the machines at both ends/on each track operate together as a pair. However, I think that the crossover in question is controlled via hand-operated switches. With that type of equipment, each switch can be operated independently and thus, the other switch may have still been lined for the normal/straight route. However, as I mentioned above, the cab signals (and any wayside signals) on *both* tracks should have reverted to STOP if the *either* end of the crossover was unlocked.

  • In reply to GREENOVERED:

    What you say makes sense.

    The question though is how these kind of incidents occur, such as the one at 59th Jct. Some former CTA employee on chicagobus.org said that it was that the operator ran over an open switch rather than the switch opening under the train, but, from what you say, the train stop should have stopped the train.

    Similarly, in this incident, if we assume that the wrong routing was established because of the north Red project, one would have thought (as the operator did) that she could have backed out, but reports are that she couldn't.

  • I've worked in signaling for a long time and the question that is asked of me most is "if you design systems that are supposed to make everything safe, why are there still so many accidents?" My answer to this typically is simply, "See that person in the operator's cab?"

    Incidentally, "backing up" a train is what has caused many past incidents at 59th Junction. Here is the typical scenario. A northbound operator is approaching the interlocking from Cottage Grove and a turnout beyond the home signal is lined for the northbound move from Ashland/63rd (lined against the operators desired route). The signal system understands this and accordingly, the facing signal is set to STOP. The operator passes the signal and gets tripped. Realizing his mistake, he attempts to back the train up clear of the signal. Unfortunately, the lead truck has "trailed" through the switch before the train stopped. In this, the flanges on the wheels have either forced the switch points open as they passed through or have ridden up and over the switch points. However, the switch machine is rigidly coupled to the points and maintains their position via a "lock rod" (the machine mechanically unlocks only at the begining of its throw and locks again upon competion of the movement). So, even though at least one axle made it past the switch points, they are still lined for the incorrect route. In fact, because the train passed the home signal and is now occupying the interlocking, the switch will also be electrically locked against any movement (any time a train occupies the track circuit over a switch, that machine is prevented from the system moving it). As the operator backs the train, the axles that managed to pass ("trail") the switch points begin moving onto the conflicting route while the rest of the wheels, i.e. those that didn't pass the points, move back on the route from which the train originally came. In short, parts of the same train move on different routes to the result of bad things happening.

    Of course, let us not forget rules. Every railroad and transit property has operating rules, the CTA being no exception. Here are a couple that must be violated in order the create the oft-seen chaos at 59th:

    1. Don't back up a train. You are supposed to operate the train from the head car in the direction that you are moving. Makes sense, right? You should see be able to view the road over which you intend to move. Of course much like an automobile, you can also back the train up/operate in reverse with the big difference being that there is no rear-view window through which you can see what is behind you. In the example that I provided above, the result would be the same regardless of which end you operated from. But this action is often where the excitement begins.

    2. If you get tripped, get out and walk your train prior to proceeding to attempt to assess the cause. Not a bad idea. It could have been a damsel in distress tied to the rails that gotcha. Or you might notice that you trailed a switch. Or that your flanges are firmly wedged between the ties.

    3. Call the rail controller over the radio and explain the situation. "But I'll get in BIG trouble," they say. On the other hand, the controller may be able to calm the operator down and help him/her think clearly about the situation. Heck, they might even be able to find a supervisor or maintainer that could come assist and prevent a little problem from becoming front-page news.

    So the operator figures, heck, I just need to back up a weeee bit and I'll get myself out of this mess. Doesn't bother checking his equipment, doesn't call anyone. The following day he is reading about his mistake in the Trib.

  • In reply to GREENOVERED:

    Thanks for your explanation. It appears thorough, although I can't say that I understand all of it.

    It does seem like it does explain what happened when the operator tried to back up at Granville.

    I challenged the "former Ravenswood operator" about an hour ago to come up with an explanation. Let's see if he comes up with anything close to this. Up to now, he has been evading.

  • Jack-

    I am happy to clarify anything that wasn't clear via this forum or via my outside email, greenovered@ymail.com. FYI, some of the best operators I knew/know were/are motoring on the Rave.

Leave a comment