1995 Russell Hill subway accident
The railway accident by Russell Hill ( Russell Hill subway accident) was a rear-end collision on the Yonge- University - Spadina line of the subway of Toronto ( Toronto Subway ) in Canada. On August 11, 1995, shortly after six clock in the evening, went a wandering to a standing in the tunnel under the Russell Hill north of Dupont station train. Three people were killed and 30 others seriously injured. This was due to human error on the one hand, on the other hand, a design flaw in a driving ban.
Expiration
The Metro Toronto was working at that time without cab signaling on the basis of optical signals. Simple route signals were working with a three-color lamp - Red means stop, yellow has a red at the next signal, and green for roadways. Before diverting the signals have two lights: the top one shows as before, the route clearance, the lower signal is also red at stop, and shows yellow at a branching position and green in performing position. In an ordinary passage to a distant signal so it can stand on yellow to green, this is to be treated as a simple yellow signal.
In addition, there are signals with another white lamp, which serve to speed optimization because so-called " grade timing" ( roughly " period method "). If a pre-signal yellow shows, and points to a subsequent main signal with red, then white can also be switched to point out that in compliance with the speed requirements for this section, the next signal will be on transit, when the train reaches the signal. On longer downhill stretches or on curvy roads with poor visibility, the track is equipped at short intervals with signals. In ordinary driving it the next signal will be on red to white after a yellow to white, and when approaching it jumps from red to white to yellow to white, and the shift lock is lowered. In the metro Toronto, drivers there have a habit of pushing the tolerances of the system, and fully close the set speed on a red and white signal - one is for a yellow-white signal by mistake but too fast, it is braked by the shift lock.
On the Streckenabschnintt of Station St. Clair West to Dupont Station, the " grade timing" system is applied throughout. The signal SP77/X38 is behind the station St. Clair a switch signal that regularly on yellow over green over white stands - the next signal is red over white, the route, however, can be expected to be free. However, on the day of railway accident, the signal was on yellow over green, so that the following signal SP71 are red and an unconditional halt to expect. The reason was that still stood a train in the following section.
The platoon leader Robert Jeffrey later testified that he remembers signal SP77/X38 than yellow over white. Although this is impossible at this signal, but perhaps a reflection on one of the lamps was misinterpreted. He was in any case of active " grade timing", and consequently drove unrestrained at next signal. Jeffrey does not remember the following signal aspects. However, these can easily be derived from the function of the system.
The following signal SP71 that consequently stood up red, the shift lock was not triggered, so that the train went through unrestrained by a design flaw. Why the platoon leader, the holding commanding signal not noticed, could not clarify - maybe he assumed that it was on red over white and the technically allowed passage simply thus is related that it jumped shortly before to yellow over white. So he went to the subsequent stretch further from an active " degrees of timing" and was based on the usual line speed.
Thus, the train continued on to the third signal SP65, from which he had to assume that it would jump in the same manner of maintenance on transit. This showed red - not red to white - and in turn disobeyed Jeffrey the maintenance requirement. He was active until the emergency brake, as the standing train came into view, which was waiting at the subsequent signal SP53.
Each train was staffed with about 200 to 300 people. Upon impact, at 18:02 with about 50 km / h 3 passengers were killed, 30 were seriously injured taken to hospitals. About 100 people came forward later because of lighter injuries for compensation. After the accident, the track was closed for five days.
Investigation
In addition to the slightly ambiguous signaling, the studies have been focused on the shift lock. As is customary in North America, there is a T-shaped lever, which is placed on the right side of the bus bar. In transit, it is down when the associated Lichsignal shows red, the lever is positioned - in this position the head element will hit a switch on the first car that triggers the emergency brake of the vehicle. Among those employed at that time in Toronto trains the H5 series, there is the trigger on the first Achsgestell where the position on the track and the distance to the train stop can be best secured.
In this case, a train stop from Ericsson was used, which is actually intended for mounting on the inside of the bus bar, and has been modified over an extension piece to make them usable on the outside of the busbar can. During the investigation, it turned out that at that time the 18 year old design, and in the bearings of the driving ban after some time have some play to protruded a few millimeters into the clearance space of the vehicle. Under normal circumstances, a vehicle will fail to complete the gauge - but if rail and wheels are a bit worn, and the signal is in a curve like this, then pushes the vehicle very far out of the center.
In the railway accident, this meant that the front of the train anstubste the established movement blockade and so easily pressed down. She swung then right up again, but not fast enough to pull the trigger on the first Achsgestell still can. Ultimately, the design of the traction lock was ineffective on this route for the command.
The investigating authorities gave in its report of 18 recommendations for improving the safety regulations. The Toronto Transit Commission created a 236 point checklist Based on these recommendations and reviewed annually thereafter their implementation ( including the replacement of all ignition interlocks ). In August 2009, only two points were open, which affect a system for speed control ( this system was then in the implementation phase ).