Back in early December, I went to a dinner for the Northstar chapter of the National Railroad Historical Society where Bill Kuebler gave a presentation about a crash of the Northern Pacific's North Coast Limited as it passed Granite Lake, Idaho at 10:18 pm on March 2nd, 1962. Both the locomotive engineer and the fireman died after the lead locomotive left the tracks at high speed, tumbled through the air, slammed into a table of land below (the fatal incident), and finally slid into the lake.
I should have written down more stuff after I heard the presentation, since a lot of the details have faded. Fortunately, Bill Kuebler does post things on the Internet on occasion, and I've been able to resurrect a few of my hazy memories by reading his comments in this discussion thread.
Anyway, the incident occurred on a high curve of nearly 8 degrees near a trestle in a mountainous area, and I think the superelevation was about 7.5 inches. That level of banking is disallowed by regulation today, though it could have saved the lives of the crew if they'd only slowed down by about 10 mph. Speeds were technically limited to 30 mph on the curve, though the overturning speed of the locomotives at that site was calculated to be 69 mph. Based on handful of skid marks left by wheels of the locomotives as they flew off the rails, it was estimated the lead engine hit the curve at 75 to 80 mph. Fortunately the rest of the train hit the curve at lower speeds, so the passenger cars slid to stop without falling into the lake, and additional fatalities were averted.
On paper, both the engineer and fireman had a lot of experience. The engineer was a pretty competitive hothead, but still good at his job: He had been operating over the territory since 1916 and knew it very well. He also knew how to push a train to its limits. The fireman was kind of the opposite: He basically followed his instructions to the letter, but was unable to respond to changing situations. He had also been operating over the territory for 20 years and was qualified as an engineer, though most of his experience was with slow freight trains that didn't require as much thinking ahead as fast passenger trains do.
Back then, it was common for train operators to flout the official speed limits as they made their way from point A to point B. A new travel time record through Montana and Idaho had just been set by another train crew, and the engineer was amped and ready to break that record.
According to the records of that night as the train flew past small stations and waypoints, the engineer had essentially achieved a new record miles before reaching the point of derailment. As the train approached Granite, the intervals for the last few reporting points showed the train operating at regular speeds. The fireman had been handed the controls, and was basically expected to guide the train in like normal—though he did have the psychological threat hanging over him that if he screwed up and wasted too much time, he'd blow the record-setting pace and get an earful from his cab-mate once they made it to their destination. (It is certain that the fireman was at the controls, since that's where his body was found.)
The curve at Granite was a tricky one—I don't know the ins and outs of train brakes, but the North Coast Limited had a fairly complex system which had been experiencing some problems earlier in the evening. It was pretty advanced, including features similar to anti-lock brakes on today's cars. An electro-pneumatic subsystem had been giving an earlier train crew grief, so that had been disabled for the rest of the run out to the west coast before this new crew even got onboard. The standard brake system was working just fine, as was shown by the fact that the engineer could race along for the first part of the journey, which included a lot of technically challenging curves.
Nonetheless, the train's brakes are important to the story: the brakes needed to be "set up" well in advance of reduced-speed zones. There was an "advance warning reduce speed" sign 3,076 feet away from the curve, but that was not far enough away for the brakes to have any significant effect if they had not already been prepared properly.
Train crews, especially those operating passenger trains with their unique brake systems and requirements, were typically made aware of this and told to watch for certain landmarks and set up the brakes by the time they reached a switch farther east along the tracks. In Kuebler's interpretation of events, the fireman lacked the situational awareness and forward thinking ability necessary to do this in time. There's some indication that he may have attempted to do the right thing, but was hesitant—at least one crew member reported feeling the brakes apply like normal ahead of the curve, but then they released seconds later. The fireman just wasn't quite sure where he was, and didn't want to screw up the speed run.
There is good indication that the fireman was paying attention up until the last second, however—just 700 feet from the curve was a grade crossing, and he blew the train horn accordingly in the standard long-long-short-long pattern. The engineer, who had most likely been taking his mind off things and yammering or resting elsewhere in the cab, appears to have been jerked back to reality by the horn and was probably reaching for the emergency brake as the engine started hurtling through the air.
The Interstate Commerce Commission was the government body responsible for investigating the crash (the formation of the National Transportation Safety Board still five years in the future at the time), though the Northern Pacific also did their own internal investigation. To their credit, the company took a multi-pronged approach to resolving many of the issues that contributed the incident as well as others that could lead to problems elsewhere. While the cause pretty much comes down to human error, that was really just the tip of the iceberg. The curve itself was pretty dangerous, coming up after a long stretch of fairly straight and boring track, so that whole segment was rebuilt in 1965 later with gentler curves. There was a culture within the railroad that allowed quite a bit of risk-taking behavior—especially speeding—and a crackdown ensued that really changed the way the railroad operated.
I was finally inspired to write this after reading about the crash of the Niagara Falls–Toronto VIA Rail train 92 in Burlington, Ontario this past Sunday, which seems to have at least a few parallels with the North Coast Limited crash. Three crew members in the locomotive were killed, apparently as the train switched tracks—an operation that's only supposed to occur at 15 mph even though the rest of the corridor supports speeds of 80 mph. The crew included two highly-experienced engineers plus one trainee—actually a 40-year-old man who had just been hired on after 20 years of service as a freight engineer.
The derailment happened just 1.3 miles away from the previous stop at Aldershot station, so it may not have reached high speed—It's not yet clear exactly how fast the train was going when it hit the switch because the onboard data recorder was damaged and is going to delay things by at least a few days. It also remains unclear who was at the controls.
Anyway, it'll be interesting to see what the investigation uncovers.