The following is pure speculation on my part, my initial impressions are based on very limited information available, including news reports which cannot be trusted.

Metro-North derailment in the Bronx - The train was coming off a 70 mph section of track into a 30 mph restrictive curve. Photos suggest the train left the track to the outside of the curve. Reports are that the cars on their side lost their windows, and my guess is that the major injuries and deaths were due to the cars rolling over and those against the window openings (think road rash and worse). Considering how far the train wandered away from the track and right of way, I really believe the train was substantially above 30 mph. There is also the reported comment by the engineer that the brakes weren't working properly.

Obviously a track failure must be considered, but probably not the primary cause. My guess is that the engine was equipped with blended braking. Dynamics are not a failure afe system, so the unexpected loss of dynamics might have resulted with the engineer coming in too hot and not realizing they needed to take immediate action to substantially increase air braking effort. Because of tight schedules, passenger trains typically run right up to a slow order, restrictive curve or signal restriction slowing or stopping at the last moment. You can't do that with freight, but passenger operations have traditionally had more leeway because of their priority, higher braking effort and shorter slowing/stopping distances. (Remember the delayed in block issue?)

But - the overturning speed on a curve is typically about double the normal safe speed limit, so the train should have been able to take the curve (assuming no track defects) at 45-55 mph without derailing. At 70 mph, my guess is that an emergency brake application should stop the train within about a 1/2 mile. If the engineer had forgot about the curve an emergency brake application should have gotten the train's speed down from 70 mph to a sppedd slow enough to make the curve within a 1/4 mile. I believe the train was at a much higher speed in the curve, and the investigators already know that based on the event recorder downloads (which they won't admit to yet).

There is always the possiblity of a brake failure due to a blocked or restricted brake pipe. The blnded braking may also be a suspect. Dpending on the software/arrangement, the system might allow for power/stretch braking. This might have been triggered simply by the engineer making a brake application with the engine still in power. Was the engine in full dynamics or power, and was it a mechanical or operator error if still in power?

My SWAG at this early stage - there was a failure of the brakes or there was a failure to apply the brakes.


Southwestern Railroad in New Mexico - Something does NOT add up here. Years ago my train went by a BN switcher that had split a switch and was defying gravity - I swear it was at 45 degrees. A few years later I was on a GP38 that dropped off a crossing (county had graded dirt over it) at 12 mph and made a right hand turn trying for the ditch. We stopped at at least 30 degrees, and I thought we were going over. (Glad we had a second engine instead of just a couple empties behind the engine.) In my experience, at 10 mph you are going to go about 150 feet before stopping in emergency even if the engine is derailed. Loads behind you on a down grade, a little farther.

Personally, I would expect to probably survive in the cab of an engine that rolled over once at 10-25 mph assuming the cab was not crushed in, though I have no desire to find out. Looking at the photos in the news link that AZebra provided in the previous thread, not in the upper photo there is a gouge from the track into to the ditch ending at the engine, and there is a stanchiron in the foreground. The right side and bottom of the engine have no obvious damage, so it only rolled onto its side, not a full rotation into the ditch. It looks to me the engine left the track rolling onto its left side and slid forward into the ditch. Another post suggested that the track was only good for 10 mph, but the photos suggest a much more violent derailment. The track might have been good for 25 mph, but I'm still not convinced it would have been that bad. Anyone in the cab would have faced the same potential meat grinder/incoming material through an open or blown out side cab window.

I agree the possibility of the engine hitting rocks and launching off the tracks previously menitioned by another poster. If the engine derailed due to rocks, it is likely the cars would have followed the engine into the ditch or piled up at the initial point of the derailment, and there is no obvious damage to the pilot to indicate it hit something or even slid along the rail, or plowed through any ground while upright. Look at the photos, the cars continued past the engine a substantial distance before derailing, with no evidence of track damage or derailed wheel adjacent to the engine. If they had been going 10-25 mph, the cars going into emergency should have stopped them closer to the engine. If the cars were loaded, there should be wheel marks, and if empty they should have stopped much sooner. There was a second derailment where most of the cars left the track and rolled in the OPPOSITE direction the engine did, to the outside starting at the beginning of a curve.

To me, the clues are obvious, and the probable cause can be summed up in one word - runaway.