I agree with the premise of what I believe to be a fact that the two were unable to discern that there were TWO trains approaching them. And I believe that at the point when #175’s engineer saw them, it was too late to save them. Unlike the movies, people cannot launch themselves off a ROW in three or four seconds. There is a discrete period of time to: 1 Sense a danger; 2 Comprehend a danger; 3 Determine a needed action; and 4 execute that action. If they had seen #175 as soon as it became in sight, I doubt they could have reacted in time. As it was they were focused on the train they saw on the other track. It matered not whether #175’s engineer applied the brake as soon as she saw them or when the train struck them. Their fate was sealed by them being on a active track without protection.
If you have seen the movie SULLY, you may recall that one of the ideas that made his defense was the time it took to determine where to safely land his plane. The “experts” said they had time to return to Laguardia with their simulations but did not include the time required to determine the problem and determine what action to take. After they added the reaction time, it became clear the Huson River was the ONLY choice and the correct choice for Sully’s plane.
The two men were in the wrong place. Hopefully, this will be a teachable moment that can be used to prevent future deaths. Does anyone know whether CSX and Amtrak have established rules and/or practices to prevent future events like this?
I don’t believe their fate was sealed until the point of impact. I see no outcome that I believe had to have occurred. Why would I believe something that is simply unknowable unless I just did not want to imply that the engineer acted impro
Once the birds took out the engines - Sully was fully engaged in his situation - mentally and physically as well as having his co-pilot fully engaged in what was taking place.
The CSX employees were never ‘engaged’ in their situation as it related to train #175 - as they were focused on train #66. Their lack of reaction to #175 was the cause of their demise.
What you say is true, but what does it have to do with the contention that the fate of the two conductors was sealed prior to their deaths? That was the point I was making, along with the point that just because Sully dealing with his emergeny took a long time, it does not mean that it would have taken that same length of time for the two conductors to become aware of their plight. Either incident could have gone either way regarding a fatal outcome, regardless of the amount of focus involved.
Since this is sneaking back into a general discussion of the accident, let me remind everyone that the original ‘source material’ is not in the defectively written report, but in NTSB docket DCA17MR010, which you can access here, for reference.
I see that in the BLET account the issue of where the conductors were has been resolved: they were walking in the gauge, and some time after they saw 66 approaching they moved across to the “tie butts.” If you look at the pictures of the train relative to the two Amtrak mains in the pictures, you can quickly see the possibility of sound from 66 reflecting from the freight consist interfering with anything coming from #175 behind.
It is in this primary material, not a cobbled-up excuse of a summary, that any insight about this accident … or actual material used for prosecution or defense … would come. The effect on the press and, through them, the general public, which will never see the detail material and may see only a highly colored interpretation of a flawed piece of expository writing, is quite different, and I have to wonder if the ‘spin’ implied by the error was intended to act there for some purpose.
Had #175 been there earlier or significatly later you might be correct, but with the sequence as it occured. the only thing that I can conjur up would have been if 175s horn had sounded first and given them enough sound to overcome the noise of #66.
It is possible that had they had something alert them to #175 approach, they might have been able to react but at 75 ft/sec there wasn’t IMHO enough reaction time to savethemselves.
I agree. Have you ever played the game where someone holds a dollar bill between your thumb and index finger and says if you catch it when I let go, you can keep it. Most of the time, it will clear your fingers before you can can close them on it. That is a demonstration of reaction time.
Depending on the horn signal structure and timing, it was possible at any moment for the signal of #66 to cease and open up a space in which the signal from #175 would have been recognized as coming from behind them, and thus alerted them to jump into the clear. It was also possible that one of them would have just spontaneously turned to check behind, which they have have done several times prior to the approach of the two trains.
Your dollar bill example is a demonstration of reaction time, but the potential
Note that reflection from the CSX train could have the effect of delaying echoes of 66’s horn to make them appear to come ‘from behind’ with about the same timing and perhaps imaging as anything that would have been coming directly from 175. We haven’t discussed the effect of Doppler shift on resolving the two-train issue; it applies in many situations, but perhaps not here. It appears the conductors were alert enough to move from the gauge to the tie-ends upon seeing 66… but not all the way out of fouling Amtrak’s line.
Perhaps I’m missing something, but I note pointedly that the event-recorder data for the locomotive on 66 was neither provided nor analyzed. There is a very important datum there – did the engineer lay on the horn continuously, or nearly so? That plus the echoes would mask anything from 175… and is consistent with their behavior to get further from what they thought was the ‘active’ rail.
Were I the engineer on Train 175 I would have used a succession of short sounds on the horn to indicate impending disaster and I would have applied the emergency brake far sooner than Ms. Reese did.
Something that has been ‘added’ since the beatings stopped is the datum that the conductors were said to be walking in the gauge, and moved to the tie ends when hearing and seeing 66. Properly-conducted testimony would have revealed this as part of the interviews for “the NTSB record” more directly, and perhaps it could have been asked of Sahara either directly or as a ‘callback’.
It does change the picture of what to do … slightly … if she saw them move over and thought they were about to go further. Still doesn’t change putting it in emergency some time … even a few milliseconds … before the actual impact.
As noted, we need an emergency code, to be learned by heart along with the other horn signals, for “emergency look around you” rather than just “emergency”. That would consist of enough ‘dead time’ between signals to allow any echoes to die to the point other horns sounding emergency could be heard and distinguished.
Such is my non-operator’s opinion, as learned in part from wiser people with running experience.
We already have that emergency whistle signal. It is exactly as what Joe said he would have done. This is part of every reference to train whistle/horn signals, going way back in history. I assume it was chosen precisely for the reasons you suggest. Leave some gaps of no tone for a better chance of hearing a tone from another whistle. The gaps also tend to give a punch feeling of urgency which is a fundamental part of emergencies. Note that it is exactly opposite of the “laying on the horn” described by the engineer of 175, and was possibly also done by the engineer of 66. That signal readily makes two horns sound like one.
Along with making a punchy horn signal, an engineer should watch for clues that a person in harms way may not realize that fact. This should have been apparent to the engineer of 175. From her viewpoint, the action of the two victims either meant they were not aware of 175 because they were distracted by 66, or they were playing chicken. Then one would need to ask themselves how often do men in green safety vests play chicken?
Particularly the fact that the two men moved away from the path of 66, but remained in the path of 175, should have made it clear that they were unware of 175 despite the laying on of the horn. They were unaware of 175 because they thought they were hearing only the horn of 66.
But the ‘new thing’ that we have to consider adding is the effect of directional echo on the emergency signal. For that, we need longer, or differently-spaced, gaps of silence between the ‘blasts’ – and to me, this means that the shorter blast clusters need to be decided as something unambiguously meaning not just “proximity danger” but “look around you danger”.
This might be as simple as the MTA’s long-short repeated three times … then delayed for a second, the whole thing being sent with as distinctive a ‘hand’ on the button as possible. Just as wireless telegraphers could recognize two signals as different, this optimizes the different ways a human ear in a multipath reflection field might discriminate two signals essentially trying to send the same priority message in the same few seconds.
I would note that a considerable amount of work on high-criticality alerts was done in ITU R10 and other working groups. One example was the stall alert in heavy aircraft, which instead of blaring in your ear calls your attention (calmly, to non-pilots, considering the importance!) speech-synthesizing something directed toward relieving the dangerous condition, or alerting to do something (pull up) or what’s becoming a potential problem (terrain). It would be delightful if we could provide a PA function via some terrifying auxetophone that would allow voice alerts at substantial distance … but I’d certainly settle in the short run for unambiguous alerting.
Euclid
We already have that emergency whistle signal. It is exactly as what Joe said he would have done. This is part of every reference to train whistle/horn signals, going way back in history.
But the ‘new thing’ that we have to consider adding is the effect of directional echo on the emergency signal. For that, we need longer, or differently-spaced, gaps of silence between the ‘blasts’ – and to me, this means that the shorter blast clusters need to be decided as something unambiguously meaning not just “proximity danger” but “look around you danger”.
This might be as simple as the MTA’s long-short repeated three times … then delayed for a second, the whole thing being sent with as distinctive a ‘hand’ on the button as possible. Just as wireless telegraphers could recognize two signals as different, this optimizes the different ways a human ear in a multipath reflection field might discriminate two signals essentially trying to send the same priority message in the same few seconds.
I would note that a considerable amount of work on high-criticality alerts was done in ITU R10 and other working groups. One example was the stall alert in heavy aircraft, which instead of blaring in your ear calls your attention (calmly, to non-pilots, considering the importance!) speech-synthesizing something directed toward relieving the dangerous condition, or alerting to do something (pull up) or what’s becoming a potential problem (terrain). It would be delightful if we could provide a PA function via some terrifying auxetophone that would allow voice alerts at substantial distance … but I’d certainly settle in the short run for unambiguous alerting.