Overmod My opinion of the recent NTSB has changed fairly dramatically since release of the final report in the Amtrak Chester accident a couple of months ago. If you want to see hackery on a grand scale and you understand the actual situation leading up to that tragedy ... read it with aspirin and timolol present; you'll probably need both by the time you're done.
My opinion of the recent NTSB has changed fairly dramatically since release of the final report in the Amtrak Chester accident a couple of months ago. If you want to see hackery on a grand scale and you understand the actual situation leading up to that tragedy ... read it with aspirin and timolol present; you'll probably need both by the time you're done.
I did read it, and didn't note any particular problems with it. I would appreciate it if you would start a new topic and make some appropriate comments on the matter. I would then re-read it with those in mind.
Ed
LithoniaOperator Whatever brake application he made (or not) seems pretty academic at this point, to me. He crashed because he was lost. If he made a speed reduction from 81 to 79 while thinking he was in 79-mph territory, that's irrevant to the crash. If he went into (Oh S#!t) emergency seconds before the crash, that's pretty much trivia in the grand scheme of things at this point. I don't get the fascination with this minutiae. The poor man has admitted to being lost. He flew into a tight curve at about 2.5 times the speed limit, and the engine wound up on the Interstate. What's the mystery?
Whatever brake application he made (or not) seems pretty academic at this point, to me.
He crashed because he was lost. If he made a speed reduction from 81 to 79 while thinking he was in 79-mph territory, that's irrevant to the crash. If he went into (Oh S#!t) emergency seconds before the crash, that's pretty much trivia in the grand scheme of things at this point.
I don't get the fascination with this minutiae. The poor man has admitted to being lost. He flew into a tight curve at about 2.5 times the speed limit, and the engine wound up on the Interstate. What's the mystery?
Also, depending on what the NTSB updates mean, he was possibly 700-1000 feet from the curve when he first saw it and realized where he was. Quick thinking would have had him make an emergency application right at that instant. That may have prevented the derailment. And even if not, it may have prevented the fatalities and many injuries.
Yet, the NTSB might be interpreted as saying the engineer made an independent brake application. So the mystery is why he would have made the least effective brake application in the face of the greatest need to slow down.
Another mystery is why the NTSB is having so much trouble making clear statements.
<greatly snipped only for brevity>
It would be regrettably in character for the ineffable Bella 'PTC would have prevented this' Dinh-Zarr to semantically "mistake" applying the brake lever in the locomotive for applying just the locomotive's physical brakes. If we had not already known about the preliminary EDR telemetry and the six-second-prior 'overspeed comment' with no sign of panic or concern (I will be watching to see if proper voice stress analysis gets done, absence of which might point to some gentle attempt at whitewashing...) I'd be more inclined to that specific interpretation of nonspecificity.
LithoniaOperator What's the mystery?
What's the mystery?
Why?
Most humans would grasp that, knowing there was a sharp curve ahead, it is of great importance not to come in too fast. Especially if you are in the part that hits first.
Perhaps he had no idea that such a curve could be encountered. But, as a professional with years of experience, he knew that he should know. And, as a professional, he took charge of the locomotive.
Perhaps he thought he'd already passed the curve. Obviously highly unlikely.
So, he knew the curve was upcoming. And that it was extremely important to "get it right". All that incentive, and yet.....
The mystery: in what particular way(s) did he screw up
He crashed because he was lost. If he made a speed reduction from 81 to 79 while thinking he was in 79-mph territory, that's irrelevant to the crash. If he went into (Oh S#!t) emergency seconds before the crash, that's pretty much trivia in the grand scheme of things at this point.
Still in training.
Euclid Overmod Euclid I conclude that the engineer never made an application of the independent brake during the approach to the curve. How you could 'conclude' anything on the basis of two conflicting NTSB reports is known only to God. The application of the "independent" brake was supposedly* captured from the locomotive event-data recorder before its oh-so-convenient shutdown and is not a subject for selective-memory reality denial because it doesn't fit later nontechnical reports. Well, I can always change my conclusion. But I am not going to wait for years to draw conclusions. So I will draw them as the NTSB dribbles out the information. And what has dribbled out says absolutely nothing about making an application of the "independent brake." Then NTSB said this: "The engineer’s actions were consistent with the application of the locomotive’s brakes just before the recording ended" I believe they are merely saying that the engineer appeared to apply the train’s brakes by manipulating the necessary brake control at the locomotive control stand, thus the term "locomotive's brakes." I do not believe they are saying that the engineer was applying brakes only to the locomotive which would be properly called an independent brake application.
Overmod Euclid I conclude that the engineer never made an application of the independent brake during the approach to the curve. How you could 'conclude' anything on the basis of two conflicting NTSB reports is known only to God. The application of the "independent" brake was supposedly* captured from the locomotive event-data recorder before its oh-so-convenient shutdown and is not a subject for selective-memory reality denial because it doesn't fit later nontechnical reports.
Euclid I conclude that the engineer never made an application of the independent brake during the approach to the curve.
I conclude that the engineer never made an application of the independent brake during the approach to the curve.
How you could 'conclude' anything on the basis of two conflicting NTSB reports is known only to God. The application of the "independent" brake was supposedly* captured from the locomotive event-data recorder before its oh-so-convenient shutdown and is not a subject for selective-memory reality denial because it doesn't fit later nontechnical reports.
Well, I can always change my conclusion. But I am not going to wait for years to draw conclusions. So I will draw them as the NTSB dribbles out the information. And what has dribbled out says absolutely nothing about making an application of the "independent brake."
Then NTSB said this:
"The engineer’s actions were consistent with the application of the locomotive’s brakes just before the recording ended"
I believe they are merely saying that the engineer appeared to apply the train’s brakes by manipulating the necessary brake control at the locomotive control stand, thus the term "locomotive's brakes." I do not believe they are saying that the engineer was applying brakes only to the locomotive which would be properly called an independent brake application.
A lack of specificity by the NTSB that creates more questions than it answers.
Never too old to have a happy childhood!
How you could 'conclude' anything on the basis of two conflicting NTSB reports is known only to God. The application of the "independent" brake was supposedly* captured from the locomotive event-data recorder before its oh-so-convenient shutdown and is not a subject for selective-memory reality denial because it doesn't fit later nontechnical reports. I am still waiting for ANY definitive proof of an intended automatic-brake application from 'sight distance' of the curve speed-restriction time up to the UDE as the train separated. I strongly recommend that y'all do, too.
*I have not seen the actual telemetry, or transcripts of depositions. But then, neither has anyone else posting in this thread. And it is fun to watch the supposition that an engineer trained as this one was would instinctively know the correct use of blended brake for the specific purpose of correcting for a locomotive overspeed warning.
I cannot get to my copy of the ACS64 manual, but if I recall correctly the blended brake does not immediately engage an automatic-brake reduction. By analogy the brakes on Siemens high-speed diesels may be similar. Someone with access to either an ACS64 or SC44 publication or even a training video might comment gainfully on this point.
oltmanndI got nothin'. Given my misunderstanding of the timing, I have no idea what the engineer was doing with the independent.
rdamon This was a Talgo set sharing a single axle between cars. Isn't the slack greatly reduced?
This was a Talgo set sharing a single axle between cars. Isn't the slack greatly reduced?
Yes. In general, passenger trains have much less free slack per coupler since type H couplers ("tightlock") mate more closely than freight couplers. Articulated joints have near zero free slack.
In addition to the free slack in couplers there is slack in the draft gear. It is basically a spring that attaches the coupler to the carbody (in practice, rubber pads in a chevron arrangment). There is no draft gear involved in articulated joints, though, so Talgo trains would be nearly slack-free.
-Don (Random stuff, mostly about trains - what else? http://blerfblog.blogspot.com/)
243129 oltmannd VOLKER LANDWEHR oltmannd Cause that's how you do it! Bunch them up gently with the independent, then dynamic brake. You have to get the little bit of slack bunched and get draft gear in compression. On page 19 we already had a short discussion of train handling. The following was stated: 243129 said: As I have stated previously use of the independent brake to reduce speed, no matter the increment, on a 'live' passenger train is indicative of poor train handling and lack of or poor training. Somehow I'm lost.Regards, Volker Back in the day, you'd stretch a passenger train, throttle open, automatic applied. Very smooth way to handle a train. Burns lots of fuel, though. Most passenger locomotives since the 1980s have blended braking, where an automatic application would engage the dynamic, as well. Not nearly as smooth, but effective. If you want to do straight dynamic to slow a train, you have to bunch the slack first. Smoothest way is with a bit of independent. There is much less slack in a passenger train than a freight train, but the principle is the same. Once again you do not, I repeat, do not use the independent brake only to slow a live passenger train. The blended brake is to be used to suppress any penalties. Please show me the information to support your assertion.
oltmannd VOLKER LANDWEHR oltmannd Cause that's how you do it! Bunch them up gently with the independent, then dynamic brake. You have to get the little bit of slack bunched and get draft gear in compression. On page 19 we already had a short discussion of train handling. The following was stated: 243129 said: As I have stated previously use of the independent brake to reduce speed, no matter the increment, on a 'live' passenger train is indicative of poor train handling and lack of or poor training. Somehow I'm lost.Regards, Volker Back in the day, you'd stretch a passenger train, throttle open, automatic applied. Very smooth way to handle a train. Burns lots of fuel, though. Most passenger locomotives since the 1980s have blended braking, where an automatic application would engage the dynamic, as well. Not nearly as smooth, but effective. If you want to do straight dynamic to slow a train, you have to bunch the slack first. Smoothest way is with a bit of independent. There is much less slack in a passenger train than a freight train, but the principle is the same.
VOLKER LANDWEHR oltmannd Cause that's how you do it! Bunch them up gently with the independent, then dynamic brake. You have to get the little bit of slack bunched and get draft gear in compression. On page 19 we already had a short discussion of train handling. The following was stated: 243129 said: As I have stated previously use of the independent brake to reduce speed, no matter the increment, on a 'live' passenger train is indicative of poor train handling and lack of or poor training. Somehow I'm lost.Regards, Volker
oltmannd Cause that's how you do it! Bunch them up gently with the independent, then dynamic brake. You have to get the little bit of slack bunched and get draft gear in compression.
On page 19 we already had a short discussion of train handling. The following was stated:
243129 said: As I have stated previously use of the independent brake to reduce speed, no matter the increment, on a 'live' passenger train is indicative of poor train handling and lack of or poor training.
Somehow I'm lost.Regards, Volker
Back in the day, you'd stretch a passenger train, throttle open, automatic applied. Very smooth way to handle a train. Burns lots of fuel, though. Most passenger locomotives since the 1980s have blended braking, where an automatic application would engage the dynamic, as well. Not nearly as smooth, but effective.
If you want to do straight dynamic to slow a train, you have to bunch the slack first. Smoothest way is with a bit of independent. There is much less slack in a passenger train than a freight train, but the principle is the same.
Once again you do not, I repeat, do not use the independent brake only to slow a live passenger train. The blended brake is to be used to suppress any penalties. Please show me the information to support your assertion.
I got nothin'. Given my misunderstanding of the timing, I have no idea what the engineer was doing with the independent. My experience with passenger power is rather limited (and old).
LithoniaOperator Could someone explain to me the terms “penalty” and “penalty application,” related to braking? Thanks.
Could someone explain to me the terms “penalty” and “penalty application,” related to braking?
Thanks.
Full service brake application to zero brake pipe pressure initiated by train control equipment such as cab signal, alerter, ATC equipment.
Since it's not an emergency brake application, it occurs more slowly and with less force and is therefore a bit less risky. It's also a bit faster to recover from.
Murphy Siding 243129 oltmannd VOLKER LANDWEHR oltmannd Cause that's how you do it! Bunch them up gently with the independent, then dynamic brake. You have to get the little bit of slack bunched and get draft gear in compression. On page 19 we already had a short discussion of train handling. The following was stated: 243129 said: As I have stated previously use of the independent brake to reduce speed, no matter the increment, on a 'live' passenger train is indicative of poor train handling and lack of or poor training. Somehow I'm lost.Regards, Volker Back in the day, you'd stretch a passenger train, throttle open, automatic applied. Very smooth way to handle a train. Burns lots of fuel, though. Most passenger locomotives since the 1980s have blended braking, where an automatic application would engage the dynamic, as well. Not nearly as smooth, but effective. If you want to do straight dynamic to slow a train, you have to bunch the slack first. Smoothest way is with a bit of independent. There is much less slack in a passenger train than a freight train, but the principle is the same. Once again you do not, I repeat, do not use the independent brake only to slow a live passenger train. The blended brake is to be used to suppress any penalties. Please show me the information to support your assertion. 243129- You speak as if you are a retired engineer perhaps? Were you ever involved in training new engineers and conductors?
243129- You speak as if you are a retired engineer perhaps? Were you ever involved in training new engineers and conductors?
BaltACDReading of the 2nd link brings the operation of the device into closer focus.
I understand in general how the system works. I haven't found anything that shows what Overmod stated. In wanswheel's second link I found the following on page 1011:
The ramps, when not electrified, causes an application of the air brakes; when electrified it energizes an electro-magnet on the engine, which prevents the operation of the brake-applying apperatus.
There is no visual or audible signal, and no speed recorder.
The last sentence somehow doesn't fit with Overmod's statement.
I have followed the "links" in the linked article:
https://archive.org/stream/railwayage53newy#page/372/mode/2up
In the links I didn't find information about any advisory action of the Miller system, but they are from the same source.Regards, Volker
One of the Wisconsin Talgo sets has apparently been pulled out of Beech Grove, and the rumor is that it will end up out here. We shall see.
Was not a steady clicking like a clock, or even an audible odometer indicating distance run -- I assumed it was intermittent relay action at 'control points'. The point is that the engineer could associate the continued clicking with assurance of safe passage, a bit like the B&O guy (also mentioned in the Westinghouse book) who said of Patenall's CPLs "any time I see a green I go like hell'. That is different from the dread in an adversarial system that clicks only when about to inflict sanctions for presumed lapses of attention.
VOLKER LANDWEHRThank you for the links. The first link isn't easy to understand without the according drawings but I'll try. I have especially problems with what Overmod said: There was a steady clicking audible feedback, which to the engineer meant 'all clear'; any action the system took in early stages of an "alert" was advisory, to concentrate attention on specific things about to happen and giving adequate time for a proper professional response.Regards, Volker
The first link isn't easy to understand without the according drawings but I'll try.
I have especially problems with what Overmod said: There was a steady clicking audible feedback, which to the engineer meant 'all clear'; any action the system took in early stages of an "alert" was advisory, to concentrate attention on specific things about to happen and giving adequate time for a proper professional response.Regards, Volker
Reading of the 2nd link brings the operation of the device into closer focus.
Thank you for the links.
Miller automatic train stop
http://digital.hagley.org/2317_50_004#page/1/mode/2up
https://archive.org/stream/railwayage57newy#page/1010/mode/2up
Most of what I knew about the Miller system was from contemporary printed primary and secondary sources when I was researching my thesis. There is one book, Brignano (et al.) "The Search for Safety", which was published for WABCO, that contains a somewhat elementary account.
You can have very effective safety or response braking without having to include the penalty aspects. Few people like to have often-gratuitous punishment even after their life might have been spared...
OvermodPart of the idea was that the train control assisted vigilant lookout, instead of merely alarming and threatening to 'down brakes' after a signal had already been passed at danger and an unsafe condition existed. There was a steady clicking audible feedback, which to the engineer meant 'all clear'; any action the system took in early stages of an "alert" was advisory, to concentrate attention on specific things about to happen and giving adequate time for a proper professional response.
I only found what I described above. Your quoted part is new to me. Can you provide a link, please?
I think engineers should be glad about a penalty brake application because otherwise in worst case they might die.Regards, Volker
The difference is not in the technology, it is how it was presented to the engineer.
Renember this is well before the USRA and the Esch Act mandated automatic train control that was expediently made part of the quid pro quo that would return railroads to private administration. So it had to be sold to stingy management, not mandated as what-is-not-mandatory-is-forbidden military-style "discipline". The Miller people chose, rightly or wrongly, to spin the IxD of their 'train control' to appeal specifically to the BLE and other labor groups, with at least the tacit idea that labor would join the public in asking for (and perhaps demanding) even better safety than the block-signal systems were providing.
Part of the idea was that the train control assisted vigilant lookout, instead of merely alarming and threatening to 'down brakes' after a signal had already been passed at danger and an unsafe condition existed. There was a steady clicking audible feedback, which to the engineer meant 'all clear'; any action the system took in early stages of an "alert" was advisory, to concentrate attention on specific things about to happen and giving adequate time for a proper professional response.
Contrast this with monkeylike yanking on forestallers or rat-pushing-levers-in-a-psych-experiment response to rising alerted beeping. I think it will be clear to you why the Miller system would be preferred.
Now in case of actual 'emergency' (in the original English sense of emergent situation with immediate danger) a Miller system might sound a dramatic alarm, slam on the brakes, or take any other appropriate 'safety' action faster than human reaction time and interlocked to prioritize events to give the shortest stop. What would be absent is any aspect of blame or shame on the engineer for making the safety nanny system have to do what his job should have been if only he were perfectly reliable.
I stand by the idea that helping an engineer run safely rather than looking for opportunities to bark his knuckles is a better paradigm for autonomic safety-augmenting systems.
OvermodIn practice, I disagree with the whole 'penalty' idea (along with the idea that ATC is a defense against prejudged worker incompetence and negligence). A good example of a 'right' metaphor for an ATC system was the Miller system, which was being promoted just before the American entry into WW1; this provided 'assurance' feedback to the engineer instead of castigating him after the fact.
I haven't found much information online. From what I found the Miller train stop was a mechanical system that let a train brake when a stop signal was run. After the shoe was off the ramp the engineer regained full control of the brakes.
Is that right so far?
On typical inductive system the engineer gets alerted. If he doesn't react on the alarm in a given timeframe there is a penalty brake application.
With the Miller system the engineer is alerted by the brake application on othe ATS systems he is alerted by the signal tone.
I don't see why the Miller system should be better. They are safety systems and shall prevent accidents caused by ignoring a signal aspect. With the Miller system the engineer could go on still ignoring a red signal. The systems are not for crew comfort, but back-ups for human failures.
But perhaps I understood something wrong with my limited information on the Miller system.Regards, Volker
Thanks to Chris / CopCarSS for my avatar.
Thanks, Overmod.
Overmod The timeline originally provided by the NTSB clearly (to me) indicated that the whole business of 'overspeed' recognition and independent-brake correction was finished at least six seconds earlier (go back and check the exact numbers). That can not possibly be a brake application made after seeing that tiny 30mph restriction sign: how sharp would your eyes have to be to make this out 800 feet away with the corresponding angular resolution -- and then there would be the issue of reaction time in conditions of surprise, another 90 feet minimum at that speed. At which point maybe he was on the brake handle, but the UDE was already dumping the pressure more quickly and to a lower value than his valve motion could command. We will need to see results of the camera video to resolve when, and really if at this point, he actually moved the brake controller as he indicated. Note that this specific detail is important enough that a "camera failure" excuse for it not being seen would set off my BS detectors.
The timeline originally provided by the NTSB clearly (to me) indicated that the whole business of 'overspeed' recognition and independent-brake correction was finished at least six seconds earlier (go back and check the exact numbers). That can not possibly be a brake application made after seeing that tiny 30mph restriction sign: how sharp would your eyes have to be to make this out 800 feet away with the corresponding angular resolution -- and then there would be the issue of reaction time in conditions of surprise, another 90 feet minimum at that speed. At which point maybe he was on the brake handle, but the UDE was already dumping the pressure more quickly and to a lower value than his valve motion could command.
We will need to see results of the camera video to resolve when, and really if at this point, he actually moved the brake controller as he indicated. Note that this specific detail is important enough that a "camera failure" excuse for it not being seen would set off my BS detectors.
[img=http://ledlighting-solutions.com/mm5/graphics/00000001/red%20stop%20sign%20lights-white_540x570.jpg] Better signage for such a hazardous curve situation for mandatory speed reduction would be in order. Perhaps an led lighted sign powered by a solar panel. Here's a link to what China can offer: https://www.alibaba.com/product-detail/Long-life-span-Slow-Down-Warning_60510749645.html My question would be was one run through enough to train an engineer on a new section? Tell me again, Amtrak does, or does not use video that monitors the cab?
LithoniaOperatorCould someone explain to me the terms “penalty” and “penalty application,” related to braking?
They relate to a particular implementation of speed control, and particularly to systems of 'automatic train stop' (as opposed to automatic train control, which sometimes does more).
The principle is that if a 'stop' signal is passed, the system imposes a 'penalty' on the operator: it causes a full brake application, right down to a stop, and then runs a timer that prevents the brake application from being released for a period of time (e.g. 30 seconds). Some systems enhanced the 'penalty' by requiring someone to get down off the engine and reset something that couldn't be reached while the train was moving.
There were arguments in the infancy of automatic train control about just how strong a brake application the system should make -- emergency application having substantial risk of damage or derailment over heavy service. A 'penalty application' would also have to serve to stop the train 'in time' in a real emergency, so it need not be limited to conservative service values.
In practice, I disagree with the whole 'penalty' idea (along with the idea that ATC is a defense against prejudged worker incompetence and negligence). A good example of a 'right' metaphor for an ATC system was the Miller system, which was being promoted just before the American entry into WW1; this provided 'assurance' feedback to the engineer instead of castigating him after the fact.
I agree, been doing this stuff since June 1967 and was always taught dont touch the indenpendent at track speed in fact dont use it at speeds above 15 MPH, you DONT need to bunch the slack with the ind. brake prior to dynamic braking, dynamic brake "set up" does that for you.
Our community is FREE to join. To participate you must either login or register for an account.