Acela derailment

Amtrak Unlimited Discussion Forum

Help Support Amtrak Unlimited Discussion Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Status
Not open for further replies.
As I was saying, if it's possible for an electronic system to know whether the switch and frog are in the same alignment, an informational warning that they were not should have avoided a situation like this. It's a layer of redundancy that simply seems to be absent from the current system.
There was an informational warning that they were not. It was the danger signal that the train was allowed to roll past.
 
We are fast reaching, if not already arrived at the point of going around in circles on this subject.

There is an expression, "Striving for perfection driving out the good."

I think what went wrong to the best of our knowledge, and how the safeguards did work as intended hs been fairly thoroughly explained.

If you want to see where people live that never make any mistakes, visit a cemetary.
 
As I was saying, if it's possible for an electronic system to know whether the switch and frog are in the same alignment, an informational warning that they were not should have avoided a situation like this. It's a layer of redundancy that simply seems to be absent from the current system.
Let's try this again.

1) There is such a system in place on that switch.

2) Something related to the control of that switch failed.

3) When that failure occurred, the system shut down the railroad with red signals. The system is already saying, "I have a problem. Something is not right here."

4) A combination of 3 human beings overrode that fail safe with the rules of manual operation and two of them blew it. They did not follow the correct procedures and derailed the train.

5) Because of that failure on the switch, the system that would have detected the misaligned frog is being overridden. The train cannot move without the override being on, as the onboard computers are saying "you have a red signal; you can't move." So even if the switch was seeing the misaligned frog, assuming that wasn't the point of the original failure in the first place, it could no longer communicate that the frog was misaligned due to the override. Actually to be more correct, if the frog or the electronics weren't what failed originally, then the switch would have been conveying that the frog was misaligned. But with the engine's computer on override, it's already ignoring the Danger signal that the switch would send for a misaligned frog.

6) This is why a derailment at 150 MPH is not possible. Under normal circumstance the switch will indeed detect a misaligned frog and it will turn every signal red. If man chooses to overrride that warning, then man and man alone must ensure that they have placed everything in the right order. In this case two men failed to ensure that everything was correct!
 
Last edited by a moderator:
If you think that manual inspection is somehow an automated system, then you think you're right. I suppose that you want us to go back to hanging out train orders with hoops, too.
Exactly!!! Now you are finally starting to get close to understanding the reality. :) Degraded operations including manual mode is part of the design of the failure chain in any automation system. So yes, when all else fails it is "train orders on hoops" time. :)

A normally operating system would display the setting of a switch accurately reflecting its current state on the CETC model displays. When the switch system fails the switch is shown as out of service on the model display, and the signal protecting the switch automatically turns to Stop and remains there until the switch communicates that it is back in order. When the CETC detected the failure they sent a MoW staff to spike the switch and get trains through manually, while the automation system is diagnosed and fixed. That is what the MoW staff was doing there. The signaling system was still operating enforcing "Restricting" speed, and enforcing Stop at the Davis Home Signal.

The reason that Rule 241 had to be invoked is because normally a Home Signal cannot be passed at danger under any circumstances, and the train's control computer would not let it through absent the release message under the rule. A Block Signal OTOH can be passed at Restricting speed normally after a full stop at it, except in a few situations where it can be passed at Restricting speed.

When a system is in a state of degraded or failed automation then indeed, notionally you hang "train orders on hoops" - virtual hoops, that is - you use a radio to read and copy a Form D or a Rule invocation or such, to the engineer/conductor and/or the train's control computer. The radio message mentioned in the last para is the equivalent of a "train order on a hoop" in this situation. No amount of insisting that all aspects of operation under all circumstances of failure should be covered fully by automation is going to make it so.

The system worked exactly as it was supposed to. Through best effort exercise of degraded automation it minimized damaged in face of failure to properly exercise manual protection, mostly by humans in the process of working around automatic protections without apparently being sufficiently attentive to details most likely due to omissions rather than commission.

For those that are interested in actually learning about such things instead of being armchair engineers proffering opinions based on lack of knowledge, I'd recommend starting with the Wikipedia article on Railway Signaling. Then you can read about the PRR Pulse Code Cab Signaling system which is the CTC/CSS system in use on the NEC. After that you can read the Wiki page on ACSES which is a good introduction to the ACSES overlay on CTC/CSS used on the NEC under which the train was operating at Davis interlocking. After that if you are even more interested I will be happy to share with you technical papers on the design of the original PRR CTC/CSS and then the ACSES overlay on it, so that we can have a meaningful discussion based on knowledge and facts, the ones that a few of us have been trying to patiently present to apparently a few quite non-receptive in the audience. Of course be prepared to exercise your brains to quite an extent since all this is not easy stuff.
 
Last edited by a moderator:
In the world of reality, even on high speed tracks, things can go wrong. One safety may I worked with a few years ago used the term "graceful degredation" meaning that if things went wrong, it would happen step by step in such a manner that there negative results would be minimuzed. This concept is the opposite of the events in the German derailment at Eschede, where the "cascade of events" to disaster consisted of a number of successive events. Many of these events were due to factors that do not exist on any high speed tracks or equipment here, and for that matter were not common in Germany, either, and the equipment related issues, at least, are no longer used.

In the real world of FRA regulation and oversight of railroads in this country, to eliminate the possibility of collision with high speed trains by equipment runaways or mishandling on other tracks, the FRA requires derails to be installed on tracks leading into track where the speed limit is above 110 mph for passenger trains. The following is quoted from the Code of Federal Regulations, Title 49 - Transportation. Subtitle B - Other Regulations Relating to Transportation (Continued). CHAPTER II - FEDERAL RAILROAD ADMINISTRATION, DEPARTMENT OF TRANSPORTATION. PART 213 - TRACK SAFETY STANDARDS, Subpart G - Train Operations at Track Classes 6 and Higher. (Class 6 means tracks having a speed limit of above 90 mph.)

§ 213.357Derails.(a) Each track, other than a main track, which connects with a Class 7, 8 or 9 main track shall be equipped with a functioning derail of the correct size and type, unless railroad equipment on the track, because of grade characteristics cannot move to foul the main track.

(b) For the purposes of this section, a derail is a device which will physically stop or divert movement of railroad rolling stock or other railroad on-track equipment past the location of the device.

© Each derail shall be clearly visible. When in a locked position, a derail shall be free of any lost motion which would prevent it from performing its intended function.

(d) Each derail shall be maintained to function as intended.

(e) Each derail shall be properly installed for the rail to which it is applied.

(f) If a track protected by a derail is occupied by standing railroad rolling stock, the derail shall be in derailing position.

(g) Each derail on a track which is connected to a Class 7, 8 or 9 main track shall be interconnected with the signal system.
This is saying that reality says that in case of screw up, it is better to put stuff on the ground than to let it get on high speed tracks. Presence of derails would have prevented the collision at Landover Chase a few years back. There were numerous other safety devices in place at that location, but all were either ignored or bypassed, and, if I recall correctly, the engineer was enjoying some grass at the time.

edit: See JIS's post below

The additional requirements of Subpart G for speeds above 90 mph are a major part of the reason that the various railroad companies are less than enthused about running passenger trains at speeds above 90 mph.
 
Last edited by a moderator:
Presence of derails would have prevented the collision at Landover a few years back. There were numerous other safety devices in place at that location, but all were either ignored or bypassed, and, if I recall correctly, the engineer was enjoying some grass at the time.
Which accident at Landover do you refer to?

The famous case of grass consumption was by Ricky Gates, a Conrail engineer, and it was at Chase Maryland, right next to the Gunpowder River bridge at Gunpow interlocking. Gates SPAD-ed the home signal at Gunpow and split the trailing switch to place his engines in path of the Colonial running in at 125mph. Amtrak had just finished removing the derails there a few year before that. That interlocking still does not have derails on the converging tracks mostly used by freight from Bayview Yard. That is because once you declare a track to be a main you don't need to do derails on them. Finally with ACSES installed it will actually become impossible absent someone bypassing it. Currently it is less likely to happen than it was back then, because now all equipment allowed on the NEC will force a penalty brake application. The Conrail units just had a warning whistle which was blocked because it was disturbing the peace of the peace pipe smokers, and it did not have the penalty brake application feature of the CSS apparently.
 
Last edited by a moderator:
Presence of derails would have prevented the collision at Landover a few years back. There were numerous other safety devices in place at that location, but all were either ignored or bypassed, and, if I recall correctly, the engineer was enjoying some grass at the time.
Which accident at Landover do you refer to?

The famous case of grass consumption was by Ricky Gates, a Conrail engineer, and it was at Chase Maryland, right next to the Gunpowder River bridge at Gunpow interlocking. Gates SPAD-ed the home signal at Gunpow and split the trailing switch to place his engines in path of the Colonial running in at 125mph. Amtrak had just finished removing the derails there a few year before that. That interlocking still does not have derails on the converging tracks mostly used by freight from Bayview Yard. That is because once you declare a track to be a main you don't need to do derails on them. Finally with ACSES installed it will actually become impossible absent someone bypassing it. Currently it is less likely to happen than it was back then, because now all equipment allowed on the NEC will force a penalty brake application. The Conrail units just had a warning whistle which was blocked because it was disturbing the peace of the peace pipe smokers, and it did not have the penalty brake application feature of the CSS apparently.
Oooops. This is what I get working from memory instead of checking up on things. Yes, I was thinking about Chase, and yes, I got some things worong. I had heard, obviously wrongly that the derails had been reinstalled.
 
No problem. Happens to the best of us from time to time.

If I recall my track numbers right at Gunpow, the two original high speed tracks were the two northmost tracks, which are numbered 2 and 3. 1, the next track voer to the south of 2 is now also a high speed track which merges into 2 at Gunpow and has no derails being a high speed main with a high speed merge switch. The next track over to the south is I think designated 0 or A or something like that and is also designated a main now hence no derail. It merges into 1. As I said my track numbering may be off by one.

Bringing it back to the Acela.....

The set that derailed was 2029/2026.

The lead unit had its trucks replaced at the Boston Southampton Street HSR Maintenance Facility on April 4th, and is back in service.
 
Last edited by a moderator:
Status
Not open for further replies.
Back
Top