Amtrak Wreck in Philadelphia

[jslader_sr]

WDGF

 

 

tree68

The event recorder will tell all...

 

 

Just out of my own ignorance and curiosity, does anyone know if the event recorder includes accelerometer and/or some form of GPS in its data? If so, I'm thinking it will answer the questions about when & where the train accelerated, and if that was normal or abnormal.

 

A log of the final 65 seconds of acceleration before the crash has already been released :

65 seconds prior to crash: 70mph

43 seconds: 80mph

31  seconds: 90 mph

16 seconds: 100mph

Using that info alone, they would  know where the train was when it started to accelerate. Additionally, the train also has GPS capabilities. In fact, the Amtrak tracker website showed the train @ 106mph as it entered the curve, something those of us who saw it didn't believe at first. That website is for allowing passengers and/or loved ones the ability to follow the progress of a train.

 The speed limit in the section of tracks 188 was traveling is 80mph on the inner tracks/70mph on the outer, which are normally used by SEPTA. Sometimes Amtrak trains overtake one another, but I don't think that was the case here, so he'd have been subject to the 80mph limit. So, up to 43 secs prior to the crash he was going the speed limit.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

[erikem]

Most informative post yet - thanks!

- Erik

[Euclid]

Jslader_sr,

Thanks for that information, but how do we know that 65 seconds before the crash, the train started to accelerate, as you say?  Is there a source for that information? 

All that we know is that the train was going 70 mph 65 seconds before the crash.

[jslader_sr]

Apparently before that 65 second mark nothing was out of the ordinary and the train was proceeding at track speed. At 65 seconds prior to the crash the train began to accelerate, although again the speed limit is 80mph, so that in itself is not out of the ordinary.

[Euclid]

All of the news reports imply that the train suddenly accelerated during the last 65 seconds.  The word, “suddenly” means that there was no acceleration occurring prior to that time.  So if it suddenly accelerated, that means that train was running at a constant speed, and the engineer advanced the throttle at 65 seconds prior to the crash.  I don’t see anything that would justify or even explain that action except for the engineer not realizing his location.

However, some person with knowledge of the railroad had been reported in the news to say that the standard procedure was to open the throttle to full power after leaving the last station, and accelerating all the way to the curve speed restriction.  Then, upon reaching that point, engineers would reduce the throttle to a level appropriate for the 50 mph speed restriction of the curve.  If that is true, then the train would indeed have accelerated during the last 65 seconds.  But it would have also been accelerating for a minute or so prior to that point.  So the 65 seconds is just one segment of a larger acceleration event.    

But the difference in the two scenarios is that with the first scenario, the engineer advanced the throttle at 65 seconds prior to the curve.  With the second scenario, the engineer failed to reduce the wide open throttle at 65 seconds prior to the curve.  The first scenario is a matter of action taken, while the second scenario is a matter of action not taken.  Both scenarios point to different causes.   

To clarify this, we would need additional speed reading prior to the 65 second mark.  Without any clarification on this point, I assume that the first scenario is correct, and the person quoted in the news advancing the second scenario is wrong. 

[Electroliner 1935]

From my reading of the engineers schedule, he would have gone on duty at NYP at about 2:30 pm to take Acela run #2163 Dp at 3:00 pm and scheduled into WAS at 5:53 pm. But he had signal and/or train equipment issues which delay him by 26 minuites making his arrival time 6:19. Train 188's departure was 7:10 pm which is enough time to use a restroom, grab something to go and board motor 602. I doubt he was excessively fatigued, but having viewed cab videos, I also doubt that road hypnosis was in play. I have heard a rumor that he had not been on this route for long. He had obviously passed a qualifying run but one of the speculations I have heard was that he might have lost his concept of where he was and thought he was past Franklin Jct. and was going to the track speed that would have been in effect after Franklin Jct. I think we need to wait and see what the results of the investigation are and hope that his memory returns. 

[Norm48327]

Electroliner said: "I think we need to wait and see what the results of the investigation are and hope that his memory returns. "

Why wait? Bucky seems to have it all figured out.

[rdgk1se3019]

Investigators believe that the engineer was using his cell phone at the time of the wreck.

[seppburgh2]

On Friday there was discussion of crimal charges being considered.  While his cell phone has been secured, and by now time lines established concerning text and voice, nothing is being released as to the phone findings.

[Anonymous]

Norm48327

Electroliner said: "I think we need to wait and see what the results of the investigation are and hope that his memory returns. "

Why wait? Bucky seems to have it all figured out.

 

Norm, I honestly believe that Bucky is a conspiracy theorist.

[Electroliner 1935]

Based on what evidence? I have read they are checking the cellphone and its use but have not seen any statements of the results of those inquirys.

 

[Wizlish]

rdgk1se3019

Investigators believe that the engineer was using his cell phone at the time of the wreck.

Hard to believe Government people can be this stupid -- or that some people reporting on it have such agendas.

Yes, Bostian was using his cell phone 'the day of the accident'.  Yes, he had it turned off in his grip as required while running.  Yes, he said he took it out, turned it on, and called 911 from it after the accident.

How this translates into the comparison with Sanchez in 2008, as I have seen on one piece of reprehensible 'news' coverage (yes, it involves what you might expect about the shared 'orientation' smear) is difficult to comprehend, but some are trying just as hard as they can to make it.  I'd prefer we not encourage it.

[schlimm]

http://www.philly.com/philly/opinion/20150518_Amtrak_worth_the_investment.html

Some reflections on Amtrak in general and the NEC in particular by an editorial writer for the Inquirer.

[Euclid]

One of the most amazing details of this disaster is the fact that the curve had a speed limiting controller for southbound trains, but not for northbound trains.  This was explained by management with the reason being that the speed limit was 110 for southbound trains, but only 80 mph for northbound trains.  They say that even though the curve is rated for 50 mph maximum, it could be successfully run through at 80 mph.  Therefore, they concluded that no speed limiter was needed for northbound trains.

Unfortunately, this reasoning does not account for an engineer confusing his location and mistakenly accelerating to the higher speed limit when approaching the curve northbound.

http://articles.philly.com/2015-05-16/news/62192679_1_amtrak-train-train-188-train-control-system

[BaltACD]

Euclid

One of the most amazing details of this disaster is the fact that the curve had a speed limiting controller for southbound trains, but not for northbound trains.  This was explained by management with the reason being that the speed limit was 110 for southbound trains, but only 80 mph for northbound trains.  They say that even though the curve is rated for 50 mph maximum, it could be successfully run through at 80 mph.  Therefore, they concluded that no speed limiter was needed for northbound trains.

Unfortunately, this reasoning does not account for an engineer confusing his location and mistakenly accelerating to the higher speed limit when approaching the curve northbound.

http://articles.philly.com/2015-05-16/news/62192679_1_amtrak-train-train-188-train-control-system 

Well it did prove conclusively that the curve could not be traversed at speeds from 102-106 MPH.

[Euclid]

In lieu of any automatic speed limiting system, is there any visual indication for northbound trains that warn an engineer to slow down for the curve?  Are there any signal indications that would call for 50 mph maximum speed?  Are there any fixed signs warning to slow down?

[wanswheel]

Amtrak was ordered to ‘begin to install’ more signs soon. Excerpt from FRA press release, May 21.

http://www.fra.dot.gov/eLib/details/L16390#p1_z5_gD_lPR

The Emergency Order requires Amtrak to immediately implement a code change to its ATC system near the Frankford Junction curve in Philadelphia, Pennsylvania. The change must enforce the passenger train speed limit of 50 mph, or lower, for northbound trains approaching the curve. Amtrak implemented this change prior to the restart of service on Monday.

In addition, Amtrak must survey the NEC to identify each main track curve where there is a reduction of more than 20 mph from the maximum authorized approach speed to that curve for passenger trains, and provide a list of each location to the FRA.

Following Amtrak’s identification of the curves referenced above, Amtrak must develop and submit an action plan to FRA that accomplishes each of the following:

Identify appropriate modifications to Amtrak’s existing ATC system or other signal systems (or alternative operational changes) to enable warning and enforcement of applicable passenger train speeds at identified curves. 

Target dates for implementing each identified modification to Amtrak’s existing ATC system or other signal systems (or alternative operational changes) to enable warning and enforcement of passenger train speeds at the identified curves. 

Amtrak must submit the action plan to the FRA within 20 days of the date of the Emergency Order.

In addition, Amtrak must begin to install additional wayside signage alerting engineers and conductors of the maximum authorized passenger train speed throughout its Northeast Corridor system no later than 30 days after the date of the order.

[Euclid]

That is why they say, "Rules are written in blood."

[zugmann]

Euclid

That is why they say, "Rules are written in blood."

 

 

Used to be.  Today rules are written for lawyers.

[Euclid]

Here is a theory.  NOTE: If you don’t like theories or speculation, please avert your eyes.

The engineer lost situational awareness of his location and opened the throttle intending to accelerate to the higher speed limit beyond the 80 mph limit.  It was a grave mistake that the engineer became aware of after he entered the curve and dumped the air.  The engineer now claims to have no memory of it as either a conscious or unconscious denial of his mistake. 

[blue streak 1]

There have been 2 overspeed wrecks just over two years.  MNRR became proactive and instituted cab signaliing to require slowing to curve speeds.  Amtrak seems to have hesitated and the FRA stepped in to require same.

The question now becomes what if anything will the other passenger RRs do to prevent a curve overspeed happening on their road ?  Those without cab signaling  or ATC are in a more difficult  situation.  Those with ATS may be able to install a timing system much like NYC TA has.

EDIT:  Completely forgot that Metrolink installed ATS on their owned and / or operated lines shortly after Chatsworth. 

 

[jslader_sr]

Euclid

All of the news reports imply that the train suddenly accelerated during the last 65 seconds.  The word, “suddenly” means that there was no acceleration occurring prior to that time.  So if it suddenly accelerated, that means that train was running at a constant speed, and the engineer advanced the throttle at 65 seconds prior to the crash.  I don’t see anything that would justify or even explain that action except for the engineer not realizing his location.

However, some person with knowledge of the railroad had been reported in the news to say that the standard procedure was to open the throttle to full power after leaving the last station, and accelerating all the way to the curve speed restriction.  Then, upon reaching that point, engineers would reduce the throttle to a level appropriate for the 50 mph speed restriction of the curve.  If that is true, then the train would indeed have accelerated during the last 65 seconds.  But it would have also been accelerating for a minute or so prior to that point.  So the 65 seconds is just one segment of a larger acceleration event.    

But the difference in the two scenarios is that with the first scenario, the engineer advanced the throttle at 65 seconds prior to the curve.  With the second scenario, the engineer failed to reduce the wide open throttle at 65 seconds prior to the curve.  The first scenario is a matter of action taken, while the second scenario is a matter of action not taken.  Both scenarios point to different causes.   

To clarify this, we would need additional speed reading prior to the 65 second mark.  Without any clarification on this point, I assume that the first scenario is correct, and the person quoted in the news advancing the second scenario is wrong. 

 

I believe the word "suddenly" is more sensationalist reporting than anything. Fact is, the train's speed would not exactly have been constant, as it passed through North Philadelphia station only moments before, and would not have done so at 70mph. North Philadelphia is only 3.4 miles from Frankford Junction; @60 mph that would be two minutes from the beginning of the 65 second timeline. He'd have been constantly accelerating just to get up to 70, and there were other curves he'd need to traverse and adjust his speed for between N. Philly and Frankford Junction. Even the NTSB spokesman Robert Sumwalt said to speculate that the rise in speed over the final 65 seconds is speculation was "sudden" and "unusual".

Hence, I don't think it was out of the ordinary for him to begin raising his track speed to go to 80 moh once again. Incidentally, 65 seconds west (railroad south) of Frankford Junction puts the train in the general vicinity of St. Christopher's Hospital for Children, which is at the end of a curve he may not have gone full speed through. From St. Chris to Frankford Junction is a tangent of track that would allow am engineer to open his throttle some before reducing the speed for SHORE interlocking,  right before the curve. The area around St. Christopher's is rather seedy, and is thought to be the general area the other two trains-SEPTA 769 and an Acela-were also struck with objects.

 

 

 

 

 

 

 

[Paul_D_North_Jr]

blue streak 1

here have been 2 overspeed wrecks just over two years.  MNRR became proactive and instituted cab signaliing to require slowing to curve speeds.  Amtrak seems to have hesitated and the FRA stepped in to require same.

The question now becomes what if anything will the other passenger RRs do to prevent a curve overspeed happening on their road ?  Those without cab signaling  or ATC are in a more difficult  situation.  Those with ATS may be able to install a timing system much like NYC TA has.

On this point, see this column by a now-retired operating official criticizing this one-railroad-by-one-railroad approach to the problem:

http://ten90solutions.com/repetition_in_the_service_of_failure 

And my recent response to him:

"Now the substantive response comment, to the portion of the last paragraph as quoted below:

"FRA will have to address and propose methods for reducing that risk on properties that operate without cab signals and speed control."

Instead of apparently putting the duty and burden on the FRA to select and design a specific system, why shouldn't the FRA act similarly to the process that the ICC used ?  In brief, in 1922 the ICC required passenger railroads operating trains in excess of 79 MPH to install a system to prevent overspeeding on at least 1 "engine division"; in 1952, the order was extended to the entire railroad(s).  As I understand it, that was the essence and extent of the requirement; it left each railroad to select and adopt a system that was best suited to its own property.  For more on this, see:

 

http://www.railway-technical.com/US-sig.shtml - "The Imposition of ATS/ATC" (near the end)

 

http://cs.trains.com/trn/f/111/p/159578/1759010.aspx - "ATS vs ATC" - has extensive discussion of the various systems, esp. on the railroads west of Chicago

 

http://www.trainorders.com/discussion/read.php?11,1273592 - "1950s Speed Restrictions"

 

http://www.trainorders.com/discussion/read.php?11,774945 - "ATS vs. ATC"

 

http://www.trainorders.com/discussion/read.php?1,31147,31319#msg-31319 - "ATS on the Southwest Chief" (near the end)

 

The advantages of that approach should be obvious - and equally valid for PTC too, although the extent of power-sharing and run-through locomotives may make it impractical today."

 

- Paul North. 

[wanswheel]

It didn’t help his memory any that the mayor declared him guilty before the police were done interrogating him at the hospital. If his memory loss is psychosomatic or something, it might be necessary to his sanity. He has an awful lot to cope with.

[Paul_D_North_Jr]

Euclid

Here is a theory.  NOTE: If you don’t like theories or speculation, please avert your eyes.

The engineer lost situational awareness of his location and opened the throttle intending to accelerate to the higher speed limit beyond the 80 mph limit.  It was a grave mistake that the engineer became aware of after he entered the curve and dumped the air.  The engineer now claims to have no memory of it as either a conscious or unconscious denial of his mistake. 

For what it's worth, the above is pretty much the same theory/ speculation as in my post on page 10 of this thread on Monday, May 18, 2015 at 9:45 PM (EDT), about 2/3 of the way down that page.

[Euclid]

Paul,

I do recall reading your conclusion, but had sort of forgotten that you had posted it.  I thought that I had read it in a news report where somebody was explaining the possibility of getting lost.  But, yes, I agree with your theory being a good explanation for increasing the throttle just ahead of the curve.   The conclusion was also mentioned by the guy in a video at this link that I found yesterday:

http://video.foxnews.com/v/4243101361001/ntsb-downplaying-idea-that-object-hit-amtrak-windshield/#sp=show-clips&v=4251866319001

I would like to take another look at it but, I can’t seem to get the video now when I open the link.  It was in that string of videos, but I can’t get to it for some reason.  My computer wants to open where I left off in the video string.  In any case, I have been mulling it over, and it makes total sense.  I had posted something earlier about a similar mix-up in location causing an N.P. wreck in Montana.  It was the same situation of multiple similar curves and landmarks making it easy to forget which is which.      

As to my thoughts about the memory loss, that is an alternative to the theory that the memory loss was caused by the possible concussion suffered in the crash. 

[blue streak 1]

Paul_D_North_Jr

The advantages of that approach should be obvious - and equally valid for PTC too, although the extent of power-sharing and run-through locomotives may make it impractical today."

 

- Paul North. 

 

 

Paul thanks for jogging my slow brain.  Timing circuit ATS inductors throughout the US protecting slower speeds might be a poor man's method to enforce  speed restrictions.  Did edit my post to add Metrolink adding ATS inductors on its lines.

One place not mentioned about slowing down is approaching a turnout.  Was on an Amtrak train that took a 25 MPH turnout at about 45 MPH.  It was a close call that jerked us around.

 

 Note a whole route would not need inductors at each signal but at approaches to slower curves or turnouts.  Maybe distance  to assure speed would not be too great  to transit safely.

[gardendance]

jslader_sr

The area around St. Christopher's is rather seedy, and is thought to be the general area the other two trains-SEPTA 769 and an Acela-were also struck with objects.

Those people should revise their thoughts. SEPTA 769 got hit BEFORE successfully stopping at North Phily, BEFORE St Christopher's hospital. I'm not sure, but I believe I've read that the Acela also got hit between North Philly and 30th St, nowhere near St Christopher's.

Of course I in no way am trying to say that there could'nt have been more than 1 rock throwing, or whatever projectile, location.

[BaltACD]

In the Philadelphia area (Chester) several nights ago my carrier had a signal bungalow struck 'through and through' with a 'projectile' that left a 1 1/2 inch exit hole after entering the structural aluminum bungalow side, going through and destroying the electronics equipment in the bungalow and exiting the other structural aluminum side.  This was not someone 'plinking' railroad property with a .22.

I have no idea of what happend in the areas that Amtrak 188 and Septa 769 traversed - but there is nasty firepower in the Greater Philadelphia area.

[jslader_sr]

gardendance

 I believed the SEPTA and Acela trains were headed towards 30th street, when they were stuck. As a matter of fact, I know the Acela was, because when it stopped at 30th Street Amtrak police reportedly boarded the train and took witness statements. The train was deemed safe to continue and went on its way.

 

jslader_sr

The area around St. Christopher's is rather seedy, and is thought to be the general area the other two trains-SEPTA 769 and an Acela-were also struck with objects.

 

 

 

Those people should revise their thoughts. SEPTA 769 got hit BEFORE successfully stopping at North Phily, BEFORE St Christopher's hospital. I'm not sure, but I believe I've read that the Acela also got hit between North Philly and 30th St, nowhere near St Christopher's.

Of course I in no way am trying to say that there could'nt have been more than 1 rock throwing, or whatever projectile, location.