Oil Train

[Euclid]

Quote from the NTSB report with color to separate the groups of thought.  This is followed by my attempt to interpret these quoted sections of the report:

Identification of Point of Derailment

NTSB investigators recovered rail pieces from the derailment area and inventoried, measured, and documented each piece; they then ordered the pieces in the same order in which they were installed. Following the reassembly, investigators agreed that the wheel marks found on the north rail and the path of the derailed equipment indicated that the point of derailment was MP CAB 146.45.

There was a break in the railhead about 30 inches east from the rail end held in place within a set of joint bars. The placement of the joint bars was a remedial action taken by the CSXT to repair an in-service rail failure found in January 2014—more than 3 months before the accident. Investigators observed and documented markings on the inside (or gage side) of the rail joint bar that was broken. Investigators found the outside rail joint bar bent but not broken.

The rail break immediately east of the joint bar location exhibited slight rail end batter on the trailing fracture edge and slight rail end batter on the receiving rail fracture edge (fracture edge in direction of travel). The only set of fracture faces exhibiting trailing or receiving rail end batter were those associated with the reverse detail fracture located about 30 inches from the end of rail joint bars at the service rail failure that occurred in January.

Reverse Detail Fracture

The derailment occurred at a sudden break of a rail originating from a reverse detail fracture on the gage corner of the railhead of the high rail in the curve.

A CSXT contractor performed ultrasonic testing in the area of the derailment the day before the accident.17 Investigators reviewed the ultrasonic test data for the failure location. The data confirm that the test equipment functioned properly and responded to known rail features that would normally be detected by the ultrasonic test probes within the area of the failed rail.

The data showed defects discovered during the ultrasonic testing including a 20 percent transverse detail fracture; this was noted as number 151 on Sperry report number 119A, dated April 29, 2014.18

 

 

UNPACKING THE REPORT—

 

First section titled:  Identification of Point of Derailment

Report states the location of the point of derailment. 

Then it states: “There was a break in the railhead about 30 inches east from the rail end held in place within a set of joint bars. The placement of the joint bars was a remedial action taken by the CSXT to repair an in-service rail failure found in January 2014—more than 3 months before the accident.”

This second statement seems to refer to two different defects: 

1)   The first is a break in the railhead. 

 

2)   The second is an in-service rail failure found in January 2014. 

 

3)   Item #1 was 30 inches east of item #2. 

 

It is not immediately clear whether the “break in the railhead” was: The point of the derailment; a new and previously undiscovered defect; or a break that was caused by the derailment.

Then the report next refers to “end batter” on the “break in the railhead,” which seems to confirm that the break was old because the end batter was on both sides of the break, indicating that there had been travel in both directions since the break. That would indicate that the break did not just happen with the passage of the derailed train.  There is no explanation of how or when this break occurred.  Nor is there any indication of whether this is the point of derailment which was said to have been discovered at the outset.  

I tend to assume that this “break in the railhead” indicates the point of derailment simply because this set of information began with announcing that the P.O.D. had been found.  But the words do not make this clear at all.

 

Second section titled: Reverse Detail Fracture

Again, the first paragraph refers to the point of derailment being at a break in the rail without any clarification of what break it is referring to.  It refers to the break as a sudden break, which would indicate that it was not the break with the end batter on both faces, which was previously clarified to be the “break in the rail” that was found 30 inches east of the fracture repaired by joint bars in January 2014.

Next, the report describes testing that confirmed that the ultrasonic testing done the day before the derailment was working properly.

Finally, the report states that said ultrasonic testing found a 20% reverse detail fracture, but does not say where this was located, or whether it had anything to do with the one or two (unclear which) defects found near the point of derailment.

However, as this report continues (not included in the above quote), it says that the defect that caused the Lynchburg derailment was a 5% reverse detail fracture, so the above reference to finding a 20% reverse detail fracture is thoroughly unclear as to where it was found and how it relates to the derailment.  Why is the 20% fracture even mentioned?

Indeed, the report says that had the fracture that caused the derailment been 20%, it would have required a different remedial response than would have been the case had it been 5%

***

[Wizlish]

Euclid

Finally, the report states that said ultrasonic testing found a 20% reverse detail fracture, but does not say where this was located, or whether it had anything to do with the one or two (unclear which) defects found near the point of derailment.

You're not doing your job right.

This is one of the important details that needs to be distinguished in that '40 feet' - one of the critical reasons in my opinion that the decision (a major proximate cause of the accident, as things turned out) to replace the whole piece of rail as defective rather than field-welding the defects was taken.

The 20% was a transverse fracture, not a reverse-detail fracture. This may look like a jargon difference to you, but it is assuredly not to people who understand rail failure.  I will leave it up to experts like mudchicken to explain this in more credible terms, but the transverse fracture observed WAS in the rolling-contact fatigue zone, and it was at 'actionable' extent.  It was also NOT the defect that proceeded to failure to cause the accident.

For heaven's sake stop using all those words and just make the points.  (And I speak as someone who uses far too many words most of the time.)

[Wizlish]

tree68

It would be imprecise at best. By the time the crew says to themselves, "boy, that was rough," and reacts by hitting the "mark this spot" button, they may well be a significant distance beyond the actual location, especially at higher speeds.

The reason for using a 'separate color' is that the responding crew would know that the defect would be displaced a distance corresponding to reaction time and (reported) train speed back from the point marked, rather than close to the site as the Sperry-car marks are.  The direction would be inferred from which 'side' has the marked rail.  And yes, the thing would have to be maintained, and Murphy or Finagle would ensure some problem in an actual accident situation, but it's not that complex a set of requirements; a glorified paint spray can with long nozzle is really all it would take.

Not that I'm actually advocating it, but it would be one added thing simplifying a track car crew's realtime work 'process'.

[Euclid]

Wizlish

 

Euclid

Finally, the report states that said ultrasonic testing found a 20% reverse detail fracture, but does not say where this was located, or whether it had anything to do with the one or two (unclear which) defects found near the point of derailment.

 

 

You're not doing your job right.

 

The 20% was a transverse fracture, not a reverse-detail fracture. This may look like a jargon difference to you, but it is assuredly not to people who understand rail failure. 

For heaven's sake stop using all those words and just make the points.  (And I speak as someone who uses far too many words most of the time.)

 

Well then strike that phrase and accept my humblest apology for the error.  Then please go back and try to understand my point of the post.  The point is only about the amibuity of the NTSB report.  I used lot of words to make that point, but every one of them is essential to the point I am making. 

The report uses a lot of words too, but they raise lots of ambiguity.  My point was to explain that ambiguity.  That required all the words I used.  I started out with many more words and then edited it down to the bare minimum.  And I put them all into the right order too.   

[Victrola1]

DES MOINES — Almost as quickly as state regulators on Thursday unanimously approved a permit for a $3.8 billion, 30-inch diameter interstate pipeline, opponents announced plans to appeal the decision to district court, saying “this is not over.”     

http://qctimes.com/news/state-and-regional/iowa/iowa-board-approves-oil-pipeline-permit/article_301fd961-70ea-5eeb-85ae-f0d7c1bebcd9.html

The Bakken pipeline has been permitted by the State of Iowa. 

 

[Wizlish]

The thing is that the fundamental ambiguities are being augmented, not reduced, when you confuse the cause of the accident with one of the ambiguities that were confused in the report.

My complaint is not about the number of words so much as it is all the words that only contribute to the confusion, and require more work to extract the point than was needed to comprehend the situation before the post was made.

My apologies too, because this is not about words or posts or even that there are 'ambiguities' (I think they are in some places actual errors) - it is about figuring out where the actual problem was, and whether there are better ways to act to prevent recurrences in the real world.  A simple observation that the RSAC committee's revised methodology 'would have prevented this accident' is simply wrong, as I think I have demonstrated, because the actual break that caused the failure was not in the RCC zone that methodology 'flagged' for more prompt response.  What is needed (and by default provided, I think, in the CSX response mentioned in the Richmond paper's article) is increased attention to other potentially critical failures -- reverse detail fractures, in particular -- and a better set of procedures to check for these and respond to them appropriately, if that is possible.

[ruderunner]

Wizlish

The thing is that the fundamental ambiguities are being augmented, not reduced, when you confuse the cause of the accident with one of the ambiguities that were confused in the report.

My complaint is not about the number of words so much as it is all the words that only contribute to the confusion, and require more work to extract the point than was needed to comprehend the situation before the post was made.

My apologies too, because this is not about words or posts or even that there are 'ambiguities' (I think they are in some places actual errors) - it is about figuring out where the actual problem was, and whether there are better ways to act to prevent recurrences in the real world.  A simple observation that the RSAC committee's revised methodology 'would have prevented this accident' is simply wrong, as I think I have demonstrated, because the actual break that caused the failure was not in the RCC zone that methodology 'flagged' for more prompt response.  What is needed (and by default provided, I think, in the CSX response mentioned in the Richmond paper's article) is increased attention to other potentially critical failures -- reverse detail fractures, in particular -- and a better set of procedures to check for these and respond to them appropriately, if that is possible.

 

 

Timeline question, so Sperry found the defect roughly 24 hours prior to the derailment. How quickly did Sperry inform CSX of the track problem? Did CSX track crews/dispatchers get the warning in time to have stopped the train? Or did the track report show up the next morning?

[Wizlish]

ruderunner

Timeline question, so Sperry found the defect roughly 24 hours prior to the derailment. How quickly did Sperry inform CSX of the track problem? Did CSX track crews/dispatchers get the warning in time to have stopped the train? Or did the track report show up the next morning?

This is part of why I want a more exact accounting of all the defects and all the times and actions in a coherent framework.  The first question I would ask is 'which defect do you mean' out of the several that were there... the reverse detail fracture not being critical enough to 'flag' by even the RSAC revised guidelines (it being out of the rolling-contact failure zone the guidelines concerned, but the 20% transverse fracture definitely being).

What I think went on here was: there was a bad piece of rail in a track that was not as heavily used by intent as the 'other main'.  There were several defects here, one of which was actionable and about to incur a remediation procedure when the decision was made to replace the whole affected section of rail directly rather than put on a 'patch'.  That was scheduled more or less 'right away' but there was a lag associated with the time to deliver the new rail to the place and coordinate the (larger and different, perhaps) crew that would do rail replacement rather than field welding.  That lag was approximately two days, and in that time the UNRELATED reverse-detail fracture went abruptly from 5% to 100%, perhaps entirely under the train (since we have neither a definition of what 'slight' meant to the investigators nor a picture of the 'batter' involved, even assuming the reference is to the reverse-detail fracture edges) and caused the reported wheel damage and then derailment.

What I want to see is the timeline of actions and priorities that the various people involved at CSX did, and see if there are other interpretations or areas where valid 'better policies' or improvements could be made.

[Euclid]

ruderunner

Timeline question, so Sperry found the defect roughly 24 hours prior to the derailment. How quickly did Sperry inform CSX of the track problem? Did CSX track crews/dispatchers get the warning in time to have stopped the train? Or did the track report show up the next morning?

As I understand it, the defect that caused the derailment was found the day before the derailment.  It was then scheduled for replacement to take place two days later.  However, derailment occurred one day later. 

From this, I must assume that CSX was informed of the defect one day before the derailment by virtue of the fact that the repair work was scheduled at that time.  I also assume that the defect did not require the immediate halting of train traffic and repair, and that the repair being scheduled for two days later was within the rules.  Apparently, the derailment proved that the rules were defective.