Trains.com

NTSB report on East Palestine accident

4789 views
40 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Friday, September 20, 2024 7:53 PM
Here is an NS train at Cresson, PA that is tipping over the rail as it goes; an unusual example of what I call “Derailed-Dragging.”  In this case, it is as if the train is passing over a switch which switches the running surface of the rails from their tops to their sides. 
 
Other variations of these freakish events can have trains running on the ground for many, miles without ever parting the trainline and jackknifing cars into a pileup.  They often tear out grade crossings and keep right on going.
 
A more common variation is when a train puts one wheelset on the ties.  They will cause track damage to some extent. 
  
  • Member since
    May 2003
  • From: US
  • 25,292 posts
Posted by BaltACD on Thursday, September 19, 2024 1:24 PM

Train becoming uncoupled -

https://www.youtube.com/watch?v=S1wytAllGkY

 

Never too old to have a happy childhood!

              

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Monday, September 16, 2024 7:45 AM

BaltACD

Once a wheel hits the ground in a derailment, all predictability of what the ultimate outcome will be is gone. 

 

Yes, you never know what will happen when a wheel leaves the rail. Among other things, it depends on speed, length and makekup of train, grades, curves, terrain, slack, DPU interaction, and train handling.   
 
What I consider the strangest result of a wheel derailing at speed, is when a wheel derails and runs on the ground for maybe 500 feet, and then it gets back on the rail and runs fine. 
 
In East Palestine, one car went on the ground and ran on the ties for 1400 feet.  Other than one wheelset running on the ties, it stayed true to the line of track well enough to keep the air hoses connected.  But for some reason, the equilibrium ended.  If it would have lasted long enough to get 40 mph off of the train, it would have saved maybe a billion dollars.  
 
  • Member since
    May 2003
  • From: US
  • 25,292 posts
Posted by BaltACD on Sunday, September 15, 2024 12:38 PM

Once a wheel hits the ground in a derailment, all predictability of what the ultimate outcome will be is gone. 

Never too old to have a happy childhood!

              

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Sunday, September 15, 2024 9:26 AM
I understand Jeff’s point that locomotive with dynamic brakes have been built in two different versions that affect how they will respond to a UDE that occurs while dynamic brakes are braking against a moving train.  I also understand the reason for the version that does not cut out dynamic braking in the advent of a UDE or any other Emergency application of air brakes.
 
In territory with heavy grades, I can see the advantage of the control system allowing applied dynamic braking to continue in the advent of a U.D.E.  Basically, it would be the situation where the more braking available, the better. 
 
Regarding these two different versions of locomotive dynamic brake operation:  In a case where dynamic braking is applied, one version will cut out dynamic braking in the advent of a UDE; and the other version will not cut out dynamic braking in the advent of a UDE.
 
In the case of the version that does not cut out dynamic braking, I assume that the engineer is able to manually cut out or otherwise operate dynamic braking controls in the same manner as usual.  Is my assumption correct?    
 
Considering relatively flat track where an engineer is using dynamic braking to decelerate the train in preparation to stop the train.  Let’s say there are 100 cars between the head end power and the first trailing DPU.  In this example, buff force caused by dynamic braking would accumulate collectively among all of the 100 cars trailing the engine.  This is because these trailing cars have no braking applied, so their inertia presses them ahead against the resistance of the dynamic braking only applied to the head end power. 
 
Then, in continuing this example of operation, suddenly a UDE occurs, originating near the head end. This sudden Emergency application runs in a wave toward the rear of the train while building its accumulating pressure of buff force against the head end locomotives.
 
In this case; both the pressure of the ongoing dynamic braking plus the pressure wave of the UDE air brake application; are combining to build the highest buff force pressure nearest the head end of the train.  For some increment of time, both sources of buff force are likely to be at their maximum, and adding to each other. 
 
To reduce the danger of this high buff force, I would expect that it would be advisable to release dynamic braking upon realizing that the train has experienced a UDE, and is making a full Emergency application to the entire train.   
 
Otherwise, if the buff force rises high enough, it will buckle the train in a horizontal plane.  The first two cars to buckle will fold together in a jackknifing action.  The next two cars will follow with the same jackknifing action. When the train finally stops, most of the cars will be laid out roughly side by side. 
  • Member since
    May 2003
  • From: US
  • 25,292 posts
Posted by BaltACD on Thursday, September 12, 2024 8:11 PM

Euclid
 
jeffhergert 
Euclid
 
Say you have a freight train in dynamic braking, and you suddenly experience a U.D.E., thus triggering an automatic air brake “Emergency” application, which will stop the train as quickly as possible.

Does the dynamic brake control system recognize the U.D.E. and respond by automatically stopping dynamic braking at the moment the U.D.E. occurs? 

Depends. A UDE will open the PCS which cuts power or dynamics. Newer engines and those retrofitted will maintain dynamics when the PCS opens.

I don't know if this feature is an option or standard on new engines. I know we have this feature on most, if not all road engines now.

Jeff 

What was the reasoning for having the PCS open to cut out dynamic braking in the event of a U.D.E., when that was in the design?  And what is the reasoning for having newer engines maintain dynamics in the event of a U.D.E.?
 
What would be the benefit of applying dynamic braking during a U.D.E.?  It seems like it might be problematic.

One element of the reasoning is to prevent a secondary impact between the two parts of a train that has broken in two because of a broken knuckle or pulled out drawbar.  Wether the reasoning is appropriate or not is open to discussion.  Releasing Dynamics when the UDE trips the PCS would leave BOTH parts of the train with only emergency air braking as the primary and only braking - theoretically allowing both parts of the train to have similar stopping distances.

If the UDE trips the PCS but the Dynamics aren't released, theoretically the head end will stop BEFORE the rear end of the train.  If the rear end runs into the stopped head end of the train there is a distinct possibility of a derailment at the impact point.

The above applies to when a train has head end power only.  With DPU's in the power make up - all bets are off.  Jeff Hergert would be the authority on that aspect of train dynamics.

Any UDE interjects wildly acting in-train forces, even on level tangent track; when those force begin acting on a train in graded terriory - all bets are off, especially when it happens on descending grades where a high level of curvature has been built to get the grade to manageable levels for ascending the grade.

Never too old to have a happy childhood!

              

  • Member since
    March 2003
  • From: Central Iowa
  • 6,900 posts
Posted by jeffhergert on Thursday, September 12, 2024 8:10 PM

The reason they want dynamics to keep working when the PCS opens, and other things can cause the PCS to open, is for trains operating in heavy grade areas. Mountains. 

The PCS doesn't know if it's open due to a train separation, air hose bursting, or if the engineer (or conductor) just placed the train in emergency. Going down a heavy grade you want to maintain all the braking effort possible. I believe it was some of the runaways that brought about changes.

Jeff

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Thursday, September 12, 2024 6:57 PM

jeffhergert

 

 
Euclid

Say you have a freight train in dynamic braking, and you suddenly experience a U.D.E., thus triggering an automatic air brake “Emergency” application, which will stop the train as quickly as possible.

Does the dynamic brake control system recognize the U.D.E. and respond by automatically stopping dynamic braking at the moment the U.D.E. occurs?

 

 

 

 

Depends. A UDE will open the PCS which cuts power or dynamics. Newer engines and those retrofitted will maintain dynamics when the PCS opens.

I don't know if this feature is an option or standard on new engines. I know we have this feature on most, if not all road engines now.

Jeff

 

What was the reasoning for having the PCS open to cut out dynamic braking in the event of a U.D.E., when that was in the design?  And what is the reasoning for having newer engines maintain dynamics in the event of a U.D.E.?
 
What would be the benefit of applying dynamic braking during a U.D.E.?  It seems like it might be problematic.
  • Member since
    March 2003
  • From: Central Iowa
  • 6,900 posts
Posted by jeffhergert on Thursday, September 12, 2024 5:15 PM

Euclid

Say you have a freight train in dynamic braking, and you suddenly experience a U.D.E., thus triggering an automatic air brake “Emergency” application, which will stop the train as quickly as possible.

Does the dynamic brake control system recognize the U.D.E. and respond by automatically stopping dynamic braking at the moment the U.D.E. occurs?

 

 

Depends. A UDE will open the PCS which cuts power or dynamics. Newer engines and those retrofitted will maintain dynamics when the PCS opens.

I don't know if this feature is an option or standard on new engines. I know we have this feature on most, if not all road engines now.

Jeff

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Thursday, September 12, 2024 2:59 PM

Say you have a freight train in dynamic braking, and you suddenly experience a U.D.E., thus triggering an automatic air brake “Emergency” application, which will stop the train as quickly as possible.

Does the dynamic brake control system recognize the U.D.E. and respond by automatically stopping dynamic braking at the moment the U.D.E. occurs?

 

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Thursday, August 15, 2024 4:21 PM
Here is some discussion that is critical of how the timeline for the rising bearing temperature trend has been covered by the NTSB report:
 
  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Sunday, July 28, 2024 7:55 AM

Reading467

Euclid,   I printed the Locomotive Event Recorders Specialist's Factual Report in an attempt to compare the lead loco (NS 4178)  with the DPU (4412) parameters during the timeframe starting at 2042 hrs, roughly 12 minutes before the emergency brake application. 

The engineer starting using the dynamics on the lead and 2nd unit (NS 4224) just before 2050 hours.  The DPU's throttle was in notch 8 until 2051:40, returning to idle at 2052:10 (the 10 seconds is an approvimate reading from Figure 4. According to the Operations Group Factual Report, the crew received the HB Warning at 2053:00, based on recordings taken from inside the cab of 4178. The head end DB was in notch 2.  The DPU was at idle. 

The engineer notched the head End DB to 5 and the DPU was also in DB 5 when the emergency application was initiated at 2054:24.  

I believe the answer to your initial question of where the pileup started is in the Track Factual Report, section 3.1.:  "Additional wheel flange derailment markings were observed in the gauge of the track starting about 120 feet west of Pleasant Drive Railroad Crossing at milepost PC 49.2.  These derailment markings were observed to the inside gauge portion of the track and run parallel with the north running rail and extend eastward to where the north derailed wheel strikes the Pleasant Drive Railroad Crossing (Note 6)." 

So, that's the apparent cause of the pile-up: the derailed north wheel of axle L1 of Covered Hopper GPLX 75465 hitting the grade crossing structure after the failed bearing on the other (south) end of axle L1 burned off, thereby separating the train line and causing an emergenecy air brake application.  

 

Well, something happened to change the 1400 ft. derailed-dragging phase into the 1600 ft. destructive-buckling-jackknifing-pileup phase.  Snagging something like a switch or grade crossing (as you mention) would be one possibility. However, (unless I am mistaken) the NTSB report does not mention finding any evidence such an event, such as a ripped out grade crossing. There are also other possibilities, including ones related to train handling. 
 
As I recall, the NTSB report says the “Derailment” began with the start of the 1400 ft. derailed-dragging phase.  But they use the term “derailment” to refer to the entire train wreck, including both the derailed-dragging phase and the buckling pileup phase. So the entire train wreck covered a distance of 3000 ft. 
 
The report mentions a variety of events during the derailed-dragging phase such as dropping the bearing race and cone, truck frame sliding on the rail top, skid marks left on top of the rail, wheel flange dents left in tie tops, and burning off of an axle.  However, this does not seem as clearly presented as it could be.  I assume there were no witnesses to any of the events over the entire 3000 ft. range of the wreck. 
  • Member since
    March 2003
  • From: Central Iowa
  • 6,900 posts
Posted by jeffhergert on Friday, July 26, 2024 7:28 PM

zugmann

 

 
jeffhergert
If you read the report, it states that NS requires a train that gets a warm bearing alarm to start slowing, but not stopping until the train has cleared the detector.  Our instructions are about the same.  Using air brakes instead of dynamics could possibly cause a false reading on another car.  

 

A critical alarm requires stopping as soon as it is received.  Even if the train is still on the detector.  Non-critical alarms you clear the detector, as the detector won't even tell you specifics until you do. 

 

I admit, I don't know NS procedures, just ours. Nor do I know what alarms might be given. We only have an alarm tone with the words, "defect detected."  Until either clear of a detector or stopped, there is no other info given. If no specific info is given, and it happens, then the entire train on both sides must be inspected.

Again for us, an immediate stop is only required for a defect tone given by a detector that only checks for dragging equipment. We have been having more dragging equipment only detectors beeing put in service, usually at wayside signals, because of the use of concrete ties. Drag a wheel over one and the tie is considered defective. 

Jeff

 

  • Member since
    March 2023
  • 11 posts
Posted by Reading467 on Wednesday, July 24, 2024 12:44 PM

Euclid,   I printed the Locomotive Event Recorders Specialist's Factual Report in an attempt to compare the lead loco (NS 4178)  with the DPU (4412) parameters during the timeframe starting at 2042 hrs, roughly 12 minutes before the emergency brake application. 

The engineer starting using the dynamics on the lead and 2nd unit (NS 4224) just before 2050 hours.  The DPU's throttle was in notch 8 until 2051:40, returning to idle at 2052:10 (the 10 seconds is an approvimate reading from Figure 4. According to the Operations Group Factual Report, the crew received the HB Warning at 2053:00, based on recordings taken from inside the cab of 4178. The head end DB was in notch 2.  The DPU was at idle. 

The engineer notched the head End DB to 5 and the DPU was also in DB 5 when the emergency application was initiated at 2054:24.  

I believe the answer to your initial question of where the pileup started is in the Track Factual Report, section 3.1.:  "Additional wheel flange derailment markings were observed in the gauge of the track starting about 120 feet west of Pleasant Drive Railroad Crossing at milepost PC 49.2.  These derailment markings were observed to the inside gauge portion of the track and run parallel with the north running rail and extend eastward to where the north derailed wheel strikes the Pleasant Drive Railroad Crossing (Note 6)." 

So, that's the apparent cause of the pile-up: the derailed north wheel of axle L1 of Covered Hopper GPLX 75465 hitting the grade crossing structure after the failed bearing on the other (south) end of axle L1 burned off, thereby separating the train line and causing an emergenecy air brake application.  

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Tuesday, July 23, 2024 9:27 PM

Overmod

Euclid, the emergency occurred due to the derailment.  The thing you want from the event-recorder data in the docket, not the report is the rate of change of the throttle control going into full dynamic.

 

 

I think what I posted above was what you are looking for.  It was not from the big final report.  It was from a docket list of subjects.  I made this google search:
 
ntsb final report docket east palestine derailment 2023
 
=====================================================
 
It opens to this access:
 
 
Feb 3, 2023 — NTSB to Meet and Approve Final Report on Norfolk Southern Train Derailment Investigation in East Palestine, Ohio. NTSB Chair will also hold ...
 
==============================
 
It opens with a short document of about 3 pages on the computer and no actual page numbers.  Then in that document, there is a list of Docket numbers.  I clicked on the first one at the top of the list.  I think I must have searched at the top for:  “EVENT RECORDER FILE” 
 
In any case, that opened another page of Docket list with their numbers with their titles names. 
 
In this large list, I found the following items as items # 16,17, & 18.
 
 
I selected all three from that list, and was able to open them.  They are listed as follows:
 
16  LOCOMOTIVE EVENT RECORDERS SPECIALIST’S FACTUAL REPORT
 
17  LOCOMOTIVE EVENT RECORDERS – ATTACHMENT 1
 
18  LOCOMOTIVE EVENT RECORDERS – ATTACHMENT 2
 
 
 
 
Docket no. 16, SPECIALIST’S FACTUAL REPORT is 10 pages long and is titled:
 
RRD23MROO5
LOCOMOTIVE EVENT RECORDERS
Specialist’s Factual Report
May 31, 2023
 
 
==================================
 
 
I think that is the one you and I were looking for since it is packed with about 5 pages of extremely detailed graphs charting the movement of the train and all of the throttle and DB setting changes; plus all the dimension, and times, speeds, etc.
 
I understand your point above about the information I am seeking not being in the whole final report.  But what I have found is not coming from the whole final report.  And it also appears to be the complete information I was looking for.   
 
I have some other thoughts on this incident.  But if you can come to any conclusions about the effect of the dynamic braking and emergency air brake application, I would be very interested in your thoughts. 
  • Member since
    May 2003
  • From: US
  • 25,292 posts
Posted by BaltACD on Tuesday, July 23, 2024 4:02 PM

While this is not a hot box induced derailment - it is enlightning as to some of the dynamics that happen when cars hit the ground - for whatever the reason from a train moving at track speed.

https://www.youtube.com/watch?v=LYubpuIe3cw

 

Never too old to have a happy childhood!

              

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Tuesday, July 23, 2024 3:48 PM

Euclid, the emergency occurred due to the derailment.  The thing you want from the event-recorder data in the docket, not the report is the rate of change of the throttle control going into full dynamic.

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Tuesday, July 23, 2024 1:44 PM

Reading467

Euclid,  go back to either the STB Public Hearing Docket or the Final Report Docket, download the Event Recorder file and then you can see what actions were taken by the engineer and the corresponding reactions of the DPU (which was located after the 109th car) both before and after the point of derailment.  There are three time frames to examine- the third set has the shortest timeframe and most relevant details     

 

Thanks Reading467 for that information.  I downloaded the 10-page report from the NTSB as shown below.  Most of the report is unclear to me as to how it affected the outcome at East Palestine.  Do have any thoughts or conclusions in that regard?
 
The report contains 5 pages of technical graphs. Most interesting to me is page 3 of 10; D. FIGURES AND TABULAR DATA; second paragraph.  That paragraph is as follows:
 
 
RRD23MR005
LOCOMOTIVE EVENT RECORDERS
Specialist’s Factual Report
May 31, 2023
 
 
 
"The event recorder data from NS 4178 and DPU 4412 indicated at 20:54:24 EST both locomotives went into emergency when NS 4178’s trainline emergency transitioned from off to emergency and DPU 4412’s pneumatic control switch (PCS) transitioned from closed to open.  At the same time of the emergencyapplication, NS 4178 was moving 41 mile per hour (mph) with the dynamic brake in notch 8 (DB8).  Approximately 38 seconds later at 20:55:02, NS 4178 came to a complete stop and had traveled approximately 1,157 feet (ft.) and DPU 4412 had traveled 1,365 ft. in that time.  One second later at 20:55:03 EST, DPU 4412 came to a complete stop and [had?] traveled an additional 3 ft."  
 
 
================================
  • Member since
    January 2002
  • From: Canterlot
  • 9,575 posts
Posted by zugmann on Tuesday, July 23, 2024 1:18 PM

jeffhergert
If you read the report, it states that NS requires a train that gets a warm bearing alarm to start slowing, but not stopping until the train has cleared the detector.  Our instructions are about the same.  Using air brakes instead of dynamics could possibly cause a false reading on another car.  

A critical alarm requires stopping as soon as it is received.  Even if the train is still on the detector.  Non-critical alarms you clear the detector, as the detector won't even tell you specifics until you do. 

It's been fun.  But it isn't much fun anymore.   Signing off for now. 


  

The opinions expressed here represent my own and not those of my employer, any other railroad, company, or person.t fun any

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Monday, July 22, 2024 4:36 PM

jeffhergert

 

I don't agree with not using air brakes to stop the train after receiving an alarm.  The admonishment is to use good train handling procedures.  I understand the argument, but anything that changes the train dynamic quickly or harshly would cause the burned off wheelset to come out of alignment and cause the train to derail.  Excessive dynamics too quickly will increase buff forces.  That could "throw" the wheelset out of line and derail, just as much so as applying the air brakes.  Heat generated from using air won't cause a hot bearing to deteriorate faster.  Heat from brakeshoes would be at the the wheel's outer edges.  A low spot in the tracks or switch frog might cause the alignment to be thrown off just enough.

 

Jeff

 

Thanks for that explanation.  I think it validates the point I was making yesterday in a few posts above, but to no avail. 
  • Member since
    March 2003
  • From: Central Iowa
  • 6,900 posts
Posted by jeffhergert on Monday, July 22, 2024 3:21 PM

daveklepper

After receiving the report from the third detector, the3 engineer nbegan slowingv the vtrain, intending to stop, using only dynamic breaking. Why not blended braking for a faster, non-emergency, service stop?  Probably because he cdecided tread braking would heat the wheel even more.

Thanks, Overmod, and comment, please?

 

If you read the report, it states that NS requires a train that gets a warm bearing alarm to start slowing, but not stopping until the train has cleared the detector.  Our instructions are about the same.  Using air brakes instead of dynamics could possibly cause a false reading on another car.  

Our detectors, and others that I've heard over the years, initially don't give what the defect is or it's location.  They only do so once has cleared the detector or the train has stopped on the detector.  Stopping movement on the detector fools it into thinking the train has cleared it.  Then the detector announces type of defect and location(s) in the train.

Most of the time, a detector that checks for hot bearings also checks for other types of defects such as dragging equipment, etc.  That's part of the reason a train is allowed to clear the detector.  Our dragging equipment only detectors do require an immediate stop if an alarm is given, but doesn't give location until stopped.  Knowing where each kind of detector is located is part of knowing one's territory.

I don't agree with not using air brakes to stop the train after receiving an alarm.  The admonishment is to use good train handling procedures.  I understand the argument, but anything that changes the train dynamic quickly or harshly would cause the burned off wheelset to come out of alignment and cause the train to derail.  Excessive dynamics too quickly will increase buff forces.  That could "throw" the wheelset out of line and derail, just as much so as applying the air brakes.  Heat generated from using air won't cause a hot bearing to deteriorate faster.  Heat from brakeshoes would be at the the wheel's outer edges.  A low spot in the tracks or switch frog might cause the alignment to be thrown off just enough.

Many years ago, when I was a conductor, my train had a bearing burn off.  We were going slow through a hand throw crossover after single tracking around a MOW track project.  This required going over a detector slow enough that it was giving trains crossing over false alarms.  Every train ahead of us got a hot box alarm and had to stop and inspect.  They all found nothing.  Our turn and we had the same alarm.  After clearing the detector and receiving the type of defect and location, we stopped and I got off.  The engineer pulled the train ahead.  (Yes, as long as it's not a key train, needed excessive power, no dust or smoke seen from back in the train, and now not operating on concrete ties, we can pull the defect location up to the conductor.)  I had him stop about 20 axles short so I could start inspecting the 20 axles before the indicated location, which was about half a mile behind the enginees.  Using the temp stick, I started checking as I went.  When I got to the indicated bearing, it wasn't there.  The truck frame was dragging on the ground.  When we stopped, the bearing had seized up and stopped rolling, twisting off like a piece of plastic.  you could see this on the axle where the bearing had been.  They sent the wheel truck out (we were about 20 miles from there was one.)  I watched them change out the wheel and then was releived on hours of certain shortly after they were done.

Jeff

  • Member since
    March 2023
  • 11 posts
Posted by Reading467 on Monday, July 22, 2024 12:00 PM

Euclid,  go back to either the STB Public Hearing Docket or the Final Report Docket, download the Event Recorder file and then you can see what actions were taken by the engineer and the corresponding reactions of the DPU (which was located after the 109th car) both before and after the point of derailment.  There are three time frames to examine- the third set has the shortest timeframe and most relevant details     

  • Member since
    May 2003
  • From: US
  • 25,292 posts
Posted by BaltACD on Sunday, July 21, 2024 6:30 PM

Euclid
 
BaltACD

Euc - are you trying to contend that if the engineer had power braked, hard enough to keep the slack stretched throughout the train that there would not have been a multi-car derailment? 

I would not draw that conclusion because of all the variables.  But yes; I do think that method of braking could have prevented the pileup; but not the way it was configured in this incident.   
 
In any case, there would not be any "power braking."  Power would not be applied as though the objective was keep the train stretched such as when going down a grade. 
 
What would act to keep the train stretched in this case would be only the dynamic braking.
 
And also, it would be only dynamic braking of DPU engines; only those behind the car with the hot bearing. They would be decelerating the train in a way that would keep it stretched through the location of the hot bearing.  There would also be no dynamic braking ahead of the hot bearing car because that would have the opposite effect of having the train bunched behind the head end, and for some distance toward the hind end.  The car with the hot bearing was very close to the head end (I recall about 20 car lengths), so probably would have been in bunched slack if dynamic braking were applied by the head end power.  So, no braking at all at the head end. 
 
This is far different from the way braking was actually applied upon hearing the detector warning.  Most of it was applied to the head end because they had at least two units and they were very near the car with the hot bearing.   
 
Even with the head end dynamic braking having the wrong effect, significant motive power using dynamic braking to resist the momentum from behind the hot bearing car, may have been able to offset the bunching coming from the head end; but only if that trailing DPU was near enough to the hot bearing car.  As I recall, the DPU first behind hot bearing car (there may have been only one DPU in that location) was quite a long distance back.  And it also had the hind end cars behind it, pushing ahead, so unless head end dynamic braking was off, braking this consist from DPU way back in the train may have been futile with the head end dynamic braking bunching the slack near the head end.  
 
And also there is this point:  If dynamic braking was off at the head end and only applied to DPU behind the hot bearing; I would not conclude that braking from a trailing DPU should be agressive for the purpose of stopping as quickly as possible. I would gamble on trading quickest stopping time for minimizing the dynamic brake tension coming forward from behind the hot bearing car.  This gamble would take a little longer to stop the train. 

You are so far over your ski's you are out jumping reality and the laws of physics to a crash landing.

Never too old to have a happy childhood!

              

  • Member since
    January 2002
  • From: Canterlot
  • 9,575 posts
Posted by zugmann on Sunday, July 21, 2024 4:34 PM

nah... nevermind. 

It's been fun.  But it isn't much fun anymore.   Signing off for now. 


  

The opinions expressed here represent my own and not those of my employer, any other railroad, company, or person.t fun any

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Sunday, July 21, 2024 4:11 PM

BaltACD

Euc - are you trying to contend that if the engineer had power braked, hard enough to keep the slack stretched throughout the train that there would not have been a multi-car derailment?

 

I would not draw that conclusion because of all the variables.  But yes; I do think that method of braking could have prevented the pileup; but not the way it was configured in this incident.   
 
In any case, there would not be any "power braking."  Power would not be applied as though the objective was keep the train stretched such as when going down a grade. 
 
What would act to keep the train stretched in this case would be only the dynamic braking.
 
And also, it would be only dynamic braking of DPU engines; only those behind the car with the hot bearing. They would be decelerating the train in a way that would keep it stretched through the location of the hot bearing.  There would also be no dynamic braking ahead of the hot bearing car because that would have the opposite effect of having the train bunched behind the head end, and for some distance toward the hind end.  The car with the hot bearing was very close to the head end (I recall about 20 car lengths), so probably would have been in bunched slack if dynamic braking were applied by the head end power.  So, no braking at all at the head end. 
 
This is far different from the way braking was actually applied upon hearing the detector warning.  Most of it was applied to the head end because they had at least two units and they were very near the car with the hot bearing.   
 
Even with the head end dynamic braking having the wrong effect, significant motive power using dynamic braking to resist the momentum from behind the hot bearing car, may have been able to offset the bunching coming from the head end; but only if that trailing DPU was near enough to the hot bearing car.  As I recall, the DPU first behind hot bearing car (there may have been only one DPU in that location) was quite a long distance back.  And it also had the hind end cars behind it, pushing ahead, so unless head end dynamic braking was off, braking this consist from DPU way back in the train may have been futile with the head end dynamic braking bunching the slack near the head end.  
 
And also there is this point:  If dynamic braking was off at the head end and only applied to DPU behind the hot bearing; I would not conclude that braking from a trailing DPU should be agressive for the purpose of stopping as quickly as possible. I would gamble on trading quickest stopping time for minimizing the dynamic brake tension coming forward from behind the hot bearing car.  This gamble would take a little longer to stop the train. 
  • Member since
    May 2003
  • From: US
  • 25,292 posts
Posted by BaltACD on Sunday, July 21, 2024 1:28 PM

Euc - are you trying to contend that if the engineer had power braked, hard enough to keep the slack stretched throughout the train that there would not have been a multi-car derailment?

Never too old to have a happy childhood!

              

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Sunday, July 21, 2024 8:35 AM

BaltACD

If you get a Hot Box indiciation from a Defect Detector or from a employee or someone along the right of way - YOU DO NOT USE AIR BRAKES to bring the train to a stop, except as a last resort.

 
If a train has a hot bearing on a railcar, and is in the process of burning off the axle, an air brake application will accelerate the rate of burn-off, as you say.  
 
If a train has a hot bearing that has already burned off an axle, and thus the railcar is dragging derailed; any amount of buff force in the train will tend to buckle the train laterally at the joint of the burned off axle.  
 
Such buff force can be produced by an air brake application; and will be highly likely to be produced by a dynamic brake application made ahead of the derailed dragging car.  The only factor that could prevent buckling in that circumstance of dynamic braking is lateral resistance in the trackwork that is capable of maintaining the derailed car tracking ability.     
 
The railcar with the failing bearing at East Palestine was derailed and dragging for at least 1,400 feet and then something caused the train to buckle at or near the joint with the hot bearing. That buckling began the process of jackknifing and pileup.  It is possible that that derailed and dragging car dug in to the roadbed, and the resistance caused by digging in caused the train to buckle and start the pileup. 
 
If there had been such digging into the roadbed, I would expect that a detailed investigation would easily find the deep furrowing in the roadbed that was plowed up when the car dug in due to its derailed dragging condition. 
 
The ideal braking in this case would have been dynamic braking only on locomotives anywhere behind the hot bearing.  I don’t know if that is possible in today’s practice, but I seems like it would be possible to make practical. 
  • Member since
    May 2003
  • From: US
  • 25,292 posts
Posted by BaltACD on Friday, July 19, 2024 9:37 PM

Euclid
...
We are now officially told that the train braking had nothing to do with the derailment. We are told that the train derailed when the failing bearing burned off the axle, and then at the same moment, the burned off axle caused the derailment.  All I would like to see is some evidence to back up that assertion. Where is the science?    

Had the engineer applied the air brakes, the axle would have rung off once the brake shoe applied pressure to the wheel tread.  Tread air brakes require that the axle be 'solid' in order to allow the pressure of the brake shoe to apply the braking force to the tread.  Once the brake shoe applies force to the wheel tread, it will then displace the wheel if the axle is no longer strong enough to hold the wheel in its designed position.

If you get a Hot Box indiciation from a Defect Detector or from a employee or someone along the right of way - YOU DO NOT USE AIR BRAKES to bring the train to a stop, except as a last resort.

Never too old to have a happy childhood!

              

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Friday, July 19, 2024 8:12 PM

BaltACD

 

 
Euclid
... 
Also, I think it is fair to say that all burned off axles cause a derailment, but they don’t always cause a pileup. 

 

Critical element in whether there is a pile up of one variety or another - what was the speed of the train when the offending car 'hit the ground'.  

The NTSB says the hot bearing melted off the axle that was next to the bearing, and the loss of that axle support caused the derailment.  It surely can and does work that way sometimes.  In such a case, the part of the axle is often found on the ground near the beginning of the heap of cars if there is a pileup.  Where, and in what sort of damaged condition was the axle found in the case of this wreck?  There is a lot of evidence left in a wreck like this, and it will disclose many facts.  Apparently, in this case, the burned off axle did not initiate the pileup.  Perhaps other parts of the truck frame assembly broke up and started the pileup.  Axles can also burn off and not cause a pileup.  Also, a train can burn off an axle and pile up for another reason than the burned off axle.  
 
As to the speed of the train, I recall reading that the speed limit for the train when entering the town was 50 mph, and their actual speed of the train was 43 mph.  I have not absorbed all of the accident report yet, but I have found what seem to be some discrepancies between what was said earlier in the investigation and what is now in the final report. 
 
At least, several months ago, the NTSB reported that as the train was entering East Palestine, it was in dynamic braking to reduce the train speed to comply with the lower speed limit when passing through the town.  Suddenly the crew got the warning to stop and inspect the train.  I assume that the crew then knew that a derailment could be imminent. There was probably plenty of discussion about the hot bearing prior to reaching East Palestine. 
 
If a derailment was imminent, there would be great benefit in stopping the train before the imminent derailment happened. They had no idea how much time they actually had.  A derailment might have occurred even before they heard the warning.  It was only the train encountering the detector that determined that a warning was made. 
 
In actuality, a derailment warned of by the detector, may have been 15 minutes or more into the future.  That would have been plenty of time to stop the train with the utmost care.   So what does an engineer do?  Does he gamble that he has time to stop easy, or does he gamble to stop as quickly as possible without causing a derailment? 
 
In this case, when the engineer heard the warning from the detector, he increased dynamic braking.  Of course we don’t know if this increase in dynamic braking created an increase in buff force which buckled the train at the joint with the hot bearing.  But we do that that very soon (less than ½ minute) after increasing dynamic braking, the train actually did pile up.  It did not just put a wheel on the ground.  It began jackknifing and piling up into a heap. 
 
We also know that the hopper car with the failing bearing was only about 25 cars back from the head end power.  That factor alone with a relatively long train, implies a lot of cars pushing against the railcar with the hot bearing as it pushes against the resistance of the head end power applying dynamic braking.
 
I assume that increasing the dynamic braking was intended to minimize stopping time. The point would have been to get stopped before the pileup occurred.  Think of the money that would have been saved if the train got stopped before it piled up.  But, stopping the train quicker would have raised the risk of causing a derailment by excess buff force. 
 
We are now officially told that the train braking had nothing to do with the derailment. We are told that the train derailed when the failing bearing burned off the axle, and then at the same moment, the burned off axle caused the derailment.  All I would like to see is some evidence to back up that assertion. Where is the science?    

Join our Community!

Our community is FREE to join. To participate you must either login or register for an account.

Search the Community

Newsletter Sign-Up

By signing up you may also receive occasional reader surveys and special offers from Trains magazine.Please view our privacy policy