I agree that we have to wait. I hope we can be shown what the engineer saw (cab video) and hear his statement without waitng the 30+ days until the NTSB preliminary report. Then let the speculation recommence.
Fred M Cain It seems like there’s a lot of speculation and arguing back and forth over the probable cause in this thread. For me, I was kinda hopin’ that someone would’ve had something more factual to report by now. I was checking online news services and this whole incident is kind of getting crowded out of the news already. I guess we’ll just have to wait for the official report to come out which could take weeks or even longer. One thing about sun kinks, isn’t true that when the stresses are so high that a sun kink is imminent that it might not occur until precipitated by the vibration of the train passing over the weakness? That’s what happened in a sun kink in Tasmania as shown in a youtube video.
So, in this case, the engineer might not have noticed anything amiss. The tracks started buckling as the engines passed over it with their vibrations. Then the track started moving out of alignment and the first few cars made it across O.K.
but by the time the last few cars reached the problem the track was so far out of alignment that the cars jumped onto the ground.
Watching the video, you can see the camera swing around and catch car 37 and 38 in a big cloud of dust. Clearly, nothing had been amiss before that, or the camera would have swung around earlier. The problem happened when 37 and 38 hit that section of track.
Let us examine that track. Note how there are ties on the inside of the curve. Note how ties only seem to be in this particular location.
Note what appears to be fresh ballast down the middle of the failure section. Note how the ties are protruding excessively on the inside of the curve.
What you're looking at is a failure in track maintenance.
A news report at the time mentioned that the railroad was investigating for overweight containers. Note that 37 and 38 each had two 20' boxes. And were surrounded by empty flats. Poor track, meet heavy load.
The train operators didn't notice anything was wrong and kept going for a mile or more. They were flagged down by a motorist. The derailed cars managed to tear up a bit more track.
Ed
If ya don't want to speculate on the causes of this wreck, don't.
If ya don't like reading what other people speculate, don't read it. You can either skip over that, or find a more entertaining topic.
If you think you should decide what other people should not talk about, I'll make a note of it.
If ya just don't like speculation, I do suggest waiting for the NTSB report. That'll be about as good as it gets.
7j43k Fred M Cain It seems like there’s a lot of speculation and arguing back and forth over the probable cause in this thread. For me, I was kinda hopin’ that someone would’ve had something more factual to report by now. I was checking online news services and this whole incident is kind of getting crowded out of the news already. I guess we’ll just have to wait for the official report to come out which could take weeks or even longer. One thing about sun kinks, isn’t true that when the stresses are so high that a sun kink is imminent that it might not occur until precipitated by the vibration of the train passing over the weakness? That’s what happened in a sun kink in Tasmania as shown in a youtube video. Posted earlier by Euclid: https://www.youtube.com/watch?v=skfalqhzpkU So, in this case, the engineer might not have noticed anything amiss. The tracks started buckling as the engines passed over it with their vibrations. Then the track started moving out of alignment and the first few cars made it across O.K. 4 locomotives and 36 cars "made it across", not exactly "the first few cars". but by the time the last few cars reached the problem the track was so far out of alignment that the cars jumped onto the ground. Watching the video, you can see the camera swing around and catch car 37 and 38 in a big cloud of dust. Clearly, nothing had been amiss before that, or the camera would have swung around earlier. The problem happened when 37 and 38 hit that section of track. Let us examine that track. Note how there are ties on the inside of the curve. Note how ties only seem to be in this particular location. Note what appears to be fresh ballast down the middle of the failure section. Note how the ties are protruding excessively on the inside of the curve. What you're looking at is a failure in track maintenance. A news report at the time mentioned that the railroad was investigating for overweight containers. Note that 37 and 38 each had two 20' boxes. And were surrounded by empty flats. Poor track, meet heavy load. The train operators didn't notice anything was wrong and kept going for a mile or more. They were flagged down by a motorist. The derailed cars managed to tear up a bit more track. Ed
I don't understand the context of this thread in which you quote me. I did post the video of the Austrailian train deraling in a sun kink.
I might have said this, but I can't find it here:
"So, in this case, the engineer might not have noticed anything amiss. The tracks started buckling as the engines passed over it with their vibrations. Then the track started moving out of alignment and the first few cars made it across O.K."
I do seem to recall saying the first few cars made it across okay. But apparently you are taking issue with that because, as you say, "four locomotives and 36 cars made it across." Apparently you are referring to the Australian train. If I said what you cite, I was referring to the Amtrak train, with which two engines and 3 cars made it across.
My only point in posting the Australian train was to show an example of the train entering an over-compressed track with the crew unaware of that condtion. Then the presence of their train triggered over-ocompressed track to manifest the sun kink that eventually derailed the train.
I am not sure about the other conclusions you have drawn about the Australian train. When the camera swung around to veiw the receeding train, the sun kink was already manifested and increasing in wave size. But the train crew never saw that sun kink because it was not yet visible when the engines passed over it.
You refer to the ties protruding off the sides of the track. I assume that is the effect of the track moving sideways in the kink and getting off of the ballast, so it dropped out and left the ties more exposed.
Are you suggesting that this was not a sun kink incident, but rather negligent maintenance work? I am just trying to understand they overall point of your post.
Euclid,
I didn't quote you at all.
I cited you as previously posting the link that I posted.
All of my quotes are from the post by Fred M Cain. My responses are to him.
7j43k Euclid, I didn't quote you at all. I cited you as previously posting the link that I posted. All of my quotes are from the post by Fred M Cain. My responses are to him. Ed
Never too old to have a happy childhood!
BaltACD
C'mon, SPECULATE! You know you want to.
Overmod Before this goes on too much further, consider the force restraining 'buckling' in the vertical plane, restricted by rail weight and fastening integrity to tie weight; force restraining movement in the longitudinal direction (controlled as noted by rail anchors and fastening including Pandrol clips when present); and lateral motion (tie friction; ballast end shoulders, etc.)
Before this goes on too much further, consider the force restraining 'buckling' in the vertical plane, restricted by rail weight and fastening integrity to tie weight; force restraining movement in the longitudinal direction (controlled as noted by rail anchors and fastening including Pandrol clips when present); and lateral motion (tie friction; ballast end shoulders, etc.)
Generally, forces restraining rail in the vertical direction are only relevant if the rail is going to buckle vertically (up-and-down), not laterally (side-to-side). I don't believe I've ever heard of this happening. The cross-section of a 136# rail is such that the resistance to up-and-down buckling for a single rail is 7-8 times greater than it's resistance to side-to-side buckling. That overcomes any of the other factors having to do with lateral or vertical restraint.
Just for the sake of being thorough (maybe event pedantic), I would point out that there is one other factor that contributes to resistance to track buckling, which is the rotational stiffness of the rail fasteners. In a track buckle, the ties normally displace laterally but don't skew very much, while the rail obviously winds up with some sharp curves (hence the term "sun kink"). That means that, in the horizontal plane, the rail has to rotate relative to the tie. The fasteners may or may not resist this rotation. As Dr. Arnold Kerr put it in _Fundamentals of Railway Track Engineering_: neglecting the rotational resistance of the fasteners "may be justified for tracks with cut-spike fasteners that have been loosened by extensive traffic. It is definitely not the case for tracks with... [elastic fasteners]."
If you were to take this to the extreme - say, if you welded the base of the rail to a plate cast into a concrete tie - then the track structure would act more like a rigid frame and less like two separate rails, and then you would get vertical buckling before you got horizontal buckling. Which is to say, you would never get either kind of buckling. What you would get is a whole lot of cracked concrete ties - you need to isolate the concrete from the rails to prevent the concrete from breaking under the impact loads of a train going over.
Dan
Ed,
I agree with your synopsis and coverage of the Australian sun kink although I think it was actually in Tasmania. I think that Tasmania is actually a separate country from Australia although I'm not sure. Perhaps it's just a province but it's on an island separated from the mainland.
Anyhow, I read somewhere that trackwork had recently been done on that curve and the track hadn't completely "settled" yet. Prbly they should've kept a 10 MPH speed restriction on it but that's water over the dam.
In any event, I believe that particular line was later closed and today is out of service although the last I knew, no moves had been made to dismantle the line. There was a group fighting to get it coverted into either a commuter rail or light rail transit line but their efforts have as yet not succeeded.
Tasmania is a separate island and is also one of six states in the Commonwealth of Australia.
I speculate that if they had any idea why the train derailed, we would be hearing about it. The full report will probably be available in two years, but it is not unusual to have some idea of cause almost immediately. And there will be a preliminary report in a few weeks. Often the cause is self-evident when the accident occurs. Or it is at least partly so, as in the case of a train upsetting from overspeed through a curve, but not knowing the reason for the overspeed.
In this case, however, the investigators seem to be stumped. If a car truck had collapsed or broke an axle, they would find that right away. They also have their clues and witness marks to tell them the story, and yet they are drawing a blank. How hard can it be? The train was running at high speed and all at once, it went on the ground for no known reason.
A sun kink has been considered, and its effect is visually obvious. But, there is the question of whether the damage caused by a sun kink would be evident if the sun kink caused a derailment.
This is because like sun kinks, derailments often tear up track, plow up the ballast, and bend rails in large bows. If a sun kink caused a derailment, the derailment damage would be added to the sun kink damage, and both damage results would look similar. Or more specifically, the obvious damage inflicted by a sun kink would likely blend into the damage caused by the derailment.
In all the news coverage, I have yet to see a meaningful photograph that sheds light on the derailment action.
Yet, the evidence available is consistent with a the effect of a sun kink, while there is no other cause that I can think of that the available evidence points to.
dpeltier If you were to take this to the extreme - say, if you welded the base of the rail to a plate cast into a concrete tie - then the track structure would act more like a rigid frame and less like two separate rails, and then you would get vertical buckling before you got horizontal buckling. Which is to say, you would never get either kind of buckling. What you would get is a whole lot of cracked concrete ties - you need to isolate the concrete from the rails to prevent the concrete from breaking under the impact loads of a train going over. Dan
I suspect sucessful railroad track is finely balanced between rigidity and flexibility, similar to a wood-frame house in an earthquake zone.
(A structural engineer suggested NOT making a certain wall in this house rigid, for that reason.)
Silly NTSB investigators. Trains.com detectives solved this one in 48 hours!
The opinions expressed here represent my own and not those of my employer, any other railroad, company, or person.
Fred M Cain Ed, I agree with your synopsis and coverage of the Australian sun kink...
I agree with your synopsis and coverage of the Australian sun kink...
I tried finding an accident report or at least more data, and didn't come up with much (except that bit about the containers being overweight).
I don't see the event as "sun kink". I see it as a track maintenance failure with likely excess stress on the track, by speed and/or weight.
Note that the trailing flats appeared to stay on the track, and wiggled on through.
Here's an interesting photo showing the siding. I do agree it needs work:
I also detected a dip in the rail underneath the second wheel on the lead unit.
Here's another:
It's a CTC siding, by the way.
Sun kinks?
A slight dip in the rails underneath anything heavy is normal.
The effects of a telephoto lens often make this look worse in photos than it actually is.
Regardless of the cause of this derailment, the equipment's behaviour and number of injuries and fatalities is very similar to this VIA derailment from 1997.
https://www.bst-tsb.gc.ca/eng/rapports-reports/rail/1997/r97h0009/r97h0009.html
Greetings from Alberta
-an Articulate Malcontent
SD70Dude A slight dip in the rails underneath anything heavy is normal.
I'm sure. How come it's only under one of the wheels?
My old eyes cannot detect this dip, Perhaps you are correct and the NTSB will elabor ate.
diningcar My old eyes cannot detect this dip, Perhaps you are correct and the NTSB will elabor ate.
I feel better now I could not see it as well.
7j43k SD70Dude A slight dip in the rails underneath anything heavy is normal. I'm sure. How come it's only under one of the wheels?
Looks pretty even to me. The track in the photos looks as good or better than our mainline.
You may be surprised to learn that ballast is not finely placed, tamped and groomed on a daily basis, and small imperfections are ok.
Track is an industrial thoroughfare, not a Japanese rock garden.......
I put a straightedge underneath the rails, and found that one spot. If y'all think it's not there, or is inconsequential, I can live with that.
SD70Dude Track is an industrial thoroughfare, not a Japanese rock garden.......
It's not even that, when there's a train on the ground.
The following post has been edited for clarification 10/3/21:
NTSB says they have not ruled out any cause, including human error. Two causes have been speculated by various experts in the field.
The one cause that has suggested in the most detail is a track buckle caused by heat. It has been characterized as being the "most likely explanation." It was said that this derailment has all the “earmarks” of having been caused by track buckle caused by heat. The experts also described those earmarks and how they fit the details of this derailment.
Euclid NTSB says they have not ruled out any cause, including human error. But they have said little about what causes would be possible. Only two causes have been speculated by them. The one that NTSB suggested in the most detail is a track buckle caused by heat. They said that is the "most likely explanation." They said this derailment has all the “earmarks” of that cause. They also described those earmarks and how they fit this derailment.
NTSB says they have not ruled out any cause, including human error. But they have said little about what causes would be possible. Only two causes have been speculated by them.
The one that NTSB suggested in the most detail is a track buckle caused by heat. They said that is the "most likely explanation." They said this derailment has all the “earmarks” of that cause. They also described those earmarks and how they fit this derailment.
Could you please supply a link to those statements?
Here is the link:
https://www.usatoday.com/story/news/nation/2021/09/27/amtrak-derailment-montana-under-investigation-three-people-killed/5886070001/
Thank you for providing the link for your source.
NTSB ("They") were not quoted as saying the "most likely explanation" was heat. THAT was attributed to Russ Quimby, a former NTSB employee.
The ONLY things NTSB (Landsberg) stated was:
"We have experts that are studying the camera footage frame by frame to make sure that we see exactly what the engineer saw--or maybe didn't see,..."
In reference to ties stacked nearby on the side of the tracks: "That will be one of the questions that we look at, maintenance will be a really big concern for us. We don't know, at this point, exactly what happened, whether it was a track issue, whether it was a mechanical issue with the train. So all of these things are open."
The NTSB did not "suggest" a heat problem. There is NO comment about heat in the above quotes. Nor did the article attribute any such thing to the NTSB or one of their employees.
The NTSB at NO TIME said anything about a heat problem in the above linked source. I cannot understand why you would mis-attribute these things. It brings into question everything you assert.
You do sloppy work. And you should be ashamed of it.
Sloppy misattributions designed to lend more authority to speculations.
7j43k I cannot understand why you would mis-attribute these things. It brings into question everything you assert. You do sloppy work. And you should be ashamed of it.
I cannot understand why you would mis-attribute these things. It brings into question everything you assert.
Euclid Ed, Yes, you are correct.
7j43k 7j43k I cannot understand why you would mis-attribute these things. It brings into question everything you assert. You do sloppy work. And you should be ashamed of it. Euclid Ed, Yes, you are correct.
Get over it.
Our community is FREE to join. To participate you must either login or register for an account.