EuclidYes, us non-experts can cover a lot of ground quickly with our freedom to speculate.
John Deere manure spreaders cover a lot of ground, too.
The opinions expressed here represent my own and not those of my employer, any other railroad, company, or person.
With that CN wreck I posted on the previous page, I find it amazing that the fasteners became a lethal booby trap waiting to spring without any warning. They would break where the unthreaded shank joins the threaded portion. But the break was hidden inside of the tie and went undiscovered.
When they inspected the track, they occasionally found that the head of a screw spike could be seen standing too high, thus indicating that screw had unscrewed a few turns. So they turned those back down and made them tight.
They did not try to tighten the screw spikes that had their heads appearing to be fully down. If they had tried to tighten those a little, they would have found that many of them turned completely freely because they were broken above the threads inside of the tie. The top portion would be loose, and it could be lifted right out the hole. Then they would have realized that those needed to be replaced because they were broken.
So one by one, lag screws broke without detection. Then when enough had broken, a point was reached where the remaining fasteners were insufficient to withstand the total loading in the broader area. So all of the unbroken fasteners then broke simultaneously under the passing of a train, just like what apparently happened at Mosier.
I wonder how extensive this sort of problem is or has been. I wonder if the problem has been solved. What would be needed is a way to quickly detect whether a spike lag screw is broken by pulling up on them to see if they withdrew from their holes. That would be a challenge to perfect a testing machine for working on a rolling pass over the track while tugging upward on every track screw.
schlimmSo now we have the UP spokesperson (who must be getting this from UP experts) saying what some of us non-experts have been saying?
If you throw enough theories against the wall, one of them is bound to stick...
Larry Resident Microferroequinologist (at least at my house) Everyone goes home; Safety begins with you My Opinion. Standard Disclaimers Apply. No Expiration Date Come ride the rails with me! There's one thing about humility - the moment you think you've got it, you've lost it...
Track west of Mosier is due for new ties.
Excerpt from UP news release, Jun. 1
http://www.up.com/media/releases/160601-oregon-investment.htm
Union Pacific plans to invest $34.6 million in 2016 to improve Oregon's transportation infrastructure…[including]…$3.6 million investment in the rail line between Bridal Veil and Hood River to replace 28,342 railroad ties.
I can’t link to an individual image, but here in the first image top-right, is the lag screw system that may be in use on the U.P. line through Mosier. I believe it is also the one that failed catastrophically in the CN derailment on 1/29/2012.
https://en.wikipedia.org/wiki/Rail_fastening_system
Note the lag screw that only holds down the tie plate and is not connected directly to the rail base. In looking at several images of these lag screws, I can see improvements that have been made to reduce the stress concentration at the thread/shank interface.
When these lag screws break inside of the tie, the top part becomes merely a loose pin resisting the shear force of the tie plate being pushed sideways. With the screw broken, it no longer provides any hold-down pressure that creates friction between the bottom of the plate and the top of the tie.
So with the broken top part acting only as a loose shear pin, the rail side force can easily exert shear force on the pin that causes it to wallow out the hole and allow the gage to widen.
So gage-widening can be seen as an indication that several lag screws have failed in an area of higher track stress such as a curve.
tree68 schlimm So now we have the UP spokesperson (who must be getting this from UP experts) saying what some of us non-experts have been saying? If you throw enough theories against the wall, one of them is bound to stick...
schlimm So now we have the UP spokesperson (who must be getting this from UP experts) saying what some of us non-experts have been saying?
That is the beauty of speculation. It gets the job done.
That article was written about a different rail securement system than is common in the US. Please tell us in your "experience" you've seen it used in North America. I don't think that system is relevant to this discussion. Yes, I'm aware lag screws are used in some cases.
Norm
EuclidThat is the beauty of speculation. It gets the job done.
And helps you keep your post count up.
[quote user="Norm48327"]
[/quote]
I am not sure where it is used. It is similar to, but not identical to the one used on the CN where this wreck occurred. See this report and look at photo #4:
http://www.tsb.gc.ca/eng/rapports-reports/rail/2012/r12e0008/r12e0008.asp
I have no idea what system is in place on the line through Mosier. But my only point is the use of lag screw spikes. They appear to be common in several different systems with other details varying.
Probably no help, UP website has a page about customer's track.
https://www.up.com/cs/groups/public/@uprr/@customers/@industrialdevelopment/@operationsspecs/@specifications/documents/up_pdf_nativedocs/pdf_up_std_0460.pdf
https://www.up.com/customers/ind-dev/operations/specs/track/
[quote user="Euclid"]
[quote user="Norm48327"]That article was written about a different rail securement system than is common in the US. Please tell us in your "experience" you've seen it used in North America. I don't think that system is relevant to this discussion. Yes, I'm aware lag screws are used in some cases.
https://en.wikipedia.org/wiki/Rail_fastening_system[/quote]
I have no idea what system is in place on the line through Mosier. But my only point is the use of lag screw spikes. They appear to be common in several different systems with other details varying.[/quote]
First: The 'lag screws' in the Canadian example are not the same thing as "spikes"; they are used to secure the baseplates that implement Pandrol-style clip retention of the rail in the (somewhat defectively engineered) wood-tie variant of that sort of system, the elastic clips performing all the rail hold-down functions just as they do in the concrete-tie version.
The only place I have seen lag screws used in place of spikes for rail retention was on the ex-Erie main line in the section where it divided at Glen Rock into two separate 'double-track' mains for commuter service. Here the construction was 'typical' spiked tieplates, with at least one lag screw driven down in place of a conventional spike (ISTR one screw and one spike per plate, but I wasn't paying as much attention to 'how many' as to the novelty of the screws). I did not think to check how this worked on curves (though I photographed on Sullivan's Curve many times!) That was to my knowledge unusual construction in the early '70s, and I don't recall seeing it on other main lines subject to much greater load or higher speed.
I think it's pointless to extend any analysis of Pandrol systems to spiked track, or to speculate that the "fastener" described in the Union Pacific spokesperson's report is a lag screw of this kind without confirming that a spring-retention fastening system was in use on the track through Mosier. There are people here (mudchicken being the likely best qualified) who KNOW the answer to that question, and could put this whole silly digression to bed with a word.
Personally, I have disliked the Pandrol 'conversion' systems as very ill-considered for the length of time I have known about them. Spring-clip retention works nicely for concrete ties, where the clips are reliably and firmly held in the tie structure, and theoretically should work in the kind of wood-tie conversion system where the plates are bushed accurately to the shanks of bolts that are threaded into large tee-nuts (sorry if there's a better technical term for these that I should use for track) in the 'bottom' of the tie, to give reliable preload and clamping force even when the grain of the tie opens up or weathers and admits water along the shaft of the bolts -- I know of nowhere in the United States that system is in use. Some of the pictures in R12E0008 graphically illustrate exactly the sorts of failure I'd expect a wrongheaded 'analogy' system to exhibit, and I would suspect that Union Pacific among others would go straight to concrete ties if they were going to adopt elastic fastening with or without elastomer cushioning under the rail base. But it's looking increasingly as if that's what the UP spokesmen are talking about.
And yes, it was Euclid that brought up the Pandrol system secured by lag bolts, and understood that there were precedents where a great many of those bolts failed progressively, without being effectively detected, in order to produce an ultimate failure.
On the other hand, he also decried the idea it was a 'fastener' failure and said UP was mentioning that only to avoid having to discuss a UDE causing the derailment...
What is the Ultimate Goal? Or was, 2 years ago.
http://railtec.illinois.edu/Crosstie/2014/presentations/Track%201%20-%20Concrete%20Crossties%20and%20Fastening%20Systems%20Track/Session%201-2_Fastening%20System%20Design%20Priorities/Ashmore%20-%20UIUC%20-%20Fastening%20System%20Design%20Priorities.pdf
Overmod,
I should not have used the term Lag Screw Spike because I am not sure of its scope. It may only refer to lag screws that have their heads bearing on the rail base. So my only point has to do with the use of lag screws in any type of rail mounting system including acting as cut spikes and bearing directly on the rail base. The details of the overall system they are used in is beside the point in my mind because the only issue is the ability of lag screws to break inside the tie and go unnoticed as they did in the CN wreck.
While we don’t know even if lag screws are used on the line through Mosier, that type of hidden progression of individual breakage leading up to a sudden and unexpected catastrophic failure seems like it fits what happened in Mosier.
For a line that is inspected every few days, I cannot see how suddenly, between inspections, 75 cut spikes somehow popped out of the ties or broke under the head. So I am speculating that the Mosier wreck was caused by the failure of some type of lag screw system where the lag screws broke and went unnoticed like they did in the CN wreck.
And yes, by all means, somebody please tell us what kind of rail securement is in place on the line through Mosier. I would like to know if it uses lag screws in any manner.
Excerpt from KREM Spokane/Coeur d’Alene, Jun. 11
http://www.krem.com/news/broken-lag-bolt-caused-train-derailment-up-officials-say/239795087
A broken lag bolt on the track caused an oil train to derail near Mosier on June 3, Union Pacific officials said in a town meeting Friday night.
"As the trains move around the curve, they apply a lateral force to that rail, shoving that rail out. What holds that rail intact are those lags," said a Union Pacific spokesperson. "If all those lags are broke for so many ties in a row, it allows that rail to spread out, which allowed the wheel to fall in. We had one wheel down at the point of derailment."
The point of derailment was about three-tenths of a mile east of where the rail cars piled up off the track.
Union Pacific officials said the lag bolt would have been detected by a gauge restraint measurement system but that test is only done once every 18 months. That test will now be done four times a year, officials said.
I have a concern that these bolts could be of defective manufacture. I am aware of subpar bolts being sold in the past and they did not meet the strength requirements. I would hope that the UP has a QA department that test a random sample. Is the cause of failure known?
Excerpt from Oregon Public Broadcasting, Jun. 11
http://www.opb.org/news/series/oil-trains/oregon-mosier-oil-train-derailment-union-pacific-rail-fastener-failure/
The Union Pacific Railroad says it has completed its investigation into an oil train derailment in the Columbia River Gorge…
“The fastener system that connects the railroad tie to the rail is what failed in this incident and is what caused the derailment,” said company spokesman Justin Jacobs.
Multiple metal fasteners, called lag bolts, failed in consecutive rail ties. That caused the rails to spread apart as the train rolled over them…
The Federal Railroad Administration is conducting its own investigation of the derailment.
There you go. Lag bolts failing in consecutive ties just like the CN wreck. Just as I speculated.
Gosh Euclid, if you take another bow without a lag you’ll break a bolt.
Euclid There you go. Lag bolts failing in consecutive ties just like the CN wreck. Just as I speculated.
I thought that you initially speculated/ were convinced/adamant that it was caused by a UDE, which again you were convinced that UP was trying to cover up with a rail fastner failure "Red Herring"
But Yep, I guess you were right, and UP should have consulted you from the begining.
Doug
May your flanges always stay BETWEEN the rails
wanswheel Excerpt from KREM Spokane/Coeur d’Alene, Jun. 11 http://www.krem.com/news/broken-lag-bolt-caused-train-derailment-up-officials-say/239795087 A broken lag bolt on the track caused an oil train to derail near Mosier on June 3, Union Pacific officials said in a town meeting Friday night. "As the trains move around the curve, they apply a lateral force to that rail, shoving that rail out. What holds that rail intact are those lags," said a Union Pacific spokesperson. "If all those lags are broke for so many ties in a row, it allows that rail to spread out, which allowed the wheel to fall in. We had one wheel down at the point of derailment." The point of derailment was about three-tenths of a mile east of where the rail cars piled up off the track. Union Pacific officials said the lag bolt would have been detected by a gauge restraint measurement system but that test is only done once every 18 months. That test will now be done four times a year, officials said.
I see that in this coverage, they are still referring to a lag bolt failing as though just one failed and caused the wreck. That seems to be a major point of confusion in the story.
Euclid wanswheel Excerpt from KREM Spokane/Coeur d’Alene, Jun. 11 http://www.krem.com/news/broken-lag-bolt-caused-train-derailment-up-officials-say/239795087 A broken lag bolt on the track caused an oil train to derail near Mosier on June 3, Union Pacific officials said in a town meeting Friday night. "As the trains move around the curve, they apply a lateral force to that rail, shoving that rail out. What holds that rail intact are those lags," said a Union Pacific spokesperson. "If all those lags are broke for so many ties in a row, it allows that rail to spread out, which allowed the wheel to fall in. We had one wheel down at the point of derailment." The point of derailment was about three-tenths of a mile east of where the rail cars piled up off the track. Union Pacific officials said the lag bolt would have been detected by a gauge restraint measurement system but that test is only done once every 18 months. That test will now be done four times a year, officials said. I see that in this coverage, they are still referring to a lag bolt failing as though just one failed and caused the wreck. That seems to be a major point of confusion in the story.
Guess someone has lost their reading powers.
Never too old to have a happy childhood!
challenger3980But Yep, I guess you were right, and UP should have consulted you from the begining.
Yep, if you want to know the cause right away hire an armchair expert. Of course their story is subject to change and modification faster than a politician's.
Electroliner 1935 I have a concern that these bolts could be of defective manufacture. I am aware of subpar bolts being sold in the past and they did not meet the strength requirements. I would hope that the UP has a QA department that test a random sample. Is the cause of failure known?
Although not industry required, I would assume that UP would require the manufacturer of these bolts be AAR M1003 certified. This in essence turns the entire rail industry into its QA department. Failures of this type are not that uncommon if indeed this installation is similar to the CN one. A good argument for the use of the GRMS inspection system.
EuclidI see that in this coverage, they are still referring to a lag bolt failing as though just one failed and caused the wreck. That seems to be a major point of confusion in the story.
"They" seems to be the media, not UP.
It is not uncommon to refer to a system failure in the singular, as in "a significant number of injuries in automobile collisions are caused by a failure to wear a seatbelt."
All of those people weren't injured due to one person's failure to wear a seatbelt.
I suspect that UP might feel that all of the lags that failed had something in common, be it corrosion, wear, location, whatever. Inasmuch as the lags are part of the track securement system, "the lag bolt" is apparently part of the problem, and may well be the proximate cause of the derailment.
A hero. Not the hero we deserved, but the hero we needed.
I will tell you the 'speculative' thing I want to know, in deep and procedural technical detail:
UP said they were inspecting these 'fasteners' every 18 months, but will now be doing it 'four times a year'.
How are they proposing to inspect lag screws driven down into wood ties? Pull them all out? Clamp onto the heads and use ultrasound? Back them off and use a camera to detect signs of rust or corrosion, or induce a magnetic field with dye sprayed on them to catch 'shank' stress raisers? What will they do if they break off a substantial number of them - cut out around them with a 'tube saw' and then drive in a treated plug?And what gets poured down into the hole in the tie when they back one of these screws out? Is any inspection of conditions inside the hole made, in addition to inspecting the screw?
It'll be interesting to see how this is done, or if the increased inspection frequency starts causing more problems than it tends to address...
Overmod I will tell you the 'speculative' thing I want to know, in deep and procedural technical detail: UP said they were inspecting these 'fasteners' every 18 months, but will now be doing it 'four times a year'. How are they proposing to inspect lag screws driven down into wood ties? Pull them all out? Clamp onto the heads and use ultrasound? Back them off and use a camera to detect signs of rust or corrosion, or induce a magnetic field with dye sprayed on them to catch 'shank' stress raisers? What will they do if they break off a substantial number of them - cut out around them with a 'tube saw' and then drive in a treated plug?And what gets poured down into the hole in the tie when they back one of these screws out? Is any inspection of conditions inside the hole made, in addition to inspecting the screw? It'll be interesting to see how this is done, or if the increased inspection frequency starts causing more problems than it tends to address...
What UP said was that their 'Gauge Restraint Test Vehicle' had been inspecting the territory every 18 months and in the future that test vehicle will test the territory quarterly.
The test vehicle applies lateral forces to the rails and measures the track structures resistance to turning the rail over - creating a record of forces applied and restistance and movement obtained - these values are then compared to the standards UP (and the other carriers that use similar testing equipment) have developed over the years. If not enough resistance and too much movement is measured, immediate actions are taken as necessary fix the situation. If the situation is measured to be bad enough, the track will be taken out of service until repairs are made.
EuclidThere you go. Lag bolts failing in consecutive ties just like the CN wreck. Just as I speculated.
Yes, after you speculated several other things, speculated the UP was trying to cover up something, stated it didn't make any sense, after I had to explain it to you, after the UP clarified their statement and said the same thing that I had to explain to you, yes after all that, your last speculation superseding all your previous speculation, that matched what other people, including the UP, had been saying for days was correct.
Congrats.
Dave H. Painted side goes up. My website : wnbranch.com
https://www.fra.dot.gov/Page/P0636
http://www.ensco.com/userfiles/file/Products_Services_PDF/07_Rail/12.0060-Deployable-Gage-Restraint-Measurement-System.pdf
BaltACDWhat UP said was that their 'Gauge Restraint Test Vehicle' had been inspecting the territory every 18 months and in the future that test vehicle will test the territory quarterly. The test vehicle applies lateral forces to the rails and measures the track structures resistance to turning the rail over - creating a record of forces applied and restistance and movement obtained - these values are then compared to the standards UP (and the other carriers that use similar testing equipment) have developed over the years. If not enough resistance and too much movement is measured, immediate actions are taken as necessary fix the situation. If the situation is measured to be bad enough, the track will be taken out of service until repairs are made.
I guess that's another example of the newsworkers reporting what they thought they heard rather than what Union Pacific actually said.
Has there been any work done since the Fabyan Bridge accident to figure out what a 'critical' number of fastener failures per unit distance is (in different curve and resistance circumstances) for these Pandrol-clip systems, and more importantly, how quickly the situation starts to deteriorate beyond that number? It might be that even with regular quarterly testing, track that passes a lateral 'simulated' test conducted by a relatively light vehicle might progress to dangerous 'invisible' damage before that is recognized either by subsequent testing or 'alert' crew recognition of some external sign in locomotive riding or train motion.
I was particularly concerned by that picture (in R12E0008) of the 'rolled steel plate' with shiny fretting polishing on its entire underside. I have to wonder whether there is some dynamic between some characteristic of the spring clips and the motion of the tieplate that is permitted when the lags slacken or break -- something that might facilitate very large and perhaps harmonic force that would 'progress' a track to failure long before a simple lateral force test would indicate.
Moreover, we might recall that almost at the outset of coverage of this story,
Spokesman Justin Jacobs said Saturday the track near Mosier, about 70 miles east of Portland, had been inspected at least six times since March 21. It was most recently checked last Tuesday, and within the past month, the company had used a detector car to check for imperfections as well as a geometry car to inspect the ground along the track.
Raquel Espinoza provided a little more detail in the original report wanswheel linked (between 2:15 and about 3:30) which indicated that Union Pacific's frequent 'safety inspections' essentially concentrated on rail integrity (through 'ultrasound' detection) and line and surface integrity (via the geometry car) -- 'everything aligned properly'. There was no indication whether the geometry car was measuring transient motion (lateral or vertical) that might indicate either a loss of vertical clamping authority or lateral 'pinning' against transverse excursions, but it would seem to follow that if there were, the tests did not disclose the damage in any form that triggered specific followup or remediation, even though the track was being 'inspected twice a week ... in excess of Federal requirements'.
Our community is FREE to join. To participate you must either login or register for an account.