Oil Trains Cause Track Defects?

[cx500]

schlimm

 

Let's set the record straight.  Most of the German Rail components were largely designed to move heavy military trains.   The same is true of those in many other European countries, though not the UK.   Passenger trains were mostly slow, with the exception of a few lines until the last 40 years.  Some very fast trains ran in the UK, and some lines in France and Germany in the 1930s.  Freight trains in Germany run fast to meet customer needs for scheduled, quick delivery primarily ("just-in-time" inventory control), since most of the important lines are multitrack.  They don't just run fast to keep out of the way of fast passenger trains.  And true HSR stretches (> 225 kmh) have their own track or even RoW.

.........

 

We should also consider which is the cause and which is the effect.  I suggest that the German freight trains run fast because that is the only way they can be assigned a timetable path in the midst of the fast moving passenger trains.  The only way to mix a high density of trains with very different speeds is to have dedicated tracks for each; otherwise the higher speed trains are forced to conform with the slowest at every pinch point where they have to share a track.  Running the freight trains fast and short minimizes that speed differential.

Because the freight trains are forced to be tightly scheduled and fast, customers can take advantage of that ability for their shipping needs. 

But running freight trains (or any train) fast costs more since the horsepower per ton needs to be much greater, hence more locomotives are needed.  Short trains mean the labor costs and pathing cost are much higher per carload.  The government may decide that there are societal benefits to encourage freight to be carried by rail to relieve pressure on congested roads.  With nationalized infrastructure it has various tools available to change the direct costs to users.

[Paul_D_North_Jr]

From a recent meeting/ seminar notice by Diana F. Marek, Assistant Director of Northwestern University's Transportation Center* (TC) and "Sandhouse Gang" [rail-oriented] group:

* http://www.transportation.northwestern.edu/ 

http://www.transportation.northwestern.edu/news_events/calendar.php 

Thursday -  Nov. 12, 2015 -  3:00 -5:00 pm

Joint Sandhouse Group & TC  Seminar
"The Business Case for Higher Speed Freight Trains"

David Burns, Railroad Industrial Engineering Consultant

[He's from the Chicago area - see: 

 https://www.linkedin.com/pub/david-burns/7/a1b/bba ]

Rail is usually considered to be the transport mode of choice for transporting freight longer distances, yet in the USA only 15% of freight, for distances more than 2000 miles, is by rail.  For long distances many shippers are spending 2 to 3 times rail freight rates as they need faster and more reliable service than the railroads can provide.  As an indication of the rail potential shippers are spending, annually, $28 billion dollars on moving freight by truck over 1000 miles.  

 There are a number of reasons rail is not able to take advantage of the long distance market. A higher speed freight train will consume up to 3 train paths of 50 mph freight trains. There is significant increase in track maintenance cost if heavy freight trains travel at higher speeds, signaling for higher speeds and longer stopping distances, and the lack of a reliable low maintenance, high axle load, track friendly freight wagon truck for speeds over 70 mph.

 As part of the FRA's research into higher speed passenger trains, research was funded for higher speed freight. As part of the justification for this research, a study was commissioned to focus on the business case for higher speed freight trains.  

The study came to interesting specific conclusions. If higher speed freight trains could be operated as the second section of a passenger train, this would minimize the impact on line capacity and could generate a rail revenue of as much as $1 million per train!  Overall, higher speed freight trains could produce about $3 billion revenue from fresh produce and other priority food products. Revenues of about $250 million from overnight intermodal and $750 million from mail and express/courier are possible.  Importantly, the potential rail revenue could reduce the freight railroads' opposition to passenger trains.

 The presentation will explain the problems of operating freight at higher speeds, the logistics chain requirements, and the potential market for this premium type of service.

 BIO:  David Burns gained his railroad experience working 8 years for the Illinois Central Railroad, and then for 38 years he has been a railroad industrial engineering consultant.  He has undertaken numerous industrial engineering, economic, and financial studies covering most aspects of railroad operations, maintenance, and the railroad's role in the logistics chain.  Recently he researched a FRA funded study for the business case for higher speed freight trains. He has written and had published 60 plus articles and papers on various aspects of technical and marketing of rail operations.  He also brings an international prospective to his experience in that he has had consulting assignments on 45 national railway systems around the World. 

- Paul North. 

[Paul_D_North_Jr]

Deggesty

As to short, fast, trains, many moons ago (more than forty-three years; I do not remember just when) there was an article in Trains which described the then practice of the Rio Grande of running such. When my bride and I rode the Rio Grande Zephyr from Denver to Salt Lake City in July of 1972, I expected to see many of them along the way--and I saw none.

Apparently the Rio Grande determined that the practice was uneconomical.

For my good friend Johnny:

"The fast and frequent railroad - Denver & Rio Grande Western's short-and-fast concept" by Shaffer, Frank E., Trains, September 1968, pg. 20 &ff. [Magazine Index 'keywords': D&RGW  - freight - operation] 

- Paul North. 

[VaRailFan]

I read the referenced LA Times story with interest as I live near the CSX spur intp Plains Western's Yorktown trans-shipment terminal. The rails were worked over when this operation started in early 2014. I see many spikes on pulled up well over 3/4 inch which I see as normal when I have observed thr rails deflect that much when a pair of trucks passes (at only 10 mph). 

However to quote the article: '“Wide gauge” is the single largest cause of accidents involving track defects.' I have searched this entire thread for the words "gauge" and "spread" and there is no post or reply addressing this issue. I could imagine side forces on the outside rail on a curve pushing sideways against the spikes, whether 3" over a period of time I do not know.

I would like to know if anyone thinks that wide guage would be less of a problem on concrete ties. 

[monon99]

The one factor that is different with oil trains is the potential for significant weight shift by lateralish sloshing. Yes,lateral-ish. We've handled big blocks of liquids before, but almost always a variety of substances with differing viscosity in different cars with differring baffling in them-if any. The unit oil train has few similarities apart from the lighter thinner unit ethanol train. From the locomotive I get a variety of lateral forces in different locations that each car is subjected to as well. When each car pushes back with potentially greater response and the a rebound-reslosh it could be pushing defects beyond anything they've been subjected to before.

[Deggesty]

Paul_D_North_Jr

 

 

Deggesty

As to short, fast, trains, many moons ago (more than forty-three years; I do not remember just when) there was an article in Trains which described the then practice of the Rio Grande of running such. When my bride and I rode the Rio Grande Zephyr from Denver to Salt Lake City in July of 1972, I expected to see many of them along the way--and I saw none.

Apparently the Rio Grande determined that the practice was uneconomical.

 

 

For my good friend Johnny:

 

"The fast and frequent railroad - Denver & Rio Grande Western's short-and-fast concept" by Shaffer, Frank E., Trains, September 1968, pg. 20 &ff. [Magazine Index 'keywords': D&RGW  - freight - operation] 

- Paul North. 

 

Thanks, Paul.

I never saw anything more about fast and frequent service after Frank Schaeffer's article was published.

Incidentally, I enjoyed a good conversation with Mr. Schaffer, in his office, that September, after my pleasant conversation with the Managing Editor.

[dehusman]

monon99

When each car pushes back with potentially greater response and the a rebound-reslosh it could be pushing defects beyond anything they've been subjected to before.

Everybody keeps talking about "damage" from sloshing.  I get that it will move the train back and forth. 

What I have yet to have anybody explain is how something rolling back and forth on the track "damages" it.  That is what tracks are designed to do, to have thing roll back and forth.  How does a car rolling 1 foot forward, then one foot back  damage the tracks?

Somebody please explain how this damage occurs.

[dehusman]

Paul_D_North_Jr

The study came to interesting specific conclusions. If higher speed freight trains could be operated as the second section of a passenger train, this would minimize the impact on line capacity and could generate a rail revenue of as much as $1 million per train!

Why didn't somebody think of that years ago!?!?!

Oh, wait, that's what railroads do, have a premium train draft the passenger train.  Works well until you have to make train meets, then not so much.

We have territory where we run 50-60 trains, about one third intermodal or automotive and one passenger train a day.  How do you operate 10 trains as a "second section" of a single passenger train?

 A higher speed freight train will consume up to 3 train paths of 50 mph freight trains. There is significant increase in track maintenance cost if heavy freight trains travel at higher speeds, signaling for higher speeds and longer stopping distances, and the lack of a reliable low maintenance, high axle load, track friendly freight wagon truck for speeds over 70 mph.

Now we are getting to the nitty gritty.  What people don't understand is that a passenger train can have a footprint that is 75-125 miles long.  On a heavy single track line one passenger train can cause trains to be held up to a crew change away (over 100 miles) because all the sidings on the subdivision are or will be filled with trains that have to be met or passed.  I have managed several territories with a mix of passenger and freight operation and having to lay down freight trains in sidings several hours in advance of the passenger train were quite common.

The gentleman quoted in article was off the IC.  It was very common to have the IC hold manual interlockings open for a passenger train up to an hour in advance of the train.  Not only were the passenger trains gumming up the works on railroad operating them, but they were delaying trains on routes of other railroads crossing the line with the passenger service.  Been there, had to explain to managment why their trains were being delayed for a passenger train that didn't even operate on their railroad.

[Paul_D_North_Jr]

No offense intended to anyone, but I now think the last few pages of this "Oil Trains Cause Track Defects?" thread deserve to have their own new and independent properly titled thread relating to operating issues, or moved or referenced in another similar thread, etc.

- Paul North. 

- Paul North.  referenced  

[Euclid]

monon99

The one factor that is different with oil trains is the potential for significant weight shift by lateralish sloshing. Yes,lateral-ish. We've handled big blocks of liquids before, but almost always a variety of substances with differing viscosity in different cars with differring baffling in them-if any. The unit oil train has few similarities apart from the lighter thinner unit ethanol train. From the locomotive I get a variety of lateral forces in different locations that each car is subjected to as well. When each car pushes back with potentially greater response and the a rebound-reslosh it could be pushing defects beyond anything they've been subjected to before.

 

Monon99,

I understand what you are saying and think it should be investigated through modeling and testing.  I assume that no testing has ever been done because tank cars have simply proven themselves in practical operation for over a century.  But as you say, oil unit trains of today’s tank cars are something relatively new and may be producing unanticipated effects.  You sound like you have practical experience as a locomotive engineer, so I am interested in your experiences with this sloshing effect.  I recently had a conversation with someone at Wabtec, and he brought up tank cars sloshing as being suspect in oil train derailments.  Many others have commented about it in these forums.  Railway Age ran this article about their intent to interview some FRA experts on the matter of tank car sloshing.  This was after FRA Acting Administrator Sarah Feinberg brought it up by saying this:

How crude oil sloshing inside moving tank cars affects train stability was under close scrutiny by the Federal Railroad Administration, the regulator’s Acting Administrator told reporters back on March 13. That was after a string of mid-winter oil train disasters exposed the prevailing focus on tank car thickness to be essentially pointless in the quest to prevent oil train derailments and explosions.

But after agreeing to be interviewed by Railway Age on the topic of sloshing, the FRA stood them up three times and the interview never happened.  So they wrote about that experience instead of sloshing:

http://www.railwayage.com/index.php/blogs/david-thomas/fra-freezes-on-tank-car-sloshing-doe-oil-volatility-bombshell-drops-like-a-dud.html

 

Here are my thoughts on the sloshing effect which I wrote on Fred Frailey’s blog a while back when sloshing was being discussed there:

Added 7 months ago

Euclid said:

I too have wondered about that sloshing effect.  When I hear the term “sloshing,” I think of fluid rolling and splashing around inside the tank with a tendency to rock the tank and tip it over.  This would be random motion varying within each tank as well as from tank to tank.

But there is another type of effect that I wonder about, and nobody has mentioned. I would not call this other effect “sloshing.”  I would call it “fluid surging” of the entire liquid load.  One person on the general forum said that when the oil is loaded to full weight for the tank car, there is 269 cubic feet of air in the tank above the oil.  I have no idea if that is true.  

If that is true, there is great potential for the mass of oil to surge forward in each car as slack runs in.  This would overload the lead truck of the car.  

In addition, perhaps this oil mass surge in many tanks could amount to a second type of slack action that is set into motion by the coupler/draft gear slack action.  If that is possible, maybe that liquid surge in each tank could accumulate force from car to car to the point where it lifts wheels off of the rails.  

Or maybe a run-in force, exaggerated by the liquid surge, could combine with the weight transfer that overloads the lead truck of each tank car.  So you would have the weight transfer overloading the lead truck as it reduces the normal loading on the rear truck.  As the truck springs react to the load change, the front of the tank car would drop as the rear rises.  

The natural line of slack action force is perfectly parallel with the track bed.  So if that line of slack force is applied to tank cars that are slightly tilted from the weight transfer, it will try to tilt the cars further.  In that sense, the slack run-in will be somewhat translated into down pressure on the lead truck of each car in addition to the added truck load created by the fluid weight transfer inside of each car.  So you would have two separate force generators combining to add loading to the lead truck of each car.  

Maybe this complex run-in and weight transfer could overload the lead truck of a tank car to point of breaking a wheel or a rail.   

[dehusman]

Euclid

If that is true, there is great potential for the mass of oil to surge forward in each car as slack runs in. This would overload the lead truck of the car.

Hyperbole. 

Do the math.  269 cu ft x 7.4 gal/cu ft = 1990 gal.

7 lb/gal x 1990 gal = 13,934 lb = 7 tons

At rest on level track, that void is across the entire length of the car, so only half of the void is at one end, so only half of the weight can be increased at one end, 3.5 tons (It will probably be less than that since the "slosh" will be more of a gradient, but 3.5 tons is a worst case scenario). 

Your "overloading" amounts to only 3.5 tons increase or less than 1 ton per wheel.  Hardly a smoking gun. 

If that is a horribly dangerous thing then you better figure out how to run those trains on perfectly flat territory, because if you put the car on a grade, the void will move to the high end of the car and "overload" the low end. 

Once again, how does increasing the load on one end of the car by less than 4 tons damage the track to the point of failure? 

[Paul_D_North_Jr]

Some other forces to consider, in addition to the pure weight transer that Euclid and Dave H. posted above:

The alleged sloshing longitudinally will also impact the end of the car, which will stop it.  That will cause a momentum type 'impulse'  longitudinal reaction in the car end, which will then cause a very slight rotation downward of that end of the car (and upward on the other end), which will likewise cause a slight increase (decrease) in the forces on the track. 

The slack action run-in - which is in line with the draft gear and frame, but below the functional frame for those tank cars that don't have a center sill and instead rely on the tank body itself to connect the 2 ends of the car - occurs below the center of gravity of the car and load.  From inertia, that will tend to bow the car upwards slightly.  Depending on the relative magnitude of the buff/ compression on the couplers and draft gear at each end of the car, there might be a similar rotational effect and possible slight increase/ decrease in the vertical loads at each end of the car.

However, keep in mind that the centroid of any such forces are about 5 ft. or less vertically from the point of resistance to them.  In comparison, the truck center distance on the cars is on the order of 50 ft.  As such, the coversion of the longitudinal force to a vertical change in load is reduced by a proportional factor of 5 / 50 = 10% (as analyzed by the "summation of moments" method, if you know what that is).   

I don't have time or means to do the math properly, but my intuitive reaction (pun !) is those forces are nowhere near enough to break rails.

To say any more would require an extremely detailed and complex dynamic modeling of the train and load forces.  To my knowledge, no one has done so in a credible manner.  In any event, the possible results would largely be governed by the range of input assumptions, which would make the results meaningless.

To me, the 'usual suspect' of dynamic impacts from flat wheels, latent rail defects, poor track surface and support, etc. are a much more likely root cause of any track-related derailments than sloshing.

- Paul North.

[Buslist]

Paul_D_North_Jr

From a recent meeting/ seminar notice by Diana F. Marek, Assistant Director of Northwestern University's Transportation Center* (TC) and "Sandhouse Gang" [rail-oriented] group:

* http://www.transportation.northwestern.edu/ 

http://www.transportation.northwestern.edu/news_events/calendar.php 

Thursday -  Nov. 12, 2015 -  3:00 -5:00 pm

Joint Sandhouse Group & TC  Seminar
"The Business Case for Higher Speed Freight Trains"

David Burns, Railroad Industrial Engineering Consultant

[He's from the Chicago area - see: 

 https://www.linkedin.com/pub/david-burns/7/a1b/bba ]

Rail is usually considered to be the transport mode of choice for transporting freight longer distances, yet in the USA only 15% of freight, for distances more than 2000 miles, is by rail.  For long distances many shippers are spending 2 to 3 times rail freight rates as they need faster and more reliable service than the railroads can provide.  As an indication of the rail potential shippers are spending, annually, $28 billion dollars on moving freight by truck over 1000 miles.  

 There are a number of reasons rail is not able to take advantage of the long distance market. A higher speed freight train will consume up to 3 train paths of 50 mph freight trains. There is significant increase in track maintenance cost if heavy freight trains travel at higher speeds, signaling for higher speeds and longer stopping distances, and the lack of a reliable low maintenance, high axle load, track friendly freight wagon truck for speeds over 70 mph.

 As part of the FRA's research into higher speed passenger trains, research was funded for higher speed freight. As part of the justification for this research, a study was commissioned to focus on the business case for higher speed freight trains.  

The study came to interesting specific conclusions. If higher speed freight trains could be operated as the second section of a passenger train, this would minimize the impact on line capacity and could generate a rail revenue of as much as $1 million per train!  Overall, higher speed freight trains could produce about $3 billion revenue from fresh produce and other priority food products. Revenues of about $250 million from overnight intermodal and $750 million from mail and express/courier are possible.  Importantly, the potential rail revenue could reduce the freight railroads' opposition to passenger trains.

 The presentation will explain the problems of operating freight at higher speeds, the logistics chain requirements, and the potential market for this premium type of service.

 BIO:  David Burns gained his railroad experience working 8 years for the Illinois Central Railroad, and then for 38 years he has been a railroad industrial engineering consultant.  He has undertaken numerous industrial engineering, economic, and financial studies covering most aspects of railroad operations, maintenance, and the railroad's role in the logistics chain.  Recently he researched a FRA funded study for the business case for higher speed freight trains. He has written and had published 60 plus articles and papers on various aspects of technical and marketing of rail operations.  He also brings an international prospective to his experience in that he has had consulting assignments on 45 national railway systems around the World. 

- Paul North. 

 

There were those that questioned Jim Blaze's career at CR as questionable, wonder how many will question 8 years at the IC? To David's credit he claims to be the IE that set up the IC's locomotive rebuild program.

[Euclid]

Paul,

I did mention what you suggest about the rotation of the car in line with the track in my above post as follows:

“The natural line of slack action force is perfectly parallel with the track bed. So if that line of slack force is applied to tank cars that are slightly tilted from the weight transfer, it will try to tilt the cars further. In that sense, the slack run-in will be somewhat translated into down pressure on the lead truck of each car in addition to the added truck load created by the fluid weight transfer inside of each car. So you would have two separate force generators combining to add loading to the lead truck of each car.”

Dave,

I am not saying that any of this is proven to be happening.  Yet you seem intent on denying with absolute certainty that it is even possible.  Yes, a few tons of weight transfer does not seem like a big deal, but I am not convinced that your math captures the entire action.  In the first place, it is not 3.5 tons of static load that is at issue.  When you factor in the sudden acceleration and then arrest of that load, it becomes a dynamic impact which I suspect would far exceed the 3.5 ton static load.

I am also not convinced that your 3.5 tons of oil is correct.  That is half of the air space, but it is the oil rather than the air space that moves to cause the impact.  So it seems to me that the moving oil would be a good size portion of the oil load, perhaps at lease several dozens of tons being accelerated by slack run-in, and then arrested by the front tank wall.   

[tree68]

Regardless of the amount/mass of the oil involved, it's going to take regular occurance of that motion over the same spot to cause the track to fail as a result of this action.  And a truly significant "slosh" would require a rather sudden change in speed.  Recall that you can drive all day without spilling your coffee if you keep your starts and stops gentle.

That does not rule out the failure of the track due to an existing defect combined with a one-time "slosh."  

I would opine that it would take the development of a "washboard" to stress the track at a particular point.  This would have to occur rather like a washboard on a gravel road - one bump causes a slight depression, which begets another, and another, etc.  

I have no idea of what the period of said washboard would have to be to perpetuate itself, but I would imagine that crews would certainly notice it.  And it would certainly show up in track testing and be remediated when the track is surfaced.

Further, rubber tires are pretty consistent in size, unlike the length of railcars (which would be the determining factor - not wheel size).

Likely?  Possible? May be.  More learned minds than mine will have to weigh in.

[Euclid]

tree68

Regardless of the amount/mass of the oil involved, it's going to take regular occurance of that motion over the same spot to cause the track to fail as a result of this action.   

What I am talking about is not slosh causing a repetitive overload in one point that eventually develops a defect that fails under a train.  I am talking about one overload event causing a failure in a rail that was not previously defective. 

As I mentioned earlier in response to the conclusion of the article:

“I cannot see why oil trains would encounter more derailment-causing track defects than other trains; or why oil trains would challenge a track defect more than other types of unit trains. 

So that leaves this possibility:  Oil trains that derail are causing track defects which then derail that same train before it passes the defect. That is the only conclusion that can explain what the article observes from the statistics.”  

[BaltACD]

Euclid

tree68

Regardless of the amount/mass of the oil involved, it's going to take regular occurance of that motion over the same spot to cause the track to fail as a result of this action. 

What I am talking about is not slosh causing a repetitive overload in one point that eventually develops a defect that fails under a train.  I am talking about one overload event causing a failure in a rail that was not previously defective. 

As I mentioned earlier in response to the conclusion of the article:

“I cannot see why oil trains would encounter more derailment-causing track defects than other trains; or why oil trains would challenge a track defect more than other types of unit trains. 

So that leaves this possibility:  Oil trains that derail are causing track defects which then derail that same train before it passes the defect. That is the only conclusion that can explain what the article observes from the statistics.”

What were the derailment statistics in the 1940's when oil was being carried on rail because of the German submarines sinking coastal oil tankers?

[schlimm]

Paul_D_North_Jr

From a recent meeting/ seminar notice by Diana F. Marek, Assistant Director of Northwestern University's Transportation Center* (TC) and "Sandhouse Gang" [rail-oriented] group:

* http://www.transportation.northwestern.edu/ 

http://www.transportation.northwestern.edu/news_events/calendar.php 

Thursday -  Nov. 12, 2015 -  3:00 -5:00 pm

Joint Sandhouse Group & TC  Seminar
"The Business Case for Higher Speed Freight Trains"

David Burns, Railroad Industrial Engineering Consultant

[He's from the Chicago area - see: 

 https://www.linkedin.com/pub/david-burns/7/a1b/bba ]

Rail is usually considered to be the transport mode of choice for transporting freight longer distances, yet in the USA only 15% of freight, for distances more than 2000 miles, is by rail.  For long distances many shippers are spending 2 to 3 times rail freight rates as they need faster and more reliable service than the railroads can provide.  As an indication of the rail potential shippers are spending, annually, $28 billion dollars on moving freight by truck over 1000 miles.  

 There are a number of reasons rail is not able to take advantage of the long distance market. A higher speed freight train will consume up to 3 train paths of 50 mph freight trains. There is significant increase in track maintenance cost if heavy freight trains travel at higher speeds, signaling for higher speeds and longer stopping distances, and the lack of a reliable low maintenance, high axle load, track friendly freight wagon truck for speeds over 70 mph.

 As part of the FRA's research into higher speed passenger trains, research was funded for higher speed freight. As part of the justification for this research, a study was commissioned to focus on the business case for higher speed freight trains.  

The study came to interesting specific conclusions. If higher speed freight trains could be operated as the second section of a passenger train, this would minimize the impact on line capacity and could generate a rail revenue of as much as $1 million per train!  Overall, higher speed freight trains could produce about $3 billion revenue from fresh produce and other priority food products. Revenues of about $250 million from overnight intermodal and $750 million from mail and express/courier are possible.  Importantly, the potential rail revenue could reduce the freight railroads' opposition to passenger trains.

 The presentation will explain the problems of operating freight at higher speeds, the logistics chain requirements, and the potential market for this premium type of service.

 BIO:  David Burns gained his railroad experience working 8 years for the Illinois Central Railroad, and then for 38 years he has been a railroad industrial engineering consultant.  He has undertaken numerous industrial engineering, economic, and financial studies covering most aspects of railroad operations, maintenance, and the railroad's role in the logistics chain.  Recently he researched a FRA funded study for the business case for higher speed freight trains. He has written and had published 60 plus articles and papers on various aspects of technical and marketing of rail operations.  He also brings an international prospective to his experience in that he has had consulting assignments on 45 national railway systems around the World. 

- Paul North. 

 

$4 bil. in additional revenue could replace some of the lost revenue from coal.  Amid the extra costs, the author overlooks an obvious savings in improved car utilization from more speed.

[Electroliner 1935]

BaltACD

What were the derailment statistics in the 1940's when oil was being carried on rail because of the German submarines sinking coastal oil tankers?

Much much smaller cars. Doubt the stresses were close to taxing the plant.

BaltACD

Oil trains that derail are causing track defects which then derail that same train before it passes the defect. That is the only conclusion that can explain what the article observes from the statistics

 

This is a possibility that needs research. I suspect that you are thinking about something similar to what was learned in the Tacoma Narrows Bridge Failure. Forces that compound and exceed the structures ability to coltrol/contain them. I've been on trains that lurched and the passengers became unstable. Then like tenpins, one takes down the next. 

[BaltACD]

Electroliner 1935

 

 

 

Euclid

Oil trains that derail are causing track defects which then derail that same train before it passes the defect. That is the only conclusion that can explain what the article observes from the statistics

This is a possibility that needs research. I suspect that you are thinking about something similar to what was learned in the Tacoma Narrows Bridge Failure. Forces that compound and exceed the structures ability to coltrol/contain them. I've been on trains that lurched and the passengers became unstable. Then like tenpins, one takes down the next.

Had to correct the quote!

[Paul_D_North_Jr]

Euclid:  OK, with that clarification I now see what you meant.  I took "tilt" to mean leaning to the side - what the aeronautical people call "roll".  Instead, they would call the motion that we're both referring to as "pitch"; I might also call it downward deflection (technically not correct, but kind of intuitive), or longitudinal rotation.

- Paul North. 

[Paul_D_North_Jr]

schlimm wrote above:

"$4 bil. in additional revenue could replace some of the lost revenue from coal."

"+1"

"Amid the extra costs, the author overlooks an obvious savings in improved car utilization from more speed"

Perhaps just not mentioned in the brief abstract ?  Nope - I quickly scanned the study (see below), and did not see anything on that item.  Good point, schlimm !

Hopefully someone here (greyhounds ?) can go and report back on this and other aspects of the study.

Here's the way to get FRA study / "Technical Report" that he's apparently referring to - "Higher Speed Freight Truck Market Analysis" (05 July 2013):

Go to: https://www.fra.dot.gov/eLib/details/L04665 and click on the "DOWLOAD" button at either the lower left or the lower right corner.  It's 56 pgs., approx. 1.64 MB electronic file size in "PDF" format. 

Also apparently referenced and described in the Sept. 2013 issues of Railway Gazette International:

http://trid.trb.org/view.aspx?id=1263487 

- Paul North.

[M636C]

Euclid wrote the following post 3 hours ago:

 

tree68

Regardless of the amount/mass of the oil involved, it's going to take regular occurance of that motion over the same spot to cause the track to fail as a result of this action.   

 

What I am talking about is not slosh causing a repetitive overload in one point that eventually develops a defect that fails under a train.  I am talking about one overload event causing a failure in a rail that was not previously defective. 

As I mentioned earlier in response to the conclusion of the article:

“I cannot see why oil trains would encounter more derailment-causing track defects than other trains; or why oil trains would challenge a track defect more than other types of unit trains. 

So that leaves this possibility:  Oil trains that derail are causing track defects which then derail that same train before it passes the defect. That is the only conclusion that can explain what the article observes from the statistics.” 

I strongly disagree with the last paragraph:

It depends upon what you mean by a "defect"....

Say we are talking about a broken rail caused by an internal fatigue fissure propagating from an inclusion (say a small piece of slag) inside the steel....

When the oil train arrives, the rail isn't broken. But the inclusion is there, as it has been since the rail was rolled. The fissure may or may not be there, but might be there as a very small defect surrounding the inclusion. This might not have shown up on any recent ultrasonic testing, if the fissure was small enough at that time.

But the fissure is a defect. It isn't a visible defect and it may not show up to a mobile Ultrasonic inspection from a Hi Rail running at say ten miles an hour. 

If the track in question has only oil trains as unit trains, and all other traffic is a mix of intermodal and mixed freights only a few axles will be heavily enough loaded to initiate enlargement of the fissure. But along comes an oil train whose cars are loaded right up to 263 000 pounds, and there are 100 identical cars. The speed and weight of the axle is just right to enlarge the fissure. After 200 repetitions, the fissure is big enough to break the rail and does so under the middle car.

Of course, if the line had unit coal trains or ore trains loaded to the same mass, and they ran at the same speed, they could initiate the failure. But if the line in question had oil trains as the only unit trains, the rails would break under the oil trains, and no VISIBLE defect would be seen before the critical oil train passed.

Suppose the defect needed 4200 cycles of a fully loaded oil train axle to develop: It would break in the middle of the eleventh loaded oil train...

I believe there are enough oil trains running on lines with no other significant unit train activity for this to be a reasonable explanation for the statistics quoted.

I'm not convinced that "sloshing", longitudinal or lateral is significant. The air space above the load is small. I believe this would act more like the bubble in a spirit level than a free wave at a surf beach. I think the volume of free air would be so small as to not significantly shift the weight of the cargo. After all most of the oil isn't moving, it's already there. Only a small top surface moves.

(My spirit level is lettered "Plasser &Theurer" by the way, and forms a key ring)

M636C

[dehusman]

Euclid

So that leaves this possibility: Oil trains that derail are causing track defects which then derail that same train before it passes the defect. That is the only conclusion that can explain what the article observes from the statistics.”

Its only a possibility if you can identify HOW sloshing creates enough force to break a rail.  (Hint : it doesn't)

Sloshing is a relatively slow movement forward and back (as opposed to the slack running in which is hugely more violent).

Sloshing side to side has to have something to cause it.  Since we are talking about a void at the top of a tank car shell that is probably less than a foot high, the amount of sloshing side to side would be minimal.  In order to get violent side to side sloshing you would have to have track or car defects would most likely deraile the car anyway.

If a thousand people say sloshing is a problem it doesn't mean it is unless somebody can quantify what the loadings are and how they are generated.  So far nobody has and the numbers I have come up with don't seem to indicate any significant loading.  I've been on trains where tank cars sloshed back and forth, the movements weren't anything even remotely close to violent enough to damage the track. 

[Euclid]

M636C

 

I'm not convinced that "sloshing", longitudinal or lateral is significant. The air space above the load is small. I believe this would act more like the bubble in a spirit level than a free wave at a surf beach. I think the volume of free air would be so small as to not significantly shift the weight of the cargo. After all most of the oil isn't moving, it's already there. Only a small top surface moves.

(My spirit level is lettered "Plasser &Theurer" by the way, and forms a key ring)

M636C

 

I believe that the way the quoted statistic is stated, it does not just compare oil unit trains to mixed freight trains.  It compares unit oil trails to all other types of trains and finds oil unit trains more like to be derailed by track defects than are trains such as ore unit trains or coal unit trains.  So what are oil unit trains doing differently?

As to air allowing slosh, it depends on how much air there is.  I do not think that the amount of air has been convincingly established, or that it does not vary from one type of oil to another.  So the potential dynamic loading from oil slosh depends on how much air is available for the oil to displace, and how hard the slack runs in to accelerate the oil mass within the tank. 

[Wizlish]

Euclid

When you factor in the sudden acceleration and then arrest of that load, it becomes a dynamic impact which I suspect would far exceed the 3.5 ton static load.

If I read this sentence correctly, you think the slosh causes hard 'bottoming' of the truck suspension, and there are no 'bump stops' on a three-piece truck.

I do not think either the overall force or the speed with which it acts to cause truck suspension deflection are even ofthe order of magnitude needed to produce such a result.  Even if they did, the force following the 'bottoming' is as likely to "splash" the oil as it is to force the wheeltread against the railhead ... and the force would be smoothly applied, as the slosh is longitudinal with relatively slight vertical component.

I think it is grossly unlikely that this action breaks railheads or webs.  Perhaps if the slosh were to reduce apring travel just at the point a geometry defect tended to kick the sideframe(s) up, you could get an augmented 'bottoming' effect, but that's something of a stretch.  And you'd see it preferentially in places where those defects are commonly encountered, like at crossings or bridges.  I do not believe that has been observed in oil-train derailment statistics.

[Euclid]

dehusman

 

Euclid

So that leaves this possibility: Oil trains that derail are causing track defects which then derail that same train before it passes the defect. That is the only conclusion that can explain what the article observes from the statistics.”

 

 

Its only a possibility if you can identify HOW sloshing creates enough force to break a rail.  (Hint : it doesn't)

Sloshing is a relatively slow movement forward and back (as opposed to the slack running in which is hugely more violent).

Sloshing side to side has to have something to cause it.  Since we are talking about a void at the top of a tank car shell that is probably less than a foot high, the amount of sloshing side to side would be minimal.  In order to get violent side to side sloshing you would have to have track or car defects would most likely deraile the car anyway.

If a thousand people say sloshing is a problem it doesn't mean it is unless somebody can quantify what the loadings are and how they are generated.  So far nobody has and the numbers I have come up with don't seem to indicate any significant loading.  I've been on trains where tank cars sloshed back and forth, the movements weren't anything even remotely close to violent enough to damage the track. 

 

I think it is a possibility whether I can identify HOW sloshing creates enough force to break a rail or not.  I will wait until you prove that it is not a possibility before I stop thinking it is a possibility. 

And just to clarify one point that you make; I do not know if a thousand people have said sloshing is a problem.  Nobody here has said that.  You are the only one who is holding to an absolute assertion, and that is that sloshing is NOT causing any problems.  I’ll wait for your proof. 

I too have seen tank cars slosh back and forth with movements that weren’t anything close to violent enough to damage the track. 

What does that prove?  And again, I am not considering a problem of oil sloshing side to side from the ride.  I am talking about oil surging forward from slack run-in. 

[Euclid]

Wizlish,

I had not considered whether the trucks would bottom or not, nor did I know that bottoming the trucks would not be possible if that is what you mean by trucks without bump stops. What happens when you push down on the truck hard enough to close the coil springs? Wouldn’t the truck be fully “bottomed out” between the rail and the top of the truck bolster?   

[Wizlish]

Euclid

What happens when you push down on the truck hard enough to close the coil springs? Wouldn’t the truck be fully “bottomed out” between the rail and the top of the truck bolster?

That would be just what I mean, with the springs already part way closed because the car is presumably loaded to near capacity.

I do have to wonder, though, if lateral roll is a significant force.  I have seen at least two cars with enough lateral slosh to produce visible roll (and eerie groaning) with the car at rest.  I had presumed this was with a load more viscous than Bakken crude, though. 

Someone with more experience in truck dynamics might calculate the effect of combined longitudinal and roll slosh as a resultant.

[M636C]

Euclid wrote the following post 39 minutes ago: 

 

 

I believe that the way the quoted statistic is stated, it does not just compare oil unit trains to mixed freight trains.  It compares unit oil trails to all other types of trains and finds oil unit trains more like to be derailed by track defects than are trains such as ore unit trains or coal unit trains.  So what are oil unit trains doing differently?

As to air allowing slosh, it depends on how much air there is.  I do not think that the amount of air has been convincingly established, or that it does not vary from one type of oil to another.  So the potential dynamic loading from oil slosh depends on how much air is available for the oil to displace, and how hard the slack runs in to accelerate the oil mass within the tank.

The statistics compare oil trains to all other types of train.

But unit coal trains and unit ore trains generally run on tracks that have carried such traffic for many years. Oil trains are a relatively new traffic and are likely to run on lines that have not carried unit trains before. This difference would not show up in the basic statistics, only in a more detailed study which compared both the number of defects and what the other traffic on the affected lines comprised.

The oil tank wagons are designed to carry oil and the variation in the total mass would relate to the density of the oil. In calculations regarding oil we tend to use a figure of 0.85 (so 85% of the density of water). I honestly don't know how this varies between crude oil and (say) automotive gasoline for example. But tankers tend to be designed for "light" or "heavy" oil based on the variation of flammability and in general we are talking about crude oil here, which is expected to be (but not always) less flammable than gasoline.

I would expect that the oil tanks are designed to allow for a small variation in density and also provide an allowance for expansion due to heat (to avoid pressure overflows on very hot days).

From what I've read, these tanks are designed for crude oil, and this would determine their volume and the allowance for expansion. I would expect there to be a small volume of air up near the top of the tank. If the tanks aren't cylindrical but bowed downward in the centre, this air might be in two pockets at each end. I don't believe these cars run half full or with any configuration that would provide a large surface area, because the density variation isn't that great.

One thing that surprised me when I took a course on the operation of oil tanker ships was that ships carrying crude oil used the crude oil to clean the tanks at the end of the trip by spraying the oil onto the tank sides using something like a garden water sprinkler. This worked because crude oil still contained the white spirit elements as well as the road bitumen content. At Lac Megantic, they found there was more white spirit and less bitumen than they thought.

M636C