Are railroads behind the curve on technological innovation?

[Wizlish]

schlimm

Putting aside the merits of the technical discussion, in term of clear syntax I fail to see where Euclid misrepresented your statement, since you had not used the term "meaningful" prior to your last in a series of rather snarky putdowns.

You are correct.  Original post has been modified accordingly.

[gardendance]

Wizlish, I don't think you modified your post sufficiently. I didn't see it before you modified it, but it sure seems still to be extremely condescending.

Euclid said that 1 more brake MIGHT have KEPT THE TRAIN PARKED.

Wizlish insultingly and off point several times said that it was NONSENSE ranked at the top, 1 more brake would not have stopped THE ROLLING TRAIN FROM DERAILING. "Might have helped keep it from moving" is extremely different from "definitely would not have prevented the derailment once it had already started rolling".

schlimm posted what I assume was a part of Wizlish's original sentence wording, which accused Euclid of "paraphrasing badly", and I got the impression Wizlish was trying to say Wizlish had always said "no MEANINGFUL difference".

Wizlish's apparently edited post has several paragraphs that condescendingly, in my opinion, ask Euclid to go to a lot of effort to research why Wizlish is right, that "just one more brake" would have made no difference. If those paragraphs are "modified accordingly" I shudder to think how horribly insulting they were before modification.

[Euclid]

Wizlish

 

Euclid

Certainly you are the only one who asserts that the defective handbrake made no difference.

 

 

What I meant is that the defective handbrake made no MEANINGFUL difference.  By the time the oil cars were derailing in downtown Lac Megantic, it didn't really matter whether one axle's worth of additional locomotive handbraking would have kept the train a couple of mph slower, when the speed at the point of accident was well over the critical speed that produced the derailment, generated the momentum that tore the cars open, and started the fires.  It's a bit like dropping an egg from 17 feet instead of 16 feet; there's a difference in the shock, but the egg is terminally damaged either way.

Don't limit yourself to the science in the reports.  Do the math yourself -- calculate the range of adhesion limits for wheel-to-rail and wheel-to-brake-shoe for the locomotive in question.  Calculate the resultant of train weight giving gravitational acceleration on the reported grade.  See how many braked axles would be required to keep the train at a balancing speed going downgrade (someone has already calculated what would have been necessary to keep the train from moving, and IIRC at what independent pressure the train started moving).  If you can't or won't do the calculations, ask someone knowledgeable like MC to do them or approximate them.  And then to explain them if you find that to be necessary.

 

 

 

Thanks Schlimm and Patrick Boylan for your supportive comments.

 

Wizlish,

My point was that one of the handbrakes that the engineer applied to secure the train failed to have any effect.  The task of that brake and all of the others applied was ONLY to prevent the train from beginning to move.  There was no intended affect for those handbrakes to control the train, limit its speed, or mitigate the effect of a derailment IF the train began to roll away from where it had been parked.  

I do not believe that you understand this point.  Otherwise you would not have said this:

“It's a bit like dropping an egg from 17 feet instead of 16 feet; there's a difference in the shock, but the egg is terminally damaged either way.”

Your comment is an indication that you believe that the applied handbrakes were intended to limit the problem once the train began to roll, and such a limiting would be similar to dropping an egg from 16 feet rather than from 17 feet.  Of course the egg will break when dropped from either height, so you use this analogy to assert that the one missing handbrake would not have made any difference.

The whole point of securement is only to prevent the train from running away.  To use your dropping egg analogy; DON’T DROP THE EGG IN THE FIRST PLACE. 

You say that the one missing handbrake would not have made any meaningful difference.  If, as I suggested, the one missing brake would have prevented the train from rolling away; that would have been a very “meaningful” difference.

So now, to add weight to your argument that the missing hand brake would not have made any “meaningful” difference, you have added several requirements to what would constitute a “meaningful” difference.  You have moved beyond the singular purpose of the handbrakes in preventing the runaway from beginning; to a whole range of additional purposes such as limiting the speed of the runaway, mitigating the effects of the derailment, or even preventing the derailment. 

You lecture me that I should do the math.  Nothing I have said requires any math.   

[Wizlish]

The only modifications I made originally were to fix the issue that schlimm identified.  I'll look at going back to do more.

What is being missed here -- since we are now discussing what Euclid considers the actual substance of his comments -- is that the one brake 'not applied' by the QRB failure would not have made a difference in holding the train with the independent released.  Euclid said yesterday

"However, you seem to be making the point that the total handbrake securement was so insufficient that the addition or subtraction of just one handbrake could not have made any difference."

.  Yes, that is the point, although I'm not entirely sure why using the word "seem" needed to be used.  I won't go into rehashing my comments about relative acceleration earlier in the thread, but go straight to that issue.

What irritated me, I think, about Euclid's original comments were that he kept indicating that the train might have stayed parked had the failed-QRB axle been effectively braked.  That is what I considered nonsense in the context of this discussion, and the relative uselessness of a single braked locomotive axle to change either the holding or the development of the Lac Megantic accident were what my comments about looking at science and mathematics were intended to address.  I regret not having the patience to calmly address repeated assertions that successful application of that one brake "might" have made a critical difference in not having the train run away in the first place.  I will do so now.

There is no particular 'magic' in static vs. sliding coefficients of friction.  There are limits to the amount of (static) braking force that a handbrake can apply to wheel treads, and limits to the amount of braking that the contact patch on two wheels can develop vs. the railhead.  I suspect that even if the locomotive axle in question had been fully locked by braking, it would have commenced sliding by the time the independent had fully bled off, but I do not think it is likely that a tired Tom Harding would have accomplished this even with a properly-functioning QRB setup -- the other locomotive axles he handbraked appear to have blued by slipping against their brake shoes rather than skidding their wheels.

I don't have the time to go through the reports this morning, but there is a section in one of them that indicates what the range of remaining independent pressure was when the train commenced moving -- IIRC that was still substantial at the instant the train started to roll, and you could calculate a measure of the retarding brake force for 'critical balance' between standing and rolling from that.  This is across all the axles on which the independent was applied, and in particular can be seen to be considerably greater than what one properly-applied single-axle GE handbrake linkage would provide. 

To me, that establishes that the one handbraked axle was not a critical factor in the runaway starting.  There would be a range of independent pressure where the additional more-succesful handbraking would have kept the train from rolling, but it would have been essentially metastable, and as the independent kept bleeding off, the handbraking would have been less and less likely to hold -- this being contingent on how fast the independent was bleeding off (and that rate is in the reports).  I think no one reading this discussion thinks that the 'seven brakes' Tom thought he had set would have kept the consist from rolling with no other effective brakes -- so the issue comes down to the range of additional time the "QRB-braked" axle would have afforded before the train started to move.  This could be measured in minutes ... and would only be important in this case if Tom Harding or a knowledgeable person from MMA were to have returned to the train and taken some action during that interval.  (As others have so helpfully noted now, the additional rolling retardation of that axle would not have materially reduced either the likelihood or the magnitude of the derailment accident itself, which I seem to have mistaken as a point of discussion here).

Euclid seems to be saying in one of his latter posts that he can claim there "might" have been no train movement without actually doing math to see how likely (or unlikely) that would actually be.  Mr. Boylan seems to think I was asking Euclid to do "a lot of effort to research why Wizlish is right".  There is not much math involved in showing the physics involved, and it has nothing whatsoever -- well, hardly whatsoever -- to do with whether I am "right" or not.  On the other hand, if Euclid chooses not to believe what the physics says, or doesn't believe what I said about the physics involved, he's going to have to do at least some analysis to show why he thinks the effectiveness of one additional partial handbrake application would have kept the whole train from rolling.  That doesn't have to involve complicated or precise numbers, but it does involve looking at some detailed information critically, and understanding some fundamental principles of (Newtonian) physics at least in proportion.

[dakotafred]

Newtonian physics, my aching hindquarters! Give us a break on New Year's Day, at least.

[BaltACD]

Yes, but .....

[Wizlish]

BaltACD

Yes, but .....

Hey, guys, as dakotafred said, it's New Year's.  I went back and de-snarked the egregious ad hominems; I think it would be good if we all did the same going forward. 

I begin to see the 'yes, but...' applying to me as well, so I'll spare everyone further expostulation on this subject, unless there are specific issues that others need to see addressed.

[Euclid]

wccobb

If this 83 year old dinosaur can butt in ... I just skimmed thru a condensed report (Canadian) on the Lac Megantic Accident Report.  The conclusion: not enough brakes set.  The report contains a recommendation for some kind of auxiliary brake system.

This discussion contains several very workable suggestions so please allow me to approach this from a different direction.  Its gotta be workable by the guys runnin' the trains.

I was not there at a certain Society of Automobile Engineers Annual Convention which was held shortly after WII.  The guest speaker was a high-ranking U.S. Army General who had directed logistics in that war.  He spoke of a design situation where a rod was connected to the end of a lever.  German engineering would develop an exotic system which allowed extreme pre-tensioning of this connection, but, should it break down, the machine was out of action until a factory-trained mechanic could be found and brought to make the needed adjustments.  The American design for the same connection was to drill a small cross hole near the end of the rod and install a cotter pin.  Then any farm kid could keep the machine running.  In the words of the general, that's how we won the war !!!

Works for railroads too.  KISS  Make it so them running the trains can keep the trains running.  This would suggest a second airline for the "parking brake" and a new, double chambered air brake cylinder (similar in function to that used by trucks) where one chamber applies the "service brakes" (today's air brake system) and a second system uses the "parking brake chamber" to set the "parking brakes" (Same brake rods, brake levers, brake shoes, wheels, etc.)  Tie this new system to a new valve on the locomotive similar to the straight (independent) air valve and & the knowledge of how to use it are already there.  Best yet, this "parking brake" can't leak off.  Once set, (air exhausted from the brake chamber) the spring in the parking brake chamber holds the brakes on until the engineer pumps them off. Total fail-safe and on the whole train. 

{True, the electronic blue-tooth stuff with solar panels probably could be made to work, but how does it work when the car's air reservoir is empty?  Not my first choice !!}

This old dinoraur remembers when unit coal trains begn to roll, to be unloaded at rotary dumpers.  This took special cars with special rotary knuckles & a small change in the location of an air hose.  It all got done.  Similar for moving the crude oil safely.  The second line for the parking brake need only be added to those cars moving crude oil and the locomotives dedicate to this sevice.  And it all can be done, I'm guessing, at the lowest cost of the various proposals. 

 

 

You bring up some good points.  I think that is basically the right approach to the “parking brake” for trains.  You add a second air line and a second cylinder on each car.  I believe that a compression spring under each car could play an effective role in the operation of this system.  But I also agree with what others have said about springs breaking, and being a maintenance issue.  That vulnerability certainly exists, but it has to be overcome.  It can be done.  All the cars and locomotives ride on springs.  And there are piston return springs in all car brake cylinders.

In any case, I would be careful about relying too much on springs.  And also, I would not use the springs to directly force the brakes to apply, as is done with truck brakes.

I see the addition of some new parts made as the conventional air brake system of levers, pins, rods, etc.  It would also need one additional pneumatic cylinder per car and a separate pneumatic pressure line extending throughout the train as pipe attached to each car and flexible hose couplings between each car.

The added pneumatic cylinder under each car would be pressurized by the added pneumatic “train line.”  When the cylinders are pressurized, they pull open a spring loaded locking pin.  That locking pin acts as a deadbolt to prevent the train brakes from releasing.  To insert that deadbolt, the train brakes must first be set by conventional air application.

Then, you exhaust the second train line, depressurizing the second pneumatic cylinders, thus allowing the spring to insert the lock pin.  To release the parking brake, you pump up the brake pipe, and set the automatic brakes; then you charge the second “train line,” which retracts the locking pins.  They you can release the automatic brakes.

But the need for parking brakes for freight trains will be a hard sell on pure economics.  It will require substantial new hardware on every railcar, added maintenance costs, new equipment on locomotives, etc.  This might be justifiable with hazardous trains, and so the conversion to this parking brake system might be confined to just certain rolling stock such as crude oil tank cars.

Nevertheless, cost would be very significant.  And also, the emerging need for a hazardous train parking brake might just suddenly go away by a federal decree that hazardous cars shall not be parked unattended; or they may be subject to a whole host of other new restrictions and securement rules.    

[oltmannd]

Wizlish

 

 

oltmannd

The only point I was trying to make was that if my mouse can talk to my computer, there's no reason one freight car can't talk to the next via radio.

 

 

But the situation is more complex than that.  If 'one freight car can't talk to the next by radio', for whatever reason, in a system relying on short-range intercar links, it can't talk to any car from there on back in the train's consist, either.  If that doesn't make you think about a certain angle cock on a certain New Haven train going to Washington...

Dumb radio links are also, as noted, indiscriminate about what they are actually communicating with.  Imagine you have two mice communicating with the same receiver on your computer.  And you are running a safety-critical application on that computer.  Now hand the mice to two separate people doing radically different things...

 

There are many ways to slay this dragon.  For one, have the car inspector tell each car who it's neighbor is when lacing up the train. 

[oltmannd]

Euclid

What is lacking in the ECP wire trainline?  I recall reading that there were problems in getting sufficiently robust connectors.

That's a biggie!

[tree68]

oltmannd

There are many ways to slay this dragon.  For one, have the car inspector tell each car who it's neighbor is when lacing up the train. 

Which is OK for through trains.  Who handles the pickups and setouts out on the road?  A westbound CSX manifest routinely picks up 10-20 cars at Utica, sometimes at oh-dark-thirty.  Someone has to ensure that all of those cars are configured, and let the existing car in the consist know who its new neighbor is.  

Likewise, when making a drop, the remaining car in the consist has to be informed that it's neighbor isn't there any more.

[BaltACD]

One reality that outsiders overlook when insisting they have the 'great idea' when it comes to railroading in the field - human requirements to hook up, configure or maintain the equipment in the 'great idea'.

Form the railroad perspective today, if you have to hire additional personnel to specifically care for the 'new idea' in addition to the personel that already exist - the idea is a non-starter for industry wide adoption.

Railroads have been squeezing costs out of their operations for generations - anything that increases overall costs will not be adopted industry wide - unless mandated by law or FRA rule making.

The industry will test many ideas - they adopt very view as they increase costs during testing and thus run up a very big red flag.

[schlimm]

BaltACD

Form the railroad perspective today, if you have to hire additional personnel to specifically care for the 'new idea' in addition to the personel that already exist - the idea is a non-starter for industry wide adoption.

 

But forunately was not the case when computers and IT came on the scene?   Or were the rails slow/;ate to adopt that technology, given that new staff had to be hired?

[tree68]

schlimm

given that new staff had to be hired?

It's been mentioned here in one thread or another that the railroads were, in fact, early adopters of computer technology.  I would opine that they saw that punched cards, printed lists, and bits and bytes over wires would actually eliminate people (and thus expense).  Yes, they had to hire new staff, but those were different positions - which essentially replaced many other jobs.

This would be roughly analogous to the Diesel mechanics and electricians that had to be hired when Diesels replaced steam.  Granted, many employees likely made the jump from one skill set to another when the jobs were available, but in the end, the workforce was much smaller.

[erikem]

schlimm

BaltACD

Form the railroad perspective today, if you have to hire additional personnel to specifically care for the 'new idea' in addition to the personel that already exist - the idea is a non-starter for industry wide adoption.

 

 

 

But forunately was not the case when computers and IT came on the scene?   Or were the rails slow/;ate to adopt that technology, given that new staff had to be hired?

Schlimm,

I strongly suggest that you find a copy of STEEL TRAILS, The Epic of the Railroads (c) 1933 by Martin D. Stevers and read the sections on how space was reserved on passenger trains and how waybills were processed. The railroads needed a large clerical staff to handle that prior to the introduction of computers and while the computers needed some extra staff for initial set-up, the ranks of the clerical staff could be reduced afterwards.

This book was one of the first to be sold through the Trains "bookstore" as a reference to the business of railroading.

- Erik

[BaltACD]

schlimm

 

BaltACD

Form the railroad perspective today, if you have to hire additional personnel to specifically care for the 'new idea' in addition to the personel that already exist - the idea is a non-starter for industry wide adoption.

 

 

 

But forunately was not the case when computers and IT came on the scene?   Or were the rails slow/;ate to adopt that technology, given that new staff had to be hired?

New staff was hired with the knowledge that more than the number hired would become unnecessary for operations after the new staff had 'completed' their activities.  Railroads used to have hundreds if not thousands employed in accounting their various functions - with the coming of computers and the successful implementation of their various programs the staffing in those accounting departments was reduced 10 to 20 fold - where at department may have had 100 employees, after computerization one or two could manage the function.

[oltmannd]

tree68

 

 

oltmannd

There are many ways to slay this dragon.  For one, have the car inspector tell each car who it's neighbor is when lacing up the train. 

 

 

Which is OK for through trains.  Who handles the pickups and setouts out on the road?  A westbound CSX manifest routinely picks up 10-20 cars at Utica, sometimes at oh-dark-thirty.  Someone has to ensure that all of those cars are configured, and let the existing car in the consist know who its new neighbor is.  

Likewise, when making a drop, the remaining car in the consist has to be informed that it's neighbor isn't there any more.

 

The RR car inventory system is another possibility. It knows the train consist and is responsible for creating wheel reports.  It's verified by line of road scanners.  It can update the head end which can then query the train.  If every car can "talk" to the car that's supposed to be it's neighbor, then you're good to go.  The conductor is ultimately responsible for the consist being correct on line of road work.

[oltmannd]

tree68

 

 

oltmannd

There are many ways to slay this dragon.  For one, have the car inspector tell each car who it's neighbor is when lacing up the train. 

 

 

Which is OK for through trains.  Who handles the pickups and setouts out on the road?  A westbound CSX manifest routinely picks up 10-20 cars at Utica, sometimes at oh-dark-thirty.  Someone has to ensure that all of those cars are configured, and let the existing car in the consist know who its new neighbor is.  

Likewise, when making a drop, the remaining car in the consist has to be informed that it's neighbor isn't there any more.

 

You HAVE to have good consists for the braking algorithm in PTC....

[schlimm]

erikem

The railroads needed a large clerical staff to handle that prior to the introduction of computers and while the computers needed some extra staff for initial set-up, the ranks of the clerical staff could be reduced afterwards.

From what you and BaltACD answered, it sounds similar to what I experienced (before I went back to grad school) in management as a buyer at Marshall Fields in Chicago.  We had a huge number of full-time clericals working in order checking, and accounts payable and receivable, with a cumbersome system, all on paper.   While I was there, the switchover to IT started (with great resistance from some of the old-timers and deadwood in management).  After it was completed, the number of clericals was ~5% of what it had been, but in that roughly 5-year transition, expenses and errors went up (again to the delight of the naysayers).

[BaltACD]

oltmannd

 

tree68 

oltmannd

There are many ways to slay this dragon.  For one, have the car inspector tell each car who it's neighbor is when lacing up the train.  

Which is OK for through trains.  Who handles the pickups and setouts out on the road?  A westbound CSX manifest routinely picks up 10-20 cars at Utica, sometimes at oh-dark-thirty.  Someone has to ensure that all of those cars are configured, and let the existing car in the consist know who its new neighbor is.  

Likewise, when making a drop, the remaining car in the consist has to be informed that it's neighbor isn't there any more. 

You HAVE to have good consists for the braking algorithm in PTC....

And before the 1st time the brakes are used in 'anger'.

[erikem]

schlimm

 From what you and BaltACD answered, it sounds similar to what I experienced (before I went back to grad school) in management as a buyer at Marshall Fields in Chicago.  We had a huge number of full-time clericals working in order checking, and accounts payable and receivable, with a cumbersome system, all on paper.   While I was there, the switchover to IT started (with great resistance from some of the old-timers and deadwood in management).  After it was completed, the number of clericals was ~5% of what it had been, but in that roughly 5-year transition, expenses and errors went up (again to the delight of the naysayers).

 

I would guess that the experience of the RR's was similar to what you observed at Marshall Fields. IIRC, the RR's employed about one million at the start of WW2, all those people had to have been doing something. Reading the Stevers book was a bit of an eye opener, having never really thought through what was involved in the reservations process in the pre-computer era.

In this case, the clerical staff was a major cost center, so the decision to install IT was driven more by cutting costs than operational improvement (e.g. electopneumatic brakes on reight trains).

- Erik