It has been suggested by PDN that the discussion of train braking be moved from the thread about urban legends to a new thread. The discussion contains some very good information and a link to a derailment report that has some good technical data that explains how the train lost its braking and became a run-away. The following is an attempt to explain in less technical terms.Braking requires two elements 1) energy conversion and 2) energy dissipation, the loss of either = NO BRAKE.The wheel absorbs and dissipates the heat energy to produce retardation. Brake applications that do not exceed the capacity of the wheels to absorb and radiate the generated heat will result in virtually continuous braking capability. I quote from the Report of Derailment of CSX Transportation Coal Train V986-26 at , January 2, 2000:<
I don’t mean to divert the discussion away from braking issues; however I wanted to mention one interesting aspect of this particular derailment. There was a general belief in Grafton that what kept the locomotive consist from derailing was the fact that its lead unit had radial – not rigid – trucks.
In my experience road locomotives with any of the common three axle trucks tend to stay on the rails better than the cars at high speeds. The trucks are not rigid (they have wheel set lateral and bolster lateral) even if they are not radial type, also they are the first through curves and switches and are the initial cause the rail mis-alignment and switch break-up that leads to car derailment. The only locomotives with a true “rigid truck” are older type switch engines such as EMD SW type that have wheel set lateral but not bolster lateral. The only exceptiion was a 3 axle truck used on later model Alco and some GE that had offset spacing of the axles, Uniion Pacific restricted their speed to 55 mph and still they would de-rail at switches in the yard at low speeds.
This is a good subject for a thread.
Having seen the NTSB report on the CSX derailment in Maryland, I was intrigued by the the 30 BHP/wheel figure and whether that would be affected by speed (realizing that braking effort would be inversely proportional to speed to maintain a given BHP) since convective heat transfer would increase with speed. With a temp limit of 500F, heat transfer is likely to be dominated by radiation (500F is 960R, where normal ambient is 60F or 520R, which gives 11.6 times more radiant energy at 500F than at 60F).
- Erik
Erik - That is an interesting observation, braking effort must be inversly proportional to speed to to maintain a constant BHP/wheel figure, however to do that the brake would have to be gradually reduced as speed increased. The freight car air brake cannot do that, it can be gradually increased but is direct release only. For any level of brake application, an increase in speed will also increase BHP and the heat generated at the wheel tread. I am not a thermal engineer, so must rely on someone else’s calculation of the proportion of heat that is transfered to the wheel by convection and radiation. The wheel temperature during braking is not limited to 500F, it is that wheel tread temperatures exceeding 500F contribute to rapidly decreasing brake shoe friction (fade) while causing rapidly increasing brake shoe friction material wear. Some of the data presented in the paper was taken at wheel tread temperatures above 500F and exceeding 1000F.
You will note the statement in the accident report that the dragging of the train through the “flats” with a heavy train brake application and high locomotive power is likely the event that increased wheel temperature to the point of significant brake fade (the kinetic energy that ended up as wheel heat actually came from the locomotive). Resuming the descent with wheel temperature in the “fade zone” resulted in BHP increasing with the train speed which con
I saw that, too, and found it very interesting. As I posted on here around August 18th or so, that’s exactly what I observed happening with a downgrade/ eastbound double-stack train at Horseshoe Curve on Sun., Aug. 16th - I estimated the rear helpers in about Run 6 or so, though only at around 20 MPH. But there, the grade is only about 1.8%, and ends just a few miles ahead.
- Paul North.
Paul and tleary,
I did notice the part about power and braking applied simultaneously, the report suggested that the runaway might have been less severe without the heating generated by applying power while the brakes were applied. The report also recommended that assuming working dynamic brakes when setting downhill speed limits was a bad idea - which kind of echoes the Milwaukee’s experience of their first runaway in an electrified zone.
My point about braking effort and speed being inversely proportional is based on the definition of mechanical power, i.e. force x speed. For example, at 30 MPH, 30 BHP corresponds to a braking force of 375 pounds, and at 15 MPH, 30 BHP corresponds to a braking force of 750 pounds (remembering that 1 HP equals 375 miles-pounds per hour).
My main question had to with whether there was enough convective heat transfer to allow an increase of braking horsepower per wheel with an increase in speed. The 30 BHP per wheel limit is set by how much heat can be dissipated while limiting the wheel tread temperature to 500F. Heat dissipation will be a combination of radiant heat transfer (essentially constant with speed) and convective (steel to air) heat transfer that will be proportional to speed (albeit with speed to a power less than 1). If the heat transfer coefficient increases, then the allowable braking power could be allowed to increase.
A related observation is that grade matters for operating double stacks due to a relatively high loaded weight per axle.
- Erik
Due in part to the conclusions and recommendations of this particular accident report the current FRA Rules 49 CFR 232.109 (g) states: All locomotives equipped with dynamic brakes and ordered on or after April 1 2006, or placed in service for the first time on or after October 1 2007, shall be designed to: (1) Conduct an electrical integrity test of the dynamic brake to determine if electrical current is being received at the grids on the system; and (2) Display in real time in the cab of the controlling (lead) locomotive the total train dynamic brake retarding force available on the train.
The idea for this rule goes back several years relative to other accidents where dynamic brakes were not fully operational and the train was operated at speeds beyond the capacity of the air brake to safely slow or stop the train. As a manager for Union Pacific locomotive air brakes I took the company line in opposition because of the expense of added equipment and maintenance costs without a clear benefit to safe train operations. I have limited operating experience (locomotive maintenance manager with an operating license) but a lot of experience in testing air brake and dynamic brake on critical grades. I did not need an indicating device to tell me if I was getting all of the expected dynamic brake (or air brake) performance on a grade, the end of train decelerometer indication and my “Butt” served just fine. Experienced engineers that understand their territory and varoius types of trains that they run know when an excessive air brake application is needed to get a “hold” on their train.
Union Pacific operating rules do allow a
DPman,
Thanks for the info and nice point about flange lubricators.
I’m assuming that the brake pipe reduction limitation for higher speed operation is for both ensuring adequate air pressure in the reservoirs and to limit wheel heating. The latter consideration is based on the 30 BHP per wheel limit for a steady state braking effort, the heat capacity of the wheel will allow for the limit to be exceeded for a short time if the wheel is substantially below 500F. I can see how an engineer could quickly get into trouble if he waited too long to apply brakes (train speeds up too much) or if he uses too light of a reduction (train spends too much time above the 30 BHP per wheel speed).
- Erik
P.S. This illustrates the point about more skill is needed to get a train down a hill than up a hill.
Virtually any idiot can operate a throttle…an Engineer knows how to operate the brakes.
Point well taken. Though the best engineer that is not familiar with the territory and the trains that operate on that territory can be blind sided when things don’t go as expected. Several accidents have been the result of assigning an engineer to an unfamiliar territory or to an unfamiliar train or both.
Or given bogus information on train weight as what happened on Cajon pass in 1989.
This was true of the CSX runaway as well. As I recall it - though the engineer was nominally qualified on the territory, this was only his first or second trip down those grades with a loaded coal train in many years. I don’t recall whether the NTSB issued a recommendation to prevent that or not.
Excellent discussion. Mr. Leary, thank you again for sharing your experiences and insights.
As to BaltACD’s point - Right on ! In Sept. 1970 I spent a whole day car-pooling with a Penn Central engineer and 2 other railfans to chase and photograph NKP 759 to, from, and on Horseshoe Curve with the High Iron Co. excursions. A good part of that afternoon was spent waiting with him and other PC personnel on the west/uphill side of the curve, above Burgoon Run/ Glen White Run, commenting on the other PC train and helper operations up and downgrade that day, especially the braking - that’s where I first heard that type of comment. It was just as fascinating and as informative as you can imagine. I’d give a lot to do that again today.
- Paul North.
Most engineers can deduce the weight of their trains by how they handle when under power, irrespective of what the consist weight is listed.
The infamous Southern Pacific “Duffy Street” run-away was the “Perfect Storn” of grade braking accidents. Lead locomotive and helper engineers unfamiliar with the grade and the train, one unit in the lead consist and one unit of the helper consist with inoperable dynamic brake, train loaded much heavier than the wheel report weight, and SP rules that used 1.5 operative brakes per loaded car for calculation of TPOB. And to complete the disaster, the wreck occured on top of a gasoline pipeline.
I knew the SP mechanical department manager that investigated on the scene. When he had to remove the event recorders from the lead consist wreckage, it required a cutting torch to to get them out. He and a carman were escorted by firemen in fire protection suits and foam guns but he was not informed about the pipeline full of gasoline just under has feet. He was as surprised and shocked as anyone when the pipeline burst a few days later and set off what was essentially a fuel bomb.