Part 1 of 3: The 'pretzel logic' of remote control
(This is the first in a three-part series of articles in which the BLET takes a look at recent actions of the Federal Railroad Administration, the Association of American Railroads, and the United Transportation Union to further expand the use of remote control locomotives to main track territory.)
(In Part I of the series, we take a look at the FRA's continuing failure to affirmatively regulate remote control, and how FRA's latest action impacts the safety of the nation's railroads. A complete copy of this series is available on the BLET website at the following link: http://www.ble-t.org/pr/pdf/pretzel.pdf)
Since the BLET's release in late July of a comprehensive study of remote control locomotive ("RCL") operations authored by renowned railroad work organization expert Dr. Frederick C. Gamst and former Federal Railroad Administration ("FRA") Associate Administrator for Safety George A. Gavalla, the silence from FRA, the Association of American Railroads ("AAR") - the industry's lobbying arm - and the United Transportation Union ("UTU") has been deafening.
However, in less than a week, FRA, AAR and UTU have joined forces in an effort to sidestep the impact of the Gamst/Gavalla Report ("Report") in a way that only underscores, once again, the industry's folly in its implementation of RCL technology, FRA's continuing abdication of its statutory duty to ensure the safety of the nation's railroad industry, and UTU's role as the industry's lapdog.
Last week - and more than four months after its Final Report to Congress on the safety of RCL became overdue - FRA fired the first salvo with the public release of a September 9, 2005 letter from Associate Administrator for Safety Daniel C. Smith to AAR and to the American Short Line and Regional Railroad Association. That letter, while purporting to caution the industry about inappropriate use of RCL technology, establishes beyond any reasonable doubt the consequences of FRA's failure to affirmatively regulate RCL operations, because the agency raised more questions than it answers.
FRA acknowledged that, as early as May of 2004, it had concerns about the use of RCL outside of yard switching operations and that it had been "surprised to learn that railroads had some RCL operations on main tracks." After "an additional year of monitoring and evaluating the non-incidental movements by RCLs over main track," FRA concluded "that, given the limitations of RCL technology, non-incidental movements over main track by RCLs can pose an unacceptable safety risk and thus should be strictly limited."
FRA admitted that a "primary problem with the current state of the RCL technology is that it is inadequate to control in-train forces during heavy-haul operations," and "similarly, that the way to operate the RCL over the main track, and its variously graded conditions, without pulling the train apart, may require manipulating the remote control device in a way that is counter-intuitive to the way it was designed to be used." Thus, FRA conceded, the current industry practice of using RCL for main track movements has created "the absurd situation of allowing an RCO [Remote Control Operator] to perform all the same operations as a train service locomotive engineer, just with different equipment and significantly less training."
In describing the operational hazards created by using RCL for main track movements, FRA stated the following:
" FRA's initial concern was that the current technology might not be suitable to control in-train forces during train movements. The speed control feature on the remote control transmitter works like an automobile's cruise control. A speed is selected, and the computer will increase locomotive power until the desired speed is reached. The computer will then automatically maintain the selected speed using locomotive power and brakes. Since plus or minus fluctuations in speeds greater than 0.5 mph often occur as trains move over the main track, the independent air brake will constantly apply and release, or locomotive power will increase or decrease, causing the train slack to run in and out as the train progresses. The longer and heavier the train, the more dramatic this slack action becomes. This function is not desirable when controlling longer trains on the main track because the computer works faster than the train air brake system. For example, under certain track profiles (short downhill, uphill track configurations), the system could apply and release the brakes before the brakes fully apply on the rear of a long train. This would create a situation where the brakes would be releasing on the head end of the train at the same time they are applying on the rear end. This condition could cause excessive in-train forces."
There are several problems with FRA's description. First, it relies upon a false analogy: an automobile's cruise control does not alternately apply throttle and brake; it increases acceleration to maintain or increase speed, and relies upon gravity and the weight of the coasting vehicle to reduce speed.
Second, the Gamst/Gavalla Report established that RCOs will occasionally manipulate the beltpack controls to compensate for the actions of the speed control circuit. See Report at pp. xvii, 31, 85. Thus, this train control deficiency also is present in switching operations, a fact conveniently omitted from FRA's letter. This operational risk is a function of the technology, and exists regardless of the setting in which it is employed.
Third, and as any experienced locomotive engineers can tell you, the repeated application and release of the automatic brake in a short time frame, as described by FRA, will result in the "dribbling away" of automatic brake pipe pressure, which is a known hazard. When that point is reached, there no longer is an ability to safely control the speed of the train regardless of manual or automatic manipulation of the airbrakes.
Behind Door #2
Notwithstanding the false analogy and incomplete description of the deficiency of the RCL speed control system, FRA stated three findings that were right on point:
Based upon the above, it would appear to be a fairly straight-forward conclusion that - other than the occasional pull-out onto a main track to clear a switch - using RCL on main tracks is fraught with danger. However, in spite of this evidence, and nearly a year and a half after being blindsided by the industry's expansion of RCL to an extent not previously disclosed, FRA has chosen to continue to let the fox guard the henhouse.
Even worse, FRA is openly encouraging the industry to plunge full-speed ahead in expanding RCL to main track operations. FRA stated that "RCOs should be as knowledgeable [concerning air brakes, train handling, signal recognition, track-train dynamics, etc. all subjects associated with the fundamentals of main track operations, regardless of speed or distance] as conventional engineers. Consequently, FRA believes that RCOs should receive additional training if they operate on main tracks."
To that end, FRA advised that it will "reopen review of railroad RCO training programs where it is clear that the railroad is committed to non-incidental main line movements" to determine whether RCOs receive "the equivalent level of classroom training as that provided for conventional train service engineers," and whether they are afforded "a minimum of 120 hours of actual, documented hands-on operating experience." If there is any doubt that this is bald encouragement, FRA also stated that it will consider crediting hours already worked in main track service in satisfaction of the 120-hour requirement.
Anticipating criticism of this green light for a practice that it has already determined to be beyond the safe capacity of RCL technology, FRA further stated the following:
"At a minimum, we recommend that the following limitations should apply to all RCL movements requiring brake tests under 49 C.F.R. Part 232:
a. Locomotive consist should not exceed 3000 horsepower, utilizing no more than eight (8) axles.
b. Train length should not exceed 1000 feet (approximately 20 car lengths).
c. Train speed should not exceed 15 mph.
d. Operations should be prohibited on any grade of 0.5 percent or greater that extends for more than 14 of a mile."
Once again, however, these "recommendations" are unenforceable by FRA, irrespective of the fact that federal air brake testing requirements have been cited.
(In Part II of this series, we examine the reactions of the AAR and UTU to FRA's September 9, 2005 letter.)
© 2005 Brotherhood of Locomotive Engineers and Trainmen