On Thursday, investigators with the Transportation Safety Board of Canada provided new details about their inquiry into the fatal Feb. 4 derailment on Kicking Horse Pass near Field, B.C., and offered suggestions to regulators on how to prevent future tragedies.
According to TSB investigators, the previous crew aboard CP train 301-349 made an emergency brake application near Partridge — the last station before the train entered the Upper Spiral Tunnel — because the train was exceeding the posted speed limit of 15 miles per hour. The crew tried to use normal brake procedures to slow the train down but was unsuccessful. According to investigators, the train was traveling at 23 miles per hour when the engineer decided to make an emergency application. According to CP operating rules, any time a train is going 5 miles per hour over the speed limit on a descending grade it is considered an “uncontrolled movement and must be stopped immediately by whatever means available,” including an emergency brake application.
After stopping, the train sat on the 2.2% grade with the air brakes applied for nearly three hours. During that time, the air brakes slowly lost their grip on the parked train.
Two hours and 45 minutes after the train was stopped at Partridge, a relief crew came on duty. The new crew was on the train for about 10 minutes when it began to roll downhill. After exiting the Upper Spiral Tunnel, the train derailed and the lead locomotive ended up in the Kicking Horse River. Conductor Dylan Paradis, engineer Andrew Dockrell, and trainee Daniel Waldenberger-Bulmer were all killed.
After the derailment, investigators took the 12 grain cars that did not derail and conducted extensive tests to the air brake systems, with weather conditions similar to the night of the derailment. The tests revealed that the car’s air brake system failed to maintain constant pressure over time. “The air brake system on these cars would not provide adequate braking effectiveness to ensure the safe operation of a loaded unit grain train in a situation where the air brakes are required to remain applied for an extended duration, such as while descending a steep grade,” TSB officials write.
TSB officials have urged government regulators to review its rules regarding air brake maintenance and inspections, especially for railroads operating in mountainous territory.
“A properly functioning air brake system is of the utmost importance in mountain territory from a safety standpoint,” officials write.
The TSB has not indicated when it hopes to complete its investigation.
Kicking Horse Pass, home the CP’s iconic Spiral Tunnels and known to local railroaders as “The Big Hill,” has been the site of numerous derailments and runaways in the past century. In fact, just a month before the fatal derailment near February, a train derailed inside the Upper Spiral Tunnel.
Mr. Rausch, I believe your reported brake action is correct. The details of operation may have been redefined by the locomotive manufacturers and railroads. Dynamic brake operation with dc locomotives depend upon the rotation of the traction motors to allow the motors to act as generators. The energy produced by the generators was converted to heat by a resistor grid. This is effectively taking the energy of motion and changing it to heat without using brake shoe friction. It is dynamic in that motion is required for the brake operation to work via the regenerative effect of the traction motors. With ac locomotives above some minimum speed, you have effectively the same operation with a much more complex control system. Below some minimum speed, the braking by the locomotive without the use of the air brakes is different. It is not regenerative. In this operation, essentially no energy of motion is being converted to heat in a resistor grid. Instead, the ac power system and controls in the locomotive are sending energy to the motors to oppose the rotation. This operation is dynamic in that it is using rotating magnetic fields in the motors to control motion and not brake pad friction. I think the latter operation is what you have observed.
I recall an ad in the early days of ac locomotives where a couple of demonstrators stopped on a grade and held the train stopped without any air brake application. (I think the ad ran in TRAINS.) I don’t know if the current control software in ac locomotives will allow such operation.
Don , as you said yourself , ECP. John, With the electronic control, there would be no problem with “bottling the air”. Again if it fails it fails to brakes on. Communication between engines and car to car could be by closed loop, radio wifi or bluetooth type link . Probably more reliable than wire connections.
David: I am referring to switching cars in a yard or industrial sidings. Individual control of each car by wifi will quickly become an operational nightmare as each car must get called up individually. Keeping all those electronic receivers and related servos working trouble free will keep a small army of technicians busy. I presume each car will have its own air compressor, also activated by wifi on request. (They are not always attached to a locomotive.)
PSR may have contributed to this tragedy indirectly.
When he took over, EHH cut nearly all experienced field managers and combined several trainmaster positions, with the few remaining managers covering an overly large territory. A lot of railroading expertise was lost that way at CN, CP and CSX. When the closest manager is several hundred miles away (and many of them being inexperienced young blood, hired straight out of management school or off the street), the train crews are mostly on their own to figure out a solution when something happens on the line.
Also, PSR tend to run under-powered train (on CN, any train with more than 0.5hp per ton is considered overpowered, despite not being able to get up to track speed). Not having enough dynamics contributed to train 301-49’s overall lack of breaking capacity, which forced an emergency stop at Partridge.
Those are not direct causes of the tragedy. But they are part of the chain of events that took place that night on the Laggan sub. Just as a botched repair on a worn-out and oil-leaking MM&A C30-7 indirectly contributed to the incineration of downtown Lac Mégantic and 47 human beings in an oil train fire in 2013
Another note – some have have made reference to PSR – please educate me on how PSR caused this tragedy.
If what has been posted is the true situation, then the tragedy appears to have been caused by the crew going off as they did not follow proper procedure and set the handbrakes, as has been mentioned. Of course we do not know the whole story and maybe there are circumstances that will become clear in the future.
What I find most disturbing is that professional union railroaders apparently do not know enough about the equipment they are using to realize they are creating a dangerous situation. They should know that brakes bleed off and need to be recharged from the train line and they should know that leaving the brakes in emergency means zero pressure in the train line so the brakes can’t recharge.
Bottom line – that engineer was not qualified to operate the equipment.
David and Carl –
A good idea, but not doable. The problem is how do you regulate the brake application along a train that’s a mile (or more) long? There is always an air pressure gradient in the trainline. If you relied on straight air to counteract spring pressure, you’d have wildly varying braking power from car to car during the application.
The “Westinghouse” system give you nearly immediate and nearly uniform braking on every car.
Carl Walter, correct. Air pressure should be used to keep the brakes off not to apply them. fail to a safe position, brakes on.
Dynamic brakes will not HOLD a stopped train. On DC locos they lose effectiveness under 10 mph; on AC locos they are effective down to 1/2 mph, but will NOT stop the train. In addition, on steep grades, supplemental air braking is always required if the dynamics are unable to control the train at the proscribed speed, as was obviously the case here.
D&C: Also how would you switch freight cars in a yard? That is not a problem with trucks! Somehow you have to allow the freight cars to be free rolling for certain moves (and not just in hump yards), and bottling the air within a car will present its own perils.
The railroad brakes as designed by George Westinghouse are bass ackwards as compared to modern truck brake systems. Trucks have their parking (emergency) brakes applied by a coiled spring, and only released by air pressure. If you run out of air, you stop.
BTW, another 3000 dollar air compressor is a lot cheaper than another 3 million dollar locomotive.
A couple of thoughts.
In the US, the PTC technology could be used to backstop the rule book. It knows where the train is, how fast it is going, what the status of the airbrake system is, how long and heavy the train is, what the status of the power is. It is entirely possible for it to be programmed to tell the crew that they need to go set handbrakes in the situation the CP train found itself in.
A good ECP system could make railroading a lot safer and simpler. (I am not convinced that a “good” ECP system has been developed and tried, yet). The Westinghouse airbrake system was an amazing invention. Using a single trainline to supply both the power and control of the airbrakes and allow for braking when the train separated was genius in the 19th century But, we can do much better, now. This CP wreck was caused in part by the difficulties that come with combining the control and power in a single trainline. Separating the power and control make life a lot simpler.
Feeding the train line slowly through the feed valve on the controlling locomotive means you can’t recharge the trainline and reservoirs very fast and you don’t have the power of all the locomotive’s air compressors and main reservoirs immediately available for recharging.
Since you’d have some smarts and power on each car (a small battery and means of charging – solar, axle bearing generator or EOT style air turbine) you could do any number of things. You could just plain maintain a stopped train at stop. You could have a parking brake that would apply after the train stopped (air actuated braking with an electrical holding device). You could have wheel slip/slide detection so that you could have maximum braking available all the time regardless of a car’s load empty status (leverage automotive “anti-lock” braking components for the sensing). No more “it takes a train a mile to stop because it’s heavy” nonsense. Stopping from 60 mph with 0.20 adhesion would take about 15 seconds and 700 feet.
The industry has enough earnings these days to cover all their capital costs, plus pay a big chunk in dividends plus buy back chunks of stock. They could easily devote some of this money to ECP development and figure out how to do the transition. They didn’t take this approach with PTC and it got jammed down their throats. ECP is the next thing. Time to get busy.
My Grandmothers older brother was an engineer pre Amtrak. His road “usually” had more locomotives on their passenger trains NOT for tractive effort but to keep enough air for breaking. Had a lot of crossings and some speed restrictions on his route.
Common sense should dictate that with longer trains in sub freezing temperatures on a mountainous rail line additional locomotives need to be added. Not for power but for breaking. If that would have happened, the first crew could possibly NOT had there issue with that train. First crew wouldn’t have gone dead on hours an gotten to normal crew change point. Second crew wouldn’t have left in a bad situation and wouldn’t have perished.
Precision Scheduling is a joke.
The TSB said the next day that the air brake handle was left in the emergency position for 3 hours by the old crew. That means the air pumps on the locomotive does nothing for the train line as that line is in zero pressure in emergency status. The air brake handle has to be put in position to charge the train line to have any brakes retained on the train. The inbound crew should be put on trial and sent to jail. They killed the three men. And CP should change their management, including adding road foreman of engines to teach the crews.
The first think EHH did on the CSX racing to PSR was to fire all the Road Foreman of Engines. Likely he did that to the CP.. They all had union right so they just moved back into being a engineer, most likely.
They should have put hand brakes on the train 10% +2 cars for the number in the train. If that crew had trouble using the brake application account of ice and snow on the shoes and had not kept a light pressure on the brake shoes to warm them up to keep the ice off of them to stop at transfer of crew location, then there was more reason why they should have put hand brakes on immediately.
Of course, I was being facetious. The air brakes didn’t work so maybe we need to resort back to the older technology. (Works for Boeing, maybe it will work for trains.)
Referring to Jack Fuller’s post just below, it sounds like this was the first emergency application, because that is the procedure followed, namely setting 75% of the retainers in the HP position. I presume another part of the special instructions was followed, namely:
“Note: If there is doubt or uncertainty regarding the continued operation of the train, then contact the RTC and request to speak directly to a Trainmaster.”
The crew would be more or less obliged to follow the instructions of the Trainmaster or other officer. As I stated earlier, we don’t know if there was argument or discussion about a supplementary need for handbrakes.
In response to another query about dynamic brakes, they can be useful, but would be of limited help here. The trio of GEs was typical power for the maximum westbound 1% grades against loaded trains. But here the train was on a much steeper 2.2% downgrade. Since the bulk trains are empty eastbound there has been little need for expensive grade reduction projects as they return from the coast. The Kicking Horse Pass is just one of a number of such places on the route.
J Robert Wayman said, ” I understand that dynamic brakes need rotation to generate energy for braking and they don’t work at a stop”.
At least on modern AC locomotives, dynamic brakes do NOT need rotation or movement to brake. From personal experience, they work very quickly and effectively at very low speeds and including from a stop.
Noel Petit:
Freight car walkways have been removed from freight car roofs since the middle to late 1960’s and employees are prohibited from being on top of a freight car. The only exception is when the car is in a shop for repair. The first new freight car I saw without a walkway was NP 5300, a new 50 foot box car at Northtown Car Shop when I started on the Northern Pacific in April of 1966. Being on top of a freight car was dangerous.
Ed Burns
Retired Class 1.
CP’s Rules and Special Instructions already describe what is to be done. From CP Prairie Timetable No. 31, Laggan Sub, Nov. 28, 2012:
11.3 The train handling procedure on page 5/7, and the
following instructions in paragraphs A, B, C and D
apply to westward freight trains in which the weight
per operative brake is 100 tons or greater.
Note: All westward trains experiencing an emergency brake application beyond mile 123.5 must communicate with the on duty Trainmaster via the RTC
and be governed by their instructions.
A. Emergency brake recovery procedure – If an
emergency brake application is experienced between Stephen and mile 125.5, the train brakes
may be released and the train allowed to proceed to mile 125.5, where it must be stopped
and the brake pipe fully recharged. Trains
which are stopped between mile 125.5 and Signal 1363 Field with the train air brakes in emergency, must be governed as follows:
First Emergency Brake Application:
Before the emergency PCS is recovered, all
crew members (ie: locomotive engineer and
conductor and Trainmaster) must perform a job
briefing to discuss with each other the use of retainer valves. In the application GOI section 15,
item 14.3 and 29.3, set retaining valves to the
HP (high pressure) position on at least 75 percent of the loaded cars.
When discussing the use of retainers and/ or
hand brakes, consider train location, amount of
train on the mountain grade weather and rail
conditions and any other conditions present that
may affect the braking of that train. If abnormal
conditions such as weather or poor braking train
dictate that the application of hand brakes is
necessary to secure the train while re-charging,
then apply a hand brake on at least 75 percent
of the cars and set retaining valves to the HP
position on at least 75 percent of the loaded
cars.
Second Emergency Brake Application:
Apply retainers on 100% of the loaded cars and
40 hand brakes on the head end of train. Allow
the train to fully recharge the brake pipe.
Note: If there is doubt or uncertainty regarding
the continued operation of the train, then contact the RTC and request to speak directly to a
Trainmaster.
I think it is time to reinstate the roving brakeman. The brakeman should run along the tops of the cars and set hand brakes upon whistle orders from the locomotive. Just remember to duck in the tunnel.
For what it’s worth, Conrail’s absolute rule (and I assume it is NS’s as well, but can’t be sure) was that when a train went into emergency on either the east or west slope of the Alleghenies, hand brakes were to be IMMEDIATELY applied on the specified number of cars depending upon tonnage. I saw cases where a road foreman would arrive to render assistance even before the road crew had completed the hand brake applications. As below, I am of the opinion that the crew going off duty was ultimately responsible for the tragedy, in part due to poor training and/or poor operating practices.
My take is that retainers should be brought back into use on these long grades. They would give crew a chance to keep brakes applied while recharging brake reservoirs. Trying to set the number of hand brakes required to secure a train in that territory with ONE man is not really feasible. Narrow ledge on a mountainside in subzero temperature is unforgiving and in my mind it is not realistic to expect a single conductor to be able to apply and/or release handbrakes on over fifty cars to secure a train in the time necessary. Old fashioned fix but I do not see a better one.
One question I have. If the locomotives can start a train on a grade, can they hold a train on a grade with either their brakes or motors in reverse? Is the Tractive Effort available to start the train on the grade from a dead stop without any slack. If so, with AC motors, do the locomotives have sufficient traction to start a train on a grade, and if so why can that not stop a train? With DC motors, they would overheat and the commutator couldn’t handle the amperage. I understand that dynamic brakes need rotation to generate energy for braking and they don’t work at a stop. but if you can start a train with motors from a stop, and they can operate at stall speed. why can’t AC motors apply force to stop a train? I am open to all thoughts.
Hopefully, you all got answers to your questions. When the brakes are in emergency, there is no air going into the cars’ brake systems. The cars’ brakes are a sealed off system that slowly (20 minutes, 3 hours it doesn’t matter) leaks off and releases the brakes. You can use retainers where if you raise the brake pipe pressure the cars’ brakes don’t release. Setting the retainers is done by turning down a valve on each car. That car becomes a drag for the whole train. The retainers will leak off as well but the car’s reservoir can be recharged.
Hand Brakes, Hand Brakes, Hand Brakes – the mountain railroading I am familiar with has grades comparable to those in this incident. TTSI require hand brakes to be applied to 50% or more as necessary when a train stops on account of a Emergency Brake Application – it doesn’t matter if the application was initiated by the Engineer or the train. SECURE THE TRAIN is the first order of business – then figure out the condition of your condition.
The original crew is more to blame for this incident than the crew that rode the runaway train to their deaths.
SECURING a train on grades is a matter of PUBLIC SAFETY.
Before pointing fingers at the original crew, we need to know what specific instructions they may have received from above, either from Calgary or a local supervisor. We don’t know how experienced they may have been. We are told nothing about the radio conversations as the situation developed. And we know little about the actual train handling as it came from Calgary.
The location is a bad place for a crew change point. It was forced to be there by the emergency brake application to control the speed. That meant the original crew no longer had enough time to both recover the air and then continue for another 30-45″ down the hill to the change-off point. Since the need for a recrew was not anticipated, none would have been arranged until after the emergency application, hence the delay of several hours while the relief crew was called, came on duty, and was transported up to an awkward location.