I've read that air may bleed from an air brake system under certain circumstances which could lead to a runaway (Lac Megantic). How would this happen? My understanding is that a sudden drop in line pressure would cause the pressurized air in each reservoir (under each car) to activate the car's brakes, and that the brakes would remain activated until the line pressure is once again raised. Each car's brakes operate independently of the others.. i.e. if car 42 has a leak in its line from reservoir to brake shoe then that leak would not affect car 85 which had an air tight line.. and so on. So how would a train that was placed into emergency begin to roll on its own?..what am I not understanding?
The weak spot in the train line is the glad hands and the gaskets therein. Most crew members have one or two such gaskets in their possesion.
If there's a good air source, such leakage normally doesn't amount to anything significant. It's slow enough that it won't usually apply the brakes.
OTOH, one facet of brake tests is leakage - if it's too high it can cause the brakes to apply (or apply further, if there's a set on), which can cause problems on a long downgrade as the brakes keep setting harder...
Larry Resident Microferroequinologist (at least at my house) Everyone goes home; Safety begins with you My Opinion. Standard Disclaimers Apply. No Expiration Date Come ride the rails with me! There's one thing about humility - the moment you think you've got it, you've lost it...
A car's emergency brake resvervoir assists in the recharging of the system by putting a little bit of air into the brake pipe during a normal release.
In a normal brake application, if the valve is leaking, the car can do this on it's own causing the ones beside it to sense a rise in pressure and release. This goes on down the whole train.
There shouldn't be anything in that reservoir if the train is stopped in emergency.
10000 feet and no dynamics? Today is going to be a good day ...
Ulrich I've read that air may bleed from an air brake system under certain circumstances which could lead to a runaway (Lac Megantic). How would this happen?
I've read that air may bleed from an air brake system under certain circumstances which could lead to a runaway (Lac Megantic). How would this happen?
At Lac Megantic the train was left with the Automatic (train air brake) released. While setting the automatic is indeed accomplished by reducing brake pipe pressure it turns out that you need a reduction rate of at least 3 PSI per minute to do so. Understanding why requires a bit more of a detailed explanation of how the car control valve works internally.
The air brake application depends on the difference in air pressures on two sides of a slide valve, the reservoir and the brake pipe. The two sides are also connected by a small (3/32 of an inch in diameter) passage through which air flows to charge the car's reservoir. The small size of this passage is also why it takes so long (~7 minutes) to fully charge a car.
Let's start with the car fully charged, both the reservoir and the brake pipe are at 95 PSI (the pressure MMA was running at). The Engineer moving the brake valve in the cab reduces the brake pipe pressure fairly quickly, so now at the car the brake pipe is at a lower pressure than the reservoir. This difference in pressure forces the slide valve over, which blocks the 3/32'' passage while also exposing the pipe to the car's brake cylinder. When the Engineer moves the brake valve to release this process is reversed, allowing the car's reservoir to be recharged from the brake pipe.
But, and here's the kicker (pun intended), if the brake pipe pressure is reduced at a very slow rate then air will simply flow out of the reservoir through the 3/32'' passage back into the brake pipe, and the pressure stays approximately the same on both sides of the slide valve. As a result the valve does not move over, no air is sent to the brake cylinder and the car's air brake stays released.
There are always many small air leaks on a train, but of note here is the effect of the air turbine on the SBU (Canadianese for EOT or FRED). The TSB found that the turbine used enough air to consistently apply the brakes on a train of 5 cars or less, but not on longer trains.
Also, the 3 PSI per minute reduction rate is only required to set the brakes intitially. Once this has happened any additional reduction, no matter how slow, will result in a stronger brake application. This is why the post-Lac Megantic Canadian rules require at least a minimum application of the automatic brake to secure unattended trains.
The independent (locomotive air brakes) and automatic brake systems are connected, and both are connected to the locomotive's main reservoir. All also suffer from myriad small leaks as one would expect from a air system out in the real world. A leak in one system, combined with a lack of MR air pressure to compensate for it will result in the slow bleeding off of both systems.
The behaviour of the air brake system at Lac-Megantic was a rare scenario that hopefully will never be repeated.
Greetings from Alberta
-an Articulate Malcontent
Ulrich I've read that air may bleed from an air brake system under certain circumstances which could lead to a runaway (Lac Megantic). How would this happen? My understanding is that a sudden drop in line pressure would cause the pressurized air in each reservoir (under each car) to activate the car's brakes, and that the brakes would remain activated until the line pressure is once again raised. Each car's brakes operate independently of the others.. i.e. if car 42 has a leak in its line from reservoir to brake shoe then that leak would not affect car 85 which had an air tight line.. and so on. So how would a train that was placed into emergency begin to roll on its own?..what am I not understanding?
Edit: Darn, SD70Dude beat me too it. With a much better explanation to boot.
The train at Lac Megantic was left with the automatic brake released. There is a small port in the control valve on the cars that allows air to flow in or out of the car's reservoir when the valve is in the release/charging position at a rate slow enough that it doesn't cause the valve to move to a braking position. When the only running locomotive shut down, it stopped supplying air to both the train's brake pipe and the locomotive's independent brake cylinders. Gradually the independent brake leaked off. Since the automatic (train) brakes had not been applied, the air in the train line leaked off at a rate slow enough that did not cause the brakes to apply. Had the automatic brake been applied, the control valve would've been in a lap position once the braking system had reached equalization. That port allowing movement of air would be closed off. Had the train brakes been set, air leaking off the train line would have set the car brakes harder, up to a certain point.
One thing that could allow a train in emergency to start to move is leakage out of the car's brake cylinder itself. You would have to have a large percentage of cars leaking off, depending on weight and grade conditions, to allow movement. I've seen strings of cars in storage for weeks with only a few that had leaked off. Still, the possible leakage is why you don't depend on air brakes alone to hold a train.
Jeff
jeffhergert Edit: Darn, SD70Dude beat me too it. With a much better explanation to boot.
I just copied and pasted what I had written on one of the old Lac-Megantic threads, had it ready to go you could say. Originally wrote it for Euclid, so had to be extra thorough.
jeffhergert One thing that could allow a train in emergency to start to move is leakage out of the car's brake cylinder itself. You would have to have a large percentage of cars leaking off, depending on weight and grade conditions, to allow movement. I've seen strings of cars in storage for weeks with only a few that had leaked off. Still, the possible leakage is why you don't depend on air brakes alone to hold a train. Jeff
Yes, a thousand times yes.
More leaks appear and they get worse as the temperature drops.
jeffhergertOne thing that could allow a train in emergency to start to move is leakage out of the car's brake cylinder itself. You would have to have a large percentage of cars leaking off, depending on weight and grade conditions, to allow movement. I've seen strings of cars in storage for weeks with only a few that had leaked off. Still, the possible leakage is why you don't depend on air brakes alone to hold a train. Jeff
As I seem to recall, those cylinders have pistons with some type of rubber seal. Would the cold stiffen and degrade those seals, allowing the emergency application air to vent and release the brakes?
You mention not depending on air brakes. In this crew change situation, are they allowed to hold the train with only an emergency application? Or are they also required to set hand brakes?
Great explanations everyone.. thank you..
traisessive1There shouldn't be anything in that reservoir if the train is stopped in emergency.
Actually, there will be - just like there will be in the service reservoir.
Either reservoir will send air toward the cylinder only until the pressure in that reservoir and the cylinder are equal. The value of the emergency reservoir is that it is still fully charged even if the service reservoir isn't (ie, if the engineer has p****d away his air).
another thing that caused it was the engine had a blown turbo and the fire department shut it down. Once the engine was shut down it dumped the air in the compressor and the train ran away and derailed. Also the crew did not tie the train down before they outlawed and only the first seven hand brakes were applied.
UlrichI've read that air may bleed from an air brake system under certain circumstances which could lead to a runaway (Lac Megantic). How would this happen?
The train that ran away into Lac Megantic was left with the automatic brake released. The independent brake was set. So the trainline/brakepipe was fully charged with all air brakes on the cars released. There is natural leakage in the trainline through all of the hose gladhand gasket connections. When the engine was shut down during the fire, trainline pressure was no longer maintained. So trainline pressure began to drop due to natural leakage.
A reduction of trainline pressure should set the automatic brakes on each car, however, if the leakage is below a certain minimum rate, trainline pressure can fall without causing the automatic brakes to set. This is what happened to that train. The trainline pressue slowly leaked off without causing any set at all. All of the car brake cylinders remained unpressurized and released. All of the car brake reservoirs remained fully charged. If their charge could have been directed to the brake cylinders, all car air brakes would have applied fully and prevented any movement. But the very slow rate of trainline leakage prevented the normal set that results from a trainline reduction.
All of this would have not mattered if sufficient handbrakes had been applied and/or if the independent brakes had not released. But the independent did release once the engine was shut down and allowed the independent set to leak off.
To expand on this topic - how are leaks detected?
The ETD will monitor and send a signal back to the head-end. When there is an air pressure warning, is that when the crew has to walk the train and find the problem? Other than separated air hoses, what are they looking for?
EuclidThe train that ran away into Lac Megantic was left with the automatic brake released. The independent brake was set. So the trainline/brakepipe was fully charged with all air brakes on the cars released. There is natural leakage in the trainline through all of the hose gladhand gasket connections.
That was not the cause of the 'leakage' that made the independent bleed off without setting the automatic. You are forgetting the very important contribution of the air turbine powering the FRED, computer-designed to use only slightly less air than would produce the necessary loss to trip the automatic brake. When the diesel engines were shut down, the turbine kept merrily spinning, and the pressure kept dropping. A bit like the famous example of the frog in the water being heated to boiling.
I believe at one point someone calculated the time that particular train would have taken to leak down to critical pressure (what was it, somewhere between 30 and 17psi?) without the FRED. It was considerably longer than the ~7hr the train would have waited before the American relief engineman arrived.
I remember hearing that the engine was left running and there was a fire happening. Someone called the FD and when they put the fire out, they turned the engine off, which led to the air brakes leaking. Not enough hand brakes had been tied down and the rolling started.
Overmod Euclid The train that ran away into Lac Megantic was left with the automatic brake released. The independent brake was set. So the trainline/brakepipe was fully charged with all air brakes on the cars released. There is natural leakage in the trainline through all of the hose gladhand gasket connections. That was not the cause of the 'leakage' that made the independent bleed off without setting the automatic. You are forgetting the very important contribution of the air turbine powering the FRED, computer-designed to use only slightly less air than would produce the necessary loss to trip the automatic brake. When the diesel engines were shut down, the turbine kept merrily spinning, and the pressure kept dropping. A bit like the famous example of the frog in the water being heated to boiling. I believe at one point someone calculated the time that particular train would have taken to leak down to critical pressure (what was it, somewhere between 30 and 17psi?) without the FRED. It was considerably longer than the ~7hr the train would have waited before the American relief engineman arrived.
Euclid The train that ran away into Lac Megantic was left with the automatic brake released. The independent brake was set. So the trainline/brakepipe was fully charged with all air brakes on the cars released. There is natural leakage in the trainline through all of the hose gladhand gasket connections.
Hmmm. I am not sure about some of that. I do recall reporting of the FRED causing the trainline pressure to drop. But there is also typical leakage in any trainline. But the point is not whether the trainline pressure fell below 17-30 psi. I don't recall how low the trainline pressure got. But the point is that it leaked so slowly that no automatic airbrake application resulted. If it had leaked a little faster, and therefore caused the car brake valves to react and start admitting reservoir pressure into the brake cylinders, it would have set the automatic brakes on each car. If trainline pressure dropped enough, it would have made a full service application on the train. I may be wrong, but I seem to recall information to the effect that the trainline leaked down to zero pressure and yet no car brakes were applied as one might expect because the rate of leakage was not high enough to cause the brake valves on the cars to react.
I don't know how or if this leakage had anything to do with the releasing of the independent brakes which had been set by the engineer. In any case, I recall in the many discussions about this wreck and the TSB report, I was struck by hearing that the trainline leaked off without applying any car air brakes. I did not think that was possible. But then we heard the explanation being that tiny orifice in each car brake valve.
So just like an emergency application needing a certain minimum rate of trainline reduction to be triggered; even a service application needs a certain minimum rate of reduction to be activated. Apparently, with the train in question, rate of leakage never achieved the necessary minimum rate of leakage to produce a service application; and of course it would have taken even a much greater rate of leakage to trigger an emergency application.
Actually in looking back through this thread, I think SD70Dude answered the intitial question posed by Ulrich.
The locomotive's main reservoir supplies the air to the train line and the independent brake cylinders. Once the engine(s) are shut down, the compressor shuts down and main reservoir pressure will also start dropping, trying to supply the train line and engine brakes, leakage, etc.
In retrospect, the train was not properly secured. However using a flawed practice (push/pull test with the independent applied) the impression was given that the train was properly secured.
Had the automatic brake been applied instead of released the train probably wouldn't have ran away on it's own. When the control valve on the cars go to "lap" after equalizing from a brake application, it closes off that charging port. Has the train line would leak down, the brakes on the cars would just set harder to a certain point.
EuclidBut the point is not whether the trainline pressure fell below 17-30 psi. I don't recall how low the trainline pressure got. But the point is that it leaked so slowly that no automatic airbrake application resulted. If it had leaked a little faster, and therefore caused the car brake valves to react and start admitting reservoir pressure into the brake cylinders, it would have set the automatic brakes on each car.
That's precisely the point of the FRED design; it is designed to use air from the trainline without inducing a brake application (inherently using the pressure-maintaining feature of the automatic brake to prevent an application).
If trainline pressure dropped enough, it would have made a full service application on the train. I may be wrong, but I seem to recall information to the effect that the trainline leaked down to zero pressure and yet no car brakes were applied as one might expect because the rate of leakage was not high enough to cause the brake valves on the cars to react.
The reason you do, in fact, recall this is because the train was left 'hanging' on the independent brakes on the consist. Independent is a straight air brake, applied with the pressure in the single trainline. When the pressure in that line bled down to a certain point, the power developed in the (locomotive) brake cylinders became inadequate to hold the train. No 'car' air brakes were ever applied, due to the quirk in the pressure maintaining mentioned above.
Had there been an absolute pressure trip on the automatic brake (as, you may recall, some of the C30-7s including I believe the one in this accident came from the factory equipped with) then the automatic would have been applied essentially in emergency when the pressure dropped through that trip valve's setpoint. I don't think anyone has argued that the pressure in the car reservoirs at that point would have been insufficient to hold the train.
I don't know how or if this leakage had anything to do with the releasing of the independent brakes which had been set by the engineer.
This is very simple once you understand that the independent brake is a straight air brake applied with air line pressure, just as the independent on many steam locomotives was applied with steam. As the pressure in the trainline bled down, the direct pressure exerted in the brake cylinders also dropped. In essence the leakage had everything to do with releasing of the independent brake, by relieving the pressure that kept the brakeshoes applied firmly enough to the locomotive wheels.
In any case, I recall in the many discussions about this wreck and the TSB report, I was struck by hearing that the trainline leaked off without applying any car air brakes. I did not think that was possible. But then we heard the explanation being that tiny orifice in each car brake valve.
You ought to read up on pressure maintaining and how it works. Does anyone 'out there' have a better source than Al Krug (which I think Euclid has read) that can explain this situation to him better?
... just like an emergency application needing a certain minimum rate of trainline reduction to be triggered; even a service application needs a certain minimum rate of reduction to be activated. Apparently, with the train in question, rate of leakage never achieved the necessary minimum rate of leakage to produce a service application; and of course it would have taken even a much greater rate of leakage to trigger an emergency application.
There are two separate things involved here: both the 'rate' at which the trainline pressure is relieved (by an application at a brake valve) and the absolute pressure drop in the trainline. There is a difference in how the mechanical system reacts -- one way I might explain it is that a service brake application is made proportionally, by reducing the pressure by xxx lb. watching the gauge, while an emergency application requires a physical trip of the related valve(s) due to imbalance of a pressure differential (e.g. through a very large opening of the brake valve or a quick pressure drop on the line side) -- look up what produces a 'dynamiting' car for a little more on how that system operates.
You are right in noting that the rate of leakage observed in the Lac Megantic situation wouldn't have produced an emergency application, either, but I don't think you quite grasp yet why it would not -- and again this is inherent in the design of the turbine in the FRED. Any discharge quick enough, at any pressure, to drop the psi a certain amount, starting at any pressure, provides the 'signal' needed to trigger the emergency.
Compare what makes the 'quick-acting' Westinghouse brake so quick-acting: what trips the brake valve is a sonic pulse sent down the trainline (in both directions from the valve producing the pulse, if in midtrain) at the effective speed of sound in compressed air. When you think about the physical factors that produce higher effective energy in such a pulse by venting air, and look at precisely how the sonic signal trips parts of the equipment on the cars, I think you will get a better understanding of how the evolved one-pipe-with-pressure-maintaining system functions.
OvermodYou are right in noting that the rate of leakage observed in the Lac Megantic situation wouldn't have produced an emergency application, either, but I don't think you quite grasp yet why it would not -- and again this is inherent in the design of the turbine in the FRED. Any discharge quick enough, at any pressure, to drop the psi a certain amount, starting at any pressure, provides the 'signal' needed to trigger the emergency. Compare what makes the 'quick-acting' Westinghouse brake so quick-acting: what trips the brake valve is a sonic pulse sent down the trainline (in both directions from the valve producing the pulse, if in midtrain) at the effective speed of sound in compressed air. When you think about the physical factors that produce higher effective energy in such a pulse by venting air, and look at precisely how the sonic signal trips parts of the equipment on the cars, I think you will get a better understanding of how the evolved one-pipe-with-pressure-maintaining system functions.
What I grasp is that the emergency application requires a minimum rate of trainline venting, or the speed of pressure drop. If the venting stays under that rate, it will only produce a service application. In the service application, all of the venting air from the entire trainline is venting from where the service application is being made. And I agree with your point that this service application depends on an absolute pressure drop. Whereas, the emergency application depends on achieving a minimum rate of trainline pressure drop.
When you exceed that required rate of pressure drop through an opening in the trainline, the control valve nearest the venting will sense the rate of pressure drop being exceeded, and that control valve will open its own vent to the trainline, and that vent will also exhaust at a rate high enough to trigger an emergency application. This second vent is completely independent of the first venting that was made to begin lowering trainline pressure. So the control valve has opened a second hole in the trainline.
Then the next control valve senses that very large rate of pressure drop, and it too then opens its vent to the trainline. One valve triggers the next, and together, they each open a vent to the trainline. So in this way, the entire trainline on 100 cars is dumped though an open vent at each car. This vents the entire trainline much quicker than a service application exhausting the trainline through only one opening. This chain reaction of each car venting a portion of the trainline is what produces the quick action.
I do not follow your reference to this chain reaction of the emergency application being a sonic pulse of energy. It is simply one control valve opening due to sensing a quick enough pressure drop; and then the next valve does the same thing when it senses the pressure drop produced by the first valve. What is sonic about that? It is purely pneumatic. It is just valves opening and letting the air out.
Our community is FREE to join. To participate you must either login or register for an account.