Air Brake Issues 101:

[Euclid]

Wizlish

 

Euclid

This seems like a lot of confusing rigmarole to assure the safe operation of a safety device as critical as a locomotive handbrake.

 

 

Without IN ANY WAY compromising your opinion that the design of the brake system is fundamentally flawed... I don't see *that* much 'rigmarole' involved with the QRB approach.  Hopefully Randy Stahl will comment specifically on what MMA operating policy calls for with respect to setting locomotive handbrakes, but the actual procedure seems fairly clear:

1) Be aware that the handbrake only applies to one axle of the locomotive, and is only intended as a parking brake for the locomotive itself, not in any sense for a train;

2) If you have to apply the handbrake with the independent applied, be sure to release the QRB valve on any locomotive so equipped -- look at the position of the brake cylinders on the truck if there's any question.

3) Be familiar enough with the QRB valve's function and construction that you know how to actuate it, how to ensure it has released (exhaust air from cylinder, brake shoes on respective axle release), and what to do if it doesn't release (Randy: what's the right set of procedures?)

4) When the QRB valve is fully released, tighten the handbrake again, and inspect the shoes for engagement again.

5) (and I'm adding this as a suggestion, not knowing whether it's part of procedure) -- if you are relying on handbrake to secure a locomotive consist from possible rollaway, release the independent and conduct a roll test after the above procedure with QRB release and retightening of handbrake is complete, to ensure that 'enough handbrake' has indeed been applied.  (This would seem common sense to me!)

How long could it possibly be, even if dog-tired and within minutes of dying on HoS, to flip a valve, listen for the air, check the shoes, and give a few more pumps or turns on the handbrake?

AGREED that this has very, very little to do with the outcome at Lac Megantic except in certain political and legal senses, so let's not go there any more.  

 

 

I believe that it is too much rigmarole, considering the potential for fatal consequences if not performed properly.  Other than that, the procedure itself is not overly complicated.

The procedure calls for listening to make sure compressed air is releasing from the brake cylinder, and looking at the brake cylinder to make sure that the piston is retracted.  If the piston is not retracted and/or no exhausting air is heard, the procedure calls for operating the QRB valve by direct hand manipulation of the operating stem.  In other words, you check to see if the QRB valve is not working, and if it is not working, you attempt to do a special workaround procedure.  If the workaround procedure does not work, you cannot rely on setting the handbrake.

In addition to the rigmarole and what is at stake in terms of safety, there is also the fact that this concern about the operation of the QRB valve is not at all intuitively obvious.   On the contrary, there is a widespread realm of operation in which handbrakes are correctly applied after air has been set, no air will be heard escaping, and the piston will remain extended.  So this very common practice and experience makes these conditions as being normal, and yet in the rare case of the QRB valve system, these conditions are supposed to signal trouble.  I think that is an interesting twist as well.

You mention the purpose of setting locomotive handbrakes as being only for securing the locomotive and not a train.  However, on the MM&A RR, the locomotive hand brakes were allowed as part of the whole train securement.  Granted, we are only talking about one handbrake being available or not.  So it is unlikely that this issue could have been decisive in the runaway cause which required many handbrakes.

But, what is interesting is that two different reasons for applying insufficient handbrakes happened to have converged here with the Lac Megantic wreck.  One was a misunderstanding of the special instructions calling for a number of handbrakes needed; and the other was lack of familiarity with the special instruction pertaining to the proper operation of the QRB Valve.  Both of these information failure modes definitely played a role in the Lac Megantic runaway. 

 

[Euclid]

BaltACD

Unfortunately, industrial design is filled with hundreds, if not thousands, of such occurances.

 

These kinds of design problems are indeed common.  They also come in lots of shades of gray.  However, I would not conclude that because there are many design problems in products, the QRB valve problem is typical and acceptable.  In my opinion, the QRB problem stands out as a glaring design defect.  It is not ever close to being typical.  It makes me wonder what happened to common sense when they came up with the idea.

 

The only answer to that question that I can see is that the designers did not recognize the problem until the related elements of the design were already locked in.  So they were blindsided by the linkage problem, but it was too late to change it.  So in true bureaucratic fashion, they produced a complex workaround to override the linkage problem.  The QRB valve was the workaround.

 

I can see how they could have overlooked the problem with the brake linkage, even though one might not expect experts to make such a mistake.  An important objective of such linkage is to have it apply the brakes no matter whether the application is powered by air or by the handbrake mechanism.  This linkage does that just fine.

 

The linkage used on freight cars does it fine too, but there is a difference.  On freight car linkage, if the air is applied and then the handbrake; both power inputs act independently on the brake linkage.  If one releases, the other continues holding.

 

However, with the QRB linkage, if air is applied and then the handbrake, both power inputs act in tandem on the brake linkage.  They depend on each other.  So, if one releases, they both release. 

 

Air can always release inadvertently by leakage, so that is why air is not allowed to be used in securing a train.  For securing a train, handbrakes are the only brake acceptable.   The problem with the QRB valve linkage is that if air is applied before the handbrake, that air is included in a way that makes it part of the securement.  Then if the air leaks off, the handbrake force also relaxes and releases.  So, fundamentally—if the air were applied first—they had air being a part of securement; and that is explicitly against the rules. 

 

That is a fatal flaw in the linkage.  Adding the QRB valve cancels out the fatal flaw, but if the QRB valve should happen to malfunction, the fatal flaw comes back.  And this may easily go undetected.

 

So I think the QRB valve is their best effort to fix the underlying flaw, but I suspect the flaw got included in the first place because they did not immediately recognize it.  I can’t see another explanation.  If there is another explanation, I would like to hear it.  There might be an interesting back story about this QRB valve application.   

 

[Randy Stahl]

Wizlish

 

Euclid

This seems like a lot of confusing rigmarole to assure the safe operation of a safety device as critical as a locomotive handbrake.

 

 

Without IN ANY WAY compromising your opinion that the design of the brake system is fundamentally flawed... I don't see *that* much 'rigmarole' involved with the QRB approach.  Hopefully Randy Stahl will comment specifically on what MMA operating policy calls for with respect to setting locomotive handbrakes, but the actual procedure seems fairly clear:

1) Be aware that the handbrake only applies to one axle of the locomotive, and is only intended as a parking brake for the locomotive itself, not in any sense for a train;

2) If you have to apply the handbrake with the independent applied, be sure to release the QRB valve on any locomotive so equipped -- look at the position of the brake cylinders on the truck if there's any question.

3) Be familiar enough with the QRB valve's function and construction that you know how to actuate it, how to ensure it has released (exhaust air from cylinder, brake shoes on respective axle release), and what to do if it doesn't release (Randy: what's the right set of procedures?)

4) When the QRB valve is fully released, tighten the handbrake again, and inspect the shoes for engagement again.

5) (and I'm adding this as a suggestion, not knowing whether it's part of procedure) -- if you are relying on handbrake to secure a locomotive consist from possible rollaway, release the independent and conduct a roll test after the above procedure with QRB release and retightening of handbrake is complete, to ensure that 'enough handbrake' has indeed been applied.  (This would seem common sense to me!)

How long could it possibly be, even if dog-tired and within minutes of dying on HoS, to flip a valve, listen for the air, check the shoes, and give a few more pumps or turns on the handbrake?

AGREED that this has very, very little to do with the outcome at Lac Megantic except in certain political and legal senses, so let's not go there any more.  

 

You pretty much summed up an operating bulletin that was issued and reissued concerning the QRB valves.

 

I may be wrong but I think the QRB style truck was implemented on trucks that needed several shock absorbers/vertical snubbers leaving no room for the cylinders to mount on the corners of the truck.

Randy

[BaltACD]

Norm48327

 

tree68

Either it sounded like a good idea at the time, or it was the only way around some design issue that the other solutions wouldn't work for.

 

 

 

 

Solve one problem create two more. Yeah, that'll work. NOT!

 

 

Unfortunately, industrial design is filled with hundreds, if not thousands, of such occurances.

[Norm48327]

tree68

Either it sounded like a good idea at the time, or it was the only way around some design issue that the other solutions wouldn't work for.

 

Solve one problem create two more. Yeah, that'll work. NOT!

[tree68]

Either it sounded like a good idea at the time, or it was the only way around some design issue that the other solutions wouldn't work for.

[Paul of Covington]

   That was pretty much my thought when I saw the diagram in the link you provided.   Why would anyone even consider such a design?

[Wizlish]

Euclid

This seems like a lot of confusing rigmarole to assure the safe operation of a safety device as critical as a locomotive handbrake.

Without IN ANY WAY compromising your opinion that the design of the brake system is fundamentally flawed... I don't see *that* much 'rigmarole' involved with the QRB approach.  Hopefully Randy Stahl will comment specifically on what MMA operating policy calls for with respect to setting locomotive handbrakes, but the actual procedure seems fairly clear:

1) Be aware that the handbrake only applies to one axle of the locomotive, and is only intended as a parking brake for the locomotive itself, not in any sense for a train;

2) If you have to apply the handbrake with the independent applied, be sure to release the QRB valve on any locomotive so equipped -- look at the position of the brake cylinders on the truck if there's any question.

3) Be familiar enough with the QRB valve's function and construction that you know how to actuate it, how to ensure it has released (exhaust air from cylinder, brake shoes on respective axle release), and what to do if it doesn't release (Randy: what's the right set of procedures?)

4) When the QRB valve is fully released, tighten the handbrake again, and inspect the shoes for engagement again.

5) (and I'm adding this as a suggestion, not knowing whether it's part of procedure) -- if you are relying on handbrake to secure a locomotive consist from possible rollaway, release the independent and conduct a roll test after the above procedure with QRB release and retightening of handbrake is complete, to ensure that 'enough handbrake' has indeed been applied.  (This would seem common sense to me!)

How long could it possibly be, even if dog-tired and within minutes of dying on HoS, to flip a valve, listen for the air, check the shoes, and give a few more pumps or turns on the handbrake?

AGREED that this has very, very little to do with the outcome at Lac Megantic except in certain political and legal senses, so let's not go there any more.  

[Euclid]

This design that requires the QRB valve strikes me as being overly risky.  It is almost surprising that it was put into production.  I wonder if they became committed to the design before realizing that the linkage was inherently flawed, and by the time they realized the problem, it was too late to abandon the design.  In any case, to make it workable, they added the QRB valve which seems like a sort of Band-Aid fix on the problem. 

The reason this brake rigging design that uses a QRB valve seems to be overly risky, because a failure of that valve could add an unacceptable amount of extra risk.  Without the QRB valve, the brake rigging caused interference between the handbrake and airbrake applications to the extent that the handbrake applications could be lost by having air first included in the application and then later released by leakage. 

That is a major danger because it is common railroad practice to apply handbrakes over brakes that are already applied with air.  It is done habitually without any hesitation.  It is only with this unique linkage associated with the underslung brake cylinders that applying a handbrake over an air brake poses the fatal danger. 

So the QRB valve is critical to safety, and yet it has many parts, and can fail.  When it fails, the failure might go unnoticed.  So, overall, it seems to me that the system is not failsafe enough.  The design seems like an accident waiting to happen.

The Lac Megantic wreck is discussed on the O-GAUGE RAILROAD forum with exceptional clarity.  I particularly like the first post by member Wyhog.  Here is the link:

   

 

ac-megantic-crude-wreck?reply=34936194200892367

Quote from member Wyhog:

 

“Also, IMHO, the cylinder QRB hand brake valves on 1970s GE locos could not have been purposely designed any better as a way to produce runaways. When I first found that and realized how it worked I was appalled. How could such a thing be approved and why would RRs accept such a design? As engineers we were never told of these valves and had to "discover" them on our own. All it takes to make one inoperable is for the QRB stub that gets activated by the hand brake chain, to get rotated 90 degrees. We had a serious head on collision when a set of helpers rolled out of Parkman, WY because of that apparatus and hit #121 head on at about 70 mph.”

 

Ultimately, the industry realized that even the QRB valve did not solve the problem because it could fail, so they added new operating instructions to check to make sure the valve is working.  This seems like a lot of confusing rigmarole to assure the safe operation of a safety device as critical as a locomotive handbrake.    

[Euclid]

.

 

[Euclid]

 

What I find most interesting about the role of the QRB valve in the Lac Megantic wreck is that it produced a handbrake failure that was entirely unrelated to the larger handbrake failure that is the more obvious cause of the wreck.  Yet both failure causes contributed independently to result in fewer handbrakes than required to secure the train.  

While the causes of these two failures are directly unrelated as executed, they are related at the highest level.  That is the level of the MM&A making sure their trainmen were properly informed about the finest details of operations which are covered in special instructions. 

1)    In the failure to set enough handbrakes:  The engineer mistakenly assumed that the independent brakes could be relied on for securement; he also did not understand what was required in proper handbrake effectiveness testing; and he also did not comply with the special instructions requiring minimum number of handbrakes.

 

2)    In the failure involving the QRB valve:   The engineer was not familiar with the special instructions procedure required to verify that the QRB valve was functioning properly.  As it happened, the valve was malfunctioning, and the failure went undetected by the engineer.  So the failure caused the later release of the handbrake associated with it. 

[Euclid]

Brake linkage with QRB valve:

http://www.tsb.gc.ca/eng/lab/rail/2013/lp1872013/Images/lp1872013-figure-04.png

Apparently, this diagrammatic illustration does not quite tell the whole story.  It appears as though setting the air brake would produce a reaction force that would work upward through the linkage and pull down on the handbrake chain.  That cannot be the case because if it pulled down on the handbrake chain, the chain would tighten and open the QRB valve.  That would in turn release the pressure applied to the air brake cylinder.

Therefore any air brake application would be self-canceling, and never fully develop.

I suspect that what the diagram omits are solid mechanical stops to prevent those two upper levers set in a “V” shape from dropping lower than shown.  That way the reaction force from an air brake application would work against the stops, and the hand brake chain would remain slack. 

[tree68]

Randy Stahl

EMD originated the underslung brakes and the QRB valve long before the GE, 6 axle "U" series. ALCo used and entirely different truck.

That's what I get for running "vintage" equipment...

[Randy Stahl]

tree68

Got it - and now I understand why the ALCO and EMD handbrake systems are superior, since the handbrake and independent work together.  Loss of either does not cause loss of the brakes.

 

EMD originated the underslung brakes and the QRB valve long before the GE, 6 axle "U" series. ALCo used and entirely different truck.

[tree68]

Got it - and now I understand why the ALCO and EMD handbrake systems are superior, since the handbrake and independent work together.  Loss of either does not cause loss of the brakes.

[Euclid]

Here is the brake rigging used on the GE C30-7 locomotive that requires the use of the QRB valve.  Fundamentally, the air cylinder and the handbrake winch work in tandem.  One or both could set the brakes.  If both set the brakes, the loss of either one will release the brakes. 

It is the air cylinder force that is likely to be lost through leakage.  So the air cylinder must be removed from the tandem linkage arrangement when a handbrake is set.  It could be removed simply by releasing the independent brakes before setting a handbrake, but then the locomotive would roll before the handbrake could be set.    

So the QRB valve is added to automatically release the air in the brake cylinder as the handbrake is tightened. If the air is not released, and if a handbrake is set, that handbrake may be effectively released due to air leakage. 

Therefore, the QRB valve is an essential safety device.  Without it, the linkage has a fatal flaw that is an accident waiting to happen.   And yet, the QRB valve is subject to mechanical failure of its own, some of which may go undetected for a while.   If the valve fails to open and not be detected by trainmen, it re-introduces the fatal flaw in the linkage.  With that, comes a lot more risk that the flaw in the linkage will cause an accident by inadequate train or locomotive securement. To remedy that problem, bulletins instruct trainmen to listen for the air releasing from the cylinder, and to check the cylinder piston position to confirm that the QRB valve has worked properly.  But these requirements are vulnerable to human error.   

If you don’t hear the escaping air, it indicates that the QRB valve is not working.  So you are told how to manually activate the QRB valve, which may or may not work.  The fine points of these bulletins never got to engineer Tom Harding.  And as it turns out, Harding’s second locomotive did have a defective QRB valve that failed to release the cylinder pressure when Harding applied the handbrake.  Then later, when the independent brakes all leaked off, the handbrake on the second locomotive was also lost.  It is just possible that the train might not have rolled, had that handbrake remained applied.  I am not saying that is what caused the runaway, but it may have played a small part in addition to the main cause of relying on air brakes to hold the train and not setting enough handbrakes. 

Brake diagram for GE C30-7 locomotive:

http://www.tsb.gc.ca/eng/lab/rail/2013/lp1872013/Images/lp1872013-figure-04.png

 

This thread is named after the false impression that the handbrake effectiveness test has been passed when it actually has failed.  Strangely, the issue with the QRB valve involves another false impression which is that a handbrake has been applied, but it really has not. 

[tree68]

Based on my experience with a 64 year old ALCO...

I want the independent set when I set the handbrake.  That, and the extra click I can get on the brake wheel with a brake club for leverage, is as tight as I'm going to set the handbrake.  

Releasing the independent isn't going to make any difference on the axle with the handbrake applied (and only one axle is applied on the RS3).  The chain might get a little tighter as the cylinder tries to retract, but the whole idea is to prevent the cylinder from retracting.  

In fact, if I fail to completely release the hand brake (ie, not turning the brake wheel enough to provide the necessary slack), it will prevent the cylinders from completely retracting.

It appears the handbrake rigging on the locomotive Ed showed operates the same way.

The same logic applies to the hand brakes on the cars.  When setting up handbrakes for parking the train for the overnight, the engineer takes a full service, so the crew member(s) pulling the handbrakes don't have to fight with the return spring in the brake cylinder(s).

I'm still having a little trouble wrapping my head around how the QRB thing works, as it seems counterintuitive.

[Euclid]

It looks like having the cylinders under the truck frames must be what requires the linkage that needs the QRB valve.   

[edblysard]

Balt, your are right, (I was going off memory of the photo instead of actually looking at it…that’s the first thing to go, huh!) the independent sure looks like it is applied. But the rest of it is still correct.

Bucky, No QRB valve, none needed.

Look at Randy’s photo to see one…

[Randy Stahl]

Here's a GE with a QRB valve.

 

http://www.railpictures.net/viewphoto.php?id=363220&nseq=32

[Paul of Covington]

   Thanks, Ed and Balt.   I see now that either the chain or the air will pull or push the piston to the right.

[Euclid]

So, as I understand it, the system in the photograph does not have a QRB valve.  And there is a different linkage system that is used with the QRB valve. 

[BaltACD]

edblysard

Paul of Covington

    Euclid, I originally got that impression about Ed's picture, but on further study I think the problem we've been talking about still applies.

   If I can ask a few questions of Ed or anyone else in the know...

   In that picture I assume the brakes are not applied.   Is this correct?

   On the independent air cylinder at the left, does the piston move to the left when the independent brake is applied?

   Does the handbrake pull the chain at the upper right?

   If these assumptions are correct, I can see that when setting the handbrake after the independent is applied, you would get little if any movement on the handbrake, and if the independent is released the chain will go slack.   Am I seeing this right? 

The independent was applied, then the handbrake applied, and the independent was released, to check the handbrake securement, the locomotive stayed put, (we are not allowed to use the independent as a sole means to secure locomotive) and the independent applied again.

Note the cylinder on the far right, the piston is retracted, indication there is no independent application.

When the independent was applied both cylinders had the pistons extended, when the independent was release, the one on the right retracted, but the one on the left cannot retract, the brake chain is holding it out.

On my carrier, the procedure is to stop the locomotive with the independent, apply the handbrake, (we have electric self-applying handbrakes on our locomotives) release the independent to ensure the handbrake is holding the locomotive secure, once assure of that, we re-apply the independent.

In the photograph, the locomotive had been sitting there for a few days, it was bad ordered for some reason I don’t remember, and we had shut it down and turned off the auto start to conserve fuel, and because it is right beside my switching lead and the noise was bugging me…the air reservoir had leaked down by the time I took the photo, but because of the design of the handbrake rigging, the handbrakes tightened, not loosened, because the cylinder on the left is trying to retract the piston, adding even further tension to the brake chain…sort of a failsafe feature, if the independent is released or leaks down, the handbrakes are tightened more.

If the handbrake is applied without or before an independent brake application, there is a slight chance some slack may be introduced into the brake chain once the independent is applied, but the independent will hold the locomotive secure, and if, after the independent is applied, it leaks down or is released the handbrake chain will tighten up, still securing the locomotive.

To give you a helping way to identify if the brakes are applied, if you see the piston sticking out of the cylinder, the brakes are on, if no piston is showing, the brakes are released.

To answer Paul’s questions in order:

No, the independent brakes are not applied, (see the above reason)

To the right, or towards the center of the truck, the piston on the left cylinder is out, held that way by the handbrake, the cylinder on the right has the piston retracted.

The handbrake chain is attached or anchored to the truck frame just above the center cylinder, runs through the pulley on the left brake beam/piston, and then to the right, where on this model, it runs through another pulley on the frame and up to the handbrake ratchet...the pull or tension is from left to right in the photo.

The opposite, if the independent is released, as in the photo, the chain tightens, applying even more force to the handbrake.

Ed -

In your picture we see 3 brake pneumatic cylinders - however this truck has 4 cylinders.  The two cylinders in the 'middle' of the truck both operate in the bellcrank for the 'weight jacker' of the middle wheel set.  The 4th cylinder is at the extreme right edge of the photo and the only thing that can be seen of it is its extended piston.  The hand brake chain goes to the snout of this piston too, and is taut which would indicate the hand brake is applied.  The piston extension of both brake cylinders is obscured by the shock absorber mounting.

[edblysard]

Paul of Covington

    Euclid, I originally got that impression about Ed's picture, but on further study I think the problem we've been talking about still applies.

   If I can ask a few questions of Ed or anyone else in the know...

   In that picture I assume the brakes are not applied.   Is this correct?

   On the independent air cylinder at the left, does the piston move to the left when the independent brake is applied?

   Does the handbrake pull the chain at the upper right?

   If these assumptions are correct, I can see that when setting the handbrake after the independent is applied, you would get little if any movement on the handbrake, and if the independent is released the chain will go slack.   Am I seeing this right? 

The independent was applied, then the handbrake applied, and the independent was released, to check the handbrake securement, the locomotive stayed put, (we are not allowed to use the independent as a sole means to secure locomotive) and the independent applied again.

Note the cylinder on the far right, the piston is retracted, indication there is no independent application.

When the independent was applied both cylinders had the pistons extended, when the independent was release, the one on the right retracted, but the one on the left cannot retract, the brake chain is holding it out.

On my carrier, the procedure is to stop the locomotive with the independent, apply the handbrake, (we have electric self-applying handbrakes on our locomotives) release the independent to ensure the handbrake is holding the locomotive secure, once assure of that, we re-apply the independent.

In the photograph, the locomotive had been sitting there for a few days, it was bad ordered for some reason I don’t remember, and we had shut it down and turned off the auto start to conserve fuel, and because it is right beside my switching lead and the noise was bugging me…the air reservoir had leaked down by the time I took the photo, but because of the design of the handbrake rigging, the handbrakes tightened, not loosened, because the cylinder on the left is trying to retract the piston, adding even further tension to the brake chain…sort of a failsafe feature, if the independent is released or leaks down, the handbrakes are tightened more.

If the handbrake is applied without or before an independent brake application, there is a slight chance some slack may be introduced into the brake chain once the independent is applied, but the independent will hold the locomotive secure, and if, after the independent is applied, it leaks down or is released the handbrake chain will tighten up, still securing the locomotive.

To give you a helping way to identify if the brakes are applied, if you see the piston sticking out of the cylinder, the brakes are on, if no piston is showing, the brakes are released.

To answer Paul’s questions in order:

No, the independent brakes are not applied, (see the above reason)

To the right, or towards the center of the truck, the piston on the left cylinder is out, held that way by the handbrake, the cylinder on the right has the piston retracted.

The handbrake chain is attached or anchored to the truck frame just above the center cylinder, runs through the pulley on the left brake beam/piston, and then to the right, where on this model, it runs through another pulley on the frame and up to the handbrake ratchet...the pull or tension is from left to right in the photo.

The opposite, if the independent is released, as in the photo, the chain tightens, applying even more force to the handbrake.

[Paul of Covington]

    Euclid, I originally got that impression about Ed's picture, but on further study I think the problem we've been talking about still applies.

   If I can ask a few questions of Ed or anyone else in the know...

   In that picture I assume the brakes are not applied.   Is this correct?

   On the independent air cylinder at the left, does the piston move to the left when the independent brake is applied?

   Does the handbrake pull the chain at the upper right?

   If these assumptions are correct, I can see that when setting the handbrake after the independent is applied, you would get little if any movement on the handbrake, and if the independent is released the chain will go slack.   Am I seeing this right? 

[Euclid]

I have been thinking about the QRB valve.  I understand the function of the QRB valve.  Ed posted a photo showing the handbrake chain relationship to the cylinder in a system where the QRB valve is not used.   I can see how that all works and that arrangement seems totally logical.  Either the brake cylinder or the handbrake winch can set the brake without one interfering with the other.  If both means were applied, either one could be released and the other would continue holding. 

However, that apparently is not the case with the alternate linkage system that requires the QRB valve.  What I am not certain about how that alternate system of linkage is configured.  Maybe a diagram of that linkage could be found.  I can see one way of building that linkage, but I don’t know if what I picture is how it is actually built.  But it would be sort of opposite the arrangement in Ed’s photo, as Ed mentioned. 

My general conclusion is that the alternate linkage serves some purpose in the basic brake cylinder location option; but it also has a fatal flaw in how it interacts with the setting of the handbrake.  Therefore to overcome that fatal flaw, the QRB valve is needed.

In the alternate linkage that I visualize, the handbrake could be set after the independent is set, BUT if the independent happens to leak off, the handbrake will be released. 

The STB report says this about the QRB valve:

“The hand brake will not tighten if the air from the R#2 brake cylinder is not exhausted. The handbrake chain will tighten and it may appear that the handbrake is set however if the R#2 brake cylinder is in the “out” position, the handbrake is not applied.”

Perhaps when they say, “if the R#2 brake cylinder is in the “out” position, the handbrake is not applied,” they mean that the handbrake is not applied in a way that it can qualify as an applied handbrake because it is subject to release if the independent brakes leak off. 

In any case, I would like to see a diagram of the brake linkage that requires the QRB valve.

[Randy Stahl]

The main reason that underslung brake cylinders went away was that the cylinders were suspended between the axles. They were subject to damage from track debris from being so close to the rail.

Most railroads that had them had specific special instructions on operating them. Its quite simple really. As you tighten the handbrake wheel you should hear the air exhausting from the QRB valve and be able to crank the brake wheel an additional amount. Another check is to visually look at the cylinders and see if the #2R piston is in or out, if the piston is out you just tightened the hand brake against the cylinder pressure. If the piston is in and the chain is tight you have a good hand brake.

 

If the chain is tight and the piston is out you can manually trip the lever on the QRB valve to release the cylinder pressure. Then you must climb back up on the locomotive and tighten the brake wheel some more.

Here's the same setup on an EMD:

http://www.railpictures.net/viewphoto.php?id=91924&nseq=163

[edblysard]

In this image, you can see where the handbrake chain attaches to the brake cylinder lever.

On the older models, if I remember right, it was all inboard, and attached to the other end of the lever.

With this design, you can tighten the handbrake with the independent applied, and if the independent leaks down, it will pull the handbrake chain tighter, the old design was almost opposite of this.

And the center cylinder is not a brake cylinder on this model, it is a weight management cylinder, it lifts the center unpowered axel up off the rail when needed.

[edblysard]

The handbrake chains pully attachment point is that cylinder’s compound lever, and if the independent is applied, that cylinder is extended and the lever at full travel, applying the hand brake will not tighten the brake shose any more to the wheel than the cylinder piston and lever travel will allow…if the cylinder bleeds down the handbrake will have slack in its chain, so exhaust that cylinder first, apply the handbrake tight, and then apply the independent.