LeChatelier's water brake hit my cue again. I haven't saved my old posts on this subject so will hit it again from the beginning. The "water brake" or "compression brake" is a form of dynamic braking that was installed on a limited number of steam locomotives operating on heavy mountain grades. The D&RGW was the primary user of this type of brake in America.
The principle of operation is to admit a small amount of superheated water from the boiler into the cylinder (I will use one cylinder for the example) just as the piston begins its return stroke (let's say the steam is in front of the piston as it begins to go forward). Since the pressure in the cylinder is nearly atmospheric (exhaust pressure) the water released from the boiler at about 200 psi is at appoximately 450 degrees f. A small amount of water will instantly flash to wet steam in the low pressure cylinder, thus filling the whole cylinder with steam at about 15 psi. As the piston moves forward it compresses the steam to a higher pressure. This is negative work that retards the motion of the piston, thus creating a braking affect on the drive wheels. When the piston reaches front dead center the highly compressed steam must be released up the stack through the valves so all the compressed energy isn't given back to the drive wheels when the piston begins to move back. All water brakes have a metering valve arrangement to admit a small amount of hot boiler water into the cylinder, usually through the valve chest and the engineer moves his reverse lever into back gear to make the valves time the port openings so all the compressed steam is bottle up in the cylinder until it reaches front dead center, then escapes through the valves up the stack.
The principle of operation is called "adiabatic compression" of steam which is exactly the opposite of "adiabatic expansion" of steam. Adiabatic expansion describes what happens when you hook an engine up, only admit steam for a small portion of the piston stroke then let it expand to a lower pressure and temperature for the rest of the stroke.
Now, there are a few practical problems with making this nice principal work in practice. First, most steam locomotives never work a cutoff much less than 30% or 40%. This means they never expand the steam more than about twice its original volume during one stroke of the piston. Steam engines though have very small clearances, which means they have enormous compression ratios -- something on the order of 20:1. This means the piston can compress atmospheric steam to a pressure and temperature far above the boiler pressure and temperature. This creates temperatures far above the working limits of steam cylinder oils and can basically burn up the engine. Second, a little bit of water will flash into a huge volume of steam at low pressures in the cylinder, thus raising the initial pressure and aggravating the high compression peak pressures and temperature problem just described. None of the brakes actually applied to steam locomotives really had a precise way of metering water into the cylinders and using the reverse lever to control braking effect was very crude because the valves could not be tuned for braking and power production at the same time.
The main reason most railroads did not use water brakes was because they were subject to much abuse on the road and caused a lot of machinery damage -- I have heard stories of narrow gauge Rio Grande engines coming off big hills into terminals with the cylinders and saddles so hot the paint was blistered off of them. I have heard similar stories about stationary condensing Unaflow steam engines doing the same thing when admission valves leaked.
As Earl described, the drifting throttles on narrow gauge D&RGW engines are more crude than LeChatelier's brake because they are just a 3/4" globe valve that lets boiler water into the valve chests -- way too crude for train control, but useful for bring a light engine down the hill without heating its tires off the drivers.
I cannot remember reading LeMassena's article, but I do think he is probably correct in whatever statements he made because he is an old time engineer that grew up around this stuff.
In answer to one of Earl's questions: One 3600 could hold back a pretty large train if the speed is high enough, just the same as diesel dynamics work better at speed than at a stall. The word "dynamic" is the key. A locomotive only has so much adhesive effect and if the braking effect required exceeds this "factor of adhesion" the drivers will simply skid and probably lock up. However, as the speed picks up the "braking horsepower" of the locomotive increases proportionately as the number of piston strokes increased per minute and its ability to hold back a train increases as well. Earl is probably right though, the real use of water brakes on the big engines was probably limited to holding tonnage in check while recharging brake cylinders and for light helper engine moves the same as on the narrow gauge. The use of dynamic brakes was not well understood on most railroads at that time. Sometime I can recite how the Milwaukee Road got into trouble with their new electrics and regenerative braking many years ago in the Saddle Mountains of Washington.
I remember suggesting modern, precise compression brakes to Steve Jackson on the D&S several years ago. I thought this might be a useful idea. Steve's response was he would rather wear out cheap brake shoes on the long downgrades than expensive locomotive machinery that would be under compression for miles on end. Good point Steve, another great idea shot full of holes.
Bill Petitjean
