This page covers (very briefly) a number of topics related to valves. These come under the following headings:
- Valve types
- Valve gear
- Valve ports
- Valve events
- Predicting valve events
- Valve travel
- Lap and Lead
Attention is also drawn to separate pages as follows:
- Enhanced Valve Events which includes subsections on
- Valve Design
The valves on a steam engine control the flow of steam in and out of the cylinders. Several types of valves were developed over the years, but most fall within three main catagories:
Slide valves largely used in the 19th century;
Piston valves which superseded slide valves in the 20th century;
Poppet valves similar in principle to those used in internal combustion engines.
"Valve gear" is the mechanism used to move the valve system that opens and closes the inlet and exhaust ports that let steam into and out of a locomotive's cylinders.
There are many different designs of valve gear, including the following more commonly used types:
Stephenson's valve gear was not invented by George Stephenson but by two of Robert Stephenson's employees, William Howe and William Williams (see http://www.timewarp.demon.co.uk/ned/howebiog.html). It was first applied by Robert Stephenson & Sons in 1842 and took its name from the company and not the man. Stephenson's valve gear was extensively used worldwide throughout the 19th century and through the first half of the 20th Century on Great Western Railway 2-cylinder locomotives. It is normally driven from "eccentrics" mounted on one of the drive axles and located inside the loco's frames. One advantage of Stephenson's valve gear is its variable lead which reduces at longer cut-offs and vice versa. See Wikipedia article for more details.
Walschaerts (no apostophe) valve gear invented by Belgian railway engineer Egide Walschaerts in 1844, gained near-universal acceptance in the early 20th century. It is commonly externally mounted and driven by an eccentric crank mounted on the end of the locomotive crank-pin. Walshaerts valve gear gives constrant lead, though this is true only in linear terms - see discussion below. See Wikipedia article for more details of the valve gear and an excellent animation.
Caprotti valve gear invented by Arturo Caprotti, an Italian engineer, and gained some popularity in mid-20th century European locomotive designs. Caprotti valve gear uses a rotary-drive from a gearbox mounted at the end of an eccentric crank on the end of the locomotive crank-pin. (See Wikipedia article for more information).
Valve ports are the slotted openings cast or cut into the circumference of each valve liner or sleeve at each end of the steam chest, though which high pressure steam passes on its way to the cylinder, and through which low pressure steam passes from the cylinder on its way to the blast pipe. At each end of the steam chest, a single port consisting of several adjacent slots (marked A in the illustration below) carries both admission steam from the steam chest to the cylinder and exhaust steam from the cylinder to atmosphere.
The valve slides back and forth along the machined surface between tapered edges C and D. The slots marked B are openings that allow exhaust steam to pass through the sleeve and into the exhaust passages leading to the blastpipe.
[Note: in the case of a Uniflow engines, separate admission and exhaust ports are fitted. The advantage gained is that the ports and cylinder surfaces are not subject to cyclical temperature fluctuations (and thus heat losses) caused by admission of hot high-pressure steam and the exhausting of cold low-pressure steam through the same openings. One consequential disadvantage is that the greater temperature differential between the steam inlets (at the cylinder end) and exhaust outlet (at the centre centre) results in differential expansion which can cause excess ring wear and (in the extreme) siezing of the piston in the cylinder.]
Valve Events may be defined as the four defining points in the cylinder power cycle - viz:
- Admission when the port opens allowing "live" (high pressure) steam from the steam chest to be admitted into the cylinder;
- Cut-off when the valve closes the port, cutting off the steam supply to the cylinder;
- Release when the valve opens the port allowing the expanded steam to be released through the exhaust passages and chimney (or to the "receiver" in the case of steam flow from high to low pressure cylinder in a compound locomotive); and
- Closure when the valve closes the port to prevent further release of exhaust steam.
These four points, in turn, define the four intervening periods as shown on the idealised Indicator Diagram (below) - viz:
- Steam admission during which period the steam pressure increases as it enters the cylinder and drives the piston;
- Steam expansion during which period, the steam expands inside the cylinder and drives the piston;
- Steam exhaust during which period the piston drives the steam out of the cylinder at a near-constant back-pressure;
- Steam compression during which period the steam remaining in the cylinder after the exhaust valve closes is compressed by the piston as it approaches the end of its stroke.
Valve events from Walschaerts valve gear are indeterminate because of the angularity effect of both connecting and eccentric rods, but they can be approximated by geometric construction using Zeuner, Reuleaux or Bilgram diagrams. Valve events can be more accurately determined by iterative computer programs such as:
- "perform" created by Prof Bill Hall and downloadable from this website;
- "WValve" created by Dr. Allan Wallace and downloadable from his own website;
- Two spreadsheets (one for Walschaerts simulation, the other for Stephensons) by Don Ashton and downloadable from his website.
- Several programs created by Charles Dockstader and downloadable from his website.
Charles Dockstader's website also includes a program for constructing a Zeuner diagram which provides insights into the variables and how they affect valve events.
Methods of constructing of Zeuner, Reuleaux and Bilgram diagrams is outlined at the SE Lounge website, and a step-by-step procedure for constructing a Reuleaux diagram (as used by Porta for preliminary estimates of valve events) is provided on the next page of this website.
Finally, some mathematical algorithms for estimating valve movements (ignoring angularity distortions) is presented in Appendix 1 of an instructive paper titled Predicting Locomotive Performance written by Prof. Bill Hall. The algorithms can be used for predicing valve openings to steam and exhaust at any position (angle) of the main crank.
Much has been written about the comparative trials conducted in 1925 between a GWR Castle 4-6-0 and the much larger LNER A1 Pacific 4-6-2 when the diminutive Castle walked away with all the honours, bettering the Pacific in time-keeping and (by a long margin) fuel consumption. The immediate result was the start of a programme of rebulding the A1s into A3s, having higher boiler pressure and long-travel valves which transformed their performance. Yet, the precise advantages that the use of long-travel valves confers has seldom been clearly explained.
Lengthening valve travel while keeping lap and lead unchanged, allows the port openings to be lengthened thereby increasing the area through which the steam passes on its way into and out of the cylinder and reducing the pressure drops in both directions. Or, if the port length is unchanged, longer valve travel requires the use of longer lap.
Lengthening valve travel also increases the speed that the valve travels, which aids lubrication (even though it is unlikely to achieve hydrodynamic conditions) and also increases the speed with which the valve opens and closes, thereby minimising the amount of "wiredrawing" (and triangular losses) that occur just before the point of closure, and giving more clearly defined valve events.
The terms "Lap" and "Lead" are often used together as if they are closely associated. However they describe very different phenomena as described below:
lap is the distance (measured in inches or millimeters) that a locomotive's valve overlaps its port when the valve is in its central position. Steam lap is the amount (or distance) that the steam side of the valve overlaps the port, while exhaust lap is the amount (or distance) that the exhaust side of the valve overlaps the port. Lap is a function of valve and port geometry;
lead is the distance (measured in inches or millimeters) that a locomotive's port is open when the piston is at dead centre. Lead is governed by the valve gear and its setting. Walschaerts valve gear is normally gives a fixed lead whereas Stephenson's valve gear gives a variable lead that increases as the cut-off is shortened. (See discussion below for a broader understanding of Lead.)
The illustrations below are taken from "Locomotive Valves and Valve Gear" by Yoder and Warren first published in 1921 and republished by Camden Miniature Steam.
In the US, the Combination Lever in Walschaerts valve gear is usually called the "Lap and Lead Lever" since its geometry defines the amount of lead that is given to the valve. The amount of valve movement that is derived from the Combination Lever equates to 2 x (lap + lead).
Purpose of Lead: Increasing lead is comparable to advancing the ignition on a petrol engine. With positive lead, the admission occurs before the piston reaches dead centre, thus ensuring that the steam has time to begin applying pressure to the piston as it begins it "power" stroke. High speed steam engines require a long lead whereas low speed freight engines require short or even negative lead. The analogy with petrol engines remains the same.
In the 5AT FDCs, Wardale describes lead as follows:
In practice the criterion to be satisfied by lead steam is to aid the obtaining of the least pressure drop between the steam chests and cylinders during the admission phase, and this is generally coincident with obtaining full steam chest pressure in the cylinder at dead centre. This criterion should be satisfied over the widest possible range of speeds and cut-offs. Factors on the 5AT favouring limited lead are as follows.
Extremely good internal streamlining at the valves and valve liner ports.
Minimal heat transfer from the inlet steam to the cylinder and piston surfaces.
Minimal steam leakage past the piston rings and piston rod and tail rod packings.
Low clearance volumes for the given level of internal streamlining.
Generally low cut-off working (which increases the crank rotation during which the valve is open to lead steam).
Minimal slackness in the valve gear, due to the extensive use of roller bearing joints, a very important factor in the ability of the valve gear to deliver the correct motion over a long period of time.
Factors on the 5AT favouring long lead are as follows.
The necessity to get steam chest pressure in the cylinders at the start of each stroke over the widest range of speeds and cut-offs, which includes at relatively high cut-offs when maximum cylinder power is required for acceleration.
The very high coupled wheel rotational speed at maximum train speed, limiting the time for lead steam to enter the cylinders.
In the design of the 5AT, Wardale has adopted a lead of 7.0 mm compared to 6.35 mm on the BR 5MT.
Porta also explains the purpose of lead in his "Compounding" paper as described in the "Triangular Losses" page of this website.
Purpose of Steam Lap: The longer the steam lap, the greater the distance that the valve has to travel in each direction to open the ports at each end of the valve chamber. Long lap valves therefore require "long-travel" valves, which Gresley famously applied to his original A1 Pacific locomotives after the 1925 comparative trials with a GW Castle.
The advantages of long steam lap (and therefore long valve travel) derives from the fact that the valve must travel further and therefore faster over the port, thereby delivering:
"sharper" events during which the period of partial valve opening is shorter, thereby reducing the period of choking or "wire drawing";
for any given cut-off, the length (and area) of port that is opened will be longer thereby increasing steam flow into and out of the cylinder.
In the design of the 5AT, Wardale has adopted a steam lap of 65mm thus giving the valve a mid-gear movement of 2 x (lap + lead) = 2 x (65 + 7) = 144mm compared to 98.4 mm on the BR 5MT giving a 46% increase.
The effect of exhaust lap is to:
- delay the point of steam release thereby extending the period of expansion and getting more "work" from the steam
- reduce the exhaust steam pressure (because of the extra expansion);
- advance the start of compression and thus the amount of compression;
- reduce the port opening available for the passage of exhaust steam;
- shorten the period of exhaust steam flow and thus reducing the draught on the fire.
The latter two effects can be seen as having "negative" effects on loco performance, however they can be counteracted by (a) the provision of large port opening (which are in any case necessitated by the provision of a long steam lap), and (b) a good exhaust system that generates high smokebox vacuum from the available exhaust steam.
The advanced compression resulting from the provision of exhaust lap is not necessarily a negative effect, since on a SGS design it compensates for the reduced pressure at which compression starts when the exhaust system is improved. Locomotives with improved exhausts may otherwise suffer inadequate compression and consequently require too much lead steam thereby showing significant indicator "triangular losses" at the start of the stroke.
A supplementary benefit of exhaust lap is that when combined with valve exhaust diffusers and a Kordina at the exhaust passage junction below the blast nozzles, it helps to reduce the exhaust pressure peak at the blast nozzles during release and the consequent draught peak on the fire. Wardale made this point in his book where he described his decision to introduce 5mm of exhaust lap on his modified 19D No 2644. He wrote:
During late May 1980 2644 was given an M repair …. and the valve heads were altered to give 5mm exhaust lap with a less rounded exhaust edge. This was motivated principally by the need to reduce the intensity of the draught peaks as a further aid towards improving firebed stability. However a little thought showed that there were other reasons why exhaust lap should be used. 2644 was generally worked at short cut-offs - 25% down to 15% and sometimes even less - at which the release was early, therefore exhaust lap was needed to lengthen the expansion period. During the return stroke compression started from a low pressure due to the improved exhaust, which meant that too little steam was compressed. Advancing the start of compression by means of exhaust lap was therefore also beneficial and it could be stated as a general rule that whenever the exhaust of a locomotive was improved there was a need far increase exhaust lap.
In the design of the 5AT, Wardale has adopted an exhaust lap of 18 mm compared to zero on the BR 5MT.