The smallest travel of the valve represented by curve No. 1 is a little less than 2¹⁄₂ in., and the ports are then opened only about ⁵⁄₁₆ in., and the steam is cut off at 8 in. on the backward and 6³⁄₄ in. on the forward stroke. The exhaust is opened or the steam is released during the backward stroke at 17 in., and during the forward stroke at 16⁵⁄₈. When the valve works with its greatest travel, as represented by curve 8, it travels 5 in., and opens the steam port wide at 3 in. of the backward stroke and 2¹⁄₄ in. of the forward stroke. The steam is cut off at 20³⁄₄ and 20¹⁄₂ in., and its release takes place at 23¹⁄₈ in. of each stroke. The following table gives the greatest width of opening, the point of cut-off, the point of release, and the lead for each motion-curve on the diagram. This table has been made up from the motion-curves drawn with the instrument described in answer to Question 191, on a locomotive which had been running about eighteen months and whose valve-gear consequently was considerably worn, as s indicated by the flatness of the motion-curves on each side at the point when the motion of the valve was reversed. This flatness was caused by the lost motion in the valve-gear, the pencil remaining for a time stationary when the motion was reversed and while the parts were moving from their bearings on one side to those on the other. The curves and the table therefore show the operation not of a theoretically perfect valve-gear, but are examples of actual practice, with such imperfections as are incidental to ordinary locomotives. It will be seen that the instrument shows not only what the valve-gear should, but what it actually does do, and delineates all its imperfections.

Question 195. What are the chief dimensions of the valve-gear represented in fig. 128?

Answer. The throw of eccentrics was 5 in., the steam-ports were 1¹⁄₄ in., and the exhaust-port 2³⁄₄ in. wide, the valve had ⁷⁄₈ in. outside and ¹⁄₁₆ inside lap and ¹⁄₁₆ in. lead at full stroke.

Question 196. What relation is there between the distance which the ports are opened by the valve, and its travel when worked by a link?

Answer. As explained in the answer to Question 52, the width which the steam-ports are opened by the valve for the admission of steam diminishes with the travel of the valve. This is shown very clearly by the motion-curves, and also in the above table, from both of which it will be seen that when the valve travels only 2¹⁄₂ in. the steam-ports are opened only ¹¹⁄₃₂ in. for the back stroke and ⁵⁄₁₆ for the front. With 2⁵⁄₈ travel the opening is ⁷⁄₁₆ and ¹³⁄₃₂ in. With 4 in. travel the port is opened 1¹⁄₈ and 1³⁄₃₂ in. and with 4¹⁄₄ in. travel they would be opened wide. With 4¹⁄₂ and 5 in. travel, as will be seen from the motion diagram, the ports are not only opened wide, but the valve throws “over” them, or travels beyond their inner edges.

Question 197. How is the point of cut-off affected by the link?

Answer. Changing the travel of a valve with a link has a very similar effect to that produced by eccentrics of different throw—that is, the period of admission is increased with the throw of the eccentric and that for expansion lessened. This is shown clearly in both the motion diagram and the table. With the first curve and a travel of 2¹⁄₂ in. the steam is cut off at 8 in. for the backward stroke and 6³⁄₄ in. for the front, and with 5 in. travel steam is admitted during 20⁵⁄₈ in. of the backward and 20¹⁄₂ in. of the forward stroke.

Question 198. How is the point of release or exhaust of the steam affected by the link?

Answer. As the travel increases, it is delayed until later in the stroke. Thus, with 2¹⁄₂ in. travel the steam is exhausted or released from the cylinder during the backward stroke when the piston has moved 17 in., and on the return stroke at 16⁵⁄₈ in., whereas, with 5 in. travel of the valve, the release is delayed until 23¹⁄₈ in. of the stroke. An examination of the diagram and table will show very clearly the relation of the point of release to the travel.

Question 199. How is the lead affected by the ordinary link motion?

Answer. It is increased as the travel is diminished, as is shown in the table, and also by the inclination of the curves at the top and bottom of the diagram.

Question 200. What is the cause of this change of the amount of lead?

Answer. This can be best explained by reference to fig. 129, which represents a link with very short eccentric rods. If now the centre from which the link was drawn was in the centre of the axle S, and the eccentric straps embraced the axle instead of the eccentrics, their ends c and d would each describe the same arc, a b, parallel with the centre line, x y, of the link, and the latter could then obviously be raised and lowered without moving the rocker-pin at all. But the eccentric straps being attached to the eccentrics, as shown by the dotted lines, when the rods are raised or lowered they describe arcs, c e and g h, from the centres s and t of the eccentrics, and not from the centre of the axle. When the link is raised then, the end of the upper rod obviously moves in the arc c e, and the top of the link is moved from the axle, as shown in fig. 130, a distance equal to the interval between the arc, a b, drawn from the centre of the axle, and e f, which the rod describes from the centre of its eccentric. When the link is lowered from back to mid-gear, a similar action takes place, as the end, d, fig. 130, of the lower rod describes an arc, f g, so that the whole link is thrown from the axle a distance equal to the space between the arcs described from the centre of the axle and the centres of the eccentrics. When the position of the eccentrics is reversed, as shown in fig. 131, the link is moved towards the axle, thus causing an increase of lead on the opposite side of the valve. We have employed for our illustrations very short eccentric rods, in order to make this action apparent by exaggerating it. It is obvious from the engravings that the difference in the lead is increased as the eccentric rods are shortened, and also as the distance between the points of connection of the rods with the link is increased. It will also be plain that increasing the throw of the eccentrics, that is, increasing the distance of the centres s, s, of the eccentrics from the centre S of the axle will also increase the variation in the lead in full and mid-gear.

Question 201. What is meant by the distribution of steam in the cylinder?

Answer. It means the admission and exhaust of steam to and from the cylinder in relation to the stroke of the piston or the revolution of the crank.

Question 202. What are the principal periods or elements of the distribution of steam by the slide-valve and link motion?

Answer. They are:

1. The pre-admission or lead, that is, the admission of steam into the cylinders in front of the piston before it has completed its stroke.

2. The admission of steam after the piston has commenced its stroke.

3. The expansion of steam in the cylinder.

4. The pre-release, or exhaust of steam before the piston has completed its stroke.

5. The release, or exhaust during the return stroke of the piston.

6. The compression of steam, or closing the exhaust before the piston has completed its return stroke.

Question 203. What is meant by the clearance of the piston?

Answer. It is the space between the piston and the cylinder-head when the former is at the end of the stroke. If the piston touched the cylinder-head at the end of each stroke, it would cause a concussion or “thump” which would injure these parts. Owing to the impossibility of constructing machinery with absolute accuracy, it is therefore necessary to leave a space, usually from ¹⁄₂ to ¹⁄₄ in. wide, between the piston and the cylinder-heads, so as to be certain that they will not strike each other should there be any slight inaccuracies in the length of the piston-rods, connecting-rods, frames or other parts.

Question 204. Why is it desirable to open the steam-port and admit steam at the end of the cylinder towards which the piston is moving BEFORE the latter has completed its stroke?

Answer. Because it is essential, in order to insure a good action of the steam, that the maximum cylinder pressure should be attained at the very commencement of the stroke. If the steam-port was not opened until after the piston had commenced its stroke, some appreciable time would be consumed in filling the clearance space and the steam-way with steam.[57] It is also found, especially if an engine is working at a high speed, that a slide-valve worked by the ordinary link-motion will not open the steam-port rapidly enough to enable steam of the maximum boiler pressure to fill the space after the receding piston, unless the valve begins to open the port before the piston reaches the end of its stroke.

Another advantage resulting from the pre-admission of steam consists in the smooth working of the engine at high speeds, a circumstance which reduces greatly the wear and tear of the working gear. As the piston approaches the end of its stroke, the pre-admitted steam forms a kind of elastic cushion, which is well calculated to absorb the momentum of the reciprocating parts at that instant. The pressure due to the momentum of these parts will, of course, depend upon their weight and the speed of working, increasing directly as the square of the speed. It follows from this that the lead should increase with the speed, and that it should be greatest at high speeds. As has been shown before, this condition is fully accomplished by the ordinary shifting-link motion.

Question 205. Upon what does the admission of steam into the cylinder depend?

Answer. It depends in the first place upon the opening of the throttle-valve, and the size of the pipes and passages through which it is conveyed from the boiler to the cylinder. In the second place, it depends upon the time and amount of opening of the steam-port by the valve.

Question 206. What should be the pressure of the steam in the cylinder during admission?

Answer. In order that the steam may be used to most advantage, it should be admitted and maintained in the cylinder at full boiler pressure during the whole period of admission. If the opening of either the throttle-valve or the steam-ports is not sufficient to allow the steam to flow into the cylinder at full boiler pressure, the steam is said to be wire-drawn, and much of the advantage of using it expansively as has already been explained in answer to Question 59, is then lost.

Question 207. Why is it difficult to admit and maintain steam at the full boiler pressure in the cylinder during admission?

Answer. Because it is necessary to reduce the travel of the slide-valve in order to cut off the steam “short,” or soon after the beginning of the stroke of the piston. When the travel is reduced, the valve opens the port only a small distance, so that the area of the opening is not then sufficient to allow the steam to flow into the cylinder with sufficient rapidity to fill it at full boiler pressure, especially if the engine is working at a high speed. Thus, by referring to the table given on page 216 and to the motion curves in fig. 128, it will be seen that when the steam is cut off at from ¹⁄₄ to ¹⁄₂ stroke, the port is opened for the admission of steam only from ¹⁄₄ to ¹⁄₂ inch wide. From the curves it will also be seen that the valve then acquires its maximum travel and the steam-port its greatest width of opening very soon after the piston begins its stroke; after which the port is gradually closed, so that before the steam is entirely cut off the opening is so much reduced in area that the steam cannot flow through it rapidly enough to maintain the steam at full boiler pressure in the cylinder when the engine is working at high speeds.

Question 208. What means are used to overcome this difficulty and thus admit steam at full boiler pressure when the valve is cutting off short?

Answer. In the first place, the steam-ports are made from ten to twelve times as long as they are wide, so that a narrow opening will have a comparatively large area. In the second place, by giving the valve lead, not only are the clearance space and the steam-way filled with steam when the piston begins its stroke, but the port is then open a distance equal to the lead. With the ordinary link motion, as has already been shown, this lead increases as the travel and period of admission diminish, so that the smaller the total distance that the port is opened, the greater is its opening at the beginning of the stroke. As the steam is usually cut off short when locomotives run at high speeds, it will be seen that the increased lead which is imparted to the valve by the shifting link is an advantage rather than a disadvantage. But while it is often possible in this way to secure a pressure of steam in the cylinder at the beginning of the stroke equal or nearly so to that in the boiler, yet it is almost impossible to maintain this pressure during the whole period of admission, when the steam is cut off short and the engine working at a high speed. To obviate this evil what is called the Allen valve was designed, which is represented in fig. 132. This valve has a channel or supplementary port, a a, which passes over the exhaust cavity, and has two openings, b, b′, in the valve-face. When the valve begins to admit or “take” steam at c, as shown in fig. 133, it will be seen that it also uncovers the opening b′ at e and admits steam at b′, which passes through the channel b′ a a b and enters the steam-port c at b, and in this way there is a double opening for the admission of steam. The opening b of the supplementary port is closed as the valve advances, but when this takes place the steam-port is uncovered far enough to admit all the steam that is required. This form of valve is very efficient when the travel and point of cut-off are very short. It then gives just twice as much opening as the ordinary valve for the admission of steam. This improved valve has been much used in Europe; but, although it is an American invention, has not received the attention in this country which its merit deserves.

Question 209. What is meant by the pre-release of steam?

Answer. First, as has already been explained in answer to Question 51, on the amount of inside lap; and second, on the outside lap of the valve and lead of the eccentrics; and third, on the travel of the valve. The less the inside lap, the greater the outside lap and consequent lead of the eccentrics; and the shorter the travel of the valve, the earlier will be the release. The proper amount of this pre-release depends upon the velocity of the piston and the quantity of steam to be discharged or the degree of expansion. From the motion-curves in fig. 128 it will be seen that it is a marked feature of the shifting-link motion that the pre-release occurs earlier in the stroke as the link approaches mid-gear, or as the travel of the valve diminishes. As the link is usually worked near that position when the engine is run at a high speed, it will be seen that in this respect again the link-motion is well adapted for working the slide-valves of locomotives.

Question 211. What governs the period of release?

Answer. The release like pre-release is dependent upon the amount of inside lap, the outside lap and consequent lead of the eccentrics, and the travel of the valve.

The addition of inside lap has the effect of closing the port earlier than it would be closed without, and thus shortening the period of release and also of reducing the area of the opening of the port. This will be apparent by referring to fig. 128, in which the valve had ¹⁄₁₆ in. lead. The dotted lines which represent the edges of the ports in relation to the exhaust edges of the valve are therefore drawn ¹⁄₁₆ in. from the centre line a b. If, however, there had been no inside lap, then the edges of the ports would have conformed to the line a b. It will be observed that the first curve crosses the dotted line g g′ at 15¹⁄₂ in. of the forward stroke, which is the point at which the port is closed to the exhaust, or where the period of release ends and compression begins. If there had been no lap and the line g g′ had therefore occupied the same position as a b, then the motion-curve would not have crossed it until the piston had reached 16 in. of its stroke, thus showing that the period of release had been lengthened and compression delayed. As the width of the opening of the port is represented by the distance of the motion-curve from the right hand side of the line g g′, which represents the edge of the port, it is obvious that if there had been no lap, so that the position of the line representing the edge of the port had occupied the position of a b, then the space between it and the motion-curve would have been greater, thus showing that the port would have been opened wider if there had been no inside lap. The width of the opening of the port to the exhaust is in fact always diminished by an amount equal to the inside lap.

With the same travel, increase of outside lap and lead shortens the period of release, but has no effect on the width of the opening of the port to the exhaust.

Increase of travel, with the same outside lap, lengthens the period of release and also increases the width of the opening of the port to the exhaust.

Question 212. What governs the period of compression?

Answer. As compression begins when release ends, or when the port is closed to the exhaust, it is controlled by exactly the same causes, and as the two events occur simultaneously, of course whatever shortens the period of release lengthens that of compression.

Question 213. What effect do the clearance spaces and steam-ways have upon the compression of the confined steam?

Answer. By referring to the motion-curves in fig. 128, it will be seen that the steam-port is closed by the exhaust edge of the valve, or compression begins some time before the piston reaches the end of the stroke. The result is that the remaining portion of the cylinder, through which the piston must move after the port is closed to the exhaust, is filled with steam of atmospheric pressure, or possibly a little above that pressure. As this is confined in the cylinder, it is compressed by the advance of the piston. If there was no room between it and the cylinder at the end of the stroke, then either the cylinder would be burst or the valve would lift so as to allow the compressed steam to flow back into the steam-chest. The clearance and the steam-passages, however, afford considerable room, into which the confined steam can be compressed without danger of bursting the cylinder, or of raising the slide-valve when there is steam in the steam-chest. As the clearance spaces and steam-ways must be filled with high-pressure steam at the beginning of each stroke, it must be obtained either by taking a supply of “live”[58] steam from the boiler, or by compressing into the clearance spaces the low-pressure steam that still remained in the cylinder when the port was closed to the exhaust. By the latter process, a certain quantity of steam is saved at the expense of increased back pressure. It should be borne in mind also that the total heat of the compressed steam increases with its pressure, and as this latter approaches that in the boiler, the temperature of the former must have been raised from that due to about atmospheric pressure to nearer the temperature of that in the boiler. These changes of temperature which the steam undergoes will affect the surface of the metal with which the steam is in contact during the period of compression; it follows from this, that the ends of the cylinder principally comprising the clearance spaces must acquire a higher temperature than those parts where expansion only takes place. This is an important consideration, since the fresh steam from the boiler comes first in contact with these spaces, and by touching surfaces which have thus previously been heated, as it were, by the high temperature of the compressed steam, less heat will be abstracted from the fresh steam, and therefore a less amount of water will be deposited in the cylinder.[59]

It will thus be seen that the effect of compression is to fill the clearance spaces and steam-ways with compressed steam before pre-admission begins. As already stated, this is done at the expense of back pressure in the cylinder. It must be remembered that all the energy, excepting that part which is wasted by loss of heat, friction, etc., which is consumed in compressing the confined steam, is again given out to the piston by expansion. The confined steam also acts as an elastic cushion to receive the piston, just as the steam which is admitted before the end of the stroke would if there were no compression. Compression, therefore, has the effect of saving the quantity of live steam which it would otherwise be necessary to admit before the end of the stroke to fill the clearance spaces and steam-ways and also to “cushion” the piston. As already stated, the momentum of the piston and other parts depends upon their weight and the speed at which they are working, increasing directly as the square of the speed, from which it follows that the compression should increase rapidly with the speed, and should be the greatest at high speeds. As the ports are prematurely closed to the exhaust with the shifting-link motion, and as the lead increases rapidly as the link approaches mid-gear, and the amount of compression is at the same time correspondingly augmented, it will be seen that the shifting-link motion fulfills these conditions very perfectly.

The pressure to which the confined steam will rise depends of course upon the amount of the period of compression, and also on the size of the clearance spaces. As it is possible to have such an amount of compression that it will exceed the boiler pressure, and thus raise the valve from its seat and be forced back into the steam-chest, some care must be exercised to proportion the one to the other, so that the degree of the confined steam may not be excessive.

Question 214. How can the effect of the distribution of the steam upon its action in the cylinder be determined by experiment?

Answer. As already explained in answer to Question 55, this can be done by an instrument called a steam indicator.

Question 215. What is the construction of this instrument?

Answer. The indicator now ordinarily used is the Richards indicator, the outside of which is represented in fig. 134 and a section in fig. 135. It consists of a cylinder, B, into which a piston, C, is accurately fitted, but so that it will move freely in the cylinder. The piston rod is surrounded with a spiral spring, D, the lower end of which is attached to the top of the piston, and the upper end to the cylinder cover. When steam is introduced below the piston it pushes it up in the cylinder and the spring is compressed. If there should be a vacuum below the piston, the air above it will press the piston downward and extend the spring. This latter occurs only when the indicator is used on condensing engines. Of course the distance which the piston is forced up by the steam pressure below it depends upon the amount of pressure and also on the tension of the spring; and therefore by attaching a pencil to the piston-rod so that it can mark on a moving card in front of it, a diagram will be drawn which would indicate the steam pressure, as was explained in answer to Question 55. But there are some practical difficulties in the way of doing this. It is found that if the pencil is attached directly to the piston-rod of the indicator, the distance through which they must move, in order to make the scale of the diagram sufficiently large to be clear, is so great that the momentum of the parts carries them further than the pressure of the steam alone would move them. The distance through which the pistons or instruments move, moreover, makes it impossible that the changes of pressure should be indicated simultaneously with the position of the piston; the latter must travel while the action is taking place, and thus the diagram shows changes of pressure later or more gradually than they occur.[60] To overcome these and other difficulties, the piston-rod of the indicator which we have illustrated is attached at h to the short arm of a lever, F G, and to the end of the long arm a piece, F I, is attached, which carries a pencil, J. By this means the piston has only one-fourth of the motion that it imparts to the pencil, so that the momentum of the moving parts is comparatively slight. If the pencil was attached directly to the end of the lever, it is obvious that it would move in the arc of a circle, and that this would be a source of error in the diagram. To avoid this the pencil is attached to what is called a “parallel motion.” This consists of a coupling-rod, F I, which connects the ends of two levers, F G and I H. The centre of the rod F I, to which the pencil is attached, will with this arrangement move in a straight line. The levers and all the parts are of course all made as light as possible, so that their weight will have little effect on the motion of the indicator piston.

The paper or card on which the diagram is drawn is wrapped around a brass cylinder, A A. This cylinder is made to revolve part of the way around by a strong twine, a b, which is wrapped around a pulley, b, at the bottom of the cylinder. The twine is attached to a lever, similar to that shown in fig. 30, which receives a reciprocating motion from the piston of the engine. The twine can of course move the cylinder only in one direction, and therefore a coiled spring similar to a clock spring is placed inside of the cylinder to draw it back when the twine is relaxed. In this way the paper cylinder or drum receives a part of a revolution at each stroke of the piston, and moves simultaneously with it. This drum is used instead of a flat card, on account of the practical difficulties of employing the latter. The motion of the paper on this drum will, however, be exactly the same in relation to the pencil as the motion of a flat card would be.

The method of attaching an indicator to a locomotive is represented in fig. 136a. It will be seen from this that it is placed over the center of the steam chest and connected to each end of the cylinder with ³⁄₄-inch pipes. A globe valve was in the case represented placed on each side of the indicator, so that it could be put into communication with either end of the cylinder, or could be completely shut off from both. A better plan, however, is to have a three-way cock at the point where the horizontal pipe connects with the vertical one leading to the indicator, as the passages in a three-way cock are more direct than those in globe valves. The arrangement of the levers for giving motion to the indicator drum, and of the seat, which is very requisite for the experimenter, will be readily understood from the engraving without further explanation. It is thought by some engineers that the indicator should be applied as near to each end of the cylinder as possible. It is believed, however, that if the pipes, cocks, and their connections are made large enough so as not to impede the motion of the steam, no appreciable error will arise from the method illustrated in fig. 136a.

Question 216. What should be the form of an indicator diagram, if the steam is distributed by a link motion so as to produce the best practicable action in the cylinders?

Answer. It should approximate to that shown in fig. 136b. In this diagram the vertical lines represent inches of the stroke, and the scale on the left the steam pressure in pounds per square inch. The atmospheric and vacuum lines are also indicated, as already explained in answer to Question 55. The points at which the different periods of the distribution begin are indicated by the letters a, b, c, d, e and f. These are in the order in which they occur: a, pre-admission; b, admission; c, expansion; d, pre-release; e, release; and f, compression. The lines forming the outline of the diagram will be designated for convenience of description as follows:

The line from a to b, the admission line.

The line from b to c, the steam line.

The line from c to d, the line or curve of expansion.

The line from d to e, the exhaust line.

The line from e to f, the line of back pressure.

The line from f to a, the line or curve of compression.

The diagram represents a distribution of steam produced by a valve having ⁷⁄₈ in. outside and ¹⁄₁₆ inside lap, and operated by the link motion represented in fig. 103. The eccentrics have 5 in. throw, and the steam-ports are 1¹⁄₄ and the exhaust 2³⁄₄ in. wide. The valve as shown by the diagram is cutting off at 8 in., or one-third of the stroke. Pre-admission begins when the piston still has ¹⁄₂ in. to move before reaching the end of its stroke. Admission of course begins with the stroke, expansion at 8 in., pre-release at 18¹⁄₂ in., release at the end of the stroke, and compression at 17¹⁄₂ in. of the return stroke. The valve is supposed to be set without any lead, or “line and line,”[61] as it is called at full stroke. When the steam is cut off at 8 in. of the stroke, the valve has 2⁵⁄₈ in. travel and ³⁄₁₆ in. lead. The steam pressure in the boiler is supposed to be 100 pounds above the atmosphere. Of course, when the valve cuts off at different points of the stroke, the periods of distribution will be somewhat changed; but from the above diagram the principal features of a good distribution can be explained.

These are: First, that the steam pressure should rise rapidly during the period of pre-admission, so that there will be full boiler pressure in the cylinder at the beginning of the stroke. When this occurs, the pre-admission line will rise from a to b, to such a point at b which will indicate full boiler pressure in the cylinder. The same pressure should then be maintained in the cylinder during the whole period of admission, and the admission line from b to c should therefore be a straight horizontal line, as shown in fig. 136b. When expansion begins, the pressure will fall, as was explained in answer to Question 55. The expansion line should approximate a hyperbolic curve, but if there is much loss of heat by radiation or other causes, the diagram will fall considerably below the theoretical curve. With cylinders well protected and with dry steam the expansion line will fall slightly below a hyperbolic curve at the beginning of the period of expansion, and rise above it during the latter part of the same period. The reason of this is that the cylinder is heated by the admission of live steam of comparatively high pressure and temperature, so that, when the pressure becomes reduced by expansion, a part of the water which is condensed in the cylinder will be re-evaporated by the heat in the latter. From the point of the pre-release, d, to the end of the stroke, e, the exhaust line should fall rapidly, so that there will be no pressure behind the piston during its return stroke. To explain the theoretical form of the exhaust line would lead us into a very abstruse discussion, which would be out of place here. It will be sufficient for our purpose to call attention to the fact that the pre-release should allow all the steam in the cylinder to escape before the piston reaches the end of the stroke, so that the back pressure during the return stroke may be as low as possible. It is, however, only at comparatively slow speeds that the steam in locomotive cylinders escapes during the period of pre-release, so that the back pressure is reduced to that of the atmosphere. It is necessary in locomotives, as has already been explained, to contract the area of the blast orifices or exhaust nozzles, in order to stimulate the draft through the fire, so that the steam cannot escape with sufficient rapidity to reduce the back pressure to that of the atmosphere if the engine is running fast. Of course every pound of back pressure on the piston is so much loss of energy, and a reduction of the amount of work done by the engine; but it is a sacrifice which must be made in order to be able to generate the requisite quantity of steam. In studying the distribution of steam, however, every effort should be made to reduce the back pressure as much as is practicable, and yet maintain a sufficient supply of steam, and therefore the line of back pressure should conform as closely as possible to the atmospheric line. The compression line should be a hyperbolic curve, beginning with the period of compression. In calculating both the compression and expansion, allowance must be made for the clearance space and steam-way. In a cylinder like that illustrated in fig. 92, their contents would be about equal to that of two inches of the cylinder. Therefore, when steam is cut off at 8 in. of the stroke, instead of having a quantity of steam which will fill a cylinder 16 in. diameter and 8 in. long, we have as much as would fill a cylinder of that diameter and 10 in. long. The same thing is true of the compression. This must occur in the above example when the piston has 6¹⁄₂ in. more to move before completing its stroke. There is therefore a quantity of steam in front of it sufficient to fill a cylinder 8¹⁄₂ in. in diameter. This steam is of course compressed by the advance of the piston, and if its pressure when compression begins is the same as that of the atmosphere, then it will be 0.9 lbs. above it when the piston has only 6 in. to move and 3.2, 6.2, 10.5, 16.9, and 27.5 lbs. effective pressure when the piston has 5, 4, 3, 2 and 1 inches to move respectively, and when pre-admission begins, the pressure will have risen to 48.7 lbs. If the back pressure is above that of the atmosphere, of course the compression will be correspondingly increased. It will also be seen that, without any or with very little clearance space, the compression would at the end of each stroke rise above the boiler pressure. It being a peculiarity of the ordinary shifting-link motion that as the period of admission is reduced that of compression is lengthened, the latter becomes very excessive when the steam is cut off at less than one-third or one-fourth of the stroke.

Question 217. In what respect would a diagram made by an indicator differ from the theoretical form represented in fig. 136b?

Answer. It would be drawn with less exactness; that is, the corners instead of being sharply defined, as in fig. 136b, would be more or less rounded, as in fig. 137, and the curves and straight lines would vary somewhat from the exact mathematical form indicated in fig. 136b. The higher the speed at which the engine is working when the diagrams are taken, the greater will be the variation from the theoretical form.