Taking a similar diagram to that used for the foregoing illustrations, as in Fig. 3371. Fix on a point a near the terminal, where the total pressure is 25 pounds. As before, this point is chosen in order that the two curves may coincide there.
Any other point might have been chosen for the point of coincidence; but a point in that vicinity is generally chosen, so that the result will show the amount of power that should be obtained from the existing terminal. This point is 3.3 inches from the clearance line, and the volume of 25 pounds 996, that is, steam of that pressure has 996 times the bulk of water.
Now if we divide the distance of a from the clearance line by 996, and multiply the quotient by each of the volumes of the other pressures indicated by similar lines, the products will be the respective lengths of the lines measured from the clearance line; the desired curve passing through their other ends. Thus, the quotient of the first or 25 lb. pressure line divided by 996 is .003313; this, multiplied by 726, the volume of 35 lbs. pressure, gives 2.4, the length of the 35 lb. pressure line; and so on for all the rest.
The application of either of the above curves will show that some diagrams are much more accurate than others, even though taken from engines of the same design and quality of workmanship.
As a general rule, those from large engines will be more correct than from small ones, and those from high more correct than from low speeds, and in either case efficiently covering the steam pipes and jacketing the cylinder, to prevent condensation, will improve the diagram.
The character of the imperfection in the expansion curve, shown by the application of a test curve, is generally too high a terminal pressure for the point of cut off, the first part of the curve being generally the most correct, and nearly all the inaccuracy appearing in the last half.
The usual explanation of this is, that the steam admitted during the live steam period condenses because of having to heat the cylinder, and that this water of condensation re-evaporates during the latter part of the stroke when this water of condensation is at a higher temperature than the expanded steam, and thus increases the pressure.
A leaky admission valve may generally, however, be looked for (or else wet steam), if the expansion curve rises much during its lower half.
TO CALCULATE THE HORSE POWER FROM A DIAGRAM.
In calculating the horse power of an engine, the only assistance given by the indicator is, that it provides a means of obtaining the average pressure of the steam throughout the piston stroke.
There are two methods of doing this, one by means of a planimeter or averaging instrument, and the other by means of lines called ordinates.
The ordinates or lines are drawn at a right angle to the atmospheric line, as shown in Fig. 3372, and each line is taken to represent the average height or length of one-half of the space between itself and the next lines.
Suppose, for example, that we require to get the area of that part of the diagram that lies between the dotted lines in the figure, and it is clear that the average height of this part of the diagram is represented by the height of the full line between them.
Any number of ordinates may be used, and the greater their number the greater the accuracy obtained. It is, however, usual to draw 10.
The end ordinates a and d, in the figure, should be only half the distance from the ends of the diagram that they are from the next ordinate, as will be seen when it is considered that the ordinate is in the middle of the space it represents.
The ordinates being drawn their lengths, are added together, and the sum so obtained is divided by the number of ordinates, which gives the average height of the ordinates.
Suppose, then, that the average height of the ordinate is two inches, and that the scale of the spring of the indicator that took the diagram was 30 lbs., then the average pressure, shown by the diagram, will be 60 lbs. per square inch. Or in other words, each inch in the height of the ordinate represents 30 lbs. pressure per square inch.
The mean effective pressure having been found, the indicated horse power (or I. H. P. as it is given in brief) is found by multiplying together the area of the piston (minus half the area of the piston rod when great accuracy is required) and the travel of the piston in feet per minute, and dividing the product by 33,000, an example having been already explained.
It is to be observed, however, that when great accuracy is required a diagram should be taken from each end of the cylinder, as the mean effective pressure at one end of the cylinder may vary considerably from that at the other.
This will be the case when a single valve is used with equal lap, because, in this case, the point of cut off will vary on one stroke as compared with the other, which occurs by reason of the angularity of the connecting rod.
When cut off valves or two admission valves are used, it may occur from improper adjustment of the valves. It occurs in all engines, because on one side of the piston the piston rod excludes the steam from the piston face, unless, indeed, the piston rod passes through both covers, in which case the rod area must be subtracted from the piston area.
If the expansion curve in a diagram from a non-condensing engine should pass below the atmospheric line, then the mean effective pressure of that part of the card that is below the atmospheric line must be subtracted from the mean effective pressure of that part that is above the atmospheric line, because the part below represents back pressure or pressure resisting the piston motion.
The planimeter affords a much quicker and more accurate method of obtaining the average steam pressure from a diagram.
Coffin’s averaging instrument or planimeter is shown in Fig. 3373. The diagram is traced by the point o, and the register wheel gives the area of the diagram.
A quick method of approximating the mean effective pressure (or M. E. P. as it is called) of a diagram is to draw a line a b, in Fig. 3374, touching the expansion curve at a, and so inclined that the space e is, as near as the eye can judge, equal to the space d. Then the line f drawn in the middle of the diagram, and measured on the scale of the spring that was used to take the diagram, represents the mean effective pressure, or M. E. P. of the diagram.
CALCULATING THE AMOUNT OF STEAM OR WATER USED.
The amount of water evaporated in the boiler is not accounted for by an indicator diagram or card, and the full reasons for this are not known.
It is obvious, however, that the loss, from the steam being unduly wet or containing water held in suspension, is not shown by the diagram, and this amount of loss will vary with the conditions.
Thus the loss from this cause will be less in proportion as the point of cut off occurs earlier in the stroke, because, as the water is at the same temperature as the steam, it will, as the temperature of the steam reduces from the expansion, evaporate more during the expansion period, doing so to a greater extent in proportion as the cut off is early, on account of there being a wider variation between the temperature of the steam at the point of cut off and at the end of the stroke. On the other hand, however, in proportion as the cut off is earlier, the proportionate loss from condensation during the live steam period is greater, because a greater length of the cylinder bore is cooled during the expansion period, and it has more time to cool in.
Whatever steam is saved by the compression, from the exhaust, must be credited to the engine in calculating the water consumption from the indicator card or diagram, since it fills, or partly fills, the clearance space.
In engines which vary the point of cut off, by varying the travel of the induction or admission valve, the amount of compression is variable with the point of cut off, and increases in proportion as the live steam period diminishes; hence to find the actual water or steam consumption per horse power per hour, diagrams would require to be taken continuously from both ends of the cylinder during the hour; assuming, however, that the point of cut off remains the same, that the amount of compression is constant, that the steam is saturated, and neither wet nor superheated, steam and the water consumption may be computed from the diagram as follows:
Water Consumption Calculations.—An engine driven by water instead of steam, at a pressure of 1 lb. per square inch, would require 859.375 lbs. per horse power per hour; the water being of such temperature and density that 1 cubic foot would weigh 621⁄2 lbs. If the mean pressure were more than 1 lb., the consumption would be proportionately less; and, if steam were used, the consumption would be as much less as the volume of steam used was greater than an equal weight of water. Hence, if we divide the number 859.375 by the mean effective pressure and by the volume of the terminal pressure, the result will be the theoretical rate of water consumption in pounds per I. H. P. per hour.
For the terminal pressure we may take the pressure at any convenient point in the expansion curve near the terminal, as at a, Fig. 3375, in which case the result found must be diminished in the proportion that the portion of stroke remaining to be made, a a, bears to the whole length of the stroke a b; and it may also be diminished by the proportion of stroke remaining to be made after the pressure at a has been reached in the compression curve at b. In other words, a b is the portion of the stroke a b, during which steam at the pressure at a is being consumed. Hence the result obtained by the above rule is multiplied by a b, and the product divided by a b.
To illustrate, suppose the mean effective pressure of the diagram to be 37.6 lbs., and the pressure at a, 25 lbs., of which the volume is 996.
Then 859.375/(37.6 × 996) = 22.94 pounds water per I. H. P. per hour, the rate that would be due to using an entire cylinder full of steam at 25 pounds pressure every stroke. But as the period of consumption is represented by b a (b a being the stroke), the following correction is required:
(22.94 × 3.03)/3.45′′ = 20.15; 3.03 inches being the portion b a, and 3.45 inches being the whole length b, a. This correction allows for the effects of clearance as well as compression, since, if more clearance had existed, the pressure at a would not have been reached till later in the stroke, and the consumption line b a would have been longer.
But such a rate can never be realized in practice. Under the best attainable conditions, such as about the load indicated on the diagram, or more on a large engine with steam tight valves and piston, and well protected cylinder and pipes, the unindicated loss will seldom be less than 10 per cent., and it will be increased by departure from any of the above conditions to almost any extent. It will increase at an accelerating ratio as the load is diminished, so that such calculations applied to light load diagrams would be deceptive and misleading; in fact, they have but little practical value, except when made for comparison with tests of actual consumption for the purpose of determining the amount of loss under certain given conditions.
DEFECTIVE DIAGRAMS.
In seeking the causes that may produce a defective diagram, the following points should be remembered:
The indicator must be kept in perfect order, thoroughly clean and well lubricated, so that its parts will move freely. It should always be cleaned throughout after using.
The motion of the indicator drum should be an exact copy, on a reduced scale, of that of the piston at every point in the stroke.
The steam pipes from the cylinder to the indicator, if any are used, must be large enough to give a free and full pressure of steam, and care must be taken that the water of condensation does not obstruct them or enter the indicator.
The cord should be as strong as possible, or if long, fine wire should be substituted.
The pencil should be held to the card with just sufficient force to make a fine line with a sharp pencil.
The diagram should be as long as the atmospheric line, any difference in this respect showing unequal tension of the cord, probably from unequal pressure of the pencil to the paper or card.
A fall in the steam line could arise from too small a steam pipe, and this could be tested by a diagram taken from the steam chest. It could also occur from too small a steam port or an obstructed steam passage as well as from a leaky piston.
An expansion curve that is higher than it should be may arise from a leaky valve, letting in steam after the cut off had occurred, or if at the later point of expansion curve, it may be caused by the steam being wet or containing water, which evaporates as the temperature falls from the expansion.
An expansion curve that is lower than it should be may be caused by a leaky piston, by a valve that leaks on the exhaust side but not on the steam side, or if the exhaust valve is separate from the steam valve, it may leak while the steam valve is tight.
It may also be caused by the cylinder being unduly cooled, as from water accumulating in a steam jacket.
There are many defects in the adjustment of the valve gear, or of improper proportion in the parts, that may be clearly shown by a diagram, while there are defects which might exist and that would not be shown on the diagram.
It is possible, for example, that a steam valve and the engine piston may both leak to the same amount, and as a result the expansion curve may appear correct and not show the leak.
Insufficient valve lead would be shown by the piston moving a certain portion of its stroke before the steam line attained its greatest height in Fig. 3376, in which from a upwards, the admission line, instead of rising vertically, is at an angle to the right, showing that the piston had moved a certain portion of its stroke before full pressure of steam was admitted.
That too small a steam port or steam pipe did not cause this defect may be known from the following reasoning:
The port opened when the pencil was at a, which shows that the valve had lead. At this time the piston was near the dead centre and moving slower than it was when the pressure reached its highest point on the diagram, and since the steam line is fairly parallel with the atmospheric line, it shows that the port was large enough to maintain the pressure when the piston was travelling fast, and therefore ample when the piston was moving slow.
The remedy in this case is to set the eccentric back.
With less compression the point a would be lower.
Excessive lead is shown in Fig. 3377 by the loop at a, where the compression curve extends up to the steam line, and the lead carries the admission line above it, because of the piston moving against the incoming steam.
To mark in the theoretical compression curve, the vacuum line and the clearance line must be drawn in as in the figure, and ordinates must be drawn.
According to the diagram, in Fig. 3377, the compression is clearly defined to have begun at c, and at that time the space filled by steam is represented by the distance from c to the clearance line. The pressure above vacuum (or total pressure) of the steam in the cylinder when the compression began is represented by the length or height of the dotted line 1.
Now suppose the piston to have moved from the point c, where compression began, to line 2 (which is midway between line 1 and the clearance line), and as the compressed steam occupies one-half the space it did when the piston was at c, therefore the steam pressure will be doubled, and line 2 may be drawn making it twice as high as line 1.
Line 2 is now the starting point for getting the next ordinate, and line 3, must be marked midway between line 2 and the clearance line, and twice as high as line 2, because at line 3 the steam will occupy half the space it did at line 2. Line 4 is obviously midway between line 3 and the clearance line.
Through the tops of these lines we may draw the theoretical compression curve, which is shown dotted in.
To find the amount of steam actually saved by the compression, we have to consider the compression curve only, beginning at the point of the diagram where it is considered that the compression actually began, and ending where the compression line joins the admission line, and the horizontal distance between these two points represents the length of the cylinder bore filled by the compression.
To find the average amount to which the steam is compressed, we must draw within this length of the diagram, and within the boundaries of the compression curve, and the line of no pressure ordinates corresponds to those given for finding the average shown pressure of a diagram, as explained with reference to that subject, taking care to have the end ordinates spaced half as wide as the intermediate ones, as explained with reference to Fig. 3372.
Chapter XLI.—AUTOMATIC CUT OFF ENGINES.
An automatic cut off engine is one in which the valve gear is so acted upon by the governor as to keep the speed of the engine uniform under variations of the load the engine drives, and notwithstanding variations in the boiler pressure. This it accomplishes by varying the point in the piston stroke at which the live steam is cut off. This is economical because it enables the engine to use the steam more expansively than is possible with engines having throttling governors, which govern the engine speed by wire drawing the steam.
There are two principal forms of automatic cut off engines, first, those in which the steam valve spindle or rod is released from the parts that move it to open for admission, while dash pots, weights, or springs close the valve to effect the cut off; and second, those in which the travel of the valve is varied so as to alter the point of cut off.
The first usually employ fly ball governors which actuate cams or stops to trip the valves for the steam cylinders. The second usually employ wheel governors or speed regulators, as they are sometimes termed.
The distinctive features in the action of the first, of which the Corliss engine is the most important, is that as two admission and two exhaust valves are used, therefore the amount of the valve lead, the point of exhaust and amount of the compression remain the same at whatever point in the piston stroke the cut off may occur; whereas in the second, the lead increases, the cut off occurs earlier, and the compression increases in proportion as the cut off occurs earlier in the piston stroke. In this class of engine the steam valve travels as quickly when opening the steam port for a short and early period of cut off as it does for a late one, hence the amount of steam port opening is as full, with reference to the piston motion, for an early as it is for a late point of cut off. In other words, there is the same amount of steam port opening for the first, second, third, and fourth inch of piston motion, let the point of cut off occur at whatever point in the piston motion it may. In engines which vary the point of cut off by reducing the travel of the slide valve, this is accomplished by using double ported valves or griddle valves.
Fig. 3378 represents the arrangement of the valves in a Corliss engine, v and v1 being the steam valves and v2 and v3 the exhaust valves. These valves, it will be seen, extend crossways of the cylinder and are circular. In the figure the valves are shown in the position they would occupy when the piston was at the crank end of the cylinder, as in the figure.
The principles of a Corliss valve gear will be understood from the following, which is derived from a book by the author of this work, and entitled Modern Steam Engines.
In 3379 and 3380 the valve gear (which is the distinctive feature of the engine) is represented with the parts in the position they occupy when the cut off occurs at half stroke, the piston having moved from the head end of the cylinder. In Figs. 3381 and 3382 the parts are shown in position with the crank on the dead centre and the piston at the crank end of the cylinder, valve v having opened its port to the amount of the lead.
| VOL. II. | THE CORLISS VALVE GEAR. | PLATE XXXIV. |
|
Large image (76 kB). |
||
| Fig. 3381. | ||
Referring to Fig. 3379, motion from the eccentric is imparted by the rod m to the wrist plate y, to which are connected the rods c, c′, for operating the admission valves, whose shafts are seen at s, s′, and the rods f, f′, for operating the exhaust valves, whose shafts are seen at t, t′.
The mechanism for the steam or admission valves may be divided into three elements: first, that for operating the valve to open the port for admission; second, that for closing the valve to effect the cut off; and third, that which determines the point in the stroke at which the cut off shall occur.
The first consists of the rod m, wrist plate y, and the rods c and c′, which operate the bell cranks r r, r′ r′ which are fast on the valve shafts s, s′. Upon the ends of bell cranks r r, r′ r′, are pivoted latch links u, u′, which have in them a recess for the latch blocks, of which one is seen at e (the rod r′ and its connection with the valve stem being shown broken away to expose e to view). During the admission the latch block abuts against the end y of the recess w and is tripped therefrom by the cam n′. The ends of arms g of the latch links abut against the hub of the arms d, d′ upon which are cams n, n′, and at a, a′ are springs for keeping the ends g of latch links u, u′ against the hubs and cams of d, d′.
Referring now to the valve mechanism at the head end only, suppose the piston to be at the head end of the cylinder, and latch block e will be seated in the recess provided in a to receive it, and as the bell crank moves, the latch block will be raised by the latch link, which is carried by a crank arm corresponding to that seen at x at the crank end of the cylinder, and as this crank arm is fast upon the valve spindle, the lifting of e will open the valve for admission. As soon, however, as the end g of the latch link meets the cam n′, the latch link will be moved so that the end y of its recess will leave contact with the latch block e and the dash pot will cause rod r′ to descend instantaneously and close the valve, thus effecting the cut off.
The period of admission, therefore, is determined by the amount of motion the latch link u′ is permitted to have before its end g meets the cam n′, which trips the latch link, and therefore frees e from the latch link recess.
The point at which the cut off will occur, therefore, is determined by the position of the cam n′, because if n′ is out of the way, the end g of the latch link will not meet it, the latch link will not disengage from the latch block e, and the cut off would be effected by the lap of the valve, and independently of the dash pot. As in Fig. 3379 the parts are shown in the positions they occupy at the instant the cut off is to occur, therefore the cam n′ has just tripped the latch link, and the end of e has just left contact with the end y of the recess w in the latch link u′.
The point in the stroke at which the tripping of u′ from e will occur and effect the cut off is determined by the governor, because d′ is connected to the governor through the rod g′. In proportion as the governor balls rise, d′ is moved from left to right, and the end of cam n′ meets g earlier, or, vice versa, in proportion as the governor balls fall, the arm d′ is moved to the left, g will meet the end of cam n′ later, and the point of cut off will be prolonged.
We now come to the means employed to close the valve quickly and without shock when the latch block is released from the latch link. Referring then to the crank end of the cylinder, the latch block for that valve is carried upon arm x, to which is attached the rod r from the dash pot piston (the arm corresponding to x, but at the head end being shown removed to expose the latch block to view). We may now turn again to the head end of the cylinder, rod r′ corresponding to rod r at the other end, and it is seen that r′ connects to a dash pot piston p′ having a stepped diameter, the lower half fitting into bore h′, and the upper half fitting into a bore h. The piston p′ fits the bore h′ and fills it when the rod r′ is at the bottom of the stroke, hence as p′ is raised there is a vacuum in h that acts to cause p′, and therefore r′ and x, to fall quickly and close the valve the instant the latch block is released from the latch link. To prevent the descent of rod r′ and piston p′ from ending in a blow, a cushion of air is given in h by the following construction:
At s and s′ are valves, threaded to screw and unscrew, the ends forming a valve for a seat entering h.
As the rod r′ and its piston p′ descend, the air in h finds exit through a hole at h until that hole is closed by the piston p′ covering it, after which the remaining air in h can only find exit through the opening left by the end of the valve s′, and this amount of opening is so regulated by the adjustment of s′ that a certain amount of air cushion is given, which prevents p′ from coming to rest with a blow. The head of valve s′ is milled or knurled, and a spring t′ fits, at its end, into the milled indentation, thus holding it in its adjusted position. The under surface of the upper part of p′ is covered by a leather disc, while the part that fits in h′ is kept air-tight by a leather-cupped packing.
The connection of the cam arms d and d′ with the governor is shown in Figs. 3381 and 3382, in which the parts are shown in the position they would occupy when the crank is on the dead centre and the piston at the crank end of the cylinder. The rod g′ connects the cam arm d′ with the upper end of lever a, which is connected to the governor and vibrates on its centre as the governor acts upon it.
Now suppose the speed to begin to diminish, and the governor balls to fall, and the direction in which a will move will be for its lower end to move to the right, thus moving d to the right and carrying its cam away from the end of the latch link, which will therefore continue to open the port for a longer period of admission. Or, referring to Fig. 3381, it is plain that, if the governor balls were to lower from a reduced governor speed, g′ would move to the left and cam n′ would be moved away from contact with the end g of the catch link, which, not being tripped, the admission would continue. On the other hand, suppose the governor balls to rise from an increase of governor speed, and d′ (Fig. 3379) would be moved to the right, and the cam n′ meeting g earlier, correspondingly hastening the cut off.
The governor is driven by a belt from a pulley on the crank shaft to the pulley w, Fig. 3381, whose shaft conveys motion to the governor spindle through the medium of a pair of bevel pinions in which v represents (referring again to Fig. 3378) the steam or admission valve for the crank end port, and v1 that for the head end port, while v2 is the exhaust valve for the crank end, and v2 that for the head end of the cylinder. All four valves are shown in the positions they would occupy when the crank was on the dead centre and the piston at the crank end of the cylinder, hence the valve positions shown correspond to the positions the parts of the valve motion occupy in Fig. 3381.
The faces of the valves are obviously arcs of circles of which the axes of the shafts s, s′ are the respective centres. Valve v has opened its port to the amount of the lead, which in this class of engine varies usually from 1⁄32 to about 1⁄16 inch. As separate exhaust valves are employed, the point of release, and (as the same valve edge that effects the release also effects the compression) therefore that of the compression, may be regulated at will by adjusting the lengths of the rods f, f′, Fig. 3379, which have at one end a right and at the other a left hand screw, so that by turning back the check nuts and then revolving the rods their lengths will be altered.
Similarly the amount of admission lead may be adjusted by an adjustment of the lengths of rods c, c′, which also have right and left hand screws. Referring now to the admission valve v, it is seen that its operating rod c is at a right angle to bell crank r, r, hence the amount of valve motion will not be diminished to any appreciable extent by reason of the wrist plate end of rod c moving in an arc of a circle, and the point of attachment of rod c to the wrist plate is such that, during the admission, the valve practically gives as quick an opening as though rod c continued at a right angle to r. But, if we turn to valve v′, which has closed its port and covers it to the amount of the lap, we find that bell crank r′ and its operating rod c′ are in such positions with relation to the wrist plate, that the motion of the latter will have but little effect in moving the bell crank r′. This is an especial feature of the Corliss valve motion and is of importance for the following reasons:
The lap of the valve (which corresponds to the lap of a plain D slide valve) is usually, in this class of engine, such as to cut off the steam at about 7⁄8 stroke, but the adjustment of the cam position is usually so made that, from the action of the governor, the latest point of cut off will occur when the piston has made 5⁄8 of its stroke, the range of cut off being from this to an admission equal to the amount of the lead.
As the eccentric is fixed upon the shaft, the speed at which the valve opens the port for the admission is the same for all corresponding piston positions. Thus suppose the piston has moved an inch from the end of the stroke, and the valve speed will be the same, whether the cut off in that stroke is to occur at quarter stroke or half stroke, and as the valve continues to open the port until it is tripped, therefore, at the moment it is tripped, the direction of valve motion must be suddenly reversed.
As the duty of its reversal falls upon the dash pot, it is desirable to make this duty as light as possible, which is accomplished by the wrist motion, which acts to reduce the valve motion after the port is opened a certain amount for the admission.
We have, therefore, that during the earlier part of the admission, the port opening is quick because of the eccentric throw being a maximum, while during the later part of the port opening, this rapid motion is offset or modified by the wrist motion, thus lessening the duty of the dash pot and enabling it to promptly close the valve.
The range of governor action, so far as the governor itself is concerned, is obviously a constant amount, because a certain amount of rise and fall of the governor balls will move the cams a given amount. But the range of cut off may be varied as follows: At z, z′, are adjustment nuts, by means of which the lengths of rods g, g′ may be varied.
Lengthening rod g obviously moves arm d and its cam n further from the end of the latch link u, and therefore prolongs the admission period.
Shortening the rod g′ causes cam n′ to move around and away from the leg g of the latch link, and prolongs the admission.
The adjustment of the lengths of g and g′ may therefore be employed for two purposes; first, to prolong the point of cut off, and maintain the speed when the engine is overloaded, or to hasten the point of cut off for a given engine speed, and thus adjust the engine for a lighter load.
HIGH SPEED AUTOMATIC CUT OFF ENGINES.
What are termed high speed engines are those whose pistons run at a velocity of more than about 600 feet per minute, some making as high as 800 or 900 feet in regular work. High speed engines are usually provided with an automatic cut off, and a majority of them vary the point of cut off, by means of shifting the eccentric across the shaft, so as to reduce the eccentric throw, and therefore the valve travel. This causes the valve to cut off the steam earlier.
The eccentric, instead of being fixed upon the crank shaft, has an elongated bore, and is hung on an arm that is pivoted at its other end after the manner of a pendulum. This arm is called the eccentric hanger.
A wheel governor is usually employed to shift the eccentric across the shaft. In some cases, however, two valves are employed, one effecting the admission, the release, and the compression, and the other the cut off.
When two valves are employed, the lead, the point of cut off, the point of release, and the point of compression may be maintained equal for all points of cut off; whereas, when a single valve is employed, the lead, the point of release, and the compression will vary with the point of cut off, or, in other words, will be different for every different point of cut off.
The general principles upon which a wheel governor is constructed is, that two weights or weighted levers in moving outwards from the engine shaft, from the action of centrifugal force, move or rather shift the eccentric across the shaft, reducing its throw, and therefore by reducing the travel of the valve hasten the point of cut off and reduce the power of the engine.
In the governor of the Buckeye engine, the centrifugal force may be varied by increasing or diminishing the distance of the weights from the pivots of the arms on which they swing.
This is shown in Fig. 3382a, in which it is seen that the weights a are adjustable along the arms a, a. The points of attachment d, d of the springs to the weight arms are also adjustable.
When reversing is done, by shifting the eccentric across the shaft, the lead cannot be kept equal, but will, if the eccentric is swung from a pivot that is on the line of centres, when the crank is on a dead centre, be greater at the head end than at the crank end of the cylinder. The discrepancy may, however, be equalized by swinging the eccentric from a pivot that is not on the line of centres at a time the crank is on a dead centre.
But this equalization will only exist at some one point in the eccentric position, or in other words, if the eccentric is shifted across the crank shaft, simply to reverse the engine, and not to vary the point of cut-off, it will naturally be moved, in reversing the engine across the shaft, to a given and constant amount, and in this case, the pivot on which its hanger is hung may be so located with reference to the line of centres and the crank (the latter being on a dead centre when the point of suspension of the eccentric hanger is found) that the lead is equal for both the backward and forward gears.
But if the eccentric is shifted across the shaft to vary the point of cut off as well as to reverse the direction of engine revolution, the lead cannot be kept equal.
It is better, in this case, to so locate the point of eccentric hanger suspension as to let the lead be the most at the head end cylinder port, because the piston travels fastest at that end of the cylinder, and therefore requires more lead, in order to cushion the piston.