Page images


When these are given, the point of the stroke of the piston at which the steam is admitted to the cylinder, cut off, exhausted, and compressed or shut up, are all deducible by model, by diagram, or by calculation. This can be done, whether the valve derives its motion from a single eccentric, or from a link-motion, as the motion of the valve is virtually the same in both cases. The way in which the valve is caused to cut off or suppress the steam earlier by the link-motion, is by shortening the travel of the valve; this is accomplished by means of the reversing gear, in such a manner that whatever be the reduction of travel communicated to the valve, the lead is always at least the same as in full gear, and with the shifting-link is rather increased.

In working out the four changes in the distribution of the steam, already enumerated, which regulate the movements of the steam, the action of the link-motion is such that, 1st, the sooner the steam is cut off, the sooner it is exhausted, the sooner the port is closed for exhaustion, and the sooner the port is opened for the admission of steam.

2d, That though every change is made earlier-as measured in parts of the stroke-there is less difference in the position of the points of exhaust, compression, and admission, than in that of the cutting off. Consequently, the shorter the admission, the longer is the expansion, as the exhaust point does not recede so much as the point of cutting off.

3d, That by the shifting link-motion, the steam may be cut off at from 1th to 4th of the stroke.

4th, That though the exhaust takes place earlier for every increase of expansion, it does not in any case take place within the first half of the stroke. For mid-gear it occurs in fact at 54 per cent. of the stroke; and the steam is expanded into 3 times the length of stroke at which it is cut off.

5th, That the period of compression, increasing as the admission is reduced, amounts to about one-half stroke in mid-gear.

6th, That the pre-admission of the steam, not above 1 per cent. of the stroke in full gear, reaches about 10 per cent. in mid-gear.

These results prove that the link-motion is capable of cutting off steam as early in the course of the stroke as can ever be advisable in practice.

It has been seen that the earlier the steam is cut off, the earlier also it is exhausted; until in mid-gear it may be released at half stroke. This has been deemed a serious objection to the use of link-motions for high expansion, as it is supposed to lead to a serious loss of expansive action, by exhausting prematurely. This loss is, however, a mere trifle in practice. The escape of the steam is by no means instantaneous, as is easily proved by the diagrams in fig. 1 (Plate I.), taken by the writer from the Caledonia, passenger engine, by means of M'Naught's Indicator, at speeds of 1 and 2 miles per hour. The numbers in the diagrams indicate the number of the sector notches to which the reversing lever was placed, while the diagrams were described. Referring to No. 1, taken under full gear, the steam is shown to be admitted to the cylinder a little before the beginning of the stroke, at A. From B to C the steam is admitted, at C shut off, expanded to D, and thence exhausted to E, the end of the

stroke, whence it continues to be exhausted till the point F in the return stroke, where the exhaust port is closed. Now, the exhaust line, D E, shows that nearly all the period of exhaust for the steam stroke is employed for the complete evacuation of the steam. And if this be the case for speeds of 1 and 2 miles an hour, it is much more so for the regular working speeds of trains. To select from a very admirable series of Indicator Diagrams, with copies of which the writer has been favored by Mr. Daniel Gooch, by whom they were taken from the cylinder of the Great Britain, locomotive, on the Great Western Railway, the figs. 4 and 5 contain diagrams taken at 17 and 55 miles per hour respectively, under the 1st, 3d, and 5th notches of the sector. The following are the conditions of the valve motion of this engine, when the diagrams were taken:

State of the Valves of the "Great Britain" Locomotive, G. W. R. Cylinder, 18 x 24 inches; wheel, 8 feet; lap, 14 inch; constant lead, full gear, 4 inches; blast orifice, 5 inches diameter.

Position of Points of Distribution.

-inch; travel in

No. of Notch.

Period of exhaust during

[blocks in formation]

steam stroke.

[blocks in formation]

On the diagrams the points of cutting off and exhaust are marked, and the steam line falls only very gradually during the period of exhaust, and especially at the high speeds. The expansion curves are shown by dotted lines, A, B, C, figs. 4 and 5, continued to the end of the stroke. These are easily calculated in terms of the relative volumes of steam, from the pressures indicated at the points of exhaust, and are such as would have been described had the exhaust been delayed till the end of the stroke. The shaded areas, A, B, C, inclosed between these dotted curves and the curves actually described, express the power lost by exhausting the steam before the stroke is completed. Averaging them for the whole stroke, they are as follows:

[merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][ocr errors][merged small][merged small][merged small][merged small][merged small]

The losses at high speeds are very small, merely nominal; and curiously enough, the loss by the earlier exhaust of the 5th notch is actually less than that under the 1st notch. The losses are of course greater at the low speeds; but even then, in the 1st notch, which is the only notch employed at very low speeds, the loss does not amount to 1 lb. per inch. The 3d and 5th notches are employed only at speeds much above 17 miles per hour, and the loss by them is of no practical moment.

Upon the whole, it follows that the possible loss by the early exhaust yielded by the link-motion is of no importance. On the contrary, it can

be proved to be beneficial, as an early exhaust is at high speeds essential to a perfect exhaust during the return-stroke. It plainly appears, therefore, that with the existing arrangements of locomotives, any attempts to eke out the power of the steam-line, by prolonging the expansion materially beyond what is accomplished by an ordinary valve and link-motion, are not only useless, but highly prejudicial.

Another objection to the link-motion is, that the steam is injuriously wire-drawn by it when under great expansion. Hence the numerous attempts to supersede the link by the employment of a separate expansion valve. The diagrams, fig. 5, may be referred to as examples of wire-drawing by the link. They were taken nearly consecutively with one opening of the regulator; and it is clear that the steam attained fully as high a pressure in the cylinder under the 5th notch as under the 1st. The pressure falls considerably towards the point of cutting off, but from the form of the steam-line, it is plain that very little additional steam is admitted for an inch or two before the cutting-off actually takes place. The most of the steam is admitted at the higher pressure, and in fact a partial expansion of the steam already admitted takes place for some distance before the expansion nominally begins. Thus the wire-drawing is, to a great extent, equivalent to an earlier cutting-off, and a greater degree of expansion. The whole possible loss by wire-drawing is comprised within the dotted line D, added to the diagram, which is merely an extension of the expansion curve to meet the steam line, drawn horizontally to represent a free admission up to an imaginary point, D, of cutting off, 5 inches from the beginning of the stroke. This shaded area, D, amounts exactly to a mean loss upon the whole stroke of one pound per square inch, by wire-drawing, under high expansion. For the 1st and 3d notches, the amount of loss by wire-drawing must obviously be still less; and, in short, the objection of wire-drawing by the link-motion, when of liberal proportions, is of no practical weight.

Another objection to the link-motion, and apparently the most formidable one, is the large fraction of power neutralized by the compression of the exhaust steam, and which increases with the degree of expansion. Compression, however, involves no loss of efficiency; for as by compression a quantity of steam is incidentally reserved and raised to a higher pressure, it gives out the power so expended in compressing it, during the next steam stroke, just as a compressed spring would do in the recoil. But, apart from this general argument, the actual efficiency of the steam in the cylinder, with and without compression, may be exactly estimated. The most direct method of doing so is, to find the quantities of water consumed as steam for one stroke, under the two conditions, and to compare them with the relative effective mean pressures. It will suffice to analyze, as an example, the high speed diagram, fig. 5, under the 5th notch, No. 5. The volume of steam admitted is measured by the product of the area of the piston, (254.47 in.,) and the period of admission, plus the total clearance in the cylinder and steam passage; the clearance being measured for simplicity in inches of stroke, we have 7+18= 8.8 inches, for the total volume admitted. The pressure of the steam when cut off is 65 lbs., for which the relative volume of water is 359. There

fore, the volume of water as steam, or the water equivalent of the steam admitted, is

[blocks in formation]

From this is to be deducted the quantity of steam reserved by compression; the volume so reserved is measured by the period of compression, plus the clearance (7·5+1·89.3), and the pressure at the point of compression is 8 lbs., for which the relative volume is 1125. Then the water equivalent of the reserved steam is—

[blocks in formation]

subtracting, there remains 6.24-2.10


4.14 cubic inches of water as steam, actually expended for one stroke of the piston. Were there to be no reservation of exhaust steam by foreclosing the exhaust port, the whole area of resistance by compression would be removed, and there would be a reserve of steam of atmospheric pressure equal in volume to the clearance only. The relative volume of atmospheric steam is 1669, and the water equivalent of the reserve would be

[blocks in formation]

the expenditure per stroke would be 6.24 0.275.97 inches of water.

Per Inch.

Now the positive mean pressure during the steam stroke, as indicated, is 40.9 lbs. And the mean resistance by compression is

[blocks in formation]

66 11.5


This effective mean pressure of 29.4 lbs. is maintained by a consumption of 4.14 inches of water per stroke; and it has just been found that with the compression removed, the positive mean pressure of 40.9 lbs. per inch would be maintained by a consumption of 5.97 inches of water per stroke. cubic inch of water is thereThe effective pressure created per

[merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small]

These quantities are expressions of the relative efficiency of steam employed with and without compression; they are virtually identical, and show that the resistance by compression in the cylinder, due to the action of the link-motion, does not in the slightest degree impair the efficiency

of the steam.

The last objection to the use of the link, requiring notice, is that at high speeds considerable back exhaust pressure is created. The amount of this is very various, and it depends also on circumstances for which the Jink-motion is not responsible; such as a deficiency of inside lead (which is regulated by the lap), small ports, a small blast-orifice, and imperfect protection of the cylinder. It suffices on the present occasion to point to what can be done by superior arrangements, as exemplified in the diagram, fig. 5, taken from the Great Britain. The cylinders of this engine are in a manner suspended in the smoke box, and thoroughly protected; the steam ways are very large, 13 x 2 inches, being in area about th of the cylinder; the exhaust passage is very direct; and the blast-orifice is 5 inches diameter, or about 1th of the area of cylinder. As a whole, these proportions are superior to those of any other engines with which the writer is acquainted; and the diagrams prove that the per centages of back exhaust pressure, in terms of the positive mean pressure, at 55 miles per hour, are

[merged small][ocr errors][ocr errors][merged small][merged small]

Better results than these should not in practice be required, for when locomotives are adapted to their work, and running at high speeds, they ought not to require an admission of steam above half stroke. However, the area of blast-orifice rules the back exhaust pressure; and, when the cylinder is duly proportioned to the boiler, it is quite practicable, by a few modifications in detail, still further to increase the orifice, sufficiently to banish all traces of back pressure of exhaust at all practicable speeds.

II. Of the Rate of Efficiency of Steam worked Expansively in the Locomotive, by the Link-Motion.-To determine this ratio experimentally, under the actual circumstances of clearance, wire-drawing, and back pressure, the writer has analyzed twenty-six of the indicator diagrams from the Great Britain, already referred to, taken at speeds of 15 to 56 miles per hour, of which the figures are examples. The following table contains in the first nine columns an analysis of these diagrams; the effective horse powers, column 10, are estimated in terms of the diameter and stroke of cylinder, the diameter of the wheel, and the effective mean pressures in the 9th column. The water equivalents, columns 11, 12, and 13, are estimated from the indicated pressures and the period of the distribution for each notch, in the way already exemplified. The expenditure of steam per hour, column 14, is deduced from column 13, in terms of the speed, the cylinder, and the wheel; and, dividing that by the effective horse power, we have the contents of column 15 in inches, and of column 16 in pounds. Column 17 contains the coke consumed per horse power per hour, deduced for the several diagrams from the consumption of water, column 16, allowing 1 lb. of coke to evaporate 8 lbs. of water. Referring to the contents of the last two columns of this table, it is obvious that the consumption of water as steam, or of coke, for a given amount of work done, becomes less the more expansively the steam is

« PreviousContinue »