From this figure it will be seen that the best point to cut off the steam with this small pressure is about half-stroke, because the steam consumed by cutting off at half-stroke is less in proportion to the power, as compared with cutting off at five-eighths; while, by cutting off at less than half-stroke, the steam line only decreases in proportion to the decrease of power. If the pressure were much greater, it might be advantageous to cut off

sooner.

Indicator diagram from the S.S. British Queen. Taken from the top of the cylinder. Vacuum 30 inches; steam in boilers 6 lbs.; expansion tried on §, 1, §, as represented by the diagram.

Fig. 101.

5:5

14

The vacuum was the same with or without expansion, as shown by the diagram; the steam ways must have been large. The condenser was on Mr. HALL's patent principle of condensing by cold surface, without the water coming in contact with the steam.

311. Difference of Effect between throttling the Steam and cutting it off by the Expansion Valve.-An experiment made in this manner to determine the difference of effect between expanding the steam in the cylinder by a proper system of expansion, and merely shutting it off by the throttle-valve when it is wished to work the engines to a low power, gave the following result:-An engine of the nominal power of 400 horses was selected, and was first worked on the lowest grade of expansion, the steam in the boiler having its full pressure of 8 lbs. Under these circumstances, it was found that 10 87 cubic feet of steam, at atmospheric pressure, were consumed per horse-power per minute.

The expansion-valves were then thrown out of gear, and the engines reduced by the throttle-valve to exactly the same number of revolutions. It was now found that although the gross horse-power, as shown by the indicator, was almost identical in both cases, in the latter case the power exerted was obtained at an expenditure of 1472 cubic feet of atmospheric steam per minute, in place of 10'87 cubic feet, as formerly; 3.85 cubic feet of steam per indicated horse-power per minute being thus saved by the principle of expansion.

It is apparent that this calculation will not give us the cause of any loss of effect by the steam escaping, or being condensed in the pipes, etc., which

must be sought for when suspected, either in the form of the indicator diagram, or by testing the engine for leakage of steam or air.

312. On the form of the Diagram when the Steam is throttled by the Link.-In Fig. 102, we have diagrams taken from top and bottom of an engine, when slowed to 4 strokes per minute, which is effected by partly closing the throttle-valve, and partly by shifting the link towards its midposition. It will be observed that in these diagrams nearly the whole areas are below the atmospheric line. Note also that on the left-hand corner of one of the figures a loop is formed, which not unfrequently appears in engines employing the link, and the meaning of which we shall now proceed to explain. The extreme point of the diagram, in every instance, answers

Fig. 102.

to the length of the stroke; and if the steam is pent up in the cylinder by the eduction passage being shut before the end of the stroke, or if it be permitted to enter before the end of the stroke, the pencil will be pushed up to its highest point before the stroke terminates, and since the paper still continues to move onward the upper part of the loop is formed. If the pressure within the cylinder, when the piston returns, were to be precisely the same as when the piston advances during this part of its course, the loop would be narrowed to a line. But as the advance of the piston, when the valve is very little opened, compresses somewhat the steam, and as its recession, when the valve is very little opened, causes it to be wire-drawn, the pressures, while the piston advances and retires through this small distance, although the cylinder is open to the boiler by means of a small orifice, will not be precisely the same; and the higher pressure will form the upper part of the loop while the lower part will be formed by the lower pressure. In following the outline of the left-hand diagram of the figure, it will be seen that the steam begins to be compressed within the cylinder when about three-fourths of the stroke has been completed, and the pencil begins to rise somewhat above its lowest point. But as the vapour within the cylinder is very rare, the rise is very little, until when the piston is about one-eighth part of its stroke, or about 8 inches from the end of the stroke, the steam valve is slightly opened, when the piston of the indicator ascends to the point answering to the pressure within the cylinder thus produced. As the opening from the boiler continues, and the piston, by advancing

against the steam, instead of receding from it, compresses rather than expands the steam admitted into the cylinder, the pressure continues to rise somewhat to the end of the stroke, when the piston of the engine having to move in the opposite direction, the steam within the cylinder will be expanded, and any that may still be entering will be wire-drawn in the contracted passage, and the pressure will fall. Under such circumstances a loop will necessarily be formed at the corner of the diagram, such as is shown to exist at the left-hand corner (see Fig. 102). The reason of there being no corresponding loop at the right-hand corner, arises simply from the valve being differently set at one end of the engine from what it is at the other; and the angles of the eccentric-rods will, in general, cause some small difference in the action of the valve at the different ends of the engine.

313. We have seen that an indicator card is run out, that is to say, the mean pressure of the card is usually ascertained by reading off with the scale the different mean pressures in each of the ten spaces, and then adding them together and dividing them by ten, and this is correct, provided each reading is a correct one; but the following is a far better and easier method. A long strip of paper, inch wide (the length varies from 6 to 18 inches, according to the nature of the card), has a starting point marked on the edge near to one end. The strip of paper is then laid along the first tenth of space, and the length of the mean pressure marked off, not read off, starting from the starting point. It is then laid on the second space, and again the mean pressure is marked off, starting from the point where the first mean pressure ended, and so on to the last of the ten mean pressures. By this means the mean pressure in each of the ten spaces are laid end to end. If we now take a rule and read off how many inches there are in the whole length, and divide them by ten, we get the number of inches in the mean pressure of the whole card.

10

Suppose when all the mean pressures for each of the ten spaces, when added together, measured 6.95 inches, then 5 is the measure in inches of the true mean pressure, and if the scale was inch, that is, I inch stood for 32 lbs., then 695 inch stands for '695 × 32 = Generally expressed, we multiply the total strip by the scale, and divide by ten.

22°24.

number of inches read off the

314. Remarks on the Quality of the Vacuum.-In designating the quality of the vacuum formed under the piston, there are two elements affecting it which ought to be noticed, viz., the weight of the atmosphere, and the temperature of the water in the condenser at the time the diagram was made. For, if the barometer stands at only 28 inches, 30 inches of mercury being equivalent to 147 lbs., 137 lbs. would be a perfect vacuum; and if the water in the condenser be at a temperature of 130°, its vapour will form a resistance of 2.17 lbs. ; therefore, the lowest attainable vacuum