The valve is then shut, thus preventing more steam from entering the cylinder a through the aperture f; at the same time the steam in the condenser h is condensed by the cold water which surrounds it, and supplied to the cistern mm, in which it stands, by the spout n, supplied by a common force-pump worked by the great beam. The vacuum being thus formed beneath the piston b, the pressure of the steam, which has free access to the upper side of the piston, forces it downwards to the bottom of the cylinder. By the action of the beam, the air-pump piston is pulled upwards, and the water withdrawn from the condenser h. By the valve mechanism, the valve g is shut, and the valve opened; steam is thus introduced beneath the piston, and an equilibrium being established between both sides, the counterpoise at the end of the great beam draws up the piston to the top of the cylinder; the air-pump is thus depressed, and the portion of condensed water lying in the lower part of the barrel passes through the clack or valve in the piston; the mechanism of the valve then shuts the valve ☀, and opens g; this taking place on the piston a reaching the top of the cylinder, the steam below the piston rushes through the valve g to the condenser, and a vacuum being formed as before, the piston is forced by the pressure of the steam on its upper side towards the bottom of the cylinder. The opening and shutting of the valves and g was effected by simple mechanism, as follows. W

Let a, fig. 21, be the spindle of the valve admitting steam to the cylinder, and b that of the valve admitting the steam to the condenser; these are connected by a joint to levers moving on the centres cc; hh is the

plug-frame, having studs or projecting pins on; these strike the handles or levers which are fixed on the rod ii, and to which the levers rp are fixed,

a

W

actuating the levers de, and lifting

or depressing the valves ab; s is the counterbalance weight which acts the tumbling bob in Beighton's valve gearing already noticed. The condenser, in its original form as introduced by Watt, consisted of a series of thin copper pipes communicating with each other, and placed in a cistern filled with cold water: in some instances flat copper pipes were used; the object, in both cases, being to present as great a surface as possible to the action of the cold water, and to effect a rapid condensation. Notwithstanding many drawbacks attendant upon this plan, it was considered an economical one, inasmuch as a comparatively small power was required to work the pump for withdrawing the air and condensed water. To receive a quick condensation by this method, it was indispensable to have a large amount of surface exposed to the cold water; this necessitated condensers of such size, that Watt was at last obliged to return to a plan which he had adopted while in Scotland during his trials under Roebuck, of condensing the steam by introducing a jet of cold water into the condenser. The clumsy outer casing was, after repeated trials, found to be possessed of inconveniences; Watt therefore discarded it, and adopted a plan of intercasing composed of thin sheet-iron, the space being only one inch and a half between it and the cylinder; this space was supplied with

fig. 21.

steam by a pipe leading from the main steampipe. This arrangement involved a radical change in the method of distributing the steam to the cylinder. The details of the new construction may be gathered from the diagram in fig 22.

Let aa be the cylinder, bb the outer casing, c the piston, d the piston-rod, e the steam-pipe leading from the boiler, f the "steam-valve," g the "equilibrium-valve," and h the "eduction-valve" in the pipe leading the steam to the equilibrium-valve. Supposing the piston at the top of the cylinder, the equilibrium-valve is closed, and the steamvalve ƒ and eduction-valve h opened; the steam from below the piston rushes through h to the condenser, and a vacuum is formed; the steam pressing on the upper side of the piston, it is forced hdownwards. On reaching the bottom, the steamvalve ƒ and eduction-valve h are closed, and the equilibrium-valve g opened; this allows the steam to gain access to the under side of the piston, as

well as to its upper side; an equilibrium of pressure is therefore formed, and the counterpoise pulls the piston to the top, the steam above it flowing through the equilibrium-valve. In this form of engine there is alternately steam and a vacuum on the under side of the piston, the steam being always above the piston.

About the year 1780, Watt introduced another modification, having for its object the attainment of a more perfect condensation: this he proposed to effect by having a perpetual vacuum below the piston, while there was alternate vacuum and steam-pressure above it; thus, on the piston having accomplished its loaded stroke, that is, from top to bottom, the vacuum being made, the whole of the time in which the piston was ascending might be occupied in freeing the cylinder from the air and steam. This idea was carried out by aid of the following arrangements.

In fig. 23, aa is the cylinder, b the piston, c the "steam-valve," d the steam-pipe, e the "eduction-valve," ff the

b

a

a

eduction pipe leading to the condenser, g the steam-port leading to the upper side of the piston. On the piston reaching the bottom of its stroke, the steam-valve c was shut, and the eduction-valve e was opened; the steam from the upper side of the piston rushed through g and e, and down the eduction-pipe f to the condenser; the vacuum was thus made on both sides of the piston, the counterpoise pulling the piston to the top; the eductionvalve remaining open during its ascent, a longer time was thus given to the formation of the vacuum above the piston. The advantages expected to flow from this ingenious arrangement did not, however, exist: it was found that in practice the condensation was so quickly performed, in fact almost instantaneously, that the longer time produced no better vacuum; and the cylinder approximating so much to the coolness of the condenser, a considerable quantity more of steam was required. Leakage to some extent also resulted from the arrangement. This arrangement of engine was therefore abandoned.

fig. 23.

We now come to notice an important improvement in the working of steam-engines, which the fertile genius of Watt added to the list of his brilliant inventions: this improvement was that of working the steam expansively. The patent for the expansive steam-engine was taken out in 1782; but the attention of Watt had been directed to the principle many years before; in 1769 he wrote to Dr. Small, as to a "method of still doubling the effect of steam, and that tolerably easy." Many matters, however, diverted his attention from this important point; and it was not until the above date that he took steps to introduce an engine in which the principle was carried out. The due understanding of its rationale is so important to the student of the steam-engine, that we propose entering into its consideration at some length.

Where steam is admitted to press on the top of a cylinder, during the whole of its descent the piston will move downwards with an accelerating

velocity, which, if not checked, will be of such amount as materially to
damage the mechanism. An able authority supposes that the value of the
expansive principle was made known through the result of some trials
which were instituted for the purpose of moderating the velocity of the
piston, and consequently the shock as the piston reached the bottom of the
cylinder. In Newcomen's engine he supposes this to have been effected
by shutting the injection-cock earlier; and in Watt's condensing-engine,
by shutting "the steam-valve at such a period of the stroke as would pre-
vent the catch-pins from striking." This shutting off the steam-communi-
cation from the boiler, at a certain part of the stroke of the piston, allow-
ing the steam to expand as the piston descends, constitutes the principle of
the method of working expansively. By the action of the well-known law
of pneumatics (see volume on Natural Philosophy in this series), the pres-
sure of the steam on the piston decreases as the space increases into which
the steam has liberty to expand itself; thus if the steam is cut off at one-
fourth of its stroke, the pressure will, at the end of the stroke, exert only a
force of one-fourth of its original pressure. By thus decreasing the power,
a simple method of equalising the tendency to an accelerated motion was
attainable. In addition, however, to this advantage, a still greater one re-
sulted from the adoption of the principle in the economisation of steam,
and the consequent saving of fuel. If steam of the temperature of 212°
"flows into a cylinder six feet long, until the piston has moved eighteen
inches downwards, when this quantity has expanded into double its former
volume, and in doing so has pressed the piston to the middle of the cylinder,
it will exert a pressure of not more than 7 pounds on each square-inch
area of the piston. When the piston has been depressed another eighteen
inches, the vapour will have expanded into three times its original bulk,
and will then urge the piston downwards with a force of not more than
?pounds on each square inch; and when it has reached the bottom
of the cylinder, and expanded into four times its original bulk, it will
not exert a greater energy than about
each
?pounds on
inch.
now we calculate the varying power of the steam from the commencement
to the termination of its stroke, beginning with a force of 14 pounds,
and ending with pounds, it will have exerted an average pressure of
nearly 8 pounds on each square inch of the piston. But if the vapour
had been permitted to flow freely into the cylinder as fast as the piston
descended, it would have pressed it with a force of 14 pounds during
the entire stroke of the piston. We thus see that one foot and a half of
steam, acting expansively, has pressed 8 pounds through six feet; while
six feet of steam, operating with its energy uniform and unimpaired, has
only carried 14 pounds through six feet; thus showing that more than one-
half of the whole steam has been saved by making it act expansively.

square

If

"Although the saving of steam is very considerable by making it work expansively, the power of the engine is reduced; thus, where the steam is cut off at one-fourth of the stroke, while the efficacy of the steam is increased four times,—that is, one-fourth the quantity of steam will complete the stroke, the power is diminished nearly one-half. In engines worked expansively, therefore, the size of cylinder must be increased in proportion to the extent to which the expansive principle is carried. But although the engine is made larger to do the same quantity of work, this work will

be done with a less consumption of fuel: this is obvious from the consideration, that at whatever point the steam is cut off, so much steam is saved; and that the steam, although it exerts a gradually decreasing force on the piston, still exerts a power of some extent, which power, whatever may be its amount, is gained without any expenditure of steam. To carry out the system of expansive working most conveniently, it is best to use steam of a pressure considerably higher than that of the atmosphere: unless this pressure is considerable, expansion cannot be carried out to any great extent with advantage; for if steam of a low pressure were used, the ultimate tension would be reduced to a point so nearly approaching that of the vapour in the condenser, that the difference would not suffice to overcome the friction of the piston, and a loss of power would be occasioned by carrying expansion to such an extent. It is clear that in the case of engines which carry expansion very far, a very perfect vacuum in the condenser is more important than it is in other cases. The advantage of applying steam expansively will be seen by an inspection of the following table: if the steam is cut off at one-half of its stroke, the performance of the engine will be multiplied 17 times; at

Watt effected the cutting off of the steam at any desired point by merely altering the position of the tappets or projecting pins in the plugframe, by which the valves were actuated upon at the proper time. As the motion of the piston was necessarily variable when the expansion principle was adopted, Watt contrived several ingenious mechanical combinations, by which the effect of the engine on the work it had to perform was uniform; he, however, did not apply these to any great extent, as he employed steam a little greater in pressure than that of the atmosphere, and cutting off only at one-third or one-fourth, according as circumstances dictated.

The reader desirous of becoming acquainted with these further evidences of Watt's inventive talent, will find several plans figured, by which this uniformity was obtained, in Stuart's Descriptive and Historical Anecdotes of Steam-Engines. We proceed to the consideration of more interesting and important matters in connection with the inventions of Watt.

Under the new arrangements it was a matter of importance to ascertain the state of the vacuum in the condenser and cylinder; for on the perfection of this obviously depended the efficiency of the engine. In order to ascertain this, Watt applied a mercurial barometer, having a connection with the inside of the pipe leading to the condenser; and another barometer was placed in connection with the boiler. The rise and fall of the mercury in the barometer attached to the condenser indicated the degree of exhaustion which had been made in it; and by the same operation in the barometer attached to the boiler, he had a measure of the pressure of steam acting in the piston: from the data thus obtained, he was able to calculate with considerable precision the amount of power given out by the engine. He afterwards, for this purpose, introduced a highly-ingenious