72

ENGINEER AND MACHINIST’S DRAWING-BOOK.

the point E, at which this arc intersects the line D E3, will
be the lower centre of the link, and the line D E will re-
present the link itself, or, in other words, the second side
of the parallelogram, which, being completed, H I will
similarly denote the centre line of the back links G, G, and
E I that of the parallel bars J, J.

The position of the fixed centre 0, of the radius bars
N, N, is all that now remains to be determined. In the
example now before us, as in most specimens of the ordi-
nary parallel motion, these rods are equal in length to
half the radius of the beam, or to D H; their fixed centre
is also in such cases, invariably in the vertical line E E2;
therefore, if from I, with the distance D H, an arc be
described cutting E E‘3, the point of intersection 0 will
be the point required. But, as the radius rods and
parallel rods should be all in the same horizontal plane
when the beam is level, it is perhaps preferable to resort
to the more direct method of fixing the position of the
point 0 by the intersection of an arc drawn from the
point D1, with a radius equal to D E, the length of the
main links. Both methods will, however, be found to
lead to the same result. Assuming another position of the
beam, we shall now endeavour to show that the point E
must still be found in the vertical line D E. Suppose,
for example, the beam to be in the horizontal position, it
will then be represented by the strong dot line C D1; the
point H, having traversed a circular arc about the centre C,
will now be in H1; and similarly the point I will have
moved along the arc I I1, and assumed a position upon
the horizontal 0 I1; and since the distance H I is invari-
able, the point I1 will be determined by the intersection
of an arc drawn with that radius from the centre H1,
within the arc I I1 I2. With regard to the position of
the point E, it is to be remarked that its distances from
the points D and I are obviously the same in all positions;
consequently, the circular arc drawn from the point I1,
with the radius I E, will meet that drawn from D1 with
the radius D E, in the point 0, which will be found to be
upon the vertical line D E2. By adopting the same con-
struction for the extreme positions of the beam (as shown
in the diagram), it will be observed that still the point E
does not deviate sensibly from the same vertical line;
and in every intermediate position the same will be
found to be the case. It is proper to add, however,
that this holds good only within certain limits, and that
when the amount of angular motion of the beam exceeds
40° the line traversed by the point E begins to deviate
considerably from a straight line, and this arrangement of
parallel motion will no longer produce the desired effect.

With respect to the vertical motion of the point of sus-
pension of the airpump-rod, it will be observed that the
arrangement of mechanism is identical with that repre-
sented in Fig. 4, Plate XXX.; the points C and 0 being
fixed centres of motion, the lines C H and 0 I being
equal, and their extremities, H and I traversing equal
and opposite arcs, it will be obvious, from what was de-
monstrated in reference to that figure, that the point K,
in the middle of the line joining H and I, will traverse a

vertical straight line bisecting the versed sine of half the
arc of vibration of either of the radii; the construction of
our present diagram will also prove this to be the case.
In fact, it is at this point K that the parallel motion is
primarily produced, the object of the parallel bars being
simply to transfer it to the piston-rod. It is worthy of
remark that, in every position of the beam, the straight
line joining the points C and K will, if produced, pass
through the point E.

Oblique Projection of an Oscillating Steam
Cylinder.—Plate XXXIII.

Although, in mechanical drawing, the planes of projec-
tion upon which objects are represented, are always taken
in a direction parallel to the faces of the objects, which
are thus shown in their actual forms and dimensions, it
frequently happens that the positions of certain parts of a
machine are such as to render an oblique or fore-shortened
view necessary. In Plate XXXIII. we add another ex-
ample, bearing reference to the subject of steam engines,
which at present occupies our attention.

The application of oscillating cylinders to steam engines
is of modern origin, and is rapidly coming into favour,
especially for marine purposes. It affords the means of
establishing a direct communication between the piston B
and the crank, the piston-rod A in this case acting also
as the connecting-rod. The cylinder C oscillates upon
trunnions E, E, working in bearings, and serving also as
the steam and exhaust pipes, and thus permitting the
piston-rod to accommodate itself to the varying angle of
the crank. The head D of the piston-rod is formed into
a socket which embraces the crank-pin, and is adjustable
by bolts. The bottom of the cylinder is cast in one piece
with the body, a central opening being left for the admis-
sion of the boring-bar, and afterwards hermetically closed
by a plate F,

Upon the exterior planed face of the cylinder is bolted
the valve-chest G, which receives the steam coming
directly from the boiler, through one of the trunnions
E, E, and which contains the short-slide steam-valve H;
this valve receives motion from an eccentric upon the
crank-shaft, at the same time that it partakes of the oscil-
lating motion of the cylinder. It is thus made to open a
communication alternately through the passages a and b,
between the steam in the valve-chest and the upper and
lower sin-faces of the piston. The valve is kept constantly
pressed against the face of the cylinder by the spring I.
The steam from the boiler is conveyed into the valve-chest
by the passage c, from one of the trunnions E, and, after
having performed its function in the cylinder, is either
conducted into the condenser, or blown off into the air, as
the case may be, by the passage d, communicating with
the other trunnion.

Figs. 3 and 7 represent two oblique views of an
oscillating cylinder, the first being an external elevation,
and the second a longitudinal section through the axis. To
obtain the first of these projections, it is necessary to lay
down the two views, Figs. 1 and 2, the former being an