This chuck has been in use for many years, and has in consequence
been of late rather neglected. It is, nevertheless, the parent of
those more elaborate contrivances included under the general title of
Geometric Chucks, of which Ibbetson's stands first in order of date,
and possibly of merit, though this last qualification may admit of
question. The epicycloid, defined mathematically, is a curve described
by the revolution of a point in the circumference of a circle, when
the latter is made to roll upon the concave or convex side of another
circle. A pin in the rim of a wheel revolving round and in contact
with another wheel, therefore, describes this curve, which constitutes
two or more loops, as will be seen by the annexed illustrations. The
number of these loops is variable, and the chuck will produce almost
any number by changing the pinions, and thus altering the relative
velocities of the revolving parts. The following description will
make the action of this chuck clear, and enable any good mechanic to
construct one for himself. In the first place, the above remarks show
a necessity for a fixed wheel, round which another may revolve. Fig.
313 represents this attached to a plate of brass, which can be fixed
to the lathe head, the mandrel passing through its centre. This is
the original pattern of plate, and need not, of course, be adhered
to, as the form can be modified to suit the lathe to which it is to
be applied. It is merely necessary to affix such a wheel to the face
of the poppet, so as to be concentric with the mandrel [a plan done
away with, however, or rather reversed, in Plant's geometric chuck].
The epicycloidal chuck, which screws to the mandrel as usual, consists
of a foundation plate of brass, A, Fig. 314, behind which is mounted
the cogwheel, B. The axis of this wheel passes through the plate,
and is carried by another plate, _e_, which is curved and adjustable
upon the former. The axis of this wheel carries a small pinion, D, so
that the whole turn together. This pinion being one of a set of change
wheels, necessitates the possibility of adjusting the plate which
carries its axis, as all the several change wheels must gear with E,
which always retains its position on the chuck. The sliding plate in
question being put in place, is clamped by the screw F. A plate of
iron, G, of the form shown, and of sufficient thickness for the secure
attachment of the wheel and of the screw which carries the ordinary
chucks, is fitted to turn on the axis of the wheel E. Its larger end
traverses within the arc H, which is graduated. The arc is bevelled
underneath, serving to hold down securely to the foundation plate the
piece of iron which is chamfered to fit it. At the left side of the
back plate is seen a stop, L, which is placed in such a position, that
when the iron plate rests against it, the screw M is concentric with
the mandrel, and work may be turned as upon an ordinary chuck. To
throw the iron plate, and consequently the nose of the chuck, on one
side, or in other words to place the work eccentrically, the screws
which retain the arc are loosened and the adjustment made by hand. The
eccentricity is marked by an index on the iron plate, which points
to the graduations seen upon the face of the arc. The eccentricity
being determined, the arc is again screwed down to retain the movable
plate in its new position. [Although this is Bergeron's method, it
appears vastly inferior to the plan of racking the edge of the iron
plate, and moving it to any required degree of eccentricity by the
aid of a tangent screw.] When made as above, the chuck will produce
loops varying in number according to the relative dimensions of the
pinion (or change wheels gearing with E) and the central wheel, M.
If the latter has 120 teeth, and the change wheel 60, two loops will
result. If the pinion has ten teeth, the number of loops will be 12,
and so forth. The practical _limit_ to the number depends on the
possibility of diminishing the pinion in size and number of cogs,
and still keeping the latter of such size and pitch as to gear with
E. If, therefore, a larger number of loops is required than can be
obtained thus, it becomes necessary so to modify the form of chuck as
to permit of intermediate change wheels, and when the modification is
carried out, we have the geometric chuck, the most perfect, but the
most complicated and expensive of all. To understand the nature of the
work, the following is given in clear language by Bergeron, and will
sufficiently explain and simplify matters.
[Illustration: FIG. 313.]
[Illustration: FIG. 314.]
[Illustration: FIG. 315.]
If one considers the movement of the piece (of work) when the wheel
M is concentric with the mandrel, it will be perceived that although
it makes two revolutions upon its axis, yet inasmuch as it has no
eccentricity it will describe no particular curve or figure; but if an
eccentricity of three divisions is given to it, two buckles or loops
will result as in Fig. 315. Before cutting the material, however,
approach the tool as near the work as possible, and putting the lathe
in motion, observe whether the buckle[23] passes too near the centre
or too far from it, and also how near it goes to the circumference.
If another change wheel with forty teeth instead of sixty is
substituted, the slide C, being adjusted accordingly, three loops
will be described (forty being one-third of one hundred and twenty),
but it is always necessary before actually cutting the material, to
try whether the buckles will pass near to the centre without going
beyond it. The result of the latter movement will be shown presently,
as it entirely alters the appearance of the pattern. The divisions
commence on the arc at the left hand, the index resting at 0° when the
plate is against the stop L, and the screw of the chuck concentric
with the mandrel. The preceding figure of three loops will become
similar to Fig. 317, retaining the same wheels and some degree of
eccentricity, but by means of the slide rest moving the tool towards
the circumference, so that the buckles overlap the centre. The effect
thus produced is, that of a set of three curvilineal triangles of
which the apex of one falls upon the base of the next. The use of this
chuck is stated to require upon the part of the operator more care
than any other, as regards the derangement of work or tool in the
least during the operation, as, if either is once moved in the least
out of position, it will be found next to impossible to strike the
line again, owing to the peculiar nature of the curve, for although
the tool may be replaced upon any one part of the line already cut
with the intention of deepening--it is by no means certain that it
will trace the same curve again. This curve, says Bergeron does not
produce an agreeable effect on the cover of a box, unless it is
very finely cut, the tool, therefore, should be very sharp in the
angle and very keen. Bergeron specially mentions this in reference
to filling the cuts with thin strips of horn or shell, a method of
inlaid ornamentation not much known or admired in the present day, but
to which allusion may probably be made again in this series. To form
the second set of loops, which are parallel with the first in these
designs, it is only necessary to use the leading screw of the slide
rest, to move the tool nearer to or further from the centre, while the
eccentricity and the arrangement of change wheels remain as before. It
is scarcely necessary to detail the formation of the larger series of
four loops and upwards, as these are simply the result of different
sized change wheels: the following principles, however, by which
the buckling of the several loops is controlled or prevented, may
perhaps be serviceable. "When, for example an eccentricity of sixteen
divisions is used--if the tool is placed at a distance of two such
divisions from the centre of the piece, a line only will be produced
of as many _curves_ as the wheel or pinion D would produce _buckles_.
If the tool is moved further from the centre by a quarter division,
the angles (connecting the curves) will be more defined, but still
no buckles will be made. A little further movement of the tool will
produce very small buckles which will thus gradually increase as the
tool is set further and farther from the centre--until at last when
the curves pass beyond the centre, the result is arrived at already
shown in Fig. 317. Another form of this chuck is shown in 318, in
which, instead of the iron plate being pivotted for the purpose of
eccentricity upon the axis of the wheel E, a parallel slide motion is
given to the main wheel by guide bars, as in the eccentric and oval
chucks. This form is figured in "Lardner's Cabinet Cyclopædia." The
large front wheel carrying the screw for chucks is pivotted to the
slide C, and protected by a plate D which nearly covers it. The wheel
L, is arranged to follow this slide, so as to remain in gear with the
large wheel without leaving the fixed wheel or ring on the face of
the poppet. In both patterns of this chuck the front wheel is used
as a division plate, being moved in either direction as many cogs as
desired to produce interlacing of the looped designs. It is better,
however, to add a racked division plate and tangent screw, as in the
eccentric chuck to act as one piece with the chuck screw, and with
the latter turning on a conical pin in the centre of the large wheel
underneath. The above apparatus requires to be used with a slow motion
owing to the complication of parts, and the whole ought to be so well
constructed, that the various wheels revolve with perfect smoothness
and without shake or noise.
[23] The word _buckle_ is used to signify the small loops--not the
large curves.
[Illustration: FIG. 316.]
[Illustration: FIG. 317.]
[Illustration: FIG. 318.]