The watchmakers' hand book

The bed P of the instrument carries a wheel R which engages with a pinion on the axis of the polishing lap m. The wheel R is mounted on a clock stud passing through a slot and fixed by a nut, so that the pitching of the two mobiles may be modified; motion is communicated to the lap by simply placing the finger against the teeth of R. The bed can move in a vertical plane, being pivoted on two screws, v, v, and the block that receives the points of these is riveted to the disc d d, which can be made to rotate, with friction, on the second disc n n. This latter is riveted to a plate e fixed at the end of a cylindrical rod F.

It will be evident that if the rod F is inserted in the T-rest support, the plate P extending towards the back of the lathe, this plate can be raised or lowered, and moved towards the right and left, so that the flat face of the lap can always be brought in contact with the pivot that is to be polished. This latter is caused to rotate by a foot-wheel while one hand holds the raised plate by the button a, and a finger of the other hand is applied to the teeth of R, causing the lap to rotate.

The upward motion of P may be limited by the edge of the top s of the button s k, which is tapped so as to rotate with stiff friction on the pillar H. The stop l is to prevent the polisher from traveling too far towards the left and thus removing too much from the shoulder that is to be polished. The screw x, giving motion to the slide y, is for securing parallelism between the pivot and the surface of the lap, according as the former is cylindrical or conical in shape.

For fine pivots it is advisable to introduce an additional wheel and pinion. The finger will then be better able to appreciate the degree of resistance opposed, and, owing to the increased velocity, it will be useless to use oilstone dust, but rouge can be applied directly after the turning. At e is a steady-pin for maintaining the position of the instrument.

BUSHING PIVOT HOLES, ETC.

560. Every watchmaker knows how to proceed in adjusting an ordinary perforated bushing or stopping. We would make a few remarks on the subject of bushings generally.

The tapped bushing is very firm, but, in order that it may be well centered, it is essential that its thread fits exactly the tube of the tool (322), and that the pointed rod is exactly central. A turned bushing, especially when a broach can be passed into it after it is in position, is more easily made central (see article 342).

When bushing holes that are rather large with solid bushings, after the hole has been marked with the pointer it must be drilled with a small drill, a larger one being subsequently passed through, so as to increase it. Otherwise there is great danger of the hole turning to one side.

If a hole, such as that of the center wheel, is bushed with a perforated bushing, it will often be found to incline towards the barrel or fusee, so that the hole is displaced. Such an inconvenience may be avoided by using a bushing with a hole smaller than is ultimately required, afterwards enlarging it with the plate centered in the lathe.

561. Riveting of Bushings. Some watchmakers have found considerable advantage in replacing the sudden and irregular impacts of a hammer by gradual pressure, without shock, obtained by a small press worked by hand on the principle of a punching machine. With a well made bushing, the flat end of which is slightly rounded, and the inside of the hole in the plate finished with a rat-tail rather than with a cross file, it is found that the riveting is always perfect. Others employ an ordinary pair of sliding tongs, the noses of which are drilled to receive two punches, one flat and the other rounded, as in the mainspring punch. Three pairs of punches suffice for all sizes of bushings, and the same tool can be used for closing up screw-holes, etc.

562. Movable Bushings. These are for use in regulator clocks and others of large dimensions, and a few words must suffice for their description. They are the invention of M. Alleaume, and will be understood from Fig. 256. It is always desirable, with a view to prevent wear, that when metal pivot-holes are used, the pivot should bear on a length equal to about three times its diameter; but for such a condition to be satisfied, it is essential that the axes of both holes and pivots be absolutely parallel. The figures will at once show in what manner such parallelism is secured. C C is the plate, in section, in which a hole is made of the form indicated by the lines that bound the cross-hatchings. The movable bushing A is held against a shoulder, and prevented from rotating by a screw, the point of which enters a small hole in the bushing. The pivot of B passes into A, and this latter is capable of such slight motion as will insure contact between the surfaces throughout their length.

DEPTHS.

563. To Secure a Good Depth. The least skillful of watchmakers can, without much difficulty, place a wheel in the depthing tool in conjunction with a pinion, and change this latter until the two are found to run easily together. But there are comparatively few that are sufficiently acquainted with the subject of depths to be able to select a pinion whose proportions are such as to satisfy the greatest number of the conditions to be fulfilled by a good depth.

This unsatisfactory state of things is due in great measure to the employment, without any correction, of tables of the sizes of pinions (544), according to which these sizes are determined by a measurement on the teeth of the wheel, taken with a pinion caliper. This method, although sufficient for ascertaining the size approximately, and even for securing a depth that runs more or less easily, cannot be accepted as an unvarying rule.

Far from resting on any mathematical law, as ignorant men urge in their attempt to instruct others, it is only true for a particular number and form of tooth in regard to the wheel, and a definite thickness of leaf and shape of the rounding in regard to the pinion. It ceases to be true if applied to other numbers of teeth, or to pinions that have their leaves thicker or thinner, or the roundings different from those of the pinion first determined upon.

564. Theoretical and practical depths. The fundamental principle of every depth is as follows: To determine what curvature should be given to the teeth of the wheel which drives, in order that the tooth that follows (whether its side be straight or formed according to a pre-determined curve) shall be driven in such a manner as to secure the best transmission of force, a transmission which is in part influenced by the uses to be made of the machine.

565. Teeth formed like the involute of a circle have very marked advantages, but they cannot be adopted in practice, especially in the case of the leaves of pinions. The epicycloid can be realized very approximately in the teeth of wheels in horology, and such teeth can be used in conjunction with pinion leaves having straight faces, the construction of which does not present any difficulty. This explains why the epicycloidal form has been adopted by watchmakers; but, although it is more easily drawn than the majority of other curves, there are still some obstacles in the way of its general application, mainly dependent on industrial requirements. The difficulty is usually got over by forming the tooth according to a circular arc, nearly identical with the epicycloidal curve, see articles (440-42).

566. When two mobiles are of the same diameter, the theoretical depth will be characterized by having teeth and spaces of equal width; but, since in practice the friction with such an arrangement would be excessive, owing to its taking place on both sides of the tooth, the teeth of the wheel that are driven are so far reduced in thickness as to secure the necessary freedom.

567. When the two mobiles are very highly numbered, the lead is short, so that the tooth of the wheel may be a trifle broader or narrower than the space without inconvenience.

But when using pinions of low number (from 6 to 10 leaves), this is not the case. In proportion as the width of the wheel tooth is reduced, its ogive becomes shorter, and the most advantageous portion of the lead (that beyond the line of centers) becomes less. And, besides this, account must be taken of the slipping towards the end of the lead, and the reduction in the difference between the geometrical and the total diameters of the wheel.

568. To secure a good depth with low numbered pinions, the leaves should not be more than half the thickness of the space. If they are thicker than this, it may be found necessary to reduce the width of the wheel teeth, when the pitching is insufficient; but the most serious objection lies in the fact that the pitch circle of the pinion will be diminished in diameter. Let there be two pinions with circular roundings and of the same total diameter, but having leaves of different thicknesses—that with the leaves thick will be found to be too small, etc.

569. To Calculate the Vibrations of a Pendulum or Balance. Multiply together the numbers of teeth of the wheels, starting with the one that carries the minute hand (which therefore makes one revolution in an hour), but exclude the escape-wheel.

Multiply together the numbers of leaves of the pinions, commencing with the one that engages with the center-wheel.

If the first product be divided by the second, the number obtained gives the number of revolutions of the escape-wheel in an hour.

Multiply this figure by twice the number of teeth of the escape-wheel, and the product is the number of single vibrations performed by the balance or pendulum in one hour.

ON THE APPLICATION OF THE GEOMETRICAL LAWS OF DEPTHS TO PRACTICE.

570. It has been urged that when the geometrical forms of the leaves and teeth, as given in scientific treatise, are accurately carried out in practice, the depths are found to be unsatisfactory and liable to cause occasional stoppage; and these facts are brought forward as evidence that theory and practice are at variance.

On the contrary, theory and practice are in perfect accord: the apparent disagreement arises from an error in the application of the geometrical laws.

In copying the theoretical forms of the teeth of wheels and leaves of pinions, it would be necessary to ascertain that they were mathematically exact, and this is impracticable. Two conditions must be borne in mind:

1. Theory shows that the mobile which drives should be made a trifle larger than the geometrical size, so as to counteract imperfections in the workmanship.

2. A pinion is never made of the exact mathematical proportions, in consequence of the processes that have to be adopted for cutting, polishing, centering, etc. If a number of pinions be taken, and if the several dimensions of each be determined by means of a micrometer measuring to hundredths of a millimeter (or from two to three-thousandths of an inch), differences that are, comparatively speaking, large will be found in the diameters, measuring between corresponding leaves; in the thickness of leaves; in the diameters of the circles at which the roundings join the straight faces, and the general truth of the pinion will nearly always leave something to be desired. It should be added that these faults will be more marked according as the leaves have been more quickly made.

The teeth of wheels will be found to be characterized by similar faults, although they are less marked.

571. It follows from these facts that, in watches and timepieces, the pinion is always a little smaller than theory would require; thus the epicycloid should be struck with a somewhat smaller generating circle, and the ogive of the tooth will be proportionately reduced.

The practical conclusion at which we arrive, then, is as follows: As it is impossible to secure absolute perfection in the teeth of small horological mechanisms, their ogives must be slightly more rounded at the points than the designs given in scientific treatises indicate, since these latter are drawn exactly in accordance with the geometrical laws.

These remarks are of the greatest possible importance to the manufacturers of both watches and timepieces; they point to the fact that not only the ogives of all wheel teeth should be lower than theory indicates, but also that, in commoner work, they must be still lower, according as the pinions are of more inferior quality.

572. To Alter a Stem Winding Pinion Depth. The depth of the Stem Winding Wheel and Pinion often occasions considerable inconvenience, and its adjustment requires to be accurately made: when the depth is too deep, its alteration is easy, as the roundings of the pinion leaves can be reduced, or the stud or other piece that carries the winding wheel can have its base a little reduced on one side, so as to set the wheel a trifle out of upright (but so slightly as not to be perceptible to the eye, and taking care that the teeth remain on a level with those of the barrel-arbor wheel). A shallow depth is somewhat more difficult to correct. If a sufficient change cannot be made by altering the support of the winding wheel, one of the following methods must be resorted to:

1. Reset the pendant of the case.

2. Make a new winding pinion of greater diameter, increasing the number of its leaves by one, to correspond to this change.

3. Alter the position of the movement in the case.

The two first methods are more especially applicable to new work, while the third is more convenient for repairers, although of course it can only be resorted to with advantage when the pinion has a bearing in the pendant. The requisite change in the position of the movement can be produced by raising the rim of the case that supports the plate, or by soldering two thin strips of metal on this rim, producing a similar effect; one on either side of the pendant will suffice, except when a considerable change is necessary, in which case they should be set at intervals around the rim to avoid an obvious inclination of the dial. Or four holes can be drilled at equal distances apart around the edge of the plate and in its plane, so that their edges overlap the position occupied by the rim of the case; pins are then set in these holes and filed away until they produce the requisite amount of elevation. Or, again, flat-headed screws may be fitted around the edge with their axes at right angles to the plane of the plate and their heads on the dial side.[7] The depth will then be adjusted by screwing these screws more or less into the plate.

It is advisable to ascertain that the dial is not forced too near the glass, as such is occasionally found to be the case, necessitating the bevelling of the edge of the former.

PALLETS.

573. To Advance a (visible) Jewel in a Pallet. Workmen that have had much experience of escapement making do this without any difficulty by holding the pallet arm in a pair of tweezers that have been slightly warmed, but ordinary repairers will not succeed with such a method: they can however, effect the required change as follows.

Make a small brass plate, E, fig. 257, with a piece c projecting upwards, which the screw v traverses with stiff friction. A saddle b is fitted to the edge of the plate by screws. A glance at the figure will suffice to show the mode of using it; the pallet arm whose jewel is to be adjusted is clamped under b with the jewel just opposite the screw v. Now turn this screw until it stands at the distance from the impulse face of a through which the jewel is to be advanced; taking the plate in a pair of long-nosed pliers, hold them over a small lamp flame, and press with a small screwdriver lightly against the point a so as to advance the stone by the requisite amount as soon as the shellac is sufficiently soft. A particle of shellac is placed at a, if any cavity forms during the process, and the plate is laid on some cool body, avoiding contact with the pallet-staff.

If the stone projects below the lower surface of the pallets, a small washer must be placed underneath before clamping the screws of b, of such a thickness that the stone is just on the level with the surface E.

574. To Alter the Form of a Pallet Face. Workmen that possess the requisite skill and steadiness of hand can alter the form of a pallet jewel, when it is necessary to modify the height or form of the impulse face, by simply using a copper polisher charged with diamond powder. The polishing material employed is always decanted in very pure oil, as otherwise it is apt to scratch instead of polish. The coarser quality is first used when a material change has to be effected, but if only a very slight alteration is necessary, and the adjustment has to be very exact, only the finest quality must be used, as there is a danger of making scratches that would be very difficult to erase. We would also add that this operation requires some skill and patience.

575. To Measure the Lift and all other Angles, etc. of the Lever Escapement. A very simple instrument for measuring these angles was designed by Curzon, one which any watchmaker can arrange for himself, and is quite sufficient for all practical purposes. This is shown in Fig. 258, and consists of an ordinary depth tool to which a scale is added. A hand adapted to the pallet-staff supported between one pair of runners of the depth tool gives motion to a curved rack (shown by dotted lines), and this causes a pinion carrying a second index to rotate, the radii being so related that the movement of the staff is magnified four times on a scale which can be observed while the glass is at the eye examining the pallets. The index which travels over the shorter scale to the left (divided up to 10° on either side of zero) is connected with the pallet-staff by a fork and a short arm passing through the circular groove; it affords a convenient means of moving the pallets while testing them, and gives a measure, in degrees, of their motion. The graduated arc shown at the top is for measuring the lever and roller.

576. Verge Pallets: to Measure their Opening. The little instrument shown in Fig. 259 may be used for this purpose; its mode of action will be easily understood from an inspection of the figure.

One of the pallets being held with its flat face against the base of the graduated semicircle by the lever and spring B, so that the axis of the verge is at right angles to the plane of the instrument through the point n, an index previously fixed to the other pallet will show by the graduations the number of degrees of opening.

This index, shown at P, Fig. 160, must be very light. It is formed in two parts, the body c d, and the small spring z z. The pallet when held in the notch c, must have its face held flat against c d by the spring z z. The face c d of the index must be quite smooth and straight, so as to avoid any error in the reading of the scale.

The pressure-block C, Fig. 259 (shown in plan and elevation at C, Fig. 260), is movable on its center, and this center, which by an engraver’s error is represented on the line n r, should be a little to the right of that line.

577. To Open or Close Verge Pallets. Some workmen cut a notch at the end of a small rod in which the verge is inserted, the two arms of the fork being then drawn together by a screw; then, holding each pallet in a pair of long-nosed pliers, one in each hand, the rod is held in the flame of a lamp and, as soon as the body of the verge becomes blue, it is gently twisted to the right or left according as the pallets require to be opened or closed.

This method is not always convenient, and the following may be recommended:

Support the verge by its shoulders between two cone-plate centers in a pair of finishing turns, as seen in Fig. 261. A carrier b is screwed to the upper pallet, and prevents the verge from rotating; c is a rod through which heat is conducted; a, shown both in plan and elevation, is another rod, which is much longer than c, and has a notch cut at the end, so that it can be forced on to the lower pallet. The end d is free, and the T-rest shown dotted at s, must be brought nearly into contact with it, the distance between them corresponding to the angle to which it is required to alter the opening of the pallets. Now hold a lamp under the free end of c and, as soon as the body of the verge changes color, d will descend by its own weight until arrested by s, the opening will thus be increased or diminished to the requisite extent.

The operation will be accomplished more quickly by directing the blow-pipe flame against the verge body.

Of course when diminishing the opening, the verge must be held in the reverse direction to that shown in the figure.

CYLINDER.

578. To Polish the Cylinder Lips Mechanically. The polishing of the lips of a cylinder is one of the most delicate operations that can be undertaken by a watchmaker; we have, therefore, endeavored to devise an instrument by which this can be done mechanically, and which should at the same time be so simple and so easily made that any watchmaker should be able to construct it for himself.

579. It consists of two distinct parts which take their place in an ordinary pair of finishing turns. 1. The plate P, Fig. 262, supported on a rod T, to take the place of the T-rest. 2. The frame E, whose axis replaces one of the runners. This much being clearly understood from the figure, there will be but little difficulty in understanding the following details.

On the plate P is mounted a bracket, b b b′, held by a screw and washer. It has a slot cut lengthwise, so that on loosening the screw it can be made to slide towards the right or left. The vertical portion b′ supports a fork-shaped piece, d c, a front view of which is given in Fig. 263, pivoted on a collet-screw, f, and this may be fixed by a pin passing through its end like a bolt. The upper end of the fork-piece is provided with teeth for a purpose that will be presently apparent.

The long spindle, g h, turns between the two supports, k, l, fixed to the plate P, under the action of the handle M. This axis carries two eccentric cams, q and R. When it rotates, the eccentric R causes the fork d c to rise and fall, thus occasioning an oscillating movement of the rack d, at the same time the other eccentric q presses against the back of the slide i n, which moves freely in the guide s, and is always held against the cam by a helical spring j; the slide thus has an oscillating motion in the direction of its length.

All the details in regard to the slide and its guide will be easily gathered from the plan in Fig. 262, and the side elevation in Fig. 264.

A small iron polisher is adapted to the slide n i. Being pivoted on a pin at one extremity, serving as an axis, its end u is pressed downward by the light spring v (Figs. 262 and 264), which might be replaced by a spiral spring below the polisher if preferred.

580. This being understood, we will pass to the frame E.

The rod that carries it is formed of thick drawn steel pinion wire, the diameter of which is less than that of the hole in the poppet-head of the turns. This spindle is provided with brass collars at y and z of such an external diameter as to be received in the poppet-head, in which the rod can rotate freely. By adopting this arrangement, not only is the frictional surface diminished, without reducing the accuracy of the adjustment, but the apparatus can be easily adapted to any pair of turns.

To the right-hand side of the frame is fixed, by two screws, the cylinder carrier x shown at X, Fig. 265. It must be removed in order to set the cylinder in position by cementing its balance to the surface; care is necessary to make sure that the back of the cylinder shall be towards the side e of the frame when the carrier is again screwed in position. After having thus replaced it, set the rack d to engage with the pinion wire z, in such a manner that, when the eccentric cam R occupies the position indicated in Fig. 263, the small iron polisher rests at the middle point of the cylinder lip. Now finally clamp the screw that fixes the support T.

The mode of action of the machine will be easily understood. If, after charging the polisher with polishing rouge the handle M is rotated, the cam R will impart an oscillating angular movement to the frame E through its axis y z, and the cam q will, at the same time, cause the polisher to move backwards and forwards, always in contact with the surface of the lip during its movement.

581. The work will be performed more rapidly, and the polish will be better if the iron have a slight lateral motion as well as that in the direction of its length. It is, however, more simple to communicate a longitudinal oscillating movement to the cylinder, and this answers the same purpose; it is only necessary to make two small additions, the spiral spring o and the little cock a. The latter is fixed to d in an inclined position (as indicated at A), and this inclination can be varied by merely turning the left-hand screw. It will be evident that when d is ascending, the cock will push the spindle y z forward; and when d descends, the spindle will be brought back to its initial position by the pressure of the spring o, which is simply placed over the end of the opposite runner. This longitudinal movement must be but slight, and it can be made as little as desired since it depends solely on the inclination of a.

582. Observations. 1. The angular motion of the frame E must be sufficient to enable the polisher to act on the entire surface of the lip. The extent of this movement is determined by the size and the degree of eccentricity of the cam R. The greatest motion will occur when the spindle passes through the hole 1 (Fig. 263), and it will gradually become less as the holes 2, 3, etc., are used. The cam q should also have two or three holes for varying its eccentricity. These cams may be made of hard wood, ivory, etc.

2. The iron polisher may be replaced by a piece of flexible spring fixed by a screw to the slide; but its pressure is less uniform.

3. The bent arm w, Figs. 262 and 265, is clamped to the plate P by a screw d, and the long arm b, Fig. 265, bears against the back of the poppet-head, and thus ensures the steadiness of P. To insure steadiness by its means, b is drawn back in the direction of the arrow, then hooked behind the poppet-head and clamped by the screw d. The firmer the support is the better.

4. The machine may be arranged so that the two lips can be polished at the same time, but it then becomes more complicated. In the tool here described, as soon as one lip is polished the cylinder carrier is unscrewed, turned around, and screwed against the left arm of the frame E, in which are two screw-holes opposite to those in the right-hand arm. Unscrewing the slide b b, the T-rest carrier is moved along the lathe bar until the polisher is over the lip; b b having been set in position is clamped, and, after seeing that w has a bearing, the second lip may be polished.

5. The cylinder carrier shown at X, Fig. 265, is used when the balance is in position. For a plain cylinder without its balance another form of carrier is employed that has at the edge of its central hole a small but solid projecting shell to which the cylinder is cemented.

583. Methods of Obtaining Continuous Motion. Rapid work is not possible when a single handle, as shown at M, is used for working the apparatus; recourse may, however, be had to one of the following methods:

1. Mount a small pinion with a square hole at its center, and make it engage with a large wheel driven by a handle. This wheel, having a great number of teeth, will proportionately increase the rate of motion.

2. Take a powerful clock movement and connect up its center arbor with the axis g h; having wound up the main spring, allow it to run down so long as it possesses sufficient power to drive the mechanism.

3. Fix a ferule at h, and drive it by the aid of a foot-wheel.

BALANCE SPRING.

584. To Select a Balance Spring. Various methods are adopted for this purpose. The most common, by which the strength is ascertained from the length of cone formed by hanging the balance from the inner coil of the spring while the outer is held in a pair of tweezers. A more exact method, based on the same principle, is to employ the small gauge shown in Fig. 266.

A vertical pillar n n is fixed on a smooth plate B, and the slide C is held by friction in any position on n n. Place C so that the distance between c′ and B is equal to the distance between the end of the lower balance-staff pivot and the balance-spring collet. Having now fitted the spring in this collet, raise the balance, by tweezers holding the outer coil, until the lower pivot just rests on B. The graduations on C will then afford a measure of the extension of the spring, and this extension should about equal the radius of the balance measured on the same scale.

When the number of vibrations performed in an hour is known, a spring may be selected by fitting it to the balance and, while holding the outer coils in the tweezers, supporting the lower pivot on a hard smooth surface; the balance is then made to vibrate and the vibrations are counted. The spring need not be pinned into the collet, but may be attached by wax to the top pivot.

585. To Fix a Balance-Spring to its Collet. A common way of doing this is to put the collet on a wire or broach which is held in one hand while the other presents the inner end of the spring, held in tweezers, to the hole in the collet; subsequently fixing it with a pin. The following is a more convenient method:

At the middle of a brass plate is a boss tapped through a vertical hole in its center to receive a small screw with flat head. When the collet is fixed by this screw passing through it, the operation of setting the spring in position and pinning it will be much facilitated, and the plate will at the same time afford a means of testing its parallelism. Two or three screws with heads of various sizes should be provided, and, in order that they may be always available, they should be screwed into holes at a corner of the plate.

This tool might be made of further use by adding an arrangement for holding the stud while drilling it, with a view to ensure that the hole is at the proper height.

586. Balance-Spring Gauge. A back view and side elevation of this are shown in Fig. 267; it can be made without difficulty by any watchmaker.

Through the middle of the plate passes a staff a b lightly pivoted between the cock p and the plate, and projecting on the left-hand side as far as the point a. Between the cock and plate it carries the collet of the spiral spring s and the stop-finger d c, and at the point z is a light finger y z that passes over the graduations on the dial.

When the stop-finger d c is free it stands in the direction of the dotted line i; on rotating the staff, by taking hold of the pivot a, in the direction of the arrow i, the extremity c of the finger will be brought round till it presses against the inclined plane r, which it will force back and, on coming against the stop near c, it will be held fast in the notch of the small bent lever that terminates at r. A spring maintains this lever always against a pin set in the plate. The index finger y z will now be standing over the zero of the scale, and will be maintained in that position until the finger d c is released by a momentary pressure of the hand on the push-piece n, when it will fly back to the initial position corresponding to the dotted line i.

587. The instrument is used as follows: The small sliding holder H, which is shown in section at E, (both of these figures being much enlarged since it is extremely fine and light), has a hole through its center that fits on to the axis at a. Having set a balance spring in the clip as indicated at E, place H on the pivot a, tightening the slide so that it can be used to rotate a b, and bring the stop-finger round to the position d c. Holding the outer coil of the spring in tweezers at v, its inner coil being held in the clip, release the bent arm by means of the push-piece n. The point on the dial at which the finger y z is arrested will give a measure of the force of the balance spring v.

It will be evident that a spring can now be easily selected of the same strength as v, or stronger or weaker within definite limits which will become well known when some use has been made of the instrument.

The entire mechanism is enclosed within a box that is covered by a glass, through the middle of which a hole is made for the passage of a. The spring s is of about the strength ordinarily used for 18-line watches.

588. To Set a Breguet Spring in Position. To test the strength of a flat spiral spring that is to be formed into a Breguet spring, it must first be attached by its collet to the balance-staff. As the outer coil cannot be held in the stud owing to its being so near to the pivot hole, it must be held in the clip b of the little appliance shown at S, Fig. 268. Holding the watch-plate between a and c, the arm D can rise or fall on the rod h, and b can be brought to such a position that it grips the spring at a point just beyond the stud, so that, when the spring is turned inward, the point held may be brought up to the stud. The springing of the watch can thus be proceeded with, and springs tried until one of the required strength is obtained. It then only remains to give the spring the double curvature, and to take care that the end of the overcoil is brought sufficiently near to the center.

Since the action of a Breguet spring is more free than that of an ordinary flat spring, the watch may be found to lose slightly; it is advisable therefore to time the watch before making the bend, so as to show a gain. A little experience will enable the watchmaker to avoid being much out, and any trifling error that there may be is corrected either by a displacement of bend or by altering the central coil. If the latter method is to be resorted to, it is better that the watch should lose rather than gain a little.

589. To Flatten an Ordinary Balance Spring. Remove the collet and stud, and clamp the spring by a central screw between two plates, which are then placed on a blueing tray and gently heated. A small piece of whitened steel is laid on the plate in order to see that the heat does not exceed what is needed to give a blue temper. Allow the plates to cool and separate them.

Ordinary springs being made of rolled steel and subsequently coiled, always open out on heating; it is therefore necessary before resorting to the above method, to coil up the spring, as otherwise the outer turn will be found to have opened beyond the stud.

590. To Diminish the Strength of a Balance Spring. Scraping the end or the entire length always renders the spring defective. Dipping in acid is very little better. It is preferable to embed the spring in cork or soft pith, and work it over a ground glass plate covered with oil stone dust that is fine and smooth. This method might be resorted to for reducing the height of a mainspring.