US113814A - Improvement in self-acting mules for spinning - Google Patents

Description

UNTTED STATES NPATENT OFFICE JAMES SUTHERLAND, OF EAST HAMPTON, MASSACHUSETTS.

IMPROVEMENT IN SELF-ACTING MULES FOR SPINNING.

Specification forming part of Letters Patent No. 113,814, dated April 18, 1871.

tion; Fig. 2, a plan view; Fig. 3, a sectional elevation; Fig. 4, a side elevation ot' the copbuilder, copping-rail, shoe P, knock-off, &c., all

the other parts being removed. Fig. 51s a plan of the scroll-drum Gr, 8mo. Fi 0*. G is a detached plan ot' the elbow-lever s3, Ste. Fig. 7 is a detached end elevation of the shoe P, Ste.'

Fig. S is a detached elevation of the mechanism by which the position of the weight is regulated. Fig. 9 is a detached end elevation of the knock-off, 85e. Fig. l0 is a detached side elevation of the parts shown otherwise in Fin". 8. Fig. 1l is a detached sectional elevation of the mechanism by which the quadrant-nut is automatically operated. Fig. 12 is a sectional elevation of the drumsE El, 85e. Fig. 13 is a plan ofthe shoe P, Svc. Fig. 14 is a detached elevation of the regulator mechanism. Fig. 15 is a plan of the apparatus shown otherwise in Fig. 11. Fig. 16 is a plan ot the followerrods F1 F2 and followers; and Fig. 17 is a detached plan of a part of the coppin grail, showing the extension-rail o.

The olice of the spinning-mule is the proper building' of the small cylindrical masses of 'thread or yarn having conical ends, which manufacturers term cops My invention consists in divers improvements in the self-acting mule, whereby its efficiency is increased, and it is made capable of building cops cheaper than heretofore.

By way of introduction to the specification of these improvements a brief general description ot' the construction and operation of some ot' the more important of those parts ot' the mule which I acknowledge to be in common use, avoiding detail, will be serviceable.

Suchparts are,rst, alongvertical stationary frame, divided by horizontal parallel bars into spaces of equal size, within which spaces and between which bars are placed in a vertical position rows of bobbins, one row in each space, on which bobbins roving is wound, and from which the tilaments or slivers77 of roving are led through little tubes placed in a row on the aforesaid frame in front of the bobbins,

the number of slivers that are led through one tube depending upon the quality of the thread to be spun, all the slivers that pass through the same tube being drawn and twisted into one thread; second, the twisting mechanism, the chief clement of which is a row of nearly vertical tapering spindles set on the top of, third, a carriage whose length is equal to that of the frame, and which is caused by mechanism which it is needless here to describe to move alternately toward the frame and away from it, or, in the language of mule-spinners, to run in77 and run out,77 the spindles bein made to revolve with extreme rapidity while the carriage is running out, and thus spin the slivers or ends of roving which the carriage draws out from theframe into thread, the movement of the carriage while running out or in being termed a stretch, the rotary motion ot' the spindles ceasing for an instant when the carriage arrives at the end of the outward stretch, and commencing again in the opposite direction at the same moment that the carriage begins to run in, the rotary motion ot' the spindles being much slower while the carriage is running in than while itis ruiming out, and the ends being all carried downward together, during the run ofthe carriage in by a `device called, fourth, a follower, and taken up by the spindles, building in this way the cop, laying upon the same one layer ot' thread at every stretch.

The foundation of the cop is a paper tube placed on the lower end of the spindle. Upon the upper part of the paper tube the base ot' the cop is built. As each successive layer of thread is wound upon the tube and spindle, the cop, of course, increases in size.

The speed of the carriage while running in being invariable in every stretch, and the same in all the stretches, it is necessary that the speed of the spindles while running in should be less during each stretch than during the one next preceding, because the cops take up the thread faster at every stretch, owing to their constantly-increasing size.

It is also necessary that this succession of decreases in the speed of the spindles should continue as long as the diameter of the cop keeps on increasing, which it does till the ta pering base of the cop is completed, a conical mass of thread having been in the meantime built around the spindle above and-resting upon saidtapering base, the cop at this stage consisting, in fact, of two cones placed base to base on the same spindle, When the diameter of the joint base of these cones has attained the size of the diameter which thecop should have, the cylindrical part of the cop begins building, and this is done by causing the follower to place upon the upper of the cone aforesaid successive layers of thread, which are, in effect, hollow cones, all of equal dimensions. the diameters of which, at their bases, are the same as the diameter of the cylindrical part of the cop', and which are laid one above another till the cop is finished, when the cylindrical portion ofthe cop begins building; therefore the series of decreases in the speed of the spindles while the carriage is running in must come to an end, and the subsequent motion of the spindles remain uniform.

But the production of a series of ydecreases in the speed ot' the spindles was not enough. In order to the proper building ofthe cop it is necessary that the series of decreases be itself a regularly-decreasing one-t'. c., that the second decrease be less than the rst, the

v third less than the second, and so on, so that there may follow in the cop the same effect that follows in the bobbin the travel of the belt that connects the concave and convex cones, each reversed with respect to the other, which communicate to the bobbin the motion by which it winds roving from the flier.

The old quadrant-screw may be depended on to produce a series of decreases in the speed of the spindles, but not a regularly-decreasing series of decreases.

The problem before me, therefore, and in the solution of which this invention mainly consists, was to superadd to the quadrant-screw some mechanism that would convert its irregular series of decreases into a regularly-decreasin series of decreases.

' For a particular setting forth of this and the other mechanisms which constitute vmy improvements, I refer to the following description.

In the drawing, A is a mule-head frame, or that part which stands at right angles to the bobbin-supporting frame above mentioned, and also at right angles to the carriage, which -runs on ways, half on one side and the other halfon the other side of the mule-head frame, the adjacent ends of the two halves of the carria'ge being connected by the center-head B,

- which travels back and forth with the carriage on Ways O.

- D is the quadrant, which is of the usual con struction, having a lower forked extremity, one branch of vwhich is pivoted in an arm, l, that extends from the upright A1 of the mulehead frame. Dl is the quadrant-screw; d, the quadrant-nut; E, a drum fixed on the shaft that turns the spindles while running in through the medium of the clip-spring and catch; d2, the chain that connects the quadrant-nut with the spindle-shaft drum E; F,'the drum that operates the quadrant, said drum being driven by'the cord f2, which is wound around the scroll-drum G, that receives motion immediately from the motor-shaft.

In all the foregoing there is nothing new; neither is there anything` new in the function which the quadrant, quadrant-screw, and quad-` rant-nut discharge,.which is,.as the carriage runs in,to unwind the chain d from the drum E by the resistance of the nut d1 moving downward and forward with the quadrant slower than the carriage moves, and thus rotating the drinn E, and, through it, the spindles, in the right direction for winding the ends upon the cops, the decrease in the speed of the spindles at every stretch while thus windin g being et'- fected by turning the quadrant-screw D1, and t-hus raisingthe quadrant-nut di. on the quadrant-screw, said nut, of course, traveling faster, when the quadrant comes down, the further it is moved on the screw aw ay from thequadrantpivot; said nut also, the faster it travels, dragging back less on the drum E, and the speed of the latter being reduced in proportion as the nut drags back less.

The first of my novel mechanisms which I shall proceed to describe is the one that so operates upon the quadrant-screw as to cause it to raise the quadrant-nut at every stretch in such` a manner as tov produce a regularly-decreasing series of` decreases in the speed of the spindles while the carriage is running in and the thread is winding on the cops.

A disk, H, Figs. l, 2,11, and l5, is cast solid with a tube, h, that extends from one side of the disk, the bore of the tube being continuous with the axial hole hl of the disk, and the function of the tube h being to inclose the stud h2, that projects outward from the forked extremity of an arm, h3, which is bolted to the outer side of the standard Al, the stud h2 serving thus to support the tube h and disk H.

A collar, a, is placed around the tube h at the opposite end of the same from the disk H, and is provided with a set-screw, by which the collar maybe made fastto the tube at any desired point. The collar a is also provided with a pin, al, which projects from its end in" a direction parallel with the tube It an d toward the disk H.

AUpon thev outside ofthe tube h a screwthread, h4, is cut over a space that abuts against the disk H at one side, the thread running out at its other end in the exterior of the tube h. From the inner side ofthe disk. H a pin, co2, projects toward the collar to and parallel with the tube h. A collar, a3, is placed upon the tube h, said collar having an interior screw-thread, 71.5, that lits the screw-thread h4' of the tube It, and admits of running the collar a3 upon the screw-thread h4, whereon is its proper position.

Beyond the screw-thread h4 the tube h is reduced in radius by an amount .equal to the depth of said screw-thread. At an interval from the screw-thread of, say, three-quarters ot' an inch in the full-sized machine, a second screw-thread, h6, is Vcut in the outside ot' the tube h, and from this screw-thread to the end ofthe tube the latter is again reduced in radius by an. amount equal to the depth ot' said screwthread.

A second collar, a4, is placed upon the tube h, said collar having at one end an interior screw-thread, hl, that fits the screwthread h6 otl the tube, and admits ot' running the collar at upon the screw-thread h", whereon is its proper place.

A pin, c5, extends from one end of the co1- ]ar c3 toward the disk H. A pin, a6, extends trom one end of the collar a4 toward the collar u..

Fixed on the end ofthe collar a3 farthest from the disk H, is a serrated disk, I, from which a pin, a7, protrudes toward the collar a4.

Parallel with the disk I, and at a little distance from it, a circumferential groove is formed in the collar a3, in which groove sits loosely a gear, b. That part ofthe outside of the collar a3 which lies between the gear b and the disk I is cut into a ratchet, b. A pawl, L2, is pivoted at one end to the side otl the gear I), and its other end engages with the ratchet b'. The pawl b2 connects the gear b and the .i collar a3 whenever the latter is turned so as to cause tlie pawl to engage with the ratchet b. Springs recessed in the collar c3 at the other side ot the gear b press against the latter with such force that when the gear is rotated it carries the collar around with it.

Fixed on the end ot the collar a4 nearest to the collar a3 is a ratchet, b3. A circumferentia-l groove is cut in the collar a4 next to the ratchet b3, in which groove sits loosely a gear, K.

A pawl, b4, is pivoted at one end to the side ofthe gear K, and engages at its other end with the ratchet b3, so as to connect the gear K and collar a4 whenever the latter rotates in the right direction.

Springs recessed'in the collar at the opposite side ot' and pressing against the gear K serve to connect the collar and gea-r with such.

tenacity that when the gear is held stationary while the tube a revolves the collar a4 is held stationary also.

A horizontal pin, al, extends from the side otl the serrated disk I toward the ratchet b, and a similar pin, a8, extends from the side ot' the ratchet b3 toward the disk I.

rIhe pins al of the collar a anda ofthe col lar a4 are placed at the saine distance from the tube h. The same is true of the pins a7 of the disk I and af of the ratchet b3; also of the pins a5 of the collar c3 and a2 of the disk H.

The gear K engages with a gear, L, which is iixed on a shaft, l, that is located in a tube, l1, which extends horizontally from the same arm to which one of the branches of the quadrant is jointed.

Fixed on the opposite extremity of the shaft l, between the two branches at the lower end of the quadrant, is a bevelgea-r, L, and this engages with the bevel-gear L2, that is placed on the lower end of the quadrant-screw.

' The disk H is provided with circumferential flanges c, projecting horizontally to each side of its rim, and with a circumferential groove, cl, in which lies the cord c2, by which the disk H is driven, one branch ot' said cord being wound around a drum, El, that sits loosely on the drum-shaft, the other branch of the cord c2 passing on thence to the belt-stretcher that keeps the cord taut.

A stud, e, is looselyT placed in an orifice that runs transversely through the disk H near the periphery ofthe same. 4

On the outer end of the stud c is fixed a ratchet, el, which is of such a diameter as to project slightly beyond the rim ofthe disk H, through an opening in the outer circumferential flange, c, ot' the disk.

On the inner end of the stud c, at the opposite side ot' the disk H, a pinion, c7, is fixed, so as to engage with the gear b.

A pawl, c2, is pivoted to the outer side ot' the disk H; and from the same side of the disk, at a point diametrically opposite from the pawl-pivot, a pin projects, to which one end of a curved spring, c3, is secured, that presses about at its middle against another projecting` pin, and at its end against the pawl c2, so as to keep the latter in place between any two of the teeth of the ratchet el.

The pawl c2 is provided with acurved arm, c4, which extends outward by the side ot Athe ratchet el through t-he opening in the flange c.

A horizontal arm, e, extends outward from the standard h3, said arm, at a point near its outer end, being bent at right angles to itself', so as toform a tongue, to which a latch, c5, is jointed, said tongue entering a slot in the butt of the latch, and thus preventing the latter from falling beyond a horizontal position.

The latch c5 extends so far toward the rim of the disk H that its point meets the arm c* of the pawl c2, and also meets the ratchet cl during each revolution of the disk I-I.

The foregoing is a full description of the construction ot' the mechanism that so operates upon the quadrantscrew as to cause it to raise the quadrant-nut at every stretch in such a manner as to produce a regularly-decreasing series of decreases in the speed of the spin dies, while the carriage is running in, from the time when the cop begins building to the time when its tapering base, with its surmounting Y cone, is completed.

I will now proceed to describe the operation of said mechanism.

The time at which this description begins is after a dofiing7 has taken place-that is, at'- ter a set of finished cops has been removed from the spindles, the carriage in the meanf time standing still at the end ot' about the iirst halt' of its run inward, at or near which point the carriage is always stopped prior to doiiing. The iirst thing to be donc is to set the collar a at the proper point on the tube hf, which point must vary according to the quality ot' the thread, a iine grade requiring the collar to be set farther out on the tube than a' coarse. grade, which will be made plain'as the description advances.

The next operation is to turn the serrated disk I with one hand forward, at the same time holding the disk H stationary with the other hand until the pin al of the disk I st-rikes the pin a8 of the ratchet b3, when, the rotation of the disk I being continued, the collar a4 is also turned forward until its pin a6 strikesthe pin al of the xed collar a.

'The term forward, applied to the rotation of the disk I, orany of the'pieces that are mounted on its tube It, is intended to mean the direction in which they revolve when the carriage is running in, and the term backward therefore signifies the direction in which these parts revolve when the carriage is ruiming out.

The turning forward of the collars a3 and a4, as above described, so as to bring the latter into contact with the fixed collar a and the former into contact with the collar a4, I term winding them up.77

Afresh set of paper tubes having been placed upon the spindles and the pawl b'Z of the gear .b having been raised from the ratchet b1 of the collar a3, iu order to break the connection between t-he gears b of the collar a3 and K of the collar a4, to the end that the latter may not be rotated, the carriage is set Vin motion to complete its unfinished inward stretch. When the carriage strikes the beam it is stopped, the

pawl b'Z let down upon the ratchet bl, and the carriage started again to make its rst outward stretch for the building of the next set of cops. The cord c2 very soon, 4though not immediately, sets the disk H in rotation. The rotation of the disk H brings the arm e4 of the pawl e2 around against the point of the latch e5, whereupon the latch presses the arm e4 inward until it disen gages the pawl c2 from the ratchet el. Immediately thereafter the ratchet c, in its turn, strikes the latch, and is by the latter, which still presses the arm e4 inward, turned forward untiltwo teeth have passed the latch. By this time the disk H has carried the arm e4 and ratchet el both at the same moment past the latch, whereupon the spring e3 presses the pawl c2 between two ofthe teeth of the ratchet, and the latter is consequently held fast, not being rotated again till it strikes the latch e5 in the course of the next backward revolution of the disk H.

The turning forward of the ratchet el also turns the pinion e7 forward, and the turning forward of the pinion c7 turns the gear b backi ward; but the gear b was already partakin g of one backward rotation--namely, thatimparted to it by the disk H through the medium of the pinion eT and the ratchet c', held fast by the pawl et. If the ratchet c1, while held fast, be the means of communicating a .backward rotation to the gear b, the ratchet, when set in motion in such direction as to communicate additional backward rotation to the said gear,

must accelerate the movement of the latter. As the original backward rotation of the gear b was of the same velocity as that of the disk H, the additional rotation imparted to the gear must give it a higher velocity backward than that of the disk. Such higher velocity, in common with its original one, the gear b imparts to the collar a3, in which it sits, through the medium of the springs recessed in the latter andpressing against the gear, and also through the medium of the spring that presses the pawl b2 of the gear b upon the ratchet bl of the collar gt3. i, Y W

The collar/c2011 receiving a higher velocity backward than that of the tube h, which has, of course, the same speed that the disk H has, turns on the screw-thread 'h6 andinoves toward the disk H as long as such higher velocity continues, or, in other words, until the ratchet el has cleared the latch e5. Then the collar a3 ceases moving toward the disk H, remaining, however, upon the tube [Lin the position it had reached when such movement ceased.

While the collar ashas been gaining ground toward the disk H in the manner above described, the collar a4 has been merely revolving with the tube It, being carried around by the contact between its pin a and the pin d ofthe fast collar a, its gear K meanwhile sitting motionless in the collar, being held stationary by its connection with the gear L and the springs that are recessed in the collar a4 and press upon the gear K, as well as the spring that presses the pawl b4 of the gear K against the ratchet b3 of the collar a4, allowing the collar to slip past the gear and the ratchet to slip under the pawl; but, as one collar moves toward the disk H and the other does not, a gap is thus formed between the two collars that were at :first wound up so close together; or, to use the term which I employ to signify the same thing, a falling-off77 takes place between the colla-rs a3 and a4. Three such failings-off occur during every outward run of the carriage.

When the carriage has reached the end of its outward stretch it stops, and the usual changes take place in the mule-head. Then the carriage enters upon its home-stretch or run inward.

As during the run out, very soon, though not immediately, the cord c2 sets the disk H` in rotation; but this time forward, instead of backward. During the forward rotation of the disk H its ratchet el and the arm e4 ofthe pawl e2 slip under and raise the latch e5 without being operated by it, and the pawl e2 holds the ratchet el fast. The ratchet el, therefore, through the medium of the pin e7, binds the gear b of the collar a3 fast to the disk H for the whole period of the forward rotation of the latter-that is to say, during the whole of the inward run of the carriage; and as the recessed sprii'lgs hold the gear b and collar a3 together, the collar a3 revolves with the tube 7L, and at'the same velocity. 4

Looking, now, to the collar a4, it will be seen that two changes have occurred with respect to it since the carriage began to run in.

Instead of revolving with the tube h, the collar a4 is now sitting upon the tube without revolving, and instead of remaining in contact with the collar a, the collar a4 is now moving away from it toward the collar a3. Both these chan ges are due to the reversing of the motion of the disk` H and tube h, together with the holding of the gear K stationary by t-he gear L.

The reversing of the motion of the tube 7b turns thepin al of the collar a away from the pin a6 of the collar a4, and leaves the latter in the gripe of the stationary gear K, which immediately clamps said collar a4 by means of its recessed springs.

. The collar a, being thus held fast while the tube 71. is revolving, is moved endwise toward the collar a3 by the screw-thread h6. The movement of the collar a4 endwise continues until the pin a8 of the ratchet b3 meets the pin al of the serrated disk l as the latter is revolving forward, which meeting of the pins is, in effect, a collision between the rotating collar a3 and the non-rotating collar a4, in which one or the other must give way. On the instant of the collision the pawl b2 of the gear b seizes the ratchet bl of the collar a3, and thus re-enforces said collar with all the power of the disk H. y The collar c3, thus strengthened, immediately overcomes the resistance of the collar a4, and sets the latter in rotation forward with itself and the tube h.V rlhe collar a4 having thus gone over to the side ofthe collar a3, without any delay the ratchet b3 thrusts itself against the pawl b4 of the gear K, and compels the latter to rotate forward with the rest. The rotation of the gear K effects the rotation of the gear L, and, in train, of the shaft I, bevel-gears Ll L2, and the quadrant-screw.

,The quadrant-screw continues rotating with the rest until the carriage has struck the beam, and during the whole period of its rotation it raises the quadrant-nut d1. The carriage having gone over its inward stretch, the forward vrotation of the disk H ceases, the usual changes take place in the mule-head, and the carriage starts to run out again. During this second run-ont the gear K is withheld from rotation by the gear L, and, consequently, the screwthread 7L draws the collar a4 endwise away from the collar a3, and into contact once more with the fast collar a, thus restoring the gap that was formed during the iirst run-out by .the falling off of the collar a3 from the collar a,- but this gap is increased during the second run-out by three more fallings-off of the fcollar a3. Vhen, therefore, the second run-in is made, the collar a4 has a wider space to move through before colliding with the collar a3 than it had during the first run-in.

As the speed of rotation of the tube h is uniform, the time occupied by the collar a* in traversing the gap must therefore be greater v during the second run-in than during the iirst 5 but the longer the collar a4 is in traversing the gap the less time does it have wherein t( rotate the gears K and L, and, by consequence, to raise the quadrant-nut, because the time of making the run-in is invariable for every stretch. 1t follows, then, that during the second inward stretch the quadrant-nut is raised less than it was during the first, and it is easy to see that during the third inward stretch the quadrant-nut will be raised less than it was during the second, and generally that during every inward stretch the quadrantnut is raisedless than during the one immediately preceding, until, finally, it altogether ceases to rise, the base of the cop being fully built. This gradually-dcereasing series of rises of the quadrant-nut produces necessarily the gradually-decreasing series of decreases in the speed of the spindles, which was the object sought to be obtained.

In order to the proper building of the two inverted cones which form the base of a copsay of No.120 yarnit is required that the quadrant-screw should make during the first stretch one and one-half turn and during the last stretch but the one-hundred-and-twentieth part of one turn, and that between these two extremes the turns'of the screw should decrease in extent in a regular series; or, to put it in another way, in building the base lof the cop the carriage has to make thirteen hundred and twenty home-stretches, and the quadrantnut has, consequently, to make thirteen hundred and twenty movements, each of less extent than the preceding one, in passing over a space of twenty inches on the quadra-ntscrew.

The bare statementof these conditions shows how impossible it is that a human operator should fulfill them, and how necessary it is that machinery should be employed to comply with them. Y

It has been stated that the Xed collar a must be placed nearer the end of the tube h for spinning a fine grade of thread than for a coarser one. The reason is that the finer the thread the longer it takes to build the base of the cop, and therefore the greater should be the number of the rises of the quadrant-nut and the consequent number of fallings-off of the collar a3 from the collar a4.

When the base of the cop is finished it is unnecessary that the quadrant-screw should be further operated. rlhe fallings-off of the collar a? finally, and at the proper time, enlarge the gap between it and the collar a4 to such an extent that the endwise movement of the collar a4 is not sufficient to bring it into collision with the collar a3 during the forward rotation of the disk H. The means thus failing whereby the said forward rotation is communicated to the quadrant-gear L, the latter, of course, ceases to receivepositive motion, and the quadrant-nut ceases to rise. Y

The pin a2 of the disk H is, in the end, overtaken by the pin a5 of the collar a3 in the course of some one of the fallings-off ofthe latter, and this falling-off is, consequently, its

last one. It is best, then, to throw back the latch e5 so that it may cease to operate the ratchet el; but if this be not done, no harm ensues beyond the mere wearing of the parts, inasmuch as when the latch turns the ratchet forward, and thus accelerates the backward rotation of the gear l) beyond that of the disk H, the collar ai is held back by the pins a5 and a? to'just the speed of the disk, ythe recessed springs of the collar a3 yield, and the gear I) slips' round on the tube la without producing any effect until the ratchet el clears the latch e5. e

It has also been stated that after the beginning both of the inward and outward runs of the carriage, very soon, though not immediately, the cord c2 sets the disk H in rotation.

The cause of the cord c2 not setting the disk H immediately in rotation is-thatr the drum El, over which the upper branch ot' the cord- 02 is wound, sits loosely in a screw-thread cnt on its shaft; E2, and when the latter begins to revolve the drum El is turned for about two and one-half revolutions 1n an opposite direc' tion, owing to its connection with the belttightener'on one side and the disk H on the other, neither of which receives motion from any othersource than thedrum El. A

On 4the shaft E2 two collars, d, are fixed by means of set-screws, one at each. side of the 'drum El, a pin, cl3, extending from the inner end of each collar toward the drum, anda pin, d4, likewise extending from each side of lthe drum toward the collars, andthe pins d3 and d4 being all at the same distance from the shaft E2.

As soon as the drum El, by its contrary rotation to that of the shaft E2, is secured along the latter so far that its pin d4 meets the pin d3 of the colla-r d, the drum begins to be rotated in the same direction'as the shaft E2, and consequently to set the disk Hin rotation.

The more idle revolutions the drum E' makes the later is the moment in the run when the disk H begins to revolve. If, say, two and a quarter revolutions of the disk H are found sufficient to vgive the quadrant-screw the proper extent of turn during the first inward stretch and all the subsequent stretches, the collars d are -placed at such a distance from each other that all the time of the run is consumed by the idle 'motion of the drum El, except the time required for making two and a quarter revolutions of the disk H.

The description of the operation of the mechanism which actuates the' quadrant-screw is now complete.

The next in order of my improvements is the apparatus for automatically regulating this same mechanism in such away as, during-thev inward runs of the carriage, to reduce the speed of the spindles whenever they revolve so rapidly as to'windthe thread too tightly upon the cops, as vit is only during that time,

dent, be exceedingly slight. The best indica-l tion of the degree of tightness to which the threads are bein g wound is obviously supplied by the threads themselves, and my aim was, therefore, to make the threads accomplish their own regulating. The problem before me, then, was to apply to the falling-off mechanism an apparatus, tov be automatically operated by the threads, that shall convert the series of decreasing speeds in which-the spindlcs are running at the time they are winding the threads too tightly upon the cops into another series having a more rapid rate of decrease.

The winding of the ends upon the spindles is brought about by the upper followenf, a

section of which, together with a section of the lower follower, f2, is seen in plan at Fig. 16, these sections being those central part-s ot1 the followers that span the space wherein is the mule-head B, between the two halves of the carriage, it being understood that the followers, with their rods F1 F2, extend toeach side of the mule .head from one end of the carriage to the other, the .followers themselves, during the winding on, being just in front, and the follower-rods Fl F2 being just in rear, of the row) of spindles, which row does not extend across above the mule-head. The ends pass between the followers on their way to thespindles. Soon after the carriage arrives at the end of the run-out the upper follower, f1, is made lto descend between the spindles and the lowerfollower, f2, below the latter, and carry the ends down with it for the purpose of placing them in the proper position to be wound upon the cops. During this descent the follower f2 keeps its place above the follower f1, and consequently receives a pressure from all the ends together, which pressure continues during the whole time ofthe run-in, or, in other words, of the winding on, gradually lessening as the follower f1 rises, and nally ceasing' when the follower fl llies above the follower f2, and thus allows-the ends to rise to the points of the spindles. A weight hung to the follower-rod Fzfcounterpoises the pressure of the threads upon the follower f2.

The upperand lower followers and theoperation of the same, as thus far described, are well known in the art of mule-spinning.

My improvement consists in converting the pressure of the ends upon the follower f2 into a force that operates to diminish this -very pressure when it becomes excessive. When the ends are wound too tightly upon the cops they draw the follower f2 lower, and consequently turn the followerrod F2 farther round than usual.

On the rod F2, near the mule-frame B, is fixed a curved finger, f3, the point of which, when the rod F2 is turned far enough, strikes the top of a vertical rod, f4, that is loosely placed in pins f5, which project inwardly from the carriage a3. A collar, j, is fixed on the rod f4, and on the rod, between the collar and the lower pin, f5, is a spiral spring, f7.

A series of teeth is cut in one side of the rodf, so as to form a ratchet, f8,- and a lever, f, having` its fiilcrum on a pin that extends horizontally from an arm, 7c, which likewise projects inward from the carriage a3, is connected with a spring that presses the lower end of the lever ff against the ratchet f8, and causes said lever to operate as a pawl and hold the rod f4 at whatever point it may be pressed ldownward to by the fingerf.

The rod f4 is of such length that, when pressed downward to any extent at all by the finger f3, its lower end, as the carriage runs out, strikes the horizontal arm k2, that extends from a collar, 7c3, which is loosely placed on the shaft l. From the opposite side ofthe collar k3 an arm, 7a4, extends, curving under the,

serrated disk I, and pointed at its extremity, so as to enter between any two of the teeth of the same.

The curved arm 7c" is connected in any suitable manner with a spiral spring, 7c5, which is placed upon a vertical stem, k6, that extends upward from the iioor beneath the said curved arm, the spring las tending to keep the curved arm, and lalso the horizontal arm k2, in proper position.

The arm k2, when struck by the rod f4, is depressed so far as to raise the curved arm kdL until the pointed extremity of the latter meets and stops the serrated disk I during the course of the backward rotation of the latter. The disk I and the collar a3, being thus checked, while the tube It continues to revolve, move endwise along the latter toward the collar a4, or a-re wound up for but a slight distance, however, as the carriage is nearly at the end of its outward run before the rod f4 strikes the arm k2. rEhe winding up of the collar a3 in this'manner diminishes its falling ott', and consequently causes the quadrant-nut to be raised higher than it would have been had the disk I not been thus checked, and the speed of the spindles during the next run-in to be decreased to a greater extent than they otherwise would have been, or, in other words, converts the series of decreasing speeds in which the spindles were running when the ends were being wound too tightly upon the cops into another series having a more rapid rate of decrease; and this was the object to be obtained. rlhis regulation of the falling-off mechanism takes place at about every fourth stretch.

After the rod f4 has discharged its office by depressing the arm 7a2, it is necessary that it should rise to the position it occupied when pressed down by the nger f3, both that it may be in its place ready for the next pressing down by the said finger, and also that it may not impinge upon the arm k2 prior to the next tightening of the ends. A horizontal pin, k7, is therefore fixed in the outer side of the mule-head frame a in such position that it may be struck by the upper arm of the lever ff just as the carriage arrives at the end of its outward run. The striking of the lever fg against the pin k7 disengages the lever from the ratchetf", and also allows the spring f7 to throw the rod f4 upward. At the same time the spring h5 draws the arms k2 and 7c* into position.

I have found it desirable during the run of the carriage iii to cause the follower fz to be held at a higher point than it has been usual heretofore to hold it, in order to reduce the strain of the follower upon the ends. At the close of the run-in, therefore, the follower]LIZ has to be brought down to a point where it shall be below the ends during the run-out. The mechanism by which the follower is thus brought down imparts to it a swift and sudden motion that carries it lower than it should go, so that it has then to rise into its proper position; but this prolonged descent of the follower is sufficient to cause the finger f3 of the follower-rod F2 t0 strike the top of the regir later-rod f4, and thus press the same down ward so far that it will impinge upon the horizontal arm 7a2 during the next run-out, and, in the manner above described, effect a reduction in the speed of the spindles beyond what is required. To prevent such a result the rod f4 is divided into two portions, the upper portion or head, m, being two or three inches in length, and jointed to an ear, ml, that extends upward from the lower portion in such a manner that the head, when vertical, forms a part of the main rod f4, and may be operated by the finger f3 in the usual way.

An arm, m2, extends horizontally outward from the roller-beam N far enough to strike the lower end of the head m when the carriage has arrived at a point about five inches from the beam, and throw said headinto a position so far inclined that it is out of the reach of the finger f3, however far the same may descend, in which position the head m is held by the arm m2 until the carriage has completed its run-in and proceeded on its run-oiit far enough to. carry the head clear ofthe arm. The head is then restored to the vertical position by means of a weighted rod, m3, that extends from its lower end in a direction away from the arm m2.

As before observed, the regulation of the falling-off mechanism, for the purpose of preventing tlie ends from being too tightly wound upon the cops, needs only to take place during the time when the double conical bases of the cops are buildin g. Sonie apparatus is, therefore, needed which shall automatically render the regulator inoperative as soon as the base is finished.

, vertically near its rear end, so as to form. an

1 spring kB, thus breaking the connection' be- `In a subsequent .part of this specication will be found a description of a certain` trav-` eling block, P. Without at this point antici-y patin g that description, it is sufficient to state that it is the block I? which is made use of to render the regulator inoperative, and this in` the following manner: The arm k2, like the rod f4, is constructed in two parts, the end piece or head k7 being jointed at its rear extremity to the front end of the main arm, and being also connected with the ma-in arm by a spiral spriu g, k", which is extended whenever the head k7 is turned forward into line with the main arm, and which,l therefore, turns the head k7 backward toward the main arm whenever the force that turned it forward is withdrawn. This force is supplied by an inclined rod, kg, that extendsl horizontally backward from the side of the block P, and is turned up arm, c, that is high enough to operate as long as lthe block P is sufciently near as a stop to hold the head k7 in line-with the main arm 7a2, and thuskeep good the connection between the regulator-rod f4 and the falling-off mechanism.

The block P. remains sufficiently near the arm k2 to enable the arm klo to operate in this way until the base of the cop is nished, by which time the block l? has traveled onward so far as to carry its arm kw clear of the head ,707, which thereupon ies back, drawn by the tween the rod f4 and serrated disk I, or, in other words, rendering the regulator mechanism inoperative.Y

Having thus completed the description of the construction and working of the regulator mechanism, I propose now, by way of preface to the explanation of the next of my improvements, to advert once more to the manner of operation of the upper follower during the windin g of the ends upon the cops.

Beginning` at the point in the cop where it places the bottom of the layer of thread at the commencement of the run of the carriage in, the follower rises slowly till it reaches the top of the cop. At this point itis my object to impart to the follower a quicker movement as it rises past the lower portions of the naked parts ofthe spindles, and before it reaches the tops of the spindles, to impart to it the still quicker motion by which it arrives atits original position above the spindles just as the carriage strikes the beam. These three separate movements of the follower I vcall its rSt, second, and third rises, respectively, and it is during the second and third rises that the threads are wound spirally upon the naked parts of the-spindles up to their very points, in order that the ends may be properly twisted during the next drawing out.

The rst rise of the follower is effected in the ordinary manner by the use ofthe coppingrail S, the sinking of which at the end of each outward run causes the roller 0, through the intermediate mechanism, to place the upper follower opposite ahigher point in the base of the cop at the beginning of each successive inward run, thus enabling said follower to place the top of each successive layer of thread at a higher point on the spindle, whereby the cop grows in height at every stretch.

It is evident that the first rise in every inward run should continue until all of each end is wound upon the cop, except just enough to compose the spiral around the spindle, and that then the second rise should begin. The second rise cannot begin till the roller o has traversed the length of the copping-rail.

The length of the copping-rail must be such as will produce a rst rise of the right length for the first inward run, in which the amount of thread required for the spiral coil is greater than in subsequent runs.

In the ordinary copping-rail there is made no provision for producing a subsequent first rise of greater length than that in the first iu ward run, it having been supposed that first rises may be invariable as to length without detriment.

Experiment, however, has proved th at every rst rise should be of greater llength than the one immediately preceding it, for the reason that the cops, being always conical where the layers of thread are laid on, take up the ends with a speed that diminishes as the ends approach the tops of the cops, so that each first rise should be prolonged till the carriage has run in so far as to leave between itself and the beam only just sufficient length -of ends to form the spiral coils; and as the length of ends requisite for the spirals is less at each successive stretch, the length of the first rises must be greater at each successive stretch. I f the first rise ends too soon, the spiral coi-ls are Wound loosely on the spindles, and kinks are consequently formed in the ends. It was, therefore, necessary for me to provide an apparatus whereby every rst rise of the followers may be prolonged till all of each end has been wound upon its cop except what is required to form the spiral coil upon the naked part of the spindle. n

The device by which this object is accomplished I term the extension-rail.77 It con sists of a block of metahp, Figs. 2 and 17, placed in a longitudinal groove made for its reception in the upper side of the coppingrail S, at the inner end of the same. The upper side of the extensionrail is at' itsrear part horizontal, and always iiush with the upper surface of the copping-rail. At its front part the upper side of the extension-rail is beveled to such an extent that when set back far enough `in its groove its incline face is flush with the inclination at the end of the coppingrail. When in this position the extension-rail forms merely a yconstituent part of the cop. ping-rail, having no function of its own to discharge, and this is its position at the placing of the lirst Ylayer of thread upon the paper tubes and spindles.

In order to cause the follower f1 to rise higher while placing the second layers of thread upon the cops than it did while placing the iirst ones, the roller o must run farther out upon the copping-rail before beginning to descend; but the roller o ran to the end of the copping-rail during the iirst stretch, and to enable it to run any farther out the coppingrail must, it is obvious, be extended. To extend the copping-rail for this purpose is the office of the rail p, which office it fulfills by moving out in its groove a very slight distance at t-he end of each outward run. The source of this intermittent motion of the extension-rail is the traveling shoe S, on the inclined upper surface of which the copping-rail S rests. To the side of the shoe S1 one extremity of a rod, S2, is jointed, the other extremity of which is connected with the lower end of a lever, S3, that has for its fulcrum, near its upper end, a stud passing transversely through the eopping-rail S. To the lever S3, at a point between its fulcruni and the connecting-rod S2, is jointed one extremity of another connecting-rod, S4, the other extremity of which is connected with the extension-rail p by a pin that projects from the side of the latter through a slot in the side of the copping-rail.

Vhen the shoe Sl is moved slightly forward, in the usual manner, by the turning of the screw Q it imparts a still slighter movement, through the medium of the lever S3 and connecting-rods S2 S4, to the extension rail p, which movement, minute as it is, is still amply sufficient to extend the copping-rail all that is necessary for the prolongation of the rst rises of the follower f1 during each successive inward run, until the length of ends between the carriage and the feeding-rolls is only just enough to form the spiral coils without producin g kinks.

The distance traversed by the extension-rail during the building of one set of cops is one and three-quarter inch. In spinning 1Z0-yarn the number of inward runs made by the carriage during that time is two thousand six hundred and forty. r)Ehe movement of the extension-rail at every run is, therefore, .00066 of an inch.

rlhe first rise of the follower f1 being completed, its second rise at once begins. This is produced by the running of the roller 0 down the inclined face ofthe extension-rail.

rThe second rise, of course, begins at a later moment in each successive stretch, owing to the prolongations of the first rise; and this it should do, as the bases of the naked parts of the spindles are changed to higher points during every inward run. rlhe second rises must, it is obvious, gradually decrease in extent the cops grow in height; but notwithstanding this no second rise should terminate till it has reached a higher point of the spindle than the one immediately preceding it, for the reason that, as the speed of the spindles is regulated to give the requisite velocity to the outside of the largest part of the cop, it is not suiicient to take up the ends into the spiral coils on the naked parts of the spindles to the degree of tightness necessary to prevent the formation of kinks without prolongation of the second rises. Noris even this sufficient, a quickening of the speed of the spindles during the second and third rises of the follower being also necessary, as will be explained hereinafter. Each third rise should, therefore, begin at a later moment in every successive stretch than the one immediately preceding it.

To produce the third rise I make use, in a novel manner, ofthe old device known as the knock-off,77 (shown in Figs. 4: and 9,) and consisting of a block, n, the upper surface of which is formed precisely like that of the extensionrail, its incline face being turned toward the copping-rail. Formerly the knock-off was rigi idly connected with the coppiug-rail, and received no motion except what it derived therefrom. The ofce of the knock-off then, as now, was to raise the fork a5 as the lower nger of the same was drawn over it by the carriage during its inward run, the rising of which fork effects the release of an arm, n3, that is connected with one of the curved fingers a4 of the follower-rod F1 from a pin which holds said arm in lock. On the release of the arm n3 the weight R rotates the follower-rod Fl suddenly backward, and thus puts the follower f1 through its third rise. The knock-off was found to operate in the same faulty manner as the copping-rail, to which it was attachedthat is, it released the upper follower too soon in all the stretches subsequent to the first, and allowed kinks to form in the ends. I therefore found it necessary to devise some method whereby the knock-ott' may be made to throw the upper follower out of lock, or, in other words, to cause each third rise of the follower to begin at a later moment of each successive stretch. The best way of accomplishing this object seemed to be to cause the knock-off to move forward a little at every stretch in the same manner as the extension-rail, and the same source of motion-namely, the intermittent movements of the shoe S1 of the coppingrail`was as ready to hand in the one case as in the other but it was necessary that the motion drawn from this source for the knock-off should be less at every stretch than that for th'e extension-rail, for the reason that inasmuch as the actual extent of each second rise of the follower must diminish at every stretch, while at the same time each second rise must end at a higher point in each successive stretch, the decrease at the end of each secondrise must be less than that at its beginning, and consequently the device which ends each second rise must move less at every stretch than the device which begins it.

The device that begins each second rise is the extension-rail. The device that ends it is the knock-olf, and the knock-off must, therefore, move less at every stretch than the eX- tension-rail.

In order to render the knockoff movable lengthwise a dovetail Vgroove is formed in it longitudinally of one side, by means of which groove the knock-oft' is slipped upon a horizontal tongue, a6, Fig. 9.

The tongue a6 is bolted at a suitable height to the side of a vertical tongue, nl, that is placed in a groove which runs lengthwise kof a vertical standard, u2, that extends upward from-the iloor at a point over which the fork a5 passes during the runs of the carriage in and out.

In order to impart to the knock-olf the requisite movements to its outer side, one extremity of a connecting-rod, nl, is jointed. the other extremity of said rod being pivoted, to the lower end of an arm, u8, that extends downward from a stud, which passes transversely through the copping-rail S, said stud being the same to which at its opposite end, on the other side of the copping-rail, the upper extremity of the lever S3 is affixed.

By means of the stud the motion derived by the lever S3 from the shoe Sl is communicated to the knock-off.` The arm a8 is made as much shorter than the distance on the lever S3 from its fulcrum to the point of its connection with the pitman S1, as may be necessary to make the movelnent of the knock-off at every stretch less to the proper degree than the movement of the extension-rail.

The functions ofthe exten sion-rail an d knock# offhaving been thus explained, the next of my proper velocity to the outside of the largest part of the cop, which is the one the spindles have during the lirst rise of the follower, is not sufficient to venable them to wind the spiral coils around the naked parts of the spindles tightly enough to prevent the formation of kinks in the ends. VIt is obvious that the said quickenin g` should begin at the moment in each inward run of the carriage when the ends have risen to the top of the cop, or, in other words, should follow the course of the iirst rises of the follower. It must, therefore, be of greatest duration in the rst stretch, and decrease in duration at every successive stretch. To accomplish this object l make use ofthe shoe P, Figs. 4, 7, 13, referred to above in the course of the explanation of the regulator mechanism.

The shoe P has a dovetail groove running lengthwise of its under straight side, by means of which groove the shoe is slipped upon a dovetail rail, P1, that is secured to the door in the path which the shoe should travel. The upper surface of the shoe is inclined backward and downward. A rod, P?, jointed at' each extremity, connects it with the copping-rail shoe or copbuilder Sl, by which rod the'intermittent motion of the cop-builder is communicated to the shoe P.

To one side of the upper curved rail of the quadrant D, and near the forward corner of the same, a bent lever is pivoted, its fulcrum being at its angle, the longer curved arm db' of which lever is bent laterally so far as to pass above the quadrant-chain d?, the major portion of said lever being on the opposite extremity strikes the shoe P just as the roller begins to descend the inclined face of the extension-rail. A spiral spring, d'0, fastened at one extremity to the quadrant and at the other to the striker d8, tends to keep the latter always in proper position, and a spiral spring, dg, fastened at one extremity to the outer end of the shorter arm (17, and at the other extremity to the front rail, D2, of the quadrant, tends to keep the bent lever in proper position.

The operation of the above-described mechanism is as follows: The shoe P having, be it understood, been so set that its highest point shall be struck by the rod d8 at the end ofthe Iirst rise of the follower f1 in the iirst stretch, the impact of the, striker against the shoe throws the form er upward, and also throws upward ythe outer extremity of the shorter arm 17. The elevation of the arm d1 causes the depression of the arm d, and the latter is so much longer than the former that the pin d5 moves downward considerably faster thanV the end of the arm dl moves upward, and, therefore, speedily strikes the chain dz and drives the same downward before it. Before the chain d2 receives the stroke of the pin d5 it is already tautened by the draft of the drum E moving forward with the carriage, and, therefore, before the bending of said chain downward by the pin (15o-an be cffected there must be 'a yielding of the chain d2 at one end or the other. Such yielding takes place at the end wound on the drum E, the latter revolving with a suddenly-increased velocity in order to let off the chain. This suddenly-increased velocity of thedrum E produces the quickening of speed in the spindle, which was the object to be obtained, said quickening beginning at the same timev with the second rise of the follower j", and continuing as long as the quadrant descendsthat is, till the endof the stretch. At every inward run the shoe P is moved as far forward as the other shoe, S, moves, and, therefore, by reason of the inclination of the upper surface of the shoe P, the latter is struck by the rod d, and the chain d2 is struck by the pin d5, each at a later moment of every successive stretch, and the duration of the quickened speed of the spindles, therefore, decreases at every run. A certain amount, however, of quickening is essential to the 'proper operation of the spindle, even to the very last stretch. The moving of the extension-rail, knock-off, and shoe P, all by the shoe S, which is usually called the cop-builder, renders it easy to regulate the motion of each in its proper relation to the motions of the others. During thebuilding of one set of cops of 1Z0-yarn, or, in other words, the running of two thousand six hundred and forty stretches, the shoe l? moves lengthwise five inches. The height of the incline in the upper side of the shoe is 1.25 inch. The extent ot' the movement of the shoe at every stretch is, therefore, .00189 of an inch, and the average decrease of length in the movements of the arm d6 at every stretch is .00047 of an inch. As the carriage runs in after completing the second stretch the drum zis revolved backward by the cords f 2 and The drum G is connected with the large gear G3 by two interlocking' ratchets, g g1, one fixed concentrically upon the inner side of the gear, and the other connected by a collar with the drum. The gear G3 turns only forward, and the drum G must, therefore, be disconnected from it before said drum can be revolved backward. The disconnecting of the drum and gear stops both the drum and the carriage, and therefore should not take place till the carriage has completed the second stretch. It is at this precise moment that it does take place, and the mechanism by which it is accomplished is one of my improvements, the ratchets g g1 themselves being old.

An arm, s, extends downward to a suitable distance from the bottom rail of the carriage a?, and from said arm, near its lower end, a horizontal pin projects inwardly.

A lever, s', is jointed at its lower extremity to a lug that extends upward from the iioor directly beneath said bottom rail of the carriage. The lever sl is of such length that when it stands upright the pin of the arm s strikes it and throws it backward just as the carriage completes the second stretch. g

A horizontal rod, s2, connects the lever s1 with the outer end of a horizontal elbow-lever, s3, pivoted upon a vertical fulcrum that extends upward from the iioor directly beneath the drum G, by which connecting-rod the backward movement of the lever s' is communicated to the elbow-lever s, causing the inner arm of the same to swing outward till it strikes and throws forward the lower end of a vertical rocking-bar, s4, that is pivoted to a horizontal arm, S5, which projects from the adjacent standard of the mule-frame.

From the upper end of the bar s4 a pin extends backward, entering a circumferential groove, s, in the collar sl, that is attached to the inner end of the drum G. vThe throwing of the lower end of the bar s4 forward draws the collar sl, and with it the drum G, inward till the collar strikes the mule-frame, and this drawing of the drum throws out of gear the ratchets g and gl, and leaves the drum free to rotate idlyon its shaft as the carriage runs in.

The means for throwing the ratehets into gear again are old.

The sixth of my improvements is the apparatus whereby, when the carriage strikes the beam at the end of the run inward, it is prevented from suddenly recoilin g with such force as to break the ends.

v To the upper surface of either of the side rails of the mule-head frame a block, r, is secured in position where it may form the foundation of said mechanism.

To a vertical lug, r1, that springs from lthe upper side of the block r, the rear extremity of a latch-bar, r2, is jointed, which latch-bar rests, near its front end, upon a spiral spring, r3, that is supported upon th'e block r.

A latch, r4, passes vertically through a slot lin the latch-bar r2, and a horizontal pin running transversely through the latch-bar, slot, and latch forms a pivot upon which the latter may be turned.

A spiral spring, r5, connects thelower part of the latch r4 with the lug r, said spring tending to draw the lowerrpart of the latch backward, and thus throw its upper part forward.

The latch r4 should be pivoted at such an elevation as to stand directly in the path ot' the follower-rod Fl as the latter moves backward and forward with the carriage.

The operation of the above-described mechanism is as follows: hen the follower-rod F1, in running in, strikes the latch r4. it throws the same forward, depresses the latch-bar r2, and passes on beyond the latch, whereupon the latch-bar is thrown upward by the spring r3 to its original position, earryin g up with it the latch and placing the latter directly in rear of the bar F1. No sooner has the latch reached this position than the recoil of the carriage brings the follower-bar back against it, which recoil is checked by the resistance offered by the spring T5 to the pushing back of the upper part of the latch.

When the carriage runs outward the follower-bar overcomes the resistance of the spring r5,`pushes back the latch, depresses the latchbar, and passes clear of the latch, which thereupon is again elevated to its original position.

The seventh of my improvements relates to t-he manner of keeping the pressure of the weight It off from the arm t until the same has been raised past the center of the followerrod F1, the necessity for which improvement arises from the fact that, prior to such passing the center, the pressure of the weight, if not taken on, would tend to draw the arm t downward at the very time when it is necessary that it should move upward.

I have therefore provided a rail, R2, beneath the carriage for the weight It to travel on, said rail being fastened to the iioor in a direction parallel to the motion of the carriage, and being inclined at its inner end to such an extent and at such a pointthat the weight reaches the incline. just before the knock-oifthrows the upper follower out of lock, and travels down the incline fast enough to impart to the follower a third rise of the vrequisite velocity. As long as the weight is on the horizontal part of the rail it is wholl y sustained thereby.

The last ot" my improvements relate to the means whereby the weight R is made to retard the second rise of the upper follower to an eX- tent that increases directly as the cops grow in height. The necessity for such increase of the retardation grows out of the fact that it' the upper follower were to move at a uniform speed during all its second rises it would get into position for the outward run sooner and sooner at every stretch, by reason of the constant decrease in the length of the second rises. But it is essential that the follower should get into position for the outward run at the vsame moment in every stretch, which moment is the same at which the carriage strikes the beam, in order that it may put the proper amount of thread into the spiral coils.

The weight R therefore is placed loosely on an arm, R3, the forward extremity of which is jointed to the carriage, while to its rear eX- tremity the lower end of the chain R1 is fastened. The weight is connected, by a rod, r6, with the peripl'ieryot' a drum, R4, that is Xed on a shat't, R5, which passes horizontally through a sleeve, R6, that. is secured to the bottcmrail ot' the carriage, and also through a ratchet, R7, fixed on the inner end of the sleeve RG.

An arm, R8, extends radially from the part of the shaft R5 that projects beyond the ratchet R7, to the outer side of which arm is pivoted a pawl, R9, which pawl a spring, r", affixed at one end tothe arm R8, presses upon, thereby causing the pawl to en gage with the ratchet R. Vhen the pawl is thus engaged with the ratchet the arm R8 is locked to the same.

An' arm, T8, projects horizontally outward from the shoe P, directly in the path of the arm R8 as the latter moves back and forth with the carriage. As the carriage runs in, the lower arm ot' the pawl R9 first strikes the arm rf", and the pawl is thereby disconnected from the ratchet, thus releasing the arm 'R8 from lock, and admitting of the throwing of the lower end of said arm backward until it clears the arm r. The throwing back of the arm rotates the drum R4 backward, and this, by means of the connecting-rod, strikes the weight It back on the arm R3.

Thearm ra is provided with a jointed head, that is kept in line with the arm by a sprin When the carriage runs out the spring yields and allows the jointed head to turn backward as the arm R8 strikes it until the latter passes clear of the arm T8, when the spring throws the jointed head forward again into position tor the neXt throwing back of the arm.

By each successive movement of the drum the weight is pushed backward, and asy the distance between the weight and the pivot of the arm R3 increases the pressure of the weight upon the chain Rl, and, by consequence, upon the arm t1 of the follower-rod f1, increases also. By reason of the increasing pressure ot' the weightupon the follower-rod as the lengths of the second rises of the follower decrease, the times of said rises remain the same.

Another advantage of this gradual retardation of the rises of the follower is the winding of the threads tightly upon the apexes or noses of the cops. The rail R2 serves to support the weight during that part of the run when the retardation of the follower is not necessary and would be injurious.

I deem it a suitable conclusion to this' specification, notwithstanding it has already attained an unusual. length, to state that my object in making the above-described improvements was to conti-ive machinery that should aut-omatically imitate, as closely as possible, the action of the hand-mule spinner, and also to set forth the resemblances'which I havesucceeded in producing between the movements of the man and the machine.

The hand-mule spinner is the only source of the rotary motion of the spindles while the carriage is ruiming in, which motion he produces by turning with his hand a fly-wheel mounted on the drum-shaft with which the spindles are connected.

He also has to operate the upper follower by depressing it just prior to the starting ot' the carriage to run in, by allowing it to move gradually upward as the carriage advances, and by letting it rise suddenly into position tor the run-out as the carriage strikes the beam.

In the operation ot' thev spindles ythe mule possessing my improvements does perfectly what the human spinner aims todo, but with only partial success.

l First, by producing a regularly-decreasing series of decreases in their speed, so that not only do they rotate more slowly during-the second stretch than during the first, and during the third stretch than during the second,

and so on till the bases of the cops are built,

but also is the ditt'erence between the respect-ive velocities ot' the second andthird stretches less than the difference between the respective velocities of the first and second stretches, andY so on; also, by diminishing their speed, when they are running too fast, just as much as may be necessary to prevent them from winding the ends too tightly upon the bases of the cops. These things the handspinner did, so far as he could, by slowing his ily-wheel from stretch to stretch, and in the course of those stretches in which ,the tightening took place to what, in his judgment, was the proper extent.

Second, by imparting to the spindles at every inward run a quickened speed ot' less and less duration vat every stretch, just as the spiral coils commence winding upon the naked partsof the spindles. This the handspinner did by increasing the velocity of his ily-wheel in the course ofthe stretch at a moment and tov an extent depending entirely upon his ownjudgment. H

In the operation of the upper follower my improvements perform the work of the ideal spinner, first, by changing to a later moment in every successive stretch both the time of beginning and of ending the quickened :movement of the follower that lays the spiral coils upon the naked parts of the spindles, and by giving to this movement the degree of quickenin g requisite to form the coils without allowing kinks to come in the ends; second, by retarding the speed of the successive rises of the the upper follower directly as their lengths decrease, by just so much as may be necessary to produce uniformity in the times of said second rises, third, by changing to a later moment in every successive stretch the time when the weight is allowed to draw the follower suddenly upward into position for the outward run of the carriage.

All these matters in the hand-mule are left entirely to the head that dictates and the hand that regulates the rise of the follower.

In preventing the recoil of the carriage after it strikes the beam my improvement imitates the pressure of the knee and hand of the human operator, by opposing to the follower-rod a force that checks it temporarily, and is withdrawn when the carriage starts to run out.

Having thus described my invention, what I claim as new, and desire to secure by Letters Patent, is

l. The fall-off mechanism herein described, substantially as set forth.

2. The combination of the quadrant-screw with the winding-on drum and with the fallingoff mechanism, as described, for the purpose of producing a regularly-decreasing series of decreases in the speed of the spindles.

3. The combination of the mechanism herein described which automatically turns the quadrant-screw with a regulating apparatus, substantially such as is herein set forth, that, during those inward runs of the carriage in which the spindles revolve too fast, is automatically operated by the ends in a manner that diminishes the falling ott' of the collar a3 to such an extent as to cause the mechanism of which said collar forms a part to decrease the speed of the spindles during the next inward run after such diminution more than it would otherwise have been decreased, and enough to prevent the spindles thereafter from winding the ends too tightly upon the double conical bases of the cops until such time as they again revolve too fast.

4. The combination of the vertical regulatorrod f4 with the jointed head m, weighted rod m3, and horizontal arm m2, in the manner described, and for the purpose of preventing the depressing of the regulator-rod out of season.

5. The arrangement of the horizontal arm k2, jointed head k7, spring 7c, or its equivalent, and rod lag, projecting from the shoe P, in the manner described, and for the purpose of automatically rendering the regulator apparatus inoperative after the double conical basesl of the cops have been built.

6. The extension rail, combined with the copping-rail and cop-builder in the manner herein described, by which the intermittent motion of the latter imparted to the extensionrail may cause the same to fpractically lengthen the copping-rail at every inward run as much as may be necessary in order to prolong the first rise of the upper follower, until all of each end has been wound upon its cop, except just enough to make the spiral coil upon the naked part of the spindle without allowing the formation of kinks in the thread.

7. A knockoft` separated from the coppingrail, and operated by a mechanism substantially such as herein described, which causes it to advance at every stretch a less distance than the extension-rail advances, but at the same time as far as may be necessary to prolong the second rise of the upper follower at ,each inward run, until the spiral coils have been wound upon the naked parts of the spindles tightly enough to prevent the formation of kinks in the ends.

8. The combination, with the quadrant and quadrant-chain, of the inclined traveling shoe l?, and the mechanism consisting substantially of the lever d di and striker d8, all constructed and operating in the manner described, and for the purpose specified.

9. The combination of thecop-builder, extension-rail, knoekoif, and shoe P, in the manner described, and for the purpose of causing them all to move in unison.

10. The combination of the scrolledrum with the rocking-bar s4, elbow-lever s3, connectin grod s2, vertical lever s1, and carriage-arm s, in the manner described, and for the purpose ot' throwing the scroll-drum out of gear at the completion of the second stretch.

1l. The combination, with themule-carriage, of a latch mechanism consisting, essentially, of the block r, lug r1, latch-bar r2, latch r4, and springs r3 v5, as and for the purpose specified.

12. The combination of the weight R with the rail R2, when the same is provided with an inclination at its inner end, in the manner described, and for the purpose of keeping the pressure of the weight ott' from the upper follower until the latter has been thrown out of lock.

13. The combination of the weight R with a mechanism, substantially such as is herein described, that automatically serves to increase the pressure of said weight upon the upper follower directly as the lengths of the second rises of said follower decrease, in the manner described, and for the purpose of causing the times of said second rises to remain uniform.

JAMES SUTHERLAND.

Witnesses:

,HARLEs MELLEN, WILLIAM STRATTON.

Images