118 



HOROLOGY. 



B*ap- ways with the same effect, because it was on the same 

 ^ ^ circle, whicheverof the pallets it rested upon ; the impulse 

 Lepaute'7 g' ven was *' so always the same on whichever pallet it 

 escapement. was g' ven > the flanches of the pallets being planes equal- 

 PLATE ly inclined. This was no doubt some improvement on 

 CCC. Graham's; but the teeth of the swing wheel in Lepaute's 



Fig- 7. consisted of sixty small pins, thirty being arranged on 

 each side of the rim of the wheel ; and where pin-teeth 

 are used, oil, which is in some degree necessary, can- 

 not easily be kept to them, the attraction of the rim of 

 the wheel constantly draining the oil from these pin sort 

 of teeth ; an evil which is perhaps not easily to be got 

 the better of, unless by using stone pallets and hard 

 tempered steel pins. 



Recoiling Notwithstanding the seeming superiority and great 

 'soapement. character which the dead beat 'scapement had long ac- 

 Yig- H- quired over that of thereonling one represented in Fig. 8. 

 this last had, however, its partizans ; and among them 

 were artists and amateurs possessed of first rate talents. 

 Such were Harrison, Professor Ludlana of Cambridge, 

 Berthoud, Smeaton, andothers Harrison, indeed, always 

 rejected the dead beat 'scapement with a sort of indigna- 

 tion. The author of the Elements of Clock and Watch 

 Making, has said a great deal in favour of the dead beat, 

 and as much against that of the recoiling one, without 

 having shown in what the difference consisted, or what 

 was the cause of the good properties in the one, or 

 what the defects in the other. It appears doubtful 

 if these causes were known to him ; yet he was very 

 deservedly allowed to be a man of considerable genius. 

 When pallets are intended to give a small recoil, their 

 form, if properly made, differs very little from those 

 made for the dead beat, as may be seen by the dotted 

 lines upon the dead beat pallets in Fig. 6. 



Comparison We shall endeavour to point out the properties and 

 of the dead defects naturally inherent in each : When the teeth of 

 beat and t}j e swing wheel, in the recoiling 'scapement, drop or 



Varments fal1 on either of the P allets > the pallets, from their form, 

 make all the wheels have a retrograde motion, opposing 

 at the same time the pendulum in its ascent, and the 

 descent, from the same cause, being equally promoted. 

 This recoil, or retrograde motion of the wheels, which 

 is imposed on them by the reaction of the pendulum, 

 is sometimes nearly a third, sometimes nearly a half 

 or more of the step previously advanced by the 

 movement. This is perliaps the greatest, or the on- 

 ly defect that can properly be imputed to the re- 

 coiling 'scapement, and is the cause of the greater 

 wearing in the holes, pivots, and pinions, than that 

 which takes place in a clock or watch having the 

 dead beat, or cylindrical 'scapement; but this defect 

 may be partly removed by making the recoil small, or 

 a little more than merely a dead beat. After a recoil- 

 ing clock has been brought to time, any additional 

 motive force that is put to it, will not greatly increase 

 the arc of vibration, yet the clock will be found to go 

 considerably faster ; and it is known that where the arc 

 of vibration is increased, the clock ought to go slower, 

 as would be the case, in some small degree, with the 

 simple pendulum. The form of the recoiling pallets 

 tends to accelerate and multiply the number of vibra- 

 tions, according to the increase of motive force impres- 

 sed upon them, and hence the clock will gain on the time 

 to which it was before regulated. Professor Ludlam, who 

 had four clocks in his house, three of them with the dead 

 beat, and the other with a recoil, said, " that none of 

 them kept time, fair or foul, like the last: This kind 

 of 'scapement gauges the pendulum ; the dead beat leaves 

 it at liberty." Were it necessary, many good proofs 



Escape- 

 ment 



could be adduced of the excellent performance of clocks 

 which had the recoiling 'scapement. 



Let us now make a similar comparative trial with the ConMTiT 

 dead beat 'scapement. An additional motive force be- ( t hc dead 

 ing put to it, we find that the arc of vibration is consid- beat anil 

 erably increased, and the clock, in consequence of this, recoiling- 

 goes very slow. There are two causes which produce ' kca l le - 

 this ; the one is, the greater pressure by the swing ni 

 wheel teeth on the circular part of the pallets during 

 the time of rest ; the other is, the increase of the arc of 

 vibration. It was observed in the case of recoil, that 

 an additional motive force made the clock go fast ; and 

 the same cause is found to make the clock having thc 

 dead beat go slow. As the causes are the same, and yet 

 produce effects diametrically opposite, does not this 

 evidently point out what is necessary to be done? 

 The pallets should be so formed, as to have very little 

 of a recoil, and as little of the dead beat ; and here any 

 variation in the motive force, or in the arc of vibration, 

 will produce no sensible deviation from its settled rate 

 of time-keeping. We have been informed, that a clock 

 was given by Mr Thomas Grignion to the Society 

 for the encouragement of Arts, Manufactures, &c. 

 " which had a dead beat 'scapement, so constructed 

 or drawn off, that any diminution or addition of motive 

 force, would not alter the time-keeping of the clock." 

 All the 'scapements of this kind which have been hith- 

 erto made, were commonly drawn off nearly in the 

 same way as Mr Grignion's, that is, the distance be- 

 tween the centre of the pallets, and the centre of the 

 swing wheel, is equal to one diameter of the wheel, 

 and the line joining the centre of the pallets, and the 

 acting part of them, is a tangent to the wheel, taking in 

 ten teeth, and 'scaping on the eleventh. This is nearly the 

 same as that represented in Plate 1 1. of MrCuming's book. 

 The only difference is, that Mr Grignion's circle of rest 

 is the same on each pallet. But whether it possesses the 

 properties which have been ascribed to it, shall be left 

 to the determination of those who may chuse to try 

 this experiment with it. 



Clock makers in general have an idea, that, in a 

 'scapement, the pallets ought to take in seven, nine, or 

 eleven teeth, thinking that an even number would net 

 answer. This opinion seems to have arisen from the old 

 crown wheel having always an odd number of teeth, 

 because an even number could not have been so fit for it. 

 There seem to be no rules (as some have imagined) 

 necessarily prescribed by either the recoiling or the 

 dead beat 'scapement, for any particular distance, which 

 the centre of the pallets ought to have from that of the 

 swing wheel. The nearer that the centres of the swing 

 wheel and pallets are, the less will be the number 

 of teeth taken in by the pallets, when a tangent for 

 them is drawn to the wheel. It is very obvious, that 

 when the arms of the pallets are long, the greater will 

 be the influence of the motive force on the vibrations 

 of the pendulum, and vice versa, when the pallets 

 are short, the angle of the 'scapement will naturally be 

 greater than may be required, but this can be easily 

 lessened by making the flanches so as to give any angle 

 required. When this angle is not quite half a degree 

 on each side, a very small motive force will keep a 

 pretty heavy pendulum in motion. We have known a 

 very good clock maker, who thought that the flanch of 

 the pallets was an arbitrary or fixed point, which could 

 be made only in one way, and it was some little while 

 before he could be convinced of the contrary. The 

 flanches may be made so long as to act something like 

 detents, so as to stop the wheel altogether by the teeth., 



