136 



1 1 R O L O G Y. 



I'ompensa- 



tiiin Hi- 

 lances. 



Rcrtluiud'i 

 first com- 

 pensation 

 iu;<iinst heat 

 suit cold. 



kludge's 

 compensa- 

 tion by a 

 spiral of 

 brass and 

 steel solder- 

 ed together. 



Le Roy's 

 compensa- 

 tion by com- 

 mon ther- 

 mometer 

 tubes and 

 balls 

 PLATE 

 CCCIV. 

 Fig. 1. 



several places. Perhaps those in the three former dif- 

 fered litttle or nothing from this. 



The first pocket watch made in Europe with a com- 

 pensation was by F. Berthoud. It was begun in 176", 

 and finished in the beginning of 1764; and was sold 

 in London in 1766, to Mr Pinchbeck, for his Britannic 

 Majesty King George the Third. The compensation 

 was effected by laminae of brass and steel pinned to- 

 gether ; one end of which being fixed to the potence 

 plate, the other acted on a short arm from a moveable 

 arbor, a longer arm having the curb pins in it, being 

 made to move nearly in the circle of the outer coil of the 

 spiral spring. It had a common crown wheel and verge 

 'scapement, and a going in time of winding. The balance 

 was so heavy as to set, being sixteen grains in weight, and 

 the vibrations were four in a second. Mr Kendal adopted 

 this mode of compensation in some of his pocket watches. 



Mudge, some time before the year 1770, made a 

 watch for Mr Smeaton, in which the compensation was 

 effected by two long slips of brass and steel soldered to- 

 gether; being dressed up, it was turned up into a 

 spiral, as close together in the coils as to be free, and 

 no more. The inner end was fixed to a circular curb 

 wheel, a short portion of the outer end had a pivot, 

 bent in the circle of the outer coil, and supported by a 

 small stud, through a hole of which it moved freely ; 

 at this end was the curb pins, between which the 

 spiral or pendulum spring passed ; the effect of heat or 

 cold on it, was counteracted by the spiral compensation 

 piece. The 'scapement of this was the cylindrical one; 

 and so long as the oil kept clean and fresh, the com- 

 pensation might be useful to it. In 1774, we made 

 one or two of the same sort for horizontal watches. At 

 that time no better 'scapement and compensation were 

 known, at least so far as came under what was then the 

 common practice. 



In the voyage undertaken for the trial of Le Roy's 

 Time-keeper in 1768, and published by Cassini in 

 1770, along with the description and drawings of it, 

 Le Roy has given that of a compensation balance, 

 which is exactly like those of the present day, only 

 the lamina- are pinned together, in place of the brass 

 being melted on the steel. The compensation of the 

 time-keeper, however, was not of this kind; it con- 

 sisted of two glass tube thermometers, bent nearly into 

 the form of a parallelogram, with a small ball at one end 

 to each, the other open, and filled partly with mercury, 

 partly with spirit of wine, fixed to the axis of the ba- 

 lance opposite one another : the balls lay very near to 

 the axis. It would appear that Le Roy had not thought 

 of a metallic compensation, until the return of the 

 time-keeper from the voyage of trial. He had taken 

 the idea of this from getting an account of Harrison's, 

 which was sent to the Royal Academy, signed by 

 Ludlam, who was one of the scientific gentlemen ap- 

 pointed by the Commissioners of the Board of Longi- 

 tude to take Harrison's account of his time-keeper, 

 previous to any part of the reward being paid him. 



In a life of Peter le Roy, the son of Julien, the me- 

 thod of compensation is erroneously ascribed to that 

 artist's father, to whom " we are indebted," says the 

 writer of the article, " for the method of compensating 

 the effects of heat and cold on the balances of chronome- 

 ters by the unequal expansion of different metals; a dis- 

 covery which has been brought by our English artists to 

 a state of great perfection, although it had been laid a- 

 side by the inventor's son Peter." 



It is difficult to understand how this mistake should 

 have been committed ; for the following account of the 



discovery, given by Peter le Roy himself, is in direct Comptma- 

 contradiction to the preceding statement. tion 



" Observation V. Sur / compensation des effcls dc 

 la chaleitr el du froid. L C 



" Selon la gazette du commerce, et la rapport sisnc compensa- 

 Ludelam, envoye a 1'academie : pour rcmedier aux ir- tion by com- 

 regularites produites dans les montres marines par la mon tllur - 



chaleur et par le froid, M. Harrison se serf d'une barre momcter 

 ,,..,. , tubes ana 



composee de deux pieces minces de cuivre el d acier de j,.jj fc 



longeur de deux ponces, rivees ensemble dans pludeurs 

 endroites,Jixccs par un bout, el ai/rint de Faitlrp. deux 

 goupilles att trovers detqaetkt passe le ressnrt du balan- 

 cier. Si cftte barre reste draite dans le terns temprre, 

 (comme le cuivre recoil phis d'imp'v.?sio?i de In chaleur 

 que 1'acier,} le cott- de cuivre deviendra convene an terns 

 chaud, et le cote d'acier le sera an l<ms froid. Ainsi les 

 goupilles Jixent tnur-a-tonr les part/ex du ressnrt, selon 

 les differens degrts de chalcitr, et l'nlottgel on le racour- 

 cissenl : d'ou nail la compensation des ejects du chciud el 

 du froid. 



" Si j'avois connu cette ingeniense methode avant 

 d 'avoir pense a mes thermometres, je n'aurois vraisem- 

 blahlement point hesite a en fake usage dans ma ma- 

 chine. 



" J'ai balance quelque terns, si je ne devois pas lui 

 donner la preference. J'ai mcme fait quelques essais 

 dans cette vue. J'en parleroi bientot : mais apres y 

 avoir pense murement, et avoir mis a part, autant qu'il 

 m'a ete po<sible, ce penchant qui nous parle en faveur 

 de nos productions, mes thermometres m'ont paru pre- 

 ferable," &c. &c. 



See Memoire sur la meilleur manicre, de me/surer le 

 lems en mer, p. 55, .56, inserted at the end of Voyage 

 fait par ordre du Roi in 176'8, pour eprouver Irs monlres 

 marines invenlees par M. le Roy, &c. par M. Cassini, 

 fils, Paris, 1770. 



As must always be the case in the infancy of any 

 branch of science, various methods are fallen on before 

 it arrives at its most improved state. Berthoud, Ar- Berthoud 

 nold, and others, had recourse to different modes of employs 

 compensation before they arrived at the one which gave different 

 them complete satisfaction. The former, in his first *" 

 machines, used small wires of brass and steel, combined ^^"f % 

 nearly like the gridiron pendulum, to effect the pur- he had it 

 pose of compensation ; to those of a later date, was ap- in the ba- 

 plied a straight piece, composed of laminte of brass and lance. 

 steel pinned together, acting on the short arm of a le- 

 ver. In the end of the other arm, which was long, 

 the curb pins were fixed. Even with those balances PLATE 

 which were afterwards made, and composed of brass CCCIV. 

 and steel pinned together, he adopted as a supple- F'g s - 2 > 3- 

 mentary aid, the straight compensation piece with the 

 moveable arm and curb pins. Considering the ta- 

 lents he possessed, and the great experience which he 

 must have had, this seems a little curious; as we think, 

 where there is a complete compensation in the balance, 

 it alone should be sufficient, and that the curb pinj 

 would tend to disturb the pendulum spring rather than 

 give any aid to the compensation. The more free the 

 pendulum spring is, the chance is more in favour of 

 good performance, when the compensation balance is 

 supposed to be fit and complete for what is required. 

 In some of Arnold's balances, two pair of laminae were Arnold 1 ! 

 placed parallel to the diametrical arms ; on the middle compensa- 

 of them was fixed a small wire which came through the tion balai- 

 rim, outside of which, and on the end of the wires, a *?? 

 small ball was fixed to each. These balls were pushed 

 out or drawri in by changes of temperature. The ar- 

 guments given in favour of such as Earnshaw's, which 



