HOROLOGY. 



155 



Camattt- indies, should the compensation be found in excess. 

 Aa it would be convenient, when shifting, to have a piece 

 **""* similar to the third lowest traverse, and three spare 

 ' pin*. Thi* piece being like a halt' of the traverse, it 

 fftimtam. may then be applied to the three rods, and pinned, but 

 f' TC not to the place where the shilling is to be made. 

 piece will prevent the rods shifting away from 

 *"* *' one another, and will allow the traverse to be moved 

 and fixed to the intended place. Two thin pieces of 

 brass pq, r i, must be provided, having nine holes in 

 them, so that all the rod* can move easily through 

 them, the two outermost hole* being kept rather a lit- 

 tle tighter than the others. Theae piece* are intend- 

 ed to prevent any tremulous motion in the rods or 

 pendulum, and are put at the same distance from the 

 upper and lower traverse*, a* shewn in the Figure. The 

 pendulum ball i* composed of two frustums of equal 

 cones ; the greater diameter is seven inches, the le-ser 

 four inches, and the height half an inch, giving, by cal- 

 culation, 24 S474 cubic inches, the weight of which in 

 lead is 9.997 Ib. The ball, when filled with lead, to- 

 gether with the shells and inside cross piece, weigh- 

 ed 10 Ib. 8 oa. : The weight of the bras* and steel 

 rod*, traverse pieces, pendulum spring, and top piece, 

 Ac. wa* 5 In. 13 ox. ; in al! os. The clock 



to which this pendulum ws applied was a month 

 one, and wa* kept going by a weight of 7 Ib. 7 oa. 

 We cannot help thinking, that this pendulum is fitted 

 up in a much better style than either Camming'* or Ber- 

 owi*. 



The third lowest traverse being shifted up tare* 

 inches, there will then be this quantity lea* for length* 

 of bra*, and steel than has been elated. The steel will 



l = 109 568 inches, and 

 Ukenat35 3 



length* and expansive ratio* may be 



o ras, an see an as een a 

 be 36+35+37 -628+3.94 3 = 

 the longeet brass rods may be take 

 S0&A9. Their lengths and expa 



Steel 109568 iaenea 



Log. 2.0396M8 



I...-. 



Ratio 1*1 . . 



3.9099135 



Log. 1.8860100 

 Log. 2.08*7854 



T he diffrrmce in exce** at 

 trendy small, and u on tfce eida 



3.906796* 



is here ex- 

 the steel rods, be- 



ing mrrrly an inch too much. In taking into 

 the length* of the reds, those on one side along with 

 the centre rod* are only taken ; those on the other side 

 serve a* kind of covnterpoue. for giving an appear. 

 anee of unifcraritr to the peadoliint, and far ptvwMaw 

 the weight of the pendulum ball from bending the 

 would do were they all on OM ride. A 



,'.H1 -.1 - I ' I .! . ' II 



d.,rl. me adopted , mr rod, :, place of l,r,M : ird. 

 from their mater , .- . ~ , , ..Hi,, 



three of steel and two of fine being ^H%'fiH for the 



but much of then a* we have seen, ap- 

 peared to have the sine rod* of a greater length than 



.1 W_ I I L__ _ * . 



they should have been, according to the comparative ex- 

 pension* of tine and Meet. There was alao a large cavity 

 in the upper part of the ball, for lodging the lower tra- 

 v-ei*e. which took away r 



.-, to adjust the compensation', being at the 

 upper end* of the rod*, rendered it very inconvenient. 



The following dimension* should answer very well Compent*. 

 for a pendulum of this sort. The length from pin to 

 pin in the upper and lower traverses of the two outside 

 steel rods should be 27 inches, and 5 or 6 inches more 

 beyond the lower traverse, to go within the ball, for the of a zinc 

 same purpose that was mentioned in one having the and teel 

 bras* rods; from the pin, in the lower end of the centre "!;";" 

 steel rod, which is in a traverse just above the lower ' 

 one, to the upper end of the pendulum spring, should 

 be 36.75 inches ; from the pin in the lower end of the 

 centre rod to the centre of the ball, 4.75 inches, being 

 in all 68.5 inches of steel. The diameter of tin- 

 rods should be .25 of an inch. In the upper traver.-e is 

 a hole, through which the centre rod must pass freely. 

 The length of the zinc rods, from the pin in the upper 

 r*e to that in the upper of the lower traverses, is 

 25.34 inches, and their diameter .27 of an inch. There 

 may be holes in the lower ends of the centre and zinc 

 rods, for adjusting the compensation, as in the case of 

 the brass rod-. The ends ot the upper and lower tra- 

 verses must have half holes, taking in with the out-idc 

 steel rods; pieces for tremulous motion, and manner of 

 fitting up the pendulum ball, and regulating nut, &c. as 

 ha* been already described. 



The length* and expansive ratios may be put 

 thus: 



Steel 68 5 inches Log. 1.8356906 



Smeaton s ratio for steel 1 33 inches Log. 2. 1 398791 



Zinc 25.34 inches ......... Log. 1.4O386O9 



Do. for sine 373 inches ...... Log. 2.5717088 



3.9755697 



The sine which is used in these pendulums must al- 

 ways be understood to be that which is hammered half 

 an inch per foot, which seems to have even more expan- 

 sion than Smtfrtim ha* stated ; but whether it expands 

 a* much a* Ward make* it, still remains to be deter. 

 lined. 



In taking into account, by this process, the compara- Method of 

 live expansion of the bras* and steel rods, it must be wmpsase. 

 observed that the steel rod* have their expansion down- tion - 

 wards, while the bra** rods expand upwards ; so that 

 when beat expand* the steel rod* downwards, tin- 

 Draw rod* acting in a contrary direction upward*, 

 tend always to keep the centre of oscillation at an 

 equal distance from the point of suspension, aud hence 

 the length of the pendulum i* constantly the same. 

 In like manner, if cold contract* the steel rod, it will 

 al-o contract the bras* rods, to a* ttill to keep the 

 dulum at aa invariable a length at can poMibly be at- 

 tained by any method that human ingenuity can pro- 

 pose. The comparative lengths of the brat* and steel 

 rod* may be computed by any other expansive ratio 

 than thone of Rerthoud'*, which we have used ; such as 

 those of fieaaaten, or of that ingcnioo* artist Troughton. 

 Their diJen-nre* will vary a little from Ik-rthoud's ; 

 bat it will be very trifling. It will require long expe- 

 rience to know which arc the best founded ; as the go- 

 ing of aetronomiral clock* may be affected by other 

 cause* which have no connection with the compensa- 

 tion of the pendulum rod. 



EUicott'r Pendulum. 



Not many year* afte 

 known, Mr EUicott, and 



after Harrison'* pendulum wa* 

 some of the French artist*, 



