156 



HOROLOGY. 



Cm.pcnsa. 



Uon 

 Fcadulumi. 



F.llicot's 

 pendulum. 

 PLATE 

 CCCV1I. 

 Fig. 3. 



Curuming's 

 improve- 

 ment upon 

 it. 



contrived compensation pendulums in different ways, 

 most of them having the ball adjustable by levers, 

 which can never lie equal to those in which the expan- 

 sion and contraction act by contact in the direct line of 

 the pendulum rod. Mr Ellicott was a very ingenious 

 artist of the old school, ns appears from several of his 

 works, and his pendulum evinces great ingenuity in its 

 construction. 



Ellicott's pendulum, shewn in Plate CCCVII. Fig. 3. 

 consisted of two bars, one of which a b was of brass, and 

 the other of iron or steel. Mr Gumming conceived, that 

 where there were two bars only, a flexure and unequal 

 bearing would take place, and therefore an exact com- 

 pensation could not be effected. To remedy this, he 

 constructed the pendulum with two steel bars, and one 

 bar of brass, as shewn in Fig. 4. No. 3. Into the lower 

 end of the brass bar a b, Fig. 4. was let one half of the 

 diameter of a small steel roller r, the other half being let 

 into a moveable brass piece having two short arms 1, 2. 

 These arms, by the levers TOO 2, no 1, moving round 

 o, o as fulcra, make the roller press equally on the lower 

 end of the brass bar. The steel roller marked r, has 

 a fillet raised up on each end, for the purpose of keep- 

 ing this part of the brass bar at an equal distance from 

 the steel bars. The length of the brass bar a b should 

 be 39.25 inches from the upper end which is square, to 

 the lower end which is rounded, its breadth three quar- 

 ters of an inch, and its thickness at least one-eighth 

 of an inch. The steel bars are in length from the upper 

 square ends to the centre of the ball, nearly where 

 the short arms of the levers act on the moveable brass 

 piece, about 39.75 inches ; and the bars are left broader 

 here, so as to be about one inch and a quarter ; and this 

 breadth is prolonged below the end of the brass bar 

 three quarters of an inch or more. The thickness of 

 the steel bars is one-tenth of an inch, and their breadth 

 three quarters. The back steel bar has no opening in it ; 

 but the front bar has one which is represented in the 

 drawing at A, Fig. 4. No. 2. In order to see the action 

 of the levers on both arms of the moveable brass piece, 

 a cavity s s Fig. 3. is left for this purpose in the ball. 

 A piece of glass is inserted in the opening, so as to pre- 

 vent dust from getting into this part of the pendulum. 

 This broad part of the steel bars serves as a kind of 

 frame, on which are screwed two pieces of brass of the 

 same figure as the steel bars, to connect them with that 

 steel part of the lower end of the pendulum rod which 

 goes through the ball, on which is a nut and screw, and 

 a strong double spring m n in Fig. 4. for the purpose 

 of taking off somewhat of the weight of the ball, ac- 

 cording as it bears too much or too little on the short 

 arms of the levers. The levers are shewn at No. 2,3; 

 the screws, and lower end of the pendulum rod in Fig. 

 4 ; the nut N, and double spring SS, in Fig. 3. The use 

 of the nut and screw is to adjust the strength of the 

 spring, as they have nothing to do in the regulation of 

 the pendulum for time to the clock, which is done by 

 an apparatus for this purpose at top connected with 

 the pendulum spring. In the brass pieces which are 

 screwed on to the broad part of the steel bars, the pi- 

 vots of the levers a, a turn. There is a piece of brass 

 put on at top, formed so as to interpose a little way 

 down between the brass bar and the steel ones, keeping 

 them at a proper distance from one another. The sides 

 near to the square ends of the bars and this piece of 

 brass are firmly pinned or screwed together. It is in 

 this piece of brass that the lower end of the pendulum 

 spring is fixed ; the upper end being fixed to a piece, 

 which is moveable up and down in a fixed frame by a 

 nut and screw. The pendulum is lengthened or short- 



ened, according as the pendulum spring is let out or Compensa- 

 brought within that part of the frame through which tion 

 the spring passes. The bars of the pendulum are con- 

 nected by four or five screws, equal spaces being taken Ellicott's" 

 for their'places between the centre of the ball and the pendulum 

 square or upper ends of the bars. The back bar is tap- improved 

 ped to receive the screws, which go through holes made ty Clln >- 

 in the front steel bar, to let them easily pass. On the nuns ' 

 shank or body of the screws, are fitted-hollow cylinders 

 either of brass or steel, and of such a length, that when 

 the screws are put home there shall be no pinching of 

 the front steel bar by the head of the screws. Rectan- 

 gular openings are made in the brass bar, for the hollow 

 cylinders to come through, whose length may be about 

 equal to twice the diameter of the cylinders, and so 

 that the brass bar may easily move on them, by any 

 small motion they may have in contracting or expand- 

 ing. On each of these cylinders is put two loose brass 

 collets or washers, one between each steel bar and the 

 brass bar to keep them free of each other. Their thick- 

 ness should be at least .04 of an inch, so as to allow 

 the air to pass freely between the bars. The small hol- 

 low cylinders through which the screws pass when con- 

 necting the bars, should go easily into the front steel 

 bar, the lower base bearing on the inside of the back 

 steel bar. The upper base should be above the surface 

 of the front bar fully more than the thickness of strong 

 writing paper, so that when the screw is put home the 

 shoulder of it may not pinch the bar. The edges of all 

 the bars should be chamfered off from each side, so as 

 to form an angle in the middle plane of their thickness, 

 for the purpose of giving them a lighter appearance, and- 

 making them less susceptible of the resistance of the air. 

 The diameter of the pendulum ball may be seven inches 

 and a half, and its thickness at the centre about two 

 inches and a quarter. In the edges, and in the line of 

 the diameter horizontally across, are placed two long 

 and stout screws, gg, Fig. 3. whose heads have gra- 

 duated circles on them, and are near the edge of the 

 ball, and an index i, i to each. The inner ends of the 

 screws shewn at *, s, are turned of such a shape, so 

 as to apply by one point only on the long arms of 

 the levers m, n, as seen in Fig. 3. The front shell 

 of the pendulum ball is fixed on by four screws. It 

 has been objected by some, that, from the weight of 

 the ball, the brass rods in a compensation pendulum are 

 compressed, and the steel rods stretched ; a matter of 

 no moment whatever in our opinion. This may, how- 

 ever, in some degree, be remedied in Ellicott's pendu- 

 lum, by making the brass bar of such a length, as to 

 come through and below the lower edge of the ball, in 

 place of the steel part, which was common to both steel 

 bars, as has just now been described. This part of the 

 brass bar is tapped, having on it the nut and strong 

 double spring, which takes off a part of the weight, as 

 has been noticed. A certain portion of the weight of 

 the ball will in this case bear on the brass bar ; suppo- 

 sing it one-third of the weight, the remaining two- thirds 

 will be carried jointly by the steel bars. Although the 

 brass bar is here carried through the ball, it is easy to 

 put a piece to it, moveable on a pivot in the middle of 

 the bar, having two lugs applying to the edges of the 

 bar, on which the short arms of the levers a, a may act, 

 as was the case in the other by the moveable brass piece. 



Things being in this situation, let us suppose that the Method of 

 bars composing the pendulum rod are lengthened by compensa- 

 heat, and that the brass lengthens more than the same t'n. 

 length of steel does ; then the brass bar a b, by its excess * 'S- 3- 

 of expansion, will press down the short ends of the le- 

 vers, m,n at b, and consequently raise up the ball, which, 





