646 



NATURE 



[Oct. 31, 1889 



It is simpler to attach to any existing clock, but not so 

 delicate as his, and is open to the same objections. It is, 

 however, capable of very good work, as may be judged 

 from the chronograph sheet of the Dunsink Observatory 

 chronograph. 



The third is the form of control which I devised for Mr. 

 Isaac Roberts, and which has been so successful with him, 

 and with Prof. Pritchard (who has had it recently attached 

 to the Oxford equatorial), that photographs have been ex- 

 posed with the telescope to which it has been attached for 

 fifteen minutes, and yielded perfect images of stars without 

 any hand and eye guiding. 



The arrangement consists, firstly, of a remontoire train, 

 driving a good mercurial or other compensated pendulum 

 — the driving of this train being of course entirely inde- 

 pendent of the equatorial clock giving motion to the 

 telescope ; secondly, of a detector apparatus, which detects 

 any difference between the rate of this standard pendulum 

 and the equatorial clock ; and thirdly, of a correcting ap- 

 paratus, which corrects automatically any error discovered 

 by the detector. This corrector itself consists of two 

 parts— an "accelerator" and a "retarder" — and these we 

 will first proceed to describe. 



In s S' s" is one of the shafts, between the driving train 

 of the equatorial clock and the worm which drives the 

 right ascension sector, this shaft being cut into three parts, 

 denoted by the letters just named. At one end the portion 

 S of the shaft carries a wheel, i, immediately adjoining 

 which is the wheel 2, mounted on the portion S' of the 

 shaft. At the other end of this last-named section of the 

 shaft is fixed a third wheel, 3, which is almost in contact 

 with the wheel 4, fixed on the end of the shaft s". The 

 shafts S and S' also have mounted freely on them the brass 

 disks, d d\ which adjoin the two pair of wheels referred 

 to above. Each of these brass disks is furnished with a 

 stud on which a small pinion is mounted, the pinion p, 

 belonging to the disk d, gearing across the pair of wheels, 

 1-2 ; while the pinion p, belonging to the disk d', gears 

 across the pair of wheels, 3-4. 



Under normal conditions, if no error exists in the 

 equatorial clock rate, the arrangement of wheels and 

 pinions just described revolves as one piece, the three 

 sections, s 3' s", of the shaft rotating at the same speed ; 

 but it is possible by an arrangement which we shall explain 

 presently, to stop the rotation of either of the disks, d d', 

 and as soon as this occurs the pinion of the stopped disk 

 has to act as a transmitter of motion from one of the 

 wheels into which it gears to the other. If the two wheels 

 of each pair had the same number of teeth, the speed of 

 both wheels would still remain the same, but in reality the 

 number of teeth in the two wheels of each pair is different, 

 and hence the stopping of one of the disks, d or d', causes 

 a variation in the rate of rotation of the two adjoining 

 wheels relatively to each other. For instance, in the case 

 of the first pair of wheels, let wheel i have 30 and wheel 

 2 have 29 teeth, and suppose that the shaft s is rotating 

 once every 60 seconds. Thus, if the disk d be stopped, 

 the wheel 2 will be made to revolve in f% of the time oc- 

 cupied by the wheel i, or in other words the rate of the 

 section s' of the shaft will be accelerated to one revolution 

 in 58 seconds. In the same way by reversing the positions 

 of the wheels in the other pair 3-4, the stoppage of the 

 disk d' can be made to effect a retardation of the portion 

 S " of the shaft relatively to s'. The edges of the disks 

 d and d' are cut into very fine teeth, and the stoppage of 

 the disks when desired is effected by causing a comb 

 attached to the armature of an electro-magnet to engage 

 with these teeth. 



The whole apparatus just described constitutes a very 

 convenient arrangement for accelerating or retarding the 

 driving motion imparted to the telescope by the equatorial 

 clock, and that it is capable of very good work is shown 

 by the photographs which have been taken by Prof. 

 Pritchard and Mr. Roberts, in which the star disks are per- 



fectly round, though exposed for 1 5 to 60 minutes, and no 

 hand guiding used. 



I have now to describe how this apparatus is, when 

 necessary, automatically brought into action by the 

 " detector." 



In Figs. 5, 6, and 7,^ w is a scape-wheel mounted on 

 the sixty-second spindle of the controlling clock, and 

 driven from that spindle through a spiral spring, x x, so 

 that no error in the equatorial clock can affect its rate or 

 that of the standard pendulum. On the same spindle 

 there is also mounted behind the scape-wheel an ebonite 

 disk, E E, Fig. 5 ; this disk, which is driven by the 

 equatorial clock, carrying two insulated rings, b b', which 

 are respectively connected metallically with two platinum 

 plates, B p.', inserted in the face of the disk. Between the 

 scape-wheel and the ebonite disk there is also mounted 

 loose on the spindle a lever, A A, which carries at one of 

 its ends a platinum bridge, B, which is of such a length as 

 to fit between the platinum plates, B B, and which in its 

 mid-position bears against a piece of rock crystal let into 

 the ebonite disk between the two plates just named. At 

 the other end the lever, A A, is formed into a fork, between 

 the arms of which projects a pin carried by the scape- 

 wheel ; the arms of the fork are provided with set screws. 



Vj^^JI-^-^^ 



Fig. 5. 



by means of which the amount of play allowed to this pin 

 in the fork can be adjusted. 



The insulating rings, b b', are electrically connected with 

 the accelerator and retarder already described by means 

 of fine platinum wires, o o', wiping against them, and the 

 action of the whole arrangement is as follows. The 

 scape-wheel, w, being driven by the control clock, has an 

 intermittent movement corresponding to the beats of the 

 pendulum, while the ebonite disk, E E, being driven by 

 the equatorial clock, has a constant movement, so that even 

 if the scape-wheel and disk make a whole revolution in the 

 same time, the pin carried by the scape-wheel will be con- 

 stantly oscillating between the pins of the fork at one end of 

 the lever. A, this lever being driven by friction from the 

 ebonite disk. The pins just named are adjusted so as to 

 allow of this oscillation taking place without interference, 

 so long as the rates of the equatorial and control clocks 

 remain uniform, but if the equatorial clock either loses or 

 gains with respect to the standard, the pin on the scape- 

 wheel comes into contact with one of the fork pins of the 

 lever, A, and shifts that lever on the spindle, bringing the 

 bridge, b, into contact with one of the platinum plates, 



* These blocks have been kindly lent by the editor of Engineering. 



