ELECTRICITY. 



527 



Practical 

 Klectricity. 



double. 



be reduced to the genei^l principles, but sufficiently cu- 

 nous to excite the meditations of persons the most ex- 

 perienced in this branch of natural philosophy. 



If the ball be connected with the lower part of Ben- 

 net's electrometer, and the plate A with the upper part, 

 and any weak electricity be communicated to the elec- 

 trometer, while the position of the apparatus is such 

 that the cross piece GH touches the two pins, a very- 

 few turns will render it perceptible. But here, as well 

 as in the common doubler, the effect is rendered un- 

 certain, by the condition that the communicated electri- 

 t city must be strong enough to destroy and predomi- 

 nate over any other electricity the plates may possess. 

 I scarcely need observe, that, if this difficulty should be 

 hereafter removed, the instrument will have great ad- 

 vantages as a multiplier of electricity in the facility of 

 its use, the rery speedy manner of its operation, and 

 the unequivocal nature of its results." See Phil. Trans. 

 1788, p. 403. 



SECT. IV. Bennet's Improvement upon hit Doubler. 



Beanet" WE have already seen, that Mr Bennet was aware, 



improve- that electricity was produced in his doubler even when 

 ' none had been previously communicated to it. In or- 

 r- der to remedy this, he 'connected the plates A and C 

 PLATE by a wire, having at its ends hooks, which went into 

 rci.. two small knobs on the back of the plates, the pillar 



figt. 1, 2. which supports the doubler being in contact with the 

 middle of the same wire. Another wire connected the 

 back of the plate- B with the brass ball D. In tliis man- 

 ner all the plates communicated with the ground, and 

 by turning the winch, all its electricity could be dis- 

 charged in every part of its revolution. Mr Bennet 

 observed, that the spontaneous discharge of the doubler 

 was always negative. With a phial positively charged 

 he touched A and C, and turned the doubler till it pro- 

 duced sparks for a long time together. He now hook- 

 ed on the wires as formerly described, and made the 

 plate B revolve about 100 times, which so completely 

 deprived the doubler of its positive electricity, that, 

 when the wires were taken off, it produced a negative 

 charge at about the same number of revolutions which 

 it formerly required. The phial charged positively was 

 again applied, and the wires being again hooked upon 

 the plates, B was made to revolve 50 times, which was 

 found sufficient to deprive it of its positive charge. In 

 many case*, five or six revolution* answered this pur- 

 pose; but Mr Bennet generally allowed 4-0 or 50 turns 

 t'the handle. 



SECT. V. Catallo's Improvement upon Sennet's Douller. 



Cavallo* MR CAVALLO having found, that, after doubling 20 

 improve- or 30 times, the doubler became strongly electrified al- 

 ment upon though no electricity had been communicated to it, ima- 

 g' ne <l tnat tms electricity arose from some friction of 

 the varnished plates. In order to avoid this source of 

 . electricity, lie constructed three plates without the least 

 varnish, and which stood within the eighth of an inch 

 of each other. Each plate stood vertically, and was sup- 

 ported by two glass sticks, which were coated with 

 sealing-wax, and fixed on a wooden pedestal 7 ! inches 

 long, 2| broad, and 1J- thick. The vertical plate is 

 about 8 inches in diameter, and is made of strong tin. 

 The stand is so narrow, that it projects only a little be- 

 yond the face of the tin plate, so as just to prevent the 

 two faces from coming in contact when they are pla- 

 ced facing one another upon a Uible. The method of 

 doubling with these plates is exactly the some as with 



Bennet's doubler, only instead of placing the one upon Practical 

 the other, they are placed facing each other ; and in ^ ^ 

 performing the operations, they are held by the wooden "*V 

 stand, so that no friction can take place! Even with 

 these precautions, however, Mr Cavallo candidly ac- 

 knowledges that the plates still produced electricity 

 from themselves, and that the instrument could not be 

 depended upon. See Phil. Trans. 1788, vol. Ixxviii. 

 page 1. 



SECT. VI. Dr Robison's Improvement upon Bcnnet's 

 Doubler. 



DR ROBISON was of opinion, that the instrument Dr Robi- 

 as originally constructed by Mr Bennet, might be freed son's im- 

 from the error of producing electricity without commu- provement 

 nications, by employing a plate of air as the intermedium "*J n rf ! 

 between the tliree plates of the doubler. To effect this, i cr> 

 he proposes'to stick on one of the plates tliree small 

 spherules, made from a capillary tube of glass, or from 

 a fibre of melted sealing-wax. The other plate being 

 made to rest on these three points, will be separated 

 from the other by a plate of air, and will scarcely re- 

 ceive any friction tliat will affect the results of the ope- 

 ration. Df Robison was also of opinion, that if the fine 

 wires, which form the connections in Nicholson's dou- 

 bler, were tipped with little balls, the dissipation would 

 be greatly prevented, and the instrument much impro- 

 ved. He likewise thinks, that an alternate motion like 

 that of a pump handle might be advantageously used, 

 as the plates would thus be permitted to approach each 

 other face to face, and admit a greater multiplication, if 

 it were thought necessary. See Robison's System of 

 Mechanical Philosophy, vol. iv. now in the press. 



For an account of other dou biers, see Journal de Phy* 

 sique, vol. xlv. p. 463. Hachette and Desorme's Dou- 

 bler, Bulletin* Soc. Philomatique, No. 83. 



CHAP. III. 



On Instruments for Multiplying Electricity. 

 SECT. I. Description of Cavallo's Multiplier. 



Tins ingenious instrument, which is represented in Description 

 Plate CCL. Fig. 3, was invented by Mr Cavallo before ol l'*Uo<s 

 the year 1 795, and appears to give less equivocal results mulu l >ller> 

 than the doubler, when made in its best form. Two PI-ATE 

 glass rods G, H, are firmly fixed in the wooden stand CL - 

 RSQ, and carry two flat brass plates A, C, of a square * 

 form. A similar brass plate B is supported by another 

 glass rod I, fixed by cement in the wooden lever KL, 

 which has a motion of rotation round a pin K, screw- 

 ed into the base RSQ. By means of this lever the 

 plate B can be moved from its present position into the 

 position B'j the lever KL having then the dotted posi- 

 tion KX. A thick brass wire N is fixed at the corner 

 Q.and there proceeds a crooked wire mO from the brass 

 socket O, on the back of the plate B. A fourth brass 

 plate D, similar to the other three, is supported upon 

 a brass wire screwed firmly at P into a piece of brass 

 FP, which slides in a groove made in the wooden base, 

 so that by pulling out this slider, the plate D can be 

 removed to any distance from the plate C. When FP 

 is pushed as far in as it will go, the distance between 

 the two surfaces of the plates D and C is every where 

 ,' 5 th of an inch. When the lever LK is pushed as far 

 round towards Q as possible, the surfaces of the plates 

 A, Bare likewise ^th of an inch distant from each other, 

 and the bent extremity of the wire Om just touches die ' 



