The mean intensity of the beam at the focus was 

 determined to be about 670 candles, while the total 

 intensity was evaluated at about 19,000 candles. 

 The electric light at Dungeness remained in inter- 

 mittent use for a dozen years, but the combination 

 of an inefficient commutator, frequent mechanical 

 breakdowns, and untrained personnel finally led 

 Trinity House to replace it with an oil light in 1874. 



Meanwhile similar efforts made across the Channel 

 in France proved more successful. These attempts were 

 begun by Florise Nollet, professor of physics at the 

 Ecole Militaire in Brussels and a descendant of Abbe 

 Nollet, the famous 18th-century electrical demon- 

 strator. In 1849 Florise Nollet added to his many 

 inventions a version of the Saxton magneto that 

 could be used either to produce hydrogen and 

 oxygen for a Drummond light by the electrolysis of 

 water or to heat a thin carbon rod to incandescence 

 in a vacuum."" He then proceeded to design a 

 multiple disk armature generator (fig. 37) which, like 

 Holmes' generator, was basically similar to that of 

 the Woolrich inachine.*- 



Nollet's magneto had not yet been constructed when 

 he died in 1853, but his specifications then were 

 being considered by a company that called itself 

 the Electric Power Corporation, with headquarters 

 in Genoa. That company obtained the drawings of 

 Nollet's proposed generator and sought to exploit it 

 in a kind of perpetual motion project. The generator 

 would be used to electrolyze water, and the resulting 

 gases would be used in turn to produce more electricity 

 in a Grove gas battery. After inveigling money out 

 of quite a few prominent people, including Napoleon 

 HI, and starting to build six machines according to 

 Nollet's plans, the company was exposed as a fraud. ^^ 

 Holmes was one of those called in to recommend a 

 possible use for the abandoned magnetoelectric 

 machines, and it was at this time he suggested they 

 be utilized for arc lights. 



Nollet's patent was sold towards the end of 1855, 

 and the Societe I'Alliance was formed with Auguste 

 Berlioz as director and with Joseph van Malderen, 

 who had been a coworker of Nollet's, as chief engi- 

 neer. The new company redesigned Nollet's genera- 



tor (fig. 38) and sought to place it in commercial use.** 

 The first attempt was made in 1856 at the illuminating 

 gas plant located at the Hotel des Invalides, where the 

 generator was used to provide hydrogen and oxygen 

 by means of electrolysis.*' This generator was formed 

 of six banks of permanent magnets, with eight magnets 

 radially arranged in each bank, and with the open ends 

 of the magnets pointing towards the axis. There 

 were five disk armatures that rotated between the 

 banks of the permanent magnets (fig. 39). The 16 

 coils on each disk were connected in series, and the 

 disks could be connected to give either high current 

 or high voltage. In spite of the new company's 

 efforts, the generator was not very successful and no 

 further commercial applications were tried for several 

 years. 



An effort was made about this time to obtain some 

 theoretical understanding of the Alliance machines. 

 F. P. Le Roux studied the variation of the current with 

 the external resistance, the variation of voltage with 

 the speed, and the efficiency of an Alliance machine 

 at the Conservatoire des Arts et Metiers in Paris 

 (fig. 40).*" He found that some two-thirds of the 

 energy from the engine driving the generator was 

 lost internally in the generator. In addition to 

 recommending certain values for the resistance of the 

 generator, he also pointed out that much energy was 

 lost through the production of sparks by the com- 

 mutator. At the suggestion of Professor Masson, 

 the commutator was removed from the Alliance 

 generator and the efficiency was found to be much 

 greater. *'' Instead of direct current, the generators 



" British patent 13302 (October 24, 1850); Mechanics Maga- 

 zine, 1851, vol. 54, pp. 358, 362-364, 410-411. 



« French patent 11649 (April 26, 1851; addition, April 24, 

 1852); British patents 14197 (July 6, 1852) and 1587 (July 1, 

 1853). 



*^ L'Electricite, 1881, vol. A, p. 154, 



" Frank Geraldy, "Les Eclairages electriques a Paris, 

 systeme de r.Alliance," La Lumiere Hecirique, 1880, vol, 2, pp. 

 259-262. It seems possible that the Compagnie TAUiance 

 was formed at a later date; Rittershaus mentions 1859 in his 

 article "Zur Geschichte der Dynamo-Maschine" in Der 

 Civilingenieur, 1893, neue Folge, vol. 39, p. 350. British patent 

 2987 (December 2, 1857); French patent 21590, (July 10, 1858; 

 additions, March 14 and December 17, 1859, August 9, 1865, 

 and December 7, 1866). 



^» F. P. Le Roux, "Memoire sur les machines magneto- 

 electriques," Comptes reiidus, 1856, vol. 43, pp. 802-805; Du 

 Moncel, op. cit. (footnote 5), vol. 1, pp. 361-364; V Ann'ee 

 scientifique, 1858, vol. 3, pp. 80-84. 



■•8 F. P. Le Roux, "Etudes sur les machines magneto-elec- 

 triques," Bibliotheque Vniverselle de Geneve, Archives des sciences 

 physiques et nalurelles, 1856, vol. 33, pp. 198-213. 



" Theodose du Moncel, L'Eclairage electrique, Paris, 1 879, ed 

 2, p. 59, E, AUard (in Les Phares: Histoire, com/ruction, 

 eclairage, Paris, 1889) says Van Malderen suppressed the com- 

 mutator. 



356 



BULLETIN 228: CONTRIBUTIONS FROM THE MUSEUM OF HISTORY AND TECHNOLOGY 



