278 E. L. Nichols — Age-coating in Incandescent Lamps. 



This phenomenon, which seems to take place in all incandes- 

 cent lamps of the present day, was first studied exhaustively 

 by Mr. W. H. Pierce, who read a paper entitled " The Relation 

 between the Initial and Average Efficiency of Incandescent 

 Electric Lamps," before the American Institute of Electrical 

 Engineers, in 1889.* Mr. Pierce made a careful study of 

 ninety-four lamps, including nearly all the types of glow-lamp 

 then in use. He found no exception to the rule of progressive 

 degeneration, f 



These changes may be ascribed to at least three causes : loss 

 of vacuum, increase of resistance, due to the disintegration of 

 the filament, and finally, the deposition of the disintegrated 

 carbon upon the inner surface of the lamp-bulb. It was the 

 object of the experiments to be described in this j3aper, to make 

 a study of this coating. We endeavored (1) to determine its- 

 character; that is to say, whether it is colorless, or is selective 

 in the particular rays it absorbs; (2) to learn something about 

 the rate and distribution of deposit; and (3) to determine 

 how far the absorption of light by the coating is accountable 

 for the diminution in the brilliancy of the lamp. 



For the determinations under the first head, we made use of 

 a form of polarizing spectro-photometer which has been more 

 than once described in the pages of this Journal.^ The 

 standards of light were incandescent lamps. These were main- 

 tained at a candle-power considerably below the normal, that 

 the changes taking place in them might be slow. Frequent 

 comparisons of their spectra showed that the relative changes 

 of brightness and quality of light were inappreciable. 



The first step in our investigation consisted in the measure- 

 ment of the absorption spectra of the bulbs of certain unused 

 lamps, which had been selected for study. These lamps were 

 then brought to degrees of incandescence, previously agreed 

 upon, and were maintained in that condition by means of the 

 current from a storage battery carefully kept constant, until 

 they showed coatings . of sufficient density to admit of ready 

 measurement. They were then taken out of circuit and the 

 absorption spectra of the blackened bulbs were determined. 

 These readings gave the amount of light of each wave-length 

 absorbed, and by repetition, the rate of deposition during dif- 

 ferent periods of the lamp's existence. The entire set of pho- 

 tometric observations were duplicated, but since the final results 



* Trans. Am. Inst. E. E., vol. vi, p. 293. 



•j- Since the above was written an extended investigation upon this subject has 

 appeared. It entirely confirms the results obtained by Mr. Pierce. (See " A 

 life and efficiency test of incandescent lamps " by Prof. B. F. Thomas and Messrs. 

 Martin and Hassler), Trans. Am. Inst. E. E., vol. ix, 1892. 



% This Journal, vol. xxxvi, p. 332 ; also Pbil. Mag., V, xxxii, p. 404. 



