﻿THE 

 LONDON, EDINBURGH, and DUBLIN 



PHILOSOPHICAL MAGAZINE 



AND 



JOURNAL OF SCIENCE. 



[FOURTH SERIES.] 



JANUARY 1871, 



I. On the Examination of the Bessemer Flame with Coloured 

 Glasses and with the Spectroscope. By J. M. Silliman, M.E., 

 Adj. Prof, of Metallurgy, Lafayette College 3 Easton } P«.* 



I. Examination with Coloured Glasses. 



IN the Bessemer process the progress of the decarbonization is 

 determined chiefly by the appearance of the smoke, flame, 

 and sparks which are emitted from the apparatus. Owing to 

 the rapidity with which the reactions take place, it is highly 

 important to catch the exact moment when the blast should be 

 turned off. This is indicated by the colour and brightness of 

 the stream of gas issuing from the converter; and by this the 

 moment of total decarbonization can generally be accurately de- 

 termined by the naked eye. When, however, pig iron of certain 

 qualities is used (manganiferous iron, for example) this determina- 

 tion is very difficult ; even those who have had much experience 

 make frequent mistakes, and find it impossible to produce the 

 same quality of steel at every blow. 



In order to intensify these flame-indications, use has been 

 made of the spectroscope, and also of various combinations of 

 coloured glasses. The former was first attempted by Dr. Roscoe, 

 and the latter by Mr. Rowan at the Atlas Works. 



Mr. Rowan experimented with a great variety of coloured 

 glasses, and obtained the best results by using three glasses — 

 two of ultramarine-blue and one of dark yellow. This little in- 

 strument (or chromopyrometer, as he terms it) is now in daily 

 use at the Atlas Works, its indications being so marked and un- 

 * From Silliman's American Journal for November 1870. 



Phil. Mag. S. 4. Vol. 41. No. 270. Jan. 1871. B 



