GEAIX-DUST EXPLOSIOXS. 13 



SERIES 2. 



Ohjcct. — Same as tliat of Series 1. 



Conditions. — ^Similar to those of Series 1, except that one of the 

 disks was blocked so as to permit foreign substances, such as nails, to 

 be in longer contact ^rith the revolving disk and thus to give off more 

 sparks. Definite amounts, usuallr 1 pint, of feed were measured out 

 for the purpose of determining the hniits of explosibility of various 

 mixtures. 



Rfsults. — 



1. In 21 attempts witli the electric arc and various grains and 

 mixtures, 16 explosions were obtained. 



2. A mixture of 1 part elevator dust to about S parts of oat hidls 

 appeared to be the limit of explosibility of these grain products. 



3. THien there was no opening the explosion was propagated to the 

 outside end of the screw conveyor, as indicated by puffs of smoke 

 (E in PL III, fig. 1, and in fig. 4). 



4. Blocking one disk appeared to give results not cUfferent from 

 tliose obtained when both disks were running. 



5. Other residts corroborated those of Series 1. 



SERIES 3. 



Ohject. — The primary object was to determine how far the flame 

 was propagated from the source of the explosion. 



Conditions. — The conditions of the test were similar to those of 

 Series 2 (one disk being blocked), but the humidity was very high, 

 inasmuch as there was a steady rain during a large part of the test. 

 A 4 by 4 mch hole was cut in the upper part of the horizontal screw 

 conveyor, about 1 foot from the chut-e leading down from the attri- 

 tion mill and about 5 feet 4 inches from the shaft of the mill. The 

 various kinds and mixtures of feed were prepared and measured 

 into small paper bags, the contents of one of which were dumped 

 into the feeding hopper at each attempt. 



Results.— 



1. Out of 39 attempts with the electric arc, 36 explosions occurred, 

 in spite of the very damp atmosphere. 



2. Explosions were obtained with elevator dust and oat huUs. and 

 elevator dust, flour, graham flour, wheat scourings, floor sweepings, 

 com, and various mixtures of these substances. 



3. The following classification of the 36 explosions may be made 

 (PL III, fig. 1, and text fig. 2) : 



27 flashes were observed at B (5 feet 4 inches from shaft of miU). 

 9 flames were observed at E (12 feet 6 inches from shaft of miU). 

 3 flames were observed at F (17 feet from shaft of mifl). 



4. 'Duruig the course of the tests 'the conveyor continually brought 

 out smoldering lumps of ground products, which could be readily 

 fanned into a red glow. 



