^*^'J^^'\ /.c;^-% /.-^i-X <? >'^Mk °- 
 
 
 ^"-^^' 
 
 
 
 
 
 ^^ ^ ' " ^ ^^. 
 
 '^l 
 
 

 
 4 O 
 
 
 \/ 
 
 
 
 
 
 
 
 <^ aP 'w^- ^ "^ 
 
 
 
 
 
^. 
 
 (: 
 
 -'. 3 4 '^ 
 
 ;^¥ 
 
 <yf GRAIN DUSTS 
 
 • estimation t.y 
 
 Mill«rfl' Committee of Buffalo^ N. V 
 Under tKe Direc- 
 Dr. Ge orge A . Hulett , ijh itf (Jhttm ist 
 US. Bureau al Min- 
 
 A PRELIMINARY REPORT 
 
 David J. Price 
 HaroH H. Browr 
 
 PITTSBURGH. PA. 
 
EXPLOSIBILITY 
 of GRAIN DUSTS 
 
 Cooperative Investigation by 
 
 Millers' Committee of Buffalo, N. Y. 
 
 Unaer tne Direction or 
 
 Dr. George A. riulett, Cnief Cnemist 
 
 U. S. Bureau or Mines 
 
 A PRELIMINARY REPORT 
 
 by 
 David J. Price 
 
 ana 
 Harola H. Brown 
 
 PITTSBURGH, PA. 
 JULY, 1914 
 
rs 
 
 THE MILLERS' COMMITTEE OF BUFFALO, N. Y. 
 
 LAWRENCE E. HARMON, CUirman, 
 
 President, Burralo Cereal Company 
 
 FRANK F. HENRY, 
 
 Manager, Wasnourn-CrosDy Company 
 
 GEORGE F. URBAN, 
 
 Secretary, George Urban Milling Company 
 
 ^<_ M- ' 
 
The Exploslbility of Grain Dusts 
 
 A PRELIMINARY REPORT 
 By David J. Price and Harold H. Brown 
 
 GENERAL INTRODUCTORY STATEMENT 
 
 UCH attention has been given to the study of the 
 inflammability of various forms of carbonaceous dusts 
 in this country and abroad, in recent years. Prelim- 
 inary conclusions indicate that under certain favorable 
 conditions nearly all forms of carbonaceous dusts are explosive. 
 Many disastrous explosions have occurred in bituminous coal 
 mines throughout the United States and European countries, 
 where firedamp or "methane gas" had not been ordinarily 
 detected. The cause of these explosions has been traced to the 
 ignition of coal dust that propagated to a violent explosion 
 throughout the entire mine, causing great loss of life and consid- 
 erable property damage. The large loss of life that usually 
 accompanies a mine disaster of this nature brings coal-dust 
 explosions forcibly to our attention at the time of the occurrence. 
 We must not conclude, however, that coal dust is the only kind of 
 dust that will explode, as experiments have clearly shown that the 
 dusts of other substances containing carbon are explosive. In 
 these various forms of carbonaceous dusts are included the dusts 
 produced in most of the industries in which grain or its products 
 are used. The object of this preliminary report is to briefly sum- 
 marize the various conditions under which explosions of this 
 nature have occurred, and to give a series of results already 
 obtained relative to the exploslbility of these dusts. 
 
 ORIGIN OF THE STUDY 
 
 I.- 
 Although many men prominently identified with the milling 
 
 industry accept the theory that carbonaceous dusts are explosive, 
 
 others feel that their plants are not in danger except when only 
 
The Explosibility of Grain Dusts 
 
 certain kinds of dusts are produced. Explosions in several flour 
 mills in 1878, clearly proved that flour dust under certain condi- 
 tions is highly explosive, and immediately caused flour millers to 
 take precautions to guard against a repetition of an occurrence 
 of this nature. 
 
 General interest was again awakened among milling men, 
 however, when an explosion occurred in a feed-grinding plant at 
 Buffalo, N. Y., on June 24, 1913, by which 33 men lost their lives, 
 and upwards of 70 were injured. To assist in the work being done 
 upon the explosibility of coal dust, the Bureau of Mines, for some 
 time has endeavored to study dust explosions in surface plants, 
 and therefore made inquiry into this disaster. The Bureau in 
 this way came into touch with the officials of the various milling 
 companies in that section. The cereal, flour and feed millers of 
 Buffalo and Western New York were desirous of obtaining any 
 possible information relative to the explosibility of grain dusts, 
 that would assist them in taking effective safety precautions, in 
 addition to the measures that had already been adopted. 
 
 A co-operative movement between milling interests gener- 
 ally and the Bureau of Mines, was arranged for the purpose of 
 making a scientific study of the explosibility of grain dusts, and 
 of methods pertaining to the prevention of such explosions. The 
 Millers were represented in the conduct of the work by Messrs. 
 Lawrence E. Harmon, President of Buffalo Cereal Company; 
 Frank F. Henry, Manager, Washburn Crosby Company, and 
 George F. Urban, Secretary of George Urban Milling Company, all 
 of Buffalo, N. Y. Prof. George A. Hulett, Chief Chemist of the 
 Bureau of Mines, was placed in charge of the study, and David J. 
 Price was assigned to the field engineering work in connection 
 with the problem. The work was started August 1, 1913, and a 
 complete series of dust samples was collected, covering the vari- 
 ous lines of milling, in order to conduct the necessary tests to 
 determine the degree of inflammability. The aim was to trace 
 the grain from the beginning of the handling, either from boats 
 or cars, through the various points in the elevators, and the 
 different stages of production, thereby securing a representative 
 series of dust samples throughout the entire plant. These were 
 carefully collected, and the conditions at the points of collection 
 were fully noted, relative to nature of dust, manner of production, 
 quantity present, and similar essential details. On February 1, 
 
Extent of Grain Industry 
 
 1914, Dr. H. H, Brown was assigned to the laboratory study of 
 these dusts, relative to their chemical nature and degree of 
 inflammability. 
 
 EXTENT OF GRAIN INDUSTRY 
 Flour and Feed Mills 
 
 ' T WILL be of interest to note the extent of the grain 
 industry, and the number of milling plants included 
 in a study of this nature. The census report for 1909, 
 relating to the milling industry in the United States, 
 incluaecl all mills grinding wheat, rye and buckwheat, corn for 
 hominy, grits or feed, but not mills doing exclusively a custom 
 business, or factories making fancy cereal food or other special 
 food preparations as a chief product. It will be of interest to note, 
 however, that the figures make milling the fifth in rank in the 
 United States. The report shows that in 1909 there were 11,691 
 establishments in this country, with a total number of 66,054 
 persons engaged.* 
 
 The following table (from Vol. X, Thirteenth Census, U. S., 
 pp. 407 and 420) shows some very interesting statistics regarding 
 the industry: 
 
 . 1909 ^ 
 
 Number of establishments 11,691 
 
 Persons engaged in the industry 66,054 
 
 Primary horsepower 853,584 
 
 Capital invested $349,151,779 
 
 Expenses 827,522,349 
 
 Salaries and wages 33,981,153 
 
 Cost of materials used 767,576,479 
 
 Miscellaneous expenses 25,964,717 
 
 Value of products 883,584,405 
 
 Value added by manufacture 116.007,926 
 
 Grain ground, bushels 800,247,961 
 
 In the total number of establishments as given above, there 
 were 5,621 mills that used over 491,000,000 bushels of wheat, 
 producing more than 105,000,000 barrels of flour, with a value 
 exceeding $546,000,000. 
 
 •From the Millers' Almanack, for 1914-1915, page 83, published by The 
 Northwestern Miller, Minneapolis, Minn. 
 
The Explosibility of Grain Dusts 
 
 There were 2,529 mills manufacturing chiefly feed for live 
 stock, and grinding the following quantity of grain : 
 
 Corn 42,442,263 bushels 
 
 Oats 14,414,992 bushels 
 
 Other grain 13,376,825 bushels 
 
 These mills produced 1,715,899 tons of feed, valued at 
 $48,242,782. The value of all other products was $2,551,259. 
 
 Glucose and Starch Industry'-' 
 
 r— 1909 N 
 
 Number of establishments 118 
 
 Persons engaged in the industry 5,827 
 
 Primary horsepower 28,257 
 
 Capital invested $38,866,419 
 
 Expenses 43,973,558 
 
 Salaries and wages 4,079,722 
 
 Cost of materials used 36,898,771 
 
 Miscellaneous expenses 2,995,065 
 
 Value of products 48,799,311 
 
 Value added by manufacture 11,900,540 
 
 The above statistics will be of interest in classifying the 
 various dust explosions according to the industries in which they 
 have occurred. Reference to the tabulated form that follows, 
 showing nature and location of explosions since 1905, makes a 
 comparison possible. 
 
 INVESTIGATIONS OF RECENT EXPLOSIONS 
 
 iNE OF the early steps taken at the beginning of the 
 study was an effort to secure data relative to occur- 
 rences of this nature in the United States and 
 European countries, in order, if possible, to assist in 
 establishing the various sources of origin, and conditions under 
 which these explosions occur. Early investigations indicated 
 that a large number of explosions have occurred in grain and 
 cereal mills throughout this country and abroad, with varying 
 degrees of violence and damage. Owing to the recognized 
 advances in milling equipment and construction in recent years, 
 it was thought advisable to devote the early part of the study to 
 the explosions that have taken place in modern plants, equipped 
 
 *Prom Vol. X, Thirteenth Census United States, page 429. 
 
Investigations of Recent Explosions 
 
 with improved milling devices and lighting systems. For this 
 reason the reported occurrences of this nature since 1905 were 
 especially investigated and carefully studied. It was thought 
 that explosions occurring since this date would include plants 
 that had taken advantage of improved milling equipment, and 
 where n£cessary provisions were being taken for dust collecting 
 and handling. 
 
 Frequency of Explosions 
 
 Explosions in grain mills were rather frequent, both in this 
 country and abroad, prior to the early '80s, as a result of which 
 it was discovered that the methods of dust collecting were not 
 efficient, and new systems accordingly were introduced. At the 
 time of these occurrences the old "stive" or "dust room'' was in 
 general use, and the dust produced during the milling process 
 was collected in this chamber. This provision made it possible 
 at all times for a dangerous amount of dust to be in suspension, 
 that would produce a violent explosion upon ignition. 
 
 The old "stive" or dust room has been replaced by dust- 
 collecting systems of various types and construction. In the 
 present modern systems of collection the dust is carried through 
 pipe lines from the different points of production by means of air 
 currents produced by an operating fan, to the dust collector, 
 where the separation takes place. 
 
 The danger due to the old "stive" or dust room was well 
 proven by explosions in Minnesota in 1878, and in Illinois in 1893. 
 In both cases very violent explosions occurred in flour mills where 
 the dust was carried to a chamber of this nature. The former 
 explosion caused the loss of 18 lives and extensive damage to 
 property; the other explosion occurred at night during a fire. 
 The mill was completely destroyed and considerable damage was 
 done to surrounding property. 
 
 Most milling plants throughout the country have taken 
 advantage of the improved systems of dust collection, and the 
 old "stive" room is obsolete in modern mills. However, it would 
 appear that in some of the European countries advances along 
 this line were not so rapid as in this country. An explosion of 
 considerable magnitude is reported as occurring in Liverpool as 
 late as the spring of 1905, in a flour mill where a dust chamber 
 or "stive" room was in use. It is thought the explosion, which 
 was followed by a disastrous fire, originated in this dust chamber. 
 
8 The Explosibility of Grain Dusts 
 
 It has been observed that the majority of explosions in 
 more recent years have occurred in mills handling or grinding 
 coarse grain. By the term "coarse grain" is meant grain such 
 as oats, corn, etc., other than wheat; and mills grinding grain 
 of this nature are termed "coarse grain mills," to distinguish 
 them from flour mills. A number of explosions in plants of this 
 nature have taken place within the last ten years, causing in some 
 cases a large loss of life and complete destruction of property. 
 
 There is given in the following table a list of grain-dust 
 explosions that have occurred in the United States since 1905, 
 which thus far have been investigated: 
 
Investigations of Recent Explosions 
 
10 The Explosibility of Grain Dusts 
 
 EXPLOSIONS IN THE UNITED STATES 
 Explosion and Fire in Iowa 
 
 CEREAL mill in Iowa was extensively damaged by fire 
 
 A\ on March 5, 1905. The fire is thought to have followed 
 ft- an explosion, reported to have originated underneath 
 a grinding machine. Two employes lost their lives, 
 and the property was completely destroyed. The fire seems to 
 have started from an attrition mill which was grinding oat hulls. 
 It was thought that the mill produced sufficient sparks to cause an 
 ignition of the dust cloud within the grinding machine. One of 
 the mill attendants heard a hard substance strike the plates 
 about the time of the occurrence. An explosion took place in the 
 conveyor line under this machine, and traveled along the conveyor 
 to an adjoining elevator leg, and thence into a storage bin. The 
 elevator leg was blown out, and a very disastrous fire followed. 
 The estimated property loss is one of the largest on record. 
 
 Explosion in Vermont, 1908 
 
 On October 7, 1908, an explosion occurred in an elevator in 
 Northern Vermont. Fifteen employes were killed, and two women 
 who were walking on the railroad tracks outside of the plant were 
 so badly burned that they died soon afterward. Fire followed 
 the explosion, and the property was very badly damaged. 
 
 This elevator was used by its operators as a feed mixing 
 plant; that is, the feed was ground, or prepared for mixing at 
 another point, and sent in cars to the elevator, where it was 
 unloaded and mixed. We can, therefore, class this as an "elevator 
 explosion," and at the same time we are forced to conclude that 
 if an explosion with such terrible force and violence could occur 
 at this elevator, that the grain elevators of the country are at all 
 times subject to the possibility of a similar occurrence if favor- 
 able conditions exist. Many milling men have maintained a false 
 sense of security in recent years, and have felt that the elevators 
 were not especially a source of danger; and that as long as an 
 explosion did not originate in the grinding mill proper, they were 
 reasonably safe. This would also seem to indicate the belief that 
 an explosion could not occur in an elevator at a plant where no 
 milling or grinding was being done. This explosion alone would 
 seem to indicate sufficiently that the dusts created during the 
 
Explosions in the United States 11 
 
 elevating and storing of grain are explosive and dangerous. Other 
 explosions in elevators prior to 1908 have occurred from time to 
 time, and are matters of record. 
 
 Explosion in Illinois, 1910 
 
 An explosion occurred in the elevator building of a glucose 
 factory in Illinois, on August 7, 1910, as a result of which two men 
 were killed and seven injured. The material that exploded was a 
 light pulverized powder called "gluten feed," which is manufac- 
 tured from corn by the removal of starch, oil, etc., and was sold 
 as cattle feed. The corn was first soaked in water for a period of 
 about 48 hours, and then delivered to a grinding mill, in which 
 it was cracked, causing the removal of the germ, which floated off 
 in a body of starchy water. The corn was then still further 
 ground in stone mills and washed in several reels and shakers in 
 order to separate out the starch for the manufacture of glucose. 
 The material that remained was gluten, which was delivered wet 
 to the filter presses, which reduced the moisture content to about 
 50 per cent. It was then dried in a rotary steam drier to about 
 12 per cent, moisture, after which it was further ground in an 
 attrition mill. The gluten feed was taken from the attrition mill 
 to a separate building, called the feed elevator, by means of an 
 air draft operating through a 16-inch galvanized iron spout about 
 375 feet long, and carrying the gluten 140 feet in elevation. The 
 air draft was maintained by a 70-inch planing mill exhaust fan, 
 driven 1,500 revolutions by a belt connected, 35-horse power 
 electric motor. This fan was located about ten feet from the 
 attrition mill, and delivered about 25 pounds of air for each 
 pound of gluten carried. The spout delivered its burden to a 
 "cyclone collector" on top of the elevator building. 
 
 The evidence at hand after the explosion led to the belief that 
 the cause of the explosion was a stream of sparks emanating 
 from the plates of the attrition mill, which was traveling at 1,500 
 revolutions per minute. These sparks were thought to have been 
 caused either by a momentary clogging of the feed of the mill, 
 which would allow the grinding plates to rub against each other, 
 or by a piece of foreign material striking against the plates. A 
 smouldering fire having originated between the mill plates, the 
 strong air draft quickly fanned it into a blaze. The amount and 
 coarseness of the gluten probably prevented an explosion in the 
 delivery pipe, but the blaze and the pressure of the resulting 
 
12 The Explosibility of Grain Dusts 
 
 gases increased enormously. When this fire and force reached 
 the elevator building enough dust was jarred into the air to make 
 favorable conditions for the explosion which followed. This 
 explosion clearly shows that the greatest violence was at a point 
 some distance away from the point of origin. Although the 
 source of origin was not in the elevator, the conditions were such 
 that when the flame and pressure reached this point the ignition 
 of the dust thrown in suspension produced a violent explosion. 
 
 Explosion in Western New York, January, 1910 
 
 An explosion occurred in a cereal mill in Western New York 
 on January 4, 1910. As a result of the fire which followed the 
 explosion several men were killed, and the plant badly damaged. 
 The explosion occurred at night, when there were only a limited 
 number of men at work. There had been a few days' suspension 
 of work for the purpose of taking an inventory, and the plant had 
 been cleaned generally throughout, the dust having been swept 
 up and removed. The weather was very cold, the temperature 
 being below zero. The cause of the explosion has never been 
 definitely determined, owing to the complete burning of that 
 section of the plant in which it occurred. 
 
 Explosion in Indiana, 1910 
 
 An explosion occurred in a starch factory in Indiana on 
 March 7, 1910, accompanied by much violence and property 
 damage. One of the causes of the original ignition is thought to 
 have been a fire that originated in the starch kiln house tunnels. 
 According to this theory, a piece of burning waste, or other 
 burning matter, was blown into these drying tunnels through the 
 medium of the large fan blowers (forcing air through large coils 
 of steam heater piping into the tunnels) , This burning substance 
 is supposed to have come in contact with some re-dried starch and 
 immediately ignited the same. The fire which started in this 
 manner spread into four of the tunnels, and the combustion was 
 so rapid and the heat so intense that a number of employes were 
 very severly burned. Owing to the presence of the fine dry starch 
 dust, the fire spread through a conveying system to an adjoining 
 building, which was destroyed. 
 
 Another theory advanced relative to this explosion was that 
 the point of origin was in the starch grinding house, and that the 
 
Explosions in the United States 13 
 
 flame was transmitted to the adjoining building by means of the 
 conveyor lines. This is made possible by the fact that any hard 
 substance or foreign material, striking against the revolving arms 
 or plates on the interior of the grinding machine, might produce a 
 series of sparks that would cause an ignition of the starch dust 
 cloud on the interior of the machine. The type of machine in use 
 consisted of a series of arms or beaters mounted onto a revolving 
 shaft or runner head that revolved at a very high rate of speed. 
 It was thought that a series of sparks produced in this manner 
 caused a fire in the conveyor line, that propagated to a very disas- 
 trous explosion. In this respect the traveling of the explosive 
 wave for some distance from the point of origin to the point where 
 the greatest damage was done resembles an explosion already 
 referred to above. It would seem to indicate that the initial 
 point of ignition is not always where the greatest amount of 
 damage is done, and also that at the point of ignition there may 
 be indications of very little damage. 
 
 Explosion in Illinois, 1912 
 
 An explosion occurred in a starch factory in Illinois on 
 November 25, 1912, and was accompanied by a large loss of life, 
 and complete loss of property. The general opinion relative to the 
 origin of this explosion seems to be that it started in the starch 
 drying house. The dust is usually very fine and dry, and when 
 in suspension in the air in sufficient quantity produces a dan- 
 gerous condition. The exact cause has been a matter of conjec- 
 ture, and owing to the workmen being killed was probably never 
 definitely determined. The possibility of one of the men igniting 
 the dust cloud by striking a match, was advanced. The explosion 
 is described as being terrific in nature, and the effect was felt 
 for some distance from the plant. 
 
 Explosion in Western New York, June, 1913 
 
 The most violent explosion of grain and cereal dusts in this 
 country in recent years occurred in a feed-grinding plant in 
 Western New York on June 24, 1913. The explosion was followed 
 by a very disastrous fire, which destroyed and damaged property 
 for some distance away. There were 33 lives lost as a result 
 of the explosion and upwards of 70 people injured. The explosion 
 occurred in the afternoon, about 4 o'clock, on a very warm, sultry 
 day, with the temperature about 90 degrees, and the humidity 
 
14 The Explosibility of Grain Dusts 
 
 about 74 per cent. This reading, however, was obtained by the 
 United States Weather Bureau, some distance away from the 
 plant, and may not represent the humidity conditions at the mill. 
 When the explosion occurred, the mill was in the regular process 
 of operation, engaged in the production of feeds. The explosion 
 is described as consisting of two reports, one following the other, 
 with the second report louder than the first. Numerous theories 
 have been advanced relative to the origin of the explosion, but 
 the exact cause will probably never be determined definitely. An 
 inquest of extended sessions was conducted by the City Court, 
 with an effort to determine the cause and circumstances of the 
 explosion. The decision of the Court, after examining 110 wit- 
 nesses, many of whom were employes, was that "the testimony, 
 without exception, has been that the plant on that day was in 
 good working order, and that all ordinary precautions for the 
 protection of the men had been taken. No evidence has been 
 produced that would throw any light upon the cause of the acci- 
 dent, and I therefore find that the men whose deaths were caused 
 by that accident, came to their deaths from fire originating from 
 causes unknown." 
 
 Explosion in Iowa, September, 1913 
 
 An explosion occurred on the night of September 11, 1913, 
 in a cereal mill in Iowa. Fortunately no lives were lost, but the 
 property was considerably damaged by the explosion and the fire 
 which followed. The explosion occurred in a section of the plant 
 where oat hulls were being ground for feed, and the greatest 
 damage was done in this section. The explosion was charac- 
 terized by a sudden puff, or report, followed by a body of flame 
 and fire. The conveyor lines from the grinding machine to the 
 storage bins gave evidence of fire, and the explosive wave seemed 
 to travel through this part of the plant. Although the grain is 
 said to have been reasonably clean, it is thought that some foreign 
 material may have been introduced into the grinding machine, 
 producing sparks sufficient to ignite the dust cloud present. The 
 flame thus produced traveled along the conveyor lines and elevator 
 legs until the receiving bins or hoppers were reached, when the 
 explosion occurred. 
 
 Accident in Western New York, September, 1913 
 
 On September 13, 1913, an accident occurred in a feed mifl 
 in Western New York. The occurrence cannot be considered as 
 
Explosions in the United States 15 
 
 a complete explosion, but rather a series of inflammations of dust 
 with limited explosive force and violence. Fire was discovered 
 in one of the conveyor lines under the fourth floor of the building, 
 and repeated efforts were made to prevent the spreading of the 
 flames. A few hours after fire was first discovered the flames 
 broke out with much force and violence, and considerably dam- 
 aged the property. For some time preceding the accident it is 
 thought trouble had been experienced with foreign material 
 entering the grinding machines, and it is supposed the original 
 sparks that ignited the dust were produced in this manner. It 
 would appear, however, that the presence of inflammable dust 
 increased the spread of the flames, and added to the extent of 
 damage and violence. 
 
 Explosion in Kansas, November, 1913 
 
 A slight explosion occurred in a flour mill in Kansas in 
 November, 1913. The explosion was of limited proportions, and 
 there was no loss of life involved nor extensive property damage 
 done. The occurrence took place about two weeks after an 
 attrition mill had been installed for grinding wheat screenings. 
 The mill was installed with a metal bin directly under the grinding 
 machine, discharging the ground material into this bin, conveying 
 the same from this bin to the elevator boot. At the time of the 
 explosion there was not sufficient dust present to cause serious 
 damage, and the only damage done was a bulging out and loosen- 
 ing of the sides of this metal bin. The explosion made a very loud 
 report and filled the mill with smoke. This occurrence would 
 indicate that a grinding machine of this nature should not dis- 
 charge directly into a bin unless provision is made in some manner 
 for excessive pressure produced in case of an ignition of dust. 
 
 Explosion in Iowa, November, 1913 
 
 An explosion took place in a cereal mill in Iowa on November 
 6, 1913. Two men were very badly burned and the property 
 suffered considerable damage as a result of the accident. The 
 explosion occurred during the noon hour, when most of the 
 employes were outside the plant. The tender at the attrition 
 mill was sitting near the machine, eating lunch, when he heard 
 an explosion inside the machine. The explosion is described as 
 being light in nature, and was followed by a later explosion of a 
 loud, rumbling nature, that shook the entire plant. The flames 
 
16 The Explosibility of Grain Dusts 
 
 were carried by means of conveyor lines and elevator legs up into 
 a large storage bin. This bin contained only a small amount of 
 ground material, and was almost empty at the time. The flames, 
 upon reaching the bin, ignited the dust in suspension, thereby 
 producing the violent explosion which followed. 
 
 Explosion in Ohio, December, 1913 
 
 An explosion occurred early in December, 1913, in an elevator, 
 in Ohio. There were no lives lost, and the damage to the property 
 was limited. However, the fact that the explosion occurred in 
 an elevator leg, during the transfer of grain, at a point in the 
 elevator building where there was no grinding machinery, would 
 indicate that grain elevators, as well as milling plants, may be 
 subject to explosions of dust produced in the process of trans- 
 ferring and elevating grain under normal conditions. 
 
 The explosion occurred in an elevator leg used for taking 
 grain from one bin and transferring to another. The leg con- 
 tained 178 buckets, 12 inches apart, and projecting 6 inches from 
 the belt. The size of the leg was 18 inches by 9 inches on the 
 outside, and the buckets were running about 471 feet per minute. 
 This elevator leg was situated in the middle of the elevator 
 building, about 20 feet from the front. It is thought the explo- 
 sion occurred in or near the first floor of the elevator. The foot of 
 the elevator extended down into the sub-basement, but there did 
 not seem to be any evidence of explosion at that point. The foot 
 was intact, the dust was not disturbed in the foot, or the conveyor 
 running into the foot. The leg on the first floor was blown olf, 
 and the head at the top of the elevator was blown apart. After 
 the explosion occurred there was an odor present that resembled 
 the odor following the discharge of a shotgun. This was quite 
 noticeable on the first floor, and on the top of the elevator where 
 the flames escaped. The officials concluded that it was possible a 
 loaded cartridge had become mixed with the grain in shipment, 
 and due to the friction in handling, was exploded, and ignited 
 the dust. 
 
 At the time of the explosion one of the men was using an 
 extension light cord, with an incandescent bulb attached, to look 
 into the leg through an opening on the first floor. This examina- 
 tion was being made at the time the explosion occurred, and later 
 it was found that the extension had been burned out. It was not 
 
Recent Explosions in Europe 17 
 
 definitely determined what relation this had to the explosion, but 
 suggests the possibility of an ignition of a dust cloud of this 
 nature by an electric spark. This occurrence in Ohio, with others 
 of its kind, indicates that the dusts arising from the handling of 
 grain, such as oats, corn, wheat, etc., are of a highly explosive 
 nature. An opinion exists among some elevator owners that the 
 only dangerous dusts are those produced during the grinding and 
 milling processes. They base this belief on the fact that grain 
 coming from the field would be mixed with sufficient incom- 
 bustible material to render the dust produced somewhat inert, 
 and therefore not explosive. The dusts, no doubt, contain a large 
 quantity of field dirt, but the dust coming from the husks or 
 shells of the grain mixes in very large proportions with the field 
 dust already present, and the mixture in this way is made inflam- 
 mable. The large number of explosions of this nature in grain 
 elevators where there is no grinding done, proves very clearly 
 that an explosion may occur under certain favorable conditions. 
 
 Explosion in Texas, March, 1914. 
 
 An explosion occurred in an elevator in Texas in March, 1914. 
 The report of the explosion is described as resembling a muffled 
 roar, similar to that of a heavy gun fired within the building. 
 Fortunately no one was hurt, but the force of the explosion was 
 sufficient to blow out the sides of the spouting floor of the 
 building. It is another example of the possibility of an explosion 
 of elevator dusts. 
 
 RECENT EXPLOSIONS IN EUROPE 
 Explosion at Glasgow, Scotland, November, 1911 
 
 N EXPLOSION occurred in a provender mill, grinding 
 
 Au) peas, beans and wheat, at Glasgow, Scotland, on 
 \^ November 10, 1911. As a result of this explosion two 
 men employed by the firm, and three children who 
 were passing along the street at the time of the accident, were 
 killed, and eight persons were injured. The explosion seems to 
 have been caused by dust, which had accumulated on an overhead 
 beam, being dislodged and falling upon the open flame of a 
 Bunsen burner. The operations carried out in this factory con- 
 sisted in grinding peas, beans and wheat. At the time of the 
 explosion all the millstones, except those used for grinding 
 
18 The Explosibility of Grain Dusts 
 
 wheat, were running. The accident occurred at about 6:30 P. M., 
 and the work was being carried on by artificial light. The mill 
 was lighted by naked gas jets, and though most of these were in 
 fairly safe positions, a light of a portable nature appears to have 
 been placed in a spot where dust could have fallen on it. The 
 explosion clearly indicates that a disastrous explosion may be 
 produced by an open gas jet, in a plant where dust of this nature 
 is produced. 
 
 Explosion at Liverpool, 1911 
 
 Probably the most disastrous explosion on record, as far as 
 loss of life is concerned, occurred in a linseed mill in Liverpool, 
 England, on November 24, 1911. As a result of this explosion 
 39 persons were killed, and 101 were injured. The explosion was 
 thought to have originated in the basement of the plant, where six 
 large disintegrators for grinding oil cakes, beans and other mate- 
 rials, and the necessary elevators, were at work. It was consid- 
 ered to be due to the ignition of a dense cloud of dust produced 
 by the breaking of a driving belt on one of the machines. This 
 belt was six inches wide, and ran at a speed of about 5,000 feet 
 per minute, and it was said that when one of these belts broke a 
 fog of dust was raised owing to the disturbance of dust on the 
 girders in the room. The definite cause of the ignition of the 
 dust cloud which was raised was not fully determined, and is only 
 a matter of conjecture. Experiments conducted with samples 
 of the dust collected from the tops of pipes and wall ledges 
 showed that an explosion of a dense cloud of such dust could be 
 caused by: 
 
 (1) The ignition of a match; 
 
 (2) Contact with a naked gas flame ; 
 
 (3) The flash caused by fusing of an electric wire; 
 
 (4) Sparks produced by breaking circuit of an electric 
 magnet. 
 
 This latter test was carried out because the disintegrators 
 were fitted with electro-magnetic separators. There were no 
 lighted gas jets in that part of the works affected, and there did 
 not seem to be any direct evidence of the burning of matches. 
 Of the two remaining possible causes, the inspector concluded 
 that the burning of a fuse on a temporary switchboard for light 
 installation, provided the flame necessary for the ignition of the 
 dust. Although this cause cannot be definitely assigned, it indi- 
 
Summary of Explosions 19 
 
 cates that we can derive valuable conclusions from this occur 
 rence relative to the relation of the dust cloud and electric sparks. 
 
 Explosions at Manchester, England, 1911 and 1913 
 
 Two explosions that appear to be somewhat similar in nature, 
 occurred in a dextrine works at Manchester, England, on March 
 21, 1911, and March 11, 1913. In the first explosion three men 
 were killed, and five others injured. As a result of the second 
 explosion eight men were injured, of whom three subsequently 
 died. The operators carried on the manufacture of dextrine 
 (British gum). Starch, ordinary flour, and a small quantity of 
 hydrochloric and nitric acids were the only substances used. The 
 inspectors in charge of the investigation concluded that the dust 
 explosions in both occurrences originated in the same manner. 
 It is thought that an ignition of dust was produced inside the fan 
 casing that was forcing the hot air to the drying stoves located 
 in the end of the plant. This ignition in the fan occurred as 
 follows: A small accumulation of dust being dislodged, possibly 
 from the beam supporting the fan, dropped onto an open gas jet 
 about tWo feet below the beam, and was ignited. This flame was 
 drawn into the fan casing, either through a badly fitting inspec- 
 tion door in the air inlet duct, or through the openings in that 
 duct, which fitted badly at its junction with the casing of the fan. 
 The flame thus produced was driven along the hot air duct, setting 
 up an explosive wave within the duct, which finally exploded the 
 contents of the drying stoves, either by firing directly the dust 
 at the bottom of the stoves, or by causing the evolution of inflam- 
 mable gases (by destructive distillation of the dust on the trays 
 and racks), which were fired by the flame and immediately pro- 
 pagated an explosion. A series of tests of samples of dust 
 collected at this plant on the day following the explosion, showed 
 that all the samples collected were highly inflammable, and when 
 in suspension in the air were capable of propagating or spreading 
 the flame very rapidly. 
 
 SUMMARY OF EXPLOSIONS 
 
 Y^'^YHE PRELIMINARY work in the study of grain 
 "V 7|T X dust explosions, as already stated, has been de- 
 j\ fii ^^^^^ ^^ ^^^ explosions occurring in more recent 
 
 0^<&«5 years. It will be observed that a large number of 
 explosions of this nature have occurred during the last ten years 
 in this country and abroad. The summary obtained up to the 
 
20 The Explosibility of Grain Dusts 
 
 present time indicates that at least thirteen of these explosions 
 have occurred in the United States since 1905, and of the number 
 stated, six have occurred since the explosion in Western New York 
 in June, 1913. This would seem to indicate that a larger number 
 of explosions may have occurred during the ten-year period than 
 already enumerated, but have not been a matter of record. The 
 fact that this study began August 1, 1913, explains the complete 
 record of these occurrences during the last year. The explosions 
 noted in this country have occurred in plants that were consid- 
 ered modern, and where advantage was taken of modern milling 
 ideas and improved mechanical devices. The explosions have not 
 been traced to carelessness in any case, but the indications are 
 that they probably have taken place under unusual conditions. 
 For this reason it may safely be concluded that a much larger 
 number than the thirteen explosions mentioned have occurred 
 during this period, and in cases where life was not involved it is 
 quite possible that the occurrence was not recorded. It was also 
 found at the beginning of the work, when an effort was being 
 made to obtain a record of explosions of this nature, that very 
 little definite information was available. Owing to the field being 
 somewhat new, data of any assistance was very limited, and 
 information pertaining to each explosion was necessarily obtained 
 from men who had little knowledge of other accidents of this 
 nature. It is possible that explosions of this kind have occurred 
 in diflferent sections of the country, reports of which have not 
 been published. However, the fact that so large a number of 
 occurrences of this kind have taken place during the past year 
 is of sufficient importance to attract our attention and demand 
 careful consideration. 
 
 Reports show that the same situation exists in European 
 countries, especially in milling plants in England.* 
 
 Four of these explosions have occurred in England during 
 the last three years, the one at Liverpool, November 24, 1911, 
 being exceptionally destructive. A careful study of occurrences 
 of this nature abroad would, no doubt, show that dust explosions 
 are frequent also in European milling plants. 
 
 ♦See reports on the explosion at the works of William Primrose & Co., Ltd., 
 by W. S. Smith, H. M. Inspector for Dangerous Trades, 1911. 
 
 Report on explosion at works of J. Bixby & Sons, Formby street, Liverpool, 
 by John Jackson, H. M. Inspector of Factories, 1911. 
 
 Report on Explosion at works of Messrs J. Laing, Son & Co., Holt Town, 
 Manchester, by John Jackson, H. M. Superintending Inspector of Factories, 1913. 
 
The Origin and Distribution of Dusts 21 
 
 ORIGIN AND DISTRIBUTION OF DUSTS 
 
 *T IS just as impossible to prevent dust from being 
 
 J\\J produced in the milling industry as it would be to pre- 
 \\ vent dust from forming during the mining of coal. 
 The milling dusts are usually fine, and owing to the 
 necessary absence of moisture are very dry, presenting dangerous 
 conditions when in suspension in the atmosphere. There are so 
 many different kinds of dust produced in the handling and milling 
 of grain that it would be impossible to enumerate them fully in 
 this preliminary report. 
 
 Samples of the following dusts have been collected, and the 
 conditions under which they have been produced have been 
 studied : 
 
 (1) Dusts produced during the process of elevating and 
 handling grain, and known as elevator dusts ; 
 
 (2) Wheat flour dusts from rolls, bolters, purifiers, con- 
 veyors, packing machines, etc. 
 
 (3) Wheat flour dusts from beams, rafters, elevator 
 heads, etc. 
 
 (4) Dusts produced during the cleaning of oats; 
 
 (5) Dust from grinding white corn; 
 
 (6) Dust from grinding yellow corn; 
 
 (7) Dust from grinding oat hulls; 
 
 (8) Oatmeal dust from packing machines; 
 
 (9) Floor dusts from elevator sweeping; 
 (10) Oat groat dusts after aspirator. 
 
 A total of about 150 samples of the various dusts mentioned 
 have been collected during the various stages of elevating and 
 grinding. 
 
 Elevating Process 
 
 During the process of elevating grain from the boats or 
 railroad cars large quantities of dust are produced. The coarse 
 or heavier dust settles on the floors, steps, machinery, etc., while 
 the very fine dust rises into the atmosphere, and settles on beams, 
 rafters and other inaccessible points. The usual system adopted 
 for the removal of this dust is by the ordinary "push broom" 
 method, and although at the time the very coarse and heavier 
 portions of the dust is removed, the finer or more dangerous 
 portion rises into the atmosphere and settles along with the other 
 fine dust already accumulated from the elevating process. 
 
22 The Explosibility of Grain Dusts 
 
 Escape from Machines and Conveyors 
 
 The bolters, rolls, purifiers, etc., all produce a large quantity 
 of dust during their process of operation, and if a satisfactory 
 system of dust collecting is not installed a portion of the dust 
 may escape into the surrounding atmosphere. During the grind- 
 ing process of any of the grains, considerable dust is produced 
 that may also escape into the air if proper provisions are not 
 taken to keep the machinery in repair. The conveyor lines and 
 elevator legs also carry continually a very large quantity of fine 
 dust, and in case of any defect in construction will allow the dust 
 to enter the surrounding air. At the discharging point of grain 
 into storage, an opportunity is given for the dust to escape into 
 the air and settle on surrounding projections. In addition to the 
 various sources enumerated, there are possibly a number of 
 others, but the ones mentioned generally cover the sources that 
 produce the largest quantity. An important consideration in the 
 prevention of the accumulation of dangerous dusts would be to 
 attack the point where the dust is produced, and not deal exclu- 
 sively with its removal after it has settled in dangerous 
 quantities. It is true that the entire escape of dust cannot be 
 prevented, but it can be reduced to such an extent that its com- 
 plete removal can be more certain. 
 
 Oat Hull Grinding 
 
 One of the lines of milling that seems to be accompanied by 
 a large number of explosions is the grinding of by-product mate- 
 rial, especially oat hulls, in the production of feeds. It is not by 
 any means to be concluded that this is the only source of danger, 
 but our attention has been drawn to the large number of occur- 
 rences of this nature in more recent years. The oat hulls are 
 ground by various types of attrition mills, and as a rule the plates 
 are set very close together for fine grinding. 
 
 EFFICIENCY OF DUST-COLLECTING SYSTEMS 
 
 I HE INTRODUCTION of dust-collecting systems in the 
 77i lit milling industry has succeeded in the practical elim- 
 ^i' (Is ination of the old dust or "stive" room. When this 
 dust room was in use, such as at the time of the 
 Minnesota explosion in 1878, and the Illinois explosion in 1893, 
 
Efficiency of Dust-Collecting Systems 23 
 
 the dust was conveyed or carried to the "stive" room from the 
 various parts of the mill. This always made an "explosive 
 chamber," as it were, allowing sufficient dust to be in suspension 
 to produce a violent explosion, if ignited. With the present 
 system of dust collection as applied to modern mills this particular 
 source of danger is done away with, and, by means of air cur- 
 rents, the dust is carried to the collector and deposited. Many 
 millers seem to feel a sense of security if such a dust-collecting 
 system is in good working order, and that this danger from explo- 
 sions is practically eliminated. Owing to the difference in types 
 of grinding machines used in flour and feed milling, the flour 
 miller feels an additional amount of protection from the fact that 
 a series of sparks necessary to ignite the dust cloud, will not be as 
 readily produced by the grinding "rolls" as when attrition or 
 friction mills are used. However, it would not be advisable to 
 conclude that flour mills are not as liable to experience explosions 
 as feed or cereal mills, owing to this one difference of machinery 
 installation. In addition to the mill proper, fires very frequently 
 occur in the elevator buildings. Since experiments have shown 
 that the elevator dusts are highly inflammable, it is possible to 
 initiate a violent explosion in the elevator annex that may spread 
 throughout the entire milling plant. The relation between the 
 elevator and the mill buildings is so close at many milling plants, 
 that an ignition of dust in one may readily propagate to a violent 
 explosion in the other. 
 
 During the progress of the study now being conducted the 
 efficiency of the different types of dust-collecting devices received 
 considerable attention. With the advances in milling many com- 
 panies utilized the by-products from the grains for feeds, which 
 necessitates, in many cases, very fine grinding, thereby producing 
 a very large amount of dust. The efficiency of the various types 
 of dust collectors installed to collect the dust produced, introduces 
 an important point for consideration. 
 
 Observations in various mills would seem to indicate that at 
 times some of the dust collectors do not work as effectively as 
 at other times. In some cases dust has been in suspension, 
 forming a noticeable cloud near the collecting device, probably 
 due to defective equipment. When this occurs the very finest 
 part of the dust leaks out into the atmosphere, and when mixed 
 with sufficient air forms a dangerous mixture. 
 
24 The Explosibility of Grain Dusts 
 
 Relation of Dust-Collecting Systems to Explosions 
 
 The relation that the dust-collecting system has to the pro- 
 pagation or spreading of the explosion is of sufficient interest to 
 demand attention. This would appear especially true when we 
 observe that many explosions have occurred in recent years where 
 dust-collecting systems were said to have been installed and in 
 operation throughout the plant at the time of the occurrence. 
 The question arises as to how efficiently these systems were 
 operating at the time of the explosion. The fact that it is 
 possible to have disastrous explosions in plants that are 
 considered modern, as far as provision for dust collecting is 
 concerned, would give sufficient reason for an investigation to 
 determine the relation between the efficiency of the system and 
 the propagation of the explosion. 
 
 Some types of dust collectors are constructed in such a 
 manner that it requires a series of "air draft lines" to complete 
 the system. The circulation of air produced by the fan allows 
 sufficient air current in circulation to keep a possibly dangerous 
 amount of dust in suspension within the enclosed area, whether 
 it be a pipe, box or conveyor line. This dust cloud, when ignited, 
 will readily propagate to a disastrous explosion. 
 
 LABORATORY INVESTIGATIONS 
 
 Introduction 
 
 >HERE are a great many questions and problems in a 
 /|T V§ grain dust investigation for both the engineer and 
 w ^ chemist. Among the latter are to be considered the 
 V^==:si^=;^ chemical nature and composition of the dusts, their 
 ignition-temperature and explosibility, — as they occur, and also 
 of the fine material contained in them, — the effect of humidity, 
 moisture and chemical composition upon their explosibility, and 
 also the effects of oxidation. The present laboratory study of 
 these dusts was undertaken to determine some of these factors, 
 and others that may suggest themselves, such as sources of 
 possible ignition. 
 
 The results given herein are preliminary, and are given 
 simply as an indication of what is being done and what may be 
 expected in future work. They should not be considered as 
 absolute, but as giving a possible comparison of the chemical 
 
Laboratory Investigations 25 
 
 composition, and also of the explosibility of the dusts tested. Any 
 suppositions or possible conclusions that may be drawn are 
 subject to change as the result of future work. 
 
 Previous Work 
 
 A large amount of work has been done upon the inflamma- 
 bility of coal dusts abroad and in the United States, but very little 
 in comparison has been done upon other carbonaceous dusts such 
 as those produced in milling industries. In 1845 Faraday and 
 Lyell gave out the supposition that coal dust in suspension in air 
 might propagate an explosion started by firedamp. This is doubt- 
 less the first suggestion made that coal dust would propagate a 
 flame. For many years it was overlooked. In 1873 Watson Smith 
 published in the Glasgow Herald the fact that a mixture of dry 
 flour and air was inflammable. 
 
 Following the diseaster in Minnesota in 1878, Professors 
 Peck and Peckham* carried out some tests upon the explosibility 
 of flour dusts. They found that two ounces of these dusts 
 together with two cubic feet of air would produce, when ignited 
 in a box with a flame, an explosion that would lift two men 
 standing on the cover. It has been calculated' that a sack of flour 
 suspended as dust in 4,000 cubic feet of air (a room 20x20x10) , 
 when ignited, would generate sufficient force to throw 2,500 tons 
 100 feet high. 
 
 More recent work upon the inflammability of carbonaceous 
 dusts other than coal has been carried out by R. V. Wheeler, 
 Chief Chemist for the Explosion in Mines Committee, England.* 
 He tested a large number of various kinds of dusts by two 
 different methods — one for the purpose of discriminating between 
 harmless and dangerous dusts; the other with the intention of 
 ascertaining the temperatures at which inflammation of the 
 dangerous dusts takes place readily. As a result of these tests 
 he divided the dusts into three classes, namely: — 
 
 "Class 1. — Dusts which ignite and propagate flame readily, the source 
 of heat required for ignition being comparatively small; such, for example, 
 as a lighted match. 
 
 *Mines and Minerals, 1908, 29, 55. 
 
 tC. E. Munroe, J. Amer. Chem. Soc, 21, 321. 
 
 JReport on the Inflammability and Capacity for Transmitting Explosions 
 of Carbonaceous Dust, Liable to Be Generated on Premises, Under the Factory 
 and Workshop Acts. 1913. R. V. Wheeler, D. Sc. 
 
26 
 
 The Explosibility of Grain Dusts 
 
 "Class II. — Dusts which are readily ignited, but which for the propaga- 
 tion of flame require a source of heat of large size and high temperature 
 (such as electric arc), or of long duration (such as the flame of a Bunsen 
 burner). 
 
 "Class III. — Dusts which do not appear to be capable of propagating 
 flame under any conditions likely to obtain in a factory; either (a) because 
 they do not readily form a cloud in air, or (l>) because they are contaminated 
 with a large quantity of incombustible matter, or (c) because the material 
 of which they are composed does not burn rapidly enough." 
 
 Dusts of interest in the present investigation were classified 
 by Wheeler as follows: 
 
 Class I: Sugar 
 Starch 
 
 Rice, meal and sugar refuse 
 Wood flour 
 Malt 
 Oat husk 
 
 Class II: Rice milling 
 
 Castor oil meal 
 Off^al grinding (bran) 
 
 Class III: Spice milling 
 Cotton seed 
 Cotton seed and soya bean 
 
 Grain (flour mill) 
 Maize 
 
 Grain (grain storage) 
 Rape seed 
 Cornflour 
 Flour (flour mill) 
 
 Grist milling 
 Horn meal 
 Mustard 
 
 Sack cleaning 
 
 Rape seed (Russian) 
 
 Grain cleaning 
 
 The classification, as here given, is according to the inflamma- 
 bility of the sample tested. Other results might be obtained 
 from other samples of the same material, especially those placed 
 in Class III. 
 
 In the first test which he used he determined the relative 
 ignition-temperatures of the dusts, that is, the temperature to 
 which the dust must be heated before it will propagate a flame. 
 For this purpose he used a glass cylinder (Fig. 1) 8 cm. (3.14 
 inches) in diameter and 140 cm. (55.14 inches) long, open at both 
 ends, and supported in a horizontal position. A platinum coil of 
 32-gauge wire, closely wound on a thin-walled tube of quartz of 
 capillary bore, passed horizontally across a cylinder at a point 
 40 cm. (15.75 inches) from one end. Through the bore of the 
 quartz tube a platinum vs. platinum-rhodium thermo-couple 
 passed, and was connected to a mille-voltmeter calibrated so as to 
 read temperatures on the Centigrade scale. By means of suitable 
 connections an electric current could be passed through the plat- 
 
Laboratory Investigations 
 
 27 
 
 inum coil so that it could be heated to any temperature up to 
 about 1400° C. (2552° F.), and maintained at a constant tem- 
 perature by the adjustment of the external resistance. The length 
 of the heated coil was 17 mm. (.669 inches) and its diameter was 
 1.5 mm. (.059 inches). 
 
 
 A 
 
 
 Z)us^ So be lesieeZ 
 
 Plaivrt t4,m. colZ 
 
 ^ea^ejTO cu^re?t,t 
 
 
 
 £rt.lajn^e<i scc^coft. ojcj'os.f c.d. 
 
 Two grams (.07 ounces) of the dust to be tested was placed 
 in an even layer along a glass tube of 2.5 cm. (0.97 inches) 
 internal diameter, 45 cm. (17.8 inches) long. This tube was 
 closed at one end by a rubber stopper carrying a glass tap of 
 
 I cm. (.39 inches) bore. It was supported in the position shown 
 in the diagram (Fig. 1), the open end being at a distance of 30 
 cm. (11.8 inches) from the heated platinum coil. The tap was 
 connected with an arrangement for giving a constant puff of air, 
 consisting of a brass cylinder 65 cm. (25.85 inches) long and of 
 
 II cm. (4.35 inches) internal diameter, fitted with a piston 
 weighted so as to give a pressure of 2 pounds per square inch. 
 
 The platinum coil having been heated to such a temperature 
 as preliminary trials may have indicated as being about that 
 required, the dust cloud was produced by suddenly opening tap B. 
 The air blast, passing over the surface of the dust in the tube 
 (A), raised the top layer and carried it into the larger tube and 
 over the heated coil in a cloud that remained uniform during the 
 
28 
 
 The Explosibility of Grain Dusts 
 
 stroke of the piston. If ignition occurred the temperature of the 
 platinum coil was lowered 10-20° C. (18-36° F.) and a fresh trial 
 made; and so on until two temperatures were obtained differing 
 by 10° C, at one of which inflammation took place, whilst at the 
 other the dust cloud passed over uninflammed. The mean tem- 
 perature was thus taken to be the ignition temperature of the 
 dust under the conditions of the test. 
 
 Using this method on the dusts listed above (all of which 
 were dried during one hour at 107° C), Wheeler gives the follow- 
 ing results, which show the relative ignition temperature of 
 the dusts: 
 
 Percentage 
 of sample as 
 
 received Igni- 
 that passed tion 
 a 200-mesh Tern. 
 Substance. sieve. Degr. C. 
 
 Wood flour Fluffy, 
 
 could not 
 sieve 
 Rice meal and sugar 
 
 refuse Fluffy, 
 
 could not 
 sieve 
 
 Rape seed Did not 
 
 sieve, 
 
 sample 
 
 too small 
 
 Rape seed (Russian) 90 
 
 Castor oil meal 50 
 
 Grain (flour mill) 50 
 
 Spice milling 50 
 
 Grist milling 30 
 
 Sack cleaning Fluffy, 
 
 could not 
 
 sieve 
 
 Rice milling 20 
 
 Offal grinding (bran).. 75 980 
 
 Sugar 50 805 
 
 Grain (grain storage)... 20 1050 
 
 Starch 80 1035 
 
 Remarks. 
 
 985 Cloud formed easily, although 
 the sample was unsieved. 
 
 970 Cloud formed easily and flame 
 traveled rapidly in it. 
 
 1050 Ignition temperature probably 
 lower, but no more dust 
 available. 
 
 No ignition. 
 
 1100 
 995 
 
 A small flare appeared at the 
 
 coil at 1000° and upward, 
 but no flame was propagated. 
 
 A small flare at coil, but there 
 
 was no propagation of flame. 
 
 No ignition. 
 
 Small flare at coil, but no pro- 
 pagation. 
 
 Flame traveled very rapidly in 
 the cloud. 
 
Laboratory Investigations 
 
 29 
 
 Percentage 
 of sample as 
 
 received Igni- 
 that passed tion 
 a '200-mesh Tem. 
 Substance sieve Deg. C Remarks 
 
 Starch 100 960 Flame propagated rapidly. 
 
 Grain cleaning No ignition. Sample much too 
 
 coarse to form cloud. 
 
 Horn meal 50 No ignition, a few sparks only. 
 
 Oat husks 90 990 
 
 Malt 50 990 Flame propagated rapidly. 
 
 Mustard 50 1050 
 
 Flour 90 1060 
 
 Maize 70 1010 
 
 Cornflour 90 1060 
 
 The second test carried out by Wheeler, to determine the 
 lowest temperature at which ignition can be effected, consisted 
 in passing the dust cloud down through a tube heated by means 
 of a copper coil. As a result of these tests he gives the following 
 data: 
 
 Ignition "Percentage of in- 
 
 teniperature combustible 
 
 Substance. Deg. C. matter. 
 
 Wood flour 610 3.2 
 
 Rice meal and sugar refuse 630 8.1 
 
 Rape seed 650 No dust for analysis 
 
 Castor oil seed 655 23.1 
 
 Grain (flour mill) 630 18.2 
 
 Spice milling 680 
 
 Very small flare 29.9 
 
 Grist milling 600 7.8 
 
 Sack cleaning No ignition could 
 
 be obtained 74.6 
 
 Rice milling 630 4.2 
 
 Offal grinding (bran) 640 30.0 
 
 Sugar 540 1.4 
 
 Grain (grain storage) 630 10.5 
 
 Starch 640 1.5 
 
 Starch 630 0.4 
 
 Grain cleaning Dust not tested; 
 
 much too coarse 8.9 
 
 Horn meal 670 5.3 
 
 Oat husk 620 13.4 
 
 Malt 600 11.7 
 
 Mustard 680 5.1 
 
 Flour 650 1.5 
 
 Maize 645 8.0 
 
 Cornflour 620 0.9 
 
30 The Explosibility of Grain Dusts 
 
 Outside of the work referred to, very little published work 
 of importance has been done upon carbonaceous dusts other than 
 coal dusts. 
 
 Experimental Work 
 
 The original laboratory work in this investigation was started 
 in the United States Food and Drug Laboratory, Chicago, under 
 the direction of Dr. A. L. Winton. Twenty-one samples of repre- 
 sentative dusts were analyzed, using the methods in vogue in 
 that laboratory. Determinations of moisture, ether extract, 
 proteins, crude fiber, ash, and nitrogen free extract or carbo- 
 hydrates, were made to determine, not only the chemical nature 
 of the materials, but also wherein they might differ from the 
 grains from which they originated. These results will be given 
 in a table below. 
 
 At the completion of this part of the work, direct experi- 
 ments upon the explosibility of these dusts were begun in the 
 Bureau of Mines laboratories, Pittsburgh, under the direction of 
 Dr. J. K. Clement. The first tests carried out were with the 
 French apparatus, as used by M. J. Taffanel. This consisted of a 
 tube in which was placed 0.5 gram (.0176 oz.) of the dust to be 
 tested. This was connected with a canvas bag, inflated with 
 oxygen under 400 cm. (15.7 in.) of water pressure. The tube 
 carried a stopcock which, when opened, allowed the dust to be 
 forced by the oxygen, in a cloud, upon a benzine lamp gas flame. 
 A relative inflammability of the dusts could be obtained by noting 
 the nature of the flame, and the intensity of the report, if any 
 resulted. Definite results could be obtained only photographically, 
 but this has been postponed to a later date. All the dusts tested, 
 which were the same as those analyzed, were found to be inflam- 
 mable, even when using air in place of oxygen to force the dust 
 cloud upon the flame. 
 
 Ignition-Temperature 
 
 Experiments were then started to determine the relative 
 inflammability of the dusts, using the method of Wheeler, out- 
 lined above in his first tests. The apparatus as used in these tests 
 was a modification of that used by Wheeler. Instead of a long 
 glass tube carrying a platinum coil 15.75 inches from one end, 
 two tubes of 3 inches internal diameter were used, these fitting 
 into a cast iron, asbestos-lined collar, which carried a coil of 
 32-gauge platinum wire, simflar to that used by Wheeler, but 
 
Laboratory Investigations 31 
 
 1 inch long, 14 inch external diameter, and covered with a thin 
 coating of alundum cement to protect the wire. The tube holding 
 the dust was % inch diameter and 6i/i inches long, slightly con- 
 stricted so as to make the dust cloud carry further. It was 
 placed in the larger tube, near the top, with its constricted end 
 12 inches from the coil, and the tube inclined so that the inner end 
 was 14 inch higher than the outer end. This was found advisable 
 to obtain a uniform dust cloud, filling the large tube at the point 
 where the coil was inserted. In place of the piston used by 
 Wheeler a 5-liter bottle, fitted with a mercury manometer, and 
 inlet and outlet tubes, was used. This was filled in each test with 
 air under 2 pounds pressure. 
 
 In determining the ignition-temperature, tests were carried 
 out until temperatures were reached where ignition was obtained 
 at least twice at one temperature, and not obtained twice at a 
 temperature 10° C. (18° F.) lower. It was noted with many 
 dusts that flames of different lengths could be obtained consist- 
 ently at temperatures varying but slightly from those where tlie 
 flame was propagated the entire length of the tube. For instance, 
 with Pittsburgh standard coal dust, the following results were 
 obtained: 
 
 Temp. Deg. C. Nature of Flame. 
 
 1193 Long; out of tube 
 
 1190 Two-thirds length of tube 
 
 1180 About 4 inches long 
 
 1175 About 2 inches long 
 
 1180 About 4 inches long 
 
 1170 About 2 inches long 
 
 1160 About 1.5 inches long 
 
 The ignition-temperature in tests of this kind was taken to 
 be the mean of the temperatures where the flame was propagated 
 over half the length of the tube, and the temperature where it 
 was short; in this case taken as 1185° C. 
 
 As the resistance of the platinum coil changes with use, and 
 the coating upon the coil was not constant, due to disintegration 
 or to the deposition of ash upon it, the tests were run against a 
 standard, and the results referred to it in each case. Since no 
 standard has been established for carbonaceous dusts, other than 
 coal dust, and since a large amount of work done by the Bureau 
 of Mines has shown that Pittsburgh Coal Dust is very uniform 
 
32 
 
 The Explosibility of Grain Dusts 
 
 and that its changes are practically negligible, and since a large 
 quantity of it was available, it was taken as a standard. After 
 determining the ignition temperature of each dust, that of the 
 standard was obtained and the results given referred to the 
 standard value. This value was arbitrarily taken as 1185° C. 
 (2165° F.), since it was the value usually obtained when the coil 
 was in the best condition. If the ignition-temperature of a certain 
 dust was found to be 1005° C, and that of the standard of the 
 same day was 1175° C, by referring these values to the standard 
 temperature for the coal dust (1185° C.) the ignition-temperature 
 of the dust tested would become 1015° C. (1859° F.). This is 
 done also to make the results more comparable. Since this is 
 arbitrary, and since the temperatures are the mean of two tem- 
 peratures, 10° apart, the results are given only to the nearest 
 5° Centigrade. 
 
 In each test three minutes were taken to raise the coil to the 
 desired temperature, and then it was held there for two minutes, 
 so that the conditions would be uniform and consistent in each 
 test. 
 
 The results of the analytical work, as well as the tests to 
 determine the relative ignition-temperature, are given here so 
 that any comparisons that may be desired can be easily drawn : 
 
 Table I 
 
 6 
 
 Kind of Dust. 
 
 Condition. 
 
 6 
 
 3 
 
 o 
 
 m 
 
 
 Protein 
 from N. 
 Factor 6.25 
 
 m 
 
 
 6 
 o 
 
 Ignition 
 Temp. 
 
 'A 
 
 4) 
 
 u 
 
 
 13 
 
 Wheat — Ele- 
 
 As received- 
 Sieved 100m 
 As received- 
 Sieved 100m 
 Sieved 80m.. 
 
 As received. 
 Sieved 100m 
 
 7.15 
 
 8.063 
 
 7.13 
 
 5.82 
 
 5.63 
 
 6.89 
 8 23 
 
 3.15 
 
 2.938 
 
 5.16 
 
 4.60 
 
 5.32 
 
 2.71 
 2 50 
 
 16.63 
 11.86 
 16.79 
 9.17 
 11.61 
 
 18.56 
 14.805 
 
 13.38 
 
 15.125 
 
 13.19 
 
 13.25 
 
 13.38 
 
 11.31 
 12.97 
 
 1 
 15.97 43.72 
 16.775 45.239 
 
 129512363 
 1 
 
 15 
 
 Wheat — Ele- 
 vator, marine 
 
 15.04 
 16.43 
 16.465 
 
 11.93 
 14.905 
 
 42.69 
 49.76 
 47.595 
 
 48.60 
 46.59 
 
 Too lumpy, 
 not form 
 cloud 
 
 17 
 
 Wheat — Bet. 
 elev. and stor- 
 
 1295 
 
 2363 
 
 23 
 
 Flour — Dust 
 
 
 
 
 
 
 
 
 
 
 from top of 
 elev. head 
 
 As received.. 
 
 9.565 
 
 1.80 
 
 0.515 
 
 13.065 
 
 0.755 
 
 74.30 
 
 1235 
 
 2255 
 
 25 
 
 Flour — Dust 
 from purifiers 
 
 As received. 
 
 9.00 
 
 3.51 
 
 2.83 
 
 12.16 
 
 1.71 
 
 70.79 
 
 Too 
 
 moist 
 
 33 
 
 Flour — Fin. 
 
 
 
 
 
 
 
 
 and lumpy 
 
 
 product from 
 pkg. room 
 
 As received. 
 
 9.57 
 
 1.38 
 
 0.20 
 
 12.51 
 
 0.56 
 
 75.78 
 
 1265 2300 
 
Laboratory Investigations 
 
 33 
 
 % Kind of Dust. 
 
 Condition. 
 
 u 
 
 3 
 m 
 
 o 
 
 Fats. 
 
 .a 
 
 « 
 
 ■a 
 
 3 
 u 
 U 
 
 01 g o 
 lit 
 
 o 
 
 Ignition 
 Temp. 
 
 05 
 
 Starch 
 Diffe 
 
 U 
 
 1 
 
 35, Wheat — Dust 
 
 As received. 
 
 7.87 " 
 
 2.655 14.675|16.22 
 
 16.21 42.37 
 
 1200 
 
 2192 
 
 from sidewall 
 
 Sieved 200m 
 
 
 
 
 
 1115 
 
 2039* 
 
 and elevator.... 
 
 
 
 
 
 
 
 
 37, Oats and corn 
 
 As received.. 
 
 8.37 
 
 2.43 
 
 7.27 8.66 
 
 13.26 
 
 60.01 
 
 1000 
 
 1832 
 
 — D u s t from 
 
 Sieved 100m. 
 
 8.29 
 
 2.463 
 
 2.30 8.44 
 
 14.765 
 
 63.74 
 
 995 
 
 1821t 
 
 unloading sta. 
 
 
 
 
 
 
 
 
 
 43, Oats and corn 
 
 
 
 
 
 
 
 
 
 — D u s t from 
 
 As received. 
 
 8.29 
 
 2.30 
 
 4.11 7.78 
 
 12.84 
 
 64.68 
 
 995 
 
 1821 
 
 top of elev 
 
 
 7.38 
 
 
 
 10.22 
 
 13.82 
 
 
 1015 
 
 1859 
 
 47, Oat s — Dust 
 
 As received. 
 
 7.67 
 
 4.80 
 
 6.48 
 
 10.16 
 
 14.70 
 
 56.19 
 
 1010 
 
 1850t 
 
 from feed oats 
 
 Sieved 100m 
 
 
 
 
 
 
 
 
 
 49, White corn — 
 
 
 
 
 
 
 
 " 
 
 
 Dust from 
 
 As received- 
 
 8.87 
 
 7.73 
 
 1.53 
 
 10.28 
 
 2.09 
 
 69.50 
 
 Too moist 
 
 drier room 
 
 53, White corn — 
 
 
 
 
 
 
 
 and lumpy 
 
 Drying of 
 
 As received. 
 
 7.46 
 
 9.53 
 
 1.92 11.43 
 
 2.59 67.09 
 
 Too moist 
 
 germ 
 
 As received- 
 
 6.65 
 
 5.285 
 
 
 4.86 61.225 
 
 and }n>^T'»'v 
 
 57, O at s — Dust 
 
 8.82 
 
 13.16 
 
 1020 
 
 1868 
 
 from ground 
 
 Sieved loom 
 
 7.23 
 
 5.665 
 
 3.90 
 
 14.63 
 
 4.445 64.13 
 
 
 
 oat hulls 
 
 
 
 
 
 61, Yellow corn — 
 
 
 
 
 
 
 
 
 
 
 First break in 
 
 As received.. 
 
 10.06 
 
 0.17 
 
 0.28 
 
 2.595 
 
 0.475 
 
 86.42 
 
 1025 
 
 1877 
 
 dry milling 
 
 
 
 
 
 
 
 
 
 
 63, Yellow cor n As received- 
 
 8.71 
 
 2.87 
 
 1.50 
 
 6.54 
 
 1.15 
 
 79.23 
 
 Too inoist 
 
 dust 
 
 65, Feed — Dust As received.. 
 
 8.40 
 8.925 
 
 1.335 
 
 7.01 
 
 1.19 7.54 
 
 9.00 
 9.96 
 
 
 and lumpy 
 Too coarse 
 
 from dust col- 
 
 Sieved lOOm 
 
 71.05 
 
 1015 1859§ 
 
 lector 
 
 
 
 
 
 
 
 
 67, Yellow corn — 
 
 
 
 
 
 
 
 
 
 Dust from 
 
 As received.. 
 
 8.72 
 
 4.125 
 
 1.525 
 
 7.065 
 
 1.275 
 
 77.29 
 
 Too moist 
 
 drier 
 
 As received.. 
 
 7.50 
 
 9.215 
 
 0.61 
 
 15.065 
 
 1.78 
 
 65.83 
 
 and lumpy 
 Too coarse 
 
 81, Oat groat dust 
 
 
 
 
 
 
 
 
 
 granular 
 
 89, Oats — Ground 
 oat hulls from 
 
 Very coarse 
 As received.. 
 
 6.79 
 
 5.39 
 
 
 14.22 
 
 4.28 
 
 
 1280 2336 
 
 attrition mill.. 
 
 Sieved lOOm 
 
 7.43 
 
 4.82 
 
 0.42 
 
 13.78 
 
 3.13 
 
 70.42 
 
 Not 
 
 97, Reduction mid- 
 
 
 
 
 
 
 
 
 enough 
 
 dliiiiirs 
 
 As received.. 
 
 9.15 
 
 1.12 
 
 0.21 i 12.75 
 
 0.47 
 
 76.30 
 
 1280|2336 
 
 103, Flour — Dust 
 
 
 
 
 
 
 
 
 
 
 from rolls and 
 
 
 
 
 
 
 
 
 
 
 purifiers b e - 
 
 
 
 
 
 
 
 
 
 
 fore reel 
 
 As received.. 
 
 8.40 
 
 1.36 
 
 0.71 
 
 14.12 
 
 0.75 
 
 74.66 
 
 1270 
 
 2318 
 
 
 
 Sieved through 200-mesh screen — *S0%, t81.2%, {81%, §57% of total. 
 
 It should be noted that these results are only relative, but 
 still they can be considered as showing a relation between the 
 inflammability of these different dusts. The results thus far 
 obtained in these tests seem to indicate that the oat and yellow 
 
34 The Explosibility of Grain Dusts 
 
 corn dusts are the most inflammable ; the fine material in the side 
 wall wheat elevator dust being the only other that has an ignition 
 temperature even approaching those for oat and yellom corn 
 dust. Sufficient work has not been done upon white corn dust to 
 warrant any statement, but present indications are that it is not 
 as inflammable as either oat or yellow corn dusts. More complete 
 data, to be obtained in future work, may change these present 
 assumptions, so they are given only as an indication of what may 
 be expected. 
 
 In order to give some idea of the way these tests are carried 
 out, as well as to show how an inflammation caused by such a 
 small source of heat may be propagated, a few photographs of 
 inflammations produced in the English apparatus are given here. 
 
 The point of ignition is within the collar, which shows dark 
 in the photographs. The tube which holds the dust is also shown 
 at the extreme right end of the pictures. The dust is blown from 
 this tube against the heated coil, where it is ignited. It is shown 
 here very conclusively that the inflammation is propagated in 
 both directions from the point of ignition, but that the largest 
 flame travels in the direction in which the dust cloud is originally 
 forced. 
 
 A distinct difference will be noted between the inflammation 
 of the oat hull and yellow corn dusts. The flame of the former 
 is the longer, while that of the latter is more bushy outside the 
 tube. This may be due to the fact that the former was the 
 coarser, and consequently burned for a longer period. Very 
 strong detonation accompanied the ignition of the corn dust, loud 
 enough to attract the attention of persons working in the same 
 building, but at least 100 feet away. The ignition of the former 
 was accompanied by a very slight detonation. 
 
 The two photographs of ignitions of oat and corn dust from 
 unloading station show very well the difference in the propagation 
 of flame at temperatures varying by only 20° C. (36° F.). 
 
 It has been thought by some milling men that certain dusts 
 were not dangerous. These photographs show beyond all ques- 
 tion that these dusts will ignite and propagate a flame under the 
 conditions of the experiment. By similar photographs it could be 
 shown that all the dusts thus far considered will propagate a 
 flame under proper conditions. The work that was first carried 
 out with the French apparatus also confirms this statement. 
 
Laboratory Investigations 
 
 35 
 
 No. 37. Oats and Corn: Dust from Unloading Station, 
 
 Sieved 200-mesh screen 
 
 Temperature 1170° C, '2138°F. 
 
 No. 37. Oats and Corn : Dust from Unloading Station, 
 
 Sieved 200-mesh screen 
 
 Temperature 1150° C, 2102° F. 
 
 5373 
 
 No. 37. Oats and Corn: Dust from Unloadinj); Station, as received. 
 Temperature 1200° C, 2192° F. 
 
36 
 
 The Explosibility of Grain Dusts 
 
 No. 57. Oats : Dust from (jround Oat hulls as received 
 Temperature ll.)0° C, 210^2° F. 
 
 No. 61. Yellow Corn: Dust from first break in Dry Millino;, as received 
 Temperature 1150° C, 2102° F. 
 
Laboratory Investigations 37 
 
 Relative Inflammability 
 
 Considerable work has been carried out in the Bureau of 
 Mines in perfecting a laboratory method for determining the rela- 
 tive inflammability of coal dusts. Dr. J. C. W. Frazer worked 
 upon a method for the determination of the pressure created 
 when the various dusts are ignited at a fixed temperature. A 
 description of the same is given in Bulletin 50 of the Bureau of 
 Mines. This apparatus has been modified by Dr. J. K. Clement, 
 and his associates in the Bureau of Mines. As used at present it 
 consists of an explosion flask of about 1400 c.c. (85.36 cu. in.) 
 capacity, a platinum coil, a device for driving a dust cloud against 
 the coil, and a Crosby pressure-gauge, for measuring the pressure 
 developed. The flask is provided with tubulures at its top and 
 bottom, the ends of which are ground true. The smaller tubulure 
 rests on a circular brass plate, carrying a small brass funnel con- 
 nected into it. A circular brass plate rests on the top of the 
 tubulure and carries the platinum coil, holding it near the center 
 of the flask by means of steel leads, passing through bakolite 
 stoppers. This plate also carries a brass tube, which is connected 
 to the Crosby gauge. The contacts between the tubulures and 
 the brass plates are made tight by rubber washers. The lower 
 plate rests in a groove in a wooden block, carrying two upright 
 rods, having a thread and nut on the upper ends. A long steel 
 plate, cut in at each end and cut out in the center to allow the 
 leads and brass tube of the upper circular brass plate to pass 
 through, rests upon the upper plate and extends over the upright 
 rods. By screwing the nuts down upon this plate the entire 
 apparatus is made gas-tight. The platinum coil is wound on a 
 hardened lavite tube one inch long, grooved, and of capillary bore. 
 The coil is covered with a coating of alundun cement. Through 
 its capillary passes a platinum vs. platinum-rhodium thermo- 
 couple, which is connected to a mille-volt meter, graduated to 
 read degrees Centigrade. The leads holding the coil are con- 
 nected to the terminals of a storage battery through a lamp 
 back, rheostats, and an ammeter, by which the amount of cur- 
 rent can be regulated and determined. 
 
 For each determination 50 m. g. (.00176 oz.) of the dust is 
 weighed into the glass funnel and shaken down as far as possible 
 into the bend in the stem of the funnel, which is connected by 
 rubber tubes to a check valve, and thence to a 150 c. c. glass bulb 
 
38 The Explosibility of Grain Dusts 
 
 used for an air reservoir. This is connected to a mercury mano- 
 meter, and then to a source of compressed air. Air is forced into 
 the small reservoir until it is under a pressure of 190 m. m. of 
 mercury (4.75 lbs.). 
 
 After weighing the sample into the funnel the flask is placed 
 over it, the coil, brass plate and gauge placed in and on the flask, 
 and the whole fastened tight, except for a small opening in the 
 brass tube. The current is turned on and the desired temperature 
 gradually reached in three minutes, and then held there for two 
 minutes to obtain uniform conditions in the flask. At the end of 
 the five minutes the small opening in the brass tube is closed, and 
 a pinch-cock between the check valve and the air reservoir is 
 opened and the dust thus forced in a cloud against the coil. The 
 pressure developed is recorded by the gauge pointer upon a 
 smoked drum and can be measured whenever convenient. 
 
 As was the case in determining the relative ignition-tem- 
 perature, no standard has been determined for carbonaceous 
 dusts other than coal dust, consequently to obviate changes in the 
 coil, and for other reasons given above, each sample was run 
 against Pittsburgh Standard Coal Dust, and all values referred 
 to it as standard. In running these tests a determination was 
 made upon the standard alternately with determinations upon the 
 dust being tested. These results were then averaged, and the 
 values which were obtained referred to the standard pressure for 
 the coal dust, which was taken as the average of the results 
 obtained each day during the tests. This value for 1200° C. was 
 9.0 pounds, for 1100° C. was 7.0 pounds, and for 1000° C. was 0.5 
 pounds. The value given below for 900° C. (0.1 pound) is that 
 obtained in previous work done in the Bureau of Mines. In 
 referring the results obtained on the cereal dusts to the standard 
 values for Pittsburgh Standard Coal dust the following proportion 
 was used : 
 
 Average pressure for cereal dust : Average pressure obtained 
 in test upon standard : : x : standard value for coal dust at 
 temperature of the tests. The value obtained for x is then com- 
 parable to the values obtained for every other dust, irrespective 
 of changes in the coil. 
 
 A certain pressure is always developed within the flask by 
 the inrush of the compressed air, but this is small, 0.3 to 0.4 of a 
 pound, and was corrected for in each case. As indicated in the 
 
Laboratory Investigations 39 
 
 following tables and curves, tests were made upon samples as 
 received, upon that portion of some samples which passed a 200- 
 mesh sieve, and also upon samples dried at 105° C. for three hours 
 in a current of dry air. 
 
 Since in most cases the dusts as received contained consid- 
 erable and variable amounts of coarse material, the results 
 obtained upon these should not be compared. For the same reason 
 it was very difficult to obtain consistent results, for it was not 
 possible to obtain a uniform sample when such a small amount 
 was used. Therefore in the subjoined table a column of high 
 results is given, which indicates what may be expected under 
 certain conditions. 
 
 Before considering the results of these tests it shall be noted 
 that the temperatures which are given are the temperatures on 
 the inside of the coil, and not at the point of ignition. The tem- 
 peratures on the outside of the coil are necessarily lower on 
 account of radiation of heat ; however, this would be constant in 
 each test, since the heating interval is always the same, and 
 therefore the results are comparable. To obtain data relative to 
 the probable temperature on the outside of the coil observations 
 were made upon it with an optical pyrometer. The averages of 
 several tests gave the following: 
 
 Temperatures inside of coil. Temperatures at surface of coil. 
 1200° C. 1075° C. 
 
 1100° C. 990° C. 
 
 1000° C. 915° C. 
 
 While these may not be accurate, due to personal error in 
 reading the optical pyrometer, they show that the outside tem- 
 peratures are considerably lower than the inside temperatures. 
 
40 
 
 The Explosibility of Grain Dusts 
 
 Table II 
 
 ji Kind of Dust 
 
 Condition 
 
 Pres. in lbs. 
 
 at 1200° C. 
 
 (2192° F.) 
 
 Referred to 
 
 Pffh. std. dust 
 
 at 9.0 lbs. 
 
 Aver 
 
 Hisrh'st 
 
 2o 
 
 5° 
 .2^ 
 
 Pres. in lbs. on dried sam- 
 ples ref d to Pgh. std. dust 
 
 1200° C 
 21920F 
 
 1100° C 1 1000° C 
 2012° PI 1832° F 
 
 900° C 
 1652o P 
 
 
 13, Wheat — Ele- 
 vator dust 
 
 As received.. 
 Sieved 100m. 
 Sieved 200m. 
 
 As received- 
 Sieved 200m. 
 
 As received... 
 Sieved 200m. 
 
 As received... 
 As received... 
 
 As received... 
 
 Sieved 200m.. 
 
 As received... 
 
 As received.. 
 
 6.5 
 
 9.2 
 
 8.5 
 9.5 
 
 17, Wheat — Dust 
 
 between top of 
 
 elev. and stor- As received. 
 
 age tanks Sieved 100m. 
 
 23, Flour — Dust 
 
 from top of I 
 
 elev. hds As received 
 
 33, Flour — Fin.l 
 
 product from 
 
 pkg. room [As received 
 
 35, Wheat — Dust 
 
 from sidewall 
 
 of elev 
 
 37, Oats and corn 
 
 — D u s t from 
 
 unloading sta. 
 43, Oats and corn 
 
 — Dust top of 
 
 elevator.... 
 47, O a t s — Dust 
 
 from feed oatsjsieved 200m. 
 49, White corn — 
 
 Dust from 
 
 drier room 
 
 57, O a t s — Dust 
 
 from ground As received. 
 
 oat hulls 
 
 61, Yellow corn — 
 
 Dust from first 
 
 break in dry 
 
 milling 
 
 63, Yellow corn — 
 
 Dust from col- 
 lector 
 
 65, Feed — Dust 
 from dust col- 
 lector Sieved 200m. 
 
 97, Reduction 
 
 middlings As received. 
 
 103, Flour — Dust| 
 from rolls andl 
 purifiers jAs received. 
 
 5.0 
 5.7 
 
 5.8 
 
 7.2 
 
 6.0 7.0 
 
 5.8 
 
 6.8 
 9.9 
 
 5.8 
 
 7.2 
 11.1 
 
 4.5 6.1 
 11.1 I 11.3 
 
 11.1 I 11.7 
 
 6.3 7.6 
 
 12.6 12.8 
 
 0.1 
 
 6.5 I 9.6 
 10.3 I 10.7 
 
 15.1 I 16.5 
 
 6.79 
 
 6.31 
 
 8.73 
 
 8.46 
 
 6.57 
 
 7.35 
 
 8.32 
 
 7.36 
 
 12.6 
 
 9.1 
 
 9.1 
 
 7.8 
 
 13.9 
 
 12.8 
 
 7.41 14.6 
 
 6.72 12.2 
 
 11.3 
 
 15.4 
 
 0.5 I • 1 8.77 I 8.3 
 
 I < I 
 
 I I I 
 
 7.9 I 8.7 I 6.44 I 14.3 
 
 0.2 I 'I 8.73 I 8.9 
 
 I I I 
 
 I t I 
 
 0.4 I * [ 8.72 I 8.2 
 
 10.4 
 
 6.9 
 
 11.6 
 
 12.5 
 
 1.0 t 
 
 9.4 
 
 7.5 
 
 8.2 
 
 4.8 
 
 0.3 
 
 10.4 
 
 0.2 
 
 6.8 
 
 0.2 
 
 0.5 
 
 0.1 
 
 4.9 
 
 14.9 
 
 9.5 
 
 9.7 
 
 8.8 
 
 14.2 
 
 14.0 
 
 16.1 
 
 12.8 
 
 13.2 
 
 8.9 
 
 •Too moist and lumpy. tCoarse; would not form cloud. |Not dried. 
 
Laboratory Investigations 
 
 41 
 
 5 
 
 z 
 
 UJ 
 
 < 
 
 CO 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 16 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 No. 47 
 
 Dust from feed oats 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 12.8 
 
 t1 
 
 - 
 
 - 
 
 
 
 
 
 
 
 
 
 
 
 1 1 
 
 
 
 
 
 
 
 
 
 \~\ — 
 
 ,_ 
 
 _ 
 
 _ 
 
 
 
 _ 
 
 _. 
 
 .- 
 
 ._ 
 
 
 
 
 
 
 
 
 
 
 
 10 
 
 .f 
 
 .- 
 
 
 ~ 
 
 — 
 
 "~ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 8 
 
 
 
 
 
 
 
 
 
 ,'' 
 
 ' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 ' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^' 
 
 ' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -I 
 
 
 ^y 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 le 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 16 
 
 
 
 
 
 No. 61 Yellow corn, dust from Orst break in dry milling: 
 
 
 
 
 
 
 
 
 .1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 94 
 
 
 
 
 , 
 
 ^ 
 
 -- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 K- 
 
 ■ — 
 
 
 
 
 
 
 
 
 
 
 
 
 8 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ''' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 y 
 
 ^' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ,5 
 
 
 ^^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 L-^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 10 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 89 
 
 
 
 
 
 No. 97 Reduction middlings 
 
 
 
 
 
 
 
 7.S 
 
 
 
 
 ^ 
 
 ._ 
 
 - 
 
 -• 
 
 — 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 5.8 
 
 
 
 
 
 
 
 
 ^. 
 
 
 t- 
 
 
 " 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 - 
 
 -- 
 
 — 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 4. 
 
 U 
 
 
 
 
 .- 
 
 — ' 
 
 -- 
 
 — 
 
 —" 
 
 r 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 6 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 9 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ,^ 
 
 -■ 
 
 --■ 
 
 
 
 
 
 
 
 Pittsburgh stand 
 
 ard coal 
 
 dust 
 
 
 
 
 
 
 
 
 ._ 
 
 — 
 
 — 
 
 " 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 y 
 
 "' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ,x 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 y 
 
 • 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 -' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 .r 
 
 
 
 
 
 
 
 
 
 * 
 
 /'' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 _. 
 
 -- 
 
 -■ 
 
 -- 
 
 - 
 
 .... 
 
 — 
 
 -r 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CENT.900 
 FAHR.I652 
 
 1000 
 1832 
 
 MOO 
 2012 
 
 1200 
 2192 
 
 TEMPERATURE —DEG. 
 
42 
 
 The Explosibility of Grain Dusts 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 lb 
 
 
 No. 13 Wheat elevator dust 
 
 
 
 
 
 
 
 
 
 
 
 12 
 
 6 
 
 — 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 10.4 
 
 
 
 
 
 
 '■\ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -■ 
 
 -- 
 
 ~ ' 
 
 
 
 
 
 
 
 8 
 
 
 
 
 
 
 
 
 "r- 
 
 
 -- 
 
 — ' 
 
 " 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 ^ 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ ' 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ,0 
 
 
 ^ ' 
 
 " 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 y 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 16 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 No. 33 Flour, finished product from packing room 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 8 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -6 ^- 
 
 
 
 
 
 
 
 7.8 
 
 
 
 
 
 
 
 
 
 4 
 
 8- 
 
 
 
 
 
 
 A 
 
 
 n 
 
 — 
 
 — 
 
 — 
 
 "" " 
 
 ■ ■" 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 • — 
 
 — ' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 ^ 
 
 ,- 
 
 ■-•' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 '"' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 — 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 IB 
 
 No. 35 Wheat, side wall of elevator 
 
 
 
 
 
 
 
 
 1 13.9 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 11.6 
 
 
 
 . _ 
 
 -- 
 
 ■- 
 
 — 
 
 * "* 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 y' 
 
 " 
 
 
 
 
 
 
 
 
 
 
 8 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 /" 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 • 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 . 
 
 ." 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 T 
 
 y 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 16 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Pittsburgh standard coal dust 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ( 
 
 ) 
 
 — 
 
 8 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 1 
 
 
 
 
 
 — - - 
 
 .- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 <r- 
 
 -- 
 
 *"" 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 • 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 
 
 
 
 
 1 
 
 
 ^' 
 
 '" 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ._ 
 
 _ 
 
 
 
 
 
 
 
 __ 
 
 .5 ^ 
 
 -f 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 CENT. 900 
 FAHR. 1652 
 
 1000 noo 
 
 1832 2012 
 
 TEMPERATUr.I- DEC. 
 
 1200 
 2192 
 
Laboratory Investigations 
 
 43 
 
 2! o 
 
 > m 
 
 X z 
 
 30 H 
 
 PRESSURE -POUNDS PER SQUARE INCH 
 
 s g 
 
 
 
 
 10 
 
 
 
 4 
 
 Sk 
 
 
 
 a> 
 
 
 
 o 
 
 
 
 o 
 
 
 
 to 
 
 
 A 
 
 
 
 a 
 
 S§ 
 
 "^ 
 
 i. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 — 
 
 > 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 N 
 
 
 
 
 
 
 I 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 S 
 
 > 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ■0 
 
 
 \ 
 
 \ 
 
 N 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i 
 
 
 \ 
 
 
 ,^ 
 
 N 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a 
 
 c 
 
 
 
 V 
 
 % 
 
 s 
 
 s 
 
 s 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 .i. 
 
 
 
 \ 
 
 \\'i 
 
 
 S 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 -<v 
 
 
 \ 
 
 \ 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^< 
 
 k - 
 
 
 \ 
 
 
 
 N 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ,8! 
 
 
 
 
 
 \ 
 
 
 
 
 s 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 S 
 s. 
 
 a 
 
 
 
 
 
 
 C^ 
 
 v 
 
 
 
 s 
 
 \ 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 i 
 
 
 
 
 \ 
 
 \ 
 
 
 
 s 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 V= 
 
 ^ 
 
 
 , 
 
 
 \ 
 
 s 
 
 
 
 
 N 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ft 
 
 
 
 
 
 
 
 \ 
 
 
 k 
 
 
 S 
 
 s 
 
 
 
 
 N 
 
 
 
 
 
 
 
 
 
 
 
 
 ?^ 9 
 
 
 
 
 
 
 
 
 
 
 V 
 
 > 
 
 
 
 S 
 
 
 
 
 
 S 
 
 
 
 
 
 
 
 
 
 
 
 ra o 
 
 ^ 
 
 ^ 
 
 
 
 
 
 
 
 
 \ 
 
 -1 
 
 o 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 ^ 
 
 
 
 
 
 
 
 \ 
 
 kl 
 
 1 
 
 ^ 
 
 
 
 
 
 
 
 
 
 ,0 
 
 
 
 
 
 
 
 
 
 
 
 
 > 
 
 ^" 
 
 s 
 
 
 % 
 
 
 
 
 \^ 
 
 '• 
 
 
 
 
 u 
 
 • 
 
 
 
 
 >> 
 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 ^ 
 
 \ 
 
 h1 
 
 
 
 \ 
 
 i 
 
 
 
 
 \< 
 
 
 
 
 \- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 \ 
 
 S 
 
 L'T. 
 
 \ 
 
 \l 
 
 
 
 
 \ 
 
 ^ 
 
 
 
 
 \ 
 
 [' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 \ 
 
 , > 
 
 n\. 
 
 A \> 
 
 
 
 
 \ 
 
 », 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \, 
 
 s 
 
 o 
 s < 
 
 i> 
 
 s 
 
 \^ 
 
 \^ 
 
 
 
 
 
 \% 
 
 1 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 s. 
 
 s 
 
 ^i 
 
 
 ^ 
 
 
 
 
 1 
 
 V 
 
 
 
 
 » 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 
 A 
 
 \ 
 
 
 
 
 
 
 \ 
 
 \ 
 
 
 
 
 \"a| 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 \ 
 
 
 
 */ 
 
 ?'* 
 
 
 
 \ 
 
 ifr 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 nV 
 
 \' 
 
 
 
 
 
 'w 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \, 
 
 
 
 
 iS 
 
 Vj 
 
 
 \ 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 lv< 
 
 >/. 
 
 
 \ 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 V 
 
 \ 
 
 
 \ 
 
 \ 
 
 ht 
 
 ^ 
 
 
 L e 
 
 fr 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 \ 
 
 \ 
 
 
 s 
 
 s** 
 
 ">>. 
 
 Y' 
 
 *'o, 
 
 
 \ 
 
 
 
 
 
 
 
 M — 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 K 
 
 \ 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 Sy 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 \ 
 
 
 
 
 ^ 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 \ 
 
 
 
 
 
 \ 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 1 \ 
 
 
 
 
 
 
 M 
 
 
 
 y 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 y 
 
 
 
 
 
 
 
 K* 
 
 r. 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 s:-^ 
 
 A 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 VM 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 \, 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 w 
 
 
 
 
 
 
 
 
 1 
 
 y 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 \.^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 In 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 \ 
 
 % 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 \ 
 
 
 <: 
 
 -> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 I 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 \ 
 
 \ 
 
 ^' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 \ 
 
 
 \ 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 \ 
 
 
 
 
 
 
 1 
 
 
 \ 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 1 
 
44 
 
 The Explosibility of Grain Dusts 
 
 16 
 
 
 
 14 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 a Screened through 200 mesh screen 
 
 
 
 
 
 
 
 
 b Screened through 100 mesh screen 
 
 12 
 
 ■o 
 
 
 
 
 
 
 
 
 10 
 
 'Z 
 
 
 
 
 
 
 
 
 
 ~'Z 
 
 
 
 
 
 
 
 i 
 
 
 c ■? 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 iS a 
 
 
 c 
 
 
 
 
 
 
 
 
 
 
 ♦- •- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a; CI 
 
 
 M 
 
 
 
 
 
 
 
 
 
 
 
 ui bo 
 
 
 c 
 
 
 
 
 
 
 
 
 
 
 
 a a 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ t- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 c 
 
 c 
 - o- 
 
 
 
 
 
 
 
 -|-|-f- 
 
 8 
 
 
 
 c 
 
 ~° 
 
 
 
 
 
 
 
 
 
 •D 
 
 C 
 
 
 
 O 
 > 
 
 bo 
 
 n 
 to 
 
 
 
 
 
 
 
 
 •8 
 
 .£ 
 
 
 
 
 
 
 c 
 
 c 
 
 
 
 
 
 
 
 
 X 
 
 
 
 J 
 
 
 
 'o 
 
 o 
 
 ~ c 
 
 o 
 
 c 
 
 3 
 X 
 
 3 
 
 - o 
 
 ~o 
 
 
 
 
 
 
 
 of elev 
 of elev 
 om pai 
 
 e 
 
 
 
 ~ o 
 
 
 
 OJ 
 
 
 1 
 
 w 
 
 ^ 
 
 £ 
 
 3 
 
 E 
 o 
 
 E 
 2 
 
 o 
 
 ■o 
 c 
 
 3 
 
 o 
 
 O 
 
 ■u 
 c 
 
 3 
 O 
 
 > 
 
 > 
 
 0) 
 
 
 
 
 
 '0 
 
 CL 
 
 a 
 
 
 c 
 
 a 
 
 
 c 
 
 3 
 
 •D 
 
 
 
 § 
 
 
 
 D 
 
 3 
 
 3 
 
 tie 
 
 tic 
 
 o 
 
 o 
 
 3 
 ■D 
 
 3 
 
 T3 
 
 -D 
 
 E 
 - 
 
 
 
 1 
 
 1 
 
 
 
 a 
 
 4 
 
 T3 
 
 _T3 
 
 _TJ 
 
 _T3 
 
 _T3 
 
 C 
 1- 
 
 C 
 
 _E 
 
 o 
 
 _£ 
 o 
 
 u 
 
 - > 
 
 ~? 
 
 ~ O 
 
 ~ 
 
 - E 
 
 
 ^ 
 
 ^ 
 
 -T3 
 
 2 
 
 C 
 
 o 
 o 
 
 E 
 o 
 
 3 
 
 E 
 
 
 
 t- 
 
 3 
 
 _Q 
 
 o 
 
 C 
 
 -o 
 
 o 
 
 ■□ 
 c 
 « 
 
 -O 
 
 o 
 
 C 
 
 -O 
 
 3 
 
 •a 
 -O 
 
 3 
 
 -O 
 
 15 
 
 -1 
 
 > 
 
 > 
 
 -i 
 
 3 
 -U. 
 
 3 
 
 
 — li. 
 
 3 
 
 ■D 
 
 1 
 
 3 
 n 
 
 X 
 
 5 
 
 "c 
 
 3 
 
 
 
 -li. 
 
 
 h- 
 
 N 
 
 CO 
 
 (~ 
 
 r^ 
 
 r~ 
 
 h- 
 
 lO 
 
 lO 
 
 CO 
 
 CO 
 
 10 
 
 CO 
 
 r~ 
 
 r~ 
 
 CO 
 
 
 o 
 
 ^ 
 
 t 
 
 t 
 
 P3 
 
 CO 
 
 lO 
 
 10 
 
 CO 
 
 CO 
 
 
 
 CD 
 
 C4 
 
 
 ~ 
 
 CO 
 
 
 o 
 
 6 
 
 d 
 
 6 
 
 <5 
 
 d 
 
 6 
 
 6 
 
 6 
 
 6 
 
 6 
 
 6 
 
 6 
 
 6 
 
 6 
 
 z 
 
 
 
 6 
 
 
 z 
 
 z 
 
 Z 
 
 Z 
 
 z 
 
 Z 
 
 z 
 
 z 
 
 Z 
 
 z 
 
 Z 
 
 Z 
 
 Z 
 
 z 
 
 Z 
 
 z 
 
 
 
 
 
 
 <o 
 
 
 "" 
 
 
 T. 
 
 
 °= 
 
 
 ns 
 
 
 ^ 
 
 
 
 ORIGINAL SAMPLES AT 1200 C. 
 2192" F. 
 
Laboratory Investigations 
 
 45 
 
 PRESSURE -POUNDS PER SQUARE INCH 
 
 r • 1 ' 
 
 No. 61 Yellow corn, dust from ftfst break in dry; milling- not dried. 
 
 No 43 Oat and corn, dust from t op of elevator 
 
 I I 1 
 
 a No. 66 Feed, dust from dust collector 
 
 a No. 35 Wheat, side wall of elevator 
 
 > m 
 
 X Z 
 
 X -^ 
 
 i; K> 
 
 CD O 
 
 >o o 
 
 a No. 37 Oat and corn dtist from unloading station 
 
 a No. 13 Wheat elevator dust 
 
 \ ^ 
 
 a No. 47 Dust from feed oats 
 
 a No. 57 Oats, dust from ground oat hulls 
 
 b No. 17 Wheat dust between top of elevator and 
 
 No. 23 Flour dust from top of elevator heads 
 
 No. 97 Reduction middlings 
 
 No. 63 Yellow corn dust from dust 
 
 collector 
 
 storage tank 
 
 No. 103 Flour dust from rolls and purifier before reel 
 
 No. 33 Flour, finished product from packing room 
 
 o o 
 5 o 
 
 No. 49 White corn dust from drier room 
 a No. 47 
 
 a No. 35 
 
 a No. 13 
 
 No. 61 -not dried. 
 
 No. 97 
 
 No. 33 
 
 a No. 47 
 
 a No. 13 
 
 00 o 
 
 u o 
 
 M O 
 
 I 
 
 No. 97 
 
 No. 33 
 
 No. 35 
 
 No. 61 '-not dried. 
 
 No.- 9 7 
 
 No. 33 
 No. 13 
 No. 47 
 Pittsburgh standard coal dust \20O 
 
 Pittsburgh standard coal dust 1 1 OCT 
 
 Pittsburgh standard coal dust 1000 
 
 1 1 i 
 
 Pittsburgh standard coal dust 900^ 
 
 tr to 
 CO CO 
 
 s ? 
 
 8 S 
 
46 The Explosibility of Grain Dusts 
 
 In making comparisons of the results given in the above 
 table and curves it should be remembered that they are only rela- 
 tive, that they simply indicate the relative inflammability of the 
 dusts. Although these values show that certain dusts are highly 
 inflammable, igniting at low temperatures, and developing enor- 
 mous pressures, it is not possible to state that the dusts which 
 produce only 0.1-0.5 pound pressure at 1200° C. are not danger- 
 ous. It is very probable that at higher temperatures, which 
 might easily be obtained from a naked flame, an electric spark, 
 or a flash from a blown-out fuse, or from a motor, these dusts 
 might be ignited and propagate a flame with explosive violence. 
 
 These results substantiate in an astonishing way the suppo- 
 sition advanced from the determination of the ignition-tempera- 
 tures of the dust, namely, that the yeflow corn and oat dusts are 
 the more inflammable. This is very plainly indicated in the 
 straight line curves of figures V and VI. However, from the 
 amount of work that has been done thus far, this supposition 
 cannot be stated as a fact, although indications are that this is 
 true. It is a fact, though, that the highest pressure, as well as 
 the lowest ignition-temperatures that have been obtained in the 
 tests made thus far, have been given by oat and yellow corn dusts. 
 
 On the contrary, the results obtained upon the sieved and 
 dried samples show that the wheat and elevator dusts are also 
 very inflammable, nearly as high pressures being developed by 
 these dusts under the condition of the experiments as by the 
 oat and corn dusts. If the pressures developed in these tests are 
 considered, it will be noted that the two giving the highest pres- 
 sures are from oats and corn, the third and fifth are wheat 
 elevator dusts, and the fourth is an oat and corn dust. These 
 values become more astonishing when we note the percentage of 
 ash in the dusts. The oat and corn dusts have from 10 to 15 per 
 cent, ash, while the wheat dusts have over 16 per cent. It is not 
 possible to state what eflfect a high ash content has upon the 
 inflammability of grain dusts, for no work has been done upon 
 this ; but from results obtained by the Bureau of Mines upon 
 coal dusts it is safe to state that the ash content and its compo- 
 sition are large factors affecting the inflammability of these dusts, 
 as they are in coal dust. If this is true it is probable that with 
 lower ash content these dusts would be much more inflammable, 
 but it cannot be assumed that the wheat elevator dusts would 
 develop as much pressures as the oat and corn dusts if they had 
 
Laboratory Investigations 47 
 
 the same percentage of ash. The astonishing fact is that these 
 dusts with such high percentage ash will develop such pressures 
 under the conditions of the experiments. 
 
 The predominating factor which determines the inflamma- 
 bility of a dust has not been determined. Many theories have 
 been advanced, such as the amount of volatile matter ; this 
 together with the moisture and ash ; the rate or ease of oxidation 
 and the degree of fineness of the dust. It is probable that all 
 these have an important bearing on the question, and should be 
 thoroughly investigated. A discussion of these theories is not 
 within the scope of this paper. However, there is one question 
 which should be considered, that of moisture. 
 
 It is well known that the humidity of the air has a consid- 
 erable effect upon the moisture content of the dusts which are 
 exposed to changes in atmospheric conditions. But the effect that 
 this has upon the inflammabilty of the dusts is not known. A 
 slight indication of what might be expected may be observed in 
 comparing the pressures developed in the above tests by the dried 
 and undried samples. It will be noted that there was a very 
 appreciable increase in pressure deeloped by the dried over the 
 undried samples, except in one case — sample 47, dust from feed 
 oats. Here a slight decrease is noticed in the average, but this 
 may be explained by the fact that fewer tests were made, and 
 consequently this is not as fair an average. Especial attention 
 should be called to samples 63, 97 and 103. It will be seen that 
 the pressures developed by these dusts increased, after drying, 
 from 0.5 pound or less to over 8.0 pounds. This can be very 
 largely attributed to the difference in moisture content, although 
 not entirely, for it is quite probable that if a cloud of the original 
 dust, undried, could have been obtained, it would have developed 
 considerably more pressure. This raises the question of how 
 much moisture would have to be removed from such a dust 
 before it could be dispersed in the form of a cloud. The whole 
 question of the effect of moisture upon the inflammability of these 
 dusts is one which is in great need of thorough investigation. 
 Work along this line has been outlined, and will be taken up as 
 soon as possible. 
 
 It will be observed that the highest pressure (16.5 pounds 
 per square inch) developed by any one dust was developed by 
 sample 61, Yellow Corn Dust from the first break in Dry Milling. 
 This was obtained upon the undried sample. What pressure would 
 
48 
 
 The Explosibility of Grain Dusts 
 
 be developed by this dust if it was dried cannot be surmised, but 
 it seems likely that it would be considerably higher, for the mois- 
 ture content of 10.06 per cent, is higher than that of any other 
 dust that has been tested. Determinations upon this sample, 
 dried, were not made, because the limit of the apparatus was 
 reached with the undried sample. 
 
 To give a better idea of the way the dusts ignite in these 
 tests the following photographs of dusts ignited at 1200° C. 
 (2192° F.) are given, with the pressures developed in each inflam- 
 mation that is shown : 
 
 5379 
 
 No. 33. Flour: Finished Product froniPackin/j Room, Dried. 
 Pressure 9.-t pounds 
 
Laboratory Investigations 
 
 49 
 
 5377 
 
 No. 35. Wheat: Dust from Sidewalls of Elevator, Dried. 
 Pressure 11.6 pounds 
 
50 
 
 The Explosibility of Grain Dusts 
 
 5378 
 
 No. 37. Oats and Corn : Dust from Unloading Station, Dried. 
 Pressure I'J.I ])ounds 
 
Laboratory Investigations 
 
 51 
 
 5380 
 
 No. 57. Oats: Dust from Ground Oat Hulls, Dried. 
 Pressure 11.1 pounds 
 
52 
 
 The Explosibility of Grain Dusts 
 
 5376 
 
 No. 61. Yellow Corn: Dust from First Break in Dry Milling, Undried. 
 
 Pressure 10.1 pounds 
 
Amount of Dust that will Propagate an Explosion 53 
 
 These show beyond any reasonable doubt that the wheat 
 dusts, as well as the others tested, are highly inflammable under 
 the conditions of these tests, although they were not as easily 
 ignited in the tests made to determine the ignition temperatures 
 of the dusts. The pressures developed in these tests are not as 
 high in the last four as were developed in the original tests made 
 upon these samples, but are not much lower except in the last 
 case. This can be explained by the fact that after the test was 
 made it was noticed that the valve in the brass tube leading to 
 the pressure gauge was not closed. 
 
 Although sufficient work has not been done to allow of any 
 absolute statements, the results thus far indicate that all dusts 
 that are made in the handling and working up of grain into food 
 products can be ignited under proper conditions, and also will 
 propagate a flame, most of them with explosive violence. This 
 statement should not be taken as meaning that the dusts will 
 ignite of themselves, that is, spontaneously ; but when heated to 
 their ignition-temperature will ignite and will propagate a flame. 
 In other words, there must be some outside source of heat. This 
 may be very small, such as a heated coil of wire, as used in the 
 above tests, if the temperature is sufficiently high ; or it may be 
 larger, as a flame, which may have a lower temperature but a 
 larger heating surface. 
 
 If it is true that all these dusts will ignite, as it doubtless is, 
 the question of the amount of dust that must be present before 
 a flame can be propagated becomes very important. In the tests 
 which gave the pressures indicated above, 0.05 gram (.00176 oz.) 
 was put in suspension in 1400 c. c. (85.36 cu. in.) of air. To 
 obtain the same proportion of dust and air, and therefore a 
 mixture as inflammable as was used in the tests, it would be 
 necessary to have only 10 pounds of the dust in a closed room 
 containing 4,466 cubic feet, or a room 10 x 30 x 15 feet, or 100 
 pounds in a room 30 x 15 x 100 feet. This mixture certainly is 
 dangerous, and it is very probable that larger amounts of the 
 dust in the same space would be more dangerous. This higher 
 amount is not as important as a knowledge of the lower limit. 
 
 AMOUNT OF DUST THAT WILL PROPAGATE 
 AN EXPLOSION 
 
 Experiments at the Pittsburgh testing station of the Bureau 
 of Mines have shown that an explosion could be produced when 
 
54 The Explosibility of Grain Dusts 
 
 there was only .032 ounce (thirty-two one thousandths — 32/1000) 
 of coal dust suspended in each cubic foot of air, or one pound in 
 500 cubic feet of air. In order to produce complete combustion 
 it takes all of the oxygen in one cubic foot of air to completely 
 burn 0.123 ounce (one hundred twenty-three thousandths) of the 
 dust used. This dust was sufficiently fine to pass through a 200- 
 mesh sieve (one with 200 openings in the length of an inch) and 
 floated easily on a strong air current.* From these tests we 
 might form a preliminary conclusion, viz., that when the mixture 
 contains between .032 and .123 ounces per cubic foot it forms 
 an explosive mixture, and one that is extremely dangerous. We 
 might term these proportions the lower and higher explosive 
 limits of the explosive coal dust mixture. 
 
 In the experiments of M, J. Taffanel at the Lieven experi- 
 ment station in France, in one instance as low a weight as .023 
 ounce of coal dust per cubic foot of space was sufficient to produce 
 an ignition.' At the German station at Derne, an ignition was 
 produced when .040 ounce of coal dust was suspended in one 
 cubic foot of air.t 
 
 These figures, as given, apply to coal dust tests, and indicate 
 that a very small amount of dust in suspension is sufficient to 
 produce the original ignition that will propagate to a very disas- 
 trous explosion. Preliminary experiments already conducted 
 indicate that many of the grain dusts have relatively a lower 
 ignition-temperature than many kinds of coal dust ; also upon 
 ignition the grain dusts give higher pressures at lower tempera- 
 tures than some of the coal dusts. This would seem to indicate 
 the possibility of securing an ignition of dust of this nature, 
 with a smaller proportion per cubic foot than is necessary for 
 coal dust. This, however, has not been definitely determined, and 
 will be a matter for future experiment. 
 
 Propagation of Dust Explosions 
 
 The investigations already conducted in connection with pre- 
 vious explosions indicate that usually two reports are heard by 
 the men who have survived the ordeal and have been able to relate 
 
 *Georg-e S. Rice, Bureau of Mines Circular No. 3 — Coal-Dust Explosions. 
 
 tExplosibility of Coal Dust. George S. Rice et al. U. S. B. M. Bulletin 
 No. 20. 
 
 $Colliery Engineer, April 1, 1914, Vol. 24, No. 9. German Coal Dust 
 Experiments. 
 
Measuring Dust in the Air 55 
 
 their experiences. The first report is described as a sharp, quick 
 sound, followed by a second of a loud, rumbling nature, and 
 lasting for a much longer period than the first report. The second 
 report is usually followed by fire, destroying the plant and sur- 
 rounding property. An explosive mixture, consisting of a very 
 small quantity of fine dust in suspension, ignited by sufficient 
 temperature would, no doubt, cause the sharp report usually 
 heard first. This original ignition, possibly only an infiammation, 
 would produce sufficient concussion to disturb the dust that is 
 settled and packed on surrounding ledges and projections, and 
 shake this fine dust into the air, making an additional explosive 
 mixture. The heat, or flame from the original small puff, or 
 inflammation, would cause an ignition of this newly formed mix- 
 ture, and the explosion would propagate throughout a very large 
 area, until the entire dust zone would be covered. This would 
 probably account for the loud, rumbling sound of long duration, 
 accompanied by a large body of flame. 
 
 This establishes an important relation between the dust in a 
 settled or packed condition and the amount of dust in suspension 
 that is necessary to originate the explosion. Many theories and 
 ideas have been advanced as to the conditions under which dust 
 explosions are produced and the amount of dust in suspension 
 necessary to originate the explosion, all probably based on differ- 
 ent tests and experiments. It is generally agreed that the dust 
 must be fine and dry, and in a state of suspension in the atmos- 
 phere, which upon being brought in contact with sufficient heat 
 or flame, causes an ignition. It is conceded that there must be a 
 proper proportion in diffusion so that the explosive mixture of 
 dust and air will ignite with sufficient force to propagate to an 
 explosion. 
 
 MEASURING DUST IN THE AIR 
 
 It is especially important to devise some accurate manner in 
 which the dust content of the air can be accurately determined. 
 Steps have already been taken since this study began, to arrange 
 an apparatus whereby dust samples could be collected from the 
 air in mills and elevators, and the content per cubic foot, deter- 
 mined. In this manner the efficiency of dust-collecting devices, 
 sweeping systems, etc., can be determined by the analysis of the 
 dust samples collected throughout the plant. When experiments 
 have shown the amount of dust required to produce the original 
 
56 The Explosibility of Grain Dusts 
 
 ignition, a system of this kind will detect the presence of a 
 dangerous dust mixture, and show the quantity of dust in 
 suspension. 
 
 CAUSES OF GRAIN-DUST EXPLOSIONS 
 
 The following causes have been assigned to many of the 
 explosions in milling plants throughout this country and abroad: 
 
 (1) Use of open lights, or naked flames, such as lamps, 
 torches, gas jets, lanterns, candles, matches, etc. 
 
 (2) Property fires. 
 
 (3) Introduction of foreign material in grinding machines. 
 
 (4) Electric sparks from motors, fuses, switches, lighting 
 systems. 
 
 (5) Static electricity produced by friction of pulleys and 
 belts, grinding machines, etc. 
 
 Use of Naked Lights 
 
 At the present time in modern milling plants the use of an 
 open flame is positively prohibited, and it is generally regarded as 
 an element of very great danger to introduce an open flame into 
 a dusty atmosphere. Many explosions in milling plants have 
 been traced to this source, and it is a practice that cannot be too 
 harshly condemned, and should be strongly prohibited. The use 
 of gas jets for lighting purposes, the introduction of lanterns or 
 open lights into grain bins or dust-collecting systems, and similar 
 practices, are very dangerous, and should be discontinued at once. 
 
 Property Fires 
 
 Many violent explosions have occurred during mill fires, due 
 to the presence of dust in large quantities. The flames and force 
 from a fire produce sufficient concussion to jar the dust into sus- 
 pension and initiate a very violent explosion. A plant where dust 
 is allowed to accumulate in large quantities, and where steps are 
 not taken to remove same as often as necessary, is always in 
 danger of fires and explosions. 
 
 Introduction of Foreign Material Into Grinding Machines 
 
 A large number of explosions in more recent years have been 
 traced to the introduction of foreign materials into grinding 
 machines. Occurrences of this nature have been especially fre- 
 quent in the grinding of oat hulls and feeds. Particles of foreign 
 
Prevention of Grain-Dust Explosions 57 
 
 material seem to pass the separating systems and, coming in 
 contact with the grinding plates of the machines, produce suf- 
 ficient sparks to cause an ignition of the dusts in the grinding 
 machines and conveyor lines. 
 
 Electric Sparks from Motors, Fuses, Etc. 
 
 Explosions have been assigned to the ignition of the dust 
 cloud by an electric arc, and by sparks from motors, blown fuses, 
 switchboards, starting boxes, lighting systems, etc. It has been 
 thought that if a trip of mine cars should run away on a dusty 
 road and break the trolley or feed wires, the electric arc produced 
 may be sufficient to cause an ignition of the dust stirred up by 
 the wrecking of the trip. As already stated, a disastrous explo- 
 sion in Liverpool, England, in 1911, was due to the ignition of 
 dust stirred up by the breaking of a belt. The cause of the 
 ignition was assigned to sparks from a blown fuse of a temporary 
 switchboard. 
 
 Static Electricity 
 
 The production of static electricity by friction of pulleys and 
 belts has been assigned as the cause of recent dust explosions. 
 Although experiments have not been conducted along this line to 
 show that a dust cloud can be ignited in this manner, a recent 
 experiment at Pittsburgh showed very clearly that sufficient 
 static electricity could be produced by a very small pulley and 
 shaft, to readily ignite gas. A Milling Company in Texas, engaged 
 in grinding cottonseed cake into meal, states, that after experi- 
 encing a series of explosions, the insulating of a certain grinding 
 machine prevented any repetition of previous occurrences. The 
 fact that explosions have been known to occur at times when the 
 feed of grinding machines was cut off, seems to indicate that an 
 unknown factor may be the responsible agent. A series of experi- 
 ments are being planned in order to determine if static electricity, 
 produced in this manner, can produce an ignition of the dust 
 while in suspension. 
 
 PREVENTION OF GRAIN-DUST EXPLOSIONS 
 
 Presence of Foreign Materials in Grain 
 
 Since only a very small quantity of dust in suspension is nec- 
 essary to present conditions favorable to ignition, it would appear 
 advisable that the proper thing to do would be to avoid the pro- 
 duction or escape of dust into the atmosphere, as far as this is 
 
58 The Explosibility of Grain Dusts 
 
 possible. From the large number of explosions thought to have 
 been due to the presence of foreign material in the grain, it 
 appears that the grain contains a portion of this material from 
 the original point of shipment. During the transportation by 
 boats, cars, etc., and also storage in the elevators, the quantity 
 of foreign material is no doubt increased and suggests the impor- 
 tance of cleaning the grain at the very first stages of handling. 
 Much of the foreign material that causes trouble later on seems 
 to originate in loading grain from the field. The quantity of 
 foreign material present may be reduced by a cleaning process 
 at the point of origin, and steps then taken to prevent the addi- 
 tion of any further quantity of foreign material during trans- 
 portation and handling. It appears to be a commercial problem, 
 but should command consideration. 
 
 Size of Receiving Bins 
 
 In addition to the source of danger presented by hard foreign 
 materials striking the plates of grinding machines, the size of 
 the bins receiving ground material has an important relation to 
 the extent of the fire or the violence of the explosion. If the bin 
 is of large dimensions and very deep, it gives a very large area 
 that may become filled with very fine dust in suspension. A 
 number of violent explosions have occurred, due to a flame coming 
 into contact with the suspended dust in bins containing only a 
 small quantity of grain. 
 
 Use of Portable Electric Lamps 
 
 It is necessary in mills and elevators, for the workmen to 
 determine at frequent intervals the amount of grain that the 
 storage bins contain. A common practice is to lower a light 
 of some kind into the bin, to observe or measure the quantity of 
 grain. Many explosions have occurred when open lights and 
 lanterns were introduced into grain bins for this purpose, and the 
 practice cannot be too strongly condemned. The relation of the 
 electric spark to the ignition of the dust cloud has not been fully 
 determined by experiment, and many companies, for this reason, 
 have discontinued the lowering of incandescent electric light bulbs 
 into dusty atmospheres. There is a tendency for the workmen to 
 become hasty in an effort to ascertain the quantity in series of 
 bins, and the bulb may, by contact with the side of the bin or 
 floor, become broken and introduce an element of possible danger. 
 
Prevention of Grain-Dust Explosions 59 
 
 The desired result can be obtained by lowering a "tape" with a 
 weight attached to the end, and the exact measurement can be 
 recorded. 
 
 When occasion renders it absolutely necessary to use a lamp 
 of some kind in a bin an approved type of portable electric lamp 
 could be used. The Electrical Section of the Bureau of Mines has 
 recently approved three different types of lamps for safety in 
 gaseous mixtures. Their value from a commercial standpoint as 
 regards economy and efficiency still remains to be ascertained. 
 Portable electric lamps are still in the early stages of perfection, 
 and new types of lamps are being introduced on the market from 
 time to time. 
 
 Electric bulbs in dusty atmospheres located near machinery 
 where there is a possibility of the lamp becoming broken, or at 
 points in the mill where workmen may strike the lamp, especially 
 when carrying a projection of some kind on their shoulder, should 
 be enclosed in strong wire guards or protectors ; and it would be 
 advisable also to enclose each bulb in a vapor-proof globe. An 
 extra safety feature would be, whenever possible, to locate all 
 fuses on light and power circuits, switches, starting boxes, 
 motors, etc., at points where dust is not present in dangerous 
 quantities. 
 
PRESS OF THE 
 
 KEYSTONE Printing Company 
 
 OF PITTSBURGH 
 
• ■qJ' 
 
 
 
 
 
 >- ./\>;^^\ oo\^;:>o ./\i.^/\ ^^^ 
 
 
 ,/ -O^^^^^O^ V^\^^' ^O.*^^/ ^/-- 
 
 
 
 
 
 
 ri^^ »■ 
 
 
 ^^-^o, 
 
 
 * AT Oi 
 
 
 
^^^^ 
 
 V<i- 
 
 
 (^ vP^ -I 
 
 
 
 
 
 
 -^0^ 
 
 'oV" 
 
 
 ^^0^ 
 
 '^oV 
 
 .^' 
 
 e--^^ 
 
 *". "- 
 
 '\1 30 
 
 I ^SSt^ INDIANA 46962 
 
 
 • <^ 
 
 
 0'