THE COLLIERY GUARDIAN ; and ■/ ■/•- • ■: JOURNAL OF THE COAL AND IRON TRADES. : Vol: cix. . FRIDAY, MARCH 19, 1915. No. .2829 Coal Dust Explosions. By J. D. The object of this paper is to show that, notwith- standing the great value which must be attached to the., research of the Home; Office Committee, an extension of that research is necessary before practical recom- ) the mendations can be made to mine owners and to Legislature. /■ ; " /"' : Inflammability of Goal Dust Clouds. / Statements . concerning/ the ffon-inflammability certain coal dusts must he accepted with reserve. MORGAN. central point, which bears against an aluminium block G, , carrying, a light rigid style .H .of “1 section. The block. is mounted on a blade spring I, secured at its ends in an adjustable clamp. Records are taken on a smoked' paper strip attached to a spring driven drum J. Time' is recorded on the strip by an electrically-driven tuning fork. - A weighed quantity of dust is placed at the ’bdttbm ./■of1 the cup, and, by means of a puff of;air producect/by> .. the .pump, is raised in the . form of a cloud jwithin’the cylinder. /On reaching the coil, ignition occurs, and the inflammation then spreads downwards throughqpt ■ ,the ; cloud. It is found that with charges up to a certain... limit, very uniform clouds can be obtained with the apparatus; but when this limit is exceeded, the cloud reaches the coil before the whole of the dust is raised, of vTt| does not appear to be possible to establish any standard .•> of inflammability that is independent of the source of/, ignition; for in all "daises the inflammation of a dust cloud is to some extent dependent on the quantity of heat in the igniting medium. The same remark is true for explosive gaseous mixtures. Gases which can readily be exploded by what are calledfat ” sparks cannot be exploded by small siparks produced under certain con- ' >nd the Pressure of explbsion closes the valve against ditions.f It may be that -a useful standard -for coal- .^.pressure of the air pump. ^Consequently, the record dusts could be established by comparing the explosi- bility: of a coal dust of .one class, with that of' another under similar, conditions. ' But, at the same time,/it - must i be j remembered that With a - different igniting medium an anthracitic dustsniight fie capable of pro- ducing as violent an explosion as a bituminous;dust. In this connection it .is necessary to point , out that'the explosibility of any dust cannot be established definitely from its chemical composition, and especially does this .apply to coal dusts. The later Home Office reports give recognition to this . fact,'and* ? the diagrams shown in ‘figs. 1 to 11,'whenr taken in ’ conjunction with the analyses, emphasise-the fact. With regard to t^e- foregoing: considerations, the question arises as’to*'Whether the relative inflammability of different coal dusts [under the conditions’ which obtain in a mine.canibe determined by small scale experiments. In thb writer's . opinion, the answer is in-the negative. Only ^operations on a large scale can yield reliable infor- mation of practical value; to .mining engineers. From small’ scale laboratory experiments - a certain well- defined result: is obtained? with all readily inflammable coal -dusts, namely, j that the rate . of /inflammation depends upon 4he density of the cloud. The validity of this result jappears:to be independent of the magni- tude of the experiment. Starting with a cloud of given • density, .the speed with which a; flame develops-’ and : spreads diminishes; as the density is increased. This-is • well shown by figs.-1 to 11. In each figure four curves are seen, which* were obtained by the explosion in a small closed chamber (having a capacity of 100 cu. cm.) of .different •quantities of dust, namely, - 0’05, dl,. *0-2; and 0-4 gramme. ; • ' . : . - • , . ; /.Description of Experimental Apparatus. The apparatus used in the tests recorded in figs. 1 to 11 is shown in fig. 14* The main-, features, are the same as’those described, in a previous paper by the writer previously cited.+■ ‘.Several improvements' suggested- by experience;have been added. It WilT'be noticed that the explosion chamber A is cylindrical for' the .greater, part of its length. At the lower. endmdeepMetachable cup B is provided, and in this is arranged a valve C. ■Con- nected to the lower end of-the cupds aru air injection pump- D. The internal surface of■ 4he ‘cup is made .to expand gradually from the valve', and^tb- merge into .the. cylinder without abrupt change- ;bf ^ectiom At the upper end of the cylinder is a platiriUm ignition coil E, Wound on a short quartz tube, and ne&r the upper end is a' branch comintifiiOating with' the pressure -indicator; This consists of’.aucdrrugated bronze disc F, fitted with a : * From a .paper).re$d before the North Staffordshire Insti- tiiW’bf"Mining and 'Mechanical Engineers.' ’ . ' See' ” The ;Tgfiitioir;of• Coal : Gas and -Methane by Momentary Electric ' Arcs,’.’., by Brof//W.- M./ Thornton, D.Sc..,. Transactions, Inst. M.E.V1912, yol. xliv., p. 145.. ' f “ Dust Explosions,” by J. D. "Morgan, Proceedings of the Institution of Civil Engineers,'1914, vol." cxcvi/', p. 384: - obtained from the explosion of a relatively large charge '\,.is not a true record for that charge,' but only for that - .portion which entered the. cloud, and. this quantity is \ mqt determinable. By using a higher injection pressure ya, greater charge can be raised; :buff this procedure. • increases the turbulence factor; in that it creates. a ^greater internal disturbance in the cloud, and results in- an artificial increase in the explosibility of the mixture. In the experiments recorded., there.- was evidence that 0-2 gramme or rather less was the maximum charge that could be completely raised. Consequently the explosion, curve, obtained with a 0*4 gramme charge does not show '' in each -case, the result which-would be obtained; if the whole of the charge entered the cloud. With the quantity which is raised a marked diminution of explosibility is found, as compared with, the smaller charges, and this diminution would be greater if the whole of the large charge could be utilised. In order to avoid the cooling effect of the ■cloud and injected- air on the wire, the latter is coiled / around a thick quartz tube or bar. In the design of apparatus for use in dust explosion experiments, . The:Goals Used. The following shows the proximate analyses of coals used jn tests illustrated in figs. 1 to.11 :—. . - ’ / ;...l ■ ■ ■ • ’ Exhall. ■ ■ '; ( Ell Slate Two-yard ■ ~ l^yder coal. coal. seam. J coal. ; ' •• (Fig. 1). (Fig. 2). (Fig. 3). (Fig. 4). Per cent . Per cent. Per cent. Per cent. ■ Water .. 9’00 ... 9-97 ... 10-15 . .. io-uo ; Vtdatile matter. .. 33*00 ... 31’24 ... 28’08 ., .. 31’15 "/ Fixed carbon ... .. 51*42 ... 56*75 57*87 .. .. 57-37 Ash.. .. 6*58 ... 2’04 ... 3’90 ., .. 1-48 [ Tptals ....... .. 100-00 ... 100-00 ... ioo-oo . .. 100:00 " ■ Exhall. Hamstead. Bare Cockshead Hardmine White coal. seam. seam. coal. (Fig. 5). (Fig. 6). (Fig. 7). (Fig. 8). ’ . ■ q ” ■ " - Per cent. Per cent. Per cent. Per cent. Water / ..[... ..’ 7’58 A. 1-54 ... 2-01 ., 10’95 Volatile matter. .. 31-80 ... 32’81 ... 33’51 . .. 35’17 . Fixed carbon .... .. 55’06. ... 63!45 ... 61’24 . .. 48’83 Ash.... .. 5-56 ... 2’20 ... 3’24 . .. 5’05 Totals .. 100-00 ... 100’00 ... 100’00 . .. 100’00 Birchenwood. Trinidad. Banbury seam. (Fig. 9). (Fig. 10). Manjak. (Fig. 11.) i. Per cent. Per cent. Per cent. Water. >... .;. 2’18 ... 1’20 . ..• 2’30 " Volatile matter ... 35’63 ... 3V70 . .. 60-05 -Fixed carbon ... ... 59’88 ... 64’10 .. .. 36’90 Ash ... 2’31 ... 3’00 . .. 0’75 Totals:.......; ... 100’00 ... 100’00 . .. 100’00 Note—All the samples were dried for 1 hour at 107 degs. Cent. 1 he volatile matter was estimated by the standard Note—All the samples were dried for 1 hour American method. Density and Pressure. • - 'Only the essential part of the explosion curve is shown __ __ It will be noticed that in almost all the tests' the time taken to reach the maximum pressure increases with the quantity of dust, and that the average rate at which the pressure increases during the explo- sion diminishes with increase of dust. This is equiva- lent to stating that an increase in the-density of the dust cloud has a diminishing effect upon the inflammability of the cloud. The fact is one of great importance in connection with the committee’s 'recommendation as to the quantity of incombustible dust to be used in mines, a great difficulty is encountered which is not experienced for it immediately suggests the questions: Is the-same in gas.: explosion work, namely, that of producing and result found in large scale experiments,, and, if so, on maintaining a uniform cloud or mixture. well brought out bymopiparison of the writer’s explo--; incombustible dust to. be computed? Assuming the for it immediately suggests the questions : Is the- same _______________________________ o- ______A \ - ", - ■ , -J- This fact, is what amount of dust in the mine is the quantity of ‘ _ / incombustible dust to. be computed? Assuming the si oil curves with that shown in fig. 15. For the latter result to be the same under mining condition, it follows the author-is indebted to Dr. R. V. Wheeler. This/ H curve should be compared with the curve corresponding to the 0-1 gramme charge in No. 6 test, both tests being made in Cockshead seam coal in clouds of approximately... the same density. It will be noticed that ...in Dy,. Wheeler’s curve,the, time • of explosion is much, quicker and: the pressure much greater than, any shown': in .the ■ writer’s test.■ ; This seems to be due entirely to'the' con-- r ditions’of the experiment. ; Dr. •Wheeler’s'apparatus comprises a spherical explosion chamber, In theibottdm, of. which-a small fan-is run at a high rate of speed. ;,In, Dr. ’ Wheeler’s ‘ apparatus a large amount of. turbulence.. is introduced by the fan, with, the -result - clearly indi- cated .by the; record. .In this ^apparatus' a. limitation of the /writer ’s apparatus is avoided., in. that by, means of the