594 THE COLLIERY GUARDIAN. March 19, 1915. cannot be obtained. Experience indicates that the same is true for combustible dust and air mixtures. Below and above certain limits a flame is incapable of travelling throughout the mixtures, but between those limits the flame appears to be capable of propagation throughout the mixture. Commencing at the lower limit, and proceeding towards the higher limit, a dust mixture behaves exactly like a gas mixture, in that at first the pressure of explosion increases concurrently with increase in the proportion of dust. But, when a certain critical proportion is exceeded, further increase of dust is attended by a diminution of the pressure of explosion, and this continues until the higher limit is reached. Summarising the foregoing results, one finds that when the amount of coal dust is below the “ lower limit,” or above the “ higher limit,” indefinite propa- gation of a flame cannot occur. When the amount lies between those two limits, but is in excess of a certain critical quantity, the coal dust-itself diminishes both the rate of inflammation and the pressure attained. It must be understood that the foregoing criticisms are based only on laboratory experiments, but it is desirable that experiments should be carried out on a large scale in order to determine the effects of excess on the explosi- bility of coal dusts, and thus provide practical informa- tion which can be obtained in no other way. lishes the usefulness of stone dust in preventing the initiation of an explosion. With regard to the effects upon an explosion that has developed beyond the igni- tion stage, the results are inconclusive, and it seems necessary that further work should be done in this direction, attention being directed to the effect of vary- ing the quantity of coal dust, as well as the proportion of added stone dust. Effect of the Surface of Mine Workings on Dust Explosions. It is well known that a gas or dust explosion is accelerated by increasing the internal motion or turbu- lence of the gas or dust mixture. The writer has noticed that when a gas explosion is produced in • a small steel tube which is open at one end and closed at the other, the explosion being initiated at the closed end of the tube, the effect is more violent when the interior of the tube is covered with a rusty incrusta- tion than when the interior-is smooth. Apparently the eddying motion set up by the rough surface over which the gas moves accelerates the explosion. Fig. 12 shows the different effects in the two cases. An instrument designed to record the pressure variations which occur during a gas explosion in an open tube* was placed at the closed end of the tube, and a mixture of coal gas and air introduced by a laboratory Bunsen burner was ignited by a sparking plug. In all cases when a smooth the gallery on dust explosions produced. therein. • ’ These constrictions were made ih the form of. rings, -each.of which was 6 in. deep. - In one series of experiment's, three constrictions were introduced at distances of 50 ft. apart, these being situated at points 300, 350, and 400 ft. distant from the point of ignition in the gallery. The difference caused by the constrictions is very remarkable.’ Fig. 13 is obtained from the diagrams published in the Fourth Report. The two curves were produced by a manometer placed 50 ft. from the open end of the gallery. No. 1 shows the record of a coal dust explosion obtained when the gallery was clear, and No. 2 when the gallery was fitted with three constrictions, as aforesaid. In the first the maximum pressure was 161b. per sq. in., and in the second 152 lb. . Now, presumably, all the, tests on which the com- mittee’s recommendation as to incombustible dust was based were carried out in the smooth or clear gallery. Having regard then to the “ constriction ” experiments, it seems highly probable that adulterated coal dust mixtures which were feebly explosive in the clear gallery would be energetically explosive in the same gallery provided with constrictions. Having regard also to the fact that a mine working has a rough interior, and is more or less obstructed at intervals, experiments in a smooth gallery are of little or no value as criteria of conditions in a mine. It would appear, therefore, that Fig. 4.- Ryder Coal, Figs. 1 to 11.—Diagrams showing the Explosibility of Various Kinds of Dusts. ♦<> 30 20 10 0-7 •05) 30 20 0.1 Fig. i.—Ell Coal, Exhall Colliery. Fig 2.—Slate Coal. Exhall Colliery Fig. 5.—Bare Coal, Exhall Colliery. Fig 3.—Two-vard ’Seam, Exhall Colliery. TIME IN SECONDS TIME IN SECONDS. 0.2 0-3 0.4 TIME IN SECONDS 0-2 0-3 0-4 0.5 TIME IN SECONDS 03 0-4 05 TIME IN SECONDS. 40 20 <0.1 Q.5 40 30 20 1-05/ .0.1 10 10 0.1 *0-2 F IG 20 30 TIME In SECONDS. 20 30 35 . TIME IN SECONDS 0.7 Fig. 10.-Bullhurst Seam, Birchenwood Colliery Fig 8 —White Coal. Hamstead Colliery TIME IN SECONDS. TIME IN SECONDS. TIME IN SECONDS. -Seven-feet Banbury Seam, Birchenwood Colliery. Fig 15—Coal-dust Explosion Curve obtained from Wheeler’s Bomb Apparatus. Fig. 6.-Cockshead Seam, Sneyd Colliery. Fig. 7 -Hard Mine Seam, Sneyd Colliery Fig. 11.—Manjak. 20 30 0.1 TIME IN SECONDS. '0-2 0-3 £ 10 Fig. 12.-Oscillograph Records of Gas-explosions in Tubes.showing Modification due to Character of Inner Surface of Tube TIME IN SECONDS. 0 0.I 0-2 TIME IN SECONDS. Fig 13.—Curves of Explosions in Home Office Gallerv showing the Effect on Explosions of a few small Constrictions in Gallery I Z 120 ui tt o 1OO co UJ & « 80 O z 60 ul a co- CO <0 LU Fig 14.—Morgan’s Explosion Apparatus __K G , so Ul 9 CO m~ ul 0 0.01 0'02 0.03 0*04 0*05 0*00 0*07 0-08 TIME IN SECONDS 0 EQUALS tIME OF eIRST appearance of pressure Ul tc 9 O £ W a M Q Z 9 O Firom the foregoing deductions it may be argued that an excessive amount of coal dust in a mine is a natural protection against dust explosions; but it must not be overlooked that, notwithstanding the retarding effect of excessive coal dust on inflammability, a real danger is introduced by excess. If we suppose that sufficient dust is disturbed to produce an extensive cloud, the density of which is considerably greater than that which will permit of the most rapid inflammation, and that the cloud is ignited,-then while the flame is spreading the cloud in advance of the flame is rapidly settling and becoming less dense. If the flame can persist until it reaches a zone of critical density, it may on reaching that zone produce a violent explosion, and the above process may then be repeated. If the addition of incombustible dust is useful, its effect should be most markedly shown in explosions of critical mixtures which give the maximum explosion effects. Presumably the quantities of coal dust employed by the committee in some of the reported tests were such as would give these mixtures. In the Fifth Report it is stated that the committee entertains no doubt that an easily raised incombustible dust, if properly placed, is capable of dealing effectively with a feeble inflammation, even under the most adverse -con- ditions; also that mixtures in equal proportions of coal and incombustible dust are difficult to ignite, but when ignited may propagate flame throughout a large gallery. Judging from the details given in the Fifth Report, it would appear that the committee’s work chiefly estab- steel tube (10 ft. in length by 1 in. in diameter) was used, the gas exploded with a prolonged rumbling report (a typical graph is shown at a). But when a rough tube of the same dimensions was employed, and the right mixture obtained, the explosion occurred with great rapidity, and produced a short and loud report (a typical graph is shown at b). To anyone familiar with records of this kind, a will be recognised as indicative of a slow and comparatively gentle explosion, and b of a rapid and violent explosion. The initial maximum pressure attained is not greatly different in the two cases, but whereas in a the pressure quickly falls and continues in an oscillatory condition while the flame is slowly travel- ling, it is maintained in b practically at its maximum during the passage of the flame at high speed through the tube. In a the instant at which the flame reached the end of the tube coincides roughly with the end of the graph, while in b the flame reached the end of the tube at the instant coinciding with the end of the first interval of positive pressure. Positive pressure is indicated above the straight zero-line, and negative pressure below that line. The Home Office testing gallery at Eskmeals consists of a comparatively smooth tube. In the Fourth Report the committee describes the effect of constrictions in * For details of this instrument, see “An Instrument for Recording Pressure Variations Due to Explosions in Tubes,’’ by J. D. Morgan, Proceedings , of the Physical Society of London, 1914, vol. xxvi., p. 172. the Home Office experiments on the efficacy of incom- bustible dust need to be continued in a gallery so arranged as to imitate closely a mine working before any recommendations can be regarded as final. In conclusion, the author desires to express his indebtedness to Mr. A. H. Clarke, of the University of Birmingham, for his analyses of the coal samples, and to Mr. E. C. R. Marks for the advantage of using the Marks and Clerk Laboratory (Birmingham) in carrying out the experimental work. Exports of coal from China in 1913 were more than double those of 1912, amounting to 1,500,000 tons, valued at nearly <£1,000,000. The figures represent chiefly coal from the Kailan and Fushun mines exported to Japan. The Kailan Mining Administration of Chihli had a successful year, and their total output was 2,036,966 tons, against 1,706,658 tons in 1912. The Pekin Syndicate mines in Honan had trouble with water, and their output fell from 549,877 tons in 1912 to 283,510 tons in 1913. The Hungshan and Fangtse mines in Shantung produced 548,600 tons, as compared with 573,600 tons in the previous year. No statistics of production are to hand for the Fushun mines, near Dairen, but the export of coal from these mines to Japan has considerably increased.