Document rp92VMZj3D50e0mgQB7991waG

512 G. W. JONES A. FACTORS AFFECTING THE LIMITS OF FLAMMABILITY Only a brief discussion of the various factors affecting the limits off mability will be given. A more complete discussion is given in published reppf The limits are affected by the direction of flame propagation; the shape, diair and length of the test apparatus; the temperature and pressure of the mikt^ the time of ignition; the percentage of water vapor present; and, indirect! the source of ignition. Wider limits are obtained for upward propagation of flame tlfa horizontal or downward propagation. Therefore, the risk of an explcf greater when the mixtures are ignited from below than when ignited from" The limits of flammability are widened as the diameter of the appara increased, rapidly at first and then more slowly as the diameter apprpf inches. An apparatus greater than 2 inches in diameter gives limits ve~ different from those obtained with 2-inch apparatus for tests at atmo pressure. The apparatus must be long enough to insure, continued propii of flame after the heat imparted to the mixture by the source of ignir been dissipated.- An apparatus 4 feet or more in length is sufficient! It found that if the apparatus is closed when the mixtures are ignifedj gases are in gentle turbulence, the lower limit is reduced slightly.4 Ordinary variations of laboratory temperatures have no appreciali on the limits of flammability. Elevated temperatures cause a widenip limits. Normal variations of atmospheric pressure have no "* on the limits.~The--effect-of-high -pres8ures-on-the-limits-is-nerth-er^! uniform, but specific for each particular combustible. As yet, np:f been developed for predicting the effect of high pressures on the! I flammability for any given combustible in air or oxygen. In both liipits are raised, while in others the limits are narrowi increased pressure wylens the limits of flammability of saturated hydfpj i The normal quantity of water vapor present in atmospheres, temperatures affects the lower limit of flammability only to a slight|e presence of water vapor reduces the upper limit because some of the mixture is displaced by the water vapor. Since the oxygen concpH| the important factor in an upper-limit mixture, as the oxygen amountvof~:combustible-that-can-be-;burned--is--d.ecreased7-and-"SO^ lowered. 1 H. F. Coward, C. Carpenter, and W. Payman, J. Chem. Soc., 115, 27;.(1919), * Hi F. Coward and G. W. Jones, V. 8. Bur. Mines Bull. No. 503, 1952."'i '* * ..'IP r Inst. Technol. Co-op. Bull. No. 30, 1926. 4 G. W. Jones, E. S. Harris, and W. E. Miller, V. S. Bur. Mines Tech. Baficri:N<h'i5Qp *G. W. Jones and R. E. Kennedy, V. S. Bur. Mines Rept. Invest. No! 3798,119C> "G. W. Jones, R. E. Kennedy, and I. Spolan, U. S. Bur. Mines Rept.lhveitftMor' 1949. ft ss'" HAZARDS OF COMBUSTIBLE GASES AND VAPORS 513 LIMITS OF FLAMMABILITY OF GASES AND VAPORS IN AIR Bill safety requires that only those values for the limits of flammability gj^^tapors in air be employed that have been obtained in apparatus ^^TK^dost limits. These conditions are met by the apparatus developed 'Misstates Bureau of Mines and shown in Figure 1. It is specially S:WsSi> !* To power line Switch [ 0ooftoJ V Essli Tran Induction coil X Drying tube -it- Vacuum pump Drying tower Ijjjffhiratus for determining the limits of flammability of gases and vapors. (Courtesy Bureau of Mines, U. S. Department of the Interior.) Hi; .... 5gn'e^^^,.determihing limits of gases, and vapors of liquids that,,are volatile j^^Sliihoratory'{emperaturesT6~giv'g'flaffifffitble'rniMti?egrTiiiFl^urdT7a"is in which the mixture is tested. Its lower end is closed-'by a lightly ||bund>gla88 plate b sealed with mercury c. It is evacuated by a pump mmliffigfliiibe j. The vapor under test is drawn from its liquidTi^ffi'O'bbhtainer ^^unig5ieflsured!-bv:'the manometer fc.- Air or other "atmosphere"- is then Sirbugh the drying tube q until atmospheric pressure is reached. The Ijjpffaojppor are then thoroughly mixed by circulation, by raising arid lowering nwuliryfyossel g repeatedly for 10 to 30 minutes, depending upon the density of combustible vapor. The mercury seal is then removed, glass plate .0 J*k <b,