Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Chapter 2 RATIONALE FOR CLAS SIFICATION OF COMBUSTIBLE DUSTS INTRODUCTION Dust explosions have plagued humanity for a great many years. Indeed, the first well-documented case occurred in a flour mill in Italy in 1785 (Palmer 1973~. Since that time, dust explosions also have occurred persistently in a wide variety of other industries (e.g., agriculture, mining, chemicals, and plastics). The possibility of such explosions often is unrecognized because a material may present little or no explosion hazard in bulk form and becomes extremely hazardous only when in the form of a dispersible dust. Until the 1981 edition was published, the National Electrical Code (NEC) classified dusts in three general categories: agricultural, carbonaceous, and metallic. Selection and installation rules for electrical equipment recognized that the hazards posed by electrically conductive dusts and electrically nonconductive dusts were different, but it was not until the 1981 edition was published that the NEC quantified the boundary values of electrical resistivity for Groups E, F. and G dusts and explicitly stated a value of resistivity below which a dust must be considered electrically conductive. These 1981 amendments to the NEC resulted, at least in part, from the earlier work of this committee and particularly its Panel on Classification of Combustible Dusts. The work of the Panel on Classification of Combustible Dusts was initiated after the Occupational Safety and Health Administration (OSHA) requested that this committee classify a submitted list of several hundred gases, vapors, and dusts in accordance with the groups presented in Article 500 of the NEC. In work previously conducted for the U.S. Coast Guard by ~ _ ~ ~ rat ~ ~~ ~ ~~ ~ nil A ~ ~ ~_:] the NRD committee on Hazardous Materials (lY/U, lY/l, 1Y/d, 1Y/~), a mernoa to classify flammable gases and vapors had been established, and this method was used by the committee to prepare a matrix of combustion-relevant properties and classifications of gases, vapors, and selected solids (Committee on Evaluation of Industrial Hazards 1979~. However, neither the NEC nor other standards give a rationale for classifying combustible dusts in Class II, Groups E, F and G as defined in NEC Article 500. Although many test methods for determining parameters of dusts that are relevant to classification were known, none were generally accepted and recognized by common use. Accordingly, the committee established the Panel on Dust Test Equipment to recommend needed test methods. 3
4 PREMISES FOR CLASSIFYING DUSTS INTO GRO US E, F. ED G The committee considered the following four premises prior to classifying dusts into Groups E, F. and G: 1. The classification method must be closely related to the standards for selection and installation of electrical apparatus in Class II locations mandated by the NEC. 2. The classification method need not encompass the classification of dusts with such inherently high fire or explosion potential that the incremental hazard caused by the presence of electrical apparatus is insignificant. Such dusts require special consideration and handling beyond the scope of the NEC. - 3. The classification method should permit classification of any unclassified dust based on easily determined parameters but also must allow, if available data petit, the identification of dusts that are a fire hazard but not a serious explosion hazard. Such dusts need not be classified according to the NEC guidelines. 4. The classification method must permit classification of plastic and chemical dusts as well as agricultural, carbonaceous, and metallic dusts. CONSEQUENCES OF THESE PREMISES Premise 1 NEC selection and installation requirements for Class II locations specify use of dust-ignition-proof enclosures or dust-tight enclosures, depending on the type of equipment enclosed and the relative hazard of the location. In addition, the NEC specifies maximum surface temperatures for electrical apparatus. Underlying these rules is the assumption that the tight enclosures will prevent dust from entering so that arcs, sparks, or hot surfaces inside the enclosure cannot ignite the dust, and the external surfaces of the apparatus will not reach a temperature sufficiently high to ignite a dust cloud or a layer of dust which may accumulate. The use of an enclosure to prevent ignition permits a dust to be classified using only the resistivity limits specified by the NEC. Ignition energy, explosive limits, and other combustion parameters need not be known. As explained below, the requirement that the surface temperature of apparatus be low enough not to ignite a layer of dust deposited on the apparatus demands a knowledge of the layer ignition temperature of the dust. This parameter is used in selecting apparatus for safe use, not for classifying a dust according to the NEC guidelines.
5 Premise 2 The objective of the NEC is to provide guidance concerning the selection and installation of electrical equipment to minimize danger from electrical shock, fire, and explosion resulting from the use of electrical equipment. Stated differently, the goal is to ensure that the risk of shock, fire, or explosion is not significantly greater when electrical apparatus is used than it would be if there were no electrical equipment present. Given this objective, it is not meaningful to classify materials that inherently pose a high fire and explosion hazard because the incremental risk resulting from the presence of electrical apparatus is low. Pyrophoric materials, strong oxidizers, and explosives that demand special precautions in use and handling need not be classified. Premise 3 As explained above, a combustible dust can be classified in Group E, F. or G if only the resistivity of the dust is known. The test for determining resistivity is simple, easy to use, and accurate enough for safe classifica- tion. However, some dusts (e.g., cereal grass, cotton [inters and some plastics dusts) can be fire hazards but are only weak explosion hazards because they are relatively difficult to ignite and, if ignited, produce relatively weak explosions. The classification of these dusts based on resistivity alone would impose economic penalties on the user of electrical apparatus (i.e., the user would have to select and install special electrical apparatus despite the fact that, in practice, the hazard level is so low that general-purpose apparatus could be used safely). Therefore, the method of classifying dusts based on pertinent combustion parameters permits the identification of low hazard potential dusts, so that they need not be classified. Classification of a low hazard potential dust is always safe, but may be uneconomic. The apparatus for use in determining the parameters needed to evaluate the severity of hazard is much more expensive to construct than that required for the resistivity test and the test methods require much greater skill than that required for the resistivity test. Also the parameters needed to evaluate the severity of hazard are much more controversial. Premise 4 The NEC requirement that the surface temperature of electrical apparatus be limited to prevent ignition of deposited layers of dust, or dust clouds, presently is implemented by imposing maximum surface temperature limits on
6 electrical apparatus, depending on the type of apparatus and the group c lassification of the dust. If NEC rules are to be applied to dusts other than metallic, carbonaceous, and agricultural dusts, a provision must be made for dusts that might ignite in cloud or layer form at a temperature lower than the maximum surface temperatures presently stipulated in the NEC. A review of published data (Dorsett and Nagy 1968, Jacobson et al. 1961, 1962, and 1964, Nagy et al. 1965 and 1968) indicated that, with some exceptions (approximately 50 out of 1500), the temperature of a heated surface required to cause ignition was always lower for a layer of dust on the surface than it was for a dust cloud in a heated chamber. In those cases in which the cloud ignition temperature was lower than the layer ignition temperature, it was apparent that either: 1. There was a change in state of the dust from a solid to a gas (a typical example would be asphalt), 2. The material in the cloud ignition temperature test was somewhat different from that in the layer ignition temperature test because of oxidation of each particle (an example would be atomized aluminum), or The difference in temperature was within the test tolerance and, as such, is probably not significant. The committee reasoned that dust cloud ignition temperatures need not be considered in the basic classification scheme because of the small number of dusts having cloud ignition temperatures lower than their layer ignition temperatures. (See Appendix D for a list of dusts that have been found, under test conditions, to have cloud ignition temperatures lower than the layer ignition temperatures.) Although the cloud ignition temperature is required to determine ignition sensitivity, it need not be considered in characterizing a combustible dust once the decision has been made that the dust is combustible and therefore needs classification. The layer ignition temperature is defined to include not only observation of ignition but also any observable change in physical state during the test. Because the layer ignition temperature may be lower than the maximum surface temperature now permitted by the NEC, the committee recommends that, for safe selection of equipment for use in the presence of such dusts, the maximum surface temperature should be required to be lower than the layer ignition temperature by a fixed differential (e.g., 25°C). Although dust characteristics such as particle size and shape, thermal insulating properties, abrasiveness, previous history (of metallic dusts), and moisture content may affect explosion parameters significantly in laboratory tests, these effects are not judged to be significant enough to change the classification of a dust. Particle size and shape are known to affect the critical properties of dust, but the availability of a relatively simple and specific test method to determine resistivity and layer ignition
temperature permits the specific dust available in the area under considera- tion to be evaluated should there be any question as to the group classifica- tion. The thermal insulating properties of a dust will have some effect on the surface temperature of the electrical equipment on which it collects, particularly equipment that generates considerable heat such as lighting fixtures and motors, however, this effect is judged to be relatively minor for most combustible dusts. The abrasiveness of a dust is not pertinent to its classification. An abrasive dust speeds wear of moving parts (e.g., motor bearings) and allows dust to enter the equipment; therefore, abrasive- ness is a parameter to be considered in equipment design, inspection, and maintenance, not a parameter related directly to explosion properties of the dust. The previous history of a dust, particularly metal dusts, is judged to be a critical factor with respect to electrical resistivity. The committee felt that for such highly conductive dusts the electrical resistivity of the solid from which the dust is formed generally should be considered to be the electrical resistivity of the dust itself.* Although moisture content is known to have an effect on the ignition properties of dust, the major effect appears to be to accelerate spontaneous ignition of non-metallic dusts when the dust is in deep layers. The effect of moisture content on ignition properties was judged to be of only minor significance with respect to ignition by the heated surface of electrical equipment because the heat of the surface would drive off the moisture. *In special situations, the method recommended in this report can be used if the samples tested are representative of those actually present in the workplace. That samples are representative is, of course, subject to agreement by the responsible code-enforcing authority. The resistivity also must be measured under conditions comparable to those to which the dust is present in practice. The test for resistivity must be conducted at the high voltage to which the dust may be exposed to ensure that high resistivity surface coatings do not break down when subjected to a voltage gradient in equipment that is higher than that used during measurement. m-a increment: .
8 REl?ERE:NCES Committee on Evaluation of Industrial Ha Bards, Matrix of Combustion-Relevant Properties and Classification of Gases, Vapors, and Selected Solids, Report N=B 3 53-1, National Academy of Sciences, Washington, D.C ., 1979. Committee on Hazardous Materials, Fire Hazard Classif ication of Chemical Vapors Re lative to Explosion Proof Equipment vol . 1, National Academy of Sciences, Washington, D.C., Reports I (February 1970 ), II ~ October 1971), III (May 1973), and INS (1974). Dorsett , H. G., Jr., and Nagy , J., Dust Explosibility of Chemicals, Drugs, Dyes, and Pe sticides, Report of Investigations 7132, U. S. Bureau of Mines, Pittsburgh, Pennsylvania, 1968. Jacobson, M., Nagy, J., Cooper, A. R., and Ball, F. J., Explosibility of Agricultural Dusts, Report of Investigations S7S 3, U. S . Bureau of Mine s, Pittsburgh, Pe nnsylvania, 1961. Jacobson, M. , Nagy, J., and Cooper, A. R., Explosibility of Dusts used in the Plastic s Industry, Report of Investigations 59 71, U. S . Bureau of Mines, Pi ttsburgh, Pe nnsylvania, 1962 . Jacobson, M., Cooper, A. R., and Nagy , J., Explosibility of Metal Powders, Report of Investigations 6516, U. S. Bureau of Mines, Pittsburgh, Pennsylvania, 1964. Nagy, J., Dorsett, H. G., Jr., and Cooper, A. R., Explosibility of Carbonaceous Dusts, Report of Investigations 659 7, U . S . Bureau of Mine s, Pittsburgh, Pennsylvania, 196S. Nagy, J., Cooper, A. R., and Dorsett, H. G., Jr., Explosibility of Miscellaneous Dusts, Report of Investigations 7208, U. S. Bureau of Mine s, Pittsburgh, Pennsylvania, 196 8. Pa leer, K. N., Dust Explosions and Fire s, Chapman and & 11 Ltd., London, p . 7-8, 197 3.
