Click for next page ( 44


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 43
1 Chapter 6 BASIS FOR REC:OMMENDAT TONS Previous reports concerning the prevention of grain dust explosions in grain~andling facilities have presented numerous recommendations. The panel found no major fault with these recommendations except that they were not ranked in term of priority, value, or economic feasibility. To be of value to an industry composed of facilities as varied in size, construction, and purpose as those handling grain, recommendations- must take into account the fact that no single recommendation will suffice to solve the problem in every facility and not every recommendation can be applied to all facilities. Further, recommendations must address factors beyond technical ones. Figure 5 illustrates the broader perspective taken by the panel to examine subtle but consequential facets of the explosion problem. For example, the personal cosmology of both grain elevator managers and workers (i.e., how they perceive who they are and the meaning of life) influences their attitude in taking action to prevent explosions. If resources were unlimited, the panel believes that the dust explosion hazard could be reduced to a negligible level in every type of facility. Recognizing, however, that resources are not unlimited, the panel concentrated its study on first determining what could be done and then on assessing each preventive action's potential for hazard control. To accomplish the latter, each preventive action was ranked as high (H), medium All, or low (L) in terms of: 1. Efficacy - the degree to which the hazard would be eliminated or . controlled by the action; 2. Feasibility - the acceptability of implementing the action in light of the economic, legal, cultural, political, social, and technical considerations depicted in Fi gore 5; 3. Ef ficiencY - the cost-e ffectiveness of the action in terms of potential dollar loss if no action i" taken versus the cost of the proposed action. The panel's recommendations fall into two main groups: (1) recc~mendations to the grain-handling industry end its trade associations concerning hazard reduction in existing facilities, needed research, and the design of future facilities and (2) recommendations to the government concerning more effective regulations. In the following discussion, the recommendations on a specific subject are presented first and then the need 43

OCR for page 43
44 .1 CULTURAL As' \ ~ TECHNICAL >\ /SOCIAL \tethod. ~ Rate of Grain Transport`/ LEGAL \ / ~ Groin Typo, Density, ~ Age; / Previous Experience,\ ~ Prevailing ~ / ~Weather, Elevator Design,, \ / Training, Expectation., ~ ~ Litigious Attitude, \ / Delegator Cteentine's,, \ / Language Proficiency, Age, ~ ~ Reduredency, Conflict, an`/07 \ ~Ign'ffon Sourcing \ Sensory Perception, Strophes, ~ Ab~enceof Regulation' ~ Laws; \ \ EXPLOSIONS / ~ Credibility of Explosion Potential, Inter- a Intra-60vernmental / \Per~onal Cosmology, Family Life,/ ECONOMIC \ Agencies, lureaucratle Inertia,/ \` Follclore, Tradition', / Tariffs, \ Impact on Foreign Pollen l \ Value Hierarchy, ~ ~ Conflict Among \ ~ Grain Value, ~ \ Custom' ~ ~ Specie! Interim/ , _ . \ / Recial/Ethnic Sensitivity, ~ GRAIN ~ Enforcement Intent, Edueation, Prejudices / DUST Criteria, ~ Capability FOURCAL / Grain-Handling Schedulee, Elevator ~ Grain Insurance; \ / By Salvage, locator Relelion. - - FIGURE 5 Facets of a systems approach to grain elevator explosions. 1

OCR for page 43
1 45 for them is explained. The panel also realizes that sane recommendations may already be in effect and that the cost of implementing these recommended actions will be different for different facilities. The panel hoper that its presentation of The recommendations in this manner will convince owners and managers to examine the list, to determine where their operation is deficient, and to take remedial action within the limits of their economic and administrative capability. The same procedure should apply to those recommendations aimed primarily at labor organizations, trade and professional organizations, and federal, state and local governments. EXIST ING FACILITTE; 1 The following recommendations address engineering matters and administrative actions. Dust Control Re oononendations Ef ficacy Feasibility Ef f~v . . 1 Btabl ish a housekeeping H M H program involving a mechanical dust collection system supplemented by manual or other means . Apply state-of-the-art H L M te chniques to reduce the concentration of airborne dust in and emanating from e levator legs . Control dust generation M M fir and airborne dust at a 11 gra in transf er and discharge points. E1 iminate all nonessential M M M horizontal surfaces . Coat all ncnhori zontal L L M surfaces exposed to airborne dust with materials that will prevent the build-up of layered dust. Apply s tate-of-the-a rt te chniques to reduce the concentration of airborne dust below the lower explosive limit where possible in enclosures other than legs. 1 L L

OCR for page 43
46 There can be no doubt that overall dust control is the most important action that can be taken to reduce the explosion hazard. It also is the step that will be the most ccst-effective. The panel's own investigation of explosions and studies of past explosions indicate that a preventable accumulation and suspension of dust was the basic feature of every explosion. There is'no question that dust explodes; the important point is that the accumulation or suspension was preventable. Of all of the locations in an elevator where duct can exist in ' suspension, the leg is by far the most hazardous. This has been recognized time and again by groups studying elevator explosions. It is the one piece of equipment with an environment of suspended dust that is continually subject to "chokings (boot fills with grain and buckets will not turn),' electrical faults, bearing failures, mechanical misalignments, ingestion of foreign material, etc. ~ of a dust collection system in the leg sufficient to reduce the-concentration of suspended dust. A means for 'preventing the accumulation of dust on the walls of 'tine leg or a means for periodically removing any accumulated dust also must be provided. A particularly difficult problem is presented by - those legs in which the middle portion of the enclosure consists of only the concrete walls of a headhouse. Manual cleaning or enclosing of the belt and buckets in a close fitting, dust-tight metal casing appear to be the only solutions. Suspensions of dust in enclosed volumes other than legs present a danger second only to that of legs. Silos, bins, garners, enclosed conveyors, etc., contain little if any equipment that can serve as an ignition source and are therefore only rarely the point of an initial explosion. However, ignition of dust concentrations in these enclosure., whose volume can be much larger than that of legs, can result in an explosion of much greater magnitude than that in a leg. These enclosures present two hazard conditions: first, the air-suspended dust that is present when they are being filled and, second, the dust clinging to the walls and ceilings that Q n be loosened by the chock of an initial explosion. The removal of the airborne dust can be accomplished using the same methods as in the leg. The dust problem in silos and the generation of additional dust can be lessened somewhat if dead boxes, grain ladders, and filling spouts that entrain a large portion of the dust in the grain stream are used. The movement of grain from one point to another results in the creation of dust and in the suspension of scene airborne dust at transfer points. A dust collection system should be used at every point where grain falls through the air (e.g., when it is transferred from one belt to another or from a spout onto a belt). There is little danger of an initial explosion occurring at these points, but without use of a dust collection-system, most of the dust will settle on the floors, walls, ledges, ducts, etc., in the work space. This dust then "n became the fuel for secondary explosions in tunnels, galleries, headhouses, and other work spaces. Thick layered dust around working equipment presents the ideal conditions for initiation and concealment - of smoldering dust f ires that can serve as the ignition source for an explos ion .

OCR for page 43
1 47 Even if dust control is applied as indicated above, there will be a need for manual housekeeping. The manual vacuuming of all exposed surfaces i. recamended over blow down or sweeping, which tends to raise dust clouds and usually does nothing more than redistribute most of the dust. (See, for example, National Fire Protection Association's NETA No. 618, Code for the Prevention of Dust Explosions in Terminal Grain Elevators.) As mentioned earlier in this report, the danger of layered dust in the work space was vividly exhibited to the panel in one of its explosion investigations. An initial explosion in a headhouse was followed by an explosion in a silo. This secondary explosion was initiated by a flame that propagated near the gallery floor due to a layer of dust until it came to an open empty silo. In numerous other explosions, headhauses, galleries, and tunnels have contained enough layered dust on exposed surfaces to fuel secondary explosions within these structures. As a complement to manual housekeeping, the panel recc~'wnends that all unnecessary horizontal surfaces be eliminated and all nonhorizontal surfaces, both those in enclosures and those in the working areas, be coated with a material that will inhibit the layering of dust. Rough concrete and wood surfaces are particularly susceptible to a buildup of layered dust. Surface coatings over metal should be somewhat conductive. In summary, dust control is most important in reducing the dust explosion hazard in grain-handling facilities. Same aspects of the dust control recommendations are relatively more expensive than others and sane may already be in effect. The value of each of the recamended actions depends, of course, on how well they are applied or performed (e.g., dust control systems that do not keep the dust concentrations below the rawer explosive limit or manual housekeeping poorly performed are dangerous since they instill a false sense of security). The total dust control efforts should be based on a performance standard and not merely on the application of the recommended actions. Every elevator having interior legs should utilize an adequate dust collection system in the leg because of the extremely hazardous condition resulting from suspended dust in proximity to potential ignition sources. The other dust control recommendations will contribute to a reduction of the hazard and are, to some degree, interdependent. For example, the application of surface coatings reduces clinging to vertical surfaces but does not eliminate the need for dust collection from enclosures other than legs; without dust collection at transfer points, the need for manual housekeeping increases greatly. Adequate manual housekeeping is possible only when there is easy access to hidden spaces and all surfaces that can support layered dust. Her~e, special attention needs to be paid to providing access ports to dust-containing enclosures to facilitate cleaning. Easy access to large expanses of walls and ceilings, such as occur in many headhouses, must be provided . me panel, of course, advises that all of these recommendations be implemented.

OCR for page 43
48 Innition Control Re commendations Use a preestablished and enforced permit procedure whenever welding, cutting, or other open flame work is to be done . Incorporate a system to indicate belt slippage and misalignment. Incorporate a method to check frequently the temperature and vibration of cri tical bearings. Use devices to extract foreign materials from theincoming grain stream. Ground all conveying and electrical equipment. Use only equipment and installation L stands rds meeting National Electrical Code requirements. Ef f icacy By Ef f iciency H H H H ~ H H ~ H M H H L H - H Next to the control of dust, the control of ignition sources is the most effective means for reducing the explosion hazard. Since the data on ignition in actual explosions are poor, it is not possible to give a meaningful ranking to ignition sources; therefore, the panel arbitrarily divided the sources into the eight categories shown in Table 2 and then assessed the probability of their occur rence and the ease of their elimination. The major deterrent to spontaneous ignition of stored grain as a likely source of ignition is the necessity of preserving the commercial value of the grain. In modern operating practice, if the grain is to be in residence for more than a few weeks, the temperature i. closely monitored. A rise of a few degrees in grain temperature indicates-insect infestation and fungus, which reduce the grain's value and mandates countermeasures, cooling of the grain by pulling air through it or by turning it. Thus, spontaneous ignition is not considered a probable source. me probability that electrical apparatus and wiring selected and installed in accordance with the provisions of the National Electrical Code (NEC) will be a source of ignition is extremely law under either normal or fault conditions. The code provisions and apparatus standards place limits on temperature of exposed surfaces and mandate enclosures that exclude dust that could contact energized parts.

OCR for page 43
1 TABLE 2 Ignition Sou roes 49 Probability Ease of Source of Occurrence Elimination Spontaneous ignition Low Easy Arcing from electrical apparatus 1 Normal operation - Low Easy fault Me ration Low . Easy Sparks from foreign materials Elevators, ferrous metals Low Easy Elevators, nonferrous metal Low Easy Elevators, other Low Easy Mills High Moderate Static electricity Moving belts High Moderate Moving grain/dust Low Difficult Hot Surfaces o Lad Low Easy Bearings High Moderate Radiators/Pipes Low Moderate Friction Rubbing head pulleys Moderate Easy Slipping. belts High Easy Scraping buckets Moderate Moderate (misaligned belts) Open f lame Hi gh Moderate Welding and cutting High Easy 1

OCR for page 43
50 If the installation is not in accordance with the NEC, the probability of the electrical apparatus or wiring becoming a source of ignition is higher, but in itself such an installation is not likely to be a source of ignition unless open contact switches or other arcing parts are exposed -directly to accumulated dust. However, if NEC guidelines are not followed there is a much higher probability that there is also little control of the use of portable equipment such as drills, hand lamps, and grinders that may be ignition sources. Failure to observe NEC requirements often can be an indication of more serious problems rather than an imminent danger itself. Even in an installation that does not strictly reflect NEC requirements, the immediate hazard may be only moderate if the equipment is nonsparking, enclosures are dust-tight, and a good standard of housekeeping is observed. This does not obviate the need, however, to follow NEC rules because a conforming installation is forgiving of other problems whereas a nonconforming installation in combination with other bad practice may become a hazard. Most investigations of the production of sparks between combinations of ferrous, nonferrous, and rocky materials have been concerned with the ignition of methane. Although there have been many investigations, no simple picture of the conditions required for dust ignition has emerged. It generally is agreed that the the rmite reaction between aluminum and rusty-steel under some conditions can ignite methane. Investigators differ on whether impact or friction is the important parameter and whether steel-steel or rock-metal impacts are ignition capable. The range of energy reported for methane ignition by sparks generated by impact is wide, ranging from 10 to 400 ft-lb, but in most reports above 200 ft-lb. It seems unlikely that a piece of tramp metal or a rock small enough to pass through a 1-1/2 in. grate would result in enough impact energy to ignite grain dust. The weight of a piece of foreign material that could pass a l-1/2 in. grate could be as high as 1 lb for the case of a steel object. A 1 lb. object weld have to fall 200 ft to have an impact energy of 200 ft-lb. Although it is not impossible, it does not seem likely that a piece would fall from a bucket and drop that distance without first striking another bucket or the leg enclosure. The above numbers are for methane. Spark ignition energies for grain dust are at least twenty times higher than that required for methane. Even if one takes into account that slower release of energy accompanies a friction spark (a hot particle) and that dust is more easily ignitable by long-duration electrical sparks (Eckhoff 1975), one concludes that sparks from tramp material falling in an elevator leg are not likely to be a prime ignition source. The potential for tramp metal to~cause a primary explosion does exist in a hammer mill. The energy released when a small piece of metal is struck by hammers in a mill is more than sufficient to ignite grain dust. Pieces of metal as well as other hard objects also can damage hammers in mills and metal buckets in an elevator. Although the objects themselves may not produce sparks sufficient to ignite dust, the damage they cause may lead to an explosion through friction heating or spark generation by the damaged hammers or buckets.

OCR for page 43
51 It has been suggested that the use of plastic buckets instead of metal buckets would eliminate ache possibility'of sparking and reduce friction heating resulting from damaged buckets striking or rubbing against the leg enclosure. More study of this za needed since it is not certain that plastic buckets will not introduce new and equally dangerous conditions. The literature on ignition by static electricity under conditions fauna in a grain elevator is sparse. Palmer {1973) offers qualitative guidance only and advises the grounding of all ducts and metal. The Canadian Grain Handling Association {1979) concluded, primarily based on the work of Morse of the National Research Council of Canada, that static electricity generated in maying grain is not'likely to be a source of ignition. A more recent study (Safety Consulting Engineers, Inc. 1980) of electrostatic properties of grain cites the need for further investigation of the properties in conjunction with grain-handling facilities. Static discharges from belting have, in conventional wisdom, been presumed to be ignition capable (University of Sa~thhampton 1980~. Industry practice has been to bond the metal framework of conveyors to ground to eliminate build-up of static charges, and this should be considered standard practice. Although the presence of large charges has not been'conclusively shown to be an ignition source, it seems likely that arcing between parts of metal framework not bonded together could release sufficient energy to ignite a dust cloud or layer. Additional experimental data on ignition by discharges from betting, the presence of static charges in the leg, and the likelihood of' static induced ignition in dust collection system are needed. me ignition temperature of grain dust layers exceeds 200C. Hot lump surfaces can serve as ignition sources if they do not meet NEC requirements for use in dusty locations. If lighting fixtures are selected and installed in accordance with the NEC, they should not be considered an ignition'source except if installed in a position that permits dust to accumulate on' the hot,- glass surface in a way that impedes beat transfer. Fires due to hot bearings have been reported, and one must conclude that bearings are a likely ignition source if only because they are so numerous in an installation using conveyors. Boot and head pulley bearing failure is especially hazardous. Two methods for reducing the hazard due to hot bearings have been proposed: locating the bearings outside of conveyor and leg enclosures so that overheating will not cause ignition of surrounding dust and monitoring the temperature of the bearings. The application of these methods will be inexpensive in new construction but only the second is applicable to an existing facility. Both still require a relatively high standard of housekeeping to keep dust layers from accumulating on external bee ring su rfaces . Slipping belts, especially at the head pulley of a leg, have been blamed for grain elevator explosions, in many cases because friction ignited the belt, which then parted and dropped dawn the leg. The universal application of underspend devices that prevent operation of the elevator under this condition can eliminate this source of ignition. A somewhat more

OCR for page 43
52 remote hazard, though it has occurred, is friction heating of the belt due to a slipping or locked boot pulley; therefore, the underspend device should monitor the speed of the belt as well as that of the head pulley. (In one explosion (Finn 1976), the pulley speed was maintained but the belt slipped because the lagging was worn off and caught fire and dropped down the leg.) This remedy is applicable to existing as well as new facilities. Interlocks to shut dawn moving systems when part of the system fails' or when dangerous choke conditions occur should be standard features. Ve nting Pe commendation . Follow, to the extent practical The Na tional Fi re Protection Ass'ociation's Standard on explosion venting (No. 6 8) for all enclosures. Concrete structure. should be vented by windows or other openings of the size dictated by this standard. l Bf ~cY Feasibility Efficiency L ~ '' H Venting can be considered to be the third most important area (following dust and ignition source control), but its effectiveness is limited since it is effective only during the occurrence of an explosion. Many also have reservations about the effectiveness of venting. The American Insurance Association (1978),-for example, states: When the rate of pressure rise exceeds 3,300 psi/e, it apparently becomes impossible to design an effective explosion relieving system." Nevertheless, since the greatest amount of damage and human injury usually is; caused by secondary explosions, venting should be considered if it can reduce the connection between primary and secondary explosions. Many locations within a grain-handling facility can be regarded literally as Loaded cannons" when they contain sufficient dust (either in layers, in a cloud, or both) to support an explosion. Examples are legs, empty bins, tunnels, duct work, headhouses, enclosed conveyors, and galleries. If there is sufficient fuel, an explosion in any of these enclosures will propagate through it until sufficient pressure is built up to rupture the walls or until the pressure is relieved at the end of the enclosure'. (If the end of the enclosure is strong-enough to withstand the pressure, the reflection of the pressure wave back dawn the enclosure adds to the magnitude of the already existing pressure.) She bursting strength of existing structures is small compared to the maximum pressures generated by most well-fueled grain dust explosions.

OCR for page 43
53 Technical information on the merits of venting grain-handling structures is practically nonexistent. Particularly hazardous is the formulation of size of vent openings based only on the total volume enclosed, without giving consideration to the effect due to the length/diameter (L/D) ratio of the enclosure, or roughness. When the ratio, L/D, equals or exceeds 10 the equivalent diameter of the vent opening exceeds D. Numerous elevator legs have a vent panel only at the top. A well-fueled explosion, initiated at the bottan of the leg, will generate enough pressure to rupture- the leg casing as it progresses up the leg long before the pressure wave reaches the vent. Indeed, the panel has investigated explosions in which the leg casing ruptured and the vent, in operable condition' was still closed. For those cases where L/D is less than 10 (e.g., garners, scales, and bag houses}, venting can provide she protection against rupture of the enclosure but a flame will propagate through the normal or vent openings of the enclosure. In addition, the vents must release to the outside of the facility if they are to protect employees in the working areas from exposure to flames. Examples of incidental venting have been noted in a number of explosions. Headhouses having numerous windows or steel sheathing walls suffer much less damage than those constructed of concrete with only a few windows or none at all. Believing explosive pressure by blowing out windows or frangible sheathing is preferable to spraying the landscape with large pieces of concrete from a concrete headhouse that has contained the explosion until the pressure tee cones sufficient to rupture its walls. The modification of existing structures to provide venting is often impossible and always very expensive. For example, nothing can be done about venting tunnels that are already completely below grade {e.g. those below silos). Thus, venting should be applied to exterior structures (e.g., bag houses, exterior legs- or other conveyors, and exterior ductwork). Sur'are As ion Be Commendation Impractical for the workplace. Possibly feasible for the interior of equipment. Devices for the suppression of explosions can be installed in legs, ductwork, and other narrow enclosures. These devices are containers-of pressurized dry pander or inert gases, usually Halons, which release the gas when triggered by actuators sensitive to flame (infrared) or pressure rise. They are very effective in suppressing explosions in enclosures, especially legs and dust collection systems, but they have two drawbacks: they are relatively expensive for small facilities and they are not 100 percent safe against false actuation, which adds to their operating cost because recharging is expensive. Research-and development being conducted by the manufacturers of these devices should be followed closely to determine if they are becoming more cost-e ffective for small facilities.

OCR for page 43
S4 Passive and active barrier systems that have been extensively employed by the coal mining industry for explosion suppression in tunnels (Cybulski 1975, Liebman et al. 1976) must be examined separately. These devices spread an extinguishing agent-water, Purple R. rock dust--across the tunnel ahead of the advancing flame front. The passive devices consist of frames supporting water containers near the tunnel roof. The airflow behind the pressure wave created by the explosion dumps the water. With the active devices, a sensor detects the explosion and actuates the dispersion devices. Extensive testing has led to optimum designs for these barriers and they may be applicable in elevator galleries and tunnels. Specifically, the water barrier. are relatively inexpensive to construct, require little maintenance, and are reliably triggered. Fire suppression by use of automatic sprinkler systems teas only marginal value in the prevention of explosions. {me protection of the physical plant from damage due to fire alone is not the subject of this report. ~ The initiation of a number of explosions can be traced back to smoldering dust that could never have triggered an automatic sprinkler system. Inert iw Pe oonanendation Do not use inerting because it is too expensive and is dangerous to personnel. Operating a-grain-handling facility in an inert atmosphere to prevent explosions has been considered in the past. It has been judged to be completely impracticable from both a mechanical and economic point of view. In addition, the inerting of large volumes is dangerous because workers- can be asphyxiated. Education Recommendations Ef ficacy Establish an information center to distribute actively all available information on elevator and mill dust explosions and the ir causes and prevention. Investigate and report on explosions in a manner that re fleets the recommendations made by the panel in its report, Report NMAB 367-1. Feasibility Efficiency . M L H L ~ L

OCR for page 43
1 55 As noted many times in this report, ignorance of the explosion hazard is very prevalent and reflects both poor dissemination of what is known and lack of knowledge concerning hazardous factors or situations. The degree of ignorance is inversely proportional to ~ the unit Blue In the 1naustry (l.e., the management or large grain-nanoling corporations is much better informed than the management of "mall mills and country elevators); therefore, given the preponderance of small facilities, lack of information is an important factor. The degree of ignorance also increases in going from overall management to the lowest paid employee. Ignorance of the dust explosion hazard can be alleviated by collecting available information in a central repository, a relatively simple but laboricus task and by distributing the available information. Numerous organizations now are engaged in disseminating information {e.g., the National Grain and Feed Association, the Grain Elevator and Processing Society, the Department of Agriculture, NIOSH, OSHA, the trade and union press, and various university and private research organizations), but it does not seem to reach the Grass roots. of the industry. Regardless of the reason for this, it is a problem that should be overcame. The panel suggests that information-be channeled through the Department of Agriculture to the state Directors of Agriculture down to the county agents of the Cooperative Extension services. me elm OSLO ensure that each graln-processlng zaclllty is informed without having to request information. Organizations such as the U.S. Fire Administration, OSHA, NIOSH, trade and professional associations, unions, insurance groups, and trade publications also should receive all available information. it. Recommendations Efficacy ~ Efficiency , Conduct rigorous Preventive H H H maintenance, especially on all parts of bucket elevators. Notify all facility managers that safety is a non delegable re sponsibility. If author) ty is delegated it must be to an employee who reports directly to the plant manager . L H H Es tablish a f ire and explos ion M L H prevention training program in each facility. If dust is returned to the grain M L L stream do it in the least hazardous manner.

OCR for page 43
S6 The manager of each facility is ultimately responsibile for the safety of his plant. Either the manager or a designated employee who reports directly to the manager should be responsible for the day-to-day safety. This person should be completely familiar with all of the plant operations and should perform a system safety analysis of the whole plant. It is important that he understand that all explosions are preceded by a sequence of events that may have begun quite sometime before actual ignition of a dust cloud. ' 'Operating'procedures affecting safety are numerous and varied but can be classified into a few categories. The diversity of sizes and types of elevators and mills dictates that the details of these procedures be left to each plant or facility; numerous examples t for each overall subject are ' readily available. The first and most important action is to insure that every employee, visitor, contract employee, local firefighter, and any others who may be in the facility are aware of the hazard of dust explosions and the means for their prevention. Numerous examples of explosions resulting from welding operations appear in the literature and the panel is aware of a number of instances when ignorance of the proper method of fighting a dust fire led' directly to an explosion. Housekeeping, including continuous surveillance for dust emissions and deposits, must be treated as a first priority activity in plant operations. The panel has seen two instances where several feet of dust had collected in boot pits and ultimately led to explosions. It was said that the pits were cleaned out regularly every few months' The importance and degree of housekeeping should be directly proportional to the degree of activity of the facility, not inversely proportional.- ' 'Maintenance is related to ignition sources in the same fashion as housekeeping is related to dust control. It is assumed that any normal plant operation should include preventive maintenance; however, in facilities where flammable dust is a problem, maintenance to prevent ignition sources assumes greater importance. Problem sites are bearings, belts, buckets, augers, pulleys (including lagging), trippers, motors, dryers, and dust collection systems. Because of poor design, however, it frequently is 'difficult to conduct effective maintenance. Head pulley gear boxes may have . . no work platform around them, tail pulley bearings may be in an unlighted boot pit next to a wall, and legs and enclosed conveyors may not have inspection ports at critical locations. Recognized problem sites and maintenance areas must be made accessible through the use of platforms, removable sections, and hinged ports that can be used by a mechanic (and possibly a helper) who may have tools in both hands. Dryers fueled by propane or butane, which are heavier-than-air gases at room temperature, can be particularly hazardous. Leaks in fuel lines can spread a layer of highly flammable, transparent gas throughout the lowest points in a plant.

OCR for page 43
1 57 Recommendations for the disposal of collected dust are certain to create vigorous responses from industry, labor, and government. As has been noted before, the question of returning the dust to the grain stream is the point at which safety and economics can come directly into conflict. Several aspects of the economic problem were considered by the panel, ranging from total cleaning of the grain to re-introduction of the collected dust into the grain stream. A properly designed and operated dust collection system collects only those dust granules that become airborne. The amount of dust of this size per bushel of grain, although a small percentage, can change radically from crop to crop, season to season, and grain type to grain type. It therefore is difficult to predict the economic impact of discarding the collected dust other than that there is some lost involved. After considering and weighing all factors, the panel ha's concluded that much more attention must be given to the method by which dust is re-introduced into the grain stream and its effect on downstream elevators. (The panel has -.een'one particularly bed 'example in which dust collected on the upside was delivered directly into the downside of an elevator leg') This conflict between safety and economics can be resolved if the industry will assume the responsibility for developing and demonstrating a method for re-introducing collected dust that will not increase the explosion hazard above that resulting from disposal of the dust. Methods of re-introducing dust and possible alternative uses of dust that would lessen the incentive to return it to the grain stream are discussed below. RESEARCH Dust Control Recommendations Efficacy Feasibility Efficiency Continue research on methods M H H for reducing the dust concen- tration in legs to a level below the lower explosive limit. Continue research on methods L M M of reducing dust concentrations below the lower explosive limit in enclosures other than legs. Conduct research to develop L M H economic uses for collected grain dust. In conformance with some of its other recommendations, the panel believes that research aimed at reducing the dust concentration in enclosed areas will produce the greatest decrease in the explosion hazard. Thus, it believes that research should be directed at developing thorough understanding 1

OCR for page 43
58 of the movement of grain from the boot to the discharge spout and of the airflow needed in the enclosure for an effective dust control system. Other topics such as'bucket material and design, belt material and design, and placement of'input and output openings also should be examined. The aim of the research should-be twofold: to identify reasonable modifications that can be made in existing elevator legs and to develop the optimum design for legs in new facilities. Inclined belt conveyors, because of their length, are not always suitable replacements for legs. Their capital and operating costs are greater than those of legs and they require a horizontal space that may not be available. , The problem with other enclosures is different from that with legs since it involves the release of dust from a falling grain stream. This was amply demonstrated to panel members who observed barge loading. When the end of the loading spot was kept level with the surface of the grain no dust appeared; when the spout was a few feet above the grain considerable dust was released into the air. Research in this area should focus on identifying methods for preventing dust from becoming airborne when filling an enclosure and collecting the duct that may become airborne. The latter task is complicated by the fact that most dust collection systems can serve as "sneak paths. for transmitting explosions from one enclosure to another. Hence, several smaller dust control systems may be preferable to a large system. me use of grain dust for pelletized animal 'feed does not require additional research, only ecoriomic development. The few pellet mills already in existence can sell all they produce and portable mills can service small elevators having only relatively small amounts of dust. If the economic value of dust can be increased, the costs of dust collection will become much more acceptable and the tendency to return dust to the grain will decrease. Information on the present disposition of collected dust, the amounts collected by elevators of various sizes and locations, and the cost of transporting dust is needed. Research also should be conducted to answer such questions as: How clean is clean? Is it dangerous if the dust layer on the floor is deep enough to show footprints? Will a primary explosion disperse enough dust to cause a secondary explosion if the floor layer is 1/8 in. thick or 1/16 in. thick or even smaller? How much dust will adhere to concrete walls? How well will various thicknesses of dust propagate a flame? No single answer to these questions will be applicable to all enclosures in elevators and mills. However, experiments conducted under rigorous, well defined conditions can establish meaningful reasonable upper or lower bounds. ' AS noted previously, the burden of research on safe ways to re-introduce dust into the grain stream and the proof of their efficacy should fall upon the industry. Using the values previously mentioned in this report of 13,000 tons of dust collected from a large elevator in a year's time, with the value of grain about $150/ton and the value of dust (for pellets) of about S50/ton, it is easy to see that there is a difference of

OCR for page 43
59 about $1,000,000 involved. This should provide some incentive for the research. On the other hand, it does seem inconsistent to place a large elevator in jeopardy to recover two thirds the grain value of airborne dust, which may amount to only a fraction of a percent of the grain handled. (A recommendation applying to this topic has already been made. See page 55.) Research on the application of a substance to grain to inhibit the formation of dust is now in progress. The results are too preliminary for the panel to make any recommendation on the subject. Ignition Sources Re commendation Investigate the e ffect of electrostatics and humidity on the explosion hazard, including an examination of conveyor belt conductivity and the 'charging of ungrounded conductive structures. Efficacy ~ Efficiency L ~ L Host of the ignition sources in elevators and mills are self-evident and the reasons for their occurrence are not mysterious. 'The one exception is electrostatic discharge. Two research topics are involved: the build-up of electrostatic charges on conveyor belts and other poorly conducting materials and the accumulation of static charges on grain and grain dust. 'This work should encompass a number of different topics. Those readily apparent are: (1) the static charging and release characteristics of conveying belts of various materials and various conductivities; (2) the potential for build-up of charges in dust clouds in silos, bins, garners' or'' other enclosures ; (3) electrostatic phenomena occurring in pneumatic svs_~ms conveying dust and grain; {4) electrostatic conditions in legs using metal buckets and using plastic buckets of various conductivitie's; and (5) the effect of atmospheric conditions on the buildup of charges and on the energy needed to ignite grain dust. Among the aims of'work done concerning topic 5 should be to establish the facts concerning the danger of low relative humidity and low absolute humidity. (The effect on the explosion hazard due to agglutination of layered dust resulting from high humidity has never been examined.) Other factors to be considered are differences in electrostatic properties for different types of dust and the basic electrostatic characteristics of grain-handling machinery (i.e., when the machinery is operating but no grain is being handled). Considerable thought and care should be given to the design of experiments in this area since electrostatic phenomena in industrial locations are so elusive and ephemeral. The results of this research obviously should be accompanied by recommendations for the elimination of any electrostatic hazards uncovered.

OCR for page 43
60 FUTURE FACILITT E S Re commendations Treat'the avoidance of dust explosion hazards as an initial design criteria in the cons traction of new mills and elevators and the modification of existing structures. Examine the overall functions of mills and elevators to develop a totally new system less subject to the hazards of dust explosions. EffiCDCV Feasibility Bffici~ L M ~ H H - The first recommendation on the design of future facilities requires little discussion. New facilities should be designed to incorporate adequate dust control, 'to avoid dust generating operations, to facilitate housekeeping, and to be well vented. Design criteria should reflect these concern. so they are not considered only after the final design i. completed when any changes become expens ive . One problem to be considered in the design of elevators and mills is the "response. of the facility if, for- sane reason, there is a primary explosion. Thus, design criteria should consider' the isolation of sites where primary explosions may occur from those that may produce secondary explosions-. The use of outside legs, pressurized electrical vaults, and isolated dryers are e'xamples that readily cane to mind. Conservative des ign has been the rule in the grain-handling industry and although most new elevators and mills incorporate advances in technology (e.g., television surveillance, electrical interlocks, and dust collection systems), they still handle and process grain in fundamentally the same manner as has been used for the past 100 years. A study of the functions of elevators and mills (e.g., grinding, blending, and storing of grain) is needed to serve as the teas 'is for totally new elevator and mill designs that will reduce the explosion hazard without decreasing efficiency or increasing cots, both capital and operating. This study should be limited only by the fact that grain must be transported from the farm to the ultimate consumer with various processing operations occurring along the way. The rapid increase in grain production in the past 20 to 25 years, in a broad sense, changed the function of elevators from storage facilities to surge tanks in a pipeline. The gradual evolution of elevators to accommodate this change unfortunately carried along the hazardous features and, in sane cases, intensified them. Considering that ache grain-handling industry accounts for more explosion" than any other single industry, it would seem worthwhile to re-examine the entire grain-handling process.

OCR for page 43
61 GOVERNMENT REGULATIONS Recommendations for actions by governments at any level--federal, state, or local--in matters of hazard reduction cannot be ignored, especially in cases pertaining to employee safety. Government regulations applying to grain-handling facilities are no different in purpose than those applying to other industries having safety problems: they are intended to point out state-of-the-art practices that will provide a safe working environment and they serve as ~laws. whose willful violation will result in punishment that, in turn, will convince other possible violators not to follow the same path. Grain industry and government understanding of safe operating practices in elevators and mills' is minimal. Although large grain corporations will disagree vigorously with this point, there are literally thousands of elevators and mills operating as independent entities whose understanding of the hazard is at best limited to a knowledge that elevator explosions are fueled by grain dust. The federal, state, and local occupational safety enforcement agencies are in no better position to decrease the dust explosion hazard for a 'number of reasons. First, there are no regulations that apply specifically to elevators and mills. Second, the protection of elevators and mills from explosions i" only a small portion of their responsibility and! consequently, they have allocated only a small portion of their staff's time and effort to the problem. {The majority of contacts between industry and safety enforcement agencies occur either after the fact--following an explosion--or during infrequent safety inspections. The panel was privately advised that in one state the available manpower was such that only about 2 percent of the elevators and mills could be inspected each year.) Gird, animosity exists between industry and regulatory agencies and, whether for real or imagined causes, 'it is a hindrance to safety. Fortunately, some progress is being made in alleviating ' the contentious situation between industry and government (OSHA). This may result in progress' in reducing the explosion hazard. Since this panel's formation, OSHA has taken two positive steps. First, it has developed a training program for its explosion investigators to enable 'them to determine belter' the "causes of explosions. If these investigators consider their primary task to be a determination of cause, the mystery attached to explosions should be reduced and better relations with industry may result. Second, it conducted a series of meetings {New Orleans, Superior, and Kansas City) in 1980 on hazards in grain-handling facilities that demonstrated its willingness to accept industry's input. Industry, itself, cited the need for performance standards at these meetings. It is, however, too early to assess the results of this effort. Appendix D consists of a report on recommended standards prepared by this panel's & bpanel on Recommended Standards and Regulations. That report should be considered as a beginning step in formulating standards by a cooperative action between industry and government.

OCR for page 43
62 Changing the grain grading system so as to penalize for the amount of dust, with the objective of decreasing the explosion hazard, has often been mentioned. The panel is of the opinion that, besides meeting considerable objection from sellers, buyers, and handlers of grain, the definition of a standard solely for.this purpose is impractical. The only effect would be to pass part of the hazard cost back to the deliverer of the grain since the grain would be accepted, clean or dusty. It is recognized that the return of dust to the grain stream places a heavier burden on the downstream handlers of grain and their dust control systems. A standard, based on the assumption that the degree of cleanliness of the grain (as now delivered to elevators and mills).is directly proportional to the safety of the facility, ignores the hazard due to dust generation in the facility. REFERENCES . American Insurance Association, Special Lies Control Bulletin,.New York .City, N.Y., 1978. Canadian Grain Handling Association! Fire and Explosion Task Force Report No. 1, Winnepeg, 1979. . Cybulski, W., Coal Dust Explosions and Their Suppressions, NTIS TT 73-57001, National Technical Information Service, Springfield, Virginia, 1975. Eckhoff, R.K., Towards absolute minimum ignition energies for dust clouds, Combustion and Flame 24 (1975~:53-64. . Finn, W.D., Report of the Commission on Health and Safety in Grain Fn evators--Burrard Terminal Fire and Explosion of October 3, 197 5, University of British Columbia, Vancouver, 1976. Liebman, I., Carry, J., and Richmond, R., H2O Barriers for Suppressing Coal Dust Explosions, U.S. Bureau of Mines, Washington, D.C., 1976. Safety Consulting Engineers, Inc., Rosemont, Illinois, Electrostatic Characterization of-Grain Products, prepared for the National Grain and feed Association, Washington, D.C., October 1, 1980. Palmer, R.N., Dust Explosions and Fires, Chapman and Hall Ltd., London, 19 73. University of Southhampton, England, Study of Static Electricity on Grain Conveying Belts, prepared for the National Grain and feed Association, Washington, D.C., November, 1980.