Document wxewmNrg0Vzq161DdEdzvYkQ

FILE NAME: Engineering (ENG) DATE: 1935 Mar DOC#: ENG004 DOCUMENT DESCRIPTION: Trade Journal Article - Dust Control: Present and Future Design Considerations MECHANICAL ENGINEERING Published by The American Society of Mechanical Engineers ----------------- ----------------- -- -- ---- N umber 3 V olume 57 _____________ _------------------------------------------------------ ------- ----------------------- -- -- Contents for March., 1935 HEAVY HYDROGEN AND HEAVY WATER . THE STUDY OF TECHNICAL HISTORY . W OODEN STEAM BOILERS . STRESSES IN RUBBER T I R E S ............................................ DUST C O N T R O L ....................................................... STRESS ANALYSIS OF FAILURES IN MACHINE PARTS G. M . Murphy 141 I. N . Liphshitz 143 . F. A. Taylor 148 . H. Hencky 149 Theodore Hatch 154 . F. L. Everett 157 STEAM RESEARCH AT THE BUREAU OF STANDARDS . . . . N. S. Osborne, H. F. Stimson, and D. C. Ginnmgs 162 THE EQUATION OF STATE FOR SUPERHEATED S T E A M ........................................... < _ F. G. Keyes, L. B. Smith, and H. T. Gerry 164 RECENT DEVELOPMENTS IN STEAM TURBINES . . C. R. Soderberg 165 THE COMMERCIAL ASPECT OF E N G IN E E R IN G William McClellan 174 PALMER CHAMBERLAINE RICKETTS, 1 8 5 6 --1 9 3 4 177 e d it o r ia l . EN G IN EERIN G PROGRESS LETTERS AND COMMENT 137 A .S.M .E. BOILER CODE 179 REVIEWS OF BOOKS 191 w h a t 's GOING ON DISPLAY ADVERTISEMENTS 1 WHAT S N E W ....................................................... 14 INDEX TO ADVERTISERS PROFESSIONAL SERVICE . CLASSIFIED ADVERTISEMENTS . . 32 194 195 198 29 30 OFFI CERS OF THE SOCIETY! R alph E. Flanders, President Erik Obbrg, Treasurer C. E. Davies, Secretary PUB LI CAT ION STAFF! G eorge A. Stetson, Editor Frederick L ask, Advertising Mgr. COMMITTEE ON PUBLICATIONS. S. W. D udley, Chairman S. F. Voorhees G . F. Bateman W. F. Ryan h - RoaEa'rs A D V I S O R Y M E M B E R S OF T H E C O M M I T T E E ON P U B L I C A T I O N S : E. L, O h ib, St- Lodi,, Mo. E. B. . T O Bu , V .Junior Member, O. B. Schier, 2d A. J . Dickie, San Francisco, C al, DUST CONTROL Present and Future Design Considerations B y THEODORE HATCH HARVARD SCHOOL O F PUBLIC HEALTH FOR many years there has been a vague understanding of the relation between industrial dust and disease. Thus we find such names as " potter's ro t" and "miner's con sumption" antedating the relatively modern word "silicosis." In a few instances, notably in the South African gold mines, silicosis has been known for some time and has been accepted as a definite liability of the industry. In the United States also, certain sections of the mining industry were made aware of the dust hazard at an early date and a few sporadic studies were made in other industries, but with these exceptions work men and employers alike were not generally cognizant of the danger of dust. This is especially true of the " new" indus tries in which the disease has not had time to establish itself as a tradition. Although scientific methods for the detection of silicosis and requirements for its control have been developed, there has been no comparable progress toward its elimination. A review of state regulations of the dusty trades shows that they do not meet the present requirements of the problem and that satisfac tory results cannot be expected with the methods of inspection and evaluation of control equipment now generally employed. Workmen's compensation acts did not originally provide for payment for incapacity resulting from occupational dust diseases, and at the present time only five states allow compen sation for silicosis. Even in these one may say th at it secured its place by default, for in no case is the disease specifically mentioned. In the face of this unsatisfactory situation there has been a tremendous increase in the number of claims during the past few years. Most of these have been presented for settlement under the common law and thus the complicated and highly technical details of the problem have come before lay juries for consideration and judgment. Conflicting testimony has been given by medical and technical witnesses and the whole situation has become more confused as a result. It is evident that, the legal aspects of the dust problem need revision and velopcd beyond the stage of trial and error, and the design of structures for the suppression of dust is based almost entirely upon empirical data together with one or two elementary for mulas. The latter were derived from the generalized experi ence in a limited number of industries, such as woodworking, where conditions and requirements are not at all like those, for example, in a sand-pulverizing plant. Wide experience and good judgment arc essential to the proper use of the scanty in formation that has been collected and systematized. Without attempting to minimize the importance of experience in engi neering design, it is necessary to point out that most of the past experience obtained in the empirical design of dust-control equipment was not evaluated by means of quantitative studies of the actual reduction in dust. Hence, past experience does not necessarily furnish the proper guide for future design. In fact, recent surveys in certain industries, employing modem methods of investigation, have shown that existing equip ment is in many cases inadequate. Dust control, as practiced today, is to some extent the off spring of dust collection, but since they are opposite in purpose, the requirements of one are not necessarily like those of the other. Thus, manufacturers of dust collectors may guarantee collection of 99Vs Per cent and in the production of a com mercial powder this is entirely satisfactory, but it has no mean ing in the performance of dust-control equipment. It is the escaping dust, not that collected, which is important, and guarantees of performance must be based upon permissible amounts of dust at the breathing zone of the workmen and in the " clean" air discharged from the equipment. For example, assume a safe limit of dustiness in the air discharged from a certain plant to be 1 mg per cu m or 0.000044 grain per cu ft and assume also a dust loading of 30 grains per cu ft. Under these conditions, which are not abnormal, the efficiency of dust "collection" must be 99.9985 per cent! PRESENT STATE CODES INADEQUATE clarification, but regardless of the merits of settlement under common law compared with operation under a compensation act, the fact is that the hygienic and economic aspects of the prob lem can be settled only through permanent elimination of dust from industry. It is desirable, therefore, to inquire into the present status of engineering design of dust-control equipment. Specifications in state codes for the design of exhaust hoods are commonly written in terms of the " static suction" to be maintained at the throat of the hood. Contrary to a wide spread idea, suction per sc has no effect in the capture of dust. The use of this index arises from the fact that it constitutes an approximate measure of the rate of airflow into a suction hood. PRESBNT STATUS OF EN G IN EERIN G DESIGN OF DUST-CONTROL EQUIPMENT Textbooks on ventilation and engineering handbooks devote little or no space to methods of control of industrial dust, and in the technical journals, papers on fundamental engineering design of equipment are conspicuous by their absence. In a recent bibliography of the literature on pneumoconiosis, less than one per cent of the articles dealt with the engineering con trol of the disease. The science of dust control has not de- Contributcd by the A.5.M.E. Safety Committee and presented at the Annual Meeting, New York, N. Y., December 3 to 7, 1934, of Tub A m erican Society of M echanical E n g in eers. This is true, however, only when the area of the throat and the size and shape of the hood are also given. In fact, as the fol lowing equation shows, these two factors are more important than the static suction, since the latter operates only at the one-half power Q = 400O A fV h where Q = rate of airflow at 65 F, cfm; A = area of throat, sq ft; h = static suction at throat, inches of water; and / = coefficient of restriction, which varies considerably with hood design. Increasing mortality rates among workmen exposed to harm ful dusts demonstrate the need for changing the regulations. 154 M a r c h , 1935 155 In Wisconsin the new general orders issued by the Industrial Commission require hazardous plants to maintain atmospheric conditions below certain permissible standards established by recognized authorities. Thus structural details of control equipment arc made secondary to results. Similar require ments have been incorporated in the proposed codes for other states. But w ith their adoption, it must be made clear that permissible values of dustiness are not absolute and may change as our knowledge increases. State requirements must therefore be kept flexible to allow for revisions dictated by new evidence and changes in " good practice." In order to facilitate this work and secure uniform requirements throughout the country, a national board is proposed whose purpose would be to review evidence and methods and prepare at stated intervals, for the assistance of state officials, standards of good practice for vari ous dusty processes and industries. The systematic correlation of actual results w ith the details of design, coupled w ith an understanding of the physical prop erties of dust and the laws governing its dispersion and behavior in the air, will provide a body of facts upon which to base the engineering design of equipment. Some of the fundamental factors arc considered in the present paper. (2) Clouds: Particles 0.1 to 1.0 micron, settle with uniform ' velocity, according to Stokes's Law. (3) Smokes: Particles 0.001 to 0.1 micron, do not settle in air. Drinker (3) has modified Gibbs's classification and combined the idea of source of material with the degree of dispersion. It is more descriptive and conveys more meaning although it is physically less exact than Gibbs's grouping. (1) Dusts: Particles 1 to 150 microns, generated by mechani cal agencies such as crushing, grinding, drilling. (2) Fumes: Particles 0.2 to 1.0 micron, resulting from chemi cal reactions such as distillation, complete and incomplete oxidation of metals. (3) Smokes: Particles less than 0.3 micron, usually resulting from incomplete combustion of carbonaceous matter. These classifications are valuable when considering methods of air cleaning and other problems related to dust control, since the method of treatment suitable to one class of suspen sion may be of no value in the case of another. Cyclones, for example, are valuable for the capture of coarse dusts, but de spite the claims sometimes made for them, they cannot be de pended upon to remove the fine material of hygienic importance. PHYSICAL PROPERTIES OP DUSTS MOTION OP PARTICLES IN AIR Suspensions of dusts and fumes in the atmosphere, although possessing greater particle size, may be compared to a certain extent w ith colloidal systems. Gibbs has given the name "aerosol" to the dispersed system in which solid or liquid par ticles are dispersed in a gas. Aerosols differ, however, in their physical properties from hydrosols (dispersing medium a liquid). First, the settling rate of the dispersed material is very much higher, and second, molecular activity and therefore Brownian motion is greater. Consequently, the particles col lide more frequently and the tendency toward flocculation is increased. Finally, aerosols differ in their electrical proper ties. Some particles carry positive charges, some negative, and some may be neutral; in a liquid medium, on the other hand, all particles generally have the same charge. The formation of aerosols results from processes of dispersion including mechanical grinding, crushing and shattering, ex plosions, etc. Energy must be introduced to overcome the force of cohesion of the original material and the amount of energy required is roughly proportional to the area of new sur face formed. The size of the particles produced depends upon the manner in which the shattering force is applied and also upon the "geological structure of the material being crushed. Thus Jones1 (1) has pointed out that in the South African "banket," the fibrous sericitc, which occurs as a cementing material between the particles of quartz, is more easily broken down during drilling and blasting than the solid grains of silica and appears in the dust, therefore, in smaller sizes. The physical properties of a material may be greatly altered Particles suspended in a gas are subject to two forces--exter nal and internal. The former includes gravitation, centrifugal force, and electrical attraction. Internal forces result from molecular activity of the gas and from electrostatic charges carried by the particles. External forces acting upon particles suspended in air cause them to move through the air with velocities that vary with the size, shape, and density of the particles, and with the density and viscosity of the air as well as with the strength of the external force. The resistance to travel of spheres through a motionless gas may be written (4) R = kanPy (2 ~ n)F*, where a = radius of the particle; p = density of gas; y = kinematic viscosity of gas; V = velocity of travel of particle; k = constant. In the region of streamline motion = 1.0 and hence R = kapyV. This is the equation developed by Stokes for the resistance of small particles moving at low velocities. The resistance varies directly with particle radius and velocity. Large particles traveling at high velocities, on the other hand, set up turbulent motion. In this region n = 2.0; the resistance to motion be comes a function of the squares of velocity and particle diame ter, and the viscosity effect is eliminated, thus R = ka2pVh. Under the attraction of gravity, suspended particles fall with increasing velocity until the gravitational force F =* 4 /3 ^ * (<r -- p)g is just balanced by the resistance R. At this point acceleration becomes zero and the settling velocity reaches a constant value which may be determined for the two forms of motion by the following equations from the original when it is finely ground and dispersed. The surface area and specific surface are enlarged many times and the V= ^ ^ ' (streamline)............... [1] rate of evaporation and solution as well as the adsorptive ca v" [ ( 3?) C ir ) al (turbulcn ...... [2] pacity are thereby greatly increased. Chemical activity also increases, since there is a larger proportion of unbalanced mole cules lying at the surface. Aerosols differ also from one an other in their physical properties, depending upon the degree Theoretically and experimentally, it has been shown that par of dispersion. It is convenient, therefore, to distinguish be ticles of quartz under approximately 75 microns in diameter tween different classes of aerosols. Gibbs's classification (2) settle according to the first equation, and that the particles is based upon the settling rate of particles in air: must be greater than 2 mm in diameter to obey the equation of (1) Dusts: Particles greater than 10 microns, settle w ith in t u r b u l e n t m o tio n * Between these two limits IS the region in creasing velocity. which, according to Allen (loc. cit.), the resistance varies 1Numbers in parentheses refer to bibliography at end of paper. with velocity and particle diameter to the 3 /2 power. 156 M ech anical E n g in e e r in g DISPERSION OP DUST INTO THE ATMOSPHERE Dust particles, generated by whatever means, require energy for their dispersion from the point of origin into the surround ing atmosphere. Physically, this is provided in only two ways: (1) Dynamic projection due to the kinetic energy of the par ticles themselves, imparted to them during their formation. (2) Dispersion by means of air currents created by the opera tion of the dusty process. The counter forces set up by the exhaust hood or other con trol device for the purpose of preventing the dispersion of dust must therefore be directed, in the first case, against the energy of the dust particles, and, in the second, against that energy which produces air currents around the machine. Since these forces of dispersion are quite different in form, it is clear that proper design of equipment depends upon an understanding of their relative magnitude and the manner in which they act. The first lends itself to approximate mathematical analysis which follows; the second depends upon the action of the dustproducing machine, and its evaluation requires careful analysis of the details of operation of the process. DYNAMIC DISPERSION A particle w ith mass m thrown off with an initial velocity Vo possesses the kinetic energy K.E. = | mVo2 and is resisted, as it travels through still air, by a force which varies w ith the type of motion and is determined by the equation R = k a W 2 ~ n)Vn Neglecting the force of gravity, the particle will travel along its path w ith a decreasing velocity until at a certain distance T the kinetic energy will be destroyed, i.e., its velocity will be zero. Thus 7 ro ro RdS = \ m V >J = / -- mVdV v Jr0 Within the region of Stokes's law, R = kV, and therefore which, after integration, becomes V = V0 -- k'S. The value of k! (k! = 3 X 10_4/<*2for spheres) varies inversely w ith the square of the particle diameter; doubling the size increases the length of travel four times for the same final velocity. For a 10-micron particle having an initial velocity of 5 X 101cm per sec (10,000 fpm), S = 4 cm when V = 0. If the particle travels w ith turbulent motion, however, the resistance varies w ith V 2and we may write and from this V = V0e~ k's Here k' = 3 X \0~i/a for rounded bodies. The particle must travel an infinite distance to reach zero velocity but will, on the other hand, attain a low finite value of V in a finite distance. A particle w ith a diameter of 2 mm, for example, projected w ith the same initial velocity as in the previous example (10,000 fpm), will have a velocity of only 10 cm per sec (20 fpm) after traveling approximately 2000 cm or 67 ft. These equations are not absolutely correct nor do they hold over the entire range indicated. The calculated values- of S arc therefore not exact. Nevertheless, they are correct in their relative orders of magnitude and to this extent they dem onstrate an important principle, namely, that whereas large particles may be dispersed by dynamic projection through still air, microscopic particles, because of their relatively enormous surface area per unit volume, do not travel any appreciable distance by virtue of their own kinetic energy. This principle is well illustrated by the action of a grinding wheel. Large particles arc thrown off by centrifugal force and may be projected a considerable distance through still air, as the path of incandescent particles so clearly indicates. This force is not great enough, however, to disperse the fine par ticles. Some may be carried in the air stream created by the ' 'drag'' of the large particles and thus escape w ith them. To a greater extent, however, their dispersion is caused by air cur rents set up by the fan action of the rotating wheel. It follows, therefore, that an exhaust hood placed in the path of the in candescent large particles may not be sufficient, for this neglects the fine (and non-luminous) dust that is dispersed with es caping air currents around the periphery of the wheel. The concept of the dual function of an exhaust hood, which arises out of the foregoing analysis, suggests that for successful dust control, three things must be done: (1) The kinetic energy of the large particles must be de stroyed. These particles are not important hygienically but they must be captured in order to prevent the escape of fine particles in the air stream created by their " drag." Their energy can be destroyed by means of barriers against which the particles impinge and lose their velocity, and the air currents induced by their motion can be dissipated by setting up counter currents in the zone of dust generation by means of air flow into the exhaust hood. (2) Air currents created by virtue of the operation of the dusty processes must be eliminated through changes in the machine, or more generally by the application of suitable baffles and housing to prevent their formation. (3) Air currents around the machine that cannot be elimi nated must be changed in direction and made to flow into the exhaust system. This is accomplished by means of the poten tial gradient established between the suction opening and the surrounding air. Through hood design, the gradient must be maintained at the proper slope from the area of dust-production and at a minimum slope from the so-called ineffective areas in which no dust is produced. A knowledge of the aerodynamic characteristics of suction openings ( 5) is therefore essential to proper design. Fundamental specifications for exhaust hoods cannot be es tablished simply from a consideration of these factors. Ex perimental investigation must remain the most valuable aid to design. Nevertheless, the creation of a real science of dust control, which has been made necessary by the widespread in dustrial dust hazard, must be based upon physical laws. Em pirical methods cannot be depended upon to relieve the heavy financial load that the dust hazard has placed upon industry. BIBLIOGRAPHY (1) "Silicosis," W. R. Jones, Institution of Mining & Metallurgy of London, Bulletin, January, 1934. (2) "Clouds and Smokes," W. E. Gibbs, P. Blakiston's Son & Co., Philadelphia, 1924. ( 3) "The Use of Owens' Jet Dust Counter and of the Electric Precipi tation in the Determination of Dusts, Fumes, and Smoke in Air," P. Drinker and R. M. Thomson, Journal, Am. Soc. Heating and Ventilating Engineers, 1924, vol. 30, p. 695. (4) "On the Motion of a Sphere in a Viscous Fluid," H. S. Allen, Phil. Mag., 1900, vol. 50, p. 323. (5) "Determining Minimum Air Velocities for Exhaust Systems," J. M. Dallavalle, Journal, Am. Soc. Heating and Ventilating Engineers, 1932, vol. 4, p. 639.