Document zbMEZG63nXg4xg5R2114BXQxg

MECHANICAL ENGINEERING Published by The American Society of TS/lechanical Engineers 1933 Volume 55 Number 10 Contents for October, 1933 THE COVER...................................................................................................BAYONNE BRIDGE the problem of purchasing power .... R. E. Flanders 595 measuring our capabilities..........................................................C. W. Squier 603 economic lot sizes in manufacturing . . . P.T. Norton, Jr. 60S problems in fast air-transport design . . . H. L. Hibbard 611 INDUSTRY AND GOVERNMENT...........................................................................................618 THB CONTROL OF INDUSTRIAL DUST.................................J. M. DallaVallt 621 CORRELATION OF METAL-CUTTING DATA.................................R. C. Deale 625 PLAINTIFFS | EXHIBIT I ASM-6 EDITORIAL........................................................ SURVEY OF ENGINEERING PROGRESS . SYNOPSES OF A.S.M.E. PAPERS . . . 628 630 643 CORRESPONDENCE................................................. 645 BOOK REVIEWS..........................................................652 WHAT's GOING ON..................................................655 DISPLAY ADVERTISEMENTS PROFESSIONAL SERVICE . 1 CLASSIFIED ADVERTISEMENTS ... 26 25 INDEX TO ADVERTISERS..................................28 OFFICBRS OF THB society: A. A. Potto., President Fjtt Obkro, Trtstmw Calvin W. Rick, Secretary PUBLICATION STAFF: Gioioi A. SmrrsON, Edit# Fuduuck La&c, Advertising Mgr. COMMITTEE ON PUBLICATIONS: L. C. Morrow, Chairman S. W. Dudley S. F. Vooihbes, Vice-Chairman M. H. Roberts W. F. Ryan C. E. Davies, Secretary to Committee on Publications ADVISOB.Y MEMBERS OF THE COMMITTEE ON PUBLICATIONS: E. L. Oklb, St. Louis, Mo. E. B. Norris, Blacksduro, Va. A. J. Dickie, San Francisco, Cauf. Mechanical Engineers. Publication office at 20th and Northampton Streets, Easton, Pa. Editorial sod Advertising departments at the head* New York, N. Y. Cable address, "Dynamic," New York. Price 60 cents a copy, $5.00 a year; to members and affiliates, 50 cents a copy, _____i jKai. Pottage to Canada, 75 cents additional, to foreign countries, $1.50 additional. Changes of address must be received at Society headquarters two weeks before they are to be eftedse on the mailing Iwt. Pleaae send old as well as new address... By-Law: The Society shall not be responsible for statements or opinions advanced in papers or ... printed in its mdihrsTirwit (B2, Par. JX Entered as aceood<Jaas martcr at the Post Office at Easroo, Pa., under the Act of March 3,1079.... Acceptance for mailing at special race of postage provided for in section 1103. Act at October 3.19U. authorised on January 17,1921- Copyrighted, 1933, by The American Society of Mechanical Engineers. The Control of INDUSTRIAL DUST The Problem of Local Exhaust and General Ventilation By J. M. DALLAVALLE1 T IS WELL KNOWN that the inhalation of certain LOCAL EXHAUST CONTROL I industrial dusts is an important factor in causing Local exhaust control depends upon the provision of occupational disabilities. Dust such as granite has adequate hoods and air velocities. The function of a been shown to produce a fibrosis of the Jungs which ishood is to direct an air movement of sufficient intensity often associated with tuberculosis, while other dusts, to prevent the escape of dust particles and to carry them such as asbestos, cement, slate, ecc., arc stated to pro to a collecting system. This fact applies to all hoods, duce a varying degree of fibrosis. In general, it has and although not always obvious, should be kept clearly been found that dusts whose quartz content is high cause in mind. a disabling fibrosis more readily. Hence, in view of the The problems involved in the design of hoods arc hazards which may prevail by the inhalation of quartz- many. There are few types of hoods which are designed containing dusts, the problem of control is extremely to handle sufficient air for the elimination of dusc, a important. Previous papers1 given before this section condition which arises from two sources: (i) a lack of the Society have been concerned with the effects of of knowledge of the nature of the air flow into the hood, exposure to industrial dusts and the methods used for and (z) insufficient data pertaining to velocities neces their quantitation. The following paragraphs are, sary to set dust in motion and to bring it under control. therefore, designed to supplement those papers and to In the following paragraphs, we shall therefore attempt present fundamental considerations with regard to the to discuss the velocity characteristics of hoods and their control of the dust hazard. application to the design of exhaust equipment. The quantity of dust produced at a given source de pends upon the type of operation performed and the HOOD VELOCITY CHARACTERISTICS method used for its control. In many cases, such fac tors as speed of production, plant layout, and the ap plication of dry or wet methods in certain operations have a direct bearing upon the quantity of dust generated. The effective removal of dust most consequently take A fact little understood is that the effects of drawing in air in the vicinity of an open duct or hood differ vastly from those of blowing out the same air. In the latter case, the condition is not unlike a stream of water issuing from a nozzle; the stream is well defined and in into account the manner in which it is produced. In tact for a considerable distance from the opening, while some instances, where dust is restricted to a small area, in the former, no such effect is noted. This may be local exhaust ventilation is generally used, while in demonstrated by merely placing an opening under suc places where dusty operations arc conducted on a large tion and noting the quickness with which thd air move scale, as in foundries and batch-mixing rooms, isolation with general exhaust may be preferable. In every case, a careful study and analysis of the problem is necessary so that dust concentrations may be reduced to safe ment diminishes as the hand is drawn outward from it. To illustrate the point further, it may be shown by ac tual measurement that the air velocity at a point one inch outward along the axis of a duct 4 in. in diameter limits. Many attempts to eliminate industrial dust is approximately one-half of what it is within the duct hazards have been unsuccessful, partly because of the itself. In other words, an air velocity of 4000 fpm in lack of attention given to the principles governing the design of control equipment and partly because of fail ures to utilize proper methods in testing the efficiency of dust removal. The necessity of giving due consideration the duct is reduced to 2.000 fpm one-quarter of a pipe diameter outward. This rate of decrease is explained by the fact that the air is drawn from all directions. Thus, if at one of the openings we circumscribe a suc to these important factors is consequently clear. cession of spheres, the areas are as the squares of their radii. The velocities are then inversely proportional 1 Assistant Sanitary Engineer, U. S. Public Health Service, Washing ton, D. C. 1 "Dust in Industry," by A. E. Russell, Mbcbafticai. Enoinhbrino, March, 1933, pp. 163-166. "Dust in Industry," by J. J. Bloomfield, HU., April, 1933, pp. 229-233. Contnbuted by the Process Industries Committee and presented at the Semi-Annual Meeting, Chicago, 111., June 23 to July 1, 1933, of Tin Awejican Society of Mechanical Enoinhbss. to the areas of the spheres; that is, as the areas increase the velocities decrease. Since the same amount of air must pass through each of the circumscribed spheres, it follows that the velocities are inversely proportional to the squares of the radii. This reasoning neglects the effect of the duct itself, but is substantially true when 621 622 Mechanical Engineering the distance from the opening is very great in comparison with the duct diameter and when there is no obstruc tion in the field of influence. It has been found as a matter of experiment3,4 that the velocity at any point along the axis of a hood under suction is closely given by the equation Y _ o.iA ioo -- Y X2 [ia] or 0.1AV0 X2 +0.1A [ib] where Y is the percentage of the velocity at the point X along the axis in terms of the velocity at the opening TABLE 1 VELOCITY CHANGE ALONG AXES OF HOODS OF VARIOUS SIZES WITH CONSTANT AIR FLOW OF 1000 CFM ^Determined f from ., Y tormula100_y- 0.1 A\ x, j Average Area of velocity opening, at *------ Distance outward along axis---X------. sq ft opening, fpm 6 in. 12 m. 18 in. 24 in. 30 in. I 1000 286 91 43 24.4 15.9 2 500 222 83 41 23.8 15-5 4 250 154 71.5 38 22.8 150 6 167 107 62.5 35 21.6 14-6 8 125 95 52.5 33.1 20.8 14.2 10 100 80 50 31 20 13.8 20 50 44.5 33 23.7 16.7 12.1 opening and are purely functions of the shape of the hood. Further, the velocity contours arc identical for similar hood shapes when the hoods arc reduced to the same basis of comparison. These important facts arc applicable to all hood problems and from them it is possible to estimate the volume of air necessary to control a given dusr source. Figs, i and z. show the contour distribution in half-plane sections of two simple FIG. I CONTOUR DISTRIBUTION WITH SQUARE OPENING Vo and A is the area of the opening. Now at great dis tances, Y is small in comparison with ioo in the de nominator on the left-hand side and consequently may be called zero. Equation [ia] then reduces to the form as already determined from elementary geometry; that is, Y (or V) is inversely proportional to the square of X. Hence, from the equation, it is at once possible to determine .the air velocity at any point along the axis of a free and unobstructed hood, given the ajr flow. Table i shows the variation of axial velocity with area of hood opening when the volume of air flow is kept constant at iooo cubic feet per minute. This table illustrates the small change of velocity at great distances from the opening. VELOCITY CONTOURS It is possible by the use of a specially constructed pitot tube* to map contours of equal velocity in any axial plane located in the field of influence. It has been found that the positions of these contours for any hood can be expressed as percentages of the velocity at the hood * "Studies ia the Design of Local Exhaust Hoods," by J. M. DallaValle and Theodore Hatch, Trans. A.S.M.E., vol. 54 0932), paper no. WDI-54-10. 4 "Velocity Characteristics of Hoods Under Suction," by T. M. DallaValle, Htating, Piping and Air Conditioning, vol. 4 (1932), pp. 370-375. FIO. X CONTOUR DISTRIBUTION WITH CIRCULAR forms of openings. The contours are typical of all 1 under suction and show the variation of velocity < the region of influence. The curves drawn ' lar to the contours at their points of intersection1: the lines of flow and give the direction of the air: Hatch, Drinker, and Choate* have applied the i of velocity contours to the design of hoods granite surfacing machines. They were enabled' development of these contours to find the air vein necessary to control the dust generated by these: ing tools and to determine the rates of air flow fore type of hood. The economic advantages of types were shown to depend upon their shapes wh pointed out, directly influenced the contour disc The development of contours for ail hood types* l "Control of the Silicosis Hazard in the Hard-Rod; I--A laboratory study of the design of dust-control systeaa i with pneumatic granite-cutting tools. By T. Hatch, P. f ' " S. P. Choate. Jour. Ind. Hjgimt, vol. 12 (1930), p. 75. October, 1933 623 valuable, often involves many difficulties. It is neces sary not only to consider the shape of the hood, but also the obstructions within the zone of influence. These obstructions frequently alter the path of the air move ment and distort it to such a degree as to render the hood ineffective in removing the dust. Air, like water, ccntration to the safe limit of io million particles per cubic foot. This is the dust concentration which has been found to produce no disabling fibrosis in the granite industry.7 The same method of correlating dust counts with air volumes had been previously used by Bloom field8 in field studies conducted among granite plants in moves over paths offering the least resistance. Vermont. It is interesting to note in this connection A typical obstruction often encountered is illustrated that Bloomfield was able to show that, under the con by a hood placed over a sand milling machine used in ditions in which granite surfacing hoods were tied, foundries. This case may be investigated by use of con an air velocity of 1500 fpm at the opening was necessary tours, but it is capable of a simple analysis which may to keep the exposure within a safe limit. For the par frequently be applied to other hoods whose area of in ticular hoods used, the air volumes handled were suf fluence is obstructed. When the milling machine is placed close to a hood, the air flow is almost entirely from the open sides between them. If, then, we disregard the air flowing from behind the edges of the hood, we may consider all the flow as taking place laterally inward. The open area may then be obtained approximately by multiplying the perimeter of the tank by the distance from the sur face of the tank to the plane of the hood opening. We then have to measure the volume of air flow into the hood, which may be ascertained either by an anemometer placed in the throat of the hood or by means of a pitot tube. This volume of air will be the same as passes through the open sides between the hood and the mill. It follows, therefore, that ficient to maintain air velocities at the cutting tool ap proximating those obtained by Hatch. The importance of quantitative measurements is fur ther emphasized by other results obtained in an investi gation of the Kelly-Hatch dust trap used in rock drilling. Since this method of dust control is not adapted to any analysis of velocity characteristics, it was necessary to determine the relation existing between the air volumes handled by the trap and the dust concentrations under varying operating conditions. In this way, it was found that a volume of 60 cfm was required to keep the dust formed to less than 3 million particles per cubic foot of air or well under the safe limit of exposure. SUCTION STANDARDS Volume of air Q = perimeter of mill X distance X velocity across open sides Hence by measuring the volume of air flow into the system and by knowing the dimensions of the mill and its distance from the hood, it is possible to deduce the average velocity across the open area. This velocity determines the effectiveness of the hood in controlling the dust arising from the mill. A more exact formula6 An important consideration entering into a discussion of local exhaust control is the present method of evaluat ing its performance. It is customary in most local ex haust work to express the characteristics of a given hood in terms of the suction maintained at the throat. It is important, however, to observe that suction measure ments only determine the air volume through the hood, as may be seen from the equation for determining the velocity close to the edge of the mill is given by the equation = M PdV Q = 4000 a / VA........................ [z] where Q = the volume of air flowing into the system, cfm where Q is the air flow, P the perimeter of the mill, and d the distance between the hood opening and the mill. The factor 1.4 is a correction to take care of the uneven distribution of flow. It is assumed, of course, that the hood is of the same shape as the mill. If the hoo4 is a = the area of the throat of the hood, that is, at the connecting duct, sq ft / = a restriction factor generally equal to 0.7 h = the suction measured by a U-tube, in. of water slightly larger, as is generally the case in practise, the In Table z the air flows for various-sized connecting formula can be used with greater assurance. ducts arc given with varying suction requirements. As EFFICIENCY OF DUST REMOVAL While the estimation of velocity characteristics has many inherent advantages, it is, nevertheless, necessary to supplement it with quantitative measurements of dust concentrations. In this way, it is possible to obtain correlations between air volumes and dust counts. In the study of the granite surfacing machines previously cited, Hatch and his coworkers6 showed that the particu lar hoods developed by them required approximately is well known, these requirements differ from state to state and vary from 1.5 to 5 in. of water displacement as measured with a U-tube. One of the chief difficulties of a high suction is the losses incurred by the high air velocities in the ducts. A suction of 5 in. in any duct is equivalent to a velocity of about 6000 fpm. It is ob vious, therefore, that the losses incurred are great and 7 "The Health of Worker! in Dusty Trades." II--Exposure to sili cosis dust (granite industry). By Albert E. Russell, R. H. Britten, L. R. Thompson, and J. J. Bloomfield. Public Health Bulletin No. 187, 300 cubic feet of air per minute to reduce the dust con- "The Guide." American Society of Heating A Ventilating Engi neers, 1933, Chapter 37. July, 1929. , "A Study of the Efficiency of Dust-Removal Systems in Granite Cutting Plants," byj. J. Bloomfield. Public Health Reprint No. 1324, October, 1929. 624 Mechanical Engineering even excessive in a pipe from i to 4 in. in diameter, as may be seen from Table 2. which is based on duct lengths of 10 ft. It would appear, since air volume determines the effectiveness of a hood, that the most efficient way to obtain air volumes corresponding to 5 in. of suction is by using larger ducts and lower suction. Thus, if a duct 3 in. in diameter is replaced by one 5 in. in diameter, the suction requirement would be reduced from 5 to 0.65 in. for the same air flow. The saving in power would be correspondingly great. Likewise, the air velocity would be one-third as large, which would en tail a further saving of energy losses. It is important to remember, however, whenever this type of reasoning is employed, that the pipe diameters are not increased so much as to nullify the conveying force of the air in the duct. The velocity should always be sufficient to carry away all the particles collected. There is rarely any need for air velocities approaching 6000 fpm. Therefore, unless it has been predetermined that a stated suction is sufficient for controlling a dust hazard, no measure of hood performance can be obtained by use of the term. If it is desired to express axial velocities in terms of suction, Equation [1] may be substituted in Equation [ib] for this purpose. Suction by itself and with no reference to the type of hood used or its location with respect to the dust source may result in serious errors. GENERAL VENTILATION The use of general ventilation for dust control is ex ceedingly complicated. The adaptability of the method depends upon a multitude of factors ranging from local conditions affecting the distribution of air movements in a building to consideration of heat losses incurred during the winter seasons. General ventilation, never theless, is an important factor in control where dusty sources are widely distributed as is the case in foundries, crushing and milling mills, and batch rooms. In these cases, dust exposure is general and involves large num bers of workers in non-dust-producing operations. TABLE 2 AIR FLOW CORRESPONDING TO SUCTION MEASURE MENTS IN HOODS WITH CONNECTING DUCTS OF VARIOUS DIAMETERS Suction, Duct diameter (inches) in. HjO 2 3 45 6 7 S 1 61 137 244 381 549 748 977 2 86 194 360 540 770 1060 1380 3 106 239 424 660 950 1300 1690 4 122 274 488 762 1098 1496 1594 5 137 306 550 854 1230 1670 2180 6 150 336 600 935 1350 1840 2390 OCCUPATIONAL SURVBY Before any attempt is made to utilize general ventila tion, it is first necessary to conduct a sanitary and occu pational survey. The method of making such a survey has been outlined by Bloomfield' and consists of an analysis of individual dusty operations. For example, in a foundry the dusting of the molds with parting com pounds, sand milling, and shakeouts should be inde pendently quantitated in terms of dust-particle-hours. Thus, if parting compounds used intermittently for a total of, say, one-half hour each day per man create 60,000,000 particles of dust, it would be said that the exposure was 30,000,000 particle-hours. The same analysis would apply to each source, and an estimate of the actual extent of dust hazards determined. The average number of particle-hours of all sources has been found by Bloomfield to check closely with the average plant dustiness. Having conducted a sanitary and occupational survey such as has been described, the next step is to attempt to segregate or control by local exhaust methods where possible the more dangerous dust sources. There can be little benefit, for example, in exposing all the workers in a foundry to the high concentrations produced by two or three men at the shakeout dumps. The same applies to pottery workers during dry batch-mixing operations. Hence, unless some direct means of control can be ap plied, some effort should be made to segregate the dan gerous dust sources. Frequently, general dusty conditions can be bettered by proper housekeeping super vision. It is a peculiar characteristic of the dusty trades to neglect this matter. The care with which plants are maintained may have an important psychological effect upon workers producing dust, and may result in greater care in the performance of duties. No specific details of the volumes re quired to obtain safe concentrations in a room can be given. There is need for a considerable research in this phase of the dust-control problem. At present it is suffi cient to point out that quantitative mea surements of dust concentrations under varying conditions of control form the most satisfactory approach to individual, problems. Cevritty JVnrfnn Company ' Daci to be published io Public Health Reports.