Document 0myp9zd0GRrrZqk60V8GYndb

^ r -D V" d p PP 0 CO 3 to f--i G *G G CO G 0 O P O o 0 > <o P r^ p g H 0 Pp G 0P CO 3 0 g; P TO H pP C H to 0 G P 0A G O o -p P > fa CD 'd .G o P d P 0 p d 0 0 CO A to Gd P >> CO 35 r^ 0 P G P *G G <-t p X H p 0 P d 0 Ca G o G CD O 0 p (' P . 0 P P o H P. d CO O u P d a* 0 Pd to A co CO A CO to .0 0 o H o a 5g H 0 P to 2 X Gp G 0 0) 2 2 o p to cO P o n CO P to P .G G 0 CO l*< p 0 P Q 0 g 'd Po to G o c P 0 G ft p pH 3 0 'd P< < 0 o G tD g r*4 P o G p H 1 p o 0 r^ P I CO J3 P .G G P G P CO o P CO to 0 o 0 to ts 0 > A p o d P 0 o TO G -d to G CO p G Oh G H < J3 rt P, p .G w <-< 2 0 P O 0 G 30 co P 3 p G (3 0 0 CO 0 P P C H o 0 P G O p P 0 'O G d o fa 0 G 0 0 d pG to O 0 >> i> p, 33 < P ^3 0 P CD PP G GP d P fit A CO co do p OOP G 3* P H Vft c G o 'd Eh P G TJ 0 d PO 0) 0 CO p 0O (0 S3 P to 0 erj o r4 aGG G 1"t o co G p Ph p o c H O G e w x: H 0 0 CO ' W -4 p > p 0 a ip CO G CO o> P rH p sG to 0 0 n <r> p P cP 3 p p c *d G G <P G 0 0 0 Op A fa 0 > > d >> c p 0p G rH fa u 0 Pi p a d fa 2 O o P G G> i 'P di> cr INDUSTRIAL HYGIENE FOUNDATION OF AMERICA, Inc. Prevent ire Knn-inrrrin^ Series, Bulletin A'o. 3. P:.. .* 1 Air Flow Measurement by the Dilution Method PITTSBURGH, PENNSYLVANIA JUNE, 1942 FOREWORD Upon recommendation of its Preventive Engineering Committee, Industrial Hygiene Foundation authorized a comprehensive study of exhaust ventilation. A grant was made to the University of Pennsyl vania, where work is now being carried on under supervision of Professor Theodore Hatch, ably assisted by Robert H. Walpole, Jr. Reports will be issued from time to time, so that members of the Foundation may benefit while the study is in progress. The reports will be in the form of bulletins in the Preventive Engineering Series; the first is on AIR FLOW MEASUREMENTS. The fundamental factor which determines the rate of exhaust ventilation for certain industrial processes is the rate of air displace ment produced by the process itself. As a starting point, therefore, in the development of a basic approach to the design of exhaust sys tems, it is desirable to measure the rate of air displacement produced by the process which is to be ventilated. In general, the air displacement occurring in a given process takes place in such a way that the ordinary air flow meters are not appli cable. It becomes necessary, therefore, to utilize a different method of measuring air flow. The method described in this bulletin was developed to meet these needs, and is capable of measuring the rate of air flow in a complex system which has multiple air inlets and outlets. A major advantage of the method is that it is not necessary to measure the amount of air escaping from every opening. Practical field tests have indicated that the technique offers a useful and practical approach to the determination of basic exhaust ventilation requirements for those industrial processes which by their operation cause air displacement. 4400 Fifth Avenue, Pittsburgh, Pa. June 1, 1942. H. B. MELLER, Managing Director. AIR FLOW MEASUREMENT BY THE DILUTION METHOD By THEODORE HATCH* and ROBERT H. WALPOLE, JR.* Problems ir, air flow measurement are encountered in whic** 11 'c not possible to connne the air stream entirely and force it through a meter or past a definite velocity-measuring station. Under these conditions, the rate of air flow may be determined by the dilution method wherein a test gas is fed into the air stream at a known rate and, after uniform mixing is established, its concentration is determined in a unit volume of air. The rate of feed of the gas divided by its resulting concentration equals the rate of air flow. The lest gas must be one that is not absorbed either chemically or physically in the system and, for genera! use. it must not be toxic or explos ive. It should lend itself to collection and measurement by means of simple sampling apparatus and analysis. Carbon dioxide meets these requirements. It is readily available in pressure cylinders and may be fed at a constant rate through a standard pressure-reducing valve and flow-meter. It is easily absorbed in an alkaline solution and is quickly determined by simple titration with as high an accuracy as may be desired. For approximate tests, a portable CO analysis apparatus of the volumetric type may be employed. It is, of course, necessary to subtract from the total amount of C02 in the sample that amount which is normally present in the air in order to determine the concentrations of the added gas. The basic apparatus required for air flow measurement by the dilution method is shown in Figure 1 and the various parts are enumerated below; 1. Cylinder of C02. 2. Constant pressure valve. 8. Flow-meter, calibrated in grams of CO-, per minute at 7C0 mm. Hg and 0* C. Correction values for other pressures and tem peratures are arranged in a convenient table. 4. Gas-feeding tube, arranged to insure uniform mixing of the test gas with the air stream. 5. Sampling tube. 6. Dust filter (when required). 7. Sampling apparatus, including absorption flask and flosv-meter. 8. Source of suction. Many combinations of feeding and sampling equipment will suggest themselves. The apparatus should be simple, rugged, and positive in opera- Department of Public Health and Preventive MvtlU-lne, School of Medicine. University of Pennsylvania. 3 tion. The flow-meter for feeding the CO.,t for example, may be built into the pressure-regulating valve while a hand-operated, constant-flow pump may be employed with the sampling apparatus. Accuracy The accuracy of the dilution method, using C02 as the test g3S, depends upon the following: 1. Uniformity of mixing of the C02 with the air stream. 2. Accuracy oi now-metev in the feeding apparatus. 3. Accuracy of the analytical procedure employed for determining the C02 concentration in the air stream. 4. Ratio of the concentration of added C02 to the normal atmos pheric content of this gas. Complete mixing of the test gas with the air stream is obviously essen tial. It is readily obtained in a simple air stream in which the air enters through a single inlet and the sampling point can be located a considerable distance downstream. In a complex- flow system, however, with numerous inlet and discharge openings and with opportunities for short-circuitir.g of the test gas, considerable care must be exercised to insure thorough mixing. In certain cases we have found it necessary to introduce the C02 into two or more of the secondary air streams before they join to form the main stream. Proof of uniform mixing is demonstrated when the concentration of the CO* is found to be the same at the several points of discharge. The accuracy of rate of feed of the test gas depends upon the design and calibration of the flow-meter, the corrections that are applied for the effects of temperature and pressure and the constancy of flow during the test. These are subject to direct control and the error may be fixed to meet the demands of the problem. This is also largely true with respect to the analytical procedure employed to determine the CO., concentration. A num ber of methods for measuring this gas are described in standard textbooks on gas analysis. The volumetric method, in winch the change in volume of the sample resulting from absorption of the CO* is measured directly, is perhaps the simplest. It requires no separate "sampling device nor is chemical titration involved. The standard instrument, however, does not measure the concentration more closely than to 1 part in 10,000 and, since the amount of gas added may not exceed 6 to 8 parts in 10,000, the accuracy of the air flow measurement will be 15 to 20 percent. More accurate methods depend upon the absorption of the C02 in an alkaline solution of known strength and determination of the amount collected by acid titration. This requires an accurate flow-meter, an efficient absorption device, protec tion of the absorption liquid from contamination with CO., from the air before and after sampling, and careful chemical analysis. It is capable, however, of measuring the C02 concentration with a high degree of accuracy. With simple apparatus and analytical procedures which are adaptable to field conditions an accuracy of 5C* is readily obtained.* The ratio of the concentration of added gas to the normal atmospheric content is, next to uniformity of mixing, the most important factor in deter- * Conductivity methods are also available. 4 mining the accuracy of air flow measurement by the dilution method. The error in air flow measurement increases as the ratio of added to normal C02 decreases, as indicated by the following equation: Eq R 1 rr --- 100 Where = Percent error in air flow measurement. Eco2 = Percent error in CO concentration. R = Ratio of added to normal C02 concentration. The relation between and R is plotted for three values of Eco2 in Figure 2. These curves exhibit sharp breaks at certain values of R below which the accuracy of air flow measurement decreases rapidly. For an error of ^5'/< in the C02 determination, the critical ratio is approximately 1.5 and this value has been selected for practical field work, thus giving an accuracy of air flow measurement of 10',<. This requires a rate of CO., feed of at least 3G grams per 1,000 cu. ft. of air, at sea level. The feeding apparatus should be provided with several interchangeable flow-meters to cover the range of air flow rates to be measured and in a particular prob lem it is necessary to select the proper rate of C02 feed from an initial estimate of the air flow and subsequent trial.4 Application The dilution method has been found particularly useful in studies of the exhaust ventilation requirements for certain industrial processes in which the principal factor controlling the ventilation needs is the rate of air dis placement established by the process itself. As an example, the discharge of crushed rock through a chute onto a belt sets up a flow of air with the falling material and the air escaping from the point of feed onto the belt disperses dust into the surrounding atmosphere. The rate of exhaust venti lation at this point must be in excess of the rate of air displacement. The latter can be measured by the dilution method, thus providing a basis for the design of the exhaust hood. The convenience of this method of meas uring air flow rates is well illustrated by the example shown in Figure 3. As indicated in the sketch, crushed ore was fed into the boot of an elevator through two chutes, one from primary screens and the other from a crusher. The material discharged from the elevator into a closed bin from which it was fed to screens, the oversize from which dropped through a bin and chute to the crusher. The problem was to determine the rate of air displacement in the elevator. Smoke tests revealed that air was entering the boot of the elevator with the ore dischaiging from the primary screens and from the crusher as well.- Above the elevator boot the housing was found to be under pressure with air escaping at the elevator head and at intermediate open ings. Four niv flow measurements were made at the locations shown in Figure 3. By means of tests 1 and 2 the Tates of air flow induced by the High rates of CO.- f*d (73 enu, per minute and more) through the pressurereducing valve caus.s (reeling \ the valve. The use of & heating element near the reducing valve helps. 6 n discharge of ore from the primary screens and the crusher respectively into the elevator were determined whereas tests 3 and 4 were designed to meas ure the total rate of air displacement in the elevator. Since the major air inflow occurred in the screen and crusher discharge chutes, it was expected that the sum of the rates of air flow determined in tests 1 and 2 would approximately equal that obtained in test 3. The results from tests 3 and 4 would be in substantial agreement provided there was no leakage in the elevator housing below the point of C02 feed in the latter test. The rates of air flow were found to be as follows:" Tet Rate of air flew, ctm. 1 ................................................................ 1810 2 ................................................................ 2750 1 + 2 .......................................................... 4560 3 ................................................................ 5000 4 ................................................................ 4600 In view of the fact that the rate of flow of ore was not constant during these tests, the agreement among the results is excellent. 6 0 ; J 53 |Oo o, .0 \ 111 --* kl I5 0 I to I Figure 1. Diagrammatic layout of apparatus employed for air flow measurement by the dilution method. 7 010 5 3 ,a / / / 8 010 5 4 I'lg u rc 2. E rro r in a ir flow measurement in relation to the ratio of added to norm al CO., concentration in the a ir. -O n e Fe e d Samples Tcsrs 3+ 4- Figurc 3. Measurement of air flow in an elevator by the dilution method. 9 010 5