Ann. occup. Hyg. Vol. 21, pp. 265-269. Pergamon Press Ltd. 1978. Printed in Great Britain EFFECT OF FLANGES ON THE VELOCITY IN FRONT OF EXHAUST VENTILATION HOODS* B. FLETCHER Safety in Mines Research Establishment, Sheffield Abstract The addition of a flange to a local exhaust ventilation hood can produce a large increase in velocity in front of the hood. It is shown that an optimum flange width exists beyond which there is little further increase in velocity; this flange width is A* where A is the area of the hood. As hood aspect ratio increases, the effect of flanges increases and the addition of even small flanges to slots can have an appreciable effect. Velocity increases of over 55 % have been measured in front of hoods of aspect ratio 16:1. List of symbols: A = face area of hood D = flange width Q = volume flow rate into the hood V centreline velocity at distance X from the hood face V o mean velocity over the hood face V F = centreline velocity (flanged hood) V v = centreline velocity (unflanged hood) W = hood or slot width X = centreline distance from hood face. A PREVIOUS paper (FLETCHER 1977) on design criteria for suction hoods and slots for use in the control of dust and fume established the dependence of centreline velocity on distance from the face of unflanged hoods. In the present paper the study has been extended to flanged hoods. Adding a flange to a hood can have two beneficial effects: firstly, the flange improves entry conditions into the hood and secondly it reduces the amount of air drawn in from the usually uncontaminated regions behind the hood. Improved entry conditions result in (a) a larger effective flow area at the vena contracta, (b) a lower entry pressure loss and hence reduced power requirements, and (c) a better velocity distribution in front of the hood. The latter effect is achieved at the expense of a decrease in the centreline velocity very close to the hood face. Farther away from the hood the effect is not noticeable and the velocity contours are pushed farther from the hood face as the air is forced to flow mainly from the region in front of the hood. On the question of how wide the flange should be, DALLAVALLE (1930) stated that a flange approx 5 in. (0.13 m) wide is sufficient for hoods up to 3 ft 2 (0.28 m 2 ) in area. Commenting on this figure, SILVERMAN (1942) wrote that the flange size might be better stated in terms of hood size. ALDEN (1948) suggested that if possible the 1 Crown Copyright 1978. 265
266 B. FLETCHER FIG. 1. Variation of centreline velocities with flange width for a square hood. flange width should be such that it intercepts the 5% velocity contour of the unflanged hood. On this basis Dallavalle's measured contours would give a flange width of approximately 0.75,4* for a square hood. Measurements of centreline velocities have been carried out on hoods of aspect ratio 1:1,4:1 and 16:1, each of area 0.01 m 2, to try to determine the effect of flange width. Several workers (FLETCHER, 1977; DALLAVALLE, 1930) have shown that such results can be scaled to hoods of larger area. Figure 1 shows the variation of velocity with flange width for the square hood. V is the centreline velocity at distance X from the hood face, A is the area of the hood. Q is the volume flow rate into the hood, V o is the mean velocity over the hood face, D is the flange width and W is the hood or slot width. It can be seen that at distances less than XjA^ = 0.3 from the hood the centreline velocity is decreased by the addition of a flange. This is due to the larger effective flow area and the more uniform velocity distribution in this region. At greater distances from the hood the velocity is increased. The optimum flange width beyond which there is little further increase in velocity is D = W = A i. Figure 2 shows the results from a similar experiment with a hood of aspect ratio 16:1. The diminution of centreline velocity in the region close to the hood by the addition of a flange seems to decrease with increasing aspect ratio. From Fig. 1 it is not possible to tell whether flange width should be expressed in terms of the hood area or of its width; however, taken in conjunction with the results shown in Fig. 2 it can be seen that hood area is the relevant parameter. Figure 2 also confirms the value D = A 1 * as the optimum flange width. Further experiments were carried out on square hoods
Effect of flanges on the velocity in front of exhaust ventilation hoods 267 A *- I I 1 I X^ X^*X*^^^^»X^^^^^»X^ x I 0-15 VA Q " 4-' X 0 0-5 I I 5 D X 0-60 0 2 4 6 2. W FIG. 2. Variation of centreline velocities with flange width for a hood of aspect ratio 16:1. FIG. 3. Effect of flange width on the velocity in front of square hoods.
268 B. FLETCHER 1-4 FIG. 4. Effect of flange width on the velocity in front of a hood of aspect ratio 4:1. VF 1-6 1-5 4 1-3 2 : n-9 / S r f Is / < ) i> ; < i ^^^ ^ FIG. 5. Effect of flange width on the velocity in front of a hood of aspect ratio 16:1. 10 0-6 015 of four different sizes. Figure 3 shows the effect of flange width on V F /V V at different distances from the hood face, where F F and V v are the centreline velocities measured with flanged and unflanged hoods respectively. It can be seen that for a flange width of D = A i an increase in velocity of approximately 25 % over that with the unflanged hood is produced at a distance of X/A i = 1 from the hood face. Figures 4 and 5 show results for hoods of area 0.01 m 2 and aspect ratios of 4:1 and 16:1 respectively. It can be seen that the velocity diminution close to the hood
Effect offlangeson the velocity in front of exhaust ventilation hoods 269 face reduces as the aspect ratio increases and that the effectiveness of the flange increases with increasing aspect ratio. In the case of the hood of aspect ratio 16:1 the percentage increase in velocity seems to be at its maximum at distances greater than X/A* = 0.6 from the hood face, the increase then being more than 55 %. It can be seen from Figs 4 and 5 that, although the optimum flange width is D = A*, large increases in velocity can be produced by much smaller flanges when they are fitted to hoods of large aspect ratio. CONCLUSIONS It has been shown that large increases in the velocity in front of exhaust hoods can be produced by the addition of a flange to the hood. Flange widths greater than D = A* give little further increase in velocity. As the aspect ratio of the hood increases the effect of the flange increases and even small flanges can make an appreciable difference to the velocity. Velocity increases of over 55% have been measured in front of hoods of aspect ratio 16:1. Acknowledgement The author would like to thank Mr A. E. Johnson for assistance in carrying out the experiments. The paper is contributed by permission of the Director of Research and Laboratory Services, Head of SMRE, Health and Safety Executive. REFERENCES ALDEN, J. L. (1948) Design of Industrial Exhaust Systems. Industrial Press, New York. DALLAVALLE, J. M. (1930) Studies in the design of local exhaust hoods. Sc. D. Thesis, Harvard Engineering School. FLETCHER, B. (1977) Ann. occup. Hyg. 20, 141-146. SILVERMAN, L. (1942) /. indust. Hyg. Toxicol. 24, 259-266.