Document e77XOOBZLEEjMZm4m9nQQRgVE

302 CHAPTER 21 1960 Guide Fj = velocity of standard air in the outlet duct, feet per minute. Ai -- area of the inlet duct, square feet. At - area of the outlet duct, square feet. Ci -- loss coefficient based on area Ax . Cj = loss coefficient based on area Aj . For a gradual symmetrical enlargement. Equation 8 changes to B-- c- (tst)" ~ c-' (as)'"CrCi {zmf <9> tchert -- pressure loss due to gradual enlargement, inches of water. C, -- coefficient of loss, as ratio of loss to loss for abrupt expansion, dependent upon the total angle included between the sides of the duct. The loss for a sudden symmetrical contraction can be ex pressed as ;i-"C-(^),-e'(4)'"C-(^5)' (10) where B, -- pressure loss due to sudden contraction (Fig. 10b). C, a loss coefficient based or orifice area A, . V, -- velocity of air through orifice, feet per minute. The loss for a gradual symmetrica) contraction can be similarly expressed as B- " c' (ioos) (11) where the coefficient of loss C, depends on the included angle of the aides of the duct and the sharpness of the edges at the junction of taper to following duct section. DUCT DESIGN The following discussion refers to ducts for commercial and industrial heating, ventilating, and air-conditioning sys tems of the central station type. The design procedures given yield the static pressure required to overcome the re sistance of the ductwork, including the supply outlets and return intakes. The fan selected for the duct system must not only produce thU pressure but also the additional pres sure required by the central equipment such as washers or spray chambers, hearing or cooling coils, and filters. Pressure losses of these components should be obtained from the manufacturers' catalogs. Special duct design procedures for bearing ducts used in residences can be found in Chapter 18 Warm Air Heating Systems. The design of ducts in industrial exhaust systems is discussed in Chapter 52. General rules which should be followed in the design of ducts are: 1. The air should be conveyed as directly as possible at the permissible velocities to obtain the desired results with mini mum noise and greatest economy of power, material, and space. 2. Sudden changes in the direction or velocity of the air should be avoided. When sudden changes are necessary at bends, turning vanes should be used to minimise the presure loss. 3. Diverging transition pieces should be made as gradual as practicable. As shown in the section on area changes, losses in abrupt enlargements are high &&d therefore such transitions should be avoided. The included angle of divergence for enlarge ment should not exceed 20 deg. Losses in contractions are low but the included angle of convergence should not be greater than 60 deg. 4. Where the greatest air carrying capacity per square foot of sheet metal is desired, rectangular ducts should be made as nearly square as possible. Aspect ratios (ratio of width to depth) greater than S to 1 should be avoided. Where possible, a ratio of 4 to 1 or less should be maintained. 5. Ducts should be constructed of smooth material, such as steel or aluminum sheet metal. For ducts made from other materials, proper allowance for the change in roughness should be made* 6. Through the design procedures which follow, a reasonably precise estimate of the flow resistances offered by the system can be obtained. However, it should be recognised that in actual installations, resistances may vary considerably from the cal culated values because of variation in the smoothness of ma terials, types of joints used, and the ability of workmen to fabricate the system in accordance with the design. Fans and motors should therefore be selected to provide at least a slight factor of safety, and dampers should be installed in each branch outlet for balancing the system. Procedure for Duct Design The general procedure is as follows; 1. Study the plan of the building and arrange the positions of the supply outlets to provide proper distribution of air within each space. Select outlet sizes from manufacturers' cata log data. 2. Draw a sketch of the most convenient system of ductwork, connecting the supply outlets and return intakes with the cen tral station apparatus, taking cognizance of the building con struction, avoiding all obstructions in steel work and equipment, and at the same time maintaining a simple design. 3. Calculate the sizes of ail main and branch ducts by one of the procedures given in the following section. 4. Determine the total pressure requirement of the supply and return duct systems. Although the loss in total pressure of each duct run connecting the fan and each supply outlet (or return intake) should be calculated and made the same for all runs, ordinarily only the pressure loss of the duct run apparently having the greatest resistance is referred to as the pressure loss of the duct system. Dampers are relied upon for balancing the system. Design Velocities It is not possible to give specific rules for selecting duct velocities, but the velocities given in Table 6 have been found to give satisfactory results in designing conventional systems. Since the fan horsepower increases approximately as the square of the velocity, and noise generation increases with static pressure, velocities should be kept low for quiet and economical operation. On the other hand, as evident from Equation 3, at a given Sow rate the duct size increases with decreasing velocity. For multistory buildings, it is some times possible to reduce the height between floors by using very small ducts, thereby effecting a considerable reduction in building investment cost. The various space-saving sys tems which are becoming increasingly numerous are disctis&ed in the section High-Velocity Systems in Chapter 19 Central Systems for Air Conditioning. The design of high-velocity ducts is discussed briefly in a later section, Design of High- Velocity Ducts, in this chapter. DESIGN METHODS In the Hpsign of air duct systems, three methods are em ployed: (1) velocity-reduction, (2) equal-friction, and (3) static-regain* The three methods and their refinements rep resent different design levels of accuracy and complexity, and they should be selected, therefore, to suit the application. Simple duct systems may be designed as quickly and easily as possible, but for large installations the system static- Air Duct Design Table 6___ Recommended and Maximum Duct Velocities for Conventional Systems CecemeiendW VeJoetfie*, fpa Detignatioa Residence* Scboab, Tbeaten, Public SoBdbtgi industrial building! Outdoor Air Intakes* Heating Coils* Air Washers Fan Outlets Branch Risers 500 500 500 250 300 350 450 500 600 500 500 500 1000-1600 1300-2000 1600-2400 700-900 600 500 1000-1300 1200-1800 600-900 800-1000 600-700 800 Maximum VeJoctfie*, Fpa Outdoor Air Intakes* Heating Coils* 800 900 1200 300 350 350 500 600 700 Fan Outlets 500 1700 500 500 1500-2200 1700-2800 Main Ducts Branch Ducts Branch Risers 800-1200 1100-1600 1300-2200 700-1000 800-1300 1000-1800 650-800 800-1200 1000-1600 * These veloeitjq ere tar late) Cue we, Dot the aet free mre; either velocities is table ere far net free area. pressure requirement must be determined as precisely as possible, the most accurate method being recommended. .... Velocity-Reduction Method This method consists of selecting the velocity at the fan discharge and designing for progressively lower velocities in the main at each branch duct. With the selected velocities and known air-flow rates, the various duct diameters are read directly from Figs. 2 or 3, and the equivalent rectangular sizes are obtained from Table 2. The pressure loss of the run having apparently the highest resistance is determined by adding the straight pipe, elbow, and transition losses; this total value represents the fan static pressure required for the supply-duet system. The return-air system is sized ami larly, starting with the lowest velocities at the return intakes and increasing them progressively in the direction of the fan inlet. Dampers are refled upon for balancing the system. A refinement of this method is to size the several branch ducts to dissipate the pressure available at the entrance to each. The pressure loss of the ductwork between the fan and first branch take-off is subtracted from the now known fan static pressure to obtain the available pressure at each junc tion. By trial, a branch velocity is found that results in the branch pressure loss being equal to, or somewhat less than, that available. The procedure is repeated for each branch. If the fan is specified so that the static pressure available for the ductwork is known, the method consists of finding, by trial, the velocities in the main duct that will result in a pressure loss-equal to the pressure available. The branch ducts are then sized as previously explained. The merits of the velocity-reduction method are that (1) duct sizes are determined very easily, and (2) velocities can 303 be limited to those known to be safe from causing noise prob lems. Its weaknesses are: (1) proper choice of velocities re quires experience and judgment, and (2) the designer cannot always determine by inspection which run probably has the highest resistance. Equal-Friction Method The principle of this method is to make the pressure loss per foot of length the same for the entire system. With this method little balancing is required for symmetrical layouts in which all runs have about the same resistance. For lay outs having both short and long runs, the shortest run will require considerable dampering. Usual practice is to select the velocity in the main duct near the fan from the standpoint of noise for the particular application. Since the flow rate (cfm) is known, this estab lishes a value of friction loss per 100 ft of duct in Figs. 2 or 3. This same friction loss value is maintained throughout the design. For example, the flow rate in the main after the first branch take-off is reduced by that handled by the branch. Therefore, proceed vertically downward in Figs. 2 or 3 to the new flow rate value, and read the velocity and diameter. Note that the velocity is reduced by this procedure. An ad vantage of this method is that it automatically reduces the duct velocities in the direction of flow, thereby insuring that problems of noise will be considerably lessened. The equivalent rectangular size of any diameter is obtained from Table 2. By continuing the procedure, all sections, including branch ducts, are sized from Figs. 2 and 3 at the same friction loss per foot of actual length. After sizing the system the pressure loss of the run having apparently the highest resistance is calculated. In so doing, the pressure losses of all elbows and transitions are included, and are expressed in terms of equivalent length of straight pipe. The principal limitation of the equal-friction method is that it does not differentiate between runs having several transitions, elbows, etc., and runs having none. Only the actual length of duct is considered; this and the flow rate fix the duct size. Moreover, when computing the system re sistance, care must be exercised that the pressure losses of transitions, elbows, etc., are included and added to the straight pipe losses. If the pressure available for the ductwork is known, as it is for packaged equipment, this pressure can be divided by the total equivalent length of the run apparently having the highest resistance to obtain a design friction loss value per foot, for use with Fig. 2 or Fig. 3. Hence, for these applica tions, it is not necessary to select an initial velocity. However, the method then has the weakness that the resistances of fittings must be expressed in terms of equivalent length. Since transitions, elbows, etc., have predominantly dynamic losses, the equivalent length of a particular fitting varies considerably with its actual size. Note, for example, the values for elbows in Fig. 7 of Chapter 18 are related to duct size, and note also, that the elbow losses in Fig. 9 of this chapter are given in terms of the number of diameters. Hence, when the available pressure is known, the method requires that the duct size be estimated in advance. The calculated duct size should therefore be compared with the initial estimate, and if considerably different, a recalculation should be made using the calculated size. Less dampering is required if the method is modified so that only the main duct is sized by the equal-friction method. Hie fan is selected for this total duct resistance and the