Document 7OzqwLk5jO5GMNO3pLxKjdXae

846 CHAPTER 42 1949' Guide Table 6. End Reflection of Plate Absorbers Percentage Free Area of Absorber 50 40 30 25 20 Attenuation, db i 2 4 5 6 jock wool blanket or 1 in. sound absorbing board preferably nailed to wood strips on the inside of the plenum. With such a lining the plenum is particularly effective in reducing low frequency fan noise. The absorp tion of the plenum in sabins is the sum of the products of each interior area of the plenum measured in square feet multiplied by its corresponding absorption coefficient. Plate Cells One of the most economical methods of applying sound. absorbent material from the standpoint of. both labor and material is the plate cell. The plate cell consists of 5 or 1 in. sound absorbent board, spaced on 2, 3 or 4 in. centers. The attenuation due to the plate cell may be divided into two parts. There is reflection at each end due to the change in area and the absorption at the ends. Values for this attenuation are given in Table 6 ` which depend on the spacing. There is also attenuation due to absorption of sound within the passages of the cell, which depends on the length and the spacing. The attenuation within the cell for 1 in. board neglecting the end effect is given approximately by Equation 5. where ..,-B = attenuation, decibels. Jj = linear length of duct, feet. S *= spacing in inches between plates up to 3 in. a = absorption coefficient for the full thickness of the cell material. value of a see Table 7. For typical An important objection to the plate cell is the increase in duct cross- sectionar area required. Often on the fan discharge, particularly with 'unitary equipment,'where a number of branch ducts take off, the plate cell may be installed with little or no difficulty. . Table 7. Attenuation Formulae for 1 In. Thick Ttpical Duct Lining Board Frequency 256 512. 1024 2048 Absorption Coefficient 0.37 0.69 0.78 . 0.78 Attenuation Reduction, db - 3.0 L P/A 7.5 L P/A '9.5 L P/A 9.5 L P/A Sound Control 847 Outlet Sound Absorbers. Outlet sound absorbers are rectangular or plate cells installed directly behind an outlet or they may be the lining of a panr or plaque outlet. They are particularly effective in the elimination of high frequency whistles which are generated by air flow in the ducts. They are also employed in large systems with long runs where only a few outlets near the fan require treatment. Frequently outlet cells are the only means of correcting existing noisy installations, as the duct sections directly behind the outlets may be the only sections accessible for treatment. (See Fig. 2). Duct Lining or Rectangular Cells One series of experiments10 made on a commonly used type of duct lining material (1 in. rock wool sheet) has shown that, subject to certain restrictions, the attenuation of single-frequency sounds may be expressed by the approximate Equation 6. This equation is accurate within plus or minus 10 per cent for duct sizes ranging from 9 x 9.in. to 18 x 18 in., for l MEM SECTION A-A ~ Fbta itowrtw Fig. 2. Outlet [Cells fob Pan Outlets ob Grilles cross-sectional dimension ratios of 1:1 to 2:1, for frequencies between 256 and 2048 cycles, and for absorption coefficients between 0.20 and 0.80. where p R 12.6L-al* A (6) ..... R -- attenuation, decibels. L = length of lined duct, feet; P = perimeter of duct, inches. A = cross-sectional area of duct, square inches. *' a = absorption coefficient of lining. /- In Table 7, the absorption coefficients at different frequencies of a material of the previously mentioned type are listed, together with the corresponding values for Equation 6. . Results of other experiments indicate, however, that Equation 6 may be in error when applied to other types of duct lining material and to. duct sizes and shapes outside of the range specified. An empirically derived chart11 representing the average experimental data on a number of different types of materials including the rock wool sheet mentioned as applicable to Equation 6 is shown in Fig. .3. Since individual materials vary, the curves in Fig. 3 are given only as representing the best available averages