Page 323 - Fundamentals of Water Treatment Unit Processes : Physical, Chemical, and Biological
P. 323

278                            Fundamentals of Water Treatment Unit Processes: Physical, Chemical, and Biological


              100                                              static mixers and back-mix reactors with respect to alum
                                                               dosages and filter effluent turbidities and found, in general,
                                                               lower alum dosages and lower effluent turbidities using a
                                                               static mixer.
                             Laminar                              Table 10.13 illustrates the nature of data available from
              10
                                                               catalogs, e.g., giving length of units for different pipe diam-
                                         Static mixer
                                                               eters ( 610 mm or 24 in.) and for different model numbers
             f                                                 (given as the number of elements). Units are available, how-
                                                               ever, as large as 1219 mm (48 in.).
                1
                                                                  Example 10.10 Static-Mixer Design

                                                                  Given
                                                                              3
                                                                  Let Q ¼ 0.696 m =s (20 mgd).
              0.1
                10 2          10 3          10 4           10 5  Required
                                      R                           Design a mixing system for alum coagulation.
            FIGURE 10.28 Static mixer friction loss coefficient as a function of  Solution
            R. (Adapted from Amirtharajah, A. et al., Mixing in Coagulation and  1. Set up a spreadsheet as shown in Table CD10.14 to
            Flocculation, Report 90841, American Water Works Research Foun-  calculate R, h L , G.
            dation, Denver, CO, pp. 37, 61–63, 2001.)               2. Add rows to explore the effects of pipe diameter,
                                                                      d(pipe), number of elements of static mixer selected,
                                                                      and variation in Q.
            design criteria for any of the foregoing parameters. They
                                                                    3. Develop ‘‘scenario’’ sets (each row is a scenario) to
            found that the number of elements made little difference
                                                                      examine each of the foregoing independent variables:
            for (#elements) > 2; using four or six elements did not, in  a. Effect of d(pipe). Select d(pipe) ¼ 762 mm (30 in.).
            general, result in lower effluent dosages or lower filter effluent  Smaller pipes result in much higher headloss and
            turbidities. Amirtharajah et al. (2001, p. 37) also compared  larger pipes give much lower G values.


            TABLE 10.13
            Length of Static-Mixer Elements for Different Pipe Diameters and Number of Elements
                                             a
                                  (a) Length (mm )                                    (b) Length (in.) b
            Pipe                    KMS Model                   Pipe                    KMS Model
            Diameter                                          Diameter
            (mm)      2     3     4     6    12    18    24     (in.)    2    3     4      6     12    18     24
             13                   114  156   292   432   568     0.50               4.50   6.1   11.5  17    22.4
             19                   140  203   384   568   754     0.75               5.50   8.0   15.1  22.4  29.7
             25                   178  257   492   727   962     1.0                7.00  10.1   19.4  28.6  37.9
             38       152   210   270   391  740   1108  1476    1.5    6     8.2  10.6   15.4   29.1  43.6  58.1
             51       191   267   333   498  965   1429  1908    2.0    7.5  10.5  13.1   19.6   38.0  56.2  75.1
             64       222   318   410   594  1153  1708  2280    2.5    8.8  12.5  16.1   23.4   45.4  67.2  90
             76       270   387   498   727  1416  2115  2813    3.0    10.6  15.2  19.6  29     55.8  83.2  111
            102       346   502   648   948  1832  2775  3680    4.0    13.6  19.8  25.5  37     72.1  109   145
            152       511   762   994  1454  2870  4277  5671    6      20.1  30.0  39    57    113   168    223
            203       664   968  1295  1924  3800                8      26.1  38.1  51    76    150
            254       841  1232  1626  2435  4820               10      33.1  48.5  64    96    190
            305       994  1451  1959  2896  5213               12      39.1  57.1  77    114   205
            356       921  1327  1759  2648  5213               14      36.2  52.2  69    104   205
            406       921  1327  1759  2648  5213               16      36.2  52.2  69    104   205
            457      1124  1645  2216  3258  6483               18      44.2  64.7  87    128   255
            508      1124  1645  2216  3258  6483               20      44.2  64.8  87    128   255
            610      1327  1949  2623  3867  7703               24      52.2  76.8  103   152   303
            Source: Kenics KMS models, 2002. (Note: Kenics is now, 2010, Chemineer-Kenics.)
            a
             Metric dimensions calculated from units in (b).
            b
              Dimensions from Kenics (2002, p. 20).
   318   319   320   321   322   323   324   325   326   327   328