Minimum Filter Media Depth Requirements

The minimum depth of filter media cannot be less than the depth specified by the manufacturer.

The performance of high-rate granular media filters at removing pathogens and particles is contingent upon the depth of the filter media (as shown in MAHC Annex Table 4.7.2.1), especially at rates of 15 gpm/ft2 (37 m/h), which is why these filters recommend at least 24 inches (61.0 cm) of filter media. The WHO recommends filtration at 10-12 gpm/ft2 (24-29 m/h) for sand filters while the PWTAG recommends 4-10 gpm/ft2 (10-24 m/h) as the maximum filtration rate for all non-domestic POOLS using sand filters.¹⁸⁷,¹⁸⁸ The STANDARD sand filter bed depth typically varies from 0.55 to 1 m (22 to 39 inches) in the UK.¹⁸⁹ Minimum Depth For swimming POOL filters with less than 24 inches (61.0 cm) of media between the top of the laterals and the top of the filter bed, lower filtration rates (e.g., 10 gpm/ft2 (24 m/h) are recommended to efficiently remove particles and pathogens. Improvements in particle removal with decreasing filtration rates have been documented.¹⁹⁰ Drinking water treatment facilities typically limit filtration to less than 4 gpm/ft2,(10 m/h), which is similar to the filtration rates recommended in AQUATIC VENUES in the 1920s.¹⁹¹,¹⁹² The minimum depth of sand in POOL filters was 36 inches (0.9 m) in 1926.¹⁹³ Sand filters are typically designed in drinking water treatment for an L/d ratio of 1000 or greater, where “L” is the depth of the media and “d” is the diameter of the media grain.¹⁹⁴ For example, a 0.6 mm effective size sand would recommend a minimum 0.6 m (23.6 inches) bed depth, and a 12 inch (30.5 cm) deep sand bed with 0.5 mm grains would have an L/d of only 610. The minimum depth of filter media above the underdrains (or laterals) is recommended be 24 inches (61.0 cm) or greater with sufficient freeboard (or space between the top of the media and the backwash overflow) to allow for a minimum of 35% filter bed expansion during backwashing. Sand or other approved granular media should be carefully graded to ensure fluidization of the entire filter bed during backwashing. A design backwash rate of at least 30% higher than the minimum fluidization velocity of the d90 size of the media in water at the larger of 86°F (30°C) or the maximum anticipated operating temperature is recommended. A backwash rate higher than the minimum could be necessary to effectively clean the media during backwashing. Variations in the media type, density, water temperature, effective size, or uniformity coefficient may cause changes in the recommended backwash flow rate and/or bed expansion, which should be subject to approval by the local authority provided hydraulic justification by the design engineer. Sand or other approved granular media should be carefully graded to ensure fluidization of the entire filter bed during backwashing. The specifications of POOL filter sand (or lack thereof) can lead to filter media being installed that cannot be effectively cleaned during backwashing. Sand that cannot be properly cleaned can lead to filter failures and/or biofilms in the bottom of a filter. Researchers have found nematodes, rotifers, ciliates, zooflagellates, amoebic trophozoites and cysts, as well as bacterial masses in the backwash water of swimming POOL sand filters.¹⁹⁵ A design backwash rate of at least 30% higher than the minimum fluidization velocity of the d90 size of the media in water at the larger of 86°F (30°C) or the maximum anticipated operating temperature is recommended, but a backwash rate higher than the minimum could be necessary to effectively clean the media during backwashing. These backwashing recommendations are based on drinking water treatment practice.¹⁹⁶ For a sample of AQUATIC VENUE filter sand examined at UNC Charlotte, the d90 size (i.e., 90% of the grains smaller than this diameter) of the media was estimated from the sieve analysis results in MAHC Annex Figure 4.7.2.1.4.1 (below) to be 1.06 mm. The calculated minimum fluidization velocity of this sized sand grain in water at 86°F (30°C) was calculated to be 16.7 gpm/ft2 (41 m/h). Since this backwash velocity would be expected to leave approximately 10% of the grains in the filter that were larger than the d90 unfluidized, common practice is to recommend a backwashing rate 30% greater than this minimum value (or 21.7 gpm/ft2) (53 m/h). The recommended backwash flow for this media by Kawamura¹⁹⁷ was graphically estimated to be 20.9 gpm/ft2 (51 m/h) at 68°F (20°C). This is the rationale for requiring at least a 15 gpm/ft2 (37 m/h) backwashing rate of all swimming POOL sand filters. To ensure compatibility with the minimum recommended backwashing rate of 15 gpm/ft2 (37 m/h), filter sand should pass through a number 20 U.S. standard sieve or equivalent (i.e., all sand grains should be smaller than approximately 0.85 mm). While this recommendation of “#20 Silica sand” is common in swimming POOL manuals and by filter manufacturers, it does not appear to be representative of the actual sand that might be installed. Sieve analyses of two brands of commercially available “pool filter sand” are provided in MAHC Annex Figures 4.7.2.1.4.1 and 4.7.2.1.4.2. Sand can also be specified by an effective size (E.S.) of 0.45 mm with a uniformity coefficient (U.C.) of less than or equal to 1.45, which is roughly equivalent to a 20/40 mesh sand. A 20/40 mesh sand would pass through a #20 (0.85 mm sieve) and be retained on a #40 (0.42 mm) sieve. In order to reduce the rate of headloss accumulation at the top of the filter bed (and the frequency of backwashing), a 20/30 mesh sand could be specified where the smallest grains at the top of the filter would be approximately 0.60 mm (30 mesh) instead of 0.42 mm (40 mesh). Figure 4.7.2.1.4.1: Grain Size Distribution of Pool Filter Sand – Brand A Figure 4.7.2.1.4.2: Grain Size Distribution of Pool Filter Sand – Brand B The depth of the expanded bed during backwashing should be at least 20% greater than the depth of the fixed bed after backwashing. Experiments to determine the backwashing rates recommended to fluidize a bed of POOL filter sand in 3-inch (7.6 cm) and 6-inch (15.2 cm) diameter clear PVC filter columns based on visual observation were conducted. Fluidization is somewhat subjective when observed visually because sand grains could be moving sluggishly prior to fluidization and because the smaller grains at the top of the filter will fluidize long before the larger grains at the bottom. For this reason, bed expansion was measured and recorded along with visual observations of when the bed actually fluidized. Fluidization was visually observed to occur between 20 and 23 gpm/ft2 (49-56 m/h), which coincided with 19-23% bed expansion in both sized columns for the unaltered commercial filter media at 68°F (20°C). Expansion data from the 3-inch (7.6 cm) diameter filter column is shown in MAHC Annex Table 4.7.2.1.4.1 (below). The 20/30 mesh fraction of the same filter media was examined under the same conditions, and the experimental results are provided in MAHC Annex Table 4.7.2.1.4.2. The media was observed to be fully fluidized at 19.9 gpm/ft2 (49 m/h)with a bed expansion of 21.8% at 68°F (20°C). Calculations based on Cleasby and Logsdon¹⁹⁸ indicate that filter backwashing rates should increase by approximately 18% for this media as the temperature is increased from 68° to 86°F (20° to 30°C) due to changes in the viscosity of water with temperature. Fluidization can be somewhat complicated to estimate, but filter bed expansion can be easily measured in the field with granular media filters that use viewing windows. Furthermore, a model exists that can be used to calculate filter bed expansion of sand in a filter during backwashing.¹⁹⁹ This model tends to be sensitive to fixed bed porosity, but using a value of 42% porosity with a sphericity of 0.85 and density of 2.65 g/cm3 yielded a bed expansion of 22.7% at 20 gpm for water at 86°F (30°C). This is the rationale for requiring the depth of the expanded bed during backwashing being at least 20% greater than the depth of the fixed bed. PWTAG recommends 15-25% bed expansion following air scouring at 32 m/h (13 gpm/ft2)(at 0.35 bar).²⁰⁰ In a study funded by PWTAG, researchers used a backwashing rate of 25 gpm/ft2 (61 m/h) to achieve 25% bed expansion of their filters.²⁰¹ Variations in the media type, density, water temperature, effective size, or uniformity coefficient may cause changes in the recommended backwash flow rate and/or bed expansion, which should be subject to approval by the local authority provided hydraulic justification by the design engineer. Table 4.7.2.1.4.1: Pool Filter Sand at 68°F (20° C) Backwash Flow Bed Expansion 12.4 gpm/ft2 (30.3 m/h) 3.6% 16.3 gpm/ft2 (39.8 m/h) 11.4% 18.5 gpm/ft2 (45.2 m/h) 16.3 % 20.3 gpm/ft2 (49.6 m/h) 19.3% 22.1 gpm/ft2 (54.0 m/h) 22.5% Table 4.7.2.1.4.2: Pool Filter Sand Sieved 20/30 mesh at 68°F (20° C) Backwash Flow Bed Expansion 12.2 gpm/ft2 (29.8 m/h) 4.8% 15.8 gpm/ft2 (38.6 m/h) 13.0% 17.9 gpm/ft2 (43.8 m/h) 17.4 % 19.9 gpm/ft2 (46.6 m/h) 21.8% 21.5 gpm/ft2 (52.6 m/h) 25.6% 23.9 gpm/ft2 (58.4 m/h) 31.2%