## Process Design

Although considerable progress has been made in our understanding of substrate transport and microbial growth in biofilms, as discussed in Chapter 15, the application of that understanding to trickling filter design has been slow. This is primarily because trickling filters are considerably more complicated than the conceptual models used to describe biofilms. For example, we saw earlier in this chapter that flow through a trickling filter is both intermittent and highly irregular, whereas biofilm models commonly assume steady flow at a constant rate. Thus, while mechanistic models such as those in Section 16.1 have helped us develop a better understanding of trickling filter performance, they have not found application in practice. Rather, the design of trickling filters is based primarily on empirical correlations and simplified models depicting pilot- and full-scale system performance. Furthermore, because BOD, has been used historically to measure the concentration of biodegradable organic matter in those systems, many design relationships are typically presented in terms of it without regard to whether the organic matter is soluble or particulate. Finally, many relationships express the performance of the trickling filter system, i.e., the bioreactor plus the settler that follows it. Although this approach has had a negative impact on our ability to isolate and understand the fundamental processes involved, it has worked satisfactorily when used by experienced designers. Thus, it represents the current state-of-fhe-art and is the approach that we use here. It is likely, however, that as more basic data are collected from large-scale systems that better approaches will evolve.

This section reviews several procedures for sizing trickling filters for a variety of applications, criteria for sizing the ventilation system, and approaches for sizing coupled TF/AS processes. When sizing a trickling filter, consideration must be given to the nature of the media being used. Trickling filters containing rock and random pack high-rate media can be built to any depth consistent with the structural constraints of the media, as discussed earlier in this chapter. Bundle high-rate media, on the other hand, is modular, and economics dictate that the depth conform to an integer number of modules. One module of such media typically is 0.61 m high, so the depth should be in increments of 0.61 m. Furthermore, as discussed in Section 19 2.3, the maximum unsupported depth to which bundle media can be stacked is about 6.7 m, or eleven modules high. These constraints must be considered when selecting the depth of a trickling filter.

19.3.1 Sizing Trickling Filters with "Black-Box" Correlations

Historically, "black-box" correlations, i.e., those simply correlating output with input, of performance data from full-scale trickling filter applications have been used to size rock trickling filters. As discussed in Section 16.4, two such correlations are those developed by the National Research Council (NRC)" and by Galler and Go-taas.'s As also discussed in that section, great care should be exercised when blackbox correlations are used. The data set upon which they are based should be reviewed carefully by the design engineer to ensure that it is similar to the proposed application. If significant differences exist, then the equation should not be used.

The NRC correlation was based on performance data from 34 rock media trickling filters operating at military installations." It reflects the impact of TOL. and recirculation on performance. The influent BOD, concentrations at the plants were relatively high (generally 200 mg/L or greater). Consequently, the correlation may not accurately reflect the performance of rock media trickling filters treating lower strength wastewater. The Galler-Gotaas'" correlation is based on 322 observations at typical municipal wastewater treatment plants. It recognizes the importance of organic loading, hydraulic loading, recirculation, media depth, and temperature on system performance. Both the NRC and the Galler-Gotaas correlation equations were developed on the basis of the BOD, concentration in the process influent and the secondary clarifier effluent. Consequently, they implicitly incorporate the impact of both the trickling filter and the secondary clarifier. When using either, it must be assumed that the secondary clarifier is properly sized and operated. Because both black-box correlations are used for sizing rock media trickling filters and few are being designed today, neither is presented here. However, they are presented in standard design manuals.""0

19.3.2 Sizing Trickling Filters with Loading Factor Relationships

The process loading factor approach to trickling filter design reflects the fact that trickling filter performance is generally correlated with the TOL or the SOL, as discussed in Section 19.2.1. The steps required when using this approach include: selection of an appropriate TOL or SOL based on experience and/or on performance correlations such as those shown in Figures 19.5 and 19.6; use of the selected relationship to calculate the required media volume; and use of typical media depths and hydraulic loadings, as described in Sections 19.2.1, 19.2.2, and 19.2.3, to dimension the trickling filter. The following examples illustrate the use of the loading factor relationships to design a roughing filter, a carbon oxidation application, and a trickling filter achieving combined carbon oxidation and nitrification. The first example considers a roughing filter.

### Example 19.3.2.1

Size a trickling filter for a roughing filter application in which 75% of the soluble BOD, is to be removed. The wastewater flow rate is 5,000 mVday, and the total BOD, concentration is 150 mg/L. Bundle media is to be used with a minimum THL of 1.8 m/hr. Assume that the relationship between soluble BOD, removal and TOL shown in Figure 19.5 is applicable.

a. What TOL should be used to achieve the treatment objective?

From Figure 19.5, a TOL of 2.5 kg BOD,/(m' -day) will achieve the treatment objective.

b. What media volume is required?

The media volume can be calculated with a rearranged form of Eq. 19.1:

What media depth is required to maintain the minimum allowable THL of 1.8 m/hr with no recirculation?

The media depth is just the volume divided by the cross-sectional area. The area can be obtained from the definition of the THL, given by Eq. 16.8. For a = 0 and A„ = 1.8 m/hr (= 43.2 m/day):

5,000

While this is acceptable, as discussed in Section 19.2.3, performance is generally reduced for media depths less than about 4 m. Thus, a greater depth should be used, which is acceptable since that will give a smaller cross-sectional area, thereby giving a greater THL. The depth of a typical sheet media bundle is 0.61 m. If we take 4 m as being the shortest acceptable depth and recognize that an integer number of bundles must be used, then the media depth should be at least 4.27 m, corresponding to 7 layers of media.

d. What THL would result if the trickling filter were 4.27 m high? 300

A,, = = 71 m/day = 2.96 m/hr This is an acceptable THL.

This example illustrates that, for roughing filter applications, the media depth may be established at minimum design values. It also illustrates that relatively high THL rates are sometimes used.

The following example uses the process loading factor approach to size a trickling filter for relatively complete removal of biodegradable organic matter, such as for secondary treatment.

### Example 19.3.2.2

Size a trickling filler for 90% removal of soluble BOD,. As in Example 19.3.2.1, the wastewater flow rate is 5,000 m'/day, the total BOD, concentration is 150 mg/L, and bundle media with a minimum THL of 1.8 m/hr is to be used. Assume that the relationship between soluble BOD, removal and TOL shown in Figure 19.5 is applicable.

What media volume is required?

From Figure 19.5 it can be seen that a TOL of 0.75 kg BODJim'day) must be used to achieve 90% removal of soluble BOD,. The media volume can be calculated with a rearranged form of Eq. 19.1:

If a media depth of 6.7 m is used, which is the maximum unsupported depth of plastic sheet bundle media and corresponds to 11 layers of bundles, what recirculation flow rate is required to maintain a THL of 1.8 m/hr (43.2 m; day)?

The recirculation ratio, a, can be calculated with Eq. 16.8 after calculating the cross-sectional area of the trickling filter.

1,000

## Post a comment