| description abstract | Flocculation is a key step in treating drinking water for pathogen removal. Baffled hydraulic flocculators (BHFs) have many advantages, including plug-flow reactor kinetics, ease of construction, and operation without moving parts. Optimal design for BHFs requires the appropriate velocity gradient, which depends primarily on the minor losses associated with the 180 degree redirection of flow around the ends of the baffles. While previous design guidelines have estimated these minor loss coefficients in the range of 2.5–4.0, and some work has been done to identify key geometric parameters impacting minor loss coefficients, no physics-based mathematical model has been presented for estimating minor loss coefficients a priori as a function of these geometric parameters. A dimensionally homogeneous physics-based model describing the relationship between the flow expansion and flow constriction length scales is derived in terms of baffle geometry and tuned to computational fluid dynamics (CFD) simulation results, using the realizable k-epsilon model and the Reynolds stress model on periodic domains. These results are then compared with baffle minor loss coefficient estimates presented in the literature, as well as with several plant-scale BHF minor loss measurements made in Central America. Across CFD simulations, hydraulic height to baffle spacing ratios varying from 2 to 10 corresponded to minor loss coefficients ranging from 9.29 to 2.55. The minor loss coefficients associated with height-to-spacing ratios less than six tended to be larger than the minor loss coefficients associated with fully expanded flow presented in the literature, with the largest minor losses occurring at the smallest height-to-spacing ratios. The mathematical model derived in this study is expected to better inform the initial design of BHFs to reduce the need for post-construction adjustments to BHFs. When designing baffled hydraulic flocculators (BHFs), it is essential to consider that the pressure drop across baffles depends not only on flow velocity, but also on baffle geometry. This is especially critical in situations where the spacing between successive baffles is large or where baffles are short. In either of these cases, the measured pressure drop across baffles may be significantly larger than the pressure drop calculated from minor loss coefficients frequently cited in the literature. This paper proposes a physics-based approach to accurately predict the minor loss coefficients associated with the flow around a baffle based on baffle geometry. The implication is that the hydraulic design of BHFs will require iteration based on the influence of baffle geometry on minor loss coefficients. These findings extend more generally to the hydraulic design of baffled channels and may be applied to other mixing or heat transfer applications, such as the design of shell-and-tube heat exchangers. | |
