Implementation of flushing tanks in combined sewer networks to comply with optimal self-cleansing properties

Combined sewer networks are designed to collect and transport the aqueous and solid wastes originating from domestic, industrial and catchment surfaces for treatment and disposal. One of the problems that arise is sedimentation of the particles within the collected water while flowing throughout such networks. In fact, many sewer pipes in combined sewer systems experience considerable fluctuations in flows, ranging from high flow during short-term storm events to longer periods of much lower dry weather flows. In low flow periods sewers experience a characteristic diurnal flow variation. In such periods when the pipe filling level is very low, minimum critical velocities might not be satisfied (Bertrand-Krajewski, 2002). Thus, deposition generally occurs during these periods and also during decelerating flows when storm runoff is receding. Although the flow of surface runoff into the sewer network generates considerable rates of shear stresses, this does not guarantee proper sediment transport in downstream sewer pipes due to lack of enough strength of the flow to constantly produce the required shear stresses. Hence deposition is likely to occur, which can generate problems such as hydraulic overloading due to a reduction in flow capacity and increase of the risk of surcharging during storm events. This could result in both surficial flooding and unwanted early operation of combined sewer overflows (CSOs). Thus, the issue of designing sewer systems to be self-cleansing becomes important. This is however not always promising, particularly in flat regions, where the necessary slopes for sewer pipes to be self-cleansing are not available (especially in the most upstream parts of the network) due to the costs of required deep excavations and pumping systems. In this regard, the use of flushing devices that generate controlled flush waves into the downstream sewer system once the tank is completely filled with runoff water could be a proper solution. The effect of such devices as external sources to help remove the settled particles from sewer pipes has been investigated by many researchers (Bertrand-Krajewski et al, 2005; Bouteligier et al, 2006; Campisano et al, 2004; Dettmar et al, 2002). In the paper an evaluation of the proper implementation of flushing tanks for eroding sediments from a combined sewer network in the village Erpe-Mere in Flanders, Belgium, is presented based on experiments and numerical simulations carried out at the Hydraulics Laboratory of the Katholieke Universiteit Leuven (K.U.Leuven). Implementation of these flushing tanks in different locations of a sewer network is studied with regard to subsequent modifications of hydrodynamic components of the flow throughout the network (shear stress, flow discharge, flow velocity). The research takes account of the hydraulic characteristics of the flushing tank (released flow rate as a function of time and the specific effects on in-sewer sediment transport). The methodology consists of utilizing version 7.5 of InfoWorks CS (Wallingford Software, UK) in order to calculate the resulting spatially distributed shear stresses as a function of the pipe diameters and slopes to evaluate eroding capabilities of the generated flush waves in an existing combined sewer network. Emphasis is given on verifying whether the simulation-based shear stresses throughout the sewer network satisfy the required self-cleansing conditions reasonably, i.e. generated shear stresses remain high enough for a small time interval. Of particular concern is the proper location of multiple flushing devices all over the combined sewer network with respect to the potential degree of sediment removal and transport, and the possible drawbacks which could occur during such flushing events such as modification of the cross section of the flow and consequent effects on sewer surcharging or flooding. In fact, there are various influencing parameters engaged in the proper installation of these devices such as sewer network characteristics, contributing catchment characteristics, etc. The evaluation results indicate that attaining a minimum shear stress criterion (e.g. 3 N/m2 for combined sewer networks based on the Flemish sewer design requirements) could not be satisfied in all parts of the considered sewer network and even some undesired effects emerged. Nevertheless, regarding the accomplished modelling analyses, the capability of such devices to produce effective forces for removal of the settled particles in combined sewer networks is well accounted for.status: publishe