Search

The present experimental study focuses on the characterization of complex mixing phenomena at pipe intersections within pressurized water distribution networks. To examine the complete mixing assumption at a cross junction, a series of experiments were conducted in the turbulent regime (R > 10,000). The experimental setup consists of a cross junction with various sensors, pumps, and a data acquisition system to accurately measure solute concentration. Selected experimental results are compared to computational fluid dynamics (CFD) results. In addition, the water quality model associated with a standard water distribution network simulator (EPANET) was reevaluated based on CFD and experimental data. Corrections based on experimental results are incorporated into EPANET (AZRED 1.0) for use in a case study. The study concludes that the complete mixing assumption can potentially create considerable errors in water quality modeling. Further, severe errors are likely to occur in systems with many cross type junctions due to bifurcation of the incoming flows. DOI: 10.1061/(ASCE)0733-9496(2008)134:3(295) CE Database subject headings: Transport phenomena; Water quality; Water distribution systems; Potable water; Experimentation.

The present study investigates solute mixing phenomena at various flow rates within a cross junction, which is commonly found in municipal drinking water distribution systems. Simulations using computational fluid dynamics are employed to model the solute concentrations leaving the junction when one inlet is comprised of clean water while the other inlet carries a solute at R > 10,000. For a few exemplary cases, the resulting velocity vectors and contours of dimensionless concentration are presented to explain the detailed mixing mechanisms at the impinging interface. The turbulent Schmidt number ([Sc.sub.t]), an important scaling parameter, is also evaluated. Experimental results were used to assess values of [Sc.sub.t] for various flow conditions that accurately captured the detailed mixing processes within the junction. The present study clearly indicates that mixing at pipe cross junctions is far from 'perfect.' Incomplete mixing results from bifurcating inlet flows that reflect off one another with minimal contact time. Improving the existing water quality model based on accurate mixing data and simulations is important not only to predict concentrations of chemical species such as chlorine in water distribution systems, but also to prepare for potential intentional and accidental contamination events. DOI: 10.1061/(ASCE)0733-9496(2008)134:3(285) CE Database subject headings: Computational fluid dynamics technique; Mixing; Water distribution systems; Numerical models; Potable water.