Approaches for estimating mixing time in a water storage tank

Abstract

The mixing performance of a water storage tank is important to ensure good drinking water quality in a distribution system. A tracer study is a commonly used approach for mixing performance evaluation but has limitations, such as limited sampling points and inability to be applied to the full scale at design stage. In practice, mixing time is usually estimated by the tracer curve measured by a probe sensor in a physical tracer study. However, the probe tracer concentration curve can be affected by many factors in the tracer release approach (e.g., instantaneous pouring or quill injecting), such as pouring volume, vertical penetration, and injecting time, making the estimated mixing time inaccurate. Coefficient of variation (CoV) decay curve is an alternative for determining mixing time. In this study, computational fluid dynamics (CFD) models were developed to study mixing in a cylindrical water tank. Mixing times were estimated from both probe tracer curve and CoV decay curve. Results show that the mixing time estimated from a probe tracer curve only represents local mixing, while that estimated from a CoV decay curve is a more appropriate index for global mixing, suggesting that the CoV decay curve should be used for mixing performance analysis in a water tank.