Design Optimization of Hydrodynamic Separators

Typically, the sizing of hydrodynamic separators (HDS) for stormwater treatment is based on two factors. The first factor is the ability to remove a designated fraction of a particle size distribution, usually with a d₅₀ of 100 μm or greater. The second factor designates the magnitude of flow that needs to be treated from a given site. The techniques for determining the magnitude of flow vary, but two basic methods dominate. The first is a fairly simplistic method that requires the HDS to be sized to remove a target percentage of a specified particle size at a defined flow rate. The flow rate can be determined using a specified rainfall intensity and storm return period with a hydrologic model output, such as the Santa Barbara Urban Hydrograph. The second method, sometimes called the average annual load method, is based on calculations which use the storm depth (or intensity) frequency distributions converted to flows through a unit, which are then used to predict the removal performance. The removal performance prediction is based on laboratory studies of specific gradations of silica sediment, which have been tested to create a performance graph. These graphs often show plots of removal performance to the point where zero removal is achieved. These flows at zero removal are sometimes termed the "peak treatment flow". This paper provides a perspective on this methodology and an approach to how sizing methods can be changed to optimize the total load reduction of a system by throttling the flow to a rate that is less the "peak treatment flow".