Date of this Version
Structural best management practices (BMPs) are used to capture and treat stormwater runoff. Most structural BMPs provide treatment by filtering runoff through a filter media or collecting it in a detention basin and slowly discharging it over an extended period of time to allow suspended solids and associated contaminants to settle out. The purpose of this study is to design an effective outlet structure that provides adequate filtration or slows discharge to 40 hours.
A model detention basin was constructed in the Civil Engineering Hydraulics Laboratory at the University of Nebraska-Lincoln (UNL) and two full scale outlet structures were tested in it. The first outlet device was an orifice controlled perforated riser. Discharge from the device was measured at many head levels and the results correlated well with discharge given by the orifice equation. The orifice controlled perforated riser adequately provided a 40 hour drain time and can be sized for various detention basin sizes using the orifice equation. At low heads, however, it was observed that perforations in the riser pipe could also control flow rates, depending on the size, elevation, and number of the lowest perforations.
The second outlet structure tested was a filtered perforated riser. An 18” diameter barrel was placed around the perforated riser and filled with coarse (D50=0.11 in) sand. Fifteen tests were run where sediment laden water was cycled through the filtered riser. The device provided good filtration but showed significant clogging. The filter media, in series with the orifice, impacted flow rates and was modeled using an unconfined aquifer equation. The unconfined aquifer equation was used to estimate changes in hydraulic conductivity as the filter began to clog with sediment. However, clogging of the filter screen was also observed and was modeled as a minor loss. Estimation of the minor loss coefficient provided a better fit to the data than the hydraulic conductivity. Therefore, it was concluded that the clogging of the filter screen was the significant driver of the head loss. Both methods of estimating clogging showed a dramatic initial increase in head loss, followed by much smaller increases. When designing the filtered riser these changes in head loss should be considered and the filtered riser should be sized based on flow rates after initial clogging.
Advisor: David M. Admiraal