Hydrocyclone Implementation at Two Wastewater Treatment Facilities To Promote Overall Settling Improvement
Hydrocyclone density-driven particle separation may offer up improved settling performance for wastewater treatment facilities experiencing poor settleability. Hydrocyclones are fed mixed liquor through the feed inlet and experience a centrifugal motion that separates solids based on density. The variation in hydrocyclone nozzle sizes will report different calculated hydraulic and mass split percentages for the overflow and underflow. Previous research conducted with hydrocyclones have at multiple full-scale facilities used a 10 m3/hr hydrocyclone to promote better settleability as well as aid the formation of aerobic granular sludge (AGS). There has been a multitude of settling improvement experiments and initiatives for full scale wastewater treatment. However, little research has been produced utilizing larger hydrocyclones (20 m3/hr) at a full-scale wastewater treatment facility during continuous operation.
Two Hampton Roads Sanitation District (HRSD) plants served as sites for this research: James River (JR) Wastewater Treatment Plant located in Newport News, VA and Urbanna (UB) Wastewater Treatment Plant located in Urbanna, VA. Both treatment facilities have utilized the hydrocyclone for more than two years, to fulfill wasting requirements. The JR plant operates the hydrocyclone continuously for wasting purposes, while UB only uses the hydrocyclone for approximately 30-45 minutes per day. In order to evaluate the effectiveness of the hydrocyclone and its overall impact on settleability at the JR plant, eight hydrocyclones were installed. JR samples were taken from the underflow sample port (representing a mixture of underflow samples representing the number of hydrocyclones operational at the sample time) and overflow samples were taken from the outfall point of a single hydrocyclone. The UB plant only operated one 5 m3/hr hydrocyclone on Treatment Train 1 during wasting operations, while Treatment Train 2 served as the control train for the duration of this research. Hydrocyclone performance at JR was assessed through direct measurement of hydraulic and mass split of the underflow and overflow components, initial settling velocity (ISV), sludge volume index (SVI), and
SVI5/SVI30 ratio. UB hydrocyclone and settling performance was measured by ISV, SVI5, SVI30, and SVI5/SVI30 ratios during different comparison experiments: hydrocyclone vs. no hydrocyclone, hydrocyclone vs. polymer addition, and hydrocyclone with polymer addition to Train 1 vs. polymer-only addition to Train 2. Nutrient concentrations from both treatment trains were collected and analyzed to determine any significant changes based on hydrocyclone use.
T-test statistical analysis, and a dose response analysis included direct measurements of the ISV, SVI5, SVI30, mass split percentages, along with the effect of polymer with and without the use of a mechanical selector.
Hydrocyclone settleability measurements at JR over time revealed a statistically significant positive correlation with the ISV, SVI5, and SVI30 measurements of the aeration effluent. Therefore, the hydrocyclone statistically had a strong impact on three settling parameters that are instrumental in determining overall settling efficiency. Statistically, no strong correlation was determined between the hydrocyclone operation and the total phosphorus (TP) concentration in the secondary effluent, or the ferric addition to the secondary clarifiers. The dose response based on the underflow ISV rate provided understanding of the nozzle comparison and the effect it provided to the underflow sample.
Hydrocyclone performance at UB was hindered by the re-seed of Train 1 (inDENSE™) due to over wasting, and most of the data were not representative. Before the re-seed, hydrocyclone performance was improving the overall settleability of the mixed liquor in comparison to Train 2 (Control). All settling parameters measured were in favor of the hydrocyclone operation. After the re-seed the plant mixed liquor changed microbial populations for a brief time and was not representative of the overall treatment efficacy. The hydrocyclone did provide a quicker settling velocity than the polymer addition when the polymer addition was steady, and through both polymeric spikes. Polymeric addition to both trains, while inDENSE™ train still employing the hydrocyclone did not provide any conclusive data as to whether polymer addition with the use of a hydrocyclone was more effective than polymer-only addition. Nutrient profiles from UB did not provide any change in NH4-N, NO3-N, NO2-N, or PO4-P, with the hydrocyclone being operational or not on the secondary clarifier effluent.