Field and CFD Analysis of Jet Aeration and Mixing Randal W. Samstag1*, Edward A. Wicklein1, Roderick D. Reardon2, Robert J. Leetch3, R. (Robbie) M. Parks3, and Colin D. Groff3 1 Carollo Engineers, Seattle, Washington. 2 Carollo Engineers, Winter Park, Florida. 3 JEA, Jacksonville, Florida. * To whom correspondence should be directed: [email protected]. ABSTRACT Field tests and three-dimensional computational fluid dynamics (CFD) modeling were conducted to evaluate effectiveness of mixing in a jet aeration system for a sequencing batch reactor (SBR). The SBR is part of the Blacks Ford Regional Water Reclamation Facility (BFRWRF) of the JEA utility in Jacksonville, Florida. Solids measurements were taken with the SBR tanks operating in both aerated and un-aerated pumped mixing modes. A density-coupled CFD model incorporating solids settling and transport as user-defined functions (UDF) was calibrated to the field solids profile data and used to identify the capacity of the system to maintain solids suspension with different power levels in both mixing modes. The modeling indicated that while the existing system was adequate to ensure complete mixing during aerated periods, the existing pumped mixing power level of approximately 7.7 W/m3 (39 horsepower per million gallons – hp/MG) would need to be increased to at least 30.8 W/m3 (156 hp/MG) to maintain a conventional mixing criterion during un-aerated mixing. As a follow up to the mixing evaluation, simulations were prepared in which density coupling of the solids transport model to the fluid momentum was turned off in the UDF. Results from these experiments indicated that the neutral density simulation (no density coupling) seriously over-predicted the degree of mixing seen in the calibrated density-coupled model. Since neutral density CFD evaluations for mixing are a common practice in the industry, this finding is significant. KEYWORDS: Mixing, solids profile, CFD, sequencing batch reactor, density-coupled, simulations INTRODUCTION One of the recommendations of a previous study for the Blacks Ford Regional Water Reclamation Facility (BFRWRF) by Carollo Engineers (2010) was to perform a computational fluid dynamic (CFD) evaluation of the jet aeration mixing system for the sequencing batch reactor (SBR) system at the BFRWRF. The BFRWRF includes influent screening, influent equalization, SBR treatment in four SBR tanks, effluent equalization, cloth media filtration, and ultraviolet (UV) disinfection. Figure 1 presents an aerial photo of the plant site. The SBR tanks at the BFWRF incorporate pumped jet mixing and jet aeration. This type of mixing / aeration system relies on recirculation of the contents of the SBR tank through a manifold with a series of high velocity discharge nozzles at the bottom of the SBR tank. The system is aerated by mixing the pumped flow with atmospheric air pressurized by blowers. The BFRWRF discharges to a reclaimed water distribution system or to the Blacks Ford Wetland Treatment System. Effluent quality requirements for wetland discharge include a limit on total nitrogen compounds of 3 mg/L. To achieve this high degree of nitrogen removal, the SBR tanks operate in cyclic aerated and un-aerated modes. During the aerated mode, ammonia in the influent wastewater is converted to oxides of nitrogen. Removal of oxidized nitrogen by biological denitrification occurs during un-aerated pumped mix cycles. It had been thought that inadequate mixing in the SBR system during un-aerated pumped mix cycles might be inhibiting the ability of the BFRWRF to remove nitrogen. The following paper presents a summary of field tests and CFD investigations undertaken to evaluate pumped and air mixing in the SBR system at the BFRWRF. MATERIALS AND METHODS Field tests were conducted at the BFRWRF to establish solids concentration profiles under normal operating conditions. Solids concentration measurements were taken at multiple depths at two locations at the edge of operating SBR tank. Solids measurements were calibrated to laboratory samples of mixed liquor suspended solids using an Insite Instrumentation Group Model 3150 portable suspended solids analyzer. Simulations were conducted using FLUENT, Release 13.0 provided by ANSYS, Inc. The geometric model and computational mesh were developed in ANSYS GAMBIT, Version 2.4.6. User defined functions (UDF) were used to model solids settling and transport and to implement density coupling with the fluid flow (See Wicklein and Samstag, 2009.) Aeration was simulated using the standard multi-phase input facilities of FLUENT. BACKGROUND SBR Data Each SBR tank at the BFRWRF is 25.9 m (85 foot - ft) in diameter and has a volume of 3,407 m3 (0.9 MG). Each tank includes a jet aeration system with three jet headers and three mixing pumps. The main jet header in each tank contains 24 discharge nozzles and receives a flow of 23,950 m3/d (4,394 gallons per minute - gpm) from a 30 kW (40 hp) end suction centrifugal pump mounted outside each tank. In addition, each tank includes two supplemental jet headers, each with six (6) discharge nozzles, which receive flow from individual 5,900 m3/d (1,080 gpm) submersible 5.6 kW (7.5 hp) pumps. SBR Tanks Figure 1. Blacks Ford Regional Water Reclamation Facility Plant Site. Table 1 presents hydraulic data for the BFRWRF SBR tanks pumped mix system. In addition to the mixing pumps, each tank system includes a 93 kW (125 hp) multistage centrifugal blower with a capacity of 1,530 – 3,300 m3/hr (900 – 1900 cubic foot per minute - cfm). The jet aeration system discharges into the SBR tank through 76 mm (3-inch – in) diameter nozzles. The air mixing system incorporates a smaller, inner nozzle, 38 mm (1-1/2 in) in diameter. Pump mixing power was estimated by applying a multiplier of 1.25 to the velocity head through the 38 mm diameter inner nozzles. This produces the approximate total dynamic head (TDH) of 6.5 m (21 ft) on the mixing pumps and a total estimated power consumption for the system of approximately 26 kW (35 hp). Table 1. Hydraulic Characteristics of Existing SBR Pump Mix System. Description Value Main Header Pump Flow (m3/day) 23,949 Number of Nozzles per Header 24 Outer Nozzles Nozzle Diameter (mm) 76.2 Jet Velocity (m/sec) 2.53 Inner Nozzles Nozzle Diameter (mm) 38 Jet Velocity (m/sec) 10.1 Supplemental Headers Number of Headers 2 Pump Flow, Each (m3/day) 5,886 Number of Nozzles per Header 6 Outer Nozzles Nozzle Diameter (mm) 76.200 Jet Velocity (m/sec) 2.49 Inner Nozzles Nozzle Diameter (mm) 38.1 Jet Velocity (m/sec) 9.96 Total Pump Flow (m3/day) 35,722 Average Velocity (m/sec) 2.52 Inner Nozzle Total Pumping Head Power (kW) 26.2 Motor horsepower rating (kW) 41 Geometric model Figure 2 presents an illustration of the three-dimensional geometric model prepared for the BFRWRF SBR tanks for CFD analysis. The model shows the three jet headers, the main header pump intake, the auxiliary header pumps and intakes, and the effluent decanters. The effluent decanters were not required for the flow simulation, but were included to simulate the fluid environment. Figure 3 illustrates the polyhedral computational mesh implemented in the meshing software projected onto model surfaces. Figure 2. SBR Tank Geometric Model. Figure 3. SBR Tank Computational Mesh. RESULTS Field test results Field tests were conducted at the BFRWRF to establish solids concentration profiles under normal operating conditions. Solids concentration measurements were taken at multiple depths at two locations at the edge of the operating SBR tank No. 3 using a calibrated optical solids measurement probe. Testing took place on July 8, 2011. Sample locations were: 1) near to the wall at one location adjacent to the main platform for Tank Number 3 between the effluent decanters and 2) adjacent to the access ladder for one auxiliary header submersible pump. Sample locations are shown schematically in Figure 4. Figure 4. Field Test Sample Locations. Mixed liquor suspended solids (MLSS) measurements were made: 1) during mixed cycles with both air and pumping operational and 2) during pumping-only mix cycles. The measured MLSS concentration for the day of the test was approximately 2,400 mg/L and the sludge volume index (SVI) was approximately 100 mL/g. Figure 5 presents results of sampling during a period when both the pumped mix system and the aeration blower were operational. The data indicate a fully mixed condition with a slight accumulation of floatable solids. Solids concentrations were in the range of 2,320 to 2,380 mg/L for the four measurement locations below the water surface and 2,550 mg/L for the measurement at the water surface. Depth (m) Air On ‐ Main Platform Legend 0.0 2550 Concentration (mg/L) 1.5 2380 < 1000 650 3.0 2350 1000 < C < 1500 1250 4.6 2370 1500 < C < 2000 1800 6.1 2320 2000 < C < 2500 2450 2500 < C < 3000 2650 > 3000 3400 Figure 5. Field Test Results with Air On. Figure 6 presents results from sampling at the location of the main platform during a period with the mixing pump in operation and with the aeration blower off. A series of four measurements were made at the main platform location at different times following shutdown of tank mixing pumps: 1) immediately after shutdown, 2) 25 minutes after shutdown, 3) 66 minutes after shutdown, and 4) 83 minutes after shutdown. The data indicate that at this location, concentrations less than 1,000 mg/l were present at the top 1.5 m level of the tank after 83 minutes of mixing and reached a maximum of 2,950 mg/L at the tank bottom after 25 minutes of mixing. At later times, the measured bottom concentrations were approximately 2,600 mg/L. Figure 7 shows results of measurements adjacent to the location of the auxiliary header submersible pump access ladder. The access location in the test tank was on the opposite side of the tank from that shown in the figure. The figure is based on the tank model, which was based on manufacturer’s installation drawings. After 42 minutes of mixing, the solids concentration at the 1.5 m level layer was 1,380 mg/L and the concentration at the bottom was measured at 2,200 mg/L. After 75 minutes of pumped mixing, the top-level concentration was 650 mg/L and the bottom concentration was 2,450 mg/L. Conclusions from the field test measurements include the following: 1) Distinct separation takes place during pumped mix cycles under the current configuration with average MLSS concentrations of 2,400 mg/L separating to less than 1,000 mg/l in the top layer and nearly 3,000 mg/L in the bottom layer. 2) Measurements indicate the dynamic mixing condition in the tanks. Solids concentrations at the bottom of the tank varied from 2,290 mg/L at the beginning of the pumped mix cycle at the main platform location to 2,950 mg/L after 25 minutes but reduced to 2,650 mg/L after 66 minutes and 2,540 mg/L after 83 minutes. Depth (m) Main Platform ‐ 0 minutes Legend 1.5 2230 Concentration (mg/L) 3.0 2210 < 1000 650 4.6 2100 1000 < C < 1500 1250 6.1 2070 1500 < C < 2000 1800 6.4 2290 2000 < C < 2500 2450 2500 < C < 3000 2650 Depth (m) Main Platform ‐ 25 minutes > 3000 3400 1.5 2150 3.0 2270 `` 4.6 2390 6.1 2950 Depth (m) Main Platform ‐ 66 minutes 1.5 2450 3.0 2350 4.6 2550 6.1 2640 Depth (m) Main Platform ‐ 83 minutes 0.0 650 1.5 2510 3.0 2590 4.6 2690 6.1 2540 Figure 6. Main Platform Field Pump Mix Test Results – Solids Concentration. Depth (m) Pump Ladder ‐ 42 minutes Legend 1.5 1380 Concentration (mg/L) 3.0 1920 < 1000 650 4.6 2180 1000 < C < 1500 1250 6.1 2200 1500 < C < 2000 1800 2000 < C < 2500 2450 Depth (m) Pump Ladder ‐ 75 minutes 2500 < C < 3000 2650 0.0 650 > 3000 3400 1.5 1590 `` 3.0 2450 4.6 2610 6.1 2450 Figure 7. Pump Ladder Field Pump Mix Test Results – Solids Concentration. Initial model simulations Using the model geometry described in the previous section, CFD simulations were initially configured to approximate current conditions of operation of the pumped mixing and mixing / aeration system. The model was configured with fixed influent boundary conditions for velocity through the inlet nozzles and outlet boundary limitations on flow to the pump inlets. Two thirds of the total inlet flow was constrained to exit the tank through the main pump outlet and one sixth of the total flow through each of the auxiliary header outlets. As shown in Table 1, existing pump capacities deliver inlet port velocities of approximately 2.5 m/sec (8.3 feet per second - fps). Figure 8 presents a graphic illustration of velocity profiles at three cross sections through the SBR tank at locations centered on inlet nozzles in the pump headers. The velocity is high near the nozzle, but the velocities rapidly diminish as the fluid jet is dissipated into the main tank contents. Velocities are less than 0.13 m/sec (0.43 fps) over a majority of the tank. Figure 9 illustrates the velocity profile centered on one of the nozzles. The model was configured with custom UDF to simulate suspended solids settling and transport as a scalar within the tank. These UDF also simulate the effect of density changes resulting from the solids concentration profile on fluid momentum. Simulations were initiated with an assumed uniform concentration of 2,400 mg/L mixed liquor suspended solids (MLSS) throughout the tank. Settling velocities were calculated based on an assumed value for the sludge volume index (SVI) of 100 mL/g, using the revised Daigger equation (Daigger 1995). These values were used to approximate conditions at the BFRWRF during the field-testing. Simulations were paused to estimate the solids concentration profile at different times after initiation of a pumped mixing cycle. Figure 10 presents the estimated solids concentration profile following different times after the start of the pumped mix cycle: 25, 44, and 66 minutes. The figure shows a small layer of relatively clear water at the top of the tank that is predicted after 25 minutes of mixing with the clear layer increasing to approximately one-eighth of the tank depth after 66 minutes. Corresponding solids concentration buildup in the bottom of the tank is predicted to concentrations in excess of 3,500 mg/L, even after only 25 minutes of mixing. Comparison of solids concentration CFD results with the results from the field test gives a reasonable qualitative agreement, although the CFD model results indicate modestly greater solids separation than seen in the field. This could be due to several factors: 1) slight over- prediction of settling velocities or 2) slight inaccuracies in the field measurements.