Evaluation of Double Perforated Baffles Installed in Rectangular Secondary Clarifiers

Transcription

1 water Article Evaluation of Double Perforated Baffles Installed in Rectangular Secondary Clarifiers Byonghi Lee Department of Environmental Energy Engineering, Kyonggi University, Suwon 6227, Korea; Tel.: Academic Editor: Andreas N. Angelakis Received: May 27; Accepted: 6 June 27; Published: 7 June 27 Abstract: Double perforated baffles in rectangular secondary clarifiers were studied to determine wher y contribute to producing high-quality effluents. The Computational Fluid Dynamics (CFD) simulations indicated that bio-flocculation occurred at front of baffle and longitudinal movement of settled sludge was hampered whenever clarifier had high inflow. Simulation results showed that rectangular clarifier with double perforated baffle produced an effluent with lower suspended solid (SS) concentrations than effluent from clarifier without baffle. To verify simulation results, a double perforated baffle was installed in two of 48 rectangular clarifiers in a 3, m 3 /d-capacity wastewater treatment plant. To study effect of baffle on solid removal, effluent turbidity of clarifier with and without double perforated baffle was measured simultaneously. Experimental data showed that double perforated baffle played a significant role in reducing effluent turbidity. The effluent turbidity reduction ratio with baffle decreased when Sludge Volume Index (SVI) of Mixed Liquor Suspended Solids (MLSS) was below ml/g. The overall average reduction ratio was 24.3% for SVI < ml/g and 45.% for SVI > ml/g. The results of this study suggest that double perforated baffles must be installed in secondary rectangular clarifiers to produce high-quality effluent regardless of operational conditions. Keywords: rectangular clarifier; secondary clarifier; double perforated baffle. Introduction Water reuse is global demand, and secondary effluent filtration has been commonly adapted to meet this demand. Whenever filtration rate flow is decreased to a certain level by accumulation of solids on filter surface, filter must be back washed. In general, backwashing water is sent back to head work of wastewater treatment facility. Since frequent backwashing entails high operational costs and high hydraulic loads for treatment facility, frequency should be decreased by lowering concentration of solids in secondary effluent that is influent to filtration facility. Many researchers studied bio-flocculation of MLSS (Mixed Liquor Suspended Solid) to reduce solids in secondary effluent. For circular secondary clarifiers, center well within clarifier was modified to a flocculation well to promote bio-flocculation []. For rectangular clarifiers, MLSS was slow-mixed before being introduced into clarifier to promote bio-flocculation [2]. In 983, ory and design concept of bio-flocculation were introduced [3]. In this concept, in rectangular secondary clarifiers, MLSS falls toward bottom of clarifier as soon as it enters clarifier owing to higher density of MLSS. The reduced MLSS creates a density flow along bottom of clarifier toward end of clarifier, while supernatant flows on top, towards front end of clarifier in opposite direction [4,5]. This density flow reduces effective volume of clarifier up to 5% [6]. Single porous baffle selection at center of clarifier Water 27, 9, 47; doi:.339/w9647

2 Water 27, 9, 47 2 of 7 was proposed to decrease reduction in effective volume by density current. The :25 (length scale) hydraulic scale-model study found that this baffle created steady flow and reduced density current [5]. When a solid wall starting from bottom and having an opening on top was installed in middle of a rectangular clarifier, wall reduced density current and created two compartments. The depth of clear water in second compartment was more than that in first compartment [7]. Krebs et al. [8] simulated rectangular clarifier with a porous baffle installed in middle of clarifier. They found that baffle created uniform flow upstream, which exhibited better sludge settling and higher sludge levels than downstream flow of baffle. Therefore, sludge settling could be increased when porous baffles were installed at multiple places within rectangular clarifiers. Furrmore, y concluded that porous baffles were advantageous over solid baffles because sludge, which settled upstream of solid baffle, might be carried over baffle which might cause high ESS (effluent suspended solid) concentrations. A study with a single porous baffle installed in middle of rectangular clarifier showed that ESS concentrations were less sensitive to influent flow variations and suggested that bio-flocculation could be taking place [9]. A solid baffle with a contracted tank bottom installed in middle of a laboratory-scale rectangular clarifier showed higher solid removal efficiency []. Many baffles having different shapes were installed in laboratory-scale clarifier to evaluate attenuation of lateral motion of liquid slush and porous baffles were best ones to decrease effect of inflow variation on liquid motion []. Four types of secondary clarifiers in Tokyo Metropolitan Area were evaluated and rectangular clarifiers with two intermediate diffusers were found to be far superior although diffuser type was not considered [2]. Hydrodynamics of secondary clarifier with intermediate solid walls was studied using computer simulations that suggested that baffle must be placed at bottom of tank to disrupt density current [3]. A laboratory-scale rectangular clarifier was constructed to determine effect of solid baffles and middle baffle with a suitable height was found to improve solid removal efficiency [4]. To calculate clarifier performance, Evaluation, Inc. surveyed secondary clarifiers with baffles and found that a larger number of internal baffles improved solid removal [5]. A three-dimensional (3-D) fluid mass conservative clarifier model was used to evaluate performance of a long rectangular secondary clarifier. The simulation results suggested installation of a solid baffle and two perforated baffles in series. The effluent launders aligned in longitudinal direction could produce an effluent with lower suspended solid concentrations. The predicted results from model agreed with field observations [6] and similar results were published [7]. Turbidity instead of SS concentrations was used to evaluate effect of bio-flocculation of MLSS at a full-scale plant [8]. For evaluation of rectangular secondary clarifier using procedure proposed by Clarifier Research Technical Committee s Protocol, effluent turbidity at secondary clarifier was measured continuously. The relationship between turbidity and secondary ESS concentrations was established and turbidity measurement data was converted to SS concentrations [9]. The SS and turbidity data in four filtration facilities were presented. The average ratio of SS-to-turbidity of secondary effluent that was influent to filtration facility was 2.55 although average ratio of effluent from filtration facility was.35 [2]. Several attempts were made to improve SS removal efficiencies by selecting solid and/or perforated baffles in rectangular clarifiers. Previous studies showed that solid baffles created a flocculation zone and perforated baffles increased effective detention time of MLSS. The double perforated baffle was proposed as a let diffuser of rectangular secondary clarifier [2]. This baffle can be used to utilize advantages of both solid and perforated baffles simultaneously. A double perforated baffle consists of two vertical slotted baffles which are located very closely to each or and slots of front baffle are aligned with boards of rear baffle. In this study, effect of double perforated baffle on solid removal in rectangular secondary clarifiers is evaluated by Computational Fluid Dynamic (CFD) simulations. This study also evaluated turbidity and SS concentrations in

3 Water 27, 9, 47 3 of 7 effluent from clarifiers with and without a double perforated baffle in a 3, m 3 /d capacity wastewater treatment facility. 2. Materials and Methods Two double perforated baffles were installed in two of 48 rectangular secondary clarifiers in Suwon Wastewater Treatment Plant located in Hwaseong City, South Korea, and a double perforated baffle was elected for each clarifier. The ESS concentration and effluent turbidity in clarifier with and without a double perforated baffle were measured to determine role of baffle in solid removal. A CFD model employed by McCorquodale et al. [7] was applied to simulate clarifier with and without a double perforated baffle. This model shows fluid and solid movement in clarifier as well as solid removal performance. The model incorporates solid settling equation proposed by Takacs et al. [22]. 2.. Physical Description of Test Facility The Suwon Wastewater Treatment Plant owned and operated by Suwon municipal government has two phases: Phases I and II have design capacities of 22, m 3 /d and 3, m 3 /d, respectively. Phase I has circular primary and secondary clarifiers and Phase II has rectangular primary and secondary clarifiers. Most of influent to plant is domestic wastewater. Phase II consists of four trains and each train has two bays. Each bay consists of six primary and secondary clarifiers and one bioreactor. At Phase II, wastewater is lifted about 3 m after being passed through preliminary treatment facility. The lifted and treated wastewater flows into 48 rectangular primary clarifiers and each unit is 3 m long and 6 m wide with an average depth of 3.5 m. The biological treatment process comprises eight bioreactors and 48 secondary clarifiers. The effluents from six primary clarifiers at each bay are combined and passed into a bioreactor. Each bioreactor contains anaerobic, anoxic, and aerobic tanks to biologically remove organic materials, nitrogen, and phosphorus. Three plug-flow passes exist in bioreactor and each pass is 77 m long and 6 m wide with an average depth of 5.5 m. At each bay, MLSS from end of third pass flows to a transfer channel that splits MLSS to six secondary clarifiers. Each rectangular clarifier has an inlet perforated wall having 27 holes with 5 mm diameter of each hole and overall opening is 58% of total area, as shown in Figure. This wall is to provide equal flow across clarifier s cross-section. Figure 2a shows longitudinal view of conventional rectangular clarifier. A skimmer flight in each clarifier moves down to tank bottom at front end of effluent weirs and reafter serves as a sludge collector flight. The sludge collector flight travels counter-currently with respect to flow through clarifier. After dumping accumulated solids to hopper at front end of clarifier, flight moves up to water surface and travels co-currently as a skimmer flight. Each clarifier is 58 m long and 6 m wide with an average depth of 5 m and floor has a slope of %. Effluent launders are at top of end wall and two side walls stretched from end of clarifier to 7.5 m toward front end. The effluent flows over V-notch weir on launders and notch on weir is on 2 cm centers. Each clarifier has a total weir length of 36 m. The design SOR (Surface Overflow Rate) for this clarifier is.74 m/h Physical Description of Test Clarifier Two double perforated baffles were placed in secondary clarifiers numbered second train, third bay, No. 5 and 6 clarifiers and each clarifier has one baffle. The double perforated baffle contains two vertical slotted baffles where boards of first baffle are opposed to slots of second baffle, as shown in Figure 3. The first baffle has 5 boards and each board is 5.75 m long and.5 m width with a thickness of 5 cm. The boards of second baffle have same shape as that of first baffle and are placed opposed to slots of first baffle.

4 turbidity measurements were hampered from March to August of 25 due to contamination of turbidity turbidity measurements sensor by grease were hampered and construction from March activities to August in of treatment 25 due to facility. contamination Starting from of September, turbidity sensor experimental by grease period and was construction shortened activities to a few days in because treatment of facility. manual Starting cleaning from of turbidity September, meter sensors. experimental At each period experimental was shortened period, to grab a few samples days because of effluent of from manual both cleaning clarifiers of were turbidity taken meter once sensors. a day for At each SS measurement. experimental period, The SS grab was measured samples of according effluent to Standard from both Methods clarifiers Water[23]. 27, were Hourly 9, taken 47 once inflow a day rates for to Phase SS measurement. II, MLSS concentrations, The SS was and measured SVI (Sludge according Volume to Standard Index) of Methods MLSS 4 of 7 in [23]. Hourly aerobic inflow tank were rates provided to Phase by II, MLSS plant concentrations, staff. and SVI (Sludge Volume Index) of MLSS in aerobic tank were provided by plant staff. Figure Figure. Inlet. Inlet perforated wall in rectangular clarifier for for this this study. study. Unit Unit is mm. is mm. Figure. Inlet perforated wall in rectangular clarifier for this study. Unit is mm. Water 27, 9, 47 5 of 8 (b) Figure 2. Longitudinal view of rectangular clarifier without (Tank B) and (b) with a double Figure 2. Longitudinal view of rectangular clarifier without (Tank B) and (b) with a double perforated baffle (Tank A) installed at one-third of longitudinal length of clarifier. Unit is mm. perforated baffle (Tank A) installed at one-third of longitudinal length of clarifier. Unit is mm.

5 (b) Water 27, Figure 9, Longitudinal view of rectangular clarifier without (Tank B) and (b) with a double 5 of 7 perforated baffle (Tank A) installed at one-third of longitudinal length of clarifier. Unit is mm. Figure Figure 3. Detailed 3. drawings of of double perforated baffle installed two in two secondary rectangular rectangular clarifiers. clarifiers. Unit Unit is mm. is mm. In this study, a double perforated baffle was placed at one-third of longitudinal length of clarifier from inlet to improve solid removal efficiency by enhancing bio-flocculation and reducing settled sludge shoving by inflow surge (Figure 2b). The effluents from second train, third bay, No. 5 clarifier that had baffle (Tank A) and No. 4 clarifier that did not have baffle (Tank B) were simultaneously tested to determine effect of baffle. For comparison of effluent quality, turbidity and SS concentrations were measured. An on-line turbidity meter (Model TB-3, DKK-TOA Corporation, Tokyo, Japan) was installed inside effluent weir of both clarifiers with 6 cm of sensor depth (Figure 4). The turbidity meters were set up to measure turbidity every 5 min and turbidity data were later downloaded to a personal computer via COM port. Initially, turbidity and SS measurements were planned for a week every month for one year. However, turbidity measurements were hampered from March to August of 25 due to contamination of turbidity sensor by grease and construction activities in treatment facility. Starting from September, experimental period was shortened to a few days because of manual cleaning of turbidity meter sensors. At each experimental period, grab samples of effluent from both clarifiers were taken once a day for SS measurement. The SS was measured according to Standard Methods [23]. Hourly inflow rates to Phase II, MLSS concentrations, and SVI (Sludge Volume Index) of MLSS in aerobic tank were provided by plant staff.

6 Water 27, 9, 47 6 of 7 Water 27, 9, 47 6 of 8 Figure 4. Top view of 2nd train 3rd bay rectangular secondary clarifiers and locations of Figure 4. Top view of 2nd train 3rd bay rectangular secondary clarifiers and locations of double double perforated baffles and turbidity meters inserted at inside of effluent weir with depth of 6 perforated cm. Unit baffles is mm. and turbidity meters inserted at inside of effluent weir with depth of 6 cm. Unit is mm. 3. Results and Discussion 3. Results and Discussion 3.. Computational Fluid Dynamics Model Simulation 3.. Computational The software Fluidpackage Dynamics developed Model Simulation by McCorquodale et al. [7] was used to simulate behavior The of fluids software and package solids in developed clarifier. by For McCorquodale fluids, equations et al. [7] and was constants used presented to simulate by McCorquodale et al. [7] were used. For solids, empirical settling equation (Equation ()) behavior of fluids and solids in clarifier. For fluids, equations and constants presented by proposed by Takacs et al. [22] was used and coefficient of K was determined by settling tests McCorquodale et al. [7] were used. For solids, empirical settling equation (Equation ()) proposed proposed by Wahlberg [24] and K2 were determined by simulation study [25], respectively. by Takacs et al. [22] was used and coefficient of K was determined by settling tests proposed by Vs = Vo Wahlberg [24] and K2 were determined by (e K(X Xmin) simulation e K2(X Xmin) study ) [25], respectively. () where, Vo = Stokes velocity (settling Vs = Vvelocity o (e K(X Xmin) of a single particle e K2(X Xmin) in clear water), ) m/h. () K = empirical coefficient for rapidly settling floc resulting from fit of batch settling data, m 3 /kg. where, Xmin = concentrations of non-settling floc, kg/m 3. K2 = a settling exponent for poorly settling particles, m 3 /kg. V o = Stokes Table velocity shows (settling coefficients velocity and SS ofconcentrations a single particle used into clear simulate water), m/h. clarifier. All physical K parameters = empirical described coefficient in Material for rapidly and Methods settlingand floc resulting estimated from SORs fit of presented batch settling Figure data, 6 were m 3 /kg. Xmin used as = physical concentrations and inflow of input non-settling data for floc, simulation, kg/m 3. respectively. Figure 5a,b illustrate simulation results with SS contours and velocity vector fields of rectangular clarifier without K2 = a settling exponent for poorly settling particles, m 3 /kg. and with double perforated baffles under constant inflow conditions, respectively. These figures are generated by TecPlot, which imports output data from model. The double perforated baffle Table shows coefficients and SS concentrations used to simulate clarifier. All physical reduces velocity and creates turbulence that provides a zone for flocculation. Moreover, baffle parameters described in Material and Methods and estimated SORs presented in Figure 6 were breaks bottom density current and induces redistribution of inflow over clarifier depth. used as This physical redistribution and inflow creates input quiescent data conditions for simulation, that enhance solid respectively. settling after Figure baffle. 5a,b illustrate simulation results with SS contours and velocity vector fields of rectangular clarifier without and with double perforated Table. Parameters bafflesused under in constant Computational inflowfluid conditions, Dynamic (CFD) respectively. simulation. These figures are generated by TecPlot, which imports Elements output Valuedata from Remarks model. The double perforated baffle reduces velocity and creates FSS, turbulence kg/m 3 that.5 provides axmin zone in Equation for flocculation. () Moreover, baffle ESS, kg/m 3.9 breaks bottom density current and induces redistribution of inflow over clarifier depth. Vo, m/h 4.78 This redistribution creates quiescent K, m 3 /kg conditions.242 that enhance solid settling after baffle. K2, m 3 /kg 7. MLSS, kg/m 3 3.

7 Water 27, 9, 47 7 of 7 Table. Parameters used in Computational Fluid Dynamic (CFD) simulation. Elements Value Remarks FSS, kg/m 3.5 Xmin in Equation () ESS, kg/m 3.9 Vo, m/h 4.78 K, m 3 /kg.242 K2, m 3 /kg 7. MLSS, kg/m 3 3. Water 27, 9, 47 7 of 8 (b) Figure 5. CFD simulation results showing suspended solid (SS) contours and velocity vector fields in Figure 5. CFD simulation results showing suspended solid (SS) contours and velocity vector fields in secondary rectangular clarifiers without (Tank B) and (b) with double perforated baffles (Tank A). secondary rectangular clarifiers without (Tank B) and (b) with double perforated baffles (Tank A). The model proposed by McCorquodale et al. [7] simulates clarifier under variable inflow The conditions. modelhourly proposed SORs by can McCorquodale be used as input et data al. to [7] determine simulates effect clarifier of changes under in inflow variable on SS inflow conditions. removal. Hourly Video output SORsis can created be used to display as input hourly data variations to determine of solids and effect velocity of vectors. changesthe invideo inflow on SS removal. shows Video turbulence output at is created front of to display baffle (Figure hourly 5b) variations and retardation of solidsof and longitudinally velocity vectors. settled The sludge movement caused by high inflow during morning hours. Figure 6 presents hourly ESS video shows turbulence at front of baffle (Figure 5b) and retardation of longitudinally concentration variations along with SORs and shows that ESS concentrations in clarifier with settled sludge movement caused by high inflow during morning hours. Figure 6 presents hourly baffle are always lower than those in clarifier without baffle. The ESS concentrations in ESS concentration variations along with SORs and shows that ESS concentrations in clarifier with clarifier without double perforated baffle were proportional to SORs; however, ESS baffle concentrations are always in lower clarifier thanwith those indouble clarifier perforated without baffle were baffle. relatively The ESS constant concentrations regardless of in clarifier SOR without changes. Bio-flocculation double perforated baffle front of were baffle proportional resulted to in SORs; lower ESS however, concentrations ESS concentrations in in clarifier, with baffle double under perforated various baffle inflow were conditions relatively (Figure constant 6). The regardless reason for of high SOR ESS changes. Bio-flocculation concentrations in front clarifier of without baffle resulted baffle in at lower high SORs ESS is concentrations that high longitudinal in clarifier, velocity with bafflecreated underby various high SOR inflow pushes conditions settled (Figure sludge 6). to The end reason of forclarifier high ESS and concentrations eventually increases ESS clarifier without concentrations. baffle at However, high SORs is baffle thatin high clarifier longitudinal attenuates velocity lateral created motion by of highsettled SORsludge pushes and diminishes effect of high inflow on ESS concentrations. According to simulation results, settled sludge to end of clarifier and eventually increases ESS concentrations. However, double perforated baffle must be installed in secondary rectangular clarifiers to achieve low ESS baffle in clarifier attenuates lateral motion of settled sludge and diminishes effect of concentrations, if inherited condition of wastewater treatment plant that has high inflow high fluctuation inflow on is ESS considered. concentrations. According to simulation results, double perforated baffle must be installed in secondary rectangular clarifiers to achieve low ESS concentrations, if inherited.8 25 condition of wastewater treatment plant that has high inflow fluctuation is considered. Surface Overflow Rate (SOR), m/hr ESS Concentration, mg/l Time, hour SOR with Double Perforated Baffle without Double Perforated Baffle Figure 6. CFD simulation results of rectangular secondary clarifier under diurnal variations of

8 concentrations in clarifier without baffle at high SORs is that high longitudinal velocity created by high SOR pushes settled sludge to end of clarifier and eventually increases ESS concentrations. However, baffle in clarifier attenuates lateral motion of settled sludge and diminishes effect of high inflow on ESS concentrations. According to simulation results, double perforated baffle must be installed in secondary rectangular clarifiers to achieve low ESS Water 27, 9, 47 8 of 7 concentrations, if inherited condition of wastewater treatment plant that has high inflow fluctuation is considered. Surface Overflow Rate (SOR), m/hr ESS Concentration, mg/l Time, hour SOR with Double Perforated Baffle without Double Perforated Baffle Figure Figure 6. CFD 6. CFD simulation simulation results results of of rectangular secondary clarifier clarifier under under diurnal diurnal variations variations of of surface overflow rate. surface overflow rate Experimental Study As described in Material and Methods, two of 48 rectangular secondary clarifiers have double perforated baffles; one double perforated baffle in each clarifier. At first, ESS concentration in clarifiers with and without double perforated baffles was considered to be a surrogate to understand effect of baffles on SS removal in rectangular clarifiers. Since SS test is prone to being unpredictable, especially at low concentrations, and cannot be performed in short intervals, turbidity was chosen as a surrogate described by Walberg [8]. An in-line turbidity meter was installed at end of clarifier with double perforated baffle (Tank A) and without double perforated baffle (Tank B) to measure turbidity every 5 min. The ESS concentrations in effluent from both clarifiers were measured once a day during experiment except for 2 July measurement. The hourly inflow data during experiment were obtained from treatment plant personnel. In beginning of this study, effluent turbidity in both clarifiers was measured for one week every month. Turbidity meters were set up to measure and store data every 5 min and stored data were downloaded to a personal computer later. These downloaded data were compiled to produce hourly average turbidity values compared with ESS concentrations and variation of inflow. Table 2 shows MLSS concentrations in aerobic tank, SVIs of MLSS, ESS concentrations of grab samples in Tanks A and B, sampling times of each grab sample, and effluent turbidity in Tanks A and B at corresponding sampling times. MLSS concentrations and SVIs were provided by plant personnel and ESS concentrations were measured by author. SVIs were very low in November and December compared to January, February, and July. The plant personnel explained that DO (Dissolved Oxygen) concentration adjustment in aerobic tank might contribute to changes in SVI. Before November, DO concentrations were maintained at 7 8 mg/l to manage operational problems caused by construction activities around aerobic tanks and clarifiers. Since all construction work was finished in November, DO concentrations in aerobic tanks were maintained at 2 mg/l. The plant operator explained that decreasing DO concentrations might have caused low SVI of MLSS.

9 Water 27, 9, 47 9 of 7 Table 2. Mixed Liquor Suspended Solids (MLSS) concentrations, Sludge Volume Index (SVI) and effluent suspended solid concentrations, corresponding effluent turbidity and grab sampling time for suspended solid concentration measurement. Date MLSS Concentration in Aerobic Tank mg/l SVI of MLSS in Aerobic Tank, ml/g Effluent SS Concentration in Tank A, mg/l Effluent SS Concentration in Tank B, mg/l Effluent Turbidity in Tank A, NTU Effluent Turbidity in Tank B, NTU Grab Sampling Time / :27 / :43 / :4 / :43 / :29 / :25 2/ : 2/ :44 2/ :58 2/ : 2/ :25 2/ : 7/ :5 7/ : : : : : : : / : / :37 2/ :4 2/ :3 2/ :2 Mean Max Min SD Note: Max, Min and SD are Maximum, Minimum and standard deviation, respectively.

10 Water 27, 9, 47 of 7 Table 2 shows that effluent turbidity in Tank A is consistently lower than that in Tank B except for one case, although effluent suspended solid concentration in Tank A exceeds concentration in Tank B in cases out of 25. Wahlberg [9] proposed a relationship between SS and turbidity. In his study, relatively high values of SS concentrations and turbidity were analyzed to present a linear equation of turbidity against SS concentration. The ranges of SS concentrations and turbidity were near to 9 mg/l and 2 to 68 NTU, respectively. However, in this study relationship between ESS concentrations and effluent turbidity is not conclusive. The low values of SS concentrations and turbidity as shown in Table 2 may cause this inconclusiveness. The ESS concentrations and turbidity have ranges of.4 to 5.5 mg/l and.5 to.78 NTU, respectively. The sample containing low SS shows inconsistent SS concentration since inclusion and/or exclusion of particles in samples can vary concentrations easily. Since ESS concentrations were not consistent, it was an obvious choice to use turbidity as a surrogate to evaluate secondary clarifier performance. In Figures 7, hourly average effluent turbidity in clarifier with and without double perforated baffles is compared with SORs and ratios of effluent turbidity reduction caused by baffles are presented. The plan was to measure turbidity for one week every month in 26. While turbidity was accurately measured in January and February, it was unpredictably high from March to June. Discussion with plant personnel revealed that effluent from a nearby food waste treatment facility that discharged to plant for furr treatment contained grease that likely attached to turbidity meter sensor and caused erratically high turbidity values. In order to eliminate effect of grease on turbidity meter, meter s sensor was manually cleaned starting from July. From August to October, experiment was halted due to construction and maintenance activities in plant. Turbidity and ESS concentration measurements in effluent from both clarifiers resumed in November and December. Figure 7a shows hourly average effluent turbidity values in clarifier with double perforated baffle as Turbidity in Tank A and without double perforated baffle as Turbidity in Tank B in January, respectively. SOR represents hourly surface overflow rate based on hourly plant inflow. Figure 7b shows ratios of effluent turbidity reduction by baffle. In Figure 7a, effluent turbidity is proportional to SOR and effluent from clarifier with double perforated baffle has less turbidity except for a few cases. The average, maximum, and minimum effluent turbidity values in Tank A were.76, 3.42, and.34 NTU and in Tank B were.22, 2.65, and.6 NTU, respectively. Figure 7b shows that effluent turbidity reduction ratios were not changed significantly by SORs except for a few cases. The average reduction ratio was 3.8% while maximum and minimum ratios were 77% and 297%, respectively. The minimum value of 297% is likely to be caused by sensor contamination. Figure 8a,b show effluent turbidity from Tanks A and B with SORs and effluent turbidity reduction ratios in February, respectively. The effluent turbidity was proportional to SORs in both tanks, while SORs influence was more significant in Tank B (Figure 8a). The double perforated baffle played a significant role in reducing effluent turbidity hike when SOR surged, consistent with simulation results presented in Figure 6. The average, maximum, and minimum effluent turbidity values were.54,.5, and.28 NTU in Tank A and.25, 2. and.58 NTU in Tank B, respectively. Figure 8b shows that all effluent turbidity reduction ratios are positive, suggesting that clarifier with baffle always has higher solid removal efficiency. Moreover, re was no turbidity sensor contamination during February experiment. When elapsed time that inflow flows to secondary clarifier is considered, increase in ratio occurs when inflow is peaked. The average reduction ratio was 54.% and maximum and minimum reduction ratios were 8.% and.4%, respectively. Note that turbidity sensor had been manually cleaned to eliminate oil attached to sensor since July. Figure 9 presents 48 h experimental data in July. Figure 9a indicates that turbidity in Tank A is always lower than in Tank B and SOR hikes do not affect effluent turbidity as in January and February. The average, maximum, and minimum effluent turbidity values were.2,.32, and. NTU in Tank A and.4,.58, and.32 NTU in Tank B, respectively. The maximum effluent turbidity was lower than. NTU regardless of baffle installation. The

11 Water 27, 9, 47 of 7 maximum effluent turbidity in Tanks A and B were 3.42 and 2.65 in January and.5 and 2. NTU in February, respectively. The effluent turbidity in July was very low compared to January and February. Tank A always had lower effluent turbidity than Tank B, suggesting that perforated baffle plays a role in reducing effluent turbidity. However, Figure 9a does not show effluent turbidity hikes in Tank B when SOR increased dramatically as depicted in Figures 7a and 8a. The SOR hike does not likely contribute to effluent turbidity increase when effluent turbidity is as low as.6 NTU. Figure 9b shows effluent turbidity reduction ratios in July. There is always a turbidity reduction by baffle with an average value of 5.4% and SOR change does not contribute to reduction ratios. Figure a,b shows effluent turbidity from Tanks A and B with SORs and effluent turbidity reduction ratios in November, respectively. Figure a shows that effluent turbidity is proportional to SOR regardless of baffle installation. In contrast to Figures 7a and 8a, effluent turbidity in Tank A is not leveled. Although Tank B has lower effluent turbidity than that of Tank A during early hours of November 29, Tank A has lower turbidity during most of experimental period. Higher turbidity in Tank A effluent may be caused by contamination of turbidity sensor since turbidity meters were installed on 27 November and turbidity had been measured before manual sensor cleaning until 28 November. Tank A having a double perforated baffle has average, maximum, and minimum effluent turbidity values of.57,.88, and.34 NTU, respectively. Tank B having no double perforated baffles has average, maximum, and minimum effluent turbidity values of.7,.5 and.36 NTU, respectively. From data from Figure a, it is not clear why baffle in Tank A cannot significantly reduce effluent turbidity when SOR increases (Figures 7a and 8a). The reason could be range of effluent turbidity or low SVI (Table 2), because December data had same SVI range as November; however, turbidity in Tank A effluent was significantly reduced with high inflow (Figure a). Figure b shows that SOR does not have any significant role in effluent turbidity reduction ratios. The average reduction ratio was 2.5%. Figure a,b shows effluent turbidity in Tanks A and B with SORs and effluent turbidity reduction ratios in December, respectively. Tank A had average, maximum, Water 27, and 9, 47 minimum effluent turbidity values of.59,.87, and.28 NTU of 8 and Tank B has values of.79,.35, and. NTU, respectively. As in January and February, effluent turbidity was Tanks A and B with SORs and effluent turbidity reduction ratios in December, respectively. Tank A proportional to had SORs average, in December. maximum, and Although minimum effluent SVI had turbidity values same of.59, range.87, in and December.28 NTU and as in November Tank B has values of.79,.35, and. NTU, respectively. As in January and February, effluent (Table 2), effluent turbidity in Tank A was significantly reduced when SOR increased (Figure a), turbidity was proportional to SORs in December. Although SVI had same range in December as different from in November data in(table November. 2), effluent Figure turbidity in b Tank shows A was significantly effluent reduced turbidity when SOR increased reduction ratios by (Figure a), different from data in November. Figure b shows effluent turbidity reduction baffle. The average reduction ratio was 28.%. ratios by baffle. The average reduction ratio was 28.% Turbidity, NTU Hour/Date in January 26 SOR Turbidity at Tank A Turbidity at Tank B Figure 7. Cont. % 5% % -5% -% -5% -2% -25% -3% Turbidity Reduction Ratio, % -35%

12 Surface Water 27, 9, 47 Hour/Date in January 26 2 of 7 SOR Turbidity at Tank A Turbidity at Tank B.2 % % % -5% -% -5% -2% -25% -3% Turbidity Reduction Ratio, % -35% Hour/Date in January 26 SOR Effluent Turbidity Reduction Ratio (b) Figure 7. Time series of Surface Overflow Rate (SOR), effluent turbidity in Tanks A and B, and effluent turbidity reduction ratio by double perforated baffle during January experimental period. SOR and effluent turbidity in Tanks A and B; (b) Effluent turbidity reduction ratio by double perforated baffle. Figure 7. Time series of Surface Overflow Rate (SOR), effluent turbidity in Tanks A and B, and effluent turbidity reduction ratio by double perforated baffle during January experimental period. SOR and effluent turbidity in Tanks A and B; (b) Effluent turbidity reduction ratio by double perforated baffle. Water 27, 9, 47 2 of Turbidity, NTU Hour/Date in February SOR Turbidity at Tank A Turbidity at Tank B 5% % 5% % -5% -% -5% -2% -25% -3% Turbidity Reduction Ratio, % -35% Hour/Date in February 26 SOR Effluent Turbidity Reducton Ratio (b) Figure 8. Time series of SOR, effluent turbidity in Tanks A and B, and effluent turbidity reduction ratio by double perforated baffle during February experimental period. SOR and effluent turbidity in Tanks A and B; (b) Effluent turbidity reduction ratio by double perforated baffle. Figure 8. Time series of SOR, effluent turbidity in Tanks A and B, and effluent turbidity reduction ratio by double perforated baffle during February experimental period. SOR and effluent turbidity in Tanks A and B; (b) Effluent turbidity reduction ratio by double perforated baffle.

13 Water 27, 9, 47 3 of 7 Water 27, 9, 47 3 of Turbidity, NTU Hour/Date in July SOR Turbidity at Tank A Turbidity at Tank B 5% % 5% % -5% -% -5% -2% -25% -3% Turbidity Reduction Ratio, % -35% Hour/Date in July 26 SOR Effluent Turbidity Reduction Ratio (b) Figure 9. Time series of SOR, effluent turbidity in Tanks A and B, and effluent turbidity reduction ratio by double perforated baffle during July experimental period. SOR and effluent turbidity in Tanks A and B; (b) Effluent turbidity reduction ratio by double perforated baffle. Figure 9. Time series of SOR, effluent turbidity in Tanks A and B, and effluent turbidity reduction ratio by double perforated baffle during July experimental period. SOR and effluent turbidity in Tanks A and B; (b) Effluent turbidity reduction ratio by double perforated baffle. Water 27, 9, 47 4 of Turbidity, NTU Hour/Date in November 26 SOR Turbidity at Tank A Turbidity at Tank B Figure. Cont Hour/Date in November 26. 5% % 5% % -5% -% -5% -2% -25% -3% -35% Turbidity Reduction Ratio,%

14 Surface Over T Water 27, 9, 47 4 of 7 Hour/Date in November SOR Turbidity at Tank A Turbidity at Tank B Hour/Date in November 26 5% % 5% % -5% -% -5% -2% -25% -3% -35% Turbidity Reduction Ratio,% SOR Effluent Turbidity Reductio Ratio (b) Figure. Time series of SOR, effluent turbidity in Tanks A and B, and effluent turbidity reduction ratio by double perforated baffle during November experimental period. SOR and effluent turbidity in Tanks A and B; (b) Effluent turbidity reduction ratio by double perforated baffle. Figure. Time series of SOR, effluent turbidity in Tanks A and B, and effluent turbidity reduction ratio by double perforated baffle during November experimental period. SOR and effluent turbidity in Tanks A and B; (b) Effluent turbidity reduction ratio by double perforated baffle. Water 27, 9, 47 5 of Turbidity, NTU Hour/Date in December SOR Turbidity at Tank A Turbidity at Tank B 5% % 5% % -5% -% -5% -2% -25% -3% Effluent Turbidity Reduction Ratio, % -35% Hour/Date in December 26 SOR Turbidity Reduction Ratio (b) Figure. Time series of SOR, effluent turbidity in Tanks A and B, and effluent turbidity reduction ratio by double perforated baffle during December experimental period. SOR and effluent turbidity in Tanks A and B; (b) Effluent turbidity reduction ratio by double perforated baffle. Figure. Time series of SOR, effluent turbidity in Tanks A and B, and effluent turbidity reduction ratio by double perforated baffle during December experimental period. SOR and effluent turbidity in Tanks A and B; (b) Effluent turbidity reduction ratio by double perforated baffle. Table 3 presents average, maximum, and minimum effluent turbidity values in Tanks A and B as well as turbidity reduction ratios in each period, respectively. The average effluent turbidity in Tank A was always lower than that in Tank B although minimum values of reduction ratios in January, November, and December were negative, suggesting that turbidity in Tank A was higher than that in Tank B. These negative reduction ratios are probably caused by contamination of turbidity sensor. Many researchers [7,2,5] claimed that election of baffles in rectangular clarifiers could reduce effluent suspended solid concentrations. Bamumer et al. [8] showed that clarifiers with perforated baffles reduced larger amounts of suspended solids and suggested that multiple installations of perforated baffles could reduce concentrations of secondary effluent suspended solids to a greater extent. The simulation and experimental results of this study support ir claims although two perforated baffles were placed very closely. Table 3 presents average, maximum, and minimum effluent turbidity values in Tanks A and B as well as turbidity reduction ratios in each period, respectively. The average effluent turbidity in

15 Water 27, 9, 47 5 of 7 Tank A was always lower than that in Tank B although minimum values of reduction ratios in January, November, and December were negative, suggesting that turbidity in Tank A was higher than that in Tank B. These negative reduction ratios are probably caused by contamination of turbidity sensor. Many researchers [7,2,5] claimed that election of baffles in rectangular clarifiers could reduce effluent suspended solid concentrations. Bamumer et al. [8] showed that clarifiers with perforated baffles reduced larger amounts of suspended solids and suggested that multiple installations of perforated baffles could reduce concentrations of secondary effluent suspended solids to a greater extent. The simulation and experimental results of this study support ir claims although two perforated baffles were placed very closely. Table 3. Summary of experimental data. Month January February July November December Elements Effluent Turbidity in Tank A NTU Effluent Turbidity in Tank B NTU Effluent Turbidity Reduction Ratio % Average Max Min Average Max Min Average Max Min Average Max Min Average Max Min The average reduction ratio in November was lowest (2.5%) and second lowest was in December (28.%). SVIs in November and December were significantly lower than those in or months (Table 2). The overall average turbidity reduction ratio in January, February, and July when SVI of MLSS was less than ml/g was estimated as 45.%. However, overall average turbidity reduction ratio in November and December when SVI of MLSS was less than ml/g was estimated as 24.3%. Therefore, double perforated baffle is less efficient when MLSS that enters clarifier has SVI < ml/g. However, se results indicate that double perforated baffle plays a significant role in decreasing effluent turbidity under any operational conditions. Therefore, double perforated baffles installed in rectangular clarifiers would produce high-quality effluents that would lower operational costs of downstream filtration units by reducing backwashing frequency. Fewer backwashes would also lower hydraulic load to main stream treatment processes. Although double perforated baffle can reduce overall operational cost of treatment facility, possible cost saving may not outweigh installation cost of baffle. When baffle is considered, diurnal profile for secondary effluent turbidity along with influent flow must be constructed and overall turbidity reduction by baffle may be estimated by results from this study. Based on this estimation, benefits from baffle can be accounted against capital cost of baffle. Since double perforated baffle is permanent structure, maintenance cost for baffle may not be considered to analyzing benefit and cost of baffle. 4. Conclusions The performance of rectangular secondary clarifiers with double perforated baffles was evaluated by CFD simulations and field tests. The results are summarized as follows:

16 Water 27, 9, 47 6 of 7 The operational data obtained from field tests are represented well by CFD simulation results. Simulation results showed lower ESS concentrations in effluent from secondary clarifier with double perforated baffle and results of field test were consistent with simulations with a few exceptions, which might have been caused by sensor contamination. The double perforated baffle was found to reduce effluent solids concentrations. The rectangular secondary clarifier with hopper at front end inherits longitudinal movement of settled sludge toward end of clarifier when high inflow enters plant. This motion results in high effluent suspended solids concentrations. The double perforated baffle was shown to inhibit longitudinal movement and decrease effluent suspended solid concentration hike when high inflow enters plant. Since double perforated baffle can produce low turbidity effluent under any operational conditions, installation of this baffle in rectangular secondary clarifiers is strongly recommended to reduce overall operational costs of filtration units of wastewater treatment plants. Acknowledgments: This work was supported by Kyonggi University Research Grant 25. The author wishes to acknowledge help received from Chan-Hui Cho at Department of Environmental Energy Engineering of Kyonggi University and Suwon Wastewater treatment plant staff. Conflicts of Interest: The author declares no conflict of interest and founding sponsors had no role in design of study; in collection, analyses, or interpretation of data; in writing of manuscript, and in decision to publish results. References. Fischer, A.J.; Hillman, A. Improved Sewage Clarification by Pre-flocculation without Chemicals. Sew. Works J. 94, 2, Knop, E. Design Studies for Emscher Mouth Treatment Plant. J. Water Pollut. Control Fed. 966, 38, Parker, D.S. Assessment of Secondary Clarification Design Concepts. J. Water Pollut. Control Fed. 983, 55, Bender, J.; Crosby, R.M. Project Summary-Hydraulic Characteristics of Activated Sludge Secondary Clarifiers; US EPA: Cincinnati, OH, USA, 984; pp Kawamura, S. Hydraulic Scale-model Simulation of Sedimentation Process. J. Am. Water Works Assoc. 98, 73, Lopez, P.R.; Lavin, A.G.; Lopez, M.M.M.; de las Heras, B. Flow Models Rectangular Sedimentation Tanks. Chem. Eng. Process. 28, 47, [CrossRef] 7. Bretscher, U.; Krebs, P.; Hager, W.H. Improvement of Flow in Final Settling Tanks. J. Environ. Eng. 992, 8, [CrossRef] 8. Krebs, P.; Vischer, D.; Gujer, W. Improvement of Secondary Clarifiers Efficiency by Porous Walls. Water Sci. Technol. 992, 26, Baumer, P.; Volkart, P.; Krebs, P. Dynamic Loading Tests for Final Settling Tank. Water Sci. Technol. 996, 34, [CrossRef]. Ahmed, F.H.; Kamel, A.; Abdel Jawad, S. Experimental Determination of Optimal Location and Contraction of Sedimentation Tank Baffles. Earth Environ. Sci. 996, 92, Younes, M.F.; Younes, Y.K.; El-Madah, M.; Ibrahim, I.M.; El-Dannanh, E.H. An Experimental Investigation of Hydrodynamic Damping due to Vertical Baffle Arrangements in a Rectangular Tank. J. Eng. Marit. Environ. 27, 22, [CrossRef] 2. Kawamura, S. Integrated Design and Operation of Water Treatment Facilities, 2nd ed.; John Wiley and Sons Inc.: New York, NY, USA, 2; pp Tamayol, A.; Firoozabadi, B.; Ashjan, M.A. Hydrodynamics of Secondary Settling Tanks and Increasing Their Performance Using Baffles. J. Environ. Eng. 2, 36, [CrossRef] 4. Asgharzadeh, H.; Frroozabadi, B.; Afshin, H. Experimental Investigation of Effects of Baffle Configurations on Performance of a Secondary Sedimentation Tank. Sci. Iran. 2, 8, [CrossRef]

17 Water 27, 9, 47 7 of 7 5. Esler, K.E. Optimizing Clarifier Design and Performance. Available online: clarifierdesignperform/part/optimize.pdf (accessed on 28 April 27). 6. Zhou, S.; Vitasovic, C.; McCorquodale, J.A.; Lipke, S.; DeNicola, M.; Saurer, P. Improving Performance of Large Rectangular Secondary Clarifier. Available online: Rectangular_Clarifiers.pdf (accessed on 28 April 27). 7. McCorquoda, J.A.; Griborio, A.; Georgiou, I. Application of a CFD Model to Improve Performance of Rectangular Clarifiers. In Proceedings of WEFTEC, Dallas, TX, USA, 26 October 26; pp Wahlberg, E.J. Activated Sludge Bioflocculation. Ph.D. Dissertation, Clemson University, Clemson, SC, USA, 992; pp Wahlberg, E.J.; Stahl, J.F.; Chen, C.; Augustus, M. Field Application of Clarifier Research Technical Committee s Protocol for Evaluating Secondary Clarifier Performance: Rectangular Co-current Sludge Removal Clarifier. In Proceedings of WEFTEC, Anaheim, CA, USA, 3 October 993; pp Asano, T. Wastewater Reclamation and Reuse; Technomic Publishing Company Inc.: Lancaster, PA, USA, 998; pp Water Environmental Federation. Clarifier Design, 2nd ed.; McGraw-Hill: New York, NY, USA, 26; pp Takacs, I.; Patry, G.G.; Nolasco, D.A. Dynamic Model of Clarification-Thickening Process. Water Res. 99, 25, [CrossRef] 23. Standard Methods for Examination of Water and Wastewater. Available online: mwwp/pdf/sm254btotalsolids.pdf (accessed on 4 June 27). 24. Wahlberg, E.J. WERF/CRTC Protocols for Evaluating Secondary Clarifier Performance; Water Environmental Research Foundation: Alexandria, VA, USA, 2; pp Taeyoung Corporation. Performance Evaluation of Double Perforated Baffle in Secondary Clarifiers (2nd Train, 3rd Bay, No. 5, 6); Construction Supervision Report No. Suwon Wastewater ; Taeyoung Corporation: Suwon, Korea, 25; pp (In Korean) 27 by author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under terms and conditions of Creative Commons Attribution (CC BY) license (