Microfiltration for Treatment of Waste Filter Washwater at a North Jersey Surface Water Treatment Plant

Microfiltration for Treatment of Waste Filter Washwater at a North Jersey Surface Water Treatment Plant Michael Furrey Supervisory Chemist, The North ...
Author: Roland Cross
5 downloads 0 Views 448KB Size
Microfiltration for Treatment of Waste Filter Washwater at a North Jersey Surface Water Treatment Plant Michael Furrey Supervisory Chemist, The North Jersey District Water Supply Commission, Wanaque, New Jersey

James Schaefer, P.E. Vice President, Water Processing, Pall Corporation, East Hills, NY

Matthew Geho WTP Analyst, The North Jersey District Water Supply Commission, Wanaque, New Jersey

Thomas Gallo R&D Engineer, Pall Corporation, Port Washington, New York

1.0 BACKGROUND The North Jersey District Water Supply Commission (Commission) serves about 2 million consumers from its 210 mgd water treatment plant (WTP) in Wanaque, New Jersey, about 25 miles northwest of Manhattan. Raw water is obtained primarily from the adjacent Wanaque Reservoir, and the nearby Pompton and Ramapo Rivers. The Wanaque WTP provides conventional treatment with chlorine as primary disinfection. Waste filter washwater and supernatant from cleaning the settling basins is recycled to the head of the WTP after passing through a holding/equalization basin. Settling basin sludge flows to either the Residuals Treatment Facility or a sludge lagoon. The lagoon also accepts decant water from sludge dewatering. Lagoon supernatant is recycled to the reservoir. Occasionally, the recycle flows disrupt the operation the WTP and impact finished water quality. A site plan for the WTP is presented on Figure 1. Because of these impacts on finished water quality and the anticipated new regulations for handling waste filter washwater, the Commission has investigated alternatives to better handle these flows and improve finished water quality. Based on successful pilot testing at another facility, one alternative is to treat the waste filter washwater and decanted settling basin supernatant with microfiltration, and to use the microfiltered water as potable water or to recycle this water to the head of the WTP. The concentrated solids from the microfiltration unit would be discharged to the Residuals Treatment Facility or the lagoon. This paper presents the results of pilot testing microfiltration for this application and a preliminary engineering evaluation of alternative plans. 1.1 New Regulations USEPA’s Long Term 1 Enhanced Surface Water Treatment and Filter Backwash Rule proposed in April 2000 requires utilities to conduct a self-assessment concerning the impacts of recycling waste filter washwater on the main treatment process. The proposed regulations for recycling

(C) 2000, American Water Works Association, Water Quality Technology Conference Proceedings

Figure 1 Wanaque WTP Site Plan waste filter washwater and sludge supernatants seek to maintain high finished water quality by recycling these flows to the head of a conventional WTP prior to coagulant addition or by providing other suitable treatment, such as equalization or settling prior to recycling. These requirements are less stringent than the requirements anticipated a year ago. Recent changes to the Surface Water Treatment Rule requiring 2 log removal of Cryptosporidium, or higher depending upon the occurrence of Cryptosporidium, set a more stringent standard for any system treating waste filter washwater for direct use as potable water. The requirement for 2 log removal of Cryptosporidium and lower filtered water turbidity may make continued recycling with the existing holding basin impractical because of its impacts on the treatment process.

(C) 2000, American Water Works Association, Water Quality Technology Conference Proceedings

1.2 Existing Facilities The Wanaque WTP provides conventional treatment with coagulation, flocculation, settling, and dual media filtration and uses chlorine for oxidation at the rapid mixer and for primary disinfection. The WTP has six flocculation-sedimentation basins and 13 dual media filters (Four flocculation/sedimentation basins placed in service in 1981 as enclosed, two level basins without sludge collection equipment). By the end of 2000, these four basins will be equipped with sludge collectors. The two other basins were constructed in 1993 as open, upflow basins with tube settlers and sludge collectors. Alum is fed at the rapid mixer as the primary coagulant with a cationic polymer added at the flocculation basins as a coagulant aid. The WTP also uses potassium permanganate as a preoxidant and lime for final pH adjustment. Waste filter washwater is discharged to the Waste Washwater Holding Basin on the north side of the WTP, which equalizes these flows. The average flow of waste washwater is approximately 3.5 mgd, while the peak day flow recently has been 6 mgd. The equalized waste washwater is pumped to either the main process flow before the rapid mix chamber, the equalization tank at the Residuals Treatment Facility, or the sludge lagoon. During manual cleaning of the settling basins, basin supernatant is discharged via a separate drain to the same holding basin and recycled to the head of the WTP. Sludge from cleaning the settling basins is discharged to the sludge lagoon for removal and storage of the residuals. The clarified supernatant from the lagoon is recycled to the reservoir. Extensive monitoring of the entire process is required in order to maintain high finished water quality. Because of the relatively high waste flows and the large amount of solids already in the lagoon, the lagoon supernatant contains high suspended solids and turbidity, which adversely impacts the performance of the WTP when it is recycled. With the recent construction of the Residuals Treatment Facility, the Commission is working towards removing the sludge lagoon from the overall operation of the WTP. Recycling larger volumes from the decanting process to the head of the plant is a significant part of the effort. 2.0 MICROFILTRATION PILOT TESTING During May 1999, the Commission initiated a pilot test of microfiltration to demonstrate the feasibility of processing the waste filter washwater and settling basin decant to produce potable water, or at least water suitable for recycle in the WTP. Pilot testing was performed by the Commission and Pall Corporation with one of Pall’s standard pilot rigs, which are designed for automatic operation and continuous data collection of important parameters, such as turbidity, pressures, and flows. The Commission provided support to install the pilot unit, on-line particle counters to analyze feed and filtered water, and laboratory analysis of basic water quality parameters. Unusual analytical work, such as analysis of Giardia and Cryptosporidium, was performed by an outside laboratory. 2.1 Study Program Goals and Objectives The pilot program was designed to demonstrate reliable performance of the microfiltration system and excellent filtered water quality. Based on preliminary discussions with the New

(C) 2000, American Water Works Association, Water Quality Technology Conference Proceedings

Jersey Department of Environmental Protection (NJDEP), filtered water from microfiltration of waste filter washwater and settling basin decant may be used as potable water, if: 1) the water meets the appropriate water quality standards; and 2) the treatment system can provide 5 log removal of Giardia and 4 log removal of Cryptosporidium to assure adequate disinfection. The design of the treatment facilities should provide the level of protection required by the Surface Water Treatment Rule and should maximize the use of existing facilities. Because iron and manganese in settled solids can dissolve under anaerobic or low pH conditions, the microfiltration system must also be capable of removing these metals. The water quality goals for the microfiltered water, as presented in Table 1, meet or exceed NJDEP standards for drinking water. Table 1 Filtered Water Quality Goals Parameter Goal Turbidity 1000 >1000 >1000 >1000 Filtered Water 3.3 5.5 5.8 6.1 11

7 7 16 18 66

Avg.

pH (units) Min.

Max.

7.2 6.9 7.1 7.4 7.8

6.9 6.8 6.4 6.7 6.5

8.3 7.0 8.3 9.3 10.4

7.8 6.9 7.1 7.5 7.6

6.9 6.8 6.4 7.0 6.4

9.1 7.0 7.1 8.5 9.8

Feed turbidity varied according to the schedule for filter backwashes. During a backwash, the waste washwater flowing into the holding basin was very turbid and created high turbulence in the basin, which kept feed high. Actual peak turbidities were much higher than the recorded values because the continuous turbidimeter on the pilot rig was set to a maximum value of 20 NTU and was occasionally blocked by solids. When the turbidimeter inlet was clogged, low turbidities were recorded even when the TMP increased sharply, such as on 6/20 – 6/21. On-site sampling data, which were as high as 153 NTU, were a better indication of feed turbidity at these times. Periods of low feed turbidity allowed the microfilter to operate successfully at the 20-gfd loading rate and recover from the periodic slugs of high turbidity feed. Color - The average feed water color was also very high at 2870 units. Most of the color is related to the high suspended solids loads in the feed water. Removal of the solids reduced color to much lower levels.

(C) 2000, American Water Works Association, Water Quality Technology Conference Proceedings

Iron, Manganese and Aluminum - High concentrations of total and dissolved iron, manganese and aluminum were present in the feed water, as shown in Table 5. Precipitates of these metals, such as hydroxides and oxides, will be removed by microfiltration; however, the dissolved fraction will pass through the microfiltration membrane. Dissolved iron, on average, was low, while the average dissolved manganese was up to 0.38 mg/l. The average dissolved aluminum was approximately 1 mg/l. Subsequent oxidation of the dissolved manganese and precipitation of aluminum will create particles and colored water after the microfiltration process. Proper oxidation and pH control prior to microfiltration will be required to control these metals. Table 5 Feed Water Metals Data (Total [Dissolved]) Period

Iron (mg/l) Avg. Max.

5/24 - 5/31 0.10 [0.01] 0.8 [0.00] 6/1 - 6/3 0.20 [0.02] 6/9 - 6/24 6/28 - 7/07 0.21 [0.01] 7/14 - 7/29 0.6 [0.00]

0.23 [0.01]

Manganese (mg/l) Avg. Max. 0.20 [0.02]

0.30 [0.03]

1.1 [0.00]

4.6 [0.06]

0.6 [0.07]

1.5 [0.38]

Aluminum (mg/l) Avg. Max. 1.1 [0.39]

1.6 [1.0]

5.9 [0.08]

28 [0.03]

34 [0.03]

4.4 [2.7]

6.0 [0.18]

16 [1.0]

0.7 [0.04]

1.0 [0.03]

3.6 [0.10]

5.8 [1.5]

18 [0.7]

0.9 [0.00]

3.0 [0.01]

5.6 [0.02]

15 [5.1]

26 [1.1]

Data for particle counts, Giardia and Cryptosporidium, and disinfection by-products are presented at the end of the next section. 3.2 Microfiltration Results The performance of the microfiltration rig was very consistent during the two month study period. Typical performance data for headloss across the membrane (TMP), feed and filtered water turbidity, and flow during a nine-day period in June are presented in Figure 3. This period of operation was selected to include data during a basin cleaning (a period of high feed turbidity, 6/16-6/17), and to show the system’s ability to clean the membrane when the feed turbidity was low. Periods of operation can be identified by the filtered water or permeate flow rate, which was typcially 1 gpm. Turbidity - Filtered turbidity according to the on-line turbidimeter was 0.02 to 0.04 NTU. As the test progressed, some iron, manganese and aluminum solids were precipitated, as evidenced by the reddish brown discoloration of clear tubing and some floc accumulation in the filtered water tanks. On-site filtered water turbidity, as presented in Table 4, was significantly higher than the on-line data. This discrepancy between the two sets of data is more fully explained below, in the section which evaluates the iron, manganese and aluminum data. TMP rose and fell in a direct and immediate relationship with fluctuations in the feed turbidity. After a slug of high turbidity feed, the TMP gradually decreased to the pre-slug value as reverse filtration and air scrubbing removed the solids from the surface of the membrane. The robustness of the system was exhibited by its ability to handle periods of high turbidity loading and still recover to continue successful treatment, while maintaining less than 0.1 NTU filtered water turbidity.

(C) 2000, American Water Works Association, Water Quality Technology Conference Proceedings

Microfiltration Performance - June 16-24, 1999 Feed Turbidity, Filtered Water Flow, and TMP

30

Feed Turbidity (NTU) Filtered Water Flow (Gpm) TMP (psid)

25

20

15

10

5

0

6/16

6/17

6/18

6/19

6/20

6/21

6/22

6/23

6/24

6/25

Date/Time

Microfiltration Performance - June 16-24, 1999 Feed and Filtered Turbidity 30

Feed Turbidity (NTU) Filtered Turbidity (NTU)

25

20

15

10

5

0 6/16

6/17

6/18

6/19

6/20

6/21

6/22

6/23

6/24

Date/Time

Figure 3 Microfiltration Performance June 16-24, 1999

(C) 2000, American Water Works Association, Water Quality Technology Conference Proceedings

6/25

Color - Filtered color was generally less than 7 units. High color values were associated with high turbidities. If the filtered turbidity was low, the filtered color also was low. Iron, Manganese and Aluminum - Removal of metals depended upon whether the metals were precipitated or dissolved before reaching the microfilter. High filtered concentrations of iron, manganese, and aluminum represented the dissolved fraction, which passed through the membrane (see Table 6). The dissolved metals were present because of anaerobic or low pH conditions in solids settled on the bottom of the settling and holding basins. When these solids were disturbed, such as during a settling basin cleaning or when waste washwater was discharged into the holding basin, some of the dissolved metals were mixed into the feed water and passed through the membrane. Post precipitation of the metals was caused by aeration in the downstream piping and small amounts of chlorine that were added to the backwash water as part of the microfiltration system design. The post precipitation was visible over time in clear filtered water piping and also in the backwash supply tank. On-site samples had precipitates from the high dissolved metals concentration, resulting in high on-site filtered turbidities. Precipitated solids were able to reach on-line instruments (turbidimeter and particle counter) and sample collection ports. Table 6 Filtered Water Total Metals Data Period

5/24 - 5/31 6/1 - 6/3 6/9 – 6/24 6/28 - 7/07 7/14 - 7/29

Iron (mg/l) Avg. Max. 0.04 0.13 0.01 0.02 0.03 0.20 0.01 0.02 0.01 0.02

Manganese (mg/l) Avg. Max. 0.01 0.08 0.05 0.08 0.37 2.7 0.03 0.03 0.01 0.03

Aluminum (mg/l) Avg. Max. 0.5 1.8 0.0 0.0 0.1 0.5 0.4 1.9 2.5 10.3

This situation became worse as the filtered water with precipitates and floc was used to backwash the microfilter. These solids were pushed onto the downstream (clean water) side of the membrane during backwashing, only to be removed after the microfilter returned to filtering operation. Immediately after a backwash, the filtered water turbidity and particle counts spiked and then decayed with time. The solids in the filtered water increased particle counts and gave the impression that the membrane had failed. Oxidation before microfiltration is required to precipitate the dissolved metals and allowed them to be removed by the filter, like in the main WTP. Particle Counts - Particle counting was also used to measure microfiltration efficiency by collecting data on the waste washwater flow before the holding basin and on the filtered water discharge. Particle concentrations in the feed water exceeded the instrument’s upper detection limit and caused that sensor to mis-read the true values. Subsequently, the feed water was diluted to obtain accurate results. The feed water particle counts were on the order of 106 to 107/ml for particle sizes greater than 2 microns (µm). Filtered water particle counts ranged from 1 to 104 particles/ml greater than 2 µm. Particle removals were estimated to be from 1.8 to 5.8 logs.

(C) 2000, American Water Works Association, Water Quality Technology Conference Proceedings

After reviewing the other operating data, the high filtered water particle counts appeared to be related to the spikes in filtered water turbidity, iron, manganese, and aluminum. These results indicated the most likely cause was post precipitation of dissolved metals creating floc and particles. Additional pilot testing with pre-oxidation is required to confirm that the microfiltration system adequately removes properly oxidized metals. Giardia and Cryptosporidium - No Giardia or Cryptosporidium were detected in two sets of feed and filtered water samples, therefore, a removal rating was not calculated. The Pall microfiltration system has achieved 4 to 6 log removals (depending upon the feed concentration) in other third party challenge tests; similar removals were expected for this test. Disinfection By-Products - One set of feed and filtered water samples was analyzed for trihalomethane and haloacetic acid formation potential (THMFP and HAAFP). The test conditions were 7 days of reaction time at pH 8 and 20o C with a chlorine dose of 10 mg/L. The feed and filtered water chlorine demands were 9.3 mg/L and 4.8 mg/L. Microfiltration achieved moderate removals of 17% and 19%, as shown in Table 7. The DBP results are similar to the available TOC data, which indicated removal of TOC associated with particles to a reasonably consistent level of dissolved organic carbon (DOC). Specifically, TOC removals were 15 to 25% when the DBP samples were collected.

Date 5/26/99 6/2/99 6/8/99 6/16/99 6/23/99 7/14/99 7/21/99 7/28/99

Table 7 DBP and TOC Data THMFP (µ µg/L) HAAFP (µ µg/L) Feed Filtered Feed Filtered 151 123 226 187 -

TOC (mg/L) Feed Filtered 3.4 3.2 14.8 2.4 10.8 2.8 3.9 3.3 2.1 1.8 2.4 1.8 2.9 2.4 3.6 2.1

CIP Frequency - During this pilot program, one chemical cleaning was performed during July, and the TMP was restored to its original clean value. The CIP frequency was in the 4 to 6 week range. With better quality feed water, the cleaning frequency may increase. 4.0 PROPOSED WASTE WASHWATER TREATMENT SYSTEM 4.1 Design Considerations The design of a full-scale microfiltration system for treating waste filter washwater and settling basin decant generated by WTP operations must meet the following objectives: •

Handle the spikes of turbidity and suspended solids

(C) 2000, American Water Works Association, Water Quality Technology Conference Proceedings

• • •

Tolerate oxidant residuals for control of iron and manganese Provide excellent removals of pathogens, such as Giardia and Cryptosporidium Produce filtered water with low turbidity.

A microfiltration system removes contaminants that are in the form of particles. For example, iron and manganese must be oxidized and precipitated, in order to be successfully removed. Giardia and Cryptosporidium are virtually completely removed because their particle size is greater than the 0.1 µm pore size of the microfiltration membrane. Spikes of Turbidity and Suspended Solids - The existing holding basin that receives the flows of waste filter washwater and settling basin decant can be redesigned to better distribute flow and remove some of the suspended solids. Currently, heavier solids settle and accumulate in the bottom of the holding basin and must be periodically removed. The recommended modifications to the holding basin include new inlet piping to evenly distribute flow across the entire basin and sludge removal equipment to frequently remove accumulated solids and discharge the solids to the existing lagoon. The easiest-to-install sludge removal equipment for this basin will be the vacuum type on rails or tracks along the floor of the basin similar to the sludge collectors being installed in the 1981 settling basins. Iron and Manganese Control - Solids which contain iron and manganese precipitates accumulate in the bottom of the settling basins or the holding basin. The iron and manganese can dissolve under the anaerobic or low pH conditions that develop in the solids accumulation, especially when water temperatures are high. The dissolved iron and manganese will be recirculated to the head of the treatment plant and result in operational problems and degraded filtered water quality. The installation of sludge removal equipment in the holding basin will reduce the amount of dissolved iron and manganese, however, some dissolved iron and manganese will continue to be generated in the settling basins. An oxidant feed to the waste washwater and settling basin supernatant flows is recommended. Because manganese is likely to be present, the optimum oxidant will be potassium permanganate. The potassium permanganate should be fed into the inlet piping into the holding basin so that the oxidized iron and manganese can precipitate and settle in the holding basin. The required dosage will depend upon the quality of the flows entering the holding basins. For planning purposes, the feed system was sized to deliver about 5 mg/L of potassium permanganate. The actual dose will vary depending on the oxidant demand of the inflows, such as the amount of dissolved iron and manganese, and the effectiveness of frequent solids removal from the holding basin. Controlling dissolved iron and manganese will reduce filtered turbidity by eliminating post precipitation. Membrane Flux - The pilot system operated successfully at a flux of 20 gfd with the regular spikes of turbidity and suspended solids. TMP increased significantly when the spikes occurred, and steadily declined to the pre-spike value with regular reverse filtration and air scrubbing to remove accumulated solids. The chemical cleaning frequency was approximately 3 or 4 months. The long chemical cleaning interval indicated that a higher flux could be sustained if the solids loading could be controlled more effectively. Longer CIP intervals may be possible by adjusting reverse filtration and air scrubbing parameters to fit the feed water quality. Another microfilter pilot test successfully treated waste filter washwater at a flux of 30 gfd.

(C) 2000, American Water Works Association, Water Quality Technology Conference Proceedings

The recommended design flux for a microfiltration system depends upon the quality of the feed water to the microfiltration unit. For feed water, which still will have the spikes of turbidity and suspended solids, a flux of 20 gfd is recommended. For feed water, which will be better equalized and settled, a flux of 25 gfd is recommended. In either case, the chemical cleaning interval is expected to be approximately two months. The proposed waste washwater treatment system is designed to handle the 3.5 mgd average flow and the peak day flows of approximately 6 mgd, plus the occasional discharges of settling basin supernatant. Two treatment alternatives were considered: •

Alternative 1 – Microfiltration without Equalization – continue to use the holding basin to receive waste washwater and basin supernatant flows and microfilter the combined flow.



Alternative 2 – Microfiltration with Equalization – modify the holding basin to remove some of the solids and equalize the waste washwater and supernatant flows, and microfilter the combined flow at a higher loading rate.

In either case, the microfiltered water will be used as potable water and microfiltration waste flows will be piped to the existing sludge lagoon. Each alternative is described below. 4.2 Alternative 1 – Microfiltration without Equalization Alternative 1 includes the installation of a new microfiltration system to treat the discharge from the holding basin and produce potable water. The existing holding basin will continue to receive and hold the waste filter washwater and settling basin decant flows prior to microfiltration. No modifications to the inlet piping or holding basin are required for this alternative. The basis of design for Alternative 1 is summarized in Table 8 and shown schematically on Figure 4. Table 8 Basis of Design for Treatment Alternatives Flow (mgd) MF Design Criteria Discharges to Lagoon AlterFlow to Flow Feed Turb. Flux Flow native Holding Basin Avg. Peak (mgd) (NTU) (gfd) Type (mgd) None from Waste WashHolding Existing 3.5 6 water Basin Waste WashMF Waste1 3.5 6 6 30 20 0.2-0.4 water water MF WasteWaste Washwater 0.2-0.4 2 3.5 6 6 15 25 water Sludge from Holding Tank 0.1-0.2 The design flow for the microfiltration system will be 6 mgd to accommodate the peak waste washwater flow. The design flow is based on recent improvements in filter operations and

(C) 2000, American Water Works Association, Water Quality Technology Conference Proceedings

Figure 4 Process Schematics for Microfiltration Alternatives backwashing procedures, and the installation of sludge collection equipment in all six settling basins, which will virtually eliminate the flows of settling basin supernatant. The microfiltration system will recover 94% of the water fed into the system; the remaining 6% of the flow, typically 0.2 to 0.4 mgd, will carry the concentrated solids into the existing sludge lagoon. The holding basin will continue to operate as it has, with some solids settling and accumulating on the bottom of the basin. Some iron and manganese will dissolve from solids accumulated on the bottom of the holding basin, so preoxidation with potassium permanganate will be needed to maintain high filtered water quality. The required modifications for Alternative 1 include: • • • •

6 mgd microfiltration system for waste washwater and occasional flows of settling basin supernatant Potassium permanganate feed facilities for preoxidation New higher head variable speed pumps in the Waste Washwater Pump Station Piping connections for microfiltered water to the finished water clearwell and microfiltration wastes to the existing sludge lagoon

(C) 2000, American Water Works Association, Water Quality Technology Conference Proceedings

The microfiltration system will consist of six racks with spaces for 100 microfiltration modules in each rack (see Figure 5). Ninety-four modules will be installed in each rack to provide 6 mgd of capacity at a loading of 20 gfd. The microfiltration system also will include the chemical cleaning equipment (tanks, pumps, chemical feed and storage), and instrumentation and controls adequate to control all parts of the system.

Figure 5 Plan of Proposed Microfiltration Facility The potassium permanganate feed system will use dry potassium permanganate and will consist of a mixing and feed tank, and metering pumps. The microfiltration system and potassium permanganate feed systems will be housed in a one story building of approximately 4,900 ft2 using finish materials to be compatible with the other buildings at the WTP. The new building will located on the southeast side of the existing Waste Washwater Holding Basin. The microfiltration system will increase the head requirements for the recycle pumps by about 30 psi, so new pumps are required. The new pumps will be variable speed units for energy efficiency and to be compatible with the microfiltration system

(C) 2000, American Water Works Association, Water Quality Technology Conference Proceedings

The filtered water will flow via 24-inch diameter piping to the north end of the existing filtered water conduit between the Administration Building and the Settling Basins. A 6-inch pipe will convey microfiltration wastes to the piping to the sludge lagoon at the southwest corner of the WTP. 4.3 Alternative 2 – Microfiltration with Equalization Alternative 2 is very similar to Alternative 1, except the holding basin will be modified to better equalize the inflows of waste washwater and settling basin supernatant and remove some of the suspended solids. Because of the improved feed water quality, the microfiltration system will be designed with a higher loading rate, which reduces the size and cost of the equipment. The basis of design for Alternative 2 is also summarized in Table 8. The modifications to the holding basin will include new inlet piping to be installed as shown in Figure 6. The new 42-inch inlet piping will carry both the waste washwater and the basin supernatant and distribute the water evenly across the east side of the basin through four 16-inch diffuser ports. The new inlet pipe will be installed about 5 feet from the east wall and the four 16-inch ports will direct water toward the east wall, which will act to dissipate the velocity and spread the flow across the basin. With the better hydraulics, more solids will settle in the holding basin, so a vacuum style sludge removal mechanism will be installed on the bottom of the basin to remove the solids as they accumulate. The removed solids will be pumped into the sludge lagoon via an 6-inch pipe.

Figure 6 Proposed Holding Basin Modifications

(C) 2000, American Water Works Association, Water Quality Technology Conference Proceedings

Because of the higher loading rate, the microfiltration system will consist of five racks with spaces for 100 microfiltration modules in each rack (see Figure 5). Eight-nine modules will be installed in each rack to provide 6 mgd of capacity at a loading of 25 gfd. The microfiltration system also will include the chemical cleaning equipment (tanks, pumps, chemical feed and storage), and instrumentation and controls adequate to control all parts of the installation. The other components for this alternative will be the same as for Alternative 1, except that the building to house the microfiltration system and potassium permanganate feed systems will be a one story building of approximately 4,400 ft2. 4.4 Estimated Costs The estimated capital costs for the alternatives are summarized in Table 9. Table 9 Estimated Capital Costs for Microfiltration Treatment Alternative 1 – Alternative 2 – Item without Equalization with Equalization Microfiltration System $3,800,000 $3,100,000 Building for Microfiltration System 490,000 440,000 New Recycle Pumps 80,000 80,000 Pre-Oxidation 50,000 50,000 Sludge Collectors 200,000 Piping and Valves 300,000 320,000 Electrical and Instrumentation 450,000 400,000 Contingencies (10%) 530,000 460,000 $5,700,000 $5,050,000 Total Estimated Cost The estimated capital cost for Alternative 1 – Microfiltration without Equalization is $5,700,000 for all of the proposed improvements including 10% for contingencies. The microfiltration equipment consists of the microfilter modules and racks, piping and valves, controls, chemical clean equipment, and related chemical storage and feed equipment. The estimated capital cost for Alternative 2 – Microfiltration with Equalization is $5,050,000 or 12% less than Alternative 1. The principal differences are the smaller microfiltration system and building to house the system, and the addition of sludge collection equipment. Estimated operating costs for the microfiltration system are summarized in Table 10. The estimated costs of power and chemicals are the same for either alternative because the system will treat the same average flow. The estimated membrane replacement cost is significantly lower for Alternative 2 because of the smaller number of membranes in the system. The estimated useful life for the membrane modules is 8 years. The estimated unit operating costs using an average flow of 3.5 mgd range between $0.26 and $0.31 per 1000 gallons of filtered water.

(C) 2000, American Water Works Association, Water Quality Technology Conference Proceedings

Table 10 Estimated Annual Operating Costs for Microfiltration System Alternative 1 – Alternative 2 – Item without Equalization with Equalization Power $60,000 $60,000 Chemicals 20,000 20,000 Membrane Replacement (8 yr life) 320,000 250,000 Total Annual Costs for Power, Chemicals, and Membranes $400,000 $330,000 Operating Cost for Filtered Water $0.31/1000 gal $0.26/1000 gal Operating labor will be 2 to 4 hours per day because the system will operate automatically and only simple chemical feeds are required for microfilter operation 5.0 CONCLUSIONS The waste washwater and settling basin supernatant were successfully treated with microfiltration to produce potable quality water. The microfiltration system with its 0.1 µm membrane was a barrier to the recycle of Giardia and Cryptosporidium, so adequate disinfection was provided by the microfilter. This system will be able to meet anticipated future regulations concerning the handling of waste washwater as well as current and future standards for drinking water. The estimated capital and operating costs for Alternative 2 – Microfiltration with Equalization are 12 to 15% lower than Alternative 1 - Microfiltration without Equalization. In addition, system operation will be more reliable if equalization is provided. The performance of the microfiltration system with preoxidation with respect to the removal of particles and turbidity should be confirmed with additional pilot testing. Pre-oxidation is required to precipitate dissolved metals so that the metals can be removed by the microfilter. Longer term pilot testing of the microfiltration system on equalized waste filter washwater and settling basin decant should be performed to test the feasibility of a 30 gfd loading (or flux) rate that was successfully pilot tested at another location. The least cost approach, Alternative 2, is based on a loading of 25 gfd. If the loading were increased to 30 gfd, the estimated capital costs would be reduced by $500,000 to $750,000 because of the smaller microfiltration system and building to house it. Estimated operating costs also would be reduced because membrane replacement costs would be reduced with fewer membrane modules in the system. The estimated operating costs at 30 gfd would be approximately $0.22/1000 gallons of water filtered. These savings are significant enough to warrant additional pilot testing to optimize the system design.

(C) 2000, American Water Works Association, Water Quality Technology Conference Proceedings

Suggest Documents