May 2003
                      NSF 03/05/EPADWCTR
Environmental Technology
Verification Report

Physical Removal of Particulate
Contaminants in Drinking Water
Polymem
Polymem UF120 S2 Ultrafiltration
Membrane Module
Luxemburg, Wisconsin
            Prepared by
          NSF International
     Under a Cooperative Agreement with
      U.S. Environmental Protection Agency

-------
         THE ENVIRONMENTAL TECHNOLOGY VERIFICATION
                                     PROGRAM ^
                                                   //
       ©ERA
E1V
    U.S. Environmental Protection Agency
                                                                    NSF International
                    ETV Joint Verification Statement
    TECHNOLOGY TYPE:   MEMBRANE FILTRATION USED IN DRINKING WATER
                           TREATMENT SYSTEMS
    APPLICATION:         PHYSICAL REMOVAL OF PARTICULATE
                           CONTAMINANTS IN DRINKING WATER
    TECHNOLOGY NAME:  POLYMEM UF120 S2 ULTRAFILTRATION MEMBRANE
                           MODULE
    COMPANY:            POLYMEM
    ADDRESS:             ROUTE DE REVEL F-31450     PHONE: 011.33.5.53.71.79.89
                           FOURQUEVAUX, FRANCE     FAX: 011.33.5.62.71.79.80
    EMAIL:                polymem@wanadoo.fr
The U.S. Environmental Protection Agency (EPA) supports the Environmental Technology Verification
(ETV)  Program to facilitate  the deployment of innovative  or  improved  environmental technologies
through performance verification and dissemination of information. The goal of the ETV program is to
further environmental protection by substantially accelerating the acceptance  and use of improved and
more cost-effective technologies. ETV seeks to achieve this goal by providing high quality, peer reviewed
data on technology performance to those involved in the design, distribution,  permitting, purchase, and
use of environmental technologies.

ETV works in partnership with recognized standards and testing organizations,  stakeholder groups
(consisting of buyers, vendor organizations, and permitters), and with the full participation of individual
technology developers. The program evaluates the performance of innovative technologies by developing
test plans that  are responsive to  the needs of stakeholders,  conducting field or laboratory tests  (as
appropriate),  collecting and analyzing data, and preparing peer reviewed  reports. All evaluations  are
conducted in accordance with rigorous quality assurance protocols to ensure that data of known and
adequate quality are generated and that the results are defensible.

NSF International (NSF) in cooperation with the EPA operates the Drinking Water Systems (DWS)
Center, one of seven ETV Centers under ETV. The DWS Center recently evaluated the performance of an
ultrafiltration membrane used in drinking water treatment system applications. This verification statement
provides a summary of the test results for the Polymem UF120 S2 Ultrafiltration Membrane Module.
Carollo Engineers, P.C., an NSF-qualified field testing organization (FTO), performed the verification
testing. NSF provided technical and quality assurance oversight  of the verification testing described in
this ETV report.
03/05/EPADWCTR       The accompanying notice is an integral part of this verification statement.           May 2003
                                          VS-i

-------
ABSTRACT

Verification testing of the Polymem UF120 S2 Ultrafiltration Membrane Module was conducted over a
46-day period at the Green Bay Water Utility Filtration Plant, Luxemburg, Wisconsin. The ETV testing
described herein was funded in conjunction with a 12-month membrane pilot study funded by the Energy
Center of Wisconsin. The Energy Center of Wisconsin chose to participate because the overall scope of
the ETV testing fit into the scope of the longer, energy focused study. The testing was performed from
March 11, 2002 through April 26, 2002, representing winter/spring conditions when, historically, feed
water quality was most difficult to treat. The feed water was Lake Michigan. Verification testing was
conducted at optimized conditions based on pilot testing conducted during the 12 months proceeding the
verification test period.  The testing was performed using a "generic"  custom membrane pilot  plant
(CMPP) capable of operating  with a variety of membrane modules that are housed in pressure vessels.
Therefore, this ETV testing verified the operation of the membrane module itself, not membrane-specific
process equipment. The membrane unit was operated in dead-end mode during two test runs, each at a
constant specific flux of 40 and 30 1/h-m2 (24 and 18 gfd), respectively. Feed water recoveries ranged
from 89-96 percent. The two test runs were operated for approximately 12.5 and 32.7 days, respectively.
The UF  module  was chemically cleaned using  a  "proof of concept" effort based  on procedures
recommended  by the  manufacturer. The cleaning procedures were effective  in restoring membrane
productivity.  The membrane  module  achieved  significant removal of particulate  contaminants and
bacteria, producing an average filtrate turbidity of 0.05 NTU and an average of 4.2 log removal of total
particles (>2  im in size). Average feed turbidity and total particle counts were 1.3 NTU  and 4,281
particles/ml, respectively.

TECHNOLOGY DESCRIPTION

The Polymem UF120  S2 Ultrafiltration Module is comprised of 19 individual polysulfone hollow-fiber
membrane bundles housed in a PVC pressure vessel. The bundles are potted on the effluent side of the
module, forming a U-shaped configuration and provide a total of 114 m2 (1227 ft2) of active membrane
surface area. The membrane, classified as an Ultrafiltration membrane, has a nominal pore size of 0.01 (im
as specified by Polymem and  was not verified in this verification test. This pore  size should provide a
physical barrier to particulate matter, bacteria, protozoans,  and viruses when membrane fibers are intact
and operated within the recommended operating ranges.

The membrane module is designed for operation in a dead-end mode, reducing power consumption over
traditional  cross  flow membrane  products,  as recirculation pumps  are  not required.  The  flow
configuration is outside to inside. This forces the accumulation of particulate matter, pathogens, and
suspended solids on the outside of the membrane fiber. The recommended backwash procedure includes
simultaneous  hydraulic backwash,  air scour,  and chlorine injection. Backwash is accomplished  by
pumping filtrate water from the inside to the outside of the fiber. This water is then discharged to waste.
An inlet for air scour is provided at the level of the potting resin via air diffusers  located inside the
module. This design makes minimum chemical cleaning intervals of 30 days possible  without exceeding
the maximum  allowable  transmembrane pressure  (terminal transmembrane pressure) of 2 bar (29 psi).
The membrane system and operating  strategy  (flux,  recovery, and  backwash intervals)  are typically
designed for a  30-day chemical cleaning interval. However, significant changes in  water quality will
effect membrane performance. Temperature fluctuation, increases in natural organic matter, turbidity, and
pH changes may have the potential to increase membrane fouling rates.

Some fraction of the particulate matter and dissolved constituents in the feed water can accumulate on the
membrane surface  and cannot be removed by hydraulic backwash and air scour. This leads to rise in
transmembrane pressure during normal operation. Once the terminal transmembrane  pressure has been
reached (29 psi), the membrane must be taken off-line to remove this matter from the membrane with a
03/05/EPADWCTR       The accompanying notice is an integral part of this verification statement.           May 2003
                                             VS-ii

-------
chemical clean. The membrane polymer is designed to be tolerant to a variety of chemicals, including
chlorine, acids, bases, and chelating agents commonly used for chemical cleaning.
Critical to this testing was the use of a "generic" CMPP. The CMPP was not provided by Polymem. The
CMPP used has the capacity to feed, backwash, and clean a variety of pressure vessel-type MF/UF
modules. Therefore, this testing verified the operation of the membrane module under a given set of
operational parameters, not membrane-specific process  equipment.

VERIFICATION TESTING DESCRIPTION

Test Site

The testing site was  the Green Bay Water Utility Filtration Plant located  at 6183  Finger Road in
Luxemberg, Wisconsin. The Green Bay Water Utility Filtration Plant is fed by one or both of two raw
water intakes located on the western  shore of Lake Michigan in Kewaunee, Wisconsin. The raw water is
pumped to the filtration plant in Luxemberg, Wisconsin. A small  amount of chlorine  (<0.30 mg/L) is
added at each intake to prevent growth of zebra mussels during transmission from intake to the treatment
facility. The CMPP used for this testing was located approximately 200 feet from the raw water channel
at the filtration plant. A submersible  pump located 3 feet below the free water surface fed the CMPP via
2-inch schedule 80 PVC pipe, and 1.5-inch PVC tubing.

Methods and Procedures

Onsite bench-top analyses including turbidity, pH, chlorine, and temperature were conducted daily at the
test site according to Standard Methods for the Examination of Water and  Waste-water, 20th Edition
(APF£A,  1998)  and by Methods for  Chemical Analysis of Water and  Wastes  (EPA, 1979),  where
applicable.  Standard Methods for the Examination of Water and Wastewater, 20th Edition (APF£A, 1998)
was followed for total coliform  analyses conducted at  Northern  Lake Service, Inc. (NLS), Crandon,
Wisconsin  and  MWH Laboratories, Pasadena, California. Other analyses  conducted by NLS  were
conducted using Standard Methods for the Examination of Water and Wastewater,  18th Edition (APF£A,
1992)  and  by Methods for Chemical Analysis of Water  and Wastes (EPA, Revision 1983),  where
applicable.  Laboratory analyses included alkalinity, total and calcium hardness, total  dissolved solids
(TDS),  total suspended  solids  (TSS),  total organic carbon  (TOC), ultraviolet  absorbance  at  254
nanometers (UVA), total coliform and heterotrophic plate count (HPC). Alkalinity and total and calcium
hardness analyses were conducted once per month. TDS analyses were conducted every other week. TOC
and UVA  analyses  were conducted twice per  week. TSS, total  coliforms,  and  HPC analyses were
conducted five days per week. Online particle  counters and turbidimeters continuously monitored both the
feed and membrane filtrate waters. The particle counters were set up to enumerate particle counts in the
following  size ranges:  total (>2 (im), 2-3  (im, 3-5 (im,  5-15 (im, and >15 (im. Data from the  online
particle counters were stored  at  5-minute intervals  on a dedicated computer.  Online turbidity
measurements were recorded at  10-minute  intervals. Challenge testing, microbial or otherwise, was not
performed  as part of this  study;  particle removal was quantified based on turbidity and particle counter
data.

VERIFICATION OF PERFORMANCE

System Operation

Verification testing conditions were established based on pilot study optimization results conducted from
May 2001 to March 2002. The membrane unit was operated at a constant specific flux of 40 L/h-m2 (24
gfd) for the first 12.5 days of operation (Run 1) and 30  L/h-m2 (18 gfd) during the remaining 32.7 days of
operation (Run 2). Production backwashes were performed at 50-minute intervals using an average
volume of 39 and 30 gallons for Runs 1 and 2, respectively. System recoveries ranged from 89-96 percent
03/05/EPADWCTR       The accompanying notice is an integral part of this verification statement.            May 2003
                                             VS-iii

-------
throughout the testing. The backwash chlorine concentration was set at 5 mg/L for the duration of the
testing.

Test Runs 1 and 2 yielded normalized specific flux decline rates of 7.2 L/h-m2/bar/day (0.29 gfd/psi/d)
and 1.7 L/h-m2/bar/day (0.069 gfd/psi/d), respectively. The improvement in fouling control during Run 2
is likely due to the lower target normalized flux. It should be noted that the 25 percent decrease in specific
flux led to a 260 percent increase in run time before a required chemical cleaning (12.5 vs. 32.7 days).

A  total of three membrane  cleanings were performed  based on the manufacturer's recommended
procedure. A high pH (11-12) chlorine solution (200 mg/L) was injected into the membrane module and
was allowed to soak for at least 4 hours. Flux data was collected after each chemical cleaning to evaluate
specific flux recovery.  The first cleaning was  performed prior  to membrane operation. Therefore,
recovery information was not available for this cleaning.  The recovery of specific normalized flux for
Chemical  Cleaning #'s 2 and 3  was  62 and 73  percent, respectively.  Cleaning #2 was performed at
ambient  water temperature,  [14-18.6°C  (57-65.5°F)], pH  = 12.2,   and  an  average total chlorine
concentration of 164 mg/L, for 8 hours. Because recovery of specific flux following Cleaning # 2 was
low, Cleaning  #  3  was performed with a similar  cleaning solution but  at elevated solution temperature
[22-31°C (72-88°F)], for an extended soaking period. Despite these changes, the specific flux recovery
was marginal (73 percent). This may be explained in part by the lack of chemical recirculation. This is
because the CMPP was not equipped with heating and recirculation equipment typically used to perform
clean-in-place (CIP) procedures on this membrane.

Membrane integrity monitoring was conducted prior-to and after this testing. Air pressure-hold tests were
conducted by opening the feed side of the  membrane to the atmosphere and  applying approximately  10
psi to the filtrate side of the membrane. Once pressurized, the loss of filtrate  side pressure was recorded
over a two-minute period. The first membrane integrity test yielded a zero pressure  loss with time. The
test at the end of system operation yielded a pressure  loss of 0.35  psi/min, which was within the
manufacturers  recommended feed side pressure loss (<0.36 psi/min). However, during this test, visual
observations showed a steady stream of air bubbles released to the feed side of the membrane.  This
suggested that a membrane fiber (or fibers) and membrane integrity may have  been compromised.
Following ETV testing, the membrane module filtrate end cap was removed to further investigate the
bubbles noted during the  final integrity  test.  This  investigation followed  the  integrity test/repair
procedures outlined in the Polymem UF120 S2 Operations and Maintenance (O&M) Manual. One broken
fiber was  identified and  repaired. One subsequent pressure decay test, performed  as  described above,
yielded a zero loss in pressure and no visual indicators of a loss of membrane integrity  (no bubbles were
detected).

Water Quality Results

The equipment verification testing described in this report  was executed using raw Lake Michigan water
obtained from the Green  Bay Water Utility Filtration Plant. Water used for CMPP operation was drawn
from the process prior to any treatment (other than C12 addition for zebra muscle control) at the water
facility and was  pumped approximately 200 feet to the skid mounted CMPP located inside a module
trailer unit. Table VS-1 below presents the results of the general water quality characterization for both
feed and filtrate waters throughout the ETV verification test.  The  feed  water had the following average
water quality during  this evaluation: C12  residual 0.05 mg/L, alkalinity 110 mg/L  as  CaCO3,  total
hardness 130 mg/L as CaCO3, calcium hardness 88 mg/L as CaCO3,  TSS  1.3 mg/L, TDS 187 mg/L,
TOC 2.3 mg/L, UVA 0.024 cm'1, algae 34 #/ml, temperature 3°C (37°F), and pH 7.8. As expected, there
was no notable change in alkalinity, total hardness, calcium hardness, or total dissolved solids across the
membrane module. However, there was a small reduction in TOC in the filtrate.
03/05/EPADWCTR       The accompanying notice is an integral part of this verification statement.           May 2003
                                             VS-iv

-------
Total suspended  solids were  measured throughout  the testing  as an indication of particle removal
potential. Filtrate  TSS was typically below the detection limit with 32 out of 37 samples reported at or
below the level of detection. Like HPC data, some of the  filtrate samples were detected at higher than
expected levels. These results  are likely due to the fact that feed and filtrate samples were so near the
detection limit of the analysis. Due to the length of time  the equipment was in use prior to the ETV
testing, it is also possible that material had built up in the portion of sample piping permanently fixed to
the CMPP skid. Although the sample ports were allowed to flush prior to sample collection, accumulated
material may have sloughed off during some of the sampling periods.

As presented in Table VS-1, average feed and filtrate bench top turbidities were 1.3  and 0.05  NTU,
respectively. Continuously monitored  filtrate turbidity was 0.035 NTU or less 90 percent of the time.
Average feed and filtrate total particle counts were 4,281  and 4 particles/ml, respectively. Table VS-2
summarizes the particulate log removal data.  Average particle log removals of 4.2, 4.1, 4.1, 3.4, 3.3, 2.9,
and  2.2 were  achieved for particle size  ranges of >2 um, 2-3, 3-5, 5-7, 7-10,  10-15, and >15 um,
respectively. The  90th percentile for feed and filtrate total  particle counts (>2 #/ml)  was approximately
9,911 and 2 particles/ml, respectively. The  membrane system removed  3.1 logs of total particles 90
percent of the time. A few of the filtrate particle count data were  recorded by the data logger as 0.00
particle/ml  (below the detection limit of the instrument). Since these data were recorded as zero values,
log removal data  could not be calculated  for these data points and were not included in the statistical
analyses. Because the membrane  system produced relatively  consistent filtrate particle counts, log
removals increased  during periods when  feed water particle counts were higher and decreased during
periods when feed water particle counts were lower. Relatively higher particle counts were measured in
the filtrate immediately following a backwash due in part  to hydraulic  and air bubble turbulence. As a
result, particle removals were decreased during these events.

A sensitivity analysis was performed  on  the data collected from one 24-hour period to determine the
potential effects of backwash events on calculated log removals. Data from March 14, 2002 were chosen
for this analysis due to the clusters of relatively lower log removal data during that time period, thereby
representing a worse case scenario. Log removals calculated for the raw data set (data including backwash
events) were 3.2  logs or greater,  90 percent of the time. Log removals calculated for the data set
excluding data obviously collected during backwash events, increased to 3.6 logs or greater, 90 percent of
the time.

Table VS-3 summarizes total coliform and HPC data. Total coliform enumeration results showed feed
concentration ranging from < 1.1-23 MPN/100 ml. Filtrate results for total coliform enumeration were
reported below the detection limit of <1 MPN/lOOml. HPC were  significantly reduced. Feed water HPC
ranged up to 330 CFU/ml. 33 of 38 filtrate HPC samples were at or below the method detection limit of 2
CFU/ml.
03/05/EPADWCTR        The accompanying notice is an integral part of this verification statement.           May 2003
                                              VS-v

-------

Table VS-1 General Water Quality for Both Feed
Parameter Units
C12 -Residual0 } mg/L
Alkalinity mg/L as CaCO3
Total Hardness mg/L as CaCO3
Calcium Hardness mg/L as CaCO3
TSS(2) mg/L
TDS mg/L
TOC mg/L
UVA cm"1
Algae #/ml
pH Units
Temperature °C (°F)
Bench Top Turbidity NTU
Particles >2 um #/ml
and Filtrate Waters
Feed Water Filtrate
0.05
110 110
130 130
88 87
1.3 1.2
187 203
2.3 2.0
0.024 0.019
34.4
7.80
3.4 (38)
1.3 0.05
4281 4
(1) Measured as part of the daily sampling activities of the Green Bay Water Utility Filtration Plant (GBWUFP).
(2) Limit of detection = 1 mg/L
Table VS-2 Particulate Log Removal
Particle Size Average Feed Count, #/ml Average
>2um 4,281
2-3 um 1,602
3-5 um 1,880
5-7 um 325
7-10um 305
10-15 um 127
>15um 41

Filtrate Count #/ml Average Log Removal
4 4.2
1 4.1
1 4.1
0 3.4
0 3.3
1 2.9
2 2.2

Table VS-3 Average Microbial Water Quality
Parameter Units Feed Water
Total Coliforms (1) MPN/100 ml 6.2
HPC (2) CFU/ml 17

Filtrate Backwash Water
<1.1 <1.1
2 24
( 1 ) Limit of detection =1.1 MPN/1 00 ml
(2) Limit of detection = 2 CFU/ml
03/05/EPADWCTR         The accompanying notice is an integral part of this verification statement.             May 2003
                                                     VS-vi

-------
Operation an d Mainten an ce Results

Operating conditions were established in a Programmable Logic Controller (PLC) prior to beginning the
test. These conditions included flux rate, production  dwell time, backwash procedures (interval and
duration), alarm condition settings, chemical feed doses, and data logging intervals. A notable exception
to the logged parameters is  air scour flow rate.   With the exception of backwash duration, these
parameters were not adjusted during operation. Backwash duration was adjusted as needed to maintain a
recovery of at least 90 percent and ranged from 60-120 seconds. Backwash chlorine was set to a dose of 5
mg/L and was checked daily through onsite analyses.

Operation of the  membrane consumed approximately 0.05 and 0.03 Ibs/day of sodium hypochlorite
during test  Runs  1  and  2,  respectively. Chemical cleanings  each consumed  0.06  Ibs  of sodium
hypochlorite and approximately 1.5-2 Ibs of sodium hydroxide.
     Original Signed by Clyde R. Dempsey                  Original Signed by
    for Hugh W. McKinnon	06/10/03          Gordon Bellen	06/13/03
    Hugh W. McKinnon                    Date          Gordon Bellen               Date
    Director                                             Vice President
    National Risk Management Research Laboratory         Research
    Office of Research and Development                   NSF International
    United States Environmental Protection Agency
    NOTICE: Verifications are  based on an evaluation of technology performance  under specific,
    predetermined criteria and the appropriate quality assurance procedures. EPA and NSF make no
    expressed or implied warranties as to the performance of the technology and do not certify that a
    technology will always operate as verified. The end user is solely responsible for complying with
    any and all  applicable federal, state, and local requirements. Mention of corporate names, trade
    names, or commercial products does not constitute endorsement or recommendation for use of
    specific products. This report is not a NSF Certification of the specific product mentioned herein.
03/05/EPADWCTR       The accompanying notice is an integral part of this verification statement.           May 2003
                                             VS-vii

-------
    Availability of Supporting Documents
    Copies  of the ETV Protocol for Equipment Verification  Testing for Physical Removal of
    Microbiological  and  Paniculate  Contaminants,  dated  May  14,  1999, the  Verification
    Statement, and the Verification Report (NSF Report # NSF 02/05/EPADWCTR) are available
    from the following sources:
    (NOTE: Appendices are not included in the Verification Report. Appendices are available
    from NSF upon request.)

    1.  ETV Drinking Water Systems Center Manager (order hard copy)
        NSF International
        P.O. Box 130140
        Ann Arbor, Michigan 48113-0140

    2.  NSF    web     site:     http://www.nsf.org/etv/dws/dws_reports.html     and    from
        http://www.nsf.org/etv/dws/dwsjroject documents.html (electronic copy)

    3.  EPA web site:  http://www.epa.gov/etv (electronic copy)
03/05/EPADWCTR        The accompanying notice is an integral part of this verification statement.           May 2003
                                             VS-viii

-------
                                                               May 2003


          Environmental Technology Verification Report

Physical Removal of Particulate Contaminants in Drinking Water


      Polymem UF120 S2 Ultrafiltration Membrane Module

                      Luxemburg, Wisconsin
                              Prepared for:

                            NSF International
                          Ann Arbor, MI 48105
                              Prepared by:

                          Carollo Engineers, P.C
                            Boise, ID 83713
     Under a cooperative agreement with the U.S. Environmental Protection Agency

                      Jeffrey Q. Adams, Project Officer
                National Risk Management Research Laboratory
                    U.S. Environmental Protection Agency
                          Cincinnati, Ohio 45268
                     Office of Research and Development
                    U.S. Environmental Protection Agency
                            Washington, DC

-------
                                       Notice

The  U.S.  Environmental Protection  Agency  (EPA)  through  its Office of Research  and
Development has financially supported and collaborated with NSF International  (NSF) under
Cooperative Agreement No. R-82833301. This verification effort was supported by the Drinking
Water  Systems (DWS) Center operating under the Environmental Technology Verification
(ETV)  Program. This document has been peer reviewed and reviewed by NSF and EPA and
recommended for public release.

-------
                                       Foreword

The U.S. Environmental Protection Agency (EPA) is charged by Congress with protecting the
Nation's land, air, and water resources. Under a mandate  of national environmental laws, the
Agency strives to formulate and implement actions leading to a compatible balance between
human activities  and the ability of natural systems to support and nurture life. To meet this
mandate,  EPA's  research  program  is providing  data  and  technical support  for  solving
environmental problems today and building a science knowledge base necessary to manage our
ecological resources wisely, understand how pollutants affect our health, and prevent or reduce
environmental risks in the future.

The National  Risk Management Research Laboratory (NRMRL) is  the Agency's center for
investigation of technological  and  management approaches for preventing and reducing risks
from pollution that threaten human health  and the environment. The focus  of the Laboratory's
research program is on methods and  their cost-effectiveness for prevention  and control  of
pollution to air, land, water, and subsurface resources; protection of water quality in public water
systems; remediation of contaminated sites, sediments and ground water; prevention and control
of indoor air pollution; and restoration of  ecosystems.  NRMRL collaborates with both public
and private  sector partners to foster technologies  that reduce the cost of compliance and to
anticipate emerging problems. NRMRL's research provides solutions to environmental problems
by: developing and promoting technologies that protect and  improve the environment; advancing
scientific and engineering information to support regulatory and policy decisions; and providing
the technical  support  and  information transfer  to ensure implementation of environmental
regulations and strategies at the national, state, and community levels.

This publication has been produced as part  of the Laboratory's strategic long-term research plan.
It is published and made available by EPA's Office of Research and Development to assist the
user community and to link researchers with their clients.
                                                            Hugh W. McKinnon, Director
                                           National Risk Management Research Laboratory
                                           in

-------
                                   Table of Contents

Section                                                                           Page

Verification Statement	VS-i
Title Page	i
Notice	ii
Foreword	iii
Table of Contents	iv
Abbreviations and Acronyms	viii
Acknowledgements	x

Chapter 1 - Introduction	1

    1.1  ETV Purpose and Program Operation	1
    1.2 Testing Participants	1
    1.3  Definition of Roles and Responsibilities of Project Participants	2
        1.3.1 NSF Responsibilities	2
        1.3.2 Field Testing Organization Responsibilities	2
        1.3.3 Manufacturer Responsibilities	3
        1.3.4 Operator and Test Site Staff Responsibilities	3
        1.3.5 Water Quality Analyst Responsibilities	4
        1.3.6 EPA Responsibilities	4
        1.3.7 Funding Source	4
    1.4 Verification Testing Site	5
        1.4.1 Site Description	5
        1.4.2 Source/Feed Water	5
        1.4.3 Pilot Effluent Discharge	6
    1.5  Background	6

Chapter 2 - Equipment Description and Operating Processes	7

    2.1  General Equipment Description	7
    2.2 UF120 S2 Ultrafiltration Membrane Module Description	7
        2.2.1 General Description	7
        2.2.2 Environmental Requirements of UF120 S2 Membrane	8
        2.2.3 Materials of Construction of UF120 S2 Membrane	8
    2.3  Operating Process	8
        2.3.1 Process Instrumentation	9
        2.3.2 Consumables	9
        2.3.3 Product Performance Capabilities	10

Chapter 3 - Methods and Procedures	11

    3.1  Environmental Technology Verification Testing Plan	11
        3.1.1 Task 1: Membrane Flux and Recovery	11
        3.1.2 Task 2: Chemical Cleaning Efficiency	12
                                           iv

-------
                                 Table of Contents (continued)

        3.1.3  Task 3: Finished Water Quality	13
        3.1.4  Task 4: Reporting of Membrane Pore Size	13
        3.1.5  Task 5: Membrane Integrity Testing	14
        3.1.6  Task 6: Data Handling Protocol	15
        3.1.7  Task 7: Quality Assurance/Quality Control	15
        3.1.8  Task 8: Operational Conditions and Maintenance	21
    3.2 Calculation of Operating Parameters	21
        3.2.1  Flux	21
        3.2.2  Specific Flux	21
        3.2.3  Normalized Specific Flux	21
        3.2.4  Feedwater System Recovery	22
    3.3 Calculation of Data Quality Indicators	22
        3.3.1  Representativeness	22
        3.3.2  Statistical Uncertainty	24
        3.3.3  Accuracy	24
        3.3.4  Precision	25
    3.4 Safety Measures	25
    3.5 Testing Schedule	26

Chapter 4 - Results and Discussion	27

    4.1 Task 1: Characterization of Membrane Flux and Recovery	27
    4.2 Task 2: Evaluation of Cleaning Efficacy	28
    4.3 Task 3: Evaluation of Finished Water Quality	29
        4.3.1  Turbidity, Particle Counts, and Particle Removal	29
        4.3.2  Microbial Removal	30
        4.3.3  Other Water Quality Parameters	31
    4.4 Task 4: Reporting Membrane Pore Size	32
    4.5 Task 5: Membrane Integrity Testing	32
    4.6 Task 6: Data Management	33
        4.6.1  Data Recording	33
        4.6.2  Data Entry, Validation ,and Reduction	33
    4.7 Task 7: Quality Assurance/Quality Control (QA/QC)	34
        4.7.1  Data Correctness	34
    4.8 Operational Conditions and Maintenance	35
        4.8.1  Overall Operation and Maintenance	35
        4.8.2  System Chemical Consumption	36
        4.8.3  Review of Operations Manual	36
        4.8.4  Equipment Deficiencies Experienced During the ETV Program	36

Chapter 5-References	37

Chapter 6 - Vendor Comments	38

Table and Figures	39

-------
                                 Table of Contents (continued)

Appendices

Appendix A.  Polymem UF120  S2 Ultrafiltration Module  Operation & Maintenance Manual
             and Photographs
Appendix B.  Custom Membrane Pilot Plant (CMPP) Operating Manual
Appendix C.  NLS and MWH Laboratory Certifications
Appendix D.  Copy of Field Notebook and Data Log Sheets
Appendix E.  Instrument Calibration Certificates
Appendix F.   Material Safety Data Sheets
Appendix G.   Laboratory Data Sheets
Appendix H.   Summary of Pore Size Characterization (By Polymem)
Appendix I.    NLS and MWH QA/QC Results
Appendix J.    Data Completeness Tables

                                    List of Tables

Table 1-1.   Lake Michigan Feed Water Quality Data	41
Table 3-1    Verification Conditions	42
Table 3-2    Operational Data	42
Table 3-3    Chemical Cleaning Procedures and Conditions	43
Table 3-4    Cleaning Verification Data	43
Table 3-5    Water Quality Sampling Schedule	44
Table 3-6    Data Management	45
Table 3-7    Methods for Measuring Precision and Accuracy	46
Table 3-8    Sample Methodology	47
Table 3-9    Corrective Action Plan	48
Table 4-1    Operational Conditions	49
Table 4-2    Summary of Operational Data	50
Table 4-3    Summary of Chemical Cleaning Water Quality  Analyses	51
Table 4-4    Summary of Chemical Cleaning Hydraulic Analyses	52
Table 4-5    Summary of Chemical Cleaning Efficacy	53
Table 4-6    Summary of Onsite  Bench-top Turbidity Data	53
Table 4-7    Summary of Online Turbidity and Particle Count Data	54
Table 4-8    Summary of Microbial Water Quality	55
Table 4-9    General Water Quality Parameters	56
Table 4-10   TSS Mass Balance	57
Table 4-11   Summary of Verification Data for the Feed Water Particle Counter	58
Table 4-12   Summary of Verification Data for the Filtrate Water Particle Counter	59
Table 4-13   Chemical Consumption Analysis	60
                                          VI

-------
                             Table of Contents (continued)

                                    List of Figures

Figure 2-1   Process and Instrumentation Diagram	61
Figure 4-1   TMP, Flux, and System Recovery Profiles	62
Figure 4-2   Operational Flux and Temperature Profiles	63
Figure 4-3   Online and Bench-top Turbidity Profiles	64
Figure 4-4   Feed and Filtrate Count Profiles for Particles >2 um	65
Figure 4-5   Feed and Filtrate Count Profiles for Particles 2-3 um	66
Figure 4-6   Feed and Filtrate Count Profiles for Particles 3-5 um	67
Figure 4-7   Feed and Filtrate Count Profiles for Particles 5-7 um	68
Figure 4-8   Feed and Filtrate Count Profiles for Particles 7-10 um	69
Figure 4-9   Feed and Filtrate Count Profiles for Particles 10-15 um	70
Figure 4-10  Feed and Filtrate Count Profiles for Particles >15 um	71
Figure 4-11  Log Removal Profile for Particles >2 um in Size	72
Figure 4-12  Log Removal Profile for Particles 2-3 um in Size	73
Figure 4-13  Log Removal Profile for Particles 3-5 um in Size	74
Figure 4-14  Log Removal Profile for Particles 5-7 um in Size	75
Figure 4-15  Log Removal Profile for Particles 7-10 um in Size	76
Figure 4-16  Log Removal Profile for Particles 10-15 um in Size	77
Figure 4-17  Log Removal Profile for Particles >15 um in Size	78
Figure 4-18  A Frequency Distribution of Filtrate Turbidity	79
Figure 4-19  A Frequency Distribution of Feed Water Turbidity	80
Figure 4-20  A Frequency Distribution for Filtrate Particles >2 um	81
Figure 4-21  A Frequency Distribution for Feed Water Particles >2 um	82
Figure 4-22  A Frequency Distribution of Log Removal for Particles >2 um	83
Figure 4-23  A Frequency Distribution of Log Removal for Particles 2-3 um	84
Figure 4-24  A Frequency Distribution of Log Removal for Particles 3-5 um	85
Figure 4-25  A Frequency Distribution of Log Removal for Particles 5-7 um	86
Figure 4-26  A Frequency Distribution of Log Removal for Particles 7-10 um	87
Figure 4-27  A Frequency Distribution of Log Removal for Particles 10-15 um	88
Figure 4-28  A Frequency Distribution of Log Removal for Particles >15 um	89
Figure 4-29  Feed and Filtrate Particle Count Profiles for Particles >2 um from 3/14/02 00:00
            to 3/14/02 20:44 Using Raw Data	90
Figure 4-30  Feed and Filtrate Particle Count Profiles for Particles >2 um from 3/14/02 00:00
            to 3/14/02 20:44 Excluding Data Collected During Backwash Events	91
Figure 4-31  A Frequency Distribution of Log Removal for Particles >2 um from 3/14/02
             00:00 to 3/14/02 20:44	92
                                          vn

-------
                             Abbreviations and Acronyms
b
°C
cfm
CPU
cn>
C12
CMPP
CPU
DWS
EPA
°F
Fe
ft2
FTO
GBWUFP
gfd
gpm
HMI
HPC
hr
IESWTR
kWh
L
LT1ESWTR

LT2ESWTR

m2
MF/UF
mg/L
mL
Bar
Degrees Celsius
Cubic feet per minute
Colony Forming Units
Clean-in-place
Chlorine
Custom Membrane Pilot Plant
Central Processing Unit
Drinking Water System
United States Environmental Protection Agency
Degrees Fahrenheit
Iron
Feet squared
Field Testing Organization
Green Bay Water Utility Filtration Plant
Gallons per square foot per day
Gallons per minute
Human-Machine Interface
Heterotrophic Plate Count
Hour
Interim Enhanced Surface Water Treatment Rule
Kilowatt hour
Liters
Long Term 1 Enhanced Surface Water Treatment
Rule
Long Term 2 Enhanced Surface Water Treatment
Rule
Meters squared
Mi crofiltrati on/Ultrafiltrati on
Milligrams per liter
Milliliter
                                         Vlll

-------
jim                                  Micrometers
MPN                                Most Probable Number
MSDS                               Material Safety Data Sheets
NaOCl                               Sodium Hypochlorite
NaOH                               Sodium Hydroxide
NLS                                 Northern Lake Service, Inc.
NSF                                 NSF International
NTU                                Nephelometric Turbidity Units
P                                    Pump
P/A                                  Presence/Absence
PLC                                 Programmable Logic Controller
ppm                                 Parts per million
psi                                   Pounds per square inch
PSTP                                Product Specific Test Plan
QA/QC                              Quality Assurance/Quality Control
SCFM                               Standard cubic feet per minute
SDWA                              Safe Drinking Water Act
STP                                 Standard Temperature and Pressure
SUVA                               Specific Ultraviolet Absorbance
TDS                                 Total Dissolved Solids
TMP                                Transmembrane Pressure
TOC                                Total Organic Carbon
TSS                                 Total Suspended Solids
VFD                                Variable Frequency Drive

-------
                               ACKNOWLEDGMENTS

The Field Testing Organization, Carollo Engineers, P.C., was responsible for elements in the
testing sequence, including collection of samples,  calibration and verification  of instruments,
data collection and analysis, data management, data interpretation and the preparation of this
report.

       Carollo Engineers, P.C.
       12592 W. Explorer Dr., Suite 200
       Boise, ID 83713
       Contact Person: Daniel A. Hugaboom, P.E.

The laboratory selected for microbiological analyses and non-microbiological analytical work of
this study was:

       Northern Lake Service, Inc.
       400 North Lake Avenue
       Crandon, WI 54520
       Contact Person: R.T. Krueger

The laboratory selected for microbiological analytical work of this study was:

       MWH Laboratories
       555 E. Walnut Street
       Pasadena, CA  91101
       Contact Person: Jim Hein

The Manufacturer of the Equipment was:

       Polymem
       Route de Revel F-31450
       Fourquevaux, France
       Contact Person: Jean-Michel Espenan

Carollo Engineers,  P.C. and Polymem wishes to thank the Energy  Center  of Wisconsin for
supporting this study.

Carollo Engineers, P.C.  wishes to thank Patricia Terry, Ph.D.  - University  of Wisconsin Green
Bay and the Green Bay Water Utility for their help during this study.

-------
                                       Chapter 1
                                     Introduction

1.1    ETV Purpose and Program Operation

The U.S. Environmental Protection Agency (EPA) has created the Environmental Technology
Verification  (ETV)  Program  to  facilitate  the deployment  of  innovative  or  improved
environmental technologies through performance verification and dissemination of information.
The goal of the ETV program is to further environmental protection by substantially accelerating
the acceptance and use of improved and more cost-effective technologies. ETV seeks to achieve
this goal by providing high quality,  peer reviewed  data on technology performance to those
involved in the design, distribution, permitting, purchase, and use of environmental technologies.

ETV works in partnership with recognized standards and testing organizations;  stakeholder
groups  which consist  of buyers,  vendor  organizations, and permitters;  and with  the  full
participation of individual technology developers. The program evaluates the performance of
innovative technologies by developing test plans that are responsive to the needs of stakeholders,
conducting field or laboratory tests (as appropriate), collecting and analyzing data, and preparing
peer reviewed reports. All  evaluations  are conducted  in accordance with rigorous  quality
assurance protocols to ensure that data of known and  adequate quality are generated and that the
results are defensible.

The EPA has partnered with NSF International (NSF), an independent, not-for-profit testing and
certification organization dedicated to public health, safety and  protection of the environment, to
verify performance of small drinking water systems that serve small communities under the ETV
Drinking Water  Systems (DWS)  Center.  A goal  of verification  testing  is  to enhance  and
facilitate the acceptance of small drinking water treatment equipment  by state drinking water
regulatory officials and consulting engineers while reducing the need for testing of equipment at
each location where the equipment's use is contemplated.  NSF will meet this goal by working
with manufacturers and NSF-qualified Field Testing Organizations (FTO) to conduct verification
testing under the approved protocols. It is important to note that verification of the  equipment
does not mean that the equipment is "certified"  by NSF  or  "accepted" by  EPA.  Rather,  it
recognizes  that the performance of the equipment has been determined and  verified by these
organizations for those conditions tested by the FTO.

The DWS Center evaluated the performance the Polymem UF120 S2 Ultrafiltration Membrane
Module, which is  a membrane filtration  technology used in drinking  water  treatment system
applications. The test evaluated the membrane module's ability to physically remove microbial
and particulate contaminants.  This  document provides the  verification  test results for the
Polymem UF120 S2 Ultrafiltration Membrane Module.

1.2    Testing Participants

The FTO was Carollo Engineers, P.C., which provided the overall management of the ETV.  The
Ultrafiltration membrane manufacturer for the ETV was Polymem. The  operations management
and staff were from the University of Wisconsin-Green Bay and Carollo Engineers. Laboratory
analyses were performed by Northern Lake Service, Inc. (NLS),  Crandon, Wisconsin  and

-------
Montgomery Watson Laboratories, Pasadena, California (total coliform enumeration only). Data
management and final report preparation were performed by the FTO, Carollo Engineers, P.C.

1.3    Definition of Roles and Responsibilities of Project Participants

The following is a brief description of each ETV participant and their roles and responsibilities.

1.3.1    NSF Responsibilities
NSF is a not-for-profit testing and certification organization dedicated to public health safety and
the protection of the environment.  Founded in 1946 and located in Ann Arbor, Michigan, NSF
has been instrumental in the development  of consensus standards for the protection of public
health and the environment. NSF  also provides testing and certification services to ensure that
products bearing the NSF Name, Logo and/or Mark meet those standards.  The EPA partnered
with the NSF to verify the performance of drinking water treatment systems through the EPA's
ETV Program.

NSF provided technical oversight of the verification testing.  An audit of the field analytical and
data gathering  and recording  procedures was conducted.  NSF also provided review of the
Product Specific Test Plan (PSTP) and this report.

Contact Information:
       NSF International
       789 N. Dixboro Rd.
       Ann Arbor, MI 48105
       Phone: 734-769-8010
       Fax: 734-769-0109
       Contact: Bruce Bartley, Project Manager
       Email: bartley@nsf.org

1.3.2    Field Testing Organization Responsibilities
Carollo Engineers,  P.C.  conducted the  verification testing of the Polymem UF120  S2
Ultrafiltration Membrane Module.  Carollo Engineers, P.C. is an NSF-qualified FTO for the
ETV Drinking Water Systems Center.

The FTO was responsible for conducting the verification testing for 30 calendar days.  FTO
employees  conducted the onsite analyses and data recording during the testing. The  specific
responsibilities of the FTO, Carollo Engineers, P.C., were to:

•  Provide the overall management of the ETV through the proj ect manager and the proj ect
   engineers.

•  Provide needed logistical support, the project communication network,  and scheduling and
   coordination of the activities  of participants.

•  Manage, evaluate, interpret and report on data generated in the ETV.

-------
•  Evaluate the performance of the ultrafiltration membrane technology according to the PSTP
   and the testing, operations, quality assurance/quality control (QA/QC), data management and
   safety protocols contained therein.

•  Provide quality control (QC) information in the ETV report.

•  Provide data generated during the ETV in  hard  copy and electronic form in a common
   spreadsheet or database form.

Contact Information:
       Carollo Engineers, P.C.
       12592 W. Explorer Dr.,  Suite 200
       Boise, ID 83713
       Phone: (208) 376-2288
       Contact Person: Daniel A. Hugaboom, P.E.
       Email: dhugaboom@carollo.com

1.3.3    Manufacturer Responsibilities

The specific responsibilities of the ultrafiltration membrane manufacturer, Polymem, were to:

•  Provide complete, field-ready equipment for the ETV at the testing site.

•  Provide logistical and technical support as required throughout the ETV.

•  Provide partial funding for the project.

•  Attend project meetings as necessary.

Contact Information:
       Polymem
       Route de Revel F-31450
       Fourquevaux, France
       Phone: Oil.33.5.65.71.79.89
       Contact Person: Jean-Michel Espenan
       Email: polymem@wanadoo.fr

1.3.4    Operator and Test Site Staff Responsibilities

The specific responsibilities of operations and test site staff from the University of Wisconsin at
Green Bay, Carollo Engineers, P.C., and Green Bay Water Utility, were to:

•  Provide set-up, shakedown,  operations, maintenance and on-site analytical services according
   to the PSTP and the testing, operations, QA/QC, data management and safety protocols.

•  Provide the necessary and appropriate space for the equipment to be tested in the ETV.

•  Provide necessary electrical power, feedwater and other utilities as required for the ETV.

•  Provide necessary drains from the test site.

-------
Contact Information:
       University of Wisconsin-Green Bay
       24200 Nicolet Dr.
       Green Bay, WI 54311-7001
       Phone: (920) 465-2278
       Contact Person: Patricia Terry, Ph.D.
       Email: terryp@uwgb.edu

1.3.5    Water Quality Analyst Responsibilities

The specific  responsibilities  of the  water  quality  analytical staff from  the NLS  and MWH
Laboratories were to:

•  Provide off-site water quality analyses prescribed in the PSTP  according to the QA/QC
   protocols contained therein.

•  Provide reports with the analytical results to the data manager.

•  Provide detailed information on the analytical procedures implemented.

Contact Information:
       Northern Lake Service, Inc.
       400 North Lake Avenue
       Cranon, WI 54520
       Phone:(715)478-2777
       Contact Person: R.T. Krueger
       E-Mail: Krueger@northernlakeservice.com

       MWH Laboratories
       555 E. Walnut Street
       Pasadena, CA 91101
       Phone: (626) 568-6400
       Contact Person: Jim Hein
1.3.6    EPA Responsibilities
The EPA through its Office of Research and  Development has financially  supported and
collaborated with NSF under Cooperative Agreement No. R-82833301. This verification effort
was supported by the DWS Center operating under the ETV Program. This document has been
peer reviewed and reviewed by NSF and EPA and recommended for public release.

1.3.7    Funding Source

The ETV testing described herein was funded in conjunction with a  12-month membrane pilot
study funded by the Energy Center of Wisconsin. The 12-month study, focused not only on the
power  requirements of membrane  filtration plants, but also on promoting the use of more
effective and less energy intensive  water treatment  technologies. The Energy  Center  of

-------
Wisconsin chose to participate because the overall scope of the ETV testing fit into of the scope
of the longer, energy focused study.

1.4    Verification Testing Site

1.4.1    Site Description

The  testing site  was the  Green Bay Water Utility Filtration Plant (GBWUFP) located at
6183  Finger Road in Luxemberg,  Wisconsin. The  plant utilizes pre-ozonation, coagulation,
flocculation, sedimentation, granular media filtration,  and chlorination prior to distribution to the
City of Green Bay. The plant also  has facilities for  recycling filter washwater, which include
settling lagoons and washwater recycle basins.

The GBWUFP is  fed by two raw water intakes located on the western shore of Lake Michigan in
Kewaunee, Wisconsin. The primary intake structure is located approximately one mile off shore
at a depth of 50 feet and is operated year round; a second intake structure located  3000 feet off
shore at a depth of 30 feet is operated during peak demand  periods to supplement the primary
intake. The raw water is pumped to the filtration  plant in Luxemberg, Wisconsin. Chlorine is
added at each intake to prevent growth of zebra mussels during transmission from intake to the
treatment  facility. Residual chlorine concentrations (as Cb)  in the feed water at the treatment
facility were collected during  the ETV test  and tested by GBWUFP  staff and are included in
Table 1-1. The chlorine residual during the test averaged 0.05  mg/L. Because chlorine is added at
the intake, "raw"  Lake Michigan water was not available at the site. However, the low reported
chlorine levels were not expected to  significantly effect membrane performance.

The pilot plant was located approximately 200 feet from the raw water channel at the filtration
plant and  was fed by either or both of the GBWUFP raw water intakes. A submersible pump
located 3 feet below the free water surface fed the pilot plant via 2-inch schedule 80 PVC pipe,
and 1.5-inch PVC tubing. A tee fitting divided flow to serve two pilot plant units and was located
approximately 20 feet from each pilot plant trailer feed connection. Ball valves located on each
branch provided flow-splitting  control. Pictures of the project  site are included in Appendix A.

1.4.2    Source/Feed Water

The  equipment verification testing described in  this report was  executed using raw Lake
Michigan  water obtained from the GBWUFP. Water used for pilot plant operation was drawn
from the process  prior to any treatment at the water facility (other than chlorine addition for
zebra mussel control) and was pumped approximately 200 feet to the skid mounted pilot plant
located inside a module trailer unit.

GBWUFP staff monitor Lake  Michigan water quality as part of normal  plant operations. In
general, Lake Michigan can be characterized  as a high quality raw water. Based on historical
data, the raw water can be characterized by low  average turbidity. However, fall and winter
storm events can  significantly  increase turbidity for several days at a  time. Seasonal variations
may also  cause  significant variation  on  water temperature  and correlate  to algae blooms.
Alkalinity, pH, and TOC show less seasonal variation than turbidity and temperature. Specific
feed water quality data collected throughout this verification testing are reported in Table 1-1.
The feed water had the following average water quality during this evaluation: Cb residual 0.05

-------
mg/L, alkalinity  110 mg/L as CaCCb, total hardness 130 mg/L as CaCCb, calcium hardness 88
mg/L as CaCOs, total dissolved solids (IDS) 187 mg/L, total suspended solids (TSS) 1.3 mg/L,
total organic carbon (TOC), 2.3 mg/L, UVA 0.024 per cm"1, algae 34/ml, pH 7.8, temperature
3°C (37°F), total coliform 5.0 MPN/lOOml, HPC  17 CFU/ml, and turbidity 1.3 NTU.

1.4.3    Pilot Effluent Discharge

Sanitary sewers do not exist at the testing site. Therefore, the effluent (filtrate and backwash) of
the pilot treatment unit was routed to the beginning of the GBWUFP facilities for recycling with
filter washwater. Chemical cleaning wastes were neutralized and disposed of off site.

1.5    Background

The  Interim  Enhanced  Surface Water  Treatment  Rule   (IESWTR),  promulgated  in  1998,
established stringent filtered water quality standards for systems serving over 10,000 people.
Promulgated at the same time as the IESTWR,  was the Stage 1 Disinfectant and Disinfection
Byproduct Rule (Stage 1 D/DBPR) to regulate levels of disinfectant residuals and byproducts in
the distribution system of water systems  of all sizes. Emerging regulations due in the calendar
year 2001, include the Long Term 1 Enhanced Surface Water Treatment Rule (LTIESWTR) that
will extend the requirements of the IESWTR to systems serving less than  10,000 people. Stage 2
LT2ESWTR will be applicable to systems of all sizes treating surface water and groundwater
under the influence of  surface water.  The  LT2ESWTR  "Agreement in Principal" includes
regulation of Cryptosporidium, and technologies that utilities must utilize to remove it from their
water source. These emerging regulations require greater removal of pathogens, lower  filter
effluent turbidity standards, and lower disinfectant byproduct concentrations in the distribution
system.

In order  to meet these more stringent  regulations, water utilities  are evaluating  numerous
treatment options including low-pressure membrane filtration such as microfiltration (MF) and
ultrafiltration (UF). Although not directly verified in this  ETV test,  these types of membrane
systems are designed to remove bacteria, protozoa, and in the case of UF membranes, viruses. In
the coming years, most water systems using surface water or groundwater under the influence of
surface water  will need  to evaluate their capacity  to meet  these objectives. These  systems
constitute the majority of the market for the product being evaluated in this report.

At the time this report was written, amendments to the Surface Water treatment Rule (SWTR), as
mandated by the Safe Drinking Water Act (SOWA) were actively being negotiated. While final
rules and  amendments were being negotiated, the implications  of these new rules  were well
known in  the  industry, and  many  water utilities were focusing their plant upgrade and new
construction on these anticipated standards.

-------
                                      Chapter 2
                    Equipment Description and Operating Processes

2.1    General Equipment Description

Critical to the verification testing was the use a "generic" custom membrane pilot plant (CMPP).
The CMPP was not provided by the membrane manufacturer, Polymem. The CMPP used has the
capacity to feed, backwash and  clean a variety of MF/UF modules that are housed in pressure
vessels. Therefore, this ETV testing verified the operation of the  membrane module under a
given set of operational parameters, not membrane-specific process equipment. A description of
the equipment used in the process is provided in this section.

2.2    UF120 S2 Ultrafiltration Membrane Module Description

2.2.1    General Description

The  Polymem  UF120  S2 Ultrafiltration  Membrane Module is comprised  of 19 individual
polysulfone hollow-fiber membrane bundles housed in a PVC pressure vessel. The bundles are
potted on the effluent side of the module, forming a U-shaped configuration and provide a total
of 114 m2 (1227 ft2) of active membrane  surface area.  The nominal pore size, as specified by
Polymem  and not verified in this ETV test,  is 0.01 jim, and is classified as an Ultrafiltration
membrane. This pore size should provide a physical barrier to bacteria, protozoans, and viruses
when membrane fibers are intact and operated within the recommended operating ranges.

The  membrane module is designed to be  operated in  a dead-end  mode, reducing  power
consumption  over traditional cross flow membrane products, as recirculation pumps  are  not
required. The flow configuration is outside to inside. This forces the accumulation of particulate
matter,  pathogens,  and  suspended  solids  on the outside  of the  membrane fiber.  The
recommended backwash procedure includes  simultaneous hydraulic backwash, air scour, and
chlorine injection. Backwash is  accomplished by pumping filtrate water from the inside to the
outside of the fiber.  This water is then discharged to waste. An inlet for air scour is provided at
the level of the potting resin via air diffusers located inside the module.

Some  fraction  of the  particulate matter  and  dissolved constituents  in  the feed  water can
accumulate on the membrane surface  and cannot be removed by hydraulic  backwash and air
scour. This leads to a rise in transmembrane pressure during normal operation.  Once the terminal
transmembrane pressure has been reached (29 psi), the  membrane must be taken off-line to
remove this  matter from the membrane with  a chemical clean. The membrane system and
operating strategy (flux, recovery, and backwash intervals) are typically designed for a 30-day
chemical cleaning interval. However, significant changes in water quality will effect membrane
performance.  Temperature fluctuation, increases in natural organic matter, turbidity, pH changes,
and high flux rates may have the  potential to increase membrane fouling rates.

The membrane polymer is tolerant to a variety of chemicals, including chlorine, acids, bases, and
chelating agents commonly used for chemical cleaning.  Photographs  of the module, detailed
specifications, and typical operating parameters are included in Appendix A. Backwash and
chemical cleaning chlorine dose tolerances and pH ranges are also included.

-------
2.2.2    Environmental Requirements of UF120 S2 Membrane

The membrane module should be placed in a location that is protected from high winds, freezing
conditions, direct sunlight,  and precipitation that may damage process piping and  electrical
equipment.

2.2.3    Materials of Construction of UF120 S2 Membrane

The membrane module's (and CMPP  skid) influent and effluent connections are standard-sized
schedule 80  PVC allowing for adaptability  to  custom  designed  membrane systems. The
membrane module itself is housed in a PVC vessel. Connection to the feed side of the module is
made via  3-inch schedule 80 PVC threaded into  the inlet of the  module. Connection on the
filtrate side of the module is made with a 12-inch stainless steel mechanical coupling and a 12-
inch by 3-inch PVC reducer. As noted, the membrane is polysulfone, potted on the filtrate side of
the module body with polyeurothane. A H-inch air scour diffuser inlet is provided at the potting
level to aid in cleaning efficacy. The membrane module weighs 99 pounds when empty and 187
pounds when full of water. The length  and diameter of the module are 37 and 12 inches,
respectively.

2.3    Operating Process

The typical UF process consists of pumps, piping, and control systems capable of performing
basic functions necessary for membrane filtration. The  process consists of a feed water pump,
membrane module, backwash pump, chemical feed systems, filtrate and chemical storage.  Feed
pumping is used during production cycles  to provide flow at the necessary pressure for driving
water across the membrane surface. At regular intervals (30-120  minutes), the feed pump is
turned off, and the backwash pump moves water into the effluent  side of the membrane at an
elevated pressure for short periods of time (60 seconds) to remove  accumulated solids from the
membrane surface. The hydraulic  backwash is used  in unison with air distributed through
diffusers located inside the  membrane module. A schematic of the CMPP process including
pumps, piping schematic, sample ports, chemical  feeds, and UF120 S2 membrane location is
included in the process and instrumentation diagram (Figure 2-1).

Because production backwashes do  not  remove all  particulate  matter from the membrane
surface, periodic  chemical  cleanings  must be performed  to restore membrane permeability.
Chemical  cleanings are performed by soaking the membrane fibers in  an appropriate solution,
followed by a rinsing cycle. This procedure takes approximately  8 hours to perform, and must be
manually initiated through the Programmable Logic Control (PLC) interface.

Periodic direct integrity monitoring checks to ensure that membrane fibers are not compromised
(therefore  allowing passage of particulate contaminants larger than the membrane pore size) are
performed at regular intervals and generally prescribed by the regulatory agency responsible for
the water system. On the CMPP used in the verification, this is a manual operation. The process,
in  general, is to pressurize the inside of the membrane fibers with air, and measure the pressure
decay rate. For the UF120 S2 membrane, the acceptable decay rate limit specified by Polymem
is  0.36 psi/min. Decay rates  in  excess  of this  limit indicate that  fiber  integrity  may be
compromised.

-------
CMPP equipment requires a 3-phase/480 V power supply to run pumps and variable frequency
drives (VFD). A transformer located in the VFD panel converts the power to 120 V single phase
for the PLC panel.

2.3.1    Process Instrumentation

The CMPP skid is equipped with a PLC unit that provides a significant degree of automation.
The PLC maintains flow rates, pressures, control valve positions, and flow during backwashes
and chemical cleans. The PLC also has the ability to shut down the system in case of a high level
alarm (high transmembrane pressure,  low liquid levels, low chemical tank levels,  etc.). The
process can be configured for the specific  application through a Windows-based  interface.
Backwash  and cleaning procedures,  flow,  pressure, alarms,  and dose  set  points can  be
customized via the PLC panel. This greatly reduces the required  operator training  and time
requirements. Functions can be adjusted, initiated, and monitored remotely. The system can be
operated without  a full-time operator; however,  operators are necessary  for  routine  system
maintenance (e.g., analytical sampling,  compressor maintenance,  and to  maintain acceptable
volumes in chemical storage tanks), maintaining water quality monitoring devices, performing
integrity tests, and trouble-shooting.

Critical data including flow rate, transmembrane pressure, flux, specific flux, temperature, and
turbidity can be downloaded remotely for analysis. However, maintenance issues  including
cleaning pre-filters, filling  feed tanks, and required sampling  must be performed by  on-site
personnel. Data collection with the exception of particle count  data and manual gage readings
can be done remotely with remote control software over a standard telephone line.

2.3.2    Consumables

2.3.2.1  Chemicals

Chemicals used by the membrane are readily available commodity products [chlorine, sodium
hydroxide (NaOH), or acid if necessary]. Typical chemical consumption includes low doses of
chlorine  [approximately 2  pounds per module per month  as  sodium hypochlorite  (NaOCl)]
during production backwashes (occurring every 30 to 120 minutes) and during chemical cleaning
(0.05 pounds per module per month) if performed once each month. NaOH is also typically used
for removal of organic compounds during  monthly chemical cleaning (2.4 pounds per module
per month). Chemical cleaning wastes can be disposed of in sanitary sewers as allowed by local
code.  If necessary, these chemicals can be quenched (chlorine) or  neutralized (acid or caustic)
prior to discharge. Sanitary sewers do not exist at the site. Therefore, chemical cleaning residuals
were neutralized and disposed of off-site.

2.3.2.2  Power

Power requirements for MF/UF membrane systems are driven primarily by feed water pumping
requirements.  Operating pressures vary significantly over the course of chemical cleaning
intervals, depending on operating strategies and feed water quality. Typical operating  pressures
(estimated from previous studies with the Polymem membrane on Lake Michigan water) range
from  5  to  15  psi. Based on a constant operating  pressure of 15 psi, feed pumping power
requirements are approximately 0.157 kWh per day per gpm of feed water flow.

-------
2.3.3    Product Performance Capabilities

The UF120 S2 membrane is capable of removal of solids primarily through a physical sieving
mechanism. The UF120 S2 membrane will remove coliforms, HPC, and other participate matter
from the feed water.  The manufacturer's stated performance capabilities were used to shape the
data quality objectives (DQO) and testing plan used for this ETV test.

Note: Challenge testing, microbial or otherwise, was not performed as part of this study; particle
removal was quantified based on turbidity and particle counter data.
                                          10

-------
                                       Chapter 3
                                Methods and Procedures

This chapter includes a detailed discussion of the ETV experimental plan,  testing conditions,
methods,  and sampling  parameters and  frequency.  In addition, this  chapter  details field
operational and maintenance procedures, quality assurance and quality  control, and  analytical
methods used throughout the ETV testing.

3.1    Environmental Technology Verification Testing Plan

3.1.1    Task 1: Membrane Flux and Recovery

The objective of this task  was to demonstrate:  1) appropriate operational  conditions for  the
membrane module; 2) the product water recovery achieved by the membrane module; and 3)  the
rate of specific flux decline observed over extended membrane filtration operation.

3.1.1.1  Work Plan

Fouling rates of a membrane are functions of both water quality and operational conditions.
Several months of pilot plant operation were completed using the test water  source prior to  the
verification testing. Optimized operational conditions were estimated based on these data, which
were collected prior to March 1, 2002.  The verification testing took place  over  a continuous
46-day  period beginning March 11, 2002 and ending April 26, 2002.  This time frame was
established to test the ability of the membrane to operate with cold water and during a period of
time when there have  historically been turbidity spikes in the GBWUFP raw water.  The
operational data used in the testing are shown in Table 3-1 and include flux, recovery, backwash
interval, data logging interval, production backwash chlorine concentration, air scour flow rate,
and terminal transmembrane pressure. The first run was set at a constant  normalized flux of 40
1/h-m2 (24 gfd) at 20°C (68°F). The second run of testing was set at a constant normalized flux of
30 1/h-m2 (18 gfd)  at 20°C (68°F). The system was operated at constant normalized flux to
minimize  temperature  related  viscosity  effects on transmembrane pressure and to  facilitate
interruption of flux loss rate.

Operating conditions were set through  the  PLC  human-machine interface (HMI)  prior to
beginning the test. These  conditions included flux rate, production dwell time, backwash
procedures (interval  and duration), alarm condition  settings,  chemical  feed doses,  and data
logging intervals. With the exception of backwash duration, these parameters were not adjusted
during  operation. Backwash duration was adjusted as needed to maintain a recovery of at least
90 percent. Recovery was determined daily by visually observing the total backwash volume
used during a production backwash cycle as indicated on the filtrate  tank site glass.  Total
production volumes were determined using the filtrate flow rate at the end of a production cycle
and multiplying by the total production cycle time (i.e. 50 minutes).

During  the test  run,  specific operational data were collected to quantify  fouling  rates and
hydraulic performance. The operational data and the collection frequency are shown in Table  3-2
and include:  feed/filtrate flow  rate, feed/filtrate  pressure,  feed temperature, transmembrane
pressure, flux and specific flux at 20°C (68°F), and recovery.
                                           11

-------
3.1.2    Task 2: Chemical Cleaning Efficiency

The objective of this task was to demonstrate the effectiveness of chemical cleaning for restoring
the specific flux of the membrane module. The cleaning procedures used were selected from the
manufacturer's  recommended  cleaning  strategies,  as  indicated  in the  Operations  and
Maintenance (O&M) Manual. The task was intended to serve as a "proof of concept" effort.

3.1.2.1  Work Plan

Prior to beginning the ETV testing, a membrane chemical cleaning procedure was performed to
ensure that the  module was  clean and that the storage solution had been purged. In addition,
chemical cleanings were performed once terminal transmembrane pressure was reached in order
to restore membrane permeability.  Lake Michigan water has low concentrations  of dissolved
minerals. Therefore, after membrane fouling, a chlorine solution at a pH of 11-12 was used to
oxidize and dissolve organic  foulants, and to remove bacterial growth. A total of three chemical
cleanings were performed for this testing as a "proof of concept" effort: 1) prior to beginning the
verification test, 2) after approximately 12.5 days, just prior to when the terminal transmembrane
pressure was reached, and 3)  at the end of the 46 days of run time.

The membrane  module was  chemically cleaned, per the manufacturers recommended cleaning
strategies, with  approximately 200 mg/L chlorine and sufficient caustic soda to maintain a pH of
11-12  (5000-7500 mg/L NaOH).  The  chemicals were dosed  automatically with  chemical
metering pumps and set points  entered into the PLC. Doses were verified by  measuring total
chlorine and pH on site immediately following chemical injection. The first two cleanings were
performed  at  ambient  temperature  while the final cleaning was  performed  at elevated
temperatures of approximately 27°C (81°F). Prior to beginning the verification test, the new
membrane was  allowed to soak in the chemical cleaning solution for a period  of 4 hours. The
two subsequent chemical cleans were allowed to  soak for 8 and 48  hours, after production
periods 1  and 2,  respectively. Following  the   chemical  cleaning  soak,  the  module was
backwashed with sufficient volume of filtrate to purge the system of the chemical solution (>30
gallons) and was placed in  filter-to-waste mode  for 90 seconds. A baseline transmembrane
pressure versus  flux trend was established immediately following each chemical  clean. Chemical
cleaning wastes were neutralized and disposed of off site.

Table 3-3 provides information on the chemicals used and  conditions of the chemical cleaning
procedure including chlorine consumption,  chemical  cleaning steps, pH  and  Cb of cleaning
residuals,  and  the  volume of water required for  chemical cleaning. Detailed procedures are
included in the  membrane and CMPP O&M Manuals in Appendices A and B. Table  3-4 shows
the timing and location of samples  collected during each chemical cleaning. The samples were
tested for parameters including temperature, pH, TDS, and turbidity. In addition, specific flux,
flux, transmembrane pressure, and flow rate was recorded. Collected data were used to  verify
cleaning conditions and assess cleaning effectiveness.

Estimates of specific flux recovery,  cleaning efficacy and assessment of irreversible loss of
membrane  specific flux  were reported  for  estimating the usable  life of the membrane. A
calculation of  specific  flux recovery and  original  specific  flux  loss, as  defined  in  the
                                          12

-------
ETV Protocol for Equipment Verification  Testing for Physical Removal of Microbial and
Particulate Contaminants (May 1999) were calculated as follows:

                                     Jsf
% Recovery of Specific Flux = 100 *
Loss of Original Specific Flux = 100 *
                                   1-
                                     Jst_

                                       Js,
Jsf= specific flux at end of current run
Jsj = specific flux at beginning of subsequent run
Jsio = specific flux at beginning of membrane test run

3.1.3    Task 3: Finished Water Quality

The objective of this task was to assess the ability of the membrane module to meet the water
quality objectives outlined by Polymem.

3.1.3.1  Work Plan

To  adequately define ETV conditions and to assess  the contaminant rejection capacity of the
UF120 S2, feed, filtrate,  and backwash water  quality were monitored over the course of the
testing. Water quality monitoring was performed per the requirements of the EPA/NSF ETV
Protocol for Equipment Verification Testing for Physical Removal of Microbial and P articulate
Contaminants (May 1999). The parameters, sampling  frequency, and specific sampling locations
are  presented in Table 3-5. On-site analysis was performed by the CMPP operator.  Off-site
analysis was performed by  NLS, Crandon, WI. Total coliform enumeration was performed by
MWH Laboratories, Pasadena,  California. NLS and MWH state  certification documents  are
provided in Appendix C. Sampling procedures are detailed in the Section 3.1.7, Task 7: QA/QC.

3.1.4    Task 4: Reporting of Membrane Pore Size

The objective of this task was to report the 90 percent and maximum pore size for the Polymem
UF120 S2 ultrafiltration membrane.

3.1.4.1  Work Plan

The nominal  pore size of the UF120 S2  membrane module used in this ETV study  was
determined by the manufacturer through the use of multiple methods including scanning  electron
microscopy (SEM), flow porometry, and particle retention testing. The UF120 S2 membrane is
an asymmetric hollow fiber with a dense layer containing  small pores on the inside and  outside,
and a less dense structure containing relatively larger pores. SEM images were used at multiple
resolutions to estimate the size of the pores in the membrane. Flow porometry (Lee, et al., 1997;
Hernandez, et al., 1999; Mietton and Courtois, 1997) was used to determine the mean size of the
pores in the membrane (including the less dense  middle layer).
                                          13

-------
3.1.5    Task 5: Membrane Integrity Testing

The objective of this test was to demonstrate the methodology to be employed for monitoring
membrane integrity, and to verify the integrity of the membrane fibers.

3.1.5.1  Work Plan

Following the initial and final chemical cleanings, an integrity test was performed consisting of
direct measurements including a pressure decay and visual bubble test. Indirect measurements of
membrane integrity were made throughout the  ETV testing by means of on-line particle count
and turbidity  data. The pressure decay and bubble test methodology is included in the membrane
O&M Manual in Appendix A and involved the steps described below.

3.1.5.2  Direct Methods of Membrane Integrity Testing

Pressure Decay Methodology
•  Operating the module without air scour for several shortened backwash cycles to remove air
   trapped between membrane fibers.

•  Pressurizing the inside of the membrane fibers with air through a valve located on the CMPP
   filtrate piping to 10 psi.

•  Opening the feed sample valve so the outside of the membrane fibers is  at atmospheric
   pressure during the test.

•  Recording measurements  from  the filtrate  pressure  transmitter on the  HMI  every
   15 seconds until the 2-minute test has been completed.

•  The membrane  technical specifications state that a pressure  decay rate  of greater than
   50 mbar over a 2-minute period (0.36 psi/min)  could indicate a compromise in membrane
   integrity.

Bubble check
•  If bubbles are noted in the vertical, transparent section of the CMPP feed water piping during
   the pressure decay test, this indicates that fibers have broken.
3.1.5.3  Indirect Methods of Membrane Integrity Testing

In-line particle counting and turbidity data  were used as an indirect method of monitoring
membrane integrity. Increases in particle counts or turbidity in the filtrate serve as indicators of
potential fiber breakage. The advantage of this method was that it could be used during normal
operation and  did not require the module to be taken out of service as with the direct methods
described above.
                                           14

-------
3.1.6   Task 6: Data Handling Protocol

The objective of this task was to establish a viable structure for the recording and transmission of
field data such that the operational data collected was sufficient and reliable for execution of the
PSTP.

3.1.6.1  Work Plan

Water quality and hydraulic data was maintained on site. Critical process parameters including
pressures, temperature, turbidity, flow rates, and chemical doses were automatically stored on the
membrane skid mounted PLC Central Processing Unit (CPU). This data was also backed up on a
separate personal computer. This  data was recorded at a maximum  of 10-minute intervals
throughout the ETV testing. In addition, online feed and filtrate turbidities were verified once per
day with bench-top measurements. Backwash turbidity was measured twice per day with bench-
top analysis and was not monitored by online instrumentation.

Feed and filtrate particle count  data were recorded at 5-minute intervals for the following size
ranges:  2-3  jim, 3-5  jim,  5-7 jim,  7-10  jim,  10-15  jim,  and >15  jim. The instrument
communicates via a RS485 protocol,  therefore  a RS485/RS232 adapter was required in order to
communicate with the computer used to collect the data. Particle count  data was collected on a
dedicated computer located adjacent  to the CMPP skid.  Both feed and filtrate particle counter
calibration  were field verified with microspheres.  Further details on the field verification
procedures are provided in Section 3.1.7.

pH, alkalinity, total  hardness, calcium hardness, TDS, TSS, TOC, and UVA data were recorded
in tabular  format as results were  received  from  the  analytical testing  laboratory.  MWH
Laboratories, Pasadena, California performed total coliform enumeration. Remaining laboratory
analyses  were  performed by NLS,  Crandon,  Wisconsin.  Likewise,  results from  biological
samples were recorded in a tabular format. Specifically designed daily log sheets were filled out
to include  other details  of the test such as  field tests, maintenance  activities, water type,
backwash procedures, operating cycles,  names of visitors,  problems, etc. and were logged by the
operator in a bound notebook.  Additionally, QA/QC information including manual checks of
pressure, flow  and  temperature transmitters, results of on-site analyses, instrumentation  flow
rates, and other pertinent information  were logged in the same manner. Chain of custody records
for off-site laboratory analysis,  on-site calibration,  and verification of on-line  instrumentation
were also recorded.  Copies of  the laboratory  notebooks and  data log sheets are included in
Appendix D. Table  3-6 summarizes the information management for this testing including how
the information was collected and stored and the reported format.

3.1.7   Task 7: Quality Assurance/Quality Control

The objective of this task was to maintain strict QA/QC methods and procedures during the test
run.
                                           15

-------
3.1.7.1  Work Plan
The quality assurance project plan (QAPP) for this verification testing specifies procedures that
were used to ensure data quality and integrity. Careful adherence to these procedures ensures that
data generated from the verification testing provided  sound analytical  results that serve as the
basis for this performance evaluation. The components of the QAPP for this ETV include:
•  Routine field procedures;
•  Analytical methods;
•  Water quality precision and accuracy;
•  Critical equipment precision and accuracy;
•  Methodology for use of blanks;
•  Description of procedures for performance evaluation samples;
•  Outline for duplicate sampling;
•  Procedures used to ensure data correctness;
•  Procedures for calculating indicators of data quality;
•  Outline for data reporting; and
•  Development of corrective action plan.
3.1.7.2  Routine Procedures
The following procedures were performed prior to the test run:
•  In-line turbidimeter reservoirs were cleaned out and calibrated against a 20 NTU standard.
•  Particle counters were field-verified using microspheres.
•  In-line flow meter was cleaned  and the meter  output was verified  with a bucket and
   stopwatch method.
•  Sample tubing was checked.
The following procedures were performed daily:
•  Routine  daily walkthroughs were conducted to  verify that  each  piece of equipment  or
   instrument was operating properly.
•  In-line turbidimeter flow rates were verified.
•  In-line turbidimeter readings were checked against  a properly calibrated bench model.
•  In-line particle counter flow rates were verified.
•  Chemical feed pump flow rates were verified.
•  Pressure transmitters were checked against pressure gauges that  were calibrated against
   NIST-traceable standards.
                                            16

-------
The following procedure was performed at the start of the run, and every 2 weeks thereafter:

•   In-line flow meter readings were verified with a bucket and stopwatch method.

3.1.7.3  Analytical Methods

Onsite bench-top analyses including turbidity, pH, chlorine, and temperature were conducted
daily  at the test  site  according to Standard Methods for the Examination of Water and
Wastewater, 20( Edition (APHA, 1998) and by Methods for Chemical Analysis of Water and
Wastes  (EPA, 1979), where applicable.  Standard Methods for the Examination of Water and
Wastewater, 20th Edition (APHA, 1998) was followed for total coliform analyses conducted at
NLS, Crandon, Wisconsin  and  MWH Laboratories,  Pasadena,  California. Other  analyses
conducted by NLS were conducted using Standard Methods for the Examination of Water and
Wastewater, 18th Edition (APHA, 1992) and by Methods for Chemical Analysis of Water and
Wastes  (EPA, Revision 1983), where applicable. The analytical methods utilized for this test to
monitor feed, filtrate, and backwash water quality are further described below:

pH - Analysis  for pH was  performed according to Standard Method 4500-H+. A 3-point
calibration of the pH meter used in this study was performed daily.

Temperature - Measurement of temperature was conducted in accordance with Standard Method
2550 B. This was done daily to check the in-line temperature transmitter on the CMPP.

In-Line  Turbidity  - Feed and filtrate water  in-line  turbidity measurements were logged
continuously on the CMPP. The feed and filtrate turbidimeters were calibrated weekly with
primary calibration standards purchased from the turbidimeter manufacturer. In addition, lenses
were cleaned weekly according to the manufacturer's instructions. The data logging readout was
checked daily against the local turbidimeter display value. Turbidimeter flow rates were checked
daily to ensure that they were within the range required by the manufacturer.

Bench-top Turbidity - Turbidity analysis was performed daily  according to Standard Methods
2130 with a bench-top turbidimeter for feed, filtrate, and backwash samples. These results were
used to  verify in-line measurements. The bench-top turbidimeter (Hach 2100 AN) was checked
daily against primary standards (0.061, 20, 200 NTU).

The method for collecting grab  samples consisted  of running  a slow, steady stream from  the
sample tap, triple-rinsing a dedicated sample beaker in this stream, allowing the sample to flow
down the side of the beaker to minimize bubble entrainment. The sample vial was double-rinsed
with the sample, carefully pouring from the beaker down the side of the sample vial, wiping the
sample  vial  clean,  inserting the sample vial into the turbidimeter,  and recording the measured
turbidity upon a stable readout.

Particle Counting  - In-line particle counting was performed on both feed and filtrate waters.
Prior to the  study,  the instrument calibration was field-verified using microspheres as described
in Section 3.1.7.5.  Current instrument and  particle standard calibration certificates and  methods
for demonstration of coincidence error are provided in Appendix E.
                                           17

-------
TOC and W/254 Absorbance - Water samples were collected by Carollo Engineers. Samples
were collected in containers supplied by NLS. Samples were held, preserved, and shipped in a
cooler at approximately 4°C  (39°F) to NLS in accordance with Standard Method 5010B. TOC
analysis followed Standard Method 5310B, UV Absorbance followed Standard Method 5910B.

Microbial Parameters - Water samples were collected by Carollo Engineers.  Samples were
collected in containers supplied by the testing laboratory and sent to NLS (or MWH) in a cooler
at approximately 4°C (39°F). Total coliform densities were reported in MPN/100 mL, following
Standard Method 9221. Total coliform was analyzed as "present" or "absent" using Standard
Method 9223. HPC densities were reported as colony forming units (cfu) per mL, per Standard
Method 9215B.

Total Alkalinity - Water samples were collected by Carollo Engineers. Samples were collected in
containers provided by NLS, and shipped in a cooler at approximately 4°C (39°F). Samples were
collected in accordance with  Standard  Method 3010B. Analysis  was  conducted per Standard
Method 2320B.

Total and Calcium Hardness - Water samples were collected by Carollo Engineers. Samples
were collected in containers  provided by NLS, and  shipped in a  cooler at approximately 4°C
(39°F). Analysis was conducted per EPA Method 200.7.

Total Dissolved Solids - Water samples were collected by Carollo Engineers.  Samples were
collected in containers provided by NLS, and shipped in a cooler at approximately 4°C (39°F).
Analysis was conducted per EPA Method 160.1.

Total Suspended Solids -  Water  samples were collected by Carollo Engineers.  Samples were
collected in containers provided by NLS, and shipped in a cooler at approximately 4°C (39°F).
Analysis was conducted per EPA Method 160.2.

Total Chlorine  - Total chlorine  was measured daily by the  Green Bay Water Utility, using
spectrophotometer methods,  following  HACK Method 8167  (equivalent to  Standard Method
4500-C1 G). Sample results were provided by Green Bay Water Utility. In addition, one chlorine
measurement was taken from the  CMPP feed water following HACK Method 8167, as noted
above, and served as verification of the readings reported by the Green Bay Water Utility.

3.1.7.4  Water Quality Precision and Accuracy

Table 3-7 describes the methodology used in this  ETV for the measurement of precision and
accuracy for  each water quality analysis performed during the test run. Duplicate samples for
analysis are  also shown  for  on-site grab  sample  analysis. The sampling location for each
duplicate grab sample was the filtrate water.

3.1.7.5  Critical Equipment Precision and Accuracy

Flow Meter  -  Water flow rates were verified prior to  the start of the testing and every
2 weeks thereafter. The verification was performed by bucket and stopwatch methods as follows.
The filtrate reservoir on the CMPP skid contains a sight glass. This sight glass contains five-
                                          18

-------
gallon graduations. Following a backwash (when the reservoir is drawn down) a constant flow
rate was established  and the operator calculated the amount of time that it takes 50 gallons of
filtrate water to accumulate in the filtrate reservoir. This data was used to calculate the average
flow rate and was checked against the flow rate as indicated on the data acquisition system of the
CMPP.

Pressure Transmitters - Pressure transmitters are located in both the feed and filtrate side of the
membrane module. These pressure transmitters output an analog signal to the PLC indicating the
gauge pressure measured as a function of time. Both of the pressure transmitters were checked
against an independent measurement in the form a pressure gauge (calibrated to NIST traceable
standards)  mounted  on both  feed  and filtrate  sides of the membrane module.  Pressure  as
indicated on the data acquisition system of the CMPP was checked against the pressure gauge
readings 7 times per week.

In-Line Particle Counters - Factory calibration of particle counters was performed prior to the
start of the study.  However, prior to the study, the instrument calibration was field-verified using
microspheres. Polystyrene sphere suspensions were purchased from EPS Analytical Standards
with concentration certifications provided (Appendix E). The testing was performed with spheres
with diameters of 2,  5 and 10  jim. The particle counter (MetOne) manufacturer's recommended
field  procedures were used. Particle suspension concentrations of 1000  particles per  ml were
used. The 2 jim particle suspension was supplied at a concentration of 1x109 particles per ml and
required field dilution with a micropipette. Particle suspensions for 5 and 10 jim were supplied at
a concentration of 1000 particles  per ml from the manufacturer and  did  not require in-field
dilution.  The  guidelines used to  perform the verification  are listed  below. Results  of this
verification are presented in Chapter 4.

•  The particle counter was  thoroughly cleaned using the  manufacturer's supplied  cleaning
   solution.

•  The particle counter was then flushed using purified reverse osmosis water for a period of at
   least 30 minutes.

•  Microsphere suspensions were continuously stirred gently, to prevent entraining air bubbles
   into the fluid.

•  A peristaltic pump was used to transfer the suspension  to the instrument at the  required
   steady flow rate of 100 ml/min for a period of at least 5 minutes to obtain stable readings.

•  Measurements were electronically logged on a data acquisition system every 2 minutes, and
   the results were  checked against the concentration and  size of microspheres in the stock
   solution.

In-line  Turbidimeters  - In-line  turbidimeters were  calibrated according to manufacturer's
methods with a 20 NTU standard, provided by the manufacturer.  Calibration of turbidimeters
was performed in advance of the verification period and once per week throughout the testing.
In-line turbidity  measurements were checked  daily against a properly calibrated bench-top
instrument.
                                           19

-------
3.1.7.6  Methodology for Use of Blanks

The methodology for using method blanks is summarized in Table 3-8.

Spiked Samples
Spike samples were utilized by NLS for the water quality parameters listed in Table 3-8. Spike
samples were not used for analyses performed in the field. NLS utilized one spike sample per
QA/QC batch. Each batch  consisted  of approximately  10  samples analyzed  for  a single
parameter.

Travel Blanks
In order to assess potential travel related contamination, a total of four travel blanks were to be
submitted to the laboratory during the verification testing.  The laboratory was to analyze half of
the travel blanks for total organic carbon and half for total coliforms. However, these travel
blanks were not performed and are a deficiency of this report.

3.1.7.7  Performance Evaluation Sample

An evaluation  sample was analyzed in accordance with the procedures of NLS. The performance
evaluation  sample  submitted was a 20 NTU turbidity  standard. This performance evaluation
sample was submitted at the start of the verification study by Carollo Engineers to NLS and was
returned with a reported value of 21 NTU.

3.1.7.8  Outline for Duplicate Sampling

Duplicate sampling was performed according to the outline presented in Table 3-7

3.1.7.9  Procedures Used to Ensure Data Correctness

The procedures used to ensure data correctness are  detailed in  Section 3.3  "Calculation of Data
Quality Indicators."

3.1.7.10 Procedures Used to Calculate Indicators of Data Quality

The  procedures  used  to  calculate indicators  of  data quality  are  detailed in Section 3.3
"Calculation of Data Quality Indicators."

3.1.7.11 Outline for Data Reporting

Reports were prepared by Carollo Engineers,  P.C. Data was reported in a draft and final ETV
report, submitted by Carollo Engineers, P.C. Status reports were not  submitted due to the short
duration of the tests.

3.1.7.12 Corrective Action Plan

The corrective action plan for  each of the tested water  quality parameters is summarized in
Table 3-9.
                                           20

-------
3.1.8   Task 8: Operational Conditions and Maintenance

The objective of this task was to maintain appropriate operational conditions and maintenance
throughout the verification testing.

3.1.8.1  Work Plan

Technical specifications for the Polymem module were provided for this testing  and include
normal operating ranges for temperature, pressure, flow, chemical compatibility, instructions for
conditioning, storage, integrity testing, fiber repair, and chemical cleaning. Carollo evaluated the
instructions and procedures in the membrane  O&M Manual for their applicability during the
verification test. The CMPP O&M Manual is included in Appendix B. It contains procedures for
filtration, backwashing, chemical cleaning, and routine maintenance. Troubleshooting guidance
is also included.

3.2    Calculation of Operating Parameters

3.2.1   Flux

Membrane flux was calculated as follows:

       ^ifiltrate
Flux =•
        Area
Where:
Qfiitrate = filtrate flow rate
Area = total active membrane surface area

3.2.2   Specific Flux

Specific Flux was calculated as follows:
                Flux
Specific Flux =
               IMP
Where:
Flux = the result of the flux calculation as shown above
IMP = transmembrane pressure

3.2.3   Normalized Specific Flux

Normalized Specific Flux was calculated as follows:
                           Stte cificFlux
Normalized Specific Flux =  O0239.(r_20)


Where:
Specific Flux = the result of the specific flux calculation as shown above
T= Temperature in Celsius
                                           21

-------
3.2.4   Feedwater System Recovery

Feed Water Recovery was calculated as follows:

                ' z?rj/
Recovery =
1-
 BW
v
 total
*100%
Where:
Vtotai = total volume of filtrate produced in a production cycle based on average flow rate and a
production cycle duration of 50 minutes.
VBW = total volume of filtrate used during a production backwash cycle

3.3    Calculation of Data Quality Indicators

Data quality parameters as specified  in the ETV protocol include representativeness, statistical
uncertainty,  accuracy, and precision. This section details how each of these  parameters was
considered throughout this testing.

3.3.1    Representativeness

As specified by the ETV Protocol (EPA/NSF  1999), representativeness of operational parameters
entails collecting a sufficient quantity of data during operation to be able to  detect a change in
operational parameters. As specified, detecting a plus  or minus 10 percent change in operating
parameter is  sufficient  for  proper  QA/QC.  Operational  parameters  including flow rate,
membrane feed, and filtrate pressures were recorded a minimum of five days a week, which NSF
specifies as sufficient for tracking changes in operational conditions that exceed this 10 percent
range.

In addition to ensuring representativeness of  operational parameters, representativeness of water
quality  samples was ensured by  executing consistent sample collection procedures. These
procedures considered sample location, timing of sample collection, sample procedures, sample
preservation, sample packaging, and sample shipping as detailed below.

Sample  Locations - The water quality  monitoring matrix was presented in Table 3-5, which
presents the water quality monitoring plan for feed, filtrate, and backwash  streams of the CMPP.
Further  guidance on  sampling locations is included in the O&M Manual/Standard Operating
Procedures for the CMPP presented in Appendix B.

Timing  of Sample Collection  - Feed  water quality sampling was performed within  one hour of
filtrate water quality sampling to ensure that the filtrate water sample was representative of the
membrane feed water quality. Filtrate water quality was relatively consistent throughout  the
duration of the production cycle. However, it is not unusual for turbidity or particle counts to be
slightly  higher at the beginning of a production cycle. Since this represents the worst case water
quality,  filtrate  water samples were collected within the first 15 minutes of a production cycle.
The PLC on the CMPP  has a timer, which  clearly indicated the  duration  of  a  particular
production cycle.
                                           22

-------
The backwash procedure  consisted of  a  three-step procedure.  Step 1  included  hydraulic
backwashing, air scour, and chlorine addition followed by a short rest period of approximately
five seconds. Step 2 repeated Step 1. Step 3 of the backwash procedure was only hydraulic; air
scour and chlorine addition were not utilized. Backwash samples were collected at the beginning
of the second backwash cycle. The backwash sample tap was opened throughout the duration of
the backwash procedure to ensure that a representative sample was collected.

Sampling Procedures, Preservation,  Packaging, and Transport - Prior to the collection of each
individual water quality sample, the  sample tap was allowed to run a minimum of 30 seconds in
order to purge the sample tap and sample line of stagnant water. Samples were then collected.
Additional considerations and procedures for individual water quality parameters are included
below:

pH - pH  samples were collected at the sample tap in polypropylene beakers and immediately
tested for pH. The temperature at which the pH reading is made was also recorded.

Temperature - In addition to temperature transmitter readings (recorded  continuously on  the
PLC), temperature gauge readings were manually recorded daily. Temperature was also recorded
while measuring pH.  Special preparation or sampling procedures were not necessary for this
measurement.

Turbidity  (Bench-top) - The method for collecting bench-top turbidity samples followed  the
procedure recommended in the testing protocol developed by the NSF. The procedure was as
follows:


•   The slow steady stream was run from the sample tap.

•   A dedicated sample cell was triple rinsed with the sample.

•   The sample was allowed to flow down the side of the cell to minimize bubble entrainment.

•   The sample cell was wiped clean.

•   The sample cell was immediately inserted into the turbidimeter.

•   The measured turbidity was recorded upon reading stabilization.
Alkalinity - Samples were collected in a polyethylene or borosilicate bottle provided by the
analytical laboratory. The sample was closed tightly and immediately placed into the sample
cooler for transport to the analytical laboratory. Sample agitation and prolonged exposure to the
air was avoided.

Total  Hardness  -  Procedures  for sampling  total hardness  followed  alkalinity  sampling
procedures.

Calcium Hardness -  Samples were collected by the procedures for alkalinity and total hardness.

Total Dissolved Solids - Resistant glass or plastic sample bottles were used as provided by the
analytical laboratory. Samples were  collected and immediately placed into the sample cooler for
                                           23

-------
transport to the  analytical laboratory. Analysis began as soon possible  as specified by  EPA
Method 160.1.

Total Suspended Solids - Sampling for total suspended solids followed the same procedure as
specified by total dissolved solids.

Total Coliform and HPC -  Sample containers were provided by the  analytical laboratory.
Aseptic sampling techniques were used as follows:


•   Sample bottles were kept closed until they were filled.

•   Sample taps were  removed, allowed to  soak in a chlorine solution for a minimum of two
    minutes, rinsed, and reconnected to the sample valve. Water was then allowed to run through
    the tap for a minimum of two minutes. The sample tap was flamed prior to sampling.

•   The cap of the sample container was removed without touching the surface of the cap or neck
    of the bottle.

•   The sample container was filled without rinsing and the cap was replaced immediately.

•   Samples were refrigerated immediately after collection and transported to the laboratory in
    coolers with frozen blue ice.

•   Samples were refrigerated upon receipt at the laboratory  and analyzed within holding times
    specified in their standard method.

3. 3.2    Statistical Uncertainty

For data sets of eight or  more, statistical  uncertainty was calculated for  grab sample analyses
including TOC, turbidity, HPC, UV254, and TSS. This was done by calculating the 95 percent
confidence interval in the following manner:

                          -         ( e
95%  Confidence Interval  = X+t   ,   J •
Where:
X = sample mean
t = student t-test with n-1 degrees of freedom
S = sample standard deviation
n = number of independent measurements

3.3.3    Accuracy

Accuracy was quantified as the percent recovery of a parameter in a sample to which a known
quantity of that parameter was added. For this testing an example of accuracy determination in
the ETV was the analysis  of a turbidity proficiency sample  in comparison  of the measured
turbidity of the known level of turbidity of the sample.

Accuracy  =  Percent Recovery =  100 * [(*_„ - x_e d) + Xknown ]
                                          24

-------
Where:
x-known = known concentration of analyte added to the sample
        = measured concentration of parameter
3.3.4   Precision

Precision refers to the degree of mutual agreement among individual measurements that provides
an estimate of random error. The standard  deviation and relative standard deviation recorded
from sample analysis were reported as a means to quantify sample precision. The percent relative
standard deviation was calculated in the following manner. The standard deviation is as follows:
 Precision Deviation  =
                                 -|l/2
                          n-\
Where:
x = sample mean
X[ = /'th data point in the data set
n = number of data points in the set

As specified  in the ETV protocol provided by the NSF, the percent relative standard deviation
for drinking  water samples must be less than  30 percent or  acceptable precision under the
verification testing program.

Relative Percent Deviation =   1 -  2  * 100%
                                x

Where:
x = sample mean
x\ = first data point of the set of two duplicate data points
X2 = second data point of the set of two duplicate data points

3.4    Safety Measures

Membrane - The membrane was inspected for visual damage prior to installation, and operated
within the manufacturer's pressure and flow ranges.

Electrical - The electrical work for the CMPP was performed prior to the start of this study by a
licensed electrical contractor in accordance with local codes.

Chemicals - Non-compatible chemicals (acids/bases/chlorine/preservation solutions) were stored
separately from one another at the pilot site. Material Safety Data Sheets were stored on site and
are provided  in Appendix F.
                                           25

-------
3.5    Testing Schedule
Field operations specific to the PSTP began February 28, 2002. Testing specific to the PSTP
began March 11, 2002. The verification testing was terminated on April 26, 2002 after 46 days
of operation. Field activities related to this testing were finalized May 3, 2002.
The following key events took place during the test run.

•  Conduct initial chemical clean to establish baseline conditions - Day 0;
•  Conduct integrity test - Day 0;
•  Start continuous membrane filtration pilot testing including sample collection, and hydraulic
   performance monitoring as described in Chapter 4 of this report - Day 1;
•  First chemical Clean - Day 12;
•  End filtration operations - Day 46;
•  Second chemical cleaning - Day 51;  and
•  Conduct integrity testing - Day 54.
                                           26

-------
                                       Chapter 4
                                 Results and Discussion

4.1    Task 1: Characterization of Membrane Flux and Recovery

The operating conditions for the Polymem UF120 S2 module are provided in Table 4-1. These
test conditions were established based on previous pilot study optimization results conducted
from May 2001 to March 2002. The membrane system ran at a constant normalized flux of 40
L/hr-m2 (24 gfd) for the first 12.5 days of operation during the ETV test (Run 1). At the end of
Run  1, the membrane  was chemically cleaned to restore flux and  to reduce  the  required
transmembrane pressure. The remaining run time was operated at a constant specific flux of 30
L/hr-m2 (18 gfd) (Run 2).

The backwash interval, or production cycle, was set at  50 minutes of operational run time
throughout the testing and was followed by a total production backwash time ranging from 60-
120 seconds. The backwash cycle  duration was  adjusted  throughout the testing to maintain a
system recovery of at least 90 percent. Backwash water was chlorinated at  a target dose of 5
mg/L for the entire duration of the testing and was verified with daily sampling.

Figure 4-1 shows  a time series plot of TMP, normalized flux, normalized  specific flux, and
system recovery  data throughout the test. Operational flux and temperature data are presented as
a time  series plot in Figure 4-2. Table 4-2 summarizes operational data collected throughout the
testing. System recovery was calculated  daily based on average flow rates and total backwash
volumes as measured  on a daily basis by plant operators. Recovery ranged from 89-96 percent
throughout the 46-days of testing. The figures show a plant shut down on 4/18/02 resulting from
a power outage  at the CMPP. When the plant was restarted, the default operational setpoints
(different from the target operational  data)  caused discontinuity in the data. When these set
points were changed back to the target values, system operations returned to normal.

At the start of Run 1  (clean membrane), the  TMP began at approximately  0.34  bar  (5 psi).
During Run 1, there was a nearly linear rise in TMP at a rate of approximately  0.087 bar per day
(1.5 psi/day). After a chemical clean, the TMP was restored to approximately  0.45 bar (6.5 psi)
and decreased to approximately 0.41 bar (6 psi) within one hour of start up. During Run 2, the
TMP remained stable near 6 psi for approximately 24 hours. The remainder of Run 2 produced a
nearly linear rise in TMP of approximately 0.039 bar per day (0.57 psi/day).

Normalized specific flux at the start of Run 1 was approximately 118 L/hr-m2-b (4.79 gfd/psi).
Due to fouling, normalized specific flux decreased to approximately 28 L/hr-m2-b (1.1  gfd/psi)
by the  end of Run 1 with a TMP of 1.43 bar (21 psi). Chemical cleaning restored the normalized
specific flux to approximately 74 L/hr-m -b (3.0 gfd/psi). Run 2 lasted nearly  33 additional days
without another chemical clean. At the time the testing was terminated (at the end of Run 2), the
normalized specific flux was approximately 17 L/hr-m2-b (0.70 gfd/psi) with a TMP of 1.7 bar
(25 psi).  The improvement in specific flux decline trends during Run 2 is likely due to the lower
target normalized flux. It should be  noted that the 25 percent decrease in normalized flux led to a
260 percent increase in run time before a required chemical cleaning (12.5 vs. 32.7 days).
                                          27

-------
Following a review of the data, the manufacturer suggested that the following changes may have
adversely impact the specific flux decline rate:

•  Low Air Scour Flow Rate: Polymem suggested that flow rates are typically 5 m3/hr  (2.9
   cfm) at standard temperature and pressure  (STP). This information is not included in the
   O&M Manual provided by Polymem.

•  The Run 1 flux rate was selected based on a recovery of 90 percent; however, the membrane
   was operated at a median recovery of 94 percent.

4.2    Task 2: Evaluation of Cleaning Efficacy

A total of three chemical cleanings were performed for this testing as a "proof of concept" effort:
1) Prior to beginning the verification test. This initial chemical cleaning was performed on 3/5/02
instead of the first day of the testing (3/11/02) because CMPP  equipment verification and the
testing of membrane baseline flux conditions required a few days of work prior to the official
ETV starting date of 3/11/02; 2) following Run  1 when the transmembrane pressure reached 1.4
bar (21 psi). Although the terminal transmembrane pressure of the membrane module was 29 psi,
during Run #1 of the ETV testing, the rate of TMP rise  was such that the  terminal pressure
would have been exceeded  during the weekend when no operator would  have been present to
stop the unit. This run was ended somewhat prematurely to avoid possible  fiber and module
damage; and 3) following Run 2 at the end of  ETV testing, when the transmembrane pressure
reached 1.7 bar (25 psi). Tables 4-3 and 4-4  summarize collected chemical cleaning water quality
and hydraulic data, respectively. It should  be noted that, for  Chemical Cleaning #3,  the TDS
samples were analyzed exceeding the EPA sampling holding times.  Table 4-5 summaries the
collected chemical cleaning efficacy data.

The recovery of specific normalized flux  for Chemical Cleaning #s 2 and  3  was 62 and 73
percent, respectively. Chemical cleaning conditions (chlorine concentration, pH) were selected
from  the manufacturer's recommended procedures. Cleaning # 2 was performed at ambient
water temperature,  [14-18.6°C  (57-65.5°F)],  pH =  12.2,  and  an  average  total  chlorine
concentration of 164 mg/L, for 8 hours.

Because recovery of specific flux was low,  Cleaning # 3 was performed with  a similar cleaning
solution but at elevated solution temperature [22 - 31°C (72-88°F)], for  an  extended soaking
period. Despite these changes the specific flux recovery was marginal (73 percent).

Following a review of the data, the manufacturer suggested that, due to the  observed calcium
hardness (average of 88 mg/L during ETV  test), some CaCOs fouling could have occurred and
an acid clean would be necessary to restore the specific flux. Additionally, recirculation of the
cleaning solution during cleaning has been shown to produce higher specific flux recoveries than
those  observed on other source waters (Hugaboom, et. al, 2001). Further optimization would be
required to improve these recoveries, however,  this was not the goal of the study. Despite the
low specific flux recoveries, the goal  of  providing "proof of concept"  was  achieved. Ideal
conditions  would likely produce much greater  cleaning efficiency. Therefore, an accurate
account of the usable membrane life cannot be estimated. The manufacturer's recommended
cleaning procedure is included in Appendix  A.
                                          28

-------
4.3    Task 3: Evaluation of Finished Water Quality

Result summary sheets for off-site analyses are provided in Appendix G.

4.3.1    Turbidity, Particle Counts, and Particle Removal

Figure 4-3 presents the on-line and bench-top turbidity profiles recorded throughout the testing.
The figure  shows data for feed,  filtrate, and backwash water. The bench-top turbidity analysis
results are summarized in Table  4-6 including feed, filtrate, and backwash water analyses. On-
line turbidity data are summarized in Table 4-7.  Overall, there  was close mutual agreement
between bench-top and online measurements. As shown in Table 4-7, feed water turbidity ranged
from  0.2 to over 19 NTU with  an average of 1.3 NTU. Bench-top backwash water turbidity
averaged about 11 NTU. Bench-top filtrate turbidity was typically  below 0.05 NTU. Online
turbidity data showed two anomalous filtrate readings dated 3-19-02 and 3-22-02 with readings
of 0.002 and 0.45 NTU, respectively. These data were not included in statistical analyses. Due to
the size of the database for online turbidity and particle counts, this data is not included in the
appendix. However, this data is on file with the NSF.

Table  4-7   also shows  online  feed  and  filtrate  particle  count data.  Total feed  particle
concentrations (>2 jim) averaged about 4,300 particles/ml. Filtrate particle counts averaged 4
particles/ml. The 90th percentile  for  feed and filtrate  total  particle  counts  (>2  #/ml) was
approximately 9,911  and  2 particles/ml, respectively. Average particle log removals of 4.2, 4.1,
4.1, 3.4, 3.3, 2.9, and 2.2 were achieved for particle size ranges of >2 urn, 2-3, 3-5, 5-7, 7-10, 10-
15, and >15 um, respectively. The membrane system removed 3.1 logs of total particles  90
percent of  the  time. It should be noted that the particle count  instruments collected sample
volumes of 100 ml per each data  point. If no particles were detected in that sample volume in the
filtrate, the filtrate particle count  data was recorded as 0.00 particle/ml (below the detection limit
for the instrument of 1 particle).  Since these data were recorded as zero values, log removal data
could not be calculated for these data points and were not included in  the statistical analyses.
Because  the membrane  system produced  relatively consistent  filtrate  particle  counts,  log
removals increased during periods when feed water particle counts were higher and decreased
during periods when feed water particle counts were lower.

Figures 4-4 through 4-10 present the time series particle count profile data (>2 jim, 2-3 jim, 3-5
|im, 5-7 um, 7-10 jim, 10-15 jim, and >15 jim) that was collected throughout the testing.  Figures
4-11 through 4-17 show log removal profiles for each particle size. The  presented data was
collected at 5-minute intervals and includes data collected during backwashes. Evaluation of the
time series data showed relatively higher particle counts during, and immediately following, a
backwash. As  a result, particle removals were decreased during these times. These  consistently
brief occurrences caused log removal data to decrease or even become negative for short periods.
The reasons for this are described below.

During backwash cycles,  the feed pump was turned off thereby halting flow to the feed particle
counter. Because the particle counter output depends on flow rate, this decrease in flow to the
particle counter resulted in lower particle counts.
                                           29

-------
The filtrate particle counter data was affected during backwash cycles as well. The surge in flow
during backwash cycles can cause  hydraulic and air bubble turbulence in the particle counter
feed weir. The stirring up  accumulated particles within the weir itself and introduction of air
bubbles lead to higher particle counts.  Furthermore, the piping used for backwashes  was fed
from the bottom of the filtrate tank that may have accumulated particles and air bubbles (caused
by increases in temperature and subsequent degassing) during the stagnant 50-minutes of a
production cycle.

The  figures  also  show higher filtrate particle  count  data  immediately  following  chemical
cleanings  for  the  same  reasons  are  described above.  After  system  stabilization (within
approximately 4 hours), filtrate particle counts reached typical low-level readings.

For evaluation  purposes the data for turbidity, particle counts, and log removal were plotted as
frequency distribution curves. Frequency distribution curves are shown in Figures 4-18 through
4-28. The 90th percentile for feed and filtrate turbidity was 3.5 and 0.035 NTU, respectively. The
90th percentile  for feed and filtrate total particle counts (>2 #/ml) was approximately 9,911 and 2
particles/ml,  respectively.  It should be noted that filtrate  particle count  average (as shown in
Table 4-7) was higher than the 90th percentile. This is due to the fact that statistical averaging is
largely effected by higher values, especially when the majority of the data  is near zero as was the
case in the filtrate particle count data set.

Figures 4-29 and 4-30 show a sensitivity analysis performed on particle count data  collected
from one 24-hour period performed to  determine the potential effects of backwash events on
calculated log  removals. Data from March 14,  2002 was chosen for this analysis due to the
clusters of relatively lower log removal data, thereby representing a worse case scenario. Figure
4-29 presents the "raw" time  series of feed and filtrate particle  count data (>2  jim) that was
collected throughout the 24-hour period and includes markers indicating likely backwash events.
The presented particle count data was  collected at 5-minute intervals and includes data collected
during backwashes. Figure 4-30 presents the time series of feed and filtrate particle count data
(>2 |im) collected throughout the same 24-hour period, with the particle count data removed that
were likely collected during backwash events. It should be noted that Figures 4-29 and 4-30 are
plotted on a log scale. As such, differences in filtrate particle counts are more apparent than with
the feed count  data. As shown in Figure 4-30, filtrate particle counts did not exceed  10 #/ml
when the filtrate particle counts recorded during the backwash events were excluded from the
data set.

Figure 4-31  shows the log removal frequency distribution curves calculated for the raw data set
(data including backwash events) and  for the data excluding particle count data likely  collected
during backwash events. Log removals (>2  jim) for the raw data set and for the data set without
particle counts likely collected during backwash events were 3.2 or greater 90 percent of the time
and 3.6 or greater 90 percent of the time, respectively.

4.3.2    Microbial Removal

The removal of naturally occurring bacteria was also monitored throughout the  ETV study. A
summary of this data is shown in Table 4-8. Total coliform bacteria were analyzed through two
means:  1) through a "presence/absence" (P/A) test and 2) through total  coliform enumeration.
                                           30

-------
One of the P/A test results, from the sample collected on 3/18/02, confirmed the presence of total
coliforms in the feed water. However, total coliforms were not detected in the corresponding
filtrate sample.  Total coliform bacteria was detected in one P/A filtrate sample on 4/12/02 and
one P/A backwash  sample on  4/22/02.  Total  coliform  enumeration  results  showed  feed
concentrations ranging from 
-------
However, each filtrate sample with detectable TSS was reported as greater than its corresponding
feed water sample. These results are likely due to the fact that feed and filtrate samples were so
near the detection limit of the analysis. Due to the length of time the equipment was in use prior
to the ETV testing, it is also possible that material had built up in the portion of sample piping
permanently fixed to the CMPP skid. Although the sample ports were allowed to flush prior to
sample collection, accumulated material may have sloughed off during some of the sampling
periods. Specifically, TSS data collected on April 5, 2002 showed  strong evidence  of a sample
mix-up. Feed,  filtrate, and  backwash TSS values were reported as  <1,  22,  and 1  mg/L,
respectively. This data was excluded from statistical analysis.

Table 4-10 presents  the mass  balance conducted  on the TSS results obtained  throughout the
testing. In the cases where the  measured TSS was below the detection limit (<1 mg/L), half of
the  detection limit was used for the  calculations. Six of the 36 calculated results showed positive
correlation (<50% relative percent deviation) between calculated  and measured backwash water
TSS. These relatively poor correlations may in part be due to the feed water TSS being close to
the  method detection limit.  In addition, backwash water samples were collected during the
second cycle  of the backwash.  It is likely that the majority of TSS were flushed out during the
first cycle of the backwash. Therefore, the reported backwash water TSS concentration may not
accurately represent the entire backwash water volume consumed.

4.4     Task 4:  Reporting Membrane Pore Size

The nominal pore size of the UF120  S2  membrane module  used  in this ETV study was
determined through the use of multiple methods; however, the manufacturer has  not determined
a pore size  distribution.  Methods  used  in determining nominal  pore  size  include scanning
electron  microscopy (SEM), flow  porometry, and particle retention  testing.  The first two
methods are described here.

The membrane  is an asymmetric hollow fiber with dense layer  containing small pores on the
inside and outside, and a less dense structure containing larger pores. SEM images were used at
multiple resolutions to estimate the size of the pores in the membrane.  Based on these images,
the  estimated  nominal pore diameter is less than 100 nanometers (nm).

Flow porometry (Lee, et  al., 1997; Hernandez, et al.,  1999; Mietton  and Courtois, 1997) was
used to determine the mean size of the pores in the membrane  (including the less dense middle
layer). The mean pore diameter used in this procedure was determined to be less than 50 nm. The
above data are taken from a letter supplied by the manufacturer that is included in Appendix H.
This data is provided for informational purposes only and the results were not verified during the
ETV testing.

4.5     Task 5:  Membrane Integrity Testing

A total of two membrane integrity tests were performed  for this testing: 1) prior to the start of the
test and 2) following the final chemical cleaning.  The first integrity test was performed with an
applied air pressure of 10 psi. Measurements of air pressure  were taken  every 15 seconds for a
period of 2 minutes. Over the  course of the 2 minutes there was zero loss in pressure and no
visual indicators  of  a  loss  of membrane integrity (no  bubbles were  detected).  The  second
                                          32

-------
integrity test was again performed with an applied pressure of 10 psi. Measurements were taken
every 15 seconds for a period of 2 minutes. The test was repeated 4 times yielding an average
pressure drop of 0.35  psi/min. This was within the allowable pressure drop as specified by the
manufacturer (<0.36 psi/min). However,  during the visual  inspection a small amount  of air
bubbles were noted in the vertical, transparent section  of the CMPP feed water piping. This
suggests that a fiber (or fibers) may have broken.

In addition to the visual air bubble test, in-line particle counting and turbidity data was used as an
indirect method of monitoring membrane integrity. Increases in particle counts or turbidity in the
filtrate serve  as  indicators of potential fiber breakage. During the ETV testing, there were no
significant increases in filtrate particle counts or turbidity that would have indicated a loss in
membrane integrity. However, following ETV testing (but prior to any further operation of the
membrane module) on 6/5/02, the membrane module filtrate end cap was removed to further
investigate the bubbles noted during the final integrity test. This investigation followed the
integrity test/repair procedures outlined in the Polymem  UF120 S2 O&M Manual. One broken
fiber was  identified and repaired. One subsequent pressure decay test, performed as  described
above, yielded a zero  loss in pressure and no visual indicators of a loss of membrane integrity
(no bubbles were detected).

4.6    Task 6: Data Management

4.6.1    Data Recording

Water quality and hydraulic data was maintained on  site. Critical  process parameters including
pressures, temperature, turbidity,  flow rates, and chemical doses were automatically stored on the
membrane skid  mounted  PLC CPU. This data was also backed up  on a separate personal
computer. This data was recorded at a  maximum of 10-minute intervals throughout the ETV
testing.  Feed and filtrate count data  was recorded at 5-minute intervals for the following size
ranges: 2-3 |im, 3-5  jim, 5-7 jim,  7-10 jim, 10-15 jim,  and >15 jim.

Data were manually recorded in tabular format for water quality parameters received from NLS
including  alkalinity, total  hardness, calcium hardness, TDS,  TSS, TOC, and UVA. Likewise,
results from biological samples were recorded in a tabular format.

Onsite bench-top analyses were recorded in specifically  designed daily log sheets. Other details
of the  ETV testing  such as field  analyses, maintenance  activities, water  type, backwash
procedures, operating  cycles, names of visitors, problems, etc. were logged by the operator in a
bound notebook. Additionally, QA/QC information including manual  checks of pressure, flow
and temperature transmitters, results of on-site analyses, instrumentation flow rates, and  other
pertinent information  were logged in the same manner. Chain  of custody  records for off-site
laboratory analysis, on-site  calibration,  and verification of  on-line instrumentation were also
recorded.  Copies of the laboratory notebooks and data log sheets are included in Appendix D.

4.6.2    Data Entry, Validation, and Reduction

Data  were entered  from  data sheets into similarly  designed data entry forms in electronic
spreadsheet format. Following data entry, the spreadsheets  were  printed and checked  against
                                           33

-------
handwritten datasheets. All corrections were noted on the hard copies and were then corrected on
the electronic copy. The hard copies of the electronic data are included in Appendix D.

4.7    Task 7: Quality Assurance/Quality Control (QA/QC)

QA/QC results and  chain  of custody records  reported by NLS and MWH are included in
Appendix I.

4.7.1    Data Correctness

There are five indicators of data correctness including representativeness, statistical uncertainty,
completeness, accuracy, and precision. This section includes the summary of analyses conducted
to ensure correctness of the data. The methods used for data analysis are outlined in Chapter 3.
Sampling and testing protocols were conducted per the requirements of the ETV Protocol for
Equipment  Verification  Testing  for  Physical Removal   of Microbial  and Particulate
Contaminants (May 1999).

4.7.1.1  Representativeness

Representativeness of the data was ensured through strict adherence to sampling and testing
methods outlined in Chapter 3. In addition, sampling efforts were coordinated so that  sampling
was conducted at a consistent time and location throughout the testing.

4.7.1.2  Statistical Uncertainty

For data sets of eight or more, statistical  uncertainty was calculated  for grab sample analysis
including TOC, turbidity,  HPC,  UV254,  and  TSS. In  addition,  statistical  uncertainty was
calculated for online measurements such as flux, flow rate, temperature, system recovery,  etc.
Statistical uncertainty was  evaluated  by calculating the  95  percent  confidence interval. The
results were presented in the previous sections.

4.7.1.3  Completeness

Data completeness refers to the amount of data collected during the ETV testing compared to the
amount of data proposed in  the  PSTP. Data completeness  was determined  for  onsite water
quality  measurement, laboratory water  quality  measurement, and  operational data recording.
Completeness tables can  be found in  Appendix J. Nearly 100 percent of the  parameters were
complete. However,  travel  blank samples were not collected during this testing.  CMPP feed
water chlorine residual was  scheduled for testing once per week to verify the daily measurements
collected as a part of the GBWUFP. Only one CMPP sample was collected during the first day of
testing to verify the readings recorded by the utility. It  should also  be noted that the backwash
water total coliform sample  collected on 4/17/02 was not analyzed due to NLS error.

4.7.1.4  Accuracy

Accuracy was quantified  as the percent recovery of a parameter in  a sample to which a known
quantity of that parameter was added. Accuracy determination in this ETV testing was performed
by the analysis  of a turbidity proficiency sample and onsite bench-top turbidimeter  standards. A
                                           34

-------
comparison  was made between the measured turbidity and the known turbidity level  of the
standard. Bench-top turbidity accuracy ranged from 84-113 percent with an  average  of 98
percent. The accuracy of the turbidity proficiency sample analyzed by NLS from a sample
collected on 3/12/02 was 95  percent.  This data is  included in Appendix I. Accuracy was also
ensured by calibration of bench-top and online turbidimeters as well as particle counter and flow
meter verification as outlined in Chapter 3.  Results  for the online particle counter calibration
verification  are  shown  in  Tables  4-11 and 4-12  for the feed and filtrate particle counters,
respectively. Standards for 2,  5 and 10 um particle sizes were used for the verification. Each
standard solution had a total  particle  concentration of  1000 particles/ml. The data show good
correlation between the tested standard and particle counter output. A minor amount of particles
were detected outside of the range of each standard. This is likely due to the standard solutions,
which contain a distribution of particle sizes rather than a pure  concentration of one particle size.
Accuracy data is presented in Appendix D.

4.7.1.5  Precision and Relative Percent Deviation

Duplicate  measurements were taken throughout the ETV testing as outlined in Chapter 3. Each
duplicate measurement was analyzed to determine the consistency of  sampling and analysis
using relative percent deviation. The relative percent deviations averaged  within 9 percent for
onsite bench-top turbidity measurements. General water quality parameters (hardness, calcium
hardness,  alkalinity, algae, TDS, TOC, and UVA) sampled for the waters (feed, filtrate, and
backwash) were as high as 40 percent with an average of 9 percent relative deviation. TSS
analyses were as high  as  189 percent  (backwash) with  an average of 50 percent relative
deviation.  HPC analyses for the feed and filtrate were as high as 150 percent with an average of
30 percent relative deviation. These data are included in Appendix G.

Relative percent deviation is greatly  effected  when  analytical measurements are close  to the
lower detection limit. Specifically filtrate HPC  and  TSS were largely effected in this manner. 33
out of 38 filtrate HPC results were at or below the detection limit. 32 of 37 TSS results were at or
below the detection  limit.

4.8    Operational Conditions and Maintenance

4.8.1    Overall Operation and Maintenance

There were  no  major reoccurring problems  experienced  during  this ETV  testing program.
However,  plant  operators  noticed  a few occasions  of air bubble entrapment  in the particle
counter feed lines  signaled  by a decrease  in particle counter  flow.  Influent  waters were
approximately 4°C (39°F) while the CMPP trailer was kept near 20°C (68°F). This temperature
difference could have caused degassing of dissolved species and subsequent entrapment of air
bubbles; however, air bubble  introduction during backwash  flow surges was the likely cause of
entrapment of air bubbles. Particle counter feed flow rates were verified  daily and corrective
action was taken if flow restrictions were found. A small tubing brush was used to remove the air
bubbles.
                                           35

-------
4.8.2    System Chemical Consumption

Chemical consumption was calculated for the system based on data collected throughout the
testing. Daily measurements of total backwash volume and chlorine dose were used to estimate
daily  consumption  of chlorine  for production backwashes.  Chemical  cleaning  chemical
consumption was based on the input parameters for sodium hydroxide and chlorine.  Chemical
dosing was verified through onsite pH and chlorine measurements taken immediately following
chemical injection.

Table 4-13 provides a summary of chemical consumption during the ETV test. For Run 1, a total
of 0.61 pounds of chlorine (as NaOCl) were consumed for production backwashes. The chemical
cleaning following Run 1 consumed 0.06 pounds of chlorine (as NaOCl) and 2.04 pounds of
NaOH (as NaOH). For Run 2,  a total of 1.06 pounds of chlorine (as NaOCl) were consumed for
production backwashes. The chemical cleaning following Run 2 used 0.06 pounds of chlorine (as
NaOCl) and 1.47 pounds of NaOH (as NaOH).

4.8.3    Review of Operations Manual

After reviewing the O&M Manual provided by Polymem, the following points were noted:

•  The O&M Manual should be reviewed for grammar and word usage;

•  The air scour flow rate during backwash should be specified;
•  Chemical  cleaning information should contain the recommendation that heated water be used
   if cleaning is  inefficient at ambient temperatures;

•  Pages 4 and 8 recommend procedures for integrity testing and for identifying a compromised
   fiber. The O&M  Manual  should be clarified to explicitly state the purpose of these two
   procedures.

4.8.4    Equipment Deficiencies Experienced During the ETV Program

4.8.4.1  Online Particle Counting

Throughout the testing there were no mechanical problems with the particle counter equipment.
However, there was one period of approximately three days from 3/29/02 3:48 to  3/31/02 17:10
in which collected particle count data was unavailable for retrieval from the data logging
software. The operators experienced similar problems prior to the start of the ETV testing. The
data logging software technical support staff were unsuccessful in diagnosing this problem. In
addition, plant operators noticed a few occasions of air bubble entrapment in the particle counter
feed lines as described in section 4.8.1.

4.8.4.2  Membrane Equipment and Online Turbidimeters

One CMPP shut down was experienced during the  testing during the night of 4/18/02 due to
lightning. The online turbidimeters experienced damage resulting from an electrical overload and
were out of commission until replacement parts were available during the first week of May. The
CMPP shut down lasted  approximately 10  hours. Operational staff successfully restarted the
CMPP with no concerns outside of the online turbidimeters.

                                          36

-------
                                      Chapter 5
                                      References

EPA/NSF International. ETVProtocol for Equipment Verification Testing for Physical Removal
of Microbiological and Particulate Contaminants, May 1999.

U.S. Environmental Protection Agency. Methods for Chemical Analysis of Water and Wastes,
EPA 600/479-020, March 1979.

U.S. Environmental Protection Agency. Methods for Chemical Analysis of Water and Wastes,
EPA 600/479-020, Revision March 1983.

Hugaboom, D. A., Roquebert, V.R., Crozes, G.F., Seacord, T.F., and Mahady, J.J., Retrofit of
Existing  Granular Media Filters With Ultrafiltration  Membrane Proceedings,  AWWA Water
Quality Technology Conference, 2001, Nashville, TN.

Hernandez, J.I., Calco, P., Pradanos, P., and Palacio, L. A Multidisciplinary Approach Towards
Pore Size Distributions of Microporous and Mesoporous Membranes In Surface Chemistry and
Electroschemistry of Membranes. Ed. T.S. Sorensen, 1999.

Jacangelo, J.G., S.S. Adham, and J-M. Laine. Mechanism of Cryptosporidium, Giardia, and
MS2 Virus Removal by MF and UF, Journal AWWA, 87(9)107-121, 1995.

Lee, Y.,  Jeong, J., Young, I.J.,  and  Lee,  W.H.  Modified Liquid Displacement  Method  for
Determination of Pore Size Distribution in Porous Membranes, Journal of Membrane Science,
130(1+2) 149-156, 1997.

Mietton Puechot, M. and Courtois, T. (1997) The Performance Limits of Bi-Liquid Porometry in
Determining the Distribution of Membrane Pore Diameters, Euromembrane '97, June 2-3, 1997,
held at University ofTwente (Netherlands).

Standard Methods for the Examination of Water and Wastewater, 1992, 18*  Edition, APHA,
AWWA,  and WEF, Edited by L.S. Clesceri, A.E. Greenberg.

Standard Methods for the Examination of Water and Wastewater, 1998, 20th Edition, APHA,
AWWA,  and WEF, Edited by L.S. Clesceri, A.E. Greenberg.
                                          37

-------
                                      Chapter 6
                                  Vendor Comments

Polymem submitted the following comments on the DRAFT report to the NSF.  These comments
were not included in the body of the text.

1. The pilot plant [CMPP] was not equipped with a dedicated air compressor for providing air for
production backwashes.  As a  result, the airflow rate dropped below the specified flow rate
during backwashing.
2.  Concerning airflow rate during backwashing, the manufacturer typically recommends a
constant flow rate of 3  to 5 m3/h  at standard  temperature and pressure. Due to the limited
capacity of the compressor, this recommendation was not strictly followed.
3. An acid cleaning was not performed following Run 1, which may have increased the specific
flux.

4. It is believed that the membrane fibers were uncompromised during the test run as evidenced
by the turbidity and particle count profiles. If membrane integrity had been compromised, it
would be expected to find a notable increase in particle concentrations in the filtrate.
                                          38

-------
Tables and Figures
       39

-------
40

-------
Table 1-1. Lake Michigan Feed
Parameter Units Count
Cl2-Residual(2)
Alkalinity
Total Hardness
Calcium
Hardness
TDS
TSS
Total Coliforms
Total Coliforms
HPC
TOC
UVA
Algae
pH
Temperature
Turbidity
(Bench-top)
mg/L
mg/L as
CaCOs
mg/L as
CaC03
mg/L as
CaC03
mg/L
mg/L
P/A
MPN/ 100
ml
CFU/ml
mg/L
cm"1
#/ml
-
°C (°F)
NTU
42
2
2
2
o
J
37
38
5
38
13
13
3
40
76
39
Water Quality Data
Median'1' Range
0.05
N/A
N/A
N/A
200
<1
N/A
<1.1
3
2.2
0.022
32.8
7.79
2.9 (37)
1.0
0.03-0.06
110-110
130-130
87-88
150-210
<1-8.0
A_p(3)

-------
Table 3-1   Verification Conditions

Operating Parameter                                              Verification Test Value
Flux                                                          30-40 1/h-m2 @ 20°C (1) (18-24 gfd)
Recovery                                                                 >90%(2)
Backwash Interval                                                         50 min(3)
Data Logging Interval                                        10 minutes (4 times per backwash cycle)
Production Backwash Chlorine Concentration                                  5 mg/L
Air Scour Flow Rate                                                  Approximately 2 cfm
Terminal Transmembrane Pressure                                            29 psi
1)  Test flux was determined based on results of pilot testing prior to the verification study.
2)  Recovery was at least 90%, however the exact values were calculated daily during verification testing and are reported in
    Chapter 4.
3)  Backwash duration was set to achieve minimum recovery criteria.


Table 3-2   Operational Data

Operation Parameter                                                    Frequency
Feed/Filtrate Water Flow(1)                                                 Continuous
Feed Pressure                                                             Continuous
Feed Temperature                                                         Continuous
Filtrate Pressure                                                           Continuous
Transmembrane Pressure                                                   Continuous
Flux @ 20°C                                                             Continuous
Specific Flux @ 20°C                                                     Continuous
Recovery                                                                  Daily
1)  Feed flow equals filtrate flow as measured on the filtrate side.
                                                  42

-------
Table 3-3    Chemical Cleaning Procedures and Conditions
Chemical
              Approximate Quantity
Sodium Hypochlorite
Sodium Hydroxide
         0.05 Ib (as NaOCl) /module/month
          2.4 Ib (as NaOH)/module/month
Cleaning Step
               Hydraulic Condition
     Duration
Step 1 - Chemical backwash

Step 2 - Cleaning soak
Step 3 - Cleaning rinse
                Backwash to drain

Soak at elevated temperatures not to exceed 90°F (32°C)
                Backwash to drain
60 seconds or approx.
       35 gal
 4, 8, and 44 hours0)
     90 seconds
Residual Characteristic
               Approximate Value
pH
Residual Chlorine
Total Volume of Residual
                      10-12
                 150 mg/L as C12
            150 gallons/module/cleaning
1)  Three chemical cleans were included as part of this testing. The first clean took place prior to membrane operation and was
    allowed to soak for 4 hours. The second and third chemical cleans were performed after elevated TMPs were reached and
    were allowed to soak for 8 and 44 hours, respectively.
Table 3-4 Cleaning Verification Data
Analytical Parameter
Temperature
pH
TDS
Turbidity
Visual observations
Operational Data
Specific Flux @ 20°C
Flux @ 20°C
Transmembrane pressure
Flow rate
Sample Timing
2,3,4,5
2,3,4,5
2,3,4,5
2,3,4,5
2,3,4,5
Sample Timing
1,5
1,5
1,5
1,5
1 ) Prior to stopping production step.
2) Immediately following chemical dosing backwash for chemical cleaning.
3) 120 minutes following chemical dosing backwash for chemical cleaning.
4) During cleaning rinse backwash at the end of the chemical soaking period.
5) During production immediately following rinsing.
                                                43

-------
Table 3-5   Water Quality Sampling Schedule

                                     Sampling
Parameter	Frequency	Feed	Filtrate	Backwash
On-Site Analysis
pH                                    I/day             X
Temperature                          Continuous          X
Turbidity (in-line)(1)(2)                 Continuous          X              X                 2/day
Particle Counts (in-line)(3)             Continuous          X              X
Total Chlorine(4)                        I/week            X
Off-Site Laboratory Analysis
Alkalinity                             I/month           X              X
Total Hardness                        I/month           X              X
Calcium Hardness                     I/month           X              X
TDS                                  2/month           X              X
TSS                                   5/week            XX                  X
Total Coliforms                         5/week            XX                  X
HPC                                   5/week            X              X
TOC                                   2/week            X              X
UVA	2/week	X	X	
1)  Turbidity data was recorded at 10-minute intervals.
2)  Online feed and filtrate turbidities were verified once per day with bench-top measurements. Backwash turbidity was
    measured twice per day with bench-top analysis and was not monitored by online instrumentation.
3)  Particle count data was recorded at 10-minute intervals.
4)  Backwash samples were analyzed once per day for total chlorine to verify the target dose of 5 mg/L. Feed water chlorine
    was monitored daily by the GBWLTFP and was verified by means of a pilot plant grab sample prior to ETV testing.
                                                   44

-------
Table 3-6   Data Management
Data Type
Documentation Format
Reporting Format
Membrane Hydraulic Data
Chemical Cleaning
  Observations/Measurements
Production Backwash
  Observations/Measurements
In-line Feed/Filtrate Turbidity
In-line Feed/Filtrate Particle Counts
In-line instrument flow measurements
Daily Walkthrough Observations
Monthly QA\QC Observations
Off-site analyses
On-site analyses
  pH, temperature, turbidity,
  conductivity
Photographs
       SCADA
      Notebook

      Notebook

       SCADA
  Dedicated Computer
      Notebook
      Notebook
      Notebook
  Feed and validated
    data from NLS

      Notebook

      Notebook
      Plots
      Table

      Table

      Plots
      Plots
      Table
      Table
      Table
   Table/Plots

   Table/Plots
                                               45

-------
Table 3-7    Methods for Measuring Precision and Accuracy
Parameter
          Precision
             Accuracy
pH

Temperature

Total Chlorine0 >
Bench-Top Turbidity


Alkalinity

Total Hardness

Calcium Hardness

Total Dissolved Solids

Total Suspended Solids

Total Organic Carbon


UV Absorbance

Total Coliform/HPC
 7 measurements per week with
      duplicates (100%)
 7 measurements per week with
      duplicates (100%)
   7 measurements per week
 14 measurements per week with
   weekly duplicates (14%)

     One duplicate (100%)

     One duplicate (100%)

     One duplicate (100%)

4 measurements with 1 duplicate
            (25%)
   60 measurements with 12
       duplicates (20%)
4 measurements per week with 1
       duplicate (25%)

4 measurements per week with 1
       duplicate (25%)
 15 each measurements per week
    with 3 duplicates (20%)
  Daily 3-point calibration with pH
       buffers at 4, 7, and 10
Initial and weekly testing against NIST
           thermometer
      Instrument "Zeroed" daily
Daily checks against primary standards
         (0, 20, 200 NTU)

      NLS standard procedures

      NLS standard procedures

      NLS standard procedures

      NLS standard procedures

      NLS standard procedures

      NLS standard procedures


      NLS standard procedures

      NLS standard procedures
1)  Samples were collected and analyzed daily by GBWLTFP as part of normal operating procedures. For verification, pilot
    plant feed waters were measured for chlorine at the beginning of the verification testing.
                                                 46

-------
Table 3-8   Sample Methodology
Parameter
   Methodology for All Samples
            Method
pH
Temperature
Turbidity (Bench-top)
Alkalinity
Total Hardness
Calcium Hardness
Total Dissolved Solids
Total Suspended Solids
Total Organic Carbon
UV Absorbance
Total Coliform/HPC
        Certified pH buffers
              None
    Ultra pure water as necessary
     NLS standard procedures
     NLS standard procedures
     NLS standard procedures
     NLS standard procedures
     NLS standard procedures
     NLS standard procedures
     NLS standard procedures
NLS and MWH Laboratories standard
           procedures
     Standard Method 4500-H+
      Standard Method 2550B
      Standard Method 2130
  Standard Methods 3010B, 2320B
        EPA Method 200.7
        EPA Method 200.7
      Standard Method 2540C
        EPA Method 160.2
      Standard Method 5310B
     Standard Methods 5910B
Standard Methods 9223/9215B (NLS)
   Standard Method 9221 (MWH)
                                               47

-------
Table 3-9   Corrective Action Plan
Parameter
     Acceptance Criteria
   Sequence of Steps for
     Corrective Action
Any Duplicate
Any Method Blank
Any Performance Evaluation
(PE) or Proficiency Sample
pH


Temperature

Turbidity (Bench-top)
Alkalinity, Total Hardness,
Calcium Hardness, Total
Dissolved Solids
Total Suspended Solids
(filtrate)
         < 10% apart
  See Table 3.8; criteria set by
    EPA-certified laboratory
    performing the analysis
  Within recovery specified for
 each PE or proficiency sample
< 10% difference from previous
day; < 1 pH unit difference from
    previous measurements
< 20% difference from previous
             day
  No increasing or decreasing
trend indicated by  results of bi-
  weekly proficiency samples.
Measurement deviates less than
      ±10% from standard.
< 20% difference from previous
           reading

    Assuming very low TSS
    concentrations, < 100%
difference from previous reading
Re-sample duplicates
Check instrument calibration;
recalibrate instrument
See Table 3.8; perform procedures
specified to each analysis as
determined by the state-certified,
third-party, or EPA-accredited
laboratory performing the analysis
Check and verify all steps in sample
collection and analysis
Re-do PE or proficiency sampling and
analysis
Resample
Check instrument calibration
Re-calibrate  instrument
Check for change in feed water source
to supply
Verify turbidimeter operation and
status of sample tap
Perform routine maintenance/cleaning
of instrument
Verify calibration using secondary
standards
Re-calibrate  using primary  standards
Verify change in feed water source or
supply

Verify corresponding increase in
turbidity
Re-sample
Check membrane integrity	
                                                   48

-------
Table 4-1 Operational
Operating Parameter
Run Period
Start Date and Time
End Date and Time
Run Length
Termination Condition
Normalized Flux (20 °C)(1)
Recovery(2)
Backwash Interval
Backwash Duration(3)
Conditions
Units
-
-
-
day:hrs
-
L/h-m2 (gfd)
%
minutes
seconds
Hydraulic Data Logging Interval minutes
Particle Count Logging Interval
Production Backwash Chlorine
Dose Setpoint
Air Scour Flow Rate
Terminal Transmembrane
Pressure
minutes
mg/L
cfm
Bar (psi)

1
3/11/0208:30
3/23/0222:17
12 days 14 hrs
Fouling
40 (24)
>90
50
60-120
10 (4 per backwash
cycle)
5
5
Approximately 2
1.4(21)

2
3/24/02 19:36
4/26/02 12:22
32 days 17 hours
Time
30 (18)
>90
50
60-120
10 (4 per backwash
cycle)
5
5
Approximately 2
1.7 (25)
1)   Test flux was determined based on results of pilot testing prior to the verification study.
2)   Recovery was at least 90%. The exact values were calculated during verification testing and are reported in Chapter 4.
3)   Backwash duration was set to achieve minimum recovery criteria.
                                                         49

-------
Table 4-2 Summary of Operational Data
Parameter
Runl
Flux
Flow Rate
Temperature
Recovery
Specific Flux
Decline at
20°C(1)
Run 2
Flux
Flow Rate
Temperature
Recovery
Specific Flux
Decline at
20°C(2)
1) Average daily
2) Average daily
Units

L/h-m2
(gfd)
gpm
°C (°F)
%
L/h - m2-
bar-d
(gfd/psi-d)

L/h-m2
(gfd)
gpm
°C (°F)
%
L/h -m2-
bar-d
(gfd/psi-d)
Count

1396
1396
1397
13
N/A

2738
2738
2738
27
N/A
Median

23.8
(14.0)
12.0
2.2 (36)
94
N/A

18.5
(10.9)
9.3
3.3 (38)
94
N/A
Range

17.9-25.4
(10.5-15.0)
9.0-12.8
1.5-3.2 (35-38)
89-96
N/A

16.1-24.2
(9.48-14.3)
8.1-12.2
1.6-6.8 (35-44)
89-95
N/A
Average

23.8
(14.0)
12.0
2.2 (36)
93
7.2
(0.29)

18.7
(11.0)
9.4
3.5 (38)
94
1.7
(0.069)
Standard
Deviation

0.5
0.3
0.3
1.3
N/A

0.98
0.49
1.31
1.2
N/A
95%
Confidence
Interval

23.8-23.8
(14.0-14.0)
12.0-12.0
2.2-2.2 (36-36)
93-94
N/A

18.8-18.8
(11.1-11.1)
9.4-9.4
3.5-3.5 (38-38)
93-94
N/A
specific flux decline for the duration of the Period 1 .
specific flux decline for the duration of the Period 2.
50

-------
Table 4-3
Parameter
Summary of Chemical
Units
Prior to
Clean
Cleaning
Start of
Clean
Water Quality Analyses
Sample Collection Time
During Clean

Backwash
Rinse

Production
after Clean
Chemical Cleaning 1(1) March 5, 2002
Time
Temperature
pH
TDS
Turbidity
Total C12
(hours)
°C (°F)

mg/L
NTU
mg/L
t<0
N/A
N/A
N/A
N/A
N/A
t=0
8.3 (47)
12.5
-
1.1
215
t=2
7.8 (46)
12.6
-
0.5
192
t=4
3.3 (38)
8.6
-
-
0.46
t= next run
3.3 (38)
7.9
-
0.1
-
Chemical Cleaning 2 <2) March 24, 2002
Time
Temperature
pH
TDS
Turbidity
Total C12
(hours)
°C (°F)

mg/L
NTU
mg/L
Chemical Cleaning 3 <3) May
Time
Temperature
pH
TDS(4)
Turbidity
Total C12
(hours)
°C (°F)

mg/L
NTU
mg/L
t<0
2(36)
7.8
-
0.032
5.1
1, 2002
to
6.3 (43)
-
-
0.037
-
t=0
14 (57)
12.2
4100
28.8
160

t=0
22 (72)
12.0
3900
31.1
186
t=2 t=8
15.8 (60.4)
12.2
4700 5900
10.9
168

t=2 t=22 t=44
25(77) 28(82) 31(88)
11.9 11.9 11.9
4500 5600 6500
17.6 17.2 15.7
180 176 180
t=8
18.6 (65.5)
12.18
130
12.3
72

t=44
16 (61)
11.9
3600
10.9
174
t=next run
7.5 (46)
8.1
120
0.121
0.7

t=next run
6.1(43)
-
-
-
-
1)  Chemical Cleaning 1 was conducted prior to membrane operation. The total soaking duration for chemical cleaning 1 was four
    hours.
2)  Chemical Cleaning 2 was allowed to soak for eight hours.
3)  Chemical Cleaning 3 was allowed to soak for 44 hours.
4)  These samples were analyzed exceeding the EPA holding times for TDS.
                                                        51

-------
Table 4-4   Summary of Chemical Cleaning Hydraulic Analyses
                                                 Value Prior to Chemical       Value Following
       Parameter                 Units                   Clean               Chemical Clean
Chemical Cleaning 1(1) March 5, 2002
Flux
Normalized Flux
Specific Flux
Normalized Specific Flux
TMP
Flow Rate
Chemical Cleaning 2 March
Flux
Normalized Flux
Specific Flux
Normalized Specific Flux
TMP
Flow Rate
Chemical Cleaning 3 May 1,
Flux
Normalized Flux
Specific Flux
Normalized Specific Flux
TMP
Flow Rate
1/h-m2 (gfd)
1/h-m2 (gfd)
l/h-m2-b (gfd/psi)
l/h-m2-b (gfd/psi)
psi
gpm
24, 2002
1/h-m2 (gfd)
1/h-m2 (gfd)
l/h-m2-b (gfd/psi)
l/h-m2-b (gfd/psi)
psi
gpm
2002
1/h-m2 (gfd)
1/h-m2 (gfd)
l/h-m2-b (gfd/psi)
l/h-m2-b (gfd/psi)
psi
gpm
N/A(1)
N/A
N/A
N/A
N/A
N/A

23.5(13.8)
39.4 (23.2)
16.5 (0.670)
27.7(1.12)
20.7
11.8

20.3 (12.0)
29.7(17.5)
11.8(0.479)
17.2 (0.698)
25.1
10.2
24 (14)
39 (23)
68 (2.8)
118(4.79)
5
12

19.2(11.3)
30.4(17.9)
45(1.8)
74.0(3.01)
6.1
10

21 (12)
32(19)
42(1.7)
63 (2.6)
7.2
10.6
1)   Chemical Cleaning 1 was performed prior to membrane operation. Therefore, this information was not available.
                                              52

-------
Table 4-5 Summary of Chemical Cleaning Efficacy
Normalized
Specific Flux
Prior to
Cleaning
L/h-m2-b
Cleaning Number Date (gfd/psi)
1 3-5-02 -(1)
2 (end of period 1)(2) 3-23-02 28 (1.1)
3 (end of period 2)(5) 5-1-02 17 (0.70)
Normalized
Specific Flux
Following
Cleaning
L/h-m2-b
(gfd/psi)
118(4.79)
74(3) (3.0)
63 (2.6)

Recovery of
Specific Flux

(%)
-J1)
62(4)
73

Loss of
Original
Specific Flux

(%)
-J1)
37(4)
47
1)   Testing began with a new module. Therefore, this information was not available.
2)   The end of period 1 was following approximately 12.5 days of non-stop operation at a constant specific flux of 40 L/hr-m
     (24 gfd).
3)   The readings were 67 L/h-m2-b immediately after chemical cleaning and stabilized after an increase to 74 L/h-m2-b.
4)   The manufacturer's recommendations include chemical recirculation at elevated temperature [90-95T (32-35°C)]. However,
     no recirculation loop or heating coil was used for this cleaning.
5)   The end of period 2 was folio wing approximately 33 days of non-stop operational a constant specific flux of 30 L/h-m2 (18
     gfd).
Table 4-6 Summary of Onsite Bench-top Turbidity Data
95%
Standard Confidence
Parameter Units Count Median Range Average Deviation Interval
Feed Water
pH 40
Temperature °C (°F) 76
Turbidity NTU 39
Filtrate
Turbidity NTU 41
Backwash
Water
Turbidity NTU 78

7.79 7.67-8.01 7.80 0.07 7.73-7.88
2.9(37) 1.9-6.5(35-44) 3.4(38) 1.6 3.4-3.4(38-38)
1.0 0.4-4.5 1.3 0.9 1.3-1.3

0.05 0.05-0.10 0.05 0.01 0.05-0.05

11 1.7-21 11 3.6 11-11
                                                          53

-------
Table 4-7 Summary of Online Turbidity and Particle Count
Parameter Units Count Median Range Average
Feed Water
Turbidity
>2 um Particles
2-3 um Particles
3-5 um Particles
5-7 um Particles
7-10 um Particles
10-15 um Particles
> 15 um Particles
Filtrate
Turbidity
>2 um Particles
2-3 um Particles
3-5 um Particles
5-7 um Particles
7-10 um Particles
10-15 um Particles
> 15 um Particles
Log Removal of
Particles'1*
>2 um Particles
2-3 um Particles
3-5 um Particles
5-7 um Particles
7-10 um Particles
10-15 um Particles
> 15 um Particles

NTU
#/ml
#/ml
#/ml
#/ml
#/ml
#/ml
#/ml

NTU
#/ml
#/ml
#/ml
#/ml
#/ml
#/ml
#/ml









4,124
11,175
11,175
11,175
11,175
11,175
11,175
11,175

4,124
11,175
11,175
11,175
11,175
11,175
11,175
11,175

8,941
6,016
6,261
3,567
4,032
3,651
6,185

0.85
2,835
1,201
1,208
178
153
60
19

0.0532
0.00
0.00
0.00
0.00
0.00
0.00
0.00

4.4
4.3
4.3
3.6
3.4
3.0
2.4

0.25-19
0-21,529
0-4,676
0-8,772
0-3,289
0-5,590
0-3,628
0-1,336

0.00-0.05
0-6,125
0-2,862
0-2,850
0-249
0-135
0-140
0-313

-1.9-5.9
-1.9-5.3
-1.9-5.5
-1.7-5.1
-1.4-5.2
-1.5-5.0
-2.8-4.7

1.3
4,281
1,602
1,880
325
305
127
41

0.05
4
1
1
0
0
1
2

4.2
4.1
4.1
3.4
3.3
2.9
2.2
Data
Standard
Deviation

1.4
3731
1,095
1,691
394
430
194
69

0.007
65
28
28
3
3
4
16

0.9
0.7
0.8
0.8
0.8
0.9
0.9
95%
Confidence 90th
Interval Percentile

1.3-1.3 2.9
4,278-4,283 9911
1,601-1,603
1,879-1,881
325-326
305-306
127-127
41-41

0.05-0.05 0.05
4-4 2
1-1
1-1
0-0
0-0
1-1
2-2
10th
Percentile
4.2-4.2 3.1
4.1-4.1 3.3
4.1-4.1 3.2
3.4-3.5 2.6
3.3-3.3 2.4
2.9-2.9 2.0
2.2-2.2 1.1
1)  Evaluation of the time series data showed relatively higher particle counts during, and immediately following, a backwash. As a
    result, particle removals were decreased during these times. Some of these consistently brief occurrences caused log removal data
    to become negative for short periods. Negative data points were not included in the statistical analysis for log removals.
                                                            54

-------
Table 4-8
Parameter
Feed Water
Total
Coliforms
HPC
Filtrate
Total
Coliforms
HPC
Backwash
Water(4)
Total
Coliforms
HPC
Summary of Microbial Water Quality
Units Count Median'1' Range Average

P/A
MPN/100 ml
CFU/ml

P/A
MPN/100 ml
CFU/ml

P/A
MPN/100 ml
CFU/ml

38 N/A(2) A-P(2) A
4 <1.1 
-------
Table 4-9 General Water Quality Parameters
Parameter Units Count Median'1' Range
Feed Water
C12-
Residual(2)
Alkalinity
Total Hardness
Calcium
Hardness
TSS
TDS
TOC
UVA
Algae
Filtrate
Alkalinity
Total Hardness
Calcium
Hardness
TSS
TDS
TOC
UVA
Backwash
Water(3)
C12
TSS

mg/L
mg/L as
CaC03
mg/L as
CaC03
mg/L as
CaC03
mg/L
mg/L
mg/L
cm"1
#/ml

mg/L as
CaCO3
mg/L as
CaC03
mg/L as
CaC03
mg/L
mg/L
mg/L
cm"1

mg/L
mg/L

42
2
2
2
37
3
13
13
o
3

2
2
2
37
o
J
19
19

40
36

0.05
N/A
N/A
N/A
<1
200
2.2
0.022
32.8

N/A
N/A
N/A
<1
210
2.0
0.019

4.8
20.

0.03-0.06
110-110
130-130
87-88
<1-8.0
150-210
1.6-3.4
0.017-0.043
32.6-37.8

110-110
130-130
87-87
<1-9.0
190-210
1.6-3.0
0.015-0.027

2.0-11.7
<1-41
Average

0.05
110
130
88
1.3
187
2.3
0.024
34.4

110
130
87
1.2
203
2.0
0.019

4.7
20
Standard
Deviation

0.01
N/A
N/A
N/A
1.8
N/A
0.5
0.007
N/A

N/A
N/A
N/A
1.9
N/A
0.3
0.003

1.5
10
95%
Confidence
Interval

0.05-0.05
N/A
N/A
N/A
1.3-1.3
N/A
2.3-2.3
0.024-0.024
N/A

N/A
N/A
N/A
1.2-1.2
N/A
2.0-2.0
0.019-0.019

4.7-4.7
20-20
1)   Values reported as non-detect were assumed to be one-half of the detection limit for the purposes of statistical evaluation.
2)   Measured as part of the daily sampling activities of the GBWUFP.
3)   Sampled during the second cycle of the production backwash.
                                                           56

-------
Table 4-10 TSS Mass Balance
Filtrate Fi"nfm
Date Flow _<*?.
Length
(gpm) (min)
Test Period 1
3/11/2002
3/11/2002
3/12/2002
3/13/2002
3/14/2002
3/15/2002
3/18/2002
3/19/2002
3/20/2002
3/21/2002
3/21/2002
3/22/2002
Test Period 2
3/25/2002
3/26/2002
3/27/2002
3/28/2002
3/28/2002
3/29/2002
4/1/2002
4/2/2002
4/3/2002
4/4/2002
4/4/2002
4/8/2002
4/9/2002
4/10/2002
4/11/2002
4/11/2002
4/15/2002
4/16/2002
4/17/2002
4/18/2002
4/18/2002
4/19/2002
4/22/2002
4/23/2002

12
12
12
12
12
12
12
12
12
12
12
12

9.4
9.4
9.4
9.4
9.4
9.4
9.4
9.4
9.4
9.4
9.4
9.4
9.4
9.4
9.4
9.4
9.4
9.4
9.4
9.4
9.4
9.4
9.4
9.4

50
50
50
50
50
50
50
50
50
50
50
50

50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
Volume
Filtered
(gal)

600
600
600
600
600
600
600
600
600
600
600
600

470
470
470
470
470
470
470
470
470
470
470
470
470
470
470
470
470
470
470
470
470
470
470
470
Backwash
Volume
(gal)

50
50
40
55
50
32
45
40
37
35
35
34

50
42
38
36
36
30
33
32
30
30
30
26
25
32
27
27
24
28
28
28
28
30
24
24
Measured
Feed
TSSd)
(mg/L)

1.0
0.5
8.0
6.0
1.0
0.5
1.0
0.5
0.5
1.0
2.0
0.5

4.0
0.5
0.5
0.5
4.0
1.0
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
Measured
Backwash
TSS(1)(2)
(mg/L)

0.5
4
26
3
23
40
41
25
31
24
24
19

14
20
14
10
13
27
14
2
20
0.5
17
29
26
23
15
20
22
12
30
20
21
36
27
16
Calculated
Backwash
TSS
(mg/L)

12
6
120
65
12
9
13
8
8
17
34
9

38
6
6
7
52
16
7
7
8
8
8
9
9
7
9
9
10
8
8
8
8
8
10
10
RPD(3)
%

184
40
129
182
63
124
102
108
117
33
35
73

91
113
77
42
120
53
65
114
87
176
74
105
94
103
53
79
77
35
113
82
86
129
94
48
1)   Values reported as non-detect (<1 mg/1) were assumed to be one-half of the detection limit (0.5 mg/1) for the purposes of this
     evaluation.
2)   Sampled during the second cycle of the production backwash.
3)   Relative Percent Deviation.
                                                           57

-------
Table 4-11 Summary of Verification Data for the Feed Water Particle Counter
Standard
Parameter Units Count Median Range Average Deviation'1'
2 um Particle Standard at a
>2 um Particles
2-3 um Particles
3-5 um Particles
5-7 um Particles
7-10 um Particles
10-15 um Particles
> 15 um Particles
#/ml
#/ml
#/ml
#/ml
#/ml
#/ml
#/ml
5 um Particle Standard at a
>2 um Particles
2-3 um Particles
3-5 um Particles
5-7 um Particles
7-10 um Particles
10-15 um Particles
> 15 um Particles
#/ml
#/ml
#/ml
#/ml
#/ml
#/ml
#/ml
10 um Particle Standard at
>2 um Particles
2-3 um Particles
3-5 um Particles
5-7 um Particles
7-10 um Particles
10-15 um Particles
> 15 um Particles
#/ml
#/ml
#/ml
#/ml
#/ml
#/ml
#/ml
95%
Confidence
Interval'1'
Concentration of 1000 #/ml
3
3
3
3
3
3
3
1143.7
720.0
303.2
43.5
41.5
27.4
13.2
1141.7-1332.2
714.8-753.7
302.2-406.8
41.9-63.4
39-60.1
25.9-33.2
12.9-15.1
1205.9
729.5
337.4
49.6
46.9
28.8
13.7
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Concentration of 1000 #/ml
3
3
3
3
3
3
3
1121 A
129.9
780.5
149.6
37.5
17.9
8.3
a Concentration of
3
3
3
3
3
3
3
952.1
43.8
70.5
26.8
585.1
212.9
11.3
1108.4-1130.9
129.5-132.3
774.8-780.9
147.5-150.9
35.8-39.5
14.8-18
6-13
1000 #/ml
944.954.6
36.5-47.3
60.6-70.7
25.5-27.2
579.2-587.2
212.4-220.1
6.1-27.7
1122.2
130.6
778.7
149.3
37.6
16.9
9.1

950.2
42.5
67.2
26.5
583.8
215.1
15.0
N/A
N/A
N/A
N/A
N/A
N/A
N/A

N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A

N/A
N/A
N/A
N/A
N/A
N/A
N/A
1)  Less than eight data points exist for this data. Therefore, this statistical analysis was not performed.
                                                             58

-------
Table 4-12  Summary of Verification Data for the Filtrate Water Particle Counter
                                                                                                 95%
                                                                               Standard      Confidence
   Parameter	Units    Count  Median	Range	Average    Deviation'1'	Interval'1'
2 um Particle Standard at a
>2 um Particles
2-3 um Particles
3-5 um Particles
5-7 um Particles
7-10 um Particles
10-15 um Particles
> 15 um Particles
#/ml
#/ml
#/ml
#/ml
#/ml
#/ml
#/ml
5 um Particle Standard at a
>2 um Particles
2-3 um Particles
3-5 um Particles
5-7 um Particles
7-10 um Particles
10-15 um Particles
> 15 um Particles
#/ml
#/ml
#/ml
#/ml
#/ml
#/ml
#/ml
10 um Particle Standard at
>2 um Particles
2-3 um Particles
3-5 um Particles
5-7 um Particles
7-10 um Particles
10-15 um Particles
> 15 um Particles
#/ml
#/ml
#/ml
#/ml
#/ml
#/ml
#/ml
Concentration of 1000 #/ml
2
2
2
2
2
2
2
N/A(2)
N/A(2)
N/A(2)
N/A(2)
N/A(2)
N/A(2)
N/A(2)
1089.7-1126.4
639.1-647.7
333.7-348.3
42.6-47.3
39.2-43.4
23.5-28.5
11.3-11.5
1108.0
643.4
341.0
45.0
41.3
26.0
11.4
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Concentration of 1000 #/ml
5
5
5
5
5
5
5
727.9
68.3
564.0
81.5
5.9
1.2
5.9
a Concentration of
4
4
4
4
4
4
4
840.4
49.7
113.3
46.6
476.0
145.9
6.4
719.4-732.8
67.1-69
557-571.6
75.4-87.4
4.7-6.3
0.7-2.5
3-8.9
1000 #/ml
830.6-849.6
47-56.1
108.9-117.4
45.1-50.3
471.1-484.6
144.4-147.1
4.4-9.2
727.4
68.1
564.3
81.9
5.7
1.5
6.0

840.3
50.6
113.2
47.2
476.9
145.8
6.6
N/A
N/A
N/A
N/A
N/A
N/A
N/A

N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A

N/A
N/A
N/A
N/A
N/A
N/A
N/A
1)  Less than eight data points exist for this data. Therefore, this statistical analysis was not performed.
2)  Less than three data points exist for this data. Therefore, this statistical analysis was not performed.
                                                    59

-------
Table 4-13 Chemical Consumption Analysis
Approximate Total
Product Volume
Produced
Operation (gal)(1)
Run 1 197,463
Chemical x T .
™ ~ NA
Clean 2

Run 2 402,388
Chemical x T .
™ , NA
Clean 3

Approximate Total
Backwash Volume
Used
(gal)<2)
13,822
35
Total for Period 1
24,143
32
Total for Period 2

Chlorine Consumed
(Ibs as NaOCl)<3)
0.61
0.06
0.67
1.06
0.06
1.12
NaOH
Consumed
(Ibs as NaOH)(4)
NA
2.04
2.04
NA
1.47
1.47
1)  Based on average flow data from Table 4.2 and a complete production and backwash cycle time of 55 minutes.
2)  Based on daily observations of production backwash volumes and observation during each chemical clean.
3)  Based on a production cycle backwash chlorine dose of 5 mg/L and chemical cleaning dose of 200 mg/L as verified
    through onsite analysis.
4)  Based on the programmed dose of 7,000 and 5,500 mg/L for Chemical Cleanings 2 and 3, respectively.
                                                      60

-------
                                                 10
                                                                    12
                                                                              13
                                                                                                 15
                                                                                                           16
                                                                                                                    17
                                                                                                                              18
                                                                                                                                       19
                                                                                                                                                 20
                                                                                                                                                                    22
                                                                                                                                                                              23
                                                                                                                                                                                       24
                                                                                               SODIUM
                                                                                                HYPO-
                                                                                               CHLORITE
                                                                                               PUMPP310
                                                                                               (0.75GPH®
                                                                                               80PSIG)
   RAW WATER
    PUMP P100
(1.5HP,45GPM@28PSIG)
                                                                              BACKWASH WATER
                                                                                 PUMP P600
                                                                            (7.5HP, 150GPM@55PSIG)
                                                                                                                                               Figure 2.1
                                                 Intuitechs.
                                      THIS DRAWING IS THE INTELLECTUAL PROPERTY OF INTUITECH AND
                                      MAY NOT BE REPRODUCED IN FULL OR IN PART FOR ANY PURPOSE
                                        ASIDE FROM THE PROJECT AS SPECIFIED ON THIS DOCUMENT
                                                 10
                                                           11
                                                                    12
                                                                              13
DIMENSIONS ARE IN INCHES
UNLESS OTHERWISE NOTED
TOLERANCES ARE AS
FOLLOWS, UNLESS
OTHERWISE NOTED

     1/X
    1/XX
     .OX
     .OOX: +.005
 ANGLES : +1°

  14         is
REFERENCE
 DRAWINGS
                                                                                                                      TITLE:
                UF PILOT
PROCESS & INSTRUMENTATION DIAGRAM
                                                                                                                       CLIENT:  CAROLLO ENGINEERS - BOISE
                                                                                                                       DESIGN BY:  EJH
                                                                                                                       DRAWN BY:  EJH
                                                                                                                       CHECK BY:
             RELEASED FOR:  FABRICATION
                                                                                                           16
                                                                                                                    17
                                                                                                                       FILE NAME:  P&ID
                                                                                                                              is
                                                 PROJECT:  UF PILOT #1
                                                 DESIGN DATE:  6-18-99
                                                 DRAWN DATE:  6-18-99
                                                 CHECK DATE:
                      P.O.:
                                                 REVISION:  3
                                                                                                                                       19
                                                                                                                                                 20
                                                                                                                                                           21
                                                                                                                                                                    22
                                                                                                                                                                              23
                                                                                                                                                                                       24

-------
0.0 4*
   3/11
                     3/16
3/21
3/26     3/31
             Transmembrane Pressure (bar)
            A System Recovery
            • Flux@20C
            I Specific Flux @ 20 C
 4/5       4/10      4/15
Time
4/20
                                                                                                             160
                                                                                                             140
                                                                                                             120
                                                              Transmembrane Pressure
                                             4 A A A/A A A A A A A A A A A     A A A A A.    AAA

                                                  /                                   \
                                                                                                                     o
                                                                                                                     o
                                                                                                                     o
                                                                                                             100  o  ~
                                                                                                                  O
                                                                                                                  CM
                                                                                                             80   N~
                                                                                                                      o
                                                                                                                      0)
                                                                                                                      Q.
                                                                                                                     V)
                                                                                                             20
                                                                                                             0
4/25     4/30
Note: The membrane unit was off-line for 21 hours and 10 minutes following
Run 1 for chemical cleaning. In addition, on 4/18/02, the membrane unit was
shut down for approximately 10 hours due to lightning.
                                                                    Figure 4-1
                                                 IMP, Flux, and System Recovery Profiles for the
                                                 Polymem UF Module UF120 S2 Membrane Module
                                                                                                                                o
                                                                                                                                es
                                                                                                                                .0.
                                                                                                                                S
                                                                                          --3   •=
                                                                                                  Q.
                                                                                                  (0

-------
O>
CO
28

26

24

22

20

18

16

14

12

10

 8

 6

 4

 2

 0
                                              Temperature
                 3/11
            3/16
             • Temperature
              l Operational Flux
3/21       3/26       3/31        4/5       4/10
                         Run Time (Days)
4/15
4/20
4/25
4/30
Note: The membrane unit was off-line for 21 hours and 10 minutes following
Run 1 for chemical cleaning. In addition, on 4/18/02, the membrane unit was
shut down for approximately 10 hours due to lightning.
                                                                                   Figure 4-2
                                                                Operational Flux and Temperature Profiles for the
                                                                Polymem UF Module UF120 S2 Membrane Module

-------
               100
  O>
                                                                                                         Raw Online Turbidity
             0.001
              0.01
                                                                             Filtrate Online Turbidity
                  3/11
3/16
3/21
3/26
3/31
       	Raw Online Turbidity
        D  Raw Benchtop Turbidity
            Filtrate Online Turbidity
        A  F&trateate Benchtop Turbidity
        O  Backwash Benchtop Turbidity
Note: The membrane unit was off-line for 21 hours and 10 minutes following
Run 1 for chemical cleaning. In addition, on 4/18/02, the membrane unit was
shut down for approximately 10 hours due to lightning.
 4/5       4/10      4/15      4/20       4/25      4/30
Time
                                                                     Figure 4-3
                                                    Online and Benchtop Turbidity Profiles for the
                                                  Polymem UF Module UF120 S2 Membrane Module

-------
CD
cn
           100000
             10000
c
3
o
o
0)
o
        s.
              1000
        =•     100
                                                                                                               Feed Particle Counts
                                                                                                             Filtrate Particle Counts
              0.01
                  3/11
                    3/16
3/21
3/26
3/31
             I Filtrate Particle Counts

            • Feed Particle Counts
 4/5

Time
4/10
4/15
4/20
4/25
4/30
Note: The membrane unit was off-line for 21 hours and 10 minutes following

Run 1 for chemical cleaning. In addition, on 4/18/02, the membrane unit was

shut down for approximately 10 hours due to lightning.
                                                                                        Figure 4-4

                                                                 Feed and Filtrate Count Profiles for Particles >2 um for the

                                                                     Polymem UF Module UF120 S2 Membrane Module

-------
  CD
  CD
               100000
c
3
o
o
0)
o

t
re
Q.
                10000
                 1000
           ~     100
                 0.01
                     3/11
                    3/16
3/21
3/26
3/31
             Filtrate Particle Counts

             l Feed Particle Counts
 4/5


Time
4/10
4/15
4/20
4/25
4/30
Note: The membrane unit was off-line for 21 hours and 10 minutes following

Run 1 for chemical cleaning. In addition, on 4/18/02, the membrane unit was

shut down for approximately 10 hours due to lightning.
                                                                                    Figure 4-5

                                                             Feed and Filtrate Count Profiles for Particles 2-3 um for the

                                                                 Polymem UF Module UF120 S2 Membrane Module

-------
O>
         10000
          1000
          100
3
o
o
0)
o

t
re
Q.
                                                                        y.M'M
                                                                                             Filtrate Particle

                                                                                         I/,1,11  Counts
                                              IB     •IBMHIMMIIi
          0.01
             3/11
                3/16
3/21
3/26
3/31
4/10
4/15
4/20
4/25
      I Filtrate Particle Counts

      • Feed Particle Counts
Note: The membrane unit was off-line for 21 hours and 10 minutes following

Run 1 for chemical cleaning. In addition, on 4/18/02, the membrane unit was

shut down for approximately 10 hours due to lightning.
 4/5

Time

                    Figure 4-6

 Feed and Filtrate Count Profiles for Particles 3-5 um for the

     Polymem UF Module UF120 S2 Membrane Module
4/30

-------
  CD
  00
             10000
              1000
               100 -
          5
          4-1
          c
          o
          o
          s.
              0.01
                 3/11
3/16
3/21
3/26
3/31
4/10
4/15
4/20
4/25
4/30
             I  Filtrate Particle Counts
             • Feed Particle Counts
Note: The membrane unit was off-line for 21 hours and 10 minutes following
Run 1 for chemical cleaning. In addition, on 4/18/02, the membrane unit was
shut down for approximately 10 hours due to lightning.
 4/5
Time
                         Figure 4-7
  Feed and Filtrate Count Profiles for Particles 5-7 um for the
      Polymem UF Module UF120 S2 Membrane Module

-------
CD
            10000
             1000
              100
         3
         o
         o
         S.
             0.01
                             III III II
                    •III
            Mill III!
              Ill
                          iiiiniiHi IIIIHIIIIII mm 111 mi in
                         HUH i iiiiiiiniMiiiiiniiini
                                                    • mini i nun
                                                     IMIII in mi
                 3/11
3/16
3/21
3/26
3/31
                Filtrate Particle Counts

               l Feed Particle Counts
 4/5       4/10

Time
4/15
4/20
4/25
4/30
Note: The membrane unit was off-line for 21 hours and 10 minutes following

Run 1 for chemical cleaning. In addition, on 4/18/02, the membrane unit was

shut down for approximately 10 hours due to lightning.
                                                                  Figure 4-8

                                          Feed and Filtrate Count Profiles for Particles 7-10 um for the

                                               Polymem UF Module UF120 S2 Membrane Module

-------
            10000
             1000
         o
         O
         re
         Q.
                   Vj'V  taflU
              0.1
             0.01
                3/11
3/16
3/21
             I Filtrate Particle Counts
             • Feed Particle Counts
3/26
3/31
 4/5
Time
4/10
4/15
4/20
4/25
4/30
Note: The membrane unit was off-line for 21 hours and 10 minutes following
Run 1 for chemical cleaning. In addition, on 4/18/02, the membrane unit was
shut down for approximately 10 hours due to lightning.
                                                                 Figure 4-9
                                         Feed and Filtrate Count Profiles for Particles 10-15 um for the
                                              Polymem UF Module UF120 S2 Membrane Module

-------
          10000
           1000
       o
       o
-J     7;
       re
       Q.
                                                                                                                      Filtrate Particle
                                                                                                                      Counts
           0.01
               3/11
3/16
3/21
3/26
3/31
          I Filtrate Particle Counts
          • Feed Particle Counts
 4/5
Time
4/10
4/15
4/20
4/25
4/30
Note: The membrane unit was off-line for 21 hours and 10 minutes following
Run 1 for chemical cleaning. In addition, on 4/18/02, the membrane unit was
shut down for approximately 10 hours due to lightning.
                                                                      Figure 4-10
                                              Feed and Filtrate Count Profiles for Particles >15 um for the
                                                   Polymem UF Module UF120 S2 Membrane Module

-------
            _  4
            re
            §
            0)
            re
            Q.
 Backwash
 Events
                                      I1',IN,'!.!

                     9
                          '.''
                                        .....
                                                 .••
                   '   i'll  .
                   I.1 11
                  <>  f
                          1  '    ,'
                                                      III    I
                                                                                  'I
                                                                     -I-
                                                                                               1 it'
                                                                                              	
               3/11
3/16
3/21
3/26
3/31
 4/5
Time
4/10
4/15
4/20
4/25
Note: The membrane unit was off-line for 21 hours and 10 minutes following
Run 1 for chemical cleaning. In addition, on 4/18/02, the membrane unit was
shut down for approximately 10 hours due to lightning.
4/30
                                                                          Figure 4-11
                                                      Log Removal Profile for Particles >2 um in Size for the
                                                        Polymem UF Module UF120 S2 Membrane Module

-------
CO
Backwash  1
Events
            3/11
3/16
4/25
Note: The membrane unit was off-line for 21 hours and 10 minutes following
Run 1 for chemical cleaning. In addition, on 4/18/02, the membrane unit was
shut down for approximately 10 hours due to lightning.
4/30
                                                                                Figure 4-12
                                                           Log Removal Profile for Particles 2-3 um in Size for the
                                                              Polymem UF Module UF120 S2 Membrane Module

-------
Backwash
Events
             3/11
3/16
4/25
4/30
Note: The membrane unit was off-line for 21 hours and 10 minutes following
Run 1 for chemical cleaning. In addition, on 4/18/02, the membrane unit was
shut down for approximately 10 hours due to lightning.
                                                                               Figure 4-13
                                                          Log Removal Profile for Particles 3-5 um in Size for the
                                                            Polymem UF Module UF120 S2 Membrane Module

-------
en
  Backwash
  Events
             3/11        3/16
                               4/25        4/30
Note: The membrane unit was off-line for 21 hours and 10 minutes following
Run 1 for chemical cleaning. In addition, on 4/18/02, the membrane unit was
shut down for approximately 10 hours due to lightning.
                     Figure 4-14
Log Removal Profile for Particles 5-7 um in Size for the
  Polymem UF Module UF120 S2 Membrane Module

-------
CD
Backwash
Events
            3/11       3/16
                               4/25       4/30
Note: The membrane unit was off-line for 21 hours and 10 minutes following
Run 1 for chemical cleaning. In addition, on 4/18/02, the membrane unit was
shut down for approximately 10 hours due to lightning.
                     Figure 4-15
Log Removal Profile for Particles 7-10 um in Size for the
   Polymem UF Module UF120 S2 Membrane Module

-------
Backwash
Events
             3/11        3/16
                                4/25        4/30
Note: The membrane unit was off-line for 21 hours and 10 minutes following
Run 1 for chemical cleaning. In addition, on 4/18/02, the membrane unit was
shut down for approximately 10 hours due to lightning.
                      Figure 4-16
Log Removal Profile for Particles 10-15 um in Size for the
   Polymem UF Module UF120 S2 Membrane Module

-------
00
  Backwash
  Events
Note: The membrane unit was off-line for 21 hours and 10 minutes following
Run 1 for chemical cleaning. In addition, on 4/18/02, the membrane unit was
shut down for approximately 10 hours due to lightning.
                                                                                                               4/25
                                         4/30
                     Figure 4-17
Log Removal Profile for Particles >15 um in Size for the
   Polymem UF Module UF120 S2 Membrane Module

-------
CD
O)
c

'•B
o>
o>
o
X
LLI
+j
O
z
4-i

0)
o

0)
Q.
Q0%
80%



40%


1 0%




i
p










1



• t *
••


















                                      0.01
                                                   0.02               0.03   0.035

                                                       Turbidity (NTU)
0.04
0.05
                                                                                                   Figure 4-18

                                                                                A Frequency Distribution of Filtrate Turbidity for the

                                                                                 Polymem UF Module UF120 S2 Membrane Module

-------
00
O
         O)


        '•&
         
-------
00
       O)
       c
       O
       X
       LU
0)
O

0)
Q.
               10%
                0%
                                                       456

                                                        Particle Count (#/ml)
                                                                                                      10
                                                                                              Figure 4-20
                                                                        A Frequency Distribution for Filtrate Partilces >2 um for the
                                                                            Polymem UF Module UF120 S2 Membrane Module

-------
00
        O)
        O
        X
        LU
        £
100%


 90%


 80%


 70%


 60%


 50%


 40%


 30%


 20%


 10%


  0%
                                5000
                                10000          15000
                                        Particle Count (#/ml)
20000
25000
30000
                                                                                       Figure 4-21
                                                                 A Frequency Distribution for Feed Water Partilces >2 um for the
                                                                       Polymem UF Module UF120 S2 Membrane Module

-------
00
CO
          O)
O
X
LU
+J

0)
O
L.
0)
Q.
                100%
                 90%
                                                            Log Removal
                                                                                      Figure 4-22
                                                              A Frequency Distribution of Log Removal for Particles >2 um for the
                                                                      Polymem UF Module UF120 S2 Membrane Module

-------
00
        O)
        c
        0)
        o
        X,
        LU
        0)
        o

        0)
        Q.
100%



 90%



 80%



 70%



 60%



 50%



 40%



 30%



 20%



 10%
                0% +
                                                234

                                                         Log Removal
                                                                                    Figure 4-23

                                                            A Frequency Distribution of Log Removal for Particles 2-3 um for the

                                                                    Polymem UF Module UF120 S2 Membrane Module

-------
oo
en
         O)
         c
         0)
         
-------
00
CD
        TO
        c

        TJ
        0)

        8
        X
        LU
        •4-1

        0)

        H
        0)
        Q.
100%



 90%



 80%



 70%



 60%



 50%



 40%



 30%



 20%



 10%



  0%
                   0
                               234

                                       Log Removal
                                                                         Figure 4-25

                                                 A Frequency Distribution of Log Removal for Particles 5-7 um for the

                                                         Polymem UF Module UF120 S2 Membrane Module

-------
00
     O)
     0)
     0)
     u
     X
     HI
     0)
     £
     0)
     0.
            100%
             90%
             80%
             70%
             60%
50%
40%
             30%
             20%
             10%
\
              0%
                                                234

                                                          Log Removal
                                                                                   Figure 4-26

                                                           A Frequency Distribution of Log Removal for Particles 7-10 um for the

                                                                   Polymem UF Module UF120 S2 Membrane Module

-------
00
00
       O)
       c
       0)
       
-------
oo
VO
        O)
        O
        X
        LU
        +J

        0)
        O

        0)
        Q.
100%



 90%



 80%



 70%



 60%



 50%



 40%



 30%



 20%



 10%



  0%
                                                              3             4

                                                         Log Removal
                                                                                   Figure 4-28

                                                           A Frequency Distribution of Log Removal for Particles >15 um for the

                                                                   Polymem UF Module UF120 S2 Membrane Module

-------
mnnn
mnn
E
5
' — ' mn
c
3
o
o
03 m
o IU
t
re
Q.
-\
D
D
0.1
001
ii, i'
'i ' 	 , 	 '
i
I I I I I II II I I I II I II II I
• /
• | • • Time of Backwash Events
" f • ' i • •• • • •"
• Ml •• • • • • MB
oooooooooooooo
oooooooooooooo
c»
c
c
c
c*
T-
?
} ^3 ^"™ C^ CO ^f LO CO 1*^ CO O5 ^3 ^~™ C^
JCN|CN|CN|CN|CN|CN|CN|CN|CN|CN|T~T~T~
JOOOOOOOOOOCNICNICNI
JT^T^T^T^T^T^T^T^T^T^^^^
"cocococococococococo^^^
•) CO CO CO
.''',
1, '"l, '
	 	 '

1 1 1 1
•
H'
> ' \'
• 4 •
MB • •
o o o o
o o o o
CN| CN| CN| CN|

''l, 	
	 ' ^"' 	
\
Feeql Particle Counts
II 1 ^ 1 II
1 Filtrate Particle Counts
- /'.
• m m
mm m • •
• • •• •
o o o o o
o o o o o
CN| CN| CN| CN| CN|






c
CN
CN
CN
^
T-
COCOCOCOCOCOCOCOCOC1"
l Feed Particle Counts T\me
Figure 4-29
• Filtrate Particle Counts Feed and FMtrate particle Count Profiles for Particles >2 urn
from 3/14/02 00:00 to 3/14/02 20:44 Using Raw Data for the
I Markers Indicating the Time of Likely Polymem UF Module UF120 S2 Membrane Module

-------
IUUUUU
mnnn
mnn
E
' — ' mn
3
o
o
03 m
o IU
t
-\
^o
0.1
Om

\ 	 	
	 '' 	
''' 1 \ I1
	 '' 	 ' 	 ' ' 	 '""^. 	
\
Feed Particle Counts
I I I I I II II I I I II I
/
• | • Time of Backwash Events
• • _•
• f-«" •• (*"" • i " • i"
" f • ' i • •• i i~
• •• • •• M • M M
M1H •• ^ • • • 1H •
ooooooooooooo
ooooooooooooo
COO-*— OJCO^LOCQI^COOOT—
CN|CN1CN1CN1CN1CN1CN1CN1CN1CN1CN1T~T~
CNIOOOOOOOOOOCNICNI
COT^T^T^T^T^T^T^T^T^T^^^
CO CO CO
1 II 1 1 1 1 1 II 1 1 1 II
Filtrate Particle Counts
- / .
• • •
~S* % mP i f i
mm mm mm m mmmm mm mm
mmm mm •••••
ooooooooooc
ooooooooooc
T-T-T-T-T-T-T-T-CNCsicN
CNICNICNICNICNICNICNICNICNICNICN
COCOCOCOCOCOCOCOCOCOf
I Feed Particle Counts Time Figure 4-30
Feed and Filtrate Particle Count Profiles for Particles >2 um
• Filtrate Particle Counts , _ .. . .__ __ __ . _ .. . .__ __ .,.,,_ •_•- r* r*. *. .1
from 3/14/02 00:00 to 3/14/02 20:44 Excluding Data Collected
During Backwash Events for the
I Markers Indicating the Time of Likely _ . ,,.-..,, » «« . ....
Backwash Event Polymem UF Module UF120 S2 Membrane Module

-------
VO
to
O)


£
0)
0)
o
X
LU
+j

0)
o
L.
0)
Q.
               100%
                90%
                70%
                  l% - -
                 0%


^r
^^
/^ .
Curve Developed with Raw Data Set /
/
/
/
/
Curve Developed Excluding Data Collected









,
\
j
A
7

















\ ,
L *L
'•\.\
\
\
\
\
"ll
\,
\
\
\
I
\
V
\
\
ll.
                                                  Log Removal
                                                               3.2  3.6
                                                                             Figure 4-31

                                                       A Frequency Distribution of Log Removal for Particles >2 um

                                                               from 3/14/02 00:00 to 3/14/02 20:44 for the

                                                            Polymem UF Module UF120 S2 Membrane Module

-------