October 2009
NSF 09/29/EPADWCTR
EPA/600/R-10/013
Environmental Technology
Verification Report
Removal of Dissolved Salts and Particulate
Contaminants from Seawater
Village Marine Tec.
Expeditionary Unit Water Purifier,
Generation 1
Prepared by
NSF International
Under a Cooperative Agreement with
U.S. Environmental Protection Agency
-------
-------
THE ENVIRONMENTAL TECHNOLOGY VERIFICATION
PROGRAM ^
f X
&EPA
ETV
U.S. Environmental Protection Agency
NSF International
ETV Joint Verification Statement
TECHNOLOGY TYPE: ULTRAFILTRATION AND REVERSE OSMOSIS
APPLICATION: REMOVAL OF DISSOLVED SALTS AND PARTICULATE
CONTAMINANTS FROM SEAWATER
PRODUCT NAME: EXPEDITIONARY UNIT WATER PURIFIER (EUWP)
VENDOR: VILLAGE MARINE TEC.
ADDRESS: 2000 W. 135TH ST.
GARDENA, CA 90249
PHONE: 310-516-9911
EMAIL: SALES@VILLAGEMARINE.COM
NSF International (NSF) manages the Drinking Water Systems (DWS) Center under the U.S.
Environmental Protection Agency's (EPA) Environmental Technology Verification (ETV) Program. The
DWS Center evaluated the performance of the Village Marine Tec. Generation 1 Expeditionary Unit
Water Purifier (EUWP). The EUWP, designed under U.S. Military specifications for civilian use,
employs ultrafiltration (UF) and reverse osmosis (RO) to produce drinking water from a variety of
sources. This document provides the verification test results for the EUWP system using seawater at
Naval Base Ventura County in Port Hueneme, California.
EPA created the 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 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 09/29/EPADWCTR The accompanying notice is an integral part of this verification statement. October 2009
VS-i
-------
PRODUCT DESCRIPTION
The following technology description was provided by the manufacturer and has not been verified.
The EUWP was developed to treat challenging water sources with variable turbidity, chemical
contamination, and very high total dissolved solids (TDS), including seawater, during emergency
situations when other water treatment facilities are incapacitated. The EUWP components include feed
pumps, a UF pretreatment system, a one or two pass RO desalination system with an energy recovery
device, storage tanks, and product pumps. It has chemical feed systems for optional pretreatment
coagulation and post treatment chlorination. Clean-in-place systems are included with the UF and RO
skids. During this verification test, coagulation pretreatment was employed, but chlorination was not.
Design specifications indicate that the UF system alone has a production capacity up to 250,000 gallons
per day (gpd) from a fresh water source with up to 500 mg/L TDS and a temperature of 25°C. The
combined UF and RO system is designed to produce from 98,000 gpd up to 162,000 gpd, depending on
the TDS of the source water and the recovery settings of the RO process.
VERIFICATION TEST DESCRIPTION
Test Site
The testing site was the Seawater Desalination Test Facility (SDTF) operated by the Naval Facilities
Engineering Service Center (NFESC) at Naval Base Ventura County (NBVC) in Port Hueneme,
California. The source water was from an open ocean intake in the Port of Hueneme, a deep-water port.
The port has no appreciable fresh water outlets; therefore, the water closely resembles that of the Pacific
Ocean salinity.
Initial characterization samples of seawater were collected in April, June and September 2006, and again
in April and August 2007. Highlights of the initial characterization data are presented in Table VS-i. In
addition to the data presented in Table VS-i, nitrite, nitrate, total silica, fluoride, and 29 metals were
analyzed and the concentrations were either below the laboratory reporting limits (not detected) or below
the National Primary Drinking Water Regulations (NPDWR) limits and are presented in the final report.
Samples for many of the metals were analyzed by EPA Method 1640, which achieved detection limits
much lower than Method 200.7 and provided data on seawater that could be compared to the NPDWR.
Table VS-i. Initial Raw Water Characterization Sampling Results
Parameter
Sample Date
04/01/06 06/08/06 09/05/06 04/24/07
pH
Conductivity (^mhos/cm)
TOC (mg/L)
UV254 (I/cm)
TSS (mg/L)
TDS (mg/L)
Alkalinity (mg/L CaCO3)
Total Hardness (mg/L as CaCO3)
Sodium (mg/L)
Heterotrophic Plate Count (CFU/mL)
Total Coliforms (CFU/100 mL)
7.77 7.96 7.8
50,000 50,000 51,100
ND (0.3)
0.016
30
34,000 37,000 35,700
100
6,580
11,000
4
80
NSF 09/29/EPADWCTR
The accompanying notice is an integral part of this verification statement.
VS-ii
October 2009
-------
Methods and Procedures
The U.S Army Tank-Automotive Research, Development, and Engineering Center (TARDEC) conducted
the EUWP test with assistance from the U.S. Bureau of Reclamation (USER). Field testing was
conducted from October 16, 2007 to November 12, 2007. The ETV test protocol calls for testing to run
for 30 days with the intent to operate the equipment until at least one chemical cleaning is performed.
NSF allowed TARDEC to stop testing two days early because over the course of testing, the UF system
was cleaned four times. Per a requirement of the ETV test, a chemical cleaning was performed on the RO
system at the end of the test, although the RO system had not yet reached its cleaning level criteria.
The testing activities followed a test/quality assurance plan (TQAP) prepared for the project. The TQAP
was developed according to ETV Protocols EPA/NSF Protocol for Equipment Verification Testing for
Removal of Inorganic Constituents, dated April 2002, and the EPA/NSF Protocol for Equipment
Verification Testing for Physical Removal of Microbiological and Paniculate Contaminants, dated
September 2005.
Turbidity and conductivity were selected as two key water quality parameters, as turbidity removal by the
system indicated the ability to remove particulate related contaminants, and a reduction in conductivity
(indicator of total dissolved solids content) showed the ability of the RO system to remove dissolved
contaminants. Flow, pressure, conductivity, and temperature recordings were collected twice per day
when possible to quantify membrane flux, specific flux, flux decline, and recovery. Grab sample turbidity
and pH readings were also recorded twice per day when possible. The UF and RO skids also included in-
line turbidimeters for the raw water, UF filtrate, and RO permeate streams. The in-line turbidimeters
recorded measurements every 15 minutes. In addition, the UF skid was equipped with in-line particle
counters that recorded particle counts every five minutes. Pressure decay tests were conducted daily on
the UF system to verify membrane integrity.
Total dissolved solids (TDS) were measured once per day on samples collected from the UF raw water
and the RO process streams and once per week on the UF discharge and RO feed water. Once per week
samples collected from the UF and RO process streams were analyzed for alkalinity, bicarbonate, total
hardness, boron, calcium, chloride, lithium, magnesium, barium, selenium, ortho-phosphate, phosphorus
(total), potassium, sodium, Stiff and Davis Stability Index (S&DSI), sulfate, total suspended solids (TSS),
UV absorbance at 254 nm (UV^s/O, and total coliforms. Samples were collected for Bacillus endospores
once per day from the UF and RO process water.
VERIFICATION OF PERFORMANCE
Finished Water Quality
The UF system reduced turbidity from a mean of 1.34 NTU in the raw water to a mean of 0.06 NTU in
the UF filtrate, as measured by the daily grab samples. This equates to a mean percent reduction of
94.9%. The 95% confidence interval shows that filtrate turbidity can be expected to be in the range of
0.05 to 0.07 NTU. The raw water turbidity, as measured by the in-line analyzer, had a mean value of 1.38
NTU. The in-line turbidity data for the UF filtrate had a mean of 0.019 NTU. The UF filtrate turbidity
levels met the NPDWR of <0.3 NTU 95% of the time and all values below 1.0 NTU throughout the test.
A second turbidity requirement is an action level of 0.15 NTU in the EPA Long Term 2 Enhanced Surface
Water Treatment Rule (LT2ESWTR). This rule states that if the filtrate turbidity exceeds 0.15 NTU over
any 15-minute period, the system must be shut down for a direct integrity test. Since the data logger
recorded turbidity every 15 minutes, the evaluation criteria was two consecutive turbidity measurements
exceeding 0.15 NTU. There were three single data points where the UF filtrate turbidity exceeded 0.15
NTU. In each instance, the previous and following turbidity values were significantly below the 0.15
NTU level. Based on these data and evaluation criteria, it appears that the UF system did not exceed the
LT2ESWTR action level during the verification test. It should be noted that the EUWP was not set up to
NSF 09/29/EPADWCTR The accompanying notice is an integral part of this verification statement. October 2009
VS-iii
-------
be compliant with the LT2ESWTR, as the in-line turbidity meters were not tied to an automatic system
shutdown if the turbidity level exceeded 0.15 NTU for any 15 minute period.
The RO system provided little additional reduction of the turbidity levels, with the RO permeate having a
mean turbidity of 0.05 NTU, based on the grab samples collected each day. The in-line RO permeate
turbidimeter measurements had a mean turbidity of 0.013 NTU. The final treated water, the RO permeate,
also met the NPDWR turbidity requirements. In addition, the RO system produced permeate with
turbidity below the LT2ESWTR action level of 0.15 NTU throughout the test. As with the UF system,
there were only three single RO permeate data points above the action level, and at no time were there
two consecutive 15-minute readings above the action level.
The RO system reduced the dissolved ions in the water, as measured by conductivity by a mean of 99%.
The mean conductivity in the RO permeate was 592 uS/cm, while that for the RO feed was 51,380 uS/cm.
The direct measurements of TDS also show 99% reduction, with the RO permeate in the 280-300 mg/L
range, compared to 34,000-39,000 mg/L in the RO feed. Sodium was reduced by 98% and chloride was
reduced by 99%. These data are consistent with the conductivity data. The other inorganic materials
measured such as hardness, alkalinity, metals, sulfate, and phosphorus were also effectively reduced in
the RO permeate.
The UF system had no impact on the pH of the water with the feed water having a mean pH of 7.78 and
the filtrate having a mean pH of 7.73. The RO system did lower the pH, the permeate having a mean pH
of6.29.
UF Membrane Integrity
Pressure decay tests, microorganism reduction, and particle counts were used to document UF membrane
integrity. Bacillus endospores and total coliforms were measured in the feed and filtrate to provide data
on the microbial reduction achieved by the UF system. In-line analyzers also collected particle count data
from the feed and filtrate streams as an additional indicator of membrane integrity and the capability of
the system to remove particulate and microbial contaminants.
Pressure decay tests on the UF system were performed on most operating days during the verification test.
The mean pressure decay rates ranged from 0.02 to 0.15 psig/min. The overall mean pressure decay rate
was 0.08 psig/min. These direct integrity test results were indicative of membrane modules with no
significant observable breaches.
The particle counters recorded the particle counts in the UF feed and UF filtrate every five minutes and
stored the data for transfer to a personal computer. The mean 2-3 (im particle count for the feed water was
5,559/mL, with a range of 53-17,843/mL. The UF filtrate had a mean 2-3 (im particle count of 42/mL,
with a range of 0-773/mL. The UF system reduced the 2-3 (im particles by a mean value of 2.3 Iogi0.
However, the maximum particle count of 773/mL may not be indicative of the typical UF separation
performance. The UF system went through a backflush cycle every half-hour, and during these
backflushes the particle counts were still being recorded. Consequently, the filtrate particle count data
included numerous spikes. The backflushes were not time-stamped, so the spikes due to backflushes
could not be identified with certainty and removed from the data set.
The mean 3-5 (im particle count for the UF feed was 3,616/mL, with a range of 1,355-9,505/mL. The
filtrate had a mean 3-5 (im particle count of 22/mL, with a range of 1-352/mL. Again, spikes due to
backflushes could not be identified with certainty. The UF system reduced the 3-5 (im particles by a mean
value of 2.5 logic.
NSF 09/29/EPADWCTR The accompanying notice is an integral part of this verification statement. October 2009
VS-iv
-------
Bacillus endospores and total coliform levels in the seawater were low during the test, with geometric
mean concentrations of 64 CFU/100 ml and 10 CFU/100 ml, respectively. The UF system reduced the
Bacillus endospores to a geometric mean of 1.3 CFU/100 ml. The UF filtrate endospores counts were 1 or
<1 CFU/lOOmL on all but two days. No total coliforms were found in any UF filtrate samples.
UF System Operation
The UF system performance operations data for the test are presented in Table VS-ii. The intake flow is
the intake from the source water into the UF feed water tank.
Table VS-ii. UF System Operations Data
95%
Standard Confidence
Parameter Count Mean Median Minimum Maximum Deviation Interval
UF Operation per day (hr) 19 18.6 19.8 7.3 22.7 4.11 +1.85
Intake Flow (gpm) 74 287 288 272 296 4.98 +1.13
Feed Flow (gpm) 74 249 251 212 279 11.4 +2.60
Filtrate Flow (gpm) 74 222 225 187 252 10.9 +2.48
Retentate Flow (gpm) 74 26 26 25 34 1.66 +0.38
Backwash Flow (gpm) 900 gallons per backwash cycle*; Backwash every 30 minutes
Feed Pressure (psig) 74 20.6 20.0 14.0 30.0 3.74 +0.85
Retentate Pressure (psig) 74 16.3 16.0 10.0 23.0 2.89 +0.66
Filtrate Temperature (°F) 74 58.3 59.0 55.0 61.0 1.62 +0.37
*Volume not measured. It was provided by the manufacturer.
The mean UF feed water flow was 249 gpm. The UF water recovery was 89.2% based on the mean feed
water and filtrate flows. The net UF filtrate production over the 28 calendar-day test period (27 - 24 hour
periods) was 4,673 kilogallons (kgal), which represents an average production rate of 173.1 kgal/day. The
total UF filtrate volume (including filtrate used for backwash) produced was 5,249 kgal, which gives an
average total production rate of 194.4 kgal/day. This production rate includes the two days when the UF
was not operated as part of the cleaning cycle and includes other days with limited production due to
cleaning or system maintenance issues.
A chemical coagulant (ferric chloride) was added to the UF feed water to improve operation of the UF
system and to lengthen run time between chemical cleanings. The coagulant addition was planned for a
feed rate of 4.37 ml/min, which would yield an iron dose (as Fe) of 0.75 mg/L in the UF feed water (4.6 x
10"6 gallons of ferric per gallon of feed water). Based on the tank records, a total of 22.4 gallons of ferric
chloride were fed into 5,259,625 gallons of feed water (4.3 x 10"6 gallons of ferric per gallon of feed
water), which is approximately 10% less than the feed rate measured by the pump calibration.
RO System Operation
The RO system operations data for the test are presented in Table VS-iii. The mean feed water flows of
115 gpm for Array 1 and 63 gpm for Array 2 were very close to the target feed rates established in the test
plan (Array 1 target 116 gpm and Array 2 target was 58 gpm) to achieve an overall RO target flowrate of
100,000 gpd. The Array 1 recovery of 61% exceeded the target specification of 50%. The Array 2
recovery of 50% also exceeded the target specification of 48%. These recoveries, in conjunction with the
feed water targets, resulted in mean permeate flow rates of 70 gpm for Array 1 and 32 gpm for Array 2.
At these flows, the RO unit would need to operate an average of approximately 16.3 hours/day to meet
the target of 100,000 gpd.
NSF 09/29/EPADWCTR The accompanying notice is an integral part of this verification statement. October 2009
VS-v
-------
Table VS-iii. RO System Operations Data
95%
Standard Confidence
Parameter Count Mean Median Minimum Maximum Deviation Interval
Array 1 Feed Flow (gpm) 74 115 115 112 117 O74 ±0.17
Array 1 Permeate Flow (gpm) 74 70 70 68 72 0.82 ±0.19
Array 1 Concentrate Flow (gpm) 74 45 45 43 48 1.03 ±0.23
Array 2 Feed Flow (gpm) 74 63 63 56 68 2.05 ±0.47
Array 2 Permeate Flow (gpm) 74 32 32 25 37 2.11 ±0.46
Array 2 Concentrate Flow (gpm) 74 31 31 30 32 0.36 ±0.08
Array 1 Feed Pressure (psig) 74 954 960 860 977 19.5 ±4.44
Array 1 Concentrate Pressure (psig) 74 905 903 870 992 15.5 ±3.53
Array 2 Feed Pressure (psig) 74 902 900 880 995 15.4 ±3.51
Array 2 Concentrate Pressure (psig) 74 868 865 850 885 7.65 ±1.74
Array 1 and 2 Combined Permeate 74 23.4 23.5 21.0 28.5 1.34 ±0.31
Pressure (psig)
Over the 28 calendar-day (27 24-hour periods) verification test, the RO feed water totalizer showed 4,673
kgal of water was fed to the RO unit. Based on the daily percent recoveries for each array (typically Array
1 at 61% and Array 2 at 50%), the total volume of permeate produced was approximately 2,671 kgal,
giving an average of 98.9 kgal/day over the 28-day test.
The primary reason the RO system did not achieve or exceed the production goal of 100 kgal/day was a
lack of feed water when the UF system was shut down for cleaning. The UF system also shutdown
anytime the RO system feed water tank was full. The test was designed to evaluate the entire system with
both UF and RO in operation. The UF system produced enough water to meet the 100 kgal/day
production goal; however, because of limited UF filtrate storage capacity, long downtime periods for the
UF system cleaning did impact the RO production. With more storage capacity for UF filtrate, the UF
system would have been able to meet the feed requirements for the RO system to achieve the overall goal
of producing 100 kgal/day, even with the more frequent cleaning schedule. Whenever, there was feed
available, the RO system operated continuously producing permeate at a flow rate of 100 to 102 gpm. The
RO system operated greater than 20 hours on 12 of the 25 actual operating days. During those days, when
the UF was also operating most hours of the day, the RO system did meet and exceed the target
production rate. The RO mean operating hours were 17.0 hours/day with a median of 19.0 hrs/day. These
mean and median hours match closely to the UF hours (mean - 16.9 hrs and median 19.1 hrs). The
maximum RO operating hours were 24 hours and the minimum was 4 hours.
Antiscalant was added to the RO feed water throughout the test. The mean dose rate was 5.7 mg/L versus
a target feed of 5 mg/L. The RO system did not appear to experience any scaling or fouling problems
during the test. The S&DSI varied from -0.71 to -0.84 during the test. This indicates that the concentrate
was a non-scaling water (S&DSI <0.0 is non-scaling). The combination of non-scaling water and the
addition of antiscalant reduced or eliminated the problems of scaling on the RO membranes.
The system operated consistently throughout the test with little change in flows or pressures. This would
suggest that for this source, the RO could have met and exceeded production targets, if sufficient water
could have been provided from the UF system. The buildup of solids on the UF system and need for
frequent UF system cleaning was the limiting factor over the test period.
NSF 09/29/EPADWCTR The accompanying notice is an integral part of this verification statement. October 2009
VS-vi
-------
The RO system specific flux was consistent over the test period and indicates that the RO membranes
were not being fouled over time. The membranes were still functioning at the end of the test at a specific
flux that was 97% of the starting specific flux; therefore, it cannot be projected when the membranes
would require cleaning. The RO system was chemically cleaned in place on November 13 and 14, 2007 at
the end of the test. This cleaning was performed because it was a requirement of the verification test to
demonstrate the cleaning process; however the RO system had not actually reached its target cleaning
level criteria.
QUALITY ASSURANCE/QUALITY CONTROL (QA/QC)
NSF provided technical and quality assurance oversight of the verification testing as described in the
verification report, including a review of 100% of the data. NSF QA personnel also conducted a technical
systems audit during testing to ensure the testing was in compliance with the test plan. A complete
description of the QA/QC procedures is provided in the verification report.
Original signed by Sally Gutierrez 08/12/10 Original signed by Robert Ferguson 05/18/10
Sally Gutierrez Date Robert Ferguson Date
Director Vice President
National Risk Management Research Water Systems
Laboratory NSF International
Office of Research and Development
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 an NSF Certification of the specific product mentioned
herein.
Availability of Supporting Documents
Copies of the test protocol, the verification statement, and the verification report (NSF
report # NSF 09/29/EPADWCTR) are available from the following sources:
1. ETV Drinking Water Systems Center Manager (order hard copy)
NSF International
P.O. Box 130140
Ann Arbor, Michigan 48113-0140
2. Electronic PDF copy
NSF web site: http://www.nsf.org/info/etv
EPA web site: http://www.epa.gov/etv
NSF 09/29/EPADWCTR The accompanying notice is an integral part of this verification statement. October 2009
VS-vii
-------
This page is intentionally blank
-------
October 2009
Environmental Technology Verification Report
Removal of Dissolved Salts and Particulate Contaminants from Seawater
Village Marine Tec.
Expeditionary Unit Water Purifier, Generation 1
Prepared by:
Michael Blumenstein, Kristie Wilhelm, and C. Bruce Bartley NSF International, Ann Arbor, MI
Dale Scherger, Scherger and Associates, Ann Arbor, MI
Michelle Chapman, United Stated Bureau of Reclamation, Denver, CO
Jeffrey Q. Adams, Project Officer, U.S. Environmental Protection Agency, Cincinnati, OH
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
-------
Notice
The U.S. Environmental Protection Agency, through its Office of Research and Development,
funded and managed, or partially funded and collaborated in, the research described herein. It
has been subjected to the Agency's peer and administrative review and has been approved for
publication. Any opinions expressed in this report are those of the author (s) and do not
necessarily reflect the views of the Agency, therefore, no official endorsement should be inferred.
Any mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
11
-------
Foreword
The EPA is charged by Congress with protecting the nation's air, water, and land 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, the EPA's Office of Research and
Development provides data and science support that can be used to solve environmental
problems and to build the scientific knowledge base needed to manage our ecological resources
wisely, to understand how pollutants affect our health, and to prevent or reduce environmental
risks.
The Environmental Technology Verification (ETV) Program has been established by the EPA to
verify the performance characteristics of innovative environmental technology across all media
and to report this objective information to permitters, buyers, and users of the technology, thus
substantially accelerating the entrance of new environmental technologies into the marketplace.
Verification organizations oversee and report verification activities based on testing and quality
assurance protocols developed with input from major stakeholders and customer groups
associated with the technology area. ETV consists of six environmental technology centers.
Information about each of these centers can be found on the internet at http://www.epa.gov/etv.
Under a cooperative agreement, NSF International has received EPA funding to plan, coordinate,
and conduct technology verification studies for the ETV "Drinking Water Systems Center" and
report the results to the community at large. The DWS Center has targeted drinking water
concerns such as arsenic reduction, microbiological contaminants, particulate removal,
disinfection by-products, radionuclides, and numerous chemical contaminants. Information
concerning specific environmental technology areas can be found on the internet at
http ://www. epa.gov/nrmrl/std/etv/verifications.html.
in
-------
Table of Contents
Verification Statement VS-i
Notice ii
Foreword iii
Table of Contents iv
Abbreviations and Acronyms ix
Abbreviations and Acronyms (continued) x
Acknowledgements xi
Chapter 1 Introduction 1
1.1 ETV Purpose and Program Operation 1
1.2 Testing Participants and Responsibilities 1
1.2.1 EPA 2
1.2.2 NSF International 2
1.2.3 ONR 3
1.2.4 TARDEC 3
1.2.5 USER 3
1.2.6 Village Marine Tec 4
1.3 Verification Testing Site 4
Chapter 2 Equipment Capabilities and Description 5
2.1 Equipment Capabilities 5
2.2 General System Description 6
2.3 Concept of Treatment Processes 8
2.3.1 UF Pretreatment/Suspended Solids Filtration 8
2.3.2 RO Desalination 9
2.4 Detailed System Description 9
2.4.1 Raw Water Intake 12
2.4.2 UF System Description 12
2.4.2.1 UF System Operation 14
2.4.2.2 UF Cleaning Procedure 15
2.4.3 RO System 17
2.4.3.1 RO skid statistics 20
2.4.3.2 RO System Operation 20
2.4.3.3 RO Cleaning Procedure 21
2.4.3.4 Pressure Exchanger 24
2.5 General Requirements and Limitations 25
2.6 Waste Generation and Permits 27
2.6.1 UFCIP 27
2.6.2 ROCIP 27
2.6.3 RO Concentrate 27
2.6.4 UF Backwash and Retentate 28
2.6.5 Discharge Permits 28
2.7 Discussion of the Operator Requirements 28
Chapter 3 Methods and Procedures 30
3.1 Quantitative and Qualitative Evaluation Criteria 30
3.2 Key Treated Water Quality and Operational Parameters 30
iv
-------
3.3 Operations and Maintenance 32
3.4 Field Operations 32
3.5 Overview of ETV Testing Plan 32
3.5.1 Task A: Characterization of Feed Water 33
3.5.2 Task B: Equipment Installation, Initial Test Runs, and Initial System Integrity
Tests 33
3.5.3 Task C: Verification Test 33
3.5.3.1 Task Cl: Membrane Flux and Recovery 33
3.5.3.2 TaskC2: Cleaning Efficiency 33
3.5.3.3 TaskC3: Finished Water Quality 33
3.5.3.4 Task C4: Membrane Module Integrity 34
3.5.3.5 TaskCS: Data Handling Protocol 34
3.5.3.6 Task C6: Quality Assurance and Quality Control 34
3.6 Task A: Characterization ofFeed Water 34
3.7 Task B: Equipment Installation, Initial Test Runs, and Initial System Integrity Tests34
3.8 Task C: Verification Testing 34
3.8.1 Task Cl: Membrane Flux and Operation 34
3.8.1.1 Work Plan 35
3.8.1.2 Evaluation Criteria 35
3.8.1.3 Equations 36
3.8.2 TaskC2: Cleaning Efficiency 40
3.8.2.1 Work Plan 40
3.8.2.2 Evaluation Criteria 40
3.8.3 TaskC3: Finished Water Quality 41
3.8.3.1 Work Plan 41
3.8.3.2 Evaluation Criteria 42
3.8.4 Task C4: Membrane Integrity Testing 42
3.8.4.1 Direct Integrity Testing: 42
3.8.4.2 Continuous Indirect Integrity Monitoring: 42
3.8.5 TaskCS: Data Handling Protocol 42
3.8.5.1 Work Plan 42
3.8.6 Task C6: Quality Assurance Project PI an 43
3.8.6.1 Experimental Objectives 43
3.8.6.2 Work Plan 43
3.8.6.3 QA/QC Verifications 43
3.8.6.4 Data Correctness 44
3.8.6.5 Operation and Maintenance 49
Chapter 4 Results and Discussion 50
4.1 Introduction 50
4.2 Equipment Installation, Start-up, and Shakedown 50
4.3 Task A: Raw Water Characterization 50
4.4 Task B: Equipment Installation, Initial Test Runs and Initial System Integrity Tests. 52
4.5 TaskC: Verification Test 55
4.5.1 Task Cl: Membrane Flux and Operation 55
4.5.1.1 UF Operating Data 56
4.5.1.2 RO System Operational Data 62
-------
4.5.2 TaskC2: Cleaning Efficiency 68
4.5.2.1 UF Backwash and Cleaning Frequency and Performance 68
4.5.2.2 RO Cleaning Frequency and Performance 69
4.5.3 Task C3: Water Quality Results 70
4.5.3.1 Water Quality Results - Turbidity, Conductivity, pH, and Temperature 71
4.5.3.2 Water Quality Results - Other Water Quality Parameters 84
4.5.3.3 Total Organic Carbon Results for Cleaning Solution 88
4.5.4 Task C4: Membrane Module Integrity 88
4.5.4.1 UF System Pressure Decay Results 89
4.5.4.2 Bacillus Endospores and Total Coliform Results 92
4.5.4.3 UF System Particle Count Data 93
4.6 Chemical Consumption 96
4.7 Quality Assurance/Quality Control 97
4.7.1 Introduction 97
4.7.2 Documentation 98
4.7.3 Quality Audits 98
4.7.4 Test Procedure QA/QC 98
4.7.5 Sample Handling 98
4.7.6 Physical and Chemical Analytical Methods QA/QC 99
4.7.7 Microbiology Laboratory QA/QC 99
4.7.7.1 Growth Media Positive Controls 99
4.7.7.2 Negative Controls 99
4.7.8 Laboratory Documentation 99
4.7.9 Data Review 99
4.7.10 Data Quality Indicators 99
4.7.10.1 Representativeness 100
4.7.10.2 Accuracy 100
4.7.10.3 Precision 100
4.7.10.4 Completeness 101
List of Figures
Figure 1-1. Photo of the concrete pad used for EUWP testing 4
Figure 2-1. Process component diagram 7
Figure 2-2. Koch UF hollow fiber modules, a single fiber, and the process flow through the
module 8
Figure 2-3. EUWP system process schematic 10
Figure 2-4. Schematic of typical EUWP layout 11
Figure 2-5. Photo of the UF skid 13
Figure 2-6. Photo of the UF cartridges mounted in the UF skid 14
Figure 2-7. Piping and instrumentation diagram of UF skid 16
Figure 2-8. Photo of the RO skid 17
Figure 2-9. Photo of the RO skid membrane vessels 18
Figure 2-10. Vessel arrangement schematic 18
Figure 2-11. Membrane arrangement schematic 19
VI
-------
Figure 2-12. P&ID of RO skid 23
Figure 2-13. PX pressure exchanger 24
Figure 4-1. Plot of UF system flow rates through the testing period 57
Figure 4-2. UF system filtrate production through the testing period 58
Figure 4-3. Plot of UF system feed and retentate pressures over the testing period 59
Figure 4-4. Plot of UF system TMP over the testing period 59
Figure 4-5. UF system specific flux over testing period 61
Figure 4-6. Change in specific flux overtime 61
Figure 4-7. Diesel fuel consumption 62
Figure 4-8. RO system flow rates 64
Figure 4-9. RO system operating pressures 64
Figure 4-10. RO system percent recoveries 65
Figure 4-11. RO system specific flux 67
Figure 4-12. Grab sample UF feed turbidity data 72
Figure 4-13. Grab sample UF filtrate turbidity data 73
Figure 4-14. UF feed andUF filtrate in-line turbidity readings 73
Figure 4-15. RO conductivity results 78
Figure 4-16. RO permeate conductivity readings from in-line meter 79
Figure 4-16. Pressure decay overtime 90
Figure 4-17. Particle count hourly averages - 2-3 jim 94
Figure 4-18. UF filtrate 2-3 jim particle count size distribution 95
Figure 4-19. Particle count hourly averages-3-5 jim 96
List of Tables
Table 2-1. Koch Membrane Systems Targa 10-48-3 5-PMC Cartridge Specifications 12
Table 2-2. UF Skid Statistics 13
Table 2-3. RO System Membrane Element Characteristics 19
Table 2-4. RO Skid Statistics 20
Table 2-5. EUWP Site Considerations and Dimensions 25
Table 2-6. Equipment Limitations 26
Table 2-7. RO Membrane Limitations 27
Table 3-1. Key Treated Water Quality Parameters 31
Table 3-2. Water Quality and Operational Parameters Measured In-Line 31
Table 3-3. Operational Parameter Sampling Locations 35
Table 3-4. Key Operating Parameters 35
Table 3-5. Operational Data Plots Appearing in Chapter 4 36
Table 3-6. Water Quality Sampling Schedule 41
Table 3-7. On-Site Analytical Equpment QA Activities 44
Table 3-8. On-Site Data Generation QC Activities 44
Table 3-9. Analytical Methods for Laboratory Analyses 46
Table 3-10. Accuracy and Precision Limits for Laboratory Analyses 48
Table 3-11. Completeness Requirements 49
Table 4-1. Initial Raw Water Characterization Sampling Results 51
Table 4-2. Results of Low Pressure Integrity Test on Individual UF Cartridges 53
vn
-------
Table 4-3. October 10, 2007 UF Full System Integrity Test Results 54
Table 4-4. UF Operational Data Statistics 56
Table 4-5. RO System Operational Measurement Statistics 63
Table 4-6. UF System CIP Cleaning Solution - Chemical Use 69
Table 4-7. RO System Specific Flux Before and After CIP 70
Table 4-8. Turbidity Results, On-Site Bench Top 74
Table 4-9. In-Line Turbidity Measurement Statistics 75
Table 4-10. Conductivity Results, On-Site Benchtop 76
Table 4-11. pH Results 80
Table 4-12. Temperature Results 82
Table 4-13. Other UF System Water Quality Data 85
Table 4-14. Cleaning Solution TOC Results 88
Table 4-15. Pressure Decay Data 91
Table 4-16. Bacillus Endospore Counts and Log Reduction Calculations 92
Table 4-17. Total Coliform Counts and Log Reduction Calculations 93
Appendices
Appendix A - Operation and Maintenance Manual
Appendix B - Field Log Sheets and Calibration Records
Appendix C - Operation Data Spreadsheets
Appendix D - In-line Turbidity and RO Permeate Conductivity Data
Appendix E - In-line Particle Count Data
Appendix F - NSF Laboratory Data Reports and Sample Chain of Custody Forms
Appendix G - Duplicate Analysis Results
Vlll
-------
Abbreviations and Acronyms
ANGB Air National Guard Base
BOD Biochemical Oxygen Demand
°C degrees Celsius
CPU colony-forming unit
CIP clean-in-place
cm centimeter
DF2 diesel fuel, grade 2
DFA diesel fuel, arctic grade
DWSC Drinking Water Systems Center
EPA United States Environmental Protection Agency
ETV Environmental Technology Verification
EUWP Expeditionary Unit Water Purifier
°F degrees Fahrenheit
FRP fiberglass reinforced plastic
ft foot (feet)
gal gallons
gfd gallons per foot per day
gpd gallons per day
gpm gallons per minute
h hour
HPC Heterotrophic plate count
in inch
JP8 j et propellent 8 (j et fuel)
kgal kilogallon
kW kilowatt
kWh kilowatt hour
L liter
Ibs pounds
LT2ESWTR Long Term 2 Enhanced Surface Water Treatment Rule
m meter
mg milligram
mL milliliter
mS milliSiemens
MWCO molecular weight cutoff
NBC nuclear, biological, and chemical
ND non-detect
NDP net driving pressure
NFESC Naval Facilities Engineering Service Center
NIST National Institute of Standards and Technology
NM not measured
NPDWR National Primary Drinking Water Regulations
NRMRL National Risk Management Research Laboratory
NSF NSF International (previously known as the National Sanitation Foundation)
NSWCCD United States Naval Surface Warfare Center - Carderock Division
IX
-------
Abbreviations and Acronyms (continued)
NTU Nephelometric turbidity units
NBVC Naval Base Ventura County
O&M operations and maintenance
ONR Office of Naval Research
ORD Office of Research and Development
P&ID piping and instrumentation diagram
PE performance evaluation
PLC programmable logic controller
ppm parts per million
psi pounds per square inch
psig pounds per square inch, gauge
PVC polyvinyl chloride
PX pressure exchanger
QA/QC Quality Assurance/Quality Control
QAPP Quality Assurance Project Plan
RO reverse osmosis
RPD relative percent difference
RPM revolutions per minute
S&DSI Stiff and Davis Stability Index
SDI Silt Density Index
SDTF Seawater Desalination Test Facility
SM Standard Methods for the Examination of Water and Wastewater
SNL Sandia National Laboratories
TARDEC Tank-Automotive Research, Development, and Engineering Center
TDS total dissolved solids
TOC total organic carbon
TQAP test/quality assurance plan
TQG Tactical Quiet Generator
TSS total suspended solids
TMP transmembrane pressure
UF ultrafiltration
USER United States Bureau of Reclamation
UV254 ultraviolet absorbance at 254 nanometers
VOC volatile organic chemicals
|im micron
|iS microSiemens
-------
Acknowledgements
The U.S. Army Tank-Automotive Research, Development, and Engineering Center (TARDEC)
was the main field testing organization. TARDEC was supported by the U.S. Bureau of
Reclamation (USER). TARDEC and USER were responsible for all elements of the tests,
including operation of the equipment, collection of samples, instrument calibration, and data
collection.
This verification report was authored by Mr. Dale Scherger of Scherger and Associates (3017
Rumsey Drive, Ann Arbor, MI 48105), Mr. Michael Blumenstein, Ms. Kristie Wilhelm, and Mr.
C. Bruce Bartley of the NSF International ETV Drinking Water Systems Center (DWSC), Ms.
Michelle Chapman of USER, and Mr. Jeffrey Q. Adams of USEPA. The verification report was
based on the project test/quality assurance plan authored by DWSC, USER, and TARDEC.
The laboratory selected for the analytical work was:
NSF International Chemistry Laboratory
789 N. Dixboro Road
Ann Arbor, Michigan 48105
Contact: Mr. Kurt Kneen
The manufacturer of the EUWP was:
Village Marine Tec.
2000 W. 13 5th St.
Gardena, CA 90249
Phone: 310-516-9911
The engineers responsible for the daily operations of the field test were Mr. Mark Miller, Mr.
Abel Juarez, Mr. Daniel Gonzalez, Mr. Andrew Tiffenbach, and Mr. Micah Ing. The USER
support staff included Ms. Michelle Chapman and Mr. Steve Dundorf
The NSF DWSC project manager was Mr. Michael Blumenstein. The DWSC is managed by
Mr. Bruce Bartley. Ms. Kristie Wilhelm of the DWSC provided valuable assistance with report
preparation.
XI
-------
This page is intentionally blank
-------
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 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; with 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 responsive to the needs of stakeholders, conducting field
demonstrations, 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) under the ETV Drinking Water Systems
Center (DWSC) to verify the performance of small drinking water systems that serve small
communities. 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 meets 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 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 DWSC evaluated the performance of the Village Marine Tec. Generation 1 Expeditionary
Unit Water Purifier (EUWP). The EUWP, developed for the U.S. Military, uses ultrafiltration
(UF) and reverse osmosis (RO) to produce drinking water from a variety of sources. This
document provides the verification test results for the EUWP system using seawater at Naval
Base Ventura County (NBVC) in Port Hueneme, California.
1.2 Testing Participants and Responsibilities
EUWP design, construction, and testing was overseen by a federal multi-agency team composed
of representatives from Office of Naval Research (ONR); Army Tank-Automotive Research,
Development, and Engineering Center (TARDEC); Naval Surface Warfare Command -
Carderock Division (NSWCCD); United States Department of Interior Bureau of Reclamation
-------
(USER); and Sandia National Laboratories (SNL). The manufacturer, Village Marine Tec., was
contracted to design and build the EUWP to the team's Generation 1 specifications using 2004
state-of-the-art technology.
The organizations involved in the verification testing project were:
• EPA
• NSF
• ONR
• TARDEC
• USER
• Village Marine Tec.
The following is a brief description of all of the ETV participants and their roles and
responsibilities.
1.2.1 EPA
EPA, through its Office of Research and Development (ORD), has financially supported and
collaborated with NSF under Cooperative Agreements R-82833301 and CR833980. This
verification effort was supported by the DWSC operating under the ETV Program. This
document has been peer-reviewed, reviewed by USEPA, and recommended for public release.
1.2.2 NSF International
NSF is an independent, not-for-profit testing and certification organization dedicated to public
health and safety and to 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 products bearing the NSF Name, Logo and/or Mark meet those standards. The
EPA partnered with NSF to verify the performance of drinking water treatment systems through
the EPA's ETV Program.
NSF authored the test plan and test report. NSF also served as the analytical laboratory for all
water quality parameters not measured in the field. NSF also provided technical oversight
during testing and conducted an audit of the field testing activities.
Contact Information:
NSF International
789 N. Dixboro Road
Ann Arbor, Michigan 48105
Contact: Mr. Bruce Bartley, Project Manager
Phone: (734)769-8010
Fax: (734)769-0109
Email: bartley@nsf.org
-------
1.2.3 ONR
The U.S. Navy ONR provided oversight of the EUWP development program which involved
developing high productivity water treatment units for land and shipboard military and civilian
emergency preparedness applications. ONR also provided funding for the EUWP ETV testing
proj ect.
Contact Information:
Office of Naval Research
Logistics Thrust Program
Operations Technology Division
800 N. Quincy St.
Arlington, VA 22217
Contact: Major Alan Stocks
Phone: 703-696-2561
Email: stocksa@onr.navy.mil
1.2.4 TARDEC
TARDEC served as the FTO for this verification. TARDEC was responsible for all on-site
testing activities, including operation of the test equipment, collection of samples, measurement
of water quality parameters, calibration and check of instrumentation, and operational data
collection.
Contact Information:
U.S. Army TARDEC/RDECOM
AMSRD-TAR-D/210, MS 110
6501 E. Eleven Mile Road
Warren, MI 48397
Contact: Mr. Bob Shalewitz, TARDEC EUWP Program Manager
Phone: 586-574-4128
Email: bob.shalewitz@us.army.mil
1.2.5 USER
USER functioned as a co-FTO, providing field operations support, and technical support for
equipment operation.
Contact Information:
U.S. Bureau of Reclamation
Denver Federal Center (D-8230)
P.O. Box 25007
Denver, CO 80225
Contact: Ms. Michelle Chapman
Phone: 303-445-2264
Email: mchapman@do.usbr.gov
-------
1.2.6 Village Marine Tec.
The EUWP manufacturer was Village Marine Tec. The manufacturer was responsible for
supplying a field-ready treatment system equipped with all necessary components, including
instrumentation and controls, and an operation and maintenance (O&M) manual. The
manufacturer was responsible for providing logistical and technical support, as needed, as well
as technical assistance to the FTO during operation and monitoring of the equipment undergoing
field verification testing.
Contact Information:
Village Marine Tec.
2000 W. 13 5th St.
Gardena, CA 90249
Phone: 310-516-9911
Email: sales@villagemarine.com
1.3 Verification Testing Site
The EUWP was tested at the Seawater Desalination Test Facility (SDTF) operated by the Naval
Facilities Engineering Service Center (NFESC) at NBVC in Port Hueneme, California. Port
Hueneme is located on the coast of California approximately 60 miles northwest of Los Angeles.
Raw seawater directly from the Port of Hueneme was used for ETV testing. The Port of
Hueneme is the only deep water port between Los Angeles and San Francisco. It has no
appreciable fresh water outlets; therefore the water closely resembles that of the Pacific Ocean
with respect to salinity. Average water temperature ranges from 55°F in the winter months to
approximately 62°F in the summer. The source water chemistry was profiled for the initial water
characterization task. The water chemistry data is presented in Section 4.3.
The EUWP was situated on a concrete pad at the SDTF as shown in Figure 1-1.
Figure 1-1. Photo of the concrete pad used for EUWP testing.
-------
Chapter 2
Equipment Capabilities and Description
The EUWP was designed to meet purified water needs in areas with challenging water sources of
very high total dissolved solids (IDS), turbidity, or hazardous contamination during emergency
situations when other water treatment facilities are incapacitated. The system uses UF and RO to
produce potable water. It is not intended to meet general municipal water treatment needs in a
cost effective manner. The design requirements - to produce 100,000 gal per day (gpd) and be
C-130 transportable - forced the use of lightweight durable materials, such as titanium, that are
more costly and would not usually be required for municipal water treatment. The requirements
to treat source water with up to 60,000 milligrams per liter (mg/L) TDS and ensure removal of
nuclear, biological, and chemical (NBC) contaminants to a safe limit, drove the design to two
parallel arrays - with a 2nd permeate pass resulting in a maximum of 65% recovery. Most
municipal water treatment systems can easily attain much higher recovery levels. The EUWP is
also intended as a demonstration of the state-of-the-art of desalination for emergency situations.
Key innovations applied in the EUWP are:
• High flux UF membrane cartridges;
• Innovative staging of RO membrane modules; and
• Small system energy recovery to pressurize a parallel array.
The EUWP was developed to meet the following objectives:
• Develop a high capacity drinking water purification unit to provide strategic water
production capability with a focus on peacekeeping, humanitarian aid, and disaster relief
missions that the military frequently supports.
• Further the state of desalination technology with a view toward reduced operational costs,
size, and weight; improved reliability; and verifying emerging technologies.
2.1 Equipment Capabilities
The objective of this verification test was to document the ability of the EUWP to meet the
following performance criteria:
The EUWP is capable of producing 100,000 gpd of water meeting EPA's National
Primary Drinking Water Regulations (NPDWR) from raw Port Hueneme sea water based
on contaminants found in the source water during the initial water characterization
phase ofETV testing.
The EUWP is intended to meet purified water needs in areas with challenging water sources of
very high TDS, turbidity, or hazardous contamination during emergency situations when other
water treatment facilities are incapacitated. The unit was designed to meet or exceed Tri-Service
Field Water Quality Standards for short-term consumption by healthy adults. However, the
technology used is capable of exceeding the EPA NPDWR.
-------
The EUWP, using the UF system only, can produce up to 250,000 gpd of potable water from a
fresh water source with up to 500 mg/L IDS and a temperature of 77 degrees Fahrenheit (°F) (25
degrees Celsius, or °C), provided that contaminants not removed by UF are not present in the
source water. Using the UF and RO system, it is designed to produce from 98,000 gpd up to
162,000 gpd depending on the TDS of the source water and the recovery settings of the RO
system. Production is decreased to 125,000 gpd (50% recovery) for higher TDS waters. It can
also produce 98,000 gpd from a NBC contaminated source with up to 45,000 mg/L TDS. NBC
contaminant removal was not verified as part of the ETV test at Port.
2.2 General System Description
• Equipment name: Expeditionary Unit Water Purifier (EUWP)
• Model number: Generation 1
• Manufacturer: Village Marine Tec., 2000 W. 135th St., Gardena, CA 90249, (310) 324-
4156.
• Power requirements: 480 volts, 250 Amp, 60 hertz, 3-phase electrical, or two 60 kilowatt
(kW) diesel Tactical Quiet Generators (TQG).
- UF Requirements - 125 amps maximum
- RO Requirements - 125 amps maximum
The EUWP is composed of feed pumps, a UF pretreatment system, a 1 or 2 pass RO desalination
system with energy recovery, storage tanks, and product pumps (Figure 2-1). It has chemical
feed systems for pretreatment and post treatment. Clean-in-place (CIP) systems are included
with the skids.
-------
UF Product Tank
RO Skid
—©•
200 urn
Figure 2-1. Process component diagram.
UF Skid Waste
L
-©—*•
200 urn
Waste UF Hollow Fiber
Membranes
2nd Pass
Feed Tank
RO Product Tank
-©—*-
-------
2.3 Concept of Treatment Processes
2.3.1 UF Pretreatment/Suspended Solids Filtration
UF is a low-pressure (5-90 pounds per square inch, gauge, or psig) membrane process that
separates particulates based on size exclusion. The UF process retains oils, paniculate
matter, bacteria, and suspended solids that contribute to turbidity and a high silt density index
(SDI). Feed water to RO systems should have turbidity less than 0.1 Nephelometric
Turbidity Units (NTU) and a SDI less than 3. UF membranes pass water, dissolved salts, and
most dissolved organic compounds. UF pore sizes range from 0.002 to 0.1 micron (|im)
(1,000-500,000 molecular weight cutoff, or MWCO). Koch Membrane Systems Targa-10
hollow fiber UF membranes are used in the EUWP. Water flows from the inside of the fiber
to the outside causing suspended solids to collect on the inside of the fiber. Periodically, the
system must be vigorously backwashed to remove this material from the system. Figure 2-2
shows example UF cartridges, a single fiber, and the flow pattern used in this system.
The key operating parameters for a UF system are the instantaneous flux and the overall
productivity taking into account the volume required for backwash. Generally, the higher the
instantaneous flux, the more often backwashing will be required. There is an optimum flux
point where overall productivity is maximized, called the critical flux. For municipal
systems, it is economical to operate the system at the critical flux. The EUWP is an
emergency supply system with extreme weight restrictions to enable transport. The weight
restrictions drove design of the UF system to operate at a maximum flux with more frequent
backwashes.
Process
Feed Flow
Hollow
Fiber Membranes
Figure 2-2. Koch UF hollow fiber modules, a single fiber, and the process flow through
the module.
-------
2.3.2 RO Desalination
Dissolved salts and larger molecular weight organic molecules can be removed by RO.
Osmosis is a naturally occurring phenomenon in which pure water is transported down a
chemical potential gradient across a semi-permeable membrane from a low concentration
solution to a high concentration solution. One measure of the chemical potential is the
osmotic pressure. Osmotic pressure is dependent on the concentration of ions and dissolved
compounds. It can be measured by pressurizing the concentrated solution until osmotic
induced flow stops. If this pressure is exceeded, then osmotic flow reverses from
concentrated solution to the dilute solution.
RO is a moderate to high-pressure (80 - 1,200 psig) membrane separation process. The
membranes in the EUWP are spiral wound with up to seven modules in a vessel. Figure 2-3
shows the construction of a spiral wound element. They are operated under cross-flow
conditions at a pressure above the osmotic pressure of the bulk solution, plus additional
pressure to overcome resistance of the modules. Water passing through the RO membrane is
called permeate, and the concentrated discharge stream is called concentrate.
The separation model is of solution and diffusion of material through the polymer of the
membrane. Dissolved salts are transported very slowly compared to water and other un-
charged molecules. Uncharged molecules may be rejected based on size exclusion,
depending on their mass and geometry.
2.4 Detailed System Description
This section provides a detailed system description. See the system operation manual in
Appendix A for further details about the system and operation. Note that the system was
designed and manufactured prior to promulgation of the final EPA Long Term 2 Enhanced
Surface Water Treatment Rule (LT2ESWTR). The EUWP, as tested, was not designed to
comply with the LT2ESWTR indirect integrity monitoring requirement that calls for the
system to shut down pending a direct integrity test, if two consecutive turbidity readings
exceed 0.15 NTU. The EUWP does have in-line turbidity meters to monitor the feed and
filtrate streams for the UF skid, but the programmable logic controller (PLC) was not
programmed to automatically shut down the system, if necessary. The RO system has an in-
line turbidity meter for the RO permeate process stream. The RO system also includes in-line
conductivity meters to monitor performance. The system process schematic and detailed
layout are shown in Figures 2-3 and 2-4, respectively.
-------
Concentrate/
Waste
Figure 2-3. EUWP system process schematic.
10
-------
Figure 2-4. Schematic of typical EUWP layout.
11
-------
2.4.1 Raw Water Intake
Raw seawater was drawn from the Port of Hueneme at a point approximately halfway into the
port from the open sea. An intake strainer was used to keep large pieces of debris from being
drawn up. Before the raw water reached the UF feed tank, ferric chloride was injected as a
coagulant, and the water was strained again through dual Amiad Filtration Systems, Ltd. model
TAF-750 strainers, operated in parallel. Each strainer was equipped with a 200 jim weave-wire
screen. The strainers did not remove any ferric chloride floe, since there was not enough time
for particles larger than 200 |im to form between the injection point and the strainer. The 3,000
gallon (gal) UF feed tank provides at least 12 minutes of retention time for floe formation.
2.4.2 UF System Description
The UF membranes used in the EUWP are model TARGA® 10-48-35-PMC, manufactured by
Koch Membrane Systems. The UF cartridge specifications are presented in Table 2-1. The UF
membranes are configured in two parallel trains of eight cartridges each, all of which are
operated in parallel. The membranes are operated such that 10% of the feed flow exits the
cartridges as retentate. Statistics of the UF skid are presented in Table 2-2. Photos of the UF
skid are shown in Figure 2-5 and Figure 2-6.
Table 2-1. Koch Membrane Systems Targa 10-48-35-PMC Cartridge Specifications
Parameter Value
Nominal Molecular Weight Cut-off 100,000
Max. Recommended Flow (per cartridge) 32.2 gpm(1)
Maximum Pressure 45 psig
Maximum Transmembrane Pressure (TMP) 30 psig
Maximum Backflush TMP 20 psig
Inner Fiber Diameter 0.035 in(2)
Membrane Area 554 ft2(3)
Cartridge Diameter 10.75 in
Cartridge Length 48 in
(1) gallons per minute
(2) inch(es)
(3) square feet
12
-------
Table 2-2. UF Skid Statistics
Parameter
Value
Production Capacity
Maximum Pressure to Membranes
Maximum Transmembrane Pressure
Water Temperature Range
Turbidity Range
Dimensions
Weight
Basic Metals
Operating Ambient Temperature Range
Storage and Transport Air Temperature
Range
Relative Humidity:
Maximum slope of unit when deployed for
operation
Power Source Requirement
Fuel Type
Fuel Capacity (60 kW Generator)
250,000 gpd
45 psig
30 psig
34-104 °F
0-150 NTU
20'Lx8'Hx8'W
15,500 Ibs dry, fully paced out for deployment, less fuel
UF System Piping: Fiberglass, Titanium, Nylon
Air System Piping: Nylon Tubing
32°F-120°F
32°F-120°F
3%-95%
5 degrees side to side, 2 degrees end to end
60 kW Generator (self contained) or power grid connection
consisting of 480 volts, 125 amps. UF system and external
pumping power requirements are 2.1 kWh/kgal(1)
DF2 (Diesel Fuel, Grade 2)
DFA (Diesel Fuel, Arctic Grade)
JP8 (jet propellent 8)
43 gal
(1) kilowatt-hours per kilogallon
Figure 2-5. Photo of the UF skid.
13
-------
Figure 2-6. Photo of the UF cartridges mounted in the UF skid.
2.4.2.1 UF System Operation
The following is a basic description of the flow path and functional description of the UF system
in normal operation for an open surface water source. The operation manual provides a full
description of UF operation. Figure 2-7 is a piping and instrumentation diagram of the UF
system.
1. Pump #1 (PI) brings water through the intake strainer #1 (ST1) (if an open intake is used)
to the UF skid. Before entering the UF feed tank, water is strained (ST2) again to 200
|im on the UF skid. The strainers serve to eliminate debris that would clog the membrane
fibers. Water exits strainer #2 and is stored in the UF feed tank (TK2) which serves as a
break tank between the feed water supply and the UF feed.
2. If necessary, ferric chloride coagulant from Chemical Pump #1 (CP1) can be added to the
feed stream before entering ST2 to enhance UF performance. The decision to use ferric
chloride is site-specific, based on the raw water quality, if known, and/or the results of a
jar test.
3. Pump #3 (P3) moves water from TK2 to the UF membranes.
4. The UF filtrate flows to tank #3 (TK3). TK3 acts as a break tank between the UF skid
and the RO skid and a back flush reservoir for the UF skid.
14
-------
5. Pump #5 (P5) pumps water from TK3 to the RO skid or directly through the disinfection
system (CL1 - calcium hypochlorite) to the distribution system when RO is not required.
The disinfection system will not be used for this verification.
2.4.2.2 UF Cleaning Procedure
The UF system must be cleaned when the TMP exceeds 35 psig after a normal backflush cycle.
This cleaning cycle is required approximately every 30 days, depending on the water source.
The CIP procedure typically uses citric acid as the low pH cleaning agent, and sodium hydroxide
as the high pH cleaning agent. Note that different cleaning agents may need to be used for
certain foulants. Citric acid, sodium hydroxide, and sodium hypochlorite (bleach) were used
during the UF system CIP procedures during the ETV test.
If system operation requires the use of ferric chloride as a coagulant, then a low pH clean must
be performed first, followed by a high pH clean. If ferric chloride is not being used, then a high
pH clean must be performed first, then a low pH clean. Ferric chloride was used during testing
at Port Hueneme. The following is a basic description of the flow path and functional
description of the UF system in normal operation. The operation manual provides a full
description of UF operation, including an operational summary described below.
1. Prior to CIP, perform a fresh back flush.
2. Following backwash, set up system for UF normal mode of operation. Activate UF drain
mode on the screen.
3. Wait for the system to drain.
4. Connect the hose from the CIP tank to the system.
5. Touch the CIP button on the screen. Select CIP Mode ON. The programmable logic
controller (PLC) will automatically move the pneumatically operated valves to the
correct positions.
6. Enable heaters to maintain CIP solution to between 96 and 100°F.
7. Turn tank mixer on using CIP display screen
8. Add the appropriate amount of chemical to achieve the desired pH.
9. Check the pH of the mixture in tank 4 at sample port V22 every 15 minutes. Use citric
acid to lower the pH to 3 or use sodium hydroxide to raise the pH to 11.
10. With high pH only, add an appropriate amount of calcium hypochlorite.
11. Start CIP by touching the CIP button at the top left of the CIP screen then start to pump
the solution using P3.
12. Allow the chemical to circulate through the selected array for 20 to 30 minutes.
13. Let the system soak for several hours after recirculation if needed to remove tough
fouling.
14. Repeat recirculation with the desired chemicals.
15. Following chemical recirculation, rinse the system as necessary with clean water.
15
-------
Figure 2-7. Piping and instrumentation diagram of UF skid.
16
-------
2.4.3 RO System
The RO skid is shown below in Figures 2-8 and 2-9.
The RO system has the capability to operate in single-pass or double-pass mode if necessary (the
double-pass mode was not used for this ETV test). The first pass of the RO system consists of a
unique combination of moderate rejection/high productivity and high rejection/moderate
productivity membranes. The first pass is composed of two parallel arrays (Figure 2-10). The
first array is fed by the high-pressure pump and has two parallel trains with two four-element
vessels each (Vessels 1, 2, 3, and 4 in Figure 2-10). The energy from the brine of this array is
used to pressurize feed water via a pressure exchanger energy recovery device to feed a second
array consisting of a single train of two four-element vessels (Vessels 5 and 6 in Figure 2-10).
The second pass RO system consists of a 2-»l array, where a second high -pressure pump boosts
permeate pressure from the first pass feeding two parallel four-element vessels (Vessels 7 and 8
in Figure 2-10). The brine from these vessels then feeds one additional four-element vessel
(Vessel 9 in Figure 2-10).
The RO design incorporates an internally staged RO element configuration on the first pass
(Figure 2-11). This configuration consists of two Dow Chemical Company FILMTEC™ SW30-
HR LE-400 elements, followed by two FILMTEC SW30-XLE400 elements, which are in turn
followed by four FILMTEC SW30-HR-12000 ultra-low-energy experimental membranes. All
membranes are polyamide thin-film composite type. The second pass RO system uses AquaPro
LE-8040UP membrane elements. Table 2-3 provides performance data for the elements used in
the system.
Figure 2-8. Photo of the RO skid.
17
-------
Figure 2-9. Photo of the RO skid membrane vessels.
1 I 2 | 3TT
~~E j g j 7j Q 1 fr Concentrate
•*! 1 I 2 | 3TT
1234
1 I 2 I 3 I 4
5 I 6 I 7 I 8
234 h-*1 5 I 6 I 7 I 8
1234
Numbers indicate pressure vessels
Figure 2-10. Vessel arrangement schematic.
18
-------
X1 | X1 | X2 | X2
X3 I X3 I X3 I X3
X1 X1 X2 X2
X3 X3 X3 X3
X1 X1 X2 X2
X3 X3 X3 X3
X4 | X4 | X4 | X4|—»
X4 X4 X4 X4
X4 X4 X4 X4
Numbers indicate pressure vessels
Figure 2-11. Membrane arrangement schematic.
Table 2-3. RO System Membrane Element Characteristics
Vessel
1st Pass
2,3,5
1st Pass
2,3,5
1st Pass
1,4,6
2nd Pass
7,8,9
Product
FILMTEC SW30-HR LE-
400
FILMTEC SW30-
XLE-400
FILMTEC SW30-HR
-12000 (experimental)
AquaPro LE-8040UP *
Designator
XI
X2
X3
X4
Nominal Active
Surface Area
ft2 (m2)
380 (35)
400 (37)
400 (37)
400 (37)
Permeate
Flowrate gpd
(m3/d)
6000 (26)
9000 (34)
12,000 (45)
10,200 (38)
Stabilized Salt
Rejection
(%)
99.8
99.7
99.7
99.7
* Toray membrane assembled by AquaPro/Village Marine
19
-------
2.4.3.1 RO skid statistics
Table 2-4 presents statistics of the RO skid.
Table 2-4. RO Skid Statistics
Parameter
Value
Production Capacity
Water Temperature Range
Dimensions
Weight
Basic Metals
Operating Ambient Temperature Range
Storage and Transport Air Temperature
Range
Relative Humidity
Maximum slope of unit when deployed for
operation
Power Source Requirement
Fuel Type (if using RO Pump Engine)*
Fuel Capacity (if using RO Pump Engine)*
~ 125,000 gpd for single pass on surface water above 25,000
mg/L TDS or groundwater above 2,500 mg/L TDS
-162,000 gpd for other lower TDS waters
-98,000 gpd in double pass mode
34_104°F
20'Lx8'Hx8'W
15,500 Ibs dry, fully paced out for deployment, less fuel
High Pressure Piping: Titanium
Production Piping: 316L Stainless Steel and fiberglass reinforced
plastic (FRP)
32°F-120°F
32°F-120°F
3%-95%
No Restrictions
Power for all but high-pressure pump is supplied from UF skid.
HP pump requirements are 480 Volts and 125 Amps. The
operational power use is 7.4 kWhr/kgal for the RO system only.
DF2, DFA, JP8
60 gal
: Electric RO pump was used for ETV testing
2.4.3.2 RO System Operation
The following is a basic description of the flow path and functional description of the RO system
in normal operation. The RO system has the capacity to operate in either a one or two pass
mode. The second pass is only used if sufficient treatment is not achieved with the first pass
(especially for NBC contamination). The operation manual provides a full description of RO
operation. Figure 2-12 is a P&ID of the RO system.
1. The UF filtrate is supplied to the RO 1st pass through P5 from TK3.
2. The RO 1st pass includes two arrays. The RO feed water (from the UF filtrate) flows into
vessels 2 and 3 (PV2, PV3). The concentrate from vessels 2 and 3 flow into vessels 1
and 4 (PV1, PV4), respectively. The combined concentrate from vessels 1 and 4 flows
through the energy recovery device, which boosts raw water pressure and feeds vessel 5
(PV5) of the second array. The concentrate from PV5 flows into vessel 6 (PV6). High
arrays and the
1st and
pressure pump #6 (P6) supplies pressure for the 1s pass
pressure exchanger #8 (P8) supplies pressure for the 1st pass 3rd array.
•>nd
20
-------
3. Sodium metabisulfite from chemical pump #2 (CP2) and tank #7 (TK7) can be added
after P5 to remove chlorine, if necessary. Free chlorine can damage RO membranes.
The maximum allowable chlorine level is membrane specific with the minimum chlorine
tolerance being non-detect.
4. Anti-sealant from chemical pump #3 (CP3) and tank #8 (TK8) is added after P5 to
minimize RO membrane scaling.
5. P6 increases the pressure to the required 1st pass 1st array operating pressure (800-1,200
psig depending on water conditions).
6. Concentrate from the 1st pass 1st array flows through the pressure exchanger P8. P8
exchanges energy from the high pressure, high salinity 1st pass concentrate to the lower
pressure, lower salinity UF filtrate feed water. The UF filtrate pressurized by P8 flows
into the 2nd array.
7. Pressure control valves #5, #6, and #7 (PCV5, PCV6, PCV7) are used to adjust pressure
within the RO 1st pass piping. When PCV5 is fully open, P8 is bypassed. When
restricted, PCV5 provides backpressure for P6.
8. As PCV6 is restricted, water is forced through P8.
9. When open, PCV7 prevents P8 overflow during start up. When restricted, it provides
additional backpressure for P6.
10. Second pass operation is optional and will not be verified in this testing. During NBC
operations or when the 1st pass permeate quality does not meet requirements, the 2n pass
is required.
11. The 2nd pass has one array with 12 membranes (PV7, PV8, PV9). The 1st pass permeate
feeds the 2nd pass. If the raw water source does not contain NBC, concentrate from the
2n pass (which is lower concentration because 2n pass feed is 1st pass permeate) is
recycled back to the raw water source to reduce the salinity of the inlet water.
12. Sodium hydroxide from chemical pump #4 (CP4) is added at the 2nd pass inlet to adjust
pH to improve the rejection of certain contaminants that are ionized at high pH such as
Boron.
13. Pump #7 (P7) pressurizes the 1st pass permeate. Pressure control valve #8 (PCV8)
provides the backpressure for pump #7 (P7).
14. The 1st pass permeate is monitored by and displayed on conductivity sensors #1 and #2
(CS1, CS2), which determine if the permeate purity meets requirements. Permeate
salinity is affected by temperature, TDS, and age of the RO membranes. If the permeate
purity does not meet requirements, CS1 de-energizes solenoid valve #1, which then
dumps the undesirable permeate back to the feed source. If the permeate purity meets
requirements, CS2 activates solenoid valve #1, allowing the handle on the dump valve to
be latched, causing the high purity permeate to flow from the RO skid to the product
water storage tanks. This diversion feature is disabled during 2nd pass operation.
15. Prior to distribution, RO permeate flows through the calcium hypochlorite disinfection
system to the product water storage tanks. This system will not be operated during this
test phase.
2.4.3.3 RO Cleaning Procedure
The RO elements should be cleaned whenever the temperature corrected product water output
drops by 10 to 15% from the initial baseline established at the beginning of operation or from the
expected output. The RO elements should also be cleaned when the TDS level of the product
21
-------
water exceeds 500 mg/L. Prior to cleaning the membranes, verify that any reduction in product
output is not the result of a corresponding variation in raw water inlet temperature or salinity by
normalizing the data to a set of initial conditions. The following is a summarization of the
operating instructions from the operations manual:
1. Set RO system in normal operation mode. Verify that valves are in the correct startup
position. Make sure that the system output is being discharged to waste.
2. Select RO clean mode on main display screen.
3. Fill tank 4 with about 300 gal of fresh, un-chlorinated water to within 12 in of the top.
4. If ferric chloride is used in the system, perform the low pH adjustment first. If ferric
chloride is not used, perform high pH adjustment first. (ETV note: ferric chloride was
used during ETV test.)
5. Dissolve the appropriate amount of alkaline detergent or citric acid in a bucket of water.
6. Check the pH of the mixture in tank 4 and adjust as needed. Use citric acid to lower pH
to 3 or use sodium hydroxide to raise the pH to 11.
7. Start P5 and allow chemical solution to circulate for 3 minutes. Check and adjust pH as
needed.
8. Allow the cleaning solution to circulate for 15 minutes.
9. Touch "RO Clean" on the screen. Then touch "Enable RO Clean."
10. Allow system to soak for 1 to 15 hours.
11. After soaking for the desired length of time, re-circulate the cleaning solution for 30
minutes.
12. Drain system and dispose of cleaning agents.
13. Repeat above steps for each desired chemical solution.
14. Rinse the RO system with fresh water.
22
-------
Figure 2-12. P&ID of RO skid.
23
-------
2.4.3.4 Pressure Exchanger
RO is an inherently power intensive process. Historically, energy from the high-pressure brine
was wasted through the utilization of a control valve to the control the process. Today, several
systems are available to recover the energy contained in the high-pressure brine to help offset the
energy required. The EUWP uses the PX® Pressure Exchanger® (Model 90S) from Energy
Recovery, Inc (Figure 2-13). The PX operates on the principle of positive displacement to allow
incoming raw water to be pressurized by direct contact with the concentrate from a high-pressure
membrane system. It uses a cylindrical rotor with longitudinal ducts parallel to its axis to
transfer the pressure energy from the concentrate stream to the feed stream. The rotor fits into a
ceramic sleeve between two ceramic end covers with precise clearances that, when filled with
high-pressure water, create an almost frictionless hydrodynamic bearing. At any given time, half
of the rotor ducts are exposed to the high-pressure stream and half of the ducts are exposed to the
low-pressure stream. As the rotor turns, the energy is transferred to the low-pressure stream,
pushing the feed water on to the booster pump. This type of energy device has been shown to be
90% efficient in transferring energy. During previous EUWP testing in Alamogordo, New
Mexico, the average observed efficiency of the energy recovery device was 78 ± 8 %.
In a typical system, the pressurized feed from the PX goes to a booster pump, which restores the
pressure lost in the exchange and feeds a second RO vessel. However, the EUWP utilizes a
parallel pass 1 train operation at approximately 10% lower pressure than the train operating
directly off the high pressure pump. PX dimensions are 24 in long x 6.5 in diameter. Wetted
materials are duplex stainless steel, ceramics, polyvinyl chloride (PVC), and fiberglass
reinforced plastic (FRP).
Feechwater
Refect fluid
Liquid piston
Rotation
High-pressure feed water
going to 2nd parallel 1st pass
High-pressure concentrate or
reject fluid from reverse osmosis membranes
High pressure
Sealed area
Low pressure
H
Low-pressure feedwater
inlet from brackish supply
pump
PX—Pressure Exchanger™
Figure 2-13. PX pressure exchanger.
Low-pressure concentrate or
reject fluid to drain
24
-------
2.5 General Requirements and Limitations
Table 2-5 lists the general environmental requirements for setup and operation of the EUWP.
Table 2-5. EUWP Site Considerations and Dimensions
Site Considerations
Site Dimensions
Drive-in access for on-road equipment
Work area required for equipment maneuvering and
setup
Fairly smooth, level, and clear ground surface
Cleared path to water source
Work area elevation above pump #1
Elevation/distance of pump #1 above the water source
Distance of pump #1 from inlet strainer #1 in water
source
Water depth from the inlet strainer #1 to the bottom of
the raw water source
Distance of distribution tanks from EUWP
Distance of distribution tanks from adjacent
distribution tank
Distance of distribution pump #9 from tee adaptors
Cleaning waste storage tank
At least 10 ft wide
At least 75 ft x 100 ft
Grade not to exceed 5° side to side and 2° end to end for
UF configured platform or skid. No restriction f-r the
RO skid. Ensure the elevation of tank #3 is equal to or
higher than the UF skid (higher is better).
Wide enough to move equipment
Maximum 25 ft vertical and 100 ft horizontal
Maximum 15 ft vertical and 50 ft horizontal
Maximum 50 ft
3 ft minimum; 5 ft or more preferred
Limited by hose length. Check hoses to determine
distance.
Limited by hose length. Check hoses to determine
distance.
Limited by hose length. Check hoses to determine
distance.
Less than 50 ft from the waste out connection
The EUWP was designed to be transported by air using a C-130 aircraft, or by land using any
number of commercial and military haul transporters. The skids have forklift pockets that allow
handling with an appropriately sized forklift.
Volume and type of consumables are site-specific depending on raw source water quality. As
recommended by the membrane manufacturer, calcium hypochlorite, citric acid, or sodium
hydroxide may be required to perform a CIP. Also as recommended by the membrane
manufacturer, citric acid, sodium hydroxide, and/or a membrane detergent may be required to
perform an RO cleaning. Depending on the raw water source quality, chemical additions may be
needed for protection of the membranes during operation. Ferric chloride may be added at the
UF skid to prevent clogging of the membranes by natural organic matter or high suspended
solids in the feed water. Antiscalant and/or sodium meta-bisulphite may be added at the RO skid
to prevent scaling and remove chlorine present in the feed water; and sodium hydroxide may be
added to raise the pH to aid rejection of constituents during the 2n pass. Calcium hypochlorite
in granular or tablet form containing 65-70% free chlorine may be added prior to filtrate or
permeate storage as a disinfectant (this did not occur as part of this ETV test). Table 2-6 covers
equipment limitations and Table 2-7 presents membrane limitations.
25
-------
Table 2-6. Equipment Limitations
Svstem
Inlet Pump #1
Strainer
Parameter
Suction head (maximum)
Differential pressure (maximum) before manual backwash
Backpressure required for strainer auto flushing
25ft
7psig
35 psig
Value
UF
Pretreatment requirements
Feed pressure (maximum)
Ambient temperature range
Water temperature range
Control air pressure
Damaging chemicals
TMP (maximum) before CIP required
Pressure surges
200 um strainer
45 psig
32 - 120°F
34 - 104°F
60 psig
Grease, Oil, Silicon
35 psig
Minimize by operating valves slowly
UF Membranes Stagnation time (maximum) before preservation required
with 1,000 - 5,000 mg/L sodium bisulfite (see operations
manual for details)
(see Table 2-1 for more details)
14 days (somewhat temperature
dependent)
UF CIP Water Turbidity
Iron
Manganese
Aluminum
Reactive silica
Colloidal silica
Total silica
Calcium sulfate
Calcium carbonate
Microbiological
SDI
pH range
Maximum feed pressure
Maximum Air Pressure
Temperature range
Filtered
All water must be free of particulate matter
<1.0NTU
O.05 mg/L
O.05 mg/L
<0.5 mg/L
ND(1)
ND
<10mg/L
< saturated at 50°C (122°F)
< saturated
no living or dead material
<3.0
1.5-13
45 psig
15 psig
32°F to 120°F
500 um prior to entering UF
RO
Water Temperature Range
SDI (maximum)
Operating Ambient Temperature Range
Storage and Transport Air Temperature Range
Relative Humidity
Pretreatment requirements
Operating concentrate pressure after backpressure valve
(maximum)
Operating permeate pressure (maximum)
2nd pass inlet pressure (maximum)
RO high pressure pump #6 maximum speed
RO high pressure pump #6 minimum inlet pressure
Stagnation time (maximum) before preservation required
34 - 104°F
5 (membrane dependent)
32°F to 120°F
32°F to 120°F
3% to 95%
UF or 200 um strainer on RO skid
200 psig
100 psig
300 psig
600 revolutions per minute (RPM)
30 psig
1 week (somewhat temperature
RO Membranes (see Table 2-6 for details)
(1) Non-detect
26
-------
Table 2-7. RO Membrane Limitations
SS
a
-
cS
S
Membrane
I
cS
£
0)
_0
U
M
sS
•_
I
8.
u
•V
•a
M
cS
•_
M
8.
•± B
3 £
% ^
tu •-
i. OJ
Q- 8.
,
II
I »
cS
S
,0
"5
^
"^ .£•
s g,
a. ,
sS
"
TARGA® 10 - 48 - 35 - PMC
FILMTEC™ SW30HRLE-400
FILMTEC™ SW30
XLE-400
FILMTEC™ SW30HR
-12000 (experimental)
AquaPro LE-8040UP(2)
45
1,000(1)
104
113
200
30
1,200 113
1,200
600
113
113
ND
2-11
2-11
2-11
2-11
-12
-12
-12
-12
15
15
15
20
50
(1) May go up to 1,200 psig under certain conditions specified by Dow Chemical
(2) Toray membrane assembled by AquaPro/Village Marine
2.6 Waste Generation and Permits
The waste streams for the EUWP consist of the following:
• Cleaning waste from UF system (UF CIP);
• Cleaning waste from the RO system (RO CIP);
• Concentrate from the RO system; and
• Backwash waste and retentate from the UF system.
2.6.1 UF CIP
The UF system CIP cycle involves use of the 300-gal CIP tank with the following chemical
cleaning cycles: acid, rinse, base with chlorine, rinse. A second base cleaning may be required.
The total volume generated with five cleaning cycles (worst case, assuming a second base
cleaning) at 300 gal each, plus 200 gal for piping/membrane volume is approximately 2,500 gal.
For this ETV verification, all cleaning solutions were captured in a storage tank. The contents of
the storage tank were pumped into the sanitary sewer.
2.6.2 RO CIP
The CIP procedure for the RO system is similar to that of the UF and uses the same 300-gal CIP
tank to dispense the cleaning solutions. The cleaning cycles consist of an acid clean followed by
a rinse, then a high pH clean with membrane cleaner followed by a final rinse. The approximate
volume of waste generated from all of the cleaning cycles is 1,200 gal of cleaning solutions, plus
200 gal of piping/membrane volume for each cycle, for a total of approximately 2,000 gal.
2.6.3 RO Concentrate
The RO concentrate was blended with the RO permeate, UF backwash, and UF retentate, and the
resulting mixture was discharged back into the port.
27
-------
2.6.4 UF Backwash and Retentate
The UF system automatically initiates a backwash every 30 minutes to remove captured material
from the membrane surface. Each backwash cycle consists of backflushing the membranes with
UF filtrate for a short period followed by a forward "fast flush" using feed water. In addition to
the backwash, the UF system also discharges a continuous retentate stream. Both waste streams
exited the system using a common discharge line that was routed to a settling tank. The settling
tank effluent was discharged into the port.
2.6.5 Discharge Permits
SDTF operates under a waiver from the South Coast Regional Water Quality Control Board.
This waiver covers the operation of desalination and filtration equipment for evaluation
purposes. The waiver stipulates that concentrate and filtrate/permeate streams be recombined
and returned to the port.
2.7 Discussion of the Operator Requirements
The following information on operator requirements is supplied by the manufacturer for
informational purposes only. A team of four water treatment specialists, with proper site
validation, layout planning and using a 10,000-lb forklift, should be able to have the EUWP
setup and producing potable water within eight hours. Depending on the distribution connection
requirements and availability of the connections, distribution of the produced potable water may
take longer.
Except for periodic O&M and data collection, once set up and operational, the EUWP is capable
of operating unattended. Staffing requirements are based on the O&M or data collection efforts
being performed. Due to the use of high pressure, electricity or diesel, and chemicals, O&M on
the equipment and piping should be performed by a minimum of two persons. Data collection
requires only one person.
The EUWP requires a skilled operator familiar with water treatment processes, equipment, and
concepts to perform O&M and collect data. A skilled operator could meet any of a variety of
requirements as discussed below. Operation of the EUWP should be performed by an individual
with similar experience, knowledge, or training as provided within these programs.
A U.S. military water treatment specialist (classified as skill level 4 through 1) supervises or
performs installation, operation of water purification equipment, water storage, and distribution
operations and activities.
The minimum skill level 4 requires the specialist to:
• Assist in water reconnaissance, site preparation, and setup of water treatment activity;
• Operate and maintain water treatment equipment;
• Receive, issue, and store potable water; and
• Perform water quality analysis testing and verification.
28
-------
Although remote operation is not available, the EUWP can be monitored remotely 24 hours per
day by use of the water system management tool, WaterEye™ (www.watereye.com). WaterEye
provides timely, critical operations monitoring information utilizing colored indicators to either
confirm system status or alert potential problems. In addition, WaterEye can assist with
managing daily, monthly, and yearly compliance requirements by monitoring compliance data
and automatically creating reports. WaterEye maintains a database of monitored instrument
readings, which are read every 15 min and uploaded to their server every 30 min. Alarm
conditions are immediately uploaded for response. WaterEye can also display/store information
calculated from uploaded instrument readings. Data must be either uploaded directly from the
PLC on the EUWP or be able to be calculated from that data.
29
-------
Chapter 3
Methods and Procedures
3.1 Quantitative and Qualitative Evaluation Criteria
The objectives of the verification test were to evaluate equipment in the following areas:
• The actual results obtained by the equipment as operated under the conditions at the test
site;
• The impacts on performance of any variations in feed water quality or process variation;
• The logistical, human, and other resources necessary to operate the equipment; and
• The reliability, ruggedness, ranges of usefulness, and ease of operation of the equipment.
There are three main components of the EUWP that were evaluated at the same time: the UF
system, the RO system, and the energy recovery system. All three components must function
successfully to meet the performance objectives. To address these objectives, the verification
test employed the quantitative and qualitative factors listed below.
Qualitative factor:
• Waste discharge requirements.
Quantitative factors:
• Water quality data
• Physical operations data - flow, membrane flux, recovery, and pressure.
• Power usage
• Chemical usage
• Waste stream generation
• Operating cycle length
3.2 Key Treated Water Quality and Operational Parameters
Treated product water must meet EPA NPDWR, and should meet EPA secondary Standards
whenever possible. As discussed in Section 2.1, the objective of this ETV verification was to
demonstrate that the EUWP can provide water that meets the requirements of the EPA NPDWR.
As such, a list of key treated water parameters was developed based the EPA regulations, and
other water quality parameters of interest. Regulated contaminants not present in raw water
samples analyzed during the characterization of feed water task were excluded from the list. The
final list is presented in Table 3-1.
Note that the test/quality assurance plan (TQAP) called for also measuring strontium and total
organic carbon (TOC), and also conducting a silt density index (SDI) test on the RO feed, but
these parameters were dropped. Strontium was to be measured for the Stiff and Davis Stability
Index (S&DSI) calculations, but it was found in the raw seawater at such a low concentration
compared to other cations that it did not need to be included in the S&DSI
30
-------
Table 3-1. Key Treated Water Quality Parameters
Parameter
pH Magnesium
Temperature Ortho -Phosphate
Specific Conductance Total Phosphorus
Turbidity Potassium
Particle Counts Selenium
Stiff and Davis Stability Index (S&DSI) Sodium
Alkalinity Sulfate
Barium Total Dissolved Solids (TDS)
Bicarbonate Total Hardness
Boron Total Suspended Solids (TSS)
Calcium Ultraviolet light absorbance at 254 nm (UV254)
Chloride Total Coliforms
Lithium Bacillus Endospores
analysis. TOC was dropped because it was measured in the raw seawater at a less than detectible
concentration (<0.3 mg/L). The SDI test was not dropped intentionally, the FTO forgot to run it.
However, this test is of questionable importance for RO feed water when the feed has been
treated by UF. The SDI test uses a 0.45 jim filter to capture and measure silt in the feed water.
The UF membrane pore size is less than 0.1 jim, so the UF should remove all of the silt that
would be captured by the SDI filter. A portion of the water quality and operational parameters
were measured continuously via in-line instrumentation, as listed in Table 3-2.
Table 3-2. Water Quality and Operational Parameters Measured In-Line
Membrane
Flow
Pressure
Conductivity
Temperature
Turbidity
Particle Count
•a
1
^
X
X
X
X
1>
-*^
c$ o>
*- +2
= ~ —
(U '- ^5
-*^ -*^ 5*
W .- w
PS ta ^
P P rt
X X
XXX
X
X
X
>* >i ^ w
^ "*^ ^ ^" ^
•s S -s o» ' o» o»
O g O | O S3 O
tf U K &H « &H «
XXX
X X
X X
X
31
-------
3.3 Operations and Maintenance
Village Marine Tec. provided an operations and maintenance manual for the EUWP, which is
included in Appendix A. The ETV test protocols call for review of the manual in regards to the
ability of the user to successfully operate the system armed with only the information in the
manual. An objective review of the manual by the field operators was not possible, because they
already had intimate knowledge of the EUWP prior to the test. Therefore, a review of the O&M
manual is not included in this report.
The following aspects of operability are addressed in Chapters 2 and 4, and in the appendices:
• Fluctuation of flow rates and pressures through unit (the time interval at which resetting
is needed);
• Presence of devices to aid the operator with flow control adjustment;
• Availability of pressure measurement;
• Measurement of raw water rate of flow;
• Pace of chemical feed with raw water; and
• Operation of the PLC control system.
3.4 Field Operations
Acting as the FTO, TARDEC conducted the testing of the EUWP as described below. TARDEC
and USER field personnel performed field analytical work using field laboratory equipment and
procedures for pH, temperature, conductivity, and turbidity. NSF performed water quality
analytical work for samples not analyzed on site. Field staff were on site each day to operate the
system and collect water quality data during the verification test.
The test plan called for the EUWP to be operated 24 hours per day, seven days per week,
excluding regular backwashes and cleaning periods. This was the case for most of the test
period, except when the system shut down during the night due to an alarm, and field personnel
were not present to restart the system. System shutdowns that occurred during the ETV test are
discussed in Chapter 4.
3.5 Overview of ETV Testing Plan
A test/quality assurance plan (TQAP) was prepared for the EUWP verification test in accordance
with the ETV Protocols EPA/NSF Protocol for Equipment Verification Testing for Removal of
Inorganic Constituents - April 2002, and the EPA/NSF Protocol for Equipment Verification
Testing for Physical Removal of Microbiological and Particulate Contaminants - September
2005. The TQAP divided the work into three main tasks (A, B, C) with Task C, the verification
test itself, divided into six subtasks. These tasks are:
Task A: Characterization of Feed Water
Task B: Equipment Installation, Initial Test Runs, and Initial System Integrity Tests
Task C: Verification Test
Task Cl: Membrane Flux and Recovery
Task C2: Cleaning Efficiency
32
-------
Task C3: Finished Water Quality
Task C4: Membrane Module Integrity
Task C5: Data Handling Protocol
Task C6: Quality Assurance/Quality Control (QA/QC)
The TQAP, which included a Quality Assurance Project Plan (QAPP), specified procedures to be
used to ensure the accurate documentation of both water quality and equipment performance.
An overview of each task is provided below with detailed information on testing procedures
presented in later sections.
3.5.1 Task A: Characterization of Feed Water
The objective of this initial operations task was to obtain a chemical, biological, and physical
characterization of the feed water prior to testing.
3.5.2 Task B: Equipment Installation, Initial Test Runs, and Initial System Integrity
Tests
The objective of this initial operations task was to evaluate equipment operation and determine
the treatment conditions that result in effective treatment of the feed water. This task was
considered shakedown testing and was carried out prior to performing Task C.
3.5.3 TaskC: Verification Test
The verification test itself consisted of six tasks described as follows:
3.5.3.1 Task Cl: Membrane Flux and Recovery
Task Cl evaluated membrane operation and entailed quantification of membrane flux decline
rates and product water recoveries. The rates of flux decline demonstrate membrane
performance at the specific operating conditions established during Task B.
3.5.3.2 TaskC2: Cleaning Efficiency
An important aspect of membrane operation is the restoration of membrane productivity after
membrane flux decline has occurred. The objective of this task was to evaluate the efficiency of
the membrane cleaning procedure. The fraction of specific flux restored following a chemical
cleaning and after successive filter runs was determined.
3.5.3.3 TaskC3: Finished Water Quality
The objective of this task was to evaluate the quality of water produced by the EUWP. Treated
water quality was evaluated in relation to feed water quality and operational conditions. The
monitored water quality parameters are listed in Table 3-1.
33
-------
3.5.3.4 TaskC4: Membrane Module Integrity
The objective of this task was to demonstrate the methodology for monitoring membrane
integrity and to verify the integrity of membrane modules.
3.5.3.5 TaskCS: Data Handling Protocol
The objective of this task was to establish an effective field protocol for data management at the
field operations site and for data transmission between TARDEC and NSF.
3.5.3.6 Task C6: Quality Assurance and Quality Control
An important aspect of verification testing is the protocol developed for QA/QC. The objective
of this task was to assure accurate measurement of operational and water quality parameters
during membrane equipment verification testing.
3.6 Task A: Characterization of Feed Water
The objective of this task was to determine the chemical, biological, and physical characteristics
of the feed water. Grab samples for seawater analysis were collected on four separate occasions
during 2006 and 2007. The samples were collected from the existing seawater intake line at
SDTF.
3.7 Task B: Equipment Installation, Initial Test Runs, and Initial System Integrity Tests
The objective of this task was to properly install the equipment and begin equipment operation,
then evaluate operation and determine whether the operating conditions resulted in effective
treatment of the water. In this task, a preliminary assessment of the treatment performance of the
equipment was made. This task was considered a shakedown testing period and was completed
before Task C. This task also included pressure decay testing of the UF membranes. See
Section 3.8.4.1 for further discussion about this test.
3.8 TaskC: Verification Testing
The verification test ran from October 16, 2007 to November 12, 2007. Note that the ETV test
protocol referenced in Section 3.5 calls for testing to run for at least 30 days. However, the main
intent of the test period length is to operate the equipment until at least one chemical cleaning is
required. NSF allowed TARDEC to stop testing two days early because over the course of
testing, four UF system cleanings were conducted. Per a requirement of the ETV test, a CIP was
performed on the RO system at the end of the ETV although the RO system had not yet reached
its cleaning level criteria.
The TQAP describes six tasks to be performed to achieve a successful verification test. Each of
these tasks is described in detail in this section.
3.8.1 TaskCl: Membrane Flux and Operation
The purpose of this task was to evaluate membrane flux during extended operation to
demonstrate membrane performance. The objectives of this task were to demonstrate the feed
34
-------
water recovery achieved by the membrane equipment, and the rate of flux decline observed over
extended membrane operation. Flow, pressure, conductivity, and temperature data were
collected daily in order to quantify the loss of productivity in terms of specific flux decline.
3.8.1.1 Work Plan
At least twice per day the operator checked the flow rates and recovery and made adjustments as
necessary to put the system on target. Thirty minutes after resetting target flow and recovery, the
operator recorded the appropriate water quality and operational data, as outlined in Table 3-3.
The set points for key operating parameters are listed in Table 3-4. Chemical usage was
monitored by recording the concentration and tank level on a daily basis.
Table 3-3. Operational Parameter Sampling Locations
Parameter
,
£
Flow X
Pressure
Conductivity
Temperature
Power usage
Operating Hours
1
P
X
X
X
4> —
2 § Js "O eS +* eS i •«
?s jy & ry *& & *& QJ ^3
ta PS pQ 'i<^HS^Hy^
-fc -fc ^ -fc C
P p p tttt^UUp
X X X X X
XX X X
XXX
X XXX
X
X
1
C^l
o
X
X
Table 3-4. Key Operating Parameters
Parameter Set Point
UF Feed Flow (gpm) 259
UF Recovery (%) 90
RO Feed Flow 1st Pass Array 1 (gpm) 116
RO Feed Flow 1st Pass Array 2 (gpm) 58
RO Recovery Levels (%) 50 (1st array) and 48 (2nd array)
3.8.1.2 Evaluation Criteria
Completion of this task involved quantification of membrane flux decline rates and product
water recoveries. Summaries of the data collected for Task Cl are presented in tabular format in
Chapter 4 for both the RO and UF systems.
The plots listed in Table 3-5 are also presented in Chapter 4 to illustrate equipment operation for
Task Cl. Note that all plots are of the parameter over time.
35
-------
Table 3-5. Operational Data Plots Appearing in Chapter 4
UF Skid RO Skid
Filtrate Production Flow Rates
Flow Rates Percent Recovery
Operating Presssures Operating Pressures
Trans-Membrane Pressures Specific Flux
Specific Flux Power Consumption
Loss of Specific Flux
Power Consumption
3.8.1.3 Equations
UF System
The following are the definitions and equations used for the verification report for the UF
system:
Filtrate: Treated water produced by the UF process.
Retentate: Water rejected by the UF system.
Feed water: Water introduced to the membrane elements after all chemical additions.
Raw water: The source water supply.
Membrane flux: The average flux across the UF membrane surface calculated by dividing the
flow rate of filtrate by the surface area of the membrane.
Membrane flux is calculated as follows:
j.=^-
' S
where:
Jt = filtrate flux at time t (gallons per square foot per day (gfd))
Qp = filtrate flow (gpd)
S = membrane surface area (ft2)
36
-------
Temperature Adjustment for Flux Calculation: Temperature corrections to 20°C for filtrate flux
and specific flux are made to correct for the variation of water viscosity with temperature. The
following empirically derived equation was used to provide temperature corrections for specific
flux calculations:
Q X(, -0.0239(7-20)
J.=-
s
where:
Jt
QP
s
T
= filtrate flux at time t (gfd)
= filtrate flow (gpd)
= membrane surface area (ft2)
= temperature of the feed water (°C)
Trammembrane Pressure: The pressure across the membrane, equal to the average feed
pressure on the membrane (average of inlet pressure and outlet pressure) minus the filtrate
(permeate) pressure:
TMP =
-P-
where:
IMP
Pf
PC
pn
transmembrane pressure (psig)
inlet pressure to the feed side of the membrane (psig)
outlet pressure on the retentate side of the membrane (psig)
filtrate pressure on the treated water side of the membrane (psig)
Specific flux: The filtrate flux that has been normalized for the TMP. The equation used for
calculation of specific flux is given by the formula provided below. Specific flux is usually
discussed with use of flux values that have been temperature-adjusted to 20°C per equation
above:
J.
Jt=-
where:
TMP
Jt
Jtm
RO System
TMP
= Transmembrane pressure across the membrane (psig)
= filtrate flux at time t (gfd) (temperature-corrected flux values were employed)
= specific flux at time t (gfd/psig)
Permeate: Water produced by the RO membrane process.
Feed Water: Water introduced to the membrane element.
Concentrate: Water rejected by the RO membrane system.
37
-------
Permeate Flux: The average permeate flux is the flow of permeate divided by the surface area of
the membrane. Permeate flux is calculated according the following formula:
Jt = ^j-
where:
Jt = permeate flux at time t (gpd))
Qp = permeate flow (gpd)
S = membrane surface area (ft)
Temperature Adjustment for Flux Calculation: Temperature corrections to 25 °C for permeate
flux and specific flux were made to correct for the variation of water viscosity with temperature.
The following empirically-derived equation were used to provide temperature corrections for
specific flux calculations:
Jt (at 25° C) = QpXS s
where:
Jt = permeate flux at time t (gfd)
Qp = permeate flow (gpd)
S = membrane surface area (ft)
T = temperature of the feed water (°C)
Net Driving Pressure: For this test, a temperature conversion chart provided by the
manufacturer was used for all temperature correction. Net Driving Pressure (NDP) is the total
average pressure available to force water through the membrane into the permeate stream. Net
driving pressure is calculated according to the following formula:
where:
NDP = net driving pressure for solvent transport across the membrane (psig)
Pf = feed water pressure to the feed side of the membrane (psig)
Pc = concentrate pressure on the concentrate side of the membrane (psig)
Pp = permeate pressure on the treated water side of the membrane (psig)
A;r = osmotic pressure (psig)
Osmotic Pressure Gradient: The term osmotic pressure gradient refers to the difference in
osmotic pressure generated across the membrane barrier as a result of different concentrations of
dissolved salts. The following equation provides an estimate of the osmotic pressure across the
semi-permeable membrane through generic use of the difference in TDS concentrations on either
side of the membrane:
A;r =
(TDSf+TDSc)
2
0.6 psi
~TDS
P
38
-------
where:
A;r = osmotic pressure (psig)
TDSf = feed water IDS concentration (mg/L)
TDSC = concentrate TDS concentration (mg/L)
TDSP = permeate TDS concentration (mg/L)
Note that the different proportions of monovalent and multivalent ions composing the TDS will
influence the actual osmotic pressure, with lower unit pressures resulting from multivalent
species. The osmotic pressure ratio of 1 psig per 100 mg/L is based upon TDS largely composed
of sodium chloride or other monovalent ions. In contrast, for TDS composed of multivalent ions,
the ratio is closer to 0.5 psig per 100 mg/L TDS. Osmotic pressure was estimated using the ionic
strength of the feed and concentrate based on the weekly data for cations and anions (Ca, Mg,
Na, K, Li, Cl, SO4, HCOs). The ratio of 1 psig per 100 mg/L TDS gave a much higher osmotic
pressure and the ratio of 0.5 psig per 100 mg/L TDS gave a lower osmotic pressure. It was
determined that the equation for TDS using a factor 0.6 psig per 100 mg/L TDS most closely
approximates the osmotic pressure calculated based on the ionic strength data available for this
water.
Specific Flux: The term specific flux is used to refer to permeate flux that has been normalized
for the net driving pressure. The equation used for calculation of specific flux is given by the
formula provided below. Specific flux is usually calculated with use of flux values that have
been temperature-adjusted to 25 °C:
Jtm=lvDP
where:
Jtm = specific flux (gfd/psig)
NDP = net driving pressure for solvent transport across the membrane (psig)
Jt = permeate flux at time t (gfd). Temperature-corrected flux values should be
employed.
Water Recovery: The recovery of feed as permeate is given as the ratio of permeate flow to feed
flow:
% System Recovery =100
where:
Qf = feed flow to the membrane (gpm)
Qp = permeate flow (gpm)
Loss of Original Specific Flux:
Percent Loss = 100- 1
J,
39
-------
where:
Jso = specific flux (gfd/psig) at time zero point of membrane testing.
Js = specific flux (gfd/psig) at time T of membrane testing.
Solute Rejection: Solute rejection is controlled by a number of operational variables that must be
reported at the time of water sample collection. Bulk rejection of a targeted inorganic chemical
contaminant may be calculated by the following equation:
C -C
Percent Solute Rejection = 100 •' f F
Cf
where:
Cf = feed concentration of specific constituent (mg/L)
Cp = permeate concentration of specific constituent (mg/L).
3.8.2 TaskC2: Cleaning Efficiency
An important aspect of membrane operation is the restoration of membrane productivity after
specific flux decline has occurred. The effectiveness of chemical cleaning to restore membrane
productivity was evaluated.
3.8.2.1 Work Plan
The manufacturer specified that the UF cleaning procedure should be executed when the TMP
drop exceeds 35 psig, even after a backwash. The manufacturer specified that the RO system be
cleaned when there is a 10 to 15% decrease in normalized permeate flowrate, 15% increase in
TMP drop or permeate TDS concentration.
Flow, pressure, and temperature data were recorded immediately before the system was shut
down for cleaning and immediately upon return to membrane operation after cleaning procedure
was complete.
Two primary indicators of cleaning efficiency and restoration of membrane productivity were
examined in this task:
• Immediate recovery of membrane productivity (percent recovery of specific flux); and
• Long term maintenance of specific flux over an equivalent time period.
The pH, temperature, conductivity, and TOC of each cleaning solution were measured after the
cleaning. Flow, pressure, and temperature data were also collected during the cleaning
procedure. Following the cleaning procedure, the specific membrane flux was calculated at the
same operating conditions used prior to the cleaning. This value was compared to the pre-
cleaning specific flux to determine the efficiency of the cleaning procedure. See Section 2.4.2.2
for the UF cleaning procedure, Section 2.4.3.3 for the RO cleaning procedure, and also the
User's Manual (Appendix A) for details on the cleaning procedures employed.
3.8.2.2 Evaluation Criteria
The outputs for this task are post-cleaning flux recoveries and the cleaning efficacy indicators
described above (including flow, pressure, and temperature data).
40
-------
3.8.3 TaskCS: Finished Water Quality
The objective of this task was to assess the ability of the membrane equipment (both UF and RO)
to meet the water quality goals specified by the manufacturer.
3.8.3.1 Work Plan
The water quality parameters in Table 3-6 were measured as indicated during the testing period.
To the extent possible, scheduled on-site analyses for each sampling point were performed on
water samples collected at the same time as the samples shipped off site.
Table 3-6. Water Quality Sampling Schedule
Parameter
pH
Temperature
Specific Conductance
Turbidity
Particle Counts
General Mineral(2)
Regulated Metals(3)
TDS
TOC
TSS
UV254
Total Coliform
Bacillus endospores
•_
4J
"S
C«
C£
DI
DI
DI
DI
DI
W
W
D2
W
W
W
D2
« 3 a
3 & '3
E E* § a
a S .a o
LI -o u '*
"5
p p S p £
W E
D! E
W
D! W
DI
W
E
W W
W W
D2 D2
-------
3.8.3.2 Evaluation Criteria
All water quality data generated during the test periods is presented in a tabular format in
Chapter 4. In addition, the UF feed and filtrate turbidity data, and the RO conductivity data is
presented in a graphical format.
3.8.4 Task C4: Membrane Integrity Testing
The objective of this task is to demonstrate the methodology to be employed for direct integrity
testing of the UF system and indirect integrity monitoring of both the RO and UF systems.
Direct testing and indirect monitoring methods were used together to provide consistent and
sensitive evaluation of membrane system integrity.
3.8.4.1 Direct Integrity Testing:
The direct integrity testing method employed on the UF system was a pressure decay test, similar
to that described in ASTM International (ASTM) Standard D6908 - Standard Practice for
Integrity Testing of Water Filtration Membrane Systems. A pressure decay test was performed
during Task B to establish a baseline pressure decay rate for the UF system. During testing, the
pressure decay test was performed daily. The pressure decay test was also performed after each
UF system cleaning.
Note that the TQAP called for conducting a marker dye test on the RO system as a direct
integrity evaluation, but this test was not conducted.
3.8.4.2 Continuous Indirect Integrity Monitoring:
Continuous indirect integrity monitoring methods were employed on both the UF and RO
systems. Turbidity was monitored continuously on the UF feed, UF filtrate, and RO permeate.
In addition to turbidity monitoring, particle counts were continuously monitored on the UF
system, and conductivity was monitored on the RO permeate stream. Turbidity readings were
recorded every fifteen minutes, while particle counts were recorded every five minutes, and
conductivity readings were recorded hourly.
Results of the direct integrity tests, and the indirect integrity monitoring data are presented in
Section 4.5.4.
3.8.5 TaskCS: Data Handling Protocol
The objectives of this task were to: 1) establish an effective structure for the recording and
transmission of test field test data, such that TARDEC provided sufficient and reliable data; and
2) develop an effective and accurate statistical analysis of the data.
3.8.5.1 Work Plan
The EUWP test system was equipped with a computer monitoring system. Some of the required
measurements (see Table 3-2) were recorded automatically by the automated system. The
remaining required measurements were recorded by hand by the field operator on-site. The data
was recorded onto specially prepared bench sheets, which are included as Appendix B.
Miscellaneous operational notes were recorded in a data logbook with numbered pages. All
42
-------
errors were crossed out with one line, and the error was initialed and dated. Completed pages
were signed, dated, and numbered by the individual responsible for the entries.
The database for the project was set up in the form of custom-designed spreadsheets. A
spreadsheet containing the operational data, including calculations, was developed by USER. A
spreadsheet containing the water quality data was developed by NSF. Following data entry,
100% of the data in the spreadsheets was checked against the numbers on the field log sheets or
laboratory analysis outputs.
3.8.6 Task C6: Quality Assurance Project Plan
QA/QC of the operation of the equipment and the measured water quality parameters was
maintained through a QAPP, as described in this section.
3.8.6.1 Experimental Objectives
The objective of this task was to maintain strict QA/QC methods and procedures during the
verification test. This included maintaining instrument calibration and operation within the
ranges specified by the manufacturer.
The elements of the QAPP for this verification test included:
• work plan;
• QA/QC verifications;
• data correctness;
• calculation of indicators of data quality; and
• corrective action plan
3.8.6.2 Work Plan
A routine daily walk-through during testing was conducted to verify that each piece of
equipment or instrumentation was operating properly. Chemical addition rates and receiving
stream flows were checked to verify that they flowed at the expected rates. Values recorded by
the automated data acquisition program were checked daily against those displayed on the
instrument displays and those measured on-site.
3.8.6.3 QA/QC Verifications
Tables 3-7 and 3-8 give the on-site QA and on-site QC activities, respectively, for the
verification test. NSF Laboratory analytical QA and QC activities followed those specified in
the NSF Laboratory Quality Assurance Manual.
43
-------
Table 3-7. On-Site Analytical Equpment QA Activities
Equipment
Action Required
Initial Flowmeters - electronic
Turbidimeter - in-line (1720E)
Turbidimeter - in-line (FilterTrak)
Particle counter - in-line
UV Spectrophotometer
Verified calibration volumetrically
Provided factory calibration certificate
Provided factory calibration certificate
Provided factory calibration certificate
Provided factory calibration certificate
Daily Chemical Feed Pump
Turbidimeter - in-line
pH meter - portable
Turbidimeter - in-line
Particle Counters - in-line
Volumetrically checked flowrate
Verified with portable turbidimeter
3-point calibration (4,7,10)
Volumetrically checked flowrate
Volumetrically checked flowrate
Weekly Rotameters
UF filtrate flow
Particle counter - in-line
Temperature - portable
Turbidimeter - portable
Conductivity meter - portable
Inspected for buildup of algae, salt, etc.
Verified volumetrically
Cleaned sensors
Verified calibration with NIST-certified thermometer
Calibrated using <0.1, 20, 100, and 800 NTU standards
Calibrated at 2 points
Every Two
Weeks
Flowmeters - electronic
Verified calibration volumetrically
Prior to Tubing
Test Particle Counter - in-line
Turbidimeter - in-line (1720E)
Turbidimeter - in-line (FilterTrak)
Checked condition, checked for leaks
Factory Calibration
Cleaned and calibrated using 20 NTU standard
Cleaned and calibrated using 0.8 NTU standard
Table 3-8. On-Site Data Generation QC Activities
Item
Action Required
Daily
Weekly
Data
Data
Reviewed system performance data since previous day
Compared field and lab water quality results when available
3.8.6.4 Data Correctness
There are five indicators of data quality that were used for this verification test:
• representativeness;
• statistical uncertainty;
• precision;
• accuracy; and
• completeness.
These five indicators are discussed in detail in the sections that follow.
44
-------
3.8.6.4.1 Representativeness
Representativeness of the data for this verification test was ensured by executing consistent
sample collection and data collection procedures, including:
• Consistency of sample locations;
• Timing of sample collection;
• Analytical methods; and
• Sampling procedures, sample preservation, packaging, and transport.
3. 8.6.4.1.1 On-Site Analytical Methods
The analytical methods for on-site monitoring of raw and treated water quality are described in
the sections below.
Analyses for pH were performed according to Standard Method 4500-H+ using a Myron L
Ultrameter II Model 6P or an Accumet Model 50. Three-point calibration (using pH 4, 7, and 10
buffer solutions) was performed daily.
Temperature
Readings for temperature were conducted in accordance with Standard Method 2550 using a
Myron L Ultrameter II Model 6P. A calibration check was performed weekly with a NIST-
traceable thermometer.
Turbidity
Turbidity was measured at all sampling points using a hand-held turbidimeter. In addition, in-
line turbidimeters were used for measurement of UF feed and filtrate. All measurements were
conducted according to EPA Method 180.1.
Hand Held Turbidimeters: A Hach 21 OOP portable turbidimeter (range 0 to 1000 NTU) was
used to measure the turbidity of the appropriate grab samples. The turbidimeter was calibrated
weekly using formazin turbidity standards of <0.1, 20, 100, and 800 NTU.
In-line Turbidimeters: In-line Hach turbidimeters were used for measurement of turbidity in the
feed (Hach 1720 E - Low Range) and UF filtrate water (Hach FilterTrak 660). The Hach 1720E
has a range from 0 to 100 NTU and uses a 20 NTU calibration standard. The Hach FilterTrak
has a range from 0.005 to 5.00 NTU and uses a 0.8 NTU calibration standard. These
turbidimeters were calibrated at the start of the test. In-line readings were periodically compared
to the readings from the hand-held turbidimeter. If the comparison suggested inaccurate
readings, the in-line turbidimeter was recalibrated. A volumetric check on the sample flowrate
was performed daily.
Conductivity
Analyses for conductivity were performed according to manufacturer's instructions using a
Myron L Ultrameter II Model 6P. A two-point calibration was performed daily.
45
-------
Particle Count
In-line particle counters were employed for measurement of particle concentrations in UF
membrane unit feed and filtrate waters. The Hach 2200 PCX in-line particle sensor selected is
able to measure particles with a range of 2 u_m to 750 um in up to 32 user-defined bins. The
particle counters were calibrated by the manufacturer prior to the ETV test.
3.8.6.4.1.2 Sample Collection, Shipment, and Storage for Laboratory Analyses
Samples were collected in bottles prepared by NSF and shipped to the test site. All samples were
preserved, if required, according to the proper analytical method. Bottles for parameters
requiring preservation were shipped to the test site containing the preservative. All samples were
kept on ice in coolers and shipped overnight to NSF. Chain of custody forms accompanied all
samples.
3.8.6.4.1.3 Laboratory Analytical Methods
A comprehensive list of laboratory analytical methods used can be found in Table 3-9. TDS
from the lab analysis was correlated to conductivity for calculation of normalized permeate flow
and rejection trends over time. TDS was used to calculate osmotic pressure gradient needed for
net driving pressure calculations.
Table 3-9. Analytical Methods for Laboratory Analyses
Parameter
Alkalinity (total, as CaCO3)
Barium
Bicarbonate (as CaCO3)
Boron
Calcium
Chloride
TDS
Hardness (total, as CaCO3)
Lithium
Magnesium
Phosphate (ortho)
Phosphate (total)
Potassium
Selenium
Sodium
Sulfate
TOC
TSS
UV254
Bacillus Endospores
Total Coliforms
Analytical Method
EPA 3 10.2
EPA 200.8
EPA 3 10.2
EPA 200.7
EPA 200.7
EPA 300.0
SM 2540 C(2)
SM 2340 B
EPA 200.8
EPA 200.7
SM 4500-P E
EPA 200.7
EPA 200.7
EPA 200.8
EPA 200.7
EPA 300.0
SM5310B
SM 2540 D
SM5910B
SM9218
SM 9222
Method Detection Limit(1)
5mg/L
lUg/L
5mg/L
lUg/L
20ug/L
0.5 mg/L
5 mg/L
2 mg/L
lUg/L
20ug/L
0.020 mg/L
0.050 mg/L
0.5 mg/L
2ug/L
0.5 mg/L
0.5 mg/L
0.1 mg/L
2 mg/L
0.000 Absorbance/cm (A/cm)
1CFU/L
1 CFU/100 mL
(1) The listed detection limits may not apply for some samples, especially the feed water samples, due to matrix
interference from the high TDS.
(2) SM=Standard Methods
46
-------
3. 8. 6. 4. 2 Statistical Uncertainty
For the water quality parameters monitored, 95% confidence intervals were calculated for data
sets of eight values or more. The following equation was used for confidence interval
calculation:
Confidence Interval = x+ [tn-i,i -(0/2) x (S/Vn)]
where:
x = sample mean
S = sample standard deviation
n = number of independent measurements included in the data set
t = Student's t distribution value with n-1 degrees of freedom
a = significance level, defined for 95% confidence as: 1 - 0.95 = 0.05
According to the 95% confidence interval approach, the a term is defined to have the value of
0.05, thus simplifying the equation for the 95% confidence interval in the following manner:
95% Confidence Interval = x + [tn-i,o.975 x (S/Vn)]
3.8.6.4.3 A ccuracy
The accuracy of on-site analytical equipment was periodically verified according to the schedule
in Table 3-7. The calibration records for the analytical equipment were recorded on bench sheets
(Appendix B). All calibrations were performed at the frequency required. All calibration data
were within the specified QC objectives on all days analyses were performed.
Accuracy for the laboratory analyses was quantified as the percent recovery of a parameter in a
sample to which a known quantity of that parameter was added. The following equation was
used to calculate accuracy:
Percent Recovery = 100 x [(X^^ - Xmeasured)
where:
= known concentration of measured parameter
= measured concentration of parameter
Accuracy also incorporates calibration procedures and use of certified standards to ensure the
calibration curves and references for analysis are near the "true value." Accuracy of analytical
readings was measured through the use of spiked samples and lab control samples. Table 3-10
presents the control sample frequency and accuracy limits for each parameter.
47
-------
Table 3-10. Accuracy and Precision Limits for Laboratory Analyses
LFM™
(spike
sample)
Parameter Frequency
Alkalinity
Barium
Bicarbonate
Boron
Calcium
Chloride
IDS
Total Hardness
Lithium
Magnesium
Manganese
Phosphate (ortho)
Phosphate (total)
Potassium
Selenium
Sodium
Sulfate
TSS
UV254
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
N/A
LFM
Acceptance
Limits
(% Recovery)
70-130%
70-130%
70-130%
70-130%
70-130%
80-120%
90-110%
70-130%
70-130%
70-130%
70-130%
70-130%
70-130%
70-130%
70-130%
70-130%
80-120%
N/A
N/A
MB(2)
Frequency
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
N/A
MB
Acceptance
Limits
-------
3.8.6.4.5 Completeness
Completeness refers to the amount of valid, acceptable data collected from a measurement
process compared to the amount expected to be obtained. Completeness was quantified
according to the following equation:
%C = (V/T)X100
where:
%C = percent completeness
V = number of measurements judged valid
T = total number of measurements
The completeness objective for data generated during this verification test was based on the
number of samples collected and analyzed for each parameter and/or method. Table 3-11
presents the completeness requirements based on the sampling frequency spelled out in the
test/QA plan.
Table 3-11. Completeness Requirements
Number of Samples per Parameter and/or Method Percent Completeness
(MO 80%
11-50 90%
> 50 95%
3.8.6.5 Operation and Maintenance
The EUWP was operated and maintained according to limits stated in Chapter 2 and the EUWP
Operation and Maintenance Manual.
49
-------
Chapter 4
Results and Discussion
4.1 Introduction
This chapter presents a summary of the water quality and operating data collected during the
verification test. Operating data are presented to describe the flow rates, volume of treated water
produced, backwash volumes and frequency, pressure differential across the UF and RO skids,
and related operating information. Water quality data are presented for the key parameters.
Information on the membrane integrity testing, microbial results, and particle counts are also
included in this Chapter. QA/QC information, as described by the QAPP in the TQAP for this
verification test, is presented at the end of the chapter.
4.2 Equipment Installation, Start-up, and Shakedown
The EUWP unit tested is maintained and stored at the NBVC. Therefore, the unit was already at
the testing site and required only minimal installation time. A series of calibrations checks and
UF integrity tests were performed prior to starting the verification test (see Section 4.4). Startup
and shakedown testing was started on September 27, 2007. The verification test began on
October 16, 2007, and ended on November 12, 2007. A post-test cleaning of the RO system
occurred on November 13 and 14, 2007.
4.3 Task A: Raw Water Characterization
Raw seawater directly from the Port of Hueneme was used for ETV testing, as discussed in
Section 1.3. The Port of Hueneme is the only deep-water port between Los Angeles and the San
Francisco. It has no appreciable fresh water outlets; therefore the water closely resembles that of
the Pacific Ocean with respect to salinity. Average water temperature ranges from 55 °F in the
winter months to approximately 62 °F in the summer. Raw seawater samples were collected in
2006 and 2007 for this task. The data for these samples are shown in Table 4-1.
Note that the metals were measured by both EPA Methods 200.7 and 200.8, and EPA Method
1640. Method 1640 includes a pre-concentration step that allows for lower detection limits in
seawater samples. The seawater detection limits for 200.7 and 200.8 are higher than those for
drinking water samples due to interferences from the high levels of sodium.
50
-------
Table 4-1. Initial Raw Water Characterization Sampling Results
Parameter
pH
Conductivity (umhos/cm)
TOC (mg/L)
UV254 (Absorbance/cm)
TSS (mg/L)
TDS (mg/L)
Alkalinity (mg/L as CaCO3)
Total Hardness (mg/L as CaCO3)
Nitrate (mg/L of N)
Nitrite (mg/L of N)
Total Silica (mg/L as SiO2)
Fluoride (mg/L)
Heterotrophic Plate Count (CFU/mL)
Total Coliforms (CFU/100 mL)
Metals by EPA 200.7 (all mg/L)
Calcium
Iron
Magnesium
Manganese
Potassium
Sodium
Metals by EPA 200.8 (all |ag/L)
Antimony
Arsenic (total)
Barium
Beryllium
Cadmium
Chromium
Copper
Lead
Lithium
Mercury
Rubidium
Selenium
Strontium
Tin
Thallium
Vanadium
Zinc
Metals by EPA 1640 (all |ag/L)
Aluminum
Antimony
Arsenic
Beryllium
Cadmium
Chromium
Cobalt
Copper
Iron
Lead
Manganese
Molybdenum
Sample Date
04/01/06 06/08/06 09/05/06 04/24/07
7.77 7.96 7.8
50,000 50,000 51,100
ND (0.3)
0.016
30
34,000 37,000 35,700
100
6,580
ND(40)
ND(10)
ND(2)
ND(O.l)
4
80
ND(20) ND(10) ND(25)
42 5.2 ND(50)
7.6 6.1 6
ND(10) ND(5) ND(5)
ND(20) ND(5) ND(5)
ND(30) ND(15) ND(25)
21 16 ND(10)
0.64 ND(5) 8
ND(0.2) 0.02 ND(0.02)
160 ND(50) 120
ND(10) 0.26 ND(5)
420
0.040
1,300
ND(0.050)
400
11,000
ND(25)
ND(50)
ND(50)
ND(250)
ND(IOO)
ND(50)
ND(50)
ND(50)
180
ND(IOO)
110
ND(IOO)
7,800
110
ND(10)
110
750
08/28/07
ND(6)
0.18
1.41
ND (0.01)
0.03
ND (0.05)
0.04
1.2
20.7
0.05
2.25
9.64
51
-------
Table 4-1. Initial Raw Water Characterization Sampling Results (continued)
Sample Date
Parameter
Nickel
Selenium
Silver
Thallium
Tin
Titanium
Vanadium
Zinc
04/01/06 06/08/06 09/05/06 04/24/07 08/28/07
0
0
ND
ND
ND
0
1
3
.24
.02
(0.04)
(0.01)
(0.01)
.81
.69
.49
4.4 Task B: Equipment Installation, Initial Test Runs and Initial System Integrity Tests
The objective of this task was to evaluate equipment operation and determine whether the
operating conditions result in effective treatment of the water. In this task, a preliminary
assessment of the treatment performance of the equipment was made. This task is considered a
shakedown testing period and was completed before the start of the verification test.
Initial equipment checks, UF integrity tests, required calibration checks, and initial test runs took
place between September 27, 2007 and the beginning of the official ETV test on October 16,
2007. During this period, sensors were calibrated, communications were established with the
particle counters and turbidimeters, and the PLC was operated to confirm programming and data
collection were operating properly. The in-line turbidimeters and conductivity meters were
calibrated. Handheld analyzers for pH, turbidity, and conductivity were checked and calibrated.
The system flow meters and pressure gauges were also calibrated during this pretest period.
A pressure decay test (integrity test) of the UF system was an important part of the initial test
runs to verify that the UF membranes and the connections were properly sealed. The first
pressure decay test of the UF system was performed on September 27. A pressure drop of 4 psig
occurred over a 20 minute period indicating that there was a leak or broken fibers in one or more
of the UF cartridges. Closer inspection found that air was leaking from Cartridge #7. This was
identified by the audible sound of air bubbles within the cartridge. Cartridge #7 was removed
from the skid and placed in a trough of water to perform a low-pressure decay test. This test
identified one broken fiber in Cartridge #7, which was repaired by plugging both ends of the
fiber.
On October 9, pressure decay tests were performed on each cartridge individually. These tests
were performed with the cartridges in place on the skid, and with the end caps in place. The
filtrate and feed outlets were capped and the cartridges filled with water. Each cartridge was then
pressurized to between 4.8 and 6.0 psig and the pressure monitored for 5 minutes. The pressure
drop over the five minutes ranged between 0.01 psig/min to 0.03 psig/min, except for Cartridge
#14, which showed a drop of 0.72 psig/min. This cartridge was removed from the system and
pressure tested in a trough with water. The identified leaks were pinned and the cartridge
reinstalled on the skid. Cartridge #14 was pressure decay tested again, and the decay rate was
measured at only 0.02psig/min. Table 4-2 shows the results of the individual cartridge low
pressure test.
52
-------
On October 10, the entire UF system was pressure decay tested once again. The system was
pressurized to 14.7 psig and the pressure drop recorded over a twenty minute period. The results
are shown in Table 4-3. These data were reviewed by NSF, and it was determined the UF system
was ready for testing.
Table 4-2. Results of Low Pressure Integrity Test on Individual UF Cartridges
Time
(minutes)
0
1
2
3
4
5
Avg.
pressure
drop/min
UF1
6.01
5.94
5.91
5.88
5.86
5.84
0.03
UF2
5.21
5.19
5.18
5.17
5.16
5.16
0.01
Applied Pressure (psig)
UF3 UF4 UF5
5.83 5.13 5.65
5.81 5.10 5.61
5.80 5.08 5.58
5.79 5.06 5.56
5.78 5.04 5.54
5.77 5.03 5.53
0.01 0.02 0.02
UF6
5.33
5.29
5.27
5.25
5.23
5.21
0.02
UF7
5.34
5.31
5.28
5.26
5.25
5.24
0.02
Time
(minutes)
0
1
2
3
4
5
Avg.
pressure
drop/min
UF8
4.82
4.80
4.79
4.78
4.77
4.76
0.01
UF9
5.00
4.98
4.96
4.94
4.93
4.92
0.02
Applied Pressure (psig)
UF10 UF11 UF12
5.26 5.14 5.06
5.24 5.10 5.05
5.22 5.06 5.04
5.21 5.03 5.04
5.19 5.00 5.03
5.17 4.98 5.02
0.02 0.03 0.01
UF13
4.94
4.92
4.91
4.90
4.89
4.88
0.01
UF14a(1)
5.45
4.26
3.34
2.63
2.15
1.85
0.72
Time
(minutes)
0
1
2
3
4
5
Avg.
pressure
drop/min
UF14b
5.18
4.75
4.37
4.03
3.74
3.44
0.35
UF14c
5.19
5.17
5.14
5.12
5.11
5.09
0.02
Applied Pressure (psig)
UF15 UF16
5.23 5.27
5.22 5.26
5.21 5.24
5.20 5.23
5.20 5.22
5.19 5.21
0.01 0.01
(1) Cartridge #14 was tested prior to repair, which is shown as UF14a. After repair, the cartridge was tested a
second time (UF14b), which it again failed. It was repaired a second time resulting in a successful test (UF14c).
53
-------
Table 4-3. October 10, 2007 UF Full System Integrity Test Results
Time
(min)
0
1
2
o
5
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Applied Pressure
(psig)
14.74
14.65
14.56
14.49
14.42
14.36
14.31
14.26
14.22
14.17
14.14
14.10
14.07
14.03
14.00
13.97
13.94
13.91
13.89
13.86
13.83
Pressure Drop
(psig/min)
NA
0.09
0.09
0.07
0.07
0.06
0.05
0.05
0.04
0.05
0.03
0.04
0.03
0.04
0.03
0.03
0.03
0.03
0.02
0.03
0.03
The UF and RO system were run for short periods of time during the pretest period, as part of the
calibration and equipment checks for the flow meters, conductivity and turbidimeters, and
particle counters. On October 11, the entire system was operated for several hours to ensure all
systems were fully operational. There were no additional lengthy runs made prior to the start of
the verification test on October 16. The RO system was operated to ensure the target flows could
be attained, but no additional checks or pretest membrane integrity tests were performed on the
RO membranes. The in-line conductivity meters were monitored at the start of the verification
test to confirm the rejection rate of the RO membranes. The conductivity measurements made on
the first day of testing showed a salt rejection rate of 98.8%.
It was known from past experience that the treatment of seawater would require the use of a
coagulant prior to the UF system and the use of an antiscalant to reduce scaling on the RO
membranes. While it is often necessary and helpful to run jar tests to determine the optimal
coagulant dose, in this case, no jar tests were performed prior to the verification test. Ferric
chloride was selected as the coagulant based on previous pilot testing experience. In 2004 a pilot
study was performed at Port Hueneme by TARDEC, in conjunction with Koch Membrane
Systems to investigate the use of UF membranes for seawater pretreatment upstream of RO
treatment. This study also evaluated the use of ferric chloride to improve UF performance. The
results from this study were used to set the initial target dosage of ferric chloride at 0.75 mg/L as
Fe. The antiscalant selected for use at this test site was ONDEO (Nalco) PermaTreat® PC-191.
The initial target dose rate was 5 mg/L.
54
-------
On October 15, the day before the verification test was scheduled to start, TARDEC scheduled a
final preliminary run with all systems operational; however, the high pressure RO pump would
not start. A service person was called and arrived at the site the same day. A problem was found
with the fuel pump. A replacement fuel pump was obtained the next morning and the system was
up and running by 11:00 a.m. on October 16.
4.5 Task C: Verification Test
The verification test was started on October 16, 2007 and ran for 28 calendar days (27 24-hour
periods), to November 12, 2007. The test had been scheduled to run for thirty days, or until the
UF system required at least two CIPs. The continuous operation portion of the verification test
was stopped on November 12, as the UF system had been cleaned four times as of November 9.
On November 13 and 14, the RO membranes were cleaned and post-cleaning operational data
was collected.
The UF system was operated each day on continuous basis, except for shutdowns for integrity
testing and routine maintenance. The UF system also automatically shut down when the RO feed
tank was full. The biggest impact on overall UF operating hours was the need to perform
chemical cleaning of the UF membranes four times during the test. A typical operating day for
the UF system was 19 to 22 hours. The mean UF operating hours per day over the entire test was
18.6 hours with a median of 19.8 hours, as shown in Table 4.4. Note that the count in Table 4-4
for the hours per day figures is only 19. For the first two weeks of the test, and also the last two
days, the operators did not record the operation hours on most of the sheets for daily operation
data.
The RO system was also setup to operate continuously, except for routine maintenance periods
and times when the UF was shutdown for integrity testing, maintenance, or cleaning. A typical
operating day for the RO system with no significant maintenance issues was 21 to 23 hours. The
mean RO operating hours per day over the entire test was 17.1 hours with a median of 19.0
hours. When alarms and shutdown occurred during unattended operation at night, the entire
system would remain shut down until an operator arrived in the morning.
The on-site operators typically collected operating data and on site water quality samples a
minimum of twice per day in accordance with the test plan schedule. The following sections
present the operating data and water quality data.
4.5.1 Task Cl: Membrane Flux and Operation
The purpose of this task was to evaluate system performance during operation. The objectives of
this task were to demonstrate the appropriate operational conditions for the system, the feed
recovery achieved by the UF and RO membranes, and the rate of flux decline observed over the
operation period.
Operational data were collected and on-site water quality measurements were made two or more
times per day throughout the test, except for three days associated with UF cleaning when only
one set of data was obtained. The data were summarized for presentation and discussion in this
55
-------
section. The complete set of data sheets can be found in Appendix B. The data spreadsheet with
the calculations is Appendix C.
4.5.1.1 UF Operating Data
4.5.1.1.1 UF Flow Rate, Filtrate Production, and IMP Results
The UF operational statistics are presented in Table 4-4. The UF skid does not have a filtrate
flow meter or filtrate pressure gauge. Therefore, the total filtrate flow was calculated as the UF
feed flow rate minus the UF retentate flow. The intake flow is the intake from the source water
into the UF feed tank. The intake pump is technically not part of the UF skid, but the intake flow
is included here as part of the overall UF treatment process. The intake pump ran at a higher flow
rate than the UF system to ensure that the UF feed water tank always contained sufficient water
to operate the UF system.
Table 4-4. UF Operational Data Statistics
Parameter
UF Operation per day (hr)
Raw Intake Flow (gpm)
Feed Flow (gpm)
Filtrate Flow (gpm)
Retentate Flow (gpm)
Backwash Flow (gpm)
Feed Pressure (psig)
Retentate Pressure (psig)
Filtrate Temperature (°F)
Count
19
74
74
74
74
74
74
74
Mean
18.6
287
249
222
26
900 gal
20.6
16.3
58.3
Median
19.8
288
251
225
26
Minimum
7.3
272
212
187
25
per backwash cycle0 };
20.0
16.0
59.0
14.0
10.0
55.0
Maximum
22.7
296
279
252
34
Standard
Deviation
4.11
4.98
11.4
10.9
1.66
95%
Confidence
Interval
+ 1.85
+ 1.13
+ 2.60
+ 2.48
+ 0.38
Backwash every 30 minutes
30.0
23.0
61.0
3.74
2.89
1.62
+ 0.85
+ 0.66
+ 0.37
(1) Volume not measured. It was provided by the manufacturer.
The mean UF feed flow of 249 gpm was below the design feed flow of 259 gpm specified for the
system (See Table 3-4). The mean filtrate flow of 222 gpm corresponds to a flow of 13.9 gpm
for each of the 16 UF membrane modules. The UF water recovery was 89.2% based on the
mean feed and filtrate flows.
Figure 4-1 shows the UF system flow rates over the duration of the verification test. The
retentate flow remained steady throughout the test. The feed flow and filtrate flow dropped as the
membranes became fouled with solids and TMP increased. Manual adjustment of the flow
control valve was made to hold the feed and filtrate flows as steady as possible.
56
-------
300
250 --
200
o.
* 150
06
o
E
100
50
0
-Feed -"-Retentate -A-Filtrate
10/16/07 10/21/07 10/26/07 10/31/07 11/05/07
Date
Figure 4-1. Plot of UF system flow rates through the testing period.
11/10/07
Total UF filtrate production was tracked using the RO feed totalizer. This production volume
was the actual filtrate used for the RO feed and does not include the filtrate used for backwash
water. The net filtrate production over the 27-day test period was 4,673 kgal, which represents an
average production rate of 173 kgal/day. The total UF filtrate volume (including filtrate used for
backwash) produced was 5,249 kgal, which gives an average total production rate of 194.4
kgal/day. This production rate includes the two days when the UF was not operated due to
cleanings, and includes the other days with limited production due to cleaning or system
maintenance issues. Figure 4-2 shows the cumulative total and net filtrate production for the UF
system over the duration of the verification test.
57
-------
UF Filtrate Production -•- UF Filtrate for RO Feed
0
10/16/07 10/21/07 10/26/07 10/31/07 11/05/07
Date
Figure 4-2. UF system filtrate production through the testing period.
11/10/07
Figure 4-3 shows the feed and retentate pressures during the test and Figure 4-4 shows the
calculated TMP results. These figures depict the impact of solids build up on the UF membranes
during operation.
A chemical coagulant (ferric chloride) was added to the UF feed to improve operation of the UF
system and to lengthen run time between chemical cleanings. The coagulant addition was
planned for a feed rate of 4.37 ml/min (0.07 gal/h), which would yield an iron dose (as Fe) of
0.75 mg/L in the UF feed (4.6 X 10"6 gal ferric chloride per gal of feed). The chemical feed pump
stroke and speed were calibrated and checked daily. In addition, the level in the ferric chloride
feed tank was recorded at least twice per day and records were maintained of the ferric chloride
added to the feed tank. Based on the tank records, a total of 22.4 gal of ferric chloride were fed
into 5,259,625 gal of feed (4.3 X 10"6 gal ferric per gal of feed), which is approximately 10% less
than the feed rate measured by the pump calibration.
58
-------
10/16/07
10/21/07
10/26/07
10/31/07
Date
11/05/07
11/10/07
Figure 4-3. Plot of UF system feed and retentate pressures over the testing period.
10/16/07
10/21/07
10/26/07
10/31/07
Date
11/05/07
11/10/07
Figure 4-4. Plot of UF system TMP over the testing period.
59
-------
4.5.1.1.2 Discussion - UF Flow Rate, Filtrate Production, and IMP
As discussed in Section 2.1, the maximum UF production rate is 250,000 gpd (not including
backwash water). Based on the net filtrate production over the 27-day verification period, the UF
system produced on average 173 kgal/day.
The EUWP included a totalizer to track the hours of UF system operation. The primary impact
on total operating hours over the 27-day test was the need to clean the UF membranes four times
during the test. There were two cleaning periods when the unit was down for more than one day,
yielding two days out of 27 with no operation. The other two cleaning periods resulted in the
daily operating hours averaging approximately 9 hours per day for the two-day cleaning periods
(each cleaning requires an overnight soak, so the cleaning covers two days). The hours of
operation varied widely, from 7.3 to 22.7 hrs, depending on the downtime for various
maintenance activities and verification related testing. The UF system was operated an average
of 18.6 hours per day with a median of 19.8 hours per day. However, a typical operation day
with no significant maintenance issues netted 21-23 hours of operation.
The first UF cleaning occurred when the TMP had only increased to 16 psig compared to a target
of 20 psig (Note: actual equipment specification says to clean when TMP exceeds 30- 35 psig).
The operators had noted an increase in feed pressure and read by mistake a backwash TMP
readout that was above 20 psig. Thus, this first cleaning was performed earlier than required.
Subsequently, as shown in Figure 4-4, the TMP increased over the next 5 to 7 days and the UF
membranes required another chemical cleaning, as the normal backwash cycle was not
sufficiently cleaning the UF membranes. After each cleaning was completed, flow rates and
TMP returned to normal ranges and similar to the values measured at the beginning of the test.
4.5.1.1.3 UF Specific Flux Results and Discussion
Figure 4-5 shows the specific flux calculations for the UF system during the test. The impact of
solids buildup on the system is clear, especially for the last three cleaning cycles. The CIPs were
successful, as the specific flux was improved after each cleaning event, but they were not able to
restore the specific flux to that at time 0. For the last cleaning event on November 9 the FTO
added an overnight soak with the low pH solution in addition to the standard overnight soak with
the high pH solution. This bolstered cleaning procedure returned the membranes to a specific
flux of 4.23 gfd/psig, which was identical to that at time 0.
Figure 4-6 shows the change in specific flux over the duration of the verification test. The
change in specific flux is calculated by comparing the specific flux on a given day to the value
calculated at the start of the test. This data shows the impact of cleaning and backwashing by
comparing a given day's specific flux to the start of the test. As can be seen, there was a steady
loss of specific flux between cleanings.
60
-------
4.5
4.0
3.5
3.0
2.5
-LTF System Cleaned -
-S"
•
S 2.0
1.5
1.0
0.5
0.0
0»
o.
10/16/07
10/21/07
10/26/07
10/31/07
Date
11/05/07
11/10/07
Figure 4-5. UF system specific flux over testing period.
10/16/07
10/21/07
10/26/07
10/31/07
Date
11/05/07
11/10/07
Figure 4-6. Change in specific flux over time.
61
-------
4.5.1.1.4 Power Supply - Fuel Consumption
For this verification test, the generator that is part of the field portable system was used to
demonstrate that the generator could sustain UF operation on a regular basis. The diesel powered
60 kWh generator supplies power to the UF skid and to the ancillary systems on the RO skid.
The RO high-pressure pump has its own diesel engine, or can operated with an electric pump.
The diesel fueled RO pump was used for this test.
The UF power requirements are stated as approximately 31 kWh when operating at full capacity
or 2.1 kWh/kgal. The generator operated throughout the test and provided adequate power for the
UF system and ancillary systems on the RO skid. Figure 4-7 shows the fuel consumption during
the test. The total fuel consumption was 3,091 gal of diesel fuel over the 27-day test. The
generator and the high pressure RO engine used the same fuel tank, so fuel usage figures are total
usage for both systems. The EUWP was actually operated for 25 of the 27 days, yielding an
average fuel consumption of 124 gpd with peak usage of 180 gpd.
3500
3000 -
2500
v25
| 2000
o.
o 1500
"3
=
1000
500
0
10/16/07
10/21/07
10/26/07
10/31/07
Date
11/05/07
11/10/07
Figure 4-7. Diesel fuel consumption.
4.5.1.2 RO System Operational Data
4.5.1.2.1 Flow rates, Operating Pressures and Percent Recovery Results
The RO operational statistics for the verification test are presented in Table 4-5. The RO system
has flow meters and pressure gauges to monitor the feed and permeate for Array 1. The
concentrate flows for Array 1 were calculated as the difference between the feed flow and
62
-------
permeate flow. Array 2 has flow meters for the permeate and concentrate, and gauges to monitor
pressure for the feed, permeate, and concentrate. The feed flow for Array 2 was calculated by
adding the permeate and concentrate flows. The UF system supplied all of the feed for the RO
system.
Table 4-5. RO System Operational Measurement Statistics
Parameter
RO Operation Hours per Day (h)
Array 1 Feed Flow (gpm)
Array 1 Permeate Flow (gpm)
Array 1 Concentrate Flow (gpm)
Array 2 Feed Flow (gpm)
Array 2 Permeate Flow (gpm)
Array 2 Concentrate Flow (gpm)
Array 1 Feed Pressure (psig)
Array 1 Concentrate Pressure (psig)
Array 2 Feed Pressure (psig)
Array 2 Concentrate Pressure (psig)
Array 1 and 2 Combined Permeate
Pressure (psig)
Count
25
74
74
74
74
74
74
70
74
70
74
74
Mean
17.1
115
70
45
63
32
31
957
905
901
868
23.4
Median
19.0
115
70
45
63
32
31
960
903
900
865
23.5
Min.
4
112
68
43
56
25
30
927
870
880
850
21.0
Max.
24
117
72
48
68
37
32
977
992
936
885
28.5
Standard
Deviation
6.12
0.74
0.82
1.03
2.05
2.11
0.36
10.8
15.5
11.0
7.65
1.34
95%
Confidence
Interval
±2.40
±0.17
±0.19
±0.23
±0.47
±0.46
±0.08
±2.39
±3.53
±2.58
±1.74
±0.31
The RO system maintained a steady permeate flow rate for both arrays throughout the
verification test. Figure 4-8 shows the daily flows for permeate and concentrate for both arrays.
Figure 4-9 shows the feed and concentrate pressures for both arrays. Feed pressure remained
steady over the duration of the test. The concentrate pressure from Array 1 was used by the
energy recovery device to provide the feed pressure for Array 2. This energy saving device
eliminated the need for a high pressure pump for the Array 2 flow, which was approximately
55% of Array 1. Without the energy saving device, additional pumping capacity and the
associated energy use would be required. The energy saving device achieved feed pressures that
were similar to the Array 1 pressures throughout the test. Based on the permeate flows from
Array 2 representing 31% of the RO water production (mean of 32 gpm out of an average of 102
gpm total), it can be roughly estimated that the energy conservation device saved 31% of the
energy that would have been required if all the permeate was produced by high pressure pumps.
63
-------
80
20
10
0
10/16/07
-Array 1 Permeate Array 2 Permeate
-Array 2 Concentrate -"-Array 1 Concentrate
10/21/07
10/26/07
10/31/07
Date
11/05/07
11/10/07
Figure 4-8. RO system flow rates.
1000
980
960
940
920
900
880
860
840
Array 1 Feed Array 2 Feed -A-Array 1 Concentrate -"-Array 2 Concentrate
10/15/07
10/20/07
10/25/07
10/30/07
Date
11/04/07
11/09/07
11/14/07
Figure 4-9. RO system operating pressures.
64
-------
Figure 4-10 shows the percent recoveries achieved by the RO system. Recoveries, calculated as
the permeate flow divided by the feed flow, were consistent throughout the test. The average
percent recovery for Array 1 was 61% with a median of 61%. The mean recovery for Array 2
was 50% with a median of 50%. As expected, the recoveries for Array 2 were lower than for
Array 1, as Array 2 operates at a lower feed pressure.
70% ^
60% -
0%
10/16/07
10/21/07
10/26/07
10/31/07
Date
11/05/07
11/10/07
Figure 4-10. RO system percent recoveries.
4.5.1.2.2 Flow rates, Operating Pressures and Percent Recovery Discussion
The mean feed flows of 115 gpm for Array 1 and 63 gpm for Array 2 were close to the target
feed rates of 116 gpm for Array 1, and 58 gpm for Array 2 listed in Table 3-4. The Array 1
recovery of 61% exceeded the target specification of 50%. The Array 2 recovery of 50% also
exceeded the target specification of 48%. These recoveries, in conjunction with the feed targets,
resulted in mean permeate flows of 70 gpm for Array 1 and 32 gpm for Array 2. At these flows,
the RO unit would need to operate an average of approximately 16.3 hours per day to meet the
claimed target of 100,000 gpd.
Over the 27-day verification test, the RO feed totalizer showed 4,673.3 kgal was fed to the RO
unit. Based the daily recoveries for each Array (typically Array 1 at 61% and Array 2 at 50%),
the total volume of permeate produced was approximately 2,671 kgal, giving an average of 98.9
kgal/day over 27-day test. This was close to the target production rate of 100,000 gpd.
65
-------
The primary reason the RO system did not meet the production goal of 100 kgal/day was lack of
feed when the UF system was shut down for cleaning. It should be noted that in addition to the
impact that frequent cleaning had on the overall UF system water production, the UF system also
shutdown anytime the RO system feed tank was full. The test was designed to verify the entire
system with both UF and RO in operation. The UF system produced enough water to meet the
100 kgal per day production goal, but because of limited UF filtrate storage capacity, long
downtime periods for the UF system cleaning did impact the RO production. With more storage
capacity for UF filtrate, the UF system would have been able to meet the feed requirements for
the RO system to achieve the overall goal of producing 100 kgal/day, even with the more
frequent cleaning schedule.
Whenever, there was feed available, the RO system operated continuously producing permeate at
a flow rate of 100 to 102 gpm. The RO system operated for more than 20 hours on 12 of the 25
actual operating days. The mean RO operating hours per day was 17.1, with a median of 19.0
hours per day. These mean and median hours match closely to the UF hours (mean 16.9 h and
median 19.1 h). The maximum RO operating hours was 24 hours and the minimum was 4 hours.
Antiscalant was added to the RO feed throughout the test. The mean dose rate was 5.7 mg/L
versus a target feed of 5 mg/L. The RO system did not seem to experience any scaling or fouling
problems during the test. The S&DSI for the concentrate water was calculated in accordance
with ASTM procedure D4582. This index can be used to determine the need for calcium
carbonate scale control measures in the operation of an RO system. Direct measurements of the
ions (Ca, Mg, Na, K, Cl, SO/t, Alkalinity) and pH of the concentrate stream provided data for the
calculation without the need to estimate these concentrations from the feed data. The S&DSI
varied from -0.71 to -0.84 during the test. This indicates that the concentrate was a non-scaling
water (S&DSI <0.0 is non-scaling). The S&DSI results for each week of the test and the
supporting water quality data are presented later in Table 4-12 in Section 4.5.3.2. The
combination of non-scaling water and the addition of antiscalant reduced or eliminated the
problems of scaling on the RO membranes.
As shown by the flow rate and pressure results, the system operated consistently throughout the
test with little change in flows or pressures. This would suggest that for this water source, the RO
could have met and exceeded the production feed targets, if sufficient water could have been
provided from the UF system. The buildup of solids on the UF system and need for frequent UF
system cleaning was the limiting factor over the verification test period.
4.5.1.2.3 RO Specific Flux - Results and Discussion
A common method of evaluating RO membrane performance is to calculate the specific flux,
which normalizes the permeate flux based on net driving pressure. The calculation of net driving
pressure (NDP) that is used in the determination of specific flux includes the calculation of
osmotic pressure. A correlation between TDS and conductivity was calculated, and this
correlation was then used with the daily conductivity measurements to calculate TDS values for
the osmotic pressure equation. The equation for the line determined for this correlation was:
y(TDS) = 733.98x(conductivity), (R2 = 0.9936)
66
-------
Figure 4-11 shows the specific flux for the two RO system arrays based on NDP and adjusted to
a temperature of 25 °C. The consistency of the specific flux over the test period further indicates
that the RO membranes were not being fouled over the time. Given that the membranes were still
functioning at the end of the test at a specific flux that was 97% of the starting specific flux, it
cannot be projected when the membranes would require cleaning.
0.045
0.040
0.000
10/16/07
10/21/07
10/26/07
10/31/07
Date
11/05/07
11/10/07
Figure 4-11. RO system specific flux.
The RO system was chemically cleaned at the end of the test on November 13 and 14. This
cleaning was performed because it was a requirement of the verification test to demonstrate the
cleaning process; even though the RO system had not actually reached its target cleaning level
criteria. Data on the cleaning is provided in Section 4.5.2.2.
4.5.1.2.4 RO System Power
The RO system uses either an electric high-pressure pump or a diesel fuel high-pressure pump
engine to pressurize the RO feed. The UF power generator described earlier provides all other
power for the RO system. For this test, the diesel fuel high-pressure pump engine was used. As
described in Section 4.5.1.1 under the Power Supply - Fuel Consumption heading, a common
fuel tank provided fuel to both the generator and the RO engine. A total of 3,091 gal of diesel
fuel was used through November 12, 2007, which was the end of the continuous flow portion of
67
-------
the test. An additional 100 gal of diesel fuel was used on the last two days when the RO
membranes were cleaned. Figure 4-7 shows the cumulative fuel usage over the verification test.
4.5.2 Task C2: Cleaning Efficiency
An important aspect of membrane operation is the ability to achieve long run times between
chemical cleanings (to maintain operation time and minimize chemical use) and to restore
membrane production after flux decline due to buildup of solids on the membrane and in the
membrane pores. The objective of this task was to evaluate the membrane cleaning procedures
and determine the fraction of specific flux restored following chemical cleaning.
4.5.2.1 UF Backwash and Cleaning Frequency and Performance
The UF system is designed to be backwashed automatically after every 30 minutes of operation.
The backwash is designed to remove solids that have accumulated on and within the membrane.
Frequent effective backwashes provide restoration of water production and lengthen the time
until chemical cleaning is required. The automatic backwash system reverses the flow through
the membrane to remove material accumulated on the membrane surface, and then a fast forward
flow flush is performed to clear the membrane. The system uses UF filtrate water for the
backwash cycle.
The automatic backwash system functioned properly during the verification test. The automatic
cycle initiated on schedule once every 30 minutes, as programmed, and the entire process was
automated. The backwash cycle counter tracked the number of backwashes performed during the
test. The backwash system used 900 gal of filtrate for each backwash cycle (volume provided by
the manufacturer). Based on the number of backwashes performed and the flow rates achieved in
the verification test, the backwash system used approximately 11-12% of the filtrate produced by
the UF system
It had been expected that the UF system would require chemical cleaning about every 15 to 30
days. However, the UF system actually was cleaned four times during the four-week test.
The first CIP for the UF system occurred on October 20, four days after startup. The specific flux
had dropped from 4.23 gpd/psig to 2.96 gpd/psig and the TMP had increased from 10 psig to 16
psig. The UF system was still producing filtrate at an acceptable rate and TMP had not reached
the target cleaning level of 20 psig. The operators had noted an increase in feed pressure and read
by mistake a backwash TMP readout that was above 20 psig. The FTO decided to clean the unit.
Thus, this first cleaning was performed earlier than required. Both a low pH and high pH
cleaning were performed. Following an overnight soaking with the high pH solution, the UF was
restarted. The specific flux increased to 3.64 gpd/psig and the TMP decreased to 12 psig. The
CIP was considered successful with an 86% recovery of specific flux.
The UF system ran from October 21 to October 28, but showed steady decrease in specific flux
and increase in TMP, as shown in Figures 4-4 and 4-5. At the end of the seven days the specific
flux had decreased to 1.82 gpd/psig and TMP had increased to 22 psig. A CIP was started on
October 28, and after an overnight high pH soaking of the membranes, the procedure was
completed on October 29. The CIP restored specific flux to 2.99 gpd/psig (71% recovery) and
decreased TMP to 14 psig.
68
-------
Two additional cleanings were required to maintain the UF system, on November 3-4 after five
more days of operation and on November 8-10 after four more days of operation. The CIP on
November 3-4 showed a specific flux recovery of only 65% and lowered the TMP to 15 psig.
The gradual decrease in cleaning performance was a concern so the procedure for the November
8-10 CIP was changed to add a low pH overnight soak followed by a regular low pH cleaning
and then a high pH over night soak. This resulted in increased down time, but it was felt that the
membranes needed to be restored closer to the original operating conditions. The November 8-10
CIP resulted in a specific flux recovery of 100%. Figures 4-4 and 4-5 show the TMP and specific
flux before and after these cleaning procedures.
Figure 4-6 shows the change in specific flux over the duration of the verification test. The
change in specific flux is calculated by comparing the specific flux on a given day to the value
calculated at the start of the test. This type of data shows the impact of cleaning and backwash by
comparing any given day's specific flux to the start of the test. As can be seen, the UF system
was being consistently fouled every five to six days. As discussed in the previous section on UF
production, this frequent cleaning resulted in significant down time for the system and reduced
the capacity for the UF system for filtrate production and subsequent RO production.
The UF CIP procedure uses three chemicals, citric acid for the low pH cleaning, and sodium
hydroxide and sodium hypochlorite (bleach) for the high pH cleaning. The amount of citric acid
and sodium hydroxide needed to make a pH 3 or pH 11 cleaning solution varied for each
cleaning. Additional chemical was added as needed during the recirculation step to maintain the
pH and chlorine concentration. The target chlorine concentration was 100 to 200 mg/L. Table 4-
6 shows the amount of each chemical that was used for each cleaning. The CIP mixing tank
contained 270 to 300 gal.
The UF cleaning solution was heated in the CIP tank with the low pH solution ranging from 35
to 39 °C and the high pH solution 32 to 37 °C. Each bank of modules was circulated with each
solution for 20 to 30 minutes. The membranes were then soaked overnight with the high pH
solution.
Table 4-6. UF System CIP Cleaning Solution - Chemical Use
Date
Oct. 20 to 21
Oct. 28 to 29
Nov. 3 to 4
Nov. 8 to 10
Citric Acid
(Solid, Lb.)
4
6
10
1 1.2/1 1.2(1)
Sodium Hydroxide
(0.5%, L)
2.4
2.0
3.8
3.8
Bleach
(12.5%, L)
12.6
12.
21.
29.
(1) Two low pH cleanings were performed with an overnight soak for each.
4.5.2.2 RO Cleaning Frequency and Performance
The RO system was cleaned on November 13 at the end of the verification test using both a low
pH and high pH cleaning. This cleaning was not required based on the system operating data, as
the specific flux for Array 1 had only decreased from an initial value of 0.0313 gpd/psig to
0.0303 gpd/psig (3% drop), and the Array 2 specific flux had only decreased from 0.0304
69
-------
gpd/psig to 0.0302 gpd/psig (1% drop). The verification test, however, required a demonstration
of the cleaning process, so it was performed at the end of the test. Figure 4-11 showed the
specific flux for the two RO system arrays over the duration of the test. These data show no
indication that the membranes were being fouled or that scaling was occurring. Based on these
data it is not possible to project when a cleaning would be required, but clearly the RO system
could sustain long run times, in excess of 30 days, in this specific application with the given
water supplied from the UF system.
The RO cleaning was performed using Avista RO Cleaner P303 for the low pH cleaning. Fifty-
four pounds of the RO cleaner were added to the 300-gal CIP tank, which resulted in a solution
with pH 3.24. The system was circulated for one hour and then the system was flushed in
preparation for the high pH cleaning. Avista RO Cleaner PI 11 was used for the high pH cleaning
solution, with 54 pounds of PI 11 added to 300 gal, yielding a pH of 10.2. The high pH cleaning
solution was circulated for two and a half hours and then the RO system was flushed and
shutdown for the night.
Table 4-7 shows the specific flux results for Arrays 1 and 2 before and after the CIP procedure.
The CIP restored Array 1 from a time 0 specific flux of 0.0341 before the cleaning, to a mean
post-cleaning flux of 0.0334, which is a recovery of 99%. For Array 2, the CIP actually restored
the membranes to an average specific flux of 0.0352, which was higher than the initial value of
0.0334, yielding a recovery of 105%.
Table 4-7. RO System Specific Flux Before and After CIP
Date
10/16/07
11/12/07
11/13/07
11/14/07
11/14/07
11/14/07
11/14/07
11/14/07
11/14/07
Time
Day 0 of test
Last Day of test
9:30
10:50
12:05
13:20
14:30
15:45
Mean post-cleaning specific flux
Array 1 Specific Flux
(gpd/psig)
0.0341
0.0328
RO System Cleaned
0.0332
0.0335
0.0335
0.0333
0.0336
0.0335
0.0334
Array 2 Specific Flux
(gpd/psig)
0.0334
0.0330
0.0365
0.0340
0.0340
0.0372
0.0343
0.0353
0.0352
4.5.3 Task C3: Water Quality Results
The primary objective of this task was to assess the ability of the membrane equipment to meet
the water quality goals, which were established as producing water that meets the USEPA
NPDWR. Several water quality parameters were selected as indicator parameters to demonstrate
the performance of the UF and RO membranes. Turbidity and conductivity were selected as two
key parameters, as turbidity removal by the system would indicate the ability to remove
paniculate related contaminants, and a reduction in conductivity (indicator of total dissolved
solids content) would show the ability of the RO system to remove dissolved contaminants. Both
turbidity and conductivity were continuously monitored using in-line meters in the EUWP, and
70
-------
grab samples were also measured onsite with portable equipment at least twice per day.
Temperature and pH were also measured onsite at least twice per day. Other water quality
parameters were monitored by collecting grab samples on a weekly basis. These parameters
included TDS, TSS, Alkalinity, Hardness, Bicarbonate, Chloride, Sulfate, Calcium, Magnesium,
Sodium, Potassium, Lithium, Boron, Barium, Selenium, Total phosphorus, Ortho-phosphate, and
UV 254 absorbance.
This section presents the water quality results for the verification test. Data on the bacteriological
samples (total coliforms and Bacillus endospores) are presented and discussed in Section 4.5.4,
Task C4: Membrane Module Integrity.
4.5.3.1 Water Quality Results - Turbidity, Conductivity, pH, and Temperature
Figures 4-12 and 4-13 graphically present the grab sample turbidity results for the raw water and
UF filtrate over the duration of the test. Table 4-8 shows this data in a tabular format, as well as
the turbidity summary statistics. Note that the data set begins on October 17, which was the
second day of the test. The October 16 turbidity readings are not included in the table because
there was a problem with the turbidimeter on that day. Over the course of the test, the UF system
reduced turbidity from a mean of 1.34 NTU in the raw water to a mean of 0.06 NTU in the UF
filtrate. The 95% confidence level shows that filtrate turbidity can be expected to be in the range
of 0.05 to 0.07 NTU. Turbidity in the raw water was reduced by a mean value of 94.9%, with a
median reduction of 95.6% through the UF system. Turbidity levels met the NPDWR
requirements of <0.3 NTU 95% of the time and all values below 1.0 NTU, except for the first
day of testing as noted above.
As discussed above, the EUWP includes in-line turbidity meters that measured the turbidity of
the raw water, UF filtrate, and RO permeate every 15 minutes as a means of monitoring
membrane integrity. The raw water and UF filtrate readings are graphed in Figure 14. Note that
there are two y-axes (different scales) in the figure, one for the raw water, and one for the UF
filtrate. Also, the gaps in the data correspond to when the system was down for UF cleanings.
The raw water turbidity, as measured by the in-line analyzer, had a mean value of 1.38 NTU, and
a median of 1.32 NTU. The in-line turbidity data for the UF filtrate had a mean of 0.019 NTU,
and a median of 0.018 NTU. Table 4-9 shows the summary statistics for the raw, UF filtrate, and
RO permeate in-line turbidity readings. All of the individual measurements from in-line turbidity
meters are listed in Appendix D.
The LT2ESWTR states that if the turbidity exceeds 0.15 NTU over any 15-minute period, the
system must be shut down and a direct integrity test performed. Since the data logger recorded
turbidity readings every 15 minutes, the evaluation criterion was two consecutive turbidity
measurements exceeding 0.15 NTU. There were only three single data points where the UF
filtrate turbidity exceeded 0.15 NTU. In each instance, the previous and following turbidity
values were significantly below the 0.15 NTU level. Based on these data, it appears that the UF
system did not exceed the LT2ESWTR action level during the verification test. It should be
noted that the EUWP was not setup to be compliant with the LT2ESWTR, as the in-line turbidity
meters were not tied to an automatic system shutdown if the turbidity level exceeded 0.15 NTU
for any 15 minute period. The in-line turbidity data was logged onto a laptop computer, which
was not connected to the EUWP for the purpose of shutting down the system. Also, it should be
71
-------
noted that the in-line turbidity meters continued to operate during UF backwash periods, and
thus, the spikes in turbidity have been caused by measurements of backwash water. The three
instances during the test when turbidity exceeded 0.15 NTU for one 15 minute reading, the
readings before and after the elevated reading were typically ten times lower, suggesting the
single high turbidity readings were most likely due to a backwash occurring at the same time that
the in-line turbidity unit reading was being recorded.
The RO system had little additional impact on the turbidity levels, with the RO permeate having
a mean turbidity of 0.05 NTU, based on the grab samples collected each day. The in-line RO
analyzer showed a mean turbidity of 0.013 NTU with a median of 0.012 NTU. The final treated
water, the RO permeate, met the NPDWR turbidity requirements (<0.3 NTU 95% of the time
and all values below 1.0 NTU), except for the first day of testing as noted above. Similar to the
UF system, the RO system produced permeate with turbidity below the LT2ESWTR action level
of 0.15 NTU throughout the test. There were only three single data points above the action level,
and at no time were there two consecutive 15-minute readings above the 0.15 NTU action level.
o
10/15/07 10/20/07 10/25/07 10/30/07
Date
Figure 4-12. Grab sample UF feed turbidity data
11/04/07
11/09/07
11/14/07
72
-------
0.25
0
10/15/07 10/20/07 10/25/07 10/30/07 11/04/07 11/09/07
Date
Figure 4-13. Grab sample UF filtrate turbidity data.
11/14/07
10/16/07 10/21/07 10/26/07 10/31/07 11/05/07 11/10/07
Date
Figure 4-14. UF feed and UF filtrate in-line turbidity readings.
73
-------
Table 4-8. Turbidity Results, On-Site Bench Top
Date
10/17/07
10/17/07
10/18/07
10/19/07
10/21/07
10/21/07
10/22/07
10/22/07
10/22/07
10/23/07
10/23/07
10/23/07
10/24/07
10/24/07
10/24/07
10/25/07
10/25/07
10/25/07
10/26/07
10/26/07
10/26/07
10/26/07
10/27/07
10/27/07
10/27/07
10/27/07
10/28/07
10/29/07
10/29/07
10/29/07
10/30/07
10/30/07
10/30/07
10/30/07
10/31/07
10/31/07
10/31/07
10/31/07
11/01/07
11/01/07
11/01/07
11/01/07
11/02/07
11/02/07
11/02/07
11/02/07
Raw Water
(NTU)
1.28
1.52
1.25
1.23
1.41
1.91
2.01
1.78
3.21
0.62
1.44
1.11
1.58
1.15
1.41
1.42
1.14
0.99
1.06
0.82
1.37
1.61
0.97
1.09
1.01
1.17
1.01
1.76
2.84
1.75
1.29
1.80
1.01
1.18
1.46
1.41
0.88
1.08
1.15
0.72
0.64
1.26
0.87
0.93
0.85
1.04
UF Filtrate
(NTU)
0.08
0.14
0.07
0.08
0.16
0.07
0.05
0.13
0.09
0.05
0.10
0.18
0.06
0.06
0.07
0.07
0.05
0.05
0.04
0.04
0.05
0.05
0.06
0.04
0.05
0.15
0.04
0.15
0.05
0.05
0.11
0.08
0.07
0.08
0.04
0.21
0.04
0.04
0.04
0.05
0.05
0.05
0.05
0.05
0.04
0.05
RO Feed
(NTU)
0.05
0.10
0.03
0.12
0.05
0.12
0.06
0.05
0.06
0.05
0.06
0.05
0.06
0.07
0.05
0.05
0.05
0.05
0.04
0.07
0.15
0.04
0.04
0.07
0.04
0.04
0.04
0.05
0.04
0.04
0.04
0.06
0.05
0.05
0.04
0.17
0.06
0.04
0.05
0.05
0.04
0.04
0.04
0.05
0.05
0.04
RO Permeate
(NTU)
0.24
0.05
0.04
0.06
0.04
0.05
0.06
0.04
0.04
0.06
0.05
0.07
0.05
0.04
0.04
0.04
0.04
0.05
0.04
0.06
0.04
0.04
0.04
0.06
0.04
0.04
0.04
0.01
0.04
0.04
0.06
0.06
0.04
0.05
0.05
0.06
NR
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
RO
Concentrate
(NTU)
1.58
0.88
0.09
0.34
1.13
0.16
0.55
0.12
0.77
0.18
0.15
0.08
0.06
0.08
0.05
0.19
0.06
0.10
0.06
0.07
0.31
0.18
0.12
0.05
0.17
0.06
0.05
0.04
1.01
0.83
0.06
0.13
0.42
0.90
0.05
0.56
0.07
0.05
0.07
0.05
0.06
0.05
0.13
0.08
0.05
0.04
UF Turbidity
Reduction (%)
93.8
90.8
94.4
93.5
88.7
96.3
97.5
92.7
97.2
91.9
93.1
83.8
96.2
94.8
95.0
95.1
95.6
94.9
96.2
95.1
96.4
96.9
93.8
96.3
95.0
87.2
96.0
91.5
98.2
97.1
91.5
95.6
93.1
93.2
97.3
85.1
95.5
96.3
96.5
93.1
92.2
96.0
94.3
94.6
95.3
95.2
74
-------
Table 4-8 Turbidity Results, On-Site Bench Top (continued)
Date
11/03/07
11/04/07
11/04/07
11/04/07
11/05/07
11/05/07
11/05/07
11/06/07
11/06/07
11/06/07
11/06/07
11/07/07
11/07/07
11/07/07
11/07/07
11/08/07
11/08/07
11/10/07
11/10/07
11/10/07
11/11/07
11/11/07
11/11/07
11/11/07
11/12/07
11/12/07
Mean:
Median:
Minimum:
Maximum:
Count:
Std. Dev.:
95% CI:
Raw Water UF Filtrate RO Feed RO
(NTU) (NTU) (NTU)
0.85
1.17
1.14
1.17
1.39
1.06
1.15
1.42
1.07
1.39
1.18
1.35
1.05
1.15
1.00
1.14
1.11
1.97
1.89
1.81
2.97
1.79
1.48
1.70
0.99
1.33
1.34
1.18
0.62
3.21
72
0.48
0.11
Table 4-9. In-Line Turbidity
Mean
Median
Minimum
Maximum
Count
Std. Dev.
95% CI
Raw Water
(NTU)
1.38
1.26
0.34
6.76
1835
0.50
±0.02
0.05
0.04
0.04
0.04
0.04
0.05
0.04
0.04
0.05
0.04
0.04
0.08
0.05
0.04
0.05
0.05
0.04
0.07
0.05
0.07
0.04
0.07
0.05
0.04
0.04
0.05
0.06
0.05
0.04
0.21
72
0.04
0.01
Measurement
UF Filtrate
(NTU)
0.019
0.018
0.003
0.712
1807
0.021
±0.001
0.05
0.05
0.05
0.04
0.04
0.06
0.04
0.05
0.05
0.04
0.04
0.06
0.04
0.04
0.04
0.04
0.05
0.02
0.05
0.05
0.04
0.04
0.04
0.04
0.05
0.02
0.05
0.05
0.02
0.17
72
0.02
0.01
Statistics
RO Permeate
(NTU)
0.013
0.012
0.001
0.333
1854
0.012
±0.0005
Permeate
(NTU)
0.04
0.01
0.04
0.03
0.04
0.04
0.03
0.04
0.04
0.03
0.03
0.06
0.04
0.04
0.03
0.04
0.04
0.04
0.05
0.05
0.04
0.04
0.04
0.05
0.04
0.04
0.05
0.04
0.01
0.24
71
0.03
0.01
RO
Concentrate
(NTU)
0.13
0.05
0.04
0.05
0.04
0.06
0.05
0.04
0.05
0.05
0.05
0.32
0.04
0.05
0.04
0.13
0.07
0.06
0.05
0.06
0.05
0.05
0.06
0.04
0.05
0.05
0.19
0.06
0.04
1.58
72
0.30
0.07
UF Turbidity
Reduction (%)
94.1
96.6
96.5
96.6
97.1
95.3
96.5
97.2
95.3
97.1
96.6
94.1
95.2
96.5
95.0
95.6
96.4
96.4
97.4
96.1
98.7
96.1
96.6
97.6
96.0
96.2
94.9
95.6
83.8
98.7
72
2.73
0.63
75
-------
The conductivity of the process streams was measured daily through bench-top analysis. In
addition, an in-line conductivity meter continuously monitored the RO permeate stream, and a
data logger recorded measurements once per hour through the test. Table 4-10 shows the bench-
top conductivity measurement data and summary statistics for the UF and RO systems. Note that
the RO permeate data is in units of microSiemens per centimeter (|iS/cm), while the data for the
rest of the process streams is in units of milliSiemens per centimeter (mS/cm). Figure 4-15
graphically presents the bench-top conductivity measurements for the RO process streams over
the duration of the test. Figure 4-16 shows the in-line meter RO permeate conductivity
measurements captured by the data logger. The in-line meter conductivity data can be found in
Appendix D. The mean conductivity in the RO permeate, as measured by the bench-top
conductivity meter, was 592 |iS/cm. The mean conductivity of the RO feed was 51,380 |iS/cm.
The mean RO permeate conductivity for the hourly data logger measurements was 587 |iS/cm.
The RO unit reduced the conductivity by a mean value of 98.9%. The direct measurement of
TDS, presented in Table 4-13, shows that the TDS concentration in the RO permeate was in the
280 to 300 mg/L range compared to the feed in the 34,000 to 39,000 mg/L range. These numbers
translate to TDS reduction of approximately 99% or greater.
Table 4-10. Conductivity Results, On-Site Benchtop
Date
10/16/07
10/17/07
10/17/07
10/18/07
10/19/07
10/21/07
10/21/07
10/22/07
10/22/07
10/22/07
10/23/07
10/23/07
10/23/07
10/24/07
10/24/07
10/24/07
10/25/07
10/25/07
10/25/07
10/26/07
10/26/07
10/26/07
10/26/07
10/27/07
Raw Water
(mS/cm)
52.91
NR
1.49
NR
52.66
51.74
51.82
52.81
50.41
50.56
53.89
50.84
50.78
53.67
50.83
50.78
51.27
51.08
51.08
51.22
50.99
51.28
51.10
51.12
UF Filtrate
(mS/cm)
53.05
NR
3.07
NR
52.68
51.78
51.88
52.62
50.70
50.72
53.68
50.78
50.68
53.43
50.77
50.66
51.54
51.23
51.16
51.36
51.20
51.28
51.25
51.28
RO Feed
(mS/cm)
53.00
50.80
50.88
51.03
52.69
51.77
51.80
52.60
50.66
50.84
53.33
50.79
50.76
53.49
50.85
50.72
51.46
51.21
51.23
51.38
51.20
51.30
51.31
51.32
RO Permeate
(uS/cm)
642.0
602.0
625.7
571.6
619.9
585.6
565.9
552.8
558.5
549.4
588.3
576.0
583.2
561.5
595.0
599.5
565.6
590.3
579.9
556.3
576.6
619.3
614.9
618.6
RO
Concentrate
(mS/cm)
90.72
34.61
87.90
87.76
89.23
88.28
88.25
89.67
86.37
86.57
90.48
86.65
86.36
90.55
86.64
86.46
87.31
87.79
87.56
87.82
87.69
88.16
87.03
88.03
RO%
Conductivity
Red.
98.8
98.8
98.8
98.9
98.8
98.9
98.9
98.9
98.9
98.9
98.9
98.9
98.9
99.0
98.8
98.8
98.9
98.8
98.9
98.9
98.9
98.8
98.8
98.8
76
-------
Table 4-10. Conductivity Results, On-Site Benchtop (continued)
Date
10/27/07
10/27/07
10/27/07
10/28/07
10/29/07
10/29/07
10/29/07
10/30/07
10/30/07
10/30/07
10/30/07
10/31/07
10/31/07
10/31/07
10/31/07
11/01/07
11/01/07
11/01/07
11/01/07
11/02/07
11/02/07
11/02/07
11/02/07
11/03/07
11/04/07
11/04/07
11/04/07
11/05/07
11/05/07
11/05/07
11/06/07
11/06/07
11/06/07
11/06/07
11/07/07
11/07/07
11/07/07
11/07/07
11/08/07
11/08/07
11/10/07
11/10/07
Raw Water
(mS/cm)
50.92
50.87
50.91
51.16
50.53
51.04
51.20
51.26
50.63
50.80
50.86
51.27
50.37
50.96
50.92
51.22
51.23
51.07
51.37
51.40
50.91
51.06
51.30
51.35
51.24
51.14
51.17
51.40
51.00
51.21
51.37
51.15
51.25
51.21
51.19
51.14
51.17
51.21
51.17
51.05
50.97
51.13
UF Filtrate
(mS/cm)
51.16
51.17
51.22
51.29
50.15
51.18
51.02
51.32
50.94
51.23
51.20
51.29
50.75
51.25
51.23
51.33
50.99
51.37
51.41
51.33
51.17
51.33
51.40
51.24
51.11
51.40
51.41
51.33
51.32
51.39
51.45
51.30
51.37
51.37
51.32
51.36
51.39
51.39
51.28
51.35
51.21
51.31
RO Feed
(mS/cm)
51.15
51.21
51.28
51.31
50.50
51.26
51.27
51.33
51.00
51.28
51.25
51.31
50.66
51.30
51.30
51.36
51.33
51.41
51.27
51.46
51.27
51.41
51.46
51.40
51.32
51.43
51.44
51.45
51.39
51.45
51.47
51.34
51.40
51.40
51.40
51.38
51.41
51.42
51.33
51.37
51.15
51.25
RO Permeate
OiS/cm)
608.3
656.3
612.4
602.1
639.5
619.6
603.1
587.1
597.7
608.9
592.5
570.6
586.1
611.7
610.0
615.9
628.7
641.5
628.6
609.4
630.4
640.9
760.1
603.3
627.7
612.0
604.3
591.2
596.4
601.1
570.1
579.3
578.5
572.2
555.1
563.7
560.6
557.6
544.6
546.3
569.4
554.8
RO
Concentrate
(mS/cm)
87.67
87.71
88.15
87.98
85.41
87.90
87.82
87.86
87.37
87.92
87.79
87.92
87.27
87.89
88.03
87.88
87.75
88.03
88.08
87.79
87.78
88.04
88.34
88.08
87.84
88.15
88.02
88.11
88.08
88.25
88.17
87.85
88.06
88.12
87.96
87.99
87.93
88.03
88.13
87.97
87.55
87.93
RO
Conductivity
Reduction (%)
98.8
98.7
98.8
98.8
98.7
98.8
98.8
98.9
98.8
98.8
98.8
98.9
98.8
98.8
98.8
98.8
98.8
98.8
98.8
98.8
98.8
98.8
98.5
98.8
98.8
98.8
98.8
98.9
98.8
98.8
98.9
98.9
98.9
98.9
98.9
98.9
98.9
98.9
98.9
98.9
98.9
98.9
77
-------
Table 4-10. Conductivity Results, On-Site Benchtop (continued)
Raw Water
Date (mS/cm)
11/10/07
11/11/07
11/11/07
11/11/07
11/11/07
11/12/07
11/12/07
Mean:
Median:
Minimum:
Maximum:
Count:
Std. Dev.:
95% CI:
51.23
51.41
51.23
51.23
51.16
51.54
51.19
50.55
51.17
1.49
53.89
71
5.94
1.38
UF Filtrate
(mS/cm)
51.31
51.45
51.24
51.24
51.18
51.55
51.34
50.68
51.29
3.07
53.68
71
5.76
1.34
RO Feed
(mS/cm)
51.36
51.49
51.19
51.28
51.41
51.49
51.35
51.38
51.33
50.50
53.49
73
0.52
0.12
RO Permeate
OiS/cm)
547.7
534.3
547.6
553.8
545.3
559.1
577.2
592.0
590.3
534.3
760.1
73
35.27
8.09
RO
Concentrate
(mS/cm)
87.82
87.73
87.56
87.63
87.63
87.92
87.96
87.16
87.90
34.61
90.72
73
6.29
1.44
RO
Conductivity
Reduction (%)
98.9
99.0
98.9
98.9
98.9
98.9
98.9
98.9
98.9
98.5
99.0
73
0.07
0.02
0
10/15/07
800
10/20/07
10/25/07
10/30/07
Date
11/04/07
11/09/07
0
11/14/07
Figure 4-15. RO conductivity results.
78
-------
750
500
10/16/07
10/21/07 10/26/07 10/31/07
Date
11/05/07
11/10/07
Figure 4-16. RO permeate conductivity readings from in-line meter.
Tables 4-11 and 4-12 present the pH and temperature data collected from the UF and RO
systems. The UF system had no impact on the pH of the water with the feed having a mean pH
of 7.78 (median 7.79) and the filtrate having a mean pH of 7.73 (median 7.73). RO treatment did
lower the pH of the treated water. The pH of the RO permeate ranged from 6.11 to 6.49 with a
mean of 6.29 (median 6.35). Note that there are only five reported pH values for the RO
permeate. For most of the test, pH measurements were made using a Myron L Ultrameter II
Model 6P, including RO permeate samples. However, this meter always reported the permeate
pH in the 8.0 to 8.5 range, which is highly unlikely for an RO permeate stream due to the loss of
dissolved ions. After a field technician realized that the RO permeate pH measurements were too
high, the FTO began using an Accumet Model 50 meter to measure the RO permeate. This meter
gave pH measurements in the expected range of 6.0 to 6.5. Only the five Accumet meter
measurements are reported in Table 4-11. It is not known why the Ultrameter II 6P meter did not
accurately measure the RO permeate pH. The meter was calibrated correctly every day at three
points using buffers of pH 4, 7, and 10. Also, the Ultrameter's results for the other process
streams agreed with confirmatory measurements made with the Accumet meter.
The UF and RO system had only a slight effect on the temperature of the water as it passed
through the systems. Water temperature in the ocean feed at the beginning of the test was in the
12.8 °C to 16.2 °C range with a mean of 14.8 °C. The mean temperature of the RO permeate was
15.7 °C with a range of 13.3 °C to 17.0 °C. Temperature variation and impact on membrane
79
-------
operating production (flux and specific flux) were accounted for in the operating section by
adjusting the data to either 20 °C or 25 °C, as described in Sections 4.5.1.1.3 and 4.5.1.2.3, the
temperature data in Table 4-12 served as the basis for the temperature adjustment calculations.
Table 4-11. pH Results
Date
10/16/07
10/17/07
10/17/07
10/18/07
10/19/07
10/21/07
10/21/07
10/22/07
10/22/07
10/22/07
10/23/07
10/23/07
10/23/07
10/24/07
10/24/07
10/24/07
10/25/07
10/25/07
10/25/07
10/26/07
10/26/07
10/26/07
10/26/07
10/27/07
10/27/07
10/27/07
10/27/07
10/28/07
10/29/07
10/29/07
10/30/07
10/30/07
10/30/07
10/30/07
10/31/07
10/31/07
10/31/07
10/31/07
11/01/07
11/01/07
Raw Water
7.76
7.63
7.50
7.73
7.87
7.59
7.79
7.67
7.68
7.71
7.86
7.74
7.80
7.73
7.78
7.68
7.79
7.79
7.81
7.77
7.78
7.83
7.82
7.86
7.75
7.79
7.79
7.81
7.80
7.82
7.86
7.73
7.79
7.78
7.79
7.94
7.71
7.37
7.76
7.84
UF Filtrate
7.81
7.66
7.69
7.55
7.83
7.72
7.74
7.64
7.71
7.62
7.79
7.68
7.73
7.76
7.72
7.67
7.72
7.76
7.74
7.71
7.70
7.69
7.76
7.80
7.75
7.75
7.74
7.76
7.72
7.78
7.76
7.71
7.75
7.73
7.73
7.86
7.69
7.64
7.69
7.70
RO Feed at
Strainer
7.40
7.47
7.72
7.59
7.75
7.65
7.61
7.66
7.60
7.64
7.80
7.73
7.71
7.71
7.71
7.69
7.73
7.72
7.76
7.70
7.73
7.73
7.76
7.81
7.75
7.75
7.73
7.75
7.73
7.78
7.75
7.70
7.74
7.73
7.70
7.80
7.68
7.65
7.67
7.76
RO Permeate
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
6.35
NM
6.11
NM
NM
NM
NM
NM
NM
NM
6.49
6.14
RO Concentrate
7.42
7.25
7.44
7.25
7.44
7.37
7.35
7.36
7.41
7.32
7.57
7.36
7.40
7.41
7.36
7.39
7.48
7.43
7.42
7.47
7.43
7.41
7.41
7.54
7.46
7.43
7.47
7.47
7.43
7.46
7.45
7.43
7.43
7.49
7.45
7.52
7.36
7.34
7.39
7.45
80
-------
Table 4-11. pH Results (continued)
Date
11/01/07
11/01/07
11/02/07
11/02/07
11/02/07
11/02/07
11/03/07
11/04/07
11/04/07
11/04/07
11/05/07
11/05/07
11/05/07
11/06/07
11/06/07
11/06/07
11/06/07
11/07/07
11/07/07
11/07/07
11/07/07
11/08/07
11/08/07
11/10/07
11/10/07
11/10/07
11/11/07
11/11/07
11/11/07
11/11/07
11/12/07
11/12/07
Mean:
Median:
Minimum:
Maximum:
Count:
Std. Dev.:
95% CI:
Raw Water
7.85
7.81
7.87
7.77
7.76
7.79
7.79
7.77
7.68
7.74
7.89
7.79
7.94
7.93
7.77
7.70
7.84
7.86
7.73
7.84
7.73
7.85
7.73
7.65
7.81
7.78
7.78
7.71
7.79
7.79
7.91
7.85
7.78
7.79
7.37
7.94
72
0.09
0.02
UF Filtrate
7.80
7.75
7.76
7.73
7.75
7.74
7.67
7.61
7.68
7.71
7.81
7.73
7.85
7.84
7.69
7.60
7.72
7.87
7.66
7.69
7.64
7.74
7.63
7.70
7.74
7.69
7.72
7.78
7.67
7.60
7.82
7.87
7.73
7.73
7.55
7.87
72
0.07
0.02
RO Feed at
Strainer
7.78
7.72
7.76
7.73
7.74
7.74
7.67
7.69
7.67
7.69
7.77
7.72
7.82
7.80
7.72
7.58
7.68
7.83
7.65
7.66
7.62
7.70
7.61
7.73
7.69
7.66
7.69
7.80
7.63
7.69
7.82
7.86
7.71
7.72
7.40
7.86
72
0.08
0.02
RO Permeate
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
6.35
NM
NM
NM
NM
NM
6.29
6.35
6.11
6.49
5
NM
NM
RO Concentrate
7.45
7.43
7.49
7.44
7.44
7.43
7.40
7.38
7.34
7.38
7.46
7.43
7.53
7.49
7.42
7.29
7.38
7.54
7.39
7.39
7.34
7.40
7.34
7.39
7.40
7.34
7.40
7.41
7.32
7.38
7.55
7.57
7.42
7.42
7.25
7.57
72
0.07
0.02
81
-------
Table 4-12. Temperature Results
Date
10/16/07
10/17/07
10/17/07
10/18/07
10/19/07
10/21/07
10/21/07
10/22/07
10/22/07
10/22/07
10/23/07
10/23/07
10/23/07
10/24/07
10/24/07
10/24/07
10/25/07
10/25/07
10/25/07
10/26/07
10/26/07
10/26/07
10/26/07
10/27/07
10/27/07
10/27/07
10/27/07
10/28/07
10/29/07
10/29/07
10/29/07
10/30/07
10/30/07
10/30/07
10/30/07
10/31/07
10/31/07
10/31/07
10/31/07
11/01/07
11/01/07
11/01/07
Raw Water
(°C)
14.4
13.9
14.2
13.1
14.1
13.1
13.0
12.8
13.3
13.1
14.4
14.0
13.9
13.8
14.3
14.3
14.0
14.7
14.4
13.8
14.7
15.4
14.9
15.5
15.6
16.0
15.5
15.3
15.2
15.1
15.0
14.9
15.2
15.5
15.1
14.6
15.5
15.4
15.4
15.4
15.8
16.2
UF Filtrate
(°C)
15.5
14.1
14.0
13.4
14.1
13.3
13.2
13.0
13.4
13.4
14.6
14.4
14.2
13.8
14.8
14.6
14.1
14.7
14.6
13.9
14.8
15.4
15.0
15.6
15.7
15.9
15.5
15.4
16.0
14.9
15.0
14.8
15.3
15.3
15.2
14.7
15.7
15.5
15.3
15.5
16.1
16.3
RO Feed
(°C)
16.4
14.7
14.0
13.3
14.3
13.5
13.5
13.1
13.7
13.5
14.8
14.3
14.3
13.9
14.8
14.7
14.2
14.7
14.6
14.1
14.9
15.4
15.1
15.7
15.7
16.2
15.6
15.4
15.8
15.2
15.1
15.0
15.6
15.4
15.3
14.9
16.2
15.7
15.3
15.7
16.1
16.3
RO Permeate
(°C)
16.0
15.2
14.7
14.1
15.3
14.2
14.1
13.7
14.2
14.1
15.5
14.9
14.9
14.6
15.4
13.3
14.9
15.3
15.2
14.8
15.6
15.9
15.7
16.4
15.5
16.7
16.2
16.3
16.4
15.8
15.8
15.6
16.2
16.1
15.9
15.5
16.6
16.3
15.3
16.3
16.8
17.0
RO Concentrate
(°C)
16.9
16.1
15.8
15.0
16.1
15.2
15.1
14.7
15.3
15.0
16.5
15.8
15.9
15.7
16.3
16.2
15.9
16.3
16.2
15.7
16.6
16.8
16.6
17.3
17.3
17.7
17.1
17.0
17.4
16.8
16.6
16.6
17.2
17.0
16.8
16.4
17.6
17.2
16.9
17.3
17.7
17.9
82
-------
Table 4-12. Temperature Results (continued)
Date
11/01/07
11/02/07
11/02/07
11/02/07
11/02/07
11/03/07
11/04/07
11/04/07
11/04/07
11/05/07
11/05/07
11/05/07
11/06/07
11/06/07
11/06/07
11/06/07
11/07/07
11/07/07
11/07/07
11/07/07
11/08/07
11/08/07
11/10/07
11/10/07
11/10/07
11/11/07
11/11/07
11/11/07
11/11/07
11/12/07
11/12/07
Mean:
Median:
Minimum:
Maximum:
Count:
Std. Dev.:
95% CI:
Raw Water
(°C)
16.1
15.9
16.1
16.2
16.0
15.9
15.8
15.8
15.8
15.5
15.8
15.7
15.1
15.5
15.0
15.1
14.8
15.0
14.6
14.7
14.5
14.6
14.4
14.2
14.0
13.6
14.3
14.1
14.0
14.5
14.8
14.8
14.9
12.8
16.2
73
0.86
0.20
UF Filtrate
(°C)
16.2
16.1
16.1
16.1
16.0
16.1
15.9
15.8
15.8
15.7
15.9
15.7
15.1
15.6
15.1
15.1
14.9
15.1
14.6
14.7
14.5
14.6
14.7
14.2
14.1
13.7
14.6
14.1
14.0
14.5
15.2
14.9
15.0
13.0
16.3
73
0.84
0.19
RO Feed
(°C)
16.4
16.1
16.3
16.2
16.2
16.1
16.0
15.9
15.8
15.7
16.0
15.8
15.3
15.8
15.2
15.2
15.0
15.2
14.7
14.8
14.7
14.8
15.0
14.3
14.1
13.9
14.6
14.2
14.2
14.6
15.4
15.1
15.1
13.1
16.4
73
0.84
0.19
RO Permeate
(°C)
17.0
16.8
17.0
16.9
16.7
16.7
16.7
16.6
16.5
16.4
16.6
16.4
16.0
16.4
15.8
15.8
15.6
15.9
15.3
15.5
15.3
15.5
15.7
15.0
14.9
14.6
15.2
14.8
14.8
15.4
16.6
15.7
15.7
13.3
17.0
73
0.87
0.20
RO Concentrate
(°C)
17.9
17.7
18.0
17.8
17.6
17.7
17.6
17.5
17.4
17.3
17.6
17.4
16.9
17.4
16.8
16.8
16.5
16.9
16.3
16.4
16.3
16.4
16.5
15.9
15.8
15.5
16.3
15.8
15.8
16.2
17.0
16.6
16.6
14.7
18.0
73
0.80
0.18
83
-------
4.5.3.2 Water Quality Results - Other Water Quality Parameters
Table 4-13 presents the other water quality data collected on a weekly basis during the
verification test. For an unknown reason, the suspended solids levels in the UF filtrate were
higher than expected. The TSS reduction through the UF skid was only 1-5 mg/L, with one set of
measurements when the TSS was actually higher in the filtrate compared to the feed. These data
are in conflict with the daily turbidity results, which show 95% reduction in turbidity (Table 4-9)
and a low turbidity in the UF filtrate (mean of 0.5 NTU). The UF system was definitely retaining
suspended solids, as evidenced by the TMP increases and four UF cleanings that were required
during the test run. Also, the steady operation of the RO system indicates that any suspended
solids in the UF filtrate (the RO feed) did not impact RO operation.
The RO system reduced the TSS in the RO feed to less than detectible levels (<2.0 mg/L) in the
RO permeate. While this good (and expected) from a final water quality perspective, if
suspended solids are getting to the RO membranes, they can build up and eventually could cause
a decrease in specific flux, and a membrane plugging issue. Membrane plugging did not occur
during this test, as shown by the minimal change in RO specific flux.
The UF system did not impact the other water quality parameters, as would be expected. These
other parameters, such as hardness, alkalinity, TDS, etc., primarily represent dissolved inorganic
constituents that are not removed by UF.
The RO system did remove many of the dissolved inorganic species, as shown by the results in
Table 4-13 for the RO permeate. Total dissolved solids were reduced by 99%, as was chloride.
Sodium was reduced by 98%. These data are consistent with the conductivity data presented
earlier, which shows a salt rejection/reduction through the RO of 98.9%. The other inorganic
materials measured, such as hardness, alkalinity, metals, sulfate, and phosphorus were also
reduced in the RO permeate. The RO concentrate increased in concentration for these parameters
above the feed levels, as would be expected. The RO membranes, at these operating conditions,
rejected the dissolved salts present in the feed throughout the test.
The TQAP called for calculating mass balances of sodium, calcium, magnesium, sulfate,
carbonate, and chloride ions, and also total dissolved solids, to determine if significant scale
formation occurred in the RO system. However, as described in Section 4.5.1.2.2, the Stiff and
Davis Stability Index was calculated, and the RO feed was found to be non-scaling. Therefore,
the mass balance exercise was not performed for this report.
84
-------
Table 4-13. Other UF System Water Quality Data
TSS (mg/L)
Date
10/17/07
10/24/07
10/30/07
11/05/07
11/12/07
ND-not
Raw
Water
7
5
9
6
8
detected (detection
UF UF
Filtrate Retentate RO Feed
5 10
4 7
7 10
7 11
3 9
limit)
5
4
5
7
4
RO
RO
Permeate Concentrate
ND(2)
ND(2)
ND(2)
ND(2)
ND(2)
8
10
7
9
9
TDS (mg/L)
Date
10/16/07
10/17/07
10/18/07
10/22/07
10/23/07
10/24/07
10/25/07
10/29/07
10/30/07
10/31/07
11/01/07
11/05/07
11/06/07
11/07/07
11/08/07
11/12/07
(1) Note
collected
Raw
Water
35000
NM
35000
35000
35000
33000
38000
35000
34000
34000
34000
33000
34000
34000
34000
NM
UF
Retentate
NM(1)
34000
NM
NM
NM
36000
NM
NM
34000
NM
NM
33000
NM
NM
NM
34000
that as listed in Table 3-4, samples of
for TDS analysis
on most days of the
UF Filtrate/
RO Feed
NM
39000
NM
NM
NM
33000
NM
NM
33000
NM
NM
34000
NM
NM
NM
34000
RO
RO
Permeate Concentrate
340
300
290
260
280
280
280
330
300
280
300
290
280
270
270
NM
the UF feed, RO permeate, and RO
test, for the purpose
of establishing
67000
67000
67000
67000
67000
67000
66000
66000
64000
65000
64000
65000
65000
65000
66000
NM
concentrate streams were
the conductivity to TDS
correlation discussed in Section 4.5. 1.2.3.
NM-not
measured
Hardness (mg/L as CaCO3)
Date
10/17/07
10/24/07
10/30/07
11/05/07
11/12/07
UF Filtrate/
Raw Water RO Feed
6200
6500
5400
5800
5500
7200
6500
6000
5600
5500
RO Permeate
4
o
J
4
o
5
o
5
RO
Concentrate
12000
12000
11000
11000
11000
Date
10/17/07
10/24/07
10/30/07
11/05/07
11/12/07
Raw Water
110
110
110
110
110
Alkalinity (mg/L
UF Filtrate/
RO Feed
110
110
110
110
120
as CaCO3)
RO Permeate
ND(5)
ND(5)
ND(5)
ND(5)
ND(5)
RO
Concentrate
200
200
200
220
220
85
-------
Table 4-13. Other UF System Water Quality Data (continued)
Date
10/17/07
10/24/07
10/30/07
11/05/07
11/12/07
Raw Water
27000
20000
20000
20000
20000
Chloride (mg/L)
UF Filtrate/
RO Feed
21000
26000
21000
21000
20000
RO Permeate
180
180
170
180
170
RO
Concentrate
44000
41000
42000
39000
42000
Date
10/17/07
10/24/07
10/30/07
11/05/07
11/12/07
Raw Water
2600
2800
2700
2600
2700
Sulfate (mg/L)
UF Filtrate/
RO Feed
2800
2800
2800
2700
2700
RO Permeate
1.7
1.6
1.9
1.6
1.4
RO
Concentrate
5100
5100
5000
5100
5000
Date
10/17/07
10/24/07
10/30/07
11/05/07
11/12/07
Raw Water
380
410
340
360
360
Calcium (mg/L)
UF Filtrate/
RO Feed
440
410
380
370
350
RO Permeate
0.24
0.21
0.26
0.21
0.19
RO
Concentrate
750
740
730
680
690
Date
10/17/07
10/24/07
10/30/07
11/05/07
11/12/07
Raw Water
1300
1300
1100
1200
1100
Magnesium (mg/L)
UF Filtrate/
RO Feed
1500
1300
1200
1100
1100
RO Permeate
0.72
0.58
0.88
0.71
0.63
RO
Concentrate
2500
2400
2300
2200
2200
Date
10/17/07
10/24/07
10/30/07
11/05/07
11/12/07
Raw Water
9300
11000
10000
10000
10000
Sodium (mg/L)
UF Filtrate/
RO Feed
9900
11000
10000
11000
10000
RO Permeate
110
100
110
120
110
RO
Concentrate
19000
20000
19000
19000
19000
Date
10/17/07
10/24/07
10/30/07
11/05/07
11/12/07
Raw Water
380
450
290
430
370
Potassium (mg/L)
UF Filtrate/
RO Feed
400
430
340
290
400
RO Permeate
3.9
4.0
4.1
4.0
11.0
RO
Concentrate
780
800
710
610
910
86
-------
Table 4-13. Other UF System Water Quality Data (continued)
Date
10/17/07
10/24/07
10/30/07
11/05/07
11/12/07
Raw Water
0.002
0.011
0.160
0.180
0.190
Lithium (mg/L)
UF Filtrate/
RO Feed
0.110
0.041
0.150
0.180
0.170
RO Permeate
0.009
0.009
0.002
0.003
0.003
RO
Concentrate
0.036
0.140
0.300
0.370
0.290
Date
10/17/07
10/24/07
10/30/07
11/05/07
11/12/07
Raw Water
6.2
6.6
4.2
4.6
4.9
Boron (mg/L)
UF Filtrate/
RO Feed
6.0
6.3
4.9
4.5
5.1
RO Permeate
1.0
1.1
1.2
1.1
1.5
RO
Concentrate
9.4
9.2
7.9
7.6
8.4
Date
10/17/07
10/24/07
10/30/07
11/05/07
11/12/07
Raw Water
ND (0.05)
0.05
ND(l.O)
ND(l.O)
ND(l.O)
Total Phosphorus (mg/L)
UF Filtrate/
RO Feed RO Permeate
ND (0.05)
0.38
ND (1.0)
ND (1.0)
ND (1.0)
ND (0.05)
ND (0.05)
ND(l.O)
ND(l.O)
ND(l.O)
RO
Concentrate
0.16
0.46
ND(l.O)
ND(l.O)
ND(l.O)
Date
10/17/07
10/24/07
10/30/07
11/05/07
11/12/07
Raw Water
0.0086
0.0478
0.0115
0.0166
0.0122
UV2s4 Absorbance
UF Filtrate/
RO Feed
0.0029
0.0120
0.0076
0.0086
ND(0)
RO Permeate
ND(0)
ND(0)
ND(0)
0.012
0.015
RO
Concentrate
0.0131
0.0073
0.0057
0.0333
0.0092
Date
10/17/07
10/24/07
Raw Water
ND (0.02)
0.02
(1) Note that this parameter was
than detectible levels.
Date
10/17/07
10/24/07
Raw Water
0.003
0.002
Ortho-Phosphate (mg/L)(1)
UF Filtrate/
RO Feed RO Permeate
ND (0.02)
ND (0.02)
ND (0.02)
ND (0.02)
dropped from the sampling plan after the
Barium (mg/L)(1)
UF Filtrate/
RO Feed
0.003
0.003
RO Permeate
ND (0.001)
ND (0.001)
RO
Concentrate
ND (0.02)
0.03
first two weeks of the test, due to the less
RO
Concentrate
0.004
0.004
(1) Note that this parameter was dropped from the sampling plan after the first two weeks of the test, due to low and
less than detectible levels.
87
-------
Table 4-13. Other UF System Water Quality Data (continued)
Selenium (mg/L)(1)
UF Filtrate/ RO
Date Raw Water RO Feed RO Permeate Concentrate
10/17/07 ND (0.020) ND (0.020) ND (0.020) ND (0.020)
10/24/07 ND (0.020) ND (0.020) ND (0.020) ND (0.020)
(1) Note that this parameter was dropped from the sampling plan after the first two weeks of the test, due to the less
than detectible levels.
Stiff and Davis Stability Index
Date
10/17/07
10/24/07
10/30/07
11/05/07
11/12/07
RO
Concentrate
PH
7.60
7.66
7.62
7.66
7.70
RO
Concentrate
PH
8.43
8.39
8.46
8.37
8.44
S&DSI
Calculation'1'
-0.83
-0.73
-0.84
-0.71
-0.74
S&DSI
Nomagraph<2)
-0.14
-0.09
-0.13
0.08
0.04
(1) Calculations in column 3 use equations from ASTM D4582
(2) S&DSI based on interpolation of nomagraphs in ASTM D4582.
4.5.3.3 Total Organic Carbon Results for Cleaning Solution
Samples of the cleaning solutions from the UF system CIP were collected from two cleaning
periods. These samples were analyzed for TOC to provide basic water quality information as
required in the ETV test protocol. The TOC results for the UF system cleaning solution are
presented in Table 4-14.
Samples of the RO cleaning solutions were also collected for TOC analyses. These results are
also shown in Table 4-14. Note that the RO cleaning solutions had higher TOC levels than the
UF cleaning solutions. This was most likely caused by the additives in the commercial RO
cleaning product that was used at the site.
Table 4-14. Cleaning Solution TOC Results
Cleaning Solution
Low pH UF solution
High pH UF solution
Low pH UF solution
High pHUF Solution
Low pH RO solution
High pH RO solution
Date
11/03/07
11/04/07
11/09/07
11/10/07
11/13/07
11/14/07
TOC (mg/L)
360
260
600
140
2100
770
4.5.4 Task C4: Membrane Module Integrity
The objective of this task was to demonstrate the methodology for integrity testing of the UF and
RO membranes and also to document system integrity. Pressure decay tests, microorganism
removal, and particulate reduction were all used to document UF membrane integrity. Bacillus
endospores and total coliforms were monitored to provide data on the microbial reduction
88
-------
achieved by the UF and RO membranes. In-line analyzers also collected particle count data, as
an additional measurement/indicator of membrane integrity and the capability of the system to
remove paniculate and microbial contaminants.
As discussed in Section 4.4, the initial UF pressure decay test on September 27, 2007 showed
that pressure was being lost from the system at a higher than desirable rate. The problem was
investigated, and one UF cartridge was found to have a broken fiber, which was repaired. On
October 9, each cartridge was tested individually. One additional cartridge was found to have a
broken fiber and it was also repaired. As shown in Table 4-3, the final UF integrity test on
October 11 before the verification test started showed an acceptable pressure decay rate.
Subsequently, the UF system was tested on a frequent basis during the verification test. The
results of those tests are presented in this section. The Bacillus endospore, total coliform, and
particle count data are also presented.
The RO system was not dye tested during this verification test. The continuous conductivity
measurements and microbial data were used as the indicator that the RO membranes were
operating properly.
4.5.4.1 UF System Pressure Decay Results
Pressure decay tests on the UF system were performed on most operating days during the
verification test. Table 4-15 presents the pressure decay data from the verification test. Data was
actually collected every minute during the pressure decay tests, but has been summarized into 2-
minute increments for ease of presentation. Figure 4-16 shows the pressure decay results on a
minute-by-minute basis in graphical format.
As shown in Table 4-15, the mean pressure decay rate on a daily basis ranged from 0.02 to 0.15
psig/min. The overall mean pressure decay rate was 0.08 psig/min. After the initial membrane
breaks found before the start of the test, there was no indication of any further problems with
membrane integrity, based on these pressure decay rate results.
Most of the pressure decay rates measured during this verification test were lower than those
measured during ETV laboratory tests on two UF cartridges from the sister EUWP unit to the
one tested for this verification (see ETV report Removal of Microbial Contaminants in Drinking
Water, Koch Membrane Systems, Inc. Tar go* 10-48-35-PMC™ Ultrafiltration Membrane, as
Used in the Village Marine Tec. Expeditionary Unit Water Purifier). Each of the two cartridges
underwent four pressure decay tests over the course of lab testing activities. Cartridge 1 had
decay rates of 0.35, 0.74, 0.6 and 0.4 psig/min, while Cartridge 2 had decay rates of 0.09, 0.1,
0.25, and 0.2 psig/min. These two cartridges had pressure decay rates ranging from 0.09 to 0.74
psig/min over four separate pressure decay tests per cartridge. These two cartridges were
challenged with Cryptosporidiumparvum oocysts, and removed greater than 4 logic.
89
-------
11
10 12 14
Time (min)
-a— 10/16/07
A 10/17/07
X 10/18/07
-0— 10/22/07
--0-- 10/23/07
-i—10/24/07
-A— 10/25/07
-•—10/26/07
—X— 10/27/07
-n— 10/29/07
-A— 10/30/07
-*—10/31/07
—X—11/01/07
11/02/07
-+-- 11/05/07
11/06/07
-*— 11/07/07
-o— 11/08/07
•-Q-- 11/11/07
16 18 20
Figure 4-16. Pressure decay over time.
90
-------
Table 4-15. Pressure Decay Data
Date
10/16/07
10/17/07
10/18/07
10/22/07
10/23/07
10/24/07
10/25/07
10/26/07
10/27/07
10/29/07
10/30/07
10/31/07
11/01/07
11/02/07
11/05/07
11/06/07
11/07/07
11/08/07
11/11/07
OMin
15.96
15.88
16.03
12.37
16.02
16.03
17.01
17.02
16.98
15.94
16.08
16.99
16.99
16.97
16.99
15.97
15.89
16.06
16.95
2Min
15.10
15.14
15.24
12.00
15.36
15.43
16.34
17.02
16.40
15.85
15.37
16.63
16.24
16.55
16.83
15.82
15.71
15.90
16.33
4Min
14.49
14.60
14.64
11.74
14.86
14.94
15.81
16.97
15.97
15.76
14.82
16.52
15.67
16.32
16.73
15.76
15.63
15.81
15.86
Pressure Readings (psig)
6 Min 8 Min 10 Min
14.05
14.21
14.18
11.58
14.50
NM
15.43
16.92
15.72
15.67
14.47
16.43
15.26
16.22
16.68
15.72
15.57
15.75
15.50
13.82
13.94
13.83
11.50
14.27
14.29
15.17
16.88
15.60
15.59
14.23
16.34
15.01
16.16
16.64
15.68
15.52
15.68
15.22
13.68
13.80
13.58
11.47
14.16
14.14
15.05
16.84
15.55
15.51
14.08
16.25
14.87
16.11
16.61
15.64
15.48
15.63
15.01
12 Min
13.59
13.73
13.44
11.45
14.11
14.03
15.00
16.82
15.52
15.45
13.99
16.15
14.80
16.07
16.57
15.61
15.43
15.58
14.82
14 Min
13.52
13.69
13.38
11.43
14.08
13.99
14.98
16.78
15.49
15.37
13.91
16.05
14.76
16.03
16.54
15.58
15.39
15.53
14.65
16 Min
13.46
13.62
13.35
11.42
14.05
13.95
14.95
16.74
15.46
15.30
13.83
15.96
14.70
16.00
16.51
15.55
15.34
15.48
NM
18 Min
13.40
13.54
13.33
11.41
14.04
13.92
14.92
16.71
15.42
15.23
13.75
15.86
14.65
15.96
16.49
15.52
15.31
15.43
14.34
Mean
Decay Rate
20 Min (psig/min)
13.35
13.47
13.31
11.40
14.02
13.90
14.90
16.68
15.40
15.16
13.67
15.75
14.62
15.93
16.46
15.49
15.26
15.38
14.20
Mean:
Median:
Minimum
Maximum
0.13
0.12
0.14
0.05
0.10
0.11
0.11
0.02
0.08
0.04
0.12
0.06
0.12
0.05
0.03
0.02
0.03
0.03
0.15
0.08
0.08
0.02
0.15
NM = not measured.
91
-------
4.5.4.2 Bacillus Endospores and Total Coliform Results
The Bacillus endospores data are shown in Table 4-16. The UF system had a mean log reduction
of 1.68 logic, with a range of 0.68 to 1.92 logic. The cumulative mean log reduction after RO
treatment was 1.73 logic, with a range of 0.73 to 1.98 logic. The UF system removed Bacillus
endospores to 1 CFU/lOOmL or <1 CFU/lOOmL on all but two days, October 22 and 23.
Similarly, the RO permeate only had one day, October 23, with Bacillus endospores above 1
CFU/lOOmL. It was noted in the logbook that on October 22 there was windy conditions at the
test site with smoke and ash in the air due to nearby wild fires. With the presence of high winds
and the smoke and ash, it was possible that the samples were contaminated when they were
collected. Similar conditions were reported on October 23 as well.
The concentration of Bacillus endospores present in the feed was low, with a geometric mean 64
CFU/lOOmL and a range of 33 to 96 CFU/lOOmL. Thus, with a detection limit of 1 CFU/lOOmL,
the maximum log reduction that could be demonstrated was 1.5 to 2.0 logic.
Table 4-16. Bacillus Endospore Counts and Log Reduction Calculations
Bacillus Endospores (CFU/lOOmL)
Sample Date
10/16/2007
10/17/2007
10/18/2007
10/22/2007
10/23/2007(1)
10/24/2007(1)
10/25/2007
10/29/2007
10/30/2007
10/31/2007
11/01/2007
11/05/2007
11/06/2007
11/07/2007
11/08/2007
Geometric Mean
Median
Maximum
Minimum
Raw
Water
56
60
66
96
65
84
81
66
61
62
33
58
54
73
65
64
65
96
33
UF
Filtrate
1
<1
1
20
4
<1
1
1
<1
<1
1
1
<1
<1
<1
1.3(2)
2.5
20
<1
UFLog
Reduction
1.8
1.8
1.8
0.7
1.2
1.9
1.9
1.8
1.8
1.8
1.5
1.8
1.7
1.9
1.8
1.6
1.7
1.9
0.7
UF RO
Retentate Permeate
50 <1
59 <1
56 <1
97 1
77 12
119 <1
79 <1
52 <1
45 <1
49 <1
35 <1
67 <1
50 <1
91 1
48 <1
61 1.2(2)
65 1.7
119 12
35 <1
UF + RO
Log
Reduction
1.8
1.8
1.8
2.0
0.7
1.9
1.9
1.8
1.8
1.8
1.5
1.8
1.7
1.9
1.8
1.7
1.7
2.0
0.7
RO
Concentrate
46
51
4
18
15
35
5
8
6
5
8
6
3
5
14
10
15
51
3
(1) Sample holding time exceeded, see Section 4.7.4 for further discussion
(2) Values below detection limits (<1) set equal to 1 for geometric mean calculation.
92
-------
The total coliform data collected during the verification test are shown in Table 4-17. The UF
system reduced the total coliform concentration to <1 CFU/lOOmL for all days tested. The feed
was low in total coliform count, ranging from 5 to 16 CFU/lOOmL. Therefore, the range of logio
reduction that could be demonstrated was only 0.7 to 1.2 logic.
Table 4-17. Total Coliform Counts and Log Reduction Calculations
Total Coliforms (CFU/lOOmL)
Date
10/18/2007
10/22/2007
10/23/2007
10/24/2007
Raw
Water
12
6
5
16
UF UF Log
Filtrate Reduction
<1 1.1
<1 0.8
<1 0.7
<1 1.2
UF RO
Discharge Permeate
4 <1
16 <1
2 <1
13 <1
UF + RO Log
Reduction
1.1
0.8
0.7
1.2
RO
Concentrate
<1
<1
<1
<1
4.5.4.3 UF System Particle Count Data
The in-line particle counters measured the particle counts in the raw water and UF filtrate every
five minutes, and stored the data for transfer to a personal computer. Particle count data can be
helpful in evaluating the integrity and performance of membrane systems and in predicting the
reduction of microbial contaminants.
The particle count data was condensed from five-minute increments to one-hour averages for
graphical presentation. The data were separated to provide information on various size ranges
(e.g. 2-3 |im, 3-5 jim), as these sizes correspond to the sizes of various microbial contaminants of
interest in drinking water, such as Cryptosporidium (3 to 5 jim)
Figure 4-17 shows the hourly averages for the raw water and UF filtrate 2-3 jim particle counts.
Some notes about this figure and the particle count data presented:
• The y-axis is in logarithmic scale.
• There is no particle count data for the first two days, and last three days of testing. The
particle count data supplied by the field operators begins at 8:30 a.m. on August 1. The
data ends at 3:35 p.m. on August 21 because the computer logging the data crashed.
• The gaps in the data are the periods when the UF system was shut down for membrane
cleanings.
• There were numerous single time point spikes in the particle counts that increased some
of the hourly averages. These spikes were likely due to the automatic backwashes
executed every half hour.
93
-------
100000
10000
e
o
1000
.8 100
o Raw Water a UF Filtrate
10/16/07 10/21/07 10/26/07 10/31/07 11/05/07 11/10/07 11/15/07
Date
Figure 4-17. Particle count hourly averages - 2-3 um.
The mean 2-3 jim particle count for the raw water was 5,559/mL with a median value of
5,533/mL. The range of particle counts for the raw water was from 53/mL to 17,843/mL. The
filtrate had a mean 2-3 |im particle count of 42/mL with a median of 25/mL and a range of 0 to
773/mL. Note that these statistics are based on individual counts, not the hourly averages
presented in the graphs. Both the mean and median 2-3 jim particle log reduction was 2.3 logic.
As evidenced by the difference between the mean and median particle counts, the particle count
distribution is skewed toward the low side of the mean, as shown in the filtrate particle count
distribution bar graph in Figure 4-18. Of the 3,464 individual particle counts used for this
analysis, 2,270 were 40/mL or less. The mean was skewed upward by a few high counts that may
have been measured during a backwash cycle.
94
-------
900
800
&
J?
2-3 Micron Particle Counts (count/mL)
Figure 4-18. UF filtrate 2-3 um particle count size distribution.
Figure 4-19 shows the hourly averages for the raw water and UF filtrate 3-5 jim particle counts.
The notes about Figure 4-16 also apply to this graph. The mean particle count for the raw water
was 3,662/mL with a median value of 3,551/mL. The range of particle counts for the feed was
from 34 to 14,750/mL. The filtrate had a mean 3-5 |im particle count of 22/mL with a median of
10/mL and a range of 0 to 620/mL. The 3-5 jim particle counts were also skewed to the low end
of the range (data not shown). As with the 2-3 |im particle counts, both the mean and median 3-5
|im particle counts were 2.5 logic.
95
-------
10000
1000
=
o
o
U
—
™
r
cS
o.
100
10
1
o Raw Water ° UF Filtrate
D
B
B
D
IP
D
n
a a a
D D D
a cP Qg
"8* o, a°
S, I
B
° ""
° a ° »SS
D D
D
B
ffi
n
10/15/07 10/20/07 10/25/07 10/30/07 11/04/07 11/09/07 11/14/07
Date
Figure 4-19. Particle count hourly averages - 3-5 um.
As can be seen, the UF system was effective in reducing the particle count in these size ranges.
The reduction of particulate matter in the smaller size range support the pressure decay tests in
showing that the UF system maintained integrity throughout the test. Further, a 2.3 to 2.5 logic
reduction would tend to predict a similar or larger reduction in equal and larger size microbial
contaminants. Combined with the pressure decay tests, these results would tend to support that
the UF system should give at least 2-3 logic reduction, if not better control of these contaminants.
Unfortunately, due to the low level of Bacillus endospores in the feed, the direct measurement of
Bacillus endospores could not confirm the results of these indicator tests of UF system
performance for microbial contaminants. These data do confirm, in conjunction with the
turbidity data, that the UF system maintained good system integrity throughout the ETV test.
4.6 Chemical Consumption
Ferric chloride was fed to the UF feed at a rate of 4.37 mL/min, or approximately 0.07 gal per
operating hour. The ferric chloride solution contained 13% iron (Fe) by weight. This yielded an
approximate dose rate of 0.75 mg/L as Fe in the feed to the UF system. Near the end of the test,
the feed pump rate was increased to 5.83 mL/min, increasing the iron dose rate to 1.0 mg/L as
Fe. The higher dose rate was only run for five operating days.
96
-------
The ferric chloride feed rate was checked on most operating days by direct measurement of the
pumping rate. The quantity of ferric chloride used was also recorded each time ferric chloride
was added to the feed tank. These measurements provided two checks on coagulant use during
the test. Based on the feed tank records, a total of 22.4 gal of ferric chloride solution was used
over the duration of the test. A total of 5,259,625 gal of feed was treated with coagulant, so the
dose rate over the entire test was 4.3 X 10~3 gal ferric chloride per 1000 gal of water treated or
0.77 mg/L as Fe.
The RO system is designed to allow addition of a scale inhibitor, if needed. For this test, the
scale inhibitor ONDEO (Nalco) PermaTreat® PC-191 was fed at target dose rate of 5 mg/L. The
antiscalant was fed as a full strength solution (1.16 specific gravity). The target pump rate, based
upon a RO feed flow rate of 174 gpm, was 2.84 mL/min (0.045 gal per hour). The average pump
rate based on daily calibration records showed an average antiscalant feed rate of 3.22 mL/min,
which yields an antiscalant dose rate of 5.7 mg/L. The quantity of antiscalant used was also
recorded each time product was added to the feed tank. Based on the feed tank records, a total of
23.5 gal of antiscalant was used over the duration of the test. Atotal of 4,673,300 gal ofROfeed
was treated with antiscalant, so the dose rate over the entire test was 5.0 X 10"3 gal antiscalant
per 1000 gal of water treated, or 5.8 mg/L.
The chemicals needed for the UF CIP were citric acid, sodium hydroxide (0.5%), and sodium
hypochlorite (12.5% bleach). Citric acid was used to lower the pH of the cleaning solution for
the low pH cleaning cycle, and sodium hydroxide was used for the high pH cleaning cycle.
Section 4.5.2.1 and Table 4-7 described and showed the details on the quantities of chemicals
used for each UF cleaning. Citric acid use ranged from 4 to 11.2 pounds per cleaning cycle;
bleach use ranged from 12 to 29 L per cleaning cycle; and sodium hydroxide use ranged from 2.0
to 3.8 L per cycle.
The RO cleaning was performed using Avista RO Cleaner P303 for the low pH cleaning, and
Avista RO Cleaner Pill for the high pH cleaning. Fifty four (54) pounds of each RO cleaner
were used for the cleaning performed at the end of the test. It should be noted again here that the
RO cleaning was performed at the end of the test, as it is a requirement of the ETV protocol to
demonstrate the cleaning process. However, the RO unit did not actually require cleaning at that
time. Because the specific flux had only decreased slightly at the end of the test, it is not possible
to project the cleaning frequency actually required for the RO in this application with this
seawater.
4.7 Quality Assurance/Quality Control
4.7.1 Introduction
An important aspect of verification testing is the QA/QC procedures and requirements. As
described in Task C6 of the methods and procedures (Section 3.8.6), a structured QAPP was
implemented to ensure the quality of collected data. Careful adherence to the procedures ensured
that the data presented in this report were of sound quality, defensible, and representative of the
equipment performance. The primary areas of evaluation were representativeness, accuracy,
precision, and completeness.
97
-------
4.7.2 Documentation
The field technicians recorded on-site data and calculations in a field logbook and on specially
prepared field log sheets. The operating logsheets include calibration records for the field
equipment used for on-site analyses. Copies of the daily data log sheets and calibration log sheets
are in Appendix B.
Data from the on-site laboratory and data log sheets were entered into Excel spreadsheets. These
spreadsheets were used to calculate various statistics (average, mean, standard deviation, etc.).
NSF DWSC staff checked 100% of the data entered into the spreadsheets to confirm the
information was correct. The spreadsheets are presented in Appendix C.
Samples collected and delivered to the NSF Laboratory for analysis were tracked using chain-of-
custody forms. Each sample was assigned a location name, date, and time of collection. The
laboratory reported the analytical results using the NSF Chemistry Laboratory management
system reports. These reports were received and reviewed by NSF DWSC staff. These laboratory
data were entered into the data spreadsheets, corrected, and verified in the same manner as the
field data. Lab reports and chain-of-custody forms are included in Appendix F.
4.7.3 Quality Audits
The NSF QA officer performed an on-site audit on October 16, 2007, which was Day 1 of
testing. The audit focused on review the field procedures, including the collection of operating
data and performance of on-site analytical methods. The TQAP requirements were used as the
basis for the audit. All deficiencies were corrected immediately.
The NSF QA Department reviewed the NSF laboratory analytical results for adherence to the
QA requirements for calibration, precision, and accuracy detailed in the project QAPP and for
compliance with the laboratory quality assurance requirements. No deficiencies were found. The
laboratory raw data records (run logs, bench sheets, calibrations records, etc.) are maintained at
NSF and are available for review.
4.7.4 Test Procedure QA/QC
The testing engineers conducted the field monitoring, measurements, and sample collection and
handling in accordance with the EPA-approved TQAP created specifically for this verification.
NSF testing laboratory staff conducted the chemical and microbiological analyses by following
the TQAP. NSF QA Department staff performed audits during testing to ensure the proper
procedures were followed. The audit yielded no significant findings.
4.7.5 Sample Handling
All samples analyzed by the NSF Chemistry and Microbiology Laboratories were labeled with
unique ID numbers. These ID numbers appear in the NSF laboratory reports for the tests. All
chemistry samples were analyzed within allowable holding times. The Bacillus endospores
samples collected on October 23 and 24 were received late due to shipping problems, so they
were not processed for analysis until two days after collection. However, exceeding the holding
time for these samples should not bias the results, since the bacteria are in a spore state, thus are
stable. As shown in Table 4-16, the endospore counts for these days were all above the means for
98
-------
the process stream, with the exception of the <1 UF filtrate count for October 24. In fact, the
maximum UF retentate count for the entire test was from the October 24 sample. Also, the late
samples were received both days at a temperature of 33 °F, so the shipping delays did not result
in the sample temperatures rising above 50 °F (10 °C), which is the maximum suggested holding
temperature in Standard Methods.
4.7.6 Physical and Chemical Analytical Methods QA/QC
The calibrations of all NSF laboratory analytical instruments and the analyses of all parameters
complied with the QA/QC provisions of the NSF Laboratories Quality Assurance Manual.
Bench top field instruments that measured turbidity, pH, temperature and specific conductance
were calibrated daily in accordance with the data quality objectives, except that the daily
calibration check lists do not indicate that the pH/temperature meter was calibrated on October
16, 17, and 21. In-line particle counters and turbidimeters were factory calibrated, and
certificates were provided as required in the TQAP.
4.7.7 Microbiology Laboratory QA/QC
4.7.7.1 Growth Media Positive Controls
All media were checked for sterility and positive growth response when prepared and when used
for microorganism enumeration. The media was discarded if growth occurred on the sterility
check media, or if there was an absence of growth in the positive response check.
4.7.7.2 Negative Controls
For each sample batch processed, an unused membrane filter and a blank with 100 mL of sterile
buffered deionized water filtered through the membrane were also placed onto the appropriate
media and incubated with the samples as negative controls. No growth was observed on any
blanks.
4.7.8 Laboratory Documentation
All laboratory activities were documented using specially prepared laboratory bench sheets and
NSF laboratory reports. Data from the bench sheets and laboratory reports were entered into
Excel spreadsheets. These spreadsheets were used to calculate average feeds and filtrates, and
logic reductions for each challenge. One hundred percent of the data entered into the
spreadsheets was checked by NSF DWSC staff to confirm all data and calculations were correct.
4.7.9 Data Review
NSF QA/QC staff reviewed the raw data records for compliance with QA/QC requirements. NSF
ETV staff checked 100% of the data in the NSF laboratory reports against the lab bench sheets.
4.7.10 Data Quality Indicators
The quality of data generated for this ETV was established through four indicators of data
quality: representativeness, accuracy, precision, and completeness.
99
-------
4.7.10.1 Representativeness
Representativeness refers to the degree to which the data accurately and precisely represent the
expected performance of the EUWP system under conditions expected for use in an emergency
response situation, or theater of war. The EUWP was operated similar to conditions of
deployment in an emergency. As stated in Chapter 2, the raw water source was seawater,
representing a possible application for the EUWP during deployment.
Representativeness was ensured by consistent execution of the test protocol and TQAP for the
test, including timing of sample collection, sampling procedures, and sample preservation.
Representativeness was also ensured by using each analytical method at its optimum capability
to provide results that represent the most accurate and precise measurement it is capable of
achieving.
4.7.10.2 Accuracy
Accuracy was quantified as the percent recovery of the parameter in a sample of known quantity.
Accuracy was measured through use of both matrix spikes of a known quantity and certified
standards during calibration of an instrument. For chemical analyses performed by the NSF
laboratory, certified QC standards and/or matrix spikes were run with each batch of samples. The
percent recoveries of all matrix spikes and standards were within the allowable limits for all
analytical methods.
The TQAP called for the FTO and NSF Chemistry Laboratory to analyze PE samples and report
the results to the NSF QA Department for review. This did not happen as part of the ETV test,
but the NSF Chemistry Laboratory regularly participates in PE studies as part of the ongoing
QA/QC program.
4.7.10.3 Precision
Precision refers to the degree of mutual agreement among individual measurements and provides
an estimate of random error. Precision of duplicate analyses was measured through calculation of
RPD. For the water quality analyses conducted at the NSF laboratory, precision was measured in
two ways. One set of field duplicates was collected for every ten samples sent to NSF. In
addition, the NSF QA program calls for one sample per analytical batch to be analyzed in
duplicate. The duplicate analysis results and RPD calculations for the field duplicates are
presented in Appendix F. The NSF internal duplicate analysis data is not presented. The samples
from this test were batched with samples from other NSF work, so most of the internal duplicates
were from samples not affiliated with the ETV test. For the field measurements, one process
stream was analyzed in duplicate every day. The field measurement duplicate analysis results
and RPD calculations are also presented in Appendix G.
100
-------
All RPD were within the allowable limit of 30% for each parameter with the following
exceptions:
• Of 57 field turbidity duplicates, seven had RPD above 30%. However, five of the seven
were measurements below 0.1 NTU, so as little as 0.02 NTU difference caused the RPD
to be above 30%.
• Of the October 24 weekly sampling duplicates, the barium and lithium samples had RPD
of 40% and 48.3%, respectively.
• Of the November 5 weekly sampling duplicates, the UV254 and Potassium samples had
RPD of 44.1% and 45.7%, respectively.
4.7.10.4 Completeness
Completeness is the proportion of valid, acceptable data generated using each method as
compared to the requirements of the test/QA plan. The completeness objective for data generated
during verification testing is based on the number of samples collected and analyzed for each
parameter and/or method, as presented in Table 3-13.
The completeness goals were met for all water quality parameters. Note that even though most of
the RO permeate pH measurements are not reported (see Section 4.5.3.1 for further discussion),
the completeness percentage for pH measurements was still met because the FTO collected more
daily operation and water quality data than was required by the TQAP. The TQAP specified that
pH measurements would be collected twice daily during the week, and once daily during the
weekend. On most days three or four sets of measurements were collected. A total of 240 pH
measurements were to be made, but there were actually 293 measurements over the course of the
test.
101
-------
References
EPA and NSF International (2002). EPA/NSF Protocol for Equipment Verification Testing for
Removal of Inorganic Constituents. NSF International.
EPA and NSF International (2005). EPA/NSF Protocol for Equipment Verification Testing for
Physical Removal of Microbiological and P articulate Contaminants.
102
------- |